diff --git a/ABOUT-NLS b/ABOUT-NLS
index b1de1b6..3b879cb 100644
--- a/ABOUT-NLS
+++ b/ABOUT-NLS
@@ -1,1282 +1,1282 @@
1 Notes on the Free Translation Project
***************************************
Free software is going international! The Free Translation Project is
a way to get maintainers of free software, translators, and users all
together, so that free software will gradually become able to speak many
languages. A few packages already provide translations for their
messages.
If you found this `ABOUT-NLS' file inside a distribution, you may
assume that the distributed package does use GNU `gettext' internally,
itself available at your nearest GNU archive site. But you do _not_
need to install GNU `gettext' prior to configuring, installing or using
this package with messages translated.
Installers will find here some useful hints. These notes also
explain how users should proceed for getting the programs to use the
available translations. They tell how people wanting to contribute and
work on translations can contact the appropriate team.
1.1 INSTALL Matters
===================
Some packages are "localizable" when properly installed; the programs
they contain can be made to speak your own native language. Most such
packages use GNU `gettext'. Other packages have their own ways to
internationalization, predating GNU `gettext'.
By default, this package will be installed to allow translation of
messages. It will automatically detect whether the system already
provides the GNU `gettext' functions. Installers may use special
options at configuration time for changing the default behaviour. The
command:
./configure --disable-nls
will _totally_ disable translation of messages.
When you already have GNU `gettext' installed on your system and run
configure without an option for your new package, `configure' will
probably detect the previously built and installed `libintl' library
and will decide to use it. If not, you may have to to use the
`--with-libintl-prefix' option to tell `configure' where to look for it.
Internationalized packages usually have many `po/LL.po' files, where
LL gives an ISO 639 two-letter code identifying the language. Unless
translations have been forbidden at `configure' time by using the
`--disable-nls' switch, all available translations are installed
together with the package. However, the environment variable `LINGUAS'
may be set, prior to configuration, to limit the installed set.
`LINGUAS' should then contain a space separated list of two-letter
codes, stating which languages are allowed.
1.2 Using This Package
======================
As a user, if your language has been installed for this package, you
only have to set the `LANG' environment variable to the appropriate
`LL_CC' combination. If you happen to have the `LC_ALL' or some other
`LC_xxx' environment variables set, you should unset them before
setting `LANG', otherwise the setting of `LANG' will not have the
desired effect. Here `LL' is an ISO 639 two-letter language code, and
`CC' is an ISO 3166 two-letter country code. For example, let's
suppose that you speak German and live in Germany. At the shell
prompt, merely execute `setenv LANG de_DE' (in `csh'),
`export LANG; LANG=de_DE' (in `sh') or `export LANG=de_DE' (in `bash').
This can be done from your `.login' or `.profile' file, once and for
all.
You might think that the country code specification is redundant.
But in fact, some languages have dialects in different countries. For
example, `de_AT' is used for Austria, and `pt_BR' for Brazil. The
country code serves to distinguish the dialects.
The locale naming convention of `LL_CC', with `LL' denoting the
language and `CC' denoting the country, is the one use on systems based
on GNU libc. On other systems, some variations of this scheme are
used, such as `LL' or `LL_CC.ENCODING'. You can get the list of
locales supported by your system for your language by running the
command `locale -a | grep '^LL''.
Not all programs have translations for all languages. By default, an
English message is shown in place of a nonexistent translation. If you
understand other languages, you can set up a priority list of languages.
This is done through a different environment variable, called
`LANGUAGE'. GNU `gettext' gives preference to `LANGUAGE' over `LANG'
for the purpose of message handling, but you still need to have `LANG'
set to the primary language; this is required by other parts of the
system libraries. For example, some Swedish users who would rather
read translations in German than English for when Swedish is not
available, set `LANGUAGE' to `sv:de' while leaving `LANG' to `sv_SE'.
Special advice for Norwegian users: The language code for Norwegian
bokma*l changed from `no' to `nb' recently (in 2003). During the
transition period, while some message catalogs for this language are
installed under `nb' and some older ones under `no', it's recommended
for Norwegian users to set `LANGUAGE' to `nb:no' so that both newer and
older translations are used.
In the `LANGUAGE' environment variable, but not in the `LANG'
environment variable, `LL_CC' combinations can be abbreviated as `LL'
to denote the language's main dialect. For example, `de' is equivalent
to `de_DE' (German as spoken in Germany), and `pt' to `pt_PT'
(Portuguese as spoken in Portugal) in this context.
1.3 Translating Teams
=====================
For the Free Translation Project to be a success, we need interested
people who like their own language and write it well, and who are also
able to synergize with other translators speaking the same language.
Each translation team has its own mailing list. The up-to-date list of
teams can be found at the Free Translation Project's homepage,
-`http://translationproject.org/', in the "Teams" area.
+`https://translationproject.org/', in the "Teams" area.
If you'd like to volunteer to _work_ at translating messages, you
should become a member of the translating team for your own language.
The subscribing address is _not_ the same as the list itself, it has
`-request' appended. For example, speakers of Swedish can send a
message to `sv-request@li.org', having this message body:
subscribe
Keep in mind that team members are expected to participate
_actively_ in translations, or at solving translational difficulties,
rather than merely lurking around. If your team does not exist yet and
you want to start one, or if you are unsure about what to do or how to
get started, please write to `coordinator@translationproject.org' to
reach the coordinator for all translator teams.
The English team is special. It works at improving and uniformizing
the terminology in use. Proven linguistic skills are praised more than
programming skills, here.
1.4 Available Packages
======================
Languages are not equally supported in all packages. The following
matrix shows the current state of internationalization, as of June
2010. The matrix shows, in regard of each package, for which languages
PO files have been submitted to translation coordination, with a
translation percentage of at least 50%.
Ready PO files af am an ar as ast az be be@latin bg bn_IN bs ca
+--------------------------------------------------+
a2ps | [] [] |
aegis | |
ant-phone | |
anubis | |
aspell | [] [] |
bash | |
bfd | |
bibshelf | [] |
binutils | |
bison | |
bison-runtime | [] |
bluez-pin | [] [] |
bombono-dvd | |
buzztard | |
cflow | |
clisp | |
coreutils | [] [] |
cpio | |
cppi | |
cpplib | [] |
cryptsetup | |
dfarc | |
dialog | [] [] |
dico | |
diffutils | [] |
dink | |
doodle | |
e2fsprogs | [] |
enscript | [] |
exif | |
fetchmail | [] |
findutils | [] |
flex | [] |
freedink | |
gas | |
gawk | [] [] |
gcal | [] |
gcc | |
gettext-examples | [] [] [] [] |
gettext-runtime | [] [] |
gettext-tools | [] [] |
gip | [] |
gjay | |
gliv | [] |
glunarclock | [] [] |
gnubiff | |
gnucash | [] |
gnuedu | |
gnulib | |
gnunet | |
gnunet-gtk | |
gnutls | |
gold | |
gpe-aerial | |
gpe-beam | |
gpe-bluetooth | |
gpe-calendar | |
gpe-clock | [] |
gpe-conf | |
gpe-contacts | |
gpe-edit | |
gpe-filemanager | |
gpe-go | |
gpe-login | |
gpe-ownerinfo | [] |
gpe-package | |
gpe-sketchbook | |
gpe-su | [] |
gpe-taskmanager | [] |
gpe-timesheet | [] |
gpe-today | [] |
gpe-todo | |
gphoto2 | |
gprof | [] |
gpsdrive | |
gramadoir | |
grep | |
grub | [] [] |
gsasl | |
gss | |
gst-plugins-bad | [] |
gst-plugins-base | [] |
gst-plugins-good | [] |
gst-plugins-ugly | [] |
gstreamer | [] [] [] |
gtick | |
gtkam | [] |
gtkorphan | [] |
gtkspell | [] [] [] |
gutenprint | |
hello | [] |
help2man | |
hylafax | |
idutils | |
indent | [] [] |
iso_15924 | |
iso_3166 | [] [] [] [] [] [] [] |
iso_3166_2 | |
iso_4217 | |
iso_639 | [] [] [] [] |
iso_639_3 | |
jwhois | |
kbd | |
keytouch | [] |
keytouch-editor | |
keytouch-keyboa... | [] |
klavaro | [] |
latrine | |
ld | [] |
leafpad | [] [] |
libc | [] [] |
libexif | () |
libextractor | |
libgnutls | |
libgpewidget | |
libgpg-error | |
libgphoto2 | |
libgphoto2_port | |
libgsasl | |
libiconv | [] |
libidn | |
lifelines | |
liferea | [] [] |
lilypond | |
linkdr | [] |
lordsawar | |
lprng | |
lynx | [] |
m4 | |
mailfromd | |
mailutils | |
make | |
man-db | |
man-db-manpages | |
minicom | |
mkisofs | |
myserver | |
nano | [] [] |
opcodes | |
parted | |
pies | |
popt | |
psmisc | |
pspp | [] |
pwdutils | |
radius | [] |
recode | [] [] |
rosegarden | |
rpm | |
rush | |
sarg | |
screem | |
scrollkeeper | [] [] [] |
sed | [] [] |
sharutils | [] [] |
shishi | |
skencil | |
solfege | |
solfege-manual | |
soundtracker | |
sp | |
sysstat | |
tar | [] |
texinfo | |
tin | |
unicode-han-tra... | |
unicode-transla... | |
util-linux-ng | [] |
vice | |
vmm | |
vorbis-tools | |
wastesedge | |
wdiff | |
wget | [] [] |
wyslij-po | |
xchat | [] [] [] [] |
xdg-user-dirs | [] [] [] [] [] [] [] [] [] |
xkeyboard-config | [] [] |
+--------------------------------------------------+
af am an ar as ast az be be@latin bg bn_IN bs ca
6 0 1 2 3 19 1 10 3 28 3 1 38
crh cs da de el en en_GB en_ZA eo es et eu fa
+-------------------------------------------------+
a2ps | [] [] [] [] [] [] [] |
aegis | [] [] [] |
ant-phone | [] () |
anubis | [] [] |
aspell | [] [] [] [] [] |
bash | [] [] [] |
bfd | [] |
bibshelf | [] [] [] |
binutils | [] |
bison | [] [] |
bison-runtime | [] [] [] [] |
bluez-pin | [] [] [] [] [] [] |
bombono-dvd | [] |
buzztard | [] [] [] |
cflow | [] [] |
clisp | [] [] [] [] |
coreutils | [] [] [] [] |
cpio | |
cppi | |
cpplib | [] [] [] |
cryptsetup | [] |
dfarc | [] [] [] |
dialog | [] [] [] [] [] |
dico | |
diffutils | [] [] [] [] [] [] |
dink | [] [] [] |
doodle | [] |
e2fsprogs | [] [] [] |
enscript | [] [] [] |
exif | () [] [] |
fetchmail | [] [] () [] [] [] |
findutils | [] [] [] |
flex | [] [] |
freedink | [] [] [] |
gas | [] |
gawk | [] [] [] |
gcal | [] |
gcc | [] [] |
gettext-examples | [] [] [] [] |
gettext-runtime | [] [] [] [] |
gettext-tools | [] [] [] |
gip | [] [] [] [] |
gjay | [] |
gliv | [] [] [] |
glunarclock | [] [] |
gnubiff | () |
gnucash | [] () () () () |
gnuedu | [] [] |
gnulib | [] [] |
gnunet | |
gnunet-gtk | [] |
gnutls | [] [] |
gold | [] |
gpe-aerial | [] [] [] [] |
gpe-beam | [] [] [] [] |
gpe-bluetooth | [] [] |
gpe-calendar | [] |
gpe-clock | [] [] [] [] |
gpe-conf | [] [] [] |
gpe-contacts | [] [] [] |
gpe-edit | [] [] |
gpe-filemanager | [] [] [] |
gpe-go | [] [] [] [] |
gpe-login | [] [] |
gpe-ownerinfo | [] [] [] [] |
gpe-package | [] [] [] |
gpe-sketchbook | [] [] [] [] |
gpe-su | [] [] [] [] |
gpe-taskmanager | [] [] [] [] |
gpe-timesheet | [] [] [] [] |
gpe-today | [] [] [] [] |
gpe-todo | [] [] [] |
gphoto2 | [] [] () [] [] [] |
gprof | [] [] [] |
gpsdrive | [] [] [] |
gramadoir | [] [] [] |
grep | [] |
grub | [] [] |
gsasl | [] |
gss | |
gst-plugins-bad | [] [] [] [] [] |
gst-plugins-base | [] [] [] [] [] |
gst-plugins-good | [] [] [] [] [] [] |
gst-plugins-ugly | [] [] [] [] [] [] |
gstreamer | [] [] [] [] [] |
gtick | [] () [] |
gtkam | [] [] () [] [] |
gtkorphan | [] [] [] [] |
gtkspell | [] [] [] [] [] [] [] |
gutenprint | [] [] [] |
hello | [] [] [] [] |
help2man | [] |
hylafax | [] [] |
idutils | [] [] |
indent | [] [] [] [] [] [] [] |
iso_15924 | [] () [] [] |
iso_3166 | [] [] [] [] () [] [] [] () |
iso_3166_2 | () |
iso_4217 | [] [] [] () [] [] |
iso_639 | [] [] [] [] () [] [] |
iso_639_3 | [] |
jwhois | [] |
kbd | [] [] [] [] [] |
keytouch | [] [] |
keytouch-editor | [] [] |
keytouch-keyboa... | [] |
klavaro | [] [] [] [] |
latrine | [] () |
ld | [] [] |
leafpad | [] [] [] [] [] [] |
libc | [] [] [] [] |
libexif | [] [] () |
libextractor | |
libgnutls | [] |
libgpewidget | [] [] |
libgpg-error | [] [] |
libgphoto2 | [] () |
libgphoto2_port | [] () [] |
libgsasl | |
libiconv | [] [] [] [] [] |
libidn | [] [] [] |
lifelines | [] () |
liferea | [] [] [] [] [] |
lilypond | [] [] [] |
linkdr | [] [] [] |
lordsawar | [] |
lprng | |
lynx | [] [] [] [] |
m4 | [] [] [] [] |
mailfromd | |
mailutils | [] |
make | [] [] [] |
man-db | |
man-db-manpages | |
minicom | [] [] [] [] |
mkisofs | |
myserver | |
nano | [] [] [] |
opcodes | [] [] |
parted | [] [] |
pies | |
popt | [] [] [] [] [] |
psmisc | [] [] [] |
pspp | [] |
pwdutils | [] |
radius | [] |
recode | [] [] [] [] [] [] |
rosegarden | () () () |
rpm | [] [] [] |
rush | |
sarg | |
screem | |
scrollkeeper | [] [] [] [] [] |
sed | [] [] [] [] [] [] |
sharutils | [] [] [] [] |
shishi | |
skencil | [] () [] |
solfege | [] [] [] |
solfege-manual | [] [] |
soundtracker | [] [] [] |
sp | [] |
sysstat | [] [] [] |
tar | [] [] [] [] |
texinfo | [] [] [] |
tin | [] [] |
unicode-han-tra... | |
unicode-transla... | |
util-linux-ng | [] [] [] [] |
vice | () () |
vmm | [] |
vorbis-tools | [] [] |
wastesedge | [] |
wdiff | [] [] |
wget | [] [] [] |
wyslij-po | |
xchat | [] [] [] [] [] |
xdg-user-dirs | [] [] [] [] [] [] [] [] [] |
xkeyboard-config | [] [] [] [] [] [] |
+-------------------------------------------------+
crh cs da de el en en_GB en_ZA eo es et eu fa
5 64 105 117 18 1 8 0 28 89 18 19 0
fi fr ga gl gu he hi hr hu hy id is it ja ka kn
+----------------------------------------------------+
a2ps | [] [] [] [] |
aegis | [] [] |
ant-phone | [] [] |
anubis | [] [] [] [] |
aspell | [] [] [] [] |
bash | [] [] [] [] |
bfd | [] [] [] |
bibshelf | [] [] [] [] [] |
binutils | [] [] [] |
bison | [] [] [] [] |
bison-runtime | [] [] [] [] [] [] |
bluez-pin | [] [] [] [] [] [] [] [] |
bombono-dvd | [] |
buzztard | [] |
cflow | [] [] [] |
clisp | [] |
coreutils | [] [] [] [] [] |
cpio | [] [] [] [] |
cppi | [] [] |
cpplib | [] [] [] |
cryptsetup | [] [] [] |
dfarc | [] [] [] |
dialog | [] [] [] [] [] [] [] |
dico | |
diffutils | [] [] [] [] [] [] [] [] [] |
dink | [] |
doodle | [] [] |
e2fsprogs | [] [] |
enscript | [] [] [] [] |
exif | [] [] [] [] [] [] |
fetchmail | [] [] [] [] |
findutils | [] [] [] [] [] [] |
flex | [] [] [] |
freedink | [] [] [] |
gas | [] [] |
gawk | [] [] [] [] () [] |
gcal | [] |
gcc | [] |
gettext-examples | [] [] [] [] [] [] [] |
gettext-runtime | [] [] [] [] [] [] |
gettext-tools | [] [] [] [] |
gip | [] [] [] [] [] [] |
gjay | [] |
gliv | [] () |
glunarclock | [] [] [] [] |
gnubiff | () [] () |
gnucash | () () () () () [] |
gnuedu | [] [] |
gnulib | [] [] [] [] [] [] |
gnunet | |
gnunet-gtk | [] |
gnutls | [] [] |
gold | [] [] |
gpe-aerial | [] [] [] |
gpe-beam | [] [] [] [] |
gpe-bluetooth | [] [] [] [] |
gpe-calendar | [] [] |
gpe-clock | [] [] [] [] [] |
gpe-conf | [] [] [] [] |
gpe-contacts | [] [] [] [] |
gpe-edit | [] [] [] |
gpe-filemanager | [] [] [] [] |
gpe-go | [] [] [] [] [] |
gpe-login | [] [] [] |
gpe-ownerinfo | [] [] [] [] [] |
gpe-package | [] [] [] |
gpe-sketchbook | [] [] [] [] |
gpe-su | [] [] [] [] [] [] |
gpe-taskmanager | [] [] [] [] [] |
gpe-timesheet | [] [] [] [] [] |
gpe-today | [] [] [] [] [] [] [] |
gpe-todo | [] [] [] |
gphoto2 | [] [] [] [] [] [] |
gprof | [] [] [] [] |
gpsdrive | [] [] [] |
gramadoir | [] [] [] |
grep | [] [] |
grub | [] [] [] [] |
gsasl | [] [] [] [] [] |
gss | [] [] [] [] [] |
gst-plugins-bad | [] [] [] [] [] [] |
gst-plugins-base | [] [] [] [] [] [] |
gst-plugins-good | [] [] [] [] [] [] |
gst-plugins-ugly | [] [] [] [] [] [] |
gstreamer | [] [] [] [] [] |
gtick | [] [] [] [] [] |
gtkam | [] [] [] [] [] |
gtkorphan | [] [] [] |
gtkspell | [] [] [] [] [] [] [] [] [] |
gutenprint | [] [] [] [] |
hello | [] [] [] |
help2man | [] [] |
hylafax | [] |
idutils | [] [] [] [] [] [] |
indent | [] [] [] [] [] [] [] [] |
iso_15924 | [] () [] [] |
iso_3166 | [] () [] [] [] [] [] [] [] [] [] [] |
iso_3166_2 | () [] [] [] |
iso_4217 | [] () [] [] [] [] |
iso_639 | [] () [] [] [] [] [] [] [] |
iso_639_3 | () [] [] |
jwhois | [] [] [] [] [] |
kbd | [] [] |
keytouch | [] [] [] [] [] [] |
keytouch-editor | [] [] [] [] [] |
keytouch-keyboa... | [] [] [] [] [] |
klavaro | [] [] |
latrine | [] [] [] |
ld | [] [] [] [] |
leafpad | [] [] [] [] [] [] [] () |
libc | [] [] [] [] [] |
libexif | [] |
libextractor | |
libgnutls | [] [] |
libgpewidget | [] [] [] [] |
libgpg-error | [] [] |
libgphoto2 | [] [] [] |
libgphoto2_port | [] [] [] |
libgsasl | [] [] [] [] [] |
libiconv | [] [] [] [] [] [] |
libidn | [] [] [] [] |
lifelines | () |
liferea | [] [] [] [] |
lilypond | [] [] |
linkdr | [] [] [] [] [] |
lordsawar | |
lprng | [] |
lynx | [] [] [] [] [] |
m4 | [] [] [] [] [] [] |
mailfromd | |
mailutils | [] [] |
make | [] [] [] [] [] [] [] [] [] |
man-db | [] [] |
man-db-manpages | [] |
minicom | [] [] [] [] [] |
mkisofs | [] [] [] [] |
myserver | |
nano | [] [] [] [] [] [] |
opcodes | [] [] [] [] |
parted | [] [] [] [] |
pies | |
popt | [] [] [] [] [] [] [] [] [] |
psmisc | [] [] [] |
pspp | |
pwdutils | [] [] |
radius | [] [] |
recode | [] [] [] [] [] [] [] [] |
rosegarden | () () () () () |
rpm | [] [] |
rush | |
sarg | [] |
screem | [] [] |
scrollkeeper | [] [] [] [] |
sed | [] [] [] [] [] [] [] [] |
sharutils | [] [] [] [] [] [] [] |
shishi | [] |
skencil | [] |
solfege | [] [] [] [] |
solfege-manual | [] [] |
soundtracker | [] [] |
sp | [] () |
sysstat | [] [] [] [] [] |
tar | [] [] [] [] [] [] [] |
texinfo | [] [] [] [] |
tin | [] |
unicode-han-tra... | |
unicode-transla... | [] [] |
util-linux-ng | [] [] [] [] [] [] |
vice | () () () |
vmm | [] |
vorbis-tools | [] |
wastesedge | () () |
wdiff | [] |
wget | [] [] [] [] [] [] [] [] |
wyslij-po | [] [] [] |
xchat | [] [] [] [] [] [] [] [] [] |
xdg-user-dirs | [] [] [] [] [] [] [] [] [] [] [] [] [] |
xkeyboard-config | [] [] [] [] [] |
+----------------------------------------------------+
fi fr ga gl gu he hi hr hu hy id is it ja ka kn
105 121 53 20 4 8 3 5 53 2 120 5 84 67 0 4
ko ku ky lg lt lv mk ml mn mr ms mt nb nds ne
+-----------------------------------------------+
a2ps | [] |
aegis | |
ant-phone | |
anubis | [] [] |
aspell | [] |
bash | |
bfd | |
bibshelf | [] [] |
binutils | |
bison | [] |
bison-runtime | [] [] [] [] [] |
bluez-pin | [] [] [] [] [] |
bombono-dvd | |
buzztard | |
cflow | |
clisp | |
coreutils | [] |
cpio | |
cppi | |
cpplib | |
cryptsetup | |
dfarc | [] |
dialog | [] [] [] [] [] |
dico | |
diffutils | [] [] |
dink | |
doodle | |
e2fsprogs | |
enscript | |
exif | [] |
fetchmail | |
findutils | |
flex | |
freedink | [] |
gas | |
gawk | |
gcal | |
gcc | |
gettext-examples | [] [] [] [] |
gettext-runtime | [] |
gettext-tools | [] |
gip | [] [] |
gjay | |
gliv | |
glunarclock | [] |
gnubiff | |
gnucash | () () () () |
gnuedu | |
gnulib | |
gnunet | |
gnunet-gtk | |
gnutls | [] |
gold | |
gpe-aerial | [] |
gpe-beam | [] |
gpe-bluetooth | [] [] |
gpe-calendar | [] |
gpe-clock | [] [] [] [] [] |
gpe-conf | [] [] |
gpe-contacts | [] [] |
gpe-edit | [] |
gpe-filemanager | [] [] |
gpe-go | [] [] [] |
gpe-login | [] |
gpe-ownerinfo | [] [] |
gpe-package | [] [] |
gpe-sketchbook | [] [] |
gpe-su | [] [] [] [] [] [] |
gpe-taskmanager | [] [] [] [] [] [] |
gpe-timesheet | [] [] |
gpe-today | [] [] [] [] |
gpe-todo | [] [] |
gphoto2 | |
gprof | [] |
gpsdrive | |
gramadoir | |
grep | |
grub | |
gsasl | |
gss | |
gst-plugins-bad | [] [] [] [] |
gst-plugins-base | [] [] |
gst-plugins-good | [] [] |
gst-plugins-ugly | [] [] [] [] [] |
gstreamer | |
gtick | |
gtkam | [] |
gtkorphan | [] [] |
gtkspell | [] [] [] [] [] [] [] |
gutenprint | |
hello | [] [] [] |
help2man | |
hylafax | |
idutils | |
indent | |
iso_15924 | [] [] |
iso_3166 | [] [] () [] [] [] [] [] |
iso_3166_2 | |
iso_4217 | [] [] |
iso_639 | [] [] |
iso_639_3 | [] |
jwhois | [] |
kbd | |
keytouch | [] |
keytouch-editor | [] |
keytouch-keyboa... | [] |
klavaro | [] |
latrine | [] |
ld | |
leafpad | [] [] [] |
libc | [] |
libexif | |
libextractor | |
libgnutls | [] |
libgpewidget | [] [] |
libgpg-error | |
libgphoto2 | |
libgphoto2_port | |
libgsasl | |
libiconv | |
libidn | |
lifelines | |
liferea | |
lilypond | |
linkdr | |
lordsawar | |
lprng | |
lynx | |
m4 | |
mailfromd | |
mailutils | |
make | [] |
man-db | |
man-db-manpages | |
minicom | [] |
mkisofs | |
myserver | |
nano | [] [] |
opcodes | |
parted | |
pies | |
popt | [] [] [] |
psmisc | |
pspp | |
pwdutils | |
radius | |
recode | |
rosegarden | |
rpm | |
rush | |
sarg | |
screem | |
scrollkeeper | [] [] |
sed | |
sharutils | |
shishi | |
skencil | |
solfege | [] |
solfege-manual | |
soundtracker | |
sp | |
sysstat | [] |
tar | [] |
texinfo | [] |
tin | |
unicode-han-tra... | |
unicode-transla... | |
util-linux-ng | |
vice | |
vmm | |
vorbis-tools | |
wastesedge | |
wdiff | |
wget | [] |
wyslij-po | |
xchat | [] [] [] |
xdg-user-dirs | [] [] [] [] [] [] [] [] |
xkeyboard-config | [] [] [] |
+-----------------------------------------------+
ko ku ky lg lt lv mk ml mn mr ms mt nb nds ne
20 5 10 1 13 48 4 2 2 4 24 10 20 3 1
nl nn or pa pl ps pt pt_BR ro ru rw sk sl sq sr
+---------------------------------------------------+
a2ps | [] [] [] [] [] [] [] [] |
aegis | [] [] [] |
ant-phone | [] [] |
anubis | [] [] [] |
aspell | [] [] [] [] [] |
bash | [] [] |
bfd | [] |
bibshelf | [] [] |
binutils | [] [] |
bison | [] [] [] |
bison-runtime | [] [] [] [] [] [] [] |
bluez-pin | [] [] [] [] [] [] [] [] |
bombono-dvd | [] () |
buzztard | [] [] |
cflow | [] |
clisp | [] [] |
coreutils | [] [] [] [] [] [] |
cpio | [] [] [] |
cppi | [] |
cpplib | [] |
cryptsetup | [] |
dfarc | [] |
dialog | [] [] [] [] |
dico | [] |
diffutils | [] [] [] [] [] [] |
dink | () |
doodle | [] [] |
e2fsprogs | [] [] |
enscript | [] [] [] [] [] |
exif | [] [] [] () [] |
fetchmail | [] [] [] [] |
findutils | [] [] [] [] [] |
flex | [] [] [] [] [] |
freedink | [] [] |
gas | |
gawk | [] [] [] [] |
gcal | |
gcc | [] |
gettext-examples | [] [] [] [] [] [] [] [] |
gettext-runtime | [] [] [] [] [] [] [] [] [] |
gettext-tools | [] [] [] [] [] [] |
gip | [] [] [] [] [] |
gjay | |
gliv | [] [] [] [] [] [] |
glunarclock | [] [] [] [] [] |
gnubiff | [] () |
gnucash | [] () () () |
gnuedu | [] |
gnulib | [] [] [] [] |
gnunet | |
gnunet-gtk | |
gnutls | [] [] |
gold | |
gpe-aerial | [] [] [] [] [] [] [] |
gpe-beam | [] [] [] [] [] [] [] |
gpe-bluetooth | [] [] |
gpe-calendar | [] [] [] [] |
gpe-clock | [] [] [] [] [] [] [] [] |
gpe-conf | [] [] [] [] [] [] [] |
gpe-contacts | [] [] [] [] [] |
gpe-edit | [] [] [] |
gpe-filemanager | [] [] [] |
gpe-go | [] [] [] [] [] [] [] [] |
gpe-login | [] [] |
gpe-ownerinfo | [] [] [] [] [] [] [] [] |
gpe-package | [] [] |
gpe-sketchbook | [] [] [] [] [] [] [] |
gpe-su | [] [] [] [] [] [] [] [] |
gpe-taskmanager | [] [] [] [] [] [] [] [] |
gpe-timesheet | [] [] [] [] [] [] [] [] |
gpe-today | [] [] [] [] [] [] [] [] |
gpe-todo | [] [] [] [] [] |
gphoto2 | [] [] [] [] [] [] [] [] |
gprof | [] [] [] |
gpsdrive | [] [] |
gramadoir | [] [] |
grep | [] [] [] [] |
grub | [] [] [] |
gsasl | [] [] [] [] |
gss | [] [] [] |
gst-plugins-bad | [] [] [] [] [] [] |
gst-plugins-base | [] [] [] [] [] |
gst-plugins-good | [] [] [] [] [] |
gst-plugins-ugly | [] [] [] [] [] [] |
gstreamer | [] [] [] [] [] |
gtick | [] [] [] |
gtkam | [] [] [] [] [] [] |
gtkorphan | [] |
gtkspell | [] [] [] [] [] [] [] [] [] [] |
gutenprint | [] [] |
hello | [] [] [] [] |
help2man | [] [] |
hylafax | [] |
idutils | [] [] [] [] [] |
indent | [] [] [] [] [] [] [] |
iso_15924 | [] [] [] [] |
iso_3166 | [] [] [] [] [] () [] [] [] [] [] [] [] [] |
iso_3166_2 | [] [] [] |
iso_4217 | [] [] [] [] [] [] [] [] |
iso_639 | [] [] [] [] [] [] [] [] [] |
iso_639_3 | [] [] |
jwhois | [] [] [] [] |
kbd | [] [] [] |
keytouch | [] [] [] |
keytouch-editor | [] [] [] |
keytouch-keyboa... | [] [] [] |
klavaro | [] [] |
latrine | [] [] |
ld | |
leafpad | [] [] [] [] [] [] [] [] [] |
libc | [] [] [] [] |
libexif | [] [] () [] |
libextractor | |
libgnutls | [] [] |
libgpewidget | [] [] [] |
libgpg-error | [] [] |
libgphoto2 | [] [] |
libgphoto2_port | [] [] [] [] [] |
libgsasl | [] [] [] [] [] |
libiconv | [] [] [] [] [] |
libidn | [] [] |
lifelines | [] [] |
liferea | [] [] [] [] [] () () [] |
lilypond | [] |
linkdr | [] [] [] |
lordsawar | |
lprng | [] |
lynx | [] [] [] |
m4 | [] [] [] [] [] |
mailfromd | [] |
mailutils | [] |
make | [] [] [] [] |
man-db | [] [] [] |
man-db-manpages | [] [] [] |
minicom | [] [] [] [] |
mkisofs | [] [] [] |
myserver | |
nano | [] [] [] [] |
opcodes | [] [] |
parted | [] [] [] [] |
pies | [] |
popt | [] [] [] [] |
psmisc | [] [] [] |
pspp | [] [] |
pwdutils | [] |
radius | [] [] [] |
recode | [] [] [] [] [] [] [] [] |
rosegarden | () () |
rpm | [] [] [] |
rush | [] [] |
sarg | |
screem | |
scrollkeeper | [] [] [] [] [] [] [] [] |
sed | [] [] [] [] [] [] [] [] [] |
sharutils | [] [] [] [] |
shishi | [] |
skencil | [] [] |
solfege | [] [] [] [] |
solfege-manual | [] [] [] |
soundtracker | [] |
sp | |
sysstat | [] [] [] [] |
tar | [] [] [] [] |
texinfo | [] [] [] [] |
tin | [] |
unicode-han-tra... | |
unicode-transla... | |
util-linux-ng | [] [] [] [] [] |
vice | [] |
vmm | [] |
vorbis-tools | [] [] |
wastesedge | [] |
wdiff | [] [] |
wget | [] [] [] [] [] [] [] |
wyslij-po | [] [] [] |
xchat | [] [] [] [] [] [] [] [] [] |
xdg-user-dirs | [] [] [] [] [] [] [] [] [] [] [] [] [] [] |
xkeyboard-config | [] [] [] |
+---------------------------------------------------+
nl nn or pa pl ps pt pt_BR ro ru rw sk sl sq sr
135 10 4 7 105 1 29 62 47 91 3 54 46 9 37
sv sw ta te tg th tr uk vi wa zh_CN zh_HK zh_TW
+---------------------------------------------------+
a2ps | [] [] [] [] [] | 27
aegis | [] | 9
ant-phone | [] [] [] [] | 9
anubis | [] [] [] [] | 15
aspell | [] [] [] | 20
bash | [] [] [] | 12
bfd | [] | 6
bibshelf | [] [] [] | 16
binutils | [] [] | 8
bison | [] [] | 12
bison-runtime | [] [] [] [] [] [] | 29
bluez-pin | [] [] [] [] [] [] [] [] | 37
bombono-dvd | [] | 4
buzztard | [] | 7
cflow | [] [] [] | 9
clisp | | 10
coreutils | [] [] [] [] | 22
cpio | [] [] [] [] [] [] | 13
cppi | [] [] | 5
cpplib | [] [] [] [] [] [] | 14
cryptsetup | [] [] | 7
dfarc | [] | 9
dialog | [] [] [] [] [] [] [] | 30
dico | [] | 2
diffutils | [] [] [] [] [] [] | 30
dink | | 4
doodle | [] [] | 7
e2fsprogs | [] [] [] | 11
enscript | [] [] [] [] | 17
exif | [] [] [] | 16
fetchmail | [] [] [] | 17
findutils | [] [] [] [] [] | 20
flex | [] [] [] [] | 15
freedink | [] | 10
gas | [] | 4
gawk | [] [] [] [] | 18
gcal | [] [] | 5
gcc | [] [] [] | 7
gettext-examples | [] [] [] [] [] [] [] | 34
gettext-runtime | [] [] [] [] [] [] [] | 29
gettext-tools | [] [] [] [] [] [] | 22
gip | [] [] [] [] | 22
gjay | [] | 3
gliv | [] [] [] | 14
glunarclock | [] [] [] [] [] | 19
gnubiff | [] [] | 4
gnucash | () [] () [] () | 10
gnuedu | [] [] | 7
gnulib | [] [] [] [] | 16
gnunet | [] | 1
gnunet-gtk | [] [] [] | 5
gnutls | [] [] [] | 10
gold | [] | 4
gpe-aerial | [] [] [] | 18
gpe-beam | [] [] [] | 19
gpe-bluetooth | [] [] [] | 13
gpe-calendar | [] [] [] [] | 12
gpe-clock | [] [] [] [] [] | 28
gpe-conf | [] [] [] [] | 20
gpe-contacts | [] [] [] | 17
gpe-edit | [] [] [] | 12
gpe-filemanager | [] [] [] [] | 16
gpe-go | [] [] [] [] [] | 25
gpe-login | [] [] [] | 11
gpe-ownerinfo | [] [] [] [] [] | 25
gpe-package | [] [] [] | 13
gpe-sketchbook | [] [] [] | 20
gpe-su | [] [] [] [] [] | 30
gpe-taskmanager | [] [] [] [] [] | 29
gpe-timesheet | [] [] [] [] [] | 25
gpe-today | [] [] [] [] [] [] | 30
gpe-todo | [] [] [] [] | 17
gphoto2 | [] [] [] [] [] | 24
gprof | [] [] [] | 15
gpsdrive | [] [] [] | 11
gramadoir | [] [] [] | 11
grep | [] [] [] | 10
grub | [] [] [] | 14
gsasl | [] [] [] [] | 14
gss | [] [] [] | 11
gst-plugins-bad | [] [] [] [] | 26
gst-plugins-base | [] [] [] [] [] | 24
gst-plugins-good | [] [] [] [] | 24
gst-plugins-ugly | [] [] [] [] [] | 29
gstreamer | [] [] [] [] | 22
gtick | [] [] [] | 13
gtkam | [] [] [] | 20
gtkorphan | [] [] [] | 14
gtkspell | [] [] [] [] [] [] [] [] [] | 45
gutenprint | [] | 10
hello | [] [] [] [] [] [] | 21
help2man | [] [] | 7
hylafax | [] | 5
idutils | [] [] [] [] | 17
indent | [] [] [] [] [] [] | 30
iso_15924 | () [] () [] [] | 16
iso_3166 | [] [] () [] [] () [] [] [] () | 53
iso_3166_2 | () [] () [] | 9
iso_4217 | [] () [] [] () [] [] | 26
iso_639 | [] [] [] () [] () [] [] [] [] | 38
iso_639_3 | [] () | 8
jwhois | [] [] [] [] [] | 16
kbd | [] [] [] [] [] | 15
keytouch | [] [] [] | 16
keytouch-editor | [] [] [] | 14
keytouch-keyboa... | [] [] [] | 14
klavaro | [] | 11
latrine | [] [] [] | 10
ld | [] [] [] [] | 11
leafpad | [] [] [] [] [] [] | 33
libc | [] [] [] [] [] | 21
libexif | [] () | 7
libextractor | [] | 1
libgnutls | [] [] [] | 9
libgpewidget | [] [] [] | 14
libgpg-error | [] [] [] | 9
libgphoto2 | [] [] | 8
libgphoto2_port | [] [] [] [] | 14
libgsasl | [] [] [] | 13
libiconv | [] [] [] [] | 21
libidn | () [] [] | 11
lifelines | [] | 4
liferea | [] [] [] | 21
lilypond | [] | 7
linkdr | [] [] [] [] [] | 17
lordsawar | | 1
lprng | [] | 3
lynx | [] [] [] [] | 17
m4 | [] [] [] [] | 19
mailfromd | [] [] | 3
mailutils | [] | 5
make | [] [] [] [] | 21
man-db | [] [] [] | 8
man-db-manpages | | 4
minicom | [] [] | 16
mkisofs | [] [] | 9
myserver | | 0
nano | [] [] [] [] | 21
opcodes | [] [] [] | 11
parted | [] [] [] [] [] | 15
pies | [] [] | 3
popt | [] [] [] [] [] [] | 27
psmisc | [] [] | 11
pspp | | 4
pwdutils | [] [] | 6
radius | [] [] | 9
recode | [] [] [] [] | 28
rosegarden | () | 0
rpm | [] [] [] | 11
rush | [] [] | 4
sarg | | 1
screem | [] | 3
scrollkeeper | [] [] [] [] [] | 27
sed | [] [] [] [] [] | 30
sharutils | [] [] [] [] [] | 22
shishi | [] | 3
skencil | [] [] | 7
solfege | [] [] [] [] | 16
solfege-manual | [] | 8
soundtracker | [] [] [] | 9
sp | [] | 3
sysstat | [] [] | 15
tar | [] [] [] [] [] [] | 23
texinfo | [] [] [] [] [] | 17
tin | | 4
unicode-han-tra... | | 0
unicode-transla... | | 2
util-linux-ng | [] [] [] [] | 20
vice | () () | 1
vmm | [] | 4
vorbis-tools | [] | 6
wastesedge | | 2
wdiff | [] [] | 7
wget | [] [] [] [] [] | 26
wyslij-po | [] [] | 8
xchat | [] [] [] [] [] [] | 36
xdg-user-dirs | [] [] [] [] [] [] [] [] [] [] | 63
xkeyboard-config | [] [] [] | 22
+---------------------------------------------------+
85 teams sv sw ta te tg th tr uk vi wa zh_CN zh_HK zh_TW
178 domains 119 1 3 3 0 10 65 51 155 17 98 7 41 2618
Some counters in the preceding matrix are higher than the number of
visible blocks let us expect. This is because a few extra PO files are
used for implementing regional variants of languages, or language
dialects.
For a PO file in the matrix above to be effective, the package to
which it applies should also have been internationalized and
distributed as such by its maintainer. There might be an observable
lag between the mere existence a PO file and its wide availability in a
distribution.
If June 2010 seems to be old, you may fetch a more recent copy of
this `ABOUT-NLS' file on most GNU archive sites. The most up-to-date
matrix with full percentage details can be found at
-`http://translationproject.org/extra/matrix.html'.
+`https://translationproject.org/extra/matrix.html'.
1.5 Using `gettext' in new packages
===================================
If you are writing a freely available program and want to
internationalize it you are welcome to use GNU `gettext' in your
package. Of course you have to respect the GNU Library General Public
License which covers the use of the GNU `gettext' library. This means
in particular that even non-free programs can use `libintl' as a shared
library, whereas only free software can use `libintl' as a static
library or use modified versions of `libintl'.
Once the sources are changed appropriately and the setup can handle
the use of `gettext' the only thing missing are the translations. The
Free Translation Project is also available for packages which are not
developed inside the GNU project. Therefore the information given above
applies also for every other Free Software Project. Contact
`coordinator@translationproject.org' to make the `.pot' files available
to the translation teams.
diff --git a/Makefile.am b/Makefile.am
index e5f3f56..4f9f094 100644
--- a/Makefile.am
+++ b/Makefile.am
@@ -1,79 +1,79 @@
# Makefile.am for libgpg-error.
# Copyright (C) 2003, 2006, 2007, 2013 g10 Code GmbH
#
# This file is part of libgpg-error.
#
# libgpg-error is free software; you can redistribute it and/or modify
# it under the terms of the GNU Lesser General Public License as
# published by the Free Software Foundation; either version 2.1 of the
# License, or (at your option) any later version.
#
# libgpg-error is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU Lesser General Public License for more details.
#
# You should have received a copy of the GNU Lesser General Public
-# License along with this program; if not, see .
+# License along with this program; if not, see .
ACLOCAL_AMFLAGS = -I m4
DISTCHECK_CONFIGURE_FLAGS = --enable-doc
# (A suitable gitlog-to-changelog script can be found in GnuPG master.)
GITLOG_TO_CHANGELOG=gitlog-to-changelog
EXTRA_DIST = autogen.sh autogen.rc libgpg-error.spec.in \
COPYING COPYING.LIB potomo doc/HACKING ChangeLog-2011 \
po/ChangeLog-2011 m4/ChangeLog-2011 contrib/ChangeLog-2011 \
build-aux/git-log-footer build-aux/git-log-fix
if LANGUAGES_SOME
lang_subdirs = lang
else
lang_subdirs =
endif
if BUILD_DOC
doc = doc
else
doc =
endif
SUBDIRS = m4 src $(doc) tests po $(lang_subdirs)
dist-hook: gen-ChangeLog
sed -e 's/@pkg_version@/$(VERSION)/g' \
$(top_srcdir)/libgpg-error.spec.in > $(distdir)/libgpg-error.spec
@set -e; echo "$(VERSION)" > $(distdir)/VERSION
gen_start_date = 2011-12-01T00:00:00
.PHONY: gen-ChangeLog
gen-ChangeLog:
set -e; \
if test -d $(top_srcdir)/.git; then \
(cd $(top_srcdir) && \
$(GITLOG_TO_CHANGELOG) --append-dot --tear-off \
--amend=build-aux/git-log-fix --tear-off \
--since=$(gen_start_date) ) > $(distdir)/cl-t; \
cat $(top_srcdir)/build-aux/git-log-footer >> $(distdir)/cl-t;\
rm -f $(distdir)/ChangeLog; \
mv $(distdir)/cl-t $(distdir)/ChangeLog; \
fi
if HAVE_W32_SYSTEM
install-data-hook:
set -e; \
for i in $$($(top_srcdir)/potomo --get-linguas $(top_srcdir)/po); do \
$(MKDIR_P) "$(DESTDIR)$(localedir)/$$i/LC_MESSAGES" || true; \
rm -f "$(DESTDIR)$(localedir)/$$i/LC_MESSAGES/libgpg-error.mo" \
2>/dev/null || true; \
$(top_srcdir)/potomo $(top_srcdir)/po/$$i.po \
"$(DESTDIR)$(localedir)/$$i/LC_MESSAGES/libgpg-error.mo" ; \
done
endif
stowinstall:
$(MAKE) $(AM_MAKEFLAGS) install prefix=/usr/local/stow/libgpg-error
diff --git a/autogen.sh b/autogen.sh
index 24da40c..91f35a6 100755
--- a/autogen.sh
+++ b/autogen.sh
@@ -1,393 +1,393 @@
#! /bin/sh
# autogen.sh
# Copyright (C) 2003, 2014 g10 Code GmbH
#
# This file is free software; as a special exception the author gives
# unlimited permission to copy and/or distribute it, with or without
# modifications, as long as this notice is preserved.
#
# This program is distributed in the hope that it will be useful, but
# WITHOUT ANY WARRANTY, to the extent permitted by law; without even the
# implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
#
# This is a generic script to create the configure script and handle cross
# build environments. It requires the presence of a autogen.rc file to
# configure it for the respective package. It is maintained as part of
# GnuPG and source copied by other packages.
#
# Version: 2014-01-10
configure_ac="configure.ac"
cvtver () {
awk 'NR==1 {split($NF,A,".");X=1000000*A[1]+1000*A[2]+A[3];print X;exit 0}'
}
check_version () {
if [ $(( `("$1" --version || echo "0") | cvtver` >= $2 )) = 1 ]; then
return 0
fi
echo "**Error**: "\`$1\'" not installed or too old." >&2
echo ' Version '$3' or newer is required.' >&2
[ -n "$4" ] && echo ' Note that this is part of '\`$4\''.' >&2
DIE="yes"
return 1
}
fatal () {
echo "autogen.sh:" "$*" >&2
DIE=yes
}
info () {
if [ -z "${SILENT}" ]; then
echo "autogen.sh:" "$*"
fi
}
die_p () {
if [ "$DIE" = "yes" ]; then
echo "autogen.sh: Stop." >&2
exit 1
fi
}
replace_sysroot () {
configure_opts=$(echo $configure_opts | sed "s#@SYSROOT@#${w32root}#g")
extraoptions=$(echo $extraoptions | sed "s#@SYSROOT@#${w32root}#g")
}
# Allow to override the default tool names
AUTOCONF=${AUTOCONF_PREFIX}${AUTOCONF:-autoconf}${AUTOCONF_SUFFIX}
AUTOHEADER=${AUTOCONF_PREFIX}${AUTOHEADER:-autoheader}${AUTOCONF_SUFFIX}
AUTOMAKE=${AUTOMAKE_PREFIX}${AUTOMAKE:-automake}${AUTOMAKE_SUFFIX}
ACLOCAL=${AUTOMAKE_PREFIX}${ACLOCAL:-aclocal}${AUTOMAKE_SUFFIX}
GETTEXT=${GETTEXT_PREFIX}${GETTEXT:-gettext}${GETTEXT_SUFFIX}
MSGMERGE=${GETTEXT_PREFIX}${MSGMERGE:-msgmerge}${GETTEXT_SUFFIX}
DIE=no
FORCE=
SILENT=
tmp=$(dirname "$0")
tsdir=$(cd "${tmp}"; pwd)
if [ -n "${AUTOGEN_SH_SILENT}" ]; then
SILENT=" --silent"
fi
if test x"$1" = x"--help"; then
echo "usage: ./autogen.sh [--silent] [--force] [--build-TYPE] [ARGS]"
exit 0
fi
if test x"$1" = x"--silent"; then
SILENT=" --silent"
shift
fi
if test x"$1" = x"--force"; then
FORCE=" --force"
shift
fi
# Reject unsafe characters in $HOME, $tsdir and cwd. We consider spaces
# as unsafe because it is too easy to get scripts wrong in this regard.
am_lf='
'
case `pwd` in
*[\;\\\"\#\$\&\'\`$am_lf\ \ ]*)
fatal "unsafe working directory name" ;;
esac
case $tsdir in
*[\;\\\"\#\$\&\'\`$am_lf\ \ ]*)
fatal "unsafe source directory: \`$tsdir'" ;;
esac
case $HOME in
*[\;\\\"\#\$\&\'\`$am_lf\ \ ]*)
fatal "unsafe home directory: \`$HOME'" ;;
esac
die_p
# List of variables sourced from autogen.rc. The strings '@SYSROOT@' in
# these variables are replaced by the actual system root.
configure_opts=
extraoptions=
# List of optional variables sourced from autogen.rc and ~/.gnupg-autogen.rc
w32_toolprefixes=
w32_extraoptions=
w32ce_toolprefixes=
w32ce_extraoptions=
w64_toolprefixes=
w64_extraoptions=
amd64_toolprefixes=
# End list of optional variables sourced from ~/.gnupg-autogen.rc
# What follows are variables which are sourced but default to
# environment variables or lacking them hardcoded values.
#w32root=
#w32ce_root=
#w64root=
#amd64root=
# Convenience option to use certain configure options for some hosts.
myhost=""
myhostsub=""
case "$1" in
--build-w32)
myhost="w32"
shift
;;
--build-w32ce)
myhost="w32"
myhostsub="ce"
shift
;;
--build-w64)
myhost="w32"
myhostsub="64"
shift
;;
--build-amd64)
myhost="amd64"
shift
;;
--build*)
fatal "**Error**: invalid build option $1"
shift
;;
*)
;;
esac
die_p
# Source our configuration
if [ -f "${tsdir}/autogen.rc" ]; then
. "${tsdir}/autogen.rc"
fi
# Source optional site specific configuration
if [ -f "$HOME/.gnupg-autogen.rc" ]; then
info "sourcing extra definitions from $HOME/.gnupg-autogen.rc"
. "$HOME/.gnupg-autogen.rc"
fi
# ******************
# W32 build script
# ******************
if [ "$myhost" = "w32" ]; then
if [ ! -f "$tsdir/build-aux/config.guess" ]; then
fatal "$tsdir/build-aux/config.guess not found"
exit 1
fi
build=`$tsdir/build-aux/config.guess`
case $myhostsub in
ce)
w32root="$w32ce_root"
[ -z "$w32root" ] && w32root="$HOME/w32ce_root"
toolprefixes="$w32ce_toolprefixes arm-mingw32ce"
extraoptions="$extraoptions $w32ce_extraoptions"
;;
64)
w32root="$w64root"
[ -z "$w32root" ] && w32root="$HOME/w64root"
toolprefixes="$w64_toolprefixes x86_64-w64-mingw32"
extraoptions="$extraoptions $w64_extraoptions"
;;
*)
[ -z "$w32root" ] && w32root="$HOME/w32root"
toolprefixes="$w32_toolprefixes i686-w64-mingw32 i586-mingw32msvc"
toolprefixes="$toolprefixes i386-mingw32msvc mingw32"
extraoptions="$extraoptions $w32_extraoptions"
;;
esac
info "Using $w32root as standard install directory"
replace_sysroot
# Locate the cross compiler
crossbindir=
for host in $toolprefixes; do
if ${host}-gcc --version >/dev/null 2>&1 ; then
crossbindir=/usr/${host}/bin
conf_CC="CC=${host}-gcc"
break;
fi
done
if [ -z "$crossbindir" ]; then
fatal "cross compiler kit not installed"
if [ -z "$myhostsub" ]; then
info "Under Debian GNU/Linux, you may install it using"
info " apt-get install mingw32 mingw32-runtime mingw32-binutils"
fi
die_p
fi
if [ -f "$tsdir/config.log" ]; then
if ! head $tsdir/config.log | grep "$host" >/dev/null; then
fatal "Please run a 'make distclean' first"
die_p
fi
fi
$tsdir/configure --enable-maintainer-mode ${SILENT} \
--prefix=${w32root} \
--host=${host} --build=${build} \
${configure_opts} ${extraoptions} "$@"
rc=$?
exit $rc
fi
# ***** end W32 build script *******
# ***** AMD64 cross build script *******
# Used to cross-compile for AMD64 (for testing)
if [ "$myhost" = "amd64" ]; then
shift
if [ ! -f $tsdir/build-aux/config.guess ]; then
echo "$tsdir/build-aux/config.guess not found" >&2
exit 1
fi
build=`$tsdir/build-aux/config.guess`
[ -z "$amd64root" ] && amd64root="$HOME/amd64root"
info "Using $amd64root as standard install directory"
replace_sysroot
toolprefixes="$amd64_toolprefixes x86_64-linux-gnu amd64-linux-gnu"
# Locate the cross compiler
crossbindir=
for host in $toolprefixes ; do
if ${host}-gcc --version >/dev/null 2>&1 ; then
crossbindir=/usr/${host}/bin
conf_CC="CC=${host}-gcc"
break;
fi
done
if [ -z "$crossbindir" ]; then
echo "Cross compiler kit not installed" >&2
echo "Stop." >&2
exit 1
fi
if [ -f "$tsdir/config.log" ]; then
if ! head $tsdir/config.log | grep "$host" >/dev/null; then
echo "Please run a 'make distclean' first" >&2
exit 1
fi
fi
$tsdir/configure --enable-maintainer-mode ${SILENT} \
--prefix=${amd64root} \
--host=${host} --build=${build} \
${configure_opts} ${extraoptions} "$@"
rc=$?
exit $rc
fi
# ***** end AMD64 cross build script *******
# Grep the required versions from configure.ac
autoconf_vers=`sed -n '/^AC_PREREQ(/ {
s/^.*(\(.*\))/\1/p
q
}' ${configure_ac}`
autoconf_vers_num=`echo "$autoconf_vers" | cvtver`
automake_vers=`sed -n '/^min_automake_version=/ {
s/^.*="\(.*\)"/\1/p
q
}' ${configure_ac}`
automake_vers_num=`echo "$automake_vers" | cvtver`
if [ -d "${tsdir}/po" ]; then
gettext_vers=`sed -n '/^AM_GNU_GETTEXT_VERSION(/ {
s/^.*\[\(.*\)])/\1/p
q
}' ${configure_ac}`
gettext_vers_num=`echo "$gettext_vers" | cvtver`
else
gettext_vers="n/a"
fi
if [ -z "$autoconf_vers" -o -z "$automake_vers" -o -z "$gettext_vers" ]
then
echo "**Error**: version information not found in "\`${configure_ac}\'"." >&2
exit 1
fi
if check_version $AUTOCONF $autoconf_vers_num $autoconf_vers ; then
check_version $AUTOHEADER $autoconf_vers_num $autoconf_vers autoconf
fi
if check_version $AUTOMAKE $automake_vers_num $automake_vers; then
check_version $ACLOCAL $automake_vers_num $autoconf_vers automake
fi
if [ "$gettext_vers" != "n/a" ]; then
if check_version $GETTEXT $gettext_vers_num $gettext_vers; then
check_version $MSGMERGE $gettext_vers_num $gettext_vers gettext
fi
fi
if [ "$DIE" = "yes" ]; then
cat <.
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2, or (at your option)
# any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
-# along with this program. If not, see .
+# along with this program. If not, see .
# As a special exception to the GNU General Public License, if you
# distribute this file as part of a program that contains a
# configuration script generated by Autoconf, you may include it under
# the same distribution terms that you use for the rest of that program.
# This file is maintained in Automake, please report
# bugs to or send patches to
# .
nl='
'
# We need space, tab and new line, in precisely that order. Quoting is
# there to prevent tools from complaining about whitespace usage.
IFS=" "" $nl"
file_conv=
# func_file_conv build_file lazy
# Convert a $build file to $host form and store it in $file
# Currently only supports Windows hosts. If the determined conversion
# type is listed in (the comma separated) LAZY, no conversion will
# take place.
func_file_conv ()
{
file=$1
case $file in
/ | /[!/]*) # absolute file, and not a UNC file
if test -z "$file_conv"; then
# lazily determine how to convert abs files
case `uname -s` in
MINGW*)
file_conv=mingw
;;
CYGWIN*)
file_conv=cygwin
;;
*)
file_conv=wine
;;
esac
fi
case $file_conv/,$2, in
*,$file_conv,*)
;;
mingw/*)
file=`cmd //C echo "$file " | sed -e 's/"\(.*\) " *$/\1/'`
;;
cygwin/*)
file=`cygpath -m "$file" || echo "$file"`
;;
wine/*)
file=`winepath -w "$file" || echo "$file"`
;;
esac
;;
esac
}
# func_cl_dashL linkdir
# Make cl look for libraries in LINKDIR
func_cl_dashL ()
{
func_file_conv "$1"
if test -z "$lib_path"; then
lib_path=$file
else
lib_path="$lib_path;$file"
fi
linker_opts="$linker_opts -LIBPATH:$file"
}
# func_cl_dashl library
# Do a library search-path lookup for cl
func_cl_dashl ()
{
lib=$1
found=no
save_IFS=$IFS
IFS=';'
for dir in $lib_path $LIB
do
IFS=$save_IFS
if $shared && test -f "$dir/$lib.dll.lib"; then
found=yes
lib=$dir/$lib.dll.lib
break
fi
if test -f "$dir/$lib.lib"; then
found=yes
lib=$dir/$lib.lib
break
fi
if test -f "$dir/lib$lib.a"; then
found=yes
lib=$dir/lib$lib.a
break
fi
done
IFS=$save_IFS
if test "$found" != yes; then
lib=$lib.lib
fi
}
# func_cl_wrapper cl arg...
# Adjust compile command to suit cl
func_cl_wrapper ()
{
# Assume a capable shell
lib_path=
shared=:
linker_opts=
for arg
do
if test -n "$eat"; then
eat=
else
case $1 in
-o)
# configure might choose to run compile as 'compile cc -o foo foo.c'.
eat=1
case $2 in
*.o | *.[oO][bB][jJ])
func_file_conv "$2"
set x "$@" -Fo"$file"
shift
;;
*)
func_file_conv "$2"
set x "$@" -Fe"$file"
shift
;;
esac
;;
-I)
eat=1
func_file_conv "$2" mingw
set x "$@" -I"$file"
shift
;;
-I*)
func_file_conv "${1#-I}" mingw
set x "$@" -I"$file"
shift
;;
-l)
eat=1
func_cl_dashl "$2"
set x "$@" "$lib"
shift
;;
-l*)
func_cl_dashl "${1#-l}"
set x "$@" "$lib"
shift
;;
-L)
eat=1
func_cl_dashL "$2"
;;
-L*)
func_cl_dashL "${1#-L}"
;;
-static)
shared=false
;;
-Wl,*)
arg=${1#-Wl,}
save_ifs="$IFS"; IFS=','
for flag in $arg; do
IFS="$save_ifs"
linker_opts="$linker_opts $flag"
done
IFS="$save_ifs"
;;
-Xlinker)
eat=1
linker_opts="$linker_opts $2"
;;
-*)
set x "$@" "$1"
shift
;;
*.cc | *.CC | *.cxx | *.CXX | *.[cC]++)
func_file_conv "$1"
set x "$@" -Tp"$file"
shift
;;
*.c | *.cpp | *.CPP | *.lib | *.LIB | *.Lib | *.OBJ | *.obj | *.[oO])
func_file_conv "$1" mingw
set x "$@" "$file"
shift
;;
*)
set x "$@" "$1"
shift
;;
esac
fi
shift
done
if test -n "$linker_opts"; then
linker_opts="-link$linker_opts"
fi
exec "$@" $linker_opts
exit 1
}
eat=
case $1 in
'')
echo "$0: No command. Try '$0 --help' for more information." 1>&2
exit 1;
;;
-h | --h*)
cat <<\EOF
Usage: compile [--help] [--version] PROGRAM [ARGS]
Wrapper for compilers which do not understand '-c -o'.
Remove '-o dest.o' from ARGS, run PROGRAM with the remaining
arguments, and rename the output as expected.
If you are trying to build a whole package this is not the
right script to run: please start by reading the file 'INSTALL'.
Report bugs to .
EOF
exit $?
;;
-v | --v*)
echo "compile $scriptversion"
exit $?
;;
cl | *[/\\]cl | cl.exe | *[/\\]cl.exe )
func_cl_wrapper "$@" # Doesn't return...
;;
esac
ofile=
cfile=
for arg
do
if test -n "$eat"; then
eat=
else
case $1 in
-o)
# configure might choose to run compile as 'compile cc -o foo foo.c'.
# So we strip '-o arg' only if arg is an object.
eat=1
case $2 in
*.o | *.obj)
ofile=$2
;;
*)
set x "$@" -o "$2"
shift
;;
esac
;;
*.c)
cfile=$1
set x "$@" "$1"
shift
;;
*)
set x "$@" "$1"
shift
;;
esac
fi
shift
done
if test -z "$ofile" || test -z "$cfile"; then
# If no '-o' option was seen then we might have been invoked from a
# pattern rule where we don't need one. That is ok -- this is a
# normal compilation that the losing compiler can handle. If no
# '.c' file was seen then we are probably linking. That is also
# ok.
exec "$@"
fi
# Name of file we expect compiler to create.
cofile=`echo "$cfile" | sed 's|^.*[\\/]||; s|^[a-zA-Z]:||; s/\.c$/.o/'`
# Create the lock directory.
# Note: use '[/\\:.-]' here to ensure that we don't use the same name
# that we are using for the .o file. Also, base the name on the expected
# object file name, since that is what matters with a parallel build.
lockdir=`echo "$cofile" | sed -e 's|[/\\:.-]|_|g'`.d
while true; do
if mkdir "$lockdir" >/dev/null 2>&1; then
break
fi
sleep 1
done
# FIXME: race condition here if user kills between mkdir and trap.
trap "rmdir '$lockdir'; exit 1" 1 2 15
# Run the compile.
"$@"
ret=$?
if test -f "$cofile"; then
test "$cofile" = "$ofile" || mv "$cofile" "$ofile"
elif test -f "${cofile}bj"; then
test "${cofile}bj" = "$ofile" || mv "${cofile}bj" "$ofile"
fi
rmdir "$lockdir"
exit $ret
# Local Variables:
# mode: shell-script
# sh-indentation: 2
# eval: (add-hook 'write-file-hooks 'time-stamp)
# time-stamp-start: "scriptversion="
# time-stamp-format: "%:y-%02m-%02d.%02H"
# time-stamp-time-zone: "UTC"
# time-stamp-end: "; # UTC"
# End:
diff --git a/build-aux/config.guess b/build-aux/config.guess
index dbfb978..7adf147 100755
--- a/build-aux/config.guess
+++ b/build-aux/config.guess
@@ -1,1421 +1,1421 @@
#! /bin/sh
# Attempt to guess a canonical system name.
# Copyright 1992-2015 Free Software Foundation, Inc.
timestamp='2015-01-01'
# This file is free software; you can redistribute it and/or modify it
# under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful, but
# WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
# General Public License for more details.
#
# You should have received a copy of the GNU General Public License
-# along with this program; if not, see .
+# along with this program; if not, see .
#
# As a special exception to the GNU General Public License, if you
# distribute this file as part of a program that contains a
# configuration script generated by Autoconf, you may include it under
# the same distribution terms that you use for the rest of that
# program. This Exception is an additional permission under section 7
# of the GNU General Public License, version 3 ("GPLv3").
#
# Originally written by Per Bothner; maintained since 2000 by Ben Elliston.
#
# You can get the latest version of this script from:
# http://git.savannah.gnu.org/gitweb/?p=config.git;a=blob_plain;f=config.guess;hb=HEAD
#
# Please send patches to .
me=`echo "$0" | sed -e 's,.*/,,'`
usage="\
Usage: $0 [OPTION]
Output the configuration name of the system \`$me' is run on.
Operation modes:
-h, --help print this help, then exit
-t, --time-stamp print date of last modification, then exit
-v, --version print version number, then exit
Report bugs and patches to ."
version="\
GNU config.guess ($timestamp)
Originally written by Per Bothner.
Copyright 1992-2015 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE."
help="
Try \`$me --help' for more information."
# Parse command line
while test $# -gt 0 ; do
case $1 in
--time-stamp | --time* | -t )
echo "$timestamp" ; exit ;;
--version | -v )
echo "$version" ; exit ;;
--help | --h* | -h )
echo "$usage"; exit ;;
-- ) # Stop option processing
shift; break ;;
- ) # Use stdin as input.
break ;;
-* )
echo "$me: invalid option $1$help" >&2
exit 1 ;;
* )
break ;;
esac
done
if test $# != 0; then
echo "$me: too many arguments$help" >&2
exit 1
fi
trap 'exit 1' 1 2 15
# CC_FOR_BUILD -- compiler used by this script. Note that the use of a
# compiler to aid in system detection is discouraged as it requires
# temporary files to be created and, as you can see below, it is a
# headache to deal with in a portable fashion.
# Historically, `CC_FOR_BUILD' used to be named `HOST_CC'. We still
# use `HOST_CC' if defined, but it is deprecated.
# Portable tmp directory creation inspired by the Autoconf team.
set_cc_for_build='
trap "exitcode=\$?; (rm -f \$tmpfiles 2>/dev/null; rmdir \$tmp 2>/dev/null) && exit \$exitcode" 0 ;
trap "rm -f \$tmpfiles 2>/dev/null; rmdir \$tmp 2>/dev/null; exit 1" 1 2 13 15 ;
: ${TMPDIR=/tmp} ;
{ tmp=`(umask 077 && mktemp -d "$TMPDIR/cgXXXXXX") 2>/dev/null` && test -n "$tmp" && test -d "$tmp" ; } ||
{ test -n "$RANDOM" && tmp=$TMPDIR/cg$$-$RANDOM && (umask 077 && mkdir $tmp) ; } ||
{ tmp=$TMPDIR/cg-$$ && (umask 077 && mkdir $tmp) && echo "Warning: creating insecure temp directory" >&2 ; } ||
{ echo "$me: cannot create a temporary directory in $TMPDIR" >&2 ; exit 1 ; } ;
dummy=$tmp/dummy ;
tmpfiles="$dummy.c $dummy.o $dummy.rel $dummy" ;
case $CC_FOR_BUILD,$HOST_CC,$CC in
,,) echo "int x;" > $dummy.c ;
for c in cc gcc c89 c99 ; do
if ($c -c -o $dummy.o $dummy.c) >/dev/null 2>&1 ; then
CC_FOR_BUILD="$c"; break ;
fi ;
done ;
if test x"$CC_FOR_BUILD" = x ; then
CC_FOR_BUILD=no_compiler_found ;
fi
;;
,,*) CC_FOR_BUILD=$CC ;;
,*,*) CC_FOR_BUILD=$HOST_CC ;;
esac ; set_cc_for_build= ;'
# This is needed to find uname on a Pyramid OSx when run in the BSD universe.
# (ghazi@noc.rutgers.edu 1994-08-24)
if (test -f /.attbin/uname) >/dev/null 2>&1 ; then
PATH=$PATH:/.attbin ; export PATH
fi
UNAME_MACHINE=`(uname -m) 2>/dev/null` || UNAME_MACHINE=unknown
UNAME_RELEASE=`(uname -r) 2>/dev/null` || UNAME_RELEASE=unknown
UNAME_SYSTEM=`(uname -s) 2>/dev/null` || UNAME_SYSTEM=unknown
UNAME_VERSION=`(uname -v) 2>/dev/null` || UNAME_VERSION=unknown
case "${UNAME_SYSTEM}" in
Linux|GNU|GNU/*)
# If the system lacks a compiler, then just pick glibc.
# We could probably try harder.
LIBC=gnu
eval $set_cc_for_build
cat <<-EOF > $dummy.c
#include
#if defined(__UCLIBC__)
LIBC=uclibc
#elif defined(__dietlibc__)
LIBC=dietlibc
#else
LIBC=gnu
#endif
EOF
eval `$CC_FOR_BUILD -E $dummy.c 2>/dev/null | grep '^LIBC' | sed 's, ,,g'`
;;
esac
# Note: order is significant - the case branches are not exclusive.
case "${UNAME_MACHINE}:${UNAME_SYSTEM}:${UNAME_RELEASE}:${UNAME_VERSION}" in
*:NetBSD:*:*)
# NetBSD (nbsd) targets should (where applicable) match one or
# more of the tuples: *-*-netbsdelf*, *-*-netbsdaout*,
# *-*-netbsdecoff* and *-*-netbsd*. For targets that recently
# switched to ELF, *-*-netbsd* would select the old
# object file format. This provides both forward
# compatibility and a consistent mechanism for selecting the
# object file format.
#
# Note: NetBSD doesn't particularly care about the vendor
# portion of the name. We always set it to "unknown".
sysctl="sysctl -n hw.machine_arch"
UNAME_MACHINE_ARCH=`(/sbin/$sysctl 2>/dev/null || \
/usr/sbin/$sysctl 2>/dev/null || echo unknown)`
case "${UNAME_MACHINE_ARCH}" in
armeb) machine=armeb-unknown ;;
arm*) machine=arm-unknown ;;
sh3el) machine=shl-unknown ;;
sh3eb) machine=sh-unknown ;;
sh5el) machine=sh5le-unknown ;;
*) machine=${UNAME_MACHINE_ARCH}-unknown ;;
esac
# The Operating System including object format, if it has switched
# to ELF recently, or will in the future.
case "${UNAME_MACHINE_ARCH}" in
arm*|i386|m68k|ns32k|sh3*|sparc|vax)
eval $set_cc_for_build
if echo __ELF__ | $CC_FOR_BUILD -E - 2>/dev/null \
| grep -q __ELF__
then
# Once all utilities can be ECOFF (netbsdecoff) or a.out (netbsdaout).
# Return netbsd for either. FIX?
os=netbsd
else
os=netbsdelf
fi
;;
*)
os=netbsd
;;
esac
# The OS release
# Debian GNU/NetBSD machines have a different userland, and
# thus, need a distinct triplet. However, they do not need
# kernel version information, so it can be replaced with a
# suitable tag, in the style of linux-gnu.
case "${UNAME_VERSION}" in
Debian*)
release='-gnu'
;;
*)
release=`echo ${UNAME_RELEASE}|sed -e 's/[-_].*/\./'`
;;
esac
# Since CPU_TYPE-MANUFACTURER-KERNEL-OPERATING_SYSTEM:
# contains redundant information, the shorter form:
# CPU_TYPE-MANUFACTURER-OPERATING_SYSTEM is used.
echo "${machine}-${os}${release}"
exit ;;
*:Bitrig:*:*)
UNAME_MACHINE_ARCH=`arch | sed 's/Bitrig.//'`
echo ${UNAME_MACHINE_ARCH}-unknown-bitrig${UNAME_RELEASE}
exit ;;
*:OpenBSD:*:*)
UNAME_MACHINE_ARCH=`arch | sed 's/OpenBSD.//'`
echo ${UNAME_MACHINE_ARCH}-unknown-openbsd${UNAME_RELEASE}
exit ;;
*:ekkoBSD:*:*)
echo ${UNAME_MACHINE}-unknown-ekkobsd${UNAME_RELEASE}
exit ;;
*:SolidBSD:*:*)
echo ${UNAME_MACHINE}-unknown-solidbsd${UNAME_RELEASE}
exit ;;
macppc:MirBSD:*:*)
echo powerpc-unknown-mirbsd${UNAME_RELEASE}
exit ;;
*:MirBSD:*:*)
echo ${UNAME_MACHINE}-unknown-mirbsd${UNAME_RELEASE}
exit ;;
alpha:OSF1:*:*)
case $UNAME_RELEASE in
*4.0)
UNAME_RELEASE=`/usr/sbin/sizer -v | awk '{print $3}'`
;;
*5.*)
UNAME_RELEASE=`/usr/sbin/sizer -v | awk '{print $4}'`
;;
esac
# According to Compaq, /usr/sbin/psrinfo has been available on
# OSF/1 and Tru64 systems produced since 1995. I hope that
# covers most systems running today. This code pipes the CPU
# types through head -n 1, so we only detect the type of CPU 0.
ALPHA_CPU_TYPE=`/usr/sbin/psrinfo -v | sed -n -e 's/^ The alpha \(.*\) processor.*$/\1/p' | head -n 1`
case "$ALPHA_CPU_TYPE" in
"EV4 (21064)")
UNAME_MACHINE="alpha" ;;
"EV4.5 (21064)")
UNAME_MACHINE="alpha" ;;
"LCA4 (21066/21068)")
UNAME_MACHINE="alpha" ;;
"EV5 (21164)")
UNAME_MACHINE="alphaev5" ;;
"EV5.6 (21164A)")
UNAME_MACHINE="alphaev56" ;;
"EV5.6 (21164PC)")
UNAME_MACHINE="alphapca56" ;;
"EV5.7 (21164PC)")
UNAME_MACHINE="alphapca57" ;;
"EV6 (21264)")
UNAME_MACHINE="alphaev6" ;;
"EV6.7 (21264A)")
UNAME_MACHINE="alphaev67" ;;
"EV6.8CB (21264C)")
UNAME_MACHINE="alphaev68" ;;
"EV6.8AL (21264B)")
UNAME_MACHINE="alphaev68" ;;
"EV6.8CX (21264D)")
UNAME_MACHINE="alphaev68" ;;
"EV6.9A (21264/EV69A)")
UNAME_MACHINE="alphaev69" ;;
"EV7 (21364)")
UNAME_MACHINE="alphaev7" ;;
"EV7.9 (21364A)")
UNAME_MACHINE="alphaev79" ;;
esac
# A Pn.n version is a patched version.
# A Vn.n version is a released version.
# A Tn.n version is a released field test version.
# A Xn.n version is an unreleased experimental baselevel.
# 1.2 uses "1.2" for uname -r.
echo ${UNAME_MACHINE}-dec-osf`echo ${UNAME_RELEASE} | sed -e 's/^[PVTX]//' | tr 'ABCDEFGHIJKLMNOPQRSTUVWXYZ' 'abcdefghijklmnopqrstuvwxyz'`
# Reset EXIT trap before exiting to avoid spurious non-zero exit code.
exitcode=$?
trap '' 0
exit $exitcode ;;
Alpha\ *:Windows_NT*:*)
# How do we know it's Interix rather than the generic POSIX subsystem?
# Should we change UNAME_MACHINE based on the output of uname instead
# of the specific Alpha model?
echo alpha-pc-interix
exit ;;
21064:Windows_NT:50:3)
echo alpha-dec-winnt3.5
exit ;;
Amiga*:UNIX_System_V:4.0:*)
echo m68k-unknown-sysv4
exit ;;
*:[Aa]miga[Oo][Ss]:*:*)
echo ${UNAME_MACHINE}-unknown-amigaos
exit ;;
*:[Mm]orph[Oo][Ss]:*:*)
echo ${UNAME_MACHINE}-unknown-morphos
exit ;;
*:OS/390:*:*)
echo i370-ibm-openedition
exit ;;
*:z/VM:*:*)
echo s390-ibm-zvmoe
exit ;;
*:OS400:*:*)
echo powerpc-ibm-os400
exit ;;
arm:RISC*:1.[012]*:*|arm:riscix:1.[012]*:*)
echo arm-acorn-riscix${UNAME_RELEASE}
exit ;;
arm*:riscos:*:*|arm*:RISCOS:*:*)
echo arm-unknown-riscos
exit ;;
SR2?01:HI-UX/MPP:*:* | SR8000:HI-UX/MPP:*:*)
echo hppa1.1-hitachi-hiuxmpp
exit ;;
Pyramid*:OSx*:*:* | MIS*:OSx*:*:* | MIS*:SMP_DC-OSx*:*:*)
# akee@wpdis03.wpafb.af.mil (Earle F. Ake) contributed MIS and NILE.
if test "`(/bin/universe) 2>/dev/null`" = att ; then
echo pyramid-pyramid-sysv3
else
echo pyramid-pyramid-bsd
fi
exit ;;
NILE*:*:*:dcosx)
echo pyramid-pyramid-svr4
exit ;;
DRS?6000:unix:4.0:6*)
echo sparc-icl-nx6
exit ;;
DRS?6000:UNIX_SV:4.2*:7* | DRS?6000:isis:4.2*:7*)
case `/usr/bin/uname -p` in
sparc) echo sparc-icl-nx7; exit ;;
esac ;;
s390x:SunOS:*:*)
echo ${UNAME_MACHINE}-ibm-solaris2`echo ${UNAME_RELEASE}|sed -e 's/[^.]*//'`
exit ;;
sun4H:SunOS:5.*:*)
echo sparc-hal-solaris2`echo ${UNAME_RELEASE}|sed -e 's/[^.]*//'`
exit ;;
sun4*:SunOS:5.*:* | tadpole*:SunOS:5.*:*)
echo sparc-sun-solaris2`echo ${UNAME_RELEASE}|sed -e 's/[^.]*//'`
exit ;;
i86pc:AuroraUX:5.*:* | i86xen:AuroraUX:5.*:*)
echo i386-pc-auroraux${UNAME_RELEASE}
exit ;;
i86pc:SunOS:5.*:* | i86xen:SunOS:5.*:*)
eval $set_cc_for_build
SUN_ARCH="i386"
# If there is a compiler, see if it is configured for 64-bit objects.
# Note that the Sun cc does not turn __LP64__ into 1 like gcc does.
# This test works for both compilers.
if [ "$CC_FOR_BUILD" != 'no_compiler_found' ]; then
if (echo '#ifdef __amd64'; echo IS_64BIT_ARCH; echo '#endif') | \
(CCOPTS= $CC_FOR_BUILD -E - 2>/dev/null) | \
grep IS_64BIT_ARCH >/dev/null
then
SUN_ARCH="x86_64"
fi
fi
echo ${SUN_ARCH}-pc-solaris2`echo ${UNAME_RELEASE}|sed -e 's/[^.]*//'`
exit ;;
sun4*:SunOS:6*:*)
# According to config.sub, this is the proper way to canonicalize
# SunOS6. Hard to guess exactly what SunOS6 will be like, but
# it's likely to be more like Solaris than SunOS4.
echo sparc-sun-solaris3`echo ${UNAME_RELEASE}|sed -e 's/[^.]*//'`
exit ;;
sun4*:SunOS:*:*)
case "`/usr/bin/arch -k`" in
Series*|S4*)
UNAME_RELEASE=`uname -v`
;;
esac
# Japanese Language versions have a version number like `4.1.3-JL'.
echo sparc-sun-sunos`echo ${UNAME_RELEASE}|sed -e 's/-/_/'`
exit ;;
sun3*:SunOS:*:*)
echo m68k-sun-sunos${UNAME_RELEASE}
exit ;;
sun*:*:4.2BSD:*)
UNAME_RELEASE=`(sed 1q /etc/motd | awk '{print substr($5,1,3)}') 2>/dev/null`
test "x${UNAME_RELEASE}" = "x" && UNAME_RELEASE=3
case "`/bin/arch`" in
sun3)
echo m68k-sun-sunos${UNAME_RELEASE}
;;
sun4)
echo sparc-sun-sunos${UNAME_RELEASE}
;;
esac
exit ;;
aushp:SunOS:*:*)
echo sparc-auspex-sunos${UNAME_RELEASE}
exit ;;
# The situation for MiNT is a little confusing. The machine name
# can be virtually everything (everything which is not
# "atarist" or "atariste" at least should have a processor
# > m68000). The system name ranges from "MiNT" over "FreeMiNT"
# to the lowercase version "mint" (or "freemint"). Finally
# the system name "TOS" denotes a system which is actually not
# MiNT. But MiNT is downward compatible to TOS, so this should
# be no problem.
atarist[e]:*MiNT:*:* | atarist[e]:*mint:*:* | atarist[e]:*TOS:*:*)
echo m68k-atari-mint${UNAME_RELEASE}
exit ;;
atari*:*MiNT:*:* | atari*:*mint:*:* | atarist[e]:*TOS:*:*)
echo m68k-atari-mint${UNAME_RELEASE}
exit ;;
*falcon*:*MiNT:*:* | *falcon*:*mint:*:* | *falcon*:*TOS:*:*)
echo m68k-atari-mint${UNAME_RELEASE}
exit ;;
milan*:*MiNT:*:* | milan*:*mint:*:* | *milan*:*TOS:*:*)
echo m68k-milan-mint${UNAME_RELEASE}
exit ;;
hades*:*MiNT:*:* | hades*:*mint:*:* | *hades*:*TOS:*:*)
echo m68k-hades-mint${UNAME_RELEASE}
exit ;;
*:*MiNT:*:* | *:*mint:*:* | *:*TOS:*:*)
echo m68k-unknown-mint${UNAME_RELEASE}
exit ;;
m68k:machten:*:*)
echo m68k-apple-machten${UNAME_RELEASE}
exit ;;
powerpc:machten:*:*)
echo powerpc-apple-machten${UNAME_RELEASE}
exit ;;
RISC*:Mach:*:*)
echo mips-dec-mach_bsd4.3
exit ;;
RISC*:ULTRIX:*:*)
echo mips-dec-ultrix${UNAME_RELEASE}
exit ;;
VAX*:ULTRIX*:*:*)
echo vax-dec-ultrix${UNAME_RELEASE}
exit ;;
2020:CLIX:*:* | 2430:CLIX:*:*)
echo clipper-intergraph-clix${UNAME_RELEASE}
exit ;;
mips:*:*:UMIPS | mips:*:*:RISCos)
eval $set_cc_for_build
sed 's/^ //' << EOF >$dummy.c
#ifdef __cplusplus
#include /* for printf() prototype */
int main (int argc, char *argv[]) {
#else
int main (argc, argv) int argc; char *argv[]; {
#endif
#if defined (host_mips) && defined (MIPSEB)
#if defined (SYSTYPE_SYSV)
printf ("mips-mips-riscos%ssysv\n", argv[1]); exit (0);
#endif
#if defined (SYSTYPE_SVR4)
printf ("mips-mips-riscos%ssvr4\n", argv[1]); exit (0);
#endif
#if defined (SYSTYPE_BSD43) || defined(SYSTYPE_BSD)
printf ("mips-mips-riscos%sbsd\n", argv[1]); exit (0);
#endif
#endif
exit (-1);
}
EOF
$CC_FOR_BUILD -o $dummy $dummy.c &&
dummyarg=`echo "${UNAME_RELEASE}" | sed -n 's/\([0-9]*\).*/\1/p'` &&
SYSTEM_NAME=`$dummy $dummyarg` &&
{ echo "$SYSTEM_NAME"; exit; }
echo mips-mips-riscos${UNAME_RELEASE}
exit ;;
Motorola:PowerMAX_OS:*:*)
echo powerpc-motorola-powermax
exit ;;
Motorola:*:4.3:PL8-*)
echo powerpc-harris-powermax
exit ;;
Night_Hawk:*:*:PowerMAX_OS | Synergy:PowerMAX_OS:*:*)
echo powerpc-harris-powermax
exit ;;
Night_Hawk:Power_UNIX:*:*)
echo powerpc-harris-powerunix
exit ;;
m88k:CX/UX:7*:*)
echo m88k-harris-cxux7
exit ;;
m88k:*:4*:R4*)
echo m88k-motorola-sysv4
exit ;;
m88k:*:3*:R3*)
echo m88k-motorola-sysv3
exit ;;
AViiON:dgux:*:*)
# DG/UX returns AViiON for all architectures
UNAME_PROCESSOR=`/usr/bin/uname -p`
if [ $UNAME_PROCESSOR = mc88100 ] || [ $UNAME_PROCESSOR = mc88110 ]
then
if [ ${TARGET_BINARY_INTERFACE}x = m88kdguxelfx ] || \
[ ${TARGET_BINARY_INTERFACE}x = x ]
then
echo m88k-dg-dgux${UNAME_RELEASE}
else
echo m88k-dg-dguxbcs${UNAME_RELEASE}
fi
else
echo i586-dg-dgux${UNAME_RELEASE}
fi
exit ;;
M88*:DolphinOS:*:*) # DolphinOS (SVR3)
echo m88k-dolphin-sysv3
exit ;;
M88*:*:R3*:*)
# Delta 88k system running SVR3
echo m88k-motorola-sysv3
exit ;;
XD88*:*:*:*) # Tektronix XD88 system running UTekV (SVR3)
echo m88k-tektronix-sysv3
exit ;;
Tek43[0-9][0-9]:UTek:*:*) # Tektronix 4300 system running UTek (BSD)
echo m68k-tektronix-bsd
exit ;;
*:IRIX*:*:*)
echo mips-sgi-irix`echo ${UNAME_RELEASE}|sed -e 's/-/_/g'`
exit ;;
????????:AIX?:[12].1:2) # AIX 2.2.1 or AIX 2.1.1 is RT/PC AIX.
echo romp-ibm-aix # uname -m gives an 8 hex-code CPU id
exit ;; # Note that: echo "'`uname -s`'" gives 'AIX '
i*86:AIX:*:*)
echo i386-ibm-aix
exit ;;
ia64:AIX:*:*)
if [ -x /usr/bin/oslevel ] ; then
IBM_REV=`/usr/bin/oslevel`
else
IBM_REV=${UNAME_VERSION}.${UNAME_RELEASE}
fi
echo ${UNAME_MACHINE}-ibm-aix${IBM_REV}
exit ;;
*:AIX:2:3)
if grep bos325 /usr/include/stdio.h >/dev/null 2>&1; then
eval $set_cc_for_build
sed 's/^ //' << EOF >$dummy.c
#include
main()
{
if (!__power_pc())
exit(1);
puts("powerpc-ibm-aix3.2.5");
exit(0);
}
EOF
if $CC_FOR_BUILD -o $dummy $dummy.c && SYSTEM_NAME=`$dummy`
then
echo "$SYSTEM_NAME"
else
echo rs6000-ibm-aix3.2.5
fi
elif grep bos324 /usr/include/stdio.h >/dev/null 2>&1; then
echo rs6000-ibm-aix3.2.4
else
echo rs6000-ibm-aix3.2
fi
exit ;;
*:AIX:*:[4567])
IBM_CPU_ID=`/usr/sbin/lsdev -C -c processor -S available | sed 1q | awk '{ print $1 }'`
if /usr/sbin/lsattr -El ${IBM_CPU_ID} | grep ' POWER' >/dev/null 2>&1; then
IBM_ARCH=rs6000
else
IBM_ARCH=powerpc
fi
if [ -x /usr/bin/lslpp ] ; then
IBM_REV=`/usr/bin/lslpp -Lqc bos.rte.libc |
awk -F: '{ print $3 }' | sed s/[0-9]*$/0/`
else
IBM_REV=${UNAME_VERSION}.${UNAME_RELEASE}
fi
echo ${IBM_ARCH}-ibm-aix${IBM_REV}
exit ;;
*:AIX:*:*)
echo rs6000-ibm-aix
exit ;;
ibmrt:4.4BSD:*|romp-ibm:BSD:*)
echo romp-ibm-bsd4.4
exit ;;
ibmrt:*BSD:*|romp-ibm:BSD:*) # covers RT/PC BSD and
echo romp-ibm-bsd${UNAME_RELEASE} # 4.3 with uname added to
exit ;; # report: romp-ibm BSD 4.3
*:BOSX:*:*)
echo rs6000-bull-bosx
exit ;;
DPX/2?00:B.O.S.:*:*)
echo m68k-bull-sysv3
exit ;;
9000/[34]??:4.3bsd:1.*:*)
echo m68k-hp-bsd
exit ;;
hp300:4.4BSD:*:* | 9000/[34]??:4.3bsd:2.*:*)
echo m68k-hp-bsd4.4
exit ;;
9000/[34678]??:HP-UX:*:*)
HPUX_REV=`echo ${UNAME_RELEASE}|sed -e 's/[^.]*.[0B]*//'`
case "${UNAME_MACHINE}" in
9000/31? ) HP_ARCH=m68000 ;;
9000/[34]?? ) HP_ARCH=m68k ;;
9000/[678][0-9][0-9])
if [ -x /usr/bin/getconf ]; then
sc_cpu_version=`/usr/bin/getconf SC_CPU_VERSION 2>/dev/null`
sc_kernel_bits=`/usr/bin/getconf SC_KERNEL_BITS 2>/dev/null`
case "${sc_cpu_version}" in
523) HP_ARCH="hppa1.0" ;; # CPU_PA_RISC1_0
528) HP_ARCH="hppa1.1" ;; # CPU_PA_RISC1_1
532) # CPU_PA_RISC2_0
case "${sc_kernel_bits}" in
32) HP_ARCH="hppa2.0n" ;;
64) HP_ARCH="hppa2.0w" ;;
'') HP_ARCH="hppa2.0" ;; # HP-UX 10.20
esac ;;
esac
fi
if [ "${HP_ARCH}" = "" ]; then
eval $set_cc_for_build
sed 's/^ //' << EOF >$dummy.c
#define _HPUX_SOURCE
#include
#include
int main ()
{
#if defined(_SC_KERNEL_BITS)
long bits = sysconf(_SC_KERNEL_BITS);
#endif
long cpu = sysconf (_SC_CPU_VERSION);
switch (cpu)
{
case CPU_PA_RISC1_0: puts ("hppa1.0"); break;
case CPU_PA_RISC1_1: puts ("hppa1.1"); break;
case CPU_PA_RISC2_0:
#if defined(_SC_KERNEL_BITS)
switch (bits)
{
case 64: puts ("hppa2.0w"); break;
case 32: puts ("hppa2.0n"); break;
default: puts ("hppa2.0"); break;
} break;
#else /* !defined(_SC_KERNEL_BITS) */
puts ("hppa2.0"); break;
#endif
default: puts ("hppa1.0"); break;
}
exit (0);
}
EOF
(CCOPTS= $CC_FOR_BUILD -o $dummy $dummy.c 2>/dev/null) && HP_ARCH=`$dummy`
test -z "$HP_ARCH" && HP_ARCH=hppa
fi ;;
esac
if [ ${HP_ARCH} = "hppa2.0w" ]
then
eval $set_cc_for_build
# hppa2.0w-hp-hpux* has a 64-bit kernel and a compiler generating
# 32-bit code. hppa64-hp-hpux* has the same kernel and a compiler
# generating 64-bit code. GNU and HP use different nomenclature:
#
# $ CC_FOR_BUILD=cc ./config.guess
# => hppa2.0w-hp-hpux11.23
# $ CC_FOR_BUILD="cc +DA2.0w" ./config.guess
# => hppa64-hp-hpux11.23
if echo __LP64__ | (CCOPTS= $CC_FOR_BUILD -E - 2>/dev/null) |
grep -q __LP64__
then
HP_ARCH="hppa2.0w"
else
HP_ARCH="hppa64"
fi
fi
echo ${HP_ARCH}-hp-hpux${HPUX_REV}
exit ;;
ia64:HP-UX:*:*)
HPUX_REV=`echo ${UNAME_RELEASE}|sed -e 's/[^.]*.[0B]*//'`
echo ia64-hp-hpux${HPUX_REV}
exit ;;
3050*:HI-UX:*:*)
eval $set_cc_for_build
sed 's/^ //' << EOF >$dummy.c
#include
int
main ()
{
long cpu = sysconf (_SC_CPU_VERSION);
/* The order matters, because CPU_IS_HP_MC68K erroneously returns
true for CPU_PA_RISC1_0. CPU_IS_PA_RISC returns correct
results, however. */
if (CPU_IS_PA_RISC (cpu))
{
switch (cpu)
{
case CPU_PA_RISC1_0: puts ("hppa1.0-hitachi-hiuxwe2"); break;
case CPU_PA_RISC1_1: puts ("hppa1.1-hitachi-hiuxwe2"); break;
case CPU_PA_RISC2_0: puts ("hppa2.0-hitachi-hiuxwe2"); break;
default: puts ("hppa-hitachi-hiuxwe2"); break;
}
}
else if (CPU_IS_HP_MC68K (cpu))
puts ("m68k-hitachi-hiuxwe2");
else puts ("unknown-hitachi-hiuxwe2");
exit (0);
}
EOF
$CC_FOR_BUILD -o $dummy $dummy.c && SYSTEM_NAME=`$dummy` &&
{ echo "$SYSTEM_NAME"; exit; }
echo unknown-hitachi-hiuxwe2
exit ;;
9000/7??:4.3bsd:*:* | 9000/8?[79]:4.3bsd:*:* )
echo hppa1.1-hp-bsd
exit ;;
9000/8??:4.3bsd:*:*)
echo hppa1.0-hp-bsd
exit ;;
*9??*:MPE/iX:*:* | *3000*:MPE/iX:*:*)
echo hppa1.0-hp-mpeix
exit ;;
hp7??:OSF1:*:* | hp8?[79]:OSF1:*:* )
echo hppa1.1-hp-osf
exit ;;
hp8??:OSF1:*:*)
echo hppa1.0-hp-osf
exit ;;
i*86:OSF1:*:*)
if [ -x /usr/sbin/sysversion ] ; then
echo ${UNAME_MACHINE}-unknown-osf1mk
else
echo ${UNAME_MACHINE}-unknown-osf1
fi
exit ;;
parisc*:Lites*:*:*)
echo hppa1.1-hp-lites
exit ;;
C1*:ConvexOS:*:* | convex:ConvexOS:C1*:*)
echo c1-convex-bsd
exit ;;
C2*:ConvexOS:*:* | convex:ConvexOS:C2*:*)
if getsysinfo -f scalar_acc
then echo c32-convex-bsd
else echo c2-convex-bsd
fi
exit ;;
C34*:ConvexOS:*:* | convex:ConvexOS:C34*:*)
echo c34-convex-bsd
exit ;;
C38*:ConvexOS:*:* | convex:ConvexOS:C38*:*)
echo c38-convex-bsd
exit ;;
C4*:ConvexOS:*:* | convex:ConvexOS:C4*:*)
echo c4-convex-bsd
exit ;;
CRAY*Y-MP:*:*:*)
echo ymp-cray-unicos${UNAME_RELEASE} | sed -e 's/\.[^.]*$/.X/'
exit ;;
CRAY*[A-Z]90:*:*:*)
echo ${UNAME_MACHINE}-cray-unicos${UNAME_RELEASE} \
| sed -e 's/CRAY.*\([A-Z]90\)/\1/' \
-e y/ABCDEFGHIJKLMNOPQRSTUVWXYZ/abcdefghijklmnopqrstuvwxyz/ \
-e 's/\.[^.]*$/.X/'
exit ;;
CRAY*TS:*:*:*)
echo t90-cray-unicos${UNAME_RELEASE} | sed -e 's/\.[^.]*$/.X/'
exit ;;
CRAY*T3E:*:*:*)
echo alphaev5-cray-unicosmk${UNAME_RELEASE} | sed -e 's/\.[^.]*$/.X/'
exit ;;
CRAY*SV1:*:*:*)
echo sv1-cray-unicos${UNAME_RELEASE} | sed -e 's/\.[^.]*$/.X/'
exit ;;
*:UNICOS/mp:*:*)
echo craynv-cray-unicosmp${UNAME_RELEASE} | sed -e 's/\.[^.]*$/.X/'
exit ;;
F30[01]:UNIX_System_V:*:* | F700:UNIX_System_V:*:*)
FUJITSU_PROC=`uname -m | tr 'ABCDEFGHIJKLMNOPQRSTUVWXYZ' 'abcdefghijklmnopqrstuvwxyz'`
FUJITSU_SYS=`uname -p | tr 'ABCDEFGHIJKLMNOPQRSTUVWXYZ' 'abcdefghijklmnopqrstuvwxyz' | sed -e 's/\///'`
FUJITSU_REL=`echo ${UNAME_RELEASE} | sed -e 's/ /_/'`
echo "${FUJITSU_PROC}-fujitsu-${FUJITSU_SYS}${FUJITSU_REL}"
exit ;;
5000:UNIX_System_V:4.*:*)
FUJITSU_SYS=`uname -p | tr 'ABCDEFGHIJKLMNOPQRSTUVWXYZ' 'abcdefghijklmnopqrstuvwxyz' | sed -e 's/\///'`
FUJITSU_REL=`echo ${UNAME_RELEASE} | tr 'ABCDEFGHIJKLMNOPQRSTUVWXYZ' 'abcdefghijklmnopqrstuvwxyz' | sed -e 's/ /_/'`
echo "sparc-fujitsu-${FUJITSU_SYS}${FUJITSU_REL}"
exit ;;
i*86:BSD/386:*:* | i*86:BSD/OS:*:* | *:Ascend\ Embedded/OS:*:*)
echo ${UNAME_MACHINE}-pc-bsdi${UNAME_RELEASE}
exit ;;
sparc*:BSD/OS:*:*)
echo sparc-unknown-bsdi${UNAME_RELEASE}
exit ;;
*:BSD/OS:*:*)
echo ${UNAME_MACHINE}-unknown-bsdi${UNAME_RELEASE}
exit ;;
*:FreeBSD:*:*)
UNAME_PROCESSOR=`/usr/bin/uname -p`
case ${UNAME_PROCESSOR} in
amd64)
echo x86_64-unknown-freebsd`echo ${UNAME_RELEASE}|sed -e 's/[-(].*//'` ;;
*)
echo ${UNAME_PROCESSOR}-unknown-freebsd`echo ${UNAME_RELEASE}|sed -e 's/[-(].*//'` ;;
esac
exit ;;
i*:CYGWIN*:*)
echo ${UNAME_MACHINE}-pc-cygwin
exit ;;
*:MINGW64*:*)
echo ${UNAME_MACHINE}-pc-mingw64
exit ;;
*:MINGW*:*)
echo ${UNAME_MACHINE}-pc-mingw32
exit ;;
*:MSYS*:*)
echo ${UNAME_MACHINE}-pc-msys
exit ;;
i*:windows32*:*)
# uname -m includes "-pc" on this system.
echo ${UNAME_MACHINE}-mingw32
exit ;;
i*:PW*:*)
echo ${UNAME_MACHINE}-pc-pw32
exit ;;
*:Interix*:*)
case ${UNAME_MACHINE} in
x86)
echo i586-pc-interix${UNAME_RELEASE}
exit ;;
authenticamd | genuineintel | EM64T)
echo x86_64-unknown-interix${UNAME_RELEASE}
exit ;;
IA64)
echo ia64-unknown-interix${UNAME_RELEASE}
exit ;;
esac ;;
[345]86:Windows_95:* | [345]86:Windows_98:* | [345]86:Windows_NT:*)
echo i${UNAME_MACHINE}-pc-mks
exit ;;
8664:Windows_NT:*)
echo x86_64-pc-mks
exit ;;
i*:Windows_NT*:* | Pentium*:Windows_NT*:*)
# How do we know it's Interix rather than the generic POSIX subsystem?
# It also conflicts with pre-2.0 versions of AT&T UWIN. Should we
# UNAME_MACHINE based on the output of uname instead of i386?
echo i586-pc-interix
exit ;;
i*:UWIN*:*)
echo ${UNAME_MACHINE}-pc-uwin
exit ;;
amd64:CYGWIN*:*:* | x86_64:CYGWIN*:*:*)
echo x86_64-unknown-cygwin
exit ;;
p*:CYGWIN*:*)
echo powerpcle-unknown-cygwin
exit ;;
prep*:SunOS:5.*:*)
echo powerpcle-unknown-solaris2`echo ${UNAME_RELEASE}|sed -e 's/[^.]*//'`
exit ;;
*:GNU:*:*)
# the GNU system
echo `echo ${UNAME_MACHINE}|sed -e 's,[-/].*$,,'`-unknown-${LIBC}`echo ${UNAME_RELEASE}|sed -e 's,/.*$,,'`
exit ;;
*:GNU/*:*:*)
# other systems with GNU libc and userland
echo ${UNAME_MACHINE}-unknown-`echo ${UNAME_SYSTEM} | sed 's,^[^/]*/,,' | tr '[A-Z]' '[a-z]'``echo ${UNAME_RELEASE}|sed -e 's/[-(].*//'`-${LIBC}
exit ;;
i*86:Minix:*:*)
echo ${UNAME_MACHINE}-pc-minix
exit ;;
aarch64:Linux:*:*)
echo ${UNAME_MACHINE}-unknown-linux-${LIBC}
exit ;;
aarch64_be:Linux:*:*)
UNAME_MACHINE=aarch64_be
echo ${UNAME_MACHINE}-unknown-linux-${LIBC}
exit ;;
alpha:Linux:*:*)
case `sed -n '/^cpu model/s/^.*: \(.*\)/\1/p' < /proc/cpuinfo` in
EV5) UNAME_MACHINE=alphaev5 ;;
EV56) UNAME_MACHINE=alphaev56 ;;
PCA56) UNAME_MACHINE=alphapca56 ;;
PCA57) UNAME_MACHINE=alphapca56 ;;
EV6) UNAME_MACHINE=alphaev6 ;;
EV67) UNAME_MACHINE=alphaev67 ;;
EV68*) UNAME_MACHINE=alphaev68 ;;
esac
objdump --private-headers /bin/sh | grep -q ld.so.1
if test "$?" = 0 ; then LIBC="gnulibc1" ; fi
echo ${UNAME_MACHINE}-unknown-linux-${LIBC}
exit ;;
arc:Linux:*:* | arceb:Linux:*:*)
echo ${UNAME_MACHINE}-unknown-linux-${LIBC}
exit ;;
arm*:Linux:*:*)
eval $set_cc_for_build
if echo __ARM_EABI__ | $CC_FOR_BUILD -E - 2>/dev/null \
| grep -q __ARM_EABI__
then
echo ${UNAME_MACHINE}-unknown-linux-${LIBC}
else
if echo __ARM_PCS_VFP | $CC_FOR_BUILD -E - 2>/dev/null \
| grep -q __ARM_PCS_VFP
then
echo ${UNAME_MACHINE}-unknown-linux-${LIBC}eabi
else
echo ${UNAME_MACHINE}-unknown-linux-${LIBC}eabihf
fi
fi
exit ;;
avr32*:Linux:*:*)
echo ${UNAME_MACHINE}-unknown-linux-${LIBC}
exit ;;
cris:Linux:*:*)
echo ${UNAME_MACHINE}-axis-linux-${LIBC}
exit ;;
crisv32:Linux:*:*)
echo ${UNAME_MACHINE}-axis-linux-${LIBC}
exit ;;
frv:Linux:*:*)
echo ${UNAME_MACHINE}-unknown-linux-${LIBC}
exit ;;
hexagon:Linux:*:*)
echo ${UNAME_MACHINE}-unknown-linux-${LIBC}
exit ;;
i*86:Linux:*:*)
echo ${UNAME_MACHINE}-pc-linux-${LIBC}
exit ;;
ia64:Linux:*:*)
echo ${UNAME_MACHINE}-unknown-linux-${LIBC}
exit ;;
m32r*:Linux:*:*)
echo ${UNAME_MACHINE}-unknown-linux-${LIBC}
exit ;;
m68*:Linux:*:*)
echo ${UNAME_MACHINE}-unknown-linux-${LIBC}
exit ;;
mips:Linux:*:* | mips64:Linux:*:*)
eval $set_cc_for_build
sed 's/^ //' << EOF >$dummy.c
#undef CPU
#undef ${UNAME_MACHINE}
#undef ${UNAME_MACHINE}el
#if defined(__MIPSEL__) || defined(__MIPSEL) || defined(_MIPSEL) || defined(MIPSEL)
CPU=${UNAME_MACHINE}el
#else
#if defined(__MIPSEB__) || defined(__MIPSEB) || defined(_MIPSEB) || defined(MIPSEB)
CPU=${UNAME_MACHINE}
#else
CPU=
#endif
#endif
EOF
eval `$CC_FOR_BUILD -E $dummy.c 2>/dev/null | grep '^CPU'`
test x"${CPU}" != x && { echo "${CPU}-unknown-linux-${LIBC}"; exit; }
;;
openrisc*:Linux:*:*)
echo or1k-unknown-linux-${LIBC}
exit ;;
or32:Linux:*:* | or1k*:Linux:*:*)
echo ${UNAME_MACHINE}-unknown-linux-${LIBC}
exit ;;
padre:Linux:*:*)
echo sparc-unknown-linux-${LIBC}
exit ;;
parisc64:Linux:*:* | hppa64:Linux:*:*)
echo hppa64-unknown-linux-${LIBC}
exit ;;
parisc:Linux:*:* | hppa:Linux:*:*)
# Look for CPU level
case `grep '^cpu[^a-z]*:' /proc/cpuinfo 2>/dev/null | cut -d' ' -f2` in
PA7*) echo hppa1.1-unknown-linux-${LIBC} ;;
PA8*) echo hppa2.0-unknown-linux-${LIBC} ;;
*) echo hppa-unknown-linux-${LIBC} ;;
esac
exit ;;
ppc64:Linux:*:*)
echo powerpc64-unknown-linux-${LIBC}
exit ;;
ppc:Linux:*:*)
echo powerpc-unknown-linux-${LIBC}
exit ;;
ppc64le:Linux:*:*)
echo powerpc64le-unknown-linux-${LIBC}
exit ;;
ppcle:Linux:*:*)
echo powerpcle-unknown-linux-${LIBC}
exit ;;
s390:Linux:*:* | s390x:Linux:*:*)
echo ${UNAME_MACHINE}-ibm-linux-${LIBC}
exit ;;
sh64*:Linux:*:*)
echo ${UNAME_MACHINE}-unknown-linux-${LIBC}
exit ;;
sh*:Linux:*:*)
echo ${UNAME_MACHINE}-unknown-linux-${LIBC}
exit ;;
sparc:Linux:*:* | sparc64:Linux:*:*)
echo ${UNAME_MACHINE}-unknown-linux-${LIBC}
exit ;;
tile*:Linux:*:*)
echo ${UNAME_MACHINE}-unknown-linux-${LIBC}
exit ;;
vax:Linux:*:*)
echo ${UNAME_MACHINE}-dec-linux-${LIBC}
exit ;;
x86_64:Linux:*:*)
echo ${UNAME_MACHINE}-unknown-linux-${LIBC}
exit ;;
xtensa*:Linux:*:*)
echo ${UNAME_MACHINE}-unknown-linux-${LIBC}
exit ;;
i*86:DYNIX/ptx:4*:*)
# ptx 4.0 does uname -s correctly, with DYNIX/ptx in there.
# earlier versions are messed up and put the nodename in both
# sysname and nodename.
echo i386-sequent-sysv4
exit ;;
i*86:UNIX_SV:4.2MP:2.*)
# Unixware is an offshoot of SVR4, but it has its own version
# number series starting with 2...
# I am not positive that other SVR4 systems won't match this,
# I just have to hope. -- rms.
# Use sysv4.2uw... so that sysv4* matches it.
echo ${UNAME_MACHINE}-pc-sysv4.2uw${UNAME_VERSION}
exit ;;
i*86:OS/2:*:*)
# If we were able to find `uname', then EMX Unix compatibility
# is probably installed.
echo ${UNAME_MACHINE}-pc-os2-emx
exit ;;
i*86:XTS-300:*:STOP)
echo ${UNAME_MACHINE}-unknown-stop
exit ;;
i*86:atheos:*:*)
echo ${UNAME_MACHINE}-unknown-atheos
exit ;;
i*86:syllable:*:*)
echo ${UNAME_MACHINE}-pc-syllable
exit ;;
i*86:LynxOS:2.*:* | i*86:LynxOS:3.[01]*:* | i*86:LynxOS:4.[02]*:*)
echo i386-unknown-lynxos${UNAME_RELEASE}
exit ;;
i*86:*DOS:*:*)
echo ${UNAME_MACHINE}-pc-msdosdjgpp
exit ;;
i*86:*:4.*:* | i*86:SYSTEM_V:4.*:*)
UNAME_REL=`echo ${UNAME_RELEASE} | sed 's/\/MP$//'`
if grep Novell /usr/include/link.h >/dev/null 2>/dev/null; then
echo ${UNAME_MACHINE}-univel-sysv${UNAME_REL}
else
echo ${UNAME_MACHINE}-pc-sysv${UNAME_REL}
fi
exit ;;
i*86:*:5:[678]*)
# UnixWare 7.x, OpenUNIX and OpenServer 6.
case `/bin/uname -X | grep "^Machine"` in
*486*) UNAME_MACHINE=i486 ;;
*Pentium) UNAME_MACHINE=i586 ;;
*Pent*|*Celeron) UNAME_MACHINE=i686 ;;
esac
echo ${UNAME_MACHINE}-unknown-sysv${UNAME_RELEASE}${UNAME_SYSTEM}${UNAME_VERSION}
exit ;;
i*86:*:3.2:*)
if test -f /usr/options/cb.name; then
UNAME_REL=`sed -n 's/.*Version //p' /dev/null >/dev/null ; then
UNAME_REL=`(/bin/uname -X|grep Release|sed -e 's/.*= //')`
(/bin/uname -X|grep i80486 >/dev/null) && UNAME_MACHINE=i486
(/bin/uname -X|grep '^Machine.*Pentium' >/dev/null) \
&& UNAME_MACHINE=i586
(/bin/uname -X|grep '^Machine.*Pent *II' >/dev/null) \
&& UNAME_MACHINE=i686
(/bin/uname -X|grep '^Machine.*Pentium Pro' >/dev/null) \
&& UNAME_MACHINE=i686
echo ${UNAME_MACHINE}-pc-sco$UNAME_REL
else
echo ${UNAME_MACHINE}-pc-sysv32
fi
exit ;;
pc:*:*:*)
# Left here for compatibility:
# uname -m prints for DJGPP always 'pc', but it prints nothing about
# the processor, so we play safe by assuming i586.
# Note: whatever this is, it MUST be the same as what config.sub
# prints for the "djgpp" host, or else GDB configury will decide that
# this is a cross-build.
echo i586-pc-msdosdjgpp
exit ;;
Intel:Mach:3*:*)
echo i386-pc-mach3
exit ;;
paragon:*:*:*)
echo i860-intel-osf1
exit ;;
i860:*:4.*:*) # i860-SVR4
if grep Stardent /usr/include/sys/uadmin.h >/dev/null 2>&1 ; then
echo i860-stardent-sysv${UNAME_RELEASE} # Stardent Vistra i860-SVR4
else # Add other i860-SVR4 vendors below as they are discovered.
echo i860-unknown-sysv${UNAME_RELEASE} # Unknown i860-SVR4
fi
exit ;;
mini*:CTIX:SYS*5:*)
# "miniframe"
echo m68010-convergent-sysv
exit ;;
mc68k:UNIX:SYSTEM5:3.51m)
echo m68k-convergent-sysv
exit ;;
M680?0:D-NIX:5.3:*)
echo m68k-diab-dnix
exit ;;
M68*:*:R3V[5678]*:*)
test -r /sysV68 && { echo 'm68k-motorola-sysv'; exit; } ;;
3[345]??:*:4.0:3.0 | 3[34]??A:*:4.0:3.0 | 3[34]??,*:*:4.0:3.0 | 3[34]??/*:*:4.0:3.0 | 4400:*:4.0:3.0 | 4850:*:4.0:3.0 | SKA40:*:4.0:3.0 | SDS2:*:4.0:3.0 | SHG2:*:4.0:3.0 | S7501*:*:4.0:3.0)
OS_REL=''
test -r /etc/.relid \
&& OS_REL=.`sed -n 's/[^ ]* [^ ]* \([0-9][0-9]\).*/\1/p' < /etc/.relid`
/bin/uname -p 2>/dev/null | grep 86 >/dev/null \
&& { echo i486-ncr-sysv4.3${OS_REL}; exit; }
/bin/uname -p 2>/dev/null | /bin/grep entium >/dev/null \
&& { echo i586-ncr-sysv4.3${OS_REL}; exit; } ;;
3[34]??:*:4.0:* | 3[34]??,*:*:4.0:*)
/bin/uname -p 2>/dev/null | grep 86 >/dev/null \
&& { echo i486-ncr-sysv4; exit; } ;;
NCR*:*:4.2:* | MPRAS*:*:4.2:*)
OS_REL='.3'
test -r /etc/.relid \
&& OS_REL=.`sed -n 's/[^ ]* [^ ]* \([0-9][0-9]\).*/\1/p' < /etc/.relid`
/bin/uname -p 2>/dev/null | grep 86 >/dev/null \
&& { echo i486-ncr-sysv4.3${OS_REL}; exit; }
/bin/uname -p 2>/dev/null | /bin/grep entium >/dev/null \
&& { echo i586-ncr-sysv4.3${OS_REL}; exit; }
/bin/uname -p 2>/dev/null | /bin/grep pteron >/dev/null \
&& { echo i586-ncr-sysv4.3${OS_REL}; exit; } ;;
m68*:LynxOS:2.*:* | m68*:LynxOS:3.0*:*)
echo m68k-unknown-lynxos${UNAME_RELEASE}
exit ;;
mc68030:UNIX_System_V:4.*:*)
echo m68k-atari-sysv4
exit ;;
TSUNAMI:LynxOS:2.*:*)
echo sparc-unknown-lynxos${UNAME_RELEASE}
exit ;;
rs6000:LynxOS:2.*:*)
echo rs6000-unknown-lynxos${UNAME_RELEASE}
exit ;;
PowerPC:LynxOS:2.*:* | PowerPC:LynxOS:3.[01]*:* | PowerPC:LynxOS:4.[02]*:*)
echo powerpc-unknown-lynxos${UNAME_RELEASE}
exit ;;
SM[BE]S:UNIX_SV:*:*)
echo mips-dde-sysv${UNAME_RELEASE}
exit ;;
RM*:ReliantUNIX-*:*:*)
echo mips-sni-sysv4
exit ;;
RM*:SINIX-*:*:*)
echo mips-sni-sysv4
exit ;;
*:SINIX-*:*:*)
if uname -p 2>/dev/null >/dev/null ; then
UNAME_MACHINE=`(uname -p) 2>/dev/null`
echo ${UNAME_MACHINE}-sni-sysv4
else
echo ns32k-sni-sysv
fi
exit ;;
PENTIUM:*:4.0*:*) # Unisys `ClearPath HMP IX 4000' SVR4/MP effort
# says
echo i586-unisys-sysv4
exit ;;
*:UNIX_System_V:4*:FTX*)
# From Gerald Hewes .
# How about differentiating between stratus architectures? -djm
echo hppa1.1-stratus-sysv4
exit ;;
*:*:*:FTX*)
# From seanf@swdc.stratus.com.
echo i860-stratus-sysv4
exit ;;
i*86:VOS:*:*)
# From Paul.Green@stratus.com.
echo ${UNAME_MACHINE}-stratus-vos
exit ;;
*:VOS:*:*)
# From Paul.Green@stratus.com.
echo hppa1.1-stratus-vos
exit ;;
mc68*:A/UX:*:*)
echo m68k-apple-aux${UNAME_RELEASE}
exit ;;
news*:NEWS-OS:6*:*)
echo mips-sony-newsos6
exit ;;
R[34]000:*System_V*:*:* | R4000:UNIX_SYSV:*:* | R*000:UNIX_SV:*:*)
if [ -d /usr/nec ]; then
echo mips-nec-sysv${UNAME_RELEASE}
else
echo mips-unknown-sysv${UNAME_RELEASE}
fi
exit ;;
BeBox:BeOS:*:*) # BeOS running on hardware made by Be, PPC only.
echo powerpc-be-beos
exit ;;
BeMac:BeOS:*:*) # BeOS running on Mac or Mac clone, PPC only.
echo powerpc-apple-beos
exit ;;
BePC:BeOS:*:*) # BeOS running on Intel PC compatible.
echo i586-pc-beos
exit ;;
BePC:Haiku:*:*) # Haiku running on Intel PC compatible.
echo i586-pc-haiku
exit ;;
x86_64:Haiku:*:*)
echo x86_64-unknown-haiku
exit ;;
SX-4:SUPER-UX:*:*)
echo sx4-nec-superux${UNAME_RELEASE}
exit ;;
SX-5:SUPER-UX:*:*)
echo sx5-nec-superux${UNAME_RELEASE}
exit ;;
SX-6:SUPER-UX:*:*)
echo sx6-nec-superux${UNAME_RELEASE}
exit ;;
SX-7:SUPER-UX:*:*)
echo sx7-nec-superux${UNAME_RELEASE}
exit ;;
SX-8:SUPER-UX:*:*)
echo sx8-nec-superux${UNAME_RELEASE}
exit ;;
SX-8R:SUPER-UX:*:*)
echo sx8r-nec-superux${UNAME_RELEASE}
exit ;;
Power*:Rhapsody:*:*)
echo powerpc-apple-rhapsody${UNAME_RELEASE}
exit ;;
*:Rhapsody:*:*)
echo ${UNAME_MACHINE}-apple-rhapsody${UNAME_RELEASE}
exit ;;
*:Darwin:*:*)
UNAME_PROCESSOR=`uname -p` || UNAME_PROCESSOR=unknown
eval $set_cc_for_build
if test "$UNAME_PROCESSOR" = unknown ; then
UNAME_PROCESSOR=powerpc
fi
if test `echo "$UNAME_RELEASE" | sed -e 's/\..*//'` -le 10 ; then
if [ "$CC_FOR_BUILD" != 'no_compiler_found' ]; then
if (echo '#ifdef __LP64__'; echo IS_64BIT_ARCH; echo '#endif') | \
(CCOPTS= $CC_FOR_BUILD -E - 2>/dev/null) | \
grep IS_64BIT_ARCH >/dev/null
then
case $UNAME_PROCESSOR in
i386) UNAME_PROCESSOR=x86_64 ;;
powerpc) UNAME_PROCESSOR=powerpc64 ;;
esac
fi
fi
elif test "$UNAME_PROCESSOR" = i386 ; then
# Avoid executing cc on OS X 10.9, as it ships with a stub
# that puts up a graphical alert prompting to install
# developer tools. Any system running Mac OS X 10.7 or
# later (Darwin 11 and later) is required to have a 64-bit
# processor. This is not true of the ARM version of Darwin
# that Apple uses in portable devices.
UNAME_PROCESSOR=x86_64
fi
echo ${UNAME_PROCESSOR}-apple-darwin${UNAME_RELEASE}
exit ;;
*:procnto*:*:* | *:QNX:[0123456789]*:*)
UNAME_PROCESSOR=`uname -p`
if test "$UNAME_PROCESSOR" = "x86"; then
UNAME_PROCESSOR=i386
UNAME_MACHINE=pc
fi
echo ${UNAME_PROCESSOR}-${UNAME_MACHINE}-nto-qnx${UNAME_RELEASE}
exit ;;
*:QNX:*:4*)
echo i386-pc-qnx
exit ;;
NEO-?:NONSTOP_KERNEL:*:*)
echo neo-tandem-nsk${UNAME_RELEASE}
exit ;;
NSE-*:NONSTOP_KERNEL:*:*)
echo nse-tandem-nsk${UNAME_RELEASE}
exit ;;
NSR-?:NONSTOP_KERNEL:*:*)
echo nsr-tandem-nsk${UNAME_RELEASE}
exit ;;
*:NonStop-UX:*:*)
echo mips-compaq-nonstopux
exit ;;
BS2000:POSIX*:*:*)
echo bs2000-siemens-sysv
exit ;;
DS/*:UNIX_System_V:*:*)
echo ${UNAME_MACHINE}-${UNAME_SYSTEM}-${UNAME_RELEASE}
exit ;;
*:Plan9:*:*)
# "uname -m" is not consistent, so use $cputype instead. 386
# is converted to i386 for consistency with other x86
# operating systems.
if test "$cputype" = "386"; then
UNAME_MACHINE=i386
else
UNAME_MACHINE="$cputype"
fi
echo ${UNAME_MACHINE}-unknown-plan9
exit ;;
*:TOPS-10:*:*)
echo pdp10-unknown-tops10
exit ;;
*:TENEX:*:*)
echo pdp10-unknown-tenex
exit ;;
KS10:TOPS-20:*:* | KL10:TOPS-20:*:* | TYPE4:TOPS-20:*:*)
echo pdp10-dec-tops20
exit ;;
XKL-1:TOPS-20:*:* | TYPE5:TOPS-20:*:*)
echo pdp10-xkl-tops20
exit ;;
*:TOPS-20:*:*)
echo pdp10-unknown-tops20
exit ;;
*:ITS:*:*)
echo pdp10-unknown-its
exit ;;
SEI:*:*:SEIUX)
echo mips-sei-seiux${UNAME_RELEASE}
exit ;;
*:DragonFly:*:*)
echo ${UNAME_MACHINE}-unknown-dragonfly`echo ${UNAME_RELEASE}|sed -e 's/[-(].*//'`
exit ;;
*:*VMS:*:*)
UNAME_MACHINE=`(uname -p) 2>/dev/null`
case "${UNAME_MACHINE}" in
A*) echo alpha-dec-vms ; exit ;;
I*) echo ia64-dec-vms ; exit ;;
V*) echo vax-dec-vms ; exit ;;
esac ;;
*:XENIX:*:SysV)
echo i386-pc-xenix
exit ;;
i*86:skyos:*:*)
echo ${UNAME_MACHINE}-pc-skyos`echo ${UNAME_RELEASE}` | sed -e 's/ .*$//'
exit ;;
i*86:rdos:*:*)
echo ${UNAME_MACHINE}-pc-rdos
exit ;;
i*86:AROS:*:*)
echo ${UNAME_MACHINE}-pc-aros
exit ;;
x86_64:VMkernel:*:*)
echo ${UNAME_MACHINE}-unknown-esx
exit ;;
esac
cat >&2 < in order to provide the needed
information to handle your system.
config.guess timestamp = $timestamp
uname -m = `(uname -m) 2>/dev/null || echo unknown`
uname -r = `(uname -r) 2>/dev/null || echo unknown`
uname -s = `(uname -s) 2>/dev/null || echo unknown`
uname -v = `(uname -v) 2>/dev/null || echo unknown`
/usr/bin/uname -p = `(/usr/bin/uname -p) 2>/dev/null`
/bin/uname -X = `(/bin/uname -X) 2>/dev/null`
hostinfo = `(hostinfo) 2>/dev/null`
/bin/universe = `(/bin/universe) 2>/dev/null`
/usr/bin/arch -k = `(/usr/bin/arch -k) 2>/dev/null`
/bin/arch = `(/bin/arch) 2>/dev/null`
/usr/bin/oslevel = `(/usr/bin/oslevel) 2>/dev/null`
/usr/convex/getsysinfo = `(/usr/convex/getsysinfo) 2>/dev/null`
UNAME_MACHINE = ${UNAME_MACHINE}
UNAME_RELEASE = ${UNAME_RELEASE}
UNAME_SYSTEM = ${UNAME_SYSTEM}
UNAME_VERSION = ${UNAME_VERSION}
EOF
exit 1
# Local variables:
# eval: (add-hook 'write-file-hooks 'time-stamp)
# time-stamp-start: "timestamp='"
# time-stamp-format: "%:y-%02m-%02d"
# time-stamp-end: "'"
# End:
diff --git a/build-aux/config.sub b/build-aux/config.sub
index 6d2e94c..0b2d816 100755
--- a/build-aux/config.sub
+++ b/build-aux/config.sub
@@ -1,1807 +1,1807 @@
#! /bin/sh
# Configuration validation subroutine script.
# Copyright 1992-2015 Free Software Foundation, Inc.
timestamp='2015-01-01'
# This file is free software; you can redistribute it and/or modify it
# under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful, but
# WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
# General Public License for more details.
#
# You should have received a copy of the GNU General Public License
-# along with this program; if not, see .
+# along with this program; if not, see .
#
# As a special exception to the GNU General Public License, if you
# distribute this file as part of a program that contains a
# configuration script generated by Autoconf, you may include it under
# the same distribution terms that you use for the rest of that
# program. This Exception is an additional permission under section 7
# of the GNU General Public License, version 3 ("GPLv3").
# Please send patches to .
#
# Configuration subroutine to validate and canonicalize a configuration type.
# Supply the specified configuration type as an argument.
# If it is invalid, we print an error message on stderr and exit with code 1.
# Otherwise, we print the canonical config type on stdout and succeed.
# You can get the latest version of this script from:
# http://git.savannah.gnu.org/gitweb/?p=config.git;a=blob_plain;f=config.sub;hb=HEAD
# This file is supposed to be the same for all GNU packages
# and recognize all the CPU types, system types and aliases
# that are meaningful with *any* GNU software.
# Each package is responsible for reporting which valid configurations
# it does not support. The user should be able to distinguish
# a failure to support a valid configuration from a meaningless
# configuration.
# The goal of this file is to map all the various variations of a given
# machine specification into a single specification in the form:
# CPU_TYPE-MANUFACTURER-OPERATING_SYSTEM
# or in some cases, the newer four-part form:
# CPU_TYPE-MANUFACTURER-KERNEL-OPERATING_SYSTEM
# It is wrong to echo any other type of specification.
me=`echo "$0" | sed -e 's,.*/,,'`
usage="\
Usage: $0 [OPTION] CPU-MFR-OPSYS
$0 [OPTION] ALIAS
Canonicalize a configuration name.
Operation modes:
-h, --help print this help, then exit
-t, --time-stamp print date of last modification, then exit
-v, --version print version number, then exit
Report bugs and patches to ."
version="\
GNU config.sub ($timestamp)
Copyright 1992-2015 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE."
help="
Try \`$me --help' for more information."
# Parse command line
while test $# -gt 0 ; do
case $1 in
--time-stamp | --time* | -t )
echo "$timestamp" ; exit ;;
--version | -v )
echo "$version" ; exit ;;
--help | --h* | -h )
echo "$usage"; exit ;;
-- ) # Stop option processing
shift; break ;;
- ) # Use stdin as input.
break ;;
-* )
echo "$me: invalid option $1$help"
exit 1 ;;
*local*)
# First pass through any local machine types.
echo $1
exit ;;
* )
break ;;
esac
done
case $# in
0) echo "$me: missing argument$help" >&2
exit 1;;
1) ;;
*) echo "$me: too many arguments$help" >&2
exit 1;;
esac
# Separate what the user gave into CPU-COMPANY and OS or KERNEL-OS (if any).
# Here we must recognize all the valid KERNEL-OS combinations.
maybe_os=`echo $1 | sed 's/^\(.*\)-\([^-]*-[^-]*\)$/\2/'`
case $maybe_os in
nto-qnx* | linux-gnu* | linux-android* | linux-dietlibc | linux-newlib* | \
linux-musl* | linux-uclibc* | uclinux-uclibc* | uclinux-gnu* | kfreebsd*-gnu* | \
knetbsd*-gnu* | netbsd*-gnu* | \
kopensolaris*-gnu* | \
storm-chaos* | os2-emx* | rtmk-nova*)
os=-$maybe_os
basic_machine=`echo $1 | sed 's/^\(.*\)-\([^-]*-[^-]*\)$/\1/'`
;;
android-linux)
os=-linux-android
basic_machine=`echo $1 | sed 's/^\(.*\)-\([^-]*-[^-]*\)$/\1/'`-unknown
;;
*)
basic_machine=`echo $1 | sed 's/-[^-]*$//'`
if [ $basic_machine != $1 ]
then os=`echo $1 | sed 's/.*-/-/'`
else os=; fi
;;
esac
### Let's recognize common machines as not being operating systems so
### that things like config.sub decstation-3100 work. We also
### recognize some manufacturers as not being operating systems, so we
### can provide default operating systems below.
case $os in
-sun*os*)
# Prevent following clause from handling this invalid input.
;;
-dec* | -mips* | -sequent* | -encore* | -pc532* | -sgi* | -sony* | \
-att* | -7300* | -3300* | -delta* | -motorola* | -sun[234]* | \
-unicom* | -ibm* | -next | -hp | -isi* | -apollo | -altos* | \
-convergent* | -ncr* | -news | -32* | -3600* | -3100* | -hitachi* |\
-c[123]* | -convex* | -sun | -crds | -omron* | -dg | -ultra | -tti* | \
-harris | -dolphin | -highlevel | -gould | -cbm | -ns | -masscomp | \
-apple | -axis | -knuth | -cray | -microblaze*)
os=
basic_machine=$1
;;
-bluegene*)
os=-cnk
;;
-sim | -cisco | -oki | -wec | -winbond)
os=
basic_machine=$1
;;
-scout)
;;
-wrs)
os=-vxworks
basic_machine=$1
;;
-chorusos*)
os=-chorusos
basic_machine=$1
;;
-chorusrdb)
os=-chorusrdb
basic_machine=$1
;;
-hiux*)
os=-hiuxwe2
;;
-sco6)
os=-sco5v6
basic_machine=`echo $1 | sed -e 's/86-.*/86-pc/'`
;;
-sco5)
os=-sco3.2v5
basic_machine=`echo $1 | sed -e 's/86-.*/86-pc/'`
;;
-sco4)
os=-sco3.2v4
basic_machine=`echo $1 | sed -e 's/86-.*/86-pc/'`
;;
-sco3.2.[4-9]*)
os=`echo $os | sed -e 's/sco3.2./sco3.2v/'`
basic_machine=`echo $1 | sed -e 's/86-.*/86-pc/'`
;;
-sco3.2v[4-9]*)
# Don't forget version if it is 3.2v4 or newer.
basic_machine=`echo $1 | sed -e 's/86-.*/86-pc/'`
;;
-sco5v6*)
# Don't forget version if it is 3.2v4 or newer.
basic_machine=`echo $1 | sed -e 's/86-.*/86-pc/'`
;;
-sco*)
os=-sco3.2v2
basic_machine=`echo $1 | sed -e 's/86-.*/86-pc/'`
;;
-udk*)
basic_machine=`echo $1 | sed -e 's/86-.*/86-pc/'`
;;
-isc)
os=-isc2.2
basic_machine=`echo $1 | sed -e 's/86-.*/86-pc/'`
;;
-clix*)
basic_machine=clipper-intergraph
;;
-isc*)
basic_machine=`echo $1 | sed -e 's/86-.*/86-pc/'`
;;
-lynx*178)
os=-lynxos178
;;
-lynx*5)
os=-lynxos5
;;
-lynx*)
os=-lynxos
;;
-ptx*)
basic_machine=`echo $1 | sed -e 's/86-.*/86-sequent/'`
;;
-windowsnt*)
os=`echo $os | sed -e 's/windowsnt/winnt/'`
;;
-psos*)
os=-psos
;;
-mint | -mint[0-9]*)
basic_machine=m68k-atari
os=-mint
;;
esac
# Decode aliases for certain CPU-COMPANY combinations.
case $basic_machine in
# Recognize the basic CPU types without company name.
# Some are omitted here because they have special meanings below.
1750a | 580 \
| a29k \
| aarch64 | aarch64_be \
| alpha | alphaev[4-8] | alphaev56 | alphaev6[78] | alphapca5[67] \
| alpha64 | alpha64ev[4-8] | alpha64ev56 | alpha64ev6[78] | alpha64pca5[67] \
| am33_2.0 \
| arc | arceb \
| arm | arm[bl]e | arme[lb] | armv[2-8] | armv[3-8][lb] | armv7[arm] \
| avr | avr32 \
| be32 | be64 \
| bfin \
| c4x | c8051 | clipper \
| d10v | d30v | dlx | dsp16xx \
| epiphany \
| fido | fr30 | frv | ft32 \
| h8300 | h8500 | hppa | hppa1.[01] | hppa2.0 | hppa2.0[nw] | hppa64 \
| hexagon \
| i370 | i860 | i960 | ia64 \
| ip2k | iq2000 \
| k1om \
| le32 | le64 \
| lm32 \
| m32c | m32r | m32rle | m68000 | m68k | m88k \
| maxq | mb | microblaze | microblazeel | mcore | mep | metag \
| mips | mipsbe | mipseb | mipsel | mipsle \
| mips16 \
| mips64 | mips64el \
| mips64octeon | mips64octeonel \
| mips64orion | mips64orionel \
| mips64r5900 | mips64r5900el \
| mips64vr | mips64vrel \
| mips64vr4100 | mips64vr4100el \
| mips64vr4300 | mips64vr4300el \
| mips64vr5000 | mips64vr5000el \
| mips64vr5900 | mips64vr5900el \
| mipsisa32 | mipsisa32el \
| mipsisa32r2 | mipsisa32r2el \
| mipsisa32r6 | mipsisa32r6el \
| mipsisa64 | mipsisa64el \
| mipsisa64r2 | mipsisa64r2el \
| mipsisa64r6 | mipsisa64r6el \
| mipsisa64sb1 | mipsisa64sb1el \
| mipsisa64sr71k | mipsisa64sr71kel \
| mipsr5900 | mipsr5900el \
| mipstx39 | mipstx39el \
| mn10200 | mn10300 \
| moxie \
| mt \
| msp430 \
| nds32 | nds32le | nds32be \
| nios | nios2 | nios2eb | nios2el \
| ns16k | ns32k \
| open8 | or1k | or1knd | or32 \
| pdp10 | pdp11 | pj | pjl \
| powerpc | powerpc64 | powerpc64le | powerpcle \
| pyramid \
| riscv32 | riscv64 \
| rl78 | rx \
| score \
| sh | sh[1234] | sh[24]a | sh[24]aeb | sh[23]e | sh[34]eb | sheb | shbe | shle | sh[1234]le | sh3ele \
| sh64 | sh64le \
| sparc | sparc64 | sparc64b | sparc64v | sparc86x | sparclet | sparclite \
| sparcv8 | sparcv9 | sparcv9b | sparcv9v \
| spu \
| tahoe | tic4x | tic54x | tic55x | tic6x | tic80 | tron \
| ubicom32 \
| v850 | v850e | v850e1 | v850e2 | v850es | v850e2v3 \
| visium \
| we32k \
| x86 | xc16x | xstormy16 | xtensa \
| z8k | z80)
basic_machine=$basic_machine-unknown
;;
c54x)
basic_machine=tic54x-unknown
;;
c55x)
basic_machine=tic55x-unknown
;;
c6x)
basic_machine=tic6x-unknown
;;
leon|leon[3-9])
basic_machine=sparc-$basic_machine
;;
m6811 | m68hc11 | m6812 | m68hc12 | m68hcs12x | nvptx | picochip)
basic_machine=$basic_machine-unknown
os=-none
;;
m88110 | m680[12346]0 | m683?2 | m68360 | m5200 | v70 | w65 | z8k)
;;
ms1)
basic_machine=mt-unknown
;;
strongarm | thumb | xscale)
basic_machine=arm-unknown
;;
xgate)
basic_machine=$basic_machine-unknown
os=-none
;;
xscaleeb)
basic_machine=armeb-unknown
;;
xscaleel)
basic_machine=armel-unknown
;;
# We use `pc' rather than `unknown'
# because (1) that's what they normally are, and
# (2) the word "unknown" tends to confuse beginning users.
i*86 | x86_64)
basic_machine=$basic_machine-pc
;;
# Object if more than one company name word.
*-*-*)
echo Invalid configuration \`$1\': machine \`$basic_machine\' not recognized 1>&2
exit 1
;;
# Recognize the basic CPU types with company name.
580-* \
| a29k-* \
| aarch64-* | aarch64_be-* \
| alpha-* | alphaev[4-8]-* | alphaev56-* | alphaev6[78]-* \
| alpha64-* | alpha64ev[4-8]-* | alpha64ev56-* | alpha64ev6[78]-* \
| alphapca5[67]-* | alpha64pca5[67]-* | arc-* | arceb-* \
| arm-* | armbe-* | armle-* | armeb-* | armv*-* \
| avr-* | avr32-* \
| be32-* | be64-* \
| bfin-* | bs2000-* \
| c[123]* | c30-* | [cjt]90-* | c4x-* \
| c8051-* | clipper-* | craynv-* | cydra-* \
| d10v-* | d30v-* | dlx-* \
| elxsi-* \
| f30[01]-* | f700-* | fido-* | fr30-* | frv-* | fx80-* \
| h8300-* | h8500-* \
| hppa-* | hppa1.[01]-* | hppa2.0-* | hppa2.0[nw]-* | hppa64-* \
| hexagon-* \
| i*86-* | i860-* | i960-* | ia64-* \
| ip2k-* | iq2000-* \
| k1om-* \
| le32-* | le64-* \
| lm32-* \
| m32c-* | m32r-* | m32rle-* \
| m68000-* | m680[012346]0-* | m68360-* | m683?2-* | m68k-* \
| m88110-* | m88k-* | maxq-* | mcore-* | metag-* \
| microblaze-* | microblazeel-* \
| mips-* | mipsbe-* | mipseb-* | mipsel-* | mipsle-* \
| mips16-* \
| mips64-* | mips64el-* \
| mips64octeon-* | mips64octeonel-* \
| mips64orion-* | mips64orionel-* \
| mips64r5900-* | mips64r5900el-* \
| mips64vr-* | mips64vrel-* \
| mips64vr4100-* | mips64vr4100el-* \
| mips64vr4300-* | mips64vr4300el-* \
| mips64vr5000-* | mips64vr5000el-* \
| mips64vr5900-* | mips64vr5900el-* \
| mipsisa32-* | mipsisa32el-* \
| mipsisa32r2-* | mipsisa32r2el-* \
| mipsisa32r6-* | mipsisa32r6el-* \
| mipsisa64-* | mipsisa64el-* \
| mipsisa64r2-* | mipsisa64r2el-* \
| mipsisa64r6-* | mipsisa64r6el-* \
| mipsisa64sb1-* | mipsisa64sb1el-* \
| mipsisa64sr71k-* | mipsisa64sr71kel-* \
| mipsr5900-* | mipsr5900el-* \
| mipstx39-* | mipstx39el-* \
| mmix-* \
| mt-* \
| msp430-* \
| nds32-* | nds32le-* | nds32be-* \
| nios-* | nios2-* | nios2eb-* | nios2el-* \
| none-* | np1-* | ns16k-* | ns32k-* \
| open8-* \
| or1k*-* \
| orion-* \
| pdp10-* | pdp11-* | pj-* | pjl-* | pn-* | power-* \
| powerpc-* | powerpc64-* | powerpc64le-* | powerpcle-* \
| pyramid-* \
| rl78-* | romp-* | rs6000-* | rx-* \
| sh-* | sh[1234]-* | sh[24]a-* | sh[24]aeb-* | sh[23]e-* | sh[34]eb-* | sheb-* | shbe-* \
| shle-* | sh[1234]le-* | sh3ele-* | sh64-* | sh64le-* \
| sparc-* | sparc64-* | sparc64b-* | sparc64v-* | sparc86x-* | sparclet-* \
| sparclite-* \
| sparcv8-* | sparcv9-* | sparcv9b-* | sparcv9v-* | sv1-* | sx?-* \
| tahoe-* \
| tic30-* | tic4x-* | tic54x-* | tic55x-* | tic6x-* | tic80-* \
| tile*-* \
| tron-* \
| ubicom32-* \
| v850-* | v850e-* | v850e1-* | v850es-* | v850e2-* | v850e2v3-* \
| vax-* \
| visium-* \
| we32k-* \
| x86-* | x86_64-* | xc16x-* | xps100-* \
| xstormy16-* | xtensa*-* \
| ymp-* \
| z8k-* | z80-*)
;;
# Recognize the basic CPU types without company name, with glob match.
xtensa*)
basic_machine=$basic_machine-unknown
;;
# Recognize the various machine names and aliases which stand
# for a CPU type and a company and sometimes even an OS.
386bsd)
basic_machine=i386-unknown
os=-bsd
;;
3b1 | 7300 | 7300-att | att-7300 | pc7300 | safari | unixpc)
basic_machine=m68000-att
;;
3b*)
basic_machine=we32k-att
;;
a29khif)
basic_machine=a29k-amd
os=-udi
;;
abacus)
basic_machine=abacus-unknown
;;
adobe68k)
basic_machine=m68010-adobe
os=-scout
;;
alliant | fx80)
basic_machine=fx80-alliant
;;
altos | altos3068)
basic_machine=m68k-altos
;;
am29k)
basic_machine=a29k-none
os=-bsd
;;
amd64)
basic_machine=x86_64-pc
;;
amd64-*)
basic_machine=x86_64-`echo $basic_machine | sed 's/^[^-]*-//'`
;;
amdahl)
basic_machine=580-amdahl
os=-sysv
;;
amiga | amiga-*)
basic_machine=m68k-unknown
;;
amigaos | amigados)
basic_machine=m68k-unknown
os=-amigaos
;;
amigaunix | amix)
basic_machine=m68k-unknown
os=-sysv4
;;
apollo68)
basic_machine=m68k-apollo
os=-sysv
;;
apollo68bsd)
basic_machine=m68k-apollo
os=-bsd
;;
aros)
basic_machine=i386-pc
os=-aros
;;
aux)
basic_machine=m68k-apple
os=-aux
;;
balance)
basic_machine=ns32k-sequent
os=-dynix
;;
blackfin)
basic_machine=bfin-unknown
os=-linux
;;
blackfin-*)
basic_machine=bfin-`echo $basic_machine | sed 's/^[^-]*-//'`
os=-linux
;;
bluegene*)
basic_machine=powerpc-ibm
os=-cnk
;;
c54x-*)
basic_machine=tic54x-`echo $basic_machine | sed 's/^[^-]*-//'`
;;
c55x-*)
basic_machine=tic55x-`echo $basic_machine | sed 's/^[^-]*-//'`
;;
c6x-*)
basic_machine=tic6x-`echo $basic_machine | sed 's/^[^-]*-//'`
;;
c90)
basic_machine=c90-cray
os=-unicos
;;
cegcc)
basic_machine=arm-unknown
os=-cegcc
;;
convex-c1)
basic_machine=c1-convex
os=-bsd
;;
convex-c2)
basic_machine=c2-convex
os=-bsd
;;
convex-c32)
basic_machine=c32-convex
os=-bsd
;;
convex-c34)
basic_machine=c34-convex
os=-bsd
;;
convex-c38)
basic_machine=c38-convex
os=-bsd
;;
cray | j90)
basic_machine=j90-cray
os=-unicos
;;
craynv)
basic_machine=craynv-cray
os=-unicosmp
;;
cr16 | cr16-*)
basic_machine=cr16-unknown
os=-elf
;;
crds | unos)
basic_machine=m68k-crds
;;
crisv32 | crisv32-* | etraxfs*)
basic_machine=crisv32-axis
;;
cris | cris-* | etrax*)
basic_machine=cris-axis
;;
crx)
basic_machine=crx-unknown
os=-elf
;;
da30 | da30-*)
basic_machine=m68k-da30
;;
decstation | decstation-3100 | pmax | pmax-* | pmin | dec3100 | decstatn)
basic_machine=mips-dec
;;
decsystem10* | dec10*)
basic_machine=pdp10-dec
os=-tops10
;;
decsystem20* | dec20*)
basic_machine=pdp10-dec
os=-tops20
;;
delta | 3300 | motorola-3300 | motorola-delta \
| 3300-motorola | delta-motorola)
basic_machine=m68k-motorola
;;
delta88)
basic_machine=m88k-motorola
os=-sysv3
;;
dicos)
basic_machine=i686-pc
os=-dicos
;;
djgpp)
basic_machine=i586-pc
os=-msdosdjgpp
;;
dpx20 | dpx20-*)
basic_machine=rs6000-bull
os=-bosx
;;
dpx2* | dpx2*-bull)
basic_machine=m68k-bull
os=-sysv3
;;
ebmon29k)
basic_machine=a29k-amd
os=-ebmon
;;
elxsi)
basic_machine=elxsi-elxsi
os=-bsd
;;
encore | umax | mmax)
basic_machine=ns32k-encore
;;
es1800 | OSE68k | ose68k | ose | OSE)
basic_machine=m68k-ericsson
os=-ose
;;
fx2800)
basic_machine=i860-alliant
;;
genix)
basic_machine=ns32k-ns
;;
gmicro)
basic_machine=tron-gmicro
os=-sysv
;;
go32)
basic_machine=i386-pc
os=-go32
;;
h3050r* | hiux*)
basic_machine=hppa1.1-hitachi
os=-hiuxwe2
;;
h8300hms)
basic_machine=h8300-hitachi
os=-hms
;;
h8300xray)
basic_machine=h8300-hitachi
os=-xray
;;
h8500hms)
basic_machine=h8500-hitachi
os=-hms
;;
harris)
basic_machine=m88k-harris
os=-sysv3
;;
hp300-*)
basic_machine=m68k-hp
;;
hp300bsd)
basic_machine=m68k-hp
os=-bsd
;;
hp300hpux)
basic_machine=m68k-hp
os=-hpux
;;
hp3k9[0-9][0-9] | hp9[0-9][0-9])
basic_machine=hppa1.0-hp
;;
hp9k2[0-9][0-9] | hp9k31[0-9])
basic_machine=m68000-hp
;;
hp9k3[2-9][0-9])
basic_machine=m68k-hp
;;
hp9k6[0-9][0-9] | hp6[0-9][0-9])
basic_machine=hppa1.0-hp
;;
hp9k7[0-79][0-9] | hp7[0-79][0-9])
basic_machine=hppa1.1-hp
;;
hp9k78[0-9] | hp78[0-9])
# FIXME: really hppa2.0-hp
basic_machine=hppa1.1-hp
;;
hp9k8[67]1 | hp8[67]1 | hp9k80[24] | hp80[24] | hp9k8[78]9 | hp8[78]9 | hp9k893 | hp893)
# FIXME: really hppa2.0-hp
basic_machine=hppa1.1-hp
;;
hp9k8[0-9][13679] | hp8[0-9][13679])
basic_machine=hppa1.1-hp
;;
hp9k8[0-9][0-9] | hp8[0-9][0-9])
basic_machine=hppa1.0-hp
;;
hppa-next)
os=-nextstep3
;;
hppaosf)
basic_machine=hppa1.1-hp
os=-osf
;;
hppro)
basic_machine=hppa1.1-hp
os=-proelf
;;
i370-ibm* | ibm*)
basic_machine=i370-ibm
;;
i*86v32)
basic_machine=`echo $1 | sed -e 's/86.*/86-pc/'`
os=-sysv32
;;
i*86v4*)
basic_machine=`echo $1 | sed -e 's/86.*/86-pc/'`
os=-sysv4
;;
i*86v)
basic_machine=`echo $1 | sed -e 's/86.*/86-pc/'`
os=-sysv
;;
i*86sol2)
basic_machine=`echo $1 | sed -e 's/86.*/86-pc/'`
os=-solaris2
;;
i386mach)
basic_machine=i386-mach
os=-mach
;;
i386-vsta | vsta)
basic_machine=i386-unknown
os=-vsta
;;
iris | iris4d)
basic_machine=mips-sgi
case $os in
-irix*)
;;
*)
os=-irix4
;;
esac
;;
isi68 | isi)
basic_machine=m68k-isi
os=-sysv
;;
leon-*|leon[3-9]-*)
basic_machine=sparc-`echo $basic_machine | sed 's/-.*//'`
;;
m68knommu)
basic_machine=m68k-unknown
os=-linux
;;
m68knommu-*)
basic_machine=m68k-`echo $basic_machine | sed 's/^[^-]*-//'`
os=-linux
;;
m88k-omron*)
basic_machine=m88k-omron
;;
magnum | m3230)
basic_machine=mips-mips
os=-sysv
;;
merlin)
basic_machine=ns32k-utek
os=-sysv
;;
microblaze*)
basic_machine=microblaze-xilinx
;;
mingw64)
basic_machine=x86_64-pc
os=-mingw64
;;
mingw32)
basic_machine=i686-pc
os=-mingw32
;;
mingw32ce)
basic_machine=arm-unknown
os=-mingw32ce
;;
miniframe)
basic_machine=m68000-convergent
;;
*mint | -mint[0-9]* | *MiNT | *MiNT[0-9]*)
basic_machine=m68k-atari
os=-mint
;;
mips3*-*)
basic_machine=`echo $basic_machine | sed -e 's/mips3/mips64/'`
;;
mips3*)
basic_machine=`echo $basic_machine | sed -e 's/mips3/mips64/'`-unknown
;;
monitor)
basic_machine=m68k-rom68k
os=-coff
;;
morphos)
basic_machine=powerpc-unknown
os=-morphos
;;
moxiebox)
basic_machine=moxie-unknown
os=-moxiebox
;;
msdos)
basic_machine=i386-pc
os=-msdos
;;
ms1-*)
basic_machine=`echo $basic_machine | sed -e 's/ms1-/mt-/'`
;;
msys)
basic_machine=i686-pc
os=-msys
;;
mvs)
basic_machine=i370-ibm
os=-mvs
;;
nacl)
basic_machine=le32-unknown
os=-nacl
;;
ncr3000)
basic_machine=i486-ncr
os=-sysv4
;;
netbsd386)
basic_machine=i386-unknown
os=-netbsd
;;
netwinder)
basic_machine=armv4l-rebel
os=-linux
;;
news | news700 | news800 | news900)
basic_machine=m68k-sony
os=-newsos
;;
news1000)
basic_machine=m68030-sony
os=-newsos
;;
news-3600 | risc-news)
basic_machine=mips-sony
os=-newsos
;;
necv70)
basic_machine=v70-nec
os=-sysv
;;
next | m*-next )
basic_machine=m68k-next
case $os in
-nextstep* )
;;
-ns2*)
os=-nextstep2
;;
*)
os=-nextstep3
;;
esac
;;
nh3000)
basic_machine=m68k-harris
os=-cxux
;;
nh[45]000)
basic_machine=m88k-harris
os=-cxux
;;
nindy960)
basic_machine=i960-intel
os=-nindy
;;
mon960)
basic_machine=i960-intel
os=-mon960
;;
nonstopux)
basic_machine=mips-compaq
os=-nonstopux
;;
np1)
basic_machine=np1-gould
;;
neo-tandem)
basic_machine=neo-tandem
;;
nse-tandem)
basic_machine=nse-tandem
;;
nsr-tandem)
basic_machine=nsr-tandem
;;
op50n-* | op60c-*)
basic_machine=hppa1.1-oki
os=-proelf
;;
openrisc | openrisc-*)
basic_machine=or32-unknown
;;
os400)
basic_machine=powerpc-ibm
os=-os400
;;
OSE68000 | ose68000)
basic_machine=m68000-ericsson
os=-ose
;;
os68k)
basic_machine=m68k-none
os=-os68k
;;
pa-hitachi)
basic_machine=hppa1.1-hitachi
os=-hiuxwe2
;;
paragon)
basic_machine=i860-intel
os=-osf
;;
parisc)
basic_machine=hppa-unknown
os=-linux
;;
parisc-*)
basic_machine=hppa-`echo $basic_machine | sed 's/^[^-]*-//'`
os=-linux
;;
pbd)
basic_machine=sparc-tti
;;
pbb)
basic_machine=m68k-tti
;;
pc532 | pc532-*)
basic_machine=ns32k-pc532
;;
pc98)
basic_machine=i386-pc
;;
pc98-*)
basic_machine=i386-`echo $basic_machine | sed 's/^[^-]*-//'`
;;
pentium | p5 | k5 | k6 | nexgen | viac3)
basic_machine=i586-pc
;;
pentiumpro | p6 | 6x86 | athlon | athlon_*)
basic_machine=i686-pc
;;
pentiumii | pentium2 | pentiumiii | pentium3)
basic_machine=i686-pc
;;
pentium4)
basic_machine=i786-pc
;;
pentium-* | p5-* | k5-* | k6-* | nexgen-* | viac3-*)
basic_machine=i586-`echo $basic_machine | sed 's/^[^-]*-//'`
;;
pentiumpro-* | p6-* | 6x86-* | athlon-*)
basic_machine=i686-`echo $basic_machine | sed 's/^[^-]*-//'`
;;
pentiumii-* | pentium2-* | pentiumiii-* | pentium3-*)
basic_machine=i686-`echo $basic_machine | sed 's/^[^-]*-//'`
;;
pentium4-*)
basic_machine=i786-`echo $basic_machine | sed 's/^[^-]*-//'`
;;
pn)
basic_machine=pn-gould
;;
power) basic_machine=power-ibm
;;
ppc | ppcbe) basic_machine=powerpc-unknown
;;
ppc-* | ppcbe-*)
basic_machine=powerpc-`echo $basic_machine | sed 's/^[^-]*-//'`
;;
ppcle | powerpclittle | ppc-le | powerpc-little)
basic_machine=powerpcle-unknown
;;
ppcle-* | powerpclittle-*)
basic_machine=powerpcle-`echo $basic_machine | sed 's/^[^-]*-//'`
;;
ppc64) basic_machine=powerpc64-unknown
;;
ppc64-*) basic_machine=powerpc64-`echo $basic_machine | sed 's/^[^-]*-//'`
;;
ppc64le | powerpc64little | ppc64-le | powerpc64-little)
basic_machine=powerpc64le-unknown
;;
ppc64le-* | powerpc64little-*)
basic_machine=powerpc64le-`echo $basic_machine | sed 's/^[^-]*-//'`
;;
ps2)
basic_machine=i386-ibm
;;
pw32)
basic_machine=i586-unknown
os=-pw32
;;
rdos | rdos64)
basic_machine=x86_64-pc
os=-rdos
;;
rdos32)
basic_machine=i386-pc
os=-rdos
;;
rom68k)
basic_machine=m68k-rom68k
os=-coff
;;
rm[46]00)
basic_machine=mips-siemens
;;
rtpc | rtpc-*)
basic_machine=romp-ibm
;;
s390 | s390-*)
basic_machine=s390-ibm
;;
s390x | s390x-*)
basic_machine=s390x-ibm
;;
sa29200)
basic_machine=a29k-amd
os=-udi
;;
sb1)
basic_machine=mipsisa64sb1-unknown
;;
sb1el)
basic_machine=mipsisa64sb1el-unknown
;;
sde)
basic_machine=mipsisa32-sde
os=-elf
;;
sei)
basic_machine=mips-sei
os=-seiux
;;
sequent)
basic_machine=i386-sequent
;;
sh)
basic_machine=sh-hitachi
os=-hms
;;
sh5el)
basic_machine=sh5le-unknown
;;
sh64)
basic_machine=sh64-unknown
;;
sparclite-wrs | simso-wrs)
basic_machine=sparclite-wrs
os=-vxworks
;;
sps7)
basic_machine=m68k-bull
os=-sysv2
;;
spur)
basic_machine=spur-unknown
;;
st2000)
basic_machine=m68k-tandem
;;
stratus)
basic_machine=i860-stratus
os=-sysv4
;;
strongarm-* | thumb-*)
basic_machine=arm-`echo $basic_machine | sed 's/^[^-]*-//'`
;;
sun2)
basic_machine=m68000-sun
;;
sun2os3)
basic_machine=m68000-sun
os=-sunos3
;;
sun2os4)
basic_machine=m68000-sun
os=-sunos4
;;
sun3os3)
basic_machine=m68k-sun
os=-sunos3
;;
sun3os4)
basic_machine=m68k-sun
os=-sunos4
;;
sun4os3)
basic_machine=sparc-sun
os=-sunos3
;;
sun4os4)
basic_machine=sparc-sun
os=-sunos4
;;
sun4sol2)
basic_machine=sparc-sun
os=-solaris2
;;
sun3 | sun3-*)
basic_machine=m68k-sun
;;
sun4)
basic_machine=sparc-sun
;;
sun386 | sun386i | roadrunner)
basic_machine=i386-sun
;;
sv1)
basic_machine=sv1-cray
os=-unicos
;;
symmetry)
basic_machine=i386-sequent
os=-dynix
;;
t3e)
basic_machine=alphaev5-cray
os=-unicos
;;
t90)
basic_machine=t90-cray
os=-unicos
;;
tile*)
basic_machine=$basic_machine-unknown
os=-linux-gnu
;;
tx39)
basic_machine=mipstx39-unknown
;;
tx39el)
basic_machine=mipstx39el-unknown
;;
toad1)
basic_machine=pdp10-xkl
os=-tops20
;;
tower | tower-32)
basic_machine=m68k-ncr
;;
tpf)
basic_machine=s390x-ibm
os=-tpf
;;
udi29k)
basic_machine=a29k-amd
os=-udi
;;
ultra3)
basic_machine=a29k-nyu
os=-sym1
;;
v810 | necv810)
basic_machine=v810-nec
os=-none
;;
vaxv)
basic_machine=vax-dec
os=-sysv
;;
vms)
basic_machine=vax-dec
os=-vms
;;
vpp*|vx|vx-*)
basic_machine=f301-fujitsu
;;
vxworks960)
basic_machine=i960-wrs
os=-vxworks
;;
vxworks68)
basic_machine=m68k-wrs
os=-vxworks
;;
vxworks29k)
basic_machine=a29k-wrs
os=-vxworks
;;
w65*)
basic_machine=w65-wdc
os=-none
;;
w89k-*)
basic_machine=hppa1.1-winbond
os=-proelf
;;
xbox)
basic_machine=i686-pc
os=-mingw32
;;
xps | xps100)
basic_machine=xps100-honeywell
;;
xscale-* | xscalee[bl]-*)
basic_machine=`echo $basic_machine | sed 's/^xscale/arm/'`
;;
ymp)
basic_machine=ymp-cray
os=-unicos
;;
z8k-*-coff)
basic_machine=z8k-unknown
os=-sim
;;
z80-*-coff)
basic_machine=z80-unknown
os=-sim
;;
none)
basic_machine=none-none
os=-none
;;
# Here we handle the default manufacturer of certain CPU types. It is in
# some cases the only manufacturer, in others, it is the most popular.
w89k)
basic_machine=hppa1.1-winbond
;;
op50n)
basic_machine=hppa1.1-oki
;;
op60c)
basic_machine=hppa1.1-oki
;;
romp)
basic_machine=romp-ibm
;;
mmix)
basic_machine=mmix-knuth
;;
rs6000)
basic_machine=rs6000-ibm
;;
vax)
basic_machine=vax-dec
;;
pdp10)
# there are many clones, so DEC is not a safe bet
basic_machine=pdp10-unknown
;;
pdp11)
basic_machine=pdp11-dec
;;
we32k)
basic_machine=we32k-att
;;
sh[1234] | sh[24]a | sh[24]aeb | sh[34]eb | sh[1234]le | sh[23]ele)
basic_machine=sh-unknown
;;
sparc | sparcv8 | sparcv9 | sparcv9b | sparcv9v)
basic_machine=sparc-sun
;;
cydra)
basic_machine=cydra-cydrome
;;
orion)
basic_machine=orion-highlevel
;;
orion105)
basic_machine=clipper-highlevel
;;
mac | mpw | mac-mpw)
basic_machine=m68k-apple
;;
pmac | pmac-mpw)
basic_machine=powerpc-apple
;;
*-unknown)
# Make sure to match an already-canonicalized machine name.
;;
*)
echo Invalid configuration \`$1\': machine \`$basic_machine\' not recognized 1>&2
exit 1
;;
esac
# Here we canonicalize certain aliases for manufacturers.
case $basic_machine in
*-digital*)
basic_machine=`echo $basic_machine | sed 's/digital.*/dec/'`
;;
*-commodore*)
basic_machine=`echo $basic_machine | sed 's/commodore.*/cbm/'`
;;
*)
;;
esac
# Decode manufacturer-specific aliases for certain operating systems.
if [ x"$os" != x"" ]
then
case $os in
# First match some system type aliases
# that might get confused with valid system types.
# -solaris* is a basic system type, with this one exception.
-auroraux)
os=-auroraux
;;
-solaris1 | -solaris1.*)
os=`echo $os | sed -e 's|solaris1|sunos4|'`
;;
-solaris)
os=-solaris2
;;
-svr4*)
os=-sysv4
;;
-unixware*)
os=-sysv4.2uw
;;
-gnu/linux*)
os=`echo $os | sed -e 's|gnu/linux|linux-gnu|'`
;;
# First accept the basic system types.
# The portable systems comes first.
# Each alternative MUST END IN A *, to match a version number.
# -sysv* is not here because it comes later, after sysvr4.
-gnu* | -bsd* | -mach* | -minix* | -genix* | -ultrix* | -irix* \
| -*vms* | -sco* | -esix* | -isc* | -aix* | -cnk* | -sunos | -sunos[34]*\
| -hpux* | -unos* | -osf* | -luna* | -dgux* | -auroraux* | -solaris* \
| -sym* | -kopensolaris* | -plan9* \
| -amigaos* | -amigados* | -msdos* | -newsos* | -unicos* | -aof* \
| -aos* | -aros* \
| -nindy* | -vxsim* | -vxworks* | -ebmon* | -hms* | -mvs* \
| -clix* | -riscos* | -uniplus* | -iris* | -rtu* | -xenix* \
| -hiux* | -386bsd* | -knetbsd* | -mirbsd* | -netbsd* \
| -bitrig* | -openbsd* | -solidbsd* \
| -ekkobsd* | -kfreebsd* | -freebsd* | -riscix* | -lynxos* \
| -bosx* | -nextstep* | -cxux* | -aout* | -elf* | -oabi* \
| -ptx* | -coff* | -ecoff* | -winnt* | -domain* | -vsta* \
| -udi* | -eabi* | -lites* | -ieee* | -go32* | -aux* \
| -chorusos* | -chorusrdb* | -cegcc* \
| -cygwin* | -msys* | -pe* | -psos* | -moss* | -proelf* | -rtems* \
| -mingw32* | -mingw64* | -linux-gnu* | -linux-android* \
| -linux-newlib* | -linux-musl* | -linux-uclibc* \
| -uxpv* | -beos* | -mpeix* | -udk* | -moxiebox* \
| -interix* | -uwin* | -mks* | -rhapsody* | -darwin* | -opened* \
| -openstep* | -oskit* | -conix* | -pw32* | -nonstopux* \
| -storm-chaos* | -tops10* | -tenex* | -tops20* | -its* \
| -os2* | -vos* | -palmos* | -uclinux* | -nucleus* \
| -morphos* | -superux* | -rtmk* | -rtmk-nova* | -windiss* \
| -powermax* | -dnix* | -nx6 | -nx7 | -sei* | -dragonfly* \
| -skyos* | -haiku* | -rdos* | -toppers* | -drops* | -es* | -tirtos*)
# Remember, each alternative MUST END IN *, to match a version number.
;;
-qnx*)
case $basic_machine in
x86-* | i*86-*)
;;
*)
os=-nto$os
;;
esac
;;
-nto-qnx*)
;;
-nto*)
os=`echo $os | sed -e 's|nto|nto-qnx|'`
;;
-sim | -es1800* | -hms* | -xray | -os68k* | -none* | -v88r* \
| -windows* | -osx | -abug | -netware* | -os9* | -beos* | -haiku* \
| -macos* | -mpw* | -magic* | -mmixware* | -mon960* | -lnews*)
;;
-mac*)
os=`echo $os | sed -e 's|mac|macos|'`
;;
-linux-dietlibc)
os=-linux-dietlibc
;;
-linux*)
os=`echo $os | sed -e 's|linux|linux-gnu|'`
;;
-sunos5*)
os=`echo $os | sed -e 's|sunos5|solaris2|'`
;;
-sunos6*)
os=`echo $os | sed -e 's|sunos6|solaris3|'`
;;
-opened*)
os=-openedition
;;
-os400*)
os=-os400
;;
-wince*)
os=-wince
;;
-osfrose*)
os=-osfrose
;;
-osf*)
os=-osf
;;
-utek*)
os=-bsd
;;
-dynix*)
os=-bsd
;;
-acis*)
os=-aos
;;
-atheos*)
os=-atheos
;;
-syllable*)
os=-syllable
;;
-386bsd)
os=-bsd
;;
-ctix* | -uts*)
os=-sysv
;;
-nova*)
os=-rtmk-nova
;;
-ns2 )
os=-nextstep2
;;
-nsk*)
os=-nsk
;;
# Preserve the version number of sinix5.
-sinix5.*)
os=`echo $os | sed -e 's|sinix|sysv|'`
;;
-sinix*)
os=-sysv4
;;
-tpf*)
os=-tpf
;;
-triton*)
os=-sysv3
;;
-oss*)
os=-sysv3
;;
-svr4)
os=-sysv4
;;
-svr3)
os=-sysv3
;;
-sysvr4)
os=-sysv4
;;
# This must come after -sysvr4.
-sysv*)
;;
-ose*)
os=-ose
;;
-es1800*)
os=-ose
;;
-xenix)
os=-xenix
;;
-*mint | -mint[0-9]* | -*MiNT | -MiNT[0-9]*)
os=-mint
;;
-aros*)
os=-aros
;;
-zvmoe)
os=-zvmoe
;;
-dicos*)
os=-dicos
;;
-nacl*)
;;
-none)
;;
*)
# Get rid of the `-' at the beginning of $os.
os=`echo $os | sed 's/[^-]*-//'`
echo Invalid configuration \`$1\': system \`$os\' not recognized 1>&2
exit 1
;;
esac
else
# Here we handle the default operating systems that come with various machines.
# The value should be what the vendor currently ships out the door with their
# machine or put another way, the most popular os provided with the machine.
# Note that if you're going to try to match "-MANUFACTURER" here (say,
# "-sun"), then you have to tell the case statement up towards the top
# that MANUFACTURER isn't an operating system. Otherwise, code above
# will signal an error saying that MANUFACTURER isn't an operating
# system, and we'll never get to this point.
case $basic_machine in
score-*)
os=-elf
;;
spu-*)
os=-elf
;;
*-acorn)
os=-riscix1.2
;;
arm*-rebel)
os=-linux
;;
arm*-semi)
os=-aout
;;
c4x-* | tic4x-*)
os=-coff
;;
c8051-*)
os=-elf
;;
hexagon-*)
os=-elf
;;
tic54x-*)
os=-coff
;;
tic55x-*)
os=-coff
;;
tic6x-*)
os=-coff
;;
# This must come before the *-dec entry.
pdp10-*)
os=-tops20
;;
pdp11-*)
os=-none
;;
*-dec | vax-*)
os=-ultrix4.2
;;
m68*-apollo)
os=-domain
;;
i386-sun)
os=-sunos4.0.2
;;
m68000-sun)
os=-sunos3
;;
m68*-cisco)
os=-aout
;;
mep-*)
os=-elf
;;
mips*-cisco)
os=-elf
;;
mips*-*)
os=-elf
;;
or32-*)
os=-coff
;;
*-tti) # must be before sparc entry or we get the wrong os.
os=-sysv3
;;
sparc-* | *-sun)
os=-sunos4.1.1
;;
*-be)
os=-beos
;;
*-haiku)
os=-haiku
;;
*-ibm)
os=-aix
;;
*-knuth)
os=-mmixware
;;
*-wec)
os=-proelf
;;
*-winbond)
os=-proelf
;;
*-oki)
os=-proelf
;;
*-hp)
os=-hpux
;;
*-hitachi)
os=-hiux
;;
i860-* | *-att | *-ncr | *-altos | *-motorola | *-convergent)
os=-sysv
;;
*-cbm)
os=-amigaos
;;
*-dg)
os=-dgux
;;
*-dolphin)
os=-sysv3
;;
m68k-ccur)
os=-rtu
;;
m88k-omron*)
os=-luna
;;
*-next )
os=-nextstep
;;
*-sequent)
os=-ptx
;;
*-crds)
os=-unos
;;
*-ns)
os=-genix
;;
i370-*)
os=-mvs
;;
*-next)
os=-nextstep3
;;
*-gould)
os=-sysv
;;
*-highlevel)
os=-bsd
;;
*-encore)
os=-bsd
;;
*-sgi)
os=-irix
;;
*-siemens)
os=-sysv4
;;
*-masscomp)
os=-rtu
;;
f30[01]-fujitsu | f700-fujitsu)
os=-uxpv
;;
*-rom68k)
os=-coff
;;
*-*bug)
os=-coff
;;
*-apple)
os=-macos
;;
*-atari*)
os=-mint
;;
*)
os=-none
;;
esac
fi
# Here we handle the case where we know the os, and the CPU type, but not the
# manufacturer. We pick the logical manufacturer.
vendor=unknown
case $basic_machine in
*-unknown)
case $os in
-riscix*)
vendor=acorn
;;
-sunos*)
vendor=sun
;;
-cnk*|-aix*)
vendor=ibm
;;
-beos*)
vendor=be
;;
-hpux*)
vendor=hp
;;
-mpeix*)
vendor=hp
;;
-hiux*)
vendor=hitachi
;;
-unos*)
vendor=crds
;;
-dgux*)
vendor=dg
;;
-luna*)
vendor=omron
;;
-genix*)
vendor=ns
;;
-mvs* | -opened*)
vendor=ibm
;;
-os400*)
vendor=ibm
;;
-ptx*)
vendor=sequent
;;
-tpf*)
vendor=ibm
;;
-vxsim* | -vxworks* | -windiss*)
vendor=wrs
;;
-aux*)
vendor=apple
;;
-hms*)
vendor=hitachi
;;
-mpw* | -macos*)
vendor=apple
;;
-*mint | -mint[0-9]* | -*MiNT | -MiNT[0-9]*)
vendor=atari
;;
-vos*)
vendor=stratus
;;
esac
basic_machine=`echo $basic_machine | sed "s/unknown/$vendor/"`
;;
esac
echo $basic_machine$os
exit
# Local variables:
# eval: (add-hook 'write-file-hooks 'time-stamp)
# time-stamp-start: "timestamp='"
# time-stamp-format: "%:y-%02m-%02d"
# time-stamp-end: "'"
# End:
diff --git a/build-aux/depcomp b/build-aux/depcomp
index 4ebd5b3..f0a474c 100755
--- a/build-aux/depcomp
+++ b/build-aux/depcomp
@@ -1,791 +1,791 @@
#! /bin/sh
# depcomp - compile a program generating dependencies as side-effects
scriptversion=2013-05-30.07; # UTC
# Copyright (C) 1999-2013 Free Software Foundation, Inc.
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2, or (at your option)
# any later version.
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
# You should have received a copy of the GNU General Public License
-# along with this program. If not, see .
+# along with this program. If not, see .
# As a special exception to the GNU General Public License, if you
# distribute this file as part of a program that contains a
# configuration script generated by Autoconf, you may include it under
# the same distribution terms that you use for the rest of that program.
# Originally written by Alexandre Oliva .
case $1 in
'')
echo "$0: No command. Try '$0 --help' for more information." 1>&2
exit 1;
;;
-h | --h*)
cat <<\EOF
Usage: depcomp [--help] [--version] PROGRAM [ARGS]
Run PROGRAMS ARGS to compile a file, generating dependencies
as side-effects.
Environment variables:
depmode Dependency tracking mode.
source Source file read by 'PROGRAMS ARGS'.
object Object file output by 'PROGRAMS ARGS'.
DEPDIR directory where to store dependencies.
depfile Dependency file to output.
tmpdepfile Temporary file to use when outputting dependencies.
libtool Whether libtool is used (yes/no).
Report bugs to .
EOF
exit $?
;;
-v | --v*)
echo "depcomp $scriptversion"
exit $?
;;
esac
# Get the directory component of the given path, and save it in the
# global variables '$dir'. Note that this directory component will
# be either empty or ending with a '/' character. This is deliberate.
set_dir_from ()
{
case $1 in
*/*) dir=`echo "$1" | sed -e 's|/[^/]*$|/|'`;;
*) dir=;;
esac
}
# Get the suffix-stripped basename of the given path, and save it the
# global variable '$base'.
set_base_from ()
{
base=`echo "$1" | sed -e 's|^.*/||' -e 's/\.[^.]*$//'`
}
# If no dependency file was actually created by the compiler invocation,
# we still have to create a dummy depfile, to avoid errors with the
# Makefile "include basename.Plo" scheme.
make_dummy_depfile ()
{
echo "#dummy" > "$depfile"
}
# Factor out some common post-processing of the generated depfile.
# Requires the auxiliary global variable '$tmpdepfile' to be set.
aix_post_process_depfile ()
{
# If the compiler actually managed to produce a dependency file,
# post-process it.
if test -f "$tmpdepfile"; then
# Each line is of the form 'foo.o: dependency.h'.
# Do two passes, one to just change these to
# $object: dependency.h
# and one to simply output
# dependency.h:
# which is needed to avoid the deleted-header problem.
{ sed -e "s,^.*\.[$lower]*:,$object:," < "$tmpdepfile"
sed -e "s,^.*\.[$lower]*:[$tab ]*,," -e 's,$,:,' < "$tmpdepfile"
} > "$depfile"
rm -f "$tmpdepfile"
else
make_dummy_depfile
fi
}
# A tabulation character.
tab=' '
# A newline character.
nl='
'
# Character ranges might be problematic outside the C locale.
# These definitions help.
upper=ABCDEFGHIJKLMNOPQRSTUVWXYZ
lower=abcdefghijklmnopqrstuvwxyz
digits=0123456789
alpha=${upper}${lower}
if test -z "$depmode" || test -z "$source" || test -z "$object"; then
echo "depcomp: Variables source, object and depmode must be set" 1>&2
exit 1
fi
# Dependencies for sub/bar.o or sub/bar.obj go into sub/.deps/bar.Po.
depfile=${depfile-`echo "$object" |
sed 's|[^\\/]*$|'${DEPDIR-.deps}'/&|;s|\.\([^.]*\)$|.P\1|;s|Pobj$|Po|'`}
tmpdepfile=${tmpdepfile-`echo "$depfile" | sed 's/\.\([^.]*\)$/.T\1/'`}
rm -f "$tmpdepfile"
# Avoid interferences from the environment.
gccflag= dashmflag=
# Some modes work just like other modes, but use different flags. We
# parameterize here, but still list the modes in the big case below,
# to make depend.m4 easier to write. Note that we *cannot* use a case
# here, because this file can only contain one case statement.
if test "$depmode" = hp; then
# HP compiler uses -M and no extra arg.
gccflag=-M
depmode=gcc
fi
if test "$depmode" = dashXmstdout; then
# This is just like dashmstdout with a different argument.
dashmflag=-xM
depmode=dashmstdout
fi
cygpath_u="cygpath -u -f -"
if test "$depmode" = msvcmsys; then
# This is just like msvisualcpp but w/o cygpath translation.
# Just convert the backslash-escaped backslashes to single forward
# slashes to satisfy depend.m4
cygpath_u='sed s,\\\\,/,g'
depmode=msvisualcpp
fi
if test "$depmode" = msvc7msys; then
# This is just like msvc7 but w/o cygpath translation.
# Just convert the backslash-escaped backslashes to single forward
# slashes to satisfy depend.m4
cygpath_u='sed s,\\\\,/,g'
depmode=msvc7
fi
if test "$depmode" = xlc; then
# IBM C/C++ Compilers xlc/xlC can output gcc-like dependency information.
gccflag=-qmakedep=gcc,-MF
depmode=gcc
fi
case "$depmode" in
gcc3)
## gcc 3 implements dependency tracking that does exactly what
## we want. Yay! Note: for some reason libtool 1.4 doesn't like
## it if -MD -MP comes after the -MF stuff. Hmm.
## Unfortunately, FreeBSD c89 acceptance of flags depends upon
## the command line argument order; so add the flags where they
## appear in depend2.am. Note that the slowdown incurred here
## affects only configure: in makefiles, %FASTDEP% shortcuts this.
for arg
do
case $arg in
-c) set fnord "$@" -MT "$object" -MD -MP -MF "$tmpdepfile" "$arg" ;;
*) set fnord "$@" "$arg" ;;
esac
shift # fnord
shift # $arg
done
"$@"
stat=$?
if test $stat -ne 0; then
rm -f "$tmpdepfile"
exit $stat
fi
mv "$tmpdepfile" "$depfile"
;;
gcc)
## Note that this doesn't just cater to obsosete pre-3.x GCC compilers.
## but also to in-use compilers like IMB xlc/xlC and the HP C compiler.
## (see the conditional assignment to $gccflag above).
## There are various ways to get dependency output from gcc. Here's
## why we pick this rather obscure method:
## - Don't want to use -MD because we'd like the dependencies to end
## up in a subdir. Having to rename by hand is ugly.
## (We might end up doing this anyway to support other compilers.)
## - The DEPENDENCIES_OUTPUT environment variable makes gcc act like
## -MM, not -M (despite what the docs say). Also, it might not be
## supported by the other compilers which use the 'gcc' depmode.
## - Using -M directly means running the compiler twice (even worse
## than renaming).
if test -z "$gccflag"; then
gccflag=-MD,
fi
"$@" -Wp,"$gccflag$tmpdepfile"
stat=$?
if test $stat -ne 0; then
rm -f "$tmpdepfile"
exit $stat
fi
rm -f "$depfile"
echo "$object : \\" > "$depfile"
# The second -e expression handles DOS-style file names with drive
# letters.
sed -e 's/^[^:]*: / /' \
-e 's/^['$alpha']:\/[^:]*: / /' < "$tmpdepfile" >> "$depfile"
## This next piece of magic avoids the "deleted header file" problem.
## The problem is that when a header file which appears in a .P file
## is deleted, the dependency causes make to die (because there is
## typically no way to rebuild the header). We avoid this by adding
## dummy dependencies for each header file. Too bad gcc doesn't do
## this for us directly.
## Some versions of gcc put a space before the ':'. On the theory
## that the space means something, we add a space to the output as
## well. hp depmode also adds that space, but also prefixes the VPATH
## to the object. Take care to not repeat it in the output.
## Some versions of the HPUX 10.20 sed can't process this invocation
## correctly. Breaking it into two sed invocations is a workaround.
tr ' ' "$nl" < "$tmpdepfile" \
| sed -e 's/^\\$//' -e '/^$/d' -e "s|.*$object$||" -e '/:$/d' \
| sed -e 's/$/ :/' >> "$depfile"
rm -f "$tmpdepfile"
;;
hp)
# This case exists only to let depend.m4 do its work. It works by
# looking at the text of this script. This case will never be run,
# since it is checked for above.
exit 1
;;
sgi)
if test "$libtool" = yes; then
"$@" "-Wp,-MDupdate,$tmpdepfile"
else
"$@" -MDupdate "$tmpdepfile"
fi
stat=$?
if test $stat -ne 0; then
rm -f "$tmpdepfile"
exit $stat
fi
rm -f "$depfile"
if test -f "$tmpdepfile"; then # yes, the sourcefile depend on other files
echo "$object : \\" > "$depfile"
# Clip off the initial element (the dependent). Don't try to be
# clever and replace this with sed code, as IRIX sed won't handle
# lines with more than a fixed number of characters (4096 in
# IRIX 6.2 sed, 8192 in IRIX 6.5). We also remove comment lines;
# the IRIX cc adds comments like '#:fec' to the end of the
# dependency line.
tr ' ' "$nl" < "$tmpdepfile" \
| sed -e 's/^.*\.o://' -e 's/#.*$//' -e '/^$/ d' \
| tr "$nl" ' ' >> "$depfile"
echo >> "$depfile"
# The second pass generates a dummy entry for each header file.
tr ' ' "$nl" < "$tmpdepfile" \
| sed -e 's/^.*\.o://' -e 's/#.*$//' -e '/^$/ d' -e 's/$/:/' \
>> "$depfile"
else
make_dummy_depfile
fi
rm -f "$tmpdepfile"
;;
xlc)
# This case exists only to let depend.m4 do its work. It works by
# looking at the text of this script. This case will never be run,
# since it is checked for above.
exit 1
;;
aix)
# The C for AIX Compiler uses -M and outputs the dependencies
# in a .u file. In older versions, this file always lives in the
# current directory. Also, the AIX compiler puts '$object:' at the
# start of each line; $object doesn't have directory information.
# Version 6 uses the directory in both cases.
set_dir_from "$object"
set_base_from "$object"
if test "$libtool" = yes; then
tmpdepfile1=$dir$base.u
tmpdepfile2=$base.u
tmpdepfile3=$dir.libs/$base.u
"$@" -Wc,-M
else
tmpdepfile1=$dir$base.u
tmpdepfile2=$dir$base.u
tmpdepfile3=$dir$base.u
"$@" -M
fi
stat=$?
if test $stat -ne 0; then
rm -f "$tmpdepfile1" "$tmpdepfile2" "$tmpdepfile3"
exit $stat
fi
for tmpdepfile in "$tmpdepfile1" "$tmpdepfile2" "$tmpdepfile3"
do
test -f "$tmpdepfile" && break
done
aix_post_process_depfile
;;
tcc)
# tcc (Tiny C Compiler) understand '-MD -MF file' since version 0.9.26
# FIXME: That version still under development at the moment of writing.
# Make that this statement remains true also for stable, released
# versions.
# It will wrap lines (doesn't matter whether long or short) with a
# trailing '\', as in:
#
# foo.o : \
# foo.c \
# foo.h \
#
# It will put a trailing '\' even on the last line, and will use leading
# spaces rather than leading tabs (at least since its commit 0394caf7
# "Emit spaces for -MD").
"$@" -MD -MF "$tmpdepfile"
stat=$?
if test $stat -ne 0; then
rm -f "$tmpdepfile"
exit $stat
fi
rm -f "$depfile"
# Each non-empty line is of the form 'foo.o : \' or ' dep.h \'.
# We have to change lines of the first kind to '$object: \'.
sed -e "s|.*:|$object :|" < "$tmpdepfile" > "$depfile"
# And for each line of the second kind, we have to emit a 'dep.h:'
# dummy dependency, to avoid the deleted-header problem.
sed -n -e 's|^ *\(.*\) *\\$|\1:|p' < "$tmpdepfile" >> "$depfile"
rm -f "$tmpdepfile"
;;
## The order of this option in the case statement is important, since the
## shell code in configure will try each of these formats in the order
## listed in this file. A plain '-MD' option would be understood by many
## compilers, so we must ensure this comes after the gcc and icc options.
pgcc)
# Portland's C compiler understands '-MD'.
# Will always output deps to 'file.d' where file is the root name of the
# source file under compilation, even if file resides in a subdirectory.
# The object file name does not affect the name of the '.d' file.
# pgcc 10.2 will output
# foo.o: sub/foo.c sub/foo.h
# and will wrap long lines using '\' :
# foo.o: sub/foo.c ... \
# sub/foo.h ... \
# ...
set_dir_from "$object"
# Use the source, not the object, to determine the base name, since
# that's sadly what pgcc will do too.
set_base_from "$source"
tmpdepfile=$base.d
# For projects that build the same source file twice into different object
# files, the pgcc approach of using the *source* file root name can cause
# problems in parallel builds. Use a locking strategy to avoid stomping on
# the same $tmpdepfile.
lockdir=$base.d-lock
trap "
echo '$0: caught signal, cleaning up...' >&2
rmdir '$lockdir'
exit 1
" 1 2 13 15
numtries=100
i=$numtries
while test $i -gt 0; do
# mkdir is a portable test-and-set.
if mkdir "$lockdir" 2>/dev/null; then
# This process acquired the lock.
"$@" -MD
stat=$?
# Release the lock.
rmdir "$lockdir"
break
else
# If the lock is being held by a different process, wait
# until the winning process is done or we timeout.
while test -d "$lockdir" && test $i -gt 0; do
sleep 1
i=`expr $i - 1`
done
fi
i=`expr $i - 1`
done
trap - 1 2 13 15
if test $i -le 0; then
echo "$0: failed to acquire lock after $numtries attempts" >&2
echo "$0: check lockdir '$lockdir'" >&2
exit 1
fi
if test $stat -ne 0; then
rm -f "$tmpdepfile"
exit $stat
fi
rm -f "$depfile"
# Each line is of the form `foo.o: dependent.h',
# or `foo.o: dep1.h dep2.h \', or ` dep3.h dep4.h \'.
# Do two passes, one to just change these to
# `$object: dependent.h' and one to simply `dependent.h:'.
sed "s,^[^:]*:,$object :," < "$tmpdepfile" > "$depfile"
# Some versions of the HPUX 10.20 sed can't process this invocation
# correctly. Breaking it into two sed invocations is a workaround.
sed 's,^[^:]*: \(.*\)$,\1,;s/^\\$//;/^$/d;/:$/d' < "$tmpdepfile" \
| sed -e 's/$/ :/' >> "$depfile"
rm -f "$tmpdepfile"
;;
hp2)
# The "hp" stanza above does not work with aCC (C++) and HP's ia64
# compilers, which have integrated preprocessors. The correct option
# to use with these is +Maked; it writes dependencies to a file named
# 'foo.d', which lands next to the object file, wherever that
# happens to be.
# Much of this is similar to the tru64 case; see comments there.
set_dir_from "$object"
set_base_from "$object"
if test "$libtool" = yes; then
tmpdepfile1=$dir$base.d
tmpdepfile2=$dir.libs/$base.d
"$@" -Wc,+Maked
else
tmpdepfile1=$dir$base.d
tmpdepfile2=$dir$base.d
"$@" +Maked
fi
stat=$?
if test $stat -ne 0; then
rm -f "$tmpdepfile1" "$tmpdepfile2"
exit $stat
fi
for tmpdepfile in "$tmpdepfile1" "$tmpdepfile2"
do
test -f "$tmpdepfile" && break
done
if test -f "$tmpdepfile"; then
sed -e "s,^.*\.[$lower]*:,$object:," "$tmpdepfile" > "$depfile"
# Add 'dependent.h:' lines.
sed -ne '2,${
s/^ *//
s/ \\*$//
s/$/:/
p
}' "$tmpdepfile" >> "$depfile"
else
make_dummy_depfile
fi
rm -f "$tmpdepfile" "$tmpdepfile2"
;;
tru64)
# The Tru64 compiler uses -MD to generate dependencies as a side
# effect. 'cc -MD -o foo.o ...' puts the dependencies into 'foo.o.d'.
# At least on Alpha/Redhat 6.1, Compaq CCC V6.2-504 seems to put
# dependencies in 'foo.d' instead, so we check for that too.
# Subdirectories are respected.
set_dir_from "$object"
set_base_from "$object"
if test "$libtool" = yes; then
# Libtool generates 2 separate objects for the 2 libraries. These
# two compilations output dependencies in $dir.libs/$base.o.d and
# in $dir$base.o.d. We have to check for both files, because
# one of the two compilations can be disabled. We should prefer
# $dir$base.o.d over $dir.libs/$base.o.d because the latter is
# automatically cleaned when .libs/ is deleted, while ignoring
# the former would cause a distcleancheck panic.
tmpdepfile1=$dir$base.o.d # libtool 1.5
tmpdepfile2=$dir.libs/$base.o.d # Likewise.
tmpdepfile3=$dir.libs/$base.d # Compaq CCC V6.2-504
"$@" -Wc,-MD
else
tmpdepfile1=$dir$base.d
tmpdepfile2=$dir$base.d
tmpdepfile3=$dir$base.d
"$@" -MD
fi
stat=$?
if test $stat -ne 0; then
rm -f "$tmpdepfile1" "$tmpdepfile2" "$tmpdepfile3"
exit $stat
fi
for tmpdepfile in "$tmpdepfile1" "$tmpdepfile2" "$tmpdepfile3"
do
test -f "$tmpdepfile" && break
done
# Same post-processing that is required for AIX mode.
aix_post_process_depfile
;;
msvc7)
if test "$libtool" = yes; then
showIncludes=-Wc,-showIncludes
else
showIncludes=-showIncludes
fi
"$@" $showIncludes > "$tmpdepfile"
stat=$?
grep -v '^Note: including file: ' "$tmpdepfile"
if test $stat -ne 0; then
rm -f "$tmpdepfile"
exit $stat
fi
rm -f "$depfile"
echo "$object : \\" > "$depfile"
# The first sed program below extracts the file names and escapes
# backslashes for cygpath. The second sed program outputs the file
# name when reading, but also accumulates all include files in the
# hold buffer in order to output them again at the end. This only
# works with sed implementations that can handle large buffers.
sed < "$tmpdepfile" -n '
/^Note: including file: *\(.*\)/ {
s//\1/
s/\\/\\\\/g
p
}' | $cygpath_u | sort -u | sed -n '
s/ /\\ /g
s/\(.*\)/'"$tab"'\1 \\/p
s/.\(.*\) \\/\1:/
H
$ {
s/.*/'"$tab"'/
G
p
}' >> "$depfile"
echo >> "$depfile" # make sure the fragment doesn't end with a backslash
rm -f "$tmpdepfile"
;;
msvc7msys)
# This case exists only to let depend.m4 do its work. It works by
# looking at the text of this script. This case will never be run,
# since it is checked for above.
exit 1
;;
#nosideeffect)
# This comment above is used by automake to tell side-effect
# dependency tracking mechanisms from slower ones.
dashmstdout)
# Important note: in order to support this mode, a compiler *must*
# always write the preprocessed file to stdout, regardless of -o.
"$@" || exit $?
# Remove the call to Libtool.
if test "$libtool" = yes; then
while test "X$1" != 'X--mode=compile'; do
shift
done
shift
fi
# Remove '-o $object'.
IFS=" "
for arg
do
case $arg in
-o)
shift
;;
$object)
shift
;;
*)
set fnord "$@" "$arg"
shift # fnord
shift # $arg
;;
esac
done
test -z "$dashmflag" && dashmflag=-M
# Require at least two characters before searching for ':'
# in the target name. This is to cope with DOS-style filenames:
# a dependency such as 'c:/foo/bar' could be seen as target 'c' otherwise.
"$@" $dashmflag |
sed "s|^[$tab ]*[^:$tab ][^:][^:]*:[$tab ]*|$object: |" > "$tmpdepfile"
rm -f "$depfile"
cat < "$tmpdepfile" > "$depfile"
# Some versions of the HPUX 10.20 sed can't process this sed invocation
# correctly. Breaking it into two sed invocations is a workaround.
tr ' ' "$nl" < "$tmpdepfile" \
| sed -e 's/^\\$//' -e '/^$/d' -e '/:$/d' \
| sed -e 's/$/ :/' >> "$depfile"
rm -f "$tmpdepfile"
;;
dashXmstdout)
# This case only exists to satisfy depend.m4. It is never actually
# run, as this mode is specially recognized in the preamble.
exit 1
;;
makedepend)
"$@" || exit $?
# Remove any Libtool call
if test "$libtool" = yes; then
while test "X$1" != 'X--mode=compile'; do
shift
done
shift
fi
# X makedepend
shift
cleared=no eat=no
for arg
do
case $cleared in
no)
set ""; shift
cleared=yes ;;
esac
if test $eat = yes; then
eat=no
continue
fi
case "$arg" in
-D*|-I*)
set fnord "$@" "$arg"; shift ;;
# Strip any option that makedepend may not understand. Remove
# the object too, otherwise makedepend will parse it as a source file.
-arch)
eat=yes ;;
-*|$object)
;;
*)
set fnord "$@" "$arg"; shift ;;
esac
done
obj_suffix=`echo "$object" | sed 's/^.*\././'`
touch "$tmpdepfile"
${MAKEDEPEND-makedepend} -o"$obj_suffix" -f"$tmpdepfile" "$@"
rm -f "$depfile"
# makedepend may prepend the VPATH from the source file name to the object.
# No need to regex-escape $object, excess matching of '.' is harmless.
sed "s|^.*\($object *:\)|\1|" "$tmpdepfile" > "$depfile"
# Some versions of the HPUX 10.20 sed can't process the last invocation
# correctly. Breaking it into two sed invocations is a workaround.
sed '1,2d' "$tmpdepfile" \
| tr ' ' "$nl" \
| sed -e 's/^\\$//' -e '/^$/d' -e '/:$/d' \
| sed -e 's/$/ :/' >> "$depfile"
rm -f "$tmpdepfile" "$tmpdepfile".bak
;;
cpp)
# Important note: in order to support this mode, a compiler *must*
# always write the preprocessed file to stdout.
"$@" || exit $?
# Remove the call to Libtool.
if test "$libtool" = yes; then
while test "X$1" != 'X--mode=compile'; do
shift
done
shift
fi
# Remove '-o $object'.
IFS=" "
for arg
do
case $arg in
-o)
shift
;;
$object)
shift
;;
*)
set fnord "$@" "$arg"
shift # fnord
shift # $arg
;;
esac
done
"$@" -E \
| sed -n -e '/^# [0-9][0-9]* "\([^"]*\)".*/ s:: \1 \\:p' \
-e '/^#line [0-9][0-9]* "\([^"]*\)".*/ s:: \1 \\:p' \
| sed '$ s: \\$::' > "$tmpdepfile"
rm -f "$depfile"
echo "$object : \\" > "$depfile"
cat < "$tmpdepfile" >> "$depfile"
sed < "$tmpdepfile" '/^$/d;s/^ //;s/ \\$//;s/$/ :/' >> "$depfile"
rm -f "$tmpdepfile"
;;
msvisualcpp)
# Important note: in order to support this mode, a compiler *must*
# always write the preprocessed file to stdout.
"$@" || exit $?
# Remove the call to Libtool.
if test "$libtool" = yes; then
while test "X$1" != 'X--mode=compile'; do
shift
done
shift
fi
IFS=" "
for arg
do
case "$arg" in
-o)
shift
;;
$object)
shift
;;
"-Gm"|"/Gm"|"-Gi"|"/Gi"|"-ZI"|"/ZI")
set fnord "$@"
shift
shift
;;
*)
set fnord "$@" "$arg"
shift
shift
;;
esac
done
"$@" -E 2>/dev/null |
sed -n '/^#line [0-9][0-9]* "\([^"]*\)"/ s::\1:p' | $cygpath_u | sort -u > "$tmpdepfile"
rm -f "$depfile"
echo "$object : \\" > "$depfile"
sed < "$tmpdepfile" -n -e 's% %\\ %g' -e '/^\(.*\)$/ s::'"$tab"'\1 \\:p' >> "$depfile"
echo "$tab" >> "$depfile"
sed < "$tmpdepfile" -n -e 's% %\\ %g' -e '/^\(.*\)$/ s::\1\::p' >> "$depfile"
rm -f "$tmpdepfile"
;;
msvcmsys)
# This case exists only to let depend.m4 do its work. It works by
# looking at the text of this script. This case will never be run,
# since it is checked for above.
exit 1
;;
none)
exec "$@"
;;
*)
echo "Unknown depmode $depmode" 1>&2
exit 1
;;
esac
exit 0
# Local Variables:
# mode: shell-script
# sh-indentation: 2
# eval: (add-hook 'write-file-hooks 'time-stamp)
# time-stamp-start: "scriptversion="
# time-stamp-format: "%:y-%02m-%02d.%02H"
# time-stamp-time-zone: "UTC"
# time-stamp-end: "; # UTC"
# End:
diff --git a/build-aux/ltmain.sh b/build-aux/ltmain.sh
index f8c3614..c3bcdc0 100644
--- a/build-aux/ltmain.sh
+++ b/build-aux/ltmain.sh
@@ -1,9677 +1,9677 @@
# libtool (GNU libtool) 2.4.2
# Written by Gordon Matzigkeit , 1996
# Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2003, 2004, 2005, 2006,
# 2007, 2008, 2009, 2010, 2011 Free Software Foundation, Inc.
# This is free software; see the source for copying conditions. There is NO
# warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
# GNU Libtool is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2 of the License, or
# (at your option) any later version.
#
# As a special exception to the GNU General Public License,
# if you distribute this file as part of a program or library that
# is built using GNU Libtool, you may include this file under the
# same distribution terms that you use for the rest of that program.
#
# GNU Libtool is distributed in the hope that it will be useful, but
# WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
# General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with GNU Libtool; see the file COPYING. If not, a copy
-# can be downloaded from http://www.gnu.org/licenses/gpl.html,
+# can be downloaded from https://www.gnu.org/licenses/gpl.html,
# or obtained by writing to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
# Usage: $progname [OPTION]... [MODE-ARG]...
#
# Provide generalized library-building support services.
#
# --config show all configuration variables
# --debug enable verbose shell tracing
# -n, --dry-run display commands without modifying any files
# --features display basic configuration information and exit
# --mode=MODE use operation mode MODE
# --preserve-dup-deps don't remove duplicate dependency libraries
# --quiet, --silent don't print informational messages
# --no-quiet, --no-silent
# print informational messages (default)
# --no-warn don't display warning messages
# --tag=TAG use configuration variables from tag TAG
# -v, --verbose print more informational messages than default
# --no-verbose don't print the extra informational messages
# --version print version information
# -h, --help, --help-all print short, long, or detailed help message
#
# MODE must be one of the following:
#
# clean remove files from the build directory
# compile compile a source file into a libtool object
# execute automatically set library path, then run a program
# finish complete the installation of libtool libraries
# install install libraries or executables
# link create a library or an executable
# uninstall remove libraries from an installed directory
#
# MODE-ARGS vary depending on the MODE. When passed as first option,
# `--mode=MODE' may be abbreviated as `MODE' or a unique abbreviation of that.
# Try `$progname --help --mode=MODE' for a more detailed description of MODE.
#
# When reporting a bug, please describe a test case to reproduce it and
# include the following information:
#
# host-triplet: $host
# shell: $SHELL
# compiler: $LTCC
# compiler flags: $LTCFLAGS
# linker: $LD (gnu? $with_gnu_ld)
# $progname: (GNU libtool) 2.4.2 Debian-2.4.2-1
# automake: $automake_version
# autoconf: $autoconf_version
#
# Report bugs to .
-# GNU libtool home page: .
-# General help using GNU software: .
+# GNU libtool home page: .
+# General help using GNU software: .
PROGRAM=libtool
PACKAGE=libtool
VERSION="2.4.2 Debian-2.4.2-1"
TIMESTAMP=""
package_revision=1.3337
# Be Bourne compatible
if test -n "${ZSH_VERSION+set}" && (emulate sh) >/dev/null 2>&1; then
emulate sh
NULLCMD=:
# Zsh 3.x and 4.x performs word splitting on ${1+"$@"}, which
# is contrary to our usage. Disable this feature.
alias -g '${1+"$@"}'='"$@"'
setopt NO_GLOB_SUBST
else
case `(set -o) 2>/dev/null` in *posix*) set -o posix;; esac
fi
BIN_SH=xpg4; export BIN_SH # for Tru64
DUALCASE=1; export DUALCASE # for MKS sh
# A function that is used when there is no print builtin or printf.
func_fallback_echo ()
{
eval 'cat <<_LTECHO_EOF
$1
_LTECHO_EOF'
}
# NLS nuisances: We save the old values to restore during execute mode.
lt_user_locale=
lt_safe_locale=
for lt_var in LANG LANGUAGE LC_ALL LC_CTYPE LC_COLLATE LC_MESSAGES
do
eval "if test \"\${$lt_var+set}\" = set; then
save_$lt_var=\$$lt_var
$lt_var=C
export $lt_var
lt_user_locale=\"$lt_var=\\\$save_\$lt_var; \$lt_user_locale\"
lt_safe_locale=\"$lt_var=C; \$lt_safe_locale\"
fi"
done
LC_ALL=C
LANGUAGE=C
export LANGUAGE LC_ALL
$lt_unset CDPATH
# Work around backward compatibility issue on IRIX 6.5. On IRIX 6.4+, sh
# is ksh but when the shell is invoked as "sh" and the current value of
# the _XPG environment variable is not equal to 1 (one), the special
# positional parameter $0, within a function call, is the name of the
# function.
progpath="$0"
: ${CP="cp -f"}
test "${ECHO+set}" = set || ECHO=${as_echo-'printf %s\n'}
: ${MAKE="make"}
: ${MKDIR="mkdir"}
: ${MV="mv -f"}
: ${RM="rm -f"}
: ${SHELL="${CONFIG_SHELL-/bin/sh}"}
: ${Xsed="$SED -e 1s/^X//"}
# Global variables:
EXIT_SUCCESS=0
EXIT_FAILURE=1
EXIT_MISMATCH=63 # $? = 63 is used to indicate version mismatch to missing.
EXIT_SKIP=77 # $? = 77 is used to indicate a skipped test to automake.
exit_status=$EXIT_SUCCESS
# Make sure IFS has a sensible default
lt_nl='
'
IFS=" $lt_nl"
dirname="s,/[^/]*$,,"
basename="s,^.*/,,"
# func_dirname file append nondir_replacement
# Compute the dirname of FILE. If nonempty, add APPEND to the result,
# otherwise set result to NONDIR_REPLACEMENT.
func_dirname ()
{
func_dirname_result=`$ECHO "${1}" | $SED "$dirname"`
if test "X$func_dirname_result" = "X${1}"; then
func_dirname_result="${3}"
else
func_dirname_result="$func_dirname_result${2}"
fi
} # func_dirname may be replaced by extended shell implementation
# func_basename file
func_basename ()
{
func_basename_result=`$ECHO "${1}" | $SED "$basename"`
} # func_basename may be replaced by extended shell implementation
# func_dirname_and_basename file append nondir_replacement
# perform func_basename and func_dirname in a single function
# call:
# dirname: Compute the dirname of FILE. If nonempty,
# add APPEND to the result, otherwise set result
# to NONDIR_REPLACEMENT.
# value returned in "$func_dirname_result"
# basename: Compute filename of FILE.
# value retuned in "$func_basename_result"
# Implementation must be kept synchronized with func_dirname
# and func_basename. For efficiency, we do not delegate to
# those functions but instead duplicate the functionality here.
func_dirname_and_basename ()
{
# Extract subdirectory from the argument.
func_dirname_result=`$ECHO "${1}" | $SED -e "$dirname"`
if test "X$func_dirname_result" = "X${1}"; then
func_dirname_result="${3}"
else
func_dirname_result="$func_dirname_result${2}"
fi
func_basename_result=`$ECHO "${1}" | $SED -e "$basename"`
} # func_dirname_and_basename may be replaced by extended shell implementation
# func_stripname prefix suffix name
# strip PREFIX and SUFFIX off of NAME.
# PREFIX and SUFFIX must not contain globbing or regex special
# characters, hashes, percent signs, but SUFFIX may contain a leading
# dot (in which case that matches only a dot).
# func_strip_suffix prefix name
func_stripname ()
{
case ${2} in
.*) func_stripname_result=`$ECHO "${3}" | $SED "s%^${1}%%; s%\\\\${2}\$%%"`;;
*) func_stripname_result=`$ECHO "${3}" | $SED "s%^${1}%%; s%${2}\$%%"`;;
esac
} # func_stripname may be replaced by extended shell implementation
# These SED scripts presuppose an absolute path with a trailing slash.
pathcar='s,^/\([^/]*\).*$,\1,'
pathcdr='s,^/[^/]*,,'
removedotparts=':dotsl
s@/\./@/@g
t dotsl
s,/\.$,/,'
collapseslashes='s@/\{1,\}@/@g'
finalslash='s,/*$,/,'
# func_normal_abspath PATH
# Remove doubled-up and trailing slashes, "." path components,
# and cancel out any ".." path components in PATH after making
# it an absolute path.
# value returned in "$func_normal_abspath_result"
func_normal_abspath ()
{
# Start from root dir and reassemble the path.
func_normal_abspath_result=
func_normal_abspath_tpath=$1
func_normal_abspath_altnamespace=
case $func_normal_abspath_tpath in
"")
# Empty path, that just means $cwd.
func_stripname '' '/' "`pwd`"
func_normal_abspath_result=$func_stripname_result
return
;;
# The next three entries are used to spot a run of precisely
# two leading slashes without using negated character classes;
# we take advantage of case's first-match behaviour.
///*)
# Unusual form of absolute path, do nothing.
;;
//*)
# Not necessarily an ordinary path; POSIX reserves leading '//'
# and for example Cygwin uses it to access remote file shares
# over CIFS/SMB, so we conserve a leading double slash if found.
func_normal_abspath_altnamespace=/
;;
/*)
# Absolute path, do nothing.
;;
*)
# Relative path, prepend $cwd.
func_normal_abspath_tpath=`pwd`/$func_normal_abspath_tpath
;;
esac
# Cancel out all the simple stuff to save iterations. We also want
# the path to end with a slash for ease of parsing, so make sure
# there is one (and only one) here.
func_normal_abspath_tpath=`$ECHO "$func_normal_abspath_tpath" | $SED \
-e "$removedotparts" -e "$collapseslashes" -e "$finalslash"`
while :; do
# Processed it all yet?
if test "$func_normal_abspath_tpath" = / ; then
# If we ascended to the root using ".." the result may be empty now.
if test -z "$func_normal_abspath_result" ; then
func_normal_abspath_result=/
fi
break
fi
func_normal_abspath_tcomponent=`$ECHO "$func_normal_abspath_tpath" | $SED \
-e "$pathcar"`
func_normal_abspath_tpath=`$ECHO "$func_normal_abspath_tpath" | $SED \
-e "$pathcdr"`
# Figure out what to do with it
case $func_normal_abspath_tcomponent in
"")
# Trailing empty path component, ignore it.
;;
..)
# Parent dir; strip last assembled component from result.
func_dirname "$func_normal_abspath_result"
func_normal_abspath_result=$func_dirname_result
;;
*)
# Actual path component, append it.
func_normal_abspath_result=$func_normal_abspath_result/$func_normal_abspath_tcomponent
;;
esac
done
# Restore leading double-slash if one was found on entry.
func_normal_abspath_result=$func_normal_abspath_altnamespace$func_normal_abspath_result
}
# func_relative_path SRCDIR DSTDIR
# generates a relative path from SRCDIR to DSTDIR, with a trailing
# slash if non-empty, suitable for immediately appending a filename
# without needing to append a separator.
# value returned in "$func_relative_path_result"
func_relative_path ()
{
func_relative_path_result=
func_normal_abspath "$1"
func_relative_path_tlibdir=$func_normal_abspath_result
func_normal_abspath "$2"
func_relative_path_tbindir=$func_normal_abspath_result
# Ascend the tree starting from libdir
while :; do
# check if we have found a prefix of bindir
case $func_relative_path_tbindir in
$func_relative_path_tlibdir)
# found an exact match
func_relative_path_tcancelled=
break
;;
$func_relative_path_tlibdir*)
# found a matching prefix
func_stripname "$func_relative_path_tlibdir" '' "$func_relative_path_tbindir"
func_relative_path_tcancelled=$func_stripname_result
if test -z "$func_relative_path_result"; then
func_relative_path_result=.
fi
break
;;
*)
func_dirname $func_relative_path_tlibdir
func_relative_path_tlibdir=${func_dirname_result}
if test "x$func_relative_path_tlibdir" = x ; then
# Have to descend all the way to the root!
func_relative_path_result=../$func_relative_path_result
func_relative_path_tcancelled=$func_relative_path_tbindir
break
fi
func_relative_path_result=../$func_relative_path_result
;;
esac
done
# Now calculate path; take care to avoid doubling-up slashes.
func_stripname '' '/' "$func_relative_path_result"
func_relative_path_result=$func_stripname_result
func_stripname '/' '/' "$func_relative_path_tcancelled"
if test "x$func_stripname_result" != x ; then
func_relative_path_result=${func_relative_path_result}/${func_stripname_result}
fi
# Normalisation. If bindir is libdir, return empty string,
# else relative path ending with a slash; either way, target
# file name can be directly appended.
if test ! -z "$func_relative_path_result"; then
func_stripname './' '' "$func_relative_path_result/"
func_relative_path_result=$func_stripname_result
fi
}
# The name of this program:
func_dirname_and_basename "$progpath"
progname=$func_basename_result
# Make sure we have an absolute path for reexecution:
case $progpath in
[\\/]*|[A-Za-z]:\\*) ;;
*[\\/]*)
progdir=$func_dirname_result
progdir=`cd "$progdir" && pwd`
progpath="$progdir/$progname"
;;
*)
save_IFS="$IFS"
IFS=${PATH_SEPARATOR-:}
for progdir in $PATH; do
IFS="$save_IFS"
test -x "$progdir/$progname" && break
done
IFS="$save_IFS"
test -n "$progdir" || progdir=`pwd`
progpath="$progdir/$progname"
;;
esac
# Sed substitution that helps us do robust quoting. It backslashifies
# metacharacters that are still active within double-quoted strings.
Xsed="${SED}"' -e 1s/^X//'
sed_quote_subst='s/\([`"$\\]\)/\\\1/g'
# Same as above, but do not quote variable references.
double_quote_subst='s/\(["`\\]\)/\\\1/g'
# Sed substitution that turns a string into a regex matching for the
# string literally.
sed_make_literal_regex='s,[].[^$\\*\/],\\&,g'
# Sed substitution that converts a w32 file name or path
# which contains forward slashes, into one that contains
# (escaped) backslashes. A very naive implementation.
lt_sed_naive_backslashify='s|\\\\*|\\|g;s|/|\\|g;s|\\|\\\\|g'
# Sed substitution to remove simple comments and empty
# lines from a Windows .def file.
sed_uncomment_deffile='/^;/d; /^[ ]*$/d'
# Re-`\' parameter expansions in output of double_quote_subst that were
# `\'-ed in input to the same. If an odd number of `\' preceded a '$'
# in input to double_quote_subst, that '$' was protected from expansion.
# Since each input `\' is now two `\'s, look for any number of runs of
# four `\'s followed by two `\'s and then a '$'. `\' that '$'.
bs='\\'
bs2='\\\\'
bs4='\\\\\\\\'
dollar='\$'
sed_double_backslash="\
s/$bs4/&\\
/g
s/^$bs2$dollar/$bs&/
s/\\([^$bs]\\)$bs2$dollar/\\1$bs2$bs$dollar/g
s/\n//g"
# Standard options:
opt_dry_run=false
opt_help=false
opt_quiet=false
opt_verbose=false
opt_warning=:
# func_echo arg...
# Echo program name prefixed message, along with the current mode
# name if it has been set yet.
func_echo ()
{
$ECHO "$progname: ${opt_mode+$opt_mode: }$*"
}
# func_verbose arg...
# Echo program name prefixed message in verbose mode only.
func_verbose ()
{
$opt_verbose && func_echo ${1+"$@"}
# A bug in bash halts the script if the last line of a function
# fails when set -e is in force, so we need another command to
# work around that:
:
}
# func_echo_all arg...
# Invoke $ECHO with all args, space-separated.
func_echo_all ()
{
$ECHO "$*"
}
# func_error arg...
# Echo program name prefixed message to standard error.
func_error ()
{
$ECHO "$progname: ${opt_mode+$opt_mode: }"${1+"$@"} 1>&2
}
# func_warning arg...
# Echo program name prefixed warning message to standard error.
func_warning ()
{
$opt_warning && $ECHO "$progname: ${opt_mode+$opt_mode: }warning: "${1+"$@"} 1>&2
# bash bug again:
:
}
# func_fatal_error arg...
# Echo program name prefixed message to standard error, and exit.
func_fatal_error ()
{
func_error ${1+"$@"}
exit $EXIT_FAILURE
}
# func_fatal_help arg...
# Echo program name prefixed message to standard error, followed by
# a help hint, and exit.
func_fatal_help ()
{
func_error ${1+"$@"}
func_fatal_error "$help"
}
help="Try \`$progname --help' for more information." ## default
# func_grep expression filename
# Check whether EXPRESSION matches any line of FILENAME, without output.
func_grep ()
{
$GREP "$1" "$2" >/dev/null 2>&1
}
# func_mkdir_p directory-path
# Make sure the entire path to DIRECTORY-PATH is available.
func_mkdir_p ()
{
my_directory_path="$1"
my_dir_list=
if test -n "$my_directory_path" && test "$opt_dry_run" != ":"; then
# Protect directory names starting with `-'
case $my_directory_path in
-*) my_directory_path="./$my_directory_path" ;;
esac
# While some portion of DIR does not yet exist...
while test ! -d "$my_directory_path"; do
# ...make a list in topmost first order. Use a colon delimited
# list incase some portion of path contains whitespace.
my_dir_list="$my_directory_path:$my_dir_list"
# If the last portion added has no slash in it, the list is done
case $my_directory_path in */*) ;; *) break ;; esac
# ...otherwise throw away the child directory and loop
my_directory_path=`$ECHO "$my_directory_path" | $SED -e "$dirname"`
done
my_dir_list=`$ECHO "$my_dir_list" | $SED 's,:*$,,'`
save_mkdir_p_IFS="$IFS"; IFS=':'
for my_dir in $my_dir_list; do
IFS="$save_mkdir_p_IFS"
# mkdir can fail with a `File exist' error if two processes
# try to create one of the directories concurrently. Don't
# stop in that case!
$MKDIR "$my_dir" 2>/dev/null || :
done
IFS="$save_mkdir_p_IFS"
# Bail out if we (or some other process) failed to create a directory.
test -d "$my_directory_path" || \
func_fatal_error "Failed to create \`$1'"
fi
}
# func_mktempdir [string]
# Make a temporary directory that won't clash with other running
# libtool processes, and avoids race conditions if possible. If
# given, STRING is the basename for that directory.
func_mktempdir ()
{
my_template="${TMPDIR-/tmp}/${1-$progname}"
if test "$opt_dry_run" = ":"; then
# Return a directory name, but don't create it in dry-run mode
my_tmpdir="${my_template}-$$"
else
# If mktemp works, use that first and foremost
my_tmpdir=`mktemp -d "${my_template}-XXXXXXXX" 2>/dev/null`
if test ! -d "$my_tmpdir"; then
# Failing that, at least try and use $RANDOM to avoid a race
my_tmpdir="${my_template}-${RANDOM-0}$$"
save_mktempdir_umask=`umask`
umask 0077
$MKDIR "$my_tmpdir"
umask $save_mktempdir_umask
fi
# If we're not in dry-run mode, bomb out on failure
test -d "$my_tmpdir" || \
func_fatal_error "cannot create temporary directory \`$my_tmpdir'"
fi
$ECHO "$my_tmpdir"
}
# func_quote_for_eval arg
# Aesthetically quote ARG to be evaled later.
# This function returns two values: FUNC_QUOTE_FOR_EVAL_RESULT
# is double-quoted, suitable for a subsequent eval, whereas
# FUNC_QUOTE_FOR_EVAL_UNQUOTED_RESULT has merely all characters
# which are still active within double quotes backslashified.
func_quote_for_eval ()
{
case $1 in
*[\\\`\"\$]*)
func_quote_for_eval_unquoted_result=`$ECHO "$1" | $SED "$sed_quote_subst"` ;;
*)
func_quote_for_eval_unquoted_result="$1" ;;
esac
case $func_quote_for_eval_unquoted_result in
# Double-quote args containing shell metacharacters to delay
# word splitting, command substitution and and variable
# expansion for a subsequent eval.
# Many Bourne shells cannot handle close brackets correctly
# in scan sets, so we specify it separately.
*[\[\~\#\^\&\*\(\)\{\}\|\;\<\>\?\'\ \ ]*|*]*|"")
func_quote_for_eval_result="\"$func_quote_for_eval_unquoted_result\""
;;
*)
func_quote_for_eval_result="$func_quote_for_eval_unquoted_result"
esac
}
# func_quote_for_expand arg
# Aesthetically quote ARG to be evaled later; same as above,
# but do not quote variable references.
func_quote_for_expand ()
{
case $1 in
*[\\\`\"]*)
my_arg=`$ECHO "$1" | $SED \
-e "$double_quote_subst" -e "$sed_double_backslash"` ;;
*)
my_arg="$1" ;;
esac
case $my_arg in
# Double-quote args containing shell metacharacters to delay
# word splitting and command substitution for a subsequent eval.
# Many Bourne shells cannot handle close brackets correctly
# in scan sets, so we specify it separately.
*[\[\~\#\^\&\*\(\)\{\}\|\;\<\>\?\'\ \ ]*|*]*|"")
my_arg="\"$my_arg\""
;;
esac
func_quote_for_expand_result="$my_arg"
}
# func_show_eval cmd [fail_exp]
# Unless opt_silent is true, then output CMD. Then, if opt_dryrun is
# not true, evaluate CMD. If the evaluation of CMD fails, and FAIL_EXP
# is given, then evaluate it.
func_show_eval ()
{
my_cmd="$1"
my_fail_exp="${2-:}"
${opt_silent-false} || {
func_quote_for_expand "$my_cmd"
eval "func_echo $func_quote_for_expand_result"
}
if ${opt_dry_run-false}; then :; else
eval "$my_cmd"
my_status=$?
if test "$my_status" -eq 0; then :; else
eval "(exit $my_status); $my_fail_exp"
fi
fi
}
# func_show_eval_locale cmd [fail_exp]
# Unless opt_silent is true, then output CMD. Then, if opt_dryrun is
# not true, evaluate CMD. If the evaluation of CMD fails, and FAIL_EXP
# is given, then evaluate it. Use the saved locale for evaluation.
func_show_eval_locale ()
{
my_cmd="$1"
my_fail_exp="${2-:}"
${opt_silent-false} || {
func_quote_for_expand "$my_cmd"
eval "func_echo $func_quote_for_expand_result"
}
if ${opt_dry_run-false}; then :; else
eval "$lt_user_locale
$my_cmd"
my_status=$?
eval "$lt_safe_locale"
if test "$my_status" -eq 0; then :; else
eval "(exit $my_status); $my_fail_exp"
fi
fi
}
# func_tr_sh
# Turn $1 into a string suitable for a shell variable name.
# Result is stored in $func_tr_sh_result. All characters
# not in the set a-zA-Z0-9_ are replaced with '_'. Further,
# if $1 begins with a digit, a '_' is prepended as well.
func_tr_sh ()
{
case $1 in
[0-9]* | *[!a-zA-Z0-9_]*)
func_tr_sh_result=`$ECHO "$1" | $SED 's/^\([0-9]\)/_\1/; s/[^a-zA-Z0-9_]/_/g'`
;;
* )
func_tr_sh_result=$1
;;
esac
}
# func_version
# Echo version message to standard output and exit.
func_version ()
{
$opt_debug
$SED -n '/(C)/!b go
:more
/\./!{
N
s/\n# / /
b more
}
:go
/^# '$PROGRAM' (GNU /,/# warranty; / {
s/^# //
s/^# *$//
s/\((C)\)[ 0-9,-]*\( [1-9][0-9]*\)/\1\2/
p
}' < "$progpath"
exit $?
}
# func_usage
# Echo short help message to standard output and exit.
func_usage ()
{
$opt_debug
$SED -n '/^# Usage:/,/^# *.*--help/ {
s/^# //
s/^# *$//
s/\$progname/'$progname'/
p
}' < "$progpath"
echo
$ECHO "run \`$progname --help | more' for full usage"
exit $?
}
# func_help [NOEXIT]
# Echo long help message to standard output and exit,
# unless 'noexit' is passed as argument.
func_help ()
{
$opt_debug
$SED -n '/^# Usage:/,/# Report bugs to/ {
:print
s/^# //
s/^# *$//
s*\$progname*'$progname'*
s*\$host*'"$host"'*
s*\$SHELL*'"$SHELL"'*
s*\$LTCC*'"$LTCC"'*
s*\$LTCFLAGS*'"$LTCFLAGS"'*
s*\$LD*'"$LD"'*
s/\$with_gnu_ld/'"$with_gnu_ld"'/
s/\$automake_version/'"`(${AUTOMAKE-automake} --version) 2>/dev/null |$SED 1q`"'/
s/\$autoconf_version/'"`(${AUTOCONF-autoconf} --version) 2>/dev/null |$SED 1q`"'/
p
d
}
/^# .* home page:/b print
/^# General help using/b print
' < "$progpath"
ret=$?
if test -z "$1"; then
exit $ret
fi
}
# func_missing_arg argname
# Echo program name prefixed message to standard error and set global
# exit_cmd.
func_missing_arg ()
{
$opt_debug
func_error "missing argument for $1."
exit_cmd=exit
}
# func_split_short_opt shortopt
# Set func_split_short_opt_name and func_split_short_opt_arg shell
# variables after splitting SHORTOPT after the 2nd character.
func_split_short_opt ()
{
my_sed_short_opt='1s/^\(..\).*$/\1/;q'
my_sed_short_rest='1s/^..\(.*\)$/\1/;q'
func_split_short_opt_name=`$ECHO "$1" | $SED "$my_sed_short_opt"`
func_split_short_opt_arg=`$ECHO "$1" | $SED "$my_sed_short_rest"`
} # func_split_short_opt may be replaced by extended shell implementation
# func_split_long_opt longopt
# Set func_split_long_opt_name and func_split_long_opt_arg shell
# variables after splitting LONGOPT at the `=' sign.
func_split_long_opt ()
{
my_sed_long_opt='1s/^\(--[^=]*\)=.*/\1/;q'
my_sed_long_arg='1s/^--[^=]*=//'
func_split_long_opt_name=`$ECHO "$1" | $SED "$my_sed_long_opt"`
func_split_long_opt_arg=`$ECHO "$1" | $SED "$my_sed_long_arg"`
} # func_split_long_opt may be replaced by extended shell implementation
exit_cmd=:
magic="%%%MAGIC variable%%%"
magic_exe="%%%MAGIC EXE variable%%%"
# Global variables.
nonopt=
preserve_args=
lo2o="s/\\.lo\$/.${objext}/"
o2lo="s/\\.${objext}\$/.lo/"
extracted_archives=
extracted_serial=0
# If this variable is set in any of the actions, the command in it
# will be execed at the end. This prevents here-documents from being
# left over by shells.
exec_cmd=
# func_append var value
# Append VALUE to the end of shell variable VAR.
func_append ()
{
eval "${1}=\$${1}\${2}"
} # func_append may be replaced by extended shell implementation
# func_append_quoted var value
# Quote VALUE and append to the end of shell variable VAR, separated
# by a space.
func_append_quoted ()
{
func_quote_for_eval "${2}"
eval "${1}=\$${1}\\ \$func_quote_for_eval_result"
} # func_append_quoted may be replaced by extended shell implementation
# func_arith arithmetic-term...
func_arith ()
{
func_arith_result=`expr "${@}"`
} # func_arith may be replaced by extended shell implementation
# func_len string
# STRING may not start with a hyphen.
func_len ()
{
func_len_result=`expr "${1}" : ".*" 2>/dev/null || echo $max_cmd_len`
} # func_len may be replaced by extended shell implementation
# func_lo2o object
func_lo2o ()
{
func_lo2o_result=`$ECHO "${1}" | $SED "$lo2o"`
} # func_lo2o may be replaced by extended shell implementation
# func_xform libobj-or-source
func_xform ()
{
func_xform_result=`$ECHO "${1}" | $SED 's/\.[^.]*$/.lo/'`
} # func_xform may be replaced by extended shell implementation
# func_fatal_configuration arg...
# Echo program name prefixed message to standard error, followed by
# a configuration failure hint, and exit.
func_fatal_configuration ()
{
func_error ${1+"$@"}
func_error "See the $PACKAGE documentation for more information."
func_fatal_error "Fatal configuration error."
}
# func_config
# Display the configuration for all the tags in this script.
func_config ()
{
re_begincf='^# ### BEGIN LIBTOOL'
re_endcf='^# ### END LIBTOOL'
# Default configuration.
$SED "1,/$re_begincf CONFIG/d;/$re_endcf CONFIG/,\$d" < "$progpath"
# Now print the configurations for the tags.
for tagname in $taglist; do
$SED -n "/$re_begincf TAG CONFIG: $tagname\$/,/$re_endcf TAG CONFIG: $tagname\$/p" < "$progpath"
done
exit $?
}
# func_features
# Display the features supported by this script.
func_features ()
{
echo "host: $host"
if test "$build_libtool_libs" = yes; then
echo "enable shared libraries"
else
echo "disable shared libraries"
fi
if test "$build_old_libs" = yes; then
echo "enable static libraries"
else
echo "disable static libraries"
fi
exit $?
}
# func_enable_tag tagname
# Verify that TAGNAME is valid, and either flag an error and exit, or
# enable the TAGNAME tag. We also add TAGNAME to the global $taglist
# variable here.
func_enable_tag ()
{
# Global variable:
tagname="$1"
re_begincf="^# ### BEGIN LIBTOOL TAG CONFIG: $tagname\$"
re_endcf="^# ### END LIBTOOL TAG CONFIG: $tagname\$"
sed_extractcf="/$re_begincf/,/$re_endcf/p"
# Validate tagname.
case $tagname in
*[!-_A-Za-z0-9,/]*)
func_fatal_error "invalid tag name: $tagname"
;;
esac
# Don't test for the "default" C tag, as we know it's
# there but not specially marked.
case $tagname in
CC) ;;
*)
if $GREP "$re_begincf" "$progpath" >/dev/null 2>&1; then
taglist="$taglist $tagname"
# Evaluate the configuration. Be careful to quote the path
# and the sed script, to avoid splitting on whitespace, but
# also don't use non-portable quotes within backquotes within
# quotes we have to do it in 2 steps:
extractedcf=`$SED -n -e "$sed_extractcf" < "$progpath"`
eval "$extractedcf"
else
func_error "ignoring unknown tag $tagname"
fi
;;
esac
}
# func_check_version_match
# Ensure that we are using m4 macros, and libtool script from the same
# release of libtool.
func_check_version_match ()
{
if test "$package_revision" != "$macro_revision"; then
if test "$VERSION" != "$macro_version"; then
if test -z "$macro_version"; then
cat >&2 <<_LT_EOF
$progname: Version mismatch error. This is $PACKAGE $VERSION, but the
$progname: definition of this LT_INIT comes from an older release.
$progname: You should recreate aclocal.m4 with macros from $PACKAGE $VERSION
$progname: and run autoconf again.
_LT_EOF
else
cat >&2 <<_LT_EOF
$progname: Version mismatch error. This is $PACKAGE $VERSION, but the
$progname: definition of this LT_INIT comes from $PACKAGE $macro_version.
$progname: You should recreate aclocal.m4 with macros from $PACKAGE $VERSION
$progname: and run autoconf again.
_LT_EOF
fi
else
cat >&2 <<_LT_EOF
$progname: Version mismatch error. This is $PACKAGE $VERSION, revision $package_revision,
$progname: but the definition of this LT_INIT comes from revision $macro_revision.
$progname: You should recreate aclocal.m4 with macros from revision $package_revision
$progname: of $PACKAGE $VERSION and run autoconf again.
_LT_EOF
fi
exit $EXIT_MISMATCH
fi
}
# Shorthand for --mode=foo, only valid as the first argument
case $1 in
clean|clea|cle|cl)
shift; set dummy --mode clean ${1+"$@"}; shift
;;
compile|compil|compi|comp|com|co|c)
shift; set dummy --mode compile ${1+"$@"}; shift
;;
execute|execut|execu|exec|exe|ex|e)
shift; set dummy --mode execute ${1+"$@"}; shift
;;
finish|finis|fini|fin|fi|f)
shift; set dummy --mode finish ${1+"$@"}; shift
;;
install|instal|insta|inst|ins|in|i)
shift; set dummy --mode install ${1+"$@"}; shift
;;
link|lin|li|l)
shift; set dummy --mode link ${1+"$@"}; shift
;;
uninstall|uninstal|uninsta|uninst|unins|unin|uni|un|u)
shift; set dummy --mode uninstall ${1+"$@"}; shift
;;
esac
# Option defaults:
opt_debug=:
opt_dry_run=false
opt_config=false
opt_preserve_dup_deps=false
opt_features=false
opt_finish=false
opt_help=false
opt_help_all=false
opt_silent=:
opt_warning=:
opt_verbose=:
opt_silent=false
opt_verbose=false
# Parse options once, thoroughly. This comes as soon as possible in the
# script to make things like `--version' happen as quickly as we can.
{
# this just eases exit handling
while test $# -gt 0; do
opt="$1"
shift
case $opt in
--debug|-x) opt_debug='set -x'
func_echo "enabling shell trace mode"
$opt_debug
;;
--dry-run|--dryrun|-n)
opt_dry_run=:
;;
--config)
opt_config=:
func_config
;;
--dlopen|-dlopen)
optarg="$1"
opt_dlopen="${opt_dlopen+$opt_dlopen
}$optarg"
shift
;;
--preserve-dup-deps)
opt_preserve_dup_deps=:
;;
--features)
opt_features=:
func_features
;;
--finish)
opt_finish=:
set dummy --mode finish ${1+"$@"}; shift
;;
--help)
opt_help=:
;;
--help-all)
opt_help_all=:
opt_help=': help-all'
;;
--mode)
test $# = 0 && func_missing_arg $opt && break
optarg="$1"
opt_mode="$optarg"
case $optarg in
# Valid mode arguments:
clean|compile|execute|finish|install|link|relink|uninstall) ;;
# Catch anything else as an error
*) func_error "invalid argument for $opt"
exit_cmd=exit
break
;;
esac
shift
;;
--no-silent|--no-quiet)
opt_silent=false
func_append preserve_args " $opt"
;;
--no-warning|--no-warn)
opt_warning=false
func_append preserve_args " $opt"
;;
--no-verbose)
opt_verbose=false
func_append preserve_args " $opt"
;;
--silent|--quiet)
opt_silent=:
func_append preserve_args " $opt"
opt_verbose=false
;;
--verbose|-v)
opt_verbose=:
func_append preserve_args " $opt"
opt_silent=false
;;
--tag)
test $# = 0 && func_missing_arg $opt && break
optarg="$1"
opt_tag="$optarg"
func_append preserve_args " $opt $optarg"
func_enable_tag "$optarg"
shift
;;
-\?|-h) func_usage ;;
--help) func_help ;;
--version) func_version ;;
# Separate optargs to long options:
--*=*)
func_split_long_opt "$opt"
set dummy "$func_split_long_opt_name" "$func_split_long_opt_arg" ${1+"$@"}
shift
;;
# Separate non-argument short options:
-\?*|-h*|-n*|-v*)
func_split_short_opt "$opt"
set dummy "$func_split_short_opt_name" "-$func_split_short_opt_arg" ${1+"$@"}
shift
;;
--) break ;;
-*) func_fatal_help "unrecognized option \`$opt'" ;;
*) set dummy "$opt" ${1+"$@"}; shift; break ;;
esac
done
# Validate options:
# save first non-option argument
if test "$#" -gt 0; then
nonopt="$opt"
shift
fi
# preserve --debug
test "$opt_debug" = : || func_append preserve_args " --debug"
case $host in
*cygwin* | *mingw* | *pw32* | *cegcc*)
# don't eliminate duplications in $postdeps and $predeps
opt_duplicate_compiler_generated_deps=:
;;
*)
opt_duplicate_compiler_generated_deps=$opt_preserve_dup_deps
;;
esac
$opt_help || {
# Sanity checks first:
func_check_version_match
if test "$build_libtool_libs" != yes && test "$build_old_libs" != yes; then
func_fatal_configuration "not configured to build any kind of library"
fi
# Darwin sucks
eval std_shrext=\"$shrext_cmds\"
# Only execute mode is allowed to have -dlopen flags.
if test -n "$opt_dlopen" && test "$opt_mode" != execute; then
func_error "unrecognized option \`-dlopen'"
$ECHO "$help" 1>&2
exit $EXIT_FAILURE
fi
# Change the help message to a mode-specific one.
generic_help="$help"
help="Try \`$progname --help --mode=$opt_mode' for more information."
}
# Bail if the options were screwed
$exit_cmd $EXIT_FAILURE
}
## ----------- ##
## Main. ##
## ----------- ##
# func_lalib_p file
# True iff FILE is a libtool `.la' library or `.lo' object file.
# This function is only a basic sanity check; it will hardly flush out
# determined imposters.
func_lalib_p ()
{
test -f "$1" &&
$SED -e 4q "$1" 2>/dev/null \
| $GREP "^# Generated by .*$PACKAGE" > /dev/null 2>&1
}
# func_lalib_unsafe_p file
# True iff FILE is a libtool `.la' library or `.lo' object file.
# This function implements the same check as func_lalib_p without
# resorting to external programs. To this end, it redirects stdin and
# closes it afterwards, without saving the original file descriptor.
# As a safety measure, use it only where a negative result would be
# fatal anyway. Works if `file' does not exist.
func_lalib_unsafe_p ()
{
lalib_p=no
if test -f "$1" && test -r "$1" && exec 5<&0 <"$1"; then
for lalib_p_l in 1 2 3 4
do
read lalib_p_line
case "$lalib_p_line" in
\#\ Generated\ by\ *$PACKAGE* ) lalib_p=yes; break;;
esac
done
exec 0<&5 5<&-
fi
test "$lalib_p" = yes
}
# func_ltwrapper_script_p file
# True iff FILE is a libtool wrapper script
# This function is only a basic sanity check; it will hardly flush out
# determined imposters.
func_ltwrapper_script_p ()
{
func_lalib_p "$1"
}
# func_ltwrapper_executable_p file
# True iff FILE is a libtool wrapper executable
# This function is only a basic sanity check; it will hardly flush out
# determined imposters.
func_ltwrapper_executable_p ()
{
func_ltwrapper_exec_suffix=
case $1 in
*.exe) ;;
*) func_ltwrapper_exec_suffix=.exe ;;
esac
$GREP "$magic_exe" "$1$func_ltwrapper_exec_suffix" >/dev/null 2>&1
}
# func_ltwrapper_scriptname file
# Assumes file is an ltwrapper_executable
# uses $file to determine the appropriate filename for a
# temporary ltwrapper_script.
func_ltwrapper_scriptname ()
{
func_dirname_and_basename "$1" "" "."
func_stripname '' '.exe' "$func_basename_result"
func_ltwrapper_scriptname_result="$func_dirname_result/$objdir/${func_stripname_result}_ltshwrapper"
}
# func_ltwrapper_p file
# True iff FILE is a libtool wrapper script or wrapper executable
# This function is only a basic sanity check; it will hardly flush out
# determined imposters.
func_ltwrapper_p ()
{
func_ltwrapper_script_p "$1" || func_ltwrapper_executable_p "$1"
}
# func_execute_cmds commands fail_cmd
# Execute tilde-delimited COMMANDS.
# If FAIL_CMD is given, eval that upon failure.
# FAIL_CMD may read-access the current command in variable CMD!
func_execute_cmds ()
{
$opt_debug
save_ifs=$IFS; IFS='~'
for cmd in $1; do
IFS=$save_ifs
eval cmd=\"$cmd\"
func_show_eval "$cmd" "${2-:}"
done
IFS=$save_ifs
}
# func_source file
# Source FILE, adding directory component if necessary.
# Note that it is not necessary on cygwin/mingw to append a dot to
# FILE even if both FILE and FILE.exe exist: automatic-append-.exe
# behavior happens only for exec(3), not for open(2)! Also, sourcing
# `FILE.' does not work on cygwin managed mounts.
func_source ()
{
$opt_debug
case $1 in
*/* | *\\*) . "$1" ;;
*) . "./$1" ;;
esac
}
# func_resolve_sysroot PATH
# Replace a leading = in PATH with a sysroot. Store the result into
# func_resolve_sysroot_result
func_resolve_sysroot ()
{
func_resolve_sysroot_result=$1
case $func_resolve_sysroot_result in
=*)
func_stripname '=' '' "$func_resolve_sysroot_result"
func_resolve_sysroot_result=$lt_sysroot$func_stripname_result
;;
esac
}
# func_replace_sysroot PATH
# If PATH begins with the sysroot, replace it with = and
# store the result into func_replace_sysroot_result.
func_replace_sysroot ()
{
case "$lt_sysroot:$1" in
?*:"$lt_sysroot"*)
func_stripname "$lt_sysroot" '' "$1"
func_replace_sysroot_result="=$func_stripname_result"
;;
*)
# Including no sysroot.
func_replace_sysroot_result=$1
;;
esac
}
# func_infer_tag arg
# Infer tagged configuration to use if any are available and
# if one wasn't chosen via the "--tag" command line option.
# Only attempt this if the compiler in the base compile
# command doesn't match the default compiler.
# arg is usually of the form 'gcc ...'
func_infer_tag ()
{
$opt_debug
if test -n "$available_tags" && test -z "$tagname"; then
CC_quoted=
for arg in $CC; do
func_append_quoted CC_quoted "$arg"
done
CC_expanded=`func_echo_all $CC`
CC_quoted_expanded=`func_echo_all $CC_quoted`
case $@ in
# Blanks in the command may have been stripped by the calling shell,
# but not from the CC environment variable when configure was run.
" $CC "* | "$CC "* | " $CC_expanded "* | "$CC_expanded "* | \
" $CC_quoted"* | "$CC_quoted "* | " $CC_quoted_expanded "* | "$CC_quoted_expanded "*) ;;
# Blanks at the start of $base_compile will cause this to fail
# if we don't check for them as well.
*)
for z in $available_tags; do
if $GREP "^# ### BEGIN LIBTOOL TAG CONFIG: $z$" < "$progpath" > /dev/null; then
# Evaluate the configuration.
eval "`${SED} -n -e '/^# ### BEGIN LIBTOOL TAG CONFIG: '$z'$/,/^# ### END LIBTOOL TAG CONFIG: '$z'$/p' < $progpath`"
CC_quoted=
for arg in $CC; do
# Double-quote args containing other shell metacharacters.
func_append_quoted CC_quoted "$arg"
done
CC_expanded=`func_echo_all $CC`
CC_quoted_expanded=`func_echo_all $CC_quoted`
case "$@ " in
" $CC "* | "$CC "* | " $CC_expanded "* | "$CC_expanded "* | \
" $CC_quoted"* | "$CC_quoted "* | " $CC_quoted_expanded "* | "$CC_quoted_expanded "*)
# The compiler in the base compile command matches
# the one in the tagged configuration.
# Assume this is the tagged configuration we want.
tagname=$z
break
;;
esac
fi
done
# If $tagname still isn't set, then no tagged configuration
# was found and let the user know that the "--tag" command
# line option must be used.
if test -z "$tagname"; then
func_echo "unable to infer tagged configuration"
func_fatal_error "specify a tag with \`--tag'"
# else
# func_verbose "using $tagname tagged configuration"
fi
;;
esac
fi
}
# func_write_libtool_object output_name pic_name nonpic_name
# Create a libtool object file (analogous to a ".la" file),
# but don't create it if we're doing a dry run.
func_write_libtool_object ()
{
write_libobj=${1}
if test "$build_libtool_libs" = yes; then
write_lobj=\'${2}\'
else
write_lobj=none
fi
if test "$build_old_libs" = yes; then
write_oldobj=\'${3}\'
else
write_oldobj=none
fi
$opt_dry_run || {
cat >${write_libobj}T </dev/null`
if test "$?" -eq 0 && test -n "${func_convert_core_file_wine_to_w32_tmp}"; then
func_convert_core_file_wine_to_w32_result=`$ECHO "$func_convert_core_file_wine_to_w32_tmp" |
$SED -e "$lt_sed_naive_backslashify"`
else
func_convert_core_file_wine_to_w32_result=
fi
fi
}
# end: func_convert_core_file_wine_to_w32
# func_convert_core_path_wine_to_w32 ARG
# Helper function used by path conversion functions when $build is *nix, and
# $host is mingw, cygwin, or some other w32 environment. Relies on a correctly
# configured wine environment available, with the winepath program in $build's
# $PATH. Assumes ARG has no leading or trailing path separator characters.
#
# ARG is path to be converted from $build format to win32.
# Result is available in $func_convert_core_path_wine_to_w32_result.
# Unconvertible file (directory) names in ARG are skipped; if no directory names
# are convertible, then the result may be empty.
func_convert_core_path_wine_to_w32 ()
{
$opt_debug
# unfortunately, winepath doesn't convert paths, only file names
func_convert_core_path_wine_to_w32_result=""
if test -n "$1"; then
oldIFS=$IFS
IFS=:
for func_convert_core_path_wine_to_w32_f in $1; do
IFS=$oldIFS
func_convert_core_file_wine_to_w32 "$func_convert_core_path_wine_to_w32_f"
if test -n "$func_convert_core_file_wine_to_w32_result" ; then
if test -z "$func_convert_core_path_wine_to_w32_result"; then
func_convert_core_path_wine_to_w32_result="$func_convert_core_file_wine_to_w32_result"
else
func_append func_convert_core_path_wine_to_w32_result ";$func_convert_core_file_wine_to_w32_result"
fi
fi
done
IFS=$oldIFS
fi
}
# end: func_convert_core_path_wine_to_w32
# func_cygpath ARGS...
# Wrapper around calling the cygpath program via LT_CYGPATH. This is used when
# when (1) $build is *nix and Cygwin is hosted via a wine environment; or (2)
# $build is MSYS and $host is Cygwin, or (3) $build is Cygwin. In case (1) or
# (2), returns the Cygwin file name or path in func_cygpath_result (input
# file name or path is assumed to be in w32 format, as previously converted
# from $build's *nix or MSYS format). In case (3), returns the w32 file name
# or path in func_cygpath_result (input file name or path is assumed to be in
# Cygwin format). Returns an empty string on error.
#
# ARGS are passed to cygpath, with the last one being the file name or path to
# be converted.
#
# Specify the absolute *nix (or w32) name to cygpath in the LT_CYGPATH
# environment variable; do not put it in $PATH.
func_cygpath ()
{
$opt_debug
if test -n "$LT_CYGPATH" && test -f "$LT_CYGPATH"; then
func_cygpath_result=`$LT_CYGPATH "$@" 2>/dev/null`
if test "$?" -ne 0; then
# on failure, ensure result is empty
func_cygpath_result=
fi
else
func_cygpath_result=
func_error "LT_CYGPATH is empty or specifies non-existent file: \`$LT_CYGPATH'"
fi
}
#end: func_cygpath
# func_convert_core_msys_to_w32 ARG
# Convert file name or path ARG from MSYS format to w32 format. Return
# result in func_convert_core_msys_to_w32_result.
func_convert_core_msys_to_w32 ()
{
$opt_debug
# awkward: cmd appends spaces to result
func_convert_core_msys_to_w32_result=`( cmd //c echo "$1" ) 2>/dev/null |
$SED -e 's/[ ]*$//' -e "$lt_sed_naive_backslashify"`
}
#end: func_convert_core_msys_to_w32
# func_convert_file_check ARG1 ARG2
# Verify that ARG1 (a file name in $build format) was converted to $host
# format in ARG2. Otherwise, emit an error message, but continue (resetting
# func_to_host_file_result to ARG1).
func_convert_file_check ()
{
$opt_debug
if test -z "$2" && test -n "$1" ; then
func_error "Could not determine host file name corresponding to"
func_error " \`$1'"
func_error "Continuing, but uninstalled executables may not work."
# Fallback:
func_to_host_file_result="$1"
fi
}
# end func_convert_file_check
# func_convert_path_check FROM_PATHSEP TO_PATHSEP FROM_PATH TO_PATH
# Verify that FROM_PATH (a path in $build format) was converted to $host
# format in TO_PATH. Otherwise, emit an error message, but continue, resetting
# func_to_host_file_result to a simplistic fallback value (see below).
func_convert_path_check ()
{
$opt_debug
if test -z "$4" && test -n "$3"; then
func_error "Could not determine the host path corresponding to"
func_error " \`$3'"
func_error "Continuing, but uninstalled executables may not work."
# Fallback. This is a deliberately simplistic "conversion" and
# should not be "improved". See libtool.info.
if test "x$1" != "x$2"; then
lt_replace_pathsep_chars="s|$1|$2|g"
func_to_host_path_result=`echo "$3" |
$SED -e "$lt_replace_pathsep_chars"`
else
func_to_host_path_result="$3"
fi
fi
}
# end func_convert_path_check
# func_convert_path_front_back_pathsep FRONTPAT BACKPAT REPL ORIG
# Modifies func_to_host_path_result by prepending REPL if ORIG matches FRONTPAT
# and appending REPL if ORIG matches BACKPAT.
func_convert_path_front_back_pathsep ()
{
$opt_debug
case $4 in
$1 ) func_to_host_path_result="$3$func_to_host_path_result"
;;
esac
case $4 in
$2 ) func_append func_to_host_path_result "$3"
;;
esac
}
# end func_convert_path_front_back_pathsep
##################################################
# $build to $host FILE NAME CONVERSION FUNCTIONS #
##################################################
# invoked via `$to_host_file_cmd ARG'
#
# In each case, ARG is the path to be converted from $build to $host format.
# Result will be available in $func_to_host_file_result.
# func_to_host_file ARG
# Converts the file name ARG from $build format to $host format. Return result
# in func_to_host_file_result.
func_to_host_file ()
{
$opt_debug
$to_host_file_cmd "$1"
}
# end func_to_host_file
# func_to_tool_file ARG LAZY
# converts the file name ARG from $build format to toolchain format. Return
# result in func_to_tool_file_result. If the conversion in use is listed
# in (the comma separated) LAZY, no conversion takes place.
func_to_tool_file ()
{
$opt_debug
case ,$2, in
*,"$to_tool_file_cmd",*)
func_to_tool_file_result=$1
;;
*)
$to_tool_file_cmd "$1"
func_to_tool_file_result=$func_to_host_file_result
;;
esac
}
# end func_to_tool_file
# func_convert_file_noop ARG
# Copy ARG to func_to_host_file_result.
func_convert_file_noop ()
{
func_to_host_file_result="$1"
}
# end func_convert_file_noop
# func_convert_file_msys_to_w32 ARG
# Convert file name ARG from (mingw) MSYS to (mingw) w32 format; automatic
# conversion to w32 is not available inside the cwrapper. Returns result in
# func_to_host_file_result.
func_convert_file_msys_to_w32 ()
{
$opt_debug
func_to_host_file_result="$1"
if test -n "$1"; then
func_convert_core_msys_to_w32 "$1"
func_to_host_file_result="$func_convert_core_msys_to_w32_result"
fi
func_convert_file_check "$1" "$func_to_host_file_result"
}
# end func_convert_file_msys_to_w32
# func_convert_file_cygwin_to_w32 ARG
# Convert file name ARG from Cygwin to w32 format. Returns result in
# func_to_host_file_result.
func_convert_file_cygwin_to_w32 ()
{
$opt_debug
func_to_host_file_result="$1"
if test -n "$1"; then
# because $build is cygwin, we call "the" cygpath in $PATH; no need to use
# LT_CYGPATH in this case.
func_to_host_file_result=`cygpath -m "$1"`
fi
func_convert_file_check "$1" "$func_to_host_file_result"
}
# end func_convert_file_cygwin_to_w32
# func_convert_file_nix_to_w32 ARG
# Convert file name ARG from *nix to w32 format. Requires a wine environment
# and a working winepath. Returns result in func_to_host_file_result.
func_convert_file_nix_to_w32 ()
{
$opt_debug
func_to_host_file_result="$1"
if test -n "$1"; then
func_convert_core_file_wine_to_w32 "$1"
func_to_host_file_result="$func_convert_core_file_wine_to_w32_result"
fi
func_convert_file_check "$1" "$func_to_host_file_result"
}
# end func_convert_file_nix_to_w32
# func_convert_file_msys_to_cygwin ARG
# Convert file name ARG from MSYS to Cygwin format. Requires LT_CYGPATH set.
# Returns result in func_to_host_file_result.
func_convert_file_msys_to_cygwin ()
{
$opt_debug
func_to_host_file_result="$1"
if test -n "$1"; then
func_convert_core_msys_to_w32 "$1"
func_cygpath -u "$func_convert_core_msys_to_w32_result"
func_to_host_file_result="$func_cygpath_result"
fi
func_convert_file_check "$1" "$func_to_host_file_result"
}
# end func_convert_file_msys_to_cygwin
# func_convert_file_nix_to_cygwin ARG
# Convert file name ARG from *nix to Cygwin format. Requires Cygwin installed
# in a wine environment, working winepath, and LT_CYGPATH set. Returns result
# in func_to_host_file_result.
func_convert_file_nix_to_cygwin ()
{
$opt_debug
func_to_host_file_result="$1"
if test -n "$1"; then
# convert from *nix to w32, then use cygpath to convert from w32 to cygwin.
func_convert_core_file_wine_to_w32 "$1"
func_cygpath -u "$func_convert_core_file_wine_to_w32_result"
func_to_host_file_result="$func_cygpath_result"
fi
func_convert_file_check "$1" "$func_to_host_file_result"
}
# end func_convert_file_nix_to_cygwin
#############################################
# $build to $host PATH CONVERSION FUNCTIONS #
#############################################
# invoked via `$to_host_path_cmd ARG'
#
# In each case, ARG is the path to be converted from $build to $host format.
# The result will be available in $func_to_host_path_result.
#
# Path separators are also converted from $build format to $host format. If
# ARG begins or ends with a path separator character, it is preserved (but
# converted to $host format) on output.
#
# All path conversion functions are named using the following convention:
# file name conversion function : func_convert_file_X_to_Y ()
# path conversion function : func_convert_path_X_to_Y ()
# where, for any given $build/$host combination the 'X_to_Y' value is the
# same. If conversion functions are added for new $build/$host combinations,
# the two new functions must follow this pattern, or func_init_to_host_path_cmd
# will break.
# func_init_to_host_path_cmd
# Ensures that function "pointer" variable $to_host_path_cmd is set to the
# appropriate value, based on the value of $to_host_file_cmd.
to_host_path_cmd=
func_init_to_host_path_cmd ()
{
$opt_debug
if test -z "$to_host_path_cmd"; then
func_stripname 'func_convert_file_' '' "$to_host_file_cmd"
to_host_path_cmd="func_convert_path_${func_stripname_result}"
fi
}
# func_to_host_path ARG
# Converts the path ARG from $build format to $host format. Return result
# in func_to_host_path_result.
func_to_host_path ()
{
$opt_debug
func_init_to_host_path_cmd
$to_host_path_cmd "$1"
}
# end func_to_host_path
# func_convert_path_noop ARG
# Copy ARG to func_to_host_path_result.
func_convert_path_noop ()
{
func_to_host_path_result="$1"
}
# end func_convert_path_noop
# func_convert_path_msys_to_w32 ARG
# Convert path ARG from (mingw) MSYS to (mingw) w32 format; automatic
# conversion to w32 is not available inside the cwrapper. Returns result in
# func_to_host_path_result.
func_convert_path_msys_to_w32 ()
{
$opt_debug
func_to_host_path_result="$1"
if test -n "$1"; then
# Remove leading and trailing path separator characters from ARG. MSYS
# behavior is inconsistent here; cygpath turns them into '.;' and ';.';
# and winepath ignores them completely.
func_stripname : : "$1"
func_to_host_path_tmp1=$func_stripname_result
func_convert_core_msys_to_w32 "$func_to_host_path_tmp1"
func_to_host_path_result="$func_convert_core_msys_to_w32_result"
func_convert_path_check : ";" \
"$func_to_host_path_tmp1" "$func_to_host_path_result"
func_convert_path_front_back_pathsep ":*" "*:" ";" "$1"
fi
}
# end func_convert_path_msys_to_w32
# func_convert_path_cygwin_to_w32 ARG
# Convert path ARG from Cygwin to w32 format. Returns result in
# func_to_host_file_result.
func_convert_path_cygwin_to_w32 ()
{
$opt_debug
func_to_host_path_result="$1"
if test -n "$1"; then
# See func_convert_path_msys_to_w32:
func_stripname : : "$1"
func_to_host_path_tmp1=$func_stripname_result
func_to_host_path_result=`cygpath -m -p "$func_to_host_path_tmp1"`
func_convert_path_check : ";" \
"$func_to_host_path_tmp1" "$func_to_host_path_result"
func_convert_path_front_back_pathsep ":*" "*:" ";" "$1"
fi
}
# end func_convert_path_cygwin_to_w32
# func_convert_path_nix_to_w32 ARG
# Convert path ARG from *nix to w32 format. Requires a wine environment and
# a working winepath. Returns result in func_to_host_file_result.
func_convert_path_nix_to_w32 ()
{
$opt_debug
func_to_host_path_result="$1"
if test -n "$1"; then
# See func_convert_path_msys_to_w32:
func_stripname : : "$1"
func_to_host_path_tmp1=$func_stripname_result
func_convert_core_path_wine_to_w32 "$func_to_host_path_tmp1"
func_to_host_path_result="$func_convert_core_path_wine_to_w32_result"
func_convert_path_check : ";" \
"$func_to_host_path_tmp1" "$func_to_host_path_result"
func_convert_path_front_back_pathsep ":*" "*:" ";" "$1"
fi
}
# end func_convert_path_nix_to_w32
# func_convert_path_msys_to_cygwin ARG
# Convert path ARG from MSYS to Cygwin format. Requires LT_CYGPATH set.
# Returns result in func_to_host_file_result.
func_convert_path_msys_to_cygwin ()
{
$opt_debug
func_to_host_path_result="$1"
if test -n "$1"; then
# See func_convert_path_msys_to_w32:
func_stripname : : "$1"
func_to_host_path_tmp1=$func_stripname_result
func_convert_core_msys_to_w32 "$func_to_host_path_tmp1"
func_cygpath -u -p "$func_convert_core_msys_to_w32_result"
func_to_host_path_result="$func_cygpath_result"
func_convert_path_check : : \
"$func_to_host_path_tmp1" "$func_to_host_path_result"
func_convert_path_front_back_pathsep ":*" "*:" : "$1"
fi
}
# end func_convert_path_msys_to_cygwin
# func_convert_path_nix_to_cygwin ARG
# Convert path ARG from *nix to Cygwin format. Requires Cygwin installed in a
# a wine environment, working winepath, and LT_CYGPATH set. Returns result in
# func_to_host_file_result.
func_convert_path_nix_to_cygwin ()
{
$opt_debug
func_to_host_path_result="$1"
if test -n "$1"; then
# Remove leading and trailing path separator characters from
# ARG. msys behavior is inconsistent here, cygpath turns them
# into '.;' and ';.', and winepath ignores them completely.
func_stripname : : "$1"
func_to_host_path_tmp1=$func_stripname_result
func_convert_core_path_wine_to_w32 "$func_to_host_path_tmp1"
func_cygpath -u -p "$func_convert_core_path_wine_to_w32_result"
func_to_host_path_result="$func_cygpath_result"
func_convert_path_check : : \
"$func_to_host_path_tmp1" "$func_to_host_path_result"
func_convert_path_front_back_pathsep ":*" "*:" : "$1"
fi
}
# end func_convert_path_nix_to_cygwin
# func_mode_compile arg...
func_mode_compile ()
{
$opt_debug
# Get the compilation command and the source file.
base_compile=
srcfile="$nonopt" # always keep a non-empty value in "srcfile"
suppress_opt=yes
suppress_output=
arg_mode=normal
libobj=
later=
pie_flag=
for arg
do
case $arg_mode in
arg )
# do not "continue". Instead, add this to base_compile
lastarg="$arg"
arg_mode=normal
;;
target )
libobj="$arg"
arg_mode=normal
continue
;;
normal )
# Accept any command-line options.
case $arg in
-o)
test -n "$libobj" && \
func_fatal_error "you cannot specify \`-o' more than once"
arg_mode=target
continue
;;
-pie | -fpie | -fPIE)
func_append pie_flag " $arg"
continue
;;
-shared | -static | -prefer-pic | -prefer-non-pic)
func_append later " $arg"
continue
;;
-no-suppress)
suppress_opt=no
continue
;;
-Xcompiler)
arg_mode=arg # the next one goes into the "base_compile" arg list
continue # The current "srcfile" will either be retained or
;; # replaced later. I would guess that would be a bug.
-Wc,*)
func_stripname '-Wc,' '' "$arg"
args=$func_stripname_result
lastarg=
save_ifs="$IFS"; IFS=','
for arg in $args; do
IFS="$save_ifs"
func_append_quoted lastarg "$arg"
done
IFS="$save_ifs"
func_stripname ' ' '' "$lastarg"
lastarg=$func_stripname_result
# Add the arguments to base_compile.
func_append base_compile " $lastarg"
continue
;;
*)
# Accept the current argument as the source file.
# The previous "srcfile" becomes the current argument.
#
lastarg="$srcfile"
srcfile="$arg"
;;
esac # case $arg
;;
esac # case $arg_mode
# Aesthetically quote the previous argument.
func_append_quoted base_compile "$lastarg"
done # for arg
case $arg_mode in
arg)
func_fatal_error "you must specify an argument for -Xcompile"
;;
target)
func_fatal_error "you must specify a target with \`-o'"
;;
*)
# Get the name of the library object.
test -z "$libobj" && {
func_basename "$srcfile"
libobj="$func_basename_result"
}
;;
esac
# Recognize several different file suffixes.
# If the user specifies -o file.o, it is replaced with file.lo
case $libobj in
*.[cCFSifmso] | \
*.ada | *.adb | *.ads | *.asm | \
*.c++ | *.cc | *.ii | *.class | *.cpp | *.cxx | \
*.[fF][09]? | *.for | *.java | *.go | *.obj | *.sx | *.cu | *.cup)
func_xform "$libobj"
libobj=$func_xform_result
;;
esac
case $libobj in
*.lo) func_lo2o "$libobj"; obj=$func_lo2o_result ;;
*)
func_fatal_error "cannot determine name of library object from \`$libobj'"
;;
esac
func_infer_tag $base_compile
for arg in $later; do
case $arg in
-shared)
test "$build_libtool_libs" != yes && \
func_fatal_configuration "can not build a shared library"
build_old_libs=no
continue
;;
-static)
build_libtool_libs=no
build_old_libs=yes
continue
;;
-prefer-pic)
pic_mode=yes
continue
;;
-prefer-non-pic)
pic_mode=no
continue
;;
esac
done
func_quote_for_eval "$libobj"
test "X$libobj" != "X$func_quote_for_eval_result" \
&& $ECHO "X$libobj" | $GREP '[]~#^*{};<>?"'"'"' &()|`$[]' \
&& func_warning "libobj name \`$libobj' may not contain shell special characters."
func_dirname_and_basename "$obj" "/" ""
objname="$func_basename_result"
xdir="$func_dirname_result"
lobj=${xdir}$objdir/$objname
test -z "$base_compile" && \
func_fatal_help "you must specify a compilation command"
# Delete any leftover library objects.
if test "$build_old_libs" = yes; then
removelist="$obj $lobj $libobj ${libobj}T"
else
removelist="$lobj $libobj ${libobj}T"
fi
# On Cygwin there's no "real" PIC flag so we must build both object types
case $host_os in
cygwin* | mingw* | pw32* | os2* | cegcc*)
pic_mode=default
;;
esac
if test "$pic_mode" = no && test "$deplibs_check_method" != pass_all; then
# non-PIC code in shared libraries is not supported
pic_mode=default
fi
# Calculate the filename of the output object if compiler does
# not support -o with -c
if test "$compiler_c_o" = no; then
output_obj=`$ECHO "$srcfile" | $SED 's%^.*/%%; s%\.[^.]*$%%'`.${objext}
lockfile="$output_obj.lock"
else
output_obj=
need_locks=no
lockfile=
fi
# Lock this critical section if it is needed
# We use this script file to make the link, it avoids creating a new file
if test "$need_locks" = yes; then
until $opt_dry_run || ln "$progpath" "$lockfile" 2>/dev/null; do
func_echo "Waiting for $lockfile to be removed"
sleep 2
done
elif test "$need_locks" = warn; then
if test -f "$lockfile"; then
$ECHO "\
*** ERROR, $lockfile exists and contains:
`cat $lockfile 2>/dev/null`
This indicates that another process is trying to use the same
temporary object file, and libtool could not work around it because
your compiler does not support \`-c' and \`-o' together. If you
repeat this compilation, it may succeed, by chance, but you had better
avoid parallel builds (make -j) in this platform, or get a better
compiler."
$opt_dry_run || $RM $removelist
exit $EXIT_FAILURE
fi
func_append removelist " $output_obj"
$ECHO "$srcfile" > "$lockfile"
fi
$opt_dry_run || $RM $removelist
func_append removelist " $lockfile"
trap '$opt_dry_run || $RM $removelist; exit $EXIT_FAILURE' 1 2 15
func_to_tool_file "$srcfile" func_convert_file_msys_to_w32
srcfile=$func_to_tool_file_result
func_quote_for_eval "$srcfile"
qsrcfile=$func_quote_for_eval_result
# Only build a PIC object if we are building libtool libraries.
if test "$build_libtool_libs" = yes; then
# Without this assignment, base_compile gets emptied.
fbsd_hideous_sh_bug=$base_compile
if test "$pic_mode" != no; then
command="$base_compile $qsrcfile $pic_flag"
else
# Don't build PIC code
command="$base_compile $qsrcfile"
fi
func_mkdir_p "$xdir$objdir"
if test -z "$output_obj"; then
# Place PIC objects in $objdir
func_append command " -o $lobj"
fi
func_show_eval_locale "$command" \
'test -n "$output_obj" && $RM $removelist; exit $EXIT_FAILURE'
if test "$need_locks" = warn &&
test "X`cat $lockfile 2>/dev/null`" != "X$srcfile"; then
$ECHO "\
*** ERROR, $lockfile contains:
`cat $lockfile 2>/dev/null`
but it should contain:
$srcfile
This indicates that another process is trying to use the same
temporary object file, and libtool could not work around it because
your compiler does not support \`-c' and \`-o' together. If you
repeat this compilation, it may succeed, by chance, but you had better
avoid parallel builds (make -j) in this platform, or get a better
compiler."
$opt_dry_run || $RM $removelist
exit $EXIT_FAILURE
fi
# Just move the object if needed, then go on to compile the next one
if test -n "$output_obj" && test "X$output_obj" != "X$lobj"; then
func_show_eval '$MV "$output_obj" "$lobj"' \
'error=$?; $opt_dry_run || $RM $removelist; exit $error'
fi
# Allow error messages only from the first compilation.
if test "$suppress_opt" = yes; then
suppress_output=' >/dev/null 2>&1'
fi
fi
# Only build a position-dependent object if we build old libraries.
if test "$build_old_libs" = yes; then
if test "$pic_mode" != yes; then
# Don't build PIC code
command="$base_compile $qsrcfile$pie_flag"
else
command="$base_compile $qsrcfile $pic_flag"
fi
if test "$compiler_c_o" = yes; then
func_append command " -o $obj"
fi
# Suppress compiler output if we already did a PIC compilation.
func_append command "$suppress_output"
func_show_eval_locale "$command" \
'$opt_dry_run || $RM $removelist; exit $EXIT_FAILURE'
if test "$need_locks" = warn &&
test "X`cat $lockfile 2>/dev/null`" != "X$srcfile"; then
$ECHO "\
*** ERROR, $lockfile contains:
`cat $lockfile 2>/dev/null`
but it should contain:
$srcfile
This indicates that another process is trying to use the same
temporary object file, and libtool could not work around it because
your compiler does not support \`-c' and \`-o' together. If you
repeat this compilation, it may succeed, by chance, but you had better
avoid parallel builds (make -j) in this platform, or get a better
compiler."
$opt_dry_run || $RM $removelist
exit $EXIT_FAILURE
fi
# Just move the object if needed
if test -n "$output_obj" && test "X$output_obj" != "X$obj"; then
func_show_eval '$MV "$output_obj" "$obj"' \
'error=$?; $opt_dry_run || $RM $removelist; exit $error'
fi
fi
$opt_dry_run || {
func_write_libtool_object "$libobj" "$objdir/$objname" "$objname"
# Unlock the critical section if it was locked
if test "$need_locks" != no; then
removelist=$lockfile
$RM "$lockfile"
fi
}
exit $EXIT_SUCCESS
}
$opt_help || {
test "$opt_mode" = compile && func_mode_compile ${1+"$@"}
}
func_mode_help ()
{
# We need to display help for each of the modes.
case $opt_mode in
"")
# Generic help is extracted from the usage comments
# at the start of this file.
func_help
;;
clean)
$ECHO \
"Usage: $progname [OPTION]... --mode=clean RM [RM-OPTION]... FILE...
Remove files from the build directory.
RM is the name of the program to use to delete files associated with each FILE
(typically \`/bin/rm'). RM-OPTIONS are options (such as \`-f') to be passed
to RM.
If FILE is a libtool library, object or program, all the files associated
with it are deleted. Otherwise, only FILE itself is deleted using RM."
;;
compile)
$ECHO \
"Usage: $progname [OPTION]... --mode=compile COMPILE-COMMAND... SOURCEFILE
Compile a source file into a libtool library object.
This mode accepts the following additional options:
-o OUTPUT-FILE set the output file name to OUTPUT-FILE
-no-suppress do not suppress compiler output for multiple passes
-prefer-pic try to build PIC objects only
-prefer-non-pic try to build non-PIC objects only
-shared do not build a \`.o' file suitable for static linking
-static only build a \`.o' file suitable for static linking
-Wc,FLAG pass FLAG directly to the compiler
COMPILE-COMMAND is a command to be used in creating a \`standard' object file
from the given SOURCEFILE.
The output file name is determined by removing the directory component from
SOURCEFILE, then substituting the C source code suffix \`.c' with the
library object suffix, \`.lo'."
;;
execute)
$ECHO \
"Usage: $progname [OPTION]... --mode=execute COMMAND [ARGS]...
Automatically set library path, then run a program.
This mode accepts the following additional options:
-dlopen FILE add the directory containing FILE to the library path
This mode sets the library path environment variable according to \`-dlopen'
flags.
If any of the ARGS are libtool executable wrappers, then they are translated
into their corresponding uninstalled binary, and any of their required library
directories are added to the library path.
Then, COMMAND is executed, with ARGS as arguments."
;;
finish)
$ECHO \
"Usage: $progname [OPTION]... --mode=finish [LIBDIR]...
Complete the installation of libtool libraries.
Each LIBDIR is a directory that contains libtool libraries.
The commands that this mode executes may require superuser privileges. Use
the \`--dry-run' option if you just want to see what would be executed."
;;
install)
$ECHO \
"Usage: $progname [OPTION]... --mode=install INSTALL-COMMAND...
Install executables or libraries.
INSTALL-COMMAND is the installation command. The first component should be
either the \`install' or \`cp' program.
The following components of INSTALL-COMMAND are treated specially:
-inst-prefix-dir PREFIX-DIR Use PREFIX-DIR as a staging area for installation
The rest of the components are interpreted as arguments to that command (only
BSD-compatible install options are recognized)."
;;
link)
$ECHO \
"Usage: $progname [OPTION]... --mode=link LINK-COMMAND...
Link object files or libraries together to form another library, or to
create an executable program.
LINK-COMMAND is a command using the C compiler that you would use to create
a program from several object files.
The following components of LINK-COMMAND are treated specially:
-all-static do not do any dynamic linking at all
-avoid-version do not add a version suffix if possible
-bindir BINDIR specify path to binaries directory (for systems where
libraries must be found in the PATH setting at runtime)
-dlopen FILE \`-dlpreopen' FILE if it cannot be dlopened at runtime
-dlpreopen FILE link in FILE and add its symbols to lt_preloaded_symbols
-export-dynamic allow symbols from OUTPUT-FILE to be resolved with dlsym(3)
-export-symbols SYMFILE
try to export only the symbols listed in SYMFILE
-export-symbols-regex REGEX
try to export only the symbols matching REGEX
-LLIBDIR search LIBDIR for required installed libraries
-lNAME OUTPUT-FILE requires the installed library libNAME
-module build a library that can dlopened
-no-fast-install disable the fast-install mode
-no-install link a not-installable executable
-no-undefined declare that a library does not refer to external symbols
-o OUTPUT-FILE create OUTPUT-FILE from the specified objects
-objectlist FILE Use a list of object files found in FILE to specify objects
-precious-files-regex REGEX
don't remove output files matching REGEX
-release RELEASE specify package release information
-rpath LIBDIR the created library will eventually be installed in LIBDIR
-R[ ]LIBDIR add LIBDIR to the runtime path of programs and libraries
-shared only do dynamic linking of libtool libraries
-shrext SUFFIX override the standard shared library file extension
-static do not do any dynamic linking of uninstalled libtool libraries
-static-libtool-libs
do not do any dynamic linking of libtool libraries
-version-info CURRENT[:REVISION[:AGE]]
specify library version info [each variable defaults to 0]
-weak LIBNAME declare that the target provides the LIBNAME interface
-Wc,FLAG
-Xcompiler FLAG pass linker-specific FLAG directly to the compiler
-Wl,FLAG
-Xlinker FLAG pass linker-specific FLAG directly to the linker
-XCClinker FLAG pass link-specific FLAG to the compiler driver (CC)
All other options (arguments beginning with \`-') are ignored.
Every other argument is treated as a filename. Files ending in \`.la' are
treated as uninstalled libtool libraries, other files are standard or library
object files.
If the OUTPUT-FILE ends in \`.la', then a libtool library is created,
only library objects (\`.lo' files) may be specified, and \`-rpath' is
required, except when creating a convenience library.
If OUTPUT-FILE ends in \`.a' or \`.lib', then a standard library is created
using \`ar' and \`ranlib', or on Windows using \`lib'.
If OUTPUT-FILE ends in \`.lo' or \`.${objext}', then a reloadable object file
is created, otherwise an executable program is created."
;;
uninstall)
$ECHO \
"Usage: $progname [OPTION]... --mode=uninstall RM [RM-OPTION]... FILE...
Remove libraries from an installation directory.
RM is the name of the program to use to delete files associated with each FILE
(typically \`/bin/rm'). RM-OPTIONS are options (such as \`-f') to be passed
to RM.
If FILE is a libtool library, all the files associated with it are deleted.
Otherwise, only FILE itself is deleted using RM."
;;
*)
func_fatal_help "invalid operation mode \`$opt_mode'"
;;
esac
echo
$ECHO "Try \`$progname --help' for more information about other modes."
}
# Now that we've collected a possible --mode arg, show help if necessary
if $opt_help; then
if test "$opt_help" = :; then
func_mode_help
else
{
func_help noexit
for opt_mode in compile link execute install finish uninstall clean; do
func_mode_help
done
} | sed -n '1p; 2,$s/^Usage:/ or: /p'
{
func_help noexit
for opt_mode in compile link execute install finish uninstall clean; do
echo
func_mode_help
done
} |
sed '1d
/^When reporting/,/^Report/{
H
d
}
$x
/information about other modes/d
/more detailed .*MODE/d
s/^Usage:.*--mode=\([^ ]*\) .*/Description of \1 mode:/'
fi
exit $?
fi
# func_mode_execute arg...
func_mode_execute ()
{
$opt_debug
# The first argument is the command name.
cmd="$nonopt"
test -z "$cmd" && \
func_fatal_help "you must specify a COMMAND"
# Handle -dlopen flags immediately.
for file in $opt_dlopen; do
test -f "$file" \
|| func_fatal_help "\`$file' is not a file"
dir=
case $file in
*.la)
func_resolve_sysroot "$file"
file=$func_resolve_sysroot_result
# Check to see that this really is a libtool archive.
func_lalib_unsafe_p "$file" \
|| func_fatal_help "\`$lib' is not a valid libtool archive"
# Read the libtool library.
dlname=
library_names=
func_source "$file"
# Skip this library if it cannot be dlopened.
if test -z "$dlname"; then
# Warn if it was a shared library.
test -n "$library_names" && \
func_warning "\`$file' was not linked with \`-export-dynamic'"
continue
fi
func_dirname "$file" "" "."
dir="$func_dirname_result"
if test -f "$dir/$objdir/$dlname"; then
func_append dir "/$objdir"
else
if test ! -f "$dir/$dlname"; then
func_fatal_error "cannot find \`$dlname' in \`$dir' or \`$dir/$objdir'"
fi
fi
;;
*.lo)
# Just add the directory containing the .lo file.
func_dirname "$file" "" "."
dir="$func_dirname_result"
;;
*)
func_warning "\`-dlopen' is ignored for non-libtool libraries and objects"
continue
;;
esac
# Get the absolute pathname.
absdir=`cd "$dir" && pwd`
test -n "$absdir" && dir="$absdir"
# Now add the directory to shlibpath_var.
if eval "test -z \"\$$shlibpath_var\""; then
eval "$shlibpath_var=\"\$dir\""
else
eval "$shlibpath_var=\"\$dir:\$$shlibpath_var\""
fi
done
# This variable tells wrapper scripts just to set shlibpath_var
# rather than running their programs.
libtool_execute_magic="$magic"
# Check if any of the arguments is a wrapper script.
args=
for file
do
case $file in
-* | *.la | *.lo ) ;;
*)
# Do a test to see if this is really a libtool program.
if func_ltwrapper_script_p "$file"; then
func_source "$file"
# Transform arg to wrapped name.
file="$progdir/$program"
elif func_ltwrapper_executable_p "$file"; then
func_ltwrapper_scriptname "$file"
func_source "$func_ltwrapper_scriptname_result"
# Transform arg to wrapped name.
file="$progdir/$program"
fi
;;
esac
# Quote arguments (to preserve shell metacharacters).
func_append_quoted args "$file"
done
if test "X$opt_dry_run" = Xfalse; then
if test -n "$shlibpath_var"; then
# Export the shlibpath_var.
eval "export $shlibpath_var"
fi
# Restore saved environment variables
for lt_var in LANG LANGUAGE LC_ALL LC_CTYPE LC_COLLATE LC_MESSAGES
do
eval "if test \"\${save_$lt_var+set}\" = set; then
$lt_var=\$save_$lt_var; export $lt_var
else
$lt_unset $lt_var
fi"
done
# Now prepare to actually exec the command.
exec_cmd="\$cmd$args"
else
# Display what would be done.
if test -n "$shlibpath_var"; then
eval "\$ECHO \"\$shlibpath_var=\$$shlibpath_var\""
echo "export $shlibpath_var"
fi
$ECHO "$cmd$args"
exit $EXIT_SUCCESS
fi
}
test "$opt_mode" = execute && func_mode_execute ${1+"$@"}
# func_mode_finish arg...
func_mode_finish ()
{
$opt_debug
libs=
libdirs=
admincmds=
for opt in "$nonopt" ${1+"$@"}
do
if test -d "$opt"; then
func_append libdirs " $opt"
elif test -f "$opt"; then
if func_lalib_unsafe_p "$opt"; then
func_append libs " $opt"
else
func_warning "\`$opt' is not a valid libtool archive"
fi
else
func_fatal_error "invalid argument \`$opt'"
fi
done
if test -n "$libs"; then
if test -n "$lt_sysroot"; then
sysroot_regex=`$ECHO "$lt_sysroot" | $SED "$sed_make_literal_regex"`
sysroot_cmd="s/\([ ']\)$sysroot_regex/\1/g;"
else
sysroot_cmd=
fi
# Remove sysroot references
if $opt_dry_run; then
for lib in $libs; do
echo "removing references to $lt_sysroot and \`=' prefixes from $lib"
done
else
tmpdir=`func_mktempdir`
for lib in $libs; do
sed -e "${sysroot_cmd} s/\([ ']-[LR]\)=/\1/g; s/\([ ']\)=/\1/g" $lib \
> $tmpdir/tmp-la
mv -f $tmpdir/tmp-la $lib
done
${RM}r "$tmpdir"
fi
fi
if test -n "$finish_cmds$finish_eval" && test -n "$libdirs"; then
for libdir in $libdirs; do
if test -n "$finish_cmds"; then
# Do each command in the finish commands.
func_execute_cmds "$finish_cmds" 'admincmds="$admincmds
'"$cmd"'"'
fi
if test -n "$finish_eval"; then
# Do the single finish_eval.
eval cmds=\"$finish_eval\"
$opt_dry_run || eval "$cmds" || func_append admincmds "
$cmds"
fi
done
fi
# Exit here if they wanted silent mode.
$opt_silent && exit $EXIT_SUCCESS
if test -n "$finish_cmds$finish_eval" && test -n "$libdirs"; then
echo "----------------------------------------------------------------------"
echo "Libraries have been installed in:"
for libdir in $libdirs; do
$ECHO " $libdir"
done
echo
echo "If you ever happen to want to link against installed libraries"
echo "in a given directory, LIBDIR, you must either use libtool, and"
echo "specify the full pathname of the library, or use the \`-LLIBDIR'"
echo "flag during linking and do at least one of the following:"
if test -n "$shlibpath_var"; then
echo " - add LIBDIR to the \`$shlibpath_var' environment variable"
echo " during execution"
fi
if test -n "$runpath_var"; then
echo " - add LIBDIR to the \`$runpath_var' environment variable"
echo " during linking"
fi
if test -n "$hardcode_libdir_flag_spec"; then
libdir=LIBDIR
eval flag=\"$hardcode_libdir_flag_spec\"
$ECHO " - use the \`$flag' linker flag"
fi
if test -n "$admincmds"; then
$ECHO " - have your system administrator run these commands:$admincmds"
fi
if test -f /etc/ld.so.conf; then
echo " - have your system administrator add LIBDIR to \`/etc/ld.so.conf'"
fi
echo
echo "See any operating system documentation about shared libraries for"
case $host in
solaris2.[6789]|solaris2.1[0-9])
echo "more information, such as the ld(1), crle(1) and ld.so(8) manual"
echo "pages."
;;
*)
echo "more information, such as the ld(1) and ld.so(8) manual pages."
;;
esac
echo "----------------------------------------------------------------------"
fi
exit $EXIT_SUCCESS
}
test "$opt_mode" = finish && func_mode_finish ${1+"$@"}
# func_mode_install arg...
func_mode_install ()
{
$opt_debug
# There may be an optional sh(1) argument at the beginning of
# install_prog (especially on Windows NT).
if test "$nonopt" = "$SHELL" || test "$nonopt" = /bin/sh ||
# Allow the use of GNU shtool's install command.
case $nonopt in *shtool*) :;; *) false;; esac; then
# Aesthetically quote it.
func_quote_for_eval "$nonopt"
install_prog="$func_quote_for_eval_result "
arg=$1
shift
else
install_prog=
arg=$nonopt
fi
# The real first argument should be the name of the installation program.
# Aesthetically quote it.
func_quote_for_eval "$arg"
func_append install_prog "$func_quote_for_eval_result"
install_shared_prog=$install_prog
case " $install_prog " in
*[\\\ /]cp\ *) install_cp=: ;;
*) install_cp=false ;;
esac
# We need to accept at least all the BSD install flags.
dest=
files=
opts=
prev=
install_type=
isdir=no
stripme=
no_mode=:
for arg
do
arg2=
if test -n "$dest"; then
func_append files " $dest"
dest=$arg
continue
fi
case $arg in
-d) isdir=yes ;;
-f)
if $install_cp; then :; else
prev=$arg
fi
;;
-g | -m | -o)
prev=$arg
;;
-s)
stripme=" -s"
continue
;;
-*)
;;
*)
# If the previous option needed an argument, then skip it.
if test -n "$prev"; then
if test "x$prev" = x-m && test -n "$install_override_mode"; then
arg2=$install_override_mode
no_mode=false
fi
prev=
else
dest=$arg
continue
fi
;;
esac
# Aesthetically quote the argument.
func_quote_for_eval "$arg"
func_append install_prog " $func_quote_for_eval_result"
if test -n "$arg2"; then
func_quote_for_eval "$arg2"
fi
func_append install_shared_prog " $func_quote_for_eval_result"
done
test -z "$install_prog" && \
func_fatal_help "you must specify an install program"
test -n "$prev" && \
func_fatal_help "the \`$prev' option requires an argument"
if test -n "$install_override_mode" && $no_mode; then
if $install_cp; then :; else
func_quote_for_eval "$install_override_mode"
func_append install_shared_prog " -m $func_quote_for_eval_result"
fi
fi
if test -z "$files"; then
if test -z "$dest"; then
func_fatal_help "no file or destination specified"
else
func_fatal_help "you must specify a destination"
fi
fi
# Strip any trailing slash from the destination.
func_stripname '' '/' "$dest"
dest=$func_stripname_result
# Check to see that the destination is a directory.
test -d "$dest" && isdir=yes
if test "$isdir" = yes; then
destdir="$dest"
destname=
else
func_dirname_and_basename "$dest" "" "."
destdir="$func_dirname_result"
destname="$func_basename_result"
# Not a directory, so check to see that there is only one file specified.
set dummy $files; shift
test "$#" -gt 1 && \
func_fatal_help "\`$dest' is not a directory"
fi
case $destdir in
[\\/]* | [A-Za-z]:[\\/]*) ;;
*)
for file in $files; do
case $file in
*.lo) ;;
*)
func_fatal_help "\`$destdir' must be an absolute directory name"
;;
esac
done
;;
esac
# This variable tells wrapper scripts just to set variables rather
# than running their programs.
libtool_install_magic="$magic"
staticlibs=
future_libdirs=
current_libdirs=
for file in $files; do
# Do each installation.
case $file in
*.$libext)
# Do the static libraries later.
func_append staticlibs " $file"
;;
*.la)
func_resolve_sysroot "$file"
file=$func_resolve_sysroot_result
# Check to see that this really is a libtool archive.
func_lalib_unsafe_p "$file" \
|| func_fatal_help "\`$file' is not a valid libtool archive"
library_names=
old_library=
relink_command=
func_source "$file"
# Add the libdir to current_libdirs if it is the destination.
if test "X$destdir" = "X$libdir"; then
case "$current_libdirs " in
*" $libdir "*) ;;
*) func_append current_libdirs " $libdir" ;;
esac
else
# Note the libdir as a future libdir.
case "$future_libdirs " in
*" $libdir "*) ;;
*) func_append future_libdirs " $libdir" ;;
esac
fi
func_dirname "$file" "/" ""
dir="$func_dirname_result"
func_append dir "$objdir"
if test -n "$relink_command"; then
# Determine the prefix the user has applied to our future dir.
inst_prefix_dir=`$ECHO "$destdir" | $SED -e "s%$libdir\$%%"`
# Don't allow the user to place us outside of our expected
# location b/c this prevents finding dependent libraries that
# are installed to the same prefix.
# At present, this check doesn't affect windows .dll's that
# are installed into $libdir/../bin (currently, that works fine)
# but it's something to keep an eye on.
test "$inst_prefix_dir" = "$destdir" && \
func_fatal_error "error: cannot install \`$file' to a directory not ending in $libdir"
if test -n "$inst_prefix_dir"; then
# Stick the inst_prefix_dir data into the link command.
relink_command=`$ECHO "$relink_command" | $SED "s%@inst_prefix_dir@%-inst-prefix-dir $inst_prefix_dir%"`
else
relink_command=`$ECHO "$relink_command" | $SED "s%@inst_prefix_dir@%%"`
fi
func_warning "relinking \`$file'"
func_show_eval "$relink_command" \
'func_fatal_error "error: relink \`$file'\'' with the above command before installing it"'
fi
# See the names of the shared library.
set dummy $library_names; shift
if test -n "$1"; then
realname="$1"
shift
srcname="$realname"
test -n "$relink_command" && srcname="$realname"T
# Install the shared library and build the symlinks.
func_show_eval "$install_shared_prog $dir/$srcname $destdir/$realname" \
'exit $?'
tstripme="$stripme"
case $host_os in
cygwin* | mingw* | pw32* | cegcc*)
case $realname in
*.dll.a)
tstripme=""
;;
esac
;;
esac
if test -n "$tstripme" && test -n "$striplib"; then
func_show_eval "$striplib $destdir/$realname" 'exit $?'
fi
if test "$#" -gt 0; then
# Delete the old symlinks, and create new ones.
# Try `ln -sf' first, because the `ln' binary might depend on
# the symlink we replace! Solaris /bin/ln does not understand -f,
# so we also need to try rm && ln -s.
for linkname
do
test "$linkname" != "$realname" \
&& func_show_eval "(cd $destdir && { $LN_S -f $realname $linkname || { $RM $linkname && $LN_S $realname $linkname; }; })"
done
fi
# Do each command in the postinstall commands.
lib="$destdir/$realname"
func_execute_cmds "$postinstall_cmds" 'exit $?'
fi
# Install the pseudo-library for information purposes.
func_basename "$file"
name="$func_basename_result"
instname="$dir/$name"i
func_show_eval "$install_prog $instname $destdir/$name" 'exit $?'
# Maybe install the static library, too.
test -n "$old_library" && func_append staticlibs " $dir/$old_library"
;;
*.lo)
# Install (i.e. copy) a libtool object.
# Figure out destination file name, if it wasn't already specified.
if test -n "$destname"; then
destfile="$destdir/$destname"
else
func_basename "$file"
destfile="$func_basename_result"
destfile="$destdir/$destfile"
fi
# Deduce the name of the destination old-style object file.
case $destfile in
*.lo)
func_lo2o "$destfile"
staticdest=$func_lo2o_result
;;
*.$objext)
staticdest="$destfile"
destfile=
;;
*)
func_fatal_help "cannot copy a libtool object to \`$destfile'"
;;
esac
# Install the libtool object if requested.
test -n "$destfile" && \
func_show_eval "$install_prog $file $destfile" 'exit $?'
# Install the old object if enabled.
if test "$build_old_libs" = yes; then
# Deduce the name of the old-style object file.
func_lo2o "$file"
staticobj=$func_lo2o_result
func_show_eval "$install_prog \$staticobj \$staticdest" 'exit $?'
fi
exit $EXIT_SUCCESS
;;
*)
# Figure out destination file name, if it wasn't already specified.
if test -n "$destname"; then
destfile="$destdir/$destname"
else
func_basename "$file"
destfile="$func_basename_result"
destfile="$destdir/$destfile"
fi
# If the file is missing, and there is a .exe on the end, strip it
# because it is most likely a libtool script we actually want to
# install
stripped_ext=""
case $file in
*.exe)
if test ! -f "$file"; then
func_stripname '' '.exe' "$file"
file=$func_stripname_result
stripped_ext=".exe"
fi
;;
esac
# Do a test to see if this is really a libtool program.
case $host in
*cygwin* | *mingw*)
if func_ltwrapper_executable_p "$file"; then
func_ltwrapper_scriptname "$file"
wrapper=$func_ltwrapper_scriptname_result
else
func_stripname '' '.exe' "$file"
wrapper=$func_stripname_result
fi
;;
*)
wrapper=$file
;;
esac
if func_ltwrapper_script_p "$wrapper"; then
notinst_deplibs=
relink_command=
func_source "$wrapper"
# Check the variables that should have been set.
test -z "$generated_by_libtool_version" && \
func_fatal_error "invalid libtool wrapper script \`$wrapper'"
finalize=yes
for lib in $notinst_deplibs; do
# Check to see that each library is installed.
libdir=
if test -f "$lib"; then
func_source "$lib"
fi
libfile="$libdir/"`$ECHO "$lib" | $SED 's%^.*/%%g'` ### testsuite: skip nested quoting test
if test -n "$libdir" && test ! -f "$libfile"; then
func_warning "\`$lib' has not been installed in \`$libdir'"
finalize=no
fi
done
relink_command=
func_source "$wrapper"
outputname=
if test "$fast_install" = no && test -n "$relink_command"; then
$opt_dry_run || {
if test "$finalize" = yes; then
tmpdir=`func_mktempdir`
func_basename "$file$stripped_ext"
file="$func_basename_result"
outputname="$tmpdir/$file"
# Replace the output file specification.
relink_command=`$ECHO "$relink_command" | $SED 's%@OUTPUT@%'"$outputname"'%g'`
$opt_silent || {
func_quote_for_expand "$relink_command"
eval "func_echo $func_quote_for_expand_result"
}
if eval "$relink_command"; then :
else
func_error "error: relink \`$file' with the above command before installing it"
$opt_dry_run || ${RM}r "$tmpdir"
continue
fi
file="$outputname"
else
func_warning "cannot relink \`$file'"
fi
}
else
# Install the binary that we compiled earlier.
file=`$ECHO "$file$stripped_ext" | $SED "s%\([^/]*\)$%$objdir/\1%"`
fi
fi
# remove .exe since cygwin /usr/bin/install will append another
# one anyway
case $install_prog,$host in
*/usr/bin/install*,*cygwin*)
case $file:$destfile in
*.exe:*.exe)
# this is ok
;;
*.exe:*)
destfile=$destfile.exe
;;
*:*.exe)
func_stripname '' '.exe' "$destfile"
destfile=$func_stripname_result
;;
esac
;;
esac
func_show_eval "$install_prog\$stripme \$file \$destfile" 'exit $?'
$opt_dry_run || if test -n "$outputname"; then
${RM}r "$tmpdir"
fi
;;
esac
done
for file in $staticlibs; do
func_basename "$file"
name="$func_basename_result"
# Set up the ranlib parameters.
oldlib="$destdir/$name"
func_to_tool_file "$oldlib" func_convert_file_msys_to_w32
tool_oldlib=$func_to_tool_file_result
func_show_eval "$install_prog \$file \$oldlib" 'exit $?'
if test -n "$stripme" && test -n "$old_striplib"; then
func_show_eval "$old_striplib $tool_oldlib" 'exit $?'
fi
# Do each command in the postinstall commands.
func_execute_cmds "$old_postinstall_cmds" 'exit $?'
done
test -n "$future_libdirs" && \
func_warning "remember to run \`$progname --finish$future_libdirs'"
if test -n "$current_libdirs"; then
# Maybe just do a dry run.
$opt_dry_run && current_libdirs=" -n$current_libdirs"
exec_cmd='$SHELL $progpath $preserve_args --finish$current_libdirs'
else
exit $EXIT_SUCCESS
fi
}
test "$opt_mode" = install && func_mode_install ${1+"$@"}
# func_generate_dlsyms outputname originator pic_p
# Extract symbols from dlprefiles and create ${outputname}S.o with
# a dlpreopen symbol table.
func_generate_dlsyms ()
{
$opt_debug
my_outputname="$1"
my_originator="$2"
my_pic_p="${3-no}"
my_prefix=`$ECHO "$my_originator" | sed 's%[^a-zA-Z0-9]%_%g'`
my_dlsyms=
if test -n "$dlfiles$dlprefiles" || test "$dlself" != no; then
if test -n "$NM" && test -n "$global_symbol_pipe"; then
my_dlsyms="${my_outputname}S.c"
else
func_error "not configured to extract global symbols from dlpreopened files"
fi
fi
if test -n "$my_dlsyms"; then
case $my_dlsyms in
"") ;;
*.c)
# Discover the nlist of each of the dlfiles.
nlist="$output_objdir/${my_outputname}.nm"
func_show_eval "$RM $nlist ${nlist}S ${nlist}T"
# Parse the name list into a source file.
func_verbose "creating $output_objdir/$my_dlsyms"
$opt_dry_run || $ECHO > "$output_objdir/$my_dlsyms" "\
/* $my_dlsyms - symbol resolution table for \`$my_outputname' dlsym emulation. */
/* Generated by $PROGRAM (GNU $PACKAGE$TIMESTAMP) $VERSION */
#ifdef __cplusplus
extern \"C\" {
#endif
#if defined(__GNUC__) && (((__GNUC__ == 4) && (__GNUC_MINOR__ >= 4)) || (__GNUC__ > 4))
#pragma GCC diagnostic ignored \"-Wstrict-prototypes\"
#endif
/* Keep this code in sync between libtool.m4, ltmain, lt_system.h, and tests. */
#if defined(_WIN32) || defined(__CYGWIN__) || defined(_WIN32_WCE)
/* DATA imports from DLLs on WIN32 con't be const, because runtime
relocations are performed -- see ld's documentation on pseudo-relocs. */
# define LT_DLSYM_CONST
#elif defined(__osf__)
/* This system does not cope well with relocations in const data. */
# define LT_DLSYM_CONST
#else
# define LT_DLSYM_CONST const
#endif
/* External symbol declarations for the compiler. */\
"
if test "$dlself" = yes; then
func_verbose "generating symbol list for \`$output'"
$opt_dry_run || echo ': @PROGRAM@ ' > "$nlist"
# Add our own program objects to the symbol list.
progfiles=`$ECHO "$objs$old_deplibs" | $SP2NL | $SED "$lo2o" | $NL2SP`
for progfile in $progfiles; do
func_to_tool_file "$progfile" func_convert_file_msys_to_w32
func_verbose "extracting global C symbols from \`$func_to_tool_file_result'"
$opt_dry_run || eval "$NM $func_to_tool_file_result | $global_symbol_pipe >> '$nlist'"
done
if test -n "$exclude_expsyms"; then
$opt_dry_run || {
eval '$EGREP -v " ($exclude_expsyms)$" "$nlist" > "$nlist"T'
eval '$MV "$nlist"T "$nlist"'
}
fi
if test -n "$export_symbols_regex"; then
$opt_dry_run || {
eval '$EGREP -e "$export_symbols_regex" "$nlist" > "$nlist"T'
eval '$MV "$nlist"T "$nlist"'
}
fi
# Prepare the list of exported symbols
if test -z "$export_symbols"; then
export_symbols="$output_objdir/$outputname.exp"
$opt_dry_run || {
$RM $export_symbols
eval "${SED} -n -e '/^: @PROGRAM@ $/d' -e 's/^.* \(.*\)$/\1/p' "'< "$nlist" > "$export_symbols"'
case $host in
*cygwin* | *mingw* | *cegcc* )
eval "echo EXPORTS "'> "$output_objdir/$outputname.def"'
eval 'cat "$export_symbols" >> "$output_objdir/$outputname.def"'
;;
esac
}
else
$opt_dry_run || {
eval "${SED} -e 's/\([].[*^$]\)/\\\\\1/g' -e 's/^/ /' -e 's/$/$/'"' < "$export_symbols" > "$output_objdir/$outputname.exp"'
eval '$GREP -f "$output_objdir/$outputname.exp" < "$nlist" > "$nlist"T'
eval '$MV "$nlist"T "$nlist"'
case $host in
*cygwin* | *mingw* | *cegcc* )
eval "echo EXPORTS "'> "$output_objdir/$outputname.def"'
eval 'cat "$nlist" >> "$output_objdir/$outputname.def"'
;;
esac
}
fi
fi
for dlprefile in $dlprefiles; do
func_verbose "extracting global C symbols from \`$dlprefile'"
func_basename "$dlprefile"
name="$func_basename_result"
case $host in
*cygwin* | *mingw* | *cegcc* )
# if an import library, we need to obtain dlname
if func_win32_import_lib_p "$dlprefile"; then
func_tr_sh "$dlprefile"
eval "curr_lafile=\$libfile_$func_tr_sh_result"
dlprefile_dlbasename=""
if test -n "$curr_lafile" && func_lalib_p "$curr_lafile"; then
# Use subshell, to avoid clobbering current variable values
dlprefile_dlname=`source "$curr_lafile" && echo "$dlname"`
if test -n "$dlprefile_dlname" ; then
func_basename "$dlprefile_dlname"
dlprefile_dlbasename="$func_basename_result"
else
# no lafile. user explicitly requested -dlpreopen .
$sharedlib_from_linklib_cmd "$dlprefile"
dlprefile_dlbasename=$sharedlib_from_linklib_result
fi
fi
$opt_dry_run || {
if test -n "$dlprefile_dlbasename" ; then
eval '$ECHO ": $dlprefile_dlbasename" >> "$nlist"'
else
func_warning "Could not compute DLL name from $name"
eval '$ECHO ": $name " >> "$nlist"'
fi
func_to_tool_file "$dlprefile" func_convert_file_msys_to_w32
eval "$NM \"$func_to_tool_file_result\" 2>/dev/null | $global_symbol_pipe |
$SED -e '/I __imp/d' -e 's/I __nm_/D /;s/_nm__//' >> '$nlist'"
}
else # not an import lib
$opt_dry_run || {
eval '$ECHO ": $name " >> "$nlist"'
func_to_tool_file "$dlprefile" func_convert_file_msys_to_w32
eval "$NM \"$func_to_tool_file_result\" 2>/dev/null | $global_symbol_pipe >> '$nlist'"
}
fi
;;
*)
$opt_dry_run || {
eval '$ECHO ": $name " >> "$nlist"'
func_to_tool_file "$dlprefile" func_convert_file_msys_to_w32
eval "$NM \"$func_to_tool_file_result\" 2>/dev/null | $global_symbol_pipe >> '$nlist'"
}
;;
esac
done
$opt_dry_run || {
# Make sure we have at least an empty file.
test -f "$nlist" || : > "$nlist"
if test -n "$exclude_expsyms"; then
$EGREP -v " ($exclude_expsyms)$" "$nlist" > "$nlist"T
$MV "$nlist"T "$nlist"
fi
# Try sorting and uniquifying the output.
if $GREP -v "^: " < "$nlist" |
if sort -k 3 /dev/null 2>&1; then
sort -k 3
else
sort +2
fi |
uniq > "$nlist"S; then
:
else
$GREP -v "^: " < "$nlist" > "$nlist"S
fi
if test -f "$nlist"S; then
eval "$global_symbol_to_cdecl"' < "$nlist"S >> "$output_objdir/$my_dlsyms"'
else
echo '/* NONE */' >> "$output_objdir/$my_dlsyms"
fi
echo >> "$output_objdir/$my_dlsyms" "\
/* The mapping between symbol names and symbols. */
typedef struct {
const char *name;
void *address;
} lt_dlsymlist;
extern LT_DLSYM_CONST lt_dlsymlist
lt_${my_prefix}_LTX_preloaded_symbols[];
LT_DLSYM_CONST lt_dlsymlist
lt_${my_prefix}_LTX_preloaded_symbols[] =
{\
{ \"$my_originator\", (void *) 0 },"
case $need_lib_prefix in
no)
eval "$global_symbol_to_c_name_address" < "$nlist" >> "$output_objdir/$my_dlsyms"
;;
*)
eval "$global_symbol_to_c_name_address_lib_prefix" < "$nlist" >> "$output_objdir/$my_dlsyms"
;;
esac
echo >> "$output_objdir/$my_dlsyms" "\
{0, (void *) 0}
};
/* This works around a problem in FreeBSD linker */
#ifdef FREEBSD_WORKAROUND
static const void *lt_preloaded_setup() {
return lt_${my_prefix}_LTX_preloaded_symbols;
}
#endif
#ifdef __cplusplus
}
#endif\
"
} # !$opt_dry_run
pic_flag_for_symtable=
case "$compile_command " in
*" -static "*) ;;
*)
case $host in
# compiling the symbol table file with pic_flag works around
# a FreeBSD bug that causes programs to crash when -lm is
# linked before any other PIC object. But we must not use
# pic_flag when linking with -static. The problem exists in
# FreeBSD 2.2.6 and is fixed in FreeBSD 3.1.
*-*-freebsd2.*|*-*-freebsd3.0*|*-*-freebsdelf3.0*)
pic_flag_for_symtable=" $pic_flag -DFREEBSD_WORKAROUND" ;;
*-*-hpux*)
pic_flag_for_symtable=" $pic_flag" ;;
*)
if test "X$my_pic_p" != Xno; then
pic_flag_for_symtable=" $pic_flag"
fi
;;
esac
;;
esac
symtab_cflags=
for arg in $LTCFLAGS; do
case $arg in
-pie | -fpie | -fPIE) ;;
*) func_append symtab_cflags " $arg" ;;
esac
done
# Now compile the dynamic symbol file.
func_show_eval '(cd $output_objdir && $LTCC$symtab_cflags -c$no_builtin_flag$pic_flag_for_symtable "$my_dlsyms")' 'exit $?'
# Clean up the generated files.
func_show_eval '$RM "$output_objdir/$my_dlsyms" "$nlist" "${nlist}S" "${nlist}T"'
# Transform the symbol file into the correct name.
symfileobj="$output_objdir/${my_outputname}S.$objext"
case $host in
*cygwin* | *mingw* | *cegcc* )
if test -f "$output_objdir/$my_outputname.def"; then
compile_command=`$ECHO "$compile_command" | $SED "s%@SYMFILE@%$output_objdir/$my_outputname.def $symfileobj%"`
finalize_command=`$ECHO "$finalize_command" | $SED "s%@SYMFILE@%$output_objdir/$my_outputname.def $symfileobj%"`
else
compile_command=`$ECHO "$compile_command" | $SED "s%@SYMFILE@%$symfileobj%"`
finalize_command=`$ECHO "$finalize_command" | $SED "s%@SYMFILE@%$symfileobj%"`
fi
;;
*)
compile_command=`$ECHO "$compile_command" | $SED "s%@SYMFILE@%$symfileobj%"`
finalize_command=`$ECHO "$finalize_command" | $SED "s%@SYMFILE@%$symfileobj%"`
;;
esac
;;
*)
func_fatal_error "unknown suffix for \`$my_dlsyms'"
;;
esac
else
# We keep going just in case the user didn't refer to
# lt_preloaded_symbols. The linker will fail if global_symbol_pipe
# really was required.
# Nullify the symbol file.
compile_command=`$ECHO "$compile_command" | $SED "s% @SYMFILE@%%"`
finalize_command=`$ECHO "$finalize_command" | $SED "s% @SYMFILE@%%"`
fi
}
# func_win32_libid arg
# return the library type of file 'arg'
#
# Need a lot of goo to handle *both* DLLs and import libs
# Has to be a shell function in order to 'eat' the argument
# that is supplied when $file_magic_command is called.
# Despite the name, also deal with 64 bit binaries.
func_win32_libid ()
{
$opt_debug
win32_libid_type="unknown"
win32_fileres=`file -L $1 2>/dev/null`
case $win32_fileres in
*ar\ archive\ import\ library*) # definitely import
win32_libid_type="x86 archive import"
;;
*ar\ archive*) # could be an import, or static
# Keep the egrep pattern in sync with the one in _LT_CHECK_MAGIC_METHOD.
if eval $OBJDUMP -f $1 | $SED -e '10q' 2>/dev/null |
$EGREP 'file format (pei*-i386(.*architecture: i386)?|pe-arm-wince|pe-x86-64)' >/dev/null; then
func_to_tool_file "$1" func_convert_file_msys_to_w32
win32_nmres=`eval $NM -f posix -A \"$func_to_tool_file_result\" |
$SED -n -e '
1,100{
/ I /{
s,.*,import,
p
q
}
}'`
case $win32_nmres in
import*) win32_libid_type="x86 archive import";;
*) win32_libid_type="x86 archive static";;
esac
fi
;;
*DLL*)
win32_libid_type="x86 DLL"
;;
*executable*) # but shell scripts are "executable" too...
case $win32_fileres in
*MS\ Windows\ PE\ Intel*)
win32_libid_type="x86 DLL"
;;
esac
;;
esac
$ECHO "$win32_libid_type"
}
# func_cygming_dll_for_implib ARG
#
# Platform-specific function to extract the
# name of the DLL associated with the specified
# import library ARG.
# Invoked by eval'ing the libtool variable
# $sharedlib_from_linklib_cmd
# Result is available in the variable
# $sharedlib_from_linklib_result
func_cygming_dll_for_implib ()
{
$opt_debug
sharedlib_from_linklib_result=`$DLLTOOL --identify-strict --identify "$1"`
}
# func_cygming_dll_for_implib_fallback_core SECTION_NAME LIBNAMEs
#
# The is the core of a fallback implementation of a
# platform-specific function to extract the name of the
# DLL associated with the specified import library LIBNAME.
#
# SECTION_NAME is either .idata$6 or .idata$7, depending
# on the platform and compiler that created the implib.
#
# Echos the name of the DLL associated with the
# specified import library.
func_cygming_dll_for_implib_fallback_core ()
{
$opt_debug
match_literal=`$ECHO "$1" | $SED "$sed_make_literal_regex"`
$OBJDUMP -s --section "$1" "$2" 2>/dev/null |
$SED '/^Contents of section '"$match_literal"':/{
# Place marker at beginning of archive member dllname section
s/.*/====MARK====/
p
d
}
# These lines can sometimes be longer than 43 characters, but
# are always uninteresting
/:[ ]*file format pe[i]\{,1\}-/d
/^In archive [^:]*:/d
# Ensure marker is printed
/^====MARK====/p
# Remove all lines with less than 43 characters
/^.\{43\}/!d
# From remaining lines, remove first 43 characters
s/^.\{43\}//' |
$SED -n '
# Join marker and all lines until next marker into a single line
/^====MARK====/ b para
H
$ b para
b
:para
x
s/\n//g
# Remove the marker
s/^====MARK====//
# Remove trailing dots and whitespace
s/[\. \t]*$//
# Print
/./p' |
# we now have a list, one entry per line, of the stringified
# contents of the appropriate section of all members of the
# archive which possess that section. Heuristic: eliminate
# all those which have a first or second character that is
# a '.' (that is, objdump's representation of an unprintable
# character.) This should work for all archives with less than
# 0x302f exports -- but will fail for DLLs whose name actually
# begins with a literal '.' or a single character followed by
# a '.'.
#
# Of those that remain, print the first one.
$SED -e '/^\./d;/^.\./d;q'
}
# func_cygming_gnu_implib_p ARG
# This predicate returns with zero status (TRUE) if
# ARG is a GNU/binutils-style import library. Returns
# with nonzero status (FALSE) otherwise.
func_cygming_gnu_implib_p ()
{
$opt_debug
func_to_tool_file "$1" func_convert_file_msys_to_w32
func_cygming_gnu_implib_tmp=`$NM "$func_to_tool_file_result" | eval "$global_symbol_pipe" | $EGREP ' (_head_[A-Za-z0-9_]+_[ad]l*|[A-Za-z0-9_]+_[ad]l*_iname)$'`
test -n "$func_cygming_gnu_implib_tmp"
}
# func_cygming_ms_implib_p ARG
# This predicate returns with zero status (TRUE) if
# ARG is an MS-style import library. Returns
# with nonzero status (FALSE) otherwise.
func_cygming_ms_implib_p ()
{
$opt_debug
func_to_tool_file "$1" func_convert_file_msys_to_w32
func_cygming_ms_implib_tmp=`$NM "$func_to_tool_file_result" | eval "$global_symbol_pipe" | $GREP '_NULL_IMPORT_DESCRIPTOR'`
test -n "$func_cygming_ms_implib_tmp"
}
# func_cygming_dll_for_implib_fallback ARG
# Platform-specific function to extract the
# name of the DLL associated with the specified
# import library ARG.
#
# This fallback implementation is for use when $DLLTOOL
# does not support the --identify-strict option.
# Invoked by eval'ing the libtool variable
# $sharedlib_from_linklib_cmd
# Result is available in the variable
# $sharedlib_from_linklib_result
func_cygming_dll_for_implib_fallback ()
{
$opt_debug
if func_cygming_gnu_implib_p "$1" ; then
# binutils import library
sharedlib_from_linklib_result=`func_cygming_dll_for_implib_fallback_core '.idata$7' "$1"`
elif func_cygming_ms_implib_p "$1" ; then
# ms-generated import library
sharedlib_from_linklib_result=`func_cygming_dll_for_implib_fallback_core '.idata$6' "$1"`
else
# unknown
sharedlib_from_linklib_result=""
fi
}
# func_extract_an_archive dir oldlib
func_extract_an_archive ()
{
$opt_debug
f_ex_an_ar_dir="$1"; shift
f_ex_an_ar_oldlib="$1"
if test "$lock_old_archive_extraction" = yes; then
lockfile=$f_ex_an_ar_oldlib.lock
until $opt_dry_run || ln "$progpath" "$lockfile" 2>/dev/null; do
func_echo "Waiting for $lockfile to be removed"
sleep 2
done
fi
func_show_eval "(cd \$f_ex_an_ar_dir && $AR x \"\$f_ex_an_ar_oldlib\")" \
'stat=$?; rm -f "$lockfile"; exit $stat'
if test "$lock_old_archive_extraction" = yes; then
$opt_dry_run || rm -f "$lockfile"
fi
if ($AR t "$f_ex_an_ar_oldlib" | sort | sort -uc >/dev/null 2>&1); then
:
else
func_fatal_error "object name conflicts in archive: $f_ex_an_ar_dir/$f_ex_an_ar_oldlib"
fi
}
# func_extract_archives gentop oldlib ...
func_extract_archives ()
{
$opt_debug
my_gentop="$1"; shift
my_oldlibs=${1+"$@"}
my_oldobjs=""
my_xlib=""
my_xabs=""
my_xdir=""
for my_xlib in $my_oldlibs; do
# Extract the objects.
case $my_xlib in
[\\/]* | [A-Za-z]:[\\/]*) my_xabs="$my_xlib" ;;
*) my_xabs=`pwd`"/$my_xlib" ;;
esac
func_basename "$my_xlib"
my_xlib="$func_basename_result"
my_xlib_u=$my_xlib
while :; do
case " $extracted_archives " in
*" $my_xlib_u "*)
func_arith $extracted_serial + 1
extracted_serial=$func_arith_result
my_xlib_u=lt$extracted_serial-$my_xlib ;;
*) break ;;
esac
done
extracted_archives="$extracted_archives $my_xlib_u"
my_xdir="$my_gentop/$my_xlib_u"
func_mkdir_p "$my_xdir"
case $host in
*-darwin*)
func_verbose "Extracting $my_xabs"
# Do not bother doing anything if just a dry run
$opt_dry_run || {
darwin_orig_dir=`pwd`
cd $my_xdir || exit $?
darwin_archive=$my_xabs
darwin_curdir=`pwd`
darwin_base_archive=`basename "$darwin_archive"`
darwin_arches=`$LIPO -info "$darwin_archive" 2>/dev/null | $GREP Architectures 2>/dev/null || true`
if test -n "$darwin_arches"; then
darwin_arches=`$ECHO "$darwin_arches" | $SED -e 's/.*are://'`
darwin_arch=
func_verbose "$darwin_base_archive has multiple architectures $darwin_arches"
for darwin_arch in $darwin_arches ; do
func_mkdir_p "unfat-$$/${darwin_base_archive}-${darwin_arch}"
$LIPO -thin $darwin_arch -output "unfat-$$/${darwin_base_archive}-${darwin_arch}/${darwin_base_archive}" "${darwin_archive}"
cd "unfat-$$/${darwin_base_archive}-${darwin_arch}"
func_extract_an_archive "`pwd`" "${darwin_base_archive}"
cd "$darwin_curdir"
$RM "unfat-$$/${darwin_base_archive}-${darwin_arch}/${darwin_base_archive}"
done # $darwin_arches
## Okay now we've a bunch of thin objects, gotta fatten them up :)
darwin_filelist=`find unfat-$$ -type f -name \*.o -print -o -name \*.lo -print | $SED -e "$basename" | sort -u`
darwin_file=
darwin_files=
for darwin_file in $darwin_filelist; do
darwin_files=`find unfat-$$ -name $darwin_file -print | sort | $NL2SP`
$LIPO -create -output "$darwin_file" $darwin_files
done # $darwin_filelist
$RM -rf unfat-$$
cd "$darwin_orig_dir"
else
cd $darwin_orig_dir
func_extract_an_archive "$my_xdir" "$my_xabs"
fi # $darwin_arches
} # !$opt_dry_run
;;
*)
func_extract_an_archive "$my_xdir" "$my_xabs"
;;
esac
my_oldobjs="$my_oldobjs "`find $my_xdir -name \*.$objext -print -o -name \*.lo -print | sort | $NL2SP`
done
func_extract_archives_result="$my_oldobjs"
}
# func_emit_wrapper [arg=no]
#
# Emit a libtool wrapper script on stdout.
# Don't directly open a file because we may want to
# incorporate the script contents within a cygwin/mingw
# wrapper executable. Must ONLY be called from within
# func_mode_link because it depends on a number of variables
# set therein.
#
# ARG is the value that the WRAPPER_SCRIPT_BELONGS_IN_OBJDIR
# variable will take. If 'yes', then the emitted script
# will assume that the directory in which it is stored is
# the $objdir directory. This is a cygwin/mingw-specific
# behavior.
func_emit_wrapper ()
{
func_emit_wrapper_arg1=${1-no}
$ECHO "\
#! $SHELL
# $output - temporary wrapper script for $objdir/$outputname
# Generated by $PROGRAM (GNU $PACKAGE$TIMESTAMP) $VERSION
#
# The $output program cannot be directly executed until all the libtool
# libraries that it depends on are installed.
#
# This wrapper script should never be moved out of the build directory.
# If it is, it will not operate correctly.
# Sed substitution that helps us do robust quoting. It backslashifies
# metacharacters that are still active within double-quoted strings.
sed_quote_subst='$sed_quote_subst'
# Be Bourne compatible
if test -n \"\${ZSH_VERSION+set}\" && (emulate sh) >/dev/null 2>&1; then
emulate sh
NULLCMD=:
# Zsh 3.x and 4.x performs word splitting on \${1+\"\$@\"}, which
# is contrary to our usage. Disable this feature.
alias -g '\${1+\"\$@\"}'='\"\$@\"'
setopt NO_GLOB_SUBST
else
case \`(set -o) 2>/dev/null\` in *posix*) set -o posix;; esac
fi
BIN_SH=xpg4; export BIN_SH # for Tru64
DUALCASE=1; export DUALCASE # for MKS sh
# The HP-UX ksh and POSIX shell print the target directory to stdout
# if CDPATH is set.
(unset CDPATH) >/dev/null 2>&1 && unset CDPATH
relink_command=\"$relink_command\"
# This environment variable determines our operation mode.
if test \"\$libtool_install_magic\" = \"$magic\"; then
# install mode needs the following variables:
generated_by_libtool_version='$macro_version'
notinst_deplibs='$notinst_deplibs'
else
# When we are sourced in execute mode, \$file and \$ECHO are already set.
if test \"\$libtool_execute_magic\" != \"$magic\"; then
file=\"\$0\""
qECHO=`$ECHO "$ECHO" | $SED "$sed_quote_subst"`
$ECHO "\
# A function that is used when there is no print builtin or printf.
func_fallback_echo ()
{
eval 'cat <<_LTECHO_EOF
\$1
_LTECHO_EOF'
}
ECHO=\"$qECHO\"
fi
# Very basic option parsing. These options are (a) specific to
# the libtool wrapper, (b) are identical between the wrapper
# /script/ and the wrapper /executable/ which is used only on
# windows platforms, and (c) all begin with the string "--lt-"
# (application programs are unlikely to have options which match
# this pattern).
#
# There are only two supported options: --lt-debug and
# --lt-dump-script. There is, deliberately, no --lt-help.
#
# The first argument to this parsing function should be the
# script's $0 value, followed by "$@".
lt_option_debug=
func_parse_lt_options ()
{
lt_script_arg0=\$0
shift
for lt_opt
do
case \"\$lt_opt\" in
--lt-debug) lt_option_debug=1 ;;
--lt-dump-script)
lt_dump_D=\`\$ECHO \"X\$lt_script_arg0\" | $SED -e 's/^X//' -e 's%/[^/]*$%%'\`
test \"X\$lt_dump_D\" = \"X\$lt_script_arg0\" && lt_dump_D=.
lt_dump_F=\`\$ECHO \"X\$lt_script_arg0\" | $SED -e 's/^X//' -e 's%^.*/%%'\`
cat \"\$lt_dump_D/\$lt_dump_F\"
exit 0
;;
--lt-*)
\$ECHO \"Unrecognized --lt- option: '\$lt_opt'\" 1>&2
exit 1
;;
esac
done
# Print the debug banner immediately:
if test -n \"\$lt_option_debug\"; then
echo \"${outputname}:${output}:\${LINENO}: libtool wrapper (GNU $PACKAGE$TIMESTAMP) $VERSION\" 1>&2
fi
}
# Used when --lt-debug. Prints its arguments to stdout
# (redirection is the responsibility of the caller)
func_lt_dump_args ()
{
lt_dump_args_N=1;
for lt_arg
do
\$ECHO \"${outputname}:${output}:\${LINENO}: newargv[\$lt_dump_args_N]: \$lt_arg\"
lt_dump_args_N=\`expr \$lt_dump_args_N + 1\`
done
}
# Core function for launching the target application
func_exec_program_core ()
{
"
case $host in
# Backslashes separate directories on plain windows
*-*-mingw | *-*-os2* | *-cegcc*)
$ECHO "\
if test -n \"\$lt_option_debug\"; then
\$ECHO \"${outputname}:${output}:\${LINENO}: newargv[0]: \$progdir\\\\\$program\" 1>&2
func_lt_dump_args \${1+\"\$@\"} 1>&2
fi
exec \"\$progdir\\\\\$program\" \${1+\"\$@\"}
"
;;
*)
$ECHO "\
if test -n \"\$lt_option_debug\"; then
\$ECHO \"${outputname}:${output}:\${LINENO}: newargv[0]: \$progdir/\$program\" 1>&2
func_lt_dump_args \${1+\"\$@\"} 1>&2
fi
exec \"\$progdir/\$program\" \${1+\"\$@\"}
"
;;
esac
$ECHO "\
\$ECHO \"\$0: cannot exec \$program \$*\" 1>&2
exit 1
}
# A function to encapsulate launching the target application
# Strips options in the --lt-* namespace from \$@ and
# launches target application with the remaining arguments.
func_exec_program ()
{
case \" \$* \" in
*\\ --lt-*)
for lt_wr_arg
do
case \$lt_wr_arg in
--lt-*) ;;
*) set x \"\$@\" \"\$lt_wr_arg\"; shift;;
esac
shift
done ;;
esac
func_exec_program_core \${1+\"\$@\"}
}
# Parse options
func_parse_lt_options \"\$0\" \${1+\"\$@\"}
# Find the directory that this script lives in.
thisdir=\`\$ECHO \"\$file\" | $SED 's%/[^/]*$%%'\`
test \"x\$thisdir\" = \"x\$file\" && thisdir=.
# Follow symbolic links until we get to the real thisdir.
file=\`ls -ld \"\$file\" | $SED -n 's/.*-> //p'\`
while test -n \"\$file\"; do
destdir=\`\$ECHO \"\$file\" | $SED 's%/[^/]*\$%%'\`
# If there was a directory component, then change thisdir.
if test \"x\$destdir\" != \"x\$file\"; then
case \"\$destdir\" in
[\\\\/]* | [A-Za-z]:[\\\\/]*) thisdir=\"\$destdir\" ;;
*) thisdir=\"\$thisdir/\$destdir\" ;;
esac
fi
file=\`\$ECHO \"\$file\" | $SED 's%^.*/%%'\`
file=\`ls -ld \"\$thisdir/\$file\" | $SED -n 's/.*-> //p'\`
done
# Usually 'no', except on cygwin/mingw when embedded into
# the cwrapper.
WRAPPER_SCRIPT_BELONGS_IN_OBJDIR=$func_emit_wrapper_arg1
if test \"\$WRAPPER_SCRIPT_BELONGS_IN_OBJDIR\" = \"yes\"; then
# special case for '.'
if test \"\$thisdir\" = \".\"; then
thisdir=\`pwd\`
fi
# remove .libs from thisdir
case \"\$thisdir\" in
*[\\\\/]$objdir ) thisdir=\`\$ECHO \"\$thisdir\" | $SED 's%[\\\\/][^\\\\/]*$%%'\` ;;
$objdir ) thisdir=. ;;
esac
fi
# Try to get the absolute directory name.
absdir=\`cd \"\$thisdir\" && pwd\`
test -n \"\$absdir\" && thisdir=\"\$absdir\"
"
if test "$fast_install" = yes; then
$ECHO "\
program=lt-'$outputname'$exeext
progdir=\"\$thisdir/$objdir\"
if test ! -f \"\$progdir/\$program\" ||
{ file=\`ls -1dt \"\$progdir/\$program\" \"\$progdir/../\$program\" 2>/dev/null | ${SED} 1q\`; \\
test \"X\$file\" != \"X\$progdir/\$program\"; }; then
file=\"\$\$-\$program\"
if test ! -d \"\$progdir\"; then
$MKDIR \"\$progdir\"
else
$RM \"\$progdir/\$file\"
fi"
$ECHO "\
# relink executable if necessary
if test -n \"\$relink_command\"; then
if relink_command_output=\`eval \$relink_command 2>&1\`; then :
else
$ECHO \"\$relink_command_output\" >&2
$RM \"\$progdir/\$file\"
exit 1
fi
fi
$MV \"\$progdir/\$file\" \"\$progdir/\$program\" 2>/dev/null ||
{ $RM \"\$progdir/\$program\";
$MV \"\$progdir/\$file\" \"\$progdir/\$program\"; }
$RM \"\$progdir/\$file\"
fi"
else
$ECHO "\
program='$outputname'
progdir=\"\$thisdir/$objdir\"
"
fi
$ECHO "\
if test -f \"\$progdir/\$program\"; then"
# fixup the dll searchpath if we need to.
#
# Fix the DLL searchpath if we need to. Do this before prepending
# to shlibpath, because on Windows, both are PATH and uninstalled
# libraries must come first.
if test -n "$dllsearchpath"; then
$ECHO "\
# Add the dll search path components to the executable PATH
PATH=$dllsearchpath:\$PATH
"
fi
# Export our shlibpath_var if we have one.
if test "$shlibpath_overrides_runpath" = yes && test -n "$shlibpath_var" && test -n "$temp_rpath"; then
$ECHO "\
# Add our own library path to $shlibpath_var
$shlibpath_var=\"$temp_rpath\$$shlibpath_var\"
# Some systems cannot cope with colon-terminated $shlibpath_var
# The second colon is a workaround for a bug in BeOS R4 sed
$shlibpath_var=\`\$ECHO \"\$$shlibpath_var\" | $SED 's/::*\$//'\`
export $shlibpath_var
"
fi
$ECHO "\
if test \"\$libtool_execute_magic\" != \"$magic\"; then
# Run the actual program with our arguments.
func_exec_program \${1+\"\$@\"}
fi
else
# The program doesn't exist.
\$ECHO \"\$0: error: \\\`\$progdir/\$program' does not exist\" 1>&2
\$ECHO \"This script is just a wrapper for \$program.\" 1>&2
\$ECHO \"See the $PACKAGE documentation for more information.\" 1>&2
exit 1
fi
fi\
"
}
# func_emit_cwrapperexe_src
# emit the source code for a wrapper executable on stdout
# Must ONLY be called from within func_mode_link because
# it depends on a number of variable set therein.
func_emit_cwrapperexe_src ()
{
cat <
#include
#ifdef _MSC_VER
# include
# include
# include
#else
# include
# include
# ifdef __CYGWIN__
# include
# endif
#endif
#include
#include
#include
#include
#include
#include
#include
#include
/* declarations of non-ANSI functions */
#if defined(__MINGW32__)
# ifdef __STRICT_ANSI__
int _putenv (const char *);
# endif
#elif defined(__CYGWIN__)
# ifdef __STRICT_ANSI__
char *realpath (const char *, char *);
int putenv (char *);
int setenv (const char *, const char *, int);
# endif
/* #elif defined (other platforms) ... */
#endif
/* portability defines, excluding path handling macros */
#if defined(_MSC_VER)
# define setmode _setmode
# define stat _stat
# define chmod _chmod
# define getcwd _getcwd
# define putenv _putenv
# define S_IXUSR _S_IEXEC
# ifndef _INTPTR_T_DEFINED
# define _INTPTR_T_DEFINED
# define intptr_t int
# endif
#elif defined(__MINGW32__)
# define setmode _setmode
# define stat _stat
# define chmod _chmod
# define getcwd _getcwd
# define putenv _putenv
#elif defined(__CYGWIN__)
# define HAVE_SETENV
# define FOPEN_WB "wb"
/* #elif defined (other platforms) ... */
#endif
#if defined(PATH_MAX)
# define LT_PATHMAX PATH_MAX
#elif defined(MAXPATHLEN)
# define LT_PATHMAX MAXPATHLEN
#else
# define LT_PATHMAX 1024
#endif
#ifndef S_IXOTH
# define S_IXOTH 0
#endif
#ifndef S_IXGRP
# define S_IXGRP 0
#endif
/* path handling portability macros */
#ifndef DIR_SEPARATOR
# define DIR_SEPARATOR '/'
# define PATH_SEPARATOR ':'
#endif
#if defined (_WIN32) || defined (__MSDOS__) || defined (__DJGPP__) || \
defined (__OS2__)
# define HAVE_DOS_BASED_FILE_SYSTEM
# define FOPEN_WB "wb"
# ifndef DIR_SEPARATOR_2
# define DIR_SEPARATOR_2 '\\'
# endif
# ifndef PATH_SEPARATOR_2
# define PATH_SEPARATOR_2 ';'
# endif
#endif
#ifndef DIR_SEPARATOR_2
# define IS_DIR_SEPARATOR(ch) ((ch) == DIR_SEPARATOR)
#else /* DIR_SEPARATOR_2 */
# define IS_DIR_SEPARATOR(ch) \
(((ch) == DIR_SEPARATOR) || ((ch) == DIR_SEPARATOR_2))
#endif /* DIR_SEPARATOR_2 */
#ifndef PATH_SEPARATOR_2
# define IS_PATH_SEPARATOR(ch) ((ch) == PATH_SEPARATOR)
#else /* PATH_SEPARATOR_2 */
# define IS_PATH_SEPARATOR(ch) ((ch) == PATH_SEPARATOR_2)
#endif /* PATH_SEPARATOR_2 */
#ifndef FOPEN_WB
# define FOPEN_WB "w"
#endif
#ifndef _O_BINARY
# define _O_BINARY 0
#endif
#define XMALLOC(type, num) ((type *) xmalloc ((num) * sizeof(type)))
#define XFREE(stale) do { \
if (stale) { free ((void *) stale); stale = 0; } \
} while (0)
#if defined(LT_DEBUGWRAPPER)
static int lt_debug = 1;
#else
static int lt_debug = 0;
#endif
const char *program_name = "libtool-wrapper"; /* in case xstrdup fails */
void *xmalloc (size_t num);
char *xstrdup (const char *string);
const char *base_name (const char *name);
char *find_executable (const char *wrapper);
char *chase_symlinks (const char *pathspec);
int make_executable (const char *path);
int check_executable (const char *path);
char *strendzap (char *str, const char *pat);
void lt_debugprintf (const char *file, int line, const char *fmt, ...);
void lt_fatal (const char *file, int line, const char *message, ...);
static const char *nonnull (const char *s);
static const char *nonempty (const char *s);
void lt_setenv (const char *name, const char *value);
char *lt_extend_str (const char *orig_value, const char *add, int to_end);
void lt_update_exe_path (const char *name, const char *value);
void lt_update_lib_path (const char *name, const char *value);
char **prepare_spawn (char **argv);
void lt_dump_script (FILE *f);
EOF
cat <= 0)
&& (st.st_mode & (S_IXUSR | S_IXGRP | S_IXOTH)))
return 1;
else
return 0;
}
int
make_executable (const char *path)
{
int rval = 0;
struct stat st;
lt_debugprintf (__FILE__, __LINE__, "(make_executable): %s\n",
nonempty (path));
if ((!path) || (!*path))
return 0;
if (stat (path, &st) >= 0)
{
rval = chmod (path, st.st_mode | S_IXOTH | S_IXGRP | S_IXUSR);
}
return rval;
}
/* Searches for the full path of the wrapper. Returns
newly allocated full path name if found, NULL otherwise
Does not chase symlinks, even on platforms that support them.
*/
char *
find_executable (const char *wrapper)
{
int has_slash = 0;
const char *p;
const char *p_next;
/* static buffer for getcwd */
char tmp[LT_PATHMAX + 1];
int tmp_len;
char *concat_name;
lt_debugprintf (__FILE__, __LINE__, "(find_executable): %s\n",
nonempty (wrapper));
if ((wrapper == NULL) || (*wrapper == '\0'))
return NULL;
/* Absolute path? */
#if defined (HAVE_DOS_BASED_FILE_SYSTEM)
if (isalpha ((unsigned char) wrapper[0]) && wrapper[1] == ':')
{
concat_name = xstrdup (wrapper);
if (check_executable (concat_name))
return concat_name;
XFREE (concat_name);
}
else
{
#endif
if (IS_DIR_SEPARATOR (wrapper[0]))
{
concat_name = xstrdup (wrapper);
if (check_executable (concat_name))
return concat_name;
XFREE (concat_name);
}
#if defined (HAVE_DOS_BASED_FILE_SYSTEM)
}
#endif
for (p = wrapper; *p; p++)
if (*p == '/')
{
has_slash = 1;
break;
}
if (!has_slash)
{
/* no slashes; search PATH */
const char *path = getenv ("PATH");
if (path != NULL)
{
for (p = path; *p; p = p_next)
{
const char *q;
size_t p_len;
for (q = p; *q; q++)
if (IS_PATH_SEPARATOR (*q))
break;
p_len = q - p;
p_next = (*q == '\0' ? q : q + 1);
if (p_len == 0)
{
/* empty path: current directory */
if (getcwd (tmp, LT_PATHMAX) == NULL)
lt_fatal (__FILE__, __LINE__, "getcwd failed: %s",
nonnull (strerror (errno)));
tmp_len = strlen (tmp);
concat_name =
XMALLOC (char, tmp_len + 1 + strlen (wrapper) + 1);
memcpy (concat_name, tmp, tmp_len);
concat_name[tmp_len] = '/';
strcpy (concat_name + tmp_len + 1, wrapper);
}
else
{
concat_name =
XMALLOC (char, p_len + 1 + strlen (wrapper) + 1);
memcpy (concat_name, p, p_len);
concat_name[p_len] = '/';
strcpy (concat_name + p_len + 1, wrapper);
}
if (check_executable (concat_name))
return concat_name;
XFREE (concat_name);
}
}
/* not found in PATH; assume curdir */
}
/* Relative path | not found in path: prepend cwd */
if (getcwd (tmp, LT_PATHMAX) == NULL)
lt_fatal (__FILE__, __LINE__, "getcwd failed: %s",
nonnull (strerror (errno)));
tmp_len = strlen (tmp);
concat_name = XMALLOC (char, tmp_len + 1 + strlen (wrapper) + 1);
memcpy (concat_name, tmp, tmp_len);
concat_name[tmp_len] = '/';
strcpy (concat_name + tmp_len + 1, wrapper);
if (check_executable (concat_name))
return concat_name;
XFREE (concat_name);
return NULL;
}
char *
chase_symlinks (const char *pathspec)
{
#ifndef S_ISLNK
return xstrdup (pathspec);
#else
char buf[LT_PATHMAX];
struct stat s;
char *tmp_pathspec = xstrdup (pathspec);
char *p;
int has_symlinks = 0;
while (strlen (tmp_pathspec) && !has_symlinks)
{
lt_debugprintf (__FILE__, __LINE__,
"checking path component for symlinks: %s\n",
tmp_pathspec);
if (lstat (tmp_pathspec, &s) == 0)
{
if (S_ISLNK (s.st_mode) != 0)
{
has_symlinks = 1;
break;
}
/* search backwards for last DIR_SEPARATOR */
p = tmp_pathspec + strlen (tmp_pathspec) - 1;
while ((p > tmp_pathspec) && (!IS_DIR_SEPARATOR (*p)))
p--;
if ((p == tmp_pathspec) && (!IS_DIR_SEPARATOR (*p)))
{
/* no more DIR_SEPARATORS left */
break;
}
*p = '\0';
}
else
{
lt_fatal (__FILE__, __LINE__,
"error accessing file \"%s\": %s",
tmp_pathspec, nonnull (strerror (errno)));
}
}
XFREE (tmp_pathspec);
if (!has_symlinks)
{
return xstrdup (pathspec);
}
tmp_pathspec = realpath (pathspec, buf);
if (tmp_pathspec == 0)
{
lt_fatal (__FILE__, __LINE__,
"could not follow symlinks for %s", pathspec);
}
return xstrdup (tmp_pathspec);
#endif
}
char *
strendzap (char *str, const char *pat)
{
size_t len, patlen;
assert (str != NULL);
assert (pat != NULL);
len = strlen (str);
patlen = strlen (pat);
if (patlen <= len)
{
str += len - patlen;
if (strcmp (str, pat) == 0)
*str = '\0';
}
return str;
}
void
lt_debugprintf (const char *file, int line, const char *fmt, ...)
{
va_list args;
if (lt_debug)
{
(void) fprintf (stderr, "%s:%s:%d: ", program_name, file, line);
va_start (args, fmt);
(void) vfprintf (stderr, fmt, args);
va_end (args);
}
}
static void
lt_error_core (int exit_status, const char *file,
int line, const char *mode,
const char *message, va_list ap)
{
fprintf (stderr, "%s:%s:%d: %s: ", program_name, file, line, mode);
vfprintf (stderr, message, ap);
fprintf (stderr, ".\n");
if (exit_status >= 0)
exit (exit_status);
}
void
lt_fatal (const char *file, int line, const char *message, ...)
{
va_list ap;
va_start (ap, message);
lt_error_core (EXIT_FAILURE, file, line, "FATAL", message, ap);
va_end (ap);
}
static const char *
nonnull (const char *s)
{
return s ? s : "(null)";
}
static const char *
nonempty (const char *s)
{
return (s && !*s) ? "(empty)" : nonnull (s);
}
void
lt_setenv (const char *name, const char *value)
{
lt_debugprintf (__FILE__, __LINE__,
"(lt_setenv) setting '%s' to '%s'\n",
nonnull (name), nonnull (value));
{
#ifdef HAVE_SETENV
/* always make a copy, for consistency with !HAVE_SETENV */
char *str = xstrdup (value);
setenv (name, str, 1);
#else
int len = strlen (name) + 1 + strlen (value) + 1;
char *str = XMALLOC (char, len);
sprintf (str, "%s=%s", name, value);
if (putenv (str) != EXIT_SUCCESS)
{
XFREE (str);
}
#endif
}
}
char *
lt_extend_str (const char *orig_value, const char *add, int to_end)
{
char *new_value;
if (orig_value && *orig_value)
{
int orig_value_len = strlen (orig_value);
int add_len = strlen (add);
new_value = XMALLOC (char, add_len + orig_value_len + 1);
if (to_end)
{
strcpy (new_value, orig_value);
strcpy (new_value + orig_value_len, add);
}
else
{
strcpy (new_value, add);
strcpy (new_value + add_len, orig_value);
}
}
else
{
new_value = xstrdup (add);
}
return new_value;
}
void
lt_update_exe_path (const char *name, const char *value)
{
lt_debugprintf (__FILE__, __LINE__,
"(lt_update_exe_path) modifying '%s' by prepending '%s'\n",
nonnull (name), nonnull (value));
if (name && *name && value && *value)
{
char *new_value = lt_extend_str (getenv (name), value, 0);
/* some systems can't cope with a ':'-terminated path #' */
int len = strlen (new_value);
while (((len = strlen (new_value)) > 0) && IS_PATH_SEPARATOR (new_value[len-1]))
{
new_value[len-1] = '\0';
}
lt_setenv (name, new_value);
XFREE (new_value);
}
}
void
lt_update_lib_path (const char *name, const char *value)
{
lt_debugprintf (__FILE__, __LINE__,
"(lt_update_lib_path) modifying '%s' by prepending '%s'\n",
nonnull (name), nonnull (value));
if (name && *name && value && *value)
{
char *new_value = lt_extend_str (getenv (name), value, 0);
lt_setenv (name, new_value);
XFREE (new_value);
}
}
EOF
case $host_os in
mingw*)
cat <<"EOF"
/* Prepares an argument vector before calling spawn().
Note that spawn() does not by itself call the command interpreter
(getenv ("COMSPEC") != NULL ? getenv ("COMSPEC") :
({ OSVERSIONINFO v; v.dwOSVersionInfoSize = sizeof(OSVERSIONINFO);
GetVersionEx(&v);
v.dwPlatformId == VER_PLATFORM_WIN32_NT;
}) ? "cmd.exe" : "command.com").
Instead it simply concatenates the arguments, separated by ' ', and calls
CreateProcess(). We must quote the arguments since Win32 CreateProcess()
interprets characters like ' ', '\t', '\\', '"' (but not '<' and '>') in a
special way:
- Space and tab are interpreted as delimiters. They are not treated as
delimiters if they are surrounded by double quotes: "...".
- Unescaped double quotes are removed from the input. Their only effect is
that within double quotes, space and tab are treated like normal
characters.
- Backslashes not followed by double quotes are not special.
- But 2*n+1 backslashes followed by a double quote become
n backslashes followed by a double quote (n >= 0):
\" -> "
\\\" -> \"
\\\\\" -> \\"
*/
#define SHELL_SPECIAL_CHARS "\"\\ \001\002\003\004\005\006\007\010\011\012\013\014\015\016\017\020\021\022\023\024\025\026\027\030\031\032\033\034\035\036\037"
#define SHELL_SPACE_CHARS " \001\002\003\004\005\006\007\010\011\012\013\014\015\016\017\020\021\022\023\024\025\026\027\030\031\032\033\034\035\036\037"
char **
prepare_spawn (char **argv)
{
size_t argc;
char **new_argv;
size_t i;
/* Count number of arguments. */
for (argc = 0; argv[argc] != NULL; argc++)
;
/* Allocate new argument vector. */
new_argv = XMALLOC (char *, argc + 1);
/* Put quoted arguments into the new argument vector. */
for (i = 0; i < argc; i++)
{
const char *string = argv[i];
if (string[0] == '\0')
new_argv[i] = xstrdup ("\"\"");
else if (strpbrk (string, SHELL_SPECIAL_CHARS) != NULL)
{
int quote_around = (strpbrk (string, SHELL_SPACE_CHARS) != NULL);
size_t length;
unsigned int backslashes;
const char *s;
char *quoted_string;
char *p;
length = 0;
backslashes = 0;
if (quote_around)
length++;
for (s = string; *s != '\0'; s++)
{
char c = *s;
if (c == '"')
length += backslashes + 1;
length++;
if (c == '\\')
backslashes++;
else
backslashes = 0;
}
if (quote_around)
length += backslashes + 1;
quoted_string = XMALLOC (char, length + 1);
p = quoted_string;
backslashes = 0;
if (quote_around)
*p++ = '"';
for (s = string; *s != '\0'; s++)
{
char c = *s;
if (c == '"')
{
unsigned int j;
for (j = backslashes + 1; j > 0; j--)
*p++ = '\\';
}
*p++ = c;
if (c == '\\')
backslashes++;
else
backslashes = 0;
}
if (quote_around)
{
unsigned int j;
for (j = backslashes; j > 0; j--)
*p++ = '\\';
*p++ = '"';
}
*p = '\0';
new_argv[i] = quoted_string;
}
else
new_argv[i] = (char *) string;
}
new_argv[argc] = NULL;
return new_argv;
}
EOF
;;
esac
cat <<"EOF"
void lt_dump_script (FILE* f)
{
EOF
func_emit_wrapper yes |
$SED -n -e '
s/^\(.\{79\}\)\(..*\)/\1\
\2/
h
s/\([\\"]\)/\\\1/g
s/$/\\n/
s/\([^\n]*\).*/ fputs ("\1", f);/p
g
D'
cat <<"EOF"
}
EOF
}
# end: func_emit_cwrapperexe_src
# func_win32_import_lib_p ARG
# True if ARG is an import lib, as indicated by $file_magic_cmd
func_win32_import_lib_p ()
{
$opt_debug
case `eval $file_magic_cmd \"\$1\" 2>/dev/null | $SED -e 10q` in
*import*) : ;;
*) false ;;
esac
}
# func_mode_link arg...
func_mode_link ()
{
$opt_debug
case $host in
*-*-cygwin* | *-*-mingw* | *-*-pw32* | *-*-os2* | *-cegcc*)
# It is impossible to link a dll without this setting, and
# we shouldn't force the makefile maintainer to figure out
# which system we are compiling for in order to pass an extra
# flag for every libtool invocation.
# allow_undefined=no
# FIXME: Unfortunately, there are problems with the above when trying
# to make a dll which has undefined symbols, in which case not
# even a static library is built. For now, we need to specify
# -no-undefined on the libtool link line when we can be certain
# that all symbols are satisfied, otherwise we get a static library.
allow_undefined=yes
;;
*)
allow_undefined=yes
;;
esac
libtool_args=$nonopt
base_compile="$nonopt $@"
compile_command=$nonopt
finalize_command=$nonopt
compile_rpath=
finalize_rpath=
compile_shlibpath=
finalize_shlibpath=
convenience=
old_convenience=
deplibs=
old_deplibs=
compiler_flags=
linker_flags=
dllsearchpath=
lib_search_path=`pwd`
inst_prefix_dir=
new_inherited_linker_flags=
avoid_version=no
bindir=
dlfiles=
dlprefiles=
dlself=no
export_dynamic=no
export_symbols=
export_symbols_regex=
generated=
libobjs=
ltlibs=
module=no
no_install=no
objs=
non_pic_objects=
precious_files_regex=
prefer_static_libs=no
preload=no
prev=
prevarg=
release=
rpath=
xrpath=
perm_rpath=
temp_rpath=
thread_safe=no
vinfo=
vinfo_number=no
weak_libs=
single_module="${wl}-single_module"
func_infer_tag $base_compile
# We need to know -static, to get the right output filenames.
for arg
do
case $arg in
-shared)
test "$build_libtool_libs" != yes && \
func_fatal_configuration "can not build a shared library"
build_old_libs=no
break
;;
-all-static | -static | -static-libtool-libs)
case $arg in
-all-static)
if test "$build_libtool_libs" = yes && test -z "$link_static_flag"; then
func_warning "complete static linking is impossible in this configuration"
fi
if test -n "$link_static_flag"; then
dlopen_self=$dlopen_self_static
fi
prefer_static_libs=yes
;;
-static)
if test -z "$pic_flag" && test -n "$link_static_flag"; then
dlopen_self=$dlopen_self_static
fi
prefer_static_libs=built
;;
-static-libtool-libs)
if test -z "$pic_flag" && test -n "$link_static_flag"; then
dlopen_self=$dlopen_self_static
fi
prefer_static_libs=yes
;;
esac
build_libtool_libs=no
build_old_libs=yes
break
;;
esac
done
# See if our shared archives depend on static archives.
test -n "$old_archive_from_new_cmds" && build_old_libs=yes
# Go through the arguments, transforming them on the way.
while test "$#" -gt 0; do
arg="$1"
shift
func_quote_for_eval "$arg"
qarg=$func_quote_for_eval_unquoted_result
func_append libtool_args " $func_quote_for_eval_result"
# If the previous option needs an argument, assign it.
if test -n "$prev"; then
case $prev in
output)
func_append compile_command " @OUTPUT@"
func_append finalize_command " @OUTPUT@"
;;
esac
case $prev in
bindir)
bindir="$arg"
prev=
continue
;;
dlfiles|dlprefiles)
if test "$preload" = no; then
# Add the symbol object into the linking commands.
func_append compile_command " @SYMFILE@"
func_append finalize_command " @SYMFILE@"
preload=yes
fi
case $arg in
*.la | *.lo) ;; # We handle these cases below.
force)
if test "$dlself" = no; then
dlself=needless
export_dynamic=yes
fi
prev=
continue
;;
self)
if test "$prev" = dlprefiles; then
dlself=yes
elif test "$prev" = dlfiles && test "$dlopen_self" != yes; then
dlself=yes
else
dlself=needless
export_dynamic=yes
fi
prev=
continue
;;
*)
if test "$prev" = dlfiles; then
func_append dlfiles " $arg"
else
func_append dlprefiles " $arg"
fi
prev=
continue
;;
esac
;;
expsyms)
export_symbols="$arg"
test -f "$arg" \
|| func_fatal_error "symbol file \`$arg' does not exist"
prev=
continue
;;
expsyms_regex)
export_symbols_regex="$arg"
prev=
continue
;;
framework)
case $host in
*-*-darwin*)
case "$deplibs " in
*" $qarg.ltframework "*) ;;
*) func_append deplibs " $qarg.ltframework" # this is fixed later
;;
esac
;;
esac
prev=
continue
;;
inst_prefix)
inst_prefix_dir="$arg"
prev=
continue
;;
objectlist)
if test -f "$arg"; then
save_arg=$arg
moreargs=
for fil in `cat "$save_arg"`
do
# func_append moreargs " $fil"
arg=$fil
# A libtool-controlled object.
# Check to see that this really is a libtool object.
if func_lalib_unsafe_p "$arg"; then
pic_object=
non_pic_object=
# Read the .lo file
func_source "$arg"
if test -z "$pic_object" ||
test -z "$non_pic_object" ||
test "$pic_object" = none &&
test "$non_pic_object" = none; then
func_fatal_error "cannot find name of object for \`$arg'"
fi
# Extract subdirectory from the argument.
func_dirname "$arg" "/" ""
xdir="$func_dirname_result"
if test "$pic_object" != none; then
# Prepend the subdirectory the object is found in.
pic_object="$xdir$pic_object"
if test "$prev" = dlfiles; then
if test "$build_libtool_libs" = yes && test "$dlopen_support" = yes; then
func_append dlfiles " $pic_object"
prev=
continue
else
# If libtool objects are unsupported, then we need to preload.
prev=dlprefiles
fi
fi
# CHECK ME: I think I busted this. -Ossama
if test "$prev" = dlprefiles; then
# Preload the old-style object.
func_append dlprefiles " $pic_object"
prev=
fi
# A PIC object.
func_append libobjs " $pic_object"
arg="$pic_object"
fi
# Non-PIC object.
if test "$non_pic_object" != none; then
# Prepend the subdirectory the object is found in.
non_pic_object="$xdir$non_pic_object"
# A standard non-PIC object
func_append non_pic_objects " $non_pic_object"
if test -z "$pic_object" || test "$pic_object" = none ; then
arg="$non_pic_object"
fi
else
# If the PIC object exists, use it instead.
# $xdir was prepended to $pic_object above.
non_pic_object="$pic_object"
func_append non_pic_objects " $non_pic_object"
fi
else
# Only an error if not doing a dry-run.
if $opt_dry_run; then
# Extract subdirectory from the argument.
func_dirname "$arg" "/" ""
xdir="$func_dirname_result"
func_lo2o "$arg"
pic_object=$xdir$objdir/$func_lo2o_result
non_pic_object=$xdir$func_lo2o_result
func_append libobjs " $pic_object"
func_append non_pic_objects " $non_pic_object"
else
func_fatal_error "\`$arg' is not a valid libtool object"
fi
fi
done
else
func_fatal_error "link input file \`$arg' does not exist"
fi
arg=$save_arg
prev=
continue
;;
precious_regex)
precious_files_regex="$arg"
prev=
continue
;;
release)
release="-$arg"
prev=
continue
;;
rpath | xrpath)
# We need an absolute path.
case $arg in
[\\/]* | [A-Za-z]:[\\/]*) ;;
*)
func_fatal_error "only absolute run-paths are allowed"
;;
esac
if test "$prev" = rpath; then
case "$rpath " in
*" $arg "*) ;;
*) func_append rpath " $arg" ;;
esac
else
case "$xrpath " in
*" $arg "*) ;;
*) func_append xrpath " $arg" ;;
esac
fi
prev=
continue
;;
shrext)
shrext_cmds="$arg"
prev=
continue
;;
weak)
func_append weak_libs " $arg"
prev=
continue
;;
xcclinker)
func_append linker_flags " $qarg"
func_append compiler_flags " $qarg"
prev=
func_append compile_command " $qarg"
func_append finalize_command " $qarg"
continue
;;
xcompiler)
func_append compiler_flags " $qarg"
prev=
func_append compile_command " $qarg"
func_append finalize_command " $qarg"
continue
;;
xlinker)
func_append linker_flags " $qarg"
func_append compiler_flags " $wl$qarg"
prev=
func_append compile_command " $wl$qarg"
func_append finalize_command " $wl$qarg"
continue
;;
*)
eval "$prev=\"\$arg\""
prev=
continue
;;
esac
fi # test -n "$prev"
prevarg="$arg"
case $arg in
-all-static)
if test -n "$link_static_flag"; then
# See comment for -static flag below, for more details.
func_append compile_command " $link_static_flag"
func_append finalize_command " $link_static_flag"
fi
continue
;;
-allow-undefined)
# FIXME: remove this flag sometime in the future.
func_fatal_error "\`-allow-undefined' must not be used because it is the default"
;;
-avoid-version)
avoid_version=yes
continue
;;
-bindir)
prev=bindir
continue
;;
-dlopen)
prev=dlfiles
continue
;;
-dlpreopen)
prev=dlprefiles
continue
;;
-export-dynamic)
export_dynamic=yes
continue
;;
-export-symbols | -export-symbols-regex)
if test -n "$export_symbols" || test -n "$export_symbols_regex"; then
func_fatal_error "more than one -exported-symbols argument is not allowed"
fi
if test "X$arg" = "X-export-symbols"; then
prev=expsyms
else
prev=expsyms_regex
fi
continue
;;
-framework)
prev=framework
continue
;;
-inst-prefix-dir)
prev=inst_prefix
continue
;;
# The native IRIX linker understands -LANG:*, -LIST:* and -LNO:*
# so, if we see these flags be careful not to treat them like -L
-L[A-Z][A-Z]*:*)
case $with_gcc/$host in
no/*-*-irix* | /*-*-irix*)
func_append compile_command " $arg"
func_append finalize_command " $arg"
;;
esac
continue
;;
-L*)
func_stripname "-L" '' "$arg"
if test -z "$func_stripname_result"; then
if test "$#" -gt 0; then
func_fatal_error "require no space between \`-L' and \`$1'"
else
func_fatal_error "need path for \`-L' option"
fi
fi
func_resolve_sysroot "$func_stripname_result"
dir=$func_resolve_sysroot_result
# We need an absolute path.
case $dir in
[\\/]* | [A-Za-z]:[\\/]*) ;;
*)
absdir=`cd "$dir" && pwd`
test -z "$absdir" && \
func_fatal_error "cannot determine absolute directory name of \`$dir'"
dir="$absdir"
;;
esac
case "$deplibs " in
*" -L$dir "* | *" $arg "*)
# Will only happen for absolute or sysroot arguments
;;
*)
# Preserve sysroot, but never include relative directories
case $dir in
[\\/]* | [A-Za-z]:[\\/]* | =*) func_append deplibs " $arg" ;;
*) func_append deplibs " -L$dir" ;;
esac
func_append lib_search_path " $dir"
;;
esac
case $host in
*-*-cygwin* | *-*-mingw* | *-*-pw32* | *-*-os2* | *-cegcc*)
testbindir=`$ECHO "$dir" | $SED 's*/lib$*/bin*'`
case :$dllsearchpath: in
*":$dir:"*) ;;
::) dllsearchpath=$dir;;
*) func_append dllsearchpath ":$dir";;
esac
case :$dllsearchpath: in
*":$testbindir:"*) ;;
::) dllsearchpath=$testbindir;;
*) func_append dllsearchpath ":$testbindir";;
esac
;;
esac
continue
;;
-l*)
if test "X$arg" = "X-lc" || test "X$arg" = "X-lm"; then
case $host in
*-*-cygwin* | *-*-mingw* | *-*-pw32* | *-*-beos* | *-cegcc* | *-*-haiku*)
# These systems don't actually have a C or math library (as such)
continue
;;
*-*-os2*)
# These systems don't actually have a C library (as such)
test "X$arg" = "X-lc" && continue
;;
*-*-openbsd* | *-*-freebsd* | *-*-dragonfly*)
# Do not include libc due to us having libc/libc_r.
test "X$arg" = "X-lc" && continue
;;
*-*-rhapsody* | *-*-darwin1.[012])
# Rhapsody C and math libraries are in the System framework
func_append deplibs " System.ltframework"
continue
;;
*-*-sco3.2v5* | *-*-sco5v6*)
# Causes problems with __ctype
test "X$arg" = "X-lc" && continue
;;
*-*-sysv4.2uw2* | *-*-sysv5* | *-*-unixware* | *-*-OpenUNIX*)
# Compiler inserts libc in the correct place for threads to work
test "X$arg" = "X-lc" && continue
;;
esac
elif test "X$arg" = "X-lc_r"; then
case $host in
*-*-openbsd* | *-*-freebsd* | *-*-dragonfly*)
# Do not include libc_r directly, use -pthread flag.
continue
;;
esac
fi
func_append deplibs " $arg"
continue
;;
-module)
module=yes
continue
;;
# Tru64 UNIX uses -model [arg] to determine the layout of C++
# classes, name mangling, and exception handling.
# Darwin uses the -arch flag to determine output architecture.
-model|-arch|-isysroot|--sysroot)
func_append compiler_flags " $arg"
func_append compile_command " $arg"
func_append finalize_command " $arg"
prev=xcompiler
continue
;;
-mt|-mthreads|-kthread|-Kthread|-pthread|-pthreads|--thread-safe \
|-threads|-fopenmp|-openmp|-mp|-xopenmp|-omp|-qsmp=*)
func_append compiler_flags " $arg"
func_append compile_command " $arg"
func_append finalize_command " $arg"
case "$new_inherited_linker_flags " in
*" $arg "*) ;;
* ) func_append new_inherited_linker_flags " $arg" ;;
esac
continue
;;
-multi_module)
single_module="${wl}-multi_module"
continue
;;
-no-fast-install)
fast_install=no
continue
;;
-no-install)
case $host in
*-*-cygwin* | *-*-mingw* | *-*-pw32* | *-*-os2* | *-*-darwin* | *-cegcc*)
# The PATH hackery in wrapper scripts is required on Windows
# and Darwin in order for the loader to find any dlls it needs.
func_warning "\`-no-install' is ignored for $host"
func_warning "assuming \`-no-fast-install' instead"
fast_install=no
;;
*) no_install=yes ;;
esac
continue
;;
-no-undefined)
allow_undefined=no
continue
;;
-objectlist)
prev=objectlist
continue
;;
-o) prev=output ;;
-precious-files-regex)
prev=precious_regex
continue
;;
-release)
prev=release
continue
;;
-rpath)
prev=rpath
continue
;;
-R)
prev=xrpath
continue
;;
-R*)
func_stripname '-R' '' "$arg"
dir=$func_stripname_result
# We need an absolute path.
case $dir in
[\\/]* | [A-Za-z]:[\\/]*) ;;
=*)
func_stripname '=' '' "$dir"
dir=$lt_sysroot$func_stripname_result
;;
*)
func_fatal_error "only absolute run-paths are allowed"
;;
esac
case "$xrpath " in
*" $dir "*) ;;
*) func_append xrpath " $dir" ;;
esac
continue
;;
-shared)
# The effects of -shared are defined in a previous loop.
continue
;;
-shrext)
prev=shrext
continue
;;
-static | -static-libtool-libs)
# The effects of -static are defined in a previous loop.
# We used to do the same as -all-static on platforms that
# didn't have a PIC flag, but the assumption that the effects
# would be equivalent was wrong. It would break on at least
# Digital Unix and AIX.
continue
;;
-thread-safe)
thread_safe=yes
continue
;;
-version-info)
prev=vinfo
continue
;;
-version-number)
prev=vinfo
vinfo_number=yes
continue
;;
-weak)
prev=weak
continue
;;
-Wc,*)
func_stripname '-Wc,' '' "$arg"
args=$func_stripname_result
arg=
save_ifs="$IFS"; IFS=','
for flag in $args; do
IFS="$save_ifs"
func_quote_for_eval "$flag"
func_append arg " $func_quote_for_eval_result"
func_append compiler_flags " $func_quote_for_eval_result"
done
IFS="$save_ifs"
func_stripname ' ' '' "$arg"
arg=$func_stripname_result
;;
-Wl,*)
func_stripname '-Wl,' '' "$arg"
args=$func_stripname_result
arg=
save_ifs="$IFS"; IFS=','
for flag in $args; do
IFS="$save_ifs"
func_quote_for_eval "$flag"
func_append arg " $wl$func_quote_for_eval_result"
func_append compiler_flags " $wl$func_quote_for_eval_result"
func_append linker_flags " $func_quote_for_eval_result"
done
IFS="$save_ifs"
func_stripname ' ' '' "$arg"
arg=$func_stripname_result
;;
-Xcompiler)
prev=xcompiler
continue
;;
-Xlinker)
prev=xlinker
continue
;;
-XCClinker)
prev=xcclinker
continue
;;
# -msg_* for osf cc
-msg_*)
func_quote_for_eval "$arg"
arg="$func_quote_for_eval_result"
;;
# Flags to be passed through unchanged, with rationale:
# -64, -mips[0-9] enable 64-bit mode for the SGI compiler
# -r[0-9][0-9]* specify processor for the SGI compiler
# -xarch=*, -xtarget=* enable 64-bit mode for the Sun compiler
# +DA*, +DD* enable 64-bit mode for the HP compiler
# -q* compiler args for the IBM compiler
# -m*, -t[45]*, -txscale* architecture-specific flags for GCC
# -F/path path to uninstalled frameworks, gcc on darwin
# -p, -pg, --coverage, -fprofile-* profiling flags for GCC
# @file GCC response files
# -tp=* Portland pgcc target processor selection
# --sysroot=* for sysroot support
# -O*, -flto*, -fwhopr*, -fuse-linker-plugin GCC link-time optimization
-64|-mips[0-9]|-r[0-9][0-9]*|-xarch=*|-xtarget=*|+DA*|+DD*|-q*|-m*| \
-t[45]*|-txscale*|-p|-pg|--coverage|-fprofile-*|-F*|@*|-tp=*|--sysroot=*| \
-O*|-flto*|-fwhopr*|-fuse-linker-plugin)
func_quote_for_eval "$arg"
arg="$func_quote_for_eval_result"
func_append compile_command " $arg"
func_append finalize_command " $arg"
func_append compiler_flags " $arg"
continue
;;
# Some other compiler flag.
-* | +*)
func_quote_for_eval "$arg"
arg="$func_quote_for_eval_result"
;;
*.$objext)
# A standard object.
func_append objs " $arg"
;;
*.lo)
# A libtool-controlled object.
# Check to see that this really is a libtool object.
if func_lalib_unsafe_p "$arg"; then
pic_object=
non_pic_object=
# Read the .lo file
func_source "$arg"
if test -z "$pic_object" ||
test -z "$non_pic_object" ||
test "$pic_object" = none &&
test "$non_pic_object" = none; then
func_fatal_error "cannot find name of object for \`$arg'"
fi
# Extract subdirectory from the argument.
func_dirname "$arg" "/" ""
xdir="$func_dirname_result"
if test "$pic_object" != none; then
# Prepend the subdirectory the object is found in.
pic_object="$xdir$pic_object"
if test "$prev" = dlfiles; then
if test "$build_libtool_libs" = yes && test "$dlopen_support" = yes; then
func_append dlfiles " $pic_object"
prev=
continue
else
# If libtool objects are unsupported, then we need to preload.
prev=dlprefiles
fi
fi
# CHECK ME: I think I busted this. -Ossama
if test "$prev" = dlprefiles; then
# Preload the old-style object.
func_append dlprefiles " $pic_object"
prev=
fi
# A PIC object.
func_append libobjs " $pic_object"
arg="$pic_object"
fi
# Non-PIC object.
if test "$non_pic_object" != none; then
# Prepend the subdirectory the object is found in.
non_pic_object="$xdir$non_pic_object"
# A standard non-PIC object
func_append non_pic_objects " $non_pic_object"
if test -z "$pic_object" || test "$pic_object" = none ; then
arg="$non_pic_object"
fi
else
# If the PIC object exists, use it instead.
# $xdir was prepended to $pic_object above.
non_pic_object="$pic_object"
func_append non_pic_objects " $non_pic_object"
fi
else
# Only an error if not doing a dry-run.
if $opt_dry_run; then
# Extract subdirectory from the argument.
func_dirname "$arg" "/" ""
xdir="$func_dirname_result"
func_lo2o "$arg"
pic_object=$xdir$objdir/$func_lo2o_result
non_pic_object=$xdir$func_lo2o_result
func_append libobjs " $pic_object"
func_append non_pic_objects " $non_pic_object"
else
func_fatal_error "\`$arg' is not a valid libtool object"
fi
fi
;;
*.$libext)
# An archive.
func_append deplibs " $arg"
func_append old_deplibs " $arg"
continue
;;
*.la)
# A libtool-controlled library.
func_resolve_sysroot "$arg"
if test "$prev" = dlfiles; then
# This library was specified with -dlopen.
func_append dlfiles " $func_resolve_sysroot_result"
prev=
elif test "$prev" = dlprefiles; then
# The library was specified with -dlpreopen.
func_append dlprefiles " $func_resolve_sysroot_result"
prev=
else
func_append deplibs " $func_resolve_sysroot_result"
fi
continue
;;
# Some other compiler argument.
*)
# Unknown arguments in both finalize_command and compile_command need
# to be aesthetically quoted because they are evaled later.
func_quote_for_eval "$arg"
arg="$func_quote_for_eval_result"
;;
esac # arg
# Now actually substitute the argument into the commands.
if test -n "$arg"; then
func_append compile_command " $arg"
func_append finalize_command " $arg"
fi
done # argument parsing loop
test -n "$prev" && \
func_fatal_help "the \`$prevarg' option requires an argument"
if test "$export_dynamic" = yes && test -n "$export_dynamic_flag_spec"; then
eval arg=\"$export_dynamic_flag_spec\"
func_append compile_command " $arg"
func_append finalize_command " $arg"
fi
oldlibs=
# calculate the name of the file, without its directory
func_basename "$output"
outputname="$func_basename_result"
libobjs_save="$libobjs"
if test -n "$shlibpath_var"; then
# get the directories listed in $shlibpath_var
eval shlib_search_path=\`\$ECHO \"\${$shlibpath_var}\" \| \$SED \'s/:/ /g\'\`
else
shlib_search_path=
fi
eval sys_lib_search_path=\"$sys_lib_search_path_spec\"
eval sys_lib_dlsearch_path=\"$sys_lib_dlsearch_path_spec\"
func_dirname "$output" "/" ""
output_objdir="$func_dirname_result$objdir"
func_to_tool_file "$output_objdir/"
tool_output_objdir=$func_to_tool_file_result
# Create the object directory.
func_mkdir_p "$output_objdir"
# Determine the type of output
case $output in
"")
func_fatal_help "you must specify an output file"
;;
*.$libext) linkmode=oldlib ;;
*.lo | *.$objext) linkmode=obj ;;
*.la) linkmode=lib ;;
*) linkmode=prog ;; # Anything else should be a program.
esac
specialdeplibs=
libs=
# Find all interdependent deplibs by searching for libraries
# that are linked more than once (e.g. -la -lb -la)
for deplib in $deplibs; do
if $opt_preserve_dup_deps ; then
case "$libs " in
*" $deplib "*) func_append specialdeplibs " $deplib" ;;
esac
fi
func_append libs " $deplib"
done
if test "$linkmode" = lib; then
libs="$predeps $libs $compiler_lib_search_path $postdeps"
# Compute libraries that are listed more than once in $predeps
# $postdeps and mark them as special (i.e., whose duplicates are
# not to be eliminated).
pre_post_deps=
if $opt_duplicate_compiler_generated_deps; then
for pre_post_dep in $predeps $postdeps; do
case "$pre_post_deps " in
*" $pre_post_dep "*) func_append specialdeplibs " $pre_post_deps" ;;
esac
func_append pre_post_deps " $pre_post_dep"
done
fi
pre_post_deps=
fi
deplibs=
newdependency_libs=
newlib_search_path=
need_relink=no # whether we're linking any uninstalled libtool libraries
notinst_deplibs= # not-installed libtool libraries
notinst_path= # paths that contain not-installed libtool libraries
case $linkmode in
lib)
passes="conv dlpreopen link"
for file in $dlfiles $dlprefiles; do
case $file in
*.la) ;;
*)
func_fatal_help "libraries can \`-dlopen' only libtool libraries: $file"
;;
esac
done
;;
prog)
compile_deplibs=
finalize_deplibs=
alldeplibs=no
newdlfiles=
newdlprefiles=
passes="conv scan dlopen dlpreopen link"
;;
*) passes="conv"
;;
esac
for pass in $passes; do
# The preopen pass in lib mode reverses $deplibs; put it back here
# so that -L comes before libs that need it for instance...
if test "$linkmode,$pass" = "lib,link"; then
## FIXME: Find the place where the list is rebuilt in the wrong
## order, and fix it there properly
tmp_deplibs=
for deplib in $deplibs; do
tmp_deplibs="$deplib $tmp_deplibs"
done
deplibs="$tmp_deplibs"
fi
if test "$linkmode,$pass" = "lib,link" ||
test "$linkmode,$pass" = "prog,scan"; then
libs="$deplibs"
deplibs=
fi
if test "$linkmode" = prog; then
case $pass in
dlopen) libs="$dlfiles" ;;
dlpreopen) libs="$dlprefiles" ;;
link)
libs="$deplibs %DEPLIBS%"
test "X$link_all_deplibs" != Xno && libs="$libs $dependency_libs"
;;
esac
fi
if test "$linkmode,$pass" = "lib,dlpreopen"; then
# Collect and forward deplibs of preopened libtool libs
for lib in $dlprefiles; do
# Ignore non-libtool-libs
dependency_libs=
func_resolve_sysroot "$lib"
case $lib in
*.la) func_source "$func_resolve_sysroot_result" ;;
esac
# Collect preopened libtool deplibs, except any this library
# has declared as weak libs
for deplib in $dependency_libs; do
func_basename "$deplib"
deplib_base=$func_basename_result
case " $weak_libs " in
*" $deplib_base "*) ;;
*) func_append deplibs " $deplib" ;;
esac
done
done
libs="$dlprefiles"
fi
if test "$pass" = dlopen; then
# Collect dlpreopened libraries
save_deplibs="$deplibs"
deplibs=
fi
for deplib in $libs; do
lib=
found=no
case $deplib in
-mt|-mthreads|-kthread|-Kthread|-pthread|-pthreads|--thread-safe \
|-threads|-fopenmp|-openmp|-mp|-xopenmp|-omp|-qsmp=*)
if test "$linkmode,$pass" = "prog,link"; then
compile_deplibs="$deplib $compile_deplibs"
finalize_deplibs="$deplib $finalize_deplibs"
else
func_append compiler_flags " $deplib"
if test "$linkmode" = lib ; then
case "$new_inherited_linker_flags " in
*" $deplib "*) ;;
* ) func_append new_inherited_linker_flags " $deplib" ;;
esac
fi
fi
continue
;;
-l*)
if test "$linkmode" != lib && test "$linkmode" != prog; then
func_warning "\`-l' is ignored for archives/objects"
continue
fi
func_stripname '-l' '' "$deplib"
name=$func_stripname_result
if test "$linkmode" = lib; then
searchdirs="$newlib_search_path $lib_search_path $compiler_lib_search_dirs $sys_lib_search_path $shlib_search_path"
else
searchdirs="$newlib_search_path $lib_search_path $sys_lib_search_path $shlib_search_path"
fi
for searchdir in $searchdirs; do
for search_ext in .la $std_shrext .so .a; do
# Search the libtool library
lib="$searchdir/lib${name}${search_ext}"
if test -f "$lib"; then
if test "$search_ext" = ".la"; then
found=yes
else
found=no
fi
break 2
fi
done
done
if test "$found" != yes; then
# deplib doesn't seem to be a libtool library
if test "$linkmode,$pass" = "prog,link"; then
compile_deplibs="$deplib $compile_deplibs"
finalize_deplibs="$deplib $finalize_deplibs"
else
deplibs="$deplib $deplibs"
test "$linkmode" = lib && newdependency_libs="$deplib $newdependency_libs"
fi
continue
else # deplib is a libtool library
# If $allow_libtool_libs_with_static_runtimes && $deplib is a stdlib,
# We need to do some special things here, and not later.
if test "X$allow_libtool_libs_with_static_runtimes" = "Xyes" ; then
case " $predeps $postdeps " in
*" $deplib "*)
if func_lalib_p "$lib"; then
library_names=
old_library=
func_source "$lib"
for l in $old_library $library_names; do
ll="$l"
done
if test "X$ll" = "X$old_library" ; then # only static version available
found=no
func_dirname "$lib" "" "."
ladir="$func_dirname_result"
lib=$ladir/$old_library
if test "$linkmode,$pass" = "prog,link"; then
compile_deplibs="$deplib $compile_deplibs"
finalize_deplibs="$deplib $finalize_deplibs"
else
deplibs="$deplib $deplibs"
test "$linkmode" = lib && newdependency_libs="$deplib $newdependency_libs"
fi
continue
fi
fi
;;
*) ;;
esac
fi
fi
;; # -l
*.ltframework)
if test "$linkmode,$pass" = "prog,link"; then
compile_deplibs="$deplib $compile_deplibs"
finalize_deplibs="$deplib $finalize_deplibs"
else
deplibs="$deplib $deplibs"
if test "$linkmode" = lib ; then
case "$new_inherited_linker_flags " in
*" $deplib "*) ;;
* ) func_append new_inherited_linker_flags " $deplib" ;;
esac
fi
fi
continue
;;
-L*)
case $linkmode in
lib)
deplibs="$deplib $deplibs"
test "$pass" = conv && continue
newdependency_libs="$deplib $newdependency_libs"
func_stripname '-L' '' "$deplib"
func_resolve_sysroot "$func_stripname_result"
func_append newlib_search_path " $func_resolve_sysroot_result"
;;
prog)
if test "$pass" = conv; then
deplibs="$deplib $deplibs"
continue
fi
if test "$pass" = scan; then
deplibs="$deplib $deplibs"
else
compile_deplibs="$deplib $compile_deplibs"
finalize_deplibs="$deplib $finalize_deplibs"
fi
func_stripname '-L' '' "$deplib"
func_resolve_sysroot "$func_stripname_result"
func_append newlib_search_path " $func_resolve_sysroot_result"
;;
*)
func_warning "\`-L' is ignored for archives/objects"
;;
esac # linkmode
continue
;; # -L
-R*)
if test "$pass" = link; then
func_stripname '-R' '' "$deplib"
func_resolve_sysroot "$func_stripname_result"
dir=$func_resolve_sysroot_result
# Make sure the xrpath contains only unique directories.
case "$xrpath " in
*" $dir "*) ;;
*) func_append xrpath " $dir" ;;
esac
fi
deplibs="$deplib $deplibs"
continue
;;
*.la)
func_resolve_sysroot "$deplib"
lib=$func_resolve_sysroot_result
;;
*.$libext)
if test "$pass" = conv; then
deplibs="$deplib $deplibs"
continue
fi
case $linkmode in
lib)
# Linking convenience modules into shared libraries is allowed,
# but linking other static libraries is non-portable.
case " $dlpreconveniencelibs " in
*" $deplib "*) ;;
*)
valid_a_lib=no
case $deplibs_check_method in
match_pattern*)
set dummy $deplibs_check_method; shift
match_pattern_regex=`expr "$deplibs_check_method" : "$1 \(.*\)"`
if eval "\$ECHO \"$deplib\"" 2>/dev/null | $SED 10q \
| $EGREP "$match_pattern_regex" > /dev/null; then
valid_a_lib=yes
fi
;;
pass_all)
valid_a_lib=yes
;;
esac
if test "$valid_a_lib" != yes; then
echo
$ECHO "*** Warning: Trying to link with static lib archive $deplib."
echo "*** I have the capability to make that library automatically link in when"
echo "*** you link to this library. But I can only do this if you have a"
echo "*** shared version of the library, which you do not appear to have"
echo "*** because the file extensions .$libext of this argument makes me believe"
echo "*** that it is just a static archive that I should not use here."
else
echo
$ECHO "*** Warning: Linking the shared library $output against the"
$ECHO "*** static library $deplib is not portable!"
deplibs="$deplib $deplibs"
fi
;;
esac
continue
;;
prog)
if test "$pass" != link; then
deplibs="$deplib $deplibs"
else
compile_deplibs="$deplib $compile_deplibs"
finalize_deplibs="$deplib $finalize_deplibs"
fi
continue
;;
esac # linkmode
;; # *.$libext
*.lo | *.$objext)
if test "$pass" = conv; then
deplibs="$deplib $deplibs"
elif test "$linkmode" = prog; then
if test "$pass" = dlpreopen || test "$dlopen_support" != yes || test "$build_libtool_libs" = no; then
# If there is no dlopen support or we're linking statically,
# we need to preload.
func_append newdlprefiles " $deplib"
compile_deplibs="$deplib $compile_deplibs"
finalize_deplibs="$deplib $finalize_deplibs"
else
func_append newdlfiles " $deplib"
fi
fi
continue
;;
%DEPLIBS%)
alldeplibs=yes
continue
;;
esac # case $deplib
if test "$found" = yes || test -f "$lib"; then :
else
func_fatal_error "cannot find the library \`$lib' or unhandled argument \`$deplib'"
fi
# Check to see that this really is a libtool archive.
func_lalib_unsafe_p "$lib" \
|| func_fatal_error "\`$lib' is not a valid libtool archive"
func_dirname "$lib" "" "."
ladir="$func_dirname_result"
dlname=
dlopen=
dlpreopen=
libdir=
library_names=
old_library=
inherited_linker_flags=
# If the library was installed with an old release of libtool,
# it will not redefine variables installed, or shouldnotlink
installed=yes
shouldnotlink=no
avoidtemprpath=
# Read the .la file
func_source "$lib"
# Convert "-framework foo" to "foo.ltframework"
if test -n "$inherited_linker_flags"; then
tmp_inherited_linker_flags=`$ECHO "$inherited_linker_flags" | $SED 's/-framework \([^ $]*\)/\1.ltframework/g'`
for tmp_inherited_linker_flag in $tmp_inherited_linker_flags; do
case " $new_inherited_linker_flags " in
*" $tmp_inherited_linker_flag "*) ;;
*) func_append new_inherited_linker_flags " $tmp_inherited_linker_flag";;
esac
done
fi
dependency_libs=`$ECHO " $dependency_libs" | $SED 's% \([^ $]*\).ltframework% -framework \1%g'`
if test "$linkmode,$pass" = "lib,link" ||
test "$linkmode,$pass" = "prog,scan" ||
{ test "$linkmode" != prog && test "$linkmode" != lib; }; then
test -n "$dlopen" && func_append dlfiles " $dlopen"
test -n "$dlpreopen" && func_append dlprefiles " $dlpreopen"
fi
if test "$pass" = conv; then
# Only check for convenience libraries
deplibs="$lib $deplibs"
if test -z "$libdir"; then
if test -z "$old_library"; then
func_fatal_error "cannot find name of link library for \`$lib'"
fi
# It is a libtool convenience library, so add in its objects.
func_append convenience " $ladir/$objdir/$old_library"
func_append old_convenience " $ladir/$objdir/$old_library"
tmp_libs=
for deplib in $dependency_libs; do
deplibs="$deplib $deplibs"
if $opt_preserve_dup_deps ; then
case "$tmp_libs " in
*" $deplib "*) func_append specialdeplibs " $deplib" ;;
esac
fi
func_append tmp_libs " $deplib"
done
elif test "$linkmode" != prog && test "$linkmode" != lib; then
func_fatal_error "\`$lib' is not a convenience library"
fi
continue
fi # $pass = conv
# Get the name of the library we link against.
linklib=
if test -n "$old_library" &&
{ test "$prefer_static_libs" = yes ||
test "$prefer_static_libs,$installed" = "built,no"; }; then
linklib=$old_library
else
for l in $old_library $library_names; do
linklib="$l"
done
fi
if test -z "$linklib"; then
func_fatal_error "cannot find name of link library for \`$lib'"
fi
# This library was specified with -dlopen.
if test "$pass" = dlopen; then
if test -z "$libdir"; then
func_fatal_error "cannot -dlopen a convenience library: \`$lib'"
fi
if test -z "$dlname" ||
test "$dlopen_support" != yes ||
test "$build_libtool_libs" = no; then
# If there is no dlname, no dlopen support or we're linking
# statically, we need to preload. We also need to preload any
# dependent libraries so libltdl's deplib preloader doesn't
# bomb out in the load deplibs phase.
func_append dlprefiles " $lib $dependency_libs"
else
func_append newdlfiles " $lib"
fi
continue
fi # $pass = dlopen
# We need an absolute path.
case $ladir in
[\\/]* | [A-Za-z]:[\\/]*) abs_ladir="$ladir" ;;
*)
abs_ladir=`cd "$ladir" && pwd`
if test -z "$abs_ladir"; then
func_warning "cannot determine absolute directory name of \`$ladir'"
func_warning "passing it literally to the linker, although it might fail"
abs_ladir="$ladir"
fi
;;
esac
func_basename "$lib"
laname="$func_basename_result"
# Find the relevant object directory and library name.
if test "X$installed" = Xyes; then
if test ! -f "$lt_sysroot$libdir/$linklib" && test -f "$abs_ladir/$linklib"; then
func_warning "library \`$lib' was moved."
dir="$ladir"
absdir="$abs_ladir"
libdir="$abs_ladir"
else
dir="$lt_sysroot$libdir"
absdir="$lt_sysroot$libdir"
fi
test "X$hardcode_automatic" = Xyes && avoidtemprpath=yes
else
if test ! -f "$ladir/$objdir/$linklib" && test -f "$abs_ladir/$linklib"; then
dir="$ladir"
absdir="$abs_ladir"
# Remove this search path later
func_append notinst_path " $abs_ladir"
else
dir="$ladir/$objdir"
absdir="$abs_ladir/$objdir"
# Remove this search path later
func_append notinst_path " $abs_ladir"
fi
fi # $installed = yes
func_stripname 'lib' '.la' "$laname"
name=$func_stripname_result
# This library was specified with -dlpreopen.
if test "$pass" = dlpreopen; then
if test -z "$libdir" && test "$linkmode" = prog; then
func_fatal_error "only libraries may -dlpreopen a convenience library: \`$lib'"
fi
case "$host" in
# special handling for platforms with PE-DLLs.
*cygwin* | *mingw* | *cegcc* )
# Linker will automatically link against shared library if both
# static and shared are present. Therefore, ensure we extract
# symbols from the import library if a shared library is present
# (otherwise, the dlopen module name will be incorrect). We do
# this by putting the import library name into $newdlprefiles.
# We recover the dlopen module name by 'saving' the la file
# name in a special purpose variable, and (later) extracting the
# dlname from the la file.
if test -n "$dlname"; then
func_tr_sh "$dir/$linklib"
eval "libfile_$func_tr_sh_result=\$abs_ladir/\$laname"
func_append newdlprefiles " $dir/$linklib"
else
func_append newdlprefiles " $dir/$old_library"
# Keep a list of preopened convenience libraries to check
# that they are being used correctly in the link pass.
test -z "$libdir" && \
func_append dlpreconveniencelibs " $dir/$old_library"
fi
;;
* )
# Prefer using a static library (so that no silly _DYNAMIC symbols
# are required to link).
if test -n "$old_library"; then
func_append newdlprefiles " $dir/$old_library"
# Keep a list of preopened convenience libraries to check
# that they are being used correctly in the link pass.
test -z "$libdir" && \
func_append dlpreconveniencelibs " $dir/$old_library"
# Otherwise, use the dlname, so that lt_dlopen finds it.
elif test -n "$dlname"; then
func_append newdlprefiles " $dir/$dlname"
else
func_append newdlprefiles " $dir/$linklib"
fi
;;
esac
fi # $pass = dlpreopen
if test -z "$libdir"; then
# Link the convenience library
if test "$linkmode" = lib; then
deplibs="$dir/$old_library $deplibs"
elif test "$linkmode,$pass" = "prog,link"; then
compile_deplibs="$dir/$old_library $compile_deplibs"
finalize_deplibs="$dir/$old_library $finalize_deplibs"
else
deplibs="$lib $deplibs" # used for prog,scan pass
fi
continue
fi
if test "$linkmode" = prog && test "$pass" != link; then
func_append newlib_search_path " $ladir"
deplibs="$lib $deplibs"
linkalldeplibs=no
if test "$link_all_deplibs" != no || test -z "$library_names" ||
test "$build_libtool_libs" = no; then
linkalldeplibs=yes
fi
tmp_libs=
for deplib in $dependency_libs; do
case $deplib in
-L*) func_stripname '-L' '' "$deplib"
func_resolve_sysroot "$func_stripname_result"
func_append newlib_search_path " $func_resolve_sysroot_result"
;;
esac
# Need to link against all dependency_libs?
if test "$linkalldeplibs" = yes; then
deplibs="$deplib $deplibs"
else
# Need to hardcode shared library paths
# or/and link against static libraries
newdependency_libs="$deplib $newdependency_libs"
fi
if $opt_preserve_dup_deps ; then
case "$tmp_libs " in
*" $deplib "*) func_append specialdeplibs " $deplib" ;;
esac
fi
func_append tmp_libs " $deplib"
done # for deplib
continue
fi # $linkmode = prog...
if test "$linkmode,$pass" = "prog,link"; then
if test -n "$library_names" &&
{ { test "$prefer_static_libs" = no ||
test "$prefer_static_libs,$installed" = "built,yes"; } ||
test -z "$old_library"; }; then
# We need to hardcode the library path
if test -n "$shlibpath_var" && test -z "$avoidtemprpath" ; then
# Make sure the rpath contains only unique directories.
case "$temp_rpath:" in
*"$absdir:"*) ;;
*) func_append temp_rpath "$absdir:" ;;
esac
fi
# Hardcode the library path.
# Skip directories that are in the system default run-time
# search path.
case " $sys_lib_dlsearch_path " in
*" $absdir "*) ;;
*)
case "$compile_rpath " in
*" $absdir "*) ;;
*) func_append compile_rpath " $absdir" ;;
esac
;;
esac
case " $sys_lib_dlsearch_path " in
*" $libdir "*) ;;
*)
case "$finalize_rpath " in
*" $libdir "*) ;;
*) func_append finalize_rpath " $libdir" ;;
esac
;;
esac
fi # $linkmode,$pass = prog,link...
if test "$alldeplibs" = yes &&
{ test "$deplibs_check_method" = pass_all ||
{ test "$build_libtool_libs" = yes &&
test -n "$library_names"; }; }; then
# We only need to search for static libraries
continue
fi
fi
link_static=no # Whether the deplib will be linked statically
use_static_libs=$prefer_static_libs
if test "$use_static_libs" = built && test "$installed" = yes; then
use_static_libs=no
fi
if test -n "$library_names" &&
{ test "$use_static_libs" = no || test -z "$old_library"; }; then
case $host in
*cygwin* | *mingw* | *cegcc*)
# No point in relinking DLLs because paths are not encoded
func_append notinst_deplibs " $lib"
need_relink=no
;;
*)
if test "$installed" = no; then
func_append notinst_deplibs " $lib"
need_relink=yes
fi
;;
esac
# This is a shared library
# Warn about portability, can't link against -module's on some
# systems (darwin). Don't bleat about dlopened modules though!
dlopenmodule=""
for dlpremoduletest in $dlprefiles; do
if test "X$dlpremoduletest" = "X$lib"; then
dlopenmodule="$dlpremoduletest"
break
fi
done
if test -z "$dlopenmodule" && test "$shouldnotlink" = yes && test "$pass" = link; then
echo
if test "$linkmode" = prog; then
$ECHO "*** Warning: Linking the executable $output against the loadable module"
else
$ECHO "*** Warning: Linking the shared library $output against the loadable module"
fi
$ECHO "*** $linklib is not portable!"
fi
if test "$linkmode" = lib &&
test "$hardcode_into_libs" = yes; then
# Hardcode the library path.
# Skip directories that are in the system default run-time
# search path.
case " $sys_lib_dlsearch_path " in
*" $absdir "*) ;;
*)
case "$compile_rpath " in
*" $absdir "*) ;;
*) func_append compile_rpath " $absdir" ;;
esac
;;
esac
case " $sys_lib_dlsearch_path " in
*" $libdir "*) ;;
*)
case "$finalize_rpath " in
*" $libdir "*) ;;
*) func_append finalize_rpath " $libdir" ;;
esac
;;
esac
fi
if test -n "$old_archive_from_expsyms_cmds"; then
# figure out the soname
set dummy $library_names
shift
realname="$1"
shift
libname=`eval "\\$ECHO \"$libname_spec\""`
# use dlname if we got it. it's perfectly good, no?
if test -n "$dlname"; then
soname="$dlname"
elif test -n "$soname_spec"; then
# bleh windows
case $host in
x86_64-*mingw32*)
func_arith $current - $age
major=$func_arith_result
versuffix="6-$major"
;;
*cygwin* | mingw* | *cegcc*)
func_arith $current - $age
major=$func_arith_result
versuffix="-$major"
;;
esac
eval soname=\"$soname_spec\"
else
soname="$realname"
fi
# Make a new name for the extract_expsyms_cmds to use
soroot="$soname"
func_basename "$soroot"
soname="$func_basename_result"
func_stripname 'lib' '.dll' "$soname"
newlib=libimp-$func_stripname_result.a
# If the library has no export list, then create one now
if test -f "$output_objdir/$soname-def"; then :
else
func_verbose "extracting exported symbol list from \`$soname'"
func_execute_cmds "$extract_expsyms_cmds" 'exit $?'
fi
# Create $newlib
if test -f "$output_objdir/$newlib"; then :; else
func_verbose "generating import library for \`$soname'"
func_execute_cmds "$old_archive_from_expsyms_cmds" 'exit $?'
fi
# make sure the library variables are pointing to the new library
dir=$output_objdir
linklib=$newlib
fi # test -n "$old_archive_from_expsyms_cmds"
if test "$linkmode" = prog || test "$opt_mode" != relink; then
add_shlibpath=
add_dir=
add=
lib_linked=yes
case $hardcode_action in
immediate | unsupported)
if test "$hardcode_direct" = no; then
add="$dir/$linklib"
case $host in
*-*-sco3.2v5.0.[024]*) add_dir="-L$dir" ;;
*-*-sysv4*uw2*) add_dir="-L$dir" ;;
*-*-sysv5OpenUNIX* | *-*-sysv5UnixWare7.[01].[10]* | \
*-*-unixware7*) add_dir="-L$dir" ;;
*-*-darwin* )
# if the lib is a (non-dlopened) module then we can not
# link against it, someone is ignoring the earlier warnings
if /usr/bin/file -L $add 2> /dev/null |
$GREP ": [^:]* bundle" >/dev/null ; then
if test "X$dlopenmodule" != "X$lib"; then
$ECHO "*** Warning: lib $linklib is a module, not a shared library"
if test -z "$old_library" ; then
echo
echo "*** And there doesn't seem to be a static archive available"
echo "*** The link will probably fail, sorry"
else
add="$dir/$old_library"
fi
elif test -n "$old_library"; then
add="$dir/$old_library"
fi
fi
esac
elif test "$hardcode_minus_L" = no; then
case $host in
*-*-sunos*) add_shlibpath="$dir" ;;
esac
add_dir="-L$dir"
add="-l$name"
elif test "$hardcode_shlibpath_var" = no; then
add_shlibpath="$dir"
add="-l$name"
else
lib_linked=no
fi
;;
relink)
if test "$hardcode_direct" = yes &&
test "$hardcode_direct_absolute" = no; then
add="$dir/$linklib"
elif test "$hardcode_minus_L" = yes; then
add_dir="-L$absdir"
# Try looking first in the location we're being installed to.
if test -n "$inst_prefix_dir"; then
case $libdir in
[\\/]*)
func_append add_dir " -L$inst_prefix_dir$libdir"
;;
esac
fi
add="-l$name"
elif test "$hardcode_shlibpath_var" = yes; then
add_shlibpath="$dir"
add="-l$name"
else
lib_linked=no
fi
;;
*) lib_linked=no ;;
esac
if test "$lib_linked" != yes; then
func_fatal_configuration "unsupported hardcode properties"
fi
if test -n "$add_shlibpath"; then
case :$compile_shlibpath: in
*":$add_shlibpath:"*) ;;
*) func_append compile_shlibpath "$add_shlibpath:" ;;
esac
fi
if test "$linkmode" = prog; then
test -n "$add_dir" && compile_deplibs="$add_dir $compile_deplibs"
test -n "$add" && compile_deplibs="$add $compile_deplibs"
else
test -n "$add_dir" && deplibs="$add_dir $deplibs"
test -n "$add" && deplibs="$add $deplibs"
if test "$hardcode_direct" != yes &&
test "$hardcode_minus_L" != yes &&
test "$hardcode_shlibpath_var" = yes; then
case :$finalize_shlibpath: in
*":$libdir:"*) ;;
*) func_append finalize_shlibpath "$libdir:" ;;
esac
fi
fi
fi
if test "$linkmode" = prog || test "$opt_mode" = relink; then
add_shlibpath=
add_dir=
add=
# Finalize command for both is simple: just hardcode it.
if test "$hardcode_direct" = yes &&
test "$hardcode_direct_absolute" = no; then
add="$libdir/$linklib"
elif test "$hardcode_minus_L" = yes; then
add_dir="-L$libdir"
add="-l$name"
elif test "$hardcode_shlibpath_var" = yes; then
case :$finalize_shlibpath: in
*":$libdir:"*) ;;
*) func_append finalize_shlibpath "$libdir:" ;;
esac
add="-l$name"
elif test "$hardcode_automatic" = yes; then
if test -n "$inst_prefix_dir" &&
test -f "$inst_prefix_dir$libdir/$linklib" ; then
add="$inst_prefix_dir$libdir/$linklib"
else
add="$libdir/$linklib"
fi
else
# We cannot seem to hardcode it, guess we'll fake it.
add_dir="-L$libdir"
# Try looking first in the location we're being installed to.
if test -n "$inst_prefix_dir"; then
case $libdir in
[\\/]*)
func_append add_dir " -L$inst_prefix_dir$libdir"
;;
esac
fi
add="-l$name"
fi
if test "$linkmode" = prog; then
test -n "$add_dir" && finalize_deplibs="$add_dir $finalize_deplibs"
test -n "$add" && finalize_deplibs="$add $finalize_deplibs"
else
test -n "$add_dir" && deplibs="$add_dir $deplibs"
test -n "$add" && deplibs="$add $deplibs"
fi
fi
elif test "$linkmode" = prog; then
# Here we assume that one of hardcode_direct or hardcode_minus_L
# is not unsupported. This is valid on all known static and
# shared platforms.
if test "$hardcode_direct" != unsupported; then
test -n "$old_library" && linklib="$old_library"
compile_deplibs="$dir/$linklib $compile_deplibs"
finalize_deplibs="$dir/$linklib $finalize_deplibs"
else
compile_deplibs="-l$name -L$dir $compile_deplibs"
finalize_deplibs="-l$name -L$dir $finalize_deplibs"
fi
elif test "$build_libtool_libs" = yes; then
# Not a shared library
if test "$deplibs_check_method" != pass_all; then
# We're trying link a shared library against a static one
# but the system doesn't support it.
# Just print a warning and add the library to dependency_libs so
# that the program can be linked against the static library.
echo
$ECHO "*** Warning: This system can not link to static lib archive $lib."
echo "*** I have the capability to make that library automatically link in when"
echo "*** you link to this library. But I can only do this if you have a"
echo "*** shared version of the library, which you do not appear to have."
if test "$module" = yes; then
echo "*** But as you try to build a module library, libtool will still create "
echo "*** a static module, that should work as long as the dlopening application"
echo "*** is linked with the -dlopen flag to resolve symbols at runtime."
if test -z "$global_symbol_pipe"; then
echo
echo "*** However, this would only work if libtool was able to extract symbol"
echo "*** lists from a program, using \`nm' or equivalent, but libtool could"
echo "*** not find such a program. So, this module is probably useless."
echo "*** \`nm' from GNU binutils and a full rebuild may help."
fi
if test "$build_old_libs" = no; then
build_libtool_libs=module
build_old_libs=yes
else
build_libtool_libs=no
fi
fi
else
deplibs="$dir/$old_library $deplibs"
link_static=yes
fi
fi # link shared/static library?
if test "$linkmode" = lib; then
if test -n "$dependency_libs" &&
{ test "$hardcode_into_libs" != yes ||
test "$build_old_libs" = yes ||
test "$link_static" = yes; }; then
# Extract -R from dependency_libs
temp_deplibs=
for libdir in $dependency_libs; do
case $libdir in
-R*) func_stripname '-R' '' "$libdir"
temp_xrpath=$func_stripname_result
case " $xrpath " in
*" $temp_xrpath "*) ;;
*) func_append xrpath " $temp_xrpath";;
esac;;
*) func_append temp_deplibs " $libdir";;
esac
done
dependency_libs="$temp_deplibs"
fi
func_append newlib_search_path " $absdir"
# Link against this library
test "$link_static" = no && newdependency_libs="$abs_ladir/$laname $newdependency_libs"
# ... and its dependency_libs
tmp_libs=
for deplib in $dependency_libs; do
newdependency_libs="$deplib $newdependency_libs"
case $deplib in
-L*) func_stripname '-L' '' "$deplib"
func_resolve_sysroot "$func_stripname_result";;
*) func_resolve_sysroot "$deplib" ;;
esac
if $opt_preserve_dup_deps ; then
case "$tmp_libs " in
*" $func_resolve_sysroot_result "*)
func_append specialdeplibs " $func_resolve_sysroot_result" ;;
esac
fi
func_append tmp_libs " $func_resolve_sysroot_result"
done
if test "$link_all_deplibs" != no; then
# Add the search paths of all dependency libraries
for deplib in $dependency_libs; do
path=
case $deplib in
-L*) path="$deplib" ;;
*.la)
func_resolve_sysroot "$deplib"
deplib=$func_resolve_sysroot_result
func_dirname "$deplib" "" "."
dir=$func_dirname_result
# We need an absolute path.
case $dir in
[\\/]* | [A-Za-z]:[\\/]*) absdir="$dir" ;;
*)
absdir=`cd "$dir" && pwd`
if test -z "$absdir"; then
func_warning "cannot determine absolute directory name of \`$dir'"
absdir="$dir"
fi
;;
esac
if $GREP "^installed=no" $deplib > /dev/null; then
case $host in
*-*-darwin*)
depdepl=
eval deplibrary_names=`${SED} -n -e 's/^library_names=\(.*\)$/\1/p' $deplib`
if test -n "$deplibrary_names" ; then
for tmp in $deplibrary_names ; do
depdepl=$tmp
done
if test -f "$absdir/$objdir/$depdepl" ; then
depdepl="$absdir/$objdir/$depdepl"
darwin_install_name=`${OTOOL} -L $depdepl | awk '{if (NR == 2) {print $1;exit}}'`
if test -z "$darwin_install_name"; then
darwin_install_name=`${OTOOL64} -L $depdepl | awk '{if (NR == 2) {print $1;exit}}'`
fi
func_append compiler_flags " ${wl}-dylib_file ${wl}${darwin_install_name}:${depdepl}"
func_append linker_flags " -dylib_file ${darwin_install_name}:${depdepl}"
path=
fi
fi
;;
*)
path="-L$absdir/$objdir"
;;
esac
else
eval libdir=`${SED} -n -e 's/^libdir=\(.*\)$/\1/p' $deplib`
test -z "$libdir" && \
func_fatal_error "\`$deplib' is not a valid libtool archive"
test "$absdir" != "$libdir" && \
func_warning "\`$deplib' seems to be moved"
path="-L$absdir"
fi
;;
esac
case " $deplibs " in
*" $path "*) ;;
*) deplibs="$path $deplibs" ;;
esac
done
fi # link_all_deplibs != no
fi # linkmode = lib
done # for deplib in $libs
if test "$pass" = link; then
if test "$linkmode" = "prog"; then
compile_deplibs="$new_inherited_linker_flags $compile_deplibs"
finalize_deplibs="$new_inherited_linker_flags $finalize_deplibs"
else
compiler_flags="$compiler_flags "`$ECHO " $new_inherited_linker_flags" | $SED 's% \([^ $]*\).ltframework% -framework \1%g'`
fi
fi
dependency_libs="$newdependency_libs"
if test "$pass" = dlpreopen; then
# Link the dlpreopened libraries before other libraries
for deplib in $save_deplibs; do
deplibs="$deplib $deplibs"
done
fi
if test "$pass" != dlopen; then
if test "$pass" != conv; then
# Make sure lib_search_path contains only unique directories.
lib_search_path=
for dir in $newlib_search_path; do
case "$lib_search_path " in
*" $dir "*) ;;
*) func_append lib_search_path " $dir" ;;
esac
done
newlib_search_path=
fi
if test "$linkmode,$pass" != "prog,link"; then
vars="deplibs"
else
vars="compile_deplibs finalize_deplibs"
fi
for var in $vars dependency_libs; do
# Add libraries to $var in reverse order
eval tmp_libs=\"\$$var\"
new_libs=
for deplib in $tmp_libs; do
# FIXME: Pedantically, this is the right thing to do, so
# that some nasty dependency loop isn't accidentally
# broken:
#new_libs="$deplib $new_libs"
# Pragmatically, this seems to cause very few problems in
# practice:
case $deplib in
-L*) new_libs="$deplib $new_libs" ;;
-R*) ;;
*)
# And here is the reason: when a library appears more
# than once as an explicit dependence of a library, or
# is implicitly linked in more than once by the
# compiler, it is considered special, and multiple
# occurrences thereof are not removed. Compare this
# with having the same library being listed as a
# dependency of multiple other libraries: in this case,
# we know (pedantically, we assume) the library does not
# need to be listed more than once, so we keep only the
# last copy. This is not always right, but it is rare
# enough that we require users that really mean to play
# such unportable linking tricks to link the library
# using -Wl,-lname, so that libtool does not consider it
# for duplicate removal.
case " $specialdeplibs " in
*" $deplib "*) new_libs="$deplib $new_libs" ;;
*)
case " $new_libs " in
*" $deplib "*) ;;
*) new_libs="$deplib $new_libs" ;;
esac
;;
esac
;;
esac
done
tmp_libs=
for deplib in $new_libs; do
case $deplib in
-L*)
case " $tmp_libs " in
*" $deplib "*) ;;
*) func_append tmp_libs " $deplib" ;;
esac
;;
*) func_append tmp_libs " $deplib" ;;
esac
done
eval $var=\"$tmp_libs\"
done # for var
fi
# Last step: remove runtime libs from dependency_libs
# (they stay in deplibs)
tmp_libs=
for i in $dependency_libs ; do
case " $predeps $postdeps $compiler_lib_search_path " in
*" $i "*)
i=""
;;
esac
if test -n "$i" ; then
func_append tmp_libs " $i"
fi
done
dependency_libs=$tmp_libs
done # for pass
if test "$linkmode" = prog; then
dlfiles="$newdlfiles"
fi
if test "$linkmode" = prog || test "$linkmode" = lib; then
dlprefiles="$newdlprefiles"
fi
case $linkmode in
oldlib)
if test -n "$dlfiles$dlprefiles" || test "$dlself" != no; then
func_warning "\`-dlopen' is ignored for archives"
fi
case " $deplibs" in
*\ -l* | *\ -L*)
func_warning "\`-l' and \`-L' are ignored for archives" ;;
esac
test -n "$rpath" && \
func_warning "\`-rpath' is ignored for archives"
test -n "$xrpath" && \
func_warning "\`-R' is ignored for archives"
test -n "$vinfo" && \
func_warning "\`-version-info/-version-number' is ignored for archives"
test -n "$release" && \
func_warning "\`-release' is ignored for archives"
test -n "$export_symbols$export_symbols_regex" && \
func_warning "\`-export-symbols' is ignored for archives"
# Now set the variables for building old libraries.
build_libtool_libs=no
oldlibs="$output"
func_append objs "$old_deplibs"
;;
lib)
# Make sure we only generate libraries of the form `libNAME.la'.
case $outputname in
lib*)
func_stripname 'lib' '.la' "$outputname"
name=$func_stripname_result
eval shared_ext=\"$shrext_cmds\"
eval libname=\"$libname_spec\"
;;
*)
test "$module" = no && \
func_fatal_help "libtool library \`$output' must begin with \`lib'"
if test "$need_lib_prefix" != no; then
# Add the "lib" prefix for modules if required
func_stripname '' '.la' "$outputname"
name=$func_stripname_result
eval shared_ext=\"$shrext_cmds\"
eval libname=\"$libname_spec\"
else
func_stripname '' '.la' "$outputname"
libname=$func_stripname_result
fi
;;
esac
if test -n "$objs"; then
if test "$deplibs_check_method" != pass_all; then
func_fatal_error "cannot build libtool library \`$output' from non-libtool objects on this host:$objs"
else
echo
$ECHO "*** Warning: Linking the shared library $output against the non-libtool"
$ECHO "*** objects $objs is not portable!"
func_append libobjs " $objs"
fi
fi
test "$dlself" != no && \
func_warning "\`-dlopen self' is ignored for libtool libraries"
set dummy $rpath
shift
test "$#" -gt 1 && \
func_warning "ignoring multiple \`-rpath's for a libtool library"
install_libdir="$1"
oldlibs=
if test -z "$rpath"; then
if test "$build_libtool_libs" = yes; then
# Building a libtool convenience library.
# Some compilers have problems with a `.al' extension so
# convenience libraries should have the same extension an
# archive normally would.
oldlibs="$output_objdir/$libname.$libext $oldlibs"
build_libtool_libs=convenience
build_old_libs=yes
fi
test -n "$vinfo" && \
func_warning "\`-version-info/-version-number' is ignored for convenience libraries"
test -n "$release" && \
func_warning "\`-release' is ignored for convenience libraries"
else
# Parse the version information argument.
save_ifs="$IFS"; IFS=':'
set dummy $vinfo 0 0 0
shift
IFS="$save_ifs"
test -n "$7" && \
func_fatal_help "too many parameters to \`-version-info'"
# convert absolute version numbers to libtool ages
# this retains compatibility with .la files and attempts
# to make the code below a bit more comprehensible
case $vinfo_number in
yes)
number_major="$1"
number_minor="$2"
number_revision="$3"
#
# There are really only two kinds -- those that
# use the current revision as the major version
# and those that subtract age and use age as
# a minor version. But, then there is irix
# which has an extra 1 added just for fun
#
case $version_type in
# correct linux to gnu/linux during the next big refactor
darwin|linux|osf|windows|none)
func_arith $number_major + $number_minor
current=$func_arith_result
age="$number_minor"
revision="$number_revision"
;;
freebsd-aout|freebsd-elf|qnx|sunos)
current="$number_major"
revision="$number_minor"
age="0"
;;
irix|nonstopux)
func_arith $number_major + $number_minor
current=$func_arith_result
age="$number_minor"
revision="$number_minor"
lt_irix_increment=no
;;
*)
func_fatal_configuration "$modename: unknown library version type \`$version_type'"
;;
esac
;;
no)
current="$1"
revision="$2"
age="$3"
;;
esac
# Check that each of the things are valid numbers.
case $current in
0|[1-9]|[1-9][0-9]|[1-9][0-9][0-9]|[1-9][0-9][0-9][0-9]|[1-9][0-9][0-9][0-9][0-9]) ;;
*)
func_error "CURRENT \`$current' must be a nonnegative integer"
func_fatal_error "\`$vinfo' is not valid version information"
;;
esac
case $revision in
0|[1-9]|[1-9][0-9]|[1-9][0-9][0-9]|[1-9][0-9][0-9][0-9]|[1-9][0-9][0-9][0-9][0-9]) ;;
*)
func_error "REVISION \`$revision' must be a nonnegative integer"
func_fatal_error "\`$vinfo' is not valid version information"
;;
esac
case $age in
0|[1-9]|[1-9][0-9]|[1-9][0-9][0-9]|[1-9][0-9][0-9][0-9]|[1-9][0-9][0-9][0-9][0-9]) ;;
*)
func_error "AGE \`$age' must be a nonnegative integer"
func_fatal_error "\`$vinfo' is not valid version information"
;;
esac
if test "$age" -gt "$current"; then
func_error "AGE \`$age' is greater than the current interface number \`$current'"
func_fatal_error "\`$vinfo' is not valid version information"
fi
# Calculate the version variables.
major=
versuffix=
verstring=
case $version_type in
none) ;;
darwin)
# Like Linux, but with the current version available in
# verstring for coding it into the library header
func_arith $current - $age
major=.$func_arith_result
versuffix="$major.$age.$revision"
# Darwin ld doesn't like 0 for these options...
func_arith $current + 1
minor_current=$func_arith_result
xlcverstring="${wl}-compatibility_version ${wl}$minor_current ${wl}-current_version ${wl}$minor_current.$revision"
verstring="-compatibility_version $minor_current -current_version $minor_current.$revision"
;;
freebsd-aout)
major=".$current"
versuffix=".$current.$revision";
;;
freebsd-elf)
major=".$current"
versuffix=".$current"
;;
irix | nonstopux)
if test "X$lt_irix_increment" = "Xno"; then
func_arith $current - $age
else
func_arith $current - $age + 1
fi
major=$func_arith_result
case $version_type in
nonstopux) verstring_prefix=nonstopux ;;
*) verstring_prefix=sgi ;;
esac
verstring="$verstring_prefix$major.$revision"
# Add in all the interfaces that we are compatible with.
loop=$revision
while test "$loop" -ne 0; do
func_arith $revision - $loop
iface=$func_arith_result
func_arith $loop - 1
loop=$func_arith_result
verstring="$verstring_prefix$major.$iface:$verstring"
done
# Before this point, $major must not contain `.'.
major=.$major
versuffix="$major.$revision"
;;
linux) # correct to gnu/linux during the next big refactor
func_arith $current - $age
major=.$func_arith_result
versuffix="$major.$age.$revision"
;;
osf)
func_arith $current - $age
major=.$func_arith_result
versuffix=".$current.$age.$revision"
verstring="$current.$age.$revision"
# Add in all the interfaces that we are compatible with.
loop=$age
while test "$loop" -ne 0; do
func_arith $current - $loop
iface=$func_arith_result
func_arith $loop - 1
loop=$func_arith_result
verstring="$verstring:${iface}.0"
done
# Make executables depend on our current version.
func_append verstring ":${current}.0"
;;
qnx)
major=".$current"
versuffix=".$current"
;;
sunos)
major=".$current"
versuffix=".$current.$revision"
;;
windows)
# Use '-' rather than '.', since we only want one
# extension on DOS 8.3 filesystems.
func_arith $current - $age
major=$func_arith_result
versuffix="-$major"
case $host in
x86_64-*mingw32*)
versuffix="6-$major"
;;
*)
versuffix="-$major"
;;
esac
;;
*)
func_fatal_configuration "unknown library version type \`$version_type'"
;;
esac
# Clear the version info if we defaulted, and they specified a release.
if test -z "$vinfo" && test -n "$release"; then
major=
case $version_type in
darwin)
# we can't check for "0.0" in archive_cmds due to quoting
# problems, so we reset it completely
verstring=
;;
*)
verstring="0.0"
;;
esac
if test "$need_version" = no; then
versuffix=
else
versuffix=".0.0"
fi
fi
# Remove version info from name if versioning should be avoided
if test "$avoid_version" = yes && test "$need_version" = no; then
major=
versuffix=
verstring=""
fi
# Check to see if the archive will have undefined symbols.
if test "$allow_undefined" = yes; then
if test "$allow_undefined_flag" = unsupported; then
func_warning "undefined symbols not allowed in $host shared libraries"
build_libtool_libs=no
build_old_libs=yes
fi
else
# Don't allow undefined symbols.
allow_undefined_flag="$no_undefined_flag"
fi
fi
func_generate_dlsyms "$libname" "$libname" "yes"
func_append libobjs " $symfileobj"
test "X$libobjs" = "X " && libobjs=
if test "$opt_mode" != relink; then
# Remove our outputs, but don't remove object files since they
# may have been created when compiling PIC objects.
removelist=
tempremovelist=`$ECHO "$output_objdir/*"`
for p in $tempremovelist; do
case $p in
*.$objext | *.gcno)
;;
$output_objdir/$outputname | $output_objdir/$libname.* | $output_objdir/${libname}${release}.*)
if test "X$precious_files_regex" != "X"; then
if $ECHO "$p" | $EGREP -e "$precious_files_regex" >/dev/null 2>&1
then
continue
fi
fi
func_append removelist " $p"
;;
*) ;;
esac
done
test -n "$removelist" && \
func_show_eval "${RM}r \$removelist"
fi
# Now set the variables for building old libraries.
if test "$build_old_libs" = yes && test "$build_libtool_libs" != convenience ; then
func_append oldlibs " $output_objdir/$libname.$libext"
# Transform .lo files to .o files.
oldobjs="$objs "`$ECHO "$libobjs" | $SP2NL | $SED "/\.${libext}$/d; $lo2o" | $NL2SP`
fi
# Eliminate all temporary directories.
#for path in $notinst_path; do
# lib_search_path=`$ECHO "$lib_search_path " | $SED "s% $path % %g"`
# deplibs=`$ECHO "$deplibs " | $SED "s% -L$path % %g"`
# dependency_libs=`$ECHO "$dependency_libs " | $SED "s% -L$path % %g"`
#done
if test -n "$xrpath"; then
# If the user specified any rpath flags, then add them.
temp_xrpath=
for libdir in $xrpath; do
func_replace_sysroot "$libdir"
func_append temp_xrpath " -R$func_replace_sysroot_result"
case "$finalize_rpath " in
*" $libdir "*) ;;
*) func_append finalize_rpath " $libdir" ;;
esac
done
if test "$hardcode_into_libs" != yes || test "$build_old_libs" = yes; then
dependency_libs="$temp_xrpath $dependency_libs"
fi
fi
# Make sure dlfiles contains only unique files that won't be dlpreopened
old_dlfiles="$dlfiles"
dlfiles=
for lib in $old_dlfiles; do
case " $dlprefiles $dlfiles " in
*" $lib "*) ;;
*) func_append dlfiles " $lib" ;;
esac
done
# Make sure dlprefiles contains only unique files
old_dlprefiles="$dlprefiles"
dlprefiles=
for lib in $old_dlprefiles; do
case "$dlprefiles " in
*" $lib "*) ;;
*) func_append dlprefiles " $lib" ;;
esac
done
if test "$build_libtool_libs" = yes; then
if test -n "$rpath"; then
case $host in
*-*-cygwin* | *-*-mingw* | *-*-pw32* | *-*-os2* | *-*-beos* | *-cegcc* | *-*-haiku*)
# these systems don't actually have a c library (as such)!
;;
*-*-rhapsody* | *-*-darwin1.[012])
# Rhapsody C library is in the System framework
func_append deplibs " System.ltframework"
;;
*-*-netbsd*)
# Don't link with libc until the a.out ld.so is fixed.
;;
*-*-openbsd* | *-*-freebsd* | *-*-dragonfly*)
# Do not include libc due to us having libc/libc_r.
;;
*-*-sco3.2v5* | *-*-sco5v6*)
# Causes problems with __ctype
;;
*-*-sysv4.2uw2* | *-*-sysv5* | *-*-unixware* | *-*-OpenUNIX*)
# Compiler inserts libc in the correct place for threads to work
;;
*)
# Add libc to deplibs on all other systems if necessary.
if test "$build_libtool_need_lc" = "yes"; then
func_append deplibs " -lc"
fi
;;
esac
fi
# Transform deplibs into only deplibs that can be linked in shared.
name_save=$name
libname_save=$libname
release_save=$release
versuffix_save=$versuffix
major_save=$major
# I'm not sure if I'm treating the release correctly. I think
# release should show up in the -l (ie -lgmp5) so we don't want to
# add it in twice. Is that correct?
release=""
versuffix=""
major=""
newdeplibs=
droppeddeps=no
case $deplibs_check_method in
pass_all)
# Don't check for shared/static. Everything works.
# This might be a little naive. We might want to check
# whether the library exists or not. But this is on
# osf3 & osf4 and I'm not really sure... Just
# implementing what was already the behavior.
newdeplibs=$deplibs
;;
test_compile)
# This code stresses the "libraries are programs" paradigm to its
# limits. Maybe even breaks it. We compile a program, linking it
# against the deplibs as a proxy for the library. Then we can check
# whether they linked in statically or dynamically with ldd.
$opt_dry_run || $RM conftest.c
cat > conftest.c </dev/null`
$nocaseglob
else
potential_libs=`ls $i/$libnameglob[.-]* 2>/dev/null`
fi
for potent_lib in $potential_libs; do
# Follow soft links.
if ls -lLd "$potent_lib" 2>/dev/null |
$GREP " -> " >/dev/null; then
continue
fi
# The statement above tries to avoid entering an
# endless loop below, in case of cyclic links.
# We might still enter an endless loop, since a link
# loop can be closed while we follow links,
# but so what?
potlib="$potent_lib"
while test -h "$potlib" 2>/dev/null; do
potliblink=`ls -ld $potlib | ${SED} 's/.* -> //'`
case $potliblink in
[\\/]* | [A-Za-z]:[\\/]*) potlib="$potliblink";;
*) potlib=`$ECHO "$potlib" | $SED 's,[^/]*$,,'`"$potliblink";;
esac
done
if eval $file_magic_cmd \"\$potlib\" 2>/dev/null |
$SED -e 10q |
$EGREP "$file_magic_regex" > /dev/null; then
func_append newdeplibs " $a_deplib"
a_deplib=""
break 2
fi
done
done
fi
if test -n "$a_deplib" ; then
droppeddeps=yes
echo
$ECHO "*** Warning: linker path does not have real file for library $a_deplib."
echo "*** I have the capability to make that library automatically link in when"
echo "*** you link to this library. But I can only do this if you have a"
echo "*** shared version of the library, which you do not appear to have"
echo "*** because I did check the linker path looking for a file starting"
if test -z "$potlib" ; then
$ECHO "*** with $libname but no candidates were found. (...for file magic test)"
else
$ECHO "*** with $libname and none of the candidates passed a file format test"
$ECHO "*** using a file magic. Last file checked: $potlib"
fi
fi
;;
*)
# Add a -L argument.
func_append newdeplibs " $a_deplib"
;;
esac
done # Gone through all deplibs.
;;
match_pattern*)
set dummy $deplibs_check_method; shift
match_pattern_regex=`expr "$deplibs_check_method" : "$1 \(.*\)"`
for a_deplib in $deplibs; do
case $a_deplib in
-l*)
func_stripname -l '' "$a_deplib"
name=$func_stripname_result
if test "X$allow_libtool_libs_with_static_runtimes" = "Xyes" ; then
case " $predeps $postdeps " in
*" $a_deplib "*)
func_append newdeplibs " $a_deplib"
a_deplib=""
;;
esac
fi
if test -n "$a_deplib" ; then
libname=`eval "\\$ECHO \"$libname_spec\""`
for i in $lib_search_path $sys_lib_search_path $shlib_search_path; do
potential_libs=`ls $i/$libname[.-]* 2>/dev/null`
for potent_lib in $potential_libs; do
potlib="$potent_lib" # see symlink-check above in file_magic test
if eval "\$ECHO \"$potent_lib\"" 2>/dev/null | $SED 10q | \
$EGREP "$match_pattern_regex" > /dev/null; then
func_append newdeplibs " $a_deplib"
a_deplib=""
break 2
fi
done
done
fi
if test -n "$a_deplib" ; then
droppeddeps=yes
echo
$ECHO "*** Warning: linker path does not have real file for library $a_deplib."
echo "*** I have the capability to make that library automatically link in when"
echo "*** you link to this library. But I can only do this if you have a"
echo "*** shared version of the library, which you do not appear to have"
echo "*** because I did check the linker path looking for a file starting"
if test -z "$potlib" ; then
$ECHO "*** with $libname but no candidates were found. (...for regex pattern test)"
else
$ECHO "*** with $libname and none of the candidates passed a file format test"
$ECHO "*** using a regex pattern. Last file checked: $potlib"
fi
fi
;;
*)
# Add a -L argument.
func_append newdeplibs " $a_deplib"
;;
esac
done # Gone through all deplibs.
;;
none | unknown | *)
newdeplibs=""
tmp_deplibs=`$ECHO " $deplibs" | $SED 's/ -lc$//; s/ -[LR][^ ]*//g'`
if test "X$allow_libtool_libs_with_static_runtimes" = "Xyes" ; then
for i in $predeps $postdeps ; do
# can't use Xsed below, because $i might contain '/'
tmp_deplibs=`$ECHO " $tmp_deplibs" | $SED "s,$i,,"`
done
fi
case $tmp_deplibs in
*[!\ \ ]*)
echo
if test "X$deplibs_check_method" = "Xnone"; then
echo "*** Warning: inter-library dependencies are not supported in this platform."
else
echo "*** Warning: inter-library dependencies are not known to be supported."
fi
echo "*** All declared inter-library dependencies are being dropped."
droppeddeps=yes
;;
esac
;;
esac
versuffix=$versuffix_save
major=$major_save
release=$release_save
libname=$libname_save
name=$name_save
case $host in
*-*-rhapsody* | *-*-darwin1.[012])
# On Rhapsody replace the C library with the System framework
newdeplibs=`$ECHO " $newdeplibs" | $SED 's/ -lc / System.ltframework /'`
;;
esac
if test "$droppeddeps" = yes; then
if test "$module" = yes; then
echo
echo "*** Warning: libtool could not satisfy all declared inter-library"
$ECHO "*** dependencies of module $libname. Therefore, libtool will create"
echo "*** a static module, that should work as long as the dlopening"
echo "*** application is linked with the -dlopen flag."
if test -z "$global_symbol_pipe"; then
echo
echo "*** However, this would only work if libtool was able to extract symbol"
echo "*** lists from a program, using \`nm' or equivalent, but libtool could"
echo "*** not find such a program. So, this module is probably useless."
echo "*** \`nm' from GNU binutils and a full rebuild may help."
fi
if test "$build_old_libs" = no; then
oldlibs="$output_objdir/$libname.$libext"
build_libtool_libs=module
build_old_libs=yes
else
build_libtool_libs=no
fi
else
echo "*** The inter-library dependencies that have been dropped here will be"
echo "*** automatically added whenever a program is linked with this library"
echo "*** or is declared to -dlopen it."
if test "$allow_undefined" = no; then
echo
echo "*** Since this library must not contain undefined symbols,"
echo "*** because either the platform does not support them or"
echo "*** it was explicitly requested with -no-undefined,"
echo "*** libtool will only create a static version of it."
if test "$build_old_libs" = no; then
oldlibs="$output_objdir/$libname.$libext"
build_libtool_libs=module
build_old_libs=yes
else
build_libtool_libs=no
fi
fi
fi
fi
# Done checking deplibs!
deplibs=$newdeplibs
fi
# Time to change all our "foo.ltframework" stuff back to "-framework foo"
case $host in
*-*-darwin*)
newdeplibs=`$ECHO " $newdeplibs" | $SED 's% \([^ $]*\).ltframework% -framework \1%g'`
new_inherited_linker_flags=`$ECHO " $new_inherited_linker_flags" | $SED 's% \([^ $]*\).ltframework% -framework \1%g'`
deplibs=`$ECHO " $deplibs" | $SED 's% \([^ $]*\).ltframework% -framework \1%g'`
;;
esac
# move library search paths that coincide with paths to not yet
# installed libraries to the beginning of the library search list
new_libs=
for path in $notinst_path; do
case " $new_libs " in
*" -L$path/$objdir "*) ;;
*)
case " $deplibs " in
*" -L$path/$objdir "*)
func_append new_libs " -L$path/$objdir" ;;
esac
;;
esac
done
for deplib in $deplibs; do
case $deplib in
-L*)
case " $new_libs " in
*" $deplib "*) ;;
*) func_append new_libs " $deplib" ;;
esac
;;
*) func_append new_libs " $deplib" ;;
esac
done
deplibs="$new_libs"
# All the library-specific variables (install_libdir is set above).
library_names=
old_library=
dlname=
# Test again, we may have decided not to build it any more
if test "$build_libtool_libs" = yes; then
# Remove ${wl} instances when linking with ld.
# FIXME: should test the right _cmds variable.
case $archive_cmds in
*\$LD\ *) wl= ;;
esac
if test "$hardcode_into_libs" = yes; then
# Hardcode the library paths
hardcode_libdirs=
dep_rpath=
rpath="$finalize_rpath"
test "$opt_mode" != relink && rpath="$compile_rpath$rpath"
for libdir in $rpath; do
if test -n "$hardcode_libdir_flag_spec"; then
if test -n "$hardcode_libdir_separator"; then
func_replace_sysroot "$libdir"
libdir=$func_replace_sysroot_result
if test -z "$hardcode_libdirs"; then
hardcode_libdirs="$libdir"
else
# Just accumulate the unique libdirs.
case $hardcode_libdir_separator$hardcode_libdirs$hardcode_libdir_separator in
*"$hardcode_libdir_separator$libdir$hardcode_libdir_separator"*)
;;
*)
func_append hardcode_libdirs "$hardcode_libdir_separator$libdir"
;;
esac
fi
else
eval flag=\"$hardcode_libdir_flag_spec\"
func_append dep_rpath " $flag"
fi
elif test -n "$runpath_var"; then
case "$perm_rpath " in
*" $libdir "*) ;;
*) func_append perm_rpath " $libdir" ;;
esac
fi
done
# Substitute the hardcoded libdirs into the rpath.
if test -n "$hardcode_libdir_separator" &&
test -n "$hardcode_libdirs"; then
libdir="$hardcode_libdirs"
eval "dep_rpath=\"$hardcode_libdir_flag_spec\""
fi
if test -n "$runpath_var" && test -n "$perm_rpath"; then
# We should set the runpath_var.
rpath=
for dir in $perm_rpath; do
func_append rpath "$dir:"
done
eval "$runpath_var='$rpath\$$runpath_var'; export $runpath_var"
fi
test -n "$dep_rpath" && deplibs="$dep_rpath $deplibs"
fi
shlibpath="$finalize_shlibpath"
test "$opt_mode" != relink && shlibpath="$compile_shlibpath$shlibpath"
if test -n "$shlibpath"; then
eval "$shlibpath_var='$shlibpath\$$shlibpath_var'; export $shlibpath_var"
fi
# Get the real and link names of the library.
eval shared_ext=\"$shrext_cmds\"
eval library_names=\"$library_names_spec\"
set dummy $library_names
shift
realname="$1"
shift
if test -n "$soname_spec"; then
eval soname=\"$soname_spec\"
else
soname="$realname"
fi
if test -z "$dlname"; then
dlname=$soname
fi
lib="$output_objdir/$realname"
linknames=
for link
do
func_append linknames " $link"
done
# Use standard objects if they are pic
test -z "$pic_flag" && libobjs=`$ECHO "$libobjs" | $SP2NL | $SED "$lo2o" | $NL2SP`
test "X$libobjs" = "X " && libobjs=
delfiles=
if test -n "$export_symbols" && test -n "$include_expsyms"; then
$opt_dry_run || cp "$export_symbols" "$output_objdir/$libname.uexp"
export_symbols="$output_objdir/$libname.uexp"
func_append delfiles " $export_symbols"
fi
orig_export_symbols=
case $host_os in
cygwin* | mingw* | cegcc*)
if test -n "$export_symbols" && test -z "$export_symbols_regex"; then
# exporting using user supplied symfile
if test "x`$SED "$sed_uncomment_deffile" $export_symbols | $SED 1q`" != xEXPORTS; then
# and it's NOT already a .def file. Must figure out
# which of the given symbols are data symbols and tag
# them as such. So, trigger use of export_symbols_cmds.
# export_symbols gets reassigned inside the "prepare
# the list of exported symbols" if statement, so the
# include_expsyms logic still works.
orig_export_symbols="$export_symbols"
export_symbols=
always_export_symbols=yes
fi
fi
;;
esac
# Prepare the list of exported symbols
if test -z "$export_symbols"; then
if test "$always_export_symbols" = yes || test -n "$export_symbols_regex"; then
func_verbose "generating symbol list for \`$libname.la'"
export_symbols="$output_objdir/$libname.exp"
$opt_dry_run || $RM $export_symbols
cmds=$export_symbols_cmds
save_ifs="$IFS"; IFS='~'
for cmd1 in $cmds; do
IFS="$save_ifs"
# Take the normal branch if the nm_file_list_spec branch
# doesn't work or if tool conversion is not needed.
case $nm_file_list_spec~$to_tool_file_cmd in
*~func_convert_file_noop | *~func_convert_file_msys_to_w32 | ~*)
try_normal_branch=yes
eval cmd=\"$cmd1\"
func_len " $cmd"
len=$func_len_result
;;
*)
try_normal_branch=no
;;
esac
if test "$try_normal_branch" = yes \
&& { test "$len" -lt "$max_cmd_len" \
|| test "$max_cmd_len" -le -1; }
then
func_show_eval "$cmd" 'exit $?'
skipped_export=false
elif test -n "$nm_file_list_spec"; then
func_basename "$output"
output_la=$func_basename_result
save_libobjs=$libobjs
save_output=$output
output=${output_objdir}/${output_la}.nm
func_to_tool_file "$output"
libobjs=$nm_file_list_spec$func_to_tool_file_result
func_append delfiles " $output"
func_verbose "creating $NM input file list: $output"
for obj in $save_libobjs; do
func_to_tool_file "$obj"
$ECHO "$func_to_tool_file_result"
done > "$output"
eval cmd=\"$cmd1\"
func_show_eval "$cmd" 'exit $?'
output=$save_output
libobjs=$save_libobjs
skipped_export=false
else
# The command line is too long to execute in one step.
func_verbose "using reloadable object file for export list..."
skipped_export=:
# Break out early, otherwise skipped_export may be
# set to false by a later but shorter cmd.
break
fi
done
IFS="$save_ifs"
if test -n "$export_symbols_regex" && test "X$skipped_export" != "X:"; then
func_show_eval '$EGREP -e "$export_symbols_regex" "$export_symbols" > "${export_symbols}T"'
func_show_eval '$MV "${export_symbols}T" "$export_symbols"'
fi
fi
fi
if test -n "$export_symbols" && test -n "$include_expsyms"; then
tmp_export_symbols="$export_symbols"
test -n "$orig_export_symbols" && tmp_export_symbols="$orig_export_symbols"
$opt_dry_run || eval '$ECHO "$include_expsyms" | $SP2NL >> "$tmp_export_symbols"'
fi
if test "X$skipped_export" != "X:" && test -n "$orig_export_symbols"; then
# The given exports_symbols file has to be filtered, so filter it.
func_verbose "filter symbol list for \`$libname.la' to tag DATA exports"
# FIXME: $output_objdir/$libname.filter potentially contains lots of
# 's' commands which not all seds can handle. GNU sed should be fine
# though. Also, the filter scales superlinearly with the number of
# global variables. join(1) would be nice here, but unfortunately
# isn't a blessed tool.
$opt_dry_run || $SED -e '/[ ,]DATA/!d;s,\(.*\)\([ \,].*\),s|^\1$|\1\2|,' < $export_symbols > $output_objdir/$libname.filter
func_append delfiles " $export_symbols $output_objdir/$libname.filter"
export_symbols=$output_objdir/$libname.def
$opt_dry_run || $SED -f $output_objdir/$libname.filter < $orig_export_symbols > $export_symbols
fi
tmp_deplibs=
for test_deplib in $deplibs; do
case " $convenience " in
*" $test_deplib "*) ;;
*)
func_append tmp_deplibs " $test_deplib"
;;
esac
done
deplibs="$tmp_deplibs"
if test -n "$convenience"; then
if test -n "$whole_archive_flag_spec" &&
test "$compiler_needs_object" = yes &&
test -z "$libobjs"; then
# extract the archives, so we have objects to list.
# TODO: could optimize this to just extract one archive.
whole_archive_flag_spec=
fi
if test -n "$whole_archive_flag_spec"; then
save_libobjs=$libobjs
eval libobjs=\"\$libobjs $whole_archive_flag_spec\"
test "X$libobjs" = "X " && libobjs=
else
gentop="$output_objdir/${outputname}x"
func_append generated " $gentop"
func_extract_archives $gentop $convenience
func_append libobjs " $func_extract_archives_result"
test "X$libobjs" = "X " && libobjs=
fi
fi
if test "$thread_safe" = yes && test -n "$thread_safe_flag_spec"; then
eval flag=\"$thread_safe_flag_spec\"
func_append linker_flags " $flag"
fi
# Make a backup of the uninstalled library when relinking
if test "$opt_mode" = relink; then
$opt_dry_run || eval '(cd $output_objdir && $RM ${realname}U && $MV $realname ${realname}U)' || exit $?
fi
# Do each of the archive commands.
if test "$module" = yes && test -n "$module_cmds" ; then
if test -n "$export_symbols" && test -n "$module_expsym_cmds"; then
eval test_cmds=\"$module_expsym_cmds\"
cmds=$module_expsym_cmds
else
eval test_cmds=\"$module_cmds\"
cmds=$module_cmds
fi
else
if test -n "$export_symbols" && test -n "$archive_expsym_cmds"; then
eval test_cmds=\"$archive_expsym_cmds\"
cmds=$archive_expsym_cmds
else
eval test_cmds=\"$archive_cmds\"
cmds=$archive_cmds
fi
fi
if test "X$skipped_export" != "X:" &&
func_len " $test_cmds" &&
len=$func_len_result &&
test "$len" -lt "$max_cmd_len" || test "$max_cmd_len" -le -1; then
:
else
# The command line is too long to link in one step, link piecewise
# or, if using GNU ld and skipped_export is not :, use a linker
# script.
# Save the value of $output and $libobjs because we want to
# use them later. If we have whole_archive_flag_spec, we
# want to use save_libobjs as it was before
# whole_archive_flag_spec was expanded, because we can't
# assume the linker understands whole_archive_flag_spec.
# This may have to be revisited, in case too many
# convenience libraries get linked in and end up exceeding
# the spec.
if test -z "$convenience" || test -z "$whole_archive_flag_spec"; then
save_libobjs=$libobjs
fi
save_output=$output
func_basename "$output"
output_la=$func_basename_result
# Clear the reloadable object creation command queue and
# initialize k to one.
test_cmds=
concat_cmds=
objlist=
last_robj=
k=1
if test -n "$save_libobjs" && test "X$skipped_export" != "X:" && test "$with_gnu_ld" = yes; then
output=${output_objdir}/${output_la}.lnkscript
func_verbose "creating GNU ld script: $output"
echo 'INPUT (' > $output
for obj in $save_libobjs
do
func_to_tool_file "$obj"
$ECHO "$func_to_tool_file_result" >> $output
done
echo ')' >> $output
func_append delfiles " $output"
func_to_tool_file "$output"
output=$func_to_tool_file_result
elif test -n "$save_libobjs" && test "X$skipped_export" != "X:" && test "X$file_list_spec" != X; then
output=${output_objdir}/${output_la}.lnk
func_verbose "creating linker input file list: $output"
: > $output
set x $save_libobjs
shift
firstobj=
if test "$compiler_needs_object" = yes; then
firstobj="$1 "
shift
fi
for obj
do
func_to_tool_file "$obj"
$ECHO "$func_to_tool_file_result" >> $output
done
func_append delfiles " $output"
func_to_tool_file "$output"
output=$firstobj\"$file_list_spec$func_to_tool_file_result\"
else
if test -n "$save_libobjs"; then
func_verbose "creating reloadable object files..."
output=$output_objdir/$output_la-${k}.$objext
eval test_cmds=\"$reload_cmds\"
func_len " $test_cmds"
len0=$func_len_result
len=$len0
# Loop over the list of objects to be linked.
for obj in $save_libobjs
do
func_len " $obj"
func_arith $len + $func_len_result
len=$func_arith_result
if test "X$objlist" = X ||
test "$len" -lt "$max_cmd_len"; then
func_append objlist " $obj"
else
# The command $test_cmds is almost too long, add a
# command to the queue.
if test "$k" -eq 1 ; then
# The first file doesn't have a previous command to add.
reload_objs=$objlist
eval concat_cmds=\"$reload_cmds\"
else
# All subsequent reloadable object files will link in
# the last one created.
reload_objs="$objlist $last_robj"
eval concat_cmds=\"\$concat_cmds~$reload_cmds~\$RM $last_robj\"
fi
last_robj=$output_objdir/$output_la-${k}.$objext
func_arith $k + 1
k=$func_arith_result
output=$output_objdir/$output_la-${k}.$objext
objlist=" $obj"
func_len " $last_robj"
func_arith $len0 + $func_len_result
len=$func_arith_result
fi
done
# Handle the remaining objects by creating one last
# reloadable object file. All subsequent reloadable object
# files will link in the last one created.
test -z "$concat_cmds" || concat_cmds=$concat_cmds~
reload_objs="$objlist $last_robj"
eval concat_cmds=\"\${concat_cmds}$reload_cmds\"
if test -n "$last_robj"; then
eval concat_cmds=\"\${concat_cmds}~\$RM $last_robj\"
fi
func_append delfiles " $output"
else
output=
fi
if ${skipped_export-false}; then
func_verbose "generating symbol list for \`$libname.la'"
export_symbols="$output_objdir/$libname.exp"
$opt_dry_run || $RM $export_symbols
libobjs=$output
# Append the command to create the export file.
test -z "$concat_cmds" || concat_cmds=$concat_cmds~
eval concat_cmds=\"\$concat_cmds$export_symbols_cmds\"
if test -n "$last_robj"; then
eval concat_cmds=\"\$concat_cmds~\$RM $last_robj\"
fi
fi
test -n "$save_libobjs" &&
func_verbose "creating a temporary reloadable object file: $output"
# Loop through the commands generated above and execute them.
save_ifs="$IFS"; IFS='~'
for cmd in $concat_cmds; do
IFS="$save_ifs"
$opt_silent || {
func_quote_for_expand "$cmd"
eval "func_echo $func_quote_for_expand_result"
}
$opt_dry_run || eval "$cmd" || {
lt_exit=$?
# Restore the uninstalled library and exit
if test "$opt_mode" = relink; then
( cd "$output_objdir" && \
$RM "${realname}T" && \
$MV "${realname}U" "$realname" )
fi
exit $lt_exit
}
done
IFS="$save_ifs"
if test -n "$export_symbols_regex" && ${skipped_export-false}; then
func_show_eval '$EGREP -e "$export_symbols_regex" "$export_symbols" > "${export_symbols}T"'
func_show_eval '$MV "${export_symbols}T" "$export_symbols"'
fi
fi
if ${skipped_export-false}; then
if test -n "$export_symbols" && test -n "$include_expsyms"; then
tmp_export_symbols="$export_symbols"
test -n "$orig_export_symbols" && tmp_export_symbols="$orig_export_symbols"
$opt_dry_run || eval '$ECHO "$include_expsyms" | $SP2NL >> "$tmp_export_symbols"'
fi
if test -n "$orig_export_symbols"; then
# The given exports_symbols file has to be filtered, so filter it.
func_verbose "filter symbol list for \`$libname.la' to tag DATA exports"
# FIXME: $output_objdir/$libname.filter potentially contains lots of
# 's' commands which not all seds can handle. GNU sed should be fine
# though. Also, the filter scales superlinearly with the number of
# global variables. join(1) would be nice here, but unfortunately
# isn't a blessed tool.
$opt_dry_run || $SED -e '/[ ,]DATA/!d;s,\(.*\)\([ \,].*\),s|^\1$|\1\2|,' < $export_symbols > $output_objdir/$libname.filter
func_append delfiles " $export_symbols $output_objdir/$libname.filter"
export_symbols=$output_objdir/$libname.def
$opt_dry_run || $SED -f $output_objdir/$libname.filter < $orig_export_symbols > $export_symbols
fi
fi
libobjs=$output
# Restore the value of output.
output=$save_output
if test -n "$convenience" && test -n "$whole_archive_flag_spec"; then
eval libobjs=\"\$libobjs $whole_archive_flag_spec\"
test "X$libobjs" = "X " && libobjs=
fi
# Expand the library linking commands again to reset the
# value of $libobjs for piecewise linking.
# Do each of the archive commands.
if test "$module" = yes && test -n "$module_cmds" ; then
if test -n "$export_symbols" && test -n "$module_expsym_cmds"; then
cmds=$module_expsym_cmds
else
cmds=$module_cmds
fi
else
if test -n "$export_symbols" && test -n "$archive_expsym_cmds"; then
cmds=$archive_expsym_cmds
else
cmds=$archive_cmds
fi
fi
fi
if test -n "$delfiles"; then
# Append the command to remove temporary files to $cmds.
eval cmds=\"\$cmds~\$RM $delfiles\"
fi
# Add any objects from preloaded convenience libraries
if test -n "$dlprefiles"; then
gentop="$output_objdir/${outputname}x"
func_append generated " $gentop"
func_extract_archives $gentop $dlprefiles
func_append libobjs " $func_extract_archives_result"
test "X$libobjs" = "X " && libobjs=
fi
save_ifs="$IFS"; IFS='~'
for cmd in $cmds; do
IFS="$save_ifs"
eval cmd=\"$cmd\"
$opt_silent || {
func_quote_for_expand "$cmd"
eval "func_echo $func_quote_for_expand_result"
}
$opt_dry_run || eval "$cmd" || {
lt_exit=$?
# Restore the uninstalled library and exit
if test "$opt_mode" = relink; then
( cd "$output_objdir" && \
$RM "${realname}T" && \
$MV "${realname}U" "$realname" )
fi
exit $lt_exit
}
done
IFS="$save_ifs"
# Restore the uninstalled library and exit
if test "$opt_mode" = relink; then
$opt_dry_run || eval '(cd $output_objdir && $RM ${realname}T && $MV $realname ${realname}T && $MV ${realname}U $realname)' || exit $?
if test -n "$convenience"; then
if test -z "$whole_archive_flag_spec"; then
func_show_eval '${RM}r "$gentop"'
fi
fi
exit $EXIT_SUCCESS
fi
# Create links to the real library.
for linkname in $linknames; do
if test "$realname" != "$linkname"; then
func_show_eval '(cd "$output_objdir" && $RM "$linkname" && $LN_S "$realname" "$linkname")' 'exit $?'
fi
done
# If -module or -export-dynamic was specified, set the dlname.
if test "$module" = yes || test "$export_dynamic" = yes; then
# On all known operating systems, these are identical.
dlname="$soname"
fi
fi
;;
obj)
if test -n "$dlfiles$dlprefiles" || test "$dlself" != no; then
func_warning "\`-dlopen' is ignored for objects"
fi
case " $deplibs" in
*\ -l* | *\ -L*)
func_warning "\`-l' and \`-L' are ignored for objects" ;;
esac
test -n "$rpath" && \
func_warning "\`-rpath' is ignored for objects"
test -n "$xrpath" && \
func_warning "\`-R' is ignored for objects"
test -n "$vinfo" && \
func_warning "\`-version-info' is ignored for objects"
test -n "$release" && \
func_warning "\`-release' is ignored for objects"
case $output in
*.lo)
test -n "$objs$old_deplibs" && \
func_fatal_error "cannot build library object \`$output' from non-libtool objects"
libobj=$output
func_lo2o "$libobj"
obj=$func_lo2o_result
;;
*)
libobj=
obj="$output"
;;
esac
# Delete the old objects.
$opt_dry_run || $RM $obj $libobj
# Objects from convenience libraries. This assumes
# single-version convenience libraries. Whenever we create
# different ones for PIC/non-PIC, this we'll have to duplicate
# the extraction.
reload_conv_objs=
gentop=
# reload_cmds runs $LD directly, so let us get rid of
# -Wl from whole_archive_flag_spec and hope we can get by with
# turning comma into space..
wl=
if test -n "$convenience"; then
if test -n "$whole_archive_flag_spec"; then
eval tmp_whole_archive_flags=\"$whole_archive_flag_spec\"
reload_conv_objs=$reload_objs\ `$ECHO "$tmp_whole_archive_flags" | $SED 's|,| |g'`
else
gentop="$output_objdir/${obj}x"
func_append generated " $gentop"
func_extract_archives $gentop $convenience
reload_conv_objs="$reload_objs $func_extract_archives_result"
fi
fi
# If we're not building shared, we need to use non_pic_objs
test "$build_libtool_libs" != yes && libobjs="$non_pic_objects"
# Create the old-style object.
reload_objs="$objs$old_deplibs "`$ECHO "$libobjs" | $SP2NL | $SED "/\.${libext}$/d; /\.lib$/d; $lo2o" | $NL2SP`" $reload_conv_objs" ### testsuite: skip nested quoting test
output="$obj"
func_execute_cmds "$reload_cmds" 'exit $?'
# Exit if we aren't doing a library object file.
if test -z "$libobj"; then
if test -n "$gentop"; then
func_show_eval '${RM}r "$gentop"'
fi
exit $EXIT_SUCCESS
fi
if test "$build_libtool_libs" != yes; then
if test -n "$gentop"; then
func_show_eval '${RM}r "$gentop"'
fi
# Create an invalid libtool object if no PIC, so that we don't
# accidentally link it into a program.
# $show "echo timestamp > $libobj"
# $opt_dry_run || eval "echo timestamp > $libobj" || exit $?
exit $EXIT_SUCCESS
fi
if test -n "$pic_flag" || test "$pic_mode" != default; then
# Only do commands if we really have different PIC objects.
reload_objs="$libobjs $reload_conv_objs"
output="$libobj"
func_execute_cmds "$reload_cmds" 'exit $?'
fi
if test -n "$gentop"; then
func_show_eval '${RM}r "$gentop"'
fi
exit $EXIT_SUCCESS
;;
prog)
case $host in
*cygwin*) func_stripname '' '.exe' "$output"
output=$func_stripname_result.exe;;
esac
test -n "$vinfo" && \
func_warning "\`-version-info' is ignored for programs"
test -n "$release" && \
func_warning "\`-release' is ignored for programs"
test "$preload" = yes \
&& test "$dlopen_support" = unknown \
&& test "$dlopen_self" = unknown \
&& test "$dlopen_self_static" = unknown && \
func_warning "\`LT_INIT([dlopen])' not used. Assuming no dlopen support."
case $host in
*-*-rhapsody* | *-*-darwin1.[012])
# On Rhapsody replace the C library is the System framework
compile_deplibs=`$ECHO " $compile_deplibs" | $SED 's/ -lc / System.ltframework /'`
finalize_deplibs=`$ECHO " $finalize_deplibs" | $SED 's/ -lc / System.ltframework /'`
;;
esac
case $host in
*-*-darwin*)
# Don't allow lazy linking, it breaks C++ global constructors
# But is supposedly fixed on 10.4 or later (yay!).
if test "$tagname" = CXX ; then
case ${MACOSX_DEPLOYMENT_TARGET-10.0} in
10.[0123])
func_append compile_command " ${wl}-bind_at_load"
func_append finalize_command " ${wl}-bind_at_load"
;;
esac
fi
# Time to change all our "foo.ltframework" stuff back to "-framework foo"
compile_deplibs=`$ECHO " $compile_deplibs" | $SED 's% \([^ $]*\).ltframework% -framework \1%g'`
finalize_deplibs=`$ECHO " $finalize_deplibs" | $SED 's% \([^ $]*\).ltframework% -framework \1%g'`
;;
esac
# move library search paths that coincide with paths to not yet
# installed libraries to the beginning of the library search list
new_libs=
for path in $notinst_path; do
case " $new_libs " in
*" -L$path/$objdir "*) ;;
*)
case " $compile_deplibs " in
*" -L$path/$objdir "*)
func_append new_libs " -L$path/$objdir" ;;
esac
;;
esac
done
for deplib in $compile_deplibs; do
case $deplib in
-L*)
case " $new_libs " in
*" $deplib "*) ;;
*) func_append new_libs " $deplib" ;;
esac
;;
*) func_append new_libs " $deplib" ;;
esac
done
compile_deplibs="$new_libs"
func_append compile_command " $compile_deplibs"
func_append finalize_command " $finalize_deplibs"
if test -n "$rpath$xrpath"; then
# If the user specified any rpath flags, then add them.
for libdir in $rpath $xrpath; do
# This is the magic to use -rpath.
case "$finalize_rpath " in
*" $libdir "*) ;;
*) func_append finalize_rpath " $libdir" ;;
esac
done
fi
# Now hardcode the library paths
rpath=
hardcode_libdirs=
for libdir in $compile_rpath $finalize_rpath; do
if test -n "$hardcode_libdir_flag_spec"; then
if test -n "$hardcode_libdir_separator"; then
if test -z "$hardcode_libdirs"; then
hardcode_libdirs="$libdir"
else
# Just accumulate the unique libdirs.
case $hardcode_libdir_separator$hardcode_libdirs$hardcode_libdir_separator in
*"$hardcode_libdir_separator$libdir$hardcode_libdir_separator"*)
;;
*)
func_append hardcode_libdirs "$hardcode_libdir_separator$libdir"
;;
esac
fi
else
eval flag=\"$hardcode_libdir_flag_spec\"
func_append rpath " $flag"
fi
elif test -n "$runpath_var"; then
case "$perm_rpath " in
*" $libdir "*) ;;
*) func_append perm_rpath " $libdir" ;;
esac
fi
case $host in
*-*-cygwin* | *-*-mingw* | *-*-pw32* | *-*-os2* | *-cegcc*)
testbindir=`${ECHO} "$libdir" | ${SED} -e 's*/lib$*/bin*'`
case :$dllsearchpath: in
*":$libdir:"*) ;;
::) dllsearchpath=$libdir;;
*) func_append dllsearchpath ":$libdir";;
esac
case :$dllsearchpath: in
*":$testbindir:"*) ;;
::) dllsearchpath=$testbindir;;
*) func_append dllsearchpath ":$testbindir";;
esac
;;
esac
done
# Substitute the hardcoded libdirs into the rpath.
if test -n "$hardcode_libdir_separator" &&
test -n "$hardcode_libdirs"; then
libdir="$hardcode_libdirs"
eval rpath=\" $hardcode_libdir_flag_spec\"
fi
compile_rpath="$rpath"
rpath=
hardcode_libdirs=
for libdir in $finalize_rpath; do
if test -n "$hardcode_libdir_flag_spec"; then
if test -n "$hardcode_libdir_separator"; then
if test -z "$hardcode_libdirs"; then
hardcode_libdirs="$libdir"
else
# Just accumulate the unique libdirs.
case $hardcode_libdir_separator$hardcode_libdirs$hardcode_libdir_separator in
*"$hardcode_libdir_separator$libdir$hardcode_libdir_separator"*)
;;
*)
func_append hardcode_libdirs "$hardcode_libdir_separator$libdir"
;;
esac
fi
else
eval flag=\"$hardcode_libdir_flag_spec\"
func_append rpath " $flag"
fi
elif test -n "$runpath_var"; then
case "$finalize_perm_rpath " in
*" $libdir "*) ;;
*) func_append finalize_perm_rpath " $libdir" ;;
esac
fi
done
# Substitute the hardcoded libdirs into the rpath.
if test -n "$hardcode_libdir_separator" &&
test -n "$hardcode_libdirs"; then
libdir="$hardcode_libdirs"
eval rpath=\" $hardcode_libdir_flag_spec\"
fi
finalize_rpath="$rpath"
if test -n "$libobjs" && test "$build_old_libs" = yes; then
# Transform all the library objects into standard objects.
compile_command=`$ECHO "$compile_command" | $SP2NL | $SED "$lo2o" | $NL2SP`
finalize_command=`$ECHO "$finalize_command" | $SP2NL | $SED "$lo2o" | $NL2SP`
fi
func_generate_dlsyms "$outputname" "@PROGRAM@" "no"
# template prelinking step
if test -n "$prelink_cmds"; then
func_execute_cmds "$prelink_cmds" 'exit $?'
fi
wrappers_required=yes
case $host in
*cegcc* | *mingw32ce*)
# Disable wrappers for cegcc and mingw32ce hosts, we are cross compiling anyway.
wrappers_required=no
;;
*cygwin* | *mingw* )
if test "$build_libtool_libs" != yes; then
wrappers_required=no
fi
;;
*)
if test "$need_relink" = no || test "$build_libtool_libs" != yes; then
wrappers_required=no
fi
;;
esac
if test "$wrappers_required" = no; then
# Replace the output file specification.
compile_command=`$ECHO "$compile_command" | $SED 's%@OUTPUT@%'"$output"'%g'`
link_command="$compile_command$compile_rpath"
# We have no uninstalled library dependencies, so finalize right now.
exit_status=0
func_show_eval "$link_command" 'exit_status=$?'
if test -n "$postlink_cmds"; then
func_to_tool_file "$output"
postlink_cmds=`func_echo_all "$postlink_cmds" | $SED -e 's%@OUTPUT@%'"$output"'%g' -e 's%@TOOL_OUTPUT@%'"$func_to_tool_file_result"'%g'`
func_execute_cmds "$postlink_cmds" 'exit $?'
fi
# Delete the generated files.
if test -f "$output_objdir/${outputname}S.${objext}"; then
func_show_eval '$RM "$output_objdir/${outputname}S.${objext}"'
fi
exit $exit_status
fi
if test -n "$compile_shlibpath$finalize_shlibpath"; then
compile_command="$shlibpath_var=\"$compile_shlibpath$finalize_shlibpath\$$shlibpath_var\" $compile_command"
fi
if test -n "$finalize_shlibpath"; then
finalize_command="$shlibpath_var=\"$finalize_shlibpath\$$shlibpath_var\" $finalize_command"
fi
compile_var=
finalize_var=
if test -n "$runpath_var"; then
if test -n "$perm_rpath"; then
# We should set the runpath_var.
rpath=
for dir in $perm_rpath; do
func_append rpath "$dir:"
done
compile_var="$runpath_var=\"$rpath\$$runpath_var\" "
fi
if test -n "$finalize_perm_rpath"; then
# We should set the runpath_var.
rpath=
for dir in $finalize_perm_rpath; do
func_append rpath "$dir:"
done
finalize_var="$runpath_var=\"$rpath\$$runpath_var\" "
fi
fi
if test "$no_install" = yes; then
# We don't need to create a wrapper script.
link_command="$compile_var$compile_command$compile_rpath"
# Replace the output file specification.
link_command=`$ECHO "$link_command" | $SED 's%@OUTPUT@%'"$output"'%g'`
# Delete the old output file.
$opt_dry_run || $RM $output
# Link the executable and exit
func_show_eval "$link_command" 'exit $?'
if test -n "$postlink_cmds"; then
func_to_tool_file "$output"
postlink_cmds=`func_echo_all "$postlink_cmds" | $SED -e 's%@OUTPUT@%'"$output"'%g' -e 's%@TOOL_OUTPUT@%'"$func_to_tool_file_result"'%g'`
func_execute_cmds "$postlink_cmds" 'exit $?'
fi
exit $EXIT_SUCCESS
fi
if test "$hardcode_action" = relink; then
# Fast installation is not supported
link_command="$compile_var$compile_command$compile_rpath"
relink_command="$finalize_var$finalize_command$finalize_rpath"
func_warning "this platform does not like uninstalled shared libraries"
func_warning "\`$output' will be relinked during installation"
else
if test "$fast_install" != no; then
link_command="$finalize_var$compile_command$finalize_rpath"
if test "$fast_install" = yes; then
relink_command=`$ECHO "$compile_var$compile_command$compile_rpath" | $SED 's%@OUTPUT@%\$progdir/\$file%g'`
else
# fast_install is set to needless
relink_command=
fi
else
link_command="$compile_var$compile_command$compile_rpath"
relink_command="$finalize_var$finalize_command$finalize_rpath"
fi
fi
# Replace the output file specification.
link_command=`$ECHO "$link_command" | $SED 's%@OUTPUT@%'"$output_objdir/$outputname"'%g'`
# Delete the old output files.
$opt_dry_run || $RM $output $output_objdir/$outputname $output_objdir/lt-$outputname
func_show_eval "$link_command" 'exit $?'
if test -n "$postlink_cmds"; then
func_to_tool_file "$output_objdir/$outputname"
postlink_cmds=`func_echo_all "$postlink_cmds" | $SED -e 's%@OUTPUT@%'"$output_objdir/$outputname"'%g' -e 's%@TOOL_OUTPUT@%'"$func_to_tool_file_result"'%g'`
func_execute_cmds "$postlink_cmds" 'exit $?'
fi
# Now create the wrapper script.
func_verbose "creating $output"
# Quote the relink command for shipping.
if test -n "$relink_command"; then
# Preserve any variables that may affect compiler behavior
for var in $variables_saved_for_relink; do
if eval test -z \"\${$var+set}\"; then
relink_command="{ test -z \"\${$var+set}\" || $lt_unset $var || { $var=; export $var; }; }; $relink_command"
elif eval var_value=\$$var; test -z "$var_value"; then
relink_command="$var=; export $var; $relink_command"
else
func_quote_for_eval "$var_value"
relink_command="$var=$func_quote_for_eval_result; export $var; $relink_command"
fi
done
relink_command="(cd `pwd`; $relink_command)"
relink_command=`$ECHO "$relink_command" | $SED "$sed_quote_subst"`
fi
# Only actually do things if not in dry run mode.
$opt_dry_run || {
# win32 will think the script is a binary if it has
# a .exe suffix, so we strip it off here.
case $output in
*.exe) func_stripname '' '.exe' "$output"
output=$func_stripname_result ;;
esac
# test for cygwin because mv fails w/o .exe extensions
case $host in
*cygwin*)
exeext=.exe
func_stripname '' '.exe' "$outputname"
outputname=$func_stripname_result ;;
*) exeext= ;;
esac
case $host in
*cygwin* | *mingw* )
func_dirname_and_basename "$output" "" "."
output_name=$func_basename_result
output_path=$func_dirname_result
cwrappersource="$output_path/$objdir/lt-$output_name.c"
cwrapper="$output_path/$output_name.exe"
$RM $cwrappersource $cwrapper
trap "$RM $cwrappersource $cwrapper; exit $EXIT_FAILURE" 1 2 15
func_emit_cwrapperexe_src > $cwrappersource
# The wrapper executable is built using the $host compiler,
# because it contains $host paths and files. If cross-
# compiling, it, like the target executable, must be
# executed on the $host or under an emulation environment.
$opt_dry_run || {
$LTCC $LTCFLAGS -o $cwrapper $cwrappersource
$STRIP $cwrapper
}
# Now, create the wrapper script for func_source use:
func_ltwrapper_scriptname $cwrapper
$RM $func_ltwrapper_scriptname_result
trap "$RM $func_ltwrapper_scriptname_result; exit $EXIT_FAILURE" 1 2 15
$opt_dry_run || {
# note: this script will not be executed, so do not chmod.
if test "x$build" = "x$host" ; then
$cwrapper --lt-dump-script > $func_ltwrapper_scriptname_result
else
func_emit_wrapper no > $func_ltwrapper_scriptname_result
fi
}
;;
* )
$RM $output
trap "$RM $output; exit $EXIT_FAILURE" 1 2 15
func_emit_wrapper no > $output
chmod +x $output
;;
esac
}
exit $EXIT_SUCCESS
;;
esac
# See if we need to build an old-fashioned archive.
for oldlib in $oldlibs; do
if test "$build_libtool_libs" = convenience; then
oldobjs="$libobjs_save $symfileobj"
addlibs="$convenience"
build_libtool_libs=no
else
if test "$build_libtool_libs" = module; then
oldobjs="$libobjs_save"
build_libtool_libs=no
else
oldobjs="$old_deplibs $non_pic_objects"
if test "$preload" = yes && test -f "$symfileobj"; then
func_append oldobjs " $symfileobj"
fi
fi
addlibs="$old_convenience"
fi
if test -n "$addlibs"; then
gentop="$output_objdir/${outputname}x"
func_append generated " $gentop"
func_extract_archives $gentop $addlibs
func_append oldobjs " $func_extract_archives_result"
fi
# Do each command in the archive commands.
if test -n "$old_archive_from_new_cmds" && test "$build_libtool_libs" = yes; then
cmds=$old_archive_from_new_cmds
else
# Add any objects from preloaded convenience libraries
if test -n "$dlprefiles"; then
gentop="$output_objdir/${outputname}x"
func_append generated " $gentop"
func_extract_archives $gentop $dlprefiles
func_append oldobjs " $func_extract_archives_result"
fi
# POSIX demands no paths to be encoded in archives. We have
# to avoid creating archives with duplicate basenames if we
# might have to extract them afterwards, e.g., when creating a
# static archive out of a convenience library, or when linking
# the entirety of a libtool archive into another (currently
# not supported by libtool).
if (for obj in $oldobjs
do
func_basename "$obj"
$ECHO "$func_basename_result"
done | sort | sort -uc >/dev/null 2>&1); then
:
else
echo "copying selected object files to avoid basename conflicts..."
gentop="$output_objdir/${outputname}x"
func_append generated " $gentop"
func_mkdir_p "$gentop"
save_oldobjs=$oldobjs
oldobjs=
counter=1
for obj in $save_oldobjs
do
func_basename "$obj"
objbase="$func_basename_result"
case " $oldobjs " in
" ") oldobjs=$obj ;;
*[\ /]"$objbase "*)
while :; do
# Make sure we don't pick an alternate name that also
# overlaps.
newobj=lt$counter-$objbase
func_arith $counter + 1
counter=$func_arith_result
case " $oldobjs " in
*[\ /]"$newobj "*) ;;
*) if test ! -f "$gentop/$newobj"; then break; fi ;;
esac
done
func_show_eval "ln $obj $gentop/$newobj || cp $obj $gentop/$newobj"
func_append oldobjs " $gentop/$newobj"
;;
*) func_append oldobjs " $obj" ;;
esac
done
fi
func_to_tool_file "$oldlib" func_convert_file_msys_to_w32
tool_oldlib=$func_to_tool_file_result
eval cmds=\"$old_archive_cmds\"
func_len " $cmds"
len=$func_len_result
if test "$len" -lt "$max_cmd_len" || test "$max_cmd_len" -le -1; then
cmds=$old_archive_cmds
elif test -n "$archiver_list_spec"; then
func_verbose "using command file archive linking..."
for obj in $oldobjs
do
func_to_tool_file "$obj"
$ECHO "$func_to_tool_file_result"
done > $output_objdir/$libname.libcmd
func_to_tool_file "$output_objdir/$libname.libcmd"
oldobjs=" $archiver_list_spec$func_to_tool_file_result"
cmds=$old_archive_cmds
else
# the command line is too long to link in one step, link in parts
func_verbose "using piecewise archive linking..."
save_RANLIB=$RANLIB
RANLIB=:
objlist=
concat_cmds=
save_oldobjs=$oldobjs
oldobjs=
# Is there a better way of finding the last object in the list?
for obj in $save_oldobjs
do
last_oldobj=$obj
done
eval test_cmds=\"$old_archive_cmds\"
func_len " $test_cmds"
len0=$func_len_result
len=$len0
for obj in $save_oldobjs
do
func_len " $obj"
func_arith $len + $func_len_result
len=$func_arith_result
func_append objlist " $obj"
if test "$len" -lt "$max_cmd_len"; then
:
else
# the above command should be used before it gets too long
oldobjs=$objlist
if test "$obj" = "$last_oldobj" ; then
RANLIB=$save_RANLIB
fi
test -z "$concat_cmds" || concat_cmds=$concat_cmds~
eval concat_cmds=\"\${concat_cmds}$old_archive_cmds\"
objlist=
len=$len0
fi
done
RANLIB=$save_RANLIB
oldobjs=$objlist
if test "X$oldobjs" = "X" ; then
eval cmds=\"\$concat_cmds\"
else
eval cmds=\"\$concat_cmds~\$old_archive_cmds\"
fi
fi
fi
func_execute_cmds "$cmds" 'exit $?'
done
test -n "$generated" && \
func_show_eval "${RM}r$generated"
# Now create the libtool archive.
case $output in
*.la)
old_library=
test "$build_old_libs" = yes && old_library="$libname.$libext"
func_verbose "creating $output"
# Preserve any variables that may affect compiler behavior
for var in $variables_saved_for_relink; do
if eval test -z \"\${$var+set}\"; then
relink_command="{ test -z \"\${$var+set}\" || $lt_unset $var || { $var=; export $var; }; }; $relink_command"
elif eval var_value=\$$var; test -z "$var_value"; then
relink_command="$var=; export $var; $relink_command"
else
func_quote_for_eval "$var_value"
relink_command="$var=$func_quote_for_eval_result; export $var; $relink_command"
fi
done
# Quote the link command for shipping.
relink_command="(cd `pwd`; $SHELL $progpath $preserve_args --mode=relink $libtool_args @inst_prefix_dir@)"
relink_command=`$ECHO "$relink_command" | $SED "$sed_quote_subst"`
if test "$hardcode_automatic" = yes ; then
relink_command=
fi
# Only create the output if not a dry run.
$opt_dry_run || {
for installed in no yes; do
if test "$installed" = yes; then
if test -z "$install_libdir"; then
break
fi
output="$output_objdir/$outputname"i
# Replace all uninstalled libtool libraries with the installed ones
newdependency_libs=
for deplib in $dependency_libs; do
case $deplib in
*.la)
func_basename "$deplib"
name="$func_basename_result"
func_resolve_sysroot "$deplib"
eval libdir=`${SED} -n -e 's/^libdir=\(.*\)$/\1/p' $func_resolve_sysroot_result`
test -z "$libdir" && \
func_fatal_error "\`$deplib' is not a valid libtool archive"
func_append newdependency_libs " ${lt_sysroot:+=}$libdir/$name"
;;
-L*)
func_stripname -L '' "$deplib"
func_replace_sysroot "$func_stripname_result"
func_append newdependency_libs " -L$func_replace_sysroot_result"
;;
-R*)
func_stripname -R '' "$deplib"
func_replace_sysroot "$func_stripname_result"
func_append newdependency_libs " -R$func_replace_sysroot_result"
;;
*) func_append newdependency_libs " $deplib" ;;
esac
done
dependency_libs="$newdependency_libs"
newdlfiles=
for lib in $dlfiles; do
case $lib in
*.la)
func_basename "$lib"
name="$func_basename_result"
eval libdir=`${SED} -n -e 's/^libdir=\(.*\)$/\1/p' $lib`
test -z "$libdir" && \
func_fatal_error "\`$lib' is not a valid libtool archive"
func_append newdlfiles " ${lt_sysroot:+=}$libdir/$name"
;;
*) func_append newdlfiles " $lib" ;;
esac
done
dlfiles="$newdlfiles"
newdlprefiles=
for lib in $dlprefiles; do
case $lib in
*.la)
# Only pass preopened files to the pseudo-archive (for
# eventual linking with the app. that links it) if we
# didn't already link the preopened objects directly into
# the library:
func_basename "$lib"
name="$func_basename_result"
eval libdir=`${SED} -n -e 's/^libdir=\(.*\)$/\1/p' $lib`
test -z "$libdir" && \
func_fatal_error "\`$lib' is not a valid libtool archive"
func_append newdlprefiles " ${lt_sysroot:+=}$libdir/$name"
;;
esac
done
dlprefiles="$newdlprefiles"
else
newdlfiles=
for lib in $dlfiles; do
case $lib in
[\\/]* | [A-Za-z]:[\\/]*) abs="$lib" ;;
*) abs=`pwd`"/$lib" ;;
esac
func_append newdlfiles " $abs"
done
dlfiles="$newdlfiles"
newdlprefiles=
for lib in $dlprefiles; do
case $lib in
[\\/]* | [A-Za-z]:[\\/]*) abs="$lib" ;;
*) abs=`pwd`"/$lib" ;;
esac
func_append newdlprefiles " $abs"
done
dlprefiles="$newdlprefiles"
fi
$RM $output
# place dlname in correct position for cygwin
# In fact, it would be nice if we could use this code for all target
# systems that can't hard-code library paths into their executables
# and that have no shared library path variable independent of PATH,
# but it turns out we can't easily determine that from inspecting
# libtool variables, so we have to hard-code the OSs to which it
# applies here; at the moment, that means platforms that use the PE
# object format with DLL files. See the long comment at the top of
# tests/bindir.at for full details.
tdlname=$dlname
case $host,$output,$installed,$module,$dlname in
*cygwin*,*lai,yes,no,*.dll | *mingw*,*lai,yes,no,*.dll | *cegcc*,*lai,yes,no,*.dll)
# If a -bindir argument was supplied, place the dll there.
if test "x$bindir" != x ;
then
func_relative_path "$install_libdir" "$bindir"
tdlname=$func_relative_path_result$dlname
else
# Otherwise fall back on heuristic.
tdlname=../bin/$dlname
fi
;;
esac
$ECHO > $output "\
# $outputname - a libtool library file
# Generated by $PROGRAM (GNU $PACKAGE$TIMESTAMP) $VERSION
#
# Please DO NOT delete this file!
# It is necessary for linking the library.
# The name that we can dlopen(3).
dlname='$tdlname'
# Names of this library.
library_names='$library_names'
# The name of the static archive.
old_library='$old_library'
# Linker flags that can not go in dependency_libs.
inherited_linker_flags='$new_inherited_linker_flags'
# Libraries that this one depends upon.
dependency_libs='$dependency_libs'
# Names of additional weak libraries provided by this library
weak_library_names='$weak_libs'
# Version information for $libname.
current=$current
age=$age
revision=$revision
# Is this an already installed library?
installed=$installed
# Should we warn about portability when linking against -modules?
shouldnotlink=$module
# Files to dlopen/dlpreopen
dlopen='$dlfiles'
dlpreopen='$dlprefiles'
# Directory that this library needs to be installed in:
libdir='$install_libdir'"
if test "$installed" = no && test "$need_relink" = yes; then
$ECHO >> $output "\
relink_command=\"$relink_command\""
fi
done
}
# Do a symbolic link so that the libtool archive can be found in
# LD_LIBRARY_PATH before the program is installed.
func_show_eval '( cd "$output_objdir" && $RM "$outputname" && $LN_S "../$outputname" "$outputname" )' 'exit $?'
;;
esac
exit $EXIT_SUCCESS
}
{ test "$opt_mode" = link || test "$opt_mode" = relink; } &&
func_mode_link ${1+"$@"}
# func_mode_uninstall arg...
func_mode_uninstall ()
{
$opt_debug
RM="$nonopt"
files=
rmforce=
exit_status=0
# This variable tells wrapper scripts just to set variables rather
# than running their programs.
libtool_install_magic="$magic"
for arg
do
case $arg in
-f) func_append RM " $arg"; rmforce=yes ;;
-*) func_append RM " $arg" ;;
*) func_append files " $arg" ;;
esac
done
test -z "$RM" && \
func_fatal_help "you must specify an RM program"
rmdirs=
for file in $files; do
func_dirname "$file" "" "."
dir="$func_dirname_result"
if test "X$dir" = X.; then
odir="$objdir"
else
odir="$dir/$objdir"
fi
func_basename "$file"
name="$func_basename_result"
test "$opt_mode" = uninstall && odir="$dir"
# Remember odir for removal later, being careful to avoid duplicates
if test "$opt_mode" = clean; then
case " $rmdirs " in
*" $odir "*) ;;
*) func_append rmdirs " $odir" ;;
esac
fi
# Don't error if the file doesn't exist and rm -f was used.
if { test -L "$file"; } >/dev/null 2>&1 ||
{ test -h "$file"; } >/dev/null 2>&1 ||
test -f "$file"; then
:
elif test -d "$file"; then
exit_status=1
continue
elif test "$rmforce" = yes; then
continue
fi
rmfiles="$file"
case $name in
*.la)
# Possibly a libtool archive, so verify it.
if func_lalib_p "$file"; then
func_source $dir/$name
# Delete the libtool libraries and symlinks.
for n in $library_names; do
func_append rmfiles " $odir/$n"
done
test -n "$old_library" && func_append rmfiles " $odir/$old_library"
case "$opt_mode" in
clean)
case " $library_names " in
*" $dlname "*) ;;
*) test -n "$dlname" && func_append rmfiles " $odir/$dlname" ;;
esac
test -n "$libdir" && func_append rmfiles " $odir/$name $odir/${name}i"
;;
uninstall)
if test -n "$library_names"; then
# Do each command in the postuninstall commands.
func_execute_cmds "$postuninstall_cmds" 'test "$rmforce" = yes || exit_status=1'
fi
if test -n "$old_library"; then
# Do each command in the old_postuninstall commands.
func_execute_cmds "$old_postuninstall_cmds" 'test "$rmforce" = yes || exit_status=1'
fi
# FIXME: should reinstall the best remaining shared library.
;;
esac
fi
;;
*.lo)
# Possibly a libtool object, so verify it.
if func_lalib_p "$file"; then
# Read the .lo file
func_source $dir/$name
# Add PIC object to the list of files to remove.
if test -n "$pic_object" &&
test "$pic_object" != none; then
func_append rmfiles " $dir/$pic_object"
fi
# Add non-PIC object to the list of files to remove.
if test -n "$non_pic_object" &&
test "$non_pic_object" != none; then
func_append rmfiles " $dir/$non_pic_object"
fi
fi
;;
*)
if test "$opt_mode" = clean ; then
noexename=$name
case $file in
*.exe)
func_stripname '' '.exe' "$file"
file=$func_stripname_result
func_stripname '' '.exe' "$name"
noexename=$func_stripname_result
# $file with .exe has already been added to rmfiles,
# add $file without .exe
func_append rmfiles " $file"
;;
esac
# Do a test to see if this is a libtool program.
if func_ltwrapper_p "$file"; then
if func_ltwrapper_executable_p "$file"; then
func_ltwrapper_scriptname "$file"
relink_command=
func_source $func_ltwrapper_scriptname_result
func_append rmfiles " $func_ltwrapper_scriptname_result"
else
relink_command=
func_source $dir/$noexename
fi
# note $name still contains .exe if it was in $file originally
# as does the version of $file that was added into $rmfiles
func_append rmfiles " $odir/$name $odir/${name}S.${objext}"
if test "$fast_install" = yes && test -n "$relink_command"; then
func_append rmfiles " $odir/lt-$name"
fi
if test "X$noexename" != "X$name" ; then
func_append rmfiles " $odir/lt-${noexename}.c"
fi
fi
fi
;;
esac
func_show_eval "$RM $rmfiles" 'exit_status=1'
done
# Try to remove the ${objdir}s in the directories where we deleted files
for dir in $rmdirs; do
if test -d "$dir"; then
func_show_eval "rmdir $dir >/dev/null 2>&1"
fi
done
exit $exit_status
}
{ test "$opt_mode" = uninstall || test "$opt_mode" = clean; } &&
func_mode_uninstall ${1+"$@"}
test -z "$opt_mode" && {
help="$generic_help"
func_fatal_help "you must specify a MODE"
}
test -z "$exec_cmd" && \
func_fatal_help "invalid operation mode \`$opt_mode'"
if test -n "$exec_cmd"; then
eval exec "$exec_cmd"
exit $EXIT_FAILURE
fi
exit $exit_status
# The TAGs below are defined such that we never get into a situation
# in which we disable both kinds of libraries. Given conflicting
# choices, we go for a static library, that is the most portable,
# since we can't tell whether shared libraries were disabled because
# the user asked for that or because the platform doesn't support
# them. This is particularly important on AIX, because we don't
# support having both static and shared libraries enabled at the same
# time on that platform, so we default to a shared-only configuration.
# If a disable-shared tag is given, we'll fallback to a static-only
# configuration. But we'll never go from static-only to shared-only.
# ### BEGIN LIBTOOL TAG CONFIG: disable-shared
build_libtool_libs=no
build_old_libs=yes
# ### END LIBTOOL TAG CONFIG: disable-shared
# ### BEGIN LIBTOOL TAG CONFIG: disable-static
build_old_libs=`case $build_libtool_libs in yes) echo no;; *) echo yes;; esac`
# ### END LIBTOOL TAG CONFIG: disable-static
# Local Variables:
# mode:shell-script
# sh-indentation:2
# End:
# vi:sw=2
diff --git a/build-aux/mdate-sh b/build-aux/mdate-sh
index b3719cf..39f48bb 100755
--- a/build-aux/mdate-sh
+++ b/build-aux/mdate-sh
@@ -1,224 +1,224 @@
#!/bin/sh
# Get modification time of a file or directory and pretty-print it.
scriptversion=2010-08-21.06; # UTC
# Copyright (C) 1995-2013 Free Software Foundation, Inc.
# written by Ulrich Drepper , June 1995
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2, or (at your option)
# any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
-# along with this program. If not, see .
+# along with this program. If not, see .
# As a special exception to the GNU General Public License, if you
# distribute this file as part of a program that contains a
# configuration script generated by Autoconf, you may include it under
# the same distribution terms that you use for the rest of that program.
# This file is maintained in Automake, please report
# bugs to or send patches to
# .
if test -n "${ZSH_VERSION+set}" && (emulate sh) >/dev/null 2>&1; then
emulate sh
NULLCMD=:
# Pre-4.2 versions of Zsh do word splitting on ${1+"$@"}, which
# is contrary to our usage. Disable this feature.
alias -g '${1+"$@"}'='"$@"'
setopt NO_GLOB_SUBST
fi
case $1 in
'')
echo "$0: No file. Try '$0 --help' for more information." 1>&2
exit 1;
;;
-h | --h*)
cat <<\EOF
Usage: mdate-sh [--help] [--version] FILE
Pretty-print the modification day of FILE, in the format:
1 January 1970
Report bugs to .
EOF
exit $?
;;
-v | --v*)
echo "mdate-sh $scriptversion"
exit $?
;;
esac
error ()
{
echo "$0: $1" >&2
exit 1
}
# Prevent date giving response in another language.
LANG=C
export LANG
LC_ALL=C
export LC_ALL
LC_TIME=C
export LC_TIME
# GNU ls changes its time format in response to the TIME_STYLE
# variable. Since we cannot assume 'unset' works, revert this
# variable to its documented default.
if test "${TIME_STYLE+set}" = set; then
TIME_STYLE=posix-long-iso
export TIME_STYLE
fi
save_arg1=$1
# Find out how to get the extended ls output of a file or directory.
if ls -L /dev/null 1>/dev/null 2>&1; then
ls_command='ls -L -l -d'
else
ls_command='ls -l -d'
fi
# Avoid user/group names that might have spaces, when possible.
if ls -n /dev/null 1>/dev/null 2>&1; then
ls_command="$ls_command -n"
fi
# A 'ls -l' line looks as follows on OS/2.
# drwxrwx--- 0 Aug 11 2001 foo
# This differs from Unix, which adds ownership information.
# drwxrwx--- 2 root root 4096 Aug 11 2001 foo
#
# To find the date, we split the line on spaces and iterate on words
# until we find a month. This cannot work with files whose owner is a
# user named "Jan", or "Feb", etc. However, it's unlikely that '/'
# will be owned by a user whose name is a month. So we first look at
# the extended ls output of the root directory to decide how many
# words should be skipped to get the date.
# On HPUX /bin/sh, "set" interprets "-rw-r--r--" as options, so the "x" below.
set x`$ls_command /`
# Find which argument is the month.
month=
command=
until test $month
do
test $# -gt 0 || error "failed parsing '$ls_command /' output"
shift
# Add another shift to the command.
command="$command shift;"
case $1 in
Jan) month=January; nummonth=1;;
Feb) month=February; nummonth=2;;
Mar) month=March; nummonth=3;;
Apr) month=April; nummonth=4;;
May) month=May; nummonth=5;;
Jun) month=June; nummonth=6;;
Jul) month=July; nummonth=7;;
Aug) month=August; nummonth=8;;
Sep) month=September; nummonth=9;;
Oct) month=October; nummonth=10;;
Nov) month=November; nummonth=11;;
Dec) month=December; nummonth=12;;
esac
done
test -n "$month" || error "failed parsing '$ls_command /' output"
# Get the extended ls output of the file or directory.
set dummy x`eval "$ls_command \"\\\$save_arg1\""`
# Remove all preceding arguments
eval $command
# Because of the dummy argument above, month is in $2.
#
# On a POSIX system, we should have
#
# $# = 5
# $1 = file size
# $2 = month
# $3 = day
# $4 = year or time
# $5 = filename
#
# On Darwin 7.7.0 and 7.6.0, we have
#
# $# = 4
# $1 = day
# $2 = month
# $3 = year or time
# $4 = filename
# Get the month.
case $2 in
Jan) month=January; nummonth=1;;
Feb) month=February; nummonth=2;;
Mar) month=March; nummonth=3;;
Apr) month=April; nummonth=4;;
May) month=May; nummonth=5;;
Jun) month=June; nummonth=6;;
Jul) month=July; nummonth=7;;
Aug) month=August; nummonth=8;;
Sep) month=September; nummonth=9;;
Oct) month=October; nummonth=10;;
Nov) month=November; nummonth=11;;
Dec) month=December; nummonth=12;;
esac
case $3 in
???*) day=$1;;
*) day=$3; shift;;
esac
# Here we have to deal with the problem that the ls output gives either
# the time of day or the year.
case $3 in
*:*) set `date`; eval year=\$$#
case $2 in
Jan) nummonthtod=1;;
Feb) nummonthtod=2;;
Mar) nummonthtod=3;;
Apr) nummonthtod=4;;
May) nummonthtod=5;;
Jun) nummonthtod=6;;
Jul) nummonthtod=7;;
Aug) nummonthtod=8;;
Sep) nummonthtod=9;;
Oct) nummonthtod=10;;
Nov) nummonthtod=11;;
Dec) nummonthtod=12;;
esac
# For the first six month of the year the time notation can also
# be used for files modified in the last year.
if (expr $nummonth \> $nummonthtod) > /dev/null;
then
year=`expr $year - 1`
fi;;
*) year=$3;;
esac
# The result.
echo $day $month $year
# Local Variables:
# mode: shell-script
# sh-indentation: 2
# eval: (add-hook 'write-file-hooks 'time-stamp)
# time-stamp-start: "scriptversion="
# time-stamp-format: "%:y-%02m-%02d.%02H"
# time-stamp-time-zone: "UTC"
# time-stamp-end: "; # UTC"
# End:
diff --git a/build-aux/missing b/build-aux/missing
index db98974..295de3f 100755
--- a/build-aux/missing
+++ b/build-aux/missing
@@ -1,215 +1,215 @@
#! /bin/sh
# Common wrapper for a few potentially missing GNU programs.
scriptversion=2013-10-28.13; # UTC
# Copyright (C) 1996-2013 Free Software Foundation, Inc.
# Originally written by Fran,cois Pinard , 1996.
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2, or (at your option)
# any later version.
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
# You should have received a copy of the GNU General Public License
-# along with this program. If not, see .
+# along with this program. If not, see .
# As a special exception to the GNU General Public License, if you
# distribute this file as part of a program that contains a
# configuration script generated by Autoconf, you may include it under
# the same distribution terms that you use for the rest of that program.
if test $# -eq 0; then
echo 1>&2 "Try '$0 --help' for more information"
exit 1
fi
case $1 in
--is-lightweight)
# Used by our autoconf macros to check whether the available missing
# script is modern enough.
exit 0
;;
--run)
# Back-compat with the calling convention used by older automake.
shift
;;
-h|--h|--he|--hel|--help)
echo "\
$0 [OPTION]... PROGRAM [ARGUMENT]...
Run 'PROGRAM [ARGUMENT]...', returning a proper advice when this fails due
to PROGRAM being missing or too old.
Options:
-h, --help display this help and exit
-v, --version output version information and exit
Supported PROGRAM values:
aclocal autoconf autoheader autom4te automake makeinfo
bison yacc flex lex help2man
Version suffixes to PROGRAM as well as the prefixes 'gnu-', 'gnu', and
'g' are ignored when checking the name.
Send bug reports to ."
exit $?
;;
-v|--v|--ve|--ver|--vers|--versi|--versio|--version)
echo "missing $scriptversion (GNU Automake)"
exit $?
;;
-*)
echo 1>&2 "$0: unknown '$1' option"
echo 1>&2 "Try '$0 --help' for more information"
exit 1
;;
esac
# Run the given program, remember its exit status.
"$@"; st=$?
# If it succeeded, we are done.
test $st -eq 0 && exit 0
# Also exit now if we it failed (or wasn't found), and '--version' was
# passed; such an option is passed most likely to detect whether the
# program is present and works.
case $2 in --version|--help) exit $st;; esac
# Exit code 63 means version mismatch. This often happens when the user
# tries to use an ancient version of a tool on a file that requires a
# minimum version.
if test $st -eq 63; then
msg="probably too old"
elif test $st -eq 127; then
# Program was missing.
msg="missing on your system"
else
# Program was found and executed, but failed. Give up.
exit $st
fi
-perl_URL=http://www.perl.org/
+perl_URL=https://www.perl.org/
flex_URL=http://flex.sourceforge.net/
-gnu_software_URL=http://www.gnu.org/software
+gnu_software_URL=https://www.gnu.org/software
program_details ()
{
case $1 in
aclocal|automake)
echo "The '$1' program is part of the GNU Automake package:"
echo "<$gnu_software_URL/automake>"
echo "It also requires GNU Autoconf, GNU m4 and Perl in order to run:"
echo "<$gnu_software_URL/autoconf>"
echo "<$gnu_software_URL/m4/>"
echo "<$perl_URL>"
;;
autoconf|autom4te|autoheader)
echo "The '$1' program is part of the GNU Autoconf package:"
echo "<$gnu_software_URL/autoconf/>"
echo "It also requires GNU m4 and Perl in order to run:"
echo "<$gnu_software_URL/m4/>"
echo "<$perl_URL>"
;;
esac
}
give_advice ()
{
# Normalize program name to check for.
normalized_program=`echo "$1" | sed '
s/^gnu-//; t
s/^gnu//; t
s/^g//; t'`
printf '%s\n' "'$1' is $msg."
configure_deps="'configure.ac' or m4 files included by 'configure.ac'"
case $normalized_program in
autoconf*)
echo "You should only need it if you modified 'configure.ac',"
echo "or m4 files included by it."
program_details 'autoconf'
;;
autoheader*)
echo "You should only need it if you modified 'acconfig.h' or"
echo "$configure_deps."
program_details 'autoheader'
;;
automake*)
echo "You should only need it if you modified 'Makefile.am' or"
echo "$configure_deps."
program_details 'automake'
;;
aclocal*)
echo "You should only need it if you modified 'acinclude.m4' or"
echo "$configure_deps."
program_details 'aclocal'
;;
autom4te*)
echo "You might have modified some maintainer files that require"
echo "the 'autom4te' program to be rebuilt."
program_details 'autom4te'
;;
bison*|yacc*)
echo "You should only need it if you modified a '.y' file."
echo "You may want to install the GNU Bison package:"
echo "<$gnu_software_URL/bison/>"
;;
lex*|flex*)
echo "You should only need it if you modified a '.l' file."
echo "You may want to install the Fast Lexical Analyzer package:"
echo "<$flex_URL>"
;;
help2man*)
echo "You should only need it if you modified a dependency" \
"of a man page."
echo "You may want to install the GNU Help2man package:"
echo "<$gnu_software_URL/help2man/>"
;;
makeinfo*)
echo "You should only need it if you modified a '.texi' file, or"
echo "any other file indirectly affecting the aspect of the manual."
echo "You might want to install the Texinfo package:"
echo "<$gnu_software_URL/texinfo/>"
echo "The spurious makeinfo call might also be the consequence of"
echo "using a buggy 'make' (AIX, DU, IRIX), in which case you might"
echo "want to install GNU make:"
echo "<$gnu_software_URL/make/>"
;;
*)
echo "You might have modified some files without having the proper"
echo "tools for further handling them. Check the 'README' file, it"
echo "often tells you about the needed prerequisites for installing"
echo "this package. You may also peek at any GNU archive site, in"
echo "case some other package contains this missing '$1' program."
;;
esac
}
give_advice "$1" | sed -e '1s/^/WARNING: /' \
-e '2,$s/^/ /' >&2
# Propagate the correct exit status (expected to be 127 for a program
# not found, 63 for a program that failed due to version mismatch).
exit $st
# Local variables:
# eval: (add-hook 'write-file-hooks 'time-stamp)
# time-stamp-start: "scriptversion="
# time-stamp-format: "%:y-%02m-%02d.%02H"
# time-stamp-time-zone: "UTC"
# time-stamp-end: "; # UTC"
# End:
diff --git a/build-aux/texinfo.tex b/build-aux/texinfo.tex
index a181898..b1f0d2e 100644
--- a/build-aux/texinfo.tex
+++ b/build-aux/texinfo.tex
@@ -1,8638 +1,8638 @@
% texinfo.tex -- TeX macros to handle Texinfo files.
%
% Load plain if necessary, i.e., if running under initex.
\expandafter\ifx\csname fmtname\endcsname\relax\input plain\fi
%
\def\texinfoversion{2007-05-03.09}
%
% Copyright (C) 1985, 1986, 1988, 1990, 1991, 1992, 1993, 1994, 1995,
% 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006,
% 2007 Free Software Foundation, Inc.
%
% This texinfo.tex file is free software; you can redistribute it and/or
% modify it under the terms of the GNU General Public License as
% published by the Free Software Foundation; either version 3, or (at
% your option) any later version.
%
% This texinfo.tex file is distributed in the hope that it will be
% useful, but WITHOUT ANY WARRANTY; without even the implied warranty
% of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
% General Public License for more details.
%
% You should have received a copy of the GNU General Public License
% along with this texinfo.tex file; see the file COPYING. If not,
-% see .
+% see .
%
% As a special exception, when this file is read by TeX when processing
% a Texinfo source document, you may use the result without
% restriction. (This has been our intent since Texinfo was invented.)
%
% Please try the latest version of texinfo.tex before submitting bug
% reports; you can get the latest version from:
-% http://www.gnu.org/software/texinfo/ (the Texinfo home page), or
+% https://www.gnu.org/software/texinfo/ (the Texinfo home page), or
% ftp://tug.org/tex/texinfo.tex
-% (and all CTAN mirrors, see http://www.ctan.org).
+% (and all CTAN mirrors, see https://www.ctan.org).
% The texinfo.tex in any given distribution could well be out
% of date, so if that's what you're using, please check.
%
% Send bug reports to bug-texinfo@gnu.org. Please include including a
% complete document in each bug report with which we can reproduce the
% problem. Patches are, of course, greatly appreciated.
%
% To process a Texinfo manual with TeX, it's most reliable to use the
% texi2dvi shell script that comes with the distribution. For a simple
% manual foo.texi, however, you can get away with this:
% tex foo.texi
% texindex foo.??
% tex foo.texi
% tex foo.texi
% dvips foo.dvi -o # or whatever; this makes foo.ps.
% The extra TeX runs get the cross-reference information correct.
% Sometimes one run after texindex suffices, and sometimes you need more
% than two; texi2dvi does it as many times as necessary.
%
% It is possible to adapt texinfo.tex for other languages, to some
% extent. You can get the existing language-specific files from the
% full Texinfo distribution.
%
-% The GNU Texinfo home page is http://www.gnu.org/software/texinfo.
+% The GNU Texinfo home page is https://www.gnu.org/software/texinfo.
\message{Loading texinfo [version \texinfoversion]:}
% If in a .fmt file, print the version number
% and turn on active characters that we couldn't do earlier because
% they might have appeared in the input file name.
\everyjob{\message{[Texinfo version \texinfoversion]}%
\catcode`+=\active \catcode`\_=\active}
\chardef\other=12
% We never want plain's \outer definition of \+ in Texinfo.
% For @tex, we can use \tabalign.
\let\+ = \relax
% Save some plain tex macros whose names we will redefine.
\let\ptexb=\b
\let\ptexbullet=\bullet
\let\ptexc=\c
\let\ptexcomma=\,
\let\ptexdot=\.
\let\ptexdots=\dots
\let\ptexend=\end
\let\ptexequiv=\equiv
\let\ptexexclam=\!
\let\ptexfootnote=\footnote
\let\ptexgtr=>
\let\ptexhat=^
\let\ptexi=\i
\let\ptexindent=\indent
\let\ptexinsert=\insert
\let\ptexlbrace=\{
\let\ptexless=<
\let\ptexnewwrite\newwrite
\let\ptexnoindent=\noindent
\let\ptexplus=+
\let\ptexrbrace=\}
\let\ptexslash=\/
\let\ptexstar=\*
\let\ptext=\t
% If this character appears in an error message or help string, it
% starts a new line in the output.
\newlinechar = `^^J
% Use TeX 3.0's \inputlineno to get the line number, for better error
% messages, but if we're using an old version of TeX, don't do anything.
%
\ifx\inputlineno\thisisundefined
\let\linenumber = \empty % Pre-3.0.
\else
\def\linenumber{l.\the\inputlineno:\space}
\fi
% Set up fixed words for English if not already set.
\ifx\putwordAppendix\undefined \gdef\putwordAppendix{Appendix}\fi
\ifx\putwordChapter\undefined \gdef\putwordChapter{Chapter}\fi
\ifx\putwordfile\undefined \gdef\putwordfile{file}\fi
\ifx\putwordin\undefined \gdef\putwordin{in}\fi
\ifx\putwordIndexIsEmpty\undefined \gdef\putwordIndexIsEmpty{(Index is empty)}\fi
\ifx\putwordIndexNonexistent\undefined \gdef\putwordIndexNonexistent{(Index is nonexistent)}\fi
\ifx\putwordInfo\undefined \gdef\putwordInfo{Info}\fi
\ifx\putwordInstanceVariableof\undefined \gdef\putwordInstanceVariableof{Instance Variable of}\fi
\ifx\putwordMethodon\undefined \gdef\putwordMethodon{Method on}\fi
\ifx\putwordNoTitle\undefined \gdef\putwordNoTitle{No Title}\fi
\ifx\putwordof\undefined \gdef\putwordof{of}\fi
\ifx\putwordon\undefined \gdef\putwordon{on}\fi
\ifx\putwordpage\undefined \gdef\putwordpage{page}\fi
\ifx\putwordsection\undefined \gdef\putwordsection{section}\fi
\ifx\putwordSection\undefined \gdef\putwordSection{Section}\fi
\ifx\putwordsee\undefined \gdef\putwordsee{see}\fi
\ifx\putwordSee\undefined \gdef\putwordSee{See}\fi
\ifx\putwordShortTOC\undefined \gdef\putwordShortTOC{Short Contents}\fi
\ifx\putwordTOC\undefined \gdef\putwordTOC{Table of Contents}\fi
%
\ifx\putwordMJan\undefined \gdef\putwordMJan{January}\fi
\ifx\putwordMFeb\undefined \gdef\putwordMFeb{February}\fi
\ifx\putwordMMar\undefined \gdef\putwordMMar{March}\fi
\ifx\putwordMApr\undefined \gdef\putwordMApr{April}\fi
\ifx\putwordMMay\undefined \gdef\putwordMMay{May}\fi
\ifx\putwordMJun\undefined \gdef\putwordMJun{June}\fi
\ifx\putwordMJul\undefined \gdef\putwordMJul{July}\fi
\ifx\putwordMAug\undefined \gdef\putwordMAug{August}\fi
\ifx\putwordMSep\undefined \gdef\putwordMSep{September}\fi
\ifx\putwordMOct\undefined \gdef\putwordMOct{October}\fi
\ifx\putwordMNov\undefined \gdef\putwordMNov{November}\fi
\ifx\putwordMDec\undefined \gdef\putwordMDec{December}\fi
%
\ifx\putwordDefmac\undefined \gdef\putwordDefmac{Macro}\fi
\ifx\putwordDefspec\undefined \gdef\putwordDefspec{Special Form}\fi
\ifx\putwordDefvar\undefined \gdef\putwordDefvar{Variable}\fi
\ifx\putwordDefopt\undefined \gdef\putwordDefopt{User Option}\fi
\ifx\putwordDeffunc\undefined \gdef\putwordDeffunc{Function}\fi
% Since the category of space is not known, we have to be careful.
\chardef\spacecat = 10
\def\spaceisspace{\catcode`\ =\spacecat}
% sometimes characters are active, so we need control sequences.
\chardef\colonChar = `\:
\chardef\commaChar = `\,
\chardef\dashChar = `\-
\chardef\dotChar = `\.
\chardef\exclamChar= `\!
\chardef\lquoteChar= `\`
\chardef\questChar = `\?
\chardef\rquoteChar= `\'
\chardef\semiChar = `\;
\chardef\underChar = `\_
% Ignore a token.
%
\def\gobble#1{}
% The following is used inside several \edef's.
\def\makecsname#1{\expandafter\noexpand\csname#1\endcsname}
% Hyphenation fixes.
\hyphenation{
Flor-i-da Ghost-script Ghost-view Mac-OS Post-Script
ap-pen-dix bit-map bit-maps
data-base data-bases eshell fall-ing half-way long-est man-u-script
man-u-scripts mini-buf-fer mini-buf-fers over-view par-a-digm
par-a-digms rath-er rec-tan-gu-lar ro-bot-ics se-vere-ly set-up spa-ces
spell-ing spell-ings
stand-alone strong-est time-stamp time-stamps which-ever white-space
wide-spread wrap-around
}
% Margin to add to right of even pages, to left of odd pages.
\newdimen\bindingoffset
\newdimen\normaloffset
\newdimen\pagewidth \newdimen\pageheight
% For a final copy, take out the rectangles
% that mark overfull boxes (in case you have decided
% that the text looks ok even though it passes the margin).
%
\def\finalout{\overfullrule=0pt}
% @| inserts a changebar to the left of the current line. It should
% surround any changed text. This approach does *not* work if the
% change spans more than two lines of output. To handle that, we would
% have adopt a much more difficult approach (putting marks into the main
% vertical list for the beginning and end of each change).
%
\def\|{%
% \vadjust can only be used in horizontal mode.
\leavevmode
%
% Append this vertical mode material after the current line in the output.
\vadjust{%
% We want to insert a rule with the height and depth of the current
% leading; that is exactly what \strutbox is supposed to record.
\vskip-\baselineskip
%
% \vadjust-items are inserted at the left edge of the type. So
% the \llap here moves out into the left-hand margin.
\llap{%
%
% For a thicker or thinner bar, change the `1pt'.
\vrule height\baselineskip width1pt
%
% This is the space between the bar and the text.
\hskip 12pt
}%
}%
}
% Sometimes it is convenient to have everything in the transcript file
% and nothing on the terminal. We don't just call \tracingall here,
% since that produces some useless output on the terminal. We also make
% some effort to order the tracing commands to reduce output in the log
% file; cf. trace.sty in LaTeX.
%
\def\gloggingall{\begingroup \globaldefs = 1 \loggingall \endgroup}%
\def\loggingall{%
\tracingstats2
\tracingpages1
\tracinglostchars2 % 2 gives us more in etex
\tracingparagraphs1
\tracingoutput1
\tracingmacros2
\tracingrestores1
\showboxbreadth\maxdimen \showboxdepth\maxdimen
\ifx\eTeXversion\undefined\else % etex gives us more logging
\tracingscantokens1
\tracingifs1
\tracinggroups1
\tracingnesting2
\tracingassigns1
\fi
\tracingcommands3 % 3 gives us more in etex
\errorcontextlines16
}%
% add check for \lastpenalty to plain's definitions. If the last thing
% we did was a \nobreak, we don't want to insert more space.
%
\def\smallbreak{\ifnum\lastpenalty<10000\par\ifdim\lastskip<\smallskipamount
\removelastskip\penalty-50\smallskip\fi\fi}
\def\medbreak{\ifnum\lastpenalty<10000\par\ifdim\lastskip<\medskipamount
\removelastskip\penalty-100\medskip\fi\fi}
\def\bigbreak{\ifnum\lastpenalty<10000\par\ifdim\lastskip<\bigskipamount
\removelastskip\penalty-200\bigskip\fi\fi}
% For @cropmarks command.
% Do @cropmarks to get crop marks.
%
\newif\ifcropmarks
\let\cropmarks = \cropmarkstrue
%
% Dimensions to add cropmarks at corners.
% Added by P. A. MacKay, 12 Nov. 1986
%
\newdimen\outerhsize \newdimen\outervsize % set by the paper size routines
\newdimen\cornerlong \cornerlong=1pc
\newdimen\cornerthick \cornerthick=.3pt
\newdimen\topandbottommargin \topandbottommargin=.75in
% Main output routine.
\chardef\PAGE = 255
\output = {\onepageout{\pagecontents\PAGE}}
\newbox\headlinebox
\newbox\footlinebox
% \onepageout takes a vbox as an argument. Note that \pagecontents
% does insertions, but you have to call it yourself.
\def\onepageout#1{%
\ifcropmarks \hoffset=0pt \else \hoffset=\normaloffset \fi
%
\ifodd\pageno \advance\hoffset by \bindingoffset
\else \advance\hoffset by -\bindingoffset\fi
%
% Do this outside of the \shipout so @code etc. will be expanded in
% the headline as they should be, not taken literally (outputting ''code).
\setbox\headlinebox = \vbox{\let\hsize=\pagewidth \makeheadline}%
\setbox\footlinebox = \vbox{\let\hsize=\pagewidth \makefootline}%
%
{%
% Have to do this stuff outside the \shipout because we want it to
% take effect in \write's, yet the group defined by the \vbox ends
% before the \shipout runs.
%
\indexdummies % don't expand commands in the output.
\normalturnoffactive % \ in index entries must not stay \, e.g., if
% the page break happens to be in the middle of an example.
% We don't want .vr (or whatever) entries like this:
% \entry{{\tt \indexbackslash }acronym}{32}{\code {\acronym}}
% "\acronym" won't work when it's read back in;
% it needs to be
% {\code {{\tt \backslashcurfont }acronym}
\shipout\vbox{%
% Do this early so pdf references go to the beginning of the page.
\ifpdfmakepagedest \pdfdest name{\the\pageno} xyz\fi
%
\ifcropmarks \vbox to \outervsize\bgroup
\hsize = \outerhsize
\vskip-\topandbottommargin
\vtop to0pt{%
\line{\ewtop\hfil\ewtop}%
\nointerlineskip
\line{%
\vbox{\moveleft\cornerthick\nstop}%
\hfill
\vbox{\moveright\cornerthick\nstop}%
}%
\vss}%
\vskip\topandbottommargin
\line\bgroup
\hfil % center the page within the outer (page) hsize.
\ifodd\pageno\hskip\bindingoffset\fi
\vbox\bgroup
\fi
%
\unvbox\headlinebox
\pagebody{#1}%
\ifdim\ht\footlinebox > 0pt
% Only leave this space if the footline is nonempty.
% (We lessened \vsize for it in \oddfootingyyy.)
% The \baselineskip=24pt in plain's \makefootline has no effect.
\vskip 24pt
\unvbox\footlinebox
\fi
%
\ifcropmarks
\egroup % end of \vbox\bgroup
\hfil\egroup % end of (centering) \line\bgroup
\vskip\topandbottommargin plus1fill minus1fill
\boxmaxdepth = \cornerthick
\vbox to0pt{\vss
\line{%
\vbox{\moveleft\cornerthick\nsbot}%
\hfill
\vbox{\moveright\cornerthick\nsbot}%
}%
\nointerlineskip
\line{\ewbot\hfil\ewbot}%
}%
\egroup % \vbox from first cropmarks clause
\fi
}% end of \shipout\vbox
}% end of group with \indexdummies
\advancepageno
\ifnum\outputpenalty>-20000 \else\dosupereject\fi
}
\newinsert\margin \dimen\margin=\maxdimen
\def\pagebody#1{\vbox to\pageheight{\boxmaxdepth=\maxdepth #1}}
{\catcode`\@ =11
\gdef\pagecontents#1{\ifvoid\topins\else\unvbox\topins\fi
% marginal hacks, juha@viisa.uucp (Juha Takala)
\ifvoid\margin\else % marginal info is present
\rlap{\kern\hsize\vbox to\z@{\kern1pt\box\margin \vss}}\fi
\dimen@=\dp#1 \unvbox#1
\ifvoid\footins\else\vskip\skip\footins\footnoterule \unvbox\footins\fi
\ifr@ggedbottom \kern-\dimen@ \vfil \fi}
}
% Here are the rules for the cropmarks. Note that they are
% offset so that the space between them is truly \outerhsize or \outervsize
% (P. A. MacKay, 12 November, 1986)
%
\def\ewtop{\vrule height\cornerthick depth0pt width\cornerlong}
\def\nstop{\vbox
{\hrule height\cornerthick depth\cornerlong width\cornerthick}}
\def\ewbot{\vrule height0pt depth\cornerthick width\cornerlong}
\def\nsbot{\vbox
{\hrule height\cornerlong depth\cornerthick width\cornerthick}}
% Parse an argument, then pass it to #1. The argument is the rest of
% the input line (except we remove a trailing comment). #1 should be a
% macro which expects an ordinary undelimited TeX argument.
%
\def\parsearg{\parseargusing{}}
\def\parseargusing#1#2{%
\def\argtorun{#2}%
\begingroup
\obeylines
\spaceisspace
#1%
\parseargline\empty% Insert the \empty token, see \finishparsearg below.
}
{\obeylines %
\gdef\parseargline#1^^M{%
\endgroup % End of the group started in \parsearg.
\argremovecomment #1\comment\ArgTerm%
}%
}
% First remove any @comment, then any @c comment.
\def\argremovecomment#1\comment#2\ArgTerm{\argremovec #1\c\ArgTerm}
\def\argremovec#1\c#2\ArgTerm{\argcheckspaces#1\^^M\ArgTerm}
% Each occurence of `\^^M' or `\^^M' is replaced by a single space.
%
% \argremovec might leave us with trailing space, e.g.,
% @end itemize @c foo
% This space token undergoes the same procedure and is eventually removed
% by \finishparsearg.
%
\def\argcheckspaces#1\^^M{\argcheckspacesX#1\^^M \^^M}
\def\argcheckspacesX#1 \^^M{\argcheckspacesY#1\^^M}
\def\argcheckspacesY#1\^^M#2\^^M#3\ArgTerm{%
\def\temp{#3}%
\ifx\temp\empty
% Do not use \next, perhaps the caller of \parsearg uses it; reuse \temp:
\let\temp\finishparsearg
\else
\let\temp\argcheckspaces
\fi
% Put the space token in:
\temp#1 #3\ArgTerm
}
% If a _delimited_ argument is enclosed in braces, they get stripped; so
% to get _exactly_ the rest of the line, we had to prevent such situation.
% We prepended an \empty token at the very beginning and we expand it now,
% just before passing the control to \argtorun.
% (Similarily, we have to think about #3 of \argcheckspacesY above: it is
% either the null string, or it ends with \^^M---thus there is no danger
% that a pair of braces would be stripped.
%
% But first, we have to remove the trailing space token.
%
\def\finishparsearg#1 \ArgTerm{\expandafter\argtorun\expandafter{#1}}
% \parseargdef\foo{...}
% is roughly equivalent to
% \def\foo{\parsearg\Xfoo}
% \def\Xfoo#1{...}
%
% Actually, I use \csname\string\foo\endcsname, ie. \\foo, as it is my
% favourite TeX trick. --kasal, 16nov03
\def\parseargdef#1{%
\expandafter \doparseargdef \csname\string#1\endcsname #1%
}
\def\doparseargdef#1#2{%
\def#2{\parsearg#1}%
\def#1##1%
}
% Several utility definitions with active space:
{
\obeyspaces
\gdef\obeyedspace{ }
% Make each space character in the input produce a normal interword
% space in the output. Don't allow a line break at this space, as this
% is used only in environments like @example, where each line of input
% should produce a line of output anyway.
%
\gdef\sepspaces{\obeyspaces\let =\tie}
% If an index command is used in an @example environment, any spaces
% therein should become regular spaces in the raw index file, not the
% expansion of \tie (\leavevmode \penalty \@M \ ).
\gdef\unsepspaces{\let =\space}
}
\def\flushcr{\ifx\par\lisppar \def\next##1{}\else \let\next=\relax \fi \next}
% Define the framework for environments in texinfo.tex. It's used like this:
%
% \envdef\foo{...}
% \def\Efoo{...}
%
% It's the responsibility of \envdef to insert \begingroup before the
% actual body; @end closes the group after calling \Efoo. \envdef also
% defines \thisenv, so the current environment is known; @end checks
% whether the environment name matches. The \checkenv macro can also be
% used to check whether the current environment is the one expected.
%
% Non-false conditionals (@iftex, @ifset) don't fit into this, so they
% are not treated as enviroments; they don't open a group. (The
% implementation of @end takes care not to call \endgroup in this
% special case.)
% At runtime, environments start with this:
\def\startenvironment#1{\begingroup\def\thisenv{#1}}
% initialize
\let\thisenv\empty
% ... but they get defined via ``\envdef\foo{...}'':
\long\def\envdef#1#2{\def#1{\startenvironment#1#2}}
\def\envparseargdef#1#2{\parseargdef#1{\startenvironment#1#2}}
% Check whether we're in the right environment:
\def\checkenv#1{%
\def\temp{#1}%
\ifx\thisenv\temp
\else
\badenverr
\fi
}
% Evironment mismatch, #1 expected:
\def\badenverr{%
\errhelp = \EMsimple
\errmessage{This command can appear only \inenvironment\temp,
not \inenvironment\thisenv}%
}
\def\inenvironment#1{%
\ifx#1\empty
out of any environment%
\else
in environment \expandafter\string#1%
\fi
}
% @end foo executes the definition of \Efoo.
% But first, it executes a specialized version of \checkenv
%
\parseargdef\end{%
\if 1\csname iscond.#1\endcsname
\else
% The general wording of \badenverr may not be ideal, but... --kasal, 06nov03
\expandafter\checkenv\csname#1\endcsname
\csname E#1\endcsname
\endgroup
\fi
}
\newhelp\EMsimple{Press RETURN to continue.}
%% Simple single-character @ commands
% @@ prints an @
% Kludge this until the fonts are right (grr).
\def\@{{\tt\char64}}
% This is turned off because it was never documented
% and you can use @w{...} around a quote to suppress ligatures.
%% Define @` and @' to be the same as ` and '
%% but suppressing ligatures.
%\def\`{{`}}
%\def\'{{'}}
% Used to generate quoted braces.
\def\mylbrace {{\tt\char123}}
\def\myrbrace {{\tt\char125}}
\let\{=\mylbrace
\let\}=\myrbrace
\begingroup
% Definitions to produce \{ and \} commands for indices,
% and @{ and @} for the aux/toc files.
\catcode`\{ = \other \catcode`\} = \other
\catcode`\[ = 1 \catcode`\] = 2
\catcode`\! = 0 \catcode`\\ = \other
!gdef!lbracecmd[\{]%
!gdef!rbracecmd[\}]%
!gdef!lbraceatcmd[@{]%
!gdef!rbraceatcmd[@}]%
!endgroup
% @comma{} to avoid , parsing problems.
\let\comma = ,
% Accents: @, @dotaccent @ringaccent @ubaraccent @udotaccent
% Others are defined by plain TeX: @` @' @" @^ @~ @= @u @v @H.
\let\, = \c
\let\dotaccent = \.
\def\ringaccent#1{{\accent23 #1}}
\let\tieaccent = \t
\let\ubaraccent = \b
\let\udotaccent = \d
% Other special characters: @questiondown @exclamdown @ordf @ordm
% Plain TeX defines: @AA @AE @O @OE @L (plus lowercase versions) @ss.
\def\questiondown{?`}
\def\exclamdown{!`}
\def\ordf{\leavevmode\raise1ex\hbox{\selectfonts\lllsize \underbar{a}}}
\def\ordm{\leavevmode\raise1ex\hbox{\selectfonts\lllsize \underbar{o}}}
% Dotless i and dotless j, used for accents.
\def\imacro{i}
\def\jmacro{j}
\def\dotless#1{%
\def\temp{#1}%
\ifx\temp\imacro \ptexi
\else\ifx\temp\jmacro \j
\else \errmessage{@dotless can be used only with i or j}%
\fi\fi
}
% The \TeX{} logo, as in plain, but resetting the spacing so that a
% period following counts as ending a sentence. (Idea found in latex.)
%
\edef\TeX{\TeX \spacefactor=1000 }
% @LaTeX{} logo. Not quite the same results as the definition in
% latex.ltx, since we use a different font for the raised A; it's most
% convenient for us to use an explicitly smaller font, rather than using
% the \scriptstyle font (since we don't reset \scriptstyle and
% \scriptscriptstyle).
%
\def\LaTeX{%
L\kern-.36em
{\setbox0=\hbox{T}%
\vbox to \ht0{\hbox{\selectfonts\lllsize A}\vss}}%
\kern-.15em
\TeX
}
% Be sure we're in horizontal mode when doing a tie, since we make space
% equivalent to this in @example-like environments. Otherwise, a space
% at the beginning of a line will start with \penalty -- and
% since \penalty is valid in vertical mode, we'd end up putting the
% penalty on the vertical list instead of in the new paragraph.
{\catcode`@ = 11
% Avoid using \@M directly, because that causes trouble
% if the definition is written into an index file.
\global\let\tiepenalty = \@M
\gdef\tie{\leavevmode\penalty\tiepenalty\ }
}
% @: forces normal size whitespace following.
\def\:{\spacefactor=1000 }
% @* forces a line break.
\def\*{\hfil\break\hbox{}\ignorespaces}
% @/ allows a line break.
\let\/=\allowbreak
% @. is an end-of-sentence period.
\def\.{.\spacefactor=\endofsentencespacefactor\space}
% @! is an end-of-sentence bang.
\def\!{!\spacefactor=\endofsentencespacefactor\space}
% @? is an end-of-sentence query.
\def\?{?\spacefactor=\endofsentencespacefactor\space}
% @frenchspacing on|off says whether to put extra space after punctuation.
%
\def\onword{on}
\def\offword{off}
%
\parseargdef\frenchspacing{%
\def\temp{#1}%
\ifx\temp\onword \plainfrenchspacing
\else\ifx\temp\offword \plainnonfrenchspacing
\else
\errhelp = \EMsimple
\errmessage{Unknown @frenchspacing option `\temp', must be on/off}%
\fi\fi
}
% @w prevents a word break. Without the \leavevmode, @w at the
% beginning of a paragraph, when TeX is still in vertical mode, would
% produce a whole line of output instead of starting the paragraph.
\def\w#1{\leavevmode\hbox{#1}}
% @group ... @end group forces ... to be all on one page, by enclosing
% it in a TeX vbox. We use \vtop instead of \vbox to construct the box
% to keep its height that of a normal line. According to the rules for
% \topskip (p.114 of the TeXbook), the glue inserted is
% max (\topskip - \ht (first item), 0). If that height is large,
% therefore, no glue is inserted, and the space between the headline and
% the text is small, which looks bad.
%
% Another complication is that the group might be very large. This can
% cause the glue on the previous page to be unduly stretched, because it
% does not have much material. In this case, it's better to add an
% explicit \vfill so that the extra space is at the bottom. The
% threshold for doing this is if the group is more than \vfilllimit
% percent of a page (\vfilllimit can be changed inside of @tex).
%
\newbox\groupbox
\def\vfilllimit{0.7}
%
\envdef\group{%
\ifnum\catcode`\^^M=\active \else
\errhelp = \groupinvalidhelp
\errmessage{@group invalid in context where filling is enabled}%
\fi
\startsavinginserts
%
\setbox\groupbox = \vtop\bgroup
% Do @comment since we are called inside an environment such as
% @example, where each end-of-line in the input causes an
% end-of-line in the output. We don't want the end-of-line after
% the `@group' to put extra space in the output. Since @group
% should appear on a line by itself (according to the Texinfo
% manual), we don't worry about eating any user text.
\comment
}
%
% The \vtop produces a box with normal height and large depth; thus, TeX puts
% \baselineskip glue before it, and (when the next line of text is done)
% \lineskip glue after it. Thus, space below is not quite equal to space
% above. But it's pretty close.
\def\Egroup{%
% To get correct interline space between the last line of the group
% and the first line afterwards, we have to propagate \prevdepth.
\endgraf % Not \par, as it may have been set to \lisppar.
\global\dimen1 = \prevdepth
\egroup % End the \vtop.
% \dimen0 is the vertical size of the group's box.
\dimen0 = \ht\groupbox \advance\dimen0 by \dp\groupbox
% \dimen2 is how much space is left on the page (more or less).
\dimen2 = \pageheight \advance\dimen2 by -\pagetotal
% if the group doesn't fit on the current page, and it's a big big
% group, force a page break.
\ifdim \dimen0 > \dimen2
\ifdim \pagetotal < \vfilllimit\pageheight
\page
\fi
\fi
\box\groupbox
\prevdepth = \dimen1
\checkinserts
}
%
% TeX puts in an \escapechar (i.e., `@') at the beginning of the help
% message, so this ends up printing `@group can only ...'.
%
\newhelp\groupinvalidhelp{%
group can only be used in environments such as @example,^^J%
where each line of input produces a line of output.}
% @need space-in-mils
% forces a page break if there is not space-in-mils remaining.
\newdimen\mil \mil=0.001in
% Old definition--didn't work.
%\parseargdef\need{\par %
%% This method tries to make TeX break the page naturally
%% if the depth of the box does not fit.
%{\baselineskip=0pt%
%\vtop to #1\mil{\vfil}\kern -#1\mil\nobreak
%\prevdepth=-1000pt
%}}
\parseargdef\need{%
% Ensure vertical mode, so we don't make a big box in the middle of a
% paragraph.
\par
%
% If the @need value is less than one line space, it's useless.
\dimen0 = #1\mil
\dimen2 = \ht\strutbox
\advance\dimen2 by \dp\strutbox
\ifdim\dimen0 > \dimen2
%
% Do a \strut just to make the height of this box be normal, so the
% normal leading is inserted relative to the preceding line.
% And a page break here is fine.
\vtop to #1\mil{\strut\vfil}%
%
% TeX does not even consider page breaks if a penalty added to the
% main vertical list is 10000 or more. But in order to see if the
% empty box we just added fits on the page, we must make it consider
% page breaks. On the other hand, we don't want to actually break the
% page after the empty box. So we use a penalty of 9999.
%
% There is an extremely small chance that TeX will actually break the
% page at this \penalty, if there are no other feasible breakpoints in
% sight. (If the user is using lots of big @group commands, which
% almost-but-not-quite fill up a page, TeX will have a hard time doing
% good page breaking, for example.) However, I could not construct an
% example where a page broke at this \penalty; if it happens in a real
% document, then we can reconsider our strategy.
\penalty9999
%
% Back up by the size of the box, whether we did a page break or not.
\kern -#1\mil
%
% Do not allow a page break right after this kern.
\nobreak
\fi
}
% @br forces paragraph break (and is undocumented).
\let\br = \par
% @page forces the start of a new page.
%
\def\page{\par\vfill\supereject}
% @exdent text....
% outputs text on separate line in roman font, starting at standard page margin
% This records the amount of indent in the innermost environment.
% That's how much \exdent should take out.
\newskip\exdentamount
% This defn is used inside fill environments such as @defun.
\parseargdef\exdent{\hfil\break\hbox{\kern -\exdentamount{\rm#1}}\hfil\break}
% This defn is used inside nofill environments such as @example.
\parseargdef\nofillexdent{{\advance \leftskip by -\exdentamount
\leftline{\hskip\leftskip{\rm#1}}}}
% @inmargin{WHICH}{TEXT} puts TEXT in the WHICH margin next to the current
% paragraph. For more general purposes, use the \margin insertion
% class. WHICH is `l' or `r'.
%
\newskip\inmarginspacing \inmarginspacing=1cm
\def\strutdepth{\dp\strutbox}
%
\def\doinmargin#1#2{\strut\vadjust{%
\nobreak
\kern-\strutdepth
\vtop to \strutdepth{%
\baselineskip=\strutdepth
\vss
% if you have multiple lines of stuff to put here, you'll need to
% make the vbox yourself of the appropriate size.
\ifx#1l%
\llap{\ignorespaces #2\hskip\inmarginspacing}%
\else
\rlap{\hskip\hsize \hskip\inmarginspacing \ignorespaces #2}%
\fi
\null
}%
}}
\def\inleftmargin{\doinmargin l}
\def\inrightmargin{\doinmargin r}
%
% @inmargin{TEXT [, RIGHT-TEXT]}
% (if RIGHT-TEXT is given, use TEXT for left page, RIGHT-TEXT for right;
% else use TEXT for both).
%
\def\inmargin#1{\parseinmargin #1,,\finish}
\def\parseinmargin#1,#2,#3\finish{% not perfect, but better than nothing.
\setbox0 = \hbox{\ignorespaces #2}%
\ifdim\wd0 > 0pt
\def\lefttext{#1}% have both texts
\def\righttext{#2}%
\else
\def\lefttext{#1}% have only one text
\def\righttext{#1}%
\fi
%
\ifodd\pageno
\def\temp{\inrightmargin\righttext}% odd page -> outside is right margin
\else
\def\temp{\inleftmargin\lefttext}%
\fi
\temp
}
% @include file insert text of that file as input.
%
\def\include{\parseargusing\filenamecatcodes\includezzz}
\def\includezzz#1{%
\pushthisfilestack
\def\thisfile{#1}%
{%
\makevalueexpandable
\def\temp{\input #1 }%
\expandafter
}\temp
\popthisfilestack
}
\def\filenamecatcodes{%
\catcode`\\=\other
\catcode`~=\other
\catcode`^=\other
\catcode`_=\other
\catcode`|=\other
\catcode`<=\other
\catcode`>=\other
\catcode`+=\other
\catcode`-=\other
}
\def\pushthisfilestack{%
\expandafter\pushthisfilestackX\popthisfilestack\StackTerm
}
\def\pushthisfilestackX{%
\expandafter\pushthisfilestackY\thisfile\StackTerm
}
\def\pushthisfilestackY #1\StackTerm #2\StackTerm {%
\gdef\popthisfilestack{\gdef\thisfile{#1}\gdef\popthisfilestack{#2}}%
}
\def\popthisfilestack{\errthisfilestackempty}
\def\errthisfilestackempty{\errmessage{Internal error:
the stack of filenames is empty.}}
\def\thisfile{}
% @center line
% outputs that line, centered.
%
\parseargdef\center{%
\ifhmode
\let\next\centerH
\else
\let\next\centerV
\fi
\next{\hfil \ignorespaces#1\unskip \hfil}%
}
\def\centerH#1{%
{%
\hfil\break
\advance\hsize by -\leftskip
\advance\hsize by -\rightskip
\line{#1}%
\break
}%
}
\def\centerV#1{\line{\kern\leftskip #1\kern\rightskip}}
% @sp n outputs n lines of vertical space
\parseargdef\sp{\vskip #1\baselineskip}
% @comment ...line which is ignored...
% @c is the same as @comment
% @ignore ... @end ignore is another way to write a comment
\def\comment{\begingroup \catcode`\^^M=\other%
\catcode`\@=\other \catcode`\{=\other \catcode`\}=\other%
\commentxxx}
{\catcode`\^^M=\other \gdef\commentxxx#1^^M{\endgroup}}
\let\c=\comment
% @paragraphindent NCHARS
% We'll use ems for NCHARS, close enough.
% NCHARS can also be the word `asis' or `none'.
% We cannot feasibly implement @paragraphindent asis, though.
%
\def\asisword{asis} % no translation, these are keywords
\def\noneword{none}
%
\parseargdef\paragraphindent{%
\def\temp{#1}%
\ifx\temp\asisword
\else
\ifx\temp\noneword
\defaultparindent = 0pt
\else
\defaultparindent = #1em
\fi
\fi
\parindent = \defaultparindent
}
% @exampleindent NCHARS
% We'll use ems for NCHARS like @paragraphindent.
% It seems @exampleindent asis isn't necessary, but
% I preserve it to make it similar to @paragraphindent.
\parseargdef\exampleindent{%
\def\temp{#1}%
\ifx\temp\asisword
\else
\ifx\temp\noneword
\lispnarrowing = 0pt
\else
\lispnarrowing = #1em
\fi
\fi
}
% @firstparagraphindent WORD
% If WORD is `none', then suppress indentation of the first paragraph
% after a section heading. If WORD is `insert', then do indent at such
% paragraphs.
%
% The paragraph indentation is suppressed or not by calling
% \suppressfirstparagraphindent, which the sectioning commands do.
% We switch the definition of this back and forth according to WORD.
% By default, we suppress indentation.
%
\def\suppressfirstparagraphindent{\dosuppressfirstparagraphindent}
\def\insertword{insert}
%
\parseargdef\firstparagraphindent{%
\def\temp{#1}%
\ifx\temp\noneword
\let\suppressfirstparagraphindent = \dosuppressfirstparagraphindent
\else\ifx\temp\insertword
\let\suppressfirstparagraphindent = \relax
\else
\errhelp = \EMsimple
\errmessage{Unknown @firstparagraphindent option `\temp'}%
\fi\fi
}
% Here is how we actually suppress indentation. Redefine \everypar to
% \kern backwards by \parindent, and then reset itself to empty.
%
% We also make \indent itself not actually do anything until the next
% paragraph.
%
\gdef\dosuppressfirstparagraphindent{%
\gdef\indent{%
\restorefirstparagraphindent
\indent
}%
\gdef\noindent{%
\restorefirstparagraphindent
\noindent
}%
\global\everypar = {%
\kern -\parindent
\restorefirstparagraphindent
}%
}
\gdef\restorefirstparagraphindent{%
\global \let \indent = \ptexindent
\global \let \noindent = \ptexnoindent
\global \everypar = {}%
}
% @asis just yields its argument. Used with @table, for example.
%
\def\asis#1{#1}
% @math outputs its argument in math mode.
%
% One complication: _ usually means subscripts, but it could also mean
% an actual _ character, as in @math{@var{some_variable} + 1}. So make
% _ active, and distinguish by seeing if the current family is \slfam,
% which is what @var uses.
{
\catcode`\_ = \active
\gdef\mathunderscore{%
\catcode`\_=\active
\def_{\ifnum\fam=\slfam \_\else\sb\fi}%
}
}
% Another complication: we want \\ (and @\) to output a \ character.
% FYI, plain.tex uses \\ as a temporary control sequence (why?), but
% this is not advertised and we don't care. Texinfo does not
% otherwise define @\.
%
% The \mathchar is class=0=ordinary, family=7=ttfam, position=5C=\.
\def\mathbackslash{\ifnum\fam=\ttfam \mathchar"075C \else\backslash \fi}
%
\def\math{%
\tex
\mathunderscore
\let\\ = \mathbackslash
\mathactive
$\finishmath
}
\def\finishmath#1{#1$\endgroup} % Close the group opened by \tex.
% Some active characters (such as <) are spaced differently in math.
% We have to reset their definitions in case the @math was an argument
% to a command which sets the catcodes (such as @item or @section).
%
{
\catcode`^ = \active
\catcode`< = \active
\catcode`> = \active
\catcode`+ = \active
\gdef\mathactive{%
\let^ = \ptexhat
\let< = \ptexless
\let> = \ptexgtr
\let+ = \ptexplus
}
}
% @bullet and @minus need the same treatment as @math, just above.
\def\bullet{$\ptexbullet$}
\def\minus{$-$}
% @dots{} outputs an ellipsis using the current font.
% We do .5em per period so that it has the same spacing in the cm
% typewriter fonts as three actual period characters; on the other hand,
% in other typewriter fonts three periods are wider than 1.5em. So do
% whichever is larger.
%
\def\dots{%
\leavevmode
\setbox0=\hbox{...}% get width of three periods
\ifdim\wd0 > 1.5em
\dimen0 = \wd0
\else
\dimen0 = 1.5em
\fi
\hbox to \dimen0{%
\hskip 0pt plus.25fil
.\hskip 0pt plus1fil
.\hskip 0pt plus1fil
.\hskip 0pt plus.5fil
}%
}
% @enddots{} is an end-of-sentence ellipsis.
%
\def\enddots{%
\dots
\spacefactor=\endofsentencespacefactor
}
% @comma{} is so commas can be inserted into text without messing up
% Texinfo's parsing.
%
\let\comma = ,
% @refill is a no-op.
\let\refill=\relax
% If working on a large document in chapters, it is convenient to
% be able to disable indexing, cross-referencing, and contents, for test runs.
% This is done with @novalidate (before @setfilename).
%
\newif\iflinks \linkstrue % by default we want the aux files.
\let\novalidate = \linksfalse
% @setfilename is done at the beginning of every texinfo file.
% So open here the files we need to have open while reading the input.
% This makes it possible to make a .fmt file for texinfo.
\def\setfilename{%
\fixbackslash % Turn off hack to swallow `\input texinfo'.
\iflinks
\tryauxfile
% Open the new aux file. TeX will close it automatically at exit.
\immediate\openout\auxfile=\jobname.aux
\fi % \openindices needs to do some work in any case.
\openindices
\let\setfilename=\comment % Ignore extra @setfilename cmds.
%
% If texinfo.cnf is present on the system, read it.
% Useful for site-wide @afourpaper, etc.
\openin 1 texinfo.cnf
\ifeof 1 \else \input texinfo.cnf \fi
\closein 1
%
\comment % Ignore the actual filename.
}
% Called from \setfilename.
%
\def\openindices{%
\newindex{cp}%
\newcodeindex{fn}%
\newcodeindex{vr}%
\newcodeindex{tp}%
\newcodeindex{ky}%
\newcodeindex{pg}%
}
% @bye.
\outer\def\bye{\pagealignmacro\tracingstats=1\ptexend}
\message{pdf,}
% adobe `portable' document format
\newcount\tempnum
\newcount\lnkcount
\newtoks\filename
\newcount\filenamelength
\newcount\pgn
\newtoks\toksA
\newtoks\toksB
\newtoks\toksC
\newtoks\toksD
\newbox\boxA
\newcount\countA
\newif\ifpdf
\newif\ifpdfmakepagedest
% when pdftex is run in dvi mode, \pdfoutput is defined (so \pdfoutput=1
% can be set). So we test for \relax and 0 as well as \undefined,
% borrowed from ifpdf.sty.
\ifx\pdfoutput\undefined
\else
\ifx\pdfoutput\relax
\else
\ifcase\pdfoutput
\else
\pdftrue
\fi
\fi
\fi
% PDF uses PostScript string constants for the names of xref targets,
% for display in the outlines, and in other places. Thus, we have to
% double any backslashes. Otherwise, a name like "\node" will be
% interpreted as a newline (\n), followed by o, d, e. Not good.
-% http://www.ntg.nl/pipermail/ntg-pdftex/2004-July/000654.html
+% https://www.ntg.nl/pipermail/ntg-pdftex/2004-July/000654.html
% (and related messages, the final outcome is that it is up to the TeX
% user to double the backslashes and otherwise make the string valid, so
% that's what we do).
% double active backslashes.
%
{\catcode`\@=0 \catcode`\\=\active
@gdef@activebackslashdouble{%
@catcode`@\=@active
@let\=@doublebackslash}
}
% To handle parens, we must adopt a different approach, since parens are
% not active characters. hyperref.dtx (which has the same problem as
% us) handles it with this amazing macro to replace tokens, with minor
% changes for Texinfo. It is included here under the GPL by permission
% from the author, Heiko Oberdiek.
%
% #1 is the tokens to replace.
% #2 is the replacement.
% #3 is the control sequence with the string.
%
\def\HyPsdSubst#1#2#3{%
\def\HyPsdReplace##1#1##2\END{%
##1%
\ifx\\##2\\%
\else
#2%
\HyReturnAfterFi{%
\HyPsdReplace##2\END
}%
\fi
}%
\xdef#3{\expandafter\HyPsdReplace#3#1\END}%
}
\long\def\HyReturnAfterFi#1\fi{\fi#1}
% #1 is a control sequence in which to do the replacements.
\def\backslashparens#1{%
\xdef#1{#1}% redefine it as its expansion; the definition is simply
% \lastnode when called from \setref -> \pdfmkdest.
\HyPsdSubst{(}{\realbackslash(}{#1}%
\HyPsdSubst{)}{\realbackslash)}{#1}%
}
\newhelp\nopdfimagehelp{Texinfo supports .png, .jpg, .jpeg, and .pdf images
with PDF output, and none of those formats could be found. (.eps cannot
be supported due to the design of the PDF format; use regular TeX (DVI
output) for that.)}
\ifpdf
\input pdfcolor
\pdfcatalog{/PageMode /UseOutlines}
%
% #1 is image name, #2 width (might be empty/whitespace), #3 height (ditto).
\def\dopdfimage#1#2#3{%
\def\imagewidth{#2}\setbox0 = \hbox{\ignorespaces #2}%
\def\imageheight{#3}\setbox2 = \hbox{\ignorespaces #3}%
%
% pdftex (and the PDF format) support .png, .jpg, .pdf (among
% others). Let's try in that order.
\let\pdfimgext=\empty
\begingroup
\openin 1 #1.png \ifeof 1
\openin 1 #1.jpg \ifeof 1
\openin 1 #1.jpeg \ifeof 1
\openin 1 #1.JPG \ifeof 1
\openin 1 #1.pdf \ifeof 1
\errhelp = \nopdfimagehelp
\errmessage{Could not find image file #1 for pdf}%
\else \gdef\pdfimgext{pdf}%
\fi
\else \gdef\pdfimgext{JPG}%
\fi
\else \gdef\pdfimgext{jpeg}%
\fi
\else \gdef\pdfimgext{jpg}%
\fi
\else \gdef\pdfimgext{png}%
\fi
\closein 1
\endgroup
%
% without \immediate, pdftex seg faults when the same image is
% included twice. (Version 3.14159-pre-1.0-unofficial-20010704.)
\ifnum\pdftexversion < 14
\immediate\pdfimage
\else
\immediate\pdfximage
\fi
\ifdim \wd0 >0pt width \imagewidth \fi
\ifdim \wd2 >0pt height \imageheight \fi
\ifnum\pdftexversion<13
#1.\pdfimgext
\else
{#1.\pdfimgext}%
\fi
\ifnum\pdftexversion < 14 \else
\pdfrefximage \pdflastximage
\fi}
%
\def\pdfmkdest#1{{%
% We have to set dummies so commands such as @code, and characters
% such as \, aren't expanded when present in a section title.
\indexnofonts
\turnoffactive
\activebackslashdouble
\makevalueexpandable
\def\pdfdestname{#1}%
\backslashparens\pdfdestname
\safewhatsit{\pdfdest name{\pdfdestname} xyz}%
}}
%
% used to mark target names; must be expandable.
\def\pdfmkpgn#1{#1}
%
% by default, use a color that is dark enough to print on paper as
% nearly black, but still distinguishable for online viewing.
% (Defined in pdfcolor.tex.)
\let\urlcolor = \BrickRed
\let\linkcolor = \BrickRed
\def\endlink{\Black\pdfendlink}
%
% Adding outlines to PDF; macros for calculating structure of outlines
% come from Petr Olsak
\def\expnumber#1{\expandafter\ifx\csname#1\endcsname\relax 0%
\else \csname#1\endcsname \fi}
\def\advancenumber#1{\tempnum=\expnumber{#1}\relax
\advance\tempnum by 1
\expandafter\xdef\csname#1\endcsname{\the\tempnum}}
%
% #1 is the section text, which is what will be displayed in the
% outline by the pdf viewer. #2 is the pdf expression for the number
% of subentries (or empty, for subsubsections). #3 is the node text,
% which might be empty if this toc entry had no corresponding node.
% #4 is the page number
%
\def\dopdfoutline#1#2#3#4{%
% Generate a link to the node text if that exists; else, use the
% page number. We could generate a destination for the section
% text in the case where a section has no node, but it doesn't
% seem worth the trouble, since most documents are normally structured.
\def\pdfoutlinedest{#3}%
\ifx\pdfoutlinedest\empty
\def\pdfoutlinedest{#4}%
\else
% Doubled backslashes in the name.
{\activebackslashdouble \xdef\pdfoutlinedest{#3}%
\backslashparens\pdfoutlinedest}%
\fi
%
% Also double the backslashes in the display string.
{\activebackslashdouble \xdef\pdfoutlinetext{#1}%
\backslashparens\pdfoutlinetext}%
%
\pdfoutline goto name{\pdfmkpgn{\pdfoutlinedest}}#2{\pdfoutlinetext}%
}
%
\def\pdfmakeoutlines{%
\begingroup
% Thanh's hack / proper braces in bookmarks
\edef\mylbrace{\iftrue \string{\else}\fi}\let\{=\mylbrace
\edef\myrbrace{\iffalse{\else\string}\fi}\let\}=\myrbrace
%
% Read toc silently, to get counts of subentries for \pdfoutline.
\def\numchapentry##1##2##3##4{%
\def\thischapnum{##2}%
\def\thissecnum{0}%
\def\thissubsecnum{0}%
}%
\def\numsecentry##1##2##3##4{%
\advancenumber{chap\thischapnum}%
\def\thissecnum{##2}%
\def\thissubsecnum{0}%
}%
\def\numsubsecentry##1##2##3##4{%
\advancenumber{sec\thissecnum}%
\def\thissubsecnum{##2}%
}%
\def\numsubsubsecentry##1##2##3##4{%
\advancenumber{subsec\thissubsecnum}%
}%
\def\thischapnum{0}%
\def\thissecnum{0}%
\def\thissubsecnum{0}%
%
% use \def rather than \let here because we redefine \chapentry et
% al. a second time, below.
\def\appentry{\numchapentry}%
\def\appsecentry{\numsecentry}%
\def\appsubsecentry{\numsubsecentry}%
\def\appsubsubsecentry{\numsubsubsecentry}%
\def\unnchapentry{\numchapentry}%
\def\unnsecentry{\numsecentry}%
\def\unnsubsecentry{\numsubsecentry}%
\def\unnsubsubsecentry{\numsubsubsecentry}%
\readdatafile{toc}%
%
% Read toc second time, this time actually producing the outlines.
% The `-' means take the \expnumber as the absolute number of
% subentries, which we calculated on our first read of the .toc above.
%
% We use the node names as the destinations.
\def\numchapentry##1##2##3##4{%
\dopdfoutline{##1}{count-\expnumber{chap##2}}{##3}{##4}}%
\def\numsecentry##1##2##3##4{%
\dopdfoutline{##1}{count-\expnumber{sec##2}}{##3}{##4}}%
\def\numsubsecentry##1##2##3##4{%
\dopdfoutline{##1}{count-\expnumber{subsec##2}}{##3}{##4}}%
\def\numsubsubsecentry##1##2##3##4{% count is always zero
\dopdfoutline{##1}{}{##3}{##4}}%
%
% PDF outlines are displayed using system fonts, instead of
% document fonts. Therefore we cannot use special characters,
% since the encoding is unknown. For example, the eogonek from
% Latin 2 (0xea) gets translated to a | character. Info from
% Staszek Wawrykiewicz, 19 Jan 2004 04:09:24 +0100.
%
% xx to do this right, we have to translate 8-bit characters to
% their "best" equivalent, based on the @documentencoding. Right
% now, I guess we'll just let the pdf reader have its way.
\indexnofonts
\setupdatafile
\catcode`\\=\active \otherbackslash
\input \jobname.toc
\endgroup
}
%
\def\skipspaces#1{\def\PP{#1}\def\D{|}%
\ifx\PP\D\let\nextsp\relax
\else\let\nextsp\skipspaces
\ifx\p\space\else\addtokens{\filename}{\PP}%
\advance\filenamelength by 1
\fi
\fi
\nextsp}
\def\getfilename#1{\filenamelength=0\expandafter\skipspaces#1|\relax}
\ifnum\pdftexversion < 14
\let \startlink \pdfannotlink
\else
\let \startlink \pdfstartlink
\fi
% make a live url in pdf output.
\def\pdfurl#1{%
\begingroup
% it seems we really need yet another set of dummies; have not
% tried to figure out what each command should do in the context
% of @url. for now, just make @/ a no-op, that's the only one
% people have actually reported a problem with.
%
\normalturnoffactive
\def\@{@}%
\let\/=\empty
\makevalueexpandable
\leavevmode\urlcolor
\startlink attr{/Border [0 0 0]}%
user{/Subtype /Link /A << /S /URI /URI (#1) >>}%
\endgroup}
\def\pdfgettoks#1.{\setbox\boxA=\hbox{\toksA={#1.}\toksB={}\maketoks}}
\def\addtokens#1#2{\edef\addtoks{\noexpand#1={\the#1#2}}\addtoks}
\def\adn#1{\addtokens{\toksC}{#1}\global\countA=1\let\next=\maketoks}
\def\poptoks#1#2|ENDTOKS|{\let\first=#1\toksD={#1}\toksA={#2}}
\def\maketoks{%
\expandafter\poptoks\the\toksA|ENDTOKS|\relax
\ifx\first0\adn0
\else\ifx\first1\adn1 \else\ifx\first2\adn2 \else\ifx\first3\adn3
\else\ifx\first4\adn4 \else\ifx\first5\adn5 \else\ifx\first6\adn6
\else\ifx\first7\adn7 \else\ifx\first8\adn8 \else\ifx\first9\adn9
\else
\ifnum0=\countA\else\makelink\fi
\ifx\first.\let\next=\done\else
\let\next=\maketoks
\addtokens{\toksB}{\the\toksD}
\ifx\first,\addtokens{\toksB}{\space}\fi
\fi
\fi\fi\fi\fi\fi\fi\fi\fi\fi\fi
\next}
\def\makelink{\addtokens{\toksB}%
{\noexpand\pdflink{\the\toksC}}\toksC={}\global\countA=0}
\def\pdflink#1{%
\startlink attr{/Border [0 0 0]} goto name{\pdfmkpgn{#1}}
\linkcolor #1\endlink}
\def\done{\edef\st{\global\noexpand\toksA={\the\toksB}}\st}
\else
\let\pdfmkdest = \gobble
\let\pdfurl = \gobble
\let\endlink = \relax
\let\linkcolor = \relax
\let\pdfmakeoutlines = \relax
\fi % \ifx\pdfoutput
\message{fonts,}
% Change the current font style to #1, remembering it in \curfontstyle.
% For now, we do not accumulate font styles: @b{@i{foo}} prints foo in
% italics, not bold italics.
%
\def\setfontstyle#1{%
\def\curfontstyle{#1}% not as a control sequence, because we are \edef'd.
\csname ten#1\endcsname % change the current font
}
% Select #1 fonts with the current style.
%
\def\selectfonts#1{\csname #1fonts\endcsname \csname\curfontstyle\endcsname}
\def\rm{\fam=0 \setfontstyle{rm}}
\def\it{\fam=\itfam \setfontstyle{it}}
\def\sl{\fam=\slfam \setfontstyle{sl}}
\def\bf{\fam=\bffam \setfontstyle{bf}}\def\bfstylename{bf}
\def\tt{\fam=\ttfam \setfontstyle{tt}}
% Texinfo sort of supports the sans serif font style, which plain TeX does not.
% So we set up a \sf.
\newfam\sffam
\def\sf{\fam=\sffam \setfontstyle{sf}}
\let\li = \sf % Sometimes we call it \li, not \sf.
% We don't need math for this font style.
\def\ttsl{\setfontstyle{ttsl}}
% Default leading.
\newdimen\textleading \textleading = 13.2pt
% Set the baselineskip to #1, and the lineskip and strut size
% correspondingly. There is no deep meaning behind these magic numbers
% used as factors; they just match (closely enough) what Knuth defined.
%
\def\lineskipfactor{.08333}
\def\strutheightpercent{.70833}
\def\strutdepthpercent {.29167}
%
\def\setleading#1{%
\normalbaselineskip = #1\relax
\normallineskip = \lineskipfactor\normalbaselineskip
\normalbaselines
\setbox\strutbox =\hbox{%
\vrule width0pt height\strutheightpercent\baselineskip
depth \strutdepthpercent \baselineskip
}%
}
%
% PDF CMaps. See also LaTeX's t1.cmap.
%
% \cmapOT1
\ifpdf
\begingroup
\catcode`\^^M=\active \def^^M{^^J}% Output line endings as the ^^J char.
\catcode`\%=12 \immediate\pdfobj stream {%!PS-Adobe-3.0 Resource-CMap
%%DocumentNeededResources: ProcSet (CIDInit)
%%IncludeResource: ProcSet (CIDInit)
%%BeginResource: CMap (TeX-OT1-0)
%%Title: (TeX-OT1-0 TeX OT1 0)
%%Version: 1.000
%%EndComments
/CIDInit /ProcSet findresource begin
12 dict begin
begincmap
/CIDSystemInfo
<< /Registry (TeX)
/Ordering (OT1)
/Supplement 0
>> def
/CMapName /TeX-OT1-0 def
/CMapType 2 def
1 begincodespacerange
<00> <7F>
endcodespacerange
8 beginbfrange
<00> <01> <0393>
<09> <0A> <03A8>
<23> <26> <0023>
<28> <3B> <0028>
<3F> <5B> <003F>
<5D> <5E> <005D>
<61> <7A> <0061>
<7B> <7C> <2013>
endbfrange
40 beginbfchar
<02> <0398>
<03> <039B>
<04> <039E>
<05> <03A0>
<06> <03A3>
<07> <03D2>
<08> <03A6>
<0B> <00660066>
<0C> <00660069>
<0D> <0066006C>
<0E> <006600660069>
<0F> <00660066006C>
<10> <0131>
<11> <0237>
<12> <0060>
<13> <00B4>
<14> <02C7>
<15> <02D8>
<16> <00AF>
<17> <02DA>
<18> <00B8>
<19> <00DF>
<1A> <00E6>
<1B> <0153>
<1C> <00F8>
<1D> <00C6>
<1E> <0152>
<1F> <00D8>
<21> <0021>
<22> <201D>
<27> <2019>
<3C> <00A1>
<3D> <003D>
<3E> <00BF>
<5C> <201C>
<5F> <02D9>
<60> <2018>
<7D> <02DD>
<7E> <007E>
<7F> <00A8>
endbfchar
endcmap
CMapName currentdict /CMap defineresource pop
end
end
%%EndResource
%%EOF
}\endgroup
\expandafter\edef\csname cmapOT1\endcsname#1{%
\pdffontattr#1{/ToUnicode \the\pdflastobj\space 0 R}%
}%
%
% \cmapOT1IT
\begingroup
\catcode`\^^M=\active \def^^M{^^J}% Output line endings as the ^^J char.
\catcode`\%=12 \immediate\pdfobj stream {%!PS-Adobe-3.0 Resource-CMap
%%DocumentNeededResources: ProcSet (CIDInit)
%%IncludeResource: ProcSet (CIDInit)
%%BeginResource: CMap (TeX-OT1IT-0)
%%Title: (TeX-OT1IT-0 TeX OT1IT 0)
%%Version: 1.000
%%EndComments
/CIDInit /ProcSet findresource begin
12 dict begin
begincmap
/CIDSystemInfo
<< /Registry (TeX)
/Ordering (OT1IT)
/Supplement 0
>> def
/CMapName /TeX-OT1IT-0 def
/CMapType 2 def
1 begincodespacerange
<00> <7F>
endcodespacerange
8 beginbfrange
<00> <01> <0393>
<09> <0A> <03A8>
<25> <26> <0025>
<28> <3B> <0028>
<3F> <5B> <003F>
<5D> <5E> <005D>
<61> <7A> <0061>
<7B> <7C> <2013>
endbfrange
42 beginbfchar
<02> <0398>
<03> <039B>
<04> <039E>
<05> <03A0>
<06> <03A3>
<07> <03D2>
<08> <03A6>
<0B> <00660066>
<0C> <00660069>
<0D> <0066006C>
<0E> <006600660069>
<0F> <00660066006C>
<10> <0131>
<11> <0237>
<12> <0060>
<13> <00B4>
<14> <02C7>
<15> <02D8>
<16> <00AF>
<17> <02DA>
<18> <00B8>
<19> <00DF>
<1A> <00E6>
<1B> <0153>
<1C> <00F8>
<1D> <00C6>
<1E> <0152>
<1F> <00D8>
<21> <0021>
<22> <201D>
<23> <0023>
<24> <00A3>
<27> <2019>
<3C> <00A1>
<3D> <003D>
<3E> <00BF>
<5C> <201C>
<5F> <02D9>
<60> <2018>
<7D> <02DD>
<7E> <007E>
<7F> <00A8>
endbfchar
endcmap
CMapName currentdict /CMap defineresource pop
end
end
%%EndResource
%%EOF
}\endgroup
\expandafter\edef\csname cmapOT1IT\endcsname#1{%
\pdffontattr#1{/ToUnicode \the\pdflastobj\space 0 R}%
}%
%
% \cmapOT1TT
\begingroup
\catcode`\^^M=\active \def^^M{^^J}% Output line endings as the ^^J char.
\catcode`\%=12 \immediate\pdfobj stream {%!PS-Adobe-3.0 Resource-CMap
%%DocumentNeededResources: ProcSet (CIDInit)
%%IncludeResource: ProcSet (CIDInit)
%%BeginResource: CMap (TeX-OT1TT-0)
%%Title: (TeX-OT1TT-0 TeX OT1TT 0)
%%Version: 1.000
%%EndComments
/CIDInit /ProcSet findresource begin
12 dict begin
begincmap
/CIDSystemInfo
<< /Registry (TeX)
/Ordering (OT1TT)
/Supplement 0
>> def
/CMapName /TeX-OT1TT-0 def
/CMapType 2 def
1 begincodespacerange
<00> <7F>
endcodespacerange
5 beginbfrange
<00> <01> <0393>
<09> <0A> <03A8>
<21> <26> <0021>
<28> <5F> <0028>
<61> <7E> <0061>
endbfrange
32 beginbfchar
<02> <0398>
<03> <039B>
<04> <039E>
<05> <03A0>
<06> <03A3>
<07> <03D2>
<08> <03A6>
<0B> <2191>
<0C> <2193>
<0D> <0027>
<0E> <00A1>
<0F> <00BF>
<10> <0131>
<11> <0237>
<12> <0060>
<13> <00B4>
<14> <02C7>
<15> <02D8>
<16> <00AF>
<17> <02DA>
<18> <00B8>
<19> <00DF>
<1A> <00E6>
<1B> <0153>
<1C> <00F8>
<1D> <00C6>
<1E> <0152>
<1F> <00D8>
<20> <2423>
<27> <2019>
<60> <2018>
<7F> <00A8>
endbfchar
endcmap
CMapName currentdict /CMap defineresource pop
end
end
%%EndResource
%%EOF
}\endgroup
\expandafter\edef\csname cmapOT1TT\endcsname#1{%
\pdffontattr#1{/ToUnicode \the\pdflastobj\space 0 R}%
}%
\else
\expandafter\let\csname cmapOT1\endcsname\gobble
\expandafter\let\csname cmapOT1IT\endcsname\gobble
\expandafter\let\csname cmapOT1TT\endcsname\gobble
\fi
% Set the font macro #1 to the font named #2, adding on the
% specified font prefix (normally `cm').
% #3 is the font's design size, #4 is a scale factor, #5 is the CMap
% encoding (currently only OT1, OT1IT and OT1TT are allowed, pass
% empty to omit).
\def\setfont#1#2#3#4#5{%
\font#1=\fontprefix#2#3 scaled #4
\csname cmap#5\endcsname#1%
}
% This is what gets called when #5 of \setfont is empty.
\let\cmap\gobble
% Use cm as the default font prefix.
% To specify the font prefix, you must define \fontprefix
% before you read in texinfo.tex.
\ifx\fontprefix\undefined
\def\fontprefix{cm}
\fi
% Support font families that don't use the same naming scheme as CM.
\def\rmshape{r}
\def\rmbshape{bx} %where the normal face is bold
\def\bfshape{b}
\def\bxshape{bx}
\def\ttshape{tt}
\def\ttbshape{tt}
\def\ttslshape{sltt}
\def\itshape{ti}
\def\itbshape{bxti}
\def\slshape{sl}
\def\slbshape{bxsl}
\def\sfshape{ss}
\def\sfbshape{ss}
\def\scshape{csc}
\def\scbshape{csc}
% Definitions for a main text size of 11pt. This is the default in
% Texinfo.
%
\def\definetextfontsizexi{%
% Text fonts (11.2pt, magstep1).
\def\textnominalsize{11pt}
\edef\mainmagstep{\magstephalf}
\setfont\textrm\rmshape{10}{\mainmagstep}{OT1}
\setfont\texttt\ttshape{10}{\mainmagstep}{OT1TT}
\setfont\textbf\bfshape{10}{\mainmagstep}{OT1}
\setfont\textit\itshape{10}{\mainmagstep}{OT1IT}
\setfont\textsl\slshape{10}{\mainmagstep}{OT1}
\setfont\textsf\sfshape{10}{\mainmagstep}{OT1}
\setfont\textsc\scshape{10}{\mainmagstep}{OT1}
\setfont\textttsl\ttslshape{10}{\mainmagstep}{OT1TT}
\font\texti=cmmi10 scaled \mainmagstep
\font\textsy=cmsy10 scaled \mainmagstep
% A few fonts for @defun names and args.
\setfont\defbf\bfshape{10}{\magstep1}{OT1}
\setfont\deftt\ttshape{10}{\magstep1}{OT1TT}
\setfont\defttsl\ttslshape{10}{\magstep1}{OT1TT}
\def\df{\let\tentt=\deftt \let\tenbf = \defbf \let\tenttsl=\defttsl \bf}
% Fonts for indices, footnotes, small examples (9pt).
\def\smallnominalsize{9pt}
\setfont\smallrm\rmshape{9}{1000}{OT1}
\setfont\smalltt\ttshape{9}{1000}{OT1TT}
\setfont\smallbf\bfshape{10}{900}{OT1}
\setfont\smallit\itshape{9}{1000}{OT1IT}
\setfont\smallsl\slshape{9}{1000}{OT1}
\setfont\smallsf\sfshape{9}{1000}{OT1}
\setfont\smallsc\scshape{10}{900}{OT1}
\setfont\smallttsl\ttslshape{10}{900}{OT1TT}
\font\smalli=cmmi9
\font\smallsy=cmsy9
% Fonts for small examples (8pt).
\def\smallernominalsize{8pt}
\setfont\smallerrm\rmshape{8}{1000}{OT1}
\setfont\smallertt\ttshape{8}{1000}{OT1TT}
\setfont\smallerbf\bfshape{10}{800}{OT1}
\setfont\smallerit\itshape{8}{1000}{OT1IT}
\setfont\smallersl\slshape{8}{1000}{OT1}
\setfont\smallersf\sfshape{8}{1000}{OT1}
\setfont\smallersc\scshape{10}{800}{OT1}
\setfont\smallerttsl\ttslshape{10}{800}{OT1TT}
\font\smalleri=cmmi8
\font\smallersy=cmsy8
% Fonts for title page (20.4pt):
\def\titlenominalsize{20pt}
\setfont\titlerm\rmbshape{12}{\magstep3}{OT1}
\setfont\titleit\itbshape{10}{\magstep4}{OT1IT}
\setfont\titlesl\slbshape{10}{\magstep4}{OT1}
\setfont\titlett\ttbshape{12}{\magstep3}{OT1TT}
\setfont\titlettsl\ttslshape{10}{\magstep4}{OT1TT}
\setfont\titlesf\sfbshape{17}{\magstep1}{OT1}
\let\titlebf=\titlerm
\setfont\titlesc\scbshape{10}{\magstep4}{OT1}
\font\titlei=cmmi12 scaled \magstep3
\font\titlesy=cmsy10 scaled \magstep4
\def\authorrm{\secrm}
\def\authortt{\sectt}
% Chapter (and unnumbered) fonts (17.28pt).
\def\chapnominalsize{17pt}
\setfont\chaprm\rmbshape{12}{\magstep2}{OT1}
\setfont\chapit\itbshape{10}{\magstep3}{OT1IT}
\setfont\chapsl\slbshape{10}{\magstep3}{OT1}
\setfont\chaptt\ttbshape{12}{\magstep2}{OT1TT}
\setfont\chapttsl\ttslshape{10}{\magstep3}{OT1TT}
\setfont\chapsf\sfbshape{17}{1000}{OT1}
\let\chapbf=\chaprm
\setfont\chapsc\scbshape{10}{\magstep3}{OT1}
\font\chapi=cmmi12 scaled \magstep2
\font\chapsy=cmsy10 scaled \magstep3
% Section fonts (14.4pt).
\def\secnominalsize{14pt}
\setfont\secrm\rmbshape{12}{\magstep1}{OT1}
\setfont\secit\itbshape{10}{\magstep2}{OT1IT}
\setfont\secsl\slbshape{10}{\magstep2}{OT1}
\setfont\sectt\ttbshape{12}{\magstep1}{OT1TT}
\setfont\secttsl\ttslshape{10}{\magstep2}{OT1TT}
\setfont\secsf\sfbshape{12}{\magstep1}{OT1}
\let\secbf\secrm
\setfont\secsc\scbshape{10}{\magstep2}{OT1}
\font\seci=cmmi12 scaled \magstep1
\font\secsy=cmsy10 scaled \magstep2
% Subsection fonts (13.15pt).
\def\ssecnominalsize{13pt}
\setfont\ssecrm\rmbshape{12}{\magstephalf}{OT1}
\setfont\ssecit\itbshape{10}{1315}{OT1IT}
\setfont\ssecsl\slbshape{10}{1315}{OT1}
\setfont\ssectt\ttbshape{12}{\magstephalf}{OT1TT}
\setfont\ssecttsl\ttslshape{10}{1315}{OT1TT}
\setfont\ssecsf\sfbshape{12}{\magstephalf}{OT1}
\let\ssecbf\ssecrm
\setfont\ssecsc\scbshape{10}{1315}{OT1}
\font\sseci=cmmi12 scaled \magstephalf
\font\ssecsy=cmsy10 scaled 1315
% Reduced fonts for @acro in text (10pt).
\def\reducednominalsize{10pt}
\setfont\reducedrm\rmshape{10}{1000}{OT1}
\setfont\reducedtt\ttshape{10}{1000}{OT1TT}
\setfont\reducedbf\bfshape{10}{1000}{OT1}
\setfont\reducedit\itshape{10}{1000}{OT1IT}
\setfont\reducedsl\slshape{10}{1000}{OT1}
\setfont\reducedsf\sfshape{10}{1000}{OT1}
\setfont\reducedsc\scshape{10}{1000}{OT1}
\setfont\reducedttsl\ttslshape{10}{1000}{OT1TT}
\font\reducedi=cmmi10
\font\reducedsy=cmsy10
% reset the current fonts
\textfonts
\rm
} % end of 11pt text font size definitions
% Definitions to make the main text be 10pt Computer Modern, with
% section, chapter, etc., sizes following suit. This is for the GNU
% Press printing of the Emacs 22 manual. Maybe other manuals in the
% future. Used with @smallbook, which sets the leading to 12pt.
%
\def\definetextfontsizex{%
% Text fonts (10pt).
\def\textnominalsize{10pt}
\edef\mainmagstep{1000}
\setfont\textrm\rmshape{10}{\mainmagstep}{OT1}
\setfont\texttt\ttshape{10}{\mainmagstep}{OT1TT}
\setfont\textbf\bfshape{10}{\mainmagstep}{OT1}
\setfont\textit\itshape{10}{\mainmagstep}{OT1IT}
\setfont\textsl\slshape{10}{\mainmagstep}{OT1}
\setfont\textsf\sfshape{10}{\mainmagstep}{OT1}
\setfont\textsc\scshape{10}{\mainmagstep}{OT1}
\setfont\textttsl\ttslshape{10}{\mainmagstep}{OT1TT}
\font\texti=cmmi10 scaled \mainmagstep
\font\textsy=cmsy10 scaled \mainmagstep
% A few fonts for @defun names and args.
\setfont\defbf\bfshape{10}{\magstephalf}{OT1}
\setfont\deftt\ttshape{10}{\magstephalf}{OT1TT}
\setfont\defttsl\ttslshape{10}{\magstephalf}{OT1TT}
\def\df{\let\tentt=\deftt \let\tenbf = \defbf \let\tenttsl=\defttsl \bf}
% Fonts for indices, footnotes, small examples (9pt).
\def\smallnominalsize{9pt}
\setfont\smallrm\rmshape{9}{1000}{OT1}
\setfont\smalltt\ttshape{9}{1000}{OT1TT}
\setfont\smallbf\bfshape{10}{900}{OT1}
\setfont\smallit\itshape{9}{1000}{OT1IT}
\setfont\smallsl\slshape{9}{1000}{OT1}
\setfont\smallsf\sfshape{9}{1000}{OT1}
\setfont\smallsc\scshape{10}{900}{OT1}
\setfont\smallttsl\ttslshape{10}{900}{OT1TT}
\font\smalli=cmmi9
\font\smallsy=cmsy9
% Fonts for small examples (8pt).
\def\smallernominalsize{8pt}
\setfont\smallerrm\rmshape{8}{1000}{OT1}
\setfont\smallertt\ttshape{8}{1000}{OT1TT}
\setfont\smallerbf\bfshape{10}{800}{OT1}
\setfont\smallerit\itshape{8}{1000}{OT1IT}
\setfont\smallersl\slshape{8}{1000}{OT1}
\setfont\smallersf\sfshape{8}{1000}{OT1}
\setfont\smallersc\scshape{10}{800}{OT1}
\setfont\smallerttsl\ttslshape{10}{800}{OT1TT}
\font\smalleri=cmmi8
\font\smallersy=cmsy8
% Fonts for title page (20.4pt):
\def\titlenominalsize{20pt}
\setfont\titlerm\rmbshape{12}{\magstep3}{OT1}
\setfont\titleit\itbshape{10}{\magstep4}{OT1IT}
\setfont\titlesl\slbshape{10}{\magstep4}{OT1}
\setfont\titlett\ttbshape{12}{\magstep3}{OT1TT}
\setfont\titlettsl\ttslshape{10}{\magstep4}{OT1TT}
\setfont\titlesf\sfbshape{17}{\magstep1}{OT1}
\let\titlebf=\titlerm
\setfont\titlesc\scbshape{10}{\magstep4}{OT1}
\font\titlei=cmmi12 scaled \magstep3
\font\titlesy=cmsy10 scaled \magstep4
\def\authorrm{\secrm}
\def\authortt{\sectt}
% Chapter fonts (14.4pt).
\def\chapnominalsize{14pt}
\setfont\chaprm\rmbshape{12}{\magstep1}{OT1}
\setfont\chapit\itbshape{10}{\magstep2}{OT1IT}
\setfont\chapsl\slbshape{10}{\magstep2}{OT1}
\setfont\chaptt\ttbshape{12}{\magstep1}{OT1TT}
\setfont\chapttsl\ttslshape{10}{\magstep2}{OT1TT}
\setfont\chapsf\sfbshape{12}{\magstep1}{OT1}
\let\chapbf\chaprm
\setfont\chapsc\scbshape{10}{\magstep2}{OT1}
\font\chapi=cmmi12 scaled \magstep1
\font\chapsy=cmsy10 scaled \magstep2
% Section fonts (12pt).
\def\secnominalsize{12pt}
\setfont\secrm\rmbshape{12}{1000}{OT1}
\setfont\secit\itbshape{10}{\magstep1}{OT1IT}
\setfont\secsl\slbshape{10}{\magstep1}{OT1}
\setfont\sectt\ttbshape{12}{1000}{OT1TT}
\setfont\secttsl\ttslshape{10}{\magstep1}{OT1TT}
\setfont\secsf\sfbshape{12}{1000}{OT1}
\let\secbf\secrm
\setfont\secsc\scbshape{10}{\magstep1}{OT1}
\font\seci=cmmi12
\font\secsy=cmsy10 scaled \magstep1
% Subsection fonts (10pt).
\def\ssecnominalsize{10pt}
\setfont\ssecrm\rmbshape{10}{1000}{OT1}
\setfont\ssecit\itbshape{10}{1000}{OT1IT}
\setfont\ssecsl\slbshape{10}{1000}{OT1}
\setfont\ssectt\ttbshape{10}{1000}{OT1TT}
\setfont\ssecttsl\ttslshape{10}{1000}{OT1TT}
\setfont\ssecsf\sfbshape{10}{1000}{OT1}
\let\ssecbf\ssecrm
\setfont\ssecsc\scbshape{10}{1000}{OT1}
\font\sseci=cmmi10
\font\ssecsy=cmsy10
% Reduced fonts for @acro in text (9pt).
\def\reducednominalsize{9pt}
\setfont\reducedrm\rmshape{9}{1000}{OT1}
\setfont\reducedtt\ttshape{9}{1000}{OT1TT}
\setfont\reducedbf\bfshape{10}{900}{OT1}
\setfont\reducedit\itshape{9}{1000}{OT1IT}
\setfont\reducedsl\slshape{9}{1000}{OT1}
\setfont\reducedsf\sfshape{9}{1000}{OT1}
\setfont\reducedsc\scshape{10}{900}{OT1}
\setfont\reducedttsl\ttslshape{10}{900}{OT1TT}
\font\reducedi=cmmi9
\font\reducedsy=cmsy9
% reduce space between paragraphs
\divide\parskip by 2
% reset the current fonts
\textfonts
\rm
} % end of 10pt text font size definitions
% We provide the user-level command
% @fonttextsize 10
% (or 11) to redefine the text font size. pt is assumed.
%
\def\xword{10}
\def\xiword{11}
%
\parseargdef\fonttextsize{%
\def\textsizearg{#1}%
\wlog{doing @fonttextsize \textsizearg}%
%
% Set \globaldefs so that documents can use this inside @tex, since
% makeinfo 4.8 does not support it, but we need it nonetheless.
%
\begingroup \globaldefs=1
\ifx\textsizearg\xword \definetextfontsizex
\else \ifx\textsizearg\xiword \definetextfontsizexi
\else
\errhelp=\EMsimple
\errmessage{@fonttextsize only supports `10' or `11', not `\textsizearg'}
\fi\fi
\endgroup
}
% In order for the font changes to affect most math symbols and letters,
% we have to define the \textfont of the standard families. Since
% texinfo doesn't allow for producing subscripts and superscripts except
% in the main text, we don't bother to reset \scriptfont and
% \scriptscriptfont (which would also require loading a lot more fonts).
%
\def\resetmathfonts{%
\textfont0=\tenrm \textfont1=\teni \textfont2=\tensy
\textfont\itfam=\tenit \textfont\slfam=\tensl \textfont\bffam=\tenbf
\textfont\ttfam=\tentt \textfont\sffam=\tensf
}
% The font-changing commands redefine the meanings of \tenSTYLE, instead
% of just \STYLE. We do this because \STYLE needs to also set the
% current \fam for math mode. Our \STYLE (e.g., \rm) commands hardwire
% \tenSTYLE to set the current font.
%
% Each font-changing command also sets the names \lsize (one size lower)
% and \lllsize (three sizes lower). These relative commands are used in
% the LaTeX logo and acronyms.
%
% This all needs generalizing, badly.
%
\def\textfonts{%
\let\tenrm=\textrm \let\tenit=\textit \let\tensl=\textsl
\let\tenbf=\textbf \let\tentt=\texttt \let\smallcaps=\textsc
\let\tensf=\textsf \let\teni=\texti \let\tensy=\textsy
\let\tenttsl=\textttsl
\def\curfontsize{text}%
\def\lsize{reduced}\def\lllsize{smaller}%
\resetmathfonts \setleading{\textleading}}
\def\titlefonts{%
\let\tenrm=\titlerm \let\tenit=\titleit \let\tensl=\titlesl
\let\tenbf=\titlebf \let\tentt=\titlett \let\smallcaps=\titlesc
\let\tensf=\titlesf \let\teni=\titlei \let\tensy=\titlesy
\let\tenttsl=\titlettsl
\def\curfontsize{title}%
\def\lsize{chap}\def\lllsize{subsec}%
\resetmathfonts \setleading{25pt}}
\def\titlefont#1{{\titlefonts\rm #1}}
\def\chapfonts{%
\let\tenrm=\chaprm \let\tenit=\chapit \let\tensl=\chapsl
\let\tenbf=\chapbf \let\tentt=\chaptt \let\smallcaps=\chapsc
\let\tensf=\chapsf \let\teni=\chapi \let\tensy=\chapsy
\let\tenttsl=\chapttsl
\def\curfontsize{chap}%
\def\lsize{sec}\def\lllsize{text}%
\resetmathfonts \setleading{19pt}}
\def\secfonts{%
\let\tenrm=\secrm \let\tenit=\secit \let\tensl=\secsl
\let\tenbf=\secbf \let\tentt=\sectt \let\smallcaps=\secsc
\let\tensf=\secsf \let\teni=\seci \let\tensy=\secsy
\let\tenttsl=\secttsl
\def\curfontsize{sec}%
\def\lsize{subsec}\def\lllsize{reduced}%
\resetmathfonts \setleading{16pt}}
\def\subsecfonts{%
\let\tenrm=\ssecrm \let\tenit=\ssecit \let\tensl=\ssecsl
\let\tenbf=\ssecbf \let\tentt=\ssectt \let\smallcaps=\ssecsc
\let\tensf=\ssecsf \let\teni=\sseci \let\tensy=\ssecsy
\let\tenttsl=\ssecttsl
\def\curfontsize{ssec}%
\def\lsize{text}\def\lllsize{small}%
\resetmathfonts \setleading{15pt}}
\let\subsubsecfonts = \subsecfonts
\def\reducedfonts{%
\let\tenrm=\reducedrm \let\tenit=\reducedit \let\tensl=\reducedsl
\let\tenbf=\reducedbf \let\tentt=\reducedtt \let\reducedcaps=\reducedsc
\let\tensf=\reducedsf \let\teni=\reducedi \let\tensy=\reducedsy
\let\tenttsl=\reducedttsl
\def\curfontsize{reduced}%
\def\lsize{small}\def\lllsize{smaller}%
\resetmathfonts \setleading{10.5pt}}
\def\smallfonts{%
\let\tenrm=\smallrm \let\tenit=\smallit \let\tensl=\smallsl
\let\tenbf=\smallbf \let\tentt=\smalltt \let\smallcaps=\smallsc
\let\tensf=\smallsf \let\teni=\smalli \let\tensy=\smallsy
\let\tenttsl=\smallttsl
\def\curfontsize{small}%
\def\lsize{smaller}\def\lllsize{smaller}%
\resetmathfonts \setleading{10.5pt}}
\def\smallerfonts{%
\let\tenrm=\smallerrm \let\tenit=\smallerit \let\tensl=\smallersl
\let\tenbf=\smallerbf \let\tentt=\smallertt \let\smallcaps=\smallersc
\let\tensf=\smallersf \let\teni=\smalleri \let\tensy=\smallersy
\let\tenttsl=\smallerttsl
\def\curfontsize{smaller}%
\def\lsize{smaller}\def\lllsize{smaller}%
\resetmathfonts \setleading{9.5pt}}
% Set the fonts to use with the @small... environments.
\let\smallexamplefonts = \smallfonts
% About \smallexamplefonts. If we use \smallfonts (9pt), @smallexample
% can fit this many characters:
% 8.5x11=86 smallbook=72 a4=90 a5=69
% If we use \scriptfonts (8pt), then we can fit this many characters:
% 8.5x11=90+ smallbook=80 a4=90+ a5=77
% For me, subjectively, the few extra characters that fit aren't worth
% the additional smallness of 8pt. So I'm making the default 9pt.
%
% By the way, for comparison, here's what fits with @example (10pt):
% 8.5x11=71 smallbook=60 a4=75 a5=58
%
% I wish the USA used A4 paper.
% --karl, 24jan03.
% Set up the default fonts, so we can use them for creating boxes.
%
\definetextfontsizexi
% Define these so they can be easily changed for other fonts.
\def\angleleft{$\langle$}
\def\angleright{$\rangle$}
% Count depth in font-changes, for error checks
\newcount\fontdepth \fontdepth=0
% Fonts for short table of contents.
\setfont\shortcontrm\rmshape{12}{1000}{OT1}
\setfont\shortcontbf\bfshape{10}{\magstep1}{OT1} % no cmb12
\setfont\shortcontsl\slshape{12}{1000}{OT1}
\setfont\shortconttt\ttshape{12}{1000}{OT1TT}
%% Add scribe-like font environments, plus @l for inline lisp (usually sans
%% serif) and @ii for TeX italic
% \smartitalic{ARG} outputs arg in italics, followed by an italic correction
% unless the following character is such as not to need one.
\def\smartitalicx{\ifx\next,\else\ifx\next-\else\ifx\next.\else
\ptexslash\fi\fi\fi}
\def\smartslanted#1{{\ifusingtt\ttsl\sl #1}\futurelet\next\smartitalicx}
\def\smartitalic#1{{\ifusingtt\ttsl\it #1}\futurelet\next\smartitalicx}
% like \smartslanted except unconditionally uses \ttsl.
% @var is set to this for defun arguments.
\def\ttslanted#1{{\ttsl #1}\futurelet\next\smartitalicx}
% like \smartslanted except unconditionally use \sl. We never want
% ttsl for book titles, do we?
\def\cite#1{{\sl #1}\futurelet\next\smartitalicx}
\let\i=\smartitalic
\let\slanted=\smartslanted
\let\var=\smartslanted
\let\dfn=\smartslanted
\let\emph=\smartitalic
% @b, explicit bold.
\def\b#1{{\bf #1}}
\let\strong=\b
% @sansserif, explicit sans.
\def\sansserif#1{{\sf #1}}
% We can't just use \exhyphenpenalty, because that only has effect at
% the end of a paragraph. Restore normal hyphenation at the end of the
% group within which \nohyphenation is presumably called.
%
\def\nohyphenation{\hyphenchar\font = -1 \aftergroup\restorehyphenation}
\def\restorehyphenation{\hyphenchar\font = `- }
% Set sfcode to normal for the chars that usually have another value.
% Can't use plain's \frenchspacing because it uses the `\x notation, and
% sometimes \x has an active definition that messes things up.
%
\catcode`@=11
\def\plainfrenchspacing{%
\sfcode\dotChar =\@m \sfcode\questChar=\@m \sfcode\exclamChar=\@m
\sfcode\colonChar=\@m \sfcode\semiChar =\@m \sfcode\commaChar =\@m
\def\endofsentencespacefactor{1000}% for @. and friends
}
\def\plainnonfrenchspacing{%
\sfcode`\.3000\sfcode`\?3000\sfcode`\!3000
\sfcode`\:2000\sfcode`\;1500\sfcode`\,1250
\def\endofsentencespacefactor{3000}% for @. and friends
}
\catcode`@=\other
\def\endofsentencespacefactor{3000}% default
\def\t#1{%
{\tt \rawbackslash \plainfrenchspacing #1}%
\null
}
\def\samp#1{`\tclose{#1}'\null}
\setfont\keyrm\rmshape{8}{1000}{OT1}
\font\keysy=cmsy9
\def\key#1{{\keyrm\textfont2=\keysy \leavevmode\hbox{%
\raise0.4pt\hbox{\angleleft}\kern-.08em\vtop{%
\vbox{\hrule\kern-0.4pt
\hbox{\raise0.4pt\hbox{\vphantom{\angleleft}}#1}}%
\kern-0.4pt\hrule}%
\kern-.06em\raise0.4pt\hbox{\angleright}}}}
\def\key #1{{\nohyphenation \uppercase{#1}}\null}
% The old definition, with no lozenge:
%\def\key #1{{\ttsl \nohyphenation \uppercase{#1}}\null}
\def\ctrl #1{{\tt \rawbackslash \hat}#1}
% @file, @option are the same as @samp.
\let\file=\samp
\let\option=\samp
% @code is a modification of @t,
% which makes spaces the same size as normal in the surrounding text.
\def\tclose#1{%
{%
% Change normal interword space to be same as for the current font.
\spaceskip = \fontdimen2\font
%
% Switch to typewriter.
\tt
%
% But `\ ' produces the large typewriter interword space.
\def\ {{\spaceskip = 0pt{} }}%
%
% Turn off hyphenation.
\nohyphenation
%
\rawbackslash
\plainfrenchspacing
#1%
}%
\null
}
% We *must* turn on hyphenation at `-' and `_' in @code.
% Otherwise, it is too hard to avoid overfull hboxes
% in the Emacs manual, the Library manual, etc.
% Unfortunately, TeX uses one parameter (\hyphenchar) to control
% both hyphenation at - and hyphenation within words.
% We must therefore turn them both off (\tclose does that)
% and arrange explicitly to hyphenate at a dash.
% -- rms.
{
\catcode`\-=\active \catcode`\_=\active
\catcode`\'=\active \catcode`\`=\active
%
\global\def\code{\begingroup
\catcode\rquoteChar=\active \catcode\lquoteChar=\active
\let'\codequoteright \let`\codequoteleft
%
\catcode\dashChar=\active \catcode\underChar=\active
\ifallowcodebreaks
\let-\codedash
\let_\codeunder
\else
\let-\realdash
\let_\realunder
\fi
\codex
}
}
\def\realdash{-}
\def\codedash{-\discretionary{}{}{}}
\def\codeunder{%
% this is all so @math{@code{var_name}+1} can work. In math mode, _
% is "active" (mathcode"8000) and \normalunderscore (or \char95, etc.)
% will therefore expand the active definition of _, which is us
% (inside @code that is), therefore an endless loop.
\ifusingtt{\ifmmode
\mathchar"075F % class 0=ordinary, family 7=ttfam, pos 0x5F=_.
\else\normalunderscore \fi
\discretionary{}{}{}}%
{\_}%
}
\def\codex #1{\tclose{#1}\endgroup}
% An additional complication: the above will allow breaks after, e.g.,
% each of the four underscores in __typeof__. This is undesirable in
% some manuals, especially if they don't have long identifiers in
% general. @allowcodebreaks provides a way to control this.
%
\newif\ifallowcodebreaks \allowcodebreakstrue
\def\keywordtrue{true}
\def\keywordfalse{false}
\parseargdef\allowcodebreaks{%
\def\txiarg{#1}%
\ifx\txiarg\keywordtrue
\allowcodebreakstrue
\else\ifx\txiarg\keywordfalse
\allowcodebreaksfalse
\else
\errhelp = \EMsimple
\errmessage{Unknown @allowcodebreaks option `\txiarg'}%
\fi\fi
}
% @kbd is like @code, except that if the argument is just one @key command,
% then @kbd has no effect.
% @kbdinputstyle -- arg is `distinct' (@kbd uses slanted tty font always),
% `example' (@kbd uses ttsl only inside of @example and friends),
% or `code' (@kbd uses normal tty font always).
\parseargdef\kbdinputstyle{%
\def\txiarg{#1}%
\ifx\txiarg\worddistinct
\gdef\kbdexamplefont{\ttsl}\gdef\kbdfont{\ttsl}%
\else\ifx\txiarg\wordexample
\gdef\kbdexamplefont{\ttsl}\gdef\kbdfont{\tt}%
\else\ifx\txiarg\wordcode
\gdef\kbdexamplefont{\tt}\gdef\kbdfont{\tt}%
\else
\errhelp = \EMsimple
\errmessage{Unknown @kbdinputstyle option `\txiarg'}%
\fi\fi\fi
}
\def\worddistinct{distinct}
\def\wordexample{example}
\def\wordcode{code}
% Default is `distinct.'
\kbdinputstyle distinct
\def\xkey{\key}
\def\kbdfoo#1#2#3\par{\def\one{#1}\def\three{#3}\def\threex{??}%
\ifx\one\xkey\ifx\threex\three \key{#2}%
\else{\tclose{\kbdfont\look}}\fi
\else{\tclose{\kbdfont\look}}\fi}
% For @indicateurl, @env, @command quotes seem unnecessary, so use \code.
\let\indicateurl=\code
\let\env=\code
\let\command=\code
% @uref (abbreviation for `urlref') takes an optional (comma-separated)
% second argument specifying the text to display and an optional third
% arg as text to display instead of (rather than in addition to) the url
% itself. First (mandatory) arg is the url. Perhaps eventually put in
% a hypertex \special here.
%
\def\uref#1{\douref #1,,,\finish}
\def\douref#1,#2,#3,#4\finish{\begingroup
\unsepspaces
\pdfurl{#1}%
\setbox0 = \hbox{\ignorespaces #3}%
\ifdim\wd0 > 0pt
\unhbox0 % third arg given, show only that
\else
\setbox0 = \hbox{\ignorespaces #2}%
\ifdim\wd0 > 0pt
\ifpdf
\unhbox0 % PDF: 2nd arg given, show only it
\else
\unhbox0\ (\code{#1})% DVI: 2nd arg given, show both it and url
\fi
\else
\code{#1}% only url given, so show it
\fi
\fi
\endlink
\endgroup}
% @url synonym for @uref, since that's how everyone uses it.
%
\let\url=\uref
% rms does not like angle brackets --karl, 17may97.
% So now @email is just like @uref, unless we are pdf.
%
%\def\email#1{\angleleft{\tt #1}\angleright}
\ifpdf
\def\email#1{\doemail#1,,\finish}
\def\doemail#1,#2,#3\finish{\begingroup
\unsepspaces
\pdfurl{mailto:#1}%
\setbox0 = \hbox{\ignorespaces #2}%
\ifdim\wd0>0pt\unhbox0\else\code{#1}\fi
\endlink
\endgroup}
\else
\let\email=\uref
\fi
% Check if we are currently using a typewriter font. Since all the
% Computer Modern typewriter fonts have zero interword stretch (and
% shrink), and it is reasonable to expect all typewriter fonts to have
% this property, we can check that font parameter.
%
\def\ifmonospace{\ifdim\fontdimen3\font=0pt }
% Typeset a dimension, e.g., `in' or `pt'. The only reason for the
% argument is to make the input look right: @dmn{pt} instead of @dmn{}pt.
%
\def\dmn#1{\thinspace #1}
\def\kbd#1{\def\look{#1}\expandafter\kbdfoo\look??\par}
% @l was never documented to mean ``switch to the Lisp font'',
% and it is not used as such in any manual I can find. We need it for
% Polish suppressed-l. --karl, 22sep96.
%\def\l#1{{\li #1}\null}
% Explicit font changes: @r, @sc, undocumented @ii.
\def\r#1{{\rm #1}} % roman font
\def\sc#1{{\smallcaps#1}} % smallcaps font
\def\ii#1{{\it #1}} % italic font
% @acronym for "FBI", "NATO", and the like.
% We print this one point size smaller, since it's intended for
% all-uppercase.
%
\def\acronym#1{\doacronym #1,,\finish}
\def\doacronym#1,#2,#3\finish{%
{\selectfonts\lsize #1}%
\def\temp{#2}%
\ifx\temp\empty \else
\space ({\unsepspaces \ignorespaces \temp \unskip})%
\fi
}
% @abbr for "Comput. J." and the like.
% No font change, but don't do end-of-sentence spacing.
%
\def\abbr#1{\doabbr #1,,\finish}
\def\doabbr#1,#2,#3\finish{%
{\plainfrenchspacing #1}%
\def\temp{#2}%
\ifx\temp\empty \else
\space ({\unsepspaces \ignorespaces \temp \unskip})%
\fi
}
% @pounds{} is a sterling sign, which Knuth put in the CM italic font.
%
\def\pounds{{\it\$}}
% @euro{} comes from a separate font, depending on the current style.
% We use the free feym* fonts from the eurosym package by Henrik
% Theiling, which support regular, slanted, bold and bold slanted (and
% "outlined" (blackboard board, sort of) versions, which we don't need).
-% It is available from http://www.ctan.org/tex-archive/fonts/eurosym.
+% It is available from https://www.ctan.org/tex-archive/fonts/eurosym.
%
% Although only regular is the truly official Euro symbol, we ignore
% that. The Euro is designed to be slightly taller than the regular
% font height.
%
% feymr - regular
% feymo - slanted
% feybr - bold
% feybo - bold slanted
%
% There is no good (free) typewriter version, to my knowledge.
% A feymr10 euro is ~7.3pt wide, while a normal cmtt10 char is ~5.25pt wide.
% Hmm.
%
% Also doesn't work in math. Do we need to do math with euro symbols?
% Hope not.
%
%
\def\euro{{\eurofont e}}
\def\eurofont{%
% We set the font at each command, rather than predefining it in
% \textfonts and the other font-switching commands, so that
% installations which never need the symbol don't have to have the
% font installed.
%
% There is only one designed size (nominal 10pt), so we always scale
% that to the current nominal size.
%
% By the way, simply using "at 1em" works for cmr10 and the like, but
% does not work for cmbx10 and other extended/shrunken fonts.
%
\def\eurosize{\csname\curfontsize nominalsize\endcsname}%
%
\ifx\curfontstyle\bfstylename
% bold:
\font\thiseurofont = \ifusingit{feybo10}{feybr10} at \eurosize
\else
% regular:
\font\thiseurofont = \ifusingit{feymo10}{feymr10} at \eurosize
\fi
\thiseurofont
}
% @registeredsymbol - R in a circle. The font for the R should really
% be smaller yet, but lllsize is the best we can do for now.
% Adapted from the plain.tex definition of \copyright.
%
\def\registeredsymbol{%
$^{{\ooalign{\hfil\raise.07ex\hbox{\selectfonts\lllsize R}%
\hfil\crcr\Orb}}%
}$%
}
% @textdegree - the normal degrees sign.
%
\def\textdegree{$^\circ$}
% Laurent Siebenmann reports \Orb undefined with:
% Textures 1.7.7 (preloaded format=plain 93.10.14) (68K) 16 APR 2004 02:38
% so we'll define it if necessary.
%
\ifx\Orb\undefined
\def\Orb{\mathhexbox20D}
\fi
\message{page headings,}
\newskip\titlepagetopglue \titlepagetopglue = 1.5in
\newskip\titlepagebottomglue \titlepagebottomglue = 2pc
% First the title page. Must do @settitle before @titlepage.
\newif\ifseenauthor
\newif\iffinishedtitlepage
% Do an implicit @contents or @shortcontents after @end titlepage if the
% user says @setcontentsaftertitlepage or @setshortcontentsaftertitlepage.
%
\newif\ifsetcontentsaftertitlepage
\let\setcontentsaftertitlepage = \setcontentsaftertitlepagetrue
\newif\ifsetshortcontentsaftertitlepage
\let\setshortcontentsaftertitlepage = \setshortcontentsaftertitlepagetrue
\parseargdef\shorttitlepage{\begingroup\hbox{}\vskip 1.5in \chaprm \centerline{#1}%
\endgroup\page\hbox{}\page}
\envdef\titlepage{%
% Open one extra group, as we want to close it in the middle of \Etitlepage.
\begingroup
\parindent=0pt \textfonts
% Leave some space at the very top of the page.
\vglue\titlepagetopglue
% No rule at page bottom unless we print one at the top with @title.
\finishedtitlepagetrue
%
% Most title ``pages'' are actually two pages long, with space
% at the top of the second. We don't want the ragged left on the second.
\let\oldpage = \page
\def\page{%
\iffinishedtitlepage\else
\finishtitlepage
\fi
\let\page = \oldpage
\page
\null
}%
}
\def\Etitlepage{%
\iffinishedtitlepage\else
\finishtitlepage
\fi
% It is important to do the page break before ending the group,
% because the headline and footline are only empty inside the group.
% If we use the new definition of \page, we always get a blank page
% after the title page, which we certainly don't want.
\oldpage
\endgroup
%
% Need this before the \...aftertitlepage checks so that if they are
% in effect the toc pages will come out with page numbers.
\HEADINGSon
%
% If they want short, they certainly want long too.
\ifsetshortcontentsaftertitlepage
\shortcontents
\contents
\global\let\shortcontents = \relax
\global\let\contents = \relax
\fi
%
\ifsetcontentsaftertitlepage
\contents
\global\let\contents = \relax
\global\let\shortcontents = \relax
\fi
}
\def\finishtitlepage{%
\vskip4pt \hrule height 2pt width \hsize
\vskip\titlepagebottomglue
\finishedtitlepagetrue
}
%%% Macros to be used within @titlepage:
\let\subtitlerm=\tenrm
\def\subtitlefont{\subtitlerm \normalbaselineskip = 13pt \normalbaselines}
\def\authorfont{\authorrm \normalbaselineskip = 16pt \normalbaselines
\let\tt=\authortt}
\parseargdef\title{%
\checkenv\titlepage
\leftline{\titlefonts\rm #1}
% print a rule at the page bottom also.
\finishedtitlepagefalse
\vskip4pt \hrule height 4pt width \hsize \vskip4pt
}
\parseargdef\subtitle{%
\checkenv\titlepage
{\subtitlefont \rightline{#1}}%
}
% @author should come last, but may come many times.
% It can also be used inside @quotation.
%
\parseargdef\author{%
\def\temp{\quotation}%
\ifx\thisenv\temp
\def\quotationauthor{#1}% printed in \Equotation.
\else
\checkenv\titlepage
\ifseenauthor\else \vskip 0pt plus 1filll \seenauthortrue \fi
{\authorfont \leftline{#1}}%
\fi
}
%%% Set up page headings and footings.
\let\thispage=\folio
\newtoks\evenheadline % headline on even pages
\newtoks\oddheadline % headline on odd pages
\newtoks\evenfootline % footline on even pages
\newtoks\oddfootline % footline on odd pages
% Now make TeX use those variables
\headline={{\textfonts\rm \ifodd\pageno \the\oddheadline
\else \the\evenheadline \fi}}
\footline={{\textfonts\rm \ifodd\pageno \the\oddfootline
\else \the\evenfootline \fi}\HEADINGShook}
\let\HEADINGShook=\relax
% Commands to set those variables.
% For example, this is what @headings on does
% @evenheading @thistitle|@thispage|@thischapter
% @oddheading @thischapter|@thispage|@thistitle
% @evenfooting @thisfile||
% @oddfooting ||@thisfile
\def\evenheading{\parsearg\evenheadingxxx}
\def\evenheadingxxx #1{\evenheadingyyy #1\|\|\|\|\finish}
\def\evenheadingyyy #1\|#2\|#3\|#4\finish{%
\global\evenheadline={\rlap{\centerline{#2}}\line{#1\hfil#3}}}
\def\oddheading{\parsearg\oddheadingxxx}
\def\oddheadingxxx #1{\oddheadingyyy #1\|\|\|\|\finish}
\def\oddheadingyyy #1\|#2\|#3\|#4\finish{%
\global\oddheadline={\rlap{\centerline{#2}}\line{#1\hfil#3}}}
\parseargdef\everyheading{\oddheadingxxx{#1}\evenheadingxxx{#1}}%
\def\evenfooting{\parsearg\evenfootingxxx}
\def\evenfootingxxx #1{\evenfootingyyy #1\|\|\|\|\finish}
\def\evenfootingyyy #1\|#2\|#3\|#4\finish{%
\global\evenfootline={\rlap{\centerline{#2}}\line{#1\hfil#3}}}
\def\oddfooting{\parsearg\oddfootingxxx}
\def\oddfootingxxx #1{\oddfootingyyy #1\|\|\|\|\finish}
\def\oddfootingyyy #1\|#2\|#3\|#4\finish{%
\global\oddfootline = {\rlap{\centerline{#2}}\line{#1\hfil#3}}%
%
% Leave some space for the footline. Hopefully ok to assume
% @evenfooting will not be used by itself.
\global\advance\pageheight by -12pt
\global\advance\vsize by -12pt
}
\parseargdef\everyfooting{\oddfootingxxx{#1}\evenfootingxxx{#1}}
% @headings double turns headings on for double-sided printing.
% @headings single turns headings on for single-sided printing.
% @headings off turns them off.
% @headings on same as @headings double, retained for compatibility.
% @headings after turns on double-sided headings after this page.
% @headings doubleafter turns on double-sided headings after this page.
% @headings singleafter turns on single-sided headings after this page.
% By default, they are off at the start of a document,
% and turned `on' after @end titlepage.
\def\headings #1 {\csname HEADINGS#1\endcsname}
\def\HEADINGSoff{%
\global\evenheadline={\hfil} \global\evenfootline={\hfil}
\global\oddheadline={\hfil} \global\oddfootline={\hfil}}
\HEADINGSoff
% When we turn headings on, set the page number to 1.
% For double-sided printing, put current file name in lower left corner,
% chapter name on inside top of right hand pages, document
% title on inside top of left hand pages, and page numbers on outside top
% edge of all pages.
\def\HEADINGSdouble{%
\global\pageno=1
\global\evenfootline={\hfil}
\global\oddfootline={\hfil}
\global\evenheadline={\line{\folio\hfil\thistitle}}
\global\oddheadline={\line{\thischapter\hfil\folio}}
\global\let\contentsalignmacro = \chapoddpage
}
\let\contentsalignmacro = \chappager
% For single-sided printing, chapter title goes across top left of page,
% page number on top right.
\def\HEADINGSsingle{%
\global\pageno=1
\global\evenfootline={\hfil}
\global\oddfootline={\hfil}
\global\evenheadline={\line{\thischapter\hfil\folio}}
\global\oddheadline={\line{\thischapter\hfil\folio}}
\global\let\contentsalignmacro = \chappager
}
\def\HEADINGSon{\HEADINGSdouble}
\def\HEADINGSafter{\let\HEADINGShook=\HEADINGSdoublex}
\let\HEADINGSdoubleafter=\HEADINGSafter
\def\HEADINGSdoublex{%
\global\evenfootline={\hfil}
\global\oddfootline={\hfil}
\global\evenheadline={\line{\folio\hfil\thistitle}}
\global\oddheadline={\line{\thischapter\hfil\folio}}
\global\let\contentsalignmacro = \chapoddpage
}
\def\HEADINGSsingleafter{\let\HEADINGShook=\HEADINGSsinglex}
\def\HEADINGSsinglex{%
\global\evenfootline={\hfil}
\global\oddfootline={\hfil}
\global\evenheadline={\line{\thischapter\hfil\folio}}
\global\oddheadline={\line{\thischapter\hfil\folio}}
\global\let\contentsalignmacro = \chappager
}
% Subroutines used in generating headings
% This produces Day Month Year style of output.
% Only define if not already defined, in case a txi-??.tex file has set
% up a different format (e.g., txi-cs.tex does this).
\ifx\today\undefined
\def\today{%
\number\day\space
\ifcase\month
\or\putwordMJan\or\putwordMFeb\or\putwordMMar\or\putwordMApr
\or\putwordMMay\or\putwordMJun\or\putwordMJul\or\putwordMAug
\or\putwordMSep\or\putwordMOct\or\putwordMNov\or\putwordMDec
\fi
\space\number\year}
\fi
% @settitle line... specifies the title of the document, for headings.
% It generates no output of its own.
\def\thistitle{\putwordNoTitle}
\def\settitle{\parsearg{\gdef\thistitle}}
\message{tables,}
% Tables -- @table, @ftable, @vtable, @item(x).
% default indentation of table text
\newdimen\tableindent \tableindent=.8in
% default indentation of @itemize and @enumerate text
\newdimen\itemindent \itemindent=.3in
% margin between end of table item and start of table text.
\newdimen\itemmargin \itemmargin=.1in
% used internally for \itemindent minus \itemmargin
\newdimen\itemmax
% Note @table, @ftable, and @vtable define @item, @itemx, etc., with
% these defs.
% They also define \itemindex
% to index the item name in whatever manner is desired (perhaps none).
\newif\ifitemxneedsnegativevskip
\def\itemxpar{\par\ifitemxneedsnegativevskip\nobreak\vskip-\parskip\nobreak\fi}
\def\internalBitem{\smallbreak \parsearg\itemzzz}
\def\internalBitemx{\itemxpar \parsearg\itemzzz}
\def\itemzzz #1{\begingroup %
\advance\hsize by -\rightskip
\advance\hsize by -\tableindent
\setbox0=\hbox{\itemindicate{#1}}%
\itemindex{#1}%
\nobreak % This prevents a break before @itemx.
%
% If the item text does not fit in the space we have, put it on a line
% by itself, and do not allow a page break either before or after that
% line. We do not start a paragraph here because then if the next
% command is, e.g., @kindex, the whatsit would get put into the
% horizontal list on a line by itself, resulting in extra blank space.
\ifdim \wd0>\itemmax
%
% Make this a paragraph so we get the \parskip glue and wrapping,
% but leave it ragged-right.
\begingroup
\advance\leftskip by-\tableindent
\advance\hsize by\tableindent
\advance\rightskip by0pt plus1fil
\leavevmode\unhbox0\par
\endgroup
%
% We're going to be starting a paragraph, but we don't want the
% \parskip glue -- logically it's part of the @item we just started.
\nobreak \vskip-\parskip
%
% Stop a page break at the \parskip glue coming up. However, if
% what follows is an environment such as @example, there will be no
% \parskip glue; then the negative vskip we just inserted would
% cause the example and the item to crash together. So we use this
% bizarre value of 10001 as a signal to \aboveenvbreak to insert
% \parskip glue after all. Section titles are handled this way also.
%
\penalty 10001
\endgroup
\itemxneedsnegativevskipfalse
\else
% The item text fits into the space. Start a paragraph, so that the
% following text (if any) will end up on the same line.
\noindent
% Do this with kerns and \unhbox so that if there is a footnote in
% the item text, it can migrate to the main vertical list and
% eventually be printed.
\nobreak\kern-\tableindent
\dimen0 = \itemmax \advance\dimen0 by \itemmargin \advance\dimen0 by -\wd0
\unhbox0
\nobreak\kern\dimen0
\endgroup
\itemxneedsnegativevskiptrue
\fi
}
\def\item{\errmessage{@item while not in a list environment}}
\def\itemx{\errmessage{@itemx while not in a list environment}}
% @table, @ftable, @vtable.
\envdef\table{%
\let\itemindex\gobble
\tablecheck{table}%
}
\envdef\ftable{%
\def\itemindex ##1{\doind {fn}{\code{##1}}}%
\tablecheck{ftable}%
}
\envdef\vtable{%
\def\itemindex ##1{\doind {vr}{\code{##1}}}%
\tablecheck{vtable}%
}
\def\tablecheck#1{%
\ifnum \the\catcode`\^^M=\active
\endgroup
\errmessage{This command won't work in this context; perhaps the problem is
that we are \inenvironment\thisenv}%
\def\next{\doignore{#1}}%
\else
\let\next\tablex
\fi
\next
}
\def\tablex#1{%
\def\itemindicate{#1}%
\parsearg\tabley
}
\def\tabley#1{%
{%
\makevalueexpandable
\edef\temp{\noexpand\tablez #1\space\space\space}%
\expandafter
}\temp \endtablez
}
\def\tablez #1 #2 #3 #4\endtablez{%
\aboveenvbreak
\ifnum 0#1>0 \advance \leftskip by #1\mil \fi
\ifnum 0#2>0 \tableindent=#2\mil \fi
\ifnum 0#3>0 \advance \rightskip by #3\mil \fi
\itemmax=\tableindent
\advance \itemmax by -\itemmargin
\advance \leftskip by \tableindent
\exdentamount=\tableindent
\parindent = 0pt
\parskip = \smallskipamount
\ifdim \parskip=0pt \parskip=2pt \fi
\let\item = \internalBitem
\let\itemx = \internalBitemx
}
\def\Etable{\endgraf\afterenvbreak}
\let\Eftable\Etable
\let\Evtable\Etable
\let\Eitemize\Etable
\let\Eenumerate\Etable
% This is the counter used by @enumerate, which is really @itemize
\newcount \itemno
\envdef\itemize{\parsearg\doitemize}
\def\doitemize#1{%
\aboveenvbreak
\itemmax=\itemindent
\advance\itemmax by -\itemmargin
\advance\leftskip by \itemindent
\exdentamount=\itemindent
\parindent=0pt
\parskip=\smallskipamount
\ifdim\parskip=0pt \parskip=2pt \fi
\def\itemcontents{#1}%
% @itemize with no arg is equivalent to @itemize @bullet.
\ifx\itemcontents\empty\def\itemcontents{\bullet}\fi
\let\item=\itemizeitem
}
% Definition of @item while inside @itemize and @enumerate.
%
\def\itemizeitem{%
\advance\itemno by 1 % for enumerations
{\let\par=\endgraf \smallbreak}% reasonable place to break
{%
% If the document has an @itemize directly after a section title, a
% \nobreak will be last on the list, and \sectionheading will have
% done a \vskip-\parskip. In that case, we don't want to zero
% parskip, or the item text will crash with the heading. On the
% other hand, when there is normal text preceding the item (as there
% usually is), we do want to zero parskip, or there would be too much
% space. In that case, we won't have a \nobreak before. At least
% that's the theory.
\ifnum\lastpenalty<10000 \parskip=0in \fi
\noindent
\hbox to 0pt{\hss \itemcontents \kern\itemmargin}%
\vadjust{\penalty 1200}}% not good to break after first line of item.
\flushcr
}
% \splitoff TOKENS\endmark defines \first to be the first token in
% TOKENS, and \rest to be the remainder.
%
\def\splitoff#1#2\endmark{\def\first{#1}\def\rest{#2}}%
% Allow an optional argument of an uppercase letter, lowercase letter,
% or number, to specify the first label in the enumerated list. No
% argument is the same as `1'.
%
\envparseargdef\enumerate{\enumeratey #1 \endenumeratey}
\def\enumeratey #1 #2\endenumeratey{%
% If we were given no argument, pretend we were given `1'.
\def\thearg{#1}%
\ifx\thearg\empty \def\thearg{1}\fi
%
% Detect if the argument is a single token. If so, it might be a
% letter. Otherwise, the only valid thing it can be is a number.
% (We will always have one token, because of the test we just made.
% This is a good thing, since \splitoff doesn't work given nothing at
% all -- the first parameter is undelimited.)
\expandafter\splitoff\thearg\endmark
\ifx\rest\empty
% Only one token in the argument. It could still be anything.
% A ``lowercase letter'' is one whose \lccode is nonzero.
% An ``uppercase letter'' is one whose \lccode is both nonzero, and
% not equal to itself.
% Otherwise, we assume it's a number.
%
% We need the \relax at the end of the \ifnum lines to stop TeX from
% continuing to look for a .
%
\ifnum\lccode\expandafter`\thearg=0\relax
\numericenumerate % a number (we hope)
\else
% It's a letter.
\ifnum\lccode\expandafter`\thearg=\expandafter`\thearg\relax
\lowercaseenumerate % lowercase letter
\else
\uppercaseenumerate % uppercase letter
\fi
\fi
\else
% Multiple tokens in the argument. We hope it's a number.
\numericenumerate
\fi
}
% An @enumerate whose labels are integers. The starting integer is
% given in \thearg.
%
\def\numericenumerate{%
\itemno = \thearg
\startenumeration{\the\itemno}%
}
% The starting (lowercase) letter is in \thearg.
\def\lowercaseenumerate{%
\itemno = \expandafter`\thearg
\startenumeration{%
% Be sure we're not beyond the end of the alphabet.
\ifnum\itemno=0
\errmessage{No more lowercase letters in @enumerate; get a bigger
alphabet}%
\fi
\char\lccode\itemno
}%
}
% The starting (uppercase) letter is in \thearg.
\def\uppercaseenumerate{%
\itemno = \expandafter`\thearg
\startenumeration{%
% Be sure we're not beyond the end of the alphabet.
\ifnum\itemno=0
\errmessage{No more uppercase letters in @enumerate; get a bigger
alphabet}
\fi
\char\uccode\itemno
}%
}
% Call \doitemize, adding a period to the first argument and supplying the
% common last two arguments. Also subtract one from the initial value in
% \itemno, since @item increments \itemno.
%
\def\startenumeration#1{%
\advance\itemno by -1
\doitemize{#1.}\flushcr
}
% @alphaenumerate and @capsenumerate are abbreviations for giving an arg
% to @enumerate.
%
\def\alphaenumerate{\enumerate{a}}
\def\capsenumerate{\enumerate{A}}
\def\Ealphaenumerate{\Eenumerate}
\def\Ecapsenumerate{\Eenumerate}
% @multitable macros
% Amy Hendrickson, 8/18/94, 3/6/96
%
% @multitable ... @end multitable will make as many columns as desired.
% Contents of each column will wrap at width given in preamble. Width
% can be specified either with sample text given in a template line,
% or in percent of \hsize, the current width of text on page.
% Table can continue over pages but will only break between lines.
% To make preamble:
%
% Either define widths of columns in terms of percent of \hsize:
% @multitable @columnfractions .25 .3 .45
% @item ...
%
% Numbers following @columnfractions are the percent of the total
% current hsize to be used for each column. You may use as many
% columns as desired.
% Or use a template:
% @multitable {Column 1 template} {Column 2 template} {Column 3 template}
% @item ...
% using the widest term desired in each column.
% Each new table line starts with @item, each subsequent new column
% starts with @tab. Empty columns may be produced by supplying @tab's
% with nothing between them for as many times as empty columns are needed,
% ie, @tab@tab@tab will produce two empty columns.
% @item, @tab do not need to be on their own lines, but it will not hurt
% if they are.
% Sample multitable:
% @multitable {Column 1 template} {Column 2 template} {Column 3 template}
% @item first col stuff @tab second col stuff @tab third col
% @item
% first col stuff
% @tab
% second col stuff
% @tab
% third col
% @item first col stuff @tab second col stuff
% @tab Many paragraphs of text may be used in any column.
%
% They will wrap at the width determined by the template.
% @item@tab@tab This will be in third column.
% @end multitable
% Default dimensions may be reset by user.
% @multitableparskip is vertical space between paragraphs in table.
% @multitableparindent is paragraph indent in table.
% @multitablecolmargin is horizontal space to be left between columns.
% @multitablelinespace is space to leave between table items, baseline
% to baseline.
% 0pt means it depends on current normal line spacing.
%
\newskip\multitableparskip
\newskip\multitableparindent
\newdimen\multitablecolspace
\newskip\multitablelinespace
\multitableparskip=0pt
\multitableparindent=6pt
\multitablecolspace=12pt
\multitablelinespace=0pt
% Macros used to set up halign preamble:
%
\let\endsetuptable\relax
\def\xendsetuptable{\endsetuptable}
\let\columnfractions\relax
\def\xcolumnfractions{\columnfractions}
\newif\ifsetpercent
% #1 is the @columnfraction, usually a decimal number like .5, but might
% be just 1. We just use it, whatever it is.
%
\def\pickupwholefraction#1 {%
\global\advance\colcount by 1
\expandafter\xdef\csname col\the\colcount\endcsname{#1\hsize}%
\setuptable
}
\newcount\colcount
\def\setuptable#1{%
\def\firstarg{#1}%
\ifx\firstarg\xendsetuptable
\let\go = \relax
\else
\ifx\firstarg\xcolumnfractions
\global\setpercenttrue
\else
\ifsetpercent
\let\go\pickupwholefraction
\else
\global\advance\colcount by 1
\setbox0=\hbox{#1\unskip\space}% Add a normal word space as a
% separator; typically that is always in the input, anyway.
\expandafter\xdef\csname col\the\colcount\endcsname{\the\wd0}%
\fi
\fi
\ifx\go\pickupwholefraction
% Put the argument back for the \pickupwholefraction call, so
% we'll always have a period there to be parsed.
\def\go{\pickupwholefraction#1}%
\else
\let\go = \setuptable
\fi%
\fi
\go
}
% multitable-only commands.
%
% @headitem starts a heading row, which we typeset in bold.
% Assignments have to be global since we are inside the implicit group
% of an alignment entry. Note that \everycr resets \everytab.
\def\headitem{\checkenv\multitable \crcr \global\everytab={\bf}\the\everytab}%
%
% A \tab used to include \hskip1sp. But then the space in a template
% line is not enough. That is bad. So let's go back to just `&' until
% we encounter the problem it was intended to solve again.
% --karl, nathan@acm.org, 20apr99.
\def\tab{\checkenv\multitable &\the\everytab}%
% @multitable ... @end multitable definitions:
%
\newtoks\everytab % insert after every tab.
%
\envdef\multitable{%
\vskip\parskip
\startsavinginserts
%
% @item within a multitable starts a normal row.
% We use \def instead of \let so that if one of the multitable entries
% contains an @itemize, we don't choke on the \item (seen as \crcr aka
% \endtemplate) expanding \doitemize.
\def\item{\crcr}%
%
\tolerance=9500
\hbadness=9500
\setmultitablespacing
\parskip=\multitableparskip
\parindent=\multitableparindent
\overfullrule=0pt
\global\colcount=0
%
\everycr = {%
\noalign{%
\global\everytab={}%
\global\colcount=0 % Reset the column counter.
% Check for saved footnotes, etc.
\checkinserts
% Keeps underfull box messages off when table breaks over pages.
%\filbreak
% Maybe so, but it also creates really weird page breaks when the
% table breaks over pages. Wouldn't \vfil be better? Wait until the
% problem manifests itself, so it can be fixed for real --karl.
}%
}%
%
\parsearg\domultitable
}
\def\domultitable#1{%
% To parse everything between @multitable and @item:
\setuptable#1 \endsetuptable
%
% This preamble sets up a generic column definition, which will
% be used as many times as user calls for columns.
% \vtop will set a single line and will also let text wrap and
% continue for many paragraphs if desired.
\halign\bgroup &%
\global\advance\colcount by 1
\multistrut
\vtop{%
% Use the current \colcount to find the correct column width:
\hsize=\expandafter\csname col\the\colcount\endcsname
%
% In order to keep entries from bumping into each other
% we will add a \leftskip of \multitablecolspace to all columns after
% the first one.
%
% If a template has been used, we will add \multitablecolspace
% to the width of each template entry.
%
% If the user has set preamble in terms of percent of \hsize we will
% use that dimension as the width of the column, and the \leftskip
% will keep entries from bumping into each other. Table will start at
% left margin and final column will justify at right margin.
%
% Make sure we don't inherit \rightskip from the outer environment.
\rightskip=0pt
\ifnum\colcount=1
% The first column will be indented with the surrounding text.
\advance\hsize by\leftskip
\else
\ifsetpercent \else
% If user has not set preamble in terms of percent of \hsize
% we will advance \hsize by \multitablecolspace.
\advance\hsize by \multitablecolspace
\fi
% In either case we will make \leftskip=\multitablecolspace:
\leftskip=\multitablecolspace
\fi
% Ignoring space at the beginning and end avoids an occasional spurious
% blank line, when TeX decides to break the line at the space before the
% box from the multistrut, so the strut ends up on a line by itself.
% For example:
% @multitable @columnfractions .11 .89
% @item @code{#}
% @tab Legal holiday which is valid in major parts of the whole country.
% Is automatically provided with highlighting sequences respectively
% marking characters.
\noindent\ignorespaces##\unskip\multistrut
}\cr
}
\def\Emultitable{%
\crcr
\egroup % end the \halign
\global\setpercentfalse
}
\def\setmultitablespacing{%
\def\multistrut{\strut}% just use the standard line spacing
%
% Compute \multitablelinespace (if not defined by user) for use in
% \multitableparskip calculation. We used define \multistrut based on
% this, but (ironically) that caused the spacing to be off.
% See bug-texinfo report from Werner Lemberg, 31 Oct 2004 12:52:20 +0100.
\ifdim\multitablelinespace=0pt
\setbox0=\vbox{X}\global\multitablelinespace=\the\baselineskip
\global\advance\multitablelinespace by-\ht0
\fi
%% Test to see if parskip is larger than space between lines of
%% table. If not, do nothing.
%% If so, set to same dimension as multitablelinespace.
\ifdim\multitableparskip>\multitablelinespace
\global\multitableparskip=\multitablelinespace
\global\advance\multitableparskip-7pt %% to keep parskip somewhat smaller
%% than skip between lines in the table.
\fi%
\ifdim\multitableparskip=0pt
\global\multitableparskip=\multitablelinespace
\global\advance\multitableparskip-7pt %% to keep parskip somewhat smaller
%% than skip between lines in the table.
\fi}
\message{conditionals,}
% @iftex, @ifnotdocbook, @ifnothtml, @ifnotinfo, @ifnotplaintext,
% @ifnotxml always succeed. They currently do nothing; we don't
% attempt to check whether the conditionals are properly nested. But we
% have to remember that they are conditionals, so that @end doesn't
% attempt to close an environment group.
%
\def\makecond#1{%
\expandafter\let\csname #1\endcsname = \relax
\expandafter\let\csname iscond.#1\endcsname = 1
}
\makecond{iftex}
\makecond{ifnotdocbook}
\makecond{ifnothtml}
\makecond{ifnotinfo}
\makecond{ifnotplaintext}
\makecond{ifnotxml}
% Ignore @ignore, @ifhtml, @ifinfo, and the like.
%
\def\direntry{\doignore{direntry}}
\def\documentdescription{\doignore{documentdescription}}
\def\docbook{\doignore{docbook}}
\def\html{\doignore{html}}
\def\ifdocbook{\doignore{ifdocbook}}
\def\ifhtml{\doignore{ifhtml}}
\def\ifinfo{\doignore{ifinfo}}
\def\ifnottex{\doignore{ifnottex}}
\def\ifplaintext{\doignore{ifplaintext}}
\def\ifxml{\doignore{ifxml}}
\def\ignore{\doignore{ignore}}
\def\menu{\doignore{menu}}
\def\xml{\doignore{xml}}
% Ignore text until a line `@end #1', keeping track of nested conditionals.
%
% A count to remember the depth of nesting.
\newcount\doignorecount
\def\doignore#1{\begingroup
% Scan in ``verbatim'' mode:
\obeylines
\catcode`\@ = \other
\catcode`\{ = \other
\catcode`\} = \other
%
% Make sure that spaces turn into tokens that match what \doignoretext wants.
\spaceisspace
%
% Count number of #1's that we've seen.
\doignorecount = 0
%
% Swallow text until we reach the matching `@end #1'.
\dodoignore{#1}%
}
{ \catcode`_=11 % We want to use \_STOP_ which cannot appear in texinfo source.
\obeylines %
%
\gdef\dodoignore#1{%
% #1 contains the command name as a string, e.g., `ifinfo'.
%
% Define a command to find the next `@end #1'.
\long\def\doignoretext##1^^M@end #1{%
\doignoretextyyy##1^^M@#1\_STOP_}%
%
% And this command to find another #1 command, at the beginning of a
% line. (Otherwise, we would consider a line `@c @ifset', for
% example, to count as an @ifset for nesting.)
\long\def\doignoretextyyy##1^^M@#1##2\_STOP_{\doignoreyyy{##2}\_STOP_}%
%
% And now expand that command.
\doignoretext ^^M%
}%
}
\def\doignoreyyy#1{%
\def\temp{#1}%
\ifx\temp\empty % Nothing found.
\let\next\doignoretextzzz
\else % Found a nested condition, ...
\advance\doignorecount by 1
\let\next\doignoretextyyy % ..., look for another.
% If we're here, #1 ends with ^^M\ifinfo (for example).
\fi
\next #1% the token \_STOP_ is present just after this macro.
}
% We have to swallow the remaining "\_STOP_".
%
\def\doignoretextzzz#1{%
\ifnum\doignorecount = 0 % We have just found the outermost @end.
\let\next\enddoignore
\else % Still inside a nested condition.
\advance\doignorecount by -1
\let\next\doignoretext % Look for the next @end.
\fi
\next
}
% Finish off ignored text.
{ \obeylines%
% Ignore anything after the last `@end #1'; this matters in verbatim
% environments, where otherwise the newline after an ignored conditional
% would result in a blank line in the output.
\gdef\enddoignore#1^^M{\endgroup\ignorespaces}%
}
% @set VAR sets the variable VAR to an empty value.
% @set VAR REST-OF-LINE sets VAR to the value REST-OF-LINE.
%
% Since we want to separate VAR from REST-OF-LINE (which might be
% empty), we can't just use \parsearg; we have to insert a space of our
% own to delimit the rest of the line, and then take it out again if we
% didn't need it.
% We rely on the fact that \parsearg sets \catcode`\ =10.
%
\parseargdef\set{\setyyy#1 \endsetyyy}
\def\setyyy#1 #2\endsetyyy{%
{%
\makevalueexpandable
\def\temp{#2}%
\edef\next{\gdef\makecsname{SET#1}}%
\ifx\temp\empty
\next{}%
\else
\setzzz#2\endsetzzz
\fi
}%
}
% Remove the trailing space \setxxx inserted.
\def\setzzz#1 \endsetzzz{\next{#1}}
% @clear VAR clears (i.e., unsets) the variable VAR.
%
\parseargdef\clear{%
{%
\makevalueexpandable
\global\expandafter\let\csname SET#1\endcsname=\relax
}%
}
% @value{foo} gets the text saved in variable foo.
\def\value{\begingroup\makevalueexpandable\valuexxx}
\def\valuexxx#1{\expandablevalue{#1}\endgroup}
{
\catcode`\- = \active \catcode`\_ = \active
%
\gdef\makevalueexpandable{%
\let\value = \expandablevalue
% We don't want these characters active, ...
\catcode`\-=\other \catcode`\_=\other
% ..., but we might end up with active ones in the argument if
% we're called from @code, as @code{@value{foo-bar_}}, though.
% So \let them to their normal equivalents.
\let-\realdash \let_\normalunderscore
}
}
% We have this subroutine so that we can handle at least some @value's
% properly in indexes (we call \makevalueexpandable in \indexdummies).
% The command has to be fully expandable (if the variable is set), since
% the result winds up in the index file. This means that if the
% variable's value contains other Texinfo commands, it's almost certain
% it will fail (although perhaps we could fix that with sufficient work
% to do a one-level expansion on the result, instead of complete).
%
\def\expandablevalue#1{%
\expandafter\ifx\csname SET#1\endcsname\relax
{[No value for ``#1'']}%
\message{Variable `#1', used in @value, is not set.}%
\else
\csname SET#1\endcsname
\fi
}
% @ifset VAR ... @end ifset reads the `...' iff VAR has been defined
% with @set.
%
% To get special treatment of `@end ifset,' call \makeond and the redefine.
%
\makecond{ifset}
\def\ifset{\parsearg{\doifset{\let\next=\ifsetfail}}}
\def\doifset#1#2{%
{%
\makevalueexpandable
\let\next=\empty
\expandafter\ifx\csname SET#2\endcsname\relax
#1% If not set, redefine \next.
\fi
\expandafter
}\next
}
\def\ifsetfail{\doignore{ifset}}
% @ifclear VAR ... @end ifclear reads the `...' iff VAR has never been
% defined with @set, or has been undefined with @clear.
%
% The `\else' inside the `\doifset' parameter is a trick to reuse the
% above code: if the variable is not set, do nothing, if it is set,
% then redefine \next to \ifclearfail.
%
\makecond{ifclear}
\def\ifclear{\parsearg{\doifset{\else \let\next=\ifclearfail}}}
\def\ifclearfail{\doignore{ifclear}}
% @dircategory CATEGORY -- specify a category of the dir file
% which this file should belong to. Ignore this in TeX.
\let\dircategory=\comment
% @defininfoenclose.
\let\definfoenclose=\comment
\message{indexing,}
% Index generation facilities
% Define \newwrite to be identical to plain tex's \newwrite
% except not \outer, so it can be used within macros and \if's.
\edef\newwrite{\makecsname{ptexnewwrite}}
% \newindex {foo} defines an index named foo.
% It automatically defines \fooindex such that
% \fooindex ...rest of line... puts an entry in the index foo.
% It also defines \fooindfile to be the number of the output channel for
% the file that accumulates this index. The file's extension is foo.
% The name of an index should be no more than 2 characters long
% for the sake of vms.
%
\def\newindex#1{%
\iflinks
\expandafter\newwrite \csname#1indfile\endcsname
\openout \csname#1indfile\endcsname \jobname.#1 % Open the file
\fi
\expandafter\xdef\csname#1index\endcsname{% % Define @#1index
\noexpand\doindex{#1}}
}
% @defindex foo == \newindex{foo}
%
\def\defindex{\parsearg\newindex}
% Define @defcodeindex, like @defindex except put all entries in @code.
%
\def\defcodeindex{\parsearg\newcodeindex}
%
\def\newcodeindex#1{%
\iflinks
\expandafter\newwrite \csname#1indfile\endcsname
\openout \csname#1indfile\endcsname \jobname.#1
\fi
\expandafter\xdef\csname#1index\endcsname{%
\noexpand\docodeindex{#1}}%
}
% @synindex foo bar makes index foo feed into index bar.
% Do this instead of @defindex foo if you don't want it as a separate index.
%
% @syncodeindex foo bar similar, but put all entries made for index foo
% inside @code.
%
\def\synindex#1 #2 {\dosynindex\doindex{#1}{#2}}
\def\syncodeindex#1 #2 {\dosynindex\docodeindex{#1}{#2}}
% #1 is \doindex or \docodeindex, #2 the index getting redefined (foo),
% #3 the target index (bar).
\def\dosynindex#1#2#3{%
% Only do \closeout if we haven't already done it, else we'll end up
% closing the target index.
\expandafter \ifx\csname donesynindex#2\endcsname \undefined
% The \closeout helps reduce unnecessary open files; the limit on the
% Acorn RISC OS is a mere 16 files.
\expandafter\closeout\csname#2indfile\endcsname
\expandafter\let\csname\donesynindex#2\endcsname = 1
\fi
% redefine \fooindfile:
\expandafter\let\expandafter\temp\expandafter=\csname#3indfile\endcsname
\expandafter\let\csname#2indfile\endcsname=\temp
% redefine \fooindex:
\expandafter\xdef\csname#2index\endcsname{\noexpand#1{#3}}%
}
% Define \doindex, the driver for all \fooindex macros.
% Argument #1 is generated by the calling \fooindex macro,
% and it is "foo", the name of the index.
% \doindex just uses \parsearg; it calls \doind for the actual work.
% This is because \doind is more useful to call from other macros.
% There is also \dosubind {index}{topic}{subtopic}
% which makes an entry in a two-level index such as the operation index.
\def\doindex#1{\edef\indexname{#1}\parsearg\singleindexer}
\def\singleindexer #1{\doind{\indexname}{#1}}
% like the previous two, but they put @code around the argument.
\def\docodeindex#1{\edef\indexname{#1}\parsearg\singlecodeindexer}
\def\singlecodeindexer #1{\doind{\indexname}{\code{#1}}}
% Take care of Texinfo commands that can appear in an index entry.
% Since there are some commands we want to expand, and others we don't,
% we have to laboriously prevent expansion for those that we don't.
%
\def\indexdummies{%
\escapechar = `\\ % use backslash in output files.
\def\@{@}% change to @@ when we switch to @ as escape char in index files.
\def\ {\realbackslash\space }%
%
% Need these in case \tex is in effect and \{ is a \delimiter again.
% But can't use \lbracecmd and \rbracecmd because texindex assumes
% braces and backslashes are used only as delimiters.
\let\{ = \mylbrace
\let\} = \myrbrace
%
% I don't entirely understand this, but when an index entry is
% generated from a macro call, the \endinput which \scanmacro inserts
% causes processing to be prematurely terminated. This is,
% apparently, because \indexsorttmp is fully expanded, and \endinput
% is an expandable command. The redefinition below makes \endinput
% disappear altogether for that purpose -- although logging shows that
% processing continues to some further point. On the other hand, it
% seems \endinput does not hurt in the printed index arg, since that
% is still getting written without apparent harm.
%
% Sample source (mac-idx3.tex, reported by Graham Percival to
% help-texinfo, 22may06):
% @macro funindex {WORD}
% @findex xyz
% @end macro
% ...
% @funindex commtest
%
% The above is not enough to reproduce the bug, but it gives the flavor.
%
% Sample whatsit resulting:
% .@write3{\entry{xyz}{@folio }{@code {xyz@endinput }}}
%
% So:
\let\endinput = \empty
%
% Do the redefinitions.
\commondummies
}
% For the aux and toc files, @ is the escape character. So we want to
% redefine everything using @ as the escape character (instead of
% \realbackslash, still used for index files). When everything uses @,
% this will be simpler.
%
\def\atdummies{%
\def\@{@@}%
\def\ {@ }%
\let\{ = \lbraceatcmd
\let\} = \rbraceatcmd
%
% Do the redefinitions.
\commondummies
\otherbackslash
}
% Called from \indexdummies and \atdummies.
%
\def\commondummies{%
%
% \definedummyword defines \#1 as \string\#1\space, thus effectively
% preventing its expansion. This is used only for control% words,
% not control letters, because the \space would be incorrect for
% control characters, but is needed to separate the control word
% from whatever follows.
%
% For control letters, we have \definedummyletter, which omits the
% space.
%
% These can be used both for control words that take an argument and
% those that do not. If it is followed by {arg} in the input, then
% that will dutifully get written to the index (or wherever).
%
\def\definedummyword ##1{\def##1{\string##1\space}}%
\def\definedummyletter##1{\def##1{\string##1}}%
\let\definedummyaccent\definedummyletter
%
\commondummiesnofonts
%
\definedummyletter\_%
%
% Non-English letters.
\definedummyword\AA
\definedummyword\AE
\definedummyword\L
\definedummyword\OE
\definedummyword\O
\definedummyword\aa
\definedummyword\ae
\definedummyword\l
\definedummyword\oe
\definedummyword\o
\definedummyword\ss
\definedummyword\exclamdown
\definedummyword\questiondown
\definedummyword\ordf
\definedummyword\ordm
%
% Although these internal commands shouldn't show up, sometimes they do.
\definedummyword\bf
\definedummyword\gtr
\definedummyword\hat
\definedummyword\less
\definedummyword\sf
\definedummyword\sl
\definedummyword\tclose
\definedummyword\tt
%
\definedummyword\LaTeX
\definedummyword\TeX
%
% Assorted special characters.
\definedummyword\bullet
\definedummyword\comma
\definedummyword\copyright
\definedummyword\registeredsymbol
\definedummyword\dots
\definedummyword\enddots
\definedummyword\equiv
\definedummyword\error
\definedummyword\euro
\definedummyword\expansion
\definedummyword\minus
\definedummyword\pounds
\definedummyword\point
\definedummyword\print
\definedummyword\result
\definedummyword\textdegree
%
% We want to disable all macros so that they are not expanded by \write.
\macrolist
%
\normalturnoffactive
%
% Handle some cases of @value -- where it does not contain any
% (non-fully-expandable) commands.
\makevalueexpandable
}
% \commondummiesnofonts: common to \commondummies and \indexnofonts.
%
\def\commondummiesnofonts{%
% Control letters and accents.
\definedummyletter\!%
\definedummyaccent\"%
\definedummyaccent\'%
\definedummyletter\*%
\definedummyaccent\,%
\definedummyletter\.%
\definedummyletter\/%
\definedummyletter\:%
\definedummyaccent\=%
\definedummyletter\?%
\definedummyaccent\^%
\definedummyaccent\`%
\definedummyaccent\~%
\definedummyword\u
\definedummyword\v
\definedummyword\H
\definedummyword\dotaccent
\definedummyword\ringaccent
\definedummyword\tieaccent
\definedummyword\ubaraccent
\definedummyword\udotaccent
\definedummyword\dotless
%
% Texinfo font commands.
\definedummyword\b
\definedummyword\i
\definedummyword\r
\definedummyword\sc
\definedummyword\t
%
% Commands that take arguments.
\definedummyword\acronym
\definedummyword\cite
\definedummyword\code
\definedummyword\command
\definedummyword\dfn
\definedummyword\emph
\definedummyword\env
\definedummyword\file
\definedummyword\kbd
\definedummyword\key
\definedummyword\math
\definedummyword\option
\definedummyword\pxref
\definedummyword\ref
\definedummyword\samp
\definedummyword\strong
\definedummyword\tie
\definedummyword\uref
\definedummyword\url
\definedummyword\var
\definedummyword\verb
\definedummyword\w
\definedummyword\xref
}
% \indexnofonts is used when outputting the strings to sort the index
% by, and when constructing control sequence names. It eliminates all
% control sequences and just writes whatever the best ASCII sort string
% would be for a given command (usually its argument).
%
\def\indexnofonts{%
% Accent commands should become @asis.
\def\definedummyaccent##1{\let##1\asis}%
% We can just ignore other control letters.
\def\definedummyletter##1{\let##1\empty}%
% Hopefully, all control words can become @asis.
\let\definedummyword\definedummyaccent
%
\commondummiesnofonts
%
% Don't no-op \tt, since it isn't a user-level command
% and is used in the definitions of the active chars like <, >, |, etc.
% Likewise with the other plain tex font commands.
%\let\tt=\asis
%
\def\ { }%
\def\@{@}%
% how to handle braces?
\def\_{\normalunderscore}%
%
% Non-English letters.
\def\AA{AA}%
\def\AE{AE}%
\def\L{L}%
\def\OE{OE}%
\def\O{O}%
\def\aa{aa}%
\def\ae{ae}%
\def\l{l}%
\def\oe{oe}%
\def\o{o}%
\def\ss{ss}%
\def\exclamdown{!}%
\def\questiondown{?}%
\def\ordf{a}%
\def\ordm{o}%
%
\def\LaTeX{LaTeX}%
\def\TeX{TeX}%
%
% Assorted special characters.
% (The following {} will end up in the sort string, but that's ok.)
\def\bullet{bullet}%
\def\comma{,}%
\def\copyright{copyright}%
\def\registeredsymbol{R}%
\def\dots{...}%
\def\enddots{...}%
\def\equiv{==}%
\def\error{error}%
\def\euro{euro}%
\def\expansion{==>}%
\def\minus{-}%
\def\pounds{pounds}%
\def\point{.}%
\def\print{-|}%
\def\result{=>}%
\def\textdegree{degrees}%
%
% We need to get rid of all macros, leaving only the arguments (if present).
% Of course this is not nearly correct, but it is the best we can do for now.
% makeinfo does not expand macros in the argument to @deffn, which ends up
% writing an index entry, and texindex isn't prepared for an index sort entry
% that starts with \.
%
% Since macro invocations are followed by braces, we can just redefine them
% to take a single TeX argument. The case of a macro invocation that
% goes to end-of-line is not handled.
%
\macrolist
}
\let\indexbackslash=0 %overridden during \printindex.
\let\SETmarginindex=\relax % put index entries in margin (undocumented)?
% Most index entries go through here, but \dosubind is the general case.
% #1 is the index name, #2 is the entry text.
\def\doind#1#2{\dosubind{#1}{#2}{}}
% Workhorse for all \fooindexes.
% #1 is name of index, #2 is stuff to put there, #3 is subentry --
% empty if called from \doind, as we usually are (the main exception
% is with most defuns, which call us directly).
%
\def\dosubind#1#2#3{%
\iflinks
{%
% Store the main index entry text (including the third arg).
\toks0 = {#2}%
% If third arg is present, precede it with a space.
\def\thirdarg{#3}%
\ifx\thirdarg\empty \else
\toks0 = \expandafter{\the\toks0 \space #3}%
\fi
%
\edef\writeto{\csname#1indfile\endcsname}%
%
\safewhatsit\dosubindwrite
}%
\fi
}
% Write the entry in \toks0 to the index file:
%
\def\dosubindwrite{%
% Put the index entry in the margin if desired.
\ifx\SETmarginindex\relax\else
\insert\margin{\hbox{\vrule height8pt depth3pt width0pt \the\toks0}}%
\fi
%
% Remember, we are within a group.
\indexdummies % Must do this here, since \bf, etc expand at this stage
\def\backslashcurfont{\indexbackslash}% \indexbackslash isn't defined now
% so it will be output as is; and it will print as backslash.
%
% Process the index entry with all font commands turned off, to
% get the string to sort by.
{\indexnofonts
\edef\temp{\the\toks0}% need full expansion
\xdef\indexsorttmp{\temp}%
}%
%
% Set up the complete index entry, with both the sort key and
% the original text, including any font commands. We write
% three arguments to \entry to the .?? file (four in the
% subentry case), texindex reduces to two when writing the .??s
% sorted result.
\edef\temp{%
\write\writeto{%
\string\entry{\indexsorttmp}{\noexpand\folio}{\the\toks0}}%
}%
\temp
}
% Take care of unwanted page breaks/skips around a whatsit:
%
% If a skip is the last thing on the list now, preserve it
% by backing up by \lastskip, doing the \write, then inserting
% the skip again. Otherwise, the whatsit generated by the
% \write or \pdfdest will make \lastskip zero. The result is that
% sequences like this:
% @end defun
% @tindex whatever
% @defun ...
% will have extra space inserted, because the \medbreak in the
% start of the @defun won't see the skip inserted by the @end of
% the previous defun.
%
% But don't do any of this if we're not in vertical mode. We
% don't want to do a \vskip and prematurely end a paragraph.
%
% Avoid page breaks due to these extra skips, too.
%
% But wait, there is a catch there:
% We'll have to check whether \lastskip is zero skip. \ifdim is not
% sufficient for this purpose, as it ignores stretch and shrink parts
% of the skip. The only way seems to be to check the textual
% representation of the skip.
%
% The following is almost like \def\zeroskipmacro{0.0pt} except that
% the ``p'' and ``t'' characters have catcode \other, not 11 (letter).
%
\edef\zeroskipmacro{\expandafter\the\csname z@skip\endcsname}
%
\newskip\whatsitskip
\newcount\whatsitpenalty
%
% ..., ready, GO:
%
\def\safewhatsit#1{%
\ifhmode
#1%
\else
% \lastskip and \lastpenalty cannot both be nonzero simultaneously.
\whatsitskip = \lastskip
\edef\lastskipmacro{\the\lastskip}%
\whatsitpenalty = \lastpenalty
%
% If \lastskip is nonzero, that means the last item was a
% skip. And since a skip is discardable, that means this
% -\skip0 glue we're inserting is preceded by a
% non-discardable item, therefore it is not a potential
% breakpoint, therefore no \nobreak needed.
\ifx\lastskipmacro\zeroskipmacro
\else
\vskip-\whatsitskip
\fi
%
#1%
%
\ifx\lastskipmacro\zeroskipmacro
% If \lastskip was zero, perhaps the last item was a penalty, and
% perhaps it was >=10000, e.g., a \nobreak. In that case, we want
% to re-insert the same penalty (values >10000 are used for various
% signals); since we just inserted a non-discardable item, any
% following glue (such as a \parskip) would be a breakpoint. For example:
%
% @deffn deffn-whatever
% @vindex index-whatever
% Description.
% would allow a break between the index-whatever whatsit
% and the "Description." paragraph.
\ifnum\whatsitpenalty>9999 \penalty\whatsitpenalty \fi
\else
% On the other hand, if we had a nonzero \lastskip,
% this make-up glue would be preceded by a non-discardable item
% (the whatsit from the \write), so we must insert a \nobreak.
\nobreak\vskip\whatsitskip
\fi
\fi
}
% The index entry written in the file actually looks like
% \entry {sortstring}{page}{topic}
% or
% \entry {sortstring}{page}{topic}{subtopic}
% The texindex program reads in these files and writes files
% containing these kinds of lines:
% \initial {c}
% before the first topic whose initial is c
% \entry {topic}{pagelist}
% for a topic that is used without subtopics
% \primary {topic}
% for the beginning of a topic that is used with subtopics
% \secondary {subtopic}{pagelist}
% for each subtopic.
% Define the user-accessible indexing commands
% @findex, @vindex, @kindex, @cindex.
\def\findex {\fnindex}
\def\kindex {\kyindex}
\def\cindex {\cpindex}
\def\vindex {\vrindex}
\def\tindex {\tpindex}
\def\pindex {\pgindex}
\def\cindexsub {\begingroup\obeylines\cindexsub}
{\obeylines %
\gdef\cindexsub "#1" #2^^M{\endgroup %
\dosubind{cp}{#2}{#1}}}
% Define the macros used in formatting output of the sorted index material.
% @printindex causes a particular index (the ??s file) to get printed.
% It does not print any chapter heading (usually an @unnumbered).
%
\parseargdef\printindex{\begingroup
\dobreak \chapheadingskip{10000}%
%
\smallfonts \rm
\tolerance = 9500
\plainfrenchspacing
\everypar = {}% don't want the \kern\-parindent from indentation suppression.
%
% See if the index file exists and is nonempty.
% Change catcode of @ here so that if the index file contains
% \initial {@}
% as its first line, TeX doesn't complain about mismatched braces
% (because it thinks @} is a control sequence).
\catcode`\@ = 11
\openin 1 \jobname.#1s
\ifeof 1
% \enddoublecolumns gets confused if there is no text in the index,
% and it loses the chapter title and the aux file entries for the
% index. The easiest way to prevent this problem is to make sure
% there is some text.
\putwordIndexNonexistent
\else
%
% If the index file exists but is empty, then \openin leaves \ifeof
% false. We have to make TeX try to read something from the file, so
% it can discover if there is anything in it.
\read 1 to \temp
\ifeof 1
\putwordIndexIsEmpty
\else
% Index files are almost Texinfo source, but we use \ as the escape
% character. It would be better to use @, but that's too big a change
% to make right now.
\def\indexbackslash{\backslashcurfont}%
\catcode`\\ = 0
\escapechar = `\\
\begindoublecolumns
\input \jobname.#1s
\enddoublecolumns
\fi
\fi
\closein 1
\endgroup}
% These macros are used by the sorted index file itself.
% Change them to control the appearance of the index.
\def\initial#1{{%
% Some minor font changes for the special characters.
\let\tentt=\sectt \let\tt=\sectt \let\sf=\sectt
%
% Remove any glue we may have, we'll be inserting our own.
\removelastskip
%
% We like breaks before the index initials, so insert a bonus.
\nobreak
\vskip 0pt plus 3\baselineskip
\penalty 0
\vskip 0pt plus -3\baselineskip
%
% Typeset the initial. Making this add up to a whole number of
% baselineskips increases the chance of the dots lining up from column
% to column. It still won't often be perfect, because of the stretch
% we need before each entry, but it's better.
%
% No shrink because it confuses \balancecolumns.
\vskip 1.67\baselineskip plus .5\baselineskip
\leftline{\secbf #1}%
% Do our best not to break after the initial.
\nobreak
\vskip .33\baselineskip plus .1\baselineskip
}}
% \entry typesets a paragraph consisting of the text (#1), dot leaders, and
% then page number (#2) flushed to the right margin. It is used for index
% and table of contents entries. The paragraph is indented by \leftskip.
%
% A straightforward implementation would start like this:
% \def\entry#1#2{...
% But this frozes the catcodes in the argument, and can cause problems to
% @code, which sets - active. This problem was fixed by a kludge---
% ``-'' was active throughout whole index, but this isn't really right.
%
% The right solution is to prevent \entry from swallowing the whole text.
% --kasal, 21nov03
\def\entry{%
\begingroup
%
% Start a new paragraph if necessary, so our assignments below can't
% affect previous text.
\par
%
% Do not fill out the last line with white space.
\parfillskip = 0in
%
% No extra space above this paragraph.
\parskip = 0in
%
% Do not prefer a separate line ending with a hyphen to fewer lines.
\finalhyphendemerits = 0
%
% \hangindent is only relevant when the entry text and page number
% don't both fit on one line. In that case, bob suggests starting the
% dots pretty far over on the line. Unfortunately, a large
% indentation looks wrong when the entry text itself is broken across
% lines. So we use a small indentation and put up with long leaders.
%
% \hangafter is reset to 1 (which is the value we want) at the start
% of each paragraph, so we need not do anything with that.
\hangindent = 2em
%
% When the entry text needs to be broken, just fill out the first line
% with blank space.
\rightskip = 0pt plus1fil
%
% A bit of stretch before each entry for the benefit of balancing
% columns.
\vskip 0pt plus1pt
%
% Swallow the left brace of the text (first parameter):
\afterassignment\doentry
\let\temp =
}
\def\doentry{%
\bgroup % Instead of the swallowed brace.
\noindent
\aftergroup\finishentry
% And now comes the text of the entry.
}
\def\finishentry#1{%
% #1 is the page number.
%
% The following is kludged to not output a line of dots in the index if
% there are no page numbers. The next person who breaks this will be
% cursed by a Unix daemon.
\def\tempa{{\rm }}%
\def\tempb{#1}%
\edef\tempc{\tempa}%
\edef\tempd{\tempb}%
\ifx\tempc\tempd
\ %
\else
%
% If we must, put the page number on a line of its own, and fill out
% this line with blank space. (The \hfil is overwhelmed with the
% fill leaders glue in \indexdotfill if the page number does fit.)
\hfil\penalty50
\null\nobreak\indexdotfill % Have leaders before the page number.
%
% The `\ ' here is removed by the implicit \unskip that TeX does as
% part of (the primitive) \par. Without it, a spurious underfull
% \hbox ensues.
\ifpdf
\pdfgettoks#1.%
\ \the\toksA
\else
\ #1%
\fi
\fi
\par
\endgroup
}
% Like plain.tex's \dotfill, except uses up at least 1 em.
\def\indexdotfill{\cleaders
\hbox{$\mathsurround=0pt \mkern1.5mu.\mkern1.5mu$}\hskip 1em plus 1fill}
\def\primary #1{\line{#1\hfil}}
\newskip\secondaryindent \secondaryindent=0.5cm
\def\secondary#1#2{{%
\parfillskip=0in
\parskip=0in
\hangindent=1in
\hangafter=1
\noindent\hskip\secondaryindent\hbox{#1}\indexdotfill
\ifpdf
\pdfgettoks#2.\ \the\toksA % The page number ends the paragraph.
\else
#2
\fi
\par
}}
% Define two-column mode, which we use to typeset indexes.
% Adapted from the TeXbook, page 416, which is to say,
% the manmac.tex format used to print the TeXbook itself.
\catcode`\@=11
\newbox\partialpage
\newdimen\doublecolumnhsize
\def\begindoublecolumns{\begingroup % ended by \enddoublecolumns
% Grab any single-column material above us.
\output = {%
%
% Here is a possibility not foreseen in manmac: if we accumulate a
% whole lot of material, we might end up calling this \output
% routine twice in a row (see the doublecol-lose test, which is
% essentially a couple of indexes with @setchapternewpage off). In
% that case we just ship out what is in \partialpage with the normal
% output routine. Generally, \partialpage will be empty when this
% runs and this will be a no-op. See the indexspread.tex test case.
\ifvoid\partialpage \else
\onepageout{\pagecontents\partialpage}%
\fi
%
\global\setbox\partialpage = \vbox{%
% Unvbox the main output page.
\unvbox\PAGE
\kern-\topskip \kern\baselineskip
}%
}%
\eject % run that output routine to set \partialpage
%
% Use the double-column output routine for subsequent pages.
\output = {\doublecolumnout}%
%
% Change the page size parameters. We could do this once outside this
% routine, in each of @smallbook, @afourpaper, and the default 8.5x11
% format, but then we repeat the same computation. Repeating a couple
% of assignments once per index is clearly meaningless for the
% execution time, so we may as well do it in one place.
%
% First we halve the line length, less a little for the gutter between
% the columns. We compute the gutter based on the line length, so it
% changes automatically with the paper format. The magic constant
% below is chosen so that the gutter has the same value (well, +-<1pt)
% as it did when we hard-coded it.
%
% We put the result in a separate register, \doublecolumhsize, so we
% can restore it in \pagesofar, after \hsize itself has (potentially)
% been clobbered.
%
\doublecolumnhsize = \hsize
\advance\doublecolumnhsize by -.04154\hsize
\divide\doublecolumnhsize by 2
\hsize = \doublecolumnhsize
%
% Double the \vsize as well. (We don't need a separate register here,
% since nobody clobbers \vsize.)
\vsize = 2\vsize
}
% The double-column output routine for all double-column pages except
% the last.
%
\def\doublecolumnout{%
\splittopskip=\topskip \splitmaxdepth=\maxdepth
% Get the available space for the double columns -- the normal
% (undoubled) page height minus any material left over from the
% previous page.
\dimen@ = \vsize
\divide\dimen@ by 2
\advance\dimen@ by -\ht\partialpage
%
% box0 will be the left-hand column, box2 the right.
\setbox0=\vsplit255 to\dimen@ \setbox2=\vsplit255 to\dimen@
\onepageout\pagesofar
\unvbox255
\penalty\outputpenalty
}
%
% Re-output the contents of the output page -- any previous material,
% followed by the two boxes we just split, in box0 and box2.
\def\pagesofar{%
\unvbox\partialpage
%
\hsize = \doublecolumnhsize
\wd0=\hsize \wd2=\hsize
\hbox to\pagewidth{\box0\hfil\box2}%
}
%
% All done with double columns.
\def\enddoublecolumns{%
% The following penalty ensures that the page builder is exercised
% _before_ we change the output routine. This is necessary in the
% following situation:
%
% The last section of the index consists only of a single entry.
% Before this section, \pagetotal is less than \pagegoal, so no
% break occurs before the last section starts. However, the last
% section, consisting of \initial and the single \entry, does not
% fit on the page and has to be broken off. Without the following
% penalty the page builder will not be exercised until \eject
% below, and by that time we'll already have changed the output
% routine to the \balancecolumns version, so the next-to-last
% double-column page will be processed with \balancecolumns, which
% is wrong: The two columns will go to the main vertical list, with
% the broken-off section in the recent contributions. As soon as
% the output routine finishes, TeX starts reconsidering the page
% break. The two columns and the broken-off section both fit on the
% page, because the two columns now take up only half of the page
% goal. When TeX sees \eject from below which follows the final
% section, it invokes the new output routine that we've set after
% \balancecolumns below; \onepageout will try to fit the two columns
% and the final section into the vbox of \pageheight (see
% \pagebody), causing an overfull box.
%
% Note that glue won't work here, because glue does not exercise the
% page builder, unlike penalties (see The TeXbook, pp. 280-281).
\penalty0
%
\output = {%
% Split the last of the double-column material. Leave it on the
% current page, no automatic page break.
\balancecolumns
%
% If we end up splitting too much material for the current page,
% though, there will be another page break right after this \output
% invocation ends. Having called \balancecolumns once, we do not
% want to call it again. Therefore, reset \output to its normal
% definition right away. (We hope \balancecolumns will never be
% called on to balance too much material, but if it is, this makes
% the output somewhat more palatable.)
\global\output = {\onepageout{\pagecontents\PAGE}}%
}%
\eject
\endgroup % started in \begindoublecolumns
%
% \pagegoal was set to the doubled \vsize above, since we restarted
% the current page. We're now back to normal single-column
% typesetting, so reset \pagegoal to the normal \vsize (after the
% \endgroup where \vsize got restored).
\pagegoal = \vsize
}
%
% Called at the end of the double column material.
\def\balancecolumns{%
\setbox0 = \vbox{\unvbox255}% like \box255 but more efficient, see p.120.
\dimen@ = \ht0
\advance\dimen@ by \topskip
\advance\dimen@ by-\baselineskip
\divide\dimen@ by 2 % target to split to
%debug\message{final 2-column material height=\the\ht0, target=\the\dimen@.}%
\splittopskip = \topskip
% Loop until we get a decent breakpoint.
{%
\vbadness = 10000
\loop
\global\setbox3 = \copy0
\global\setbox1 = \vsplit3 to \dimen@
\ifdim\ht3>\dimen@
\global\advance\dimen@ by 1pt
\repeat
}%
%debug\message{split to \the\dimen@, column heights: \the\ht1, \the\ht3.}%
\setbox0=\vbox to\dimen@{\unvbox1}%
\setbox2=\vbox to\dimen@{\unvbox3}%
%
\pagesofar
}
\catcode`\@ = \other
\message{sectioning,}
% Chapters, sections, etc.
% \unnumberedno is an oxymoron, of course. But we count the unnumbered
% sections so that we can refer to them unambiguously in the pdf
% outlines by their "section number". We avoid collisions with chapter
% numbers by starting them at 10000. (If a document ever has 10000
% chapters, we're in trouble anyway, I'm sure.)
\newcount\unnumberedno \unnumberedno = 10000
\newcount\chapno
\newcount\secno \secno=0
\newcount\subsecno \subsecno=0
\newcount\subsubsecno \subsubsecno=0
% This counter is funny since it counts through charcodes of letters A, B, ...
\newcount\appendixno \appendixno = `\@
%
% \def\appendixletter{\char\the\appendixno}
% We do the following ugly conditional instead of the above simple
% construct for the sake of pdftex, which needs the actual
% letter in the expansion, not just typeset.
%
\def\appendixletter{%
\ifnum\appendixno=`A A%
\else\ifnum\appendixno=`B B%
\else\ifnum\appendixno=`C C%
\else\ifnum\appendixno=`D D%
\else\ifnum\appendixno=`E E%
\else\ifnum\appendixno=`F F%
\else\ifnum\appendixno=`G G%
\else\ifnum\appendixno=`H H%
\else\ifnum\appendixno=`I I%
\else\ifnum\appendixno=`J J%
\else\ifnum\appendixno=`K K%
\else\ifnum\appendixno=`L L%
\else\ifnum\appendixno=`M M%
\else\ifnum\appendixno=`N N%
\else\ifnum\appendixno=`O O%
\else\ifnum\appendixno=`P P%
\else\ifnum\appendixno=`Q Q%
\else\ifnum\appendixno=`R R%
\else\ifnum\appendixno=`S S%
\else\ifnum\appendixno=`T T%
\else\ifnum\appendixno=`U U%
\else\ifnum\appendixno=`V V%
\else\ifnum\appendixno=`W W%
\else\ifnum\appendixno=`X X%
\else\ifnum\appendixno=`Y Y%
\else\ifnum\appendixno=`Z Z%
% The \the is necessary, despite appearances, because \appendixletter is
% expanded while writing the .toc file. \char\appendixno is not
% expandable, thus it is written literally, thus all appendixes come out
% with the same letter (or @) in the toc without it.
\else\char\the\appendixno
\fi\fi\fi\fi\fi\fi\fi\fi\fi\fi\fi\fi\fi
\fi\fi\fi\fi\fi\fi\fi\fi\fi\fi\fi\fi\fi}
% Each @chapter defines this as the name of the chapter.
% page headings and footings can use it. @section does likewise.
% However, they are not reliable, because we don't use marks.
\def\thischapter{}
\def\thissection{}
\newcount\absseclevel % used to calculate proper heading level
\newcount\secbase\secbase=0 % @raisesections/@lowersections modify this count
% @raisesections: treat @section as chapter, @subsection as section, etc.
\def\raisesections{\global\advance\secbase by -1}
\let\up=\raisesections % original BFox name
% @lowersections: treat @chapter as section, @section as subsection, etc.
\def\lowersections{\global\advance\secbase by 1}
\let\down=\lowersections % original BFox name
% we only have subsub.
\chardef\maxseclevel = 3
%
% A numbered section within an unnumbered changes to unnumbered too.
% To achive this, remember the "biggest" unnum. sec. we are currently in:
\chardef\unmlevel = \maxseclevel
%
% Trace whether the current chapter is an appendix or not:
% \chapheadtype is "N" or "A", unnumbered chapters are ignored.
\def\chapheadtype{N}
% Choose a heading macro
% #1 is heading type
% #2 is heading level
% #3 is text for heading
\def\genhead#1#2#3{%
% Compute the abs. sec. level:
\absseclevel=#2
\advance\absseclevel by \secbase
% Make sure \absseclevel doesn't fall outside the range:
\ifnum \absseclevel < 0
\absseclevel = 0
\else
\ifnum \absseclevel > 3
\absseclevel = 3
\fi
\fi
% The heading type:
\def\headtype{#1}%
\if \headtype U%
\ifnum \absseclevel < \unmlevel
\chardef\unmlevel = \absseclevel
\fi
\else
% Check for appendix sections:
\ifnum \absseclevel = 0
\edef\chapheadtype{\headtype}%
\else
\if \headtype A\if \chapheadtype N%
\errmessage{@appendix... within a non-appendix chapter}%
\fi\fi
\fi
% Check for numbered within unnumbered:
\ifnum \absseclevel > \unmlevel
\def\headtype{U}%
\else
\chardef\unmlevel = 3
\fi
\fi
% Now print the heading:
\if \headtype U%
\ifcase\absseclevel
\unnumberedzzz{#3}%
\or \unnumberedseczzz{#3}%
\or \unnumberedsubseczzz{#3}%
\or \unnumberedsubsubseczzz{#3}%
\fi
\else
\if \headtype A%
\ifcase\absseclevel
\appendixzzz{#3}%
\or \appendixsectionzzz{#3}%
\or \appendixsubseczzz{#3}%
\or \appendixsubsubseczzz{#3}%
\fi
\else
\ifcase\absseclevel
\chapterzzz{#3}%
\or \seczzz{#3}%
\or \numberedsubseczzz{#3}%
\or \numberedsubsubseczzz{#3}%
\fi
\fi
\fi
\suppressfirstparagraphindent
}
% an interface:
\def\numhead{\genhead N}
\def\apphead{\genhead A}
\def\unnmhead{\genhead U}
% @chapter, @appendix, @unnumbered. Increment top-level counter, reset
% all lower-level sectioning counters to zero.
%
% Also set \chaplevelprefix, which we prepend to @float sequence numbers
% (e.g., figures), q.v. By default (before any chapter), that is empty.
\let\chaplevelprefix = \empty
%
\outer\parseargdef\chapter{\numhead0{#1}} % normally numhead0 calls chapterzzz
\def\chapterzzz#1{%
% section resetting is \global in case the chapter is in a group, such
% as an @include file.
\global\secno=0 \global\subsecno=0 \global\subsubsecno=0
\global\advance\chapno by 1
%
% Used for \float.
\gdef\chaplevelprefix{\the\chapno.}%
\resetallfloatnos
%
\message{\putwordChapter\space \the\chapno}%
%
% Write the actual heading.
\chapmacro{#1}{Ynumbered}{\the\chapno}%
%
% So @section and the like are numbered underneath this chapter.
\global\let\section = \numberedsec
\global\let\subsection = \numberedsubsec
\global\let\subsubsection = \numberedsubsubsec
}
\outer\parseargdef\appendix{\apphead0{#1}} % normally apphead0 calls appendixzzz
\def\appendixzzz#1{%
\global\secno=0 \global\subsecno=0 \global\subsubsecno=0
\global\advance\appendixno by 1
\gdef\chaplevelprefix{\appendixletter.}%
\resetallfloatnos
%
\def\appendixnum{\putwordAppendix\space \appendixletter}%
\message{\appendixnum}%
%
\chapmacro{#1}{Yappendix}{\appendixletter}%
%
\global\let\section = \appendixsec
\global\let\subsection = \appendixsubsec
\global\let\subsubsection = \appendixsubsubsec
}
\outer\parseargdef\unnumbered{\unnmhead0{#1}} % normally unnmhead0 calls unnumberedzzz
\def\unnumberedzzz#1{%
\global\secno=0 \global\subsecno=0 \global\subsubsecno=0
\global\advance\unnumberedno by 1
%
% Since an unnumbered has no number, no prefix for figures.
\global\let\chaplevelprefix = \empty
\resetallfloatnos
%
% This used to be simply \message{#1}, but TeX fully expands the
% argument to \message. Therefore, if #1 contained @-commands, TeX
% expanded them. For example, in `@unnumbered The @cite{Book}', TeX
% expanded @cite (which turns out to cause errors because \cite is meant
% to be executed, not expanded).
%
% Anyway, we don't want the fully-expanded definition of @cite to appear
% as a result of the \message, we just want `@cite' itself. We use
% \the to achieve this: TeX expands \the only once,
% simply yielding the contents of . (We also do this for
% the toc entries.)
\toks0 = {#1}%
\message{(\the\toks0)}%
%
\chapmacro{#1}{Ynothing}{\the\unnumberedno}%
%
\global\let\section = \unnumberedsec
\global\let\subsection = \unnumberedsubsec
\global\let\subsubsection = \unnumberedsubsubsec
}
% @centerchap is like @unnumbered, but the heading is centered.
\outer\parseargdef\centerchap{%
% Well, we could do the following in a group, but that would break
% an assumption that \chapmacro is called at the outermost level.
% Thus we are safer this way: --kasal, 24feb04
\let\centerparametersmaybe = \centerparameters
\unnmhead0{#1}%
\let\centerparametersmaybe = \relax
}
% @top is like @unnumbered.
\let\top\unnumbered
% Sections.
\outer\parseargdef\numberedsec{\numhead1{#1}} % normally calls seczzz
\def\seczzz#1{%
\global\subsecno=0 \global\subsubsecno=0 \global\advance\secno by 1
\sectionheading{#1}{sec}{Ynumbered}{\the\chapno.\the\secno}%
}
\outer\parseargdef\appendixsection{\apphead1{#1}} % normally calls appendixsectionzzz
\def\appendixsectionzzz#1{%
\global\subsecno=0 \global\subsubsecno=0 \global\advance\secno by 1
\sectionheading{#1}{sec}{Yappendix}{\appendixletter.\the\secno}%
}
\let\appendixsec\appendixsection
\outer\parseargdef\unnumberedsec{\unnmhead1{#1}} % normally calls unnumberedseczzz
\def\unnumberedseczzz#1{%
\global\subsecno=0 \global\subsubsecno=0 \global\advance\secno by 1
\sectionheading{#1}{sec}{Ynothing}{\the\unnumberedno.\the\secno}%
}
% Subsections.
\outer\parseargdef\numberedsubsec{\numhead2{#1}} % normally calls numberedsubseczzz
\def\numberedsubseczzz#1{%
\global\subsubsecno=0 \global\advance\subsecno by 1
\sectionheading{#1}{subsec}{Ynumbered}{\the\chapno.\the\secno.\the\subsecno}%
}
\outer\parseargdef\appendixsubsec{\apphead2{#1}} % normally calls appendixsubseczzz
\def\appendixsubseczzz#1{%
\global\subsubsecno=0 \global\advance\subsecno by 1
\sectionheading{#1}{subsec}{Yappendix}%
{\appendixletter.\the\secno.\the\subsecno}%
}
\outer\parseargdef\unnumberedsubsec{\unnmhead2{#1}} %normally calls unnumberedsubseczzz
\def\unnumberedsubseczzz#1{%
\global\subsubsecno=0 \global\advance\subsecno by 1
\sectionheading{#1}{subsec}{Ynothing}%
{\the\unnumberedno.\the\secno.\the\subsecno}%
}
% Subsubsections.
\outer\parseargdef\numberedsubsubsec{\numhead3{#1}} % normally numberedsubsubseczzz
\def\numberedsubsubseczzz#1{%
\global\advance\subsubsecno by 1
\sectionheading{#1}{subsubsec}{Ynumbered}%
{\the\chapno.\the\secno.\the\subsecno.\the\subsubsecno}%
}
\outer\parseargdef\appendixsubsubsec{\apphead3{#1}} % normally appendixsubsubseczzz
\def\appendixsubsubseczzz#1{%
\global\advance\subsubsecno by 1
\sectionheading{#1}{subsubsec}{Yappendix}%
{\appendixletter.\the\secno.\the\subsecno.\the\subsubsecno}%
}
\outer\parseargdef\unnumberedsubsubsec{\unnmhead3{#1}} %normally unnumberedsubsubseczzz
\def\unnumberedsubsubseczzz#1{%
\global\advance\subsubsecno by 1
\sectionheading{#1}{subsubsec}{Ynothing}%
{\the\unnumberedno.\the\secno.\the\subsecno.\the\subsubsecno}%
}
% These macros control what the section commands do, according
% to what kind of chapter we are in (ordinary, appendix, or unnumbered).
% Define them by default for a numbered chapter.
\let\section = \numberedsec
\let\subsection = \numberedsubsec
\let\subsubsection = \numberedsubsubsec
% Define @majorheading, @heading and @subheading
% NOTE on use of \vbox for chapter headings, section headings, and such:
% 1) We use \vbox rather than the earlier \line to permit
% overlong headings to fold.
% 2) \hyphenpenalty is set to 10000 because hyphenation in a
% heading is obnoxious; this forbids it.
% 3) Likewise, headings look best if no \parindent is used, and
% if justification is not attempted. Hence \raggedright.
\def\majorheading{%
{\advance\chapheadingskip by 10pt \chapbreak }%
\parsearg\chapheadingzzz
}
\def\chapheading{\chapbreak \parsearg\chapheadingzzz}
\def\chapheadingzzz#1{%
{\chapfonts \vbox{\hyphenpenalty=10000\tolerance=5000
\parindent=0pt\raggedright
\rm #1\hfill}}%
\bigskip \par\penalty 200\relax
\suppressfirstparagraphindent
}
% @heading, @subheading, @subsubheading.
\parseargdef\heading{\sectionheading{#1}{sec}{Yomitfromtoc}{}
\suppressfirstparagraphindent}
\parseargdef\subheading{\sectionheading{#1}{subsec}{Yomitfromtoc}{}
\suppressfirstparagraphindent}
\parseargdef\subsubheading{\sectionheading{#1}{subsubsec}{Yomitfromtoc}{}
\suppressfirstparagraphindent}
% These macros generate a chapter, section, etc. heading only
% (including whitespace, linebreaking, etc. around it),
% given all the information in convenient, parsed form.
%%% Args are the skip and penalty (usually negative)
\def\dobreak#1#2{\par\ifdim\lastskip<#1\removelastskip\penalty#2\vskip#1\fi}
%%% Define plain chapter starts, and page on/off switching for it
% Parameter controlling skip before chapter headings (if needed)
\newskip\chapheadingskip
\def\chapbreak{\dobreak \chapheadingskip {-4000}}
\def\chappager{\par\vfill\supereject}
\def\chapoddpage{\chappager \ifodd\pageno \else \hbox to 0pt{} \chappager\fi}
\def\setchapternewpage #1 {\csname CHAPPAG#1\endcsname}
\def\CHAPPAGoff{%
\global\let\contentsalignmacro = \chappager
\global\let\pchapsepmacro=\chapbreak
\global\let\pagealignmacro=\chappager}
\def\CHAPPAGon{%
\global\let\contentsalignmacro = \chappager
\global\let\pchapsepmacro=\chappager
\global\let\pagealignmacro=\chappager
\global\def\HEADINGSon{\HEADINGSsingle}}
\def\CHAPPAGodd{%
\global\let\contentsalignmacro = \chapoddpage
\global\let\pchapsepmacro=\chapoddpage
\global\let\pagealignmacro=\chapoddpage
\global\def\HEADINGSon{\HEADINGSdouble}}
\CHAPPAGon
% Chapter opening.
%
% #1 is the text, #2 is the section type (Ynumbered, Ynothing,
% Yappendix, Yomitfromtoc), #3 the chapter number.
%
% To test against our argument.
\def\Ynothingkeyword{Ynothing}
\def\Yomitfromtockeyword{Yomitfromtoc}
\def\Yappendixkeyword{Yappendix}
%
\def\chapmacro#1#2#3{%
\pchapsepmacro
{%
\chapfonts \rm
%
% Have to define \thissection before calling \donoderef, because the
% xref code eventually uses it. On the other hand, it has to be called
% after \pchapsepmacro, or the headline will change too soon.
\gdef\thissection{#1}%
\gdef\thischaptername{#1}%
%
% Only insert the separating space if we have a chapter/appendix
% number, and don't print the unnumbered ``number''.
\def\temptype{#2}%
\ifx\temptype\Ynothingkeyword
\setbox0 = \hbox{}%
\def\toctype{unnchap}%
\gdef\thischapternum{}%
\gdef\thischapter{#1}%
\else\ifx\temptype\Yomitfromtockeyword
\setbox0 = \hbox{}% contents like unnumbered, but no toc entry
\def\toctype{omit}%
\gdef\thischapternum{}%
\gdef\thischapter{}%
\else\ifx\temptype\Yappendixkeyword
\setbox0 = \hbox{\putwordAppendix{} #3\enspace}%
\def\toctype{app}%
\xdef\thischapternum{\appendixletter}%
% We don't substitute the actual chapter name into \thischapter
% because we don't want its macros evaluated now. And we don't
% use \thissection because that changes with each section.
%
\xdef\thischapter{\putwordAppendix{} \appendixletter:
\noexpand\thischaptername}%
\else
\setbox0 = \hbox{#3\enspace}%
\def\toctype{numchap}%
\xdef\thischapternum{\the\chapno}%
\xdef\thischapter{\putwordChapter{} \the\chapno:
\noexpand\thischaptername}%
\fi\fi\fi
%
% Write the toc entry for this chapter. Must come before the
% \donoderef, because we include the current node name in the toc
% entry, and \donoderef resets it to empty.
\writetocentry{\toctype}{#1}{#3}%
%
% For pdftex, we have to write out the node definition (aka, make
% the pdfdest) after any page break, but before the actual text has
% been typeset. If the destination for the pdf outline is after the
% text, then jumping from the outline may wind up with the text not
% being visible, for instance under high magnification.
\donoderef{#2}%
%
% Typeset the actual heading.
\vbox{\hyphenpenalty=10000 \tolerance=5000 \parindent=0pt \raggedright
\hangindent=\wd0 \centerparametersmaybe
\unhbox0 #1\par}%
}%
\nobreak\bigskip % no page break after a chapter title
\nobreak
}
% @centerchap -- centered and unnumbered.
\let\centerparametersmaybe = \relax
\def\centerparameters{%
\advance\rightskip by 3\rightskip
\leftskip = \rightskip
\parfillskip = 0pt
}
% I don't think this chapter style is supported any more, so I'm not
% updating it with the new noderef stuff. We'll see. --karl, 11aug03.
%
\def\setchapterstyle #1 {\csname CHAPF#1\endcsname}
%
\def\unnchfopen #1{%
\chapoddpage {\chapfonts \vbox{\hyphenpenalty=10000\tolerance=5000
\parindent=0pt\raggedright
\rm #1\hfill}}\bigskip \par\nobreak
}
\def\chfopen #1#2{\chapoddpage {\chapfonts
\vbox to 3in{\vfil \hbox to\hsize{\hfil #2} \hbox to\hsize{\hfil #1} \vfil}}%
\par\penalty 5000 %
}
\def\centerchfopen #1{%
\chapoddpage {\chapfonts \vbox{\hyphenpenalty=10000\tolerance=5000
\parindent=0pt
\hfill {\rm #1}\hfill}}\bigskip \par\nobreak
}
\def\CHAPFopen{%
\global\let\chapmacro=\chfopen
\global\let\centerchapmacro=\centerchfopen}
% Section titles. These macros combine the section number parts and
% call the generic \sectionheading to do the printing.
%
\newskip\secheadingskip
\def\secheadingbreak{\dobreak \secheadingskip{-1000}}
% Subsection titles.
\newskip\subsecheadingskip
\def\subsecheadingbreak{\dobreak \subsecheadingskip{-500}}
% Subsubsection titles.
\def\subsubsecheadingskip{\subsecheadingskip}
\def\subsubsecheadingbreak{\subsecheadingbreak}
% Print any size, any type, section title.
%
% #1 is the text, #2 is the section level (sec/subsec/subsubsec), #3 is
% the section type for xrefs (Ynumbered, Ynothing, Yappendix), #4 is the
% section number.
%
\def\sectionheading#1#2#3#4{%
{%
% Switch to the right set of fonts.
\csname #2fonts\endcsname \rm
%
% Insert space above the heading.
\csname #2headingbreak\endcsname
%
% Only insert the space after the number if we have a section number.
\def\sectionlevel{#2}%
\def\temptype{#3}%
%
\ifx\temptype\Ynothingkeyword
\setbox0 = \hbox{}%
\def\toctype{unn}%
\gdef\thissection{#1}%
\else\ifx\temptype\Yomitfromtockeyword
% for @headings -- no section number, don't include in toc,
% and don't redefine \thissection.
\setbox0 = \hbox{}%
\def\toctype{omit}%
\let\sectionlevel=\empty
\else\ifx\temptype\Yappendixkeyword
\setbox0 = \hbox{#4\enspace}%
\def\toctype{app}%
\gdef\thissection{#1}%
\else
\setbox0 = \hbox{#4\enspace}%
\def\toctype{num}%
\gdef\thissection{#1}%
\fi\fi\fi
%
% Write the toc entry (before \donoderef). See comments in \chapmacro.
\writetocentry{\toctype\sectionlevel}{#1}{#4}%
%
% Write the node reference (= pdf destination for pdftex).
% Again, see comments in \chapmacro.
\donoderef{#3}%
%
% Interline glue will be inserted when the vbox is completed.
% That glue will be a valid breakpoint for the page, since it'll be
% preceded by a whatsit (usually from the \donoderef, or from the
% \writetocentry if there was no node). We don't want to allow that
% break, since then the whatsits could end up on page n while the
% section is on page n+1, thus toc/etc. are wrong. Debian bug 276000.
\nobreak
%
% Output the actual section heading.
\vbox{\hyphenpenalty=10000 \tolerance=5000 \parindent=0pt \raggedright
\hangindent=\wd0 % zero if no section number
\unhbox0 #1}%
}%
% Add extra space after the heading -- half of whatever came above it.
% Don't allow stretch, though.
\kern .5 \csname #2headingskip\endcsname
%
% Do not let the kern be a potential breakpoint, as it would be if it
% was followed by glue.
\nobreak
%
% We'll almost certainly start a paragraph next, so don't let that
% glue accumulate. (Not a breakpoint because it's preceded by a
% discardable item.)
\vskip-\parskip
%
% This is purely so the last item on the list is a known \penalty >
% 10000. This is so \startdefun can avoid allowing breakpoints after
% section headings. Otherwise, it would insert a valid breakpoint between:
%
% @section sec-whatever
% @deffn def-whatever
\penalty 10001
}
\message{toc,}
% Table of contents.
\newwrite\tocfile
% Write an entry to the toc file, opening it if necessary.
% Called from @chapter, etc.
%
% Example usage: \writetocentry{sec}{Section Name}{\the\chapno.\the\secno}
% We append the current node name (if any) and page number as additional
% arguments for the \{chap,sec,...}entry macros which will eventually
% read this. The node name is used in the pdf outlines as the
% destination to jump to.
%
% We open the .toc file for writing here instead of at @setfilename (or
% any other fixed time) so that @contents can be anywhere in the document.
% But if #1 is `omit', then we don't do anything. This is used for the
% table of contents chapter openings themselves.
%
\newif\iftocfileopened
\def\omitkeyword{omit}%
%
\def\writetocentry#1#2#3{%
\edef\writetoctype{#1}%
\ifx\writetoctype\omitkeyword \else
\iftocfileopened\else
\immediate\openout\tocfile = \jobname.toc
\global\tocfileopenedtrue
\fi
%
\iflinks
{\atdummies
\edef\temp{%
\write\tocfile{@#1entry{#2}{#3}{\lastnode}{\noexpand\folio}}}%
\temp
}%
\fi
\fi
%
% Tell \shipout to create a pdf destination on each page, if we're
% writing pdf. These are used in the table of contents. We can't
% just write one on every page because the title pages are numbered
% 1 and 2 (the page numbers aren't printed), and so are the first
% two pages of the document. Thus, we'd have two destinations named
% `1', and two named `2'.
\ifpdf \global\pdfmakepagedesttrue \fi
}
% These characters do not print properly in the Computer Modern roman
% fonts, so we must take special care. This is more or less redundant
% with the Texinfo input format setup at the end of this file.
%
\def\activecatcodes{%
\catcode`\"=\active
\catcode`\$=\active
\catcode`\<=\active
\catcode`\>=\active
\catcode`\\=\active
\catcode`\^=\active
\catcode`\_=\active
\catcode`\|=\active
\catcode`\~=\active
}
% Read the toc file, which is essentially Texinfo input.
\def\readtocfile{%
\setupdatafile
\activecatcodes
\input \jobname.toc
}
\newskip\contentsrightmargin \contentsrightmargin=1in
\newcount\savepageno
\newcount\lastnegativepageno \lastnegativepageno = -1
% Prepare to read what we've written to \tocfile.
%
\def\startcontents#1{%
% If @setchapternewpage on, and @headings double, the contents should
% start on an odd page, unlike chapters. Thus, we maintain
% \contentsalignmacro in parallel with \pagealignmacro.
% From: Torbjorn Granlund
\contentsalignmacro
\immediate\closeout\tocfile
%
% Don't need to put `Contents' or `Short Contents' in the headline.
% It is abundantly clear what they are.
\def\thischapter{}%
\chapmacro{#1}{Yomitfromtoc}{}%
%
\savepageno = \pageno
\begingroup % Set up to handle contents files properly.
\raggedbottom % Worry more about breakpoints than the bottom.
\advance\hsize by -\contentsrightmargin % Don't use the full line length.
%
% Roman numerals for page numbers.
\ifnum \pageno>0 \global\pageno = \lastnegativepageno \fi
}
% Normal (long) toc.
\def\contents{%
\startcontents{\putwordTOC}%
\openin 1 \jobname.toc
\ifeof 1 \else
\readtocfile
\fi
\vfill \eject
\contentsalignmacro % in case @setchapternewpage odd is in effect
\ifeof 1 \else
\pdfmakeoutlines
\fi
\closein 1
\endgroup
\lastnegativepageno = \pageno
\global\pageno = \savepageno
}
% And just the chapters.
\def\summarycontents{%
\startcontents{\putwordShortTOC}%
%
\let\numchapentry = \shortchapentry
\let\appentry = \shortchapentry
\let\unnchapentry = \shortunnchapentry
% We want a true roman here for the page numbers.
\secfonts
\let\rm=\shortcontrm \let\bf=\shortcontbf
\let\sl=\shortcontsl \let\tt=\shortconttt
\rm
\hyphenpenalty = 10000
\advance\baselineskip by 1pt % Open it up a little.
\def\numsecentry##1##2##3##4{}
\let\appsecentry = \numsecentry
\let\unnsecentry = \numsecentry
\let\numsubsecentry = \numsecentry
\let\appsubsecentry = \numsecentry
\let\unnsubsecentry = \numsecentry
\let\numsubsubsecentry = \numsecentry
\let\appsubsubsecentry = \numsecentry
\let\unnsubsubsecentry = \numsecentry
\openin 1 \jobname.toc
\ifeof 1 \else
\readtocfile
\fi
\closein 1
\vfill \eject
\contentsalignmacro % in case @setchapternewpage odd is in effect
\endgroup
\lastnegativepageno = \pageno
\global\pageno = \savepageno
}
\let\shortcontents = \summarycontents
% Typeset the label for a chapter or appendix for the short contents.
% The arg is, e.g., `A' for an appendix, or `3' for a chapter.
%
\def\shortchaplabel#1{%
% This space should be enough, since a single number is .5em, and the
% widest letter (M) is 1em, at least in the Computer Modern fonts.
% But use \hss just in case.
% (This space doesn't include the extra space that gets added after
% the label; that gets put in by \shortchapentry above.)
%
% We'd like to right-justify chapter numbers, but that looks strange
% with appendix letters. And right-justifying numbers and
% left-justifying letters looks strange when there is less than 10
% chapters. Have to read the whole toc once to know how many chapters
% there are before deciding ...
\hbox to 1em{#1\hss}%
}
% These macros generate individual entries in the table of contents.
% The first argument is the chapter or section name.
% The last argument is the page number.
% The arguments in between are the chapter number, section number, ...
% Chapters, in the main contents.
\def\numchapentry#1#2#3#4{\dochapentry{#2\labelspace#1}{#4}}
%
% Chapters, in the short toc.
% See comments in \dochapentry re vbox and related settings.
\def\shortchapentry#1#2#3#4{%
\tocentry{\shortchaplabel{#2}\labelspace #1}{\doshortpageno\bgroup#4\egroup}%
}
% Appendices, in the main contents.
% Need the word Appendix, and a fixed-size box.
%
\def\appendixbox#1{%
% We use M since it's probably the widest letter.
\setbox0 = \hbox{\putwordAppendix{} M}%
\hbox to \wd0{\putwordAppendix{} #1\hss}}
%
\def\appentry#1#2#3#4{\dochapentry{\appendixbox{#2}\labelspace#1}{#4}}
% Unnumbered chapters.
\def\unnchapentry#1#2#3#4{\dochapentry{#1}{#4}}
\def\shortunnchapentry#1#2#3#4{\tocentry{#1}{\doshortpageno\bgroup#4\egroup}}
% Sections.
\def\numsecentry#1#2#3#4{\dosecentry{#2\labelspace#1}{#4}}
\let\appsecentry=\numsecentry
\def\unnsecentry#1#2#3#4{\dosecentry{#1}{#4}}
% Subsections.
\def\numsubsecentry#1#2#3#4{\dosubsecentry{#2\labelspace#1}{#4}}
\let\appsubsecentry=\numsubsecentry
\def\unnsubsecentry#1#2#3#4{\dosubsecentry{#1}{#4}}
% And subsubsections.
\def\numsubsubsecentry#1#2#3#4{\dosubsubsecentry{#2\labelspace#1}{#4}}
\let\appsubsubsecentry=\numsubsubsecentry
\def\unnsubsubsecentry#1#2#3#4{\dosubsubsecentry{#1}{#4}}
% This parameter controls the indentation of the various levels.
% Same as \defaultparindent.
\newdimen\tocindent \tocindent = 15pt
% Now for the actual typesetting. In all these, #1 is the text and #2 is the
% page number.
%
% If the toc has to be broken over pages, we want it to be at chapters
% if at all possible; hence the \penalty.
\def\dochapentry#1#2{%
\penalty-300 \vskip1\baselineskip plus.33\baselineskip minus.25\baselineskip
\begingroup
\chapentryfonts
\tocentry{#1}{\dopageno\bgroup#2\egroup}%
\endgroup
\nobreak\vskip .25\baselineskip plus.1\baselineskip
}
\def\dosecentry#1#2{\begingroup
\secentryfonts \leftskip=\tocindent
\tocentry{#1}{\dopageno\bgroup#2\egroup}%
\endgroup}
\def\dosubsecentry#1#2{\begingroup
\subsecentryfonts \leftskip=2\tocindent
\tocentry{#1}{\dopageno\bgroup#2\egroup}%
\endgroup}
\def\dosubsubsecentry#1#2{\begingroup
\subsubsecentryfonts \leftskip=3\tocindent
\tocentry{#1}{\dopageno\bgroup#2\egroup}%
\endgroup}
% We use the same \entry macro as for the index entries.
\let\tocentry = \entry
% Space between chapter (or whatever) number and the title.
\def\labelspace{\hskip1em \relax}
\def\dopageno#1{{\rm #1}}
\def\doshortpageno#1{{\rm #1}}
\def\chapentryfonts{\secfonts \rm}
\def\secentryfonts{\textfonts}
\def\subsecentryfonts{\textfonts}
\def\subsubsecentryfonts{\textfonts}
\message{environments,}
% @foo ... @end foo.
% @point{}, @result{}, @expansion{}, @print{}, @equiv{}.
%
% Since these characters are used in examples, it should be an even number of
% \tt widths. Each \tt character is 1en, so two makes it 1em.
%
\def\point{$\star$}
\def\result{\leavevmode\raise.15ex\hbox to 1em{\hfil$\Rightarrow$\hfil}}
\def\expansion{\leavevmode\raise.1ex\hbox to 1em{\hfil$\mapsto$\hfil}}
\def\print{\leavevmode\lower.1ex\hbox to 1em{\hfil$\dashv$\hfil}}
\def\equiv{\leavevmode\lower.1ex\hbox to 1em{\hfil$\ptexequiv$\hfil}}
% The @error{} command.
% Adapted from the TeXbook's \boxit.
%
\newbox\errorbox
%
{\tentt \global\dimen0 = 3em}% Width of the box.
\dimen2 = .55pt % Thickness of rules
% The text. (`r' is open on the right, `e' somewhat less so on the left.)
\setbox0 = \hbox{\kern-.75pt \reducedsf error\kern-1.5pt}
%
\setbox\errorbox=\hbox to \dimen0{\hfil
\hsize = \dimen0 \advance\hsize by -5.8pt % Space to left+right.
\advance\hsize by -2\dimen2 % Rules.
\vbox{%
\hrule height\dimen2
\hbox{\vrule width\dimen2 \kern3pt % Space to left of text.
\vtop{\kern2.4pt \box0 \kern2.4pt}% Space above/below.
\kern3pt\vrule width\dimen2}% Space to right.
\hrule height\dimen2}
\hfil}
%
\def\error{\leavevmode\lower.7ex\copy\errorbox}
% @tex ... @end tex escapes into raw Tex temporarily.
% One exception: @ is still an escape character, so that @end tex works.
% But \@ or @@ will get a plain tex @ character.
\envdef\tex{%
\catcode `\\=0 \catcode `\{=1 \catcode `\}=2
\catcode `\$=3 \catcode `\&=4 \catcode `\#=6
\catcode `\^=7 \catcode `\_=8 \catcode `\~=\active \let~=\tie
\catcode `\%=14
\catcode `\+=\other
\catcode `\"=\other
\catcode `\|=\other
\catcode `\<=\other
\catcode `\>=\other
\escapechar=`\\
%
\let\b=\ptexb
\let\bullet=\ptexbullet
\let\c=\ptexc
\let\,=\ptexcomma
\let\.=\ptexdot
\let\dots=\ptexdots
\let\equiv=\ptexequiv
\let\!=\ptexexclam
\let\i=\ptexi
\let\indent=\ptexindent
\let\noindent=\ptexnoindent
\let\{=\ptexlbrace
\let\+=\tabalign
\let\}=\ptexrbrace
\let\/=\ptexslash
\let\*=\ptexstar
\let\t=\ptext
\let\frenchspacing=\plainfrenchspacing
%
\def\endldots{\mathinner{\ldots\ldots\ldots\ldots}}%
\def\enddots{\relax\ifmmode\endldots\else$\mathsurround=0pt \endldots\,$\fi}%
\def\@{@}%
}
% There is no need to define \Etex.
% Define @lisp ... @end lisp.
% @lisp environment forms a group so it can rebind things,
% including the definition of @end lisp (which normally is erroneous).
% Amount to narrow the margins by for @lisp.
\newskip\lispnarrowing \lispnarrowing=0.4in
% This is the definition that ^^M gets inside @lisp, @example, and other
% such environments. \null is better than a space, since it doesn't
% have any width.
\def\lisppar{\null\endgraf}
% This space is always present above and below environments.
\newskip\envskipamount \envskipamount = 0pt
% Make spacing and below environment symmetrical. We use \parskip here
% to help in doing that, since in @example-like environments \parskip
% is reset to zero; thus the \afterenvbreak inserts no space -- but the
% start of the next paragraph will insert \parskip.
%
\def\aboveenvbreak{{%
% =10000 instead of <10000 because of a special case in \itemzzz and
% \sectionheading, q.v.
\ifnum \lastpenalty=10000 \else
\advance\envskipamount by \parskip
\endgraf
\ifdim\lastskip<\envskipamount
\removelastskip
% it's not a good place to break if the last penalty was \nobreak
% or better ...
\ifnum\lastpenalty<10000 \penalty-50 \fi
\vskip\envskipamount
\fi
\fi
}}
\let\afterenvbreak = \aboveenvbreak
% \nonarrowing is a flag. If "set", @lisp etc don't narrow margins; it will
% also clear it, so that its embedded environments do the narrowing again.
\let\nonarrowing=\relax
% @cartouche ... @end cartouche: draw rectangle w/rounded corners around
% environment contents.
\font\circle=lcircle10
\newdimen\circthick
\newdimen\cartouter\newdimen\cartinner
\newskip\normbskip\newskip\normpskip\newskip\normlskip
\circthick=\fontdimen8\circle
%
\def\ctl{{\circle\char'013\hskip -6pt}}% 6pt from pl file: 1/2charwidth
\def\ctr{{\hskip 6pt\circle\char'010}}
\def\cbl{{\circle\char'012\hskip -6pt}}
\def\cbr{{\hskip 6pt\circle\char'011}}
\def\carttop{\hbox to \cartouter{\hskip\lskip
\ctl\leaders\hrule height\circthick\hfil\ctr
\hskip\rskip}}
\def\cartbot{\hbox to \cartouter{\hskip\lskip
\cbl\leaders\hrule height\circthick\hfil\cbr
\hskip\rskip}}
%
\newskip\lskip\newskip\rskip
\envdef\cartouche{%
\ifhmode\par\fi % can't be in the midst of a paragraph.
\startsavinginserts
\lskip=\leftskip \rskip=\rightskip
\leftskip=0pt\rightskip=0pt % we want these *outside*.
\cartinner=\hsize \advance\cartinner by-\lskip
\advance\cartinner by-\rskip
\cartouter=\hsize
\advance\cartouter by 18.4pt % allow for 3pt kerns on either
% side, and for 6pt waste from
% each corner char, and rule thickness
\normbskip=\baselineskip \normpskip=\parskip \normlskip=\lineskip
% Flag to tell @lisp, etc., not to narrow margin.
\let\nonarrowing = t%
\vbox\bgroup
\baselineskip=0pt\parskip=0pt\lineskip=0pt
\carttop
\hbox\bgroup
\hskip\lskip
\vrule\kern3pt
\vbox\bgroup
\kern3pt
\hsize=\cartinner
\baselineskip=\normbskip
\lineskip=\normlskip
\parskip=\normpskip
\vskip -\parskip
\comment % For explanation, see the end of \def\group.
}
\def\Ecartouche{%
\ifhmode\par\fi
\kern3pt
\egroup
\kern3pt\vrule
\hskip\rskip
\egroup
\cartbot
\egroup
\checkinserts
}
% This macro is called at the beginning of all the @example variants,
% inside a group.
\def\nonfillstart{%
\aboveenvbreak
\hfuzz = 12pt % Don't be fussy
\sepspaces % Make spaces be word-separators rather than space tokens.
\let\par = \lisppar % don't ignore blank lines
\obeylines % each line of input is a line of output
\parskip = 0pt
\parindent = 0pt
\emergencystretch = 0pt % don't try to avoid overfull boxes
\ifx\nonarrowing\relax
\advance \leftskip by \lispnarrowing
\exdentamount=\lispnarrowing
\else
\let\nonarrowing = \relax
\fi
\let\exdent=\nofillexdent
}
% If you want all examples etc. small: @set dispenvsize small.
% If you want even small examples the full size: @set dispenvsize nosmall.
% This affects the following displayed environments:
% @example, @display, @format, @lisp
%
\def\smallword{small}
\def\nosmallword{nosmall}
\let\SETdispenvsize\relax
\def\setnormaldispenv{%
\ifx\SETdispenvsize\smallword
% end paragraph for sake of leading, in case document has no blank
% line. This is redundant with what happens in \aboveenvbreak, but
% we need to do it before changing the fonts, and it's inconvenient
% to change the fonts afterward.
\ifnum \lastpenalty=10000 \else \endgraf \fi
\smallexamplefonts \rm
\fi
}
\def\setsmalldispenv{%
\ifx\SETdispenvsize\nosmallword
\else
\ifnum \lastpenalty=10000 \else \endgraf \fi
\smallexamplefonts \rm
\fi
}
% We often define two environments, @foo and @smallfoo.
% Let's do it by one command:
\def\makedispenv #1#2{
\expandafter\envdef\csname#1\endcsname {\setnormaldispenv #2}
\expandafter\envdef\csname small#1\endcsname {\setsmalldispenv #2}
\expandafter\let\csname E#1\endcsname \afterenvbreak
\expandafter\let\csname Esmall#1\endcsname \afterenvbreak
}
% Define two synonyms:
\def\maketwodispenvs #1#2#3{
\makedispenv{#1}{#3}
\makedispenv{#2}{#3}
}
% @lisp: indented, narrowed, typewriter font; @example: same as @lisp.
%
% @smallexample and @smalllisp: use smaller fonts.
% Originally contributed by Pavel@xerox.
%
\maketwodispenvs {lisp}{example}{%
\nonfillstart
\tt\quoteexpand
\let\kbdfont = \kbdexamplefont % Allow @kbd to do something special.
\gobble % eat return
}
% @display/@smalldisplay: same as @lisp except keep current font.
%
\makedispenv {display}{%
\nonfillstart
\gobble
}
% @format/@smallformat: same as @display except don't narrow margins.
%
\makedispenv{format}{%
\let\nonarrowing = t%
\nonfillstart
\gobble
}
% @flushleft: same as @format, but doesn't obey \SETdispenvsize.
\envdef\flushleft{%
\let\nonarrowing = t%
\nonfillstart
\gobble
}
\let\Eflushleft = \afterenvbreak
% @flushright.
%
\envdef\flushright{%
\let\nonarrowing = t%
\nonfillstart
\advance\leftskip by 0pt plus 1fill
\gobble
}
\let\Eflushright = \afterenvbreak
% @quotation does normal linebreaking (hence we can't use \nonfillstart)
% and narrows the margins. We keep \parskip nonzero in general, since
% we're doing normal filling. So, when using \aboveenvbreak and
% \afterenvbreak, temporarily make \parskip 0.
%
\envdef\quotation{%
{\parskip=0pt \aboveenvbreak}% because \aboveenvbreak inserts \parskip
\parindent=0pt
%
% @cartouche defines \nonarrowing to inhibit narrowing at next level down.
\ifx\nonarrowing\relax
\advance\leftskip by \lispnarrowing
\advance\rightskip by \lispnarrowing
\exdentamount = \lispnarrowing
\else
\let\nonarrowing = \relax
\fi
\parsearg\quotationlabel
}
% We have retained a nonzero parskip for the environment, since we're
% doing normal filling.
%
\def\Equotation{%
\par
\ifx\quotationauthor\undefined\else
% indent a bit.
\leftline{\kern 2\leftskip \sl ---\quotationauthor}%
\fi
{\parskip=0pt \afterenvbreak}%
}
% If we're given an argument, typeset it in bold with a colon after.
\def\quotationlabel#1{%
\def\temp{#1}%
\ifx\temp\empty \else
{\bf #1: }%
\fi
}
% LaTeX-like @verbatim...@end verbatim and @verb{...}
% If we want to allow any as delimiter,
% we need the curly braces so that makeinfo sees the @verb command, eg:
% `@verbx...x' would look like the '@verbx' command. --janneke@gnu.org
%
% [Knuth]: Donald Ervin Knuth, 1996. The TeXbook.
%
% [Knuth] p.344; only we need to do the other characters Texinfo sets
% active too. Otherwise, they get lost as the first character on a
% verbatim line.
\def\dospecials{%
\do\ \do\\\do\{\do\}\do\$\do\&%
\do\#\do\^\do\^^K\do\_\do\^^A\do\%\do\~%
\do\<\do\>\do\|\do\@\do+\do\"%
}
%
% [Knuth] p. 380
\def\uncatcodespecials{%
\def\do##1{\catcode`##1=\other}\dospecials}
%
% [Knuth] pp. 380,381,391
% Disable Spanish ligatures ?` and !` of \tt font
\begingroup
\catcode`\`=\active\gdef`{\relax\lq}
\endgroup
%
% Setup for the @verb command.
%
% Eight spaces for a tab
\begingroup
\catcode`\^^I=\active
\gdef\tabeightspaces{\catcode`\^^I=\active\def^^I{\ \ \ \ \ \ \ \ }}
\endgroup
%
\def\setupverb{%
\tt % easiest (and conventionally used) font for verbatim
\def\par{\leavevmode\endgraf}%
\catcode`\`=\active
\tabeightspaces
% Respect line breaks,
% print special symbols as themselves, and
% make each space count
% must do in this order:
\obeylines \uncatcodespecials \sepspaces
}
% Setup for the @verbatim environment
%
% Real tab expansion
\newdimen\tabw \setbox0=\hbox{\tt\space} \tabw=8\wd0 % tab amount
%
\def\starttabbox{\setbox0=\hbox\bgroup}
% Allow an option to not replace quotes with a regular directed right
% quote/apostrophe (char 0x27), but instead use the undirected quote
% from cmtt (char 0x0d). The undirected quote is ugly, so don't make it
% the default, but it works for pasting with more pdf viewers (at least
% evince), the lilypond developers report. xpdf does work with the
% regular 0x27.
%
\def\codequoteright{%
\expandafter\ifx\csname SETcodequoteundirected\endcsname\relax
'%
\else
\char'15
\fi
}
%
% and a similar option for the left quote char vs. a grave accent.
% Modern fonts display ASCII 0x60 as a grave accent, so some people like
% the code environments to do likewise.
%
\def\codequoteleft{%
\expandafter\ifx\csname SETcodequotebacktick\endcsname\relax
`%
\else
\char'22
\fi
}
%
\begingroup
\catcode`\^^I=\active
\gdef\tabexpand{%
\catcode`\^^I=\active
\def^^I{\leavevmode\egroup
\dimen0=\wd0 % the width so far, or since the previous tab
\divide\dimen0 by\tabw
\multiply\dimen0 by\tabw % compute previous multiple of \tabw
\advance\dimen0 by\tabw % advance to next multiple of \tabw
\wd0=\dimen0 \box0 \starttabbox
}%
}
\catcode`\'=\active
\gdef\rquoteexpand{\catcode\rquoteChar=\active \def'{\codequoteright}}%
%
\catcode`\`=\active
\gdef\lquoteexpand{\catcode\lquoteChar=\active \def`{\codequoteleft}}%
%
\gdef\quoteexpand{\rquoteexpand \lquoteexpand}%
\endgroup
% start the verbatim environment.
\def\setupverbatim{%
\let\nonarrowing = t%
\nonfillstart
% Easiest (and conventionally used) font for verbatim
\tt
\def\par{\leavevmode\egroup\box0\endgraf}%
\catcode`\`=\active
\tabexpand
\quoteexpand
% Respect line breaks,
% print special symbols as themselves, and
% make each space count
% must do in this order:
\obeylines \uncatcodespecials \sepspaces
\everypar{\starttabbox}%
}
% Do the @verb magic: verbatim text is quoted by unique
% delimiter characters. Before first delimiter expect a
% right brace, after last delimiter expect closing brace:
%
% \def\doverb'{'#1'}'{#1}
%
% [Knuth] p. 382; only eat outer {}
\begingroup
\catcode`[=1\catcode`]=2\catcode`\{=\other\catcode`\}=\other
\gdef\doverb{#1[\def\next##1#1}[##1\endgroup]\next]
\endgroup
%
\def\verb{\begingroup\setupverb\doverb}
%
%
% Do the @verbatim magic: define the macro \doverbatim so that
% the (first) argument ends when '@end verbatim' is reached, ie:
%
% \def\doverbatim#1@end verbatim{#1}
%
% For Texinfo it's a lot easier than for LaTeX,
% because texinfo's \verbatim doesn't stop at '\end{verbatim}':
% we need not redefine '\', '{' and '}'.
%
% Inspired by LaTeX's verbatim command set [latex.ltx]
%
\begingroup
\catcode`\ =\active
\obeylines %
% ignore everything up to the first ^^M, that's the newline at the end
% of the @verbatim input line itself. Otherwise we get an extra blank
% line in the output.
\xdef\doverbatim#1^^M#2@end verbatim{#2\noexpand\end\gobble verbatim}%
% We really want {...\end verbatim} in the body of the macro, but
% without the active space; thus we have to use \xdef and \gobble.
\endgroup
%
\envdef\verbatim{%
\setupverbatim\doverbatim
}
\let\Everbatim = \afterenvbreak
% @verbatiminclude FILE - insert text of file in verbatim environment.
%
\def\verbatiminclude{\parseargusing\filenamecatcodes\doverbatiminclude}
%
\def\doverbatiminclude#1{%
{%
\makevalueexpandable
\setupverbatim
\input #1
\afterenvbreak
}%
}
% @copying ... @end copying.
% Save the text away for @insertcopying later.
%
% We save the uninterpreted tokens, rather than creating a box.
% Saving the text in a box would be much easier, but then all the
% typesetting commands (@smallbook, font changes, etc.) have to be done
% beforehand -- and a) we want @copying to be done first in the source
% file; b) letting users define the frontmatter in as flexible order as
% possible is very desirable.
%
\def\copying{\checkenv{}\begingroup\scanargctxt\docopying}
\def\docopying#1@end copying{\endgroup\def\copyingtext{#1}}
%
\def\insertcopying{%
\begingroup
\parindent = 0pt % paragraph indentation looks wrong on title page
\scanexp\copyingtext
\endgroup
}
\message{defuns,}
% @defun etc.
\newskip\defbodyindent \defbodyindent=.4in
\newskip\defargsindent \defargsindent=50pt
\newskip\deflastargmargin \deflastargmargin=18pt
\newcount\defunpenalty
% Start the processing of @deffn:
\def\startdefun{%
\ifnum\lastpenalty<10000
\medbreak
\defunpenalty=10003 % Will keep this @deffn together with the
% following @def command, see below.
\else
% If there are two @def commands in a row, we'll have a \nobreak,
% which is there to keep the function description together with its
% header. But if there's nothing but headers, we need to allow a
% break somewhere. Check specifically for penalty 10002, inserted
% by \printdefunline, instead of 10000, since the sectioning
% commands also insert a nobreak penalty, and we don't want to allow
% a break between a section heading and a defun.
%
% As a minor refinement, we avoid "club" headers by signalling
% with penalty of 10003 after the very first @deffn in the
% sequence (see above), and penalty of 10002 after any following
% @def command.
\ifnum\lastpenalty=10002 \penalty2000 \else \defunpenalty=10002 \fi
%
% Similarly, after a section heading, do not allow a break.
% But do insert the glue.
\medskip % preceded by discardable penalty, so not a breakpoint
\fi
%
\parindent=0in
\advance\leftskip by \defbodyindent
\exdentamount=\defbodyindent
}
\def\dodefunx#1{%
% First, check whether we are in the right environment:
\checkenv#1%
%
% As above, allow line break if we have multiple x headers in a row.
% It's not a great place, though.
\ifnum\lastpenalty=10002 \penalty3000 \else \defunpenalty=10002 \fi
%
% And now, it's time to reuse the body of the original defun:
\expandafter\gobbledefun#1%
}
\def\gobbledefun#1\startdefun{}
% \printdefunline \deffnheader{text}
%
\def\printdefunline#1#2{%
\begingroup
% call \deffnheader:
#1#2 \endheader
% common ending:
\interlinepenalty = 10000
\advance\rightskip by 0pt plus 1fil
\endgraf
\nobreak\vskip -\parskip
\penalty\defunpenalty % signal to \startdefun and \dodefunx
% Some of the @defun-type tags do not enable magic parentheses,
% rendering the following check redundant. But we don't optimize.
\checkparencounts
\endgroup
}
\def\Edefun{\endgraf\medbreak}
% \makedefun{deffn} creates \deffn, \deffnx and \Edeffn;
% the only thing remainnig is to define \deffnheader.
%
\def\makedefun#1{%
\expandafter\let\csname E#1\endcsname = \Edefun
\edef\temp{\noexpand\domakedefun
\makecsname{#1}\makecsname{#1x}\makecsname{#1header}}%
\temp
}
% \domakedefun \deffn \deffnx \deffnheader
%
% Define \deffn and \deffnx, without parameters.
% \deffnheader has to be defined explicitly.
%
\def\domakedefun#1#2#3{%
\envdef#1{%
\startdefun
\parseargusing\activeparens{\printdefunline#3}%
}%
\def#2{\dodefunx#1}%
\def#3%
}
%%% Untyped functions:
% @deffn category name args
\makedefun{deffn}{\deffngeneral{}}
% @deffn category class name args
\makedefun{defop}#1 {\defopon{#1\ \putwordon}}
% \defopon {category on}class name args
\def\defopon#1#2 {\deffngeneral{\putwordon\ \code{#2}}{#1\ \code{#2}} }
% \deffngeneral {subind}category name args
%
\def\deffngeneral#1#2 #3 #4\endheader{%
% Remember that \dosubind{fn}{foo}{} is equivalent to \doind{fn}{foo}.
\dosubind{fn}{\code{#3}}{#1}%
\defname{#2}{}{#3}\magicamp\defunargs{#4\unskip}%
}
%%% Typed functions:
% @deftypefn category type name args
\makedefun{deftypefn}{\deftypefngeneral{}}
% @deftypeop category class type name args
\makedefun{deftypeop}#1 {\deftypeopon{#1\ \putwordon}}
% \deftypeopon {category on}class type name args
\def\deftypeopon#1#2 {\deftypefngeneral{\putwordon\ \code{#2}}{#1\ \code{#2}} }
% \deftypefngeneral {subind}category type name args
%
\def\deftypefngeneral#1#2 #3 #4 #5\endheader{%
\dosubind{fn}{\code{#4}}{#1}%
\defname{#2}{#3}{#4}\defunargs{#5\unskip}%
}
%%% Typed variables:
% @deftypevr category type var args
\makedefun{deftypevr}{\deftypecvgeneral{}}
% @deftypecv category class type var args
\makedefun{deftypecv}#1 {\deftypecvof{#1\ \putwordof}}
% \deftypecvof {category of}class type var args
\def\deftypecvof#1#2 {\deftypecvgeneral{\putwordof\ \code{#2}}{#1\ \code{#2}} }
% \deftypecvgeneral {subind}category type var args
%
\def\deftypecvgeneral#1#2 #3 #4 #5\endheader{%
\dosubind{vr}{\code{#4}}{#1}%
\defname{#2}{#3}{#4}\defunargs{#5\unskip}%
}
%%% Untyped variables:
% @defvr category var args
\makedefun{defvr}#1 {\deftypevrheader{#1} {} }
% @defcv category class var args
\makedefun{defcv}#1 {\defcvof{#1\ \putwordof}}
% \defcvof {category of}class var args
\def\defcvof#1#2 {\deftypecvof{#1}#2 {} }
%%% Type:
% @deftp category name args
\makedefun{deftp}#1 #2 #3\endheader{%
\doind{tp}{\code{#2}}%
\defname{#1}{}{#2}\defunargs{#3\unskip}%
}
% Remaining @defun-like shortcuts:
\makedefun{defun}{\deffnheader{\putwordDeffunc} }
\makedefun{defmac}{\deffnheader{\putwordDefmac} }
\makedefun{defspec}{\deffnheader{\putwordDefspec} }
\makedefun{deftypefun}{\deftypefnheader{\putwordDeffunc} }
\makedefun{defvar}{\defvrheader{\putwordDefvar} }
\makedefun{defopt}{\defvrheader{\putwordDefopt} }
\makedefun{deftypevar}{\deftypevrheader{\putwordDefvar} }
\makedefun{defmethod}{\defopon\putwordMethodon}
\makedefun{deftypemethod}{\deftypeopon\putwordMethodon}
\makedefun{defivar}{\defcvof\putwordInstanceVariableof}
\makedefun{deftypeivar}{\deftypecvof\putwordInstanceVariableof}
% \defname, which formats the name of the @def (not the args).
% #1 is the category, such as "Function".
% #2 is the return type, if any.
% #3 is the function name.
%
% We are followed by (but not passed) the arguments, if any.
%
\def\defname#1#2#3{%
% Get the values of \leftskip and \rightskip as they were outside the @def...
\advance\leftskip by -\defbodyindent
%
% How we'll format the type name. Putting it in brackets helps
% distinguish it from the body text that may end up on the next line
% just below it.
\def\temp{#1}%
\setbox0=\hbox{\kern\deflastargmargin \ifx\temp\empty\else [\rm\temp]\fi}
%
% Figure out line sizes for the paragraph shape.
% The first line needs space for \box0; but if \rightskip is nonzero,
% we need only space for the part of \box0 which exceeds it:
\dimen0=\hsize \advance\dimen0 by -\wd0 \advance\dimen0 by \rightskip
% The continuations:
\dimen2=\hsize \advance\dimen2 by -\defargsindent
% (plain.tex says that \dimen1 should be used only as global.)
\parshape 2 0in \dimen0 \defargsindent \dimen2
%
% Put the type name to the right margin.
\noindent
\hbox to 0pt{%
\hfil\box0 \kern-\hsize
% \hsize has to be shortened this way:
\kern\leftskip
% Intentionally do not respect \rightskip, since we need the space.
}%
%
% Allow all lines to be underfull without complaint:
\tolerance=10000 \hbadness=10000
\exdentamount=\defbodyindent
{%
% defun fonts. We use typewriter by default (used to be bold) because:
% . we're printing identifiers, they should be in tt in principle.
% . in languages with many accents, such as Czech or French, it's
% common to leave accents off identifiers. The result looks ok in
% tt, but exceedingly strange in rm.
% . we don't want -- and --- to be treated as ligatures.
% . this still does not fix the ?` and !` ligatures, but so far no
% one has made identifiers using them :).
\df \tt
\def\temp{#2}% return value type
\ifx\temp\empty\else \tclose{\temp} \fi
#3% output function name
}%
{\rm\enskip}% hskip 0.5 em of \tenrm
%
\boldbrax
% arguments will be output next, if any.
}
% Print arguments in slanted roman (not ttsl), inconsistently with using
% tt for the name. This is because literal text is sometimes needed in
% the argument list (groff manual), and ttsl and tt are not very
% distinguishable. Prevent hyphenation at `-' chars.
%
\def\defunargs#1{%
% use sl by default (not ttsl),
% tt for the names.
\df \sl \hyphenchar\font=0
%
% On the other hand, if an argument has two dashes (for instance), we
% want a way to get ttsl. Let's try @var for that.
\let\var=\ttslanted
#1%
\sl\hyphenchar\font=45
}
% We want ()&[] to print specially on the defun line.
%
\def\activeparens{%
\catcode`\(=\active \catcode`\)=\active
\catcode`\[=\active \catcode`\]=\active
\catcode`\&=\active
}
% Make control sequences which act like normal parenthesis chars.
\let\lparen = ( \let\rparen = )
% Be sure that we always have a definition for `(', etc. For example,
% if the fn name has parens in it, \boldbrax will not be in effect yet,
% so TeX would otherwise complain about undefined control sequence.
{
\activeparens
\global\let(=\lparen \global\let)=\rparen
\global\let[=\lbrack \global\let]=\rbrack
\global\let& = \&
\gdef\boldbrax{\let(=\opnr\let)=\clnr\let[=\lbrb\let]=\rbrb}
\gdef\magicamp{\let&=\amprm}
}
\newcount\parencount
% If we encounter &foo, then turn on ()-hacking afterwards
\newif\ifampseen
\def\amprm#1 {\ampseentrue{\bf\ }}
\def\parenfont{%
\ifampseen
% At the first level, print parens in roman,
% otherwise use the default font.
\ifnum \parencount=1 \rm \fi
\else
% The \sf parens (in \boldbrax) actually are a little bolder than
% the contained text. This is especially needed for [ and ] .
\sf
\fi
}
\def\infirstlevel#1{%
\ifampseen
\ifnum\parencount=1
#1%
\fi
\fi
}
\def\bfafterword#1 {#1 \bf}
\def\opnr{%
\global\advance\parencount by 1
{\parenfont(}%
\infirstlevel \bfafterword
}
\def\clnr{%
{\parenfont)}%
\infirstlevel \sl
\global\advance\parencount by -1
}
\newcount\brackcount
\def\lbrb{%
\global\advance\brackcount by 1
{\bf[}%
}
\def\rbrb{%
{\bf]}%
\global\advance\brackcount by -1
}
\def\checkparencounts{%
\ifnum\parencount=0 \else \badparencount \fi
\ifnum\brackcount=0 \else \badbrackcount \fi
}
\def\badparencount{%
\errmessage{Unbalanced parentheses in @def}%
\global\parencount=0
}
\def\badbrackcount{%
\errmessage{Unbalanced square braces in @def}%
\global\brackcount=0
}
\message{macros,}
% @macro.
% To do this right we need a feature of e-TeX, \scantokens,
% which we arrange to emulate with a temporary file in ordinary TeX.
\ifx\eTeXversion\undefined
\newwrite\macscribble
\def\scantokens#1{%
\toks0={#1}%
\immediate\openout\macscribble=\jobname.tmp
\immediate\write\macscribble{\the\toks0}%
\immediate\closeout\macscribble
\input \jobname.tmp
}
\fi
\def\scanmacro#1{%
\begingroup
\newlinechar`\^^M
\let\xeatspaces\eatspaces
% Undo catcode changes of \startcontents and \doprintindex
% When called from @insertcopying or (short)caption, we need active
% backslash to get it printed correctly. Previously, we had
% \catcode`\\=\other instead. We'll see whether a problem appears
% with macro expansion. --kasal, 19aug04
\catcode`\@=0 \catcode`\\=\active \escapechar=`\@
% ... and \example
\spaceisspace
%
% Append \endinput to make sure that TeX does not see the ending newline.
% I've verified that it is necessary both for e-TeX and for ordinary TeX
% --kasal, 29nov03
\scantokens{#1\endinput}%
\endgroup
}
\def\scanexp#1{%
\edef\temp{\noexpand\scanmacro{#1}}%
\temp
}
\newcount\paramno % Count of parameters
\newtoks\macname % Macro name
\newif\ifrecursive % Is it recursive?
% List of all defined macros in the form
% \definedummyword\macro1\definedummyword\macro2...
% Currently is also contains all @aliases; the list can be split
% if there is a need.
\def\macrolist{}
% Add the macro to \macrolist
\def\addtomacrolist#1{\expandafter \addtomacrolistxxx \csname#1\endcsname}
\def\addtomacrolistxxx#1{%
\toks0 = \expandafter{\macrolist\definedummyword#1}%
\xdef\macrolist{\the\toks0}%
}
% Utility routines.
% This does \let #1 = #2, with \csnames; that is,
% \let \csname#1\endcsname = \csname#2\endcsname
% (except of course we have to play expansion games).
%
\def\cslet#1#2{%
\expandafter\let
\csname#1\expandafter\endcsname
\csname#2\endcsname
}
% Trim leading and trailing spaces off a string.
% Concepts from aro-bend problem 15 (see CTAN).
{\catcode`\@=11
\gdef\eatspaces #1{\expandafter\trim@\expandafter{#1 }}
\gdef\trim@ #1{\trim@@ @#1 @ #1 @ @@}
\gdef\trim@@ #1@ #2@ #3@@{\trim@@@\empty #2 @}
\def\unbrace#1{#1}
\unbrace{\gdef\trim@@@ #1 } #2@{#1}
}
% Trim a single trailing ^^M off a string.
{\catcode`\^^M=\other \catcode`\Q=3%
\gdef\eatcr #1{\eatcra #1Q^^MQ}%
\gdef\eatcra#1^^MQ{\eatcrb#1Q}%
\gdef\eatcrb#1Q#2Q{#1}%
}
% Macro bodies are absorbed as an argument in a context where
% all characters are catcode 10, 11 or 12, except \ which is active
% (as in normal texinfo). It is necessary to change the definition of \.
% It's necessary to have hard CRs when the macro is executed. This is
% done by making ^^M (\endlinechar) catcode 12 when reading the macro
% body, and then making it the \newlinechar in \scanmacro.
\def\scanctxt{%
\catcode`\"=\other
\catcode`\+=\other
\catcode`\<=\other
\catcode`\>=\other
\catcode`\@=\other
\catcode`\^=\other
\catcode`\_=\other
\catcode`\|=\other
\catcode`\~=\other
}
\def\scanargctxt{%
\scanctxt
\catcode`\\=\other
\catcode`\^^M=\other
}
\def\macrobodyctxt{%
\scanctxt
\catcode`\{=\other
\catcode`\}=\other
\catcode`\^^M=\other
\usembodybackslash
}
\def\macroargctxt{%
\scanctxt
\catcode`\\=\other
}
% \mbodybackslash is the definition of \ in @macro bodies.
% It maps \foo\ => \csname macarg.foo\endcsname => #N
% where N is the macro parameter number.
% We define \csname macarg.\endcsname to be \realbackslash, so
% \\ in macro replacement text gets you a backslash.
{\catcode`@=0 @catcode`@\=@active
@gdef@usembodybackslash{@let\=@mbodybackslash}
@gdef@mbodybackslash#1\{@csname macarg.#1@endcsname}
}
\expandafter\def\csname macarg.\endcsname{\realbackslash}
\def\macro{\recursivefalse\parsearg\macroxxx}
\def\rmacro{\recursivetrue\parsearg\macroxxx}
\def\macroxxx#1{%
\getargs{#1}% now \macname is the macname and \argl the arglist
\ifx\argl\empty % no arguments
\paramno=0%
\else
\expandafter\parsemargdef \argl;%
\fi
\if1\csname ismacro.\the\macname\endcsname
\message{Warning: redefining \the\macname}%
\else
\expandafter\ifx\csname \the\macname\endcsname \relax
\else \errmessage{Macro name \the\macname\space already defined}\fi
\global\cslet{macsave.\the\macname}{\the\macname}%
\global\expandafter\let\csname ismacro.\the\macname\endcsname=1%
\addtomacrolist{\the\macname}%
\fi
\begingroup \macrobodyctxt
\ifrecursive \expandafter\parsermacbody
\else \expandafter\parsemacbody
\fi}
\parseargdef\unmacro{%
\if1\csname ismacro.#1\endcsname
\global\cslet{#1}{macsave.#1}%
\global\expandafter\let \csname ismacro.#1\endcsname=0%
% Remove the macro name from \macrolist:
\begingroup
\expandafter\let\csname#1\endcsname \relax
\let\definedummyword\unmacrodo
\xdef\macrolist{\macrolist}%
\endgroup
\else
\errmessage{Macro #1 not defined}%
\fi
}
% Called by \do from \dounmacro on each macro. The idea is to omit any
% macro definitions that have been changed to \relax.
%
\def\unmacrodo#1{%
\ifx #1\relax
% remove this
\else
\noexpand\definedummyword \noexpand#1%
\fi
}
% This makes use of the obscure feature that if the last token of a
% is #, then the preceding argument is delimited by
% an opening brace, and that opening brace is not consumed.
\def\getargs#1{\getargsxxx#1{}}
\def\getargsxxx#1#{\getmacname #1 \relax\getmacargs}
\def\getmacname #1 #2\relax{\macname={#1}}
\def\getmacargs#1{\def\argl{#1}}
% Parse the optional {params} list. Set up \paramno and \paramlist
% so \defmacro knows what to do. Define \macarg.blah for each blah
% in the params list, to be ##N where N is the position in that list.
% That gets used by \mbodybackslash (above).
% We need to get `macro parameter char #' into several definitions.
% The technique used is stolen from LaTeX: let \hash be something
% unexpandable, insert that wherever you need a #, and then redefine
% it to # just before using the token list produced.
%
% The same technique is used to protect \eatspaces till just before
% the macro is used.
\def\parsemargdef#1;{\paramno=0\def\paramlist{}%
\let\hash\relax\let\xeatspaces\relax\parsemargdefxxx#1,;,}
\def\parsemargdefxxx#1,{%
\if#1;\let\next=\relax
\else \let\next=\parsemargdefxxx
\advance\paramno by 1%
\expandafter\edef\csname macarg.\eatspaces{#1}\endcsname
{\xeatspaces{\hash\the\paramno}}%
\edef\paramlist{\paramlist\hash\the\paramno,}%
\fi\next}
% These two commands read recursive and nonrecursive macro bodies.
% (They're different since rec and nonrec macros end differently.)
\long\def\parsemacbody#1@end macro%
{\xdef\temp{\eatcr{#1}}\endgroup\defmacro}%
\long\def\parsermacbody#1@end rmacro%
{\xdef\temp{\eatcr{#1}}\endgroup\defmacro}%
% This defines the macro itself. There are six cases: recursive and
% nonrecursive macros of zero, one, and many arguments.
% Much magic with \expandafter here.
% \xdef is used so that macro definitions will survive the file
% they're defined in; @include reads the file inside a group.
\def\defmacro{%
\let\hash=##% convert placeholders to macro parameter chars
\ifrecursive
\ifcase\paramno
% 0
\expandafter\xdef\csname\the\macname\endcsname{%
\noexpand\scanmacro{\temp}}%
\or % 1
\expandafter\xdef\csname\the\macname\endcsname{%
\bgroup\noexpand\macroargctxt
\noexpand\braceorline
\expandafter\noexpand\csname\the\macname xxx\endcsname}%
\expandafter\xdef\csname\the\macname xxx\endcsname##1{%
\egroup\noexpand\scanmacro{\temp}}%
\else % many
\expandafter\xdef\csname\the\macname\endcsname{%
\bgroup\noexpand\macroargctxt
\noexpand\csname\the\macname xx\endcsname}%
\expandafter\xdef\csname\the\macname xx\endcsname##1{%
\expandafter\noexpand\csname\the\macname xxx\endcsname ##1,}%
\expandafter\expandafter
\expandafter\xdef
\expandafter\expandafter
\csname\the\macname xxx\endcsname
\paramlist{\egroup\noexpand\scanmacro{\temp}}%
\fi
\else
\ifcase\paramno
% 0
\expandafter\xdef\csname\the\macname\endcsname{%
\noexpand\norecurse{\the\macname}%
\noexpand\scanmacro{\temp}\egroup}%
\or % 1
\expandafter\xdef\csname\the\macname\endcsname{%
\bgroup\noexpand\macroargctxt
\noexpand\braceorline
\expandafter\noexpand\csname\the\macname xxx\endcsname}%
\expandafter\xdef\csname\the\macname xxx\endcsname##1{%
\egroup
\noexpand\norecurse{\the\macname}%
\noexpand\scanmacro{\temp}\egroup}%
\else % many
\expandafter\xdef\csname\the\macname\endcsname{%
\bgroup\noexpand\macroargctxt
\expandafter\noexpand\csname\the\macname xx\endcsname}%
\expandafter\xdef\csname\the\macname xx\endcsname##1{%
\expandafter\noexpand\csname\the\macname xxx\endcsname ##1,}%
\expandafter\expandafter
\expandafter\xdef
\expandafter\expandafter
\csname\the\macname xxx\endcsname
\paramlist{%
\egroup
\noexpand\norecurse{\the\macname}%
\noexpand\scanmacro{\temp}\egroup}%
\fi
\fi}
\def\norecurse#1{\bgroup\cslet{#1}{macsave.#1}}
% \braceorline decides whether the next nonwhitespace character is a
% {. If so it reads up to the closing }, if not, it reads the whole
% line. Whatever was read is then fed to the next control sequence
% as an argument (by \parsebrace or \parsearg)
\def\braceorline#1{\let\macnamexxx=#1\futurelet\nchar\braceorlinexxx}
\def\braceorlinexxx{%
\ifx\nchar\bgroup\else
\expandafter\parsearg
\fi \macnamexxx}
% @alias.
% We need some trickery to remove the optional spaces around the equal
% sign. Just make them active and then expand them all to nothing.
\def\alias{\parseargusing\obeyspaces\aliasxxx}
\def\aliasxxx #1{\aliasyyy#1\relax}
\def\aliasyyy #1=#2\relax{%
{%
\expandafter\let\obeyedspace=\empty
\addtomacrolist{#1}%
\xdef\next{\global\let\makecsname{#1}=\makecsname{#2}}%
}%
\next
}
\message{cross references,}
\newwrite\auxfile
\newif\ifhavexrefs % True if xref values are known.
\newif\ifwarnedxrefs % True if we warned once that they aren't known.
% @inforef is relatively simple.
\def\inforef #1{\inforefzzz #1,,,,**}
\def\inforefzzz #1,#2,#3,#4**{\putwordSee{} \putwordInfo{} \putwordfile{} \file{\ignorespaces #3{}},
node \samp{\ignorespaces#1{}}}
% @node's only job in TeX is to define \lastnode, which is used in
% cross-references. The @node line might or might not have commas, and
% might or might not have spaces before the first comma, like:
% @node foo , bar , ...
% We don't want such trailing spaces in the node name.
%
\parseargdef\node{\checkenv{}\donode #1 ,\finishnodeparse}
%
% also remove a trailing comma, in case of something like this:
% @node Help-Cross, , , Cross-refs
\def\donode#1 ,#2\finishnodeparse{\dodonode #1,\finishnodeparse}
\def\dodonode#1,#2\finishnodeparse{\gdef\lastnode{#1}}
\let\nwnode=\node
\let\lastnode=\empty
% Write a cross-reference definition for the current node. #1 is the
% type (Ynumbered, Yappendix, Ynothing).
%
\def\donoderef#1{%
\ifx\lastnode\empty\else
\setref{\lastnode}{#1}%
\global\let\lastnode=\empty
\fi
}
% @anchor{NAME} -- define xref target at arbitrary point.
%
\newcount\savesfregister
%
\def\savesf{\relax \ifhmode \savesfregister=\spacefactor \fi}
\def\restoresf{\relax \ifhmode \spacefactor=\savesfregister \fi}
\def\anchor#1{\savesf \setref{#1}{Ynothing}\restoresf \ignorespaces}
% \setref{NAME}{SNT} defines a cross-reference point NAME (a node or an
% anchor), which consists of three parts:
% 1) NAME-title - the current sectioning name taken from \thissection,
% or the anchor name.
% 2) NAME-snt - section number and type, passed as the SNT arg, or
% empty for anchors.
% 3) NAME-pg - the page number.
%
% This is called from \donoderef, \anchor, and \dofloat. In the case of
% floats, there is an additional part, which is not written here:
% 4) NAME-lof - the text as it should appear in a @listoffloats.
%
\def\setref#1#2{%
\pdfmkdest{#1}%
\iflinks
{%
\atdummies % preserve commands, but don't expand them
\edef\writexrdef##1##2{%
\write\auxfile{@xrdef{#1-% #1 of \setref, expanded by the \edef
##1}{##2}}% these are parameters of \writexrdef
}%
\toks0 = \expandafter{\thissection}%
\immediate \writexrdef{title}{\the\toks0 }%
\immediate \writexrdef{snt}{\csname #2\endcsname}% \Ynumbered etc.
\safewhatsit{\writexrdef{pg}{\folio}}% will be written later, during \shipout
}%
\fi
}
% @xref, @pxref, and @ref generate cross-references. For \xrefX, #1 is
% the node name, #2 the name of the Info cross-reference, #3 the printed
% node name, #4 the name of the Info file, #5 the name of the printed
% manual. All but the node name can be omitted.
%
\def\pxref#1{\putwordsee{} \xrefX[#1,,,,,,,]}
\def\xref#1{\putwordSee{} \xrefX[#1,,,,,,,]}
\def\ref#1{\xrefX[#1,,,,,,,]}
\def\xrefX[#1,#2,#3,#4,#5,#6]{\begingroup
\unsepspaces
\def\printedmanual{\ignorespaces #5}%
\def\printedrefname{\ignorespaces #3}%
\setbox1=\hbox{\printedmanual\unskip}%
\setbox0=\hbox{\printedrefname\unskip}%
\ifdim \wd0 = 0pt
% No printed node name was explicitly given.
\expandafter\ifx\csname SETxref-automatic-section-title\endcsname\relax
% Use the node name inside the square brackets.
\def\printedrefname{\ignorespaces #1}%
\else
% Use the actual chapter/section title appear inside
% the square brackets. Use the real section title if we have it.
\ifdim \wd1 > 0pt
% It is in another manual, so we don't have it.
\def\printedrefname{\ignorespaces #1}%
\else
\ifhavexrefs
% We know the real title if we have the xref values.
\def\printedrefname{\refx{#1-title}{}}%
\else
% Otherwise just copy the Info node name.
\def\printedrefname{\ignorespaces #1}%
\fi%
\fi
\fi
\fi
%
% Make link in pdf output.
\ifpdf
\leavevmode
\getfilename{#4}%
{\indexnofonts
\turnoffactive
% See comments at \activebackslashdouble.
{\activebackslashdouble \xdef\pdfxrefdest{#1}%
\backslashparens\pdfxrefdest}%
%
\ifnum\filenamelength>0
\startlink attr{/Border [0 0 0]}%
goto file{\the\filename.pdf} name{\pdfxrefdest}%
\else
\startlink attr{/Border [0 0 0]}%
goto name{\pdfmkpgn{\pdfxrefdest}}%
\fi
}%
\linkcolor
\fi
%
% Float references are printed completely differently: "Figure 1.2"
% instead of "[somenode], p.3". We distinguish them by the
% LABEL-title being set to a magic string.
{%
% Have to otherify everything special to allow the \csname to
% include an _ in the xref name, etc.
\indexnofonts
\turnoffactive
\expandafter\global\expandafter\let\expandafter\Xthisreftitle
\csname XR#1-title\endcsname
}%
\iffloat\Xthisreftitle
% If the user specified the print name (third arg) to the ref,
% print it instead of our usual "Figure 1.2".
\ifdim\wd0 = 0pt
\refx{#1-snt}{}%
\else
\printedrefname
\fi
%
% if the user also gave the printed manual name (fifth arg), append
% "in MANUALNAME".
\ifdim \wd1 > 0pt
\space \putwordin{} \cite{\printedmanual}%
\fi
\else
% node/anchor (non-float) references.
%
% If we use \unhbox0 and \unhbox1 to print the node names, TeX does not
% insert empty discretionaries after hyphens, which means that it will
% not find a line break at a hyphen in a node names. Since some manuals
% are best written with fairly long node names, containing hyphens, this
% is a loss. Therefore, we give the text of the node name again, so it
% is as if TeX is seeing it for the first time.
\ifdim \wd1 > 0pt
\putwordsection{} ``\printedrefname'' \putwordin{} \cite{\printedmanual}%
\else
% _ (for example) has to be the character _ for the purposes of the
% control sequence corresponding to the node, but it has to expand
% into the usual \leavevmode...\vrule stuff for purposes of
% printing. So we \turnoffactive for the \refx-snt, back on for the
% printing, back off for the \refx-pg.
{\turnoffactive
% Only output a following space if the -snt ref is nonempty; for
% @unnumbered and @anchor, it won't be.
\setbox2 = \hbox{\ignorespaces \refx{#1-snt}{}}%
\ifdim \wd2 > 0pt \refx{#1-snt}\space\fi
}%
% output the `[mynode]' via a macro so it can be overridden.
\xrefprintnodename\printedrefname
%
% But we always want a comma and a space:
,\space
%
% output the `page 3'.
\turnoffactive \putwordpage\tie\refx{#1-pg}{}%
\fi
\fi
\endlink
\endgroup}
% This macro is called from \xrefX for the `[nodename]' part of xref
% output. It's a separate macro only so it can be changed more easily,
% since square brackets don't work well in some documents. Particularly
% one that Bob is working on :).
%
\def\xrefprintnodename#1{[#1]}
% Things referred to by \setref.
%
\def\Ynothing{}
\def\Yomitfromtoc{}
\def\Ynumbered{%
\ifnum\secno=0
\putwordChapter@tie \the\chapno
\else \ifnum\subsecno=0
\putwordSection@tie \the\chapno.\the\secno
\else \ifnum\subsubsecno=0
\putwordSection@tie \the\chapno.\the\secno.\the\subsecno
\else
\putwordSection@tie \the\chapno.\the\secno.\the\subsecno.\the\subsubsecno
\fi\fi\fi
}
\def\Yappendix{%
\ifnum\secno=0
\putwordAppendix@tie @char\the\appendixno{}%
\else \ifnum\subsecno=0
\putwordSection@tie @char\the\appendixno.\the\secno
\else \ifnum\subsubsecno=0
\putwordSection@tie @char\the\appendixno.\the\secno.\the\subsecno
\else
\putwordSection@tie
@char\the\appendixno.\the\secno.\the\subsecno.\the\subsubsecno
\fi\fi\fi
}
% Define \refx{NAME}{SUFFIX} to reference a cross-reference string named NAME.
% If its value is nonempty, SUFFIX is output afterward.
%
\def\refx#1#2{%
{%
\indexnofonts
\otherbackslash
\expandafter\global\expandafter\let\expandafter\thisrefX
\csname XR#1\endcsname
}%
\ifx\thisrefX\relax
% If not defined, say something at least.
\angleleft un\-de\-fined\angleright
\iflinks
\ifhavexrefs
\message{\linenumber Undefined cross reference `#1'.}%
\else
\ifwarnedxrefs\else
\global\warnedxrefstrue
\message{Cross reference values unknown; you must run TeX again.}%
\fi
\fi
\fi
\else
% It's defined, so just use it.
\thisrefX
\fi
#2% Output the suffix in any case.
}
% This is the macro invoked by entries in the aux file. Usually it's
% just a \def (we prepend XR to the control sequence name to avoid
% collisions). But if this is a float type, we have more work to do.
%
\def\xrdef#1#2{%
{% The node name might contain 8-bit characters, which in our current
% implementation are changed to commands like @'e. Don't let these
% mess up the control sequence name.
\indexnofonts
\turnoffactive
\xdef\safexrefname{#1}%
}%
%
\expandafter\gdef\csname XR\safexrefname\endcsname{#2}% remember this xref
%
% Was that xref control sequence that we just defined for a float?
\expandafter\iffloat\csname XR\safexrefname\endcsname
% it was a float, and we have the (safe) float type in \iffloattype.
\expandafter\let\expandafter\floatlist
\csname floatlist\iffloattype\endcsname
%
% Is this the first time we've seen this float type?
\expandafter\ifx\floatlist\relax
\toks0 = {\do}% yes, so just \do
\else
% had it before, so preserve previous elements in list.
\toks0 = \expandafter{\floatlist\do}%
\fi
%
% Remember this xref in the control sequence \floatlistFLOATTYPE,
% for later use in \listoffloats.
\expandafter\xdef\csname floatlist\iffloattype\endcsname{\the\toks0
{\safexrefname}}%
\fi
}
% Read the last existing aux file, if any. No error if none exists.
%
\def\tryauxfile{%
\openin 1 \jobname.aux
\ifeof 1 \else
\readdatafile{aux}%
\global\havexrefstrue
\fi
\closein 1
}
\def\setupdatafile{%
\catcode`\^^@=\other
\catcode`\^^A=\other
\catcode`\^^B=\other
\catcode`\^^C=\other
\catcode`\^^D=\other
\catcode`\^^E=\other
\catcode`\^^F=\other
\catcode`\^^G=\other
\catcode`\^^H=\other
\catcode`\^^K=\other
\catcode`\^^L=\other
\catcode`\^^N=\other
\catcode`\^^P=\other
\catcode`\^^Q=\other
\catcode`\^^R=\other
\catcode`\^^S=\other
\catcode`\^^T=\other
\catcode`\^^U=\other
\catcode`\^^V=\other
\catcode`\^^W=\other
\catcode`\^^X=\other
\catcode`\^^Z=\other
\catcode`\^^[=\other
\catcode`\^^\=\other
\catcode`\^^]=\other
\catcode`\^^^=\other
\catcode`\^^_=\other
% It was suggested to set the catcode of ^ to 7, which would allow ^^e4 etc.
% in xref tags, i.e., node names. But since ^^e4 notation isn't
% supported in the main text, it doesn't seem desirable. Furthermore,
% that is not enough: for node names that actually contain a ^
% character, we would end up writing a line like this: 'xrdef {'hat
% b-title}{'hat b} and \xrdef does a \csname...\endcsname on the first
% argument, and \hat is not an expandable control sequence. It could
% all be worked out, but why? Either we support ^^ or we don't.
%
% The other change necessary for this was to define \auxhat:
% \def\auxhat{\def^{'hat }}% extra space so ok if followed by letter
% and then to call \auxhat in \setq.
%
\catcode`\^=\other
%
% Special characters. Should be turned off anyway, but...
\catcode`\~=\other
\catcode`\[=\other
\catcode`\]=\other
\catcode`\"=\other
\catcode`\_=\other
\catcode`\|=\other
\catcode`\<=\other
\catcode`\>=\other
\catcode`\$=\other
\catcode`\#=\other
\catcode`\&=\other
\catcode`\%=\other
\catcode`+=\other % avoid \+ for paranoia even though we've turned it off
%
% This is to support \ in node names and titles, since the \
% characters end up in a \csname. It's easier than
% leaving it active and making its active definition an actual \
% character. What I don't understand is why it works in the *value*
% of the xrdef. Seems like it should be a catcode12 \, and that
% should not typeset properly. But it works, so I'm moving on for
% now. --karl, 15jan04.
\catcode`\\=\other
%
% Make the characters 128-255 be printing characters.
{%
\count1=128
\def\loop{%
\catcode\count1=\other
\advance\count1 by 1
\ifnum \count1<256 \loop \fi
}%
}%
%
% @ is our escape character in .aux files, and we need braces.
\catcode`\{=1
\catcode`\}=2
\catcode`\@=0
}
\def\readdatafile#1{%
\begingroup
\setupdatafile
\input\jobname.#1
\endgroup}
\message{insertions,}
% including footnotes.
\newcount \footnoteno
% The trailing space in the following definition for supereject is
% vital for proper filling; pages come out unaligned when you do a
% pagealignmacro call if that space before the closing brace is
% removed. (Generally, numeric constants should always be followed by a
% space to prevent strange expansion errors.)
\def\supereject{\par\penalty -20000\footnoteno =0 }
% @footnotestyle is meaningful for info output only.
\let\footnotestyle=\comment
{\catcode `\@=11
%
% Auto-number footnotes. Otherwise like plain.
\gdef\footnote{%
\let\indent=\ptexindent
\let\noindent=\ptexnoindent
\global\advance\footnoteno by \@ne
\edef\thisfootno{$^{\the\footnoteno}$}%
%
% In case the footnote comes at the end of a sentence, preserve the
% extra spacing after we do the footnote number.
\let\@sf\empty
\ifhmode\edef\@sf{\spacefactor\the\spacefactor}\ptexslash\fi
%
% Remove inadvertent blank space before typesetting the footnote number.
\unskip
\thisfootno\@sf
\dofootnote
}%
% Don't bother with the trickery in plain.tex to not require the
% footnote text as a parameter. Our footnotes don't need to be so general.
%
% Oh yes, they do; otherwise, @ifset (and anything else that uses
% \parseargline) fails inside footnotes because the tokens are fixed when
% the footnote is read. --karl, 16nov96.
%
\gdef\dofootnote{%
\insert\footins\bgroup
% We want to typeset this text as a normal paragraph, even if the
% footnote reference occurs in (for example) a display environment.
% So reset some parameters.
\hsize=\pagewidth
\interlinepenalty\interfootnotelinepenalty
\splittopskip\ht\strutbox % top baseline for broken footnotes
\splitmaxdepth\dp\strutbox
\floatingpenalty\@MM
\leftskip\z@skip
\rightskip\z@skip
\spaceskip\z@skip
\xspaceskip\z@skip
\parindent\defaultparindent
%
\smallfonts \rm
%
% Because we use hanging indentation in footnotes, a @noindent appears
% to exdent this text, so make it be a no-op. makeinfo does not use
% hanging indentation so @noindent can still be needed within footnote
% text after an @example or the like (not that this is good style).
\let\noindent = \relax
%
% Hang the footnote text off the number. Use \everypar in case the
% footnote extends for more than one paragraph.
\everypar = {\hang}%
\textindent{\thisfootno}%
%
% Don't crash into the line above the footnote text. Since this
% expands into a box, it must come within the paragraph, lest it
% provide a place where TeX can split the footnote.
\footstrut
\futurelet\next\fo@t
}
}%end \catcode `\@=11
% In case a @footnote appears in a vbox, save the footnote text and create
% the real \insert just after the vbox finished. Otherwise, the insertion
% would be lost.
% Similarily, if a @footnote appears inside an alignment, save the footnote
% text to a box and make the \insert when a row of the table is finished.
% And the same can be done for other insert classes. --kasal, 16nov03.
% Replace the \insert primitive by a cheating macro.
% Deeper inside, just make sure that the saved insertions are not spilled
% out prematurely.
%
\def\startsavinginserts{%
\ifx \insert\ptexinsert
\let\insert\saveinsert
\else
\let\checkinserts\relax
\fi
}
% This \insert replacement works for both \insert\footins{foo} and
% \insert\footins\bgroup foo\egroup, but it doesn't work for \insert27{foo}.
%
\def\saveinsert#1{%
\edef\next{\noexpand\savetobox \makeSAVEname#1}%
\afterassignment\next
% swallow the left brace
\let\temp =
}
\def\makeSAVEname#1{\makecsname{SAVE\expandafter\gobble\string#1}}
\def\savetobox#1{\global\setbox#1 = \vbox\bgroup \unvbox#1}
\def\checksaveins#1{\ifvoid#1\else \placesaveins#1\fi}
\def\placesaveins#1{%
\ptexinsert \csname\expandafter\gobblesave\string#1\endcsname
{\box#1}%
}
% eat @SAVE -- beware, all of them have catcode \other:
{
\def\dospecials{\do S\do A\do V\do E} \uncatcodespecials % ;-)
\gdef\gobblesave @SAVE{}
}
% initialization:
\def\newsaveins #1{%
\edef\next{\noexpand\newsaveinsX \makeSAVEname#1}%
\next
}
\def\newsaveinsX #1{%
\csname newbox\endcsname #1%
\expandafter\def\expandafter\checkinserts\expandafter{\checkinserts
\checksaveins #1}%
}
% initialize:
\let\checkinserts\empty
\newsaveins\footins
\newsaveins\margin
% @image. We use the macros from epsf.tex to support this.
% If epsf.tex is not installed and @image is used, we complain.
%
% Check for and read epsf.tex up front. If we read it only at @image
% time, we might be inside a group, and then its definitions would get
% undone and the next image would fail.
\openin 1 = epsf.tex
\ifeof 1 \else
% Do not bother showing banner with epsf.tex v2.7k (available in
% doc/epsf.tex and on ctan).
\def\epsfannounce{\toks0 = }%
\input epsf.tex
\fi
\closein 1
%
% We will only complain once about lack of epsf.tex.
\newif\ifwarnednoepsf
\newhelp\noepsfhelp{epsf.tex must be installed for images to
work. It is also included in the Texinfo distribution, or you can get
it from ftp://tug.org/tex/epsf.tex.}
%
\def\image#1{%
\ifx\epsfbox\undefined
\ifwarnednoepsf \else
\errhelp = \noepsfhelp
\errmessage{epsf.tex not found, images will be ignored}%
\global\warnednoepsftrue
\fi
\else
\imagexxx #1,,,,,\finish
\fi
}
%
% Arguments to @image:
% #1 is (mandatory) image filename; we tack on .eps extension.
% #2 is (optional) width, #3 is (optional) height.
% #4 is (ignored optional) html alt text.
% #5 is (ignored optional) extension.
% #6 is just the usual extra ignored arg for parsing this stuff.
\newif\ifimagevmode
\def\imagexxx#1,#2,#3,#4,#5,#6\finish{\begingroup
\catcode`\^^M = 5 % in case we're inside an example
\normalturnoffactive % allow _ et al. in names
% If the image is by itself, center it.
\ifvmode
\imagevmodetrue
\nobreak\bigskip
% Usually we'll have text after the image which will insert
% \parskip glue, so insert it here too to equalize the space
% above and below.
\nobreak\vskip\parskip
\nobreak
\line\bgroup
\fi
%
% Output the image.
\ifpdf
\dopdfimage{#1}{#2}{#3}%
\else
% \epsfbox itself resets \epsf?size at each figure.
\setbox0 = \hbox{\ignorespaces #2}\ifdim\wd0 > 0pt \epsfxsize=#2\relax \fi
\setbox0 = \hbox{\ignorespaces #3}\ifdim\wd0 > 0pt \epsfysize=#3\relax \fi
\epsfbox{#1.eps}%
\fi
%
\ifimagevmode \egroup \bigbreak \fi % space after the image
\endgroup}
% @float FLOATTYPE,LABEL,LOC ... @end float for displayed figures, tables,
% etc. We don't actually implement floating yet, we always include the
% float "here". But it seemed the best name for the future.
%
\envparseargdef\float{\eatcommaspace\eatcommaspace\dofloat#1, , ,\finish}
% There may be a space before second and/or third parameter; delete it.
\def\eatcommaspace#1, {#1,}
% #1 is the optional FLOATTYPE, the text label for this float, typically
% "Figure", "Table", "Example", etc. Can't contain commas. If omitted,
% this float will not be numbered and cannot be referred to.
%
% #2 is the optional xref label. Also must be present for the float to
% be referable.
%
% #3 is the optional positioning argument; for now, it is ignored. It
% will somehow specify the positions allowed to float to (here, top, bottom).
%
% We keep a separate counter for each FLOATTYPE, which we reset at each
% chapter-level command.
\let\resetallfloatnos=\empty
%
\def\dofloat#1,#2,#3,#4\finish{%
\let\thiscaption=\empty
\let\thisshortcaption=\empty
%
% don't lose footnotes inside @float.
%
% BEWARE: when the floats start float, we have to issue warning whenever an
% insert appears inside a float which could possibly float. --kasal, 26may04
%
\startsavinginserts
%
% We can't be used inside a paragraph.
\par
%
\vtop\bgroup
\def\floattype{#1}%
\def\floatlabel{#2}%
\def\floatloc{#3}% we do nothing with this yet.
%
\ifx\floattype\empty
\let\safefloattype=\empty
\else
{%
% the floattype might have accents or other special characters,
% but we need to use it in a control sequence name.
\indexnofonts
\turnoffactive
\xdef\safefloattype{\floattype}%
}%
\fi
%
% If label is given but no type, we handle that as the empty type.
\ifx\floatlabel\empty \else
% We want each FLOATTYPE to be numbered separately (Figure 1,
% Table 1, Figure 2, ...). (And if no label, no number.)
%
\expandafter\getfloatno\csname\safefloattype floatno\endcsname
\global\advance\floatno by 1
%
{%
% This magic value for \thissection is output by \setref as the
% XREFLABEL-title value. \xrefX uses it to distinguish float
% labels (which have a completely different output format) from
% node and anchor labels. And \xrdef uses it to construct the
% lists of floats.
%
\edef\thissection{\floatmagic=\safefloattype}%
\setref{\floatlabel}{Yfloat}%
}%
\fi
%
% start with \parskip glue, I guess.
\vskip\parskip
%
% Don't suppress indentation if a float happens to start a section.
\restorefirstparagraphindent
}
% we have these possibilities:
% @float Foo,lbl & @caption{Cap}: Foo 1.1: Cap
% @float Foo,lbl & no caption: Foo 1.1
% @float Foo & @caption{Cap}: Foo: Cap
% @float Foo & no caption: Foo
% @float ,lbl & Caption{Cap}: 1.1: Cap
% @float ,lbl & no caption: 1.1
% @float & @caption{Cap}: Cap
% @float & no caption:
%
\def\Efloat{%
\let\floatident = \empty
%
% In all cases, if we have a float type, it comes first.
\ifx\floattype\empty \else \def\floatident{\floattype}\fi
%
% If we have an xref label, the number comes next.
\ifx\floatlabel\empty \else
\ifx\floattype\empty \else % if also had float type, need tie first.
\appendtomacro\floatident{\tie}%
\fi
% the number.
\appendtomacro\floatident{\chaplevelprefix\the\floatno}%
\fi
%
% Start the printed caption with what we've constructed in
% \floatident, but keep it separate; we need \floatident again.
\let\captionline = \floatident
%
\ifx\thiscaption\empty \else
\ifx\floatident\empty \else
\appendtomacro\captionline{: }% had ident, so need a colon between
\fi
%
% caption text.
\appendtomacro\captionline{\scanexp\thiscaption}%
\fi
%
% If we have anything to print, print it, with space before.
% Eventually this needs to become an \insert.
\ifx\captionline\empty \else
\vskip.5\parskip
\captionline
%
% Space below caption.
\vskip\parskip
\fi
%
% If have an xref label, write the list of floats info. Do this
% after the caption, to avoid chance of it being a breakpoint.
\ifx\floatlabel\empty \else
% Write the text that goes in the lof to the aux file as
% \floatlabel-lof. Besides \floatident, we include the short
% caption if specified, else the full caption if specified, else nothing.
{%
\atdummies
%
% since we read the caption text in the macro world, where ^^M
% is turned into a normal character, we have to scan it back, so
% we don't write the literal three characters "^^M" into the aux file.
\scanexp{%
\xdef\noexpand\gtemp{%
\ifx\thisshortcaption\empty
\thiscaption
\else
\thisshortcaption
\fi
}%
}%
\immediate\write\auxfile{@xrdef{\floatlabel-lof}{\floatident
\ifx\gtemp\empty \else : \gtemp \fi}}%
}%
\fi
\egroup % end of \vtop
%
% place the captured inserts
%
% BEWARE: when the floats start floating, we have to issue warning
% whenever an insert appears inside a float which could possibly
% float. --kasal, 26may04
%
\checkinserts
}
% Append the tokens #2 to the definition of macro #1, not expanding either.
%
\def\appendtomacro#1#2{%
\expandafter\def\expandafter#1\expandafter{#1#2}%
}
% @caption, @shortcaption
%
\def\caption{\docaption\thiscaption}
\def\shortcaption{\docaption\thisshortcaption}
\def\docaption{\checkenv\float \bgroup\scanargctxt\defcaption}
\def\defcaption#1#2{\egroup \def#1{#2}}
% The parameter is the control sequence identifying the counter we are
% going to use. Create it if it doesn't exist and assign it to \floatno.
\def\getfloatno#1{%
\ifx#1\relax
% Haven't seen this figure type before.
\csname newcount\endcsname #1%
%
% Remember to reset this floatno at the next chap.
\expandafter\gdef\expandafter\resetallfloatnos
\expandafter{\resetallfloatnos #1=0 }%
\fi
\let\floatno#1%
}
% \setref calls this to get the XREFLABEL-snt value. We want an @xref
% to the FLOATLABEL to expand to "Figure 3.1". We call \setref when we
% first read the @float command.
%
\def\Yfloat{\floattype@tie \chaplevelprefix\the\floatno}%
% Magic string used for the XREFLABEL-title value, so \xrefX can
% distinguish floats from other xref types.
\def\floatmagic{!!float!!}
% #1 is the control sequence we are passed; we expand into a conditional
% which is true if #1 represents a float ref. That is, the magic
% \thissection value which we \setref above.
%
\def\iffloat#1{\expandafter\doiffloat#1==\finish}
%
% #1 is (maybe) the \floatmagic string. If so, #2 will be the
% (safe) float type for this float. We set \iffloattype to #2.
%
\def\doiffloat#1=#2=#3\finish{%
\def\temp{#1}%
\def\iffloattype{#2}%
\ifx\temp\floatmagic
}
% @listoffloats FLOATTYPE - print a list of floats like a table of contents.
%
\parseargdef\listoffloats{%
\def\floattype{#1}% floattype
{%
% the floattype might have accents or other special characters,
% but we need to use it in a control sequence name.
\indexnofonts
\turnoffactive
\xdef\safefloattype{\floattype}%
}%
%
% \xrdef saves the floats as a \do-list in \floatlistSAFEFLOATTYPE.
\expandafter\ifx\csname floatlist\safefloattype\endcsname \relax
\ifhavexrefs
% if the user said @listoffloats foo but never @float foo.
\message{\linenumber No `\safefloattype' floats to list.}%
\fi
\else
\begingroup
\leftskip=\tocindent % indent these entries like a toc
\let\do=\listoffloatsdo
\csname floatlist\safefloattype\endcsname
\endgroup
\fi
}
% This is called on each entry in a list of floats. We're passed the
% xref label, in the form LABEL-title, which is how we save it in the
% aux file. We strip off the -title and look up \XRLABEL-lof, which
% has the text we're supposed to typeset here.
%
% Figures without xref labels will not be included in the list (since
% they won't appear in the aux file).
%
\def\listoffloatsdo#1{\listoffloatsdoentry#1\finish}
\def\listoffloatsdoentry#1-title\finish{{%
% Can't fully expand XR#1-lof because it can contain anything. Just
% pass the control sequence. On the other hand, XR#1-pg is just the
% page number, and we want to fully expand that so we can get a link
% in pdf output.
\toksA = \expandafter{\csname XR#1-lof\endcsname}%
%
% use the same \entry macro we use to generate the TOC and index.
\edef\writeentry{\noexpand\entry{\the\toksA}{\csname XR#1-pg\endcsname}}%
\writeentry
}}
\message{localization,}
% @documentlanguage is usually given very early, just after
% @setfilename. If done too late, it may not override everything
% properly. Single argument is the language abbreviation.
% It would be nice if we could set up a hyphenation file here.
%
\parseargdef\documentlanguage{%
\tex % read txi-??.tex file in plain TeX.
% Read the file if it exists.
\openin 1 txi-#1.tex
\ifeof 1
\errhelp = \nolanghelp
\errmessage{Cannot read language file txi-#1.tex}%
\else
\input txi-#1.tex
\fi
\closein 1
\endgroup
}
\newhelp\nolanghelp{The given language definition file cannot be found or
is empty. Maybe you need to install it? In the current directory
should work if nowhere else does.}
% Set the catcode of characters 128 through 255 to the specified number.
%
\def\setnonasciicharscatcode#1{%
\count255=128
\loop\ifnum\count255<256
\global\catcode\count255=#1
\advance\count255 by 1
\repeat
}
% @documentencoding sets the definition of non-ASCII characters
% according to the specified encoding.
%
\parseargdef\documentencoding{%
% Encoding being declared for the document.
\def\declaredencoding{\csname #1.enc\endcsname}%
%
% Supported encodings: names converted to tokens in order to be able
% to compare them with \ifx.
\def\ascii{\csname US-ASCII.enc\endcsname}%
\def\latnine{\csname ISO-8859-15.enc\endcsname}%
\def\latone{\csname ISO-8859-1.enc\endcsname}%
\def\lattwo{\csname ISO-8859-2.enc\endcsname}%
\def\utfeight{\csname UTF-8.enc\endcsname}%
%
\ifx \declaredencoding \ascii
\asciichardefs
%
\else \ifx \declaredencoding \lattwo
\setnonasciicharscatcode\active
\lattwochardefs
%
\else \ifx \declaredencoding \latone
\setnonasciicharscatcode\active
\latonechardefs
%
\else \ifx \declaredencoding \latnine
\setnonasciicharscatcode\active
\latninechardefs
%
\else \ifx \declaredencoding \utfeight
\setnonasciicharscatcode\active
\utfeightchardefs
%
\else
\message{Unknown document encoding #1, ignoring.}%
%
\fi % utfeight
\fi % latnine
\fi % latone
\fi % lattwo
\fi % ascii
}
% A message to be logged when using a character that isn't available
% the default font encoding (OT1).
%
\def\missingcharmsg#1{\message{Character missing in OT1 encoding: #1.}}
% Take account of \c (plain) vs. \, (Texinfo) difference.
\def\cedilla#1{\ifx\c\ptexc\c{#1}\else\,{#1}\fi}
% First, make active non-ASCII characters in order for them to be
% correctly categorized when TeX reads the replacement text of
% macros containing the character definitions.
\setnonasciicharscatcode\active
%
% Latin1 (ISO-8859-1) character definitions.
\def\latonechardefs{%
\gdef^^a0{~}
\gdef^^a1{\exclamdown}
\gdef^^a2{\missingcharmsg{CENT SIGN}}
\gdef^^a3{{\pounds}}
\gdef^^a4{\missingcharmsg{CURRENCY SIGN}}
\gdef^^a5{\missingcharmsg{YEN SIGN}}
\gdef^^a6{\missingcharmsg{BROKEN BAR}}
\gdef^^a7{\S}
\gdef^^a8{\"{}}
\gdef^^a9{\copyright}
\gdef^^aa{\ordf}
\gdef^^ab{\missingcharmsg{LEFT-POINTING DOUBLE ANGLE QUOTATION MARK}}
\gdef^^ac{$\lnot$}
\gdef^^ad{\-}
\gdef^^ae{\registeredsymbol}
\gdef^^af{\={}}
%
\gdef^^b0{\textdegree}
\gdef^^b1{$\pm$}
\gdef^^b2{$^2$}
\gdef^^b3{$^3$}
\gdef^^b4{\'{}}
\gdef^^b5{$\mu$}
\gdef^^b6{\P}
%
\gdef^^b7{$^.$}
\gdef^^b8{\cedilla\ }
\gdef^^b9{$^1$}
\gdef^^ba{\ordm}
%
\gdef^^bb{\missingcharmsg{RIGHT-POINTING DOUBLE ANGLE QUOTATION MARK}}
\gdef^^bc{$1\over4$}
\gdef^^bd{$1\over2$}
\gdef^^be{$3\over4$}
\gdef^^bf{\questiondown}
%
\gdef^^c0{\`A}
\gdef^^c1{\'A}
\gdef^^c2{\^A}
\gdef^^c3{\~A}
\gdef^^c4{\"A}
\gdef^^c5{\ringaccent A}
\gdef^^c6{\AE}
\gdef^^c7{\cedilla C}
\gdef^^c8{\`E}
\gdef^^c9{\'E}
\gdef^^ca{\^E}
\gdef^^cb{\"E}
\gdef^^cc{\`I}
\gdef^^cd{\'I}
\gdef^^ce{\^I}
\gdef^^cf{\"I}
%
\gdef^^d0{\missingcharmsg{LATIN CAPITAL LETTER ETH}}
\gdef^^d1{\~N}
\gdef^^d2{\`O}
\gdef^^d3{\'O}
\gdef^^d4{\^O}
\gdef^^d5{\~O}
\gdef^^d6{\"O}
\gdef^^d7{$\times$}
\gdef^^d8{\O}
\gdef^^d9{\`U}
\gdef^^da{\'U}
\gdef^^db{\^U}
\gdef^^dc{\"U}
\gdef^^dd{\'Y}
\gdef^^de{\missingcharmsg{LATIN CAPITAL LETTER THORN}}
\gdef^^df{\ss}
%
\gdef^^e0{\`a}
\gdef^^e1{\'a}
\gdef^^e2{\^a}
\gdef^^e3{\~a}
\gdef^^e4{\"a}
\gdef^^e5{\ringaccent a}
\gdef^^e6{\ae}
\gdef^^e7{\cedilla c}
\gdef^^e8{\`e}
\gdef^^e9{\'e}
\gdef^^ea{\^e}
\gdef^^eb{\"e}
\gdef^^ec{\`{\dotless i}}
\gdef^^ed{\'{\dotless i}}
\gdef^^ee{\^{\dotless i}}
\gdef^^ef{\"{\dotless i}}
%
\gdef^^f0{\missingcharmsg{LATIN SMALL LETTER ETH}}
\gdef^^f1{\~n}
\gdef^^f2{\`o}
\gdef^^f3{\'o}
\gdef^^f4{\^o}
\gdef^^f5{\~o}
\gdef^^f6{\"o}
\gdef^^f7{$\div$}
\gdef^^f8{\o}
\gdef^^f9{\`u}
\gdef^^fa{\'u}
\gdef^^fb{\^u}
\gdef^^fc{\"u}
\gdef^^fd{\'y}
\gdef^^fe{\missingcharmsg{LATIN SMALL LETTER THORN}}
\gdef^^ff{\"y}
}
% Latin9 (ISO-8859-15) encoding character definitions.
\def\latninechardefs{%
% Encoding is almost identical to Latin1.
\latonechardefs
%
\gdef^^a4{\euro}
\gdef^^a6{\v S}
\gdef^^a8{\v s}
\gdef^^b4{\v Z}
\gdef^^b8{\v z}
\gdef^^bc{\OE}
\gdef^^bd{\oe}
\gdef^^be{\"Y}
}
% Latin2 (ISO-8859-2) character definitions.
\def\lattwochardefs{%
\gdef^^a0{~}
\gdef^^a1{\missingcharmsg{LATIN CAPITAL LETTER A WITH OGONEK}}
\gdef^^a2{\u{}}
\gdef^^a3{\L}
\gdef^^a4{\missingcharmsg{CURRENCY SIGN}}
\gdef^^a5{\v L}
\gdef^^a6{\'S}
\gdef^^a7{\S}
\gdef^^a8{\"{}}
\gdef^^a9{\v S}
\gdef^^aa{\cedilla S}
\gdef^^ab{\v T}
\gdef^^ac{\'Z}
\gdef^^ad{\-}
\gdef^^ae{\v Z}
\gdef^^af{\dotaccent Z}
%
\gdef^^b0{\textdegree}
\gdef^^b1{\missingcharmsg{LATIN SMALL LETTER A WITH OGONEK}}
\gdef^^b2{\missingcharmsg{OGONEK}}
\gdef^^b3{\l}
\gdef^^b4{\'{}}
\gdef^^b5{\v l}
\gdef^^b6{\'s}
\gdef^^b7{\v{}}
\gdef^^b8{\cedilla\ }
\gdef^^b9{\v s}
\gdef^^ba{\cedilla s}
\gdef^^bb{\v t}
\gdef^^bc{\'z}
\gdef^^bd{\H{}}
\gdef^^be{\v z}
\gdef^^bf{\dotaccent z}
%
\gdef^^c0{\'R}
\gdef^^c1{\'A}
\gdef^^c2{\^A}
\gdef^^c3{\u A}
\gdef^^c4{\"A}
\gdef^^c5{\'L}
\gdef^^c6{\'C}
\gdef^^c7{\cedilla C}
\gdef^^c8{\v C}
\gdef^^c9{\'E}
\gdef^^ca{\missingcharmsg{LATIN CAPITAL LETTER E WITH OGONEK}}
\gdef^^cb{\"E}
\gdef^^cc{\v E}
\gdef^^cd{\'I}
\gdef^^ce{\^I}
\gdef^^cf{\v D}
%
\gdef^^d0{\missingcharmsg{LATIN CAPITAL LETTER D WITH STROKE}}
\gdef^^d1{\'N}
\gdef^^d2{\v N}
\gdef^^d3{\'O}
\gdef^^d4{\^O}
\gdef^^d5{\H O}
\gdef^^d6{\"O}
\gdef^^d7{$\times$}
\gdef^^d8{\v R}
\gdef^^d9{\ringaccent U}
\gdef^^da{\'U}
\gdef^^db{\H U}
\gdef^^dc{\"U}
\gdef^^dd{\'Y}
\gdef^^de{\cedilla T}
\gdef^^df{\ss}
%
\gdef^^e0{\'r}
\gdef^^e1{\'a}
\gdef^^e2{\^a}
\gdef^^e3{\u a}
\gdef^^e4{\"a}
\gdef^^e5{\'l}
\gdef^^e6{\'c}
\gdef^^e7{\cedilla c}
\gdef^^e8{\v c}
\gdef^^e9{\'e}
\gdef^^ea{\missingcharmsg{LATIN SMALL LETTER E WITH OGONEK}}
\gdef^^eb{\"e}
\gdef^^ec{\v e}
\gdef^^ed{\'\i}
\gdef^^ee{\^\i}
\gdef^^ef{\v d}
%
\gdef^^f0{\missingcharmsg{LATIN SMALL LETTER D WITH STROKE}}
\gdef^^f1{\'n}
\gdef^^f2{\v n}
\gdef^^f3{\'o}
\gdef^^f4{\^o}
\gdef^^f5{\H o}
\gdef^^f6{\"o}
\gdef^^f7{$\div$}
\gdef^^f8{\v r}
\gdef^^f9{\ringaccent u}
\gdef^^fa{\'u}
\gdef^^fb{\H u}
\gdef^^fc{\"u}
\gdef^^fd{\'y}
\gdef^^fe{\cedilla t}
\gdef^^ff{\dotaccent{}}
}
% UTF-8 character definitions.
%
% This code to support UTF-8 is based on LaTeX's utf8.def, with some
% changes for Texinfo conventions. It is included here under the GPL by
% permission from Frank Mittelbach and the LaTeX team.
%
\newcount\countUTFx
\newcount\countUTFy
\newcount\countUTFz
\gdef\UTFviiiTwoOctets#1#2{\expandafter
\UTFviiiDefined\csname u8:#1\string #2\endcsname}
%
\gdef\UTFviiiThreeOctets#1#2#3{\expandafter
\UTFviiiDefined\csname u8:#1\string #2\string #3\endcsname}
%
\gdef\UTFviiiFourOctets#1#2#3#4{\expandafter
\UTFviiiDefined\csname u8:#1\string #2\string #3\string #4\endcsname}
\gdef\UTFviiiDefined#1{%
\ifx #1\relax
\message{\linenumber Unicode char \string #1 not defined for Texinfo}%
\else
\expandafter #1%
\fi
}
\begingroup
\catcode`\~13
\catcode`\"12
\def\UTFviiiLoop{%
\global\catcode\countUTFx\active
\uccode`\~\countUTFx
\uppercase\expandafter{\UTFviiiTmp}%
\advance\countUTFx by 1
\ifnum\countUTFx < \countUTFy
\expandafter\UTFviiiLoop
\fi}
\countUTFx = "C2
\countUTFy = "E0
\def\UTFviiiTmp{%
\xdef~{\noexpand\UTFviiiTwoOctets\string~}}
\UTFviiiLoop
\countUTFx = "E0
\countUTFy = "F0
\def\UTFviiiTmp{%
\xdef~{\noexpand\UTFviiiThreeOctets\string~}}
\UTFviiiLoop
\countUTFx = "F0
\countUTFy = "F4
\def\UTFviiiTmp{%
\xdef~{\noexpand\UTFviiiFourOctets\string~}}
\UTFviiiLoop
\endgroup
\begingroup
\catcode`\"=12
\catcode`\<=12
\catcode`\.=12
\catcode`\,=12
\catcode`\;=12
\catcode`\!=12
\catcode`\~=13
\gdef\DeclareUnicodeCharacter#1#2{%
\countUTFz = "#1\relax
\wlog{\space\space defining Unicode char U+#1 (decimal \the\countUTFz)}%
\begingroup
\parseXMLCharref
\def\UTFviiiTwoOctets##1##2{%
\csname u8:##1\string ##2\endcsname}%
\def\UTFviiiThreeOctets##1##2##3{%
\csname u8:##1\string ##2\string ##3\endcsname}%
\def\UTFviiiFourOctets##1##2##3##4{%
\csname u8:##1\string ##2\string ##3\string ##4\endcsname}%
\expandafter\expandafter\expandafter\expandafter
\expandafter\expandafter\expandafter
\gdef\UTFviiiTmp{#2}%
\endgroup}
\gdef\parseXMLCharref{%
\ifnum\countUTFz < "A0\relax
\errhelp = \EMsimple
\errmessage{Cannot define Unicode char value < 00A0}%
\else\ifnum\countUTFz < "800\relax
\parseUTFviiiA,%
\parseUTFviiiB C\UTFviiiTwoOctets.,%
\else\ifnum\countUTFz < "10000\relax
\parseUTFviiiA;%
\parseUTFviiiA,%
\parseUTFviiiB E\UTFviiiThreeOctets.{,;}%
\else
\parseUTFviiiA;%
\parseUTFviiiA,%
\parseUTFviiiA!%
\parseUTFviiiB F\UTFviiiFourOctets.{!,;}%
\fi\fi\fi
}
\gdef\parseUTFviiiA#1{%
\countUTFx = \countUTFz
\divide\countUTFz by 64
\countUTFy = \countUTFz
\multiply\countUTFz by 64
\advance\countUTFx by -\countUTFz
\advance\countUTFx by 128
\uccode `#1\countUTFx
\countUTFz = \countUTFy}
\gdef\parseUTFviiiB#1#2#3#4{%
\advance\countUTFz by "#10\relax
\uccode `#3\countUTFz
\uppercase{\gdef\UTFviiiTmp{#2#3#4}}}
\endgroup
\def\utfeightchardefs{%
\DeclareUnicodeCharacter{00A0}{\tie}
\DeclareUnicodeCharacter{00A1}{\exclamdown}
\DeclareUnicodeCharacter{00A3}{\pounds}
\DeclareUnicodeCharacter{00A8}{\"{ }}
\DeclareUnicodeCharacter{00A9}{\copyright}
\DeclareUnicodeCharacter{00AA}{\ordf}
\DeclareUnicodeCharacter{00AD}{\-}
\DeclareUnicodeCharacter{00AE}{\registeredsymbol}
\DeclareUnicodeCharacter{00AF}{\={ }}
\DeclareUnicodeCharacter{00B0}{\ringaccent{ }}
\DeclareUnicodeCharacter{00B4}{\'{ }}
\DeclareUnicodeCharacter{00B8}{\cedilla{ }}
\DeclareUnicodeCharacter{00BA}{\ordm}
\DeclareUnicodeCharacter{00BF}{\questiondown}
\DeclareUnicodeCharacter{00C0}{\`A}
\DeclareUnicodeCharacter{00C1}{\'A}
\DeclareUnicodeCharacter{00C2}{\^A}
\DeclareUnicodeCharacter{00C3}{\~A}
\DeclareUnicodeCharacter{00C4}{\"A}
\DeclareUnicodeCharacter{00C5}{\AA}
\DeclareUnicodeCharacter{00C6}{\AE}
\DeclareUnicodeCharacter{00C7}{\cedilla{C}}
\DeclareUnicodeCharacter{00C8}{\`E}
\DeclareUnicodeCharacter{00C9}{\'E}
\DeclareUnicodeCharacter{00CA}{\^E}
\DeclareUnicodeCharacter{00CB}{\"E}
\DeclareUnicodeCharacter{00CC}{\`I}
\DeclareUnicodeCharacter{00CD}{\'I}
\DeclareUnicodeCharacter{00CE}{\^I}
\DeclareUnicodeCharacter{00CF}{\"I}
\DeclareUnicodeCharacter{00D1}{\~N}
\DeclareUnicodeCharacter{00D2}{\`O}
\DeclareUnicodeCharacter{00D3}{\'O}
\DeclareUnicodeCharacter{00D4}{\^O}
\DeclareUnicodeCharacter{00D5}{\~O}
\DeclareUnicodeCharacter{00D6}{\"O}
\DeclareUnicodeCharacter{00D8}{\O}
\DeclareUnicodeCharacter{00D9}{\`U}
\DeclareUnicodeCharacter{00DA}{\'U}
\DeclareUnicodeCharacter{00DB}{\^U}
\DeclareUnicodeCharacter{00DC}{\"U}
\DeclareUnicodeCharacter{00DD}{\'Y}
\DeclareUnicodeCharacter{00DF}{\ss}
\DeclareUnicodeCharacter{00E0}{\`a}
\DeclareUnicodeCharacter{00E1}{\'a}
\DeclareUnicodeCharacter{00E2}{\^a}
\DeclareUnicodeCharacter{00E3}{\~a}
\DeclareUnicodeCharacter{00E4}{\"a}
\DeclareUnicodeCharacter{00E5}{\aa}
\DeclareUnicodeCharacter{00E6}{\ae}
\DeclareUnicodeCharacter{00E7}{\cedilla{c}}
\DeclareUnicodeCharacter{00E8}{\`e}
\DeclareUnicodeCharacter{00E9}{\'e}
\DeclareUnicodeCharacter{00EA}{\^e}
\DeclareUnicodeCharacter{00EB}{\"e}
\DeclareUnicodeCharacter{00EC}{\`{\dotless{i}}}
\DeclareUnicodeCharacter{00ED}{\'{\dotless{i}}}
\DeclareUnicodeCharacter{00EE}{\^{\dotless{i}}}
\DeclareUnicodeCharacter{00EF}{\"{\dotless{i}}}
\DeclareUnicodeCharacter{00F1}{\~n}
\DeclareUnicodeCharacter{00F2}{\`o}
\DeclareUnicodeCharacter{00F3}{\'o}
\DeclareUnicodeCharacter{00F4}{\^o}
\DeclareUnicodeCharacter{00F5}{\~o}
\DeclareUnicodeCharacter{00F6}{\"o}
\DeclareUnicodeCharacter{00F8}{\o}
\DeclareUnicodeCharacter{00F9}{\`u}
\DeclareUnicodeCharacter{00FA}{\'u}
\DeclareUnicodeCharacter{00FB}{\^u}
\DeclareUnicodeCharacter{00FC}{\"u}
\DeclareUnicodeCharacter{00FD}{\'y}
\DeclareUnicodeCharacter{00FF}{\"y}
\DeclareUnicodeCharacter{0100}{\=A}
\DeclareUnicodeCharacter{0101}{\=a}
\DeclareUnicodeCharacter{0102}{\u{A}}
\DeclareUnicodeCharacter{0103}{\u{a}}
\DeclareUnicodeCharacter{0106}{\'C}
\DeclareUnicodeCharacter{0107}{\'c}
\DeclareUnicodeCharacter{0108}{\^C}
\DeclareUnicodeCharacter{0109}{\^c}
\DeclareUnicodeCharacter{010A}{\dotaccent{C}}
\DeclareUnicodeCharacter{010B}{\dotaccent{c}}
\DeclareUnicodeCharacter{010C}{\v{C}}
\DeclareUnicodeCharacter{010D}{\v{c}}
\DeclareUnicodeCharacter{010E}{\v{D}}
\DeclareUnicodeCharacter{0112}{\=E}
\DeclareUnicodeCharacter{0113}{\=e}
\DeclareUnicodeCharacter{0114}{\u{E}}
\DeclareUnicodeCharacter{0115}{\u{e}}
\DeclareUnicodeCharacter{0116}{\dotaccent{E}}
\DeclareUnicodeCharacter{0117}{\dotaccent{e}}
\DeclareUnicodeCharacter{011A}{\v{E}}
\DeclareUnicodeCharacter{011B}{\v{e}}
\DeclareUnicodeCharacter{011C}{\^G}
\DeclareUnicodeCharacter{011D}{\^g}
\DeclareUnicodeCharacter{011E}{\u{G}}
\DeclareUnicodeCharacter{011F}{\u{g}}
\DeclareUnicodeCharacter{0120}{\dotaccent{G}}
\DeclareUnicodeCharacter{0121}{\dotaccent{g}}
\DeclareUnicodeCharacter{0124}{\^H}
\DeclareUnicodeCharacter{0125}{\^h}
\DeclareUnicodeCharacter{0128}{\~I}
\DeclareUnicodeCharacter{0129}{\~{\dotless{i}}}
\DeclareUnicodeCharacter{012A}{\=I}
\DeclareUnicodeCharacter{012B}{\={\dotless{i}}}
\DeclareUnicodeCharacter{012C}{\u{I}}
\DeclareUnicodeCharacter{012D}{\u{\dotless{i}}}
\DeclareUnicodeCharacter{0130}{\dotaccent{I}}
\DeclareUnicodeCharacter{0131}{\dotless{i}}
\DeclareUnicodeCharacter{0132}{IJ}
\DeclareUnicodeCharacter{0133}{ij}
\DeclareUnicodeCharacter{0134}{\^J}
\DeclareUnicodeCharacter{0135}{\^{\dotless{j}}}
\DeclareUnicodeCharacter{0139}{\'L}
\DeclareUnicodeCharacter{013A}{\'l}
\DeclareUnicodeCharacter{0141}{\L}
\DeclareUnicodeCharacter{0142}{\l}
\DeclareUnicodeCharacter{0143}{\'N}
\DeclareUnicodeCharacter{0144}{\'n}
\DeclareUnicodeCharacter{0147}{\v{N}}
\DeclareUnicodeCharacter{0148}{\v{n}}
\DeclareUnicodeCharacter{014C}{\=O}
\DeclareUnicodeCharacter{014D}{\=o}
\DeclareUnicodeCharacter{014E}{\u{O}}
\DeclareUnicodeCharacter{014F}{\u{o}}
\DeclareUnicodeCharacter{0150}{\H{O}}
\DeclareUnicodeCharacter{0151}{\H{o}}
\DeclareUnicodeCharacter{0152}{\OE}
\DeclareUnicodeCharacter{0153}{\oe}
\DeclareUnicodeCharacter{0154}{\'R}
\DeclareUnicodeCharacter{0155}{\'r}
\DeclareUnicodeCharacter{0158}{\v{R}}
\DeclareUnicodeCharacter{0159}{\v{r}}
\DeclareUnicodeCharacter{015A}{\'S}
\DeclareUnicodeCharacter{015B}{\'s}
\DeclareUnicodeCharacter{015C}{\^S}
\DeclareUnicodeCharacter{015D}{\^s}
\DeclareUnicodeCharacter{015E}{\cedilla{S}}
\DeclareUnicodeCharacter{015F}{\cedilla{s}}
\DeclareUnicodeCharacter{0160}{\v{S}}
\DeclareUnicodeCharacter{0161}{\v{s}}
\DeclareUnicodeCharacter{0162}{\cedilla{t}}
\DeclareUnicodeCharacter{0163}{\cedilla{T}}
\DeclareUnicodeCharacter{0164}{\v{T}}
\DeclareUnicodeCharacter{0168}{\~U}
\DeclareUnicodeCharacter{0169}{\~u}
\DeclareUnicodeCharacter{016A}{\=U}
\DeclareUnicodeCharacter{016B}{\=u}
\DeclareUnicodeCharacter{016C}{\u{U}}
\DeclareUnicodeCharacter{016D}{\u{u}}
\DeclareUnicodeCharacter{016E}{\ringaccent{U}}
\DeclareUnicodeCharacter{016F}{\ringaccent{u}}
\DeclareUnicodeCharacter{0170}{\H{U}}
\DeclareUnicodeCharacter{0171}{\H{u}}
\DeclareUnicodeCharacter{0174}{\^W}
\DeclareUnicodeCharacter{0175}{\^w}
\DeclareUnicodeCharacter{0176}{\^Y}
\DeclareUnicodeCharacter{0177}{\^y}
\DeclareUnicodeCharacter{0178}{\"Y}
\DeclareUnicodeCharacter{0179}{\'Z}
\DeclareUnicodeCharacter{017A}{\'z}
\DeclareUnicodeCharacter{017B}{\dotaccent{Z}}
\DeclareUnicodeCharacter{017C}{\dotaccent{z}}
\DeclareUnicodeCharacter{017D}{\v{Z}}
\DeclareUnicodeCharacter{017E}{\v{z}}
\DeclareUnicodeCharacter{01C4}{D\v{Z}}
\DeclareUnicodeCharacter{01C5}{D\v{z}}
\DeclareUnicodeCharacter{01C6}{d\v{z}}
\DeclareUnicodeCharacter{01C7}{LJ}
\DeclareUnicodeCharacter{01C8}{Lj}
\DeclareUnicodeCharacter{01C9}{lj}
\DeclareUnicodeCharacter{01CA}{NJ}
\DeclareUnicodeCharacter{01CB}{Nj}
\DeclareUnicodeCharacter{01CC}{nj}
\DeclareUnicodeCharacter{01CD}{\v{A}}
\DeclareUnicodeCharacter{01CE}{\v{a}}
\DeclareUnicodeCharacter{01CF}{\v{I}}
\DeclareUnicodeCharacter{01D0}{\v{\dotless{i}}}
\DeclareUnicodeCharacter{01D1}{\v{O}}
\DeclareUnicodeCharacter{01D2}{\v{o}}
\DeclareUnicodeCharacter{01D3}{\v{U}}
\DeclareUnicodeCharacter{01D4}{\v{u}}
\DeclareUnicodeCharacter{01E2}{\={\AE}}
\DeclareUnicodeCharacter{01E3}{\={\ae}}
\DeclareUnicodeCharacter{01E6}{\v{G}}
\DeclareUnicodeCharacter{01E7}{\v{g}}
\DeclareUnicodeCharacter{01E8}{\v{K}}
\DeclareUnicodeCharacter{01E9}{\v{k}}
\DeclareUnicodeCharacter{01F0}{\v{\dotless{j}}}
\DeclareUnicodeCharacter{01F1}{DZ}
\DeclareUnicodeCharacter{01F2}{Dz}
\DeclareUnicodeCharacter{01F3}{dz}
\DeclareUnicodeCharacter{01F4}{\'G}
\DeclareUnicodeCharacter{01F5}{\'g}
\DeclareUnicodeCharacter{01F8}{\`N}
\DeclareUnicodeCharacter{01F9}{\`n}
\DeclareUnicodeCharacter{01FC}{\'{\AE}}
\DeclareUnicodeCharacter{01FD}{\'{\ae}}
\DeclareUnicodeCharacter{01FE}{\'{\O}}
\DeclareUnicodeCharacter{01FF}{\'{\o}}
\DeclareUnicodeCharacter{021E}{\v{H}}
\DeclareUnicodeCharacter{021F}{\v{h}}
\DeclareUnicodeCharacter{0226}{\dotaccent{A}}
\DeclareUnicodeCharacter{0227}{\dotaccent{a}}
\DeclareUnicodeCharacter{0228}{\cedilla{E}}
\DeclareUnicodeCharacter{0229}{\cedilla{e}}
\DeclareUnicodeCharacter{022E}{\dotaccent{O}}
\DeclareUnicodeCharacter{022F}{\dotaccent{o}}
\DeclareUnicodeCharacter{0232}{\=Y}
\DeclareUnicodeCharacter{0233}{\=y}
\DeclareUnicodeCharacter{0237}{\dotless{j}}
\DeclareUnicodeCharacter{1E02}{\dotaccent{B}}
\DeclareUnicodeCharacter{1E03}{\dotaccent{b}}
\DeclareUnicodeCharacter{1E04}{\udotaccent{B}}
\DeclareUnicodeCharacter{1E05}{\udotaccent{b}}
\DeclareUnicodeCharacter{1E06}{\ubaraccent{B}}
\DeclareUnicodeCharacter{1E07}{\ubaraccent{b}}
\DeclareUnicodeCharacter{1E0A}{\dotaccent{D}}
\DeclareUnicodeCharacter{1E0B}{\dotaccent{d}}
\DeclareUnicodeCharacter{1E0C}{\udotaccent{D}}
\DeclareUnicodeCharacter{1E0D}{\udotaccent{d}}
\DeclareUnicodeCharacter{1E0E}{\ubaraccent{D}}
\DeclareUnicodeCharacter{1E0F}{\ubaraccent{d}}
\DeclareUnicodeCharacter{1E1E}{\dotaccent{F}}
\DeclareUnicodeCharacter{1E1F}{\dotaccent{f}}
\DeclareUnicodeCharacter{1E20}{\=G}
\DeclareUnicodeCharacter{1E21}{\=g}
\DeclareUnicodeCharacter{1E22}{\dotaccent{H}}
\DeclareUnicodeCharacter{1E23}{\dotaccent{h}}
\DeclareUnicodeCharacter{1E24}{\udotaccent{H}}
\DeclareUnicodeCharacter{1E25}{\udotaccent{h}}
\DeclareUnicodeCharacter{1E26}{\"H}
\DeclareUnicodeCharacter{1E27}{\"h}
\DeclareUnicodeCharacter{1E30}{\'K}
\DeclareUnicodeCharacter{1E31}{\'k}
\DeclareUnicodeCharacter{1E32}{\udotaccent{K}}
\DeclareUnicodeCharacter{1E33}{\udotaccent{k}}
\DeclareUnicodeCharacter{1E34}{\ubaraccent{K}}
\DeclareUnicodeCharacter{1E35}{\ubaraccent{k}}
\DeclareUnicodeCharacter{1E36}{\udotaccent{L}}
\DeclareUnicodeCharacter{1E37}{\udotaccent{l}}
\DeclareUnicodeCharacter{1E3A}{\ubaraccent{L}}
\DeclareUnicodeCharacter{1E3B}{\ubaraccent{l}}
\DeclareUnicodeCharacter{1E3E}{\'M}
\DeclareUnicodeCharacter{1E3F}{\'m}
\DeclareUnicodeCharacter{1E40}{\dotaccent{M}}
\DeclareUnicodeCharacter{1E41}{\dotaccent{m}}
\DeclareUnicodeCharacter{1E42}{\udotaccent{M}}
\DeclareUnicodeCharacter{1E43}{\udotaccent{m}}
\DeclareUnicodeCharacter{1E44}{\dotaccent{N}}
\DeclareUnicodeCharacter{1E45}{\dotaccent{n}}
\DeclareUnicodeCharacter{1E46}{\udotaccent{N}}
\DeclareUnicodeCharacter{1E47}{\udotaccent{n}}
\DeclareUnicodeCharacter{1E48}{\ubaraccent{N}}
\DeclareUnicodeCharacter{1E49}{\ubaraccent{n}}
\DeclareUnicodeCharacter{1E54}{\'P}
\DeclareUnicodeCharacter{1E55}{\'p}
\DeclareUnicodeCharacter{1E56}{\dotaccent{P}}
\DeclareUnicodeCharacter{1E57}{\dotaccent{p}}
\DeclareUnicodeCharacter{1E58}{\dotaccent{R}}
\DeclareUnicodeCharacter{1E59}{\dotaccent{r}}
\DeclareUnicodeCharacter{1E5A}{\udotaccent{R}}
\DeclareUnicodeCharacter{1E5B}{\udotaccent{r}}
\DeclareUnicodeCharacter{1E5E}{\ubaraccent{R}}
\DeclareUnicodeCharacter{1E5F}{\ubaraccent{r}}
\DeclareUnicodeCharacter{1E60}{\dotaccent{S}}
\DeclareUnicodeCharacter{1E61}{\dotaccent{s}}
\DeclareUnicodeCharacter{1E62}{\udotaccent{S}}
\DeclareUnicodeCharacter{1E63}{\udotaccent{s}}
\DeclareUnicodeCharacter{1E6A}{\dotaccent{T}}
\DeclareUnicodeCharacter{1E6B}{\dotaccent{t}}
\DeclareUnicodeCharacter{1E6C}{\udotaccent{T}}
\DeclareUnicodeCharacter{1E6D}{\udotaccent{t}}
\DeclareUnicodeCharacter{1E6E}{\ubaraccent{T}}
\DeclareUnicodeCharacter{1E6F}{\ubaraccent{t}}
\DeclareUnicodeCharacter{1E7C}{\~V}
\DeclareUnicodeCharacter{1E7D}{\~v}
\DeclareUnicodeCharacter{1E7E}{\udotaccent{V}}
\DeclareUnicodeCharacter{1E7F}{\udotaccent{v}}
\DeclareUnicodeCharacter{1E80}{\`W}
\DeclareUnicodeCharacter{1E81}{\`w}
\DeclareUnicodeCharacter{1E82}{\'W}
\DeclareUnicodeCharacter{1E83}{\'w}
\DeclareUnicodeCharacter{1E84}{\"W}
\DeclareUnicodeCharacter{1E85}{\"w}
\DeclareUnicodeCharacter{1E86}{\dotaccent{W}}
\DeclareUnicodeCharacter{1E87}{\dotaccent{w}}
\DeclareUnicodeCharacter{1E88}{\udotaccent{W}}
\DeclareUnicodeCharacter{1E89}{\udotaccent{w}}
\DeclareUnicodeCharacter{1E8A}{\dotaccent{X}}
\DeclareUnicodeCharacter{1E8B}{\dotaccent{x}}
\DeclareUnicodeCharacter{1E8C}{\"X}
\DeclareUnicodeCharacter{1E8D}{\"x}
\DeclareUnicodeCharacter{1E8E}{\dotaccent{Y}}
\DeclareUnicodeCharacter{1E8F}{\dotaccent{y}}
\DeclareUnicodeCharacter{1E90}{\^Z}
\DeclareUnicodeCharacter{1E91}{\^z}
\DeclareUnicodeCharacter{1E92}{\udotaccent{Z}}
\DeclareUnicodeCharacter{1E93}{\udotaccent{z}}
\DeclareUnicodeCharacter{1E94}{\ubaraccent{Z}}
\DeclareUnicodeCharacter{1E95}{\ubaraccent{z}}
\DeclareUnicodeCharacter{1E96}{\ubaraccent{h}}
\DeclareUnicodeCharacter{1E97}{\"t}
\DeclareUnicodeCharacter{1E98}{\ringaccent{w}}
\DeclareUnicodeCharacter{1E99}{\ringaccent{y}}
\DeclareUnicodeCharacter{1EA0}{\udotaccent{A}}
\DeclareUnicodeCharacter{1EA1}{\udotaccent{a}}
\DeclareUnicodeCharacter{1EB8}{\udotaccent{E}}
\DeclareUnicodeCharacter{1EB9}{\udotaccent{e}}
\DeclareUnicodeCharacter{1EBC}{\~E}
\DeclareUnicodeCharacter{1EBD}{\~e}
\DeclareUnicodeCharacter{1ECA}{\udotaccent{I}}
\DeclareUnicodeCharacter{1ECB}{\udotaccent{i}}
\DeclareUnicodeCharacter{1ECC}{\udotaccent{O}}
\DeclareUnicodeCharacter{1ECD}{\udotaccent{o}}
\DeclareUnicodeCharacter{1EE4}{\udotaccent{U}}
\DeclareUnicodeCharacter{1EE5}{\udotaccent{u}}
\DeclareUnicodeCharacter{1EF2}{\`Y}
\DeclareUnicodeCharacter{1EF3}{\`y}
\DeclareUnicodeCharacter{1EF4}{\udotaccent{Y}}
\DeclareUnicodeCharacter{1EF8}{\~Y}
\DeclareUnicodeCharacter{1EF9}{\~y}
\DeclareUnicodeCharacter{2013}{--}
\DeclareUnicodeCharacter{2014}{---}
\DeclareUnicodeCharacter{2022}{\bullet}
\DeclareUnicodeCharacter{2026}{\dots}
\DeclareUnicodeCharacter{20AC}{\euro}
\DeclareUnicodeCharacter{2192}{\expansion}
\DeclareUnicodeCharacter{21D2}{\result}
\DeclareUnicodeCharacter{2212}{\minus}
\DeclareUnicodeCharacter{2217}{\point}
\DeclareUnicodeCharacter{2261}{\equiv}
}% end of \utfeightchardefs
% US-ASCII character definitions.
\def\asciichardefs{% nothing need be done
\relax
}
% Make non-ASCII characters printable again for compatibility with
% existing Texinfo documents that may use them, even without declaring a
% document encoding.
%
\setnonasciicharscatcode \other
\message{formatting,}
\newdimen\defaultparindent \defaultparindent = 15pt
\chapheadingskip = 15pt plus 4pt minus 2pt
\secheadingskip = 12pt plus 3pt minus 2pt
\subsecheadingskip = 9pt plus 2pt minus 2pt
% Prevent underfull vbox error messages.
\vbadness = 10000
% Don't be so finicky about underfull hboxes, either.
\hbadness = 2000
% Following George Bush, just get rid of widows and orphans.
\widowpenalty=10000
\clubpenalty=10000
% Use TeX 3.0's \emergencystretch to help line breaking, but if we're
% using an old version of TeX, don't do anything. We want the amount of
% stretch added to depend on the line length, hence the dependence on
% \hsize. We call this whenever the paper size is set.
%
\def\setemergencystretch{%
\ifx\emergencystretch\thisisundefined
% Allow us to assign to \emergencystretch anyway.
\def\emergencystretch{\dimen0}%
\else
\emergencystretch = .15\hsize
\fi
}
% Parameters in order: 1) textheight; 2) textwidth;
% 3) voffset; 4) hoffset; 5) binding offset; 6) topskip;
% 7) physical page height; 8) physical page width.
%
% We also call \setleading{\textleading}, so the caller should define
% \textleading. The caller should also set \parskip.
%
\def\internalpagesizes#1#2#3#4#5#6#7#8{%
\voffset = #3\relax
\topskip = #6\relax
\splittopskip = \topskip
%
\vsize = #1\relax
\advance\vsize by \topskip
\outervsize = \vsize
\advance\outervsize by 2\topandbottommargin
\pageheight = \vsize
%
\hsize = #2\relax
\outerhsize = \hsize
\advance\outerhsize by 0.5in
\pagewidth = \hsize
%
\normaloffset = #4\relax
\bindingoffset = #5\relax
%
\ifpdf
\pdfpageheight #7\relax
\pdfpagewidth #8\relax
\fi
%
\setleading{\textleading}
%
\parindent = \defaultparindent
\setemergencystretch
}
% @letterpaper (the default).
\def\letterpaper{{\globaldefs = 1
\parskip = 3pt plus 2pt minus 1pt
\textleading = 13.2pt
%
% If page is nothing but text, make it come out even.
\internalpagesizes{46\baselineskip}{6in}%
{\voffset}{.25in}%
{\bindingoffset}{36pt}%
{11in}{8.5in}%
}}
% Use @smallbook to reset parameters for 7x9.25 trim size.
\def\smallbook{{\globaldefs = 1
\parskip = 2pt plus 1pt
\textleading = 12pt
%
\internalpagesizes{7.5in}{5in}%
{\voffset}{.25in}%
{\bindingoffset}{16pt}%
{9.25in}{7in}%
%
\lispnarrowing = 0.3in
\tolerance = 700
\hfuzz = 1pt
\contentsrightmargin = 0pt
\defbodyindent = .5cm
}}
% Use @smallerbook to reset parameters for 6x9 trim size.
% (Just testing, parameters still in flux.)
\def\smallerbook{{\globaldefs = 1
\parskip = 1.5pt plus 1pt
\textleading = 12pt
%
\internalpagesizes{7.4in}{4.8in}%
{-.2in}{-.4in}%
{0pt}{14pt}%
{9in}{6in}%
%
\lispnarrowing = 0.25in
\tolerance = 700
\hfuzz = 1pt
\contentsrightmargin = 0pt
\defbodyindent = .4cm
}}
% Use @afourpaper to print on European A4 paper.
\def\afourpaper{{\globaldefs = 1
\parskip = 3pt plus 2pt minus 1pt
\textleading = 13.2pt
%
% Double-side printing via postscript on Laserjet 4050
% prints double-sided nicely when \bindingoffset=10mm and \hoffset=-6mm.
% To change the settings for a different printer or situation, adjust
% \normaloffset until the front-side and back-side texts align. Then
% do the same for \bindingoffset. You can set these for testing in
% your texinfo source file like this:
% @tex
% \global\normaloffset = -6mm
% \global\bindingoffset = 10mm
% @end tex
\internalpagesizes{51\baselineskip}{160mm}
{\voffset}{\hoffset}%
{\bindingoffset}{44pt}%
{297mm}{210mm}%
%
\tolerance = 700
\hfuzz = 1pt
\contentsrightmargin = 0pt
\defbodyindent = 5mm
}}
% Use @afivepaper to print on European A5 paper.
% From romildo@urano.iceb.ufop.br, 2 July 2000.
% He also recommends making @example and @lisp be small.
\def\afivepaper{{\globaldefs = 1
\parskip = 2pt plus 1pt minus 0.1pt
\textleading = 12.5pt
%
\internalpagesizes{160mm}{120mm}%
{\voffset}{\hoffset}%
{\bindingoffset}{8pt}%
{210mm}{148mm}%
%
\lispnarrowing = 0.2in
\tolerance = 800
\hfuzz = 1.2pt
\contentsrightmargin = 0pt
\defbodyindent = 2mm
\tableindent = 12mm
}}
% A specific text layout, 24x15cm overall, intended for A4 paper.
\def\afourlatex{{\globaldefs = 1
\afourpaper
\internalpagesizes{237mm}{150mm}%
{\voffset}{4.6mm}%
{\bindingoffset}{7mm}%
{297mm}{210mm}%
%
% Must explicitly reset to 0 because we call \afourpaper.
\globaldefs = 0
}}
% Use @afourwide to print on A4 paper in landscape format.
\def\afourwide{{\globaldefs = 1
\afourpaper
\internalpagesizes{241mm}{165mm}%
{\voffset}{-2.95mm}%
{\bindingoffset}{7mm}%
{297mm}{210mm}%
\globaldefs = 0
}}
% @pagesizes TEXTHEIGHT[,TEXTWIDTH]
% Perhaps we should allow setting the margins, \topskip, \parskip,
% and/or leading, also. Or perhaps we should compute them somehow.
%
\parseargdef\pagesizes{\pagesizesyyy #1,,\finish}
\def\pagesizesyyy#1,#2,#3\finish{{%
\setbox0 = \hbox{\ignorespaces #2}\ifdim\wd0 > 0pt \hsize=#2\relax \fi
\globaldefs = 1
%
\parskip = 3pt plus 2pt minus 1pt
\setleading{\textleading}%
%
\dimen0 = #1
\advance\dimen0 by \voffset
%
\dimen2 = \hsize
\advance\dimen2 by \normaloffset
%
\internalpagesizes{#1}{\hsize}%
{\voffset}{\normaloffset}%
{\bindingoffset}{44pt}%
{\dimen0}{\dimen2}%
}}
% Set default to letter.
%
\letterpaper
\message{and turning on texinfo input format.}
% Define macros to output various characters with catcode for normal text.
\catcode`\"=\other
\catcode`\~=\other
\catcode`\^=\other
\catcode`\_=\other
\catcode`\|=\other
\catcode`\<=\other
\catcode`\>=\other
\catcode`\+=\other
\catcode`\$=\other
\def\normaldoublequote{"}
\def\normaltilde{~}
\def\normalcaret{^}
\def\normalunderscore{_}
\def\normalverticalbar{|}
\def\normalless{<}
\def\normalgreater{>}
\def\normalplus{+}
\def\normaldollar{$}%$ font-lock fix
% This macro is used to make a character print one way in \tt
% (where it can probably be output as-is), and another way in other fonts,
% where something hairier probably needs to be done.
%
% #1 is what to print if we are indeed using \tt; #2 is what to print
% otherwise. Since all the Computer Modern typewriter fonts have zero
% interword stretch (and shrink), and it is reasonable to expect all
% typewriter fonts to have this, we can check that font parameter.
%
\def\ifusingtt#1#2{\ifdim \fontdimen3\font=0pt #1\else #2\fi}
% Same as above, but check for italic font. Actually this also catches
% non-italic slanted fonts since it is impossible to distinguish them from
% italic fonts. But since this is only used by $ and it uses \sl anyway
% this is not a problem.
\def\ifusingit#1#2{\ifdim \fontdimen1\font>0pt #1\else #2\fi}
% Turn off all special characters except @
% (and those which the user can use as if they were ordinary).
% Most of these we simply print from the \tt font, but for some, we can
% use math or other variants that look better in normal text.
\catcode`\"=\active
\def\activedoublequote{{\tt\char34}}
\let"=\activedoublequote
\catcode`\~=\active
\def~{{\tt\char126}}
\chardef\hat=`\^
\catcode`\^=\active
\def^{{\tt \hat}}
\catcode`\_=\active
\def_{\ifusingtt\normalunderscore\_}
\let\realunder=_
% Subroutine for the previous macro.
\def\_{\leavevmode \kern.07em \vbox{\hrule width.3em height.1ex}\kern .07em }
\catcode`\|=\active
\def|{{\tt\char124}}
\chardef \less=`\<
\catcode`\<=\active
\def<{{\tt \less}}
\chardef \gtr=`\>
\catcode`\>=\active
\def>{{\tt \gtr}}
\catcode`\+=\active
\def+{{\tt \char 43}}
\catcode`\$=\active
\def${\ifusingit{{\sl\$}}\normaldollar}%$ font-lock fix
% If a .fmt file is being used, characters that might appear in a file
% name cannot be active until we have parsed the command line.
% So turn them off again, and have \everyjob (or @setfilename) turn them on.
% \otherifyactive is called near the end of this file.
\def\otherifyactive{\catcode`+=\other \catcode`\_=\other}
% Used sometimes to turn off (effectively) the active characters even after
% parsing them.
\def\turnoffactive{%
\normalturnoffactive
\otherbackslash
}
\catcode`\@=0
% \backslashcurfont outputs one backslash character in current font,
% as in \char`\\.
\global\chardef\backslashcurfont=`\\
\global\let\rawbackslashxx=\backslashcurfont % let existing .??s files work
% \realbackslash is an actual character `\' with catcode other, and
% \doublebackslash is two of them (for the pdf outlines).
{\catcode`\\=\other @gdef@realbackslash{\} @gdef@doublebackslash{\\}}
% In texinfo, backslash is an active character; it prints the backslash
% in fixed width font.
\catcode`\\=\active
@def@normalbackslash{{@tt@backslashcurfont}}
% On startup, @fixbackslash assigns:
% @let \ = @normalbackslash
% \rawbackslash defines an active \ to do \backslashcurfont.
% \otherbackslash defines an active \ to be a literal `\' character with
% catcode other.
@gdef@rawbackslash{@let\=@backslashcurfont}
@gdef@otherbackslash{@let\=@realbackslash}
% Same as @turnoffactive except outputs \ as {\tt\char`\\} instead of
% the literal character `\'.
%
@def@normalturnoffactive{%
@let\=@normalbackslash
@let"=@normaldoublequote
@let~=@normaltilde
@let^=@normalcaret
@let_=@normalunderscore
@let|=@normalverticalbar
@let<=@normalless
@let>=@normalgreater
@let+=@normalplus
@let$=@normaldollar %$ font-lock fix
@unsepspaces
}
% Make _ and + \other characters, temporarily.
% This is canceled by @fixbackslash.
@otherifyactive
% If a .fmt file is being used, we don't want the `\input texinfo' to show up.
% That is what \eatinput is for; after that, the `\' should revert to printing
% a backslash.
%
@gdef@eatinput input texinfo{@fixbackslash}
@global@let\ = @eatinput
% On the other hand, perhaps the file did not have a `\input texinfo'. Then
% the first `\' in the file would cause an error. This macro tries to fix
% that, assuming it is called before the first `\' could plausibly occur.
% Also turn back on active characters that might appear in the input
% file name, in case not using a pre-dumped format.
%
@gdef@fixbackslash{%
@ifx\@eatinput @let\ = @normalbackslash @fi
@catcode`+=@active
@catcode`@_=@active
}
% Say @foo, not \foo, in error messages.
@escapechar = `@@
% These look ok in all fonts, so just make them not special.
@catcode`@& = @other
@catcode`@# = @other
@catcode`@% = @other
@c Local variables:
@c eval: (add-hook 'write-file-hooks 'time-stamp)
@c page-delimiter: "^\\\\message"
@c time-stamp-start: "def\\\\texinfoversion{"
@c time-stamp-format: "%:y-%02m-%02d.%02H"
@c time-stamp-end: "}"
@c End:
@c vim:sw=2:
@ignore
arch-tag: e1b36e32-c96e-4135-a41a-0b2efa2ea115
@end ignore
diff --git a/configure.ac b/configure.ac
index aec3685..19ae9c8 100644
--- a/configure.ac
+++ b/configure.ac
@@ -1,554 +1,554 @@
# configure.ac for libgpg-error
# Copyright (C) 2003, 2004, 2006, 2010, 2013, 2014, 2015 g10 Code GmbH
#
# This file is part of libgpg-error.
#
# libgpg-error is free software; you can redistribute it and/or modify
# it under the terms of the GNU Lesser General Public License as
# published by the Free Software Foundation; either version 2.1 of the
# License, or (at your option) any later version.
#
# libgpg-error is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU Lesser General Public License for more details.
#
# You should have received a copy of the GNU General Public License
-# along with this program; if not, see .
+# along with this program; if not, see .
# (Process this file with autoconf to produce a configure script.)
# The following lines are used by ./autogen.sh.
AC_PREREQ(2.59)
min_automake_version="1.14"
# To build a release you need to create a tag with the version number
# (git tag -s libgpg-error-n.m) and run "./autogen.sh --force".
# Please bump the version number immediately after the release, do
# another commit, and a push so that the git magic is able to work.
# See below for the LT versions.
m4_define([mym4_version_major], [1])
m4_define([mym4_version_minor], [22])
# Below is m4 magic to extract and compute the revision number, the
# decimalized short revision number, a beta version string, and a flag
# indicating a development version (mym4_isgit). Note that the m4
# processing is done by autoconf and not during the configure run.
m4_define([mym4_version], [mym4_version_major.mym4_version_minor])
m4_define([mym4_revision],
m4_esyscmd([git rev-parse --short HEAD | tr -d '\n\r']))
m4_define([mym4_revision_dec],
m4_esyscmd_s([echo $((0x$(echo ]mym4_revision[|head -c 4)))]))
m4_define([mym4_betastring],
m4_esyscmd_s([git describe --match 'libgpg-error-[0-9].*[0-9]' --long|\
awk -F- '$4!=0{print"-beta"$4}']))
m4_define([mym4_isgit],m4_if(mym4_betastring,[],[no],[yes]))
m4_define([mym4_full_version],[mym4_version[]mym4_betastring])
-AC_INIT([libgpg-error],[mym4_full_version],[http://bugs.gnupg.org])
+AC_INIT([libgpg-error],[mym4_full_version],[https://bugs.gnupg.org])
# LT Version numbers, remember to change them just *before* a release.
# (Code changed: REVISION++)
# (Interfaces added/removed/changed: CURRENT++, REVISION=0)
# (Interfaces added: AGE++)
# (Interfaces removed: AGE=0)
# Note that added error codes don't constitute an interface change.
LIBGPG_ERROR_LT_CURRENT=17
LIBGPG_ERROR_LT_AGE=17
LIBGPG_ERROR_LT_REVISION=0
################################################
AC_SUBST(LIBGPG_ERROR_LT_CURRENT)
AC_SUBST(LIBGPG_ERROR_LT_AGE)
AC_SUBST(LIBGPG_ERROR_LT_REVISION)
VERSION_NUMBER=m4_esyscmd(printf "0x%02x%02x00" mym4_version_major \
mym4_version_minor)
AC_SUBST(VERSION_NUMBER)
AC_CONFIG_AUX_DIR([build-aux])
AM_INIT_AUTOMAKE([serial-tests dist-bzip2])
AM_MAINTAINER_MODE
AC_CONFIG_SRCDIR([src/err-sources.h.in])
AC_CONFIG_HEADER([config.h])
AC_CONFIG_MACRO_DIR([m4])
AM_SILENT_RULES
# We need to know about the host architecture to avoid spurious
# warnings.
AC_CANONICAL_HOST
AB_INIT
# Checks for programs.
AC_PROG_CC
AM_PROG_CC_C_O
AC_PROG_CPP
AC_PROG_AWK
AC_CHECK_TOOL(AR, ar, :)
AC_GNU_SOURCE
# Set some internal variables depending on the platform for later use.
have_w32_system=no
have_w64_system=no
have_w32ce_system=no
case "${host}" in
x86_64-*mingw32*)
have_w32_system=yes
have_w64_system=yes
;;
*-mingw32ce*)
have_w32_system=yes
have_w32ce_system=yes
;;
*-mingw32*)
have_w32_system=yes
;;
*)
;;
esac
if test "$have_w32_system" != yes; then
gl_THREADLIB_EARLY
fi
# We build libgpg-error with large file support so that we have a 64
# bit off_t. Our external interface uses the gpgrt_off_t which is
# anyway specified as 64 bit. Thus the same libgpg-error can be used
# by software which is not build with large file support.
AC_SYS_LARGEFILE
LT_PREREQ([2.2.6])
LT_INIT([win32-dll disable-static])
LT_LANG([Windows Resource])
# We need to compile and run a program on the build machine.
dnl The AC_PROG_CC_FOR_BUILD macro in the AC archive is broken for
dnl autoconf 2.57.
dnl AC_PROG_CC_FOR_BUILD
AC_MSG_CHECKING(for cc for build)
if test "$cross_compiling" = "yes"; then
CC_FOR_BUILD="${CC_FOR_BUILD-cc}"
else
CC_FOR_BUILD="${CC_FOR_BUILD-$CC}"
fi
AC_MSG_RESULT($CC_FOR_BUILD)
AC_ARG_VAR(CC_FOR_BUILD,[build system C compiler])
AH_BOTTOM([
/* Force using of NLS for W32 even if no libintl has been found. This is
okay because we have our own gettext implementation for W32. */
#if defined(HAVE_W32_SYSTEM) && !defined(ENABLE_NLS)
#define ENABLE_NLS 1
#endif
/* Connect the generic estream-printf.c to our framework. */
#define _ESTREAM_PRINTF_REALLOC _gpgrt_realloc
#define _ESTREAM_PRINTF_EXTRA_INCLUDE "gpgrt-int.h"
/* For building we need to define these macro. */
#define GPG_ERR_ENABLE_GETTEXT_MACROS 1
#define GPG_ERR_ENABLE_ERRNO_MACROS 1
#define GPGRT_ENABLE_ES_MACROS 1
])
# Note, that autogen.sh greps for the next line.
AM_GNU_GETTEXT_VERSION([0.19.3])
AM_GNU_GETTEXT([external])
# Checks for header files.
AC_HEADER_STDC
AC_CHECK_HEADERS([stdlib.h locale.h stdint.h sys/select.h sys/time.h])
AC_FUNC_STRERROR_R
case "${host_os}" in
solaris*)
# All versions of Solaris from 2.4 have a thread-safe strerror().
# Since Solaris 10, in addition strerror_r() exists.
;;
*)
AC_CHECK_FUNC([strerror_r], [],
AC_MSG_WARN([[Without strerror_r, gpg_strerror_r might not be thread-safe]]))
;;
esac
AC_CHECK_FUNCS([flockfile vasprintf])
#
# Checks for typedefs, structures, and compiler characteristics.
#
AC_C_CONST
AC_CHECK_SIZEOF(int)
AC_CHECK_SIZEOF(long)
AC_CHECK_SIZEOF(long long)
# Find a 64 bit integer type to be used instead of off_t. We prefer
# the standard integer types over int64_t and finally try long long.
if test "$ac_cv_sizeof_int" = "8"; then
replacement_for_off_t="int"
elif test "$ac_cv_sizeof_long" = "8"; then
replacement_for_off_t="long"
elif test "$ac_cv_header_stdint_h" = yes; then
replacement_for_off_t="int64_t"
elif test "$ac_cv_sizeof_long_long" = "8"; then
replacement_for_off_t="long long"
else
AC_MSG_ERROR([[
***
*** No 64 bit signed integer type found. Can't build this library.
***]])
fi
AC_DEFINE_UNQUOTED(REPLACEMENT_FOR_OFF_T, "$replacement_for_off_t",
[Used by mkheader to insert the replacement type.])
#
# Setup gcc specific options
#
AC_MSG_NOTICE([checking for cc features])
if test "$GCC" = yes; then
# Check whether gcc does not emit a diagnostic for unknown -Wno-*
# options. This is the case for gcc >= 4.6
AC_MSG_CHECKING([if gcc ignores unknown -Wno-* options])
AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[
#if __GNUC__ < 4 || (__GNUC__ == 4 && __GNUC_MINOR__ < 6 )
#kickerror
#endif]],[])],[_gcc_silent_wno=yes],[_gcc_silent_wno=no])
AC_MSG_RESULT($_gcc_silent_wno)
# Note that it is okay to use CFLAGS here because these are just
# warning options and the user should have a chance of overriding
# them.
if test "$USE_MAINTAINER_MODE" = "yes"; then
CFLAGS="$CFLAGS -O3 -Wall -Wcast-align -Wshadow -Wstrict-prototypes"
CFLAGS="$CFLAGS -Wformat -Wno-format-y2k -Wformat-security"
if test x"$_gcc_silent_wno" = xyes ; then
_gcc_wopt=yes
else
AC_MSG_CHECKING([if gcc supports -Wno-missing-field-initializers])
_gcc_cflags_save=$CFLAGS
CFLAGS="-Wno-missing-field-initializers"
AC_COMPILE_IFELSE([AC_LANG_PROGRAM([],[])],
[_gcc_wopt=yes],[_gcc_wopt=no])
AC_MSG_RESULT($_gcc_wopt)
CFLAGS=$_gcc_cflags_save;
fi
if test x"$_gcc_wopt" = xyes ; then
CFLAGS="$CFLAGS -W -Wno-sign-compare -Wno-missing-field-initializers"
fi
AC_MSG_CHECKING([if gcc supports -Wdeclaration-after-statement])
_gcc_cflags_save=$CFLAGS
CFLAGS="-Wdeclaration-after-statement"
AC_COMPILE_IFELSE([AC_LANG_PROGRAM([],[])],_gcc_wopt=yes,_gcc_wopt=no)
AC_MSG_RESULT($_gcc_wopt)
CFLAGS=$_gcc_cflags_save;
if test x"$_gcc_wopt" = xyes ; then
CFLAGS="$CFLAGS -Wdeclaration-after-statement"
fi
else
CFLAGS="$CFLAGS -Wall"
fi
AC_MSG_CHECKING([if gcc supports -Wpointer-arith])
_gcc_cflags_save=$CFLAGS
CFLAGS="-Wpointer-arith"
AC_COMPILE_IFELSE([AC_LANG_PROGRAM([],[])],_gcc_psign=yes,_gcc_psign=no)
AC_MSG_RESULT($_gcc_psign)
CFLAGS=$_gcc_cflags_save;
if test x"$_gcc_psign" = xyes ; then
CFLAGS="$CFLAGS -Wpointer-arith"
fi
# The undocumented option -Wno-psabi suppresses the annoying
# "the ABI of passing union with long double has changed in GCC 4.4"
# which is emitted in estream-printf.c but entirely irrelvant
# because that union is local to the file.
if test x"$_gcc_silent_wno" = xyes ; then
CFLAGS="$CFLAGS -Wno-psabi"
fi
fi
#
# Check whether the compiler supports the GCC style aligned attribute
#
AC_CACHE_CHECK([whether the GCC style aligned attribute is supported],
[gcry_cv_gcc_attribute_aligned],
[gcry_cv_gcc_attribute_aligned=no
AC_COMPILE_IFELSE([AC_LANG_SOURCE(
[[struct { int a; } foo __attribute__ ((aligned (16)));]])],
[gcry_cv_gcc_attribute_aligned=yes])])
if test "$gcry_cv_gcc_attribute_aligned" = "yes" ; then
AC_DEFINE(HAVE_GCC_ATTRIBUTE_ALIGNED,1,
[Defined if a GCC style "__attribute__ ((aligned (n))" is supported])
fi
#
# Check for ELF visibility support.
#
AC_CACHE_CHECK(whether the visibility attribute is supported,
gcry_cv_visibility_attribute,
[gcry_cv_visibility_attribute=no
AC_LANG_CONFTEST([AC_LANG_SOURCE(
[[int foo __attribute__ ((visibility ("hidden"))) = 1;
int bar __attribute__ ((visibility ("protected"))) = 1;
]])])
if ${CC-cc} -Werror -S conftest.c -o conftest.s \
1>&AS_MESSAGE_LOG_FD 2>&AS_MESSAGE_LOG_FD ; then
if grep '\.hidden.*foo' conftest.s >/dev/null 2>&1 ; then
if grep '\.protected.*bar' conftest.s >/dev/null 2>&1; then
gcry_cv_visibility_attribute=yes
fi
fi
fi
])
if test "$gcry_cv_visibility_attribute" = "yes"; then
AC_CACHE_CHECK(for broken visibility attribute,
gcry_cv_broken_visibility_attribute,
[gcry_cv_broken_visibility_attribute=yes
AC_LANG_CONFTEST([AC_LANG_SOURCE(
[[int foo (int x);
int bar (int x) __asm__ ("foo")
__attribute__ ((visibility ("hidden")));
int bar (int x) { return x; }
]])])
if ${CC-cc} -Werror -S conftest.c -o conftest.s \
1>&AS_MESSAGE_LOG_FD 2>&AS_MESSAGE_LOG_FD ; then
if grep '\.hidden@<:@ _@:>@foo' conftest.s >/dev/null 2>&1;
then
gcry_cv_broken_visibility_attribute=no
fi
fi
])
fi
if test "$gcry_cv_visibility_attribute" = "yes"; then
AC_CACHE_CHECK(for broken alias attribute,
gcry_cv_broken_alias_attribute,
[gcry_cv_broken_alias_attribute=yes
AC_LANG_CONFTEST([AC_LANG_SOURCE(
[[extern int foo (int x) __asm ("xyzzy");
int bar (int x) { return x; }
extern __typeof (bar) foo __attribute ((weak, alias ("bar")));
extern int dfoo;
extern __typeof (dfoo) dfoo __asm ("abccb");
int dfoo = 1;
]])])
if ${CC-cc} -Werror -S conftest.c -o conftest.s \
1>&AS_MESSAGE_LOG_FD 2>&AS_MESSAGE_LOG_FD ; then
if grep 'xyzzy' conftest.s >/dev/null 2>&1 && \
grep 'abccb' conftest.s >/dev/null 2>&1; then
gcry_cv_broken_alias_attribute=no
fi
fi
])
fi
if test "$gcry_cv_visibility_attribute" = "yes"; then
AC_CACHE_CHECK(if gcc supports -fvisibility=hidden,
gcry_cv_gcc_has_f_visibility,
[gcry_cv_gcc_has_f_visibility=no
_gcc_cflags_save=$CFLAGS
CFLAGS="-fvisibility=hidden"
AC_COMPILE_IFELSE([AC_LANG_PROGRAM([],[])],
gcry_cv_gcc_has_f_visibility=yes)
CFLAGS=$_gcc_cflags_save;
])
fi
if test "$gcry_cv_visibility_attribute" = "yes" \
&& test "$gcry_cv_broken_visibility_attribute" != "yes" \
&& test "$gcry_cv_broken_alias_attribute" != "yes" \
&& test "$gcry_cv_gcc_has_f_visibility" = "yes"
then
AC_DEFINE(GPGRT_USE_VISIBILITY, 1,
[Define to use the GNU C visibility attribute.])
CFLAGS="$CFLAGS -fvisibility=hidden"
fi
#
# Check whether ld supports a version script.
# (Actually not a check but a list of systems which are known to support it.)
#
have_ld_version_script=no
case "${host}" in
*-*-linux*)
have_ld_version_script=yes
;;
*-*-gnu*)
have_ld_version_script=yes
;;
esac
AC_ARG_ENABLE([ld-version-script],
AC_HELP_STRING([--enable-ld-version-script],
[enable/disable use of linker version script.
(default is system dependent)]),
[have_ld_version_script=$enableval],
[ : ] )
AM_CONDITIONAL(HAVE_LD_VERSION_SCRIPT, test "$have_ld_version_script" = "yes")
#
# Check for thread library.
#
# Windows has always thread support; thus we don't bother to test for
# it as it may lead to false results when cross building.
if test "$have_w32_system" = yes; then
AC_DEFINE([USE_WINDOWS_THREADS], [1])
LIBTHREAD=
LTLIBTHREAD=
LIBMULTITHREAD=
LTLIBMULTITHREAD=
THREADLIB_CPPFLAGS=""
AC_SUBST([LIBTHREAD])
AC_SUBST([LTLIBTHREAD])
AC_SUBST([LIBMULTITHREAD])
AC_SUBST([LTLIBMULTITHREAD])
else
gl_LOCK
if test "$gl_threads_api" = posix; then
AC_CHECK_SIZEOF(pthread_mutex_t,,[AC_INCLUDES_DEFAULT
#include ])
fi
fi
#
# Prepare building of estream
#
estream_INIT
#
# Substitution used for gpg-error-config
#
GPG_ERROR_CONFIG_LIBS="-lgpg-error"
if test "x$LIBTHREAD" != x; then
GPG_ERROR_CONFIG_LIBS="${GPG_ERROR_CONFIG_LIBS} ${LIBTHREAD}"
fi
if test "x$LIBMULTITHREAD" != x; then
GPG_ERROR_CONFIG_MT_LIBS="${LIBMULTITHREAD}"
else
GPG_ERROR_CONFIG_MT_LIBS=""
fi
GPG_ERROR_CONFIG_CFLAGS=""
if test "x$THREADLIB_CPPFLAGS" != x; then
GPG_ERROR_CONFIG_MT_CFLAGS="${THREADLIB_CPPFLAGS}"
else
GPG_ERROR_CONFIG_MT_CFLAGS=""
fi
GPG_ERROR_CONFIG_ISUBDIRAFTER=""
GPG_ERROR_CONFIG_HOST="$host"
AC_SUBST(GPG_ERROR_CONFIG_LIBS)
AC_SUBST(GPG_ERROR_CONFIG_CFLAGS)
AC_SUBST(GPG_ERROR_CONFIG_MT_LIBS)
AC_SUBST(GPG_ERROR_CONFIG_MT_CFLAGS)
AC_SUBST(GPG_ERROR_CONFIG_ISUBDIRAFTER)
AC_SUBST(GPG_ERROR_CONFIG_HOST)
#
# Special defines for certain platforms
#
force_use_syscfg=no
if test "$have_w32_system" = yes; then
AC_DEFINE(HAVE_W32_SYSTEM,1,[Defined if we run on a W32 API based system])
if test "$have_w64_system" = yes; then
AC_DEFINE(HAVE_W64_SYSTEM,1,[Defined if we run on 64 bit W32 API system])
fi
if test "$have_w32ce_system" = yes; then
AC_DEFINE(HAVE_W32CE_SYSTEM,1,[Defined if we run on WindowsCE])
GPG_ERROR_CONFIG_ISUBDIRAFTER="gpg-extra"
fi
force_use_syscfg=yes
fi
if test x$cross_compiling = xyes; then
force_use_syscfg=yes
fi
AM_CONDITIONAL(HAVE_W32_SYSTEM, test "$have_w32_system" = yes)
AM_CONDITIONAL(HAVE_W64_SYSTEM, test "$have_w64_system" = yes)
AM_CONDITIONAL(HAVE_W32CE_SYSTEM, test "$have_w32ce_system" = yes)
AM_CONDITIONAL(CROSS_COMPILING, test x$cross_compiling = xyes)
AM_CONDITIONAL(FORCE_USE_SYSCFG, test x$force_use_syscfg = xyes)
AC_DEFINE_UNQUOTED(HOST_TRIPLET_STRING, "$host", [The host triplet])
#
# Provide information about the build.
#
BUILD_REVISION="mym4_revision"
AC_SUBST(BUILD_REVISION)
AC_DEFINE_UNQUOTED(BUILD_REVISION, "$BUILD_REVISION",
[GIT commit id revision used to build this package])
changequote(,)dnl
BUILD_VERSION=`echo "$PACKAGE_VERSION"|sed 's/\([0-9.]*\).*/\1./'`
changequote([,])dnl
BUILD_VERSION="${BUILD_VERSION}0.mym4_revision_dec"
BUILD_FILEVERSION=`echo "${BUILD_VERSION}" | tr . ,`
AC_SUBST(BUILD_VERSION)
AC_SUBST(BUILD_FILEVERSION)
AC_ARG_ENABLE([build-timestamp],
AC_HELP_STRING([--enable-build-timestamp],
[set an explicit build timestamp for reproducibility.
(default is the current time in ISO-8601 format)]),
[if test "$enableval" = "yes"; then
BUILD_TIMESTAMP=`date -u +%Y-%m-%dT%H:%M+0000 2>/dev/null || date`
else
BUILD_TIMESTAMP="$enableval"
fi],
[BUILD_TIMESTAMP=""])
AC_SUBST(BUILD_TIMESTAMP)
AC_DEFINE_UNQUOTED(BUILD_TIMESTAMP, "$BUILD_TIMESTAMP",
[The time this package was configured for a build])
AC_ARG_ENABLE(languages, AC_HELP_STRING([--disable-languages],
[do not build support for other languages than C]))
AM_CONDITIONAL([LANGUAGES_SOME], [test "x$enable_languages" != xno])
build_doc=yes
AC_ARG_ENABLE([doc], AC_HELP_STRING([--disable-doc],
[do not build the documentation]),
build_doc=$enableval, build_doc=yes)
AM_CONDITIONAL([BUILD_DOC], [test "x$build_doc" != xno])
#
# Substitution
#
AC_CONFIG_FILES([Makefile])
AC_CONFIG_FILES([doc/Makefile po/Makefile.in m4/Makefile])
AC_CONFIG_FILES([src/Makefile tests/Makefile])
AC_CONFIG_FILES([lang/Makefile lang/cl/Makefile lang/cl/gpg-error.asd])
AC_CONFIG_FILES([src/versioninfo.rc src/gpg-error.w32-manifest])
AC_CONFIG_FILES([src/gpg-error-config], [chmod +x src/gpg-error-config])
AC_OUTPUT
tmp=
if test "$have_w32_system" != yes; then
if test x"$gl_use_threads" = xno; then
tmp=" NO-THREADS"
fi
fi
echo "
$PACKAGE_NAME-$PACKAGE_VERSION prepared for make
Revision: mym4_revision (mym4_revision_dec)
Platform: $host$tmp
"
if test "$gcry_cv_gcc_attribute_aligned" != "yes" ; then
cat <.
+# License along with this program; if not, see .
EXTRA_DIST = HACKING errorref.txt \
yat2m.c
DISTCLEANFILES = gpgrt.cps yat2m-stamp.tmp yat2m-stamp $(myman_pages)
CLEANFILES = yat2m
info_TEXINFOS = gpgrt.texi
gpgrt_TEXINFOS = lgpl.texi gpl.texi
YAT2M_OPTIONS = -I $(srcdir) \
--release "Libgpg-error @PACKAGE_VERSION@" --source "GnuPG"
myman_sources = gpgrt.texi
myman_pages = gpg-error-config.1
man_MANS = $(myman_pages)
yat2m: yat2m.c
$(CC_FOR_BUILD) -o $@ $(srcdir)/yat2m.c
yat2m-stamp: $(myman_sources)
@rm -f yat2m-stamp.tmp
@touch yat2m-stamp.tmp
for file in $(myman_sources) ; do \
./yat2m $(YAT2M_OPTIONS) --store \
`test -f '$$file' || echo '$(srcdir)/'`$$file ; done
@mv -f yat2m-stamp.tmp $@
yat2m-stamp: yat2m
$(myman_pages) : yat2m-stamp
@if test -f $@; then :; else \
trap 'rm -rf yat2m-stamp yat2m-lock' 1 2 13 15; \
if mkdir yat2m-lock 2>/dev/null; then \
rm -f yat2m-stamp; \
$(MAKE) $(AM_MAKEFLAGS) yat2m-stamp; \
rmdir yat2m-lock; \
else \
while test -d yat2m-lock; do sleep 1; done; \
test -f yat2m-stamp; exit $$?; \
fi; \
fi
# Make sure that gpgrt.texi is touched if any other texi file has
# been modified. This is required so that the version.texi magic
# updates the release date.
gpgrt.texi : $(gpgrt_TEXINFOS)
touch $(srcdir)/gcrypt.texi
diff --git a/doc/ldap2gpgerr.c b/doc/ldap2gpgerr.c
index 515bf40..f6fd3cf 100644
--- a/doc/ldap2gpgerr.c
+++ b/doc/ldap2gpgerr.c
@@ -1,184 +1,184 @@
/* ldap2gpgerr.c - Mapping of LDAP error codes to gpg-error codes.
* Written in 2015 by Werner Koch
*
* To the extent possible under law, the author(s) have dedicated all
* copyright and related and neighboring rights to this software to
* the public domain worldwide. This software is distributed without
* any warranty.
*
* You should have received a copy of the CC0 Public Domain Dedication
* along with this software. If not, see
- * .
+ * .
*/
/*
* These functions are not part of libgpg-error so not to introduce a
* dependency on a specific LDAP implementation. Feel free to copy
* and distribute them with your code.
*/
#ifdef _WIN32
# include
# include
#else
# include
#endif
#include
/* Windows uses a few other names. Re-map them. */
#ifdef _WIN32
# define LDAP_ADMINLIMIT_EXCEEDED LDAP_ADMIN_LIMIT_EXCEEDED
# define LDAP_UNAVAILABLE_CRITICAL_EXTENSION LDAP_UNAVAILABLE_CRIT_EXTENSION
# define LDAP_TYPE_OR_VALUE_EXISTS LDAP_ATTRIBUTE_OR_VALUE_EXISTS
# define LDAP_INSUFFICIENT_ACCESS LDAP_INSUFFICIENT_RIGHTS
# define LDAP_VLV_ERROR LDAP_VIRTUAL_LIST_VIEW_ERROR
#endif
/* Map LDAP error CODE to an gpg_err_code_t. */
gpg_err_code_t
map_ldap_to_gpg_error (int code)
{
gpg_err_code_t ec;
switch (code)
{
#ifdef LDAP_X_CONNECTING
case LDAP_X_CONNECTING: ec = GPG_ERR_LDAP_X_CONNECTING; break;
#endif
case LDAP_REFERRAL_LIMIT_EXCEEDED: ec = GPG_ERR_LDAP_REFERRAL_LIMIT; break;
case LDAP_CLIENT_LOOP: ec = GPG_ERR_LDAP_CLIENT_LOOP; break;
case LDAP_NO_RESULTS_RETURNED: ec = GPG_ERR_LDAP_NO_RESULTS; break;
case LDAP_CONTROL_NOT_FOUND: ec = GPG_ERR_LDAP_CONTROL_NOT_FOUND; break;
case LDAP_NOT_SUPPORTED: ec = GPG_ERR_LDAP_NOT_SUPPORTED; break;
case LDAP_CONNECT_ERROR: ec = GPG_ERR_LDAP_CONNECT; break;
case LDAP_NO_MEMORY: ec = GPG_ERR_LDAP_NO_MEMORY; break;
case LDAP_PARAM_ERROR: ec = GPG_ERR_LDAP_PARAM; break;
case LDAP_USER_CANCELLED: ec = GPG_ERR_LDAP_USER_CANCELLED; break;
case LDAP_FILTER_ERROR: ec = GPG_ERR_LDAP_FILTER; break;
case LDAP_AUTH_UNKNOWN: ec = GPG_ERR_LDAP_AUTH_UNKNOWN; break;
case LDAP_TIMEOUT: ec = GPG_ERR_LDAP_TIMEOUT; break;
case LDAP_DECODING_ERROR: ec = GPG_ERR_LDAP_DECODING; break;
case LDAP_ENCODING_ERROR: ec = GPG_ERR_LDAP_ENCODING; break;
case LDAP_LOCAL_ERROR: ec = GPG_ERR_LDAP_LOCAL; break;
case LDAP_SERVER_DOWN: ec = GPG_ERR_LDAP_SERVER_DOWN; break;
case LDAP_SUCCESS: ec = GPG_ERR_LDAP_SUCCESS; break;
case LDAP_OPERATIONS_ERROR: ec = GPG_ERR_LDAP_OPERATIONS; break;
case LDAP_PROTOCOL_ERROR: ec = GPG_ERR_LDAP_PROTOCOL; break;
case LDAP_TIMELIMIT_EXCEEDED: ec = GPG_ERR_LDAP_TIMELIMIT; break;
case LDAP_SIZELIMIT_EXCEEDED: ec = GPG_ERR_LDAP_SIZELIMIT; break;
case LDAP_COMPARE_FALSE: ec = GPG_ERR_LDAP_COMPARE_FALSE; break;
case LDAP_COMPARE_TRUE: ec = GPG_ERR_LDAP_COMPARE_TRUE; break;
case LDAP_AUTH_METHOD_NOT_SUPPORTED: ec=GPG_ERR_LDAP_UNSUPPORTED_AUTH;break;
case LDAP_STRONG_AUTH_REQUIRED: ec = GPG_ERR_LDAP_STRONG_AUTH_RQRD; break;
case LDAP_PARTIAL_RESULTS: ec = GPG_ERR_LDAP_PARTIAL_RESULTS; break;
case LDAP_REFERRAL: ec = GPG_ERR_LDAP_REFERRAL; break;
#ifdef LDAP_ADMINLIMIT_EXCEEDED
case LDAP_ADMINLIMIT_EXCEEDED: ec = GPG_ERR_LDAP_ADMINLIMIT; break;
#endif
#ifdef LDAP_UNAVAILABLE_CRITICAL_EXTENSION
case LDAP_UNAVAILABLE_CRITICAL_EXTENSION:
ec = GPG_ERR_LDAP_UNAVAIL_CRIT_EXTN; break;
#endif
case LDAP_CONFIDENTIALITY_REQUIRED: ec = GPG_ERR_LDAP_CONFIDENT_RQRD; break;
case LDAP_SASL_BIND_IN_PROGRESS: ec = GPG_ERR_LDAP_SASL_BIND_INPROG; break;
case LDAP_NO_SUCH_ATTRIBUTE: ec = GPG_ERR_LDAP_NO_SUCH_ATTRIBUTE; break;
case LDAP_UNDEFINED_TYPE: ec = GPG_ERR_LDAP_UNDEFINED_TYPE; break;
case LDAP_INAPPROPRIATE_MATCHING: ec = GPG_ERR_LDAP_BAD_MATCHING; break;
case LDAP_CONSTRAINT_VIOLATION: ec = GPG_ERR_LDAP_CONST_VIOLATION; break;
#ifdef LDAP_TYPE_OR_VALUE_EXISTS
case LDAP_TYPE_OR_VALUE_EXISTS: ec = GPG_ERR_LDAP_TYPE_VALUE_EXISTS; break;
#endif
case LDAP_INVALID_SYNTAX: ec = GPG_ERR_LDAP_INV_SYNTAX; break;
case LDAP_NO_SUCH_OBJECT: ec = GPG_ERR_LDAP_NO_SUCH_OBJ; break;
case LDAP_ALIAS_PROBLEM: ec = GPG_ERR_LDAP_ALIAS_PROBLEM; break;
case LDAP_INVALID_DN_SYNTAX: ec = GPG_ERR_LDAP_INV_DN_SYNTAX; break;
case LDAP_IS_LEAF: ec = GPG_ERR_LDAP_IS_LEAF; break;
case LDAP_ALIAS_DEREF_PROBLEM: ec = GPG_ERR_LDAP_ALIAS_DEREF; break;
#ifdef LDAP_X_PROXY_AUTHZ_FAILURE
case LDAP_X_PROXY_AUTHZ_FAILURE: ec = GPG_ERR_LDAP_X_PROXY_AUTH_FAIL; break;
#endif
case LDAP_INAPPROPRIATE_AUTH: ec = GPG_ERR_LDAP_BAD_AUTH; break;
case LDAP_INVALID_CREDENTIALS: ec = GPG_ERR_LDAP_INV_CREDENTIALS; break;
#ifdef LDAP_INSUFFICIENT_ACCESS
case LDAP_INSUFFICIENT_ACCESS: ec = GPG_ERR_LDAP_INSUFFICIENT_ACC; break;
#endif
case LDAP_BUSY: ec = GPG_ERR_LDAP_BUSY; break;
case LDAP_UNAVAILABLE: ec = GPG_ERR_LDAP_UNAVAILABLE; break;
case LDAP_UNWILLING_TO_PERFORM: ec = GPG_ERR_LDAP_UNWILL_TO_PERFORM; break;
case LDAP_LOOP_DETECT: ec = GPG_ERR_LDAP_LOOP_DETECT; break;
case LDAP_NAMING_VIOLATION: ec = GPG_ERR_LDAP_NAMING_VIOLATION; break;
case LDAP_OBJECT_CLASS_VIOLATION: ec = GPG_ERR_LDAP_OBJ_CLS_VIOLATION; break;
case LDAP_NOT_ALLOWED_ON_NONLEAF: ec=GPG_ERR_LDAP_NOT_ALLOW_NONLEAF;break;
case LDAP_NOT_ALLOWED_ON_RDN: ec = GPG_ERR_LDAP_NOT_ALLOW_ON_RDN; break;
case LDAP_ALREADY_EXISTS: ec = GPG_ERR_LDAP_ALREADY_EXISTS; break;
case LDAP_NO_OBJECT_CLASS_MODS: ec = GPG_ERR_LDAP_NO_OBJ_CLASS_MODS; break;
case LDAP_RESULTS_TOO_LARGE: ec = GPG_ERR_LDAP_RESULTS_TOO_LARGE; break;
case LDAP_AFFECTS_MULTIPLE_DSAS: ec = GPG_ERR_LDAP_AFFECTS_MULT_DSAS; break;
#ifdef LDAP_VLV_ERROR
case LDAP_VLV_ERROR: ec = GPG_ERR_LDAP_VLV; break;
#endif
case LDAP_OTHER: ec = GPG_ERR_LDAP_OTHER; break;
#ifdef LDAP_CUP_RESOURCES_EXHAUSTED
case LDAP_CUP_RESOURCES_EXHAUSTED: ec=GPG_ERR_LDAP_CUP_RESOURCE_LIMIT;break;
case LDAP_CUP_SECURITY_VIOLATION: ec=GPG_ERR_LDAP_CUP_SEC_VIOLATION; break;
case LDAP_CUP_INVALID_DATA: ec = GPG_ERR_LDAP_CUP_INV_DATA; break;
case LDAP_CUP_UNSUPPORTED_SCHEME: ec = GPG_ERR_LDAP_CUP_UNSUP_SCHEME; break;
case LDAP_CUP_RELOAD_REQUIRED: ec = GPG_ERR_LDAP_CUP_RELOAD; break;
#endif
#ifdef LDAP_CANCELLED
case LDAP_CANCELLED: ec = GPG_ERR_LDAP_CANCELLED; break;
#endif
#ifdef LDAP_NO_SUCH_OPERATION
case LDAP_NO_SUCH_OPERATION: ec = GPG_ERR_LDAP_NO_SUCH_OPERATION; break;
#endif
#ifdef LDAP_TOO_LATE
case LDAP_TOO_LATE: ec = GPG_ERR_LDAP_TOO_LATE; break;
#endif
#ifdef LDAP_CANNOT_CANCEL
case LDAP_CANNOT_CANCEL: ec = GPG_ERR_LDAP_CANNOT_CANCEL; break;
#endif
#ifdef LDAP_ASSERTION_FAILED
case LDAP_ASSERTION_FAILED: ec = GPG_ERR_LDAP_ASSERTION_FAILED; break;
#endif
#ifdef LDAP_PROXIED_AUTHORIZATION_DENIED
case LDAP_PROXIED_AUTHORIZATION_DENIED:
ec = GPG_ERR_LDAP_PROX_AUTH_DENIED; break;
#endif
default:
#if defined(LDAP_E_ERROR) && defined(LDAP_X_ERROR)
if (LDAP_E_ERROR (code))
ec = GPG_ERR_LDAP_E_GENERAL;
else if (LDAP_X_ERROR (code))
ec = GPG_ERR_LDAP_X_GENERAL;
else
#endif
ec = GPG_ERR_LDAP_GENERAL;
break;
}
return ec;
}
diff --git a/doc/yat2m.c b/doc/yat2m.c
index f780952..5039cc2 100644
--- a/doc/yat2m.c
+++ b/doc/yat2m.c
@@ -1,1582 +1,1582 @@
/* yat2m.c - Yet Another Texi 2 Man converter
* Copyright (C) 2005, 2013 g10 Code GmbH
* Copyright (C) 2006, 2008, 2011 Free Software Foundation, Inc.
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 3 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
- * along with this program; if not, see .
+ * along with this program; if not, see .
*/
/*
This is a simple texinfo to man page converter. It needs some
special markup in th e texinfo and tries best to get a create man
page. It has been designed for the GnuPG man pages and thus only
a few texinfo commands are supported.
To use this you need to add the following macros into your texinfo
source:
@macro manpage {a}
@end macro
@macro mansect {a}
@end macro
@macro manpause
@end macro
@macro mancont
@end macro
They are used by yat2m to select parts of the Texinfo which should
go into the man page. These macros need to be used without leading
left space. Processing starts after a "manpage" macro has been
seen. "mansect" identifies the section and yat2m make sure to
emit the sections in the proper order. Note that @mansect skips
the next input line if that line begins with @section, @subsection or
@chapheading.
To insert verbatim troff markup, the following texinfo code may be
used:
@ifset manverb
.B whateever you want
@end ifset
alternativly a special comment may be used:
@c man:.B whatever you want
This is useful in case you need just one line. If you want to
include parts only in the man page but keep the texinfo
translation you may use:
@ifset isman
stuff to be rendered only on man pages
@end ifset
or to exclude stuff from man pages:
@ifclear isman
stuff not to be rendered on man pages
@end ifclear
the keyword @section is ignored, however @subsection gets rendered
as ".SS". @menu is completely skipped. Several man pages may be
extracted from one file, either using the --store or the --select
option.
If you want to indent tables in the source use this style:
@table foo
@item
@item
@table
@item
@end
@end
Don't change the indentation within a table and keep the same
number of white space at the start of the line. yat2m simply
detects the number of white spaces in front of an @item and remove
this number of spaces from all following lines until a new @item
is found or there are less spaces than for the last @item.
Note that @* does only work correctly if used at the end of an
input line.
*/
#include
#include
#include
#include
#include
#include
#include
#include
#include
#define PGM "yat2m"
#define VERSION "1.0"
/* The maximum length of a line including the linefeed and one extra
character. */
#define LINESIZE 1024
/* Number of allowed condition nestings. */
#define MAX_CONDITION_NESTING 10
/* Option flags. */
static int verbose;
static int quiet;
static int debug;
static const char *opt_source;
static const char *opt_release;
static const char *opt_select;
static const char *opt_include;
static int opt_store;
/* Flag to keep track whether any error occurred. */
static int any_error;
/* Object to keep macro definitions. */
struct macro_s
{
struct macro_s *next;
char *value; /* Malloced value. */
char name[1];
};
typedef struct macro_s *macro_t;
/* List of all defined macros. */
static macro_t macrolist;
/* List of variables set by @set. */
static macro_t variablelist;
/* List of global macro names. The value part is not used. */
static macro_t predefinedmacrolist;
/* Object to keep track of @isset and @ifclear. */
struct condition_s
{
int manverb; /* "manverb" needs special treatment. */
int isset; /* This is an @isset condition. */
char name[1]; /* Name of the condition macro. */
};
typedef struct condition_s *condition_t;
/* The stack used to evaluate conditions. And the current states. */
static condition_t condition_stack[MAX_CONDITION_NESTING];
static int condition_stack_idx;
static int cond_is_active; /* State of ifset/ifclear */
static int cond_in_verbatim; /* State of "manverb". */
/* Object to store one line of content. */
struct line_buffer_s
{
struct line_buffer_s *next;
int verbatim; /* True if LINE contains verbatim data. The default
is Texinfo source. */
char *line;
};
typedef struct line_buffer_s *line_buffer_t;
/* Object to collect the data of a section. */
struct section_buffer_s
{
char *name; /* Malloced name of the section. This may be
NULL to indicate this slot is not used. */
line_buffer_t lines; /* Linked list with the lines of the section. */
line_buffer_t *lines_tail; /* Helper for faster appending to the
linked list. */
line_buffer_t last_line; /* Points to the last line appended. */
};
typedef struct section_buffer_s *section_buffer_t;
/* Variable to keep info about the current page together. */
static struct
{
/* Filename of the current page or NULL if no page is active. Malloced. */
char *name;
/* Number of allocated elements in SECTIONS below. */
size_t n_sections;
/* Array with the data of the sections. */
section_buffer_t sections;
} thepage;
/* The list of standard section names. COMMANDS and ASSUAN are GnuPG
specific. */
static const char * const standard_sections[] =
{ "NAME", "SYNOPSIS", "DESCRIPTION",
"RETURN VALUE", "EXIT STATUS", "ERROR HANDLING", "ERRORS",
"COMMANDS", "OPTIONS", "USAGE", "EXAMPLES", "FILES",
"ENVIRONMENT", "DIAGNOSTICS", "SECURITY", "CONFORMING TO",
"ASSUAN", "NOTES", "BUGS", "AUTHOR", "SEE ALSO", NULL };
/*-- Local prototypes. --*/
static void proc_texi_buffer (FILE *fp, const char *line, size_t len,
int *table_level, int *eol_action);
/* Print diagnostic message and exit with failure. */
static void
die (const char *format, ...)
{
va_list arg_ptr;
fflush (stdout);
fprintf (stderr, "%s: ", PGM);
va_start (arg_ptr, format);
vfprintf (stderr, format, arg_ptr);
va_end (arg_ptr);
putc ('\n', stderr);
exit (1);
}
/* Print diagnostic message. */
static void
err (const char *format, ...)
{
va_list arg_ptr;
fflush (stdout);
if (strncmp (format, "%s:%d:", 6))
fprintf (stderr, "%s: ", PGM);
va_start (arg_ptr, format);
vfprintf (stderr, format, arg_ptr);
va_end (arg_ptr);
putc ('\n', stderr);
any_error = 1;
}
/* Print diagnostic message. */
static void
inf (const char *format, ...)
{
va_list arg_ptr;
fflush (stdout);
fprintf (stderr, "%s: ", PGM);
va_start (arg_ptr, format);
vfprintf (stderr, format, arg_ptr);
va_end (arg_ptr);
putc ('\n', stderr);
}
static void *
xmalloc (size_t n)
{
void *p = malloc (n);
if (!p)
die ("out of core: %s", strerror (errno));
return p;
}
static void *
xcalloc (size_t n, size_t m)
{
void *p = calloc (n, m);
if (!p)
die ("out of core: %s", strerror (errno));
return p;
}
static void *
xrealloc (void *old, size_t n)
{
void *p = realloc (old, n);
if (!p)
die ("out of core: %s", strerror (errno));
return p;
}
static char *
xstrdup (const char *string)
{
void *p = malloc (strlen (string)+1);
if (!p)
die ("out of core: %s", strerror (errno));
strcpy (p, string);
return p;
}
/* Uppercase the ascii characters in STRING. */
static char *
ascii_strupr (char *string)
{
char *p;
for (p = string; *p; p++)
if (!(*p & 0x80))
*p = toupper (*p);
return string;
}
/* Return the current date as an ISO string. */
const char *
isodatestring (void)
{
static char buffer[11+5];
struct tm *tp;
time_t atime = time (NULL);
if (atime < 0)
strcpy (buffer, "????" "-??" "-??");
else
{
tp = gmtime (&atime);
sprintf (buffer,"%04d-%02d-%02d",
1900+tp->tm_year, tp->tm_mon+1, tp->tm_mday );
}
return buffer;
}
/* Add NAME to the list of predefined macros which are global for all
files. */
static void
add_predefined_macro (const char *name)
{
macro_t m;
for (m=predefinedmacrolist; m; m = m->next)
if (!strcmp (m->name, name))
break;
if (!m)
{
m = xcalloc (1, sizeof *m + strlen (name));
strcpy (m->name, name);
m->next = predefinedmacrolist;
predefinedmacrolist = m;
}
}
/* Create or update a macro with name MACRONAME and set its values TO
MACROVALUE. Note that ownership of the macro value is transferred
to this function. */
static void
set_macro (const char *macroname, char *macrovalue)
{
macro_t m;
for (m=macrolist; m; m = m->next)
if (!strcmp (m->name, macroname))
break;
if (m)
free (m->value);
else
{
m = xcalloc (1, sizeof *m + strlen (macroname));
strcpy (m->name, macroname);
m->next = macrolist;
macrolist = m;
}
m->value = macrovalue;
macrovalue = NULL;
}
/* Create or update a variable with name and value given in NAMEANDVALUE. */
static void
set_variable (char *nameandvalue)
{
macro_t m;
const char *value;
char *p;
for (p = nameandvalue; *p && *p != ' ' && *p != '\t'; p++)
;
if (!*p)
value = "";
else
{
*p++ = 0;
while (*p == ' ' || *p == '\t')
p++;
value = p;
}
for (m=variablelist; m; m = m->next)
if (!strcmp (m->name, nameandvalue))
break;
if (m)
free (m->value);
else
{
m = xcalloc (1, sizeof *m + strlen (nameandvalue));
strcpy (m->name, nameandvalue);
m->next = variablelist;
variablelist = m;
}
m->value = xstrdup (value);
}
/* Return true if the macro or variable NAME is set, i.e. not the
empty string and not evaluating to 0. */
static int
macro_set_p (const char *name)
{
macro_t m;
for (m = macrolist; m ; m = m->next)
if (!strcmp (m->name, name))
break;
if (!m)
for (m = variablelist; m ; m = m->next)
if (!strcmp (m->name, name))
break;
if (!m || !m->value || !*m->value)
return 0;
if ((*m->value & 0x80) || !isdigit (*m->value))
return 1; /* Not a digit but some other string. */
return !!atoi (m->value);
}
/* Evaluate the current conditions. */
static void
evaluate_conditions (const char *fname, int lnr)
{
int i;
/* for (i=0; i < condition_stack_idx; i++) */
/* inf ("%s:%d: stack[%d] %s %s %c", */
/* fname, lnr, i, condition_stack[i]->isset? "set":"clr", */
/* condition_stack[i]->name, */
/* (macro_set_p (condition_stack[i]->name) */
/* ^ !condition_stack[i]->isset)? 't':'f'); */
cond_is_active = 1;
cond_in_verbatim = 0;
if (condition_stack_idx)
{
for (i=0; i < condition_stack_idx; i++)
{
if (condition_stack[i]->manverb)
cond_in_verbatim = (macro_set_p (condition_stack[i]->name)
^ !condition_stack[i]->isset);
else if (!(macro_set_p (condition_stack[i]->name)
^ !condition_stack[i]->isset))
{
cond_is_active = 0;
break;
}
}
}
/* inf ("%s:%d: active=%d verbatim=%d", */
/* fname, lnr, cond_is_active, cond_in_verbatim); */
}
/* Push a condition with condition macro NAME onto the stack. If
ISSET is true, a @isset condition is pushed. */
static void
push_condition (const char *name, int isset, const char *fname, int lnr)
{
condition_t cond;
int manverb = 0;
if (condition_stack_idx >= MAX_CONDITION_NESTING)
{
err ("%s:%d: condition nested too deep", fname, lnr);
return;
}
if (!strcmp (name, "manverb"))
{
if (!isset)
{
err ("%s:%d: using \"@ifclear manverb\" is not allowed", fname, lnr);
return;
}
manverb = 1;
}
cond = xcalloc (1, sizeof *cond + strlen (name));
cond->manverb = manverb;
cond->isset = isset;
strcpy (cond->name, name);
condition_stack[condition_stack_idx++] = cond;
evaluate_conditions (fname, lnr);
}
/* Remove the last condition from the stack. ISSET is used for error
reporting. */
static void
pop_condition (int isset, const char *fname, int lnr)
{
if (!condition_stack_idx)
{
err ("%s:%d: unbalanced \"@end %s\"",
fname, lnr, isset?"isset":"isclear");
return;
}
condition_stack_idx--;
free (condition_stack[condition_stack_idx]);
condition_stack[condition_stack_idx] = NULL;
evaluate_conditions (fname, lnr);
}
/* Return a section buffer for the section NAME. Allocate a new buffer
if this is a new section. Keep track of the sections in THEPAGE.
This function may reallocate the section array in THEPAGE. */
static section_buffer_t
get_section_buffer (const char *name)
{
int i;
section_buffer_t sect;
/* If there is no section we put everything into the required NAME
section. Given that this is the first one listed it is likely
that error are easily visible. */
if (!name)
name = "NAME";
for (i=0; i < thepage.n_sections; i++)
{
sect = thepage.sections + i;
if (sect->name && !strcmp (name, sect->name))
return sect;
}
for (i=0; i < thepage.n_sections; i++)
if (!thepage.sections[i].name)
break;
if (i < thepage.n_sections)
sect = thepage.sections + i;
else
{
/* We need to allocate or reallocate the section array. */
size_t old_n = thepage.n_sections;
size_t new_n = 20;
if (!old_n)
thepage.sections = xcalloc (new_n, sizeof *thepage.sections);
else
{
thepage.sections = xrealloc (thepage.sections,
((old_n + new_n)
* sizeof *thepage.sections));
memset (thepage.sections + old_n, 0,
new_n * sizeof *thepage.sections);
}
thepage.n_sections += new_n;
/* Setup the tail pointers. */
for (i=old_n; i < thepage.n_sections; i++)
{
sect = thepage.sections + i;
sect->lines_tail = §->lines;
}
sect = thepage.sections + old_n;
}
/* Store the name. */
assert (!sect->name);
sect->name = xstrdup (name);
return sect;
}
/* Add the content of LINE to the section named SECTNAME. */
static void
add_content (const char *sectname, char *line, int verbatim)
{
section_buffer_t sect;
line_buffer_t lb;
sect = get_section_buffer (sectname);
if (sect->last_line && !sect->last_line->verbatim == !verbatim)
{
/* Lets append that line to the last one. We do this to keep
all lines of the same kind (i.e.verbatim or not) together in
one large buffer. */
size_t n1, n;
lb = sect->last_line;
n1 = strlen (lb->line);
n = n1 + 1 + strlen (line) + 1;
lb->line = xrealloc (lb->line, n);
strcpy (lb->line+n1, "\n");
strcpy (lb->line+n1+1, line);
}
else
{
lb = xcalloc (1, sizeof *lb);
lb->verbatim = verbatim;
lb->line = xstrdup (line);
sect->last_line = lb;
*sect->lines_tail = lb;
sect->lines_tail = &lb->next;
}
}
/* Prepare for a new man page using the filename NAME. */
static void
start_page (char *name)
{
if (verbose)
inf ("starting page '%s'", name);
assert (!thepage.name);
thepage.name = xstrdup (name);
thepage.n_sections = 0;
}
/* Write the .TH entry of the current page. Return -1 if there is a
problem with the page. */
static int
write_th (FILE *fp)
{
char *name, *p;
fputs (".\\\" Created from Texinfo source by yat2m " VERSION "\n", fp);
name = ascii_strupr (xstrdup (thepage.name));
p = strrchr (name, '.');
if (!p || !p[1])
{
err ("no section name in man page '%s'", thepage.name);
free (name);
return -1;
}
*p++ = 0;
fprintf (fp, ".TH %s %s %s \"%s\" \"%s\"\n",
name, p, isodatestring (), opt_release, opt_source);
return 0;
}
/* Process the texinfo command COMMAND (without the leading @) and
write output if needed to FP. REST is the remainer of the line
which should either point to an opening brace or to a white space.
The function returns the number of characters already processed
from REST. LEN is the usable length of REST. TABLE_LEVEL is used to
control the indentation of tables. */
static size_t
proc_texi_cmd (FILE *fp, const char *command, const char *rest, size_t len,
int *table_level, int *eol_action)
{
static struct {
const char *name; /* Name of the command. */
int what; /* What to do with this command. */
const char *lead_in; /* String to print with a opening brace. */
const char *lead_out;/* String to print with the closing brace. */
} cmdtbl[] = {
{ "command", 0, "\\fB", "\\fR" },
{ "code", 0, "\\fB", "\\fR" },
{ "sc", 0, "\\fB", "\\fR" },
{ "var", 0, "\\fI", "\\fR" },
{ "samp", 0, "\\(aq", "\\(aq" },
{ "file", 0, "\\(oq\\fI","\\fR\\(cq" },
{ "env", 0, "\\(oq\\fI","\\fR\\(cq" },
{ "acronym", 0 },
{ "dfn", 0 },
{ "option", 0, "\\fB", "\\fR" },
{ "example", 1, ".RS 2\n.nf\n" },
{ "smallexample", 1, ".RS 2\n.nf\n" },
{ "asis", 7 },
{ "anchor", 7 },
{ "cartouche", 1 },
{ "xref", 0, "see: [", "]" },
{ "pxref", 0, "see: [", "]" },
{ "uref", 0, "(\\fB", "\\fR)" },
{ "footnote",0, " ([", "])" },
{ "emph", 0, "\\fI", "\\fR" },
{ "w", 1 },
{ "c", 5 },
{ "opindex", 1 },
{ "cpindex", 1 },
{ "cindex", 1 },
{ "noindent", 0 },
{ "section", 1 },
{ "chapter", 1 },
{ "subsection", 6, "\n.SS " },
{ "chapheading", 0},
{ "item", 2, ".TP\n.B " },
{ "itemx", 2, ".TP\n.B " },
{ "table", 3 },
{ "itemize", 3 },
{ "bullet", 0, "* " },
{ "*", 0, "\n.br"},
{ "/", 0 },
{ "end", 4 },
{ "quotation",1, ".RS\n\\fB" },
{ "value", 8 },
{ NULL }
};
size_t n;
int i;
const char *s;
const char *lead_out = NULL;
int ignore_args = 0;
for (i=0; cmdtbl[i].name && strcmp (cmdtbl[i].name, command); i++)
;
if (cmdtbl[i].name)
{
s = cmdtbl[i].lead_in;
if (s)
fputs (s, fp);
lead_out = cmdtbl[i].lead_out;
switch (cmdtbl[i].what)
{
case 1: /* Throw away the entire line. */
s = memchr (rest, '\n', len);
return s? (s-rest)+1 : len;
case 2: /* Handle @item. */
break;
case 3: /* Handle table. */
if (++(*table_level) > 1)
fputs (".RS\n", fp);
/* Now throw away the entire line. */
s = memchr (rest, '\n', len);
return s? (s-rest)+1 : len;
break;
case 4: /* Handle end. */
for (s=rest, n=len; n && (*s == ' ' || *s == '\t'); s++, n--)
;
if (n >= 5 && !memcmp (s, "table", 5)
&& (!n || s[5] == ' ' || s[5] == '\t' || s[5] == '\n'))
{
if ((*table_level)-- > 1)
fputs (".RE\n", fp);
}
else if (n >= 7 && !memcmp (s, "example", 7)
&& (!n || s[7] == ' ' || s[7] == '\t' || s[7] == '\n'))
{
fputs (".fi\n.RE\n", fp);
}
else if (n >= 12 && !memcmp (s, "smallexample", 12)
&& (!n || s[12] == ' ' || s[12] == '\t' || s[12] == '\n'))
{
fputs (".fi\n.RE\n", fp);
}
else if (n >= 9 && !memcmp (s, "quotation", 9)
&& (!n || s[9] == ' ' || s[9] == '\t' || s[9] == '\n'))
{
fputs ("\\fR\n.RE\n", fp);
}
/* Now throw away the entire line. */
s = memchr (rest, '\n', len);
return s? (s-rest)+1 : len;
case 5: /* Handle special comments. */
for (s=rest, n=len; n && (*s == ' ' || *s == '\t'); s++, n--)
;
if (n >= 4 && !memcmp (s, "man:", 4))
{
for (s+=4, n-=4; n && *s != '\n'; n--, s++)
putc (*s, fp);
putc ('\n', fp);
}
/* Now throw away the entire line. */
s = memchr (rest, '\n', len);
return s? (s-rest)+1 : len;
case 6:
*eol_action = 1;
break;
case 7:
ignore_args = 1;
break;
case 8:
ignore_args = 1;
if (*rest != '{')
{
err ("opening brace for command '%s' missing", command);
return len;
}
else
{
/* Find closing brace. */
for (s=rest+1, n=1; *s && n < len; s++, n++)
if (*s == '}')
break;
if (*s != '}')
{
err ("closing brace for command '%s' not found", command);
return len;
}
else
{
size_t len = s - (rest + 1);
macro_t m;
for (m = variablelist; m; m = m->next)
if (strlen (m->name) == len
&&!strncmp (m->name, rest+1, len))
break;
if (m)
fputs (m->value, fp);
else
inf ("texinfo variable '%.*s' is not set",
(int)len, rest+1);
}
}
break;
default:
break;
}
}
else /* macro */
{
macro_t m;
for (m = macrolist; m ; m = m->next)
if (!strcmp (m->name, command))
break;
if (m)
{
proc_texi_buffer (fp, m->value, strlen (m->value),
table_level, eol_action);
ignore_args = 1; /* Parameterized macros are not yet supported. */
}
else
inf ("texinfo command '%s' not supported (%.*s)", command,
((s = memchr (rest, '\n', len)), (s? (s-rest) : len)), rest);
}
if (*rest == '{')
{
/* Find matching closing brace. */
for (s=rest+1, n=1, i=1; i && *s && n < len; s++, n++)
if (*s == '{')
i++;
else if (*s == '}')
i--;
if (i)
{
err ("closing brace for command '%s' not found", command);
return len;
}
if (n > 2 && !ignore_args)
proc_texi_buffer (fp, rest+1, n-2, table_level, eol_action);
}
else
n = 0;
if (lead_out)
fputs (lead_out, fp);
return n;
}
/* Process the string LINE with LEN bytes of Texinfo content. */
static void
proc_texi_buffer (FILE *fp, const char *line, size_t len,
int *table_level, int *eol_action)
{
const char *s;
char cmdbuf[256];
int cmdidx = 0;
int in_cmd = 0;
size_t n;
for (s=line; *s && len; s++, len--)
{
if (in_cmd)
{
if (in_cmd == 1)
{
switch (*s)
{
case '@': case '{': case '}':
putc (*s, fp); in_cmd = 0;
break;
case ':': /* Not ending a sentence flag. */
in_cmd = 0;
break;
case '.': case '!': case '?': /* Ending a sentence. */
putc (*s, fp); in_cmd = 0;
break;
case ' ': case '\t': case '\n': /* Non collapsing spaces. */
putc (*s, fp); in_cmd = 0;
break;
default:
cmdidx = 0;
cmdbuf[cmdidx++] = *s;
in_cmd++;
break;
}
}
else if (*s == '{' || *s == ' ' || *s == '\t' || *s == '\n')
{
cmdbuf[cmdidx] = 0;
n = proc_texi_cmd (fp, cmdbuf, s, len, table_level, eol_action);
assert (n <= len);
s += n; len -= n;
s--; len++;
in_cmd = 0;
}
else if (cmdidx < sizeof cmdbuf -1)
cmdbuf[cmdidx++] = *s;
else
{
err ("texinfo command too long - ignored");
in_cmd = 0;
}
}
else if (*s == '@')
in_cmd = 1;
else if (*s == '\n')
{
switch (*eol_action)
{
case 1: /* Create a dummy paragraph. */
fputs ("\n\\ \n", fp);
break;
default:
putc (*s, fp);
}
*eol_action = 0;
}
else if (*s == '\\')
fputs ("\\\\", fp);
else
putc (*s, fp);
}
if (in_cmd > 1)
{
cmdbuf[cmdidx] = 0;
n = proc_texi_cmd (fp, cmdbuf, s, len, table_level, eol_action);
assert (n <= len);
s += n; len -= n;
s--; len++;
in_cmd = 0;
}
}
/* Do something with the Texinfo line LINE. */
static void
parse_texi_line (FILE *fp, const char *line, int *table_level)
{
int eol_action = 0;
/* A quick test whether there are any texinfo commands. */
if (!strchr (line, '@'))
{
fputs (line, fp);
putc ('\n', fp);
return;
}
proc_texi_buffer (fp, line, strlen (line), table_level, &eol_action);
putc ('\n', fp);
}
/* Write all the lines LINES to FP. */
static void
write_content (FILE *fp, line_buffer_t lines)
{
line_buffer_t line;
int table_level = 0;
for (line = lines; line; line = line->next)
{
if (line->verbatim)
{
fputs (line->line, fp);
putc ('\n', fp);
}
else
{
/* fputs ("TEXI---", fp); */
/* fputs (line->line, fp); */
/* fputs ("---\n", fp); */
parse_texi_line (fp, line->line, &table_level);
}
}
}
static int
is_standard_section (const char *name)
{
int i;
const char *s;
for (i=0; (s=standard_sections[i]); i++)
if (!strcmp (s, name))
return 1;
return 0;
}
/* Finish a page; that is sort the data and write it out to the file. */
static void
finish_page (void)
{
FILE *fp;
section_buffer_t sect = NULL;
int idx;
const char *s;
int i;
if (!thepage.name)
return; /* No page active. */
if (verbose)
inf ("finishing page '%s'", thepage.name);
if (opt_select)
{
if (!strcmp (opt_select, thepage.name))
{
inf ("selected '%s'", thepage.name );
fp = stdout;
}
else
{
fp = fopen ( "/dev/null", "w" );
if (!fp)
die ("failed to open /dev/null: %s\n", strerror (errno));
}
}
else if (opt_store)
{
inf ("writing '%s'", thepage.name );
fp = fopen ( thepage.name, "w" );
if (!fp)
die ("failed to create '%s': %s\n", thepage.name, strerror (errno));
}
else
fp = stdout;
if (write_th (fp))
goto leave;
for (idx=0; (s=standard_sections[idx]); idx++)
{
for (i=0; i < thepage.n_sections; i++)
{
sect = thepage.sections + i;
if (sect->name && !strcmp (s, sect->name))
break;
}
if (i == thepage.n_sections)
sect = NULL;
if (sect)
{
fprintf (fp, ".SH %s\n", sect->name);
write_content (fp, sect->lines);
/* Now continue with all non standard sections directly
following this one. */
for (i++; i < thepage.n_sections; i++)
{
sect = thepage.sections + i;
if (sect->name && is_standard_section (sect->name))
break;
if (sect->name)
{
fprintf (fp, ".SH %s\n", sect->name);
write_content (fp, sect->lines);
}
}
}
}
leave:
if (fp != stdout)
fclose (fp);
free (thepage.name);
thepage.name = NULL;
/* FIXME: Cleanup the content. */
}
/* Parse one Texinfo file and create manpages according to the
embedded instructions. */
static void
parse_file (const char *fname, FILE *fp, char **section_name, int in_pause)
{
char *line;
int lnr = 0;
/* Fixme: The following state variables don't carry over to include
files. */
int skip_to_end = 0; /* Used to skip over menu entries. */
int skip_sect_line = 0; /* Skip after @mansect. */
int item_indent = 0; /* How far is the current @item indented. */
/* Helper to define a macro. */
char *macroname = NULL;
char *macrovalue = NULL;
size_t macrovaluesize = 0;
size_t macrovalueused = 0;
line = xmalloc (LINESIZE);
while (fgets (line, LINESIZE, fp))
{
size_t n = strlen (line);
int got_line = 0;
char *p, *pend;
lnr++;
if (!n || line[n-1] != '\n')
{
err ("%s:%d: trailing linefeed missing, line too long or "
"embedded Nul character", fname, lnr);
break;
}
line[--n] = 0;
/* Kludge to allow indentation of tables. */
for (p=line; *p == ' ' || *p == '\t'; p++)
;
if (*p)
{
if (*p == '@' && !strncmp (p+1, "item", 4))
item_indent = p - line; /* Set a new indent level. */
else if (p - line < item_indent)
item_indent = 0; /* Switch off indention. */
if (item_indent)
{
memmove (line, line+item_indent, n - item_indent + 1);
n -= item_indent;
}
}
if (*line == '@')
{
for (p=line+1, n=1; *p && *p != ' ' && *p != '\t'; p++)
n++;
while (*p == ' ' || *p == '\t')
p++;
}
else
p = line;
/* Take action on macro. */
if (macroname)
{
if (n == 4 && !memcmp (line, "@end", 4)
&& (line[4]==' '||line[4]=='\t'||!line[4])
&& !strncmp (p, "macro", 5)
&& (p[5]==' '||p[5]=='\t'||!p[5]))
{
if (macrovalueused)
macrovalue[--macrovalueused] = 0; /* Kill the last LF. */
macrovalue[macrovalueused] = 0; /* Terminate macro. */
macrovalue = xrealloc (macrovalue, macrovalueused+1);
set_macro (macroname, macrovalue);
macrovalue = NULL;
free (macroname);
macroname = NULL;
}
else
{
if (macrovalueused + strlen (line) + 2 >= macrovaluesize)
{
macrovaluesize += strlen (line) + 256;
macrovalue = xrealloc (macrovalue, macrovaluesize);
}
strcpy (macrovalue+macrovalueused, line);
macrovalueused += strlen (line);
macrovalue[macrovalueused++] = '\n';
}
continue;
}
if (n >= 5 && !memcmp (line, "@node", 5)
&& (line[5]==' '||line[5]=='\t'||!line[5]))
{
/* Completey ignore @node lines. */
continue;
}
if (skip_sect_line)
{
skip_sect_line = 0;
if (!strncmp (line, "@section", 8)
|| !strncmp (line, "@subsection", 11)
|| !strncmp (line, "@chapheading", 12))
continue;
}
/* We only parse lines we need and ignore the rest. There are a
few macros used to control this as well as one @ifset
command. Parts we know about are saved away into containers
separate for each section. */
/* First process ifset/ifclear commands. */
if (*line == '@')
{
if (n == 6 && !memcmp (line, "@ifset", 6)
&& (line[6]==' '||line[6]=='\t'))
{
for (p=line+7; *p == ' ' || *p == '\t'; p++)
;
if (!*p)
{
err ("%s:%d: name missing after \"@ifset\"", fname, lnr);
continue;
}
for (pend=p; *pend && *pend != ' ' && *pend != '\t'; pend++)
;
*pend = 0; /* Ignore rest of the line. */
push_condition (p, 1, fname, lnr);
continue;
}
else if (n == 8 && !memcmp (line, "@ifclear", 8)
&& (line[8]==' '||line[8]=='\t'))
{
for (p=line+9; *p == ' ' || *p == '\t'; p++)
;
if (!*p)
{
err ("%s:%d: name missing after \"@ifsclear\"", fname, lnr);
continue;
}
for (pend=p; *pend && *pend != ' ' && *pend != '\t'; pend++)
;
*pend = 0; /* Ignore rest of the line. */
push_condition (p, 0, fname, lnr);
continue;
}
else if (n == 4 && !memcmp (line, "@end", 4)
&& (line[4]==' '||line[4]=='\t')
&& !strncmp (p, "ifset", 5)
&& (p[5]==' '||p[5]=='\t'||!p[5]))
{
pop_condition (1, fname, lnr);
continue;
}
else if (n == 4 && !memcmp (line, "@end", 4)
&& (line[4]==' '||line[4]=='\t')
&& !strncmp (p, "ifclear", 7)
&& (p[7]==' '||p[7]=='\t'||!p[7]))
{
pop_condition (0, fname, lnr);
continue;
}
}
/* Take action on ifset/ifclear. */
if (!cond_is_active)
continue;
/* Process commands. */
if (*line == '@')
{
if (skip_to_end
&& n == 4 && !memcmp (line, "@end", 4)
&& (line[4]==' '||line[4]=='\t'||!line[4]))
{
skip_to_end = 0;
}
else if (cond_in_verbatim)
{
got_line = 1;
}
else if (n == 6 && !memcmp (line, "@macro", 6))
{
macroname = xstrdup (p);
macrovalue = xmalloc ((macrovaluesize = 1024));
macrovalueused = 0;
}
else if (n == 4 && !memcmp (line, "@set", 4))
{
set_variable (p);
}
else if (n == 8 && !memcmp (line, "@manpage", 8))
{
free (*section_name);
*section_name = NULL;
finish_page ();
start_page (p);
in_pause = 0;
}
else if (n == 8 && !memcmp (line, "@mansect", 8))
{
if (!thepage.name)
err ("%s:%d: section outside of a man page", fname, lnr);
else
{
free (*section_name);
*section_name = ascii_strupr (xstrdup (p));
in_pause = 0;
skip_sect_line = 1;
}
}
else if (n == 9 && !memcmp (line, "@manpause", 9))
{
if (!*section_name)
err ("%s:%d: pausing outside of a man section", fname, lnr);
else if (in_pause)
err ("%s:%d: already pausing", fname, lnr);
else
in_pause = 1;
}
else if (n == 8 && !memcmp (line, "@mancont", 8))
{
if (!*section_name)
err ("%s:%d: continue outside of a man section", fname, lnr);
else if (!in_pause)
err ("%s:%d: continue while not pausing", fname, lnr);
else
in_pause = 0;
}
else if (n == 5 && !memcmp (line, "@menu", 5)
&& (line[5]==' '||line[5]=='\t'||!line[5]))
{
skip_to_end = 1;
}
else if (n == 8 && !memcmp (line, "@include", 8)
&& (line[8]==' '||line[8]=='\t'||!line[8]))
{
char *incname = xstrdup (p);
FILE *incfp = fopen (incname, "r");
if (!incfp && opt_include && *opt_include && *p != '/')
{
free (incname);
incname = xmalloc (strlen (opt_include) + 1
+ strlen (p) + 1);
strcpy (incname, opt_include);
if ( incname[strlen (incname)-1] != '/' )
strcat (incname, "/");
strcat (incname, p);
incfp = fopen (incname, "r");
}
if (!incfp)
err ("can't open include file '%s':%s",
incname, strerror (errno));
else
{
parse_file (incname, incfp, section_name, in_pause);
fclose (incfp);
}
free (incname);
}
else if (n == 4 && !memcmp (line, "@bye", 4)
&& (line[4]==' '||line[4]=='\t'||!line[4]))
{
break;
}
else if (!skip_to_end)
got_line = 1;
}
else if (!skip_to_end)
got_line = 1;
if (got_line && cond_in_verbatim)
add_content (*section_name, line, 1);
else if (got_line && thepage.name && *section_name && !in_pause)
add_content (*section_name, line, 0);
}
if (ferror (fp))
err ("%s:%d: read error: %s", fname, lnr, strerror (errno));
free (macroname);
free (macrovalue);
free (line);
}
static void
top_parse_file (const char *fname, FILE *fp)
{
char *section_name = NULL; /* Name of the current section or NULL
if not in a section. */
macro_t m;
while (macrolist)
{
macro_t next = macrolist->next;
free (macrolist->value);
free (macrolist);
macrolist = next;
}
while (variablelist)
{
macro_t next = variablelist->next;
free (variablelist->value);
free (variablelist);
variablelist = next;
}
for (m=predefinedmacrolist; m; m = m->next)
set_macro (m->name, xstrdup ("1"));
cond_is_active = 1;
cond_in_verbatim = 0;
parse_file (fname, fp, §ion_name, 0);
free (section_name);
finish_page ();
}
int
main (int argc, char **argv)
{
int last_argc = -1;
opt_source = "GNU";
opt_release = "";
/* Define default macros. The trick is that these macros are not
defined when using the actual texinfo renderer. */
add_predefined_macro ("isman");
add_predefined_macro ("manverb");
/* Option parsing. */
if (argc)
{
argc--; argv++;
}
while (argc && last_argc != argc )
{
last_argc = argc;
if (!strcmp (*argv, "--"))
{
argc--; argv++;
break;
}
else if (!strcmp (*argv, "--help"))
{
puts (
"Usage: " PGM " [OPTION] [FILE]\n"
"Extract man pages from a Texinfo source.\n\n"
" --source NAME use NAME as source field\n"
" --release STRING use STRING as the release field\n"
" --store write output using @manpage name\n"
" --select NAME only output pages with @manpage NAME\n"
" --verbose enable extra informational output\n"
" --debug enable additional debug output\n"
" --help display this help and exit\n"
" -I DIR also search in include DIR\n"
" -D gpgone the only useable define\n\n"
"With no FILE, or when FILE is -, read standard input.\n\n"
"Report bugs to .");
exit (0);
}
else if (!strcmp (*argv, "--version"))
{
puts (PGM " " VERSION "\n"
"Copyright (C) 2005 g10 Code GmbH\n"
"This program comes with ABSOLUTELY NO WARRANTY.\n"
"This is free software, and you are welcome to redistribute it\n"
"under certain conditions. See the file COPYING for details.");
exit (0);
}
else if (!strcmp (*argv, "--verbose"))
{
verbose = 1;
argc--; argv++;
}
else if (!strcmp (*argv, "--quiet"))
{
quiet = 1;
argc--; argv++;
}
else if (!strcmp (*argv, "--debug"))
{
verbose = debug = 1;
argc--; argv++;
}
else if (!strcmp (*argv, "--source"))
{
argc--; argv++;
if (argc)
{
opt_source = *argv;
argc--; argv++;
}
}
else if (!strcmp (*argv, "--release"))
{
argc--; argv++;
if (argc)
{
opt_release = *argv;
argc--; argv++;
}
}
else if (!strcmp (*argv, "--store"))
{
opt_store = 1;
argc--; argv++;
}
else if (!strcmp (*argv, "--select"))
{
argc--; argv++;
if (argc)
{
opt_select = strrchr (*argv, '/');
if (opt_select)
opt_select++;
else
opt_select = *argv;
argc--; argv++;
}
}
else if (!strcmp (*argv, "-I"))
{
argc--; argv++;
if (argc)
{
opt_include = *argv;
argc--; argv++;
}
}
else if (!strcmp (*argv, "-D"))
{
argc--; argv++;
if (argc)
{
add_predefined_macro (*argv);
argc--; argv++;
}
}
}
if (argc > 1)
die ("usage: " PGM " [OPTION] [FILE] (try --help for more information)\n");
/* Start processing. */
if (argc && strcmp (*argv, "-"))
{
FILE *fp = fopen (*argv, "rb");
if (!fp)
die ("%s:0: can't open file: %s", *argv, strerror (errno));
top_parse_file (*argv, fp);
fclose (fp);
}
else
top_parse_file ("-", stdin);
return !!any_error;
}
/*
Local Variables:
compile-command: "gcc -Wall -g -Wall -o yat2m yat2m.c"
End:
*/
diff --git a/m4/ac_prog_cc_for_build.m4 b/m4/ac_prog_cc_for_build.m4
index 42c91ce..32329c7 100644
--- a/m4/ac_prog_cc_for_build.m4
+++ b/m4/ac_prog_cc_for_build.m4
@@ -1,149 +1,149 @@
dnl Available from the GNU Autoconf Macro Archive at:
-dnl http://www.gnu.org/software/ac-archive/htmldoc/ac_prog_cc_for_build.html
+dnl https://www.gnu.org/software/ac-archive/htmldoc/ac_prog_cc_for_build.html
dnl
dnl All content of this archive is free software;
dnl you can redistribute it and/or modify it under the terms of the GNU
dnl General Public License as published by the Free Software Foundation;
dnl either version 2, or (at your option) any later version.
dnl
dnl As a special exception, the respective Autoconf Macro's copyright
dnl owner gives unlimited permission to copy, distribute and modify the
dnl configure scripts that are the output of Autoconf when processing the
dnl Macro. You need not follow the terms of the GNU General Public License
dnl when using or distributing such scripts, even though portions of the
dnl text of the Macro appear in them. The GNU General Public License (GPL)
dnl does govern all other use of the material that constitutes the
dnl Autoconf Macro.
dnl
dnl This special exception to the GPL applies to versions of the Autoconf
dnl Macro released by the GNU Autoconf Macro Archive. When you make and
dnl distribute a modified version of the Autoconf Macro, you may extend
dnl this special exception to the GPL to apply to your modified version as
dnl well.
dnl Synopsis AC_PROG_CC_FOR_BUILD
dnl
dnl This macro searches for a C compiler that generates native
dnl executables, that is a C compiler that surely is not a cross-compiler.
dnl This can be useful if you have to generate source code at compile-time
dnl like for example GCC does.
dnl
dnl The macro sets the CC_FOR_BUILD and CPP_FOR_BUILD macros to anything
dnl needed to compile or link (CC_FOR_BUILD) and preprocess
dnl (CPP_FOR_BUILD). The value of these variables can be overridden by
dnl the user by specifying a compiler with an environment variable (like
dnl you do for standard CC).
dnl
dnl It also sets BUILD_EXEEXT and BUILD_OBJEXT to the executable and
dnl object file extensions for the build platform, and GCC_FOR_BUILD
dnl to `yes' if the compiler we found is GCC. All these variables but
dnl GCC_FOR_BUILD are substituted in the Makefile.
dnl
dnl Author Paolo Bonzini
dnl
AC_DEFUN([AC_PROG_CC_FOR_BUILD], [dnl
AC_REQUIRE([AC_PROG_CC])dnl
AC_REQUIRE([AC_PROG_CPP])dnl
AC_REQUIRE([AC_EXEEXT])dnl
AC_REQUIRE([AC_CANONICAL_SYSTEM])dnl
dnl
pushdef([AC_TRY_COMPILER], [
cat > conftest.$ac_ext << EOF
#line __oline__ "configure"
#include "confdefs.h"
[$1]
EOF
# If we can't run a trivial program, we are probably using a cross
compiler.
# Fail miserably.
if AC_TRY_EVAL(ac_link) && test -s conftest${ac_exeext} && (./conftest;
exit) 2>/dev/null; then
[$2]=yes
else
echo "configure: failed program was:" >&AC_FD_CC
cat conftest.$ac_ext >&AC_FD_CC
[$2]=no
fi
[$3]=no
rm -fr conftest*])dnl
dnl Use the standard macros, but make them use other variable names
dnl
pushdef([cross_compiling], [#])dnl
pushdef([ac_cv_prog_CPP], ac_cv_build_prog_CPP)dnl
pushdef([ac_cv_prog_gcc], ac_cv_build_prog_gcc)dnl
pushdef([ac_cv_prog_cc_works], ac_cv_build_prog_cc_works)dnl
pushdef([ac_cv_prog_cc_cross], ac_cv_build_prog_cc_cross)dnl
pushdef([ac_cv_prog_cc_g], ac_cv_build_prog_cc_g)dnl
pushdef([ac_cv_exeext], ac_cv_build_exeext)dnl
pushdef([ac_cv_objext], ac_cv_build_objext)dnl
pushdef([ac_exeext], ac_build_exeext)dnl
pushdef([ac_objext], ac_build_objext)dnl
pushdef([CC], CC_FOR_BUILD)dnl
pushdef([CPP], CPP_FOR_BUILD)dnl
pushdef([CFLAGS], CFLAGS_FOR_BUILD)dnl
pushdef([CPPFLAGS], CPPFLAGS_FOR_BUILD)dnl
pushdef([host], build)dnl
pushdef([host_alias], build_alias)dnl
pushdef([host_cpu], build_cpu)dnl
pushdef([host_vendor], build_vendor)dnl
pushdef([host_os], build_os)dnl
pushdef([ac_cv_host], ac_cv_build)dnl
pushdef([ac_cv_host_alias], ac_cv_build_alias)dnl
pushdef([ac_cv_host_cpu], ac_cv_build_cpu)dnl
pushdef([ac_cv_host_vendor], ac_cv_build_vendor)dnl
pushdef([ac_cv_host_os], ac_cv_build_os)dnl
pushdef([ac_cpp], ac_build_cpp)dnl
pushdef([ac_compile], ac_build_compile)dnl
pushdef([ac_link], ac_build_link)dnl
dnl Defeat the anti-duplication mechanism
dnl
dnl undefine([AC_PROVIDE_AC_PROG_CPP])dnl
dnl undefine([AC_PROVIDE_AC_PROG_C])dnl
dnl undefine([AC_PROVIDE_AC_EXEEXT])dnl
AC_PROG_CC
AC_PROG_CPP
AC_EXEEXT
dnl Restore the old definitions
dnl
popdef([AC_TRY_COMPILER])dnl
popdef([ac_link])dnl
popdef([ac_compile])dnl
popdef([ac_cpp])dnl
popdef([ac_cv_host_os])dnl
popdef([ac_cv_host_vendor])dnl
popdef([ac_cv_host_cpu])dnl
popdef([ac_cv_host_alias])dnl
popdef([ac_cv_host])dnl
popdef([host_os])dnl
popdef([host_vendor])dnl
popdef([host_cpu])dnl
popdef([host_alias])dnl
popdef([host])dnl
popdef([CPPFLAGS])dnl
popdef([CFLAGS])dnl
popdef([CPP])dnl
popdef([CC])dnl
popdef([ac_objext])dnl
popdef([ac_exeext])dnl
popdef([ac_cv_objext])dnl
popdef([ac_cv_exeext])dnl
popdef([ac_cv_prog_cc_g])dnl
popdef([ac_cv_prog_cc_works])dnl
popdef([ac_cv_prog_cc_cross])dnl
popdef([ac_cv_prog_gcc])dnl
popdef([ac_cv_prog_CPP])dnl
popdef([cross_compiling])dnl
dnl Finally, set Makefile variables
dnl
BUILD_EXEEXT=$ac_build_exeext
BUILD_OBJEXT=$ac_build_objext
AC_SUBST(BUILD_EXEEXT)dnl
AC_SUBST(BUILD_OBJEXT)dnl
AC_SUBST([CFLAGS_FOR_BUILD])dnl
AC_SUBST([CPPFLAGS_FOR_BUILD])dnl
])
diff --git a/m4/libtool.m4 b/m4/libtool.m4
index 1d62b05..0cd84af 100644
--- a/m4/libtool.m4
+++ b/m4/libtool.m4
@@ -1,8027 +1,8027 @@
# libtool.m4 - Configure libtool for the host system. -*-Autoconf-*-
#
# Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2003, 2004, 2005,
# 2006, 2007, 2008, 2009, 2010, 2011 Free Software
# Foundation, Inc.
# Written by Gordon Matzigkeit, 1996
#
# This file is free software; the Free Software Foundation gives
# unlimited permission to copy and/or distribute it, with or without
# modifications, as long as this notice is preserved.
m4_define([_LT_COPYING], [dnl
# Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2003, 2004, 2005,
# 2006, 2007, 2008, 2009, 2010, 2011 Free Software
# Foundation, Inc.
# Written by Gordon Matzigkeit, 1996
#
# This file is part of GNU Libtool.
#
# GNU Libtool is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License as
# published by the Free Software Foundation; either version 2 of
# the License, or (at your option) any later version.
#
# As a special exception to the GNU General Public License,
# if you distribute this file as part of a program or library that
# is built using GNU Libtool, you may include this file under the
# same distribution terms that you use for the rest of that program.
#
# GNU Libtool is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with GNU Libtool; see the file COPYING. If not, a copy
-# can be downloaded from http://www.gnu.org/licenses/gpl.html, or
+# can be downloaded from https://www.gnu.org/licenses/gpl.html, or
# obtained by writing to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
])
# serial 57 LT_INIT
# LT_PREREQ(VERSION)
# ------------------
# Complain and exit if this libtool version is less that VERSION.
m4_defun([LT_PREREQ],
[m4_if(m4_version_compare(m4_defn([LT_PACKAGE_VERSION]), [$1]), -1,
[m4_default([$3],
[m4_fatal([Libtool version $1 or higher is required],
63)])],
[$2])])
# _LT_CHECK_BUILDDIR
# ------------------
# Complain if the absolute build directory name contains unusual characters
m4_defun([_LT_CHECK_BUILDDIR],
[case `pwd` in
*\ * | *\ *)
AC_MSG_WARN([Libtool does not cope well with whitespace in `pwd`]) ;;
esac
])
# LT_INIT([OPTIONS])
# ------------------
AC_DEFUN([LT_INIT],
[AC_PREREQ([2.58])dnl We use AC_INCLUDES_DEFAULT
AC_REQUIRE([AC_CONFIG_AUX_DIR_DEFAULT])dnl
AC_BEFORE([$0], [LT_LANG])dnl
AC_BEFORE([$0], [LT_OUTPUT])dnl
AC_BEFORE([$0], [LTDL_INIT])dnl
m4_require([_LT_CHECK_BUILDDIR])dnl
dnl Autoconf doesn't catch unexpanded LT_ macros by default:
m4_pattern_forbid([^_?LT_[A-Z_]+$])dnl
m4_pattern_allow([^(_LT_EOF|LT_DLGLOBAL|LT_DLLAZY_OR_NOW|LT_MULTI_MODULE)$])dnl
dnl aclocal doesn't pull ltoptions.m4, ltsugar.m4, or ltversion.m4
dnl unless we require an AC_DEFUNed macro:
AC_REQUIRE([LTOPTIONS_VERSION])dnl
AC_REQUIRE([LTSUGAR_VERSION])dnl
AC_REQUIRE([LTVERSION_VERSION])dnl
AC_REQUIRE([LTOBSOLETE_VERSION])dnl
m4_require([_LT_PROG_LTMAIN])dnl
_LT_SHELL_INIT([SHELL=${CONFIG_SHELL-/bin/sh}])
dnl Parse OPTIONS
_LT_SET_OPTIONS([$0], [$1])
# This can be used to rebuild libtool when needed
LIBTOOL_DEPS="$ltmain"
# Always use our own libtool.
LIBTOOL='$(SHELL) $(top_builddir)/libtool'
AC_SUBST(LIBTOOL)dnl
_LT_SETUP
# Only expand once:
m4_define([LT_INIT])
])# LT_INIT
# Old names:
AU_ALIAS([AC_PROG_LIBTOOL], [LT_INIT])
AU_ALIAS([AM_PROG_LIBTOOL], [LT_INIT])
dnl aclocal-1.4 backwards compatibility:
dnl AC_DEFUN([AC_PROG_LIBTOOL], [])
dnl AC_DEFUN([AM_PROG_LIBTOOL], [])
# _LT_CC_BASENAME(CC)
# -------------------
# Calculate cc_basename. Skip known compiler wrappers and cross-prefix.
m4_defun([_LT_CC_BASENAME],
[for cc_temp in $1""; do
case $cc_temp in
compile | *[[\\/]]compile | ccache | *[[\\/]]ccache ) ;;
distcc | *[[\\/]]distcc | purify | *[[\\/]]purify ) ;;
\-*) ;;
*) break;;
esac
done
cc_basename=`$ECHO "$cc_temp" | $SED "s%.*/%%; s%^$host_alias-%%"`
])
# _LT_FILEUTILS_DEFAULTS
# ----------------------
# It is okay to use these file commands and assume they have been set
# sensibly after `m4_require([_LT_FILEUTILS_DEFAULTS])'.
m4_defun([_LT_FILEUTILS_DEFAULTS],
[: ${CP="cp -f"}
: ${MV="mv -f"}
: ${RM="rm -f"}
])# _LT_FILEUTILS_DEFAULTS
# _LT_SETUP
# ---------
m4_defun([_LT_SETUP],
[AC_REQUIRE([AC_CANONICAL_HOST])dnl
AC_REQUIRE([AC_CANONICAL_BUILD])dnl
AC_REQUIRE([_LT_PREPARE_SED_QUOTE_VARS])dnl
AC_REQUIRE([_LT_PROG_ECHO_BACKSLASH])dnl
_LT_DECL([], [PATH_SEPARATOR], [1], [The PATH separator for the build system])dnl
dnl
_LT_DECL([], [host_alias], [0], [The host system])dnl
_LT_DECL([], [host], [0])dnl
_LT_DECL([], [host_os], [0])dnl
dnl
_LT_DECL([], [build_alias], [0], [The build system])dnl
_LT_DECL([], [build], [0])dnl
_LT_DECL([], [build_os], [0])dnl
dnl
AC_REQUIRE([AC_PROG_CC])dnl
AC_REQUIRE([LT_PATH_LD])dnl
AC_REQUIRE([LT_PATH_NM])dnl
dnl
AC_REQUIRE([AC_PROG_LN_S])dnl
test -z "$LN_S" && LN_S="ln -s"
_LT_DECL([], [LN_S], [1], [Whether we need soft or hard links])dnl
dnl
AC_REQUIRE([LT_CMD_MAX_LEN])dnl
_LT_DECL([objext], [ac_objext], [0], [Object file suffix (normally "o")])dnl
_LT_DECL([], [exeext], [0], [Executable file suffix (normally "")])dnl
dnl
m4_require([_LT_FILEUTILS_DEFAULTS])dnl
m4_require([_LT_CHECK_SHELL_FEATURES])dnl
m4_require([_LT_PATH_CONVERSION_FUNCTIONS])dnl
m4_require([_LT_CMD_RELOAD])dnl
m4_require([_LT_CHECK_MAGIC_METHOD])dnl
m4_require([_LT_CHECK_SHAREDLIB_FROM_LINKLIB])dnl
m4_require([_LT_CMD_OLD_ARCHIVE])dnl
m4_require([_LT_CMD_GLOBAL_SYMBOLS])dnl
m4_require([_LT_WITH_SYSROOT])dnl
_LT_CONFIG_LIBTOOL_INIT([
# See if we are running on zsh, and set the options which allow our
# commands through without removal of \ escapes INIT.
if test -n "\${ZSH_VERSION+set}" ; then
setopt NO_GLOB_SUBST
fi
])
if test -n "${ZSH_VERSION+set}" ; then
setopt NO_GLOB_SUBST
fi
_LT_CHECK_OBJDIR
m4_require([_LT_TAG_COMPILER])dnl
case $host_os in
aix3*)
# AIX sometimes has problems with the GCC collect2 program. For some
# reason, if we set the COLLECT_NAMES environment variable, the problems
# vanish in a puff of smoke.
if test "X${COLLECT_NAMES+set}" != Xset; then
COLLECT_NAMES=
export COLLECT_NAMES
fi
;;
esac
# Global variables:
ofile=libtool
can_build_shared=yes
# All known linkers require a `.a' archive for static linking (except MSVC,
# which needs '.lib').
libext=a
with_gnu_ld="$lt_cv_prog_gnu_ld"
old_CC="$CC"
old_CFLAGS="$CFLAGS"
# Set sane defaults for various variables
test -z "$CC" && CC=cc
test -z "$LTCC" && LTCC=$CC
test -z "$LTCFLAGS" && LTCFLAGS=$CFLAGS
test -z "$LD" && LD=ld
test -z "$ac_objext" && ac_objext=o
_LT_CC_BASENAME([$compiler])
# Only perform the check for file, if the check method requires it
test -z "$MAGIC_CMD" && MAGIC_CMD=file
case $deplibs_check_method in
file_magic*)
if test "$file_magic_cmd" = '$MAGIC_CMD'; then
_LT_PATH_MAGIC
fi
;;
esac
# Use C for the default configuration in the libtool script
LT_SUPPORTED_TAG([CC])
_LT_LANG_C_CONFIG
_LT_LANG_DEFAULT_CONFIG
_LT_CONFIG_COMMANDS
])# _LT_SETUP
# _LT_PREPARE_SED_QUOTE_VARS
# --------------------------
# Define a few sed substitution that help us do robust quoting.
m4_defun([_LT_PREPARE_SED_QUOTE_VARS],
[# Backslashify metacharacters that are still active within
# double-quoted strings.
sed_quote_subst='s/\([["`$\\]]\)/\\\1/g'
# Same as above, but do not quote variable references.
double_quote_subst='s/\([["`\\]]\)/\\\1/g'
# Sed substitution to delay expansion of an escaped shell variable in a
# double_quote_subst'ed string.
delay_variable_subst='s/\\\\\\\\\\\$/\\\\\\$/g'
# Sed substitution to delay expansion of an escaped single quote.
delay_single_quote_subst='s/'\''/'\'\\\\\\\'\''/g'
# Sed substitution to avoid accidental globbing in evaled expressions
no_glob_subst='s/\*/\\\*/g'
])
# _LT_PROG_LTMAIN
# ---------------
# Note that this code is called both from `configure', and `config.status'
# now that we use AC_CONFIG_COMMANDS to generate libtool. Notably,
# `config.status' has no value for ac_aux_dir unless we are using Automake,
# so we pass a copy along to make sure it has a sensible value anyway.
m4_defun([_LT_PROG_LTMAIN],
[m4_ifdef([AC_REQUIRE_AUX_FILE], [AC_REQUIRE_AUX_FILE([ltmain.sh])])dnl
_LT_CONFIG_LIBTOOL_INIT([ac_aux_dir='$ac_aux_dir'])
ltmain="$ac_aux_dir/ltmain.sh"
])# _LT_PROG_LTMAIN
## ------------------------------------- ##
## Accumulate code for creating libtool. ##
## ------------------------------------- ##
# So that we can recreate a full libtool script including additional
# tags, we accumulate the chunks of code to send to AC_CONFIG_COMMANDS
# in macros and then make a single call at the end using the `libtool'
# label.
# _LT_CONFIG_LIBTOOL_INIT([INIT-COMMANDS])
# ----------------------------------------
# Register INIT-COMMANDS to be passed to AC_CONFIG_COMMANDS later.
m4_define([_LT_CONFIG_LIBTOOL_INIT],
[m4_ifval([$1],
[m4_append([_LT_OUTPUT_LIBTOOL_INIT],
[$1
])])])
# Initialize.
m4_define([_LT_OUTPUT_LIBTOOL_INIT])
# _LT_CONFIG_LIBTOOL([COMMANDS])
# ------------------------------
# Register COMMANDS to be passed to AC_CONFIG_COMMANDS later.
m4_define([_LT_CONFIG_LIBTOOL],
[m4_ifval([$1],
[m4_append([_LT_OUTPUT_LIBTOOL_COMMANDS],
[$1
])])])
# Initialize.
m4_define([_LT_OUTPUT_LIBTOOL_COMMANDS])
# _LT_CONFIG_SAVE_COMMANDS([COMMANDS], [INIT_COMMANDS])
# -----------------------------------------------------
m4_defun([_LT_CONFIG_SAVE_COMMANDS],
[_LT_CONFIG_LIBTOOL([$1])
_LT_CONFIG_LIBTOOL_INIT([$2])
])
# _LT_FORMAT_COMMENT([COMMENT])
# -----------------------------
# Add leading comment marks to the start of each line, and a trailing
# full-stop to the whole comment if one is not present already.
m4_define([_LT_FORMAT_COMMENT],
[m4_ifval([$1], [
m4_bpatsubst([m4_bpatsubst([$1], [^ *], [# ])],
[['`$\]], [\\\&])]m4_bmatch([$1], [[!?.]$], [], [.])
)])
## ------------------------ ##
## FIXME: Eliminate VARNAME ##
## ------------------------ ##
# _LT_DECL([CONFIGNAME], VARNAME, VALUE, [DESCRIPTION], [IS-TAGGED?])
# -------------------------------------------------------------------
# CONFIGNAME is the name given to the value in the libtool script.
# VARNAME is the (base) name used in the configure script.
# VALUE may be 0, 1 or 2 for a computed quote escaped value based on
# VARNAME. Any other value will be used directly.
m4_define([_LT_DECL],
[lt_if_append_uniq([lt_decl_varnames], [$2], [, ],
[lt_dict_add_subkey([lt_decl_dict], [$2], [libtool_name],
[m4_ifval([$1], [$1], [$2])])
lt_dict_add_subkey([lt_decl_dict], [$2], [value], [$3])
m4_ifval([$4],
[lt_dict_add_subkey([lt_decl_dict], [$2], [description], [$4])])
lt_dict_add_subkey([lt_decl_dict], [$2],
[tagged?], [m4_ifval([$5], [yes], [no])])])
])
# _LT_TAGDECL([CONFIGNAME], VARNAME, VALUE, [DESCRIPTION])
# --------------------------------------------------------
m4_define([_LT_TAGDECL], [_LT_DECL([$1], [$2], [$3], [$4], [yes])])
# lt_decl_tag_varnames([SEPARATOR], [VARNAME1...])
# ------------------------------------------------
m4_define([lt_decl_tag_varnames],
[_lt_decl_filter([tagged?], [yes], $@)])
# _lt_decl_filter(SUBKEY, VALUE, [SEPARATOR], [VARNAME1..])
# ---------------------------------------------------------
m4_define([_lt_decl_filter],
[m4_case([$#],
[0], [m4_fatal([$0: too few arguments: $#])],
[1], [m4_fatal([$0: too few arguments: $#: $1])],
[2], [lt_dict_filter([lt_decl_dict], [$1], [$2], [], lt_decl_varnames)],
[3], [lt_dict_filter([lt_decl_dict], [$1], [$2], [$3], lt_decl_varnames)],
[lt_dict_filter([lt_decl_dict], $@)])[]dnl
])
# lt_decl_quote_varnames([SEPARATOR], [VARNAME1...])
# --------------------------------------------------
m4_define([lt_decl_quote_varnames],
[_lt_decl_filter([value], [1], $@)])
# lt_decl_dquote_varnames([SEPARATOR], [VARNAME1...])
# ---------------------------------------------------
m4_define([lt_decl_dquote_varnames],
[_lt_decl_filter([value], [2], $@)])
# lt_decl_varnames_tagged([SEPARATOR], [VARNAME1...])
# ---------------------------------------------------
m4_define([lt_decl_varnames_tagged],
[m4_assert([$# <= 2])dnl
_$0(m4_quote(m4_default([$1], [[, ]])),
m4_ifval([$2], [[$2]], [m4_dquote(lt_decl_tag_varnames)]),
m4_split(m4_normalize(m4_quote(_LT_TAGS)), [ ]))])
m4_define([_lt_decl_varnames_tagged],
[m4_ifval([$3], [lt_combine([$1], [$2], [_], $3)])])
# lt_decl_all_varnames([SEPARATOR], [VARNAME1...])
# ------------------------------------------------
m4_define([lt_decl_all_varnames],
[_$0(m4_quote(m4_default([$1], [[, ]])),
m4_if([$2], [],
m4_quote(lt_decl_varnames),
m4_quote(m4_shift($@))))[]dnl
])
m4_define([_lt_decl_all_varnames],
[lt_join($@, lt_decl_varnames_tagged([$1],
lt_decl_tag_varnames([[, ]], m4_shift($@))))dnl
])
# _LT_CONFIG_STATUS_DECLARE([VARNAME])
# ------------------------------------
# Quote a variable value, and forward it to `config.status' so that its
# declaration there will have the same value as in `configure'. VARNAME
# must have a single quote delimited value for this to work.
m4_define([_LT_CONFIG_STATUS_DECLARE],
[$1='`$ECHO "$][$1" | $SED "$delay_single_quote_subst"`'])
# _LT_CONFIG_STATUS_DECLARATIONS
# ------------------------------
# We delimit libtool config variables with single quotes, so when
# we write them to config.status, we have to be sure to quote all
# embedded single quotes properly. In configure, this macro expands
# each variable declared with _LT_DECL (and _LT_TAGDECL) into:
#
# ='`$ECHO "$" | $SED "$delay_single_quote_subst"`'
m4_defun([_LT_CONFIG_STATUS_DECLARATIONS],
[m4_foreach([_lt_var], m4_quote(lt_decl_all_varnames),
[m4_n([_LT_CONFIG_STATUS_DECLARE(_lt_var)])])])
# _LT_LIBTOOL_TAGS
# ----------------
# Output comment and list of tags supported by the script
m4_defun([_LT_LIBTOOL_TAGS],
[_LT_FORMAT_COMMENT([The names of the tagged configurations supported by this script])dnl
available_tags="_LT_TAGS"dnl
])
# _LT_LIBTOOL_DECLARE(VARNAME, [TAG])
# -----------------------------------
# Extract the dictionary values for VARNAME (optionally with TAG) and
# expand to a commented shell variable setting:
#
# # Some comment about what VAR is for.
# visible_name=$lt_internal_name
m4_define([_LT_LIBTOOL_DECLARE],
[_LT_FORMAT_COMMENT(m4_quote(lt_dict_fetch([lt_decl_dict], [$1],
[description])))[]dnl
m4_pushdef([_libtool_name],
m4_quote(lt_dict_fetch([lt_decl_dict], [$1], [libtool_name])))[]dnl
m4_case(m4_quote(lt_dict_fetch([lt_decl_dict], [$1], [value])),
[0], [_libtool_name=[$]$1],
[1], [_libtool_name=$lt_[]$1],
[2], [_libtool_name=$lt_[]$1],
[_libtool_name=lt_dict_fetch([lt_decl_dict], [$1], [value])])[]dnl
m4_ifval([$2], [_$2])[]m4_popdef([_libtool_name])[]dnl
])
# _LT_LIBTOOL_CONFIG_VARS
# -----------------------
# Produce commented declarations of non-tagged libtool config variables
# suitable for insertion in the LIBTOOL CONFIG section of the `libtool'
# script. Tagged libtool config variables (even for the LIBTOOL CONFIG
# section) are produced by _LT_LIBTOOL_TAG_VARS.
m4_defun([_LT_LIBTOOL_CONFIG_VARS],
[m4_foreach([_lt_var],
m4_quote(_lt_decl_filter([tagged?], [no], [], lt_decl_varnames)),
[m4_n([_LT_LIBTOOL_DECLARE(_lt_var)])])])
# _LT_LIBTOOL_TAG_VARS(TAG)
# -------------------------
m4_define([_LT_LIBTOOL_TAG_VARS],
[m4_foreach([_lt_var], m4_quote(lt_decl_tag_varnames),
[m4_n([_LT_LIBTOOL_DECLARE(_lt_var, [$1])])])])
# _LT_TAGVAR(VARNAME, [TAGNAME])
# ------------------------------
m4_define([_LT_TAGVAR], [m4_ifval([$2], [$1_$2], [$1])])
# _LT_CONFIG_COMMANDS
# -------------------
# Send accumulated output to $CONFIG_STATUS. Thanks to the lists of
# variables for single and double quote escaping we saved from calls
# to _LT_DECL, we can put quote escaped variables declarations
# into `config.status', and then the shell code to quote escape them in
# for loops in `config.status'. Finally, any additional code accumulated
# from calls to _LT_CONFIG_LIBTOOL_INIT is expanded.
m4_defun([_LT_CONFIG_COMMANDS],
[AC_PROVIDE_IFELSE([LT_OUTPUT],
dnl If the libtool generation code has been placed in $CONFIG_LT,
dnl instead of duplicating it all over again into config.status,
dnl then we will have config.status run $CONFIG_LT later, so it
dnl needs to know what name is stored there:
[AC_CONFIG_COMMANDS([libtool],
[$SHELL $CONFIG_LT || AS_EXIT(1)], [CONFIG_LT='$CONFIG_LT'])],
dnl If the libtool generation code is destined for config.status,
dnl expand the accumulated commands and init code now:
[AC_CONFIG_COMMANDS([libtool],
[_LT_OUTPUT_LIBTOOL_COMMANDS], [_LT_OUTPUT_LIBTOOL_COMMANDS_INIT])])
])#_LT_CONFIG_COMMANDS
# Initialize.
m4_define([_LT_OUTPUT_LIBTOOL_COMMANDS_INIT],
[
# The HP-UX ksh and POSIX shell print the target directory to stdout
# if CDPATH is set.
(unset CDPATH) >/dev/null 2>&1 && unset CDPATH
sed_quote_subst='$sed_quote_subst'
double_quote_subst='$double_quote_subst'
delay_variable_subst='$delay_variable_subst'
_LT_CONFIG_STATUS_DECLARATIONS
LTCC='$LTCC'
LTCFLAGS='$LTCFLAGS'
compiler='$compiler_DEFAULT'
# A function that is used when there is no print builtin or printf.
func_fallback_echo ()
{
eval 'cat <<_LTECHO_EOF
\$[]1
_LTECHO_EOF'
}
# Quote evaled strings.
for var in lt_decl_all_varnames([[ \
]], lt_decl_quote_varnames); do
case \`eval \\\\\$ECHO \\\\""\\\\\$\$var"\\\\"\` in
*[[\\\\\\\`\\"\\\$]]*)
eval "lt_\$var=\\\\\\"\\\`\\\$ECHO \\"\\\$\$var\\" | \\\$SED \\"\\\$sed_quote_subst\\"\\\`\\\\\\""
;;
*)
eval "lt_\$var=\\\\\\"\\\$\$var\\\\\\""
;;
esac
done
# Double-quote double-evaled strings.
for var in lt_decl_all_varnames([[ \
]], lt_decl_dquote_varnames); do
case \`eval \\\\\$ECHO \\\\""\\\\\$\$var"\\\\"\` in
*[[\\\\\\\`\\"\\\$]]*)
eval "lt_\$var=\\\\\\"\\\`\\\$ECHO \\"\\\$\$var\\" | \\\$SED -e \\"\\\$double_quote_subst\\" -e \\"\\\$sed_quote_subst\\" -e \\"\\\$delay_variable_subst\\"\\\`\\\\\\""
;;
*)
eval "lt_\$var=\\\\\\"\\\$\$var\\\\\\""
;;
esac
done
_LT_OUTPUT_LIBTOOL_INIT
])
# _LT_GENERATED_FILE_INIT(FILE, [COMMENT])
# ------------------------------------
# Generate a child script FILE with all initialization necessary to
# reuse the environment learned by the parent script, and make the
# file executable. If COMMENT is supplied, it is inserted after the
# `#!' sequence but before initialization text begins. After this
# macro, additional text can be appended to FILE to form the body of
# the child script. The macro ends with non-zero status if the
# file could not be fully written (such as if the disk is full).
m4_ifdef([AS_INIT_GENERATED],
[m4_defun([_LT_GENERATED_FILE_INIT],[AS_INIT_GENERATED($@)])],
[m4_defun([_LT_GENERATED_FILE_INIT],
[m4_require([AS_PREPARE])]dnl
[m4_pushdef([AS_MESSAGE_LOG_FD])]dnl
[lt_write_fail=0
cat >$1 <<_ASEOF || lt_write_fail=1
#! $SHELL
# Generated by $as_me.
$2
SHELL=\${CONFIG_SHELL-$SHELL}
export SHELL
_ASEOF
cat >>$1 <<\_ASEOF || lt_write_fail=1
AS_SHELL_SANITIZE
_AS_PREPARE
exec AS_MESSAGE_FD>&1
_ASEOF
test $lt_write_fail = 0 && chmod +x $1[]dnl
m4_popdef([AS_MESSAGE_LOG_FD])])])# _LT_GENERATED_FILE_INIT
# LT_OUTPUT
# ---------
# This macro allows early generation of the libtool script (before
# AC_OUTPUT is called), incase it is used in configure for compilation
# tests.
AC_DEFUN([LT_OUTPUT],
[: ${CONFIG_LT=./config.lt}
AC_MSG_NOTICE([creating $CONFIG_LT])
_LT_GENERATED_FILE_INIT(["$CONFIG_LT"],
[# Run this file to recreate a libtool stub with the current configuration.])
cat >>"$CONFIG_LT" <<\_LTEOF
lt_cl_silent=false
exec AS_MESSAGE_LOG_FD>>config.log
{
echo
AS_BOX([Running $as_me.])
} >&AS_MESSAGE_LOG_FD
lt_cl_help="\
\`$as_me' creates a local libtool stub from the current configuration,
for use in further configure time tests before the real libtool is
generated.
Usage: $[0] [[OPTIONS]]
-h, --help print this help, then exit
-V, --version print version number, then exit
-q, --quiet do not print progress messages
-d, --debug don't remove temporary files
Report bugs to ."
lt_cl_version="\
m4_ifset([AC_PACKAGE_NAME], [AC_PACKAGE_NAME ])config.lt[]dnl
m4_ifset([AC_PACKAGE_VERSION], [ AC_PACKAGE_VERSION])
configured by $[0], generated by m4_PACKAGE_STRING.
Copyright (C) 2011 Free Software Foundation, Inc.
This config.lt script is free software; the Free Software Foundation
gives unlimited permision to copy, distribute and modify it."
while test $[#] != 0
do
case $[1] in
--version | --v* | -V )
echo "$lt_cl_version"; exit 0 ;;
--help | --h* | -h )
echo "$lt_cl_help"; exit 0 ;;
--debug | --d* | -d )
debug=: ;;
--quiet | --q* | --silent | --s* | -q )
lt_cl_silent=: ;;
-*) AC_MSG_ERROR([unrecognized option: $[1]
Try \`$[0] --help' for more information.]) ;;
*) AC_MSG_ERROR([unrecognized argument: $[1]
Try \`$[0] --help' for more information.]) ;;
esac
shift
done
if $lt_cl_silent; then
exec AS_MESSAGE_FD>/dev/null
fi
_LTEOF
cat >>"$CONFIG_LT" <<_LTEOF
_LT_OUTPUT_LIBTOOL_COMMANDS_INIT
_LTEOF
cat >>"$CONFIG_LT" <<\_LTEOF
AC_MSG_NOTICE([creating $ofile])
_LT_OUTPUT_LIBTOOL_COMMANDS
AS_EXIT(0)
_LTEOF
chmod +x "$CONFIG_LT"
# configure is writing to config.log, but config.lt does its own redirection,
# appending to config.log, which fails on DOS, as config.log is still kept
# open by configure. Here we exec the FD to /dev/null, effectively closing
# config.log, so it can be properly (re)opened and appended to by config.lt.
lt_cl_success=:
test "$silent" = yes &&
lt_config_lt_args="$lt_config_lt_args --quiet"
exec AS_MESSAGE_LOG_FD>/dev/null
$SHELL "$CONFIG_LT" $lt_config_lt_args || lt_cl_success=false
exec AS_MESSAGE_LOG_FD>>config.log
$lt_cl_success || AS_EXIT(1)
])# LT_OUTPUT
# _LT_CONFIG(TAG)
# ---------------
# If TAG is the built-in tag, create an initial libtool script with a
# default configuration from the untagged config vars. Otherwise add code
# to config.status for appending the configuration named by TAG from the
# matching tagged config vars.
m4_defun([_LT_CONFIG],
[m4_require([_LT_FILEUTILS_DEFAULTS])dnl
_LT_CONFIG_SAVE_COMMANDS([
m4_define([_LT_TAG], m4_if([$1], [], [C], [$1]))dnl
m4_if(_LT_TAG, [C], [
# See if we are running on zsh, and set the options which allow our
# commands through without removal of \ escapes.
if test -n "${ZSH_VERSION+set}" ; then
setopt NO_GLOB_SUBST
fi
cfgfile="${ofile}T"
trap "$RM \"$cfgfile\"; exit 1" 1 2 15
$RM "$cfgfile"
cat <<_LT_EOF >> "$cfgfile"
#! $SHELL
# `$ECHO "$ofile" | sed 's%^.*/%%'` - Provide generalized library-building support services.
# Generated automatically by $as_me ($PACKAGE$TIMESTAMP) $VERSION
# Libtool was configured on host `(hostname || uname -n) 2>/dev/null | sed 1q`:
# NOTE: Changes made to this file will be lost: look at ltmain.sh.
#
_LT_COPYING
_LT_LIBTOOL_TAGS
# ### BEGIN LIBTOOL CONFIG
_LT_LIBTOOL_CONFIG_VARS
_LT_LIBTOOL_TAG_VARS
# ### END LIBTOOL CONFIG
_LT_EOF
case $host_os in
aix3*)
cat <<\_LT_EOF >> "$cfgfile"
# AIX sometimes has problems with the GCC collect2 program. For some
# reason, if we set the COLLECT_NAMES environment variable, the problems
# vanish in a puff of smoke.
if test "X${COLLECT_NAMES+set}" != Xset; then
COLLECT_NAMES=
export COLLECT_NAMES
fi
_LT_EOF
;;
esac
_LT_PROG_LTMAIN
# We use sed instead of cat because bash on DJGPP gets confused if
# if finds mixed CR/LF and LF-only lines. Since sed operates in
# text mode, it properly converts lines to CR/LF. This bash problem
# is reportedly fixed, but why not run on old versions too?
sed '$q' "$ltmain" >> "$cfgfile" \
|| (rm -f "$cfgfile"; exit 1)
_LT_PROG_REPLACE_SHELLFNS
mv -f "$cfgfile" "$ofile" ||
(rm -f "$ofile" && cp "$cfgfile" "$ofile" && rm -f "$cfgfile")
chmod +x "$ofile"
],
[cat <<_LT_EOF >> "$ofile"
dnl Unfortunately we have to use $1 here, since _LT_TAG is not expanded
dnl in a comment (ie after a #).
# ### BEGIN LIBTOOL TAG CONFIG: $1
_LT_LIBTOOL_TAG_VARS(_LT_TAG)
# ### END LIBTOOL TAG CONFIG: $1
_LT_EOF
])dnl /m4_if
],
[m4_if([$1], [], [
PACKAGE='$PACKAGE'
VERSION='$VERSION'
TIMESTAMP='$TIMESTAMP'
RM='$RM'
ofile='$ofile'], [])
])dnl /_LT_CONFIG_SAVE_COMMANDS
])# _LT_CONFIG
# LT_SUPPORTED_TAG(TAG)
# ---------------------
# Trace this macro to discover what tags are supported by the libtool
# --tag option, using:
# autoconf --trace 'LT_SUPPORTED_TAG:$1'
AC_DEFUN([LT_SUPPORTED_TAG], [])
# C support is built-in for now
m4_define([_LT_LANG_C_enabled], [])
m4_define([_LT_TAGS], [])
# LT_LANG(LANG)
# -------------
# Enable libtool support for the given language if not already enabled.
AC_DEFUN([LT_LANG],
[AC_BEFORE([$0], [LT_OUTPUT])dnl
m4_case([$1],
[C], [_LT_LANG(C)],
[C++], [_LT_LANG(CXX)],
[Go], [_LT_LANG(GO)],
[Java], [_LT_LANG(GCJ)],
[Fortran 77], [_LT_LANG(F77)],
[Fortran], [_LT_LANG(FC)],
[Windows Resource], [_LT_LANG(RC)],
[m4_ifdef([_LT_LANG_]$1[_CONFIG],
[_LT_LANG($1)],
[m4_fatal([$0: unsupported language: "$1"])])])dnl
])# LT_LANG
# _LT_LANG(LANGNAME)
# ------------------
m4_defun([_LT_LANG],
[m4_ifdef([_LT_LANG_]$1[_enabled], [],
[LT_SUPPORTED_TAG([$1])dnl
m4_append([_LT_TAGS], [$1 ])dnl
m4_define([_LT_LANG_]$1[_enabled], [])dnl
_LT_LANG_$1_CONFIG($1)])dnl
])# _LT_LANG
m4_ifndef([AC_PROG_GO], [
############################################################
# NOTE: This macro has been submitted for inclusion into #
# GNU Autoconf as AC_PROG_GO. When it is available in #
# a released version of Autoconf we should remove this #
# macro and use it instead. #
############################################################
m4_defun([AC_PROG_GO],
[AC_LANG_PUSH(Go)dnl
AC_ARG_VAR([GOC], [Go compiler command])dnl
AC_ARG_VAR([GOFLAGS], [Go compiler flags])dnl
_AC_ARG_VAR_LDFLAGS()dnl
AC_CHECK_TOOL(GOC, gccgo)
if test -z "$GOC"; then
if test -n "$ac_tool_prefix"; then
AC_CHECK_PROG(GOC, [${ac_tool_prefix}gccgo], [${ac_tool_prefix}gccgo])
fi
fi
if test -z "$GOC"; then
AC_CHECK_PROG(GOC, gccgo, gccgo, false)
fi
])#m4_defun
])#m4_ifndef
# _LT_LANG_DEFAULT_CONFIG
# -----------------------
m4_defun([_LT_LANG_DEFAULT_CONFIG],
[AC_PROVIDE_IFELSE([AC_PROG_CXX],
[LT_LANG(CXX)],
[m4_define([AC_PROG_CXX], defn([AC_PROG_CXX])[LT_LANG(CXX)])])
AC_PROVIDE_IFELSE([AC_PROG_F77],
[LT_LANG(F77)],
[m4_define([AC_PROG_F77], defn([AC_PROG_F77])[LT_LANG(F77)])])
AC_PROVIDE_IFELSE([AC_PROG_FC],
[LT_LANG(FC)],
[m4_define([AC_PROG_FC], defn([AC_PROG_FC])[LT_LANG(FC)])])
dnl The call to [A][M_PROG_GCJ] is quoted like that to stop aclocal
dnl pulling things in needlessly.
AC_PROVIDE_IFELSE([AC_PROG_GCJ],
[LT_LANG(GCJ)],
[AC_PROVIDE_IFELSE([A][M_PROG_GCJ],
[LT_LANG(GCJ)],
[AC_PROVIDE_IFELSE([LT_PROG_GCJ],
[LT_LANG(GCJ)],
[m4_ifdef([AC_PROG_GCJ],
[m4_define([AC_PROG_GCJ], defn([AC_PROG_GCJ])[LT_LANG(GCJ)])])
m4_ifdef([A][M_PROG_GCJ],
[m4_define([A][M_PROG_GCJ], defn([A][M_PROG_GCJ])[LT_LANG(GCJ)])])
m4_ifdef([LT_PROG_GCJ],
[m4_define([LT_PROG_GCJ], defn([LT_PROG_GCJ])[LT_LANG(GCJ)])])])])])
AC_PROVIDE_IFELSE([AC_PROG_GO],
[LT_LANG(GO)],
[m4_define([AC_PROG_GO], defn([AC_PROG_GO])[LT_LANG(GO)])])
AC_PROVIDE_IFELSE([LT_PROG_RC],
[LT_LANG(RC)],
[m4_define([LT_PROG_RC], defn([LT_PROG_RC])[LT_LANG(RC)])])
])# _LT_LANG_DEFAULT_CONFIG
# Obsolete macros:
AU_DEFUN([AC_LIBTOOL_CXX], [LT_LANG(C++)])
AU_DEFUN([AC_LIBTOOL_F77], [LT_LANG(Fortran 77)])
AU_DEFUN([AC_LIBTOOL_FC], [LT_LANG(Fortran)])
AU_DEFUN([AC_LIBTOOL_GCJ], [LT_LANG(Java)])
AU_DEFUN([AC_LIBTOOL_RC], [LT_LANG(Windows Resource)])
dnl aclocal-1.4 backwards compatibility:
dnl AC_DEFUN([AC_LIBTOOL_CXX], [])
dnl AC_DEFUN([AC_LIBTOOL_F77], [])
dnl AC_DEFUN([AC_LIBTOOL_FC], [])
dnl AC_DEFUN([AC_LIBTOOL_GCJ], [])
dnl AC_DEFUN([AC_LIBTOOL_RC], [])
# _LT_TAG_COMPILER
# ----------------
m4_defun([_LT_TAG_COMPILER],
[AC_REQUIRE([AC_PROG_CC])dnl
_LT_DECL([LTCC], [CC], [1], [A C compiler])dnl
_LT_DECL([LTCFLAGS], [CFLAGS], [1], [LTCC compiler flags])dnl
_LT_TAGDECL([CC], [compiler], [1], [A language specific compiler])dnl
_LT_TAGDECL([with_gcc], [GCC], [0], [Is the compiler the GNU compiler?])dnl
# If no C compiler was specified, use CC.
LTCC=${LTCC-"$CC"}
# If no C compiler flags were specified, use CFLAGS.
LTCFLAGS=${LTCFLAGS-"$CFLAGS"}
# Allow CC to be a program name with arguments.
compiler=$CC
])# _LT_TAG_COMPILER
# _LT_COMPILER_BOILERPLATE
# ------------------------
# Check for compiler boilerplate output or warnings with
# the simple compiler test code.
m4_defun([_LT_COMPILER_BOILERPLATE],
[m4_require([_LT_DECL_SED])dnl
ac_outfile=conftest.$ac_objext
echo "$lt_simple_compile_test_code" >conftest.$ac_ext
eval "$ac_compile" 2>&1 >/dev/null | $SED '/^$/d; /^ *+/d' >conftest.err
_lt_compiler_boilerplate=`cat conftest.err`
$RM conftest*
])# _LT_COMPILER_BOILERPLATE
# _LT_LINKER_BOILERPLATE
# ----------------------
# Check for linker boilerplate output or warnings with
# the simple link test code.
m4_defun([_LT_LINKER_BOILERPLATE],
[m4_require([_LT_DECL_SED])dnl
ac_outfile=conftest.$ac_objext
echo "$lt_simple_link_test_code" >conftest.$ac_ext
eval "$ac_link" 2>&1 >/dev/null | $SED '/^$/d; /^ *+/d' >conftest.err
_lt_linker_boilerplate=`cat conftest.err`
$RM -r conftest*
])# _LT_LINKER_BOILERPLATE
# _LT_REQUIRED_DARWIN_CHECKS
# -------------------------
m4_defun_once([_LT_REQUIRED_DARWIN_CHECKS],[
case $host_os in
rhapsody* | darwin*)
AC_CHECK_TOOL([DSYMUTIL], [dsymutil], [:])
AC_CHECK_TOOL([NMEDIT], [nmedit], [:])
AC_CHECK_TOOL([LIPO], [lipo], [:])
AC_CHECK_TOOL([OTOOL], [otool], [:])
AC_CHECK_TOOL([OTOOL64], [otool64], [:])
_LT_DECL([], [DSYMUTIL], [1],
[Tool to manipulate archived DWARF debug symbol files on Mac OS X])
_LT_DECL([], [NMEDIT], [1],
[Tool to change global to local symbols on Mac OS X])
_LT_DECL([], [LIPO], [1],
[Tool to manipulate fat objects and archives on Mac OS X])
_LT_DECL([], [OTOOL], [1],
[ldd/readelf like tool for Mach-O binaries on Mac OS X])
_LT_DECL([], [OTOOL64], [1],
[ldd/readelf like tool for 64 bit Mach-O binaries on Mac OS X 10.4])
AC_CACHE_CHECK([for -single_module linker flag],[lt_cv_apple_cc_single_mod],
[lt_cv_apple_cc_single_mod=no
if test -z "${LT_MULTI_MODULE}"; then
# By default we will add the -single_module flag. You can override
# by either setting the environment variable LT_MULTI_MODULE
# non-empty at configure time, or by adding -multi_module to the
# link flags.
rm -rf libconftest.dylib*
echo "int foo(void){return 1;}" > conftest.c
echo "$LTCC $LTCFLAGS $LDFLAGS -o libconftest.dylib \
-dynamiclib -Wl,-single_module conftest.c" >&AS_MESSAGE_LOG_FD
$LTCC $LTCFLAGS $LDFLAGS -o libconftest.dylib \
-dynamiclib -Wl,-single_module conftest.c 2>conftest.err
_lt_result=$?
# If there is a non-empty error log, and "single_module"
# appears in it, assume the flag caused a linker warning
if test -s conftest.err && $GREP single_module conftest.err; then
cat conftest.err >&AS_MESSAGE_LOG_FD
# Otherwise, if the output was created with a 0 exit code from
# the compiler, it worked.
elif test -f libconftest.dylib && test $_lt_result -eq 0; then
lt_cv_apple_cc_single_mod=yes
else
cat conftest.err >&AS_MESSAGE_LOG_FD
fi
rm -rf libconftest.dylib*
rm -f conftest.*
fi])
AC_CACHE_CHECK([for -exported_symbols_list linker flag],
[lt_cv_ld_exported_symbols_list],
[lt_cv_ld_exported_symbols_list=no
save_LDFLAGS=$LDFLAGS
echo "_main" > conftest.sym
LDFLAGS="$LDFLAGS -Wl,-exported_symbols_list,conftest.sym"
AC_LINK_IFELSE([AC_LANG_PROGRAM([],[])],
[lt_cv_ld_exported_symbols_list=yes],
[lt_cv_ld_exported_symbols_list=no])
LDFLAGS="$save_LDFLAGS"
])
AC_CACHE_CHECK([for -force_load linker flag],[lt_cv_ld_force_load],
[lt_cv_ld_force_load=no
cat > conftest.c << _LT_EOF
int forced_loaded() { return 2;}
_LT_EOF
echo "$LTCC $LTCFLAGS -c -o conftest.o conftest.c" >&AS_MESSAGE_LOG_FD
$LTCC $LTCFLAGS -c -o conftest.o conftest.c 2>&AS_MESSAGE_LOG_FD
echo "$AR cru libconftest.a conftest.o" >&AS_MESSAGE_LOG_FD
$AR cru libconftest.a conftest.o 2>&AS_MESSAGE_LOG_FD
echo "$RANLIB libconftest.a" >&AS_MESSAGE_LOG_FD
$RANLIB libconftest.a 2>&AS_MESSAGE_LOG_FD
cat > conftest.c << _LT_EOF
int main() { return 0;}
_LT_EOF
echo "$LTCC $LTCFLAGS $LDFLAGS -o conftest conftest.c -Wl,-force_load,./libconftest.a" >&AS_MESSAGE_LOG_FD
$LTCC $LTCFLAGS $LDFLAGS -o conftest conftest.c -Wl,-force_load,./libconftest.a 2>conftest.err
_lt_result=$?
if test -s conftest.err && $GREP force_load conftest.err; then
cat conftest.err >&AS_MESSAGE_LOG_FD
elif test -f conftest && test $_lt_result -eq 0 && $GREP forced_load conftest >/dev/null 2>&1 ; then
lt_cv_ld_force_load=yes
else
cat conftest.err >&AS_MESSAGE_LOG_FD
fi
rm -f conftest.err libconftest.a conftest conftest.c
rm -rf conftest.dSYM
])
case $host_os in
rhapsody* | darwin1.[[012]])
_lt_dar_allow_undefined='${wl}-undefined ${wl}suppress' ;;
darwin1.*)
_lt_dar_allow_undefined='${wl}-flat_namespace ${wl}-undefined ${wl}suppress' ;;
darwin*) # darwin 5.x on
# if running on 10.5 or later, the deployment target defaults
# to the OS version, if on x86, and 10.4, the deployment
# target defaults to 10.4. Don't you love it?
case ${MACOSX_DEPLOYMENT_TARGET-10.0},$host in
10.0,*86*-darwin8*|10.0,*-darwin[[91]]*)
_lt_dar_allow_undefined='${wl}-undefined ${wl}dynamic_lookup' ;;
10.[[012]]*)
_lt_dar_allow_undefined='${wl}-flat_namespace ${wl}-undefined ${wl}suppress' ;;
10.*)
_lt_dar_allow_undefined='${wl}-undefined ${wl}dynamic_lookup' ;;
esac
;;
esac
if test "$lt_cv_apple_cc_single_mod" = "yes"; then
_lt_dar_single_mod='$single_module'
fi
if test "$lt_cv_ld_exported_symbols_list" = "yes"; then
_lt_dar_export_syms=' ${wl}-exported_symbols_list,$output_objdir/${libname}-symbols.expsym'
else
_lt_dar_export_syms='~$NMEDIT -s $output_objdir/${libname}-symbols.expsym ${lib}'
fi
if test "$DSYMUTIL" != ":" && test "$lt_cv_ld_force_load" = "no"; then
_lt_dsymutil='~$DSYMUTIL $lib || :'
else
_lt_dsymutil=
fi
;;
esac
])
# _LT_DARWIN_LINKER_FEATURES([TAG])
# ---------------------------------
# Checks for linker and compiler features on darwin
m4_defun([_LT_DARWIN_LINKER_FEATURES],
[
m4_require([_LT_REQUIRED_DARWIN_CHECKS])
_LT_TAGVAR(archive_cmds_need_lc, $1)=no
_LT_TAGVAR(hardcode_direct, $1)=no
_LT_TAGVAR(hardcode_automatic, $1)=yes
_LT_TAGVAR(hardcode_shlibpath_var, $1)=unsupported
if test "$lt_cv_ld_force_load" = "yes"; then
_LT_TAGVAR(whole_archive_flag_spec, $1)='`for conv in $convenience\"\"; do test -n \"$conv\" && new_convenience=\"$new_convenience ${wl}-force_load,$conv\"; done; func_echo_all \"$new_convenience\"`'
m4_case([$1], [F77], [_LT_TAGVAR(compiler_needs_object, $1)=yes],
[FC], [_LT_TAGVAR(compiler_needs_object, $1)=yes])
else
_LT_TAGVAR(whole_archive_flag_spec, $1)=''
fi
_LT_TAGVAR(link_all_deplibs, $1)=yes
_LT_TAGVAR(allow_undefined_flag, $1)="$_lt_dar_allow_undefined"
case $cc_basename in
ifort*) _lt_dar_can_shared=yes ;;
*) _lt_dar_can_shared=$GCC ;;
esac
if test "$_lt_dar_can_shared" = "yes"; then
output_verbose_link_cmd=func_echo_all
_LT_TAGVAR(archive_cmds, $1)="\$CC -dynamiclib \$allow_undefined_flag -o \$lib \$libobjs \$deplibs \$compiler_flags -install_name \$rpath/\$soname \$verstring $_lt_dar_single_mod${_lt_dsymutil}"
_LT_TAGVAR(module_cmds, $1)="\$CC \$allow_undefined_flag -o \$lib -bundle \$libobjs \$deplibs \$compiler_flags${_lt_dsymutil}"
_LT_TAGVAR(archive_expsym_cmds, $1)="sed 's,^,_,' < \$export_symbols > \$output_objdir/\${libname}-symbols.expsym~\$CC -dynamiclib \$allow_undefined_flag -o \$lib \$libobjs \$deplibs \$compiler_flags -install_name \$rpath/\$soname \$verstring ${_lt_dar_single_mod}${_lt_dar_export_syms}${_lt_dsymutil}"
_LT_TAGVAR(module_expsym_cmds, $1)="sed -e 's,^,_,' < \$export_symbols > \$output_objdir/\${libname}-symbols.expsym~\$CC \$allow_undefined_flag -o \$lib -bundle \$libobjs \$deplibs \$compiler_flags${_lt_dar_export_syms}${_lt_dsymutil}"
m4_if([$1], [CXX],
[ if test "$lt_cv_apple_cc_single_mod" != "yes"; then
_LT_TAGVAR(archive_cmds, $1)="\$CC -r -keep_private_externs -nostdlib -o \${lib}-master.o \$libobjs~\$CC -dynamiclib \$allow_undefined_flag -o \$lib \${lib}-master.o \$deplibs \$compiler_flags -install_name \$rpath/\$soname \$verstring${_lt_dsymutil}"
_LT_TAGVAR(archive_expsym_cmds, $1)="sed 's,^,_,' < \$export_symbols > \$output_objdir/\${libname}-symbols.expsym~\$CC -r -keep_private_externs -nostdlib -o \${lib}-master.o \$libobjs~\$CC -dynamiclib \$allow_undefined_flag -o \$lib \${lib}-master.o \$deplibs \$compiler_flags -install_name \$rpath/\$soname \$verstring${_lt_dar_export_syms}${_lt_dsymutil}"
fi
],[])
else
_LT_TAGVAR(ld_shlibs, $1)=no
fi
])
# _LT_SYS_MODULE_PATH_AIX([TAGNAME])
# ----------------------------------
# Links a minimal program and checks the executable
# for the system default hardcoded library path. In most cases,
# this is /usr/lib:/lib, but when the MPI compilers are used
# the location of the communication and MPI libs are included too.
# If we don't find anything, use the default library path according
# to the aix ld manual.
# Store the results from the different compilers for each TAGNAME.
# Allow to override them for all tags through lt_cv_aix_libpath.
m4_defun([_LT_SYS_MODULE_PATH_AIX],
[m4_require([_LT_DECL_SED])dnl
if test "${lt_cv_aix_libpath+set}" = set; then
aix_libpath=$lt_cv_aix_libpath
else
AC_CACHE_VAL([_LT_TAGVAR([lt_cv_aix_libpath_], [$1])],
[AC_LINK_IFELSE([AC_LANG_PROGRAM],[
lt_aix_libpath_sed='[
/Import File Strings/,/^$/ {
/^0/ {
s/^0 *\([^ ]*\) *$/\1/
p
}
}]'
_LT_TAGVAR([lt_cv_aix_libpath_], [$1])=`dump -H conftest$ac_exeext 2>/dev/null | $SED -n -e "$lt_aix_libpath_sed"`
# Check for a 64-bit object if we didn't find anything.
if test -z "$_LT_TAGVAR([lt_cv_aix_libpath_], [$1])"; then
_LT_TAGVAR([lt_cv_aix_libpath_], [$1])=`dump -HX64 conftest$ac_exeext 2>/dev/null | $SED -n -e "$lt_aix_libpath_sed"`
fi],[])
if test -z "$_LT_TAGVAR([lt_cv_aix_libpath_], [$1])"; then
_LT_TAGVAR([lt_cv_aix_libpath_], [$1])="/usr/lib:/lib"
fi
])
aix_libpath=$_LT_TAGVAR([lt_cv_aix_libpath_], [$1])
fi
])# _LT_SYS_MODULE_PATH_AIX
# _LT_SHELL_INIT(ARG)
# -------------------
m4_define([_LT_SHELL_INIT],
[m4_divert_text([M4SH-INIT], [$1
])])# _LT_SHELL_INIT
# _LT_PROG_ECHO_BACKSLASH
# -----------------------
# Find how we can fake an echo command that does not interpret backslash.
# In particular, with Autoconf 2.60 or later we add some code to the start
# of the generated configure script which will find a shell with a builtin
# printf (which we can use as an echo command).
m4_defun([_LT_PROG_ECHO_BACKSLASH],
[ECHO='\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\'
ECHO=$ECHO$ECHO$ECHO$ECHO$ECHO
ECHO=$ECHO$ECHO$ECHO$ECHO$ECHO$ECHO
AC_MSG_CHECKING([how to print strings])
# Test print first, because it will be a builtin if present.
if test "X`( print -r -- -n ) 2>/dev/null`" = X-n && \
test "X`print -r -- $ECHO 2>/dev/null`" = "X$ECHO"; then
ECHO='print -r --'
elif test "X`printf %s $ECHO 2>/dev/null`" = "X$ECHO"; then
ECHO='printf %s\n'
else
# Use this function as a fallback that always works.
func_fallback_echo ()
{
eval 'cat <<_LTECHO_EOF
$[]1
_LTECHO_EOF'
}
ECHO='func_fallback_echo'
fi
# func_echo_all arg...
# Invoke $ECHO with all args, space-separated.
func_echo_all ()
{
$ECHO "$*"
}
case "$ECHO" in
printf*) AC_MSG_RESULT([printf]) ;;
print*) AC_MSG_RESULT([print -r]) ;;
*) AC_MSG_RESULT([cat]) ;;
esac
m4_ifdef([_AS_DETECT_SUGGESTED],
[_AS_DETECT_SUGGESTED([
test -n "${ZSH_VERSION+set}${BASH_VERSION+set}" || (
ECHO='\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\'
ECHO=$ECHO$ECHO$ECHO$ECHO$ECHO
ECHO=$ECHO$ECHO$ECHO$ECHO$ECHO$ECHO
PATH=/empty FPATH=/empty; export PATH FPATH
test "X`printf %s $ECHO`" = "X$ECHO" \
|| test "X`print -r -- $ECHO`" = "X$ECHO" )])])
_LT_DECL([], [SHELL], [1], [Shell to use when invoking shell scripts])
_LT_DECL([], [ECHO], [1], [An echo program that protects backslashes])
])# _LT_PROG_ECHO_BACKSLASH
# _LT_WITH_SYSROOT
# ----------------
AC_DEFUN([_LT_WITH_SYSROOT],
[AC_MSG_CHECKING([for sysroot])
AC_ARG_WITH([sysroot],
[ --with-sysroot[=DIR] Search for dependent libraries within DIR
(or the compiler's sysroot if not specified).],
[], [with_sysroot=no])
dnl lt_sysroot will always be passed unquoted. We quote it here
dnl in case the user passed a directory name.
lt_sysroot=
case ${with_sysroot} in #(
yes)
if test "$GCC" = yes; then
lt_sysroot=`$CC --print-sysroot 2>/dev/null`
fi
;; #(
/*)
lt_sysroot=`echo "$with_sysroot" | sed -e "$sed_quote_subst"`
;; #(
no|'')
;; #(
*)
AC_MSG_RESULT([${with_sysroot}])
AC_MSG_ERROR([The sysroot must be an absolute path.])
;;
esac
AC_MSG_RESULT([${lt_sysroot:-no}])
_LT_DECL([], [lt_sysroot], [0], [The root where to search for ]dnl
[dependent libraries, and in which our libraries should be installed.])])
# _LT_ENABLE_LOCK
# ---------------
m4_defun([_LT_ENABLE_LOCK],
[AC_ARG_ENABLE([libtool-lock],
[AS_HELP_STRING([--disable-libtool-lock],
[avoid locking (might break parallel builds)])])
test "x$enable_libtool_lock" != xno && enable_libtool_lock=yes
# Some flags need to be propagated to the compiler or linker for good
# libtool support.
case $host in
ia64-*-hpux*)
# Find out which ABI we are using.
echo 'int i;' > conftest.$ac_ext
if AC_TRY_EVAL(ac_compile); then
case `/usr/bin/file conftest.$ac_objext` in
*ELF-32*)
HPUX_IA64_MODE="32"
;;
*ELF-64*)
HPUX_IA64_MODE="64"
;;
esac
fi
rm -rf conftest*
;;
*-*-irix6*)
# Find out which ABI we are using.
echo '[#]line '$LINENO' "configure"' > conftest.$ac_ext
if AC_TRY_EVAL(ac_compile); then
if test "$lt_cv_prog_gnu_ld" = yes; then
case `/usr/bin/file conftest.$ac_objext` in
*32-bit*)
LD="${LD-ld} -melf32bsmip"
;;
*N32*)
LD="${LD-ld} -melf32bmipn32"
;;
*64-bit*)
LD="${LD-ld} -melf64bmip"
;;
esac
else
case `/usr/bin/file conftest.$ac_objext` in
*32-bit*)
LD="${LD-ld} -32"
;;
*N32*)
LD="${LD-ld} -n32"
;;
*64-bit*)
LD="${LD-ld} -64"
;;
esac
fi
fi
rm -rf conftest*
;;
x86_64-*kfreebsd*-gnu|x86_64-*linux*|ppc*-*linux*|powerpc*-*linux*| \
s390*-*linux*|s390*-*tpf*|sparc*-*linux*)
# Find out which ABI we are using.
echo 'int i;' > conftest.$ac_ext
if AC_TRY_EVAL(ac_compile); then
case `/usr/bin/file conftest.o` in
*32-bit*)
case $host in
x86_64-*kfreebsd*-gnu)
LD="${LD-ld} -m elf_i386_fbsd"
;;
x86_64-*linux*)
LD="${LD-ld} -m elf_i386"
;;
powerpc64le-*)
LD="${LD-ld} -m elf32lppclinux"
;;
powerpc64-*)
LD="${LD-ld} -m elf32ppclinux"
;;
s390x-*linux*)
LD="${LD-ld} -m elf_s390"
;;
sparc64-*linux*)
LD="${LD-ld} -m elf32_sparc"
;;
esac
;;
*64-bit*)
case $host in
x86_64-*kfreebsd*-gnu)
LD="${LD-ld} -m elf_x86_64_fbsd"
;;
x86_64-*linux*)
LD="${LD-ld} -m elf_x86_64"
;;
powerpcle-*)
LD="${LD-ld} -m elf64lppc"
;;
powerpc-*)
LD="${LD-ld} -m elf64ppc"
;;
s390*-*linux*|s390*-*tpf*)
LD="${LD-ld} -m elf64_s390"
;;
sparc*-*linux*)
LD="${LD-ld} -m elf64_sparc"
;;
esac
;;
esac
fi
rm -rf conftest*
;;
*-*-sco3.2v5*)
# On SCO OpenServer 5, we need -belf to get full-featured binaries.
SAVE_CFLAGS="$CFLAGS"
CFLAGS="$CFLAGS -belf"
AC_CACHE_CHECK([whether the C compiler needs -belf], lt_cv_cc_needs_belf,
[AC_LANG_PUSH(C)
AC_LINK_IFELSE([AC_LANG_PROGRAM([[]],[[]])],[lt_cv_cc_needs_belf=yes],[lt_cv_cc_needs_belf=no])
AC_LANG_POP])
if test x"$lt_cv_cc_needs_belf" != x"yes"; then
# this is probably gcc 2.8.0, egcs 1.0 or newer; no need for -belf
CFLAGS="$SAVE_CFLAGS"
fi
;;
*-*solaris*)
# Find out which ABI we are using.
echo 'int i;' > conftest.$ac_ext
if AC_TRY_EVAL(ac_compile); then
case `/usr/bin/file conftest.o` in
*64-bit*)
case $lt_cv_prog_gnu_ld in
yes*)
case $host in
i?86-*-solaris*)
LD="${LD-ld} -m elf_x86_64"
;;
sparc*-*-solaris*)
LD="${LD-ld} -m elf64_sparc"
;;
esac
# GNU ld 2.21 introduced _sol2 emulations. Use them if available.
if ${LD-ld} -V | grep _sol2 >/dev/null 2>&1; then
LD="${LD-ld}_sol2"
fi
;;
*)
if ${LD-ld} -64 -r -o conftest2.o conftest.o >/dev/null 2>&1; then
LD="${LD-ld} -64"
fi
;;
esac
;;
esac
fi
rm -rf conftest*
;;
esac
need_locks="$enable_libtool_lock"
])# _LT_ENABLE_LOCK
# _LT_PROG_AR
# -----------
m4_defun([_LT_PROG_AR],
[AC_CHECK_TOOLS(AR, [ar], false)
: ${AR=ar}
: ${AR_FLAGS=cru}
_LT_DECL([], [AR], [1], [The archiver])
_LT_DECL([], [AR_FLAGS], [1], [Flags to create an archive])
AC_CACHE_CHECK([for archiver @FILE support], [lt_cv_ar_at_file],
[lt_cv_ar_at_file=no
AC_COMPILE_IFELSE([AC_LANG_PROGRAM],
[echo conftest.$ac_objext > conftest.lst
lt_ar_try='$AR $AR_FLAGS libconftest.a @conftest.lst >&AS_MESSAGE_LOG_FD'
AC_TRY_EVAL([lt_ar_try])
if test "$ac_status" -eq 0; then
# Ensure the archiver fails upon bogus file names.
rm -f conftest.$ac_objext libconftest.a
AC_TRY_EVAL([lt_ar_try])
if test "$ac_status" -ne 0; then
lt_cv_ar_at_file=@
fi
fi
rm -f conftest.* libconftest.a
])
])
if test "x$lt_cv_ar_at_file" = xno; then
archiver_list_spec=
else
archiver_list_spec=$lt_cv_ar_at_file
fi
_LT_DECL([], [archiver_list_spec], [1],
[How to feed a file listing to the archiver])
])# _LT_PROG_AR
# _LT_CMD_OLD_ARCHIVE
# -------------------
m4_defun([_LT_CMD_OLD_ARCHIVE],
[_LT_PROG_AR
AC_CHECK_TOOL(STRIP, strip, :)
test -z "$STRIP" && STRIP=:
_LT_DECL([], [STRIP], [1], [A symbol stripping program])
AC_CHECK_TOOL(RANLIB, ranlib, :)
test -z "$RANLIB" && RANLIB=:
_LT_DECL([], [RANLIB], [1],
[Commands used to install an old-style archive])
# Determine commands to create old-style static archives.
old_archive_cmds='$AR $AR_FLAGS $oldlib$oldobjs'
old_postinstall_cmds='chmod 644 $oldlib'
old_postuninstall_cmds=
if test -n "$RANLIB"; then
case $host_os in
openbsd*)
old_postinstall_cmds="$old_postinstall_cmds~\$RANLIB -t \$tool_oldlib"
;;
*)
old_postinstall_cmds="$old_postinstall_cmds~\$RANLIB \$tool_oldlib"
;;
esac
old_archive_cmds="$old_archive_cmds~\$RANLIB \$tool_oldlib"
fi
case $host_os in
darwin*)
lock_old_archive_extraction=yes ;;
*)
lock_old_archive_extraction=no ;;
esac
_LT_DECL([], [old_postinstall_cmds], [2])
_LT_DECL([], [old_postuninstall_cmds], [2])
_LT_TAGDECL([], [old_archive_cmds], [2],
[Commands used to build an old-style archive])
_LT_DECL([], [lock_old_archive_extraction], [0],
[Whether to use a lock for old archive extraction])
])# _LT_CMD_OLD_ARCHIVE
# _LT_COMPILER_OPTION(MESSAGE, VARIABLE-NAME, FLAGS,
# [OUTPUT-FILE], [ACTION-SUCCESS], [ACTION-FAILURE])
# ----------------------------------------------------------------
# Check whether the given compiler option works
AC_DEFUN([_LT_COMPILER_OPTION],
[m4_require([_LT_FILEUTILS_DEFAULTS])dnl
m4_require([_LT_DECL_SED])dnl
AC_CACHE_CHECK([$1], [$2],
[$2=no
m4_if([$4], , [ac_outfile=conftest.$ac_objext], [ac_outfile=$4])
echo "$lt_simple_compile_test_code" > conftest.$ac_ext
lt_compiler_flag="$3"
# Insert the option either (1) after the last *FLAGS variable, or
# (2) before a word containing "conftest.", or (3) at the end.
# Note that $ac_compile itself does not contain backslashes and begins
# with a dollar sign (not a hyphen), so the echo should work correctly.
# The option is referenced via a variable to avoid confusing sed.
lt_compile=`echo "$ac_compile" | $SED \
-e 's:.*FLAGS}\{0,1\} :&$lt_compiler_flag :; t' \
-e 's: [[^ ]]*conftest\.: $lt_compiler_flag&:; t' \
-e 's:$: $lt_compiler_flag:'`
(eval echo "\"\$as_me:$LINENO: $lt_compile\"" >&AS_MESSAGE_LOG_FD)
(eval "$lt_compile" 2>conftest.err)
ac_status=$?
cat conftest.err >&AS_MESSAGE_LOG_FD
echo "$as_me:$LINENO: \$? = $ac_status" >&AS_MESSAGE_LOG_FD
if (exit $ac_status) && test -s "$ac_outfile"; then
# The compiler can only warn and ignore the option if not recognized
# So say no if there are warnings other than the usual output.
$ECHO "$_lt_compiler_boilerplate" | $SED '/^$/d' >conftest.exp
$SED '/^$/d; /^ *+/d' conftest.err >conftest.er2
if test ! -s conftest.er2 || diff conftest.exp conftest.er2 >/dev/null; then
$2=yes
fi
fi
$RM conftest*
])
if test x"[$]$2" = xyes; then
m4_if([$5], , :, [$5])
else
m4_if([$6], , :, [$6])
fi
])# _LT_COMPILER_OPTION
# Old name:
AU_ALIAS([AC_LIBTOOL_COMPILER_OPTION], [_LT_COMPILER_OPTION])
dnl aclocal-1.4 backwards compatibility:
dnl AC_DEFUN([AC_LIBTOOL_COMPILER_OPTION], [])
# _LT_LINKER_OPTION(MESSAGE, VARIABLE-NAME, FLAGS,
# [ACTION-SUCCESS], [ACTION-FAILURE])
# ----------------------------------------------------
# Check whether the given linker option works
AC_DEFUN([_LT_LINKER_OPTION],
[m4_require([_LT_FILEUTILS_DEFAULTS])dnl
m4_require([_LT_DECL_SED])dnl
AC_CACHE_CHECK([$1], [$2],
[$2=no
save_LDFLAGS="$LDFLAGS"
LDFLAGS="$LDFLAGS $3"
echo "$lt_simple_link_test_code" > conftest.$ac_ext
if (eval $ac_link 2>conftest.err) && test -s conftest$ac_exeext; then
# The linker can only warn and ignore the option if not recognized
# So say no if there are warnings
if test -s conftest.err; then
# Append any errors to the config.log.
cat conftest.err 1>&AS_MESSAGE_LOG_FD
$ECHO "$_lt_linker_boilerplate" | $SED '/^$/d' > conftest.exp
$SED '/^$/d; /^ *+/d' conftest.err >conftest.er2
if diff conftest.exp conftest.er2 >/dev/null; then
$2=yes
fi
else
$2=yes
fi
fi
$RM -r conftest*
LDFLAGS="$save_LDFLAGS"
])
if test x"[$]$2" = xyes; then
m4_if([$4], , :, [$4])
else
m4_if([$5], , :, [$5])
fi
])# _LT_LINKER_OPTION
# Old name:
AU_ALIAS([AC_LIBTOOL_LINKER_OPTION], [_LT_LINKER_OPTION])
dnl aclocal-1.4 backwards compatibility:
dnl AC_DEFUN([AC_LIBTOOL_LINKER_OPTION], [])
# LT_CMD_MAX_LEN
#---------------
AC_DEFUN([LT_CMD_MAX_LEN],
[AC_REQUIRE([AC_CANONICAL_HOST])dnl
# find the maximum length of command line arguments
AC_MSG_CHECKING([the maximum length of command line arguments])
AC_CACHE_VAL([lt_cv_sys_max_cmd_len], [dnl
i=0
teststring="ABCD"
case $build_os in
msdosdjgpp*)
# On DJGPP, this test can blow up pretty badly due to problems in libc
# (any single argument exceeding 2000 bytes causes a buffer overrun
# during glob expansion). Even if it were fixed, the result of this
# check would be larger than it should be.
lt_cv_sys_max_cmd_len=12288; # 12K is about right
;;
gnu*)
# Under GNU Hurd, this test is not required because there is
# no limit to the length of command line arguments.
# Libtool will interpret -1 as no limit whatsoever
lt_cv_sys_max_cmd_len=-1;
;;
cygwin* | mingw* | cegcc*)
# On Win9x/ME, this test blows up -- it succeeds, but takes
# about 5 minutes as the teststring grows exponentially.
# Worse, since 9x/ME are not pre-emptively multitasking,
# you end up with a "frozen" computer, even though with patience
# the test eventually succeeds (with a max line length of 256k).
# Instead, let's just punt: use the minimum linelength reported by
# all of the supported platforms: 8192 (on NT/2K/XP).
lt_cv_sys_max_cmd_len=8192;
;;
mint*)
# On MiNT this can take a long time and run out of memory.
lt_cv_sys_max_cmd_len=8192;
;;
amigaos*)
# On AmigaOS with pdksh, this test takes hours, literally.
# So we just punt and use a minimum line length of 8192.
lt_cv_sys_max_cmd_len=8192;
;;
netbsd* | freebsd* | openbsd* | darwin* | dragonfly*)
# This has been around since 386BSD, at least. Likely further.
if test -x /sbin/sysctl; then
lt_cv_sys_max_cmd_len=`/sbin/sysctl -n kern.argmax`
elif test -x /usr/sbin/sysctl; then
lt_cv_sys_max_cmd_len=`/usr/sbin/sysctl -n kern.argmax`
else
lt_cv_sys_max_cmd_len=65536 # usable default for all BSDs
fi
# And add a safety zone
lt_cv_sys_max_cmd_len=`expr $lt_cv_sys_max_cmd_len \/ 4`
lt_cv_sys_max_cmd_len=`expr $lt_cv_sys_max_cmd_len \* 3`
;;
interix*)
# We know the value 262144 and hardcode it with a safety zone (like BSD)
lt_cv_sys_max_cmd_len=196608
;;
os2*)
# The test takes a long time on OS/2.
lt_cv_sys_max_cmd_len=8192
;;
osf*)
# Dr. Hans Ekkehard Plesser reports seeing a kernel panic running configure
# due to this test when exec_disable_arg_limit is 1 on Tru64. It is not
# nice to cause kernel panics so lets avoid the loop below.
# First set a reasonable default.
lt_cv_sys_max_cmd_len=16384
#
if test -x /sbin/sysconfig; then
case `/sbin/sysconfig -q proc exec_disable_arg_limit` in
*1*) lt_cv_sys_max_cmd_len=-1 ;;
esac
fi
;;
sco3.2v5*)
lt_cv_sys_max_cmd_len=102400
;;
sysv5* | sco5v6* | sysv4.2uw2*)
kargmax=`grep ARG_MAX /etc/conf/cf.d/stune 2>/dev/null`
if test -n "$kargmax"; then
lt_cv_sys_max_cmd_len=`echo $kargmax | sed 's/.*[[ ]]//'`
else
lt_cv_sys_max_cmd_len=32768
fi
;;
*)
lt_cv_sys_max_cmd_len=`(getconf ARG_MAX) 2> /dev/null`
if test -n "$lt_cv_sys_max_cmd_len"; then
lt_cv_sys_max_cmd_len=`expr $lt_cv_sys_max_cmd_len \/ 4`
lt_cv_sys_max_cmd_len=`expr $lt_cv_sys_max_cmd_len \* 3`
else
# Make teststring a little bigger before we do anything with it.
# a 1K string should be a reasonable start.
for i in 1 2 3 4 5 6 7 8 ; do
teststring=$teststring$teststring
done
SHELL=${SHELL-${CONFIG_SHELL-/bin/sh}}
# If test is not a shell built-in, we'll probably end up computing a
# maximum length that is only half of the actual maximum length, but
# we can't tell.
while { test "X"`env echo "$teststring$teststring" 2>/dev/null` \
= "X$teststring$teststring"; } >/dev/null 2>&1 &&
test $i != 17 # 1/2 MB should be enough
do
i=`expr $i + 1`
teststring=$teststring$teststring
done
# Only check the string length outside the loop.
lt_cv_sys_max_cmd_len=`expr "X$teststring" : ".*" 2>&1`
teststring=
# Add a significant safety factor because C++ compilers can tack on
# massive amounts of additional arguments before passing them to the
# linker. It appears as though 1/2 is a usable value.
lt_cv_sys_max_cmd_len=`expr $lt_cv_sys_max_cmd_len \/ 2`
fi
;;
esac
])
if test -n $lt_cv_sys_max_cmd_len ; then
AC_MSG_RESULT($lt_cv_sys_max_cmd_len)
else
AC_MSG_RESULT(none)
fi
max_cmd_len=$lt_cv_sys_max_cmd_len
_LT_DECL([], [max_cmd_len], [0],
[What is the maximum length of a command?])
])# LT_CMD_MAX_LEN
# Old name:
AU_ALIAS([AC_LIBTOOL_SYS_MAX_CMD_LEN], [LT_CMD_MAX_LEN])
dnl aclocal-1.4 backwards compatibility:
dnl AC_DEFUN([AC_LIBTOOL_SYS_MAX_CMD_LEN], [])
# _LT_HEADER_DLFCN
# ----------------
m4_defun([_LT_HEADER_DLFCN],
[AC_CHECK_HEADERS([dlfcn.h], [], [], [AC_INCLUDES_DEFAULT])dnl
])# _LT_HEADER_DLFCN
# _LT_TRY_DLOPEN_SELF (ACTION-IF-TRUE, ACTION-IF-TRUE-W-USCORE,
# ACTION-IF-FALSE, ACTION-IF-CROSS-COMPILING)
# ----------------------------------------------------------------
m4_defun([_LT_TRY_DLOPEN_SELF],
[m4_require([_LT_HEADER_DLFCN])dnl
if test "$cross_compiling" = yes; then :
[$4]
else
lt_dlunknown=0; lt_dlno_uscore=1; lt_dlneed_uscore=2
lt_status=$lt_dlunknown
cat > conftest.$ac_ext <<_LT_EOF
[#line $LINENO "configure"
#include "confdefs.h"
#if HAVE_DLFCN_H
#include
#endif
#include
#ifdef RTLD_GLOBAL
# define LT_DLGLOBAL RTLD_GLOBAL
#else
# ifdef DL_GLOBAL
# define LT_DLGLOBAL DL_GLOBAL
# else
# define LT_DLGLOBAL 0
# endif
#endif
/* We may have to define LT_DLLAZY_OR_NOW in the command line if we
find out it does not work in some platform. */
#ifndef LT_DLLAZY_OR_NOW
# ifdef RTLD_LAZY
# define LT_DLLAZY_OR_NOW RTLD_LAZY
# else
# ifdef DL_LAZY
# define LT_DLLAZY_OR_NOW DL_LAZY
# else
# ifdef RTLD_NOW
# define LT_DLLAZY_OR_NOW RTLD_NOW
# else
# ifdef DL_NOW
# define LT_DLLAZY_OR_NOW DL_NOW
# else
# define LT_DLLAZY_OR_NOW 0
# endif
# endif
# endif
# endif
#endif
/* When -fvisbility=hidden is used, assume the code has been annotated
correspondingly for the symbols needed. */
#if defined(__GNUC__) && (((__GNUC__ == 3) && (__GNUC_MINOR__ >= 3)) || (__GNUC__ > 3))
int fnord () __attribute__((visibility("default")));
#endif
int fnord () { return 42; }
int main ()
{
void *self = dlopen (0, LT_DLGLOBAL|LT_DLLAZY_OR_NOW);
int status = $lt_dlunknown;
if (self)
{
if (dlsym (self,"fnord")) status = $lt_dlno_uscore;
else
{
if (dlsym( self,"_fnord")) status = $lt_dlneed_uscore;
else puts (dlerror ());
}
/* dlclose (self); */
}
else
puts (dlerror ());
return status;
}]
_LT_EOF
if AC_TRY_EVAL(ac_link) && test -s conftest${ac_exeext} 2>/dev/null; then
(./conftest; exit; ) >&AS_MESSAGE_LOG_FD 2>/dev/null
lt_status=$?
case x$lt_status in
x$lt_dlno_uscore) $1 ;;
x$lt_dlneed_uscore) $2 ;;
x$lt_dlunknown|x*) $3 ;;
esac
else :
# compilation failed
$3
fi
fi
rm -fr conftest*
])# _LT_TRY_DLOPEN_SELF
# LT_SYS_DLOPEN_SELF
# ------------------
AC_DEFUN([LT_SYS_DLOPEN_SELF],
[m4_require([_LT_HEADER_DLFCN])dnl
if test "x$enable_dlopen" != xyes; then
enable_dlopen=unknown
enable_dlopen_self=unknown
enable_dlopen_self_static=unknown
else
lt_cv_dlopen=no
lt_cv_dlopen_libs=
case $host_os in
beos*)
lt_cv_dlopen="load_add_on"
lt_cv_dlopen_libs=
lt_cv_dlopen_self=yes
;;
mingw* | pw32* | cegcc*)
lt_cv_dlopen="LoadLibrary"
lt_cv_dlopen_libs=
;;
cygwin*)
lt_cv_dlopen="dlopen"
lt_cv_dlopen_libs=
;;
darwin*)
# if libdl is installed we need to link against it
AC_CHECK_LIB([dl], [dlopen],
[lt_cv_dlopen="dlopen" lt_cv_dlopen_libs="-ldl"],[
lt_cv_dlopen="dyld"
lt_cv_dlopen_libs=
lt_cv_dlopen_self=yes
])
;;
*)
AC_CHECK_FUNC([shl_load],
[lt_cv_dlopen="shl_load"],
[AC_CHECK_LIB([dld], [shl_load],
[lt_cv_dlopen="shl_load" lt_cv_dlopen_libs="-ldld"],
[AC_CHECK_FUNC([dlopen],
[lt_cv_dlopen="dlopen"],
[AC_CHECK_LIB([dl], [dlopen],
[lt_cv_dlopen="dlopen" lt_cv_dlopen_libs="-ldl"],
[AC_CHECK_LIB([svld], [dlopen],
[lt_cv_dlopen="dlopen" lt_cv_dlopen_libs="-lsvld"],
[AC_CHECK_LIB([dld], [dld_link],
[lt_cv_dlopen="dld_link" lt_cv_dlopen_libs="-ldld"])
])
])
])
])
])
;;
esac
if test "x$lt_cv_dlopen" != xno; then
enable_dlopen=yes
else
enable_dlopen=no
fi
case $lt_cv_dlopen in
dlopen)
save_CPPFLAGS="$CPPFLAGS"
test "x$ac_cv_header_dlfcn_h" = xyes && CPPFLAGS="$CPPFLAGS -DHAVE_DLFCN_H"
save_LDFLAGS="$LDFLAGS"
wl=$lt_prog_compiler_wl eval LDFLAGS=\"\$LDFLAGS $export_dynamic_flag_spec\"
save_LIBS="$LIBS"
LIBS="$lt_cv_dlopen_libs $LIBS"
AC_CACHE_CHECK([whether a program can dlopen itself],
lt_cv_dlopen_self, [dnl
_LT_TRY_DLOPEN_SELF(
lt_cv_dlopen_self=yes, lt_cv_dlopen_self=yes,
lt_cv_dlopen_self=no, lt_cv_dlopen_self=cross)
])
if test "x$lt_cv_dlopen_self" = xyes; then
wl=$lt_prog_compiler_wl eval LDFLAGS=\"\$LDFLAGS $lt_prog_compiler_static\"
AC_CACHE_CHECK([whether a statically linked program can dlopen itself],
lt_cv_dlopen_self_static, [dnl
_LT_TRY_DLOPEN_SELF(
lt_cv_dlopen_self_static=yes, lt_cv_dlopen_self_static=yes,
lt_cv_dlopen_self_static=no, lt_cv_dlopen_self_static=cross)
])
fi
CPPFLAGS="$save_CPPFLAGS"
LDFLAGS="$save_LDFLAGS"
LIBS="$save_LIBS"
;;
esac
case $lt_cv_dlopen_self in
yes|no) enable_dlopen_self=$lt_cv_dlopen_self ;;
*) enable_dlopen_self=unknown ;;
esac
case $lt_cv_dlopen_self_static in
yes|no) enable_dlopen_self_static=$lt_cv_dlopen_self_static ;;
*) enable_dlopen_self_static=unknown ;;
esac
fi
_LT_DECL([dlopen_support], [enable_dlopen], [0],
[Whether dlopen is supported])
_LT_DECL([dlopen_self], [enable_dlopen_self], [0],
[Whether dlopen of programs is supported])
_LT_DECL([dlopen_self_static], [enable_dlopen_self_static], [0],
[Whether dlopen of statically linked programs is supported])
])# LT_SYS_DLOPEN_SELF
# Old name:
AU_ALIAS([AC_LIBTOOL_DLOPEN_SELF], [LT_SYS_DLOPEN_SELF])
dnl aclocal-1.4 backwards compatibility:
dnl AC_DEFUN([AC_LIBTOOL_DLOPEN_SELF], [])
# _LT_COMPILER_C_O([TAGNAME])
# ---------------------------
# Check to see if options -c and -o are simultaneously supported by compiler.
# This macro does not hard code the compiler like AC_PROG_CC_C_O.
m4_defun([_LT_COMPILER_C_O],
[m4_require([_LT_DECL_SED])dnl
m4_require([_LT_FILEUTILS_DEFAULTS])dnl
m4_require([_LT_TAG_COMPILER])dnl
AC_CACHE_CHECK([if $compiler supports -c -o file.$ac_objext],
[_LT_TAGVAR(lt_cv_prog_compiler_c_o, $1)],
[_LT_TAGVAR(lt_cv_prog_compiler_c_o, $1)=no
$RM -r conftest 2>/dev/null
mkdir conftest
cd conftest
mkdir out
echo "$lt_simple_compile_test_code" > conftest.$ac_ext
lt_compiler_flag="-o out/conftest2.$ac_objext"
# Insert the option either (1) after the last *FLAGS variable, or
# (2) before a word containing "conftest.", or (3) at the end.
# Note that $ac_compile itself does not contain backslashes and begins
# with a dollar sign (not a hyphen), so the echo should work correctly.
lt_compile=`echo "$ac_compile" | $SED \
-e 's:.*FLAGS}\{0,1\} :&$lt_compiler_flag :; t' \
-e 's: [[^ ]]*conftest\.: $lt_compiler_flag&:; t' \
-e 's:$: $lt_compiler_flag:'`
(eval echo "\"\$as_me:$LINENO: $lt_compile\"" >&AS_MESSAGE_LOG_FD)
(eval "$lt_compile" 2>out/conftest.err)
ac_status=$?
cat out/conftest.err >&AS_MESSAGE_LOG_FD
echo "$as_me:$LINENO: \$? = $ac_status" >&AS_MESSAGE_LOG_FD
if (exit $ac_status) && test -s out/conftest2.$ac_objext
then
# The compiler can only warn and ignore the option if not recognized
# So say no if there are warnings
$ECHO "$_lt_compiler_boilerplate" | $SED '/^$/d' > out/conftest.exp
$SED '/^$/d; /^ *+/d' out/conftest.err >out/conftest.er2
if test ! -s out/conftest.er2 || diff out/conftest.exp out/conftest.er2 >/dev/null; then
_LT_TAGVAR(lt_cv_prog_compiler_c_o, $1)=yes
fi
fi
chmod u+w . 2>&AS_MESSAGE_LOG_FD
$RM conftest*
# SGI C++ compiler will create directory out/ii_files/ for
# template instantiation
test -d out/ii_files && $RM out/ii_files/* && rmdir out/ii_files
$RM out/* && rmdir out
cd ..
$RM -r conftest
$RM conftest*
])
_LT_TAGDECL([compiler_c_o], [lt_cv_prog_compiler_c_o], [1],
[Does compiler simultaneously support -c and -o options?])
])# _LT_COMPILER_C_O
# _LT_COMPILER_FILE_LOCKS([TAGNAME])
# ----------------------------------
# Check to see if we can do hard links to lock some files if needed
m4_defun([_LT_COMPILER_FILE_LOCKS],
[m4_require([_LT_ENABLE_LOCK])dnl
m4_require([_LT_FILEUTILS_DEFAULTS])dnl
_LT_COMPILER_C_O([$1])
hard_links="nottested"
if test "$_LT_TAGVAR(lt_cv_prog_compiler_c_o, $1)" = no && test "$need_locks" != no; then
# do not overwrite the value of need_locks provided by the user
AC_MSG_CHECKING([if we can lock with hard links])
hard_links=yes
$RM conftest*
ln conftest.a conftest.b 2>/dev/null && hard_links=no
touch conftest.a
ln conftest.a conftest.b 2>&5 || hard_links=no
ln conftest.a conftest.b 2>/dev/null && hard_links=no
AC_MSG_RESULT([$hard_links])
if test "$hard_links" = no; then
AC_MSG_WARN([`$CC' does not support `-c -o', so `make -j' may be unsafe])
need_locks=warn
fi
else
need_locks=no
fi
_LT_DECL([], [need_locks], [1], [Must we lock files when doing compilation?])
])# _LT_COMPILER_FILE_LOCKS
# _LT_CHECK_OBJDIR
# ----------------
m4_defun([_LT_CHECK_OBJDIR],
[AC_CACHE_CHECK([for objdir], [lt_cv_objdir],
[rm -f .libs 2>/dev/null
mkdir .libs 2>/dev/null
if test -d .libs; then
lt_cv_objdir=.libs
else
# MS-DOS does not allow filenames that begin with a dot.
lt_cv_objdir=_libs
fi
rmdir .libs 2>/dev/null])
objdir=$lt_cv_objdir
_LT_DECL([], [objdir], [0],
[The name of the directory that contains temporary libtool files])dnl
m4_pattern_allow([LT_OBJDIR])dnl
AC_DEFINE_UNQUOTED(LT_OBJDIR, "$lt_cv_objdir/",
[Define to the sub-directory in which libtool stores uninstalled libraries.])
])# _LT_CHECK_OBJDIR
# _LT_LINKER_HARDCODE_LIBPATH([TAGNAME])
# --------------------------------------
# Check hardcoding attributes.
m4_defun([_LT_LINKER_HARDCODE_LIBPATH],
[AC_MSG_CHECKING([how to hardcode library paths into programs])
_LT_TAGVAR(hardcode_action, $1)=
if test -n "$_LT_TAGVAR(hardcode_libdir_flag_spec, $1)" ||
test -n "$_LT_TAGVAR(runpath_var, $1)" ||
test "X$_LT_TAGVAR(hardcode_automatic, $1)" = "Xyes" ; then
# We can hardcode non-existent directories.
if test "$_LT_TAGVAR(hardcode_direct, $1)" != no &&
# If the only mechanism to avoid hardcoding is shlibpath_var, we
# have to relink, otherwise we might link with an installed library
# when we should be linking with a yet-to-be-installed one
## test "$_LT_TAGVAR(hardcode_shlibpath_var, $1)" != no &&
test "$_LT_TAGVAR(hardcode_minus_L, $1)" != no; then
# Linking always hardcodes the temporary library directory.
_LT_TAGVAR(hardcode_action, $1)=relink
else
# We can link without hardcoding, and we can hardcode nonexisting dirs.
_LT_TAGVAR(hardcode_action, $1)=immediate
fi
else
# We cannot hardcode anything, or else we can only hardcode existing
# directories.
_LT_TAGVAR(hardcode_action, $1)=unsupported
fi
AC_MSG_RESULT([$_LT_TAGVAR(hardcode_action, $1)])
if test "$_LT_TAGVAR(hardcode_action, $1)" = relink ||
test "$_LT_TAGVAR(inherit_rpath, $1)" = yes; then
# Fast installation is not supported
enable_fast_install=no
elif test "$shlibpath_overrides_runpath" = yes ||
test "$enable_shared" = no; then
# Fast installation is not necessary
enable_fast_install=needless
fi
_LT_TAGDECL([], [hardcode_action], [0],
[How to hardcode a shared library path into an executable])
])# _LT_LINKER_HARDCODE_LIBPATH
# _LT_CMD_STRIPLIB
# ----------------
m4_defun([_LT_CMD_STRIPLIB],
[m4_require([_LT_DECL_EGREP])
striplib=
old_striplib=
AC_MSG_CHECKING([whether stripping libraries is possible])
if test -n "$STRIP" && $STRIP -V 2>&1 | $GREP "GNU strip" >/dev/null; then
test -z "$old_striplib" && old_striplib="$STRIP --strip-debug"
test -z "$striplib" && striplib="$STRIP --strip-unneeded"
AC_MSG_RESULT([yes])
else
# FIXME - insert some real tests, host_os isn't really good enough
case $host_os in
darwin*)
if test -n "$STRIP" ; then
striplib="$STRIP -x"
old_striplib="$STRIP -S"
AC_MSG_RESULT([yes])
else
AC_MSG_RESULT([no])
fi
;;
*)
AC_MSG_RESULT([no])
;;
esac
fi
_LT_DECL([], [old_striplib], [1], [Commands to strip libraries])
_LT_DECL([], [striplib], [1])
])# _LT_CMD_STRIPLIB
# _LT_SYS_DYNAMIC_LINKER([TAG])
# -----------------------------
# PORTME Fill in your ld.so characteristics
m4_defun([_LT_SYS_DYNAMIC_LINKER],
[AC_REQUIRE([AC_CANONICAL_HOST])dnl
m4_require([_LT_DECL_EGREP])dnl
m4_require([_LT_FILEUTILS_DEFAULTS])dnl
m4_require([_LT_DECL_OBJDUMP])dnl
m4_require([_LT_DECL_SED])dnl
m4_require([_LT_CHECK_SHELL_FEATURES])dnl
AC_MSG_CHECKING([dynamic linker characteristics])
m4_if([$1],
[], [
if test "$GCC" = yes; then
case $host_os in
darwin*) lt_awk_arg="/^libraries:/,/LR/" ;;
*) lt_awk_arg="/^libraries:/" ;;
esac
case $host_os in
mingw* | cegcc*) lt_sed_strip_eq="s,=\([[A-Za-z]]:\),\1,g" ;;
*) lt_sed_strip_eq="s,=/,/,g" ;;
esac
lt_search_path_spec=`$CC -print-search-dirs | awk $lt_awk_arg | $SED -e "s/^libraries://" -e $lt_sed_strip_eq`
case $lt_search_path_spec in
*\;*)
# if the path contains ";" then we assume it to be the separator
# otherwise default to the standard path separator (i.e. ":") - it is
# assumed that no part of a normal pathname contains ";" but that should
# okay in the real world where ";" in dirpaths is itself problematic.
lt_search_path_spec=`$ECHO "$lt_search_path_spec" | $SED 's/;/ /g'`
;;
*)
lt_search_path_spec=`$ECHO "$lt_search_path_spec" | $SED "s/$PATH_SEPARATOR/ /g"`
;;
esac
# Ok, now we have the path, separated by spaces, we can step through it
# and add multilib dir if necessary.
lt_tmp_lt_search_path_spec=
lt_multi_os_dir=`$CC $CPPFLAGS $CFLAGS $LDFLAGS -print-multi-os-directory 2>/dev/null`
for lt_sys_path in $lt_search_path_spec; do
if test -d "$lt_sys_path/$lt_multi_os_dir"; then
lt_tmp_lt_search_path_spec="$lt_tmp_lt_search_path_spec $lt_sys_path/$lt_multi_os_dir"
else
test -d "$lt_sys_path" && \
lt_tmp_lt_search_path_spec="$lt_tmp_lt_search_path_spec $lt_sys_path"
fi
done
lt_search_path_spec=`$ECHO "$lt_tmp_lt_search_path_spec" | awk '
BEGIN {RS=" "; FS="/|\n";} {
lt_foo="";
lt_count=0;
for (lt_i = NF; lt_i > 0; lt_i--) {
if ($lt_i != "" && $lt_i != ".") {
if ($lt_i == "..") {
lt_count++;
} else {
if (lt_count == 0) {
lt_foo="/" $lt_i lt_foo;
} else {
lt_count--;
}
}
}
}
if (lt_foo != "") { lt_freq[[lt_foo]]++; }
if (lt_freq[[lt_foo]] == 1) { print lt_foo; }
}'`
# AWK program above erroneously prepends '/' to C:/dos/paths
# for these hosts.
case $host_os in
mingw* | cegcc*) lt_search_path_spec=`$ECHO "$lt_search_path_spec" |\
$SED 's,/\([[A-Za-z]]:\),\1,g'` ;;
esac
sys_lib_search_path_spec=`$ECHO "$lt_search_path_spec" | $lt_NL2SP`
else
sys_lib_search_path_spec="/lib /usr/lib /usr/local/lib"
fi])
library_names_spec=
libname_spec='lib$name'
soname_spec=
shrext_cmds=".so"
postinstall_cmds=
postuninstall_cmds=
finish_cmds=
finish_eval=
shlibpath_var=
shlibpath_overrides_runpath=unknown
version_type=none
dynamic_linker="$host_os ld.so"
sys_lib_dlsearch_path_spec="/lib /usr/lib"
need_lib_prefix=unknown
hardcode_into_libs=no
# when you set need_version to no, make sure it does not cause -set_version
# flags to be left without arguments
need_version=unknown
case $host_os in
aix3*)
version_type=linux # correct to gnu/linux during the next big refactor
library_names_spec='${libname}${release}${shared_ext}$versuffix $libname.a'
shlibpath_var=LIBPATH
# AIX 3 has no versioning support, so we append a major version to the name.
soname_spec='${libname}${release}${shared_ext}$major'
;;
aix[[4-9]]*)
version_type=linux # correct to gnu/linux during the next big refactor
need_lib_prefix=no
need_version=no
hardcode_into_libs=yes
if test "$host_cpu" = ia64; then
# AIX 5 supports IA64
library_names_spec='${libname}${release}${shared_ext}$major ${libname}${release}${shared_ext}$versuffix $libname${shared_ext}'
shlibpath_var=LD_LIBRARY_PATH
else
# With GCC up to 2.95.x, collect2 would create an import file
# for dependence libraries. The import file would start with
# the line `#! .'. This would cause the generated library to
# depend on `.', always an invalid library. This was fixed in
# development snapshots of GCC prior to 3.0.
case $host_os in
aix4 | aix4.[[01]] | aix4.[[01]].*)
if { echo '#if __GNUC__ > 2 || (__GNUC__ == 2 && __GNUC_MINOR__ >= 97)'
echo ' yes '
echo '#endif'; } | ${CC} -E - | $GREP yes > /dev/null; then
:
else
can_build_shared=no
fi
;;
esac
# AIX (on Power*) has no versioning support, so currently we can not hardcode correct
# soname into executable. Probably we can add versioning support to
# collect2, so additional links can be useful in future.
if test "$aix_use_runtimelinking" = yes; then
# If using run time linking (on AIX 4.2 or later) use lib.so
# instead of lib.a to let people know that these are not
# typical AIX shared libraries.
library_names_spec='${libname}${release}${shared_ext}$versuffix ${libname}${release}${shared_ext}$major $libname${shared_ext}'
else
# We preserve .a as extension for shared libraries through AIX4.2
# and later when we are not doing run time linking.
library_names_spec='${libname}${release}.a $libname.a'
soname_spec='${libname}${release}${shared_ext}$major'
fi
shlibpath_var=LIBPATH
fi
;;
amigaos*)
case $host_cpu in
powerpc)
# Since July 2007 AmigaOS4 officially supports .so libraries.
# When compiling the executable, add -use-dynld -Lsobjs: to the compileline.
library_names_spec='${libname}${release}${shared_ext}$versuffix ${libname}${release}${shared_ext}$major $libname${shared_ext}'
;;
m68k)
library_names_spec='$libname.ixlibrary $libname.a'
# Create ${libname}_ixlibrary.a entries in /sys/libs.
finish_eval='for lib in `ls $libdir/*.ixlibrary 2>/dev/null`; do libname=`func_echo_all "$lib" | $SED '\''s%^.*/\([[^/]]*\)\.ixlibrary$%\1%'\''`; test $RM /sys/libs/${libname}_ixlibrary.a; $show "cd /sys/libs && $LN_S $lib ${libname}_ixlibrary.a"; cd /sys/libs && $LN_S $lib ${libname}_ixlibrary.a || exit 1; done'
;;
esac
;;
beos*)
library_names_spec='${libname}${shared_ext}'
dynamic_linker="$host_os ld.so"
shlibpath_var=LIBRARY_PATH
;;
bsdi[[45]]*)
version_type=linux # correct to gnu/linux during the next big refactor
need_version=no
library_names_spec='${libname}${release}${shared_ext}$versuffix ${libname}${release}${shared_ext}$major $libname${shared_ext}'
soname_spec='${libname}${release}${shared_ext}$major'
finish_cmds='PATH="\$PATH:/sbin" ldconfig $libdir'
shlibpath_var=LD_LIBRARY_PATH
sys_lib_search_path_spec="/shlib /usr/lib /usr/X11/lib /usr/contrib/lib /lib /usr/local/lib"
sys_lib_dlsearch_path_spec="/shlib /usr/lib /usr/local/lib"
# the default ld.so.conf also contains /usr/contrib/lib and
# /usr/X11R6/lib (/usr/X11 is a link to /usr/X11R6), but let us allow
# libtool to hard-code these into programs
;;
cygwin* | mingw* | pw32* | cegcc*)
version_type=windows
shrext_cmds=".dll"
need_version=no
need_lib_prefix=no
case $GCC,$cc_basename in
yes,*)
# gcc
library_names_spec='$libname.dll.a'
# DLL is installed to $(libdir)/../bin by postinstall_cmds
postinstall_cmds='base_file=`basename \${file}`~
dlpath=`$SHELL 2>&1 -c '\''. $dir/'\''\${base_file}'\''i; echo \$dlname'\''`~
dldir=$destdir/`dirname \$dlpath`~
test -d \$dldir || mkdir -p \$dldir~
$install_prog $dir/$dlname \$dldir/$dlname~
chmod a+x \$dldir/$dlname~
if test -n '\''$stripme'\'' && test -n '\''$striplib'\''; then
eval '\''$striplib \$dldir/$dlname'\'' || exit \$?;
fi'
postuninstall_cmds='dldll=`$SHELL 2>&1 -c '\''. $file; echo \$dlname'\''`~
dlpath=$dir/\$dldll~
$RM \$dlpath'
shlibpath_overrides_runpath=yes
case $host_os in
cygwin*)
# Cygwin DLLs use 'cyg' prefix rather than 'lib'
soname_spec='`echo ${libname} | sed -e 's/^lib/cyg/'``echo ${release} | $SED -e 's/[[.]]/-/g'`${versuffix}${shared_ext}'
m4_if([$1], [],[
sys_lib_search_path_spec="$sys_lib_search_path_spec /usr/lib/w32api"])
;;
mingw* | cegcc*)
# MinGW DLLs use traditional 'lib' prefix
soname_spec='${libname}`echo ${release} | $SED -e 's/[[.]]/-/g'`${versuffix}${shared_ext}'
;;
pw32*)
# pw32 DLLs use 'pw' prefix rather than 'lib'
library_names_spec='`echo ${libname} | sed -e 's/^lib/pw/'``echo ${release} | $SED -e 's/[[.]]/-/g'`${versuffix}${shared_ext}'
;;
esac
dynamic_linker='Win32 ld.exe'
;;
*,cl*)
# Native MSVC
libname_spec='$name'
soname_spec='${libname}`echo ${release} | $SED -e 's/[[.]]/-/g'`${versuffix}${shared_ext}'
library_names_spec='${libname}.dll.lib'
case $build_os in
mingw*)
sys_lib_search_path_spec=
lt_save_ifs=$IFS
IFS=';'
for lt_path in $LIB
do
IFS=$lt_save_ifs
# Let DOS variable expansion print the short 8.3 style file name.
lt_path=`cd "$lt_path" 2>/dev/null && cmd //C "for %i in (".") do @echo %~si"`
sys_lib_search_path_spec="$sys_lib_search_path_spec $lt_path"
done
IFS=$lt_save_ifs
# Convert to MSYS style.
sys_lib_search_path_spec=`$ECHO "$sys_lib_search_path_spec" | sed -e 's|\\\\|/|g' -e 's| \\([[a-zA-Z]]\\):| /\\1|g' -e 's|^ ||'`
;;
cygwin*)
# Convert to unix form, then to dos form, then back to unix form
# but this time dos style (no spaces!) so that the unix form looks
# like /cygdrive/c/PROGRA~1:/cygdr...
sys_lib_search_path_spec=`cygpath --path --unix "$LIB"`
sys_lib_search_path_spec=`cygpath --path --dos "$sys_lib_search_path_spec" 2>/dev/null`
sys_lib_search_path_spec=`cygpath --path --unix "$sys_lib_search_path_spec" | $SED -e "s/$PATH_SEPARATOR/ /g"`
;;
*)
sys_lib_search_path_spec="$LIB"
if $ECHO "$sys_lib_search_path_spec" | [$GREP ';[c-zC-Z]:/' >/dev/null]; then
# It is most probably a Windows format PATH.
sys_lib_search_path_spec=`$ECHO "$sys_lib_search_path_spec" | $SED -e 's/;/ /g'`
else
sys_lib_search_path_spec=`$ECHO "$sys_lib_search_path_spec" | $SED -e "s/$PATH_SEPARATOR/ /g"`
fi
# FIXME: find the short name or the path components, as spaces are
# common. (e.g. "Program Files" -> "PROGRA~1")
;;
esac
# DLL is installed to $(libdir)/../bin by postinstall_cmds
postinstall_cmds='base_file=`basename \${file}`~
dlpath=`$SHELL 2>&1 -c '\''. $dir/'\''\${base_file}'\''i; echo \$dlname'\''`~
dldir=$destdir/`dirname \$dlpath`~
test -d \$dldir || mkdir -p \$dldir~
$install_prog $dir/$dlname \$dldir/$dlname'
postuninstall_cmds='dldll=`$SHELL 2>&1 -c '\''. $file; echo \$dlname'\''`~
dlpath=$dir/\$dldll~
$RM \$dlpath'
shlibpath_overrides_runpath=yes
dynamic_linker='Win32 link.exe'
;;
*)
# Assume MSVC wrapper
library_names_spec='${libname}`echo ${release} | $SED -e 's/[[.]]/-/g'`${versuffix}${shared_ext} $libname.lib'
dynamic_linker='Win32 ld.exe'
;;
esac
# FIXME: first we should search . and the directory the executable is in
shlibpath_var=PATH
;;
darwin* | rhapsody*)
dynamic_linker="$host_os dyld"
version_type=darwin
need_lib_prefix=no
need_version=no
library_names_spec='${libname}${release}${major}$shared_ext ${libname}$shared_ext'
soname_spec='${libname}${release}${major}$shared_ext'
shlibpath_overrides_runpath=yes
shlibpath_var=DYLD_LIBRARY_PATH
shrext_cmds='`test .$module = .yes && echo .so || echo .dylib`'
m4_if([$1], [],[
sys_lib_search_path_spec="$sys_lib_search_path_spec /usr/local/lib"])
sys_lib_dlsearch_path_spec='/usr/local/lib /lib /usr/lib'
;;
dgux*)
version_type=linux # correct to gnu/linux during the next big refactor
need_lib_prefix=no
need_version=no
library_names_spec='${libname}${release}${shared_ext}$versuffix ${libname}${release}${shared_ext}$major $libname$shared_ext'
soname_spec='${libname}${release}${shared_ext}$major'
shlibpath_var=LD_LIBRARY_PATH
;;
freebsd* | dragonfly*)
# DragonFly does not have aout. When/if they implement a new
# versioning mechanism, adjust this.
if test -x /usr/bin/objformat; then
objformat=`/usr/bin/objformat`
else
case $host_os in
freebsd[[23]].*) objformat=aout ;;
*) objformat=elf ;;
esac
fi
version_type=freebsd-$objformat
case $version_type in
freebsd-elf*)
library_names_spec='${libname}${release}${shared_ext}$versuffix ${libname}${release}${shared_ext} $libname${shared_ext}'
need_version=no
need_lib_prefix=no
;;
freebsd-*)
library_names_spec='${libname}${release}${shared_ext}$versuffix $libname${shared_ext}$versuffix'
need_version=yes
;;
esac
shlibpath_var=LD_LIBRARY_PATH
case $host_os in
freebsd2.*)
shlibpath_overrides_runpath=yes
;;
freebsd3.[[01]]* | freebsdelf3.[[01]]*)
shlibpath_overrides_runpath=yes
hardcode_into_libs=yes
;;
freebsd3.[[2-9]]* | freebsdelf3.[[2-9]]* | \
freebsd4.[[0-5]] | freebsdelf4.[[0-5]] | freebsd4.1.1 | freebsdelf4.1.1)
shlibpath_overrides_runpath=no
hardcode_into_libs=yes
;;
*) # from 4.6 on, and DragonFly
shlibpath_overrides_runpath=yes
hardcode_into_libs=yes
;;
esac
;;
gnu*)
version_type=linux # correct to gnu/linux during the next big refactor
need_lib_prefix=no
need_version=no
library_names_spec='${libname}${release}${shared_ext}$versuffix ${libname}${release}${shared_ext}${major} ${libname}${shared_ext}'
soname_spec='${libname}${release}${shared_ext}$major'
shlibpath_var=LD_LIBRARY_PATH
shlibpath_overrides_runpath=no
hardcode_into_libs=yes
;;
haiku*)
version_type=linux # correct to gnu/linux during the next big refactor
need_lib_prefix=no
need_version=no
dynamic_linker="$host_os runtime_loader"
library_names_spec='${libname}${release}${shared_ext}$versuffix ${libname}${release}${shared_ext}${major} ${libname}${shared_ext}'
soname_spec='${libname}${release}${shared_ext}$major'
shlibpath_var=LIBRARY_PATH
shlibpath_overrides_runpath=yes
sys_lib_dlsearch_path_spec='/boot/home/config/lib /boot/common/lib /boot/system/lib'
hardcode_into_libs=yes
;;
hpux9* | hpux10* | hpux11*)
# Give a soname corresponding to the major version so that dld.sl refuses to
# link against other versions.
version_type=sunos
need_lib_prefix=no
need_version=no
case $host_cpu in
ia64*)
shrext_cmds='.so'
hardcode_into_libs=yes
dynamic_linker="$host_os dld.so"
shlibpath_var=LD_LIBRARY_PATH
shlibpath_overrides_runpath=yes # Unless +noenvvar is specified.
library_names_spec='${libname}${release}${shared_ext}$versuffix ${libname}${release}${shared_ext}$major $libname${shared_ext}'
soname_spec='${libname}${release}${shared_ext}$major'
if test "X$HPUX_IA64_MODE" = X32; then
sys_lib_search_path_spec="/usr/lib/hpux32 /usr/local/lib/hpux32 /usr/local/lib"
else
sys_lib_search_path_spec="/usr/lib/hpux64 /usr/local/lib/hpux64"
fi
sys_lib_dlsearch_path_spec=$sys_lib_search_path_spec
;;
hppa*64*)
shrext_cmds='.sl'
hardcode_into_libs=yes
dynamic_linker="$host_os dld.sl"
shlibpath_var=LD_LIBRARY_PATH # How should we handle SHLIB_PATH
shlibpath_overrides_runpath=yes # Unless +noenvvar is specified.
library_names_spec='${libname}${release}${shared_ext}$versuffix ${libname}${release}${shared_ext}$major $libname${shared_ext}'
soname_spec='${libname}${release}${shared_ext}$major'
sys_lib_search_path_spec="/usr/lib/pa20_64 /usr/ccs/lib/pa20_64"
sys_lib_dlsearch_path_spec=$sys_lib_search_path_spec
;;
*)
shrext_cmds='.sl'
dynamic_linker="$host_os dld.sl"
shlibpath_var=SHLIB_PATH
shlibpath_overrides_runpath=no # +s is required to enable SHLIB_PATH
library_names_spec='${libname}${release}${shared_ext}$versuffix ${libname}${release}${shared_ext}$major $libname${shared_ext}'
soname_spec='${libname}${release}${shared_ext}$major'
;;
esac
# HP-UX runs *really* slowly unless shared libraries are mode 555, ...
postinstall_cmds='chmod 555 $lib'
# or fails outright, so override atomically:
install_override_mode=555
;;
interix[[3-9]]*)
version_type=linux # correct to gnu/linux during the next big refactor
need_lib_prefix=no
need_version=no
library_names_spec='${libname}${release}${shared_ext}$versuffix ${libname}${release}${shared_ext}$major ${libname}${shared_ext}'
soname_spec='${libname}${release}${shared_ext}$major'
dynamic_linker='Interix 3.x ld.so.1 (PE, like ELF)'
shlibpath_var=LD_LIBRARY_PATH
shlibpath_overrides_runpath=no
hardcode_into_libs=yes
;;
irix5* | irix6* | nonstopux*)
case $host_os in
nonstopux*) version_type=nonstopux ;;
*)
if test "$lt_cv_prog_gnu_ld" = yes; then
version_type=linux # correct to gnu/linux during the next big refactor
else
version_type=irix
fi ;;
esac
need_lib_prefix=no
need_version=no
soname_spec='${libname}${release}${shared_ext}$major'
library_names_spec='${libname}${release}${shared_ext}$versuffix ${libname}${release}${shared_ext}$major ${libname}${release}${shared_ext} $libname${shared_ext}'
case $host_os in
irix5* | nonstopux*)
libsuff= shlibsuff=
;;
*)
case $LD in # libtool.m4 will add one of these switches to LD
*-32|*"-32 "|*-melf32bsmip|*"-melf32bsmip ")
libsuff= shlibsuff= libmagic=32-bit;;
*-n32|*"-n32 "|*-melf32bmipn32|*"-melf32bmipn32 ")
libsuff=32 shlibsuff=N32 libmagic=N32;;
*-64|*"-64 "|*-melf64bmip|*"-melf64bmip ")
libsuff=64 shlibsuff=64 libmagic=64-bit;;
*) libsuff= shlibsuff= libmagic=never-match;;
esac
;;
esac
shlibpath_var=LD_LIBRARY${shlibsuff}_PATH
shlibpath_overrides_runpath=no
sys_lib_search_path_spec="/usr/lib${libsuff} /lib${libsuff} /usr/local/lib${libsuff}"
sys_lib_dlsearch_path_spec="/usr/lib${libsuff} /lib${libsuff}"
hardcode_into_libs=yes
;;
# No shared lib support for Linux oldld, aout, or coff.
linux*oldld* | linux*aout* | linux*coff*)
dynamic_linker=no
;;
linux*android*)
version_type=none # Android doesn't support versioned libraries.
need_lib_prefix=no
need_version=no
library_names_spec='$libname$release$shared_ext'
soname_spec='$libname$release$shared_ext'
finish_cmds=
shlibpath_var=LD_LIBRARY_PATH
shlibpath_overrides_runpath=yes
# This implies no fast_install, which is unacceptable.
# Some rework will be needed to allow for fast_install
# before this can be enabled.
hardcode_into_libs=yes
dynamic_linker='Android linker'
# Don't embed -rpath directories since the linker doesn't support them.
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='-L$libdir'
;;
# This must be glibc/ELF.
linux* | k*bsd*-gnu | kopensolaris*-gnu)
version_type=linux # correct to gnu/linux during the next big refactor
need_lib_prefix=no
need_version=no
library_names_spec='${libname}${release}${shared_ext}$versuffix ${libname}${release}${shared_ext}$major $libname${shared_ext}'
soname_spec='${libname}${release}${shared_ext}$major'
finish_cmds='PATH="\$PATH:/sbin" ldconfig -n $libdir'
shlibpath_var=LD_LIBRARY_PATH
shlibpath_overrides_runpath=no
# Some binutils ld are patched to set DT_RUNPATH
AC_CACHE_VAL([lt_cv_shlibpath_overrides_runpath],
[lt_cv_shlibpath_overrides_runpath=no
save_LDFLAGS=$LDFLAGS
save_libdir=$libdir
eval "libdir=/foo; wl=\"$_LT_TAGVAR(lt_prog_compiler_wl, $1)\"; \
LDFLAGS=\"\$LDFLAGS $_LT_TAGVAR(hardcode_libdir_flag_spec, $1)\""
AC_LINK_IFELSE([AC_LANG_PROGRAM([],[])],
[AS_IF([ ($OBJDUMP -p conftest$ac_exeext) 2>/dev/null | grep "RUNPATH.*$libdir" >/dev/null],
[lt_cv_shlibpath_overrides_runpath=yes])])
LDFLAGS=$save_LDFLAGS
libdir=$save_libdir
])
shlibpath_overrides_runpath=$lt_cv_shlibpath_overrides_runpath
# This implies no fast_install, which is unacceptable.
# Some rework will be needed to allow for fast_install
# before this can be enabled.
hardcode_into_libs=yes
# Append ld.so.conf contents to the search path
if test -f /etc/ld.so.conf; then
lt_ld_extra=`awk '/^include / { system(sprintf("cd /etc; cat %s 2>/dev/null", \[$]2)); skip = 1; } { if (!skip) print \[$]0; skip = 0; }' < /etc/ld.so.conf | $SED -e 's/#.*//;/^[ ]*hwcap[ ]/d;s/[:, ]/ /g;s/=[^=]*$//;s/=[^= ]* / /g;s/"//g;/^$/d' | tr '\n' ' '`
sys_lib_dlsearch_path_spec="/lib /usr/lib $lt_ld_extra"
fi
# We used to test for /lib/ld.so.1 and disable shared libraries on
# powerpc, because MkLinux only supported shared libraries with the
# GNU dynamic linker. Since this was broken with cross compilers,
# most powerpc-linux boxes support dynamic linking these days and
# people can always --disable-shared, the test was removed, and we
# assume the GNU/Linux dynamic linker is in use.
dynamic_linker='GNU/Linux ld.so'
;;
netbsdelf*-gnu)
version_type=linux
need_lib_prefix=no
need_version=no
library_names_spec='${libname}${release}${shared_ext}$versuffix ${libname}${release}${shared_ext}$major ${libname}${shared_ext}'
soname_spec='${libname}${release}${shared_ext}$major'
shlibpath_var=LD_LIBRARY_PATH
shlibpath_overrides_runpath=no
hardcode_into_libs=yes
dynamic_linker='NetBSD ld.elf_so'
;;
netbsd*)
version_type=sunos
need_lib_prefix=no
need_version=no
if echo __ELF__ | $CC -E - | $GREP __ELF__ >/dev/null; then
library_names_spec='${libname}${release}${shared_ext}$versuffix ${libname}${shared_ext}$versuffix'
finish_cmds='PATH="\$PATH:/sbin" ldconfig -m $libdir'
dynamic_linker='NetBSD (a.out) ld.so'
else
library_names_spec='${libname}${release}${shared_ext}$versuffix ${libname}${release}${shared_ext}$major ${libname}${shared_ext}'
soname_spec='${libname}${release}${shared_ext}$major'
dynamic_linker='NetBSD ld.elf_so'
fi
shlibpath_var=LD_LIBRARY_PATH
shlibpath_overrides_runpath=yes
hardcode_into_libs=yes
;;
newsos6)
version_type=linux # correct to gnu/linux during the next big refactor
library_names_spec='${libname}${release}${shared_ext}$versuffix ${libname}${release}${shared_ext}$major $libname${shared_ext}'
shlibpath_var=LD_LIBRARY_PATH
shlibpath_overrides_runpath=yes
;;
*nto* | *qnx*)
version_type=qnx
need_lib_prefix=no
need_version=no
library_names_spec='${libname}${release}${shared_ext}$versuffix ${libname}${release}${shared_ext}$major $libname${shared_ext}'
soname_spec='${libname}${release}${shared_ext}$major'
shlibpath_var=LD_LIBRARY_PATH
shlibpath_overrides_runpath=no
hardcode_into_libs=yes
dynamic_linker='ldqnx.so'
;;
openbsd*)
version_type=sunos
sys_lib_dlsearch_path_spec="/usr/lib"
need_lib_prefix=no
# Some older versions of OpenBSD (3.3 at least) *do* need versioned libs.
case $host_os in
openbsd3.3 | openbsd3.3.*) need_version=yes ;;
*) need_version=no ;;
esac
library_names_spec='${libname}${release}${shared_ext}$versuffix ${libname}${shared_ext}$versuffix'
finish_cmds='PATH="\$PATH:/sbin" ldconfig -m $libdir'
shlibpath_var=LD_LIBRARY_PATH
if test -z "`echo __ELF__ | $CC -E - | $GREP __ELF__`" || test "$host_os-$host_cpu" = "openbsd2.8-powerpc"; then
case $host_os in
openbsd2.[[89]] | openbsd2.[[89]].*)
shlibpath_overrides_runpath=no
;;
*)
shlibpath_overrides_runpath=yes
;;
esac
else
shlibpath_overrides_runpath=yes
fi
;;
os2*)
libname_spec='$name'
shrext_cmds=".dll"
need_lib_prefix=no
library_names_spec='$libname${shared_ext} $libname.a'
dynamic_linker='OS/2 ld.exe'
shlibpath_var=LIBPATH
;;
osf3* | osf4* | osf5*)
version_type=osf
need_lib_prefix=no
need_version=no
soname_spec='${libname}${release}${shared_ext}$major'
library_names_spec='${libname}${release}${shared_ext}$versuffix ${libname}${release}${shared_ext}$major $libname${shared_ext}'
shlibpath_var=LD_LIBRARY_PATH
sys_lib_search_path_spec="/usr/shlib /usr/ccs/lib /usr/lib/cmplrs/cc /usr/lib /usr/local/lib /var/shlib"
sys_lib_dlsearch_path_spec="$sys_lib_search_path_spec"
;;
rdos*)
dynamic_linker=no
;;
solaris*)
version_type=linux # correct to gnu/linux during the next big refactor
need_lib_prefix=no
need_version=no
library_names_spec='${libname}${release}${shared_ext}$versuffix ${libname}${release}${shared_ext}$major $libname${shared_ext}'
soname_spec='${libname}${release}${shared_ext}$major'
shlibpath_var=LD_LIBRARY_PATH
shlibpath_overrides_runpath=yes
hardcode_into_libs=yes
# ldd complains unless libraries are executable
postinstall_cmds='chmod +x $lib'
;;
sunos4*)
version_type=sunos
library_names_spec='${libname}${release}${shared_ext}$versuffix ${libname}${shared_ext}$versuffix'
finish_cmds='PATH="\$PATH:/usr/etc" ldconfig $libdir'
shlibpath_var=LD_LIBRARY_PATH
shlibpath_overrides_runpath=yes
if test "$with_gnu_ld" = yes; then
need_lib_prefix=no
fi
need_version=yes
;;
sysv4 | sysv4.3*)
version_type=linux # correct to gnu/linux during the next big refactor
library_names_spec='${libname}${release}${shared_ext}$versuffix ${libname}${release}${shared_ext}$major $libname${shared_ext}'
soname_spec='${libname}${release}${shared_ext}$major'
shlibpath_var=LD_LIBRARY_PATH
case $host_vendor in
sni)
shlibpath_overrides_runpath=no
need_lib_prefix=no
runpath_var=LD_RUN_PATH
;;
siemens)
need_lib_prefix=no
;;
motorola)
need_lib_prefix=no
need_version=no
shlibpath_overrides_runpath=no
sys_lib_search_path_spec='/lib /usr/lib /usr/ccs/lib'
;;
esac
;;
sysv4*MP*)
if test -d /usr/nec ;then
version_type=linux # correct to gnu/linux during the next big refactor
library_names_spec='$libname${shared_ext}.$versuffix $libname${shared_ext}.$major $libname${shared_ext}'
soname_spec='$libname${shared_ext}.$major'
shlibpath_var=LD_LIBRARY_PATH
fi
;;
sysv5* | sco3.2v5* | sco5v6* | unixware* | OpenUNIX* | sysv4*uw2*)
version_type=freebsd-elf
need_lib_prefix=no
need_version=no
library_names_spec='${libname}${release}${shared_ext}$versuffix ${libname}${release}${shared_ext} $libname${shared_ext}'
soname_spec='${libname}${release}${shared_ext}$major'
shlibpath_var=LD_LIBRARY_PATH
shlibpath_overrides_runpath=yes
hardcode_into_libs=yes
if test "$with_gnu_ld" = yes; then
sys_lib_search_path_spec='/usr/local/lib /usr/gnu/lib /usr/ccs/lib /usr/lib /lib'
else
sys_lib_search_path_spec='/usr/ccs/lib /usr/lib'
case $host_os in
sco3.2v5*)
sys_lib_search_path_spec="$sys_lib_search_path_spec /lib"
;;
esac
fi
sys_lib_dlsearch_path_spec='/usr/lib'
;;
tpf*)
# TPF is a cross-target only. Preferred cross-host = GNU/Linux.
version_type=linux # correct to gnu/linux during the next big refactor
need_lib_prefix=no
need_version=no
library_names_spec='${libname}${release}${shared_ext}$versuffix ${libname}${release}${shared_ext}$major $libname${shared_ext}'
shlibpath_var=LD_LIBRARY_PATH
shlibpath_overrides_runpath=no
hardcode_into_libs=yes
;;
uts4*)
version_type=linux # correct to gnu/linux during the next big refactor
library_names_spec='${libname}${release}${shared_ext}$versuffix ${libname}${release}${shared_ext}$major $libname${shared_ext}'
soname_spec='${libname}${release}${shared_ext}$major'
shlibpath_var=LD_LIBRARY_PATH
;;
*)
dynamic_linker=no
;;
esac
AC_MSG_RESULT([$dynamic_linker])
test "$dynamic_linker" = no && can_build_shared=no
variables_saved_for_relink="PATH $shlibpath_var $runpath_var"
if test "$GCC" = yes; then
variables_saved_for_relink="$variables_saved_for_relink GCC_EXEC_PREFIX COMPILER_PATH LIBRARY_PATH"
fi
if test "${lt_cv_sys_lib_search_path_spec+set}" = set; then
sys_lib_search_path_spec="$lt_cv_sys_lib_search_path_spec"
fi
if test "${lt_cv_sys_lib_dlsearch_path_spec+set}" = set; then
sys_lib_dlsearch_path_spec="$lt_cv_sys_lib_dlsearch_path_spec"
fi
_LT_DECL([], [variables_saved_for_relink], [1],
[Variables whose values should be saved in libtool wrapper scripts and
restored at link time])
_LT_DECL([], [need_lib_prefix], [0],
[Do we need the "lib" prefix for modules?])
_LT_DECL([], [need_version], [0], [Do we need a version for libraries?])
_LT_DECL([], [version_type], [0], [Library versioning type])
_LT_DECL([], [runpath_var], [0], [Shared library runtime path variable])
_LT_DECL([], [shlibpath_var], [0],[Shared library path variable])
_LT_DECL([], [shlibpath_overrides_runpath], [0],
[Is shlibpath searched before the hard-coded library search path?])
_LT_DECL([], [libname_spec], [1], [Format of library name prefix])
_LT_DECL([], [library_names_spec], [1],
[[List of archive names. First name is the real one, the rest are links.
The last name is the one that the linker finds with -lNAME]])
_LT_DECL([], [soname_spec], [1],
[[The coded name of the library, if different from the real name]])
_LT_DECL([], [install_override_mode], [1],
[Permission mode override for installation of shared libraries])
_LT_DECL([], [postinstall_cmds], [2],
[Command to use after installation of a shared archive])
_LT_DECL([], [postuninstall_cmds], [2],
[Command to use after uninstallation of a shared archive])
_LT_DECL([], [finish_cmds], [2],
[Commands used to finish a libtool library installation in a directory])
_LT_DECL([], [finish_eval], [1],
[[As "finish_cmds", except a single script fragment to be evaled but
not shown]])
_LT_DECL([], [hardcode_into_libs], [0],
[Whether we should hardcode library paths into libraries])
_LT_DECL([], [sys_lib_search_path_spec], [2],
[Compile-time system search path for libraries])
_LT_DECL([], [sys_lib_dlsearch_path_spec], [2],
[Run-time system search path for libraries])
])# _LT_SYS_DYNAMIC_LINKER
# _LT_PATH_TOOL_PREFIX(TOOL)
# --------------------------
# find a file program which can recognize shared library
AC_DEFUN([_LT_PATH_TOOL_PREFIX],
[m4_require([_LT_DECL_EGREP])dnl
AC_MSG_CHECKING([for $1])
AC_CACHE_VAL(lt_cv_path_MAGIC_CMD,
[case $MAGIC_CMD in
[[\\/*] | ?:[\\/]*])
lt_cv_path_MAGIC_CMD="$MAGIC_CMD" # Let the user override the test with a path.
;;
*)
lt_save_MAGIC_CMD="$MAGIC_CMD"
lt_save_ifs="$IFS"; IFS=$PATH_SEPARATOR
dnl $ac_dummy forces splitting on constant user-supplied paths.
dnl POSIX.2 word splitting is done only on the output of word expansions,
dnl not every word. This closes a longstanding sh security hole.
ac_dummy="m4_if([$2], , $PATH, [$2])"
for ac_dir in $ac_dummy; do
IFS="$lt_save_ifs"
test -z "$ac_dir" && ac_dir=.
if test -f $ac_dir/$1; then
lt_cv_path_MAGIC_CMD="$ac_dir/$1"
if test -n "$file_magic_test_file"; then
case $deplibs_check_method in
"file_magic "*)
file_magic_regex=`expr "$deplibs_check_method" : "file_magic \(.*\)"`
MAGIC_CMD="$lt_cv_path_MAGIC_CMD"
if eval $file_magic_cmd \$file_magic_test_file 2> /dev/null |
$EGREP "$file_magic_regex" > /dev/null; then
:
else
cat <<_LT_EOF 1>&2
*** Warning: the command libtool uses to detect shared libraries,
*** $file_magic_cmd, produces output that libtool cannot recognize.
*** The result is that libtool may fail to recognize shared libraries
*** as such. This will affect the creation of libtool libraries that
*** depend on shared libraries, but programs linked with such libtool
*** libraries will work regardless of this problem. Nevertheless, you
*** may want to report the problem to your system manager and/or to
*** bug-libtool@gnu.org
_LT_EOF
fi ;;
esac
fi
break
fi
done
IFS="$lt_save_ifs"
MAGIC_CMD="$lt_save_MAGIC_CMD"
;;
esac])
MAGIC_CMD="$lt_cv_path_MAGIC_CMD"
if test -n "$MAGIC_CMD"; then
AC_MSG_RESULT($MAGIC_CMD)
else
AC_MSG_RESULT(no)
fi
_LT_DECL([], [MAGIC_CMD], [0],
[Used to examine libraries when file_magic_cmd begins with "file"])dnl
])# _LT_PATH_TOOL_PREFIX
# Old name:
AU_ALIAS([AC_PATH_TOOL_PREFIX], [_LT_PATH_TOOL_PREFIX])
dnl aclocal-1.4 backwards compatibility:
dnl AC_DEFUN([AC_PATH_TOOL_PREFIX], [])
# _LT_PATH_MAGIC
# --------------
# find a file program which can recognize a shared library
m4_defun([_LT_PATH_MAGIC],
[_LT_PATH_TOOL_PREFIX(${ac_tool_prefix}file, /usr/bin$PATH_SEPARATOR$PATH)
if test -z "$lt_cv_path_MAGIC_CMD"; then
if test -n "$ac_tool_prefix"; then
_LT_PATH_TOOL_PREFIX(file, /usr/bin$PATH_SEPARATOR$PATH)
else
MAGIC_CMD=:
fi
fi
])# _LT_PATH_MAGIC
# LT_PATH_LD
# ----------
# find the pathname to the GNU or non-GNU linker
AC_DEFUN([LT_PATH_LD],
[AC_REQUIRE([AC_PROG_CC])dnl
AC_REQUIRE([AC_CANONICAL_HOST])dnl
AC_REQUIRE([AC_CANONICAL_BUILD])dnl
m4_require([_LT_DECL_SED])dnl
m4_require([_LT_DECL_EGREP])dnl
m4_require([_LT_PROG_ECHO_BACKSLASH])dnl
AC_ARG_WITH([gnu-ld],
[AS_HELP_STRING([--with-gnu-ld],
[assume the C compiler uses GNU ld @<:@default=no@:>@])],
[test "$withval" = no || with_gnu_ld=yes],
[with_gnu_ld=no])dnl
ac_prog=ld
if test "$GCC" = yes; then
# Check if gcc -print-prog-name=ld gives a path.
AC_MSG_CHECKING([for ld used by $CC])
case $host in
*-*-mingw*)
# gcc leaves a trailing carriage return which upsets mingw
ac_prog=`($CC -print-prog-name=ld) 2>&5 | tr -d '\015'` ;;
*)
ac_prog=`($CC -print-prog-name=ld) 2>&5` ;;
esac
case $ac_prog in
# Accept absolute paths.
[[\\/]]* | ?:[[\\/]]*)
re_direlt='/[[^/]][[^/]]*/\.\./'
# Canonicalize the pathname of ld
ac_prog=`$ECHO "$ac_prog"| $SED 's%\\\\%/%g'`
while $ECHO "$ac_prog" | $GREP "$re_direlt" > /dev/null 2>&1; do
ac_prog=`$ECHO $ac_prog| $SED "s%$re_direlt%/%"`
done
test -z "$LD" && LD="$ac_prog"
;;
"")
# If it fails, then pretend we aren't using GCC.
ac_prog=ld
;;
*)
# If it is relative, then search for the first ld in PATH.
with_gnu_ld=unknown
;;
esac
elif test "$with_gnu_ld" = yes; then
AC_MSG_CHECKING([for GNU ld])
else
AC_MSG_CHECKING([for non-GNU ld])
fi
AC_CACHE_VAL(lt_cv_path_LD,
[if test -z "$LD"; then
lt_save_ifs="$IFS"; IFS=$PATH_SEPARATOR
for ac_dir in $PATH; do
IFS="$lt_save_ifs"
test -z "$ac_dir" && ac_dir=.
if test -f "$ac_dir/$ac_prog" || test -f "$ac_dir/$ac_prog$ac_exeext"; then
lt_cv_path_LD="$ac_dir/$ac_prog"
# Check to see if the program is GNU ld. I'd rather use --version,
# but apparently some variants of GNU ld only accept -v.
# Break only if it was the GNU/non-GNU ld that we prefer.
case `"$lt_cv_path_LD" -v 2>&1 &1 /dev/null 2>&1; then
lt_cv_deplibs_check_method='file_magic ^x86 archive import|^x86 DLL'
lt_cv_file_magic_cmd='func_win32_libid'
else
# Keep this pattern in sync with the one in func_win32_libid.
lt_cv_deplibs_check_method='file_magic file format (pei*-i386(.*architecture: i386)?|pe-arm-wince|pe-x86-64)'
lt_cv_file_magic_cmd='$OBJDUMP -f'
fi
;;
cegcc*)
# use the weaker test based on 'objdump'. See mingw*.
lt_cv_deplibs_check_method='file_magic file format pe-arm-.*little(.*architecture: arm)?'
lt_cv_file_magic_cmd='$OBJDUMP -f'
;;
darwin* | rhapsody*)
lt_cv_deplibs_check_method=pass_all
;;
freebsd* | dragonfly*)
if echo __ELF__ | $CC -E - | $GREP __ELF__ > /dev/null; then
case $host_cpu in
i*86 )
# Not sure whether the presence of OpenBSD here was a mistake.
# Let's accept both of them until this is cleared up.
lt_cv_deplibs_check_method='file_magic (FreeBSD|OpenBSD|DragonFly)/i[[3-9]]86 (compact )?demand paged shared library'
lt_cv_file_magic_cmd=/usr/bin/file
lt_cv_file_magic_test_file=`echo /usr/lib/libc.so.*`
;;
esac
else
lt_cv_deplibs_check_method=pass_all
fi
;;
gnu*)
lt_cv_deplibs_check_method=pass_all
;;
haiku*)
lt_cv_deplibs_check_method=pass_all
;;
hpux10.20* | hpux11*)
lt_cv_file_magic_cmd=/usr/bin/file
case $host_cpu in
ia64*)
lt_cv_deplibs_check_method='file_magic (s[[0-9]][[0-9]][[0-9]]|ELF-[[0-9]][[0-9]]) shared object file - IA64'
lt_cv_file_magic_test_file=/usr/lib/hpux32/libc.so
;;
hppa*64*)
[lt_cv_deplibs_check_method='file_magic (s[0-9][0-9][0-9]|ELF[ -][0-9][0-9])(-bit)?( [LM]SB)? shared object( file)?[, -]* PA-RISC [0-9]\.[0-9]']
lt_cv_file_magic_test_file=/usr/lib/pa20_64/libc.sl
;;
*)
lt_cv_deplibs_check_method='file_magic (s[[0-9]][[0-9]][[0-9]]|PA-RISC[[0-9]]\.[[0-9]]) shared library'
lt_cv_file_magic_test_file=/usr/lib/libc.sl
;;
esac
;;
interix[[3-9]]*)
# PIC code is broken on Interix 3.x, that's why |\.a not |_pic\.a here
lt_cv_deplibs_check_method='match_pattern /lib[[^/]]+(\.so|\.a)$'
;;
irix5* | irix6* | nonstopux*)
case $LD in
*-32|*"-32 ") libmagic=32-bit;;
*-n32|*"-n32 ") libmagic=N32;;
*-64|*"-64 ") libmagic=64-bit;;
*) libmagic=never-match;;
esac
lt_cv_deplibs_check_method=pass_all
;;
# This must be glibc/ELF.
linux* | k*bsd*-gnu | kopensolaris*-gnu)
lt_cv_deplibs_check_method=pass_all
;;
netbsd* | netbsdelf*-gnu)
if echo __ELF__ | $CC -E - | $GREP __ELF__ > /dev/null; then
lt_cv_deplibs_check_method='match_pattern /lib[[^/]]+(\.so\.[[0-9]]+\.[[0-9]]+|_pic\.a)$'
else
lt_cv_deplibs_check_method='match_pattern /lib[[^/]]+(\.so|_pic\.a)$'
fi
;;
newos6*)
lt_cv_deplibs_check_method='file_magic ELF [[0-9]][[0-9]]*-bit [[ML]]SB (executable|dynamic lib)'
lt_cv_file_magic_cmd=/usr/bin/file
lt_cv_file_magic_test_file=/usr/lib/libnls.so
;;
*nto* | *qnx*)
lt_cv_deplibs_check_method=pass_all
;;
openbsd*)
if test -z "`echo __ELF__ | $CC -E - | $GREP __ELF__`" || test "$host_os-$host_cpu" = "openbsd2.8-powerpc"; then
lt_cv_deplibs_check_method='match_pattern /lib[[^/]]+(\.so\.[[0-9]]+\.[[0-9]]+|\.so|_pic\.a)$'
else
lt_cv_deplibs_check_method='match_pattern /lib[[^/]]+(\.so\.[[0-9]]+\.[[0-9]]+|_pic\.a)$'
fi
;;
osf3* | osf4* | osf5*)
lt_cv_deplibs_check_method=pass_all
;;
rdos*)
lt_cv_deplibs_check_method=pass_all
;;
solaris*)
lt_cv_deplibs_check_method=pass_all
;;
sysv5* | sco3.2v5* | sco5v6* | unixware* | OpenUNIX* | sysv4*uw2*)
lt_cv_deplibs_check_method=pass_all
;;
sysv4 | sysv4.3*)
case $host_vendor in
motorola)
lt_cv_deplibs_check_method='file_magic ELF [[0-9]][[0-9]]*-bit [[ML]]SB (shared object|dynamic lib) M[[0-9]][[0-9]]* Version [[0-9]]'
lt_cv_file_magic_test_file=`echo /usr/lib/libc.so*`
;;
ncr)
lt_cv_deplibs_check_method=pass_all
;;
sequent)
lt_cv_file_magic_cmd='/bin/file'
lt_cv_deplibs_check_method='file_magic ELF [[0-9]][[0-9]]*-bit [[LM]]SB (shared object|dynamic lib )'
;;
sni)
lt_cv_file_magic_cmd='/bin/file'
lt_cv_deplibs_check_method="file_magic ELF [[0-9]][[0-9]]*-bit [[LM]]SB dynamic lib"
lt_cv_file_magic_test_file=/lib/libc.so
;;
siemens)
lt_cv_deplibs_check_method=pass_all
;;
pc)
lt_cv_deplibs_check_method=pass_all
;;
esac
;;
tpf*)
lt_cv_deplibs_check_method=pass_all
;;
esac
])
file_magic_glob=
want_nocaseglob=no
if test "$build" = "$host"; then
case $host_os in
mingw* | pw32*)
if ( shopt | grep nocaseglob ) >/dev/null 2>&1; then
want_nocaseglob=yes
else
file_magic_glob=`echo aAbBcCdDeEfFgGhHiIjJkKlLmMnNoOpPqQrRsStTuUvVwWxXyYzZ | $SED -e "s/\(..\)/s\/[[\1]]\/[[\1]]\/g;/g"`
fi
;;
esac
fi
file_magic_cmd=$lt_cv_file_magic_cmd
deplibs_check_method=$lt_cv_deplibs_check_method
test -z "$deplibs_check_method" && deplibs_check_method=unknown
_LT_DECL([], [deplibs_check_method], [1],
[Method to check whether dependent libraries are shared objects])
_LT_DECL([], [file_magic_cmd], [1],
[Command to use when deplibs_check_method = "file_magic"])
_LT_DECL([], [file_magic_glob], [1],
[How to find potential files when deplibs_check_method = "file_magic"])
_LT_DECL([], [want_nocaseglob], [1],
[Find potential files using nocaseglob when deplibs_check_method = "file_magic"])
])# _LT_CHECK_MAGIC_METHOD
# LT_PATH_NM
# ----------
# find the pathname to a BSD- or MS-compatible name lister
AC_DEFUN([LT_PATH_NM],
[AC_REQUIRE([AC_PROG_CC])dnl
AC_CACHE_CHECK([for BSD- or MS-compatible name lister (nm)], lt_cv_path_NM,
[if test -n "$NM"; then
# Let the user override the test.
lt_cv_path_NM="$NM"
else
lt_nm_to_check="${ac_tool_prefix}nm"
if test -n "$ac_tool_prefix" && test "$build" = "$host"; then
lt_nm_to_check="$lt_nm_to_check nm"
fi
for lt_tmp_nm in $lt_nm_to_check; do
lt_save_ifs="$IFS"; IFS=$PATH_SEPARATOR
for ac_dir in $PATH /usr/ccs/bin/elf /usr/ccs/bin /usr/ucb /bin; do
IFS="$lt_save_ifs"
test -z "$ac_dir" && ac_dir=.
tmp_nm="$ac_dir/$lt_tmp_nm"
if test -f "$tmp_nm" || test -f "$tmp_nm$ac_exeext" ; then
# Check to see if the nm accepts a BSD-compat flag.
# Adding the `sed 1q' prevents false positives on HP-UX, which says:
# nm: unknown option "B" ignored
# Tru64's nm complains that /dev/null is an invalid object file
case `"$tmp_nm" -B /dev/null 2>&1 | sed '1q'` in
*/dev/null* | *'Invalid file or object type'*)
lt_cv_path_NM="$tmp_nm -B"
break
;;
*)
case `"$tmp_nm" -p /dev/null 2>&1 | sed '1q'` in
*/dev/null*)
lt_cv_path_NM="$tmp_nm -p"
break
;;
*)
lt_cv_path_NM=${lt_cv_path_NM="$tmp_nm"} # keep the first match, but
continue # so that we can try to find one that supports BSD flags
;;
esac
;;
esac
fi
done
IFS="$lt_save_ifs"
done
: ${lt_cv_path_NM=no}
fi])
if test "$lt_cv_path_NM" != "no"; then
NM="$lt_cv_path_NM"
else
# Didn't find any BSD compatible name lister, look for dumpbin.
if test -n "$DUMPBIN"; then :
# Let the user override the test.
else
AC_CHECK_TOOLS(DUMPBIN, [dumpbin "link -dump"], :)
case `$DUMPBIN -symbols /dev/null 2>&1 | sed '1q'` in
*COFF*)
DUMPBIN="$DUMPBIN -symbols"
;;
*)
DUMPBIN=:
;;
esac
fi
AC_SUBST([DUMPBIN])
if test "$DUMPBIN" != ":"; then
NM="$DUMPBIN"
fi
fi
test -z "$NM" && NM=nm
AC_SUBST([NM])
_LT_DECL([], [NM], [1], [A BSD- or MS-compatible name lister])dnl
AC_CACHE_CHECK([the name lister ($NM) interface], [lt_cv_nm_interface],
[lt_cv_nm_interface="BSD nm"
echo "int some_variable = 0;" > conftest.$ac_ext
(eval echo "\"\$as_me:$LINENO: $ac_compile\"" >&AS_MESSAGE_LOG_FD)
(eval "$ac_compile" 2>conftest.err)
cat conftest.err >&AS_MESSAGE_LOG_FD
(eval echo "\"\$as_me:$LINENO: $NM \\\"conftest.$ac_objext\\\"\"" >&AS_MESSAGE_LOG_FD)
(eval "$NM \"conftest.$ac_objext\"" 2>conftest.err > conftest.out)
cat conftest.err >&AS_MESSAGE_LOG_FD
(eval echo "\"\$as_me:$LINENO: output\"" >&AS_MESSAGE_LOG_FD)
cat conftest.out >&AS_MESSAGE_LOG_FD
if $GREP 'External.*some_variable' conftest.out > /dev/null; then
lt_cv_nm_interface="MS dumpbin"
fi
rm -f conftest*])
])# LT_PATH_NM
# Old names:
AU_ALIAS([AM_PROG_NM], [LT_PATH_NM])
AU_ALIAS([AC_PROG_NM], [LT_PATH_NM])
dnl aclocal-1.4 backwards compatibility:
dnl AC_DEFUN([AM_PROG_NM], [])
dnl AC_DEFUN([AC_PROG_NM], [])
# _LT_CHECK_SHAREDLIB_FROM_LINKLIB
# --------------------------------
# how to determine the name of the shared library
# associated with a specific link library.
# -- PORTME fill in with the dynamic library characteristics
m4_defun([_LT_CHECK_SHAREDLIB_FROM_LINKLIB],
[m4_require([_LT_DECL_EGREP])
m4_require([_LT_DECL_OBJDUMP])
m4_require([_LT_DECL_DLLTOOL])
AC_CACHE_CHECK([how to associate runtime and link libraries],
lt_cv_sharedlib_from_linklib_cmd,
[lt_cv_sharedlib_from_linklib_cmd='unknown'
case $host_os in
cygwin* | mingw* | pw32* | cegcc*)
# two different shell functions defined in ltmain.sh
# decide which to use based on capabilities of $DLLTOOL
case `$DLLTOOL --help 2>&1` in
*--identify-strict*)
lt_cv_sharedlib_from_linklib_cmd=func_cygming_dll_for_implib
;;
*)
lt_cv_sharedlib_from_linklib_cmd=func_cygming_dll_for_implib_fallback
;;
esac
;;
*)
# fallback: assume linklib IS sharedlib
lt_cv_sharedlib_from_linklib_cmd="$ECHO"
;;
esac
])
sharedlib_from_linklib_cmd=$lt_cv_sharedlib_from_linklib_cmd
test -z "$sharedlib_from_linklib_cmd" && sharedlib_from_linklib_cmd=$ECHO
_LT_DECL([], [sharedlib_from_linklib_cmd], [1],
[Command to associate shared and link libraries])
])# _LT_CHECK_SHAREDLIB_FROM_LINKLIB
# _LT_PATH_MANIFEST_TOOL
# ----------------------
# locate the manifest tool
m4_defun([_LT_PATH_MANIFEST_TOOL],
[AC_CHECK_TOOL(MANIFEST_TOOL, mt, :)
test -z "$MANIFEST_TOOL" && MANIFEST_TOOL=mt
AC_CACHE_CHECK([if $MANIFEST_TOOL is a manifest tool], [lt_cv_path_mainfest_tool],
[lt_cv_path_mainfest_tool=no
echo "$as_me:$LINENO: $MANIFEST_TOOL '-?'" >&AS_MESSAGE_LOG_FD
$MANIFEST_TOOL '-?' 2>conftest.err > conftest.out
cat conftest.err >&AS_MESSAGE_LOG_FD
if $GREP 'Manifest Tool' conftest.out > /dev/null; then
lt_cv_path_mainfest_tool=yes
fi
rm -f conftest*])
if test "x$lt_cv_path_mainfest_tool" != xyes; then
MANIFEST_TOOL=:
fi
_LT_DECL([], [MANIFEST_TOOL], [1], [Manifest tool])dnl
])# _LT_PATH_MANIFEST_TOOL
# LT_LIB_M
# --------
# check for math library
AC_DEFUN([LT_LIB_M],
[AC_REQUIRE([AC_CANONICAL_HOST])dnl
LIBM=
case $host in
*-*-beos* | *-*-cegcc* | *-*-cygwin* | *-*-haiku* | *-*-pw32* | *-*-darwin*)
# These system don't have libm, or don't need it
;;
*-ncr-sysv4.3*)
AC_CHECK_LIB(mw, _mwvalidcheckl, LIBM="-lmw")
AC_CHECK_LIB(m, cos, LIBM="$LIBM -lm")
;;
*)
AC_CHECK_LIB(m, cos, LIBM="-lm")
;;
esac
AC_SUBST([LIBM])
])# LT_LIB_M
# Old name:
AU_ALIAS([AC_CHECK_LIBM], [LT_LIB_M])
dnl aclocal-1.4 backwards compatibility:
dnl AC_DEFUN([AC_CHECK_LIBM], [])
# _LT_COMPILER_NO_RTTI([TAGNAME])
# -------------------------------
m4_defun([_LT_COMPILER_NO_RTTI],
[m4_require([_LT_TAG_COMPILER])dnl
_LT_TAGVAR(lt_prog_compiler_no_builtin_flag, $1)=
if test "$GCC" = yes; then
case $cc_basename in
nvcc*)
_LT_TAGVAR(lt_prog_compiler_no_builtin_flag, $1)=' -Xcompiler -fno-builtin' ;;
*)
_LT_TAGVAR(lt_prog_compiler_no_builtin_flag, $1)=' -fno-builtin' ;;
esac
_LT_COMPILER_OPTION([if $compiler supports -fno-rtti -fno-exceptions],
lt_cv_prog_compiler_rtti_exceptions,
[-fno-rtti -fno-exceptions], [],
[_LT_TAGVAR(lt_prog_compiler_no_builtin_flag, $1)="$_LT_TAGVAR(lt_prog_compiler_no_builtin_flag, $1) -fno-rtti -fno-exceptions"])
fi
_LT_TAGDECL([no_builtin_flag], [lt_prog_compiler_no_builtin_flag], [1],
[Compiler flag to turn off builtin functions])
])# _LT_COMPILER_NO_RTTI
# _LT_CMD_GLOBAL_SYMBOLS
# ----------------------
m4_defun([_LT_CMD_GLOBAL_SYMBOLS],
[AC_REQUIRE([AC_CANONICAL_HOST])dnl
AC_REQUIRE([AC_PROG_CC])dnl
AC_REQUIRE([AC_PROG_AWK])dnl
AC_REQUIRE([LT_PATH_NM])dnl
AC_REQUIRE([LT_PATH_LD])dnl
m4_require([_LT_DECL_SED])dnl
m4_require([_LT_DECL_EGREP])dnl
m4_require([_LT_TAG_COMPILER])dnl
# Check for command to grab the raw symbol name followed by C symbol from nm.
AC_MSG_CHECKING([command to parse $NM output from $compiler object])
AC_CACHE_VAL([lt_cv_sys_global_symbol_pipe],
[
# These are sane defaults that work on at least a few old systems.
# [They come from Ultrix. What could be older than Ultrix?!! ;)]
# Character class describing NM global symbol codes.
symcode='[[BCDEGRST]]'
# Regexp to match symbols that can be accessed directly from C.
sympat='\([[_A-Za-z]][[_A-Za-z0-9]]*\)'
# Define system-specific variables.
case $host_os in
aix*)
symcode='[[BCDT]]'
;;
cygwin* | mingw* | pw32* | cegcc*)
symcode='[[ABCDGISTW]]'
;;
hpux*)
if test "$host_cpu" = ia64; then
symcode='[[ABCDEGRST]]'
fi
;;
irix* | nonstopux*)
symcode='[[BCDEGRST]]'
;;
osf*)
symcode='[[BCDEGQRST]]'
;;
solaris*)
symcode='[[BDRT]]'
;;
sco3.2v5*)
symcode='[[DT]]'
;;
sysv4.2uw2*)
symcode='[[DT]]'
;;
sysv5* | sco5v6* | unixware* | OpenUNIX*)
symcode='[[ABDT]]'
;;
sysv4)
symcode='[[DFNSTU]]'
;;
esac
# If we're using GNU nm, then use its standard symbol codes.
case `$NM -V 2>&1` in
*GNU* | *'with BFD'*)
symcode='[[ABCDGIRSTW]]' ;;
esac
# Transform an extracted symbol line into a proper C declaration.
# Some systems (esp. on ia64) link data and code symbols differently,
# so use this general approach.
lt_cv_sys_global_symbol_to_cdecl="sed -n -e 's/^T .* \(.*\)$/extern int \1();/p' -e 's/^$symcode* .* \(.*\)$/extern char \1;/p'"
# Transform an extracted symbol line into symbol name and symbol address
lt_cv_sys_global_symbol_to_c_name_address="sed -n -e 's/^: \([[^ ]]*\)[[ ]]*$/ {\\\"\1\\\", (void *) 0},/p' -e 's/^$symcode* \([[^ ]]*\) \([[^ ]]*\)$/ {\"\2\", (void *) \&\2},/p'"
lt_cv_sys_global_symbol_to_c_name_address_lib_prefix="sed -n -e 's/^: \([[^ ]]*\)[[ ]]*$/ {\\\"\1\\\", (void *) 0},/p' -e 's/^$symcode* \([[^ ]]*\) \(lib[[^ ]]*\)$/ {\"\2\", (void *) \&\2},/p' -e 's/^$symcode* \([[^ ]]*\) \([[^ ]]*\)$/ {\"lib\2\", (void *) \&\2},/p'"
# Handle CRLF in mingw tool chain
opt_cr=
case $build_os in
mingw*)
opt_cr=`$ECHO 'x\{0,1\}' | tr x '\015'` # option cr in regexp
;;
esac
# Try without a prefix underscore, then with it.
for ac_symprfx in "" "_"; do
# Transform symcode, sympat, and symprfx into a raw symbol and a C symbol.
symxfrm="\\1 $ac_symprfx\\2 \\2"
# Write the raw and C identifiers.
if test "$lt_cv_nm_interface" = "MS dumpbin"; then
# Fake it for dumpbin and say T for any non-static function
# and D for any global variable.
# Also find C++ and __fastcall symbols from MSVC++,
# which start with @ or ?.
lt_cv_sys_global_symbol_pipe="$AWK ['"\
" {last_section=section; section=\$ 3};"\
" /^COFF SYMBOL TABLE/{for(i in hide) delete hide[i]};"\
" /Section length .*#relocs.*(pick any)/{hide[last_section]=1};"\
" \$ 0!~/External *\|/{next};"\
" / 0+ UNDEF /{next}; / UNDEF \([^|]\)*()/{next};"\
" {if(hide[section]) next};"\
" {f=0}; \$ 0~/\(\).*\|/{f=1}; {printf f ? \"T \" : \"D \"};"\
" {split(\$ 0, a, /\||\r/); split(a[2], s)};"\
" s[1]~/^[@?]/{print s[1], s[1]; next};"\
" s[1]~prfx {split(s[1],t,\"@\"); print t[1], substr(t[1],length(prfx))}"\
" ' prfx=^$ac_symprfx]"
else
lt_cv_sys_global_symbol_pipe="sed -n -e 's/^.*[[ ]]\($symcode$symcode*\)[[ ]][[ ]]*$ac_symprfx$sympat$opt_cr$/$symxfrm/p'"
fi
lt_cv_sys_global_symbol_pipe="$lt_cv_sys_global_symbol_pipe | sed '/ __gnu_lto/d'"
# Check to see that the pipe works correctly.
pipe_works=no
rm -f conftest*
cat > conftest.$ac_ext <<_LT_EOF
#ifdef __cplusplus
extern "C" {
#endif
char nm_test_var;
void nm_test_func(void);
void nm_test_func(void){}
#ifdef __cplusplus
}
#endif
int main(){nm_test_var='a';nm_test_func();return(0);}
_LT_EOF
if AC_TRY_EVAL(ac_compile); then
# Now try to grab the symbols.
nlist=conftest.nm
if AC_TRY_EVAL(NM conftest.$ac_objext \| "$lt_cv_sys_global_symbol_pipe" \> $nlist) && test -s "$nlist"; then
# Try sorting and uniquifying the output.
if sort "$nlist" | uniq > "$nlist"T; then
mv -f "$nlist"T "$nlist"
else
rm -f "$nlist"T
fi
# Make sure that we snagged all the symbols we need.
if $GREP ' nm_test_var$' "$nlist" >/dev/null; then
if $GREP ' nm_test_func$' "$nlist" >/dev/null; then
cat <<_LT_EOF > conftest.$ac_ext
/* Keep this code in sync between libtool.m4, ltmain, lt_system.h, and tests. */
#if defined(_WIN32) || defined(__CYGWIN__) || defined(_WIN32_WCE)
/* DATA imports from DLLs on WIN32 con't be const, because runtime
relocations are performed -- see ld's documentation on pseudo-relocs. */
# define LT@&t@_DLSYM_CONST
#elif defined(__osf__)
/* This system does not cope well with relocations in const data. */
# define LT@&t@_DLSYM_CONST
#else
# define LT@&t@_DLSYM_CONST const
#endif
#ifdef __cplusplus
extern "C" {
#endif
_LT_EOF
# Now generate the symbol file.
eval "$lt_cv_sys_global_symbol_to_cdecl"' < "$nlist" | $GREP -v main >> conftest.$ac_ext'
cat <<_LT_EOF >> conftest.$ac_ext
/* The mapping between symbol names and symbols. */
LT@&t@_DLSYM_CONST struct {
const char *name;
void *address;
}
lt__PROGRAM__LTX_preloaded_symbols[[]] =
{
{ "@PROGRAM@", (void *) 0 },
_LT_EOF
$SED "s/^$symcode$symcode* \(.*\) \(.*\)$/ {\"\2\", (void *) \&\2},/" < "$nlist" | $GREP -v main >> conftest.$ac_ext
cat <<\_LT_EOF >> conftest.$ac_ext
{0, (void *) 0}
};
/* This works around a problem in FreeBSD linker */
#ifdef FREEBSD_WORKAROUND
static const void *lt_preloaded_setup() {
return lt__PROGRAM__LTX_preloaded_symbols;
}
#endif
#ifdef __cplusplus
}
#endif
_LT_EOF
# Now try linking the two files.
mv conftest.$ac_objext conftstm.$ac_objext
lt_globsym_save_LIBS=$LIBS
lt_globsym_save_CFLAGS=$CFLAGS
LIBS="conftstm.$ac_objext"
CFLAGS="$CFLAGS$_LT_TAGVAR(lt_prog_compiler_no_builtin_flag, $1)"
if AC_TRY_EVAL(ac_link) && test -s conftest${ac_exeext}; then
pipe_works=yes
fi
LIBS=$lt_globsym_save_LIBS
CFLAGS=$lt_globsym_save_CFLAGS
else
echo "cannot find nm_test_func in $nlist" >&AS_MESSAGE_LOG_FD
fi
else
echo "cannot find nm_test_var in $nlist" >&AS_MESSAGE_LOG_FD
fi
else
echo "cannot run $lt_cv_sys_global_symbol_pipe" >&AS_MESSAGE_LOG_FD
fi
else
echo "$progname: failed program was:" >&AS_MESSAGE_LOG_FD
cat conftest.$ac_ext >&5
fi
rm -rf conftest* conftst*
# Do not use the global_symbol_pipe unless it works.
if test "$pipe_works" = yes; then
break
else
lt_cv_sys_global_symbol_pipe=
fi
done
])
if test -z "$lt_cv_sys_global_symbol_pipe"; then
lt_cv_sys_global_symbol_to_cdecl=
fi
if test -z "$lt_cv_sys_global_symbol_pipe$lt_cv_sys_global_symbol_to_cdecl"; then
AC_MSG_RESULT(failed)
else
AC_MSG_RESULT(ok)
fi
# Response file support.
if test "$lt_cv_nm_interface" = "MS dumpbin"; then
nm_file_list_spec='@'
elif $NM --help 2>/dev/null | grep '[[@]]FILE' >/dev/null; then
nm_file_list_spec='@'
fi
_LT_DECL([global_symbol_pipe], [lt_cv_sys_global_symbol_pipe], [1],
[Take the output of nm and produce a listing of raw symbols and C names])
_LT_DECL([global_symbol_to_cdecl], [lt_cv_sys_global_symbol_to_cdecl], [1],
[Transform the output of nm in a proper C declaration])
_LT_DECL([global_symbol_to_c_name_address],
[lt_cv_sys_global_symbol_to_c_name_address], [1],
[Transform the output of nm in a C name address pair])
_LT_DECL([global_symbol_to_c_name_address_lib_prefix],
[lt_cv_sys_global_symbol_to_c_name_address_lib_prefix], [1],
[Transform the output of nm in a C name address pair when lib prefix is needed])
_LT_DECL([], [nm_file_list_spec], [1],
[Specify filename containing input files for $NM])
]) # _LT_CMD_GLOBAL_SYMBOLS
# _LT_COMPILER_PIC([TAGNAME])
# ---------------------------
m4_defun([_LT_COMPILER_PIC],
[m4_require([_LT_TAG_COMPILER])dnl
_LT_TAGVAR(lt_prog_compiler_wl, $1)=
_LT_TAGVAR(lt_prog_compiler_pic, $1)=
_LT_TAGVAR(lt_prog_compiler_static, $1)=
m4_if([$1], [CXX], [
# C++ specific cases for pic, static, wl, etc.
if test "$GXX" = yes; then
_LT_TAGVAR(lt_prog_compiler_wl, $1)='-Wl,'
_LT_TAGVAR(lt_prog_compiler_static, $1)='-static'
case $host_os in
aix*)
# All AIX code is PIC.
if test "$host_cpu" = ia64; then
# AIX 5 now supports IA64 processor
_LT_TAGVAR(lt_prog_compiler_static, $1)='-Bstatic'
fi
;;
amigaos*)
case $host_cpu in
powerpc)
# see comment about AmigaOS4 .so support
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-fPIC'
;;
m68k)
# FIXME: we need at least 68020 code to build shared libraries, but
# adding the `-m68020' flag to GCC prevents building anything better,
# like `-m68040'.
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-m68020 -resident32 -malways-restore-a4'
;;
esac
;;
beos* | irix5* | irix6* | nonstopux* | osf3* | osf4* | osf5*)
# PIC is the default for these OSes.
;;
mingw* | cygwin* | os2* | pw32* | cegcc*)
# This hack is so that the source file can tell whether it is being
# built for inclusion in a dll (and should export symbols for example).
# Although the cygwin gcc ignores -fPIC, still need this for old-style
# (--disable-auto-import) libraries
m4_if([$1], [GCJ], [],
[_LT_TAGVAR(lt_prog_compiler_pic, $1)='-DDLL_EXPORT'])
;;
darwin* | rhapsody*)
# PIC is the default on this platform
# Common symbols not allowed in MH_DYLIB files
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-fno-common'
;;
*djgpp*)
# DJGPP does not support shared libraries at all
_LT_TAGVAR(lt_prog_compiler_pic, $1)=
;;
haiku*)
# PIC is the default for Haiku.
# The "-static" flag exists, but is broken.
_LT_TAGVAR(lt_prog_compiler_static, $1)=
;;
interix[[3-9]]*)
# Interix 3.x gcc -fpic/-fPIC options generate broken code.
# Instead, we relocate shared libraries at runtime.
;;
sysv4*MP*)
if test -d /usr/nec; then
_LT_TAGVAR(lt_prog_compiler_pic, $1)=-Kconform_pic
fi
;;
hpux*)
# PIC is the default for 64-bit PA HP-UX, but not for 32-bit
# PA HP-UX. On IA64 HP-UX, PIC is the default but the pic flag
# sets the default TLS model and affects inlining.
case $host_cpu in
hppa*64*)
;;
*)
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-fPIC'
;;
esac
;;
*qnx* | *nto*)
# QNX uses GNU C++, but need to define -shared option too, otherwise
# it will coredump.
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-fPIC -shared'
;;
*)
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-fPIC'
;;
esac
else
case $host_os in
aix[[4-9]]*)
# All AIX code is PIC.
if test "$host_cpu" = ia64; then
# AIX 5 now supports IA64 processor
_LT_TAGVAR(lt_prog_compiler_static, $1)='-Bstatic'
else
_LT_TAGVAR(lt_prog_compiler_static, $1)='-bnso -bI:/lib/syscalls.exp'
fi
;;
chorus*)
case $cc_basename in
cxch68*)
# Green Hills C++ Compiler
# _LT_TAGVAR(lt_prog_compiler_static, $1)="--no_auto_instantiation -u __main -u __premain -u _abort -r $COOL_DIR/lib/libOrb.a $MVME_DIR/lib/CC/libC.a $MVME_DIR/lib/classix/libcx.s.a"
;;
esac
;;
mingw* | cygwin* | os2* | pw32* | cegcc*)
# This hack is so that the source file can tell whether it is being
# built for inclusion in a dll (and should export symbols for example).
m4_if([$1], [GCJ], [],
[_LT_TAGVAR(lt_prog_compiler_pic, $1)='-DDLL_EXPORT'])
;;
dgux*)
case $cc_basename in
ec++*)
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-KPIC'
;;
ghcx*)
# Green Hills C++ Compiler
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-pic'
;;
*)
;;
esac
;;
freebsd* | dragonfly*)
# FreeBSD uses GNU C++
;;
hpux9* | hpux10* | hpux11*)
case $cc_basename in
CC*)
_LT_TAGVAR(lt_prog_compiler_wl, $1)='-Wl,'
_LT_TAGVAR(lt_prog_compiler_static, $1)='${wl}-a ${wl}archive'
if test "$host_cpu" != ia64; then
_LT_TAGVAR(lt_prog_compiler_pic, $1)='+Z'
fi
;;
aCC*)
_LT_TAGVAR(lt_prog_compiler_wl, $1)='-Wl,'
_LT_TAGVAR(lt_prog_compiler_static, $1)='${wl}-a ${wl}archive'
case $host_cpu in
hppa*64*|ia64*)
# +Z the default
;;
*)
_LT_TAGVAR(lt_prog_compiler_pic, $1)='+Z'
;;
esac
;;
*)
;;
esac
;;
interix*)
# This is c89, which is MS Visual C++ (no shared libs)
# Anyone wants to do a port?
;;
irix5* | irix6* | nonstopux*)
case $cc_basename in
CC*)
_LT_TAGVAR(lt_prog_compiler_wl, $1)='-Wl,'
_LT_TAGVAR(lt_prog_compiler_static, $1)='-non_shared'
# CC pic flag -KPIC is the default.
;;
*)
;;
esac
;;
linux* | k*bsd*-gnu | kopensolaris*-gnu)
case $cc_basename in
KCC*)
# KAI C++ Compiler
_LT_TAGVAR(lt_prog_compiler_wl, $1)='--backend -Wl,'
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-fPIC'
;;
ecpc* )
# old Intel C++ for x86_64 which still supported -KPIC.
_LT_TAGVAR(lt_prog_compiler_wl, $1)='-Wl,'
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-KPIC'
_LT_TAGVAR(lt_prog_compiler_static, $1)='-static'
;;
icpc* )
# Intel C++, used to be incompatible with GCC.
# ICC 10 doesn't accept -KPIC any more.
_LT_TAGVAR(lt_prog_compiler_wl, $1)='-Wl,'
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-fPIC'
_LT_TAGVAR(lt_prog_compiler_static, $1)='-static'
;;
pgCC* | pgcpp*)
# Portland Group C++ compiler
_LT_TAGVAR(lt_prog_compiler_wl, $1)='-Wl,'
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-fpic'
_LT_TAGVAR(lt_prog_compiler_static, $1)='-Bstatic'
;;
cxx*)
# Compaq C++
# Make sure the PIC flag is empty. It appears that all Alpha
# Linux and Compaq Tru64 Unix objects are PIC.
_LT_TAGVAR(lt_prog_compiler_pic, $1)=
_LT_TAGVAR(lt_prog_compiler_static, $1)='-non_shared'
;;
xlc* | xlC* | bgxl[[cC]]* | mpixl[[cC]]*)
# IBM XL 8.0, 9.0 on PPC and BlueGene
_LT_TAGVAR(lt_prog_compiler_wl, $1)='-Wl,'
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-qpic'
_LT_TAGVAR(lt_prog_compiler_static, $1)='-qstaticlink'
;;
*)
case `$CC -V 2>&1 | sed 5q` in
*Sun\ C*)
# Sun C++ 5.9
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-KPIC'
_LT_TAGVAR(lt_prog_compiler_static, $1)='-Bstatic'
_LT_TAGVAR(lt_prog_compiler_wl, $1)='-Qoption ld '
;;
esac
;;
esac
;;
lynxos*)
;;
m88k*)
;;
mvs*)
case $cc_basename in
cxx*)
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-W c,exportall'
;;
*)
;;
esac
;;
netbsd* | netbsdelf*-gnu)
;;
*qnx* | *nto*)
# QNX uses GNU C++, but need to define -shared option too, otherwise
# it will coredump.
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-fPIC -shared'
;;
osf3* | osf4* | osf5*)
case $cc_basename in
KCC*)
_LT_TAGVAR(lt_prog_compiler_wl, $1)='--backend -Wl,'
;;
RCC*)
# Rational C++ 2.4.1
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-pic'
;;
cxx*)
# Digital/Compaq C++
_LT_TAGVAR(lt_prog_compiler_wl, $1)='-Wl,'
# Make sure the PIC flag is empty. It appears that all Alpha
# Linux and Compaq Tru64 Unix objects are PIC.
_LT_TAGVAR(lt_prog_compiler_pic, $1)=
_LT_TAGVAR(lt_prog_compiler_static, $1)='-non_shared'
;;
*)
;;
esac
;;
psos*)
;;
solaris*)
case $cc_basename in
CC* | sunCC*)
# Sun C++ 4.2, 5.x and Centerline C++
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-KPIC'
_LT_TAGVAR(lt_prog_compiler_static, $1)='-Bstatic'
_LT_TAGVAR(lt_prog_compiler_wl, $1)='-Qoption ld '
;;
gcx*)
# Green Hills C++ Compiler
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-PIC'
;;
*)
;;
esac
;;
sunos4*)
case $cc_basename in
CC*)
# Sun C++ 4.x
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-pic'
_LT_TAGVAR(lt_prog_compiler_static, $1)='-Bstatic'
;;
lcc*)
# Lucid
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-pic'
;;
*)
;;
esac
;;
sysv5* | unixware* | sco3.2v5* | sco5v6* | OpenUNIX*)
case $cc_basename in
CC*)
_LT_TAGVAR(lt_prog_compiler_wl, $1)='-Wl,'
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-KPIC'
_LT_TAGVAR(lt_prog_compiler_static, $1)='-Bstatic'
;;
esac
;;
tandem*)
case $cc_basename in
NCC*)
# NonStop-UX NCC 3.20
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-KPIC'
;;
*)
;;
esac
;;
vxworks*)
;;
*)
_LT_TAGVAR(lt_prog_compiler_can_build_shared, $1)=no
;;
esac
fi
],
[
if test "$GCC" = yes; then
_LT_TAGVAR(lt_prog_compiler_wl, $1)='-Wl,'
_LT_TAGVAR(lt_prog_compiler_static, $1)='-static'
case $host_os in
aix*)
# All AIX code is PIC.
if test "$host_cpu" = ia64; then
# AIX 5 now supports IA64 processor
_LT_TAGVAR(lt_prog_compiler_static, $1)='-Bstatic'
fi
;;
amigaos*)
case $host_cpu in
powerpc)
# see comment about AmigaOS4 .so support
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-fPIC'
;;
m68k)
# FIXME: we need at least 68020 code to build shared libraries, but
# adding the `-m68020' flag to GCC prevents building anything better,
# like `-m68040'.
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-m68020 -resident32 -malways-restore-a4'
;;
esac
;;
beos* | irix5* | irix6* | nonstopux* | osf3* | osf4* | osf5*)
# PIC is the default for these OSes.
;;
mingw* | cygwin* | pw32* | os2* | cegcc*)
# This hack is so that the source file can tell whether it is being
# built for inclusion in a dll (and should export symbols for example).
# Although the cygwin gcc ignores -fPIC, still need this for old-style
# (--disable-auto-import) libraries
m4_if([$1], [GCJ], [],
[_LT_TAGVAR(lt_prog_compiler_pic, $1)='-DDLL_EXPORT'])
;;
darwin* | rhapsody*)
# PIC is the default on this platform
# Common symbols not allowed in MH_DYLIB files
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-fno-common'
;;
haiku*)
# PIC is the default for Haiku.
# The "-static" flag exists, but is broken.
_LT_TAGVAR(lt_prog_compiler_static, $1)=
;;
hpux*)
# PIC is the default for 64-bit PA HP-UX, but not for 32-bit
# PA HP-UX. On IA64 HP-UX, PIC is the default but the pic flag
# sets the default TLS model and affects inlining.
case $host_cpu in
hppa*64*)
# +Z the default
;;
*)
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-fPIC'
;;
esac
;;
interix[[3-9]]*)
# Interix 3.x gcc -fpic/-fPIC options generate broken code.
# Instead, we relocate shared libraries at runtime.
;;
msdosdjgpp*)
# Just because we use GCC doesn't mean we suddenly get shared libraries
# on systems that don't support them.
_LT_TAGVAR(lt_prog_compiler_can_build_shared, $1)=no
enable_shared=no
;;
*nto* | *qnx*)
# QNX uses GNU C++, but need to define -shared option too, otherwise
# it will coredump.
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-fPIC -shared'
;;
sysv4*MP*)
if test -d /usr/nec; then
_LT_TAGVAR(lt_prog_compiler_pic, $1)=-Kconform_pic
fi
;;
*)
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-fPIC'
;;
esac
case $cc_basename in
nvcc*) # Cuda Compiler Driver 2.2
_LT_TAGVAR(lt_prog_compiler_wl, $1)='-Xlinker '
if test -n "$_LT_TAGVAR(lt_prog_compiler_pic, $1)"; then
_LT_TAGVAR(lt_prog_compiler_pic, $1)="-Xcompiler $_LT_TAGVAR(lt_prog_compiler_pic, $1)"
fi
;;
esac
else
# PORTME Check for flag to pass linker flags through the system compiler.
case $host_os in
aix*)
_LT_TAGVAR(lt_prog_compiler_wl, $1)='-Wl,'
if test "$host_cpu" = ia64; then
# AIX 5 now supports IA64 processor
_LT_TAGVAR(lt_prog_compiler_static, $1)='-Bstatic'
else
_LT_TAGVAR(lt_prog_compiler_static, $1)='-bnso -bI:/lib/syscalls.exp'
fi
;;
mingw* | cygwin* | pw32* | os2* | cegcc*)
# This hack is so that the source file can tell whether it is being
# built for inclusion in a dll (and should export symbols for example).
m4_if([$1], [GCJ], [],
[_LT_TAGVAR(lt_prog_compiler_pic, $1)='-DDLL_EXPORT'])
;;
hpux9* | hpux10* | hpux11*)
_LT_TAGVAR(lt_prog_compiler_wl, $1)='-Wl,'
# PIC is the default for IA64 HP-UX and 64-bit HP-UX, but
# not for PA HP-UX.
case $host_cpu in
hppa*64*|ia64*)
# +Z the default
;;
*)
_LT_TAGVAR(lt_prog_compiler_pic, $1)='+Z'
;;
esac
# Is there a better lt_prog_compiler_static that works with the bundled CC?
_LT_TAGVAR(lt_prog_compiler_static, $1)='${wl}-a ${wl}archive'
;;
irix5* | irix6* | nonstopux*)
_LT_TAGVAR(lt_prog_compiler_wl, $1)='-Wl,'
# PIC (with -KPIC) is the default.
_LT_TAGVAR(lt_prog_compiler_static, $1)='-non_shared'
;;
linux* | k*bsd*-gnu | kopensolaris*-gnu)
case $cc_basename in
# old Intel for x86_64 which still supported -KPIC.
ecc*)
_LT_TAGVAR(lt_prog_compiler_wl, $1)='-Wl,'
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-KPIC'
_LT_TAGVAR(lt_prog_compiler_static, $1)='-static'
;;
# icc used to be incompatible with GCC.
# ICC 10 doesn't accept -KPIC any more.
icc* | ifort*)
_LT_TAGVAR(lt_prog_compiler_wl, $1)='-Wl,'
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-fPIC'
_LT_TAGVAR(lt_prog_compiler_static, $1)='-static'
;;
# Lahey Fortran 8.1.
lf95*)
_LT_TAGVAR(lt_prog_compiler_wl, $1)='-Wl,'
_LT_TAGVAR(lt_prog_compiler_pic, $1)='--shared'
_LT_TAGVAR(lt_prog_compiler_static, $1)='--static'
;;
nagfor*)
# NAG Fortran compiler
_LT_TAGVAR(lt_prog_compiler_wl, $1)='-Wl,-Wl,,'
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-PIC'
_LT_TAGVAR(lt_prog_compiler_static, $1)='-Bstatic'
;;
pgcc* | pgf77* | pgf90* | pgf95* | pgfortran*)
# Portland Group compilers (*not* the Pentium gcc compiler,
# which looks to be a dead project)
_LT_TAGVAR(lt_prog_compiler_wl, $1)='-Wl,'
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-fpic'
_LT_TAGVAR(lt_prog_compiler_static, $1)='-Bstatic'
;;
ccc*)
_LT_TAGVAR(lt_prog_compiler_wl, $1)='-Wl,'
# All Alpha code is PIC.
_LT_TAGVAR(lt_prog_compiler_static, $1)='-non_shared'
;;
xl* | bgxl* | bgf* | mpixl*)
# IBM XL C 8.0/Fortran 10.1, 11.1 on PPC and BlueGene
_LT_TAGVAR(lt_prog_compiler_wl, $1)='-Wl,'
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-qpic'
_LT_TAGVAR(lt_prog_compiler_static, $1)='-qstaticlink'
;;
*)
case `$CC -V 2>&1 | sed 5q` in
*Sun\ Ceres\ Fortran* | *Sun*Fortran*\ [[1-7]].* | *Sun*Fortran*\ 8.[[0-3]]*)
# Sun Fortran 8.3 passes all unrecognized flags to the linker
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-KPIC'
_LT_TAGVAR(lt_prog_compiler_static, $1)='-Bstatic'
_LT_TAGVAR(lt_prog_compiler_wl, $1)=''
;;
*Sun\ F* | *Sun*Fortran*)
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-KPIC'
_LT_TAGVAR(lt_prog_compiler_static, $1)='-Bstatic'
_LT_TAGVAR(lt_prog_compiler_wl, $1)='-Qoption ld '
;;
*Sun\ C*)
# Sun C 5.9
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-KPIC'
_LT_TAGVAR(lt_prog_compiler_static, $1)='-Bstatic'
_LT_TAGVAR(lt_prog_compiler_wl, $1)='-Wl,'
;;
*Intel*\ [[CF]]*Compiler*)
_LT_TAGVAR(lt_prog_compiler_wl, $1)='-Wl,'
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-fPIC'
_LT_TAGVAR(lt_prog_compiler_static, $1)='-static'
;;
*Portland\ Group*)
_LT_TAGVAR(lt_prog_compiler_wl, $1)='-Wl,'
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-fpic'
_LT_TAGVAR(lt_prog_compiler_static, $1)='-Bstatic'
;;
esac
;;
esac
;;
newsos6)
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-KPIC'
_LT_TAGVAR(lt_prog_compiler_static, $1)='-Bstatic'
;;
*nto* | *qnx*)
# QNX uses GNU C++, but need to define -shared option too, otherwise
# it will coredump.
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-fPIC -shared'
;;
osf3* | osf4* | osf5*)
_LT_TAGVAR(lt_prog_compiler_wl, $1)='-Wl,'
# All OSF/1 code is PIC.
_LT_TAGVAR(lt_prog_compiler_static, $1)='-non_shared'
;;
rdos*)
_LT_TAGVAR(lt_prog_compiler_static, $1)='-non_shared'
;;
solaris*)
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-KPIC'
_LT_TAGVAR(lt_prog_compiler_static, $1)='-Bstatic'
case $cc_basename in
f77* | f90* | f95* | sunf77* | sunf90* | sunf95*)
_LT_TAGVAR(lt_prog_compiler_wl, $1)='-Qoption ld ';;
*)
_LT_TAGVAR(lt_prog_compiler_wl, $1)='-Wl,';;
esac
;;
sunos4*)
_LT_TAGVAR(lt_prog_compiler_wl, $1)='-Qoption ld '
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-PIC'
_LT_TAGVAR(lt_prog_compiler_static, $1)='-Bstatic'
;;
sysv4 | sysv4.2uw2* | sysv4.3*)
_LT_TAGVAR(lt_prog_compiler_wl, $1)='-Wl,'
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-KPIC'
_LT_TAGVAR(lt_prog_compiler_static, $1)='-Bstatic'
;;
sysv4*MP*)
if test -d /usr/nec ;then
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-Kconform_pic'
_LT_TAGVAR(lt_prog_compiler_static, $1)='-Bstatic'
fi
;;
sysv5* | unixware* | sco3.2v5* | sco5v6* | OpenUNIX*)
_LT_TAGVAR(lt_prog_compiler_wl, $1)='-Wl,'
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-KPIC'
_LT_TAGVAR(lt_prog_compiler_static, $1)='-Bstatic'
;;
unicos*)
_LT_TAGVAR(lt_prog_compiler_wl, $1)='-Wl,'
_LT_TAGVAR(lt_prog_compiler_can_build_shared, $1)=no
;;
uts4*)
_LT_TAGVAR(lt_prog_compiler_pic, $1)='-pic'
_LT_TAGVAR(lt_prog_compiler_static, $1)='-Bstatic'
;;
*)
_LT_TAGVAR(lt_prog_compiler_can_build_shared, $1)=no
;;
esac
fi
])
case $host_os in
# For platforms which do not support PIC, -DPIC is meaningless:
*djgpp*)
_LT_TAGVAR(lt_prog_compiler_pic, $1)=
;;
*)
_LT_TAGVAR(lt_prog_compiler_pic, $1)="$_LT_TAGVAR(lt_prog_compiler_pic, $1)@&t@m4_if([$1],[],[ -DPIC],[m4_if([$1],[CXX],[ -DPIC],[])])"
;;
esac
AC_CACHE_CHECK([for $compiler option to produce PIC],
[_LT_TAGVAR(lt_cv_prog_compiler_pic, $1)],
[_LT_TAGVAR(lt_cv_prog_compiler_pic, $1)=$_LT_TAGVAR(lt_prog_compiler_pic, $1)])
_LT_TAGVAR(lt_prog_compiler_pic, $1)=$_LT_TAGVAR(lt_cv_prog_compiler_pic, $1)
#
# Check to make sure the PIC flag actually works.
#
if test -n "$_LT_TAGVAR(lt_prog_compiler_pic, $1)"; then
_LT_COMPILER_OPTION([if $compiler PIC flag $_LT_TAGVAR(lt_prog_compiler_pic, $1) works],
[_LT_TAGVAR(lt_cv_prog_compiler_pic_works, $1)],
[$_LT_TAGVAR(lt_prog_compiler_pic, $1)@&t@m4_if([$1],[],[ -DPIC],[m4_if([$1],[CXX],[ -DPIC],[])])], [],
[case $_LT_TAGVAR(lt_prog_compiler_pic, $1) in
"" | " "*) ;;
*) _LT_TAGVAR(lt_prog_compiler_pic, $1)=" $_LT_TAGVAR(lt_prog_compiler_pic, $1)" ;;
esac],
[_LT_TAGVAR(lt_prog_compiler_pic, $1)=
_LT_TAGVAR(lt_prog_compiler_can_build_shared, $1)=no])
fi
_LT_TAGDECL([pic_flag], [lt_prog_compiler_pic], [1],
[Additional compiler flags for building library objects])
_LT_TAGDECL([wl], [lt_prog_compiler_wl], [1],
[How to pass a linker flag through the compiler])
#
# Check to make sure the static flag actually works.
#
wl=$_LT_TAGVAR(lt_prog_compiler_wl, $1) eval lt_tmp_static_flag=\"$_LT_TAGVAR(lt_prog_compiler_static, $1)\"
_LT_LINKER_OPTION([if $compiler static flag $lt_tmp_static_flag works],
_LT_TAGVAR(lt_cv_prog_compiler_static_works, $1),
$lt_tmp_static_flag,
[],
[_LT_TAGVAR(lt_prog_compiler_static, $1)=])
_LT_TAGDECL([link_static_flag], [lt_prog_compiler_static], [1],
[Compiler flag to prevent dynamic linking])
])# _LT_COMPILER_PIC
# _LT_LINKER_SHLIBS([TAGNAME])
# ----------------------------
# See if the linker supports building shared libraries.
m4_defun([_LT_LINKER_SHLIBS],
[AC_REQUIRE([LT_PATH_LD])dnl
AC_REQUIRE([LT_PATH_NM])dnl
m4_require([_LT_PATH_MANIFEST_TOOL])dnl
m4_require([_LT_FILEUTILS_DEFAULTS])dnl
m4_require([_LT_DECL_EGREP])dnl
m4_require([_LT_DECL_SED])dnl
m4_require([_LT_CMD_GLOBAL_SYMBOLS])dnl
m4_require([_LT_TAG_COMPILER])dnl
AC_MSG_CHECKING([whether the $compiler linker ($LD) supports shared libraries])
m4_if([$1], [CXX], [
_LT_TAGVAR(export_symbols_cmds, $1)='$NM $libobjs $convenience | $global_symbol_pipe | $SED '\''s/.* //'\'' | sort | uniq > $export_symbols'
_LT_TAGVAR(exclude_expsyms, $1)=['_GLOBAL_OFFSET_TABLE_|_GLOBAL__F[ID]_.*']
case $host_os in
aix[[4-9]]*)
# If we're using GNU nm, then we don't want the "-C" option.
# -C means demangle to AIX nm, but means don't demangle with GNU nm
# Also, AIX nm treats weak defined symbols like other global defined
# symbols, whereas GNU nm marks them as "W".
if $NM -V 2>&1 | $GREP 'GNU' > /dev/null; then
_LT_TAGVAR(export_symbols_cmds, $1)='$NM -Bpg $libobjs $convenience | awk '\''{ if (((\$ 2 == "T") || (\$ 2 == "D") || (\$ 2 == "B") || (\$ 2 == "W")) && ([substr](\$ 3,1,1) != ".")) { print \$ 3 } }'\'' | sort -u > $export_symbols'
else
_LT_TAGVAR(export_symbols_cmds, $1)='$NM -BCpg $libobjs $convenience | awk '\''{ if (((\$ 2 == "T") || (\$ 2 == "D") || (\$ 2 == "B")) && ([substr](\$ 3,1,1) != ".")) { print \$ 3 } }'\'' | sort -u > $export_symbols'
fi
;;
pw32*)
_LT_TAGVAR(export_symbols_cmds, $1)="$ltdll_cmds"
;;
cygwin* | mingw* | cegcc*)
case $cc_basename in
cl*)
_LT_TAGVAR(exclude_expsyms, $1)='_NULL_IMPORT_DESCRIPTOR|_IMPORT_DESCRIPTOR_.*'
;;
*)
_LT_TAGVAR(export_symbols_cmds, $1)='$NM $libobjs $convenience | $global_symbol_pipe | $SED -e '\''/^[[BCDGRS]][[ ]]/s/.*[[ ]]\([[^ ]]*\)/\1 DATA/;s/^.*[[ ]]__nm__\([[^ ]]*\)[[ ]][[^ ]]*/\1 DATA/;/^I[[ ]]/d;/^[[AITW]][[ ]]/s/.* //'\'' | sort | uniq > $export_symbols'
_LT_TAGVAR(exclude_expsyms, $1)=['[_]+GLOBAL_OFFSET_TABLE_|[_]+GLOBAL__[FID]_.*|[_]+head_[A-Za-z0-9_]+_dll|[A-Za-z0-9_]+_dll_iname']
;;
esac
;;
linux* | k*bsd*-gnu | gnu*)
_LT_TAGVAR(link_all_deplibs, $1)=no
;;
*)
_LT_TAGVAR(export_symbols_cmds, $1)='$NM $libobjs $convenience | $global_symbol_pipe | $SED '\''s/.* //'\'' | sort | uniq > $export_symbols'
;;
esac
], [
runpath_var=
_LT_TAGVAR(allow_undefined_flag, $1)=
_LT_TAGVAR(always_export_symbols, $1)=no
_LT_TAGVAR(archive_cmds, $1)=
_LT_TAGVAR(archive_expsym_cmds, $1)=
_LT_TAGVAR(compiler_needs_object, $1)=no
_LT_TAGVAR(enable_shared_with_static_runtimes, $1)=no
_LT_TAGVAR(export_dynamic_flag_spec, $1)=
_LT_TAGVAR(export_symbols_cmds, $1)='$NM $libobjs $convenience | $global_symbol_pipe | $SED '\''s/.* //'\'' | sort | uniq > $export_symbols'
_LT_TAGVAR(hardcode_automatic, $1)=no
_LT_TAGVAR(hardcode_direct, $1)=no
_LT_TAGVAR(hardcode_direct_absolute, $1)=no
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)=
_LT_TAGVAR(hardcode_libdir_separator, $1)=
_LT_TAGVAR(hardcode_minus_L, $1)=no
_LT_TAGVAR(hardcode_shlibpath_var, $1)=unsupported
_LT_TAGVAR(inherit_rpath, $1)=no
_LT_TAGVAR(link_all_deplibs, $1)=unknown
_LT_TAGVAR(module_cmds, $1)=
_LT_TAGVAR(module_expsym_cmds, $1)=
_LT_TAGVAR(old_archive_from_new_cmds, $1)=
_LT_TAGVAR(old_archive_from_expsyms_cmds, $1)=
_LT_TAGVAR(thread_safe_flag_spec, $1)=
_LT_TAGVAR(whole_archive_flag_spec, $1)=
# include_expsyms should be a list of space-separated symbols to be *always*
# included in the symbol list
_LT_TAGVAR(include_expsyms, $1)=
# exclude_expsyms can be an extended regexp of symbols to exclude
# it will be wrapped by ` (' and `)$', so one must not match beginning or
# end of line. Example: `a|bc|.*d.*' will exclude the symbols `a' and `bc',
# as well as any symbol that contains `d'.
_LT_TAGVAR(exclude_expsyms, $1)=['_GLOBAL_OFFSET_TABLE_|_GLOBAL__F[ID]_.*']
# Although _GLOBAL_OFFSET_TABLE_ is a valid symbol C name, most a.out
# platforms (ab)use it in PIC code, but their linkers get confused if
# the symbol is explicitly referenced. Since portable code cannot
# rely on this symbol name, it's probably fine to never include it in
# preloaded symbol tables.
# Exclude shared library initialization/finalization symbols.
dnl Note also adjust exclude_expsyms for C++ above.
extract_expsyms_cmds=
case $host_os in
cygwin* | mingw* | pw32* | cegcc*)
# FIXME: the MSVC++ port hasn't been tested in a loooong time
# When not using gcc, we currently assume that we are using
# Microsoft Visual C++.
if test "$GCC" != yes; then
with_gnu_ld=no
fi
;;
interix*)
# we just hope/assume this is gcc and not c89 (= MSVC++)
with_gnu_ld=yes
;;
openbsd*)
with_gnu_ld=no
;;
linux* | k*bsd*-gnu | gnu*)
_LT_TAGVAR(link_all_deplibs, $1)=no
;;
esac
_LT_TAGVAR(ld_shlibs, $1)=yes
# On some targets, GNU ld is compatible enough with the native linker
# that we're better off using the native interface for both.
lt_use_gnu_ld_interface=no
if test "$with_gnu_ld" = yes; then
case $host_os in
aix*)
# The AIX port of GNU ld has always aspired to compatibility
# with the native linker. However, as the warning in the GNU ld
# block says, versions before 2.19.5* couldn't really create working
# shared libraries, regardless of the interface used.
case `$LD -v 2>&1` in
*\ \(GNU\ Binutils\)\ 2.19.5*) ;;
*\ \(GNU\ Binutils\)\ 2.[[2-9]]*) ;;
*\ \(GNU\ Binutils\)\ [[3-9]]*) ;;
*)
lt_use_gnu_ld_interface=yes
;;
esac
;;
*)
lt_use_gnu_ld_interface=yes
;;
esac
fi
if test "$lt_use_gnu_ld_interface" = yes; then
# If archive_cmds runs LD, not CC, wlarc should be empty
wlarc='${wl}'
# Set some defaults for GNU ld with shared library support. These
# are reset later if shared libraries are not supported. Putting them
# here allows them to be overridden if necessary.
runpath_var=LD_RUN_PATH
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='${wl}-rpath ${wl}$libdir'
_LT_TAGVAR(export_dynamic_flag_spec, $1)='${wl}--export-dynamic'
# ancient GNU ld didn't support --whole-archive et. al.
if $LD --help 2>&1 | $GREP 'no-whole-archive' > /dev/null; then
_LT_TAGVAR(whole_archive_flag_spec, $1)="$wlarc"'--whole-archive$convenience '"$wlarc"'--no-whole-archive'
else
_LT_TAGVAR(whole_archive_flag_spec, $1)=
fi
supports_anon_versioning=no
case `$LD -v 2>&1` in
*GNU\ gold*) supports_anon_versioning=yes ;;
*\ [[01]].* | *\ 2.[[0-9]].* | *\ 2.10.*) ;; # catch versions < 2.11
*\ 2.11.93.0.2\ *) supports_anon_versioning=yes ;; # RH7.3 ...
*\ 2.11.92.0.12\ *) supports_anon_versioning=yes ;; # Mandrake 8.2 ...
*\ 2.11.*) ;; # other 2.11 versions
*) supports_anon_versioning=yes ;;
esac
# See if GNU ld supports shared libraries.
case $host_os in
aix[[3-9]]*)
# On AIX/PPC, the GNU linker is very broken
if test "$host_cpu" != ia64; then
_LT_TAGVAR(ld_shlibs, $1)=no
cat <<_LT_EOF 1>&2
*** Warning: the GNU linker, at least up to release 2.19, is reported
*** to be unable to reliably create shared libraries on AIX.
*** Therefore, libtool is disabling shared libraries support. If you
*** really care for shared libraries, you may want to install binutils
*** 2.20 or above, or modify your PATH so that a non-GNU linker is found.
*** You will then need to restart the configuration process.
_LT_EOF
fi
;;
amigaos*)
case $host_cpu in
powerpc)
# see comment about AmigaOS4 .so support
_LT_TAGVAR(archive_cmds, $1)='$CC -shared $libobjs $deplibs $compiler_flags ${wl}-soname $wl$soname -o $lib'
_LT_TAGVAR(archive_expsym_cmds, $1)=''
;;
m68k)
_LT_TAGVAR(archive_cmds, $1)='$RM $output_objdir/a2ixlibrary.data~$ECHO "#define NAME $libname" > $output_objdir/a2ixlibrary.data~$ECHO "#define LIBRARY_ID 1" >> $output_objdir/a2ixlibrary.data~$ECHO "#define VERSION $major" >> $output_objdir/a2ixlibrary.data~$ECHO "#define REVISION $revision" >> $output_objdir/a2ixlibrary.data~$AR $AR_FLAGS $lib $libobjs~$RANLIB $lib~(cd $output_objdir && a2ixlibrary -32)'
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='-L$libdir'
_LT_TAGVAR(hardcode_minus_L, $1)=yes
;;
esac
;;
beos*)
if $LD --help 2>&1 | $GREP ': supported targets:.* elf' > /dev/null; then
_LT_TAGVAR(allow_undefined_flag, $1)=unsupported
# Joseph Beckenbach says some releases of gcc
# support --undefined. This deserves some investigation. FIXME
_LT_TAGVAR(archive_cmds, $1)='$CC -nostart $libobjs $deplibs $compiler_flags ${wl}-soname $wl$soname -o $lib'
else
_LT_TAGVAR(ld_shlibs, $1)=no
fi
;;
cygwin* | mingw* | pw32* | cegcc*)
# _LT_TAGVAR(hardcode_libdir_flag_spec, $1) is actually meaningless,
# as there is no search path for DLLs.
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='-L$libdir'
_LT_TAGVAR(export_dynamic_flag_spec, $1)='${wl}--export-all-symbols'
_LT_TAGVAR(allow_undefined_flag, $1)=unsupported
_LT_TAGVAR(always_export_symbols, $1)=no
_LT_TAGVAR(enable_shared_with_static_runtimes, $1)=yes
_LT_TAGVAR(export_symbols_cmds, $1)='$NM $libobjs $convenience | $global_symbol_pipe | $SED -e '\''/^[[BCDGRS]][[ ]]/s/.*[[ ]]\([[^ ]]*\)/\1 DATA/;s/^.*[[ ]]__nm__\([[^ ]]*\)[[ ]][[^ ]]*/\1 DATA/;/^I[[ ]]/d;/^[[AITW]][[ ]]/s/.* //'\'' | sort | uniq > $export_symbols'
_LT_TAGVAR(exclude_expsyms, $1)=['[_]+GLOBAL_OFFSET_TABLE_|[_]+GLOBAL__[FID]_.*|[_]+head_[A-Za-z0-9_]+_dll|[A-Za-z0-9_]+_dll_iname']
if $LD --help 2>&1 | $GREP 'auto-import' > /dev/null; then
_LT_TAGVAR(archive_cmds, $1)='$CC -shared $libobjs $deplibs $compiler_flags -o $output_objdir/$soname ${wl}--enable-auto-image-base -Xlinker --out-implib -Xlinker $lib'
# If the export-symbols file already is a .def file (1st line
# is EXPORTS), use it as is; otherwise, prepend...
_LT_TAGVAR(archive_expsym_cmds, $1)='if test "x`$SED \"$sed_uncomment_deffile\" $export_symbols | $SED 1q`" = xEXPORTS; then
cp $export_symbols $output_objdir/$soname.def;
else
echo EXPORTS > $output_objdir/$soname.def;
cat $export_symbols >> $output_objdir/$soname.def;
fi~
$CC -shared $output_objdir/$soname.def $libobjs $deplibs $compiler_flags -o $output_objdir/$soname ${wl}--enable-auto-image-base -Xlinker --out-implib -Xlinker $lib'
else
_LT_TAGVAR(ld_shlibs, $1)=no
fi
;;
haiku*)
_LT_TAGVAR(archive_cmds, $1)='$CC -shared $libobjs $deplibs $compiler_flags ${wl}-soname $wl$soname -o $lib'
_LT_TAGVAR(link_all_deplibs, $1)=yes
;;
interix[[3-9]]*)
_LT_TAGVAR(hardcode_direct, $1)=no
_LT_TAGVAR(hardcode_shlibpath_var, $1)=no
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='${wl}-rpath,$libdir'
_LT_TAGVAR(export_dynamic_flag_spec, $1)='${wl}-E'
# Hack: On Interix 3.x, we cannot compile PIC because of a broken gcc.
# Instead, shared libraries are loaded at an image base (0x10000000 by
# default) and relocated if they conflict, which is a slow very memory
# consuming and fragmenting process. To avoid this, we pick a random,
# 256 KiB-aligned image base between 0x50000000 and 0x6FFC0000 at link
# time. Moving up from 0x10000000 also allows more sbrk(2) space.
_LT_TAGVAR(archive_cmds, $1)='$CC -shared $pic_flag $libobjs $deplibs $compiler_flags ${wl}-h,$soname ${wl}--image-base,`expr ${RANDOM-$$} % 4096 / 2 \* 262144 + 1342177280` -o $lib'
_LT_TAGVAR(archive_expsym_cmds, $1)='sed "s,^,_," $export_symbols >$output_objdir/$soname.expsym~$CC -shared $pic_flag $libobjs $deplibs $compiler_flags ${wl}-h,$soname ${wl}--retain-symbols-file,$output_objdir/$soname.expsym ${wl}--image-base,`expr ${RANDOM-$$} % 4096 / 2 \* 262144 + 1342177280` -o $lib'
;;
gnu* | linux* | tpf* | k*bsd*-gnu | kopensolaris*-gnu)
tmp_diet=no
if test "$host_os" = linux-dietlibc; then
case $cc_basename in
diet\ *) tmp_diet=yes;; # linux-dietlibc with static linking (!diet-dyn)
esac
fi
if $LD --help 2>&1 | $EGREP ': supported targets:.* elf' > /dev/null \
&& test "$tmp_diet" = no
then
tmp_addflag=' $pic_flag'
tmp_sharedflag='-shared'
case $cc_basename,$host_cpu in
pgcc*) # Portland Group C compiler
_LT_TAGVAR(whole_archive_flag_spec, $1)='${wl}--whole-archive`for conv in $convenience\"\"; do test -n \"$conv\" && new_convenience=\"$new_convenience,$conv\"; done; func_echo_all \"$new_convenience\"` ${wl}--no-whole-archive'
tmp_addflag=' $pic_flag'
;;
pgf77* | pgf90* | pgf95* | pgfortran*)
# Portland Group f77 and f90 compilers
_LT_TAGVAR(whole_archive_flag_spec, $1)='${wl}--whole-archive`for conv in $convenience\"\"; do test -n \"$conv\" && new_convenience=\"$new_convenience,$conv\"; done; func_echo_all \"$new_convenience\"` ${wl}--no-whole-archive'
tmp_addflag=' $pic_flag -Mnomain' ;;
ecc*,ia64* | icc*,ia64*) # Intel C compiler on ia64
tmp_addflag=' -i_dynamic' ;;
efc*,ia64* | ifort*,ia64*) # Intel Fortran compiler on ia64
tmp_addflag=' -i_dynamic -nofor_main' ;;
ifc* | ifort*) # Intel Fortran compiler
tmp_addflag=' -nofor_main' ;;
lf95*) # Lahey Fortran 8.1
_LT_TAGVAR(whole_archive_flag_spec, $1)=
tmp_sharedflag='--shared' ;;
xl[[cC]]* | bgxl[[cC]]* | mpixl[[cC]]*) # IBM XL C 8.0 on PPC (deal with xlf below)
tmp_sharedflag='-qmkshrobj'
tmp_addflag= ;;
nvcc*) # Cuda Compiler Driver 2.2
_LT_TAGVAR(whole_archive_flag_spec, $1)='${wl}--whole-archive`for conv in $convenience\"\"; do test -n \"$conv\" && new_convenience=\"$new_convenience,$conv\"; done; func_echo_all \"$new_convenience\"` ${wl}--no-whole-archive'
_LT_TAGVAR(compiler_needs_object, $1)=yes
;;
esac
case `$CC -V 2>&1 | sed 5q` in
*Sun\ C*) # Sun C 5.9
_LT_TAGVAR(whole_archive_flag_spec, $1)='${wl}--whole-archive`new_convenience=; for conv in $convenience\"\"; do test -z \"$conv\" || new_convenience=\"$new_convenience,$conv\"; done; func_echo_all \"$new_convenience\"` ${wl}--no-whole-archive'
_LT_TAGVAR(compiler_needs_object, $1)=yes
tmp_sharedflag='-G' ;;
*Sun\ F*) # Sun Fortran 8.3
tmp_sharedflag='-G' ;;
esac
_LT_TAGVAR(archive_cmds, $1)='$CC '"$tmp_sharedflag""$tmp_addflag"' $libobjs $deplibs $compiler_flags ${wl}-soname $wl$soname -o $lib'
if test "x$supports_anon_versioning" = xyes; then
_LT_TAGVAR(archive_expsym_cmds, $1)='echo "{ global:" > $output_objdir/$libname.ver~
cat $export_symbols | sed -e "s/\(.*\)/\1;/" >> $output_objdir/$libname.ver~
echo "local: *; };" >> $output_objdir/$libname.ver~
$CC '"$tmp_sharedflag""$tmp_addflag"' $libobjs $deplibs $compiler_flags ${wl}-soname $wl$soname ${wl}-version-script ${wl}$output_objdir/$libname.ver -o $lib'
fi
case $cc_basename in
xlf* | bgf* | bgxlf* | mpixlf*)
# IBM XL Fortran 10.1 on PPC cannot create shared libs itself
_LT_TAGVAR(whole_archive_flag_spec, $1)='--whole-archive$convenience --no-whole-archive'
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='${wl}-rpath ${wl}$libdir'
_LT_TAGVAR(archive_cmds, $1)='$LD -shared $libobjs $deplibs $linker_flags -soname $soname -o $lib'
if test "x$supports_anon_versioning" = xyes; then
_LT_TAGVAR(archive_expsym_cmds, $1)='echo "{ global:" > $output_objdir/$libname.ver~
cat $export_symbols | sed -e "s/\(.*\)/\1;/" >> $output_objdir/$libname.ver~
echo "local: *; };" >> $output_objdir/$libname.ver~
$LD -shared $libobjs $deplibs $linker_flags -soname $soname -version-script $output_objdir/$libname.ver -o $lib'
fi
;;
esac
else
_LT_TAGVAR(ld_shlibs, $1)=no
fi
;;
netbsd* | netbsdelf*-gnu)
if echo __ELF__ | $CC -E - | $GREP __ELF__ >/dev/null; then
_LT_TAGVAR(archive_cmds, $1)='$LD -Bshareable $libobjs $deplibs $linker_flags -o $lib'
wlarc=
else
_LT_TAGVAR(archive_cmds, $1)='$CC -shared $pic_flag $libobjs $deplibs $compiler_flags ${wl}-soname $wl$soname -o $lib'
_LT_TAGVAR(archive_expsym_cmds, $1)='$CC -shared $pic_flag $libobjs $deplibs $compiler_flags ${wl}-soname $wl$soname ${wl}-retain-symbols-file $wl$export_symbols -o $lib'
fi
;;
solaris*)
if $LD -v 2>&1 | $GREP 'BFD 2\.8' > /dev/null; then
_LT_TAGVAR(ld_shlibs, $1)=no
cat <<_LT_EOF 1>&2
*** Warning: The releases 2.8.* of the GNU linker cannot reliably
*** create shared libraries on Solaris systems. Therefore, libtool
*** is disabling shared libraries support. We urge you to upgrade GNU
*** binutils to release 2.9.1 or newer. Another option is to modify
*** your PATH or compiler configuration so that the native linker is
*** used, and then restart.
_LT_EOF
elif $LD --help 2>&1 | $GREP ': supported targets:.* elf' > /dev/null; then
_LT_TAGVAR(archive_cmds, $1)='$CC -shared $pic_flag $libobjs $deplibs $compiler_flags ${wl}-soname $wl$soname -o $lib'
_LT_TAGVAR(archive_expsym_cmds, $1)='$CC -shared $pic_flag $libobjs $deplibs $compiler_flags ${wl}-soname $wl$soname ${wl}-retain-symbols-file $wl$export_symbols -o $lib'
else
_LT_TAGVAR(ld_shlibs, $1)=no
fi
;;
sysv5* | sco3.2v5* | sco5v6* | unixware* | OpenUNIX*)
case `$LD -v 2>&1` in
*\ [[01]].* | *\ 2.[[0-9]].* | *\ 2.1[[0-5]].*)
_LT_TAGVAR(ld_shlibs, $1)=no
cat <<_LT_EOF 1>&2
*** Warning: Releases of the GNU linker prior to 2.16.91.0.3 can not
*** reliably create shared libraries on SCO systems. Therefore, libtool
*** is disabling shared libraries support. We urge you to upgrade GNU
*** binutils to release 2.16.91.0.3 or newer. Another option is to modify
*** your PATH or compiler configuration so that the native linker is
*** used, and then restart.
_LT_EOF
;;
*)
# For security reasons, it is highly recommended that you always
# use absolute paths for naming shared libraries, and exclude the
# DT_RUNPATH tag from executables and libraries. But doing so
# requires that you compile everything twice, which is a pain.
if $LD --help 2>&1 | $GREP ': supported targets:.* elf' > /dev/null; then
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='${wl}-rpath ${wl}$libdir'
_LT_TAGVAR(archive_cmds, $1)='$CC -shared $libobjs $deplibs $compiler_flags ${wl}-soname $wl$soname -o $lib'
_LT_TAGVAR(archive_expsym_cmds, $1)='$CC -shared $libobjs $deplibs $compiler_flags ${wl}-soname $wl$soname ${wl}-retain-symbols-file $wl$export_symbols -o $lib'
else
_LT_TAGVAR(ld_shlibs, $1)=no
fi
;;
esac
;;
sunos4*)
_LT_TAGVAR(archive_cmds, $1)='$LD -assert pure-text -Bshareable -o $lib $libobjs $deplibs $linker_flags'
wlarc=
_LT_TAGVAR(hardcode_direct, $1)=yes
_LT_TAGVAR(hardcode_shlibpath_var, $1)=no
;;
*)
if $LD --help 2>&1 | $GREP ': supported targets:.* elf' > /dev/null; then
_LT_TAGVAR(archive_cmds, $1)='$CC -shared $pic_flag $libobjs $deplibs $compiler_flags ${wl}-soname $wl$soname -o $lib'
_LT_TAGVAR(archive_expsym_cmds, $1)='$CC -shared $pic_flag $libobjs $deplibs $compiler_flags ${wl}-soname $wl$soname ${wl}-retain-symbols-file $wl$export_symbols -o $lib'
else
_LT_TAGVAR(ld_shlibs, $1)=no
fi
;;
esac
if test "$_LT_TAGVAR(ld_shlibs, $1)" = no; then
runpath_var=
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)=
_LT_TAGVAR(export_dynamic_flag_spec, $1)=
_LT_TAGVAR(whole_archive_flag_spec, $1)=
fi
else
# PORTME fill in a description of your system's linker (not GNU ld)
case $host_os in
aix3*)
_LT_TAGVAR(allow_undefined_flag, $1)=unsupported
_LT_TAGVAR(always_export_symbols, $1)=yes
_LT_TAGVAR(archive_expsym_cmds, $1)='$LD -o $output_objdir/$soname $libobjs $deplibs $linker_flags -bE:$export_symbols -T512 -H512 -bM:SRE~$AR $AR_FLAGS $lib $output_objdir/$soname'
# Note: this linker hardcodes the directories in LIBPATH if there
# are no directories specified by -L.
_LT_TAGVAR(hardcode_minus_L, $1)=yes
if test "$GCC" = yes && test -z "$lt_prog_compiler_static"; then
# Neither direct hardcoding nor static linking is supported with a
# broken collect2.
_LT_TAGVAR(hardcode_direct, $1)=unsupported
fi
;;
aix[[4-9]]*)
if test "$host_cpu" = ia64; then
# On IA64, the linker does run time linking by default, so we don't
# have to do anything special.
aix_use_runtimelinking=no
exp_sym_flag='-Bexport'
no_entry_flag=""
else
# If we're using GNU nm, then we don't want the "-C" option.
# -C means demangle to AIX nm, but means don't demangle with GNU nm
# Also, AIX nm treats weak defined symbols like other global
# defined symbols, whereas GNU nm marks them as "W".
if $NM -V 2>&1 | $GREP 'GNU' > /dev/null; then
_LT_TAGVAR(export_symbols_cmds, $1)='$NM -Bpg $libobjs $convenience | awk '\''{ if (((\$ 2 == "T") || (\$ 2 == "D") || (\$ 2 == "B") || (\$ 2 == "W")) && ([substr](\$ 3,1,1) != ".")) { print \$ 3 } }'\'' | sort -u > $export_symbols'
else
_LT_TAGVAR(export_symbols_cmds, $1)='$NM -BCpg $libobjs $convenience | awk '\''{ if (((\$ 2 == "T") || (\$ 2 == "D") || (\$ 2 == "B")) && ([substr](\$ 3,1,1) != ".")) { print \$ 3 } }'\'' | sort -u > $export_symbols'
fi
aix_use_runtimelinking=no
# Test if we are trying to use run time linking or normal
# AIX style linking. If -brtl is somewhere in LDFLAGS, we
# need to do runtime linking.
case $host_os in aix4.[[23]]|aix4.[[23]].*|aix[[5-9]]*)
for ld_flag in $LDFLAGS; do
if (test $ld_flag = "-brtl" || test $ld_flag = "-Wl,-brtl"); then
aix_use_runtimelinking=yes
break
fi
done
;;
esac
exp_sym_flag='-bexport'
no_entry_flag='-bnoentry'
fi
# When large executables or shared objects are built, AIX ld can
# have problems creating the table of contents. If linking a library
# or program results in "error TOC overflow" add -mminimal-toc to
# CXXFLAGS/CFLAGS for g++/gcc. In the cases where that is not
# enough to fix the problem, add -Wl,-bbigtoc to LDFLAGS.
_LT_TAGVAR(archive_cmds, $1)=''
_LT_TAGVAR(hardcode_direct, $1)=yes
_LT_TAGVAR(hardcode_direct_absolute, $1)=yes
_LT_TAGVAR(hardcode_libdir_separator, $1)=':'
_LT_TAGVAR(link_all_deplibs, $1)=yes
_LT_TAGVAR(file_list_spec, $1)='${wl}-f,'
if test "$GCC" = yes; then
case $host_os in aix4.[[012]]|aix4.[[012]].*)
# We only want to do this on AIX 4.2 and lower, the check
# below for broken collect2 doesn't work under 4.3+
collect2name=`${CC} -print-prog-name=collect2`
if test -f "$collect2name" &&
strings "$collect2name" | $GREP resolve_lib_name >/dev/null
then
# We have reworked collect2
:
else
# We have old collect2
_LT_TAGVAR(hardcode_direct, $1)=unsupported
# It fails to find uninstalled libraries when the uninstalled
# path is not listed in the libpath. Setting hardcode_minus_L
# to unsupported forces relinking
_LT_TAGVAR(hardcode_minus_L, $1)=yes
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='-L$libdir'
_LT_TAGVAR(hardcode_libdir_separator, $1)=
fi
;;
esac
shared_flag='-shared'
if test "$aix_use_runtimelinking" = yes; then
shared_flag="$shared_flag "'${wl}-G'
fi
_LT_TAGVAR(link_all_deplibs, $1)=no
else
# not using gcc
if test "$host_cpu" = ia64; then
# VisualAge C++, Version 5.5 for AIX 5L for IA-64, Beta 3 Release
# chokes on -Wl,-G. The following line is correct:
shared_flag='-G'
else
if test "$aix_use_runtimelinking" = yes; then
shared_flag='${wl}-G'
else
shared_flag='${wl}-bM:SRE'
fi
fi
fi
_LT_TAGVAR(export_dynamic_flag_spec, $1)='${wl}-bexpall'
# It seems that -bexpall does not export symbols beginning with
# underscore (_), so it is better to generate a list of symbols to export.
_LT_TAGVAR(always_export_symbols, $1)=yes
if test "$aix_use_runtimelinking" = yes; then
# Warning - without using the other runtime loading flags (-brtl),
# -berok will link without error, but may produce a broken library.
_LT_TAGVAR(allow_undefined_flag, $1)='-berok'
# Determine the default libpath from the value encoded in an
# empty executable.
_LT_SYS_MODULE_PATH_AIX([$1])
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='${wl}-blibpath:$libdir:'"$aix_libpath"
_LT_TAGVAR(archive_expsym_cmds, $1)='$CC -o $output_objdir/$soname $libobjs $deplibs '"\${wl}$no_entry_flag"' $compiler_flags `if test "x${allow_undefined_flag}" != "x"; then func_echo_all "${wl}${allow_undefined_flag}"; else :; fi` '"\${wl}$exp_sym_flag:\$export_symbols $shared_flag"
else
if test "$host_cpu" = ia64; then
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='${wl}-R $libdir:/usr/lib:/lib'
_LT_TAGVAR(allow_undefined_flag, $1)="-z nodefs"
_LT_TAGVAR(archive_expsym_cmds, $1)="\$CC $shared_flag"' -o $output_objdir/$soname $libobjs $deplibs '"\${wl}$no_entry_flag"' $compiler_flags ${wl}${allow_undefined_flag} '"\${wl}$exp_sym_flag:\$export_symbols"
else
# Determine the default libpath from the value encoded in an
# empty executable.
_LT_SYS_MODULE_PATH_AIX([$1])
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='${wl}-blibpath:$libdir:'"$aix_libpath"
# Warning - without using the other run time loading flags,
# -berok will link without error, but may produce a broken library.
_LT_TAGVAR(no_undefined_flag, $1)=' ${wl}-bernotok'
_LT_TAGVAR(allow_undefined_flag, $1)=' ${wl}-berok'
if test "$with_gnu_ld" = yes; then
# We only use this code for GNU lds that support --whole-archive.
_LT_TAGVAR(whole_archive_flag_spec, $1)='${wl}--whole-archive$convenience ${wl}--no-whole-archive'
else
# Exported symbols can be pulled into shared objects from archives
_LT_TAGVAR(whole_archive_flag_spec, $1)='$convenience'
fi
_LT_TAGVAR(archive_cmds_need_lc, $1)=yes
# This is similar to how AIX traditionally builds its shared libraries.
_LT_TAGVAR(archive_expsym_cmds, $1)="\$CC $shared_flag"' -o $output_objdir/$soname $libobjs $deplibs ${wl}-bnoentry $compiler_flags ${wl}-bE:$export_symbols${allow_undefined_flag}~$AR $AR_FLAGS $output_objdir/$libname$release.a $output_objdir/$soname'
fi
fi
;;
amigaos*)
case $host_cpu in
powerpc)
# see comment about AmigaOS4 .so support
_LT_TAGVAR(archive_cmds, $1)='$CC -shared $libobjs $deplibs $compiler_flags ${wl}-soname $wl$soname -o $lib'
_LT_TAGVAR(archive_expsym_cmds, $1)=''
;;
m68k)
_LT_TAGVAR(archive_cmds, $1)='$RM $output_objdir/a2ixlibrary.data~$ECHO "#define NAME $libname" > $output_objdir/a2ixlibrary.data~$ECHO "#define LIBRARY_ID 1" >> $output_objdir/a2ixlibrary.data~$ECHO "#define VERSION $major" >> $output_objdir/a2ixlibrary.data~$ECHO "#define REVISION $revision" >> $output_objdir/a2ixlibrary.data~$AR $AR_FLAGS $lib $libobjs~$RANLIB $lib~(cd $output_objdir && a2ixlibrary -32)'
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='-L$libdir'
_LT_TAGVAR(hardcode_minus_L, $1)=yes
;;
esac
;;
bsdi[[45]]*)
_LT_TAGVAR(export_dynamic_flag_spec, $1)=-rdynamic
;;
cygwin* | mingw* | pw32* | cegcc*)
# When not using gcc, we currently assume that we are using
# Microsoft Visual C++.
# hardcode_libdir_flag_spec is actually meaningless, as there is
# no search path for DLLs.
case $cc_basename in
cl*)
# Native MSVC
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)=' '
_LT_TAGVAR(allow_undefined_flag, $1)=unsupported
_LT_TAGVAR(always_export_symbols, $1)=yes
_LT_TAGVAR(file_list_spec, $1)='@'
# Tell ltmain to make .lib files, not .a files.
libext=lib
# Tell ltmain to make .dll files, not .so files.
shrext_cmds=".dll"
# FIXME: Setting linknames here is a bad hack.
_LT_TAGVAR(archive_cmds, $1)='$CC -o $output_objdir/$soname $libobjs $compiler_flags $deplibs -Wl,-dll~linknames='
_LT_TAGVAR(archive_expsym_cmds, $1)='if test "x`$SED \"$sed_uncomment_deffile\" $export_symbols | $SED 1q`" = xEXPORTS; then
sed -n -e 's/\\\\\\\(.*\\\\\\\)/-link\\\ -EXPORT:\\\\\\\1/' -e '1\\\!p' < $export_symbols > $output_objdir/$soname.exp;
else
sed -e 's/\\\\\\\(.*\\\\\\\)/-link\\\ -EXPORT:\\\\\\\1/' < $export_symbols > $output_objdir/$soname.exp;
fi~
$CC -o $tool_output_objdir$soname $libobjs $compiler_flags $deplibs "@$tool_output_objdir$soname.exp" -Wl,-DLL,-IMPLIB:"$tool_output_objdir$libname.dll.lib"~
linknames='
# The linker will not automatically build a static lib if we build a DLL.
# _LT_TAGVAR(old_archive_from_new_cmds, $1)='true'
_LT_TAGVAR(enable_shared_with_static_runtimes, $1)=yes
_LT_TAGVAR(exclude_expsyms, $1)='_NULL_IMPORT_DESCRIPTOR|_IMPORT_DESCRIPTOR_.*'
_LT_TAGVAR(export_symbols_cmds, $1)='$NM $libobjs $convenience | $global_symbol_pipe | $SED -e '\''/^[[BCDGRS]][[ ]]/s/.*[[ ]]\([[^ ]]*\)/\1,DATA/'\'' | $SED -e '\''/^[[AITW]][[ ]]/s/.*[[ ]]//'\'' | sort | uniq > $export_symbols'
# Don't use ranlib
_LT_TAGVAR(old_postinstall_cmds, $1)='chmod 644 $oldlib'
_LT_TAGVAR(postlink_cmds, $1)='lt_outputfile="@OUTPUT@"~
lt_tool_outputfile="@TOOL_OUTPUT@"~
case $lt_outputfile in
*.exe|*.EXE) ;;
*)
lt_outputfile="$lt_outputfile.exe"
lt_tool_outputfile="$lt_tool_outputfile.exe"
;;
esac~
if test "$MANIFEST_TOOL" != ":" && test -f "$lt_outputfile.manifest"; then
$MANIFEST_TOOL -manifest "$lt_tool_outputfile.manifest" -outputresource:"$lt_tool_outputfile" || exit 1;
$RM "$lt_outputfile.manifest";
fi'
;;
*)
# Assume MSVC wrapper
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)=' '
_LT_TAGVAR(allow_undefined_flag, $1)=unsupported
# Tell ltmain to make .lib files, not .a files.
libext=lib
# Tell ltmain to make .dll files, not .so files.
shrext_cmds=".dll"
# FIXME: Setting linknames here is a bad hack.
_LT_TAGVAR(archive_cmds, $1)='$CC -o $lib $libobjs $compiler_flags `func_echo_all "$deplibs" | $SED '\''s/ -lc$//'\''` -link -dll~linknames='
# The linker will automatically build a .lib file if we build a DLL.
_LT_TAGVAR(old_archive_from_new_cmds, $1)='true'
# FIXME: Should let the user specify the lib program.
_LT_TAGVAR(old_archive_cmds, $1)='lib -OUT:$oldlib$oldobjs$old_deplibs'
_LT_TAGVAR(enable_shared_with_static_runtimes, $1)=yes
;;
esac
;;
darwin* | rhapsody*)
_LT_DARWIN_LINKER_FEATURES($1)
;;
dgux*)
_LT_TAGVAR(archive_cmds, $1)='$LD -G -h $soname -o $lib $libobjs $deplibs $linker_flags'
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='-L$libdir'
_LT_TAGVAR(hardcode_shlibpath_var, $1)=no
;;
# FreeBSD 2.2.[012] allows us to include c++rt0.o to get C++ constructor
# support. Future versions do this automatically, but an explicit c++rt0.o
# does not break anything, and helps significantly (at the cost of a little
# extra space).
freebsd2.2*)
_LT_TAGVAR(archive_cmds, $1)='$LD -Bshareable -o $lib $libobjs $deplibs $linker_flags /usr/lib/c++rt0.o'
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='-R$libdir'
_LT_TAGVAR(hardcode_direct, $1)=yes
_LT_TAGVAR(hardcode_shlibpath_var, $1)=no
;;
# Unfortunately, older versions of FreeBSD 2 do not have this feature.
freebsd2.*)
_LT_TAGVAR(archive_cmds, $1)='$LD -Bshareable -o $lib $libobjs $deplibs $linker_flags'
_LT_TAGVAR(hardcode_direct, $1)=yes
_LT_TAGVAR(hardcode_minus_L, $1)=yes
_LT_TAGVAR(hardcode_shlibpath_var, $1)=no
;;
# FreeBSD 3 and greater uses gcc -shared to do shared libraries.
freebsd* | dragonfly*)
_LT_TAGVAR(archive_cmds, $1)='$CC -shared $pic_flag -o $lib $libobjs $deplibs $compiler_flags'
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='-R$libdir'
_LT_TAGVAR(hardcode_direct, $1)=yes
_LT_TAGVAR(hardcode_shlibpath_var, $1)=no
;;
hpux9*)
if test "$GCC" = yes; then
_LT_TAGVAR(archive_cmds, $1)='$RM $output_objdir/$soname~$CC -shared $pic_flag ${wl}+b ${wl}$install_libdir -o $output_objdir/$soname $libobjs $deplibs $compiler_flags~test $output_objdir/$soname = $lib || mv $output_objdir/$soname $lib'
else
_LT_TAGVAR(archive_cmds, $1)='$RM $output_objdir/$soname~$LD -b +b $install_libdir -o $output_objdir/$soname $libobjs $deplibs $linker_flags~test $output_objdir/$soname = $lib || mv $output_objdir/$soname $lib'
fi
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='${wl}+b ${wl}$libdir'
_LT_TAGVAR(hardcode_libdir_separator, $1)=:
_LT_TAGVAR(hardcode_direct, $1)=yes
# hardcode_minus_L: Not really in the search PATH,
# but as the default location of the library.
_LT_TAGVAR(hardcode_minus_L, $1)=yes
_LT_TAGVAR(export_dynamic_flag_spec, $1)='${wl}-E'
;;
hpux10*)
if test "$GCC" = yes && test "$with_gnu_ld" = no; then
_LT_TAGVAR(archive_cmds, $1)='$CC -shared $pic_flag ${wl}+h ${wl}$soname ${wl}+b ${wl}$install_libdir -o $lib $libobjs $deplibs $compiler_flags'
else
_LT_TAGVAR(archive_cmds, $1)='$LD -b +h $soname +b $install_libdir -o $lib $libobjs $deplibs $linker_flags'
fi
if test "$with_gnu_ld" = no; then
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='${wl}+b ${wl}$libdir'
_LT_TAGVAR(hardcode_libdir_separator, $1)=:
_LT_TAGVAR(hardcode_direct, $1)=yes
_LT_TAGVAR(hardcode_direct_absolute, $1)=yes
_LT_TAGVAR(export_dynamic_flag_spec, $1)='${wl}-E'
# hardcode_minus_L: Not really in the search PATH,
# but as the default location of the library.
_LT_TAGVAR(hardcode_minus_L, $1)=yes
fi
;;
hpux11*)
if test "$GCC" = yes && test "$with_gnu_ld" = no; then
case $host_cpu in
hppa*64*)
_LT_TAGVAR(archive_cmds, $1)='$CC -shared ${wl}+h ${wl}$soname -o $lib $libobjs $deplibs $compiler_flags'
;;
ia64*)
_LT_TAGVAR(archive_cmds, $1)='$CC -shared $pic_flag ${wl}+h ${wl}$soname ${wl}+nodefaultrpath -o $lib $libobjs $deplibs $compiler_flags'
;;
*)
_LT_TAGVAR(archive_cmds, $1)='$CC -shared $pic_flag ${wl}+h ${wl}$soname ${wl}+b ${wl}$install_libdir -o $lib $libobjs $deplibs $compiler_flags'
;;
esac
else
case $host_cpu in
hppa*64*)
_LT_TAGVAR(archive_cmds, $1)='$CC -b ${wl}+h ${wl}$soname -o $lib $libobjs $deplibs $compiler_flags'
;;
ia64*)
_LT_TAGVAR(archive_cmds, $1)='$CC -b ${wl}+h ${wl}$soname ${wl}+nodefaultrpath -o $lib $libobjs $deplibs $compiler_flags'
;;
*)
m4_if($1, [], [
# Older versions of the 11.00 compiler do not understand -b yet
# (HP92453-01 A.11.01.20 doesn't, HP92453-01 B.11.X.35175-35176.GP does)
_LT_LINKER_OPTION([if $CC understands -b],
_LT_TAGVAR(lt_cv_prog_compiler__b, $1), [-b],
[_LT_TAGVAR(archive_cmds, $1)='$CC -b ${wl}+h ${wl}$soname ${wl}+b ${wl}$install_libdir -o $lib $libobjs $deplibs $compiler_flags'],
[_LT_TAGVAR(archive_cmds, $1)='$LD -b +h $soname +b $install_libdir -o $lib $libobjs $deplibs $linker_flags'])],
[_LT_TAGVAR(archive_cmds, $1)='$CC -b ${wl}+h ${wl}$soname ${wl}+b ${wl}$install_libdir -o $lib $libobjs $deplibs $compiler_flags'])
;;
esac
fi
if test "$with_gnu_ld" = no; then
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='${wl}+b ${wl}$libdir'
_LT_TAGVAR(hardcode_libdir_separator, $1)=:
case $host_cpu in
hppa*64*|ia64*)
_LT_TAGVAR(hardcode_direct, $1)=no
_LT_TAGVAR(hardcode_shlibpath_var, $1)=no
;;
*)
_LT_TAGVAR(hardcode_direct, $1)=yes
_LT_TAGVAR(hardcode_direct_absolute, $1)=yes
_LT_TAGVAR(export_dynamic_flag_spec, $1)='${wl}-E'
# hardcode_minus_L: Not really in the search PATH,
# but as the default location of the library.
_LT_TAGVAR(hardcode_minus_L, $1)=yes
;;
esac
fi
;;
irix5* | irix6* | nonstopux*)
if test "$GCC" = yes; then
_LT_TAGVAR(archive_cmds, $1)='$CC -shared $pic_flag $libobjs $deplibs $compiler_flags ${wl}-soname ${wl}$soname `test -n "$verstring" && func_echo_all "${wl}-set_version ${wl}$verstring"` ${wl}-update_registry ${wl}${output_objdir}/so_locations -o $lib'
# Try to use the -exported_symbol ld option, if it does not
# work, assume that -exports_file does not work either and
# implicitly export all symbols.
# This should be the same for all languages, so no per-tag cache variable.
AC_CACHE_CHECK([whether the $host_os linker accepts -exported_symbol],
[lt_cv_irix_exported_symbol],
[save_LDFLAGS="$LDFLAGS"
LDFLAGS="$LDFLAGS -shared ${wl}-exported_symbol ${wl}foo ${wl}-update_registry ${wl}/dev/null"
AC_LINK_IFELSE(
[AC_LANG_SOURCE(
[AC_LANG_CASE([C], [[int foo (void) { return 0; }]],
[C++], [[int foo (void) { return 0; }]],
[Fortran 77], [[
subroutine foo
end]],
[Fortran], [[
subroutine foo
end]])])],
[lt_cv_irix_exported_symbol=yes],
[lt_cv_irix_exported_symbol=no])
LDFLAGS="$save_LDFLAGS"])
if test "$lt_cv_irix_exported_symbol" = yes; then
_LT_TAGVAR(archive_expsym_cmds, $1)='$CC -shared $pic_flag $libobjs $deplibs $compiler_flags ${wl}-soname ${wl}$soname `test -n "$verstring" && func_echo_all "${wl}-set_version ${wl}$verstring"` ${wl}-update_registry ${wl}${output_objdir}/so_locations ${wl}-exports_file ${wl}$export_symbols -o $lib'
fi
else
_LT_TAGVAR(archive_cmds, $1)='$CC -shared $libobjs $deplibs $compiler_flags -soname $soname `test -n "$verstring" && func_echo_all "-set_version $verstring"` -update_registry ${output_objdir}/so_locations -o $lib'
_LT_TAGVAR(archive_expsym_cmds, $1)='$CC -shared $libobjs $deplibs $compiler_flags -soname $soname `test -n "$verstring" && func_echo_all "-set_version $verstring"` -update_registry ${output_objdir}/so_locations -exports_file $export_symbols -o $lib'
fi
_LT_TAGVAR(archive_cmds_need_lc, $1)='no'
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='${wl}-rpath ${wl}$libdir'
_LT_TAGVAR(hardcode_libdir_separator, $1)=:
_LT_TAGVAR(inherit_rpath, $1)=yes
_LT_TAGVAR(link_all_deplibs, $1)=yes
;;
netbsd* | netbsdelf*-gnu)
if echo __ELF__ | $CC -E - | $GREP __ELF__ >/dev/null; then
_LT_TAGVAR(archive_cmds, $1)='$LD -Bshareable -o $lib $libobjs $deplibs $linker_flags' # a.out
else
_LT_TAGVAR(archive_cmds, $1)='$LD -shared -o $lib $libobjs $deplibs $linker_flags' # ELF
fi
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='-R$libdir'
_LT_TAGVAR(hardcode_direct, $1)=yes
_LT_TAGVAR(hardcode_shlibpath_var, $1)=no
;;
newsos6)
_LT_TAGVAR(archive_cmds, $1)='$LD -G -h $soname -o $lib $libobjs $deplibs $linker_flags'
_LT_TAGVAR(hardcode_direct, $1)=yes
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='${wl}-rpath ${wl}$libdir'
_LT_TAGVAR(hardcode_libdir_separator, $1)=:
_LT_TAGVAR(hardcode_shlibpath_var, $1)=no
;;
*nto* | *qnx*)
;;
openbsd*)
if test -f /usr/libexec/ld.so; then
_LT_TAGVAR(hardcode_direct, $1)=yes
_LT_TAGVAR(hardcode_shlibpath_var, $1)=no
_LT_TAGVAR(hardcode_direct_absolute, $1)=yes
if test -z "`echo __ELF__ | $CC -E - | $GREP __ELF__`" || test "$host_os-$host_cpu" = "openbsd2.8-powerpc"; then
_LT_TAGVAR(archive_cmds, $1)='$CC -shared $pic_flag -o $lib $libobjs $deplibs $compiler_flags'
_LT_TAGVAR(archive_expsym_cmds, $1)='$CC -shared $pic_flag -o $lib $libobjs $deplibs $compiler_flags ${wl}-retain-symbols-file,$export_symbols'
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='${wl}-rpath,$libdir'
_LT_TAGVAR(export_dynamic_flag_spec, $1)='${wl}-E'
else
case $host_os in
openbsd[[01]].* | openbsd2.[[0-7]] | openbsd2.[[0-7]].*)
_LT_TAGVAR(archive_cmds, $1)='$LD -Bshareable -o $lib $libobjs $deplibs $linker_flags'
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='-R$libdir'
;;
*)
_LT_TAGVAR(archive_cmds, $1)='$CC -shared $pic_flag -o $lib $libobjs $deplibs $compiler_flags'
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='${wl}-rpath,$libdir'
;;
esac
fi
else
_LT_TAGVAR(ld_shlibs, $1)=no
fi
;;
os2*)
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='-L$libdir'
_LT_TAGVAR(hardcode_minus_L, $1)=yes
_LT_TAGVAR(allow_undefined_flag, $1)=unsupported
_LT_TAGVAR(archive_cmds, $1)='$ECHO "LIBRARY $libname INITINSTANCE" > $output_objdir/$libname.def~$ECHO "DESCRIPTION \"$libname\"" >> $output_objdir/$libname.def~echo DATA >> $output_objdir/$libname.def~echo " SINGLE NONSHARED" >> $output_objdir/$libname.def~echo EXPORTS >> $output_objdir/$libname.def~emxexp $libobjs >> $output_objdir/$libname.def~$CC -Zdll -Zcrtdll -o $lib $libobjs $deplibs $compiler_flags $output_objdir/$libname.def'
_LT_TAGVAR(old_archive_from_new_cmds, $1)='emximp -o $output_objdir/$libname.a $output_objdir/$libname.def'
;;
osf3*)
if test "$GCC" = yes; then
_LT_TAGVAR(allow_undefined_flag, $1)=' ${wl}-expect_unresolved ${wl}\*'
_LT_TAGVAR(archive_cmds, $1)='$CC -shared${allow_undefined_flag} $libobjs $deplibs $compiler_flags ${wl}-soname ${wl}$soname `test -n "$verstring" && func_echo_all "${wl}-set_version ${wl}$verstring"` ${wl}-update_registry ${wl}${output_objdir}/so_locations -o $lib'
else
_LT_TAGVAR(allow_undefined_flag, $1)=' -expect_unresolved \*'
_LT_TAGVAR(archive_cmds, $1)='$CC -shared${allow_undefined_flag} $libobjs $deplibs $compiler_flags -soname $soname `test -n "$verstring" && func_echo_all "-set_version $verstring"` -update_registry ${output_objdir}/so_locations -o $lib'
fi
_LT_TAGVAR(archive_cmds_need_lc, $1)='no'
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='${wl}-rpath ${wl}$libdir'
_LT_TAGVAR(hardcode_libdir_separator, $1)=:
;;
osf4* | osf5*) # as osf3* with the addition of -msym flag
if test "$GCC" = yes; then
_LT_TAGVAR(allow_undefined_flag, $1)=' ${wl}-expect_unresolved ${wl}\*'
_LT_TAGVAR(archive_cmds, $1)='$CC -shared${allow_undefined_flag} $pic_flag $libobjs $deplibs $compiler_flags ${wl}-msym ${wl}-soname ${wl}$soname `test -n "$verstring" && func_echo_all "${wl}-set_version ${wl}$verstring"` ${wl}-update_registry ${wl}${output_objdir}/so_locations -o $lib'
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='${wl}-rpath ${wl}$libdir'
else
_LT_TAGVAR(allow_undefined_flag, $1)=' -expect_unresolved \*'
_LT_TAGVAR(archive_cmds, $1)='$CC -shared${allow_undefined_flag} $libobjs $deplibs $compiler_flags -msym -soname $soname `test -n "$verstring" && func_echo_all "-set_version $verstring"` -update_registry ${output_objdir}/so_locations -o $lib'
_LT_TAGVAR(archive_expsym_cmds, $1)='for i in `cat $export_symbols`; do printf "%s %s\\n" -exported_symbol "\$i" >> $lib.exp; done; printf "%s\\n" "-hidden">> $lib.exp~
$CC -shared${allow_undefined_flag} ${wl}-input ${wl}$lib.exp $compiler_flags $libobjs $deplibs -soname $soname `test -n "$verstring" && $ECHO "-set_version $verstring"` -update_registry ${output_objdir}/so_locations -o $lib~$RM $lib.exp'
# Both c and cxx compiler support -rpath directly
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='-rpath $libdir'
fi
_LT_TAGVAR(archive_cmds_need_lc, $1)='no'
_LT_TAGVAR(hardcode_libdir_separator, $1)=:
;;
solaris*)
_LT_TAGVAR(no_undefined_flag, $1)=' -z defs'
if test "$GCC" = yes; then
wlarc='${wl}'
_LT_TAGVAR(archive_cmds, $1)='$CC -shared $pic_flag ${wl}-z ${wl}text ${wl}-h ${wl}$soname -o $lib $libobjs $deplibs $compiler_flags'
_LT_TAGVAR(archive_expsym_cmds, $1)='echo "{ global:" > $lib.exp~cat $export_symbols | $SED -e "s/\(.*\)/\1;/" >> $lib.exp~echo "local: *; };" >> $lib.exp~
$CC -shared $pic_flag ${wl}-z ${wl}text ${wl}-M ${wl}$lib.exp ${wl}-h ${wl}$soname -o $lib $libobjs $deplibs $compiler_flags~$RM $lib.exp'
else
case `$CC -V 2>&1` in
*"Compilers 5.0"*)
wlarc=''
_LT_TAGVAR(archive_cmds, $1)='$LD -G${allow_undefined_flag} -h $soname -o $lib $libobjs $deplibs $linker_flags'
_LT_TAGVAR(archive_expsym_cmds, $1)='echo "{ global:" > $lib.exp~cat $export_symbols | $SED -e "s/\(.*\)/\1;/" >> $lib.exp~echo "local: *; };" >> $lib.exp~
$LD -G${allow_undefined_flag} -M $lib.exp -h $soname -o $lib $libobjs $deplibs $linker_flags~$RM $lib.exp'
;;
*)
wlarc='${wl}'
_LT_TAGVAR(archive_cmds, $1)='$CC -G${allow_undefined_flag} -h $soname -o $lib $libobjs $deplibs $compiler_flags'
_LT_TAGVAR(archive_expsym_cmds, $1)='echo "{ global:" > $lib.exp~cat $export_symbols | $SED -e "s/\(.*\)/\1;/" >> $lib.exp~echo "local: *; };" >> $lib.exp~
$CC -G${allow_undefined_flag} -M $lib.exp -h $soname -o $lib $libobjs $deplibs $compiler_flags~$RM $lib.exp'
;;
esac
fi
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='-R$libdir'
_LT_TAGVAR(hardcode_shlibpath_var, $1)=no
case $host_os in
solaris2.[[0-5]] | solaris2.[[0-5]].*) ;;
*)
# The compiler driver will combine and reorder linker options,
# but understands `-z linker_flag'. GCC discards it without `$wl',
# but is careful enough not to reorder.
# Supported since Solaris 2.6 (maybe 2.5.1?)
if test "$GCC" = yes; then
_LT_TAGVAR(whole_archive_flag_spec, $1)='${wl}-z ${wl}allextract$convenience ${wl}-z ${wl}defaultextract'
else
_LT_TAGVAR(whole_archive_flag_spec, $1)='-z allextract$convenience -z defaultextract'
fi
;;
esac
_LT_TAGVAR(link_all_deplibs, $1)=yes
;;
sunos4*)
if test "x$host_vendor" = xsequent; then
# Use $CC to link under sequent, because it throws in some extra .o
# files that make .init and .fini sections work.
_LT_TAGVAR(archive_cmds, $1)='$CC -G ${wl}-h $soname -o $lib $libobjs $deplibs $compiler_flags'
else
_LT_TAGVAR(archive_cmds, $1)='$LD -assert pure-text -Bstatic -o $lib $libobjs $deplibs $linker_flags'
fi
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='-L$libdir'
_LT_TAGVAR(hardcode_direct, $1)=yes
_LT_TAGVAR(hardcode_minus_L, $1)=yes
_LT_TAGVAR(hardcode_shlibpath_var, $1)=no
;;
sysv4)
case $host_vendor in
sni)
_LT_TAGVAR(archive_cmds, $1)='$LD -G -h $soname -o $lib $libobjs $deplibs $linker_flags'
_LT_TAGVAR(hardcode_direct, $1)=yes # is this really true???
;;
siemens)
## LD is ld it makes a PLAMLIB
## CC just makes a GrossModule.
_LT_TAGVAR(archive_cmds, $1)='$LD -G -o $lib $libobjs $deplibs $linker_flags'
_LT_TAGVAR(reload_cmds, $1)='$CC -r -o $output$reload_objs'
_LT_TAGVAR(hardcode_direct, $1)=no
;;
motorola)
_LT_TAGVAR(archive_cmds, $1)='$LD -G -h $soname -o $lib $libobjs $deplibs $linker_flags'
_LT_TAGVAR(hardcode_direct, $1)=no #Motorola manual says yes, but my tests say they lie
;;
esac
runpath_var='LD_RUN_PATH'
_LT_TAGVAR(hardcode_shlibpath_var, $1)=no
;;
sysv4.3*)
_LT_TAGVAR(archive_cmds, $1)='$LD -G -h $soname -o $lib $libobjs $deplibs $linker_flags'
_LT_TAGVAR(hardcode_shlibpath_var, $1)=no
_LT_TAGVAR(export_dynamic_flag_spec, $1)='-Bexport'
;;
sysv4*MP*)
if test -d /usr/nec; then
_LT_TAGVAR(archive_cmds, $1)='$LD -G -h $soname -o $lib $libobjs $deplibs $linker_flags'
_LT_TAGVAR(hardcode_shlibpath_var, $1)=no
runpath_var=LD_RUN_PATH
hardcode_runpath_var=yes
_LT_TAGVAR(ld_shlibs, $1)=yes
fi
;;
sysv4*uw2* | sysv5OpenUNIX* | sysv5UnixWare7.[[01]].[[10]]* | unixware7* | sco3.2v5.0.[[024]]*)
_LT_TAGVAR(no_undefined_flag, $1)='${wl}-z,text'
_LT_TAGVAR(archive_cmds_need_lc, $1)=no
_LT_TAGVAR(hardcode_shlibpath_var, $1)=no
runpath_var='LD_RUN_PATH'
if test "$GCC" = yes; then
_LT_TAGVAR(archive_cmds, $1)='$CC -shared ${wl}-h,$soname -o $lib $libobjs $deplibs $compiler_flags'
_LT_TAGVAR(archive_expsym_cmds, $1)='$CC -shared ${wl}-Bexport:$export_symbols ${wl}-h,$soname -o $lib $libobjs $deplibs $compiler_flags'
else
_LT_TAGVAR(archive_cmds, $1)='$CC -G ${wl}-h,$soname -o $lib $libobjs $deplibs $compiler_flags'
_LT_TAGVAR(archive_expsym_cmds, $1)='$CC -G ${wl}-Bexport:$export_symbols ${wl}-h,$soname -o $lib $libobjs $deplibs $compiler_flags'
fi
;;
sysv5* | sco3.2v5* | sco5v6*)
# Note: We can NOT use -z defs as we might desire, because we do not
# link with -lc, and that would cause any symbols used from libc to
# always be unresolved, which means just about no library would
# ever link correctly. If we're not using GNU ld we use -z text
# though, which does catch some bad symbols but isn't as heavy-handed
# as -z defs.
_LT_TAGVAR(no_undefined_flag, $1)='${wl}-z,text'
_LT_TAGVAR(allow_undefined_flag, $1)='${wl}-z,nodefs'
_LT_TAGVAR(archive_cmds_need_lc, $1)=no
_LT_TAGVAR(hardcode_shlibpath_var, $1)=no
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='${wl}-R,$libdir'
_LT_TAGVAR(hardcode_libdir_separator, $1)=':'
_LT_TAGVAR(link_all_deplibs, $1)=yes
_LT_TAGVAR(export_dynamic_flag_spec, $1)='${wl}-Bexport'
runpath_var='LD_RUN_PATH'
if test "$GCC" = yes; then
_LT_TAGVAR(archive_cmds, $1)='$CC -shared ${wl}-h,$soname -o $lib $libobjs $deplibs $compiler_flags'
_LT_TAGVAR(archive_expsym_cmds, $1)='$CC -shared ${wl}-Bexport:$export_symbols ${wl}-h,$soname -o $lib $libobjs $deplibs $compiler_flags'
else
_LT_TAGVAR(archive_cmds, $1)='$CC -G ${wl}-h,$soname -o $lib $libobjs $deplibs $compiler_flags'
_LT_TAGVAR(archive_expsym_cmds, $1)='$CC -G ${wl}-Bexport:$export_symbols ${wl}-h,$soname -o $lib $libobjs $deplibs $compiler_flags'
fi
;;
uts4*)
_LT_TAGVAR(archive_cmds, $1)='$LD -G -h $soname -o $lib $libobjs $deplibs $linker_flags'
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='-L$libdir'
_LT_TAGVAR(hardcode_shlibpath_var, $1)=no
;;
*)
_LT_TAGVAR(ld_shlibs, $1)=no
;;
esac
if test x$host_vendor = xsni; then
case $host in
sysv4 | sysv4.2uw2* | sysv4.3* | sysv5*)
_LT_TAGVAR(export_dynamic_flag_spec, $1)='${wl}-Blargedynsym'
;;
esac
fi
fi
])
AC_MSG_RESULT([$_LT_TAGVAR(ld_shlibs, $1)])
test "$_LT_TAGVAR(ld_shlibs, $1)" = no && can_build_shared=no
_LT_TAGVAR(with_gnu_ld, $1)=$with_gnu_ld
_LT_DECL([], [libext], [0], [Old archive suffix (normally "a")])dnl
_LT_DECL([], [shrext_cmds], [1], [Shared library suffix (normally ".so")])dnl
_LT_DECL([], [extract_expsyms_cmds], [2],
[The commands to extract the exported symbol list from a shared archive])
#
# Do we need to explicitly link libc?
#
case "x$_LT_TAGVAR(archive_cmds_need_lc, $1)" in
x|xyes)
# Assume -lc should be added
_LT_TAGVAR(archive_cmds_need_lc, $1)=yes
if test "$enable_shared" = yes && test "$GCC" = yes; then
case $_LT_TAGVAR(archive_cmds, $1) in
*'~'*)
# FIXME: we may have to deal with multi-command sequences.
;;
'$CC '*)
# Test whether the compiler implicitly links with -lc since on some
# systems, -lgcc has to come before -lc. If gcc already passes -lc
# to ld, don't add -lc before -lgcc.
AC_CACHE_CHECK([whether -lc should be explicitly linked in],
[lt_cv_]_LT_TAGVAR(archive_cmds_need_lc, $1),
[$RM conftest*
echo "$lt_simple_compile_test_code" > conftest.$ac_ext
if AC_TRY_EVAL(ac_compile) 2>conftest.err; then
soname=conftest
lib=conftest
libobjs=conftest.$ac_objext
deplibs=
wl=$_LT_TAGVAR(lt_prog_compiler_wl, $1)
pic_flag=$_LT_TAGVAR(lt_prog_compiler_pic, $1)
compiler_flags=-v
linker_flags=-v
verstring=
output_objdir=.
libname=conftest
lt_save_allow_undefined_flag=$_LT_TAGVAR(allow_undefined_flag, $1)
_LT_TAGVAR(allow_undefined_flag, $1)=
if AC_TRY_EVAL(_LT_TAGVAR(archive_cmds, $1) 2\>\&1 \| $GREP \" -lc \" \>/dev/null 2\>\&1)
then
lt_cv_[]_LT_TAGVAR(archive_cmds_need_lc, $1)=no
else
lt_cv_[]_LT_TAGVAR(archive_cmds_need_lc, $1)=yes
fi
_LT_TAGVAR(allow_undefined_flag, $1)=$lt_save_allow_undefined_flag
else
cat conftest.err 1>&5
fi
$RM conftest*
])
_LT_TAGVAR(archive_cmds_need_lc, $1)=$lt_cv_[]_LT_TAGVAR(archive_cmds_need_lc, $1)
;;
esac
fi
;;
esac
_LT_TAGDECL([build_libtool_need_lc], [archive_cmds_need_lc], [0],
[Whether or not to add -lc for building shared libraries])
_LT_TAGDECL([allow_libtool_libs_with_static_runtimes],
[enable_shared_with_static_runtimes], [0],
[Whether or not to disallow shared libs when runtime libs are static])
_LT_TAGDECL([], [export_dynamic_flag_spec], [1],
[Compiler flag to allow reflexive dlopens])
_LT_TAGDECL([], [whole_archive_flag_spec], [1],
[Compiler flag to generate shared objects directly from archives])
_LT_TAGDECL([], [compiler_needs_object], [1],
[Whether the compiler copes with passing no objects directly])
_LT_TAGDECL([], [old_archive_from_new_cmds], [2],
[Create an old-style archive from a shared archive])
_LT_TAGDECL([], [old_archive_from_expsyms_cmds], [2],
[Create a temporary old-style archive to link instead of a shared archive])
_LT_TAGDECL([], [archive_cmds], [2], [Commands used to build a shared archive])
_LT_TAGDECL([], [archive_expsym_cmds], [2])
_LT_TAGDECL([], [module_cmds], [2],
[Commands used to build a loadable module if different from building
a shared archive.])
_LT_TAGDECL([], [module_expsym_cmds], [2])
_LT_TAGDECL([], [with_gnu_ld], [1],
[Whether we are building with GNU ld or not])
_LT_TAGDECL([], [allow_undefined_flag], [1],
[Flag that allows shared libraries with undefined symbols to be built])
_LT_TAGDECL([], [no_undefined_flag], [1],
[Flag that enforces no undefined symbols])
_LT_TAGDECL([], [hardcode_libdir_flag_spec], [1],
[Flag to hardcode $libdir into a binary during linking.
This must work even if $libdir does not exist])
_LT_TAGDECL([], [hardcode_libdir_separator], [1],
[Whether we need a single "-rpath" flag with a separated argument])
_LT_TAGDECL([], [hardcode_direct], [0],
[Set to "yes" if using DIR/libNAME${shared_ext} during linking hardcodes
DIR into the resulting binary])
_LT_TAGDECL([], [hardcode_direct_absolute], [0],
[Set to "yes" if using DIR/libNAME${shared_ext} during linking hardcodes
DIR into the resulting binary and the resulting library dependency is
"absolute", i.e impossible to change by setting ${shlibpath_var} if the
library is relocated])
_LT_TAGDECL([], [hardcode_minus_L], [0],
[Set to "yes" if using the -LDIR flag during linking hardcodes DIR
into the resulting binary])
_LT_TAGDECL([], [hardcode_shlibpath_var], [0],
[Set to "yes" if using SHLIBPATH_VAR=DIR during linking hardcodes DIR
into the resulting binary])
_LT_TAGDECL([], [hardcode_automatic], [0],
[Set to "yes" if building a shared library automatically hardcodes DIR
into the library and all subsequent libraries and executables linked
against it])
_LT_TAGDECL([], [inherit_rpath], [0],
[Set to yes if linker adds runtime paths of dependent libraries
to runtime path list])
_LT_TAGDECL([], [link_all_deplibs], [0],
[Whether libtool must link a program against all its dependency libraries])
_LT_TAGDECL([], [always_export_symbols], [0],
[Set to "yes" if exported symbols are required])
_LT_TAGDECL([], [export_symbols_cmds], [2],
[The commands to list exported symbols])
_LT_TAGDECL([], [exclude_expsyms], [1],
[Symbols that should not be listed in the preloaded symbols])
_LT_TAGDECL([], [include_expsyms], [1],
[Symbols that must always be exported])
_LT_TAGDECL([], [prelink_cmds], [2],
[Commands necessary for linking programs (against libraries) with templates])
_LT_TAGDECL([], [postlink_cmds], [2],
[Commands necessary for finishing linking programs])
_LT_TAGDECL([], [file_list_spec], [1],
[Specify filename containing input files])
dnl FIXME: Not yet implemented
dnl _LT_TAGDECL([], [thread_safe_flag_spec], [1],
dnl [Compiler flag to generate thread safe objects])
])# _LT_LINKER_SHLIBS
# _LT_LANG_C_CONFIG([TAG])
# ------------------------
# Ensure that the configuration variables for a C compiler are suitably
# defined. These variables are subsequently used by _LT_CONFIG to write
# the compiler configuration to `libtool'.
m4_defun([_LT_LANG_C_CONFIG],
[m4_require([_LT_DECL_EGREP])dnl
lt_save_CC="$CC"
AC_LANG_PUSH(C)
# Source file extension for C test sources.
ac_ext=c
# Object file extension for compiled C test sources.
objext=o
_LT_TAGVAR(objext, $1)=$objext
# Code to be used in simple compile tests
lt_simple_compile_test_code="int some_variable = 0;"
# Code to be used in simple link tests
lt_simple_link_test_code='int main(){return(0);}'
_LT_TAG_COMPILER
# Save the default compiler, since it gets overwritten when the other
# tags are being tested, and _LT_TAGVAR(compiler, []) is a NOP.
compiler_DEFAULT=$CC
# save warnings/boilerplate of simple test code
_LT_COMPILER_BOILERPLATE
_LT_LINKER_BOILERPLATE
## CAVEAT EMPTOR:
## There is no encapsulation within the following macros, do not change
## the running order or otherwise move them around unless you know exactly
## what you are doing...
if test -n "$compiler"; then
_LT_COMPILER_NO_RTTI($1)
_LT_COMPILER_PIC($1)
_LT_COMPILER_C_O($1)
_LT_COMPILER_FILE_LOCKS($1)
_LT_LINKER_SHLIBS($1)
_LT_SYS_DYNAMIC_LINKER($1)
_LT_LINKER_HARDCODE_LIBPATH($1)
LT_SYS_DLOPEN_SELF
_LT_CMD_STRIPLIB
# Report which library types will actually be built
AC_MSG_CHECKING([if libtool supports shared libraries])
AC_MSG_RESULT([$can_build_shared])
AC_MSG_CHECKING([whether to build shared libraries])
test "$can_build_shared" = "no" && enable_shared=no
# On AIX, shared libraries and static libraries use the same namespace, and
# are all built from PIC.
case $host_os in
aix3*)
test "$enable_shared" = yes && enable_static=no
if test -n "$RANLIB"; then
archive_cmds="$archive_cmds~\$RANLIB \$lib"
postinstall_cmds='$RANLIB $lib'
fi
;;
aix[[4-9]]*)
if test "$host_cpu" != ia64 && test "$aix_use_runtimelinking" = no ; then
test "$enable_shared" = yes && enable_static=no
fi
;;
esac
AC_MSG_RESULT([$enable_shared])
AC_MSG_CHECKING([whether to build static libraries])
# Make sure either enable_shared or enable_static is yes.
test "$enable_shared" = yes || enable_static=yes
AC_MSG_RESULT([$enable_static])
_LT_CONFIG($1)
fi
AC_LANG_POP
CC="$lt_save_CC"
])# _LT_LANG_C_CONFIG
# _LT_LANG_CXX_CONFIG([TAG])
# --------------------------
# Ensure that the configuration variables for a C++ compiler are suitably
# defined. These variables are subsequently used by _LT_CONFIG to write
# the compiler configuration to `libtool'.
m4_defun([_LT_LANG_CXX_CONFIG],
[m4_require([_LT_FILEUTILS_DEFAULTS])dnl
m4_require([_LT_DECL_EGREP])dnl
m4_require([_LT_PATH_MANIFEST_TOOL])dnl
if test -n "$CXX" && ( test "X$CXX" != "Xno" &&
( (test "X$CXX" = "Xg++" && `g++ -v >/dev/null 2>&1` ) ||
(test "X$CXX" != "Xg++"))) ; then
AC_PROG_CXXCPP
else
_lt_caught_CXX_error=yes
fi
AC_LANG_PUSH(C++)
_LT_TAGVAR(archive_cmds_need_lc, $1)=no
_LT_TAGVAR(allow_undefined_flag, $1)=
_LT_TAGVAR(always_export_symbols, $1)=no
_LT_TAGVAR(archive_expsym_cmds, $1)=
_LT_TAGVAR(compiler_needs_object, $1)=no
_LT_TAGVAR(export_dynamic_flag_spec, $1)=
_LT_TAGVAR(hardcode_direct, $1)=no
_LT_TAGVAR(hardcode_direct_absolute, $1)=no
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)=
_LT_TAGVAR(hardcode_libdir_separator, $1)=
_LT_TAGVAR(hardcode_minus_L, $1)=no
_LT_TAGVAR(hardcode_shlibpath_var, $1)=unsupported
_LT_TAGVAR(hardcode_automatic, $1)=no
_LT_TAGVAR(inherit_rpath, $1)=no
_LT_TAGVAR(module_cmds, $1)=
_LT_TAGVAR(module_expsym_cmds, $1)=
_LT_TAGVAR(link_all_deplibs, $1)=unknown
_LT_TAGVAR(old_archive_cmds, $1)=$old_archive_cmds
_LT_TAGVAR(reload_flag, $1)=$reload_flag
_LT_TAGVAR(reload_cmds, $1)=$reload_cmds
_LT_TAGVAR(no_undefined_flag, $1)=
_LT_TAGVAR(whole_archive_flag_spec, $1)=
_LT_TAGVAR(enable_shared_with_static_runtimes, $1)=no
# Source file extension for C++ test sources.
ac_ext=cpp
# Object file extension for compiled C++ test sources.
objext=o
_LT_TAGVAR(objext, $1)=$objext
# No sense in running all these tests if we already determined that
# the CXX compiler isn't working. Some variables (like enable_shared)
# are currently assumed to apply to all compilers on this platform,
# and will be corrupted by setting them based on a non-working compiler.
if test "$_lt_caught_CXX_error" != yes; then
# Code to be used in simple compile tests
lt_simple_compile_test_code="int some_variable = 0;"
# Code to be used in simple link tests
lt_simple_link_test_code='int main(int, char *[[]]) { return(0); }'
# ltmain only uses $CC for tagged configurations so make sure $CC is set.
_LT_TAG_COMPILER
# save warnings/boilerplate of simple test code
_LT_COMPILER_BOILERPLATE
_LT_LINKER_BOILERPLATE
# Allow CC to be a program name with arguments.
lt_save_CC=$CC
lt_save_CFLAGS=$CFLAGS
lt_save_LD=$LD
lt_save_GCC=$GCC
GCC=$GXX
lt_save_with_gnu_ld=$with_gnu_ld
lt_save_path_LD=$lt_cv_path_LD
if test -n "${lt_cv_prog_gnu_ldcxx+set}"; then
lt_cv_prog_gnu_ld=$lt_cv_prog_gnu_ldcxx
else
$as_unset lt_cv_prog_gnu_ld
fi
if test -n "${lt_cv_path_LDCXX+set}"; then
lt_cv_path_LD=$lt_cv_path_LDCXX
else
$as_unset lt_cv_path_LD
fi
test -z "${LDCXX+set}" || LD=$LDCXX
CC=${CXX-"c++"}
CFLAGS=$CXXFLAGS
compiler=$CC
_LT_TAGVAR(compiler, $1)=$CC
_LT_CC_BASENAME([$compiler])
if test -n "$compiler"; then
# We don't want -fno-exception when compiling C++ code, so set the
# no_builtin_flag separately
if test "$GXX" = yes; then
_LT_TAGVAR(lt_prog_compiler_no_builtin_flag, $1)=' -fno-builtin'
else
_LT_TAGVAR(lt_prog_compiler_no_builtin_flag, $1)=
fi
if test "$GXX" = yes; then
# Set up default GNU C++ configuration
LT_PATH_LD
# Check if GNU C++ uses GNU ld as the underlying linker, since the
# archiving commands below assume that GNU ld is being used.
if test "$with_gnu_ld" = yes; then
_LT_TAGVAR(archive_cmds, $1)='$CC $pic_flag -shared -nostdlib $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags ${wl}-soname $wl$soname -o $lib'
_LT_TAGVAR(archive_expsym_cmds, $1)='$CC $pic_flag -shared -nostdlib $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags ${wl}-soname $wl$soname ${wl}-retain-symbols-file $wl$export_symbols -o $lib'
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='${wl}-rpath ${wl}$libdir'
_LT_TAGVAR(export_dynamic_flag_spec, $1)='${wl}--export-dynamic'
# If archive_cmds runs LD, not CC, wlarc should be empty
# XXX I think wlarc can be eliminated in ltcf-cxx, but I need to
# investigate it a little bit more. (MM)
wlarc='${wl}'
# ancient GNU ld didn't support --whole-archive et. al.
if eval "`$CC -print-prog-name=ld` --help 2>&1" |
$GREP 'no-whole-archive' > /dev/null; then
_LT_TAGVAR(whole_archive_flag_spec, $1)="$wlarc"'--whole-archive$convenience '"$wlarc"'--no-whole-archive'
else
_LT_TAGVAR(whole_archive_flag_spec, $1)=
fi
else
with_gnu_ld=no
wlarc=
# A generic and very simple default shared library creation
# command for GNU C++ for the case where it uses the native
# linker, instead of GNU ld. If possible, this setting should
# overridden to take advantage of the native linker features on
# the platform it is being used on.
_LT_TAGVAR(archive_cmds, $1)='$CC -shared -nostdlib $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags -o $lib'
fi
# Commands to make compiler produce verbose output that lists
# what "hidden" libraries, object files and flags are used when
# linking a shared library.
output_verbose_link_cmd='$CC -shared $CFLAGS -v conftest.$objext 2>&1 | $GREP -v "^Configured with:" | $GREP "\-L"'
else
GXX=no
with_gnu_ld=no
wlarc=
fi
# PORTME: fill in a description of your system's C++ link characteristics
AC_MSG_CHECKING([whether the $compiler linker ($LD) supports shared libraries])
_LT_TAGVAR(ld_shlibs, $1)=yes
case $host_os in
aix3*)
# FIXME: insert proper C++ library support
_LT_TAGVAR(ld_shlibs, $1)=no
;;
aix[[4-9]]*)
if test "$host_cpu" = ia64; then
# On IA64, the linker does run time linking by default, so we don't
# have to do anything special.
aix_use_runtimelinking=no
exp_sym_flag='-Bexport'
no_entry_flag=""
else
aix_use_runtimelinking=no
# Test if we are trying to use run time linking or normal
# AIX style linking. If -brtl is somewhere in LDFLAGS, we
# need to do runtime linking.
case $host_os in aix4.[[23]]|aix4.[[23]].*|aix[[5-9]]*)
for ld_flag in $LDFLAGS; do
case $ld_flag in
*-brtl*)
aix_use_runtimelinking=yes
break
;;
esac
done
;;
esac
exp_sym_flag='-bexport'
no_entry_flag='-bnoentry'
fi
# When large executables or shared objects are built, AIX ld can
# have problems creating the table of contents. If linking a library
# or program results in "error TOC overflow" add -mminimal-toc to
# CXXFLAGS/CFLAGS for g++/gcc. In the cases where that is not
# enough to fix the problem, add -Wl,-bbigtoc to LDFLAGS.
_LT_TAGVAR(archive_cmds, $1)=''
_LT_TAGVAR(hardcode_direct, $1)=yes
_LT_TAGVAR(hardcode_direct_absolute, $1)=yes
_LT_TAGVAR(hardcode_libdir_separator, $1)=':'
_LT_TAGVAR(link_all_deplibs, $1)=yes
_LT_TAGVAR(file_list_spec, $1)='${wl}-f,'
if test "$GXX" = yes; then
case $host_os in aix4.[[012]]|aix4.[[012]].*)
# We only want to do this on AIX 4.2 and lower, the check
# below for broken collect2 doesn't work under 4.3+
collect2name=`${CC} -print-prog-name=collect2`
if test -f "$collect2name" &&
strings "$collect2name" | $GREP resolve_lib_name >/dev/null
then
# We have reworked collect2
:
else
# We have old collect2
_LT_TAGVAR(hardcode_direct, $1)=unsupported
# It fails to find uninstalled libraries when the uninstalled
# path is not listed in the libpath. Setting hardcode_minus_L
# to unsupported forces relinking
_LT_TAGVAR(hardcode_minus_L, $1)=yes
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='-L$libdir'
_LT_TAGVAR(hardcode_libdir_separator, $1)=
fi
esac
shared_flag='-shared'
if test "$aix_use_runtimelinking" = yes; then
shared_flag="$shared_flag "'${wl}-G'
fi
else
# not using gcc
if test "$host_cpu" = ia64; then
# VisualAge C++, Version 5.5 for AIX 5L for IA-64, Beta 3 Release
# chokes on -Wl,-G. The following line is correct:
shared_flag='-G'
else
if test "$aix_use_runtimelinking" = yes; then
shared_flag='${wl}-G'
else
shared_flag='${wl}-bM:SRE'
fi
fi
fi
_LT_TAGVAR(export_dynamic_flag_spec, $1)='${wl}-bexpall'
# It seems that -bexpall does not export symbols beginning with
# underscore (_), so it is better to generate a list of symbols to
# export.
_LT_TAGVAR(always_export_symbols, $1)=yes
if test "$aix_use_runtimelinking" = yes; then
# Warning - without using the other runtime loading flags (-brtl),
# -berok will link without error, but may produce a broken library.
_LT_TAGVAR(allow_undefined_flag, $1)='-berok'
# Determine the default libpath from the value encoded in an empty
# executable.
_LT_SYS_MODULE_PATH_AIX([$1])
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='${wl}-blibpath:$libdir:'"$aix_libpath"
_LT_TAGVAR(archive_expsym_cmds, $1)='$CC -o $output_objdir/$soname $libobjs $deplibs '"\${wl}$no_entry_flag"' $compiler_flags `if test "x${allow_undefined_flag}" != "x"; then func_echo_all "${wl}${allow_undefined_flag}"; else :; fi` '"\${wl}$exp_sym_flag:\$export_symbols $shared_flag"
else
if test "$host_cpu" = ia64; then
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='${wl}-R $libdir:/usr/lib:/lib'
_LT_TAGVAR(allow_undefined_flag, $1)="-z nodefs"
_LT_TAGVAR(archive_expsym_cmds, $1)="\$CC $shared_flag"' -o $output_objdir/$soname $libobjs $deplibs '"\${wl}$no_entry_flag"' $compiler_flags ${wl}${allow_undefined_flag} '"\${wl}$exp_sym_flag:\$export_symbols"
else
# Determine the default libpath from the value encoded in an
# empty executable.
_LT_SYS_MODULE_PATH_AIX([$1])
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='${wl}-blibpath:$libdir:'"$aix_libpath"
# Warning - without using the other run time loading flags,
# -berok will link without error, but may produce a broken library.
_LT_TAGVAR(no_undefined_flag, $1)=' ${wl}-bernotok'
_LT_TAGVAR(allow_undefined_flag, $1)=' ${wl}-berok'
if test "$with_gnu_ld" = yes; then
# We only use this code for GNU lds that support --whole-archive.
_LT_TAGVAR(whole_archive_flag_spec, $1)='${wl}--whole-archive$convenience ${wl}--no-whole-archive'
else
# Exported symbols can be pulled into shared objects from archives
_LT_TAGVAR(whole_archive_flag_spec, $1)='$convenience'
fi
_LT_TAGVAR(archive_cmds_need_lc, $1)=yes
# This is similar to how AIX traditionally builds its shared
# libraries.
_LT_TAGVAR(archive_expsym_cmds, $1)="\$CC $shared_flag"' -o $output_objdir/$soname $libobjs $deplibs ${wl}-bnoentry $compiler_flags ${wl}-bE:$export_symbols${allow_undefined_flag}~$AR $AR_FLAGS $output_objdir/$libname$release.a $output_objdir/$soname'
fi
fi
;;
beos*)
if $LD --help 2>&1 | $GREP ': supported targets:.* elf' > /dev/null; then
_LT_TAGVAR(allow_undefined_flag, $1)=unsupported
# Joseph Beckenbach says some releases of gcc
# support --undefined. This deserves some investigation. FIXME
_LT_TAGVAR(archive_cmds, $1)='$CC -nostart $libobjs $deplibs $compiler_flags ${wl}-soname $wl$soname -o $lib'
else
_LT_TAGVAR(ld_shlibs, $1)=no
fi
;;
chorus*)
case $cc_basename in
*)
# FIXME: insert proper C++ library support
_LT_TAGVAR(ld_shlibs, $1)=no
;;
esac
;;
cygwin* | mingw* | pw32* | cegcc*)
case $GXX,$cc_basename in
,cl* | no,cl*)
# Native MSVC
# hardcode_libdir_flag_spec is actually meaningless, as there is
# no search path for DLLs.
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)=' '
_LT_TAGVAR(allow_undefined_flag, $1)=unsupported
_LT_TAGVAR(always_export_symbols, $1)=yes
_LT_TAGVAR(file_list_spec, $1)='@'
# Tell ltmain to make .lib files, not .a files.
libext=lib
# Tell ltmain to make .dll files, not .so files.
shrext_cmds=".dll"
# FIXME: Setting linknames here is a bad hack.
_LT_TAGVAR(archive_cmds, $1)='$CC -o $output_objdir/$soname $libobjs $compiler_flags $deplibs -Wl,-dll~linknames='
_LT_TAGVAR(archive_expsym_cmds, $1)='if test "x`$SED \"$sed_uncomment_deffile\" $export_symbols | $SED 1q`" = xEXPORTS; then
$SED -n -e 's/\\\\\\\(.*\\\\\\\)/-link\\\ -EXPORT:\\\\\\\1/' -e '1\\\!p' < $export_symbols > $output_objdir/$soname.exp;
else
$SED -e 's/\\\\\\\(.*\\\\\\\)/-link\\\ -EXPORT:\\\\\\\1/' < $export_symbols > $output_objdir/$soname.exp;
fi~
$CC -o $tool_output_objdir$soname $libobjs $compiler_flags $deplibs "@$tool_output_objdir$soname.exp" -Wl,-DLL,-IMPLIB:"$tool_output_objdir$libname.dll.lib"~
linknames='
# The linker will not automatically build a static lib if we build a DLL.
# _LT_TAGVAR(old_archive_from_new_cmds, $1)='true'
_LT_TAGVAR(enable_shared_with_static_runtimes, $1)=yes
# Don't use ranlib
_LT_TAGVAR(old_postinstall_cmds, $1)='chmod 644 $oldlib'
_LT_TAGVAR(postlink_cmds, $1)='lt_outputfile="@OUTPUT@"~
lt_tool_outputfile="@TOOL_OUTPUT@"~
case $lt_outputfile in
*.exe|*.EXE) ;;
*)
lt_outputfile="$lt_outputfile.exe"
lt_tool_outputfile="$lt_tool_outputfile.exe"
;;
esac~
func_to_tool_file "$lt_outputfile"~
if test "$MANIFEST_TOOL" != ":" && test -f "$lt_outputfile.manifest"; then
$MANIFEST_TOOL -manifest "$lt_tool_outputfile.manifest" -outputresource:"$lt_tool_outputfile" || exit 1;
$RM "$lt_outputfile.manifest";
fi'
;;
*)
# g++
# _LT_TAGVAR(hardcode_libdir_flag_spec, $1) is actually meaningless,
# as there is no search path for DLLs.
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='-L$libdir'
_LT_TAGVAR(export_dynamic_flag_spec, $1)='${wl}--export-all-symbols'
_LT_TAGVAR(allow_undefined_flag, $1)=unsupported
_LT_TAGVAR(always_export_symbols, $1)=no
_LT_TAGVAR(enable_shared_with_static_runtimes, $1)=yes
if $LD --help 2>&1 | $GREP 'auto-import' > /dev/null; then
_LT_TAGVAR(archive_cmds, $1)='$CC -shared -nostdlib $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags -o $output_objdir/$soname ${wl}--enable-auto-image-base -Xlinker --out-implib -Xlinker $lib'
# If the export-symbols file already is a .def file (1st line
# is EXPORTS), use it as is; otherwise, prepend...
_LT_TAGVAR(archive_expsym_cmds, $1)='if test "x`$SED \"$sed_uncomment_deffile\" $export_symbols | $SED 1q`" = xEXPORTS; then
cp $export_symbols $output_objdir/$soname.def;
else
echo EXPORTS > $output_objdir/$soname.def;
cat $export_symbols >> $output_objdir/$soname.def;
fi~
$CC -shared -nostdlib $output_objdir/$soname.def $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags -o $output_objdir/$soname ${wl}--enable-auto-image-base -Xlinker --out-implib -Xlinker $lib'
else
_LT_TAGVAR(ld_shlibs, $1)=no
fi
;;
esac
;;
darwin* | rhapsody*)
_LT_DARWIN_LINKER_FEATURES($1)
;;
dgux*)
case $cc_basename in
ec++*)
# FIXME: insert proper C++ library support
_LT_TAGVAR(ld_shlibs, $1)=no
;;
ghcx*)
# Green Hills C++ Compiler
# FIXME: insert proper C++ library support
_LT_TAGVAR(ld_shlibs, $1)=no
;;
*)
# FIXME: insert proper C++ library support
_LT_TAGVAR(ld_shlibs, $1)=no
;;
esac
;;
freebsd2.*)
# C++ shared libraries reported to be fairly broken before
# switch to ELF
_LT_TAGVAR(ld_shlibs, $1)=no
;;
freebsd-elf*)
_LT_TAGVAR(archive_cmds_need_lc, $1)=no
;;
freebsd* | dragonfly*)
# FreeBSD 3 and later use GNU C++ and GNU ld with standard ELF
# conventions
_LT_TAGVAR(ld_shlibs, $1)=yes
;;
gnu*)
;;
haiku*)
_LT_TAGVAR(archive_cmds, $1)='$CC -shared $libobjs $deplibs $compiler_flags ${wl}-soname $wl$soname -o $lib'
_LT_TAGVAR(link_all_deplibs, $1)=yes
;;
hpux9*)
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='${wl}+b ${wl}$libdir'
_LT_TAGVAR(hardcode_libdir_separator, $1)=:
_LT_TAGVAR(export_dynamic_flag_spec, $1)='${wl}-E'
_LT_TAGVAR(hardcode_direct, $1)=yes
_LT_TAGVAR(hardcode_minus_L, $1)=yes # Not in the search PATH,
# but as the default
# location of the library.
case $cc_basename in
CC*)
# FIXME: insert proper C++ library support
_LT_TAGVAR(ld_shlibs, $1)=no
;;
aCC*)
_LT_TAGVAR(archive_cmds, $1)='$RM $output_objdir/$soname~$CC -b ${wl}+b ${wl}$install_libdir -o $output_objdir/$soname $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags~test $output_objdir/$soname = $lib || mv $output_objdir/$soname $lib'
# Commands to make compiler produce verbose output that lists
# what "hidden" libraries, object files and flags are used when
# linking a shared library.
#
# There doesn't appear to be a way to prevent this compiler from
# explicitly linking system object files so we need to strip them
# from the output so that they don't get included in the library
# dependencies.
output_verbose_link_cmd='templist=`($CC -b $CFLAGS -v conftest.$objext 2>&1) | $EGREP "\-L"`; list=""; for z in $templist; do case $z in conftest.$objext) list="$list $z";; *.$objext);; *) list="$list $z";;esac; done; func_echo_all "$list"'
;;
*)
if test "$GXX" = yes; then
_LT_TAGVAR(archive_cmds, $1)='$RM $output_objdir/$soname~$CC -shared -nostdlib $pic_flag ${wl}+b ${wl}$install_libdir -o $output_objdir/$soname $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags~test $output_objdir/$soname = $lib || mv $output_objdir/$soname $lib'
else
# FIXME: insert proper C++ library support
_LT_TAGVAR(ld_shlibs, $1)=no
fi
;;
esac
;;
hpux10*|hpux11*)
if test $with_gnu_ld = no; then
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='${wl}+b ${wl}$libdir'
_LT_TAGVAR(hardcode_libdir_separator, $1)=:
case $host_cpu in
hppa*64*|ia64*)
;;
*)
_LT_TAGVAR(export_dynamic_flag_spec, $1)='${wl}-E'
;;
esac
fi
case $host_cpu in
hppa*64*|ia64*)
_LT_TAGVAR(hardcode_direct, $1)=no
_LT_TAGVAR(hardcode_shlibpath_var, $1)=no
;;
*)
_LT_TAGVAR(hardcode_direct, $1)=yes
_LT_TAGVAR(hardcode_direct_absolute, $1)=yes
_LT_TAGVAR(hardcode_minus_L, $1)=yes # Not in the search PATH,
# but as the default
# location of the library.
;;
esac
case $cc_basename in
CC*)
# FIXME: insert proper C++ library support
_LT_TAGVAR(ld_shlibs, $1)=no
;;
aCC*)
case $host_cpu in
hppa*64*)
_LT_TAGVAR(archive_cmds, $1)='$CC -b ${wl}+h ${wl}$soname -o $lib $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags'
;;
ia64*)
_LT_TAGVAR(archive_cmds, $1)='$CC -b ${wl}+h ${wl}$soname ${wl}+nodefaultrpath -o $lib $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags'
;;
*)
_LT_TAGVAR(archive_cmds, $1)='$CC -b ${wl}+h ${wl}$soname ${wl}+b ${wl}$install_libdir -o $lib $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags'
;;
esac
# Commands to make compiler produce verbose output that lists
# what "hidden" libraries, object files and flags are used when
# linking a shared library.
#
# There doesn't appear to be a way to prevent this compiler from
# explicitly linking system object files so we need to strip them
# from the output so that they don't get included in the library
# dependencies.
output_verbose_link_cmd='templist=`($CC -b $CFLAGS -v conftest.$objext 2>&1) | $GREP "\-L"`; list=""; for z in $templist; do case $z in conftest.$objext) list="$list $z";; *.$objext);; *) list="$list $z";;esac; done; func_echo_all "$list"'
;;
*)
if test "$GXX" = yes; then
if test $with_gnu_ld = no; then
case $host_cpu in
hppa*64*)
_LT_TAGVAR(archive_cmds, $1)='$CC -shared -nostdlib -fPIC ${wl}+h ${wl}$soname -o $lib $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags'
;;
ia64*)
_LT_TAGVAR(archive_cmds, $1)='$CC -shared -nostdlib $pic_flag ${wl}+h ${wl}$soname ${wl}+nodefaultrpath -o $lib $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags'
;;
*)
_LT_TAGVAR(archive_cmds, $1)='$CC -shared -nostdlib $pic_flag ${wl}+h ${wl}$soname ${wl}+b ${wl}$install_libdir -o $lib $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags'
;;
esac
fi
else
# FIXME: insert proper C++ library support
_LT_TAGVAR(ld_shlibs, $1)=no
fi
;;
esac
;;
interix[[3-9]]*)
_LT_TAGVAR(hardcode_direct, $1)=no
_LT_TAGVAR(hardcode_shlibpath_var, $1)=no
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='${wl}-rpath,$libdir'
_LT_TAGVAR(export_dynamic_flag_spec, $1)='${wl}-E'
# Hack: On Interix 3.x, we cannot compile PIC because of a broken gcc.
# Instead, shared libraries are loaded at an image base (0x10000000 by
# default) and relocated if they conflict, which is a slow very memory
# consuming and fragmenting process. To avoid this, we pick a random,
# 256 KiB-aligned image base between 0x50000000 and 0x6FFC0000 at link
# time. Moving up from 0x10000000 also allows more sbrk(2) space.
_LT_TAGVAR(archive_cmds, $1)='$CC -shared $pic_flag $libobjs $deplibs $compiler_flags ${wl}-h,$soname ${wl}--image-base,`expr ${RANDOM-$$} % 4096 / 2 \* 262144 + 1342177280` -o $lib'
_LT_TAGVAR(archive_expsym_cmds, $1)='sed "s,^,_," $export_symbols >$output_objdir/$soname.expsym~$CC -shared $pic_flag $libobjs $deplibs $compiler_flags ${wl}-h,$soname ${wl}--retain-symbols-file,$output_objdir/$soname.expsym ${wl}--image-base,`expr ${RANDOM-$$} % 4096 / 2 \* 262144 + 1342177280` -o $lib'
;;
irix5* | irix6*)
case $cc_basename in
CC*)
# SGI C++
_LT_TAGVAR(archive_cmds, $1)='$CC -shared -all -multigot $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags -soname $soname `test -n "$verstring" && func_echo_all "-set_version $verstring"` -update_registry ${output_objdir}/so_locations -o $lib'
# Archives containing C++ object files must be created using
# "CC -ar", where "CC" is the IRIX C++ compiler. This is
# necessary to make sure instantiated templates are included
# in the archive.
_LT_TAGVAR(old_archive_cmds, $1)='$CC -ar -WR,-u -o $oldlib $oldobjs'
;;
*)
if test "$GXX" = yes; then
if test "$with_gnu_ld" = no; then
_LT_TAGVAR(archive_cmds, $1)='$CC -shared $pic_flag -nostdlib $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags ${wl}-soname ${wl}$soname `test -n "$verstring" && func_echo_all "${wl}-set_version ${wl}$verstring"` ${wl}-update_registry ${wl}${output_objdir}/so_locations -o $lib'
else
_LT_TAGVAR(archive_cmds, $1)='$CC -shared $pic_flag -nostdlib $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags ${wl}-soname ${wl}$soname `test -n "$verstring" && func_echo_all "${wl}-set_version ${wl}$verstring"` -o $lib'
fi
fi
_LT_TAGVAR(link_all_deplibs, $1)=yes
;;
esac
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='${wl}-rpath ${wl}$libdir'
_LT_TAGVAR(hardcode_libdir_separator, $1)=:
_LT_TAGVAR(inherit_rpath, $1)=yes
;;
linux* | k*bsd*-gnu | kopensolaris*-gnu)
case $cc_basename in
KCC*)
# Kuck and Associates, Inc. (KAI) C++ Compiler
# KCC will only create a shared library if the output file
# ends with ".so" (or ".sl" for HP-UX), so rename the library
# to its proper name (with version) after linking.
_LT_TAGVAR(archive_cmds, $1)='tempext=`echo $shared_ext | $SED -e '\''s/\([[^()0-9A-Za-z{}]]\)/\\\\\1/g'\''`; templib=`echo $lib | $SED -e "s/\${tempext}\..*/.so/"`; $CC $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags --soname $soname -o \$templib; mv \$templib $lib'
_LT_TAGVAR(archive_expsym_cmds, $1)='tempext=`echo $shared_ext | $SED -e '\''s/\([[^()0-9A-Za-z{}]]\)/\\\\\1/g'\''`; templib=`echo $lib | $SED -e "s/\${tempext}\..*/.so/"`; $CC $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags --soname $soname -o \$templib ${wl}-retain-symbols-file,$export_symbols; mv \$templib $lib'
# Commands to make compiler produce verbose output that lists
# what "hidden" libraries, object files and flags are used when
# linking a shared library.
#
# There doesn't appear to be a way to prevent this compiler from
# explicitly linking system object files so we need to strip them
# from the output so that they don't get included in the library
# dependencies.
output_verbose_link_cmd='templist=`$CC $CFLAGS -v conftest.$objext -o libconftest$shared_ext 2>&1 | $GREP "ld"`; rm -f libconftest$shared_ext; list=""; for z in $templist; do case $z in conftest.$objext) list="$list $z";; *.$objext);; *) list="$list $z";;esac; done; func_echo_all "$list"'
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='${wl}-rpath,$libdir'
_LT_TAGVAR(export_dynamic_flag_spec, $1)='${wl}--export-dynamic'
# Archives containing C++ object files must be created using
# "CC -Bstatic", where "CC" is the KAI C++ compiler.
_LT_TAGVAR(old_archive_cmds, $1)='$CC -Bstatic -o $oldlib $oldobjs'
;;
icpc* | ecpc* )
# Intel C++
with_gnu_ld=yes
# version 8.0 and above of icpc choke on multiply defined symbols
# if we add $predep_objects and $postdep_objects, however 7.1 and
# earlier do not add the objects themselves.
case `$CC -V 2>&1` in
*"Version 7."*)
_LT_TAGVAR(archive_cmds, $1)='$CC -shared $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags ${wl}-soname $wl$soname -o $lib'
_LT_TAGVAR(archive_expsym_cmds, $1)='$CC -shared $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags ${wl}-soname $wl$soname ${wl}-retain-symbols-file $wl$export_symbols -o $lib'
;;
*) # Version 8.0 or newer
tmp_idyn=
case $host_cpu in
ia64*) tmp_idyn=' -i_dynamic';;
esac
_LT_TAGVAR(archive_cmds, $1)='$CC -shared'"$tmp_idyn"' $libobjs $deplibs $compiler_flags ${wl}-soname $wl$soname -o $lib'
_LT_TAGVAR(archive_expsym_cmds, $1)='$CC -shared'"$tmp_idyn"' $libobjs $deplibs $compiler_flags ${wl}-soname $wl$soname ${wl}-retain-symbols-file $wl$export_symbols -o $lib'
;;
esac
_LT_TAGVAR(archive_cmds_need_lc, $1)=no
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='${wl}-rpath,$libdir'
_LT_TAGVAR(export_dynamic_flag_spec, $1)='${wl}--export-dynamic'
_LT_TAGVAR(whole_archive_flag_spec, $1)='${wl}--whole-archive$convenience ${wl}--no-whole-archive'
;;
pgCC* | pgcpp*)
# Portland Group C++ compiler
case `$CC -V` in
*pgCC\ [[1-5]].* | *pgcpp\ [[1-5]].*)
_LT_TAGVAR(prelink_cmds, $1)='tpldir=Template.dir~
rm -rf $tpldir~
$CC --prelink_objects --instantiation_dir $tpldir $objs $libobjs $compile_deplibs~
compile_command="$compile_command `find $tpldir -name \*.o | sort | $NL2SP`"'
_LT_TAGVAR(old_archive_cmds, $1)='tpldir=Template.dir~
rm -rf $tpldir~
$CC --prelink_objects --instantiation_dir $tpldir $oldobjs$old_deplibs~
$AR $AR_FLAGS $oldlib$oldobjs$old_deplibs `find $tpldir -name \*.o | sort | $NL2SP`~
$RANLIB $oldlib'
_LT_TAGVAR(archive_cmds, $1)='tpldir=Template.dir~
rm -rf $tpldir~
$CC --prelink_objects --instantiation_dir $tpldir $predep_objects $libobjs $deplibs $convenience $postdep_objects~
$CC -shared $pic_flag $predep_objects $libobjs $deplibs `find $tpldir -name \*.o | sort | $NL2SP` $postdep_objects $compiler_flags ${wl}-soname ${wl}$soname -o $lib'
_LT_TAGVAR(archive_expsym_cmds, $1)='tpldir=Template.dir~
rm -rf $tpldir~
$CC --prelink_objects --instantiation_dir $tpldir $predep_objects $libobjs $deplibs $convenience $postdep_objects~
$CC -shared $pic_flag $predep_objects $libobjs $deplibs `find $tpldir -name \*.o | sort | $NL2SP` $postdep_objects $compiler_flags ${wl}-soname ${wl}$soname ${wl}-retain-symbols-file ${wl}$export_symbols -o $lib'
;;
*) # Version 6 and above use weak symbols
_LT_TAGVAR(archive_cmds, $1)='$CC -shared $pic_flag $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags ${wl}-soname ${wl}$soname -o $lib'
_LT_TAGVAR(archive_expsym_cmds, $1)='$CC -shared $pic_flag $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags ${wl}-soname ${wl}$soname ${wl}-retain-symbols-file ${wl}$export_symbols -o $lib'
;;
esac
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='${wl}--rpath ${wl}$libdir'
_LT_TAGVAR(export_dynamic_flag_spec, $1)='${wl}--export-dynamic'
_LT_TAGVAR(whole_archive_flag_spec, $1)='${wl}--whole-archive`for conv in $convenience\"\"; do test -n \"$conv\" && new_convenience=\"$new_convenience,$conv\"; done; func_echo_all \"$new_convenience\"` ${wl}--no-whole-archive'
;;
cxx*)
# Compaq C++
_LT_TAGVAR(archive_cmds, $1)='$CC -shared $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags ${wl}-soname $wl$soname -o $lib'
_LT_TAGVAR(archive_expsym_cmds, $1)='$CC -shared $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags ${wl}-soname $wl$soname -o $lib ${wl}-retain-symbols-file $wl$export_symbols'
runpath_var=LD_RUN_PATH
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='-rpath $libdir'
_LT_TAGVAR(hardcode_libdir_separator, $1)=:
# Commands to make compiler produce verbose output that lists
# what "hidden" libraries, object files and flags are used when
# linking a shared library.
#
# There doesn't appear to be a way to prevent this compiler from
# explicitly linking system object files so we need to strip them
# from the output so that they don't get included in the library
# dependencies.
output_verbose_link_cmd='templist=`$CC -shared $CFLAGS -v conftest.$objext 2>&1 | $GREP "ld"`; templist=`func_echo_all "$templist" | $SED "s/\(^.*ld.*\)\( .*ld .*$\)/\1/"`; list=""; for z in $templist; do case $z in conftest.$objext) list="$list $z";; *.$objext);; *) list="$list $z";;esac; done; func_echo_all "X$list" | $Xsed'
;;
xl* | mpixl* | bgxl*)
# IBM XL 8.0 on PPC, with GNU ld
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='${wl}-rpath ${wl}$libdir'
_LT_TAGVAR(export_dynamic_flag_spec, $1)='${wl}--export-dynamic'
_LT_TAGVAR(archive_cmds, $1)='$CC -qmkshrobj $libobjs $deplibs $compiler_flags ${wl}-soname $wl$soname -o $lib'
if test "x$supports_anon_versioning" = xyes; then
_LT_TAGVAR(archive_expsym_cmds, $1)='echo "{ global:" > $output_objdir/$libname.ver~
cat $export_symbols | sed -e "s/\(.*\)/\1;/" >> $output_objdir/$libname.ver~
echo "local: *; };" >> $output_objdir/$libname.ver~
$CC -qmkshrobj $libobjs $deplibs $compiler_flags ${wl}-soname $wl$soname ${wl}-version-script ${wl}$output_objdir/$libname.ver -o $lib'
fi
;;
*)
case `$CC -V 2>&1 | sed 5q` in
*Sun\ C*)
# Sun C++ 5.9
_LT_TAGVAR(no_undefined_flag, $1)=' -zdefs'
_LT_TAGVAR(archive_cmds, $1)='$CC -G${allow_undefined_flag} -h$soname -o $lib $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags'
_LT_TAGVAR(archive_expsym_cmds, $1)='$CC -G${allow_undefined_flag} -h$soname -o $lib $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags ${wl}-retain-symbols-file ${wl}$export_symbols'
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='-R$libdir'
_LT_TAGVAR(whole_archive_flag_spec, $1)='${wl}--whole-archive`new_convenience=; for conv in $convenience\"\"; do test -z \"$conv\" || new_convenience=\"$new_convenience,$conv\"; done; func_echo_all \"$new_convenience\"` ${wl}--no-whole-archive'
_LT_TAGVAR(compiler_needs_object, $1)=yes
# Not sure whether something based on
# $CC $CFLAGS -v conftest.$objext -o libconftest$shared_ext 2>&1
# would be better.
output_verbose_link_cmd='func_echo_all'
# Archives containing C++ object files must be created using
# "CC -xar", where "CC" is the Sun C++ compiler. This is
# necessary to make sure instantiated templates are included
# in the archive.
_LT_TAGVAR(old_archive_cmds, $1)='$CC -xar -o $oldlib $oldobjs'
;;
esac
;;
esac
;;
lynxos*)
# FIXME: insert proper C++ library support
_LT_TAGVAR(ld_shlibs, $1)=no
;;
m88k*)
# FIXME: insert proper C++ library support
_LT_TAGVAR(ld_shlibs, $1)=no
;;
mvs*)
case $cc_basename in
cxx*)
# FIXME: insert proper C++ library support
_LT_TAGVAR(ld_shlibs, $1)=no
;;
*)
# FIXME: insert proper C++ library support
_LT_TAGVAR(ld_shlibs, $1)=no
;;
esac
;;
netbsd*)
if echo __ELF__ | $CC -E - | $GREP __ELF__ >/dev/null; then
_LT_TAGVAR(archive_cmds, $1)='$LD -Bshareable -o $lib $predep_objects $libobjs $deplibs $postdep_objects $linker_flags'
wlarc=
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='-R$libdir'
_LT_TAGVAR(hardcode_direct, $1)=yes
_LT_TAGVAR(hardcode_shlibpath_var, $1)=no
fi
# Workaround some broken pre-1.5 toolchains
output_verbose_link_cmd='$CC -shared $CFLAGS -v conftest.$objext 2>&1 | $GREP conftest.$objext | $SED -e "s:-lgcc -lc -lgcc::"'
;;
*nto* | *qnx*)
_LT_TAGVAR(ld_shlibs, $1)=yes
;;
openbsd2*)
# C++ shared libraries are fairly broken
_LT_TAGVAR(ld_shlibs, $1)=no
;;
openbsd*)
if test -f /usr/libexec/ld.so; then
_LT_TAGVAR(hardcode_direct, $1)=yes
_LT_TAGVAR(hardcode_shlibpath_var, $1)=no
_LT_TAGVAR(hardcode_direct_absolute, $1)=yes
_LT_TAGVAR(archive_cmds, $1)='$CC -shared $pic_flag $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags -o $lib'
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='${wl}-rpath,$libdir'
if test -z "`echo __ELF__ | $CC -E - | grep __ELF__`" || test "$host_os-$host_cpu" = "openbsd2.8-powerpc"; then
_LT_TAGVAR(archive_expsym_cmds, $1)='$CC -shared $pic_flag $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags ${wl}-retain-symbols-file,$export_symbols -o $lib'
_LT_TAGVAR(export_dynamic_flag_spec, $1)='${wl}-E'
_LT_TAGVAR(whole_archive_flag_spec, $1)="$wlarc"'--whole-archive$convenience '"$wlarc"'--no-whole-archive'
fi
output_verbose_link_cmd=func_echo_all
else
_LT_TAGVAR(ld_shlibs, $1)=no
fi
;;
osf3* | osf4* | osf5*)
case $cc_basename in
KCC*)
# Kuck and Associates, Inc. (KAI) C++ Compiler
# KCC will only create a shared library if the output file
# ends with ".so" (or ".sl" for HP-UX), so rename the library
# to its proper name (with version) after linking.
_LT_TAGVAR(archive_cmds, $1)='tempext=`echo $shared_ext | $SED -e '\''s/\([[^()0-9A-Za-z{}]]\)/\\\\\1/g'\''`; templib=`echo "$lib" | $SED -e "s/\${tempext}\..*/.so/"`; $CC $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags --soname $soname -o \$templib; mv \$templib $lib'
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='${wl}-rpath,$libdir'
_LT_TAGVAR(hardcode_libdir_separator, $1)=:
# Archives containing C++ object files must be created using
# the KAI C++ compiler.
case $host in
osf3*) _LT_TAGVAR(old_archive_cmds, $1)='$CC -Bstatic -o $oldlib $oldobjs' ;;
*) _LT_TAGVAR(old_archive_cmds, $1)='$CC -o $oldlib $oldobjs' ;;
esac
;;
RCC*)
# Rational C++ 2.4.1
# FIXME: insert proper C++ library support
_LT_TAGVAR(ld_shlibs, $1)=no
;;
cxx*)
case $host in
osf3*)
_LT_TAGVAR(allow_undefined_flag, $1)=' ${wl}-expect_unresolved ${wl}\*'
_LT_TAGVAR(archive_cmds, $1)='$CC -shared${allow_undefined_flag} $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags ${wl}-soname $soname `test -n "$verstring" && func_echo_all "${wl}-set_version $verstring"` -update_registry ${output_objdir}/so_locations -o $lib'
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='${wl}-rpath ${wl}$libdir'
;;
*)
_LT_TAGVAR(allow_undefined_flag, $1)=' -expect_unresolved \*'
_LT_TAGVAR(archive_cmds, $1)='$CC -shared${allow_undefined_flag} $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags -msym -soname $soname `test -n "$verstring" && func_echo_all "-set_version $verstring"` -update_registry ${output_objdir}/so_locations -o $lib'
_LT_TAGVAR(archive_expsym_cmds, $1)='for i in `cat $export_symbols`; do printf "%s %s\\n" -exported_symbol "\$i" >> $lib.exp; done~
echo "-hidden">> $lib.exp~
$CC -shared$allow_undefined_flag $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags -msym -soname $soname ${wl}-input ${wl}$lib.exp `test -n "$verstring" && $ECHO "-set_version $verstring"` -update_registry ${output_objdir}/so_locations -o $lib~
$RM $lib.exp'
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='-rpath $libdir'
;;
esac
_LT_TAGVAR(hardcode_libdir_separator, $1)=:
# Commands to make compiler produce verbose output that lists
# what "hidden" libraries, object files and flags are used when
# linking a shared library.
#
# There doesn't appear to be a way to prevent this compiler from
# explicitly linking system object files so we need to strip them
# from the output so that they don't get included in the library
# dependencies.
output_verbose_link_cmd='templist=`$CC -shared $CFLAGS -v conftest.$objext 2>&1 | $GREP "ld" | $GREP -v "ld:"`; templist=`func_echo_all "$templist" | $SED "s/\(^.*ld.*\)\( .*ld.*$\)/\1/"`; list=""; for z in $templist; do case $z in conftest.$objext) list="$list $z";; *.$objext);; *) list="$list $z";;esac; done; func_echo_all "$list"'
;;
*)
if test "$GXX" = yes && test "$with_gnu_ld" = no; then
_LT_TAGVAR(allow_undefined_flag, $1)=' ${wl}-expect_unresolved ${wl}\*'
case $host in
osf3*)
_LT_TAGVAR(archive_cmds, $1)='$CC -shared -nostdlib ${allow_undefined_flag} $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags ${wl}-soname ${wl}$soname `test -n "$verstring" && func_echo_all "${wl}-set_version ${wl}$verstring"` ${wl}-update_registry ${wl}${output_objdir}/so_locations -o $lib'
;;
*)
_LT_TAGVAR(archive_cmds, $1)='$CC -shared $pic_flag -nostdlib ${allow_undefined_flag} $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags ${wl}-msym ${wl}-soname ${wl}$soname `test -n "$verstring" && func_echo_all "${wl}-set_version ${wl}$verstring"` ${wl}-update_registry ${wl}${output_objdir}/so_locations -o $lib'
;;
esac
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='${wl}-rpath ${wl}$libdir'
_LT_TAGVAR(hardcode_libdir_separator, $1)=:
# Commands to make compiler produce verbose output that lists
# what "hidden" libraries, object files and flags are used when
# linking a shared library.
output_verbose_link_cmd='$CC -shared $CFLAGS -v conftest.$objext 2>&1 | $GREP -v "^Configured with:" | $GREP "\-L"'
else
# FIXME: insert proper C++ library support
_LT_TAGVAR(ld_shlibs, $1)=no
fi
;;
esac
;;
psos*)
# FIXME: insert proper C++ library support
_LT_TAGVAR(ld_shlibs, $1)=no
;;
sunos4*)
case $cc_basename in
CC*)
# Sun C++ 4.x
# FIXME: insert proper C++ library support
_LT_TAGVAR(ld_shlibs, $1)=no
;;
lcc*)
# Lucid
# FIXME: insert proper C++ library support
_LT_TAGVAR(ld_shlibs, $1)=no
;;
*)
# FIXME: insert proper C++ library support
_LT_TAGVAR(ld_shlibs, $1)=no
;;
esac
;;
solaris*)
case $cc_basename in
CC* | sunCC*)
# Sun C++ 4.2, 5.x and Centerline C++
_LT_TAGVAR(archive_cmds_need_lc,$1)=yes
_LT_TAGVAR(no_undefined_flag, $1)=' -zdefs'
_LT_TAGVAR(archive_cmds, $1)='$CC -G${allow_undefined_flag} -h$soname -o $lib $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags'
_LT_TAGVAR(archive_expsym_cmds, $1)='echo "{ global:" > $lib.exp~cat $export_symbols | $SED -e "s/\(.*\)/\1;/" >> $lib.exp~echo "local: *; };" >> $lib.exp~
$CC -G${allow_undefined_flag} ${wl}-M ${wl}$lib.exp -h$soname -o $lib $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags~$RM $lib.exp'
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='-R$libdir'
_LT_TAGVAR(hardcode_shlibpath_var, $1)=no
case $host_os in
solaris2.[[0-5]] | solaris2.[[0-5]].*) ;;
*)
# The compiler driver will combine and reorder linker options,
# but understands `-z linker_flag'.
# Supported since Solaris 2.6 (maybe 2.5.1?)
_LT_TAGVAR(whole_archive_flag_spec, $1)='-z allextract$convenience -z defaultextract'
;;
esac
_LT_TAGVAR(link_all_deplibs, $1)=yes
output_verbose_link_cmd='func_echo_all'
# Archives containing C++ object files must be created using
# "CC -xar", where "CC" is the Sun C++ compiler. This is
# necessary to make sure instantiated templates are included
# in the archive.
_LT_TAGVAR(old_archive_cmds, $1)='$CC -xar -o $oldlib $oldobjs'
;;
gcx*)
# Green Hills C++ Compiler
_LT_TAGVAR(archive_cmds, $1)='$CC -shared $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags ${wl}-h $wl$soname -o $lib'
# The C++ compiler must be used to create the archive.
_LT_TAGVAR(old_archive_cmds, $1)='$CC $LDFLAGS -archive -o $oldlib $oldobjs'
;;
*)
# GNU C++ compiler with Solaris linker
if test "$GXX" = yes && test "$with_gnu_ld" = no; then
_LT_TAGVAR(no_undefined_flag, $1)=' ${wl}-z ${wl}defs'
if $CC --version | $GREP -v '^2\.7' > /dev/null; then
_LT_TAGVAR(archive_cmds, $1)='$CC -shared $pic_flag -nostdlib $LDFLAGS $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags ${wl}-h $wl$soname -o $lib'
_LT_TAGVAR(archive_expsym_cmds, $1)='echo "{ global:" > $lib.exp~cat $export_symbols | $SED -e "s/\(.*\)/\1;/" >> $lib.exp~echo "local: *; };" >> $lib.exp~
$CC -shared $pic_flag -nostdlib ${wl}-M $wl$lib.exp -o $lib $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags~$RM $lib.exp'
# Commands to make compiler produce verbose output that lists
# what "hidden" libraries, object files and flags are used when
# linking a shared library.
output_verbose_link_cmd='$CC -shared $CFLAGS -v conftest.$objext 2>&1 | $GREP -v "^Configured with:" | $GREP "\-L"'
else
# g++ 2.7 appears to require `-G' NOT `-shared' on this
# platform.
_LT_TAGVAR(archive_cmds, $1)='$CC -G -nostdlib $LDFLAGS $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags ${wl}-h $wl$soname -o $lib'
_LT_TAGVAR(archive_expsym_cmds, $1)='echo "{ global:" > $lib.exp~cat $export_symbols | $SED -e "s/\(.*\)/\1;/" >> $lib.exp~echo "local: *; };" >> $lib.exp~
$CC -G -nostdlib ${wl}-M $wl$lib.exp -o $lib $predep_objects $libobjs $deplibs $postdep_objects $compiler_flags~$RM $lib.exp'
# Commands to make compiler produce verbose output that lists
# what "hidden" libraries, object files and flags are used when
# linking a shared library.
output_verbose_link_cmd='$CC -G $CFLAGS -v conftest.$objext 2>&1 | $GREP -v "^Configured with:" | $GREP "\-L"'
fi
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='${wl}-R $wl$libdir'
case $host_os in
solaris2.[[0-5]] | solaris2.[[0-5]].*) ;;
*)
_LT_TAGVAR(whole_archive_flag_spec, $1)='${wl}-z ${wl}allextract$convenience ${wl}-z ${wl}defaultextract'
;;
esac
fi
;;
esac
;;
sysv4*uw2* | sysv5OpenUNIX* | sysv5UnixWare7.[[01]].[[10]]* | unixware7* | sco3.2v5.0.[[024]]*)
_LT_TAGVAR(no_undefined_flag, $1)='${wl}-z,text'
_LT_TAGVAR(archive_cmds_need_lc, $1)=no
_LT_TAGVAR(hardcode_shlibpath_var, $1)=no
runpath_var='LD_RUN_PATH'
case $cc_basename in
CC*)
_LT_TAGVAR(archive_cmds, $1)='$CC -G ${wl}-h,$soname -o $lib $libobjs $deplibs $compiler_flags'
_LT_TAGVAR(archive_expsym_cmds, $1)='$CC -G ${wl}-Bexport:$export_symbols ${wl}-h,$soname -o $lib $libobjs $deplibs $compiler_flags'
;;
*)
_LT_TAGVAR(archive_cmds, $1)='$CC -shared ${wl}-h,$soname -o $lib $libobjs $deplibs $compiler_flags'
_LT_TAGVAR(archive_expsym_cmds, $1)='$CC -shared ${wl}-Bexport:$export_symbols ${wl}-h,$soname -o $lib $libobjs $deplibs $compiler_flags'
;;
esac
;;
sysv5* | sco3.2v5* | sco5v6*)
# Note: We can NOT use -z defs as we might desire, because we do not
# link with -lc, and that would cause any symbols used from libc to
# always be unresolved, which means just about no library would
# ever link correctly. If we're not using GNU ld we use -z text
# though, which does catch some bad symbols but isn't as heavy-handed
# as -z defs.
_LT_TAGVAR(no_undefined_flag, $1)='${wl}-z,text'
_LT_TAGVAR(allow_undefined_flag, $1)='${wl}-z,nodefs'
_LT_TAGVAR(archive_cmds_need_lc, $1)=no
_LT_TAGVAR(hardcode_shlibpath_var, $1)=no
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)='${wl}-R,$libdir'
_LT_TAGVAR(hardcode_libdir_separator, $1)=':'
_LT_TAGVAR(link_all_deplibs, $1)=yes
_LT_TAGVAR(export_dynamic_flag_spec, $1)='${wl}-Bexport'
runpath_var='LD_RUN_PATH'
case $cc_basename in
CC*)
_LT_TAGVAR(archive_cmds, $1)='$CC -G ${wl}-h,$soname -o $lib $libobjs $deplibs $compiler_flags'
_LT_TAGVAR(archive_expsym_cmds, $1)='$CC -G ${wl}-Bexport:$export_symbols ${wl}-h,$soname -o $lib $libobjs $deplibs $compiler_flags'
_LT_TAGVAR(old_archive_cmds, $1)='$CC -Tprelink_objects $oldobjs~
'"$_LT_TAGVAR(old_archive_cmds, $1)"
_LT_TAGVAR(reload_cmds, $1)='$CC -Tprelink_objects $reload_objs~
'"$_LT_TAGVAR(reload_cmds, $1)"
;;
*)
_LT_TAGVAR(archive_cmds, $1)='$CC -shared ${wl}-h,$soname -o $lib $libobjs $deplibs $compiler_flags'
_LT_TAGVAR(archive_expsym_cmds, $1)='$CC -shared ${wl}-Bexport:$export_symbols ${wl}-h,$soname -o $lib $libobjs $deplibs $compiler_flags'
;;
esac
;;
tandem*)
case $cc_basename in
NCC*)
# NonStop-UX NCC 3.20
# FIXME: insert proper C++ library support
_LT_TAGVAR(ld_shlibs, $1)=no
;;
*)
# FIXME: insert proper C++ library support
_LT_TAGVAR(ld_shlibs, $1)=no
;;
esac
;;
vxworks*)
# FIXME: insert proper C++ library support
_LT_TAGVAR(ld_shlibs, $1)=no
;;
*)
# FIXME: insert proper C++ library support
_LT_TAGVAR(ld_shlibs, $1)=no
;;
esac
AC_MSG_RESULT([$_LT_TAGVAR(ld_shlibs, $1)])
test "$_LT_TAGVAR(ld_shlibs, $1)" = no && can_build_shared=no
_LT_TAGVAR(GCC, $1)="$GXX"
_LT_TAGVAR(LD, $1)="$LD"
## CAVEAT EMPTOR:
## There is no encapsulation within the following macros, do not change
## the running order or otherwise move them around unless you know exactly
## what you are doing...
_LT_SYS_HIDDEN_LIBDEPS($1)
_LT_COMPILER_PIC($1)
_LT_COMPILER_C_O($1)
_LT_COMPILER_FILE_LOCKS($1)
_LT_LINKER_SHLIBS($1)
_LT_SYS_DYNAMIC_LINKER($1)
_LT_LINKER_HARDCODE_LIBPATH($1)
_LT_CONFIG($1)
fi # test -n "$compiler"
CC=$lt_save_CC
CFLAGS=$lt_save_CFLAGS
LDCXX=$LD
LD=$lt_save_LD
GCC=$lt_save_GCC
with_gnu_ld=$lt_save_with_gnu_ld
lt_cv_path_LDCXX=$lt_cv_path_LD
lt_cv_path_LD=$lt_save_path_LD
lt_cv_prog_gnu_ldcxx=$lt_cv_prog_gnu_ld
lt_cv_prog_gnu_ld=$lt_save_with_gnu_ld
fi # test "$_lt_caught_CXX_error" != yes
AC_LANG_POP
])# _LT_LANG_CXX_CONFIG
# _LT_FUNC_STRIPNAME_CNF
# ----------------------
# func_stripname_cnf prefix suffix name
# strip PREFIX and SUFFIX off of NAME.
# PREFIX and SUFFIX must not contain globbing or regex special
# characters, hashes, percent signs, but SUFFIX may contain a leading
# dot (in which case that matches only a dot).
#
# This function is identical to the (non-XSI) version of func_stripname,
# except this one can be used by m4 code that may be executed by configure,
# rather than the libtool script.
m4_defun([_LT_FUNC_STRIPNAME_CNF],[dnl
AC_REQUIRE([_LT_DECL_SED])
AC_REQUIRE([_LT_PROG_ECHO_BACKSLASH])
func_stripname_cnf ()
{
case ${2} in
.*) func_stripname_result=`$ECHO "${3}" | $SED "s%^${1}%%; s%\\\\${2}\$%%"`;;
*) func_stripname_result=`$ECHO "${3}" | $SED "s%^${1}%%; s%${2}\$%%"`;;
esac
} # func_stripname_cnf
])# _LT_FUNC_STRIPNAME_CNF
# _LT_SYS_HIDDEN_LIBDEPS([TAGNAME])
# ---------------------------------
# Figure out "hidden" library dependencies from verbose
# compiler output when linking a shared library.
# Parse the compiler output and extract the necessary
# objects, libraries and library flags.
m4_defun([_LT_SYS_HIDDEN_LIBDEPS],
[m4_require([_LT_FILEUTILS_DEFAULTS])dnl
AC_REQUIRE([_LT_FUNC_STRIPNAME_CNF])dnl
# Dependencies to place before and after the object being linked:
_LT_TAGVAR(predep_objects, $1)=
_LT_TAGVAR(postdep_objects, $1)=
_LT_TAGVAR(predeps, $1)=
_LT_TAGVAR(postdeps, $1)=
_LT_TAGVAR(compiler_lib_search_path, $1)=
dnl we can't use the lt_simple_compile_test_code here,
dnl because it contains code intended for an executable,
dnl not a library. It's possible we should let each
dnl tag define a new lt_????_link_test_code variable,
dnl but it's only used here...
m4_if([$1], [], [cat > conftest.$ac_ext <<_LT_EOF
int a;
void foo (void) { a = 0; }
_LT_EOF
], [$1], [CXX], [cat > conftest.$ac_ext <<_LT_EOF
class Foo
{
public:
Foo (void) { a = 0; }
private:
int a;
};
_LT_EOF
], [$1], [F77], [cat > conftest.$ac_ext <<_LT_EOF
subroutine foo
implicit none
integer*4 a
a=0
return
end
_LT_EOF
], [$1], [FC], [cat > conftest.$ac_ext <<_LT_EOF
subroutine foo
implicit none
integer a
a=0
return
end
_LT_EOF
], [$1], [GCJ], [cat > conftest.$ac_ext <<_LT_EOF
public class foo {
private int a;
public void bar (void) {
a = 0;
}
};
_LT_EOF
], [$1], [GO], [cat > conftest.$ac_ext <<_LT_EOF
package foo
func foo() {
}
_LT_EOF
])
_lt_libdeps_save_CFLAGS=$CFLAGS
case "$CC $CFLAGS " in #(
*\ -flto*\ *) CFLAGS="$CFLAGS -fno-lto" ;;
*\ -fwhopr*\ *) CFLAGS="$CFLAGS -fno-whopr" ;;
*\ -fuse-linker-plugin*\ *) CFLAGS="$CFLAGS -fno-use-linker-plugin" ;;
esac
dnl Parse the compiler output and extract the necessary
dnl objects, libraries and library flags.
if AC_TRY_EVAL(ac_compile); then
# Parse the compiler output and extract the necessary
# objects, libraries and library flags.
# Sentinel used to keep track of whether or not we are before
# the conftest object file.
pre_test_object_deps_done=no
for p in `eval "$output_verbose_link_cmd"`; do
case ${prev}${p} in
-L* | -R* | -l*)
# Some compilers place space between "-{L,R}" and the path.
# Remove the space.
if test $p = "-L" ||
test $p = "-R"; then
prev=$p
continue
fi
# Expand the sysroot to ease extracting the directories later.
if test -z "$prev"; then
case $p in
-L*) func_stripname_cnf '-L' '' "$p"; prev=-L; p=$func_stripname_result ;;
-R*) func_stripname_cnf '-R' '' "$p"; prev=-R; p=$func_stripname_result ;;
-l*) func_stripname_cnf '-l' '' "$p"; prev=-l; p=$func_stripname_result ;;
esac
fi
case $p in
=*) func_stripname_cnf '=' '' "$p"; p=$lt_sysroot$func_stripname_result ;;
esac
if test "$pre_test_object_deps_done" = no; then
case ${prev} in
-L | -R)
# Internal compiler library paths should come after those
# provided the user. The postdeps already come after the
# user supplied libs so there is no need to process them.
if test -z "$_LT_TAGVAR(compiler_lib_search_path, $1)"; then
_LT_TAGVAR(compiler_lib_search_path, $1)="${prev}${p}"
else
_LT_TAGVAR(compiler_lib_search_path, $1)="${_LT_TAGVAR(compiler_lib_search_path, $1)} ${prev}${p}"
fi
;;
# The "-l" case would never come before the object being
# linked, so don't bother handling this case.
esac
else
if test -z "$_LT_TAGVAR(postdeps, $1)"; then
_LT_TAGVAR(postdeps, $1)="${prev}${p}"
else
_LT_TAGVAR(postdeps, $1)="${_LT_TAGVAR(postdeps, $1)} ${prev}${p}"
fi
fi
prev=
;;
*.lto.$objext) ;; # Ignore GCC LTO objects
*.$objext)
# This assumes that the test object file only shows up
# once in the compiler output.
if test "$p" = "conftest.$objext"; then
pre_test_object_deps_done=yes
continue
fi
if test "$pre_test_object_deps_done" = no; then
if test -z "$_LT_TAGVAR(predep_objects, $1)"; then
_LT_TAGVAR(predep_objects, $1)="$p"
else
_LT_TAGVAR(predep_objects, $1)="$_LT_TAGVAR(predep_objects, $1) $p"
fi
else
if test -z "$_LT_TAGVAR(postdep_objects, $1)"; then
_LT_TAGVAR(postdep_objects, $1)="$p"
else
_LT_TAGVAR(postdep_objects, $1)="$_LT_TAGVAR(postdep_objects, $1) $p"
fi
fi
;;
*) ;; # Ignore the rest.
esac
done
# Clean up.
rm -f a.out a.exe
else
echo "libtool.m4: error: problem compiling $1 test program"
fi
$RM -f confest.$objext
CFLAGS=$_lt_libdeps_save_CFLAGS
# PORTME: override above test on systems where it is broken
m4_if([$1], [CXX],
[case $host_os in
interix[[3-9]]*)
# Interix 3.5 installs completely hosed .la files for C++, so rather than
# hack all around it, let's just trust "g++" to DTRT.
_LT_TAGVAR(predep_objects,$1)=
_LT_TAGVAR(postdep_objects,$1)=
_LT_TAGVAR(postdeps,$1)=
;;
linux*)
case `$CC -V 2>&1 | sed 5q` in
*Sun\ C*)
# Sun C++ 5.9
# The more standards-conforming stlport4 library is
# incompatible with the Cstd library. Avoid specifying
# it if it's in CXXFLAGS. Ignore libCrun as
# -library=stlport4 depends on it.
case " $CXX $CXXFLAGS " in
*" -library=stlport4 "*)
solaris_use_stlport4=yes
;;
esac
if test "$solaris_use_stlport4" != yes; then
_LT_TAGVAR(postdeps,$1)='-library=Cstd -library=Crun'
fi
;;
esac
;;
solaris*)
case $cc_basename in
CC* | sunCC*)
# The more standards-conforming stlport4 library is
# incompatible with the Cstd library. Avoid specifying
# it if it's in CXXFLAGS. Ignore libCrun as
# -library=stlport4 depends on it.
case " $CXX $CXXFLAGS " in
*" -library=stlport4 "*)
solaris_use_stlport4=yes
;;
esac
# Adding this requires a known-good setup of shared libraries for
# Sun compiler versions before 5.6, else PIC objects from an old
# archive will be linked into the output, leading to subtle bugs.
if test "$solaris_use_stlport4" != yes; then
_LT_TAGVAR(postdeps,$1)='-library=Cstd -library=Crun'
fi
;;
esac
;;
esac
])
case " $_LT_TAGVAR(postdeps, $1) " in
*" -lc "*) _LT_TAGVAR(archive_cmds_need_lc, $1)=no ;;
esac
_LT_TAGVAR(compiler_lib_search_dirs, $1)=
if test -n "${_LT_TAGVAR(compiler_lib_search_path, $1)}"; then
_LT_TAGVAR(compiler_lib_search_dirs, $1)=`echo " ${_LT_TAGVAR(compiler_lib_search_path, $1)}" | ${SED} -e 's! -L! !g' -e 's!^ !!'`
fi
_LT_TAGDECL([], [compiler_lib_search_dirs], [1],
[The directories searched by this compiler when creating a shared library])
_LT_TAGDECL([], [predep_objects], [1],
[Dependencies to place before and after the objects being linked to
create a shared library])
_LT_TAGDECL([], [postdep_objects], [1])
_LT_TAGDECL([], [predeps], [1])
_LT_TAGDECL([], [postdeps], [1])
_LT_TAGDECL([], [compiler_lib_search_path], [1],
[The library search path used internally by the compiler when linking
a shared library])
])# _LT_SYS_HIDDEN_LIBDEPS
# _LT_LANG_F77_CONFIG([TAG])
# --------------------------
# Ensure that the configuration variables for a Fortran 77 compiler are
# suitably defined. These variables are subsequently used by _LT_CONFIG
# to write the compiler configuration to `libtool'.
m4_defun([_LT_LANG_F77_CONFIG],
[AC_LANG_PUSH(Fortran 77)
if test -z "$F77" || test "X$F77" = "Xno"; then
_lt_disable_F77=yes
fi
_LT_TAGVAR(archive_cmds_need_lc, $1)=no
_LT_TAGVAR(allow_undefined_flag, $1)=
_LT_TAGVAR(always_export_symbols, $1)=no
_LT_TAGVAR(archive_expsym_cmds, $1)=
_LT_TAGVAR(export_dynamic_flag_spec, $1)=
_LT_TAGVAR(hardcode_direct, $1)=no
_LT_TAGVAR(hardcode_direct_absolute, $1)=no
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)=
_LT_TAGVAR(hardcode_libdir_separator, $1)=
_LT_TAGVAR(hardcode_minus_L, $1)=no
_LT_TAGVAR(hardcode_automatic, $1)=no
_LT_TAGVAR(inherit_rpath, $1)=no
_LT_TAGVAR(module_cmds, $1)=
_LT_TAGVAR(module_expsym_cmds, $1)=
_LT_TAGVAR(link_all_deplibs, $1)=unknown
_LT_TAGVAR(old_archive_cmds, $1)=$old_archive_cmds
_LT_TAGVAR(reload_flag, $1)=$reload_flag
_LT_TAGVAR(reload_cmds, $1)=$reload_cmds
_LT_TAGVAR(no_undefined_flag, $1)=
_LT_TAGVAR(whole_archive_flag_spec, $1)=
_LT_TAGVAR(enable_shared_with_static_runtimes, $1)=no
# Source file extension for f77 test sources.
ac_ext=f
# Object file extension for compiled f77 test sources.
objext=o
_LT_TAGVAR(objext, $1)=$objext
# No sense in running all these tests if we already determined that
# the F77 compiler isn't working. Some variables (like enable_shared)
# are currently assumed to apply to all compilers on this platform,
# and will be corrupted by setting them based on a non-working compiler.
if test "$_lt_disable_F77" != yes; then
# Code to be used in simple compile tests
lt_simple_compile_test_code="\
subroutine t
return
end
"
# Code to be used in simple link tests
lt_simple_link_test_code="\
program t
end
"
# ltmain only uses $CC for tagged configurations so make sure $CC is set.
_LT_TAG_COMPILER
# save warnings/boilerplate of simple test code
_LT_COMPILER_BOILERPLATE
_LT_LINKER_BOILERPLATE
# Allow CC to be a program name with arguments.
lt_save_CC="$CC"
lt_save_GCC=$GCC
lt_save_CFLAGS=$CFLAGS
CC=${F77-"f77"}
CFLAGS=$FFLAGS
compiler=$CC
_LT_TAGVAR(compiler, $1)=$CC
_LT_CC_BASENAME([$compiler])
GCC=$G77
if test -n "$compiler"; then
AC_MSG_CHECKING([if libtool supports shared libraries])
AC_MSG_RESULT([$can_build_shared])
AC_MSG_CHECKING([whether to build shared libraries])
test "$can_build_shared" = "no" && enable_shared=no
# On AIX, shared libraries and static libraries use the same namespace, and
# are all built from PIC.
case $host_os in
aix3*)
test "$enable_shared" = yes && enable_static=no
if test -n "$RANLIB"; then
archive_cmds="$archive_cmds~\$RANLIB \$lib"
postinstall_cmds='$RANLIB $lib'
fi
;;
aix[[4-9]]*)
if test "$host_cpu" != ia64 && test "$aix_use_runtimelinking" = no ; then
test "$enable_shared" = yes && enable_static=no
fi
;;
esac
AC_MSG_RESULT([$enable_shared])
AC_MSG_CHECKING([whether to build static libraries])
# Make sure either enable_shared or enable_static is yes.
test "$enable_shared" = yes || enable_static=yes
AC_MSG_RESULT([$enable_static])
_LT_TAGVAR(GCC, $1)="$G77"
_LT_TAGVAR(LD, $1)="$LD"
## CAVEAT EMPTOR:
## There is no encapsulation within the following macros, do not change
## the running order or otherwise move them around unless you know exactly
## what you are doing...
_LT_COMPILER_PIC($1)
_LT_COMPILER_C_O($1)
_LT_COMPILER_FILE_LOCKS($1)
_LT_LINKER_SHLIBS($1)
_LT_SYS_DYNAMIC_LINKER($1)
_LT_LINKER_HARDCODE_LIBPATH($1)
_LT_CONFIG($1)
fi # test -n "$compiler"
GCC=$lt_save_GCC
CC="$lt_save_CC"
CFLAGS="$lt_save_CFLAGS"
fi # test "$_lt_disable_F77" != yes
AC_LANG_POP
])# _LT_LANG_F77_CONFIG
# _LT_LANG_FC_CONFIG([TAG])
# -------------------------
# Ensure that the configuration variables for a Fortran compiler are
# suitably defined. These variables are subsequently used by _LT_CONFIG
# to write the compiler configuration to `libtool'.
m4_defun([_LT_LANG_FC_CONFIG],
[AC_LANG_PUSH(Fortran)
if test -z "$FC" || test "X$FC" = "Xno"; then
_lt_disable_FC=yes
fi
_LT_TAGVAR(archive_cmds_need_lc, $1)=no
_LT_TAGVAR(allow_undefined_flag, $1)=
_LT_TAGVAR(always_export_symbols, $1)=no
_LT_TAGVAR(archive_expsym_cmds, $1)=
_LT_TAGVAR(export_dynamic_flag_spec, $1)=
_LT_TAGVAR(hardcode_direct, $1)=no
_LT_TAGVAR(hardcode_direct_absolute, $1)=no
_LT_TAGVAR(hardcode_libdir_flag_spec, $1)=
_LT_TAGVAR(hardcode_libdir_separator, $1)=
_LT_TAGVAR(hardcode_minus_L, $1)=no
_LT_TAGVAR(hardcode_automatic, $1)=no
_LT_TAGVAR(inherit_rpath, $1)=no
_LT_TAGVAR(module_cmds, $1)=
_LT_TAGVAR(module_expsym_cmds, $1)=
_LT_TAGVAR(link_all_deplibs, $1)=unknown
_LT_TAGVAR(old_archive_cmds, $1)=$old_archive_cmds
_LT_TAGVAR(reload_flag, $1)=$reload_flag
_LT_TAGVAR(reload_cmds, $1)=$reload_cmds
_LT_TAGVAR(no_undefined_flag, $1)=
_LT_TAGVAR(whole_archive_flag_spec, $1)=
_LT_TAGVAR(enable_shared_with_static_runtimes, $1)=no
# Source file extension for fc test sources.
ac_ext=${ac_fc_srcext-f}
# Object file extension for compiled fc test sources.
objext=o
_LT_TAGVAR(objext, $1)=$objext
# No sense in running all these tests if we already determined that
# the FC compiler isn't working. Some variables (like enable_shared)
# are currently assumed to apply to all compilers on this platform,
# and will be corrupted by setting them based on a non-working compiler.
if test "$_lt_disable_FC" != yes; then
# Code to be used in simple compile tests
lt_simple_compile_test_code="\
subroutine t
return
end
"
# Code to be used in simple link tests
lt_simple_link_test_code="\
program t
end
"
# ltmain only uses $CC for tagged configurations so make sure $CC is set.
_LT_TAG_COMPILER
# save warnings/boilerplate of simple test code
_LT_COMPILER_BOILERPLATE
_LT_LINKER_BOILERPLATE
# Allow CC to be a program name with arguments.
lt_save_CC="$CC"
lt_save_GCC=$GCC
lt_save_CFLAGS=$CFLAGS
CC=${FC-"f95"}
CFLAGS=$FCFLAGS
compiler=$CC
GCC=$ac_cv_fc_compiler_gnu
_LT_TAGVAR(compiler, $1)=$CC
_LT_CC_BASENAME([$compiler])
if test -n "$compiler"; then
AC_MSG_CHECKING([if libtool supports shared libraries])
AC_MSG_RESULT([$can_build_shared])
AC_MSG_CHECKING([whether to build shared libraries])
test "$can_build_shared" = "no" && enable_shared=no
# On AIX, shared libraries and static libraries use the same namespace, and
# are all built from PIC.
case $host_os in
aix3*)
test "$enable_shared" = yes && enable_static=no
if test -n "$RANLIB"; then
archive_cmds="$archive_cmds~\$RANLIB \$lib"
postinstall_cmds='$RANLIB $lib'
fi
;;
aix[[4-9]]*)
if test "$host_cpu" != ia64 && test "$aix_use_runtimelinking" = no ; then
test "$enable_shared" = yes && enable_static=no
fi
;;
esac
AC_MSG_RESULT([$enable_shared])
AC_MSG_CHECKING([whether to build static libraries])
# Make sure either enable_shared or enable_static is yes.
test "$enable_shared" = yes || enable_static=yes
AC_MSG_RESULT([$enable_static])
_LT_TAGVAR(GCC, $1)="$ac_cv_fc_compiler_gnu"
_LT_TAGVAR(LD, $1)="$LD"
## CAVEAT EMPTOR:
## There is no encapsulation within the following macros, do not change
## the running order or otherwise move them around unless you know exactly
## what you are doing...
_LT_SYS_HIDDEN_LIBDEPS($1)
_LT_COMPILER_PIC($1)
_LT_COMPILER_C_O($1)
_LT_COMPILER_FILE_LOCKS($1)
_LT_LINKER_SHLIBS($1)
_LT_SYS_DYNAMIC_LINKER($1)
_LT_LINKER_HARDCODE_LIBPATH($1)
_LT_CONFIG($1)
fi # test -n "$compiler"
GCC=$lt_save_GCC
CC=$lt_save_CC
CFLAGS=$lt_save_CFLAGS
fi # test "$_lt_disable_FC" != yes
AC_LANG_POP
])# _LT_LANG_FC_CONFIG
# _LT_LANG_GCJ_CONFIG([TAG])
# --------------------------
# Ensure that the configuration variables for the GNU Java Compiler compiler
# are suitably defined. These variables are subsequently used by _LT_CONFIG
# to write the compiler configuration to `libtool'.
m4_defun([_LT_LANG_GCJ_CONFIG],
[AC_REQUIRE([LT_PROG_GCJ])dnl
AC_LANG_SAVE
# Source file extension for Java test sources.
ac_ext=java
# Object file extension for compiled Java test sources.
objext=o
_LT_TAGVAR(objext, $1)=$objext
# Code to be used in simple compile tests
lt_simple_compile_test_code="class foo {}"
# Code to be used in simple link tests
lt_simple_link_test_code='public class conftest { public static void main(String[[]] argv) {}; }'
# ltmain only uses $CC for tagged configurations so make sure $CC is set.
_LT_TAG_COMPILER
# save warnings/boilerplate of simple test code
_LT_COMPILER_BOILERPLATE
_LT_LINKER_BOILERPLATE
# Allow CC to be a program name with arguments.
lt_save_CC=$CC
lt_save_CFLAGS=$CFLAGS
lt_save_GCC=$GCC
GCC=yes
CC=${GCJ-"gcj"}
CFLAGS=$GCJFLAGS
compiler=$CC
_LT_TAGVAR(compiler, $1)=$CC
_LT_TAGVAR(LD, $1)="$LD"
_LT_CC_BASENAME([$compiler])
# GCJ did not exist at the time GCC didn't implicitly link libc in.
_LT_TAGVAR(archive_cmds_need_lc, $1)=no
_LT_TAGVAR(old_archive_cmds, $1)=$old_archive_cmds
_LT_TAGVAR(reload_flag, $1)=$reload_flag
_LT_TAGVAR(reload_cmds, $1)=$reload_cmds
## CAVEAT EMPTOR:
## There is no encapsulation within the following macros, do not change
## the running order or otherwise move them around unless you know exactly
## what you are doing...
if test -n "$compiler"; then
_LT_COMPILER_NO_RTTI($1)
_LT_COMPILER_PIC($1)
_LT_COMPILER_C_O($1)
_LT_COMPILER_FILE_LOCKS($1)
_LT_LINKER_SHLIBS($1)
_LT_LINKER_HARDCODE_LIBPATH($1)
_LT_CONFIG($1)
fi
AC_LANG_RESTORE
GCC=$lt_save_GCC
CC=$lt_save_CC
CFLAGS=$lt_save_CFLAGS
])# _LT_LANG_GCJ_CONFIG
# _LT_LANG_GO_CONFIG([TAG])
# --------------------------
# Ensure that the configuration variables for the GNU Go compiler
# are suitably defined. These variables are subsequently used by _LT_CONFIG
# to write the compiler configuration to `libtool'.
m4_defun([_LT_LANG_GO_CONFIG],
[AC_REQUIRE([LT_PROG_GO])dnl
AC_LANG_SAVE
# Source file extension for Go test sources.
ac_ext=go
# Object file extension for compiled Go test sources.
objext=o
_LT_TAGVAR(objext, $1)=$objext
# Code to be used in simple compile tests
lt_simple_compile_test_code="package main; func main() { }"
# Code to be used in simple link tests
lt_simple_link_test_code='package main; func main() { }'
# ltmain only uses $CC for tagged configurations so make sure $CC is set.
_LT_TAG_COMPILER
# save warnings/boilerplate of simple test code
_LT_COMPILER_BOILERPLATE
_LT_LINKER_BOILERPLATE
# Allow CC to be a program name with arguments.
lt_save_CC=$CC
lt_save_CFLAGS=$CFLAGS
lt_save_GCC=$GCC
GCC=yes
CC=${GOC-"gccgo"}
CFLAGS=$GOFLAGS
compiler=$CC
_LT_TAGVAR(compiler, $1)=$CC
_LT_TAGVAR(LD, $1)="$LD"
_LT_CC_BASENAME([$compiler])
# Go did not exist at the time GCC didn't implicitly link libc in.
_LT_TAGVAR(archive_cmds_need_lc, $1)=no
_LT_TAGVAR(old_archive_cmds, $1)=$old_archive_cmds
_LT_TAGVAR(reload_flag, $1)=$reload_flag
_LT_TAGVAR(reload_cmds, $1)=$reload_cmds
## CAVEAT EMPTOR:
## There is no encapsulation within the following macros, do not change
## the running order or otherwise move them around unless you know exactly
## what you are doing...
if test -n "$compiler"; then
_LT_COMPILER_NO_RTTI($1)
_LT_COMPILER_PIC($1)
_LT_COMPILER_C_O($1)
_LT_COMPILER_FILE_LOCKS($1)
_LT_LINKER_SHLIBS($1)
_LT_LINKER_HARDCODE_LIBPATH($1)
_LT_CONFIG($1)
fi
AC_LANG_RESTORE
GCC=$lt_save_GCC
CC=$lt_save_CC
CFLAGS=$lt_save_CFLAGS
])# _LT_LANG_GO_CONFIG
# _LT_LANG_RC_CONFIG([TAG])
# -------------------------
# Ensure that the configuration variables for the Windows resource compiler
# are suitably defined. These variables are subsequently used by _LT_CONFIG
# to write the compiler configuration to `libtool'.
m4_defun([_LT_LANG_RC_CONFIG],
[AC_REQUIRE([LT_PROG_RC])dnl
AC_LANG_SAVE
# Source file extension for RC test sources.
ac_ext=rc
# Object file extension for compiled RC test sources.
objext=o
_LT_TAGVAR(objext, $1)=$objext
# Code to be used in simple compile tests
lt_simple_compile_test_code='sample MENU { MENUITEM "&Soup", 100, CHECKED }'
# Code to be used in simple link tests
lt_simple_link_test_code="$lt_simple_compile_test_code"
# ltmain only uses $CC for tagged configurations so make sure $CC is set.
_LT_TAG_COMPILER
# save warnings/boilerplate of simple test code
_LT_COMPILER_BOILERPLATE
_LT_LINKER_BOILERPLATE
# Allow CC to be a program name with arguments.
lt_save_CC="$CC"
lt_save_CFLAGS=$CFLAGS
lt_save_GCC=$GCC
GCC=
CC=${RC-"windres"}
CFLAGS=
compiler=$CC
_LT_TAGVAR(compiler, $1)=$CC
_LT_CC_BASENAME([$compiler])
_LT_TAGVAR(lt_cv_prog_compiler_c_o, $1)=yes
if test -n "$compiler"; then
:
_LT_CONFIG($1)
fi
GCC=$lt_save_GCC
AC_LANG_RESTORE
CC=$lt_save_CC
CFLAGS=$lt_save_CFLAGS
])# _LT_LANG_RC_CONFIG
# LT_PROG_GCJ
# -----------
AC_DEFUN([LT_PROG_GCJ],
[m4_ifdef([AC_PROG_GCJ], [AC_PROG_GCJ],
[m4_ifdef([A][M_PROG_GCJ], [A][M_PROG_GCJ],
[AC_CHECK_TOOL(GCJ, gcj,)
test "x${GCJFLAGS+set}" = xset || GCJFLAGS="-g -O2"
AC_SUBST(GCJFLAGS)])])[]dnl
])
# Old name:
AU_ALIAS([LT_AC_PROG_GCJ], [LT_PROG_GCJ])
dnl aclocal-1.4 backwards compatibility:
dnl AC_DEFUN([LT_AC_PROG_GCJ], [])
# LT_PROG_GO
# ----------
AC_DEFUN([LT_PROG_GO],
[AC_CHECK_TOOL(GOC, gccgo,)
])
# LT_PROG_RC
# ----------
AC_DEFUN([LT_PROG_RC],
[AC_CHECK_TOOL(RC, windres,)
])
# Old name:
AU_ALIAS([LT_AC_PROG_RC], [LT_PROG_RC])
dnl aclocal-1.4 backwards compatibility:
dnl AC_DEFUN([LT_AC_PROG_RC], [])
# _LT_DECL_EGREP
# --------------
# If we don't have a new enough Autoconf to choose the best grep
# available, choose the one first in the user's PATH.
m4_defun([_LT_DECL_EGREP],
[AC_REQUIRE([AC_PROG_EGREP])dnl
AC_REQUIRE([AC_PROG_FGREP])dnl
test -z "$GREP" && GREP=grep
_LT_DECL([], [GREP], [1], [A grep program that handles long lines])
_LT_DECL([], [EGREP], [1], [An ERE matcher])
_LT_DECL([], [FGREP], [1], [A literal string matcher])
dnl Non-bleeding-edge autoconf doesn't subst GREP, so do it here too
AC_SUBST([GREP])
])
# _LT_DECL_OBJDUMP
# --------------
# If we don't have a new enough Autoconf to choose the best objdump
# available, choose the one first in the user's PATH.
m4_defun([_LT_DECL_OBJDUMP],
[AC_CHECK_TOOL(OBJDUMP, objdump, false)
test -z "$OBJDUMP" && OBJDUMP=objdump
_LT_DECL([], [OBJDUMP], [1], [An object symbol dumper])
AC_SUBST([OBJDUMP])
])
# _LT_DECL_DLLTOOL
# ----------------
# Ensure DLLTOOL variable is set.
m4_defun([_LT_DECL_DLLTOOL],
[AC_CHECK_TOOL(DLLTOOL, dlltool, false)
test -z "$DLLTOOL" && DLLTOOL=dlltool
_LT_DECL([], [DLLTOOL], [1], [DLL creation program])
AC_SUBST([DLLTOOL])
])
# _LT_DECL_SED
# ------------
# Check for a fully-functional sed program, that truncates
# as few characters as possible. Prefer GNU sed if found.
m4_defun([_LT_DECL_SED],
[AC_PROG_SED
test -z "$SED" && SED=sed
Xsed="$SED -e 1s/^X//"
_LT_DECL([], [SED], [1], [A sed program that does not truncate output])
_LT_DECL([], [Xsed], ["\$SED -e 1s/^X//"],
[Sed that helps us avoid accidentally triggering echo(1) options like -n])
])# _LT_DECL_SED
m4_ifndef([AC_PROG_SED], [
############################################################
# NOTE: This macro has been submitted for inclusion into #
# GNU Autoconf as AC_PROG_SED. When it is available in #
# a released version of Autoconf we should remove this #
# macro and use it instead. #
############################################################
m4_defun([AC_PROG_SED],
[AC_MSG_CHECKING([for a sed that does not truncate output])
AC_CACHE_VAL(lt_cv_path_SED,
[# Loop through the user's path and test for sed and gsed.
# Then use that list of sed's as ones to test for truncation.
as_save_IFS=$IFS; IFS=$PATH_SEPARATOR
for as_dir in $PATH
do
IFS=$as_save_IFS
test -z "$as_dir" && as_dir=.
for lt_ac_prog in sed gsed; do
for ac_exec_ext in '' $ac_executable_extensions; do
if $as_executable_p "$as_dir/$lt_ac_prog$ac_exec_ext"; then
lt_ac_sed_list="$lt_ac_sed_list $as_dir/$lt_ac_prog$ac_exec_ext"
fi
done
done
done
IFS=$as_save_IFS
lt_ac_max=0
lt_ac_count=0
# Add /usr/xpg4/bin/sed as it is typically found on Solaris
# along with /bin/sed that truncates output.
for lt_ac_sed in $lt_ac_sed_list /usr/xpg4/bin/sed; do
test ! -f $lt_ac_sed && continue
cat /dev/null > conftest.in
lt_ac_count=0
echo $ECHO_N "0123456789$ECHO_C" >conftest.in
# Check for GNU sed and select it if it is found.
if "$lt_ac_sed" --version 2>&1 < /dev/null | grep 'GNU' > /dev/null; then
lt_cv_path_SED=$lt_ac_sed
break
fi
while true; do
cat conftest.in conftest.in >conftest.tmp
mv conftest.tmp conftest.in
cp conftest.in conftest.nl
echo >>conftest.nl
$lt_ac_sed -e 's/a$//' < conftest.nl >conftest.out || break
cmp -s conftest.out conftest.nl || break
# 10000 chars as input seems more than enough
test $lt_ac_count -gt 10 && break
lt_ac_count=`expr $lt_ac_count + 1`
if test $lt_ac_count -gt $lt_ac_max; then
lt_ac_max=$lt_ac_count
lt_cv_path_SED=$lt_ac_sed
fi
done
done
])
SED=$lt_cv_path_SED
AC_SUBST([SED])
AC_MSG_RESULT([$SED])
])#AC_PROG_SED
])#m4_ifndef
# Old name:
AU_ALIAS([LT_AC_PROG_SED], [AC_PROG_SED])
dnl aclocal-1.4 backwards compatibility:
dnl AC_DEFUN([LT_AC_PROG_SED], [])
# _LT_CHECK_SHELL_FEATURES
# ------------------------
# Find out whether the shell is Bourne or XSI compatible,
# or has some other useful features.
m4_defun([_LT_CHECK_SHELL_FEATURES],
[AC_MSG_CHECKING([whether the shell understands some XSI constructs])
# Try some XSI features
xsi_shell=no
( _lt_dummy="a/b/c"
test "${_lt_dummy##*/},${_lt_dummy%/*},${_lt_dummy#??}"${_lt_dummy%"$_lt_dummy"}, \
= c,a/b,b/c, \
&& eval 'test $(( 1 + 1 )) -eq 2 \
&& test "${#_lt_dummy}" -eq 5' ) >/dev/null 2>&1 \
&& xsi_shell=yes
AC_MSG_RESULT([$xsi_shell])
_LT_CONFIG_LIBTOOL_INIT([xsi_shell='$xsi_shell'])
AC_MSG_CHECKING([whether the shell understands "+="])
lt_shell_append=no
( foo=bar; set foo baz; eval "$[1]+=\$[2]" && test "$foo" = barbaz ) \
>/dev/null 2>&1 \
&& lt_shell_append=yes
AC_MSG_RESULT([$lt_shell_append])
_LT_CONFIG_LIBTOOL_INIT([lt_shell_append='$lt_shell_append'])
if ( (MAIL=60; unset MAIL) || exit) >/dev/null 2>&1; then
lt_unset=unset
else
lt_unset=false
fi
_LT_DECL([], [lt_unset], [0], [whether the shell understands "unset"])dnl
# test EBCDIC or ASCII
case `echo X|tr X '\101'` in
A) # ASCII based system
# \n is not interpreted correctly by Solaris 8 /usr/ucb/tr
lt_SP2NL='tr \040 \012'
lt_NL2SP='tr \015\012 \040\040'
;;
*) # EBCDIC based system
lt_SP2NL='tr \100 \n'
lt_NL2SP='tr \r\n \100\100'
;;
esac
_LT_DECL([SP2NL], [lt_SP2NL], [1], [turn spaces into newlines])dnl
_LT_DECL([NL2SP], [lt_NL2SP], [1], [turn newlines into spaces])dnl
])# _LT_CHECK_SHELL_FEATURES
# _LT_PROG_FUNCTION_REPLACE (FUNCNAME, REPLACEMENT-BODY)
# ------------------------------------------------------
# In `$cfgfile', look for function FUNCNAME delimited by `^FUNCNAME ()$' and
# '^} FUNCNAME ', and replace its body with REPLACEMENT-BODY.
m4_defun([_LT_PROG_FUNCTION_REPLACE],
[dnl {
sed -e '/^$1 ()$/,/^} # $1 /c\
$1 ()\
{\
m4_bpatsubsts([$2], [$], [\\], [^\([ ]\)], [\\\1])
} # Extended-shell $1 implementation' "$cfgfile" > $cfgfile.tmp \
&& mv -f "$cfgfile.tmp" "$cfgfile" \
|| (rm -f "$cfgfile" && cp "$cfgfile.tmp" "$cfgfile" && rm -f "$cfgfile.tmp")
test 0 -eq $? || _lt_function_replace_fail=:
])
# _LT_PROG_REPLACE_SHELLFNS
# -------------------------
# Replace existing portable implementations of several shell functions with
# equivalent extended shell implementations where those features are available..
m4_defun([_LT_PROG_REPLACE_SHELLFNS],
[if test x"$xsi_shell" = xyes; then
_LT_PROG_FUNCTION_REPLACE([func_dirname], [dnl
case ${1} in
*/*) func_dirname_result="${1%/*}${2}" ;;
* ) func_dirname_result="${3}" ;;
esac])
_LT_PROG_FUNCTION_REPLACE([func_basename], [dnl
func_basename_result="${1##*/}"])
_LT_PROG_FUNCTION_REPLACE([func_dirname_and_basename], [dnl
case ${1} in
*/*) func_dirname_result="${1%/*}${2}" ;;
* ) func_dirname_result="${3}" ;;
esac
func_basename_result="${1##*/}"])
_LT_PROG_FUNCTION_REPLACE([func_stripname], [dnl
# pdksh 5.2.14 does not do ${X%$Y} correctly if both X and Y are
# positional parameters, so assign one to ordinary parameter first.
func_stripname_result=${3}
func_stripname_result=${func_stripname_result#"${1}"}
func_stripname_result=${func_stripname_result%"${2}"}])
_LT_PROG_FUNCTION_REPLACE([func_split_long_opt], [dnl
func_split_long_opt_name=${1%%=*}
func_split_long_opt_arg=${1#*=}])
_LT_PROG_FUNCTION_REPLACE([func_split_short_opt], [dnl
func_split_short_opt_arg=${1#??}
func_split_short_opt_name=${1%"$func_split_short_opt_arg"}])
_LT_PROG_FUNCTION_REPLACE([func_lo2o], [dnl
case ${1} in
*.lo) func_lo2o_result=${1%.lo}.${objext} ;;
*) func_lo2o_result=${1} ;;
esac])
_LT_PROG_FUNCTION_REPLACE([func_xform], [ func_xform_result=${1%.*}.lo])
_LT_PROG_FUNCTION_REPLACE([func_arith], [ func_arith_result=$(( $[*] ))])
_LT_PROG_FUNCTION_REPLACE([func_len], [ func_len_result=${#1}])
fi
if test x"$lt_shell_append" = xyes; then
_LT_PROG_FUNCTION_REPLACE([func_append], [ eval "${1}+=\\${2}"])
_LT_PROG_FUNCTION_REPLACE([func_append_quoted], [dnl
func_quote_for_eval "${2}"
dnl m4 expansion turns \\\\ into \\, and then the shell eval turns that into \
eval "${1}+=\\\\ \\$func_quote_for_eval_result"])
# Save a `func_append' function call where possible by direct use of '+='
sed -e 's%func_append \([[a-zA-Z_]]\{1,\}\) "%\1+="%g' $cfgfile > $cfgfile.tmp \
&& mv -f "$cfgfile.tmp" "$cfgfile" \
|| (rm -f "$cfgfile" && cp "$cfgfile.tmp" "$cfgfile" && rm -f "$cfgfile.tmp")
test 0 -eq $? || _lt_function_replace_fail=:
else
# Save a `func_append' function call even when '+=' is not available
sed -e 's%func_append \([[a-zA-Z_]]\{1,\}\) "%\1="$\1%g' $cfgfile > $cfgfile.tmp \
&& mv -f "$cfgfile.tmp" "$cfgfile" \
|| (rm -f "$cfgfile" && cp "$cfgfile.tmp" "$cfgfile" && rm -f "$cfgfile.tmp")
test 0 -eq $? || _lt_function_replace_fail=:
fi
if test x"$_lt_function_replace_fail" = x":"; then
AC_MSG_WARN([Unable to substitute extended shell functions in $ofile])
fi
])
# _LT_PATH_CONVERSION_FUNCTIONS
# -----------------------------
# Determine which file name conversion functions should be used by
# func_to_host_file (and, implicitly, by func_to_host_path). These are needed
# for certain cross-compile configurations and native mingw.
m4_defun([_LT_PATH_CONVERSION_FUNCTIONS],
[AC_REQUIRE([AC_CANONICAL_HOST])dnl
AC_REQUIRE([AC_CANONICAL_BUILD])dnl
AC_MSG_CHECKING([how to convert $build file names to $host format])
AC_CACHE_VAL(lt_cv_to_host_file_cmd,
[case $host in
*-*-mingw* )
case $build in
*-*-mingw* ) # actually msys
lt_cv_to_host_file_cmd=func_convert_file_msys_to_w32
;;
*-*-cygwin* )
lt_cv_to_host_file_cmd=func_convert_file_cygwin_to_w32
;;
* ) # otherwise, assume *nix
lt_cv_to_host_file_cmd=func_convert_file_nix_to_w32
;;
esac
;;
*-*-cygwin* )
case $build in
*-*-mingw* ) # actually msys
lt_cv_to_host_file_cmd=func_convert_file_msys_to_cygwin
;;
*-*-cygwin* )
lt_cv_to_host_file_cmd=func_convert_file_noop
;;
* ) # otherwise, assume *nix
lt_cv_to_host_file_cmd=func_convert_file_nix_to_cygwin
;;
esac
;;
* ) # unhandled hosts (and "normal" native builds)
lt_cv_to_host_file_cmd=func_convert_file_noop
;;
esac
])
to_host_file_cmd=$lt_cv_to_host_file_cmd
AC_MSG_RESULT([$lt_cv_to_host_file_cmd])
_LT_DECL([to_host_file_cmd], [lt_cv_to_host_file_cmd],
[0], [convert $build file names to $host format])dnl
AC_MSG_CHECKING([how to convert $build file names to toolchain format])
AC_CACHE_VAL(lt_cv_to_tool_file_cmd,
[#assume ordinary cross tools, or native build.
lt_cv_to_tool_file_cmd=func_convert_file_noop
case $host in
*-*-mingw* )
case $build in
*-*-mingw* ) # actually msys
lt_cv_to_tool_file_cmd=func_convert_file_msys_to_w32
;;
esac
;;
esac
])
to_tool_file_cmd=$lt_cv_to_tool_file_cmd
AC_MSG_RESULT([$lt_cv_to_tool_file_cmd])
_LT_DECL([to_tool_file_cmd], [lt_cv_to_tool_file_cmd],
[0], [convert $build files to toolchain format])dnl
])# _LT_PATH_CONVERSION_FUNCTIONS
diff --git a/m4/threadlib.m4 b/m4/threadlib.m4
index b015365..32c2ea5 100644
--- a/m4/threadlib.m4
+++ b/m4/threadlib.m4
@@ -1,349 +1,349 @@
# threadlib.m4 serial 10 (gettext-0.18.2) modified by wk 2014-01-24.
dnl Copyright (C) 2005-2014 Free Software Foundation, Inc.
dnl This file is free software; the Free Software Foundation
dnl gives unlimited permission to copy and/or distribute it,
dnl with or without modifications, as long as this notice is preserved.
dnl From Bruno Haible.
dnl gl_THREADLIB
dnl ------------
dnl Tests for a multithreading library to be used.
dnl If the configure.ac contains a definition of the gl_THREADLIB_DEFAULT_NO
dnl (it must be placed before the invocation of gl_THREADLIB_EARLY!), then the
dnl default is 'no', otherwise it is system dependent. In both cases, the user
dnl can change the choice through the options --enable-threads=choice or
dnl --disable-threads.
dnl Defines at most one of the macros USE_POSIX_THREADS, USE_SOLARIS_THREADS,
dnl USE_WINDOWS_THREADS
dnl Sets the variables LIBTHREAD and LTLIBTHREAD to the linker options for use
dnl in a Makefile (LIBTHREAD for use without libtool, LTLIBTHREAD for use with
dnl libtool).
dnl Sets the variables LIBMULTITHREAD and LTLIBMULTITHREAD similarly, for
dnl programs that really need multithread functionality. The difference
dnl between LIBTHREAD and LIBMULTITHREAD is that on platforms supporting weak
dnl symbols, typically LIBTHREAD="" whereas LIBMULTITHREAD="-lpthread".
dnl Sets THREADLIB_CPPFLAGS to -D_REENTRANT or -D_THREAD_SAFE if needed for
dnl multithread-safe programs and adds THREADLIB_CPPFLAGS to CPPFLAGS.
AC_DEFUN([gl_THREADLIB_EARLY],
[
AC_REQUIRE([gl_THREADLIB_EARLY_BODY])
])
dnl The guts of gl_THREADLIB_EARLY. Needs to be expanded only once.
AC_DEFUN([gl_THREADLIB_EARLY_BODY],
[
dnl Ordering constraints: This macro modifies CPPFLAGS in a way that
dnl influences the result of the autoconf tests that test for *_unlocked
dnl declarations, on AIX 5 at least. Therefore it must come early.
AC_BEFORE([$0], [gl_FUNC_GLIBC_UNLOCKED_IO])dnl
AC_BEFORE([$0], [gl_ARGP])dnl
AC_REQUIRE([AC_CANONICAL_HOST])
dnl _GNU_SOURCE is needed for pthread_rwlock_t on glibc systems.
dnl AC_USE_SYSTEM_EXTENSIONS was introduced in autoconf 2.60 and obsoletes
dnl AC_GNU_SOURCE.
m4_ifdef([AC_USE_SYSTEM_EXTENSIONS],
[AC_REQUIRE([AC_USE_SYSTEM_EXTENSIONS])],
[AC_REQUIRE([AC_GNU_SOURCE])])
dnl Check for multithreading.
THREADLIB_CPPFLAGS=""
m4_ifdef([gl_THREADLIB_DEFAULT_NO],
[m4_divert_text([DEFAULTS], [gl_use_threads_default=no])],
[m4_divert_text([DEFAULTS], [gl_use_threads_default=])])
AC_ARG_ENABLE([threads],
AC_HELP_STRING([--enable-threads={posix|solaris|windows}], [specify multithreading API])m4_ifdef([gl_THREADLIB_DEFAULT_NO], [], [
AC_HELP_STRING([--disable-threads], [build without multithread safety])]),
[gl_use_threads=$enableval],
[if test -n "$gl_use_threads_default"; then
gl_use_threads="$gl_use_threads_default"
else
changequote(,)dnl
case "$host_os" in
dnl Disable multithreading by default on OSF/1, because it interferes
dnl with fork()/exec(): When msgexec is linked with -lpthread, its
dnl child process gets an endless segmentation fault inside execvp().
dnl Disable multithreading by default on Cygwin 1.5.x, because it has
dnl bugs that lead to endless loops or crashes. See
- dnl .
+ dnl .
osf*) gl_use_threads=no ;;
cygwin*)
case `uname -r` in
1.[0-5].*) gl_use_threads=no ;;
*) gl_use_threads=yes ;;
esac
;;
*) gl_use_threads=yes ;;
esac
changequote([,])dnl
fi
])
if test "$gl_use_threads" = yes || test "$gl_use_threads" = posix; then
# For using :
case "$host_os" in
osf*)
# On OSF/1, the compiler needs the flag -D_REENTRANT so that it
# groks . cc also understands the flag -pthread, but
# we don't use it because 1. gcc-2.95 doesn't understand -pthread,
# 2. putting a flag into CPPFLAGS that has an effect on the linker
# causes the AC_LINK_IFELSE test below to succeed unexpectedly,
# leading to wrong values of LIBTHREAD and LTLIBTHREAD.
THREADLIB_CPPFLAGS="$THREADLIB_CPPFLAGS -D_REENTRANT"
;;
esac
# Some systems optimize for single-threaded programs by default, and
# need special flags to disable these optimizations. For example, the
# definition of 'errno' in .
case "$host_os" in
aix* | freebsd*)
THREADLIB_CPPFLAGS="$THREADLIB_CPPFLAGS -D_THREAD_SAFE"
;;
solaris*)
THREADLIB_CPPFLAGS="$THREADLIB_CPPFLAGS -D_REENTRANT"
;;
esac
fi
if test x"$THREADLIB_CPPFLAGS" != x ; then
CPPFLAGS="$CPPFLAGS $THREADLIB_CPPFLAGS"
fi
])
dnl The guts of gl_THREADLIB. Needs to be expanded only once.
AC_DEFUN([gl_THREADLIB_BODY],
[
AC_REQUIRE([gl_THREADLIB_EARLY_BODY])
gl_threads_api=none
LIBTHREAD=
LTLIBTHREAD=
LIBMULTITHREAD=
LTLIBMULTITHREAD=
if test "$gl_use_threads" != no; then
dnl Check whether the compiler and linker support weak declarations.
AC_CACHE_CHECK([whether imported symbols can be declared weak],
[gl_cv_have_weak],
[gl_cv_have_weak=no
dnl First, test whether the compiler accepts it syntactically.
AC_LINK_IFELSE(
[AC_LANG_PROGRAM(
[[extern void xyzzy ();
#pragma weak xyzzy]],
[[xyzzy();]])],
[gl_cv_have_weak=maybe])
if test $gl_cv_have_weak = maybe; then
dnl Second, test whether it actually works. On Cygwin 1.7.2, with
dnl gcc 4.3, symbols declared weak always evaluate to the address 0.
AC_RUN_IFELSE(
[AC_LANG_SOURCE([[
#include
#pragma weak fputs
int main ()
{
return (fputs == NULL);
}]])],
[gl_cv_have_weak=yes],
[gl_cv_have_weak=no],
[dnl When cross-compiling, assume that only ELF platforms support
dnl weak symbols.
AC_EGREP_CPP([Extensible Linking Format],
[#ifdef __ELF__
Extensible Linking Format
#endif
],
[gl_cv_have_weak="guessing yes"],
[gl_cv_have_weak="guessing no"])
])
fi
])
if test "$gl_use_threads" = yes || test "$gl_use_threads" = posix; then
# On OSF/1, the compiler needs the flag -pthread or -D_REENTRANT so that
# it groks . It's added above, in gl_THREADLIB_EARLY_BODY.
AC_CHECK_HEADER([pthread.h],
[gl_have_pthread_h=yes], [gl_have_pthread_h=no])
if test "$gl_have_pthread_h" = yes; then
# Other possible tests:
# -lpthreads (FSU threads, PCthreads)
# -lgthreads
gl_have_pthread=
# Test whether both pthread_mutex_lock and pthread_mutexattr_init exist
# in libc. IRIX 6.5 has the first one in both libc and libpthread, but
# the second one only in libpthread, and lock.c needs it.
AC_LINK_IFELSE(
[AC_LANG_PROGRAM(
[[#include ]],
[[pthread_mutex_lock((pthread_mutex_t*)0);
pthread_mutexattr_init((pthread_mutexattr_t*)0);]])],
[gl_have_pthread=yes])
# Test for libpthread by looking for pthread_kill. (Not pthread_self,
# since it is defined as a macro on OSF/1.)
if test -n "$gl_have_pthread"; then
# The program links fine without libpthread. But it may actually
# need to link with libpthread in order to create multiple threads.
AC_CHECK_LIB([pthread], [pthread_kill],
[LIBMULTITHREAD=-lpthread LTLIBMULTITHREAD=-lpthread
# On Solaris and HP-UX, most pthread functions exist also in libc.
# Therefore pthread_in_use() needs to actually try to create a
# thread: pthread_create from libc will fail, whereas
# pthread_create will actually create a thread.
case "$host_os" in
solaris* | hpux*)
AC_DEFINE([PTHREAD_IN_USE_DETECTION_HARD], [1],
[Define if the pthread_in_use() detection is hard.])
esac
])
else
# Some library is needed. Try libpthread and libc_r.
AC_CHECK_LIB([pthread], [pthread_kill],
[gl_have_pthread=yes
LIBTHREAD=-lpthread LTLIBTHREAD=-lpthread
LIBMULTITHREAD=-lpthread LTLIBMULTITHREAD=-lpthread])
if test -z "$gl_have_pthread"; then
# For FreeBSD 4.
AC_CHECK_LIB([c_r], [pthread_kill],
[gl_have_pthread=yes
LIBTHREAD=-lc_r LTLIBTHREAD=-lc_r
LIBMULTITHREAD=-lc_r LTLIBMULTITHREAD=-lc_r])
fi
fi
if test -n "$gl_have_pthread"; then
gl_threads_api=posix
AC_DEFINE([USE_POSIX_THREADS], [1],
[Define if the POSIX multithreading library can be used.])
if test -n "$LIBMULTITHREAD" || test -n "$LTLIBMULTITHREAD"; then
if case "$gl_cv_have_weak" in *yes) true;; *) false;; esac; then
AC_DEFINE([USE_POSIX_THREADS_WEAK], [1],
[Define if references to the POSIX multithreading library should be made weak.])
LIBTHREAD=
LTLIBTHREAD=
fi
fi
fi
fi
fi
if test -z "$gl_have_pthread"; then
if test "$gl_use_threads" = yes || test "$gl_use_threads" = solaris; then
gl_have_solaristhread=
gl_save_LIBS="$LIBS"
LIBS="$LIBS -lthread"
AC_LINK_IFELSE(
[AC_LANG_PROGRAM(
[[
#include
#include
]],
[[thr_self();]])],
[gl_have_solaristhread=yes])
LIBS="$gl_save_LIBS"
if test -n "$gl_have_solaristhread"; then
gl_threads_api=solaris
LIBTHREAD=-lthread
LTLIBTHREAD=-lthread
LIBMULTITHREAD="$LIBTHREAD"
LTLIBMULTITHREAD="$LTLIBTHREAD"
AC_DEFINE([USE_SOLARIS_THREADS], [1],
[Define if the old Solaris multithreading library can be used.])
if case "$gl_cv_have_weak" in *yes) true;; *) false;; esac; then
AC_DEFINE([USE_SOLARIS_THREADS_WEAK], [1],
[Define if references to the old Solaris multithreading library should be made weak.])
LIBTHREAD=
LTLIBTHREAD=
fi
fi
fi
fi
if test -z "$gl_have_pthread"; then
case "$gl_use_threads" in
yes | windows | win32) # The 'win32' is for backward compatibility.
if { case "$host_os" in
mingw*) true;;
*) false;;
esac
}; then
gl_threads_api=windows
AC_DEFINE([USE_WINDOWS_THREADS], [1],
[Define if the native Windows multithreading API can be used.])
fi
;;
esac
fi
fi
AC_MSG_CHECKING([for multithread API to use])
AC_MSG_RESULT([$gl_threads_api])
AC_SUBST([LIBTHREAD])
AC_SUBST([LTLIBTHREAD])
AC_SUBST([LIBMULTITHREAD])
AC_SUBST([LTLIBMULTITHREAD])
])
AC_DEFUN([gl_THREADLIB],
[
AC_REQUIRE([gl_THREADLIB_EARLY])
AC_REQUIRE([gl_THREADLIB_BODY])
])
dnl gl_DISABLE_THREADS
dnl ------------------
dnl Sets the gl_THREADLIB default so that threads are not used by default.
dnl The user can still override it at installation time, by using the
dnl configure option '--enable-threads'.
AC_DEFUN([gl_DISABLE_THREADS], [
m4_divert_text([INIT_PREPARE], [gl_use_threads_default=no])
])
dnl Survey of platforms:
dnl
dnl Platform Available Compiler Supports test-lock
dnl flavours option weak result
dnl --------------- --------- --------- -------- ---------
dnl Linux 2.4/glibc posix -lpthread Y OK
dnl
dnl GNU Hurd/glibc posix
dnl
dnl FreeBSD 5.3 posix -lc_r Y
dnl posix -lkse ? Y
dnl posix -lpthread ? Y
dnl posix -lthr Y
dnl
dnl FreeBSD 5.2 posix -lc_r Y
dnl posix -lkse Y
dnl posix -lthr Y
dnl
dnl FreeBSD 4.0,4.10 posix -lc_r Y OK
dnl
dnl NetBSD 1.6 --
dnl
dnl OpenBSD 3.4 posix -lpthread Y OK
dnl
dnl Mac OS X 10.[123] posix -lpthread Y OK
dnl
dnl Solaris 7,8,9 posix -lpthread Y Sol 7,8: 0.0; Sol 9: OK
dnl solaris -lthread Y Sol 7,8: 0.0; Sol 9: OK
dnl
dnl HP-UX 11 posix -lpthread N (cc) OK
dnl Y (gcc)
dnl
dnl IRIX 6.5 posix -lpthread Y 0.5
dnl
dnl AIX 4.3,5.1 posix -lpthread N AIX 4: 0.5; AIX 5: OK
dnl
dnl OSF/1 4.0,5.1 posix -pthread (cc) N OK
dnl -lpthread (gcc) Y
dnl
dnl Cygwin posix -lpthread Y OK
dnl
dnl Any of the above pth -lpth 0.0
dnl
dnl Mingw windows N OK
dnl
dnl BeOS 5 --
dnl
dnl The test-lock result shows what happens if in test-lock.c EXPLICIT_YIELD is
dnl turned off:
dnl OK if all three tests terminate OK,
dnl 0.5 if the first test terminates OK but the second one loops endlessly,
dnl 0.0 if the first test already loops endlessly.
diff --git a/po/en@boldquot.header b/po/en@boldquot.header
index fedb6a0..506ca9e 100644
--- a/po/en@boldquot.header
+++ b/po/en@boldquot.header
@@ -1,25 +1,25 @@
# All this catalog "translates" are quotation characters.
# The msgids must be ASCII and therefore cannot contain real quotation
# characters, only substitutes like grave accent (0x60), apostrophe (0x27)
# and double quote (0x22). These substitutes look strange; see
-# http://www.cl.cam.ac.uk/~mgk25/ucs/quotes.html
+# https://www.cl.cam.ac.uk/~mgk25/ucs/quotes.html
#
# This catalog translates grave accent (0x60) and apostrophe (0x27) to
# left single quotation mark (U+2018) and right single quotation mark (U+2019).
# It also translates pairs of apostrophe (0x27) to
# left single quotation mark (U+2018) and right single quotation mark (U+2019)
# and pairs of quotation mark (0x22) to
# left double quotation mark (U+201C) and right double quotation mark (U+201D).
#
# When output to an UTF-8 terminal, the quotation characters appear perfectly.
# When output to an ISO-8859-1 terminal, the single quotation marks are
# transliterated to apostrophes (by iconv in glibc 2.2 or newer) or to
# grave/acute accent (by libiconv), and the double quotation marks are
# transliterated to 0x22.
# When output to an ASCII terminal, the single quotation marks are
# transliterated to apostrophes, and the double quotation marks are
# transliterated to 0x22.
#
# This catalog furthermore displays the text between the quotation marks in
# bold face, assuming the VT100/XTerm escape sequences.
#
diff --git a/po/en@quot.header b/po/en@quot.header
index a9647fc..6522f0c 100644
--- a/po/en@quot.header
+++ b/po/en@quot.header
@@ -1,22 +1,22 @@
# All this catalog "translates" are quotation characters.
# The msgids must be ASCII and therefore cannot contain real quotation
# characters, only substitutes like grave accent (0x60), apostrophe (0x27)
# and double quote (0x22). These substitutes look strange; see
-# http://www.cl.cam.ac.uk/~mgk25/ucs/quotes.html
+# https://www.cl.cam.ac.uk/~mgk25/ucs/quotes.html
#
# This catalog translates grave accent (0x60) and apostrophe (0x27) to
# left single quotation mark (U+2018) and right single quotation mark (U+2019).
# It also translates pairs of apostrophe (0x27) to
# left single quotation mark (U+2018) and right single quotation mark (U+2019)
# and pairs of quotation mark (0x22) to
# left double quotation mark (U+201C) and right double quotation mark (U+201D).
#
# When output to an UTF-8 terminal, the quotation characters appear perfectly.
# When output to an ISO-8859-1 terminal, the single quotation marks are
# transliterated to apostrophes (by iconv in glibc 2.2 or newer) or to
# grave/acute accent (by libiconv), and the double quotation marks are
# transliterated to 0x22.
# When output to an ASCII terminal, the single quotation marks are
# transliterated to apostrophes, and the double quotation marks are
# transliterated to 0x22.
#
diff --git a/src/Makefile.am b/src/Makefile.am
index 8aca7df..b7cb023 100644
--- a/src/Makefile.am
+++ b/src/Makefile.am
@@ -1,306 +1,306 @@
# Makefile.am for libgpg-error.
# Copyright (C) 2003, 2004, 2014 g10 Code GmbH
#
# This file is part of libgpg-error.
#
# libgpg-error is free software; you can redistribute it and/or modify
# it under the terms of the GNU Lesser General Public License as
# published by the Free Software Foundation; either version 2.1 of the
# License, or (at your option) any later version.
#
# libgpg-error is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU Lesser General Public License for more details.
#
# You should have received a copy of the GNU Lesser General Public
-# License along with this program; if not, see .
+# License along with this program; if not, see .
# We distribute the generated sources err-sources.h and err-codes.h,
# because they are needed to build the po directory, and they don't
# depend on the configuration anyway.
if HAVE_W32CE_SYSTEM
gpg_extra_headers = gpg-extra/errno.h
extra_cppflags = -idirafter gpg-extra
else
gpg_extra_headers =
extra_cppflags =
endif
localedir = $(datadir)/locale
bin_PROGRAMS = gpg-error
if HAVE_W32_SYSTEM
noinst_PROGRAMS = gen-w32-lock-obj
else
noinst_PROGRAMS = gen-posix-lock-obj
endif
# Distributed lock object definitions for cross compilation.
lock_obj_pub = \
syscfg/lock-obj-pub.aarch64-unknown-linux-gnu.h \
syscfg/lock-obj-pub.aarch64-apple-darwin.h \
syscfg/lock-obj-pub.alpha-unknown-linux-gnu.h \
syscfg/lock-obj-pub.arm-unknown-linux-androideabi.h \
syscfg/lock-obj-pub.arm-unknown-linux-gnueabi.h \
syscfg/lock-obj-pub.arm-unknown-linux-gnueabihf.h \
syscfg/lock-obj-pub.arm-apple-darwin.h \
syscfg/lock-obj-pub.hppa-unknown-linux-gnu.h \
syscfg/lock-obj-pub.i686-pc-gnu.h \
syscfg/lock-obj-pub.i686-pc-kfreebsd-gnu.h \
syscfg/lock-obj-pub.i686-pc-linux-gnu.h \
syscfg/lock-obj-pub.m68k-unknown-linux-gnu.h \
syscfg/lock-obj-pub.mips-unknown-linux-gnu.h \
syscfg/lock-obj-pub.mips64el-unknown-linux-gnuabi64.h \
syscfg/lock-obj-pub.mipsel-unknown-linux-gnu.h \
syscfg/lock-obj-pub.nios2-unknown-linux-gnu.h \
syscfg/lock-obj-pub.or1k-unknown-linux-gnu.h \
syscfg/lock-obj-pub.powerpc-unknown-linux-gnu.h \
syscfg/lock-obj-pub.powerpc64-unknown-linux-gnu.h \
syscfg/lock-obj-pub.powerpc64le-unknown-linux-gnu.h \
syscfg/lock-obj-pub.s390x-ibm-linux-gnu.h \
syscfg/lock-obj-pub.sh4-unknown-linux-gnu.h \
syscfg/lock-obj-pub.sparc-unknown-linux-gnu.h \
syscfg/lock-obj-pub.sparc64-unknown-linux-gnu.h \
syscfg/lock-obj-pub.x86_64-pc-kfreebsd-gnu.h \
syscfg/lock-obj-pub.x86_64-pc-linux-gnu.h \
syscfg/lock-obj-pub.x86_64-pc-linux-gnux32.h \
syscfg/lock-obj-pub.mingw32.h
lib_LTLIBRARIES = libgpg-error.la
nodist_include_HEADERS = gpg-error.h
bin_SCRIPTS = gpg-error-config
m4datadir = $(datadir)/aclocal
m4data_DATA = gpg-error.m4
EXTRA_DIST = mkstrtable.awk err-sources.h.in err-codes.h.in \
mkerrnos.awk errnos.in README \
mkerrcodes.awk mkerrcodes1.awk mkerrcodes2.awk mkerrcodes.c \
mkheader.c gpg-error.h.in mkw32errmap.c w32-add.h w32ce-add.h \
err-sources.h err-codes.h gpg-error-config.in gpg-error.m4 \
gpg-error.vers gpg-error.def.in \
versioninfo.rc.in gpg-error.w32-manifest.in \
$(lock_obj_pub)
BUILT_SOURCES = err-sources.h err-codes.h code-to-errno.h code-from-errno.h \
err-sources-sym.h err-codes-sym.h errnos-sym.h gpg-error.h \
gpg-error.def mkw32errmap.map.c
tmp_files = _mkerrcodes.h _gpg-error.def.h mkw32errmap.tab.h mkw32errmap.map.c
CLEANFILES = err-sources.h err-codes.h code-to-errno.h code-from-errno.h \
gpg-error.h mkerrcodes mkerrcodes.h gpg-error.def mkw32errmap.tab.h \
mkw32errmap.map.c err-sources-sym.h err-codes-sym.h errnos-sym.h \
gpg-extra/errno.h mkheader $(tmp_files) lock-obj-pub.native.h
#
# {{{ Begin Windows part
#
if HAVE_W32_SYSTEM
arch_sources = w32-gettext.c w32-lock.c w32-lock-obj.h w32-thread.c
RCCOMPILE = $(RC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) \
-DLOCALEDIR=\"$(localedir)\" $(AM_CPPFLAGS) $(CPPFLAGS)
LTRCCOMPILE = $(LIBTOOL) --mode=compile --tag=RC $(RCCOMPILE)
SUFFIXES = .rc .lo
.rc.lo:
$(LTRCCOMPILE) -i "$<" -o "$@"
gpg_error_res = versioninfo.lo
no_undefined = -no-undefined
export_symbols = -export-symbols gpg-error.def
# i686-w64-mingw32.gcc version 4.9.1 takes the long long helper
# functions from libgcc_s_sjlj-1.dll and not from a static libgcc. As
# a plain C program we do not use exception handler and thus there is
# no need to use this DLL. Thus we force gcc to link that statically.
extra_ltoptions = -XCClinker -static-libgcc
versioninfo.lo : gpg-error.w32-manifest
install-def-file: gpg-error.def
-$(INSTALL) -d $(DESTDIR)$(libdir)
$(INSTALL) gpg-error.def $(DESTDIR)$(libdir)/gpg-error.def
uninstall-def-file:
-rm $(DESTDIR)$(libdir)/gpg-error.def
libgpg_error_la_DEPENDENCIES = $(gpg_error_res) gpg-error.def
intllibs =
#
# }}} End Windows part
#
else
#
# {{{ Begin Unix part
#
arch_sources = posix-lock.c posix-lock-obj.h posix-thread.c
gpg_error_res =
no_undefined =
export_symbols =
extra_ltoptions =
install-def-file:
uninstall-def-file:
intllibs = @LTLIBINTL@
endif
#
# }}} End Unix part
#
if HAVE_LD_VERSION_SCRIPT
libgpg_error_vers_opt = -Wl,--version-script=$(srcdir)/gpg-error.vers
else
libgpg_error_vers_opt =
endif
libgpg_error_la_LDFLAGS = \
$(no_undefined) $(export_symbols) $(libgpg_error_vers_opt) \
$(extra_ltoptions) -version-info \
@LIBGPG_ERROR_LT_CURRENT@:@LIBGPG_ERROR_LT_REVISION@:@LIBGPG_ERROR_LT_AGE@
libgpg_error_la_SOURCES = gettext.h $(arch_sources) \
gpgrt-int.h init.c init.h version.c lock.h thread.h \
estream.c estream-printf.c estream-printf.h \
strsource.c strerror.c code-to-errno.c code-from-errno.c \
visibility.c visibility.h
nodist_libgpg_error_la_SOURCES = gpg-error.h
# libgpg_error_la_DEPENDENCIES = \
# $(srcdir)/gpg-error.vers
# Note that RCCOMPILE needs the same defines as ..._la_CPPFLAGS but
# without the extra_cppflags because they may include am -idirafter
# which is not supported by the RC compiler.
libgpg_error_la_CPPFLAGS = -DLOCALEDIR=\"$(localedir)\" $(extra_cppflags)
libgpg_error_la_LIBADD = $(gpg_error_res) $(intllibs) $(LIBTHREAD)
gpg_error_SOURCES = strsource-sym.c strerror-sym.c gpg-error.c
gpg_error_CPPFLAGS = -DLOCALEDIR=\"$(localedir)\" $(extra_cppflags)
gpg_error_LDADD = ./libgpg-error.la @LTLIBINTL@
# We build err-sources.h and err-codes.h in the source directory.
# This is needed because gettext does only look into the source
# directory to find the files listed in po/POTFILE.in. To make these
# rules work we also need to depend on Makefile.am and not on the
# generated files Makefile.in or Makefile.
$(srcdir)/err-sources.h: Makefile.am mkstrtable.awk err-sources.h.in
$(AWK) -f $(srcdir)/mkstrtable.awk -v textidx=3 \
$(srcdir)/err-sources.h.in >$@
err-sources-sym.h: Makefile mkstrtable.awk err-sources.h.in
$(AWK) -f $(srcdir)/mkstrtable.awk -v textidx=2 -v nogettext=1 \
$(srcdir)/err-sources.h.in >$@
$(srcdir)/err-codes.h: Makefile.am mkstrtable.awk err-codes.h.in
$(AWK) -f $(srcdir)/mkstrtable.awk -v textidx=3 \
$(srcdir)/err-codes.h.in >$@
err-codes-sym.h: Makefile mkstrtable.awk err-codes.h.in
$(AWK) -f $(srcdir)/mkstrtable.awk -v textidx=2 -v nogettext=1 \
$(srcdir)/err-codes.h.in >$@
code-to-errno.h: Makefile mkerrnos.awk errnos.in
$(AWK) -f $(srcdir)/mkerrnos.awk $(srcdir)/errnos.in >$@
# It is correct to use $(CPP). We want the host's idea of the error codes.
mkerrcodes.h: Makefile mkerrcodes.awk $(gpg_extra_headers)
$(AWK) -f $(srcdir)/mkerrcodes1.awk $(srcdir)/errnos.in >_$@
$(CPP) $(CPPFLAGS) $(extra_cppflags) -P _$@ | grep GPG_ERR_ | \
$(AWK) -f $(srcdir)/mkerrcodes.awk >$@
-rm _$@
if HAVE_W32CE_SYSTEM
# It is correct to use $(CPP). We want the host's idea of the error codes.
mkw32errmap.tab.h: Makefile mkw32errmap.c
$(CPP) -DRESOLVE_MACROS $(srcdir)/mkw32errmap.c | \
grep '{&mkw32errmap_marker' >$@
mkw32errmap.map.c: mkw32errmap
./mkw32errmap --map > $@
gpg-extra/errno.h: mkw32errmap
-$(MKDIR_P) gpg-extra
./mkw32errmap > $@
else
mkw32errmap.map.c:
echo "/*dummy*/" > $@
endif
# We use CC proper for preprocessing thus we have to convince it that
# the data is really to be preprocessed.
gpg-error.def: Makefile gpg-error.def.in
cat $(srcdir)/gpg-error.def.in >_$@.h
$(CPP) $(DEFAULT_INCLUDES) $(INCLUDES) $(extra_cppflags) _$@.h | \
grep -v '^#' >$@
-rm _$@.h
# It is correct to use $(CC_FOR_BUILD) here. We want to run the
# program at build time.
mkerrcodes: mkerrcodes.c mkerrcodes.h Makefile
$(CC_FOR_BUILD) -I. -I$(srcdir) -o $@ $(srcdir)/mkerrcodes.c
if HAVE_W32CE_SYSTEM
# It is correct to use $(CC_FOR_BUILD) here. We want to run the
# program at build time.
mkw32errmap: mkw32errmap.c mkw32errmap.tab.h Makefile
$(CC_FOR_BUILD) -I. -I$(srcdir) -o $@ $(srcdir)/mkw32errmap.c
endif
code-from-errno.h: mkerrcodes Makefile
./mkerrcodes | $(AWK) -f $(srcdir)/mkerrcodes2.awk >$@
errnos-sym.h: Makefile mkstrtable.awk errnos.in
$(AWK) -f $(srcdir)/mkstrtable.awk -v textidx=2 -v nogettext=1 \
-v prefix=GPG_ERR_ -v namespace=errnos_ \
$(srcdir)/errnos.in >$@
mkheader: mkheader.c Makefile
$(CC_FOR_BUILD) -g -O0 -I. -I$(srcdir) -o $@ $(srcdir)/mkheader.c
parts_of_gpg_error_h = \
gpg-error.h.in \
err-sources.h.in \
err-codes.h.in \
errnos.in \
w32-add.h \
w32ce-add.h \
$(lock_obj_pub)
# If we are cross-compiling or building on Windows we better make sure
# that no stale native lock include file will be found by mkheader.
if FORCE_USE_SYSCFG
pre_mkheader_cmds = if test -f lock-obj-pub.native.h; \
then rm lock-obj-pub.native.h; fi
else
pre_mkheader_cmds = :
parts_of_gpg_error_h += ./lock-obj-pub.native.h
./lock-obj-pub.native.h: Makefile gen-posix-lock-obj posix-lock-obj.h
./gen-posix-lock-obj >$@
endif
# We also depend on versioninfo.rc because that is build by
# config.status and thus has up-to-date version numbers.
gpg-error.h: Makefile mkheader $(parts_of_gpg_error_h) \
versioninfo.rc ../config.h
$(pre_mkheader_cmds)
./mkheader $(host_os) $(host_triplet) $(srcdir)/gpg-error.h.in \
../config.h $(PACKAGE_VERSION) $(VERSION_NUMBER) >$@
install-data-local:
if HAVE_W32CE_SYSTEM
-$(MKDIR_P) "$(DESTDIR)$(includedir)/gpg-extra"
$(INSTALL_DATA) gpg-extra/errno.h \
"$(DESTDIR)$(includedir)/gpg-extra/errno.h"
else
:
endif
diff --git a/src/estream-printf.c b/src/estream-printf.c
index 39a813f..091ff7d 100644
--- a/src/estream-printf.c
+++ b/src/estream-printf.c
@@ -1,1875 +1,1875 @@
/* estream-printf.c - Versatile mostly C-99 compliant printf formatting
* Copyright (C) 2007, 2008, 2009, 2010, 2012, 2014 g10 Code GmbH
*
* This file is part of Libestream.
*
* Libestream is free software; you can redistribute it and/or modify
* it under the terms of the GNU Lesser General Public License as
* published by the Free Software Foundation; either version 2.1 of
* the License, or (at your option) any later version.
*
* Libestream is distributed in the hope that it will be useful, but
* WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
- * License along with Libestream; if not, see .
+ * License along with Libestream; if not, see .
*
* ALTERNATIVELY, Libestream may be distributed under the terms of the
* following license, in which case the provisions of this license are
* required INSTEAD OF the GNU General Public License. If you wish to
* allow use of your version of this file only under the terms of the
* GNU General Public License, and not to allow others to use your
* version of this file under the terms of the following license,
* indicate your decision by deleting this paragraph and the license
* below.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, and the entire permission notice in its entirety,
* including the disclaimer of warranties.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. The name of the author may not be used to endorse or promote
* products derived from this software without specific prior
* written permission.
*
* THIS SOFTWARE IS PROVIDED "AS IS" AND ANY EXPRESS OR IMPLIED
* WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
* OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
* DISCLAIMED. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT,
* INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
* (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
* SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
* STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED
* OF THE POSSIBILITY OF SUCH DAMAGE.
*/
/* Required autoconf tests:
AC_TYPE_LONG_LONG_INT defines HAVE_LONG_LONG_INT
AC_TYPE_LONG_DOUBLE defines HAVE_LONG_DOUBLE
AC_TYPE_INTMAX_T defines HAVE_INTMAX_T
AC_TYPE_UINTMAX_T defines HAVE_UINTMAX_T
AC_CHECK_TYPES([ptrdiff_t]) defines HAVE_PTRDIFF_T
AC_CHECK_SIZEOF([unsigned long]) defines SIZEOF_UNSIGNED_LONG
AC_CHECK_SIZEOF([void *]) defines SIZEOF_VOID_P
HAVE_LANGINFO_THOUSANDS_SEP
Note that the file estream.m4 provides the autoconf macro
ESTREAM_PRINTF_INIT which runs all required checks.
See estream-printf.h for ways to tune this code.
Missing stuff: wchar and wint_t
thousands_sep in pr_float.
*/
#ifdef HAVE_CONFIG_H
# include
#endif
#if defined(_WIN32) && !defined(HAVE_W32_SYSTEM)
# define HAVE_W32_SYSTEM 1
# if defined(__MINGW32CE__) && !defined (HAVE_W32CE_SYSTEM)
# define HAVE_W32CE_SYSTEM
# endif
#endif
#include
#include
#include
#include
#include
#include
#include
#include
#if defined(HAVE_INTMAX_T) || defined(HAVE_UINTMAX_T)
# ifdef HAVE_STDINT_H
# include
# endif
#endif
#ifdef HAVE_LANGINFO_THOUSANDS_SEP
#include
#endif
#ifdef HAVE_W32CE_SYSTEM
#include /* ERRNO replacement. */
#endif
#ifdef _ESTREAM_PRINTF_EXTRA_INCLUDE
# include _ESTREAM_PRINTF_EXTRA_INCLUDE
#endif
#include "estream-printf.h"
/* #define DEBUG 1 */
/* Allow redefinition of asprintf used realloc function. */
#if defined(_ESTREAM_PRINTF_REALLOC)
#define my_printf_realloc(a,b) _ESTREAM_PRINTF_REALLOC((a),(b))
#else
#define my_printf_realloc(a,b) fixed_realloc((a),(b))
#endif
/* A wrapper to set ERRNO. */
#ifdef HAVE_W32CE_SYSTEM
# define _set_errno(a) gpg_err_set_errno ((a))
#else
# define _set_errno(a) do { errno = (a); } while (0)
#endif
/* Calculate array dimension. */
#ifndef DIM
#define DIM(array) (sizeof (array) / sizeof (*array))
#endif
/* We allow for that many args without requiring malloced memory. */
#define DEFAULT_MAX_ARGSPECS 5
/* We allow for that many values without requiring malloced memory. */
#define DEFAULT_MAX_VALUES 8
/* We allocate this many new array argspec elements each time. */
#define ARGSPECS_BUMP_VALUE 10
/* Special values for the field width and the precision. */
#define NO_FIELD_VALUE (-1)
#define STAR_FIELD_VALUE (-2)
/* Bit valuues used for the conversion flags. */
#define FLAG_GROUPING 1
#define FLAG_LEFT_JUST 2
#define FLAG_PLUS_SIGN 4
#define FLAG_SPACE_PLUS 8
#define FLAG_ALT_CONV 16
#define FLAG_ZERO_PAD 32
/* Constants used the length modifiers. */
typedef enum
{
LENMOD_NONE = 0,
LENMOD_CHAR, /* "hh" */
LENMOD_SHORT, /* "h" */
LENMOD_LONG, /* "l" */
LENMOD_LONGLONG, /* "ll" */
LENMOD_INTMAX, /* "j" */
LENMOD_SIZET, /* "z" */
LENMOD_PTRDIFF, /* "t" */
LENMOD_LONGDBL /* "L" */
} lenmod_t;
/* All the conversion specifiers. */
typedef enum
{
CONSPEC_UNKNOWN = 0,
CONSPEC_DECIMAL,
CONSPEC_OCTAL,
CONSPEC_UNSIGNED,
CONSPEC_HEX,
CONSPEC_HEX_UP,
CONSPEC_FLOAT,
CONSPEC_FLOAT_UP,
CONSPEC_EXP,
CONSPEC_EXP_UP,
CONSPEC_F_OR_G,
CONSPEC_F_OR_G_UP,
CONSPEC_HEX_EXP,
CONSPEC_HEX_EXP_UP,
CONSPEC_CHAR,
CONSPEC_STRING,
CONSPEC_POINTER,
CONSPEC_STRERROR,
CONSPEC_BYTES_SO_FAR
} conspec_t;
/* Constants describing all the suppoorted types. Note that we list
all the types we know about even if certain types are not available
on this system. */
typedef enum
{
VALTYPE_UNSUPPORTED = 0, /* Artificial type for error detection. */
VALTYPE_CHAR,
VALTYPE_SCHAR,
VALTYPE_UCHAR,
VALTYPE_SHORT,
VALTYPE_USHORT,
VALTYPE_INT,
VALTYPE_UINT,
VALTYPE_LONG,
VALTYPE_ULONG,
VALTYPE_LONGLONG,
VALTYPE_ULONGLONG,
VALTYPE_DOUBLE,
VALTYPE_LONGDOUBLE,
VALTYPE_STRING,
VALTYPE_INTMAX,
VALTYPE_UINTMAX,
VALTYPE_SIZE,
VALTYPE_PTRDIFF,
VALTYPE_POINTER,
VALTYPE_CHAR_PTR,
VALTYPE_SCHAR_PTR,
VALTYPE_SHORT_PTR,
VALTYPE_INT_PTR,
VALTYPE_LONG_PTR,
VALTYPE_LONGLONG_PTR,
VALTYPE_INTMAX_PTR,
VALTYPE_SIZE_PTR,
VALTYPE_PTRDIFF_PTR
} valtype_t;
/* A union used to store the actual values. */
typedef union
{
char a_char;
signed char a_schar;
unsigned char a_uchar;
short a_short;
unsigned short a_ushort;
int a_int;
unsigned int a_uint;
long int a_long;
unsigned long int a_ulong;
#ifdef HAVE_LONG_LONG_INT
long long int a_longlong;
unsigned long long int a_ulonglong;
#endif
double a_double;
#ifdef HAVE_LONG_DOUBLE
long double a_longdouble;
#endif
const char *a_string;
#ifdef HAVE_INTMAX_T
intmax_t a_intmax;
#endif
#ifdef HAVE_UINTMAX_T
intmax_t a_uintmax;
#endif
size_t a_size;
#ifdef HAVE_PTRDIFF_T
ptrdiff_t a_ptrdiff;
#endif
void *a_void_ptr;
char *a_char_ptr;
signed char *a_schar_ptr;
short *a_short_ptr;
int *a_int_ptr;
long *a_long_ptr;
#ifdef HAVE_LONG_LONG_INT
long long int *a_longlong_ptr;
#endif
#ifdef HAVE_INTMAX_T
intmax_t *a_intmax_ptr;
#endif
size_t *a_size_ptr;
#ifdef HAVE_PTRDIFF_T
ptrdiff_t *a_ptrdiff_ptr;
#endif
} value_t;
/* An object used to keep track of a format option and arguments. */
struct argspec_s
{
size_t length; /* The length of these args including the percent. */
unsigned int flags; /* The conversion flags (bits defined by FLAG_foo). */
int width; /* The field width. */
int precision; /* The precision. */
lenmod_t lenmod; /* The length modifier. */
conspec_t conspec; /* The conversion specifier. */
int arg_pos; /* The position of the argument. This one may
be -1 to indicate that no value is expected
(e.g. for "%m"). */
int width_pos; /* The position of the argument for a field
width star's value. 0 for not used. */
int precision_pos; /* The position of the argument for the a
precision star's value. 0 for not used. */
valtype_t vt; /* The type of the corresponding argument. */
};
typedef struct argspec_s *argspec_t;
/* An object to build up a table of values and their types. */
struct valueitem_s
{
valtype_t vt; /* The type of the value. */
value_t value; /* The value. */
};
typedef struct valueitem_s *valueitem_t;
/* Not all systems have a C-90 compliant realloc. To cope with this
we use this simple wrapper. */
#ifndef _ESTREAM_PRINTF_REALLOC
static void *
fixed_realloc (void *a, size_t n)
{
if (!a)
return malloc (n);
if (!n)
{
free (a);
return NULL;
}
return realloc (a, n);
}
#endif /*!_ESTREAM_PRINTF_REALLOC*/
#ifdef DEBUG
static void
dump_argspecs (argspec_t arg, size_t argcount)
{
int idx;
for (idx=0; argcount; argcount--, arg++, idx++)
fprintf (stderr,
"%2d: len=%u flags=%u width=%d prec=%d mod=%d "
"con=%d vt=%d pos=%d-%d-%d\n",
idx,
(unsigned int)arg->length,
arg->flags,
arg->width,
arg->precision,
arg->lenmod,
arg->conspec,
arg->vt,
arg->arg_pos,
arg->width_pos,
arg->precision_pos);
}
#endif /*DEBUG*/
/* Set the vt field for ARG. */
static void
compute_type (argspec_t arg)
{
switch (arg->conspec)
{
case CONSPEC_UNKNOWN:
arg->vt = VALTYPE_UNSUPPORTED;
break;
case CONSPEC_DECIMAL:
switch (arg->lenmod)
{
case LENMOD_CHAR: arg->vt = VALTYPE_SCHAR; break;
case LENMOD_SHORT: arg->vt = VALTYPE_SHORT; break;
case LENMOD_LONG: arg->vt = VALTYPE_LONG; break;
case LENMOD_LONGLONG: arg->vt = VALTYPE_LONGLONG; break;
case LENMOD_INTMAX: arg->vt = VALTYPE_INTMAX; break;
case LENMOD_SIZET: arg->vt = VALTYPE_SIZE; break;
case LENMOD_PTRDIFF: arg->vt = VALTYPE_PTRDIFF; break;
default: arg->vt = VALTYPE_INT; break;
}
break;
case CONSPEC_OCTAL:
case CONSPEC_UNSIGNED:
case CONSPEC_HEX:
case CONSPEC_HEX_UP:
switch (arg->lenmod)
{
case LENMOD_CHAR: arg->vt = VALTYPE_UCHAR; break;
case LENMOD_SHORT: arg->vt = VALTYPE_USHORT; break;
case LENMOD_LONG: arg->vt = VALTYPE_ULONG; break;
case LENMOD_LONGLONG: arg->vt = VALTYPE_ULONGLONG; break;
case LENMOD_INTMAX: arg->vt = VALTYPE_UINTMAX; break;
case LENMOD_SIZET: arg->vt = VALTYPE_SIZE; break;
case LENMOD_PTRDIFF: arg->vt = VALTYPE_PTRDIFF; break;
default: arg->vt = VALTYPE_UINT; break;
}
break;
case CONSPEC_FLOAT:
case CONSPEC_FLOAT_UP:
case CONSPEC_EXP:
case CONSPEC_EXP_UP:
case CONSPEC_F_OR_G:
case CONSPEC_F_OR_G_UP:
case CONSPEC_HEX_EXP:
case CONSPEC_HEX_EXP_UP:
switch (arg->lenmod)
{
case LENMOD_LONGDBL: arg->vt = VALTYPE_LONGDOUBLE; break;
case LENMOD_LONG: arg->vt = VALTYPE_DOUBLE; break;
default: arg->vt = VALTYPE_DOUBLE; break;
}
break;
case CONSPEC_CHAR:
arg->vt = VALTYPE_INT;
break;
case CONSPEC_STRING:
arg->vt = VALTYPE_STRING;
break;
case CONSPEC_POINTER:
arg->vt = VALTYPE_POINTER;
break;
case CONSPEC_STRERROR:
arg->vt = VALTYPE_STRING;
break;
case CONSPEC_BYTES_SO_FAR:
switch (arg->lenmod)
{
case LENMOD_CHAR: arg->vt = VALTYPE_SCHAR_PTR; break;
case LENMOD_SHORT: arg->vt = VALTYPE_SHORT_PTR; break;
case LENMOD_LONG: arg->vt = VALTYPE_LONG_PTR; break;
case LENMOD_LONGLONG: arg->vt = VALTYPE_LONGLONG_PTR; break;
case LENMOD_INTMAX: arg->vt = VALTYPE_INTMAX_PTR; break;
case LENMOD_SIZET: arg->vt = VALTYPE_SIZE_PTR; break;
case LENMOD_PTRDIFF: arg->vt = VALTYPE_PTRDIFF_PTR; break;
default: arg->vt = VALTYPE_INT_PTR; break;
}
break;
}
}
/* Parse the FORMAT string and populate the specification array stored
at the address ARGSPECS_ADDR. The caller has provided enough space
to store up to MAX_ARGSPECS in that buffer. The function may
however ignore the provided buffer and malloc a larger one. On
success the address of that larger buffer will be stored at
ARGSPECS_ADDR. The actual number of specifications will be
returned at R_ARGSPECS_COUNT. */
static int
parse_format (const char *format,
argspec_t *argspecs_addr, size_t max_argspecs,
size_t *r_argspecs_count)
{
const char *s;
argspec_t argspecs = *argspecs_addr;
argspec_t arg;
size_t argcount = 0;
if (!format)
goto leave_einval;
for (; *format; format++)
{
unsigned int flags;
int width, precision;
lenmod_t lenmod;
conspec_t conspec;
int arg_pos, width_pos, precision_pos;
if (*format != '%')
continue;
s = ++format;
if (!*s)
goto leave_einval;
if (*s == '%')
continue; /* Just a quoted percent. */
/* First check whether there is a positional argument. */
arg_pos = 0; /* No positional argument given. */
if (*s >= '1' && *s <= '9')
{
const char *save_s = s;
arg_pos = (*s++ - '0');
for (; *s >= '0' && *s <= '9'; s++)
arg_pos = 10*arg_pos + (*s - '0');
if (arg_pos < 0)
goto leave_einval; /* Overflow during conversion. */
if (*s == '$')
s++;
else
{
arg_pos = 0;
s = save_s;
}
}
/* Parse the flags. */
flags = 0;
for ( ; *s; s++)
{
switch (*s)
{
case '\'': flags |= FLAG_GROUPING; break;
case '-': flags |= FLAG_LEFT_JUST; break;
case '+': flags |= FLAG_PLUS_SIGN; break;
case ' ': flags |= FLAG_SPACE_PLUS; break;
case '#': flags |= FLAG_ALT_CONV; break;
case '0': flags |= FLAG_ZERO_PAD; break;
default:
goto flags_parsed;
}
}
flags_parsed:
/* Parse the field width. */
width_pos = 0;
if (*s == '*')
{
width = STAR_FIELD_VALUE;
s++;
/* If we have a positional argument, another one might also
be used to give the position of the star's value. */
if (arg_pos && *s >= '1' && *s <= '9')
{
width_pos = (*s++ - '0');
for (; *s >= '0' && *s <= '9'; s++)
width_pos = 10*width_pos + (*s - '0');
if (width_pos < 1)
goto leave_einval; /* Overflow during conversion. */
if (*s != '$')
goto leave_einval; /* Not followed by $. */
s++;
}
}
else if ( *s >= '0' && *s <= '9')
{
width = (*s++ - '0');
for (; *s >= '0' && *s <= '9'; s++)
{
if (!width && *s == '0')
goto leave_einval; /* Leading zeroes are not allowed.
Fixme: check what other
implementations do. */
width = 10*width + (*s - '0');
}
if (width < 0)
goto leave_einval; /* Overflow during conversion. */
}
else
width = NO_FIELD_VALUE;
/* Parse the precision. */
precision_pos = 0;
precision = NO_FIELD_VALUE;
if (*s == '.')
{
int ignore_value = (s[1] == '-');
s++;
if (*s == '*')
{
precision = STAR_FIELD_VALUE;
s++;
/* If we have a positional argument, another one might also
be used to give the position of the star's value. */
if (arg_pos && *s >= '1' && *s <= '9')
{
precision_pos = (*s++ - '0');
for (; *s >= '0' && *s <= '9'; s++)
precision_pos = 10*precision_pos + (*s - '0');
if (precision_pos < 1)
goto leave_einval; /* Overflow during conversion. */
if (*s != '$')
goto leave_einval; /* Not followed by $. */
s++;
}
}
else if ( *s >= '0' && *s <= '9')
{
precision = (*s++ - '0');
for (; *s >= '0' && *s <= '9'; s++)
{
if (!precision && *s == '0')
goto leave_einval; /* Leading zeroes are not allowed.
Fixme: check what other
implementations do. */
precision = 10*precision + (*s - '0');
}
if (precision < 0)
goto leave_einval; /* Overflow during conversion. */
}
else
precision = 0;
if (ignore_value)
precision = NO_FIELD_VALUE;
}
/* Parse the length modifiers. */
switch (*s)
{
case 'h':
if (s[1] == 'h')
{
lenmod = LENMOD_CHAR;
s++;
}
else
lenmod = LENMOD_SHORT;
s++;
break;
case 'l':
if (s[1] == 'l')
{
lenmod = LENMOD_LONGLONG;
s++;
}
else
lenmod = LENMOD_LONG;
s++;
break;
case 'j': lenmod = LENMOD_INTMAX; s++; break;
case 'z': lenmod = LENMOD_SIZET; s++; break;
case 't': lenmod = LENMOD_PTRDIFF; s++; break;
case 'L': lenmod = LENMOD_LONGDBL; s++; break;
default: lenmod = LENMOD_NONE; break;
}
/* Parse the conversion specifier. */
switch (*s)
{
case 'd':
case 'i': conspec = CONSPEC_DECIMAL; break;
case 'o': conspec = CONSPEC_OCTAL; break;
case 'u': conspec = CONSPEC_UNSIGNED; break;
case 'x': conspec = CONSPEC_HEX; break;
case 'X': conspec = CONSPEC_HEX_UP; break;
case 'f': conspec = CONSPEC_FLOAT; break;
case 'F': conspec = CONSPEC_FLOAT_UP; break;
case 'e': conspec = CONSPEC_EXP; break;
case 'E': conspec = CONSPEC_EXP_UP; break;
case 'g': conspec = CONSPEC_F_OR_G; break;
case 'G': conspec = CONSPEC_F_OR_G_UP; break;
case 'a': conspec = CONSPEC_HEX_EXP; break;
case 'A': conspec = CONSPEC_HEX_EXP_UP; break;
case 'c': conspec = CONSPEC_CHAR; break;
case 's': conspec = CONSPEC_STRING; break;
case 'p': conspec = CONSPEC_POINTER; break;
case 'n': conspec = CONSPEC_BYTES_SO_FAR; break;
case 'C': conspec = CONSPEC_CHAR; lenmod = LENMOD_LONG; break;
case 'S': conspec = CONSPEC_STRING; lenmod = LENMOD_LONG; break;
case 'm': conspec = CONSPEC_STRERROR; arg_pos = -1; break;
default: conspec = CONSPEC_UNKNOWN;
}
/* Save the args. */
if (argcount >= max_argspecs)
{
/* We either need to allocate a new array instead of the
caller provided one or realloc the array. Instead of
using realloc we allocate a new one and release the
original one then. */
size_t n, newmax;
argspec_t newarg;
newmax = max_argspecs + ARGSPECS_BUMP_VALUE;
if (newmax <= max_argspecs)
goto leave_einval; /* Too many arguments. */
newarg = calloc (newmax, sizeof *newarg);
if (!newarg)
goto leave;
for (n=0; n < argcount; n++)
newarg[n] = argspecs[n];
if (argspecs != *argspecs_addr)
free (argspecs);
argspecs = newarg;
max_argspecs = newmax;
}
arg = argspecs + argcount;
arg->length = s - format + 2;
arg->flags = flags;
arg->width = width;
arg->precision = precision;
arg->lenmod = lenmod;
arg->conspec = conspec;
arg->arg_pos = arg_pos;
arg->width_pos = width_pos;
arg->precision_pos = precision_pos;
compute_type (arg);
argcount++;
format = s;
}
*argspecs_addr = argspecs;
*r_argspecs_count = argcount;
return 0; /* Success. */
leave_einval:
_set_errno (EINVAL);
leave:
if (argspecs != *argspecs_addr)
free (argspecs);
*argspecs_addr = NULL;
return -1;
}
/* This function reads all the values as specified by VALUETABLE into
VALUETABLE. The values are expected in VAARGS. The function
returns -1 if a specified type is not supported. */
static int
read_values (valueitem_t valuetable, size_t valuetable_len, va_list vaargs)
{
int validx;
for (validx=0; validx < valuetable_len; validx++)
{
value_t *value = &valuetable[validx].value;
valtype_t vt = valuetable[validx].vt;
switch (vt)
{
case VALTYPE_CHAR: value->a_char = va_arg (vaargs, int); break;
case VALTYPE_CHAR_PTR:
value->a_char_ptr = va_arg (vaargs, char *);
break;
case VALTYPE_SCHAR: value->a_schar = va_arg (vaargs, int); break;
case VALTYPE_SCHAR_PTR:
value->a_schar_ptr = va_arg (vaargs, signed char *);
break;
case VALTYPE_UCHAR: value->a_uchar = va_arg (vaargs, int); break;
case VALTYPE_SHORT: value->a_short = va_arg (vaargs, int); break;
case VALTYPE_USHORT: value->a_ushort = va_arg (vaargs, int); break;
case VALTYPE_SHORT_PTR:
value->a_short_ptr = va_arg (vaargs, short *);
break;
case VALTYPE_INT:
value->a_int = va_arg (vaargs, int);
break;
case VALTYPE_INT_PTR:
value->a_int_ptr = va_arg (vaargs, int *);
break;
case VALTYPE_UINT:
value->a_uint = va_arg (vaargs, unsigned int);
break;
case VALTYPE_LONG:
value->a_long = va_arg (vaargs, long);
break;
case VALTYPE_ULONG:
value->a_ulong = va_arg (vaargs, unsigned long);
break;
case VALTYPE_LONG_PTR:
value->a_long_ptr = va_arg (vaargs, long *);
break;
#ifdef HAVE_LONG_LONG_INT
case VALTYPE_LONGLONG:
value->a_longlong = va_arg (vaargs, long long int);
break;
case VALTYPE_ULONGLONG:
value->a_ulonglong = va_arg (vaargs, unsigned long long int);
break;
case VALTYPE_LONGLONG_PTR:
value->a_longlong_ptr = va_arg (vaargs, long long *);
break;
#endif
case VALTYPE_DOUBLE:
value->a_double = va_arg (vaargs, double);
break;
#ifdef HAVE_LONG_DOUBLE
case VALTYPE_LONGDOUBLE:
value->a_longdouble = va_arg (vaargs, long double);
break;
#endif
case VALTYPE_STRING:
value->a_string = va_arg (vaargs, const char *);
break;
case VALTYPE_POINTER:
value->a_void_ptr = va_arg (vaargs, void *);
break;
#ifdef HAVE_INTMAX_T
case VALTYPE_INTMAX:
value->a_intmax = va_arg (vaargs, intmax_t);
break;
case VALTYPE_INTMAX_PTR:
value->a_intmax_ptr = va_arg (vaargs, intmax_t *);
break;
#endif
#ifdef HAVE_UINTMAX_T
case VALTYPE_UINTMAX:
value->a_uintmax = va_arg (vaargs, uintmax_t);
break;
#endif
case VALTYPE_SIZE:
value->a_size = va_arg (vaargs, size_t);
break;
case VALTYPE_SIZE_PTR:
value->a_size_ptr = va_arg (vaargs, size_t *);
break;
#ifdef HAVE_PTRDIFF_T
case VALTYPE_PTRDIFF:
value->a_ptrdiff = va_arg (vaargs, ptrdiff_t);
break;
case VALTYPE_PTRDIFF_PTR:
value->a_ptrdiff_ptr = va_arg (vaargs, ptrdiff_t *);
break;
#endif
default: /* Unsupported type. */
return -1;
}
}
return 0;
}
/* Output COUNT padding characters PADCHAR and update NBYTES by the
number of bytes actually written. */
static int
pad_out (estream_printf_out_t outfnc, void *outfncarg,
int padchar, int count, size_t *nbytes)
{
char buf[32];
size_t n;
int rc;
while (count > 0)
{
n = (count <= sizeof buf)? count : sizeof buf;
memset (buf, padchar, n);
rc = outfnc (outfncarg, buf, n);
if (rc)
return rc;
*nbytes += n;
count -= n;
}
return 0;
}
/* "d,i,o,u,x,X" formatting. OUTFNC and OUTFNCARG describes the
output routine, ARG gives the argument description and VALUE the
actual value (its type is available through arg->vt). */
static int
pr_integer (estream_printf_out_t outfnc, void *outfncarg,
argspec_t arg, value_t value, size_t *nbytes)
{
int rc;
#ifdef HAVE_LONG_LONG_INT
unsigned long long aulong;
#else
unsigned long aulong;
#endif
char numbuf[100];
char *p, *pend;
size_t n;
char signchar = 0;
int n_prec; /* Number of extra precision digits required. */
int n_extra; /* Extra number of prefix or sign characters. */
if (arg->conspec == CONSPEC_DECIMAL)
{
#ifdef HAVE_LONG_LONG_INT
long long along;
#else
long along;
#endif
switch (arg->vt)
{
case VALTYPE_SHORT: along = value.a_short; break;
case VALTYPE_INT: along = value.a_int; break;
case VALTYPE_LONG: along = value.a_long; break;
#ifdef HAVE_LONG_LONG_INT
case VALTYPE_LONGLONG: along = value.a_longlong; break;
case VALTYPE_SIZE: along = value.a_size; break;
# ifdef HAVE_INTMAX_T
case VALTYPE_INTMAX: along = value.a_intmax; break;
# endif
# ifdef HAVE_PTRDIFF_T
case VALTYPE_PTRDIFF: along = value.a_ptrdiff; break;
# endif
#endif /*HAVE_LONG_LONG_INT*/
default:
return -1;
}
if (along < 0)
{
aulong = -along;
signchar = '-';
}
else
aulong = along;
}
else
{
switch (arg->vt)
{
case VALTYPE_USHORT: aulong = value.a_ushort; break;
case VALTYPE_UINT: aulong = value.a_uint; break;
case VALTYPE_ULONG: aulong = value.a_ulong; break;
#ifdef HAVE_LONG_LONG_INT
case VALTYPE_ULONGLONG: aulong = value.a_ulonglong; break;
case VALTYPE_SIZE: aulong = value.a_size; break;
# ifdef HAVE_UINTMAX_T
case VALTYPE_UINTMAX: aulong = value.a_uintmax; break;
# endif
# ifdef HAVE_PTRDIFF_T
case VALTYPE_PTRDIFF: aulong = value.a_ptrdiff; break;
# endif
#endif /*HAVE_LONG_LONG_INT*/
default:
return -1;
}
}
if (signchar == '-')
;
else if ((arg->flags & FLAG_PLUS_SIGN))
signchar = '+';
else if ((arg->flags & FLAG_SPACE_PLUS))
signchar = ' ';
n_extra = !!signchar;
/* We build the string up backwards. */
p = pend = numbuf + DIM(numbuf);
if ((!aulong && !arg->precision))
;
else if (arg->conspec == CONSPEC_DECIMAL
|| arg->conspec == CONSPEC_UNSIGNED)
{
int grouping = -1;
const char * grouping_string =
#ifdef HAVE_LANGINFO_THOUSANDS_SEP
nl_langinfo(THOUSANDS_SEP);
#else
"'";
#endif
do
{
if ((arg->flags & FLAG_GROUPING)
&& (++grouping == 3) && *grouping_string)
{
*--p = *grouping_string;
grouping = 0;
}
*--p = '0' + (aulong % 10);
aulong /= 10;
}
while (aulong);
}
else if (arg->conspec == CONSPEC_OCTAL)
{
do
{
*--p = '0' + (aulong % 8);
aulong /= 8;
}
while (aulong);
if ((arg->flags & FLAG_ALT_CONV) && *p != '0')
*--p = '0';
}
else /* HEX or HEXUP */
{
const char *digits = ((arg->conspec == CONSPEC_HEX)
? "0123456789abcdef" : "0123456789ABCDEF");
do
{
*--p = digits[(aulong % 16)];
aulong /= 16;
}
while (aulong);
if ((arg->flags & FLAG_ALT_CONV))
n_extra += 2;
}
n = pend - p;
if ((arg->flags & FLAG_ZERO_PAD)
&& arg->precision == NO_FIELD_VALUE && !(arg->flags & FLAG_LEFT_JUST)
&& n && arg->width - n_extra > n )
n_prec = arg->width - n_extra - n;
else if (arg->precision > 0 && arg->precision > n)
n_prec = arg->precision - n;
else
n_prec = 0;
if (!(arg->flags & FLAG_LEFT_JUST)
&& arg->width >= 0 && arg->width - n_extra > n
&& arg->width - n_extra - n >= n_prec )
{
rc = pad_out (outfnc, outfncarg, ' ',
arg->width - n_extra - n - n_prec, nbytes);
if (rc)
return rc;
}
if (signchar)
{
rc = outfnc (outfncarg, &signchar, 1);
if (rc)
return rc;
*nbytes += 1;
}
if ((arg->flags & FLAG_ALT_CONV)
&& (arg->conspec == CONSPEC_HEX || arg->conspec == CONSPEC_HEX_UP))
{
rc = outfnc (outfncarg, arg->conspec == CONSPEC_HEX? "0x": "0X", 2);
if (rc)
return rc;
*nbytes += 2;
}
if (n_prec)
{
rc = pad_out (outfnc, outfncarg, '0', n_prec, nbytes);
if (rc)
return rc;
}
rc = outfnc (outfncarg, p, pend - p);
if (rc)
return rc;
*nbytes += pend - p;
if ((arg->flags & FLAG_LEFT_JUST)
&& arg->width >= 0 && arg->width - n_extra - n_prec > n)
{
rc = pad_out (outfnc, outfncarg, ' ',
arg->width - n_extra - n_prec - n, nbytes);
if (rc)
return rc;
}
return 0;
}
/* "e,E,f,F,g,G,a,A" formatting. OUTFNC and OUTFNCARG describes the
output routine, ARG gives the argument description and VALUE the
actual value (its type is available through arg->vt). For
portability reasons sprintf is used for the actual formatting.
This is useful because sprint is the only standard function to
convert a floating number into its ascii representation. To avoid
using malloc we just pass the precision to sprintf and do the final
formatting with our own code. */
static int
pr_float (estream_printf_out_t outfnc, void *outfncarg,
argspec_t arg, value_t value, size_t *nbytes)
{
int rc;
#ifdef HAVE_LONG_DOUBLE
long double adblfloat = 0; /* Just to please gcc. */
int use_dbl = 0;
#endif
double afloat;
char numbuf[350];
char formatstr[20];
char *p, *pend;
size_t n;
char signchar = 0;
int n_extra; /* Extra number of prefix or sign characters. */
switch (arg->vt)
{
case VALTYPE_DOUBLE: afloat = value.a_double; break;
#ifdef HAVE_LONG_DOUBLE
case VALTYPE_LONGDOUBLE:
afloat = 0; /* Just to please gcc. */
adblfloat = value.a_longdouble;
use_dbl=1; break;
#endif
default:
return -1;
}
/* We build the string using sprint. */
p = formatstr + sizeof formatstr;
*--p = 0;
switch (arg->conspec)
{
case CONSPEC_FLOAT: *--p = 'f'; break;
case CONSPEC_FLOAT_UP: *--p = 'F'; break;
case CONSPEC_EXP: *--p = 'e'; break;
case CONSPEC_EXP_UP: *--p = 'E'; break;
case CONSPEC_F_OR_G: *--p = 'g'; break;
case CONSPEC_F_OR_G_UP: *--p = 'G'; break;
case CONSPEC_HEX_EXP: *--p = 'a'; break;
case CONSPEC_HEX_EXP_UP: *--p = 'A'; break;
default:
return -1; /* Actually a bug. */
}
#ifdef HAVE_LONG_DOUBLE
if (use_dbl)
*--p = 'L';
#endif
if (arg->precision != NO_FIELD_VALUE)
{
/* Limit it to a meaningful value so that even a stupid sprintf
won't overflow our buffer. */
n = arg->precision <= 100? arg->precision : 100;
do
{
*--p = '0' + (n % 10);
n /= 10;
}
while (n);
*--p = '.';
}
if ((arg->flags & FLAG_ALT_CONV))
*--p = '#';
*--p = '%';
#ifdef HAVE_LONG_DOUBLE
if (use_dbl)
sprintf (numbuf, p, adblfloat);
else
#endif /*HAVE_LONG_DOUBLE*/
sprintf (numbuf, p, afloat);
p = numbuf;
n = strlen (numbuf);
pend = p + n;
if (*p =='-')
{
signchar = '-';
p++;
n--;
}
else if ((arg->flags & FLAG_PLUS_SIGN))
signchar = '+';
else if ((arg->flags & FLAG_SPACE_PLUS))
signchar = ' ';
n_extra = !!signchar;
if (!(arg->flags & FLAG_LEFT_JUST)
&& arg->width >= 0 && arg->width - n_extra > n)
{
rc = pad_out (outfnc, outfncarg, ' ', arg->width - n_extra - n, nbytes);
if (rc)
return rc;
}
if (signchar)
{
rc = outfnc (outfncarg, &signchar, 1);
if (rc)
return rc;
*nbytes += 1;
}
rc = outfnc (outfncarg, p, pend - p);
if (rc)
return rc;
*nbytes += pend - p;
if ((arg->flags & FLAG_LEFT_JUST)
&& arg->width >= 0 && arg->width - n_extra > n)
{
rc = pad_out (outfnc, outfncarg, ' ', arg->width - n_extra - n, nbytes);
if (rc)
return rc;
}
return 0;
}
/* "c" formatting. */
static int
pr_char (estream_printf_out_t outfnc, void *outfncarg,
argspec_t arg, value_t value, size_t *nbytes)
{
int rc;
char buf[1];
if (arg->vt != VALTYPE_INT)
return -1;
buf[0] = (unsigned int)value.a_int;
rc = outfnc (outfncarg, buf, 1);
if(rc)
return rc;
*nbytes += 1;
return 0;
}
/* "s" formatting. */
static int
pr_string (estream_printf_out_t outfnc, void *outfncarg,
argspec_t arg, value_t value, size_t *nbytes)
{
int rc;
size_t n;
const char *string, *s;
if (arg->vt != VALTYPE_STRING)
return -1;
string = value.a_string;
if (!string)
string = "(null)";
if (arg->precision >= 0)
{
/* Test for nul after N so that we can pass a non-nul terminated
string. */
for (n=0,s=string; n < arg->precision && *s; s++)
n++;
}
else
n = strlen (string);
if (!(arg->flags & FLAG_LEFT_JUST)
&& arg->width >= 0 && arg->width > n )
{
rc = pad_out (outfnc, outfncarg, ' ', arg->width - n, nbytes);
if (rc)
return rc;
}
rc = outfnc (outfncarg, string, n);
if (rc)
return rc;
*nbytes += n;
if ((arg->flags & FLAG_LEFT_JUST)
&& arg->width >= 0 && arg->width > n)
{
rc = pad_out (outfnc, outfncarg, ' ', arg->width - n, nbytes);
if (rc)
return rc;
}
return 0;
}
/* "p" formatting. */
static int
pr_pointer (estream_printf_out_t outfnc, void *outfncarg,
argspec_t arg, value_t value, size_t *nbytes)
{
int rc;
#if defined(HAVE_LONG_LONG_INT) && (SIZEOF_UNSIGNED_LONG < SIZEOF_VOID_P)
unsigned long long aulong;
#else
unsigned long aulong;
#endif
char numbuf[100];
char *p, *pend;
if (arg->vt != VALTYPE_POINTER)
return -1;
/* We assume that a pointer can be converted to an unsigned long.
That is not correct for a 64 bit Windows, but then we assume that
long long is supported and usable for storing a pointer. */
#if defined(HAVE_LONG_LONG_INT) && (SIZEOF_UNSIGNED_LONG < SIZEOF_VOID_P)
aulong = (unsigned long long)value.a_void_ptr;
#else
aulong = (unsigned long)value.a_void_ptr;
#endif
p = pend = numbuf + DIM(numbuf);
do
{
*--p = "0123456789abcdefx"[(aulong % 16)];
aulong /= 16;
}
while (aulong);
while ((pend-p) < 2*sizeof (aulong))
*--p = '0';
*--p = 'x';
*--p = '0';
rc = outfnc (outfncarg, p, pend - p);
if (rc)
return rc;
*nbytes += pend - p;
return 0;
}
/* "n" pesudo format operation. */
static int
pr_bytes_so_far (estream_printf_out_t outfnc, void *outfncarg,
argspec_t arg, value_t value, size_t *nbytes)
{
(void)outfnc;
(void)outfncarg;
switch (arg->vt)
{
case VALTYPE_SCHAR_PTR:
*value.a_schar_ptr = (signed char)(unsigned int)(*nbytes);
break;
case VALTYPE_SHORT_PTR:
*value.a_short_ptr = (short)(unsigned int)(*nbytes);
break;
case VALTYPE_LONG_PTR:
*value.a_long_ptr = (long)(*nbytes);
break;
#ifdef HAVE_LONG_LONG_INT
case VALTYPE_LONGLONG_PTR:
*value.a_longlong_ptr = (long long)(*nbytes);
break;
#endif
#ifdef HAVE_INTMAX_T
case VALTYPE_INTMAX_PTR:
*value.a_intmax_ptr = (intmax_t)(*nbytes);
break;
#endif
case VALTYPE_SIZE_PTR:
*value.a_size_ptr = (*nbytes);
break;
#ifdef HAVE_PTRDIFF_T
case VALTYPE_PTRDIFF_PTR:
*value.a_ptrdiff_ptr = (ptrdiff_t)(*nbytes);
break;
#endif
case VALTYPE_INT_PTR:
*value.a_int_ptr = (int)(*nbytes);
break;
default:
return -1; /* An unsupported type has been used. */
}
return 0;
}
/* Run the actual formatting. OUTFNC and OUTFNCARG are the output
functions. FORMAT is format string ARGSPECS is the parsed format
string, ARGSPECS_LEN the number of items in ARGSPECS. VALUETABLE
holds the values and may be directly addressed using the position
arguments given by ARGSPECS. MYERRNO is used for the "%m"
conversion. NBYTES well be updated to reflect the number of bytes
send to the output function. */
static int
do_format (estream_printf_out_t outfnc, void *outfncarg,
const char *format, argspec_t argspecs, size_t argspecs_len,
valueitem_t valuetable, int myerrno, size_t *nbytes)
{
int rc = 0;
const char *s;
argspec_t arg = argspecs;
int argidx = 0; /* Only used for assertion. */
size_t n;
value_t value;
s = format;
while ( *s )
{
if (*s != '%')
{
s++;
continue;
}
if (s != format)
{
rc = outfnc (outfncarg, format, (n=s-format));
if (rc)
return rc;
*nbytes += n;
}
if (s[1] == '%')
{
/* Note that this code ignores one trailing percent escape -
this is however okay as the args parser must have
detected this already. */
rc = outfnc (outfncarg, s, 1);
if (rc)
return rc;
*nbytes += 1;
s += 2;
format = s;
continue;
}
/* Save the next start. */
s += arg->length;
format = s;
assert (argidx < argspecs_len);
argidx++;
/* Apply indirect field width and precision values. */
if (arg->width == STAR_FIELD_VALUE)
{
assert (valuetable[arg->width_pos-1].vt == VALTYPE_INT);
arg->width = valuetable[arg->width_pos-1].value.a_int;
if (arg->width < 0)
{
arg->width = -arg->width;
arg->flags |= FLAG_LEFT_JUST;
}
}
if (arg->precision == STAR_FIELD_VALUE)
{
assert (valuetable[arg->precision_pos-1].vt == VALTYPE_INT);
arg->precision = valuetable[arg->precision_pos-1].value.a_int;
if (arg->precision < 0)
arg->precision = NO_FIELD_VALUE;
}
if (arg->arg_pos == -1 && arg->conspec == CONSPEC_STRERROR)
value.a_string = strerror (myerrno);
else
{
assert (arg->vt == valuetable[arg->arg_pos-1].vt);
value = valuetable[arg->arg_pos-1].value;
}
switch (arg->conspec)
{
case CONSPEC_UNKNOWN: assert (!"bug"); break;
case CONSPEC_DECIMAL:
case CONSPEC_UNSIGNED:
case CONSPEC_OCTAL:
case CONSPEC_HEX:
case CONSPEC_HEX_UP:
rc = pr_integer (outfnc, outfncarg, arg, value, nbytes);
break;
case CONSPEC_FLOAT:
case CONSPEC_FLOAT_UP:
case CONSPEC_EXP:
case CONSPEC_EXP_UP:
case CONSPEC_F_OR_G:
case CONSPEC_F_OR_G_UP:
case CONSPEC_HEX_EXP:
case CONSPEC_HEX_EXP_UP:
rc = pr_float (outfnc, outfncarg, arg, value, nbytes);
break;
case CONSPEC_CHAR:
rc = pr_char (outfnc, outfncarg, arg, value, nbytes);
break;
case CONSPEC_STRING:
case CONSPEC_STRERROR:
rc = pr_string (outfnc, outfncarg, arg, value, nbytes);
break;
case CONSPEC_POINTER:
rc = pr_pointer (outfnc, outfncarg, arg, value, nbytes);
break;
case CONSPEC_BYTES_SO_FAR:
rc = pr_bytes_so_far (outfnc, outfncarg, arg, value, nbytes);
break;
}
if (rc)
return rc;
arg++;
}
/* Print out any trailing stuff. */
n = s - format;
rc = n? outfnc (outfncarg, format, n) : 0;
if (!rc)
*nbytes += n;
return rc;
}
/* The versatile printf formatting routine. It expects a callback
function OUTFNC and an opaque argument OUTFNCARG used for actual
output of the formatted stuff. FORMAT is the format specification
and VAARGS a variable argumemt list matching the arguments of
FORMAT. */
int
_gpgrt_estream_format (estream_printf_out_t outfnc,
void *outfncarg,
const char *format, va_list vaargs)
{
/* Buffer to hold the argspecs and a pointer to it.*/
struct argspec_s argspecs_buffer[DEFAULT_MAX_ARGSPECS];
argspec_t argspecs = argspecs_buffer;
size_t argspecs_len; /* Number of specifications in ARGSPECS. */
/* Buffer to hold the description for the values. */
struct valueitem_s valuetable_buffer[DEFAULT_MAX_VALUES];
valueitem_t valuetable = valuetable_buffer;
int rc; /* Return code. */
size_t argidx; /* Used to index the argspecs array. */
size_t validx; /* Used to index the valuetable. */
int max_pos;/* Highest argument position. */
size_t nbytes = 0; /* Keep track of the number of bytes passed to
the output function. */
int myerrno = errno; /* Save the errno for use with "%m". */
/* Parse the arguments to come up with descriptive list. We can't
do this on the fly because we need to support positional
arguments. */
rc = parse_format (format, &argspecs, DIM(argspecs_buffer), &argspecs_len);
if (rc)
goto leave;
/* Check that all ARG_POS fields are set. */
for (argidx=0,max_pos=0; argidx < argspecs_len; argidx++)
{
if (argspecs[argidx].arg_pos != -1
&& argspecs[argidx].arg_pos > max_pos)
max_pos = argspecs[argidx].arg_pos;
if (argspecs[argidx].width_pos > max_pos)
max_pos = argspecs[argidx].width_pos;
if (argspecs[argidx].precision_pos > max_pos)
max_pos = argspecs[argidx].precision_pos;
}
if (!max_pos)
{
/* Fill in all the positions. */
for (argidx=0; argidx < argspecs_len; argidx++)
{
if (argspecs[argidx].width == STAR_FIELD_VALUE)
argspecs[argidx].width_pos = ++max_pos;
if (argspecs[argidx].precision == STAR_FIELD_VALUE)
argspecs[argidx].precision_pos = ++max_pos;
if (argspecs[argidx].arg_pos != -1 )
argspecs[argidx].arg_pos = ++max_pos;
}
}
else
{
/* Check that they are all filled. More test are done later. */
for (argidx=0; argidx < argspecs_len; argidx++)
{
if (!argspecs[argidx].arg_pos
|| (argspecs[argidx].width == STAR_FIELD_VALUE
&& !argspecs[argidx].width_pos)
|| (argspecs[argidx].precision == STAR_FIELD_VALUE
&& !argspecs[argidx].precision_pos))
goto leave_einval;
}
}
/* Check that there is no overflow in max_pos and that it has a
reasonable length. There may never be more elements than the
number of characters in FORMAT. */
if (max_pos < 0 || max_pos >= strlen (format))
goto leave_einval;
#ifdef DEBUG
dump_argspecs (argspecs, argspecs_len);
#endif
/* Allocate a table to hold the values. If it is small enough we
use a stack allocated buffer. */
if (max_pos > DIM(valuetable_buffer))
{
valuetable = calloc (max_pos, sizeof *valuetable);
if (!valuetable)
goto leave_error;
}
else
{
for (validx=0; validx < DIM(valuetable_buffer); validx++)
valuetable[validx].vt = VALTYPE_UNSUPPORTED;
}
for (argidx=0; argidx < argspecs_len; argidx++)
{
if (argspecs[argidx].arg_pos != - 1)
{
validx = argspecs[argidx].arg_pos - 1;
if (valuetable[validx].vt)
goto leave_einval; /* Already defined. */
valuetable[validx].vt = argspecs[argidx].vt;
}
if (argspecs[argidx].width == STAR_FIELD_VALUE)
{
validx = argspecs[argidx].width_pos - 1;
if (valuetable[validx].vt)
goto leave_einval; /* Already defined. */
valuetable[validx].vt = VALTYPE_INT;
}
if (argspecs[argidx].precision == STAR_FIELD_VALUE)
{
validx = argspecs[argidx].precision_pos - 1;
if (valuetable[validx].vt)
goto leave_einval; /* Already defined. */
valuetable[validx].vt = VALTYPE_INT;
}
}
/* Read all the arguments. This will error out for unsupported
types and for not given positional arguments. */
rc = read_values (valuetable, max_pos, vaargs);
if (rc)
goto leave_einval;
/* for (validx=0; validx < max_pos; validx++) */
/* fprintf (stderr, "%2d: vt=%d\n", validx, valuetable[validx].vt); */
/* Everything has been collected, go ahead with the formatting. */
rc = do_format (outfnc, outfncarg, format,
argspecs, argspecs_len, valuetable, myerrno, &nbytes);
goto leave;
leave_einval:
_set_errno (EINVAL);
leave_error:
rc = -1;
leave:
if (valuetable != valuetable_buffer)
free (valuetable);
if (argspecs != argspecs_buffer)
free (argspecs);
return rc;
}
/* A simple output handler utilizing stdio. */
static int
plain_stdio_out (void *outfncarg, const char *buf, size_t buflen)
{
FILE *fp = (FILE*)outfncarg;
if ( fwrite (buf, buflen, 1, fp) != 1 )
return -1;
return 0;
}
/* A replacement for printf. */
int
_gpgrt_estream_printf (const char *format, ...)
{
int rc;
va_list arg_ptr;
va_start (arg_ptr, format);
rc = _gpgrt_estream_format (plain_stdio_out, stderr, format, arg_ptr);
va_end (arg_ptr);
return rc;
}
/* A replacement for fprintf. */
int
_gpgrt_estream_fprintf (FILE *fp, const char *format, ...)
{
int rc;
va_list arg_ptr;
va_start (arg_ptr, format);
rc = _gpgrt_estream_format (plain_stdio_out, fp, format, arg_ptr);
va_end (arg_ptr);
return rc;
}
/* A replacement for vfprintf. */
int
_gpgrt_estream_vfprintf (FILE *fp, const char *format, va_list arg_ptr)
{
return _gpgrt_estream_format (plain_stdio_out, fp, format, arg_ptr);
}
/* Communication object used between estream_snprintf and
fixed_buffer_out. */
struct fixed_buffer_parm_s
{
size_t size; /* Size of the buffer. */
size_t count; /* Number of bytes requested for output. */
size_t used; /* Used size of the buffer. */
char *buffer; /* Provided buffer. */
};
/* A simple malloced buffer output handler. */
static int
fixed_buffer_out (void *outfncarg, const char *buf, size_t buflen)
{
struct fixed_buffer_parm_s *parm = outfncarg;
parm->count += buflen;
if (!parm->buffer)
;
else if (parm->used + buflen < parm->size)
{
/* Handle the common case that everything fits into the buffer
separately. */
memcpy (parm->buffer + parm->used, buf, buflen);
parm->used += buflen;
}
else
{
/* The slow version of above. */
for ( ;buflen && parm->used < parm->size; buflen--)
parm->buffer[parm->used++] = *buf++;
}
return 0;
}
/* A replacement for vsnprintf. */
int
_gpgrt_estream_vsnprintf (char *buf, size_t bufsize,
const char *format, va_list arg_ptr)
{
struct fixed_buffer_parm_s parm;
int rc;
parm.size = bufsize;
parm.count = 0;
parm.used = 0;
parm.buffer = bufsize?buf:NULL;
rc = _gpgrt_estream_format (fixed_buffer_out, &parm, format, arg_ptr);
if (!rc)
rc = fixed_buffer_out (&parm, "", 1); /* Print terminating Nul. */
if (rc == -1)
return -1;
if (bufsize && buf && parm.size && parm.count >= parm.size)
buf[parm.size-1] = 0;
parm.count--; /* Do not count the trailing nul. */
return (int)parm.count; /* Return number of bytes which would have
been written. */
}
/* A replacement for snprintf. */
int
_gpgrt_estream_snprintf (char *buf, size_t bufsize, const char *format, ...)
{
int rc;
va_list arg_ptr;
va_start (arg_ptr, format);
rc = _gpgrt_estream_vsnprintf (buf, bufsize, format, arg_ptr);
va_end (arg_ptr);
return rc;
}
/* Communication object used between estream_asprintf and
dynamic_buffer_out. */
struct dynamic_buffer_parm_s
{
int error_flag; /* Internal helper. */
size_t alloced; /* Allocated size of the buffer. */
size_t used; /* Used size of the buffer. */
char *buffer; /* Malloced buffer. */
};
/* A simple malloced buffer output handler. */
static int
dynamic_buffer_out (void *outfncarg, const char *buf, size_t buflen)
{
struct dynamic_buffer_parm_s *parm = outfncarg;
if (parm->error_flag)
{
/* Just in case some formatting routine did not checked for an
error. */
_set_errno (parm->error_flag);
return -1;
}
if (parm->used + buflen >= parm->alloced)
{
char *p;
parm->alloced += buflen + 512;
p = my_printf_realloc (parm->buffer, parm->alloced);
if (!p)
{
parm->error_flag = errno ? errno : ENOMEM;
/* Wipe out what we already accumulated. This is useful in
case sensitive data is formated. */
memset (parm->buffer, 0, parm->used);
return -1;
}
parm->buffer = p;
}
memcpy (parm->buffer + parm->used, buf, buflen);
parm->used += buflen;
return 0;
}
/* A replacement for vasprintf. As with the BSD version of vasprintf
-1 will be returned on error and NULL stored at BUFP. On success
the number of bytes printed will be returned. */
int
_gpgrt_estream_vasprintf (char **bufp, const char *format, va_list arg_ptr)
{
struct dynamic_buffer_parm_s parm;
int rc;
parm.error_flag = 0;
parm.alloced = 512;
parm.used = 0;
parm.buffer = my_printf_realloc (NULL, parm.alloced);
if (!parm.buffer)
{
*bufp = NULL;
return -1;
}
rc = _gpgrt_estream_format (dynamic_buffer_out, &parm, format, arg_ptr);
if (!rc)
rc = dynamic_buffer_out (&parm, "", 1); /* Print terminating Nul. */
/* Fixme: Should we shrink the resulting buffer? */
if (rc != -1 && parm.error_flag)
{
rc = -1;
_set_errno (parm.error_flag);
}
if (rc == -1)
{
memset (parm.buffer, 0, parm.used);
if (parm.buffer)
my_printf_realloc (parm.buffer, 0);
*bufp = NULL;
return -1;
}
assert (parm.used); /* We have at least the terminating Nul. */
*bufp = parm.buffer;
return parm.used - 1; /* Do not include that Nul. */
}
/* A replacement for asprintf. As with the BSD of asprintf version -1
will be returned on error and NULL stored at BUFP. On success the
number of bytes printed will be returned. */
int
_gpgrt_estream_asprintf (char **bufp, const char *format, ...)
{
int rc;
va_list arg_ptr;
va_start (arg_ptr, format);
rc = _gpgrt_estream_vasprintf (bufp, format, arg_ptr);
va_end (arg_ptr);
return rc;
}
/* A variant of asprintf. The function returns the allocated buffer
or NULL on error; ERRNO is set in the error case. The caller
should use es_free to release the buffer. This function actually
belongs into estream-printf but we put it here as a convenience
and because es_free is required anyway. */
char *
_gpgrt_estream_bsprintf (const char *format, ...)
{
int rc;
va_list ap;
char *buf;
va_start (ap, format);
rc = _gpgrt_estream_vasprintf (&buf, format, ap);
va_end (ap);
if (rc < 0)
return NULL;
return buf;
}
diff --git a/src/estream-printf.h b/src/estream-printf.h
index e82d8fb..ae303a7 100644
--- a/src/estream-printf.h
+++ b/src/estream-printf.h
@@ -1,151 +1,151 @@
/* estream-printf.h - Versatile mostly C-99 compliant printf formatting.
* Copyright (C) 2007, 2010, 2012 g10 Code GmbH
*
* This file is part of Libestream.
*
* Libestream is free software; you can redistribute it and/or modify
* it under the terms of the GNU Lesser General Public License as
* published by the Free Software Foundation; either version 2.1 of
* the License, or (at your option) any later version.
*
* Libestream is distributed in the hope that it will be useful, but
* WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
- * License along with Libestream; if not, see .
+ * License along with Libestream; if not, see .
*
* ALTERNATIVELY, Libestream may be distributed under the terms of the
* following license, in which case the provisions of this license are
* required INSTEAD OF the GNU General Public License. If you wish to
* allow use of your version of this file only under the terms of the
* GNU General Public License, and not to allow others to use your
* version of this file under the terms of the following license,
* indicate your decision by deleting this paragraph and the license
* below.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, and the entire permission notice in its entirety,
* including the disclaimer of warranties.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. The name of the author may not be used to endorse or promote
* products derived from this software without specific prior
* written permission.
*
* THIS SOFTWARE IS PROVIDED "AS IS" AND ANY EXPRESS OR IMPLIED
* WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
* OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
* DISCLAIMED. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT,
* INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
* (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
* SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
* STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED
* OF THE POSSIBILITY OF SUCH DAMAGE.
*/
#ifndef ESTREAM_PRINTF_H
#define ESTREAM_PRINTF_H
#include
#include
/* To use this file with libraries the following macro is useful:
#define _ESTREAM_EXT_SYM_PREFIX _foo_
This prefixes all external symbols with "_foo_".
For the implementation of the code (estream-printf.c) the following
macros may be used to tune the implementation for certain systems:
#define _ESTREAM_PRINTF_REALLOC foo_realloc
Make estream_asprintf and estream_vasprintf use foo_realloc
instead of the standard realloc to allocate memory returned to
the caller. Note that foo_realloc needs to be C-90 compliant:
foo_realloc (NULL,n) is the same as a call to malloc(n) and
foo_realloc (a, 0) is the same as a call to free (a).
#define _ESTREAM_PRINTF_EXTRA_INCLUDE "foo.h"
This includes the file "foo.h" which may provide prototypes for
the custom memory allocation functions.
*/
#ifdef _ESTREAM_EXT_SYM_PREFIX
#ifndef _ESTREAM_PREFIX
#define _ESTREAM_PREFIX1(x,y) x ## y
#define _ESTREAM_PREFIX2(x,y) _ESTREAM_PREFIX1(x,y)
#define _ESTREAM_PREFIX(x) _ESTREAM_PREFIX2(_ESTREAM_EXT_SYM_PREFIX,x)
#endif /*_ESTREAM_PREFIX*/
#define estream_printf_out_t _ESTREAM_PREFIX(estream_printf_out_t)
#define estream_format _ESTREAM_PREFIX(estream_format)
#define estream_printf _ESTREAM_PREFIX(estream_printf)
#define estream_fprintf _ESTREAM_PREFIX(estream_fprintf)
#define estream_vfprintf _ESTREAM_PREFIX(estream_vfprintf)
#define estream_snprintf _ESTREAM_PREFIX(estream_snprintf)
#define estream_vsnprintf _ESTREAM_PREFIX(estream_vsnprintf)
#define estream_asprintf _ESTREAM_PREFIX(estream_asprintf)
#define estream_vasprintf _ESTREAM_PREFIX(estream_vasprintf)
#endif /*_ESTREAM_EXT_SYM_PREFIX*/
#ifndef _ESTREAM_GCC_A_PRINTF
# if __GNUC__ > 4 || (__GNUC__ == 4 && __GNUC_MINOR__ >= 4 )
# define _ESTREAM_GCC_A_PRINTF( f, a ) \
__attribute__ ((format (__gnu_printf__,f,a)))
# elif __GNUC__ > 2 || (__GNUC__ == 2 && __GNUC_MINOR__ >= 5 )
# define _ESTREAM_GCC_A_PRINTF( f, a ) \
__attribute__ ((format (printf,f,a)))
# else
# define _ESTREAM_GCC_A_PRINTF( f, a )
# endif
#endif /*_ESTREAM_GCC_A_PRINTF*/
#ifdef __cplusplus
extern "C"
{
#if 0
}
#endif
#endif
typedef int (*estream_printf_out_t)
(void *outfncarg, const char *buf, size_t buflen);
int _gpgrt_estream_format (estream_printf_out_t outfnc, void *outfncarg,
const char *format, va_list vaargs)
_ESTREAM_GCC_A_PRINTF(3,0);
int _gpgrt_estream_printf (const char *format, ...)
_ESTREAM_GCC_A_PRINTF(1,2);
int _gpgrt_estream_fprintf (FILE *fp, const char *format, ... )
_ESTREAM_GCC_A_PRINTF(2,3);
int _gpgrt_estream_vfprintf (FILE *fp, const char *format, va_list arg_ptr)
_ESTREAM_GCC_A_PRINTF(2,0);
int _gpgrt_estream_snprintf (char *buf, size_t bufsize, const char *format, ...)
_ESTREAM_GCC_A_PRINTF(3,4);
int _gpgrt_estream_vsnprintf (char *buf,size_t bufsize,
const char *format, va_list arg_ptr)
_ESTREAM_GCC_A_PRINTF(3,0);
int _gpgrt_estream_asprintf (char **bufp, const char *format, ...)
_ESTREAM_GCC_A_PRINTF(2,3);
int _gpgrt_estream_vasprintf (char **bufp, const char *format, va_list arg_ptr)
_ESTREAM_GCC_A_PRINTF(2,0);
char *_gpgrt_estream_bsprintf (const char *format, ...)
_ESTREAM_GCC_A_PRINTF(1,2);
#ifdef __cplusplus
}
#endif
#endif /*ESTREAM_PRINTF_H*/
diff --git a/src/estream.c b/src/estream.c
index 3a89868..abce0bf 100644
--- a/src/estream.c
+++ b/src/estream.c
@@ -1,4865 +1,4865 @@
/* estream.c - Extended Stream I/O Library
* Copyright (C) 2004, 2005, 2006, 2007, 2009, 2010, 2011,
* 2014, 2015 g10 Code GmbH
*
* This file is part of Libestream.
*
* Libestream is free software; you can redistribute it and/or modify
* it under the terms of the GNU Lesser General Public License as
* published by the Free Software Foundation; either version 2.1 of
* the License, or (at your option) any later version.
*
* Libestream is distributed in the hope that it will be useful, but
* WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
- * License along with Libestream; if not, see .
+ * License along with Libestream; if not, see .
*
* ALTERNATIVELY, Libestream may be distributed under the terms of the
* following license, in which case the provisions of this license are
* required INSTEAD OF the GNU General Public License. If you wish to
* allow use of your version of this file only under the terms of the
* GNU General Public License, and not to allow others to use your
* version of this file under the terms of the following license,
* indicate your decision by deleting this paragraph and the license
* below.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, and the entire permission notice in its entirety,
* including the disclaimer of warranties.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. The name of the author may not be used to endorse or promote
* products derived from this software without specific prior
* written permission.
*
* THIS SOFTWARE IS PROVIDED "AS IS" AND ANY EXPRESS OR IMPLIED
* WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
* OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
* DISCLAIMED. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT,
* INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
* (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
* SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
* STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED
* OF THE POSSIBILITY OF SUCH DAMAGE.
*/
#ifdef USE_ESTREAM_SUPPORT_H
# include
#endif
#ifdef HAVE_CONFIG_H
# include
#endif
#if defined(_WIN32) && !defined(HAVE_W32_SYSTEM)
# define HAVE_W32_SYSTEM 1
# if defined(__MINGW32CE__) && !defined (HAVE_W32CE_SYSTEM)
# define HAVE_W32CE_SYSTEM
# endif
#endif
#ifdef HAVE_SYS_SELECT_H
# include
#endif
#ifdef HAVE_SYS_TIME_H
# include
#endif
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#ifdef HAVE_W32_SYSTEM
# ifdef HAVE_WINSOCK2_H
# include
# endif
# include
#endif
#include "gpgrt-int.h"
#include "estream-printf.h"
#ifndef O_BINARY
# define O_BINARY 0
#endif
#ifndef HAVE_DOSISH_SYSTEM
# ifdef HAVE_W32_SYSTEM
# define HAVE_DOSISH_SYSTEM 1
# endif
#endif
#ifdef HAVE_W32_SYSTEM
# define S_IRGRP S_IRUSR
# define S_IROTH S_IRUSR
# define S_IWGRP S_IWUSR
# define S_IWOTH S_IWUSR
# define S_IXGRP S_IXUSR
# define S_IXOTH S_IXUSR
# define O_NONBLOCK 0 /* FIXME: Not yet supported. */
#endif
#ifndef EAGAIN
# define EAGAIN EWOULDBLOCK
#endif
#ifdef HAVE_W32CE_SYSTEM
# define _set_errno(a) gpg_err_set_errno ((a))
/* Setmode is missing in cegcc but available since CE 5.0. */
int _setmode (int handle, int mode);
# define setmode(a,b) _setmode ((a),(b))
#else
# define _set_errno(a) do { errno = (a); } while (0)
#endif
#ifdef HAVE_W32_SYSTEM
# define IS_INVALID_FD(a) ((void*)(a) == (void*)(-1)) /* ?? FIXME. */
#else
# define IS_INVALID_FD(a) ((a) == -1)
#endif
/* Generally used types. */
typedef void *(*func_realloc_t) (void *mem, size_t size);
typedef void (*func_free_t) (void *mem);
/* Buffer management layer. */
#define BUFFER_BLOCK_SIZE BUFSIZ
#define BUFFER_UNREAD_SIZE 16
/* A linked list to hold notification functions. */
struct notify_list_s
{
struct notify_list_s *next;
void (*fnc) (estream_t, void*); /* The notification function. */
void *fnc_value; /* The value to be passed to FNC. */
};
typedef struct notify_list_s *notify_list_t;
/* A private cookie function to implement an internal IOCTL
service. */
typedef int (*cookie_ioctl_function_t) (void *cookie, int cmd,
void *ptr, size_t *len);
/* IOCTL commands for the private cookie function. */
#define COOKIE_IOCTL_SNATCH_BUFFER 1
#define COOKIE_IOCTL_NONBLOCK 2
/* The internal stream object. */
struct _gpgrt_stream_internal
{
unsigned char buffer[BUFFER_BLOCK_SIZE];
unsigned char unread_buffer[BUFFER_UNREAD_SIZE];
gpgrt_lock_t lock; /* Lock. */
void *cookie; /* Cookie. */
void *opaque; /* Opaque data. */
unsigned int modeflags; /* Flags for the backend. */
char *printable_fname; /* Malloced filename for es_fname_get. */
gpgrt_off_t offset;
gpgrt_cookie_read_function_t func_read;
gpgrt_cookie_write_function_t func_write;
gpgrt_cookie_seek_function_t func_seek;
gpgrt_cookie_close_function_t func_close;
cookie_ioctl_function_t func_ioctl;
int strategy;
es_syshd_t syshd; /* A copy of the system handle. */
struct
{
unsigned int err: 1;
unsigned int eof: 1;
unsigned int hup: 1;
} indicators;
unsigned int deallocate_buffer: 1;
unsigned int is_stdstream:1; /* This is a standard stream. */
unsigned int stdstream_fd:2; /* 0, 1 or 2 for a standard stream. */
unsigned int printable_fname_inuse: 1; /* es_fname_get has been used. */
unsigned int samethread: 1; /* The "samethread" mode keyword. */
size_t print_ntotal; /* Bytes written from in print_writer. */
notify_list_t onclose; /* On close notify function list. */
};
typedef struct _gpgrt_stream_internal *estream_internal_t;
/* A linked list to hold active stream objects. */
struct estream_list_s
{
struct estream_list_s *next;
estream_t stream; /* Entry is not used if NULL. */
};
typedef struct estream_list_s *estream_list_t;
static estream_list_t estream_list;
/* A lock object for the estream list and the custom_std_fds array. */
GPGRT_LOCK_DEFINE (estream_list_lock);
/* File descriptors registered to be used as the standard file handles. */
static int custom_std_fds[3];
static unsigned char custom_std_fds_valid[3];
/* Functions called before and after blocking syscalls. */
static void (*pre_syscall_func)(void);
static void (*post_syscall_func)(void);
/* Error code replacements. */
#ifndef EOPNOTSUPP
# define EOPNOTSUPP ENOSYS
#endif
/* Local prototypes. */
static void fname_set_internal (estream_t stream, const char *fname, int quote);
/* Macros. */
/* Calculate array dimension. */
#ifndef DIM
#define DIM(array) (sizeof (array) / sizeof (*array))
#endif
#define tohex(n) ((n) < 10 ? ((n) + '0') : (((n) - 10) + 'A'))
/* Evaluate EXPRESSION, setting VARIABLE to the return code, if
VARIABLE is zero. */
#define SET_UNLESS_NONZERO(variable, tmp_variable, expression) \
do \
{ \
tmp_variable = expression; \
if ((! variable) && tmp_variable) \
variable = tmp_variable; \
} \
while (0)
static void *
mem_alloc (size_t n)
{
return _gpgrt_malloc (n);
}
static void *
mem_realloc (void *p, size_t n)
{
return _gpgrt_realloc (p, n);
}
static void
mem_free (void *p)
{
if (p)
_gpgrt_free (p);
}
#ifdef HAVE_W32_SYSTEM
static int
map_w32_to_errno (DWORD w32_err)
{
switch (w32_err)
{
case 0:
return 0;
case ERROR_FILE_NOT_FOUND:
return ENOENT;
case ERROR_PATH_NOT_FOUND:
return ENOENT;
case ERROR_ACCESS_DENIED:
return EPERM;
case ERROR_INVALID_HANDLE:
case ERROR_INVALID_BLOCK:
return EINVAL;
case ERROR_NOT_ENOUGH_MEMORY:
return ENOMEM;
case ERROR_NO_DATA:
return EPIPE;
default:
return EIO;
}
}
#endif /*HAVE_W32_SYSTEM*/
/* Replacement fucntions. */
#ifndef HAVE_MEMRCHR
static void *
memrchr (const void *buffer, int c, size_t n)
{
const unsigned char *p = buffer;
for (p += n; n ; n--)
if (*--p == c)
return (void *)p;
return NULL;
}
#endif /*HAVE_MEMRCHR*/
/*
* Lock wrappers
*/
#if 0
# define dbg_lock_0(f) fprintf (stderr, "estream: " f);
# define dbg_lock_1(f, a) fprintf (stderr, "estream: " f, (a));
# define dbg_lock_2(f, a, b) fprintf (stderr, "estream: " f, (a), (b));
#else
# define dbg_lock_0(f)
# define dbg_lock_1(f, a)
# define dbg_lock_2(f, a, b)
#endif
static int
init_stream_lock (estream_t _GPGRT__RESTRICT stream)
{
int rc;
if (!stream->intern->samethread)
{
dbg_lock_1 ("enter init_stream_lock for %p\n", stream);
memset (&stream->intern->lock, 0 , sizeof stream->intern->lock);
rc = _gpgrt_lock_init (&stream->intern->lock);
dbg_lock_2 ("leave init_stream_lock for %p: rc=%d\n", stream, rc);
}
else
rc = 0;
return rc;
}
static void
destroy_stream_lock (estream_t _GPGRT__RESTRICT stream)
{
if (!stream->intern->samethread)
{
dbg_lock_1 ("enter destroy_stream_lock for %p\n", stream);
_gpgrt_lock_destroy (&stream->intern->lock);
dbg_lock_1 ("leave destroy_stream_lock for %p\n", stream);
}
}
static void
lock_stream (estream_t _GPGRT__RESTRICT stream)
{
if (!stream->intern->samethread)
{
dbg_lock_1 ("enter lock_stream for %p\n", stream);
_gpgrt_lock_lock (&stream->intern->lock);
dbg_lock_1 ("leave lock_stream for %p\n", stream);
}
}
static int
trylock_stream (estream_t _GPGRT__RESTRICT stream)
{
int rc;
if (!stream->intern->samethread)
{
dbg_lock_1 ("enter trylock_stream for %p\n", stream);
rc = _gpgrt_lock_trylock (&stream->intern->lock)? 0 : -1;
dbg_lock_2 ("leave trylock_stream for %p: rc=%d\n", stream, rc);
}
else
rc = 0;
return rc;
}
static void
unlock_stream (estream_t _GPGRT__RESTRICT stream)
{
if (!stream->intern->samethread)
{
dbg_lock_1 ("enter unlock_stream for %p\n", stream);
_gpgrt_lock_unlock (&stream->intern->lock);
dbg_lock_1 ("leave unlock_stream for %p\n", stream);
}
}
static void
lock_list (void)
{
dbg_lock_0 ("enter lock_list\n");
_gpgrt_lock_lock (&estream_list_lock);
dbg_lock_0 ("leave lock_list\n");
}
static void
unlock_list (void)
{
dbg_lock_0 ("enter unlock_list\n");
_gpgrt_lock_unlock (&estream_list_lock);
dbg_lock_0 ("leave unlock_list\n");
}
#undef dbg_lock_0
#undef dbg_lock_1
#undef dbg_lock_2
/*
* List manipulation.
*/
/* Add STREAM to the list of registered stream objects. If
WITH_LOCKED_LIST is true it is assumed that the list of streams is
already locked. The implementation is straightforward: We first
look for an unused entry in the list and use that; if none is
available we put a new item at the head. We drawback of the
strategy never to shorten the list is that a one time allocation of
many streams will lead to scanning unused entries later. If that
turns out to be a problem, we may either free some items from the
list or append new entries at the end; or use a table. Returns 0
on success; on error or non-zero is returned and ERRNO set. */
static int
do_list_add (estream_t stream, int with_locked_list)
{
estream_list_t item;
if (!with_locked_list)
lock_list ();
for (item = estream_list; item && item->stream; item = item->next)
;
if (!item)
{
item = mem_alloc (sizeof *item);
if (item)
{
item->next = estream_list;
estream_list = item;
}
}
if (item)
item->stream = stream;
if (!with_locked_list)
unlock_list ();
return item? 0 : -1;
}
/* Remove STREAM from the list of registered stream objects. */
static void
do_list_remove (estream_t stream, int with_locked_list)
{
estream_list_t item;
if (!with_locked_list)
lock_list ();
for (item = estream_list; item; item = item->next)
if (item->stream == stream)
{
item->stream = NULL;
break;
}
if (!with_locked_list)
unlock_list ();
}
static void
do_deinit (void)
{
/* Flush all streams. */
_gpgrt_fflush (NULL);
/* We should release the estream_list. However there is one
problem: That list is also used to search for the standard
estream file descriptors. If we would remove the entire list,
any use of es_foo in another atexit function may re-create the
list and the streams with possible undesirable effects. Given
that we don't close the stream either, it should not matter that
we keep the list and let the OS clean it up at process end. */
/* Reset the syscall clamp. */
pre_syscall_func = NULL;
post_syscall_func = NULL;
}
/*
* Initialization.
*/
int
_gpgrt_es_init (void)
{
static int initialized;
if (!initialized)
{
initialized = 1;
atexit (do_deinit);
}
return 0;
}
/* Register the syscall clamp. These two functions are called
immediately before and after a possible blocking system call. This
should be used before any I/O happens. The function is commonly
used with the nPth library:
gpgrt_set_syscall_clamp (npth_unprotect, npth_protect);
These functions may not modify ERRNO.
*/
void
_gpgrt_set_syscall_clamp (void (*pre)(void), void (*post)(void))
{
pre_syscall_func = pre;
post_syscall_func = post;
}
/*
* I/O methods.
*/
/* Implementation of Memory I/O. */
/* Cookie for memory objects. */
typedef struct estream_cookie_mem
{
unsigned int modeflags; /* Open flags. */
unsigned char *memory; /* Allocated data buffer. */
size_t memory_size; /* Allocated size of MEMORY. */
size_t memory_limit; /* Caller supplied maximum allowed
allocation size or 0 for no limit. */
size_t offset; /* Current offset in MEMORY. */
size_t data_len; /* Used length of data in MEMORY. */
size_t block_size; /* Block size. */
struct {
unsigned int grow: 1; /* MEMORY is allowed to grow. */
} flags;
func_realloc_t func_realloc;
func_free_t func_free;
} *estream_cookie_mem_t;
/* Create function for memory objects. DATA is either NULL or a user
supplied buffer with the initial conetnt of the memory buffer. If
DATA is NULL, DATA_N and DATA_LEN need to be 0 as well. If DATA is
not NULL, DATA_N gives the allocated size of DATA and DATA_LEN the
used length in DATA. If this function succeeds DATA is now owned
by this function. If GROW is false FUNC_REALLOC is not
required. */
static int
func_mem_create (void *_GPGRT__RESTRICT *_GPGRT__RESTRICT cookie,
unsigned char *_GPGRT__RESTRICT data, size_t data_n,
size_t data_len,
size_t block_size, unsigned int grow,
func_realloc_t func_realloc, func_free_t func_free,
unsigned int modeflags,
size_t memory_limit)
{
estream_cookie_mem_t mem_cookie;
int err;
if (!data && (data_n || data_len))
{
_set_errno (EINVAL);
return -1;
}
if (grow && func_free && !func_realloc)
{
_set_errno (EINVAL);
return -1;
}
mem_cookie = mem_alloc (sizeof (*mem_cookie));
if (!mem_cookie)
err = -1;
else
{
mem_cookie->modeflags = modeflags;
mem_cookie->memory = data;
mem_cookie->memory_size = data_n;
mem_cookie->memory_limit = memory_limit;
mem_cookie->offset = 0;
mem_cookie->data_len = data_len;
mem_cookie->block_size = block_size;
mem_cookie->flags.grow = !!grow;
mem_cookie->func_realloc
= grow? (func_realloc ? func_realloc : mem_realloc) : NULL;
mem_cookie->func_free = func_free ? func_free : mem_free;
*cookie = mem_cookie;
err = 0;
}
return err;
}
/* Read function for memory objects. */
static gpgrt_ssize_t
es_func_mem_read (void *cookie, void *buffer, size_t size)
{
estream_cookie_mem_t mem_cookie = cookie;
gpgrt_ssize_t ret;
if (!size) /* Just the pending data check. */
return (mem_cookie->data_len - mem_cookie->offset)? 0 : -1;
if (size > mem_cookie->data_len - mem_cookie->offset)
size = mem_cookie->data_len - mem_cookie->offset;
if (size)
{
memcpy (buffer, mem_cookie->memory + mem_cookie->offset, size);
mem_cookie->offset += size;
}
ret = size;
return ret;
}
/* Write function for memory objects. */
static gpgrt_ssize_t
es_func_mem_write (void *cookie, const void *buffer, size_t size)
{
estream_cookie_mem_t mem_cookie = cookie;
gpgrt_ssize_t ret;
size_t nleft;
if (!size)
return 0; /* A flush is a NOP for memory objects. */
if (mem_cookie->modeflags & O_APPEND)
{
/* Append to data. */
mem_cookie->offset = mem_cookie->data_len;
}
assert (mem_cookie->memory_size >= mem_cookie->offset);
nleft = mem_cookie->memory_size - mem_cookie->offset;
/* If we are not allowed to grow the buffer, limit the size to the
left space. */
if (!mem_cookie->flags.grow && size > nleft)
size = nleft;
/* Enlarge the memory buffer if needed. */
if (size > nleft)
{
unsigned char *newbuf;
size_t newsize;
if (!mem_cookie->memory_size)
newsize = size; /* Not yet allocated. */
else
newsize = mem_cookie->memory_size + (size - nleft);
if (newsize < mem_cookie->offset)
{
_set_errno (EINVAL);
return -1;
}
/* Round up to the next block length. BLOCK_SIZE should always
be set; we check anyway. */
if (mem_cookie->block_size)
{
newsize += mem_cookie->block_size - 1;
if (newsize < mem_cookie->offset)
{
_set_errno (EINVAL);
return -1;
}
newsize /= mem_cookie->block_size;
newsize *= mem_cookie->block_size;
}
/* Check for a total limit. */
if (mem_cookie->memory_limit && newsize > mem_cookie->memory_limit)
{
_set_errno (ENOSPC);
return -1;
}
assert (mem_cookie->func_realloc);
newbuf = mem_cookie->func_realloc (mem_cookie->memory, newsize);
if (!newbuf)
return -1;
mem_cookie->memory = newbuf;
mem_cookie->memory_size = newsize;
assert (mem_cookie->memory_size >= mem_cookie->offset);
nleft = mem_cookie->memory_size - mem_cookie->offset;
assert (size <= nleft);
}
memcpy (mem_cookie->memory + mem_cookie->offset, buffer, size);
if (mem_cookie->offset + size > mem_cookie->data_len)
mem_cookie->data_len = mem_cookie->offset + size;
mem_cookie->offset += size;
ret = size;
return ret;
}
/* Seek function for memory objects. */
static int
es_func_mem_seek (void *cookie, gpgrt_off_t *offset, int whence)
{
estream_cookie_mem_t mem_cookie = cookie;
gpgrt_off_t pos_new;
switch (whence)
{
case SEEK_SET:
pos_new = *offset;
break;
case SEEK_CUR:
pos_new = mem_cookie->offset += *offset;
break;
case SEEK_END:
pos_new = mem_cookie->data_len += *offset;
break;
default:
_set_errno (EINVAL);
return -1;
}
if (pos_new > mem_cookie->memory_size)
{
size_t newsize;
void *newbuf;
if (!mem_cookie->flags.grow)
{
_set_errno (ENOSPC);
return -1;
}
newsize = pos_new + mem_cookie->block_size - 1;
if (newsize < pos_new)
{
_set_errno (EINVAL);
return -1;
}
newsize /= mem_cookie->block_size;
newsize *= mem_cookie->block_size;
if (mem_cookie->memory_limit && newsize > mem_cookie->memory_limit)
{
_set_errno (ENOSPC);
return -1;
}
assert (mem_cookie->func_realloc);
newbuf = mem_cookie->func_realloc (mem_cookie->memory, newsize);
if (!newbuf)
return -1;
mem_cookie->memory = newbuf;
mem_cookie->memory_size = newsize;
}
if (pos_new > mem_cookie->data_len)
{
/* Fill spare space with zeroes. */
memset (mem_cookie->memory + mem_cookie->data_len,
0, pos_new - mem_cookie->data_len);
mem_cookie->data_len = pos_new;
}
mem_cookie->offset = pos_new;
*offset = pos_new;
return 0;
}
/* An IOCTL function for memory objects. */
static int
es_func_mem_ioctl (void *cookie, int cmd, void *ptr, size_t *len)
{
estream_cookie_mem_t mem_cookie = cookie;
int ret;
if (cmd == COOKIE_IOCTL_SNATCH_BUFFER)
{
/* Return the internal buffer of the stream to the caller and
invalidate it for the stream. */
*(void**)ptr = mem_cookie->memory;
*len = mem_cookie->offset;
mem_cookie->memory = NULL;
mem_cookie->memory_size = 0;
mem_cookie->offset = 0;
ret = 0;
}
else
{
_set_errno (EINVAL);
ret = -1;
}
return ret;
}
/* Destroy function for memory objects. */
static int
es_func_mem_destroy (void *cookie)
{
estream_cookie_mem_t mem_cookie = cookie;
if (cookie)
{
mem_cookie->func_free (mem_cookie->memory);
mem_free (mem_cookie);
}
return 0;
}
static gpgrt_cookie_io_functions_t estream_functions_mem =
{
es_func_mem_read,
es_func_mem_write,
es_func_mem_seek,
es_func_mem_destroy
};
/* Implementation of file descriptor based I/O. */
/* Cookie for fd objects. */
typedef struct estream_cookie_fd
{
int fd; /* The file descriptor we are using for actual output. */
int no_close; /* If set we won't close the file descriptor. */
int nonblock; /* Non-blocking mode is enabled. */
} *estream_cookie_fd_t;
/* Create function for objects indentified by a libc file descriptor. */
static int
func_fd_create (void **cookie, int fd, unsigned int modeflags, int no_close)
{
estream_cookie_fd_t fd_cookie;
int err;
fd_cookie = mem_alloc (sizeof (*fd_cookie));
if (! fd_cookie)
err = -1;
else
{
#ifdef HAVE_DOSISH_SYSTEM
/* Make sure it is in binary mode if requested. */
if ( (modeflags & O_BINARY) )
setmode (fd, O_BINARY);
#endif
fd_cookie->fd = fd;
fd_cookie->no_close = no_close;
fd_cookie->nonblock = !!(modeflags & O_NONBLOCK);
*cookie = fd_cookie;
err = 0;
}
return err;
}
/* Read function for fd objects. */
static gpgrt_ssize_t
es_func_fd_read (void *cookie, void *buffer, size_t size)
{
estream_cookie_fd_t file_cookie = cookie;
gpgrt_ssize_t bytes_read;
if (!size)
bytes_read = -1; /* We don't know whether anything is pending. */
else if (IS_INVALID_FD (file_cookie->fd))
{
_gpgrt_yield ();
bytes_read = 0;
}
else
{
if (pre_syscall_func)
pre_syscall_func ();
do
{
bytes_read = read (file_cookie->fd, buffer, size);
}
while (bytes_read == -1 && errno == EINTR);
if (post_syscall_func)
post_syscall_func ();
}
return bytes_read;
}
/* Write function for fd objects. */
static gpgrt_ssize_t
es_func_fd_write (void *cookie, const void *buffer, size_t size)
{
estream_cookie_fd_t file_cookie = cookie;
gpgrt_ssize_t bytes_written;
if (IS_INVALID_FD (file_cookie->fd))
{
_gpgrt_yield ();
bytes_written = size; /* Yeah: Success writing to the bit bucket. */
}
else if (buffer)
{
if (pre_syscall_func)
pre_syscall_func ();
do
{
bytes_written = write (file_cookie->fd, buffer, size);
}
while (bytes_written == -1 && errno == EINTR);
if (post_syscall_func)
post_syscall_func ();
}
else
bytes_written = size; /* Note that for a flush SIZE should be 0. */
return bytes_written;
}
/* Seek function for fd objects. */
static int
es_func_fd_seek (void *cookie, gpgrt_off_t *offset, int whence)
{
estream_cookie_fd_t file_cookie = cookie;
gpgrt_off_t offset_new;
int err;
if (IS_INVALID_FD (file_cookie->fd))
{
_set_errno (ESPIPE);
err = -1;
}
else
{
if (pre_syscall_func)
pre_syscall_func ();
offset_new = lseek (file_cookie->fd, *offset, whence);
if (post_syscall_func)
post_syscall_func ();
if (offset_new == -1)
err = -1;
else
{
*offset = offset_new;
err = 0;
}
}
return err;
}
/* An IOCTL function for fd objects. */
static int
es_func_fd_ioctl (void *cookie, int cmd, void *ptr, size_t *len)
{
estream_cookie_fd_t fd_cookie = cookie;
int ret;
if (cmd == COOKIE_IOCTL_NONBLOCK && !len)
{
fd_cookie->nonblock = !!ptr;
if (IS_INVALID_FD (fd_cookie->fd))
{
_set_errno (EINVAL);
ret = -1;
}
else
{
#ifdef _WIN32
_set_errno (EOPNOTSUPP); /* FIXME: Implement for Windows. */
ret = -1;
#else
_set_errno (0);
ret = fcntl (fd_cookie->fd, F_GETFL, 0);
if (ret == -1 && errno)
;
else if (fd_cookie->nonblock)
ret = fcntl (fd_cookie->fd, F_SETFL, (ret | O_NONBLOCK));
else
ret = fcntl (fd_cookie->fd, F_SETFL, (ret & ~O_NONBLOCK));
#endif
}
}
else
{
_set_errno (EINVAL);
ret = -1;
}
return ret;
}
/* Destroy function for fd objects. */
static int
es_func_fd_destroy (void *cookie)
{
estream_cookie_fd_t fd_cookie = cookie;
int err;
if (fd_cookie)
{
if (IS_INVALID_FD (fd_cookie->fd))
err = 0;
else
err = fd_cookie->no_close? 0 : close (fd_cookie->fd);
mem_free (fd_cookie);
}
else
err = 0;
return err;
}
static gpgrt_cookie_io_functions_t estream_functions_fd =
{
es_func_fd_read,
es_func_fd_write,
es_func_fd_seek,
es_func_fd_destroy
};
#ifdef HAVE_W32_SYSTEM
/* Implementation of W32 handle based I/O. */
/* Cookie for fd objects. */
typedef struct estream_cookie_w32
{
HANDLE hd; /* The handle we are using for actual output. */
int no_close; /* If set we won't close the handle. */
} *estream_cookie_w32_t;
/* Create function for w32 handle objects. */
static int
es_func_w32_create (void **cookie, HANDLE hd,
unsigned int modeflags, int no_close)
{
estream_cookie_w32_t w32_cookie;
int err;
w32_cookie = mem_alloc (sizeof (*w32_cookie));
if (!w32_cookie)
err = -1;
else
{
/* CR/LF translations are not supported when using the bare W32
API. If that is really required we need to implemented that
in the upper layer. */
(void)modeflags;
w32_cookie->hd = hd;
w32_cookie->no_close = no_close;
*cookie = w32_cookie;
err = 0;
}
return err;
}
/* Read function for W32 handle objects. */
static gpgrt_ssize_t
es_func_w32_read (void *cookie, void *buffer, size_t size)
{
estream_cookie_w32_t w32_cookie = cookie;
gpgrt_ssize_t bytes_read;
if (!size)
bytes_read = -1; /* We don't know whether anything is pending. */
else if (w32_cookie->hd == INVALID_HANDLE_VALUE)
{
_gpgrt_yield ();
bytes_read = 0;
}
else
{
if (pre_syscall_func)
pre_syscall_func ();
do
{
DWORD nread, ec;
if (!ReadFile (w32_cookie->hd, buffer, size, &nread, NULL))
{
ec = GetLastError ();
if (ec == ERROR_BROKEN_PIPE)
bytes_read = 0; /* Like our pth_read we handle this as EOF. */
else
{
_set_errno (map_w32_to_errno (ec));
bytes_read = -1;
}
}
else
bytes_read = (int)nread;
}
while (bytes_read == -1 && errno == EINTR);
if (post_syscall_func)
post_syscall_func ();
}
return bytes_read;
}
/* Write function for W32 handle objects. */
static gpgrt_ssize_t
es_func_w32_write (void *cookie, const void *buffer, size_t size)
{
estream_cookie_w32_t w32_cookie = cookie;
gpgrt_ssize_t bytes_written;
if (w32_cookie->hd == INVALID_HANDLE_VALUE)
{
_gpgrt_yield ();
bytes_written = size; /* Yeah: Success writing to the bit bucket. */
}
else if (buffer)
{
if (pre_syscall_func)
pre_syscall_func ();
do
{
DWORD nwritten;
if (!WriteFile (w32_cookie->hd, buffer, size, &nwritten, NULL))
{
_set_errno (map_w32_to_errno (GetLastError ()));
bytes_written = -1;
}
else
bytes_written = (int)nwritten;
}
while (bytes_written == -1 && errno == EINTR);
if (post_syscall_func)
post_syscall_func ();
}
else
bytes_written = size; /* Note that for a flush SIZE should be 0. */
return bytes_written;
}
/* Seek function for W32 handle objects. */
static int
es_func_w32_seek (void *cookie, gpgrt_off_t *offset, int whence)
{
estream_cookie_w32_t w32_cookie = cookie;
DWORD method;
LARGE_INTEGER distance, newoff;
if (w32_cookie->hd == INVALID_HANDLE_VALUE)
{
_set_errno (ESPIPE);
return -1;
}
if (whence == SEEK_SET)
{
method = FILE_BEGIN;
distance.QuadPart = (unsigned long long)(*offset);
}
else if (whence == SEEK_CUR)
{
method = FILE_CURRENT;
distance.QuadPart = (long long)(*offset);
}
else if (whence == SEEK_END)
{
method = FILE_END;
distance.QuadPart = (long long)(*offset);
}
else
{
_set_errno (EINVAL);
return -1;
}
#ifdef HAVE_W32CE_SYSTEM
# warning need to use SetFilePointer
#else
if (pre_syscall_func)
pre_syscall_func ();
if (!SetFilePointerEx (w32_cookie->hd, distance, &newoff, method))
{
_set_errno (map_w32_to_errno (GetLastError ()));
if (post_syscall_func)
post_syscall_func ();
return -1;
}
if (post_syscall_func)
post_syscall_func ();
#endif
/* Note that gpgrt_off_t is always 64 bit. */
*offset = (gpgrt_off_t)newoff.QuadPart;
return 0;
}
/* Destroy function for W32 handle objects. */
static int
es_func_w32_destroy (void *cookie)
{
estream_cookie_w32_t w32_cookie = cookie;
int err;
if (w32_cookie)
{
if (w32_cookie->hd == INVALID_HANDLE_VALUE)
err = 0;
else if (w32_cookie->no_close)
err = 0;
else
{
if (!CloseHandle (w32_cookie->hd))
{
_set_errno (map_w32_to_errno (GetLastError ()));
err = -1;
}
else
err = 0;
}
mem_free (w32_cookie);
}
else
err = 0;
return err;
}
static gpgrt_cookie_io_functions_t estream_functions_w32 =
{
es_func_w32_read,
es_func_w32_write,
es_func_w32_seek,
es_func_w32_destroy
};
#endif /*HAVE_W32_SYSTEM*/
/* Implementation of FILE* I/O. */
/* Cookie for fp objects. */
typedef struct estream_cookie_fp
{
FILE *fp; /* The file pointer we are using for actual output. */
int no_close; /* If set we won't close the file pointer. */
} *estream_cookie_fp_t;
/* Create function for FILE objects. */
static int
func_fp_create (void **cookie, FILE *fp,
unsigned int modeflags, int no_close)
{
estream_cookie_fp_t fp_cookie;
int err;
fp_cookie = mem_alloc (sizeof *fp_cookie);
if (!fp_cookie)
err = -1;
else
{
#ifdef HAVE_DOSISH_SYSTEM
/* Make sure it is in binary mode if requested. */
if ( (modeflags & O_BINARY) )
setmode (fileno (fp), O_BINARY);
#else
(void)modeflags;
#endif
fp_cookie->fp = fp;
fp_cookie->no_close = no_close;
*cookie = fp_cookie;
err = 0;
}
return err;
}
/* Read function for FILE* objects. */
static gpgrt_ssize_t
es_func_fp_read (void *cookie, void *buffer, size_t size)
{
estream_cookie_fp_t file_cookie = cookie;
gpgrt_ssize_t bytes_read;
if (!size)
return -1; /* We don't know whether anything is pending. */
if (file_cookie->fp)
{
if (pre_syscall_func)
pre_syscall_func ();
bytes_read = fread (buffer, 1, size, file_cookie->fp);
if (post_syscall_func)
post_syscall_func ();
}
else
bytes_read = 0;
if (!bytes_read && ferror (file_cookie->fp))
return -1;
return bytes_read;
}
/* Write function for FILE* objects. */
static gpgrt_ssize_t
es_func_fp_write (void *cookie, const void *buffer, size_t size)
{
estream_cookie_fp_t file_cookie = cookie;
size_t bytes_written;
if (file_cookie->fp)
{
if (pre_syscall_func)
pre_syscall_func ();
if (buffer)
{
#ifdef HAVE_W32_SYSTEM
/* Using an fwrite to stdout connected to the console fails
with the error "Not enough space" for an fwrite size of
>= 52KB (tested on Windows XP SP2). To solve this we
always chunk the writes up into smaller blocks. */
bytes_written = 0;
while (bytes_written < size)
{
size_t cnt = size - bytes_written;
if (cnt > 32*1024)
cnt = 32*1024;
if (fwrite ((const char*)buffer + bytes_written,
cnt, 1, file_cookie->fp) != 1)
break; /* Write error. */
bytes_written += cnt;
}
#else
bytes_written = fwrite (buffer, 1, size, file_cookie->fp);
#endif
}
else /* Only flush requested. */
bytes_written = size;
fflush (file_cookie->fp);
if (post_syscall_func)
post_syscall_func ();
}
else
bytes_written = size; /* Successfully written to the bit bucket. */
if (bytes_written != size)
return -1;
return bytes_written;
}
/* Seek function for FILE* objects. */
static int
es_func_fp_seek (void *cookie, gpgrt_off_t *offset, int whence)
{
estream_cookie_fp_t file_cookie = cookie;
long int offset_new;
if (!file_cookie->fp)
{
_set_errno (ESPIPE);
return -1;
}
if (pre_syscall_func)
pre_syscall_func ();
if ( fseek (file_cookie->fp, (long int)*offset, whence) )
{
/* fprintf (stderr, "\nfseek failed: errno=%d (%s)\n", */
/* errno,strerror (errno)); */
if (post_syscall_func)
post_syscall_func ();
return -1;
}
offset_new = ftell (file_cookie->fp);
if (post_syscall_func)
post_syscall_func ();
if (offset_new == -1)
{
/* fprintf (stderr, "\nftell failed: errno=%d (%s)\n", */
/* errno,strerror (errno)); */
return -1;
}
*offset = offset_new;
return 0;
}
/* Destroy function for FILE* objects. */
static int
es_func_fp_destroy (void *cookie)
{
estream_cookie_fp_t fp_cookie = cookie;
int err;
if (fp_cookie)
{
if (fp_cookie->fp)
{
if (pre_syscall_func)
pre_syscall_func ();
fflush (fp_cookie->fp);
if (post_syscall_func)
post_syscall_func ();
err = fp_cookie->no_close? 0 : fclose (fp_cookie->fp);
}
else
err = 0;
mem_free (fp_cookie);
}
else
err = 0;
return err;
}
static gpgrt_cookie_io_functions_t estream_functions_fp =
{
es_func_fp_read,
es_func_fp_write,
es_func_fp_seek,
es_func_fp_destroy
};
/* Implementation of file I/O. */
/* Create function for objects identified by a file name. */
static int
func_file_create (void **cookie, int *filedes,
const char *path, unsigned int modeflags, unsigned int cmode)
{
estream_cookie_fd_t file_cookie;
int err;
int fd;
err = 0;
fd = -1;
file_cookie = mem_alloc (sizeof (*file_cookie));
if (! file_cookie)
{
err = -1;
goto out;
}
fd = open (path, modeflags, cmode);
if (fd == -1)
{
err = -1;
goto out;
}
#ifdef HAVE_DOSISH_SYSTEM
/* Make sure it is in binary mode if requested. */
if ( (modeflags & O_BINARY) )
setmode (fd, O_BINARY);
#endif
file_cookie->fd = fd;
file_cookie->no_close = 0;
*cookie = file_cookie;
*filedes = fd;
out:
if (err)
mem_free (file_cookie);
return err;
}
/* Parse the mode flags of fopen et al. In addition to the POSIX
defined mode flags keyword parameters are supported. These are
key/value pairs delimited by comma and optional white spaces.
Keywords and values may not contain a comma or white space; unknown
keywords are skipped. Supported keywords are:
mode=
Creates a file and gives the new file read and write permissions
for the user and read permission for the group. The format of
the string is the same as shown by the -l option of the ls(1)
command. However the first letter must be a dash and it is
allowed to leave out trailing dashes. If this keyword parameter
is not given the default mode for creating files is "-rw-rw-r--"
(664). Note that the system still applies the current umask to
the mode when crating a file. Example:
"wb,mode=-rw-r--"
samethread
Assumes that the object is only used by the creating thread and
disables any internal locking. This keyword is also found on
IBM systems.
nonblock
The object is opened in non-blocking mode. This is the same as
calling gpgrt_set_nonblock on the file.
Note: R_CMODE is optional because is only required by functions
which are able to creat a file. */
static int
parse_mode (const char *modestr,
unsigned int *modeflags, int *samethread,
unsigned int *r_cmode)
{
unsigned int omode, oflags, cmode;
int got_cmode = 0;
*samethread = 0;
switch (*modestr)
{
case 'r':
omode = O_RDONLY;
oflags = 0;
break;
case 'w':
omode = O_WRONLY;
oflags = O_TRUNC | O_CREAT;
break;
case 'a':
omode = O_WRONLY;
oflags = O_APPEND | O_CREAT;
break;
default:
_set_errno (EINVAL);
return -1;
}
for (modestr++; *modestr; modestr++)
{
switch (*modestr)
{
case '+':
omode = O_RDWR;
break;
case 'b':
oflags |= O_BINARY;
break;
case 'x':
oflags |= O_EXCL;
break;
case ',':
goto keyvalue;
default: /* Ignore unknown flags. */
break;
}
}
keyvalue:
/* Parse key/value pairs (similar to fopen on mainframes). */
for (cmode=0; *modestr == ','; modestr += strcspn (modestr, ","))
{
modestr++;
modestr += strspn (modestr, " \t");
if (!strncmp (modestr, "mode=", 5))
{
static struct {
char letter;
unsigned int value;
} table[] = { { '-', 0 },
{ 'r', S_IRUSR }, { 'w', S_IWUSR }, { 'x', S_IXUSR },
{ 'r', S_IRGRP }, { 'w', S_IWGRP }, { 'x', S_IXGRP },
{ 'r', S_IROTH }, { 'w', S_IWOTH }, { 'x', S_IXOTH }};
int idx;
got_cmode = 1;
modestr += 5;
/* For now we only support a string as used by ls(1) and no
octal numbers. The first character must be a dash. */
for (idx=0; idx < 10 && *modestr; idx++, modestr++)
{
if (*modestr == table[idx].letter)
cmode |= table[idx].value;
else if (*modestr != '-')
break;
}
if (*modestr && !strchr (" \t,", *modestr))
{
_set_errno (EINVAL);
return -1;
}
}
else if (!strncmp (modestr, "samethread", 10))
{
modestr += 10;
if (*modestr && !strchr (" \t,", *modestr))
{
_set_errno (EINVAL);
return -1;
}
*samethread = 1;
}
else if (!strncmp (modestr, "nonblock", 8))
{
modestr += 8;
if (*modestr && !strchr (" \t,", *modestr))
{
_set_errno (EINVAL);
return -1;
}
oflags |= O_NONBLOCK;
}
}
if (!got_cmode)
cmode = (S_IRUSR | S_IWUSR | S_IRGRP | S_IWGRP | S_IROTH);
*modeflags = (omode | oflags);
if (r_cmode)
*r_cmode = cmode;
return 0;
}
/*
* Low level stream functionality.
*/
static int
es_fill (estream_t stream)
{
size_t bytes_read = 0;
int err;
if (!stream->intern->func_read)
{
_set_errno (EOPNOTSUPP);
err = -1;
}
else if (!stream->buffer_size)
err = 0;
else
{
gpgrt_cookie_read_function_t func_read = stream->intern->func_read;
gpgrt_ssize_t ret;
ret = (*func_read) (stream->intern->cookie,
stream->buffer, stream->buffer_size);
if (ret == -1)
{
bytes_read = 0;
err = -1;
#if EWOULDBLOCK != EAGAIN
if (errno == EWOULDBLOCK)
_set_errno (EAGAIN);
#endif
}
else
{
bytes_read = ret;
err = 0;
}
}
if (err)
{
if (errno != EAGAIN)
{
if (errno == EPIPE)
stream->intern->indicators.hup = 1;
stream->intern->indicators.err = 1;
}
}
else if (!bytes_read)
stream->intern->indicators.eof = 1;
stream->intern->offset += stream->data_len;
stream->data_len = bytes_read;
stream->data_offset = 0;
return err;
}
static int
es_flush (estream_t stream)
{
gpgrt_cookie_write_function_t func_write = stream->intern->func_write;
int err;
assert (stream->flags.writing);
if (stream->data_offset)
{
size_t bytes_written;
size_t data_flushed;
gpgrt_ssize_t ret;
if (! func_write)
{
err = EOPNOTSUPP;
goto out;
}
/* Note: to prevent an endless loop caused by user-provided
write-functions that pretend to have written more bytes than
they were asked to write, we have to check for
"(stream->data_offset - data_flushed) > 0" instead of
"stream->data_offset - data_flushed". */
data_flushed = 0;
err = 0;
while ((((gpgrt_ssize_t) (stream->data_offset - data_flushed)) > 0)
&& !err)
{
ret = (*func_write) (stream->intern->cookie,
stream->buffer + data_flushed,
stream->data_offset - data_flushed);
if (ret == -1)
{
bytes_written = 0;
err = -1;
#if EWOULDBLOCK != EAGAIN
if (errno == EWOULDBLOCK)
_set_errno (EAGAIN);
#endif
}
else
bytes_written = ret;
data_flushed += bytes_written;
if (err)
break;
}
stream->data_flushed += data_flushed;
if (stream->data_offset == data_flushed)
{
stream->intern->offset += stream->data_offset;
stream->data_offset = 0;
stream->data_flushed = 0;
/* Propagate flush event. */
(*func_write) (stream->intern->cookie, NULL, 0);
}
}
else
err = 0;
out:
if (err && errno != EAGAIN)
{
if (errno == EPIPE)
stream->intern->indicators.hup = 1;
stream->intern->indicators.err = 1;
}
return err;
}
/* Discard buffered data for STREAM. */
static void
es_empty (estream_t stream)
{
assert (!stream->flags.writing);
stream->data_len = 0;
stream->data_offset = 0;
stream->unread_data_len = 0;
}
/* Initialize STREAM. */
static void
init_stream_obj (estream_t stream,
void *cookie, es_syshd_t *syshd,
gpgrt_cookie_io_functions_t functions,
unsigned int modeflags, int samethread)
{
stream->intern->cookie = cookie;
stream->intern->opaque = NULL;
stream->intern->offset = 0;
stream->intern->func_read = functions.func_read;
stream->intern->func_write = functions.func_write;
stream->intern->func_seek = functions.func_seek;
stream->intern->func_ioctl = NULL;
stream->intern->func_close = functions.func_close;
stream->intern->strategy = _IOFBF;
stream->intern->syshd = *syshd;
stream->intern->print_ntotal = 0;
stream->intern->indicators.err = 0;
stream->intern->indicators.eof = 0;
stream->intern->indicators.hup = 0;
stream->intern->is_stdstream = 0;
stream->intern->stdstream_fd = 0;
stream->intern->deallocate_buffer = 0;
stream->intern->printable_fname = NULL;
stream->intern->printable_fname_inuse = 0;
stream->intern->samethread = !!samethread;
stream->intern->onclose = NULL;
stream->data_len = 0;
stream->data_offset = 0;
stream->data_flushed = 0;
stream->unread_data_len = 0;
/* Depending on the modeflags we set whether we start in writing or
reading mode. This is required in case we are working on a
stream which is not seeekable (like stdout). Without this
pre-initialization we would do a seek at the first write call and
as this will fail no output will be delivered. */
if ((modeflags & O_WRONLY) || (modeflags & O_RDWR) )
stream->flags.writing = 1;
else
stream->flags.writing = 0;
}
/* Deinitialize STREAM. */
static int
es_deinitialize (estream_t stream)
{
gpgrt_cookie_close_function_t func_close;
int err, tmp_err;
func_close = stream->intern->func_close;
err = 0;
if (stream->flags.writing)
SET_UNLESS_NONZERO (err, tmp_err, es_flush (stream));
if (func_close)
SET_UNLESS_NONZERO (err, tmp_err, (*func_close) (stream->intern->cookie));
mem_free (stream->intern->printable_fname);
stream->intern->printable_fname = NULL;
stream->intern->printable_fname_inuse = 0;
while (stream->intern->onclose)
{
notify_list_t tmp = stream->intern->onclose->next;
mem_free (stream->intern->onclose);
stream->intern->onclose = tmp;
}
return err;
}
/* Create a new stream object, initialize it. */
static int
es_create (estream_t *stream, void *cookie, es_syshd_t *syshd,
gpgrt_cookie_io_functions_t functions, unsigned int modeflags,
int samethread, int with_locked_list)
{
estream_internal_t stream_internal_new;
estream_t stream_new;
int err;
stream_new = NULL;
stream_internal_new = NULL;
stream_new = mem_alloc (sizeof (*stream_new));
if (! stream_new)
{
err = -1;
goto out;
}
stream_internal_new = mem_alloc (sizeof (*stream_internal_new));
if (! stream_internal_new)
{
err = -1;
goto out;
}
stream_new->buffer = stream_internal_new->buffer;
stream_new->buffer_size = sizeof (stream_internal_new->buffer);
stream_new->unread_buffer = stream_internal_new->unread_buffer;
stream_new->unread_buffer_size = sizeof (stream_internal_new->unread_buffer);
stream_new->intern = stream_internal_new;
init_stream_obj (stream_new, cookie, syshd, functions, modeflags, samethread);
init_stream_lock (stream_new);
err = do_list_add (stream_new, with_locked_list);
if (err)
goto out;
*stream = stream_new;
out:
if (err)
{
if (stream_new)
{
es_deinitialize (stream_new);
destroy_stream_lock (stream_new);
mem_free (stream_new->intern);
mem_free (stream_new);
}
}
return err;
}
/* Deinitialize a stream object and destroy it. */
static int
do_close (estream_t stream, int with_locked_list)
{
int err;
if (stream)
{
do_list_remove (stream, with_locked_list);
while (stream->intern->onclose)
{
notify_list_t tmp = stream->intern->onclose->next;
if (stream->intern->onclose->fnc)
stream->intern->onclose->fnc (stream,
stream->intern->onclose->fnc_value);
mem_free (stream->intern->onclose);
stream->intern->onclose = tmp;
}
err = es_deinitialize (stream);
destroy_stream_lock (stream);
mem_free (stream->intern);
mem_free (stream);
}
else
err = 0;
return err;
}
/* This worker function is called with a locked stream. */
static int
do_onclose (estream_t stream, int mode,
void (*fnc) (estream_t, void*), void *fnc_value)
{
notify_list_t item;
if (!mode)
{
for (item = stream->intern->onclose; item; item = item->next)
if (item->fnc && item->fnc == fnc && item->fnc_value == fnc_value)
item->fnc = NULL; /* Disable this notification. */
}
else
{
item = mem_alloc (sizeof *item);
if (!item)
return -1;
item->fnc = fnc;
item->fnc_value = fnc_value;
item->next = stream->intern->onclose;
stream->intern->onclose = item;
}
return 0;
}
/* Try to read BYTES_TO_READ bytes FROM STREAM into BUFFER in
unbuffered-mode, storing the amount of bytes read in
*BYTES_READ. */
static int
es_read_nbf (estream_t _GPGRT__RESTRICT stream,
unsigned char *_GPGRT__RESTRICT buffer,
size_t bytes_to_read, size_t *_GPGRT__RESTRICT bytes_read)
{
gpgrt_cookie_read_function_t func_read = stream->intern->func_read;
size_t data_read;
gpgrt_ssize_t ret;
int err;
data_read = 0;
err = 0;
while (bytes_to_read - data_read)
{
ret = (*func_read) (stream->intern->cookie,
buffer + data_read, bytes_to_read - data_read);
if (ret == -1)
{
err = -1;
#if EWOULDBLOCK != EAGAIN
if (errno == EWOULDBLOCK)
_set_errno (EAGAIN);
#endif
break;
}
else if (ret)
data_read += ret;
else
break;
}
stream->intern->offset += data_read;
*bytes_read = data_read;
return err;
}
static int
check_pending_nbf (estream_t _GPGRT__RESTRICT stream)
{
gpgrt_cookie_read_function_t func_read = stream->intern->func_read;
char buffer[1];
if (!(*func_read) (stream->intern->cookie, buffer, 0))
return 1; /* Pending bytes. */
return 0; /* No pending bytes or error. */
}
/* Try to read BYTES_TO_READ bytes FROM STREAM into BUFFER in
fully-buffered-mode, storing the amount of bytes read in
*BYTES_READ. */
static int
es_read_fbf (estream_t _GPGRT__RESTRICT stream,
unsigned char *_GPGRT__RESTRICT buffer,
size_t bytes_to_read, size_t *_GPGRT__RESTRICT bytes_read)
{
size_t data_available;
size_t data_to_read;
size_t data_read;
int err;
data_read = 0;
err = 0;
while ((bytes_to_read - data_read) && (! err))
{
if (stream->data_offset == stream->data_len)
{
/* Nothing more to read in current container, try to
fill container with new data. */
err = es_fill (stream);
if (! err)
if (! stream->data_len)
/* Filling did not result in any data read. */
break;
}
if (! err)
{
/* Filling resulted in some new data. */
data_to_read = bytes_to_read - data_read;
data_available = stream->data_len - stream->data_offset;
if (data_to_read > data_available)
data_to_read = data_available;
memcpy (buffer + data_read,
stream->buffer + stream->data_offset, data_to_read);
stream->data_offset += data_to_read;
data_read += data_to_read;
}
}
*bytes_read = data_read;
return err;
}
static int
check_pending_fbf (estream_t _GPGRT__RESTRICT stream)
{
gpgrt_cookie_read_function_t func_read = stream->intern->func_read;
char buffer[1];
if (stream->data_offset == stream->data_len)
{
/* Nothing more to read in current container, check whether it
would be possible to fill the container with new data. */
if (!(*func_read) (stream->intern->cookie, buffer, 0))
return 1; /* Pending bytes. */
}
else
return 1;
return 0;
}
/* Try to read BYTES_TO_READ bytes FROM STREAM into BUFFER in
line-buffered-mode, storing the amount of bytes read in
*BYTES_READ. */
static int
es_read_lbf (estream_t _GPGRT__RESTRICT stream,
unsigned char *_GPGRT__RESTRICT buffer,
size_t bytes_to_read, size_t *_GPGRT__RESTRICT bytes_read)
{
int err;
err = es_read_fbf (stream, buffer, bytes_to_read, bytes_read);
return err;
}
/* Try to read BYTES_TO_READ bytes FROM STREAM into BUFFER, storing
the amount of bytes read in BYTES_READ. */
static int
es_readn (estream_t _GPGRT__RESTRICT stream,
void *_GPGRT__RESTRICT buffer_arg,
size_t bytes_to_read, size_t *_GPGRT__RESTRICT bytes_read)
{
unsigned char *buffer = (unsigned char *)buffer_arg;
size_t data_read_unread, data_read;
int err;
data_read_unread = 0;
data_read = 0;
err = 0;
if (stream->flags.writing)
{
/* Switching to reading mode -> flush output. */
err = es_flush (stream);
if (err)
goto out;
stream->flags.writing = 0;
}
/* Read unread data first. */
while ((bytes_to_read - data_read_unread) && stream->unread_data_len)
{
buffer[data_read_unread]
= stream->unread_buffer[stream->unread_data_len - 1];
stream->unread_data_len--;
data_read_unread++;
}
switch (stream->intern->strategy)
{
case _IONBF:
err = es_read_nbf (stream,
buffer + data_read_unread,
bytes_to_read - data_read_unread, &data_read);
break;
case _IOLBF:
err = es_read_lbf (stream,
buffer + data_read_unread,
bytes_to_read - data_read_unread, &data_read);
break;
case _IOFBF:
err = es_read_fbf (stream,
buffer + data_read_unread,
bytes_to_read - data_read_unread, &data_read);
break;
}
out:
if (bytes_read)
*bytes_read = data_read_unread + data_read;
return err;
}
/* Return true if at least one byte is pending for read. This is a
best effort check and it it possible that bytes are still pending
even if false is returned. If the stream is in writing mode it is
switched to read mode. */
static int
check_pending (estream_t _GPGRT__RESTRICT stream)
{
if (stream->flags.writing)
{
/* Switching to reading mode -> flush output. */
if (es_flush (stream))
return 0; /* Better return 0 on error. */
stream->flags.writing = 0;
}
/* Check unread data first. */
if (stream->unread_data_len)
return 1;
switch (stream->intern->strategy)
{
case _IONBF:
return check_pending_nbf (stream);
case _IOLBF:
case _IOFBF:
return check_pending_fbf (stream);
}
return 0;
}
/* Try to unread DATA_N bytes from DATA into STREAM, storing the
amount of bytes successfully unread in *BYTES_UNREAD. */
static void
es_unreadn (estream_t _GPGRT__RESTRICT stream,
const unsigned char *_GPGRT__RESTRICT data, size_t data_n,
size_t *_GPGRT__RESTRICT bytes_unread)
{
size_t space_left;
space_left = stream->unread_buffer_size - stream->unread_data_len;
if (data_n > space_left)
data_n = space_left;
if (! data_n)
goto out;
memcpy (stream->unread_buffer + stream->unread_data_len, data, data_n);
stream->unread_data_len += data_n;
stream->intern->indicators.eof = 0;
out:
if (bytes_unread)
*bytes_unread = data_n;
}
/* Seek in STREAM. */
static int
es_seek (estream_t _GPGRT__RESTRICT stream, gpgrt_off_t offset, int whence,
gpgrt_off_t *_GPGRT__RESTRICT offset_new)
{
gpgrt_cookie_seek_function_t func_seek = stream->intern->func_seek;
int err, ret;
gpgrt_off_t off;
if (! func_seek)
{
_set_errno (EOPNOTSUPP);
err = -1;
goto out;
}
if (stream->flags.writing)
{
/* Flush data first in order to prevent flushing it to the wrong
offset. */
err = es_flush (stream);
if (err)
goto out;
stream->flags.writing = 0;
}
off = offset;
if (whence == SEEK_CUR)
{
off = off - stream->data_len + stream->data_offset;
off -= stream->unread_data_len;
}
ret = (*func_seek) (stream->intern->cookie, &off, whence);
if (ret == -1)
{
err = -1;
#if EWOULDBLOCK != EAGAIN
if (errno == EWOULDBLOCK)
_set_errno (EAGAIN);
#endif
goto out;
}
err = 0;
es_empty (stream);
if (offset_new)
*offset_new = off;
stream->intern->indicators.eof = 0;
stream->intern->offset = off;
out:
if (err)
{
if (errno == EPIPE)
stream->intern->indicators.hup = 1;
stream->intern->indicators.err = 1;
}
return err;
}
/* Write BYTES_TO_WRITE bytes from BUFFER into STREAM in
unbuffered-mode, storing the amount of bytes written in
*BYTES_WRITTEN. */
static int
es_write_nbf (estream_t _GPGRT__RESTRICT stream,
const unsigned char *_GPGRT__RESTRICT buffer,
size_t bytes_to_write, size_t *_GPGRT__RESTRICT bytes_written)
{
gpgrt_cookie_write_function_t func_write = stream->intern->func_write;
size_t data_written;
gpgrt_ssize_t ret;
int err;
if (bytes_to_write && (! func_write))
{
err = EOPNOTSUPP;
goto out;
}
data_written = 0;
err = 0;
while (bytes_to_write - data_written)
{
ret = (*func_write) (stream->intern->cookie,
buffer + data_written,
bytes_to_write - data_written);
if (ret == -1)
{
err = -1;
#if EWOULDBLOCK != EAGAIN
if (errno == EWOULDBLOCK)
_set_errno (EAGAIN);
#endif
break;
}
else
data_written += ret;
}
stream->intern->offset += data_written;
*bytes_written = data_written;
out:
return err;
}
/* Write BYTES_TO_WRITE bytes from BUFFER into STREAM in
fully-buffered-mode, storing the amount of bytes written in
*BYTES_WRITTEN. */
static int
es_write_fbf (estream_t _GPGRT__RESTRICT stream,
const unsigned char *_GPGRT__RESTRICT buffer,
size_t bytes_to_write, size_t *_GPGRT__RESTRICT bytes_written)
{
size_t space_available;
size_t data_to_write;
size_t data_written;
int err;
data_written = 0;
err = 0;
while ((bytes_to_write - data_written) && (! err))
{
if (stream->data_offset == stream->buffer_size)
/* Container full, flush buffer. */
err = es_flush (stream);
if (! err)
{
/* Flushing resulted in empty container. */
data_to_write = bytes_to_write - data_written;
space_available = stream->buffer_size - stream->data_offset;
if (data_to_write > space_available)
data_to_write = space_available;
memcpy (stream->buffer + stream->data_offset,
buffer + data_written, data_to_write);
stream->data_offset += data_to_write;
data_written += data_to_write;
}
}
*bytes_written = data_written;
return err;
}
/* Write BYTES_TO_WRITE bytes from BUFFER into STREAM in
line-buffered-mode, storing the amount of bytes written in
*BYTES_WRITTEN. */
static int
es_write_lbf (estream_t _GPGRT__RESTRICT stream,
const unsigned char *_GPGRT__RESTRICT buffer,
size_t bytes_to_write, size_t *_GPGRT__RESTRICT bytes_written)
{
size_t data_flushed = 0;
size_t data_buffered = 0;
unsigned char *nlp;
int err = 0;
nlp = memrchr (buffer, '\n', bytes_to_write);
if (nlp)
{
/* Found a newline, directly write up to (including) this
character. */
err = es_flush (stream);
if (!err)
err = es_write_nbf (stream, buffer, nlp - buffer + 1, &data_flushed);
}
if (!err)
{
/* Write remaining data fully buffered. */
err = es_write_fbf (stream, buffer + data_flushed,
bytes_to_write - data_flushed, &data_buffered);
}
*bytes_written = data_flushed + data_buffered;
return err;
}
/* Write BYTES_TO_WRITE bytes from BUFFER into STREAM in, storing the
amount of bytes written in BYTES_WRITTEN. */
static int
es_writen (estream_t _GPGRT__RESTRICT stream,
const void *_GPGRT__RESTRICT buffer,
size_t bytes_to_write, size_t *_GPGRT__RESTRICT bytes_written)
{
size_t data_written;
int err;
data_written = 0;
err = 0;
if (!stream->flags.writing)
{
/* Switching to writing mode -> discard input data and seek to
position at which reading has stopped. We can do this only
if a seek function has been registered. */
if (stream->intern->func_seek)
{
err = es_seek (stream, 0, SEEK_CUR, NULL);
if (err)
{
if (errno == ESPIPE)
err = 0;
else
goto out;
}
}
}
switch (stream->intern->strategy)
{
case _IONBF:
err = es_write_nbf (stream, buffer, bytes_to_write, &data_written);
break;
case _IOLBF:
err = es_write_lbf (stream, buffer, bytes_to_write, &data_written);
break;
case _IOFBF:
err = es_write_fbf (stream, buffer, bytes_to_write, &data_written);
break;
}
out:
if (bytes_written)
*bytes_written = data_written;
if (data_written)
if (!stream->flags.writing)
stream->flags.writing = 1;
return err;
}
static int
es_peek (estream_t _GPGRT__RESTRICT stream,
unsigned char **_GPGRT__RESTRICT data,
size_t *_GPGRT__RESTRICT data_len)
{
int err;
if (stream->flags.writing)
{
/* Switching to reading mode -> flush output. */
err = es_flush (stream);
if (err)
goto out;
stream->flags.writing = 0;
}
if (stream->data_offset == stream->data_len)
{
/* Refill container. */
err = es_fill (stream);
if (err)
goto out;
}
if (data)
*data = stream->buffer + stream->data_offset;
if (data_len)
*data_len = stream->data_len - stream->data_offset;
err = 0;
out:
return err;
}
/* Skip SIZE bytes of input data contained in buffer. */
static int
es_skip (estream_t stream, size_t size)
{
int err;
if (stream->data_offset + size > stream->data_len)
{
_set_errno (EINVAL);
err = -1;
}
else
{
stream->data_offset += size;
err = 0;
}
return err;
}
static int
doreadline (estream_t _GPGRT__RESTRICT stream, size_t max_length,
char *_GPGRT__RESTRICT *_GPGRT__RESTRICT line,
size_t *_GPGRT__RESTRICT line_length)
{
size_t space_left;
size_t line_size;
estream_t line_stream;
char *line_new;
void *line_stream_cookie;
char *newline;
unsigned char *data;
size_t data_len;
int err;
es_syshd_t syshd;
line_new = NULL;
line_stream = NULL;
line_stream_cookie = NULL;
err = func_mem_create (&line_stream_cookie, NULL, 0, 0,
BUFFER_BLOCK_SIZE, 1,
mem_realloc, mem_free,
O_RDWR,
0);
if (err)
goto out;
memset (&syshd, 0, sizeof syshd);
err = es_create (&line_stream, line_stream_cookie, &syshd,
estream_functions_mem, O_RDWR, 1, 0);
if (err)
goto out;
space_left = max_length;
line_size = 0;
while (1)
{
if (max_length && (space_left == 1))
break;
err = es_peek (stream, &data, &data_len);
if (err || (! data_len))
break;
if (data_len > (space_left - 1))
data_len = space_left - 1;
newline = memchr (data, '\n', data_len);
if (newline)
{
data_len = (newline - (char *) data) + 1;
err = _gpgrt_write (line_stream, data, data_len, NULL);
if (! err)
{
space_left -= data_len;
line_size += data_len;
es_skip (stream, data_len);
break;
}
}
else
{
err = _gpgrt_write (line_stream, data, data_len, NULL);
if (! err)
{
space_left -= data_len;
line_size += data_len;
es_skip (stream, data_len);
}
}
if (err)
break;
}
if (err)
goto out;
/* Complete line has been written to line_stream. */
if ((max_length > 1) && (! line_size))
{
stream->intern->indicators.eof = 1;
goto out;
}
err = es_seek (line_stream, 0, SEEK_SET, NULL);
if (err)
goto out;
if (! *line)
{
line_new = mem_alloc (line_size + 1);
if (! line_new)
{
err = -1;
goto out;
}
}
else
line_new = *line;
err = _gpgrt_read (line_stream, line_new, line_size, NULL);
if (err)
goto out;
line_new[line_size] = '\0';
if (! *line)
*line = line_new;
if (line_length)
*line_length = line_size;
out:
if (line_stream)
do_close (line_stream, 0);
else if (line_stream_cookie)
es_func_mem_destroy (line_stream_cookie);
if (err)
{
if (! *line)
mem_free (line_new);
stream->intern->indicators.err = 1;
}
return err;
}
/* Output function used by estream_format. */
static int
print_writer (void *outfncarg, const char *buf, size_t buflen)
{
estream_t stream = outfncarg;
size_t nwritten;
int rc;
nwritten = 0;
rc = es_writen (stream, buf, buflen, &nwritten);
stream->intern->print_ntotal += nwritten;
return rc;
}
/* The core of our printf function. This is called in locked state. */
static int
es_print (estream_t _GPGRT__RESTRICT stream,
const char *_GPGRT__RESTRICT format, va_list ap)
{
int rc;
stream->intern->print_ntotal = 0;
rc = _gpgrt_estream_format (print_writer, stream, format, ap);
if (rc)
return -1;
return (int)stream->intern->print_ntotal;
}
static int
es_set_buffering (estream_t _GPGRT__RESTRICT stream,
char *_GPGRT__RESTRICT buffer, int mode, size_t size)
{
int err;
/* Flush or empty buffer depending on mode. */
if (stream->flags.writing)
{
err = es_flush (stream);
if (err)
goto out;
}
else
es_empty (stream);
stream->intern->indicators.eof = 0;
/* Free old buffer in case that was allocated by this function. */
if (stream->intern->deallocate_buffer)
{
stream->intern->deallocate_buffer = 0;
mem_free (stream->buffer);
stream->buffer = NULL;
}
if (mode == _IONBF)
stream->buffer_size = 0;
else
{
void *buffer_new;
if (buffer)
buffer_new = buffer;
else
{
if (!size)
size = BUFSIZ;
buffer_new = mem_alloc (size);
if (! buffer_new)
{
err = -1;
goto out;
}
}
stream->buffer = buffer_new;
stream->buffer_size = size;
if (! buffer)
stream->intern->deallocate_buffer = 1;
}
stream->intern->strategy = mode;
err = 0;
out:
return err;
}
static gpgrt_off_t
es_offset_calculate (estream_t stream)
{
gpgrt_off_t offset;
offset = stream->intern->offset + stream->data_offset;
if (offset < stream->unread_data_len)
/* Offset undefined. */
offset = 0;
else
offset -= stream->unread_data_len;
return offset;
}
static void
es_opaque_ctrl (estream_t _GPGRT__RESTRICT stream,
void *_GPGRT__RESTRICT opaque_new,
void **_GPGRT__RESTRICT opaque_old)
{
if (opaque_old)
*opaque_old = stream->intern->opaque;
if (opaque_new)
stream->intern->opaque = opaque_new;
}
/* API. */
estream_t
_gpgrt_fopen (const char *_GPGRT__RESTRICT path,
const char *_GPGRT__RESTRICT mode)
{
unsigned int modeflags, cmode;
int samethread, create_called;
estream_t stream;
void *cookie;
int err;
int fd;
es_syshd_t syshd;
stream = NULL;
cookie = NULL;
create_called = 0;
err = parse_mode (mode, &modeflags, &samethread, &cmode);
if (err)
goto out;
err = func_file_create (&cookie, &fd, path, modeflags, cmode);
if (err)
goto out;
syshd.type = ES_SYSHD_FD;
syshd.u.fd = fd;
create_called = 1;
err = es_create (&stream, cookie, &syshd, estream_functions_fd, modeflags,
samethread, 0);
if (err)
goto out;
if (stream && path)
fname_set_internal (stream, path, 1);
out:
if (err && create_called)
(*estream_functions_fd.func_close) (cookie);
return stream;
}
/* Create a new estream object in memory. If DATA is not NULL this
buffer will be used as the memory buffer; thus after this functions
returns with the success the the memory at DATA belongs to the new
estream. The allocated length of DATA is given by DATA_LEN and its
used length by DATA_N. Usually this is malloced buffer; if a
static buffer is provided, the caller must pass false for GROW and
provide a dummy function for FUNC_FREE. FUNC_FREE and FUNC_REALLOC
allow the caller to provide custom functions for realloc and free
to be used by the new estream object. Note that the realloc
function is also used for initial allocation. If DATA is NULL a
buffer is internally allocated; either using internal function or
those provide by the caller. It is an error to provide a realloc
function but no free function. Providing only a free function is
allowed as long as GROW is false. */
estream_t
_gpgrt_mopen (void *_GPGRT__RESTRICT data, size_t data_n, size_t data_len,
unsigned int grow,
func_realloc_t func_realloc, func_free_t func_free,
const char *_GPGRT__RESTRICT mode)
{
int create_called = 0;
estream_t stream = NULL;
void *cookie = NULL;
unsigned int modeflags;
int samethread;
int err;
es_syshd_t syshd;
err = parse_mode (mode, &modeflags, &samethread, NULL);
if (err)
goto out;
err = func_mem_create (&cookie, data, data_n, data_len,
BUFFER_BLOCK_SIZE, grow,
func_realloc, func_free, modeflags, 0);
if (err)
goto out;
memset (&syshd, 0, sizeof syshd);
create_called = 1;
err = es_create (&stream, cookie, &syshd,
estream_functions_mem, modeflags, samethread, 0);
out:
if (err && create_called)
(*estream_functions_mem.func_close) (cookie);
return stream;
}
estream_t
_gpgrt_fopenmem (size_t memlimit, const char *_GPGRT__RESTRICT mode)
{
unsigned int modeflags;
int samethread;
estream_t stream = NULL;
void *cookie = NULL;
es_syshd_t syshd;
/* Memory streams are always read/write. We use MODE only to get
the append flag. */
if (parse_mode (mode, &modeflags, &samethread, NULL))
return NULL;
modeflags |= O_RDWR;
if (func_mem_create (&cookie, NULL, 0, 0,
BUFFER_BLOCK_SIZE, 1,
mem_realloc, mem_free, modeflags,
memlimit))
return NULL;
memset (&syshd, 0, sizeof syshd);
if (es_create (&stream, cookie, &syshd, estream_functions_mem, modeflags,
samethread, 0))
(*estream_functions_mem.func_close) (cookie);
if (stream)
stream->intern->func_ioctl = es_func_mem_ioctl;
return stream;
}
/* This is the same as es_fopenmem but intializes the memory with a
copy of (DATA,DATALEN). The stream is initially set to the
beginning. If MEMLIMIT is not 0 but shorter than DATALEN it
DATALEN will be used as the value for MEMLIMIT. */
estream_t
_gpgrt_fopenmem_init (size_t memlimit, const char *_GPGRT__RESTRICT mode,
const void *data, size_t datalen)
{
estream_t stream;
if (memlimit && memlimit < datalen)
memlimit = datalen;
stream = _gpgrt_fopenmem (memlimit, mode);
if (stream && data && datalen)
{
if (es_writen (stream, data, datalen, NULL))
{
int saveerrno = errno;
_gpgrt_fclose (stream);
stream = NULL;
_set_errno (saveerrno);
}
else
{
es_seek (stream, 0L, SEEK_SET, NULL);
stream->intern->indicators.eof = 0;
stream->intern->indicators.err = 0;
}
}
return stream;
}
estream_t
_gpgrt_fopencookie (void *_GPGRT__RESTRICT cookie,
const char *_GPGRT__RESTRICT mode,
gpgrt_cookie_io_functions_t functions)
{
unsigned int modeflags;
int samethread;
estream_t stream;
int err;
es_syshd_t syshd;
stream = NULL;
modeflags = 0;
err = parse_mode (mode, &modeflags, &samethread, NULL);
if (err)
goto out;
memset (&syshd, 0, sizeof syshd);
err = es_create (&stream, cookie, &syshd, functions, modeflags,
samethread, 0);
if (err)
goto out;
out:
return stream;
}
static estream_t
do_fdopen (int filedes, const char *mode, int no_close, int with_locked_list)
{
unsigned int modeflags;
int samethread, create_called;
estream_t stream;
void *cookie;
int err;
es_syshd_t syshd;
stream = NULL;
cookie = NULL;
create_called = 0;
err = parse_mode (mode, &modeflags, &samethread, NULL);
if (err)
goto out;
err = func_fd_create (&cookie, filedes, modeflags, no_close);
if (err)
goto out;
syshd.type = ES_SYSHD_FD;
syshd.u.fd = filedes;
create_called = 1;
err = es_create (&stream, cookie, &syshd, estream_functions_fd,
modeflags, samethread, with_locked_list);
if (!err && stream)
{
stream->intern->func_ioctl = es_func_fd_ioctl;
if ((modeflags & O_NONBLOCK))
err = es_func_fd_ioctl (cookie, COOKIE_IOCTL_NONBLOCK, "", NULL);
}
out:
if (err && create_called)
(*estream_functions_fd.func_close) (cookie);
return stream;
}
estream_t
_gpgrt_fdopen (int filedes, const char *mode)
{
return do_fdopen (filedes, mode, 0, 0);
}
/* A variant of es_fdopen which does not close FILEDES at the end. */
estream_t
_gpgrt_fdopen_nc (int filedes, const char *mode)
{
return do_fdopen (filedes, mode, 1, 0);
}
static estream_t
do_fpopen (FILE *fp, const char *mode, int no_close, int with_locked_list)
{
unsigned int modeflags, cmode;
int samethread, create_called;
estream_t stream;
void *cookie;
int err;
es_syshd_t syshd;
stream = NULL;
cookie = NULL;
create_called = 0;
err = parse_mode (mode, &modeflags, &samethread, &cmode);
if (err)
goto out;
if (fp)
fflush (fp);
err = func_fp_create (&cookie, fp, modeflags, no_close);
if (err)
goto out;
syshd.type = ES_SYSHD_FD;
syshd.u.fd = fp? fileno (fp): -1;
create_called = 1;
err = es_create (&stream, cookie, &syshd, estream_functions_fp,
modeflags, samethread, with_locked_list);
out:
if (err && create_called)
(*estream_functions_fp.func_close) (cookie);
return stream;
}
/* Create an estream from the stdio stream FP. This mechanism is
useful in case the stdio streams have special properties and may
not be mixed with fd based functions. This is for example the case
under Windows where the 3 standard streams are associated with the
console whereas a duped and fd-opened stream of one of this stream
won't be associated with the console. As this messes things up it
is easier to keep on using the standard I/O stream as a backend for
estream. */
estream_t
_gpgrt_fpopen (FILE *fp, const char *mode)
{
return do_fpopen (fp, mode, 0, 0);
}
/* Same as es_fpopen but does not close FP at the end. */
estream_t
_gpgrt_fpopen_nc (FILE *fp, const char *mode)
{
return do_fpopen (fp, mode, 1, 0);
}
#ifdef HAVE_W32_SYSTEM
estream_t
do_w32open (HANDLE hd, const char *mode,
int no_close, int with_locked_list)
{
unsigned int modeflags, cmode;
int samethread;
int create_called = 0;
estream_t stream = NULL;
void *cookie = NULL;
int err;
es_syshd_t syshd;
err = parse_mode (mode, &modeflags, &samethread, &cmode);
if (err)
goto leave;
err = es_func_w32_create (&cookie, hd, modeflags, no_close);
if (err)
goto leave;
syshd.type = ES_SYSHD_HANDLE;
syshd.u.handle = hd;
create_called = 1;
err = es_create (&stream, cookie, &syshd, estream_functions_w32,
modeflags, samethread, with_locked_list);
leave:
if (err && create_called)
(*estream_functions_w32.func_close) (cookie);
return stream;
}
#endif /*HAVE_W32_SYSTEM*/
static estream_t
do_sysopen (es_syshd_t *syshd, const char *mode, int no_close)
{
estream_t stream;
switch (syshd->type)
{
case ES_SYSHD_FD:
case ES_SYSHD_SOCK:
stream = do_fdopen (syshd->u.fd, mode, no_close, 0);
break;
#ifdef HAVE_W32_SYSTEM
case ES_SYSHD_HANDLE:
stream = do_w32open (syshd->u.handle, mode, no_close, 0);
break;
#endif
/* FIXME: Support RVIDs under Wince? */
default:
_set_errno (EINVAL);
stream = NULL;
}
return stream;
}
/* On POSIX systems this function is an alias for es_fdopen. Under
Windows it uses the bare W32 API and thus a HANDLE instead of a
file descriptor. */
estream_t
_gpgrt_sysopen (es_syshd_t *syshd, const char *mode)
{
return do_sysopen (syshd, mode, 0);
}
/* Same as es_sysopen but the handle/fd will not be closed by
es_fclose. */
estream_t
_gpgrt_sysopen_nc (es_syshd_t *syshd, const char *mode)
{
return do_sysopen (syshd, mode, 1);
}
/* Set custom standard descriptors to be used for stdin, stdout and
stderr. This function needs to be called before any of the
standard streams are accessed. This internal version uses a double
dash inside its name. */
void
_gpgrt__set_std_fd (int no, int fd)
{
/* fprintf (stderr, "es_set_std_fd(%d, %d)\n", no, fd); */
lock_list ();
if (no >= 0 && no < 3 && !custom_std_fds_valid[no])
{
custom_std_fds[no] = fd;
custom_std_fds_valid[no] = 1;
}
unlock_list ();
}
/* Return the stream used for stdin, stdout or stderr.
This internal version uses a double dash inside its name. */
estream_t
_gpgrt__get_std_stream (int fd)
{
estream_list_t list_obj;
estream_t stream = NULL;
fd %= 3; /* We only allow 0, 1 or 2 but we don't want to return an error. */
lock_list ();
for (list_obj = estream_list; list_obj; list_obj = list_obj->next)
if (list_obj->stream && list_obj->stream->intern->is_stdstream
&& list_obj->stream->intern->stdstream_fd == fd)
{
stream = list_obj->stream;
break;
}
if (!stream)
{
/* Standard stream not yet created. We first try to create them
from registered file descriptors. */
if (!fd && custom_std_fds_valid[0])
stream = do_fdopen (custom_std_fds[0], "r", 1, 1);
else if (fd == 1 && custom_std_fds_valid[1])
stream = do_fdopen (custom_std_fds[1], "a", 1, 1);
else if (custom_std_fds_valid[2])
stream = do_fdopen (custom_std_fds[2], "a", 1, 1);
if (!stream)
{
/* Second try is to use the standard C streams. */
if (!fd)
stream = do_fpopen (stdin, "r", 1, 1);
else if (fd == 1)
stream = do_fpopen (stdout, "a", 1, 1);
else
stream = do_fpopen (stderr, "a", 1, 1);
}
if (!stream)
{
/* Last try: Create a bit bucket. */
stream = do_fpopen (NULL, fd? "a":"r", 0, 1);
if (!stream)
{
fprintf (stderr, "fatal: error creating a dummy estream"
" for %d: %s\n", fd, strerror (errno));
abort();
}
}
stream->intern->is_stdstream = 1;
stream->intern->stdstream_fd = fd;
if (fd == 2)
es_set_buffering (stream, NULL, _IOLBF, 0);
fname_set_internal (stream,
fd == 0? "[stdin]" :
fd == 1? "[stdout]" : "[stderr]", 0);
}
unlock_list ();
return stream;
}
/* Note: A "samethread" keyword given in "mode" is ignored and the
value used by STREAM is used instead. */
estream_t
_gpgrt_freopen (const char *_GPGRT__RESTRICT path,
const char *_GPGRT__RESTRICT mode,
estream_t _GPGRT__RESTRICT stream)
{
int err;
if (path)
{
unsigned int modeflags, cmode;
int dummy, samethread, create_called;
void *cookie;
int fd;
es_syshd_t syshd;
cookie = NULL;
create_called = 0;
samethread = stream->intern->samethread;
lock_stream (stream);
es_deinitialize (stream);
err = parse_mode (mode, &modeflags, &dummy, &cmode);
if (err)
goto leave;
(void)dummy;
err = func_file_create (&cookie, &fd, path, modeflags, cmode);
if (err)
goto leave;
syshd.type = ES_SYSHD_FD;
syshd.u.fd = fd;
create_called = 1;
init_stream_obj (stream, cookie, &syshd, estream_functions_fd,
modeflags, samethread);
leave:
if (err)
{
if (create_called)
es_func_fd_destroy (cookie);
do_close (stream, 0);
stream = NULL;
}
else
{
if (path)
fname_set_internal (stream, path, 1);
unlock_stream (stream);
}
}
else
{
/* FIXME? We don't support re-opening at the moment. */
_set_errno (EINVAL);
es_deinitialize (stream);
do_close (stream, 0);
stream = NULL;
}
return stream;
}
int
_gpgrt_fclose (estream_t stream)
{
int err;
err = do_close (stream, 0);
return err;
}
/* This is a special version of es_fclose which can be used with
es_fopenmem to return the memory buffer. This is feature is useful
to write to a memory buffer using estream. Note that the function
does not close the stream if the stream does not support snatching
the buffer. On error NULL is stored at R_BUFFER. Note that if no
write operation has happened, NULL may also be stored at BUFFER on
success. The caller needs to release the returned memory using
gpgrt_free. */
int
_gpgrt_fclose_snatch (estream_t stream, void **r_buffer, size_t *r_buflen)
{
int err;
/* Note: There is no need to lock the stream in a close call. The
object will be destroyed after the close and thus any other
contender for the lock would work on a closed stream. */
if (r_buffer)
{
cookie_ioctl_function_t func_ioctl = stream->intern->func_ioctl;
size_t buflen;
*r_buffer = NULL;
if (!func_ioctl)
{
_set_errno (EOPNOTSUPP);
err = -1;
goto leave;
}
if (stream->flags.writing)
{
err = es_flush (stream);
if (err)
goto leave;
stream->flags.writing = 0;
}
err = func_ioctl (stream->intern->cookie, COOKIE_IOCTL_SNATCH_BUFFER,
r_buffer, &buflen);
if (err)
goto leave;
if (r_buflen)
*r_buflen = buflen;
}
err = do_close (stream, 0);
leave:
if (err && r_buffer)
{
mem_free (*r_buffer);
*r_buffer = NULL;
}
return err;
}
/* Register or unregister a close notification function for STREAM.
FNC is the function to call and FNC_VALUE the value passed as
second argument. To register the notification the value for MODE
must be 1. If mode is 0 the function tries to remove or disable an
already registered notification; for this to work the value of FNC
and FNC_VALUE must be the same as with the registration and
FNC_VALUE must be a unique value. No error will be returned if
MODE is 0.
FIXME: I think the next comment is not anymore correct:
Unregister should only be used in the error case because it may not
be able to remove memory internally allocated for the onclose
handler.
FIXME: Unregister is not thread safe.
The notification will be called right before the stream is closed.
It may not call any estream function for STREAM, neither direct nor
indirectly. */
int
_gpgrt_onclose (estream_t stream, int mode,
void (*fnc) (estream_t, void*), void *fnc_value)
{
int err;
lock_stream (stream);
err = do_onclose (stream, mode, fnc, fnc_value);
unlock_stream (stream);
return err;
}
int
_gpgrt_fileno_unlocked (estream_t stream)
{
es_syshd_t syshd;
if (_gpgrt_syshd_unlocked (stream, &syshd))
return -1;
switch (syshd.type)
{
case ES_SYSHD_FD: return syshd.u.fd;
case ES_SYSHD_SOCK: return syshd.u.sock;
default:
_set_errno (EINVAL);
return -1;
}
}
/* Return the handle of a stream which has been opened by es_sysopen.
The caller needs to pass a structure which will be filled with the
sys handle. Return 0 on success or true on error and sets errno.
This is the unlocked version. */
int
_gpgrt_syshd_unlocked (estream_t stream, es_syshd_t *syshd)
{
if (!stream || !syshd || stream->intern->syshd.type == ES_SYSHD_NONE)
{
if (syshd)
syshd->type = ES_SYSHD_NONE;
_set_errno (EINVAL);
return -1;
}
*syshd = stream->intern->syshd;
return 0;
}
void
_gpgrt_flockfile (estream_t stream)
{
lock_stream (stream);
}
int
_gpgrt_ftrylockfile (estream_t stream)
{
return trylock_stream (stream);
}
void
_gpgrt_funlockfile (estream_t stream)
{
unlock_stream (stream);
}
int
_gpgrt_fileno (estream_t stream)
{
int ret;
lock_stream (stream);
ret = _gpgrt_fileno_unlocked (stream);
unlock_stream (stream);
return ret;
}
/* Return the handle of a stream which has been opened by es_sysopen.
The caller needs to pass a structure which will be filled with the
sys handle. Return 0 on success or true on error and sets errno.
This is the unlocked version. */
int
_gpgrt_syshd (estream_t stream, es_syshd_t *syshd)
{
int ret;
lock_stream (stream);
ret = _gpgrt_syshd_unlocked (stream, syshd);
unlock_stream (stream);
return ret;
}
int
_gpgrt__pending_unlocked (estream_t stream)
{
return check_pending (stream);
}
/* Return true if there is at least one byte pending for read on
STREAM. This does only work if the backend supports checking for
pending bytes and is thus mostly useful with cookie based backends.
Note that if this function is used with cookie based functions, the
read cookie may be called with 0 for the SIZE argument. If bytes
are pending the function is expected to return -1 in this case and
thus deviates from the standard behavior of read(2). */
int
_gpgrt__pending (estream_t stream)
{
int ret;
lock_stream (stream);
ret = _gpgrt__pending_unlocked (stream);
unlock_stream (stream);
return ret;
}
int
_gpgrt_feof_unlocked (estream_t stream)
{
return stream->intern->indicators.eof;
}
int
_gpgrt_feof (estream_t stream)
{
int ret;
lock_stream (stream);
ret = _gpgrt_feof_unlocked (stream);
unlock_stream (stream);
return ret;
}
int
_gpgrt_ferror_unlocked (estream_t stream)
{
return stream->intern->indicators.err;
}
int
_gpgrt_ferror (estream_t stream)
{
int ret;
lock_stream (stream);
ret = _gpgrt_ferror_unlocked (stream);
unlock_stream (stream);
return ret;
}
void
_gpgrt_clearerr_unlocked (estream_t stream)
{
stream->intern->indicators.eof = 0;
stream->intern->indicators.err = 0;
/* We do not reset the HUP indicator because there is no way to
get out of this state. */
}
void
_gpgrt_clearerr (estream_t stream)
{
lock_stream (stream);
_gpgrt_clearerr_unlocked (stream);
unlock_stream (stream);
}
static int
do_fflush (estream_t stream)
{
int err;
if (stream->flags.writing)
err = es_flush (stream);
else
{
es_empty (stream);
err = 0;
}
return err;
}
int
_gpgrt_fflush (estream_t stream)
{
int err;
if (stream)
{
lock_stream (stream);
err = do_fflush (stream);
unlock_stream (stream);
}
else
{
estream_list_t item;
err = 0;
lock_list ();
for (item = estream_list; item; item = item->next)
if (item->stream)
{
lock_stream (item->stream);
err |= do_fflush (item->stream);
unlock_stream (item->stream);
}
unlock_list ();
}
return err ? EOF : 0;
}
int
_gpgrt_fseek (estream_t stream, long int offset, int whence)
{
int err;
lock_stream (stream);
err = es_seek (stream, offset, whence, NULL);
unlock_stream (stream);
return err;
}
int
_gpgrt_fseeko (estream_t stream, gpgrt_off_t offset, int whence)
{
int err;
lock_stream (stream);
err = es_seek (stream, offset, whence, NULL);
unlock_stream (stream);
return err;
}
long int
_gpgrt_ftell (estream_t stream)
{
long int ret;
lock_stream (stream);
ret = es_offset_calculate (stream);
unlock_stream (stream);
return ret;
}
gpgrt_off_t
_gpgrt_ftello (estream_t stream)
{
gpgrt_off_t ret = -1;
lock_stream (stream);
ret = es_offset_calculate (stream);
unlock_stream (stream);
return ret;
}
void
_gpgrt_rewind (estream_t stream)
{
lock_stream (stream);
es_seek (stream, 0L, SEEK_SET, NULL);
/* Note that es_seek already cleared the EOF flag. */
stream->intern->indicators.err = 0;
unlock_stream (stream);
}
int
_gpgrt__getc_underflow (estream_t stream)
{
int err;
unsigned char c;
size_t bytes_read;
err = es_readn (stream, &c, 1, &bytes_read);
return (err || (! bytes_read)) ? EOF : c;
}
int
_gpgrt__putc_overflow (int c, estream_t stream)
{
unsigned char d = c;
int err;
err = es_writen (stream, &d, 1, NULL);
return err ? EOF : c;
}
int
_gpgrt_fgetc (estream_t stream)
{
int ret;
lock_stream (stream);
ret = _gpgrt_getc_unlocked (stream);
unlock_stream (stream);
return ret;
}
int
_gpgrt_fputc (int c, estream_t stream)
{
int ret;
lock_stream (stream);
ret = _gpgrt_putc_unlocked (c, stream);
unlock_stream (stream);
return ret;
}
int
_gpgrt_ungetc (int c, estream_t stream)
{
unsigned char data = (unsigned char) c;
size_t data_unread;
lock_stream (stream);
es_unreadn (stream, &data, 1, &data_unread);
unlock_stream (stream);
return data_unread ? c : EOF;
}
int
_gpgrt_read (estream_t _GPGRT__RESTRICT stream,
void *_GPGRT__RESTRICT buffer, size_t bytes_to_read,
size_t *_GPGRT__RESTRICT bytes_read)
{
int err;
if (bytes_to_read)
{
lock_stream (stream);
err = es_readn (stream, buffer, bytes_to_read, bytes_read);
unlock_stream (stream);
}
else
err = 0;
return err;
}
int
_gpgrt_write (estream_t _GPGRT__RESTRICT stream,
const void *_GPGRT__RESTRICT buffer, size_t bytes_to_write,
size_t *_GPGRT__RESTRICT bytes_written)
{
int err;
if (bytes_to_write)
{
lock_stream (stream);
err = es_writen (stream, buffer, bytes_to_write, bytes_written);
unlock_stream (stream);
}
else
err = 0;
return err;
}
size_t
_gpgrt_fread (void *_GPGRT__RESTRICT ptr, size_t size, size_t nitems,
estream_t _GPGRT__RESTRICT stream)
{
size_t ret, bytes;
if (size * nitems)
{
lock_stream (stream);
es_readn (stream, ptr, size * nitems, &bytes);
unlock_stream (stream);
ret = bytes / size;
}
else
ret = 0;
return ret;
}
size_t
_gpgrt_fwrite (const void *_GPGRT__RESTRICT ptr, size_t size, size_t nitems,
estream_t _GPGRT__RESTRICT stream)
{
size_t ret, bytes;
if (size * nitems)
{
lock_stream (stream);
es_writen (stream, ptr, size * nitems, &bytes);
unlock_stream (stream);
ret = bytes / size;
}
else
ret = 0;
return ret;
}
char *
_gpgrt_fgets (char *_GPGRT__RESTRICT buffer, int length,
estream_t _GPGRT__RESTRICT stream)
{
unsigned char *s = (unsigned char*)buffer;
int c;
if (!length)
return NULL;
c = EOF;
lock_stream (stream);
while (length > 1 && (c = _gpgrt_getc_unlocked (stream)) != EOF && c != '\n')
{
*s++ = c;
length--;
}
unlock_stream (stream);
if (c == EOF && s == (unsigned char*)buffer)
return NULL; /* Nothing read. */
if (c != EOF && length > 1)
*s++ = c;
*s = 0;
return buffer;
}
int
_gpgrt_fputs_unlocked (const char *_GPGRT__RESTRICT s,
estream_t _GPGRT__RESTRICT stream)
{
size_t length;
int err;
length = strlen (s);
err = es_writen (stream, s, length, NULL);
return err ? EOF : 0;
}
int
_gpgrt_fputs (const char *_GPGRT__RESTRICT s, estream_t _GPGRT__RESTRICT stream)
{
size_t length;
int err;
length = strlen (s);
lock_stream (stream);
err = es_writen (stream, s, length, NULL);
unlock_stream (stream);
return err ? EOF : 0;
}
gpgrt_ssize_t
_gpgrt_getline (char *_GPGRT__RESTRICT *_GPGRT__RESTRICT lineptr,
size_t *_GPGRT__RESTRICT n, estream_t _GPGRT__RESTRICT stream)
{
char *line = NULL;
size_t line_n = 0;
int err;
lock_stream (stream);
err = doreadline (stream, 0, &line, &line_n);
unlock_stream (stream);
if (err)
goto out;
if (*n)
{
/* Caller wants us to use his buffer. */
if (*n < (line_n + 1))
{
/* Provided buffer is too small -> resize. */
void *p;
p = mem_realloc (*lineptr, line_n + 1);
if (! p)
err = -1;
else
{
if (*lineptr != p)
*lineptr = p;
}
}
if (! err)
{
memcpy (*lineptr, line, line_n + 1);
if (*n != line_n)
*n = line_n;
}
mem_free (line);
}
else
{
/* Caller wants new buffers. */
*lineptr = line;
*n = line_n;
}
out:
return err ? err : (gpgrt_ssize_t)line_n;
}
/* Same as fgets() but if the provided buffer is too short a larger
one will be allocated. This is similar to getline. A line is
considered a byte stream ending in a LF.
If MAX_LENGTH is not NULL, it shall point to a value with the
maximum allowed allocation.
Returns the length of the line. EOF is indicated by a line of
length zero. A truncated line is indicated my setting the value at
MAX_LENGTH to 0. If the returned value is less then 0 not enough
memory was enable or another error occurred; ERRNO is then set
accordingly.
If a line has been truncated, the file pointer is moved forward to
the end of the line so that the next read starts with the next
line. Note that MAX_LENGTH must be re-initialzied in this case.
The caller initially needs to provide the address of a variable,
initialized to NULL, at ADDR_OF_BUFFER and don't change this value
anymore with the following invocations. LENGTH_OF_BUFFER should be
the address of a variable, initialized to 0, which is also
maintained by this function. Thus, both paramaters should be
considered the state of this function.
Note: The returned buffer is allocated with enough extra space to
allow the caller to append a CR,LF,Nul. The buffer should be
released using gpgrt_free.
*/
gpgrt_ssize_t
_gpgrt_read_line (estream_t stream,
char **addr_of_buffer, size_t *length_of_buffer,
size_t *max_length)
{
int c;
char *buffer = *addr_of_buffer;
size_t length = *length_of_buffer;
size_t nbytes = 0;
size_t maxlen = max_length? *max_length : 0;
char *p;
if (!buffer)
{
/* No buffer given - allocate a new one. */
length = 256;
buffer = mem_alloc (length);
*addr_of_buffer = buffer;
if (!buffer)
{
*length_of_buffer = 0;
if (max_length)
*max_length = 0;
return -1;
}
*length_of_buffer = length;
}
if (length < 4)
{
/* This should never happen. If it does, the function has been
called with wrong arguments. */
_set_errno (EINVAL);
return -1;
}
length -= 3; /* Reserve 3 bytes for CR,LF,EOL. */
lock_stream (stream);
p = buffer;
while ((c = _gpgrt_getc_unlocked (stream)) != EOF)
{
if (nbytes == length)
{
/* Enlarge the buffer. */
if (maxlen && length > maxlen)
{
/* We are beyond our limit: Skip the rest of the line. */
while (c != '\n' && (c=_gpgrt_getc_unlocked (stream)) != EOF)
;
*p++ = '\n'; /* Always append a LF (we reserved some space). */
nbytes++;
if (max_length)
*max_length = 0; /* Indicate truncation. */
break; /* the while loop. */
}
length += 3; /* Adjust for the reserved bytes. */
length += length < 1024? 256 : 1024;
*addr_of_buffer = mem_realloc (buffer, length);
if (!*addr_of_buffer)
{
int save_errno = errno;
mem_free (buffer);
*length_of_buffer = 0;
if (max_length)
*max_length = 0;
unlock_stream (stream);
_set_errno (save_errno);
return -1;
}
buffer = *addr_of_buffer;
*length_of_buffer = length;
length -= 3;
p = buffer + nbytes;
}
*p++ = c;
nbytes++;
if (c == '\n')
break;
}
*p = 0; /* Make sure the line is a string. */
unlock_stream (stream);
return nbytes;
}
/* Wrapper around free() to match the memory allocation system used by
estream. Should be used for all buffers returned to the caller by
libestream. If a custom allocation handler has been set with
gpgrt_set_alloc_func that register function may be used
instead. This function has been moved to init.c. */
/* void */
/* _gpgrt_free (void *a) */
/* { */
/* mem_free (a); */
/* } */
int
_gpgrt_vfprintf_unlocked (estream_t _GPGRT__RESTRICT stream,
const char *_GPGRT__RESTRICT format,
va_list ap)
{
return es_print (stream, format, ap);
}
int
_gpgrt_vfprintf (estream_t _GPGRT__RESTRICT stream,
const char *_GPGRT__RESTRICT format,
va_list ap)
{
int ret;
lock_stream (stream);
ret = es_print (stream, format, ap);
unlock_stream (stream);
return ret;
}
int
_gpgrt_fprintf_unlocked (estream_t _GPGRT__RESTRICT stream,
const char *_GPGRT__RESTRICT format, ...)
{
int ret;
va_list ap;
va_start (ap, format);
ret = es_print (stream, format, ap);
va_end (ap);
return ret;
}
int
_gpgrt_fprintf (estream_t _GPGRT__RESTRICT stream,
const char *_GPGRT__RESTRICT format, ...)
{
int ret;
va_list ap;
va_start (ap, format);
lock_stream (stream);
ret = es_print (stream, format, ap);
unlock_stream (stream);
va_end (ap);
return ret;
}
static int
tmpfd (void)
{
#ifdef HAVE_W32_SYSTEM
int attempts, n;
#ifdef HAVE_W32CE_SYSTEM
wchar_t buffer[MAX_PATH+9+12+1];
# define mystrlen(a) wcslen (a)
wchar_t *name, *p;
#else
char buffer[MAX_PATH+9+12+1];
# define mystrlen(a) strlen (a)
char *name, *p;
#endif
HANDLE file;
int pid = GetCurrentProcessId ();
unsigned int value;
int i;
n = GetTempPath (MAX_PATH+1, buffer);
if (!n || n > MAX_PATH || mystrlen (buffer) > MAX_PATH)
{
_set_errno (ENOENT);
return -1;
}
p = buffer + mystrlen (buffer);
#ifdef HAVE_W32CE_SYSTEM
wcscpy (p, L"_estream");
#else
strcpy (p, "_estream");
#endif
p += 8;
/* We try to create the directory but don't care about an error as
it may already exist and the CreateFile would throw an error
anyway. */
CreateDirectory (buffer, NULL);
*p++ = '\\';
name = p;
for (attempts=0; attempts < 10; attempts++)
{
p = name;
value = (GetTickCount () ^ ((pid<<16) & 0xffff0000));
for (i=0; i < 8; i++)
{
*p++ = tohex (((value >> 28) & 0x0f));
value <<= 4;
}
#ifdef HAVE_W32CE_SYSTEM
wcscpy (p, L".tmp");
#else
strcpy (p, ".tmp");
#endif
file = CreateFile (buffer,
GENERIC_READ | GENERIC_WRITE,
0,
NULL,
CREATE_NEW,
FILE_ATTRIBUTE_TEMPORARY | FILE_FLAG_DELETE_ON_CLOSE,
NULL);
if (file != INVALID_HANDLE_VALUE)
{
#ifdef HAVE_W32CE_SYSTEM
int fd = (int)file;
#else
int fd = _open_osfhandle ((long)file, 0);
if (fd == -1)
{
CloseHandle (file);
return -1;
}
#endif
return fd;
}
Sleep (1); /* One ms as this is the granularity of GetTickCount. */
}
_set_errno (ENOENT);
return -1;
#else /*!HAVE_W32_SYSTEM*/
FILE *fp;
int fp_fd;
int fd;
fp = NULL;
fd = -1;
fp = tmpfile ();
if (! fp)
goto out;
fp_fd = fileno (fp);
fd = dup (fp_fd);
out:
if (fp)
fclose (fp);
return fd;
#endif /*!HAVE_W32_SYSTEM*/
}
estream_t
_gpgrt_tmpfile (void)
{
unsigned int modeflags;
int create_called;
estream_t stream;
void *cookie;
int err;
int fd;
es_syshd_t syshd;
create_called = 0;
stream = NULL;
modeflags = O_RDWR | O_TRUNC | O_CREAT;
cookie = NULL;
fd = tmpfd ();
if (fd == -1)
{
err = -1;
goto out;
}
err = func_fd_create (&cookie, fd, modeflags, 0);
if (err)
goto out;
syshd.type = ES_SYSHD_FD;
syshd.u.fd = fd;
create_called = 1;
err = es_create (&stream, cookie, &syshd, estream_functions_fd, modeflags,
0, 0);
out:
if (err)
{
if (create_called)
es_func_fd_destroy (cookie);
else if (fd != -1)
close (fd);
stream = NULL;
}
return stream;
}
int
_gpgrt_setvbuf (estream_t _GPGRT__RESTRICT stream,
char *_GPGRT__RESTRICT buf, int type, size_t size)
{
int err;
if ((type == _IOFBF || type == _IOLBF || type == _IONBF)
&& (!buf || size || type == _IONBF))
{
lock_stream (stream);
err = es_set_buffering (stream, buf, type, size);
unlock_stream (stream);
}
else
{
_set_errno (EINVAL);
err = -1;
}
return err;
}
/* Put a stream into binary mode. This is only needed for the
standard streams if they are to be used in a binary way. On Unix
systems it is never needed but MSDOS based systems require such a
call. It needs to be called before any I/O is done on STREAM. */
void
_gpgrt_set_binary (estream_t stream)
{
lock_stream (stream);
if (!(stream->intern->modeflags & O_BINARY))
{
stream->intern->modeflags |= O_BINARY;
#ifdef HAVE_DOSISH_SYSTEM
if (stream->intern->func_read == es_func_fd_read)
{
estream_cookie_fd_t fd_cookie = stream->intern->cookie;
if (!IS_INVALID_FD (fd_cookie->fd))
setmode (fd_cookie->fd, O_BINARY);
}
else if (stream->intern->func_read == es_func_fp_read)
{
estream_cookie_fp_t fp_cookie = stream->intern->cookie;
if (fp_cookie->fp)
setmode (fileno (fp_cookie->fp), O_BINARY);
}
#endif
}
unlock_stream (stream);
}
/* Set non-blocking mode for STREAM. Use true for ONOFF to enable and
false to disable non-blocking mode. Returns 0 on success or -1 on
error and sets ERRNO. Note that not all backends support
non-blocking mode.
In non-blocking mode a system call will not block but return an
error and set errno to EAGAIN. The estream API always uses EAGAIN
and not EWOULDBLOCK. If a buffered function like es_fgetc() or
es_fgets() returns an error and both, feof() and ferror() return
false the caller may assume that the error condition was EAGAIN.
Switching back from non-blocking to blocking may raise problems
with buffering, thus care should be taken. Although read+write
sockets are supported in theory, switching from write to read may
result into problems because estream may first flush the write
buffers and there is no way to handle that non-blocking (EAGAIN)
case. Explicit flushing should thus be done before before
switching to read. */
int
_gpgrt_set_nonblock (estream_t stream, int onoff)
{
cookie_ioctl_function_t func_ioctl;
int ret;
lock_stream (stream);
func_ioctl = stream->intern->func_ioctl;
if (!func_ioctl)
{
_set_errno (EOPNOTSUPP);
ret = -1;
}
else
{
unsigned int save_flags = stream->intern->modeflags;
if (onoff)
stream->intern->modeflags |= O_NONBLOCK;
else
stream->intern->modeflags &= ~O_NONBLOCK;
ret = func_ioctl (stream->intern->cookie, COOKIE_IOCTL_NONBLOCK,
onoff?"":NULL, NULL);
if (ret)
stream->intern->modeflags = save_flags;
}
unlock_stream (stream);
return ret;
}
/* Return true if STREAM is in non-blocking mode. */
int
_gpgrt_get_nonblock (estream_t stream)
{
int ret;
lock_stream (stream);
ret = !!(stream->intern->modeflags & O_NONBLOCK);
unlock_stream (stream);
return ret;
}
/* A version of poll(2) working on estream handles. Note that not all
estream types work with this function. In contrast to the standard
poll function the gpgrt_poll_t object uses a set of names bit flags
instead of the EVENTS and REVENTS members. An item with the IGNORE
flag set is entirely ignored. The TIMEOUT values is given in
milliseconds, a value of -1 waits indefinitely, and a value of 0
returns immediately.
A positive return value gives the number of fds with new
information. A return value of 0 indicates a timeout and -1
indicates an error in which case ERRNO is set. */
int
_gpgrt_poll (gpgrt_poll_t *fds, unsigned int nfds, int timeout)
{
gpgrt_poll_t *item;
int count = 0;
fd_set readfds, writefds, exceptfds;
int any_readfd, any_writefd, any_exceptfd;
int idx;
int max_fd;
int fd, ret, any;
if (!fds)
{
_set_errno (EINVAL);
return -1;
}
/* Clear all response fields (even for ignored items). */
for (item = fds, idx = 0; idx < nfds; item++, idx++)
{
item->got_read = 0;
item->got_write = 0;
item->got_oob = 0;
item->got_rdhup = 0;
item->got_err = 0;
item->got_hup = 0;
item->got_nval = 0;
}
/* Check for pending reads. */
for (item = fds, idx = 0; idx < nfds; item++, idx++)
{
if (item->ignore)
continue;
if (!item->want_read)
continue;
if (_gpgrt__pending (item->stream))
{
item->got_read = 1;
count++;
}
}
/* Check for space in the write buffers. */
for (item = fds, idx = 0; idx < nfds; item++, idx++)
{
if (item->ignore)
continue;
if (!item->want_write)
continue;
/* FIXME */
}
if (count)
return count; /* Early return without waiting. */
/* Now do the real select. */
any_readfd = any_writefd = any_exceptfd = 0;
max_fd = 0;
for (item = fds, idx = 0; idx < nfds; item++, idx++)
{
if (item->ignore)
continue;
fd = _gpgrt_fileno (item->stream);
if (fd == -1)
continue; /* Stream does not support polling. */
if (item->want_read)
{
if (!any_readfd)
{
FD_ZERO (&readfds);
any_readfd = 1;
}
FD_SET (fd, &readfds);
if (fd > max_fd)
max_fd = fd;
}
if (item->want_write)
{
if (!any_writefd)
{
FD_ZERO (&writefds);
any_writefd = 1;
}
FD_SET (fd, &writefds);
if (fd > max_fd)
max_fd = fd;
}
if (item->want_oob)
{
if (!any_exceptfd)
{
FD_ZERO (&exceptfds);
any_exceptfd = 1;
}
FD_SET (fd, &exceptfds);
if (fd > max_fd)
max_fd = fd;
}
}
#ifdef _WIN32
(void)timeout;
ret = -1;
_set_errno (EOPNOTSUPP);
#else
if (pre_syscall_func)
pre_syscall_func ();
do
{
struct timeval timeout_val;
timeout_val.tv_sec = timeout / 1000;
timeout_val.tv_usec = (timeout % 1000) * 1000;
ret = select (max_fd+1,
any_readfd? &readfds : NULL,
any_writefd? &writefds : NULL,
any_exceptfd? &exceptfds : NULL,
timeout == -1 ? NULL : &timeout_val);
}
while (ret == -1 && errno == EINTR);
if (post_syscall_func)
post_syscall_func ();
#endif
if (ret == -1)
return -1;
if (!ret)
return 0; /* Timeout. Note that in this case we can't return
got_err for an invalid stream. */
for (item = fds, idx = 0; idx < nfds; item++, idx++)
{
if (item->ignore)
continue;
fd = _gpgrt_fileno (item->stream);
if (fd == -1)
{
item->got_err = 1; /* Stream does not support polling. */
count++;
continue;
}
any = 0;
if (item->stream->intern->indicators.hup)
{
item->got_hup = 1;
any = 1;
}
if (item->want_read && FD_ISSET (fd, &readfds))
{
item->got_read = 1;
any = 1;
}
if (item->want_write && FD_ISSET (fd, &writefds))
{
item->got_write = 1;
any = 1;
}
if (item->want_oob && FD_ISSET (fd, &exceptfds))
{
item->got_oob = 1;
any = 1;
}
if (any)
count++;
}
return count;
}
void
_gpgrt_opaque_set (estream_t stream, void *opaque)
{
lock_stream (stream);
es_opaque_ctrl (stream, opaque, NULL);
unlock_stream (stream);
}
void *
_gpgrt_opaque_get (estream_t stream)
{
void *opaque;
lock_stream (stream);
es_opaque_ctrl (stream, NULL, &opaque);
unlock_stream (stream);
return opaque;
}
static void
fname_set_internal (estream_t stream, const char *fname, int quote)
{
if (stream->intern->printable_fname
&& !stream->intern->printable_fname_inuse)
{
mem_free (stream->intern->printable_fname);
stream->intern->printable_fname = NULL;
}
if (stream->intern->printable_fname)
return; /* Can't change because it is in use. */
if (*fname != '[')
quote = 0;
else
quote = !!quote;
stream->intern->printable_fname = mem_alloc (strlen (fname) + quote + 1);
if (quote)
stream->intern->printable_fname[0] = '\\';
strcpy (stream->intern->printable_fname+quote, fname);
}
/* Set the filename attribute of STREAM. There is no error return.
as long as STREAM is valid. This function is called internally by
functions which open a filename. */
void
_gpgrt_fname_set (estream_t stream, const char *fname)
{
if (fname)
{
lock_stream (stream);
fname_set_internal (stream, fname, 1);
unlock_stream (stream);
}
}
/* Return the filename attribute of STREAM. In case no filename has
been set, "[?]" will be returned. The returned file name is valid
as long as STREAM is valid. */
const char *
_gpgrt_fname_get (estream_t stream)
{
const char *fname;
lock_stream (stream);
fname = stream->intern->printable_fname;
if (fname)
stream->intern->printable_fname_inuse = 1;
unlock_stream (stream);
if (!fname)
fname = "[?]";
return fname;
}
/* Print a BUFFER to STREAM while replacing all control characters and
the characters in DELIMITERS by standard C escape sequences.
Returns 0 on success or -1 on error. If BYTES_WRITTEN is not NULL
the number of bytes actually written are stored at this
address. */
int
_gpgrt_write_sanitized (estream_t _GPGRT__RESTRICT stream,
const void * _GPGRT__RESTRICT buffer, size_t length,
const char * delimiters,
size_t * _GPGRT__RESTRICT bytes_written)
{
const unsigned char *p = buffer;
size_t count = 0;
int ret;
lock_stream (stream);
for (; length; length--, p++, count++)
{
if (*p < 0x20
|| *p == 0x7f
|| (delimiters
&& (strchr (delimiters, *p) || *p == '\\')))
{
_gpgrt_putc_unlocked ('\\', stream);
count++;
if (*p == '\n')
{
_gpgrt_putc_unlocked ('n', stream);
count++;
}
else if (*p == '\r')
{
_gpgrt_putc_unlocked ('r', stream);
count++;
}
else if (*p == '\f')
{
_gpgrt_putc_unlocked ('f', stream);
count++;
}
else if (*p == '\v')
{
_gpgrt_putc_unlocked ('v', stream);
count++;
}
else if (*p == '\b')
{
_gpgrt_putc_unlocked ('b', stream);
count++;
}
else if (!*p)
{
_gpgrt_putc_unlocked('0', stream);
count++;
}
else
{
_gpgrt_fprintf_unlocked (stream, "x%02x", *p);
count += 3;
}
}
else
{
_gpgrt_putc_unlocked (*p, stream);
count++;
}
}
if (bytes_written)
*bytes_written = count;
ret = _gpgrt_ferror_unlocked (stream)? -1 : 0;
unlock_stream (stream);
return ret;
}
/* Write LENGTH bytes of BUFFER to STREAM as a hex encoded string.
RESERVED must be 0. Returns 0 on success or -1 on error. If
BYTES_WRITTEN is not NULL the number of bytes actually written are
stored at this address. */
int
_gpgrt_write_hexstring (estream_t _GPGRT__RESTRICT stream,
const void *_GPGRT__RESTRICT buffer, size_t length,
int reserved, size_t *_GPGRT__RESTRICT bytes_written )
{
int ret;
const unsigned char *s;
size_t count = 0;
(void)reserved;
#define tohex(n) ((n) < 10 ? ((n) + '0') : (((n) - 10) + 'A'))
if (!length)
return 0;
lock_stream (stream);
for (s = buffer; length; s++, length--)
{
_gpgrt_putc_unlocked ( tohex ((*s>>4)&15), stream);
_gpgrt_putc_unlocked ( tohex (*s&15), stream);
count += 2;
}
if (bytes_written)
*bytes_written = count;
ret = _gpgrt_ferror_unlocked (stream)? -1 : 0;
unlock_stream (stream);
return ret;
#undef tohex
}
diff --git a/src/estream.h b/src/estream.h
index cfb7eec..91f2bc0 100644
--- a/src/estream.h
+++ b/src/estream.h
@@ -1,30 +1,30 @@
/* gpgrt-int.h - Internal definitions
* Copyright (C) 2014 g10 Code GmbH
*
* This file is part of libgpg-error.
*
* libgpg-error is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public License
* as published by the Free Software Foundation; either version 2.1 of
* the License, or (at your option) any later version.
*
* libgpg-error is distributed in the hope that it will be useful, but
* WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
- * License along with this program; if not, see .
+ * License along with this program; if not, see .
*/
#ifndef ESTREAM_H
#define ESTREAM_H
#include "gpg-error.h"
/* Local prototypes for estream. */
int _gpgrt_es_init (void);
#endif /*ESTREAM_H*/
diff --git a/src/gen-posix-lock-obj.c b/src/gen-posix-lock-obj.c
index 595d379..22de456 100644
--- a/src/gen-posix-lock-obj.c
+++ b/src/gen-posix-lock-obj.c
@@ -1,163 +1,163 @@
/* gen-posix-lock-obj.c - Build tool to construct the lock object.
Copyright (C) 2014 g10 Code GmbH
This file is part of libgpg-error.
libgpg-error is free software; you can redistribute it and/or
modify it under the terms of the GNU Lesser General Public License
as published by the Free Software Foundation; either version 2.1 of
the License, or (at your option) any later version.
libgpg-error is distributed in the hope that it will be useful, but
WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public
- License along with this program; if not, see .
+ License along with this program; if not, see .
*/
#if HAVE_CONFIG_H
#include
#endif
#ifdef HAVE_W32_SYSTEM
# error This module may not be build for Windows.
#endif
#include
#include
#include
#include
#include
#include "posix-lock-obj.h"
#define PGM "gen-posix-lock-obj"
/* Check that configure did its job. */
#ifdef USE_POSIX_THREADS
#if SIZEOF_PTHREAD_MUTEX_T < 4
# error sizeof pthread_mutex_t is not known.
#endif
#endif
/* Special requirements for certain platforms. */
#if defined(__hppa__) && defined(__linux__)
# define USE_16BYTE_ALIGNMENT 1
#else
# define USE_16BYTE_ALIGNMENT 0
#endif
#if USE_16BYTE_ALIGNMENT && !HAVE_GCC_ATTRIBUTE_ALIGNED
# error compiler is not able to enforce a 16 byte alignment
#endif
#ifdef USE_POSIX_THREADS
static pthread_mutex_t mtx = PTHREAD_MUTEX_INITIALIZER;
#endif
int
main (void)
{
#ifdef USE_POSIX_THREADS
unsigned char *p;
int i;
#endif
struct {
long vers;
#ifdef USE_POSIX_THREADS
pthread_mutex_t mtx;
#endif
} dummyobj;
#ifdef USE_POSIX_THREADS
if (sizeof mtx != SIZEOF_PTHREAD_MUTEX_T)
{
fprintf (stderr, PGM ": pthread_mutex_t mismatch\n");
exit (1);
}
#endif /*USE_POSIX_THREADS*/
if (sizeof (dummyobj) != sizeof (_gpgrt_lock_t))
{
fprintf (stderr, PGM ": internal and external lock object mismatch\n");
exit (1);
}
printf ("## lock-obj-pub.%s.h%s\n"
"## File created by " PGM " - DO NOT EDIT\n"
"## To be included by mkheader into gpg-error.h\n"
"\n",
HOST_TRIPLET_STRING,
#ifdef USE_POSIX_THREADS
""
#else
" - NO LOCK SUPPORT"
#endif
);
#ifdef USE_POSIX_THREADS
/* To force a probably suitable alignment of the structure we use a
union and include a long and a pointer to a long. */
printf ("typedef struct\n"
"{\n"
" long _vers;\n"
" union {\n"
" volatile char _priv[%d];\n"
"%s"
" long _x_align;\n"
" long *_xp_align;\n"
" } u;\n"
"} gpgrt_lock_t;\n"
"\n"
"#define GPGRT_LOCK_INITIALIZER {%d,{{",
SIZEOF_PTHREAD_MUTEX_T,
# if USE_16BYTE_ALIGNMENT
" int _x16_align __attribute__ ((aligned (16)));\n",
# else
"",
# endif
LOCK_ABI_VERSION);
p = (unsigned char *)&mtx;
for (i=0; i < sizeof mtx; i++)
{
if (i && !(i % 8))
printf (" \\\n%*s", 36, "");
printf ("%u", p[i]);
if (i < sizeof mtx - 1)
putchar (',');
}
fputs ("}}}\n", stdout);
#else /*!USE_POSIX_THREADS*/
printf ("/* Dummy object - no locking available. */\n"
"typedef struct\n"
"{\n"
" long _vers;\n"
"} gpgrt_lock_t;\n"
"\n"
"#define GPGRT_LOCK_INITIALIZER {%d}\n",
LOCK_ABI_VERSION);
#endif /*!USE_POSIX_THREADS*/
fputs ("##\n"
"## Loc" "al Variables:\n"
"## mode: c\n"
"## buffer-read-only: t\n"
"## End:\n"
"##\n", stdout);
if (ferror (stdout))
{
fprintf (stderr, PGM ": error writing to stdout: %s\n", strerror (errno));
return 1;
}
return 0;
}
diff --git a/src/gen-w32-lock-obj.c b/src/gen-w32-lock-obj.c
index 9e49d1e..f8da67f 100644
--- a/src/gen-w32-lock-obj.c
+++ b/src/gen-w32-lock-obj.c
@@ -1,55 +1,55 @@
/* gen-w32-lock-obj.c - Build tool to get the size of the lock object.
Copyright (C) 2014 g10 Code GmbH
This file is part of libgpg-error.
libgpg-error is free software; you can redistribute it and/or
modify it under the terms of the GNU Lesser General Public License
as published by the Free Software Foundation; either version 2.1 of
the License, or (at your option) any later version.
libgpg-error is distributed in the hope that it will be useful, but
WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public
- License along with this program; if not, see .
+ License along with this program; if not, see .
*/
#if HAVE_CONFIG_H
#include
#endif
#ifndef HAVE_W32_SYSTEM
# error This module may only be build for Windows.
#endif
#include
#include
#include
#include "w32-lock-obj.h"
int
main (void)
{
_gpgrt_lock_t lk;
unsigned char *p;
int i;
printf ("sizeof CRITICAL_SECTION = %u\n", (int)sizeof (CRITICAL_SECTION));
printf ("sizeof _gpgrt_lock_t = %u\n", (int)sizeof lk);
memset (&lk, 0, sizeof lk);
lk.vers = LOCK_ABI_VERSION;
lk.started = -1;
printf ("#define GPGRT_LOCK_INITIALIZER {");
p = (unsigned char *)&lk;
for (i=0; i < sizeof lk - 1; i++)
printf ("%u,", p[i]);
printf ("%u}\n", p[sizeof(lk)-1]);
return 0;
}
diff --git a/src/gpg-error.def.in b/src/gpg-error.def.in
index 16de809..f8b9ebc 100644
--- a/src/gpg-error.def.in
+++ b/src/gpg-error.def.in
@@ -1,147 +1,147 @@
/* libgpg-error.def - Exported symbols for W32
* Copyright (C) 2014 g10 Code GmbH
*
* This file is part of libgpg-error.
*
* libgpg-error is free software; you can redistribute it and/or modify
* it under the terms of the GNU Lesser General Public License as
* published by the Free Software Foundation; either version 2.1 of
* the License, or (at your option) any later version.
*
* libgpg-error is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
- * License along with this program; if not, see .
+ * License along with this program; if not, see .
*
* Note: This file should be updated manually and the ordinals shall
* never be changed. Also check gpg-error.vers and visibility.h.
*
* This file needs to be pre-processed.
*/
#include
EXPORTS
gpg_strerror @1
gpg_strerror_r @2
gpg_strsource @3
gpg_err_code_from_errno @4
gpg_err_code_to_errno @5
/* @6 - Not anymore used. */
gpg_err_code_from_syserror @7
gpg_err_set_errno @8
#ifdef HAVE_W32CE_SYSTEM
_gpg_w32ce_get_errno @9
_gpg_w32ce_strerror @10
#endif
#ifdef HAVE_W32_SYSTEM
_gpg_w32_bindtextdomain @11
_gpg_w32_textdomain @12
_gpg_w32_gettext @13
_gpg_w32_dgettext @14
_gpg_w32_dngettext @15
_gpg_w32_gettext_localename @16
_gpg_w32_gettext_use_utf8 @17
#endif
/* @18 - Not anymore used. */
gpg_error_check_version @19
gpgrt_lock_init @20
gpgrt_lock_lock @21
gpgrt_lock_unlock @22
gpgrt_lock_destroy @23
gpgrt_yield @24
gpgrt_lock_trylock @25
gpgrt_set_syscall_clamp @26
gpgrt_fopen @27
gpgrt_mopen @28
gpgrt_fopenmem @29
gpgrt_fopenmem_init @30
gpgrt_fdopen @31
gpgrt_fdopen_nc @32
gpgrt_sysopen @33
gpgrt_sysopen_nc @34
gpgrt_fpopen @35
gpgrt_fpopen_nc @36
gpgrt_freopen @37
gpgrt_fopencookie @38
gpgrt_fclose @39
gpgrt_fclose_snatch @40
gpgrt_onclose @41
gpgrt_fileno @42
gpgrt_fileno_unlocked @43
gpgrt_syshd @44
gpgrt_syshd_unlocked @45
_gpgrt_set_std_fd @46
_gpgrt_get_std_stream @47
gpgrt_flockfile @48
gpgrt_ftrylockfile @49
gpgrt_funlockfile @50
gpgrt_feof @51
gpgrt_feof_unlocked @52
gpgrt_ferror @53
gpgrt_ferror_unlocked @54
gpgrt_clearerr @55
gpgrt_clearerr_unlocked @56
gpgrt_fflush @57
gpgrt_fseek @58
gpgrt_fseeko @59
gpgrt_ftell @60
gpgrt_ftello @61
gpgrt_rewind @62
gpgrt_fgetc @63
_gpgrt_getc_underflow @64
gpgrt_fputc @65
_gpgrt_putc_overflow @66
gpgrt_ungetc @67
gpgrt_read @68
gpgrt_write @69
gpgrt_write_sanitized @70
gpgrt_write_hexstring @71
gpgrt_fread @72
gpgrt_fwrite @73
gpgrt_fgets @74
gpgrt_fputs @75
gpgrt_fputs_unlocked @76
gpgrt_getline @77
gpgrt_read_line @78
gpgrt_free @79
gpgrt_fprintf @80
gpgrt_fprintf_unlocked @81
gpgrt_printf @82
gpgrt_printf_unlocked @83
gpgrt_vfprintf @84
gpgrt_vfprintf_unlocked @85
gpgrt_setvbuf @86
gpgrt_setbuf @87
gpgrt_set_binary @88
gpgrt_tmpfile @89
gpgrt_opaque_set @90
gpgrt_opaque_get @91
gpgrt_fname_set @92
gpgrt_fname_get @93
gpgrt_asprintf @94
gpgrt_vasprintf @95
gpgrt_bsprintf @96
gpgrt_vbsprintf @97
gpgrt_snprintf @98
gpgrt_vsnprintf @99
gpgrt_check_version @100
gpg_err_init @101
gpg_err_deinit @102
gpgrt_set_alloc_func @103
_gpgrt_pending @104
_gpgrt_pending_unlocked @105
gpgrt_set_nonblock @106
gpgrt_get_nonblock @107
gpgrt_poll @108
;; end of file with public symbols for Windows.
diff --git a/src/gpg-error.h.in b/src/gpg-error.h.in
index 28581e0..b32b4c4 100644
--- a/src/gpg-error.h.in
+++ b/src/gpg-error.h.in
@@ -1,827 +1,827 @@
/* gpg-error.h - Public interface to libgpg-error. -*- c -*-
* Copyright (C) 2003, 2004, 2010, 2013, 2014, 2015 g10 Code GmbH
*
* This file is part of libgpg-error.
*
* libgpg-error is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public License
* as published by the Free Software Foundation; either version 2.1 of
* the License, or (at your option) any later version.
*
* libgpg-error is distributed in the hope that it will be useful, but
* WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
- * License along with this program; if not, see .
+ * License along with this program; if not, see .
*
* @configure_input@
*/
#ifndef GPG_ERROR_H
#define GPG_ERROR_H 1
#include
#include
#include
/* The version string of this header. */
#define GPG_ERROR_VERSION @version@
#define GPGRT_VERSION @version@
/* The version number of this header. */
#define GPG_ERROR_VERSION_NUMBER @version-number@
#define GPGRT_VERSION_NUMBER @version-number@
#ifdef __GNUC__
# define GPG_ERR_INLINE __inline__
#elif defined(_MSC_VER) && _MSC_VER >= 1300
# define GPG_ERR_INLINE __inline
#elif defined (__STDC_VERSION__) && __STDC_VERSION__ >= 199901L
# define GPG_ERR_INLINE inline
#else
# ifndef GPG_ERR_INLINE
# define GPG_ERR_INLINE
# endif
#endif
#ifdef __cplusplus
extern "C" {
#if 0 /* just to make Emacs auto-indent happy */
}
#endif
#endif /* __cplusplus */
/* The GnuPG project consists of many components. Error codes are
exchanged between all components. The common error codes and their
user-presentable descriptions are kept into a shared library to
allow adding new error codes and components without recompiling any
of the other components. The interface will not change in a
backward incompatible way.
An error code together with an error source build up an error
value. As the error value is been passed from one component to
another, it preserver the information about the source and nature
of the error.
A component of the GnuPG project can define the following macros to
tune the behaviour of the library:
GPG_ERR_SOURCE_DEFAULT: Define to an error source of type
gpg_err_source_t to make that source the default for gpg_error().
Otherwise GPG_ERR_SOURCE_UNKNOWN is used as default.
GPG_ERR_ENABLE_GETTEXT_MACROS: Define to provide macros to map the
internal gettext API to standard names. This has only an effect on
Windows platforms.
GPGRT_ENABLE_ES_MACROS: Define to provide "es_" macros for the
estream functions.
In addition to the error codes, Libgpg-error also provides a set of
functions used by most GnuPG components. */
/* The error source type gpg_err_source_t.
Where as the Poo out of a welle small
Taketh his firste springing and his sours.
--Chaucer. */
/* Only use free slots, never change or reorder the existing
entries. */
typedef enum
{
@include:err-sources@
/* This is one more than the largest allowed entry. */
GPG_ERR_SOURCE_DIM = 128
} gpg_err_source_t;
/* The error code type gpg_err_code_t. */
/* Only use free slots, never change or reorder the existing
entries. */
typedef enum
{
@include:err-codes@
/* The following error codes are used to map system errors. */
#define GPG_ERR_SYSTEM_ERROR (1 << 15)
@include:errnos@
/* This is one more than the largest allowed entry. */
GPG_ERR_CODE_DIM = 65536
} gpg_err_code_t;
/* The error value type gpg_error_t. */
/* We would really like to use bit-fields in a struct, but using
structs as return values can cause binary compatibility issues, in
particular if you want to do it efficiently (also see
-freg-struct-return option to GCC). */
typedef unsigned int gpg_error_t;
/* We use the lowest 16 bits of gpg_error_t for error codes. The 16th
bit indicates system errors. */
#define GPG_ERR_CODE_MASK (GPG_ERR_CODE_DIM - 1)
/* Bits 17 to 24 are reserved. */
/* We use the upper 7 bits of gpg_error_t for error sources. */
#define GPG_ERR_SOURCE_MASK (GPG_ERR_SOURCE_DIM - 1)
#define GPG_ERR_SOURCE_SHIFT 24
/* The highest bit is reserved. It shouldn't be used to prevent
potential negative numbers when transmitting error values as
text. */
/* GCC feature test. */
#if __GNUC__
# define _GPG_ERR_GCC_VERSION (__GNUC__ * 10000 \
+ __GNUC_MINOR__ * 100 \
+ __GNUC_PATCHLEVEL__)
#else
# define _GPG_ERR_GCC_VERSION 0
#endif
#undef _GPG_ERR_HAVE_CONSTRUCTOR
#if _GPG_ERR_GCC_VERSION > 30100
# define _GPG_ERR_CONSTRUCTOR __attribute__ ((__constructor__))
# define _GPG_ERR_HAVE_CONSTRUCTOR
#else
# define _GPG_ERR_CONSTRUCTOR
#endif
#define GPGRT_GCC_VERSION _GCC_ERR_GCC_VERSION
#if _GPG_ERR_GCC_VERSION >= 29200
# define _GPGRT__RESTRICT __restrict__
#else
# define _GPGRT__RESTRICT
#endif
/* The noreturn attribute. */
#if _GPG_ERR_GCC_VERSION >= 20500
# define GPGRT_ATTR_NORETURN __attribute__ ((noreturn))
#else
# define GPGRT_ATTR_NORETURN
#endif
/* The printf attributes. */
#if _GPG_ERR_GCC_VERSION >= 40400
# define GPGRT_ATTR_PRINTF(f, a) \
__attribute__ ((format(__gnu_printf__,f,a)))
# define GPGRT_ATTR_NR_PRINTF(f, a) \
__attribute__ ((noreturn, format(__gnu_printf__,f,a)))
#elif _GPG_ERR_GCC_VERSION >= 20500
# define GPGRT_ATTR_PRINTF(f, a) \
__attribute__ ((format(printf,f,a)))
# define GPGRT_ATTR_NR_PRINTF(f, a) \
__attribute__ ((noreturn, format(printf,f,a)))
#else
# define GPGRT_ATTR_PRINTF(f, a)
# define GPGRT_ATTR_NR_PRINTF(f, a)
#endif
#if _GPG_ERR_GCC_VERSION >= 20800
# define GPGRT_ATTR_FORMAT_ARG(a) __attribute__ ((__format_arg__ (a)))
#else
# define GPGRT_ATTR_FORMAT_ARG(a)
#endif
/* The sentinel attribute. */
#if _GPG_ERR_GCC_VERSION >= 40000
# define GPGRT_ATTR_SENTINEL(a) __attribute__ ((sentinel(a)))
#else
# define GPGRT_ATTR_SENTINEL(a)
#endif
/* The used and unused attributes.
I am not sure since when the unused attribute is really supported.
In any case it it only needed for gcc versions which print a
warning. Thus let us require gcc >= 3.5. */
#if _GPG_ERR_GCC_VERSION >= 40000
# define GPGRT_ATTR_USED __attribute__ ((used))
#else
# define GPGRT_ATTR_USED
#endif
#if _GPG_ERR_GCC_VERSION >= 30500
# define GPGRT_ATTR_UNUSED __attribute__ ((unused))
#else
# define GPGRT_ATTR_UNUSED
#endif
/* The deprecated attribute. */
#if _GPG_ERR_GCC_VERSION >= 30100
# define GPGRT_ATTR_DEPRECATED __attribute__ ((__deprecated__))
#else
# define GPGRT_ATTR_DEPRECATED
#endif
/* The pure attribute. */
#if _GPG_ERR_GCC_VERSION >= 29600
# define GPGRT_ATTR_PURE __attribute__ ((__pure__))
#else
# define GPGRT_ATTR_PURE
#endif
/* The malloc attribute. */
#if _GPG_ERR_GCC_VERSION >= 30200
# define GPGRT_ATTR_MALLOC __attribute__ ((__malloc__))
#else
# define GPGRT_ATTR_MALLOC
#endif
/* A macro defined if a GCC style __FUNCTION__ macro is available. */
#undef GPGRT_HAVE_MACRO_FUNCTION
#if _GPG_ERR_GCC_VERSION >= 20500
# define GPGRT_HAVE_MACRO_FUNCTION 1
#endif
/* A macro defined if the pragma GCC push_options is available. */
#undef GPGRT_HAVE_PRAGMA_GCC_PUSH
#if _GPG_ERR_GCC_VERSION >= 40400
# define GPGRT_HAVE_PRAGMA_GCC_PUSH 1
#endif
/* The new name for the inline macro. */
#define GPGRT_INLINE GPG_ERR_INLINE
/* Initialization function. */
/* Initialize the library. This function should be run early. */
gpg_error_t gpg_err_init (void) _GPG_ERR_CONSTRUCTOR;
/* If this is defined, the library is already initialized by the
constructor and does not need to be initialized explicitely. */
#undef GPG_ERR_INITIALIZED
#ifdef _GPG_ERR_HAVE_CONSTRUCTOR
# define GPG_ERR_INITIALIZED 1
# define gpgrt_init() do { gpg_err_init (); } while (0)
#else
# define gpgrt_init() do { ; } while (0)
#endif
/* See the source on how to use the deinit function; it is usually not
required. */
void gpg_err_deinit (int mode);
/* Register blocking system I/O clamping functions. */
void gpgrt_set_syscall_clamp (void (*pre)(void), void (*post)(void));
/* Register a custom malloc/realloc/free function. */
void gpgrt_set_alloc_func (void *(*f)(void *a, size_t n));
/* Constructor and accessor functions. */
/* Construct an error value from an error code and source. Within a
subsystem, use gpg_error. */
static GPG_ERR_INLINE gpg_error_t
gpg_err_make (gpg_err_source_t source, gpg_err_code_t code)
{
return code == GPG_ERR_NO_ERROR ? GPG_ERR_NO_ERROR
: (((source & GPG_ERR_SOURCE_MASK) << GPG_ERR_SOURCE_SHIFT)
| (code & GPG_ERR_CODE_MASK));
}
/* The user should define GPG_ERR_SOURCE_DEFAULT before including this
file to specify a default source for gpg_error. */
#ifndef GPG_ERR_SOURCE_DEFAULT
#define GPG_ERR_SOURCE_DEFAULT GPG_ERR_SOURCE_UNKNOWN
#endif
static GPG_ERR_INLINE gpg_error_t
gpg_error (gpg_err_code_t code)
{
return gpg_err_make (GPG_ERR_SOURCE_DEFAULT, code);
}
/* Retrieve the error code from an error value. */
static GPG_ERR_INLINE gpg_err_code_t
gpg_err_code (gpg_error_t err)
{
return (gpg_err_code_t) (err & GPG_ERR_CODE_MASK);
}
/* Retrieve the error source from an error value. */
static GPG_ERR_INLINE gpg_err_source_t
gpg_err_source (gpg_error_t err)
{
return (gpg_err_source_t) ((err >> GPG_ERR_SOURCE_SHIFT)
& GPG_ERR_SOURCE_MASK);
}
/* String functions. */
/* Return a pointer to a string containing a description of the error
code in the error value ERR. This function is not thread-safe. */
const char *gpg_strerror (gpg_error_t err);
/* Return the error string for ERR in the user-supplied buffer BUF of
size BUFLEN. This function is, in contrast to gpg_strerror,
thread-safe if a thread-safe strerror_r() function is provided by
the system. If the function succeeds, 0 is returned and BUF
contains the string describing the error. If the buffer was not
large enough, ERANGE is returned and BUF contains as much of the
beginning of the error string as fits into the buffer. */
int gpg_strerror_r (gpg_error_t err, char *buf, size_t buflen);
/* Return a pointer to a string containing a description of the error
source in the error value ERR. */
const char *gpg_strsource (gpg_error_t err);
/* Mapping of system errors (errno). */
/* Retrieve the error code for the system error ERR. This returns
GPG_ERR_UNKNOWN_ERRNO if the system error is not mapped (report
this). */
gpg_err_code_t gpg_err_code_from_errno (int err);
/* Retrieve the system error for the error code CODE. This returns 0
if CODE is not a system error code. */
int gpg_err_code_to_errno (gpg_err_code_t code);
/* Retrieve the error code directly from the ERRNO variable. This
returns GPG_ERR_UNKNOWN_ERRNO if the system error is not mapped
(report this) and GPG_ERR_MISSING_ERRNO if ERRNO has the value 0. */
gpg_err_code_t gpg_err_code_from_syserror (void);
/* Set the ERRNO variable. This function is the preferred way to set
ERRNO due to peculiarities on WindowsCE. */
void gpg_err_set_errno (int err);
/* Return or check the version. Both functions are identical. */
const char *gpgrt_check_version (const char *req_version);
const char *gpg_error_check_version (const char *req_version);
/* System specific type definitions. */
@define:gpgrt_ssize_t@
@define:gpgrt_off_t@
@include:os-add@
/* Self-documenting convenience functions. */
static GPG_ERR_INLINE gpg_error_t
gpg_err_make_from_errno (gpg_err_source_t source, int err)
{
return gpg_err_make (source, gpg_err_code_from_errno (err));
}
static GPG_ERR_INLINE gpg_error_t
gpg_error_from_errno (int err)
{
return gpg_error (gpg_err_code_from_errno (err));
}
static GPG_ERR_INLINE gpg_error_t
gpg_error_from_syserror (void)
{
return gpg_error (gpg_err_code_from_syserror ());
}
/* Lock functions. */
@include:lock-obj@
#define GPGRT_LOCK_DEFINE(name) \
static gpgrt_lock_t name = GPGRT_LOCK_INITIALIZER
/* NB: If GPGRT_LOCK_DEFINE is not used, zero out the lock variable
before passing it to gpgrt_lock_init. */
gpg_err_code_t gpgrt_lock_init (gpgrt_lock_t *lockhd);
gpg_err_code_t gpgrt_lock_lock (gpgrt_lock_t *lockhd);
gpg_err_code_t gpgrt_lock_trylock (gpgrt_lock_t *lockhd);
gpg_err_code_t gpgrt_lock_unlock (gpgrt_lock_t *lockhd);
gpg_err_code_t gpgrt_lock_destroy (gpgrt_lock_t *lockhd);
/* Thread functions. */
gpg_err_code_t gpgrt_yield (void);
/* Estream */
/* The definition of this struct is entirely private. You must not
use it for anything. It is only here so some functions can be
implemented as macros. */
struct _gpgrt_stream_internal;
struct _gpgrt__stream
{
/* The layout of this struct must never change. It may be grown,
but only if all functions which access the new members are
versioned. */
/* Various flags. */
struct {
unsigned int magic: 16;
unsigned int writing: 1;
unsigned int reserved: 15;
} flags;
/* A pointer to the stream buffer. */
unsigned char *buffer;
/* The size of the buffer in bytes. */
size_t buffer_size;
/* The length of the usable data in the buffer, only valid when in
read mode (see flags). */
size_t data_len;
/* The current position of the offset pointer, valid in read and
write mode. */
size_t data_offset;
size_t data_flushed;
unsigned char *unread_buffer;
size_t unread_buffer_size;
/* The number of unread bytes. */
size_t unread_data_len;
/* A pointer to our internal data for this stream. */
struct _gpgrt_stream_internal *intern;
};
/* The opaque type for an estream. */
typedef struct _gpgrt__stream *gpgrt_stream_t;
#ifdef GPGRT_ENABLE_ES_MACROS
typedef struct _gpgrt__stream *estream_t;
#endif
typedef @api_ssize_t@ (*gpgrt_cookie_read_function_t) (void *cookie,
void *buffer, size_t size);
typedef @api_ssize_t@ (*gpgrt_cookie_write_function_t) (void *cookie,
const void *buffer,
size_t size);
typedef int (*gpgrt_cookie_seek_function_t) (void *cookie,
gpgrt_off_t *pos, int whence);
typedef int (*gpgrt_cookie_close_function_t) (void *cookie);
struct _gpgrt_cookie_io_functions
{
gpgrt_cookie_read_function_t func_read;
gpgrt_cookie_write_function_t func_write;
gpgrt_cookie_seek_function_t func_seek;
gpgrt_cookie_close_function_t func_close;
};
typedef struct _gpgrt_cookie_io_functions gpgrt_cookie_io_functions_t;
#ifdef GPGRT_ENABLE_ES_MACROS
typedef struct _gpgrt_cookie_io_functions es_cookie_io_functions_t;
#define es_cookie_read_function_t gpgrt_cookie_read_function_t
#define es_cookie_write_function_t gpgrt_cookie_read_function_t
#define es_cookie_seek_function_t gpgrt_cookie_read_function_t
#define es_cookie_close_function_t gpgrt_cookie_read_function_t
#endif
enum gpgrt_syshd_types
{
GPGRT_SYSHD_NONE = 0, /* No system handle available. */
GPGRT_SYSHD_FD = 1, /* A file descriptor as returned by open(). */
GPGRT_SYSHD_SOCK = 2, /* A socket as returned by socket(). */
GPGRT_SYSHD_RVID = 3, /* A rendezvous id (see libassuan's gpgcedev.c). */
GPGRT_SYSHD_HANDLE = 4 /* A HANDLE object (Windows). */
};
struct _gpgrt_syshd
{
enum gpgrt_syshd_types type;
union {
int fd;
int sock;
int rvid;
void *handle;
} u;
};
typedef struct _gpgrt_syshd gpgrt_syshd_t;
#ifdef GPGRT_ENABLE_ES_MACROS
typedef struct _gpgrt_syshd es_syshd_t;
#define ES_SYSHD_NONE GPGRT_SYSHD_NONE
#define ES_SYSHD_FD GPGRT_SYSHD_FD
#define ES_SYSHD_SOCK GPGRT_SYSHD_SOCK
#define ES_SYSHD_RVID GPGRT_SYSHD_RVID
#define ES_SYSHD_HANDLE GPGRT_SYSHD_HANDLE
#endif
/* The object used with gpgrt_poll. */
struct _gpgrt_poll_s
{
gpgrt_stream_t stream;
unsigned int want_read:1;
unsigned int want_write:1;
unsigned int want_oob:1;
unsigned int want_rdhup:1;
unsigned int _reserv1:4;
unsigned int got_read:1;
unsigned int got_write:1;
unsigned int got_oob:1;
unsigned int got_rdhup:1;
unsigned int _reserv2:4;
unsigned int got_err:1;
unsigned int got_hup:1;
unsigned int got_nval:1;
unsigned int _reserv3:4;
unsigned int ignore:1;
unsigned int user:8; /* For application use. */
};
typedef struct _gpgrt_poll_s gpgrt_poll_t;
#ifdef GPGRT_ENABLE_ES_MACROS
typedef struct _gpgrt_poll_s es_poll_t;
#endif
gpgrt_stream_t gpgrt_fopen (const char *_GPGRT__RESTRICT path,
const char *_GPGRT__RESTRICT mode);
gpgrt_stream_t gpgrt_mopen (void *_GPGRT__RESTRICT data,
size_t data_n, size_t data_len,
unsigned int grow,
void *(*func_realloc) (void *mem, size_t size),
void (*func_free) (void *mem),
const char *_GPGRT__RESTRICT mode);
gpgrt_stream_t gpgrt_fopenmem (size_t memlimit,
const char *_GPGRT__RESTRICT mode);
gpgrt_stream_t gpgrt_fopenmem_init (size_t memlimit,
const char *_GPGRT__RESTRICT mode,
const void *data, size_t datalen);
gpgrt_stream_t gpgrt_fdopen (int filedes, const char *mode);
gpgrt_stream_t gpgrt_fdopen_nc (int filedes, const char *mode);
gpgrt_stream_t gpgrt_sysopen (gpgrt_syshd_t *syshd, const char *mode);
gpgrt_stream_t gpgrt_sysopen_nc (gpgrt_syshd_t *syshd, const char *mode);
gpgrt_stream_t gpgrt_fpopen (FILE *fp, const char *mode);
gpgrt_stream_t gpgrt_fpopen_nc (FILE *fp, const char *mode);
gpgrt_stream_t gpgrt_freopen (const char *_GPGRT__RESTRICT path,
const char *_GPGRT__RESTRICT mode,
gpgrt_stream_t _GPGRT__RESTRICT stream);
gpgrt_stream_t gpgrt_fopencookie (void *_GPGRT__RESTRICT cookie,
const char *_GPGRT__RESTRICT mode,
gpgrt_cookie_io_functions_t functions);
int gpgrt_fclose (gpgrt_stream_t stream);
int gpgrt_fclose_snatch (gpgrt_stream_t stream,
void **r_buffer, size_t *r_buflen);
int gpgrt_onclose (gpgrt_stream_t stream, int mode,
void (*fnc) (gpgrt_stream_t, void*), void *fnc_value);
int gpgrt_fileno (gpgrt_stream_t stream);
int gpgrt_fileno_unlocked (gpgrt_stream_t stream);
int gpgrt_syshd (gpgrt_stream_t stream, gpgrt_syshd_t *syshd);
int gpgrt_syshd_unlocked (gpgrt_stream_t stream, gpgrt_syshd_t *syshd);
void _gpgrt_set_std_fd (int no, int fd);
gpgrt_stream_t _gpgrt_get_std_stream (int fd);
#define gpgrt_stdin _gpgrt_get_std_stream (0)
#define gpgrt_stdout _gpgrt_get_std_stream (1)
#define gpgrt_stderr _gpgrt_get_std_stream (2)
void gpgrt_flockfile (gpgrt_stream_t stream);
int gpgrt_ftrylockfile (gpgrt_stream_t stream);
void gpgrt_funlockfile (gpgrt_stream_t stream);
int gpgrt_feof (gpgrt_stream_t stream);
int gpgrt_feof_unlocked (gpgrt_stream_t stream);
int gpgrt_ferror (gpgrt_stream_t stream);
int gpgrt_ferror_unlocked (gpgrt_stream_t stream);
void gpgrt_clearerr (gpgrt_stream_t stream);
void gpgrt_clearerr_unlocked (gpgrt_stream_t stream);
int _gpgrt_pending (gpgrt_stream_t stream); /* (private) */
int _gpgrt_pending_unlocked (gpgrt_stream_t stream); /* (private) */
#define gpgrt_pending(stream) _gpgrt_pending (stream)
#define gpgrt_pending_unlocked(stream) \
(((!(stream)->flags.writing) \
&& (((stream)->data_offset < (stream)->data_len) \
|| ((stream)->unread_data_len))) \
? 1 : _gpgrt_pending_unlocked ((stream)))
int gpgrt_fflush (gpgrt_stream_t stream);
int gpgrt_fseek (gpgrt_stream_t stream, long int offset, int whence);
int gpgrt_fseeko (gpgrt_stream_t stream, gpgrt_off_t offset, int whence);
long int gpgrt_ftell (gpgrt_stream_t stream);
gpgrt_off_t gpgrt_ftello (gpgrt_stream_t stream);
void gpgrt_rewind (gpgrt_stream_t stream);
int gpgrt_fgetc (gpgrt_stream_t stream);
int gpgrt_fputc (int c, gpgrt_stream_t stream);
int _gpgrt_getc_underflow (gpgrt_stream_t stream); /* (private) */
int _gpgrt_putc_overflow (int c, gpgrt_stream_t stream); /* (private) */
#define gpgrt_getc_unlocked(stream) \
(((!(stream)->flags.writing) \
&& ((stream)->data_offset < (stream)->data_len) \
&& (! (stream)->unread_data_len)) \
? ((int) (stream)->buffer[((stream)->data_offset)++]) \
: _gpgrt_getc_underflow ((stream)))
#define gpgrt_putc_unlocked(c, stream) \
(((stream)->flags.writing \
&& ((stream)->data_offset < (stream)->buffer_size) \
&& (c != '\n')) \
? ((int) ((stream)->buffer[((stream)->data_offset)++] = (c))) \
: _gpgrt_putc_overflow ((c), (stream)))
#define gpgrt_getc(stream) gpgrt_fgetc (stream)
#define gpgrt_putc(c, stream) gpgrt_fputc (c, stream)
int gpgrt_ungetc (int c, gpgrt_stream_t stream);
int gpgrt_read (gpgrt_stream_t _GPGRT__RESTRICT stream,
void *_GPGRT__RESTRICT buffer, size_t bytes_to_read,
size_t *_GPGRT__RESTRICT bytes_read);
int gpgrt_write (gpgrt_stream_t _GPGRT__RESTRICT stream,
const void *_GPGRT__RESTRICT buffer, size_t bytes_to_write,
size_t *_GPGRT__RESTRICT bytes_written);
int gpgrt_write_sanitized (gpgrt_stream_t _GPGRT__RESTRICT stream,
const void *_GPGRT__RESTRICT buffer, size_t length,
const char *delimiters,
size_t *_GPGRT__RESTRICT bytes_written);
int gpgrt_write_hexstring (gpgrt_stream_t _GPGRT__RESTRICT stream,
const void *_GPGRT__RESTRICT buffer, size_t length,
int reserved,
size_t *_GPGRT__RESTRICT bytes_written);
size_t gpgrt_fread (void *_GPGRT__RESTRICT ptr, size_t size, size_t nitems,
gpgrt_stream_t _GPGRT__RESTRICT stream);
size_t gpgrt_fwrite (const void *_GPGRT__RESTRICT ptr, size_t size, size_t memb,
gpgrt_stream_t _GPGRT__RESTRICT stream);
char *gpgrt_fgets (char *_GPGRT__RESTRICT s, int n,
gpgrt_stream_t _GPGRT__RESTRICT stream);
int gpgrt_fputs (const char *_GPGRT__RESTRICT s,
gpgrt_stream_t _GPGRT__RESTRICT stream);
int gpgrt_fputs_unlocked (const char *_GPGRT__RESTRICT s,
gpgrt_stream_t _GPGRT__RESTRICT stream);
@api_ssize_t@ gpgrt_getline (char *_GPGRT__RESTRICT *_GPGRT__RESTRICT lineptr,
size_t *_GPGRT__RESTRICT n,
gpgrt_stream_t stream);
@api_ssize_t@ gpgrt_read_line (gpgrt_stream_t stream,
char **addr_of_buffer, size_t *length_of_buffer,
size_t *max_length);
void gpgrt_free (void *a);
int gpgrt_fprintf (gpgrt_stream_t _GPGRT__RESTRICT stream,
const char *_GPGRT__RESTRICT format, ...)
GPGRT_ATTR_PRINTF(2,3);
int gpgrt_fprintf_unlocked (gpgrt_stream_t _GPGRT__RESTRICT stream,
const char *_GPGRT__RESTRICT format, ...)
GPGRT_ATTR_PRINTF(2,3);
int gpgrt_printf (const char *_GPGRT__RESTRICT format, ...)
GPGRT_ATTR_PRINTF(1,2);
int gpgrt_printf_unlocked (const char *_GPGRT__RESTRICT format, ...)
GPGRT_ATTR_PRINTF(1,2);
int gpgrt_vfprintf (gpgrt_stream_t _GPGRT__RESTRICT stream,
const char *_GPGRT__RESTRICT format, va_list ap)
GPGRT_ATTR_PRINTF(2,0);
int gpgrt_vfprintf_unlocked (gpgrt_stream_t _GPGRT__RESTRICT stream,
const char *_GPGRT__RESTRICT format, va_list ap)
GPGRT_ATTR_PRINTF(2,0);
int gpgrt_setvbuf (gpgrt_stream_t _GPGRT__RESTRICT stream,
char *_GPGRT__RESTRICT buf, int mode, size_t size);
void gpgrt_setbuf (gpgrt_stream_t _GPGRT__RESTRICT stream,
char *_GPGRT__RESTRICT buf);
void gpgrt_set_binary (gpgrt_stream_t stream);
int gpgrt_set_nonblock (gpgrt_stream_t stream, int onoff);
int gpgrt_get_nonblock (gpgrt_stream_t stream);
int gpgrt_poll (gpgrt_poll_t *fdlist, unsigned int nfds, int timeout);
gpgrt_stream_t gpgrt_tmpfile (void);
void gpgrt_opaque_set (gpgrt_stream_t _GPGRT__RESTRICT stream,
void *_GPGRT__RESTRICT opaque);
void *gpgrt_opaque_get (gpgrt_stream_t stream);
void gpgrt_fname_set (gpgrt_stream_t stream, const char *fname);
const char *gpgrt_fname_get (gpgrt_stream_t stream);
int gpgrt_asprintf (char **r_buf, const char * _GPGRT__RESTRICT format, ...)
GPGRT_ATTR_PRINTF(2,3);
int gpgrt_vasprintf (char **r_buf, const char * _GPGRT__RESTRICT format,
va_list ap)
GPGRT_ATTR_PRINTF(2,0);
char *gpgrt_bsprintf (const char * _GPGRT__RESTRICT format, ...)
GPGRT_ATTR_PRINTF(1,2);
char *gpgrt_vbsprintf (const char * _GPGRT__RESTRICT format, va_list ap)
GPGRT_ATTR_PRINTF(1,0);
int gpgrt_snprintf (char *buf, size_t bufsize,
const char * _GPGRT__RESTRICT format, ...)
GPGRT_ATTR_PRINTF(3,4);
int gpgrt_vsnprintf (char *buf,size_t bufsize,
const char * _GPGRT__RESTRICT format, va_list arg_ptr)
GPGRT_ATTR_PRINTF(3,0);
#ifdef GPGRT_ENABLE_ES_MACROS
# define es_fopen gpgrt_fopen
# define es_mopen gpgrt_mopen
# define es_fopenmem gpgrt_fopenmem
# define es_fopenmem_init gpgrt_fopenmem_init
# define es_fdopen gpgrt_fdopen
# define es_fdopen_nc gpgrt_fdopen_nc
# define es_sysopen gpgrt_sysopen
# define es_sysopen_nc gpgrt_sysopen_nc
# define es_fpopen gpgrt_fpopen
# define es_fpopen_nc gpgrt_fpopen_nc
# define es_freopen gpgrt_freopen
# define es_fopencookie gpgrt_fopencookie
# define es_fclose gpgrt_fclose
# define es_fclose_snatch gpgrt_fclose_snatch
# define es_onclose gpgrt_onclose
# define es_fileno gpgrt_fileno
# define es_fileno_unlocked gpgrt_fileno_unlocked
# define es_syshd gpgrt_syshd
# define es_syshd_unlocked gpgrt_syshd_unlocked
# define es_stdin _gpgrt_get_std_stream (0)
# define es_stdout _gpgrt_get_std_stream (1)
# define es_stderr _gpgrt_get_std_stream (2)
# define es_flockfile gpgrt_flockfile
# define es_ftrylockfile gpgrt_ftrylockfile
# define es_funlockfile gpgrt_funlockfile
# define es_feof gpgrt_feof
# define es_feof_unlocked gpgrt_feof_unlocked
# define es_ferror gpgrt_ferror
# define es_ferror_unlocked gpgrt_ferror_unlocked
# define es_clearerr gpgrt_clearerr
# define es_clearerr_unlocked gpgrt_clearerr_unlocked
# define es_pending gpgrt_pending
# define es_pending_unlocked gpgrt_pending_unlocked
# define es_fflush gpgrt_fflush
# define es_fseek gpgrt_fseek
# define es_fseeko gpgrt_fseeko
# define es_ftell gpgrt_ftell
# define es_ftello gpgrt_ftello
# define es_rewind gpgrt_rewind
# define es_fgetc gpgrt_fgetc
# define es_fputc gpgrt_fputc
# define es_getc_unlocked gpgrt_getc_unlocked
# define es_putc_unlocked gpgrt_putc_unlocked
# define es_getc gpgrt_getc
# define es_putc gpgrt_putc
# define es_ungetc gpgrt_ungetc
# define es_read gpgrt_read
# define es_write gpgrt_write
# define es_write_sanitized gpgrt_write_sanitized
# define es_write_hexstring gpgrt_write_hexstring
# define es_fread gpgrt_fread
# define es_fwrite gpgrt_fwrite
# define es_fgets gpgrt_fgets
# define es_fputs gpgrt_fputs
# define es_fputs_unlocked gpgrt_fputs_unlocked
# define es_getline gpgrt_getline
# define es_read_line gpgrt_read_line
# define es_free gpgrt_free
# define es_fprintf gpgrt_fprintf
# define es_fprintf_unlocked gpgrt_fprintf_unlocked
# define es_printf gpgrt_printf
# define es_printf_unlocked gpgrt_printf_unlocked
# define es_vfprintf gpgrt_vfprintf
# define es_vfprintf_unlocked gpgrt_vfprintf_unlocked
# define es_setvbuf gpgrt_setvbuf
# define es_setbuf gpgrt_setbuf
# define es_set_binary gpgrt_set_binary
# define es_set_nonblock gpgrt_set_nonblock
# define es_get_nonblock gpgrt_get_nonblock
# define es_poll gpgrt_poll
# define es_tmpfile gpgrt_tmpfile
# define es_opaque_set gpgrt_opaque_set
# define es_opaque_get gpgrt_opaque_get
# define es_fname_set gpgrt_fname_set
# define es_fname_get gpgrt_fname_get
# define es_asprintf gpgrt_asprintf
# define es_vasprintf gpgrt_vasprintf
# define es_bsprintf gpgrt_bsprintf
# define es_vbsprintf gpgrt_vbsprintf
#endif /*GPGRT_ENABLE_ES_MACROS*/
#ifdef __cplusplus
}
#endif
#endif /* GPG_ERROR_H */
diff --git a/src/gpg-error.vers b/src/gpg-error.vers
index 067bb29..cdff0e3 100644
--- a/src/gpg-error.vers
+++ b/src/gpg-error.vers
@@ -1,130 +1,130 @@
# libgpg-error.vers - What symbols to export -*- std -*-
# Copyright (C) 2014 g10 Code GmbH
#
# This file is part of libgpg-error.
#
# libgpg-error is free software; you can redistribute it and/or modify
# it under the terms of the GNU Lesser General Public License as
# published by the Free Software Foundation; either version 2.1 of the
# License, or (at your option) any later version.
#
# libgpg-error is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU Lesser General Public License for more details.
#
# You should have received a copy of the GNU Lesser General Public
-# License along with this program; if not, see .
+# License along with this program; if not, see .
#
# NOTE: When adding new functions, please make sure to add them to
# visibility.h and gpg-error.def.in as well.
GPG_ERROR_1.0 {
global:
gpg_strerror;
gpg_strerror_r;
gpg_strsource;
gpg_err_code_from_errno;
gpg_err_code_to_errno;
gpg_err_code_from_syserror;
gpg_err_set_errno;
gpg_error_check_version;
gpgrt_lock_init;
gpgrt_lock_lock;
gpgrt_lock_unlock;
gpgrt_lock_destroy;
gpgrt_yield;
gpgrt_lock_trylock;
gpgrt_set_syscall_clamp;
gpgrt_fopen;
gpgrt_mopen;
gpgrt_fopenmem;
gpgrt_fopenmem_init;
gpgrt_fdopen;
gpgrt_fdopen_nc;
gpgrt_sysopen;
gpgrt_sysopen_nc;
gpgrt_fpopen;
gpgrt_fpopen_nc;
gpgrt_freopen;
gpgrt_fopencookie;
gpgrt_fclose;
gpgrt_fclose_snatch;
gpgrt_onclose;
gpgrt_fileno;
gpgrt_fileno_unlocked;
gpgrt_syshd;
gpgrt_syshd_unlocked;
_gpgrt_set_std_fd;
_gpgrt_get_std_stream;
gpgrt_flockfile;
gpgrt_ftrylockfile;
gpgrt_funlockfile;
_gpgrt_pending;
_gpgrt_pending_unlocked;
gpgrt_feof;
gpgrt_feof_unlocked;
gpgrt_ferror;
gpgrt_ferror_unlocked;
gpgrt_clearerr;
gpgrt_clearerr_unlocked;
gpgrt_fflush;
gpgrt_fseek;
gpgrt_fseeko;
gpgrt_ftell;
gpgrt_ftello;
gpgrt_rewind;
gpgrt_fgetc;
_gpgrt_getc_underflow;
gpgrt_fputc;
_gpgrt_putc_overflow;
gpgrt_ungetc;
gpgrt_read;
gpgrt_write;
gpgrt_write_sanitized;
gpgrt_write_hexstring;
gpgrt_fread;
gpgrt_fwrite;
gpgrt_fgets;
gpgrt_fputs;
gpgrt_fputs_unlocked;
gpgrt_getline;
gpgrt_read_line;
gpgrt_free;
gpgrt_fprintf;
gpgrt_fprintf_unlocked;
gpgrt_printf;
gpgrt_printf_unlocked;
gpgrt_vfprintf;
gpgrt_vfprintf_unlocked;
gpgrt_setvbuf;
gpgrt_setbuf;
gpgrt_set_binary;
gpgrt_set_nonblock;
gpgrt_get_nonblock;
gpgrt_poll;
gpgrt_tmpfile;
gpgrt_opaque_set;
gpgrt_opaque_get;
gpgrt_fname_set;
gpgrt_fname_get;
gpgrt_asprintf;
gpgrt_vasprintf;
gpgrt_bsprintf;
gpgrt_vbsprintf;
gpgrt_snprintf;
gpgrt_vsnprintf;
gpgrt_check_version;
gpg_err_init;
gpg_err_deinit;
gpgrt_set_alloc_func;
local:
*;
};
diff --git a/src/gpgrt-int.h b/src/gpgrt-int.h
index 0f7b29b..d69fe2c 100644
--- a/src/gpgrt-int.h
+++ b/src/gpgrt-int.h
@@ -1,209 +1,209 @@
/* gpgrt-int.h - Internal definitions
* Copyright (C) 2014 g10 Code GmbH
*
* This file is part of libgpg-error.
*
* libgpg-error is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public License
* as published by the Free Software Foundation; either version 2.1 of
* the License, or (at your option) any later version.
*
* libgpg-error is distributed in the hope that it will be useful, but
* WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
- * License along with this program; if not, see .
+ * License along with this program; if not, see .
*/
#ifndef _GPGRT_GPGRT_INT_H
#define _GPGRT_GPGRT_INT_H
#include "gpg-error.h"
#include "visibility.h"
/* Local error function prototypes. */
const char *_gpg_strerror (gpg_error_t err);
int _gpg_strerror_r (gpg_error_t err, char *buf, size_t buflen);
const char *_gpg_strsource (gpg_error_t err);
gpg_err_code_t _gpg_err_code_from_errno (int err);
int _gpg_err_code_to_errno (gpg_err_code_t code);
gpg_err_code_t _gpg_err_code_from_syserror (void);
void _gpg_err_set_errno (int err);
gpg_error_t _gpg_err_init (void);
void _gpg_err_deinit (int mode);
void _gpgrt_set_alloc_func (void *(*f)(void *a, size_t n));
void *_gpgrt_realloc (void *a, size_t n);
void *_gpgrt_malloc (size_t n);
void _gpgrt_free (void *a);
const char *_gpg_error_check_version (const char *req_version);
gpg_err_code_t _gpgrt_lock_init (gpgrt_lock_t *lockhd);
gpg_err_code_t _gpgrt_lock_lock (gpgrt_lock_t *lockhd);
gpg_err_code_t _gpgrt_lock_trylock (gpgrt_lock_t *lockhd);
gpg_err_code_t _gpgrt_lock_unlock (gpgrt_lock_t *lockhd);
gpg_err_code_t _gpgrt_lock_destroy (gpgrt_lock_t *lockhd);
gpg_err_code_t _gpgrt_yield (void);
/* Local prototypes for estream. */
int _gpgrt_es_init (void);
void _gpgrt_set_syscall_clamp (void (*pre)(void), void (*post)(void));
gpgrt_stream_t _gpgrt_fopen (const char *_GPGRT__RESTRICT path,
const char *_GPGRT__RESTRICT mode);
gpgrt_stream_t _gpgrt_mopen (void *_GPGRT__RESTRICT data,
size_t data_n, size_t data_len,
unsigned int grow,
void *(*func_realloc) (void *mem, size_t size),
void (*func_free) (void *mem),
const char *_GPGRT__RESTRICT mode);
gpgrt_stream_t _gpgrt_fopenmem (size_t memlimit,
const char *_GPGRT__RESTRICT mode);
gpgrt_stream_t _gpgrt_fopenmem_init (size_t memlimit,
const char *_GPGRT__RESTRICT mode,
const void *data, size_t datalen);
gpgrt_stream_t _gpgrt_fdopen (int filedes, const char *mode);
gpgrt_stream_t _gpgrt_fdopen_nc (int filedes, const char *mode);
gpgrt_stream_t _gpgrt_sysopen (gpgrt_syshd_t *syshd, const char *mode);
gpgrt_stream_t _gpgrt_sysopen_nc (gpgrt_syshd_t *syshd, const char *mode);
gpgrt_stream_t _gpgrt_fpopen (FILE *fp, const char *mode);
gpgrt_stream_t _gpgrt_fpopen_nc (FILE *fp, const char *mode);
gpgrt_stream_t _gpgrt_freopen (const char *_GPGRT__RESTRICT path,
const char *_GPGRT__RESTRICT mode,
gpgrt_stream_t _GPGRT__RESTRICT stream);
gpgrt_stream_t _gpgrt_fopencookie (void *_GPGRT__RESTRICT cookie,
const char *_GPGRT__RESTRICT mode,
gpgrt_cookie_io_functions_t functions);
int _gpgrt_fclose (gpgrt_stream_t stream);
int _gpgrt_fclose_snatch (gpgrt_stream_t stream,
void **r_buffer, size_t *r_buflen);
int _gpgrt_onclose (gpgrt_stream_t stream, int mode,
void (*fnc) (gpgrt_stream_t, void*), void *fnc_value);
int _gpgrt_fileno (gpgrt_stream_t stream);
int _gpgrt_fileno_unlocked (gpgrt_stream_t stream);
int _gpgrt_syshd (gpgrt_stream_t stream, gpgrt_syshd_t *syshd);
int _gpgrt_syshd_unlocked (gpgrt_stream_t stream, gpgrt_syshd_t *syshd);
void _gpgrt__set_std_fd (int no, int fd);
gpgrt_stream_t _gpgrt__get_std_stream (int fd);
void _gpgrt_flockfile (gpgrt_stream_t stream);
int _gpgrt_ftrylockfile (gpgrt_stream_t stream);
void _gpgrt_funlockfile (gpgrt_stream_t stream);
int _gpgrt_feof (gpgrt_stream_t stream);
int _gpgrt_feof_unlocked (gpgrt_stream_t stream);
int _gpgrt_ferror (gpgrt_stream_t stream);
int _gpgrt_ferror_unlocked (gpgrt_stream_t stream);
void _gpgrt_clearerr (gpgrt_stream_t stream);
void _gpgrt_clearerr_unlocked (gpgrt_stream_t stream);
int _gpgrt__pending (gpgrt_stream_t stream);
int _gpgrt__pending_unlocked (gpgrt_stream_t stream);
int _gpgrt_fflush (gpgrt_stream_t stream);
int _gpgrt_fseek (gpgrt_stream_t stream, long int offset, int whence);
int _gpgrt_fseeko (gpgrt_stream_t stream, gpgrt_off_t offset, int whence);
long int _gpgrt_ftell (gpgrt_stream_t stream);
gpgrt_off_t _gpgrt_ftello (gpgrt_stream_t stream);
void _gpgrt_rewind (gpgrt_stream_t stream);
int _gpgrt_fgetc (gpgrt_stream_t stream);
int _gpgrt_fputc (int c, gpgrt_stream_t stream);
int _gpgrt__getc_underflow (gpgrt_stream_t stream);
int _gpgrt__putc_overflow (int c, gpgrt_stream_t stream);
/* Note: Keeps the next two macros in sync
with their counterparts in gpg-error.h. */
#define _gpgrt_getc_unlocked(stream) \
(((!(stream)->flags.writing) \
&& ((stream)->data_offset < (stream)->data_len) \
&& (! (stream)->unread_data_len)) \
? ((int) (stream)->buffer[((stream)->data_offset)++]) \
: _gpgrt__getc_underflow ((stream)))
#define _gpgrt_putc_unlocked(c, stream) \
(((stream)->flags.writing \
&& ((stream)->data_offset < (stream)->buffer_size) \
&& (c != '\n')) \
? ((int) ((stream)->buffer[((stream)->data_offset)++] = (c))) \
: _gpgrt__putc_overflow ((c), (stream)))
int _gpgrt_ungetc (int c, gpgrt_stream_t stream);
int _gpgrt_read (gpgrt_stream_t _GPGRT__RESTRICT stream,
void *_GPGRT__RESTRICT buffer, size_t bytes_to_read,
size_t *_GPGRT__RESTRICT bytes_read);
int _gpgrt_write (gpgrt_stream_t _GPGRT__RESTRICT stream,
const void *_GPGRT__RESTRICT buffer, size_t bytes_to_write,
size_t *_GPGRT__RESTRICT bytes_written);
int _gpgrt_write_sanitized (gpgrt_stream_t _GPGRT__RESTRICT stream,
const void *_GPGRT__RESTRICT buffer, size_t length,
const char *delimiters,
size_t *_GPGRT__RESTRICT bytes_written);
int _gpgrt_write_hexstring (gpgrt_stream_t _GPGRT__RESTRICT stream,
const void *_GPGRT__RESTRICT buffer, size_t length,
int reserved,
size_t *_GPGRT__RESTRICT bytes_written);
size_t _gpgrt_fread (void *_GPGRT__RESTRICT ptr, size_t size, size_t nitems,
gpgrt_stream_t _GPGRT__RESTRICT stream);
size_t _gpgrt_fwrite (const void *_GPGRT__RESTRICT ptr,
size_t size, size_t memb,
gpgrt_stream_t _GPGRT__RESTRICT stream);
char *_gpgrt_fgets (char *_GPGRT__RESTRICT s, int n,
gpgrt_stream_t _GPGRT__RESTRICT stream);
int _gpgrt_fputs (const char *_GPGRT__RESTRICT s,
gpgrt_stream_t _GPGRT__RESTRICT stream);
int _gpgrt_fputs_unlocked (const char *_GPGRT__RESTRICT s,
gpgrt_stream_t _GPGRT__RESTRICT stream);
gpgrt_ssize_t _gpgrt_getline (char *_GPGRT__RESTRICT *_GPGRT__RESTRICT lineptr,
size_t *_GPGRT__RESTRICT n,
gpgrt_stream_t stream);
gpgrt_ssize_t _gpgrt_read_line (gpgrt_stream_t stream,
char **addr_of_buffer, size_t *length_of_buffer,
size_t *max_length);
int _gpgrt_fprintf (gpgrt_stream_t _GPGRT__RESTRICT stream,
const char *_GPGRT__RESTRICT format, ...)
GPGRT_ATTR_PRINTF(2,3);
int _gpgrt_fprintf_unlocked (gpgrt_stream_t _GPGRT__RESTRICT stream,
const char *_GPGRT__RESTRICT format, ...)
GPGRT_ATTR_PRINTF(2,3);
int _gpgrt_vfprintf (gpgrt_stream_t _GPGRT__RESTRICT stream,
const char *_GPGRT__RESTRICT format, va_list ap)
GPGRT_ATTR_PRINTF(2,0);
int _gpgrt_vfprintf_unlocked (gpgrt_stream_t _GPGRT__RESTRICT stream,
const char *_GPGRT__RESTRICT format, va_list ap)
GPGRT_ATTR_PRINTF(2,0);
int _gpgrt_setvbuf (gpgrt_stream_t _GPGRT__RESTRICT stream,
char *_GPGRT__RESTRICT buf, int mode, size_t size);
void _gpgrt_set_binary (gpgrt_stream_t stream);
int _gpgrt_set_nonblock (gpgrt_stream_t stream, int onoff);
int _gpgrt_get_nonblock (gpgrt_stream_t stream);
int _gpgrt_poll (gpgrt_poll_t *fds, unsigned int nfds, int timeout);
gpgrt_stream_t _gpgrt_tmpfile (void);
void _gpgrt_opaque_set (gpgrt_stream_t _GPGRT__RESTRICT stream,
void *_GPGRT__RESTRICT opaque);
void *_gpgrt_opaque_get (gpgrt_stream_t stream);
void _gpgrt_fname_set (gpgrt_stream_t stream, const char *fname);
const char *_gpgrt_fname_get (gpgrt_stream_t stream);
#include "estream-printf.h"
#endif /*_GPGRT_GPGRT_INT_H*/
diff --git a/src/init.c b/src/init.c
index 7abb6ff..8de54b6 100644
--- a/src/init.c
+++ b/src/init.c
@@ -1,427 +1,427 @@
/* init.c - Initialize the GnuPG error library.
Copyright (C) 2005, 2010 g10 Code GmbH
This file is part of libgpg-error.
libgpg-error is free software; you can redistribute it and/or
modify it under the terms of the GNU Lesser General Public License
as published by the Free Software Foundation; either version 2.1 of
the License, or (at your option) any later version.
libgpg-error is distributed in the hope that it will be useful, but
WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public
- License along with this program; if not, see .
+ License along with this program; if not, see .
*/
#if HAVE_CONFIG_H
#include
#endif
#include
#include
#include
#include
#include "gpgrt-int.h"
#include "gettext.h"
#include "init.h"
#ifdef HAVE_W32CE_SYSTEM
# include "mkw32errmap.map.c" /* Generated map_w32codes () */
# ifndef TLS_OUT_OF_INDEXES
# define TLS_OUT_OF_INDEXES 0xFFFFFFFF
# endif
# ifndef __MINGW32CE__
# /* Replace the Mingw32CE provided abort function. */
# define abort() do { TerminateProcess (GetCurrentProcess(), 8); } while (0)
# endif
#endif
/* Locale directory support. */
#if HAVE_W32_SYSTEM
#include
static int tls_index = TLS_OUT_OF_INDEXES; /* Index for the TLS functions. */
static char *get_locale_dir (void);
static void drop_locale_dir (char *locale_dir);
#else /*!HAVE_W32_SYSTEM*/
#define get_locale_dir() LOCALEDIR
#define drop_locale_dir(dir)
#endif /*!HAVE_W32_SYSTEM*/
/* The realloc function as set by gpgrt_set_alloc_func. */
static void *(*custom_realloc)(void *a, size_t n);
static void
real_init (void)
{
#ifdef ENABLE_NLS
char *locale_dir;
/* We only have to bind our locale directory to our text domain. */
locale_dir = get_locale_dir ();
if (locale_dir)
{
bindtextdomain (PACKAGE, locale_dir);
drop_locale_dir (locale_dir);
}
#endif
_gpgrt_es_init ();
}
/* Initialize the library. This function should be run early. */
gpg_error_t
_gpg_err_init (void)
{
#ifdef HAVE_W32_SYSTEM
# ifdef DLL_EXPORT
/* We always have a constructor and thus this function is called
automatically. Due to the way the C init code of mingw works,
the constructors are called before our DllMain function is
called. The problem with that is that the TLS has not been setup
and w32-gettext.c requires TLS. To solve this we do nothing here
but call the actual init code from our DllMain. */
# else /*!DLL_EXPORT*/
/* Note that if the TLS is actually used, we can't release the TLS
as there is no way to know when a thread terminates (i.e. no
thread-specific-atexit). You are really better off to use the
DLL! */
if (tls_index == TLS_OUT_OF_INDEXES)
{
tls_index = TlsAlloc ();
if (tls_index == TLS_OUT_OF_INDEXES)
{
/* No way to continue - commit suicide. */
abort ();
}
_gpg_w32__init_gettext_module ();
real_init ();
}
# endif /*!DLL_EXPORT*/
#else
real_init ();
#endif
return 0;
}
/* Deinitialize libgpg-error. This function is only used in special
circumstances. No gpg-error function should be used after this
function has been called. A value of 0 passed for MODE
deinitializes the entire libgpg-error, a value of 1 releases
resources allocated for the current thread and only that thread may
not anymore access libgpg-error after such a call. Under Windows
this function may be called from the DllMain function of a DLL
which statically links to libgpg-error. */
void
_gpg_err_deinit (int mode)
{
#if defined (HAVE_W32_SYSTEM) && !defined(DLL_EXPORT)
struct tls_space_s *tls;
tls = TlsGetValue (tls_index);
if (tls)
{
TlsSetValue (tls_index, NULL);
LocalFree (tls);
}
if (mode == 0)
{
TlsFree (tls_index);
tls_index = TLS_OUT_OF_INDEXES;
}
#else
(void)mode;
#endif
}
/* Register F as allocation function. This function is used for all
APIs which return an allocated buffer. F needs to have standard
realloc semantics. It should be called as early as possible and
not changed later. */
void
_gpgrt_set_alloc_func (void *(*f)(void *a, size_t n))
{
custom_realloc = f;
}
/* The realloc to be used for data returned by the public API. */
void *
_gpgrt_realloc (void *a, size_t n)
{
if (custom_realloc)
return custom_realloc (a, n);
if (!n)
{
free (a);
return NULL;
}
if (!a)
return malloc (n);
return realloc (a, n);
}
/* The malloc to be used for data returned by the public API. */
void *
_gpgrt_malloc (size_t n)
{
if (!n)
n++;
return _gpgrt_realloc (NULL, n);
}
/* The free to be used for data returned by the public API. */
void
_gpgrt_free (void *a)
{
_gpgrt_realloc (a, 0);
}
void
_gpg_err_set_errno (int err)
{
#ifdef HAVE_W32CE_SYSTEM
SetLastError (err);
#else /*!HAVE_W32CE_SYSTEM*/
errno = err;
#endif /*!HAVE_W32CE_SYSTEM*/
}
#ifdef HAVE_W32_SYSTEM
/*****************************************
******** Below is only Windows code. ****
*****************************************/
static char *
get_locale_dir (void)
{
static wchar_t moddir[MAX_PATH+5];
char *result, *p;
int nbytes;
if (!GetModuleFileNameW (NULL, moddir, MAX_PATH))
*moddir = 0;
#define SLDIR "\\share\\locale"
if (*moddir)
{
nbytes = WideCharToMultiByte (CP_UTF8, 0, moddir, -1, NULL, 0, NULL, NULL);
if (nbytes < 0)
return NULL;
result = malloc (nbytes + strlen (SLDIR) + 1);
if (result)
{
nbytes = WideCharToMultiByte (CP_UTF8, 0, moddir, -1,
result, nbytes, NULL, NULL);
if (nbytes < 0)
{
free (result);
result = NULL;
}
else
{
p = strrchr (result, '\\');
if (p)
*p = 0;
/* If we are installed below "bin" strip that part and
use the top directory instead.
Background: Under Windows we don't install GnuPG
below bin/ but in the top directory with only share/,
lib/, and etc/ below it. One of the reasons is to
keep the the length of the filenames at bay so not to
increase the limited length of the PATH envvar.
Another and more important reason, however, is that
the very first GPG versions on W32 were installed
into a flat directory structure and for best
compatibility with these versions we didn't changed
that later. For WindowsCE we can right away install
it under bin, though. The hack with detection of the
bin directory part allows us to eventually migrate to
such a directory layout under plain Windows without
the need to change libgpg-error. */
p = strrchr (result, '\\');
if (p && !strcmp (p+1, "bin"))
*p = 0;
/* Append the static part. */
strcat (result, SLDIR);
}
}
}
else /* Use the old default value. */
{
result = malloc (10 + strlen (SLDIR) + 1);
if (result)
{
strcpy (result, "c:\\gnupg");
strcat (result, SLDIR);
}
}
#undef SLDIR
return result;
}
static void
drop_locale_dir (char *locale_dir)
{
free (locale_dir);
}
/* Return the tls object. This function is guaranteed to return a
valid non-NULL object. */
struct tls_space_s *
get_tls (void)
{
struct tls_space_s *tls;
tls = TlsGetValue (tls_index);
if (!tls)
{
/* Called by a thread which existed before this DLL was loaded.
Allocate the space. */
tls = LocalAlloc (LPTR, sizeof *tls);
if (!tls)
{
/* No way to continue - commit suicide. */
abort ();
}
tls->gt_use_utf8 = 0;
TlsSetValue (tls_index, tls);
}
return tls;
}
/* Return the value of the ERRNO variable. This needs to be a
function so that we can have a per-thread ERRNO. This is used only
on WindowsCE because that OS misses an errno. */
#ifdef HAVE_W32CE_SYSTEM
int
_gpg_w32ce_get_errno (void)
{
return map_w32codes ( GetLastError () );
}
#endif /*HAVE_W32CE_SYSTEM*/
/* Replacement strerror function for WindowsCE. */
#ifdef HAVE_W32CE_SYSTEM
char *
_gpg_w32ce_strerror (int err)
{
struct tls_space_s *tls = get_tls ();
wchar_t tmpbuf[STRBUFFER_SIZE];
int n;
if (err == -1)
err = _gpg_w32ce_get_errno ();
/* Note: On a German HTC Touch Pro2 device I also tried
LOCALE_USER_DEFAULT and LOCALE_SYSTEM_DEFAULT - both returned
English messages. */
if (FormatMessageW (FORMAT_MESSAGE_FROM_SYSTEM, NULL, err,
MAKELANGID (LANG_NEUTRAL, SUBLANG_DEFAULT),
tmpbuf, STRBUFFER_SIZE -1,
NULL))
{
n = WideCharToMultiByte (CP_UTF8, 0, tmpbuf, -1,
tls->strerror_buffer,
sizeof tls->strerror_buffer -1,
NULL, NULL);
}
else
n = -1;
if (n < 0)
snprintf (tls->strerror_buffer, sizeof tls->strerror_buffer -1,
"[w32err=%d]", err);
return tls->strerror_buffer;
}
#endif /*HAVE_W32CE_SYSTEM*/
/* Entry point called by the DLL loader. */
#ifdef DLL_EXPORT
int WINAPI
DllMain (HINSTANCE hinst, DWORD reason, LPVOID reserved)
{
struct tls_space_s *tls;
(void)reserved;
(void)hinst;
switch (reason)
{
case DLL_PROCESS_ATTACH:
tls_index = TlsAlloc ();
if (tls_index == TLS_OUT_OF_INDEXES)
return FALSE;
#ifndef _GPG_ERR_HAVE_CONSTRUCTOR
/* If we have not constructors (e.g. MSC) we call it here. */
_gpg_w32__init_gettext_module ();
#endif
/* falltru. */
case DLL_THREAD_ATTACH:
tls = LocalAlloc (LPTR, sizeof *tls);
if (!tls)
return FALSE;
tls->gt_use_utf8 = 0;
TlsSetValue (tls_index, tls);
if (reason == DLL_PROCESS_ATTACH)
{
real_init ();
}
break;
case DLL_THREAD_DETACH:
tls = TlsGetValue (tls_index);
if (tls)
LocalFree (tls);
break;
case DLL_PROCESS_DETACH:
tls = TlsGetValue (tls_index);
if (tls)
LocalFree (tls);
TlsFree (tls_index);
break;
default:
break;
}
return TRUE;
}
#endif /*DLL_EXPORT*/
#endif /*HAVE_W32_SYSTEM*/
diff --git a/src/init.h b/src/init.h
index 0a31fd7..90a716a 100644
--- a/src/init.h
+++ b/src/init.h
@@ -1,70 +1,70 @@
/* init.h - Declarations for init.c
Copyright (C) 2010 g10 Code GmbH
This file is part of libgpg-error.
libgpg-error is free software; you can redistribute it and/or
modify it under the terms of the GNU Lesser General Public License
as published by the Free Software Foundation; either version 2.1 of
the License, or (at your option) any later version.
libgpg-error is distributed in the hope that it will be useful, but
WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public
- License along with this program; if not, see .
+ License along with this program; if not, see .
*/
#ifndef INIT_H
#define INIT_H
#if HAVE_W32_SYSTEM
/* Forward declaration - defined in w32-gettext.c. */
struct loaded_domain;
/* An item for a linked list of loaded domains. */
struct domainlist_s
{
struct domainlist_s *next;
char *dname; /* Directory name for the mo file. */
char *fname; /* File name for the MO file. */
int load_failed; /* True if loading the domain failed. */
struct loaded_domain *domain; /* NULL if not loaded. Never changed
once set to non-NULL. */
char name[1]; /* Name of the domain. Never changed
once set. */
};
/* 119 bytes for an error message should be enough. With this size we
can assume that the allocation does not take up more than 128 bytes
per thread. Note that this is only used for W32CE. */
#define STRBUFFER_SIZE 120
/* The TLS space definition. */
struct tls_space_s
{
/* Flag used by w32-gettext. */
int gt_use_utf8;
#ifdef HAVE_W32CE_SYSTEM
char strerror_buffer[STRBUFFER_SIZE];
#endif
};
/* Return the TLS. */
struct tls_space_s *get_tls (void);
/* Explicit constructor for w32-gettext.c */
#ifndef DLL_EXPORT
void _gpg_w32__init_gettext_module (void);
#endif
#endif /*HAVE_W32_SYSTEM*/
#endif /*INIT_H*/
diff --git a/src/lock.h b/src/lock.h
index b60f2c2..a830b36 100644
--- a/src/lock.h
+++ b/src/lock.h
@@ -1,24 +1,24 @@
/* lock.h - Declarations for *-lock.c
Copyright (C) 2014 g10 Code GmbH
This file is part of libgpg-error.
libgpg-error is free software; you can redistribute it and/or
modify it under the terms of the GNU Lesser General Public License
as published by the Free Software Foundation; either version 2.1 of
the License, or (at your option) any later version.
libgpg-error is distributed in the hope that it will be useful, but
WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public
- License along with this program; if not, see .
+ License along with this program; if not, see .
*/
#ifndef LOCK_H
#define LOCK_H
#endif /*LOCK_H*/
diff --git a/src/mkw32errmap.c b/src/mkw32errmap.c
index 68d0f05..508a513 100644
--- a/src/mkw32errmap.c
+++ b/src/mkw32errmap.c
@@ -1,178 +1,178 @@
/* mkw32errmap.c - Generate mapping sources for Windows.
Copyright (C) 2010 g10 Code GmbH
This file is part of libgpg-error.
libgpg-error is free software; you can redistribute it and/or
modify it under the terms of the GNU Lesser General Public License
as published by the Free Software Foundation; either version 2.1 of
the License, or (at your option) any later version.
libgpg-error is distributed in the hope that it will be useful, but
WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public
- License along with this program; if not, see .
+ License along with this program; if not, see .
*/
#ifdef RESOLVE_MACROS
# include
#endif
#include
#include
static const char header_gpg_extra_errno_h[] =
"/* errno.h - WindowsCE errno.h substitute\n"
" Copyright (C) 2010 g10 Code GmbH\n"
"\n"
" This file is free software; as a special exception the author gives\n"
" unlimited permission to copy and/or distribute it, with or without\n"
" modifications, as long as this notice is preserved.\n"
"\n"
" This file is distributed in the hope that it will be useful, but\n"
" WITHOUT ANY WARRANTY, to the extent permitted by law; without even\n"
" the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR\n"
" PURPOSE.\n"
"\n"
" +++ Do not edit! File has been generated by mkw32errmap.c +++\n"
"\n"
" This file is intended to be used with ming32ce-gcc to implement an\n"
" errno substitute under WindowsCE. It must be included via gcc's\n"
" -idirafter option. The gpg-error-config script emits the\n"
" appropriate option snippet. The actual implementation of the errno\n"
" related functions are part of libgpg-error. A separate header file\n"
" is required because errno.h is often included before gpg-error.h.\n"
" */\n"
"\n"
"#ifndef _GPG_ERROR_EXTRA_ERRNO_H\n"
"#define _GPG_ERROR_EXTRA_ERRNO_H\n"
"\n"
"/* Due to peculiarities in W32 we can't implement ERRNO as an\n"
" writable lvalue. This also allows us to easily find places\n"
" where ERRNO is being written to. See also gpg_err_set_errno. */\n"
"int _gpg_w32ce_get_errno (void);\n"
"#define errno (_gpg_w32ce_get_errno ())\n"
"\n";
static const char footer_gpg_extra_errno_h[] =
"\n"
"#endif /*_GPG_ERROR_EXTRA_ERRNO_H*/\n";
/* The table below is used in two modes. First we run the host
preprocessor over it to generate a new include file. This include
file has the same content but the Windows error macros are
resolved. This is so that we don't need to include winerror.h into
the generated errno.h. The mkw32errmap_marker variable is only
here to have something to grep for after preprocessing. */
static int mkw32errmap_marker;
struct table_s
{
int *dummy;
const char *name;
int w32code;
int w32code2;
};
struct table_s table[] =
{
#ifdef RESOLVE_MACROS
#define X(a,b,c) \
{&mkw32errmap_marker, (a), (b), (c)}
X( "EPERM", ERROR_CANNOT_MAKE , 0 ),
X( "ENOENT", ERROR_FILE_NOT_FOUND , ERROR_PATH_NOT_FOUND ),
X( "EINTR", ERROR_INVALID_AT_INTERRUPT_TIME , 0 ),
X( "EIO", ERROR_IO_DEVICE , 0 ),
X( "ENXIO", ERROR_FILE_INVALID , 0 ),
X( "EBADF", ERROR_INVALID_HANDLE , 0 ),
X( "EAGAIN", ERROR_MORE_DATA , WSAEWOULDBLOCK ),
X( "ENOMEM", ERROR_NOT_ENOUGH_MEMORY , 0 ),
X( "EACCES", ERROR_ACCESS_DENIED , 0 ),
X( "EFAULT", ERROR_PROCESS_ABORTED , 0 ),
X( "EBUSY", ERROR_BUSY , 0 ),
X( "EEXIST", ERROR_FILE_EXISTS , WSAEADDRINUSE ),
X( "EXDEV", ERROR_NOT_SAME_DEVICE , 0 ),
X( "ENODEV", ERROR_BAD_DEVICE , ERROR_DEV_NOT_EXIST ),
X( "ENOTDIR",ERROR_DIRECTORY , 0 ),
X( "EINVAL", ERROR_INVALID_PARAMETER , 0 ),
X( "ENFILE", ERROR_NO_MORE_FILES , 0 ),
X( "EMFILE", ERROR_TOO_MANY_OPEN_FILES , 0 ),
X( "ENOSPC", ERROR_DISK_FULL , 0 ),
X( "EROFS", ERROR_WRITE_PROTECT , 0 ),
X( "EPIPE", ERROR_BROKEN_PIPE , 0 ),
X( "ERANGE", ERROR_ARITHMETIC_OVERFLOW , 0 ),
X( "EDEADLOCK",ERROR_POSSIBLE_DEADLOCK , 0 ),
X( "ENAMETOOLONG", ERROR_FILENAME_EXCED_RANGE, 0 ),
X( "ENOLCK", ERROR_SHARING_BUFFER_EXCEEDED , 0 ),
X( "ENOSYS", ERROR_NOT_SUPPORTED , 0 ),
X( "ENOTEMPTY",ERROR_DIR_NOT_EMPTY , 0 ),
X( "ESPIPE", ERROR_SEEK_ON_DEVICE , 0 ),
#if 0 /* FIXME: Find appropriate mappings. */
X( "EILSEQ", ),
X( "EDOM", ),
X( "EMLINK", ),
X( "ESRCH", ), /* No such process */
X( "E2BIG", ), /* Arg list too long */
X( "ENOEXEC", ), /* Exec format error */
X( "ECHILD", ), /* No child processes */
X( "EISDIR", ), /* Is a directory */
X( "ENOTTY", ), /* Inappropriate I/O control operation */
X( "EFBIG", ), /* File too large */
#endif
#undef X
#else /*!RESOLVE_MACROS*/
# include "mkw32errmap.tab.h"
#endif /*!RESOLVE_MACROS*/
{ NULL, 0 }
};
static int
compare_table (const void *a_v, const void *b_v)
{
const struct table_s *a = a_v;
const struct table_s *b = b_v;
return (a->w32code - b->w32code);
}
int
main (int argc, char **argv)
{
int idx;
for (idx=0; table[idx].name; idx++)
;
qsort (table, idx, sizeof *table, compare_table);
if (argc == 2 && !strcmp (argv[1], "--map"))
{
fputs ("static int\n"
"map_w32codes (int err)\n"
"{\n"
" switch (err)\n"
" {\n", stdout );
for (idx=0; table[idx].name; idx++)
if (table[idx].w32code2)
printf (" case %d: return %d;\n",
table[idx].w32code2, table[idx].w32code);
fputs (" default: return err;\n"
" }\n"
"}\n", stdout);
}
else
{
fputs (header_gpg_extra_errno_h, stdout);
for (idx=0; table[idx].name; idx++)
printf ("#define %-12s %5d\n", table[idx].name, table[idx].w32code);
fputs (footer_gpg_extra_errno_h, stdout);
}
return 0;
}
diff --git a/src/posix-lock-obj.h b/src/posix-lock-obj.h
index 872e55a..08e0bac 100644
--- a/src/posix-lock-obj.h
+++ b/src/posix-lock-obj.h
@@ -1,42 +1,42 @@
/* posic-lock-obj.h - Declaration of the POSIX lock object
Copyright (C) 2014 g10 Code GmbH
This file is part of libgpg-error.
libgpg-error is free software; you can redistribute it and/or
modify it under the terms of the GNU Lesser General Public License
as published by the Free Software Foundation; either version 2.1 of
the License, or (at your option) any later version.
libgpg-error is distributed in the hope that it will be useful, but
WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public
- License along with this program; if not, see .
+ License along with this program; if not, see .
*/
#ifndef POSIX_LOCK_OBJ_H
#define POSIX_LOCK_OBJ_H
#define LOCK_ABI_NOT_AVAILABLE (-1)
#if USE_POSIX_THREADS
# define LOCK_ABI_VERSION 1
#else
# define LOCK_ABI_VERSION LOCK_ABI_NOT_AVAILABLE
#endif
typedef struct
{
long vers;
#if USE_POSIX_THREADS
union {
pthread_mutex_t mtx;
long *dummy;
} u;
#endif
} _gpgrt_lock_t;
#endif /*POSIX_LOCK_OBJ_H*/
diff --git a/src/posix-lock.c b/src/posix-lock.c
index d8f5465..2e0ae92 100644
--- a/src/posix-lock.c
+++ b/src/posix-lock.c
@@ -1,262 +1,262 @@
/* posix-lock.c - GPGRT lock functions for POSIX systems
Copyright (C) 2005-2009 Free Software Foundation, Inc.
Copyright (C) 2014 g10 Code GmbH
This file is part of libgpg-error.
libgpg-error is free software; you can redistribute it and/or
modify it under the terms of the GNU Lesser General Public License
as published by the Free Software Foundation; either version 2.1 of
the License, or (at your option) any later version.
libgpg-error is distributed in the hope that it will be useful, but
WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public
- License along with this program; if not, see .
+ License along with this program; if not, see .
Parts of the code, in particular use_pthreads_p, are based on code
from gettext, written by Bruno Haible , 2005.
*/
#if HAVE_CONFIG_H
#include
#endif
#ifdef HAVE_W32_SYSTEM
# error This module may not be build for Windows.
#endif
#include
#include
#include
#include
#include
#if USE_POSIX_THREADS
# include
#endif
#include "gpg-error.h"
#include "lock.h"
#include "posix-lock-obj.h"
#if USE_POSIX_THREADS
# if USE_POSIX_THREADS_WEAK
/* On ELF systems it is easy to use pthreads using weak
references. Take care not to test the address of a weak
referenced function we actually use; some GCC versions have a
bug were &foo != NULL is always evaluated to true in PIC mode. */
# pragma weak pthread_cancel
# pragma weak pthread_mutex_init
# pragma weak pthread_mutex_lock
# pragma weak pthread_mutex_trylock
# pragma weak pthread_mutex_unlock
# pragma weak pthread_mutex_destroy
# if ! PTHREAD_IN_USE_DETECTION_HARD
# define use_pthread_p() (!!pthread_cancel)
# endif
# else /*!USE_POSIX_THREADS_WEAK*/
# if ! PTHREAD_IN_USE_DETECTION_HARD
# define use_pthread_p() (1)
# endif
# endif /*!USE_POSIX_THREADS_WEAK*/
# if PTHREAD_IN_USE_DETECTION_HARD
/* The function to be executed by a dummy thread. */
static void *
dummy_thread_func (void *arg)
{
return arg;
}
static int
use_pthread_p (void)
{
static int tested;
static int result; /* 1: linked with -lpthread, 0: only with libc */
if (!tested)
{
pthread_t thread;
if (pthread_create (&thread, NULL, dummy_thread_func, NULL))
result = 0; /* Thread creation failed. */
else
{
/* Thread creation works. */
void *retval;
if (pthread_join (thread, &retval) != 0)
{
assert (!"pthread_join");
abort ();
}
result = 1;
}
tested = 1;
}
return result;
}
#endif /*PTHREAD_IN_USE_DETECTION_HARD*/
#endif /*USE_POSIX_THREADS*/
static _gpgrt_lock_t *
get_lock_object (gpgrt_lock_t *lockhd)
{
_gpgrt_lock_t *lock = (_gpgrt_lock_t*)lockhd;
if (lock->vers != LOCK_ABI_VERSION)
{
assert (!"lock ABI version");
abort ();
}
if (sizeof (gpgrt_lock_t) < sizeof (_gpgrt_lock_t))
{
assert (!"sizeof lock obj");
abort ();
}
return lock;
}
gpg_err_code_t
_gpgrt_lock_init (gpgrt_lock_t *lockhd)
{
_gpgrt_lock_t *lock = (_gpgrt_lock_t*)lockhd;
int rc;
/* If VERS is zero we assume that no static initialization has been
done, so we setup our ABI version right here. The caller might
have called us to test whether lock support is at all available. */
if (!lock->vers)
{
if (sizeof (gpgrt_lock_t) < sizeof (_gpgrt_lock_t))
{
assert (!"sizeof lock obj");
abort ();
}
lock->vers = LOCK_ABI_VERSION;
}
else /* Run the usual check. */
lock = get_lock_object (lockhd);
#if USE_POSIX_THREADS
if (use_pthread_p())
{
rc = pthread_mutex_init (&lock->u.mtx, NULL);
if (rc)
rc = gpg_err_code_from_errno (rc);
}
else
rc = 0; /* Threads are not used. */
#else /* Unknown thread system. */
rc = lock->vers == LOCK_ABI_NOT_AVAILABLE? 0 : GPG_ERR_NOT_IMPLEMENTED;
#endif /* Unknown thread system. */
return rc;
}
gpg_err_code_t
_gpgrt_lock_lock (gpgrt_lock_t *lockhd)
{
_gpgrt_lock_t *lock = get_lock_object (lockhd);
int rc;
#if USE_POSIX_THREADS
if (use_pthread_p())
{
rc = pthread_mutex_lock (&lock->u.mtx);
if (rc)
rc = gpg_err_code_from_errno (rc);
}
else
rc = 0; /* Threads are not used. */
#else /* Unknown thread system. */
rc = lock->vers == LOCK_ABI_NOT_AVAILABLE? 0 : GPG_ERR_NOT_IMPLEMENTED;
#endif /* Unknown thread system. */
return rc;
}
gpg_err_code_t
_gpgrt_lock_trylock (gpgrt_lock_t *lockhd)
{
_gpgrt_lock_t *lock = get_lock_object (lockhd);
int rc;
#if USE_POSIX_THREADS
if (use_pthread_p())
{
rc = pthread_mutex_trylock (&lock->u.mtx);
if (rc)
rc = gpg_err_code_from_errno (rc);
}
else
rc = 0; /* Threads are not used. */
#else /* Unknown thread system. */
rc = lock->vers == LOCK_ABI_NOT_AVAILABLE? 0 : GPG_ERR_NOT_IMPLEMENTED;
#endif /* Unknown thread system. */
return rc;
}
gpg_err_code_t
_gpgrt_lock_unlock (gpgrt_lock_t *lockhd)
{
_gpgrt_lock_t *lock = get_lock_object (lockhd);
int rc;
#if USE_POSIX_THREADS
if (use_pthread_p())
{
rc = pthread_mutex_unlock (&lock->u.mtx);
if (rc)
rc = gpg_err_code_from_errno (rc);
}
else
rc = 0; /* Threads are not used. */
#else /* Unknown thread system. */
rc = lock->vers == LOCK_ABI_NOT_AVAILABLE? 0 : GPG_ERR_NOT_IMPLEMENTED;
#endif /* Unknown thread system. */
return rc;
}
/* Note: Use this function only if no other thread holds or waits for
this lock. */
gpg_err_code_t
_gpgrt_lock_destroy (gpgrt_lock_t *lockhd)
{
_gpgrt_lock_t *lock = get_lock_object (lockhd);
int rc;
#if USE_POSIX_THREADS
if (use_pthread_p())
{
rc = pthread_mutex_destroy (&lock->u.mtx);
if (rc)
rc = gpg_err_code_from_errno (rc);
else
{
/* Re-init the mutex so that it can be re-used. */
gpgrt_lock_t tmp = GPGRT_LOCK_INITIALIZER;
memcpy (lockhd, &tmp, sizeof tmp);
}
}
else
rc = 0; /* Threads are not used. */
#else /* Unknown thread system. */
rc = lock->vers == LOCK_ABI_NOT_AVAILABLE? 0 : GPG_ERR_NOT_IMPLEMENTED;
#endif /* Unknown thread system. */
return rc;
}
diff --git a/src/posix-thread.c b/src/posix-thread.c
index bef0386..270dc91 100644
--- a/src/posix-thread.c
+++ b/src/posix-thread.c
@@ -1,63 +1,63 @@
/* posix-thread.c - GPGRT thread functions for POSIX systems
Copyright (C) 2014 g10 Code GmbH
This file is part of libgpg-error.
libgpg-error is free software; you can redistribute it and/or
modify it under the terms of the GNU Lesser General Public License
as published by the Free Software Foundation; either version 2.1 of
the License, or (at your option) any later version.
libgpg-error is distributed in the hope that it will be useful, but
WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public
- License along with this program; if not, see .
+ License along with this program; if not, see .
*/
#if HAVE_CONFIG_H
#include
#endif
#ifdef HAVE_W32_SYSTEM
# error This module may not be build for Windows.
#endif
#include
#include
#include
#include
#include /* Get posix option macros. */
#if USE_POSIX_THREADS
# ifdef _POSIX_PRIORITY_SCHEDULING
# include
# endif
#elif USE_SOLARIS_THREADS
# include
#endif
#include "gpg-error.h"
#include "thread.h"
gpg_err_code_t
_gpgrt_yield (void)
{
#if USE_POSIX_THREADS
# ifdef _POSIX_PRIORITY_SCHEDULING
sched_yield ();
# else
return GPG_ERR_NOT_SUPPORTED;
# endif
#elif USE_SOLARIS_THREADS
thr_yield ();
#else
return GPG_ERR_NOT_SUPPORTED;
#endif
return 0;
}
diff --git a/src/thread.h b/src/thread.h
index 0b2cf04..c650a99 100644
--- a/src/thread.h
+++ b/src/thread.h
@@ -1,24 +1,24 @@
/* thread.h - Declarations for *-thread.c
Copyright (C) 2014 g10 Code GmbH
This file is part of libgpg-error.
libgpg-error is free software; you can redistribute it and/or
modify it under the terms of the GNU Lesser General Public License
as published by the Free Software Foundation; either version 2.1 of
the License, or (at your option) any later version.
libgpg-error is distributed in the hope that it will be useful, but
WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public
- License along with this program; if not, see .
+ License along with this program; if not, see .
*/
#ifndef THREAD_H
#define THREAD_H
#endif /*THREAD_H*/
diff --git a/src/version.c b/src/version.c
index 216fee4..d0c408d 100644
--- a/src/version.c
+++ b/src/version.c
@@ -1,121 +1,121 @@
/* version.c - Version checking
* Copyright (C) 2001, 2002, 2012, 2013, 2014 g10 Code GmbH
*
* This file is part of libgpg-error.
*
* libgpg-error is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public License
* as published by the Free Software Foundation; either version 2.1 of
* the License, or (at your option) any later version.
*
* libgpg-error is distributed in the hope that it will be useful, but
* WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
- * License along with this program; if not, see .
+ * License along with this program; if not, see .
*/
#if HAVE_CONFIG_H
#include
#endif
#include
#include
#include
#include
#define digitp(a) ((a) >= '0' && (a) <= '9')
/* This is actually a dummy function to make sure that is module is
not empty. Some compilers barf on empty modules. */
static const char *
cright_blurb (void)
{
static const char blurb[] =
"\n\n"
"This is Libgpg-error " PACKAGE_VERSION " - An error code library\n"
"Copyright 2003, 2004, 2010, 2013, 2014, 2015 g10 Code GmbH\n"
"\n"
"(" BUILD_REVISION " " BUILD_TIMESTAMP ")\n"
"\n\n";
return blurb;
}
static const char*
parse_version_number (const char *s, int *number)
{
int val = 0;
if (*s == '0' && digitp (s[1]))
return NULL; /* Leading zeros are not allowed. */
for (; digitp (*s); s++)
{
val *= 10;
val += *s - '0';
}
*number = val;
return val < 0 ? NULL : s;
}
static const char *
parse_version_string (const char *s, int *major, int *minor)
{
s = parse_version_number (s, major);
if (!s || *s != '.')
return NULL;
s++;
s = parse_version_number (s, minor);
if (!s)
return NULL;
return s; /* Patchlevel. */
}
static const char *
compare_versions (const char *my_version, const char *req_version)
{
int my_major, my_minor;
int rq_major, rq_minor;
const char *my_plvl, *rq_plvl;
if (!req_version)
return my_version;
if (!my_version)
return NULL;
my_plvl = parse_version_string (my_version, &my_major, &my_minor);
if (!my_plvl)
return NULL; /* Very strange: our own version is bogus. */
rq_plvl = parse_version_string(req_version, &rq_major, &rq_minor);
if (!rq_plvl)
return NULL; /* Requested version string is invalid. */
if (my_major > rq_major
|| (my_major == rq_major && my_minor >= rq_minor))
{
return my_version;
}
return NULL;
}
/*
* Check that the the version of the library is at minimum REQ_VERSION
* and return the actual version string; return NULL if the condition
* is not met. If NULL is passed to this function, no check is done
* and the version string is simply returned.
*/
const char *
_gpg_error_check_version (const char *req_version)
{
if (req_version && req_version[0] == 1 && req_version[1] == 1)
return cright_blurb ();
return compare_versions (PACKAGE_VERSION, req_version);
}
diff --git a/src/visibility.c b/src/visibility.c
index 4750685..e3ac8a7 100644
--- a/src/visibility.c
+++ b/src/visibility.c
@@ -1,707 +1,707 @@
/* visibility.c - Wrapper for all public functions.
* Copyright (C) 2014 g10 Code GmbH
*
* This file is part of libgpg-error.
*
* libgpg-error is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public License
* as published by the Free Software Foundation; either version 2.1 of
* the License, or (at your option) any later version.
*
* libgpg-error is distributed in the hope that it will be useful, but
* WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
- * License along with this program; if not, see .
+ * License along with this program; if not, see .
*/
#include
#include
#define _GPGRT_INCL_BY_VISIBILITY_C 1
#include "gpgrt-int.h"
const char *
gpg_strerror (gpg_error_t err)
{
return _gpg_strerror (err);
}
int
gpg_strerror_r (gpg_error_t err, char *buf, size_t buflen)
{
return _gpg_strerror_r (err, buf, buflen);
}
const char *
gpg_strsource (gpg_error_t err)
{
return _gpg_strsource (err);
}
gpg_err_code_t
gpg_err_code_from_errno (int err)
{
return _gpg_err_code_from_errno (err);
}
int
gpg_err_code_to_errno (gpg_err_code_t code)
{
return _gpg_err_code_to_errno (code);
}
gpg_err_code_t
gpg_err_code_from_syserror (void)
{
return _gpg_err_code_from_syserror ();
}
void
gpg_err_set_errno (int err)
{
_gpg_err_set_errno (err);
}
gpg_error_t
gpg_err_init (void)
{
return _gpg_err_init ();
}
void
gpg_err_deinit (int mode)
{
_gpg_err_deinit (mode);
}
const char *
gpg_error_check_version (const char *req_version)
{
return _gpg_error_check_version (req_version);
}
const char *
gpgrt_check_version (const char *req_version)
{
return _gpg_error_check_version (req_version);
}
void
gpgrt_set_syscall_clamp (void (*pre)(void), void (*post)(void))
{
_gpgrt_set_syscall_clamp (pre, post);
}
void
gpgrt_set_alloc_func (void *(*f)(void *a, size_t n))
{
_gpgrt_set_alloc_func (f);
}
gpg_err_code_t
gpgrt_lock_init (gpgrt_lock_t *lockhd)
{
return _gpgrt_lock_init (lockhd);
}
gpg_err_code_t
gpgrt_lock_lock (gpgrt_lock_t *lockhd)
{
return _gpgrt_lock_lock (lockhd);
}
gpg_err_code_t
gpgrt_lock_trylock (gpgrt_lock_t *lockhd)
{
return _gpgrt_lock_trylock (lockhd);
}
gpg_err_code_t
gpgrt_lock_unlock (gpgrt_lock_t *lockhd)
{
return _gpgrt_lock_unlock (lockhd);
}
gpg_err_code_t
gpgrt_lock_destroy (gpgrt_lock_t *lockhd)
{
return _gpgrt_lock_destroy (lockhd);
}
gpg_err_code_t
gpgrt_yield (void)
{
return _gpgrt_yield ();
}
estream_t
gpgrt_fopen (const char *_GPGRT__RESTRICT path,
const char *_GPGRT__RESTRICT mode)
{
return _gpgrt_fopen (path, mode);
}
estream_t
gpgrt_mopen (void *_GPGRT__RESTRICT data, size_t data_n, size_t data_len,
unsigned int grow,
void *(*func_realloc) (void *mem, size_t size),
void (*func_free) (void *mem),
const char *_GPGRT__RESTRICT mode)
{
return _gpgrt_mopen (data, data_n, data_len, grow, func_realloc, func_free,
mode);
}
estream_t
gpgrt_fopenmem (size_t memlimit, const char *_GPGRT__RESTRICT mode)
{
return _gpgrt_fopenmem (memlimit, mode);
}
estream_t
gpgrt_fopenmem_init (size_t memlimit, const char *_GPGRT__RESTRICT mode,
const void *data, size_t datalen)
{
return _gpgrt_fopenmem_init (memlimit, mode, data, datalen);
}
estream_t
gpgrt_fdopen (int filedes, const char *mode)
{
return _gpgrt_fdopen (filedes, mode);
}
estream_t
gpgrt_fdopen_nc (int filedes, const char *mode)
{
return _gpgrt_fdopen_nc (filedes, mode);
}
estream_t
gpgrt_sysopen (es_syshd_t *syshd, const char *mode)
{
return _gpgrt_sysopen (syshd, mode);
}
estream_t
gpgrt_sysopen_nc (es_syshd_t *syshd, const char *mode)
{
return _gpgrt_sysopen_nc (syshd, mode);
}
estream_t
gpgrt_fpopen (FILE *fp, const char *mode)
{
return _gpgrt_fpopen (fp, mode);
}
estream_t
gpgrt_fpopen_nc (FILE *fp, const char *mode)
{
return _gpgrt_fpopen_nc (fp, mode);
}
estream_t
gpgrt_freopen (const char *_GPGRT__RESTRICT path,
const char *_GPGRT__RESTRICT mode,
estream_t _GPGRT__RESTRICT stream)
{
return _gpgrt_freopen (path, mode, stream);
}
estream_t
gpgrt_fopencookie (void *_GPGRT__RESTRICT cookie,
const char *_GPGRT__RESTRICT mode,
gpgrt_cookie_io_functions_t functions)
{
return _gpgrt_fopencookie (cookie, mode, functions);
}
int
gpgrt_fclose (estream_t stream)
{
return _gpgrt_fclose (stream);
}
int
gpgrt_fclose_snatch (estream_t stream, void **r_buffer, size_t *r_buflen)
{
return _gpgrt_fclose_snatch (stream, r_buffer, r_buflen);
}
int
gpgrt_onclose (estream_t stream, int mode,
void (*fnc) (estream_t, void*), void *fnc_value)
{
return _gpgrt_onclose (stream, mode, fnc, fnc_value);
}
int
gpgrt_fileno (estream_t stream)
{
return _gpgrt_fileno (stream);
}
int
gpgrt_fileno_unlocked (estream_t stream)
{
return _gpgrt_fileno_unlocked (stream);
}
int
gpgrt_syshd (estream_t stream, es_syshd_t *syshd)
{
return _gpgrt_syshd (stream, syshd);
}
int
gpgrt_syshd_unlocked (estream_t stream, es_syshd_t *syshd)
{
return _gpgrt_syshd_unlocked (stream, syshd);
}
void
_gpgrt_set_std_fd (int no, int fd)
{
_gpgrt__set_std_fd (no, fd); /* (double dash in name) */
}
estream_t
_gpgrt_get_std_stream (int fd)
{
return _gpgrt__get_std_stream (fd); /* (double dash in name) */
}
void
gpgrt_flockfile (estream_t stream)
{
_gpgrt_flockfile (stream);
}
int
gpgrt_ftrylockfile (estream_t stream)
{
return _gpgrt_ftrylockfile (stream);
}
void
gpgrt_funlockfile (estream_t stream)
{
_gpgrt_funlockfile (stream);
}
int
_gpgrt_pending (estream_t stream)
{
return _gpgrt__pending (stream);
}
int
_gpgrt_pending_unlocked (estream_t stream)
{
return _gpgrt__pending_unlocked (stream);
}
int
gpgrt_feof (estream_t stream)
{
return _gpgrt_feof (stream);
}
int
gpgrt_feof_unlocked (estream_t stream)
{
return _gpgrt_feof_unlocked (stream);
}
int
gpgrt_ferror (estream_t stream)
{
return _gpgrt_ferror (stream);
}
int
gpgrt_ferror_unlocked (estream_t stream)
{
return _gpgrt_ferror_unlocked (stream);
}
void
gpgrt_clearerr (estream_t stream)
{
_gpgrt_clearerr (stream);
}
void
gpgrt_clearerr_unlocked (estream_t stream)
{
_gpgrt_clearerr_unlocked (stream);
}
int
gpgrt_fflush (estream_t stream)
{
return _gpgrt_fflush (stream);
}
int
gpgrt_fseek (estream_t stream, long int offset, int whence)
{
return _gpgrt_fseek (stream, offset, whence);
}
int
gpgrt_fseeko (estream_t stream, gpgrt_off_t offset, int whence)
{
return _gpgrt_fseeko (stream, offset, whence);
}
long int
gpgrt_ftell (estream_t stream)
{
return _gpgrt_ftell (stream);
}
gpgrt_off_t
gpgrt_ftello (estream_t stream)
{
return _gpgrt_ftello (stream);
}
void
gpgrt_rewind (estream_t stream)
{
_gpgrt_rewind (stream);
}
int
gpgrt_fgetc (estream_t stream)
{
return _gpgrt_fgetc (stream);
}
int
_gpgrt_getc_underflow (estream_t stream)
{
return _gpgrt__getc_underflow (stream);
}
int
gpgrt_fputc (int c, estream_t stream)
{
return _gpgrt_fputc (c, stream);
}
int
_gpgrt_putc_overflow (int c, estream_t stream)
{
return _gpgrt__putc_overflow (c, stream);
}
int
gpgrt_ungetc (int c, estream_t stream)
{
return _gpgrt_ungetc (c, stream);
}
int
gpgrt_read (estream_t _GPGRT__RESTRICT stream,
void *_GPGRT__RESTRICT buffer, size_t bytes_to_read,
size_t *_GPGRT__RESTRICT bytes_read)
{
return _gpgrt_read (stream, buffer, bytes_to_read, bytes_read);
}
int
gpgrt_write (estream_t _GPGRT__RESTRICT stream,
const void *_GPGRT__RESTRICT buffer, size_t bytes_to_write,
size_t *_GPGRT__RESTRICT bytes_written)
{
return _gpgrt_write (stream, buffer, bytes_to_write, bytes_written);
}
int
gpgrt_write_sanitized (estream_t _GPGRT__RESTRICT stream,
const void * _GPGRT__RESTRICT buffer, size_t length,
const char * delimiters,
size_t * _GPGRT__RESTRICT bytes_written)
{
return _gpgrt_write_sanitized (stream, buffer, length, delimiters,
bytes_written);
}
int
gpgrt_write_hexstring (estream_t _GPGRT__RESTRICT stream,
const void *_GPGRT__RESTRICT buffer, size_t length,
int reserved, size_t *_GPGRT__RESTRICT bytes_written )
{
return _gpgrt_write_hexstring (stream, buffer, length, reserved,
bytes_written);
}
size_t
gpgrt_fread (void *_GPGRT__RESTRICT ptr, size_t size, size_t nitems,
estream_t _GPGRT__RESTRICT stream)
{
return _gpgrt_fread (ptr, size, nitems, stream);
}
size_t
gpgrt_fwrite (const void *_GPGRT__RESTRICT ptr, size_t size, size_t nitems,
estream_t _GPGRT__RESTRICT stream)
{
return _gpgrt_fwrite (ptr, size, nitems, stream);
}
char *
gpgrt_fgets (char *_GPGRT__RESTRICT buffer, int length,
estream_t _GPGRT__RESTRICT stream)
{
return _gpgrt_fgets (buffer, length, stream);
}
int
gpgrt_fputs (const char *_GPGRT__RESTRICT s, estream_t _GPGRT__RESTRICT stream)
{
return _gpgrt_fputs (s, stream);
}
int
gpgrt_fputs_unlocked (const char *_GPGRT__RESTRICT s,
estream_t _GPGRT__RESTRICT stream)
{
return _gpgrt_fputs_unlocked (s, stream);
}
gpgrt_ssize_t
gpgrt_getline (char *_GPGRT__RESTRICT *_GPGRT__RESTRICT lineptr,
size_t *_GPGRT__RESTRICT n, estream_t _GPGRT__RESTRICT stream)
{
return _gpgrt_getline (lineptr, n, stream);
}
gpgrt_ssize_t
gpgrt_read_line (estream_t stream,
char **addr_of_buffer, size_t *length_of_buffer,
size_t *max_length)
{
return _gpgrt_read_line (stream, addr_of_buffer, length_of_buffer,
max_length);
}
void
gpgrt_free (void *a)
{
if (a)
_gpgrt_free (a);
}
int
gpgrt_vfprintf (estream_t _GPGRT__RESTRICT stream,
const char *_GPGRT__RESTRICT format,
va_list ap)
{
return _gpgrt_vfprintf (stream, format, ap);
}
int
gpgrt_vfprintf_unlocked (estream_t _GPGRT__RESTRICT stream,
const char *_GPGRT__RESTRICT format,
va_list ap)
{
return _gpgrt_vfprintf_unlocked (stream, format, ap);
}
int
gpgrt_printf (const char *_GPGRT__RESTRICT format, ...)
{
va_list ap;
int rc;
va_start (ap, format);
rc = _gpgrt_vfprintf (es_stdout, format, ap);
va_end (ap);
return rc;
}
int
gpgrt_printf_unlocked (const char *_GPGRT__RESTRICT format, ...)
{
va_list ap;
int rc;
va_start (ap, format);
rc = _gpgrt_vfprintf_unlocked (es_stdout, format, ap);
va_end (ap);
return rc;
}
int
gpgrt_fprintf (estream_t _GPGRT__RESTRICT stream,
const char *_GPGRT__RESTRICT format, ...)
{
va_list ap;
int rc;
va_start (ap, format);
rc = _gpgrt_vfprintf (stream, format, ap);
va_end (ap);
return rc;
}
int
gpgrt_fprintf_unlocked (estream_t _GPGRT__RESTRICT stream,
const char *_GPGRT__RESTRICT format, ...)
{
va_list ap;
int rc;
va_start (ap, format);
rc = _gpgrt_vfprintf_unlocked (stream, format, ap);
va_end (ap);
return rc;
}
int
gpgrt_setvbuf (estream_t _GPGRT__RESTRICT stream,
char *_GPGRT__RESTRICT buf, int type, size_t size)
{
return _gpgrt_setvbuf (stream, buf, type, size);
}
void
gpgrt_setbuf (estream_t _GPGRT__RESTRICT stream, char *_GPGRT__RESTRICT buf)
{
_gpgrt_setvbuf (stream, buf, buf? _IOFBF : _IONBF, BUFSIZ);
}
void
gpgrt_set_binary (estream_t stream)
{
_gpgrt_set_binary (stream);
}
int
gpgrt_set_nonblock (estream_t stream, int onoff)
{
return _gpgrt_set_nonblock (stream, onoff);
}
int
gpgrt_get_nonblock (estream_t stream)
{
return _gpgrt_get_nonblock (stream);
}
int
gpgrt_poll (gpgrt_poll_t *fds, unsigned int nfds, int timeout)
{
return _gpgrt_poll (fds, nfds, timeout);
}
estream_t
gpgrt_tmpfile (void)
{
return _gpgrt_tmpfile ();
}
void
gpgrt_opaque_set (estream_t stream, void *opaque)
{
_gpgrt_opaque_set (stream, opaque);
}
void *
gpgrt_opaque_get (estream_t stream)
{
return _gpgrt_opaque_get (stream);
}
void
gpgrt_fname_set (estream_t stream, const char *fname)
{
_gpgrt_fname_set (stream, fname);
}
const char *
gpgrt_fname_get (estream_t stream)
{
return _gpgrt_fname_get (stream);
}
int
gpgrt_asprintf (char **r_buf, const char *_GPGRT__RESTRICT format, ...)
{
va_list ap;
int rc;
va_start (ap, format);
rc = _gpgrt_estream_vasprintf (r_buf, format, ap);
va_end (ap);
return rc;
}
int
gpgrt_vasprintf (char **r_buf, const char *_GPGRT__RESTRICT format, va_list ap)
{
return _gpgrt_estream_vasprintf (r_buf, format, ap);
}
char *
gpgrt_bsprintf (const char *_GPGRT__RESTRICT format, ...)
{
int rc;
va_list ap;
char *buf;
va_start (ap, format);
rc = _gpgrt_estream_vasprintf (&buf, format, ap);
va_end (ap);
if (rc < 0)
return NULL;
return buf;
}
char *
gpgrt_vbsprintf (const char *_GPGRT__RESTRICT format, va_list ap)
{
int rc;
char *buf;
rc = _gpgrt_estream_vasprintf (&buf, format, ap);
if (rc < 0)
return NULL;
return buf;
}
int
gpgrt_snprintf (char *buf, size_t bufsize, const char *format, ...)
{
int rc;
va_list arg_ptr;
va_start (arg_ptr, format);
rc = _gpgrt_estream_vsnprintf (buf, bufsize, format, arg_ptr);
va_end (arg_ptr);
return rc;
}
int
gpgrt_vsnprintf (char *buf, size_t bufsize,
const char *format, va_list arg_ptr)
{
return _gpgrt_estream_vsnprintf (buf, bufsize, format, arg_ptr);
}
diff --git a/src/visibility.h b/src/visibility.h
index ec3a124..1de6c62 100644
--- a/src/visibility.h
+++ b/src/visibility.h
@@ -1,259 +1,259 @@
/* visibility.h - Set visibility attribute
* Copyright (C) 2014 g10 Code GmbH
*
* This file is part of libgpg-error.
*
* libgpg-error is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public License
* as published by the Free Software Foundation; either version 2.1 of
* the License, or (at your option) any later version.
*
* libgpg-error is distributed in the hope that it will be useful, but
* WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
- * License along with this program; if not, see .
+ * License along with this program; if not, see .
*/
#ifndef _GPGRT_VISIBILITY_H
#define _GPGRT_VISIBILITY_H
/* Include the main header here so that public symbols are mapped to
the internal underscored ones. */
#ifdef _GPGRT_INCL_BY_VISIBILITY_C
# include "gpgrt-int.h"
#endif
/* Our use of the ELF visibility feature works by passing
-fvisibiliy=hidden on the command line and by explicitly marking
all exported functions as visible.
NOTE: When adding new functions, please make sure to add them to
gpg-error.vers and gpg-error.def.in as well. */
#ifdef _GPGRT_INCL_BY_VISIBILITY_C
# ifdef GPGRT_USE_VISIBILITY
# define MARK_VISIBLE(name) \
extern __typeof__ (name) name __attribute__ ((visibility("default")));
# else
# define MARK_VISIBLE(name) /* */
# endif
MARK_VISIBLE (gpg_strerror)
MARK_VISIBLE (gpg_strerror_r)
MARK_VISIBLE (gpg_strsource)
MARK_VISIBLE (gpg_err_code_from_errno)
MARK_VISIBLE (gpg_err_code_to_errno)
MARK_VISIBLE (gpg_err_code_from_syserror)
MARK_VISIBLE (gpg_err_set_errno)
MARK_VISIBLE (gpg_err_init)
MARK_VISIBLE (gpg_err_deinit)
MARK_VISIBLE (gpg_error_check_version)
MARK_VISIBLE (gpgrt_check_version)
MARK_VISIBLE (gpgrt_lock_init)
MARK_VISIBLE (gpgrt_lock_lock)
MARK_VISIBLE (gpgrt_lock_unlock)
MARK_VISIBLE (gpgrt_lock_destroy)
MARK_VISIBLE (gpgrt_yield)
MARK_VISIBLE (gpgrt_lock_trylock)
MARK_VISIBLE (gpgrt_fopen)
MARK_VISIBLE (gpgrt_mopen)
MARK_VISIBLE (gpgrt_fopenmem)
MARK_VISIBLE (gpgrt_fopenmem_init)
MARK_VISIBLE (gpgrt_fdopen)
MARK_VISIBLE (gpgrt_fdopen_nc)
MARK_VISIBLE (gpgrt_sysopen)
MARK_VISIBLE (gpgrt_sysopen_nc)
MARK_VISIBLE (gpgrt_fpopen)
MARK_VISIBLE (gpgrt_fpopen_nc)
MARK_VISIBLE (gpgrt_freopen)
MARK_VISIBLE (gpgrt_fopencookie)
MARK_VISIBLE (gpgrt_fclose)
MARK_VISIBLE (gpgrt_fclose_snatch)
MARK_VISIBLE (gpgrt_onclose)
MARK_VISIBLE (gpgrt_fileno)
MARK_VISIBLE (gpgrt_fileno_unlocked)
MARK_VISIBLE (gpgrt_syshd)
MARK_VISIBLE (gpgrt_syshd_unlocked)
MARK_VISIBLE (_gpgrt_set_std_fd)
MARK_VISIBLE (_gpgrt_get_std_stream)
MARK_VISIBLE (gpgrt_flockfile)
MARK_VISIBLE (gpgrt_ftrylockfile)
MARK_VISIBLE (gpgrt_funlockfile)
MARK_VISIBLE (_gpgrt_pending)
MARK_VISIBLE (_gpgrt_pending_unlocked)
MARK_VISIBLE (gpgrt_feof)
MARK_VISIBLE (gpgrt_feof_unlocked)
MARK_VISIBLE (gpgrt_ferror)
MARK_VISIBLE (gpgrt_ferror_unlocked)
MARK_VISIBLE (gpgrt_clearerr)
MARK_VISIBLE (gpgrt_clearerr_unlocked)
MARK_VISIBLE (gpgrt_fflush)
MARK_VISIBLE (gpgrt_fseek)
MARK_VISIBLE (gpgrt_fseeko)
MARK_VISIBLE (gpgrt_ftell)
MARK_VISIBLE (gpgrt_ftello)
MARK_VISIBLE (gpgrt_rewind)
MARK_VISIBLE (gpgrt_fgetc)
MARK_VISIBLE (_gpgrt_getc_underflow)
MARK_VISIBLE (gpgrt_fputc)
MARK_VISIBLE (_gpgrt_putc_overflow)
MARK_VISIBLE (gpgrt_ungetc)
MARK_VISIBLE (gpgrt_read)
MARK_VISIBLE (gpgrt_write)
MARK_VISIBLE (gpgrt_write_sanitized)
MARK_VISIBLE (gpgrt_write_hexstring)
MARK_VISIBLE (gpgrt_fread)
MARK_VISIBLE (gpgrt_fwrite)
MARK_VISIBLE (gpgrt_fgets)
MARK_VISIBLE (gpgrt_fputs)
MARK_VISIBLE (gpgrt_fputs_unlocked)
MARK_VISIBLE (gpgrt_getline)
MARK_VISIBLE (gpgrt_read_line)
MARK_VISIBLE (gpgrt_free)
MARK_VISIBLE (gpgrt_fprintf)
MARK_VISIBLE (gpgrt_fprintf_unlocked)
MARK_VISIBLE (gpgrt_printf)
MARK_VISIBLE (gpgrt_printf_unlocked)
MARK_VISIBLE (gpgrt_vfprintf)
MARK_VISIBLE (gpgrt_vfprintf_unlocked)
MARK_VISIBLE (gpgrt_setvbuf)
MARK_VISIBLE (gpgrt_setbuf)
MARK_VISIBLE (gpgrt_set_binary)
MARK_VISIBLE (gpgrt_set_nonblock)
MARK_VISIBLE (gpgrt_get_nonblock)
MARK_VISIBLE (gpgrt_poll)
MARK_VISIBLE (gpgrt_tmpfile)
MARK_VISIBLE (gpgrt_opaque_set)
MARK_VISIBLE (gpgrt_opaque_get)
MARK_VISIBLE (gpgrt_fname_set)
MARK_VISIBLE (gpgrt_fname_get)
MARK_VISIBLE (gpgrt_asprintf)
MARK_VISIBLE (gpgrt_vasprintf)
MARK_VISIBLE (gpgrt_bsprintf)
MARK_VISIBLE (gpgrt_vbsprintf)
MARK_VISIBLE (gpgrt_snprintf)
MARK_VISIBLE (gpgrt_vsnprintf)
MARK_VISIBLE (gpgrt_set_syscall_clamp)
MARK_VISIBLE (gpgrt_set_alloc_func)
#undef MARK_VISIBLE
#else /*!_GPGRT_INCL_BY_VISIBILITY_C*/
/* To avoid accidental use of the public functions inside Libgpg-error,
we redefine them to catch such errors. */
#define gpg_strerror _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpg_strerror_r _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpg_strsource _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpg_err_code_from_errno _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpg_err_code_to_errno _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpg_err_code_from_syserror _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpg_err_set_errno _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpg_err_init _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpg_err_deinit _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpg_error_check_version _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_check_version _gpgrt_USE_OTHER_FUNCTION
#define gpgrt_lock_init _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_lock_lock _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_lock_unlock _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_lock_destroy _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_yield _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_lock_trylock _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_fopen _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_mopen _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_fopenmem _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_fopenmem_init _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_fdopen _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_fdopen_nc _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_sysopen _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_sysopen_nc _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_fpopen _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_fpopen_nc _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_freopen _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_fopencookie _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_fclose _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_fclose_snatch _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_onclose _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_fileno _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_fileno_unlocked _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_syshd _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_syshd_unlocked _gpgrt_USE_UNDERSCORED_FUNCTION
#define _gpgrt_set_std_fd _gpgrt_USE_UNDERSCORED_FUNCTION
#define _gpgrt_get_std_stream _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_flockfile _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_ftrylockfile _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_funlockfile _gpgrt_USE_UNDERSCORED_FUNCTION
#define _gpgrt_pending _gpgrt_USE_UNDERSCORED_FUNCTION
#define _gpgrt_pending_unlocked _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_feof _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_feof_unlocked _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_ferror _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_ferror_unlocked _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_clearerr _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_clearerr_unlocked _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_fflush _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_fseek _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_fseeko _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_ftell _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_ftello _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_rewind _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_fgetc _gpgrt_USE_UNDERSCORED_FUNCTION
#define _gpgrt_getc_underflow _gpgrt_USE_DBLUNDERSCO_FUNCTION
#define gpgrt_fputc _gpgrt_USE_UNDERSCORED_FUNCTION
#define _gpgrt_putc_overflow _gpgrt_USE_DBLUNDERSCO_FUNCTION
#define gpgrt_ungetc _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_read _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_write _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_write_sanitized _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_write_hexstring _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_fread _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_fwrite _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_fgets _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_fputs _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_fputs_unlocked _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_getline _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_read_line _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_free _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_fprintf _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_fprintf_unlocked _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_printf _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_printf_unlocked _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_vfprintf _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_vfprintf_unlocked _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_setvbuf _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_setbuf _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_set_binary _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_set_nonblock _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_get_nonblock _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_poll _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_tmpfile _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_opaque_set _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_opaque_get _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_fname_set _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_fname_get _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_asprintf _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_vasprintf _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_bsprintf _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_vbsprintf _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_snprintf _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_vsnprintf _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_set_syscall_clamp _gpgrt_USE_UNDERSCORED_FUNCTION
#define gpgrt_set_alloc_func _gpgrt_USE_UNDERSCORED_FUNCTION
#endif /*!_GPGRT_INCL_BY_VISIBILITY_C*/
#endif /*_GPGRT_VISIBILITY_H*/
diff --git a/src/w32-gettext.c b/src/w32-gettext.c
index 146de53..734b9c1 100644
--- a/src/w32-gettext.c
+++ b/src/w32-gettext.c
@@ -1,1963 +1,1963 @@
/* w32-gettext.h - A simple gettext implementation for Windows targets.
Copyright (C) 1995, 1996, 1997, 1999, 2005, 2007,
2008, 2010 Free Software Foundation, Inc.
This file is part of libgpg-error.
libgpg-error is free software; you can redistribute it and/or
modify it under the terms of the GNU Lesser General Public License
as published by the Free Software Foundation; either version 2.1 of
the License, or (at your option) any later version.
libgpg-error is distributed in the hope that it will be useful, but
WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public
- License along with this program; if not, see .
+ License along with this program; if not, see .
*/
#if HAVE_CONFIG_H
#include
#endif
#if !defined (_WIN32) && !defined (__CYGWIN32__)
# error This module may only be build for Windows or Cygwin32
#endif
#include
#include
#include
#include
#include
#ifdef HAVE_SYS_TYPES_H
#include
#endif
#include
#ifndef HAVE_W32CE_SYSTEM
# include
#endif /*HAVE_W32CE_SYSTEM*/
#include
#ifdef JNLIB_IN_JNLIB
#include "libjnlib-config.h"
#endif
#ifndef jnlib_malloc
# define jnlib_malloc(a) malloc ((a))
# define jnlib_calloc(a,b) calloc ((a), (b))
# define jnlib_free(a) free ((a))
# define jnlib_xstrdup(a) not_used
#endif /*!jnlib_malloc*/
#include "init.h"
#include "gpg-error.h"
#ifdef HAVE_W32CE_SYSTEM
/* Forward declaration. */
static wchar_t *utf8_to_wchar (const char *string, size_t length, size_t *retlen);
static HANDLE
MyCreateFileA (LPCSTR lpFileName, DWORD dwDesiredAccess, DWORD dwSharedMode,
LPSECURITY_ATTRIBUTES lpSecurityAttributes,
DWORD dwCreationDisposition, DWORD dwFlagsAndAttributes,
HANDLE hTemplateFile)
{
wchar_t *filename;
HANDLE result;
int err;
size_t size;
filename = utf8_to_wchar (lpFileName, -1, &size);
if (!filename)
return INVALID_HANDLE_VALUE;
result = CreateFileW (filename, dwDesiredAccess, dwSharedMode,
lpSecurityAttributes, dwCreationDisposition,
dwFlagsAndAttributes, hTemplateFile);
err = GetLastError ();
free (filename);
SetLastError (err);
return result;
}
#undef CreateFileA
#define CreateFileA MyCreateFileA
#endif
/* localname.c from gettext BEGIN. */
/* Determine the current selected locale.
Copyright (C) 1995-1999, 2000-2003 Free Software Foundation, Inc.
This program is free software; you can redistribute it and/or modify it
under the terms of the GNU Library General Public License as published
by the Free Software Foundation; either version 2, or (at your option)
any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
Library General Public License for more details.
You should have received a copy of the GNU Library General Public
License along with this program; if not, write to the Free Software
Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307,
USA. */
/* Written by Ulrich Drepper , 1995. */
/* Win32 code written by Tor Lillqvist . */
/* Renamed _nl_locale_name, removed unused args, removed include files,
non-W32 code and changed comments . */
/* Mingw headers don't have latest language and sublanguage codes. */
#ifndef LANG_AFRIKAANS
#define LANG_AFRIKAANS 0x36
#endif
#ifndef LANG_ALBANIAN
#define LANG_ALBANIAN 0x1c
#endif
#ifndef LANG_AMHARIC
#define LANG_AMHARIC 0x5e
#endif
#ifndef LANG_ARABIC
#define LANG_ARABIC 0x01
#endif
#ifndef LANG_ARMENIAN
#define LANG_ARMENIAN 0x2b
#endif
#ifndef LANG_ASSAMESE
#define LANG_ASSAMESE 0x4d
#endif
#ifndef LANG_AZERI
#define LANG_AZERI 0x2c
#endif
#ifndef LANG_BASQUE
#define LANG_BASQUE 0x2d
#endif
#ifndef LANG_BELARUSIAN
#define LANG_BELARUSIAN 0x23
#endif
#ifndef LANG_BENGALI
#define LANG_BENGALI 0x45
#endif
#ifndef LANG_BURMESE
#define LANG_BURMESE 0x55
#endif
#ifndef LANG_CAMBODIAN
#define LANG_CAMBODIAN 0x53
#endif
#ifndef LANG_CATALAN
#define LANG_CATALAN 0x03
#endif
#ifndef LANG_CHEROKEE
#define LANG_CHEROKEE 0x5c
#endif
#ifndef LANG_DIVEHI
#define LANG_DIVEHI 0x65
#endif
#ifndef LANG_EDO
#define LANG_EDO 0x66
#endif
#ifndef LANG_ESTONIAN
#define LANG_ESTONIAN 0x25
#endif
#ifndef LANG_FAEROESE
#define LANG_FAEROESE 0x38
#endif
#ifndef LANG_FARSI
#define LANG_FARSI 0x29
#endif
#ifndef LANG_FRISIAN
#define LANG_FRISIAN 0x62
#endif
#ifndef LANG_FULFULDE
#define LANG_FULFULDE 0x67
#endif
#ifndef LANG_GAELIC
#define LANG_GAELIC 0x3c
#endif
#ifndef LANG_GALICIAN
#define LANG_GALICIAN 0x56
#endif
#ifndef LANG_GEORGIAN
#define LANG_GEORGIAN 0x37
#endif
#ifndef LANG_GUARANI
#define LANG_GUARANI 0x74
#endif
#ifndef LANG_GUJARATI
#define LANG_GUJARATI 0x47
#endif
#ifndef LANG_HAUSA
#define LANG_HAUSA 0x68
#endif
#ifndef LANG_HAWAIIAN
#define LANG_HAWAIIAN 0x75
#endif
#ifndef LANG_HEBREW
#define LANG_HEBREW 0x0d
#endif
#ifndef LANG_HINDI
#define LANG_HINDI 0x39
#endif
#ifndef LANG_IBIBIO
#define LANG_IBIBIO 0x69
#endif
#ifndef LANG_IGBO
#define LANG_IGBO 0x70
#endif
#ifndef LANG_INDONESIAN
#define LANG_INDONESIAN 0x21
#endif
#ifndef LANG_INUKTITUT
#define LANG_INUKTITUT 0x5d
#endif
#ifndef LANG_KANNADA
#define LANG_KANNADA 0x4b
#endif
#ifndef LANG_KANURI
#define LANG_KANURI 0x71
#endif
#ifndef LANG_KASHMIRI
#define LANG_KASHMIRI 0x60
#endif
#ifndef LANG_KAZAK
#define LANG_KAZAK 0x3f
#endif
#ifndef LANG_KONKANI
#define LANG_KONKANI 0x57
#endif
#ifndef LANG_KYRGYZ
#define LANG_KYRGYZ 0x40
#endif
#ifndef LANG_LAO
#define LANG_LAO 0x54
#endif
#ifndef LANG_LATIN
#define LANG_LATIN 0x76
#endif
#ifndef LANG_LATVIAN
#define LANG_LATVIAN 0x26
#endif
#ifndef LANG_LITHUANIAN
#define LANG_LITHUANIAN 0x27
#endif
#ifndef LANG_MACEDONIAN
#define LANG_MACEDONIAN 0x2f
#endif
#ifndef LANG_MALAY
#define LANG_MALAY 0x3e
#endif
#ifndef LANG_MALAYALAM
#define LANG_MALAYALAM 0x4c
#endif
#ifndef LANG_MALTESE
#define LANG_MALTESE 0x3a
#endif
#ifndef LANG_MANIPURI
#define LANG_MANIPURI 0x58
#endif
#ifndef LANG_MARATHI
#define LANG_MARATHI 0x4e
#endif
#ifndef LANG_MONGOLIAN
#define LANG_MONGOLIAN 0x50
#endif
#ifndef LANG_NEPALI
#define LANG_NEPALI 0x61
#endif
#ifndef LANG_ORIYA
#define LANG_ORIYA 0x48
#endif
#ifndef LANG_OROMO
#define LANG_OROMO 0x72
#endif
#ifndef LANG_PAPIAMENTU
#define LANG_PAPIAMENTU 0x79
#endif
#ifndef LANG_PASHTO
#define LANG_PASHTO 0x63
#endif
#ifndef LANG_PUNJABI
#define LANG_PUNJABI 0x46
#endif
#ifndef LANG_RHAETO_ROMANCE
#define LANG_RHAETO_ROMANCE 0x17
#endif
#ifndef LANG_SAAMI
#define LANG_SAAMI 0x3b
#endif
#ifndef LANG_SANSKRIT
#define LANG_SANSKRIT 0x4f
#endif
#ifndef LANG_SERBIAN
#define LANG_SERBIAN 0x1a
#endif
#ifndef LANG_SINDHI
#define LANG_SINDHI 0x59
#endif
#ifndef LANG_SINHALESE
#define LANG_SINHALESE 0x5b
#endif
#ifndef LANG_SLOVAK
#define LANG_SLOVAK 0x1b
#endif
#ifndef LANG_SOMALI
#define LANG_SOMALI 0x77
#endif
#ifndef LANG_SORBIAN
#define LANG_SORBIAN 0x2e
#endif
#ifndef LANG_SUTU
#define LANG_SUTU 0x30
#endif
#ifndef LANG_SWAHILI
#define LANG_SWAHILI 0x41
#endif
#ifndef LANG_SYRIAC
#define LANG_SYRIAC 0x5a
#endif
#ifndef LANG_TAGALOG
#define LANG_TAGALOG 0x64
#endif
#ifndef LANG_TAJIK
#define LANG_TAJIK 0x28
#endif
#ifndef LANG_TAMAZIGHT
#define LANG_TAMAZIGHT 0x5f
#endif
#ifndef LANG_TAMIL
#define LANG_TAMIL 0x49
#endif
#ifndef LANG_TATAR
#define LANG_TATAR 0x44
#endif
#ifndef LANG_TELUGU
#define LANG_TELUGU 0x4a
#endif
#ifndef LANG_THAI
#define LANG_THAI 0x1e
#endif
#ifndef LANG_TIBETAN
#define LANG_TIBETAN 0x51
#endif
#ifndef LANG_TIGRINYA
#define LANG_TIGRINYA 0x73
#endif
#ifndef LANG_TSONGA
#define LANG_TSONGA 0x31
#endif
#ifndef LANG_TSWANA
#define LANG_TSWANA 0x32
#endif
#ifndef LANG_TURKMEN
#define LANG_TURKMEN 0x42
#endif
#ifndef LANG_UKRAINIAN
#define LANG_UKRAINIAN 0x22
#endif
#ifndef LANG_URDU
#define LANG_URDU 0x20
#endif
#ifndef LANG_UZBEK
#define LANG_UZBEK 0x43
#endif
#ifndef LANG_VENDA
#define LANG_VENDA 0x33
#endif
#ifndef LANG_VIETNAMESE
#define LANG_VIETNAMESE 0x2a
#endif
#ifndef LANG_WELSH
#define LANG_WELSH 0x52
#endif
#ifndef LANG_XHOSA
#define LANG_XHOSA 0x34
#endif
#ifndef LANG_YI
#define LANG_YI 0x78
#endif
#ifndef LANG_YIDDISH
#define LANG_YIDDISH 0x3d
#endif
#ifndef LANG_YORUBA
#define LANG_YORUBA 0x6a
#endif
#ifndef LANG_ZULU
#define LANG_ZULU 0x35
#endif
#ifndef SUBLANG_ARABIC_SAUDI_ARABIA
#define SUBLANG_ARABIC_SAUDI_ARABIA 0x01
#endif
#ifndef SUBLANG_ARABIC_IRAQ
#define SUBLANG_ARABIC_IRAQ 0x02
#endif
#ifndef SUBLANG_ARABIC_EGYPT
#define SUBLANG_ARABIC_EGYPT 0x03
#endif
#ifndef SUBLANG_ARABIC_LIBYA
#define SUBLANG_ARABIC_LIBYA 0x04
#endif
#ifndef SUBLANG_ARABIC_ALGERIA
#define SUBLANG_ARABIC_ALGERIA 0x05
#endif
#ifndef SUBLANG_ARABIC_MOROCCO
#define SUBLANG_ARABIC_MOROCCO 0x06
#endif
#ifndef SUBLANG_ARABIC_TUNISIA
#define SUBLANG_ARABIC_TUNISIA 0x07
#endif
#ifndef SUBLANG_ARABIC_OMAN
#define SUBLANG_ARABIC_OMAN 0x08
#endif
#ifndef SUBLANG_ARABIC_YEMEN
#define SUBLANG_ARABIC_YEMEN 0x09
#endif
#ifndef SUBLANG_ARABIC_SYRIA
#define SUBLANG_ARABIC_SYRIA 0x0a
#endif
#ifndef SUBLANG_ARABIC_JORDAN
#define SUBLANG_ARABIC_JORDAN 0x0b
#endif
#ifndef SUBLANG_ARABIC_LEBANON
#define SUBLANG_ARABIC_LEBANON 0x0c
#endif
#ifndef SUBLANG_ARABIC_KUWAIT
#define SUBLANG_ARABIC_KUWAIT 0x0d
#endif
#ifndef SUBLANG_ARABIC_UAE
#define SUBLANG_ARABIC_UAE 0x0e
#endif
#ifndef SUBLANG_ARABIC_BAHRAIN
#define SUBLANG_ARABIC_BAHRAIN 0x0f
#endif
#ifndef SUBLANG_ARABIC_QATAR
#define SUBLANG_ARABIC_QATAR 0x10
#endif
#ifndef SUBLANG_AZERI_LATIN
#define SUBLANG_AZERI_LATIN 0x01
#endif
#ifndef SUBLANG_AZERI_CYRILLIC
#define SUBLANG_AZERI_CYRILLIC 0x02
#endif
#ifndef SUBLANG_BENGALI_INDIA
#define SUBLANG_BENGALI_INDIA 0x01
#endif
#ifndef SUBLANG_BENGALI_BANGLADESH
#define SUBLANG_BENGALI_BANGLADESH 0x02
#endif
#ifndef SUBLANG_CHINESE_MACAU
#define SUBLANG_CHINESE_MACAU 0x05
#endif
#ifndef SUBLANG_ENGLISH_SOUTH_AFRICA
#define SUBLANG_ENGLISH_SOUTH_AFRICA 0x07
#endif
#ifndef SUBLANG_ENGLISH_JAMAICA
#define SUBLANG_ENGLISH_JAMAICA 0x08
#endif
#ifndef SUBLANG_ENGLISH_CARIBBEAN
#define SUBLANG_ENGLISH_CARIBBEAN 0x09
#endif
#ifndef SUBLANG_ENGLISH_BELIZE
#define SUBLANG_ENGLISH_BELIZE 0x0a
#endif
#ifndef SUBLANG_ENGLISH_TRINIDAD
#define SUBLANG_ENGLISH_TRINIDAD 0x0b
#endif
#ifndef SUBLANG_ENGLISH_ZIMBABWE
#define SUBLANG_ENGLISH_ZIMBABWE 0x0c
#endif
#ifndef SUBLANG_ENGLISH_PHILIPPINES
#define SUBLANG_ENGLISH_PHILIPPINES 0x0d
#endif
#ifndef SUBLANG_ENGLISH_INDONESIA
#define SUBLANG_ENGLISH_INDONESIA 0x0e
#endif
#ifndef SUBLANG_ENGLISH_HONGKONG
#define SUBLANG_ENGLISH_HONGKONG 0x0f
#endif
#ifndef SUBLANG_ENGLISH_INDIA
#define SUBLANG_ENGLISH_INDIA 0x10
#endif
#ifndef SUBLANG_ENGLISH_MALAYSIA
#define SUBLANG_ENGLISH_MALAYSIA 0x11
#endif
#ifndef SUBLANG_ENGLISH_SINGAPORE
#define SUBLANG_ENGLISH_SINGAPORE 0x12
#endif
#ifndef SUBLANG_FRENCH_LUXEMBOURG
#define SUBLANG_FRENCH_LUXEMBOURG 0x05
#endif
#ifndef SUBLANG_FRENCH_MONACO
#define SUBLANG_FRENCH_MONACO 0x06
#endif
#ifndef SUBLANG_FRENCH_WESTINDIES
#define SUBLANG_FRENCH_WESTINDIES 0x07
#endif
#ifndef SUBLANG_FRENCH_REUNION
#define SUBLANG_FRENCH_REUNION 0x08
#endif
#ifndef SUBLANG_FRENCH_CONGO
#define SUBLANG_FRENCH_CONGO 0x09
#endif
#ifndef SUBLANG_FRENCH_SENEGAL
#define SUBLANG_FRENCH_SENEGAL 0x0a
#endif
#ifndef SUBLANG_FRENCH_CAMEROON
#define SUBLANG_FRENCH_CAMEROON 0x0b
#endif
#ifndef SUBLANG_FRENCH_COTEDIVOIRE
#define SUBLANG_FRENCH_COTEDIVOIRE 0x0c
#endif
#ifndef SUBLANG_FRENCH_MALI
#define SUBLANG_FRENCH_MALI 0x0d
#endif
#ifndef SUBLANG_FRENCH_MOROCCO
#define SUBLANG_FRENCH_MOROCCO 0x0e
#endif
#ifndef SUBLANG_FRENCH_HAITI
#define SUBLANG_FRENCH_HAITI 0x0f
#endif
#ifndef SUBLANG_GERMAN_LUXEMBOURG
#define SUBLANG_GERMAN_LUXEMBOURG 0x04
#endif
#ifndef SUBLANG_GERMAN_LIECHTENSTEIN
#define SUBLANG_GERMAN_LIECHTENSTEIN 0x05
#endif
#ifndef SUBLANG_KASHMIRI_INDIA
#define SUBLANG_KASHMIRI_INDIA 0x02
#endif
#ifndef SUBLANG_MALAY_MALAYSIA
#define SUBLANG_MALAY_MALAYSIA 0x01
#endif
#ifndef SUBLANG_MALAY_BRUNEI_DARUSSALAM
#define SUBLANG_MALAY_BRUNEI_DARUSSALAM 0x02
#endif
#ifndef SUBLANG_NEPALI_INDIA
#define SUBLANG_NEPALI_INDIA 0x02
#endif
#ifndef SUBLANG_PUNJABI_INDIA
#define SUBLANG_PUNJABI_INDIA 0x01
#endif
#ifndef SUBLANG_ROMANIAN_ROMANIA
#define SUBLANG_ROMANIAN_ROMANIA 0x01
#endif
#ifndef SUBLANG_SERBIAN_LATIN
#define SUBLANG_SERBIAN_LATIN 0x02
#endif
#ifndef SUBLANG_SERBIAN_CYRILLIC
#define SUBLANG_SERBIAN_CYRILLIC 0x03
#endif
#ifndef SUBLANG_SINDHI_INDIA
#define SUBLANG_SINDHI_INDIA 0x00
#endif
#ifndef SUBLANG_SINDHI_PAKISTAN
#define SUBLANG_SINDHI_PAKISTAN 0x01
#endif
#ifndef SUBLANG_SPANISH_GUATEMALA
#define SUBLANG_SPANISH_GUATEMALA 0x04
#endif
#ifndef SUBLANG_SPANISH_COSTA_RICA
#define SUBLANG_SPANISH_COSTA_RICA 0x05
#endif
#ifndef SUBLANG_SPANISH_PANAMA
#define SUBLANG_SPANISH_PANAMA 0x06
#endif
#ifndef SUBLANG_SPANISH_DOMINICAN_REPUBLIC
#define SUBLANG_SPANISH_DOMINICAN_REPUBLIC 0x07
#endif
#ifndef SUBLANG_SPANISH_VENEZUELA
#define SUBLANG_SPANISH_VENEZUELA 0x08
#endif
#ifndef SUBLANG_SPANISH_COLOMBIA
#define SUBLANG_SPANISH_COLOMBIA 0x09
#endif
#ifndef SUBLANG_SPANISH_PERU
#define SUBLANG_SPANISH_PERU 0x0a
#endif
#ifndef SUBLANG_SPANISH_ARGENTINA
#define SUBLANG_SPANISH_ARGENTINA 0x0b
#endif
#ifndef SUBLANG_SPANISH_ECUADOR
#define SUBLANG_SPANISH_ECUADOR 0x0c
#endif
#ifndef SUBLANG_SPANISH_CHILE
#define SUBLANG_SPANISH_CHILE 0x0d
#endif
#ifndef SUBLANG_SPANISH_URUGUAY
#define SUBLANG_SPANISH_URUGUAY 0x0e
#endif
#ifndef SUBLANG_SPANISH_PARAGUAY
#define SUBLANG_SPANISH_PARAGUAY 0x0f
#endif
#ifndef SUBLANG_SPANISH_BOLIVIA
#define SUBLANG_SPANISH_BOLIVIA 0x10
#endif
#ifndef SUBLANG_SPANISH_EL_SALVADOR
#define SUBLANG_SPANISH_EL_SALVADOR 0x11
#endif
#ifndef SUBLANG_SPANISH_HONDURAS
#define SUBLANG_SPANISH_HONDURAS 0x12
#endif
#ifndef SUBLANG_SPANISH_NICARAGUA
#define SUBLANG_SPANISH_NICARAGUA 0x13
#endif
#ifndef SUBLANG_SPANISH_PUERTO_RICO
#define SUBLANG_SPANISH_PUERTO_RICO 0x14
#endif
#ifndef SUBLANG_SWEDISH_FINLAND
#define SUBLANG_SWEDISH_FINLAND 0x02
#endif
#ifndef SUBLANG_TAMAZIGHT_ARABIC
#define SUBLANG_TAMAZIGHT_ARABIC 0x01
#endif
#ifndef SUBLANG_TAMAZIGHT_LATIN
#define SUBLANG_TAMAZIGHT_LATIN 0x02
#endif
#ifndef SUBLANG_TIGRINYA_ETHIOPIA
#define SUBLANG_TIGRINYA_ETHIOPIA 0x00
#endif
#ifndef SUBLANG_TIGRINYA_ERITREA
#define SUBLANG_TIGRINYA_ERITREA 0x01
#endif
#ifndef SUBLANG_URDU_PAKISTAN
#define SUBLANG_URDU_PAKISTAN 0x01
#endif
#ifndef SUBLANG_URDU_INDIA
#define SUBLANG_URDU_INDIA 0x02
#endif
#ifndef SUBLANG_UZBEK_LATIN
#define SUBLANG_UZBEK_LATIN 0x01
#endif
#ifndef SUBLANG_UZBEK_CYRILLIC
#define SUBLANG_UZBEK_CYRILLIC 0x02
#endif
/* Return an XPG style locale name
language[_territory[.codeset]][@modifier].
Don't even bother determining the codeset; it's not useful in this
context, because message catalogs are not specific to a single
codeset. The result must not be freed; it is statically
allocated. */
static const char *
my_nl_locale_name (const char *categoryname)
{
#ifndef HAVE_W32CE_SYSTEM
const char *retval;
#endif
LCID lcid;
LANGID langid;
int primary, sub;
/* Let the user override the system settings through environment
variables, as on POSIX systems. */
#ifndef HAVE_W32CE_SYSTEM
retval = getenv ("LC_ALL");
if (retval != NULL && retval[0] != '\0')
return retval;
retval = getenv (categoryname);
if (retval != NULL && retval[0] != '\0')
return retval;
retval = getenv ("LANG");
if (retval != NULL && retval[0] != '\0')
return retval;
#endif /*!HAVE_W32CE_SYSTEM*/
/* Use native Win32 API locale ID. */
#ifdef HAVE_W32CE_SYSTEM
lcid = GetSystemDefaultLCID ();
#else
lcid = GetThreadLocale ();
#endif
/* Strip off the sorting rules, keep only the language part. */
langid = LANGIDFROMLCID (lcid);
/* Split into language and territory part. */
primary = PRIMARYLANGID (langid);
sub = SUBLANGID (langid);
/* Dispatch on language.
See also http://www.unicode.org/unicode/onlinedat/languages.html .
- For details about languages, see http://www.ethnologue.com/ . */
+ For details about languages, see https://www.ethnologue.com/ . */
switch (primary)
{
case LANG_AFRIKAANS: return "af_ZA";
case LANG_ALBANIAN: return "sq_AL";
case LANG_AMHARIC: return "am_ET";
case LANG_ARABIC:
switch (sub)
{
case SUBLANG_ARABIC_SAUDI_ARABIA: return "ar_SA";
case SUBLANG_ARABIC_IRAQ: return "ar_IQ";
case SUBLANG_ARABIC_EGYPT: return "ar_EG";
case SUBLANG_ARABIC_LIBYA: return "ar_LY";
case SUBLANG_ARABIC_ALGERIA: return "ar_DZ";
case SUBLANG_ARABIC_MOROCCO: return "ar_MA";
case SUBLANG_ARABIC_TUNISIA: return "ar_TN";
case SUBLANG_ARABIC_OMAN: return "ar_OM";
case SUBLANG_ARABIC_YEMEN: return "ar_YE";
case SUBLANG_ARABIC_SYRIA: return "ar_SY";
case SUBLANG_ARABIC_JORDAN: return "ar_JO";
case SUBLANG_ARABIC_LEBANON: return "ar_LB";
case SUBLANG_ARABIC_KUWAIT: return "ar_KW";
case SUBLANG_ARABIC_UAE: return "ar_AE";
case SUBLANG_ARABIC_BAHRAIN: return "ar_BH";
case SUBLANG_ARABIC_QATAR: return "ar_QA";
}
return "ar";
case LANG_ARMENIAN: return "hy_AM";
case LANG_ASSAMESE: return "as_IN";
case LANG_AZERI:
switch (sub)
{
/* FIXME: Adjust this when Azerbaijani locales appear on Unix. */
case SUBLANG_AZERI_LATIN: return "az_AZ@latin";
case SUBLANG_AZERI_CYRILLIC: return "az_AZ@cyrillic";
}
return "az";
case LANG_BASQUE:
return "eu"; /* Ambiguous: could be "eu_ES" or "eu_FR". */
case LANG_BELARUSIAN: return "be_BY";
case LANG_BENGALI:
switch (sub)
{
case SUBLANG_BENGALI_INDIA: return "bn_IN";
case SUBLANG_BENGALI_BANGLADESH: return "bn_BD";
}
return "bn";
case LANG_BULGARIAN: return "bg_BG";
case LANG_BURMESE: return "my_MM";
case LANG_CAMBODIAN: return "km_KH";
case LANG_CATALAN: return "ca_ES";
case LANG_CHEROKEE: return "chr_US";
case LANG_CHINESE:
switch (sub)
{
case SUBLANG_CHINESE_TRADITIONAL: return "zh_TW";
case SUBLANG_CHINESE_SIMPLIFIED: return "zh_CN";
case SUBLANG_CHINESE_HONGKONG: return "zh_HK";
case SUBLANG_CHINESE_SINGAPORE: return "zh_SG";
case SUBLANG_CHINESE_MACAU: return "zh_MO";
}
return "zh";
case LANG_CROATIAN: /* LANG_CROATIAN == LANG_SERBIAN
* What used to be called Serbo-Croatian
* should really now be two separate
* languages because of political reasons.
* (Says tml, who knows nothing about Serbian
* or Croatian.)
* (I can feel those flames coming already.)
*/
switch (sub)
{
case SUBLANG_DEFAULT: return "hr_HR";
case SUBLANG_SERBIAN_LATIN: return "sr_CS";
case SUBLANG_SERBIAN_CYRILLIC: return "sr_CS@cyrillic";
}
return "hr";
case LANG_CZECH: return "cs_CZ";
case LANG_DANISH: return "da_DK";
case LANG_DIVEHI: return "div_MV";
case LANG_DUTCH:
switch (sub)
{
case SUBLANG_DUTCH: return "nl_NL";
case SUBLANG_DUTCH_BELGIAN: /* FLEMISH, VLAAMS */ return "nl_BE";
}
return "nl";
case LANG_EDO: return "bin_NG";
case LANG_ENGLISH:
switch (sub)
{
/* SUBLANG_ENGLISH_US == SUBLANG_DEFAULT. Heh. I thought
* English was the language spoken in England.
* Oh well.
*/
case SUBLANG_ENGLISH_US: return "en_US";
case SUBLANG_ENGLISH_UK: return "en_GB";
case SUBLANG_ENGLISH_AUS: return "en_AU";
case SUBLANG_ENGLISH_CAN: return "en_CA";
case SUBLANG_ENGLISH_NZ: return "en_NZ";
case SUBLANG_ENGLISH_EIRE: return "en_IE";
case SUBLANG_ENGLISH_SOUTH_AFRICA: return "en_ZA";
case SUBLANG_ENGLISH_JAMAICA: return "en_JM";
case SUBLANG_ENGLISH_CARIBBEAN: return "en_GD"; /* Grenada? */
case SUBLANG_ENGLISH_BELIZE: return "en_BZ";
case SUBLANG_ENGLISH_TRINIDAD: return "en_TT";
case SUBLANG_ENGLISH_ZIMBABWE: return "en_ZW";
case SUBLANG_ENGLISH_PHILIPPINES: return "en_PH";
case SUBLANG_ENGLISH_INDONESIA: return "en_ID";
case SUBLANG_ENGLISH_HONGKONG: return "en_HK";
case SUBLANG_ENGLISH_INDIA: return "en_IN";
case SUBLANG_ENGLISH_MALAYSIA: return "en_MY";
case SUBLANG_ENGLISH_SINGAPORE: return "en_SG";
}
return "en";
case LANG_ESTONIAN: return "et_EE";
case LANG_FAEROESE: return "fo_FO";
case LANG_FARSI: return "fa_IR";
case LANG_FINNISH: return "fi_FI";
case LANG_FRENCH:
switch (sub)
{
case SUBLANG_FRENCH: return "fr_FR";
case SUBLANG_FRENCH_BELGIAN: /* WALLOON */ return "fr_BE";
case SUBLANG_FRENCH_CANADIAN: return "fr_CA";
case SUBLANG_FRENCH_SWISS: return "fr_CH";
case SUBLANG_FRENCH_LUXEMBOURG: return "fr_LU";
case SUBLANG_FRENCH_MONACO: return "fr_MC";
case SUBLANG_FRENCH_WESTINDIES: return "fr"; /* Caribbean? */
case SUBLANG_FRENCH_REUNION: return "fr_RE";
case SUBLANG_FRENCH_CONGO: return "fr_CG";
case SUBLANG_FRENCH_SENEGAL: return "fr_SN";
case SUBLANG_FRENCH_CAMEROON: return "fr_CM";
case SUBLANG_FRENCH_COTEDIVOIRE: return "fr_CI";
case SUBLANG_FRENCH_MALI: return "fr_ML";
case SUBLANG_FRENCH_MOROCCO: return "fr_MA";
case SUBLANG_FRENCH_HAITI: return "fr_HT";
}
return "fr";
case LANG_FRISIAN: return "fy_NL";
case LANG_FULFULDE: return "ful_NG";
case LANG_GAELIC:
switch (sub)
{
case 0x01: /* SCOTTISH */ return "gd_GB";
case 0x02: /* IRISH */ return "ga_IE";
}
return "C";
case LANG_GALICIAN: return "gl_ES";
case LANG_GEORGIAN: return "ka_GE";
case LANG_GERMAN:
switch (sub)
{
case SUBLANG_GERMAN: return "de_DE";
case SUBLANG_GERMAN_SWISS: return "de_CH";
case SUBLANG_GERMAN_AUSTRIAN: return "de_AT";
case SUBLANG_GERMAN_LUXEMBOURG: return "de_LU";
case SUBLANG_GERMAN_LIECHTENSTEIN: return "de_LI";
}
return "de";
case LANG_GREEK: return "el_GR";
case LANG_GUARANI: return "gn_PY";
case LANG_GUJARATI: return "gu_IN";
case LANG_HAUSA: return "ha_NG";
case LANG_HAWAIIAN:
/* FIXME: Do they mean Hawaiian ("haw_US", 1000 speakers)
or Hawaii Creole English ("cpe_US", 600000 speakers)? */
return "cpe_US";
case LANG_HEBREW: return "he_IL";
case LANG_HINDI: return "hi_IN";
case LANG_HUNGARIAN: return "hu_HU";
case LANG_IBIBIO: return "nic_NG";
case LANG_ICELANDIC: return "is_IS";
case LANG_IGBO: return "ibo_NG";
case LANG_INDONESIAN: return "id_ID";
case LANG_INUKTITUT: return "iu_CA";
case LANG_ITALIAN:
switch (sub)
{
case SUBLANG_ITALIAN: return "it_IT";
case SUBLANG_ITALIAN_SWISS: return "it_CH";
}
return "it";
case LANG_JAPANESE: return "ja_JP";
case LANG_KANNADA: return "kn_IN";
case LANG_KANURI: return "kau_NG";
case LANG_KASHMIRI:
switch (sub)
{
case SUBLANG_DEFAULT: return "ks_PK";
case SUBLANG_KASHMIRI_INDIA: return "ks_IN";
}
return "ks";
case LANG_KAZAK: return "kk_KZ";
case LANG_KONKANI:
/* FIXME: Adjust this when such locales appear on Unix. */
return "kok_IN";
case LANG_KOREAN: return "ko_KR";
case LANG_KYRGYZ: return "ky_KG";
case LANG_LAO: return "lo_LA";
case LANG_LATIN: return "la_VA";
case LANG_LATVIAN: return "lv_LV";
case LANG_LITHUANIAN: return "lt_LT";
case LANG_MACEDONIAN: return "mk_MK";
case LANG_MALAY:
switch (sub)
{
case SUBLANG_MALAY_MALAYSIA: return "ms_MY";
case SUBLANG_MALAY_BRUNEI_DARUSSALAM: return "ms_BN";
}
return "ms";
case LANG_MALAYALAM: return "ml_IN";
case LANG_MALTESE: return "mt_MT";
case LANG_MANIPURI:
/* FIXME: Adjust this when such locales appear on Unix. */
return "mni_IN";
case LANG_MARATHI: return "mr_IN";
case LANG_MONGOLIAN:
return "mn"; /* Ambiguous: could be "mn_CN" or "mn_MN". */
case LANG_NEPALI:
switch (sub)
{
case SUBLANG_DEFAULT: return "ne_NP";
case SUBLANG_NEPALI_INDIA: return "ne_IN";
}
return "ne";
case LANG_NORWEGIAN:
switch (sub)
{
case SUBLANG_NORWEGIAN_BOKMAL: return "no_NO";
case SUBLANG_NORWEGIAN_NYNORSK: return "nn_NO";
}
return "no";
case LANG_ORIYA: return "or_IN";
case LANG_OROMO: return "om_ET";
case LANG_PAPIAMENTU: return "pap_AN";
case LANG_PASHTO:
return "ps"; /* Ambiguous: could be "ps_PK" or "ps_AF". */
case LANG_POLISH: return "pl_PL";
case LANG_PORTUGUESE:
switch (sub)
{
case SUBLANG_PORTUGUESE: return "pt_PT";
/* Hmm. SUBLANG_PORTUGUESE_BRAZILIAN == SUBLANG_DEFAULT.
Same phenomenon as SUBLANG_ENGLISH_US == SUBLANG_DEFAULT. */
case SUBLANG_PORTUGUESE_BRAZILIAN: return "pt_BR";
}
return "pt";
case LANG_PUNJABI:
switch (sub)
{
case SUBLANG_PUNJABI_INDIA: return "pa_IN"; /* Gurmukhi script */
}
return "pa";
case LANG_RHAETO_ROMANCE: return "rm_CH";
case LANG_ROMANIAN:
switch (sub)
{
case SUBLANG_ROMANIAN_ROMANIA: return "ro_RO";
}
return "ro";
case LANG_RUSSIAN:
return "ru"; /* Ambiguous: could be "ru_RU" or "ru_UA" or "ru_MD". */
case LANG_SAAMI: /* actually Northern Sami */ return "se_NO";
case LANG_SANSKRIT: return "sa_IN";
case LANG_SINDHI:
switch (sub)
{
case SUBLANG_SINDHI_INDIA: return "sd_IN";
case SUBLANG_SINDHI_PAKISTAN: return "sd_PK";
}
return "sd";
case LANG_SINHALESE: return "si_LK";
case LANG_SLOVAK: return "sk_SK";
case LANG_SLOVENIAN: return "sl_SI";
case LANG_SOMALI: return "so_SO";
case LANG_SORBIAN:
/* FIXME: Adjust this when such locales appear on Unix. */
return "wen_DE";
case LANG_SPANISH:
switch (sub)
{
case SUBLANG_SPANISH: return "es_ES";
case SUBLANG_SPANISH_MEXICAN: return "es_MX";
case SUBLANG_SPANISH_MODERN:
return "es_ES@modern"; /* not seen on Unix */
case SUBLANG_SPANISH_GUATEMALA: return "es_GT";
case SUBLANG_SPANISH_COSTA_RICA: return "es_CR";
case SUBLANG_SPANISH_PANAMA: return "es_PA";
case SUBLANG_SPANISH_DOMINICAN_REPUBLIC: return "es_DO";
case SUBLANG_SPANISH_VENEZUELA: return "es_VE";
case SUBLANG_SPANISH_COLOMBIA: return "es_CO";
case SUBLANG_SPANISH_PERU: return "es_PE";
case SUBLANG_SPANISH_ARGENTINA: return "es_AR";
case SUBLANG_SPANISH_ECUADOR: return "es_EC";
case SUBLANG_SPANISH_CHILE: return "es_CL";
case SUBLANG_SPANISH_URUGUAY: return "es_UY";
case SUBLANG_SPANISH_PARAGUAY: return "es_PY";
case SUBLANG_SPANISH_BOLIVIA: return "es_BO";
case SUBLANG_SPANISH_EL_SALVADOR: return "es_SV";
case SUBLANG_SPANISH_HONDURAS: return "es_HN";
case SUBLANG_SPANISH_NICARAGUA: return "es_NI";
case SUBLANG_SPANISH_PUERTO_RICO: return "es_PR";
}
return "es";
case LANG_SUTU: return "bnt_TZ"; /* or "st_LS" or "nso_ZA"? */
case LANG_SWAHILI: return "sw_KE";
case LANG_SWEDISH:
switch (sub)
{
case SUBLANG_DEFAULT: return "sv_SE";
case SUBLANG_SWEDISH_FINLAND: return "sv_FI";
}
return "sv";
case LANG_SYRIAC: return "syr_TR"; /* An extinct language. */
case LANG_TAGALOG: return "tl_PH";
case LANG_TAJIK: return "tg_TJ";
case LANG_TAMAZIGHT:
switch (sub)
{
/* FIXME: Adjust this when Tamazight locales appear on Unix. */
case SUBLANG_TAMAZIGHT_ARABIC: return "ber_MA@arabic";
case SUBLANG_TAMAZIGHT_LATIN: return "ber_MA@latin";
}
return "ber_MA";
case LANG_TAMIL:
return "ta"; /* Ambiguous: could be "ta_IN" or "ta_LK" or "ta_SG". */
case LANG_TATAR: return "tt_RU";
case LANG_TELUGU: return "te_IN";
case LANG_THAI: return "th_TH";
case LANG_TIBETAN: return "bo_CN";
case LANG_TIGRINYA:
switch (sub)
{
case SUBLANG_TIGRINYA_ETHIOPIA: return "ti_ET";
case SUBLANG_TIGRINYA_ERITREA: return "ti_ER";
}
return "ti";
case LANG_TSONGA: return "ts_ZA";
case LANG_TSWANA: return "tn_BW";
case LANG_TURKISH: return "tr_TR";
case LANG_TURKMEN: return "tk_TM";
case LANG_UKRAINIAN: return "uk_UA";
case LANG_URDU:
switch (sub)
{
case SUBLANG_URDU_PAKISTAN: return "ur_PK";
case SUBLANG_URDU_INDIA: return "ur_IN";
}
return "ur";
case LANG_UZBEK:
switch (sub)
{
case SUBLANG_UZBEK_LATIN: return "uz_UZ";
case SUBLANG_UZBEK_CYRILLIC: return "uz_UZ@cyrillic";
}
return "uz";
case LANG_VENDA:
/* FIXME: It's not clear whether Venda has the ISO 639-2 two-letter code
"ve" or not.
http://www.loc.gov/standards/iso639-2/englangn.html has it, but
http://lcweb.loc.gov/standards/iso639-2/codechanges.html doesn't, */
return "ven_ZA"; /* or "ve_ZA"? */
case LANG_VIETNAMESE: return "vi_VN";
case LANG_WELSH: return "cy_GB";
case LANG_XHOSA: return "xh_ZA";
case LANG_YI: return "sit_CN";
case LANG_YIDDISH: return "yi_IL";
case LANG_YORUBA: return "yo_NG";
case LANG_ZULU: return "zu_ZA";
default: return "C";
}
}
/* localname.c from gettext END. */
/* Support functions. */
static GPG_ERR_INLINE uint32_t
do_swap_u32 (uint32_t i)
{
return (i << 24) | ((i & 0xff00) << 8) | ((i >> 8) & 0xff00) | (i >> 24);
}
#define SWAPIT(flag, data) ((flag) ? do_swap_u32(data) : (data))
/* We assume to have `unsigned long int' value with at least 32 bits. */
#define HASHWORDBITS 32
/* The so called `hashpjw' function by P.J. Weinberger
[see Aho/Sethi/Ullman, COMPILERS: Principles, Techniques and Tools,
1986, 1987 Bell Telephone Laboratories, Inc.] */
static GPG_ERR_INLINE unsigned long
hash_string (const char *str_param)
{
unsigned long int hval, g;
const char *str = str_param;
hval = 0;
while (*str != '\0')
{
hval <<= 4;
hval += (unsigned long int) *str++;
g = hval & ((unsigned long int) 0xf << (HASHWORDBITS - 4));
if (g != 0)
{
hval ^= g >> (HASHWORDBITS - 8);
hval ^= g;
}
}
return hval;
}
/* Generic message catalog and gettext stuff. */
/* The magic number of the GNU message catalog format. */
#define MAGIC 0x950412de
#define MAGIC_SWAPPED 0xde120495
/* Revision number of the currently used .mo (binary) file format. */
#define MO_REVISION_NUMBER 0
/* Header for binary .mo file format. */
struct mo_file_header
{
/* The magic number. */
uint32_t magic;
/* The revision number of the file format. */
uint32_t revision;
/* The number of strings pairs. */
uint32_t nstrings;
/* Offset of table with start offsets of original strings. */
uint32_t orig_tab_offset;
/* Offset of table with start offsets of translation strings. */
uint32_t trans_tab_offset;
/* Size of hashing table. */
uint32_t hash_tab_size;
/* Offset of first hashing entry. */
uint32_t hash_tab_offset;
};
struct string_desc
{
/* Length of addressed string. */
uint32_t length;
/* Offset of string in file. */
uint32_t offset;
};
struct overflow_space_s
{
struct overflow_space_s *next;
uint32_t idx;
uint32_t length;
char d[1];
};
struct loaded_domain
{
char *data;
char *data_native; /* Data mapped to the native version of the
string. (Allocated along with DATA). */
int must_swap;
uint16_t nstrings; /* Number of strings. */
uint16_t *mapped; /* Array of mapping indicators:
0 := Not mapped (original utf8).
1 := Mapped to native encoding in overflow space.
>=2 := Mapped to native encoding. The value
gives the length of the mapped string.
Because the terminating nul is included
in the length and an empty string is
not allowed, values are always > 1. */
struct overflow_space_s *overflow_space;
struct string_desc *orig_tab;
struct string_desc *trans_tab;
uint32_t hash_size;
uint32_t *hash_tab;
};
/* The list of domains we we are aware of. This list is protected by
the criticla section DOMAINLIST_ACCESS_CS. */
static struct domainlist_s *domainlist;
/* A critical section to guard access to the domainlist. */
static CRITICAL_SECTION domainlist_access_cs;
/* The name of the current domain. This is a malloced string. This
is a gobal variable which is not thread-safe. */
static char *current_domainname;
/* Constructor for this module. This can only be used if we are a
DLL. If used as a static lib we can't control the process set; for
example it might be used with a main module which is not build with
mingw and thus does not know how to call the constructors. */
#ifdef DLL_EXPORT
static void module_init (void) _GPG_ERR_CONSTRUCTOR;
#endif
static void
module_init (void)
{
static int init_done;
if (!init_done)
{
InitializeCriticalSection (&domainlist_access_cs);
init_done = 1;
}
}
#if !defined(DLL_EXPORT) || !defined(_GPG_ERR_HAVE_CONSTRUCTOR)
void
_gpg_w32__init_gettext_module (void)
{
module_init ();
}
#endif
/* Free the domain data. */
static void
free_domain (struct loaded_domain *domain)
{
struct overflow_space_s *os, *os2;
jnlib_free (domain->data);
jnlib_free (domain->mapped);
for (os = domain->overflow_space; os; os = os2)
{
os2 = os->next;
jnlib_free (os);
}
jnlib_free (domain);
}
static struct loaded_domain *
load_domain (const char *filename)
{
HANDLE fh;
DWORD size;
struct mo_file_header *data = NULL;
struct loaded_domain *domain = NULL;
size_t to_read;
char *read_ptr;
fh = CreateFileA (filename, GENERIC_READ, FILE_SHARE_WRITE, NULL,
OPEN_EXISTING, 0, NULL);
if (fh == INVALID_HANDLE_VALUE)
return NULL;
size = GetFileSize (fh, NULL);
if (size == INVALID_FILE_SIZE)
{
CloseHandle (fh);
return NULL;
}
data = (2*size <= size)? NULL : jnlib_malloc (2*size);
if (!data)
{
CloseHandle (fh);
return NULL;
}
to_read = size;
read_ptr = (char *) data;
do
{
BOOL res;
DWORD nb;
res = ReadFile (fh, read_ptr, to_read, &nb, NULL);
if (! res || nb < to_read)
{
CloseHandle (fh);
jnlib_free (data);
return NULL;
}
read_ptr += nb;
to_read -= nb;
}
while (to_read > 0);
CloseHandle (fh);
/* Using the magic number we can test whether it really is a message
catalog file. */
if (data->magic != MAGIC && data->magic != MAGIC_SWAPPED)
{
/* The magic number is wrong: not a message catalog file. */
jnlib_free (data);
return NULL;
}
domain = jnlib_calloc (1, sizeof *domain);
if (!domain)
{
jnlib_free (data);
return NULL;
}
domain->data = (char *) data;
domain->data_native = (char *) data + size;
domain->must_swap = data->magic != MAGIC;
/* Fill in the information about the available tables. */
switch (SWAPIT (domain->must_swap, data->revision))
{
case MO_REVISION_NUMBER:
{
uint32_t nstrings;
/* Because we use 16 bit values for the mapping array, we
can't support more that 65534 strings (65535 would be okay,
but it is often used as a special value). A PO file with
that many translations is very unlikely given that GnuPG
with its very large number of strings has only about 1600
strings + variants. */
nstrings = SWAPIT (domain->must_swap, data->nstrings);
if (nstrings > 65534)
goto bailout;
domain->nstrings = nstrings;
domain->orig_tab = (struct string_desc *)
((char *) data + SWAPIT (domain->must_swap, data->orig_tab_offset));
domain->trans_tab = (struct string_desc *)
((char *) data + SWAPIT (domain->must_swap, data->trans_tab_offset));
domain->hash_size = SWAPIT (domain->must_swap, data->hash_tab_size);
domain->hash_tab = (uint32_t *)
((char *) data + SWAPIT (domain->must_swap, data->hash_tab_offset));
}
break;
default: /* This is an invalid revision. */
goto bailout;
}
/* Allocate an array to keep track of code page mappings. */
domain->mapped = jnlib_calloc (domain->nstrings, sizeof *domain->mapped);
if (domain->mapped)
return domain; /* Okay. */
bailout:
jnlib_free (data);
jnlib_free (domain);
return NULL;
}
/* Return a malloced wide char string from an UTF-8 encoded input
string STRING. Caller must free this value. On failure returns
NULL. The result of calling this function with STRING set to NULL
is not defined. */
static wchar_t *
utf8_to_wchar (const char *string, size_t length, size_t *retlen)
{
int n;
wchar_t *result;
size_t nbytes;
n = MultiByteToWideChar (CP_UTF8, 0, string, length, NULL, 0);
if (n < 0 || (n+1) <= 0)
return NULL;
nbytes = (size_t)(n+1) * sizeof(*result);
if (nbytes / sizeof(*result) != (n+1))
{
gpg_err_set_errno (ENOMEM);
return NULL;
}
result = jnlib_malloc (nbytes);
if (!result)
return NULL;
n = MultiByteToWideChar (CP_UTF8, 0, string, length, result, n);
if (n < 0)
{
jnlib_free (result);
return NULL;
}
*retlen = n;
return result;
}
/* Return a malloced string encoded in the native console codepage
from the wide char input string STRING.
Caller must free this value. On failure returns NULL.
The result of calling this function with STRING set to NULL
is not defined. */
static char *
wchar_to_native (const wchar_t *string, size_t length, size_t *retlen)
{
int n;
char *result;
unsigned int cpno = GetConsoleOutputCP ();
/* GetConsoleOutputCP returns the 8-Bit codepage that should be used
for console output. If the codepage is not returned we fall back
to the codepage GUI programs should use (CP_ACP). */
if (!cpno)
cpno = GetACP ();
n = WideCharToMultiByte (cpno, 0, string, length, NULL, 0, NULL, NULL);
if (n < 0 || (n+1) <= 0)
return NULL;
result = jnlib_malloc (n+1);
if (!result)
return NULL;
n = WideCharToMultiByte (cpno, 0, string, length, result, n, NULL, NULL);
if (n < 0)
{
jnlib_free (result);
return NULL;
}
*retlen = n;
return result;
}
/* Convert UTF8 to the native codepage. Caller must free the return value. */
static char *
utf8_to_native (const char *string, size_t length, size_t *retlen)
{
wchar_t *wstring;
char *result;
size_t newlen;
wstring = utf8_to_wchar (string, length, &newlen);
if (wstring)
{
result = wchar_to_native (wstring, newlen, &newlen);
jnlib_free (wstring);
}
else
result = NULL;
*retlen = result? newlen : 0;
return result;
}
/* Specify that the DOMAINNAME message catalog will be found
in DIRNAME rather than in the system locale data base. */
const char *
_gpg_w32_bindtextdomain (const char *domainname, const char *dirname)
{
const char *catval_full;
char *catval;
char *fname;
const char *retvalue;
if (!dirname)
{
struct domainlist_s *dl;
retvalue = NULL;
EnterCriticalSection (&domainlist_access_cs);
{
for (dl = domainlist; dl; dl = dl->next)
if (!strcmp (dl->name, domainname))
{
retvalue = dl->dname;
break;
}
}
LeaveCriticalSection (&domainlist_access_cs);
return retvalue;
}
/* DIRNAME is "$INSTALLDIR\share\locale". */
/* First find out the category value. */
catval = NULL;
catval_full = my_nl_locale_name ("LC_MESSAGES");
/* Normally we would have to loop over all returned locales and
search for the right file. See gettext intl/dcigettext.c for all
the gory details. Here, we only support the basic category, and
ignore everything else. */
if (catval_full)
{
char *p;
catval = jnlib_malloc (strlen (catval_full) + 1);
if (catval)
{
strcpy (catval, catval_full);
p = strchr (catval, '_');
if (p)
*p = '\0';
}
}
if (!catval)
return NULL;
/* Now build the filename string. The complete filename is this:
DIRNAME + \ + CATVAL + \LC_MESSAGES\ + DOMAINNAME + .mo */
{
int len = (strlen (dirname) + 1 + strlen (catval) + 13
+ strlen (domainname) + 3 + 1);
char *p;
fname = jnlib_malloc (len);
if (!fname)
{
jnlib_free (catval);
return NULL;
}
p = fname;
strcpy (p, dirname);
p += strlen (dirname);
*(p++) = '\\';
strcpy (p, catval);
p += strlen (catval);
strcpy (p, "\\LC_MESSAGES\\");
p += 13;
strcpy (p, domainname);
p += strlen (domainname);
strcpy (p, ".mo");
}
jnlib_free (catval);
/* Store the domain information in the domainlist. */
{
struct domainlist_s *item, *dl;
char *rel_ptr1 = NULL;
char *rel_ptr2 = NULL;
item = jnlib_calloc (1, sizeof *dl + strlen (domainname));
if (!item)
{
jnlib_free (fname);
return NULL;
}
strcpy (item->name, domainname);
item->dname = jnlib_malloc (strlen (dirname) +1);
if(!item->dname)
{
jnlib_free (item);
jnlib_free (fname);
return NULL;
}
strcpy (item->dname, dirname);
retvalue = item->dname;
EnterCriticalSection (&domainlist_access_cs);
{
for (dl = domainlist; dl; dl = dl->next)
if (!strcmp (dl->name, domainname))
break;
if (!dl) /* First time called for this domainname. */
{
item->fname = fname;
fname = NULL;
item->next = domainlist;
domainlist = item;
item = NULL;
}
else /* Update only. */
{
rel_ptr1 = dl->fname;
dl->fname = fname;
fname = NULL;
rel_ptr2 = dl->dname;
dl->dname = item->dname;
item->dname = NULL;
}
}
LeaveCriticalSection (&domainlist_access_cs);
jnlib_free (item);
jnlib_free (rel_ptr1);
jnlib_free (rel_ptr2);
}
return retvalue;
}
static const char *
get_plural (const char *data, size_t datalen, unsigned long nplural)
{
const char *p;
int idx;
/* We only support the Germanic rule. */
idx = (nplural == 1? 0 : 1);
for (; idx; idx--)
{
p = strchr (data, 0) + 1;
if (p >= data+datalen)
return "ERROR in GETTEXT (bad plural entry)";
datalen -= (p-data);
data = p;
}
return data;
}
static const char*
get_string (struct loaded_domain *domain, uint32_t idx,
int use_plural, unsigned long nplural)
{
struct tls_space_s *tls = get_tls ();
struct overflow_space_s *os;
const char *trans; /* Pointer to the translated entry. */
size_t translen; /* Length of that entry. */
if (idx > 65534)
return "ERROR in GETTEXT (too many strings)";
if (tls->gt_use_utf8)
{
trans = (domain->data
+ SWAPIT(domain->must_swap, domain->trans_tab[idx].offset));
translen = SWAPIT(domain->must_swap, domain->trans_tab[idx].length);
}
else if (!domain->mapped[idx])
{
/* Not yet mapped. Map from utf-8 to native encoding now. */
const char *p_utf8;
size_t plen_utf8, buflen;
char *buf;
p_utf8 = (domain->data
+ SWAPIT(domain->must_swap, domain->trans_tab[idx].offset));
plen_utf8 = SWAPIT(domain->must_swap, domain->trans_tab[idx].length);
/* We need to include the nul, so that the utf8->wchar->native
conversion chain works correctly and the nul is stored after
the conversion. */
if (p_utf8[plen_utf8])
{
trans = "ERROR in MO file"; /* Terminating zero is missing. */
translen = 0;
goto leave;
}
plen_utf8++;
buf = utf8_to_native (p_utf8, plen_utf8, &buflen);
if (!buf)
{
trans = "ERROR in GETTEXT MALLOC";
translen = 0;
}
else if (buflen <= plen_utf8 && buflen > 1)
{
/* Copy into the DATA_NATIVE area. */
char *p_tmp;
p_tmp = (domain->data_native
+ SWAPIT(domain->must_swap, domain->trans_tab[idx].offset));
memcpy (p_tmp, buf, buflen);
domain->mapped[idx] = buflen;
trans = p_tmp;
translen = buflen;
}
else
{
/* There is not enough space for the translation (or for
whatever reason an empty string is used): Store it in the
overflow_space and mark that in the mapped array.
Because UTF-8 strings are in general longer than the
Windows native encoding, we expect that this won't happen
too often and thus we use a linked list to manage this
space. */
os = jnlib_malloc (sizeof *os + buflen);
if (os)
{
os->idx = idx;
memcpy (os->d, buf, buflen);
os->length = buflen;
os->next = domain->overflow_space;
domain->overflow_space = os;
domain->mapped[idx] = 1;
trans = os->d;
translen = os->length;
}
else
{
trans = "ERROR in GETTEXT MALLOC";
translen = 0;
}
}
if (translen)
translen--; /* TRANSLEN shall be the size without the nul. */
jnlib_free (buf);
}
else if (domain->mapped[idx] == 1)
{
/* The translated string is in the overflow_space. */
for (os=domain->overflow_space; os; os = os->next)
if (os->idx == idx)
break;
if (os)
{
trans = os->d;
translen = os->length;
}
else
{
trans = "ERROR in GETTEXT (overflow space)\n";
translen = 0;
}
}
else
{
trans = (domain->data_native
+ SWAPIT(domain->must_swap, domain->trans_tab[idx].offset));
translen = domain->mapped[idx];
}
leave:
if (use_plural && translen)
return get_plural (trans, translen, nplural);
else
return trans;
}
static const char *
do_gettext (const char *domainname,
const char *msgid, const char *msgid2, unsigned long nplural)
{
struct domainlist_s *dl;
struct loaded_domain *domain;
int load_failed;
uint32_t top, bottom, nstr;
char *filename;
if (!domainname)
domainname = current_domainname? current_domainname : "";
/* FIXME: The whole locking stuff is a bit questionable because
gettext does not claim to be thread-safe. We need to investigate
this further. */
load_failed = 0;
domain = NULL;
filename = NULL;
EnterCriticalSection (&domainlist_access_cs);
{
for (dl = domainlist; dl; dl = dl->next)
if (!strcmp (dl->name, domainname))
{
load_failed = dl->load_failed;
domain = dl->domain;
break;
}
if (dl && !domain && !load_failed && dl->fname)
{
filename = jnlib_malloc (strlen (dl->fname) + 1);
if (filename)
strcpy (filename, dl->fname);
}
}
LeaveCriticalSection (&domainlist_access_cs);
if (!dl)
goto not_found; /* DOMAINNAME not bound. */
if (filename)
{
/* No attempt so far to load the MO file. Try now. */
int updated = 0;
domain = load_domain (filename);
jnlib_free (filename);
filename = NULL;
EnterCriticalSection (&domainlist_access_cs);
{
for (dl = domainlist; dl; dl = dl->next)
if (!strcmp (dl->name, domainname))
{
if (domain)
dl->domain = domain;
else
dl->load_failed = 1;
updated = 1;
break;
}
}
LeaveCriticalSection (&domainlist_access_cs);
if (!updated)
{
/* Ooops - lost the domain. */
free_domain (domain);
domain = NULL;
}
}
if (!domain)
goto not_found; /* No MO file. */
/* First try to use the hash table. */
if (domain->hash_size > 2 && domain->hash_tab)
{
/* Use the hashing table. */
uint32_t len = strlen (msgid);
uint32_t hash_val = hash_string (msgid);
uint32_t idx = hash_val % domain->hash_size;
uint32_t incr = 1 + (hash_val % (domain->hash_size - 2));
while ( (nstr = SWAPIT (domain->must_swap, domain->hash_tab[idx])) )
{
nstr--;
if (nstr < domain->nstrings
&& SWAPIT(domain->must_swap,
domain->orig_tab[nstr].length) >= len
&& !strcmp (msgid, (domain->data
+ SWAPIT(domain->must_swap,
domain->orig_tab[nstr].offset))))
{
return get_string (domain, nstr, !!msgid2, nplural);
}
if (idx >= domain->hash_size - incr)
idx -= domain->hash_size - incr;
else
idx += incr;
}
}
/* Now we try the default method: binary search in the sorted array
of messages. */
bottom = 0;
top = domain->nstrings;
while (bottom < top)
{
int cmp_val;
nstr = (bottom + top) / 2;
cmp_val = strcmp (msgid, (domain->data
+ SWAPIT(domain->must_swap,
domain->orig_tab[nstr].offset)));
if (cmp_val < 0)
top = nstr;
else if (cmp_val > 0)
bottom = nstr + 1;
else
{
return get_string (domain, nstr, !!msgid2, nplural);
}
}
not_found:
/* We use the standard Germanic rule if plural has been requested. */
return msgid2? (nplural == 1? msgid : msgid2) : msgid;
}
const char *
_gpg_w32_textdomain (const char *domainname)
{
char *string;
if (!domainname)
{
if (!current_domainname)
gpg_err_set_errno (0);
}
else
{
string = malloc (strlen (domainname) + 1);
if (!string)
return NULL;
strcpy (string, domainname);
current_domainname = string;
}
return current_domainname;
}
/* A direct implementation of gettext instead of a macro calling
dngettext. This is so that the caller does not need to push dummy
values on the stack. The used domain is the first one registered
with bindtextdomain. */
const char *
_gpg_w32_gettext (const char *msgid)
{
return do_gettext (NULL, msgid, NULL, 0);
}
/* A direct implementation of dgettext instead of a macro calling
dngettext. This is so that the caller does not need to push dummy
values on the stack. */
const char *
_gpg_w32_dgettext (const char *domainname, const char *msgid)
{
return do_gettext (domainname, msgid, NULL, 0);
}
/* Our implementation of dngettext. This is the most genereic
function we have; a macro implements ngettext. */
const char *
_gpg_w32_dngettext (const char *domainname, const char *msgid1,
const char *msgid2, unsigned long int n)
{
/* We use the simple Germanic plural rule. */
return do_gettext (domainname, msgid1, msgid2, n);
}
/* Return the locale name as used by gettext. The return value will
never be NULL. */
const char *
_gpg_w32_gettext_localename (void)
{
const char *s;
s = my_nl_locale_name ("LC_MESSAGES");
return s? s:"";
}
/* With a VALUE of 1 switch the gettext functions into utf8 mode.
That is the strings are returned without translation to the native
charset. A VALUE of 0 switches back to translated strings. A VALUE
of -1 returns the current value. */
int
_gpg_w32_gettext_use_utf8 (int value)
{
struct tls_space_s *tls = get_tls ();
int last = tls->gt_use_utf8;
if (value != -1)
tls->gt_use_utf8 = value;
return last;
}
#ifdef TEST
int
main (int argc, char **argv)
{
const char atext1[] =
"Warning: You have entered an insecure passphrase.%%0A"
"A passphrase should be at least %u character long.";
const char atext2[] =
"Warning: You have entered an insecure passphrase.%%0A"
"A passphrase should be at least %u characters long.";
if (argc)
{
argc--;
argv++;
}
_gpg_err_w32_bindtextdomain ("gnupg2", "c:/programme/gnu/gnupg/share/locale");
printf ("locale is `%s'\n", _gpg_err_w32_gettext_localename ());
fputs ("text with N=1:\n", stdout);
fputs (_gpg_err_w32_ngettext (atext1, atext2, 1), stdout);
fputs ("\n\ntext with N=2:\n", stdout);
fputs (_gpg_err_w32_ngettext (atext1, atext2, 2), stdout);
fputs ("\nready\n", stdout);
return 0;
}
/*
* Local Variables:
* compile-command: "i586-mingw32msvc-gcc -DTEST -Wall -g w32-gettext.c"
* End:
*/
#endif /*TEST*/
diff --git a/src/w32-lock-obj.h b/src/w32-lock-obj.h
index ef53546..8ed3084 100644
--- a/src/w32-lock-obj.h
+++ b/src/w32-lock-obj.h
@@ -1,38 +1,38 @@
/* w32-lock-obj.h - Declaration of the Windows lock object
Copyright (C) 2014 g10 Code GmbH
This file is part of libgpg-error.
libgpg-error is free software; you can redistribute it and/or
modify it under the terms of the GNU Lesser General Public License
as published by the Free Software Foundation; either version 2.1 of
the License, or (at your option) any later version.
libgpg-error is distributed in the hope that it will be useful, but
WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public
- License along with this program; if not, see .
+ License along with this program; if not, see .
*/
#ifndef W32_LOCK_OBJ_H
#define W32_LOCK_OBJ_H
#define LOCK_ABI_VERSION 1
/* The real definition of our lock object. The public definition is
named gpgrt_lock_t and hides this internal structure. */
#pragma pack(push, 8)
typedef struct
{
long vers;
volatile long initdone;
volatile long started;
CRITICAL_SECTION csec;
} _gpgrt_lock_t;
#pragma pack(pop)
#endif /*W32_LOCK_OBJ_H*/
diff --git a/src/w32-lock.c b/src/w32-lock.c
index 8c086f9..d1decc9 100644
--- a/src/w32-lock.c
+++ b/src/w32-lock.c
@@ -1,158 +1,158 @@
/* w32-lock.c - GPGRT lock functions for Windows
Copyright (C) 2014 g10 Code GmbH
This file is part of libgpg-error.
libgpg-error is free software; you can redistribute it and/or
modify it under the terms of the GNU Lesser General Public License
as published by the Free Software Foundation; either version 2.1 of
the License, or (at your option) any later version.
libgpg-error is distributed in the hope that it will be useful, but
WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public
- License along with this program; if not, see .
+ License along with this program; if not, see .
*/
#if HAVE_CONFIG_H
#include
#endif
#ifndef HAVE_W32_SYSTEM
# error This module may only be build for Windows.
#endif
#include
#include
#include
#include
#include
#define WIN32_LEAN_AND_MEAN
#include
#include "lock.h"
#include "w32-lock-obj.h"
static _gpgrt_lock_t *
get_lock_object (gpgrt_lock_t *lockhd)
{
_gpgrt_lock_t *lock = (_gpgrt_lock_t*)lockhd;
if (lock->vers != LOCK_ABI_VERSION)
abort ();
return lock;
}
gpg_err_code_t
_gpgrt_lock_init (gpgrt_lock_t *lockhd)
{
_gpgrt_lock_t *lock = (_gpgrt_lock_t*)lockhd;
/* If VERS is zero we assume that no static initialization has been
done, so we setup our ABI version right here. The caller might
have called us to test whether lock support is at all available. */
if (!lock->vers)
{
if (sizeof (gpgrt_lock_t) < sizeof (_gpgrt_lock_t))
abort ();
lock->vers = LOCK_ABI_VERSION;
}
else /* Run the usual check. */
{
lock = get_lock_object (lockhd);
if (sizeof (gpgrt_lock_t) < sizeof (_gpgrt_lock_t))
abort ();
}
InitializeCriticalSection (&lock->csec);
lock->initdone = 1;
return 0;
}
gpg_err_code_t
_gpgrt_lock_lock (gpgrt_lock_t *lockhd)
{
_gpgrt_lock_t *lock = get_lock_object (lockhd);
if (!lock->initdone)
{
if (!InterlockedIncrement (&lock->started))
{
/* The new value of started is 0. Because the initial value
if the variable was -1 we known that this thread is the
first who needs this lock. Thus we initialize now. All
other threads won't get 0 back from InterlockedIncrement
and thus fall into the wait loop below. We ignore that
STARTED may in theory overflow if this thread starves for
too long. */
gpgrt_lock_init (lockhd);
}
else
{
while (!lock->initdone)
Sleep (0);
}
}
EnterCriticalSection (&lock->csec);
return 0;
}
gpg_err_code_t
_gpgrt_lock_trylock (gpgrt_lock_t *lockhd)
{
_gpgrt_lock_t *lock = get_lock_object (lockhd);
if (!lock->initdone)
{
if (!InterlockedIncrement (&lock->started))
{
gpgrt_lock_init (lockhd);
}
else
{
while (!lock->initdone)
Sleep (0);
}
}
if (!TryEnterCriticalSection (&lock->csec))
return GPG_ERR_EBUSY;
return 0;
}
gpg_err_code_t
_gpgrt_lock_unlock (gpgrt_lock_t *lockhd)
{
_gpgrt_lock_t *lock = get_lock_object (lockhd);
if (!lock->initdone)
return GPG_ERR_INV_LOCK_OBJ;
LeaveCriticalSection (&lock->csec);
return 0;
}
/* Note: Use this function only if no other thread holds or waits for
this lock. */
gpg_err_code_t
_gpgrt_lock_destroy (gpgrt_lock_t *lockhd)
{
_gpgrt_lock_t *lock = get_lock_object (lockhd);
if (!lock->initdone)
return GPG_ERR_INV_LOCK_OBJ;
DeleteCriticalSection (&lock->csec);
lock->initdone = 0;
lock->started = -1;
return 0;
}
diff --git a/src/w32-thread.c b/src/w32-thread.c
index 53d26b4..6860075 100644
--- a/src/w32-thread.c
+++ b/src/w32-thread.c
@@ -1,44 +1,44 @@
/* w32-thread.c - GPGRT thread functions for Windows
Copyright (C) 2014 g10 Code GmbH
This file is part of libgpg-error.
libgpg-error is free software; you can redistribute it and/or
modify it under the terms of the GNU Lesser General Public License
as published by the Free Software Foundation; either version 2.1 of
the License, or (at your option) any later version.
libgpg-error is distributed in the hope that it will be useful, but
WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public
- License along with this program; if not, see .
+ License along with this program; if not, see .
*/
#if HAVE_CONFIG_H
#include
#endif
#ifndef HAVE_W32_SYSTEM
# error This module may only be build for Windows.
#endif
#include
#include
#include
#include
#include
#define WIN32_LEAN_AND_MEAN
#include
#include "thread.h"
gpg_err_code_t
_gpgrt_yield (void)
{
Sleep (0);
return 0;
}
diff --git a/tests/t-common.h b/tests/t-common.h
index c6dcd12..e11bca9 100644
--- a/tests/t-common.h
+++ b/tests/t-common.h
@@ -1,102 +1,102 @@
/* t-common.h - Common code for the tests.
* Copyright (C) 2013 g10 Code GmbH
*
* This file is part of libgpg-error.
*
* libgpg-error is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public License
* as published by the Free Software Foundation; either version 2.1 of
* the License, or (at your option) any later version.
*
* libgpg-error is distributed in the hope that it will be useful, but
* WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
- * License along with this program; if not, see .
+ * License along with this program; if not, see .
*/
#include
#include "../src/gpg-error.h"
#ifndef PGM
# error Macro PGM not defined.
#endif
#ifndef DIM
# define DIM(array) (sizeof (array) / sizeof (*array))
#endif
static int verbose;
static int debug;
static int errorcount;
static void
die (const char *format, ...)
{
va_list arg_ptr ;
fflush (stdout);
#ifdef HAVE_FLOCKFILE
flockfile (stderr);
#endif
fprintf (stderr, "%s: ", PGM);
va_start (arg_ptr, format) ;
vfprintf (stderr, format, arg_ptr);
va_end (arg_ptr);
if (*format && format[strlen(format)-1] != '\n')
putc ('\n', stderr);
#ifdef HAVE_FLOCKFILE
funlockfile (stderr);
#endif
exit (1);
}
static void
fail (const char *format, ...)
{
va_list arg_ptr;
fflush (stdout);
#ifdef HAVE_FLOCKFILE
flockfile (stderr);
#endif
fprintf (stderr, "%s: ", PGM);
va_start (arg_ptr, format);
vfprintf (stderr, format, arg_ptr);
va_end (arg_ptr);
if (*format && format[strlen(format)-1] != '\n')
putc ('\n', stderr);
#ifdef HAVE_FLOCKFILE
funlockfile (stderr);
#endif
errorcount++;
if (errorcount >= 50)
die ("stopped after 50 errors.");
}
static void
show (const char *format, ...)
{
va_list arg_ptr;
if (!verbose)
return;
#ifdef HAVE_FLOCKFILE
flockfile (stderr);
#endif
fprintf (stderr, "%s: ", PGM);
va_start (arg_ptr, format);
vfprintf (stderr, format, arg_ptr);
if (*format && format[strlen(format)-1] != '\n')
putc ('\n', stderr);
va_end (arg_ptr);
#ifdef HAVE_FLOCKFILE
funlockfile (stderr);
#endif
}
diff --git a/tests/t-lock.c b/tests/t-lock.c
index 1ccf815..38c9cec 100644
--- a/tests/t-lock.c
+++ b/tests/t-lock.c
@@ -1,331 +1,331 @@
/* t-lock.c - Check the lock functions
* Copyright (C) 2013, 2015 g10 Code GmbH
*
* This file is part of libgpg-error.
*
* libgpg-error is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public License
* as published by the Free Software Foundation; either version 2.1 of
* the License, or (at your option) any later version.
*
* libgpg-error is distributed in the hope that it will be useful, but
* WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
- * License along with this program; if not, see .
+ * License along with this program; if not, see .
*/
#if HAVE_CONFIG_H
# include
#endif
#include
#include
#include
#include
#include
#include
#ifdef _WIN32
# include
# include
#else
# include
#endif
#define PGM "t-lock"
#include "t-common.h"
#ifdef _WIN32
# define THREAD_RET_TYPE DWORD WINAPI
# define THREAD_RET_VALUE 0
#else
# define THREAD_RET_TYPE void *
# define THREAD_RET_VALUE NULL
#endif
/* Our tests works by having a a couple of accountant threads which do
random transactions between accounts and a revision threads which
checks that the balance of all accounts is invariant. The idea for
this check is due to Bruno Haible. */
#define N_ACCOUNT 8
#define ACCOUNT_VALUE 42
static int account[N_ACCOUNT];
GPGRT_LOCK_DEFINE (accounts_lock);
/* Number of transactions done by each accountant. */
#define N_TRANSACTIONS 1000
/* Number of accountants to run. */
#define N_ACCOUNTANTS 5
/* Maximum transaction value. A quite low value is used so that we
would get an integer overflow. */
#define MAX_TRANSACTION_VALUE 50
/* Flag to tell the revision thread to finish. */
static volatile int stop_revision_thread;
/* Initialze all accounts. */
static void
init_accounts (void)
{
int i;
for (i=0; i < N_ACCOUNT; i++)
account[i] = ACCOUNT_VALUE;
}
/* Check that the sum of all accounts matches the initial sum. */
static void
check_accounts (void)
{
int i, sum;
sum = 0;
for (i = 0; i < N_ACCOUNT; i++)
sum += account[i];
if (sum != N_ACCOUNT * ACCOUNT_VALUE)
die ("accounts out of balance");
}
static void
print_accounts (void)
{
int i;
for (i=0; i < N_ACCOUNT; i++)
printf ("account %d: %6d\n", i, account[i]);
}
#if defined(_WIN32) || defined(USE_POSIX_THREADS)
/* Get a a random integer value in the range 0 to HIGH. */
static unsigned int
get_rand (int high)
{
return (unsigned int)(1+(int)((double)(high+1)*rand ()/(RAND_MAX+1.0))) - 1;
}
/* Pick a random account. Note that this function is not
thread-safe. */
static int
pick_account (void)
{
return get_rand (N_ACCOUNT - 1);
}
/* Pick a random value for a transaction. This is not thread-safe. */
static int
pick_value (void)
{
return get_rand (MAX_TRANSACTION_VALUE);
}
/* This is the revision department. */
static THREAD_RET_TYPE
revision_thread (void *arg)
{
gpg_err_code_t rc;
int i = 0;
(void)arg;
while (!stop_revision_thread)
{
rc = gpgrt_lock_lock (&accounts_lock);
if (rc)
fail ("gpgrt_lock_lock failed at %d: %s", __LINE__, gpg_strerror (rc));
check_accounts ();
rc = gpgrt_lock_unlock (&accounts_lock);
if (rc)
fail ("gpgrt_lock_unlock failed at %d: %s", __LINE__,gpg_strerror (rc));
if (!(++i%7))
gpgrt_yield ();
}
return THREAD_RET_VALUE;
}
/* This is one of our accountants. */
static THREAD_RET_TYPE
accountant_thread (void *arg)
{
gpg_err_code_t rc;
int i;
int acc1, acc2;
int value;
(void)arg;
#ifdef _WIN32
srand (time(NULL)*getpid()); /* Windows needs it per thread. */
#endif
for (i = 0; i < N_TRANSACTIONS; i++)
{
rc = gpgrt_lock_lock (&accounts_lock);
if (rc)
fail ("gpgrt_lock_lock failed at %d: %s", __LINE__, gpg_strerror (rc));
acc1 = pick_account ();
acc2 = pick_account ();
value = pick_value ();
account[acc1] += value;
account[acc2] -= value;
rc = gpgrt_lock_unlock (&accounts_lock);
if (rc)
fail ("gpgrt_lock_unlock failed at %d: %s", __LINE__,gpg_strerror (rc));
if (i && !(i%8))
gpgrt_yield ();
}
return THREAD_RET_VALUE;
}
#endif /*_WIN32||USE_POSIX_THREADS*/
static void
run_test (void)
{
#ifdef _WIN32
HANDLE rthread;
HANDLE athreads[N_ACCOUNTANTS];
int i;
int rc;
stop_revision_thread = 0;
rthread = CreateThread (NULL, 0, revision_thread, NULL, 0, NULL);
if (!rthread)
die ("error creating revision thread: rc=%d", (int)GetLastError ());
for (i=0; i < N_ACCOUNTANTS; i++)
{
athreads[i] = CreateThread (NULL, 0, accountant_thread, NULL, 0, NULL);
if (!athreads[i])
die ("error creating accountant thread %d: rc=%d",
i, (int)GetLastError ());
}
for (i=0; i < N_ACCOUNTANTS; i++)
{
rc = WaitForSingleObject (athreads[i], INFINITE);
if (rc == WAIT_OBJECT_0)
show ("accountant thread %d has terminated", i);
else
fail ("waiting for accountant thread %d failed: %d",
i, (int)GetLastError ());
CloseHandle (athreads[i]);
}
stop_revision_thread = 1;
rc = WaitForSingleObject (rthread, INFINITE);
if (rc == WAIT_OBJECT_0)
show ("revision thread has terminated");
else
fail ("waiting for revision thread failed: %d", (int)GetLastError ());
CloseHandle (rthread);
#else /*!_WIN32*/
# ifdef USE_POSIX_THREADS
pthread_t rthread;
pthread_t athreads[N_ACCOUNTANTS];
int i;
stop_revision_thread = 0;
pthread_create (&rthread, NULL, revision_thread, NULL);
for (i=0; i < N_ACCOUNTANTS; i++)
pthread_create (&athreads[i], NULL, accountant_thread, NULL);
for (i=0; i < N_ACCOUNTANTS; i++)
{
pthread_join (athreads[i], NULL);
show ("accountant thread %d has terminated", i);
}
stop_revision_thread = 1;
pthread_join (rthread, NULL);
show ("revision thread has terminated");
# else /*!USE_POSIX_THREADS*/
verbose++;
show ("no thread support - skipping test\n", PGM);
verbose--;
# endif /*!USE_POSIX_THREADS*/
#endif /*!_WIN32*/
gpgrt_lock_destroy (&accounts_lock);
}
int
main (int argc, char **argv)
{
int last_argc = -1;
int rc;
if (argc)
{
argc--; argv++;
}
while (argc && last_argc != argc )
{
last_argc = argc;
if (!strcmp (*argv, "--help"))
{
puts (
"usage: ./t-lock [options]\n"
"\n"
"Options:\n"
" --verbose Show what is going on\n"
" --debug Flyswatter\n"
);
exit (0);
}
if (!strcmp (*argv, "--verbose"))
{
verbose = 1;
argc--; argv++;
}
else if (!strcmp (*argv, "--debug"))
{
verbose = debug = 1;
argc--; argv++;
}
}
srand (time(NULL)*getpid());
if (!gpg_error_check_version (GPG_ERROR_VERSION))
{
die ("gpg_error_check_version returned an error");
errorcount++;
}
init_accounts ();
check_accounts ();
run_test ();
check_accounts ();
/* Run a second time to check deinit code. */
run_test ();
check_accounts ();
/* And a third time to test an explicit init. */
rc = gpgrt_lock_init (&accounts_lock);
if (rc)
fail ("gpgrt_lock_init failed at %d: %s", __LINE__, gpg_strerror (rc));
run_test ();
check_accounts ();
if (verbose)
print_accounts ();
return errorcount ? 1 : 0;
}
diff --git a/tests/t-poll.c b/tests/t-poll.c
index 57cdb75..56b29c8 100644
--- a/tests/t-poll.c
+++ b/tests/t-poll.c
@@ -1,385 +1,385 @@
/* t-poll.c - Check the poll function
* Copyright (C) 2015 g10 Code GmbH
*
* This file is part of libgpg-error.
*
* libgpg-error is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public License
* as published by the Free Software Foundation; either version 2.1 of
* the License, or (at your option) any later version.
*
* libgpg-error is distributed in the hope that it will be useful, but
* WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
- * License along with this program; if not, see .
+ * License along with this program; if not, see .
*/
/* FIXME: We need much better tests that this very basic one. */
#if HAVE_CONFIG_H
# include
#endif
#include
#include
#include
#include
#include
#include
#include
#ifdef _WIN32
# include
# include
#else
# include
#endif
#define PGM "t-lock"
#include "t-common.h"
#ifdef _WIN32
# define THREAD_RET_TYPE DWORD WINAPI
# define THREAD_RET_VALUE 0
#else
# define THREAD_RET_TYPE void *
# define THREAD_RET_VALUE NULL
#endif
/* Object to convey data to a thread. */
struct thread_arg
{
const char *name;
estream_t stream;
volatile int stop_me;
#ifdef USE_POSIX_THREADS
pthread_t thread;
#elif _WIN32
HANDLE thread;
#endif
};
static struct thread_arg peer_stdin; /* Thread to feed the stdin. */
static struct thread_arg peer_stdout; /* Thread to feed the stdout. */
static struct thread_arg peer_stderr; /* Thread to feed the stderr. */
static estream_t test_stdin;
static estream_t test_stdout;
static estream_t test_stderr;
#if defined(_WIN32) || defined(USE_POSIX_THREADS)
/* This thread feeds data to the given stream. */
static THREAD_RET_TYPE
producer_thread (void *argaddr)
{
struct thread_arg *arg = argaddr;
int i = 0;
(void)arg;
while (!arg->stop_me && i++ < 3)
{
show ("thread '%s' about to write\n", arg->name);
es_fprintf (arg->stream, "This is '%s' count=%d\n", arg->name, i);
es_fflush (arg->stream);
}
es_fclose (arg->stream);
return THREAD_RET_VALUE;
}
/* This thread eats data from the given stream. */
static THREAD_RET_TYPE
consumer_thread (void *argaddr)
{
struct thread_arg *arg = argaddr;
char buf[15];
(void)arg;
while (!arg->stop_me)
{
show ("thread '%s' ready to read\n", arg->name);
if (!es_fgets (buf, sizeof buf, arg->stream))
{
show ("Thread '%s' received EOF or error\n", arg->name);
break;
}
show ("Thread '%s' got: '%s'\n", arg->name, buf);
}
es_fclose (arg->stream);
return THREAD_RET_VALUE;
}
#endif /*_WIN32 || USE_POSIX_THREADS */
static void
launch_thread (THREAD_RET_TYPE (*fnc)(void *), struct thread_arg *th)
{
#ifdef _WIN32
th->thread = CreateThread (NULL, 0, fnc, th, 0, NULL);
if (!th->thread)
die ("creating thread '%s' failed: rc=%d", th->name, (int)GetLastError ());
show ("thread '%s' launched (fd=%d)\n", th->name, es_fileno (th->stream));
#elif USE_POSIX_THREADS
th->stop_me = 0;
if (pthread_create (&th->thread, NULL, fnc, th))
die ("creating thread '%s' failed: %s\n", th->name, strerror (errno));
show ("thread '%s' launched (fd=%d)\n", th->name, es_fileno (th->stream));
# else /* no thread support */
verbose++;
show ("no thread support - skipping test\n", PGM);
verbose--;
#endif /* no thread support */
}
static void
join_thread (struct thread_arg *th)
{
#ifdef _WIN32
int rc;
rc = WaitForSingleObject (th->thread, INFINITE);
if (rc == WAIT_OBJECT_0)
show ("thread '%s' has terminated\n", th->name);
else
fail ("waiting for thread '%s' failed: %d", th->name, (int)GetLastError ());
CloseHandle (th->thread);
#elif USE_POSIX_THREADS
pthread_join (th->thread, NULL);
show ("thread '%s' has terminated\n", th->name);
#endif
}
static void
create_pipe (estream_t *r_in, estream_t *r_out)
{
gpg_error_t err;
int filedes[2];
#ifdef _WIN32
if (_pipe (filedes, 512, 0) == -1)
#else
if (pipe (filedes) == -1)
#endif
{
err = gpg_error_from_syserror ();
die ("error creating a pipe: %s\n", gpg_strerror (err));
}
show ("created pipe [%d, %d]\n", filedes[0], filedes[1]);
*r_in = es_fdopen (filedes[0], "r");
if (!*r_in)
{
err = gpg_error_from_syserror ();
die ("error creating a stream for a pipe: %s\n", gpg_strerror (err));
}
*r_out = es_fdopen (filedes[1], "w");
if (!*r_out)
{
err = gpg_error_from_syserror ();
die ("error creating a stream for a pipe: %s\n", gpg_strerror (err));
}
}
static void
test_poll (void)
{
int ret;
gpgrt_poll_t fds[3];
char buffer[16];
size_t used, nwritten;
int c;
memset (fds, 0, sizeof fds);
fds[0].stream = test_stdin;
fds[0].want_read = 1;
fds[1].stream = test_stdout;
fds[1].want_write = 1;
/* FIXME: We don't use the next stream at all. */
fds[2].stream = test_stderr;
fds[2].want_write = 1;
fds[2].ignore = 1;
used = 0;
while (used || !fds[0].ignore)
{
ret = gpgrt_poll (fds, DIM(fds), -1);
if (ret == -1)
{
fail ("gpgrt_poll failed: %s\n", strerror (errno));
continue;
}
if (!ret)
{
fail ("gpgrt_poll unexpectedly timed out\n");
continue;
}
show ("gpgrt_poll detected %d events\n", ret);
if (fds[0].got_read)
{
/* Read from the producer. */
for (;;)
{
c = es_fgetc (fds[0].stream);
if (c == EOF)
{
if (es_feof (fds[0].stream))
{
show ("reading '%s': EOF\n", peer_stdin.name);
fds[0].ignore = 1; /* Not anymore needed. */
peer_stdin.stop_me = 1; /* Tell the thread to stop. */
}
else if (es_ferror (fds[0].stream))
{
fail ("error reading '%s': %s\n",
peer_stdin.name, strerror (errno));
fds[0].ignore = 1; /* Disable. */
peer_stdin.stop_me = 1; /* Tell the thread to stop. */
}
else
show ("reading '%s': EAGAIN\n", peer_stdin.name);
break;
}
else
{
if (used <= sizeof buffer -1)
buffer[used++] = c;
if (used == sizeof buffer)
{
show ("throttling reading from '%s'\n", peer_stdin.name);
fds[0].ignore = 1;
break;
}
}
}
show ("read from '%s': %zu bytes\n", peer_stdin.name, used);
if (used)
fds[1].ignore = 0; /* Data to send. */
}
if (fds[1].got_write)
{
if (used)
{
ret = es_write (fds[1].stream, buffer, used, &nwritten);
show ("result for writing to '%s': ret=%d, n=%zu, nwritten=%zu\n",
peer_stdout.name, ret, used, nwritten);
if (!ret)
{
assert (nwritten <= used);
/* Move the remaining data to the front of buffer. */
memmove (buffer, buffer + nwritten,
sizeof buffer - nwritten);
used -= nwritten;
}
ret = es_fflush (fds[1].stream);
if (ret)
fail ("Flushing for '%s' failed: %s\n",
peer_stdout.name, strerror (errno));
}
if (!used)
fds[1].ignore = 1; /* No need to send data. */
}
if (used < sizeof buffer / 2 && !peer_stdin.stop_me && fds[0].ignore)
{
show ("accelerate reading from '%s'\n", peer_stdin.name);
fds[0].ignore = 0;
}
}
}
int
main (int argc, char **argv)
{
int last_argc = -1;
if (argc)
{
argc--; argv++;
}
while (argc && last_argc != argc )
{
last_argc = argc;
if (!strcmp (*argv, "--help"))
{
puts (
"usage: ./t-poll [options]\n"
"\n"
"Options:\n"
" --verbose Show what is going on\n"
" --debug Flyswatter\n"
);
exit (0);
}
if (!strcmp (*argv, "--verbose"))
{
verbose = 1;
argc--; argv++;
}
else if (!strcmp (*argv, "--debug"))
{
verbose = debug = 1;
argc--; argv++;
}
}
if (!gpg_error_check_version (GPG_ERROR_VERSION))
{
die ("gpg_error_check_version returned an error");
errorcount++;
}
peer_stdin.name = "stdin producer";
create_pipe (&test_stdin, &peer_stdin.stream);
peer_stdout.name = "stdout consumer";
create_pipe (&peer_stdout.stream, &test_stdout);
peer_stderr.name = "stderr consumer";
create_pipe (&peer_stderr.stream, &test_stderr);
if (es_set_nonblock (test_stdin, 1))
fail ("error setting test_stdin to nonblock: %s\n", strerror (errno));
if (es_set_nonblock (test_stdout, 1))
fail ("error setting test_stdout to nonblock: %s\n", strerror (errno));
if (es_set_nonblock (test_stderr, 1))
fail ("error setting test_stderr to nonblock: %s\n", strerror (errno));
launch_thread (producer_thread, &peer_stdin );
launch_thread (consumer_thread, &peer_stdout);
launch_thread (consumer_thread, &peer_stderr);
test_poll ();
show ("Waiting for threads to terminate...\n");
es_fclose (test_stdin);
es_fclose (test_stdout);
es_fclose (test_stderr);
peer_stdin.stop_me = 1;
peer_stdout.stop_me = 1;
peer_stderr.stop_me = 1;
join_thread (&peer_stdin);
join_thread (&peer_stdout);
join_thread (&peer_stderr);
return errorcount ? 1 : 0;
}
diff --git a/tests/t-printf.c b/tests/t-printf.c
index 069bdc6..7fba012 100644
--- a/tests/t-printf.c
+++ b/tests/t-printf.c
@@ -1,429 +1,429 @@
/* t-printf.c - Check the estream printf fucntions.
* Copyright (C) 2013 g10 Code GmbH
*
* This file is part of libgpg-error.
*
* libgpg-error is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public License
* as published by the Free Software Foundation; either version 2.1 of
* the License, or (at your option) any later version.
*
* libgpg-error is distributed in the hope that it will be useful, but
* WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
- * License along with this program; if not, see .
+ * License along with this program; if not, see .
*/
/* Note that these tests check against glibc behaviour. On non glibc
systems expect non matching return codes in some border cases. */
#if HAVE_CONFIG_H
# include
#endif
#include
#include
#include
#include
#include
#include
#define PGM "t-printf"
#include "t-common.h"
static char *one_test_buf1;
static int one_test_rc1;
static void
one_test_x0 (const char *format, ...)
{
va_list arg_ptr;
show ("format: ->%s<-\n", format);
errno = ENOENT; /* For the "%m" test. */
va_start (arg_ptr, format);
#ifdef HAVE_VASPRINTF
one_test_rc1 = vasprintf (&one_test_buf1, format, arg_ptr);
#else
one_test_rc1 = -1;
#endif
va_end (arg_ptr);
if (one_test_rc1 == -1)
{
fail (" sys: errno=%d (%s)\n", errno, strerror (errno));
one_test_buf1 = NULL;
}
else
show (" sys: ->%s<-\n", one_test_buf1);
}
static void
one_test_x1 (const char *format, ...)
{
int rc2;
va_list arg_ptr;
char *buf2;
errno = ENOENT;
va_start (arg_ptr, format);
rc2 = gpgrt_vasprintf (&buf2, format, arg_ptr);
va_end (arg_ptr);
if (rc2 == -1)
{
fail (" our: errno=%d (%s)\n", errno, strerror (errno));
}
else
show (" our: ->%s<-\n", buf2);
if (one_test_rc1 != -1 && rc2 != -1 && strcmp (one_test_buf1, buf2))
{
fail ("error: output does not match\n"
"format: ->%s<-\n sys: ->%s<-\n our: ->%s<-\n",
format, one_test_buf1, buf2);
}
else if ( one_test_rc1 != rc2 )
{
fail ("error: return codes are different: sys_rc=%d our_rc=%d\n",
one_test_rc1, rc2);
}
free (buf2);
}
static void
one_test_x2 (const char *format, ...)
{
va_list arg_ptr;
char *buf2;
/* Test once more using the bsprintf variant. */
errno = ENOENT;
va_start (arg_ptr, format);
buf2 = gpgrt_vbsprintf (format, arg_ptr);
va_end (arg_ptr);
if (!buf2)
{
fail (" our(2): errno=%d (%s)\n", errno, strerror (errno));
}
else if (verbose)
show (" our: ->%s<-\n", buf2);
if (one_test_rc1 != -1 && buf2 && strcmp (one_test_buf1, buf2))
{
fail ("error: output does not match\n"
"format(2): ->%s<-\n sys: ->%s<-\n our: ->%s<-\n",
format, one_test_buf1, buf2);
}
es_free (buf2);
free (one_test_buf1);
one_test_buf1 = NULL;
}
#define one_test_0(a) \
one_test_x0 (a); \
one_test_x1 (a); \
one_test_x2 (a)
#define one_test_1(a, b) \
one_test_x0 (a, b); \
one_test_x1 (a, b); \
one_test_x2 (a, b)
#define one_test_2(a, b, c) \
one_test_x0 (a, b, c); \
one_test_x1 (a, b, c); \
one_test_x2 (a, b, c)
#define one_test_3(a, b, c, d) \
one_test_x0 (a, b, c, d); \
one_test_x1 (a, b, c, d); \
one_test_x2 (a, b, c, d)
static void
run_tests (void)
{
#ifndef HAVE_VASPRINTF
/* We do not have a system vasprintf. */
show ("run-tests: disabled due to missing vasprintf.\n");
#else /*HAVE_VASPRINTF */
/*one_test ("%d %% %'d", 17, 19681977);*/
one_test_2 ("%d %% %d", 17, 768114563);
one_test_2 ("%d %% %d", 17, -768114563);
one_test_1 ("%d", 17);
one_test_1 ("%4d", 17);
one_test_1 ("%40d", 17);
one_test_1 ("%-d", 17);
one_test_1 ("%-4d", 17);
one_test_1 ("%-140d", 17);
one_test_1 ("%d", -17);
one_test_1 ("%4d", -17);
one_test_1 ("%40d", -17);
one_test_1 ("%-d", -17);
one_test_1 ("%-4d", -17);
one_test_1 ("%-40d", -17);
one_test_1 ("%+4d", 17);
one_test_1 ("%+4d", -17);
one_test_1 ("%-+4d", 17);
one_test_1 ("%-+4d", -17);
one_test_1 ("% 4d", 17);
one_test_1 ("% 4d", -17);
one_test_1 ("%- +4d", 17);
one_test_1 ("%- +4d", -17);
one_test_1 ("%.4d", 17);
one_test_1 ("%.0d", 17);
one_test_1 ("%.0d", 0);
one_test_1 ("%.4d", -17);
one_test_1 ("%.0d", -17);
one_test_1 ("%6.4d", 17);
one_test_1 ("%6.4d", -17);
one_test_1 ("%6.0d", 0);
one_test_1 ("%4.6d", 17);
one_test_1 ("%4.6d", -17);
one_test_1 ("% 4.6d", 17);
one_test_1 ("% 6.0d", 0);
one_test_1 ("%.4d", 17);
one_test_1 ("%04d", 17);
one_test_1 ("%.4d", -17);
one_test_1 ("%04d", -17);
one_test_1 ("%0.d", 0);
one_test_2 ("%*d", 7, 42);
one_test_2 ("%*d", -7, 42);
one_test_2 ("%.*d", 7, 42);
one_test_2 ("%.*d", -7, 42);
one_test_3 ("%*.*d", 10, 7, 42);
one_test_3 ("%*.*d", 10, -7, 42);
one_test_3 ("%*.*d", -10, 7, 42);
one_test_3 ("%*.*d", -10, -7, 42);
one_test_2 ("%*x", 7, 42);
one_test_2 ("%*x", -7, 42);
one_test_2 ("%.*x", 7, 42);
one_test_2 ("%.*x", -7, 42);
one_test_3 ("%*.*x", 10, 7, 42);
one_test_3 ("%*.*x", 10, -7, 42);
one_test_3 ("%*.*x", -10, 7, 42);
one_test_3 ("%*.*x", -10, -7, 42);
one_test_2 ("%#*x", 7, 42);
one_test_2 ("%#*x", -7, 42);
one_test_2 ("%#.*x", 7, 42);
one_test_2 ("%#.*x", -7, 42);
one_test_3 ("%#*.*x", 10, 7, 42);
one_test_3 ("%#*.*x", 10, -7, 42);
one_test_3 ("%#*.*x", -10, 7, 42);
one_test_3 ("%#*.*x", -10, -7, 42);
one_test_2 ("%*X", 7, 42);
one_test_2 ("%*X", -7, 42);
one_test_2 ("%.*X", 7, 42);
one_test_2 ("%.*X", -7, 42);
one_test_3 ("%*.*X", 10, 7, 42);
one_test_3 ("%*.*X", 10, -7, 42);
one_test_3 ("%*.*X", -10, 7, 42);
one_test_3 ("%*.*X", -10, -7, 42);
one_test_2 ("%#*X", 7, 42);
one_test_2 ("%#*X", -7, 42);
one_test_2 ("%#.*X", 7, 42);
one_test_2 ("%#.*X", -7, 42);
one_test_3 ("%#*.*X", 10, 7, 42);
one_test_3 ("%#*.*X", 10, -7, 42);
one_test_3 ("%#*.*X", -10, 7, 42);
one_test_3 ("%#*.*X", -10, -7, 42);
one_test_2 ("%*o", 7, 42);
one_test_2 ("%*o", -7, 42);
one_test_2 ("%.*o", 7, 42);
one_test_2 ("%.*o", -7, 42);
one_test_3 ("%*.*o", 10, 7, 42);
one_test_3 ("%*.*o", 10, -7, 42);
one_test_3 ("%*.*o", -10, 7, 42);
one_test_3 ("%*.*o", -10, -7, 42);
one_test_2 ("%#*o", 7, 42);
one_test_2 ("%#*o", -7, 42);
one_test_2 ("%#.*o", 7, 42);
one_test_2 ("%#.*o", -7, 42);
one_test_3 ("%#*.*o", 10, 7, 42);
one_test_3 ("%#*.*o", 10, -7, 42);
one_test_3 ("%#*.*o", -10, 7, 42);
one_test_3 ("%#*.*o", -10, -7, 42);
one_test_1 ("%s", "the quick brown fox jumps over the lazy dogs back");
one_test_1 ("%.0s", "the quick brown fox jumps over the lazy dogs back");
one_test_1 ("%.10s", "the quick brown fox jumps over the lazy dogs back");
one_test_1 ("%.48s", "the quick brown fox jumps over the lazy dogs back");
one_test_1 ("%.49s", "the quick brown fox jumps over the lazy dogs back");
one_test_1 ("%.50s", "the quick brown fox jumps over the lazy dogs back");
one_test_1 ("%.51s", "the quick brown fox jumps over the lazy dogs back");
one_test_1 ("%48s", "the quick brown fox jumps over the lazy dogs back");
one_test_1 ("%49s", "the quick brown fox jumps over the lazy dogs back");
one_test_1 ("%50s", "the quick brown fox jumps over the lazy dogs back");
one_test_1 ("%51s", "the quick brown fox jumps over the lazy dogs back");
one_test_1 ("%-51s", "the quick brown fox jumps over the lazy dogs back");
one_test_1 ("/%s=", "CN");
one_test_1 ("%f", 3.1415926535);
one_test_1 ("%f", -3.1415926535);
one_test_1 ("%.10f", 3.1415926535);
one_test_1 ("%.2f", 3.1415926535);
one_test_1 ("%.1f", 3.1415926535);
one_test_1 ("%.0f", 3.1415926535);
one_test_1 ("%.20f", 3.1415926535);
one_test_1 ("%10.10f", 3.1415926535);
one_test_1 ("%10.2f", 3.1415926535);
one_test_1 ("%10.1f", 3.1415926535);
one_test_1 ("%10.0f", 3.1415926535);
one_test_1 ("%30.20f", 3.1415926535);
one_test_1 ("%10.10f", -3.1415926535);
one_test_1 ("%10.2f", -3.1415926535);
one_test_1 ("%10.1f", -3.1415926535);
one_test_1 ("%10.0f", -3.1415926535);
one_test_1 ("%30.20f", -3.1415926535);
one_test_1 ("%-10f", 3.1415926535);
one_test_1 ("%-10.10f", 3.1415926535);
one_test_1 ("%-10.2f", 3.1415926535);
one_test_1 ("%-10.1f", 3.1415926535);
one_test_1 ("%-10.0f", 3.1415926535);
one_test_1 ("%-30.20f", 3.1415926535);
one_test_1 ("%-10f", -3.1415926535);
one_test_1 ("%-10.10f", -3.1415926535);
one_test_1 ("%-10.2f", -3.1415926535);
one_test_1 ("%-10.1f", -3.1415926535);
one_test_1 ("%-10.0f", -3.1415926535);
one_test_1 ("%-30.20f", -3.1415926535);
one_test_1 ("%#.0f", 3.1415926535);
one_test_1 ("%#10.0f", 3.1415926535);
one_test_1 ("%#10.0f", -3.1415926535);
one_test_1 ("%-#10.0f", 3.1415926535);
one_test_1 ("%-#10.0f", -3.1415926535);
one_test_1 ("%e", 3.1415926535);
one_test_1 ("%g", 3.1415926535);
one_test_1 ("%a", 1.0);
one_test_1 ("%a", -1.0);
one_test_1 ("%a", 3.1415926535);
#ifdef HAVE_LONG_DOUBLE
one_test_1 ("%La", (long double)1.0);
one_test_1 ("%La", (long double)-1.0);
one_test_1 ("%La", (long double)3.1415926535);
#endif
#ifdef __GLIBC__
/* "%m" is a glibc extension so this _test_ will only work on such a
system. */
one_test_0 ("%m");
one_test_1 ("%d=%m", 17);
one_test_2 ("%2$d:%m:%1$d", 42, 17);
#endif /*__GLIBC__*/
#endif /*HAVE_VASPRINTF */
}
static void
check_snprintf (void)
{
char buffer[20];
int rc, rc2;
size_t tmplen, blen, blen2;
rc = gpgrt_snprintf (buffer, 0, "%*s", 18, "");
if (rc != 18)
printf ("rc=%d\n", rc );
rc = gpgrt_snprintf (buffer, sizeof buffer, "%*s", 18, "");
if (rc != 18)
printf ("rc=%d, strlen(buffer)=%d\n", rc, (int)strlen (buffer));
rc = gpgrt_snprintf (buffer, sizeof buffer, "%*s", 19, "");
if (rc != 19)
printf ("rc=%d, strlen(buffer)=%d\n", rc, (int)strlen (buffer));
rc = gpgrt_snprintf (buffer, sizeof buffer, "%*s", 20, "");
if (rc != 20)
printf ("rc=%d, strlen(buffer)=%d\n", rc, (int)strlen (buffer));
rc = gpgrt_snprintf (buffer, sizeof buffer, "%*s", 21, "");
if (rc != 21)
printf ("rc=%d, strlen(buffer)=%d\n", rc, (int)strlen (buffer));
for (tmplen = 0; tmplen <= sizeof buffer; tmplen++)
{
rc = gpgrt_snprintf (buffer, tmplen, "%04d%02d%02dT%02d%02d%02d",
1998, 9, 7, 16, 56, 05);
blen = strlen (buffer);
rc2 = snprintf (buffer, tmplen, "%04d%02d%02dT%02d%02d%02d",
1998, 9, 7, 16, 56, 05);
blen2 = strlen (buffer);
if (rc != rc2 || blen != blen2)
printf ("snprintf test with len %u gives %d instead of %d (%u,%u)\n",
(unsigned int)tmplen, rc, rc2,
(unsigned int)blen, (unsigned int)blen2);
}
}
int
main (int argc, char **argv)
{
int last_argc = -1;
if (argc)
{
argc--; argv++;
}
while (argc && last_argc != argc )
{
last_argc = argc;
if (!strcmp (*argv, "--help"))
{
puts (
"usage: ./" PGM " [options]\n"
"\n"
"Options:\n"
" --verbose Show what is going on\n"
" --debug Flyswatter\n"
);
exit (0);
}
if (!strcmp (*argv, "--verbose"))
{
verbose = 1;
argc--; argv++;
}
else if (!strcmp (*argv, "--debug"))
{
verbose = debug = 1;
argc--; argv++;
}
}
setlocale (LC_NUMERIC, "");
if (!gpg_error_check_version (GPG_ERROR_VERSION))
{
die ("gpg_error_check_version returned an error");
errorcount++;
}
run_tests ();
check_snprintf ();
#ifdef __GLIBC__
return !!errorcount;
#else
return 0;
#endif
}
diff --git a/tests/t-version.c b/tests/t-version.c
index ce8f41b..4606dbc 100644
--- a/tests/t-version.c
+++ b/tests/t-version.c
@@ -1,98 +1,98 @@
/* t-version.c - Check the version info function
* Copyright (C) 2013 g10 Code GmbH
*
* This file is part of libgpg-error.
*
* libgpg-error is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public License
* as published by the Free Software Foundation; either version 2.1 of
* the License, or (at your option) any later version.
*
* libgpg-error is distributed in the hope that it will be useful, but
* WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
- * License along with this program; if not, see .
+ * License along with this program; if not, see .
*/
#ifdef HAVE_CONFIG_H
#include
#endif
#include
#include
#include
#include
#include "../src/gpg-error.h"
static const char *logpfx = "";
static int verbose;
static int debug;
static int errorcount;
int
main (int argc, char **argv)
{
int last_argc = -1;
if (argc)
{
logpfx = *argv;
argc--; argv++;
}
while (argc && last_argc != argc )
{
last_argc = argc;
if (!strcmp (*argv, "--help"))
{
puts (
"usage: ./version [options]\n"
"\n"
"Options:\n"
" --verbose Show what is going on\n"
);
exit (0);
}
if (!strcmp (*argv, "--verbose"))
{
verbose = 1;
argc--; argv++;
}
else if (!strcmp (*argv, "--debug"))
{
verbose = debug = 1;
argc--; argv++;
}
}
if (!gpg_error_check_version (GPG_ERROR_VERSION))
{
fprintf (stderr, "%s: gpg_error_check_version returned an error\n",
logpfx);
errorcount++;
}
if (!gpg_error_check_version ("1.10"))
{
fprintf (stderr, "%s: gpg_error_check_version returned an "
"error for an old version\n", logpfx);
errorcount++;
}
if (gpg_error_check_version ("15"))
{
fprintf (stderr, "%s: gpg_error_check_version did not return an error"
" for a newer version\n", logpfx);
errorcount++;
}
if (verbose || errorcount)
{
printf ("Version from header: %s (0x%06x)\n",
GPG_ERROR_VERSION, GPG_ERROR_VERSION_NUMBER);
printf ("Version from binary: %s\n", gpg_error_check_version (NULL));
printf ("Copyright blurb ...:%s\n", gpg_error_check_version ("\x01\x01"));
}
return errorcount ? 1 : 0;
}