Page MenuHome GnuPG
Files F19741823

No OneTemporary
Actions

View Options

This file is larger than 256 KB, so syntax highlighting was skipped.
This document is not UTF8. It was detected as Shift JIS and converted to UTF8 for display.
diff --git a/ChangeLog b/ChangeLog
index 08b06c0c..2cecbae2 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,1032 +1,1042 @@
+2006-08-28 Werner Koch <wk@g10code.com>
+
+ Released 1.2.3.
+
+ * configure.ac: Set LT to C13/A1/R2.
+
+2006-07-26 Werner Koch <wk@g10code.com>
+
+ * configure.ac: New option --disable-optimization.
+
2005-12-08 Werner Koch <wk@g10code.com>
* configure.ac: Changed the random device names for netbsd. From
Christian Biere.
2005-10-05 Werner Koch <wk@g10code.com>
Released 1.2.2.
* configure.ac: Set LT to C13/A1/R1.
2005-09-14 Werner Koch <wk@g10code.com>
* configure.ac: New option --enable-noexecstack.
2005-06-10 Werner Koch <wk@g10code.com>
* configure.ac: Moved basic tests to the top.
2005-02-03 Moritz Schulte <moritz@g10code.com>
* THANKS: Updated.
2005-01-05 Werner Koch <wk@g10code.com>
Released 1.2.1.
* configure.ac: Set LT to C13/A2/R0.
2005-01-04 Werner Koch <wk@g10code.com>
* Makefile.am (ACLOCAL_AMFLAGS): New for -I m4.
(AUTOMAKE_OPTIONS): New to create a bzip archive.
Updated to automake 1.9.
* acinclude.m4: Updated for automake 1.9.
2004-12-18 Werner Koch <wk@g10code.com>
* autogen.sh: Added option --build-w32 to run W32 cross compiling
configure.
2004-08-09 Moritz Schulte <moritz@g10code.com>
* THANKS: Updated.
* libgcrypt.txt: Slightly updated.
2004-07-04 Moritz Schulte <moritz@g10code.com>
* THANKS: Updated.
2004-04-21 Werner Koch <wk@gnupg.org>
* configure.ac: Don't print a warning if GNU make was not found.
2004-05-07 Moritz Schulte <moritz@g10code.de>
* THANKS: Updated.
2004-04-02 Thomas Schwinge <schwinge@nic-nac-project.de>
* autogen.sh: Added ACLOCAL_FLAGS.
2004-04-15 Werner Koch <wk@gnupg.org>
Released 1.2.0.
* configure.ac: Set LT to C12/A1/R1.
2004-04-06 Werner Koch <wk@gnupg.org>
* config.guess, config.sub, ltmain.sh: Updated to those from
libtools 1.5.4.
2004-03-29 Werner Koch <wk@gnupg.org>
Released 1.1.94.
* configure.ac: Set LT to C12/A1/R0.
2004-03-10 Marcus Brinkmann <marcus@g10code.de>
* configure.ac (LIBGCRYPT_CONFIG_LIBS_PTHREAD,
LIBGCRYPT_CONFIG_CFLAGS_PTHREAD, LIBGCRYPT_CONFIG_LIBS_PTH,
LIBGCRYPT_CONFIG_CFLAGS_PTH, have_pth, have_pthread, AC_CHECK_PTH,
AC_CHECK_LIB(pthread), HAVE_PTH, HAVE_PTHREAD): Removed.
2004-03-06 Werner Koch <wk@gnupg.org>
Released 1.1.93.
* configure.ac (LIBGCRYPT_CONFIG_SONAME_NUMBER): Replaced by
LIBGCRYPT_CONPIG_API_VERSION. Set it to 1. Set LT to C11/A0/R1.
2004-03-05 Werner Koch <wk@gnupg.org>
* configure.ac (LIBGCRYPT_CONFIG_SONAME_NUMBER): New.
2004-02-20 Werner Koch <wk@gnupg.org>
Released 1.1.92.
* configure.ac: Set LT to C11/A0/R0.
2004-02-11 Werner Koch <wk@gnupg.org>
* autogen.sh (check_version): Removed bashism and simplified.
2004-02-06 Werner Koch <wk@gnupg.org>
* configure.ac: Add rfc2268 cipher algorithm.
2004-01-25 Moritz Schulte <mo@g10code.com>
* THANKS: Updated.
2003-12-19 Werner Koch <wk@gnupg.org>
Released 1.1.91.
* configure.ac: Bumbed LT version to C10/A3/R1.
2003-12-08 Werner Koch <wk@gnupg.org>
* Makefile.am (dist-hook): Don't distribute stuff from the now
obsolete scripts dir.
(EXTRA_DIST): Remove README_alpha
* README-alpha: Removed.
* configure.ac (AM_CONFIG_AUX_DIR): Removed.
* COPYING.DOC: Removed.
* Makefile.am (EXTRA_DIST): Added README.CVS and
autogen.sh. Removed COPYING.DOC.
2003-11-14 Werner Koch <wk@gnupg.org>
Released 1.1.90.
* configure.ac: Bumbed LT version to C10/A3/R0.
* configure.ac (have_ld_version_script): Set the default in
a separate test.
(PRINTABLE_OS_NAME): Don't handle the Hurd extra, this leads to
conflicts with BSD based GNU systems. The Hurd has now a working
uname.
2003-11-04 Werner Koch <wk@gnupg.org>
* configure.ac (USE_SHA1): Make sure it is always included.
(USE_RMD160): Removed this AM conditional.
2003-10-31 Werner Koch <wk@gnupg.org>
* configure.ac: Bumbed version number to 1.1.90-cvs for futher
development
Released 1.1.44.
* acinclude.m4 (AC_CHECK_PTH): Added.
* configure.ac: Use it here instead of the generic lib test.
Bumbed LT vesion to C9/A2/R0.
2003-10-27 Werner Koch <wk@gnupg.org>
* configure.ac: Give a hint on where libgpg-error is available.
Reformatted long lines. Don't include gcrypt-defs.h.
(--enable-gcc-warnings): New option.
2003-10-24 Moritz Schulte <mo@g10code.com>
* configure.ac: Check for socklen_t.
2003-10-11 Moritz Schulte <mo@g10code.com>
* acinclude.m4: Update AM_PATH_GPG_ERROR macro.
2003-09-04 Werner Koch <wk@gnupg.org>
Released 1.1.43.
* configure.ac: Require libgpg-error 0.4 due to the prime interface.
2003-08-29 Werner Koch <wk@gnupg.org>
* acinclude.m4 (GNUPG_SYS_SYMBOL_UNDERSCORE): Re-implemented.
* configure.ac: Use it here.
2003-08-27 Moritz Schulte <mo@g10code.com>
* configure.ac: Substitute: LIBGCRYPT_CONFIG_LIBS_PTHREAD,
LIBGCRYPT_CONFIG_CFLAGS_PTHREAD, LIBGCRYPT_CONFIG_LIBS_PTH,
LIBGCRYPT_CONFIG_CFLAGS_PTH, LIBGCRYPT_THREAD_MODULES.
2003-08-07 Moritz Schulte <moritz@g10code.com>
* configure.ac: Fail, if libgpg-error could not be found.
2003-07-31 Werner Koch <wk@gnupg.org>
Released 1.1.42.
* configure.ac: Set LT version to 7/0/0.
2003-07-30 Werner Koch <wk@gnupg.org>
* AUTHORS (Maintainer): Assigned Moritz as Maintainer.
2003-07-30 Moritz Schulte <moritz@g10code.com>
* NEWS: Include much more complete list of `Interface changes
relative to the 1.1.12 release'.
2003-07-14 Moritz Schulte <moritz@g10code.com>
* configure.ac: Bumbed version number up to 1.1.42-cvs.
2003-07-09 Moritz Schulte <moritz@g10code.com>
* configure.ac: Reintroduce --disable-asm, since it is needed by
mpi/config.links.
2003-07-05 Moritz Schulte <moritz@g10code.com>
* README: Few changes, mention libgpg-error.
2003-06-18 Moritz Schulte <moritz@g10code.com>
* configure.ac (available_ciphers): Removed Serpent, hrrm.
2003-06-17 Moritz Schulte <moritz@g10code.com>
* acinclude.m4: Removed macro definitions: GNUPG_CHECK_FAQPROG,
GNUPG_CHECK_ENDIAN, GNUPG_CHECK_CACHE, GNUPG_CHECK_PIC,
GNUPG_CHECK_EXPORTDYNAMIC, GNUPG_CHECK_IPC, GNUPG_PROG_NM,
GNUPG_SYS_SYMBOL_UNDERSCORE, GNUPG_FUNC_MKDIR_TAKES_ONE_ARG,
GPH_PROG_DB2ANY.
Added macro definitions: AM_PATH_GPG_ERROR.
* configure.ac: Use alternative approach for building based on
conditional sources, which does not make automake eat all your
memory, etc.
Removed unused tests.
Renamed --enable-static-rnd to --enable-random.
Use Autoconf's AC_C_BIGENDIAN macro instead of our own.
Re-organized the whole file.
2003-06-16 Moritz Schulte <moritz@g10code.com>
* configure.ac (AC_CONFIG_FILES): Removed doc/version.sgml.
2003-06-11 Moritz Schulte <moritz@g10code.com>
* configure.ac: Remove --enable-libgpg-error flag.
Ue AC_PATH_GPG_ERROR.
2003-06-09 Moritz Schulte <moritz@g10code.com>
* NEWS: Mention API changes and libgpg-error.
2003-05-25 Moritz Schulte <moritz@g10code.com>
* configure.ac (USE_LIBGPG_ERROR): Implementation of the
--enable-libgpg-error switch.
Define USE_LIBGPG_ERROR in LIBGCRYPT_CONFIG_FLAGS, in case
libgpg-error is used.
2003-05-22 Moritz Schulte <moritz@g10code.com>
* configure.ac (AC_CHECK_HEADERS): Removed unused headers:
termio.h, langinfo.h.
(AC_CHECK_FUNCS): Removed unused functions: strsep, strlwr,
tcgetattr, setrlimit, strftime, nl_langinfo, sigaction,
sigprocmask, fopen64, fstat64.
2003-04-27 Moritz Schulte <moritz@g10code.com>
* README: Documented new configure switches.
Mention the --enable-maintainer-switch.
* configure.ac: Merged some code from GnuPG's configure.ac for
disabling sha512/tiger in case no 64 data types are available.
2003-04-17 Moritz Schulte <moritz@g10code.com>
* configure.ac: Include support for sha512.
2003-04-17 Moritz Schulte <moritz@g10code.com>
* AUTHORS: Updated.
2003-04-16 Moritz Schulte <moritz@g10code.com>
* configure.ac: Implement command line switches: --enable-ciphers,
--enable-pubkey-ciphers and --enable-digests.
Set Automake conditionals and config.h symbols depending on the
selected ciphers, pubkey-ciphers, digests and random-modules.
* acinclude.m4 (LIST_MEMBER): New macro.
* configure.ac: Simplified, removed code for parsing
EXTRA_PROGRAMS from Makefile.am.
2003-04-08 Moritz Schulte <moritz@g10code.com>
* configure.ac: Merged random-module selection code from GnuPG's
configure.ac.
2003-04-07 Moritz Schulte <moritz@g10code.com>
* configure.ac: Removed code for generating contruct.c.
Remove digest modules from the static_modules list, only handle
random module selection.
2003-03-24 Moritz Schulte <moritz@g10code.com>
* NEWS: Mention new CBC_MAC flag.
* AUTHORS (Maintainer): Update entry for Simon Josefsson.
2003-03-04 Moritz Schulte <moritz@g10code.com>
* TODO: Remove item about resetting handles, since
gcry_cipher_reset is implemented by now.
* NEWS: Mentioned gcry_cipher_reset.
2003-01-21 Werner Koch <wk@gnupg.org>
* README (Configure options): New.
* configure.ac (have_ld_version_script): New option
--enable-ld-version-script.
2003-01-20 Simon Josefsson <jas@extundo.com>
* configure.ac (MODULES_IN_CIPHER): Add crc.
2003-01-20 Werner Koch <wk@gnupg.org>
Released 1.1.12.
* configure.ac (LIBGCRYPT_LT_REVISION): Bumbed up.
2002-12-21 Werner Koch <wk@gnupg.org>
Released 1.1.11.
* configure.ac (LIBGCRYPT_LT_CURRENT: Bumbed to 6/5/0 due to a new
interface
2002-12-19 Werner Koch <wk@gnupg.org>
* configure.ac (have_pthread): Check for pthreads in libc.
(have_ld_version_script): New.
2002-11-10 Werner Koch <wk@gnupg.org>
* configure.ac (MODULES_IN_CIPHER): Add md4.c. By Simon Josefsson.
2002-09-20 Werner Koch <wk@gnupg.org>
Released 1.1.10.
* configure.ac (HAVE_DEV_RANDOM_IOCTL): Don't check for it; it is
not used.
(AS_CHECK_HEADERS): Check for sys/select.h.
* Makefile.am (DIST_SUBDIRS): New to include the w32-dll directory
2002-09-18 Timo Schulz <ts@winpt.org>
* configure.ac: Added makefile for the W32 DLL.
2002-09-17 Werner Koch <wk@gnupg.org>
* configure.ac: Check for Pth and Pthreads.
2002-08-23 Werner Koch <wk@gnupg.org>
Released 1.1.9.
* configure.ac (LIBGCRYPT_CONFIG_CFLAGS): Renamed from
LIBGCRYPT_CFLAGS and removed the libpath because it is set by the
config script.
(LIBGCRYPT_LT_REVISION): Set LT version to 5/4/1.
2002-06-25 Werner Koch <wk@gnupg.org>
Released 1.1.8.
* configure.ac: Set LT version to 5/4/0.
2002-05-21 Werner Koch <wk@gnupg.org>
Released 1.1.7.
* configure.ac: Set LT version to 4/3/0.
2002-05-17 Werner Koch <wk@gnupg.org>
* configure.ac: Removed all the dynamic loading stuff.
2002-05-16 Werner Koch <wk@gnupg.org>
* configure.ac: Reordered the C_CHECK_FUNCS.
2002-05-15 Werner Koch <wk@gnupg.org>
* configure.ac: Adjusted for new MPI module stuff.
2002-05-14 Werner Koch <wk@gnupg.org>
Changed license to the LGPL.
2002-05-02 Werner Koch <wk@gnupg.org>
* jnlib/: Removed.
* Makefile.am (SUBDIRS): Removed jnlib.
* configure.ac (jnlib/Makefile): Removed.
* configure.ac: Define _REENTRANT.
2002-02-18 Werner Koch <wk@gnupg.org>
* configure.ac (MPI_EXTRA_ASM_OBJS): Use .lo suffix.
(AC_CANONICAL_TARGET): Added.
2002-02-07 Werner Koch <wk@gnupg.org>
Released 1.1.6.
2002-01-24 Werner Koch <wk@gnupg.org>
* jnlib/: Replaced by a fresh copy from GnuPG (actually the NewPG
development branch). Adjusted Makefile.am and jnlib-config.h
accordingly.
2001-12-18 Werner Koch <wk@gnupg.org>
Released 1.1.5.
* Makefile.am (dist-hook): Only look in mpi and scripts for
distfiles; this way we don't include those of a stale "make dist"
directory.
* acinclude.m4 (GNUPG_FIX_HDR_VERSION): Make it work with the new
automake.
* configure.ac: Don't chmod db2any.
2001-08-06 Werner Koch <wk@gnupg.org>
* configure.ac: Removed cross compiling hacks.
2001-08-03 Werner Koch <wk@gnupg.org>
Released 1.1.4.
* acinclude.m4 (GNUPG_CHECK_TYPEDEF): Define GNU Source.
Migrated to autoconf 2.52.
* acinclude.m4: Removed GNUPG_LINK_FILES and converted.
* acconfig.h: Removed
* configure.in: Replaced by...
* configure.ac: and modified for use with autoconf 2.52. Replaced
GNUPG_LINK_FILES with AC_CONFIG_LINKS and moved some informational
messages to the end. Removed --enable-m-debug
* tests/: New.
* Makefile.am: Included tests directory
* configure.in (DYNLINK_MOD_CFLAGS): Use -shared with dec-osf.
Reported by Chris Adams. Merged some cases.
2001-05-31 Werner Koch <wk@gnupg.org>
Released 1.1.3.
* configure.in: Use _gcry_ prefix when creating the cipher constructor.
* acconfig.h (_GCRYPT_IN_LIBGCRYPT): Define it here.
2001-05-28 Werner Koch <wk@gnupg.org>
* acinclude.m4 (GPH_PROG_DOCBOOK): Removed.
(GPH_PROG_DB2ANY): New. Taken from GPH.
* configure.in: Use it here.
2000-12-19 Werner Koch <wk@gnupg.org>
Major change:
Removed all GnuPG stuff and renamed this piece of software
to gcrypt. The directory gcrypt has been renamed to src.
2000-11-14 Werner Koch <wk@gnupg.org>
Version 1.1.2 released.
2000-11-13 Werner Koch <wk@gnupg.org>
* acinclude.m4 (GNUPG_FIX_HDR_VERSION): VPATH build fix.
2000-10-10 Werner Koch <wk@gnupg.org>
* Makefile.am (dist-hook): Create the version file.
* configure.in: Set the libtool version here, removed the need
for the version file.
Mon Sep 18 16:35:45 CEST 2000 Werner Koch <wk@openit.de>
* acinclude.m4 (GNUPG_CHECK_MLOCK): Removed that silly mkdir().
* configure.in: Changes to allow for Solaris random device.
By Nils Ellmenreich.
(--with-egd-socket): New.
* configure.in (GNUPG_HOMEDIR): New.
* configure.in: Check for fstat64 and fopen64
* acinclude.m4 (GNUPG_CHECK_FAQPROG): New.
* configure.in: Test for this.
* configure.in (DYNLINK_MOD_CFLAGS): Fix by David Champion.
Tue Aug 22 14:31:15 CEST 2000 Werner Koch <wk@openit.de>
Version 1.1.1
Fri Aug 18 14:27:14 CEST 2000 Werner Koch <wk@openit.de>
* agent/: New.
* Makefile.am, configure.in: Support for the new directory.
Mon Jul 17 16:35:47 CEST 2000 Werner Koch <wk@>
* configure.in (mingw32): Changes to allow for mingw32msvc
Fri Jul 14 19:38:23 CEST 2000 Werner Koch <wk@>
The big merge between this one and the stable branch 1.0. Still need
to merge TNANKS, AUTHORS and such. It probaly does not compile yet.
* acinclude.m4 (GNUPG_CHECK_MLOCK): Fixed syntax error in C code.
* configure.in: Add check for termio.h, wait unctiosn and sigaction.
* acinclude.m4, configure.in (GNUPG_CHECK_GNUMAKE): New.
* acinclude.m4 (MKDIR_TAKES_ONE_ARG): Check some headers. By Ga? Qu駻i.
* configure.in (AM_INIT_AUTOMAKE): Use this now. By Ga?.
* acinclude.m4 (GNUPG_CHECK_EXPORTDYNAMIC): Replacement for
GNUPG_CHECK_RDYNAMIC which should handle gcc with non GNU ld nicer.
Contributed by Dave Dykstra.
* configure.in (GNYPG_CHECK_RDYNAMIC): Replaced by the new check.
* configure.in: Add a test for unisgned long long.
* configure.in (DYNLINK_MOD_CFLAGS): Set different for NetBSD.
* configure.in: Add check for clock_gettime
* configure.in (ALL_LINGUAS): Add nl.
* configure.in (ALL_LINGUAS): Add Esperanto.
* configure.in (ALL_LINGUAS): Add sv and ja.
* configure.in: Use /usr/local for CFLAGS and LDFLAGS when
target is freebsd. By R駑i.
* configure.in: Do not set development version when the version has
a dash in it. Suggested by Dave Dykstra.
* configure.in: Removed substitution for doc/gph/Makefile.
Do all the gcc warning only in maintainer mode.
* configure.in (dlopen): Use CHECK_FUNC for a test of dlopen in libc.
Suggested by Alexandre Oliva.
(-Wall): Moved the settting of gcc warning options near to the end
so that tests don't get confused. Suggested by Paul D. Smith.
* acinclude.m4 (GNUPG_SYS_NM_PARSE): Added BSDI support.
(GNUPG_CHECK_RDYNAMIC): Ditto.
* acinclude.m4 (GNUPG_CHECK_MLOCK): Changed the way to test for
librt. Test suggested by Jeff Long.
* acinclude.m4 (GNUPG_CHECK_MLOCK): Do librt check only when
we can't link a test program. This way GNU systems don't need
to link against linrt.
(GNUPG_CHECK_IPC): Fixed use of TRY_COMPILE macro. From Tim Mooney.
* acinclude.m4 (GNUPG_SYS_SYMBOL_UNDERSCORE): Add support for
DJGPP.
(GNUPG_CHECK_MLOCK): Check whether mlock sits in librt.
* acinclude.m4 (GNUPG_CHECK_RDYNAMIC): Add NetBSD. By Thomas Klausner.
* acconfig.h (HAVE_MLOCK): Added
Mon Mar 13 19:22:46 CET 2000 Werner Koch <wk@openit.de>
* configure.in: Now uses the Docbook M4s from GPH.
Mon Jan 31 17:46:35 CET 2000 Werner Koch <wk@>
* Makefile.am: Re-added tools. By R駑i.
Mon Jan 31 16:37:34 CET 2000 Werner Koch <wk@gnupg.de>
* configure.in: Create a symlink for types.h in gcrypt/.
Thu Jan 27 18:00:44 CET 2000 Werner Koch <wk@gnupg.de>
* configure.in (g10defs.h): Replaced by gnupg-defs.h
Mon Jan 24 13:04:28 CET 2000 Werner Koch <wk@gnupg.de>
* jnlib/ : New.
* configure.in: Do set development version when the version has
a dash in it. Suggested by Dave Dykstra.
Thu Dec 9 17:22:27 CET 1999 Werner Koch <wk@gnupg.de>
* acinclude.m4 (GNUPG_FIX_HDR_VERSION): New.
* configure.in: Check and fix the version number of gcrypt/gcrypt.h
so that it is always the save as VERSION.
Thu Oct 28 16:17:46 CEST 1999 Werner Koch <wk@gnupg.de>
* Started with development series 1.1 on 1999-10-26
Tue Oct 26 14:10:21 CEST 1999 Werner Koch <wk@gnupg.de>
* README-alpha: New
* configure.in: Fixed quoting in test for development version.
* THANKS: Add entries for Michael, Brenno and J Horacio who did
very nice Howto documents - I apoligize for forgetting to mention them
earlier.
Fri Sep 17 12:56:42 CEST 1999 Werner Koch <wk@isil.d.shuttle.de>
* configure.in: Add "-lcap" when capabilities are requested.
Add the conditional CROSS_COMPILING.
* Makefile.am: Don't use checks when CROSS_COMPILING.
Wed Sep 15 16:22:17 CEST 1999 Werner Koch <wk@isil.d.shuttle.de>
* configure.in (ALL_LINGUAS): Add pt_PT.
* configure.in: Some tweaks for cross compiling under MingW32
* acconfig.h (USE_STATIC_RNDW32): New.
Tue Sep 7 17:08:10 CEST 1999 Werner Koch <wk@isil.d.shuttle.de>
* VERSION: Set to 1.0.0.
Mon Sep 6 19:59:08 CEST 1999 Werner Koch <wk@isil.d.shuttle.de>
* configure.in: Create makefile in doc/gph
* acinclude.m4 (GNUPG_FUNC_MKDIR_TAKES_ONE_ARG): New
* configure.in: use the above.
Thu Sep 2 16:40:55 CEST 1999 Werner Koch <wk@isil.d.shuttle.de>
* VERSION: Set to 0.9.11.
Tue Aug 31 17:20:44 CEST 1999 Werner Koch <wk@isil.d.shuttle.de>
* configure.in: Minor changes to the OS/2 and Mingw32 system labels.
Add a printable name for Hurd.
Mon Aug 30 20:38:33 CEST 1999 Werner Koch <wk@isil.d.shuttle.de>
* configure.in: Some support for DJGPP (Mark Elbrecht)
Wed Aug 4 10:34:46 CEST 1999 Werner Koch <wk@isil.d.shuttle.de>
* VERSION: Set to 0.9.10.
Mon Jul 26 09:34:46 CEST 1999 Werner Koch <wk@isil.d.shuttle.de>
* acinclude.m4 (GNUPG_SYS_SYMBOL_UNDERSCORE): remove init of ac_cv_...
* Makefile.am (DISCLEANFILES): New
Fri Jul 23 13:53:03 CEST 1999 Werner Koch <wk@isil.d.shuttle.de>
* VERSION: Set to 0.9.9.
* configure.in: Print a notice when rndunix is used.
Thu Jul 15 10:15:35 CEST 1999 Werner Koch <wk@isil.d.shuttle.de>
* acinclude.m4 (GNUPG_SYS_SYMBOL_UNDERSCORE): Fixed last modification.
Wed Jul 7 13:08:40 CEST 1999 Werner Koch <wk@isil.d.shuttle.de>
* Makefile.am: Support for libtool.
* configure.in: Ditto.
Tue Jun 29 21:44:25 CEST 1999 Werner Koch <wk@isil.d.shuttle.de>
* configure.in (use_local_zlib): The lost dollar is back.
* acinclude.m4 (GNUPG_SYS_SYMBOL_UNDERSCORE): Add EMX case.
* configure.in: Another variant of the MX vendor string
* configure.in (--with-capabilities): Some test code (Remi).
Sat Jun 26 12:15:59 CEST 1999 Werner Koch <wk@isil.d.shuttle.de>
* acinclude.m4 (GNUPG_CHECK_RDYNAMIC): Support for HPUX and IRIX.
* configure.in (HAVE_DL_SHL_LOAD): New for HPUX (Dave Dykstra).
* VERSION: Now 0.9.8
Wed Jun 16 20:16:21 CEST 1999 Werner Koch <wk@isil.d.shuttle.de>
* configure.in: Add test for docbook-to-man
Tue Jun 15 12:21:08 CEST 1999 Werner Koch <wk@isil.d.shuttle.de>
* acinclude.m4 (GNUPG_SYS_NM_PARSE): Support for {net,free}bsd,
Thu Jun 10 14:18:23 CEST 1999 Werner Koch <wk@isil.d.shuttle.de>
* configure.in (ZLIB,GDBM): Check both, header and lib.
Sat Jun 5 15:30:33 CEST 1999 Werner Koch <wk@isil.d.shuttle.de>
* pkclist.c (key_present_in_pk_list): New (Michael).
Tue May 25 19:50:32 CEST 1999 Werner Koch <wk@isil.d.shuttle.de>
* configure.in (IS_DEVELOPMENT_VERSION): Fixed detection.
Sun May 23 14:20:22 CEST 1999 Werner Koch <wk@isil.d.shuttle.de>
* acinclude.m4 (GNUPG_SYS_SYMBOL_UNDERSCORE): assume yes when
cross-compiling.
Mon May 17 21:54:43 CEST 1999 Werner Koch <wk@isil.d.shuttle.de>
* configure.in (socket): Fix for Unisys by Katsuhiro Kondou.
Sat May 8 19:28:08 CEST 1999 Werner Koch <wk@isil.d.shuttle.de>
* NEWS: Add a marker line which I forgot to do for 0.9.6.
Thu May 6 14:18:17 CEST 1999 Werner Koch <wk@isil.d.shuttle.de>
* README: Minor updates
* VERSION: Now 0.9.6
Thu Apr 8 09:35:53 CEST 1999 Werner Koch <wk@isil.d.shuttle.de>
* acinclude.m4 (GNUPG_CHECK_RDYNAMIC): Fix for
amiga-openbsd (Peter Reich)
(GNUPG_PROG_NM): Ditto
Wed Apr 7 20:51:39 CEST 1999 Werner Koch <wk@isil.d.shuttle.de>
* Makefile.am (g10defs.h): Removed.
* configure.in (AC_OUTPUT_COMMANDS): Create g10defs.h
Sat Mar 20 12:55:33 CET 1999 Werner Koch <wk@isil.d.shuttle.de>
* VERSION: Now 0.9.5
Sun Mar 14 19:34:36 CET 1999 Werner Koch <wk@isil.d.shuttle.de>
* acinclude.m4 (AM_SYS_SYMBOL_UNDERSCORE): Removed because it is
now in the latest libtool.
Thu Mar 11 16:39:46 CET 1999 Werner Koch <wk@isil.d.shuttle.de>
* configure.in: Removed the need for libtool
Mon Mar 8 20:47:17 CET 1999 Werner Koch <wk@isil.d.shuttle.de>
* configure.in (DLSYM_NEEDS_UNDERSCORE): Replaced.
* acinclude.in (AM_SYS_SYMBOL_UNDERSCORE): New.
* VERSION: Now 0.9.4
Sun Feb 28 19:11:00 CET 1999 Werner Koch <wk@isil.d.shuttle.de>
* configure.in (dld): Test disabled.
Fri Feb 26 17:55:41 CET 1999 Werner Koch <wk@isil.d.shuttle.de>
* encode.c (encode_simple): temporary fix.
Wed Feb 24 11:07:27 CET 1999 Werner Koch <wk@isil.d.shuttle.de>
* configure.in: New option --enable-static-rnd.
Mon Feb 22 20:04:00 CET 1999 Werner Koch <wk@isil.d.shuttle.de>
* BUGS: Now we assign bug numbers.
* OBUGS: New to keep rack o fixed bugs (CVS only)
Fri Feb 19 18:01:54 CET 1999 Werner Koch <wk@isil.d.shuttle.de>
* VERSION: Released 0.9.3
Fri Feb 19 15:49:15 CET 1999 Werner Koch <wk@isil.d.shuttle.de>
* acinclude.m4: Removed gettext macros.
Tue Feb 16 14:10:02 CET 1999 Werner Koch <wk@isil.d.shuttle.de>
* configure.in (socket): Check for -lsocket and -lnsl.
(osf4): Disable all warnings for DEC's cc.
(-Wall): Add more warning options for gcc
Sat Feb 13 12:04:43 CET 1999 Werner Koch <wk@isil.d.shuttle.de>
* configure.in: Changed detection of compiler flags.
* intl/ : Removed directory
Wed Feb 10 17:15:39 CET 1999 Werner Koch <wk@isil.d.shuttle.de>
* acinclude.m4 (GNUPG_CHECK_RDYNAMIC): Fix for freebsd 2.2
* configure.in: a lot of changes to allow selection of modules.
Add support for OS/2.
* acinclude.m4: add some more caching
* README: Spelling and grammar corrections (John A. Martin)
* INSTALL: Ditto.
Wed Jan 20 21:40:21 CET 1999 Werner Koch <wk@isil.d.shuttle.de>
* configure.in: --enable-m-guard is now default
Wed Jan 13 12:49:36 CET 1999 Werner Koch <wk@isil.d.shuttle.de>
* INSTALL: Applied new information how to build rpms by Fabio Coatti
* Makefile.in (gnupg.spec): Changed the names.
Tue Jan 12 11:17:18 CET 1999 Werner Koch <wk@isil.d.shuttle.de>
* config.links (m68k-atari-mint): New
Tue Jan 12 09:17:19 CET 1999 Ga? Qu駻i <gqueri@mail.dotcom.fr>
* all: Fixed typos all over the place
Sat Jan 9 16:02:23 CET 1999 Werner Koch <wk@isil.d.shuttle.de>
* configure.in: Add a way to statically link rndunix
Sun Jan 3 15:28:44 CET 1999 Werner Koch <wk@isil.d.shuttle.de>
* acinclude.m4 (GNUPG_CHECK_RDYNAMIC): New.
* configure.in (DYNLOAD_CFLAGS): Use result from CHECK_RDYNAMIC
Wed Dec 23 13:18:14 CET 1998 Werner Koch <wk@isil.d.shuttle.de>
* README: Replaced the command overview with a short intro.
Sat Dec 12 18:40:32 CET 1998 Werner Koch <wk@isil.d.shuttle.de>
* configure.in: Add check for dlopen in libc (Greg Troxel)
and a new define
* acconfig.h (DLSYM_NEEDS_UNDERSCORE): New.
Thu Dec 10 20:15:36 CET 1998 Werner Koch <wk@isil.d.shuttle.de>
* acinclude.m (GNUPG_CHECK_PIC): New
* configure.in, acinclude.m4: Renamed all WK_ to GNUPG_
Tue Dec 8 15:09:29 CET 1998 Werner Koch <wk@isil.d.shuttle.de>
* VERSION: Set to 0.4.5
Wed Nov 25 12:38:29 1998 Werner Koch (wk@isil.d.shuttle.de)
* configure.in (USE_RNDLINUX): New.
Fri Nov 20 19:34:57 1998 Werner Koch (wk@isil.d.shuttle.de)
* VERSION: Released 0.4.4
* configure.in (try_asm_modules): For option --disable-asm
Tue Nov 10 19:32:40 1998 Werner Koch (wk@isil.d.shuttle.de)
* configure.in (MPI_SFLAGS): New.
Tue Nov 10 13:44:53 1998 Werner Koch (wk@isil.d.shuttle.de)
* ABOUT-NLS: New
* configure.in (AC_REVISION): New.
Sun Nov 8 18:20:35 1998 Werner Koch (wk@isil.d.shuttle.de)
* VERSION: Set to 0.4.3
Sun Oct 25 19:49:37 1998 Werner Koch (wk@isil.d.shuttle.de)
* Makefile.am (g10defs.h): New macro GNUPG_DATADIR.
Wed Oct 21 17:24:24 1998 Werner Koch (wk@isil.d.shuttle.de)
* configure.in: Removed gettext kludge
* acinclude.m4: Add patched AM_WITH_NKS macro
Tue Oct 20 19:03:36 1998 Werner Koch (wk@isil.d.shuttle.de)
* configure.in: Kludge to make AM_GNU_GETTEXT work,
changed some macors to more modern versions. Also
changeg the all makefiles to remove duplicate ../intl.
* acinclude.m4: Removed the gettext stuff, as this
already comes with automake now.
Wed Oct 14 12:11:34 1998 Werner Koch (wk@isil.d.shuttle.de)
* configure.in (NAME_OF_DEV_RANDOM): New.
(DYNLINK_MOD_CFLAGS): New.
Thu Oct 8 10:55:15 1998 Werner Koch (wk@isil.d.shuttle.de)
* Makefile.am (g10defs.h): creates include file
* acconfig.h: now includes g10defs.h
* configure.in: Removed G10_LOCALEDIR and GNUPG_LIB
Thu Sep 17 18:49:40 1998 Werner Koch (wk@(none))
* Makefile.am (dist-hook): Now creates RPM file.
* scripts/gnupg.spec: New template file for RPMs
Thu Jul 30 19:17:07 1998 Werner Koch (wk@(none))
* acinclude.h (WK_CHECK_IPC): New
* configure.in : Add checks for SysV IPC
Thu Jun 25 11:18:49 1998 Werner Koch (wk@isil.d.shuttle.de)
* configure.in (--disable-dynload): New.
Wed Jun 10 07:48:59 1998 Werner Koch,mobil,,, (wk@tobold)
* configure.in (GNUPG_LIBDIR): New.
Mon May 25 19:10:59 1998 Werner Koch (wk@isil.d.shuttle.de)
* rand-unix.c (fast_random_poll): fixed syntax bug.
Mon May 11 10:21:31 1998 Werner Koch (wk@isil.d.shuttle.de)
* configure.in (PRINTABLE_OS_NAME): Linux is now GNU/Linux
Tue Apr 14 19:08:05 1998 Werner Koch (wk@isil.d.shuttle.de)
* [all files]: Applied Matthew Skala's typo and grammar fixes.
Wed Mar 4 10:32:40 1998 Werner Koch (wk@isil.d.shuttle.de)
* configure.in (getrusage,gettimeofday): New tests.
Fri Feb 27 13:14:17 1998 Werner Koch (wk@isil.d.shuttle.de)
* configure.in (--disable-m-guard): New.
Thu Feb 26 17:09:27 1998 Werner Koch (wk@isil.d.shuttle.de)
* configure.in, acinclude.m4, intl/, po/: New macros taken
from GNOME, switched to automake 1.2f
Thu Feb 26 09:05:46 1998 Werner Koch (wk@isil.d.shuttle.de)
* configure.in (doc/Makefile): New
Thu Feb 26 07:40:47 1998 Werner Koch (wk@isil.d.shuttle.de)
* configure.in: Changed gettext stuff
Wed Feb 25 11:44:10 1998 Werner Koch (wk@isil.d.shuttle.de)
* checks/*test : restructured the directory.
Tue Feb 24 15:59:12 1998 Werner Koch (wk@isil.d.shuttle.de)
* configure.in: Changed the name of the package to GNUPG and
chnaged several other names too.
Wed Feb 18 17:36:45 1998 Werner Koch (wk@isil.d.shuttle.de)
* Makefile.am (checks): New.
Sat Feb 14 15:37:55 1998 Werner Koch (wk@isil.d.shuttle.de)
* configure.in (mpi_config_done): Removed asm links caching.
Sat Feb 14 14:02:20 1998 Werner Koch (wk@isil.d.shuttle.de)
* configure.in (PRINTABLE_OS_NAME): New.
* acconfig.h: Likewise.
Fri Feb 13 19:43:41 1998 Werner Koch (wk@isil.d.shuttle.de)
* configure.in : Fixed zlib stuff
* Makefile.am: Likewise
Copyright 1998, 1999, 2000, 2001, 2002, 2003,
2004 Free Software Foundation, Inc.
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 file 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.
diff --git a/INSTALL b/INSTALL
index 095b1eb4..23e5f25d 100644
--- a/INSTALL
+++ b/INSTALL
@@ -1,231 +1,236 @@
Installation Instructions
*************************
-Copyright (C) 1994, 1995, 1996, 1999, 2000, 2001, 2002, 2004 Free
+Copyright (C) 1994, 1995, 1996, 1999, 2000, 2001, 2002, 2004, 2005 Free
Software Foundation, Inc.
This file is free documentation; the Free Software Foundation gives
unlimited permission to copy, distribute and modify it.
Basic Installation
==================
These are generic installation instructions.
The `configure' shell script attempts to guess correct values for
various system-dependent variables used during compilation. It uses
those values to create a `Makefile' in each directory of the package.
It may also create one or more `.h' files containing system-dependent
definitions. Finally, it creates a shell script `config.status' that
you can run in the future to recreate the current configuration, and a
file `config.log' containing compiler output (useful mainly for
debugging `configure').
It can also use an optional file (typically called `config.cache'
and enabled with `--cache-file=config.cache' or simply `-C') that saves
the results of its tests to speed up reconfiguring. (Caching is
disabled by default to prevent problems with accidental use of stale
cache files.)
If you need to do unusual things to compile the package, please try
to figure out how `configure' could check whether to do them, and mail
diffs or instructions to the address given in the `README' so they can
be considered for the next release. If you are using the cache, and at
some point `config.cache' contains results you don't want to keep, you
may remove or edit it.
The file `configure.ac' (or `configure.in') is used to create
`configure' by a program called `autoconf'. You only need
`configure.ac' if you want to change it or regenerate `configure' using
a newer version of `autoconf'.
The simplest way to compile this package is:
1. `cd' to the directory containing the package's source code and type
`./configure' to configure the package for your system. If you're
using `csh' on an old version of System V, you might need to type
`sh ./configure' instead to prevent `csh' from trying to execute
`configure' itself.
Running `configure' takes awhile. While running, it prints some
messages telling which features it is checking for.
2. Type `make' to compile the package.
3. Optionally, type `make check' to run any self-tests that come with
the package.
4. Type `make install' to install the programs and any data files and
documentation.
5. You can remove the program binaries and object files from the
source code directory by typing `make clean'. To also remove the
files that `configure' created (so you can compile the package for
a different kind of computer), type `make distclean'. There is
also a `make maintainer-clean' target, but that is intended mainly
for the package's developers. If you use it, you may have to get
all sorts of other programs in order to regenerate files that came
with the distribution.
Compilers and Options
=====================
Some systems require unusual options for compilation or linking that the
`configure' script does not know about. Run `./configure --help' for
details on some of the pertinent environment variables.
You can give `configure' initial values for configuration parameters
by setting variables in the command line or in the environment. Here
is an example:
./configure CC=c89 CFLAGS=-O2 LIBS=-lposix
*Note Defining Variables::, for more details.
Compiling For Multiple Architectures
====================================
You can compile the package for more than one kind of computer at the
same time, by placing the object files for each architecture in their
own directory. To do this, you must use a version of `make' that
supports the `VPATH' variable, such as GNU `make'. `cd' to the
directory where you want the object files and executables to go and run
the `configure' script. `configure' automatically checks for the
source code in the directory that `configure' is in and in `..'.
If you have to use a `make' that does not support the `VPATH'
variable, you have to compile the package for one architecture at a
time in the source code directory. After you have installed the
package for one architecture, use `make distclean' before reconfiguring
for another architecture.
Installation Names
==================
-By default, `make install' will install the package's files in
-`/usr/local/bin', `/usr/local/man', etc. You can specify an
-installation prefix other than `/usr/local' by giving `configure' the
-option `--prefix=PREFIX'.
+By default, `make install' installs the package's commands under
+`/usr/local/bin', include files under `/usr/local/include', etc. You
+can specify an installation prefix other than `/usr/local' by giving
+`configure' the option `--prefix=PREFIX'.
You can specify separate installation prefixes for
architecture-specific files and architecture-independent files. If you
-give `configure' the option `--exec-prefix=PREFIX', the package will
-use PREFIX as the prefix for installing programs and libraries.
-Documentation and other data files will still use the regular prefix.
+pass the option `--exec-prefix=PREFIX' to `configure', the package uses
+PREFIX as the prefix for installing programs and libraries.
+Documentation and other data files still use the regular prefix.
In addition, if you use an unusual directory layout you can give
options like `--bindir=DIR' to specify different values for particular
kinds of files. Run `configure --help' for a list of the directories
you can set and what kinds of files go in them.
If the package supports it, you can cause programs to be installed
with an extra prefix or suffix on their names by giving `configure' the
option `--program-prefix=PREFIX' or `--program-suffix=SUFFIX'.
Optional Features
=================
Some packages pay attention to `--enable-FEATURE' options to
`configure', where FEATURE indicates an optional part of the package.
They may also pay attention to `--with-PACKAGE' options, where PACKAGE
is something like `gnu-as' or `x' (for the X Window System). The
`README' should mention any `--enable-' and `--with-' options that the
package recognizes.
For packages that use the X Window System, `configure' can usually
find the X include and library files automatically, but if it doesn't,
you can use the `configure' options `--x-includes=DIR' and
`--x-libraries=DIR' to specify their locations.
Specifying the System Type
==========================
There may be some features `configure' cannot figure out automatically,
but needs to determine by the type of machine the package will run on.
Usually, assuming the package is built to be run on the _same_
architectures, `configure' can figure that out, but if it prints a
message saying it cannot guess the machine type, give it the
`--build=TYPE' option. TYPE can either be a short name for the system
type, such as `sun4', or a canonical name which has the form:
CPU-COMPANY-SYSTEM
where SYSTEM can have one of these forms:
OS KERNEL-OS
See the file `config.sub' for the possible values of each field. If
`config.sub' isn't included in this package, then this package doesn't
need to know the machine type.
If you are _building_ compiler tools for cross-compiling, you should
-use the `--target=TYPE' option to select the type of system they will
+use the option `--target=TYPE' to select the type of system they will
produce code for.
If you want to _use_ a cross compiler, that generates code for a
platform different from the build platform, you should specify the
"host" platform (i.e., that on which the generated programs will
eventually be run) with `--host=TYPE'.
Sharing Defaults
================
If you want to set default values for `configure' scripts to share, you
can create a site shell script called `config.site' that gives default
values for variables like `CC', `cache_file', and `prefix'.
`configure' looks for `PREFIX/share/config.site' if it exists, then
`PREFIX/etc/config.site' if it exists. Or, you can set the
`CONFIG_SITE' environment variable to the location of the site script.
A warning: not all `configure' scripts look for a site script.
Defining Variables
==================
Variables not defined in a site shell script can be set in the
environment passed to `configure'. However, some packages may run
configure again during the build, and the customized values of these
variables may be lost. In order to avoid this problem, you should set
them in the `configure' command line, using `VAR=value'. For example:
./configure CC=/usr/local2/bin/gcc
-will cause the specified gcc to be used as the C compiler (unless it is
-overridden in the site shell script).
+causes the specified `gcc' to be used as the C compiler (unless it is
+overridden in the site shell script). Here is a another example:
+
+ /bin/bash ./configure CONFIG_SHELL=/bin/bash
+
+Here the `CONFIG_SHELL=/bin/bash' operand causes subsequent
+configuration-related scripts to be executed by `/bin/bash'.
`configure' Invocation
======================
`configure' recognizes the following options to control how it operates.
`--help'
`-h'
Print a summary of the options to `configure', and exit.
`--version'
`-V'
Print the version of Autoconf used to generate the `configure'
script, and exit.
`--cache-file=FILE'
Enable the cache: use and save the results of the tests in FILE,
traditionally `config.cache'. FILE defaults to `/dev/null' to
disable caching.
`--config-cache'
`-C'
Alias for `--cache-file=config.cache'.
`--quiet'
`--silent'
`-q'
Do not print messages saying which checks are being made. To
suppress all normal output, redirect it to `/dev/null' (any error
messages will still be shown).
`--srcdir=DIR'
Look for the package's source code in directory DIR. Usually
`configure' can determine that directory automatically.
`configure' also accepts some other, not widely useful, options. Run
`configure --help' for more details.
diff --git a/NEWS b/NEWS
index 40cfe87d..830452e5 100644
--- a/NEWS
+++ b/NEWS
@@ -1,436 +1,441 @@
-Noteworthy changes in version 1.2.3
+Noteworthy changes in version 1.2.3 (2006-08-28)
------------------------------------------------
-Noteworthy changes in version 1.2.2 (2005-10-05}
+ * Rewrote gcry_mpi_rshift to allow arbitrary shift counts.
+
+ * Minor bug fixes.
+
+
+Noteworthy changes in version 1.2.2 (2005-10-05)
------------------------------------------------
* Made the RNG immune against fork without exec.
* Minor changes to some function declarations. Buffer arguments are
now typed as void pointer. This should not affect any compilation.
* A bug in the definition of gcry_cipher_register has been fixed.
* Interface changes relative to the 1.2.1 release:
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
gcry_cipher_encrypt CHANGED: Arguments IN and OUT are now void*.
gcry_cipher_decrypt CHANGED: Arguments IN and OUT are now void*.
gcry_create_nonce CHANGED: Argument BUFFER is now void*.
gcry_md_ctl CHANGED: Argument BUFFER is now void*.
gcry_sexp_sprint CHANGED: Argument BUFFER is now void*.
gcry_mpi_scan CHANGED: Argument BUFFER is now void*.
gcry_cipher_register CHANGED: Argument ALGORITHM_ID is now int*.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Noteworthy changes in version 1.2.1 (2005-01-05)
------------------------------------------------
* Portability fixes, memory allocation fixes and other minor things.
* Support to build as a W32 static library.
* Changed the way the RNG gets initialized. This allows to keep it
uninitialized as long as no random numbers are used. To override
this, the new macro gcry_fast_random_poll may be used. It is in
general a good idea to spread this macro into the application code
to make sure that these polls happen often enough.
* Interface changes relative to the 1.2.0 release:
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
gcry_fast_random_poll NEW
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Noteworthy changes in version 1.2.0 (2004-04-15)
------------------------------------------------
* First stable release.
Noteworthy changes in version 1.1.94 (2004-03-29)
-------------------------------------------------
* The support for multi-threaded users goes into its third
incarnation. We removed compile time support for thread libraries.
To support the thread library of your choice, you have to set up
callback handlers at initialization time. New data structures, a
new control command, and default initializers are provided for this
purpose.
* Interface changes relative to the 1.1.93 release:
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
libgcrypt-config --thread OBSOLETE
libgcrypt-pth.la REMOVED
libgcrypt-pthread.la REMOVED
GCRYCTL_SET_THREAD_CBS NEW
struct gcrypt_thread_cbs NEW
enum gcry_thread_option NEW
GCRY_THREAD_OPTION_PTH_IMPL NEW
GCRY_THREAD_OPTION_PTHREAD_IMPL NEW
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Noteworthy changes in version 1.1.93 (2004-03-06)
-------------------------------------------------
* The automatic thread library detection has finally been removed.
From now on, only linking explicitely to libgcrypt, libgcrypt-pth
or libgcrypt-pthread is supported.
Noteworthy changes in version 1.1.92 (2004-02-20)
-------------------------------------------------
* Minor bug fixes.
* Included a limited implementation of RFC2268.
* Changed API of the gcry_ac_ functions. Only a very few programs
should be affected by this.
* Interface changes relative to the 1.1.91 release:
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
GCRY_CIPHER_RFC2268_40 NEW.
gcry_ac_data_set CHANGED: New argument FLAGS.
gcry_ac_data_get_name CHANGED: New argument FLAGS.
gcry_ac_data_get_index CHANGED: New argument FLAGS.
gcry_ac_key_pair_generate CHANGED: New and reordered arguments.
gcry_ac_key_test CHANGED: New argument HANDLE.
gcry_ac_key_get_nbits CHANGED: New argument HANDLE.
gcry_ac_key_get_grip CHANGED: New argument HANDLE.
gcry_ac_data_search REMOVED.
gcry_ac_data_add REMOVED.
GCRY_AC_DATA_FLAG_NO_BLINDING REMOVED.
GCRY_AC_FLAG_NO_BLINDING NEW: Replaces above.
Noteworthy changes in version 1.1.91 (2003-12-19)
-------------------------------------------------
* Code cleanups and minor bug fixes.
Noteworthy changes in version 1.1.90 (2003-11-14)
-------------------------------------------------
* The use of the GCRY_WEAK_RANDOM level is now deprecated in favor of
the new gcry_create_nonce function.
* gcry_sexp_build now supports a "%b" format to include a memory buffer.
* Minor configuration fixes.
* Interface changes relative to the 1.1.44 release:
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
gcry_create_nonce NEW
gcry_sexp_build ENHANCED
Noteworthy changes in version 1.1.44 (2003-10-31)
-------------------------------------------------
* Bug fixes and more code cleanups.
* Enhanced the prime API.
* Interface changes relative to the 1.1.43 release:
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
gcry_prime_group_generator NEW
gcry_prime_release_factors NEW
Noteworthy changes in version 1.1.43 (2003-09-04)
-------------------------------------------------
* Bug fixes and internal code cleanups.
* Support for the Serpent cipher algorithm.
* Interface changes relative to the 1.1.42 release:
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
gcry_prime_generate NEW
gcry_prime_check NEW
Noteworthy changes in version 1.1.42 (2003-07-31)
-------------------------------------------------
* Major API cleanup. Applications need to be converted to the new
API. See README.apichanges for hints on how to do that. Backward
compatibility is provided where it was possible without too much
effort and did not collide with the overall sanitization effort.
However, this is only for ease of transition. NO DEPRECATED
FUNCTION OR DATA TYPE IS CONSIDERED A PART OF THE API OR ABI AND
WILL BE DROPPED IN THE FUTURE WITHOUT CHANGING THE SONAME OF THE
LIBRARY.
* If gcrypt.h is included in sources compiled by GCC 3.1 or later,
deprecated attributes will warn about use of obsolete functions and
type definitions. You can suppress these warnings by passing
-Wno-deprecated-declarations to the gcc command.
* gcry_check_version must be called from now on to initialize the
library, it is not longer optional.
* Removed `libgcrypt errno' concept.
* Libgcrypt depends on libgpg-error, a library that provides error
codes and according functions for all GnuPG components. Functions
that used to return error codes asa `int' have been changed to
return a code of type `gcry_error_t'. All GCRYERR_* error symbols
have been removed, since they are now contained in libgpg-error
(GPG_ERR_*). All functions and types in libgpg-error have also been
wrapped in Libgcrypt. The new types are gcry_err_code_t and
gcry_err_source_t. The new functions are gcry_err_code,
gcry_err_source, gcry_error, gcry_err_make, gcry_error_from_errno,
gcry_err_make_from_errno, gcry_err_code_from_errno,
gcry_err_code_to_errno, gcry_strsource.
* New function gcry_mpi_dump to help in debugging.
* Added alternative interface for asymmetric cryptography.
* CRC-32, CRC-32 a'la RFC 1510, CRC-24 a'la RFC 2440 are now
supported.
* SHA-256, SHA-384 and SHA-512 are now supported.
* 128 bit Twofish is now supported.
* The random module won't print the "not enough random bytes
available" anymore. A new progress status is issued instead.
* CBC-MAC for block ciphers is now supported, by using a
GCRY_CIPHER_CBC_MAC cipher flag.
* CTR mode for block ciphers is now supported.
* The public RSA exponent can now be specified in key generation.
* RSA blinding is now supported and is used automatically for RSA
decryption. It can be explicitely disabled by using the
`no-blinding' symbol in the `flags' S-Expression or by using the
GCRY_AC_FLAG_DATA_NO_BLINDING flag when using the ac interface.
* gcry_sexp_canon_len does not use a `historically encoded' error
code anymore.
* Interface changes relative to the 1.1.12 release:
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
GCRY_MPI DEPRECATED; Use: gcry_mpi_t
GcryMPI DEPRECATED; Use: gcry_mpi_t
GCRY_SEXP DEPRECATED; Use: gcry_sexp_t
GcrySexp DEPRECATED; Use: gcry_sexp_t
GCRY_CIPHER_HD DEPRECATED; Use: gcry_cipher_hd_t
GcryCipherHd DEPRECATED; Use: gcry_cipher_hd_t
GCRY_MD_HD DEPRECATED; Use: gcry_md_hd_t
GcryMDHd DEPRECATED; Use: gcry_md_hd_t
gcry_error_t NEW
gcry_err_code_t NEW
gcry_err_source_t NEW
gcry_err_make NEW
gcry_error NEW
gcry_err_code NEW
gcry_err_source NEW
gcry_err_code_from_errno NEW
gcry_err_code_to_errno NEW
gcry_err_make_from_errno NEW
gcry_error_from_errno NEW
gcry_strsource NEW
GCRYERR_{some error code} REMOVED; Use GPG_ERR_*
from libgpg-error instead.
gcry_errno REMOVED
gcry_sexp_canon_len CHANGED
gcry_sexp_build_array NEW
gcry_mpi_scan CHANGED: New argument to separate in/out args.
gcry_mpi_print CHANGED: Ditto.
gcry_mpi_dump NEW
gcry_cipher_open CHANGED
gcry_cipher_reset NEW
gcry_cipher_register NEW
gcry_cipher_unregister NEW
gcry_cipher_list NEW
gcry_cipher_algo_keylen REPLACED macro with function.
gcry_cipher_algo_blklen REPLACED macro with function.
gcry_pk_register NEW
gcry_pk_unregister NEW
gcry_pk_list NEW
gcry_pk_decrypt ENHANCED: Allows flag to return
complete S-expression.
gcry_md_open CHANGED
gcry_md_copy CHANGED
gcry_md_is_enabled NEW
gcry_md_is_secure NEW
gcry_md_register NEW
gcry_md_unregister NEW
gcry_md_list NEW
gcry_ac_data_t NEW
gcry_ac_key_t NEW
gcry_ac_key_pair_t NEW
gcry_ac_handle_t NEW
gcry_ac_key_spec_rsa_t NEW
gcry_ac_data_new NEW
gcry_ac_data_destroy NEW
gcry_ac_data_set NEW
gcry_ac_data_copy NEW
gcry_ac_data_length NEW
gcry_ac_data_get_name NEW
gcry_ac_data_get_index NEW
gcry_ac_data_clear NEW
gcry_ac_open NEW
gcry_ac_close NEW
gcry_ac_key_init NEW
gcry_ac_key_pair_generate NEW
gcry_ac_key_pair_extract NEW
gcry_ac_key_data_get NEW
gcry_ac_key_test NEW
gcry_ac_key_get_nbits NEW
gcry_ac_key_get_grip NEW
gcry_ac_key_destroy NEW
gcry_ac_key_pair_destroy NEW
gcry_ac_data_encrypt NEW
gcry_ac_data_decrypt NEW
gcry_ac_data_sign NEW
gcry_ac_data_verify NEW
gcry_ac_id_to_name NEW
gcry_ac_name_to_id NEW
gcry_handler_progress_t NEW
gcry_handler_alloc_t NEW
gcry_handler_secure_check_t NEW
gcry_handle_realloc_t NEW
gcry_handler_free_t NEW
gcry_handler_no_mem_t NEW
gcry_handler_error_t NEW
gcry_handler_log_t NEW
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Noteworthy changes in version 1.1.12 (2003-01-20)
-------------------------------------------------
* gcry_pk_sign, gcry_pk_verify and gcry_pk_encrypt can now handle an
optional pkcs1 flags parameter in the S-expression. A similar flag
may be passed to gcry_pk_decrypt but it is only syntactically
implemented.
* New convenience macro gcry_md_get_asnoid.
* There is now some real stuff in the manual.
Noteworthy changes in version 1.1.11 (2002-12-21)
-------------------------------------------------
* Don't export internal symbols anymore (currently only for GNU systems)
* New algorithm: MD4
* Implemented ciphertext stealing.
* Smaller bugs fixes and a few new OIDs.
* Interface changes relative to the 1.1.8 release:
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
gcry_cipher_cts NEW
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Noteworthy changes in version 1.1.10 (2002-09-20)
-------------------------------------------------
* Fixed shared library builds for i386, PPC and Sparc.
* Added simple benchmark tool.
* Replaced the internal mutexes by code which automatically adapts to
the used threading library. Currently Pth and Pthread are
supported. For non-ELF systems the GNU toolchain is now required..
* Added untested support to build Windows DLLs.
Noteworthy changes in version 1.1.9 (2002-08-23)
------------------------------------------------
* Support for plain old DES.
Noteworthy changes in version 1.1.8 (2002-06-25)
------------------------------------------------
* Minor cleanups and exported a few new functions.
* Interface changes relative to the 1.1.7 release:
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
gcry_mpi_div NEW
gcry_mpi_mod NEW
gcry_mpi_invm NEW
gcry_mpi_swap NEW
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Noteworthy changes in version 1.1.7 (2002-05-21)
------------------------------------------------
* Libgcrypt is now distributed under the terms of the GNU Lesser
General Public License; see the README file for details.
* It is possible to use libgcrypt w/o intialized secure memory.
* Libgcrypt should now be thread safe after the initialization.
gcry_control (GCRYCRL_INITIALIZATION_FINISHED,NULL,0) should have
been called before creating additional threads.
* Interface changes relative to the 1.1.6 release:
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
GCRYCTL_DISABLE_INTERNAL_LOCKING NEW
GCRYCTL_DISABLE_SECMEM NEW
GCRYCTL_INITIALIZATION_FINISHED NEW
GCRYCTL_INITIALIZATION_FINISHED_P NEW
GCRYCTL_ANY_INITIALIZATION_P NEW
gcry_strdup NEW
gcry_sexp_create NEW
gcry_sexp_new NEW
gcry_set_progress_handler NEW
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Noteworthy changes in version 1.1.6 (2002-02-07)
------------------------------------------------
* Enhanced the S-expression conversion functions.
Noteworthy changes in version 1.1.5 (2001-12-18)
------------------------------------------------
* gcry_{cipher,md}_map_name are now able to map stringified object IDs.
* New functions gcry_sexp_canon_len and gcry_cipher_mode_from_oid.
* Closed some memory leaks.
Noteworthy changes in version 1.1.4 (2001-08-03)
------------------------------------------------
* Arcfour does now work.
* Some minor fixes.
* Added a first test program
* Migrated to autoconf 2.52.
Noteworthy changes in version 1.1.3 (2001-05-31)
------------------------------------------------
* First release of Libgcrypt which is a result of splitting GnuPG
into into libgcrypt and GnuPG.
Copyright 2001, 2002, 2003, 2004 Free Software Foundation, Inc.
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 file 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.
diff --git a/TODO b/TODO
index d35aa464..c819045c 100644
--- a/TODO
+++ b/TODO
@@ -1,35 +1,37 @@
What's left to do -*- outline -*-
+* Updated the FSF directory.
+
* Add more tests. Even basic is very minimal.
* udiv-qrnbd.o should get build as *.lo [HPUX]
* Allow operation using RSA keys consisting of the OpenSSL list of
parameters and allow for a third form where the private Exponent
is not given (saves space).
* Add a warning to the manual, to check that libgcrypt actually has
been compiled with thread support when used by a threaded
application.
* write an autoconf test to check whether the linker supports a
version script.
* Make use of the forthcoming visibility attribute.
* Add attributes to the MPI functions.
* Write tests for the progress function
* In case the ac interface will be more popular than the pk interface,
the asymmetric ciphers could be changed for convenient interaction
with the ac interface (i.e. by using ac's `data sets') and the pk
interface could be changed to be a wrapper for the ac interface.
* HMAC won't work with sha-512 due to the different block size. OTOH,
I can imagine no cryptographic reason to use it.
* cipher/pubkey.c and pubkey implementaions.
Don't rely on the secure memory based wiping function but add an
extra wiping.
diff --git a/compile b/compile
index 3d217032..1b1d2321 100755
--- a/compile
+++ b/compile
@@ -1,142 +1,142 @@
#! /bin/sh
# Wrapper for compilers which do not understand `-c -o'.
-scriptversion=2004-10-12.08
+scriptversion=2005-05-14.22
-# Copyright (C) 1999, 2000, 2003, 2004 Free Software Foundation, Inc.
+# Copyright (C) 1999, 2000, 2003, 2004, 2005 Free Software Foundation, Inc.
# Written by Tom Tromey <tromey@cygnus.com>.
#
# 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, write to the Free Software
-# Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
# 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 <bug-automake@gnu.org> or send patches to
# <automake-patches@gnu.org>.
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 <bug-automake@gnu.org>.
EOF
- exit 0
+ exit $?
;;
-v | --v*)
echo "compile $scriptversion"
- exit 0
+ exit $?
;;
esac
ofile=
cfile=
eat=
for arg
do
if test -n "$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 "$cfile" | sed -e 's|^.*/||' -e 's/\.c$/.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
mv "$cofile" "$ofile"
elif test -f "${cofile}bj"; then
mv "${cofile}bj" "$ofile"
fi
rmdir "$lockdir"
exit $ret
# Local Variables:
# mode: shell-script
# sh-indentation: 2
# eval: (add-hook 'write-file-hooks 'time-stamp)
# time-stamp-start: "scriptversion="
# time-stamp-format: "%:y-%02m-%02d.%02H"
# time-stamp-end: "$"
# End:
diff --git a/config.guess b/config.guess
index 77c7cbab..c38553dc 100755
--- a/config.guess
+++ b/config.guess
@@ -1,1441 +1,1497 @@
#! /bin/sh
# Attempt to guess a canonical system name.
# Copyright (C) 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999,
-# 2000, 2001, 2002, 2003, 2004 Free Software Foundation, Inc.
+# 2000, 2001, 2002, 2003, 2004, 2005 Free Software Foundation, Inc.
-timestamp='2004-08-13'
+timestamp='2006-02-23'
# 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 2 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, write to the Free Software
-# Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+# Foundation, Inc., 51 Franklin Street - Fifth Floor, Boston, MA
+# 02110-1301, USA.
#
# 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 Per Bothner <per@bothner.com>.
# Please send patches to <config-patches@gnu.org>. Submit a context
# diff and a properly formatted ChangeLog entry.
#
# This script attempts to guess a canonical system name similar to
# config.sub. If it succeeds, it prints the system name on stdout, and
# exits with 0. Otherwise, it exits with 1.
#
# The plan is that this can be called by configure scripts if you
# don't specify an explicit build system type.
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 <config-patches@gnu.org>."
version="\
GNU config.guess ($timestamp)
Originally written by Per Bothner.
-Copyright (C) 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004
+Copyright (C) 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005
Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE."
help="
Try \`$me --help' for more information."
# Parse command line
while test $# -gt 0 ; do
case $1 in
--time-stamp | --time* | -t )
- echo "$timestamp" ; exit 0 ;;
+ echo "$timestamp" ; exit ;;
--version | -v )
- echo "$version" ; exit 0 ;;
+ echo "$version" ; exit ;;
--help | --h* | -h )
- echo "$usage"; exit 0 ;;
+ 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 ;
: ${TMPDIR=/tmp} ;
- { tmp=`(umask 077 && mktemp -d -q "$TMPDIR/cgXXXXXX") 2>/dev/null` && test -n "$tmp" && test -d "$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=$HOST_CC ;;
-esac ;'
+esac ; set_cc_for_build= ;'
# This is needed to find uname on a Pyramid OSx when run in the BSD universe.
# (ghazi@noc.rutgers.edu 1994-08-24)
if (test -f /.attbin/uname) >/dev/null 2>&1 ; then
PATH=$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
# 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 tupples: *-*-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 "${UNAME_MACHINE_ARCH}" in
armeb) machine=armeb-unknown ;;
arm*) machine=arm-unknown ;;
sh3el) machine=shl-unknown ;;
sh3eb) machine=sh-unknown ;;
*) machine=${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__ | $CC_FOR_BUILD -E - 2>/dev/null \
| grep __ELF__ >/dev/null
then
# Once all utilities can be ECOFF (netbsdecoff) or a.out (netbsdaout).
# Return netbsd for either. FIX?
os=netbsd
else
os=netbsdelf
fi
;;
*)
os=netbsd
;;
esac
# The OS release
# Debian GNU/NetBSD machines have a different userland, and
# thus, need a distinct triplet. However, they do not need
# kernel version information, so it can be replaced with a
# suitable tag, in the style of linux-gnu.
case "${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 0 ;;
- amd64:OpenBSD:*:*)
- echo x86_64-unknown-openbsd${UNAME_RELEASE}
- exit 0 ;;
- amiga:OpenBSD:*:*)
- echo m68k-unknown-openbsd${UNAME_RELEASE}
- exit 0 ;;
- cats:OpenBSD:*:*)
- echo arm-unknown-openbsd${UNAME_RELEASE}
- exit 0 ;;
- hp300:OpenBSD:*:*)
- echo m68k-unknown-openbsd${UNAME_RELEASE}
- exit 0 ;;
- luna88k:OpenBSD:*:*)
- echo m88k-unknown-openbsd${UNAME_RELEASE}
- exit 0 ;;
- mac68k:OpenBSD:*:*)
- echo m68k-unknown-openbsd${UNAME_RELEASE}
- exit 0 ;;
- macppc:OpenBSD:*:*)
- echo powerpc-unknown-openbsd${UNAME_RELEASE}
- exit 0 ;;
- mvme68k:OpenBSD:*:*)
- echo m68k-unknown-openbsd${UNAME_RELEASE}
- exit 0 ;;
- mvme88k:OpenBSD:*:*)
- echo m88k-unknown-openbsd${UNAME_RELEASE}
- exit 0 ;;
- mvmeppc:OpenBSD:*:*)
- echo powerpc-unknown-openbsd${UNAME_RELEASE}
- exit 0 ;;
- sgi:OpenBSD:*:*)
- echo mips64-unknown-openbsd${UNAME_RELEASE}
- exit 0 ;;
- sun3:OpenBSD:*:*)
- echo m68k-unknown-openbsd${UNAME_RELEASE}
- exit 0 ;;
+ exit ;;
*:OpenBSD:*:*)
- echo ${UNAME_MACHINE}-unknown-openbsd${UNAME_RELEASE}
- exit 0 ;;
+ 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 0 ;;
+ exit ;;
+ *:SolidBSD:*:*)
+ echo ${UNAME_MACHINE}-unknown-solidbsd${UNAME_RELEASE}
+ exit ;;
macppc:MirBSD:*:*)
echo powerppc-unknown-mirbsd${UNAME_RELEASE}
- exit 0 ;;
+ exit ;;
*:MirBSD:*:*)
echo ${UNAME_MACHINE}-unknown-mirbsd${UNAME_RELEASE}
- exit 0 ;;
+ 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'`
- exit 0 ;;
+ exit ;;
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 0 ;;
+ exit ;;
21064:Windows_NT:50:3)
echo alpha-dec-winnt3.5
- exit 0 ;;
+ exit ;;
Amiga*:UNIX_System_V:4.0:*)
echo m68k-unknown-sysv4
- exit 0;;
+ exit ;;
*:[Aa]miga[Oo][Ss]:*:*)
echo ${UNAME_MACHINE}-unknown-amigaos
- exit 0 ;;
+ exit ;;
*:[Mm]orph[Oo][Ss]:*:*)
echo ${UNAME_MACHINE}-unknown-morphos
- exit 0 ;;
+ exit ;;
*:OS/390:*:*)
echo i370-ibm-openedition
- exit 0 ;;
+ exit ;;
+ *:z/VM:*:*)
+ echo s390-ibm-zvmoe
+ exit ;;
*:OS400:*:*)
echo powerpc-ibm-os400
- exit 0 ;;
+ exit ;;
arm:RISC*:1.[012]*:*|arm:riscix:1.[012]*:*)
echo arm-acorn-riscix${UNAME_RELEASE}
- exit 0;;
+ 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 0;;
+ 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 0 ;;
+ exit ;;
NILE*:*:*:dcosx)
echo pyramid-pyramid-svr4
- exit 0 ;;
+ exit ;;
DRS?6000:unix:4.0:6*)
echo sparc-icl-nx6
- exit 0 ;;
- DRS?6000:UNIX_SV:4.2*:7*)
+ 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 0 ;;
+ sparc) echo sparc-icl-nx7; exit ;;
esac ;;
sun4H:SunOS:5.*:*)
echo sparc-hal-solaris2`echo ${UNAME_RELEASE}|sed -e 's/[^.]*//'`
- exit 0 ;;
+ exit ;;
sun4*:SunOS:5.*:* | tadpole*:SunOS:5.*:*)
echo sparc-sun-solaris2`echo ${UNAME_RELEASE}|sed -e 's/[^.]*//'`
- exit 0 ;;
+ exit ;;
i86pc:SunOS:5.*:*)
echo i386-pc-solaris2`echo ${UNAME_RELEASE}|sed -e 's/[^.]*//'`
- exit 0 ;;
+ 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 0 ;;
+ 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 ${UNAME_RELEASE}|sed -e 's/-/_/'`
- exit 0 ;;
+ exit ;;
sun3*:SunOS:*:*)
echo m68k-sun-sunos${UNAME_RELEASE}
- exit 0 ;;
+ 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 0 ;;
+ exit ;;
aushp:SunOS:*:*)
echo sparc-auspex-sunos${UNAME_RELEASE}
- exit 0 ;;
+ exit ;;
# The situation for MiNT is a little confusing. The machine name
# can be virtually everything (everything which is not
# "atarist" or "atariste" at least should have a processor
# > m68000). The system name ranges from "MiNT" over "FreeMiNT"
# to the lowercase version "mint" (or "freemint"). Finally
# the system name "TOS" denotes a system which is actually not
# MiNT. But MiNT is downward compatible to TOS, so this should
# be no problem.
atarist[e]:*MiNT:*:* | atarist[e]:*mint:*:* | atarist[e]:*TOS:*:*)
echo m68k-atari-mint${UNAME_RELEASE}
- exit 0 ;;
+ exit ;;
atari*:*MiNT:*:* | atari*:*mint:*:* | atarist[e]:*TOS:*:*)
echo m68k-atari-mint${UNAME_RELEASE}
- exit 0 ;;
+ exit ;;
*falcon*:*MiNT:*:* | *falcon*:*mint:*:* | *falcon*:*TOS:*:*)
echo m68k-atari-mint${UNAME_RELEASE}
- exit 0 ;;
+ exit ;;
milan*:*MiNT:*:* | milan*:*mint:*:* | *milan*:*TOS:*:*)
echo m68k-milan-mint${UNAME_RELEASE}
- exit 0 ;;
+ exit ;;
hades*:*MiNT:*:* | hades*:*mint:*:* | *hades*:*TOS:*:*)
echo m68k-hades-mint${UNAME_RELEASE}
- exit 0 ;;
+ exit ;;
*:*MiNT:*:* | *:*mint:*:* | *:*TOS:*:*)
echo m68k-unknown-mint${UNAME_RELEASE}
- exit 0 ;;
+ exit ;;
m68k:machten:*:*)
echo m68k-apple-machten${UNAME_RELEASE}
- exit 0 ;;
+ exit ;;
powerpc:machten:*:*)
echo powerpc-apple-machten${UNAME_RELEASE}
- exit 0 ;;
+ exit ;;
RISC*:Mach:*:*)
echo mips-dec-mach_bsd4.3
- exit 0 ;;
+ exit ;;
RISC*:ULTRIX:*:*)
echo mips-dec-ultrix${UNAME_RELEASE}
- exit 0 ;;
+ exit ;;
VAX*:ULTRIX*:*:*)
echo vax-dec-ultrix${UNAME_RELEASE}
- exit 0 ;;
+ exit ;;
2020:CLIX:*:* | 2430:CLIX:*:*)
echo clipper-intergraph-clix${UNAME_RELEASE}
- exit 0 ;;
+ exit ;;
mips:*:*:UMIPS | mips:*:*:RISCos)
eval $set_cc_for_build
sed 's/^ //' << EOF >$dummy.c
#ifdef __cplusplus
#include <stdio.h> /* 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 \
- && $dummy `echo "${UNAME_RELEASE}" | sed -n 's/\([0-9]*\).*/\1/p'` \
- && exit 0
+ $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 0 ;;
+ exit ;;
Motorola:PowerMAX_OS:*:*)
echo powerpc-motorola-powermax
- exit 0 ;;
+ exit ;;
Motorola:*:4.3:PL8-*)
echo powerpc-harris-powermax
- exit 0 ;;
+ exit ;;
Night_Hawk:*:*:PowerMAX_OS | Synergy:PowerMAX_OS:*:*)
echo powerpc-harris-powermax
- exit 0 ;;
+ exit ;;
Night_Hawk:Power_UNIX:*:*)
echo powerpc-harris-powerunix
- exit 0 ;;
+ exit ;;
m88k:CX/UX:7*:*)
echo m88k-harris-cxux7
- exit 0 ;;
+ exit ;;
m88k:*:4*:R4*)
echo m88k-motorola-sysv4
- exit 0 ;;
+ exit ;;
m88k:*:3*:R3*)
echo m88k-motorola-sysv3
- exit 0 ;;
+ 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 0 ;;
+ exit ;;
M88*:DolphinOS:*:*) # DolphinOS (SVR3)
echo m88k-dolphin-sysv3
- exit 0 ;;
+ exit ;;
M88*:*:R3*:*)
# Delta 88k system running SVR3
echo m88k-motorola-sysv3
- exit 0 ;;
+ exit ;;
XD88*:*:*:*) # Tektronix XD88 system running UTekV (SVR3)
echo m88k-tektronix-sysv3
- exit 0 ;;
+ exit ;;
Tek43[0-9][0-9]:UTek:*:*) # Tektronix 4300 system running UTek (BSD)
echo m68k-tektronix-bsd
- exit 0 ;;
+ exit ;;
*:IRIX*:*:*)
echo mips-sgi-irix`echo ${UNAME_RELEASE}|sed -e 's/-/_/g'`
- exit 0 ;;
+ exit ;;
????????:AIX?:[12].1:2) # AIX 2.2.1 or AIX 2.1.1 is RT/PC AIX.
- echo romp-ibm-aix # uname -m gives an 8 hex-code CPU id
- exit 0 ;; # Note that: echo "'`uname -s`'" gives 'AIX '
+ echo romp-ibm-aix # uname -m gives an 8 hex-code CPU id
+ exit ;; # Note that: echo "'`uname -s`'" gives 'AIX '
i*86:AIX:*:*)
echo i386-ibm-aix
- exit 0 ;;
+ exit ;;
ia64:AIX:*:*)
if [ -x /usr/bin/oslevel ] ; then
IBM_REV=`/usr/bin/oslevel`
else
IBM_REV=${UNAME_VERSION}.${UNAME_RELEASE}
fi
echo ${UNAME_MACHINE}-ibm-aix${IBM_REV}
- exit 0 ;;
+ 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 <sys/systemcfg.h>
main()
{
if (!__power_pc())
exit(1);
puts("powerpc-ibm-aix3.2.5");
exit(0);
}
EOF
- $CC_FOR_BUILD -o $dummy $dummy.c && $dummy && exit 0
- echo rs6000-ibm-aix3.2.5
+ 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 0 ;;
+ exit ;;
*:AIX:*:[45])
IBM_CPU_ID=`/usr/sbin/lsdev -C -c processor -S available | sed 1q | awk '{ print $1 }'`
if /usr/sbin/lsattr -El ${IBM_CPU_ID} | grep ' POWER' >/dev/null 2>&1; then
IBM_ARCH=rs6000
else
IBM_ARCH=powerpc
fi
if [ -x /usr/bin/oslevel ] ; then
IBM_REV=`/usr/bin/oslevel`
else
IBM_REV=${UNAME_VERSION}.${UNAME_RELEASE}
fi
echo ${IBM_ARCH}-ibm-aix${IBM_REV}
- exit 0 ;;
+ exit ;;
*:AIX:*:*)
echo rs6000-ibm-aix
- exit 0 ;;
+ exit ;;
ibmrt:4.4BSD:*|romp-ibm:BSD:*)
echo romp-ibm-bsd4.4
- exit 0 ;;
+ exit ;;
ibmrt:*BSD:*|romp-ibm:BSD:*) # covers RT/PC BSD and
echo romp-ibm-bsd${UNAME_RELEASE} # 4.3 with uname added to
- exit 0 ;; # report: romp-ibm BSD 4.3
+ exit ;; # report: romp-ibm BSD 4.3
*:BOSX:*:*)
echo rs6000-bull-bosx
- exit 0 ;;
+ exit ;;
DPX/2?00:B.O.S.:*:*)
echo m68k-bull-sysv3
- exit 0 ;;
+ exit ;;
9000/[34]??:4.3bsd:1.*:*)
echo m68k-hp-bsd
- exit 0 ;;
+ exit ;;
hp300:4.4BSD:*:* | 9000/[34]??:4.3bsd:2.*:*)
echo m68k-hp-bsd4.4
- exit 0 ;;
+ 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 <stdlib.h>
#include <unistd.h>
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
- # avoid double evaluation of $set_cc_for_build
- test -n "$CC_FOR_BUILD" || eval $set_cc_for_build
- if echo __LP64__ | (CCOPTS= $CC_FOR_BUILD -E -) | grep __LP64__ >/dev/null
+ 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 __LP64__ >/dev/null
then
HP_ARCH="hppa2.0w"
else
HP_ARCH="hppa64"
fi
fi
echo ${HP_ARCH}-hp-hpux${HPUX_REV}
- exit 0 ;;
+ exit ;;
ia64:HP-UX:*:*)
HPUX_REV=`echo ${UNAME_RELEASE}|sed -e 's/[^.]*.[0B]*//'`
echo ia64-hp-hpux${HPUX_REV}
- exit 0 ;;
+ exit ;;
3050*:HI-UX:*:*)
eval $set_cc_for_build
sed 's/^ //' << EOF >$dummy.c
#include <unistd.h>
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 && $dummy && exit 0
+ $CC_FOR_BUILD -o $dummy $dummy.c && SYSTEM_NAME=`$dummy` &&
+ { echo "$SYSTEM_NAME"; exit; }
echo unknown-hitachi-hiuxwe2
- exit 0 ;;
+ exit ;;
9000/7??:4.3bsd:*:* | 9000/8?[79]:4.3bsd:*:* )
echo hppa1.1-hp-bsd
- exit 0 ;;
+ exit ;;
9000/8??:4.3bsd:*:*)
echo hppa1.0-hp-bsd
- exit 0 ;;
+ exit ;;
*9??*:MPE/iX:*:* | *3000*:MPE/iX:*:*)
echo hppa1.0-hp-mpeix
- exit 0 ;;
+ exit ;;
hp7??:OSF1:*:* | hp8?[79]:OSF1:*:* )
echo hppa1.1-hp-osf
- exit 0 ;;
+ exit ;;
hp8??:OSF1:*:*)
echo hppa1.0-hp-osf
- exit 0 ;;
+ exit ;;
i*86:OSF1:*:*)
if [ -x /usr/sbin/sysversion ] ; then
echo ${UNAME_MACHINE}-unknown-osf1mk
else
echo ${UNAME_MACHINE}-unknown-osf1
fi
- exit 0 ;;
+ exit ;;
parisc*:Lites*:*:*)
echo hppa1.1-hp-lites
- exit 0 ;;
+ exit ;;
C1*:ConvexOS:*:* | convex:ConvexOS:C1*:*)
echo c1-convex-bsd
- exit 0 ;;
+ exit ;;
C2*:ConvexOS:*:* | convex:ConvexOS:C2*:*)
if getsysinfo -f scalar_acc
then echo c32-convex-bsd
else echo c2-convex-bsd
fi
- exit 0 ;;
+ exit ;;
C34*:ConvexOS:*:* | convex:ConvexOS:C34*:*)
echo c34-convex-bsd
- exit 0 ;;
+ exit ;;
C38*:ConvexOS:*:* | convex:ConvexOS:C38*:*)
echo c38-convex-bsd
- exit 0 ;;
+ exit ;;
C4*:ConvexOS:*:* | convex:ConvexOS:C4*:*)
echo c4-convex-bsd
- exit 0 ;;
+ exit ;;
CRAY*Y-MP:*:*:*)
echo ymp-cray-unicos${UNAME_RELEASE} | sed -e 's/\.[^.]*$/.X/'
- exit 0 ;;
+ 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/\.[^.]*$/.X/'
- exit 0 ;;
+ exit ;;
CRAY*TS:*:*:*)
echo t90-cray-unicos${UNAME_RELEASE} | sed -e 's/\.[^.]*$/.X/'
- exit 0 ;;
+ exit ;;
CRAY*T3E:*:*:*)
echo alphaev5-cray-unicosmk${UNAME_RELEASE} | sed -e 's/\.[^.]*$/.X/'
- exit 0 ;;
+ exit ;;
CRAY*SV1:*:*:*)
echo sv1-cray-unicos${UNAME_RELEASE} | sed -e 's/\.[^.]*$/.X/'
- exit 0 ;;
+ exit ;;
*:UNICOS/mp:*:*)
echo craynv-cray-unicosmp${UNAME_RELEASE} | sed -e 's/\.[^.]*$/.X/'
- exit 0 ;;
+ 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 0 ;;
+ 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 0 ;;
+ exit ;;
i*86:BSD/386:*:* | i*86:BSD/OS:*:* | *:Ascend\ Embedded/OS:*:*)
echo ${UNAME_MACHINE}-pc-bsdi${UNAME_RELEASE}
- exit 0 ;;
+ exit ;;
sparc*:BSD/OS:*:*)
echo sparc-unknown-bsdi${UNAME_RELEASE}
- exit 0 ;;
+ exit ;;
*:BSD/OS:*:*)
echo ${UNAME_MACHINE}-unknown-bsdi${UNAME_RELEASE}
- exit 0 ;;
+ exit ;;
*:FreeBSD:*:*)
- echo ${UNAME_MACHINE}-unknown-freebsd`echo ${UNAME_RELEASE}|sed -e 's/[-(].*//'`
- exit 0 ;;
+ case ${UNAME_MACHINE} in
+ pc98)
+ echo i386-unknown-freebsd`echo ${UNAME_RELEASE}|sed -e 's/[-(].*//'` ;;
+ *)
+ echo ${UNAME_MACHINE}-unknown-freebsd`echo ${UNAME_RELEASE}|sed -e 's/[-(].*//'` ;;
+ esac
+ exit ;;
i*:CYGWIN*:*)
echo ${UNAME_MACHINE}-pc-cygwin
- exit 0 ;;
+ exit ;;
i*:MINGW*:*)
echo ${UNAME_MACHINE}-pc-mingw32
- exit 0 ;;
+ exit ;;
+ i*:MSYS_NT-*:*:*)
+ echo ${UNAME_MACHINE}-pc-mingw32
+ exit ;;
+ i*:windows32*:*)
+ # uname -m includes "-pc" on this system.
+ echo ${UNAME_MACHINE}-mingw32
+ exit ;;
i*:PW*:*)
echo ${UNAME_MACHINE}-pc-pw32
- exit 0 ;;
- x86:Interix*:[34]*)
- echo i586-pc-interix${UNAME_RELEASE}|sed -e 's/\..*//'
- exit 0 ;;
+ exit ;;
+ x86:Interix*:[345]*)
+ echo i586-pc-interix${UNAME_RELEASE}
+ exit ;;
+ EM64T:Interix*:[345]*)
+ echo x86_64-unknown-interix${UNAME_RELEASE}
+ exit ;;
[345]86:Windows_95:* | [345]86:Windows_98:* | [345]86:Windows_NT:*)
echo i${UNAME_MACHINE}-pc-mks
- exit 0 ;;
+ 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 0 ;;
+ exit ;;
i*:UWIN*:*)
echo ${UNAME_MACHINE}-pc-uwin
- exit 0 ;;
+ exit ;;
+ amd64:CYGWIN*:*:* | x86_64:CYGWIN*:*:*)
+ echo x86_64-unknown-cygwin
+ exit ;;
p*:CYGWIN*:*)
echo powerpcle-unknown-cygwin
- exit 0 ;;
+ exit ;;
prep*:SunOS:5.*:*)
echo powerpcle-unknown-solaris2`echo ${UNAME_RELEASE}|sed -e 's/[^.]*//'`
- exit 0 ;;
+ exit ;;
*:GNU:*:*)
# the GNU system
echo `echo ${UNAME_MACHINE}|sed -e 's,[-/].*$,,'`-unknown-gnu`echo ${UNAME_RELEASE}|sed -e 's,/.*$,,'`
- exit 0 ;;
+ 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/[-(].*//'`-gnu
- exit 0 ;;
+ exit ;;
i*86:Minix:*:*)
echo ${UNAME_MACHINE}-pc-minix
- exit 0 ;;
+ exit ;;
arm*:Linux:*:*)
echo ${UNAME_MACHINE}-unknown-linux-gnu
- exit 0 ;;
+ exit ;;
cris:Linux:*:*)
echo cris-axis-linux-gnu
- exit 0 ;;
+ exit ;;
+ crisv32:Linux:*:*)
+ echo crisv32-axis-linux-gnu
+ exit ;;
+ frv:Linux:*:*)
+ echo frv-unknown-linux-gnu
+ exit ;;
ia64:Linux:*:*)
echo ${UNAME_MACHINE}-unknown-linux-gnu
- exit 0 ;;
+ exit ;;
m32r*:Linux:*:*)
echo ${UNAME_MACHINE}-unknown-linux-gnu
- exit 0 ;;
+ exit ;;
m68*:Linux:*:*)
echo ${UNAME_MACHINE}-unknown-linux-gnu
- exit 0 ;;
+ exit ;;
mips:Linux:*:*)
eval $set_cc_for_build
sed 's/^ //' << EOF >$dummy.c
#undef CPU
#undef mips
#undef mipsel
#if defined(__MIPSEL__) || defined(__MIPSEL) || defined(_MIPSEL) || defined(MIPSEL)
CPU=mipsel
#else
#if defined(__MIPSEB__) || defined(__MIPSEB) || defined(_MIPSEB) || defined(MIPSEB)
CPU=mips
#else
CPU=
#endif
#endif
EOF
- eval `$CC_FOR_BUILD -E $dummy.c 2>/dev/null | grep ^CPU=`
- test x"${CPU}" != x && echo "${CPU}-unknown-linux-gnu" && exit 0
+ eval "`$CC_FOR_BUILD -E $dummy.c 2>/dev/null | sed -n '
+ /^CPU/{
+ s: ::g
+ p
+ }'`"
+ test x"${CPU}" != x && { echo "${CPU}-unknown-linux-gnu"; exit; }
;;
mips64:Linux:*:*)
eval $set_cc_for_build
sed 's/^ //' << EOF >$dummy.c
#undef CPU
#undef mips64
#undef mips64el
#if defined(__MIPSEL__) || defined(__MIPSEL) || defined(_MIPSEL) || defined(MIPSEL)
CPU=mips64el
#else
#if defined(__MIPSEB__) || defined(__MIPSEB) || defined(_MIPSEB) || defined(MIPSEB)
CPU=mips64
#else
CPU=
#endif
#endif
EOF
- eval `$CC_FOR_BUILD -E $dummy.c 2>/dev/null | grep ^CPU=`
- test x"${CPU}" != x && echo "${CPU}-unknown-linux-gnu" && exit 0
+ eval "`$CC_FOR_BUILD -E $dummy.c 2>/dev/null | sed -n '
+ /^CPU/{
+ s: ::g
+ p
+ }'`"
+ test x"${CPU}" != x && { echo "${CPU}-unknown-linux-gnu"; exit; }
;;
+ or32:Linux:*:*)
+ echo or32-unknown-linux-gnu
+ exit ;;
ppc:Linux:*:*)
echo powerpc-unknown-linux-gnu
- exit 0 ;;
+ exit ;;
ppc64:Linux:*:*)
echo powerpc64-unknown-linux-gnu
- exit 0 ;;
+ 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 ld.so.1 >/dev/null
if test "$?" = 0 ; then LIBC="libc1" ; else LIBC="" ; fi
echo ${UNAME_MACHINE}-unknown-linux-gnu${LIBC}
- exit 0 ;;
+ 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-gnu ;;
PA8*) echo hppa2.0-unknown-linux-gnu ;;
*) echo hppa-unknown-linux-gnu ;;
esac
- exit 0 ;;
+ exit ;;
parisc64:Linux:*:* | hppa64:Linux:*:*)
echo hppa64-unknown-linux-gnu
- exit 0 ;;
+ exit ;;
s390:Linux:*:* | s390x:Linux:*:*)
echo ${UNAME_MACHINE}-ibm-linux
- exit 0 ;;
+ exit ;;
sh64*:Linux:*:*)
echo ${UNAME_MACHINE}-unknown-linux-gnu
- exit 0 ;;
+ exit ;;
sh*:Linux:*:*)
echo ${UNAME_MACHINE}-unknown-linux-gnu
- exit 0 ;;
+ exit ;;
sparc:Linux:*:* | sparc64:Linux:*:*)
echo ${UNAME_MACHINE}-unknown-linux-gnu
- exit 0 ;;
+ exit ;;
+ vax:Linux:*:*)
+ echo ${UNAME_MACHINE}-dec-linux-gnu
+ exit ;;
x86_64:Linux:*:*)
echo x86_64-unknown-linux-gnu
- exit 0 ;;
+ exit ;;
i*86:Linux:*:*)
# The BFD linker knows what the default object file format is, so
# first see if it will tell us. cd to the root directory to prevent
# problems with other programs or directories called `ld' in the path.
# Set LC_ALL=C to ensure ld outputs messages in English.
ld_supported_targets=`cd /; LC_ALL=C ld --help 2>&1 \
| sed -ne '/supported targets:/!d
s/[ ][ ]*/ /g
s/.*supported targets: *//
s/ .*//
p'`
case "$ld_supported_targets" in
elf32-i386)
TENTATIVE="${UNAME_MACHINE}-pc-linux-gnu"
;;
a.out-i386-linux)
echo "${UNAME_MACHINE}-pc-linux-gnuaout"
- exit 0 ;;
+ exit ;;
coff-i386)
echo "${UNAME_MACHINE}-pc-linux-gnucoff"
- exit 0 ;;
+ exit ;;
"")
# Either a pre-BFD a.out linker (linux-gnuoldld) or
# one that does not give us useful --help.
echo "${UNAME_MACHINE}-pc-linux-gnuoldld"
- exit 0 ;;
+ exit ;;
esac
# Determine whether the default compiler is a.out or elf
eval $set_cc_for_build
sed 's/^ //' << EOF >$dummy.c
#include <features.h>
#ifdef __ELF__
# ifdef __GLIBC__
# if __GLIBC__ >= 2
LIBC=gnu
# else
LIBC=gnulibc1
# endif
# else
LIBC=gnulibc1
# endif
#else
- #ifdef __INTEL_COMPILER
+ #if defined(__INTEL_COMPILER) || defined(__PGI) || defined(__sun)
LIBC=gnu
#else
LIBC=gnuaout
#endif
#endif
#ifdef __dietlibc__
LIBC=dietlibc
#endif
EOF
- eval `$CC_FOR_BUILD -E $dummy.c 2>/dev/null | grep ^LIBC=`
- test x"${LIBC}" != x && echo "${UNAME_MACHINE}-pc-linux-${LIBC}" && exit 0
- test x"${TENTATIVE}" != x && echo "${TENTATIVE}" && exit 0
+ eval "`$CC_FOR_BUILD -E $dummy.c 2>/dev/null | sed -n '
+ /^LIBC/{
+ s: ::g
+ p
+ }'`"
+ test x"${LIBC}" != x && {
+ echo "${UNAME_MACHINE}-pc-linux-${LIBC}"
+ exit
+ }
+ test x"${TENTATIVE}" != x && { echo "${TENTATIVE}"; 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 0 ;;
+ 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 0 ;;
+ 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 0 ;;
+ exit ;;
i*86:XTS-300:*:STOP)
echo ${UNAME_MACHINE}-unknown-stop
- exit 0 ;;
+ exit ;;
i*86:atheos:*:*)
echo ${UNAME_MACHINE}-unknown-atheos
- exit 0 ;;
- i*86:syllable:*:*)
+ exit ;;
+ i*86:syllable:*:*)
echo ${UNAME_MACHINE}-pc-syllable
- exit 0 ;;
+ exit ;;
i*86:LynxOS:2.*:* | i*86:LynxOS:3.[01]*:* | i*86:LynxOS:4.0*:*)
echo i386-unknown-lynxos${UNAME_RELEASE}
- exit 0 ;;
+ exit ;;
i*86:*DOS:*:*)
echo ${UNAME_MACHINE}-pc-msdosdjgpp
- exit 0 ;;
+ 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 0 ;;
- i*86:*:5:[78]*)
+ 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 0 ;;
+ exit ;;
i*86:*:3.2:*)
if test -f /usr/options/cb.name; then
UNAME_REL=`sed -n 's/.*Version //p' </usr/options/cb.name`
echo ${UNAME_MACHINE}-pc-isc$UNAME_REL
elif /bin/uname -X 2>/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 ${UNAME_MACHINE}-pc-sysv32
fi
- exit 0 ;;
+ exit ;;
pc:*:*:*)
# Left here for compatibility:
# uname -m prints for DJGPP always 'pc', but it prints nothing about
# the processor, so we play safe by assuming i386.
echo i386-pc-msdosdjgpp
- exit 0 ;;
+ exit ;;
Intel:Mach:3*:*)
echo i386-pc-mach3
- exit 0 ;;
+ exit ;;
paragon:*:*:*)
echo i860-intel-osf1
- exit 0 ;;
+ exit ;;
i860:*:4.*:*) # i860-SVR4
if grep Stardent /usr/include/sys/uadmin.h >/dev/null 2>&1 ; then
echo i860-stardent-sysv${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 0 ;;
+ exit ;;
mini*:CTIX:SYS*5:*)
# "miniframe"
echo m68010-convergent-sysv
- exit 0 ;;
+ exit ;;
mc68k:UNIX:SYSTEM5:3.51m)
echo m68k-convergent-sysv
- exit 0 ;;
+ exit ;;
M680?0:D-NIX:5.3:*)
echo m68k-diab-dnix
- exit 0 ;;
+ exit ;;
M68*:*:R3V[5678]*:*)
- test -r /sysV68 && echo 'm68k-motorola-sysv' && exit 0 ;;
+ 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${OS_REL} && exit 0
+ && { echo i486-ncr-sysv4.3${OS_REL}; exit; }
/bin/uname -p 2>/dev/null | /bin/grep entium >/dev/null \
- && echo i586-ncr-sysv4.3${OS_REL} && exit 0 ;;
+ && { echo i586-ncr-sysv4.3${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 0 ;;
+ && { echo i486-ncr-sysv4; exit; } ;;
m68*:LynxOS:2.*:* | m68*:LynxOS:3.0*:*)
echo m68k-unknown-lynxos${UNAME_RELEASE}
- exit 0 ;;
+ exit ;;
mc68030:UNIX_System_V:4.*:*)
echo m68k-atari-sysv4
- exit 0 ;;
+ exit ;;
TSUNAMI:LynxOS:2.*:*)
echo sparc-unknown-lynxos${UNAME_RELEASE}
- exit 0 ;;
+ exit ;;
rs6000:LynxOS:2.*:*)
echo rs6000-unknown-lynxos${UNAME_RELEASE}
- exit 0 ;;
+ exit ;;
PowerPC:LynxOS:2.*:* | PowerPC:LynxOS:3.[01]*:* | PowerPC:LynxOS:4.0*:*)
echo powerpc-unknown-lynxos${UNAME_RELEASE}
- exit 0 ;;
+ exit ;;
SM[BE]S:UNIX_SV:*:*)
echo mips-dde-sysv${UNAME_RELEASE}
- exit 0 ;;
+ exit ;;
RM*:ReliantUNIX-*:*:*)
echo mips-sni-sysv4
- exit 0 ;;
+ exit ;;
RM*:SINIX-*:*:*)
echo mips-sni-sysv4
- exit 0 ;;
+ 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 0 ;;
+ exit ;;
PENTIUM:*:4.0*:*) # Unisys `ClearPath HMP IX 4000' SVR4/MP effort
# says <Richard.M.Bartel@ccMail.Census.GOV>
echo i586-unisys-sysv4
- exit 0 ;;
+ exit ;;
*:UNIX_System_V:4*:FTX*)
# From Gerald Hewes <hewes@openmarket.com>.
# How about differentiating between stratus architectures? -djm
echo hppa1.1-stratus-sysv4
- exit 0 ;;
+ exit ;;
*:*:*:FTX*)
# From seanf@swdc.stratus.com.
echo i860-stratus-sysv4
- exit 0 ;;
+ exit ;;
+ i*86:VOS:*:*)
+ # From Paul.Green@stratus.com.
+ echo ${UNAME_MACHINE}-stratus-vos
+ exit ;;
*:VOS:*:*)
# From Paul.Green@stratus.com.
echo hppa1.1-stratus-vos
- exit 0 ;;
+ exit ;;
mc68*:A/UX:*:*)
echo m68k-apple-aux${UNAME_RELEASE}
- exit 0 ;;
+ exit ;;
news*:NEWS-OS:6*:*)
echo mips-sony-newsos6
- exit 0 ;;
+ 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 0 ;;
+ exit ;;
BeBox:BeOS:*:*) # BeOS running on hardware made by Be, PPC only.
echo powerpc-be-beos
- exit 0 ;;
+ exit ;;
BeMac:BeOS:*:*) # BeOS running on Mac or Mac clone, PPC only.
echo powerpc-apple-beos
- exit 0 ;;
+ exit ;;
BePC:BeOS:*:*) # BeOS running on Intel PC compatible.
echo i586-pc-beos
- exit 0 ;;
+ exit ;;
SX-4:SUPER-UX:*:*)
echo sx4-nec-superux${UNAME_RELEASE}
- exit 0 ;;
+ exit ;;
SX-5:SUPER-UX:*:*)
echo sx5-nec-superux${UNAME_RELEASE}
- exit 0 ;;
+ exit ;;
SX-6:SUPER-UX:*:*)
echo sx6-nec-superux${UNAME_RELEASE}
- exit 0 ;;
+ exit ;;
Power*:Rhapsody:*:*)
echo powerpc-apple-rhapsody${UNAME_RELEASE}
- exit 0 ;;
+ exit ;;
*:Rhapsody:*:*)
echo ${UNAME_MACHINE}-apple-rhapsody${UNAME_RELEASE}
- exit 0 ;;
+ exit ;;
*:Darwin:*:*)
UNAME_PROCESSOR=`uname -p` || UNAME_PROCESSOR=unknown
case $UNAME_PROCESSOR in
- *86) UNAME_PROCESSOR=i686 ;;
unknown) UNAME_PROCESSOR=powerpc ;;
esac
echo ${UNAME_PROCESSOR}-apple-darwin${UNAME_RELEASE}
- exit 0 ;;
+ 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${UNAME_RELEASE}
- exit 0 ;;
+ exit ;;
*:QNX:*:4*)
echo i386-pc-qnx
- exit 0 ;;
+ exit ;;
+ NSE-?:NONSTOP_KERNEL:*:*)
+ echo nse-tandem-nsk${UNAME_RELEASE}
+ exit ;;
NSR-?:NONSTOP_KERNEL:*:*)
echo nsr-tandem-nsk${UNAME_RELEASE}
- exit 0 ;;
+ exit ;;
*:NonStop-UX:*:*)
echo mips-compaq-nonstopux
- exit 0 ;;
+ exit ;;
BS2000:POSIX*:*:*)
echo bs2000-siemens-sysv
- exit 0 ;;
+ exit ;;
DS/*:UNIX_System_V:*:*)
echo ${UNAME_MACHINE}-${UNAME_SYSTEM}-${UNAME_RELEASE}
- exit 0 ;;
+ 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 ${UNAME_MACHINE}-unknown-plan9
- exit 0 ;;
+ exit ;;
*:TOPS-10:*:*)
echo pdp10-unknown-tops10
- exit 0 ;;
+ exit ;;
*:TENEX:*:*)
echo pdp10-unknown-tenex
- exit 0 ;;
+ exit ;;
KS10:TOPS-20:*:* | KL10:TOPS-20:*:* | TYPE4:TOPS-20:*:*)
echo pdp10-dec-tops20
- exit 0 ;;
+ exit ;;
XKL-1:TOPS-20:*:* | TYPE5:TOPS-20:*:*)
echo pdp10-xkl-tops20
- exit 0 ;;
+ exit ;;
*:TOPS-20:*:*)
echo pdp10-unknown-tops20
- exit 0 ;;
+ exit ;;
*:ITS:*:*)
echo pdp10-unknown-its
- exit 0 ;;
+ exit ;;
SEI:*:*:SEIUX)
echo mips-sei-seiux${UNAME_RELEASE}
- exit 0 ;;
+ exit ;;
*:DragonFly:*:*)
echo ${UNAME_MACHINE}-unknown-dragonfly`echo ${UNAME_RELEASE}|sed -e 's/[-(].*//'`
- exit 0 ;;
+ exit ;;
*:*VMS:*:*)
UNAME_MACHINE=`(uname -p) 2>/dev/null`
case "${UNAME_MACHINE}" in
- A*) echo alpha-dec-vms && exit 0 ;;
- I*) echo ia64-dec-vms && exit 0 ;;
- V*) echo vax-dec-vms && exit 0 ;;
- esac
+ 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 ${UNAME_MACHINE}-pc-rdos
+ exit ;;
esac
#echo '(No uname command or uname output not recognized.)' 1>&2
#echo "${UNAME_MACHINE}:${UNAME_SYSTEM}:${UNAME_RELEASE}:${UNAME_VERSION}" 1>&2
eval $set_cc_for_build
cat >$dummy.c <<EOF
#ifdef _SEQUENT_
# include <sys/types.h>
# include <sys/utsname.h>
#endif
main ()
{
#if defined (sony)
#if defined (MIPSEB)
/* BFD wants "bsd" instead of "newsos". Perhaps BFD should be changed,
I don't know.... */
printf ("mips-sony-bsd\n"); exit (0);
#else
#include <sys/param.h>
printf ("m68k-sony-newsos%s\n",
#ifdef NEWSOS4
"4"
#else
""
#endif
); exit (0);
#endif
#endif
#if defined (__arm) && defined (__acorn) && defined (__unix)
- printf ("arm-acorn-riscix"); exit (0);
+ printf ("arm-acorn-riscix\n"); exit (0);
#endif
#if defined (hp300) && !defined (hpux)
printf ("m68k-hp-bsd\n"); exit (0);
#endif
#if defined (NeXT)
#if !defined (__ARCHITECTURE__)
#define __ARCHITECTURE__ "m68k"
#endif
int version;
version=`(hostinfo | sed -n 's/.*NeXT Mach \([0-9]*\).*/\1/p') 2>/dev/null`;
if (version < 4)
printf ("%s-next-nextstep%d\n", __ARCHITECTURE__, version);
else
printf ("%s-next-openstep%d\n", __ARCHITECTURE__, version);
exit (0);
#endif
#if defined (MULTIMAX) || defined (n16)
#if defined (UMAXV)
printf ("ns32k-encore-sysv\n"); exit (0);
#else
#if defined (CMU)
printf ("ns32k-encore-mach\n"); exit (0);
#else
printf ("ns32k-encore-bsd\n"); exit (0);
#endif
#endif
#endif
#if defined (__386BSD__)
printf ("i386-pc-bsd\n"); exit (0);
#endif
#if defined (sequent)
#if defined (i386)
printf ("i386-sequent-dynix\n"); exit (0);
#endif
#if defined (ns32000)
printf ("ns32k-sequent-dynix\n"); exit (0);
#endif
#endif
#if defined (_SEQUENT_)
struct utsname un;
uname(&un);
if (strncmp(un.version, "V2", 2) == 0) {
printf ("i386-sequent-ptx2\n"); exit (0);
}
if (strncmp(un.version, "V1", 2) == 0) { /* XXX is V1 correct? */
printf ("i386-sequent-ptx1\n"); exit (0);
}
printf ("i386-sequent-ptx\n"); exit (0);
#endif
#if defined (vax)
# if !defined (ultrix)
# include <sys/param.h>
# if defined (BSD)
# if BSD == 43
printf ("vax-dec-bsd4.3\n"); exit (0);
# else
# if BSD == 199006
printf ("vax-dec-bsd4.3reno\n"); exit (0);
# else
printf ("vax-dec-bsd\n"); exit (0);
# endif
# endif
# else
printf ("vax-dec-bsd\n"); exit (0);
# endif
# else
printf ("vax-dec-ultrix\n"); exit (0);
# endif
#endif
#if defined (alliant) && defined (i860)
printf ("i860-alliant-bsd\n"); exit (0);
#endif
exit (1);
}
EOF
-$CC_FOR_BUILD -o $dummy $dummy.c 2>/dev/null && $dummy && exit 0
+$CC_FOR_BUILD -o $dummy $dummy.c 2>/dev/null && SYSTEM_NAME=`$dummy` &&
+ { echo "$SYSTEM_NAME"; exit; }
# Apollos put the system type in the environment.
-test -d /usr/apollo && { echo ${ISP}-apollo-${SYSTYPE}; exit 0; }
+test -d /usr/apollo && { echo ${ISP}-apollo-${SYSTYPE}; exit; }
# Convex versions that predate uname can use getsysinfo(1)
if [ -x /usr/convex/getsysinfo ]
then
case `getsysinfo -f cpu_type` in
c1*)
echo c1-convex-bsd
- exit 0 ;;
+ exit ;;
c2*)
if getsysinfo -f scalar_acc
then echo c32-convex-bsd
else echo c2-convex-bsd
fi
- exit 0 ;;
+ exit ;;
c34*)
echo c34-convex-bsd
- exit 0 ;;
+ exit ;;
c38*)
echo c38-convex-bsd
- exit 0 ;;
+ exit ;;
c4*)
echo c4-convex-bsd
- exit 0 ;;
+ exit ;;
esac
fi
cat >&2 <<EOF
$0: unable to guess system type
This script, last modified $timestamp, has failed to recognize
the operating system you are using. It is advised that you
download the most up to date version of the config scripts from
- ftp://ftp.gnu.org/pub/gnu/config/
+ http://savannah.gnu.org/cgi-bin/viewcvs/*checkout*/config/config/config.guess
+and
+ http://savannah.gnu.org/cgi-bin/viewcvs/*checkout*/config/config/config.sub
If the version you run ($0) is already up to date, please
send the following data and any information you think might be
pertinent to <config-patches@gnu.org> 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/config.sub b/config.sub
index ac6de986..ad9f3957 100755
--- a/config.sub
+++ b/config.sub
@@ -1,1552 +1,1608 @@
#! /bin/sh
# Configuration validation subroutine script.
# Copyright (C) 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999,
-# 2000, 2001, 2002, 2003, 2004 Free Software Foundation, Inc.
+# 2000, 2001, 2002, 2003, 2004, 2005 Free Software Foundation, Inc.
-timestamp='2004-06-24'
+timestamp='2006-02-23'
# This file is (in principle) common to ALL GNU software.
# The presence of a machine in this file suggests that SOME GNU software
# can handle that machine. It does not imply ALL GNU software can.
#
# 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 2 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, write to the Free Software
-# Foundation, Inc., 59 Temple Place - Suite 330,
-# Boston, MA 02111-1307, USA.
-
+# Foundation, Inc., 51 Franklin Street - Fifth Floor, Boston, MA
+# 02110-1301, USA.
+#
# 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.
+
# Please send patches to <config-patches@gnu.org>. Submit a context
# diff and a properly formatted ChangeLog entry.
#
# 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.
# 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
$0 [OPTION] ALIAS
Canonicalize a configuration name.
Operation modes:
-h, --help print this help, then exit
-t, --time-stamp print date of last modification, then exit
-v, --version print version number, then exit
Report bugs and patches to <config-patches@gnu.org>."
version="\
GNU config.sub ($timestamp)
-Copyright (C) 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004
+Copyright (C) 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005
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 0 ;;
+ echo "$timestamp" ; exit ;;
--version | -v )
- echo "$version" ; exit 0 ;;
+ echo "$version" ; exit ;;
--help | --h* | -h )
- echo "$usage"; exit 0 ;;
+ 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 0;;
+ 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 $maybe_os in
- nto-qnx* | linux-gnu* | linux-dietlibc | linux-uclibc* | uclinux-uclibc* | uclinux-gnu* | \
- kfreebsd*-gnu* | knetbsd*-gnu* | netbsd*-gnu* | storm-chaos* | os2-emx* | rtmk-nova*)
+ nto-qnx* | linux-gnu* | linux-dietlibc | linux-newlib* | linux-uclibc* | \
+ uclinux-uclibc* | uclinux-gnu* | kfreebsd*-gnu* | knetbsd*-gnu* | netbsd*-gnu* | \
+ storm-chaos* | os2-emx* | rtmk-nova*)
os=-$maybe_os
basic_machine=`echo $1 | sed 's/^\(.*\)-\([^-]*-[^-]*\)$/\1/'`
;;
*)
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)
os=
basic_machine=$1
;;
-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*)
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 $basic_machine in
# Recognize the basic CPU types without company name.
# Some are omitted here because they have special meanings below.
1750a | 580 \
| a29k \
| alpha | alphaev[4-8] | alphaev56 | alphaev6[78] | alphapca5[67] \
| alpha64 | alpha64ev[4-8] | alpha64ev56 | alpha64ev6[78] | alpha64pca5[67] \
| am33_2.0 \
| arc | arm | arm[bl]e | arme[lb] | armv[2345] | armv[345][lb] | avr \
+ | bfin \
| c4x | clipper \
| d10v | d30v | dlx | dsp16xx \
| fr30 | frv \
| h8300 | h8500 | hppa | hppa1.[01] | hppa2.0 | hppa2.0[nw] | hppa64 \
| i370 | i860 | i960 | ia64 \
| ip2k | iq2000 \
- | m32r | m32rle | m68000 | m68k | m88k | mcore \
+ | m32r | m32rle | m68000 | m68k | m88k | maxq | mb | microblaze | mcore \
| mips | mipsbe | mipseb | mipsel | mipsle \
| mips16 \
| mips64 | mips64el \
| mips64vr | mips64vrel \
| mips64orion | mips64orionel \
| mips64vr4100 | mips64vr4100el \
| mips64vr4300 | mips64vr4300el \
| mips64vr5000 | mips64vr5000el \
+ | mips64vr5900 | mips64vr5900el \
| mipsisa32 | mipsisa32el \
| mipsisa32r2 | mipsisa32r2el \
| mipsisa64 | mipsisa64el \
| mipsisa64r2 | mipsisa64r2el \
| mipsisa64sb1 | mipsisa64sb1el \
| mipsisa64sr71k | mipsisa64sr71kel \
| mipstx39 | mipstx39el \
| mn10200 | mn10300 \
+ | mt \
| msp430 \
+ | nios | nios2 \
| ns16k | ns32k \
- | openrisc | or32 \
+ | or32 \
| pdp10 | pdp11 | pj | pjl \
| powerpc | powerpc64 | powerpc64le | powerpcle | ppcbe \
| pyramid \
- | sh | sh[1234] | sh[23]e | sh[34]eb | shbe | shle | sh[1234]le | sh3ele \
+ | sh | sh[1234] | sh[24]a | sh[23]e | sh[34]eb | shbe | shle | sh[1234]le | sh3ele \
| sh64 | sh64le \
- | sparc | sparc64 | sparc86x | sparclet | sparclite | sparcv8 | sparcv9 | sparcv9b \
+ | sparc | sparc64 | sparc64b | sparc86x | sparclet | sparclite \
+ | sparcv8 | sparcv9 | sparcv9b \
| strongarm \
| tahoe | thumb | tic4x | tic80 | tron \
| v850 | v850e \
| we32k \
- | x86 | xscale | xstormy16 | xtensa \
+ | x86 | xscale | xscalee[bl] | xstormy16 | xtensa \
| z8k)
basic_machine=$basic_machine-unknown
;;
+ m32c)
+ basic_machine=$basic_machine-unknown
+ ;;
m6811 | m68hc11 | m6812 | m68hc12)
# Motorola 68HC11/12.
basic_machine=$basic_machine-unknown
os=-none
;;
m88110 | m680[12346]0 | m683?2 | m68360 | m5200 | v70 | w65 | z8k)
;;
+ ms1)
+ basic_machine=mt-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-* \
| alpha-* | alphaev[4-8]-* | alphaev56-* | alphaev6[78]-* \
| alpha64-* | alpha64ev[4-8]-* | alpha64ev56-* | alpha64ev6[78]-* \
| alphapca5[67]-* | alpha64pca5[67]-* | arc-* \
| arm-* | armbe-* | armle-* | armeb-* | armv*-* \
| avr-* \
- | bs2000-* \
+ | bfin-* | bs2000-* \
| c[123]* | c30-* | [cjt]90-* | c4x-* | c54x-* | c55x-* | c6x-* \
| clipper-* | craynv-* | cydra-* \
| d10v-* | d30v-* | dlx-* \
| elxsi-* \
| f30[01]-* | f700-* | fr30-* | frv-* | fx80-* \
| h8300-* | h8500-* \
| hppa-* | hppa1.[01]-* | hppa2.0-* | hppa2.0[nw]-* | hppa64-* \
| i*86-* | i860-* | i960-* | ia64-* \
| ip2k-* | iq2000-* \
| m32r-* | m32rle-* \
| m68000-* | m680[012346]0-* | m68360-* | m683?2-* | m68k-* \
- | m88110-* | m88k-* | mcore-* \
+ | m88110-* | m88k-* | maxq-* | mcore-* \
| mips-* | mipsbe-* | mipseb-* | mipsel-* | mipsle-* \
| mips16-* \
| mips64-* | mips64el-* \
| mips64vr-* | mips64vrel-* \
| mips64orion-* | mips64orionel-* \
| mips64vr4100-* | mips64vr4100el-* \
| mips64vr4300-* | mips64vr4300el-* \
| mips64vr5000-* | mips64vr5000el-* \
+ | mips64vr5900-* | mips64vr5900el-* \
| mipsisa32-* | mipsisa32el-* \
| mipsisa32r2-* | mipsisa32r2el-* \
| mipsisa64-* | mipsisa64el-* \
| mipsisa64r2-* | mipsisa64r2el-* \
| mipsisa64sb1-* | mipsisa64sb1el-* \
| mipsisa64sr71k-* | mipsisa64sr71kel-* \
| mipstx39-* | mipstx39el-* \
| mmix-* \
+ | mt-* \
| msp430-* \
+ | nios-* | nios2-* \
| none-* | np1-* | ns16k-* | ns32k-* \
| orion-* \
| pdp10-* | pdp11-* | pj-* | pjl-* | pn-* | power-* \
| powerpc-* | powerpc64-* | powerpc64le-* | powerpcle-* | ppcbe-* \
| pyramid-* \
| romp-* | rs6000-* \
- | sh-* | sh[1234]-* | sh[23]e-* | sh[34]eb-* | shbe-* \
+ | sh-* | sh[1234]-* | sh[24]a-* | sh[23]e-* | sh[34]eb-* | shbe-* \
| shle-* | sh[1234]le-* | sh3ele-* | sh64-* | sh64le-* \
- | sparc-* | sparc64-* | sparc86x-* | sparclet-* | sparclite-* \
+ | sparc-* | sparc64-* | sparc64b-* | sparc86x-* | sparclet-* \
+ | sparclite-* \
| sparcv8-* | sparcv9-* | sparcv9b-* | strongarm-* | sv1-* | sx?-* \
| tahoe-* | thumb-* \
| tic30-* | tic4x-* | tic54x-* | tic55x-* | tic6x-* | tic80-* \
| tron-* \
| v850-* | v850e-* | vax-* \
| we32k-* \
- | x86-* | x86_64-* | xps100-* | xscale-* | xstormy16-* \
- | xtensa-* \
+ | x86-* | x86_64-* | xps100-* | xscale-* | xscalee[bl]-* \
+ | xstormy16-* | xtensa-* \
| ymp-* \
| z8k-*)
;;
+ m32c-*)
+ ;;
# 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
;;
aux)
basic_machine=m68k-apple
os=-aux
;;
balance)
basic_machine=ns32k-sequent
os=-dynix
;;
c90)
basic_machine=c90-cray
os=-unicos
;;
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
;;
cr16c)
basic_machine=cr16c-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
;;
+ 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'm not sure what "Sysv32" means. Should this be sysv3.2?
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
;;
m88k-omron*)
basic_machine=m88k-omron
;;
magnum | m3230)
basic_machine=mips-mips
os=-sysv
;;
merlin)
basic_machine=ns32k-utek
os=-sysv
;;
mingw32)
basic_machine=i386-pc
os=-mingw32
;;
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
;;
msdos)
basic_machine=i386-pc
os=-msdos
;;
+ ms1-*)
+ basic_machine=`echo $basic_machine | sed -e 's/ms1-/mt-/'`
+ ;;
mvs)
basic_machine=i370-ibm
os=-mvs
;;
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
;;
nsr-tandem)
basic_machine=nsr-tandem
;;
op50n-* | op60c-*)
basic_machine=hppa1.1-oki
os=-proelf
;;
- or32 | or32-*)
+ openrisc | openrisc-*)
basic_machine=or32-unknown
- os=-coff
;;
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
;;
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) basic_machine=powerpc-unknown
;;
ppc-*) 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)
+ 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
;;
sei)
basic_machine=mips-sei
os=-seiux
;;
sequent)
basic_machine=i386-sequent
;;
sh)
basic_machine=sh-hitachi
os=-hms
;;
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
;;
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
;;
tic54x | c54x*)
basic_machine=tic54x-unknown
os=-coff
;;
tic55x | c55x*)
basic_machine=tic55x-unknown
os=-coff
;;
tic6x | c6x*)
basic_machine=tic6x-unknown
os=-coff
;;
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
;;
ymp)
basic_machine=ymp-cray
os=-unicos
;;
z8k-*-coff)
basic_machine=z8k-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
;;
- sh3 | sh4 | sh[34]eb | sh[1234]le | sh[23]ele)
+ sh[1234] | sh[24]a | sh[34]eb | sh[1234]le | sh[23]ele)
basic_machine=sh-unknown
;;
- sh64)
- basic_machine=sh64-unknown
- ;;
sparc | sparcv8 | sparcv9 | sparcv9b)
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.
-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* | -sunos | -sunos[34]*\
| -hpux* | -unos* | -osf* | -luna* | -dgux* | -solaris* | -sym* \
| -amigaos* | -amigados* | -msdos* | -newsos* | -unicos* | -aof* \
| -aos* \
| -nindy* | -vxsim* | -vxworks* | -ebmon* | -hms* | -mvs* \
| -clix* | -riscos* | -uniplus* | -iris* | -rtu* | -xenix* \
- | -hiux* | -386bsd* | -knetbsd* | -mirbsd* | -netbsd* | -openbsd* \
+ | -hiux* | -386bsd* | -knetbsd* | -mirbsd* | -netbsd* \
+ | -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* \
| -cygwin* | -pe* | -psos* | -moss* | -proelf* | -rtems* \
- | -mingw32* | -linux-gnu* | -linux-uclibc* | -uxpv* | -beos* | -mpeix* | -udk* \
+ | -mingw32* | -linux-gnu* | -linux-newlib* | -linux-uclibc* \
+ | -uxpv* | -beos* | -mpeix* | -udk* \
| -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*)
+ | -powermax* | -dnix* | -nx6 | -nx7 | -sei* | -dragonfly* \
+ | -skyos* | -haiku* | -rdos*)
# 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* \
+ | -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
;;
-kaos*)
os=-kaos
;;
+ -zvmoe)
+ os=-zvmoe
+ ;;
-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
*-acorn)
os=-riscix1.2
;;
arm*-rebel)
os=-linux
;;
arm*-semi)
os=-aout
;;
c4x-* | tic4x-*)
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
# This also exists in the configure program, but was not the
# default.
# os=-sunos4
;;
m68*-cisco)
os=-aout
;;
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
;;
-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 0
+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/configure.ac b/configure.ac
index d48c14b9..26f45b87 100644
--- a/configure.ac
+++ b/configure.ac
@@ -1,843 +1,849 @@
# Configure.ac script for Libgcrypt
# Copyright (C) 1998, 1999, 2000, 2001, 2002
# 2003, 2004 Free Software Foundation, Inc.
#
# This file is part of Libgcrypt.
#
# Libgcrypt 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.
#
# Libgcrypt 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, write to the Free Software
# Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA
#
# (Process this file with autoconf to produce a configure script.)
AC_REVISION($Revision$)
AC_PREREQ(2.59)
min_automake_version="1.9.3"
# Version number: Remember to change it immediately *after* a release
# but remove a "-cvs" prefix right *before* a release and append it
# soon later.
-AC_INIT(libgcrypt, 1.2.3-cvs, bug-libgcrypt@gnupg.org)
+AC_INIT(libgcrypt, 1.2.3, bug-libgcrypt@gnupg.org)
# LT Version numbers, remember to change them just *before* a release.
# (Interfaces removed: CURRENT++, AGE=0, REVISION=0)
# (Interfaces added: CURRENT++, AGE++, REVISION=0)
# (No interfaces changed: REVISION++)
LIBGCRYPT_LT_CURRENT=13
LIBGCRYPT_LT_AGE=2
-LIBGCRYPT_LT_REVISION=1
+LIBGCRYPT_LT_REVISION=2
# If the API is changed in an incompatible way: increment the next counter.
LIBGCRYPT_CONFIG_API_VERSION=1
NEED_GPG_ERROR_VERSION=0.5
PACKAGE=$PACKAGE_NAME
VERSION=$PACKAGE_VERSION
AC_CONFIG_SRCDIR([src/gcrypt.h])
AC_CANONICAL_TARGET()
AM_INIT_AUTOMAKE($PACKAGE, $VERSION)
AM_CONFIG_HEADER(config.h)
AM_MAINTAINER_MODE
AH_TOP([
/* need this, because some autoconf tests rely on this (e.g. stpcpy)
* and it should be used for new programs */
#define _GNU_SOURCE 1
])
AH_BOTTOM([
#define _GCRYPT_IN_LIBGCRYPT 1
])
AH_VERBATIM([_REENTRANT],
[/* To allow the use of Libgcrypt in multithreaded programs we have to use
special features from the library. */
#ifndef _REENTRANT
# define _REENTRANT 1
#endif
])
AC_SUBST(LIBGCRYPT_LT_CURRENT)
AC_SUBST(LIBGCRYPT_LT_AGE)
AC_SUBST(LIBGCRYPT_LT_REVISION)
AC_SUBST(PACKAGE)
AC_SUBST(VERSION)
AC_DEFINE_UNQUOTED(PACKAGE, "$PACKAGE", [Name of this package])
AC_DEFINE_UNQUOTED(VERSION, "$VERSION", [Version of this package])
AC_PROG_MAKE_SET
missing_dir=`cd $ac_aux_dir && pwd`
AM_MISSING_PROG(ACLOCAL, aclocal, $missing_dir)
AM_MISSING_PROG(AUTOCONF, autoconf, $missing_dir)
AM_MISSING_PROG(AUTOMAKE, automake, $missing_dir)
AM_MISSING_PROG(AUTOHEADER, autoheader, $missing_dir)
# AM_MISSING_PROG(MAKEINFO, makeinfo, $missing_dir)
AC_PROG_CC
AC_PROG_CPP
AM_PROG_AS
AC_ISC_POSIX
AC_PROG_INSTALL
AC_PROG_AWK
AM_PROG_LIBTOOL
##########################
## General definitions. ##
##########################
# Used by libgcrypt-config
LIBGCRYPT_CONFIG_LIBS="-lgcrypt"
LIBGCRYPT_CONFIG_CFLAGS=""
# Definitions for symmetric ciphers.
available_ciphers="arcfour blowfish cast5 des aes twofish serpent rfc2268"
enabled_ciphers=""
# Definitions for public-key ciphers.
available_pubkey_ciphers="dsa elgamal rsa"
enabled_pubkey_ciphers=""
# Definitions for message digests.
available_digests="crc md4 md5 rmd160 sha1 sha256"
available_digests_64="sha512 tiger"
enabled_digests=""
# Definitions for random modules.
available_random_modules="linux egd unix"
auto_random_modules="$available_random_modules"
# Supported thread backends.
LIBGCRYPT_THREAD_MODULES=""
# Other definitions.
print_egd_notice=no
# Setup some stuff depending on host/target.
case "${target}" in
*-*-mingw32*)
available_random_modules="w32"
ac_cv_have_dev_random=no
AC_DEFINE(USE_ONLY_8DOT3,1,
[set this to limit filenames to the 8.3 format])
AC_DEFINE(HAVE_DRIVE_LETTERS,1,
[defined if we must run on a stupid file system])
AC_DEFINE(HAVE_DOSISH_SYSTEM,1,
[defined if we run on some of the PCDOS like systems
(DOS, Windoze. OS/2) with special properties like
no file modes])
;;
i?86-emx-os2 | i?86-*-os2*emx)
# OS/2 with the EMX environment
ac_cv_have_dev_random=no
AC_DEFINE(HAVE_DRIVE_LETTERS)
AC_DEFINE(HAVE_DOSISH_SYSTEM)
;;
i?86-*-msdosdjgpp*)
# DOS with the DJGPP environment
ac_cv_have_dev_random=no
AC_DEFINE(HAVE_DRIVE_LETTERS)
AC_DEFINE(HAVE_DOSISH_SYSTEM)
;;
*-*-freebsd*)
# FreeBSD
CPPFLAGS="$CPPFLAGS -I/usr/local/include"
LDFLAGS="$LDFLAGS -L/usr/local/lib"
;;
*-*-hpux*)
if test -z "$GCC" ; then
CFLAGS="$CFLAGS -Ae -D_HPUX_SOURCE"
fi
;;
*-dec-osf4*)
if test -z "$GCC" ; then
# Suppress all warnings
# to get rid of the unsigned/signed char mismatch warnings.
CFLAGS="$CFLAGS -w"
fi
;;
m68k-atari-mint)
;;
*)
;;
esac
# A printable OS Name is sometime useful.
case "${target}" in
*-*-mingw32*)
PRINTABLE_OS_NAME="MingW32"
;;
i?86-emx-os2 | i?86-*-os2*emx )
PRINTABLE_OS_NAME="OS/2"
;;
i?86-*-msdosdjgpp*)
PRINTABLE_OS_NAME="MSDOS/DJGPP"
;;
*-linux*)
PRINTABLE_OS_NAME="GNU/Linux"
;;
*)
PRINTABLE_OS_NAME=`uname -s || echo "Unknown"`
;;
esac
case "${target}" in
*-openbsd*)
# FIXME: Are these the best flags for OpenBSD?
NAME_OF_DEV_RANDOM="/dev/srandom"
NAME_OF_DEV_URANDOM="/dev/urandom"
# DYNLINK_MOD_CFLAGS="-shared -rdynamic $CFLAGS_PIC -Wl,-Bshareable -Wl,-x"
;;
*-solaris* | *-irix* | *-dec-osf* | *-netbsd* )
NAME_OF_DEV_RANDOM="/dev/random"
NAME_OF_DEV_URANDOM="/dev/random"
# DYNLINK_MOD_CFLAGS="-shared $CFLAGS_PIC"
;;
*)
NAME_OF_DEV_RANDOM="/dev/random"
NAME_OF_DEV_URANDOM="/dev/urandom"
# -shared is a gcc-ism. Find pic flags from GNUPG_CHECK_PIC.
# if test -n "$GCC" ; then
# DYNLINK_MOD_CFLAGS="-shared $CFLAGS_PIC"
# else
# DYNLINK_MOD_CFLAGS="$CFLAGS_PIC"
# fi
;;
esac
AC_C_BIGENDIAN
AC_CHECK_SIZEOF(unsigned short, 2)
AC_CHECK_SIZEOF(unsigned int, 4)
AC_CHECK_SIZEOF(unsigned long, 4)
AC_CHECK_SIZEOF(unsigned long long, 0)
if test "$ac_cv_sizeof_unsigned_short" = "0" \
|| test "$ac_cv_sizeof_unsigned_int" = "0" \
|| test "$ac_cv_sizeof_unsigned_long" = "0"; then
AC_MSG_WARN([Hmmm, something is wrong with the sizes - using defaults]);
fi
# Do we have any 64-bit data types?
if test "$ac_cv_sizeof_unsigned_int" != "8" \
&& test "$ac_cv_sizeof_unsigned_long" != "8" \
&& test "$ac_cv_sizeof_unsigned_long_long" != "8" \
&& test "$ac_cv_sizeof_uint64_t" != "8"; then
AC_MSG_WARN([No 64-bit types. Disabling TIGER/192, SHA-384, and SHA-512])
else
available_digests="$available_digests $available_digests_64"
fi
# If not specified otherwise, all available algorithms will be
# included.
default_ciphers="$available_ciphers"
default_pubkey_ciphers="$available_pubkey_ciphers"
default_digests="$available_digests"
############################
## Command line switches. ##
############################
# Implementation of the --enable-ciphers switch.
AC_ARG_ENABLE(ciphers,
AC_HELP_STRING([--enable-ciphers=ciphers],
[select the symmetric ciphers to include]),
[enabled_ciphers=$(echo $enableval | tr , ' ' | tr '[A-Z]' '[a-z]')],
[enabled_ciphers=""])
if test "x$enabled_ciphers" = "x" \
-o "$enabled_ciphers" = "yes" \
-o "$enabled_ciphers" = "no"; then
enabled_ciphers=$default_ciphers
fi
AC_MSG_CHECKING([which symmetric ciphers to include])
for cipher in $enabled_ciphers; do
LIST_MEMBER($cipher, $available_ciphers)
if test "$found" = "0"; then
AC_MSG_ERROR([unsupported cipher specified])
fi
done
AC_MSG_RESULT([$enabled_ciphers])
# Implementation of the --enable-pubkey-ciphers switch.
AC_ARG_ENABLE(pubkey-ciphers,
AC_HELP_STRING([--enable-pubkey-ciphers=ciphers],
[select the public-key ciphers to include]),
[enabled_pubkey_ciphers=$(echo $enableval | tr , ' ' | tr '[A-Z]' '[a-z]')],
[enabled_pubkey_ciphers=""])
if test "x$enabled_pubkey_ciphers" = "x" \
-o "$enabled_pubkey_ciphers" = "yes" \
-o "$enabled_pubkey_ciphers" = "no"; then
enabled_pubkey_ciphers=$default_pubkey_ciphers
fi
AC_MSG_CHECKING([which public-key ciphers to include])
for cipher in $enabled_pubkey_ciphers; do
LIST_MEMBER($cipher, $available_pubkey_ciphers)
if test "$found" = "0"; then
AC_MSG_ERROR([unsupported public-key cipher specified])
fi
done
AC_MSG_RESULT([$enabled_pubkey_ciphers])
# Implementation of the --enable-digests switch.
AC_ARG_ENABLE(digests,
AC_HELP_STRING([--enable-digests=digests],
[select the message digests to include]),
[enabled_digests=$(echo $enableval | tr , ' ' | tr '[A-Z]' '[a-z]')],
[enabled_digests=""])
if test "x$enabled_digests" = "x" \
-o "$enabled_digests" = "yes" \
-o "$enabled_digests" = "no"; then
enabled_digests=$default_digests
fi
AC_MSG_CHECKING([which message digests to include])
for digest in $enabled_digests; do
LIST_MEMBER($digest, $available_digests)
if test "$found" = "0"; then
AC_MSG_ERROR([unsupported message digest specified])
fi
done
AC_MSG_RESULT([$enabled_digests])
# Implementation of the --enable-random switch.
AC_ARG_ENABLE(random,
AC_HELP_STRING([--enable-random=name],
[select which random number generator to use]),
[random=$(echo $enableval | tr '[A-Z]' '[a-z]')],
[])
if test "x$random" = "x" -o "$random" = "yes" -o "$random" = "no"; then
random=default
fi
AC_MSG_CHECKING([which random module to use])
if test "$random" != "default" -a "$random" != "auto"; then
LIST_MEMBER($random, $available_random_modules)
if test "$found" = "0"; then
AC_MSG_ERROR([unsupported random module specified])
fi
fi
AC_MSG_RESULT($random)
# Implementation of the --disable-dev-random switch.
AC_MSG_CHECKING([whether use of /dev/random is requested])
AC_ARG_ENABLE(dev-random,
[ --disable-dev-random disable the use of dev random],
try_dev_random=$enableval, try_dev_random=yes)
AC_MSG_RESULT($try_dev_random)
# Implementation of the --with-egd-socket switch.
AC_ARG_WITH(egd-socket,
[ --with-egd-socket=NAME Use NAME for the EGD socket)],
egd_socket_name="$withval", egd_socket_name="" )
AC_DEFINE_UNQUOTED(EGD_SOCKET_NAME, "$egd_socket_name",
[Define if you don't want the default EGD socket name.
For details see cipher/rndegd.c])
# Implementation of --disable-asm.
AC_MSG_CHECKING([whether assembler modules are requested])
AC_ARG_ENABLE([asm],
AC_HELP_STRING([--disable-asm],
[Disable assembler modules]),
[try_asm_modules=$enableval],
[try_asm_modules=yes])
AC_MSG_RESULT($try_asm_modules)
# Implementation of the --enable-m-guard switch.
AC_MSG_CHECKING([whether memory guard is requested])
AC_ARG_ENABLE(m-guard,
[ --enable-m-guard enable memory guard facility],
use_m_guard=$enableval, use_m_guard=no)
AC_MSG_RESULT($use_m_guard)
if test "$use_m_guard" = yes ; then
AC_DEFINE(M_GUARD,1,[Define to use the (obsolete) malloc guarding feature])
fi
# Implementation of the --with-capabilities switch.
# Check whether we want to use Linux capabilities
AC_MSG_CHECKING([whether use of capabilities is requested])
AC_ARG_WITH(capabilities,
[ --with-capabilities use linux capabilities [default=no]],
[use_capabilities="$withval"],[use_capabilities=no])
AC_MSG_RESULT($use_capabilities)
# We don't have a test to check whether as(1) knows about the
# non executable stack option. Thus we provide an option to enable it.
AC_MSG_CHECKING([whether non excutable stack support is requested])
AC_ARG_ENABLE(noexecstack,
AC_HELP_STRING([--enable-noexecstack],
[enable non executable stack support (gcc only)]),
noexecstack_support=$enableval, noexecstack_support=no)
AC_MSG_RESULT($noexecstack_support)
AC_DEFINE_UNQUOTED(PRINTABLE_OS_NAME, "$PRINTABLE_OS_NAME",
[A human readable text with the name of the OS])
# For some systems we know that we have ld_version scripts.
# Use it then as default.
have_ld_version_script=no
case "${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")
AC_DEFINE_UNQUOTED(NAME_OF_DEV_RANDOM, "$NAME_OF_DEV_RANDOM",
[defined to the name of the strong random device])
AC_DEFINE_UNQUOTED(NAME_OF_DEV_URANDOM, "$NAME_OF_DEV_URANDOM",
[defined to the name of the weaker random device])
###############################
#### Checks for libraries. ####
###############################
AM_PATH_GPG_ERROR("$NEED_GPG_ERROR_VERSION")
if test "x$GPG_ERROR_LIBS" = "x"; then
AC_MSG_ERROR([libgpg-error is needed.
See ftp://ftp.gnupg.org/gcrypt/libgpg-error/ .])
fi
AC_DEFINE(GPG_ERR_SOURCE_DEFAULT, GPG_ERR_SOURCE_GCRYPT,
[The default error source for libgcrypt.])
# Solaris needs -lsocket and -lnsl. Unisys system includes
# gethostbyname in libsocket but needs libnsl for socket.
AC_CHECK_LIB(nsl, gethostbyname)
AC_CHECK_LIB(socket, socket, ac_need_libsocket=1, ac_try_nsl=1)
if test x$ac_need_libsocket = x1; then
LIBS="$LIBS -lsocket"
fi
if test x$ac_try_nsl = x1; then
AC_CHECK_LIB(nsl, gethostbyname, ac_need_libnsl=1)
if test x$ac_need_libnsl = x1
then
LIBS="$LIBS -lnsl"
fi
fi
##################################
#### Checks for header files. ####
##################################
AC_HEADER_STDC
AC_CHECK_HEADERS(unistd.h sys/select.h)
##########################################
#### Checks for typedefs, structures, ####
#### and compiler characteristics. ####
##########################################
AC_C_CONST
AC_C_INLINE
AC_TYPE_SIZE_T
AC_TYPE_SIGNAL
AC_DECL_SYS_SIGLIST
GNUPG_CHECK_TYPEDEF(byte, HAVE_BYTE_TYPEDEF)
GNUPG_CHECK_TYPEDEF(ushort, HAVE_USHORT_TYPEDEF)
GNUPG_CHECK_TYPEDEF(ulong, HAVE_ULONG_TYPEDEF)
GNUPG_CHECK_TYPEDEF(u16, HAVE_U16_TYPEDEF)
GNUPG_CHECK_TYPEDEF(u32, HAVE_U32_TYPEDEF)
AC_CHECK_TYPE(socklen_t,,
[AC_DEFINE_UNQUOTED([socklen_t],
[int],
[Define to `int' if unavailable.])],
[
#include <sys/types.h>
#include <sys/socket.h>
])
#######################################
#### Checks for library functions. ####
#######################################
AC_FUNC_VPRINTF
# We have replacements for these in src/missing-string.c
AC_CHECK_FUNCS(stpcpy strcasecmp)
# We have replacements for these in src/g10lib.h
AC_CHECK_FUNCS(strtoul memmove stricmp atexit raise)
# Other checks
AC_CHECK_FUNCS(strerror rand mmap getpagesize waitpid wait4)
AC_CHECK_FUNCS(gettimeofday getrusage gethrtime clock_gettime)
GNUPG_CHECK_MLOCK
# Check whether we can use Linux capabilities as requested.
if test "$use_capabilities" = "yes" ; then
use_capabilities=no
AC_CHECK_HEADERS(sys/capability.h)
if test "$ac_cv_header_sys_capability_h" = "yes" ; then
AC_CHECK_LIB(cap, cap_init, ac_need_libcap=1)
if test "$ac_cv_lib_cap_cap_init" = "yes"; then
AC_DEFINE(USE_CAPABILITIES,1,
[define if capabilities should be used])
LIBS="$LIBS -lcap"
use_capabilities=yes
fi
fi
if test "$use_capabilities" = "no" ; then
AC_MSG_WARN([[
***
*** The use of capabilities on this system is not possible.
*** You need a recent Linux kernel and some patches:
*** fcaps-2.2.9-990610.patch (kernel patch for 2.2.9)
*** fcap-module-990613.tar.gz (kernel module)
*** libcap-1.92.tar.gz (user mode library and utilities)
*** And you have to configure the kernel with CONFIG_VFS_CAP_PLUGIN
*** set (filesystems menu). Be warned: This code is *really* ALPHA.
***]])
fi
fi
# Check whether a random device is available.
if test "$try_dev_random" = yes ; then
AC_CACHE_CHECK(for random device, ac_cv_have_dev_random,
[if test -r "$NAME_OF_DEV_RANDOM" && test -r "$NAME_OF_DEV_URANDOM" ; then
ac_cv_have_dev_random=yes; else ac_cv_have_dev_random=no; fi])
if test "$ac_cv_have_dev_random" = yes; then
AC_DEFINE(HAVE_DEV_RANDOM,1,
[defined if the system supports a random device] )
fi
else
AC_MSG_CHECKING(for random device)
ac_cv_have_dev_random=no
AC_MSG_RESULT(has been disabled)
fi
# Figure out the random modules for this configuration.
if test "$random" = "default"; then
# Select default value.
if test "$ac_cv_have_dev_random" = yes; then
# Try Linuxish random device.
random_modules="linux"
else
case "${target}" in
*-*-mingw32*|*-*-cygwin*)
# Windows random device.
random_modules="w32"
;;
*)
# Build everything, allow to select at runtime.
random_modules="$auto_random_modules"
;;
esac
fi
else
if test "$random" = "auto"; then
# Build everything, allow to select at runtime.
random_modules="$auto_random_modules"
fi
fi
#
# Setup assembler stuff.
#
GNUPG_SYS_SYMBOL_UNDERSCORE()
AC_MSG_CHECKING(for mpi assembler functions)
if test -f $srcdir/mpi/config.links ; then
. $srcdir/mpi/config.links
AC_CONFIG_LINKS("$mpi_ln_list")
ac_cv_mpi_mod_list="$mpi_mod_list"
ac_cv_mpi_sflags="$mpi_sflags"
ac_cv_mpi_config_done="yes"
AC_MSG_RESULT(done)
else
AC_MSG_RESULT(failed)
AC_MSG_ERROR([mpi/config.links missing!])
fi
MPI_MOD_LIST_LO=""
MPI_MOD_LIST_O=""
if test "$ac_cv_mpi_mod_list" != ""; then
for i in $ac_cv_mpi_mod_list; do
MPI_MOD_LIST_LO="$MPI_MOD_LIST_LO $i.lo"
MPI_MOD_LIST_O="$MPI_MOD_LIST_O $i.o"
done
fi
AC_SUBST(MPI_MOD_LIST_LO)
AC_SUBST(MPI_MOD_LIST_O)
MPI_SFLAGS="$ac_cv_mpi_sflags"
AC_SUBST(MPI_SFLAGS)
# Allow users to append something to the version string without
# flagging it as development version. The user version part is
# considered everything after a dash.
changequote(,)#
tmp_pat='[a-zA-Z]'
changequote([,])#
if echo "$VERSION" | sed 's/-.*//' | grep "$tmp_pat" >/dev/null ; then
AC_DEFINE(IS_DEVELOPMENT_VERSION,1,
[Defined if this is not a regular release])
fi
AM_CONDITIONAL(CROSS_COMPILING, test x$cross_compiling = xyes)
-# We don't check for GNU make anymore - automake should not have the
-# old flaws anymore.
-#GNUPG_CHECK_GNUMAKE
+
+# This is handy for debugging so the compiler doesn't rearrange
+# things and eliminate variables.
+AC_ARG_ENABLE(optimization,
+ AC_HELP_STRING([--disable-optimization],
+ [disable compiler optimization]),
+ [if test $enableval = no ; then
+ CFLAGS=`echo $CFLAGS | sed 's/-O[[0-9]]//'`
+ fi])
AC_ARG_ENABLE(gcc-warnings,
AC_HELP_STRING([--enable-gcc-warnings],
[enable more verbose gcc warnings]),
[more_gcc_warnings="$enableval"],
[more_gcc_warnings="no"])
if test "$GCC" = yes; then
if test "$USE_MAINTAINER_MODE" = "yes" ||
test "$more_gcc_warnings" = "yes"; then
CFLAGS="$CFLAGS -Wall -Wcast-align -Wshadow -Wstrict-prototypes"
if test "$more_gcc_warnings" = "yes"; then
CFLAGS="$CFLAGS -W -Wpointer-arith -Wbad-function-cast"
CFLAGS="$CFLAGS -Wwrite-strings -Wunreachable-code"
CFLAGS="$CFLAGS -Wno-sign-compare"
fi
else
CFLAGS="$CFLAGS -Wall"
fi
# Non exec stack hack. Fixme: Write a test to check whether as
# can cope with it and use the enable-noexecstack option only to
# disable it in case it is required on some platforms.
if test "$noexecstack_support" = yes; then
NOEXECSTACK_FLAGS="-Wa,--noexecstack"
fi
fi
AC_SUBST(NOEXECSTACK_FLAGS)
#
# Make the version number in src/gcrypt.h the same as the one here.
# (this is easier than to have a *.in file just for one substitution)
#
GNUPG_FIX_HDR_VERSION([src/gcrypt.h], GCRYPT_VERSION)
AC_SUBST(LIBGCRYPT_CONFIG_API_VERSION)
AC_SUBST(LIBGCRYPT_CONFIG_LIBS)
AC_SUBST(LIBGCRYPT_CONFIG_CFLAGS)
AC_SUBST(LIBGCRYPT_THREAD_MODULES)
AC_CONFIG_COMMANDS([gcrypt-conf],[[
chmod +x src/libgcrypt-config
]],[[
prefix=$prefix
exec_prefix=$exec_prefix
libdir=$libdir
datadir=$datadir
DATADIRNAME=$DATADIRNAME
]])
#####################
#### Conclusion. ####
#####################
# Define conditional sources and config.h symbols depending on the
# selected ciphers, pubkey-ciphers, digests and random modules.
LIST_MEMBER(arcfour, $enabled_ciphers)
test "$found" = "1" && GCRYPT_CIPHERS="$GCRYPT_CIPHERS arcfour.lo"
AC_DEFINE_UNQUOTED(USE_ARCFOUR, $found,
[Defined if this module should be included])
LIST_MEMBER(blowfish, $enabled_ciphers)
test "$found" = "1" && GCRYPT_CIPHERS="$GCRYPT_CIPHERS blowfish.lo"
AC_DEFINE_UNQUOTED(USE_BLOWFISH, $found,
[Defined if this module should be included])
LIST_MEMBER(cast5, $enabled_ciphers)
test "$found" = "1" && GCRYPT_CIPHERS="$GCRYPT_CIPHERS cast5.lo"
AC_DEFINE_UNQUOTED(USE_CAST5, $found,
[Defined if this module should be included])
LIST_MEMBER(des, $enabled_ciphers)
test "$found" = "1" && GCRYPT_CIPHERS="$GCRYPT_CIPHERS des.lo"
AC_DEFINE_UNQUOTED(USE_DES, $found,
[Defined if this module should be included])
LIST_MEMBER(aes, $enabled_ciphers)
test "$found" = "1" && GCRYPT_CIPHERS="$GCRYPT_CIPHERS rijndael.lo"
AC_DEFINE_UNQUOTED(USE_AES, $found,
[Defined if this module should be included])
LIST_MEMBER(twofish, $enabled_ciphers)
test "$found" = "1" && GCRYPT_CIPHERS="$GCRYPT_CIPHERS twofish.lo"
AC_DEFINE_UNQUOTED(USE_TWOFISH, $found,
[Defined if this module should be included])
LIST_MEMBER(serpent, $enabled_ciphers)
test "$found" = "1" && GCRYPT_CIPHERS="$GCRYPT_CIPHERS serpent.lo"
AC_DEFINE_UNQUOTED(USE_SERPENT, $found,
[Defined if this module should be included])
LIST_MEMBER(rfc2268, $enabled_ciphers)
test "$found" = "1" && GCRYPT_CIPHERS="$GCRYPT_CIPHERS rfc2268.lo"
AC_DEFINE_UNQUOTED(USE_RFC2268, $found,
[Defined if this module should be included])
LIST_MEMBER(dsa, $enabled_pubkey_ciphers)
test "$found" = "1" && GCRYPT_PUBKEY_CIPHERS="$GCRYPT_PUBKEY_CIPHERS dsa.lo"
AC_DEFINE_UNQUOTED(USE_DSA, $found,
[Defined if this module should be included])
LIST_MEMBER(rsa, $enabled_pubkey_ciphers)
test "$found" = "1" && GCRYPT_PUBKEY_CIPHERS="$GCRYPT_PUBKEY_CIPHERS rsa.lo"
AC_DEFINE_UNQUOTED(USE_RSA, $found,
[Defined if this module should be included])
LIST_MEMBER(elgamal, $enabled_pubkey_ciphers)
test "$found" = "1" \
&& GCRYPT_PUBKEY_CIPHERS="$GCRYPT_PUBKEY_CIPHERS elgamal.lo"
AC_DEFINE_UNQUOTED(USE_ELGAMAL, $found,
[Defined if this module should be included])
LIST_MEMBER(crc, $enabled_digests)
test "$found" = "1" && GCRYPT_DIGESTS="$GCRYPT_DIGESTS crc.lo"
AC_DEFINE_UNQUOTED(USE_CRC, $found,
[Defined if this module should be included])
LIST_MEMBER(md4, $enabled_digests)
test "$found" = "1" && GCRYPT_DIGESTS="$GCRYPT_DIGESTS md4.lo"
AC_DEFINE_UNQUOTED(USE_MD4, $found,
[Defined if this module should be included])
LIST_MEMBER(md5, $enabled_digests)
test "$found" = "1" && GCRYPT_DIGESTS="$GCRYPT_DIGESTS md5.lo"
AC_DEFINE_UNQUOTED(USE_MD5, $found,
[Defined if this module should be included])
LIST_MEMBER(sha256, $enabled_digests)
test "$found" = "1" && GCRYPT_DIGESTS="$GCRYPT_DIGESTS sha256.lo"
AC_DEFINE_UNQUOTED(USE_SHA256, $found,
[Defined if this module should be included])
LIST_MEMBER(sha512, $enabled_digests)
test "$found" = "1" && GCRYPT_DIGESTS="$GCRYPT_DIGESTS sha512.lo"
AC_DEFINE_UNQUOTED(USE_SHA512, $found,
[Defined if this module should be included])
LIST_MEMBER(tiger, $enabled_digests)
test "$found" = "1" && GCRYPT_DIGESTS="$GCRYPT_DIGESTS tiger.lo"
AC_DEFINE_UNQUOTED(USE_TIGER, $found,
[Defined if this module should be included])
# rmd160 and sha1 should be included always.
GCRYPT_DIGESTS="$GCRYPT_DIGESTS rmd160.lo sha1.lo"
AC_DEFINE_UNQUOTED(USE_RMD160, 1, [Defined if this module should be included])
AC_DEFINE_UNQUOTED(USE_SHA1, 1, [Defined if this module should be included])
LIST_MEMBER(linux, $random_modules)
test "$found" = "1" && GCRYPT_RANDOM="$GCRYPT_RANDOM rndlinux.lo"
AC_DEFINE_UNQUOTED(USE_RNDLINUX, $found,
[Defined if the /dev/random based RNG should be used.])
LIST_MEMBER(unix, $random_modules)
test "$found" = "1" && GCRYPT_RANDOM="$GCRYPT_RANDOM rndunix.lo"
AC_DEFINE_UNQUOTED(USE_RNDUNIX, $found,
[Defined if the default Unix RNG should be used.])
if test "$found" = "1"; then
print_egd_notice=yes
fi
LIST_MEMBER(egd, $random_modules)
test "$found" = "1" && GCRYPT_RANDOM="$GCRYPT_RANDOM rndegd.lo"
AC_DEFINE_UNQUOTED(USE_RNDEGD, $found,
[Defined if the EGD based RNG should be used.])
LIST_MEMBER(w32, $random_modules)
test "$found" = "1" && GCRYPT_RANDOM="$GCRYPT_RANDOM rndw32.lo"
AC_DEFINE_UNQUOTED(USE_RNDW32, $found,
[Defined if the Windows specific RNG should be used.])
AC_SUBST([GCRYPT_CIPHERS])
AC_SUBST([GCRYPT_PUBKEY_CIPHERS])
AC_SUBST([GCRYPT_DIGESTS])
AC_SUBST([GCRYPT_RANDOM])
AC_SUBST(LIBGCRYPT_CIPHERS, $enabled_ciphers)
AC_SUBST(LIBGCRYPT_PUBKEY_CIPHERS, $enabled_pubkey_ciphers)
AC_SUBST(LIBGCRYPT_DIGESTS, $enabled_digests)
AC_CONFIG_FILES([
Makefile
mpi/Makefile
cipher/Makefile
doc/Makefile
src/Makefile
src/libgcrypt-config
tests/Makefile
w32-dll/Makefile
])
AC_OUTPUT
# Give some feedback
echo
if test "$print_egd_notice" = "yes"; then
cat <<G10EOF
The performance of the Unix random gatherer module (rndunix) is not
very good and it does not keep the entropy pool over multiple
invocations of GnuPG. The suggested way to overcome this problem is
to use the
Entropy Gathering Daemon (EGD)
which provides a entropy source for the whole system. It is written
in Perl and available at the GnuPG FTP servers. To enable EGD you
should rerun configure with the option "--enable-static-rnd=egd".
For more information consult the GnuPG webpages:
http://www.gnupg.org/download.html#egd
G10EOF
fi
warn=""
for file in "random_modules"; do
case "$file" in rndunix | rndw32) warn="$warn $file";; esac
done
if test -n "$warn"; then
echo "Please note that you are building a version of Libgcrypt with"
echo " $warn"
echo "included. These parts are licensed under the GPL and thus the"
echo "use of this library has to comply with the conditions of the GPL."
fi
# Give some feedback
echo " Configured for: $PRINTABLE_OS_NAME ($target)"
echo
diff --git a/depcomp b/depcomp
index 11e2d3bf..04701da5 100755
--- a/depcomp
+++ b/depcomp
@@ -1,522 +1,530 @@
#! /bin/sh
# depcomp - compile a program generating dependencies as side-effects
-scriptversion=2004-05-31.23
+scriptversion=2005-07-09.11
-# Copyright (C) 1999, 2000, 2003, 2004 Free Software Foundation, Inc.
+# Copyright (C) 1999, 2000, 2003, 2004, 2005 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, write to the Free Software
-# Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA
-# 02111-1307, USA.
+# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+# 02110-1301, USA.
# 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 <oliva@dcc.unicamp.br>.
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 outputing dependencies.
libtool Whether libtool is used (yes/no).
Report bugs to <bug-automake@gnu.org>.
EOF
- exit 0
+ exit $?
;;
-v | --v*)
echo "depcomp $scriptversion"
- exit 0
+ exit $?
;;
esac
if test -z "$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"
# 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
case "$depmode" in
gcc3)
## gcc 3 implements dependency tracking that does exactly what
## we want. Yay! Note: for some reason libtool 1.4 doesn't like
## it if -MD -MP comes after the -MF stuff. Hmm.
"$@" -MT "$object" -MD -MP -MF "$tmpdepfile"
stat=$?
if test $stat -eq 0; then :
else
rm -f "$tmpdepfile"
exit $stat
fi
mv "$tmpdepfile" "$depfile"
;;
gcc)
## There are various ways to get dependency output from gcc. Here's
## why we pick this rather obscure method:
## - Don't want to use -MD because we'd like the dependencies to end
## up in a subdir. Having to rename by hand is ugly.
## (We might end up doing this anyway to support other compilers.)
## - The DEPENDENCIES_OUTPUT environment variable makes gcc act like
## -MM, not -M (despite what the docs say).
## - Using -M directly means running the compiler twice (even worse
## than renaming).
if test -z "$gccflag"; then
gccflag=-MD,
fi
"$@" -Wp,"$gccflag$tmpdepfile"
stat=$?
if test $stat -eq 0; then :
else
rm -f "$tmpdepfile"
exit $stat
fi
rm -f "$depfile"
echo "$object : \\" > "$depfile"
alpha=ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz
## The second -e expression handles DOS-style file names with drive letters.
sed -e 's/^[^:]*: / /' \
-e 's/^['$alpha']:\/[^:]*: / /' < "$tmpdepfile" >> "$depfile"
## This next piece of magic avoids the `deleted header file' problem.
## The problem is that when a header file which appears in a .P file
## is deleted, the dependency causes make to die (because there is
## typically no way to rebuild the header). We avoid this by adding
## dummy dependencies for each header file. Too bad gcc doesn't do
## this for us directly.
tr ' ' '
' < "$tmpdepfile" |
## 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.
## Some versions of the HPUX 10.20 sed can't process this invocation
## correctly. Breaking it into two sed invocations is a workaround.
sed -e 's/^\\$//' -e '/^$/d' -e '/:$/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 $stat -eq 0; then :
else
rm -f "$tmpdepfile"
exit $stat
fi
rm -f "$depfile"
if test -f "$tmpdepfile"; then # yes, the sourcefile depend on other files
echo "$object : \\" > "$depfile"
# Clip off the initial element (the dependent). Don't try to be
# clever and replace this with sed code, as IRIX sed won't handle
# lines with more than a fixed number of characters (4096 in
# IRIX 6.2 sed, 8192 in IRIX 6.5). We also remove comment lines;
# the IRIX cc adds comments like `#:fec' to the end of the
# dependency line.
tr ' ' '
' < "$tmpdepfile" \
| sed -e 's/^.*\.o://' -e 's/#.*$//' -e '/^$/ d' | \
tr '
' ' ' >> $depfile
echo >> $depfile
# The second pass generates a dummy entry for each header file.
tr ' ' '
' < "$tmpdepfile" \
| sed -e 's/^.*\.o://' -e 's/#.*$//' -e '/^$/ d' -e 's/$/:/' \
>> $depfile
else
# The sourcefile does not contain any dependencies, so just
# store a dummy comment line, to avoid errors with the Makefile
# "include basename.Plo" scheme.
echo "#dummy" > "$depfile"
fi
rm -f "$tmpdepfile"
;;
aix)
# The C for AIX Compiler uses -M and outputs the dependencies
# in a .u file. In older versions, this file always lives in the
# current directory. Also, the AIX compiler puts `$object:' at the
# start of each line; $object doesn't have directory information.
# Version 6 uses the directory in both cases.
stripped=`echo "$object" | sed 's/\(.*\)\..*$/\1/'`
tmpdepfile="$stripped.u"
if test "$libtool" = yes; then
"$@" -Wc,-M
else
"$@" -M
fi
stat=$?
if test -f "$tmpdepfile"; then :
else
stripped=`echo "$stripped" | sed 's,^.*/,,'`
tmpdepfile="$stripped.u"
fi
if test $stat -eq 0; then :
else
rm -f "$tmpdepfile"
exit $stat
fi
if test -f "$tmpdepfile"; then
outname="$stripped.o"
# Each line is of the form `foo.o: dependent.h'.
# Do two passes, one to just change these to
# `$object: dependent.h' and one to simply `dependent.h:'.
sed -e "s,^$outname:,$object :," < "$tmpdepfile" > "$depfile"
sed -e "s,^$outname: \(.*\)$,\1:," < "$tmpdepfile" >> "$depfile"
else
# The sourcefile does not contain any dependencies, so just
# store a dummy comment line, to avoid errors with the Makefile
# "include basename.Plo" scheme.
echo "#dummy" > "$depfile"
fi
rm -f "$tmpdepfile"
;;
icc)
# Intel's C compiler understands `-MD -MF file'. However on
# icc -MD -MF foo.d -c -o sub/foo.o sub/foo.c
# ICC 7.0 will fill foo.d with something like
# foo.o: sub/foo.c
# foo.o: sub/foo.h
# which is wrong. We want:
# sub/foo.o: sub/foo.c
# sub/foo.o: sub/foo.h
# sub/foo.c:
# sub/foo.h:
# ICC 7.1 will output
# foo.o: sub/foo.c sub/foo.h
# and will wrap long lines using \ :
# foo.o: sub/foo.c ... \
# sub/foo.h ... \
# ...
"$@" -MD -MF "$tmpdepfile"
stat=$?
if test $stat -eq 0; then :
else
rm -f "$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"
;;
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.
dir=`echo "$object" | sed -e 's|/[^/]*$|/|'`
test "x$dir" = "x$object" && dir=
base=`echo "$object" | sed -e 's|^.*/||' -e 's/\.o$//' -e 's/\.lo$//'`
if test "$libtool" = yes; then
- # Dependencies are output in .lo.d with libtool 1.4.
- # With libtool 1.5 they are output both in $dir.libs/$base.o.d
- # and in $dir.libs/$base.o.d and $dir$base.o.d. We process the
- # latter, because the former will be cleaned when $dir.libs is
- # erased.
- tmpdepfile1="$dir.libs/$base.lo.d"
- tmpdepfile2="$dir$base.o.d"
- tmpdepfile3="$dir.libs/$base.d"
+ # With Tru64 cc, shared objects can also be used to make a
+ # static library. This mecanism is used in libtool 1.4 series to
+ # handle both shared and static libraries in a single compilation.
+ # With libtool 1.4, dependencies were output in $dir.libs/$base.lo.d.
+ #
+ # With libtool 1.5 this exception was removed, and libtool now
+ # generates 2 separate objects for the 2 libraries. These two
+ # compilations output dependencies in 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=$dir.libs/$base.lo.d # libtool 1.4
+ tmpdepfile2=$dir$base.o.d # libtool 1.5
+ tmpdepfile3=$dir.libs/$base.o.d # libtool 1.5
+ tmpdepfile4=$dir.libs/$base.d # Compaq CCC V6.2-504
"$@" -Wc,-MD
else
- tmpdepfile1="$dir$base.o.d"
- tmpdepfile2="$dir$base.d"
- tmpdepfile3="$dir$base.d"
+ tmpdepfile1=$dir$base.o.d
+ tmpdepfile2=$dir$base.d
+ tmpdepfile3=$dir$base.d
+ tmpdepfile4=$dir$base.d
"$@" -MD
fi
stat=$?
if test $stat -eq 0; then :
else
- rm -f "$tmpdepfile1" "$tmpdepfile2" "$tmpdepfile3"
+ rm -f "$tmpdepfile1" "$tmpdepfile2" "$tmpdepfile3" "$tmpdepfile4"
exit $stat
fi
- if test -f "$tmpdepfile1"; then
- tmpdepfile="$tmpdepfile1"
- elif test -f "$tmpdepfile2"; then
- tmpdepfile="$tmpdepfile2"
- else
- tmpdepfile="$tmpdepfile3"
- fi
+ for tmpdepfile in "$tmpdepfile1" "$tmpdepfile2" "$tmpdepfile3" "$tmpdepfile4"
+ do
+ test -f "$tmpdepfile" && break
+ done
if test -f "$tmpdepfile"; then
sed -e "s,^.*\.[a-z]*:,$object:," < "$tmpdepfile" > "$depfile"
# That's a tab and a space in the [].
sed -e 's,^.*\.[a-z]*:[ ]*,,' -e 's,$,:,' < "$tmpdepfile" >> "$depfile"
else
echo "#dummy" > "$depfile"
fi
rm -f "$tmpdepfile"
;;
#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 $1 != '--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.
"$@" $dashmflag |
sed 's:^[ ]*[^: ][^:][^:]*\:[ ]*:'"$object"'\: :' > "$tmpdepfile"
rm -f "$depfile"
cat < "$tmpdepfile" > "$depfile"
tr ' ' '
' < "$tmpdepfile" | \
## Some versions of the HPUX 10.20 sed can't process this invocation
## correctly. Breaking it into two sed invocations is a workaround.
sed -e 's/^\\$//' -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 $1 != '--mode=compile'; do
shift
done
shift
fi
# X makedepend
shift
cleared=no
for arg in "$@"; do
case $cleared in
no)
set ""; shift
cleared=yes ;;
esac
case "$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.
-*|$object)
;;
*)
set fnord "$@" "$arg"; shift ;;
esac
done
obj_suffix="`echo $object | sed 's/^.*\././'`"
touch "$tmpdepfile"
${MAKEDEPEND-makedepend} -o"$obj_suffix" -f"$tmpdepfile" "$@"
rm -f "$depfile"
cat < "$tmpdepfile" > "$depfile"
sed '1,2d' "$tmpdepfile" | tr ' ' '
' | \
## Some versions of the HPUX 10.20 sed can't process this invocation
## correctly. Breaking it into two sed invocations is a workaround.
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 $1 != '--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 '/^# [0-9][0-9]* "\([^"]*\)".*/ s:: \1 \\:p' |
+ 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 "$tmpdepfile"
;;
msvisualcpp)
# Important note: in order to support this mode, a compiler *must*
# always write the preprocessed file to stdout, regardless of -o,
# because we must use -o when running libtool.
"$@" || exit $?
IFS=" "
for arg
do
case "$arg" in
"-Gm"|"/Gm"|"-Gi"|"/Gi"|"-ZI"|"/ZI")
set fnord "$@"
shift
shift
;;
*)
set fnord "$@" "$arg"
shift
shift
;;
esac
done
"$@" -E |
sed -n '/^#line [0-9][0-9]* "\([^"]*\)"/ s::echo "`cygpath -u \\"\1\\"`":p' | sort | uniq > "$tmpdepfile"
rm -f "$depfile"
echo "$object : \\" > "$depfile"
. "$tmpdepfile" | sed 's% %\\ %g' | sed -n '/^\(.*\)$/ s:: \1 \\:p' >> "$depfile"
echo " " >> "$depfile"
. "$tmpdepfile" | sed 's% %\\ %g' | sed -n '/^\(.*\)$/ s::\1\::p' >> "$depfile"
rm -f "$tmpdepfile"
;;
none)
exec "$@"
;;
*)
echo "Unknown depmode $depmode" 1>&2
exit 1
;;
esac
exit 0
# Local Variables:
# mode: shell-script
# sh-indentation: 2
# eval: (add-hook 'write-file-hooks 'time-stamp)
# time-stamp-start: "scriptversion="
# time-stamp-format: "%:y-%02m-%02d.%02H"
# time-stamp-end: "$"
# End:
diff --git a/doc/mdate-sh b/doc/mdate-sh
index 05e828b0..cd916c0a 100755
--- a/doc/mdate-sh
+++ b/doc/mdate-sh
@@ -1,170 +1,201 @@
#!/bin/sh
# Get modification time of a file or directory and pretty-print it.
-scriptversion=2003-11-09.00
+scriptversion=2005-06-29.22
-# Copyright (C) 1995, 1996, 1997, 2003 Free Software Foundation, Inc.
+# Copyright (C) 1995, 1996, 1997, 2003, 2004, 2005 Free Software
+# Foundation, Inc.
# written by Ulrich Drepper <drepper@gnu.ai.mit.edu>, 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, write to the Free Software Foundation,
-# Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+# Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
# 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 <bug-automake@gnu.org> or send patches to
# <automake-patches@gnu.org>.
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 time of FILE.
Report bugs to <bug-automake@gnu.org>.
EOF
- exit 0
+ exit $?
;;
-v | --v*)
echo "mdate-sh $scriptversion"
- exit 0
+ exit $?
;;
esac
# Prevent date giving response in another language.
LANG=C
export LANG
LC_ALL=C
export LC_ALL
LC_TIME=C
export LC_TIME
-save_arg1="$1"
+# GNU ls changes its time format in response to the TIME_STYLE
+# variable. Since we cannot assume `unset' works, revert this
+# variable to its documented default.
+if test "${TIME_STYLE+set}" = set; then
+ TIME_STYLE=posix-long-iso
+ export TIME_STYLE
+fi
+
+save_arg1=$1
# Find out how to get the extended ls output of a file or directory.
if ls -L /dev/null 1>/dev/null 2>&1; then
ls_command='ls -L -l -d'
else
ls_command='ls -l -d'
fi
# A `ls -l' line looks as follows on OS/2.
# drwxrwx--- 0 Aug 11 2001 foo
# This differs from Unix, which adds ownership information.
# drwxrwx--- 2 root root 4096 Aug 11 2001 foo
#
# To find the date, we split the line on spaces and iterate on words
# until we find a month. This cannot work with files whose owner is a
# user named `Jan', or `Feb', etc. However, it's unlikely that `/'
# will be owned by a user whose name is a month. So we first look at
# the extended ls output of the root directory to decide how many
# words should be skipped to get the date.
# On HPUX /bin/sh, "set" interprets "-rw-r--r--" as options, so the "x" below.
-set - x`$ls_command /`
+set x`ls -l -d /`
# Find which argument is the month.
month=
command=
until test $month
do
shift
# Add another shift to the command.
command="$command shift;"
case $1 in
Jan) month=January; nummonth=1;;
Feb) month=February; nummonth=2;;
Mar) month=March; nummonth=3;;
Apr) month=April; nummonth=4;;
May) month=May; nummonth=5;;
Jun) month=June; nummonth=6;;
Jul) month=July; nummonth=7;;
Aug) month=August; nummonth=8;;
Sep) month=September; nummonth=9;;
Oct) month=October; nummonth=10;;
Nov) month=November; nummonth=11;;
Dec) month=December; nummonth=12;;
esac
done
# Get the extended ls output of the file or directory.
-set - x`eval "$ls_command \"\$save_arg1\""`
+set dummy x`eval "$ls_command \"\$save_arg1\""`
# Remove all preceding arguments
eval $command
-# Get the month. Next argument is day, followed by the year or time.
-case $1 in
+# 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
-day=$2
+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 $year
# Local Variables:
# mode: shell-script
# sh-indentation: 2
# eval: (add-hook 'write-file-hooks 'time-stamp)
# time-stamp-start: "scriptversion="
# time-stamp-format: "%:y-%02m-%02d.%02H"
# time-stamp-end: "$"
# End:
diff --git a/doc/texinfo.tex b/doc/texinfo.tex
index 47372211..ff2c4065 100644
--- a/doc/texinfo.tex
+++ b/doc/texinfo.tex
@@ -1,7023 +1,7210 @@
% 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{2004-10-31.06}
+\def\texinfoversion{2005-07-05.19}
%
% Copyright (C) 1985, 1986, 1988, 1990, 1991, 1992, 1993, 1994, 1995,
-% 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004 Free Software
+% 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005 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 2, 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, write
-% to the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
-% Boston, MA 02111-1307, USA.
+% to the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
+% Boston, MA 02110-1301, USA.
%
% 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
% ftp://tug.org/tex/texinfo.tex
% (and all CTAN mirrors, see http://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.
\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}
\message{Basics,}
\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\ptexnoindent=\noindent
\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
% In some macros, we cannot use the `\? notation---the left quote is
% in some cases the escape char.
+\chardef\backChar = `\\
\chardef\colonChar = `\:
\chardef\commaChar = `\,
\chardef\dotChar = `\.
\chardef\exclamChar= `\!
+\chardef\plusChar = `\+
\chardef\questChar = `\?
\chardef\semiChar = `\;
\chardef\underChar = `\_
\chardef\spaceChar = `\ %
\chardef\spacecat = 10
\def\spaceisspace{\catcode\spaceChar=\spacecat}
+{% for help with debugging.
+ % example usage: \expandafter\show\activebackslash
+ \catcode`\! = 0 \catcode`\\ = \active
+ !global!def!activebackslash{\}
+}
+
% 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.
%
- \escapechar = `\\ % use backslash in output files.
\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.
\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 \oddfootingxxx.)
% The \baselineskip=24pt in plain's \makefootline has no effect.
\vskip 2\baselineskip
\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 \normalturnoffactive
+ }% 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\next{#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 `<space>\^^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
% We cannot use \next here, as it holds the macro to run;
% thus we 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 \next.
% (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\next\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 file.
+ % 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=3000 }
+\def\.{.\spacefactor=\endofsentencespacefactor\space}
% @! is an end-of-sentence bang.
-\def\!{!\spacefactor=3000 }
+\def\!{!\spacefactor=\endofsentencespacefactor\space}
% @? is an end-of-sentence query.
-\def\?{?\spacefactor=3000 }
+\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\underChar = \active
\gdef\mathunderscore{%
\catcode\underChar=\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 a typewriter
% font as three actual period characters.
%
\def\dots{%
\leavevmode
\hbox to 1.5em{%
\hskip 0pt plus 0.25fil
.\hfil.\hfil.%
\hskip 0pt plus 0.5fil
}%
}
% @enddots{} is an end-of-sentence ellipsis.
%
\def\enddots{%
\dots
- \spacefactor=3000
+ \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, to
+% 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
+% (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 we do).
+
+% double active backslashes.
+%
+{\catcode`\@=0 \catcode`\\=\active
+ @gdef@activebackslash{@catcode`@\=@active @otherbackslash}
+ @gdef@activebackslashdouble{%
+ @catcode@backChar=@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. I've
+% tinkered with it a little for texinfo, but it's definitely from there.
+%
+% #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{(}{\backslashlparen}{#1}%
+ \HyPsdSubst{)}{\backslashrparen}{#1}%
+}
+
+{\catcode\exclamChar = 0 \catcode\backChar = \other
+ !gdef!backslashlparen{\(}%
+ !gdef!backslashrparen{\)}%
+}
+
\ifpdf
\input pdfcolor
\pdfcatalog{/PageMode /UseOutlines}%
\def\dopdfimage#1#2#3{%
\def\imagewidth{#2}%
\def\imageheight{#3}%
% 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
\ifx\empty\imagewidth\else width \imagewidth \fi
\ifx\empty\imageheight\else height \imageheight \fi
\ifnum\pdftexversion<13
#1.pdf%
\else
{#1.pdf}%
\fi
\ifnum\pdftexversion < 14 \else
\pdfrefximage \pdflastximage
\fi}
\def\pdfmkdest#1{{%
- % We have to set dummies so commands such as @code in a section title
- % aren't expanded.
+ % We have to set dummies so commands such as @code, and characters
+ % such as \, aren't expanded when present in a section title.
\atdummies
- \normalturnoffactive
- \pdfdest name{#1} xyz%
- }}
- \def\pdfmkpgn#1{#1}
+ \activebackslashdouble
+ \def\pdfdestname{#1}%
+ \backslashparens\pdfdestname
+ \pdfdest name{\pdfdestname} xyz%
+ }}%
+ %
+ % used to mark target names; must be expandable.
+ \def\pdfmkpgn#1{#1}%
+ %
\let\linkcolor = \Blue % was Cyan, but that seems light?
\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. #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.
+ % #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 worthwhile, since most documents are normally structured.
+ % seem worth the trouble, since most documents are normally structured.
\def\pdfoutlinedest{#3}%
- \ifx\pdfoutlinedest\empty \def\pdfoutlinedest{#4}\fi
+ \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{#1}%
+ \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}%
- \let\thissecnum\empty
- \let\thissubsecnum\empty
+ \def\thissecnum{0}%
+ \def\thissubsecnum{0}%
}%
\def\numsecentry##1##2##3##4{%
\advancenumber{chap\thischapnum}%
\def\thissecnum{##2}%
- \let\thissubsecnum\empty
+ \def\thissubsecnum{0}%
}%
\def\numsubsecentry##1##2##3##4{%
\advancenumber{sec\thissecnum}%
\def\thissubsecnum{##2}%
}%
\def\numsubsubsecentry##1##2##3##4{%
\advancenumber{subsec\thissubsecnum}%
}%
- \let\thischapnum\empty
- \let\thissecnum\empty
- \let\thissubsecnum\empty
+ \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}%
- \input \jobname.toc
+ \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
- \turnoffactive
+ \setupdatafile
+ \activebackslash
\input \jobname.toc
\endgroup
}
%
- \def\makelinks #1,{%
- \def\params{#1}\def\E{END}%
- \ifx\params\E
- \let\nextmakelinks=\relax
- \else
- \let\nextmakelinks=\makelinks
- \ifnum\lnkcount>0,\fi
- \picknum{#1}%
- \startlink attr{/Border [0 0 0]}
- goto name{\pdfmkpgn{\the\pgn}}%
- \linkcolor #1%
- \advance\lnkcount by 1%
- \endlink
- \fi
- \nextmakelinks
- }
- \def\picknum#1{\expandafter\pn#1}
- \def\pn#1{%
- \def\p{#1}%
- \ifx\p\lbrace
- \let\nextpn=\ppn
- \else
- \let\nextpn=\ppnn
- \def\first{#1}
- \fi
- \nextpn
- }
- \def\ppn#1{\pgn=#1\gobble}
- \def\ppnn{\pgn=\first}
- \def\pdfmklnk#1{\lnkcount=0\makelinks #1,END,}
\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
\def\pdfurl#1{%
\begingroup
\normalturnoffactive\def\@{@}%
\makevalueexpandable
\leavevmode\Red
\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\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
}%
}
% 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
\def\setfont#1#2#3#4{\font#1=\fontprefix#2#3 scaled #4}
% 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}
% Text fonts (11.2pt, magstep1).
-\newcount\mainmagstep
-\ifx\bigger\relax
- % not really supported.
- \mainmagstep=\magstep1
- \setfont\textrm\rmshape{12}{1000}
- \setfont\texttt\ttshape{12}{1000}
-\else
- \mainmagstep=\magstephalf
- \setfont\textrm\rmshape{10}{\mainmagstep}
- \setfont\texttt\ttshape{10}{\mainmagstep}
-\fi
+\def\textnominalsize{11pt}
+\edef\mainmagstep{\magstephalf}
+\setfont\textrm\rmshape{10}{\mainmagstep}
+\setfont\texttt\ttshape{10}{\mainmagstep}
\setfont\textbf\bfshape{10}{\mainmagstep}
\setfont\textit\itshape{10}{\mainmagstep}
\setfont\textsl\slshape{10}{\mainmagstep}
\setfont\textsf\sfshape{10}{\mainmagstep}
\setfont\textsc\scshape{10}{\mainmagstep}
\setfont\textttsl\ttslshape{10}{\mainmagstep}
\font\texti=cmmi10 scaled \mainmagstep
\font\textsy=cmsy10 scaled \mainmagstep
% A few fonts for @defun names and args.
\setfont\defbf\bfshape{10}{\magstep1}
\setfont\deftt\ttshape{10}{\magstep1}
\setfont\defttsl\ttslshape{10}{\magstep1}
\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}
\setfont\smalltt\ttshape{9}{1000}
\setfont\smallbf\bfshape{10}{900}
\setfont\smallit\itshape{9}{1000}
\setfont\smallsl\slshape{9}{1000}
\setfont\smallsf\sfshape{9}{1000}
\setfont\smallsc\scshape{10}{900}
\setfont\smallttsl\ttslshape{10}{900}
\font\smalli=cmmi9
\font\smallsy=cmsy9
% Fonts for small examples (8pt).
+\def\smallernominalsize{8pt}
\setfont\smallerrm\rmshape{8}{1000}
\setfont\smallertt\ttshape{8}{1000}
\setfont\smallerbf\bfshape{10}{800}
\setfont\smallerit\itshape{8}{1000}
\setfont\smallersl\slshape{8}{1000}
\setfont\smallersf\sfshape{8}{1000}
\setfont\smallersc\scshape{10}{800}
\setfont\smallerttsl\ttslshape{10}{800}
\font\smalleri=cmmi8
\font\smallersy=cmsy8
% Fonts for title page (20.4pt):
+\def\titlenominalsize{20pt}
\setfont\titlerm\rmbshape{12}{\magstep3}
\setfont\titleit\itbshape{10}{\magstep4}
\setfont\titlesl\slbshape{10}{\magstep4}
\setfont\titlett\ttbshape{12}{\magstep3}
\setfont\titlettsl\ttslshape{10}{\magstep4}
\setfont\titlesf\sfbshape{17}{\magstep1}
\let\titlebf=\titlerm
\setfont\titlesc\scbshape{10}{\magstep4}
\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}
\setfont\chapit\itbshape{10}{\magstep3}
\setfont\chapsl\slbshape{10}{\magstep3}
\setfont\chaptt\ttbshape{12}{\magstep2}
\setfont\chapttsl\ttslshape{10}{\magstep3}
\setfont\chapsf\sfbshape{17}{1000}
\let\chapbf=\chaprm
\setfont\chapsc\scbshape{10}{\magstep3}
\font\chapi=cmmi12 scaled \magstep2
\font\chapsy=cmsy10 scaled \magstep3
% Section fonts (14.4pt).
+\def\secnominalsize{14pt}
\setfont\secrm\rmbshape{12}{\magstep1}
\setfont\secit\itbshape{10}{\magstep2}
\setfont\secsl\slbshape{10}{\magstep2}
\setfont\sectt\ttbshape{12}{\magstep1}
\setfont\secttsl\ttslshape{10}{\magstep2}
\setfont\secsf\sfbshape{12}{\magstep1}
\let\secbf\secrm
\setfont\secsc\scbshape{10}{\magstep2}
\font\seci=cmmi12 scaled \magstep1
\font\secsy=cmsy10 scaled \magstep2
% Subsection fonts (13.15pt).
+\def\ssecnominalsize{13pt}
\setfont\ssecrm\rmbshape{12}{\magstephalf}
\setfont\ssecit\itbshape{10}{1315}
\setfont\ssecsl\slbshape{10}{1315}
\setfont\ssectt\ttbshape{12}{\magstephalf}
\setfont\ssecttsl\ttslshape{10}{1315}
\setfont\ssecsf\sfbshape{12}{\magstephalf}
\let\ssecbf\ssecrm
\setfont\ssecsc\scbshape{10}{1315}
\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}
\setfont\reducedtt\ttshape{10}{1000}
\setfont\reducedbf\bfshape{10}{1000}
\setfont\reducedit\itshape{10}{1000}
\setfont\reducedsl\slshape{10}{1000}
\setfont\reducedsf\sfshape{10}{1000}
\setfont\reducedsc\scshape{10}{1000}
\setfont\reducedttsl\ttslshape{10}{1000}
\font\reducedi=cmmi10
\font\reducedsy=cmsy10
% 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
+ \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.
%
\textfonts \rm
% 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}
\setfont\shortcontbf\bfshape{10}{\magstep1} % no cmb12
\setfont\shortcontsl\slshape{12}{1000}
\setfont\shortconttt\ttshape{12}{1000}
%% 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\frenchspacing{%
+ \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 \frenchspacing #1}%
+ {\tt \rawbackslash \plainfrenchspacing #1}%
\null
}
\def\samp#1{`\tclose{#1}'\null}
\setfont\keyrm\rmshape{8}{1000}
\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}}}}
% 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
- \frenchspacing
+ \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
%
\global\def\code{\begingroup
- \catcode`\-=\active \let-\codedash
- \catcode`\_=\active \let_\codeunder
+ \catcode`\-=\active \catcode`\_=\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\arg{#1}%
- \ifx\arg\worddistinct
+ \def\txiarg{#1}%
+ \ifx\txiarg\worddistinct
\gdef\kbdexamplefont{\ttsl}\gdef\kbdfont{\ttsl}%
- \else\ifx\arg\wordexample
+ \else\ifx\txiarg\wordexample
\gdef\kbdexamplefont{\ttsl}\gdef\kbdfont{\tt}%
- \else\ifx\arg\wordcode
+ \else\ifx\txiarg\wordcode
\gdef\kbdexamplefont{\tt}\gdef\kbdfont{\tt}%
\else
\errhelp = \EMsimple
- \errmessage{Unknown @kbdinputstyle option `\arg'}%
+ \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{%
- {\frenchspacing #1}%
+ {\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.
+%
+% 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{%
$^{{\ooalign{\hfil\raise.07ex\hbox{\selectfonts\lllsize R}%
\hfil\crcr\Orb}}%
}$%
}
% 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 -\baselineskip
\global\advance\vsize by -\baselineskip
}
\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 <number>.
%
\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:
\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', which must be on a line
% by itself.
\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.
\obeylines %
\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.
\def\enddoignore{\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 \newindex.
-{\catcode`\@=11
-\gdef\newwrite{\alloc@7\write\chardef\sixt@@n}}
+% 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
%
- % \definedummyword defines \#1 as \realbackslash #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{%
- \expandafter\def\csname ##1\endcsname{\realbackslash ##1\space}%
- }%
- \def\definedummyletter##1{%
- \expandafter\def\csname ##1\endcsname{\realbackslash ##1}%
- }%
- \let\definedummyaccent\definedummyletter
- %
% Do the redefinitions.
\commondummies
}
-% For the aux file, @ is the escape character. So we want to redefine
-% everything using @ instead of \realbackslash. When everything uses
-% @, this will be simpler.
+% 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
%
- % (See comments in \indexdummies.)
- \def\definedummyword##1{%
- \expandafter\def\csname ##1\endcsname{@##1\space}%
- }%
- \def\definedummyletter##1{%
- \expandafter\def\csname ##1\endcsname{@##1}%
- }%
- \let\definedummyaccent\definedummyletter
- %
% Do the redefinitions.
\commondummies
}
-% Called from \indexdummies and \atdummies. \definedummyword and
-% \definedummyletter must be defined first.
+% Called from \indexdummies and \atdummies.
%
\def\commondummies{%
%
- \normalturnoffactive
+ % \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{_}%
+ \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}%
+ \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}%
+ \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{expansion}%
- \definedummyword{minus}%
- \definedummyword{pounds}%
- \definedummyword{point}%
- \definedummyword{print}%
- \definedummyword{result}%
+ \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
+ %
+ % 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
- %
- % Normal spaces, not active ones.
- \unsepspaces
- %
- % No macro expansion.
- \turnoffmacros
}
% \commondummiesnofonts: common to \commondummies and \indexnofonts.
%
% Better have this without active chars.
{
\catcode`\~=\other
\gdef\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}%
+ \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}%
+ \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{samp}%
- \definedummyword{strong}%
- \definedummyword{tie}%
- \definedummyword{uref}%
- \definedummyword{url}%
- \definedummyword{var}%
- \definedummyword{verb}%
- \definedummyword{w}%
+ \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\samp
+ \definedummyword\strong
+ \definedummyword\tie
+ \definedummyword\uref
+ \definedummyword\url
+ \definedummyword\var
+ \definedummyword\verb
+ \definedummyword\w
}
}
% \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{%
- \expandafter\let\csname ##1\endcsname\asis
- }%
+ \def\definedummyaccent##1{\let##1\asis}%
% We can just ignore other control letters.
- \def\definedummyletter##1{%
- \expandafter\def\csname ##1\endcsname{}%
- }%
+ \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{=>}%
%
- % Don't write macro names.
- \emptyusermacros
+ % 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}%
%
\ifvmode
\dosubindsanitize
\else
\dosubindwrite
\fi
}%
\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
- \escapechar=`\\
\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:
%
% 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 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}
%
% ..., ready, GO:
%
\def\dosubindsanitize{%
% \lastskip and \lastpenalty cannot both be nonzero simultaneously.
\skip0 = \lastskip
\edef\lastskipmacro{\the\lastskip}%
\count255 = \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-\skip0
\fi
%
\dosubindwrite
%
\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\count255>9999 \penalty\count255 \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\skip0
\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
\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 \dotfill except takes at least 1 em.
\def\indexdotfill{\cleaders
\hbox{$\mathsurround=0pt \mkern1.5mu ${\it .}$ \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{%
\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<toks register> to achieve this: TeX expands \the<toks> only once,
% simply yielding the contents of <toks register>. (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}%
- \def\thischapter{#1}%
+ \gdef\thischapter{#1}%
\else\ifx\temptype\Yomitfromtockeyword
\setbox0 = \hbox{}% contents like unnumbered, but no toc entry
\def\toctype{omit}%
- \xdef\thischapter{}%
+ \gdef\thischapter{}%
\else\ifx\temptype\Yappendixkeyword
\setbox0 = \hbox{\putwordAppendix{} #3\enspace}%
\def\toctype{app}%
% 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\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 \chfplain.
\writetocentry{\toctype\sectionlevel}{#1}{#4}%
%
% Write the node reference (= pdf destination for pdftex).
% Again, see comments in \chfplain.
\donoderef{#3}%
%
% 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
- \toks0 = {#2}%
- \toks2 = \expandafter{\lastnode}%
- \edef\temp{\write\tocfile{\realbackslash #1entry{\the\toks0}{#3}%
- {\the\toks2}{\noexpand\folio}}}%
- \temp
+ {\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 <tege@matematik.su.se>
\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.
- \catcode`\\=0 \catcode`\{=1 \catcode`\}=2 \catcode`\@=11
- % We can't do this, because then an actual ^ in a section
- % title fails, e.g., @chapter ^ -- exponentiation. --karl, 9jul97.
- %\catcode`\^=7 % to see ^^e4 as \"a etc. juha@piuha.ydi.vtt.fi
- \raggedbottom % Worry more about breakpoints than the bottom.
+ \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
- \input \jobname.toc
+ \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
- \input \jobname.toc
+ \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$\hfil}}
% The @error{} command.
% Adapted from the TeXbook's \boxit.
%
\newbox\errorbox
%
{\tentt \global\dimen0 = 3em}% Width of the box.
\dimen2 = .55pt % Thickness of rules
% The text. (`r' is open on the right, `e' somewhat less so on the left.)
\setbox0 = \hbox{\kern-.75pt \tensf error\kern-1.5pt}
%
\setbox\errorbox=\hbox to \dimen0{\hfil
\hsize = \dimen0 \advance\hsize by -5.8pt % Space to left+right.
\advance\hsize by -2\dimen2 % Rules.
\vbox{%
\hrule height\dimen2
\hbox{\vrule width\dimen2 \kern3pt % Space to left of text.
\vtop{\kern2.4pt \box0 \kern2.4pt}% Space above/below.
\kern3pt\vrule width\dimen2}% Space to right.
\hrule height\dimen2}
\hfil}
%
\def\error{\leavevmode\lower.7ex\copy\errorbox}
% @tex ... @end tex escapes into raw Tex temporarily.
% One exception: @ is still an escape character, so that @end tex works.
% But \@ or @@ will get a plain tex @ character.
\envdef\tex{%
\catcode `\\=0 \catcode `\{=1 \catcode `\}=2
\catcode `\$=3 \catcode `\&=4 \catcode `\#=6
\catcode `\^=7 \catcode `\_=8 \catcode `\~=\active \let~=\tie
\catcode `\%=14
\catcode `\+=\other
\catcode `\"=\other
\catcode `\|=\other
\catcode `\<=\other
\catcode `\>=\other
\escapechar=`\\
%
\let\b=\ptexb
\let\bullet=\ptexbullet
\let\c=\ptexc
\let\,=\ptexcomma
\let\.=\ptexdot
\let\dots=\ptexdots
\let\equiv=\ptexequiv
\let\!=\ptexexclam
\let\i=\ptexi
\let\indent=\ptexindent
\let\noindent=\ptexnoindent
\let\{=\ptexlbrace
\let\+=\tabalign
\let\}=\ptexrbrace
\let\/=\ptexslash
\let\*=\ptexstar
\let\t=\ptext
+ \let\frenchspacing=\plainfrenchspacing
%
\def\endldots{\mathinner{\ldots\ldots\ldots\ldots}}%
\def\enddots{\relax\ifmmode\endldots\else$\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.
+% \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=\comment
+ \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
- % @cartouche defines \nonarrowing to inhibit narrowing
- % at next level down.
\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
\smallexamplefonts \rm
\fi
}
\def\setsmalldispenv{%
\ifx\SETdispenvsize\nosmallword
\else
\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
\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{<char>...<char>}
% If we want to allow any <char> 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}
\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
}%
}
\endgroup
\def\setupverbatim{%
+ \let\nonarrowing = t%
\nonfillstart
- \advance\leftskip by -\defbodyindent
% Easiest (and conventionally used) font for verbatim
\tt
\def\par{\leavevmode\egroup\box0\endgraf}%
\catcode`\`=\active
\tabexpand
% 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'{'<char>#1<char>'}'{#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
% Start the processing of @deffn:
\def\startdefun{%
\ifnum\lastpenalty<10000
\medbreak
\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 \defargscommonending, 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.
%
\ifnum\lastpenalty=10002 \penalty2000 \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 \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 10002 % 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\&#1 }}
\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?
-\def\macrolist{} % List of all defined macros in the form
- % \do\macro1\do\macro2...
+
+% 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%
- % Add the macroname to \macrolist
- \toks0 = \expandafter{\macrolist\do}%
- \xdef\macrolist{\the\toks0
- \expandafter\noexpand\csname\the\macname\endcsname}%
+ \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\do\unmacrodo
+ \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
+ \ifx #1\relax
% remove this
\else
- \noexpand\do \noexpand #1%
+ \noexpand\definedummyword \noexpand#1%
\fi
}
% This makes use of the obscure feature that if the last token of a
% <parameter list> 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\next=#1\futurelet\nchar\braceorlinexxx}
\def\braceorlinexxx{%
\ifx\nchar\bgroup\else
\expandafter\parsearg
\fi \next}
-% We want to disable all macros during \shipout so that they are not
-% expanded by \write.
-\def\turnoffmacros{\begingroup \def\do##1{\let\noexpand##1=\relax}%
- \edef\next{\macrolist}\expandafter\endgroup\next}
-
-% For \indexnofonts, 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.
-%
-\def\emptyusermacros{\begingroup
- \def\do##1{\let\noexpand##1=\noexpand\asis}%
- \edef\next{\macrolist}\expandafter\endgroup\next}
-
% @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
- \turnoffactive
- \otherbackslash
\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.
\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}%
- {\turnoffactive \otherbackslash
+ {\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{#1}%
+ goto file{\the\filename.pdf} name{\pdfxrefdest}%
\else
\startlink attr{/Border [0 0 0]}%
- goto name{\pdfmkpgn{#1}}%
+ 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
- \otherbackslash
\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 \otherbackslash
+ {\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 \otherbackslash \putwordpage\tie\refx{#1-pg}{}%
+ \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{%
\expandafter\gdef\csname XR#1\endcsname{#2}% remember this xref value.
%
% Was that xref control sequence that we just defined for a float?
\expandafter\iffloat\csname XR#1\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{#1}}%
\fi
}
% Read the last existing aux file, if any. No error if none exists.
%
\def\tryauxfile{%
\openin 1 \jobname.aux
\ifeof 1 \else
- \readauxfile
+ \readdatafile{aux}%
\global\havexrefstrue
\fi
\closein 1
}
-\def\readauxfile{\begingroup
+\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.
{%
- \count 1=128
+ \count1=128
\def\loop{%
- \catcode\count 1=\other
- \advance\count 1 by 1
- \ifnum \count 1<256 \loop \fi
+ \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
- %
- \input \jobname.aux
-\endgroup}
+}
+\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\hss
\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 \hss \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 \turnoffactive \otherbackslash
+ \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 float, 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,}
% and i18n.
% @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.}
% @documentencoding should change something in TeX eventually, most
% likely, but for now just recognize it.
\let\documentencoding = \comment
% Page size parameters.
%
\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.
+% 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 7x9.5 (or so) format.
+% 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}
\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
% \rawbackslash defines an active \ to do \backslashcurfont.
% \otherbackslash defines an active \ to be a literal `\' character with
% catcode other.
{\catcode`\\=\active
@gdef@rawbackslash{@let\=@backslashcurfont}
@gdef@otherbackslash{@let\=@realbackslash}
}
-% \realbackslash is an actual character `\' with catcode other.
-{\catcode`\\=\other @gdef@realbackslash{\}}
+% \realbackslash is an actual character `\' with catcode other, and
+% \doublebackslash is two of them (for the pdf outlines).
+{\catcode`\\=\other @gdef@realbackslash{\} @gdef@doublebackslash{\\}}
% \normalbackslash outputs one backslash in fixed width font.
\def\normalbackslash{{\tt\backslashcurfont}}
\catcode`\\=\active
% Used sometimes to turn off (effectively) the active characters
% even after parsing them.
@def@turnoffactive{%
@let"=@normaldoublequote
@let\=@realbackslash
@let~=@normaltilde
@let^=@normalcaret
@let_=@normalunderscore
@let|=@normalverticalbar
@let<=@normalless
@let>=@normalgreater
@let+=@normalplus
@let$=@normaldollar %$ font-lock fix
@unsepspaces
}
% Same as @turnoffactive except outputs \ as {\tt\char`\\} instead of
% the literal character `\'. (Thus, \ is not expandable when this is in
% effect.)
%
@def@normalturnoffactive{@turnoffactive @let\=@normalbackslash}
% 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 back turn on active characters that might appear in the input
+% 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/install-sh b/install-sh
index 0b65ee87..4d4a9519 100755
--- a/install-sh
+++ b/install-sh
@@ -1,323 +1,323 @@
#!/bin/sh
# install - install a program, script, or datafile
-scriptversion=2004-10-22.00
+scriptversion=2005-05-14.22
# This originates from X11R5 (mit/util/scripts/install.sh), which was
# later released in X11R6 (xc/config/util/install.sh) with the
# following copyright and license.
#
# Copyright (C) 1994 X Consortium
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to
# deal in the Software without restriction, including without limitation the
# rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
# sell copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in
# all copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# X CONSORTIUM BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN
# AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNEC-
# TION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
#
# Except as contained in this notice, the name of the X Consortium shall not
# be used in advertising or otherwise to promote the sale, use or other deal-
# ings in this Software without prior written authorization from the X Consor-
# tium.
#
#
# FSF changes to this file are in the public domain.
#
# Calling this script install-sh is preferred over install.sh, to prevent
# `make' implicit rules from creating a file called install from it
# when there is no Makefile.
#
# This script is compatible with the BSD install script, but was written
# from scratch. It can only install one file at a time, a restriction
# shared with many OS's install programs.
# set DOITPROG to echo to test this script
# Don't use :- since 4.3BSD and earlier shells don't like it.
doit="${DOITPROG-}"
# put in absolute paths if you don't have them in your path; or use env. vars.
mvprog="${MVPROG-mv}"
cpprog="${CPPROG-cp}"
chmodprog="${CHMODPROG-chmod}"
chownprog="${CHOWNPROG-chown}"
chgrpprog="${CHGRPPROG-chgrp}"
stripprog="${STRIPPROG-strip}"
rmprog="${RMPROG-rm}"
mkdirprog="${MKDIRPROG-mkdir}"
chmodcmd="$chmodprog 0755"
chowncmd=
chgrpcmd=
stripcmd=
rmcmd="$rmprog -f"
mvcmd="$mvprog"
src=
dst=
dir_arg=
dstarg=
no_target_directory=
usage="Usage: $0 [OPTION]... [-T] SRCFILE DSTFILE
or: $0 [OPTION]... SRCFILES... DIRECTORY
or: $0 [OPTION]... -t DIRECTORY SRCFILES...
or: $0 [OPTION]... -d DIRECTORIES...
In the 1st form, copy SRCFILE to DSTFILE.
In the 2nd and 3rd, copy all SRCFILES to DIRECTORY.
In the 4th, create DIRECTORIES.
Options:
-c (ignored)
-d create directories instead of installing files.
-g GROUP $chgrpprog installed files to GROUP.
-m MODE $chmodprog installed files to MODE.
-o USER $chownprog installed files to USER.
-s $stripprog installed files.
-t DIRECTORY install into DIRECTORY.
-T report an error if DSTFILE is a directory.
--help display this help and exit.
--version display version info and exit.
Environment variables override the default commands:
CHGRPPROG CHMODPROG CHOWNPROG CPPROG MKDIRPROG MVPROG RMPROG STRIPPROG
"
while test -n "$1"; do
case $1 in
-c) shift
continue;;
-d) dir_arg=true
shift
continue;;
-g) chgrpcmd="$chgrpprog $2"
shift
shift
continue;;
- --help) echo "$usage"; exit 0;;
+ --help) echo "$usage"; exit $?;;
-m) chmodcmd="$chmodprog $2"
shift
shift
continue;;
-o) chowncmd="$chownprog $2"
shift
shift
continue;;
-s) stripcmd=$stripprog
shift
continue;;
-t) dstarg=$2
shift
shift
continue;;
-T) no_target_directory=true
shift
continue;;
- --version) echo "$0 $scriptversion"; exit 0;;
+ --version) echo "$0 $scriptversion"; exit $?;;
*) # When -d is used, all remaining arguments are directories to create.
# When -t is used, the destination is already specified.
test -n "$dir_arg$dstarg" && break
# Otherwise, the last argument is the destination. Remove it from $@.
for arg
do
if test -n "$dstarg"; then
# $@ is not empty: it contains at least $arg.
set fnord "$@" "$dstarg"
shift # fnord
fi
shift # arg
dstarg=$arg
done
break;;
esac
done
if test -z "$1"; then
if test -z "$dir_arg"; then
echo "$0: no input file specified." >&2
exit 1
fi
# It's OK to call `install-sh -d' without argument.
# This can happen when creating conditional directories.
exit 0
fi
for src
do
# Protect names starting with `-'.
case $src in
-*) src=./$src ;;
esac
if test -n "$dir_arg"; then
dst=$src
src=
if test -d "$dst"; then
mkdircmd=:
chmodcmd=
else
mkdircmd=$mkdirprog
fi
else
# Waiting for this to be detected by the "$cpprog $src $dsttmp" command
# might cause directories to be created, which would be especially bad
# if $src (and thus $dsttmp) contains '*'.
if test ! -f "$src" && test ! -d "$src"; then
echo "$0: $src does not exist." >&2
exit 1
fi
if test -z "$dstarg"; then
echo "$0: no destination specified." >&2
exit 1
fi
dst=$dstarg
# Protect names starting with `-'.
case $dst in
-*) dst=./$dst ;;
esac
# If destination is a directory, append the input filename; won't work
# if double slashes aren't ignored.
if test -d "$dst"; then
if test -n "$no_target_directory"; then
echo "$0: $dstarg: Is a directory" >&2
exit 1
fi
dst=$dst/`basename "$src"`
fi
fi
# This sed command emulates the dirname command.
dstdir=`echo "$dst" | sed -e 's,/*$,,;s,[^/]*$,,;s,/*$,,;s,^$,.,'`
# Make sure that the destination directory exists.
# Skip lots of stat calls in the usual case.
if test ! -d "$dstdir"; then
defaultIFS='
'
IFS="${IFS-$defaultIFS}"
oIFS=$IFS
# Some sh's can't handle IFS=/ for some reason.
IFS='%'
set x `echo "$dstdir" | sed -e 's@/@%@g' -e 's@^%@/@'`
shift
IFS=$oIFS
pathcomp=
while test $# -ne 0 ; do
pathcomp=$pathcomp$1
shift
if test ! -d "$pathcomp"; then
$mkdirprog "$pathcomp"
# mkdir can fail with a `File exist' error in case several
# install-sh are creating the directory concurrently. This
# is OK.
test -d "$pathcomp" || exit
fi
pathcomp=$pathcomp/
done
fi
if test -n "$dir_arg"; then
$doit $mkdircmd "$dst" \
&& { test -z "$chowncmd" || $doit $chowncmd "$dst"; } \
&& { test -z "$chgrpcmd" || $doit $chgrpcmd "$dst"; } \
&& { test -z "$stripcmd" || $doit $stripcmd "$dst"; } \
&& { test -z "$chmodcmd" || $doit $chmodcmd "$dst"; }
else
dstfile=`basename "$dst"`
# Make a couple of temp file names in the proper directory.
dsttmp=$dstdir/_inst.$$_
rmtmp=$dstdir/_rm.$$_
# Trap to clean up those temp files at exit.
trap 'ret=$?; rm -f "$dsttmp" "$rmtmp" && exit $ret' 0
trap '(exit $?); exit' 1 2 13 15
# Copy the file name to the temp name.
$doit $cpprog "$src" "$dsttmp" &&
# and set any options; do chmod last to preserve setuid bits.
#
# If any of these fail, we abort the whole thing. If we want to
# ignore errors from any of these, just make sure not to ignore
# errors from the above "$doit $cpprog $src $dsttmp" command.
#
{ test -z "$chowncmd" || $doit $chowncmd "$dsttmp"; } \
&& { test -z "$chgrpcmd" || $doit $chgrpcmd "$dsttmp"; } \
&& { test -z "$stripcmd" || $doit $stripcmd "$dsttmp"; } \
&& { test -z "$chmodcmd" || $doit $chmodcmd "$dsttmp"; } &&
# Now rename the file to the real destination.
{ $doit $mvcmd -f "$dsttmp" "$dstdir/$dstfile" 2>/dev/null \
|| {
# The rename failed, perhaps because mv can't rename something else
# to itself, or perhaps because mv is so ancient that it does not
# support -f.
# Now remove or move aside any old file at destination location.
# We try this two ways since rm can't unlink itself on some
# systems and the destination file might be busy for other
# reasons. In this case, the final cleanup might fail but the new
# file should still install successfully.
{
if test -f "$dstdir/$dstfile"; then
$doit $rmcmd -f "$dstdir/$dstfile" 2>/dev/null \
|| $doit $mvcmd -f "$dstdir/$dstfile" "$rmtmp" 2>/dev/null \
|| {
echo "$0: cannot unlink or rename $dstdir/$dstfile" >&2
- (exit 1); exit
+ (exit 1); exit 1
}
else
:
fi
} &&
# Now rename the file to the real destination.
$doit $mvcmd "$dsttmp" "$dstdir/$dstfile"
}
}
- fi || { (exit 1); exit; }
+ fi || { (exit 1); exit 1; }
done
# The final little trick to "correctly" pass the exit status to the exit trap.
{
- (exit 0); exit
+ (exit 0); exit 0
}
# Local variables:
# eval: (add-hook 'write-file-hooks 'time-stamp)
# time-stamp-start: "scriptversion="
# time-stamp-format: "%:y-%02m-%02d.%02H"
# time-stamp-end: "$"
# End:
diff --git a/missing b/missing
index 64b5f901..894e786e 100755
--- a/missing
+++ b/missing
@@ -1,353 +1,360 @@
#! /bin/sh
# Common stub for a few missing GNU programs while installing.
-scriptversion=2004-09-07.08
+scriptversion=2005-06-08.21
-# Copyright (C) 1996, 1997, 1999, 2000, 2002, 2003, 2004
+# Copyright (C) 1996, 1997, 1999, 2000, 2002, 2003, 2004, 2005
# Free Software Foundation, Inc.
# Originally by Fran,cois Pinard <pinard@iro.umontreal.ca>, 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, write to the Free Software
-# Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA
-# 02111-1307, USA.
+# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+# 02110-1301, USA.
# 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
run=:
# In the cases where this matters, `missing' is being run in the
# srcdir already.
if test -f configure.ac; then
configure_ac=configure.ac
else
configure_ac=configure.in
fi
msg="missing on your system"
case "$1" in
--run)
# Try to run requested program, and just exit if it succeeds.
run=
shift
"$@" && exit 0
# Exit code 63 means version mismatch. This often happens
# when the user try to use an ancient version of a tool on
# a file that requires a minimum version. In this case we
# we should proceed has if the program had been absent, or
# if --run hadn't been passed.
if test $? = 63; then
run=:
msg="probably too old"
fi
;;
-h|--h|--he|--hel|--help)
echo "\
$0 [OPTION]... PROGRAM [ARGUMENT]...
Handle \`PROGRAM [ARGUMENT]...' for when PROGRAM is missing, or return an
error status if there is no known handling for PROGRAM.
Options:
-h, --help display this help and exit
-v, --version output version information and exit
--run try to run the given command, and emulate it if it fails
Supported PROGRAM values:
aclocal touch file \`aclocal.m4'
autoconf touch file \`configure'
autoheader touch file \`config.h.in'
automake touch all \`Makefile.in' files
bison create \`y.tab.[ch]', if possible, from existing .[ch]
flex create \`lex.yy.c', if possible, from existing .c
help2man touch the output file
lex create \`lex.yy.c', if possible, from existing .c
makeinfo touch the output file
tar try tar, gnutar, gtar, then tar without non-portable flags
yacc create \`y.tab.[ch]', if possible, from existing .[ch]
Send bug reports to <bug-automake@gnu.org>."
- exit 0
+ exit $?
;;
-v|--v|--ve|--ver|--vers|--versi|--versio|--version)
echo "missing $scriptversion (GNU Automake)"
- exit 0
+ exit $?
;;
-*)
echo 1>&2 "$0: Unknown \`$1' option"
echo 1>&2 "Try \`$0 --help' for more information"
exit 1
;;
esac
# Now exit if we have it, but it failed. Also exit now if we
# don't have it and --version was passed (most likely to detect
# the program).
case "$1" in
lex|yacc)
# Not GNU programs, they don't have --version.
;;
tar)
if test -n "$run"; then
echo 1>&2 "ERROR: \`tar' requires --run"
exit 1
elif test "x$2" = "x--version" || test "x$2" = "x--help"; then
exit 1
fi
;;
*)
if test -z "$run" && ($1 --version) > /dev/null 2>&1; then
# We have it, but it failed.
exit 1
elif test "x$2" = "x--version" || test "x$2" = "x--help"; then
# Could not run --version or --help. This is probably someone
# running `$TOOL --version' or `$TOOL --help' to check whether
# $TOOL exists and not knowing $TOOL uses missing.
exit 1
fi
;;
esac
# If it does not exist, or fails to run (possibly an outdated version),
# try to emulate it.
case "$1" in
aclocal*)
echo 1>&2 "\
WARNING: \`$1' is $msg. You should only need it if
you modified \`acinclude.m4' or \`${configure_ac}'. You might want
to install the \`Automake' and \`Perl' packages. Grab them from
any GNU archive site."
touch aclocal.m4
;;
autoconf)
echo 1>&2 "\
WARNING: \`$1' is $msg. You should only need it if
you modified \`${configure_ac}'. You might want to install the
\`Autoconf' and \`GNU m4' packages. Grab them from any GNU
archive site."
touch configure
;;
autoheader)
echo 1>&2 "\
WARNING: \`$1' is $msg. You should only need it if
you modified \`acconfig.h' or \`${configure_ac}'. You might want
to install the \`Autoconf' and \`GNU m4' packages. Grab them
from any GNU archive site."
files=`sed -n 's/^[ ]*A[CM]_CONFIG_HEADER(\([^)]*\)).*/\1/p' ${configure_ac}`
test -z "$files" && files="config.h"
touch_files=
for f in $files; do
case "$f" in
*:*) touch_files="$touch_files "`echo "$f" |
sed -e 's/^[^:]*://' -e 's/:.*//'`;;
*) touch_files="$touch_files $f.in";;
esac
done
touch $touch_files
;;
automake*)
echo 1>&2 "\
WARNING: \`$1' is $msg. You should only need it if
you modified \`Makefile.am', \`acinclude.m4' or \`${configure_ac}'.
You might want to install the \`Automake' and \`Perl' packages.
Grab them from any GNU archive site."
find . -type f -name Makefile.am -print |
sed 's/\.am$/.in/' |
while read f; do touch "$f"; done
;;
autom4te)
echo 1>&2 "\
WARNING: \`$1' is needed, but is $msg.
You might have modified some files without having the
proper tools for further handling them.
You can get \`$1' as part of \`Autoconf' from any GNU
archive site."
file=`echo "$*" | sed -n 's/.*--output[ =]*\([^ ]*\).*/\1/p'`
test -z "$file" && file=`echo "$*" | sed -n 's/.*-o[ ]*\([^ ]*\).*/\1/p'`
if test -f "$file"; then
touch $file
else
test -z "$file" || exec >$file
echo "#! /bin/sh"
echo "# Created by GNU Automake missing as a replacement of"
echo "# $ $@"
echo "exit 0"
chmod +x $file
exit 1
fi
;;
bison|yacc)
echo 1>&2 "\
WARNING: \`$1' $msg. You should only need it if
you modified a \`.y' file. You may need the \`Bison' package
in order for those modifications to take effect. You can get
\`Bison' from any GNU archive site."
rm -f y.tab.c y.tab.h
if [ $# -ne 1 ]; then
eval LASTARG="\${$#}"
case "$LASTARG" in
*.y)
SRCFILE=`echo "$LASTARG" | sed 's/y$/c/'`
if [ -f "$SRCFILE" ]; then
cp "$SRCFILE" y.tab.c
fi
SRCFILE=`echo "$LASTARG" | sed 's/y$/h/'`
if [ -f "$SRCFILE" ]; then
cp "$SRCFILE" y.tab.h
fi
;;
esac
fi
if [ ! -f y.tab.h ]; then
echo >y.tab.h
fi
if [ ! -f y.tab.c ]; then
echo 'main() { return 0; }' >y.tab.c
fi
;;
lex|flex)
echo 1>&2 "\
WARNING: \`$1' is $msg. You should only need it if
you modified a \`.l' file. You may need the \`Flex' package
in order for those modifications to take effect. You can get
\`Flex' from any GNU archive site."
rm -f lex.yy.c
if [ $# -ne 1 ]; then
eval LASTARG="\${$#}"
case "$LASTARG" in
*.l)
SRCFILE=`echo "$LASTARG" | sed 's/l$/c/'`
if [ -f "$SRCFILE" ]; then
cp "$SRCFILE" lex.yy.c
fi
;;
esac
fi
if [ ! -f lex.yy.c ]; then
echo 'main() { return 0; }' >lex.yy.c
fi
;;
help2man)
echo 1>&2 "\
WARNING: \`$1' is $msg. You should only need it if
you modified a dependency of a manual page. You may need the
\`Help2man' package in order for those modifications to take
effect. You can get \`Help2man' from any GNU archive site."
file=`echo "$*" | sed -n 's/.*-o \([^ ]*\).*/\1/p'`
if test -z "$file"; then
file=`echo "$*" | sed -n 's/.*--output=\([^ ]*\).*/\1/p'`
fi
if [ -f "$file" ]; then
touch $file
else
test -z "$file" || exec >$file
echo ".ab help2man is required to generate this page"
exit 1
fi
;;
makeinfo)
echo 1>&2 "\
WARNING: \`$1' is $msg. You should only need it if
you modified a \`.texi' or \`.texinfo' file, or any other file
indirectly affecting the aspect of the manual. The spurious
call might also be the consequence of using a buggy \`make' (AIX,
DU, IRIX). You might want to install the \`Texinfo' package or
the \`GNU make' package. Grab either from any GNU archive site."
+ # The file to touch is that specified with -o ...
file=`echo "$*" | sed -n 's/.*-o \([^ ]*\).*/\1/p'`
if test -z "$file"; then
- file=`echo "$*" | sed 's/.* \([^ ]*\) *$/\1/'`
- file=`sed -n '/^@setfilename/ { s/.* \([^ ]*\) *$/\1/; p; q; }' $file`
+ # ... or it is the one specified with @setfilename ...
+ infile=`echo "$*" | sed 's/.* \([^ ]*\) *$/\1/'`
+ file=`sed -n '/^@setfilename/ { s/.* \([^ ]*\) *$/\1/; p; q; }' $infile`
+ # ... or it is derived from the source name (dir/f.texi becomes f.info)
+ test -z "$file" && file=`echo "$infile" | sed 's,.*/,,;s,.[^.]*$,,'`.info
fi
+ # If the file does not exist, the user really needs makeinfo;
+ # let's fail without touching anything.
+ test -f $file || exit 1
touch $file
;;
tar)
shift
# We have already tried tar in the generic part.
# Look for gnutar/gtar before invocation to avoid ugly error
# messages.
if (gnutar --version > /dev/null 2>&1); then
gnutar "$@" && exit 0
fi
if (gtar --version > /dev/null 2>&1); then
gtar "$@" && exit 0
fi
firstarg="$1"
if shift; then
case "$firstarg" in
*o*)
firstarg=`echo "$firstarg" | sed s/o//`
tar "$firstarg" "$@" && exit 0
;;
esac
case "$firstarg" in
*h*)
firstarg=`echo "$firstarg" | sed s/h//`
tar "$firstarg" "$@" && exit 0
;;
esac
fi
echo 1>&2 "\
WARNING: I can't seem to be able to run \`tar' with the given arguments.
You may want to install GNU tar or Free paxutils, or check the
command line arguments."
exit 1
;;
*)
echo 1>&2 "\
WARNING: \`$1' is needed, and is $msg.
You might have modified some files without having the
proper tools for further handling them. Check the \`README' file,
it often tells you about the needed prerequisites for installing
this package. You may also peek at any GNU archive site, in case
some other package would contain this missing \`$1' program."
exit 1
;;
esac
exit 0
# Local variables:
# eval: (add-hook 'write-file-hooks 'time-stamp)
# time-stamp-start: "scriptversion="
# time-stamp-format: "%:y-%02m-%02d.%02H"
# time-stamp-end: "$"
# End:
diff --git a/mkinstalldirs b/mkinstalldirs
index 6fbe5e11..259dbfcd 100755
--- a/mkinstalldirs
+++ b/mkinstalldirs
@@ -1,150 +1,158 @@
#! /bin/sh
# mkinstalldirs --- make directory hierarchy
-scriptversion=2004-02-15.20
+scriptversion=2005-06-29.22
# Original author: Noah Friedman <friedman@prep.ai.mit.edu>
# Created: 1993-05-16
# Public domain.
#
# This file is maintained in Automake, please report
# bugs to <bug-automake@gnu.org> or send patches to
# <automake-patches@gnu.org>.
errstatus=0
-dirmode=""
+dirmode=
usage="\
Usage: mkinstalldirs [-h] [--help] [--version] [-m MODE] DIR ...
Create each directory DIR (with mode MODE, if specified), including all
leading file name components.
Report bugs to <bug-automake@gnu.org>."
# process command line arguments
while test $# -gt 0 ; do
case $1 in
-h | --help | --h*) # -h for help
echo "$usage"
- exit 0
+ exit $?
;;
-m) # -m PERM arg
shift
test $# -eq 0 && { echo "$usage" 1>&2; exit 1; }
dirmode=$1
shift
;;
--version)
echo "$0 $scriptversion"
- exit 0
+ exit $?
;;
--) # stop option processing
shift
break
;;
-*) # unknown option
echo "$usage" 1>&2
exit 1
;;
*) # first non-opt arg
break
;;
esac
done
for file
do
if test -d "$file"; then
shift
else
break
fi
done
case $# in
0) exit 0 ;;
esac
# Solaris 8's mkdir -p isn't thread-safe. If you mkdir -p a/b and
# mkdir -p a/c at the same time, both will detect that a is missing,
# one will create a, then the other will try to create a and die with
# a "File exists" error. This is a problem when calling mkinstalldirs
# from a parallel make. We use --version in the probe to restrict
# ourselves to GNU mkdir, which is thread-safe.
case $dirmode in
'')
if mkdir -p --version . >/dev/null 2>&1 && test ! -d ./--version; then
echo "mkdir -p -- $*"
exec mkdir -p -- "$@"
else
# On NextStep and OpenStep, the `mkdir' command does not
# recognize any option. It will interpret all options as
# directories to create, and then abort because `.' already
# exists.
test -d ./-p && rmdir ./-p
test -d ./--version && rmdir ./--version
fi
;;
*)
if mkdir -m "$dirmode" -p --version . >/dev/null 2>&1 &&
test ! -d ./--version; then
echo "mkdir -m $dirmode -p -- $*"
exec mkdir -m "$dirmode" -p -- "$@"
else
# Clean up after NextStep and OpenStep mkdir.
for d in ./-m ./-p ./--version "./$dirmode";
do
test -d $d && rmdir $d
done
fi
;;
esac
for file
do
- set fnord `echo ":$file" | sed -ne 's/^:\//#/;s/^://;s/\// /g;s/^#/\//;p'`
+ case $file in
+ /*) pathcomp=/ ;;
+ *) pathcomp= ;;
+ esac
+ oIFS=$IFS
+ IFS=/
+ set fnord $file
shift
+ IFS=$oIFS
- pathcomp=
for d
do
- pathcomp="$pathcomp$d"
+ test "x$d" = x && continue
+
+ pathcomp=$pathcomp$d
case $pathcomp in
-*) pathcomp=./$pathcomp ;;
esac
if test ! -d "$pathcomp"; then
echo "mkdir $pathcomp"
mkdir "$pathcomp" || lasterr=$?
if test ! -d "$pathcomp"; then
errstatus=$lasterr
else
if test ! -z "$dirmode"; then
echo "chmod $dirmode $pathcomp"
- lasterr=""
+ lasterr=
chmod "$dirmode" "$pathcomp" || lasterr=$?
if test ! -z "$lasterr"; then
errstatus=$lasterr
fi
fi
fi
fi
- pathcomp="$pathcomp/"
+ pathcomp=$pathcomp/
done
done
exit $errstatus
# Local Variables:
# mode: shell-script
# sh-indentation: 2
# eval: (add-hook 'write-file-hooks 'time-stamp)
# time-stamp-start: "scriptversion="
# time-stamp-format: "%:y-%02m-%02d.%02H"
# time-stamp-end: "$"
# End:
diff --git a/mpi/ChangeLog b/mpi/ChangeLog
index 2ee49814..a3888c59 100644
--- a/mpi/ChangeLog
+++ b/mpi/ChangeLog
@@ -1,605 +1,615 @@
+2006-08-28 Werner Koch <wk@g10code.com>
+
+ * mpiutil.c (_gcry_mpi_resize): Clear the new part of the resized
+ limb space.
+ * mpi-bit.c (gcry_mpi_set_bit, gcry_mpi_set_highbit): Fixed
+ allocation. Reported by bpgcrypt at itaparica.org.
+
+ * mpi-bit.c (gcry_mpi_rshift): Rewritten to remove the limitation
+ on N (which used to be less than BITS_PER_MPI_LIMB).
+
2005-09-14 Werner Koch <wk@g10code.com>
* Makefile.am (AM_CCASFLAGS): Add AM_CCASFLAGS.
2005-06-16 Werner Koch <wk@g10code.com>
* mpiutil.c (gcry_mpi_randomize): Made P unsigned.
* mpicoder.c (gcry_mpi_scan): Changed arg BUFFER to void*.
(mpi_read_from_buffer): Made BUFFER arg const.
(gcry_mpi_scan): Removed now needless cast. Add cast for arg to
mpi_fromstr.
(gcry_mpi_print): Made TMP unsigned.
2005-05-29 Moritz Schulte <moritz@g10code.com>
* mpiutil.c (gcry_mpi_randomize): Store random data in secure
memory if the given MPI is secure - not the other way around (argl).
2005-02-16 Moritz Schulte <moritz@g10code.com>
* mpiutil.c (_gcry_mpi_alloc_limb_space): Rewritten, fixed memory
corruption.
2005-01-05 Werner Koch <wk@g10code.com>
* hppa1.1/udiv-qrnnd.S: Reverted change of 2004-03-02 but kept the
.align directive.
2004-12-16 Werner Koch <wk@g10code.com>
* config.links (mpi_optional_modules): Move entry for powerpc64
before generic powerpc. Suggested by Rafael ﾃ」ila de Espﾃｭndola.
2004-03-02 Werner Koch <wk@gnupg.org>
* hppa1.1/udiv-qrnnd.S: Alignment fix from Lamont Jones for
Debian. Taken from gnupg-1.3.
* longlong.h: Added PowerPC 64 bit code from GPM-4.1.2 but didn't
enable it yet. Some whitespace changes in HPPA to fix assembler
problems on HP-UX. From gnupg 1.3
* mpiutil.c (_gcry_mpi_alloc_limb_space): Better allocate
something even if NLIMBS is passed as 0.
* config.links: Updated system list to match gnupg 1.3.
2003-12-19 Werner Koch <wk@gnupg.org>
* mpi-internal.h [M_DEBUG]: Removed this unused code.
(struct karatsuba_ctx): Added TSPACE_NLIMBS and TP_NLIMBS.
* mpiutil.c (_gcry_mpi_free_limb_space): Add arg NLIMBS and wipe
out the memory. Changed all callers.
* mpih-mul.c (_gcry_mpih_mul_karatsuba_case): Keep track of
allocated limbs.
* mpi-div.c (_gcry_mpi_tdiv_qr): Keep track of allocated limbs.
* mpi-mul.c (gcry_mpi_mul): Ditto.
* mpi-pow.c (gcry_mpi_powm): Ditto.
* Manifest: Empty new file. Also add Manifest files to all CPU
specific directories.
* Makefile.am: Added.
* mpiutil.c (gcry_mpi_randomize): Use gcry_create_nonce if WEAK
random has been requested.
2003-10-31 Werner Koch <wk@gnupg.org>
* i386/mpih-rshift.S, i386/mpih-lshift.S: Use %dl and not %edx for
testb; this avoids an assembler warning.
* mpi-pow.c (gcry_mpi_powm): s/exp/expo/ to avoid shadowing warning.
2003-08-19 Marcus Brinkmann <marcus@g10code.de>
* Makefile.am (SUFFIXES): New variable.
(.S.o, .S.lo, .S.obj): Rewritten.
2003-07-30 Moritz Schulte <moritz@g10code.com>
* longlong.h (__clz_tab): Renamed to _gcry_clz_tab.
* mpi-bit.c (__clz_tab): Likewise.
2003-07-27 Werner Koch <wk@gnupg.org>
* mpicoder.c (gcry_mpi_scan): New argument BUFLEN to replace the
use of the intial value of NBYTES. Changed BUFFER to unsigned.
(gcry_mpi_print): Likewise.
(gcry_mpi_dump): New.
(_gcry_log_mpidump): Make use of gcry_mpi_dump.
(mpi_print): Removed.
(gcry_mpi_scan): Allocated mpi in secure memory when required.
(gcry_mpi_aprint): Changed BUFFER to unsigned char*.
2003-07-14 Moritz Schulte <moritz@g10code.com>
* mpicoder.c: Used gcry_err* wrappers for libgpg-error symbols.
2003-06-16 Moritz Schulte <moritz@g10code.com>
* mpi-add.c: Replace last occurences of old type names with newer
names (i.e. replace MPI with gcry_mpi_t).
* mpi-bit.c: Likewise.
* mpi-cmp.c: Likewise.
* mpi-div.c: Likewise.
* mpi-gcd.c: Likewise.
* mpi-internal.h: Likewise.
* mpi-inv.c: Likewise.
* mpi-mpow.c: Likewise.
* mpi-mul.c: Likewise.
* mpi-pow.c: Likewise.
* mpi-scan.c: Likewise.
* mpicoder.c: Likewise.
* mpiutil.c: Likewise.
2003-06-09 Moritz Schulte <moritz@g10code.com>
* mpicoder.c (gcry_mpi_scan): Adjust for libgpg-error.
(gcry_mpi_print): Likewise.
(gcry_mpi_aprint): Likewise.
2003-06-07 Moritz Schulte <moritz@g10code.com>
* longlong.h, mpi-add.c, mpi-bit.c, mpi-cmp.c, mpi-div.c,
mpi-gcd.c, mpi-inline.c, mpi-inline.h, mpi-internal.h, mpi-inv.c,
mpi-mpow.c, mpi-mul.c, mpi-pow.c, mpi-scan.c, mpicoder.c,
mpih-div.c, mpih-mul.c, mpiutil.c, generic/mpi-asm-defs.h,
generic/mpih-add1.c, generic/mpih-lshift.c, generic/mpih-mul1.c,
generic/mpih-mul2.c, generic/mpih-mul3.c, generic/mpih-rshift.c,
generic/mpih-sub1.c, generic/udiv-w-sdiv.c, i386/syntax.h,
m68k/syntax.h, mips3/mpi-asm-defs.h, powerpc32/syntax.h: Edited
all preprocessor instructions to remove whitespace before the '#'.
This is not required by C89, but there are some compilers out
there that don't like it. Replaced any occurence of the now
deprecated type names with the new ones.
2003-05-21 Moritz Schulte <moritz@g10code.com>
* mpiutil.c (_gcry_mpi_alloc_limb_space): Only try to allocate
memory in case the amount of bytes to allocate is non-zero.
2003-04-27 Moritz Schulte <moritz@g10code.com>
* mpiutil.c (_gcry_mpi_resize): Allocate secure memory, in case
bit zero of `flags' is set.
* mpi-add.c (gcry_mpi_sub): Simplify function; always use a
temporary variable now.
2003-04-15 Werner Koch <wk@gnupg.org>
* longlong.h (umul_ppmm): Support SH3 and SH4. Thanks to
kazuya.s@jp.yokogawa.com.
2003-04-02 Werner Koch <wk@gnupg.org>
* mpicoder.c (gcry_mpi_print): Fixed testing against possible
uninitialized LEN. Valgrinded by Nikos Mavroyanopoulos.
2003-01-15 Werner Koch <wk@gnupg.org>
* longlong.h: Removed some spaces between backslashes and newlines.
2002-09-20 Werner Koch <wk@gnupg.org>
* mpi-mul.c (gcry_mpi_mul_2exp): New. This was declared in
gcrypt.h but only implemented as internal function. Noted by Timo
but a few minutes to late for today's release.
* Makefile.am (DISTCLEANFILES): Include mpi-asm-defs.h
2002-09-18 Werner Koch <wk@gnupg.org>
* Makefile.am (.S.lo): Pass -DPIC. i386, PPC and Sparc code
require it. It worked for me because I am using the i586 code.
2002-08-23 Werner Koch <wk@gnupg.org>
* Makefile.am (.S.lo): Fixed for libtool build with --disable-shared.
2002-07-24 Werner Koch <wk@gnupg.org>
* longlong.h: Replaced all K&R multiline strings by ISO ones for
the sake of modern compilers. Suggested by Marco Parrone.
2002-06-24 Werner Koch <wk@gnupg.org>
* mpiutil.c (gcry_mpi_swap): New.
* mpi-div.c (gcry_mpi_div): New.
(gcry_mpi_mod): New.
* mpi-inv.c (gcry_mpi_invm): New.
* mpicoder.c (do_get_buffer): Make sure that we allocate at least
one byte.
2002-06-12 Werner Koch <wk@gnupg.org>
* hppa1.1/udiv-qrnnd.S: Changes for PIC by Randolph Chung.
2002-05-15 Werner Koch <wk@gnupg.org>
* config.links: Chnage the way the mpi modules are determined.
* Makefile.am: Revamped to better handle modules
2002-05-14 Werner Koch <wk@gnupg.org>
Changed license of all files to the LGPL.
2002-04-18 Werner Koch <wk@gnupg.org>
* mpicoder.c (gcry_mpi_scan): Don't use normalize on a NULL MPI.
2002-03-20 Werner Koch <wk@gnupg.org>
* mpicoder.c (mpi_read_from_buffer): Bail out on a zero length
buffer because we can't eventually do an malloc of this size.
Reported by Timo.
2002-01-14 Werner Koch <wk@gnupg.org>
* mpi-inv.c (_gcry_mpi_invm): Typo fixes, noted by Carlo Perassi.
2001-11-01 Werner Koch <wk@gnupg.org>
* mpicoder.c (gcry_mpi_scan): Allow to pass a nbytes as NULL or
with value 0 for format GCRY_FMT_SSH, so that the length is not
used for any checks, only the length stored in the bufer is used.
This is a nice format becuase we can just pass a buffer around and
don't need to care about its length.
2001-08-03 Werner Koch <wk@gnupg.org>
* config.links: Changed the way the list of files to be
symlinked is returned.
2001-05-31 Werner Koch <wk@gnupg.org>
* mpih-cmp.c: Removed and moved mpihelp_cmp to ..
* mpi-inline.h: .. here.
Major function renaming. All global functions are now prefixed
with _gcry_ or gcry_. Renamed also all mpihelp_ to just mpih_ so
that functions names are not getting to long an unreadable and for
better matching with the filenames.
2001-05-28 Werner Koch <wk@gnupg.org>
* mpicoder.c (mpi_fromstr): Made static and assume that all input
is in hexformat.
Updated all CPU specific code with the one from GnuPG-1.0.5. This
is just a change of text formatting and the use of .label
instead of labels for hppa and pa7100.
* longlong.h: Fixes for ARM by Phil Blundell.
2001-03-29 Werner Koch <wk@gnupg.org>
* mpi-mul.c (mpi_mul): Make sure that secret temporary results are
not stored in w. Suggested by Florian Weimer.
* config.links: Use i386 code for i386. According to tests by
Kevin Ryde the i586 code runs slow on i386 CPUs. Ditto for i786.
2001-01-11 Werner Koch <wk@gnupg.org>
* Makefile.am: Removed mpi.h.
2000-12-19 Werner Koch <wk@gnupg.org>
* mpi-internal.h: Put limb_t definition in an ifdef.
Major change:
Removed all GnuPG stuff and renamed this piece of software
to gcrypt.
2000-11-14 Werner Koch <wk@gnupg.org>
* mpi-internal.h, mpi.h: Changed the way they are called and
introduced DID_MPI_LIMP_TYPEDEF hack. Very ugly, should all be
revamped.
* Makefile.am (OMIT_DEPENDENCIES): Hack to work around dependency
problems.
2000-10-11 Werner Koch <wk@gnupg.org>
* generic/mpi-asm-defs.h: New.
* mips3/mpi-asm-defs.h: New.
* config.links: Create a link to one of the above files.
Fri Jul 28 18:19:11 CEST 2000 Werner Koch <wk@openit.de>
* mpicoder.c (gcry_mpi_scan): Normalize the returned MPI.
Tue Jul 25 17:44:15 CEST 2000 Werner Koch <wk@openit.de>
* config.links: Support for powerpc--netbsd by Gabriel Rosenkoetter.
Mon Jul 17 16:35:47 CEST 2000 Werner Koch <wk@>
* power/: Add all files from GMP for this CPU. Converted comments to
CPP comments because some ASes complain about ' in comments.
* config.links: Support for BSDI 4.x; by Wayne Chapeskie. Add support
for FreeBSD 5 and made the case stmt looking nicer; by Jun Kuriyama.
Add support for NetBSD.
(sparc8): Made the search path the same as sparc9
(sparc64-unknown-linux-gnu): use udiv module; by Adam Mitchell.
* Makefile.am: c/SFLAGS/ASFLAGS/. This has only been used by the
powerpc and actually never passed the -Wa,foo to the cc.
* mpih-div.c (mpihelp_divrem): The MPN_COPY_DECR copied one element
too many. This is a gmp2.0.2p9.txt patch.
* longlong.h (umul_ppmm): Fixes for ARM-4. By Sean MacLennan.
* mpi-internal.h (karatsuba_ctx): New.
* mpih-mul.c (mpihelp_release_karatsuba_ctx): New.
(mpihelp_mul_karatsuba_case): New.
(mpihelp_mul): Splitted to make use of the new functions.
* mpi-pow.c (mpi_powm): Make use of the new splitted function to avoid
multiple allocation of temporary memory during the karatsuba operations.
* mpi_mpow.c: Removed the unused Barrett code.
2000-03-21 16:17:30 Werner Koch (wk@habibti.openit.de)
* config.links: Add support for FreeBSD 5.
Mon Jan 24 22:24:38 CET 2000 Werner Koch <wk@gnupg.de>
* mpicoder.c (gcry_mpi_aprint): Now really returns the length.
Mon Jan 24 13:04:28 CET 2000 Werner Koch <wk@gnupg.de>
* mpiutil.c: Removed all memory debugging code.
* mpicoder.c (gcry_mpi_aprint): New.
* Replaced all m_ memory functions by g10_ ones.
Fri Dec 31 14:06:56 CET 1999 Werner Koch <wk@gnupg.de>
* mpi-bit.c (gcry_mpi_get_nbits): New.
* mpiutil.c (mpi_set_secure): made static.
(gcry_mpi_get_flag): New.
(gcry_mpi_set_flag): New.
(gcry_mpi_clear_flag): New.
(mpi_set_opaque): renamed to gcry_mpi_set_opaque.
(mpi_get_opaque): renamed to gcry_mpi_get_opaque.
Fri Dec 31 12:48:31 CET 1999 Werner Koch <wk@gnupg.de>
* mpicoder.c (mpi_read_from_buffer): Made static.
(gcry_mpi_print): A buffer of NULL is now allowed to get the required
length back.
(mpi_get_keyid): Removed.
(mpi_print): Made static - should be removed.
Wed Dec 8 21:58:32 CET 1999 Werner Koch <wk@gnupg.de>
* Makefile.am (INCLUDES): Add ../gcrypt.
* g10m.c : Removed.
* mpicoder.c (mpi_write): Removed.
(mpi_read): Removed.
(gcry_mpi_scan): New. Taken from ../gcrypt/mpiapi.c.
(gcry_mpi_print): Ditto.
* mpi-pow.c (mpi_powm): Renamed to ...
(gcry_mpi_powm): ... this.
* mpiutil.c (gcry_mpi_new): New as a wrapper around the old function.
Taken from ../gcrypt/mpiapi.c.
(gcry_mpi_snew): Ditto.
(gcry_mpi_release): Ditto.
(gcry_mpi_copy): Ditto.
(gcry_mpi_set): Ditto.
(gcry_mpi_set_ui): Ditto.
(gcry_mpi_cmp): Ditto.
(gcry_mpi_cmp_ui): Ditto.
(gcry_mpi_randomize): Ditto.
* mpicoder.c (mpi_print): Removed the nbit_info kludge.
* mpi-bits.c (mpi_get_nbits): Replaced the is_protected stuff by
checking whether it is an opaque mpi and then returns it's length
in bits.
* mpiutil.c (mpi_set_opaque): Changed the interface to take a number
of bits for the length. Adjusted all users.
(mpi_get_opaque): Ditto.
Fri Nov 19 17:15:20 CET 1999 Werner Koch <wk@gnupg.de>
* mpicoder.c (g10_log_mpidump): Add a temporary workaround
* mpih-mul.c (mpihelp_mul_n): s/m_is_ecure/g10_is_secure/
* mpiutil.c (mpi_alloc): Remved the debug mode because it has turned
out, that this feature was not very useful in the past. Use the
new alloc functions.
(mpi_alloc_secure): Ditto.
(mpi_alloc_limb_space): Ditto.
(mpi_free_limb_space): Ditto.
(mpi_resize): Ditto.
(mpi_free): Ditto.
(mpi_set_secure): Removed the debug stuff.
(mpi_set_opaque): Ditto.
(mpi_copy): Ditto.
(mpi_alloc_set_ui): Ditto.
(mpi_m_check): Use g10_ wrapper.
Mon Aug 30 20:38:33 CEST 1999 Werner Koch <wk@isil.d.shuttle.de>
* config.links: Add case label for DJGPP
Wed Jul 14 19:42:08 CEST 1999 Werner Koch <wk@isil.d.shuttle.de>
* Makefile.am: Use .s files as temporaries, disabled other .S rules.
Wed Jul 7 13:08:40 CEST 1999 Werner Koch <wk@isil.d.shuttle.de>
* mpicoder.c (g10_log_mpidump): New.
* Makefile.am: Support for libtool.
Fri Jul 2 11:45:54 CEST 1999 Werner Koch <wk@isil.d.shuttle.de>
* mpi-bit.c (mpi_lshift_limbs,mpi_rshift_limbs): New.
* mpi-mpow.c (barrett_mulm): New but diabled.
Tue Jun 1 16:01:46 CEST 1999 Werner Koch <wk@isil.d.shuttle.de>
* config.links (i[56]86*-*-freebsdelf*): New.
Sun May 23 14:20:22 CEST 1999 Werner Koch <wk@isil.d.shuttle.de>
* config.links (sysdep.h): Not any more conditionally created.
Tue May 4 15:47:53 CEST 1999 Werner Koch <wk@isil.d.shuttle.de>
* mpiutil.c (mpi_alloc_like): New.
Mon Apr 26 17:48:15 CEST 1999 Werner Koch <wk@isil.d.shuttle.de>
* mpih-add.c, mpih-sub.c: Removed
* mpi-inline.c: New.
* mpi-inline.h: Make it usable by mpi-inline.c.
Sun Apr 18 10:11:28 CEST 1999 Werner Koch <wk@isil.d.shuttle.de>
* mpih-mul.c (mpihelp_mul_n): Fixed use of memory region.
(mpihelp_mul): Ditto.
Wed Apr 7 20:51:39 CEST 1999 Werner Koch <wk@isil.d.shuttle.de>
* Makefile.am: Explicit rules to invoke cpp on *.S
Mon Mar 8 20:47:17 CET 1999 Werner Koch <wk@isil.d.shuttle.de>
* config.links: Take advantage of the with_symbol_underscore macro.
Add support for freebsd 4.
Wed Feb 24 11:07:27 CET 1999 Werner Koch <wk@isil.d.shuttle.de>
* mips3/mpih-sub1.S: Removed left over junk in last line. (Should I
blame me or my editor?).
Sat Feb 13 12:04:43 CET 1999 Werner Koch <wk@isil.d.shuttle.de>
* Makefile.am: Removed the +=. Add MPI_OPT_FLAGS.
Sat Jan 9 16:02:23 CET 1999 Werner Koch <wk@isil.d.shuttle.de>
* mpi-cmp.c (mpi_cmp_ui): Normalized the arg.
Thu Jan 7 18:00:58 CET 1999 Werner Koch <wk@isil.d.shuttle.de>
* mpi-bit.c (mpi_normalize): New.
(mpi_get_nbits): Normalize the MPI.
* mpi-bit.c (mpi_cmp): Normalize the MPI before the compare.
Tue Dec 8 13:15:16 CET 1998 Werner Koch <wk@isil.d.shuttle.de>
* config.links: Moved the case for powerpc*linux
* powerpcp32/*.S: Removed some underscores.
Thu Nov 26 07:27:52 1998 Werner Koch <werner.koch@guug.de>
* config.links: Support for ppc with ELF
* powerpc32/syntax.h: New.
* powerpc32/*.S: Applied ELF patches (glibc patches)
Tue Nov 10 19:31:37 1998 Werner Koch (wk@isil.d.shuttle.de)
* power*/ : Started with stuff for PPC
* config.links: Some stuff for PPC.
* generic/udiv-w-sdiv.c: New but disabled.
Tue Oct 27 12:37:46 1998 Werner Koch (wk@isil.d.shuttle.de)
* config.links (freebsd): Fixes for FreeBSD 3.0
Wed Oct 14 09:59:30 1998 Werner Koch (wk@isil.d.shuttle.de)
* config.links (freebsd): ELF patches from Jun Kuriyama.
Thu Oct 8 13:28:17 1998 Werner Koch (wk@isil.d.shuttle.de)
* mpi-mpow.c (mpi_mulpowm): Fixed mem leak (m_free/mpi_free).
Thu Sep 17 18:08:50 1998 Werner Koch (wk@(none))
* hppa1.1/udiv-qrnnd.S: Fix from Steffen Zahn for HPUX 10.20
Thu Aug 6 16:39:28 1998 Werner Koch,mobil,,, (wk@tobold)
* mpi-bit.c (mpi_set_bytes): Removed.
Wed Aug 5 15:11:12 1998 Werner Koch (wk@(none))
* mpicoder.c (mpi_read_from_buffer): New.
* mpiutil.c (mpi_set_opaque): New.
(mpi_get_opaque): New.
(mpi_copy): Changed to support opauqe flag
(mpi_free): Ditto.
Sat Jul 4 10:11:11 1998 Werner Koch (wk@isil.d.shuttle.de)
* mpiutil.c (mpi_clear): Reset flags.
(mpi_set): Ditto.
(mpi_alloc_secure): Set flag to 1 and not ored the 1 in, tsss..
Fri Jun 26 11:19:06 1998 Werner Koch (wk@isil.d.shuttle.de)
* mpiutil.c (mpi_alloc): set nbits to 0.
(mpi_alloc_secure): Ditto.
(mpi_clear): Ditto.
Thu Jun 25 11:50:01 1998 Werner Koch (wk@isil.d.shuttle.de)
* mips3/*.S: New
Mon May 18 13:47:06 1998 Werner Koch (wk@isil.d.shuttle.de)
* config.links: split mpih-shift into mpih-[lr]shift and
changed all implementations.
* mpi/alpha: add some new assembler stuff.
Wed May 13 11:04:29 1998 Werner Koch (wk@isil.d.shuttle.de)
* config.links: Add support for MIPS
Thu Apr 9 11:31:36 1998 Werner Koch (wk@isil.d.shuttle.de)
* mpicoder.c (mpi_get_secure_buffer): New.
Wed Apr 8 09:44:33 1998 Werner Koch (wk@isil.d.shuttle.de)
* config.links: Applied small fix from Ulf Mﾃδｶller.
Mon Apr 6 12:38:52 1998 Werner Koch (wk@isil.d.shuttle.de)
* mpicoder.c (mpi_get_buffer): Removed returned leading zeroes
and changed all callers.
Tue Mar 10 13:40:34 1998 Werner Koch (wk@isil.d.shuttle.de)
* mpi-bit.c (mpi_clear_highbit): New.
Mon Mar 2 19:29:00 1998 Werner Koch (wk@isil.d.shuttle.de)
* Makefile.am (DISTCLEANFILES): New
Thu Feb 26 06:48:54 1998 Werner Koch (wk@isil.d.shuttle.de)
* config.links (X86_BROKEN_ALIGN): Added for some systems.
Mon Feb 23 12:21:40 1998 Werner Koch (wk@isil.d.shuttle.de)
* mpi/m68k/mpih-shift.S (Lspecial): Changed duplicate symbol.
Mon Feb 16 13:00:27 1998 Werner Koch (wk@isil.d.shuttle.de)
* config.links : Add detection of m68k cpus
Copyright 1998,1999,2000,2001,2002,2003 Free Software Foundation, Inc.
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 file 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.
diff --git a/mpi/mpi-bit.c b/mpi/mpi-bit.c
index 26853efd..afedabe3 100644
--- a/mpi/mpi-bit.c
+++ b/mpi/mpi-bit.c
@@ -1,257 +1,316 @@
/* mpi-bit.c - MPI bit level fucntions
* Copyright (C) 1998, 1999, 2001, 2002 Free Software Foundation, Inc.
*
* This file is part of Libgcrypt.
*
* Libgcrypt 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.
*
* Libgcrypt 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, write to the Free Software
* Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA
*/
#include <config.h>
#include <stdio.h>
#include <stdlib.h>
#include <assert.h>
#include "mpi-internal.h"
#include "longlong.h"
#ifdef MPI_INTERNAL_NEED_CLZ_TAB
#ifdef __STDC__
const
#endif
unsigned char
_gcry_clz_tab[] =
{
0,1,2,2,3,3,3,3,4,4,4,4,4,4,4,4,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,
6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,
7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,
7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,
8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,
8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,
8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,
8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,
};
#endif
#define A_LIMB_1 ((mpi_limb_t)1)
/****************
* Sometimes we have MSL (most significant limbs) which are 0;
* this is for some reasons not good, so this function removes them.
*/
void
_gcry_mpi_normalize( gcry_mpi_t a )
{
if( mpi_is_opaque(a) )
return;
for( ; a->nlimbs && !a->d[a->nlimbs-1]; a->nlimbs-- )
;
}
/****************
* Return the number of bits in A.
*/
unsigned int
gcry_mpi_get_nbits( gcry_mpi_t a )
{
unsigned n;
if( mpi_is_opaque(a) ) {
return a->sign; /* which holds the number of bits */
}
_gcry_mpi_normalize( a );
if( a->nlimbs ) {
mpi_limb_t alimb = a->d[a->nlimbs-1];
if( alimb )
count_leading_zeros( n, alimb );
else
n = BITS_PER_MPI_LIMB;
n = BITS_PER_MPI_LIMB - n + (a->nlimbs-1) * BITS_PER_MPI_LIMB;
}
else
n = 0;
return n;
}
/****************
* Test whether bit N is set.
*/
int
gcry_mpi_test_bit( gcry_mpi_t a, unsigned int n )
{
unsigned int limbno, bitno;
mpi_limb_t limb;
limbno = n / BITS_PER_MPI_LIMB;
bitno = n % BITS_PER_MPI_LIMB;
if( limbno >= a->nlimbs )
return 0; /* too far left: this is a 0 */
limb = a->d[limbno];
return (limb & (A_LIMB_1 << bitno))? 1: 0;
}
/****************
* Set bit N of A.
*/
void
gcry_mpi_set_bit( gcry_mpi_t a, unsigned int n )
{
unsigned int limbno, bitno;
limbno = n / BITS_PER_MPI_LIMB;
bitno = n % BITS_PER_MPI_LIMB;
if( limbno >= a->nlimbs ) { /* resize */
- if( a->alloced >= limbno )
- mpi_resize(a, limbno+1 );
+ mpi_resize (a, limbno+1 );
a->nlimbs = limbno+1;
}
a->d[limbno] |= (A_LIMB_1<<bitno);
}
/****************
* Set bit N of A. and clear all bits above
*/
void
gcry_mpi_set_highbit( gcry_mpi_t a, unsigned int n )
{
unsigned int limbno, bitno;
limbno = n / BITS_PER_MPI_LIMB;
bitno = n % BITS_PER_MPI_LIMB;
if( limbno >= a->nlimbs ) { /* resize */
- if( a->alloced >= limbno )
- mpi_resize(a, limbno+1 );
+ mpi_resize (a, limbno+1 );
a->nlimbs = limbno+1;
}
a->d[limbno] |= (A_LIMB_1<<bitno);
for( bitno++; bitno < BITS_PER_MPI_LIMB; bitno++ )
a->d[limbno] &= ~(A_LIMB_1 << bitno);
a->nlimbs = limbno+1;
}
/****************
* clear bit N of A and all bits above
*/
void
gcry_mpi_clear_highbit( gcry_mpi_t a, unsigned int n )
{
unsigned int limbno, bitno;
limbno = n / BITS_PER_MPI_LIMB;
bitno = n % BITS_PER_MPI_LIMB;
if( limbno >= a->nlimbs )
return; /* not allocated, so need to clear bits :-) */
for( ; bitno < BITS_PER_MPI_LIMB; bitno++ )
a->d[limbno] &= ~(A_LIMB_1 << bitno);
a->nlimbs = limbno+1;
}
/****************
* Clear bit N of A.
*/
void
gcry_mpi_clear_bit( gcry_mpi_t a, unsigned int n )
{
unsigned int limbno, bitno;
limbno = n / BITS_PER_MPI_LIMB;
bitno = n % BITS_PER_MPI_LIMB;
if( limbno >= a->nlimbs )
return; /* don't need to clear this bit, it's to far to left */
a->d[limbno] &= ~(A_LIMB_1 << bitno);
}
-/****************
- * Shift A by N bits to the right
- * FIXME: should use alloc_limb if X and A are same.
+/*
+ * Shift A by N bits to the right.
*/
void
gcry_mpi_rshift( gcry_mpi_t x, gcry_mpi_t a, unsigned n )
{
- mpi_ptr_t xp;
- mpi_size_t xsize;
-
- xsize = a->nlimbs;
- x->sign = a->sign;
- RESIZE_IF_NEEDED(x, xsize);
- xp = x->d;
-
- if( xsize ) {
- _gcry_mpih_rshift( xp, a->d, xsize, n);
- MPN_NORMALIZE( xp, xsize);
+ mpi_size_t xsize;
+ unsigned int i;
+ unsigned int nlimbs = (n/BITS_PER_MPI_LIMB);
+ unsigned int nbits = (n%BITS_PER_MPI_LIMB);
+
+ if ( x == a )
+ {
+ /* In-place operation. */
+ if ( nlimbs >= x->nlimbs )
+ {
+ x->nlimbs = 0;
+ return;
+ }
+
+ if (nlimbs)
+ {
+ for (i=0; i < x->nlimbs - nlimbs; i++ )
+ x->d[i] = x->d[i+nlimbs];
+ x->d[i] = 0;
+ x->nlimbs -= nlimbs;
+
+ }
+ if ( x->nlimbs && nbits )
+ _gcry_mpih_rshift ( x->d, x->d, x->nlimbs, nbits );
+ }
+ else if ( nlimbs )
+ {
+ /* Copy and shift by more or equal bits than in a limb. */
+ xsize = a->nlimbs;
+ x->sign = a->sign;
+ RESIZE_IF_NEEDED (x, xsize);
+ x->nlimbs = xsize;
+ for (i=0; i < a->nlimbs; i++ )
+ x->d[i] = a->d[i];
+ x->nlimbs = i;
+
+ if ( nlimbs >= x->nlimbs )
+ {
+ x->nlimbs = 0;
+ return;
+ }
+
+ if (nlimbs)
+ {
+ for (i=0; i < x->nlimbs - nlimbs; i++ )
+ x->d[i] = x->d[i+nlimbs];
+ x->d[i] = 0;
+ x->nlimbs -= nlimbs;
+ }
+
+ if ( x->nlimbs && nbits )
+ _gcry_mpih_rshift ( x->d, x->d, x->nlimbs, nbits );
+ }
+ else
+ {
+ /* Copy and shift by less than bits in a limb. */
+ xsize = a->nlimbs;
+ x->sign = a->sign;
+ RESIZE_IF_NEEDED (x, xsize);
+ x->nlimbs = xsize;
+
+ if ( xsize )
+ {
+ if (nbits )
+ _gcry_mpih_rshift (x->d, a->d, x->nlimbs, nbits );
+ else
+ {
+ /* The rshift helper function is not specified for
+ NBITS==0, thus we do a plain copy here. */
+ for (i=0; i < x->nlimbs; i++ )
+ x->d[i] = a->d[i];
+ }
+ }
}
- x->nlimbs = xsize;
+ MPN_NORMALIZE (x->d, x->nlimbs);
}
/****************
* Shift A by COUNT limbs to the left
* This is used only within the MPI library
*/
void
_gcry_mpi_lshift_limbs( gcry_mpi_t a, unsigned int count )
{
mpi_ptr_t ap = a->d;
int n = a->nlimbs;
int i;
if( !count || !n )
return;
RESIZE_IF_NEEDED( a, n+count );
for( i = n-1; i >= 0; i-- )
ap[i+count] = ap[i];
for(i=0; i < count; i++ )
ap[i] = 0;
a->nlimbs += count;
}
/****************
* Shift A by COUNT limbs to the right
* This is used only within the MPI library
*/
void
_gcry_mpi_rshift_limbs( gcry_mpi_t a, unsigned int count )
{
mpi_ptr_t ap = a->d;
mpi_size_t n = a->nlimbs;
unsigned int i;
if( count >= n ) {
a->nlimbs = 0;
return;
}
for( i = 0; i < n - count; i++ )
ap[i] = ap[i+count];
ap[i] = 0;
a->nlimbs -= count;
}
diff --git a/mpi/mpiutil.c b/mpi/mpiutil.c
index b8ec2330..f6ce6575 100644
--- a/mpi/mpiutil.c
+++ b/mpi/mpiutil.c
@@ -1,428 +1,440 @@
/* mpiutil.ac - Utility functions for MPI
* Copyright (C) 1998, 2000, 2001, 2002, 2003 Free Software Foundation, Inc.
*
* This file is part of Libgcrypt.
*
* Libgcrypt 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.
*
* Libgcrypt 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, write to the Free Software
* Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA
*/
#include <config.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <assert.h>
#include "g10lib.h"
#include "mpi-internal.h"
#include "memory.h"
/****************
* Note: It was a bad idea to use the number of limbs to allocate
* because on a alpha the limbs are large but we normally need
* integers of n bits - So we should chnage this to bits (or bytes).
*
* But mpi_alloc is used in a lot of places :-)
*/
gcry_mpi_t
_gcry_mpi_alloc( unsigned nlimbs )
{
gcry_mpi_t a;
a = gcry_xmalloc( sizeof *a );
a->d = nlimbs? mpi_alloc_limb_space( nlimbs, 0 ) : NULL;
a->alloced = nlimbs;
a->nlimbs = 0;
a->sign = 0;
a->flags = 0;
return a;
}
void
_gcry_mpi_m_check( gcry_mpi_t a )
{
_gcry_check_heap(a);
_gcry_check_heap(a->d);
}
gcry_mpi_t
_gcry_mpi_alloc_secure( unsigned nlimbs )
{
gcry_mpi_t a;
a = gcry_xmalloc( sizeof *a );
a->d = nlimbs? mpi_alloc_limb_space( nlimbs, 1 ) : NULL;
a->alloced = nlimbs;
a->flags = 1;
a->nlimbs = 0;
a->sign = 0;
return a;
}
mpi_ptr_t
_gcry_mpi_alloc_limb_space( unsigned int nlimbs, int secure )
{
mpi_ptr_t p;
size_t len;
len = (nlimbs ? nlimbs : 1) * sizeof (mpi_limb_t);
p = secure ? gcry_xmalloc_secure (len) : gcry_xmalloc (len);
if (! nlimbs)
*p = 0;
return p;
}
void
_gcry_mpi_free_limb_space( mpi_ptr_t a, unsigned int nlimbs)
{
if (a)
{
size_t len = nlimbs * sizeof(mpi_limb_t);
/* If we have information on the number of allocated limbs, we
better wipe that space out. This is a failsafe feature if
secure memory has been disabled or was not properly
implemented in user provided allocation functions. */
if (len)
wipememory (a, len);
gcry_free(a);
}
}
void
_gcry_mpi_assign_limb_space( gcry_mpi_t a, mpi_ptr_t ap, unsigned int nlimbs )
{
_gcry_mpi_free_limb_space (a->d, a->alloced);
a->d = ap;
a->alloced = nlimbs;
}
/****************
- * Resize the array of A to NLIMBS. the additional space is cleared
- * (set to 0) [done by gcry_realloc()]
+ * Resize the array of A to NLIMBS. The additional space is cleared
+ * (set to 0).
*/
void
_gcry_mpi_resize (gcry_mpi_t a, unsigned nlimbs)
{
+ size_t i;
+
if (nlimbs <= a->alloced)
- return; /* no need to do it */
+ {
+ /* We only need to clear the new space (this is a nop if the
+ limb space is already of the correct size. */
+ for (i=a->nlimbs; i < a->alloced; i++)
+ a->d[i] = 0;
+ return;
+ }
if (a->d)
- a->d = gcry_xrealloc (a->d, nlimbs * sizeof (mpi_limb_t));
+ {
+ a->d = gcry_xrealloc (a->d, nlimbs * sizeof (mpi_limb_t));
+ for (i=a->alloced; i < nlimbs; i++)
+ a->d[i] = 0;
+ }
else
{
if (a->flags & 1)
/* Secure memory is wanted. */
a->d = gcry_xcalloc_secure (nlimbs , sizeof (mpi_limb_t));
else
/* Standard memory. */
a->d = gcry_xcalloc (nlimbs , sizeof (mpi_limb_t));
}
a->alloced = nlimbs;
}
void
_gcry_mpi_clear( gcry_mpi_t a )
{
a->nlimbs = 0;
a->flags = 0;
}
void
_gcry_mpi_free( gcry_mpi_t a )
{
if (!a )
return;
if ((a->flags & 4))
gcry_free( a->d );
else
{
_gcry_mpi_free_limb_space(a->d, a->alloced);
}
if ((a->flags & ~7))
log_bug("invalid flag value in mpi\n");
gcry_free(a);
}
static void
mpi_set_secure( gcry_mpi_t a )
{
mpi_ptr_t ap, bp;
if ( (a->flags & 1) )
return;
a->flags |= 1;
ap = a->d;
if (!a->nlimbs)
{
assert(!ap);
return;
}
bp = mpi_alloc_limb_space (a->nlimbs, 1);
MPN_COPY( bp, ap, a->nlimbs );
a->d = bp;
_gcry_mpi_free_limb_space (ap, a->alloced);
}
gcry_mpi_t
gcry_mpi_set_opaque( gcry_mpi_t a, void *p, unsigned int nbits )
{
if (!a)
a = mpi_alloc(0);
if( a->flags & 4 )
gcry_free( a->d );
else
_gcry_mpi_free_limb_space (a->d, a->alloced);
a->d = p;
a->alloced = 0;
a->nlimbs = 0;
a->sign = nbits;
a->flags = 4;
return a;
}
void *
gcry_mpi_get_opaque( gcry_mpi_t a, unsigned int *nbits )
{
if( !(a->flags & 4) )
log_bug("mpi_get_opaque on normal mpi\n");
if( nbits )
*nbits = a->sign;
return a->d;
}
/****************
* Note: This copy function should not interpret the MPI
* but copy it transparently.
*/
gcry_mpi_t
_gcry_mpi_copy( gcry_mpi_t a )
{
int i;
gcry_mpi_t b;
if( a && (a->flags & 4) ) {
void *p = gcry_is_secure(a->d)? gcry_xmalloc_secure( (a->sign+7)/8 )
: gcry_xmalloc( (a->sign+7)/8 );
memcpy( p, a->d, (a->sign+7)/8 );
b = gcry_mpi_set_opaque( NULL, p, a->sign );
}
else if( a ) {
b = mpi_is_secure(a)? mpi_alloc_secure( a->nlimbs )
: mpi_alloc( a->nlimbs );
b->nlimbs = a->nlimbs;
b->sign = a->sign;
b->flags = a->flags;
for(i=0; i < b->nlimbs; i++ )
b->d[i] = a->d[i];
}
else
b = NULL;
return b;
}
/****************
* This function allocates an MPI which is optimized to hold
* a value as large as the one given in the argument and allocates it
* with the same flags as A.
*/
gcry_mpi_t
_gcry_mpi_alloc_like( gcry_mpi_t a )
{
gcry_mpi_t b;
if( a && (a->flags & 4) ) {
int n = (a->sign+7)/8;
void *p = gcry_is_secure(a->d)? gcry_malloc_secure( n )
: gcry_malloc( n );
memcpy( p, a->d, n );
b = gcry_mpi_set_opaque( NULL, p, a->sign );
}
else if( a ) {
b = mpi_is_secure(a)? mpi_alloc_secure( a->nlimbs )
: mpi_alloc( a->nlimbs );
b->nlimbs = 0;
b->sign = 0;
b->flags = a->flags;
}
else
b = NULL;
return b;
}
void
_gcry_mpi_set( gcry_mpi_t w, gcry_mpi_t u)
{
mpi_ptr_t wp, up;
mpi_size_t usize = u->nlimbs;
int usign = u->sign;
RESIZE_IF_NEEDED(w, usize);
wp = w->d;
up = u->d;
MPN_COPY( wp, up, usize );
w->nlimbs = usize;
w->flags = u->flags;
w->sign = usign;
}
void
_gcry_mpi_set_ui( gcry_mpi_t w, unsigned long u)
{
RESIZE_IF_NEEDED(w, 1);
w->d[0] = u;
w->nlimbs = u? 1:0;
w->sign = 0;
w->flags = 0;
}
gcry_mpi_t
_gcry_mpi_alloc_set_ui( unsigned long u)
{
gcry_mpi_t w = mpi_alloc(1);
w->d[0] = u;
w->nlimbs = u? 1:0;
w->sign = 0;
return w;
}
void
_gcry_mpi_swap( gcry_mpi_t a, gcry_mpi_t b)
{
struct gcry_mpi tmp;
tmp = *a; *a = *b; *b = tmp;
}
void
gcry_mpi_swap( gcry_mpi_t a, gcry_mpi_t b)
{
_gcry_mpi_swap (a, b);
}
gcry_mpi_t
gcry_mpi_new( unsigned int nbits )
{
return _gcry_mpi_alloc( (nbits+BITS_PER_MPI_LIMB-1) / BITS_PER_MPI_LIMB );
}
gcry_mpi_t
gcry_mpi_snew( unsigned int nbits )
{
return _gcry_mpi_alloc_secure( (nbits+BITS_PER_MPI_LIMB-1) / BITS_PER_MPI_LIMB );
}
void
gcry_mpi_release( gcry_mpi_t a )
{
_gcry_mpi_free( a );
}
gcry_mpi_t
gcry_mpi_copy( const gcry_mpi_t a )
{
return _gcry_mpi_copy( (gcry_mpi_t)a );
}
gcry_mpi_t
gcry_mpi_set( gcry_mpi_t w, const gcry_mpi_t u )
{
if( !w )
w = _gcry_mpi_alloc( mpi_get_nlimbs(u) );
_gcry_mpi_set( w, (gcry_mpi_t)u );
return w;
}
gcry_mpi_t
gcry_mpi_set_ui( gcry_mpi_t w, unsigned long u )
{
if( !w )
w = _gcry_mpi_alloc(1);
_gcry_mpi_set_ui( w, u );
return w;
}
void
gcry_mpi_randomize( gcry_mpi_t w,
unsigned int nbits, enum gcry_random_level level )
{
unsigned char *p;
size_t nbytes = (nbits+7)/8;
if (level == GCRY_WEAK_RANDOM)
{
p = mpi_is_secure(w) ? gcry_xmalloc_secure (nbytes)
: gcry_xmalloc (nbytes);
gcry_create_nonce (p, nbytes);
}
else
{
p = mpi_is_secure(w) ? gcry_random_bytes_secure (nbytes, level)
: gcry_random_bytes (nbytes, level);
}
_gcry_mpi_set_buffer( w, p, nbytes, 0 );
gcry_free (p);
}
void
gcry_mpi_set_flag( gcry_mpi_t a, enum gcry_mpi_flag flag )
{
switch( flag ) {
case GCRYMPI_FLAG_SECURE: mpi_set_secure(a); break;
case GCRYMPI_FLAG_OPAQUE:
default: log_bug("invalid flag value\n");
}
}
void
gcry_mpi_clear_flag( gcry_mpi_t a, enum gcry_mpi_flag flag )
{
switch( flag ) {
case GCRYMPI_FLAG_SECURE:
case GCRYMPI_FLAG_OPAQUE:
default: log_bug("invalid flag value\n");
}
}
int
gcry_mpi_get_flag( gcry_mpi_t a, enum gcry_mpi_flag flag )
{
switch( flag ) {
case GCRYMPI_FLAG_SECURE: return (a->flags & 1);
case GCRYMPI_FLAG_OPAQUE: return (a->flags & 4);
default: log_bug("invalid flag value\n");
}
}
diff --git a/src/ChangeLog b/src/ChangeLog
index 56d9c37f..bebf185e 100644
--- a/src/ChangeLog
+++ b/src/ChangeLog
@@ -1,1455 +1,1460 @@
+2006-08-28 Werner Koch <wk@g10code.com>
+
+ * global.c (gcry_control): Clandestine API change for
+ GCRYCTL_FAKED_RANDOM_P.
+
2006-06-21 Werner Koch <wk@g10code.com>
- * global.c (gcry_xcalloc, gcry_xcalloc_secure): made safe against
+ * global.c (gcry_xcalloc, gcry_xcalloc_secure): Made safe against
integer overflow.
2005-06-17 Moritz Schulte <moritz@g10code.com>
* global.c (gcry_xstrdup): Removed superfluous strcpy call.
2005-06-16 Werner Koch <wk@g10code.com>
* sexp.c (normalize): Made P unsigned.
(gcry_sexp_nth_data): Cast return value to char*.
(sexp_sscan): Fixed sign/usnigned conflicts.
(whitespacep): Changed arg to char*.
(unquote_string): Changed arg STRING to char*.
* gcrypt-module.h (gcry_cipher_register): Changed arg ALGORITHM_ID
from unsigned int * to int*.
* gcrypt.h (gcry_mpi_scan): Changed arg BUFFER from unsigned char*
to void*.
2005-06-15 Werner Koch <wk@g10code.com>
* gcrypt.h, sexp.c (gcry_sexp_sprint): Changed arg BUFFER from
char* to void*.
* gcrypt.h (gcry_md_ctl): Changed arg BUFFER from unsigned char*
to void*.
(gcry_randomize): Ditto.
(gcry_create_nonce): Ditto.
(gcry_cipher_encrypt, gcry_cipher_decrypt): Changed args IN and
OUT from unsigned char* to void*.
2005-04-16 Moritz Schulte <moritz@g10code.com>
* g10lib.h (_gcry_ac_init): Declare.
* global.c (global_init): Call _gcry_ac_init; don't forget to set err.
2005-04-14 Werner Koch <wk@g10code.com>
* sexp.c (whitespacep): New.
(sexp_sscan): Replace isdigit and isspace by whitespacep and
digitp.
2005-02-22 Werner Koch <wk@g10code.com>
* global.c (_gcry_malloc): Make sure ERRNO is set if we return
NULL. Remove unneeded initialization of M to allow the compiler
to catch errors.
(gcry_realloc): Make sure ERRNO is set if we return NULL.
2005-01-05 Werner Koch <wk@g10code.com>
* gcrypt.h: Added GCRYCTL_FAST_POLL.
(gcry_fast_random_poll): New.
* global.c (gcry_control) <FAST_POLL>: New.
2004-12-20 Werner Koch <wk@g10code.com>
* gcrypt.h (GCRY_THREAD_OPTION_PTH_IMPL): Tweaks for W32.
2004-12-06 Werner Koch <wk@g10code.com>
* g10lib.h (stpcpy, strcasecmp): Prefix them with _gcry_ and add
macros to define them to the real name. This is needed to avoid
double symbols when using static linkage.
* missing-string.c: Likewise.
(vasprintf): Removed as it is not used anywhere.
2004-11-26 Werner Koch <wk@g10code.com>
* types.h [_WIN32]: Avoid warning about double defined type byte.
2004-11-25 Werner Koch <wk@g10code.com>
* ath.h [_WIN32]: Adjusted for modern mingw cross compiler.
2004-11-09 Werner Koch <wk@g10code.de>
* gcrypt.h: Removed 3 trailing commas from enums. Reported by
Heiko Stamer.
2004-09-21 Werner Koch <wk@g10code.de>
* sexp.c (sexp_sscan): Removed c++ style commets. Noted by Yoann
Vandoorselaere.
2004-09-15 Werner Koch <wk@g10code.de>
* global.c (gcry_control): Adjusted random into call for the
changed API in random.c. Seemed we never compiled the entire
thing since May.
2004-08-22 Moritz Schulte <moritz@g10code.com>
* secmem.c (_gcry_secmem_init): Try to lock pool into core not
only when running with root privileges.
2004-08-16 Werner Koch <wk@g10code.de>
* secmem.c (_gcry_secmem_init): Defer printing of the warning.
2004-08-10 Moritz Schulte <moritz@g10code.com>
* gcrypt.h: Include <sys/time.h>, thanks to Simon Josefsson.
2004-04-22 Marcus Brinkmann <marcus@g10code.de>
* libgcrypt.m4: Quote first argument to AC_DEFUN.
2004-04-15 Werner Koch <wk@gnupg.org>
* secmem.c (_gcry_secmem_malloc_internal): Removed old extra info
error output.
(_gcry_secmem_term): Use wipememory2 here.
* misc.c (_gcry_burn_stack): Use wipememory to avoid optimizations.
* string.c: Removed. Was never used.
* global.c (gcry_strdup): Replaced by the version from string.c
(gcry_xstrdup): Rewritten.
* gcrypt.h: Removed duplicate prototype for gcry_strdup.
2004-03-29 Werner Koch <wk@gnupg.org>
* secmem.c (_gcry_secmem_realloc): Fixed double unlock; bug
manifested itself due to the more rigorous checking in the changed
ath.h
* libgcrypt-config.in (Options): Ignore the obsolete --threads
option for now.
2004-03-17 Marcus Brinkmann <marcus@g10code.de>
* libgcrypt-config.in (includedir, libdir): Quote'em. Use
$gpg_error_cflags and $gpg_error_libs. Fix construction of
$includes.
2004-03-14 Marcus Brinkmann <marcus@g10code.de>
* libgcrypt-config.in (includedir, libdir): New variables. For
--cflags, don't test $cflags. Also check against /include for the
GNU/Hurd. Don't overwrite but extend $cflags_final. Likewise for
--libs.
2004-03-10 Marcus Brinkmann <marcus@g10code.de>
* Makefile.am (ltlib_libgcrypt_pthread, ltlib_libgcrypt_pth): Removed.
(lib_LTLIBRARIES): Remove those variables from here.
(libgcrypt_pthread_la_SOURCES, libgcrypt_pthread_la_LDFLAGS,
(libgcrypt_pthread_la_DEPENDENCIES, libgcrypt_pthread_la_LIBADD,
(libgcrypt_pth_la_SOURCES, libgcrypt_pth_la_LDFLAGS,
(libgcrypt_pth_la_DEPENDENCIES, libgcrypt_pth_la_LIBADD,
(noinst_LTLIBRARIES): Removed.
(libgcrypt_real_la_SOURCES): Merge with ...
(libgcrypt_la_SOURCES): ... likewise.
(libgcrypt_real_la_DEPENDENCIES): Merge with ...
(libgcrypt_la_DEPENDENCIES): ... this.
(libgcrypt_real_la_LIBADD): Merge with ...
(libgcrypt_la_LIBADD): ... this.
* libgcrypt-config.in (libs_pthread, libs_pth, cflags_pth)
(cflags_pthread, thread_module, thread_modules): Removed.
(Options): Remove --thread option from help output. If the option
is specified, output an error and exit.
For --cflags and --libs option, remove pth and pthread from output.
* gcrypt.h: Include <sys/types.h> and <sys/socket.h>.
(enum gcry_ctl_cmds): Add GCRYCTL_SET_THREAD_CBS.
(gcry_thread_cbs): New struct.
* global.c (gcry_control): Implement GCRYCTL_SET_THREAD_CBS.
(global_init): Don't call ath_init here.
* ath.h: Rewritten.
* ath.c: Rewritten.
2004-03-06 Werner Koch <wk@gnupg.org>
* libgcrypt-config.in: s/--soname-number/--api-version/
* libgcrypt.m4: Changed test for API version.
2004-03-05 Werner Koch <wk@gnupg.org>
* libgcrypt.m4: Optionally check the SONAME number.
* libgcrypt-config.in: Add option --soname-number
2004-03-01 Marcus Brinkmann <marcus@g10code.de>
* Makefile.am (libgcrypt_la_SOURCES): Add ath.c.
* ath.c (ath_init): Add missing function.
* Makefile.am (ath_pth_src): Removed.
(ath_pthread_src): Removed.
(libgcrypt_la_SOURCES): Remove ath-compat, $(ath_pth_src) and
$(ath_pthread_src).
* ath-compat.c, ath-pth-compat.c, ath-pthread-compat.c: Files
removed.
2004-02-20 Werner Koch <wk@gnupg.org>
* gcrypt.h (GCRY_PRIME_CHECK_AT_GOT_PRIME)
(GCRY_PRIME_CHECK_AT_FINISH),
(GCRY_PRIME_CHECK_AT_MAYBE_PRIME): New.
2004-02-18 Werner Koch <wk@gnupg.org>
* libgcrypt-config.in: Ignore setting of --prefix.
2004-02-13 Werner Koch <wk@gnupg.org>
* gcrypt.h: Added GCRY_CIPHER_RFC2268_128, alsthough not yet
supported.
2004-02-06 Werner Koch <wk@gnupg.org>
* gcrypt.h: Added GCRY_CIPHER_RFC2268_40.
2004-02-03 Werner Koch <wk@gnupg.org>
* secmem.c (_gcry_secmem_init): Do not print the "not locked into
core warning" if the NO_WARNING flag has been set.
* sexp.c (sexp_sscan): Allocate result in secure memory if BUFFER
is in secure memory. Switch to secure memory for the a secure %b
format item. Extra paranoid wipe on error.
(gcry_sexp_release): Added paranoid wiping for securely allocated
S-expressions.
2004-01-25 Moritz Schulte <mo@g10code.com>
* ath.h: Include <config.h>.
2004-01-12 Moritz Schulte <mo@g10code.com>
* gcrypt.h: Adjusted declarations of: gcry_ac_data_set,
gcry_ac_data_get_name, gcry_ac_data_get_index,
gcry_ac_key_pair_generate, gcry_ac_key_test,
gcry_ac_key_get_nbits, gcry_ac_key_get_grip.
* gcrypt.h (GCRY_AC_FLAG_DATA_NO_BLINDING): Removed symbol.
(GCRY_AC_FLAG_DEALLOC, GCRY_AC_FLAG_COPY)
(GCRY_AC_FLAG_NO_BLINDING): New symbols.
* global.c (gcry_strdup): Removed function.
* string.c: New file.
* Makefile.am (libgcrypt_real_la_SOURCES): Added: string.c.
* string.c (gcry_strdup): New function.
* gcrypt.h (gcry_strdup): Declare.
2003-12-19 Werner Koch <wk@gnupg.org>
* g10lib.h (wipememory, wipememory2): New; taken from gnupg.
2003-11-14 Werner Koch <wk@gnupg.org>
* global.c (gcry_strdup): Don't copy the string after a malloc
error.
2003-11-11 Werner Koch <wk@gnupg.org>
* sexp.c (sexp_sscan): Implemented "%b" format specifier.
2003-11-11 Moritz Schulte <mo@g10code.com>
* libgcrypt.m4: Do not set prefix when calling libgcrypt-config.
Thanks to Nikos Mavroyanopoulos.
2003-11-08 Moritz Schulte <mo@g10code.com>
* cipher.h (small_prime_numbers): Removed declaration.
(PUBKEY_FLAG_NO_BLINDING): Put braces around shift.
2003-11-04 Werner Koch <wk@gnupg.org>
* cipher.h (_gcry_sha1_has_buffer): New.
* gcrypt.h (gcry_create_nonce): New.
2003-10-31 Werner Koch <wk@gnupg.org>
* libgcrypt.vers (_gcry_generate_elg_prime): Removed this symbol;
gnutls does not need it anymore.
* secmem.c (mb_get_new): s/pool/block/ due to global pool.
* misc.c (gcry_set_log_handler): s/logf/f/ to avoid shadowing
warning against a builtin.
* ath-pth-compat.c: cast pth_connect to get rid of the const
prototype.
2003-10-27 Werner Koch <wk@gnupg.org>
* ath.h (ATH_MUTEX_INITIALIZER): Removed spurious semicolon.
2003-10-27 Moritz Schulte <mo@g10code.com>
* libgcrypt-config.in: Include libs/cflags of libgpg-error.
* sexp.c (sexp_sscan): Cleaned up, deallocate scanned sexp on
error.
* module.c (MODULE_ID_MIN): New symbol, use it.
2003-10-27 Werner Koch <wk@gnupg.org>
* gcrypt.h (gcry_pk_testkey): Doc fix.
2003-09-29 Moritz Schulte <mo@g10code.com>
* libgcrypt-config.in: Fix --algorithms option.
2003-10-23 Werner Koch <wk@gnupg.org>
* gcrypt.h (gcry_err_code): Use GPG_ERR_INLINE instead of
__inline__.
* secmem.c (lock_pool): Don't print the warning for certain
systems, handle ENOMEM.
2003-10-21 Werner Koch <wk@gnupg.org>
* secmem.c (_gcry_secmem_dump_stats): Fixed format sepcifier for a
size_t. Reported by Stephane Corthesy.
2003-10-10 Werner Koch <wk@gnupg.org>
* global.c (_gcry_malloc): Handle the no_secure_memory option.
* gcrypt.h (gcry_prime_group_generator): New.
(gcry_prime_release_factors): New.
2003-10-07 Werner Koch <wk@gnupg.org>
* sexp.c (sexp_sscan): Check that parenthesis are matching.
2003-09-28 Moritz Schulte <mo@g10code.com>
* g10lib.h: Declare: _gcry_malloc.
(GCRY_ALLOC_FLAG_SECURE): New symbol.
* global.c (_gcry_malloc): New function...
(gcry_malloc): ... use it.
(gcry_malloc_secure): Likewise.
* ath.c: Change License to LGPL.
* ath-pthread-compat.c: Likewise.
* ath-pthread.c: Likewise.
* ath-pth-compat.c: Likewise.
* ath-pth.c: Likewise.
* ath.h: Likewise.
* ath-compat.c: Likewise.
* secmem.c (_gcry_secmem_realloc): Do not forget to release secmem
lock. Thanks to low halo for triggering this bug.
2003-09-04 Werner Koch <wk@gnupg.org>
* gcrypt.h (_GCRY_ERR_SOURCE_DEFAULT): Removed cruft.
(gcry_prime_check_func_t): Renamed arg for clarity.
2003-09-02 Moritz Schulte <mo@g10code.com>
* gcrypt.h (GCRY_PRIME_FLAG_SPECIAL_FACTOR): New symbol.
2003-09-01 Moritz Schulte <mo@g10code.com>
* gcrypt.h (gcry_random_level_t): New type.
(gcry_prime_check_func_t): Likewise.
(GCRY_PRIME_FLAG_SECRET): New symbol.
(gcry_prime_generate, gcry_prime_check): Declare functions.
2003-08-28 Werner Koch <wk@gnupg.org>
* Makefile.am (libgcrypt_pth_la_LDFLAGS): Removed PTH_CFLAGS cruft.
2003-08-27 Moritz Schulte <mo@g10code.com>
* global.c (gcry_control): Remove call to ath_deinit.
* Makefile.am (libgcrypt_real_la_DEPENDENCIES): Fixed.
(libgcrypt_real_la_LIBADD): Fixed.
Removed unecessary variables.
* libgcrypt-config.in: Adjusted script for new thread handling.
* Makefile.am: New version, based on GPGMEs Makefile.am.
* ath.c, ath-compat.c, ath.h, ath-pth.c, ath-pth-compat.c,
ath-pthread.c, ath-pthread-compat.c: New files, merged from GPGME.
* ath.c, ath.h, ath-pthread.c, ath-pth.c: Removed files.
2003-08-08 Moritz Schulte <moritz@g10code.com>
* global.c (gcry_realloc): Remove FIXME about `clearing out
realloced memory', since _gcry_secmem_realloc takes care of
overwriting old memory.
2003-08-07 Werner Koch <wk@gnupg.org>
* module.c (_gcry_module_release): Don't act if module is NULL.
2003-07-30 Moritz Schulte <moritz@g10code.com>
* gcrypt.h (enum gcry_ac_id): Added: GCRY_AC_ELG_E.
Reverted change: use gcry_md_flags enumeration list instead of
defines.
2003-07-29 Werner Koch <wk@gnupg.org>
* global.c (gcry_control): Add GCRYCTL_SET_RANDOM_SEED_FILE and
GCRYCTL_UPDATE_RANDOM_SEED_FILE.
* gcrypt.h: Ditto. Renamed index to idx, so avoid warning
related to the old index function.
2003-07-28 Moritz Schulte <moritz@g10code.com>
* global.c (gcry_err_code_from_errno, gcry_err_code_to_errno)
(gcry_err_make_from_errno, gcry_error_from_errno): New functions.
* gcrypt.h: Declared: gcry_err_code_from_errno,
gcry_err_code_to_errno, gcry_err_make_from_errno,
gcry_error_from_errno.
* Makefile.am (include_HEADERS): Added: gcrypt-module.h.
* gcrypt.h: Include <gcrypt-module.h>.
* gcrypt-module.h: New file.
2003-07-27 Werner Koch <wk@gnupg.org>
* gcrypt.h (gcry_mpi_scan, gcry_mpi_print): API change.
(gcry_mpi_dump): New.
2003-07-21 Moritz Schulte <moritz@g10code.com>
* gcrypt.h: Declared: gcry_ac_key_data_get.
(gcry_pk_spec): Renamed member `sexp_names' into `aliases'.
2003-07-20 Moritz Schulte <moritz@g10code.com>
* gcrypt.h (gcry_md_oid_spec_t): New type.
(gcry_md_spec): New member: oids.
2003-07-19 Moritz Schulte <moritz@g10code.com>
* gcrypt.h (gcry_cipher_oid_spec_t): New type.
(gcry_cipher_spec): New member: oids;
2003-07-18 Werner Koch <wk@gnupg.org>
* gcrypt.h (gcry_mpi_set_opaque): Add a warning comment.
2003-07-15 Moritz Schulte <moritz@g10code.com>
* secmem.c (compress_pool): Remove function, since unused blocks
are automatically concatenad.
* gcrypt.h: Bumped version number up to 1.1.42-cvs.
2003-07-14 Moritz Schulte <moritz@g10code.com>
* gcrypt.h (gcry_cipher_spec): New member: aliases.
* Makefile.am (noinst_PROGRAMS, testapi_SOURCES, testapai_LDADD,
benchmark_SOURCES, benchmark_LDADD): Removed.
* benchmark.c, testapi.c: Removed files.
* mpi.h: Removed disabled typedef.
* g10lib.h: Likewise.
* benchmark.c, g10lib.h, gcrypt.h, global.c, module.c, sexp.c:
Used gcry_err* wrappers for libgpg-error symbols.
2003-07-12 Moritz Schulte <moritz@g10code.com>
* global.c: Likewise.
* gcrypt.h: New type: gcry_error_t, gcry_err_code_t and
gcry_err_source_t.
(gcry_err_make, gcry_error, gcry_err_code, gcry_err_source): New
functions.
* global.c (gcry_strerror): New function.
(gcry_strsource): New function.
* gcrypt.h: New symbol: GCRY_CIPHER_TWOFISH128.
2003-07-09 Moritz Schulte <moritz@g10code.com>
* gcrypt.h (enum gcry_md_flags): Removed, used define instead,
since that is more common than an enumeration list when it comes
to flags that can be bitwise ORed.
2003-07-08 Moritz Schulte <moritz@g10code.com>
* global.c: Use new types for handlers.
* gcrypt.h: Declare: gcry_ac_data_copy.
2003-07-07 Moritz Schulte <moritz@g10code.com>
* sexp.c (gcry_sexp_build_array): Use dummy argument pointer.
Thanks to Simon Josefsson <jas@extunde.com>.
* gcrypt.h: Declare: gcry_cipher_list, gcry_pk_list, gcry_md_list.
2003-07-05 Moritz Schulte <moritz@g10code.com>
* gcrypt.h: Declare: gcry_cipher_register, gcry_cipher_unregister,
gcry_md_register, gcry_md_unregister, gcry_pk_register,
gcry_pk_unregister.
(gcry_cipher_spec): Removed member: algorithm.
(gcry_pk_spec): Likewise.
(gcry_md_spec): Likewise.
Adjusted declarations: gcry_cipher_register, gcry_pk_register,
gcry_md_register.
* module.c: Replaced all occurences of `id' with `mod_id', since
`id' is a keyword in obj-c.
* gcrypt.h (gcry_cipher_spec): Renamed member `id' to `algorithm'.
(gcry_pk_spec): Likewise.
(gcry_md_spec): Likewise.
* cipher.h: Removed types: gcry_pubkey_generate_t,
gcry_pubkey_check_secret_key_t, gcry_pubkey_encrypt_t,
gcry_pubkey_decrypt_t, gcry_pubkey_sign_t, gcry_pubkey_verify_t,
gcry_pubkey_get_nbits_t, gcry_pk_spec_t, gcry_digest_init_t,
gcry_digest_write_t, gcry_digest_final_t, gcry_digest_read_t,
gcry_digest_spec_t, gcry_cipher_setkey_t, gcry_cipher_encrypt_t,
gcry_cipher_decrypt_t, gcry_cipher_stencrypt_t,
gcry_cipher_stdecrypt_t, gcry_cipher_spec_t.
* gcrypt.h: New types: gcry_pk_generate_t,
gcry_pk_check_secret_key_t, gcry_pk_encrypt_t, gcry_pk_decrypt_t,
gcry_pk_sign_t, gcry_pk_verify_t, gcry_pk_get_nbits_t,
gcry_pk_spec_t, gcry_md_init_t, gcry_md_write_t, gcry_md_final_t,
gcry_md_read_t, gcry_md_spec_t, gcry_cipher_setkey_t,
gcry_cipher_encrypt_t, gcry_cipher_decrypt_t,
gcry_cipher_stencrypt_t, gcry_cipher_stdecrypt_t,
gcry_cipher_spec_t, gcry_module_t.
2003-07-04 Moritz Schulte <moritz@g10code.com>
* module.c (_gcry_module_list): New function.
2003-07-02 Moritz Schulte <moritz@g10code.com>
* module.c (_gcry_module_lookup): Fixed typo.
* gcrypt.h: Added all definitions and declarations necessary for
the new ac interface.
2003-06-30 Moritz Schulte <moritz@g10code.com>
* g10lib.h: Added declarations: _gcry_pk_module_lookup,
_gcry_pk_module_release.
2003-06-18 Werner Koch <wk@gnupg.org>
* benchmark.c (cipher_bench): Adjusted for new API of get_blklen
and get_keylen.
* gcrypt.h (gcry_cipher_get_algo_blklen)
(gcry_cipher_get_algo_keylen): Replaced macro by funcion.
2003-06-18 Moritz Schulte <moritz@g10code.com>
* cipher.h: Renamed types GcryDigestSpec, GcryCipherSpec and
GcryPubkeySpec into: gcry_digest_spec_t, gcry_cipher_spec_t and
gcry_pubkey_spec_t.
(gcry_pubkey_spec): Defined member `id' as unsigned.
(gcry_digest_spec): Likewise.
(gcry_cipher_spec): Likewise.
* module.c (_gcry_module_id_new): New function.
(_gcry_module_add): Generate a new ID via _gcry_module_id_new in
case `id' is zero.
* g10lib.h, module.c: Replace old type GcryModule with newer one:
gcry_module_t.
* module.c (_gcry_module_add): Added argument `id', use it.
* g10lib.h: Added declaration: _gcry_module_lookup_id.
(_gcry_module_add): Added argument `id'.
* module.c (_gcry_module_lookup_id): New function.
* g10lib.h (struct gcry_module): New member: id.
* gcrypt.h: New type: gcry_handler_progress_t,
gcry_handler_alloc_t, gcry_haandler_secure_check_t,
gcry_handler_realloc_t, gcry_handler_free_t,
gcry_handler_no_mem_t, gcry_handler_error_t, gcry_handler_log_t.
Use new types.
* cipher.h: Include <gcrypt.h>.
New types: gcry_pk_generate_t, gcry_pk_check_secret_key_t,
gcry_pk_encrypt_t, gcry_pk_decrypt_t, gcry_pk_sign_t,
gcry_pk_verify_t, gcry_pk_get_nbits_t, gcry_md_init_t,
gcry_md_write_t, gcry_md_final_t, gcry_md_read_t,
gcry_cipher_setkey_t, gcry_cipher_encrypt_t,
gcry_cipher_decrypt_t, gcry_cipher_stencrypt_t,
gcry_cipher_stdecrypt_t.
Use new types.
2003-06-17 Moritz Schulte <moritz@g10code.com>
* Makefile.am (AM_CFLAGS): Added: @GPG_ERROR_CFLAGS@.
2003-06-16 Moritz Schulte <moritz@g10code.com>
* g10lib.h: Replace last occurences of old type names with newer
names (i.e. replace MPI with gcry_mpi_t).
* mpi.h: Likewise.
* sexp.c: Likewise.
2003-06-15 Moritz Schulte <moritz@g10code.com>
* testapi.c (test_genkey): Use gpg_strerror instead of
gcry_strerror.
* global.c (gcry_control): Fixed typo.
* misc.c (_gcry_fatal_error): Use gpg_strerror instead of
gcry_strerror.
* types.h (STRLIST): Removed type since it is not used.
2003-06-11 Moritz Schulte <moritz@g10code.com>
* global.c (global_init): Call: _gcry_cipher_init, _gcry_md_init,
_gcry_pk_init.
* g10lib.h: Declare: _gcry_cipher_init, _gcry_md_init,
_gcry_pk_init.
* global.c (gcry_strerror): Remove compatibility code.
* Makefile.am: Remove support libgpg-error special handling.
(AM_CPPFLAGS): Add @GPG_ERROR_CFLAGS@
* gcrypt.h: Likewise.
2003-06-13 Werner Koch <wk@gnupg.org>
* gcrypt.h (gcry_md_get_algo): Reverted to old API. This is a
convenience function anyway and error checking is not approriate.
(gcry_md_is_enabled): New.
(gcry_md_is_secure): Replaced macro by function and reverted to old
API.
2003-06-11 Werner Koch <wk@gnupg.org>
* gcrypt.h (GCRYERR): Define _GCRY_ERR_SOURCE_DEFAULT instead of
GPG_ERR_SOURCE_DEFAULT, so that libgpg-error still works despite
the use of the old gcrypt error codes.
(gcry_md_copy): Swapped arguments.
2003-06-09 Moritz Schulte <moritz@g10code.com>
* Makefile.am: Support for libgpg-error.
2003-06-08 Moritz Schulte <moritz@g10code.com>
* sexp.c (gcry_sexp_create): Expect sane error values from
gcry_sexp_canon_len instead of the `historical' values.
2003-06-07 Moritz Schulte <moritz@g10code.com>
* ath.c, ath.c, ath-pth.c, ath-pthread.c, benchmark.c, cipher.h,
g10lib.h, gcrypt.h, global.c, misc.c, missing-string.c, module.c,
mpi.h, secmem.c, secmem.h, sexp.c, stdmem.c, stdmem.h, testapi.c,
types.h: Edited all preprocessor instructions to remove whitespace
before the '#'. This is not required by C89, but there are some
compilers out there that don't like it. Replaced any occurence of
the now deprecated type names with the new ones.
* gcrypt.h: Re-organized checking for gcc features; New macro:
_GCRY_GCC_ATTR_DEPRECATED.
Include copy of libgpg-error's gpg-error.h in order to make it
easy to build libgcrypt without needing libgpg-error.h.
(GCRY_MPI, GcryMPI, GCRY_SEXP, GcrySexp, GCRY_CIPHER_HD,
GcryCipherHd, GCRY_MD_HD, GcryMDHd): Declared deprecated.
(gcry_mpi_t, gcry_sexp_t, gcry_cipher_hd_t, gcry_md_hd_t): New
types.
2003-06-04 Moritz Schulte <moritz@g10code.com>
* sexp.c (sexp_sscan): New argument: arg_list, adjusted all
callers.
(ARG_NEXT): New macro.
(sexp_sscan): Use ARG_NEXT for receiving format string arguments.
(gcry_sexp_build_array): New function.
2003-06-02 Moritz Schulte <moritz@g10code.com>
* gcrypt.h: Added some comments describing the gcry_sexp_*
functions.
Include <gpg-error.h> instead of <gpg/error.h>.
2003-06-01 Moritz Schulte <moritz@g10code.com>
* sexp.c (OLDPARSECODE): Removed macro...
(gcry_sexp_canon_len): ... and do not use it.
* gcrypt.h (gcry_errno): Removed declaration.
* g10lib.h (string_to_pubkey_algo, pubkey_algo_to_string,
pubkey_nbits): Removed declarations for non-existing functions.
2003-05-31 Moritz Schulte <moritz@g10code.com>
* cipher.h (is_RSA, is_ELGAMAL): Removed macros.
* g10lib.h (set_lasterr): Removed macro.
(_gcry_set_lasterr): Removed declaration.
* gcrypt.h: Changed declarations for: gcry_pk_algo_info,
gcry_md_open, gcry_md_copy, gcry_md_algo_info, gcry_md_info,
gcry_md_get_algo, gcry_random_add_bytes.
(gcry_md_is_secure): Adjust macro for new API.
2003-05-29 Moritz Schulte <moritz@g10code.com>
* gcrypt.h: Changed declarations for: gcry_cipher_open,
gcry_cipher_info, gcry_cipher_algo_info.
(gcry_cipher_get_algo_keylen): Adjuster for new
gcry_cipher_algo_info interface.
(gcry_cipher_get_algo_blklen): Likewise.
* global.c (gcry_errno): Removed function.
(gcry_strerror): Do not use gcry_errno.
(_gcry_set_lasterr): Removed function.
(last_ec): Removed variable.
2003-05-27 Moritz Schulte <moritz@g10code.com>
* gcrypt.h (enum gcry_cipher_algos): Make Serpent IDs do not
conflict with OpenPGP. Reported by Timo Schulz.
* global.c (gcry_control): Fixed name of enum list.
2003-05-25 Moritz Schulte <moritz@g10code.com>
* cipher.h (gcry_cipher_spec): Adjust return type of `setkey' for
libgpg-error.
(gcry_pubkey_spec): Adjust return type of `generate',
`check_secret_key', `encrypt', `decrypt', `sign' and `verify' for
libgpg-error.
* sexp.c (gcry_sexp_canon_len): Adjusted for libgpg-error.
(gcry_sexp_create): Likewise.
(gcry_sexp_new): Likewise.
(sexp_sscan): Likewise.
(gcry_sexp_build): Likewise.
(gcry_sexp_sscan): Likewise.
* module.c (_gcry_module_add): Likewise.
* global.c (last_ec): Change type to gpg_error_t.
(gcry_control): Adjust for libgpg-error.
(gcry_errno): Likewise.
(gcry_strerror): Likewise.
(_gcry_set_lasterr): Likewise.
(gcry_xmalloc): Likewise.
(gcry_xrealloc): Likewise.
2003-05-22 Moritz Schulte <moritz@g10code.com>
* types.h: Merged code from GnuPG regarding U64_C.
* missing-string.c (strsep): Removed function.
* g10lib.h: Removed declarations: strsep, strlwr.
* secmem.c (secmem_lock): New variable.
(SECMEM_LOCK, SECMEM_UNLOCK): New macros.
(_gcry_secmem_set_flags): Use SECMEM_LOCK and SECMEM_UNLOCK.
(_gcry_secmem_get_flags): Likewise.
(_gcry_secmem_init): Likewie.
(_gcry_secmem_malloc): Likewise.
(_gcry_secmem_free): Likewise.
(_gcry_secmem_malloc): Renamed to ...
(_gcry_secmem_malloc_internal): ... this.
(_gcry_secmem_malloc): New function, use SECMEM_LOCK,
SECMEM_UNLOCK, call _gcry_secmem_malloc_internal.
(_gcry_secmem_free): Renamed to ...
(_gcry_secmem_free_internal): ... this.
(_gcry_secmem_free): New function, use SECMEM_LOCK, SECMEM_UNLOCK,
call _gcry_secmem_free_internal.
(_gcry_secmem_realloc): Use SECMEM_LOCK, SECMEM_UNLOCK, call
_gcry_secmem_malloc_internal and _gcry_secmem_free_internal.
(_gcry_private_is_secure): Use SECMEM_LOCK, SECMEM_UNLOCK.
(_gcry_secmem_dump_stats): Likewise.
(_gcry_secmem_malloc_internal): Removed unused variable:
compressed.
Include "ath.h".
2003-05-21 Moritz Schulte <moritz@g10code.com>
* gcrypt.h (GCRY_CIPHER_SERPENT128, GCRY_CIPHER_SERPENT192,
GCRY_CIPHER_SERPENT256): New symbols.
2003-05-19 Moritz Schulte <moritz@g10code.com>
* gcrypt.h: Reversed changes from 2003-03-03 since they would have
been an unnecessary ABI break.
2003-05-13 Moritz Schulte <moritz@g10code.com>
* secmem.c (stats_update): New function.
(BLOCK_HEAD_SIZE): New symbol.
(MB_FLAG_ACTIVE): New symbol.
(ADDR_TO_BLOCK, BLOCK_VALID): New macros.
(mb_get_next): New function.
(mb_get_prev): New function.
(mb_merge): New function.
(mb_get_new): New function.
(unused_blocks): Removed variable.
(init_pool): Initialize new memory pool.
(_gcry_secmem_malloc): Use new heap management code.
(_gcry_secmem_free): Likewise.
(_gcry_secmem_realloc): Likewise.
Renamed type MEMBLOCK to memblock_t.
2003-04-27 Moritz Schulte <moritz@g10code.com>
* cipher.h (gcry_pubkey_spec): New member: sexp_names.
2003-04-23 Moritz Schulte <moritz@g10code.com>
* cipher.h (gcry_pubkey_spec): Removed members: npkey, nskey,
nenc, nsig.
(gcry_pubkey_spec): Added members: elements_pkey, elements_skey,
elements_enc, elements_sig, elements_grip.
2003-04-17 Moritz Schulte <moritz@g10code.com>
* g10lib.h (GcryModule): New typedef.
* gcrypt.h (gcry_cipher_register, gcry_cipher_unregister,
gcry_digest_register, gcry_digest_unregister,
gcry_pubkey_register, gcry_pubkey_unregister): Function
declarations removed - for now.
* gcrypt.h (GcryModule): Declaration removed.
* gcrypt.h (GcryPubkeySpec, GcryDigestSpec, GcryCipherSpec):
Types Moved...
* cipher.h: ... here.
2003-04-17 Moritz Schulte <moritz@g10code.com>
* cipher.h: Declare digest_spec_sha512 and digest_spec_384.
2003-04-16 Moritz Schulte <moritz@g10code.com>
* module.c (_gcry_module_use): New function.
* g10lib.h (_gcry_module_use): Declare function.
* libgcrypt-config.in: Support for --algorithms switch, which
prints the algorithms included in the built libgcrypt.
* global.c (gcry_set_progress_handler): Register progress
functions depending on the enabled algorithms.
2003-04-07 Moritz Schulte <moritz@g10code.com>
* Makefile.am (libgcrypt_la_SOURCES): Added module.c
* module.c: New file.
(_gcry_module_add): New function.
(_gcry_module_drop): New function.
(_gcry_module_lookup): New function.
(_gcry_module_release): New function.
* g10lib.h (GcryModule): New types.
(FLAG_MODULE_DISABLED): New symbol.
Added declarations for _gcry_module_add, _gcry_module_release and
_gcry_module_lookup.
* gcrypt.h: New types: GcryPubkeySpec, GcryDigestSpec,
GcryCipherSpec.
Added declarations for: gcry_cipher_register,
gcry_cipher_unregister, gcry_digest_register,
gcry_digest_unregister, gcry_pubkey_register and
gcry_pubkey_unregister.
* cipher.h: Removed symbols: CIPHER_ALGO_NONE, CIPHER_ALGO_IDEA,
CIPHER_ALGO_3DES, CIPHER_ALGO_CAST5, CIPHER_ALGO_BLOWFISH,
CIPHER_ALGO_SAFER_SK128, CIPHER_ALGO_DES_SK, CIPHER_ALGO_TWOFISH,
CIPHER_ALGO_TWOFISH_OLD, CIPHER_ALGO_DUMMY, PUBKEY_USAGE_SIG,
PUBKEY_USAGE_ENC, DIGEST_ALGO_MD5, DIGEST_ALGO_SHA1,
DIGEST_ALGO_RMD160, DIGEST_ALGO_TIGER, PUBKEY_ALGO_RSA,
PUBKEY_ALGO_RSA_E, PUBKEY_ALGO_RSA_S, PUBKEY_ALGO_DSA,
PUBKEY_ALGO_ELGAMAL, PUBKEY_ALGO_ELGAMAL_E.
2003-04-02 Moritz Schulte <moritz@g10code.com>
* benchmark.c (md_bench): Fix error message.
2003-03-31 Moritz Schulte <moritz@g10code.com>
* benchmark.c (cipher_bench): Added CTR mode.
2003-03-30 Simon Josefsson <jas@extundo.com>
* gcrypt.h (enum gcry_control_cmds): Add GCRY_SET_CTR.
(enum gcry_cipher_modes): Add GCRY_CIPHER_MODE_CTR.
(gcry_cipher_setctr): New macro to set counter.
2003-03-19 Moritz Schulte <moritz@g10code.com>
* cipher.h (PUBKEY_FLAG_NO_BLINDING): New symbol.
2003-03-22 Simon Josefsson <jas@extundo.com>
* gcrypt.h: Add GCRYCTL_SET_CBC_MAC and GCRY_CIPHER_CBC_MAC.
2003-03-19 Werner Koch <wk@gnupg.org>
* g10lib.h: Adjusted primegen.c prototypes.
2003-03-12 Werner Koch <wk@gnupg.org>
* sexp.c (sexp_sscan): Initialize NM. Thanks to Ian Peters for
valgrinding this.
2003-03-06 Moritz Schulte <mo@g10code.com>
* secmem.h (GCRY_SECMEM_FLAG_NO_WARNING,
GCRY_SECMEM_FLAG_SUSPEND_WARNING): New symbols.
* global.c (gcry_control): Use
GCRY_SECMEM_FLAG_{NO,SUSPEND}_WARNING, instead of hard-coded
values.
* secmem.c (_gcry_secmem_set_flags): Likewise.
* secmem.c (_gcry_secmem_get_flags): Likewise.
2003-03-03 Moritz Schulte <moritz@g10code.com>
* misc.c: Removed old FIXME, since there is already a function to
set the value of `verbosity_level'.
* gcrypt.h: Removed enumeration list: gcry_ctl_cmds.
New enumeration lists: gcry_global_control_cmds,
gcry_control_cmds, gcry_info_cmds, gcry_algo_info_cmds.
2003-03-02 Moritz Schulte <moritz@g10code.com>
* gcrypt.h (gcry_cipher_reset): New macro for resetting a handle.
2003-02-28 Moritz Schulte <moritz@g10code.com>
* secmem.c (DEFAULT_PAGESIZE): New symbol.
(init_pool): Use DEFAULT_PAGESIZE.
2003-02-23 Moritz Schulte <moritz@g10code.com>
* secmem.h: Fix typo in declaration of _gcry_secmem_term.
* sexp.c: Move macro definitions of `digitp', `octdigit', `alphap'
and `hexdigit' ...
* g10lib.h: ... here.
* misc.c (_gcry_burn_stack): New function (former name:
burn_stack).
* g10lib.h (burn_stack): Declare _gcry_burn_stack().
2003-01-24 Werner Koch <wk@gnupg.org>
* global.c (gcry_set_progress_handler): Register a random progress
handler.
2003-01-23 Werner Koch <wk@gnupg.org>
* gcrypt.h (GCRY_ENABLE_QUICK_RANDOM): New.
* global.c (gcry_control): Make use of it.
2003-01-21 Werner Koch <wk@gnupg.org>
* gcrypt.h (gcry_random_add_bytes): Add QUALITY argument.
2003-01-21 Timo Schulz <twoaday@freakmail.de>
* gcrypt.h (gcry_random_add_bytes): New.
2003-01-20 Simon Josefsson <jas@extundo.com>
* gcrypt.h (gcry_md_algos): Add GCRY_MD_CRC32,
GCRY_MD_CRC32_RFC1510, GCRY_MD_CRC24_RFC2440.
2003-01-16 Werner Koch <wk@gnupg.org>
* gcrypt.h (gcry_md_write): Changed type of 2nd argument to void*.
(gcry_md_hash_buffer): Changed type of both buffers to void*.
(gcry_md_setkey): Changed type of 2nd argument to void*.
(gcry_md_get_asnoid): New.
2003-01-15 Werner Koch <wk@gnupg.org>
* sexp.c (gcry_sexp_length): Fixed. This was seriously broken.
2003-01-14 Werner Koch <wk@gnupg.org>
* gcrypt.h (GCRYERR_INV_FLAG), global.c (gcry_strerror): New.
2003-01-02 Werner Koch <wk@gnupg.org>
* libgcrypt.vers: Temporary export _gcry_generate_elg_prime for
use by GNUTLS.
2002-12-21 Werner Koch <wk@gnupg.org>
* gcrypt.h: Make use of gcc's pure and malloc attributes
(gcry_md_putc): Use a helper variable to avoid multiple
evaluation of H.
* g10lib.h, stdmem.h, secmem.h: Use gcc attributes pure and malloc.
* stdmem.c (use_m_guard): Don't default to yes.
2002-12-19 Werner Koch <wk@gnupg.org>
* global.c (global_init): The meat was never run due to a faulty
check. Thanks to Nikos for pointing this out.
* global.c (gcry_control): Return 1 and not -1 for the
initialization tests.
* libgcrypt.vers: New.
* Makefile.am: Use this instead of the build symbol file.
* global.c (gcry_control) <initialization>: Call the random module
initializer to make sure that the pool lock flag has been
initialized.
2002-12-09 Werner Koch <wk@gnupg.org>
* global.c (gcry_calloc,gcry_calloc_secure): Check for overflow.
Noted by Florian Weimer.
2002-11-10 Simon Josefsson <jas@extundo.com>
* gcrypt.h (gcry_ctl_cmds): New GCRYCTL_SET_CBC_CTS control flag.
(gcry_cipher_flags): New GCRY_CIPHER_CBC_CTS gcry_cipher_open() flag.
(gcry_cipher_cts): New macro for toggling CTS.
2002-11-10 Werner Koch <wk@gnupg.org>
* gcrypt.h (GCRY_MD_MD4): New. We use a non OpenPGP value here.
2002-09-20 Werner Koch <wk@gnupg.org>
* ath.c: Include sys.time.h if sys/select.h does not exist.
(ath_select, ath_waitpid): Shortcut for Windows.
* ath.h: Include some Windows headers. By Timo.
2002-09-18 Werner Koch <wk@gnupg.org>
* ath.h: Prefix ath_deinit.
2002-09-17 Werner Koch <wk@gnupg.org>
* benchmark.c: New.
(mpi_bench, do_powm): Add a a simple test for RSA.
* global.c (global_init): New. Use it instead of the setting
any_init_done. Initialize the ATH system.
(gcry_check_version): Hook global_init in. This is the suggested
way to initialize the library.
(_gcry_no_internal_locking): Removed. We simply call a ath_deinit
and leave it to ATH to disbale the locking.
* ath.c, ath.h, ath-pth.c, ath-pthread.c: New. Taken from GPGME.
* mutex.h: Removed.
* Makefile.am (ath_components): New.
2002-09-16 Werner Koch <wk@gnupg.org>
* secmem.c (_gcry_secmem_dump_stats): Replaced fprintf by log_*.
2002-08-23 Werner Koch <wk@gnupg.org>
* missing-string.c: Removed unneeded strlwr.
* libgcrypt.m4: Made much more simple.
* libgcrypt-config.in: Made --prefix work for --libs.
2002-08-14 Werner Koch <wk@gnupg.org>
* gcrypt.h: Add GCRY_CIPGER_DES. Included string.h for size_t.
Suggested by Simon Josefsson.
2002-07-25 Werner Koch <wk@gnupg.org>
* cipher.h: Added prototypes for progress functions.
* global.c: Include cipher.h for those prototypes.
* stdmem.c (_gcry_private_realloc): Replaced void* by char * for
pointer arithmetic reasons. Noted by Stephan Austermuehle.
2002-06-24 Werner Koch <wk@gnupg.org>
* missing-string.c: Include ctype.h.
* gcrypt.h (gcry_mpi_invm, gcry_mpi_div, gcry_mpi_mod)
(gcry_mpi_swap): New.
2002-06-18 Werner Koch <wk@gnupg.org>
* gcrypt.h: Added a bunch of brief function descriptions.
2002-05-21 Werner Koch <wk@gnupg.org>
* misc.c (_gcry_log_printf): Don't initialize a va_list. Noted by
Jeff Johnson.
* global.c (gcry_set_progress_handler): New.
* gcrypt.h: Replaced the typedef for byte.
2002-05-16 Werner Koch <wk@gnupg.org>
* missing-string.c: New.
* gcrypt.h: Add new error codes GCRYERR_SEXP_ and typedefs
GcryMPI, GcrySexp, GcryCipherHd, GcryMDHd as aliases for the old
ones using an underscore.
* global.c (gcry_strerror): Add strings fro the new error codes.
* sexp.c (gcry_sexp_canon_len): Use a macro to convert from new to
old error codes.
(gcry_sexp_create,gcry_sexp_new): New.
2002-05-15 Werner Koch <wk@gnupg.org>
* mutex.h (DEFINE_LOCAL_MUTEX): Macro to define a mutex and
initialize it so that we can detect an unitialized mutex and don't
read from stdin.
2002-05-14 Werner Koch <wk@gnupg.org>
Changed license of all files to the LGPL.
2002-05-07 Werner Koch <wk@gnupg.org>
* global.c (gcry_control): Add commands
GCRYCTL_ANY_INITIALIZATION_P and GCRYCTL_INITIALIZATION_FINISHED_P
so that other libraries are able to check for required
initializations.
2002-05-02 Werner Koch <wk@gnupg.org>
* gcrypt.h (GCRYCTL_DISABLE_INTERNAL_LOCKING): New.
* global.c (gcry_control): Implement it.
(_gcry_no_internal_locking): New.
* mutex.h: Prefixed all fucntions with _gcry_. Bypass all
functions when desired.
* gcrypt.h (GCRYCTL_DISABLE_SECMEM): New.
* global.c (gcry_control,gcry_malloc_secure,gcry_is_secure):
Implement it here.
* secmem.c (_gcry_private_is_secure): Return false if the pool is
not initialized.
* gcrypt.h (GCRYCTL_INITIALIZATION_FINISHED): New.
* gcrypt.h (gcry_cipher_algos): Replaced RINDAEL by AES and change
the macros to expand from rijdael to aes.
* stdmem.c (_gcry_private_malloc): Return NULL for 0 byte allocation.
(_gcry_private_malloc_secure): Ditto.
* g10lib.h: Copied the JNLIB_GCC macros from ../jnlib/mischelp.h
and removed the inclusion of that file.
2002-04-15 Werner Koch <wk@gnupg.org>
* global.c (gcry_strdup): New.
2002-03-18 Werner Koch <wk@gnupg.org>
* mutex.h: New file with a portable thread mutex implementation
written by Marcus Brinkmann. Taken from GPGME.
2002-02-18 Werner Koch <wk@gnupg.org>
* sexp.c (gcry_sexp_sscan): Don't initialize the dummy
variable. Suggested by Jordi Mallach.
2002-01-31 Werner Koch <wk@gnupg.org>
* sexp.c (suitable_encoding,convert_to_hex,convert_to_string)
(convert_to_token): New.
(gcry_sexp_sprint): Better formatting of advanced encoding, does
now insert LFs and escapes all unprintable characters.
(unquote_string): New.
(sexp_sscan): Implemented the missing conversion of quoted strings.
2002-01-26 Werner Koch <wk@gnupg.org>
* libgcrypt-config.in: Add copyright notice.
2002-01-11 Werner Koch <wk@gnupg.org>
* sexp.c (gcry_sexp_canon_len): Fixed last change.
2002-01-01 Timo Schulz <ts@winpt.org>
* stdmem.c (_gcry_private_realloc): If pointer is NULL now realloc
behaves like malloc.
2001-12-20 Werner Koch <wk@gnupg.org>
* sexp.c (gcry_sexp_canon_len): Describe the error codes and
return an error if this is not a S-Exp; i.e. it does not start
with an open parenthesis.
2001-12-18 Werner Koch <wk@gnupg.org>
* sexp.c (gcry_sexp_canon_len): Fixed the test on NULL buffer.
* Makefile.am (DISTCLEANFILES): Include libgcrypt.sym
* sexp.c: Removed the commented test code because we now have a
test in ../tests/
2001-12-17 Werner Koch <wk@gnupg.org>
* sexp.c (gcry_sexp_canon_len): New.
2001-12-11 Werner Koch <wk@gnupg.org>
* gcrypt.h: Fixed AES128 macro, add enum for OFB mode.
2001-12-05 Werner Koch <wk@gnupg.org>
* misc.c (_gcry_log_printf): New.
* sexp.c (dump_string,gcry_sexp_dump): Use logging functions
instead of stderr.
2001-11-16 Werner Koch <wk@gnupg.org>
* gcrypt.h: New constant GCRYCTL_IS_ALGO_ENABLED.
2001-10-02 Werner Koch <wk@gnupg.org>
* gcrypt.h: Removed a couple of trailing commas.
2001-08-28 Werner Koch <wk@gnupg.org>
* sexp.c (sexp_sscan): Add an argument to enable the
arg_ptr. Changed all callers. Suggested by Tom Holroyd.
2001-08-03 Werner Koch <wk@gnupg.org>
* global.c (gcry_strerror): Updated list of error codes.
2001-07-23 Werner Koch <wk@gnupg.org>
* gcrypt.h: Replaced the last ulong. Noted by Rami Lehti.
2001-05-31 Werner Koch <wk@gnupg.org>
* gcrypt.h, mpi.h: Made some mpi functions public.
* wrapper.c: Removed.
* global.c: Renamed all g10_ prefixed functions which had wrappers
to gcry_xxx. So we now use the exported memory functions inernally.
Renamed all g10_ prefixed functions to _gcry_ prefixed ones.
* g10lib.h (_GCRYPT_IN_LIBGCRYPT): Replace defintion by a test on it.
2001-05-28 Werner Koch <wk@gnupg.org>
* libgcrypt.m4: Check GCRYPT_VERSION macro and not LIBGCRYPT_VERSION.
* mpi.h: Removed mpi_fromstr prototype.
2001-01-11 Werner Koch <wk@gnupg.org>
* Makefile.am (libgcrypt_la_SOURCES): Add mpi.h
2000-12-19 Werner Koch <wk@gnupg.org>
* types.h: Moved from ../include to here.
Major change:
Removed all GnuPG stuff and renamed this piece of software
to gcrypt.
2000-11-14 Werner Koch <wk@gnupg.org>
* mpi.h: Moved to ../mpi.
* Makefile.am (OMIT_DEPENDENCIES): Hack to work around dependency
problems.
2000-10-11 Werner Koch <wk@gnupg.org>
* mpi.h: Changed the way mpi_limb_t is defined.
2000-10-10 Werner Koch <wk@gnupg.org>
* Makefile.am: Take version-info from configure.
2000-10-09 Werner Koch <wk@gnupg.org>
* gcrypt.h: New cipher mode, new algo Arcfour and new error code
GCRYERR_INV_CIPHER_MODE.
* global.c (gcry_strerror): New errorcode.
Wed Oct 4 13:16:18 CEST 2000 Werner Koch <wk@openit.de>
* gcrypt.h (gcry_md_setkey): Replaced macro by function prototype.
Mon Sep 18 16:35:45 CEST 2000 Werner Koch <wk@openit.de>
* gcrypt.h (GCRYCTL_GET_ALGO_USAGE): New.
* secmem.c (secmem_realloc): check for failed secmem_malloc. By
Matt Kraai.
Mon Jul 31 10:04:47 CEST 2000 Werner Koch <wk@openit.de>
* sexp.c: Removed the datalen fields from list tags.
(gcry_sexp_car_data,gcry_sexp_cdr_data,gcry_sexp_car_mpi,
gcry_sexp_cdr_mpi): Removed.
(gcry_sexp_nth,gcry_sexp_nth_data,gcry_sexp_nth_mpi): New.
Fri Jul 28 18:19:11 CEST 2000 Werner Koch <wk@openit.de>
* sexp.c (sexp_sscan): Fixed reallocation to secure memory.
(new_empty_list): Removed
(gcry_sexp_length): New.
(gcry_sexp_enum): Removed.
(normalize): New. Reworked the whole thing to use NULL for an empty list.
(make_space): New instead of the macro.
Tue Jul 25 17:44:15 CEST 2000 Werner Koch <wk@openit.de>
* sexp.c: Major rewrite.
(gcry_sexp_sscan): Reordered arguments. Moved functionality to ..
(sexp_sscan): .. this.
(gcry_sexp_build): New.
(gcry_sexp_new_name_mpi, gcry_sexp_new_name_data, gcry_sexp_new_data,
gcry_sexp_new_mpi): Removed.
Fri Jul 14 19:38:23 CEST 2000 Werner Koch <wk@>
* gcrypt.h (gcry_md_start_debug, gcry_md_stop_debug): New.
(gcry_ctl_cmds): New control values
* sexp.c (gcry_sexp_sscan): Add hex format parsing.
* secmem.c (lock_pool): Check for ENOSYS return my mlock() on old SCOs.
(pool_is_mmapped): Made volatile.
(lock_pool): No more warning for QNX. By Sam Roberts.
(lock_pool,secmem_init): Additional check for dropped privs.
2000-03-21 09:18:48 Werner Koch (wk@habibti.gnupg.de)
* gcrypt.h (gcry_md_setkey): New.
(GCRY_MD_FLAG_HMAC): New.
Mon Jan 31 16:37:34 CET 2000 Werner Koch <wk@gnupg.de>
* Makefile.am: Add g10lib.h
Thu Jan 27 18:00:44 CET 2000 Werner Koch <wk@gnupg.de>
* sexp.c (gcry_sexp_sscan): Allow NULL for erroff.
Mon Jan 24 22:24:38 CET 2000 Werner Koch <wk@gnupg.de>
* sexp.c (gcry_sexp_alist): New.
Mon Jan 24 13:04:28 CET 2000 Werner Koch <wk@gnupg.de>
* secmem.c: Moved from ../util to here.
* secmem.h: New.
* stdmem.c: New. Based on the old ../util/memory.c.
* stdmem.h: New.
Wed Dec 8 21:58:32 CET 1999 Werner Koch <wk@gnupg.de>
* gcrypt.m4: New.
* gcrypt-config: New.
* mpi.h (mpi_get_nbit_info): Removed
(mpi_set_nbit_info): Removed.
(struct gcry_mpi): Removed the nbits field.
* misc.c (g10_log_verbosity): New.
* global.c (g10_xstrdup): New.
* mpiapi.c: Removed.
* mpi.h: Moved from ../include to here. Removed some obsolete
prototypes and the iobuf.h header.
* cipher.h: Moved from ../include to here. Removed the mpi.h header.
* g10lib.h: Moved from ../include to here.
Fri Nov 19 17:15:20 CET 1999 Werner Koch <wk@gnupg.de>
* sexp.c (dump_string): New. Taken from gnupg/util/miscutil.c.
(do_dump_list): s/print_string/dump_string/.
* testapi.c: New.
* mpiapi.c (gcry_mpi_randomize): Use new random API.
Sat Nov 13 17:44:23 CET 1999 Werner Koch <wk@gnupg.de>
* gloabl.c (gcry_control): Add cases for dumping random
and secmem stats.
Tue Oct 26 14:10:21 CEST 1999 Werner Koch <wk@gnupg.de>
* pkapi.c: Removed.
* symapi.c: Removed.
* g10lib.h: Moved to ../include.
* mdapi.c: Removed.
Wed Jul 7 13:08:40 CEST 1999 Werner Koch <wk@isil.d.shuttle.de>
* sexp.c: New.
Tue Dec 8 13:15:16 CET 1998 Werner Koch <wk@isil.d.shuttle.de>
* gcrypt.h: New
* mpiapi.c: New
Copyright (C) 1998,1999,2000,2001,2002,2003
- 2004,2005 Free Software Foundation, Inc.
+ 2004,2005,2006 Free Software Foundation, Inc.
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 file 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.
diff --git a/src/global.c b/src/global.c
index 73e02b48..77c2c99f 100644
--- a/src/global.c
+++ b/src/global.c
@@ -1,714 +1,726 @@
/* global.c - global control functions
* Copyright (C) 1998, 1999, 2000, 2001, 2002, 2003
* 2004, 2005, 2006 Free Software Foundation, Inc.
*
* This file is part of Libgcrypt.
*
* Libgcrypt 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.
*
* Libgcrypt 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, write to the Free Software
* Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA
*/
#include <config.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <stdarg.h>
#include <ctype.h>
#include <assert.h>
#include <limits.h>
#include <errno.h>
#include "g10lib.h"
#include "cipher.h"
#include "stdmem.h" /* our own memory allocator */
#include "secmem.h" /* our own secmem allocator */
#include "ath.h"
/****************
* flag bits: 0 : general cipher debug
* 1 : general MPI debug
*/
static unsigned int debug_flags;
static gcry_handler_alloc_t alloc_func;
static gcry_handler_alloc_t alloc_secure_func;
static gcry_handler_secure_check_t is_secure_func;
static gcry_handler_realloc_t realloc_func;
static gcry_handler_free_t free_func;
static gcry_handler_no_mem_t outofcore_handler;
static void *outofcore_handler_value = NULL;
static int no_secure_memory = 0;
static int any_init_done;
/* This is our handmade constructor. It gets called by any function
likely to be called at startup. The suggested way for an
application to make sure that this has been called is by using
gcry_check_version. */
static void
global_init (void)
{
gcry_err_code_t err = GPG_ERR_NO_ERROR;
if (any_init_done)
return;
any_init_done = 1;
err = ath_init ();
if (! err)
err = _gcry_cipher_init ();
if (! err)
err = _gcry_md_init ();
if (! err)
err = _gcry_pk_init ();
#if 0
if (! err)
err = _gcry_ac_init ();
#endif
if (err)
/* FIXME? */
BUG ();
}
static const char*
parse_version_number( const char *s, int *number )
{
int val = 0;
if( *s == '0' && isdigit(s[1]) )
return NULL; /* leading zeros are not allowed */
for ( ; isdigit(*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, int *micro )
{
s = parse_version_number( s, major );
if( !s || *s != '.' )
return NULL;
s++;
s = parse_version_number( s, minor );
if( !s || *s != '.' )
return NULL;
s++;
s = parse_version_number( s, micro );
if( !s )
return NULL;
return s; /* patchlevel */
}
/****************
* Check that the the version of the library is at minimum the requested one
* and return the version string; return NULL if the condition is not
* satisfied. If a NULL is passed to this function, no check is done,
* but the version string is simply returned.
*/
const char *
gcry_check_version( const char *req_version )
{
const char *ver = VERSION;
int my_major, my_minor, my_micro;
int rq_major, rq_minor, rq_micro;
const char *my_plvl, *rq_plvl;
global_init ();
if ( !req_version )
return ver;
my_plvl = parse_version_string( ver, &my_major, &my_minor, &my_micro );
if ( !my_plvl )
return NULL; /* very strange our own version is bogus */
rq_plvl = parse_version_string( req_version, &rq_major, &rq_minor,
&rq_micro );
if ( !rq_plvl )
return NULL; /* req version string is invalid */
if ( my_major > rq_major
|| (my_major == rq_major && my_minor > rq_minor)
|| (my_major == rq_major && my_minor == rq_minor
&& my_micro > rq_micro)
|| (my_major == rq_major && my_minor == rq_minor
&& my_micro == rq_micro
&& strcmp( my_plvl, rq_plvl ) >= 0) ) {
return ver;
}
return NULL;
}
gcry_error_t
gcry_control (enum gcry_ctl_cmds cmd, ...)
{
gcry_err_code_t err = GPG_ERR_NO_ERROR;
static int init_finished = 0;
va_list arg_ptr;
va_start (arg_ptr, cmd);
switch (cmd)
{
case GCRYCTL_ENABLE_M_GUARD:
_gcry_private_enable_m_guard ();
break;
case GCRYCTL_ENABLE_QUICK_RANDOM:
_gcry_quick_random_gen (1);
break;
+ case 51: /* Should be GCRYCTL_FAKED_RANDOM_P but we want to sneak
+ that into the API for the sake of GnuPG 1.9 - there
+ we check the error code and print a warning message
+ if that call is not supported. The literal number is
+ used there as well. */
+ /* Return an error if the RNG is faked one (i.e. enabled by
+ ENABLE_QUICK_RANDOM. */
+ if (_gcry_random_is_faked ())
+ err = GPG_ERR_GENERAL;
+ break;
+
+
case GCRYCTL_DUMP_RANDOM_STATS:
_gcry_random_dump_stats ();
break;
case GCRYCTL_DUMP_MEMORY_STATS:
/*m_print_stats("[fixme: prefix]");*/
break;
case GCRYCTL_DUMP_SECMEM_STATS:
_gcry_secmem_dump_stats ();
break;
case GCRYCTL_DROP_PRIVS:
global_init ();
_gcry_secmem_init (0);
break;
case GCRYCTL_DISABLE_SECMEM:
global_init ();
no_secure_memory = 1;
break;
case GCRYCTL_INIT_SECMEM:
global_init ();
_gcry_secmem_init (va_arg (arg_ptr, unsigned int));
break;
case GCRYCTL_TERM_SECMEM:
global_init ();
_gcry_secmem_term ();
break;
case GCRYCTL_DISABLE_SECMEM_WARN:
_gcry_secmem_set_flags ((_gcry_secmem_get_flags ()
| GCRY_SECMEM_FLAG_NO_WARNING));
break;
case GCRYCTL_SUSPEND_SECMEM_WARN:
_gcry_secmem_set_flags ((_gcry_secmem_get_flags ()
| GCRY_SECMEM_FLAG_SUSPEND_WARNING));
break;
case GCRYCTL_RESUME_SECMEM_WARN:
_gcry_secmem_set_flags ((_gcry_secmem_get_flags ()
& ~GCRY_SECMEM_FLAG_SUSPEND_WARNING));
break;
case GCRYCTL_USE_SECURE_RNDPOOL:
global_init ();
_gcry_secure_random_alloc (); /* put random number into secure memory */
break;
case GCRYCTL_SET_RANDOM_SEED_FILE:
_gcry_set_random_seed_file (va_arg (arg_ptr, const char *));
break;
case GCRYCTL_UPDATE_RANDOM_SEED_FILE:
_gcry_update_random_seed_file ();
break;
case GCRYCTL_SET_VERBOSITY:
_gcry_set_log_verbosity (va_arg (arg_ptr, int));
break;
case GCRYCTL_SET_DEBUG_FLAGS:
debug_flags |= va_arg (arg_ptr, unsigned int);
break;
case GCRYCTL_CLEAR_DEBUG_FLAGS:
debug_flags &= ~va_arg (arg_ptr, unsigned int);
break;
case GCRYCTL_DISABLE_INTERNAL_LOCKING:
global_init ();
break;
case GCRYCTL_ANY_INITIALIZATION_P:
if (any_init_done)
err = GPG_ERR_GENERAL;
break;
case GCRYCTL_INITIALIZATION_FINISHED_P:
if (init_finished)
err = GPG_ERR_GENERAL;
break;
case GCRYCTL_INITIALIZATION_FINISHED:
/* This is a hook which should be used by an application after
all initialization has been done and right before any threads
are started. It is not really needed but the only way to be
really sure that all initialization for thread-safety has
been done. */
if (! init_finished)
{
global_init ();
/* Do only a basic random initialization, i.e. init the
mutexes. */
_gcry_random_initialize (0);
init_finished = 1;
}
break;
case GCRYCTL_SET_THREAD_CBS:
err = ath_install (va_arg (arg_ptr, void *), any_init_done);
break;
case GCRYCTL_FAST_POLL:
/* We need to do make sure that the random pool is really
initialized so that the poll fucntion is not a NOP. */
_gcry_random_initialize (1);
_gcry_fast_random_poll ();
break;
default:
err = GPG_ERR_INV_OP;
}
va_end(arg_ptr);
return gcry_error (err);
}
/* Return a pointer to a string containing a description of the error
code in the error value ERR. */
const char *
gcry_strerror (gcry_error_t err)
{
return gpg_strerror (err);
}
/* Return a pointer to a string containing a description of the error
source in the error value ERR. */
const char *
gcry_strsource (gcry_error_t err)
{
return gpg_strsource (err);
}
/* Retrieve the error code for the system error ERR. This returns
GPG_ERR_UNKNOWN_ERRNO if the system error is not mapped (report
this). */
gcry_err_code_t
gcry_err_code_from_errno (int err)
{
return gpg_err_code_from_errno (err);
}
/* Retrieve the system error for the error code CODE. This returns 0
if CODE is not a system error code. */
int
gcry_err_code_to_errno (gcry_err_code_t code)
{
return gpg_err_code_from_errno (code);
}
/* Return an error value with the error source SOURCE and the system
error ERR. */
gcry_error_t
gcry_err_make_from_errno (gpg_err_source_t source, int err)
{
return gpg_err_make_from_errno (source, err);
}
/* Return an error value with the system error ERR. */
gcry_err_code_t
gcry_error_from_errno (int err)
{
return gcry_error (gpg_err_code_from_errno (err));
}
/****************
* NOTE: All 5 functions should be set. */
void
gcry_set_allocation_handler (gcry_handler_alloc_t new_alloc_func,
gcry_handler_alloc_t new_alloc_secure_func,
gcry_handler_secure_check_t new_is_secure_func,
gcry_handler_realloc_t new_realloc_func,
gcry_handler_free_t new_free_func)
{
global_init ();
alloc_func = new_alloc_func;
alloc_secure_func = new_alloc_secure_func;
is_secure_func = new_is_secure_func;
realloc_func = new_realloc_func;
free_func = new_free_func;
}
/****************
* Set an optional handler which is called in case the xmalloc functions
* ran out of memory. This handler may do one of these things:
* o free some memory and return true, so that the xmalloc function
* tries again.
* o Do whatever it like and return false, so that the xmalloc functions
* use the default fatal error handler.
* o Terminate the program and don't return.
*
* The handler function is called with 3 arguments: The opaque value set with
* this function, the requested memory size, and a flag with these bits
* currently defined:
* bit 0 set = secure memory has been requested.
*/
void
gcry_set_outofcore_handler( int (*f)( void*, size_t, unsigned int ),
void *value )
{
global_init ();
outofcore_handler = f;
outofcore_handler_value = value;
}
gcry_err_code_t
_gcry_malloc (size_t n, unsigned int flags, void **mem)
{
gcry_err_code_t err = 0;
void *m;
if ((flags & GCRY_ALLOC_FLAG_SECURE) && !no_secure_memory)
{
if (alloc_secure_func)
m = (*alloc_secure_func) (n);
else
m = _gcry_private_malloc_secure (n);
}
else
{
if (alloc_func)
m = (*alloc_func) (n);
else
m = _gcry_private_malloc (n);
}
if (!m)
{
if (!errno)
errno = ENOMEM;
err = gpg_err_code_from_errno (errno);
}
else
*mem = m;
return err;
}
void *
gcry_malloc (size_t n)
{
void *mem = NULL;
_gcry_malloc (n, 0, &mem);
return mem;
}
void *
gcry_malloc_secure (size_t n)
{
void *mem = NULL;
_gcry_malloc (n, GCRY_ALLOC_FLAG_SECURE, &mem);
return mem;
}
int
gcry_is_secure (const void *a)
{
if (no_secure_memory)
return 0;
if (is_secure_func)
return is_secure_func (a) ;
return _gcry_private_is_secure (a);
}
void
_gcry_check_heap( const void *a )
{
/* FIXME: implement this*/
#if 0
if( some_handler )
some_handler(a)
else
_gcry_private_check_heap(a)
#endif
}
void *
gcry_realloc (void *a, size_t n)
{
void *p;
if (realloc_func)
p = realloc_func (a, n);
else
p = _gcry_private_realloc (a, n);
if (!p && !errno)
errno = ENOMEM;
return p;
}
void
gcry_free( void *p )
{
if( !p )
return;
if (free_func)
free_func (p);
else
_gcry_private_free (p);
}
void *
gcry_calloc (size_t n, size_t m)
{
size_t bytes;
void *p;
bytes = n * m; /* size_t is unsigned so the behavior on overflow is
defined. */
if (m && bytes / m != n)
{
errno = ENOMEM;
return NULL;
}
p = gcry_malloc (bytes);
if (p)
memset (p, 0, bytes);
return p;
}
void *
gcry_calloc_secure (size_t n, size_t m)
{
size_t bytes;
void *p;
bytes = n * m; /* size_t is unsigned so the behavior on overflow is
defined. */
if (m && bytes / m != n)
{
errno = ENOMEM;
return NULL;
}
p = gcry_malloc_secure (bytes);
if (p)
memset (p, 0, bytes);
return p;
}
/* Create and return a copy of the null-terminated string STRING. If
it is contained in secure memory, the copy will be contained in
secure memory as well. In an out-of-memory condition, NULL is
returned. */
char *
gcry_strdup (const char *string)
{
char *string_cp = NULL;
size_t string_n = 0;
string_n = strlen (string);
if (gcry_is_secure (string))
string_cp = gcry_malloc_secure (string_n + 1);
else
string_cp = gcry_malloc (string_n + 1);
if (string_cp)
strcpy (string_cp, string);
return string_cp;
}
void *
gcry_xmalloc( size_t n )
{
void *p;
while ( !(p = gcry_malloc( n )) ) {
if( !outofcore_handler
|| !outofcore_handler( outofcore_handler_value, n, 0 ) ) {
_gcry_fatal_error(gpg_err_code_from_errno (errno), NULL );
}
}
return p;
}
void *
gcry_xrealloc( void *a, size_t n )
{
void *p;
while ( !(p = gcry_realloc( a, n )) ) {
if( !outofcore_handler
|| !outofcore_handler( outofcore_handler_value, n, 2 ) ) {
_gcry_fatal_error(gpg_err_code_from_errno (errno), NULL );
}
}
return p;
}
void *
gcry_xmalloc_secure( size_t n )
{
void *p;
while ( !(p = gcry_malloc_secure( n )) ) {
if( !outofcore_handler
|| !outofcore_handler( outofcore_handler_value, n, 1 ) ) {
_gcry_fatal_error(gpg_err_code_from_errno (errno),
_("out of core in secure memory"));
}
}
return p;
}
void *
gcry_xcalloc( size_t n, size_t m )
{
size_t nbytes;
void *p;
nbytes = n * m;
if (m && nbytes / m != n)
{
errno = ENOMEM;
_gcry_fatal_error(gpg_err_code_from_errno (errno), NULL );
}
p = gcry_xmalloc ( nbytes );
memset ( p, 0, nbytes );
return p;
}
void *
gcry_xcalloc_secure( size_t n, size_t m )
{
size_t nbytes;
void *p;
nbytes = n * m;
if (m && nbytes / m != n)
{
errno = ENOMEM;
_gcry_fatal_error(gpg_err_code_from_errno (errno), NULL );
}
p = gcry_xmalloc_secure ( nbytes );
memset ( p, 0, nbytes );
return p;
}
char *
gcry_xstrdup (const char *string)
{
char *p;
while ( !(p = gcry_strdup (string)) )
{
size_t n = strlen (string);
int is_sec = !!gcry_is_secure (string);
if (!outofcore_handler
|| !outofcore_handler (outofcore_handler_value, n, is_sec) )
{
_gcry_fatal_error (gpg_err_code_from_errno (errno),
is_sec? _("out of core in secure memory"):NULL);
}
}
return p;
}
int
_gcry_get_debug_flag( unsigned int mask )
{
return debug_flags & mask;
}
/* It is often useful to get some feedback of long running operations.
This function may be used to register a handler for this.
The callback function CB is used as:
void cb (void *opaque, const char *what, int printchar,
int current, int total);
Where WHAT is a string identifying the the type of the progress
output, PRINTCHAR the character usually printed, CURRENT the amount
of progress currently done and TOTAL the expected amount of
progress. A value of 0 for TOTAL indicates that there is no
estimation available.
Defined values for WHAT:
"need_entropy" X 0 number-of-bytes-required
When running low on entropy
"primegen" '\n' 0 0
Prime generated
'!'
Need to refresh the prime pool
'<','>'
Number of bits adjusted
'^'
Looking for a generator
'.'
Fermat tests on 10 candidates failed
':'
Restart with a new random value
'+'
Rabin Miller test passed
"pk_elg" '+','-','.','\n' 0 0
Only used in debugging mode.
"pk_dsa"
Only used in debugging mode.
*/
void
gcry_set_progress_handler (void (*cb)(void *,const char*,int, int, int),
void *cb_data)
{
#if USE_DSA
_gcry_register_pk_dsa_progress (cb, cb_data);
#endif
#if USE_ELGAMAL
_gcry_register_pk_elg_progress (cb, cb_data);
#endif
_gcry_register_primegen_progress (cb, cb_data);
_gcry_register_random_progress (cb, cb_data);
}
diff --git a/tests/ChangeLog b/tests/ChangeLog
index c3b8b1ac..a91f749b 100644
--- a/tests/ChangeLog
+++ b/tests/ChangeLog
@@ -1,313 +1,317 @@
+2006-08-28 Werner Koch <wk@g10code.com>
+
+ * t-mpi-bit.c: New. Copied from development branch.
+
2005-09-13 Werner Koch <wk@g10code.com>
* random.c: New.
2005-06-16 Werner Koch <wk@g10code.com>
* basic.c (check_one_md): Made P unsigned.
2004-02-25 Werner Koch <wk@gnupg.org>
* Makefile.am (TEST): Add benchmark.
* benchmark.c (md_bench, cipher_bench): Allow NULL arg to to run
tests for all algorithms.
(main): Run all tests by default.
2004-02-03 Werner Koch <wk@gnupg.org>
* tsexp.c (basic): New pass to check secure memory switching.
2004-01-12 Moritz Schulte <mo@g10code.com>
* ac.c (check_one): Adjust to new ac API.
2003-11-22 Werner Koch <wk@gnupg.org>
* pubkey.c (check_keys_crypt): Fixed my last patch.
2003-11-11 Werner Koch <wk@gnupg.org>
* tsexp.c (basic): Add pass structure and a test for the %b
format.
2003-11-04 Werner Koch <wk@gnupg.org>
* Makefile.am (noinst_PROGRAMS): Use this so that test programs
get always build.
* keygen.c (check_nonce): New.
(main): Add a basic check for the nocen function.
2003-10-31 Werner Koch <wk@gnupg.org>
* basic.c (check_aes128_cbc_cts_cipher): Make it a prototype
* ac.c (check_run): Comment unused variable.
2003-10-10 Werner Koch <wk@gnupg.org>
* prime.c (check_primes): Generate a generator and avoid printing
unless in verbose mode.
2003-10-07 Werner Koch <wk@gnupg.org>
* tsexp.c (check_sscan): New.
2003-09-04 Werner Koch <wk@gnupg.org>
* pubkey.c (check_keys_crypt): Fix for compatibility mode.
2003-09-02 Moritz Schulte <mo@g10code.com>
* Makefile.am (TESTS): Added: prime.
* prime.c: New file.
2003-08-27 Moritz Schulte <mo@g10code.com>
* basic.c (check_ciphers): Added: Serpent.
Write braces around flags.
2003-08-04 Moritz Schulte <moritz@g10code.com>
* benchmark.c (do_powm): Adjust for new gcry_mpi_scan interface.
2003-07-23 Moritz Schulte <moritz@g10code.com>
* ac.c (key_copy): New function...
(check_one): ... use it.
2003-07-22 Moritz Schulte <moritz@g10code.com>
* basic.c (check_ciphers): Use gcry_cipher_map_name.
2003-07-18 Moritz Schulte <moritz@g10code.com>
* ac.c (check_run): Renamed to ...
(check_one): ... this, changed calling interface.
(check_run): New function.
* register.c: Adjust gcry_cipher_spec_t structure.
2003-07-14 Moritz Schulte <moritz@g10code.com>
* register.c: Adjust cipher specification structure.
* benchmark.c: New file.
* testapi.c: New file.
* Makefile.am (EXTRA_PROGRAMS): Set to: benchmark testapi.
(check_PROGRAMS): Set to: $(TESTS).
2003-07-12 Moritz Schulte <moritz@g10code.com>
* ac.c, basic.c, keygen.c, register.c, sexp.c, tsexp.c: Used
gcry_err* wrappers for libgpg symbols.
* basic.c (check_ciphers): Added: GCRY_CIPHER_TWOFISH128.
2003-07-08 Moritz Schulte <moritz@g10code.com>
* Makefile.am (LIBS): Remove: -lpthread.
* basic.c (check_one_cipher): Fix variable initialization. Thanks
to Simon Joseffson <jas@extundo.com>.
2003-07-07 Moritz Schulte <moritz@g10code.com>
* Makefile.am (TESTS): Added: register.
2003-07-05 Moritz Schulte <moritz@g10code.com>
* register.c (check_run): Adjusted for new gcry_cipher_register API.
2003-07-02 Moritz Schulte <moritz@g10code.com>
* Makefile.am (TESTS): Added: ac.
* ac.c: New file.
2003-06-18 Werner Koch <wk@gnupg.org>
* basic.c (check_cbc_mac_cipher): Adjusted for new API of get_blklen
and get_keylen.
(check_ctr_cipher): Ditto.
(check_one_cipher): Ditto.
(check_one_md): Adjusted for new API of gcry_md_copy.
2003-06-18 Moritz Schulte <moritz@g10code.com>
* register.c: Replace old type GcryModule with newer one:
gcry_module_t.
Adjusted for new API.
* Makefile.am (AM_CFLAGS): Added: @GPG_ERROR_CFLAGS@.
2003-06-15 Moritz Schulte <moritz@g10code.com>
* basic.c (get_keys_new): New function.
(do_check_one_pubkey): New function ...
(check_one_pubkey): ... use it.
(progress_handler): New function.
(main): Use gcry_set_progress_handler.
2003-06-14 Moritz Schulte <moritz@g10code.com>
* basic.c: Replaced calls to gcry_strerror with calls to
gpg_strerror.
(check_one_md): Adjust for new gcry_md_copy API.
* tsexp.c: Likewise.
* keygen.c: Likewise.
2003-06-12 Moritz Schulte <moritz@g10code.com>
* basic.c: Changed here and there, reorganized pubkey checks,
added DSA and ELG keys.
2003-06-09 Moritz Schulte <moritz@g10code.com>
* basic.c, keygen.c, pubkey.c, register.c, tsexp.c: Changed to use
new API.
2003-06-01 Moritz Schulte <moritz@g10code.com>
* tsexp.c (canon_len): Adjust for new gcry_sexp_canon_len API.
2003-05-26 Moritz Schulte <moritz@g10code.com>
* basic.c (verify_one_signature): Adjust for libgpg-error.
(check_pubkey_sign): Likewise.
(check_pubkey): Likewise.
* basic.c (check_pubkey_sign): Likewise.
* tsexp.c (canon_len): Likewise.
(back_and_forth_one): Likewise.
2003-04-27 Moritz Schulte <moritz@g10code.com>
* pubkey.c: Changed the sample private key to contain the
identifier `openpgp-rsa' instead of `rsa'.
* basic.c (check_digests): Enabled/fixed some tests for TIGER.
2003-04-17 Moritz Schulte <moritz@g10code.com>
* Makefile.am (TESTS): Removed `register' for now.
2003-04-17 Moritz Schulte <moritz@g10code.com>
* basic.c (check_digests): Include checks for SHA512 and SHA384.
2003-04-16 Moritz Schulte <moritz@g10code.com>
* basic.c (check_one_md): Also test md_copy.
2003-04-07 Moritz Schulte <moritz@g10code.com>
* Makefile.am (TESTS): Added register.
* register.c: New file.
2003-03-30 Simon Josefsson <jas@extundo.com>
* basic.c (check_one_cipher): New. Test CTR.
(main): Call it.
(check_ciphers): Check CTR mode.
2003-03-26 Moritz Schulte <moritz@g10code.com>
* Makefile.am (TESTS): Added pubkey.
* pubkey.c: New file.
2003-03-22 Simon Josefsson <jas@extundo.com>
* basic.c (check_cbc_mac_cipher): New.
(main): Use it.
2003-03-19 Werner Koch <wk@gnupg.org>
* keygen.c (check_rsa_keys): Don't expect an exponent when asking
for e=0.
(check_generated_rsa_key): Just print exponent if EXPECTED_E is 0.
2003-03-02 Moritz Schulte <moritz@g10code.com>
* basic.c (check_one_cipher): Use gcry_cipher_reset() instead of
gcry_cipher_close(), gcry_cipher_open and gcry_cipher_setkey().
2003-01-23 Werner Koch <wk@gnupg.org>
* keygen.c: New.
2003-01-20 Simon Josefsson <jas@extundo.com>
* basic.c (check_digests): Add CRC.
(check_one_md): Print computed and expected values on error.
2003-01-20 Werner Koch <wk@gnupg.org>
* basic.c (check_one_md): Kludge to check a one million "a".
(check_digests): Add checks for SHA-256.
2003-01-20 Werner Koch <wk@gnupg.org>
* basic.c (check_pubkey): Check the keygrip for the sample key.
2003-01-15 Werner Koch <wk@gnupg.org>
* basic.c (verify_one_signature,check_pubkey_sign)
(check_pubkey): New.
(main): Check public key functions. Add a --debug option.
2002-11-23 Werner Koch <wk@gnupg.org>
* basic.c (check_digests): Add another test for MD4. By Simon
Josefsson.
2002-11-10 Simon Josefsson <jas@extundo.com>
* basic.c (check_aes128_cbc_cts_cipher): New function.
(check_one_cipher): Add flags parameter.
(check_ciphers): Support flags parameter.
(main): Check CTS.
2002-11-10 Werner Koch <wk@gnupg.org>
* basic.c (check_one_md): New. By Simon Josefsson.
(check_digests): New tests for MD4. By Simon.
2002-08-26 Werner Koch <wk@gnupg.org>
* basic.c (check_ciphers): Check simple DES.
2002-05-16 Werner Koch <wk@gnupg.org>
* tsexp.c (back_and_forth): Very minimal test of the new functions.
2002-05-14 Werner Koch <wk@gnupg.org>
Changed license of all files to the LGPL.
2002-05-02 Werner Koch <wk@gnupg.org>
* basic.c: Add option --verbose.
2002-01-11 Werner Koch <wk@gnupg.org>
* tsexp.c (canon_len): Fixed tests.
2001-12-18 Werner Koch <wk@gnupg.org>
* tsexp.c: New.
Copyright 2001, 2002, 2003 Free Software Foundation, Inc.
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 file 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.
diff --git a/tests/Makefile.am b/tests/Makefile.am
index 36aede8c..f997c250 100644
--- a/tests/Makefile.am
+++ b/tests/Makefile.am
@@ -1,29 +1,30 @@
# Copyright (C) 2001, 2002, 2003 Free Software Foundation, Inc.
#
# This file is part of Libgcrypt.
#
# Libgcrypt 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.
#
# Libgcrypt 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, write to the Free Software
# Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA
## Process this file with automake to produce Makefile.in
-TESTS = prime register ac basic tsexp keygen pubkey benchmark random
+TESTS = prime register ac basic tsexp keygen pubkey benchmark random \
+ t-mpi-bit
INCLUDES = -I$(top_srcdir)/src
LDADD = ../src/libgcrypt.la
EXTRA_PROGRAMS = testapi
noinst_PROGRAMS = $(TESTS)
AM_CFLAGS = @GPG_ERROR_CFLAGS@
diff --git a/tests/t-mpi-bit.c b/tests/t-mpi-bit.c
new file mode 100644
index 00000000..2465a8fd
--- /dev/null
+++ b/tests/t-mpi-bit.c
@@ -0,0 +1,241 @@
+/* t-mpi-bit.c - Tests for bit level functions
+ * Copyright (C) 2006 Free Software Foundation, Inc.
+ *
+ * This file is part of Libgcrypt.
+ *
+ * Libgcrypt 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.
+ *
+ * Libgcrypt 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, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
+ * MA 02110-1301, USA.
+ */
+
+#ifdef HAVE_CONFIG_H
+#include <config.h>
+#endif
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <assert.h>
+#include <stdarg.h>
+
+#include "../src/gcrypt.h"
+
+#define PGM "t-mpi-bit"
+
+static const char *wherestr;
+static int verbose;
+static int error_count;
+
+#define xmalloc(a) gcry_xmalloc ((a))
+#define xcalloc(a,b) gcry_xcalloc ((a),(b))
+#define xfree(a) gcry_free ((a))
+#define pass() do { ; } while (0)
+
+static void
+show (const char *format, ...)
+{
+ va_list arg_ptr;
+
+ if (!verbose)
+ return;
+ fprintf (stderr, "%s: ", PGM);
+ va_start (arg_ptr, format);
+ vfprintf (stderr, format, arg_ptr);
+ va_end (arg_ptr);
+}
+
+static void
+fail (const char *format, ...)
+{
+ va_list arg_ptr;
+
+ fflush (stdout);
+ fprintf (stderr, "%s: ", PGM);
+ if (wherestr)
+ fprintf (stderr, "%s: ", wherestr);
+ va_start (arg_ptr, format);
+ vfprintf (stderr, format, arg_ptr);
+ va_end (arg_ptr);
+ error_count++;
+}
+
+static void
+die (const char *format, ...)
+{
+ va_list arg_ptr;
+
+ fflush (stdout);
+ fprintf (stderr, "%s: ", PGM);
+ if (wherestr)
+ fprintf (stderr, "%s: ", wherestr);
+ va_start (arg_ptr, format);
+ vfprintf (stderr, format, arg_ptr);
+ va_end (arg_ptr);
+ exit (1);
+}
+
+/* Allocate a bit string consisting of '0' and '1' from the MPI
+ A. Return the LENGTH least significant bits. Caller needs to xfree
+ the result. */
+static char *
+mpi2bitstr (gcry_mpi_t a, size_t length)
+{
+ char *p, *buf;
+
+ buf = p = xmalloc (length+1);
+ while (length--)
+ *p++ = gcry_mpi_test_bit (a, length) ? '1':'0';
+ *p = 0;
+
+ return buf;
+}
+
+/* Shift a bit string to the right. */
+static void
+rshiftbitstring (char *string, size_t n)
+{
+ size_t len = strlen (string);
+
+ if (n > len)
+ n = len;
+
+ memmove (string+n, string, len-n);
+ memset (string, '0', n);
+}
+
+
+/* This is to check a bug reported by bpgcrypt at itaparica.org on
+ 2006-07-31 against libgcrypt 1.2.2. */
+static void
+one_bit_only (int highbit)
+{
+ gcry_mpi_t a;
+ char *result;
+ int i;
+
+ wherestr = "one_bit_only";
+ show ("checking that set_%sbit does only set one bit\n", highbit?"high":"");
+
+ a = gcry_mpi_new (0);
+ gcry_mpi_randomize (a, 70, GCRY_WEAK_RANDOM);
+ gcry_mpi_set_ui (a, 0);
+
+ if (highbit)
+ gcry_mpi_set_highbit (a, 42);
+ else
+ gcry_mpi_set_bit (a, 42);
+ if (!gcry_mpi_test_bit (a, 42))
+ fail ("failed to set a bit\n");
+ gcry_mpi_clear_bit (a, 42);
+ if (gcry_mpi_test_bit (a, 42))
+ fail ("failed to clear a bit\n");
+ result = mpi2bitstr (a, 70);
+ assert (strlen (result) == 70);
+ show ("r=%s\n", result);
+ for (i=0; result[i]; i++)
+ if ( result[i] != '0' )
+ break;
+ if (result[i])
+ fail ("spurious bits detected\n");
+ xfree (result);
+ gcry_mpi_release (a);
+}
+
+/* Check that the shifting actually works for an amount larger than
+ the number of bits per limb. */
+static void
+test_rshift (int pass)
+{
+ gcry_mpi_t a, b;
+ char *result, *result2;
+ int i;
+
+ wherestr = "test_rshift";
+ show ("checking that rshift works as expected (pass %d)\n", pass);
+
+ a = gcry_mpi_new (0);
+ b = gcry_mpi_new (0);
+ gcry_mpi_randomize (a, 70, GCRY_WEAK_RANDOM);
+
+ for (i=0; i < 75; i++)
+ {
+ gcry_mpi_rshift (b, a, i);
+
+ result = mpi2bitstr (b, 72);
+ result2 = mpi2bitstr (a, 72);
+ rshiftbitstring (result2, i);
+ if (strcmp (result, result2))
+ {
+ show ("got =%s\n", result);
+ show ("want=%s\n", result2);
+ fail ("rshift by %d failed\n", i);
+ }
+ xfree (result);
+ xfree (result2);
+ }
+
+ /* Again. This time using in-place operation. */
+ gcry_mpi_randomize (a, 70, GCRY_WEAK_RANDOM);
+
+ for (i=0; i < 75; i++)
+ {
+ gcry_mpi_release (b);
+ b = gcry_mpi_copy (a);
+ gcry_mpi_rshift (b, b, i);
+
+ result = mpi2bitstr (b, 72);
+ result2 = mpi2bitstr (a, 72);
+ rshiftbitstring (result2, i);
+ if (strcmp (result, result2))
+ {
+ show ("got =%s\n", result);
+ show ("want=%s\n", result2);
+ fail ("in-place rshift by %d failed\n", i);
+ }
+ xfree (result2);
+ xfree (result);
+ }
+
+ gcry_mpi_release (b);
+ gcry_mpi_release (a);
+}
+
+
+int
+main (int argc, char **argv)
+{
+ int debug = 0;
+ int i;
+
+ if (argc > 1 && !strcmp (argv[1], "--verbose"))
+ verbose = 1;
+ else if (argc > 1 && !strcmp (argv[1], "--debug"))
+ verbose = debug = 1;
+
+ if (!gcry_check_version (GCRYPT_VERSION))
+ die ("version mismatch\n");
+
+ gcry_control (GCRYCTL_DISABLE_SECMEM, 0);
+ gcry_control (GCRYCTL_INITIALIZATION_FINISHED, 0);
+ if (debug)
+ gcry_control (GCRYCTL_SET_DEBUG_FLAGS, 1u, 0);
+ gcry_control (GCRYCTL_ENABLE_QUICK_RANDOM, 0);
+
+ one_bit_only (0);
+ one_bit_only (1);
+ for (i=0; i < 5; i++)
+ test_rshift (i); /* Run several times due to random initializations. */
+
+ show ("All tests completed. Errors: %d\n", error_count);
+ return error_count ? 1 : 0;
+}

File Metadata

Mime Type
text/x-diff
Expires
Sat, Feb 1, 9:17 AM (1 d, 7 h)
Storage Engine
local-disk
Storage Format
Raw Data
Storage Handle
30/c4/0d517c2cb9334cd9291800ff3183

Event Timeline