Page MenuHome GnuPG
Files F26446528

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 ISO-8859-1 (Latin 1) and converted to UTF8 for display.
diff --git a/ChangeLog-old b/ChangeLog-old
index cad8c1a..02a8010 100644
--- a/ChangeLog-old
+++ b/ChangeLog-old
@@ -1,2175 +1,2175 @@
2012-08-08 Werner Koch <wk@gnupg.org>
NB: ChangeLog files are no longer manually maintained. Starting
on August 8, 2012 we put change information only in the GIT commit
log, and generate a top-level ChangeLog file from logs at "make
dist". See doc/HACKING for details.
2011-08-10 Werner Koch <wk@g10code.com>
* configure.ac: Change bug report address to the BTS. Fix
autoconf warnings.
2011-06-28 Werner Koch <wk@g10code.com>
* gtk+-2/gtksecentry.c (PROP_EDITING_CANCELED): New.
(gtk_secure_entry_class_init, gtk_secure_entry_set_property)
(gtk_secure_entry_get_property): Implement as dummy it to get rid
of a runtime warning.
2010-12-16 Werner Koch <wk@g10code.com>
Release 0.8.1.
* doc/pinentry.texi: Remove the custom font setting for underscores.
Update to automake 1.10.
* missing, config.guess, config.sub: Update from GnuPG.
* w32/main.c (debugfp): Define to stderr.
(raise_sip) [!W32CE]: Make it a dummy function.
* qt4/pinentrydialog.cpp (raiseWindow): Use SetForegroundWindow
API directly.
(SetForegroundWindowEx): Comment out.
2010-10-19 Werner Koch <wk@g10code.com>
* pinentry/pinentry.c (pinentry_have_display): Also detect
--display=DISPID.
2010-10-12 Werner Koch <wk@g10code.com>
* qt/main.cpp (escape_accel): New. Adapt from from qt4/.
(qt_cmd_handler): Implement default-ok and default-cancel. Escape
accelerators.
2010-09-30 Werner Koch <wk@g10code.com>
* pinentry/pinentry.h (struct pinentry): Add CLOSE_BUTTON.
* pinentry/pinentry.c (cmd_getpin, cmd_confirm, cmd_message): Send
BUTTON_INFO.
* gtk+-2/pinentry-gtk-2.c (delete_event): Set close_button.
2010-09-22 Werner Koch <wk@g10code.com>
* gtk+-2/pinentry-gtk-2.c (create_window): Make use of
default_cancel and default_ok.
2010-08-19 Werner Koch <wk@g10code.com>
* gtk+-2/Makefile.am (pinentry_gtk_2_SOURCES): Add gseal-gtk-compat.h.
* configure.ac (AH_TOP, AH_BOTTOM): New. Define
GPG_ERR_ENABLE_ERRNO_MACROS.
2010-06-08 Marc Mutz <marc.mutz@kdab.com> (wk)
* qt4/pinentrydialog.cpp (PinEntryDialog): Add WindowStaysOnTopHint.
* qt/pinentrydialog.cpp (PinEntryDialog): Add WStyle_StaysOnTop.
2010-05-12 Guido Günther <agx@sigxcpu.org> (wk)
* gtk+-2/gtksecentry.c: Make -DSEAL safe.
* gtk+-2/gseal-gtk-compat.h: New.
* gtk+-2/gtksecentry.h: Include new header.
2010-05-07 Werner Koch <wk@g10code.com>
* gtk+-2/pinentry-gtk-2.c (create_window): Keep the window above
and make sure it has the focus. Fixes bug#1162.
2010-05-03 Werner Koch <wk@g10code.com>
* gtk+-2/pinentry-gtk-2.c (create_window): Use
visibility-notify-event instead of map-event. Suggested by Ed
Marten.
2010-04-26 Werner Koch <wk@g10code.com>
* gtk+-2/pinentry-gtk-2.c (grab_failed): New.
(grab_keyboard): s/g_error/g_critical/. Set GRAB_FAILED.
(gtk_cmd_handler): Return Cancel if a grab failed.
2010-04-20 Marcus Brinkmann <marcus@g10code.de>
* pinentry/pinentry.c: Move include config.h to top!
2010-04-19 Werner Koch <wk@g10code.com>
* gtk+-2/gtksecentry.c (gtk_secure_entry_delete_cb)
(gtk_secure_entry_toggle_overwrite, activate_cb)
(unichar_chosen_func): Comment out these unused functions.
(secentry_malloc): Fix format string error.
2010-04-19 Stanislav Ochotnicky <sochotnicky@redhat.com> (wk)
* gtk+-2/pinentry-gtk-2.c (ungrab_keyboard): Fix segv when using
option -g.
2010-04-16 Marcus Brinkmann <marcus@g10code.de>
* secmem/util.c: Fix inverted test.
2010-04-12 Werner Koch <wk@g10code.com>
* w32/main.c (dlg_proc): Act on VK_RETURN.
* w32/pinentry-w32.rc: Include config.h. Add basic support for
small screens as used with WindowsCE.
* w32/Makefile.am (pinentry-w32.o): Add -I.. to the command.
2010-04-08 Werner Koch <wk@g10code.com>
* pinentry/pinentry.c [W32CE]: Do not include errno.h and
locale.h.
(pinentry_have_display) [W32CE]: Do not use getenv.
(pinentry_parse_opts) [W32CE]: No error diagnostics due to missing
errno.
(sleep) [W32CE]: New.
(option_handler) [W32CE]: Ignore option --debug-wait.
(getpid) [W32CE]: New.
(pinentry_loop): Use STDIN_FILENO and STDOUT_FILENO.
(pinentry_loop): Factor code out to ...
(pinentry_loop2): New.
* w32/main.c (w32_strerror) [W32CE]: Do not use FormatMessage.
(main) [W32CE]: Do not use LockSetForegroundWindow.
(set_dlg_item_text): Use wchar_t for static string.
* secmem/util.c [W32CE]: Do not include errno.h.
(xwrite) [W32CE]: Shortcut EINTR loop.
2010-04-07 Werner Koch <wk@g10code.com>
* autogen.sh: Modernized.
2010-03-03 Werner Koch <wk@g10code.com>
Release 0.8.0.
2010-03-02 Werner Koch <wk@g10code.com>
* qt4/main.cpp (from_utf8): Add extra braces for clarity.
(main): Copy the fixed --display string.
2010-02-25 Marc Mutz <marc@kdab.com> (wk)
* qt4/main.cpp (from_utf8): don't throw InvalidUtf8(); for now,
fall back to QString::fromLocal8Bit()
2010-02-22 Marc Mutz <marc@kdab.com> (wk)
* qt4/main.cpp (InvalidUtf8): New exception class.
(from_utf8): wrapper around QString::fromUtf8 throwing
InvalidUtf8.
(qt_cmd_handler): Use from_utf8 instead of QString::fromUtf8.
(qt_cmd_handler_ex): New wrapper around qt_cmd_handler.
(pinentry_cmd_handler): Use qt_cmd_handler_ex.
* qt4/main.cpp (qt_cmd_handler): Also handle accels in
SET{OK,NOTOK,CANCEL} string.
2010-02-19 Marc Mutz <marc@kdab.com> (wk)
* qt4/main.cpp (qt_cmd_handler), qt4/pinentrydialog.cpp
(PinentryDialog::PinEntryDialog): Use the button icons that come
with Qt.
* qt4/pinentry_qrc, qt4/qrc_pinentry.cpp: New.
* qt4/gtk-cancel.png, qt4/gtk-ok.png: Removed.
* qt4/Makefile.am (EXTRA_DIST): Remove the GTK icons, add
pinentry.qrc.
(pinentry_qt4_SOURCES): add qrc_pinentry.cpp.
* qt4/main.cpp (qt_cmd_handler): Remove i18n support again.
* qt4/pinentrydialog.cpp (PinentryDialog::PinentryDialog):
set PIN entry field as buddy of its prompt
* qt4/main.cpp (escape_accel): New.
(qt_cmd_handler): use it on default_cancel, default_ok, prompt.
* qt4/pinentrydialog.h, qt4/pinentrydialog.cpp
(PinEntryDialog::PinEntryDialog): don't use a nested vbox layout.
* qt4/pinentrydialog.h: replace include file with forward decl.
* qt4/pinentrydialog.cpp: add include file here.
* qt4/pinentrydialog.h, qt4/pinentrydialog.cpp
(PinentryDialog::keyPressEvent()): Removed.
* qt/4pinentrydialog.cpp (PinEntryDialog::PinEntryDialog): use
QLayout::SetFixedSize instead of rolling our own solution.
* qt4/pinentrydialog.h: (PinentryDialog::setFixedSize): Removed.
* qt4/pinentrydialog.moc: Re-created.
* qt4/main.cpp: (qt_cmd_handler) make Cancel the default button
for CONFIRM
* qt4/main.cpp: (qt_cmd_handler) implement the confirm/message
logic correctly
* qt4/pinentrydialog.h (accepted, rejected): Removed; already
provided by base class.
* qt4/pinentrydialog.cpp (PinEntryDialog::PinEntryDialog): Use
QDialog::{accepted,rejected}.
* qt4/pinentrydialog.moc: Re-created.
* qt4/pinentrydialog.moc, qt4/qsecurelineedit.moc: Make compatible
with newer Qt version.
2010-02-19 Werner Koch <wk@g10code.com>
* gtk+-2/pinentry-gtk-2.c (create_window): Allow mnemonics in
button labels.
* pinentry/pinentry.h (struct pinentry): Add default_prompt.
* pinentry/pinentry.c (option_handler): Support "default-prompt".
(cmd_getpin): Use it.
(noprefix_strdup): Remove; should be done by caller.
2010-02-17 Werner Koch <wk@g10code.com>
* pinentry/pinentry.h (struct pinentry): Add default_ok and
default_cancel.
* pinentry/pinentry.c (noprefix_strdup): New.
(option_handler): Support "default-ok" and "default-cancel".
(strcpy_escaped): Change second arg from unsigend char to char.
2010-02-16 Werner Koch <wk@g10code.com>
* qt4/document-encrypt.png, qt4/gtk-cancel.png, qt4/gtk-ok.png: New.
* qt4/Makefile.am (EXTRA_DIST): Add them.
2010-02-15 Christoph Schleifenbaum <christoph@kdab.com> (wk)
Also attributed to Marc Mutz <marc@kdab.com>.
* qt4/main.cpp, qt/pinentrydialog.cpp: If the style says so,
set (GTK) icons on OK and Cancel buttons.
2010-02-15 Christoph Schleifenbaum <christoph@kdab.com> (wk)
* qt4/main.cpp: Use an explicit QMessageBox instead of the
'information' convenience method. We do this for two reasons:
1. So we can set our own icon (icon()). 2. So we can call our
raiseWindow() function on it.
2010-02-15 Christoph Schleifenbaum <christoph@kdab.com>
* qt4/main.cpp: Use title. Use Qt translation of "OK" and "Cancel"
* qt4/pinentrydialog.cpp: Don't show the question-mark icon in the
window title.
2010-02-15 Christoph Schleifenbaum <christoph@kdab.com> (wk)
* qt4/pinentrydialog.h, qt4/pinentrydialog.cpp: Use the
document-encrypt icon instead of the
QMessageBox::{Information,Critical} ones.
2010-02-15 Christoph Schleifenbaum <christoph@kdab.com> (wk)
Also attributed to Marc Mutz <marc@kdab.com>.
* qt4/pinentrydialog.cpp: Properly show/hide optional widgets.
2010-02-15 Christoph Schleifenbaum <christoph@kdab.com>
* qt4/pinentrydialog.cpp: Set text color or error label to red.
2010-02-15 Christoph Schleifenbaum <christoph@kdab.com> (wk)
Also attributed to Marc Mutz <marc@kdab.com>.
* qt4/pinentrydialog.cpp: Gather scattered layouting code, use a
single gridlayout for all, and use QDialogButtonBox.
2010-02-15 Christoph Schleifenbaum <christoph@kdab.com> (wk)
* qt4/pinentrydialog.cpp: Call SetForegroundWindow on showEvent()
and set a fixed size on the dialog.
* qt4/main.cpp: Set document-encrypt as the window icon.
2010-02-15 Marc Mutz <marc@kdab.com> (wk)
* qt4/main.cpp: Load the correct Qt translation file based on
either lc_messages or the current locale.
* qt4/main.cpp: Use pinentry_t->parent_wid also for message-box
case.
* qt4/main.cpp: Properly delete the QApplication.
2010-02-15 Christoph Schleifenbaum <christoph@kdab.com> (wk)
* qt4/main.cpp: Remove guard clause with impossible condition.
* qt4/main.cpp: Respect pinentry_t->title.
* qt4/main.cpp: Include config.h.
2009-12-16 Marcus Brinkmann <marcus@g10code.de>
* gtk+-2/gtksecentry.c: Don't translate properties.
Patch by Guido Guenther <agx@sigxcpu.org>.
2009-12-02 Werner Koch <wk@g10code.com>
* secmem/util.c: Re-indent function names.
(TEMP_FAILURE_RETRY): Remove because it is non-portable.
(xwrite): Open code the EINTR retry.
* pinentry/pinentry-curses.c (dialog_run): Replace '\e' by a
standard escape. Reported by Wyllys Ingersoll.
2009-08-10 Marcus Brinkmann <marcus@g10code.de>
* pinentry/pinentry.c [!ENABLE_ENHANCED]: Remove option enhanced.
* pinentry/pinentry.h (struct pinentry) [!ENABLE_ENHANCED]: Remove
member enhanced.
* gtk+-2/pinentry-gtk-2.c: New macro.
(button_clicked, create_window) [!ENABLE_ENHANCED]: Disable code.
* gtk/pinentry-gtk.c (button_clicked, create_window): Likewise.
* doc/pinentry.texi (Using pinentry): Comment out enhanced option.
2009-07-10 Till Adam <till@kdab.com>
* qt4/pinentrydialog.cpp (PinEntryDialog): Redesign the dialog to
be prettier.
2009-06-19 Marcus Brinkmann <marcus@g10code.de>
* configure.ac: Post-release changes.
Release version 0.7.6.
* configure.ac: Bump version.
2009-01-29 Bernhard Reiter <bernhard@intevation.de>
* doc/pinentry.texi (Using pinentry): Fix typo.
2009-04-24 Guido Guenther <agx@sigxcpu.org>
* gtk+-2/gtksecentry.c (gtk_secure_entry_class_init): Don't
install "gtk-entry-select-on-focus" settings property since it
clashes with the one from GtkEntry.
(gtk_secure_entry_grab_focus): Create a temporary GtkEntry widget
to read that value instead. Allows to have GtkEntry and
GtkSecureEntry in the same process.
2009-06-19 Marcus Brinkmann <marcus@g10code.de>
Suggested by Simon McVittie:
* gtk+-2/gtksecentry.c (g_malloc, g_realloc, g_free): Renamed to
secentry_malloc, secentry_realloc.
(g_malloc0): Removed.
(GMALLOC_SIZE): Move to ...
* gtk+-2/gtksecentry.h (GMALLOC_SIZE): ... here.
(secentry_malloc, secentry_realloc, secentry_free): New prototypes.
* gtk+-2/pinentry-gtk-2.c (main): Set memory allocation vtable.
2009-05-04 Till Adam <till@kdab.net>
* qt4/main.cpp (qt_cmd_handler): Use WId instead of HWND.
2009-04-14 Marcus Brinkmann <marcus@g10code.de>
* pinentry/pinentry.h (struct pinentry): New member NOTOK.
Rename member USER_CLOSED to CANCELED.
* pinentry/pinentry.c: Add initializer for NOTOK.
(register_commands): Add SETNOTOK.
* pinentry/pinentry-curses.c (STRING_NOTOK): New macro.
(dialog_pos_t): New value DIALOG_POS_NOTOK.
(struct dialog): New members notok, notok_x, notok_y.
(dialog_create): Implement NOTOK.
(dialog_switch_pos): Likewise.
(dialog_run): Likewise.
* gtk+-2/pinentry-gtk-2.c: Rename confirm_yes to confirm_value.
(confirm_value_t): New type. Use it for confirm_value.
(window_closed): Remove.
(confirm_button_clicked): Set confirm_value.
(create_window): Implement NOTOK.
(gtk_cmd_handler): Implement cancel behaviour a bit differently.
2009-04-09 Marcus Brinkmann <marcus@g10code.de>
* qt4/qsecurelineedit.h (QSecureLineEdit): Don't export.
2009-04-06 Marcus Brinkmann <marcus@g10code.de>
* configure.ac: Portability fixes.
2009-04-03 Marcus Brinkmann <marcus@g10code.de>
* qt4/Makefile.am (pinentry_qt4_SOURCES): Add qsecurelineedit_p.h.
* secmem/Makefile.am (libsecmem_a_SOURCES): Add secmem++.h.
* qt4/Makefile.am (pinentry_qt4_SOURCES): Add pinentrydialog.moc
and qsecurelineedit.
* configure.ac: Incorporate revision number in version string.
2009-04-02 Till Adam <till@kdab.net>
* qt4/qsecurelineedit.moc, qt4/main.cpp, qt4/pinentrydialog.moc,
qt4/qsecurelineedit.cpp, qt4/pinentrydialog.cpp,
qt4/pinentrydialog.h: Implement quality bar and grabbing.
* qt4/pinentrydialog.cpp (PinEntryDialog): Enable word wrapping.
2009-04-03 Marcus Brinkmann <marcus@g10code.de>
* qt4/main.cpp (qt_cmd_handler): Cast window handle ID.
2009-03-19 Werner Koch <wk@g10code.com>
* pinentry/pinentry.h (struct pinentry): Add field user_closed.
* pinentry/pinentry.c (cmd_getpin, cmd_confirm, cmd_message):
Reset this field.
(cmd_confirm): Return cancel if the user closed the window.
* gtk+-2/pinentry-gtk-2.c (window_closed): New.
(gtk_cmd_handler, delete_event): Set it.
2009-02-19 Marcus Brinkmann <marcus@g10code.de>
* secmem/secmem++.h: New file.
* qt4/Makefile.am, qt4/main.cpp, qt4/pinentrydialog.cpp,
qt4/pinentrydialog.h: Changes to use secure memory.
* qt4/pinentrydialog.moc, qt4/qsecurelineedit.moc: New files.
* qt4/qsecurelineedit.cpp, qt4/qsecurelineedit.h,
qt4/qsecurelineedit_p.h, qt4/secstring.h, qt4/secstring.cpp: New
files.
Submitted by Marc Mutz <marc@kdab.net>.
2008-12-08 Werner Koch <wk@g10code.com>
* qt/main.cpp (qt_cmd_handler) <!want_pass>: Translate from utf8.
* qt4/main.cpp (qt_cmd_handler) <!want_pass>: Translate from utf8.
2008-11-12 Ludovic Rousseau <ludovic.rousseau@gmail.com> (wk)
* pinentry/pinentry.h (struct pinentry): Add field TITLE.
* pinentry/pinentry.c (cmd_settitle): New.
(register_commands): Add command SETTITLE.
* gtk+-2/pinentry-gtk-2.c (create_window): Set the title.
2008-10-17 Werner Koch <wk@g10code.com>
* secmem/memory.h: Add C++ extern declaration.
* secmem/secmem.c (secmem_get_max_size): New.
2008-07-25 Marcus Brinkmann <marcus@ulysses.g10code.com>
* qt4/Makefile.am (DISTCLEANFILES): Rename to CLEANFILES.
(BUILD_SOURCES): New.
2008-07-25 Marcus Brinkmann <marcus@g10code.de>
* configure.ac: Prefer pinentry gtk2 over gtk. Now also choose
qt4 (after qt and before w32 native).
* README: Document Qt.
2008-07-24 Marcus Brinkmann <marcus@g10code.de>
* qt4/pinentrydialog.h, qt4/pinentrydialog.cpp, qt4/main.cpp,
qt4/Makefile.am: New files.
* configure.ac: Add check for Qt4.
(AC_CONFIG_FILES): Add qt4/Makefile.am.
* Makefile.am (pinentry_qt4): New variable.
(SUBDIRS): Add ${pinentry_qt4}.
2008-07-23 Werner Koch <wk@g10code.com>
* secmem/secmem.c (secmem_init): Fix apostrophe in error
message. Reported by Petr Uzel.
2008-05-05 Marcus Brinkmann <marcus@ulysses.g10code.com>
* gtk+-2/gtksecentry.c (GMALLOC_SIZE): New symbol.
(g_malloc, g_malloc0, g_realloc): Use GMALLOC_SIZE instead of
gulong for size attribute.
2008-03-03 Marcus Brinkmann <marcus@g10code.de>
* gtk+-2/pinentry-gtk-2.c (create_window): Install realize callback.
(ungrab_keyboard): Unmake dialog transient.
(make_transient): New function.
Submitted by Joachim Breitner.
2008-02-15 Werner Koch <wk@g10code.com>
Release 0.7.4.
2008-02-14 Werner Koch <wk@g10code.com>
* configure.ac: Check for -Wno-pointer-sign.
* pinentry/pinentry.c (cmd_getinfo): New.
(register_commands): Register it.
2008-01-02 Marcus Brinkmann <marcus@g10code.de>
* configure.ac: Use PKG_CONFIG instead of PKGCONFIG and
pkg-config. Use AC_PATH_PROG instead of AC_CHECK_PROG.
* Makefile.am (install-exec-local): Add exe extension to link.
2007-11-29 Marcus Brinkmann <marcus@g10code.de>
Released 0.7.4.
2007-11-19 Werner Koch <wk@g10code.com>
* doc/pinentry.texi (Protocol): Typo fixes by Bernhard Herzog.
Describe SETQUALITYBAR_TT.
2007-11-19 Bernhard Herzog <bernhard.herzog@intevation.de> (wk)
* qt/pinentrydialog.cpp (PinEntryDialog): Fixed crash
2007-11-19 Werner Koch <wk@g10code.com>
* gtk+-2/pinentry-gtk-2.c (create_window): Use again map-event and
unmap-event as this works on my setup far more reliable than
expose-event/no-expose-event.
* gtk+-2/gtksecentry.c (get_cursor_time): s/time/blinktime/ t
avoid shadowing warning.
* pinentry/pinentry.h (struct pinentry): Change QUALITY_BAR to a
char ptr.
(struct pinentry): Add QUALITY_BAR_TT.
* pinentry/pinentry.c (cmd_setqualitybar): Allow to set a label text.
(cmd_setqualitybar_tt): New.
* gtk+-2/pinentry-gtk-2.c (create_window): Take label and tooltip
from global.
* qt/pinentrydialog.h (setQualityBar, setQualityBarTT)
(_quality_bar_label): New.
* qt/pinentrydialog.cpp (setQualityBar, setQualityBarTT): New.
(PinEntryDialog): Remove setting of tooltip.
* qt/main.cpp (qt_cmd_handler): Propagate quality bar label and tootip.
2007-11-19 Bernhard Herzog <bernhard.herzog@intevation.de> (wk)
* qt/pinentrydialog.cpp (PinEntryDialog): Move the quality bar
below the text entry and align them properly. Show a tooltip.
* gtk+-2/pinentry-gtk-2.c (create_window): Ditto. Don't let it
change its height.
(QUALITYBAR_EMPTY_TEXT): New.
2007-09-18 Werner Koch <wk@g10code.com>
* qt/secqlineedit.h (SecQLineEdit): New signal textModified.
* qt/secqlineedit.cpp (finishChange): Emit it.
* qt/pinentrydialog.cpp (setPinentryInfo): New.
(PinEntryDialog): Add arg ENABLE_QUALITY_BAR.
* qt/pinentrydialog.h (setPinentryInfo): New.
(PinEntryDialog): Add arg ENABLE_QUALITY_BAR.
* pinentry/pinentry.h (struct pinentry): Add member QUALITY_BAR
and CTX_ASSUAN.
* pinentry/pinentry.c (cmd_setqualitybar): New.
(copy_and_escape): New.
(pinentry_inq_quality): New.
2007-07-09 Werner Koch <wk@g10code.com>
* doc/pinentry.texi: Fixed direntry syntax.
* configure.ac: Add --without-libcap. From the Gentoo patch archive.
* gtk+-2/pinentry-gtk-2.c (create_window): Use expose-event
instead of map-event. From Alon Bar-Lev.
2007-07-06 Werner Koch <wk@g10code.com>
Released 0.7.3.
* config.sub, config.guess: Updated from current Savannah CVS.
2007-06-27 Werner Koch <wk@g10code.com>
* w32/main.c: Revamped the SetFocus stuff. It is all not that easy.
2007-06-26 Werner Koch <wk@g10code.com>
* w32/Makefile.am (pinentry_w32_LDFLAGS): Add -mconsole again.
* w32/main.c (dlg_proc): Set focus.
(resize_button): New. No code yet.
(dlg_proc): Call it for the buttons.
(w32_cmd_handler): Restore old foreground window.
2007-06-20 Werner Koch <wk@g10code.com>
* w32/Makefile.am (pinentry_w32_LDFLAGS): Remove -mconsole.
* w32/main.c (wchar_to_utf8): New.
(ok_button_clicked): Use it.
(utf8_to_wchar): New.
(set_dlg_item_text): New.
(dlg_proc): Use new function so that we are able to correctly
display all prompts.
(main): Load LockSetForegroundWindow.
(dlg_proc): Call LockSetForegroundWindow via its fnc ptr.
(center_window): New. Taken from GPGol.
(dlg_proc): Call it.
(w32_cmd_handler): Revamped the confirm mode.
2007-06-18 Werner Koch <wk@g10code.com>
* w32/main.c (dlg_proc): Call LockSetForegroundWindow.
* Makefile.am (signed-dist, %.sig): Remove.
* autogen.sh: Modernized.
2007-05-10 Marcus Brinkmann <marcus@g10code.de>
* pinentry/pinentry.h (pinentry_color_t): New type.
(struct pinentry): New members COLOR_FG, COLOR_FG_BRIGHT,
COLOR_BG, COLOR_SO.
* pinentry/pinentry.c (pinentry_parse_opts): Support new option --colors.
(parse_color): New function.
* pinentry/pinentry-curses.c (USE_COLORS): New macro.
(pinentry_color): New static variable.
(dialog_create): Redo color management.
* pinentry/pinentry-curses.c (dialog_create): Re-add calculation
of cancel button position. Adjust calculation of OK button
position if it is the only one used.
2007-04-13 Marcus Brinkmann <marcus@g10code.de>
* qt/secqlineedit.h (SecQLineEdit::contextMenuEvent,
SecQLineEdit::createPopupMenu): Remove prototype.
* qt/secqlineedit.cpp (SecQLineEdit::contextMenuEvent,
SecQLineEdit::createPopupMenu): Remove implementation.
Submitted by Tobias Koenig <tokoe@kde.org>.
2007-02-14 Werner Koch <wk@g10code.com>
* pinentry/pinentry.h (struct pinentry): Add TOUCH_FILE.
* pinentry/pinentry.c (option_handler): New option "touch-file".
(pinentry_have_display): Ignore an empty DISPLAY.
* pinentry/pinentry-curses.c (do_touch_file): New.
(curses_cmd_handler): Call it.
* configure.ac: Check for utime.h.
2007-01-24 Werner Koch <wk@g10code.com>
* pinentry/pinentry.c (cmd_message): New.
(cmd_confirm): New command option --one-button.
(cmd_getpin): Zeroise ONE_BUTTON.
* pinentry/pinentry.h (struct pinentry): Add field ONE_BUTTON.
* gtk/pinentry-gtk.c (create_window): Take care of new option.
* gtk+-2/pinentry-gtk-2.c (create_window): Ditto.
* pinentry/pinentry-curses.c (dialog_create): Ditto.
(dialog_create, dialog_switch_pos): Allow CANCEL to be optional.
2006-07-29 Marcus Brinkmann <marcus@g10code.de>
* secmem/secmem.c (init_pool): Close FD after establishing the
mapping.
2005-09-28 Marcus Brinkmann <marcus@g10code.de>
* configure.ac (PINENTRY_GTK): Error out if iconv is not found but
needed.
* config.rpath: New file from gettext. Needed by iconv.m4.
2005-07-11 Marcus Brinkmann <marcus@g10code.de>
* pinentry/pinentry-curses.c (dialog_switch_pos): Set the cursor
state to invisible before moving around. Move the cursor to the
beginning of the dialog buttons for accessibility.
2005-06-16 Marcus Brinkmann <marcus@g10code.de>
* pinentry/pinentry-curses.c (dialog_run): Only convert pin if a
pin actually exists.
2005-01-27 Werner Koch <wk@g10code.com>
Released 0.7.2.
* gtk+-2/Makefile.am: Removed padlock-keyhole.xpm.
* configure.ac: Use AC_GNU_SOURCE instead of the custom define.
Prefer gtk2 pinentry over qt.
2005-01-21 Marcus Brinkmann <marcus@g10code.de>
* doc/pinentry.texi: Fix spelling errors.
Submitted by Ville Skyttä.
2004-12-23 Werner Koch <wk@g10code.com>
* w32/pinentry-w32.rc: Remove the default texts for description,
prompt and error. Make it system modal. Enlarge the description
field.
2004-12-22 Timo Schulz <twoaday@g10code.com>
* w32/main.c: Remove all helper functions and use a callback
to set the dialog items directly.
(dlg_proc): Set 'result' to -1 to indicate cancel.
(ok_button_clicked): Adjusted. Set 'result' to the len of
the PIN to indicate success.
2004-12-22 Werner Koch <wk@g10code.com>
* w32/main.c: Simplified.
* w32/dialog.h, w32/dialog.c, w32/controller.h, w32/controller.c:
Removed
* w32/resource.h, w32/pinentry-w32.rc, w32/main.c, w32/dialog.h
* w32/dialog.c, w32/controller.h, w32/controller.c
* w32/Makefile.am: New. Based on Timo's work.
Update to automake 1.9.
* autogen.sh (configure_ac): Add --build-w32 option.
* Makefile.am: Support for the W32 pinentry.
* configure.ac: Ditto. Reformatted some error messages.
Define the usual conditionals for W32.
Check for a couple of more usually required headers.
* pinentry/pinentry.h (sleep) [W32]: New.
* pinentry/pinentry.c: Include langinfo.h only if available.
(pinentry_loop) [DOSISH]: Don't do uid check.
* secmem/util.c [DOSISH]: Disable UID stuff.
2004-09-27 Marcus Brinkmann <marcus@g10code.de>
* acinclude.m4 (IU_LIB_NCURSES, IU_LIB_CURSES, IU_LIB_TERMCAP):
Moved to m4/curses.m4.
(AM_ICONV): Moved to m4/iconv.m4.
(AM_PATH_GLIB): Moved to m4/glib.m4.
(QT_FIND_FILE, QT_PATH_MOC, QT_PATH_X, QT_PRINT_PROGRAM,
QT_CHECK_VERSION, QT_PATH_1_3, QT_PATH, QT_CHECK_COMPILER_FLAG,
QT_REMOVE_FORBIDDEN, QT_VALIDIFY_CXXFLAGS, QT_CHECK_COMPILERS,
QT_CHECK_RPATH, QT_CHECK_LIBPTHREAD, QT_CHECK_PTHREAD_OPTION,
QT_CHECK_THREADING): Move to m4/qt.m4.
2004-09-21 Marcus Brinkmann <marcus@g10code.de>
* qt/Makefile.am (pinentry_qt_LDFLAGS): Add $(QT_RPATH).
Requested by Leo Savernik <l.savernik@aon.at>.
2004-09-02 Marcus Brinkmann <marcus@g10code.de>
* gtk+-2/padlock-keyhole.xpm: File removed.
* gtk+-2/pinentry-gtk-2.c (create_window): Use stock icon.
* gtk+-2/gtksecentry.h, gtk+-2/gtksecentry.c: Fix copyright notice.
Submitted by Albrecht Dress albrecht.dress@arcor.de.
2004-08-17 Marcus Brinkmann <marcus@g10code.de>
* configure.ac: Invoke AC_PROG_LN_S.
(PINENTRY_DEFAULT): New variable. Substitute it. Fail if no
default can be determined.
* Makefile.am (install-exec-local): Install pinentry default link.
* configure.ac: Check for Gtk+-2.
* gtk+-2: New directory with gtk+-2 pinentry.
* gtk+-2/Makefile.am, gtk+-2/gtksecentry.h, gtk+-2/gtksecentry.c,
gtk+-2/pinentry-gtk-2.c, gtk+-2/padlock-keyhole.xpm: New files.
* Makefile.am (pinentry_gtk_2): New variable.
(SUBDIRS): Add pinentry_gtk_2.
Submitted by Albrecht Dress albrecht.dress@arcor.de.
2004-08-04 Werner Koch <wk@g10code.de>
* pinentry/pinentry.c (usage): Print help to stdout.
2004-07-30 Moritz Schulte <moritz@g10code.com>
* qt/Makefile.am (ncurses_include): Removed
-I$(top_srcdir)/pinentry ...
(AM_CPPFLAGS): ... added: -I$(top_srcdir)/pinentry. Thanks to
Peter Eisentraut.
* pinentry/pinentry.c (pinentry_utf8_to_local): Declare INPUT
const.
(pinentry_local_to_utf8): Likewise.
(pinentry_utf8_to_local, pinentry_local_to_utf8): Compile only, if
either Curses or GTK+ support is enabled.
* configure.ac: Do also check for libiconv when the GTK+ version
of pinentry is to be build. Define PINENTRY_CURSES, PINENTRY_GTK,
PINENTRY_QT depending on which versions of pinentry should be
build.
2004-05-21 Marcus Brinkmann <marcus@g10code.de>
* acinclude.m4 (QT_CHECK_DIRECT): Removed.
(QT_PATH_1_3): Do never invoke QT_CHECK_DIRECT.
2004-04-21 Werner Koch <wk@gnupg.org>
Released 0.7.1.
2004-04-20 Werner Koch <wk@gnupg.org>
* secmem/secmem.c [!ORIGINAL_GPG_VERSION]: Include util.h for some
typedefs and protos.
(secmem_free, secmem_term): Use wipememory2 instead of memset.
* autogen.sh (configure_ac): Fixed version check (s/==/=/).
2004-04-02 Thomas Schwinge <schwinge@nic-nac-project.de>
* autogen.sh: Added ACLOCAL_FLAGS.
2004-02-23 Marcus Brinkmann <marcus@g10code.de>
* qt/main.cpp: Include "pinentry.h", not <pinentry.h>.
2004-01-30 Werner Koch <wk@gnupg.org>
* configure.ac (fopencookie): Remove that test.
2004-01-28 Moritz Schulte <mo@g10code.com>
* gtk/gtksecentry.c:
(gtk_secure_entry_key_press): Treat GDK_KP_Enter just like
GDK_Return.
2004-01-18 Marcus Brinkmann <marcus@g10code.de>
* qt/secqstring.cpp: Do not include "private/qunicodetables_p.h".
(isRightToLeft): De-optimize direction query.
* qt/secqinternal_p.h, qt/secqinternal.cpp: New files.
* qt/Makefile.am (pinentry_qt_SOURCES): Add secqinternal_p.h
and secqinternal.cpp.
(EXTRA_DIST): Add README.SecQ.
* qt/secqlineedit.cpp: Include "secqinternal_p.h".
(drawContents): Use SecQSharedDoubleBuffer.
2004-01-02 Werner Koch <wk@gnupg.org>
* configure.ac: Early check for a c++ compiler.
2003-12-23 Werner Koch <wk@gnupg.org>
Released 0.7.0.
Added Manifest files to all directories.
2003-12-22 Werner Koch <wk@gnupg.org>
* qt/main.cpp: Include errno.h.
(main): Translate the --display option to -display, so that the Qt
init code can grasp it.
* doc/ChangeLog: Removed and merged with this file.
* doc/pinentry.texi: Cleaned up.
* doc/fdl.texi: Removed.
* pinentry/pinentry.h (struct pinentry): Added LOCALE_ERR.
* gtk/pinentry-gtk.c (button_clicked): Set the LOCAE_ERR flag.
* pinentry/pinentry-curses.c (dialog_run): Ditto.
* pinentry/pinentry.c (cmd_getpin, cmd_confirm): Check this flag.
(pinentry_local_to_utf8): Release the correct buffer in the error
case. Print diagnostics.
(pinentry_utf8_to_local): Print diagnostics.
(pinentry_parse_opts): Make short options work.
(pinentry_utf8_to_local): Pass nl_langinfo to iconv_open.
* gtk/pinentry-gtk.c (button_clicked): Use the right value as
input for the conversion.
* pinentry/pinentry.c: New variable THIS_PGMNAME.
(pinentry_init): Add arg PGMNAME and store it. Use it at all
marked placed instead of the constant "pinentry".
(usage): Use it here too.
* curses/pinentry-curses.c (main): Call pinentry_init with our name.
* qt/main.cpp (main): Ditto.
* gtk/pinentry-gtk.c (main): Ditto.
* configure.ac: Check for mmap.
* secmem/util.h (wipememory2,wipememory,wipe): New.
* secmem/util.c (wipe): Removed.
* secmem/util.c (lower_privs, raise_privs): Commented out.
* pinentry/pinentry.c (pinentry_loop): Add paranoia check for
dropped privs.
* secmem/secmem.c (lock_pool): Cleanup syntax of cpp directives.
* gtk/pinentry-gtk.c (main): Print package name in the version line.
* curses/pinentry-curses.c (main): Ditto.
* qt/main.cpp (main): Ditto. Fixed typo.
* gtk/pinentry-gtk.c: Include memory.h.
2003-12-20 Marcus Brinkmann <marcus@g10code.de>
* pinentry/pinentry.h (struct pinentry): New member PARENT_WID.
* pinentry/pinentry.c (pinentry): Add new member here.
(usage): Add --parent-wid.
(pinentry_parse_opts): Add case for "parent-wid".
(option_handler): Same here.
2003-12-19 Marcus Brinkmann <marcus@g10code.de>
* pinentry/pinentry.c (cmd_setcancel): Use strcpy_escaped.
(cmd_setok): Likewise.
(cmd_setprompt): Likewise.
(pinentry_utf8_to_local): Don't use
nl_langinfo, but just lc_ctype directly.
* pinentry/pinentry.c (cmd_getpin): Do not convert passphrase to
UTF-8 here.
* gtk/pinentry-gtk.c (button_clicked): Convert passphrase to UTF8
here.
* pinentry/pinentry-curses.c (dialog_run): Likewise.
2003-12-14 Marcus Brinkmann <marcus@g10code.de>
* pinentry/pinentry.c (pinentry_init): Register secmem_term as
atexit function. Set assuan malloc hooks to secmem.
(pinentry_parse_opts): Add break statement to silence gcc warning.
* pinentry/pinentry.c (cmd_getpin): If canceled, release and clear
PINENTRY->pin nevertheless.
* acinclude.m4 (qt_incdirs): Add /usr/include/qt3.
* qt/Makefile.am (pinentry_qt_SOURCES): Remove cppmemory.h,
cppmemory.cpp, pinentrycontroller.h, pinentrycontroller.cpp.
(nodist_pinentry_qt_SOURCES): Remove pinentrycontroller.moc.cpp.
(libcurses): Move ../pinentry/libpinentry.a from here to ...
(pinentry_qt_LDADD): ... here. Change order a bit to make it
work.
* qt/cppmemory.h, qt/cppmemory.cpp, qt/pinentrycontroller.h,
qt/pinentrycontroller.cpp: Files removed.
* qt/secqstring.h, qt/secqstring.cpp, secqlineedit.h,
secqlineedit.cpp: New files.
* qt/Makefile.am (pinentry_qt_SOURCES): Add secqstring.h,
secqstring.cpp, secqlineedit.h, and secqlineedit.cpp.
(nodist_pinentry_qt_SOURCES): Add secqlineedit.moc.cpp.
* qt/main.cpp: Do not include "memory.h" or "secmem-util.h", nor
<new> or "pinentrycontroller.h". Include <qapplication.h>,
<qmessagebox.h>, <qwidget.h> and "secqstring.h". Always include
<pinentry.h>.
[USE_KDE]: Remove all instances.
(curses_main): Function removed.
(my_new_handler): Likewise.
(qt_main): Likewise.
(qt_cmd_handler): New function.
(pinentry_cmd_handler): Define always (to qt_cmd_handler).
(main): Rewritten.
* qt/pinentrydialog.cpp: Do not include <qlineedit.h>, but
"secqlineedit.h".
(PinEntryDialog::PinEntryDialog): Make _edit a SecQLineEdit
object. Connect accepted SIGNAL to accept SLOT, and rejected
SIGNAL to reject SLOT.
(PinEntryDialog::setText): Make argument SecQString rather than
QString.
(PinEntryDialog::text): Likewise for return value.
* qt/pinentrydialog.h: Declare SecQString and SecQLineEdit classes.
(class PinEntryDialog): Disable property text (for now). Adjust
argument of setText and return value of text, as well as type of
_edit.
2003-12-09 Werner Koch <wk@gnupg.org>
* README.CVS: New.
* Makefile.am (EXTRA_DIST): Add README.CVS
(ACLOCAL_AMFLAGS): New.
* configure.ac: Added min_automake_versions.
* autogen.sh: Revamped.
2003-04-23 Steffen Hansen <steffen@klaralvdalens-datakonsult.se>
* configure.ac: Version 0.6.10-cvs
2003-04-23 Steffen Hansen <steffen@klaralvdalens-datakonsult.se>
* configure.ac: Version 0.6.9
* qt/Makefile.am: Added moc files to DISTCLEANFILES
- * qt/pinentrycontroller.cpp: Dont spew assuan debug stuff out on stderr.
+ * qt/pinentrycontroller.cpp: Don't spew assuan debug stuff out on stderr.
2003-03-26 Steffen Hansen <steffen@klaralvdalens-datakonsult.se>
* qt/cppmemory.cpp, qt/main.cpp: Only override array allocation
operators. This should take care of the reported memory-problems
and even make pinentry-qt use a bit less memory.
2003-02-15 Steffen Hansen <steffen@klaralvdalens-datakonsult.se>
* qt/pinentrydialog.h, qt/pinentrydialog.cpp: Added icons
for error/non-error messages.
2003-02-07 Marcus Brinkmann <marcus@g10code.de>
Released 0.6.8.
2003-02-04 Steffen Hansen <steffen@klaralvdalens-datakonsult.se>
* qt/main.cpp: Work around '--display' option. This fixes
the pinentry-qt problem reported by several people lately.
2003-01-24 Werner Koch <wk@gnupg.org>
* autogen.sh: Print a hint to use maintainer mode.
2002-12-24 Marcus Brinkmann <marcus@g10code.de>
* pinentry/pinentry-curses.c (collect_line): New function.
(COPY_OUT, MAKE_BUTTON): New macros.
(dialog_create): Rewrite the initializing code and the description
calculation routine with word wrapping.
2002-11-20 Werner Koch <wk@gnupg.org>
Released 0.6.7.
* pinentry/pinentry-curses.c (dialog_create): Better truncate
lines than to go into an infinite loop. We need to implement word
wrap.
(dialog_run): Add DIALOG_POS_NONE to switch so prevent a warning.
2002-11-12 Werner Koch <wk@gnupg.org>
* config.sub, config.guess: Updated from ftp.gnu.org/gnu/config
to version 2002-11-08.
2002-11-09 Werner Koch <wk@gnupg.org>
Released 0.6.6.
2002-11-08 Werner Koch <wk@gnupg.org>
* pinentry/pinentry-curses.c (convert_utf8_string): Renamed to
* pinentry/pinentry.c (pinentry_utf8_to_local): this. Changed
callers.
(pinentry_local_to_utf8): New.
(cmd_getpin): Convert result back to UTF-8.
* gtk/pinentry-gtk.c (create_utf8_label): New.
(create_window): Use it here to set the prompts.
2002-11-06 Werner Koch <wk@gnupg.org>
* pinentry/pinentry-curses.c (dialog_run): Fixed retrun value tests
for fopen.
2002-11-05 Werner Koch <wk@gnupg.org>
* secmem/util.c (init_uids): Make it a prototype.
* gtk/pinentry-gtk.c (enter_callback): Changed argument name to
avoid shadowing warning.
(create_window): Removed unused variable I.
(ok): Not used, commented.
* pinentry/pinentry.c: Include headers for getpid and sleep
prototypes.
* secmem/util.h: Correctly declare functions taking no args.
* gtk/pinentry-gtk.c: Move gtk headers to the top to avoid
compiler warnings about shadowing index etc.
* curses/pinentry-curses.c: Include stdio.h for the printf prototype.
* pinentry/pinentry-curses.c (dialog_switch_pos): Return a value.
* pinentry/pinentry.c (pinentry_have_display): New.
(pinentry_setbufferlen): Must return a value. Fixed documentation.
(usage): Print a question mark as a substitue for the program name.
* gtk/pinentry-gtk.c (main): use it here instead of getenv().
* qt/main.cpp (main): Ditto.
2002-10-11 Werner Koch <wk@gnupg.org>
* configure.ac, Makefile.am: Added doc/ and tests for makeinfo
* doc/pinentry.texi, doc/Makefile.am: New.
* doc/gpl.texi, doc/fdl.texi: Added these standard files.
2002-09-30 Werner Koch <wk@gnupg.org>
Released 0.6.5.
* qt/pinentrycontroller.cpp (optionHandler): Make sure that a
value is returned.
* configure.ac: Use -Wall also for C++.
2002-08-19 Steffen Hansen <steffen@hrhansen.dk>
* Relased 0.6.4.
2002-08-11 Steffen Hansen <steffen@hrhansen.dk>
* Adapted pinentry-qt to new CONFIRM spec.
2002-06-26 Werner Koch <wk@gnupg.org>
Release 0.6.3.
2002-05-24 Werner Koch <wk@gnupg.org>
* AUTHORS: Added Marcus
* README: Fixed spelling of Quintuple-Agent.
2002-05-13 Marcus Brinkmann <marcus@g10code.de>
Released 0.6.2.
* configure.ac: Set version number to 0.6.2.
* NEWS: Add information for 0.6.2.
* README: Update for release.
2002-05-09 Marcus Brinkmann <marcus@g10code.de>
* configure.ac: Add option --enable-fallback-curses and bind it to
the ncurses check. Add automake conditional
BUILD_LIBPINENTRY_CURSES and FALLBACK_CURSES. Add preprocessor
symbol FALLBACK_CURSES.
* curses/Makefile.am (AM_CPPFLAGS): Add $(NCURSES_INCLUDE).
(LDADD): Add ../pinentry/libpinentry-curses.a.
* curses/pinentry-curses.c: Include "pinentry-curses.h".
Moved most of the meat to ...
* pinentry/pinentry-curses.c: ... here. New file. Make all
functions and global variables static.
(dialog_cmd_handler): Rename to ...
(curses_cmd_handler): ... this.
* pinentry/pinentry-curses.h: New file.
* pinentry/Makefile.am (noinst_LIBRARIES)
[BUILD_LIBPINENTRY_CURSES]: Add libpinentry-curses.a.
(libpinentry_curses_a_SOURCES): New target.
* gtk/Makefile.am (INCLUDES): Moved all to ...
(AM_CPPFLAGS): ... here.
[CURSES_FALLBACK]: Define ncurses_include and libcurses.
(AM_CPPFLAGS): Add $(ncurses_include).
(LDADD): Add $(libcurses).
* gtk/pinentry-gtk.c: Rename TIMEOUT to TIME_OUT to avoid conflict
with curses.h.
[CURSES_FALLBACK]: Include "pinentry-curses.h".
(button_clicked): Likewise.
(create_window): Likewise.
(cmd_handler): Renamed to ...
(gtk_cmd_handler): ... this.
(pinentry_cmd_handler): Set to gtk_cmd_handler.
(main) [CURSES_FALLBACK]: Initialize GTK+ only if environment
variable DISPLAY is set, otherwise fall back to curses dialog.
* pinentry/pinentry.h: Protect against multiple inclusion.
2002-05-09 Marcus Brinkmann <marcus@g10code.de>
* curses/pinentry-curses.c (dialog_create): Allow multi-line error
texts.
* pinentry/pinentry.c (cmd_seterror): Call strcpy_escaped, rather
than strcpy, to allow percent-escaping the error text.
2002-04-25 Steffen Hansen <steffen@hrhansen.dk>
* pinentry-qt: Use ok and cancel value if
provided.
2002-04-25 Marcus Brinkmann <marcus@g10code.de>
* gtk/pinentry-gtk.c (create_window): Use ok and cancel value if
provided.
2002-04-25 Marcus Brinkmann <marcus@g10code.de>
* qt/pinentrycontroller.h: New members _ok and _cancel.
* qt/pinentrycontroller.cpp (registerCommands): Add SETOK and SETCANCEL.
(assuanOk): New method.
(assuanCancel): Likewise.
2002-04-25 Marcus Brinkmann <marcus@g10code.de>
* curses/pinentry-curses.c (dialog_create): Grok the new ok and
cancel members to set the pbutton texts, rather than parsing the
prompt in the confirm case.
* pinentry/pinentry.h (struct pinentry): Add new members ok and cancel.
* pinentry/pinentry.c (register_commands): Add new commands SETOK
and SETCANCEL to set button texts.
(struct pinentry pinentry): Add initializers for new members.
(cmd_setok): New function.
(cmd_setcancel): Likewise.
2002-04-24 Marcus Brinkmann <marcus@g10code.de>
* curses/pinentry-curses.c (dialog_create): Add '<' and '>' around
the user provided button texts. Replace sizeof by strlen to fix
size calculation of ok and cancel button.
2002-04-23 Marcus Brinkmann <marcus@g10code.de>
* pinentry/pinentry.h (struct pinentry): New variables lc_ctype
and lc_messages.
* pinentry/pinentry.c (usage): New options --lc-ctype and
--lc-messages.
(pinentry_parse_opts): Likewise.
(option_handler): Likewise.
(struct pinentry pinentry): New initializers for new members.
* curses/pinentry-curses.c (convert_utf8_string): New function.
(struct dialog): New members ok and cancel.
(dialog_create): New variables ERR, DESCRIPTION, ERROR, PROMPT,
OK, and CANCEL. Initialize them with the localised versions of
the pinentry strings. If in confirm mode, split up the prompt at
'|' and use the values as button texts. Use localised strings.
(dialog_switch_pos): Use localised strings.
(dialog_run): Free dialog strings.
* acinclude.m4 (AM_ICONV): New check from gettext.
* configure.ac: Run AM_ICONV if curses pinentry is build.
Don't check for inttypes.h, don't check size of unsigned int or
unsigned long.
(LIBCAP): Move check to interface independent part.
2002-04-21 Steffen Hansen <steffen@hrhansen.dk>
* Removed X11 dependency and use Qt for grabbing the keyboard.
* Clear the lineedit before asking the user for the PIN.
2002-04-12 Steffen Hansen <steffen@hrhansen.dk>
* Enable pinentry-qt if Qt is found
2002-04-06 Marcus Brinkmann <marcus@g10code.de>
* qt: New directory.
* qt/Makefile.am, qt/cppmemory.cpp, qt/main.cpp,
pinentrycontroller.cpp, qt/pinentrycontroller.h,
qt/pinentrydialog.cpp, qt/pinentrydialog.h: New file, copied from
kde/.
* kde: Directory removed.
* kde/Makefile.am, kde/cppmemory.cpp, kde/main.cpp,
kde/pinentry.desktop, kde/pinentrycontroller.cpp,
kde/pinentrycontroller.h, kde/pinentrydialog.cpp,
kde/pinentrydialog.h: Files removed.
* acinclude.m4: Removed the KE checks and completely overhauled
the Qt checks, putting all Qt checks in their own namespace QT_,
and make it declare QT_-prefixed variables for linking and
compilation.
* configure.ac: Use the new Qt checks instead the KDE checks.
Replace "kde" with "qt" everywhere.
* Makefile.am: Replace "kde" with "qt" everywhere.
2002-04-06 Marcus Brinkmann <marcus@g10code.de>
* acinclude.m4: Reworked the Qt and KDE checks, cutting out a lot
of dead and not-so-dead wood. Gave all KDE checks proper names.
* configure.ac: Use the new names for the KDE checks.
2002-04-06 Marcus Brinkmann <marcus@g10code.de>
* kde/Makefile.am (EXTRA_DIST): Remove variable.
(install-data-local): Remove target.
(uninstall-local): Likewise.
2002-04-05 Marcus Brinkmann <marcus@g10code.de>
Released 0.6.0.
* configure.ac: Set version number to 0.6.
* NEWS: Add information for 0.6.0.
* secmem/Makefile.am (libsecmem_a_SOURCES): Replace secmem.h with memory.
* kde/Makefile.am (pinentry_kde_SOURCES): Move
pinentrydialog.moc.cpp and pinentrycontroller.moc.cpp to ...
(nodist_pinentry_kde_SOURCES): ... this new target.
2002-04-05 Marcus Brinkmann <marcus@g10code.de>
* acinclude.m4: A lot of new checks more or less straight from KDE's
admin/acinclude.m4.in.
* configure.ac (AC_CANONICAL_HOST): Call that macro.
(AC_CHECK_COMPILERS, AC_PATH_KDE): Call those macros if KDE
pinentry is enabled.
* kde/Makefile.am (CXXFLAGS, XXX_PREFIX, XXX_KDE_DEFINES, LIB_QT,
LIB_KDECORE, LIB_KDEUI, KDE_RPATH, MOC, kde_appsdir): Variables removed.
(AM_CPPFLAGS): Replace XXX_KDE_DEFINES by all_includes.
(pinentry_kde_LDFLAGS): Add all_libraries.
* README: Document that automatic check is not possible for KDE.
2002-04-05 Marcus Brinkmann <marcus@g10code.de>
* curses/pinentry-curses.c (dialog_run): Add handling for TAB key.
(dialog_create): New variable description_x. Calculate dimension
of multi-line description correctly.
2002-04-04 Marcus Brinkmann <marcus@g10code.de>
* pinentry/pinentry.h (struct pinentry): New members DISPLAY,
TTYNAME and TTYTYPE.
* pinentry/pinentry.c (pinentry): Likewise.
* pinentry/pinentry.c: Include <string.h>.
(usage): Add new options --display,
--ttyname and --ttytype.
(option_handler): Likewise.
(pinentry_parse_opts): Likewise.
* curses/pinentry-curses.c (dialog_cmd_handler): Use
PINENTRY->ttyname and PINENTRY->ttytype.
2002-03-30 Marcus Brinkmann <marcus@g10code.de>
* acinclude.m4: Add AM_PATH_GLIB and AM_PATH_GTK.
2002-03-29 Marcus Brinkmann <marcus@g10code.de>
* configure.ac: Choose a more appropriate AC_CONFIG_SRCDIR.
2002-03-29 Marcus Brinkmann <marcus@g10code.de>
* kde/Makefile.am (pinentry_kde_LDADD): Link with $(LIBCAP).
* gtk/Makefile.am (install-exec-local): Moved to ...
* Makefile.am (install-exec-local): ... here.
2002-03-29 Marcus Brinkmann <marcus@g10code.de>
* kde/Makefile.am (kde_appsdir): New variable.
* kde/Makefile.am (install-data-local): Use DESTDIR.
(uninstall-local): Likewise.
2002-03-29 Marcus Brinkmann <marcus@g10code.de>
Merge of the gpinentry and curses pinentry program into the
pinentry distribution. For this, the structure of the repository
has been thoroughly overhauled. Some of the changes:
* secmem: New directory with secure memory allocation code.
* pinentry: New directory with pinentry support library.
* curses: New directory with curses frontend.
* gtk: New directory with GTK+ frontend.
* kde: New directory with only the core of the old kpinentry program.
* admin: Directory removed.
* po: Directory removed.
* kpinentry: Directory removed.
* doc: Directory removed.
* jnlib: Directory removed.x
The changes in more detail:
* AUTHORS: Add authors of other pinentry frontends.
* ChangeLog: Add the one from gpinentry.
* Makefile.am: Completely rewritten.
* README: Add content.
* TODO: Add content.
* NEWS: New file from gpinentry.
* THANKS: New file from gpinentry.
* acinclude.m4: New file.
* configure.ac: New file which configures for all frontends.
* curses/Makefile.am, curses/pinentry-curses.c: New files for
curses frontend.
* gtk/Makefile.am, gtk/gtksecentry.c, gtk/gtksecentry.h,
gtk/pinentry-gtk.c: New files, modified from gpinentry, for GTK+
frontend.
* kde/Makefile.am, kde/cppmemory.cpp, kde/main.cpp,
kde/pinentry.desktop, kde/pinentrycontroller.cpp,
kde/pinentrycontroller.h, kde/pinentrydialog.cpp,
kde/pinentrydialog.h: New files, modified from kpinentry, for KDE
frontend.
* pinentry/Makefile.am, pinentry/pinentry.c, pinentry/pinentry.h:
New files containing pinentry support library, partly factored out
from gpinentry.
* secmem/Makefile.am, secmem/memory.h, secmem/secmem-util.h,
secmem/secmem.c, secmem/util.c, secmem/util.h: New files
containing secure memory allocation code common to all pinentry
frontends.
* Makefile.dist, acconfig.h, configure.files, configure.in.in:
Files removed in favor of new configure.ac.
* pinentry.lsm: Removed file never used.
* stamp-h.in: Removed generated file.
* admin/ChangeLog, admin/Makefile.common, admin/acinclude.m4.in,
admin/am_edit, admin/am_edit.py, admin/conf.change.pl,
admin/config.guess, admin/config.pl, admin/config.sub,
admin/configure.in.min, admin/debianrules, admin/depcomp,
admin/install-sh, admin/libtool.m4.in, admin/ltcf-c.sh,
admin/ltcf-cxx.sh, admin/ltcf-gcj.sh, admin/ltconfig,
admin/ltmain.sh, admin/missing, admin/mkinstalldirs, admin/ylwrap:
Removed KDE build suite in favor of configure.ac.
* doc/Makefile.am, doc/en/Makefile.am, doc/en/index.docbook:
Removed files never used.
* jnlib/ChangeLog, jnlib/Makefile.am, jnlib/argparse.c,
jnlib/argparse.h jnlib/dotlock.c, jnlib/dotlock.h,
jnlib/libjnlib-config.h, jnlib/logging.c, jnlib/logging.h,
jnlib/mischelp.h, jnlib/stringhelp.c, jnlib/stringhelp.h,
jnlib/strlist.c, jnlib/strlist.h, jnlib/types.h, jnlib/xmalloc.c,
jnlib/xmalloc.h: Removed files no longer used.
* kpinentry/Makefile.am, kpinentry/cppmemory.cpp,
kpinentry/cppmemory.h, kpinentry/i18n.h, kpinentry/main.cpp,
kpinentry/memory.h, kpinentry/pinentry.desktop,
kpinentry/pinentrycontroller.cpp, kpinentry/pinentrycontroller.h,
kpinentry/pinentrydialog.cpp, kpinentry/pinentrydialog.h,
kpinentry/secmem.cpp, kpinentry/util.cpp, kpinentry/util.h:
Removed files in favor of new files in kde/.
* po/Makefile.am, po/pinentry.pot: Removed files never used.
* autogen.sh: New file.
2002-03-04 Werner Koch <wk@gnupg.org>
* gpinentry.c (enter_callback): New
(create_window): Connect it to the entry field.
2002-02-18 Werner Koch <wk@gnupg.org>
Released 0.5.1.
* gpinentry.c (create_window): Add CONFIRM_MODE.
(cmd_confirm): Implemented.
* assuan/: Updated from NewPG.
2002-01-20 Werner Koch <wk@gnupg.org>
* gpinentry.c (option_handler): New to allow changing of the grab
status.
(grab_keyboard): Shortcut this when global grab is not set.
2002-01-04 Werner Koch <wk@gnupg.org>
Released 0.5.0.
* configure.ac: Bumbed version
* util.h (xtoi_1, xtoi_2): New.
* gpinentry.c (strcpy_escaped): New
(cmd_setdesc, cmd_seterror): Use it here to allo multiline texts.
* gpinentry: Removed debugging outbut
(create_window): Tweaked layout.
2001-12-07 Werner Koch <wk@gnupg.org>
New package gpinentry based on quintuple-agent.
Removed all stuff except for the basic configuration stuff and
what is needed to build gpinentry. Also removed i18n support.
* gpinentry.c: Renamed from secret-query.c
2001-02-12 Robert Bihlmeyer <robbe@orcus.priv.at>
* 1.0.0 released. Woo-hoo!
* configure.in, NEWS: Bumped version.
2001-02-04 Robert Bihlmeyer <robbe@orcus.priv.at>
* agent.c (make_tmpdir): Honor $TMPDIR.
(agent): Would exit on every minor problem. Now, we just close the
offending connection.
Ignore SIGPIPE, so that EPIPE will close connection.
2001-01-11 Robert Bihlmeyer <robbe@orcus.priv.at>
* secmem.c: Move one include statement so that the thing compiles.
* secret-query.c (main): Minor source cosmetics.
2000-11-16 Robert Bihlmeyer <robbe@orcus.priv.at>
* gtksecentry.c (gtk_secure_entry_insert_text): Secured a couple
of memory (de)allocations that were missed. Thanks to John Steele
<jhs28@eng.cam.ac.uk> for spotting these.
2000-11-10 Robert Bihlmeyer <robbe@orcus.priv.at>
* Thoughts: Removed in favor of new TODO.
* README: Remove content and refer to doc/manual.info instead.
* agent.c (do_get): Would burn badly on a premature exit of the
query program (which would occur routinely if you selected /cancel/).
2000-10-25 Robert Bihlmeyer <robbe@orcus.priv.at>
* memory.h: Include sys/types.h for size_t.
2000-10-08 Robert Bihlmeyer <robbe@orcus.priv.at>
* README: Removed apology about missing documentation.
Updated paragraph about Linux capability patch.
Typo & Refill.
2000-10-03 Robert Bihlmeyer <robbe@orcus.priv.at>
* 0.9 released.
* Makefile.am (EXTRA_DIST): Distribute BUGS (the file, that is).
* configure.in, NEWS: Bumped version.
* Makefile.am, configure.in: Add debian subdir.
* README: Recommend GTK+.
Update list of checked platforms.
2000-10-02 Robert Bihlmeyer <robbe@orcus.priv.at>
* secret-query.c (ok): Simplify.
Put empty line between headers and secret.
(usage): Document '--help' and '--version'.
* agent.c (do_get): Use enhanced secret-query output to fill in
options.
2000-10-01 Robert Bihlmeyer <robbe@orcus.priv.at>
* secret-query.c (main): Clarify error.
* client.c (query_options): New global variable.
(main): New option '--query-options' to pass options to the query
program.
(xgetpass): Use it.
* agent.c (main): New option '--query-options' to pass options to
the query program.
(do_get): Use 'query_options'.
* acconfig.h, configure.in: Add QUERY_PROGRAM definition.
* client.c: Remove here.
* agent.c (do_get): Use it here, too.
* secret-query.c (main): New option '--no-global-grab' introduced,
that prevents keyboard grabbing unless the window has focus.
2000-09-11 Robert Bihlmeyer <robbe@orcus.priv.at>
* secret-query.c (usage): Add two missing pieces of "\n\".
* agent.c (main): --nofork is now the default, and the option is
deprecated. New option --fork added to turn forking on again.
Close stdout (and stderr unless debugging) even when not forking,
so that normal usage inside eval is still possible.
(agent): Exit gracefully on HUP, so that logging out now kills the
agent.
* README (Contact Information): Old URL - duh!
(Using Secret Agent): We no longer fork per default.
* Makefile.am (lib/libutil.a): New target, allows targets that not
automatically recurse (but still depend on libutil.a) to succeed.
2000-07-20 Robert Bihlmeyer <robbe@orcus.priv.at>
* secret-query.c (usage): New function.
(main): Parse options: debug, enhanced, help, version.
Turn on locale support.
If enhanced, insert widgets to ask for timeout and insurance.
(ok): If enhanced, print more information on exit.
(grab_keyboard): Die if grab was unsuccessful.
2000-05-31 Robert Bihlmeyer <robbe@orcus.priv.at>
* 0.8 released.
* configure.in, NEWS: Bumped version.
* Makefile.am (SUBDIRS): Include doc.
* configure.in, acconfig.h: Check for ssize_t.
Check for vsnprintf(), strdup().
Generate doc/Makefile.
* apgp.c, agpg.c, agentlib.c, util.c: Include more stuff.
* agent.c (do_get): Use asprintf() instead of snprintf() so we
don't need to roll our own for yet another function.
Fix some includes.
* acinclude.m4: gettext macros copied from automake and fixed.
2000-05-30 Robert Bihlmeyer <robbe@orcus.priv.at>
* configure.in, Makefile.am: Properly include doc subdir.
2000-05-29 Robert Bihlmeyer <robbe@orcus.priv.at>
* configure.in, acconfig.h: Add test for XMESSAGE path.
* client.c (main): Decode command from string to integer code
first, then evaluate that in ifs.
* secret-ask.c: New file, external ask-for-confirmation utility.
* configure.in, Makefile.am: Add secret-ask to programs being
built when GTK is available. Rename QUERY to more descriptive
GTK_PROGRAMS.
* agent.c (do_get): If GTK is available, try executing secret-ask
first.
Only put the comment into the insure-question if there is a
comment.
(main): --csh was missing from usage message.
2000-04-23 Robert Bihlmeyer <robbe@orcus.priv.at>
* secret-query.c (constrain_size): Lower window max_width to
- accomodate bugs in GTK and Scwm.
+ accommodate bugs in GTK and Scwm.
* agent.c (main): Set x_enabled if X appears to be available.
Use it to selectively make FLAGS_INSURE supported.
(do_get): Use it instead of testing at every call.
(do_put): Requests containing unsupported flags fail.
(forget_old_stuff): Would not set next_deadline correctly.
(do_get): Implement FLAGS_INSURE.
* client.c (main): "list" format changed so that comment is to the
far right. Display the deadline as proper date/time, too.
1999-11-11 Robert Bihlmeyer <robbe@orcus.priv.at>
* agent.c (next_deadline): New global variable, holds time when
next secret has to be killed.
(store): Keep it up-to-date.
(forget_old_stuff): New function, reaps secrets ready to kill,
keeps next_deadline updated.
(agent): Use it on all secrets, whenever a deadline is active.
1999-11-08 Robert Bihlmeyer <robbe@orcus.priv.at>
* agent.h (request_put): Add flags, deadline. Increase REQUEST_MAGIC.
(reply_get): Ditto, and increase REPLY_MAGIC.
* agent.c (store): Store flags, deadline in reply.
(do_put): Hand flags, deadline from request on to store().
(do_get): Store on-demand queried secrets without deadline or
special flags, for now.
(agent): Do not ignore obsolete clients, return an error reply.
* agentlib.c (agent_put): Add flags, deadline arguments, and copy
them into the request.
* agentlib.h (agent_put): Update prototype.
* * client.c (main): Added --time-to-live (-t) and --insure (-i)
options, influencing PUT's deadline and flags, respectively.
(main): Change list format to include new attributes.
1999-11-05 Robert Bihlmeyer <robbe@orcus.priv.at>
* configure.in: <getopt.h> check had redundant definition.
* acconfig.h: Remove here, too.
* configure.in: Check for <inttypes.h>. If not found, check for
unsigned {int, long} sizes.
* agent.h: Either include <inttypes.h>, or try to define uint32_t
yourself.
Need to include "config.h".
1999-11-04 Robert Bihlmeyer <robbe@orcus.priv.at>
* Makefile.am (signed-dist): New rule, generates sig for dist.
(%.sig): New rule, for detached signatures in general.
* configure.in: Check for ulong.
* acconfig.h: Document it.
* secmem.c: Unconditionally defining it is no longer necessary
here. But do include <sys/types.h> in all cases.
* 0.7 released.
* secmem.c: ulong is not defined on all systems.
* agent.h: <inttypes.h> should define `uint32_t' as per Unix98, so
we use that.
1999-11-02 Robert Bihlmeyer <robbe@orcus.priv.at>
* agent.h: Augmented requests and replies with magic numbers.
Data structures heavily commented.
All structures and enums typedef'd.
* agent.c: Adapted.
(store): Set magic number in stored reply.
(do_put): Set magic number in reply.
(do_delete): Ditto.
(do_list): Ditto.
(agent): Check magic number in request.
* agentlib.c: Adapted.
(send_request): Set magic number in request. Check it in reply.
* agentlib.h: Adapted.
* agpg.c: Adapted.
* apgp.c: Adapted.
* client.c: Adapted.
1999-10-31 Robert Bihlmeyer <robbe@orcus.priv.at>
* secmem.c: Instead of defining ulong directly, include
<sys/types.h>
* secret-query.c (constrain_size): New function, puts constrains
on size of toplevel window.
(grab_keyboard): Removed protection against multiple calls.
(ungrab_keyboard): New function, cancelling a keyboard grab.
(main): Hang `grab_keyboard' onto map-event which makes it
actually work, hang `ungrab_keyboard' onto unmap-event.
Hang `constrain_size' onto size-request.
* secmem.c: ulong was undefined on some systems.
1999-10-19 Robert Bihlmeyer <robbe@orcus.priv.at>
* Makefile.am (install-exec-local): Ignore setcap errors.
* apgp.c: New program, based on agpg.c, but for pgp2.6.
* Makefile.am (bin_PROGRAMS, apgp_SOURCES): Added it.
* agpg.c (GPG): New constant.
(find_id, main): Use it throughout.
(find_id): Forgot to pclose on success.
* secret-query.c (main): Don't expand anything.
Use a button box for the buttons.
Prompt label can be overridden from the commandline.
* agent.c (do_get): Make spawned secret-query show the id.
* client.c (xgetpass): Pass prompt to secret-query.
(main): Include id in xgetpass prompt.
1999-10-14 Robert Bihlmeyer <robbe@orcus.priv.at>
* configure.in: setcap must be searched outside the usual user
PATH, too.
* Makefile.am (install-exec-local): Set cap_ipc_lock permitted on
installed binaries, if possible.
* agent.c (xdup2): New function, dup2 with error handling.
(move_fd): New function, moves fds.
(store): New function, abstracted out from do_put.
(do_put): Use it.
(do_get): If secret was not found, and DISPLAY is set, try to
query the user about it. If successful, store it.
(main): Route standard file descriptors to /dev/null rather than
just closing them. The latter would confuse children.
1999-10-13 Robert Bihlmeyer <robbe@orcus.priv.at>
* gtksecentry.c, gtksecentry.h: New files, being slightly modified
versions of GTK+'s gtkentry.[ch], spiffed up to use secure memory.
* secret-query.c (ok, unselect, main): Replace GtkEntry with
GtkSecureEntry.
(main): Initialize secure memory.
* Makefile.am (secret_query_SOURCES): Added gtksecentry.[ch].
* README (Security): New chapter.
* configure.in, acconfig.h: Check for POSIX capabilities, and the
setcap program.
* Makefile.am: Link LIBCAP to those binaries using secmem.c.
* util.h: Include <sys/types.h> for size_t.
1999-09-21 Robert Bihlmeyer <robbe@orcus.priv.at>
* secmem.c: Synced with gnupg-1.0 (top new feature: capabilities).
(log_fatal): New function, logs to stderr, and dies.
1999-09-08 Robert Bihlmeyer <robbe@orcus.priv.at>
* secmem.c (log_info): New function, logs to stderr.
* agent.c (agent): select() expects the number of fds, not the
highest fd. So remember that number.
* agent.c (agent): Don't use FD_SETSIZE, which is not defined on
all systems. Remember the number of the highest descriptor
instead.
* configure.in: Replace getline() instead of getdelim() because
this is the function we really need. Still check for getdelim(),
though - there are systems out there where this is provided, but
getline() is not.
1999-09-01 Robert Bihlmeyer <robbe@orcus.priv.at>
* 0.6 released.
* configure.in: Bump version.
* NEWS: Updated.
1999-08-31 Robert Bihlmeyer <robbe@orcus.priv.at>
* util.c (init_uids, lower_privs, raise_privs, drop_privs): New
functions, for setuid binaries, extracted from agent.c.
* util.h: Add prototypes for them.
* agpg.c (main): Use them.
* client.c (main): Ditto.
* agent.c (main): Ditto. Removed code that did the same.
Unconditionally include "asprintf.h" (it protects itself now).
1999-08-25 Robert Bihlmeyer <robbe@orcus.priv.at>
* Makefile.am (SUBDIRS): Process . before test so that "make
check" always builds all in . first.
1999-08-21 Robert Bihlmeyer <robbe@orcus.priv.at>
* secret-query.c: Include "config.h".
* agent.h (reply_list_entry, reply_list): New reply structures.
* agent.c (send_list_entry): New function.
(do_list): First send number of entries, then each entry via
send_list_entry().
* agentlib.c (agent_list): Read entries returned by LIST request.
* client.c (main): Output all entries returned by agent_list().
* agentlib.c (agent_put): Don't construct PUT request in insecure
stack space.
1999-08-20 Robert Bihlmeyer <robbe@orcus.priv.at>
* NEWS: Bump patchlevel.
* configure.in: Bump patchlevel.
Check for missing setenv().
1999-08-09 Robert Bihlmeyer <robbe@orcus.priv.at>
* configure.in: Check for strsignal().
* client-test: Obsoleted by test/client.
* Makefile.am (SUBDIRS): New subdirectory.
* configure.in (AC_OUTPUT): Add here, too.
* 0.5 released.
* README: Explain why secret-client will not output secrets to a
tty, and mention the cat-trick.
* agent.c (main): If seteuid is not available, don't use it and
issue a warning if running setuid.
* configure.in: Run together two REPLACE_FUNCS.
Check for seteuid.
* client-test: Mask out insecure memory warnings.
* agentlib.c (send_request): Let the calling functions reserve
space for the reply, but offer a simple way for simple requests.
(agent_get): Allocate secure memory.
* agent.c (main): Moved secmem_init() after the fork, since that
seemingly munlock's all pages.
- Drop priviledges just in case somebody wants to install this
+ Drop privileges just in case somebody wants to install this
suid-root.
Flush stdout.
* agpg.c (find_id): Would reorder arguments.
Initialize opt_version.
(main): Initialize secure memory.
* Makefile.am (agpg_SOURCES): Link with secure memory module.
* client.c (usage): Fixed another program name reference.
1999-08-06 Robert Bihlmeyer <robbe@orcus.priv.at>
* configure.in: Conditionally define HAVE_GTK.
* acconfig.h: Add here too.
* client.c (xgetpass): Use "secret-query" only if it was built.
(main): Don't output secret (GET command) to ttys.
* agpg.c (find_id): Also print own version if "--version" is given.
(main): Check agent_init() errors.
Print error if exec fails.
* agent.c (main): Added an option to produce csh-compatible
output.
* agent.c, client.c: Forgot the terminating NULL in long options.
Fixed the program names in usage and version output.
* agent.c (create_socket): AF_UNIX and PF_UNIX are Unix98, so
that's what we use. AF_LOCAL, PF_LOCAL removed.
* agentlib.c (agent_init): Ditto.
1999-08-05 Robert Bihlmeyer <robbe@orcus.priv.at>
* 0.4 released.
* configure.in: Bumped version.
Check for missing getdelim.
* cgpg: Removed, obsoleted by agpg.
* Makefile.am: Here, too.
* agpg.c (find_id): New function.
(main): Use it.
* Makefile.am (INCLUDES): Put GTK_FLAGS and GLIB_FLAGS here. It
is the easiest way for sources needing it, and it won't hurt those
that don't.
(agent.o): Explicit command removed accordingly.
1999-08-04 Robert Bihlmeyer <robbe@orcus.priv.at>
* configure.in: Need double quoting in nested AC_MSG_WARN.
* client.c (xgetpass): If no tty is available, but a DISPLAY is,
fork off "secure-query" to read the secret.
Put the fgets into a loop that keeps reading until all of the
secret is read.
* client-test: Unset DISPLAY, so that "secret-query" is never used.
1999-08-03 Robert Bihlmeyer <robbe@orcus.priv.at>
* agpg.c: New file, first cut at a C version of the gpg wrapper,
written in a hurry (20 keys waiting to be signed, and a growling
stomach).
* Makefile.am: Add it to built programs.
(LDADD): New default.
(secret_client_LDADD): Removed, since it was identical to default.
1999-08-01 Robert Bihlmeyer <robbe@orcus.priv.at>
* secret-query.c: New program, queries the user for a password.
* Makefile.am (bin_PROGRAMS): Added it.
* configure.in: Check for GTK+, build "secret-query" only when
that is available.
* cgpg: Extra argument for ID is no longer necessary. cgpg will
scan the gpg args for switches that affect user-id, and determine
the right key itself. Per convention, the key-id is used by "GET".
* configure.in: The project name is now "secret-agent".
* Makefile.am: "agent" & "client" renamed to "secret-agent" &
"secret-client", respectively.
* client-test: Adapt to new names.
* Thoughts: Removed discussion of other names.
Added indication of which things already work.
- * agent.c (make_tmpdir): Removed occurance of "gpg-agent."
+ * agent.c (make_tmpdir): Removed occurrence of "gpg-agent."
* README: First proper version.
* client-test: Context diffs are more portable then unified diffs.
* agent.c, agent.h, agentlib.c, agentlib.h, client.c, memory.h,
util.c, util.h: Banner updated to new name.
1999-07-29 Robert Bihlmeyer <robbe@orcus.priv.at>
* configure.in: Check for missing asprintf.
Check if -lsocket is needed.
* secmem.c (secmem_dump_stats): Replace usage of ulong.
* Makefile.am (INCLUDES): Add the lib subdirectory to include search.
* agent.c: Forgot to include <string.h>.
Include RYO asprintf header if this function is missing.
For the sake of compatibility, provide a definition for AF_LOCAL,
PF_LOCAL, if missing.
* agentlib.c: Ditto.
1999-07-28 Robert Bihlmeyer <robbe@orcus.priv.at>
* Makefile.am (client_SOURCES): Add "secmem.c", "memory.h".
* client.c (xgetpass): Use secmem_malloc() instead of RYO.
(main): Init and shutdown secmem.
1999-07-27 Robert Bihlmeyer <robbe@orcus.priv.at>
* Makefile.am (agent_SOURCES): Add "secmem.c", "i18n.h", "memory.h".
(client_SOURCES): Add "i18n.h".
* client.c (main): Exit on agent_init() failure.
* agent.c (main): Init secmem.
Make --debug switch cumulative.
(cleanup): Shutdown secmem.
(do_put): Use secmem for storage of secrets.
(do_delete): Use secmem_free(). Since this wipes the memory on its
own, wipe() is superflous now.
(agent): Use secmem for inbound requests.
(delete_secret): New function, takes part of do_delete's
functionality.
(do_put): Use it to remove old versions stored under the same id.
(do_delete): Use it to delete secrets.
* memory.h: New file.
* secmem.c: New file, snarfed from GnuPG and modified slightly.
* acinclude.m4: New file.
* configure.in: (ALL_LINGUAS): Expanded list of available
languages. Most of them only have a few translations from gpg,
tough ...
getopt_long test was commented out for debugging,
and left such. Fixed.
Check for mlock.
* acconfig.h: Comment HAVE_BROKEN_MLOCK.
* cgpg: A space was missing.
* i18n.h: New file, centralizing the gettext macro defs.
* agent.c (BLIND): New macro, that blinds out a secret if debug
level is too low.
(do_put): Use it.
(do_get): Use it.
Include i18n.h.
* agentlib.c: Include i18n.h
* client.c (usage): New function.
Usage-message made gettext-friendly.
(xgetpass): Use perror() instead of fprintf().
(main): Use it.
Make comment an optional argument of PUT.
Include i18n.h.
1999-07-26 Robert Bihlmeyer <robbe@orcus.priv.at>
* Makefile.am (client_LDADD): Add lib/libutil.a for portability.
(agent_LDADD): Ditto.
(SUBDIRS): Add lib directory.
(bin_SCRIPTS): New with cgpg, so it gets installed, too.
* configure.in: Add lib/Makefile to output.
1999-07-24 Robert Bihlmeyer <robbe@orcus.priv.at>
* cgpg: New file.
* Makefile.am (EXTRA_DIST): Added it.
* agent.c: Moved inclusion of config.h before inclusion of
libintl.h since the latter needs HAVE_LC_MESSAGE.
* client.c: Ditto.
* clientlib.c: Ditto.
* client-test: Update for new client semantics.
* client.c (check_status): Use debugmsg(). Do nothing if not
debugging.
(xgetpass): New function, getpass replacement that uses mlock'ed
memory.
(main): PUT now asks for the secret rather then getting it from
the commandline.
GET prints only the secret to stdout.
* configure.in: Rearranged.
Check for socklen_t.
* acconfig.h: Added a definition for it.
* client.c: Include packaged getopt.h if the system doesn't provide
one.
* agent.c: Ditto.
(create_socket): Replace AF_FILE, PF_FILE with AF_LOCAL, PF_LOCAL
for portability.
* agentlib.c (agent_init): Ditto.
Explicitly cast addr to a sockaddr pointer.
* Makefile.am (client_LDADD): Added @INTLLIBS@.
(agent_LDADD): Ditto.
1999-07-19 Robert Bihlmeyer <robbe@orcus.priv.at>
* configure.in: Check for getopt.h and getopt_long.
* acconfig.h (HAVE_GETOPT_H): New define.
* lib/getopt.c, lib/getopt1.c, lib/getopt.h: Added.
1999-07-18 Robert Bihlmeyer <robbe@orcus.priv.at>
* 0.2 released.
* NEWS: Updated.
* Makefile.am (agent.o): Mentioning the source explicitly does not
work for srcdir!=builddir.
* client.c (main): Function arguments are not always evaluated in
order, so drop the neat ++optind in favor of optind+1, optind+2,
etc.
* configure.in: Upped version.
1999-06-28 Robert Bihlmeyer <robbe@orcus.priv.at>
* util.h: Added multi-inclusion guard.
* agent.c Include "util.h".
(main): Forgot to exit at end.
(do_delete): Assume that value is a string and wipe it accordingly.
* configure.in (--enable-debug): New switch.
* agent.h (status_t): Added STATUS_COMM_ERR code.
Added multi-inclusion guard.
* client.c (main): Abstracted out most functionality into a
function library, namely:
* agentlib.c: New file.
* agentlib.h: New file.
* Makefile.am (client_SOURCES): Added agentlib.c, agentlib.h.
* Makefile.am (INCLUDES): GLIB_CFLAGS moved again, this time to
the agent.o target.
1999-06-15 Robert Bihlmeyer <robbe@orcus.priv.at>
* 0.1 released.
* Makefile.am (agent_CFLAGS): Removed - did not work.
(INCLUDES): Moved the GLIB stuff here.
1999-06-14 Robert Bihlmeyer <robbe@orcus.priv.at>
* configure.in (ALL_LINGUAS): Added `de'.
* agent.c (do_get): Added more debugmsgs.
(do_put): Wouldn't allocate enough for `value'.
(main): New option "--nofork" prevents forking.
Use macros for the std filedescriptor numbers.
Only close stderr if not debugging.
(main): Make Usage string gettext-friendly.
* client-test (cleanup): New function.
Call it on shell exit.
(client): New function.
Use it instead of calling client binary directly.
diff client output with expected one in GET testcases.
1999-06-13 Robert Bihlmeyer <robbe@orcus.priv.at>
* agent.c (failed_reply): New constant.
(do_list): Use it.
(do_put): The hash key was overwritten - strdup it.
Construct a GET reply and save that in the hash.
(do_get): Just send the preconstructed reply if the id is present,
and failed_reply otherwise.
(do_delete): Actually free the hashed stuff.
* client.c: Exit with error if agent returned STATUS_FAIL.
* Makefile.am (EXTRA_DIST): Added autogen.sh, Thoughts,
client-test.
(TESTS): Added client-test.
(AUTOMAKE_OPTIONS): Added gnits.
Copyright 2002, 2003, 2010 g10 Code GmbH
This file is free software; as a special exception the author gives
unlimited permission to copy and/or distribute it, with or without
modifications, as long as this notice is preserved.
This 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/NEWS b/NEWS
index 4278865..7a7ca76 100644
--- a/NEWS
+++ b/NEWS
@@ -1,342 +1,342 @@
Noteworthy changes in version 0.9.8 (unreleased)
------------------------------------------------
* Qt pinentry now supports repeat mode in one dialog.
* Qt and GTK pinentries now make it possible to show the
entered value.
* Qt pinentry now only grabs the keyboard if an entry field
is focused.
* Fixed foreground handling in pinentry-qt if compiled
with Qt5 for Windows.
* Fixed error output for cached passwords.
* Fixed potential crash in Qt qualitybar calculation.
* GTK keyboard grabbing is now a bit more robust.
Noteworthy changes in version 0.9.7 (2015-12-07)
------------------------------------------------
* Fix regressions in the Qt pinentry.
* Fix minor problems pinnetyr-tty.
* New option --invisible-char.
Noteworthy changes in version 0.9.6 (2015-09-10)
------------------------------------------------
* Many improvements for the dump tty pinentry.
* Use the standard GTK+-2 text entry widget instead of our outdated
and back-then-it-was-more-secure text widget.
* Use the standard Qt text widget.
* Allow for building a static Qt variant.
* Fix regression in w32 pinentry.
Noteworthy changes in version 0.9.5 (2015-07-01)
------------------------------------------------
* Replaced the internal Assuan and gpg-error code by the standard
libassuan and libgpg-error libraries.
* Add a new Emacs pinentry and use as fallback for GUI programs.
* gnome3: The use-password-manager checkbox does now work.
* Gtk: Improved fallback to curses feature.
* curses: Recognize DEL as backspace.
Noteworthy changes in version 0.9.4 (2015-06-05)
------------------------------------------------
* Fix regression in GTK+ and curses pinentries.
Noteworthy changes in version 0.9.3 (2015-06-01)
------------------------------------------------
* Improved documentation
* New pinentry-gnome3
* More improvements for pinentry-tty.
* Fixes for pinentry-curses including support for Ctrl-W, Ctrl-U,
Ctrl-H, Ctrl-L, and Alt-Backspace
* New Assuan command to request clearing an external cache.
* Fixed problems linking to ncursesw.
* All kind of other minor fixes.
Noteworthy changes in version 0.9.2 (2015-05-11)
------------------------------------------------
* Support for saving the passphrase with libsecret.
* Escape key works in the Gtk+ pinentry.
* Improvements for pinentry-tty.
* Minor cleanups for the native Windows pinentry.
Noteworthy changes in version 0.9.1 (2015-03-18)
------------------------------------------------
* Fixed build problems for systems without ncurses.
* Reworked the option parser to allow building on systems without
getopt_long.
* Fixed Qt4 build problems.
Noteworthy changes in version 0.9.0 (2014-10-26)
------------------------------------------------
* New command SETREPEAT. Currently only supported for Gtk+-2.
* Gtk+-2: Pasting using the mouse is now supported.
* curses: Check that it is actually connected to a tty.
* Removed the old qt-3 and gtk+-1 pinentries.
Noteworthy changes in version 0.8.4 (2014-09-18)
------------------------------------------------
* New pinentry-tty version for dumb terminals.
* Qt4: New option to enable pasting the passphrase from clipboard
* Qt4: Improved accessiblity
* Qt4: Raise confirm message windows into foreground
* Qt4 (Windows): Improve the way pinentry-qt raises itself in the
foreground.
* Improved the build system.
Noteworthy changes in version 0.8.3 (2013-04-26)
------------------------------------------------
* Build fixes for newer mingw32 toolchains.
* Add SETTIMEOUT command for the gtk+-2 pinentry.
Noteworthy changes in version 0.8.2 (2012-08-08)
------------------------------------------------
* New SETTIMEOUT command for the qt4 pinentry.
* Wide character support for the curses pinentry.
* Various bug fixes.
Noteworthy changes in version 0.8.1 (2010-12-16)
------------------------------------------------
* The W32 pinentry now supports WindowsCE.
* The GTK pinentry now always sticks to the top and properly grabs
the keyboard.
* The protocol options default-cancel and default-ok now work for the
pinentry-gtk2 and pinentry-qt (that is QT3).
Noteworthy changes in version 0.8.0 (2010-03-03)
------------------------------------------------
* Beautified the qt4 pinentry
* Minor enhancements.
Noteworthy changes in version 0.7.6 (2009-06-19)
------------------------------------------------
* Make Gtk+-2 pinentry transient to the root window.
* Add Qt4 pinentry.
* Add native W32 pinentry.
* Fix utf-8 problem in Qt pinentries.
* Return GPG_ERR_CANCELED if during a "CONFIRM" command the user
closed the window.
* Add quality bar.
Noteworthy changes in version 0.7.5 (2008-02-15)
------------------------------------------------
* Fix cross compilation for Gtk+-2 pinentry.
* New Assuan command GETINFO with subcommands "version" and "pid".
Noteworthy changes in version 0.7.4 (2007-11-29)
------------------------------------------------
* Pinentry-gtk-2 and pinentry-qt now support a simple passphrase
quality indicator.
Noteworthy changes in version 0.7.3 (2007-07-06)
------------------------------------------------
* New command MESSAGE and --one-button compatibility option to
CONFIRM.
* New Assuan option touch-file to set a file which will be touched
after ncurses does not need the display anymore.
* New option --colors=FG,BG,SO to set the colors for the curses
pinentry.
- * Pinentry-w32 does now basicaly work. It needs some finishing
+ * Pinentry-w32 does now basically work. It needs some finishing
though. For example the buttons should resize themself according
to the size of the text.
Noteworthy changes in version 0.7.2 (2005-01-27)
------------------------------------------------
* Remove bug in configure script that would use installed version of
Qt even if another path was explicitely specified with QTDIR.
* Honor the rpath setting for Qt.
* Add GTK+-2 pinentry.
* Install a symbolic link under the name "pinentry" that defaults to
pinentry-gtk, pinentry-qt, pinentry-gtk-2, or pinentry-curses, in
that order.
Noteworthy changes in version 0.7.1 (2004-04-21)
------------------------------------------------
* Removed unneeded Assuan cruft.
* Fixes for *BSD.
Noteworthy changes in version 0.7.0 (2003-12-23)
------------------------------------------------
* Make UTF8 description (prompt, error message, button texts) work.
* Make sure that secmem_term is called before program termination.
* Make assuan in Gtk and Curses pinentry use secure memory for
storage.
* Fixed a bug that would occur if a canceled GETPIN was immediately
followed by a CONFIRM.
* Disabled undo/redo in Qt pinentry.
* Print diagnostics for locale problems and return a new error code
in that case.
Noteworthy changes in version 0.6.8 (2003-02-07)
------------------------------------------------
* Bug fix in pinentry-qt.
Noteworthy changes in version 0.6.7 (2002-11-20)
------------------------------------------------
* Workaround for a bug in the curses version which led to an infinite
loop.
Noteworthy changes in version 0.6.6 (2002-11-09)
------------------------------------------------
* Fixed handling of DISPLAY and --display for the sake of the curses
fallback.
* UTF-8 conversion does now work for the GTK+ and CURSES version.
Noteworthy changes in version 0.6.5 (2002-09-30)
------------------------------------------------
* Handle Assuan options in the qt version.
Noteworthy changes in version 0.6.4 (2002-08-19)
------------------------------------------------
* Handle CONFIRM command in the qt version.
Noteworthy changes in version 0.6.3 (2002-06-26)
------------------------------------------------
* Minor bug fixes to the qt version.
Noteworthy changes in version 0.6.2 (2002-05-13)
------------------------------------------------
* Error texts can now be percent-escaped.
* The Curses pinentry supports multi-line error texts.
* The GTK+ and Qt pinentry can fall back to curses if no display is
available.
Noteworthy changes in version 0.6.1 (2002-04-25)
------------------------------------------------
* The Curses pinentry supports user-provided button texts via the
new SETOK and SETCANCEL commands.
* The Curses pinentry supports setting the desired character set
locale with --lc-ctype and correctly translates the UTF-8 strings
into that.
Noteworthy changes in version 0.6.0 (2002-04-05)
------------------------------------------------
* Merged all pinentry frontends into a single module.
* There is now a Curses frontend.
* The curses pinentry supports --ttyname and --ttytype options to
set the desired input/output terminal and its type.
Noteworthy changes in version 0.5.1 (2002-02-18)
------------------------------------------------
* CONFIRM command works
Noteworthy changes in version 0.5.0 (2002-01-04)
------------------------------------------------
* Window layout is somewhat nicer
* percent escape sequences do now work for SETDESC and SETERROR
diff --git a/build-aux/texinfo.tex b/build-aux/texinfo.tex
index a181898..5552e50 100644
--- a/build-aux/texinfo.tex
+++ b/build-aux/texinfo.tex
@@ -1,8638 +1,8638 @@
% texinfo.tex -- TeX macros to handle Texinfo files.
%
% Load plain if necessary, i.e., if running under initex.
\expandafter\ifx\csname fmtname\endcsname\relax\input plain\fi
%
-\def\texinfoversion{2007-05-03.09}
+\def\texinfoversion{2016-11-03.12}
%
% Copyright (C) 1985, 1986, 1988, 1990, 1991, 1992, 1993, 1994, 1995,
% 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006,
% 2007 Free Software Foundation, Inc.
%
% This texinfo.tex file is free software; you can redistribute it and/or
% modify it under the terms of the GNU General Public License as
% published by the Free Software Foundation; either version 3, or (at
% your option) any later version.
%
% This texinfo.tex file is distributed in the hope that it will be
% useful, but WITHOUT ANY WARRANTY; without even the implied warranty
% of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
% General Public License for more details.
%
% You should have received a copy of the GNU General Public License
% along with this texinfo.tex file; see the file COPYING. If not,
% see <http://www.gnu.org/licenses/>.
%
% 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}
\chardef\other=12
% We never want plain's \outer definition of \+ in Texinfo.
% For @tex, we can use \tabalign.
\let\+ = \relax
% Save some plain tex macros whose names we will redefine.
\let\ptexb=\b
\let\ptexbullet=\bullet
\let\ptexc=\c
\let\ptexcomma=\,
\let\ptexdot=\.
\let\ptexdots=\dots
\let\ptexend=\end
\let\ptexequiv=\equiv
\let\ptexexclam=\!
\let\ptexfootnote=\footnote
\let\ptexgtr=>
\let\ptexhat=^
\let\ptexi=\i
\let\ptexindent=\indent
\let\ptexinsert=\insert
\let\ptexlbrace=\{
\let\ptexless=<
\let\ptexnewwrite\newwrite
\let\ptexnoindent=\noindent
\let\ptexplus=+
\let\ptexrbrace=\}
\let\ptexslash=\/
\let\ptexstar=\*
\let\ptext=\t
% If this character appears in an error message or help string, it
% starts a new line in the output.
\newlinechar = `^^J
% Use TeX 3.0's \inputlineno to get the line number, for better error
% messages, but if we're using an old version of TeX, don't do anything.
%
\ifx\inputlineno\thisisundefined
\let\linenumber = \empty % Pre-3.0.
\else
\def\linenumber{l.\the\inputlineno:\space}
\fi
% Set up fixed words for English if not already set.
\ifx\putwordAppendix\undefined \gdef\putwordAppendix{Appendix}\fi
\ifx\putwordChapter\undefined \gdef\putwordChapter{Chapter}\fi
\ifx\putwordfile\undefined \gdef\putwordfile{file}\fi
\ifx\putwordin\undefined \gdef\putwordin{in}\fi
\ifx\putwordIndexIsEmpty\undefined \gdef\putwordIndexIsEmpty{(Index is empty)}\fi
\ifx\putwordIndexNonexistent\undefined \gdef\putwordIndexNonexistent{(Index is nonexistent)}\fi
\ifx\putwordInfo\undefined \gdef\putwordInfo{Info}\fi
\ifx\putwordInstanceVariableof\undefined \gdef\putwordInstanceVariableof{Instance Variable of}\fi
\ifx\putwordMethodon\undefined \gdef\putwordMethodon{Method on}\fi
\ifx\putwordNoTitle\undefined \gdef\putwordNoTitle{No Title}\fi
\ifx\putwordof\undefined \gdef\putwordof{of}\fi
\ifx\putwordon\undefined \gdef\putwordon{on}\fi
\ifx\putwordpage\undefined \gdef\putwordpage{page}\fi
\ifx\putwordsection\undefined \gdef\putwordsection{section}\fi
\ifx\putwordSection\undefined \gdef\putwordSection{Section}\fi
\ifx\putwordsee\undefined \gdef\putwordsee{see}\fi
\ifx\putwordSee\undefined \gdef\putwordSee{See}\fi
\ifx\putwordShortTOC\undefined \gdef\putwordShortTOC{Short Contents}\fi
\ifx\putwordTOC\undefined \gdef\putwordTOC{Table of Contents}\fi
%
\ifx\putwordMJan\undefined \gdef\putwordMJan{January}\fi
\ifx\putwordMFeb\undefined \gdef\putwordMFeb{February}\fi
\ifx\putwordMMar\undefined \gdef\putwordMMar{March}\fi
\ifx\putwordMApr\undefined \gdef\putwordMApr{April}\fi
\ifx\putwordMMay\undefined \gdef\putwordMMay{May}\fi
\ifx\putwordMJun\undefined \gdef\putwordMJun{June}\fi
\ifx\putwordMJul\undefined \gdef\putwordMJul{July}\fi
\ifx\putwordMAug\undefined \gdef\putwordMAug{August}\fi
\ifx\putwordMSep\undefined \gdef\putwordMSep{September}\fi
\ifx\putwordMOct\undefined \gdef\putwordMOct{October}\fi
\ifx\putwordMNov\undefined \gdef\putwordMNov{November}\fi
\ifx\putwordMDec\undefined \gdef\putwordMDec{December}\fi
%
\ifx\putwordDefmac\undefined \gdef\putwordDefmac{Macro}\fi
\ifx\putwordDefspec\undefined \gdef\putwordDefspec{Special Form}\fi
\ifx\putwordDefvar\undefined \gdef\putwordDefvar{Variable}\fi
\ifx\putwordDefopt\undefined \gdef\putwordDefopt{User Option}\fi
\ifx\putwordDeffunc\undefined \gdef\putwordDeffunc{Function}\fi
% Since the category of space is not known, we have to be careful.
\chardef\spacecat = 10
\def\spaceisspace{\catcode`\ =\spacecat}
% sometimes characters are active, so we need control sequences.
\chardef\colonChar = `\:
\chardef\commaChar = `\,
\chardef\dashChar = `\-
\chardef\dotChar = `\.
\chardef\exclamChar= `\!
\chardef\lquoteChar= `\`
\chardef\questChar = `\?
\chardef\rquoteChar= `\'
\chardef\semiChar = `\;
\chardef\underChar = `\_
% Ignore a token.
%
\def\gobble#1{}
% The following is used inside several \edef's.
\def\makecsname#1{\expandafter\noexpand\csname#1\endcsname}
% Hyphenation fixes.
\hyphenation{
Flor-i-da Ghost-script Ghost-view Mac-OS Post-Script
ap-pen-dix bit-map bit-maps
data-base data-bases eshell fall-ing half-way long-est man-u-script
man-u-scripts mini-buf-fer mini-buf-fers over-view par-a-digm
par-a-digms rath-er rec-tan-gu-lar ro-bot-ics se-vere-ly set-up spa-ces
spell-ing spell-ings
stand-alone strong-est time-stamp time-stamps which-ever white-space
wide-spread wrap-around
}
% Margin to add to right of even pages, to left of odd pages.
\newdimen\bindingoffset
\newdimen\normaloffset
\newdimen\pagewidth \newdimen\pageheight
% For a final copy, take out the rectangles
% that mark overfull boxes (in case you have decided
% that the text looks ok even though it passes the margin).
%
\def\finalout{\overfullrule=0pt}
% @| inserts a changebar to the left of the current line. It should
% surround any changed text. This approach does *not* work if the
% change spans more than two lines of output. To handle that, we would
% have adopt a much more difficult approach (putting marks into the main
% vertical list for the beginning and end of each change).
%
\def\|{%
% \vadjust can only be used in horizontal mode.
\leavevmode
%
% Append this vertical mode material after the current line in the output.
\vadjust{%
% We want to insert a rule with the height and depth of the current
% leading; that is exactly what \strutbox is supposed to record.
\vskip-\baselineskip
%
% \vadjust-items are inserted at the left edge of the type. So
% the \llap here moves out into the left-hand margin.
\llap{%
%
% For a thicker or thinner bar, change the `1pt'.
\vrule height\baselineskip width1pt
%
% This is the space between the bar and the text.
\hskip 12pt
}%
}%
}
% Sometimes it is convenient to have everything in the transcript file
% and nothing on the terminal. We don't just call \tracingall here,
% since that produces some useless output on the terminal. We also make
% some effort to order the tracing commands to reduce output in the log
% file; cf. trace.sty in LaTeX.
%
\def\gloggingall{\begingroup \globaldefs = 1 \loggingall \endgroup}%
\def\loggingall{%
\tracingstats2
\tracingpages1
\tracinglostchars2 % 2 gives us more in etex
\tracingparagraphs1
\tracingoutput1
\tracingmacros2
\tracingrestores1
\showboxbreadth\maxdimen \showboxdepth\maxdimen
\ifx\eTeXversion\undefined\else % etex gives us more logging
\tracingscantokens1
\tracingifs1
\tracinggroups1
\tracingnesting2
\tracingassigns1
\fi
\tracingcommands3 % 3 gives us more in etex
\errorcontextlines16
}%
% add check for \lastpenalty to plain's definitions. If the last thing
% we did was a \nobreak, we don't want to insert more space.
%
\def\smallbreak{\ifnum\lastpenalty<10000\par\ifdim\lastskip<\smallskipamount
\removelastskip\penalty-50\smallskip\fi\fi}
\def\medbreak{\ifnum\lastpenalty<10000\par\ifdim\lastskip<\medskipamount
\removelastskip\penalty-100\medskip\fi\fi}
\def\bigbreak{\ifnum\lastpenalty<10000\par\ifdim\lastskip<\bigskipamount
\removelastskip\penalty-200\bigskip\fi\fi}
% For @cropmarks command.
% Do @cropmarks to get crop marks.
%
\newif\ifcropmarks
\let\cropmarks = \cropmarkstrue
%
% Dimensions to add cropmarks at corners.
% Added by P. A. MacKay, 12 Nov. 1986
%
\newdimen\outerhsize \newdimen\outervsize % set by the paper size routines
\newdimen\cornerlong \cornerlong=1pc
\newdimen\cornerthick \cornerthick=.3pt
\newdimen\topandbottommargin \topandbottommargin=.75in
% Main output routine.
\chardef\PAGE = 255
\output = {\onepageout{\pagecontents\PAGE}}
\newbox\headlinebox
\newbox\footlinebox
% \onepageout takes a vbox as an argument. Note that \pagecontents
% does insertions, but you have to call it yourself.
\def\onepageout#1{%
\ifcropmarks \hoffset=0pt \else \hoffset=\normaloffset \fi
%
\ifodd\pageno \advance\hoffset by \bindingoffset
\else \advance\hoffset by -\bindingoffset\fi
%
% Do this outside of the \shipout so @code etc. will be expanded in
% the headline as they should be, not taken literally (outputting ''code).
\setbox\headlinebox = \vbox{\let\hsize=\pagewidth \makeheadline}%
\setbox\footlinebox = \vbox{\let\hsize=\pagewidth \makefootline}%
%
{%
% Have to do this stuff outside the \shipout because we want it to
% take effect in \write's, yet the group defined by the \vbox ends
% before the \shipout runs.
%
\indexdummies % don't expand commands in the output.
\normalturnoffactive % \ in index entries must not stay \, e.g., if
% the page break happens to be in the middle of an example.
% We don't want .vr (or whatever) entries like this:
% \entry{{\tt \indexbackslash }acronym}{32}{\code {\acronym}}
% "\acronym" won't work when it's read back in;
% it needs to be
% {\code {{\tt \backslashcurfont }acronym}
\shipout\vbox{%
% Do this early so pdf references go to the beginning of the page.
\ifpdfmakepagedest \pdfdest name{\the\pageno} xyz\fi
%
\ifcropmarks \vbox to \outervsize\bgroup
\hsize = \outerhsize
\vskip-\topandbottommargin
\vtop to0pt{%
\line{\ewtop\hfil\ewtop}%
\nointerlineskip
\line{%
\vbox{\moveleft\cornerthick\nstop}%
\hfill
\vbox{\moveright\cornerthick\nstop}%
}%
\vss}%
\vskip\topandbottommargin
\line\bgroup
\hfil % center the page within the outer (page) hsize.
\ifodd\pageno\hskip\bindingoffset\fi
\vbox\bgroup
\fi
%
\unvbox\headlinebox
\pagebody{#1}%
\ifdim\ht\footlinebox > 0pt
% Only leave this space if the footline is nonempty.
% (We lessened \vsize for it in \oddfootingyyy.)
% The \baselineskip=24pt in plain's \makefootline has no effect.
\vskip 24pt
\unvbox\footlinebox
\fi
%
\ifcropmarks
\egroup % end of \vbox\bgroup
\hfil\egroup % end of (centering) \line\bgroup
\vskip\topandbottommargin plus1fill minus1fill
\boxmaxdepth = \cornerthick
\vbox to0pt{\vss
\line{%
\vbox{\moveleft\cornerthick\nsbot}%
\hfill
\vbox{\moveright\cornerthick\nsbot}%
}%
\nointerlineskip
\line{\ewbot\hfil\ewbot}%
}%
\egroup % \vbox from first cropmarks clause
\fi
}% end of \shipout\vbox
}% end of group with \indexdummies
\advancepageno
\ifnum\outputpenalty>-20000 \else\dosupereject\fi
}
\newinsert\margin \dimen\margin=\maxdimen
\def\pagebody#1{\vbox to\pageheight{\boxmaxdepth=\maxdepth #1}}
{\catcode`\@ =11
\gdef\pagecontents#1{\ifvoid\topins\else\unvbox\topins\fi
% marginal hacks, juha@viisa.uucp (Juha Takala)
\ifvoid\margin\else % marginal info is present
\rlap{\kern\hsize\vbox to\z@{\kern1pt\box\margin \vss}}\fi
\dimen@=\dp#1 \unvbox#1
\ifvoid\footins\else\vskip\skip\footins\footnoterule \unvbox\footins\fi
\ifr@ggedbottom \kern-\dimen@ \vfil \fi}
}
% Here are the rules for the cropmarks. Note that they are
% offset so that the space between them is truly \outerhsize or \outervsize
% (P. A. MacKay, 12 November, 1986)
%
\def\ewtop{\vrule height\cornerthick depth0pt width\cornerlong}
\def\nstop{\vbox
{\hrule height\cornerthick depth\cornerlong width\cornerthick}}
\def\ewbot{\vrule height0pt depth\cornerthick width\cornerlong}
\def\nsbot{\vbox
{\hrule height\cornerlong depth\cornerthick width\cornerthick}}
% Parse an argument, then pass it to #1. The argument is the rest of
% the input line (except we remove a trailing comment). #1 should be a
% macro which expects an ordinary undelimited TeX argument.
%
\def\parsearg{\parseargusing{}}
\def\parseargusing#1#2{%
\def\argtorun{#2}%
\begingroup
\obeylines
\spaceisspace
#1%
\parseargline\empty% Insert the \empty token, see \finishparsearg below.
}
{\obeylines %
\gdef\parseargline#1^^M{%
\endgroup % End of the group started in \parsearg.
\argremovecomment #1\comment\ArgTerm%
}%
}
% First remove any @comment, then any @c comment.
\def\argremovecomment#1\comment#2\ArgTerm{\argremovec #1\c\ArgTerm}
\def\argremovec#1\c#2\ArgTerm{\argcheckspaces#1\^^M\ArgTerm}
-% Each occurence of `\^^M' or `<space>\^^M' is replaced by a single space.
+% Each occurrence 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
% Do not use \next, perhaps the caller of \parsearg uses it; reuse \temp:
\let\temp\finishparsearg
\else
\let\temp\argcheckspaces
\fi
% Put the space token in:
\temp#1 #3\ArgTerm
}
% If a _delimited_ argument is enclosed in braces, they get stripped; so
% to get _exactly_ the rest of the line, we had to prevent such situation.
% We prepended an \empty token at the very beginning and we expand it now,
% just before passing the control to \argtorun.
-% (Similarily, we have to think about #3 of \argcheckspacesY above: it is
+% (Similarly, we have to think about #3 of \argcheckspacesY above: it is
% either the null string, or it ends with \^^M---thus there is no danger
% that a pair of braces would be stripped.
%
% But first, we have to remove the trailing space token.
%
\def\finishparsearg#1 \ArgTerm{\expandafter\argtorun\expandafter{#1}}
% \parseargdef\foo{...}
% is roughly equivalent to
% \def\foo{\parsearg\Xfoo}
% \def\Xfoo#1{...}
%
% Actually, I use \csname\string\foo\endcsname, ie. \\foo, as it is my
% favourite TeX trick. --kasal, 16nov03
\def\parseargdef#1{%
\expandafter \doparseargdef \csname\string#1\endcsname #1%
}
\def\doparseargdef#1#2{%
\def#2{\parsearg#1}%
\def#1##1%
}
% Several utility definitions with active space:
{
\obeyspaces
\gdef\obeyedspace{ }
% Make each space character in the input produce a normal interword
% space in the output. Don't allow a line break at this space, as this
% is used only in environments like @example, where each line of input
% should produce a line of output anyway.
%
\gdef\sepspaces{\obeyspaces\let =\tie}
% If an index command is used in an @example environment, any spaces
% therein should become regular spaces in the raw index file, not the
% expansion of \tie (\leavevmode \penalty \@M \ ).
\gdef\unsepspaces{\let =\space}
}
\def\flushcr{\ifx\par\lisppar \def\next##1{}\else \let\next=\relax \fi \next}
% Define the framework for environments in texinfo.tex. It's used like this:
%
% \envdef\foo{...}
% \def\Efoo{...}
%
% It's the responsibility of \envdef to insert \begingroup before the
% actual body; @end closes the group after calling \Efoo. \envdef also
% defines \thisenv, so the current environment is known; @end checks
% whether the environment name matches. The \checkenv macro can also be
% used to check whether the current environment is the one expected.
%
% Non-false conditionals (@iftex, @ifset) don't fit into this, so they
-% are not treated as enviroments; they don't open a group. (The
+% are not treated as environments; they don't open a group. (The
% implementation of @end takes care not to call \endgroup in this
% special case.)
% At runtime, environments start with this:
\def\startenvironment#1{\begingroup\def\thisenv{#1}}
% initialize
\let\thisenv\empty
% ... but they get defined via ``\envdef\foo{...}'':
\long\def\envdef#1#2{\def#1{\startenvironment#1#2}}
\def\envparseargdef#1#2{\parseargdef#1{\startenvironment#1#2}}
% Check whether we're in the right environment:
\def\checkenv#1{%
\def\temp{#1}%
\ifx\thisenv\temp
\else
\badenverr
\fi
}
% Evironment mismatch, #1 expected:
\def\badenverr{%
\errhelp = \EMsimple
\errmessage{This command can appear only \inenvironment\temp,
not \inenvironment\thisenv}%
}
\def\inenvironment#1{%
\ifx#1\empty
out of any environment%
\else
in environment \expandafter\string#1%
\fi
}
% @end foo executes the definition of \Efoo.
% But first, it executes a specialized version of \checkenv
%
\parseargdef\end{%
\if 1\csname iscond.#1\endcsname
\else
% The general wording of \badenverr may not be ideal, but... --kasal, 06nov03
\expandafter\checkenv\csname#1\endcsname
\csname E#1\endcsname
\endgroup
\fi
}
\newhelp\EMsimple{Press RETURN to continue.}
%% Simple single-character @ commands
% @@ prints an @
% Kludge this until the fonts are right (grr).
\def\@{{\tt\char64}}
% This is turned off because it was never documented
% and you can use @w{...} around a quote to suppress ligatures.
%% Define @` and @' to be the same as ` and '
%% but suppressing ligatures.
%\def\`{{`}}
%\def\'{{'}}
% Used to generate quoted braces.
\def\mylbrace {{\tt\char123}}
\def\myrbrace {{\tt\char125}}
\let\{=\mylbrace
\let\}=\myrbrace
\begingroup
% Definitions to produce \{ and \} commands for indices,
% and @{ and @} for the aux/toc files.
\catcode`\{ = \other \catcode`\} = \other
\catcode`\[ = 1 \catcode`\] = 2
\catcode`\! = 0 \catcode`\\ = \other
!gdef!lbracecmd[\{]%
!gdef!rbracecmd[\}]%
!gdef!lbraceatcmd[@{]%
!gdef!rbraceatcmd[@}]%
!endgroup
% @comma{} to avoid , parsing problems.
\let\comma = ,
% Accents: @, @dotaccent @ringaccent @ubaraccent @udotaccent
% Others are defined by plain TeX: @` @' @" @^ @~ @= @u @v @H.
\let\, = \c
\let\dotaccent = \.
\def\ringaccent#1{{\accent23 #1}}
\let\tieaccent = \t
\let\ubaraccent = \b
\let\udotaccent = \d
% Other special characters: @questiondown @exclamdown @ordf @ordm
% Plain TeX defines: @AA @AE @O @OE @L (plus lowercase versions) @ss.
\def\questiondown{?`}
\def\exclamdown{!`}
\def\ordf{\leavevmode\raise1ex\hbox{\selectfonts\lllsize \underbar{a}}}
\def\ordm{\leavevmode\raise1ex\hbox{\selectfonts\lllsize \underbar{o}}}
% Dotless i and dotless j, used for accents.
\def\imacro{i}
\def\jmacro{j}
\def\dotless#1{%
\def\temp{#1}%
\ifx\temp\imacro \ptexi
\else\ifx\temp\jmacro \j
\else \errmessage{@dotless can be used only with i or j}%
\fi\fi
}
% The \TeX{} logo, as in plain, but resetting the spacing so that a
% period following counts as ending a sentence. (Idea found in latex.)
%
\edef\TeX{\TeX \spacefactor=1000 }
% @LaTeX{} logo. Not quite the same results as the definition in
% latex.ltx, since we use a different font for the raised A; it's most
% convenient for us to use an explicitly smaller font, rather than using
% the \scriptstyle font (since we don't reset \scriptstyle and
% \scriptscriptstyle).
%
\def\LaTeX{%
L\kern-.36em
{\setbox0=\hbox{T}%
\vbox to \ht0{\hbox{\selectfonts\lllsize A}\vss}}%
\kern-.15em
\TeX
}
% Be sure we're in horizontal mode when doing a tie, since we make space
% equivalent to this in @example-like environments. Otherwise, a space
% at the beginning of a line will start with \penalty -- and
% since \penalty is valid in vertical mode, we'd end up putting the
% penalty on the vertical list instead of in the new paragraph.
{\catcode`@ = 11
% Avoid using \@M directly, because that causes trouble
% if the definition is written into an index file.
\global\let\tiepenalty = \@M
\gdef\tie{\leavevmode\penalty\tiepenalty\ }
}
% @: forces normal size whitespace following.
\def\:{\spacefactor=1000 }
% @* forces a line break.
\def\*{\hfil\break\hbox{}\ignorespaces}
% @/ allows a line break.
\let\/=\allowbreak
% @. is an end-of-sentence period.
\def\.{.\spacefactor=\endofsentencespacefactor\space}
% @! is an end-of-sentence bang.
\def\!{!\spacefactor=\endofsentencespacefactor\space}
% @? is an end-of-sentence query.
\def\?{?\spacefactor=\endofsentencespacefactor\space}
% @frenchspacing on|off says whether to put extra space after punctuation.
%
\def\onword{on}
\def\offword{off}
%
\parseargdef\frenchspacing{%
\def\temp{#1}%
\ifx\temp\onword \plainfrenchspacing
\else\ifx\temp\offword \plainnonfrenchspacing
\else
\errhelp = \EMsimple
\errmessage{Unknown @frenchspacing option `\temp', must be on/off}%
\fi\fi
}
% @w prevents a word break. Without the \leavevmode, @w at the
% beginning of a paragraph, when TeX is still in vertical mode, would
% produce a whole line of output instead of starting the paragraph.
\def\w#1{\leavevmode\hbox{#1}}
% @group ... @end group forces ... to be all on one page, by enclosing
% it in a TeX vbox. We use \vtop instead of \vbox to construct the box
% to keep its height that of a normal line. According to the rules for
% \topskip (p.114 of the TeXbook), the glue inserted is
% max (\topskip - \ht (first item), 0). If that height is large,
% therefore, no glue is inserted, and the space between the headline and
% the text is small, which looks bad.
%
% Another complication is that the group might be very large. This can
% cause the glue on the previous page to be unduly stretched, because it
% does not have much material. In this case, it's better to add an
% explicit \vfill so that the extra space is at the bottom. The
% threshold for doing this is if the group is more than \vfilllimit
% percent of a page (\vfilllimit can be changed inside of @tex).
%
\newbox\groupbox
\def\vfilllimit{0.7}
%
\envdef\group{%
\ifnum\catcode`\^^M=\active \else
\errhelp = \groupinvalidhelp
\errmessage{@group invalid in context where filling is enabled}%
\fi
\startsavinginserts
%
\setbox\groupbox = \vtop\bgroup
% Do @comment since we are called inside an environment such as
% @example, where each end-of-line in the input causes an
% end-of-line in the output. We don't want the end-of-line after
% the `@group' to put extra space in the output. Since @group
% should appear on a line by itself (according to the Texinfo
% manual), we don't worry about eating any user text.
\comment
}
%
% The \vtop produces a box with normal height and large depth; thus, TeX puts
% \baselineskip glue before it, and (when the next line of text is done)
% \lineskip glue after it. Thus, space below is not quite equal to space
% above. But it's pretty close.
\def\Egroup{%
% To get correct interline space between the last line of the group
% and the first line afterwards, we have to propagate \prevdepth.
\endgraf % Not \par, as it may have been set to \lisppar.
\global\dimen1 = \prevdepth
\egroup % End the \vtop.
% \dimen0 is the vertical size of the group's box.
\dimen0 = \ht\groupbox \advance\dimen0 by \dp\groupbox
% \dimen2 is how much space is left on the page (more or less).
\dimen2 = \pageheight \advance\dimen2 by -\pagetotal
% if the group doesn't fit on the current page, and it's a big big
% group, force a page break.
\ifdim \dimen0 > \dimen2
\ifdim \pagetotal < \vfilllimit\pageheight
\page
\fi
\fi
\box\groupbox
\prevdepth = \dimen1
\checkinserts
}
%
% TeX puts in an \escapechar (i.e., `@') at the beginning of the help
% message, so this ends up printing `@group can only ...'.
%
\newhelp\groupinvalidhelp{%
group can only be used in environments such as @example,^^J%
where each line of input produces a line of output.}
% @need space-in-mils
% forces a page break if there is not space-in-mils remaining.
\newdimen\mil \mil=0.001in
% Old definition--didn't work.
%\parseargdef\need{\par %
%% This method tries to make TeX break the page naturally
%% if the depth of the box does not fit.
%{\baselineskip=0pt%
%\vtop to #1\mil{\vfil}\kern -#1\mil\nobreak
%\prevdepth=-1000pt
%}}
\parseargdef\need{%
% Ensure vertical mode, so we don't make a big box in the middle of a
% paragraph.
\par
%
% If the @need value is less than one line space, it's useless.
\dimen0 = #1\mil
\dimen2 = \ht\strutbox
\advance\dimen2 by \dp\strutbox
\ifdim\dimen0 > \dimen2
%
% Do a \strut just to make the height of this box be normal, so the
% normal leading is inserted relative to the preceding line.
% And a page break here is fine.
\vtop to #1\mil{\strut\vfil}%
%
% TeX does not even consider page breaks if a penalty added to the
% main vertical list is 10000 or more. But in order to see if the
% empty box we just added fits on the page, we must make it consider
% page breaks. On the other hand, we don't want to actually break the
% page after the empty box. So we use a penalty of 9999.
%
% There is an extremely small chance that TeX will actually break the
% page at this \penalty, if there are no other feasible breakpoints in
% sight. (If the user is using lots of big @group commands, which
% almost-but-not-quite fill up a page, TeX will have a hard time doing
% good page breaking, for example.) However, I could not construct an
% example where a page broke at this \penalty; if it happens in a real
% document, then we can reconsider our strategy.
\penalty9999
%
% Back up by the size of the box, whether we did a page break or not.
\kern -#1\mil
%
% Do not allow a page break right after this kern.
\nobreak
\fi
}
% @br forces paragraph break (and is undocumented).
\let\br = \par
% @page forces the start of a new page.
%
\def\page{\par\vfill\supereject}
% @exdent text....
% outputs text on separate line in roman font, starting at standard page margin
% This records the amount of indent in the innermost environment.
% That's how much \exdent should take out.
\newskip\exdentamount
% This defn is used inside fill environments such as @defun.
\parseargdef\exdent{\hfil\break\hbox{\kern -\exdentamount{\rm#1}}\hfil\break}
% This defn is used inside nofill environments such as @example.
\parseargdef\nofillexdent{{\advance \leftskip by -\exdentamount
\leftline{\hskip\leftskip{\rm#1}}}}
% @inmargin{WHICH}{TEXT} puts TEXT in the WHICH margin next to the current
% paragraph. For more general purposes, use the \margin insertion
% class. WHICH is `l' or `r'.
%
\newskip\inmarginspacing \inmarginspacing=1cm
\def\strutdepth{\dp\strutbox}
%
\def\doinmargin#1#2{\strut\vadjust{%
\nobreak
\kern-\strutdepth
\vtop to \strutdepth{%
\baselineskip=\strutdepth
\vss
% if you have multiple lines of stuff to put here, you'll need to
% make the vbox yourself of the appropriate size.
\ifx#1l%
\llap{\ignorespaces #2\hskip\inmarginspacing}%
\else
\rlap{\hskip\hsize \hskip\inmarginspacing \ignorespaces #2}%
\fi
\null
}%
}}
\def\inleftmargin{\doinmargin l}
\def\inrightmargin{\doinmargin r}
%
% @inmargin{TEXT [, RIGHT-TEXT]}
% (if RIGHT-TEXT is given, use TEXT for left page, RIGHT-TEXT for right;
% else use TEXT for both).
%
\def\inmargin#1{\parseinmargin #1,,\finish}
\def\parseinmargin#1,#2,#3\finish{% not perfect, but better than nothing.
\setbox0 = \hbox{\ignorespaces #2}%
\ifdim\wd0 > 0pt
\def\lefttext{#1}% have both texts
\def\righttext{#2}%
\else
\def\lefttext{#1}% have only one text
\def\righttext{#1}%
\fi
%
\ifodd\pageno
\def\temp{\inrightmargin\righttext}% odd page -> outside is right margin
\else
\def\temp{\inleftmargin\lefttext}%
\fi
\temp
}
% @include file insert text of that file as input.
%
\def\include{\parseargusing\filenamecatcodes\includezzz}
\def\includezzz#1{%
\pushthisfilestack
\def\thisfile{#1}%
{%
\makevalueexpandable
\def\temp{\input #1 }%
\expandafter
}\temp
\popthisfilestack
}
\def\filenamecatcodes{%
\catcode`\\=\other
\catcode`~=\other
\catcode`^=\other
\catcode`_=\other
\catcode`|=\other
\catcode`<=\other
\catcode`>=\other
\catcode`+=\other
\catcode`-=\other
}
\def\pushthisfilestack{%
\expandafter\pushthisfilestackX\popthisfilestack\StackTerm
}
\def\pushthisfilestackX{%
\expandafter\pushthisfilestackY\thisfile\StackTerm
}
\def\pushthisfilestackY #1\StackTerm #2\StackTerm {%
\gdef\popthisfilestack{\gdef\thisfile{#1}\gdef\popthisfilestack{#2}}%
}
\def\popthisfilestack{\errthisfilestackempty}
\def\errthisfilestackempty{\errmessage{Internal error:
the stack of filenames is empty.}}
\def\thisfile{}
% @center line
% outputs that line, centered.
%
\parseargdef\center{%
\ifhmode
\let\next\centerH
\else
\let\next\centerV
\fi
\next{\hfil \ignorespaces#1\unskip \hfil}%
}
\def\centerH#1{%
{%
\hfil\break
\advance\hsize by -\leftskip
\advance\hsize by -\rightskip
\line{#1}%
\break
}%
}
\def\centerV#1{\line{\kern\leftskip #1\kern\rightskip}}
% @sp n outputs n lines of vertical space
\parseargdef\sp{\vskip #1\baselineskip}
% @comment ...line which is ignored...
% @c is the same as @comment
% @ignore ... @end ignore is another way to write a comment
\def\comment{\begingroup \catcode`\^^M=\other%
\catcode`\@=\other \catcode`\{=\other \catcode`\}=\other%
\commentxxx}
{\catcode`\^^M=\other \gdef\commentxxx#1^^M{\endgroup}}
\let\c=\comment
% @paragraphindent NCHARS
% We'll use ems for NCHARS, close enough.
% NCHARS can also be the word `asis' or `none'.
% We cannot feasibly implement @paragraphindent asis, though.
%
\def\asisword{asis} % no translation, these are keywords
\def\noneword{none}
%
\parseargdef\paragraphindent{%
\def\temp{#1}%
\ifx\temp\asisword
\else
\ifx\temp\noneword
\defaultparindent = 0pt
\else
\defaultparindent = #1em
\fi
\fi
\parindent = \defaultparindent
}
% @exampleindent NCHARS
% We'll use ems for NCHARS like @paragraphindent.
% It seems @exampleindent asis isn't necessary, but
% I preserve it to make it similar to @paragraphindent.
\parseargdef\exampleindent{%
\def\temp{#1}%
\ifx\temp\asisword
\else
\ifx\temp\noneword
\lispnarrowing = 0pt
\else
\lispnarrowing = #1em
\fi
\fi
}
% @firstparagraphindent WORD
% If WORD is `none', then suppress indentation of the first paragraph
% after a section heading. If WORD is `insert', then do indent at such
% paragraphs.
%
% The paragraph indentation is suppressed or not by calling
% \suppressfirstparagraphindent, which the sectioning commands do.
% We switch the definition of this back and forth according to WORD.
% By default, we suppress indentation.
%
\def\suppressfirstparagraphindent{\dosuppressfirstparagraphindent}
\def\insertword{insert}
%
\parseargdef\firstparagraphindent{%
\def\temp{#1}%
\ifx\temp\noneword
\let\suppressfirstparagraphindent = \dosuppressfirstparagraphindent
\else\ifx\temp\insertword
\let\suppressfirstparagraphindent = \relax
\else
\errhelp = \EMsimple
\errmessage{Unknown @firstparagraphindent option `\temp'}%
\fi\fi
}
% Here is how we actually suppress indentation. Redefine \everypar to
% \kern backwards by \parindent, and then reset itself to empty.
%
% We also make \indent itself not actually do anything until the next
% paragraph.
%
\gdef\dosuppressfirstparagraphindent{%
\gdef\indent{%
\restorefirstparagraphindent
\indent
}%
\gdef\noindent{%
\restorefirstparagraphindent
\noindent
}%
\global\everypar = {%
\kern -\parindent
\restorefirstparagraphindent
}%
}
\gdef\restorefirstparagraphindent{%
\global \let \indent = \ptexindent
\global \let \noindent = \ptexnoindent
\global \everypar = {}%
}
% @asis just yields its argument. Used with @table, for example.
%
\def\asis#1{#1}
% @math outputs its argument in math mode.
%
% One complication: _ usually means subscripts, but it could also mean
% an actual _ character, as in @math{@var{some_variable} + 1}. So make
% _ active, and distinguish by seeing if the current family is \slfam,
% which is what @var uses.
{
\catcode`\_ = \active
\gdef\mathunderscore{%
\catcode`\_=\active
\def_{\ifnum\fam=\slfam \_\else\sb\fi}%
}
}
% Another complication: we want \\ (and @\) to output a \ character.
% FYI, plain.tex uses \\ as a temporary control sequence (why?), but
% this is not advertised and we don't care. Texinfo does not
% otherwise define @\.
%
% The \mathchar is class=0=ordinary, family=7=ttfam, position=5C=\.
\def\mathbackslash{\ifnum\fam=\ttfam \mathchar"075C \else\backslash \fi}
%
\def\math{%
\tex
\mathunderscore
\let\\ = \mathbackslash
\mathactive
$\finishmath
}
\def\finishmath#1{#1$\endgroup} % Close the group opened by \tex.
% Some active characters (such as <) are spaced differently in math.
% We have to reset their definitions in case the @math was an argument
% to a command which sets the catcodes (such as @item or @section).
%
{
\catcode`^ = \active
\catcode`< = \active
\catcode`> = \active
\catcode`+ = \active
\gdef\mathactive{%
\let^ = \ptexhat
\let< = \ptexless
\let> = \ptexgtr
\let+ = \ptexplus
}
}
% @bullet and @minus need the same treatment as @math, just above.
\def\bullet{$\ptexbullet$}
\def\minus{$-$}
% @dots{} outputs an ellipsis using the current font.
% We do .5em per period so that it has the same spacing in the cm
% typewriter fonts as three actual period characters; on the other hand,
% in other typewriter fonts three periods are wider than 1.5em. So do
% whichever is larger.
%
\def\dots{%
\leavevmode
\setbox0=\hbox{...}% get width of three periods
\ifdim\wd0 > 1.5em
\dimen0 = \wd0
\else
\dimen0 = 1.5em
\fi
\hbox to \dimen0{%
\hskip 0pt plus.25fil
.\hskip 0pt plus1fil
.\hskip 0pt plus1fil
.\hskip 0pt plus.5fil
}%
}
% @enddots{} is an end-of-sentence ellipsis.
%
\def\enddots{%
\dots
\spacefactor=\endofsentencespacefactor
}
% @comma{} is so commas can be inserted into text without messing up
% Texinfo's parsing.
%
\let\comma = ,
% @refill is a no-op.
\let\refill=\relax
% If working on a large document in chapters, it is convenient to
% be able to disable indexing, cross-referencing, and contents, for test runs.
% This is done with @novalidate (before @setfilename).
%
\newif\iflinks \linkstrue % by default we want the aux files.
\let\novalidate = \linksfalse
% @setfilename is done at the beginning of every texinfo file.
% So open here the files we need to have open while reading the input.
% This makes it possible to make a .fmt file for texinfo.
\def\setfilename{%
\fixbackslash % Turn off hack to swallow `\input texinfo'.
\iflinks
\tryauxfile
% Open the new aux file. TeX will close it automatically at exit.
\immediate\openout\auxfile=\jobname.aux
\fi % \openindices needs to do some work in any case.
\openindices
\let\setfilename=\comment % Ignore extra @setfilename cmds.
%
% If texinfo.cnf is present on the system, read it.
% Useful for site-wide @afourpaper, etc.
\openin 1 texinfo.cnf
\ifeof 1 \else \input texinfo.cnf \fi
\closein 1
%
\comment % Ignore the actual filename.
}
% Called from \setfilename.
%
\def\openindices{%
\newindex{cp}%
\newcodeindex{fn}%
\newcodeindex{vr}%
\newcodeindex{tp}%
\newcodeindex{ky}%
\newcodeindex{pg}%
}
% @bye.
\outer\def\bye{\pagealignmacro\tracingstats=1\ptexend}
\message{pdf,}
% adobe `portable' document format
\newcount\tempnum
\newcount\lnkcount
\newtoks\filename
\newcount\filenamelength
\newcount\pgn
\newtoks\toksA
\newtoks\toksB
\newtoks\toksC
\newtoks\toksD
\newbox\boxA
\newcount\countA
\newif\ifpdf
\newif\ifpdfmakepagedest
% when pdftex is run in dvi mode, \pdfoutput is defined (so \pdfoutput=1
% can be set). So we test for \relax and 0 as well as \undefined,
% borrowed from ifpdf.sty.
\ifx\pdfoutput\undefined
\else
\ifx\pdfoutput\relax
\else
\ifcase\pdfoutput
\else
\pdftrue
\fi
\fi
\fi
% PDF uses PostScript string constants for the names of xref targets,
% for display in the outlines, and in other places. Thus, we have to
% double any backslashes. Otherwise, a name like "\node" will be
% interpreted as a newline (\n), followed by o, d, e. Not good.
% http://www.ntg.nl/pipermail/ntg-pdftex/2004-July/000654.html
% (and related messages, the final outcome is that it is up to the TeX
% user to double the backslashes and otherwise make the string valid, so
% that's what we do).
% double active backslashes.
%
{\catcode`\@=0 \catcode`\\=\active
@gdef@activebackslashdouble{%
@catcode`@\=@active
@let\=@doublebackslash}
}
% To handle parens, we must adopt a different approach, since parens are
% not active characters. hyperref.dtx (which has the same problem as
% us) handles it with this amazing macro to replace tokens, with minor
% changes for Texinfo. It is included here under the GPL by permission
% from the author, Heiko Oberdiek.
%
% #1 is the tokens to replace.
% #2 is the replacement.
% #3 is the control sequence with the string.
%
\def\HyPsdSubst#1#2#3{%
\def\HyPsdReplace##1#1##2\END{%
##1%
\ifx\\##2\\%
\else
#2%
\HyReturnAfterFi{%
\HyPsdReplace##2\END
}%
\fi
}%
\xdef#3{\expandafter\HyPsdReplace#3#1\END}%
}
\long\def\HyReturnAfterFi#1\fi{\fi#1}
% #1 is a control sequence in which to do the replacements.
\def\backslashparens#1{%
\xdef#1{#1}% redefine it as its expansion; the definition is simply
% \lastnode when called from \setref -> \pdfmkdest.
\HyPsdSubst{(}{\realbackslash(}{#1}%
\HyPsdSubst{)}{\realbackslash)}{#1}%
}
\newhelp\nopdfimagehelp{Texinfo supports .png, .jpg, .jpeg, and .pdf images
with PDF output, and none of those formats could be found. (.eps cannot
be supported due to the design of the PDF format; use regular TeX (DVI
output) for that.)}
\ifpdf
\input pdfcolor
\pdfcatalog{/PageMode /UseOutlines}
%
% #1 is image name, #2 width (might be empty/whitespace), #3 height (ditto).
\def\dopdfimage#1#2#3{%
\def\imagewidth{#2}\setbox0 = \hbox{\ignorespaces #2}%
\def\imageheight{#3}\setbox2 = \hbox{\ignorespaces #3}%
%
% pdftex (and the PDF format) support .png, .jpg, .pdf (among
% others). Let's try in that order.
\let\pdfimgext=\empty
\begingroup
\openin 1 #1.png \ifeof 1
\openin 1 #1.jpg \ifeof 1
\openin 1 #1.jpeg \ifeof 1
\openin 1 #1.JPG \ifeof 1
\openin 1 #1.pdf \ifeof 1
\errhelp = \nopdfimagehelp
\errmessage{Could not find image file #1 for pdf}%
\else \gdef\pdfimgext{pdf}%
\fi
\else \gdef\pdfimgext{JPG}%
\fi
\else \gdef\pdfimgext{jpeg}%
\fi
\else \gdef\pdfimgext{jpg}%
\fi
\else \gdef\pdfimgext{png}%
\fi
\closein 1
\endgroup
%
% without \immediate, pdftex seg faults when the same image is
% included twice. (Version 3.14159-pre-1.0-unofficial-20010704.)
\ifnum\pdftexversion < 14
\immediate\pdfimage
\else
\immediate\pdfximage
\fi
\ifdim \wd0 >0pt width \imagewidth \fi
\ifdim \wd2 >0pt height \imageheight \fi
\ifnum\pdftexversion<13
#1.\pdfimgext
\else
{#1.\pdfimgext}%
\fi
\ifnum\pdftexversion < 14 \else
\pdfrefximage \pdflastximage
\fi}
%
\def\pdfmkdest#1{{%
% We have to set dummies so commands such as @code, and characters
% such as \, aren't expanded when present in a section title.
\indexnofonts
\turnoffactive
\activebackslashdouble
\makevalueexpandable
\def\pdfdestname{#1}%
\backslashparens\pdfdestname
\safewhatsit{\pdfdest name{\pdfdestname} xyz}%
}}
%
% used to mark target names; must be expandable.
\def\pdfmkpgn#1{#1}
%
% by default, use a color that is dark enough to print on paper as
% nearly black, but still distinguishable for online viewing.
% (Defined in pdfcolor.tex.)
\let\urlcolor = \BrickRed
\let\linkcolor = \BrickRed
\def\endlink{\Black\pdfendlink}
%
% Adding outlines to PDF; macros for calculating structure of outlines
% come from Petr Olsak
\def\expnumber#1{\expandafter\ifx\csname#1\endcsname\relax 0%
\else \csname#1\endcsname \fi}
\def\advancenumber#1{\tempnum=\expnumber{#1}\relax
\advance\tempnum by 1
\expandafter\xdef\csname#1\endcsname{\the\tempnum}}
%
% #1 is the section text, which is what will be displayed in the
% outline by the pdf viewer. #2 is the pdf expression for the number
% of subentries (or empty, for subsubsections). #3 is the node text,
% which might be empty if this toc entry had no corresponding node.
% #4 is the page number
%
\def\dopdfoutline#1#2#3#4{%
% Generate a link to the node text if that exists; else, use the
% page number. We could generate a destination for the section
% text in the case where a section has no node, but it doesn't
% seem worth the trouble, since most documents are normally structured.
\def\pdfoutlinedest{#3}%
\ifx\pdfoutlinedest\empty
\def\pdfoutlinedest{#4}%
\else
% Doubled backslashes in the name.
{\activebackslashdouble \xdef\pdfoutlinedest{#3}%
\backslashparens\pdfoutlinedest}%
\fi
%
% Also double the backslashes in the display string.
{\activebackslashdouble \xdef\pdfoutlinetext{#1}%
\backslashparens\pdfoutlinetext}%
%
\pdfoutline goto name{\pdfmkpgn{\pdfoutlinedest}}#2{\pdfoutlinetext}%
}
%
\def\pdfmakeoutlines{%
\begingroup
% Thanh's hack / proper braces in bookmarks
\edef\mylbrace{\iftrue \string{\else}\fi}\let\{=\mylbrace
\edef\myrbrace{\iffalse{\else\string}\fi}\let\}=\myrbrace
%
% Read toc silently, to get counts of subentries for \pdfoutline.
\def\numchapentry##1##2##3##4{%
\def\thischapnum{##2}%
\def\thissecnum{0}%
\def\thissubsecnum{0}%
}%
\def\numsecentry##1##2##3##4{%
\advancenumber{chap\thischapnum}%
\def\thissecnum{##2}%
\def\thissubsecnum{0}%
}%
\def\numsubsecentry##1##2##3##4{%
\advancenumber{sec\thissecnum}%
\def\thissubsecnum{##2}%
}%
\def\numsubsubsecentry##1##2##3##4{%
\advancenumber{subsec\thissubsecnum}%
}%
\def\thischapnum{0}%
\def\thissecnum{0}%
\def\thissubsecnum{0}%
%
% use \def rather than \let here because we redefine \chapentry et
% al. a second time, below.
\def\appentry{\numchapentry}%
\def\appsecentry{\numsecentry}%
\def\appsubsecentry{\numsubsecentry}%
\def\appsubsubsecentry{\numsubsubsecentry}%
\def\unnchapentry{\numchapentry}%
\def\unnsecentry{\numsecentry}%
\def\unnsubsecentry{\numsubsecentry}%
\def\unnsubsubsecentry{\numsubsubsecentry}%
\readdatafile{toc}%
%
% Read toc second time, this time actually producing the outlines.
% The `-' means take the \expnumber as the absolute number of
% subentries, which we calculated on our first read of the .toc above.
%
% We use the node names as the destinations.
\def\numchapentry##1##2##3##4{%
\dopdfoutline{##1}{count-\expnumber{chap##2}}{##3}{##4}}%
\def\numsecentry##1##2##3##4{%
\dopdfoutline{##1}{count-\expnumber{sec##2}}{##3}{##4}}%
\def\numsubsecentry##1##2##3##4{%
\dopdfoutline{##1}{count-\expnumber{subsec##2}}{##3}{##4}}%
\def\numsubsubsecentry##1##2##3##4{% count is always zero
\dopdfoutline{##1}{}{##3}{##4}}%
%
% PDF outlines are displayed using system fonts, instead of
% document fonts. Therefore we cannot use special characters,
% since the encoding is unknown. For example, the eogonek from
% Latin 2 (0xea) gets translated to a | character. Info from
% Staszek Wawrykiewicz, 19 Jan 2004 04:09:24 +0100.
%
% xx to do this right, we have to translate 8-bit characters to
% their "best" equivalent, based on the @documentencoding. Right
% now, I guess we'll just let the pdf reader have its way.
\indexnofonts
\setupdatafile
\catcode`\\=\active \otherbackslash
\input \jobname.toc
\endgroup
}
%
\def\skipspaces#1{\def\PP{#1}\def\D{|}%
\ifx\PP\D\let\nextsp\relax
\else\let\nextsp\skipspaces
\ifx\p\space\else\addtokens{\filename}{\PP}%
\advance\filenamelength by 1
\fi
\fi
\nextsp}
\def\getfilename#1{\filenamelength=0\expandafter\skipspaces#1|\relax}
\ifnum\pdftexversion < 14
\let \startlink \pdfannotlink
\else
\let \startlink \pdfstartlink
\fi
% make a live url in pdf output.
\def\pdfurl#1{%
\begingroup
% it seems we really need yet another set of dummies; have not
% tried to figure out what each command should do in the context
% of @url. for now, just make @/ a no-op, that's the only one
% people have actually reported a problem with.
%
\normalturnoffactive
\def\@{@}%
\let\/=\empty
\makevalueexpandable
\leavevmode\urlcolor
\startlink attr{/Border [0 0 0]}%
user{/Subtype /Link /A << /S /URI /URI (#1) >>}%
\endgroup}
\def\pdfgettoks#1.{\setbox\boxA=\hbox{\toksA={#1.}\toksB={}\maketoks}}
\def\addtokens#1#2{\edef\addtoks{\noexpand#1={\the#1#2}}\addtoks}
\def\adn#1{\addtokens{\toksC}{#1}\global\countA=1\let\next=\maketoks}
\def\poptoks#1#2|ENDTOKS|{\let\first=#1\toksD={#1}\toksA={#2}}
\def\maketoks{%
\expandafter\poptoks\the\toksA|ENDTOKS|\relax
\ifx\first0\adn0
\else\ifx\first1\adn1 \else\ifx\first2\adn2 \else\ifx\first3\adn3
\else\ifx\first4\adn4 \else\ifx\first5\adn5 \else\ifx\first6\adn6
\else\ifx\first7\adn7 \else\ifx\first8\adn8 \else\ifx\first9\adn9
\else
\ifnum0=\countA\else\makelink\fi
\ifx\first.\let\next=\done\else
\let\next=\maketoks
\addtokens{\toksB}{\the\toksD}
\ifx\first,\addtokens{\toksB}{\space}\fi
\fi
\fi\fi\fi\fi\fi\fi\fi\fi\fi\fi
\next}
\def\makelink{\addtokens{\toksB}%
{\noexpand\pdflink{\the\toksC}}\toksC={}\global\countA=0}
\def\pdflink#1{%
\startlink attr{/Border [0 0 0]} goto name{\pdfmkpgn{#1}}
\linkcolor #1\endlink}
\def\done{\edef\st{\global\noexpand\toksA={\the\toksB}}\st}
\else
\let\pdfmkdest = \gobble
\let\pdfurl = \gobble
\let\endlink = \relax
\let\linkcolor = \relax
\let\pdfmakeoutlines = \relax
\fi % \ifx\pdfoutput
\message{fonts,}
% Change the current font style to #1, remembering it in \curfontstyle.
% For now, we do not accumulate font styles: @b{@i{foo}} prints foo in
% italics, not bold italics.
%
\def\setfontstyle#1{%
\def\curfontstyle{#1}% not as a control sequence, because we are \edef'd.
\csname ten#1\endcsname % change the current font
}
% Select #1 fonts with the current style.
%
\def\selectfonts#1{\csname #1fonts\endcsname \csname\curfontstyle\endcsname}
\def\rm{\fam=0 \setfontstyle{rm}}
\def\it{\fam=\itfam \setfontstyle{it}}
\def\sl{\fam=\slfam \setfontstyle{sl}}
\def\bf{\fam=\bffam \setfontstyle{bf}}\def\bfstylename{bf}
\def\tt{\fam=\ttfam \setfontstyle{tt}}
% Texinfo sort of supports the sans serif font style, which plain TeX does not.
% So we set up a \sf.
\newfam\sffam
\def\sf{\fam=\sffam \setfontstyle{sf}}
\let\li = \sf % Sometimes we call it \li, not \sf.
% We don't need math for this font style.
\def\ttsl{\setfontstyle{ttsl}}
% Default leading.
\newdimen\textleading \textleading = 13.2pt
% Set the baselineskip to #1, and the lineskip and strut size
% correspondingly. There is no deep meaning behind these magic numbers
% used as factors; they just match (closely enough) what Knuth defined.
%
\def\lineskipfactor{.08333}
\def\strutheightpercent{.70833}
\def\strutdepthpercent {.29167}
%
\def\setleading#1{%
\normalbaselineskip = #1\relax
\normallineskip = \lineskipfactor\normalbaselineskip
\normalbaselines
\setbox\strutbox =\hbox{%
\vrule width0pt height\strutheightpercent\baselineskip
depth \strutdepthpercent \baselineskip
}%
}
%
% PDF CMaps. See also LaTeX's t1.cmap.
%
% \cmapOT1
\ifpdf
\begingroup
\catcode`\^^M=\active \def^^M{^^J}% Output line endings as the ^^J char.
\catcode`\%=12 \immediate\pdfobj stream {%!PS-Adobe-3.0 Resource-CMap
%%DocumentNeededResources: ProcSet (CIDInit)
%%IncludeResource: ProcSet (CIDInit)
%%BeginResource: CMap (TeX-OT1-0)
%%Title: (TeX-OT1-0 TeX OT1 0)
%%Version: 1.000
%%EndComments
/CIDInit /ProcSet findresource begin
12 dict begin
begincmap
/CIDSystemInfo
<< /Registry (TeX)
/Ordering (OT1)
/Supplement 0
>> def
/CMapName /TeX-OT1-0 def
/CMapType 2 def
1 begincodespacerange
<00> <7F>
endcodespacerange
8 beginbfrange
<00> <01> <0393>
<09> <0A> <03A8>
<23> <26> <0023>
<28> <3B> <0028>
<3F> <5B> <003F>
<5D> <5E> <005D>
<61> <7A> <0061>
<7B> <7C> <2013>
endbfrange
40 beginbfchar
<02> <0398>
<03> <039B>
<04> <039E>
<05> <03A0>
<06> <03A3>
<07> <03D2>
<08> <03A6>
<0B> <00660066>
<0C> <00660069>
<0D> <0066006C>
<0E> <006600660069>
<0F> <00660066006C>
<10> <0131>
<11> <0237>
<12> <0060>
<13> <00B4>
<14> <02C7>
<15> <02D8>
<16> <00AF>
<17> <02DA>
<18> <00B8>
<19> <00DF>
<1A> <00E6>
<1B> <0153>
<1C> <00F8>
<1D> <00C6>
<1E> <0152>
<1F> <00D8>
<21> <0021>
<22> <201D>
<27> <2019>
<3C> <00A1>
<3D> <003D>
<3E> <00BF>
<5C> <201C>
<5F> <02D9>
<60> <2018>
<7D> <02DD>
<7E> <007E>
<7F> <00A8>
endbfchar
endcmap
CMapName currentdict /CMap defineresource pop
end
end
%%EndResource
%%EOF
}\endgroup
\expandafter\edef\csname cmapOT1\endcsname#1{%
\pdffontattr#1{/ToUnicode \the\pdflastobj\space 0 R}%
}%
%
% \cmapOT1IT
\begingroup
\catcode`\^^M=\active \def^^M{^^J}% Output line endings as the ^^J char.
\catcode`\%=12 \immediate\pdfobj stream {%!PS-Adobe-3.0 Resource-CMap
%%DocumentNeededResources: ProcSet (CIDInit)
%%IncludeResource: ProcSet (CIDInit)
%%BeginResource: CMap (TeX-OT1IT-0)
%%Title: (TeX-OT1IT-0 TeX OT1IT 0)
%%Version: 1.000
%%EndComments
/CIDInit /ProcSet findresource begin
12 dict begin
begincmap
/CIDSystemInfo
<< /Registry (TeX)
/Ordering (OT1IT)
/Supplement 0
>> def
/CMapName /TeX-OT1IT-0 def
/CMapType 2 def
1 begincodespacerange
<00> <7F>
endcodespacerange
8 beginbfrange
<00> <01> <0393>
<09> <0A> <03A8>
<25> <26> <0025>
<28> <3B> <0028>
<3F> <5B> <003F>
<5D> <5E> <005D>
<61> <7A> <0061>
<7B> <7C> <2013>
endbfrange
42 beginbfchar
<02> <0398>
<03> <039B>
<04> <039E>
<05> <03A0>
<06> <03A3>
<07> <03D2>
<08> <03A6>
<0B> <00660066>
<0C> <00660069>
<0D> <0066006C>
<0E> <006600660069>
<0F> <00660066006C>
<10> <0131>
<11> <0237>
<12> <0060>
<13> <00B4>
<14> <02C7>
<15> <02D8>
<16> <00AF>
<17> <02DA>
<18> <00B8>
<19> <00DF>
<1A> <00E6>
<1B> <0153>
<1C> <00F8>
<1D> <00C6>
<1E> <0152>
<1F> <00D8>
<21> <0021>
<22> <201D>
<23> <0023>
<24> <00A3>
<27> <2019>
<3C> <00A1>
<3D> <003D>
<3E> <00BF>
<5C> <201C>
<5F> <02D9>
<60> <2018>
<7D> <02DD>
<7E> <007E>
<7F> <00A8>
endbfchar
endcmap
CMapName currentdict /CMap defineresource pop
end
end
%%EndResource
%%EOF
}\endgroup
\expandafter\edef\csname cmapOT1IT\endcsname#1{%
\pdffontattr#1{/ToUnicode \the\pdflastobj\space 0 R}%
}%
%
% \cmapOT1TT
\begingroup
\catcode`\^^M=\active \def^^M{^^J}% Output line endings as the ^^J char.
\catcode`\%=12 \immediate\pdfobj stream {%!PS-Adobe-3.0 Resource-CMap
%%DocumentNeededResources: ProcSet (CIDInit)
%%IncludeResource: ProcSet (CIDInit)
%%BeginResource: CMap (TeX-OT1TT-0)
%%Title: (TeX-OT1TT-0 TeX OT1TT 0)
%%Version: 1.000
%%EndComments
/CIDInit /ProcSet findresource begin
12 dict begin
begincmap
/CIDSystemInfo
<< /Registry (TeX)
/Ordering (OT1TT)
/Supplement 0
>> def
/CMapName /TeX-OT1TT-0 def
/CMapType 2 def
1 begincodespacerange
<00> <7F>
endcodespacerange
5 beginbfrange
<00> <01> <0393>
<09> <0A> <03A8>
<21> <26> <0021>
<28> <5F> <0028>
<61> <7E> <0061>
endbfrange
32 beginbfchar
<02> <0398>
<03> <039B>
<04> <039E>
<05> <03A0>
<06> <03A3>
<07> <03D2>
<08> <03A6>
<0B> <2191>
<0C> <2193>
<0D> <0027>
<0E> <00A1>
<0F> <00BF>
<10> <0131>
<11> <0237>
<12> <0060>
<13> <00B4>
<14> <02C7>
<15> <02D8>
<16> <00AF>
<17> <02DA>
<18> <00B8>
<19> <00DF>
<1A> <00E6>
<1B> <0153>
<1C> <00F8>
<1D> <00C6>
<1E> <0152>
<1F> <00D8>
<20> <2423>
<27> <2019>
<60> <2018>
<7F> <00A8>
endbfchar
endcmap
CMapName currentdict /CMap defineresource pop
end
end
%%EndResource
%%EOF
}\endgroup
\expandafter\edef\csname cmapOT1TT\endcsname#1{%
\pdffontattr#1{/ToUnicode \the\pdflastobj\space 0 R}%
}%
\else
\expandafter\let\csname cmapOT1\endcsname\gobble
\expandafter\let\csname cmapOT1IT\endcsname\gobble
\expandafter\let\csname cmapOT1TT\endcsname\gobble
\fi
% Set the font macro #1 to the font named #2, adding on the
% specified font prefix (normally `cm').
% #3 is the font's design size, #4 is a scale factor, #5 is the CMap
% encoding (currently only OT1, OT1IT and OT1TT are allowed, pass
% empty to omit).
\def\setfont#1#2#3#4#5{%
\font#1=\fontprefix#2#3 scaled #4
\csname cmap#5\endcsname#1%
}
% This is what gets called when #5 of \setfont is empty.
\let\cmap\gobble
% Use cm as the default font prefix.
% To specify the font prefix, you must define \fontprefix
% before you read in texinfo.tex.
\ifx\fontprefix\undefined
\def\fontprefix{cm}
\fi
% Support font families that don't use the same naming scheme as CM.
\def\rmshape{r}
\def\rmbshape{bx} %where the normal face is bold
\def\bfshape{b}
\def\bxshape{bx}
\def\ttshape{tt}
\def\ttbshape{tt}
\def\ttslshape{sltt}
\def\itshape{ti}
\def\itbshape{bxti}
\def\slshape{sl}
\def\slbshape{bxsl}
\def\sfshape{ss}
\def\sfbshape{ss}
\def\scshape{csc}
\def\scbshape{csc}
% Definitions for a main text size of 11pt. This is the default in
% Texinfo.
%
\def\definetextfontsizexi{%
% Text fonts (11.2pt, magstep1).
\def\textnominalsize{11pt}
\edef\mainmagstep{\magstephalf}
\setfont\textrm\rmshape{10}{\mainmagstep}{OT1}
\setfont\texttt\ttshape{10}{\mainmagstep}{OT1TT}
\setfont\textbf\bfshape{10}{\mainmagstep}{OT1}
\setfont\textit\itshape{10}{\mainmagstep}{OT1IT}
\setfont\textsl\slshape{10}{\mainmagstep}{OT1}
\setfont\textsf\sfshape{10}{\mainmagstep}{OT1}
\setfont\textsc\scshape{10}{\mainmagstep}{OT1}
\setfont\textttsl\ttslshape{10}{\mainmagstep}{OT1TT}
\font\texti=cmmi10 scaled \mainmagstep
\font\textsy=cmsy10 scaled \mainmagstep
% A few fonts for @defun names and args.
\setfont\defbf\bfshape{10}{\magstep1}{OT1}
\setfont\deftt\ttshape{10}{\magstep1}{OT1TT}
\setfont\defttsl\ttslshape{10}{\magstep1}{OT1TT}
\def\df{\let\tentt=\deftt \let\tenbf = \defbf \let\tenttsl=\defttsl \bf}
% Fonts for indices, footnotes, small examples (9pt).
\def\smallnominalsize{9pt}
\setfont\smallrm\rmshape{9}{1000}{OT1}
\setfont\smalltt\ttshape{9}{1000}{OT1TT}
\setfont\smallbf\bfshape{10}{900}{OT1}
\setfont\smallit\itshape{9}{1000}{OT1IT}
\setfont\smallsl\slshape{9}{1000}{OT1}
\setfont\smallsf\sfshape{9}{1000}{OT1}
\setfont\smallsc\scshape{10}{900}{OT1}
\setfont\smallttsl\ttslshape{10}{900}{OT1TT}
\font\smalli=cmmi9
\font\smallsy=cmsy9
% Fonts for small examples (8pt).
\def\smallernominalsize{8pt}
\setfont\smallerrm\rmshape{8}{1000}{OT1}
\setfont\smallertt\ttshape{8}{1000}{OT1TT}
\setfont\smallerbf\bfshape{10}{800}{OT1}
\setfont\smallerit\itshape{8}{1000}{OT1IT}
\setfont\smallersl\slshape{8}{1000}{OT1}
\setfont\smallersf\sfshape{8}{1000}{OT1}
\setfont\smallersc\scshape{10}{800}{OT1}
\setfont\smallerttsl\ttslshape{10}{800}{OT1TT}
\font\smalleri=cmmi8
\font\smallersy=cmsy8
% Fonts for title page (20.4pt):
\def\titlenominalsize{20pt}
\setfont\titlerm\rmbshape{12}{\magstep3}{OT1}
\setfont\titleit\itbshape{10}{\magstep4}{OT1IT}
\setfont\titlesl\slbshape{10}{\magstep4}{OT1}
\setfont\titlett\ttbshape{12}{\magstep3}{OT1TT}
\setfont\titlettsl\ttslshape{10}{\magstep4}{OT1TT}
\setfont\titlesf\sfbshape{17}{\magstep1}{OT1}
\let\titlebf=\titlerm
\setfont\titlesc\scbshape{10}{\magstep4}{OT1}
\font\titlei=cmmi12 scaled \magstep3
\font\titlesy=cmsy10 scaled \magstep4
\def\authorrm{\secrm}
\def\authortt{\sectt}
% Chapter (and unnumbered) fonts (17.28pt).
\def\chapnominalsize{17pt}
\setfont\chaprm\rmbshape{12}{\magstep2}{OT1}
\setfont\chapit\itbshape{10}{\magstep3}{OT1IT}
\setfont\chapsl\slbshape{10}{\magstep3}{OT1}
\setfont\chaptt\ttbshape{12}{\magstep2}{OT1TT}
\setfont\chapttsl\ttslshape{10}{\magstep3}{OT1TT}
\setfont\chapsf\sfbshape{17}{1000}{OT1}
\let\chapbf=\chaprm
\setfont\chapsc\scbshape{10}{\magstep3}{OT1}
\font\chapi=cmmi12 scaled \magstep2
\font\chapsy=cmsy10 scaled \magstep3
% Section fonts (14.4pt).
\def\secnominalsize{14pt}
\setfont\secrm\rmbshape{12}{\magstep1}{OT1}
\setfont\secit\itbshape{10}{\magstep2}{OT1IT}
\setfont\secsl\slbshape{10}{\magstep2}{OT1}
\setfont\sectt\ttbshape{12}{\magstep1}{OT1TT}
\setfont\secttsl\ttslshape{10}{\magstep2}{OT1TT}
\setfont\secsf\sfbshape{12}{\magstep1}{OT1}
\let\secbf\secrm
\setfont\secsc\scbshape{10}{\magstep2}{OT1}
\font\seci=cmmi12 scaled \magstep1
\font\secsy=cmsy10 scaled \magstep2
% Subsection fonts (13.15pt).
\def\ssecnominalsize{13pt}
\setfont\ssecrm\rmbshape{12}{\magstephalf}{OT1}
\setfont\ssecit\itbshape{10}{1315}{OT1IT}
\setfont\ssecsl\slbshape{10}{1315}{OT1}
\setfont\ssectt\ttbshape{12}{\magstephalf}{OT1TT}
\setfont\ssecttsl\ttslshape{10}{1315}{OT1TT}
\setfont\ssecsf\sfbshape{12}{\magstephalf}{OT1}
\let\ssecbf\ssecrm
\setfont\ssecsc\scbshape{10}{1315}{OT1}
\font\sseci=cmmi12 scaled \magstephalf
\font\ssecsy=cmsy10 scaled 1315
% Reduced fonts for @acro in text (10pt).
\def\reducednominalsize{10pt}
\setfont\reducedrm\rmshape{10}{1000}{OT1}
\setfont\reducedtt\ttshape{10}{1000}{OT1TT}
\setfont\reducedbf\bfshape{10}{1000}{OT1}
\setfont\reducedit\itshape{10}{1000}{OT1IT}
\setfont\reducedsl\slshape{10}{1000}{OT1}
\setfont\reducedsf\sfshape{10}{1000}{OT1}
\setfont\reducedsc\scshape{10}{1000}{OT1}
\setfont\reducedttsl\ttslshape{10}{1000}{OT1TT}
\font\reducedi=cmmi10
\font\reducedsy=cmsy10
% reset the current fonts
\textfonts
\rm
} % end of 11pt text font size definitions
% Definitions to make the main text be 10pt Computer Modern, with
% section, chapter, etc., sizes following suit. This is for the GNU
% Press printing of the Emacs 22 manual. Maybe other manuals in the
% future. Used with @smallbook, which sets the leading to 12pt.
%
\def\definetextfontsizex{%
% Text fonts (10pt).
\def\textnominalsize{10pt}
\edef\mainmagstep{1000}
\setfont\textrm\rmshape{10}{\mainmagstep}{OT1}
\setfont\texttt\ttshape{10}{\mainmagstep}{OT1TT}
\setfont\textbf\bfshape{10}{\mainmagstep}{OT1}
\setfont\textit\itshape{10}{\mainmagstep}{OT1IT}
\setfont\textsl\slshape{10}{\mainmagstep}{OT1}
\setfont\textsf\sfshape{10}{\mainmagstep}{OT1}
\setfont\textsc\scshape{10}{\mainmagstep}{OT1}
\setfont\textttsl\ttslshape{10}{\mainmagstep}{OT1TT}
\font\texti=cmmi10 scaled \mainmagstep
\font\textsy=cmsy10 scaled \mainmagstep
% A few fonts for @defun names and args.
\setfont\defbf\bfshape{10}{\magstephalf}{OT1}
\setfont\deftt\ttshape{10}{\magstephalf}{OT1TT}
\setfont\defttsl\ttslshape{10}{\magstephalf}{OT1TT}
\def\df{\let\tentt=\deftt \let\tenbf = \defbf \let\tenttsl=\defttsl \bf}
% Fonts for indices, footnotes, small examples (9pt).
\def\smallnominalsize{9pt}
\setfont\smallrm\rmshape{9}{1000}{OT1}
\setfont\smalltt\ttshape{9}{1000}{OT1TT}
\setfont\smallbf\bfshape{10}{900}{OT1}
\setfont\smallit\itshape{9}{1000}{OT1IT}
\setfont\smallsl\slshape{9}{1000}{OT1}
\setfont\smallsf\sfshape{9}{1000}{OT1}
\setfont\smallsc\scshape{10}{900}{OT1}
\setfont\smallttsl\ttslshape{10}{900}{OT1TT}
\font\smalli=cmmi9
\font\smallsy=cmsy9
% Fonts for small examples (8pt).
\def\smallernominalsize{8pt}
\setfont\smallerrm\rmshape{8}{1000}{OT1}
\setfont\smallertt\ttshape{8}{1000}{OT1TT}
\setfont\smallerbf\bfshape{10}{800}{OT1}
\setfont\smallerit\itshape{8}{1000}{OT1IT}
\setfont\smallersl\slshape{8}{1000}{OT1}
\setfont\smallersf\sfshape{8}{1000}{OT1}
\setfont\smallersc\scshape{10}{800}{OT1}
\setfont\smallerttsl\ttslshape{10}{800}{OT1TT}
\font\smalleri=cmmi8
\font\smallersy=cmsy8
% Fonts for title page (20.4pt):
\def\titlenominalsize{20pt}
\setfont\titlerm\rmbshape{12}{\magstep3}{OT1}
\setfont\titleit\itbshape{10}{\magstep4}{OT1IT}
\setfont\titlesl\slbshape{10}{\magstep4}{OT1}
\setfont\titlett\ttbshape{12}{\magstep3}{OT1TT}
\setfont\titlettsl\ttslshape{10}{\magstep4}{OT1TT}
\setfont\titlesf\sfbshape{17}{\magstep1}{OT1}
\let\titlebf=\titlerm
\setfont\titlesc\scbshape{10}{\magstep4}{OT1}
\font\titlei=cmmi12 scaled \magstep3
\font\titlesy=cmsy10 scaled \magstep4
\def\authorrm{\secrm}
\def\authortt{\sectt}
% Chapter fonts (14.4pt).
\def\chapnominalsize{14pt}
\setfont\chaprm\rmbshape{12}{\magstep1}{OT1}
\setfont\chapit\itbshape{10}{\magstep2}{OT1IT}
\setfont\chapsl\slbshape{10}{\magstep2}{OT1}
\setfont\chaptt\ttbshape{12}{\magstep1}{OT1TT}
\setfont\chapttsl\ttslshape{10}{\magstep2}{OT1TT}
\setfont\chapsf\sfbshape{12}{\magstep1}{OT1}
\let\chapbf\chaprm
\setfont\chapsc\scbshape{10}{\magstep2}{OT1}
\font\chapi=cmmi12 scaled \magstep1
\font\chapsy=cmsy10 scaled \magstep2
% Section fonts (12pt).
\def\secnominalsize{12pt}
\setfont\secrm\rmbshape{12}{1000}{OT1}
\setfont\secit\itbshape{10}{\magstep1}{OT1IT}
\setfont\secsl\slbshape{10}{\magstep1}{OT1}
\setfont\sectt\ttbshape{12}{1000}{OT1TT}
\setfont\secttsl\ttslshape{10}{\magstep1}{OT1TT}
\setfont\secsf\sfbshape{12}{1000}{OT1}
\let\secbf\secrm
\setfont\secsc\scbshape{10}{\magstep1}{OT1}
\font\seci=cmmi12
\font\secsy=cmsy10 scaled \magstep1
% Subsection fonts (10pt).
\def\ssecnominalsize{10pt}
\setfont\ssecrm\rmbshape{10}{1000}{OT1}
\setfont\ssecit\itbshape{10}{1000}{OT1IT}
\setfont\ssecsl\slbshape{10}{1000}{OT1}
\setfont\ssectt\ttbshape{10}{1000}{OT1TT}
\setfont\ssecttsl\ttslshape{10}{1000}{OT1TT}
\setfont\ssecsf\sfbshape{10}{1000}{OT1}
\let\ssecbf\ssecrm
\setfont\ssecsc\scbshape{10}{1000}{OT1}
\font\sseci=cmmi10
\font\ssecsy=cmsy10
% Reduced fonts for @acro in text (9pt).
\def\reducednominalsize{9pt}
\setfont\reducedrm\rmshape{9}{1000}{OT1}
\setfont\reducedtt\ttshape{9}{1000}{OT1TT}
\setfont\reducedbf\bfshape{10}{900}{OT1}
\setfont\reducedit\itshape{9}{1000}{OT1IT}
\setfont\reducedsl\slshape{9}{1000}{OT1}
\setfont\reducedsf\sfshape{9}{1000}{OT1}
\setfont\reducedsc\scshape{10}{900}{OT1}
\setfont\reducedttsl\ttslshape{10}{900}{OT1TT}
\font\reducedi=cmmi9
\font\reducedsy=cmsy9
% reduce space between paragraphs
\divide\parskip by 2
% reset the current fonts
\textfonts
\rm
} % end of 10pt text font size definitions
% We provide the user-level command
% @fonttextsize 10
% (or 11) to redefine the text font size. pt is assumed.
%
\def\xword{10}
\def\xiword{11}
%
\parseargdef\fonttextsize{%
\def\textsizearg{#1}%
\wlog{doing @fonttextsize \textsizearg}%
%
% Set \globaldefs so that documents can use this inside @tex, since
% makeinfo 4.8 does not support it, but we need it nonetheless.
%
\begingroup \globaldefs=1
\ifx\textsizearg\xword \definetextfontsizex
\else \ifx\textsizearg\xiword \definetextfontsizexi
\else
\errhelp=\EMsimple
\errmessage{@fonttextsize only supports `10' or `11', not `\textsizearg'}
\fi\fi
\endgroup
}
% In order for the font changes to affect most math symbols and letters,
% we have to define the \textfont of the standard families. Since
% texinfo doesn't allow for producing subscripts and superscripts except
% in the main text, we don't bother to reset \scriptfont and
% \scriptscriptfont (which would also require loading a lot more fonts).
%
\def\resetmathfonts{%
\textfont0=\tenrm \textfont1=\teni \textfont2=\tensy
\textfont\itfam=\tenit \textfont\slfam=\tensl \textfont\bffam=\tenbf
\textfont\ttfam=\tentt \textfont\sffam=\tensf
}
% The font-changing commands redefine the meanings of \tenSTYLE, instead
% of just \STYLE. We do this because \STYLE needs to also set the
% current \fam for math mode. Our \STYLE (e.g., \rm) commands hardwire
% \tenSTYLE to set the current font.
%
% Each font-changing command also sets the names \lsize (one size lower)
% and \lllsize (three sizes lower). These relative commands are used in
% the LaTeX logo and acronyms.
%
% This all needs generalizing, badly.
%
\def\textfonts{%
\let\tenrm=\textrm \let\tenit=\textit \let\tensl=\textsl
\let\tenbf=\textbf \let\tentt=\texttt \let\smallcaps=\textsc
\let\tensf=\textsf \let\teni=\texti \let\tensy=\textsy
\let\tenttsl=\textttsl
\def\curfontsize{text}%
\def\lsize{reduced}\def\lllsize{smaller}%
\resetmathfonts \setleading{\textleading}}
\def\titlefonts{%
\let\tenrm=\titlerm \let\tenit=\titleit \let\tensl=\titlesl
\let\tenbf=\titlebf \let\tentt=\titlett \let\smallcaps=\titlesc
\let\tensf=\titlesf \let\teni=\titlei \let\tensy=\titlesy
\let\tenttsl=\titlettsl
\def\curfontsize{title}%
\def\lsize{chap}\def\lllsize{subsec}%
\resetmathfonts \setleading{25pt}}
\def\titlefont#1{{\titlefonts\rm #1}}
\def\chapfonts{%
\let\tenrm=\chaprm \let\tenit=\chapit \let\tensl=\chapsl
\let\tenbf=\chapbf \let\tentt=\chaptt \let\smallcaps=\chapsc
\let\tensf=\chapsf \let\teni=\chapi \let\tensy=\chapsy
\let\tenttsl=\chapttsl
\def\curfontsize{chap}%
\def\lsize{sec}\def\lllsize{text}%
\resetmathfonts \setleading{19pt}}
\def\secfonts{%
\let\tenrm=\secrm \let\tenit=\secit \let\tensl=\secsl
\let\tenbf=\secbf \let\tentt=\sectt \let\smallcaps=\secsc
\let\tensf=\secsf \let\teni=\seci \let\tensy=\secsy
\let\tenttsl=\secttsl
\def\curfontsize{sec}%
\def\lsize{subsec}\def\lllsize{reduced}%
\resetmathfonts \setleading{16pt}}
\def\subsecfonts{%
\let\tenrm=\ssecrm \let\tenit=\ssecit \let\tensl=\ssecsl
\let\tenbf=\ssecbf \let\tentt=\ssectt \let\smallcaps=\ssecsc
\let\tensf=\ssecsf \let\teni=\sseci \let\tensy=\ssecsy
\let\tenttsl=\ssecttsl
\def\curfontsize{ssec}%
\def\lsize{text}\def\lllsize{small}%
\resetmathfonts \setleading{15pt}}
\let\subsubsecfonts = \subsecfonts
\def\reducedfonts{%
\let\tenrm=\reducedrm \let\tenit=\reducedit \let\tensl=\reducedsl
\let\tenbf=\reducedbf \let\tentt=\reducedtt \let\reducedcaps=\reducedsc
\let\tensf=\reducedsf \let\teni=\reducedi \let\tensy=\reducedsy
\let\tenttsl=\reducedttsl
\def\curfontsize{reduced}%
\def\lsize{small}\def\lllsize{smaller}%
\resetmathfonts \setleading{10.5pt}}
\def\smallfonts{%
\let\tenrm=\smallrm \let\tenit=\smallit \let\tensl=\smallsl
\let\tenbf=\smallbf \let\tentt=\smalltt \let\smallcaps=\smallsc
\let\tensf=\smallsf \let\teni=\smalli \let\tensy=\smallsy
\let\tenttsl=\smallttsl
\def\curfontsize{small}%
\def\lsize{smaller}\def\lllsize{smaller}%
\resetmathfonts \setleading{10.5pt}}
\def\smallerfonts{%
\let\tenrm=\smallerrm \let\tenit=\smallerit \let\tensl=\smallersl
\let\tenbf=\smallerbf \let\tentt=\smallertt \let\smallcaps=\smallersc
\let\tensf=\smallersf \let\teni=\smalleri \let\tensy=\smallersy
\let\tenttsl=\smallerttsl
\def\curfontsize{smaller}%
\def\lsize{smaller}\def\lllsize{smaller}%
\resetmathfonts \setleading{9.5pt}}
% Set the fonts to use with the @small... environments.
\let\smallexamplefonts = \smallfonts
% About \smallexamplefonts. If we use \smallfonts (9pt), @smallexample
% can fit this many characters:
% 8.5x11=86 smallbook=72 a4=90 a5=69
% If we use \scriptfonts (8pt), then we can fit this many characters:
% 8.5x11=90+ smallbook=80 a4=90+ a5=77
% For me, subjectively, the few extra characters that fit aren't worth
% the additional smallness of 8pt. So I'm making the default 9pt.
%
% By the way, for comparison, here's what fits with @example (10pt):
% 8.5x11=71 smallbook=60 a4=75 a5=58
%
% I wish the USA used A4 paper.
% --karl, 24jan03.
% Set up the default fonts, so we can use them for creating boxes.
%
\definetextfontsizexi
% Define these so they can be easily changed for other fonts.
\def\angleleft{$\langle$}
\def\angleright{$\rangle$}
% Count depth in font-changes, for error checks
\newcount\fontdepth \fontdepth=0
% Fonts for short table of contents.
\setfont\shortcontrm\rmshape{12}{1000}{OT1}
\setfont\shortcontbf\bfshape{10}{\magstep1}{OT1} % no cmb12
\setfont\shortcontsl\slshape{12}{1000}{OT1}
\setfont\shortconttt\ttshape{12}{1000}{OT1TT}
%% Add scribe-like font environments, plus @l for inline lisp (usually sans
%% serif) and @ii for TeX italic
% \smartitalic{ARG} outputs arg in italics, followed by an italic correction
% unless the following character is such as not to need one.
\def\smartitalicx{\ifx\next,\else\ifx\next-\else\ifx\next.\else
\ptexslash\fi\fi\fi}
\def\smartslanted#1{{\ifusingtt\ttsl\sl #1}\futurelet\next\smartitalicx}
\def\smartitalic#1{{\ifusingtt\ttsl\it #1}\futurelet\next\smartitalicx}
% like \smartslanted except unconditionally uses \ttsl.
% @var is set to this for defun arguments.
\def\ttslanted#1{{\ttsl #1}\futurelet\next\smartitalicx}
% like \smartslanted except unconditionally use \sl. We never want
% ttsl for book titles, do we?
\def\cite#1{{\sl #1}\futurelet\next\smartitalicx}
\let\i=\smartitalic
\let\slanted=\smartslanted
\let\var=\smartslanted
\let\dfn=\smartslanted
\let\emph=\smartitalic
% @b, explicit bold.
\def\b#1{{\bf #1}}
\let\strong=\b
% @sansserif, explicit sans.
\def\sansserif#1{{\sf #1}}
% We can't just use \exhyphenpenalty, because that only has effect at
% the end of a paragraph. Restore normal hyphenation at the end of the
% group within which \nohyphenation is presumably called.
%
\def\nohyphenation{\hyphenchar\font = -1 \aftergroup\restorehyphenation}
\def\restorehyphenation{\hyphenchar\font = `- }
% Set sfcode to normal for the chars that usually have another value.
% Can't use plain's \frenchspacing because it uses the `\x notation, and
% sometimes \x has an active definition that messes things up.
%
\catcode`@=11
\def\plainfrenchspacing{%
\sfcode\dotChar =\@m \sfcode\questChar=\@m \sfcode\exclamChar=\@m
\sfcode\colonChar=\@m \sfcode\semiChar =\@m \sfcode\commaChar =\@m
\def\endofsentencespacefactor{1000}% for @. and friends
}
\def\plainnonfrenchspacing{%
\sfcode`\.3000\sfcode`\?3000\sfcode`\!3000
\sfcode`\:2000\sfcode`\;1500\sfcode`\,1250
\def\endofsentencespacefactor{3000}% for @. and friends
}
\catcode`@=\other
\def\endofsentencespacefactor{3000}% default
\def\t#1{%
{\tt \rawbackslash \plainfrenchspacing #1}%
\null
}
\def\samp#1{`\tclose{#1}'\null}
\setfont\keyrm\rmshape{8}{1000}{OT1}
\font\keysy=cmsy9
\def\key#1{{\keyrm\textfont2=\keysy \leavevmode\hbox{%
\raise0.4pt\hbox{\angleleft}\kern-.08em\vtop{%
\vbox{\hrule\kern-0.4pt
\hbox{\raise0.4pt\hbox{\vphantom{\angleleft}}#1}}%
\kern-0.4pt\hrule}%
\kern-.06em\raise0.4pt\hbox{\angleright}}}}
\def\key #1{{\nohyphenation \uppercase{#1}}\null}
% The old definition, with no lozenge:
%\def\key #1{{\ttsl \nohyphenation \uppercase{#1}}\null}
\def\ctrl #1{{\tt \rawbackslash \hat}#1}
% @file, @option are the same as @samp.
\let\file=\samp
\let\option=\samp
% @code is a modification of @t,
% which makes spaces the same size as normal in the surrounding text.
\def\tclose#1{%
{%
% Change normal interword space to be same as for the current font.
\spaceskip = \fontdimen2\font
%
% Switch to typewriter.
\tt
%
% But `\ ' produces the large typewriter interword space.
\def\ {{\spaceskip = 0pt{} }}%
%
% Turn off hyphenation.
\nohyphenation
%
\rawbackslash
\plainfrenchspacing
#1%
}%
\null
}
% We *must* turn on hyphenation at `-' and `_' in @code.
% Otherwise, it is too hard to avoid overfull hboxes
% in the Emacs manual, the Library manual, etc.
% Unfortunately, TeX uses one parameter (\hyphenchar) to control
% both hyphenation at - and hyphenation within words.
% We must therefore turn them both off (\tclose does that)
% and arrange explicitly to hyphenate at a dash.
% -- rms.
{
\catcode`\-=\active \catcode`\_=\active
\catcode`\'=\active \catcode`\`=\active
%
\global\def\code{\begingroup
\catcode\rquoteChar=\active \catcode\lquoteChar=\active
\let'\codequoteright \let`\codequoteleft
%
\catcode\dashChar=\active \catcode\underChar=\active
\ifallowcodebreaks
\let-\codedash
\let_\codeunder
\else
\let-\realdash
\let_\realunder
\fi
\codex
}
}
\def\realdash{-}
\def\codedash{-\discretionary{}{}{}}
\def\codeunder{%
% this is all so @math{@code{var_name}+1} can work. In math mode, _
% is "active" (mathcode"8000) and \normalunderscore (or \char95, etc.)
% will therefore expand the active definition of _, which is us
% (inside @code that is), therefore an endless loop.
\ifusingtt{\ifmmode
\mathchar"075F % class 0=ordinary, family 7=ttfam, pos 0x5F=_.
\else\normalunderscore \fi
\discretionary{}{}{}}%
{\_}%
}
\def\codex #1{\tclose{#1}\endgroup}
% An additional complication: the above will allow breaks after, e.g.,
% each of the four underscores in __typeof__. This is undesirable in
% some manuals, especially if they don't have long identifiers in
% general. @allowcodebreaks provides a way to control this.
%
\newif\ifallowcodebreaks \allowcodebreakstrue
\def\keywordtrue{true}
\def\keywordfalse{false}
\parseargdef\allowcodebreaks{%
\def\txiarg{#1}%
\ifx\txiarg\keywordtrue
\allowcodebreakstrue
\else\ifx\txiarg\keywordfalse
\allowcodebreaksfalse
\else
\errhelp = \EMsimple
\errmessage{Unknown @allowcodebreaks option `\txiarg'}%
\fi\fi
}
% @kbd is like @code, except that if the argument is just one @key command,
% then @kbd has no effect.
% @kbdinputstyle -- arg is `distinct' (@kbd uses slanted tty font always),
% `example' (@kbd uses ttsl only inside of @example and friends),
% or `code' (@kbd uses normal tty font always).
\parseargdef\kbdinputstyle{%
\def\txiarg{#1}%
\ifx\txiarg\worddistinct
\gdef\kbdexamplefont{\ttsl}\gdef\kbdfont{\ttsl}%
\else\ifx\txiarg\wordexample
\gdef\kbdexamplefont{\ttsl}\gdef\kbdfont{\tt}%
\else\ifx\txiarg\wordcode
\gdef\kbdexamplefont{\tt}\gdef\kbdfont{\tt}%
\else
\errhelp = \EMsimple
\errmessage{Unknown @kbdinputstyle option `\txiarg'}%
\fi\fi\fi
}
\def\worddistinct{distinct}
\def\wordexample{example}
\def\wordcode{code}
% Default is `distinct.'
\kbdinputstyle distinct
\def\xkey{\key}
\def\kbdfoo#1#2#3\par{\def\one{#1}\def\three{#3}\def\threex{??}%
\ifx\one\xkey\ifx\threex\three \key{#2}%
\else{\tclose{\kbdfont\look}}\fi
\else{\tclose{\kbdfont\look}}\fi}
% For @indicateurl, @env, @command quotes seem unnecessary, so use \code.
\let\indicateurl=\code
\let\env=\code
\let\command=\code
% @uref (abbreviation for `urlref') takes an optional (comma-separated)
% second argument specifying the text to display and an optional third
% arg as text to display instead of (rather than in addition to) the url
% itself. First (mandatory) arg is the url. Perhaps eventually put in
% a hypertex \special here.
%
\def\uref#1{\douref #1,,,\finish}
\def\douref#1,#2,#3,#4\finish{\begingroup
\unsepspaces
\pdfurl{#1}%
\setbox0 = \hbox{\ignorespaces #3}%
\ifdim\wd0 > 0pt
\unhbox0 % third arg given, show only that
\else
\setbox0 = \hbox{\ignorespaces #2}%
\ifdim\wd0 > 0pt
\ifpdf
\unhbox0 % PDF: 2nd arg given, show only it
\else
\unhbox0\ (\code{#1})% DVI: 2nd arg given, show both it and url
\fi
\else
\code{#1}% only url given, so show it
\fi
\fi
\endlink
\endgroup}
% @url synonym for @uref, since that's how everyone uses it.
%
\let\url=\uref
% rms does not like angle brackets --karl, 17may97.
% So now @email is just like @uref, unless we are pdf.
%
%\def\email#1{\angleleft{\tt #1}\angleright}
\ifpdf
\def\email#1{\doemail#1,,\finish}
\def\doemail#1,#2,#3\finish{\begingroup
\unsepspaces
\pdfurl{mailto:#1}%
\setbox0 = \hbox{\ignorespaces #2}%
\ifdim\wd0>0pt\unhbox0\else\code{#1}\fi
\endlink
\endgroup}
\else
\let\email=\uref
\fi
% Check if we are currently using a typewriter font. Since all the
% Computer Modern typewriter fonts have zero interword stretch (and
% shrink), and it is reasonable to expect all typewriter fonts to have
% this property, we can check that font parameter.
%
\def\ifmonospace{\ifdim\fontdimen3\font=0pt }
% Typeset a dimension, e.g., `in' or `pt'. The only reason for the
% argument is to make the input look right: @dmn{pt} instead of @dmn{}pt.
%
\def\dmn#1{\thinspace #1}
\def\kbd#1{\def\look{#1}\expandafter\kbdfoo\look??\par}
% @l was never documented to mean ``switch to the Lisp font'',
% and it is not used as such in any manual I can find. We need it for
% Polish suppressed-l. --karl, 22sep96.
%\def\l#1{{\li #1}\null}
% Explicit font changes: @r, @sc, undocumented @ii.
\def\r#1{{\rm #1}} % roman font
\def\sc#1{{\smallcaps#1}} % smallcaps font
\def\ii#1{{\it #1}} % italic font
% @acronym for "FBI", "NATO", and the like.
% We print this one point size smaller, since it's intended for
% all-uppercase.
%
\def\acronym#1{\doacronym #1,,\finish}
\def\doacronym#1,#2,#3\finish{%
{\selectfonts\lsize #1}%
\def\temp{#2}%
\ifx\temp\empty \else
\space ({\unsepspaces \ignorespaces \temp \unskip})%
\fi
}
% @abbr for "Comput. J." and the like.
% No font change, but don't do end-of-sentence spacing.
%
\def\abbr#1{\doabbr #1,,\finish}
\def\doabbr#1,#2,#3\finish{%
{\plainfrenchspacing #1}%
\def\temp{#2}%
\ifx\temp\empty \else
\space ({\unsepspaces \ignorespaces \temp \unskip})%
\fi
}
% @pounds{} is a sterling sign, which Knuth put in the CM italic font.
%
\def\pounds{{\it\$}}
% @euro{} comes from a separate font, depending on the current style.
% We use the free feym* fonts from the eurosym package by Henrik
% Theiling, which support regular, slanted, bold and bold slanted (and
% "outlined" (blackboard board, sort of) versions, which we don't need).
% It is available from http://www.ctan.org/tex-archive/fonts/eurosym.
%
% Although only regular is the truly official Euro symbol, we ignore
% that. The Euro is designed to be slightly taller than the regular
% font height.
%
% feymr - regular
% feymo - slanted
% feybr - bold
% feybo - bold slanted
%
% There is no good (free) typewriter version, to my knowledge.
% A feymr10 euro is ~7.3pt wide, while a normal cmtt10 char is ~5.25pt wide.
% Hmm.
%
% Also doesn't work in math. Do we need to do math with euro symbols?
% Hope not.
%
%
\def\euro{{\eurofont e}}
\def\eurofont{%
% We set the font at each command, rather than predefining it in
% \textfonts and the other font-switching commands, so that
% installations which never need the symbol don't have to have the
% font installed.
%
% There is only one designed size (nominal 10pt), so we always scale
% that to the current nominal size.
%
% By the way, simply using "at 1em" works for cmr10 and the like, but
% does not work for cmbx10 and other extended/shrunken fonts.
%
\def\eurosize{\csname\curfontsize nominalsize\endcsname}%
%
\ifx\curfontstyle\bfstylename
% bold:
\font\thiseurofont = \ifusingit{feybo10}{feybr10} at \eurosize
\else
% regular:
\font\thiseurofont = \ifusingit{feymo10}{feymr10} at \eurosize
\fi
\thiseurofont
}
% @registeredsymbol - R in a circle. The font for the R should really
% be smaller yet, but lllsize is the best we can do for now.
% Adapted from the plain.tex definition of \copyright.
%
\def\registeredsymbol{%
$^{{\ooalign{\hfil\raise.07ex\hbox{\selectfonts\lllsize R}%
\hfil\crcr\Orb}}%
}$%
}
% @textdegree - the normal degrees sign.
%
\def\textdegree{$^\circ$}
% Laurent Siebenmann reports \Orb undefined with:
% Textures 1.7.7 (preloaded format=plain 93.10.14) (68K) 16 APR 2004 02:38
% so we'll define it if necessary.
%
\ifx\Orb\undefined
\def\Orb{\mathhexbox20D}
\fi
\message{page headings,}
\newskip\titlepagetopglue \titlepagetopglue = 1.5in
\newskip\titlepagebottomglue \titlepagebottomglue = 2pc
% First the title page. Must do @settitle before @titlepage.
\newif\ifseenauthor
\newif\iffinishedtitlepage
% Do an implicit @contents or @shortcontents after @end titlepage if the
% user says @setcontentsaftertitlepage or @setshortcontentsaftertitlepage.
%
\newif\ifsetcontentsaftertitlepage
\let\setcontentsaftertitlepage = \setcontentsaftertitlepagetrue
\newif\ifsetshortcontentsaftertitlepage
\let\setshortcontentsaftertitlepage = \setshortcontentsaftertitlepagetrue
\parseargdef\shorttitlepage{\begingroup\hbox{}\vskip 1.5in \chaprm \centerline{#1}%
\endgroup\page\hbox{}\page}
\envdef\titlepage{%
% Open one extra group, as we want to close it in the middle of \Etitlepage.
\begingroup
\parindent=0pt \textfonts
% Leave some space at the very top of the page.
\vglue\titlepagetopglue
% No rule at page bottom unless we print one at the top with @title.
\finishedtitlepagetrue
%
% Most title ``pages'' are actually two pages long, with space
% at the top of the second. We don't want the ragged left on the second.
\let\oldpage = \page
\def\page{%
\iffinishedtitlepage\else
\finishtitlepage
\fi
\let\page = \oldpage
\page
\null
}%
}
\def\Etitlepage{%
\iffinishedtitlepage\else
\finishtitlepage
\fi
% It is important to do the page break before ending the group,
% because the headline and footline are only empty inside the group.
% If we use the new definition of \page, we always get a blank page
% after the title page, which we certainly don't want.
\oldpage
\endgroup
%
% Need this before the \...aftertitlepage checks so that if they are
% in effect the toc pages will come out with page numbers.
\HEADINGSon
%
% If they want short, they certainly want long too.
\ifsetshortcontentsaftertitlepage
\shortcontents
\contents
\global\let\shortcontents = \relax
\global\let\contents = \relax
\fi
%
\ifsetcontentsaftertitlepage
\contents
\global\let\contents = \relax
\global\let\shortcontents = \relax
\fi
}
\def\finishtitlepage{%
\vskip4pt \hrule height 2pt width \hsize
\vskip\titlepagebottomglue
\finishedtitlepagetrue
}
%%% Macros to be used within @titlepage:
\let\subtitlerm=\tenrm
\def\subtitlefont{\subtitlerm \normalbaselineskip = 13pt \normalbaselines}
\def\authorfont{\authorrm \normalbaselineskip = 16pt \normalbaselines
\let\tt=\authortt}
\parseargdef\title{%
\checkenv\titlepage
\leftline{\titlefonts\rm #1}
% print a rule at the page bottom also.
\finishedtitlepagefalse
\vskip4pt \hrule height 4pt width \hsize \vskip4pt
}
\parseargdef\subtitle{%
\checkenv\titlepage
{\subtitlefont \rightline{#1}}%
}
% @author should come last, but may come many times.
% It can also be used inside @quotation.
%
\parseargdef\author{%
\def\temp{\quotation}%
\ifx\thisenv\temp
\def\quotationauthor{#1}% printed in \Equotation.
\else
\checkenv\titlepage
\ifseenauthor\else \vskip 0pt plus 1filll \seenauthortrue \fi
{\authorfont \leftline{#1}}%
\fi
}
%%% Set up page headings and footings.
\let\thispage=\folio
\newtoks\evenheadline % headline on even pages
\newtoks\oddheadline % headline on odd pages
\newtoks\evenfootline % footline on even pages
\newtoks\oddfootline % footline on odd pages
% Now make TeX use those variables
\headline={{\textfonts\rm \ifodd\pageno \the\oddheadline
\else \the\evenheadline \fi}}
\footline={{\textfonts\rm \ifodd\pageno \the\oddfootline
\else \the\evenfootline \fi}\HEADINGShook}
\let\HEADINGShook=\relax
% Commands to set those variables.
% For example, this is what @headings on does
% @evenheading @thistitle|@thispage|@thischapter
% @oddheading @thischapter|@thispage|@thistitle
% @evenfooting @thisfile||
% @oddfooting ||@thisfile
\def\evenheading{\parsearg\evenheadingxxx}
\def\evenheadingxxx #1{\evenheadingyyy #1\|\|\|\|\finish}
\def\evenheadingyyy #1\|#2\|#3\|#4\finish{%
\global\evenheadline={\rlap{\centerline{#2}}\line{#1\hfil#3}}}
\def\oddheading{\parsearg\oddheadingxxx}
\def\oddheadingxxx #1{\oddheadingyyy #1\|\|\|\|\finish}
\def\oddheadingyyy #1\|#2\|#3\|#4\finish{%
\global\oddheadline={\rlap{\centerline{#2}}\line{#1\hfil#3}}}
\parseargdef\everyheading{\oddheadingxxx{#1}\evenheadingxxx{#1}}%
\def\evenfooting{\parsearg\evenfootingxxx}
\def\evenfootingxxx #1{\evenfootingyyy #1\|\|\|\|\finish}
\def\evenfootingyyy #1\|#2\|#3\|#4\finish{%
\global\evenfootline={\rlap{\centerline{#2}}\line{#1\hfil#3}}}
\def\oddfooting{\parsearg\oddfootingxxx}
\def\oddfootingxxx #1{\oddfootingyyy #1\|\|\|\|\finish}
\def\oddfootingyyy #1\|#2\|#3\|#4\finish{%
\global\oddfootline = {\rlap{\centerline{#2}}\line{#1\hfil#3}}%
%
% Leave some space for the footline. Hopefully ok to assume
% @evenfooting will not be used by itself.
\global\advance\pageheight by -12pt
\global\advance\vsize by -12pt
}
\parseargdef\everyfooting{\oddfootingxxx{#1}\evenfootingxxx{#1}}
% @headings double turns headings on for double-sided printing.
% @headings single turns headings on for single-sided printing.
% @headings off turns them off.
% @headings on same as @headings double, retained for compatibility.
% @headings after turns on double-sided headings after this page.
% @headings doubleafter turns on double-sided headings after this page.
% @headings singleafter turns on single-sided headings after this page.
% By default, they are off at the start of a document,
% and turned `on' after @end titlepage.
\def\headings #1 {\csname HEADINGS#1\endcsname}
\def\HEADINGSoff{%
\global\evenheadline={\hfil} \global\evenfootline={\hfil}
\global\oddheadline={\hfil} \global\oddfootline={\hfil}}
\HEADINGSoff
% When we turn headings on, set the page number to 1.
% For double-sided printing, put current file name in lower left corner,
% chapter name on inside top of right hand pages, document
% title on inside top of left hand pages, and page numbers on outside top
% edge of all pages.
\def\HEADINGSdouble{%
\global\pageno=1
\global\evenfootline={\hfil}
\global\oddfootline={\hfil}
\global\evenheadline={\line{\folio\hfil\thistitle}}
\global\oddheadline={\line{\thischapter\hfil\folio}}
\global\let\contentsalignmacro = \chapoddpage
}
\let\contentsalignmacro = \chappager
% For single-sided printing, chapter title goes across top left of page,
% page number on top right.
\def\HEADINGSsingle{%
\global\pageno=1
\global\evenfootline={\hfil}
\global\oddfootline={\hfil}
\global\evenheadline={\line{\thischapter\hfil\folio}}
\global\oddheadline={\line{\thischapter\hfil\folio}}
\global\let\contentsalignmacro = \chappager
}
\def\HEADINGSon{\HEADINGSdouble}
\def\HEADINGSafter{\let\HEADINGShook=\HEADINGSdoublex}
\let\HEADINGSdoubleafter=\HEADINGSafter
\def\HEADINGSdoublex{%
\global\evenfootline={\hfil}
\global\oddfootline={\hfil}
\global\evenheadline={\line{\folio\hfil\thistitle}}
\global\oddheadline={\line{\thischapter\hfil\folio}}
\global\let\contentsalignmacro = \chapoddpage
}
\def\HEADINGSsingleafter{\let\HEADINGShook=\HEADINGSsinglex}
\def\HEADINGSsinglex{%
\global\evenfootline={\hfil}
\global\oddfootline={\hfil}
\global\evenheadline={\line{\thischapter\hfil\folio}}
\global\oddheadline={\line{\thischapter\hfil\folio}}
\global\let\contentsalignmacro = \chappager
}
% Subroutines used in generating headings
% This produces Day Month Year style of output.
% Only define if not already defined, in case a txi-??.tex file has set
% up a different format (e.g., txi-cs.tex does this).
\ifx\today\undefined
\def\today{%
\number\day\space
\ifcase\month
\or\putwordMJan\or\putwordMFeb\or\putwordMMar\or\putwordMApr
\or\putwordMMay\or\putwordMJun\or\putwordMJul\or\putwordMAug
\or\putwordMSep\or\putwordMOct\or\putwordMNov\or\putwordMDec
\fi
\space\number\year}
\fi
% @settitle line... specifies the title of the document, for headings.
% It generates no output of its own.
\def\thistitle{\putwordNoTitle}
\def\settitle{\parsearg{\gdef\thistitle}}
\message{tables,}
% Tables -- @table, @ftable, @vtable, @item(x).
% default indentation of table text
\newdimen\tableindent \tableindent=.8in
% default indentation of @itemize and @enumerate text
\newdimen\itemindent \itemindent=.3in
% margin between end of table item and start of table text.
\newdimen\itemmargin \itemmargin=.1in
% used internally for \itemindent minus \itemmargin
\newdimen\itemmax
% Note @table, @ftable, and @vtable define @item, @itemx, etc., with
% these defs.
% They also define \itemindex
% to index the item name in whatever manner is desired (perhaps none).
\newif\ifitemxneedsnegativevskip
\def\itemxpar{\par\ifitemxneedsnegativevskip\nobreak\vskip-\parskip\nobreak\fi}
\def\internalBitem{\smallbreak \parsearg\itemzzz}
\def\internalBitemx{\itemxpar \parsearg\itemzzz}
\def\itemzzz #1{\begingroup %
\advance\hsize by -\rightskip
\advance\hsize by -\tableindent
\setbox0=\hbox{\itemindicate{#1}}%
\itemindex{#1}%
\nobreak % This prevents a break before @itemx.
%
% If the item text does not fit in the space we have, put it on a line
% by itself, and do not allow a page break either before or after that
% line. We do not start a paragraph here because then if the next
% command is, e.g., @kindex, the whatsit would get put into the
% horizontal list on a line by itself, resulting in extra blank space.
\ifdim \wd0>\itemmax
%
% Make this a paragraph so we get the \parskip glue and wrapping,
% but leave it ragged-right.
\begingroup
\advance\leftskip by-\tableindent
\advance\hsize by\tableindent
\advance\rightskip by0pt plus1fil
\leavevmode\unhbox0\par
\endgroup
%
% We're going to be starting a paragraph, but we don't want the
% \parskip glue -- logically it's part of the @item we just started.
\nobreak \vskip-\parskip
%
% Stop a page break at the \parskip glue coming up. However, if
% what follows is an environment such as @example, there will be no
% \parskip glue; then the negative vskip we just inserted would
% cause the example and the item to crash together. So we use this
% bizarre value of 10001 as a signal to \aboveenvbreak to insert
% \parskip glue after all. Section titles are handled this way also.
%
\penalty 10001
\endgroup
\itemxneedsnegativevskipfalse
\else
% The item text fits into the space. Start a paragraph, so that the
% following text (if any) will end up on the same line.
\noindent
% Do this with kerns and \unhbox so that if there is a footnote in
% the item text, it can migrate to the main vertical list and
% eventually be printed.
\nobreak\kern-\tableindent
\dimen0 = \itemmax \advance\dimen0 by \itemmargin \advance\dimen0 by -\wd0
\unhbox0
\nobreak\kern\dimen0
\endgroup
\itemxneedsnegativevskiptrue
\fi
}
\def\item{\errmessage{@item while not in a list environment}}
\def\itemx{\errmessage{@itemx while not in a list environment}}
% @table, @ftable, @vtable.
\envdef\table{%
\let\itemindex\gobble
\tablecheck{table}%
}
\envdef\ftable{%
\def\itemindex ##1{\doind {fn}{\code{##1}}}%
\tablecheck{ftable}%
}
\envdef\vtable{%
\def\itemindex ##1{\doind {vr}{\code{##1}}}%
\tablecheck{vtable}%
}
\def\tablecheck#1{%
\ifnum \the\catcode`\^^M=\active
\endgroup
\errmessage{This command won't work in this context; perhaps the problem is
that we are \inenvironment\thisenv}%
\def\next{\doignore{#1}}%
\else
\let\next\tablex
\fi
\next
}
\def\tablex#1{%
\def\itemindicate{#1}%
\parsearg\tabley
}
\def\tabley#1{%
{%
\makevalueexpandable
\edef\temp{\noexpand\tablez #1\space\space\space}%
\expandafter
}\temp \endtablez
}
\def\tablez #1 #2 #3 #4\endtablez{%
\aboveenvbreak
\ifnum 0#1>0 \advance \leftskip by #1\mil \fi
\ifnum 0#2>0 \tableindent=#2\mil \fi
\ifnum 0#3>0 \advance \rightskip by #3\mil \fi
\itemmax=\tableindent
\advance \itemmax by -\itemmargin
\advance \leftskip by \tableindent
\exdentamount=\tableindent
\parindent = 0pt
\parskip = \smallskipamount
\ifdim \parskip=0pt \parskip=2pt \fi
\let\item = \internalBitem
\let\itemx = \internalBitemx
}
\def\Etable{\endgraf\afterenvbreak}
\let\Eftable\Etable
\let\Evtable\Etable
\let\Eitemize\Etable
\let\Eenumerate\Etable
% This is the counter used by @enumerate, which is really @itemize
\newcount \itemno
\envdef\itemize{\parsearg\doitemize}
\def\doitemize#1{%
\aboveenvbreak
\itemmax=\itemindent
\advance\itemmax by -\itemmargin
\advance\leftskip by \itemindent
\exdentamount=\itemindent
\parindent=0pt
\parskip=\smallskipamount
\ifdim\parskip=0pt \parskip=2pt \fi
\def\itemcontents{#1}%
% @itemize with no arg is equivalent to @itemize @bullet.
\ifx\itemcontents\empty\def\itemcontents{\bullet}\fi
\let\item=\itemizeitem
}
% Definition of @item while inside @itemize and @enumerate.
%
\def\itemizeitem{%
\advance\itemno by 1 % for enumerations
{\let\par=\endgraf \smallbreak}% reasonable place to break
{%
% If the document has an @itemize directly after a section title, a
% \nobreak will be last on the list, and \sectionheading will have
% done a \vskip-\parskip. In that case, we don't want to zero
% parskip, or the item text will crash with the heading. On the
% other hand, when there is normal text preceding the item (as there
% usually is), we do want to zero parskip, or there would be too much
% space. In that case, we won't have a \nobreak before. At least
% that's the theory.
\ifnum\lastpenalty<10000 \parskip=0in \fi
\noindent
\hbox to 0pt{\hss \itemcontents \kern\itemmargin}%
\vadjust{\penalty 1200}}% not good to break after first line of item.
\flushcr
}
% \splitoff TOKENS\endmark defines \first to be the first token in
% TOKENS, and \rest to be the remainder.
%
\def\splitoff#1#2\endmark{\def\first{#1}\def\rest{#2}}%
% Allow an optional argument of an uppercase letter, lowercase letter,
% or number, to specify the first label in the enumerated list. No
% argument is the same as `1'.
%
\envparseargdef\enumerate{\enumeratey #1 \endenumeratey}
\def\enumeratey #1 #2\endenumeratey{%
% If we were given no argument, pretend we were given `1'.
\def\thearg{#1}%
\ifx\thearg\empty \def\thearg{1}\fi
%
% Detect if the argument is a single token. If so, it might be a
% letter. Otherwise, the only valid thing it can be is a number.
% (We will always have one token, because of the test we just made.
% This is a good thing, since \splitoff doesn't work given nothing at
% all -- the first parameter is undelimited.)
\expandafter\splitoff\thearg\endmark
\ifx\rest\empty
% Only one token in the argument. It could still be anything.
% A ``lowercase letter'' is one whose \lccode is nonzero.
% An ``uppercase letter'' is one whose \lccode is both nonzero, and
% not equal to itself.
% Otherwise, we assume it's a number.
%
% We need the \relax at the end of the \ifnum lines to stop TeX from
% continuing to look for a <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:
\obeylines
\catcode`\@ = \other
\catcode`\{ = \other
\catcode`\} = \other
%
% Make sure that spaces turn into tokens that match what \doignoretext wants.
\spaceisspace
%
% Count number of #1's that we've seen.
\doignorecount = 0
%
% Swallow text until we reach the matching `@end #1'.
\dodoignore{#1}%
}
{ \catcode`_=11 % We want to use \_STOP_ which cannot appear in texinfo source.
\obeylines %
%
\gdef\dodoignore#1{%
% #1 contains the command name as a string, e.g., `ifinfo'.
%
% Define a command to find the next `@end #1'.
\long\def\doignoretext##1^^M@end #1{%
\doignoretextyyy##1^^M@#1\_STOP_}%
%
% And this command to find another #1 command, at the beginning of a
% line. (Otherwise, we would consider a line `@c @ifset', for
% example, to count as an @ifset for nesting.)
\long\def\doignoretextyyy##1^^M@#1##2\_STOP_{\doignoreyyy{##2}\_STOP_}%
%
% And now expand that command.
\doignoretext ^^M%
}%
}
\def\doignoreyyy#1{%
\def\temp{#1}%
\ifx\temp\empty % Nothing found.
\let\next\doignoretextzzz
\else % Found a nested condition, ...
\advance\doignorecount by 1
\let\next\doignoretextyyy % ..., look for another.
% If we're here, #1 ends with ^^M\ifinfo (for example).
\fi
\next #1% the token \_STOP_ is present just after this macro.
}
% We have to swallow the remaining "\_STOP_".
%
\def\doignoretextzzz#1{%
\ifnum\doignorecount = 0 % We have just found the outermost @end.
\let\next\enddoignore
\else % Still inside a nested condition.
\advance\doignorecount by -1
\let\next\doignoretext % Look for the next @end.
\fi
\next
}
% Finish off ignored text.
{ \obeylines%
% Ignore anything after the last `@end #1'; this matters in verbatim
% environments, where otherwise the newline after an ignored conditional
% would result in a blank line in the output.
\gdef\enddoignore#1^^M{\endgroup\ignorespaces}%
}
% @set VAR sets the variable VAR to an empty value.
% @set VAR REST-OF-LINE sets VAR to the value REST-OF-LINE.
%
% Since we want to separate VAR from REST-OF-LINE (which might be
% empty), we can't just use \parsearg; we have to insert a space of our
% own to delimit the rest of the line, and then take it out again if we
% didn't need it.
% We rely on the fact that \parsearg sets \catcode`\ =10.
%
\parseargdef\set{\setyyy#1 \endsetyyy}
\def\setyyy#1 #2\endsetyyy{%
{%
\makevalueexpandable
\def\temp{#2}%
\edef\next{\gdef\makecsname{SET#1}}%
\ifx\temp\empty
\next{}%
\else
\setzzz#2\endsetzzz
\fi
}%
}
% Remove the trailing space \setxxx inserted.
\def\setzzz#1 \endsetzzz{\next{#1}}
% @clear VAR clears (i.e., unsets) the variable VAR.
%
\parseargdef\clear{%
{%
\makevalueexpandable
\global\expandafter\let\csname SET#1\endcsname=\relax
}%
}
% @value{foo} gets the text saved in variable foo.
\def\value{\begingroup\makevalueexpandable\valuexxx}
\def\valuexxx#1{\expandablevalue{#1}\endgroup}
{
\catcode`\- = \active \catcode`\_ = \active
%
\gdef\makevalueexpandable{%
\let\value = \expandablevalue
% We don't want these characters active, ...
\catcode`\-=\other \catcode`\_=\other
% ..., but we might end up with active ones in the argument if
% we're called from @code, as @code{@value{foo-bar_}}, though.
% So \let them to their normal equivalents.
\let-\realdash \let_\normalunderscore
}
}
% We have this subroutine so that we can handle at least some @value's
% properly in indexes (we call \makevalueexpandable in \indexdummies).
% The command has to be fully expandable (if the variable is set), since
% the result winds up in the index file. This means that if the
% variable's value contains other Texinfo commands, it's almost certain
% it will fail (although perhaps we could fix that with sufficient work
% to do a one-level expansion on the result, instead of complete).
%
\def\expandablevalue#1{%
\expandafter\ifx\csname SET#1\endcsname\relax
{[No value for ``#1'']}%
\message{Variable `#1', used in @value, is not set.}%
\else
\csname SET#1\endcsname
\fi
}
% @ifset VAR ... @end ifset reads the `...' iff VAR has been defined
% with @set.
%
% To get special treatment of `@end ifset,' call \makeond and the redefine.
%
\makecond{ifset}
\def\ifset{\parsearg{\doifset{\let\next=\ifsetfail}}}
\def\doifset#1#2{%
{%
\makevalueexpandable
\let\next=\empty
\expandafter\ifx\csname SET#2\endcsname\relax
#1% If not set, redefine \next.
\fi
\expandafter
}\next
}
\def\ifsetfail{\doignore{ifset}}
% @ifclear VAR ... @end ifclear reads the `...' iff VAR has never been
% defined with @set, or has been undefined with @clear.
%
% The `\else' inside the `\doifset' parameter is a trick to reuse the
% above code: if the variable is not set, do nothing, if it is set,
% then redefine \next to \ifclearfail.
%
\makecond{ifclear}
\def\ifclear{\parsearg{\doifset{\else \let\next=\ifclearfail}}}
\def\ifclearfail{\doignore{ifclear}}
% @dircategory CATEGORY -- specify a category of the dir file
% which this file should belong to. Ignore this in TeX.
\let\dircategory=\comment
% @defininfoenclose.
\let\definfoenclose=\comment
\message{indexing,}
% Index generation facilities
% Define \newwrite to be identical to plain tex's \newwrite
% except not \outer, so it can be used within macros and \if's.
\edef\newwrite{\makecsname{ptexnewwrite}}
% \newindex {foo} defines an index named foo.
% It automatically defines \fooindex such that
% \fooindex ...rest of line... puts an entry in the index foo.
% It also defines \fooindfile to be the number of the output channel for
% the file that accumulates this index. The file's extension is foo.
% The name of an index should be no more than 2 characters long
% for the sake of vms.
%
\def\newindex#1{%
\iflinks
\expandafter\newwrite \csname#1indfile\endcsname
\openout \csname#1indfile\endcsname \jobname.#1 % Open the file
\fi
\expandafter\xdef\csname#1index\endcsname{% % Define @#1index
\noexpand\doindex{#1}}
}
% @defindex foo == \newindex{foo}
%
\def\defindex{\parsearg\newindex}
% Define @defcodeindex, like @defindex except put all entries in @code.
%
\def\defcodeindex{\parsearg\newcodeindex}
%
\def\newcodeindex#1{%
\iflinks
\expandafter\newwrite \csname#1indfile\endcsname
\openout \csname#1indfile\endcsname \jobname.#1
\fi
\expandafter\xdef\csname#1index\endcsname{%
\noexpand\docodeindex{#1}}%
}
% @synindex foo bar makes index foo feed into index bar.
% Do this instead of @defindex foo if you don't want it as a separate index.
%
% @syncodeindex foo bar similar, but put all entries made for index foo
% inside @code.
%
\def\synindex#1 #2 {\dosynindex\doindex{#1}{#2}}
\def\syncodeindex#1 #2 {\dosynindex\docodeindex{#1}{#2}}
% #1 is \doindex or \docodeindex, #2 the index getting redefined (foo),
% #3 the target index (bar).
\def\dosynindex#1#2#3{%
% Only do \closeout if we haven't already done it, else we'll end up
% closing the target index.
\expandafter \ifx\csname donesynindex#2\endcsname \undefined
% The \closeout helps reduce unnecessary open files; the limit on the
% Acorn RISC OS is a mere 16 files.
\expandafter\closeout\csname#2indfile\endcsname
\expandafter\let\csname\donesynindex#2\endcsname = 1
\fi
% redefine \fooindfile:
\expandafter\let\expandafter\temp\expandafter=\csname#3indfile\endcsname
\expandafter\let\csname#2indfile\endcsname=\temp
% redefine \fooindex:
\expandafter\xdef\csname#2index\endcsname{\noexpand#1{#3}}%
}
% Define \doindex, the driver for all \fooindex macros.
% Argument #1 is generated by the calling \fooindex macro,
% and it is "foo", the name of the index.
% \doindex just uses \parsearg; it calls \doind for the actual work.
% This is because \doind is more useful to call from other macros.
% There is also \dosubind {index}{topic}{subtopic}
% which makes an entry in a two-level index such as the operation index.
\def\doindex#1{\edef\indexname{#1}\parsearg\singleindexer}
\def\singleindexer #1{\doind{\indexname}{#1}}
% like the previous two, but they put @code around the argument.
\def\docodeindex#1{\edef\indexname{#1}\parsearg\singlecodeindexer}
\def\singlecodeindexer #1{\doind{\indexname}{\code{#1}}}
% Take care of Texinfo commands that can appear in an index entry.
% Since there are some commands we want to expand, and others we don't,
% we have to laboriously prevent expansion for those that we don't.
%
\def\indexdummies{%
\escapechar = `\\ % use backslash in output files.
\def\@{@}% change to @@ when we switch to @ as escape char in index files.
\def\ {\realbackslash\space }%
%
% Need these in case \tex is in effect and \{ is a \delimiter again.
% But can't use \lbracecmd and \rbracecmd because texindex assumes
% braces and backslashes are used only as delimiters.
\let\{ = \mylbrace
\let\} = \myrbrace
%
% I don't entirely understand this, but when an index entry is
% generated from a macro call, the \endinput which \scanmacro inserts
% causes processing to be prematurely terminated. This is,
% apparently, because \indexsorttmp is fully expanded, and \endinput
% is an expandable command. The redefinition below makes \endinput
% disappear altogether for that purpose -- although logging shows that
% processing continues to some further point. On the other hand, it
% seems \endinput does not hurt in the printed index arg, since that
% is still getting written without apparent harm.
%
% Sample source (mac-idx3.tex, reported by Graham Percival to
% help-texinfo, 22may06):
% @macro funindex {WORD}
% @findex xyz
% @end macro
% ...
% @funindex commtest
%
% The above is not enough to reproduce the bug, but it gives the flavor.
%
% Sample whatsit resulting:
% .@write3{\entry{xyz}{@folio }{@code {xyz@endinput }}}
%
% So:
\let\endinput = \empty
%
% Do the redefinitions.
\commondummies
}
% For the aux and toc files, @ is the escape character. So we want to
% redefine everything using @ as the escape character (instead of
% \realbackslash, still used for index files). When everything uses @,
% this will be simpler.
%
\def\atdummies{%
\def\@{@@}%
\def\ {@ }%
\let\{ = \lbraceatcmd
\let\} = \rbraceatcmd
%
% Do the redefinitions.
\commondummies
\otherbackslash
}
% Called from \indexdummies and \atdummies.
%
\def\commondummies{%
%
% \definedummyword defines \#1 as \string\#1\space, thus effectively
% preventing its expansion. This is used only for control% words,
% not control letters, because the \space would be incorrect for
% control characters, but is needed to separate the control word
% from whatever follows.
%
% For control letters, we have \definedummyletter, which omits the
% space.
%
% These can be used both for control words that take an argument and
% those that do not. If it is followed by {arg} in the input, then
% that will dutifully get written to the index (or wherever).
%
\def\definedummyword ##1{\def##1{\string##1\space}}%
\def\definedummyletter##1{\def##1{\string##1}}%
\let\definedummyaccent\definedummyletter
%
\commondummiesnofonts
%
\definedummyletter\_%
%
% Non-English letters.
\definedummyword\AA
\definedummyword\AE
\definedummyword\L
\definedummyword\OE
\definedummyword\O
\definedummyword\aa
\definedummyword\ae
\definedummyword\l
\definedummyword\oe
\definedummyword\o
\definedummyword\ss
\definedummyword\exclamdown
\definedummyword\questiondown
\definedummyword\ordf
\definedummyword\ordm
%
% Although these internal commands shouldn't show up, sometimes they do.
\definedummyword\bf
\definedummyword\gtr
\definedummyword\hat
\definedummyword\less
\definedummyword\sf
\definedummyword\sl
\definedummyword\tclose
\definedummyword\tt
%
\definedummyword\LaTeX
\definedummyword\TeX
%
% Assorted special characters.
\definedummyword\bullet
\definedummyword\comma
\definedummyword\copyright
\definedummyword\registeredsymbol
\definedummyword\dots
\definedummyword\enddots
\definedummyword\equiv
\definedummyword\error
\definedummyword\euro
\definedummyword\expansion
\definedummyword\minus
\definedummyword\pounds
\definedummyword\point
\definedummyword\print
\definedummyword\result
\definedummyword\textdegree
%
% We want to disable all macros so that they are not expanded by \write.
\macrolist
%
\normalturnoffactive
%
% Handle some cases of @value -- where it does not contain any
% (non-fully-expandable) commands.
\makevalueexpandable
}
% \commondummiesnofonts: common to \commondummies and \indexnofonts.
%
\def\commondummiesnofonts{%
% Control letters and accents.
\definedummyletter\!%
\definedummyaccent\"%
\definedummyaccent\'%
\definedummyletter\*%
\definedummyaccent\,%
\definedummyletter\.%
\definedummyletter\/%
\definedummyletter\:%
\definedummyaccent\=%
\definedummyletter\?%
\definedummyaccent\^%
\definedummyaccent\`%
\definedummyaccent\~%
\definedummyword\u
\definedummyword\v
\definedummyword\H
\definedummyword\dotaccent
\definedummyword\ringaccent
\definedummyword\tieaccent
\definedummyword\ubaraccent
\definedummyword\udotaccent
\definedummyword\dotless
%
% Texinfo font commands.
\definedummyword\b
\definedummyword\i
\definedummyword\r
\definedummyword\sc
\definedummyword\t
%
% Commands that take arguments.
\definedummyword\acronym
\definedummyword\cite
\definedummyword\code
\definedummyword\command
\definedummyword\dfn
\definedummyword\emph
\definedummyword\env
\definedummyword\file
\definedummyword\kbd
\definedummyword\key
\definedummyword\math
\definedummyword\option
\definedummyword\pxref
\definedummyword\ref
\definedummyword\samp
\definedummyword\strong
\definedummyword\tie
\definedummyword\uref
\definedummyword\url
\definedummyword\var
\definedummyword\verb
\definedummyword\w
\definedummyword\xref
}
% \indexnofonts is used when outputting the strings to sort the index
% by, and when constructing control sequence names. It eliminates all
% control sequences and just writes whatever the best ASCII sort string
% would be for a given command (usually its argument).
%
\def\indexnofonts{%
% Accent commands should become @asis.
\def\definedummyaccent##1{\let##1\asis}%
% We can just ignore other control letters.
\def\definedummyletter##1{\let##1\empty}%
% Hopefully, all control words can become @asis.
\let\definedummyword\definedummyaccent
%
\commondummiesnofonts
%
% Don't no-op \tt, since it isn't a user-level command
% and is used in the definitions of the active chars like <, >, |, etc.
% Likewise with the other plain tex font commands.
%\let\tt=\asis
%
\def\ { }%
\def\@{@}%
% how to handle braces?
\def\_{\normalunderscore}%
%
% Non-English letters.
\def\AA{AA}%
\def\AE{AE}%
\def\L{L}%
\def\OE{OE}%
\def\O{O}%
\def\aa{aa}%
\def\ae{ae}%
\def\l{l}%
\def\oe{oe}%
\def\o{o}%
\def\ss{ss}%
\def\exclamdown{!}%
\def\questiondown{?}%
\def\ordf{a}%
\def\ordm{o}%
%
\def\LaTeX{LaTeX}%
\def\TeX{TeX}%
%
% Assorted special characters.
% (The following {} will end up in the sort string, but that's ok.)
\def\bullet{bullet}%
\def\comma{,}%
\def\copyright{copyright}%
\def\registeredsymbol{R}%
\def\dots{...}%
\def\enddots{...}%
\def\equiv{==}%
\def\error{error}%
\def\euro{euro}%
\def\expansion{==>}%
\def\minus{-}%
\def\pounds{pounds}%
\def\point{.}%
\def\print{-|}%
\def\result{=>}%
\def\textdegree{degrees}%
%
% We need to get rid of all macros, leaving only the arguments (if present).
% Of course this is not nearly correct, but it is the best we can do for now.
% makeinfo does not expand macros in the argument to @deffn, which ends up
% writing an index entry, and texindex isn't prepared for an index sort entry
% that starts with \.
%
% Since macro invocations are followed by braces, we can just redefine them
% to take a single TeX argument. The case of a macro invocation that
% goes to end-of-line is not handled.
%
\macrolist
}
\let\indexbackslash=0 %overridden during \printindex.
\let\SETmarginindex=\relax % put index entries in margin (undocumented)?
% Most index entries go through here, but \dosubind is the general case.
% #1 is the index name, #2 is the entry text.
\def\doind#1#2{\dosubind{#1}{#2}{}}
% Workhorse for all \fooindexes.
% #1 is name of index, #2 is stuff to put there, #3 is subentry --
% empty if called from \doind, as we usually are (the main exception
% is with most defuns, which call us directly).
%
\def\dosubind#1#2#3{%
\iflinks
{%
% Store the main index entry text (including the third arg).
\toks0 = {#2}%
% If third arg is present, precede it with a space.
\def\thirdarg{#3}%
\ifx\thirdarg\empty \else
\toks0 = \expandafter{\the\toks0 \space #3}%
\fi
%
\edef\writeto{\csname#1indfile\endcsname}%
%
\safewhatsit\dosubindwrite
}%
\fi
}
% Write the entry in \toks0 to the index file:
%
\def\dosubindwrite{%
% Put the index entry in the margin if desired.
\ifx\SETmarginindex\relax\else
\insert\margin{\hbox{\vrule height8pt depth3pt width0pt \the\toks0}}%
\fi
%
% Remember, we are within a group.
\indexdummies % Must do this here, since \bf, etc expand at this stage
\def\backslashcurfont{\indexbackslash}% \indexbackslash isn't defined now
% so it will be output as is; and it will print as backslash.
%
% Process the index entry with all font commands turned off, to
% get the string to sort by.
{\indexnofonts
\edef\temp{\the\toks0}% need full expansion
\xdef\indexsorttmp{\temp}%
}%
%
% Set up the complete index entry, with both the sort key and
% the original text, including any font commands. We write
% three arguments to \entry to the .?? file (four in the
% subentry case), texindex reduces to two when writing the .??s
% sorted result.
\edef\temp{%
\write\writeto{%
\string\entry{\indexsorttmp}{\noexpand\folio}{\the\toks0}}%
}%
\temp
}
% Take care of unwanted page breaks/skips around a whatsit:
%
% If a skip is the last thing on the list now, preserve it
% by backing up by \lastskip, doing the \write, then inserting
% the skip again. Otherwise, the whatsit generated by the
% \write or \pdfdest will make \lastskip zero. The result is that
% sequences like this:
% @end defun
% @tindex whatever
% @defun ...
% will have extra space inserted, because the \medbreak in the
% start of the @defun won't see the skip inserted by the @end of
% the previous defun.
%
% But don't do any of this if we're not in vertical mode. We
% don't want to do a \vskip and prematurely end a paragraph.
%
% Avoid page breaks due to these extra skips, too.
%
% But wait, there is a catch there:
% We'll have to check whether \lastskip is zero skip. \ifdim is not
% sufficient for this purpose, as it ignores stretch and shrink parts
% of the skip. The only way seems to be to check the textual
% representation of the skip.
%
% The following is almost like \def\zeroskipmacro{0.0pt} except that
% the ``p'' and ``t'' characters have catcode \other, not 11 (letter).
%
\edef\zeroskipmacro{\expandafter\the\csname z@skip\endcsname}
%
\newskip\whatsitskip
\newcount\whatsitpenalty
%
% ..., ready, GO:
%
\def\safewhatsit#1{%
\ifhmode
#1%
\else
% \lastskip and \lastpenalty cannot both be nonzero simultaneously.
\whatsitskip = \lastskip
\edef\lastskipmacro{\the\lastskip}%
\whatsitpenalty = \lastpenalty
%
% If \lastskip is nonzero, that means the last item was a
% skip. And since a skip is discardable, that means this
% -\skip0 glue we're inserting is preceded by a
% non-discardable item, therefore it is not a potential
% breakpoint, therefore no \nobreak needed.
\ifx\lastskipmacro\zeroskipmacro
\else
\vskip-\whatsitskip
\fi
%
#1%
%
\ifx\lastskipmacro\zeroskipmacro
% If \lastskip was zero, perhaps the last item was a penalty, and
% perhaps it was >=10000, e.g., a \nobreak. In that case, we want
% to re-insert the same penalty (values >10000 are used for various
% signals); since we just inserted a non-discardable item, any
% following glue (such as a \parskip) would be a breakpoint. For example:
%
% @deffn deffn-whatever
% @vindex index-whatever
% Description.
% would allow a break between the index-whatever whatsit
% and the "Description." paragraph.
\ifnum\whatsitpenalty>9999 \penalty\whatsitpenalty \fi
\else
% On the other hand, if we had a nonzero \lastskip,
% this make-up glue would be preceded by a non-discardable item
% (the whatsit from the \write), so we must insert a \nobreak.
\nobreak\vskip\whatsitskip
\fi
\fi
}
% The index entry written in the file actually looks like
% \entry {sortstring}{page}{topic}
% or
% \entry {sortstring}{page}{topic}{subtopic}
% The texindex program reads in these files and writes files
% containing these kinds of lines:
% \initial {c}
% before the first topic whose initial is c
% \entry {topic}{pagelist}
% for a topic that is used without subtopics
% \primary {topic}
% for the beginning of a topic that is used with subtopics
% \secondary {subtopic}{pagelist}
% for each subtopic.
% Define the user-accessible indexing commands
% @findex, @vindex, @kindex, @cindex.
\def\findex {\fnindex}
\def\kindex {\kyindex}
\def\cindex {\cpindex}
\def\vindex {\vrindex}
\def\tindex {\tpindex}
\def\pindex {\pgindex}
\def\cindexsub {\begingroup\obeylines\cindexsub}
{\obeylines %
\gdef\cindexsub "#1" #2^^M{\endgroup %
\dosubind{cp}{#2}{#1}}}
% Define the macros used in formatting output of the sorted index material.
% @printindex causes a particular index (the ??s file) to get printed.
% It does not print any chapter heading (usually an @unnumbered).
%
\parseargdef\printindex{\begingroup
\dobreak \chapheadingskip{10000}%
%
\smallfonts \rm
\tolerance = 9500
\plainfrenchspacing
\everypar = {}% don't want the \kern\-parindent from indentation suppression.
%
% See if the index file exists and is nonempty.
% Change catcode of @ here so that if the index file contains
% \initial {@}
% as its first line, TeX doesn't complain about mismatched braces
% (because it thinks @} is a control sequence).
\catcode`\@ = 11
\openin 1 \jobname.#1s
\ifeof 1
% \enddoublecolumns gets confused if there is no text in the index,
% and it loses the chapter title and the aux file entries for the
% index. The easiest way to prevent this problem is to make sure
% there is some text.
\putwordIndexNonexistent
\else
%
% If the index file exists but is empty, then \openin leaves \ifeof
% false. We have to make TeX try to read something from the file, so
% it can discover if there is anything in it.
\read 1 to \temp
\ifeof 1
\putwordIndexIsEmpty
\else
% Index files are almost Texinfo source, but we use \ as the escape
% character. It would be better to use @, but that's too big a change
% to make right now.
\def\indexbackslash{\backslashcurfont}%
\catcode`\\ = 0
\escapechar = `\\
\begindoublecolumns
\input \jobname.#1s
\enddoublecolumns
\fi
\fi
\closein 1
\endgroup}
% These macros are used by the sorted index file itself.
% Change them to control the appearance of the index.
\def\initial#1{{%
% Some minor font changes for the special characters.
\let\tentt=\sectt \let\tt=\sectt \let\sf=\sectt
%
% Remove any glue we may have, we'll be inserting our own.
\removelastskip
%
% We like breaks before the index initials, so insert a bonus.
\nobreak
\vskip 0pt plus 3\baselineskip
\penalty 0
\vskip 0pt plus -3\baselineskip
%
% Typeset the initial. Making this add up to a whole number of
% baselineskips increases the chance of the dots lining up from column
% to column. It still won't often be perfect, because of the stretch
% we need before each entry, but it's better.
%
% No shrink because it confuses \balancecolumns.
\vskip 1.67\baselineskip plus .5\baselineskip
\leftline{\secbf #1}%
% Do our best not to break after the initial.
\nobreak
\vskip .33\baselineskip plus .1\baselineskip
}}
% \entry typesets a paragraph consisting of the text (#1), dot leaders, and
% then page number (#2) flushed to the right margin. It is used for index
% and table of contents entries. The paragraph is indented by \leftskip.
%
% A straightforward implementation would start like this:
% \def\entry#1#2{...
% But this frozes the catcodes in the argument, and can cause problems to
% @code, which sets - active. This problem was fixed by a kludge---
% ``-'' was active throughout whole index, but this isn't really right.
%
% The right solution is to prevent \entry from swallowing the whole text.
% --kasal, 21nov03
\def\entry{%
\begingroup
%
% Start a new paragraph if necessary, so our assignments below can't
% affect previous text.
\par
%
% Do not fill out the last line with white space.
\parfillskip = 0in
%
% No extra space above this paragraph.
\parskip = 0in
%
% Do not prefer a separate line ending with a hyphen to fewer lines.
\finalhyphendemerits = 0
%
% \hangindent is only relevant when the entry text and page number
% don't both fit on one line. In that case, bob suggests starting the
% dots pretty far over on the line. Unfortunately, a large
% indentation looks wrong when the entry text itself is broken across
% lines. So we use a small indentation and put up with long leaders.
%
% \hangafter is reset to 1 (which is the value we want) at the start
% of each paragraph, so we need not do anything with that.
\hangindent = 2em
%
% When the entry text needs to be broken, just fill out the first line
% with blank space.
\rightskip = 0pt plus1fil
%
% A bit of stretch before each entry for the benefit of balancing
% columns.
\vskip 0pt plus1pt
%
% Swallow the left brace of the text (first parameter):
\afterassignment\doentry
\let\temp =
}
\def\doentry{%
\bgroup % Instead of the swallowed brace.
\noindent
\aftergroup\finishentry
% And now comes the text of the entry.
}
\def\finishentry#1{%
% #1 is the page number.
%
% The following is kludged to not output a line of dots in the index if
% there are no page numbers. The next person who breaks this will be
% cursed by a Unix daemon.
\def\tempa{{\rm }}%
\def\tempb{#1}%
\edef\tempc{\tempa}%
\edef\tempd{\tempb}%
\ifx\tempc\tempd
\ %
\else
%
% If we must, put the page number on a line of its own, and fill out
% this line with blank space. (The \hfil is overwhelmed with the
% fill leaders glue in \indexdotfill if the page number does fit.)
\hfil\penalty50
\null\nobreak\indexdotfill % Have leaders before the page number.
%
% The `\ ' here is removed by the implicit \unskip that TeX does as
% part of (the primitive) \par. Without it, a spurious underfull
% \hbox ensues.
\ifpdf
\pdfgettoks#1.%
\ \the\toksA
\else
\ #1%
\fi
\fi
\par
\endgroup
}
% Like plain.tex's \dotfill, except uses up at least 1 em.
\def\indexdotfill{\cleaders
\hbox{$\mathsurround=0pt \mkern1.5mu.\mkern1.5mu$}\hskip 1em plus 1fill}
\def\primary #1{\line{#1\hfil}}
\newskip\secondaryindent \secondaryindent=0.5cm
\def\secondary#1#2{{%
\parfillskip=0in
\parskip=0in
\hangindent=1in
\hangafter=1
\noindent\hskip\secondaryindent\hbox{#1}\indexdotfill
\ifpdf
\pdfgettoks#2.\ \the\toksA % The page number ends the paragraph.
\else
#2
\fi
\par
}}
% Define two-column mode, which we use to typeset indexes.
% Adapted from the TeXbook, page 416, which is to say,
% the manmac.tex format used to print the TeXbook itself.
\catcode`\@=11
\newbox\partialpage
\newdimen\doublecolumnhsize
\def\begindoublecolumns{\begingroup % ended by \enddoublecolumns
% Grab any single-column material above us.
\output = {%
%
% Here is a possibility not foreseen in manmac: if we accumulate a
% whole lot of material, we might end up calling this \output
% routine twice in a row (see the doublecol-lose test, which is
% essentially a couple of indexes with @setchapternewpage off). In
% that case we just ship out what is in \partialpage with the normal
% output routine. Generally, \partialpage will be empty when this
% runs and this will be a no-op. See the indexspread.tex test case.
\ifvoid\partialpage \else
\onepageout{\pagecontents\partialpage}%
\fi
%
\global\setbox\partialpage = \vbox{%
% Unvbox the main output page.
\unvbox\PAGE
\kern-\topskip \kern\baselineskip
}%
}%
\eject % run that output routine to set \partialpage
%
% Use the double-column output routine for subsequent pages.
\output = {\doublecolumnout}%
%
% Change the page size parameters. We could do this once outside this
% routine, in each of @smallbook, @afourpaper, and the default 8.5x11
% format, but then we repeat the same computation. Repeating a couple
% of assignments once per index is clearly meaningless for the
% execution time, so we may as well do it in one place.
%
% First we halve the line length, less a little for the gutter between
% the columns. We compute the gutter based on the line length, so it
% changes automatically with the paper format. The magic constant
% below is chosen so that the gutter has the same value (well, +-<1pt)
% as it did when we hard-coded it.
%
% We put the result in a separate register, \doublecolumhsize, so we
% can restore it in \pagesofar, after \hsize itself has (potentially)
% been clobbered.
%
\doublecolumnhsize = \hsize
\advance\doublecolumnhsize by -.04154\hsize
\divide\doublecolumnhsize by 2
\hsize = \doublecolumnhsize
%
% Double the \vsize as well. (We don't need a separate register here,
% since nobody clobbers \vsize.)
\vsize = 2\vsize
}
% The double-column output routine for all double-column pages except
% the last.
%
\def\doublecolumnout{%
\splittopskip=\topskip \splitmaxdepth=\maxdepth
% Get the available space for the double columns -- the normal
% (undoubled) page height minus any material left over from the
% previous page.
\dimen@ = \vsize
\divide\dimen@ by 2
\advance\dimen@ by -\ht\partialpage
%
% box0 will be the left-hand column, box2 the right.
\setbox0=\vsplit255 to\dimen@ \setbox2=\vsplit255 to\dimen@
\onepageout\pagesofar
\unvbox255
\penalty\outputpenalty
}
%
% Re-output the contents of the output page -- any previous material,
% followed by the two boxes we just split, in box0 and box2.
\def\pagesofar{%
\unvbox\partialpage
%
\hsize = \doublecolumnhsize
\wd0=\hsize \wd2=\hsize
\hbox to\pagewidth{\box0\hfil\box2}%
}
%
% All done with double columns.
\def\enddoublecolumns{%
% The following penalty ensures that the page builder is exercised
% _before_ we change the output routine. This is necessary in the
% following situation:
%
% The last section of the index consists only of a single entry.
% Before this section, \pagetotal is less than \pagegoal, so no
% break occurs before the last section starts. However, the last
% section, consisting of \initial and the single \entry, does not
% fit on the page and has to be broken off. Without the following
% penalty the page builder will not be exercised until \eject
% below, and by that time we'll already have changed the output
% routine to the \balancecolumns version, so the next-to-last
% double-column page will be processed with \balancecolumns, which
% is wrong: The two columns will go to the main vertical list, with
% the broken-off section in the recent contributions. As soon as
% the output routine finishes, TeX starts reconsidering the page
% break. The two columns and the broken-off section both fit on the
% page, because the two columns now take up only half of the page
% goal. When TeX sees \eject from below which follows the final
% section, it invokes the new output routine that we've set after
% \balancecolumns below; \onepageout will try to fit the two columns
% and the final section into the vbox of \pageheight (see
% \pagebody), causing an overfull box.
%
% Note that glue won't work here, because glue does not exercise the
% page builder, unlike penalties (see The TeXbook, pp. 280-281).
\penalty0
%
\output = {%
% Split the last of the double-column material. Leave it on the
% current page, no automatic page break.
\balancecolumns
%
% If we end up splitting too much material for the current page,
% though, there will be another page break right after this \output
% invocation ends. Having called \balancecolumns once, we do not
% want to call it again. Therefore, reset \output to its normal
% definition right away. (We hope \balancecolumns will never be
% called on to balance too much material, but if it is, this makes
% the output somewhat more palatable.)
\global\output = {\onepageout{\pagecontents\PAGE}}%
}%
\eject
\endgroup % started in \begindoublecolumns
%
% \pagegoal was set to the doubled \vsize above, since we restarted
% the current page. We're now back to normal single-column
% typesetting, so reset \pagegoal to the normal \vsize (after the
% \endgroup where \vsize got restored).
\pagegoal = \vsize
}
%
% Called at the end of the double column material.
\def\balancecolumns{%
\setbox0 = \vbox{\unvbox255}% like \box255 but more efficient, see p.120.
\dimen@ = \ht0
\advance\dimen@ by \topskip
\advance\dimen@ by-\baselineskip
\divide\dimen@ by 2 % target to split to
%debug\message{final 2-column material height=\the\ht0, target=\the\dimen@.}%
\splittopskip = \topskip
% Loop until we get a decent breakpoint.
{%
\vbadness = 10000
\loop
\global\setbox3 = \copy0
\global\setbox1 = \vsplit3 to \dimen@
\ifdim\ht3>\dimen@
\global\advance\dimen@ by 1pt
\repeat
}%
%debug\message{split to \the\dimen@, column heights: \the\ht1, \the\ht3.}%
\setbox0=\vbox to\dimen@{\unvbox1}%
\setbox2=\vbox to\dimen@{\unvbox3}%
%
\pagesofar
}
\catcode`\@ = \other
\message{sectioning,}
% Chapters, sections, etc.
% \unnumberedno is an oxymoron, of course. But we count the unnumbered
% sections so that we can refer to them unambiguously in the pdf
% outlines by their "section number". We avoid collisions with chapter
% numbers by starting them at 10000. (If a document ever has 10000
% chapters, we're in trouble anyway, I'm sure.)
\newcount\unnumberedno \unnumberedno = 10000
\newcount\chapno
\newcount\secno \secno=0
\newcount\subsecno \subsecno=0
\newcount\subsubsecno \subsubsecno=0
% This counter is funny since it counts through charcodes of letters A, B, ...
\newcount\appendixno \appendixno = `\@
%
% \def\appendixletter{\char\the\appendixno}
% We do the following ugly conditional instead of the above simple
% construct for the sake of pdftex, which needs the actual
% letter in the expansion, not just typeset.
%
\def\appendixletter{%
\ifnum\appendixno=`A A%
\else\ifnum\appendixno=`B B%
\else\ifnum\appendixno=`C C%
\else\ifnum\appendixno=`D D%
\else\ifnum\appendixno=`E E%
\else\ifnum\appendixno=`F F%
\else\ifnum\appendixno=`G G%
\else\ifnum\appendixno=`H H%
\else\ifnum\appendixno=`I I%
\else\ifnum\appendixno=`J J%
\else\ifnum\appendixno=`K K%
\else\ifnum\appendixno=`L L%
\else\ifnum\appendixno=`M M%
\else\ifnum\appendixno=`N N%
\else\ifnum\appendixno=`O O%
\else\ifnum\appendixno=`P P%
\else\ifnum\appendixno=`Q Q%
\else\ifnum\appendixno=`R R%
\else\ifnum\appendixno=`S S%
\else\ifnum\appendixno=`T T%
\else\ifnum\appendixno=`U U%
\else\ifnum\appendixno=`V V%
\else\ifnum\appendixno=`W W%
\else\ifnum\appendixno=`X X%
\else\ifnum\appendixno=`Y Y%
\else\ifnum\appendixno=`Z Z%
% The \the is necessary, despite appearances, because \appendixletter is
% expanded while writing the .toc file. \char\appendixno is not
% expandable, thus it is written literally, thus all appendixes come out
% with the same letter (or @) in the toc without it.
\else\char\the\appendixno
\fi\fi\fi\fi\fi\fi\fi\fi\fi\fi\fi\fi\fi
\fi\fi\fi\fi\fi\fi\fi\fi\fi\fi\fi\fi\fi}
% Each @chapter defines this as the name of the chapter.
% page headings and footings can use it. @section does likewise.
% However, they are not reliable, because we don't use marks.
\def\thischapter{}
\def\thissection{}
\newcount\absseclevel % used to calculate proper heading level
\newcount\secbase\secbase=0 % @raisesections/@lowersections modify this count
% @raisesections: treat @section as chapter, @subsection as section, etc.
\def\raisesections{\global\advance\secbase by -1}
\let\up=\raisesections % original BFox name
% @lowersections: treat @chapter as section, @section as subsection, etc.
\def\lowersections{\global\advance\secbase by 1}
\let\down=\lowersections % original BFox name
% we only have subsub.
\chardef\maxseclevel = 3
%
% A numbered section within an unnumbered changes to unnumbered too.
-% To achive this, remember the "biggest" unnum. sec. we are currently in:
+% To achieve 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}%
\gdef\thischapternum{}%
\gdef\thischapter{#1}%
\else\ifx\temptype\Yomitfromtockeyword
\setbox0 = \hbox{}% contents like unnumbered, but no toc entry
\def\toctype{omit}%
\gdef\thischapternum{}%
\gdef\thischapter{}%
\else\ifx\temptype\Yappendixkeyword
\setbox0 = \hbox{\putwordAppendix{} #3\enspace}%
\def\toctype{app}%
\xdef\thischapternum{\appendixletter}%
% We don't substitute the actual chapter name into \thischapter
% because we don't want its macros evaluated now. And we don't
% use \thissection because that changes with each section.
%
\xdef\thischapter{\putwordAppendix{} \appendixletter:
\noexpand\thischaptername}%
\else
\setbox0 = \hbox{#3\enspace}%
\def\toctype{numchap}%
\xdef\thischapternum{\the\chapno}%
\xdef\thischapter{\putwordChapter{} \the\chapno:
\noexpand\thischaptername}%
\fi\fi\fi
%
% Write the toc entry for this chapter. Must come before the
% \donoderef, because we include the current node name in the toc
% entry, and \donoderef resets it to empty.
\writetocentry{\toctype}{#1}{#3}%
%
% For pdftex, we have to write out the node definition (aka, make
% the pdfdest) after any page break, but before the actual text has
% been typeset. If the destination for the pdf outline is after the
% text, then jumping from the outline may wind up with the text not
% being visible, for instance under high magnification.
\donoderef{#2}%
%
% Typeset the actual heading.
\vbox{\hyphenpenalty=10000 \tolerance=5000 \parindent=0pt \raggedright
\hangindent=\wd0 \centerparametersmaybe
\unhbox0 #1\par}%
}%
\nobreak\bigskip % no page break after a chapter title
\nobreak
}
% @centerchap -- centered and unnumbered.
\let\centerparametersmaybe = \relax
\def\centerparameters{%
\advance\rightskip by 3\rightskip
\leftskip = \rightskip
\parfillskip = 0pt
}
% I don't think this chapter style is supported any more, so I'm not
% updating it with the new noderef stuff. We'll see. --karl, 11aug03.
%
\def\setchapterstyle #1 {\csname CHAPF#1\endcsname}
%
\def\unnchfopen #1{%
\chapoddpage {\chapfonts \vbox{\hyphenpenalty=10000\tolerance=5000
\parindent=0pt\raggedright
\rm #1\hfill}}\bigskip \par\nobreak
}
\def\chfopen #1#2{\chapoddpage {\chapfonts
\vbox to 3in{\vfil \hbox to\hsize{\hfil #2} \hbox to\hsize{\hfil #1} \vfil}}%
\par\penalty 5000 %
}
\def\centerchfopen #1{%
\chapoddpage {\chapfonts \vbox{\hyphenpenalty=10000\tolerance=5000
\parindent=0pt
\hfill {\rm #1}\hfill}}\bigskip \par\nobreak
}
\def\CHAPFopen{%
\global\let\chapmacro=\chfopen
\global\let\centerchapmacro=\centerchfopen}
% Section titles. These macros combine the section number parts and
% call the generic \sectionheading to do the printing.
%
\newskip\secheadingskip
\def\secheadingbreak{\dobreak \secheadingskip{-1000}}
% Subsection titles.
\newskip\subsecheadingskip
\def\subsecheadingbreak{\dobreak \subsecheadingskip{-500}}
% Subsubsection titles.
\def\subsubsecheadingskip{\subsecheadingskip}
\def\subsubsecheadingbreak{\subsecheadingbreak}
% Print any size, any type, section title.
%
% #1 is the text, #2 is the section level (sec/subsec/subsubsec), #3 is
% the section type for xrefs (Ynumbered, Ynothing, Yappendix), #4 is the
% section number.
%
\def\sectionheading#1#2#3#4{%
{%
% Switch to the right set of fonts.
\csname #2fonts\endcsname \rm
%
% Insert space above the heading.
\csname #2headingbreak\endcsname
%
% Only insert the space after the number if we have a section number.
\def\sectionlevel{#2}%
\def\temptype{#3}%
%
\ifx\temptype\Ynothingkeyword
\setbox0 = \hbox{}%
\def\toctype{unn}%
\gdef\thissection{#1}%
\else\ifx\temptype\Yomitfromtockeyword
% for @headings -- no section number, don't include in toc,
% and don't redefine \thissection.
\setbox0 = \hbox{}%
\def\toctype{omit}%
\let\sectionlevel=\empty
\else\ifx\temptype\Yappendixkeyword
\setbox0 = \hbox{#4\enspace}%
\def\toctype{app}%
\gdef\thissection{#1}%
\else
\setbox0 = \hbox{#4\enspace}%
\def\toctype{num}%
\gdef\thissection{#1}%
\fi\fi\fi
%
% Write the toc entry (before \donoderef). See comments in \chapmacro.
\writetocentry{\toctype\sectionlevel}{#1}{#4}%
%
% Write the node reference (= pdf destination for pdftex).
% Again, see comments in \chapmacro.
\donoderef{#3}%
%
% Interline glue will be inserted when the vbox is completed.
% That glue will be a valid breakpoint for the page, since it'll be
% preceded by a whatsit (usually from the \donoderef, or from the
% \writetocentry if there was no node). We don't want to allow that
% break, since then the whatsits could end up on page n while the
% section is on page n+1, thus toc/etc. are wrong. Debian bug 276000.
\nobreak
%
% Output the actual section heading.
\vbox{\hyphenpenalty=10000 \tolerance=5000 \parindent=0pt \raggedright
\hangindent=\wd0 % zero if no section number
\unhbox0 #1}%
}%
% Add extra space after the heading -- half of whatever came above it.
% Don't allow stretch, though.
\kern .5 \csname #2headingskip\endcsname
%
% Do not let the kern be a potential breakpoint, as it would be if it
% was followed by glue.
\nobreak
%
% We'll almost certainly start a paragraph next, so don't let that
% glue accumulate. (Not a breakpoint because it's preceded by a
% discardable item.)
\vskip-\parskip
%
% This is purely so the last item on the list is a known \penalty >
% 10000. This is so \startdefun can avoid allowing breakpoints after
% section headings. Otherwise, it would insert a valid breakpoint between:
%
% @section sec-whatever
% @deffn def-whatever
\penalty 10001
}
\message{toc,}
% Table of contents.
\newwrite\tocfile
% Write an entry to the toc file, opening it if necessary.
% Called from @chapter, etc.
%
% Example usage: \writetocentry{sec}{Section Name}{\the\chapno.\the\secno}
% We append the current node name (if any) and page number as additional
% arguments for the \{chap,sec,...}entry macros which will eventually
% read this. The node name is used in the pdf outlines as the
% destination to jump to.
%
% We open the .toc file for writing here instead of at @setfilename (or
% any other fixed time) so that @contents can be anywhere in the document.
% But if #1 is `omit', then we don't do anything. This is used for the
% table of contents chapter openings themselves.
%
\newif\iftocfileopened
\def\omitkeyword{omit}%
%
\def\writetocentry#1#2#3{%
\edef\writetoctype{#1}%
\ifx\writetoctype\omitkeyword \else
\iftocfileopened\else
\immediate\openout\tocfile = \jobname.toc
\global\tocfileopenedtrue
\fi
%
\iflinks
{\atdummies
\edef\temp{%
\write\tocfile{@#1entry{#2}{#3}{\lastnode}{\noexpand\folio}}}%
\temp
}%
\fi
\fi
%
% Tell \shipout to create a pdf destination on each page, if we're
% writing pdf. These are used in the table of contents. We can't
% just write one on every page because the title pages are numbered
% 1 and 2 (the page numbers aren't printed), and so are the first
% two pages of the document. Thus, we'd have two destinations named
% `1', and two named `2'.
\ifpdf \global\pdfmakepagedesttrue \fi
}
% These characters do not print properly in the Computer Modern roman
% fonts, so we must take special care. This is more or less redundant
% with the Texinfo input format setup at the end of this file.
%
\def\activecatcodes{%
\catcode`\"=\active
\catcode`\$=\active
\catcode`\<=\active
\catcode`\>=\active
\catcode`\\=\active
\catcode`\^=\active
\catcode`\_=\active
\catcode`\|=\active
\catcode`\~=\active
}
% Read the toc file, which is essentially Texinfo input.
\def\readtocfile{%
\setupdatafile
\activecatcodes
\input \jobname.toc
}
\newskip\contentsrightmargin \contentsrightmargin=1in
\newcount\savepageno
\newcount\lastnegativepageno \lastnegativepageno = -1
% Prepare to read what we've written to \tocfile.
%
\def\startcontents#1{%
% If @setchapternewpage on, and @headings double, the contents should
% start on an odd page, unlike chapters. Thus, we maintain
% \contentsalignmacro in parallel with \pagealignmacro.
% From: Torbjorn Granlund <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.
\raggedbottom % Worry more about breakpoints than the bottom.
\advance\hsize by -\contentsrightmargin % Don't use the full line length.
%
% Roman numerals for page numbers.
\ifnum \pageno>0 \global\pageno = \lastnegativepageno \fi
}
% Normal (long) toc.
\def\contents{%
\startcontents{\putwordTOC}%
\openin 1 \jobname.toc
\ifeof 1 \else
\readtocfile
\fi
\vfill \eject
\contentsalignmacro % in case @setchapternewpage odd is in effect
\ifeof 1 \else
\pdfmakeoutlines
\fi
\closein 1
\endgroup
\lastnegativepageno = \pageno
\global\pageno = \savepageno
}
% And just the chapters.
\def\summarycontents{%
\startcontents{\putwordShortTOC}%
%
\let\numchapentry = \shortchapentry
\let\appentry = \shortchapentry
\let\unnchapentry = \shortunnchapentry
% We want a true roman here for the page numbers.
\secfonts
\let\rm=\shortcontrm \let\bf=\shortcontbf
\let\sl=\shortcontsl \let\tt=\shortconttt
\rm
\hyphenpenalty = 10000
\advance\baselineskip by 1pt % Open it up a little.
\def\numsecentry##1##2##3##4{}
\let\appsecentry = \numsecentry
\let\unnsecentry = \numsecentry
\let\numsubsecentry = \numsecentry
\let\appsubsecentry = \numsecentry
\let\unnsubsecentry = \numsecentry
\let\numsubsubsecentry = \numsecentry
\let\appsubsubsecentry = \numsecentry
\let\unnsubsubsecentry = \numsecentry
\openin 1 \jobname.toc
\ifeof 1 \else
\readtocfile
\fi
\closein 1
\vfill \eject
\contentsalignmacro % in case @setchapternewpage odd is in effect
\endgroup
\lastnegativepageno = \pageno
\global\pageno = \savepageno
}
\let\shortcontents = \summarycontents
% Typeset the label for a chapter or appendix for the short contents.
% The arg is, e.g., `A' for an appendix, or `3' for a chapter.
%
\def\shortchaplabel#1{%
% This space should be enough, since a single number is .5em, and the
% widest letter (M) is 1em, at least in the Computer Modern fonts.
% But use \hss just in case.
% (This space doesn't include the extra space that gets added after
% the label; that gets put in by \shortchapentry above.)
%
% We'd like to right-justify chapter numbers, but that looks strange
% with appendix letters. And right-justifying numbers and
% left-justifying letters looks strange when there is less than 10
% chapters. Have to read the whole toc once to know how many chapters
% there are before deciding ...
\hbox to 1em{#1\hss}%
}
% These macros generate individual entries in the table of contents.
% The first argument is the chapter or section name.
% The last argument is the page number.
% The arguments in between are the chapter number, section number, ...
% Chapters, in the main contents.
\def\numchapentry#1#2#3#4{\dochapentry{#2\labelspace#1}{#4}}
%
% Chapters, in the short toc.
% See comments in \dochapentry re vbox and related settings.
\def\shortchapentry#1#2#3#4{%
\tocentry{\shortchaplabel{#2}\labelspace #1}{\doshortpageno\bgroup#4\egroup}%
}
% Appendices, in the main contents.
% Need the word Appendix, and a fixed-size box.
%
\def\appendixbox#1{%
% We use M since it's probably the widest letter.
\setbox0 = \hbox{\putwordAppendix{} M}%
\hbox to \wd0{\putwordAppendix{} #1\hss}}
%
\def\appentry#1#2#3#4{\dochapentry{\appendixbox{#2}\labelspace#1}{#4}}
% Unnumbered chapters.
\def\unnchapentry#1#2#3#4{\dochapentry{#1}{#4}}
\def\shortunnchapentry#1#2#3#4{\tocentry{#1}{\doshortpageno\bgroup#4\egroup}}
% Sections.
\def\numsecentry#1#2#3#4{\dosecentry{#2\labelspace#1}{#4}}
\let\appsecentry=\numsecentry
\def\unnsecentry#1#2#3#4{\dosecentry{#1}{#4}}
% Subsections.
\def\numsubsecentry#1#2#3#4{\dosubsecentry{#2\labelspace#1}{#4}}
\let\appsubsecentry=\numsubsecentry
\def\unnsubsecentry#1#2#3#4{\dosubsecentry{#1}{#4}}
% And subsubsections.
\def\numsubsubsecentry#1#2#3#4{\dosubsubsecentry{#2\labelspace#1}{#4}}
\let\appsubsubsecentry=\numsubsubsecentry
\def\unnsubsubsecentry#1#2#3#4{\dosubsubsecentry{#1}{#4}}
% This parameter controls the indentation of the various levels.
% Same as \defaultparindent.
\newdimen\tocindent \tocindent = 15pt
% Now for the actual typesetting. In all these, #1 is the text and #2 is the
% page number.
%
% If the toc has to be broken over pages, we want it to be at chapters
% if at all possible; hence the \penalty.
\def\dochapentry#1#2{%
\penalty-300 \vskip1\baselineskip plus.33\baselineskip minus.25\baselineskip
\begingroup
\chapentryfonts
\tocentry{#1}{\dopageno\bgroup#2\egroup}%
\endgroup
\nobreak\vskip .25\baselineskip plus.1\baselineskip
}
\def\dosecentry#1#2{\begingroup
\secentryfonts \leftskip=\tocindent
\tocentry{#1}{\dopageno\bgroup#2\egroup}%
\endgroup}
\def\dosubsecentry#1#2{\begingroup
\subsecentryfonts \leftskip=2\tocindent
\tocentry{#1}{\dopageno\bgroup#2\egroup}%
\endgroup}
\def\dosubsubsecentry#1#2{\begingroup
\subsubsecentryfonts \leftskip=3\tocindent
\tocentry{#1}{\dopageno\bgroup#2\egroup}%
\endgroup}
% We use the same \entry macro as for the index entries.
\let\tocentry = \entry
% Space between chapter (or whatever) number and the title.
\def\labelspace{\hskip1em \relax}
\def\dopageno#1{{\rm #1}}
\def\doshortpageno#1{{\rm #1}}
\def\chapentryfonts{\secfonts \rm}
\def\secentryfonts{\textfonts}
\def\subsecentryfonts{\textfonts}
\def\subsubsecentryfonts{\textfonts}
\message{environments,}
% @foo ... @end foo.
% @point{}, @result{}, @expansion{}, @print{}, @equiv{}.
%
% Since these characters are used in examples, it should be an even number of
% \tt widths. Each \tt character is 1en, so two makes it 1em.
%
\def\point{$\star$}
\def\result{\leavevmode\raise.15ex\hbox to 1em{\hfil$\Rightarrow$\hfil}}
\def\expansion{\leavevmode\raise.1ex\hbox to 1em{\hfil$\mapsto$\hfil}}
\def\print{\leavevmode\lower.1ex\hbox to 1em{\hfil$\dashv$\hfil}}
\def\equiv{\leavevmode\lower.1ex\hbox to 1em{\hfil$\ptexequiv$\hfil}}
% The @error{} command.
% Adapted from the TeXbook's \boxit.
%
\newbox\errorbox
%
{\tentt \global\dimen0 = 3em}% Width of the box.
\dimen2 = .55pt % Thickness of rules
% The text. (`r' is open on the right, `e' somewhat less so on the left.)
\setbox0 = \hbox{\kern-.75pt \reducedsf error\kern-1.5pt}
%
\setbox\errorbox=\hbox to \dimen0{\hfil
\hsize = \dimen0 \advance\hsize by -5.8pt % Space to left+right.
\advance\hsize by -2\dimen2 % Rules.
\vbox{%
\hrule height\dimen2
\hbox{\vrule width\dimen2 \kern3pt % Space to left of text.
\vtop{\kern2.4pt \box0 \kern2.4pt}% Space above/below.
\kern3pt\vrule width\dimen2}% Space to right.
\hrule height\dimen2}
\hfil}
%
\def\error{\leavevmode\lower.7ex\copy\errorbox}
% @tex ... @end tex escapes into raw Tex temporarily.
% One exception: @ is still an escape character, so that @end tex works.
% But \@ or @@ will get a plain tex @ character.
\envdef\tex{%
\catcode `\\=0 \catcode `\{=1 \catcode `\}=2
\catcode `\$=3 \catcode `\&=4 \catcode `\#=6
\catcode `\^=7 \catcode `\_=8 \catcode `\~=\active \let~=\tie
\catcode `\%=14
\catcode `\+=\other
\catcode `\"=\other
\catcode `\|=\other
\catcode `\<=\other
\catcode `\>=\other
\escapechar=`\\
%
\let\b=\ptexb
\let\bullet=\ptexbullet
\let\c=\ptexc
\let\,=\ptexcomma
\let\.=\ptexdot
\let\dots=\ptexdots
\let\equiv=\ptexequiv
\let\!=\ptexexclam
\let\i=\ptexi
\let\indent=\ptexindent
\let\noindent=\ptexnoindent
\let\{=\ptexlbrace
\let\+=\tabalign
\let\}=\ptexrbrace
\let\/=\ptexslash
\let\*=\ptexstar
\let\t=\ptext
\let\frenchspacing=\plainfrenchspacing
%
\def\endldots{\mathinner{\ldots\ldots\ldots\ldots}}%
\def\enddots{\relax\ifmmode\endldots\else$\mathsurround=0pt \endldots\,$\fi}%
\def\@{@}%
}
% There is no need to define \Etex.
% Define @lisp ... @end lisp.
% @lisp environment forms a group so it can rebind things,
% including the definition of @end lisp (which normally is erroneous).
% Amount to narrow the margins by for @lisp.
\newskip\lispnarrowing \lispnarrowing=0.4in
% This is the definition that ^^M gets inside @lisp, @example, and other
% such environments. \null is better than a space, since it doesn't
% have any width.
\def\lisppar{\null\endgraf}
% This space is always present above and below environments.
\newskip\envskipamount \envskipamount = 0pt
% Make spacing and below environment symmetrical. We use \parskip here
% to help in doing that, since in @example-like environments \parskip
% is reset to zero; thus the \afterenvbreak inserts no space -- but the
% start of the next paragraph will insert \parskip.
%
\def\aboveenvbreak{{%
% =10000 instead of <10000 because of a special case in \itemzzz and
% \sectionheading, q.v.
\ifnum \lastpenalty=10000 \else
\advance\envskipamount by \parskip
\endgraf
\ifdim\lastskip<\envskipamount
\removelastskip
% it's not a good place to break if the last penalty was \nobreak
% or better ...
\ifnum\lastpenalty<10000 \penalty-50 \fi
\vskip\envskipamount
\fi
\fi
}}
\let\afterenvbreak = \aboveenvbreak
% \nonarrowing is a flag. If "set", @lisp etc don't narrow margins; it will
% also clear it, so that its embedded environments do the narrowing again.
\let\nonarrowing=\relax
% @cartouche ... @end cartouche: draw rectangle w/rounded corners around
% environment contents.
\font\circle=lcircle10
\newdimen\circthick
\newdimen\cartouter\newdimen\cartinner
\newskip\normbskip\newskip\normpskip\newskip\normlskip
\circthick=\fontdimen8\circle
%
\def\ctl{{\circle\char'013\hskip -6pt}}% 6pt from pl file: 1/2charwidth
\def\ctr{{\hskip 6pt\circle\char'010}}
\def\cbl{{\circle\char'012\hskip -6pt}}
\def\cbr{{\hskip 6pt\circle\char'011}}
\def\carttop{\hbox to \cartouter{\hskip\lskip
\ctl\leaders\hrule height\circthick\hfil\ctr
\hskip\rskip}}
\def\cartbot{\hbox to \cartouter{\hskip\lskip
\cbl\leaders\hrule height\circthick\hfil\cbr
\hskip\rskip}}
%
\newskip\lskip\newskip\rskip
\envdef\cartouche{%
\ifhmode\par\fi % can't be in the midst of a paragraph.
\startsavinginserts
\lskip=\leftskip \rskip=\rightskip
\leftskip=0pt\rightskip=0pt % we want these *outside*.
\cartinner=\hsize \advance\cartinner by-\lskip
\advance\cartinner by-\rskip
\cartouter=\hsize
\advance\cartouter by 18.4pt % allow for 3pt kerns on either
% side, and for 6pt waste from
% each corner char, and rule thickness
\normbskip=\baselineskip \normpskip=\parskip \normlskip=\lineskip
% Flag to tell @lisp, etc., not to narrow margin.
\let\nonarrowing = t%
\vbox\bgroup
\baselineskip=0pt\parskip=0pt\lineskip=0pt
\carttop
\hbox\bgroup
\hskip\lskip
\vrule\kern3pt
\vbox\bgroup
\kern3pt
\hsize=\cartinner
\baselineskip=\normbskip
\lineskip=\normlskip
\parskip=\normpskip
\vskip -\parskip
\comment % For explanation, see the end of \def\group.
}
\def\Ecartouche{%
\ifhmode\par\fi
\kern3pt
\egroup
\kern3pt\vrule
\hskip\rskip
\egroup
\cartbot
\egroup
\checkinserts
}
% This macro is called at the beginning of all the @example variants,
% inside a group.
\def\nonfillstart{%
\aboveenvbreak
\hfuzz = 12pt % Don't be fussy
\sepspaces % Make spaces be word-separators rather than space tokens.
\let\par = \lisppar % don't ignore blank lines
\obeylines % each line of input is a line of output
\parskip = 0pt
\parindent = 0pt
\emergencystretch = 0pt % don't try to avoid overfull boxes
\ifx\nonarrowing\relax
\advance \leftskip by \lispnarrowing
\exdentamount=\lispnarrowing
\else
\let\nonarrowing = \relax
\fi
\let\exdent=\nofillexdent
}
% If you want all examples etc. small: @set dispenvsize small.
% If you want even small examples the full size: @set dispenvsize nosmall.
% This affects the following displayed environments:
% @example, @display, @format, @lisp
%
\def\smallword{small}
\def\nosmallword{nosmall}
\let\SETdispenvsize\relax
\def\setnormaldispenv{%
\ifx\SETdispenvsize\smallword
% end paragraph for sake of leading, in case document has no blank
% line. This is redundant with what happens in \aboveenvbreak, but
% we need to do it before changing the fonts, and it's inconvenient
% to change the fonts afterward.
\ifnum \lastpenalty=10000 \else \endgraf \fi
\smallexamplefonts \rm
\fi
}
\def\setsmalldispenv{%
\ifx\SETdispenvsize\nosmallword
\else
\ifnum \lastpenalty=10000 \else \endgraf \fi
\smallexamplefonts \rm
\fi
}
% We often define two environments, @foo and @smallfoo.
% Let's do it by one command:
\def\makedispenv #1#2{
\expandafter\envdef\csname#1\endcsname {\setnormaldispenv #2}
\expandafter\envdef\csname small#1\endcsname {\setsmalldispenv #2}
\expandafter\let\csname E#1\endcsname \afterenvbreak
\expandafter\let\csname Esmall#1\endcsname \afterenvbreak
}
% Define two synonyms:
\def\maketwodispenvs #1#2#3{
\makedispenv{#1}{#3}
\makedispenv{#2}{#3}
}
% @lisp: indented, narrowed, typewriter font; @example: same as @lisp.
%
% @smallexample and @smalllisp: use smaller fonts.
% Originally contributed by Pavel@xerox.
%
\maketwodispenvs {lisp}{example}{%
\nonfillstart
\tt\quoteexpand
\let\kbdfont = \kbdexamplefont % Allow @kbd to do something special.
\gobble % eat return
}
% @display/@smalldisplay: same as @lisp except keep current font.
%
\makedispenv {display}{%
\nonfillstart
\gobble
}
% @format/@smallformat: same as @display except don't narrow margins.
%
\makedispenv{format}{%
\let\nonarrowing = t%
\nonfillstart
\gobble
}
% @flushleft: same as @format, but doesn't obey \SETdispenvsize.
\envdef\flushleft{%
\let\nonarrowing = t%
\nonfillstart
\gobble
}
\let\Eflushleft = \afterenvbreak
% @flushright.
%
\envdef\flushright{%
\let\nonarrowing = t%
\nonfillstart
\advance\leftskip by 0pt plus 1fill
\gobble
}
\let\Eflushright = \afterenvbreak
% @quotation does normal linebreaking (hence we can't use \nonfillstart)
% and narrows the margins. We keep \parskip nonzero in general, since
% we're doing normal filling. So, when using \aboveenvbreak and
% \afterenvbreak, temporarily make \parskip 0.
%
\envdef\quotation{%
{\parskip=0pt \aboveenvbreak}% because \aboveenvbreak inserts \parskip
\parindent=0pt
%
% @cartouche defines \nonarrowing to inhibit narrowing at next level down.
\ifx\nonarrowing\relax
\advance\leftskip by \lispnarrowing
\advance\rightskip by \lispnarrowing
\exdentamount = \lispnarrowing
\else
\let\nonarrowing = \relax
\fi
\parsearg\quotationlabel
}
% We have retained a nonzero parskip for the environment, since we're
% doing normal filling.
%
\def\Equotation{%
\par
\ifx\quotationauthor\undefined\else
% indent a bit.
\leftline{\kern 2\leftskip \sl ---\quotationauthor}%
\fi
{\parskip=0pt \afterenvbreak}%
}
% If we're given an argument, typeset it in bold with a colon after.
\def\quotationlabel#1{%
\def\temp{#1}%
\ifx\temp\empty \else
{\bf #1: }%
\fi
}
% LaTeX-like @verbatim...@end verbatim and @verb{<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}
% Allow an option to not replace quotes with a regular directed right
% quote/apostrophe (char 0x27), but instead use the undirected quote
% from cmtt (char 0x0d). The undirected quote is ugly, so don't make it
% the default, but it works for pasting with more pdf viewers (at least
% evince), the lilypond developers report. xpdf does work with the
% regular 0x27.
%
\def\codequoteright{%
\expandafter\ifx\csname SETcodequoteundirected\endcsname\relax
'%
\else
\char'15
\fi
}
%
% and a similar option for the left quote char vs. a grave accent.
% Modern fonts display ASCII 0x60 as a grave accent, so some people like
% the code environments to do likewise.
%
\def\codequoteleft{%
\expandafter\ifx\csname SETcodequotebacktick\endcsname\relax
`%
\else
\char'22
\fi
}
%
\begingroup
\catcode`\^^I=\active
\gdef\tabexpand{%
\catcode`\^^I=\active
\def^^I{\leavevmode\egroup
\dimen0=\wd0 % the width so far, or since the previous tab
\divide\dimen0 by\tabw
\multiply\dimen0 by\tabw % compute previous multiple of \tabw
\advance\dimen0 by\tabw % advance to next multiple of \tabw
\wd0=\dimen0 \box0 \starttabbox
}%
}
\catcode`\'=\active
\gdef\rquoteexpand{\catcode\rquoteChar=\active \def'{\codequoteright}}%
%
\catcode`\`=\active
\gdef\lquoteexpand{\catcode\lquoteChar=\active \def`{\codequoteleft}}%
%
\gdef\quoteexpand{\rquoteexpand \lquoteexpand}%
\endgroup
% start the verbatim environment.
\def\setupverbatim{%
\let\nonarrowing = t%
\nonfillstart
% Easiest (and conventionally used) font for verbatim
\tt
\def\par{\leavevmode\egroup\box0\endgraf}%
\catcode`\`=\active
\tabexpand
\quoteexpand
% Respect line breaks,
% print special symbols as themselves, and
% make each space count
% must do in this order:
\obeylines \uncatcodespecials \sepspaces
\everypar{\starttabbox}%
}
% Do the @verb magic: verbatim text is quoted by unique
% delimiter characters. Before first delimiter expect a
% right brace, after last delimiter expect closing brace:
%
% \def\doverb'{'<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
\newcount\defunpenalty
% Start the processing of @deffn:
\def\startdefun{%
\ifnum\lastpenalty<10000
\medbreak
\defunpenalty=10003 % Will keep this @deffn together with the
% following @def command, see below.
\else
% If there are two @def commands in a row, we'll have a \nobreak,
% which is there to keep the function description together with its
% header. But if there's nothing but headers, we need to allow a
% break somewhere. Check specifically for penalty 10002, inserted
% by \printdefunline, instead of 10000, since the sectioning
% commands also insert a nobreak penalty, and we don't want to allow
% a break between a section heading and a defun.
%
% As a minor refinement, we avoid "club" headers by signalling
% with penalty of 10003 after the very first @deffn in the
% sequence (see above), and penalty of 10002 after any following
% @def command.
\ifnum\lastpenalty=10002 \penalty2000 \else \defunpenalty=10002 \fi
%
% Similarly, after a section heading, do not allow a break.
% But do insert the glue.
\medskip % preceded by discardable penalty, so not a breakpoint
\fi
%
\parindent=0in
\advance\leftskip by \defbodyindent
\exdentamount=\defbodyindent
}
\def\dodefunx#1{%
% First, check whether we are in the right environment:
\checkenv#1%
%
% As above, allow line break if we have multiple x headers in a row.
% It's not a great place, though.
\ifnum\lastpenalty=10002 \penalty3000 \else \defunpenalty=10002 \fi
%
% And now, it's time to reuse the body of the original defun:
\expandafter\gobbledefun#1%
}
\def\gobbledefun#1\startdefun{}
% \printdefunline \deffnheader{text}
%
\def\printdefunline#1#2{%
\begingroup
% call \deffnheader:
#1#2 \endheader
% common ending:
\interlinepenalty = 10000
\advance\rightskip by 0pt plus 1fil
\endgraf
\nobreak\vskip -\parskip
\penalty\defunpenalty % signal to \startdefun and \dodefunx
% Some of the @defun-type tags do not enable magic parentheses,
% rendering the following check redundant. But we don't optimize.
\checkparencounts
\endgroup
}
\def\Edefun{\endgraf\medbreak}
% \makedefun{deffn} creates \deffn, \deffnx and \Edeffn;
% the only thing remainnig is to define \deffnheader.
%
\def\makedefun#1{%
\expandafter\let\csname E#1\endcsname = \Edefun
\edef\temp{\noexpand\domakedefun
\makecsname{#1}\makecsname{#1x}\makecsname{#1header}}%
\temp
}
% \domakedefun \deffn \deffnx \deffnheader
%
% Define \deffn and \deffnx, without parameters.
% \deffnheader has to be defined explicitly.
%
\def\domakedefun#1#2#3{%
\envdef#1{%
\startdefun
\parseargusing\activeparens{\printdefunline#3}%
}%
\def#2{\dodefunx#1}%
\def#3%
}
%%% Untyped functions:
% @deffn category name args
\makedefun{deffn}{\deffngeneral{}}
% @deffn category class name args
\makedefun{defop}#1 {\defopon{#1\ \putwordon}}
% \defopon {category on}class name args
\def\defopon#1#2 {\deffngeneral{\putwordon\ \code{#2}}{#1\ \code{#2}} }
% \deffngeneral {subind}category name args
%
\def\deffngeneral#1#2 #3 #4\endheader{%
% Remember that \dosubind{fn}{foo}{} is equivalent to \doind{fn}{foo}.
\dosubind{fn}{\code{#3}}{#1}%
\defname{#2}{}{#3}\magicamp\defunargs{#4\unskip}%
}
%%% Typed functions:
% @deftypefn category type name args
\makedefun{deftypefn}{\deftypefngeneral{}}
% @deftypeop category class type name args
\makedefun{deftypeop}#1 {\deftypeopon{#1\ \putwordon}}
% \deftypeopon {category on}class type name args
\def\deftypeopon#1#2 {\deftypefngeneral{\putwordon\ \code{#2}}{#1\ \code{#2}} }
% \deftypefngeneral {subind}category type name args
%
\def\deftypefngeneral#1#2 #3 #4 #5\endheader{%
\dosubind{fn}{\code{#4}}{#1}%
\defname{#2}{#3}{#4}\defunargs{#5\unskip}%
}
%%% Typed variables:
% @deftypevr category type var args
\makedefun{deftypevr}{\deftypecvgeneral{}}
% @deftypecv category class type var args
\makedefun{deftypecv}#1 {\deftypecvof{#1\ \putwordof}}
% \deftypecvof {category of}class type var args
\def\deftypecvof#1#2 {\deftypecvgeneral{\putwordof\ \code{#2}}{#1\ \code{#2}} }
% \deftypecvgeneral {subind}category type var args
%
\def\deftypecvgeneral#1#2 #3 #4 #5\endheader{%
\dosubind{vr}{\code{#4}}{#1}%
\defname{#2}{#3}{#4}\defunargs{#5\unskip}%
}
%%% Untyped variables:
% @defvr category var args
\makedefun{defvr}#1 {\deftypevrheader{#1} {} }
% @defcv category class var args
\makedefun{defcv}#1 {\defcvof{#1\ \putwordof}}
% \defcvof {category of}class var args
\def\defcvof#1#2 {\deftypecvof{#1}#2 {} }
%%% Type:
% @deftp category name args
\makedefun{deftp}#1 #2 #3\endheader{%
\doind{tp}{\code{#2}}%
\defname{#1}{}{#2}\defunargs{#3\unskip}%
}
% Remaining @defun-like shortcuts:
\makedefun{defun}{\deffnheader{\putwordDeffunc} }
\makedefun{defmac}{\deffnheader{\putwordDefmac} }
\makedefun{defspec}{\deffnheader{\putwordDefspec} }
\makedefun{deftypefun}{\deftypefnheader{\putwordDeffunc} }
\makedefun{defvar}{\defvrheader{\putwordDefvar} }
\makedefun{defopt}{\defvrheader{\putwordDefopt} }
\makedefun{deftypevar}{\deftypevrheader{\putwordDefvar} }
\makedefun{defmethod}{\defopon\putwordMethodon}
\makedefun{deftypemethod}{\deftypeopon\putwordMethodon}
\makedefun{defivar}{\defcvof\putwordInstanceVariableof}
\makedefun{deftypeivar}{\deftypecvof\putwordInstanceVariableof}
% \defname, which formats the name of the @def (not the args).
% #1 is the category, such as "Function".
% #2 is the return type, if any.
% #3 is the function name.
%
% We are followed by (but not passed) the arguments, if any.
%
\def\defname#1#2#3{%
% Get the values of \leftskip and \rightskip as they were outside the @def...
\advance\leftskip by -\defbodyindent
%
% How we'll format the type name. Putting it in brackets helps
% distinguish it from the body text that may end up on the next line
% just below it.
\def\temp{#1}%
\setbox0=\hbox{\kern\deflastargmargin \ifx\temp\empty\else [\rm\temp]\fi}
%
% Figure out line sizes for the paragraph shape.
% The first line needs space for \box0; but if \rightskip is nonzero,
% we need only space for the part of \box0 which exceeds it:
\dimen0=\hsize \advance\dimen0 by -\wd0 \advance\dimen0 by \rightskip
% The continuations:
\dimen2=\hsize \advance\dimen2 by -\defargsindent
% (plain.tex says that \dimen1 should be used only as global.)
\parshape 2 0in \dimen0 \defargsindent \dimen2
%
% Put the type name to the right margin.
\noindent
\hbox to 0pt{%
\hfil\box0 \kern-\hsize
% \hsize has to be shortened this way:
\kern\leftskip
% Intentionally do not respect \rightskip, since we need the space.
}%
%
% Allow all lines to be underfull without complaint:
\tolerance=10000 \hbadness=10000
\exdentamount=\defbodyindent
{%
% defun fonts. We use typewriter by default (used to be bold) because:
% . we're printing identifiers, they should be in tt in principle.
% . in languages with many accents, such as Czech or French, it's
% common to leave accents off identifiers. The result looks ok in
% tt, but exceedingly strange in rm.
% . we don't want -- and --- to be treated as ligatures.
% . this still does not fix the ?` and !` ligatures, but so far no
% one has made identifiers using them :).
\df \tt
\def\temp{#2}% return value type
\ifx\temp\empty\else \tclose{\temp} \fi
#3% output function name
}%
{\rm\enskip}% hskip 0.5 em of \tenrm
%
\boldbrax
% arguments will be output next, if any.
}
% Print arguments in slanted roman (not ttsl), inconsistently with using
% tt for the name. This is because literal text is sometimes needed in
% the argument list (groff manual), and ttsl and tt are not very
% distinguishable. Prevent hyphenation at `-' chars.
%
\def\defunargs#1{%
% use sl by default (not ttsl),
% tt for the names.
\df \sl \hyphenchar\font=0
%
% On the other hand, if an argument has two dashes (for instance), we
% want a way to get ttsl. Let's try @var for that.
\let\var=\ttslanted
#1%
\sl\hyphenchar\font=45
}
% We want ()&[] to print specially on the defun line.
%
\def\activeparens{%
\catcode`\(=\active \catcode`\)=\active
\catcode`\[=\active \catcode`\]=\active
\catcode`\&=\active
}
% Make control sequences which act like normal parenthesis chars.
\let\lparen = ( \let\rparen = )
% Be sure that we always have a definition for `(', etc. For example,
% if the fn name has parens in it, \boldbrax will not be in effect yet,
% so TeX would otherwise complain about undefined control sequence.
{
\activeparens
\global\let(=\lparen \global\let)=\rparen
\global\let[=\lbrack \global\let]=\rbrack
\global\let& = \&
\gdef\boldbrax{\let(=\opnr\let)=\clnr\let[=\lbrb\let]=\rbrb}
\gdef\magicamp{\let&=\amprm}
}
\newcount\parencount
% If we encounter &foo, then turn on ()-hacking afterwards
\newif\ifampseen
\def\amprm#1 {\ampseentrue{\bf\&#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?
% List of all defined macros in the form
% \definedummyword\macro1\definedummyword\macro2...
% Currently is also contains all @aliases; the list can be split
% if there is a need.
\def\macrolist{}
% Add the macro to \macrolist
\def\addtomacrolist#1{\expandafter \addtomacrolistxxx \csname#1\endcsname}
\def\addtomacrolistxxx#1{%
\toks0 = \expandafter{\macrolist\definedummyword#1}%
\xdef\macrolist{\the\toks0}%
}
% Utility routines.
% This does \let #1 = #2, with \csnames; that is,
% \let \csname#1\endcsname = \csname#2\endcsname
% (except of course we have to play expansion games).
%
\def\cslet#1#2{%
\expandafter\let
\csname#1\expandafter\endcsname
\csname#2\endcsname
}
% Trim leading and trailing spaces off a string.
% Concepts from aro-bend problem 15 (see CTAN).
{\catcode`\@=11
\gdef\eatspaces #1{\expandafter\trim@\expandafter{#1 }}
\gdef\trim@ #1{\trim@@ @#1 @ #1 @ @@}
\gdef\trim@@ #1@ #2@ #3@@{\trim@@@\empty #2 @}
\def\unbrace#1{#1}
\unbrace{\gdef\trim@@@ #1 } #2@{#1}
}
% Trim a single trailing ^^M off a string.
{\catcode`\^^M=\other \catcode`\Q=3%
\gdef\eatcr #1{\eatcra #1Q^^MQ}%
\gdef\eatcra#1^^MQ{\eatcrb#1Q}%
\gdef\eatcrb#1Q#2Q{#1}%
}
% Macro bodies are absorbed as an argument in a context where
% all characters are catcode 10, 11 or 12, except \ which is active
% (as in normal texinfo). It is necessary to change the definition of \.
% It's necessary to have hard CRs when the macro is executed. This is
% done by making ^^M (\endlinechar) catcode 12 when reading the macro
% body, and then making it the \newlinechar in \scanmacro.
\def\scanctxt{%
\catcode`\"=\other
\catcode`\+=\other
\catcode`\<=\other
\catcode`\>=\other
\catcode`\@=\other
\catcode`\^=\other
\catcode`\_=\other
\catcode`\|=\other
\catcode`\~=\other
}
\def\scanargctxt{%
\scanctxt
\catcode`\\=\other
\catcode`\^^M=\other
}
\def\macrobodyctxt{%
\scanctxt
\catcode`\{=\other
\catcode`\}=\other
\catcode`\^^M=\other
\usembodybackslash
}
\def\macroargctxt{%
\scanctxt
\catcode`\\=\other
}
% \mbodybackslash is the definition of \ in @macro bodies.
% It maps \foo\ => \csname macarg.foo\endcsname => #N
% where N is the macro parameter number.
% We define \csname macarg.\endcsname to be \realbackslash, so
% \\ in macro replacement text gets you a backslash.
{\catcode`@=0 @catcode`@\=@active
@gdef@usembodybackslash{@let\=@mbodybackslash}
@gdef@mbodybackslash#1\{@csname macarg.#1@endcsname}
}
\expandafter\def\csname macarg.\endcsname{\realbackslash}
\def\macro{\recursivefalse\parsearg\macroxxx}
\def\rmacro{\recursivetrue\parsearg\macroxxx}
\def\macroxxx#1{%
\getargs{#1}% now \macname is the macname and \argl the arglist
\ifx\argl\empty % no arguments
\paramno=0%
\else
\expandafter\parsemargdef \argl;%
\fi
\if1\csname ismacro.\the\macname\endcsname
\message{Warning: redefining \the\macname}%
\else
\expandafter\ifx\csname \the\macname\endcsname \relax
\else \errmessage{Macro name \the\macname\space already defined}\fi
\global\cslet{macsave.\the\macname}{\the\macname}%
\global\expandafter\let\csname ismacro.\the\macname\endcsname=1%
\addtomacrolist{\the\macname}%
\fi
\begingroup \macrobodyctxt
\ifrecursive \expandafter\parsermacbody
\else \expandafter\parsemacbody
\fi}
\parseargdef\unmacro{%
\if1\csname ismacro.#1\endcsname
\global\cslet{#1}{macsave.#1}%
\global\expandafter\let \csname ismacro.#1\endcsname=0%
% Remove the macro name from \macrolist:
\begingroup
\expandafter\let\csname#1\endcsname \relax
\let\definedummyword\unmacrodo
\xdef\macrolist{\macrolist}%
\endgroup
\else
\errmessage{Macro #1 not defined}%
\fi
}
% Called by \do from \dounmacro on each macro. The idea is to omit any
% macro definitions that have been changed to \relax.
%
\def\unmacrodo#1{%
\ifx #1\relax
% remove this
\else
\noexpand\definedummyword \noexpand#1%
\fi
}
% This makes use of the obscure feature that if the last token of a
% <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\macnamexxx=#1\futurelet\nchar\braceorlinexxx}
\def\braceorlinexxx{%
\ifx\nchar\bgroup\else
\expandafter\parsearg
\fi \macnamexxx}
% @alias.
% We need some trickery to remove the optional spaces around the equal
% sign. Just make them active and then expand them all to nothing.
\def\alias{\parseargusing\obeyspaces\aliasxxx}
\def\aliasxxx #1{\aliasyyy#1\relax}
\def\aliasyyy #1=#2\relax{%
{%
\expandafter\let\obeyedspace=\empty
\addtomacrolist{#1}%
\xdef\next{\global\let\makecsname{#1}=\makecsname{#2}}%
}%
\next
}
\message{cross references,}
\newwrite\auxfile
\newif\ifhavexrefs % True if xref values are known.
\newif\ifwarnedxrefs % True if we warned once that they aren't known.
% @inforef is relatively simple.
\def\inforef #1{\inforefzzz #1,,,,**}
\def\inforefzzz #1,#2,#3,#4**{\putwordSee{} \putwordInfo{} \putwordfile{} \file{\ignorespaces #3{}},
node \samp{\ignorespaces#1{}}}
% @node's only job in TeX is to define \lastnode, which is used in
% cross-references. The @node line might or might not have commas, and
% might or might not have spaces before the first comma, like:
% @node foo , bar , ...
% We don't want such trailing spaces in the node name.
%
\parseargdef\node{\checkenv{}\donode #1 ,\finishnodeparse}
%
% also remove a trailing comma, in case of something like this:
% @node Help-Cross, , , Cross-refs
\def\donode#1 ,#2\finishnodeparse{\dodonode #1,\finishnodeparse}
\def\dodonode#1,#2\finishnodeparse{\gdef\lastnode{#1}}
\let\nwnode=\node
\let\lastnode=\empty
% Write a cross-reference definition for the current node. #1 is the
% type (Ynumbered, Yappendix, Ynothing).
%
\def\donoderef#1{%
\ifx\lastnode\empty\else
\setref{\lastnode}{#1}%
\global\let\lastnode=\empty
\fi
}
% @anchor{NAME} -- define xref target at arbitrary point.
%
\newcount\savesfregister
%
\def\savesf{\relax \ifhmode \savesfregister=\spacefactor \fi}
\def\restoresf{\relax \ifhmode \spacefactor=\savesfregister \fi}
\def\anchor#1{\savesf \setref{#1}{Ynothing}\restoresf \ignorespaces}
% \setref{NAME}{SNT} defines a cross-reference point NAME (a node or an
% anchor), which consists of three parts:
% 1) NAME-title - the current sectioning name taken from \thissection,
% or the anchor name.
% 2) NAME-snt - section number and type, passed as the SNT arg, or
% empty for anchors.
% 3) NAME-pg - the page number.
%
% This is called from \donoderef, \anchor, and \dofloat. In the case of
% floats, there is an additional part, which is not written here:
% 4) NAME-lof - the text as it should appear in a @listoffloats.
%
\def\setref#1#2{%
\pdfmkdest{#1}%
\iflinks
{%
\atdummies % preserve commands, but don't expand them
\edef\writexrdef##1##2{%
\write\auxfile{@xrdef{#1-% #1 of \setref, expanded by the \edef
##1}{##2}}% these are parameters of \writexrdef
}%
\toks0 = \expandafter{\thissection}%
\immediate \writexrdef{title}{\the\toks0 }%
\immediate \writexrdef{snt}{\csname #2\endcsname}% \Ynumbered etc.
\safewhatsit{\writexrdef{pg}{\folio}}% will be written later, during \shipout
}%
\fi
}
% @xref, @pxref, and @ref generate cross-references. For \xrefX, #1 is
% the node name, #2 the name of the Info cross-reference, #3 the printed
% node name, #4 the name of the Info file, #5 the name of the printed
% manual. All but the node name can be omitted.
%
\def\pxref#1{\putwordsee{} \xrefX[#1,,,,,,,]}
\def\xref#1{\putwordSee{} \xrefX[#1,,,,,,,]}
\def\ref#1{\xrefX[#1,,,,,,,]}
\def\xrefX[#1,#2,#3,#4,#5,#6]{\begingroup
\unsepspaces
\def\printedmanual{\ignorespaces #5}%
\def\printedrefname{\ignorespaces #3}%
\setbox1=\hbox{\printedmanual\unskip}%
\setbox0=\hbox{\printedrefname\unskip}%
\ifdim \wd0 = 0pt
% No printed node name was explicitly given.
\expandafter\ifx\csname SETxref-automatic-section-title\endcsname\relax
% Use the node name inside the square brackets.
\def\printedrefname{\ignorespaces #1}%
\else
% Use the actual chapter/section title appear inside
% the square brackets. Use the real section title if we have it.
\ifdim \wd1 > 0pt
% It is in another manual, so we don't have it.
\def\printedrefname{\ignorespaces #1}%
\else
\ifhavexrefs
% We know the real title if we have the xref values.
\def\printedrefname{\refx{#1-title}{}}%
\else
% Otherwise just copy the Info node name.
\def\printedrefname{\ignorespaces #1}%
\fi%
\fi
\fi
\fi
%
% Make link in pdf output.
\ifpdf
\leavevmode
\getfilename{#4}%
{\indexnofonts
\turnoffactive
% See comments at \activebackslashdouble.
{\activebackslashdouble \xdef\pdfxrefdest{#1}%
\backslashparens\pdfxrefdest}%
%
\ifnum\filenamelength>0
\startlink attr{/Border [0 0 0]}%
goto file{\the\filename.pdf} name{\pdfxrefdest}%
\else
\startlink attr{/Border [0 0 0]}%
goto name{\pdfmkpgn{\pdfxrefdest}}%
\fi
}%
\linkcolor
\fi
%
% Float references are printed completely differently: "Figure 1.2"
% instead of "[somenode], p.3". We distinguish them by the
% LABEL-title being set to a magic string.
{%
% Have to otherify everything special to allow the \csname to
% include an _ in the xref name, etc.
\indexnofonts
\turnoffactive
\expandafter\global\expandafter\let\expandafter\Xthisreftitle
\csname XR#1-title\endcsname
}%
\iffloat\Xthisreftitle
% If the user specified the print name (third arg) to the ref,
% print it instead of our usual "Figure 1.2".
\ifdim\wd0 = 0pt
\refx{#1-snt}{}%
\else
\printedrefname
\fi
%
% if the user also gave the printed manual name (fifth arg), append
% "in MANUALNAME".
\ifdim \wd1 > 0pt
\space \putwordin{} \cite{\printedmanual}%
\fi
\else
% node/anchor (non-float) references.
%
% If we use \unhbox0 and \unhbox1 to print the node names, TeX does not
% insert empty discretionaries after hyphens, which means that it will
% not find a line break at a hyphen in a node names. Since some manuals
% are best written with fairly long node names, containing hyphens, this
% is a loss. Therefore, we give the text of the node name again, so it
% is as if TeX is seeing it for the first time.
\ifdim \wd1 > 0pt
\putwordsection{} ``\printedrefname'' \putwordin{} \cite{\printedmanual}%
\else
% _ (for example) has to be the character _ for the purposes of the
% control sequence corresponding to the node, but it has to expand
% into the usual \leavevmode...\vrule stuff for purposes of
% printing. So we \turnoffactive for the \refx-snt, back on for the
% printing, back off for the \refx-pg.
{\turnoffactive
% Only output a following space if the -snt ref is nonempty; for
% @unnumbered and @anchor, it won't be.
\setbox2 = \hbox{\ignorespaces \refx{#1-snt}{}}%
\ifdim \wd2 > 0pt \refx{#1-snt}\space\fi
}%
% output the `[mynode]' via a macro so it can be overridden.
\xrefprintnodename\printedrefname
%
% But we always want a comma and a space:
,\space
%
% output the `page 3'.
\turnoffactive \putwordpage\tie\refx{#1-pg}{}%
\fi
\fi
\endlink
\endgroup}
% This macro is called from \xrefX for the `[nodename]' part of xref
% output. It's a separate macro only so it can be changed more easily,
% since square brackets don't work well in some documents. Particularly
% one that Bob is working on :).
%
\def\xrefprintnodename#1{[#1]}
% Things referred to by \setref.
%
\def\Ynothing{}
\def\Yomitfromtoc{}
\def\Ynumbered{%
\ifnum\secno=0
\putwordChapter@tie \the\chapno
\else \ifnum\subsecno=0
\putwordSection@tie \the\chapno.\the\secno
\else \ifnum\subsubsecno=0
\putwordSection@tie \the\chapno.\the\secno.\the\subsecno
\else
\putwordSection@tie \the\chapno.\the\secno.\the\subsecno.\the\subsubsecno
\fi\fi\fi
}
\def\Yappendix{%
\ifnum\secno=0
\putwordAppendix@tie @char\the\appendixno{}%
\else \ifnum\subsecno=0
\putwordSection@tie @char\the\appendixno.\the\secno
\else \ifnum\subsubsecno=0
\putwordSection@tie @char\the\appendixno.\the\secno.\the\subsecno
\else
\putwordSection@tie
@char\the\appendixno.\the\secno.\the\subsecno.\the\subsubsecno
\fi\fi\fi
}
% Define \refx{NAME}{SUFFIX} to reference a cross-reference string named NAME.
% If its value is nonempty, SUFFIX is output afterward.
%
\def\refx#1#2{%
{%
\indexnofonts
\otherbackslash
\expandafter\global\expandafter\let\expandafter\thisrefX
\csname XR#1\endcsname
}%
\ifx\thisrefX\relax
% If not defined, say something at least.
\angleleft un\-de\-fined\angleright
\iflinks
\ifhavexrefs
\message{\linenumber Undefined cross reference `#1'.}%
\else
\ifwarnedxrefs\else
\global\warnedxrefstrue
\message{Cross reference values unknown; you must run TeX again.}%
\fi
\fi
\fi
\else
% It's defined, so just use it.
\thisrefX
\fi
#2% Output the suffix in any case.
}
% This is the macro invoked by entries in the aux file. Usually it's
% just a \def (we prepend XR to the control sequence name to avoid
% collisions). But if this is a float type, we have more work to do.
%
\def\xrdef#1#2{%
{% The node name might contain 8-bit characters, which in our current
% implementation are changed to commands like @'e. Don't let these
% mess up the control sequence name.
\indexnofonts
\turnoffactive
\xdef\safexrefname{#1}%
}%
%
\expandafter\gdef\csname XR\safexrefname\endcsname{#2}% remember this xref
%
% Was that xref control sequence that we just defined for a float?
\expandafter\iffloat\csname XR\safexrefname\endcsname
% it was a float, and we have the (safe) float type in \iffloattype.
\expandafter\let\expandafter\floatlist
\csname floatlist\iffloattype\endcsname
%
% Is this the first time we've seen this float type?
\expandafter\ifx\floatlist\relax
\toks0 = {\do}% yes, so just \do
\else
% had it before, so preserve previous elements in list.
\toks0 = \expandafter{\floatlist\do}%
\fi
%
% Remember this xref in the control sequence \floatlistFLOATTYPE,
% for later use in \listoffloats.
\expandafter\xdef\csname floatlist\iffloattype\endcsname{\the\toks0
{\safexrefname}}%
\fi
}
% Read the last existing aux file, if any. No error if none exists.
%
\def\tryauxfile{%
\openin 1 \jobname.aux
\ifeof 1 \else
\readdatafile{aux}%
\global\havexrefstrue
\fi
\closein 1
}
\def\setupdatafile{%
\catcode`\^^@=\other
\catcode`\^^A=\other
\catcode`\^^B=\other
\catcode`\^^C=\other
\catcode`\^^D=\other
\catcode`\^^E=\other
\catcode`\^^F=\other
\catcode`\^^G=\other
\catcode`\^^H=\other
\catcode`\^^K=\other
\catcode`\^^L=\other
\catcode`\^^N=\other
\catcode`\^^P=\other
\catcode`\^^Q=\other
\catcode`\^^R=\other
\catcode`\^^S=\other
\catcode`\^^T=\other
\catcode`\^^U=\other
\catcode`\^^V=\other
\catcode`\^^W=\other
\catcode`\^^X=\other
\catcode`\^^Z=\other
\catcode`\^^[=\other
\catcode`\^^\=\other
\catcode`\^^]=\other
\catcode`\^^^=\other
\catcode`\^^_=\other
% It was suggested to set the catcode of ^ to 7, which would allow ^^e4 etc.
% in xref tags, i.e., node names. But since ^^e4 notation isn't
% supported in the main text, it doesn't seem desirable. Furthermore,
% that is not enough: for node names that actually contain a ^
% character, we would end up writing a line like this: 'xrdef {'hat
% b-title}{'hat b} and \xrdef does a \csname...\endcsname on the first
% argument, and \hat is not an expandable control sequence. It could
% all be worked out, but why? Either we support ^^ or we don't.
%
% The other change necessary for this was to define \auxhat:
% \def\auxhat{\def^{'hat }}% extra space so ok if followed by letter
% and then to call \auxhat in \setq.
%
\catcode`\^=\other
%
% Special characters. Should be turned off anyway, but...
\catcode`\~=\other
\catcode`\[=\other
\catcode`\]=\other
\catcode`\"=\other
\catcode`\_=\other
\catcode`\|=\other
\catcode`\<=\other
\catcode`\>=\other
\catcode`\$=\other
\catcode`\#=\other
\catcode`\&=\other
\catcode`\%=\other
\catcode`+=\other % avoid \+ for paranoia even though we've turned it off
%
% This is to support \ in node names and titles, since the \
% characters end up in a \csname. It's easier than
% leaving it active and making its active definition an actual \
% character. What I don't understand is why it works in the *value*
% of the xrdef. Seems like it should be a catcode12 \, and that
% should not typeset properly. But it works, so I'm moving on for
% now. --karl, 15jan04.
\catcode`\\=\other
%
% Make the characters 128-255 be printing characters.
{%
\count1=128
\def\loop{%
\catcode\count1=\other
\advance\count1 by 1
\ifnum \count1<256 \loop \fi
}%
}%
%
% @ is our escape character in .aux files, and we need braces.
\catcode`\{=1
\catcode`\}=2
\catcode`\@=0
}
\def\readdatafile#1{%
\begingroup
\setupdatafile
\input\jobname.#1
\endgroup}
\message{insertions,}
% including footnotes.
\newcount \footnoteno
% The trailing space in the following definition for supereject is
% vital for proper filling; pages come out unaligned when you do a
% pagealignmacro call if that space before the closing brace is
% removed. (Generally, numeric constants should always be followed by a
% space to prevent strange expansion errors.)
\def\supereject{\par\penalty -20000\footnoteno =0 }
% @footnotestyle is meaningful for info output only.
\let\footnotestyle=\comment
{\catcode `\@=11
%
% Auto-number footnotes. Otherwise like plain.
\gdef\footnote{%
\let\indent=\ptexindent
\let\noindent=\ptexnoindent
\global\advance\footnoteno by \@ne
\edef\thisfootno{$^{\the\footnoteno}$}%
%
% In case the footnote comes at the end of a sentence, preserve the
% extra spacing after we do the footnote number.
\let\@sf\empty
\ifhmode\edef\@sf{\spacefactor\the\spacefactor}\ptexslash\fi
%
% Remove inadvertent blank space before typesetting the footnote number.
\unskip
\thisfootno\@sf
\dofootnote
}%
% Don't bother with the trickery in plain.tex to not require the
% footnote text as a parameter. Our footnotes don't need to be so general.
%
% Oh yes, they do; otherwise, @ifset (and anything else that uses
% \parseargline) fails inside footnotes because the tokens are fixed when
% the footnote is read. --karl, 16nov96.
%
\gdef\dofootnote{%
\insert\footins\bgroup
% We want to typeset this text as a normal paragraph, even if the
% footnote reference occurs in (for example) a display environment.
% So reset some parameters.
\hsize=\pagewidth
\interlinepenalty\interfootnotelinepenalty
\splittopskip\ht\strutbox % top baseline for broken footnotes
\splitmaxdepth\dp\strutbox
\floatingpenalty\@MM
\leftskip\z@skip
\rightskip\z@skip
\spaceskip\z@skip
\xspaceskip\z@skip
\parindent\defaultparindent
%
\smallfonts \rm
%
% Because we use hanging indentation in footnotes, a @noindent appears
% to exdent this text, so make it be a no-op. makeinfo does not use
% hanging indentation so @noindent can still be needed within footnote
% text after an @example or the like (not that this is good style).
\let\noindent = \relax
%
% Hang the footnote text off the number. Use \everypar in case the
% footnote extends for more than one paragraph.
\everypar = {\hang}%
\textindent{\thisfootno}%
%
% Don't crash into the line above the footnote text. Since this
% expands into a box, it must come within the paragraph, lest it
% provide a place where TeX can split the footnote.
\footstrut
\futurelet\next\fo@t
}
}%end \catcode `\@=11
% In case a @footnote appears in a vbox, save the footnote text and create
% the real \insert just after the vbox finished. Otherwise, the insertion
% would be lost.
-% Similarily, if a @footnote appears inside an alignment, save the footnote
+% Similarly, if a @footnote appears inside an alignment, save the footnote
% text to a box and make the \insert when a row of the table is finished.
% And the same can be done for other insert classes. --kasal, 16nov03.
% Replace the \insert primitive by a cheating macro.
% Deeper inside, just make sure that the saved insertions are not spilled
% out prematurely.
%
\def\startsavinginserts{%
\ifx \insert\ptexinsert
\let\insert\saveinsert
\else
\let\checkinserts\relax
\fi
}
% This \insert replacement works for both \insert\footins{foo} and
% \insert\footins\bgroup foo\egroup, but it doesn't work for \insert27{foo}.
%
\def\saveinsert#1{%
\edef\next{\noexpand\savetobox \makeSAVEname#1}%
\afterassignment\next
% swallow the left brace
\let\temp =
}
\def\makeSAVEname#1{\makecsname{SAVE\expandafter\gobble\string#1}}
\def\savetobox#1{\global\setbox#1 = \vbox\bgroup \unvbox#1}
\def\checksaveins#1{\ifvoid#1\else \placesaveins#1\fi}
\def\placesaveins#1{%
\ptexinsert \csname\expandafter\gobblesave\string#1\endcsname
{\box#1}%
}
% eat @SAVE -- beware, all of them have catcode \other:
{
\def\dospecials{\do S\do A\do V\do E} \uncatcodespecials % ;-)
\gdef\gobblesave @SAVE{}
}
% initialization:
\def\newsaveins #1{%
\edef\next{\noexpand\newsaveinsX \makeSAVEname#1}%
\next
}
\def\newsaveinsX #1{%
\csname newbox\endcsname #1%
\expandafter\def\expandafter\checkinserts\expandafter{\checkinserts
\checksaveins #1}%
}
% initialize:
\let\checkinserts\empty
\newsaveins\footins
\newsaveins\margin
% @image. We use the macros from epsf.tex to support this.
% If epsf.tex is not installed and @image is used, we complain.
%
% Check for and read epsf.tex up front. If we read it only at @image
% time, we might be inside a group, and then its definitions would get
% undone and the next image would fail.
\openin 1 = epsf.tex
\ifeof 1 \else
% Do not bother showing banner with epsf.tex v2.7k (available in
% doc/epsf.tex and on ctan).
\def\epsfannounce{\toks0 = }%
\input epsf.tex
\fi
\closein 1
%
% We will only complain once about lack of epsf.tex.
\newif\ifwarnednoepsf
\newhelp\noepsfhelp{epsf.tex must be installed for images to
work. It is also included in the Texinfo distribution, or you can get
it from ftp://tug.org/tex/epsf.tex.}
%
\def\image#1{%
\ifx\epsfbox\undefined
\ifwarnednoepsf \else
\errhelp = \noepsfhelp
\errmessage{epsf.tex not found, images will be ignored}%
\global\warnednoepsftrue
\fi
\else
\imagexxx #1,,,,,\finish
\fi
}
%
% Arguments to @image:
% #1 is (mandatory) image filename; we tack on .eps extension.
% #2 is (optional) width, #3 is (optional) height.
% #4 is (ignored optional) html alt text.
% #5 is (ignored optional) extension.
% #6 is just the usual extra ignored arg for parsing this stuff.
\newif\ifimagevmode
\def\imagexxx#1,#2,#3,#4,#5,#6\finish{\begingroup
\catcode`\^^M = 5 % in case we're inside an example
\normalturnoffactive % allow _ et al. in names
% If the image is by itself, center it.
\ifvmode
\imagevmodetrue
\nobreak\bigskip
% Usually we'll have text after the image which will insert
% \parskip glue, so insert it here too to equalize the space
% above and below.
\nobreak\vskip\parskip
\nobreak
\line\bgroup
\fi
%
% Output the image.
\ifpdf
\dopdfimage{#1}{#2}{#3}%
\else
% \epsfbox itself resets \epsf?size at each figure.
\setbox0 = \hbox{\ignorespaces #2}\ifdim\wd0 > 0pt \epsfxsize=#2\relax \fi
\setbox0 = \hbox{\ignorespaces #3}\ifdim\wd0 > 0pt \epsfysize=#3\relax \fi
\epsfbox{#1.eps}%
\fi
%
\ifimagevmode \egroup \bigbreak \fi % space after the image
\endgroup}
% @float FLOATTYPE,LABEL,LOC ... @end float for displayed figures, tables,
% etc. We don't actually implement floating yet, we always include the
% float "here". But it seemed the best name for the future.
%
\envparseargdef\float{\eatcommaspace\eatcommaspace\dofloat#1, , ,\finish}
% There may be a space before second and/or third parameter; delete it.
\def\eatcommaspace#1, {#1,}
% #1 is the optional FLOATTYPE, the text label for this float, typically
% "Figure", "Table", "Example", etc. Can't contain commas. If omitted,
% this float will not be numbered and cannot be referred to.
%
% #2 is the optional xref label. Also must be present for the float to
% be referable.
%
% #3 is the optional positioning argument; for now, it is ignored. It
% will somehow specify the positions allowed to float to (here, top, bottom).
%
% We keep a separate counter for each FLOATTYPE, which we reset at each
% chapter-level command.
\let\resetallfloatnos=\empty
%
\def\dofloat#1,#2,#3,#4\finish{%
\let\thiscaption=\empty
\let\thisshortcaption=\empty
%
% don't lose footnotes inside @float.
%
% BEWARE: when the floats start float, we have to issue warning whenever an
% insert appears inside a float which could possibly float. --kasal, 26may04
%
\startsavinginserts
%
% We can't be used inside a paragraph.
\par
%
\vtop\bgroup
\def\floattype{#1}%
\def\floatlabel{#2}%
\def\floatloc{#3}% we do nothing with this yet.
%
\ifx\floattype\empty
\let\safefloattype=\empty
\else
{%
% the floattype might have accents or other special characters,
% but we need to use it in a control sequence name.
\indexnofonts
\turnoffactive
\xdef\safefloattype{\floattype}%
}%
\fi
%
% If label is given but no type, we handle that as the empty type.
\ifx\floatlabel\empty \else
% We want each FLOATTYPE to be numbered separately (Figure 1,
% Table 1, Figure 2, ...). (And if no label, no number.)
%
\expandafter\getfloatno\csname\safefloattype floatno\endcsname
\global\advance\floatno by 1
%
{%
% This magic value for \thissection is output by \setref as the
% XREFLABEL-title value. \xrefX uses it to distinguish float
% labels (which have a completely different output format) from
% node and anchor labels. And \xrdef uses it to construct the
% lists of floats.
%
\edef\thissection{\floatmagic=\safefloattype}%
\setref{\floatlabel}{Yfloat}%
}%
\fi
%
% start with \parskip glue, I guess.
\vskip\parskip
%
% Don't suppress indentation if a float happens to start a section.
\restorefirstparagraphindent
}
% we have these possibilities:
% @float Foo,lbl & @caption{Cap}: Foo 1.1: Cap
% @float Foo,lbl & no caption: Foo 1.1
% @float Foo & @caption{Cap}: Foo: Cap
% @float Foo & no caption: Foo
% @float ,lbl & Caption{Cap}: 1.1: Cap
% @float ,lbl & no caption: 1.1
% @float & @caption{Cap}: Cap
% @float & no caption:
%
\def\Efloat{%
\let\floatident = \empty
%
% In all cases, if we have a float type, it comes first.
\ifx\floattype\empty \else \def\floatident{\floattype}\fi
%
% If we have an xref label, the number comes next.
\ifx\floatlabel\empty \else
\ifx\floattype\empty \else % if also had float type, need tie first.
\appendtomacro\floatident{\tie}%
\fi
% the number.
\appendtomacro\floatident{\chaplevelprefix\the\floatno}%
\fi
%
% Start the printed caption with what we've constructed in
% \floatident, but keep it separate; we need \floatident again.
\let\captionline = \floatident
%
\ifx\thiscaption\empty \else
\ifx\floatident\empty \else
\appendtomacro\captionline{: }% had ident, so need a colon between
\fi
%
% caption text.
\appendtomacro\captionline{\scanexp\thiscaption}%
\fi
%
% If we have anything to print, print it, with space before.
% Eventually this needs to become an \insert.
\ifx\captionline\empty \else
\vskip.5\parskip
\captionline
%
% Space below caption.
\vskip\parskip
\fi
%
% If have an xref label, write the list of floats info. Do this
% after the caption, to avoid chance of it being a breakpoint.
\ifx\floatlabel\empty \else
% Write the text that goes in the lof to the aux file as
% \floatlabel-lof. Besides \floatident, we include the short
% caption if specified, else the full caption if specified, else nothing.
{%
\atdummies
%
% since we read the caption text in the macro world, where ^^M
% is turned into a normal character, we have to scan it back, so
% we don't write the literal three characters "^^M" into the aux file.
\scanexp{%
\xdef\noexpand\gtemp{%
\ifx\thisshortcaption\empty
\thiscaption
\else
\thisshortcaption
\fi
}%
}%
\immediate\write\auxfile{@xrdef{\floatlabel-lof}{\floatident
\ifx\gtemp\empty \else : \gtemp \fi}}%
}%
\fi
\egroup % end of \vtop
%
% place the captured inserts
%
% BEWARE: when the floats start floating, we have to issue warning
% whenever an insert appears inside a float which could possibly
% float. --kasal, 26may04
%
\checkinserts
}
% Append the tokens #2 to the definition of macro #1, not expanding either.
%
\def\appendtomacro#1#2{%
\expandafter\def\expandafter#1\expandafter{#1#2}%
}
% @caption, @shortcaption
%
\def\caption{\docaption\thiscaption}
\def\shortcaption{\docaption\thisshortcaption}
\def\docaption{\checkenv\float \bgroup\scanargctxt\defcaption}
\def\defcaption#1#2{\egroup \def#1{#2}}
% The parameter is the control sequence identifying the counter we are
% going to use. Create it if it doesn't exist and assign it to \floatno.
\def\getfloatno#1{%
\ifx#1\relax
% Haven't seen this figure type before.
\csname newcount\endcsname #1%
%
% Remember to reset this floatno at the next chap.
\expandafter\gdef\expandafter\resetallfloatnos
\expandafter{\resetallfloatnos #1=0 }%
\fi
\let\floatno#1%
}
% \setref calls this to get the XREFLABEL-snt value. We want an @xref
% to the FLOATLABEL to expand to "Figure 3.1". We call \setref when we
% first read the @float command.
%
\def\Yfloat{\floattype@tie \chaplevelprefix\the\floatno}%
% Magic string used for the XREFLABEL-title value, so \xrefX can
% distinguish floats from other xref types.
\def\floatmagic{!!float!!}
% #1 is the control sequence we are passed; we expand into a conditional
% which is true if #1 represents a float ref. That is, the magic
% \thissection value which we \setref above.
%
\def\iffloat#1{\expandafter\doiffloat#1==\finish}
%
% #1 is (maybe) the \floatmagic string. If so, #2 will be the
% (safe) float type for this float. We set \iffloattype to #2.
%
\def\doiffloat#1=#2=#3\finish{%
\def\temp{#1}%
\def\iffloattype{#2}%
\ifx\temp\floatmagic
}
% @listoffloats FLOATTYPE - print a list of floats like a table of contents.
%
\parseargdef\listoffloats{%
\def\floattype{#1}% floattype
{%
% the floattype might have accents or other special characters,
% but we need to use it in a control sequence name.
\indexnofonts
\turnoffactive
\xdef\safefloattype{\floattype}%
}%
%
% \xrdef saves the floats as a \do-list in \floatlistSAFEFLOATTYPE.
\expandafter\ifx\csname floatlist\safefloattype\endcsname \relax
\ifhavexrefs
% if the user said @listoffloats foo but never @float foo.
\message{\linenumber No `\safefloattype' floats to list.}%
\fi
\else
\begingroup
\leftskip=\tocindent % indent these entries like a toc
\let\do=\listoffloatsdo
\csname floatlist\safefloattype\endcsname
\endgroup
\fi
}
% This is called on each entry in a list of floats. We're passed the
% xref label, in the form LABEL-title, which is how we save it in the
% aux file. We strip off the -title and look up \XRLABEL-lof, which
% has the text we're supposed to typeset here.
%
% Figures without xref labels will not be included in the list (since
% they won't appear in the aux file).
%
\def\listoffloatsdo#1{\listoffloatsdoentry#1\finish}
\def\listoffloatsdoentry#1-title\finish{{%
% Can't fully expand XR#1-lof because it can contain anything. Just
% pass the control sequence. On the other hand, XR#1-pg is just the
% page number, and we want to fully expand that so we can get a link
% in pdf output.
\toksA = \expandafter{\csname XR#1-lof\endcsname}%
%
% use the same \entry macro we use to generate the TOC and index.
\edef\writeentry{\noexpand\entry{\the\toksA}{\csname XR#1-pg\endcsname}}%
\writeentry
}}
\message{localization,}
% @documentlanguage is usually given very early, just after
% @setfilename. If done too late, it may not override everything
% properly. Single argument is the language abbreviation.
% It would be nice if we could set up a hyphenation file here.
%
\parseargdef\documentlanguage{%
\tex % read txi-??.tex file in plain TeX.
% Read the file if it exists.
\openin 1 txi-#1.tex
\ifeof 1
\errhelp = \nolanghelp
\errmessage{Cannot read language file txi-#1.tex}%
\else
\input txi-#1.tex
\fi
\closein 1
\endgroup
}
\newhelp\nolanghelp{The given language definition file cannot be found or
is empty. Maybe you need to install it? In the current directory
should work if nowhere else does.}
% Set the catcode of characters 128 through 255 to the specified number.
%
\def\setnonasciicharscatcode#1{%
\count255=128
\loop\ifnum\count255<256
\global\catcode\count255=#1
\advance\count255 by 1
\repeat
}
% @documentencoding sets the definition of non-ASCII characters
% according to the specified encoding.
%
\parseargdef\documentencoding{%
% Encoding being declared for the document.
\def\declaredencoding{\csname #1.enc\endcsname}%
%
% Supported encodings: names converted to tokens in order to be able
% to compare them with \ifx.
\def\ascii{\csname US-ASCII.enc\endcsname}%
\def\latnine{\csname ISO-8859-15.enc\endcsname}%
\def\latone{\csname ISO-8859-1.enc\endcsname}%
\def\lattwo{\csname ISO-8859-2.enc\endcsname}%
\def\utfeight{\csname UTF-8.enc\endcsname}%
%
\ifx \declaredencoding \ascii
\asciichardefs
%
\else \ifx \declaredencoding \lattwo
\setnonasciicharscatcode\active
\lattwochardefs
%
\else \ifx \declaredencoding \latone
\setnonasciicharscatcode\active
\latonechardefs
%
\else \ifx \declaredencoding \latnine
\setnonasciicharscatcode\active
\latninechardefs
%
\else \ifx \declaredencoding \utfeight
\setnonasciicharscatcode\active
\utfeightchardefs
%
\else
\message{Unknown document encoding #1, ignoring.}%
%
\fi % utfeight
\fi % latnine
\fi % latone
\fi % lattwo
\fi % ascii
}
% A message to be logged when using a character that isn't available
% the default font encoding (OT1).
%
\def\missingcharmsg#1{\message{Character missing in OT1 encoding: #1.}}
% Take account of \c (plain) vs. \, (Texinfo) difference.
\def\cedilla#1{\ifx\c\ptexc\c{#1}\else\,{#1}\fi}
% First, make active non-ASCII characters in order for them to be
% correctly categorized when TeX reads the replacement text of
% macros containing the character definitions.
\setnonasciicharscatcode\active
%
% Latin1 (ISO-8859-1) character definitions.
\def\latonechardefs{%
\gdef^^a0{~}
\gdef^^a1{\exclamdown}
\gdef^^a2{\missingcharmsg{CENT SIGN}}
\gdef^^a3{{\pounds}}
\gdef^^a4{\missingcharmsg{CURRENCY SIGN}}
\gdef^^a5{\missingcharmsg{YEN SIGN}}
\gdef^^a6{\missingcharmsg{BROKEN BAR}}
\gdef^^a7{\S}
\gdef^^a8{\"{}}
\gdef^^a9{\copyright}
\gdef^^aa{\ordf}
\gdef^^ab{\missingcharmsg{LEFT-POINTING DOUBLE ANGLE QUOTATION MARK}}
\gdef^^ac{$\lnot$}
\gdef^^ad{\-}
\gdef^^ae{\registeredsymbol}
\gdef^^af{\={}}
%
\gdef^^b0{\textdegree}
\gdef^^b1{$\pm$}
\gdef^^b2{$^2$}
\gdef^^b3{$^3$}
\gdef^^b4{\'{}}
\gdef^^b5{$\mu$}
\gdef^^b6{\P}
%
\gdef^^b7{$^.$}
\gdef^^b8{\cedilla\ }
\gdef^^b9{$^1$}
\gdef^^ba{\ordm}
%
\gdef^^bb{\missingcharmsg{RIGHT-POINTING DOUBLE ANGLE QUOTATION MARK}}
\gdef^^bc{$1\over4$}
\gdef^^bd{$1\over2$}
\gdef^^be{$3\over4$}
\gdef^^bf{\questiondown}
%
\gdef^^c0{\`A}
\gdef^^c1{\'A}
\gdef^^c2{\^A}
\gdef^^c3{\~A}
\gdef^^c4{\"A}
\gdef^^c5{\ringaccent A}
\gdef^^c6{\AE}
\gdef^^c7{\cedilla C}
\gdef^^c8{\`E}
\gdef^^c9{\'E}
\gdef^^ca{\^E}
\gdef^^cb{\"E}
\gdef^^cc{\`I}
\gdef^^cd{\'I}
\gdef^^ce{\^I}
\gdef^^cf{\"I}
%
\gdef^^d0{\missingcharmsg{LATIN CAPITAL LETTER ETH}}
\gdef^^d1{\~N}
\gdef^^d2{\`O}
\gdef^^d3{\'O}
\gdef^^d4{\^O}
\gdef^^d5{\~O}
\gdef^^d6{\"O}
\gdef^^d7{$\times$}
\gdef^^d8{\O}
\gdef^^d9{\`U}
\gdef^^da{\'U}
\gdef^^db{\^U}
\gdef^^dc{\"U}
\gdef^^dd{\'Y}
\gdef^^de{\missingcharmsg{LATIN CAPITAL LETTER THORN}}
\gdef^^df{\ss}
%
\gdef^^e0{\`a}
\gdef^^e1{\'a}
\gdef^^e2{\^a}
\gdef^^e3{\~a}
\gdef^^e4{\"a}
\gdef^^e5{\ringaccent a}
\gdef^^e6{\ae}
\gdef^^e7{\cedilla c}
\gdef^^e8{\`e}
\gdef^^e9{\'e}
\gdef^^ea{\^e}
\gdef^^eb{\"e}
\gdef^^ec{\`{\dotless i}}
\gdef^^ed{\'{\dotless i}}
\gdef^^ee{\^{\dotless i}}
\gdef^^ef{\"{\dotless i}}
%
\gdef^^f0{\missingcharmsg{LATIN SMALL LETTER ETH}}
\gdef^^f1{\~n}
\gdef^^f2{\`o}
\gdef^^f3{\'o}
\gdef^^f4{\^o}
\gdef^^f5{\~o}
\gdef^^f6{\"o}
\gdef^^f7{$\div$}
\gdef^^f8{\o}
\gdef^^f9{\`u}
\gdef^^fa{\'u}
\gdef^^fb{\^u}
\gdef^^fc{\"u}
\gdef^^fd{\'y}
\gdef^^fe{\missingcharmsg{LATIN SMALL LETTER THORN}}
\gdef^^ff{\"y}
}
% Latin9 (ISO-8859-15) encoding character definitions.
\def\latninechardefs{%
% Encoding is almost identical to Latin1.
\latonechardefs
%
\gdef^^a4{\euro}
\gdef^^a6{\v S}
\gdef^^a8{\v s}
\gdef^^b4{\v Z}
\gdef^^b8{\v z}
\gdef^^bc{\OE}
\gdef^^bd{\oe}
\gdef^^be{\"Y}
}
% Latin2 (ISO-8859-2) character definitions.
\def\lattwochardefs{%
\gdef^^a0{~}
\gdef^^a1{\missingcharmsg{LATIN CAPITAL LETTER A WITH OGONEK}}
\gdef^^a2{\u{}}
\gdef^^a3{\L}
\gdef^^a4{\missingcharmsg{CURRENCY SIGN}}
\gdef^^a5{\v L}
\gdef^^a6{\'S}
\gdef^^a7{\S}
\gdef^^a8{\"{}}
\gdef^^a9{\v S}
\gdef^^aa{\cedilla S}
\gdef^^ab{\v T}
\gdef^^ac{\'Z}
\gdef^^ad{\-}
\gdef^^ae{\v Z}
\gdef^^af{\dotaccent Z}
%
\gdef^^b0{\textdegree}
\gdef^^b1{\missingcharmsg{LATIN SMALL LETTER A WITH OGONEK}}
\gdef^^b2{\missingcharmsg{OGONEK}}
\gdef^^b3{\l}
\gdef^^b4{\'{}}
\gdef^^b5{\v l}
\gdef^^b6{\'s}
\gdef^^b7{\v{}}
\gdef^^b8{\cedilla\ }
\gdef^^b9{\v s}
\gdef^^ba{\cedilla s}
\gdef^^bb{\v t}
\gdef^^bc{\'z}
\gdef^^bd{\H{}}
\gdef^^be{\v z}
\gdef^^bf{\dotaccent z}
%
\gdef^^c0{\'R}
\gdef^^c1{\'A}
\gdef^^c2{\^A}
\gdef^^c3{\u A}
\gdef^^c4{\"A}
\gdef^^c5{\'L}
\gdef^^c6{\'C}
\gdef^^c7{\cedilla C}
\gdef^^c8{\v C}
\gdef^^c9{\'E}
\gdef^^ca{\missingcharmsg{LATIN CAPITAL LETTER E WITH OGONEK}}
\gdef^^cb{\"E}
\gdef^^cc{\v E}
\gdef^^cd{\'I}
\gdef^^ce{\^I}
\gdef^^cf{\v D}
%
\gdef^^d0{\missingcharmsg{LATIN CAPITAL LETTER D WITH STROKE}}
\gdef^^d1{\'N}
\gdef^^d2{\v N}
\gdef^^d3{\'O}
\gdef^^d4{\^O}
\gdef^^d5{\H O}
\gdef^^d6{\"O}
\gdef^^d7{$\times$}
\gdef^^d8{\v R}
\gdef^^d9{\ringaccent U}
\gdef^^da{\'U}
\gdef^^db{\H U}
\gdef^^dc{\"U}
\gdef^^dd{\'Y}
\gdef^^de{\cedilla T}
\gdef^^df{\ss}
%
\gdef^^e0{\'r}
\gdef^^e1{\'a}
\gdef^^e2{\^a}
\gdef^^e3{\u a}
\gdef^^e4{\"a}
\gdef^^e5{\'l}
\gdef^^e6{\'c}
\gdef^^e7{\cedilla c}
\gdef^^e8{\v c}
\gdef^^e9{\'e}
\gdef^^ea{\missingcharmsg{LATIN SMALL LETTER E WITH OGONEK}}
\gdef^^eb{\"e}
\gdef^^ec{\v e}
\gdef^^ed{\'\i}
\gdef^^ee{\^\i}
\gdef^^ef{\v d}
%
\gdef^^f0{\missingcharmsg{LATIN SMALL LETTER D WITH STROKE}}
\gdef^^f1{\'n}
\gdef^^f2{\v n}
\gdef^^f3{\'o}
\gdef^^f4{\^o}
\gdef^^f5{\H o}
\gdef^^f6{\"o}
\gdef^^f7{$\div$}
\gdef^^f8{\v r}
\gdef^^f9{\ringaccent u}
\gdef^^fa{\'u}
\gdef^^fb{\H u}
\gdef^^fc{\"u}
\gdef^^fd{\'y}
\gdef^^fe{\cedilla t}
\gdef^^ff{\dotaccent{}}
}
% UTF-8 character definitions.
%
% This code to support UTF-8 is based on LaTeX's utf8.def, with some
% changes for Texinfo conventions. It is included here under the GPL by
% permission from Frank Mittelbach and the LaTeX team.
%
\newcount\countUTFx
\newcount\countUTFy
\newcount\countUTFz
\gdef\UTFviiiTwoOctets#1#2{\expandafter
\UTFviiiDefined\csname u8:#1\string #2\endcsname}
%
\gdef\UTFviiiThreeOctets#1#2#3{\expandafter
\UTFviiiDefined\csname u8:#1\string #2\string #3\endcsname}
%
\gdef\UTFviiiFourOctets#1#2#3#4{\expandafter
\UTFviiiDefined\csname u8:#1\string #2\string #3\string #4\endcsname}
\gdef\UTFviiiDefined#1{%
\ifx #1\relax
\message{\linenumber Unicode char \string #1 not defined for Texinfo}%
\else
\expandafter #1%
\fi
}
\begingroup
\catcode`\~13
\catcode`\"12
\def\UTFviiiLoop{%
\global\catcode\countUTFx\active
\uccode`\~\countUTFx
\uppercase\expandafter{\UTFviiiTmp}%
\advance\countUTFx by 1
\ifnum\countUTFx < \countUTFy
\expandafter\UTFviiiLoop
\fi}
\countUTFx = "C2
\countUTFy = "E0
\def\UTFviiiTmp{%
\xdef~{\noexpand\UTFviiiTwoOctets\string~}}
\UTFviiiLoop
\countUTFx = "E0
\countUTFy = "F0
\def\UTFviiiTmp{%
\xdef~{\noexpand\UTFviiiThreeOctets\string~}}
\UTFviiiLoop
\countUTFx = "F0
\countUTFy = "F4
\def\UTFviiiTmp{%
\xdef~{\noexpand\UTFviiiFourOctets\string~}}
\UTFviiiLoop
\endgroup
\begingroup
\catcode`\"=12
\catcode`\<=12
\catcode`\.=12
\catcode`\,=12
\catcode`\;=12
\catcode`\!=12
\catcode`\~=13
\gdef\DeclareUnicodeCharacter#1#2{%
\countUTFz = "#1\relax
\wlog{\space\space defining Unicode char U+#1 (decimal \the\countUTFz)}%
\begingroup
\parseXMLCharref
\def\UTFviiiTwoOctets##1##2{%
\csname u8:##1\string ##2\endcsname}%
\def\UTFviiiThreeOctets##1##2##3{%
\csname u8:##1\string ##2\string ##3\endcsname}%
\def\UTFviiiFourOctets##1##2##3##4{%
\csname u8:##1\string ##2\string ##3\string ##4\endcsname}%
\expandafter\expandafter\expandafter\expandafter
\expandafter\expandafter\expandafter
\gdef\UTFviiiTmp{#2}%
\endgroup}
\gdef\parseXMLCharref{%
\ifnum\countUTFz < "A0\relax
\errhelp = \EMsimple
\errmessage{Cannot define Unicode char value < 00A0}%
\else\ifnum\countUTFz < "800\relax
\parseUTFviiiA,%
\parseUTFviiiB C\UTFviiiTwoOctets.,%
\else\ifnum\countUTFz < "10000\relax
\parseUTFviiiA;%
\parseUTFviiiA,%
\parseUTFviiiB E\UTFviiiThreeOctets.{,;}%
\else
\parseUTFviiiA;%
\parseUTFviiiA,%
\parseUTFviiiA!%
\parseUTFviiiB F\UTFviiiFourOctets.{!,;}%
\fi\fi\fi
}
\gdef\parseUTFviiiA#1{%
\countUTFx = \countUTFz
\divide\countUTFz by 64
\countUTFy = \countUTFz
\multiply\countUTFz by 64
\advance\countUTFx by -\countUTFz
\advance\countUTFx by 128
\uccode `#1\countUTFx
\countUTFz = \countUTFy}
\gdef\parseUTFviiiB#1#2#3#4{%
\advance\countUTFz by "#10\relax
\uccode `#3\countUTFz
\uppercase{\gdef\UTFviiiTmp{#2#3#4}}}
\endgroup
\def\utfeightchardefs{%
\DeclareUnicodeCharacter{00A0}{\tie}
\DeclareUnicodeCharacter{00A1}{\exclamdown}
\DeclareUnicodeCharacter{00A3}{\pounds}
\DeclareUnicodeCharacter{00A8}{\"{ }}
\DeclareUnicodeCharacter{00A9}{\copyright}
\DeclareUnicodeCharacter{00AA}{\ordf}
\DeclareUnicodeCharacter{00AD}{\-}
\DeclareUnicodeCharacter{00AE}{\registeredsymbol}
\DeclareUnicodeCharacter{00AF}{\={ }}
\DeclareUnicodeCharacter{00B0}{\ringaccent{ }}
\DeclareUnicodeCharacter{00B4}{\'{ }}
\DeclareUnicodeCharacter{00B8}{\cedilla{ }}
\DeclareUnicodeCharacter{00BA}{\ordm}
\DeclareUnicodeCharacter{00BF}{\questiondown}
\DeclareUnicodeCharacter{00C0}{\`A}
\DeclareUnicodeCharacter{00C1}{\'A}
\DeclareUnicodeCharacter{00C2}{\^A}
\DeclareUnicodeCharacter{00C3}{\~A}
\DeclareUnicodeCharacter{00C4}{\"A}
\DeclareUnicodeCharacter{00C5}{\AA}
\DeclareUnicodeCharacter{00C6}{\AE}
\DeclareUnicodeCharacter{00C7}{\cedilla{C}}
\DeclareUnicodeCharacter{00C8}{\`E}
\DeclareUnicodeCharacter{00C9}{\'E}
\DeclareUnicodeCharacter{00CA}{\^E}
\DeclareUnicodeCharacter{00CB}{\"E}
\DeclareUnicodeCharacter{00CC}{\`I}
\DeclareUnicodeCharacter{00CD}{\'I}
\DeclareUnicodeCharacter{00CE}{\^I}
\DeclareUnicodeCharacter{00CF}{\"I}
\DeclareUnicodeCharacter{00D1}{\~N}
\DeclareUnicodeCharacter{00D2}{\`O}
\DeclareUnicodeCharacter{00D3}{\'O}
\DeclareUnicodeCharacter{00D4}{\^O}
\DeclareUnicodeCharacter{00D5}{\~O}
\DeclareUnicodeCharacter{00D6}{\"O}
\DeclareUnicodeCharacter{00D8}{\O}
\DeclareUnicodeCharacter{00D9}{\`U}
\DeclareUnicodeCharacter{00DA}{\'U}
\DeclareUnicodeCharacter{00DB}{\^U}
\DeclareUnicodeCharacter{00DC}{\"U}
\DeclareUnicodeCharacter{00DD}{\'Y}
\DeclareUnicodeCharacter{00DF}{\ss}
\DeclareUnicodeCharacter{00E0}{\`a}
\DeclareUnicodeCharacter{00E1}{\'a}
\DeclareUnicodeCharacter{00E2}{\^a}
\DeclareUnicodeCharacter{00E3}{\~a}
\DeclareUnicodeCharacter{00E4}{\"a}
\DeclareUnicodeCharacter{00E5}{\aa}
\DeclareUnicodeCharacter{00E6}{\ae}
\DeclareUnicodeCharacter{00E7}{\cedilla{c}}
\DeclareUnicodeCharacter{00E8}{\`e}
\DeclareUnicodeCharacter{00E9}{\'e}
\DeclareUnicodeCharacter{00EA}{\^e}
\DeclareUnicodeCharacter{00EB}{\"e}
\DeclareUnicodeCharacter{00EC}{\`{\dotless{i}}}
\DeclareUnicodeCharacter{00ED}{\'{\dotless{i}}}
\DeclareUnicodeCharacter{00EE}{\^{\dotless{i}}}
\DeclareUnicodeCharacter{00EF}{\"{\dotless{i}}}
\DeclareUnicodeCharacter{00F1}{\~n}
\DeclareUnicodeCharacter{00F2}{\`o}
\DeclareUnicodeCharacter{00F3}{\'o}
\DeclareUnicodeCharacter{00F4}{\^o}
\DeclareUnicodeCharacter{00F5}{\~o}
\DeclareUnicodeCharacter{00F6}{\"o}
\DeclareUnicodeCharacter{00F8}{\o}
\DeclareUnicodeCharacter{00F9}{\`u}
\DeclareUnicodeCharacter{00FA}{\'u}
\DeclareUnicodeCharacter{00FB}{\^u}
\DeclareUnicodeCharacter{00FC}{\"u}
\DeclareUnicodeCharacter{00FD}{\'y}
\DeclareUnicodeCharacter{00FF}{\"y}
\DeclareUnicodeCharacter{0100}{\=A}
\DeclareUnicodeCharacter{0101}{\=a}
\DeclareUnicodeCharacter{0102}{\u{A}}
\DeclareUnicodeCharacter{0103}{\u{a}}
\DeclareUnicodeCharacter{0106}{\'C}
\DeclareUnicodeCharacter{0107}{\'c}
\DeclareUnicodeCharacter{0108}{\^C}
\DeclareUnicodeCharacter{0109}{\^c}
\DeclareUnicodeCharacter{010A}{\dotaccent{C}}
\DeclareUnicodeCharacter{010B}{\dotaccent{c}}
\DeclareUnicodeCharacter{010C}{\v{C}}
\DeclareUnicodeCharacter{010D}{\v{c}}
\DeclareUnicodeCharacter{010E}{\v{D}}
\DeclareUnicodeCharacter{0112}{\=E}
\DeclareUnicodeCharacter{0113}{\=e}
\DeclareUnicodeCharacter{0114}{\u{E}}
\DeclareUnicodeCharacter{0115}{\u{e}}
\DeclareUnicodeCharacter{0116}{\dotaccent{E}}
\DeclareUnicodeCharacter{0117}{\dotaccent{e}}
\DeclareUnicodeCharacter{011A}{\v{E}}
\DeclareUnicodeCharacter{011B}{\v{e}}
\DeclareUnicodeCharacter{011C}{\^G}
\DeclareUnicodeCharacter{011D}{\^g}
\DeclareUnicodeCharacter{011E}{\u{G}}
\DeclareUnicodeCharacter{011F}{\u{g}}
\DeclareUnicodeCharacter{0120}{\dotaccent{G}}
\DeclareUnicodeCharacter{0121}{\dotaccent{g}}
\DeclareUnicodeCharacter{0124}{\^H}
\DeclareUnicodeCharacter{0125}{\^h}
\DeclareUnicodeCharacter{0128}{\~I}
\DeclareUnicodeCharacter{0129}{\~{\dotless{i}}}
\DeclareUnicodeCharacter{012A}{\=I}
\DeclareUnicodeCharacter{012B}{\={\dotless{i}}}
\DeclareUnicodeCharacter{012C}{\u{I}}
\DeclareUnicodeCharacter{012D}{\u{\dotless{i}}}
\DeclareUnicodeCharacter{0130}{\dotaccent{I}}
\DeclareUnicodeCharacter{0131}{\dotless{i}}
\DeclareUnicodeCharacter{0132}{IJ}
\DeclareUnicodeCharacter{0133}{ij}
\DeclareUnicodeCharacter{0134}{\^J}
\DeclareUnicodeCharacter{0135}{\^{\dotless{j}}}
\DeclareUnicodeCharacter{0139}{\'L}
\DeclareUnicodeCharacter{013A}{\'l}
\DeclareUnicodeCharacter{0141}{\L}
\DeclareUnicodeCharacter{0142}{\l}
\DeclareUnicodeCharacter{0143}{\'N}
\DeclareUnicodeCharacter{0144}{\'n}
\DeclareUnicodeCharacter{0147}{\v{N}}
\DeclareUnicodeCharacter{0148}{\v{n}}
\DeclareUnicodeCharacter{014C}{\=O}
\DeclareUnicodeCharacter{014D}{\=o}
\DeclareUnicodeCharacter{014E}{\u{O}}
\DeclareUnicodeCharacter{014F}{\u{o}}
\DeclareUnicodeCharacter{0150}{\H{O}}
\DeclareUnicodeCharacter{0151}{\H{o}}
\DeclareUnicodeCharacter{0152}{\OE}
\DeclareUnicodeCharacter{0153}{\oe}
\DeclareUnicodeCharacter{0154}{\'R}
\DeclareUnicodeCharacter{0155}{\'r}
\DeclareUnicodeCharacter{0158}{\v{R}}
\DeclareUnicodeCharacter{0159}{\v{r}}
\DeclareUnicodeCharacter{015A}{\'S}
\DeclareUnicodeCharacter{015B}{\'s}
\DeclareUnicodeCharacter{015C}{\^S}
\DeclareUnicodeCharacter{015D}{\^s}
\DeclareUnicodeCharacter{015E}{\cedilla{S}}
\DeclareUnicodeCharacter{015F}{\cedilla{s}}
\DeclareUnicodeCharacter{0160}{\v{S}}
\DeclareUnicodeCharacter{0161}{\v{s}}
\DeclareUnicodeCharacter{0162}{\cedilla{t}}
\DeclareUnicodeCharacter{0163}{\cedilla{T}}
\DeclareUnicodeCharacter{0164}{\v{T}}
\DeclareUnicodeCharacter{0168}{\~U}
\DeclareUnicodeCharacter{0169}{\~u}
\DeclareUnicodeCharacter{016A}{\=U}
\DeclareUnicodeCharacter{016B}{\=u}
\DeclareUnicodeCharacter{016C}{\u{U}}
\DeclareUnicodeCharacter{016D}{\u{u}}
\DeclareUnicodeCharacter{016E}{\ringaccent{U}}
\DeclareUnicodeCharacter{016F}{\ringaccent{u}}
\DeclareUnicodeCharacter{0170}{\H{U}}
\DeclareUnicodeCharacter{0171}{\H{u}}
\DeclareUnicodeCharacter{0174}{\^W}
\DeclareUnicodeCharacter{0175}{\^w}
\DeclareUnicodeCharacter{0176}{\^Y}
\DeclareUnicodeCharacter{0177}{\^y}
\DeclareUnicodeCharacter{0178}{\"Y}
\DeclareUnicodeCharacter{0179}{\'Z}
\DeclareUnicodeCharacter{017A}{\'z}
\DeclareUnicodeCharacter{017B}{\dotaccent{Z}}
\DeclareUnicodeCharacter{017C}{\dotaccent{z}}
\DeclareUnicodeCharacter{017D}{\v{Z}}
\DeclareUnicodeCharacter{017E}{\v{z}}
\DeclareUnicodeCharacter{01C4}{D\v{Z}}
\DeclareUnicodeCharacter{01C5}{D\v{z}}
\DeclareUnicodeCharacter{01C6}{d\v{z}}
\DeclareUnicodeCharacter{01C7}{LJ}
\DeclareUnicodeCharacter{01C8}{Lj}
\DeclareUnicodeCharacter{01C9}{lj}
\DeclareUnicodeCharacter{01CA}{NJ}
\DeclareUnicodeCharacter{01CB}{Nj}
\DeclareUnicodeCharacter{01CC}{nj}
\DeclareUnicodeCharacter{01CD}{\v{A}}
\DeclareUnicodeCharacter{01CE}{\v{a}}
\DeclareUnicodeCharacter{01CF}{\v{I}}
\DeclareUnicodeCharacter{01D0}{\v{\dotless{i}}}
\DeclareUnicodeCharacter{01D1}{\v{O}}
\DeclareUnicodeCharacter{01D2}{\v{o}}
\DeclareUnicodeCharacter{01D3}{\v{U}}
\DeclareUnicodeCharacter{01D4}{\v{u}}
\DeclareUnicodeCharacter{01E2}{\={\AE}}
\DeclareUnicodeCharacter{01E3}{\={\ae}}
\DeclareUnicodeCharacter{01E6}{\v{G}}
\DeclareUnicodeCharacter{01E7}{\v{g}}
\DeclareUnicodeCharacter{01E8}{\v{K}}
\DeclareUnicodeCharacter{01E9}{\v{k}}
\DeclareUnicodeCharacter{01F0}{\v{\dotless{j}}}
\DeclareUnicodeCharacter{01F1}{DZ}
\DeclareUnicodeCharacter{01F2}{Dz}
\DeclareUnicodeCharacter{01F3}{dz}
\DeclareUnicodeCharacter{01F4}{\'G}
\DeclareUnicodeCharacter{01F5}{\'g}
\DeclareUnicodeCharacter{01F8}{\`N}
\DeclareUnicodeCharacter{01F9}{\`n}
\DeclareUnicodeCharacter{01FC}{\'{\AE}}
\DeclareUnicodeCharacter{01FD}{\'{\ae}}
\DeclareUnicodeCharacter{01FE}{\'{\O}}
\DeclareUnicodeCharacter{01FF}{\'{\o}}
\DeclareUnicodeCharacter{021E}{\v{H}}
\DeclareUnicodeCharacter{021F}{\v{h}}
\DeclareUnicodeCharacter{0226}{\dotaccent{A}}
\DeclareUnicodeCharacter{0227}{\dotaccent{a}}
\DeclareUnicodeCharacter{0228}{\cedilla{E}}
\DeclareUnicodeCharacter{0229}{\cedilla{e}}
\DeclareUnicodeCharacter{022E}{\dotaccent{O}}
\DeclareUnicodeCharacter{022F}{\dotaccent{o}}
\DeclareUnicodeCharacter{0232}{\=Y}
\DeclareUnicodeCharacter{0233}{\=y}
\DeclareUnicodeCharacter{0237}{\dotless{j}}
\DeclareUnicodeCharacter{1E02}{\dotaccent{B}}
\DeclareUnicodeCharacter{1E03}{\dotaccent{b}}
\DeclareUnicodeCharacter{1E04}{\udotaccent{B}}
\DeclareUnicodeCharacter{1E05}{\udotaccent{b}}
\DeclareUnicodeCharacter{1E06}{\ubaraccent{B}}
\DeclareUnicodeCharacter{1E07}{\ubaraccent{b}}
\DeclareUnicodeCharacter{1E0A}{\dotaccent{D}}
\DeclareUnicodeCharacter{1E0B}{\dotaccent{d}}
\DeclareUnicodeCharacter{1E0C}{\udotaccent{D}}
\DeclareUnicodeCharacter{1E0D}{\udotaccent{d}}
\DeclareUnicodeCharacter{1E0E}{\ubaraccent{D}}
\DeclareUnicodeCharacter{1E0F}{\ubaraccent{d}}
\DeclareUnicodeCharacter{1E1E}{\dotaccent{F}}
\DeclareUnicodeCharacter{1E1F}{\dotaccent{f}}
\DeclareUnicodeCharacter{1E20}{\=G}
\DeclareUnicodeCharacter{1E21}{\=g}
\DeclareUnicodeCharacter{1E22}{\dotaccent{H}}
\DeclareUnicodeCharacter{1E23}{\dotaccent{h}}
\DeclareUnicodeCharacter{1E24}{\udotaccent{H}}
\DeclareUnicodeCharacter{1E25}{\udotaccent{h}}
\DeclareUnicodeCharacter{1E26}{\"H}
\DeclareUnicodeCharacter{1E27}{\"h}
\DeclareUnicodeCharacter{1E30}{\'K}
\DeclareUnicodeCharacter{1E31}{\'k}
\DeclareUnicodeCharacter{1E32}{\udotaccent{K}}
\DeclareUnicodeCharacter{1E33}{\udotaccent{k}}
\DeclareUnicodeCharacter{1E34}{\ubaraccent{K}}
\DeclareUnicodeCharacter{1E35}{\ubaraccent{k}}
\DeclareUnicodeCharacter{1E36}{\udotaccent{L}}
\DeclareUnicodeCharacter{1E37}{\udotaccent{l}}
\DeclareUnicodeCharacter{1E3A}{\ubaraccent{L}}
\DeclareUnicodeCharacter{1E3B}{\ubaraccent{l}}
\DeclareUnicodeCharacter{1E3E}{\'M}
\DeclareUnicodeCharacter{1E3F}{\'m}
\DeclareUnicodeCharacter{1E40}{\dotaccent{M}}
\DeclareUnicodeCharacter{1E41}{\dotaccent{m}}
\DeclareUnicodeCharacter{1E42}{\udotaccent{M}}
\DeclareUnicodeCharacter{1E43}{\udotaccent{m}}
\DeclareUnicodeCharacter{1E44}{\dotaccent{N}}
\DeclareUnicodeCharacter{1E45}{\dotaccent{n}}
\DeclareUnicodeCharacter{1E46}{\udotaccent{N}}
\DeclareUnicodeCharacter{1E47}{\udotaccent{n}}
\DeclareUnicodeCharacter{1E48}{\ubaraccent{N}}
\DeclareUnicodeCharacter{1E49}{\ubaraccent{n}}
\DeclareUnicodeCharacter{1E54}{\'P}
\DeclareUnicodeCharacter{1E55}{\'p}
\DeclareUnicodeCharacter{1E56}{\dotaccent{P}}
\DeclareUnicodeCharacter{1E57}{\dotaccent{p}}
\DeclareUnicodeCharacter{1E58}{\dotaccent{R}}
\DeclareUnicodeCharacter{1E59}{\dotaccent{r}}
\DeclareUnicodeCharacter{1E5A}{\udotaccent{R}}
\DeclareUnicodeCharacter{1E5B}{\udotaccent{r}}
\DeclareUnicodeCharacter{1E5E}{\ubaraccent{R}}
\DeclareUnicodeCharacter{1E5F}{\ubaraccent{r}}
\DeclareUnicodeCharacter{1E60}{\dotaccent{S}}
\DeclareUnicodeCharacter{1E61}{\dotaccent{s}}
\DeclareUnicodeCharacter{1E62}{\udotaccent{S}}
\DeclareUnicodeCharacter{1E63}{\udotaccent{s}}
\DeclareUnicodeCharacter{1E6A}{\dotaccent{T}}
\DeclareUnicodeCharacter{1E6B}{\dotaccent{t}}
\DeclareUnicodeCharacter{1E6C}{\udotaccent{T}}
\DeclareUnicodeCharacter{1E6D}{\udotaccent{t}}
\DeclareUnicodeCharacter{1E6E}{\ubaraccent{T}}
\DeclareUnicodeCharacter{1E6F}{\ubaraccent{t}}
\DeclareUnicodeCharacter{1E7C}{\~V}
\DeclareUnicodeCharacter{1E7D}{\~v}
\DeclareUnicodeCharacter{1E7E}{\udotaccent{V}}
\DeclareUnicodeCharacter{1E7F}{\udotaccent{v}}
\DeclareUnicodeCharacter{1E80}{\`W}
\DeclareUnicodeCharacter{1E81}{\`w}
\DeclareUnicodeCharacter{1E82}{\'W}
\DeclareUnicodeCharacter{1E83}{\'w}
\DeclareUnicodeCharacter{1E84}{\"W}
\DeclareUnicodeCharacter{1E85}{\"w}
\DeclareUnicodeCharacter{1E86}{\dotaccent{W}}
\DeclareUnicodeCharacter{1E87}{\dotaccent{w}}
\DeclareUnicodeCharacter{1E88}{\udotaccent{W}}
\DeclareUnicodeCharacter{1E89}{\udotaccent{w}}
\DeclareUnicodeCharacter{1E8A}{\dotaccent{X}}
\DeclareUnicodeCharacter{1E8B}{\dotaccent{x}}
\DeclareUnicodeCharacter{1E8C}{\"X}
\DeclareUnicodeCharacter{1E8D}{\"x}
\DeclareUnicodeCharacter{1E8E}{\dotaccent{Y}}
\DeclareUnicodeCharacter{1E8F}{\dotaccent{y}}
\DeclareUnicodeCharacter{1E90}{\^Z}
\DeclareUnicodeCharacter{1E91}{\^z}
\DeclareUnicodeCharacter{1E92}{\udotaccent{Z}}
\DeclareUnicodeCharacter{1E93}{\udotaccent{z}}
\DeclareUnicodeCharacter{1E94}{\ubaraccent{Z}}
\DeclareUnicodeCharacter{1E95}{\ubaraccent{z}}
\DeclareUnicodeCharacter{1E96}{\ubaraccent{h}}
\DeclareUnicodeCharacter{1E97}{\"t}
\DeclareUnicodeCharacter{1E98}{\ringaccent{w}}
\DeclareUnicodeCharacter{1E99}{\ringaccent{y}}
\DeclareUnicodeCharacter{1EA0}{\udotaccent{A}}
\DeclareUnicodeCharacter{1EA1}{\udotaccent{a}}
\DeclareUnicodeCharacter{1EB8}{\udotaccent{E}}
\DeclareUnicodeCharacter{1EB9}{\udotaccent{e}}
\DeclareUnicodeCharacter{1EBC}{\~E}
\DeclareUnicodeCharacter{1EBD}{\~e}
\DeclareUnicodeCharacter{1ECA}{\udotaccent{I}}
\DeclareUnicodeCharacter{1ECB}{\udotaccent{i}}
\DeclareUnicodeCharacter{1ECC}{\udotaccent{O}}
\DeclareUnicodeCharacter{1ECD}{\udotaccent{o}}
\DeclareUnicodeCharacter{1EE4}{\udotaccent{U}}
\DeclareUnicodeCharacter{1EE5}{\udotaccent{u}}
\DeclareUnicodeCharacter{1EF2}{\`Y}
\DeclareUnicodeCharacter{1EF3}{\`y}
\DeclareUnicodeCharacter{1EF4}{\udotaccent{Y}}
\DeclareUnicodeCharacter{1EF8}{\~Y}
\DeclareUnicodeCharacter{1EF9}{\~y}
\DeclareUnicodeCharacter{2013}{--}
\DeclareUnicodeCharacter{2014}{---}
\DeclareUnicodeCharacter{2022}{\bullet}
\DeclareUnicodeCharacter{2026}{\dots}
\DeclareUnicodeCharacter{20AC}{\euro}
\DeclareUnicodeCharacter{2192}{\expansion}
\DeclareUnicodeCharacter{21D2}{\result}
\DeclareUnicodeCharacter{2212}{\minus}
\DeclareUnicodeCharacter{2217}{\point}
\DeclareUnicodeCharacter{2261}{\equiv}
}% end of \utfeightchardefs
% US-ASCII character definitions.
\def\asciichardefs{% nothing need be done
\relax
}
% Make non-ASCII characters printable again for compatibility with
% existing Texinfo documents that may use them, even without declaring a
% document encoding.
%
\setnonasciicharscatcode \other
\message{formatting,}
\newdimen\defaultparindent \defaultparindent = 15pt
\chapheadingskip = 15pt plus 4pt minus 2pt
\secheadingskip = 12pt plus 3pt minus 2pt
\subsecheadingskip = 9pt plus 2pt minus 2pt
% Prevent underfull vbox error messages.
\vbadness = 10000
% Don't be so finicky about underfull hboxes, either.
\hbadness = 2000
% Following George Bush, just get rid of widows and orphans.
\widowpenalty=10000
\clubpenalty=10000
% Use TeX 3.0's \emergencystretch to help line breaking, but if we're
% using an old version of TeX, don't do anything. We want the amount of
% stretch added to depend on the line length, hence the dependence on
% \hsize. We call this whenever the paper size is set.
%
\def\setemergencystretch{%
\ifx\emergencystretch\thisisundefined
% Allow us to assign to \emergencystretch anyway.
\def\emergencystretch{\dimen0}%
\else
\emergencystretch = .15\hsize
\fi
}
% Parameters in order: 1) textheight; 2) textwidth;
% 3) voffset; 4) hoffset; 5) binding offset; 6) topskip;
% 7) physical page height; 8) physical page width.
%
% We also call \setleading{\textleading}, so the caller should define
% \textleading. The caller should also set \parskip.
%
\def\internalpagesizes#1#2#3#4#5#6#7#8{%
\voffset = #3\relax
\topskip = #6\relax
\splittopskip = \topskip
%
\vsize = #1\relax
\advance\vsize by \topskip
\outervsize = \vsize
\advance\outervsize by 2\topandbottommargin
\pageheight = \vsize
%
\hsize = #2\relax
\outerhsize = \hsize
\advance\outerhsize by 0.5in
\pagewidth = \hsize
%
\normaloffset = #4\relax
\bindingoffset = #5\relax
%
\ifpdf
\pdfpageheight #7\relax
\pdfpagewidth #8\relax
\fi
%
\setleading{\textleading}
%
\parindent = \defaultparindent
\setemergencystretch
}
% @letterpaper (the default).
\def\letterpaper{{\globaldefs = 1
\parskip = 3pt plus 2pt minus 1pt
\textleading = 13.2pt
%
% If page is nothing but text, make it come out even.
\internalpagesizes{46\baselineskip}{6in}%
{\voffset}{.25in}%
{\bindingoffset}{36pt}%
{11in}{8.5in}%
}}
% Use @smallbook to reset parameters for 7x9.25 trim size.
\def\smallbook{{\globaldefs = 1
\parskip = 2pt plus 1pt
\textleading = 12pt
%
\internalpagesizes{7.5in}{5in}%
{\voffset}{.25in}%
{\bindingoffset}{16pt}%
{9.25in}{7in}%
%
\lispnarrowing = 0.3in
\tolerance = 700
\hfuzz = 1pt
\contentsrightmargin = 0pt
\defbodyindent = .5cm
}}
% Use @smallerbook to reset parameters for 6x9 trim size.
% (Just testing, parameters still in flux.)
\def\smallerbook{{\globaldefs = 1
\parskip = 1.5pt plus 1pt
\textleading = 12pt
%
\internalpagesizes{7.4in}{4.8in}%
{-.2in}{-.4in}%
{0pt}{14pt}%
{9in}{6in}%
%
\lispnarrowing = 0.25in
\tolerance = 700
\hfuzz = 1pt
\contentsrightmargin = 0pt
\defbodyindent = .4cm
}}
% Use @afourpaper to print on European A4 paper.
\def\afourpaper{{\globaldefs = 1
\parskip = 3pt plus 2pt minus 1pt
\textleading = 13.2pt
%
% Double-side printing via postscript on Laserjet 4050
% prints double-sided nicely when \bindingoffset=10mm and \hoffset=-6mm.
% To change the settings for a different printer or situation, adjust
% \normaloffset until the front-side and back-side texts align. Then
% do the same for \bindingoffset. You can set these for testing in
% your texinfo source file like this:
% @tex
% \global\normaloffset = -6mm
% \global\bindingoffset = 10mm
% @end tex
\internalpagesizes{51\baselineskip}{160mm}
{\voffset}{\hoffset}%
{\bindingoffset}{44pt}%
{297mm}{210mm}%
%
\tolerance = 700
\hfuzz = 1pt
\contentsrightmargin = 0pt
\defbodyindent = 5mm
}}
% Use @afivepaper to print on European A5 paper.
% From romildo@urano.iceb.ufop.br, 2 July 2000.
% He also recommends making @example and @lisp be small.
\def\afivepaper{{\globaldefs = 1
\parskip = 2pt plus 1pt minus 0.1pt
\textleading = 12.5pt
%
\internalpagesizes{160mm}{120mm}%
{\voffset}{\hoffset}%
{\bindingoffset}{8pt}%
{210mm}{148mm}%
%
\lispnarrowing = 0.2in
\tolerance = 800
\hfuzz = 1.2pt
\contentsrightmargin = 0pt
\defbodyindent = 2mm
\tableindent = 12mm
}}
% A specific text layout, 24x15cm overall, intended for A4 paper.
\def\afourlatex{{\globaldefs = 1
\afourpaper
\internalpagesizes{237mm}{150mm}%
{\voffset}{4.6mm}%
{\bindingoffset}{7mm}%
{297mm}{210mm}%
%
% Must explicitly reset to 0 because we call \afourpaper.
\globaldefs = 0
}}
% Use @afourwide to print on A4 paper in landscape format.
\def\afourwide{{\globaldefs = 1
\afourpaper
\internalpagesizes{241mm}{165mm}%
{\voffset}{-2.95mm}%
{\bindingoffset}{7mm}%
{297mm}{210mm}%
\globaldefs = 0
}}
% @pagesizes TEXTHEIGHT[,TEXTWIDTH]
% Perhaps we should allow setting the margins, \topskip, \parskip,
% and/or leading, also. Or perhaps we should compute them somehow.
%
\parseargdef\pagesizes{\pagesizesyyy #1,,\finish}
\def\pagesizesyyy#1,#2,#3\finish{{%
\setbox0 = \hbox{\ignorespaces #2}\ifdim\wd0 > 0pt \hsize=#2\relax \fi
\globaldefs = 1
%
\parskip = 3pt plus 2pt minus 1pt
\setleading{\textleading}%
%
\dimen0 = #1
\advance\dimen0 by \voffset
%
\dimen2 = \hsize
\advance\dimen2 by \normaloffset
%
\internalpagesizes{#1}{\hsize}%
{\voffset}{\normaloffset}%
{\bindingoffset}{44pt}%
{\dimen0}{\dimen2}%
}}
% Set default to letter.
%
\letterpaper
\message{and turning on texinfo input format.}
% Define macros to output various characters with catcode for normal text.
\catcode`\"=\other
\catcode`\~=\other
\catcode`\^=\other
\catcode`\_=\other
\catcode`\|=\other
\catcode`\<=\other
\catcode`\>=\other
\catcode`\+=\other
\catcode`\$=\other
\def\normaldoublequote{"}
\def\normaltilde{~}
\def\normalcaret{^}
\def\normalunderscore{_}
\def\normalverticalbar{|}
\def\normalless{<}
\def\normalgreater{>}
\def\normalplus{+}
\def\normaldollar{$}%$ font-lock fix
% This macro is used to make a character print one way in \tt
% (where it can probably be output as-is), and another way in other fonts,
% where something hairier probably needs to be done.
%
% #1 is what to print if we are indeed using \tt; #2 is what to print
% otherwise. Since all the Computer Modern typewriter fonts have zero
% interword stretch (and shrink), and it is reasonable to expect all
% typewriter fonts to have this, we can check that font parameter.
%
\def\ifusingtt#1#2{\ifdim \fontdimen3\font=0pt #1\else #2\fi}
% Same as above, but check for italic font. Actually this also catches
% non-italic slanted fonts since it is impossible to distinguish them from
% italic fonts. But since this is only used by $ and it uses \sl anyway
% this is not a problem.
\def\ifusingit#1#2{\ifdim \fontdimen1\font>0pt #1\else #2\fi}
% Turn off all special characters except @
% (and those which the user can use as if they were ordinary).
% Most of these we simply print from the \tt font, but for some, we can
% use math or other variants that look better in normal text.
\catcode`\"=\active
\def\activedoublequote{{\tt\char34}}
\let"=\activedoublequote
\catcode`\~=\active
\def~{{\tt\char126}}
\chardef\hat=`\^
\catcode`\^=\active
\def^{{\tt \hat}}
\catcode`\_=\active
\def_{\ifusingtt\normalunderscore\_}
\let\realunder=_
% Subroutine for the previous macro.
\def\_{\leavevmode \kern.07em \vbox{\hrule width.3em height.1ex}\kern .07em }
\catcode`\|=\active
\def|{{\tt\char124}}
\chardef \less=`\<
\catcode`\<=\active
\def<{{\tt \less}}
\chardef \gtr=`\>
\catcode`\>=\active
\def>{{\tt \gtr}}
\catcode`\+=\active
\def+{{\tt \char 43}}
\catcode`\$=\active
\def${\ifusingit{{\sl\$}}\normaldollar}%$ font-lock fix
% If a .fmt file is being used, characters that might appear in a file
% name cannot be active until we have parsed the command line.
% So turn them off again, and have \everyjob (or @setfilename) turn them on.
% \otherifyactive is called near the end of this file.
\def\otherifyactive{\catcode`+=\other \catcode`\_=\other}
% Used sometimes to turn off (effectively) the active characters even after
% parsing them.
\def\turnoffactive{%
\normalturnoffactive
\otherbackslash
}
\catcode`\@=0
% \backslashcurfont outputs one backslash character in current font,
% as in \char`\\.
\global\chardef\backslashcurfont=`\\
\global\let\rawbackslashxx=\backslashcurfont % let existing .??s files work
% \realbackslash is an actual character `\' with catcode other, and
% \doublebackslash is two of them (for the pdf outlines).
{\catcode`\\=\other @gdef@realbackslash{\} @gdef@doublebackslash{\\}}
% In texinfo, backslash is an active character; it prints the backslash
% in fixed width font.
\catcode`\\=\active
@def@normalbackslash{{@tt@backslashcurfont}}
% On startup, @fixbackslash assigns:
% @let \ = @normalbackslash
% \rawbackslash defines an active \ to do \backslashcurfont.
% \otherbackslash defines an active \ to be a literal `\' character with
% catcode other.
@gdef@rawbackslash{@let\=@backslashcurfont}
@gdef@otherbackslash{@let\=@realbackslash}
% Same as @turnoffactive except outputs \ as {\tt\char`\\} instead of
% the literal character `\'.
%
@def@normalturnoffactive{%
@let\=@normalbackslash
@let"=@normaldoublequote
@let~=@normaltilde
@let^=@normalcaret
@let_=@normalunderscore
@let|=@normalverticalbar
@let<=@normalless
@let>=@normalgreater
@let+=@normalplus
@let$=@normaldollar %$ font-lock fix
@unsepspaces
}
% Make _ and + \other characters, temporarily.
% This is canceled by @fixbackslash.
@otherifyactive
% If a .fmt file is being used, we don't want the `\input texinfo' to show up.
% That is what \eatinput is for; after that, the `\' should revert to printing
% a backslash.
%
@gdef@eatinput input texinfo{@fixbackslash}
@global@let\ = @eatinput
% On the other hand, perhaps the file did not have a `\input texinfo'. Then
% the first `\' in the file would cause an error. This macro tries to fix
% that, assuming it is called before the first `\' could plausibly occur.
% Also turn back on active characters that might appear in the input
% file name, in case not using a pre-dumped format.
%
@gdef@fixbackslash{%
@ifx\@eatinput @let\ = @normalbackslash @fi
@catcode`+=@active
@catcode`@_=@active
}
% Say @foo, not \foo, in error messages.
@escapechar = `@@
% These look ok in all fonts, so just make them not special.
@catcode`@& = @other
@catcode`@# = @other
@catcode`@% = @other
@c Local variables:
@c eval: (add-hook 'write-file-hooks 'time-stamp)
@c page-delimiter: "^\\\\message"
@c time-stamp-start: "def\\\\texinfoversion{"
@c time-stamp-format: "%:y-%02m-%02d.%02H"
@c time-stamp-end: "}"
@c End:
@c vim:sw=2:
@ignore
arch-tag: e1b36e32-c96e-4135-a41a-0b2efa2ea115
@end ignore
diff --git a/doc/pinentry.texi b/doc/pinentry.texi
index dcff886..e9da4ee 100644
--- a/doc/pinentry.texi
+++ b/doc/pinentry.texi
@@ -1,757 +1,757 @@
\input texinfo @c -*-texinfo-*-
@c %**start of header
@setfilename pinentry.info
@include version.texi
@macro copyrightnotice
Copyright @copyright{} 2002, 2005, 2015 g10 Code GmbH
@end macro
@macro permissionnotice
Permission is granted to copy, distribute and/or modify this document
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. The text of the license can be found in the
section entitled ``Copying''.
@end macro
@macro pinentry
@sc{pinentry}
@end macro
@settitle Using the PIN-Entry
@c Create a separate index for command line options.
@defcodeindex op
@c Merge the standard indexes into a single one.
@syncodeindex fn cp
@syncodeindex vr cp
@syncodeindex ky cp
@syncodeindex pg cp
@syncodeindex tp cp
@c printing stuff taken from gcc.
@macro gnupgtabopt{body}
@code{\body\}
@end macro
@macro gnupgoptlist{body}
@smallexample
\body\
@end smallexample
@end macro
@c Makeinfo handles the above macro OK, TeX needs manual line breaks;
@c they get lost at some point in handling the macro. But if @macro is
@c used here rather than @alias, it produces double line breaks.
@iftex
@alias gol = *
@end iftex
@ifnottex
@macro gol
@end macro
@end ifnottex
@c %**end of header
@ifnottex
@dircategory GNU Utilities
@direntry
* pinentry: (pinentry). Securely ask for a passphrase or PIN.
@end direntry
This file documents the use and the internals of the @pinentry{}.
This is edition @value{EDITION}, last updated @value{UPDATED}, of
@cite{The `PINEntry' Manual}, for version @value{VERSION}.
@sp 1
Published by g10 Code GmbH@*
Hüttenstr. 61@*
40699 Erkrath, Germany
@sp 1
@copyrightnotice{}
@sp 1
@permissionnotice{}
@end ifnottex
@setchapternewpage odd
@titlepage
@title Using the PIN-Entry
@subtitle Version @value{VERSION}
@subtitle @value{UPDATED}
@author Werner Koch @code{(wk@@gnupg.org)}
@page
@vskip 0pt plus 1filll
@copyrightnotice{}
@sp 2
@permissionnotice{}
@end titlepage
@summarycontents
@contents
@page
@node Top
@top Introduction
@cindex introduction
This manual documents how to use the @pinentry{} and its protocol.
The @pinentry{} is a small GUI application used to enter PINs or
passphrases. It is usually invoked by @sc{gpg-agent}
(@pxref{Invoking GPG-AGENT, ,Invoking the gpg-agent, gnupg,
The `GNU Privacy Guard' Manual}, for details).
@pinentry{} comes in several flavors to fit the look and feel of the
used GUI toolkit: A @sc{GTK+} based one named @code{pinentry-gtk}; a
@sc{Qt} based one named @code{pinentry-qt}; and, two non-graphical
ones @code{pinentry-curses}, which uses curses, and
@code{pinentry-tty}, which doesn't require anything more than a simple
terminal. Not all of them are necessarily available on your
installation. If curses is supported on your system, the GUI-based
flavors fall back to curses when the @code{DISPLAY} variable is not
set.
@menu
* Using pinentry:: How to use the beast.
* Front ends:: Description and comparison of the front ends
Developer information
* Protocol:: The Assuan protocol description.
* Implementation Details:: For those extending or writing a new pinentry.
Miscellaneous
* Copying:: GNU General Public License says
how you can copy and share PIN-Entry
as well as this manual.
Indices
* Option Index:: Index to command line options.
* Index:: Index of concepts and symbol names.
@end menu
@node Using pinentry
@chapter How to use the @pinentry{}
@c man begin DESCRIPTION
You may run @pinentry{} directly from the command line and pass the
commands according to the Assuan protocol via stdin/stdout.
@c man end
@c man begin OPTIONS
Here is a list of options supported by all flavors of pinentry:
@table @gnupgtabopt
@item --version
@opindex version
Print the program version and licensing information.
@item --help
@opindex help
Print a usage message summarizing the most useful command line options.
@item --debug
@itemx -d
@opindex debug
@opindex d
Turn on some debugging. Mostly useful for the maintainers. Note that
this may reveal sensitive information like the entered passphrase.
@c @item --enhanced
@c @itemx -e
@c @opindex enhanced
@c @opindex e
@c Ask for timeouts and insurance, too. Note that this is currently not
@c fully supported.
@item --no-global-grab
@itemx -g
@opindex no-global-grab
@opindex g
Grab the keyboard only when the window is focused. Use this option if
you are debugging software using the @pinentry{}; otherwise you may
not be able to to access your X session anymore (unless you have other
means to connect to the machine to kill the @pinentry{}).
@item --parent-wid @var{n}
@opindex parent-wid
Use window ID @var{n} as the parent window for positioning the window.
Note, that this is not fully supported by all flavors of @pinentry{}.
@item --timeout @var{seconds}
@opindex timeout
Give up waiting for input from the user after the specified number of
seconds and return an error. The error returned is the same as if the
Cancel button was selected. To disable the timeout and wait
indefinitely, set this to 0, which is the default.
@item --display @var{string}
@itemx --ttyname @var{string}
@itemx --ttytype @var{string}
@itemx --lc-ctype @var{string}
@itemx --lc-messages @var{string}
@opindex display
@opindex ttyname
@opindex ttytype
@opindex lc-ctype
@opindex lc-messa
These options are used to pass localization information to
@pinentry{}. They are required because @pinentry{} is usually called
by some background process which does not have any information about
the locale and terminal to use. It is also possible to pass these
options using Assuan protocol options.
@end table
@node Front ends
@chapter Front Ends
There are several different flavors of @pinentry{}. Concretely, there
are Gtk+2, Qt@tie{}4, Gnome@tie{}3, Emacs, curses and tty variants.
These different implementations provide higher levels of integration
with a specific environment. For instance, the Gnome@tie{}3
@pinentry{} uses Gnome@tie{}3 widgets to display the prompts. For
Gnome@tie{}3 users, this higher level of integration provides a more
consistent aesthetic. However, this comes at a cost. Because this
@pinentry{} uses so many components, there is a larger chance of a
failure. In particular, there is a larger chance that the passphrase
is saved in memory and that memory is exposed to an attacker (consider
the OpenSSL Heartbeat vulnerability).
To understand how many components touch the passphrase, consider again
the Gnome@tie{}3 implementation. When a user presses a button on the
keyboard, the key is passed from the kernel to the X@tie{}server to
the toolkit (Gtk+) and to the actual text entry widget. Along the
way, the key is saved in memory and processed. In fact, the key
presses are probably read using standard C library functions, which
buffer the input. None of this code is careful to make sure the
contents of the memory are not leaked by keeping the data in unpagable
memory and wiping it when the buffer is freed. However, even if they
did, there is still the problem that when a computer hibernates, the
system writes unpagable memory to disk anyway. Further, many
installations are virtualized (e.g., running on Xen) and have little
control over their actual environment.
The curses variant uses a significant smaller software stack and the
tty variant uses an even smaller one. However, if they are run in an
X@tie{}terminal, then a similar number of components are handling the
passphrase as in the Gnome@tie{}3 case! Thus, to be most secure, you
need to direct GPG@tie{}Agent to use a fixed virtual console. Since
you need to remain logged in for GPG@tie{}Agent to use that console,
you should run there and have @code{screen} or @code{tmux} lock the
tty.
The Emacs pinentry implementation interacts with a running Emacs
session and directs the Emacs instance to display the passphrase
prompt. Since this doesn't work very well if there is no Emacs
running, the generic @pinentry{} backend checks if a
@pinentry{}-enabled Emacs should be used. Specifically, it looks to
see if the @code{INSIDE_EMACS} variable is set and then attempts to
establish a connection to the specified address. If this is the case,
then instead of, e.g., @code{pinentry-gtk2} displaying a Gtk+2
pinentry, it interacts with the Emacs session. This functionality can
be explicitly disabled by passing @code{--disable-inside-emacs} to
@code{configure} when building @pinentry{}.
Having Emacs get the passphrase is convenient, however, it is a
significant security risk. Emacs keeps all key presses buffered.
(You can see the recent key presses by typing @code{C-h l}
(@code{view-lossage}) in emacs.) Further, Emacs is a huge program,
which doesn't provide any process isolation to speak of. As such,
having it handle the passphrase adds a huge chunk of code to the
user's trusted computing base. Because of this concern, Emacs doesn't
enable this by default (the user has to run @code{(pinentry-start)},
e.g., from his or her @code{.emacs} file, explicitly).
Similar to the inside-emacs check, the @pinentry{} frontends check
whether the @code{DISPLAY} variable is set and a working X server is
available. If this is not the case, then they fallback to the curses
front end. This can also be disabled by passing
@code{--disable-fallback-curses} to @code{configure} at build time.
@c
@c Assuan Protocol
@c
@node Protocol
@chapter @pinentry{}'s Assuan Protocol
The @pinentry{} should never service more than one connection at once.
It is reasonable to exec the @pinentry{} prior to a request.
The @pinentry{} does not need to stay in memory because the
@sc{gpg-agent} has the ability to cache passphrases. The usual way to
run the @pinentry{} is by setting up a pipe (not a socket) and then
fork/exec the @pinentry{}. The communication is then done by means of
the protocol described here until the client is satisfied with the
result.
Although it is called a @pinentry{}, it allows entering reasonably
long strings (strings that are up to 2048 characters long are
supported by every pinentry). The client using the @pinentry{} has to
check for correctness.
Note that all strings are expected to be encoded as UTF-8; @pinentry{}
takes care of converting it to the locally used codeset. To include
linefeeds or other special characters, you may percent-escape them
(e.g., a line feed is encoded as @code{%0A}, the percent sign itself
is encoded as @code{%25}, etc.).
The following is a list of supported commands:
@table @gnupgtabopt
@item Set the timeout before returning an error
@example
C: SETTIMEOUT 30
S: OK
@end example
@item Set the descriptive text to display
@example
C: SETDESC Enter PIN for Richard Nixon <nobody@@trickydicky.gov>
S: OK
@end example
@item Set the prompt to show
When asking for a PIN, set the text just before the widget for
passphrase entry.
@example
C: SETPROMPT PIN:
S: OK
@end example
You should use an underscore in the text only if you know that a
modern version of pinentry is used. Modern versions underline the
next character after the underscore and use the first such underlined
character as a keyboard accelerator. Use a double underscore to
escape an underscore.
@item Set the window title
This command may be used to change the default window title. When
using this feature you should take care that the window is still
identifiable as the pinentry.
@example
C: SETTITLE Tape Recorder Room
S: OK
@end example
@item Set the button texts
There are three texts which should be used to override the English
defaults:
To set the text for the button signaling confirmation (in UTF-8).
See SETPROMPT on how to use an keyboard accelerator.
@example
C: SETOK Yes
S: OK
@end example
To set the text for the button signaling cancellation or disagreement
(in UTF-8). See SETPROMPT on how to use an keyboard accelerator.
@example
C: SETCANCEL No
S: OK
@end example
In case three buttons are required, use the following command to set
the text (UTF-8) for the non-affirmative response button. The
affirmative button text is still set using SETOK and the CANCEL button
text with SETCANCEL. See SETPROMPT on how to use an keyboard
accelerator.
@example
C: SETNOTOK Do not do this
S: OK
@end example
@item Set the Error text
This is used by the client to display an error message. In contrast
to the other commands, the error message is automatically reset with
a GETPIN or CONFIRM, and is only displayed when asking for a PIN.
@example
C: SETERROR Invalid PIN entered - please try again
S: OK
@end example
@item Enable a passphrase quality indicator
Adds a quality indicator to the GETPIN window. This indicator is
updated as the passphrase is typed. The clients needs to implement an
inquiry named "QUALITY" which gets passed the current passpharse
(percent-plus escaped) and should send back a string with a single
numerical vauelue between -100 and 100. Negative values will be
displayed in red.
@example
C: SETQUALITYBAR
S: OK
@end example
If a custom label for the quality bar is required, just add that label
as an argument as a percent-escaped string. You will need this
feature to translate the label because @pinentry{} has no internal
gettext except for stock strings from the toolkit library.
If you want to show a tooltip for the quality bar, you may use
@example
C: SETQUALITYBAR_TT string
S: OK
@end example
@noindent
With STRING being a percent escaped string shown as the tooltip.
@item Ask for a PIN
The meat of this tool is to ask for a passphrase of PIN, it is done with
this command:
@example
C: GETPIN
S: D no more tapes
S: OK
@end example
Note that the passphrase is transmitted in clear using standard data
responses. Expect it to be in UTF-8.
@item Ask for confirmation
To ask for a confirmation (yes or no), you can use this command:
@example
C: CONFIRM
S: OK
@end example
The client should use SETDESC to set an appropriate text before
issuing this command, and may use SETPROMPT to set the button texts.
The value returned is either OK for YES or the error code
@code{ASSUAN_Not_Confirmed}.
@item Show a message
To show a message, you can use this command:
@example
C: MESSAGE
S: OK
@end example
alternatively you may add an option to confirm:
@example
C: CONFIRM --one-button
S: OK
@end example
The client should use SETDESC to set an appropriate text before issuing
this command, and may use SETOK to set the text for the dismiss button.
The value returned is OK or an error message.
@item Set the output device
When using X, the @pinentry{} program must be invoked with an
appropriate @code{DISPLAY} environment variable or the
@code{--display} option.
When using a text terminal:
@example
C: OPTION ttyname=/dev/tty3
S: OK
C: OPTION ttytype=vt100
S: OK
C: OPTION lc-ctype=de_DE.UTF-8
S: OK
@end example
The client should use the @code{ttyname} option to set the output TTY
file name, the @code{ttytype} option to the @code{TERM} variable
appropriate for this tty and @code{lc-ctype} to the locale which
defines the character set to use for this terminal.
@item Set the default strings
To avoid having translations in Pinentry proper, the caller may set
certain translated strings which are used by @pinentry{} as default
strings.
@example
C: OPTION default-ok=_Korrekt
S: OK
C: OPTION default-cancel=Abbruch
S: OK
C: OPTION default-prompt=PIN eingeben:
S: OK
@end example
The strings are subject to accelerator marking, see SETPROMPT for
details.
@item Passphrase caching
Some environments, such as GNOME, cache passwords and passphrases.
The @pinentry{} should only use an external cache if the
@code{allow-external-password-cache} option was set and a stable key
identifier (using SETKEYINFO) was provided. In this case, if the
passphrase was read from the cache, the @pinentry{} should send the
@code{PASSWORD_FROM_CACHE} status message before returning the
passphrase. This indicates to GPG Agent that it should not increment
the passphrase retry counter.
@example
C: OPTION allow-external-password-cache
S: OK
C: SETKEYINFO key-grip
S: OK
C: getpin
S: S PASSWORD_FROM_CACHE
S: D 1234
C: OK
@end example
Note: if @code{allow-external-password-cache} is not specified, an
external password cache must not be used: this can lead to subtle
bugs. In particular, if this option is not specified, then GPG Agent
does not recognize the @code{PASSWORD_FROM_CACHE} status message and
will count trying a cached password against the password retry count.
If the password retry count is 1, then the user will never have the
opportunity to correct the cached password.
Note: it is strongly recommended that a pinentry supporting this
feature provide the user an option to enable it manually. That is,
saving a passphrase in an external password manager should be opt-in.
The key identifier provided SETKEYINFO must be considered opaque and
may change in the future. It currently has the form
@code{X/HEXSTRING} where @code{X} is either @code{n}, @code{s}, or
@code{u}. In the former two cases, the HEXSTRING corresponds to the
key grip. The key grip is not the OpenPGP Key ID, but it can be
mapped to the key using the following:
@example
# gpg2 --with-keygrip --list-secret-keys
@end example
@noindent
and searching the output for the key grip. The same command-line
options can also be used with gpgsm.
@end table
@node Implementation Details
@chapter Implementation Details
The pinentry source code can be divided into three categories. There
is a backend module, which lives in @code{pinentry/}, there are
utility functions, e.g., in @code{secmem/}, and there are various
frontends.
All of the low-level logic lives in the backend. This frees the
frontends from having to implement, e.g., the Assuan protocol. When
the backend receives an option, it updates the state in a
@code{pinentry_t} struct. The frontend is called when the client
either calls @code{GETPIN}, @code{CONFIRM} or @code{MESSAGE}. In
these cases, the backend invokes the @code{pinentry_cmd_handler},
which is passed the @code{pinentry_t} struct.
When the callback is invoked, the frontend should create a window
based on the state in the @code{pinentry_t} struct. For instance, the
title to use for the dialog's window (if any) is stored in the
@code{title} field. If the is @code{NULL}, the frontend should choose
a reasonable default value. (Default is not always provided, because
different tool kits and environments have different reasonable
defaults.)
The widget needs to support a number of different interactions with
the user. Each of them is described below.
@table @gnupgtabopt
@item Passphrase Confirmation
When creating a new key, the passphrase should be entered twice. The
client (typically GPG Agent) indicates this to the @pinentry{} by
invoking @code{SETREPEAT}. In this case, the backend sets the
@code{repeat_passphrase} field to a copy of the passed string. The
value of this field should be used to label a second text input.
It is the frontend's responsibility to check that the passwords match.
If they don't match, the frontend should display an error message and
continue to prompt the user.
If the passwords do match, then, when the user presses the okay
button, the @code{repeat_okay} field should be set to @code{1} (this
causes the backend to emit the @code{S PIN_REPEATED} status message).
@item Message Box
Sometimes GPG Agent needs to display a message. In this case, the
@code{pin} variable is @code{NULL}.
At the Assuan level, this mode is selected by using either the
@code{MESSAGE} or the @code{CONFIRM} command instead of the
@code{GETPIN} command. The @code{MESSAGE} command never shows the
cancel or an other button. The same holds for @code{CONFIRM} if it
was passed the ``--one-button'' argument. If @code{CONFIRM} was not
passed this argument, the dialog for @code{CONFIRM} should show both
the @code{ok} and the @code{cancel} buttons and optionally the
@code{notok} button. The frontend can determine whether the dialog is
a one-button dialog by inspecting the @code{one_button} variable.
@item Passphrase Entry
If neither of the above cases holds, then GPG Agent is simply
requesting the passphrase. In this case, the @code{ok} and
@code{cancel} buttons should be displayed.
@end table
The layout of the three variants is quite similar. Here are the
relevant elements that describe the layout:
@table @gnupgtabopt
@item @code{title}
The window's title.
@item @code{description}
The reason for the dialog. When requesting a passphrase, this
describes the key. When showing a message box, this is the message to
show.
@item @code{error}
If GPG Agent determines that the passphrase was incorrect, it will
call @code{GETPIN} again (up to a configurable number of times) to
again prompt the user. In this case, this variable contains a
description of the error message. This text should typically be
highlighted in someway.
@item @code{prompt}, @code{default-prompt}
The string to associate with the passphrase entry box.
There is a subtle difference between @code{prompt} and
@code{default-prompt}. @code{default-prompt} means that a stylized
prompt (e.g., an icon suggesting a prompt) may be used. @code{prompt}
means that the entry's meaning is not consistent with such a style
and, as such, no icon should be used.
If both variables are set, the @code{prompt} variant takes precedence.
@item @code{repeat_passphrase}
The string to associate with the second passphrase entry box. The
second passphrase entry box should only be shown if this is not
@code{NULL}.
@item @code{ok}, @code{default-ok}
The string to show in the @code{ok} button.
If there are any @code{_} characters, the following character should
be used as an accelerator. (A double underscore means a plain
underscore should be shown.) If the frontend does not support
accelerators, then the underscores should be removed manually.
There is a subtle difference between @code{ok} and @code{default-ok}.
@code{default-ok} means that a stylized OK button should be used. For
instance, it could include a check mark. @code{ok} means that the
button's meaning is not consistent with such an icon and, as such, no
icon should be used. Thus, if the @code{ok} button should have the
text ``No password required'' then @code{ok} should be used because a
check mark icon doesn't make sense.
If this variable is @code{NULL}, the frontend should choose a
reasonable default.
If both variables are set, the @code{ok} variant takes precedence.
@item @code{cancel}, @code{default-cancel}
Like the @code{ok} and @code{default-ok} buttons except these strings
are used for the cancel button.
This button should not be shown if @code{one_button} is set.
@code{default-notok}
Like the @code{default-ok} button except this string is used for the
other button.
This button should only be displayed when showing a message box. If
these variables are @code{NULL} or @code{one_button} is set, this
button should not be displayed.
@item @code{quality_bar}
If this is set, a widget should be used to show the password's
quality. The value of this field is a label for the widget.
Note: to update the password quality, whenever the password changes,
call the @code{pinentry_inq_quality} function and then update the
password quality widget correspondingly.
@item @code{quality_bar_tt}
A tooltip for the quality bar.
@item @code{default_pwmngr}
If @code{may_cache_password} and @code{keyinfo} are set and the user
consents, then the @pinentry{} may cache the password with an external
manager. Note: getting the user's consent is essential, because
password managers often provide a different level of security. If the
above condition is true and @code{tried_password_cache} is false, then
a check box with the specified string should be displayed. The check
box must default to off.
@item @code{default-cf-visi}
The string to show with a question if you want to confirm that
the user wants to change the visibility of the password.
@item @code{default-tt-visi}
Tooltip for an action that would reveal the entered password.
@item @code{default-tt-hide}
Tooltip for an action that would hide the password revealed
by the action labeld with @code{default-tt-visi}
@end table
When the handler is done, it should store the passphrase in
@code{pin}, if appropriate. This variable is allocated in secure
memory. Use @code{pinentry_setbufferlen} to size the buffer.
The actual return code is dependent on whether the dialog is in
message mode or in passphrase mode.
If the dialog is in message mode and the user pressed ok, return 1.
-Otherwise, return 0. If an error occured, indicate this by setting it
+Otherwise, return 0. If an error occurred, indicate this by setting it
in @code{specific_err} or setting @code{locale_err} to @code{1} (for
locale specific errors). If the dialog was canceled, then the handler
should set the @code{canceled} variable to @code{1}. If the not ok
button was pressed, don't do anything else.
If the dialog is in passphrase mode return @code{1} if the user
-entered a password and pressed ok. If an error occured, return
+entered a password and pressed ok. If an error occurred, return
@code{-1} and set @code{specific_err} or @code{locale_err}, as above.
If the user canceled the dialog box, return @code{-1}.
If the window was closed, then the handler should set the
@code{close_button} variable and otherwise act as if the cancel button
was pressed.
@c ---------------------------------------------------------------------
@c Legal Blurbs
@c ---------------------------------------------------------------------
@include gpl.texi
@c ---------------------------------------------------------------------
@c Indexes
@c ---------------------------------------------------------------------
@node Option Index
@unnumbered Option Index
@printindex op
@node Index
@unnumbered Index
@printindex cp
@c ---------------------------------------------------------------------
@c Epilogue
@c ---------------------------------------------------------------------
@bye
diff --git a/doc/texinfo.tex b/doc/texinfo.tex
index 5063065..919d85d 100644
--- a/doc/texinfo.tex
+++ b/doc/texinfo.tex
@@ -1,8638 +1,8638 @@
% texinfo.tex -- TeX macros to handle Texinfo files.
%
% Load plain if necessary, i.e., if running under initex.
\expandafter\ifx\csname fmtname\endcsname\relax\input plain\fi
%
-\def\texinfoversion{2007-05-03.09}
+\def\texinfoversion{2016-11-03.12}
%
% Copyright (C) 1985, 1986, 1988, 1990, 1991, 1992, 1993, 1994, 1995,
% 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006,
% 2007 Free Software Foundation, Inc.
%
% This texinfo.tex file is free software; you can redistribute it and/or
% modify it under the terms of the GNU General Public License as
% published by the Free Software Foundation; either version 3, or (at
% your option) any later version.
%
% This texinfo.tex file is distributed in the hope that it will be
% useful, but WITHOUT ANY WARRANTY; without even the implied warranty
% of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
% General Public License for more details.
%
% You should have received a copy of the GNU General Public License
% along with this texinfo.tex file; see the file COPYING. If not,
% see <http://www.gnu.org/licenses/>.
%
% 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}
\chardef\other=12
% We never want plain's \outer definition of \+ in Texinfo.
% For @tex, we can use \tabalign.
\let\+ = \relax
% Save some plain tex macros whose names we will redefine.
\let\ptexb=\b
\let\ptexbullet=\bullet
\let\ptexc=\c
\let\ptexcomma=\,
\let\ptexdot=\.
\let\ptexdots=\dots
\let\ptexend=\end
\let\ptexequiv=\equiv
\let\ptexexclam=\!
\let\ptexfootnote=\footnote
\let\ptexgtr=>
\let\ptexhat=^
\let\ptexi=\i
\let\ptexindent=\indent
\let\ptexinsert=\insert
\let\ptexlbrace=\{
\let\ptexless=<
\let\ptexnewwrite\newwrite
\let\ptexnoindent=\noindent
\let\ptexplus=+
\let\ptexrbrace=\}
\let\ptexslash=\/
\let\ptexstar=\*
\let\ptext=\t
% If this character appears in an error message or help string, it
% starts a new line in the output.
\newlinechar = `^^J
% Use TeX 3.0's \inputlineno to get the line number, for better error
% messages, but if we're using an old version of TeX, don't do anything.
%
\ifx\inputlineno\thisisundefined
\let\linenumber = \empty % Pre-3.0.
\else
\def\linenumber{l.\the\inputlineno:\space}
\fi
% Set up fixed words for English if not already set.
\ifx\putwordAppendix\undefined \gdef\putwordAppendix{Appendix}\fi
\ifx\putwordChapter\undefined \gdef\putwordChapter{Chapter}\fi
\ifx\putwordfile\undefined \gdef\putwordfile{file}\fi
\ifx\putwordin\undefined \gdef\putwordin{in}\fi
\ifx\putwordIndexIsEmpty\undefined \gdef\putwordIndexIsEmpty{(Index is empty)}\fi
\ifx\putwordIndexNonexistent\undefined \gdef\putwordIndexNonexistent{(Index is nonexistent)}\fi
\ifx\putwordInfo\undefined \gdef\putwordInfo{Info}\fi
\ifx\putwordInstanceVariableof\undefined \gdef\putwordInstanceVariableof{Instance Variable of}\fi
\ifx\putwordMethodon\undefined \gdef\putwordMethodon{Method on}\fi
\ifx\putwordNoTitle\undefined \gdef\putwordNoTitle{No Title}\fi
\ifx\putwordof\undefined \gdef\putwordof{of}\fi
\ifx\putwordon\undefined \gdef\putwordon{on}\fi
\ifx\putwordpage\undefined \gdef\putwordpage{page}\fi
\ifx\putwordsection\undefined \gdef\putwordsection{section}\fi
\ifx\putwordSection\undefined \gdef\putwordSection{Section}\fi
\ifx\putwordsee\undefined \gdef\putwordsee{see}\fi
\ifx\putwordSee\undefined \gdef\putwordSee{See}\fi
\ifx\putwordShortTOC\undefined \gdef\putwordShortTOC{Short Contents}\fi
\ifx\putwordTOC\undefined \gdef\putwordTOC{Table of Contents}\fi
%
\ifx\putwordMJan\undefined \gdef\putwordMJan{January}\fi
\ifx\putwordMFeb\undefined \gdef\putwordMFeb{February}\fi
\ifx\putwordMMar\undefined \gdef\putwordMMar{March}\fi
\ifx\putwordMApr\undefined \gdef\putwordMApr{April}\fi
\ifx\putwordMMay\undefined \gdef\putwordMMay{May}\fi
\ifx\putwordMJun\undefined \gdef\putwordMJun{June}\fi
\ifx\putwordMJul\undefined \gdef\putwordMJul{July}\fi
\ifx\putwordMAug\undefined \gdef\putwordMAug{August}\fi
\ifx\putwordMSep\undefined \gdef\putwordMSep{September}\fi
\ifx\putwordMOct\undefined \gdef\putwordMOct{October}\fi
\ifx\putwordMNov\undefined \gdef\putwordMNov{November}\fi
\ifx\putwordMDec\undefined \gdef\putwordMDec{December}\fi
%
\ifx\putwordDefmac\undefined \gdef\putwordDefmac{Macro}\fi
\ifx\putwordDefspec\undefined \gdef\putwordDefspec{Special Form}\fi
\ifx\putwordDefvar\undefined \gdef\putwordDefvar{Variable}\fi
\ifx\putwordDefopt\undefined \gdef\putwordDefopt{User Option}\fi
\ifx\putwordDeffunc\undefined \gdef\putwordDeffunc{Function}\fi
% Since the category of space is not known, we have to be careful.
\chardef\spacecat = 10
\def\spaceisspace{\catcode`\ =\spacecat}
% sometimes characters are active, so we need control sequences.
\chardef\colonChar = `\:
\chardef\commaChar = `\,
\chardef\dashChar = `\-
\chardef\dotChar = `\.
\chardef\exclamChar= `\!
\chardef\lquoteChar= `\`
\chardef\questChar = `\?
\chardef\rquoteChar= `\'
\chardef\semiChar = `\;
\chardef\underChar = `\_
% Ignore a token.
%
\def\gobble#1{}
% The following is used inside several \edef's.
\def\makecsname#1{\expandafter\noexpand\csname#1\endcsname}
% Hyphenation fixes.
\hyphenation{
Flor-i-da Ghost-script Ghost-view Mac-OS Post-Script
ap-pen-dix bit-map bit-maps
data-base data-bases eshell fall-ing half-way long-est man-u-script
man-u-scripts mini-buf-fer mini-buf-fers over-view par-a-digm
par-a-digms rath-er rec-tan-gu-lar ro-bot-ics se-vere-ly set-up spa-ces
spell-ing spell-ings
stand-alone strong-est time-stamp time-stamps which-ever white-space
wide-spread wrap-around
}
% Margin to add to right of even pages, to left of odd pages.
\newdimen\bindingoffset
\newdimen\normaloffset
\newdimen\pagewidth \newdimen\pageheight
% For a final copy, take out the rectangles
% that mark overfull boxes (in case you have decided
% that the text looks ok even though it passes the margin).
%
\def\finalout{\overfullrule=0pt}
% @| inserts a changebar to the left of the current line. It should
% surround any changed text. This approach does *not* work if the
% change spans more than two lines of output. To handle that, we would
% have adopt a much more difficult approach (putting marks into the main
% vertical list for the beginning and end of each change).
%
\def\|{%
% \vadjust can only be used in horizontal mode.
\leavevmode
%
% Append this vertical mode material after the current line in the output.
\vadjust{%
% We want to insert a rule with the height and depth of the current
% leading; that is exactly what \strutbox is supposed to record.
\vskip-\baselineskip
%
% \vadjust-items are inserted at the left edge of the type. So
% the \llap here moves out into the left-hand margin.
\llap{%
%
% For a thicker or thinner bar, change the `1pt'.
\vrule height\baselineskip width1pt
%
% This is the space between the bar and the text.
\hskip 12pt
}%
}%
}
% Sometimes it is convenient to have everything in the transcript file
% and nothing on the terminal. We don't just call \tracingall here,
% since that produces some useless output on the terminal. We also make
% some effort to order the tracing commands to reduce output in the log
% file; cf. trace.sty in LaTeX.
%
\def\gloggingall{\begingroup \globaldefs = 1 \loggingall \endgroup}%
\def\loggingall{%
\tracingstats2
\tracingpages1
\tracinglostchars2 % 2 gives us more in etex
\tracingparagraphs1
\tracingoutput1
\tracingmacros2
\tracingrestores1
\showboxbreadth\maxdimen \showboxdepth\maxdimen
\ifx\eTeXversion\undefined\else % etex gives us more logging
\tracingscantokens1
\tracingifs1
\tracinggroups1
\tracingnesting2
\tracingassigns1
\fi
\tracingcommands3 % 3 gives us more in etex
\errorcontextlines16
}%
% add check for \lastpenalty to plain's definitions. If the last thing
% we did was a \nobreak, we don't want to insert more space.
%
\def\smallbreak{\ifnum\lastpenalty<10000\par\ifdim\lastskip<\smallskipamount
\removelastskip\penalty-50\smallskip\fi\fi}
\def\medbreak{\ifnum\lastpenalty<10000\par\ifdim\lastskip<\medskipamount
\removelastskip\penalty-100\medskip\fi\fi}
\def\bigbreak{\ifnum\lastpenalty<10000\par\ifdim\lastskip<\bigskipamount
\removelastskip\penalty-200\bigskip\fi\fi}
% For @cropmarks command.
% Do @cropmarks to get crop marks.
%
\newif\ifcropmarks
\let\cropmarks = \cropmarkstrue
%
% Dimensions to add cropmarks at corners.
% Added by P. A. MacKay, 12 Nov. 1986
%
\newdimen\outerhsize \newdimen\outervsize % set by the paper size routines
\newdimen\cornerlong \cornerlong=1pc
\newdimen\cornerthick \cornerthick=.3pt
\newdimen\topandbottommargin \topandbottommargin=.75in
% Main output routine.
\chardef\PAGE = 255
\output = {\onepageout{\pagecontents\PAGE}}
\newbox\headlinebox
\newbox\footlinebox
% \onepageout takes a vbox as an argument. Note that \pagecontents
% does insertions, but you have to call it yourself.
\def\onepageout#1{%
\ifcropmarks \hoffset=0pt \else \hoffset=\normaloffset \fi
%
\ifodd\pageno \advance\hoffset by \bindingoffset
\else \advance\hoffset by -\bindingoffset\fi
%
% Do this outside of the \shipout so @code etc. will be expanded in
% the headline as they should be, not taken literally (outputting ''code).
\setbox\headlinebox = \vbox{\let\hsize=\pagewidth \makeheadline}%
\setbox\footlinebox = \vbox{\let\hsize=\pagewidth \makefootline}%
%
{%
% Have to do this stuff outside the \shipout because we want it to
% take effect in \write's, yet the group defined by the \vbox ends
% before the \shipout runs.
%
\indexdummies % don't expand commands in the output.
\normalturnoffactive % \ in index entries must not stay \, e.g., if
% the page break happens to be in the middle of an example.
% We don't want .vr (or whatever) entries like this:
% \entry{{\tt \indexbackslash }acronym}{32}{\code {\acronym}}
% "\acronym" won't work when it's read back in;
% it needs to be
% {\code {{\tt \backslashcurfont }acronym}
\shipout\vbox{%
% Do this early so pdf references go to the beginning of the page.
\ifpdfmakepagedest \pdfdest name{\the\pageno} xyz\fi
%
\ifcropmarks \vbox to \outervsize\bgroup
\hsize = \outerhsize
\vskip-\topandbottommargin
\vtop to0pt{%
\line{\ewtop\hfil\ewtop}%
\nointerlineskip
\line{%
\vbox{\moveleft\cornerthick\nstop}%
\hfill
\vbox{\moveright\cornerthick\nstop}%
}%
\vss}%
\vskip\topandbottommargin
\line\bgroup
\hfil % center the page within the outer (page) hsize.
\ifodd\pageno\hskip\bindingoffset\fi
\vbox\bgroup
\fi
%
\unvbox\headlinebox
\pagebody{#1}%
\ifdim\ht\footlinebox > 0pt
% Only leave this space if the footline is nonempty.
% (We lessened \vsize for it in \oddfootingyyy.)
% The \baselineskip=24pt in plain's \makefootline has no effect.
\vskip 24pt
\unvbox\footlinebox
\fi
%
\ifcropmarks
\egroup % end of \vbox\bgroup
\hfil\egroup % end of (centering) \line\bgroup
\vskip\topandbottommargin plus1fill minus1fill
\boxmaxdepth = \cornerthick
\vbox to0pt{\vss
\line{%
\vbox{\moveleft\cornerthick\nsbot}%
\hfill
\vbox{\moveright\cornerthick\nsbot}%
}%
\nointerlineskip
\line{\ewbot\hfil\ewbot}%
}%
\egroup % \vbox from first cropmarks clause
\fi
}% end of \shipout\vbox
}% end of group with \indexdummies
\advancepageno
\ifnum\outputpenalty>-20000 \else\dosupereject\fi
}
\newinsert\margin \dimen\margin=\maxdimen
\def\pagebody#1{\vbox to\pageheight{\boxmaxdepth=\maxdepth #1}}
{\catcode`\@ =11
\gdef\pagecontents#1{\ifvoid\topins\else\unvbox\topins\fi
% marginal hacks, juha@viisa.uucp (Juha Takala)
\ifvoid\margin\else % marginal info is present
\rlap{\kern\hsize\vbox to\z@{\kern1pt\box\margin \vss}}\fi
\dimen@=\dp#1 \unvbox#1
\ifvoid\footins\else\vskip\skip\footins\footnoterule \unvbox\footins\fi
\ifr@ggedbottom \kern-\dimen@ \vfil \fi}
}
% Here are the rules for the cropmarks. Note that they are
% offset so that the space between them is truly \outerhsize or \outervsize
% (P. A. MacKay, 12 November, 1986)
%
\def\ewtop{\vrule height\cornerthick depth0pt width\cornerlong}
\def\nstop{\vbox
{\hrule height\cornerthick depth\cornerlong width\cornerthick}}
\def\ewbot{\vrule height0pt depth\cornerthick width\cornerlong}
\def\nsbot{\vbox
{\hrule height\cornerlong depth\cornerthick width\cornerthick}}
% Parse an argument, then pass it to #1. The argument is the rest of
% the input line (except we remove a trailing comment). #1 should be a
% macro which expects an ordinary undelimited TeX argument.
%
\def\parsearg{\parseargusing{}}
\def\parseargusing#1#2{%
\def\argtorun{#2}%
\begingroup
\obeylines
\spaceisspace
#1%
\parseargline\empty% Insert the \empty token, see \finishparsearg below.
}
{\obeylines %
\gdef\parseargline#1^^M{%
\endgroup % End of the group started in \parsearg.
\argremovecomment #1\comment\ArgTerm%
}%
}
% First remove any @comment, then any @c comment.
\def\argremovecomment#1\comment#2\ArgTerm{\argremovec #1\c\ArgTerm}
\def\argremovec#1\c#2\ArgTerm{\argcheckspaces#1\^^M\ArgTerm}
-% Each occurence of `\^^M' or `<space>\^^M' is replaced by a single space.
+% Each occurrence 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
% Do not use \next, perhaps the caller of \parsearg uses it; reuse \temp:
\let\temp\finishparsearg
\else
\let\temp\argcheckspaces
\fi
% Put the space token in:
\temp#1 #3\ArgTerm
}
% If a _delimited_ argument is enclosed in braces, they get stripped; so
% to get _exactly_ the rest of the line, we had to prevent such situation.
% We prepended an \empty token at the very beginning and we expand it now,
% just before passing the control to \argtorun.
-% (Similarily, we have to think about #3 of \argcheckspacesY above: it is
+% (Similarly, we have to think about #3 of \argcheckspacesY above: it is
% either the null string, or it ends with \^^M---thus there is no danger
% that a pair of braces would be stripped.
%
% But first, we have to remove the trailing space token.
%
\def\finishparsearg#1 \ArgTerm{\expandafter\argtorun\expandafter{#1}}
% \parseargdef\foo{...}
% is roughly equivalent to
% \def\foo{\parsearg\Xfoo}
% \def\Xfoo#1{...}
%
% Actually, I use \csname\string\foo\endcsname, ie. \\foo, as it is my
% favourite TeX trick. --kasal, 16nov03
\def\parseargdef#1{%
\expandafter \doparseargdef \csname\string#1\endcsname #1%
}
\def\doparseargdef#1#2{%
\def#2{\parsearg#1}%
\def#1##1%
}
% Several utility definitions with active space:
{
\obeyspaces
\gdef\obeyedspace{ }
% Make each space character in the input produce a normal interword
% space in the output. Don't allow a line break at this space, as this
% is used only in environments like @example, where each line of input
% should produce a line of output anyway.
%
\gdef\sepspaces{\obeyspaces\let =\tie}
% If an index command is used in an @example environment, any spaces
% therein should become regular spaces in the raw index file, not the
% expansion of \tie (\leavevmode \penalty \@M \ ).
\gdef\unsepspaces{\let =\space}
}
\def\flushcr{\ifx\par\lisppar \def\next##1{}\else \let\next=\relax \fi \next}
% Define the framework for environments in texinfo.tex. It's used like this:
%
% \envdef\foo{...}
% \def\Efoo{...}
%
% It's the responsibility of \envdef to insert \begingroup before the
% actual body; @end closes the group after calling \Efoo. \envdef also
% defines \thisenv, so the current environment is known; @end checks
% whether the environment name matches. The \checkenv macro can also be
% used to check whether the current environment is the one expected.
%
% Non-false conditionals (@iftex, @ifset) don't fit into this, so they
-% are not treated as enviroments; they don't open a group. (The
+% are not treated as environments; they don't open a group. (The
% implementation of @end takes care not to call \endgroup in this
% special case.)
% At runtime, environments start with this:
\def\startenvironment#1{\begingroup\def\thisenv{#1}}
% initialize
\let\thisenv\empty
% ... but they get defined via ``\envdef\foo{...}'':
\long\def\envdef#1#2{\def#1{\startenvironment#1#2}}
\def\envparseargdef#1#2{\parseargdef#1{\startenvironment#1#2}}
% Check whether we're in the right environment:
\def\checkenv#1{%
\def\temp{#1}%
\ifx\thisenv\temp
\else
\badenverr
\fi
}
% Evironment mismatch, #1 expected:
\def\badenverr{%
\errhelp = \EMsimple
\errmessage{This command can appear only \inenvironment\temp,
not \inenvironment\thisenv}%
}
\def\inenvironment#1{%
\ifx#1\empty
out of any environment%
\else
in environment \expandafter\string#1%
\fi
}
% @end foo executes the definition of \Efoo.
% But first, it executes a specialized version of \checkenv
%
\parseargdef\end{%
\if 1\csname iscond.#1\endcsname
\else
% The general wording of \badenverr may not be ideal, but... --kasal, 06nov03
\expandafter\checkenv\csname#1\endcsname
\csname E#1\endcsname
\endgroup
\fi
}
\newhelp\EMsimple{Press RETURN to continue.}
%% Simple single-character @ commands
% @@ prints an @
% Kludge this until the fonts are right (grr).
\def\@{{\tt\char64}}
% This is turned off because it was never documented
% and you can use @w{...} around a quote to suppress ligatures.
%% Define @` and @' to be the same as ` and '
%% but suppressing ligatures.
%\def\`{{`}}
%\def\'{{'}}
% Used to generate quoted braces.
\def\mylbrace {{\tt\char123}}
\def\myrbrace {{\tt\char125}}
\let\{=\mylbrace
\let\}=\myrbrace
\begingroup
% Definitions to produce \{ and \} commands for indices,
% and @{ and @} for the aux/toc files.
\catcode`\{ = \other \catcode`\} = \other
\catcode`\[ = 1 \catcode`\] = 2
\catcode`\! = 0 \catcode`\\ = \other
!gdef!lbracecmd[\{]%
!gdef!rbracecmd[\}]%
!gdef!lbraceatcmd[@{]%
!gdef!rbraceatcmd[@}]%
!endgroup
% @comma{} to avoid , parsing problems.
\let\comma = ,
% Accents: @, @dotaccent @ringaccent @ubaraccent @udotaccent
% Others are defined by plain TeX: @` @' @" @^ @~ @= @u @v @H.
\let\, = \c
\let\dotaccent = \.
\def\ringaccent#1{{\accent23 #1}}
\let\tieaccent = \t
\let\ubaraccent = \b
\let\udotaccent = \d
% Other special characters: @questiondown @exclamdown @ordf @ordm
% Plain TeX defines: @AA @AE @O @OE @L (plus lowercase versions) @ss.
\def\questiondown{?`}
\def\exclamdown{!`}
\def\ordf{\leavevmode\raise1ex\hbox{\selectfonts\lllsize \underbar{a}}}
\def\ordm{\leavevmode\raise1ex\hbox{\selectfonts\lllsize \underbar{o}}}
% Dotless i and dotless j, used for accents.
\def\imacro{i}
\def\jmacro{j}
\def\dotless#1{%
\def\temp{#1}%
\ifx\temp\imacro \ptexi
\else\ifx\temp\jmacro \j
\else \errmessage{@dotless can be used only with i or j}%
\fi\fi
}
% The \TeX{} logo, as in plain, but resetting the spacing so that a
% period following counts as ending a sentence. (Idea found in latex.)
%
\edef\TeX{\TeX \spacefactor=1000 }
% @LaTeX{} logo. Not quite the same results as the definition in
% latex.ltx, since we use a different font for the raised A; it's most
% convenient for us to use an explicitly smaller font, rather than using
% the \scriptstyle font (since we don't reset \scriptstyle and
% \scriptscriptstyle).
%
\def\LaTeX{%
L\kern-.36em
{\setbox0=\hbox{T}%
\vbox to \ht0{\hbox{\selectfonts\lllsize A}\vss}}%
\kern-.15em
\TeX
}
% Be sure we're in horizontal mode when doing a tie, since we make space
% equivalent to this in @example-like environments. Otherwise, a space
% at the beginning of a line will start with \penalty -- and
% since \penalty is valid in vertical mode, we'd end up putting the
% penalty on the vertical list instead of in the new paragraph.
{\catcode`@ = 11
% Avoid using \@M directly, because that causes trouble
% if the definition is written into an index file.
\global\let\tiepenalty = \@M
\gdef\tie{\leavevmode\penalty\tiepenalty\ }
}
% @: forces normal size whitespace following.
\def\:{\spacefactor=1000 }
% @* forces a line break.
\def\*{\hfil\break\hbox{}\ignorespaces}
% @/ allows a line break.
\let\/=\allowbreak
% @. is an end-of-sentence period.
\def\.{.\spacefactor=\endofsentencespacefactor\space}
% @! is an end-of-sentence bang.
\def\!{!\spacefactor=\endofsentencespacefactor\space}
% @? is an end-of-sentence query.
\def\?{?\spacefactor=\endofsentencespacefactor\space}
% @frenchspacing on|off says whether to put extra space after punctuation.
%
\def\onword{on}
\def\offword{off}
%
\parseargdef\frenchspacing{%
\def\temp{#1}%
\ifx\temp\onword \plainfrenchspacing
\else\ifx\temp\offword \plainnonfrenchspacing
\else
\errhelp = \EMsimple
\errmessage{Unknown @frenchspacing option `\temp', must be on/off}%
\fi\fi
}
% @w prevents a word break. Without the \leavevmode, @w at the
% beginning of a paragraph, when TeX is still in vertical mode, would
% produce a whole line of output instead of starting the paragraph.
\def\w#1{\leavevmode\hbox{#1}}
% @group ... @end group forces ... to be all on one page, by enclosing
% it in a TeX vbox. We use \vtop instead of \vbox to construct the box
% to keep its height that of a normal line. According to the rules for
% \topskip (p.114 of the TeXbook), the glue inserted is
% max (\topskip - \ht (first item), 0). If that height is large,
% therefore, no glue is inserted, and the space between the headline and
% the text is small, which looks bad.
%
% Another complication is that the group might be very large. This can
% cause the glue on the previous page to be unduly stretched, because it
% does not have much material. In this case, it's better to add an
% explicit \vfill so that the extra space is at the bottom. The
% threshold for doing this is if the group is more than \vfilllimit
% percent of a page (\vfilllimit can be changed inside of @tex).
%
\newbox\groupbox
\def\vfilllimit{0.7}
%
\envdef\group{%
\ifnum\catcode`\^^M=\active \else
\errhelp = \groupinvalidhelp
\errmessage{@group invalid in context where filling is enabled}%
\fi
\startsavinginserts
%
\setbox\groupbox = \vtop\bgroup
% Do @comment since we are called inside an environment such as
% @example, where each end-of-line in the input causes an
% end-of-line in the output. We don't want the end-of-line after
% the `@group' to put extra space in the output. Since @group
% should appear on a line by itself (according to the Texinfo
% manual), we don't worry about eating any user text.
\comment
}
%
% The \vtop produces a box with normal height and large depth; thus, TeX puts
% \baselineskip glue before it, and (when the next line of text is done)
% \lineskip glue after it. Thus, space below is not quite equal to space
% above. But it's pretty close.
\def\Egroup{%
% To get correct interline space between the last line of the group
% and the first line afterwards, we have to propagate \prevdepth.
\endgraf % Not \par, as it may have been set to \lisppar.
\global\dimen1 = \prevdepth
\egroup % End the \vtop.
% \dimen0 is the vertical size of the group's box.
\dimen0 = \ht\groupbox \advance\dimen0 by \dp\groupbox
% \dimen2 is how much space is left on the page (more or less).
\dimen2 = \pageheight \advance\dimen2 by -\pagetotal
% if the group doesn't fit on the current page, and it's a big big
% group, force a page break.
\ifdim \dimen0 > \dimen2
\ifdim \pagetotal < \vfilllimit\pageheight
\page
\fi
\fi
\box\groupbox
\prevdepth = \dimen1
\checkinserts
}
%
% TeX puts in an \escapechar (i.e., `@') at the beginning of the help
% message, so this ends up printing `@group can only ...'.
%
\newhelp\groupinvalidhelp{%
group can only be used in environments such as @example,^^J%
where each line of input produces a line of output.}
% @need space-in-mils
% forces a page break if there is not space-in-mils remaining.
\newdimen\mil \mil=0.001in
% Old definition--didn't work.
%\parseargdef\need{\par %
%% This method tries to make TeX break the page naturally
%% if the depth of the box does not fit.
%{\baselineskip=0pt%
%\vtop to #1\mil{\vfil}\kern -#1\mil\nobreak
%\prevdepth=-1000pt
%}}
\parseargdef\need{%
% Ensure vertical mode, so we don't make a big box in the middle of a
% paragraph.
\par
%
% If the @need value is less than one line space, it's useless.
\dimen0 = #1\mil
\dimen2 = \ht\strutbox
\advance\dimen2 by \dp\strutbox
\ifdim\dimen0 > \dimen2
%
% Do a \strut just to make the height of this box be normal, so the
% normal leading is inserted relative to the preceding line.
% And a page break here is fine.
\vtop to #1\mil{\strut\vfil}%
%
% TeX does not even consider page breaks if a penalty added to the
% main vertical list is 10000 or more. But in order to see if the
% empty box we just added fits on the page, we must make it consider
% page breaks. On the other hand, we don't want to actually break the
% page after the empty box. So we use a penalty of 9999.
%
% There is an extremely small chance that TeX will actually break the
% page at this \penalty, if there are no other feasible breakpoints in
% sight. (If the user is using lots of big @group commands, which
% almost-but-not-quite fill up a page, TeX will have a hard time doing
% good page breaking, for example.) However, I could not construct an
% example where a page broke at this \penalty; if it happens in a real
% document, then we can reconsider our strategy.
\penalty9999
%
% Back up by the size of the box, whether we did a page break or not.
\kern -#1\mil
%
% Do not allow a page break right after this kern.
\nobreak
\fi
}
% @br forces paragraph break (and is undocumented).
\let\br = \par
% @page forces the start of a new page.
%
\def\page{\par\vfill\supereject}
% @exdent text....
% outputs text on separate line in roman font, starting at standard page margin
% This records the amount of indent in the innermost environment.
% That's how much \exdent should take out.
\newskip\exdentamount
% This defn is used inside fill environments such as @defun.
\parseargdef\exdent{\hfil\break\hbox{\kern -\exdentamount{\rm#1}}\hfil\break}
% This defn is used inside nofill environments such as @example.
\parseargdef\nofillexdent{{\advance \leftskip by -\exdentamount
\leftline{\hskip\leftskip{\rm#1}}}}
% @inmargin{WHICH}{TEXT} puts TEXT in the WHICH margin next to the current
% paragraph. For more general purposes, use the \margin insertion
% class. WHICH is `l' or `r'.
%
\newskip\inmarginspacing \inmarginspacing=1cm
\def\strutdepth{\dp\strutbox}
%
\def\doinmargin#1#2{\strut\vadjust{%
\nobreak
\kern-\strutdepth
\vtop to \strutdepth{%
\baselineskip=\strutdepth
\vss
% if you have multiple lines of stuff to put here, you'll need to
% make the vbox yourself of the appropriate size.
\ifx#1l%
\llap{\ignorespaces #2\hskip\inmarginspacing}%
\else
\rlap{\hskip\hsize \hskip\inmarginspacing \ignorespaces #2}%
\fi
\null
}%
}}
\def\inleftmargin{\doinmargin l}
\def\inrightmargin{\doinmargin r}
%
% @inmargin{TEXT [, RIGHT-TEXT]}
% (if RIGHT-TEXT is given, use TEXT for left page, RIGHT-TEXT for right;
% else use TEXT for both).
%
\def\inmargin#1{\parseinmargin #1,,\finish}
\def\parseinmargin#1,#2,#3\finish{% not perfect, but better than nothing.
\setbox0 = \hbox{\ignorespaces #2}%
\ifdim\wd0 > 0pt
\def\lefttext{#1}% have both texts
\def\righttext{#2}%
\else
\def\lefttext{#1}% have only one text
\def\righttext{#1}%
\fi
%
\ifodd\pageno
\def\temp{\inrightmargin\righttext}% odd page -> outside is right margin
\else
\def\temp{\inleftmargin\lefttext}%
\fi
\temp
}
% @include file insert text of that file as input.
%
\def\include{\parseargusing\filenamecatcodes\includezzz}
\def\includezzz#1{%
\pushthisfilestack
\def\thisfile{#1}%
{%
\makevalueexpandable
\def\temp{\input #1 }%
\expandafter
}\temp
\popthisfilestack
}
\def\filenamecatcodes{%
\catcode`\\=\other
\catcode`~=\other
\catcode`^=\other
\catcode`_=\other
\catcode`|=\other
\catcode`<=\other
\catcode`>=\other
\catcode`+=\other
\catcode`-=\other
}
\def\pushthisfilestack{%
\expandafter\pushthisfilestackX\popthisfilestack\StackTerm
}
\def\pushthisfilestackX{%
\expandafter\pushthisfilestackY\thisfile\StackTerm
}
\def\pushthisfilestackY #1\StackTerm #2\StackTerm {%
\gdef\popthisfilestack{\gdef\thisfile{#1}\gdef\popthisfilestack{#2}}%
}
\def\popthisfilestack{\errthisfilestackempty}
\def\errthisfilestackempty{\errmessage{Internal error:
the stack of filenames is empty.}}
\def\thisfile{}
% @center line
% outputs that line, centered.
%
\parseargdef\center{%
\ifhmode
\let\next\centerH
\else
\let\next\centerV
\fi
\next{\hfil \ignorespaces#1\unskip \hfil}%
}
\def\centerH#1{%
{%
\hfil\break
\advance\hsize by -\leftskip
\advance\hsize by -\rightskip
\line{#1}%
\break
}%
}
\def\centerV#1{\line{\kern\leftskip #1\kern\rightskip}}
% @sp n outputs n lines of vertical space
\parseargdef\sp{\vskip #1\baselineskip}
% @comment ...line which is ignored...
% @c is the same as @comment
% @ignore ... @end ignore is another way to write a comment
\def\comment{\begingroup \catcode`\^^M=\other%
\catcode`\@=\other \catcode`\{=\other \catcode`\}=\other%
\commentxxx}
{\catcode`\^^M=\other \gdef\commentxxx#1^^M{\endgroup}}
\let\c=\comment
% @paragraphindent NCHARS
% We'll use ems for NCHARS, close enough.
% NCHARS can also be the word `asis' or `none'.
% We cannot feasibly implement @paragraphindent asis, though.
%
\def\asisword{asis} % no translation, these are keywords
\def\noneword{none}
%
\parseargdef\paragraphindent{%
\def\temp{#1}%
\ifx\temp\asisword
\else
\ifx\temp\noneword
\defaultparindent = 0pt
\else
\defaultparindent = #1em
\fi
\fi
\parindent = \defaultparindent
}
% @exampleindent NCHARS
% We'll use ems for NCHARS like @paragraphindent.
% It seems @exampleindent asis isn't necessary, but
% I preserve it to make it similar to @paragraphindent.
\parseargdef\exampleindent{%
\def\temp{#1}%
\ifx\temp\asisword
\else
\ifx\temp\noneword
\lispnarrowing = 0pt
\else
\lispnarrowing = #1em
\fi
\fi
}
% @firstparagraphindent WORD
% If WORD is `none', then suppress indentation of the first paragraph
% after a section heading. If WORD is `insert', then do indent at such
% paragraphs.
%
% The paragraph indentation is suppressed or not by calling
% \suppressfirstparagraphindent, which the sectioning commands do.
% We switch the definition of this back and forth according to WORD.
% By default, we suppress indentation.
%
\def\suppressfirstparagraphindent{\dosuppressfirstparagraphindent}
\def\insertword{insert}
%
\parseargdef\firstparagraphindent{%
\def\temp{#1}%
\ifx\temp\noneword
\let\suppressfirstparagraphindent = \dosuppressfirstparagraphindent
\else\ifx\temp\insertword
\let\suppressfirstparagraphindent = \relax
\else
\errhelp = \EMsimple
\errmessage{Unknown @firstparagraphindent option `\temp'}%
\fi\fi
}
% Here is how we actually suppress indentation. Redefine \everypar to
% \kern backwards by \parindent, and then reset itself to empty.
%
% We also make \indent itself not actually do anything until the next
% paragraph.
%
\gdef\dosuppressfirstparagraphindent{%
\gdef\indent{%
\restorefirstparagraphindent
\indent
}%
\gdef\noindent{%
\restorefirstparagraphindent
\noindent
}%
\global\everypar = {%
\kern -\parindent
\restorefirstparagraphindent
}%
}
\gdef\restorefirstparagraphindent{%
\global \let \indent = \ptexindent
\global \let \noindent = \ptexnoindent
\global \everypar = {}%
}
% @asis just yields its argument. Used with @table, for example.
%
\def\asis#1{#1}
% @math outputs its argument in math mode.
%
% One complication: _ usually means subscripts, but it could also mean
% an actual _ character, as in @math{@var{some_variable} + 1}. So make
% _ active, and distinguish by seeing if the current family is \slfam,
% which is what @var uses.
{
\catcode`\_ = \active
\gdef\mathunderscore{%
\catcode`\_=\active
\def_{\ifnum\fam=\slfam \_\else\sb\fi}%
}
}
% Another complication: we want \\ (and @\) to output a \ character.
% FYI, plain.tex uses \\ as a temporary control sequence (why?), but
% this is not advertised and we don't care. Texinfo does not
% otherwise define @\.
%
% The \mathchar is class=0=ordinary, family=7=ttfam, position=5C=\.
\def\mathbackslash{\ifnum\fam=\ttfam \mathchar"075C \else\backslash \fi}
%
\def\math{%
\tex
\mathunderscore
\let\\ = \mathbackslash
\mathactive
$\finishmath
}
\def\finishmath#1{#1$\endgroup} % Close the group opened by \tex.
% Some active characters (such as <) are spaced differently in math.
% We have to reset their definitions in case the @math was an argument
% to a command which sets the catcodes (such as @item or @section).
%
{
\catcode`^ = \active
\catcode`< = \active
\catcode`> = \active
\catcode`+ = \active
\gdef\mathactive{%
\let^ = \ptexhat
\let< = \ptexless
\let> = \ptexgtr
\let+ = \ptexplus
}
}
% @bullet and @minus need the same treatment as @math, just above.
\def\bullet{$\ptexbullet$}
\def\minus{$-$}
% @dots{} outputs an ellipsis using the current font.
% We do .5em per period so that it has the same spacing in the cm
% typewriter fonts as three actual period characters; on the other hand,
% in other typewriter fonts three periods are wider than 1.5em. So do
% whichever is larger.
%
\def\dots{%
\leavevmode
\setbox0=\hbox{...}% get width of three periods
\ifdim\wd0 > 1.5em
\dimen0 = \wd0
\else
\dimen0 = 1.5em
\fi
\hbox to \dimen0{%
\hskip 0pt plus.25fil
.\hskip 0pt plus1fil
.\hskip 0pt plus1fil
.\hskip 0pt plus.5fil
}%
}
% @enddots{} is an end-of-sentence ellipsis.
%
\def\enddots{%
\dots
\spacefactor=\endofsentencespacefactor
}
% @comma{} is so commas can be inserted into text without messing up
% Texinfo's parsing.
%
\let\comma = ,
% @refill is a no-op.
\let\refill=\relax
% If working on a large document in chapters, it is convenient to
% be able to disable indexing, cross-referencing, and contents, for test runs.
% This is done with @novalidate (before @setfilename).
%
\newif\iflinks \linkstrue % by default we want the aux files.
\let\novalidate = \linksfalse
% @setfilename is done at the beginning of every texinfo file.
% So open here the files we need to have open while reading the input.
% This makes it possible to make a .fmt file for texinfo.
\def\setfilename{%
\fixbackslash % Turn off hack to swallow `\input texinfo'.
\iflinks
\tryauxfile
% Open the new aux file. TeX will close it automatically at exit.
\immediate\openout\auxfile=\jobname.aux
\fi % \openindices needs to do some work in any case.
\openindices
\let\setfilename=\comment % Ignore extra @setfilename cmds.
%
% If texinfo.cnf is present on the system, read it.
% Useful for site-wide @afourpaper, etc.
\openin 1 texinfo.cnf
\ifeof 1 \else \input texinfo.cnf \fi
\closein 1
%
\comment % Ignore the actual filename.
}
% Called from \setfilename.
%
\def\openindices{%
\newindex{cp}%
\newcodeindex{fn}%
\newcodeindex{vr}%
\newcodeindex{tp}%
\newcodeindex{ky}%
\newcodeindex{pg}%
}
% @bye.
\outer\def\bye{\pagealignmacro\tracingstats=1\ptexend}
\message{pdf,}
% adobe `portable' document format
\newcount\tempnum
\newcount\lnkcount
\newtoks\filename
\newcount\filenamelength
\newcount\pgn
\newtoks\toksA
\newtoks\toksB
\newtoks\toksC
\newtoks\toksD
\newbox\boxA
\newcount\countA
\newif\ifpdf
\newif\ifpdfmakepagedest
% when pdftex is run in dvi mode, \pdfoutput is defined (so \pdfoutput=1
% can be set). So we test for \relax and 0 as well as \undefined,
% borrowed from ifpdf.sty.
\ifx\pdfoutput\undefined
\else
\ifx\pdfoutput\relax
\else
\ifcase\pdfoutput
\else
\pdftrue
\fi
\fi
\fi
% PDF uses PostScript string constants for the names of xref targets,
% for display in the outlines, and in other places. Thus, we have to
% double any backslashes. Otherwise, a name like "\node" will be
% interpreted as a newline (\n), followed by o, d, e. Not good.
% http://www.ntg.nl/pipermail/ntg-pdftex/2004-July/000654.html
% (and related messages, the final outcome is that it is up to the TeX
% user to double the backslashes and otherwise make the string valid, so
% that's what we do).
% double active backslashes.
%
{\catcode`\@=0 \catcode`\\=\active
@gdef@activebackslashdouble{%
@catcode`@\=@active
@let\=@doublebackslash}
}
% To handle parens, we must adopt a different approach, since parens are
% not active characters. hyperref.dtx (which has the same problem as
% us) handles it with this amazing macro to replace tokens, with minor
% changes for Texinfo. It is included here under the GPL by permission
% from the author, Heiko Oberdiek.
%
% #1 is the tokens to replace.
% #2 is the replacement.
% #3 is the control sequence with the string.
%
\def\HyPsdSubst#1#2#3{%
\def\HyPsdReplace##1#1##2\END{%
##1%
\ifx\\##2\\%
\else
#2%
\HyReturnAfterFi{%
\HyPsdReplace##2\END
}%
\fi
}%
\xdef#3{\expandafter\HyPsdReplace#3#1\END}%
}
\long\def\HyReturnAfterFi#1\fi{\fi#1}
% #1 is a control sequence in which to do the replacements.
\def\backslashparens#1{%
\xdef#1{#1}% redefine it as its expansion; the definition is simply
% \lastnode when called from \setref -> \pdfmkdest.
\HyPsdSubst{(}{\realbackslash(}{#1}%
\HyPsdSubst{)}{\realbackslash)}{#1}%
}
\newhelp\nopdfimagehelp{Texinfo supports .png, .jpg, .jpeg, and .pdf images
with PDF output, and none of those formats could be found. (.eps cannot
be supported due to the design of the PDF format; use regular TeX (DVI
output) for that.)}
\ifpdf
\input pdfcolor
\pdfcatalog{/PageMode /UseOutlines}
%
% #1 is image name, #2 width (might be empty/whitespace), #3 height (ditto).
\def\dopdfimage#1#2#3{%
\def\imagewidth{#2}\setbox0 = \hbox{\ignorespaces #2}%
\def\imageheight{#3}\setbox2 = \hbox{\ignorespaces #3}%
%
% pdftex (and the PDF format) support .png, .jpg, .pdf (among
% others). Let's try in that order.
\let\pdfimgext=\empty
\begingroup
\openin 1 #1.png \ifeof 1
\openin 1 #1.jpg \ifeof 1
\openin 1 #1.jpeg \ifeof 1
\openin 1 #1.JPG \ifeof 1
\openin 1 #1.pdf \ifeof 1
\errhelp = \nopdfimagehelp
\errmessage{Could not find image file #1 for pdf}%
\else \gdef\pdfimgext{pdf}%
\fi
\else \gdef\pdfimgext{JPG}%
\fi
\else \gdef\pdfimgext{jpeg}%
\fi
\else \gdef\pdfimgext{jpg}%
\fi
\else \gdef\pdfimgext{png}%
\fi
\closein 1
\endgroup
%
% without \immediate, pdftex seg faults when the same image is
% included twice. (Version 3.14159-pre-1.0-unofficial-20010704.)
\ifnum\pdftexversion < 14
\immediate\pdfimage
\else
\immediate\pdfximage
\fi
\ifdim \wd0 >0pt width \imagewidth \fi
\ifdim \wd2 >0pt height \imageheight \fi
\ifnum\pdftexversion<13
#1.\pdfimgext
\else
{#1.\pdfimgext}%
\fi
\ifnum\pdftexversion < 14 \else
\pdfrefximage \pdflastximage
\fi}
%
\def\pdfmkdest#1{{%
% We have to set dummies so commands such as @code, and characters
% such as \, aren't expanded when present in a section title.
\indexnofonts
\turnoffactive
\activebackslashdouble
\makevalueexpandable
\def\pdfdestname{#1}%
\backslashparens\pdfdestname
\safewhatsit{\pdfdest name{\pdfdestname} xyz}%
}}
%
% used to mark target names; must be expandable.
\def\pdfmkpgn#1{#1}
%
% by default, use a color that is dark enough to print on paper as
% nearly black, but still distinguishable for online viewing.
% (Defined in pdfcolor.tex.)
\let\urlcolor = \BrickRed
\let\linkcolor = \BrickRed
\def\endlink{\Black\pdfendlink}
%
% Adding outlines to PDF; macros for calculating structure of outlines
% come from Petr Olsak
\def\expnumber#1{\expandafter\ifx\csname#1\endcsname\relax 0%
\else \csname#1\endcsname \fi}
\def\advancenumber#1{\tempnum=\expnumber{#1}\relax
\advance\tempnum by 1
\expandafter\xdef\csname#1\endcsname{\the\tempnum}}
%
% #1 is the section text, which is what will be displayed in the
% outline by the pdf viewer. #2 is the pdf expression for the number
% of subentries (or empty, for subsubsections). #3 is the node text,
% which might be empty if this toc entry had no corresponding node.
% #4 is the page number
%
\def\dopdfoutline#1#2#3#4{%
% Generate a link to the node text if that exists; else, use the
% page number. We could generate a destination for the section
% text in the case where a section has no node, but it doesn't
% seem worth the trouble, since most documents are normally structured.
\def\pdfoutlinedest{#3}%
\ifx\pdfoutlinedest\empty
\def\pdfoutlinedest{#4}%
\else
% Doubled backslashes in the name.
{\activebackslashdouble \xdef\pdfoutlinedest{#3}%
\backslashparens\pdfoutlinedest}%
\fi
%
% Also double the backslashes in the display string.
{\activebackslashdouble \xdef\pdfoutlinetext{#1}%
\backslashparens\pdfoutlinetext}%
%
\pdfoutline goto name{\pdfmkpgn{\pdfoutlinedest}}#2{\pdfoutlinetext}%
}
%
\def\pdfmakeoutlines{%
\begingroup
% Thanh's hack / proper braces in bookmarks
\edef\mylbrace{\iftrue \string{\else}\fi}\let\{=\mylbrace
\edef\myrbrace{\iffalse{\else\string}\fi}\let\}=\myrbrace
%
% Read toc silently, to get counts of subentries for \pdfoutline.
\def\numchapentry##1##2##3##4{%
\def\thischapnum{##2}%
\def\thissecnum{0}%
\def\thissubsecnum{0}%
}%
\def\numsecentry##1##2##3##4{%
\advancenumber{chap\thischapnum}%
\def\thissecnum{##2}%
\def\thissubsecnum{0}%
}%
\def\numsubsecentry##1##2##3##4{%
\advancenumber{sec\thissecnum}%
\def\thissubsecnum{##2}%
}%
\def\numsubsubsecentry##1##2##3##4{%
\advancenumber{subsec\thissubsecnum}%
}%
\def\thischapnum{0}%
\def\thissecnum{0}%
\def\thissubsecnum{0}%
%
% use \def rather than \let here because we redefine \chapentry et
% al. a second time, below.
\def\appentry{\numchapentry}%
\def\appsecentry{\numsecentry}%
\def\appsubsecentry{\numsubsecentry}%
\def\appsubsubsecentry{\numsubsubsecentry}%
\def\unnchapentry{\numchapentry}%
\def\unnsecentry{\numsecentry}%
\def\unnsubsecentry{\numsubsecentry}%
\def\unnsubsubsecentry{\numsubsubsecentry}%
\readdatafile{toc}%
%
% Read toc second time, this time actually producing the outlines.
% The `-' means take the \expnumber as the absolute number of
% subentries, which we calculated on our first read of the .toc above.
%
% We use the node names as the destinations.
\def\numchapentry##1##2##3##4{%
\dopdfoutline{##1}{count-\expnumber{chap##2}}{##3}{##4}}%
\def\numsecentry##1##2##3##4{%
\dopdfoutline{##1}{count-\expnumber{sec##2}}{##3}{##4}}%
\def\numsubsecentry##1##2##3##4{%
\dopdfoutline{##1}{count-\expnumber{subsec##2}}{##3}{##4}}%
\def\numsubsubsecentry##1##2##3##4{% count is always zero
\dopdfoutline{##1}{}{##3}{##4}}%
%
% PDF outlines are displayed using system fonts, instead of
% document fonts. Therefore we cannot use special characters,
% since the encoding is unknown. For example, the eogonek from
% Latin 2 (0xea) gets translated to a | character. Info from
% Staszek Wawrykiewicz, 19 Jan 2004 04:09:24 +0100.
%
% xx to do this right, we have to translate 8-bit characters to
% their "best" equivalent, based on the @documentencoding. Right
% now, I guess we'll just let the pdf reader have its way.
\indexnofonts
\setupdatafile
\catcode`\\=\active \otherbackslash
\input \jobname.toc
\endgroup
}
%
\def\skipspaces#1{\def\PP{#1}\def\D{|}%
\ifx\PP\D\let\nextsp\relax
\else\let\nextsp\skipspaces
\ifx\p\space\else\addtokens{\filename}{\PP}%
\advance\filenamelength by 1
\fi
\fi
\nextsp}
\def\getfilename#1{\filenamelength=0\expandafter\skipspaces#1|\relax}
\ifnum\pdftexversion < 14
\let \startlink \pdfannotlink
\else
\let \startlink \pdfstartlink
\fi
% make a live url in pdf output.
\def\pdfurl#1{%
\begingroup
% it seems we really need yet another set of dummies; have not
% tried to figure out what each command should do in the context
% of @url. for now, just make @/ a no-op, that's the only one
% people have actually reported a problem with.
%
\normalturnoffactive
\def\@{@}%
\let\/=\empty
\makevalueexpandable
\leavevmode\urlcolor
\startlink attr{/Border [0 0 0]}%
user{/Subtype /Link /A << /S /URI /URI (#1) >>}%
\endgroup}
\def\pdfgettoks#1.{\setbox\boxA=\hbox{\toksA={#1.}\toksB={}\maketoks}}
\def\addtokens#1#2{\edef\addtoks{\noexpand#1={\the#1#2}}\addtoks}
\def\adn#1{\addtokens{\toksC}{#1}\global\countA=1\let\next=\maketoks}
\def\poptoks#1#2|ENDTOKS|{\let\first=#1\toksD={#1}\toksA={#2}}
\def\maketoks{%
\expandafter\poptoks\the\toksA|ENDTOKS|\relax
\ifx\first0\adn0
\else\ifx\first1\adn1 \else\ifx\first2\adn2 \else\ifx\first3\adn3
\else\ifx\first4\adn4 \else\ifx\first5\adn5 \else\ifx\first6\adn6
\else\ifx\first7\adn7 \else\ifx\first8\adn8 \else\ifx\first9\adn9
\else
\ifnum0=\countA\else\makelink\fi
\ifx\first.\let\next=\done\else
\let\next=\maketoks
\addtokens{\toksB}{\the\toksD}
\ifx\first,\addtokens{\toksB}{\space}\fi
\fi
\fi\fi\fi\fi\fi\fi\fi\fi\fi\fi
\next}
\def\makelink{\addtokens{\toksB}%
{\noexpand\pdflink{\the\toksC}}\toksC={}\global\countA=0}
\def\pdflink#1{%
\startlink attr{/Border [0 0 0]} goto name{\pdfmkpgn{#1}}
\linkcolor #1\endlink}
\def\done{\edef\st{\global\noexpand\toksA={\the\toksB}}\st}
\else
\let\pdfmkdest = \gobble
\let\pdfurl = \gobble
\let\endlink = \relax
\let\linkcolor = \relax
\let\pdfmakeoutlines = \relax
\fi % \ifx\pdfoutput
\message{fonts,}
% Change the current font style to #1, remembering it in \curfontstyle.
% For now, we do not accumulate font styles: @b{@i{foo}} prints foo in
% italics, not bold italics.
%
\def\setfontstyle#1{%
\def\curfontstyle{#1}% not as a control sequence, because we are \edef'd.
\csname ten#1\endcsname % change the current font
}
% Select #1 fonts with the current style.
%
\def\selectfonts#1{\csname #1fonts\endcsname \csname\curfontstyle\endcsname}
\def\rm{\fam=0 \setfontstyle{rm}}
\def\it{\fam=\itfam \setfontstyle{it}}
\def\sl{\fam=\slfam \setfontstyle{sl}}
\def\bf{\fam=\bffam \setfontstyle{bf}}\def\bfstylename{bf}
\def\tt{\fam=\ttfam \setfontstyle{tt}}
% Texinfo sort of supports the sans serif font style, which plain TeX does not.
% So we set up a \sf.
\newfam\sffam
\def\sf{\fam=\sffam \setfontstyle{sf}}
\let\li = \sf % Sometimes we call it \li, not \sf.
% We don't need math for this font style.
\def\ttsl{\setfontstyle{ttsl}}
% Default leading.
\newdimen\textleading \textleading = 13.2pt
% Set the baselineskip to #1, and the lineskip and strut size
% correspondingly. There is no deep meaning behind these magic numbers
% used as factors; they just match (closely enough) what Knuth defined.
%
\def\lineskipfactor{.08333}
\def\strutheightpercent{.70833}
\def\strutdepthpercent {.29167}
%
\def\setleading#1{%
\normalbaselineskip = #1\relax
\normallineskip = \lineskipfactor\normalbaselineskip
\normalbaselines
\setbox\strutbox =\hbox{%
\vrule width0pt height\strutheightpercent\baselineskip
depth \strutdepthpercent \baselineskip
}%
}
%
% PDF CMaps. See also LaTeX's t1.cmap.
%
% \cmapOT1
\ifpdf
\begingroup
\catcode`\^^M=\active \def^^M{^^J}% Output line endings as the ^^J char.
\catcode`\%=12 \immediate\pdfobj stream {%!PS-Adobe-3.0 Resource-CMap
%%DocumentNeededResources: ProcSet (CIDInit)
%%IncludeResource: ProcSet (CIDInit)
%%BeginResource: CMap (TeX-OT1-0)
%%Title: (TeX-OT1-0 TeX OT1 0)
%%Version: 1.000
%%EndComments
/CIDInit /ProcSet findresource begin
12 dict begin
begincmap
/CIDSystemInfo
<< /Registry (TeX)
/Ordering (OT1)
/Supplement 0
>> def
/CMapName /TeX-OT1-0 def
/CMapType 2 def
1 begincodespacerange
<00> <7F>
endcodespacerange
8 beginbfrange
<00> <01> <0393>
<09> <0A> <03A8>
<23> <26> <0023>
<28> <3B> <0028>
<3F> <5B> <003F>
<5D> <5E> <005D>
<61> <7A> <0061>
<7B> <7C> <2013>
endbfrange
40 beginbfchar
<02> <0398>
<03> <039B>
<04> <039E>
<05> <03A0>
<06> <03A3>
<07> <03D2>
<08> <03A6>
<0B> <00660066>
<0C> <00660069>
<0D> <0066006C>
<0E> <006600660069>
<0F> <00660066006C>
<10> <0131>
<11> <0237>
<12> <0060>
<13> <00B4>
<14> <02C7>
<15> <02D8>
<16> <00AF>
<17> <02DA>
<18> <00B8>
<19> <00DF>
<1A> <00E6>
<1B> <0153>
<1C> <00F8>
<1D> <00C6>
<1E> <0152>
<1F> <00D8>
<21> <0021>
<22> <201D>
<27> <2019>
<3C> <00A1>
<3D> <003D>
<3E> <00BF>
<5C> <201C>
<5F> <02D9>
<60> <2018>
<7D> <02DD>
<7E> <007E>
<7F> <00A8>
endbfchar
endcmap
CMapName currentdict /CMap defineresource pop
end
end
%%EndResource
%%EOF
}\endgroup
\expandafter\edef\csname cmapOT1\endcsname#1{%
\pdffontattr#1{/ToUnicode \the\pdflastobj\space 0 R}%
}%
%
% \cmapOT1IT
\begingroup
\catcode`\^^M=\active \def^^M{^^J}% Output line endings as the ^^J char.
\catcode`\%=12 \immediate\pdfobj stream {%!PS-Adobe-3.0 Resource-CMap
%%DocumentNeededResources: ProcSet (CIDInit)
%%IncludeResource: ProcSet (CIDInit)
%%BeginResource: CMap (TeX-OT1IT-0)
%%Title: (TeX-OT1IT-0 TeX OT1IT 0)
%%Version: 1.000
%%EndComments
/CIDInit /ProcSet findresource begin
12 dict begin
begincmap
/CIDSystemInfo
<< /Registry (TeX)
/Ordering (OT1IT)
/Supplement 0
>> def
/CMapName /TeX-OT1IT-0 def
/CMapType 2 def
1 begincodespacerange
<00> <7F>
endcodespacerange
8 beginbfrange
<00> <01> <0393>
<09> <0A> <03A8>
<25> <26> <0025>
<28> <3B> <0028>
<3F> <5B> <003F>
<5D> <5E> <005D>
<61> <7A> <0061>
<7B> <7C> <2013>
endbfrange
42 beginbfchar
<02> <0398>
<03> <039B>
<04> <039E>
<05> <03A0>
<06> <03A3>
<07> <03D2>
<08> <03A6>
<0B> <00660066>
<0C> <00660069>
<0D> <0066006C>
<0E> <006600660069>
<0F> <00660066006C>
<10> <0131>
<11> <0237>
<12> <0060>
<13> <00B4>
<14> <02C7>
<15> <02D8>
<16> <00AF>
<17> <02DA>
<18> <00B8>
<19> <00DF>
<1A> <00E6>
<1B> <0153>
<1C> <00F8>
<1D> <00C6>
<1E> <0152>
<1F> <00D8>
<21> <0021>
<22> <201D>
<23> <0023>
<24> <00A3>
<27> <2019>
<3C> <00A1>
<3D> <003D>
<3E> <00BF>
<5C> <201C>
<5F> <02D9>
<60> <2018>
<7D> <02DD>
<7E> <007E>
<7F> <00A8>
endbfchar
endcmap
CMapName currentdict /CMap defineresource pop
end
end
%%EndResource
%%EOF
}\endgroup
\expandafter\edef\csname cmapOT1IT\endcsname#1{%
\pdffontattr#1{/ToUnicode \the\pdflastobj\space 0 R}%
}%
%
% \cmapOT1TT
\begingroup
\catcode`\^^M=\active \def^^M{^^J}% Output line endings as the ^^J char.
\catcode`\%=12 \immediate\pdfobj stream {%!PS-Adobe-3.0 Resource-CMap
%%DocumentNeededResources: ProcSet (CIDInit)
%%IncludeResource: ProcSet (CIDInit)
%%BeginResource: CMap (TeX-OT1TT-0)
%%Title: (TeX-OT1TT-0 TeX OT1TT 0)
%%Version: 1.000
%%EndComments
/CIDInit /ProcSet findresource begin
12 dict begin
begincmap
/CIDSystemInfo
<< /Registry (TeX)
/Ordering (OT1TT)
/Supplement 0
>> def
/CMapName /TeX-OT1TT-0 def
/CMapType 2 def
1 begincodespacerange
<00> <7F>
endcodespacerange
5 beginbfrange
<00> <01> <0393>
<09> <0A> <03A8>
<21> <26> <0021>
<28> <5F> <0028>
<61> <7E> <0061>
endbfrange
32 beginbfchar
<02> <0398>
<03> <039B>
<04> <039E>
<05> <03A0>
<06> <03A3>
<07> <03D2>
<08> <03A6>
<0B> <2191>
<0C> <2193>
<0D> <0027>
<0E> <00A1>
<0F> <00BF>
<10> <0131>
<11> <0237>
<12> <0060>
<13> <00B4>
<14> <02C7>
<15> <02D8>
<16> <00AF>
<17> <02DA>
<18> <00B8>
<19> <00DF>
<1A> <00E6>
<1B> <0153>
<1C> <00F8>
<1D> <00C6>
<1E> <0152>
<1F> <00D8>
<20> <2423>
<27> <2019>
<60> <2018>
<7F> <00A8>
endbfchar
endcmap
CMapName currentdict /CMap defineresource pop
end
end
%%EndResource
%%EOF
}\endgroup
\expandafter\edef\csname cmapOT1TT\endcsname#1{%
\pdffontattr#1{/ToUnicode \the\pdflastobj\space 0 R}%
}%
\else
\expandafter\let\csname cmapOT1\endcsname\gobble
\expandafter\let\csname cmapOT1IT\endcsname\gobble
\expandafter\let\csname cmapOT1TT\endcsname\gobble
\fi
% Set the font macro #1 to the font named #2, adding on the
% specified font prefix (normally `cm').
% #3 is the font's design size, #4 is a scale factor, #5 is the CMap
% encoding (currently only OT1, OT1IT and OT1TT are allowed, pass
% empty to omit).
\def\setfont#1#2#3#4#5{%
\font#1=\fontprefix#2#3 scaled #4
\csname cmap#5\endcsname#1%
}
% This is what gets called when #5 of \setfont is empty.
\let\cmap\gobble
% Use cm as the default font prefix.
% To specify the font prefix, you must define \fontprefix
% before you read in texinfo.tex.
\ifx\fontprefix\undefined
\def\fontprefix{cm}
\fi
% Support font families that don't use the same naming scheme as CM.
\def\rmshape{r}
\def\rmbshape{bx} %where the normal face is bold
\def\bfshape{b}
\def\bxshape{bx}
\def\ttshape{tt}
\def\ttbshape{tt}
\def\ttslshape{sltt}
\def\itshape{ti}
\def\itbshape{bxti}
\def\slshape{sl}
\def\slbshape{bxsl}
\def\sfshape{ss}
\def\sfbshape{ss}
\def\scshape{csc}
\def\scbshape{csc}
% Definitions for a main text size of 11pt. This is the default in
% Texinfo.
%
\def\definetextfontsizexi{%
% Text fonts (11.2pt, magstep1).
\def\textnominalsize{11pt}
\edef\mainmagstep{\magstephalf}
\setfont\textrm\rmshape{10}{\mainmagstep}{OT1}
\setfont\texttt\ttshape{10}{\mainmagstep}{OT1TT}
\setfont\textbf\bfshape{10}{\mainmagstep}{OT1}
\setfont\textit\itshape{10}{\mainmagstep}{OT1IT}
\setfont\textsl\slshape{10}{\mainmagstep}{OT1}
\setfont\textsf\sfshape{10}{\mainmagstep}{OT1}
\setfont\textsc\scshape{10}{\mainmagstep}{OT1}
\setfont\textttsl\ttslshape{10}{\mainmagstep}{OT1TT}
\font\texti=cmmi10 scaled \mainmagstep
\font\textsy=cmsy10 scaled \mainmagstep
% A few fonts for @defun names and args.
\setfont\defbf\bfshape{10}{\magstep1}{OT1}
\setfont\deftt\ttshape{10}{\magstep1}{OT1TT}
\setfont\defttsl\ttslshape{10}{\magstep1}{OT1TT}
\def\df{\let\tentt=\deftt \let\tenbf = \defbf \let\tenttsl=\defttsl \bf}
% Fonts for indices, footnotes, small examples (9pt).
\def\smallnominalsize{9pt}
\setfont\smallrm\rmshape{9}{1000}{OT1}
\setfont\smalltt\ttshape{9}{1000}{OT1TT}
\setfont\smallbf\bfshape{10}{900}{OT1}
\setfont\smallit\itshape{9}{1000}{OT1IT}
\setfont\smallsl\slshape{9}{1000}{OT1}
\setfont\smallsf\sfshape{9}{1000}{OT1}
\setfont\smallsc\scshape{10}{900}{OT1}
\setfont\smallttsl\ttslshape{10}{900}{OT1TT}
\font\smalli=cmmi9
\font\smallsy=cmsy9
% Fonts for small examples (8pt).
\def\smallernominalsize{8pt}
\setfont\smallerrm\rmshape{8}{1000}{OT1}
\setfont\smallertt\ttshape{8}{1000}{OT1TT}
\setfont\smallerbf\bfshape{10}{800}{OT1}
\setfont\smallerit\itshape{8}{1000}{OT1IT}
\setfont\smallersl\slshape{8}{1000}{OT1}
\setfont\smallersf\sfshape{8}{1000}{OT1}
\setfont\smallersc\scshape{10}{800}{OT1}
\setfont\smallerttsl\ttslshape{10}{800}{OT1TT}
\font\smalleri=cmmi8
\font\smallersy=cmsy8
% Fonts for title page (20.4pt):
\def\titlenominalsize{20pt}
\setfont\titlerm\rmbshape{12}{\magstep3}{OT1}
\setfont\titleit\itbshape{10}{\magstep4}{OT1IT}
\setfont\titlesl\slbshape{10}{\magstep4}{OT1}
\setfont\titlett\ttbshape{12}{\magstep3}{OT1TT}
\setfont\titlettsl\ttslshape{10}{\magstep4}{OT1TT}
\setfont\titlesf\sfbshape{17}{\magstep1}{OT1}
\let\titlebf=\titlerm
\setfont\titlesc\scbshape{10}{\magstep4}{OT1}
\font\titlei=cmmi12 scaled \magstep3
\font\titlesy=cmsy10 scaled \magstep4
\def\authorrm{\secrm}
\def\authortt{\sectt}
% Chapter (and unnumbered) fonts (17.28pt).
\def\chapnominalsize{17pt}
\setfont\chaprm\rmbshape{12}{\magstep2}{OT1}
\setfont\chapit\itbshape{10}{\magstep3}{OT1IT}
\setfont\chapsl\slbshape{10}{\magstep3}{OT1}
\setfont\chaptt\ttbshape{12}{\magstep2}{OT1TT}
\setfont\chapttsl\ttslshape{10}{\magstep3}{OT1TT}
\setfont\chapsf\sfbshape{17}{1000}{OT1}
\let\chapbf=\chaprm
\setfont\chapsc\scbshape{10}{\magstep3}{OT1}
\font\chapi=cmmi12 scaled \magstep2
\font\chapsy=cmsy10 scaled \magstep3
% Section fonts (14.4pt).
\def\secnominalsize{14pt}
\setfont\secrm\rmbshape{12}{\magstep1}{OT1}
\setfont\secit\itbshape{10}{\magstep2}{OT1IT}
\setfont\secsl\slbshape{10}{\magstep2}{OT1}
\setfont\sectt\ttbshape{12}{\magstep1}{OT1TT}
\setfont\secttsl\ttslshape{10}{\magstep2}{OT1TT}
\setfont\secsf\sfbshape{12}{\magstep1}{OT1}
\let\secbf\secrm
\setfont\secsc\scbshape{10}{\magstep2}{OT1}
\font\seci=cmmi12 scaled \magstep1
\font\secsy=cmsy10 scaled \magstep2
% Subsection fonts (13.15pt).
\def\ssecnominalsize{13pt}
\setfont\ssecrm\rmbshape{12}{\magstephalf}{OT1}
\setfont\ssecit\itbshape{10}{1315}{OT1IT}
\setfont\ssecsl\slbshape{10}{1315}{OT1}
\setfont\ssectt\ttbshape{12}{\magstephalf}{OT1TT}
\setfont\ssecttsl\ttslshape{10}{1315}{OT1TT}
\setfont\ssecsf\sfbshape{12}{\magstephalf}{OT1}
\let\ssecbf\ssecrm
\setfont\ssecsc\scbshape{10}{1315}{OT1}
\font\sseci=cmmi12 scaled \magstephalf
\font\ssecsy=cmsy10 scaled 1315
% Reduced fonts for @acro in text (10pt).
\def\reducednominalsize{10pt}
\setfont\reducedrm\rmshape{10}{1000}{OT1}
\setfont\reducedtt\ttshape{10}{1000}{OT1TT}
\setfont\reducedbf\bfshape{10}{1000}{OT1}
\setfont\reducedit\itshape{10}{1000}{OT1IT}
\setfont\reducedsl\slshape{10}{1000}{OT1}
\setfont\reducedsf\sfshape{10}{1000}{OT1}
\setfont\reducedsc\scshape{10}{1000}{OT1}
\setfont\reducedttsl\ttslshape{10}{1000}{OT1TT}
\font\reducedi=cmmi10
\font\reducedsy=cmsy10
% reset the current fonts
\textfonts
\rm
} % end of 11pt text font size definitions
% Definitions to make the main text be 10pt Computer Modern, with
% section, chapter, etc., sizes following suit. This is for the GNU
% Press printing of the Emacs 22 manual. Maybe other manuals in the
% future. Used with @smallbook, which sets the leading to 12pt.
%
\def\definetextfontsizex{%
% Text fonts (10pt).
\def\textnominalsize{10pt}
\edef\mainmagstep{1000}
\setfont\textrm\rmshape{10}{\mainmagstep}{OT1}
\setfont\texttt\ttshape{10}{\mainmagstep}{OT1TT}
\setfont\textbf\bfshape{10}{\mainmagstep}{OT1}
\setfont\textit\itshape{10}{\mainmagstep}{OT1IT}
\setfont\textsl\slshape{10}{\mainmagstep}{OT1}
\setfont\textsf\sfshape{10}{\mainmagstep}{OT1}
\setfont\textsc\scshape{10}{\mainmagstep}{OT1}
\setfont\textttsl\ttslshape{10}{\mainmagstep}{OT1TT}
\font\texti=cmmi10 scaled \mainmagstep
\font\textsy=cmsy10 scaled \mainmagstep
% A few fonts for @defun names and args.
\setfont\defbf\bfshape{10}{\magstephalf}{OT1}
\setfont\deftt\ttshape{10}{\magstephalf}{OT1TT}
\setfont\defttsl\ttslshape{10}{\magstephalf}{OT1TT}
\def\df{\let\tentt=\deftt \let\tenbf = \defbf \let\tenttsl=\defttsl \bf}
% Fonts for indices, footnotes, small examples (9pt).
\def\smallnominalsize{9pt}
\setfont\smallrm\rmshape{9}{1000}{OT1}
\setfont\smalltt\ttshape{9}{1000}{OT1TT}
\setfont\smallbf\bfshape{10}{900}{OT1}
\setfont\smallit\itshape{9}{1000}{OT1IT}
\setfont\smallsl\slshape{9}{1000}{OT1}
\setfont\smallsf\sfshape{9}{1000}{OT1}
\setfont\smallsc\scshape{10}{900}{OT1}
\setfont\smallttsl\ttslshape{10}{900}{OT1TT}
\font\smalli=cmmi9
\font\smallsy=cmsy9
% Fonts for small examples (8pt).
\def\smallernominalsize{8pt}
\setfont\smallerrm\rmshape{8}{1000}{OT1}
\setfont\smallertt\ttshape{8}{1000}{OT1TT}
\setfont\smallerbf\bfshape{10}{800}{OT1}
\setfont\smallerit\itshape{8}{1000}{OT1IT}
\setfont\smallersl\slshape{8}{1000}{OT1}
\setfont\smallersf\sfshape{8}{1000}{OT1}
\setfont\smallersc\scshape{10}{800}{OT1}
\setfont\smallerttsl\ttslshape{10}{800}{OT1TT}
\font\smalleri=cmmi8
\font\smallersy=cmsy8
% Fonts for title page (20.4pt):
\def\titlenominalsize{20pt}
\setfont\titlerm\rmbshape{12}{\magstep3}{OT1}
\setfont\titleit\itbshape{10}{\magstep4}{OT1IT}
\setfont\titlesl\slbshape{10}{\magstep4}{OT1}
\setfont\titlett\ttbshape{12}{\magstep3}{OT1TT}
\setfont\titlettsl\ttslshape{10}{\magstep4}{OT1TT}
\setfont\titlesf\sfbshape{17}{\magstep1}{OT1}
\let\titlebf=\titlerm
\setfont\titlesc\scbshape{10}{\magstep4}{OT1}
\font\titlei=cmmi12 scaled \magstep3
\font\titlesy=cmsy10 scaled \magstep4
\def\authorrm{\secrm}
\def\authortt{\sectt}
% Chapter fonts (14.4pt).
\def\chapnominalsize{14pt}
\setfont\chaprm\rmbshape{12}{\magstep1}{OT1}
\setfont\chapit\itbshape{10}{\magstep2}{OT1IT}
\setfont\chapsl\slbshape{10}{\magstep2}{OT1}
\setfont\chaptt\ttbshape{12}{\magstep1}{OT1TT}
\setfont\chapttsl\ttslshape{10}{\magstep2}{OT1TT}
\setfont\chapsf\sfbshape{12}{\magstep1}{OT1}
\let\chapbf\chaprm
\setfont\chapsc\scbshape{10}{\magstep2}{OT1}
\font\chapi=cmmi12 scaled \magstep1
\font\chapsy=cmsy10 scaled \magstep2
% Section fonts (12pt).
\def\secnominalsize{12pt}
\setfont\secrm\rmbshape{12}{1000}{OT1}
\setfont\secit\itbshape{10}{\magstep1}{OT1IT}
\setfont\secsl\slbshape{10}{\magstep1}{OT1}
\setfont\sectt\ttbshape{12}{1000}{OT1TT}
\setfont\secttsl\ttslshape{10}{\magstep1}{OT1TT}
\setfont\secsf\sfbshape{12}{1000}{OT1}
\let\secbf\secrm
\setfont\secsc\scbshape{10}{\magstep1}{OT1}
\font\seci=cmmi12
\font\secsy=cmsy10 scaled \magstep1
% Subsection fonts (10pt).
\def\ssecnominalsize{10pt}
\setfont\ssecrm\rmbshape{10}{1000}{OT1}
\setfont\ssecit\itbshape{10}{1000}{OT1IT}
\setfont\ssecsl\slbshape{10}{1000}{OT1}
\setfont\ssectt\ttbshape{10}{1000}{OT1TT}
\setfont\ssecttsl\ttslshape{10}{1000}{OT1TT}
\setfont\ssecsf\sfbshape{10}{1000}{OT1}
\let\ssecbf\ssecrm
\setfont\ssecsc\scbshape{10}{1000}{OT1}
\font\sseci=cmmi10
\font\ssecsy=cmsy10
% Reduced fonts for @acro in text (9pt).
\def\reducednominalsize{9pt}
\setfont\reducedrm\rmshape{9}{1000}{OT1}
\setfont\reducedtt\ttshape{9}{1000}{OT1TT}
\setfont\reducedbf\bfshape{10}{900}{OT1}
\setfont\reducedit\itshape{9}{1000}{OT1IT}
\setfont\reducedsl\slshape{9}{1000}{OT1}
\setfont\reducedsf\sfshape{9}{1000}{OT1}
\setfont\reducedsc\scshape{10}{900}{OT1}
\setfont\reducedttsl\ttslshape{10}{900}{OT1TT}
\font\reducedi=cmmi9
\font\reducedsy=cmsy9
% reduce space between paragraphs
\divide\parskip by 2
% reset the current fonts
\textfonts
\rm
} % end of 10pt text font size definitions
% We provide the user-level command
% @fonttextsize 10
% (or 11) to redefine the text font size. pt is assumed.
%
\def\xword{10}
\def\xiword{11}
%
\parseargdef\fonttextsize{%
\def\textsizearg{#1}%
\wlog{doing @fonttextsize \textsizearg}%
%
% Set \globaldefs so that documents can use this inside @tex, since
% makeinfo 4.8 does not support it, but we need it nonetheless.
%
\begingroup \globaldefs=1
\ifx\textsizearg\xword \definetextfontsizex
\else \ifx\textsizearg\xiword \definetextfontsizexi
\else
\errhelp=\EMsimple
\errmessage{@fonttextsize only supports `10' or `11', not `\textsizearg'}
\fi\fi
\endgroup
}
% In order for the font changes to affect most math symbols and letters,
% we have to define the \textfont of the standard families. Since
% texinfo doesn't allow for producing subscripts and superscripts except
% in the main text, we don't bother to reset \scriptfont and
% \scriptscriptfont (which would also require loading a lot more fonts).
%
\def\resetmathfonts{%
\textfont0=\tenrm \textfont1=\teni \textfont2=\tensy
\textfont\itfam=\tenit \textfont\slfam=\tensl \textfont\bffam=\tenbf
\textfont\ttfam=\tentt \textfont\sffam=\tensf
}
% The font-changing commands redefine the meanings of \tenSTYLE, instead
% of just \STYLE. We do this because \STYLE needs to also set the
% current \fam for math mode. Our \STYLE (e.g., \rm) commands hardwire
% \tenSTYLE to set the current font.
%
% Each font-changing command also sets the names \lsize (one size lower)
% and \lllsize (three sizes lower). These relative commands are used in
% the LaTeX logo and acronyms.
%
% This all needs generalizing, badly.
%
\def\textfonts{%
\let\tenrm=\textrm \let\tenit=\textit \let\tensl=\textsl
\let\tenbf=\textbf \let\tentt=\texttt \let\smallcaps=\textsc
\let\tensf=\textsf \let\teni=\texti \let\tensy=\textsy
\let\tenttsl=\textttsl
\def\curfontsize{text}%
\def\lsize{reduced}\def\lllsize{smaller}%
\resetmathfonts \setleading{\textleading}}
\def\titlefonts{%
\let\tenrm=\titlerm \let\tenit=\titleit \let\tensl=\titlesl
\let\tenbf=\titlebf \let\tentt=\titlett \let\smallcaps=\titlesc
\let\tensf=\titlesf \let\teni=\titlei \let\tensy=\titlesy
\let\tenttsl=\titlettsl
\def\curfontsize{title}%
\def\lsize{chap}\def\lllsize{subsec}%
\resetmathfonts \setleading{25pt}}
\def\titlefont#1{{\titlefonts\rm #1}}
\def\chapfonts{%
\let\tenrm=\chaprm \let\tenit=\chapit \let\tensl=\chapsl
\let\tenbf=\chapbf \let\tentt=\chaptt \let\smallcaps=\chapsc
\let\tensf=\chapsf \let\teni=\chapi \let\tensy=\chapsy
\let\tenttsl=\chapttsl
\def\curfontsize{chap}%
\def\lsize{sec}\def\lllsize{text}%
\resetmathfonts \setleading{19pt}}
\def\secfonts{%
\let\tenrm=\secrm \let\tenit=\secit \let\tensl=\secsl
\let\tenbf=\secbf \let\tentt=\sectt \let\smallcaps=\secsc
\let\tensf=\secsf \let\teni=\seci \let\tensy=\secsy
\let\tenttsl=\secttsl
\def\curfontsize{sec}%
\def\lsize{subsec}\def\lllsize{reduced}%
\resetmathfonts \setleading{16pt}}
\def\subsecfonts{%
\let\tenrm=\ssecrm \let\tenit=\ssecit \let\tensl=\ssecsl
\let\tenbf=\ssecbf \let\tentt=\ssectt \let\smallcaps=\ssecsc
\let\tensf=\ssecsf \let\teni=\sseci \let\tensy=\ssecsy
\let\tenttsl=\ssecttsl
\def\curfontsize{ssec}%
\def\lsize{text}\def\lllsize{small}%
\resetmathfonts \setleading{15pt}}
\let\subsubsecfonts = \subsecfonts
\def\reducedfonts{%
\let\tenrm=\reducedrm \let\tenit=\reducedit \let\tensl=\reducedsl
\let\tenbf=\reducedbf \let\tentt=\reducedtt \let\reducedcaps=\reducedsc
\let\tensf=\reducedsf \let\teni=\reducedi \let\tensy=\reducedsy
\let\tenttsl=\reducedttsl
\def\curfontsize{reduced}%
\def\lsize{small}\def\lllsize{smaller}%
\resetmathfonts \setleading{10.5pt}}
\def\smallfonts{%
\let\tenrm=\smallrm \let\tenit=\smallit \let\tensl=\smallsl
\let\tenbf=\smallbf \let\tentt=\smalltt \let\smallcaps=\smallsc
\let\tensf=\smallsf \let\teni=\smalli \let\tensy=\smallsy
\let\tenttsl=\smallttsl
\def\curfontsize{small}%
\def\lsize{smaller}\def\lllsize{smaller}%
\resetmathfonts \setleading{10.5pt}}
\def\smallerfonts{%
\let\tenrm=\smallerrm \let\tenit=\smallerit \let\tensl=\smallersl
\let\tenbf=\smallerbf \let\tentt=\smallertt \let\smallcaps=\smallersc
\let\tensf=\smallersf \let\teni=\smalleri \let\tensy=\smallersy
\let\tenttsl=\smallerttsl
\def\curfontsize{smaller}%
\def\lsize{smaller}\def\lllsize{smaller}%
\resetmathfonts \setleading{9.5pt}}
% Set the fonts to use with the @small... environments.
\let\smallexamplefonts = \smallfonts
% About \smallexamplefonts. If we use \smallfonts (9pt), @smallexample
% can fit this many characters:
% 8.5x11=86 smallbook=72 a4=90 a5=69
% If we use \scriptfonts (8pt), then we can fit this many characters:
% 8.5x11=90+ smallbook=80 a4=90+ a5=77
% For me, subjectively, the few extra characters that fit aren't worth
% the additional smallness of 8pt. So I'm making the default 9pt.
%
% By the way, for comparison, here's what fits with @example (10pt):
% 8.5x11=71 smallbook=60 a4=75 a5=58
%
% I wish the USA used A4 paper.
% --karl, 24jan03.
% Set up the default fonts, so we can use them for creating boxes.
%
\definetextfontsizexi
% Define these so they can be easily changed for other fonts.
\def\angleleft{$\langle$}
\def\angleright{$\rangle$}
% Count depth in font-changes, for error checks
\newcount\fontdepth \fontdepth=0
% Fonts for short table of contents.
\setfont\shortcontrm\rmshape{12}{1000}{OT1}
\setfont\shortcontbf\bfshape{10}{\magstep1}{OT1} % no cmb12
\setfont\shortcontsl\slshape{12}{1000}{OT1}
\setfont\shortconttt\ttshape{12}{1000}{OT1TT}
%% Add scribe-like font environments, plus @l for inline lisp (usually sans
%% serif) and @ii for TeX italic
% \smartitalic{ARG} outputs arg in italics, followed by an italic correction
% unless the following character is such as not to need one.
\def\smartitalicx{\ifx\next,\else\ifx\next-\else\ifx\next.\else
\ptexslash\fi\fi\fi}
\def\smartslanted#1{{\ifusingtt\ttsl\sl #1}\futurelet\next\smartitalicx}
\def\smartitalic#1{{\ifusingtt\ttsl\it #1}\futurelet\next\smartitalicx}
% like \smartslanted except unconditionally uses \ttsl.
% @var is set to this for defun arguments.
\def\ttslanted#1{{\ttsl #1}\futurelet\next\smartitalicx}
% like \smartslanted except unconditionally use \sl. We never want
% ttsl for book titles, do we?
\def\cite#1{{\sl #1}\futurelet\next\smartitalicx}
\let\i=\smartitalic
\let\slanted=\smartslanted
\let\var=\smartslanted
\let\dfn=\smartslanted
\let\emph=\smartitalic
% @b, explicit bold.
\def\b#1{{\bf #1}}
\let\strong=\b
% @sansserif, explicit sans.
\def\sansserif#1{{\sf #1}}
% We can't just use \exhyphenpenalty, because that only has effect at
% the end of a paragraph. Restore normal hyphenation at the end of the
% group within which \nohyphenation is presumably called.
%
\def\nohyphenation{\hyphenchar\font = -1 \aftergroup\restorehyphenation}
\def\restorehyphenation{\hyphenchar\font = `- }
% Set sfcode to normal for the chars that usually have another value.
% Can't use plain's \frenchspacing because it uses the `\x notation, and
% sometimes \x has an active definition that messes things up.
%
\catcode`@=11
\def\plainfrenchspacing{%
\sfcode\dotChar =\@m \sfcode\questChar=\@m \sfcode\exclamChar=\@m
\sfcode\colonChar=\@m \sfcode\semiChar =\@m \sfcode\commaChar =\@m
\def\endofsentencespacefactor{1000}% for @. and friends
}
\def\plainnonfrenchspacing{%
\sfcode`\.3000\sfcode`\?3000\sfcode`\!3000
\sfcode`\:2000\sfcode`\;1500\sfcode`\,1250
\def\endofsentencespacefactor{3000}% for @. and friends
}
\catcode`@=\other
\def\endofsentencespacefactor{3000}% default
\def\t#1{%
{\tt \rawbackslash \plainfrenchspacing #1}%
\null
}
\def\samp#1{`\tclose{#1}'\null}
\setfont\keyrm\rmshape{8}{1000}{OT1}
\font\keysy=cmsy9
\def\key#1{{\keyrm\textfont2=\keysy \leavevmode\hbox{%
\raise0.4pt\hbox{\angleleft}\kern-.08em\vtop{%
\vbox{\hrule\kern-0.4pt
\hbox{\raise0.4pt\hbox{\vphantom{\angleleft}}#1}}%
\kern-0.4pt\hrule}%
\kern-.06em\raise0.4pt\hbox{\angleright}}}}
\def\key #1{{\nohyphenation \uppercase{#1}}\null}
% The old definition, with no lozenge:
%\def\key #1{{\ttsl \nohyphenation \uppercase{#1}}\null}
\def\ctrl #1{{\tt \rawbackslash \hat}#1}
% @file, @option are the same as @samp.
\let\file=\samp
\let\option=\samp
% @code is a modification of @t,
% which makes spaces the same size as normal in the surrounding text.
\def\tclose#1{%
{%
% Change normal interword space to be same as for the current font.
\spaceskip = \fontdimen2\font
%
% Switch to typewriter.
\tt
%
% But `\ ' produces the large typewriter interword space.
\def\ {{\spaceskip = 0pt{} }}%
%
% Turn off hyphenation.
\nohyphenation
%
\rawbackslash
\plainfrenchspacing
#1%
}%
\null
}
% We *must* turn on hyphenation at `-' and `_' in @code.
% Otherwise, it is too hard to avoid overfull hboxes
% in the Emacs manual, the Library manual, etc.
% Unfortunately, TeX uses one parameter (\hyphenchar) to control
% both hyphenation at - and hyphenation within words.
% We must therefore turn them both off (\tclose does that)
% and arrange explicitly to hyphenate at a dash.
% -- rms.
{
\catcode`\-=\active \catcode`\_=\active
\catcode`\'=\active \catcode`\`=\active
%
\global\def\code{\begingroup
\catcode\rquoteChar=\active \catcode\lquoteChar=\active
\let'\codequoteright \let`\codequoteleft
%
\catcode\dashChar=\active \catcode\underChar=\active
\ifallowcodebreaks
\let-\codedash
\let_\codeunder
\else
\let-\realdash
\let_\realunder
\fi
\codex
}
}
\def\realdash{-}
\def\codedash{-\discretionary{}{}{}}
\def\codeunder{%
% this is all so @math{@code{var_name}+1} can work. In math mode, _
% is "active" (mathcode"8000) and \normalunderscore (or \char95, etc.)
% will therefore expand the active definition of _, which is us
% (inside @code that is), therefore an endless loop.
\ifusingtt{\ifmmode
\mathchar"075F % class 0=ordinary, family 7=ttfam, pos 0x5F=_.
\else\normalunderscore \fi
\discretionary{}{}{}}%
{\_}%
}
\def\codex #1{\tclose{#1}\endgroup}
% An additional complication: the above will allow breaks after, e.g.,
% each of the four underscores in __typeof__. This is undesirable in
% some manuals, especially if they don't have long identifiers in
% general. @allowcodebreaks provides a way to control this.
%
\newif\ifallowcodebreaks \allowcodebreakstrue
\def\keywordtrue{true}
\def\keywordfalse{false}
\parseargdef\allowcodebreaks{%
\def\txiarg{#1}%
\ifx\txiarg\keywordtrue
\allowcodebreakstrue
\else\ifx\txiarg\keywordfalse
\allowcodebreaksfalse
\else
\errhelp = \EMsimple
\errmessage{Unknown @allowcodebreaks option `\txiarg'}%
\fi\fi
}
% @kbd is like @code, except that if the argument is just one @key command,
% then @kbd has no effect.
% @kbdinputstyle -- arg is `distinct' (@kbd uses slanted tty font always),
% `example' (@kbd uses ttsl only inside of @example and friends),
% or `code' (@kbd uses normal tty font always).
\parseargdef\kbdinputstyle{%
\def\txiarg{#1}%
\ifx\txiarg\worddistinct
\gdef\kbdexamplefont{\ttsl}\gdef\kbdfont{\ttsl}%
\else\ifx\txiarg\wordexample
\gdef\kbdexamplefont{\ttsl}\gdef\kbdfont{\tt}%
\else\ifx\txiarg\wordcode
\gdef\kbdexamplefont{\tt}\gdef\kbdfont{\tt}%
\else
\errhelp = \EMsimple
\errmessage{Unknown @kbdinputstyle option `\txiarg'}%
\fi\fi\fi
}
\def\worddistinct{distinct}
\def\wordexample{example}
\def\wordcode{code}
% Default is `distinct.'
\kbdinputstyle distinct
\def\xkey{\key}
\def\kbdfoo#1#2#3\par{\def\one{#1}\def\three{#3}\def\threex{??}%
\ifx\one\xkey\ifx\threex\three \key{#2}%
\else{\tclose{\kbdfont\look}}\fi
\else{\tclose{\kbdfont\look}}\fi}
% For @indicateurl, @env, @command quotes seem unnecessary, so use \code.
\let\indicateurl=\code
\let\env=\code
\let\command=\code
% @uref (abbreviation for `urlref') takes an optional (comma-separated)
% second argument specifying the text to display and an optional third
% arg as text to display instead of (rather than in addition to) the url
% itself. First (mandatory) arg is the url. Perhaps eventually put in
% a hypertex \special here.
%
\def\uref#1{\douref #1,,,\finish}
\def\douref#1,#2,#3,#4\finish{\begingroup
\unsepspaces
\pdfurl{#1}%
\setbox0 = \hbox{\ignorespaces #3}%
\ifdim\wd0 > 0pt
\unhbox0 % third arg given, show only that
\else
\setbox0 = \hbox{\ignorespaces #2}%
\ifdim\wd0 > 0pt
\ifpdf
\unhbox0 % PDF: 2nd arg given, show only it
\else
\unhbox0\ (\code{#1})% DVI: 2nd arg given, show both it and url
\fi
\else
\code{#1}% only url given, so show it
\fi
\fi
\endlink
\endgroup}
% @url synonym for @uref, since that's how everyone uses it.
%
\let\url=\uref
% rms does not like angle brackets --karl, 17may97.
% So now @email is just like @uref, unless we are pdf.
%
%\def\email#1{\angleleft{\tt #1}\angleright}
\ifpdf
\def\email#1{\doemail#1,,\finish}
\def\doemail#1,#2,#3\finish{\begingroup
\unsepspaces
\pdfurl{mailto:#1}%
\setbox0 = \hbox{\ignorespaces #2}%
\ifdim\wd0>0pt\unhbox0\else\code{#1}\fi
\endlink
\endgroup}
\else
\let\email=\uref
\fi
% Check if we are currently using a typewriter font. Since all the
% Computer Modern typewriter fonts have zero interword stretch (and
% shrink), and it is reasonable to expect all typewriter fonts to have
% this property, we can check that font parameter.
%
\def\ifmonospace{\ifdim\fontdimen3\font=0pt }
% Typeset a dimension, e.g., `in' or `pt'. The only reason for the
% argument is to make the input look right: @dmn{pt} instead of @dmn{}pt.
%
\def\dmn#1{\thinspace #1}
\def\kbd#1{\def\look{#1}\expandafter\kbdfoo\look??\par}
% @l was never documented to mean ``switch to the Lisp font'',
% and it is not used as such in any manual I can find. We need it for
% Polish suppressed-l. --karl, 22sep96.
%\def\l#1{{\li #1}\null}
% Explicit font changes: @r, @sc, undocumented @ii.
\def\r#1{{\rm #1}} % roman font
\def\sc#1{{\smallcaps#1}} % smallcaps font
\def\ii#1{{\it #1}} % italic font
% @acronym for "FBI", "NATO", and the like.
% We print this one point size smaller, since it's intended for
% all-uppercase.
%
\def\acronym#1{\doacronym #1,,\finish}
\def\doacronym#1,#2,#3\finish{%
{\selectfonts\lsize #1}%
\def\temp{#2}%
\ifx\temp\empty \else
\space ({\unsepspaces \ignorespaces \temp \unskip})%
\fi
}
% @abbr for "Comput. J." and the like.
% No font change, but don't do end-of-sentence spacing.
%
\def\abbr#1{\doabbr #1,,\finish}
\def\doabbr#1,#2,#3\finish{%
{\plainfrenchspacing #1}%
\def\temp{#2}%
\ifx\temp\empty \else
\space ({\unsepspaces \ignorespaces \temp \unskip})%
\fi
}
% @pounds{} is a sterling sign, which Knuth put in the CM italic font.
%
\def\pounds{{\it\$}}
% @euro{} comes from a separate font, depending on the current style.
% We use the free feym* fonts from the eurosym package by Henrik
% Theiling, which support regular, slanted, bold and bold slanted (and
% "outlined" (blackboard board, sort of) versions, which we don't need).
% It is available from http://www.ctan.org/tex-archive/fonts/eurosym.
%
% Although only regular is the truly official Euro symbol, we ignore
% that. The Euro is designed to be slightly taller than the regular
% font height.
%
% feymr - regular
% feymo - slanted
% feybr - bold
% feybo - bold slanted
%
% There is no good (free) typewriter version, to my knowledge.
% A feymr10 euro is ~7.3pt wide, while a normal cmtt10 char is ~5.25pt wide.
% Hmm.
%
% Also doesn't work in math. Do we need to do math with euro symbols?
% Hope not.
%
%
\def\euro{{\eurofont e}}
\def\eurofont{%
% We set the font at each command, rather than predefining it in
% \textfonts and the other font-switching commands, so that
% installations which never need the symbol don't have to have the
% font installed.
%
% There is only one designed size (nominal 10pt), so we always scale
% that to the current nominal size.
%
% By the way, simply using "at 1em" works for cmr10 and the like, but
% does not work for cmbx10 and other extended/shrunken fonts.
%
\def\eurosize{\csname\curfontsize nominalsize\endcsname}%
%
\ifx\curfontstyle\bfstylename
% bold:
\font\thiseurofont = \ifusingit{feybo10}{feybr10} at \eurosize
\else
% regular:
\font\thiseurofont = \ifusingit{feymo10}{feymr10} at \eurosize
\fi
\thiseurofont
}
% @registeredsymbol - R in a circle. The font for the R should really
% be smaller yet, but lllsize is the best we can do for now.
% Adapted from the plain.tex definition of \copyright.
%
\def\registeredsymbol{%
$^{{\ooalign{\hfil\raise.07ex\hbox{\selectfonts\lllsize R}%
\hfil\crcr\Orb}}%
}$%
}
% @textdegree - the normal degrees sign.
%
\def\textdegree{$^\circ$}
% Laurent Siebenmann reports \Orb undefined with:
% Textures 1.7.7 (preloaded format=plain 93.10.14) (68K) 16 APR 2004 02:38
% so we'll define it if necessary.
%
\ifx\Orb\undefined
\def\Orb{\mathhexbox20D}
\fi
\message{page headings,}
\newskip\titlepagetopglue \titlepagetopglue = 1.5in
\newskip\titlepagebottomglue \titlepagebottomglue = 2pc
% First the title page. Must do @settitle before @titlepage.
\newif\ifseenauthor
\newif\iffinishedtitlepage
% Do an implicit @contents or @shortcontents after @end titlepage if the
% user says @setcontentsaftertitlepage or @setshortcontentsaftertitlepage.
%
\newif\ifsetcontentsaftertitlepage
\let\setcontentsaftertitlepage = \setcontentsaftertitlepagetrue
\newif\ifsetshortcontentsaftertitlepage
\let\setshortcontentsaftertitlepage = \setshortcontentsaftertitlepagetrue
\parseargdef\shorttitlepage{\begingroup\hbox{}\vskip 1.5in \chaprm \centerline{#1}%
\endgroup\page\hbox{}\page}
\envdef\titlepage{%
% Open one extra group, as we want to close it in the middle of \Etitlepage.
\begingroup
\parindent=0pt \textfonts
% Leave some space at the very top of the page.
\vglue\titlepagetopglue
% No rule at page bottom unless we print one at the top with @title.
\finishedtitlepagetrue
%
% Most title ``pages'' are actually two pages long, with space
% at the top of the second. We don't want the ragged left on the second.
\let\oldpage = \page
\def\page{%
\iffinishedtitlepage\else
\finishtitlepage
\fi
\let\page = \oldpage
\page
\null
}%
}
\def\Etitlepage{%
\iffinishedtitlepage\else
\finishtitlepage
\fi
% It is important to do the page break before ending the group,
% because the headline and footline are only empty inside the group.
% If we use the new definition of \page, we always get a blank page
% after the title page, which we certainly don't want.
\oldpage
\endgroup
%
% Need this before the \...aftertitlepage checks so that if they are
% in effect the toc pages will come out with page numbers.
\HEADINGSon
%
% If they want short, they certainly want long too.
\ifsetshortcontentsaftertitlepage
\shortcontents
\contents
\global\let\shortcontents = \relax
\global\let\contents = \relax
\fi
%
\ifsetcontentsaftertitlepage
\contents
\global\let\contents = \relax
\global\let\shortcontents = \relax
\fi
}
\def\finishtitlepage{%
\vskip4pt \hrule height 2pt width \hsize
\vskip\titlepagebottomglue
\finishedtitlepagetrue
}
%%% Macros to be used within @titlepage:
\let\subtitlerm=\tenrm
\def\subtitlefont{\subtitlerm \normalbaselineskip = 13pt \normalbaselines}
\def\authorfont{\authorrm \normalbaselineskip = 16pt \normalbaselines
\let\tt=\authortt}
\parseargdef\title{%
\checkenv\titlepage
\leftline{\titlefonts\rm #1}
% print a rule at the page bottom also.
\finishedtitlepagefalse
\vskip4pt \hrule height 4pt width \hsize \vskip4pt
}
\parseargdef\subtitle{%
\checkenv\titlepage
{\subtitlefont \rightline{#1}}%
}
% @author should come last, but may come many times.
% It can also be used inside @quotation.
%
\parseargdef\author{%
\def\temp{\quotation}%
\ifx\thisenv\temp
\def\quotationauthor{#1}% printed in \Equotation.
\else
\checkenv\titlepage
\ifseenauthor\else \vskip 0pt plus 1filll \seenauthortrue \fi
{\authorfont \leftline{#1}}%
\fi
}
%%% Set up page headings and footings.
\let\thispage=\folio
\newtoks\evenheadline % headline on even pages
\newtoks\oddheadline % headline on odd pages
\newtoks\evenfootline % footline on even pages
\newtoks\oddfootline % footline on odd pages
% Now make TeX use those variables
\headline={{\textfonts\rm \ifodd\pageno \the\oddheadline
\else \the\evenheadline \fi}}
\footline={{\textfonts\rm \ifodd\pageno \the\oddfootline
\else \the\evenfootline \fi}\HEADINGShook}
\let\HEADINGShook=\relax
% Commands to set those variables.
% For example, this is what @headings on does
% @evenheading @thistitle|@thispage|@thischapter
% @oddheading @thischapter|@thispage|@thistitle
% @evenfooting @thisfile||
% @oddfooting ||@thisfile
\def\evenheading{\parsearg\evenheadingxxx}
\def\evenheadingxxx #1{\evenheadingyyy #1\|\|\|\|\finish}
\def\evenheadingyyy #1\|#2\|#3\|#4\finish{%
\global\evenheadline={\rlap{\centerline{#2}}\line{#1\hfil#3}}}
\def\oddheading{\parsearg\oddheadingxxx}
\def\oddheadingxxx #1{\oddheadingyyy #1\|\|\|\|\finish}
\def\oddheadingyyy #1\|#2\|#3\|#4\finish{%
\global\oddheadline={\rlap{\centerline{#2}}\line{#1\hfil#3}}}
\parseargdef\everyheading{\oddheadingxxx{#1}\evenheadingxxx{#1}}%
\def\evenfooting{\parsearg\evenfootingxxx}
\def\evenfootingxxx #1{\evenfootingyyy #1\|\|\|\|\finish}
\def\evenfootingyyy #1\|#2\|#3\|#4\finish{%
\global\evenfootline={\rlap{\centerline{#2}}\line{#1\hfil#3}}}
\def\oddfooting{\parsearg\oddfootingxxx}
\def\oddfootingxxx #1{\oddfootingyyy #1\|\|\|\|\finish}
\def\oddfootingyyy #1\|#2\|#3\|#4\finish{%
\global\oddfootline = {\rlap{\centerline{#2}}\line{#1\hfil#3}}%
%
% Leave some space for the footline. Hopefully ok to assume
% @evenfooting will not be used by itself.
\global\advance\pageheight by -12pt
\global\advance\vsize by -12pt
}
\parseargdef\everyfooting{\oddfootingxxx{#1}\evenfootingxxx{#1}}
% @headings double turns headings on for double-sided printing.
% @headings single turns headings on for single-sided printing.
% @headings off turns them off.
% @headings on same as @headings double, retained for compatibility.
% @headings after turns on double-sided headings after this page.
% @headings doubleafter turns on double-sided headings after this page.
% @headings singleafter turns on single-sided headings after this page.
% By default, they are off at the start of a document,
% and turned `on' after @end titlepage.
\def\headings #1 {\csname HEADINGS#1\endcsname}
\def\HEADINGSoff{%
\global\evenheadline={\hfil} \global\evenfootline={\hfil}
\global\oddheadline={\hfil} \global\oddfootline={\hfil}}
\HEADINGSoff
% When we turn headings on, set the page number to 1.
% For double-sided printing, put current file name in lower left corner,
% chapter name on inside top of right hand pages, document
% title on inside top of left hand pages, and page numbers on outside top
% edge of all pages.
\def\HEADINGSdouble{%
\global\pageno=1
\global\evenfootline={\hfil}
\global\oddfootline={\hfil}
\global\evenheadline={\line{\folio\hfil\thistitle}}
\global\oddheadline={\line{\thischapter\hfil\folio}}
\global\let\contentsalignmacro = \chapoddpage
}
\let\contentsalignmacro = \chappager
% For single-sided printing, chapter title goes across top left of page,
% page number on top right.
\def\HEADINGSsingle{%
\global\pageno=1
\global\evenfootline={\hfil}
\global\oddfootline={\hfil}
\global\evenheadline={\line{\thischapter\hfil\folio}}
\global\oddheadline={\line{\thischapter\hfil\folio}}
\global\let\contentsalignmacro = \chappager
}
\def\HEADINGSon{\HEADINGSdouble}
\def\HEADINGSafter{\let\HEADINGShook=\HEADINGSdoublex}
\let\HEADINGSdoubleafter=\HEADINGSafter
\def\HEADINGSdoublex{%
\global\evenfootline={\hfil}
\global\oddfootline={\hfil}
\global\evenheadline={\line{\folio\hfil\thistitle}}
\global\oddheadline={\line{\thischapter\hfil\folio}}
\global\let\contentsalignmacro = \chapoddpage
}
\def\HEADINGSsingleafter{\let\HEADINGShook=\HEADINGSsinglex}
\def\HEADINGSsinglex{%
\global\evenfootline={\hfil}
\global\oddfootline={\hfil}
\global\evenheadline={\line{\thischapter\hfil\folio}}
\global\oddheadline={\line{\thischapter\hfil\folio}}
\global\let\contentsalignmacro = \chappager
}
% Subroutines used in generating headings
% This produces Day Month Year style of output.
% Only define if not already defined, in case a txi-??.tex file has set
% up a different format (e.g., txi-cs.tex does this).
\ifx\today\undefined
\def\today{%
\number\day\space
\ifcase\month
\or\putwordMJan\or\putwordMFeb\or\putwordMMar\or\putwordMApr
\or\putwordMMay\or\putwordMJun\or\putwordMJul\or\putwordMAug
\or\putwordMSep\or\putwordMOct\or\putwordMNov\or\putwordMDec
\fi
\space\number\year}
\fi
% @settitle line... specifies the title of the document, for headings.
% It generates no output of its own.
\def\thistitle{\putwordNoTitle}
\def\settitle{\parsearg{\gdef\thistitle}}
\message{tables,}
% Tables -- @table, @ftable, @vtable, @item(x).
% default indentation of table text
\newdimen\tableindent \tableindent=.8in
% default indentation of @itemize and @enumerate text
\newdimen\itemindent \itemindent=.3in
% margin between end of table item and start of table text.
\newdimen\itemmargin \itemmargin=.1in
% used internally for \itemindent minus \itemmargin
\newdimen\itemmax
% Note @table, @ftable, and @vtable define @item, @itemx, etc., with
% these defs.
% They also define \itemindex
% to index the item name in whatever manner is desired (perhaps none).
\newif\ifitemxneedsnegativevskip
\def\itemxpar{\par\ifitemxneedsnegativevskip\nobreak\vskip-\parskip\nobreak\fi}
\def\internalBitem{\smallbreak \parsearg\itemzzz}
\def\internalBitemx{\itemxpar \parsearg\itemzzz}
\def\itemzzz #1{\begingroup %
\advance\hsize by -\rightskip
\advance\hsize by -\tableindent
\setbox0=\hbox{\itemindicate{#1}}%
\itemindex{#1}%
\nobreak % This prevents a break before @itemx.
%
% If the item text does not fit in the space we have, put it on a line
% by itself, and do not allow a page break either before or after that
% line. We do not start a paragraph here because then if the next
% command is, e.g., @kindex, the whatsit would get put into the
% horizontal list on a line by itself, resulting in extra blank space.
\ifdim \wd0>\itemmax
%
% Make this a paragraph so we get the \parskip glue and wrapping,
% but leave it ragged-right.
\begingroup
\advance\leftskip by-\tableindent
\advance\hsize by\tableindent
\advance\rightskip by0pt plus1fil
\leavevmode\unhbox0\par
\endgroup
%
% We're going to be starting a paragraph, but we don't want the
% \parskip glue -- logically it's part of the @item we just started.
\nobreak \vskip-\parskip
%
% Stop a page break at the \parskip glue coming up. However, if
% what follows is an environment such as @example, there will be no
% \parskip glue; then the negative vskip we just inserted would
% cause the example and the item to crash together. So we use this
% bizarre value of 10001 as a signal to \aboveenvbreak to insert
% \parskip glue after all. Section titles are handled this way also.
%
\penalty 10001
\endgroup
\itemxneedsnegativevskipfalse
\else
% The item text fits into the space. Start a paragraph, so that the
% following text (if any) will end up on the same line.
\noindent
% Do this with kerns and \unhbox so that if there is a footnote in
% the item text, it can migrate to the main vertical list and
% eventually be printed.
\nobreak\kern-\tableindent
\dimen0 = \itemmax \advance\dimen0 by \itemmargin \advance\dimen0 by -\wd0
\unhbox0
\nobreak\kern\dimen0
\endgroup
\itemxneedsnegativevskiptrue
\fi
}
\def\item{\errmessage{@item while not in a list environment}}
\def\itemx{\errmessage{@itemx while not in a list environment}}
% @table, @ftable, @vtable.
\envdef\table{%
\let\itemindex\gobble
\tablecheck{table}%
}
\envdef\ftable{%
\def\itemindex ##1{\doind {fn}{\code{##1}}}%
\tablecheck{ftable}%
}
\envdef\vtable{%
\def\itemindex ##1{\doind {vr}{\code{##1}}}%
\tablecheck{vtable}%
}
\def\tablecheck#1{%
\ifnum \the\catcode`\^^M=\active
\endgroup
\errmessage{This command won't work in this context; perhaps the problem is
that we are \inenvironment\thisenv}%
\def\next{\doignore{#1}}%
\else
\let\next\tablex
\fi
\next
}
\def\tablex#1{%
\def\itemindicate{#1}%
\parsearg\tabley
}
\def\tabley#1{%
{%
\makevalueexpandable
\edef\temp{\noexpand\tablez #1\space\space\space}%
\expandafter
}\temp \endtablez
}
\def\tablez #1 #2 #3 #4\endtablez{%
\aboveenvbreak
\ifnum 0#1>0 \advance \leftskip by #1\mil \fi
\ifnum 0#2>0 \tableindent=#2\mil \fi
\ifnum 0#3>0 \advance \rightskip by #3\mil \fi
\itemmax=\tableindent
\advance \itemmax by -\itemmargin
\advance \leftskip by \tableindent
\exdentamount=\tableindent
\parindent = 0pt
\parskip = \smallskipamount
\ifdim \parskip=0pt \parskip=2pt \fi
\let\item = \internalBitem
\let\itemx = \internalBitemx
}
\def\Etable{\endgraf\afterenvbreak}
\let\Eftable\Etable
\let\Evtable\Etable
\let\Eitemize\Etable
\let\Eenumerate\Etable
% This is the counter used by @enumerate, which is really @itemize
\newcount \itemno
\envdef\itemize{\parsearg\doitemize}
\def\doitemize#1{%
\aboveenvbreak
\itemmax=\itemindent
\advance\itemmax by -\itemmargin
\advance\leftskip by \itemindent
\exdentamount=\itemindent
\parindent=0pt
\parskip=\smallskipamount
\ifdim\parskip=0pt \parskip=2pt \fi
\def\itemcontents{#1}%
% @itemize with no arg is equivalent to @itemize @bullet.
\ifx\itemcontents\empty\def\itemcontents{\bullet}\fi
\let\item=\itemizeitem
}
% Definition of @item while inside @itemize and @enumerate.
%
\def\itemizeitem{%
\advance\itemno by 1 % for enumerations
{\let\par=\endgraf \smallbreak}% reasonable place to break
{%
% If the document has an @itemize directly after a section title, a
% \nobreak will be last on the list, and \sectionheading will have
% done a \vskip-\parskip. In that case, we don't want to zero
% parskip, or the item text will crash with the heading. On the
% other hand, when there is normal text preceding the item (as there
% usually is), we do want to zero parskip, or there would be too much
% space. In that case, we won't have a \nobreak before. At least
% that's the theory.
\ifnum\lastpenalty<10000 \parskip=0in \fi
\noindent
\hbox to 0pt{\hss \itemcontents \kern\itemmargin}%
\vadjust{\penalty 1200}}% not good to break after first line of item.
\flushcr
}
% \splitoff TOKENS\endmark defines \first to be the first token in
% TOKENS, and \rest to be the remainder.
%
\def\splitoff#1#2\endmark{\def\first{#1}\def\rest{#2}}%
% Allow an optional argument of an uppercase letter, lowercase letter,
% or number, to specify the first label in the enumerated list. No
% argument is the same as `1'.
%
\envparseargdef\enumerate{\enumeratey #1 \endenumeratey}
\def\enumeratey #1 #2\endenumeratey{%
% If we were given no argument, pretend we were given `1'.
\def\thearg{#1}%
\ifx\thearg\empty \def\thearg{1}\fi
%
% Detect if the argument is a single token. If so, it might be a
% letter. Otherwise, the only valid thing it can be is a number.
% (We will always have one token, because of the test we just made.
% This is a good thing, since \splitoff doesn't work given nothing at
% all -- the first parameter is undelimited.)
\expandafter\splitoff\thearg\endmark
\ifx\rest\empty
% Only one token in the argument. It could still be anything.
% A ``lowercase letter'' is one whose \lccode is nonzero.
% An ``uppercase letter'' is one whose \lccode is both nonzero, and
% not equal to itself.
% Otherwise, we assume it's a number.
%
% We need the \relax at the end of the \ifnum lines to stop TeX from
% continuing to look for a <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:
\obeylines
\catcode`\@ = \other
\catcode`\{ = \other
\catcode`\} = \other
%
% Make sure that spaces turn into tokens that match what \doignoretext wants.
\spaceisspace
%
% Count number of #1's that we've seen.
\doignorecount = 0
%
% Swallow text until we reach the matching `@end #1'.
\dodoignore{#1}%
}
{ \catcode`_=11 % We want to use \_STOP_ which cannot appear in texinfo source.
\obeylines %
%
\gdef\dodoignore#1{%
% #1 contains the command name as a string, e.g., `ifinfo'.
%
% Define a command to find the next `@end #1'.
\long\def\doignoretext##1^^M@end #1{%
\doignoretextyyy##1^^M@#1\_STOP_}%
%
% And this command to find another #1 command, at the beginning of a
% line. (Otherwise, we would consider a line `@c @ifset', for
% example, to count as an @ifset for nesting.)
\long\def\doignoretextyyy##1^^M@#1##2\_STOP_{\doignoreyyy{##2}\_STOP_}%
%
% And now expand that command.
\doignoretext ^^M%
}%
}
\def\doignoreyyy#1{%
\def\temp{#1}%
\ifx\temp\empty % Nothing found.
\let\next\doignoretextzzz
\else % Found a nested condition, ...
\advance\doignorecount by 1
\let\next\doignoretextyyy % ..., look for another.
% If we're here, #1 ends with ^^M\ifinfo (for example).
\fi
\next #1% the token \_STOP_ is present just after this macro.
}
% We have to swallow the remaining "\_STOP_".
%
\def\doignoretextzzz#1{%
\ifnum\doignorecount = 0 % We have just found the outermost @end.
\let\next\enddoignore
\else % Still inside a nested condition.
\advance\doignorecount by -1
\let\next\doignoretext % Look for the next @end.
\fi
\next
}
% Finish off ignored text.
{ \obeylines%
% Ignore anything after the last `@end #1'; this matters in verbatim
% environments, where otherwise the newline after an ignored conditional
% would result in a blank line in the output.
\gdef\enddoignore#1^^M{\endgroup\ignorespaces}%
}
% @set VAR sets the variable VAR to an empty value.
% @set VAR REST-OF-LINE sets VAR to the value REST-OF-LINE.
%
% Since we want to separate VAR from REST-OF-LINE (which might be
% empty), we can't just use \parsearg; we have to insert a space of our
% own to delimit the rest of the line, and then take it out again if we
% didn't need it.
% We rely on the fact that \parsearg sets \catcode`\ =10.
%
\parseargdef\set{\setyyy#1 \endsetyyy}
\def\setyyy#1 #2\endsetyyy{%
{%
\makevalueexpandable
\def\temp{#2}%
\edef\next{\gdef\makecsname{SET#1}}%
\ifx\temp\empty
\next{}%
\else
\setzzz#2\endsetzzz
\fi
}%
}
% Remove the trailing space \setxxx inserted.
\def\setzzz#1 \endsetzzz{\next{#1}}
% @clear VAR clears (i.e., unsets) the variable VAR.
%
\parseargdef\clear{%
{%
\makevalueexpandable
\global\expandafter\let\csname SET#1\endcsname=\relax
}%
}
% @value{foo} gets the text saved in variable foo.
\def\value{\begingroup\makevalueexpandable\valuexxx}
\def\valuexxx#1{\expandablevalue{#1}\endgroup}
{
\catcode`\- = \active \catcode`\_ = \active
%
\gdef\makevalueexpandable{%
\let\value = \expandablevalue
% We don't want these characters active, ...
\catcode`\-=\other \catcode`\_=\other
% ..., but we might end up with active ones in the argument if
% we're called from @code, as @code{@value{foo-bar_}}, though.
% So \let them to their normal equivalents.
\let-\realdash \let_\normalunderscore
}
}
% We have this subroutine so that we can handle at least some @value's
% properly in indexes (we call \makevalueexpandable in \indexdummies).
% The command has to be fully expandable (if the variable is set), since
% the result winds up in the index file. This means that if the
% variable's value contains other Texinfo commands, it's almost certain
% it will fail (although perhaps we could fix that with sufficient work
% to do a one-level expansion on the result, instead of complete).
%
\def\expandablevalue#1{%
\expandafter\ifx\csname SET#1\endcsname\relax
{[No value for ``#1'']}%
\message{Variable `#1', used in @value, is not set.}%
\else
\csname SET#1\endcsname
\fi
}
% @ifset VAR ... @end ifset reads the `...' iff VAR has been defined
% with @set.
%
% To get special treatment of `@end ifset,' call \makeond and the redefine.
%
\makecond{ifset}
\def\ifset{\parsearg{\doifset{\let\next=\ifsetfail}}}
\def\doifset#1#2{%
{%
\makevalueexpandable
\let\next=\empty
\expandafter\ifx\csname SET#2\endcsname\relax
#1% If not set, redefine \next.
\fi
\expandafter
}\next
}
\def\ifsetfail{\doignore{ifset}}
% @ifclear VAR ... @end ifclear reads the `...' iff VAR has never been
% defined with @set, or has been undefined with @clear.
%
% The `\else' inside the `\doifset' parameter is a trick to reuse the
% above code: if the variable is not set, do nothing, if it is set,
% then redefine \next to \ifclearfail.
%
\makecond{ifclear}
\def\ifclear{\parsearg{\doifset{\else \let\next=\ifclearfail}}}
\def\ifclearfail{\doignore{ifclear}}
% @dircategory CATEGORY -- specify a category of the dir file
% which this file should belong to. Ignore this in TeX.
\let\dircategory=\comment
% @defininfoenclose.
\let\definfoenclose=\comment
\message{indexing,}
% Index generation facilities
% Define \newwrite to be identical to plain tex's \newwrite
% except not \outer, so it can be used within macros and \if's.
\edef\newwrite{\makecsname{ptexnewwrite}}
% \newindex {foo} defines an index named foo.
% It automatically defines \fooindex such that
% \fooindex ...rest of line... puts an entry in the index foo.
% It also defines \fooindfile to be the number of the output channel for
% the file that accumulates this index. The file's extension is foo.
% The name of an index should be no more than 2 characters long
% for the sake of vms.
%
\def\newindex#1{%
\iflinks
\expandafter\newwrite \csname#1indfile\endcsname
\openout \csname#1indfile\endcsname \jobname.#1 % Open the file
\fi
\expandafter\xdef\csname#1index\endcsname{% % Define @#1index
\noexpand\doindex{#1}}
}
% @defindex foo == \newindex{foo}
%
\def\defindex{\parsearg\newindex}
% Define @defcodeindex, like @defindex except put all entries in @code.
%
\def\defcodeindex{\parsearg\newcodeindex}
%
\def\newcodeindex#1{%
\iflinks
\expandafter\newwrite \csname#1indfile\endcsname
\openout \csname#1indfile\endcsname \jobname.#1
\fi
\expandafter\xdef\csname#1index\endcsname{%
\noexpand\docodeindex{#1}}%
}
% @synindex foo bar makes index foo feed into index bar.
% Do this instead of @defindex foo if you don't want it as a separate index.
%
% @syncodeindex foo bar similar, but put all entries made for index foo
% inside @code.
%
\def\synindex#1 #2 {\dosynindex\doindex{#1}{#2}}
\def\syncodeindex#1 #2 {\dosynindex\docodeindex{#1}{#2}}
% #1 is \doindex or \docodeindex, #2 the index getting redefined (foo),
% #3 the target index (bar).
\def\dosynindex#1#2#3{%
% Only do \closeout if we haven't already done it, else we'll end up
% closing the target index.
\expandafter \ifx\csname donesynindex#2\endcsname \undefined
% The \closeout helps reduce unnecessary open files; the limit on the
% Acorn RISC OS is a mere 16 files.
\expandafter\closeout\csname#2indfile\endcsname
\expandafter\let\csname\donesynindex#2\endcsname = 1
\fi
% redefine \fooindfile:
\expandafter\let\expandafter\temp\expandafter=\csname#3indfile\endcsname
\expandafter\let\csname#2indfile\endcsname=\temp
% redefine \fooindex:
\expandafter\xdef\csname#2index\endcsname{\noexpand#1{#3}}%
}
% Define \doindex, the driver for all \fooindex macros.
% Argument #1 is generated by the calling \fooindex macro,
% and it is "foo", the name of the index.
% \doindex just uses \parsearg; it calls \doind for the actual work.
% This is because \doind is more useful to call from other macros.
% There is also \dosubind {index}{topic}{subtopic}
% which makes an entry in a two-level index such as the operation index.
\def\doindex#1{\edef\indexname{#1}\parsearg\singleindexer}
\def\singleindexer #1{\doind{\indexname}{#1}}
% like the previous two, but they put @code around the argument.
\def\docodeindex#1{\edef\indexname{#1}\parsearg\singlecodeindexer}
\def\singlecodeindexer #1{\doind{\indexname}{\code{#1}}}
% Take care of Texinfo commands that can appear in an index entry.
% Since there are some commands we want to expand, and others we don't,
% we have to laboriously prevent expansion for those that we don't.
%
\def\indexdummies{%
\escapechar = `\\ % use backslash in output files.
\def\@{@}% change to @@ when we switch to @ as escape char in index files.
\def\ {\realbackslash\space }%
%
% Need these in case \tex is in effect and \{ is a \delimiter again.
% But can't use \lbracecmd and \rbracecmd because texindex assumes
% braces and backslashes are used only as delimiters.
\let\{ = \mylbrace
\let\} = \myrbrace
%
% I don't entirely understand this, but when an index entry is
% generated from a macro call, the \endinput which \scanmacro inserts
% causes processing to be prematurely terminated. This is,
% apparently, because \indexsorttmp is fully expanded, and \endinput
% is an expandable command. The redefinition below makes \endinput
% disappear altogether for that purpose -- although logging shows that
% processing continues to some further point. On the other hand, it
% seems \endinput does not hurt in the printed index arg, since that
% is still getting written without apparent harm.
%
% Sample source (mac-idx3.tex, reported by Graham Percival to
% help-texinfo, 22may06):
% @macro funindex {WORD}
% @findex xyz
% @end macro
% ...
% @funindex commtest
%
% The above is not enough to reproduce the bug, but it gives the flavor.
%
% Sample whatsit resulting:
% .@write3{\entry{xyz}{@folio }{@code {xyz@endinput }}}
%
% So:
\let\endinput = \empty
%
% Do the redefinitions.
\commondummies
}
% For the aux and toc files, @ is the escape character. So we want to
% redefine everything using @ as the escape character (instead of
% \realbackslash, still used for index files). When everything uses @,
% this will be simpler.
%
\def\atdummies{%
\def\@{@@}%
\def\ {@ }%
\let\{ = \lbraceatcmd
\let\} = \rbraceatcmd
%
% Do the redefinitions.
\commondummies
\otherbackslash
}
% Called from \indexdummies and \atdummies.
%
\def\commondummies{%
%
% \definedummyword defines \#1 as \string\#1\space, thus effectively
% preventing its expansion. This is used only for control% words,
% not control letters, because the \space would be incorrect for
% control characters, but is needed to separate the control word
% from whatever follows.
%
% For control letters, we have \definedummyletter, which omits the
% space.
%
% These can be used both for control words that take an argument and
% those that do not. If it is followed by {arg} in the input, then
% that will dutifully get written to the index (or wherever).
%
\def\definedummyword ##1{\def##1{\string##1\space}}%
\def\definedummyletter##1{\def##1{\string##1}}%
\let\definedummyaccent\definedummyletter
%
\commondummiesnofonts
%
\definedummyletter\_%
%
% Non-English letters.
\definedummyword\AA
\definedummyword\AE
\definedummyword\L
\definedummyword\OE
\definedummyword\O
\definedummyword\aa
\definedummyword\ae
\definedummyword\l
\definedummyword\oe
\definedummyword\o
\definedummyword\ss
\definedummyword\exclamdown
\definedummyword\questiondown
\definedummyword\ordf
\definedummyword\ordm
%
% Although these internal commands shouldn't show up, sometimes they do.
\definedummyword\bf
\definedummyword\gtr
\definedummyword\hat
\definedummyword\less
\definedummyword\sf
\definedummyword\sl
\definedummyword\tclose
\definedummyword\tt
%
\definedummyword\LaTeX
\definedummyword\TeX
%
% Assorted special characters.
\definedummyword\bullet
\definedummyword\comma
\definedummyword\copyright
\definedummyword\registeredsymbol
\definedummyword\dots
\definedummyword\enddots
\definedummyword\equiv
\definedummyword\error
\definedummyword\euro
\definedummyword\expansion
\definedummyword\minus
\definedummyword\pounds
\definedummyword\point
\definedummyword\print
\definedummyword\result
\definedummyword\textdegree
%
% We want to disable all macros so that they are not expanded by \write.
\macrolist
%
\normalturnoffactive
%
% Handle some cases of @value -- where it does not contain any
% (non-fully-expandable) commands.
\makevalueexpandable
}
% \commondummiesnofonts: common to \commondummies and \indexnofonts.
%
\def\commondummiesnofonts{%
% Control letters and accents.
\definedummyletter\!%
\definedummyaccent\"%
\definedummyaccent\'%
\definedummyletter\*%
\definedummyaccent\,%
\definedummyletter\.%
\definedummyletter\/%
\definedummyletter\:%
\definedummyaccent\=%
\definedummyletter\?%
\definedummyaccent\^%
\definedummyaccent\`%
\definedummyaccent\~%
\definedummyword\u
\definedummyword\v
\definedummyword\H
\definedummyword\dotaccent
\definedummyword\ringaccent
\definedummyword\tieaccent
\definedummyword\ubaraccent
\definedummyword\udotaccent
\definedummyword\dotless
%
% Texinfo font commands.
\definedummyword\b
\definedummyword\i
\definedummyword\r
\definedummyword\sc
\definedummyword\t
%
% Commands that take arguments.
\definedummyword\acronym
\definedummyword\cite
\definedummyword\code
\definedummyword\command
\definedummyword\dfn
\definedummyword\emph
\definedummyword\env
\definedummyword\file
\definedummyword\kbd
\definedummyword\key
\definedummyword\math
\definedummyword\option
\definedummyword\pxref
\definedummyword\ref
\definedummyword\samp
\definedummyword\strong
\definedummyword\tie
\definedummyword\uref
\definedummyword\url
\definedummyword\var
\definedummyword\verb
\definedummyword\w
\definedummyword\xref
}
% \indexnofonts is used when outputting the strings to sort the index
% by, and when constructing control sequence names. It eliminates all
% control sequences and just writes whatever the best ASCII sort string
% would be for a given command (usually its argument).
%
\def\indexnofonts{%
% Accent commands should become @asis.
\def\definedummyaccent##1{\let##1\asis}%
% We can just ignore other control letters.
\def\definedummyletter##1{\let##1\empty}%
% Hopefully, all control words can become @asis.
\let\definedummyword\definedummyaccent
%
\commondummiesnofonts
%
% Don't no-op \tt, since it isn't a user-level command
% and is used in the definitions of the active chars like <, >, |, etc.
% Likewise with the other plain tex font commands.
%\let\tt=\asis
%
\def\ { }%
\def\@{@}%
% how to handle braces?
\def\_{\normalunderscore}%
%
% Non-English letters.
\def\AA{AA}%
\def\AE{AE}%
\def\L{L}%
\def\OE{OE}%
\def\O{O}%
\def\aa{aa}%
\def\ae{ae}%
\def\l{l}%
\def\oe{oe}%
\def\o{o}%
\def\ss{ss}%
\def\exclamdown{!}%
\def\questiondown{?}%
\def\ordf{a}%
\def\ordm{o}%
%
\def\LaTeX{LaTeX}%
\def\TeX{TeX}%
%
% Assorted special characters.
% (The following {} will end up in the sort string, but that's ok.)
\def\bullet{bullet}%
\def\comma{,}%
\def\copyright{copyright}%
\def\registeredsymbol{R}%
\def\dots{...}%
\def\enddots{...}%
\def\equiv{==}%
\def\error{error}%
\def\euro{euro}%
\def\expansion{==>}%
\def\minus{-}%
\def\pounds{pounds}%
\def\point{.}%
\def\print{-|}%
\def\result{=>}%
\def\textdegree{degrees}%
%
% We need to get rid of all macros, leaving only the arguments (if present).
% Of course this is not nearly correct, but it is the best we can do for now.
% makeinfo does not expand macros in the argument to @deffn, which ends up
% writing an index entry, and texindex isn't prepared for an index sort entry
% that starts with \.
%
% Since macro invocations are followed by braces, we can just redefine them
% to take a single TeX argument. The case of a macro invocation that
% goes to end-of-line is not handled.
%
\macrolist
}
\let\indexbackslash=0 %overridden during \printindex.
\let\SETmarginindex=\relax % put index entries in margin (undocumented)?
% Most index entries go through here, but \dosubind is the general case.
% #1 is the index name, #2 is the entry text.
\def\doind#1#2{\dosubind{#1}{#2}{}}
% Workhorse for all \fooindexes.
% #1 is name of index, #2 is stuff to put there, #3 is subentry --
% empty if called from \doind, as we usually are (the main exception
% is with most defuns, which call us directly).
%
\def\dosubind#1#2#3{%
\iflinks
{%
% Store the main index entry text (including the third arg).
\toks0 = {#2}%
% If third arg is present, precede it with a space.
\def\thirdarg{#3}%
\ifx\thirdarg\empty \else
\toks0 = \expandafter{\the\toks0 \space #3}%
\fi
%
\edef\writeto{\csname#1indfile\endcsname}%
%
\safewhatsit\dosubindwrite
}%
\fi
}
% Write the entry in \toks0 to the index file:
%
\def\dosubindwrite{%
% Put the index entry in the margin if desired.
\ifx\SETmarginindex\relax\else
\insert\margin{\hbox{\vrule height8pt depth3pt width0pt \the\toks0}}%
\fi
%
% Remember, we are within a group.
\indexdummies % Must do this here, since \bf, etc expand at this stage
\def\backslashcurfont{\indexbackslash}% \indexbackslash isn't defined now
% so it will be output as is; and it will print as backslash.
%
% Process the index entry with all font commands turned off, to
% get the string to sort by.
{\indexnofonts
\edef\temp{\the\toks0}% need full expansion
\xdef\indexsorttmp{\temp}%
}%
%
% Set up the complete index entry, with both the sort key and
% the original text, including any font commands. We write
% three arguments to \entry to the .?? file (four in the
% subentry case), texindex reduces to two when writing the .??s
% sorted result.
\edef\temp{%
\write\writeto{%
\string\entry{\indexsorttmp}{\noexpand\folio}{\the\toks0}}%
}%
\temp
}
% Take care of unwanted page breaks/skips around a whatsit:
%
% If a skip is the last thing on the list now, preserve it
% by backing up by \lastskip, doing the \write, then inserting
% the skip again. Otherwise, the whatsit generated by the
% \write or \pdfdest will make \lastskip zero. The result is that
% sequences like this:
% @end defun
% @tindex whatever
% @defun ...
% will have extra space inserted, because the \medbreak in the
% start of the @defun won't see the skip inserted by the @end of
% the previous defun.
%
% But don't do any of this if we're not in vertical mode. We
% don't want to do a \vskip and prematurely end a paragraph.
%
% Avoid page breaks due to these extra skips, too.
%
% But wait, there is a catch there:
% We'll have to check whether \lastskip is zero skip. \ifdim is not
% sufficient for this purpose, as it ignores stretch and shrink parts
% of the skip. The only way seems to be to check the textual
% representation of the skip.
%
% The following is almost like \def\zeroskipmacro{0.0pt} except that
% the ``p'' and ``t'' characters have catcode \other, not 11 (letter).
%
\edef\zeroskipmacro{\expandafter\the\csname z@skip\endcsname}
%
\newskip\whatsitskip
\newcount\whatsitpenalty
%
% ..., ready, GO:
%
\def\safewhatsit#1{%
\ifhmode
#1%
\else
% \lastskip and \lastpenalty cannot both be nonzero simultaneously.
\whatsitskip = \lastskip
\edef\lastskipmacro{\the\lastskip}%
\whatsitpenalty = \lastpenalty
%
% If \lastskip is nonzero, that means the last item was a
% skip. And since a skip is discardable, that means this
% -\skip0 glue we're inserting is preceded by a
% non-discardable item, therefore it is not a potential
% breakpoint, therefore no \nobreak needed.
\ifx\lastskipmacro\zeroskipmacro
\else
\vskip-\whatsitskip
\fi
%
#1%
%
\ifx\lastskipmacro\zeroskipmacro
% If \lastskip was zero, perhaps the last item was a penalty, and
% perhaps it was >=10000, e.g., a \nobreak. In that case, we want
% to re-insert the same penalty (values >10000 are used for various
% signals); since we just inserted a non-discardable item, any
% following glue (such as a \parskip) would be a breakpoint. For example:
%
% @deffn deffn-whatever
% @vindex index-whatever
% Description.
% would allow a break between the index-whatever whatsit
% and the "Description." paragraph.
\ifnum\whatsitpenalty>9999 \penalty\whatsitpenalty \fi
\else
% On the other hand, if we had a nonzero \lastskip,
% this make-up glue would be preceded by a non-discardable item
% (the whatsit from the \write), so we must insert a \nobreak.
\nobreak\vskip\whatsitskip
\fi
\fi
}
% The index entry written in the file actually looks like
% \entry {sortstring}{page}{topic}
% or
% \entry {sortstring}{page}{topic}{subtopic}
% The texindex program reads in these files and writes files
% containing these kinds of lines:
% \initial {c}
% before the first topic whose initial is c
% \entry {topic}{pagelist}
% for a topic that is used without subtopics
% \primary {topic}
% for the beginning of a topic that is used with subtopics
% \secondary {subtopic}{pagelist}
% for each subtopic.
% Define the user-accessible indexing commands
% @findex, @vindex, @kindex, @cindex.
\def\findex {\fnindex}
\def\kindex {\kyindex}
\def\cindex {\cpindex}
\def\vindex {\vrindex}
\def\tindex {\tpindex}
\def\pindex {\pgindex}
\def\cindexsub {\begingroup\obeylines\cindexsub}
{\obeylines %
\gdef\cindexsub "#1" #2^^M{\endgroup %
\dosubind{cp}{#2}{#1}}}
% Define the macros used in formatting output of the sorted index material.
% @printindex causes a particular index (the ??s file) to get printed.
% It does not print any chapter heading (usually an @unnumbered).
%
\parseargdef\printindex{\begingroup
\dobreak \chapheadingskip{10000}%
%
\smallfonts \rm
\tolerance = 9500
\plainfrenchspacing
\everypar = {}% don't want the \kern\-parindent from indentation suppression.
%
% See if the index file exists and is nonempty.
% Change catcode of @ here so that if the index file contains
% \initial {@}
% as its first line, TeX doesn't complain about mismatched braces
% (because it thinks @} is a control sequence).
\catcode`\@ = 11
\openin 1 \jobname.#1s
\ifeof 1
% \enddoublecolumns gets confused if there is no text in the index,
% and it loses the chapter title and the aux file entries for the
% index. The easiest way to prevent this problem is to make sure
% there is some text.
\putwordIndexNonexistent
\else
%
% If the index file exists but is empty, then \openin leaves \ifeof
% false. We have to make TeX try to read something from the file, so
% it can discover if there is anything in it.
\read 1 to \temp
\ifeof 1
\putwordIndexIsEmpty
\else
% Index files are almost Texinfo source, but we use \ as the escape
% character. It would be better to use @, but that's too big a change
% to make right now.
\def\indexbackslash{\backslashcurfont}%
\catcode`\\ = 0
\escapechar = `\\
\begindoublecolumns
\input \jobname.#1s
\enddoublecolumns
\fi
\fi
\closein 1
\endgroup}
% These macros are used by the sorted index file itself.
% Change them to control the appearance of the index.
\def\initial#1{{%
% Some minor font changes for the special characters.
\let\tentt=\sectt \let\tt=\sectt \let\sf=\sectt
%
% Remove any glue we may have, we'll be inserting our own.
\removelastskip
%
% We like breaks before the index initials, so insert a bonus.
\nobreak
\vskip 0pt plus 3\baselineskip
\penalty 0
\vskip 0pt plus -3\baselineskip
%
% Typeset the initial. Making this add up to a whole number of
% baselineskips increases the chance of the dots lining up from column
% to column. It still won't often be perfect, because of the stretch
% we need before each entry, but it's better.
%
% No shrink because it confuses \balancecolumns.
\vskip 1.67\baselineskip plus .5\baselineskip
\leftline{\secbf #1}%
% Do our best not to break after the initial.
\nobreak
\vskip .33\baselineskip plus .1\baselineskip
}}
% \entry typesets a paragraph consisting of the text (#1), dot leaders, and
% then page number (#2) flushed to the right margin. It is used for index
% and table of contents entries. The paragraph is indented by \leftskip.
%
% A straightforward implementation would start like this:
% \def\entry#1#2{...
% But this frozes the catcodes in the argument, and can cause problems to
% @code, which sets - active. This problem was fixed by a kludge---
% ``-'' was active throughout whole index, but this isn't really right.
%
% The right solution is to prevent \entry from swallowing the whole text.
% --kasal, 21nov03
\def\entry{%
\begingroup
%
% Start a new paragraph if necessary, so our assignments below can't
% affect previous text.
\par
%
% Do not fill out the last line with white space.
\parfillskip = 0in
%
% No extra space above this paragraph.
\parskip = 0in
%
% Do not prefer a separate line ending with a hyphen to fewer lines.
\finalhyphendemerits = 0
%
% \hangindent is only relevant when the entry text and page number
% don't both fit on one line. In that case, bob suggests starting the
% dots pretty far over on the line. Unfortunately, a large
% indentation looks wrong when the entry text itself is broken across
% lines. So we use a small indentation and put up with long leaders.
%
% \hangafter is reset to 1 (which is the value we want) at the start
% of each paragraph, so we need not do anything with that.
\hangindent = 2em
%
% When the entry text needs to be broken, just fill out the first line
% with blank space.
\rightskip = 0pt plus1fil
%
% A bit of stretch before each entry for the benefit of balancing
% columns.
\vskip 0pt plus1pt
%
% Swallow the left brace of the text (first parameter):
\afterassignment\doentry
\let\temp =
}
\def\doentry{%
\bgroup % Instead of the swallowed brace.
\noindent
\aftergroup\finishentry
% And now comes the text of the entry.
}
\def\finishentry#1{%
% #1 is the page number.
%
% The following is kludged to not output a line of dots in the index if
% there are no page numbers. The next person who breaks this will be
% cursed by a Unix daemon.
\def\tempa{{\rm }}%
\def\tempb{#1}%
\edef\tempc{\tempa}%
\edef\tempd{\tempb}%
\ifx\tempc\tempd
\ %
\else
%
% If we must, put the page number on a line of its own, and fill out
% this line with blank space. (The \hfil is overwhelmed with the
% fill leaders glue in \indexdotfill if the page number does fit.)
\hfil\penalty50
\null\nobreak\indexdotfill % Have leaders before the page number.
%
% The `\ ' here is removed by the implicit \unskip that TeX does as
% part of (the primitive) \par. Without it, a spurious underfull
% \hbox ensues.
\ifpdf
\pdfgettoks#1.%
\ \the\toksA
\else
\ #1%
\fi
\fi
\par
\endgroup
}
% Like plain.tex's \dotfill, except uses up at least 1 em.
\def\indexdotfill{\cleaders
\hbox{$\mathsurround=0pt \mkern1.5mu.\mkern1.5mu$}\hskip 1em plus 1fill}
\def\primary #1{\line{#1\hfil}}
\newskip\secondaryindent \secondaryindent=0.5cm
\def\secondary#1#2{{%
\parfillskip=0in
\parskip=0in
\hangindent=1in
\hangafter=1
\noindent\hskip\secondaryindent\hbox{#1}\indexdotfill
\ifpdf
\pdfgettoks#2.\ \the\toksA % The page number ends the paragraph.
\else
#2
\fi
\par
}}
% Define two-column mode, which we use to typeset indexes.
% Adapted from the TeXbook, page 416, which is to say,
% the manmac.tex format used to print the TeXbook itself.
\catcode`\@=11
\newbox\partialpage
\newdimen\doublecolumnhsize
\def\begindoublecolumns{\begingroup % ended by \enddoublecolumns
% Grab any single-column material above us.
\output = {%
%
% Here is a possibility not foreseen in manmac: if we accumulate a
% whole lot of material, we might end up calling this \output
% routine twice in a row (see the doublecol-lose test, which is
% essentially a couple of indexes with @setchapternewpage off). In
% that case we just ship out what is in \partialpage with the normal
% output routine. Generally, \partialpage will be empty when this
% runs and this will be a no-op. See the indexspread.tex test case.
\ifvoid\partialpage \else
\onepageout{\pagecontents\partialpage}%
\fi
%
\global\setbox\partialpage = \vbox{%
% Unvbox the main output page.
\unvbox\PAGE
\kern-\topskip \kern\baselineskip
}%
}%
\eject % run that output routine to set \partialpage
%
% Use the double-column output routine for subsequent pages.
\output = {\doublecolumnout}%
%
% Change the page size parameters. We could do this once outside this
% routine, in each of @smallbook, @afourpaper, and the default 8.5x11
% format, but then we repeat the same computation. Repeating a couple
% of assignments once per index is clearly meaningless for the
% execution time, so we may as well do it in one place.
%
% First we halve the line length, less a little for the gutter between
% the columns. We compute the gutter based on the line length, so it
% changes automatically with the paper format. The magic constant
% below is chosen so that the gutter has the same value (well, +-<1pt)
% as it did when we hard-coded it.
%
% We put the result in a separate register, \doublecolumhsize, so we
% can restore it in \pagesofar, after \hsize itself has (potentially)
% been clobbered.
%
\doublecolumnhsize = \hsize
\advance\doublecolumnhsize by -.04154\hsize
\divide\doublecolumnhsize by 2
\hsize = \doublecolumnhsize
%
% Double the \vsize as well. (We don't need a separate register here,
% since nobody clobbers \vsize.)
\vsize = 2\vsize
}
% The double-column output routine for all double-column pages except
% the last.
%
\def\doublecolumnout{%
\splittopskip=\topskip \splitmaxdepth=\maxdepth
% Get the available space for the double columns -- the normal
% (undoubled) page height minus any material left over from the
% previous page.
\dimen@ = \vsize
\divide\dimen@ by 2
\advance\dimen@ by -\ht\partialpage
%
% box0 will be the left-hand column, box2 the right.
\setbox0=\vsplit255 to\dimen@ \setbox2=\vsplit255 to\dimen@
\onepageout\pagesofar
\unvbox255
\penalty\outputpenalty
}
%
% Re-output the contents of the output page -- any previous material,
% followed by the two boxes we just split, in box0 and box2.
\def\pagesofar{%
\unvbox\partialpage
%
\hsize = \doublecolumnhsize
\wd0=\hsize \wd2=\hsize
\hbox to\pagewidth{\box0\hfil\box2}%
}
%
% All done with double columns.
\def\enddoublecolumns{%
% The following penalty ensures that the page builder is exercised
% _before_ we change the output routine. This is necessary in the
% following situation:
%
% The last section of the index consists only of a single entry.
% Before this section, \pagetotal is less than \pagegoal, so no
% break occurs before the last section starts. However, the last
% section, consisting of \initial and the single \entry, does not
% fit on the page and has to be broken off. Without the following
% penalty the page builder will not be exercised until \eject
% below, and by that time we'll already have changed the output
% routine to the \balancecolumns version, so the next-to-last
% double-column page will be processed with \balancecolumns, which
% is wrong: The two columns will go to the main vertical list, with
% the broken-off section in the recent contributions. As soon as
% the output routine finishes, TeX starts reconsidering the page
% break. The two columns and the broken-off section both fit on the
% page, because the two columns now take up only half of the page
% goal. When TeX sees \eject from below which follows the final
% section, it invokes the new output routine that we've set after
% \balancecolumns below; \onepageout will try to fit the two columns
% and the final section into the vbox of \pageheight (see
% \pagebody), causing an overfull box.
%
% Note that glue won't work here, because glue does not exercise the
% page builder, unlike penalties (see The TeXbook, pp. 280-281).
\penalty0
%
\output = {%
% Split the last of the double-column material. Leave it on the
% current page, no automatic page break.
\balancecolumns
%
% If we end up splitting too much material for the current page,
% though, there will be another page break right after this \output
% invocation ends. Having called \balancecolumns once, we do not
% want to call it again. Therefore, reset \output to its normal
% definition right away. (We hope \balancecolumns will never be
% called on to balance too much material, but if it is, this makes
% the output somewhat more palatable.)
\global\output = {\onepageout{\pagecontents\PAGE}}%
}%
\eject
\endgroup % started in \begindoublecolumns
%
% \pagegoal was set to the doubled \vsize above, since we restarted
% the current page. We're now back to normal single-column
% typesetting, so reset \pagegoal to the normal \vsize (after the
% \endgroup where \vsize got restored).
\pagegoal = \vsize
}
%
% Called at the end of the double column material.
\def\balancecolumns{%
\setbox0 = \vbox{\unvbox255}% like \box255 but more efficient, see p.120.
\dimen@ = \ht0
\advance\dimen@ by \topskip
\advance\dimen@ by-\baselineskip
\divide\dimen@ by 2 % target to split to
%debug\message{final 2-column material height=\the\ht0, target=\the\dimen@.}%
\splittopskip = \topskip
% Loop until we get a decent breakpoint.
{%
\vbadness = 10000
\loop
\global\setbox3 = \copy0
\global\setbox1 = \vsplit3 to \dimen@
\ifdim\ht3>\dimen@
\global\advance\dimen@ by 1pt
\repeat
}%
%debug\message{split to \the\dimen@, column heights: \the\ht1, \the\ht3.}%
\setbox0=\vbox to\dimen@{\unvbox1}%
\setbox2=\vbox to\dimen@{\unvbox3}%
%
\pagesofar
}
\catcode`\@ = \other
\message{sectioning,}
% Chapters, sections, etc.
% \unnumberedno is an oxymoron, of course. But we count the unnumbered
% sections so that we can refer to them unambiguously in the pdf
% outlines by their "section number". We avoid collisions with chapter
% numbers by starting them at 10000. (If a document ever has 10000
% chapters, we're in trouble anyway, I'm sure.)
\newcount\unnumberedno \unnumberedno = 10000
\newcount\chapno
\newcount\secno \secno=0
\newcount\subsecno \subsecno=0
\newcount\subsubsecno \subsubsecno=0
% This counter is funny since it counts through charcodes of letters A, B, ...
\newcount\appendixno \appendixno = `\@
%
% \def\appendixletter{\char\the\appendixno}
% We do the following ugly conditional instead of the above simple
% construct for the sake of pdftex, which needs the actual
% letter in the expansion, not just typeset.
%
\def\appendixletter{%
\ifnum\appendixno=`A A%
\else\ifnum\appendixno=`B B%
\else\ifnum\appendixno=`C C%
\else\ifnum\appendixno=`D D%
\else\ifnum\appendixno=`E E%
\else\ifnum\appendixno=`F F%
\else\ifnum\appendixno=`G G%
\else\ifnum\appendixno=`H H%
\else\ifnum\appendixno=`I I%
\else\ifnum\appendixno=`J J%
\else\ifnum\appendixno=`K K%
\else\ifnum\appendixno=`L L%
\else\ifnum\appendixno=`M M%
\else\ifnum\appendixno=`N N%
\else\ifnum\appendixno=`O O%
\else\ifnum\appendixno=`P P%
\else\ifnum\appendixno=`Q Q%
\else\ifnum\appendixno=`R R%
\else\ifnum\appendixno=`S S%
\else\ifnum\appendixno=`T T%
\else\ifnum\appendixno=`U U%
\else\ifnum\appendixno=`V V%
\else\ifnum\appendixno=`W W%
\else\ifnum\appendixno=`X X%
\else\ifnum\appendixno=`Y Y%
\else\ifnum\appendixno=`Z Z%
% The \the is necessary, despite appearances, because \appendixletter is
% expanded while writing the .toc file. \char\appendixno is not
% expandable, thus it is written literally, thus all appendixes come out
% with the same letter (or @) in the toc without it.
\else\char\the\appendixno
\fi\fi\fi\fi\fi\fi\fi\fi\fi\fi\fi\fi\fi
\fi\fi\fi\fi\fi\fi\fi\fi\fi\fi\fi\fi\fi}
% Each @chapter defines this as the name of the chapter.
% page headings and footings can use it. @section does likewise.
% However, they are not reliable, because we don't use marks.
\def\thischapter{}
\def\thissection{}
\newcount\absseclevel % used to calculate proper heading level
\newcount\secbase\secbase=0 % @raisesections/@lowersections modify this count
% @raisesections: treat @section as chapter, @subsection as section, etc.
\def\raisesections{\global\advance\secbase by -1}
\let\up=\raisesections % original BFox name
% @lowersections: treat @chapter as section, @section as subsection, etc.
\def\lowersections{\global\advance\secbase by 1}
\let\down=\lowersections % original BFox name
% we only have subsub.
\chardef\maxseclevel = 3
%
% A numbered section within an unnumbered changes to unnumbered too.
-% To achive this, remember the "biggest" unnum. sec. we are currently in:
+% To achieve 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}%
\gdef\thischapternum{}%
\gdef\thischapter{#1}%
\else\ifx\temptype\Yomitfromtockeyword
\setbox0 = \hbox{}% contents like unnumbered, but no toc entry
\def\toctype{omit}%
\gdef\thischapternum{}%
\gdef\thischapter{}%
\else\ifx\temptype\Yappendixkeyword
\setbox0 = \hbox{\putwordAppendix{} #3\enspace}%
\def\toctype{app}%
\xdef\thischapternum{\appendixletter}%
% We don't substitute the actual chapter name into \thischapter
% because we don't want its macros evaluated now. And we don't
% use \thissection because that changes with each section.
%
\xdef\thischapter{\putwordAppendix{} \appendixletter:
\noexpand\thischaptername}%
\else
\setbox0 = \hbox{#3\enspace}%
\def\toctype{numchap}%
\xdef\thischapternum{\the\chapno}%
\xdef\thischapter{\putwordChapter{} \the\chapno:
\noexpand\thischaptername}%
\fi\fi\fi
%
% Write the toc entry for this chapter. Must come before the
% \donoderef, because we include the current node name in the toc
% entry, and \donoderef resets it to empty.
\writetocentry{\toctype}{#1}{#3}%
%
% For pdftex, we have to write out the node definition (aka, make
% the pdfdest) after any page break, but before the actual text has
% been typeset. If the destination for the pdf outline is after the
% text, then jumping from the outline may wind up with the text not
% being visible, for instance under high magnification.
\donoderef{#2}%
%
% Typeset the actual heading.
\vbox{\hyphenpenalty=10000 \tolerance=5000 \parindent=0pt \raggedright
\hangindent=\wd0 \centerparametersmaybe
\unhbox0 #1\par}%
}%
\nobreak\bigskip % no page break after a chapter title
\nobreak
}
% @centerchap -- centered and unnumbered.
\let\centerparametersmaybe = \relax
\def\centerparameters{%
\advance\rightskip by 3\rightskip
\leftskip = \rightskip
\parfillskip = 0pt
}
% I don't think this chapter style is supported any more, so I'm not
% updating it with the new noderef stuff. We'll see. --karl, 11aug03.
%
\def\setchapterstyle #1 {\csname CHAPF#1\endcsname}
%
\def\unnchfopen #1{%
\chapoddpage {\chapfonts \vbox{\hyphenpenalty=10000\tolerance=5000
\parindent=0pt\raggedright
\rm #1\hfill}}\bigskip \par\nobreak
}
\def\chfopen #1#2{\chapoddpage {\chapfonts
\vbox to 3in{\vfil \hbox to\hsize{\hfil #2} \hbox to\hsize{\hfil #1} \vfil}}%
\par\penalty 5000 %
}
\def\centerchfopen #1{%
\chapoddpage {\chapfonts \vbox{\hyphenpenalty=10000\tolerance=5000
\parindent=0pt
\hfill {\rm #1}\hfill}}\bigskip \par\nobreak
}
\def\CHAPFopen{%
\global\let\chapmacro=\chfopen
\global\let\centerchapmacro=\centerchfopen}
% Section titles. These macros combine the section number parts and
% call the generic \sectionheading to do the printing.
%
\newskip\secheadingskip
\def\secheadingbreak{\dobreak \secheadingskip{-1000}}
% Subsection titles.
\newskip\subsecheadingskip
\def\subsecheadingbreak{\dobreak \subsecheadingskip{-500}}
% Subsubsection titles.
\def\subsubsecheadingskip{\subsecheadingskip}
\def\subsubsecheadingbreak{\subsecheadingbreak}
% Print any size, any type, section title.
%
% #1 is the text, #2 is the section level (sec/subsec/subsubsec), #3 is
% the section type for xrefs (Ynumbered, Ynothing, Yappendix), #4 is the
% section number.
%
\def\sectionheading#1#2#3#4{%
{%
% Switch to the right set of fonts.
\csname #2fonts\endcsname \rm
%
% Insert space above the heading.
\csname #2headingbreak\endcsname
%
% Only insert the space after the number if we have a section number.
\def\sectionlevel{#2}%
\def\temptype{#3}%
%
\ifx\temptype\Ynothingkeyword
\setbox0 = \hbox{}%
\def\toctype{unn}%
\gdef\thissection{#1}%
\else\ifx\temptype\Yomitfromtockeyword
% for @headings -- no section number, don't include in toc,
% and don't redefine \thissection.
\setbox0 = \hbox{}%
\def\toctype{omit}%
\let\sectionlevel=\empty
\else\ifx\temptype\Yappendixkeyword
\setbox0 = \hbox{#4\enspace}%
\def\toctype{app}%
\gdef\thissection{#1}%
\else
\setbox0 = \hbox{#4\enspace}%
\def\toctype{num}%
\gdef\thissection{#1}%
\fi\fi\fi
%
% Write the toc entry (before \donoderef). See comments in \chapmacro.
\writetocentry{\toctype\sectionlevel}{#1}{#4}%
%
% Write the node reference (= pdf destination for pdftex).
% Again, see comments in \chapmacro.
\donoderef{#3}%
%
% Interline glue will be inserted when the vbox is completed.
% That glue will be a valid breakpoint for the page, since it'll be
% preceded by a whatsit (usually from the \donoderef, or from the
% \writetocentry if there was no node). We don't want to allow that
% break, since then the whatsits could end up on page n while the
% section is on page n+1, thus toc/etc. are wrong. Debian bug 276000.
\nobreak
%
% Output the actual section heading.
\vbox{\hyphenpenalty=10000 \tolerance=5000 \parindent=0pt \raggedright
\hangindent=\wd0 % zero if no section number
\unhbox0 #1}%
}%
% Add extra space after the heading -- half of whatever came above it.
% Don't allow stretch, though.
\kern .5 \csname #2headingskip\endcsname
%
% Do not let the kern be a potential breakpoint, as it would be if it
% was followed by glue.
\nobreak
%
% We'll almost certainly start a paragraph next, so don't let that
% glue accumulate. (Not a breakpoint because it's preceded by a
% discardable item.)
\vskip-\parskip
%
% This is purely so the last item on the list is a known \penalty >
% 10000. This is so \startdefun can avoid allowing breakpoints after
% section headings. Otherwise, it would insert a valid breakpoint between:
%
% @section sec-whatever
% @deffn def-whatever
\penalty 10001
}
\message{toc,}
% Table of contents.
\newwrite\tocfile
% Write an entry to the toc file, opening it if necessary.
% Called from @chapter, etc.
%
% Example usage: \writetocentry{sec}{Section Name}{\the\chapno.\the\secno}
% We append the current node name (if any) and page number as additional
% arguments for the \{chap,sec,...}entry macros which will eventually
% read this. The node name is used in the pdf outlines as the
% destination to jump to.
%
% We open the .toc file for writing here instead of at @setfilename (or
% any other fixed time) so that @contents can be anywhere in the document.
% But if #1 is `omit', then we don't do anything. This is used for the
% table of contents chapter openings themselves.
%
\newif\iftocfileopened
\def\omitkeyword{omit}%
%
\def\writetocentry#1#2#3{%
\edef\writetoctype{#1}%
\ifx\writetoctype\omitkeyword \else
\iftocfileopened\else
\immediate\openout\tocfile = \jobname.toc
\global\tocfileopenedtrue
\fi
%
\iflinks
{\atdummies
\edef\temp{%
\write\tocfile{@#1entry{#2}{#3}{\lastnode}{\noexpand\folio}}}%
\temp
}%
\fi
\fi
%
% Tell \shipout to create a pdf destination on each page, if we're
% writing pdf. These are used in the table of contents. We can't
% just write one on every page because the title pages are numbered
% 1 and 2 (the page numbers aren't printed), and so are the first
% two pages of the document. Thus, we'd have two destinations named
% `1', and two named `2'.
\ifpdf \global\pdfmakepagedesttrue \fi
}
% These characters do not print properly in the Computer Modern roman
% fonts, so we must take special care. This is more or less redundant
% with the Texinfo input format setup at the end of this file.
%
\def\activecatcodes{%
\catcode`\"=\active
\catcode`\$=\active
\catcode`\<=\active
\catcode`\>=\active
\catcode`\\=\active
\catcode`\^=\active
\catcode`\_=\active
\catcode`\|=\active
\catcode`\~=\active
}
% Read the toc file, which is essentially Texinfo input.
\def\readtocfile{%
\setupdatafile
\activecatcodes
\input \jobname.toc
}
\newskip\contentsrightmargin \contentsrightmargin=1in
\newcount\savepageno
\newcount\lastnegativepageno \lastnegativepageno = -1
% Prepare to read what we've written to \tocfile.
%
\def\startcontents#1{%
% If @setchapternewpage on, and @headings double, the contents should
% start on an odd page, unlike chapters. Thus, we maintain
% \contentsalignmacro in parallel with \pagealignmacro.
% From: Torbjorn Granlund <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.
\raggedbottom % Worry more about breakpoints than the bottom.
\advance\hsize by -\contentsrightmargin % Don't use the full line length.
%
% Roman numerals for page numbers.
\ifnum \pageno>0 \global\pageno = \lastnegativepageno \fi
}
% Normal (long) toc.
\def\contents{%
\startcontents{\putwordTOC}%
\openin 1 \jobname.toc
\ifeof 1 \else
\readtocfile
\fi
\vfill \eject
\contentsalignmacro % in case @setchapternewpage odd is in effect
\ifeof 1 \else
\pdfmakeoutlines
\fi
\closein 1
\endgroup
\lastnegativepageno = \pageno
\global\pageno = \savepageno
}
% And just the chapters.
\def\summarycontents{%
\startcontents{\putwordShortTOC}%
%
\let\numchapentry = \shortchapentry
\let\appentry = \shortchapentry
\let\unnchapentry = \shortunnchapentry
% We want a true roman here for the page numbers.
\secfonts
\let\rm=\shortcontrm \let\bf=\shortcontbf
\let\sl=\shortcontsl \let\tt=\shortconttt
\rm
\hyphenpenalty = 10000
\advance\baselineskip by 1pt % Open it up a little.
\def\numsecentry##1##2##3##4{}
\let\appsecentry = \numsecentry
\let\unnsecentry = \numsecentry
\let\numsubsecentry = \numsecentry
\let\appsubsecentry = \numsecentry
\let\unnsubsecentry = \numsecentry
\let\numsubsubsecentry = \numsecentry
\let\appsubsubsecentry = \numsecentry
\let\unnsubsubsecentry = \numsecentry
\openin 1 \jobname.toc
\ifeof 1 \else
\readtocfile
\fi
\closein 1
\vfill \eject
\contentsalignmacro % in case @setchapternewpage odd is in effect
\endgroup
\lastnegativepageno = \pageno
\global\pageno = \savepageno
}
\let\shortcontents = \summarycontents
% Typeset the label for a chapter or appendix for the short contents.
% The arg is, e.g., `A' for an appendix, or `3' for a chapter.
%
\def\shortchaplabel#1{%
% This space should be enough, since a single number is .5em, and the
% widest letter (M) is 1em, at least in the Computer Modern fonts.
% But use \hss just in case.
% (This space doesn't include the extra space that gets added after
% the label; that gets put in by \shortchapentry above.)
%
% We'd like to right-justify chapter numbers, but that looks strange
% with appendix letters. And right-justifying numbers and
% left-justifying letters looks strange when there is less than 10
% chapters. Have to read the whole toc once to know how many chapters
% there are before deciding ...
\hbox to 1em{#1\hss}%
}
% These macros generate individual entries in the table of contents.
% The first argument is the chapter or section name.
% The last argument is the page number.
% The arguments in between are the chapter number, section number, ...
% Chapters, in the main contents.
\def\numchapentry#1#2#3#4{\dochapentry{#2\labelspace#1}{#4}}
%
% Chapters, in the short toc.
% See comments in \dochapentry re vbox and related settings.
\def\shortchapentry#1#2#3#4{%
\tocentry{\shortchaplabel{#2}\labelspace #1}{\doshortpageno\bgroup#4\egroup}%
}
% Appendices, in the main contents.
% Need the word Appendix, and a fixed-size box.
%
\def\appendixbox#1{%
% We use M since it's probably the widest letter.
\setbox0 = \hbox{\putwordAppendix{} M}%
\hbox to \wd0{\putwordAppendix{} #1\hss}}
%
\def\appentry#1#2#3#4{\dochapentry{\appendixbox{#2}\labelspace#1}{#4}}
% Unnumbered chapters.
\def\unnchapentry#1#2#3#4{\dochapentry{#1}{#4}}
\def\shortunnchapentry#1#2#3#4{\tocentry{#1}{\doshortpageno\bgroup#4\egroup}}
% Sections.
\def\numsecentry#1#2#3#4{\dosecentry{#2\labelspace#1}{#4}}
\let\appsecentry=\numsecentry
\def\unnsecentry#1#2#3#4{\dosecentry{#1}{#4}}
% Subsections.
\def\numsubsecentry#1#2#3#4{\dosubsecentry{#2\labelspace#1}{#4}}
\let\appsubsecentry=\numsubsecentry
\def\unnsubsecentry#1#2#3#4{\dosubsecentry{#1}{#4}}
% And subsubsections.
\def\numsubsubsecentry#1#2#3#4{\dosubsubsecentry{#2\labelspace#1}{#4}}
\let\appsubsubsecentry=\numsubsubsecentry
\def\unnsubsubsecentry#1#2#3#4{\dosubsubsecentry{#1}{#4}}
% This parameter controls the indentation of the various levels.
% Same as \defaultparindent.
\newdimen\tocindent \tocindent = 15pt
% Now for the actual typesetting. In all these, #1 is the text and #2 is the
% page number.
%
% If the toc has to be broken over pages, we want it to be at chapters
% if at all possible; hence the \penalty.
\def\dochapentry#1#2{%
\penalty-300 \vskip1\baselineskip plus.33\baselineskip minus.25\baselineskip
\begingroup
\chapentryfonts
\tocentry{#1}{\dopageno\bgroup#2\egroup}%
\endgroup
\nobreak\vskip .25\baselineskip plus.1\baselineskip
}
\def\dosecentry#1#2{\begingroup
\secentryfonts \leftskip=\tocindent
\tocentry{#1}{\dopageno\bgroup#2\egroup}%
\endgroup}
\def\dosubsecentry#1#2{\begingroup
\subsecentryfonts \leftskip=2\tocindent
\tocentry{#1}{\dopageno\bgroup#2\egroup}%
\endgroup}
\def\dosubsubsecentry#1#2{\begingroup
\subsubsecentryfonts \leftskip=3\tocindent
\tocentry{#1}{\dopageno\bgroup#2\egroup}%
\endgroup}
% We use the same \entry macro as for the index entries.
\let\tocentry = \entry
% Space between chapter (or whatever) number and the title.
\def\labelspace{\hskip1em \relax}
\def\dopageno#1{{\rm #1}}
\def\doshortpageno#1{{\rm #1}}
\def\chapentryfonts{\secfonts \rm}
\def\secentryfonts{\textfonts}
\def\subsecentryfonts{\textfonts}
\def\subsubsecentryfonts{\textfonts}
\message{environments,}
% @foo ... @end foo.
% @point{}, @result{}, @expansion{}, @print{}, @equiv{}.
%
% Since these characters are used in examples, it should be an even number of
% \tt widths. Each \tt character is 1en, so two makes it 1em.
%
\def\point{$\star$}
\def\result{\leavevmode\raise.15ex\hbox to 1em{\hfil$\Rightarrow$\hfil}}
\def\expansion{\leavevmode\raise.1ex\hbox to 1em{\hfil$\mapsto$\hfil}}
\def\print{\leavevmode\lower.1ex\hbox to 1em{\hfil$\dashv$\hfil}}
\def\equiv{\leavevmode\lower.1ex\hbox to 1em{\hfil$\ptexequiv$\hfil}}
% The @error{} command.
% Adapted from the TeXbook's \boxit.
%
\newbox\errorbox
%
{\tentt \global\dimen0 = 3em}% Width of the box.
\dimen2 = .55pt % Thickness of rules
% The text. (`r' is open on the right, `e' somewhat less so on the left.)
\setbox0 = \hbox{\kern-.75pt \reducedsf error\kern-1.5pt}
%
\setbox\errorbox=\hbox to \dimen0{\hfil
\hsize = \dimen0 \advance\hsize by -5.8pt % Space to left+right.
\advance\hsize by -2\dimen2 % Rules.
\vbox{%
\hrule height\dimen2
\hbox{\vrule width\dimen2 \kern3pt % Space to left of text.
\vtop{\kern2.4pt \box0 \kern2.4pt}% Space above/below.
\kern3pt\vrule width\dimen2}% Space to right.
\hrule height\dimen2}
\hfil}
%
\def\error{\leavevmode\lower.7ex\copy\errorbox}
% @tex ... @end tex escapes into raw Tex temporarily.
% One exception: @ is still an escape character, so that @end tex works.
% But \@ or @@ will get a plain tex @ character.
\envdef\tex{%
\catcode `\\=0 \catcode `\{=1 \catcode `\}=2
\catcode `\$=3 \catcode `\&=4 \catcode `\#=6
\catcode `\^=7 \catcode `\_=8 \catcode `\~=\active \let~=\tie
\catcode `\%=14
\catcode `\+=\other
\catcode `\"=\other
\catcode `\|=\other
\catcode `\<=\other
\catcode `\>=\other
\escapechar=`\\
%
\let\b=\ptexb
\let\bullet=\ptexbullet
\let\c=\ptexc
\let\,=\ptexcomma
\let\.=\ptexdot
\let\dots=\ptexdots
\let\equiv=\ptexequiv
\let\!=\ptexexclam
\let\i=\ptexi
\let\indent=\ptexindent
\let\noindent=\ptexnoindent
\let\{=\ptexlbrace
\let\+=\tabalign
\let\}=\ptexrbrace
\let\/=\ptexslash
\let\*=\ptexstar
\let\t=\ptext
\let\frenchspacing=\plainfrenchspacing
%
\def\endldots{\mathinner{\ldots\ldots\ldots\ldots}}%
\def\enddots{\relax\ifmmode\endldots\else$\mathsurround=0pt \endldots\,$\fi}%
\def\@{@}%
}
% There is no need to define \Etex.
% Define @lisp ... @end lisp.
% @lisp environment forms a group so it can rebind things,
% including the definition of @end lisp (which normally is erroneous).
% Amount to narrow the margins by for @lisp.
\newskip\lispnarrowing \lispnarrowing=0.4in
% This is the definition that ^^M gets inside @lisp, @example, and other
% such environments. \null is better than a space, since it doesn't
% have any width.
\def\lisppar{\null\endgraf}
% This space is always present above and below environments.
\newskip\envskipamount \envskipamount = 0pt
% Make spacing and below environment symmetrical. We use \parskip here
% to help in doing that, since in @example-like environments \parskip
% is reset to zero; thus the \afterenvbreak inserts no space -- but the
% start of the next paragraph will insert \parskip.
%
\def\aboveenvbreak{{%
% =10000 instead of <10000 because of a special case in \itemzzz and
% \sectionheading, q.v.
\ifnum \lastpenalty=10000 \else
\advance\envskipamount by \parskip
\endgraf
\ifdim\lastskip<\envskipamount
\removelastskip
% it's not a good place to break if the last penalty was \nobreak
% or better ...
\ifnum\lastpenalty<10000 \penalty-50 \fi
\vskip\envskipamount
\fi
\fi
}}
\let\afterenvbreak = \aboveenvbreak
% \nonarrowing is a flag. If "set", @lisp etc don't narrow margins; it will
% also clear it, so that its embedded environments do the narrowing again.
\let\nonarrowing=\relax
% @cartouche ... @end cartouche: draw rectangle w/rounded corners around
% environment contents.
\font\circle=lcircle10
\newdimen\circthick
\newdimen\cartouter\newdimen\cartinner
\newskip\normbskip\newskip\normpskip\newskip\normlskip
\circthick=\fontdimen8\circle
%
\def\ctl{{\circle\char'013\hskip -6pt}}% 6pt from pl file: 1/2charwidth
\def\ctr{{\hskip 6pt\circle\char'010}}
\def\cbl{{\circle\char'012\hskip -6pt}}
\def\cbr{{\hskip 6pt\circle\char'011}}
\def\carttop{\hbox to \cartouter{\hskip\lskip
\ctl\leaders\hrule height\circthick\hfil\ctr
\hskip\rskip}}
\def\cartbot{\hbox to \cartouter{\hskip\lskip
\cbl\leaders\hrule height\circthick\hfil\cbr
\hskip\rskip}}
%
\newskip\lskip\newskip\rskip
\envdef\cartouche{%
\ifhmode\par\fi % can't be in the midst of a paragraph.
\startsavinginserts
\lskip=\leftskip \rskip=\rightskip
\leftskip=0pt\rightskip=0pt % we want these *outside*.
\cartinner=\hsize \advance\cartinner by-\lskip
\advance\cartinner by-\rskip
\cartouter=\hsize
\advance\cartouter by 18.4pt % allow for 3pt kerns on either
% side, and for 6pt waste from
% each corner char, and rule thickness
\normbskip=\baselineskip \normpskip=\parskip \normlskip=\lineskip
% Flag to tell @lisp, etc., not to narrow margin.
\let\nonarrowing = t%
\vbox\bgroup
\baselineskip=0pt\parskip=0pt\lineskip=0pt
\carttop
\hbox\bgroup
\hskip\lskip
\vrule\kern3pt
\vbox\bgroup
\kern3pt
\hsize=\cartinner
\baselineskip=\normbskip
\lineskip=\normlskip
\parskip=\normpskip
\vskip -\parskip
\comment % For explanation, see the end of \def\group.
}
\def\Ecartouche{%
\ifhmode\par\fi
\kern3pt
\egroup
\kern3pt\vrule
\hskip\rskip
\egroup
\cartbot
\egroup
\checkinserts
}
% This macro is called at the beginning of all the @example variants,
% inside a group.
\def\nonfillstart{%
\aboveenvbreak
\hfuzz = 12pt % Don't be fussy
\sepspaces % Make spaces be word-separators rather than space tokens.
\let\par = \lisppar % don't ignore blank lines
\obeylines % each line of input is a line of output
\parskip = 0pt
\parindent = 0pt
\emergencystretch = 0pt % don't try to avoid overfull boxes
\ifx\nonarrowing\relax
\advance \leftskip by \lispnarrowing
\exdentamount=\lispnarrowing
\else
\let\nonarrowing = \relax
\fi
\let\exdent=\nofillexdent
}
% If you want all examples etc. small: @set dispenvsize small.
% If you want even small examples the full size: @set dispenvsize nosmall.
% This affects the following displayed environments:
% @example, @display, @format, @lisp
%
\def\smallword{small}
\def\nosmallword{nosmall}
\let\SETdispenvsize\relax
\def\setnormaldispenv{%
\ifx\SETdispenvsize\smallword
% end paragraph for sake of leading, in case document has no blank
% line. This is redundant with what happens in \aboveenvbreak, but
% we need to do it before changing the fonts, and it's inconvenient
% to change the fonts afterward.
\ifnum \lastpenalty=10000 \else \endgraf \fi
\smallexamplefonts \rm
\fi
}
\def\setsmalldispenv{%
\ifx\SETdispenvsize\nosmallword
\else
\ifnum \lastpenalty=10000 \else \endgraf \fi
\smallexamplefonts \rm
\fi
}
% We often define two environments, @foo and @smallfoo.
% Let's do it by one command:
\def\makedispenv #1#2{
\expandafter\envdef\csname#1\endcsname {\setnormaldispenv #2}
\expandafter\envdef\csname small#1\endcsname {\setsmalldispenv #2}
\expandafter\let\csname E#1\endcsname \afterenvbreak
\expandafter\let\csname Esmall#1\endcsname \afterenvbreak
}
% Define two synonyms:
\def\maketwodispenvs #1#2#3{
\makedispenv{#1}{#3}
\makedispenv{#2}{#3}
}
% @lisp: indented, narrowed, typewriter font; @example: same as @lisp.
%
% @smallexample and @smalllisp: use smaller fonts.
% Originally contributed by Pavel@xerox.
%
\maketwodispenvs {lisp}{example}{%
\nonfillstart
\tt\quoteexpand
\let\kbdfont = \kbdexamplefont % Allow @kbd to do something special.
\gobble % eat return
}
% @display/@smalldisplay: same as @lisp except keep current font.
%
\makedispenv {display}{%
\nonfillstart
\gobble
}
% @format/@smallformat: same as @display except don't narrow margins.
%
\makedispenv{format}{%
\let\nonarrowing = t%
\nonfillstart
\gobble
}
% @flushleft: same as @format, but doesn't obey \SETdispenvsize.
\envdef\flushleft{%
\let\nonarrowing = t%
\nonfillstart
\gobble
}
\let\Eflushleft = \afterenvbreak
% @flushright.
%
\envdef\flushright{%
\let\nonarrowing = t%
\nonfillstart
\advance\leftskip by 0pt plus 1fill
\gobble
}
\let\Eflushright = \afterenvbreak
% @quotation does normal linebreaking (hence we can't use \nonfillstart)
% and narrows the margins. We keep \parskip nonzero in general, since
% we're doing normal filling. So, when using \aboveenvbreak and
% \afterenvbreak, temporarily make \parskip 0.
%
\envdef\quotation{%
{\parskip=0pt \aboveenvbreak}% because \aboveenvbreak inserts \parskip
\parindent=0pt
%
% @cartouche defines \nonarrowing to inhibit narrowing at next level down.
\ifx\nonarrowing\relax
\advance\leftskip by \lispnarrowing
\advance\rightskip by \lispnarrowing
\exdentamount = \lispnarrowing
\else
\let\nonarrowing = \relax
\fi
\parsearg\quotationlabel
}
% We have retained a nonzero parskip for the environment, since we're
% doing normal filling.
%
\def\Equotation{%
\par
\ifx\quotationauthor\undefined\else
% indent a bit.
\leftline{\kern 2\leftskip \sl ---\quotationauthor}%
\fi
{\parskip=0pt \afterenvbreak}%
}
% If we're given an argument, typeset it in bold with a colon after.
\def\quotationlabel#1{%
\def\temp{#1}%
\ifx\temp\empty \else
{\bf #1: }%
\fi
}
% LaTeX-like @verbatim...@end verbatim and @verb{<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}
% Allow an option to not replace quotes with a regular directed right
% quote/apostrophe (char 0x27), but instead use the undirected quote
% from cmtt (char 0x0d). The undirected quote is ugly, so don't make it
% the default, but it works for pasting with more pdf viewers (at least
% evince), the lilypond developers report. xpdf does work with the
% regular 0x27.
%
\def\codequoteright{%
\expandafter\ifx\csname SETcodequoteundirected\endcsname\relax
'%
\else
\char'15
\fi
}
%
% and a similar option for the left quote char vs. a grave accent.
% Modern fonts display ASCII 0x60 as a grave accent, so some people like
% the code environments to do likewise.
%
\def\codequoteleft{%
\expandafter\ifx\csname SETcodequotebacktick\endcsname\relax
`%
\else
\char'22
\fi
}
%
\begingroup
\catcode`\^^I=\active
\gdef\tabexpand{%
\catcode`\^^I=\active
\def^^I{\leavevmode\egroup
\dimen0=\wd0 % the width so far, or since the previous tab
\divide\dimen0 by\tabw
\multiply\dimen0 by\tabw % compute previous multiple of \tabw
\advance\dimen0 by\tabw % advance to next multiple of \tabw
\wd0=\dimen0 \box0 \starttabbox
}%
}
\catcode`\'=\active
\gdef\rquoteexpand{\catcode\rquoteChar=\active \def'{\codequoteright}}%
%
\catcode`\`=\active
\gdef\lquoteexpand{\catcode\lquoteChar=\active \def`{\codequoteleft}}%
%
\gdef\quoteexpand{\rquoteexpand \lquoteexpand}%
\endgroup
% start the verbatim environment.
\def\setupverbatim{%
\let\nonarrowing = t%
\nonfillstart
% Easiest (and conventionally used) font for verbatim
\tt
\def\par{\leavevmode\egroup\box0\endgraf}%
\catcode`\`=\active
\tabexpand
\quoteexpand
% Respect line breaks,
% print special symbols as themselves, and
% make each space count
% must do in this order:
\obeylines \uncatcodespecials \sepspaces
\everypar{\starttabbox}%
}
% Do the @verb magic: verbatim text is quoted by unique
% delimiter characters. Before first delimiter expect a
% right brace, after last delimiter expect closing brace:
%
% \def\doverb'{'<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
\newcount\defunpenalty
% Start the processing of @deffn:
\def\startdefun{%
\ifnum\lastpenalty<10000
\medbreak
\defunpenalty=10003 % Will keep this @deffn together with the
% following @def command, see below.
\else
% If there are two @def commands in a row, we'll have a \nobreak,
% which is there to keep the function description together with its
% header. But if there's nothing but headers, we need to allow a
% break somewhere. Check specifically for penalty 10002, inserted
% by \printdefunline, instead of 10000, since the sectioning
% commands also insert a nobreak penalty, and we don't want to allow
% a break between a section heading and a defun.
%
% As a minor refinement, we avoid "club" headers by signalling
% with penalty of 10003 after the very first @deffn in the
% sequence (see above), and penalty of 10002 after any following
% @def command.
\ifnum\lastpenalty=10002 \penalty2000 \else \defunpenalty=10002 \fi
%
% Similarly, after a section heading, do not allow a break.
% But do insert the glue.
\medskip % preceded by discardable penalty, so not a breakpoint
\fi
%
\parindent=0in
\advance\leftskip by \defbodyindent
\exdentamount=\defbodyindent
}
\def\dodefunx#1{%
% First, check whether we are in the right environment:
\checkenv#1%
%
% As above, allow line break if we have multiple x headers in a row.
% It's not a great place, though.
\ifnum\lastpenalty=10002 \penalty3000 \else \defunpenalty=10002 \fi
%
% And now, it's time to reuse the body of the original defun:
\expandafter\gobbledefun#1%
}
\def\gobbledefun#1\startdefun{}
% \printdefunline \deffnheader{text}
%
\def\printdefunline#1#2{%
\begingroup
% call \deffnheader:
#1#2 \endheader
% common ending:
\interlinepenalty = 10000
\advance\rightskip by 0pt plus 1fil
\endgraf
\nobreak\vskip -\parskip
\penalty\defunpenalty % signal to \startdefun and \dodefunx
% Some of the @defun-type tags do not enable magic parentheses,
% rendering the following check redundant. But we don't optimize.
\checkparencounts
\endgroup
}
\def\Edefun{\endgraf\medbreak}
% \makedefun{deffn} creates \deffn, \deffnx and \Edeffn;
% the only thing remainnig is to define \deffnheader.
%
\def\makedefun#1{%
\expandafter\let\csname E#1\endcsname = \Edefun
\edef\temp{\noexpand\domakedefun
\makecsname{#1}\makecsname{#1x}\makecsname{#1header}}%
\temp
}
% \domakedefun \deffn \deffnx \deffnheader
%
% Define \deffn and \deffnx, without parameters.
% \deffnheader has to be defined explicitly.
%
\def\domakedefun#1#2#3{%
\envdef#1{%
\startdefun
\parseargusing\activeparens{\printdefunline#3}%
}%
\def#2{\dodefunx#1}%
\def#3%
}
%%% Untyped functions:
% @deffn category name args
\makedefun{deffn}{\deffngeneral{}}
% @deffn category class name args
\makedefun{defop}#1 {\defopon{#1\ \putwordon}}
% \defopon {category on}class name args
\def\defopon#1#2 {\deffngeneral{\putwordon\ \code{#2}}{#1\ \code{#2}} }
% \deffngeneral {subind}category name args
%
\def\deffngeneral#1#2 #3 #4\endheader{%
% Remember that \dosubind{fn}{foo}{} is equivalent to \doind{fn}{foo}.
\dosubind{fn}{\code{#3}}{#1}%
\defname{#2}{}{#3}\magicamp\defunargs{#4\unskip}%
}
%%% Typed functions:
% @deftypefn category type name args
\makedefun{deftypefn}{\deftypefngeneral{}}
% @deftypeop category class type name args
\makedefun{deftypeop}#1 {\deftypeopon{#1\ \putwordon}}
% \deftypeopon {category on}class type name args
\def\deftypeopon#1#2 {\deftypefngeneral{\putwordon\ \code{#2}}{#1\ \code{#2}} }
% \deftypefngeneral {subind}category type name args
%
\def\deftypefngeneral#1#2 #3 #4 #5\endheader{%
\dosubind{fn}{\code{#4}}{#1}%
\defname{#2}{#3}{#4}\defunargs{#5\unskip}%
}
%%% Typed variables:
% @deftypevr category type var args
\makedefun{deftypevr}{\deftypecvgeneral{}}
% @deftypecv category class type var args
\makedefun{deftypecv}#1 {\deftypecvof{#1\ \putwordof}}
% \deftypecvof {category of}class type var args
\def\deftypecvof#1#2 {\deftypecvgeneral{\putwordof\ \code{#2}}{#1\ \code{#2}} }
% \deftypecvgeneral {subind}category type var args
%
\def\deftypecvgeneral#1#2 #3 #4 #5\endheader{%
\dosubind{vr}{\code{#4}}{#1}%
\defname{#2}{#3}{#4}\defunargs{#5\unskip}%
}
%%% Untyped variables:
% @defvr category var args
\makedefun{defvr}#1 {\deftypevrheader{#1} {} }
% @defcv category class var args
\makedefun{defcv}#1 {\defcvof{#1\ \putwordof}}
% \defcvof {category of}class var args
\def\defcvof#1#2 {\deftypecvof{#1}#2 {} }
%%% Type:
% @deftp category name args
\makedefun{deftp}#1 #2 #3\endheader{%
\doind{tp}{\code{#2}}%
\defname{#1}{}{#2}\defunargs{#3\unskip}%
}
% Remaining @defun-like shortcuts:
\makedefun{defun}{\deffnheader{\putwordDeffunc} }
\makedefun{defmac}{\deffnheader{\putwordDefmac} }
\makedefun{defspec}{\deffnheader{\putwordDefspec} }
\makedefun{deftypefun}{\deftypefnheader{\putwordDeffunc} }
\makedefun{defvar}{\defvrheader{\putwordDefvar} }
\makedefun{defopt}{\defvrheader{\putwordDefopt} }
\makedefun{deftypevar}{\deftypevrheader{\putwordDefvar} }
\makedefun{defmethod}{\defopon\putwordMethodon}
\makedefun{deftypemethod}{\deftypeopon\putwordMethodon}
\makedefun{defivar}{\defcvof\putwordInstanceVariableof}
\makedefun{deftypeivar}{\deftypecvof\putwordInstanceVariableof}
% \defname, which formats the name of the @def (not the args).
% #1 is the category, such as "Function".
% #2 is the return type, if any.
% #3 is the function name.
%
% We are followed by (but not passed) the arguments, if any.
%
\def\defname#1#2#3{%
% Get the values of \leftskip and \rightskip as they were outside the @def...
\advance\leftskip by -\defbodyindent
%
% How we'll format the type name. Putting it in brackets helps
% distinguish it from the body text that may end up on the next line
% just below it.
\def\temp{#1}%
\setbox0=\hbox{\kern\deflastargmargin \ifx\temp\empty\else [\rm\temp]\fi}
%
% Figure out line sizes for the paragraph shape.
% The first line needs space for \box0; but if \rightskip is nonzero,
% we need only space for the part of \box0 which exceeds it:
\dimen0=\hsize \advance\dimen0 by -\wd0 \advance\dimen0 by \rightskip
% The continuations:
\dimen2=\hsize \advance\dimen2 by -\defargsindent
% (plain.tex says that \dimen1 should be used only as global.)
\parshape 2 0in \dimen0 \defargsindent \dimen2
%
% Put the type name to the right margin.
\noindent
\hbox to 0pt{%
\hfil\box0 \kern-\hsize
% \hsize has to be shortened this way:
\kern\leftskip
% Intentionally do not respect \rightskip, since we need the space.
}%
%
% Allow all lines to be underfull without complaint:
\tolerance=10000 \hbadness=10000
\exdentamount=\defbodyindent
{%
% defun fonts. We use typewriter by default (used to be bold) because:
% . we're printing identifiers, they should be in tt in principle.
% . in languages with many accents, such as Czech or French, it's
% common to leave accents off identifiers. The result looks ok in
% tt, but exceedingly strange in rm.
% . we don't want -- and --- to be treated as ligatures.
% . this still does not fix the ?` and !` ligatures, but so far no
% one has made identifiers using them :).
\df \tt
\def\temp{#2}% return value type
\ifx\temp\empty\else \tclose{\temp} \fi
#3% output function name
}%
{\rm\enskip}% hskip 0.5 em of \tenrm
%
\boldbrax
% arguments will be output next, if any.
}
% Print arguments in slanted roman (not ttsl), inconsistently with using
% tt for the name. This is because literal text is sometimes needed in
% the argument list (groff manual), and ttsl and tt are not very
% distinguishable. Prevent hyphenation at `-' chars.
%
\def\defunargs#1{%
% use sl by default (not ttsl),
% tt for the names.
\df \sl \hyphenchar\font=0
%
% On the other hand, if an argument has two dashes (for instance), we
% want a way to get ttsl. Let's try @var for that.
\let\var=\ttslanted
#1%
\sl\hyphenchar\font=45
}
% We want ()&[] to print specially on the defun line.
%
\def\activeparens{%
\catcode`\(=\active \catcode`\)=\active
\catcode`\[=\active \catcode`\]=\active
\catcode`\&=\active
}
% Make control sequences which act like normal parenthesis chars.
\let\lparen = ( \let\rparen = )
% Be sure that we always have a definition for `(', etc. For example,
% if the fn name has parens in it, \boldbrax will not be in effect yet,
% so TeX would otherwise complain about undefined control sequence.
{
\activeparens
\global\let(=\lparen \global\let)=\rparen
\global\let[=\lbrack \global\let]=\rbrack
\global\let& = \&
\gdef\boldbrax{\let(=\opnr\let)=\clnr\let[=\lbrb\let]=\rbrb}
\gdef\magicamp{\let&=\amprm}
}
\newcount\parencount
% If we encounter &foo, then turn on ()-hacking afterwards
\newif\ifampseen
\def\amprm#1 {\ampseentrue{\bf\&#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?
% List of all defined macros in the form
% \definedummyword\macro1\definedummyword\macro2...
% Currently is also contains all @aliases; the list can be split
% if there is a need.
\def\macrolist{}
% Add the macro to \macrolist
\def\addtomacrolist#1{\expandafter \addtomacrolistxxx \csname#1\endcsname}
\def\addtomacrolistxxx#1{%
\toks0 = \expandafter{\macrolist\definedummyword#1}%
\xdef\macrolist{\the\toks0}%
}
% Utility routines.
% This does \let #1 = #2, with \csnames; that is,
% \let \csname#1\endcsname = \csname#2\endcsname
% (except of course we have to play expansion games).
%
\def\cslet#1#2{%
\expandafter\let
\csname#1\expandafter\endcsname
\csname#2\endcsname
}
% Trim leading and trailing spaces off a string.
% Concepts from aro-bend problem 15 (see CTAN).
{\catcode`\@=11
\gdef\eatspaces #1{\expandafter\trim@\expandafter{#1 }}
\gdef\trim@ #1{\trim@@ @#1 @ #1 @ @@}
\gdef\trim@@ #1@ #2@ #3@@{\trim@@@\empty #2 @}
\def\unbrace#1{#1}
\unbrace{\gdef\trim@@@ #1 } #2@{#1}
}
% Trim a single trailing ^^M off a string.
{\catcode`\^^M=\other \catcode`\Q=3%
\gdef\eatcr #1{\eatcra #1Q^^MQ}%
\gdef\eatcra#1^^MQ{\eatcrb#1Q}%
\gdef\eatcrb#1Q#2Q{#1}%
}
% Macro bodies are absorbed as an argument in a context where
% all characters are catcode 10, 11 or 12, except \ which is active
% (as in normal texinfo). It is necessary to change the definition of \.
% It's necessary to have hard CRs when the macro is executed. This is
% done by making ^^M (\endlinechar) catcode 12 when reading the macro
% body, and then making it the \newlinechar in \scanmacro.
\def\scanctxt{%
\catcode`\"=\other
\catcode`\+=\other
\catcode`\<=\other
\catcode`\>=\other
\catcode`\@=\other
\catcode`\^=\other
\catcode`\_=\other
\catcode`\|=\other
\catcode`\~=\other
}
\def\scanargctxt{%
\scanctxt
\catcode`\\=\other
\catcode`\^^M=\other
}
\def\macrobodyctxt{%
\scanctxt
\catcode`\{=\other
\catcode`\}=\other
\catcode`\^^M=\other
\usembodybackslash
}
\def\macroargctxt{%
\scanctxt
\catcode`\\=\other
}
% \mbodybackslash is the definition of \ in @macro bodies.
% It maps \foo\ => \csname macarg.foo\endcsname => #N
% where N is the macro parameter number.
% We define \csname macarg.\endcsname to be \realbackslash, so
% \\ in macro replacement text gets you a backslash.
{\catcode`@=0 @catcode`@\=@active
@gdef@usembodybackslash{@let\=@mbodybackslash}
@gdef@mbodybackslash#1\{@csname macarg.#1@endcsname}
}
\expandafter\def\csname macarg.\endcsname{\realbackslash}
\def\macro{\recursivefalse\parsearg\macroxxx}
\def\rmacro{\recursivetrue\parsearg\macroxxx}
\def\macroxxx#1{%
\getargs{#1}% now \macname is the macname and \argl the arglist
\ifx\argl\empty % no arguments
\paramno=0%
\else
\expandafter\parsemargdef \argl;%
\fi
\if1\csname ismacro.\the\macname\endcsname
\message{Warning: redefining \the\macname}%
\else
\expandafter\ifx\csname \the\macname\endcsname \relax
\else \errmessage{Macro name \the\macname\space already defined}\fi
\global\cslet{macsave.\the\macname}{\the\macname}%
\global\expandafter\let\csname ismacro.\the\macname\endcsname=1%
\addtomacrolist{\the\macname}%
\fi
\begingroup \macrobodyctxt
\ifrecursive \expandafter\parsermacbody
\else \expandafter\parsemacbody
\fi}
\parseargdef\unmacro{%
\if1\csname ismacro.#1\endcsname
\global\cslet{#1}{macsave.#1}%
\global\expandafter\let \csname ismacro.#1\endcsname=0%
% Remove the macro name from \macrolist:
\begingroup
\expandafter\let\csname#1\endcsname \relax
\let\definedummyword\unmacrodo
\xdef\macrolist{\macrolist}%
\endgroup
\else
\errmessage{Macro #1 not defined}%
\fi
}
% Called by \do from \dounmacro on each macro. The idea is to omit any
% macro definitions that have been changed to \relax.
%
\def\unmacrodo#1{%
\ifx #1\relax
% remove this
\else
\noexpand\definedummyword \noexpand#1%
\fi
}
% This makes use of the obscure feature that if the last token of a
% <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\macnamexxx=#1\futurelet\nchar\braceorlinexxx}
\def\braceorlinexxx{%
\ifx\nchar\bgroup\else
\expandafter\parsearg
\fi \macnamexxx}
% @alias.
% We need some trickery to remove the optional spaces around the equal
% sign. Just make them active and then expand them all to nothing.
\def\alias{\parseargusing\obeyspaces\aliasxxx}
\def\aliasxxx #1{\aliasyyy#1\relax}
\def\aliasyyy #1=#2\relax{%
{%
\expandafter\let\obeyedspace=\empty
\addtomacrolist{#1}%
\xdef\next{\global\let\makecsname{#1}=\makecsname{#2}}%
}%
\next
}
\message{cross references,}
\newwrite\auxfile
\newif\ifhavexrefs % True if xref values are known.
\newif\ifwarnedxrefs % True if we warned once that they aren't known.
% @inforef is relatively simple.
\def\inforef #1{\inforefzzz #1,,,,**}
\def\inforefzzz #1,#2,#3,#4**{\putwordSee{} \putwordInfo{} \putwordfile{} \file{\ignorespaces #3{}},
node \samp{\ignorespaces#1{}}}
% @node's only job in TeX is to define \lastnode, which is used in
% cross-references. The @node line might or might not have commas, and
% might or might not have spaces before the first comma, like:
% @node foo , bar , ...
% We don't want such trailing spaces in the node name.
%
\parseargdef\node{\checkenv{}\donode #1 ,\finishnodeparse}
%
% also remove a trailing comma, in case of something like this:
% @node Help-Cross, , , Cross-refs
\def\donode#1 ,#2\finishnodeparse{\dodonode #1,\finishnodeparse}
\def\dodonode#1,#2\finishnodeparse{\gdef\lastnode{#1}}
\let\nwnode=\node
\let\lastnode=\empty
% Write a cross-reference definition for the current node. #1 is the
% type (Ynumbered, Yappendix, Ynothing).
%
\def\donoderef#1{%
\ifx\lastnode\empty\else
\setref{\lastnode}{#1}%
\global\let\lastnode=\empty
\fi
}
% @anchor{NAME} -- define xref target at arbitrary point.
%
\newcount\savesfregister
%
\def\savesf{\relax \ifhmode \savesfregister=\spacefactor \fi}
\def\restoresf{\relax \ifhmode \spacefactor=\savesfregister \fi}
\def\anchor#1{\savesf \setref{#1}{Ynothing}\restoresf \ignorespaces}
% \setref{NAME}{SNT} defines a cross-reference point NAME (a node or an
% anchor), which consists of three parts:
% 1) NAME-title - the current sectioning name taken from \thissection,
% or the anchor name.
% 2) NAME-snt - section number and type, passed as the SNT arg, or
% empty for anchors.
% 3) NAME-pg - the page number.
%
% This is called from \donoderef, \anchor, and \dofloat. In the case of
% floats, there is an additional part, which is not written here:
% 4) NAME-lof - the text as it should appear in a @listoffloats.
%
\def\setref#1#2{%
\pdfmkdest{#1}%
\iflinks
{%
\atdummies % preserve commands, but don't expand them
\edef\writexrdef##1##2{%
\write\auxfile{@xrdef{#1-% #1 of \setref, expanded by the \edef
##1}{##2}}% these are parameters of \writexrdef
}%
\toks0 = \expandafter{\thissection}%
\immediate \writexrdef{title}{\the\toks0 }%
\immediate \writexrdef{snt}{\csname #2\endcsname}% \Ynumbered etc.
\safewhatsit{\writexrdef{pg}{\folio}}% will be written later, during \shipout
}%
\fi
}
% @xref, @pxref, and @ref generate cross-references. For \xrefX, #1 is
% the node name, #2 the name of the Info cross-reference, #3 the printed
% node name, #4 the name of the Info file, #5 the name of the printed
% manual. All but the node name can be omitted.
%
\def\pxref#1{\putwordsee{} \xrefX[#1,,,,,,,]}
\def\xref#1{\putwordSee{} \xrefX[#1,,,,,,,]}
\def\ref#1{\xrefX[#1,,,,,,,]}
\def\xrefX[#1,#2,#3,#4,#5,#6]{\begingroup
\unsepspaces
\def\printedmanual{\ignorespaces #5}%
\def\printedrefname{\ignorespaces #3}%
\setbox1=\hbox{\printedmanual\unskip}%
\setbox0=\hbox{\printedrefname\unskip}%
\ifdim \wd0 = 0pt
% No printed node name was explicitly given.
\expandafter\ifx\csname SETxref-automatic-section-title\endcsname\relax
% Use the node name inside the square brackets.
\def\printedrefname{\ignorespaces #1}%
\else
% Use the actual chapter/section title appear inside
% the square brackets. Use the real section title if we have it.
\ifdim \wd1 > 0pt
% It is in another manual, so we don't have it.
\def\printedrefname{\ignorespaces #1}%
\else
\ifhavexrefs
% We know the real title if we have the xref values.
\def\printedrefname{\refx{#1-title}{}}%
\else
% Otherwise just copy the Info node name.
\def\printedrefname{\ignorespaces #1}%
\fi%
\fi
\fi
\fi
%
% Make link in pdf output.
\ifpdf
\leavevmode
\getfilename{#4}%
{\indexnofonts
\turnoffactive
% See comments at \activebackslashdouble.
{\activebackslashdouble \xdef\pdfxrefdest{#1}%
\backslashparens\pdfxrefdest}%
%
\ifnum\filenamelength>0
\startlink attr{/Border [0 0 0]}%
goto file{\the\filename.pdf} name{\pdfxrefdest}%
\else
\startlink attr{/Border [0 0 0]}%
goto name{\pdfmkpgn{\pdfxrefdest}}%
\fi
}%
\linkcolor
\fi
%
% Float references are printed completely differently: "Figure 1.2"
% instead of "[somenode], p.3". We distinguish them by the
% LABEL-title being set to a magic string.
{%
% Have to otherify everything special to allow the \csname to
% include an _ in the xref name, etc.
\indexnofonts
\turnoffactive
\expandafter\global\expandafter\let\expandafter\Xthisreftitle
\csname XR#1-title\endcsname
}%
\iffloat\Xthisreftitle
% If the user specified the print name (third arg) to the ref,
% print it instead of our usual "Figure 1.2".
\ifdim\wd0 = 0pt
\refx{#1-snt}{}%
\else
\printedrefname
\fi
%
% if the user also gave the printed manual name (fifth arg), append
% "in MANUALNAME".
\ifdim \wd1 > 0pt
\space \putwordin{} \cite{\printedmanual}%
\fi
\else
% node/anchor (non-float) references.
%
% If we use \unhbox0 and \unhbox1 to print the node names, TeX does not
% insert empty discretionaries after hyphens, which means that it will
% not find a line break at a hyphen in a node names. Since some manuals
% are best written with fairly long node names, containing hyphens, this
% is a loss. Therefore, we give the text of the node name again, so it
% is as if TeX is seeing it for the first time.
\ifdim \wd1 > 0pt
\putwordsection{} ``\printedrefname'' \putwordin{} \cite{\printedmanual}%
\else
% _ (for example) has to be the character _ for the purposes of the
% control sequence corresponding to the node, but it has to expand
% into the usual \leavevmode...\vrule stuff for purposes of
% printing. So we \turnoffactive for the \refx-snt, back on for the
% printing, back off for the \refx-pg.
{\turnoffactive
% Only output a following space if the -snt ref is nonempty; for
% @unnumbered and @anchor, it won't be.
\setbox2 = \hbox{\ignorespaces \refx{#1-snt}{}}%
\ifdim \wd2 > 0pt \refx{#1-snt}\space\fi
}%
% output the `[mynode]' via a macro so it can be overridden.
\xrefprintnodename\printedrefname
%
% But we always want a comma and a space:
,\space
%
% output the `page 3'.
\turnoffactive \putwordpage\tie\refx{#1-pg}{}%
\fi
\fi
\endlink
\endgroup}
% This macro is called from \xrefX for the `[nodename]' part of xref
% output. It's a separate macro only so it can be changed more easily,
% since square brackets don't work well in some documents. Particularly
% one that Bob is working on :).
%
\def\xrefprintnodename#1{[#1]}
% Things referred to by \setref.
%
\def\Ynothing{}
\def\Yomitfromtoc{}
\def\Ynumbered{%
\ifnum\secno=0
\putwordChapter@tie \the\chapno
\else \ifnum\subsecno=0
\putwordSection@tie \the\chapno.\the\secno
\else \ifnum\subsubsecno=0
\putwordSection@tie \the\chapno.\the\secno.\the\subsecno
\else
\putwordSection@tie \the\chapno.\the\secno.\the\subsecno.\the\subsubsecno
\fi\fi\fi
}
\def\Yappendix{%
\ifnum\secno=0
\putwordAppendix@tie @char\the\appendixno{}%
\else \ifnum\subsecno=0
\putwordSection@tie @char\the\appendixno.\the\secno
\else \ifnum\subsubsecno=0
\putwordSection@tie @char\the\appendixno.\the\secno.\the\subsecno
\else
\putwordSection@tie
@char\the\appendixno.\the\secno.\the\subsecno.\the\subsubsecno
\fi\fi\fi
}
% Define \refx{NAME}{SUFFIX} to reference a cross-reference string named NAME.
% If its value is nonempty, SUFFIX is output afterward.
%
\def\refx#1#2{%
{%
\indexnofonts
\otherbackslash
\expandafter\global\expandafter\let\expandafter\thisrefX
\csname XR#1\endcsname
}%
\ifx\thisrefX\relax
% If not defined, say something at least.
\angleleft un\-de\-fined\angleright
\iflinks
\ifhavexrefs
\message{\linenumber Undefined cross reference `#1'.}%
\else
\ifwarnedxrefs\else
\global\warnedxrefstrue
\message{Cross reference values unknown; you must run TeX again.}%
\fi
\fi
\fi
\else
% It's defined, so just use it.
\thisrefX
\fi
#2% Output the suffix in any case.
}
% This is the macro invoked by entries in the aux file. Usually it's
% just a \def (we prepend XR to the control sequence name to avoid
% collisions). But if this is a float type, we have more work to do.
%
\def\xrdef#1#2{%
{% The node name might contain 8-bit characters, which in our current
% implementation are changed to commands like @'e. Don't let these
% mess up the control sequence name.
\indexnofonts
\turnoffactive
\xdef\safexrefname{#1}%
}%
%
\expandafter\gdef\csname XR\safexrefname\endcsname{#2}% remember this xref
%
% Was that xref control sequence that we just defined for a float?
\expandafter\iffloat\csname XR\safexrefname\endcsname
% it was a float, and we have the (safe) float type in \iffloattype.
\expandafter\let\expandafter\floatlist
\csname floatlist\iffloattype\endcsname
%
% Is this the first time we've seen this float type?
\expandafter\ifx\floatlist\relax
\toks0 = {\do}% yes, so just \do
\else
% had it before, so preserve previous elements in list.
\toks0 = \expandafter{\floatlist\do}%
\fi
%
% Remember this xref in the control sequence \floatlistFLOATTYPE,
% for later use in \listoffloats.
\expandafter\xdef\csname floatlist\iffloattype\endcsname{\the\toks0
{\safexrefname}}%
\fi
}
% Read the last existing aux file, if any. No error if none exists.
%
\def\tryauxfile{%
\openin 1 \jobname.aux
\ifeof 1 \else
\readdatafile{aux}%
\global\havexrefstrue
\fi
\closein 1
}
\def\setupdatafile{%
\catcode`\^^@=\other
\catcode`\^^A=\other
\catcode`\^^B=\other
\catcode`\^^C=\other
\catcode`\^^D=\other
\catcode`\^^E=\other
\catcode`\^^F=\other
\catcode`\^^G=\other
\catcode`\^^H=\other
\catcode`\^^K=\other
\catcode`\^^L=\other
\catcode`\^^N=\other
\catcode`\^^P=\other
\catcode`\^^Q=\other
\catcode`\^^R=\other
\catcode`\^^S=\other
\catcode`\^^T=\other
\catcode`\^^U=\other
\catcode`\^^V=\other
\catcode`\^^W=\other
\catcode`\^^X=\other
\catcode`\^^Z=\other
\catcode`\^^[=\other
\catcode`\^^\=\other
\catcode`\^^]=\other
\catcode`\^^^=\other
\catcode`\^^_=\other
% It was suggested to set the catcode of ^ to 7, which would allow ^^e4 etc.
% in xref tags, i.e., node names. But since ^^e4 notation isn't
% supported in the main text, it doesn't seem desirable. Furthermore,
% that is not enough: for node names that actually contain a ^
% character, we would end up writing a line like this: 'xrdef {'hat
% b-title}{'hat b} and \xrdef does a \csname...\endcsname on the first
% argument, and \hat is not an expandable control sequence. It could
% all be worked out, but why? Either we support ^^ or we don't.
%
% The other change necessary for this was to define \auxhat:
% \def\auxhat{\def^{'hat }}% extra space so ok if followed by letter
% and then to call \auxhat in \setq.
%
\catcode`\^=\other
%
% Special characters. Should be turned off anyway, but...
\catcode`\~=\other
\catcode`\[=\other
\catcode`\]=\other
\catcode`\"=\other
\catcode`\_=\other
\catcode`\|=\other
\catcode`\<=\other
\catcode`\>=\other
\catcode`\$=\other
\catcode`\#=\other
\catcode`\&=\other
\catcode`\%=\other
\catcode`+=\other % avoid \+ for paranoia even though we've turned it off
%
% This is to support \ in node names and titles, since the \
% characters end up in a \csname. It's easier than
% leaving it active and making its active definition an actual \
% character. What I don't understand is why it works in the *value*
% of the xrdef. Seems like it should be a catcode12 \, and that
% should not typeset properly. But it works, so I'm moving on for
% now. --karl, 15jan04.
\catcode`\\=\other
%
% Make the characters 128-255 be printing characters.
{%
\count1=128
\def\loop{%
\catcode\count1=\other
\advance\count1 by 1
\ifnum \count1<256 \loop \fi
}%
}%
%
% @ is our escape character in .aux files, and we need braces.
\catcode`\{=1
\catcode`\}=2
\catcode`\@=0
}
\def\readdatafile#1{%
\begingroup
\setupdatafile
\input\jobname.#1
\endgroup}
\message{insertions,}
% including footnotes.
\newcount \footnoteno
% The trailing space in the following definition for supereject is
% vital for proper filling; pages come out unaligned when you do a
% pagealignmacro call if that space before the closing brace is
% removed. (Generally, numeric constants should always be followed by a
% space to prevent strange expansion errors.)
\def\supereject{\par\penalty -20000\footnoteno =0 }
% @footnotestyle is meaningful for info output only.
\let\footnotestyle=\comment
{\catcode `\@=11
%
% Auto-number footnotes. Otherwise like plain.
\gdef\footnote{%
\let\indent=\ptexindent
\let\noindent=\ptexnoindent
\global\advance\footnoteno by \@ne
\edef\thisfootno{$^{\the\footnoteno}$}%
%
% In case the footnote comes at the end of a sentence, preserve the
% extra spacing after we do the footnote number.
\let\@sf\empty
\ifhmode\edef\@sf{\spacefactor\the\spacefactor}\ptexslash\fi
%
% Remove inadvertent blank space before typesetting the footnote number.
\unskip
\thisfootno\@sf
\dofootnote
}%
% Don't bother with the trickery in plain.tex to not require the
% footnote text as a parameter. Our footnotes don't need to be so general.
%
% Oh yes, they do; otherwise, @ifset (and anything else that uses
% \parseargline) fails inside footnotes because the tokens are fixed when
% the footnote is read. --karl, 16nov96.
%
\gdef\dofootnote{%
\insert\footins\bgroup
% We want to typeset this text as a normal paragraph, even if the
% footnote reference occurs in (for example) a display environment.
% So reset some parameters.
\hsize=\pagewidth
\interlinepenalty\interfootnotelinepenalty
\splittopskip\ht\strutbox % top baseline for broken footnotes
\splitmaxdepth\dp\strutbox
\floatingpenalty\@MM
\leftskip\z@skip
\rightskip\z@skip
\spaceskip\z@skip
\xspaceskip\z@skip
\parindent\defaultparindent
%
\smallfonts \rm
%
% Because we use hanging indentation in footnotes, a @noindent appears
% to exdent this text, so make it be a no-op. makeinfo does not use
% hanging indentation so @noindent can still be needed within footnote
% text after an @example or the like (not that this is good style).
\let\noindent = \relax
%
% Hang the footnote text off the number. Use \everypar in case the
% footnote extends for more than one paragraph.
\everypar = {\hang}%
\textindent{\thisfootno}%
%
% Don't crash into the line above the footnote text. Since this
% expands into a box, it must come within the paragraph, lest it
% provide a place where TeX can split the footnote.
\footstrut
\futurelet\next\fo@t
}
}%end \catcode `\@=11
% In case a @footnote appears in a vbox, save the footnote text and create
% the real \insert just after the vbox finished. Otherwise, the insertion
% would be lost.
-% Similarily, if a @footnote appears inside an alignment, save the footnote
+% Similarly, if a @footnote appears inside an alignment, save the footnote
% text to a box and make the \insert when a row of the table is finished.
% And the same can be done for other insert classes. --kasal, 16nov03.
% Replace the \insert primitive by a cheating macro.
% Deeper inside, just make sure that the saved insertions are not spilled
% out prematurely.
%
\def\startsavinginserts{%
\ifx \insert\ptexinsert
\let\insert\saveinsert
\else
\let\checkinserts\relax
\fi
}
% This \insert replacement works for both \insert\footins{foo} and
% \insert\footins\bgroup foo\egroup, but it doesn't work for \insert27{foo}.
%
\def\saveinsert#1{%
\edef\next{\noexpand\savetobox \makeSAVEname#1}%
\afterassignment\next
% swallow the left brace
\let\temp =
}
\def\makeSAVEname#1{\makecsname{SAVE\expandafter\gobble\string#1}}
\def\savetobox#1{\global\setbox#1 = \vbox\bgroup \unvbox#1}
\def\checksaveins#1{\ifvoid#1\else \placesaveins#1\fi}
\def\placesaveins#1{%
\ptexinsert \csname\expandafter\gobblesave\string#1\endcsname
{\box#1}%
}
% eat @SAVE -- beware, all of them have catcode \other:
{
\def\dospecials{\do S\do A\do V\do E} \uncatcodespecials % ;-)
\gdef\gobblesave @SAVE{}
}
% initialization:
\def\newsaveins #1{%
\edef\next{\noexpand\newsaveinsX \makeSAVEname#1}%
\next
}
\def\newsaveinsX #1{%
\csname newbox\endcsname #1%
\expandafter\def\expandafter\checkinserts\expandafter{\checkinserts
\checksaveins #1}%
}
% initialize:
\let\checkinserts\empty
\newsaveins\footins
\newsaveins\margin
% @image. We use the macros from epsf.tex to support this.
% If epsf.tex is not installed and @image is used, we complain.
%
% Check for and read epsf.tex up front. If we read it only at @image
% time, we might be inside a group, and then its definitions would get
% undone and the next image would fail.
\openin 1 = epsf.tex
\ifeof 1 \else
% Do not bother showing banner with epsf.tex v2.7k (available in
% doc/epsf.tex and on ctan).
\def\epsfannounce{\toks0 = }%
\input epsf.tex
\fi
\closein 1
%
% We will only complain once about lack of epsf.tex.
\newif\ifwarnednoepsf
\newhelp\noepsfhelp{epsf.tex must be installed for images to
work. It is also included in the Texinfo distribution, or you can get
it from ftp://tug.org/tex/epsf.tex.}
%
\def\image#1{%
\ifx\epsfbox\undefined
\ifwarnednoepsf \else
\errhelp = \noepsfhelp
\errmessage{epsf.tex not found, images will be ignored}%
\global\warnednoepsftrue
\fi
\else
\imagexxx #1,,,,,\finish
\fi
}
%
% Arguments to @image:
% #1 is (mandatory) image filename; we tack on .eps extension.
% #2 is (optional) width, #3 is (optional) height.
% #4 is (ignored optional) html alt text.
% #5 is (ignored optional) extension.
% #6 is just the usual extra ignored arg for parsing this stuff.
\newif\ifimagevmode
\def\imagexxx#1,#2,#3,#4,#5,#6\finish{\begingroup
\catcode`\^^M = 5 % in case we're inside an example
\normalturnoffactive % allow _ et al. in names
% If the image is by itself, center it.
\ifvmode
\imagevmodetrue
\nobreak\bigskip
% Usually we'll have text after the image which will insert
% \parskip glue, so insert it here too to equalize the space
% above and below.
\nobreak\vskip\parskip
\nobreak
\line\bgroup
\fi
%
% Output the image.
\ifpdf
\dopdfimage{#1}{#2}{#3}%
\else
% \epsfbox itself resets \epsf?size at each figure.
\setbox0 = \hbox{\ignorespaces #2}\ifdim\wd0 > 0pt \epsfxsize=#2\relax \fi
\setbox0 = \hbox{\ignorespaces #3}\ifdim\wd0 > 0pt \epsfysize=#3\relax \fi
\epsfbox{#1.eps}%
\fi
%
\ifimagevmode \egroup \bigbreak \fi % space after the image
\endgroup}
% @float FLOATTYPE,LABEL,LOC ... @end float for displayed figures, tables,
% etc. We don't actually implement floating yet, we always include the
% float "here". But it seemed the best name for the future.
%
\envparseargdef\float{\eatcommaspace\eatcommaspace\dofloat#1, , ,\finish}
% There may be a space before second and/or third parameter; delete it.
\def\eatcommaspace#1, {#1,}
% #1 is the optional FLOATTYPE, the text label for this float, typically
% "Figure", "Table", "Example", etc. Can't contain commas. If omitted,
% this float will not be numbered and cannot be referred to.
%
% #2 is the optional xref label. Also must be present for the float to
% be referable.
%
% #3 is the optional positioning argument; for now, it is ignored. It
% will somehow specify the positions allowed to float to (here, top, bottom).
%
% We keep a separate counter for each FLOATTYPE, which we reset at each
% chapter-level command.
\let\resetallfloatnos=\empty
%
\def\dofloat#1,#2,#3,#4\finish{%
\let\thiscaption=\empty
\let\thisshortcaption=\empty
%
% don't lose footnotes inside @float.
%
% BEWARE: when the floats start float, we have to issue warning whenever an
% insert appears inside a float which could possibly float. --kasal, 26may04
%
\startsavinginserts
%
% We can't be used inside a paragraph.
\par
%
\vtop\bgroup
\def\floattype{#1}%
\def\floatlabel{#2}%
\def\floatloc{#3}% we do nothing with this yet.
%
\ifx\floattype\empty
\let\safefloattype=\empty
\else
{%
% the floattype might have accents or other special characters,
% but we need to use it in a control sequence name.
\indexnofonts
\turnoffactive
\xdef\safefloattype{\floattype}%
}%
\fi
%
% If label is given but no type, we handle that as the empty type.
\ifx\floatlabel\empty \else
% We want each FLOATTYPE to be numbered separately (Figure 1,
% Table 1, Figure 2, ...). (And if no label, no number.)
%
\expandafter\getfloatno\csname\safefloattype floatno\endcsname
\global\advance\floatno by 1
%
{%
% This magic value for \thissection is output by \setref as the
% XREFLABEL-title value. \xrefX uses it to distinguish float
% labels (which have a completely different output format) from
% node and anchor labels. And \xrdef uses it to construct the
% lists of floats.
%
\edef\thissection{\floatmagic=\safefloattype}%
\setref{\floatlabel}{Yfloat}%
}%
\fi
%
% start with \parskip glue, I guess.
\vskip\parskip
%
% Don't suppress indentation if a float happens to start a section.
\restorefirstparagraphindent
}
% we have these possibilities:
% @float Foo,lbl & @caption{Cap}: Foo 1.1: Cap
% @float Foo,lbl & no caption: Foo 1.1
% @float Foo & @caption{Cap}: Foo: Cap
% @float Foo & no caption: Foo
% @float ,lbl & Caption{Cap}: 1.1: Cap
% @float ,lbl & no caption: 1.1
% @float & @caption{Cap}: Cap
% @float & no caption:
%
\def\Efloat{%
\let\floatident = \empty
%
% In all cases, if we have a float type, it comes first.
\ifx\floattype\empty \else \def\floatident{\floattype}\fi
%
% If we have an xref label, the number comes next.
\ifx\floatlabel\empty \else
\ifx\floattype\empty \else % if also had float type, need tie first.
\appendtomacro\floatident{\tie}%
\fi
% the number.
\appendtomacro\floatident{\chaplevelprefix\the\floatno}%
\fi
%
% Start the printed caption with what we've constructed in
% \floatident, but keep it separate; we need \floatident again.
\let\captionline = \floatident
%
\ifx\thiscaption\empty \else
\ifx\floatident\empty \else
\appendtomacro\captionline{: }% had ident, so need a colon between
\fi
%
% caption text.
\appendtomacro\captionline{\scanexp\thiscaption}%
\fi
%
% If we have anything to print, print it, with space before.
% Eventually this needs to become an \insert.
\ifx\captionline\empty \else
\vskip.5\parskip
\captionline
%
% Space below caption.
\vskip\parskip
\fi
%
% If have an xref label, write the list of floats info. Do this
% after the caption, to avoid chance of it being a breakpoint.
\ifx\floatlabel\empty \else
% Write the text that goes in the lof to the aux file as
% \floatlabel-lof. Besides \floatident, we include the short
% caption if specified, else the full caption if specified, else nothing.
{%
\atdummies
%
% since we read the caption text in the macro world, where ^^M
% is turned into a normal character, we have to scan it back, so
% we don't write the literal three characters "^^M" into the aux file.
\scanexp{%
\xdef\noexpand\gtemp{%
\ifx\thisshortcaption\empty
\thiscaption
\else
\thisshortcaption
\fi
}%
}%
\immediate\write\auxfile{@xrdef{\floatlabel-lof}{\floatident
\ifx\gtemp\empty \else : \gtemp \fi}}%
}%
\fi
\egroup % end of \vtop
%
% place the captured inserts
%
% BEWARE: when the floats start floating, we have to issue warning
% whenever an insert appears inside a float which could possibly
% float. --kasal, 26may04
%
\checkinserts
}
% Append the tokens #2 to the definition of macro #1, not expanding either.
%
\def\appendtomacro#1#2{%
\expandafter\def\expandafter#1\expandafter{#1#2}%
}
% @caption, @shortcaption
%
\def\caption{\docaption\thiscaption}
\def\shortcaption{\docaption\thisshortcaption}
\def\docaption{\checkenv\float \bgroup\scanargctxt\defcaption}
\def\defcaption#1#2{\egroup \def#1{#2}}
% The parameter is the control sequence identifying the counter we are
% going to use. Create it if it doesn't exist and assign it to \floatno.
\def\getfloatno#1{%
\ifx#1\relax
% Haven't seen this figure type before.
\csname newcount\endcsname #1%
%
% Remember to reset this floatno at the next chap.
\expandafter\gdef\expandafter\resetallfloatnos
\expandafter{\resetallfloatnos #1=0 }%
\fi
\let\floatno#1%
}
% \setref calls this to get the XREFLABEL-snt value. We want an @xref
% to the FLOATLABEL to expand to "Figure 3.1". We call \setref when we
% first read the @float command.
%
\def\Yfloat{\floattype@tie \chaplevelprefix\the\floatno}%
% Magic string used for the XREFLABEL-title value, so \xrefX can
% distinguish floats from other xref types.
\def\floatmagic{!!float!!}
% #1 is the control sequence we are passed; we expand into a conditional
% which is true if #1 represents a float ref. That is, the magic
% \thissection value which we \setref above.
%
\def\iffloat#1{\expandafter\doiffloat#1==\finish}
%
% #1 is (maybe) the \floatmagic string. If so, #2 will be the
% (safe) float type for this float. We set \iffloattype to #2.
%
\def\doiffloat#1=#2=#3\finish{%
\def\temp{#1}%
\def\iffloattype{#2}%
\ifx\temp\floatmagic
}
% @listoffloats FLOATTYPE - print a list of floats like a table of contents.
%
\parseargdef\listoffloats{%
\def\floattype{#1}% floattype
{%
% the floattype might have accents or other special characters,
% but we need to use it in a control sequence name.
\indexnofonts
\turnoffactive
\xdef\safefloattype{\floattype}%
}%
%
% \xrdef saves the floats as a \do-list in \floatlistSAFEFLOATTYPE.
\expandafter\ifx\csname floatlist\safefloattype\endcsname \relax
\ifhavexrefs
% if the user said @listoffloats foo but never @float foo.
\message{\linenumber No `\safefloattype' floats to list.}%
\fi
\else
\begingroup
\leftskip=\tocindent % indent these entries like a toc
\let\do=\listoffloatsdo
\csname floatlist\safefloattype\endcsname
\endgroup
\fi
}
% This is called on each entry in a list of floats. We're passed the
% xref label, in the form LABEL-title, which is how we save it in the
% aux file. We strip off the -title and look up \XRLABEL-lof, which
% has the text we're supposed to typeset here.
%
% Figures without xref labels will not be included in the list (since
% they won't appear in the aux file).
%
\def\listoffloatsdo#1{\listoffloatsdoentry#1\finish}
\def\listoffloatsdoentry#1-title\finish{{%
% Can't fully expand XR#1-lof because it can contain anything. Just
% pass the control sequence. On the other hand, XR#1-pg is just the
% page number, and we want to fully expand that so we can get a link
% in pdf output.
\toksA = \expandafter{\csname XR#1-lof\endcsname}%
%
% use the same \entry macro we use to generate the TOC and index.
\edef\writeentry{\noexpand\entry{\the\toksA}{\csname XR#1-pg\endcsname}}%
\writeentry
}}
\message{localization,}
% @documentlanguage is usually given very early, just after
% @setfilename. If done too late, it may not override everything
% properly. Single argument is the language abbreviation.
% It would be nice if we could set up a hyphenation file here.
%
\parseargdef\documentlanguage{%
\tex % read txi-??.tex file in plain TeX.
% Read the file if it exists.
\openin 1 txi-#1.tex
\ifeof 1
\errhelp = \nolanghelp
\errmessage{Cannot read language file txi-#1.tex}%
\else
\input txi-#1.tex
\fi
\closein 1
\endgroup
}
\newhelp\nolanghelp{The given language definition file cannot be found or
is empty. Maybe you need to install it? In the current directory
should work if nowhere else does.}
% Set the catcode of characters 128 through 255 to the specified number.
%
\def\setnonasciicharscatcode#1{%
\count255=128
\loop\ifnum\count255<256
\global\catcode\count255=#1
\advance\count255 by 1
\repeat
}
% @documentencoding sets the definition of non-ASCII characters
% according to the specified encoding.
%
\parseargdef\documentencoding{%
% Encoding being declared for the document.
\def\declaredencoding{\csname #1.enc\endcsname}%
%
% Supported encodings: names converted to tokens in order to be able
% to compare them with \ifx.
\def\ascii{\csname US-ASCII.enc\endcsname}%
\def\latnine{\csname ISO-8859-15.enc\endcsname}%
\def\latone{\csname ISO-8859-1.enc\endcsname}%
\def\lattwo{\csname ISO-8859-2.enc\endcsname}%
\def\utfeight{\csname UTF-8.enc\endcsname}%
%
\ifx \declaredencoding \ascii
\asciichardefs
%
\else \ifx \declaredencoding \lattwo
\setnonasciicharscatcode\active
\lattwochardefs
%
\else \ifx \declaredencoding \latone
\setnonasciicharscatcode\active
\latonechardefs
%
\else \ifx \declaredencoding \latnine
\setnonasciicharscatcode\active
\latninechardefs
%
\else \ifx \declaredencoding \utfeight
\setnonasciicharscatcode\active
\utfeightchardefs
%
\else
\message{Unknown document encoding #1, ignoring.}%
%
\fi % utfeight
\fi % latnine
\fi % latone
\fi % lattwo
\fi % ascii
}
% A message to be logged when using a character that isn't available
% the default font encoding (OT1).
%
\def\missingcharmsg#1{\message{Character missing in OT1 encoding: #1.}}
% Take account of \c (plain) vs. \, (Texinfo) difference.
\def\cedilla#1{\ifx\c\ptexc\c{#1}\else\,{#1}\fi}
% First, make active non-ASCII characters in order for them to be
% correctly categorized when TeX reads the replacement text of
% macros containing the character definitions.
\setnonasciicharscatcode\active
%
% Latin1 (ISO-8859-1) character definitions.
\def\latonechardefs{%
\gdef^^a0{~}
\gdef^^a1{\exclamdown}
\gdef^^a2{\missingcharmsg{CENT SIGN}}
\gdef^^a3{{\pounds}}
\gdef^^a4{\missingcharmsg{CURRENCY SIGN}}
\gdef^^a5{\missingcharmsg{YEN SIGN}}
\gdef^^a6{\missingcharmsg{BROKEN BAR}}
\gdef^^a7{\S}
\gdef^^a8{\"{}}
\gdef^^a9{\copyright}
\gdef^^aa{\ordf}
\gdef^^ab{\missingcharmsg{LEFT-POINTING DOUBLE ANGLE QUOTATION MARK}}
\gdef^^ac{$\lnot$}
\gdef^^ad{\-}
\gdef^^ae{\registeredsymbol}
\gdef^^af{\={}}
%
\gdef^^b0{\textdegree}
\gdef^^b1{$\pm$}
\gdef^^b2{$^2$}
\gdef^^b3{$^3$}
\gdef^^b4{\'{}}
\gdef^^b5{$\mu$}
\gdef^^b6{\P}
%
\gdef^^b7{$^.$}
\gdef^^b8{\cedilla\ }
\gdef^^b9{$^1$}
\gdef^^ba{\ordm}
%
\gdef^^bb{\missingcharmsg{RIGHT-POINTING DOUBLE ANGLE QUOTATION MARK}}
\gdef^^bc{$1\over4$}
\gdef^^bd{$1\over2$}
\gdef^^be{$3\over4$}
\gdef^^bf{\questiondown}
%
\gdef^^c0{\`A}
\gdef^^c1{\'A}
\gdef^^c2{\^A}
\gdef^^c3{\~A}
\gdef^^c4{\"A}
\gdef^^c5{\ringaccent A}
\gdef^^c6{\AE}
\gdef^^c7{\cedilla C}
\gdef^^c8{\`E}
\gdef^^c9{\'E}
\gdef^^ca{\^E}
\gdef^^cb{\"E}
\gdef^^cc{\`I}
\gdef^^cd{\'I}
\gdef^^ce{\^I}
\gdef^^cf{\"I}
%
\gdef^^d0{\missingcharmsg{LATIN CAPITAL LETTER ETH}}
\gdef^^d1{\~N}
\gdef^^d2{\`O}
\gdef^^d3{\'O}
\gdef^^d4{\^O}
\gdef^^d5{\~O}
\gdef^^d6{\"O}
\gdef^^d7{$\times$}
\gdef^^d8{\O}
\gdef^^d9{\`U}
\gdef^^da{\'U}
\gdef^^db{\^U}
\gdef^^dc{\"U}
\gdef^^dd{\'Y}
\gdef^^de{\missingcharmsg{LATIN CAPITAL LETTER THORN}}
\gdef^^df{\ss}
%
\gdef^^e0{\`a}
\gdef^^e1{\'a}
\gdef^^e2{\^a}
\gdef^^e3{\~a}
\gdef^^e4{\"a}
\gdef^^e5{\ringaccent a}
\gdef^^e6{\ae}
\gdef^^e7{\cedilla c}
\gdef^^e8{\`e}
\gdef^^e9{\'e}
\gdef^^ea{\^e}
\gdef^^eb{\"e}
\gdef^^ec{\`{\dotless i}}
\gdef^^ed{\'{\dotless i}}
\gdef^^ee{\^{\dotless i}}
\gdef^^ef{\"{\dotless i}}
%
\gdef^^f0{\missingcharmsg{LATIN SMALL LETTER ETH}}
\gdef^^f1{\~n}
\gdef^^f2{\`o}
\gdef^^f3{\'o}
\gdef^^f4{\^o}
\gdef^^f5{\~o}
\gdef^^f6{\"o}
\gdef^^f7{$\div$}
\gdef^^f8{\o}
\gdef^^f9{\`u}
\gdef^^fa{\'u}
\gdef^^fb{\^u}
\gdef^^fc{\"u}
\gdef^^fd{\'y}
\gdef^^fe{\missingcharmsg{LATIN SMALL LETTER THORN}}
\gdef^^ff{\"y}
}
% Latin9 (ISO-8859-15) encoding character definitions.
\def\latninechardefs{%
% Encoding is almost identical to Latin1.
\latonechardefs
%
\gdef^^a4{\euro}
\gdef^^a6{\v S}
\gdef^^a8{\v s}
\gdef^^b4{\v Z}
\gdef^^b8{\v z}
\gdef^^bc{\OE}
\gdef^^bd{\oe}
\gdef^^be{\"Y}
}
% Latin2 (ISO-8859-2) character definitions.
\def\lattwochardefs{%
\gdef^^a0{~}
\gdef^^a1{\missingcharmsg{LATIN CAPITAL LETTER A WITH OGONEK}}
\gdef^^a2{\u{}}
\gdef^^a3{\L}
\gdef^^a4{\missingcharmsg{CURRENCY SIGN}}
\gdef^^a5{\v L}
\gdef^^a6{\'S}
\gdef^^a7{\S}
\gdef^^a8{\"{}}
\gdef^^a9{\v S}
\gdef^^aa{\cedilla S}
\gdef^^ab{\v T}
\gdef^^ac{\'Z}
\gdef^^ad{\-}
\gdef^^ae{\v Z}
\gdef^^af{\dotaccent Z}
%
\gdef^^b0{\textdegree}
\gdef^^b1{\missingcharmsg{LATIN SMALL LETTER A WITH OGONEK}}
\gdef^^b2{\missingcharmsg{OGONEK}}
\gdef^^b3{\l}
\gdef^^b4{\'{}}
\gdef^^b5{\v l}
\gdef^^b6{\'s}
\gdef^^b7{\v{}}
\gdef^^b8{\cedilla\ }
\gdef^^b9{\v s}
\gdef^^ba{\cedilla s}
\gdef^^bb{\v t}
\gdef^^bc{\'z}
\gdef^^bd{\H{}}
\gdef^^be{\v z}
\gdef^^bf{\dotaccent z}
%
\gdef^^c0{\'R}
\gdef^^c1{\'A}
\gdef^^c2{\^A}
\gdef^^c3{\u A}
\gdef^^c4{\"A}
\gdef^^c5{\'L}
\gdef^^c6{\'C}
\gdef^^c7{\cedilla C}
\gdef^^c8{\v C}
\gdef^^c9{\'E}
\gdef^^ca{\missingcharmsg{LATIN CAPITAL LETTER E WITH OGONEK}}
\gdef^^cb{\"E}
\gdef^^cc{\v E}
\gdef^^cd{\'I}
\gdef^^ce{\^I}
\gdef^^cf{\v D}
%
\gdef^^d0{\missingcharmsg{LATIN CAPITAL LETTER D WITH STROKE}}
\gdef^^d1{\'N}
\gdef^^d2{\v N}
\gdef^^d3{\'O}
\gdef^^d4{\^O}
\gdef^^d5{\H O}
\gdef^^d6{\"O}
\gdef^^d7{$\times$}
\gdef^^d8{\v R}
\gdef^^d9{\ringaccent U}
\gdef^^da{\'U}
\gdef^^db{\H U}
\gdef^^dc{\"U}
\gdef^^dd{\'Y}
\gdef^^de{\cedilla T}
\gdef^^df{\ss}
%
\gdef^^e0{\'r}
\gdef^^e1{\'a}
\gdef^^e2{\^a}
\gdef^^e3{\u a}
\gdef^^e4{\"a}
\gdef^^e5{\'l}
\gdef^^e6{\'c}
\gdef^^e7{\cedilla c}
\gdef^^e8{\v c}
\gdef^^e9{\'e}
\gdef^^ea{\missingcharmsg{LATIN SMALL LETTER E WITH OGONEK}}
\gdef^^eb{\"e}
\gdef^^ec{\v e}
\gdef^^ed{\'\i}
\gdef^^ee{\^\i}
\gdef^^ef{\v d}
%
\gdef^^f0{\missingcharmsg{LATIN SMALL LETTER D WITH STROKE}}
\gdef^^f1{\'n}
\gdef^^f2{\v n}
\gdef^^f3{\'o}
\gdef^^f4{\^o}
\gdef^^f5{\H o}
\gdef^^f6{\"o}
\gdef^^f7{$\div$}
\gdef^^f8{\v r}
\gdef^^f9{\ringaccent u}
\gdef^^fa{\'u}
\gdef^^fb{\H u}
\gdef^^fc{\"u}
\gdef^^fd{\'y}
\gdef^^fe{\cedilla t}
\gdef^^ff{\dotaccent{}}
}
% UTF-8 character definitions.
%
% This code to support UTF-8 is based on LaTeX's utf8.def, with some
% changes for Texinfo conventions. It is included here under the GPL by
% permission from Frank Mittelbach and the LaTeX team.
%
\newcount\countUTFx
\newcount\countUTFy
\newcount\countUTFz
\gdef\UTFviiiTwoOctets#1#2{\expandafter
\UTFviiiDefined\csname u8:#1\string #2\endcsname}
%
\gdef\UTFviiiThreeOctets#1#2#3{\expandafter
\UTFviiiDefined\csname u8:#1\string #2\string #3\endcsname}
%
\gdef\UTFviiiFourOctets#1#2#3#4{\expandafter
\UTFviiiDefined\csname u8:#1\string #2\string #3\string #4\endcsname}
\gdef\UTFviiiDefined#1{%
\ifx #1\relax
\message{\linenumber Unicode char \string #1 not defined for Texinfo}%
\else
\expandafter #1%
\fi
}
\begingroup
\catcode`\~13
\catcode`\"12
\def\UTFviiiLoop{%
\global\catcode\countUTFx\active
\uccode`\~\countUTFx
\uppercase\expandafter{\UTFviiiTmp}%
\advance\countUTFx by 1
\ifnum\countUTFx < \countUTFy
\expandafter\UTFviiiLoop
\fi}
\countUTFx = "C2
\countUTFy = "E0
\def\UTFviiiTmp{%
\xdef~{\noexpand\UTFviiiTwoOctets\string~}}
\UTFviiiLoop
\countUTFx = "E0
\countUTFy = "F0
\def\UTFviiiTmp{%
\xdef~{\noexpand\UTFviiiThreeOctets\string~}}
\UTFviiiLoop
\countUTFx = "F0
\countUTFy = "F4
\def\UTFviiiTmp{%
\xdef~{\noexpand\UTFviiiFourOctets\string~}}
\UTFviiiLoop
\endgroup
\begingroup
\catcode`\"=12
\catcode`\<=12
\catcode`\.=12
\catcode`\,=12
\catcode`\;=12
\catcode`\!=12
\catcode`\~=13
\gdef\DeclareUnicodeCharacter#1#2{%
\countUTFz = "#1\relax
\wlog{\space\space defining Unicode char U+#1 (decimal \the\countUTFz)}%
\begingroup
\parseXMLCharref
\def\UTFviiiTwoOctets##1##2{%
\csname u8:##1\string ##2\endcsname}%
\def\UTFviiiThreeOctets##1##2##3{%
\csname u8:##1\string ##2\string ##3\endcsname}%
\def\UTFviiiFourOctets##1##2##3##4{%
\csname u8:##1\string ##2\string ##3\string ##4\endcsname}%
\expandafter\expandafter\expandafter\expandafter
\expandafter\expandafter\expandafter
\gdef\UTFviiiTmp{#2}%
\endgroup}
\gdef\parseXMLCharref{%
\ifnum\countUTFz < "A0\relax
\errhelp = \EMsimple
\errmessage{Cannot define Unicode char value < 00A0}%
\else\ifnum\countUTFz < "800\relax
\parseUTFviiiA,%
\parseUTFviiiB C\UTFviiiTwoOctets.,%
\else\ifnum\countUTFz < "10000\relax
\parseUTFviiiA;%
\parseUTFviiiA,%
\parseUTFviiiB E\UTFviiiThreeOctets.{,;}%
\else
\parseUTFviiiA;%
\parseUTFviiiA,%
\parseUTFviiiA!%
\parseUTFviiiB F\UTFviiiFourOctets.{!,;}%
\fi\fi\fi
}
\gdef\parseUTFviiiA#1{%
\countUTFx = \countUTFz
\divide\countUTFz by 64
\countUTFy = \countUTFz
\multiply\countUTFz by 64
\advance\countUTFx by -\countUTFz
\advance\countUTFx by 128
\uccode `#1\countUTFx
\countUTFz = \countUTFy}
\gdef\parseUTFviiiB#1#2#3#4{%
\advance\countUTFz by "#10\relax
\uccode `#3\countUTFz
\uppercase{\gdef\UTFviiiTmp{#2#3#4}}}
\endgroup
\def\utfeightchardefs{%
\DeclareUnicodeCharacter{00A0}{\tie}
\DeclareUnicodeCharacter{00A1}{\exclamdown}
\DeclareUnicodeCharacter{00A3}{\pounds}
\DeclareUnicodeCharacter{00A8}{\"{ }}
\DeclareUnicodeCharacter{00A9}{\copyright}
\DeclareUnicodeCharacter{00AA}{\ordf}
\DeclareUnicodeCharacter{00AD}{\-}
\DeclareUnicodeCharacter{00AE}{\registeredsymbol}
\DeclareUnicodeCharacter{00AF}{\={ }}
\DeclareUnicodeCharacter{00B0}{\ringaccent{ }}
\DeclareUnicodeCharacter{00B4}{\'{ }}
\DeclareUnicodeCharacter{00B8}{\cedilla{ }}
\DeclareUnicodeCharacter{00BA}{\ordm}
\DeclareUnicodeCharacter{00BF}{\questiondown}
\DeclareUnicodeCharacter{00C0}{\`A}
\DeclareUnicodeCharacter{00C1}{\'A}
\DeclareUnicodeCharacter{00C2}{\^A}
\DeclareUnicodeCharacter{00C3}{\~A}
\DeclareUnicodeCharacter{00C4}{\"A}
\DeclareUnicodeCharacter{00C5}{\AA}
\DeclareUnicodeCharacter{00C6}{\AE}
\DeclareUnicodeCharacter{00C7}{\cedilla{C}}
\DeclareUnicodeCharacter{00C8}{\`E}
\DeclareUnicodeCharacter{00C9}{\'E}
\DeclareUnicodeCharacter{00CA}{\^E}
\DeclareUnicodeCharacter{00CB}{\"E}
\DeclareUnicodeCharacter{00CC}{\`I}
\DeclareUnicodeCharacter{00CD}{\'I}
\DeclareUnicodeCharacter{00CE}{\^I}
\DeclareUnicodeCharacter{00CF}{\"I}
\DeclareUnicodeCharacter{00D1}{\~N}
\DeclareUnicodeCharacter{00D2}{\`O}
\DeclareUnicodeCharacter{00D3}{\'O}
\DeclareUnicodeCharacter{00D4}{\^O}
\DeclareUnicodeCharacter{00D5}{\~O}
\DeclareUnicodeCharacter{00D6}{\"O}
\DeclareUnicodeCharacter{00D8}{\O}
\DeclareUnicodeCharacter{00D9}{\`U}
\DeclareUnicodeCharacter{00DA}{\'U}
\DeclareUnicodeCharacter{00DB}{\^U}
\DeclareUnicodeCharacter{00DC}{\"U}
\DeclareUnicodeCharacter{00DD}{\'Y}
\DeclareUnicodeCharacter{00DF}{\ss}
\DeclareUnicodeCharacter{00E0}{\`a}
\DeclareUnicodeCharacter{00E1}{\'a}
\DeclareUnicodeCharacter{00E2}{\^a}
\DeclareUnicodeCharacter{00E3}{\~a}
\DeclareUnicodeCharacter{00E4}{\"a}
\DeclareUnicodeCharacter{00E5}{\aa}
\DeclareUnicodeCharacter{00E6}{\ae}
\DeclareUnicodeCharacter{00E7}{\cedilla{c}}
\DeclareUnicodeCharacter{00E8}{\`e}
\DeclareUnicodeCharacter{00E9}{\'e}
\DeclareUnicodeCharacter{00EA}{\^e}
\DeclareUnicodeCharacter{00EB}{\"e}
\DeclareUnicodeCharacter{00EC}{\`{\dotless{i}}}
\DeclareUnicodeCharacter{00ED}{\'{\dotless{i}}}
\DeclareUnicodeCharacter{00EE}{\^{\dotless{i}}}
\DeclareUnicodeCharacter{00EF}{\"{\dotless{i}}}
\DeclareUnicodeCharacter{00F1}{\~n}
\DeclareUnicodeCharacter{00F2}{\`o}
\DeclareUnicodeCharacter{00F3}{\'o}
\DeclareUnicodeCharacter{00F4}{\^o}
\DeclareUnicodeCharacter{00F5}{\~o}
\DeclareUnicodeCharacter{00F6}{\"o}
\DeclareUnicodeCharacter{00F8}{\o}
\DeclareUnicodeCharacter{00F9}{\`u}
\DeclareUnicodeCharacter{00FA}{\'u}
\DeclareUnicodeCharacter{00FB}{\^u}
\DeclareUnicodeCharacter{00FC}{\"u}
\DeclareUnicodeCharacter{00FD}{\'y}
\DeclareUnicodeCharacter{00FF}{\"y}
\DeclareUnicodeCharacter{0100}{\=A}
\DeclareUnicodeCharacter{0101}{\=a}
\DeclareUnicodeCharacter{0102}{\u{A}}
\DeclareUnicodeCharacter{0103}{\u{a}}
\DeclareUnicodeCharacter{0106}{\'C}
\DeclareUnicodeCharacter{0107}{\'c}
\DeclareUnicodeCharacter{0108}{\^C}
\DeclareUnicodeCharacter{0109}{\^c}
\DeclareUnicodeCharacter{010A}{\dotaccent{C}}
\DeclareUnicodeCharacter{010B}{\dotaccent{c}}
\DeclareUnicodeCharacter{010C}{\v{C}}
\DeclareUnicodeCharacter{010D}{\v{c}}
\DeclareUnicodeCharacter{010E}{\v{D}}
\DeclareUnicodeCharacter{0112}{\=E}
\DeclareUnicodeCharacter{0113}{\=e}
\DeclareUnicodeCharacter{0114}{\u{E}}
\DeclareUnicodeCharacter{0115}{\u{e}}
\DeclareUnicodeCharacter{0116}{\dotaccent{E}}
\DeclareUnicodeCharacter{0117}{\dotaccent{e}}
\DeclareUnicodeCharacter{011A}{\v{E}}
\DeclareUnicodeCharacter{011B}{\v{e}}
\DeclareUnicodeCharacter{011C}{\^G}
\DeclareUnicodeCharacter{011D}{\^g}
\DeclareUnicodeCharacter{011E}{\u{G}}
\DeclareUnicodeCharacter{011F}{\u{g}}
\DeclareUnicodeCharacter{0120}{\dotaccent{G}}
\DeclareUnicodeCharacter{0121}{\dotaccent{g}}
\DeclareUnicodeCharacter{0124}{\^H}
\DeclareUnicodeCharacter{0125}{\^h}
\DeclareUnicodeCharacter{0128}{\~I}
\DeclareUnicodeCharacter{0129}{\~{\dotless{i}}}
\DeclareUnicodeCharacter{012A}{\=I}
\DeclareUnicodeCharacter{012B}{\={\dotless{i}}}
\DeclareUnicodeCharacter{012C}{\u{I}}
\DeclareUnicodeCharacter{012D}{\u{\dotless{i}}}
\DeclareUnicodeCharacter{0130}{\dotaccent{I}}
\DeclareUnicodeCharacter{0131}{\dotless{i}}
\DeclareUnicodeCharacter{0132}{IJ}
\DeclareUnicodeCharacter{0133}{ij}
\DeclareUnicodeCharacter{0134}{\^J}
\DeclareUnicodeCharacter{0135}{\^{\dotless{j}}}
\DeclareUnicodeCharacter{0139}{\'L}
\DeclareUnicodeCharacter{013A}{\'l}
\DeclareUnicodeCharacter{0141}{\L}
\DeclareUnicodeCharacter{0142}{\l}
\DeclareUnicodeCharacter{0143}{\'N}
\DeclareUnicodeCharacter{0144}{\'n}
\DeclareUnicodeCharacter{0147}{\v{N}}
\DeclareUnicodeCharacter{0148}{\v{n}}
\DeclareUnicodeCharacter{014C}{\=O}
\DeclareUnicodeCharacter{014D}{\=o}
\DeclareUnicodeCharacter{014E}{\u{O}}
\DeclareUnicodeCharacter{014F}{\u{o}}
\DeclareUnicodeCharacter{0150}{\H{O}}
\DeclareUnicodeCharacter{0151}{\H{o}}
\DeclareUnicodeCharacter{0152}{\OE}
\DeclareUnicodeCharacter{0153}{\oe}
\DeclareUnicodeCharacter{0154}{\'R}
\DeclareUnicodeCharacter{0155}{\'r}
\DeclareUnicodeCharacter{0158}{\v{R}}
\DeclareUnicodeCharacter{0159}{\v{r}}
\DeclareUnicodeCharacter{015A}{\'S}
\DeclareUnicodeCharacter{015B}{\'s}
\DeclareUnicodeCharacter{015C}{\^S}
\DeclareUnicodeCharacter{015D}{\^s}
\DeclareUnicodeCharacter{015E}{\cedilla{S}}
\DeclareUnicodeCharacter{015F}{\cedilla{s}}
\DeclareUnicodeCharacter{0160}{\v{S}}
\DeclareUnicodeCharacter{0161}{\v{s}}
\DeclareUnicodeCharacter{0162}{\cedilla{t}}
\DeclareUnicodeCharacter{0163}{\cedilla{T}}
\DeclareUnicodeCharacter{0164}{\v{T}}
\DeclareUnicodeCharacter{0168}{\~U}
\DeclareUnicodeCharacter{0169}{\~u}
\DeclareUnicodeCharacter{016A}{\=U}
\DeclareUnicodeCharacter{016B}{\=u}
\DeclareUnicodeCharacter{016C}{\u{U}}
\DeclareUnicodeCharacter{016D}{\u{u}}
\DeclareUnicodeCharacter{016E}{\ringaccent{U}}
\DeclareUnicodeCharacter{016F}{\ringaccent{u}}
\DeclareUnicodeCharacter{0170}{\H{U}}
\DeclareUnicodeCharacter{0171}{\H{u}}
\DeclareUnicodeCharacter{0174}{\^W}
\DeclareUnicodeCharacter{0175}{\^w}
\DeclareUnicodeCharacter{0176}{\^Y}
\DeclareUnicodeCharacter{0177}{\^y}
\DeclareUnicodeCharacter{0178}{\"Y}
\DeclareUnicodeCharacter{0179}{\'Z}
\DeclareUnicodeCharacter{017A}{\'z}
\DeclareUnicodeCharacter{017B}{\dotaccent{Z}}
\DeclareUnicodeCharacter{017C}{\dotaccent{z}}
\DeclareUnicodeCharacter{017D}{\v{Z}}
\DeclareUnicodeCharacter{017E}{\v{z}}
\DeclareUnicodeCharacter{01C4}{D\v{Z}}
\DeclareUnicodeCharacter{01C5}{D\v{z}}
\DeclareUnicodeCharacter{01C6}{d\v{z}}
\DeclareUnicodeCharacter{01C7}{LJ}
\DeclareUnicodeCharacter{01C8}{Lj}
\DeclareUnicodeCharacter{01C9}{lj}
\DeclareUnicodeCharacter{01CA}{NJ}
\DeclareUnicodeCharacter{01CB}{Nj}
\DeclareUnicodeCharacter{01CC}{nj}
\DeclareUnicodeCharacter{01CD}{\v{A}}
\DeclareUnicodeCharacter{01CE}{\v{a}}
\DeclareUnicodeCharacter{01CF}{\v{I}}
\DeclareUnicodeCharacter{01D0}{\v{\dotless{i}}}
\DeclareUnicodeCharacter{01D1}{\v{O}}
\DeclareUnicodeCharacter{01D2}{\v{o}}
\DeclareUnicodeCharacter{01D3}{\v{U}}
\DeclareUnicodeCharacter{01D4}{\v{u}}
\DeclareUnicodeCharacter{01E2}{\={\AE}}
\DeclareUnicodeCharacter{01E3}{\={\ae}}
\DeclareUnicodeCharacter{01E6}{\v{G}}
\DeclareUnicodeCharacter{01E7}{\v{g}}
\DeclareUnicodeCharacter{01E8}{\v{K}}
\DeclareUnicodeCharacter{01E9}{\v{k}}
\DeclareUnicodeCharacter{01F0}{\v{\dotless{j}}}
\DeclareUnicodeCharacter{01F1}{DZ}
\DeclareUnicodeCharacter{01F2}{Dz}
\DeclareUnicodeCharacter{01F3}{dz}
\DeclareUnicodeCharacter{01F4}{\'G}
\DeclareUnicodeCharacter{01F5}{\'g}
\DeclareUnicodeCharacter{01F8}{\`N}
\DeclareUnicodeCharacter{01F9}{\`n}
\DeclareUnicodeCharacter{01FC}{\'{\AE}}
\DeclareUnicodeCharacter{01FD}{\'{\ae}}
\DeclareUnicodeCharacter{01FE}{\'{\O}}
\DeclareUnicodeCharacter{01FF}{\'{\o}}
\DeclareUnicodeCharacter{021E}{\v{H}}
\DeclareUnicodeCharacter{021F}{\v{h}}
\DeclareUnicodeCharacter{0226}{\dotaccent{A}}
\DeclareUnicodeCharacter{0227}{\dotaccent{a}}
\DeclareUnicodeCharacter{0228}{\cedilla{E}}
\DeclareUnicodeCharacter{0229}{\cedilla{e}}
\DeclareUnicodeCharacter{022E}{\dotaccent{O}}
\DeclareUnicodeCharacter{022F}{\dotaccent{o}}
\DeclareUnicodeCharacter{0232}{\=Y}
\DeclareUnicodeCharacter{0233}{\=y}
\DeclareUnicodeCharacter{0237}{\dotless{j}}
\DeclareUnicodeCharacter{1E02}{\dotaccent{B}}
\DeclareUnicodeCharacter{1E03}{\dotaccent{b}}
\DeclareUnicodeCharacter{1E04}{\udotaccent{B}}
\DeclareUnicodeCharacter{1E05}{\udotaccent{b}}
\DeclareUnicodeCharacter{1E06}{\ubaraccent{B}}
\DeclareUnicodeCharacter{1E07}{\ubaraccent{b}}
\DeclareUnicodeCharacter{1E0A}{\dotaccent{D}}
\DeclareUnicodeCharacter{1E0B}{\dotaccent{d}}
\DeclareUnicodeCharacter{1E0C}{\udotaccent{D}}
\DeclareUnicodeCharacter{1E0D}{\udotaccent{d}}
\DeclareUnicodeCharacter{1E0E}{\ubaraccent{D}}
\DeclareUnicodeCharacter{1E0F}{\ubaraccent{d}}
\DeclareUnicodeCharacter{1E1E}{\dotaccent{F}}
\DeclareUnicodeCharacter{1E1F}{\dotaccent{f}}
\DeclareUnicodeCharacter{1E20}{\=G}
\DeclareUnicodeCharacter{1E21}{\=g}
\DeclareUnicodeCharacter{1E22}{\dotaccent{H}}
\DeclareUnicodeCharacter{1E23}{\dotaccent{h}}
\DeclareUnicodeCharacter{1E24}{\udotaccent{H}}
\DeclareUnicodeCharacter{1E25}{\udotaccent{h}}
\DeclareUnicodeCharacter{1E26}{\"H}
\DeclareUnicodeCharacter{1E27}{\"h}
\DeclareUnicodeCharacter{1E30}{\'K}
\DeclareUnicodeCharacter{1E31}{\'k}
\DeclareUnicodeCharacter{1E32}{\udotaccent{K}}
\DeclareUnicodeCharacter{1E33}{\udotaccent{k}}
\DeclareUnicodeCharacter{1E34}{\ubaraccent{K}}
\DeclareUnicodeCharacter{1E35}{\ubaraccent{k}}
\DeclareUnicodeCharacter{1E36}{\udotaccent{L}}
\DeclareUnicodeCharacter{1E37}{\udotaccent{l}}
\DeclareUnicodeCharacter{1E3A}{\ubaraccent{L}}
\DeclareUnicodeCharacter{1E3B}{\ubaraccent{l}}
\DeclareUnicodeCharacter{1E3E}{\'M}
\DeclareUnicodeCharacter{1E3F}{\'m}
\DeclareUnicodeCharacter{1E40}{\dotaccent{M}}
\DeclareUnicodeCharacter{1E41}{\dotaccent{m}}
\DeclareUnicodeCharacter{1E42}{\udotaccent{M}}
\DeclareUnicodeCharacter{1E43}{\udotaccent{m}}
\DeclareUnicodeCharacter{1E44}{\dotaccent{N}}
\DeclareUnicodeCharacter{1E45}{\dotaccent{n}}
\DeclareUnicodeCharacter{1E46}{\udotaccent{N}}
\DeclareUnicodeCharacter{1E47}{\udotaccent{n}}
\DeclareUnicodeCharacter{1E48}{\ubaraccent{N}}
\DeclareUnicodeCharacter{1E49}{\ubaraccent{n}}
\DeclareUnicodeCharacter{1E54}{\'P}
\DeclareUnicodeCharacter{1E55}{\'p}
\DeclareUnicodeCharacter{1E56}{\dotaccent{P}}
\DeclareUnicodeCharacter{1E57}{\dotaccent{p}}
\DeclareUnicodeCharacter{1E58}{\dotaccent{R}}
\DeclareUnicodeCharacter{1E59}{\dotaccent{r}}
\DeclareUnicodeCharacter{1E5A}{\udotaccent{R}}
\DeclareUnicodeCharacter{1E5B}{\udotaccent{r}}
\DeclareUnicodeCharacter{1E5E}{\ubaraccent{R}}
\DeclareUnicodeCharacter{1E5F}{\ubaraccent{r}}
\DeclareUnicodeCharacter{1E60}{\dotaccent{S}}
\DeclareUnicodeCharacter{1E61}{\dotaccent{s}}
\DeclareUnicodeCharacter{1E62}{\udotaccent{S}}
\DeclareUnicodeCharacter{1E63}{\udotaccent{s}}
\DeclareUnicodeCharacter{1E6A}{\dotaccent{T}}
\DeclareUnicodeCharacter{1E6B}{\dotaccent{t}}
\DeclareUnicodeCharacter{1E6C}{\udotaccent{T}}
\DeclareUnicodeCharacter{1E6D}{\udotaccent{t}}
\DeclareUnicodeCharacter{1E6E}{\ubaraccent{T}}
\DeclareUnicodeCharacter{1E6F}{\ubaraccent{t}}
\DeclareUnicodeCharacter{1E7C}{\~V}
\DeclareUnicodeCharacter{1E7D}{\~v}
\DeclareUnicodeCharacter{1E7E}{\udotaccent{V}}
\DeclareUnicodeCharacter{1E7F}{\udotaccent{v}}
\DeclareUnicodeCharacter{1E80}{\`W}
\DeclareUnicodeCharacter{1E81}{\`w}
\DeclareUnicodeCharacter{1E82}{\'W}
\DeclareUnicodeCharacter{1E83}{\'w}
\DeclareUnicodeCharacter{1E84}{\"W}
\DeclareUnicodeCharacter{1E85}{\"w}
\DeclareUnicodeCharacter{1E86}{\dotaccent{W}}
\DeclareUnicodeCharacter{1E87}{\dotaccent{w}}
\DeclareUnicodeCharacter{1E88}{\udotaccent{W}}
\DeclareUnicodeCharacter{1E89}{\udotaccent{w}}
\DeclareUnicodeCharacter{1E8A}{\dotaccent{X}}
\DeclareUnicodeCharacter{1E8B}{\dotaccent{x}}
\DeclareUnicodeCharacter{1E8C}{\"X}
\DeclareUnicodeCharacter{1E8D}{\"x}
\DeclareUnicodeCharacter{1E8E}{\dotaccent{Y}}
\DeclareUnicodeCharacter{1E8F}{\dotaccent{y}}
\DeclareUnicodeCharacter{1E90}{\^Z}
\DeclareUnicodeCharacter{1E91}{\^z}
\DeclareUnicodeCharacter{1E92}{\udotaccent{Z}}
\DeclareUnicodeCharacter{1E93}{\udotaccent{z}}
\DeclareUnicodeCharacter{1E94}{\ubaraccent{Z}}
\DeclareUnicodeCharacter{1E95}{\ubaraccent{z}}
\DeclareUnicodeCharacter{1E96}{\ubaraccent{h}}
\DeclareUnicodeCharacter{1E97}{\"t}
\DeclareUnicodeCharacter{1E98}{\ringaccent{w}}
\DeclareUnicodeCharacter{1E99}{\ringaccent{y}}
\DeclareUnicodeCharacter{1EA0}{\udotaccent{A}}
\DeclareUnicodeCharacter{1EA1}{\udotaccent{a}}
\DeclareUnicodeCharacter{1EB8}{\udotaccent{E}}
\DeclareUnicodeCharacter{1EB9}{\udotaccent{e}}
\DeclareUnicodeCharacter{1EBC}{\~E}
\DeclareUnicodeCharacter{1EBD}{\~e}
\DeclareUnicodeCharacter{1ECA}{\udotaccent{I}}
\DeclareUnicodeCharacter{1ECB}{\udotaccent{i}}
\DeclareUnicodeCharacter{1ECC}{\udotaccent{O}}
\DeclareUnicodeCharacter{1ECD}{\udotaccent{o}}
\DeclareUnicodeCharacter{1EE4}{\udotaccent{U}}
\DeclareUnicodeCharacter{1EE5}{\udotaccent{u}}
\DeclareUnicodeCharacter{1EF2}{\`Y}
\DeclareUnicodeCharacter{1EF3}{\`y}
\DeclareUnicodeCharacter{1EF4}{\udotaccent{Y}}
\DeclareUnicodeCharacter{1EF8}{\~Y}
\DeclareUnicodeCharacter{1EF9}{\~y}
\DeclareUnicodeCharacter{2013}{--}
\DeclareUnicodeCharacter{2014}{---}
\DeclareUnicodeCharacter{2022}{\bullet}
\DeclareUnicodeCharacter{2026}{\dots}
\DeclareUnicodeCharacter{20AC}{\euro}
\DeclareUnicodeCharacter{2192}{\expansion}
\DeclareUnicodeCharacter{21D2}{\result}
\DeclareUnicodeCharacter{2212}{\minus}
\DeclareUnicodeCharacter{2217}{\point}
\DeclareUnicodeCharacter{2261}{\equiv}
}% end of \utfeightchardefs
% US-ASCII character definitions.
\def\asciichardefs{% nothing need be done
\relax
}
% Make non-ASCII characters printable again for compatibility with
% existing Texinfo documents that may use them, even without declaring a
% document encoding.
%
\setnonasciicharscatcode \other
\message{formatting,}
\newdimen\defaultparindent \defaultparindent = 15pt
\chapheadingskip = 15pt plus 4pt minus 2pt
\secheadingskip = 12pt plus 3pt minus 2pt
\subsecheadingskip = 9pt plus 2pt minus 2pt
% Prevent underfull vbox error messages.
\vbadness = 10000
% Don't be so finicky about underfull hboxes, either.
\hbadness = 2000
% Following George Bush, just get rid of widows and orphans.
\widowpenalty=10000
\clubpenalty=10000
% Use TeX 3.0's \emergencystretch to help line breaking, but if we're
% using an old version of TeX, don't do anything. We want the amount of
% stretch added to depend on the line length, hence the dependence on
% \hsize. We call this whenever the paper size is set.
%
\def\setemergencystretch{%
\ifx\emergencystretch\thisisundefined
% Allow us to assign to \emergencystretch anyway.
\def\emergencystretch{\dimen0}%
\else
\emergencystretch = .15\hsize
\fi
}
% Parameters in order: 1) textheight; 2) textwidth;
% 3) voffset; 4) hoffset; 5) binding offset; 6) topskip;
% 7) physical page height; 8) physical page width.
%
% We also call \setleading{\textleading}, so the caller should define
% \textleading. The caller should also set \parskip.
%
\def\internalpagesizes#1#2#3#4#5#6#7#8{%
\voffset = #3\relax
\topskip = #6\relax
\splittopskip = \topskip
%
\vsize = #1\relax
\advance\vsize by \topskip
\outervsize = \vsize
\advance\outervsize by 2\topandbottommargin
\pageheight = \vsize
%
\hsize = #2\relax
\outerhsize = \hsize
\advance\outerhsize by 0.5in
\pagewidth = \hsize
%
\normaloffset = #4\relax
\bindingoffset = #5\relax
%
\ifpdf
\pdfpageheight #7\relax
\pdfpagewidth #8\relax
\fi
%
\setleading{\textleading}
%
\parindent = \defaultparindent
\setemergencystretch
}
% @letterpaper (the default).
\def\letterpaper{{\globaldefs = 1
\parskip = 3pt plus 2pt minus 1pt
\textleading = 13.2pt
%
% If page is nothing but text, make it come out even.
\internalpagesizes{46\baselineskip}{6in}%
{\voffset}{.25in}%
{\bindingoffset}{36pt}%
{11in}{8.5in}%
}}
% Use @smallbook to reset parameters for 7x9.25 trim size.
\def\smallbook{{\globaldefs = 1
\parskip = 2pt plus 1pt
\textleading = 12pt
%
\internalpagesizes{7.5in}{5in}%
{\voffset}{.25in}%
{\bindingoffset}{16pt}%
{9.25in}{7in}%
%
\lispnarrowing = 0.3in
\tolerance = 700
\hfuzz = 1pt
\contentsrightmargin = 0pt
\defbodyindent = .5cm
}}
% Use @smallerbook to reset parameters for 6x9 trim size.
% (Just testing, parameters still in flux.)
\def\smallerbook{{\globaldefs = 1
\parskip = 1.5pt plus 1pt
\textleading = 12pt
%
\internalpagesizes{7.4in}{4.8in}%
{-.2in}{-.4in}%
{0pt}{14pt}%
{9in}{6in}%
%
\lispnarrowing = 0.25in
\tolerance = 700
\hfuzz = 1pt
\contentsrightmargin = 0pt
\defbodyindent = .4cm
}}
% Use @afourpaper to print on European A4 paper.
\def\afourpaper{{\globaldefs = 1
\parskip = 3pt plus 2pt minus 1pt
\textleading = 13.2pt
%
% Double-side printing via postscript on Laserjet 4050
% prints double-sided nicely when \bindingoffset=10mm and \hoffset=-6mm.
% To change the settings for a different printer or situation, adjust
% \normaloffset until the front-side and back-side texts align. Then
% do the same for \bindingoffset. You can set these for testing in
% your texinfo source file like this:
% @tex
% \global\normaloffset = -6mm
% \global\bindingoffset = 10mm
% @end tex
\internalpagesizes{51\baselineskip}{160mm}
{\voffset}{\hoffset}%
{\bindingoffset}{44pt}%
{297mm}{210mm}%
%
\tolerance = 700
\hfuzz = 1pt
\contentsrightmargin = 0pt
\defbodyindent = 5mm
}}
% Use @afivepaper to print on European A5 paper.
% From romildo@urano.iceb.ufop.br, 2 July 2000.
% He also recommends making @example and @lisp be small.
\def\afivepaper{{\globaldefs = 1
\parskip = 2pt plus 1pt minus 0.1pt
\textleading = 12.5pt
%
\internalpagesizes{160mm}{120mm}%
{\voffset}{\hoffset}%
{\bindingoffset}{8pt}%
{210mm}{148mm}%
%
\lispnarrowing = 0.2in
\tolerance = 800
\hfuzz = 1.2pt
\contentsrightmargin = 0pt
\defbodyindent = 2mm
\tableindent = 12mm
}}
% A specific text layout, 24x15cm overall, intended for A4 paper.
\def\afourlatex{{\globaldefs = 1
\afourpaper
\internalpagesizes{237mm}{150mm}%
{\voffset}{4.6mm}%
{\bindingoffset}{7mm}%
{297mm}{210mm}%
%
% Must explicitly reset to 0 because we call \afourpaper.
\globaldefs = 0
}}
% Use @afourwide to print on A4 paper in landscape format.
\def\afourwide{{\globaldefs = 1
\afourpaper
\internalpagesizes{241mm}{165mm}%
{\voffset}{-2.95mm}%
{\bindingoffset}{7mm}%
{297mm}{210mm}%
\globaldefs = 0
}}
% @pagesizes TEXTHEIGHT[,TEXTWIDTH]
% Perhaps we should allow setting the margins, \topskip, \parskip,
% and/or leading, also. Or perhaps we should compute them somehow.
%
\parseargdef\pagesizes{\pagesizesyyy #1,,\finish}
\def\pagesizesyyy#1,#2,#3\finish{{%
\setbox0 = \hbox{\ignorespaces #2}\ifdim\wd0 > 0pt \hsize=#2\relax \fi
\globaldefs = 1
%
\parskip = 3pt plus 2pt minus 1pt
\setleading{\textleading}%
%
\dimen0 = #1
\advance\dimen0 by \voffset
%
\dimen2 = \hsize
\advance\dimen2 by \normaloffset
%
\internalpagesizes{#1}{\hsize}%
{\voffset}{\normaloffset}%
{\bindingoffset}{44pt}%
{\dimen0}{\dimen2}%
}}
% Set default to letter.
%
\letterpaper
\message{and turning on texinfo input format.}
% Define macros to output various characters with catcode for normal text.
\catcode`\"=\other
\catcode`\~=\other
\catcode`\^=\other
\catcode`\_=\other
\catcode`\|=\other
\catcode`\<=\other
\catcode`\>=\other
\catcode`\+=\other
\catcode`\$=\other
\def\normaldoublequote{"}
\def\normaltilde{~}
\def\normalcaret{^}
\def\normalunderscore{_}
\def\normalverticalbar{|}
\def\normalless{<}
\def\normalgreater{>}
\def\normalplus{+}
\def\normaldollar{$}%$ font-lock fix
% This macro is used to make a character print one way in \tt
% (where it can probably be output as-is), and another way in other fonts,
% where something hairier probably needs to be done.
%
% #1 is what to print if we are indeed using \tt; #2 is what to print
% otherwise. Since all the Computer Modern typewriter fonts have zero
% interword stretch (and shrink), and it is reasonable to expect all
% typewriter fonts to have this, we can check that font parameter.
%
\def\ifusingtt#1#2{\ifdim \fontdimen3\font=0pt #1\else #2\fi}
% Same as above, but check for italic font. Actually this also catches
% non-italic slanted fonts since it is impossible to distinguish them from
% italic fonts. But since this is only used by $ and it uses \sl anyway
% this is not a problem.
\def\ifusingit#1#2{\ifdim \fontdimen1\font>0pt #1\else #2\fi}
% Turn off all special characters except @
% (and those which the user can use as if they were ordinary).
% Most of these we simply print from the \tt font, but for some, we can
% use math or other variants that look better in normal text.
\catcode`\"=\active
\def\activedoublequote{{\tt\char34}}
\let"=\activedoublequote
\catcode`\~=\active
\def~{{\tt\char126}}
\chardef\hat=`\^
\catcode`\^=\active
\def^{{\tt \hat}}
\catcode`\_=\active
\def_{\ifusingtt\normalunderscore\_}
\let\realunder=_
% Subroutine for the previous macro.
\def\_{\leavevmode \kern.07em \vbox{\hrule width.3em height.1ex}\kern .07em }
\catcode`\|=\active
\def|{{\tt\char124}}
\chardef \less=`\<
\catcode`\<=\active
\def<{{\tt \less}}
\chardef \gtr=`\>
\catcode`\>=\active
\def>{{\tt \gtr}}
\catcode`\+=\active
\def+{{\tt \char 43}}
\catcode`\$=\active
\def${\ifusingit{{\sl\$}}\normaldollar}%$ font-lock fix
% If a .fmt file is being used, characters that might appear in a file
% name cannot be active until we have parsed the command line.
% So turn them off again, and have \everyjob (or @setfilename) turn them on.
% \otherifyactive is called near the end of this file.
\def\otherifyactive{\catcode`+=\other \catcode`\_=\other}
% Used sometimes to turn off (effectively) the active characters even after
% parsing them.
\def\turnoffactive{%
\normalturnoffactive
\otherbackslash
}
\catcode`\@=0
% \backslashcurfont outputs one backslash character in current font,
% as in \char`\\.
\global\chardef\backslashcurfont=`\\
\global\let\rawbackslashxx=\backslashcurfont % let existing .??s files work
% \realbackslash is an actual character `\' with catcode other, and
% \doublebackslash is two of them (for the pdf outlines).
{\catcode`\\=\other @gdef@realbackslash{\} @gdef@doublebackslash{\\}}
% In texinfo, backslash is an active character; it prints the backslash
% in fixed width font.
\catcode`\\=\active
@def@normalbackslash{{@tt@backslashcurfont}}
% On startup, @fixbackslash assigns:
% @let \ = @normalbackslash
% \rawbackslash defines an active \ to do \backslashcurfont.
% \otherbackslash defines an active \ to be a literal `\' character with
% catcode other.
@gdef@rawbackslash{@let\=@backslashcurfont}
@gdef@otherbackslash{@let\=@realbackslash}
% Same as @turnoffactive except outputs \ as {\tt\char`\\} instead of
% the literal character `\'.
%
@def@normalturnoffactive{%
@let\=@normalbackslash
@let"=@normaldoublequote
@let~=@normaltilde
@let^=@normalcaret
@let_=@normalunderscore
@let|=@normalverticalbar
@let<=@normalless
@let>=@normalgreater
@let+=@normalplus
@let$=@normaldollar %$ font-lock fix
@unsepspaces
}
% Make _ and + \other characters, temporarily.
% This is canceled by @fixbackslash.
@otherifyactive
% If a .fmt file is being used, we don't want the `\input texinfo' to show up.
% That is what \eatinput is for; after that, the `\' should revert to printing
% a backslash.
%
@gdef@eatinput input texinfo{@fixbackslash}
@global@let\ = @eatinput
% On the other hand, perhaps the file did not have a `\input texinfo'. Then
% the first `\' in the file would cause an error. This macro tries to fix
% that, assuming it is called before the first `\' could plausibly occur.
% Also turn back on active characters that might appear in the input
% file name, in case not using a pre-dumped format.
%
@gdef@fixbackslash{%
@ifx\@eatinput @let\ = @normalbackslash @fi
@catcode`+=@active
@catcode`@_=@active
}
% Say @foo, not \foo, in error messages.
@escapechar = `@@
% These look ok in all fonts, so just make them not special.
@catcode`@& = @other
@catcode`@# = @other
@catcode`@% = @other
@c Local variables:
@c eval: (add-hook 'write-file-hooks 'time-stamp)
@c page-delimiter: "^\\\\message"
@c time-stamp-start: "def\\\\texinfoversion{"
@c time-stamp-format: "%:y-%02m-%02d.%02H"
@c time-stamp-end: "}"
@c End:
@c vim:sw=2:
@ignore
arch-tag: e1b36e32-c96e-4135-a41a-0b2efa2ea115
@end ignore
diff --git a/gnome3/pinentry-gnome3.c b/gnome3/pinentry-gnome3.c
index d6d7d16..90ff60f 100644
--- a/gnome3/pinentry-gnome3.c
+++ b/gnome3/pinentry-gnome3.c
@@ -1,281 +1,281 @@
/* pinentry-gnome3.c
Copyright (C) 2015 g10 Code GmbH
pinentry-gnome-3 is a pinentry application for GNOME 3. It tries
to follow the Gnome Human Interface Guide as close as possible.
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 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., 675 Mass Ave, Cambridge, MA 02139, USA. */
#ifdef HAVE_CONFIG_H
# include "config.h"
#endif
#include <gcr/gcr-base.h>
#include <string.h>
#include <stdlib.h>
#include <assuan.h>
#include "memory.h"
#include "pinentry.h"
#ifdef FALLBACK_CURSES
#include "pinentry-curses.h"
#endif
#define PGMNAME "pinentry-gnome3"
#ifndef VERSION
# define VERSION
#endif
static gchar *
pinentry_utf8_validate (gchar *text)
{
gchar *result;
if (!text)
return NULL;
if (g_utf8_validate (text, -1, NULL))
return g_strdup (text);
/* Failure: Assume that it was encoded in the current locale and
convert it to utf-8. */
result = g_locale_to_utf8 (text, -1, NULL, NULL, NULL);
if (!result)
{
gchar *p;
result = p = g_strdup (text);
while (!g_utf8_validate (p, -1, (const gchar **) &p))
*p = '?';
}
return result;
}
static GcrPrompt *
create_prompt (pinentry_t pe, int confirm)
{
GcrPrompt *prompt;
GError *error = NULL;
char *msg;
/* Create the prompt. */
prompt = GCR_PROMPT (gcr_system_prompt_open (-1, NULL, &error));
if (! prompt)
{
fprintf (stderr, "couldn't create prompt for gnupg passphrase: %s\n",
error->message);
pe->specific_err_loc = "gcr_prompt";
pe->specific_err_info = strdup (error->message);
pe->specific_err = gpg_error (GPG_ERR_CONFIGURATION);
g_error_free (error);
return NULL;
}
/* Set the messages for the various buttons, etc. */
if (pe->title)
{
msg = pinentry_utf8_validate (pe->title);
gcr_prompt_set_title (prompt, msg);
g_free (msg);
}
if (pe->description)
{
msg = pinentry_utf8_validate (pe->description);
gcr_prompt_set_description (prompt, msg);
g_free (msg);
}
- /* An error occured during the last prompt. */
+ /* An error occurred during the last prompt. */
if (pe->error)
{
msg = pinentry_utf8_validate (pe->error);
gcr_prompt_set_warning (prompt, msg);
g_free (msg);
}
if (! pe->prompt && confirm)
gcr_prompt_set_message (prompt, "Message");
else if (! pe->prompt && ! confirm)
gcr_prompt_set_message (prompt, "Enter Passphrase");
else
{
msg = pinentry_utf8_validate (pe->prompt);
gcr_prompt_set_message (prompt, msg);
g_free (msg);
}
if (! confirm)
gcr_prompt_set_password_new (prompt, !!pe->repeat_passphrase);
if (pe->ok || pe->default_ok)
{
msg = pinentry_utf8_validate (pe->ok ?: pe->default_ok);
gcr_prompt_set_continue_label (prompt, msg);
g_free (msg);
}
/* XXX: Disable this button if pe->one_button is set. */
if (pe->cancel || pe->default_cancel)
{
msg = pinentry_utf8_validate (pe->cancel ?: pe->default_cancel);
gcr_prompt_set_cancel_label (prompt, msg);
g_free (msg);
}
if (confirm && pe->notok)
{
/* XXX: Add support for the third option. */
}
/* XXX: gcr expects a string; we have a int. */
// gcr_prompt_set_caller_window (prompt, pe->parent_wid);
#ifdef HAVE_LIBSECRET
if (! confirm && pe->allow_external_password_cache && pe->keyinfo)
{
if (pe->default_pwmngr)
{
msg = pinentry_utf8_validate (pe->default_pwmngr);
gcr_prompt_set_choice_label (prompt, msg);
g_free (msg);
}
else
gcr_prompt_set_choice_label
(prompt, "Automatically unlock this key, whenever I'm logged in");
}
#endif
return prompt;
}
static int
gnome3_cmd_handler (pinentry_t pe)
{
GcrPrompt *prompt = NULL;
GError *error = NULL;
int ret = -1;
if (pe->pin)
/* Passphrase mode. */
{
const char *password;
prompt = create_prompt (pe, 0);
if (! prompt)
/* Something went wrong. */
{
pe->canceled = 1;
return -1;
}
/* "The returned password is valid until the next time a method
is called to display another prompt." */
password = gcr_prompt_password (prompt, NULL, &error);
if (error)
/* Error. */
{
pe->specific_err = gpg_error (GPG_ERR_ASS_GENERAL);
g_error_free (error);
ret = -1;
}
else if (! password && ! error)
/* User cancelled the operation. */
ret = -1;
else
{
pinentry_setbufferlen (pe, strlen (password) + 1);
if (pe->pin)
strcpy (pe->pin, password);
if (pe->repeat_passphrase)
pe->repeat_okay = 1;
#ifdef HAVE_LIBSECRET
if (pe->allow_external_password_cache && pe->keyinfo)
pe->may_cache_password = gcr_prompt_get_choice_chosen (prompt);
#endif
ret = 1;
}
}
else
/* Message box mode. */
{
GcrPromptReply reply;
prompt = create_prompt (pe, 1);
if (! prompt)
/* Something went wrong. */
{
pe->canceled = 1;
return -1;
}
/* XXX: We don't support a third button! */
reply = gcr_prompt_confirm_run (prompt, NULL, &error);
if (error)
{
pe->specific_err = gpg_error (GPG_ERR_ASS_GENERAL);
ret = 0;
}
else if (reply == GCR_PROMPT_REPLY_CONTINUE
/* XXX: Hack since gcr doesn't yet support one button
message boxes treat cancel the same as okay. */
|| pe->one_button)
/* Confirmation. */
ret = 1;
else
/* GCR_PROMPT_REPLY_CANCEL */
{
pe->canceled = 1;
ret = 0;
}
}
if (prompt)
g_clear_object (&prompt);
return ret;
}
pinentry_cmd_handler_t pinentry_cmd_handler = gnome3_cmd_handler;
int
main (int argc, char *argv[])
{
pinentry_init (PGMNAME);
#ifdef FALLBACK_CURSES
if (!getenv ("DBUS_SESSION_BUS_ADDRESS"))
{
fprintf (stderr, "No $DBUS_SESSION_BUS_ADDRESS found,"
" falling back to curses\n");
pinentry_cmd_handler = curses_cmd_handler;
}
#endif
pinentry_parse_opts (argc, argv);
if (pinentry_loop ())
return 1;
return 0;
}
diff --git a/m4/glib.m4 b/m4/glib.m4
index b3c632b..7338acb 100644
--- a/m4/glib.m4
+++ b/m4/glib.m4
@@ -1,196 +1,196 @@
# Configure paths for GLIB
# Owen Taylor 97-11-3
dnl AM_PATH_GLIB([MINIMUM-VERSION, [ACTION-IF-FOUND [, ACTION-IF-NOT-FOUND [, MODULES]]]])
dnl Test for GLIB, and define GLIB_CFLAGS and GLIB_LIBS, if "gmodule" or
dnl gthread is specified in MODULES, pass to glib-config
dnl
AC_DEFUN([AM_PATH_GLIB],
[dnl
dnl Get the cflags and libraries from the glib-config script
dnl
AC_ARG_WITH(glib-prefix,[ --with-glib-prefix=PFX Prefix where GLIB is installed (optional)],
glib_config_prefix="$withval", glib_config_prefix="")
AC_ARG_WITH(glib-exec-prefix,[ --with-glib-exec-prefix=PFX Exec prefix where GLIB is installed (optional)],
glib_config_exec_prefix="$withval", glib_config_exec_prefix="")
AC_ARG_ENABLE(glibtest, [ --disable-glibtest Do not try to compile and run a test GLIB program],
, enable_glibtest=yes)
if test x$glib_config_exec_prefix != x ; then
glib_config_args="$glib_config_args --exec-prefix=$glib_config_exec_prefix"
if test x${GLIB_CONFIG+set} != xset ; then
GLIB_CONFIG=$glib_config_exec_prefix/bin/glib-config
fi
fi
if test x$glib_config_prefix != x ; then
glib_config_args="$glib_config_args --prefix=$glib_config_prefix"
if test x${GLIB_CONFIG+set} != xset ; then
GLIB_CONFIG=$glib_config_prefix/bin/glib-config
fi
fi
for module in . $4
do
case "$module" in
gmodule)
glib_config_args="$glib_config_args gmodule"
;;
gthread)
glib_config_args="$glib_config_args gthread"
;;
esac
done
AC_PATH_PROG(GLIB_CONFIG, glib-config, no)
min_glib_version=ifelse([$1], ,0.99.7,$1)
AC_MSG_CHECKING(for GLIB - version >= $min_glib_version)
no_glib=""
if test "$GLIB_CONFIG" = "no" ; then
no_glib=yes
else
GLIB_CFLAGS=`$GLIB_CONFIG $glib_config_args --cflags`
GLIB_LIBS=`$GLIB_CONFIG $glib_config_args --libs`
glib_config_major_version=`$GLIB_CONFIG $glib_config_args --version | \
sed 's/\([[0-9]]*\).\([[0-9]]*\).\([[0-9]]*\)/\1/'`
glib_config_minor_version=`$GLIB_CONFIG $glib_config_args --version | \
sed 's/\([[0-9]]*\).\([[0-9]]*\).\([[0-9]]*\)/\2/'`
glib_config_micro_version=`$GLIB_CONFIG $glib_config_args --version | \
sed 's/\([[0-9]]*\).\([[0-9]]*\).\([[0-9]]*\)/\3/'`
if test "x$enable_glibtest" = "xyes" ; then
ac_save_CFLAGS="$CFLAGS"
ac_save_LIBS="$LIBS"
CFLAGS="$CFLAGS $GLIB_CFLAGS"
LIBS="$GLIB_LIBS $LIBS"
dnl
dnl Now check if the installed GLIB is sufficiently new. (Also sanity
dnl checks the results of glib-config to some extent
dnl
rm -f conf.glibtest
AC_TRY_RUN([
#include <glib.h>
#include <stdio.h>
#include <stdlib.h>
int
main ()
{
int major, minor, micro;
char *tmp_version;
system ("touch conf.glibtest");
/* HP/UX 9 (%@#!) writes to sscanf strings */
tmp_version = g_strdup("$min_glib_version");
if (sscanf(tmp_version, "%d.%d.%d", &major, &minor, &micro) != 3) {
printf("%s, bad version string\n", "$min_glib_version");
exit(1);
}
if ((glib_major_version != $glib_config_major_version) ||
(glib_minor_version != $glib_config_minor_version) ||
(glib_micro_version != $glib_config_micro_version))
{
printf("\n*** 'glib-config --version' returned %d.%d.%d, but GLIB (%d.%d.%d)\n",
$glib_config_major_version, $glib_config_minor_version, $glib_config_micro_version,
glib_major_version, glib_minor_version, glib_micro_version);
printf ("*** was found! If glib-config was correct, then it is best\n");
printf ("*** to remove the old version of GLIB. You may also be able to fix the error\n");
- printf("*** by modifying your LD_LIBRARY_PATH enviroment variable, or by editing\n");
+ printf("*** by modifying your LD_LIBRARY_PATH environment variable, or by editing\n");
printf("*** /etc/ld.so.conf. Make sure you have run ldconfig if that is\n");
printf("*** required on your system.\n");
printf("*** If glib-config was wrong, set the environment variable GLIB_CONFIG\n");
printf("*** to point to the correct copy of glib-config, and remove the file config.cache\n");
printf("*** before re-running configure\n");
}
else if ((glib_major_version != GLIB_MAJOR_VERSION) ||
(glib_minor_version != GLIB_MINOR_VERSION) ||
(glib_micro_version != GLIB_MICRO_VERSION))
{
printf("*** GLIB header files (version %d.%d.%d) do not match\n",
GLIB_MAJOR_VERSION, GLIB_MINOR_VERSION, GLIB_MICRO_VERSION);
printf("*** library (version %d.%d.%d)\n",
glib_major_version, glib_minor_version, glib_micro_version);
}
else
{
if ((glib_major_version > major) ||
((glib_major_version == major) && (glib_minor_version > minor)) ||
((glib_major_version == major) && (glib_minor_version == minor) && (glib_micro_version >= micro)))
{
return 0;
}
else
{
printf("\n*** An old version of GLIB (%d.%d.%d) was found.\n",
glib_major_version, glib_minor_version, glib_micro_version);
printf("*** You need a version of GLIB newer than %d.%d.%d. The latest version of\n",
major, minor, micro);
printf("*** GLIB is always available from ftp://ftp.gtk.org.\n");
printf("***\n");
printf("*** If you have already installed a sufficiently new version, this error\n");
printf("*** probably means that the wrong copy of the glib-config shell script is\n");
printf("*** being found. The easiest way to fix this is to remove the old version\n");
printf("*** of GLIB, but you can also set the GLIB_CONFIG environment to point to the\n");
printf("*** correct copy of glib-config. (In this case, you will have to\n");
- printf("*** modify your LD_LIBRARY_PATH enviroment variable, or edit /etc/ld.so.conf\n");
+ printf("*** modify your LD_LIBRARY_PATH environment variable, or edit /etc/ld.so.conf\n");
printf("*** so that the correct libraries are found at run-time))\n");
}
}
return 1;
}
],, no_glib=yes,[echo $ac_n "cross compiling; assumed OK... $ac_c"])
CFLAGS="$ac_save_CFLAGS"
LIBS="$ac_save_LIBS"
fi
fi
if test "x$no_glib" = x ; then
AC_MSG_RESULT(yes)
ifelse([$2], , :, [$2])
else
AC_MSG_RESULT(no)
if test "$GLIB_CONFIG" = "no" ; then
echo "*** The glib-config script installed by GLIB could not be found"
echo "*** If GLIB was installed in PREFIX, make sure PREFIX/bin is in"
echo "*** your path, or set the GLIB_CONFIG environment variable to the"
echo "*** full path to glib-config."
else
if test -f conf.glibtest ; then
:
else
echo "*** Could not run GLIB test program, checking why..."
CFLAGS="$CFLAGS $GLIB_CFLAGS"
LIBS="$LIBS $GLIB_LIBS"
AC_TRY_LINK([
#include <glib.h>
#include <stdio.h>
], [ return ((glib_major_version) || (glib_minor_version) || (glib_micro_version)); ],
[ echo "*** The test program compiled, but did not run. This usually means"
echo "*** that the run-time linker is not finding GLIB or finding the wrong"
echo "*** version of GLIB. If it is not finding GLIB, you'll need to set your"
echo "*** LD_LIBRARY_PATH environment variable, or edit /etc/ld.so.conf to point"
echo "*** to the installed location Also, make sure you have run ldconfig if that"
echo "*** is required on your system"
echo "***"
echo "*** If you have an old version installed, it is best to remove it, although"
echo "*** you may also be able to get things to work by modifying LD_LIBRARY_PATH"
echo "***"
echo "*** If you have a RedHat 5.0 system, you should remove the GTK package that"
echo "*** came with the system with the command"
echo "***"
echo "*** rpm --erase --nodeps gtk gtk-devel" ],
[ echo "*** The test program failed to compile or link. See the file config.log for the"
- echo "*** exact error that occured. This usually means GLIB was incorrectly installed"
+ echo "*** exact error that occurred. This usually means GLIB was incorrectly installed"
echo "*** or that you have moved GLIB since it was installed. In the latter case, you"
echo "*** may want to edit the glib-config script: $GLIB_CONFIG" ])
CFLAGS="$ac_save_CFLAGS"
LIBS="$ac_save_LIBS"
fi
fi
GLIB_CFLAGS=""
GLIB_LIBS=""
ifelse([$3], , :, [$3])
fi
AC_SUBST(GLIB_CFLAGS)
AC_SUBST(GLIB_LIBS)
rm -f conf.glibtest
])
diff --git a/m4/gtk.m4 b/m4/gtk.m4
index f2dd472..6f6a422 100644
--- a/m4/gtk.m4
+++ b/m4/gtk.m4
@@ -1,194 +1,194 @@
# Configure paths for GTK+
# Owen Taylor 97-11-3
dnl AM_PATH_GTK([MINIMUM-VERSION, [ACTION-IF-FOUND [, ACTION-IF-NOT-FOUND [, MODULES]]]])
dnl Test for GTK, and define GTK_CFLAGS and GTK_LIBS
dnl
AC_DEFUN([AM_PATH_GTK],
[dnl
dnl Get the cflags and libraries from the gtk-config script
dnl
AC_ARG_WITH(gtk-prefix,[ --with-gtk-prefix=PFX Prefix where GTK is installed (optional)],
gtk_config_prefix="$withval", gtk_config_prefix="")
AC_ARG_WITH(gtk-exec-prefix,[ --with-gtk-exec-prefix=PFX Exec prefix where GTK is installed (optional)],
gtk_config_exec_prefix="$withval", gtk_config_exec_prefix="")
AC_ARG_ENABLE(gtktest, [ --disable-gtktest Do not try to compile and run a test GTK program],
, enable_gtktest=yes)
for module in . $4
do
case "$module" in
gthread)
gtk_config_args="$gtk_config_args gthread"
;;
esac
done
if test x$gtk_config_exec_prefix != x ; then
gtk_config_args="$gtk_config_args --exec-prefix=$gtk_config_exec_prefix"
if test x${GTK_CONFIG+set} != xset ; then
GTK_CONFIG=$gtk_config_exec_prefix/bin/gtk-config
fi
fi
if test x$gtk_config_prefix != x ; then
gtk_config_args="$gtk_config_args --prefix=$gtk_config_prefix"
if test x${GTK_CONFIG+set} != xset ; then
GTK_CONFIG=$gtk_config_prefix/bin/gtk-config
fi
fi
AC_PATH_PROG(GTK_CONFIG, gtk-config, no)
min_gtk_version=ifelse([$1], ,0.99.7,$1)
AC_MSG_CHECKING(for GTK - version >= $min_gtk_version)
no_gtk=""
if test "$GTK_CONFIG" = "no" ; then
no_gtk=yes
else
GTK_CFLAGS=`$GTK_CONFIG $gtk_config_args --cflags`
GTK_LIBS=`$GTK_CONFIG $gtk_config_args --libs`
gtk_config_major_version=`$GTK_CONFIG $gtk_config_args --version | \
sed 's/\([[0-9]]*\).\([[0-9]]*\).\([[0-9]]*\)/\1/'`
gtk_config_minor_version=`$GTK_CONFIG $gtk_config_args --version | \
sed 's/\([[0-9]]*\).\([[0-9]]*\).\([[0-9]]*\)/\2/'`
gtk_config_micro_version=`$GTK_CONFIG $gtk_config_args --version | \
sed 's/\([[0-9]]*\).\([[0-9]]*\).\([[0-9]]*\)/\3/'`
if test "x$enable_gtktest" = "xyes" ; then
ac_save_CFLAGS="$CFLAGS"
ac_save_LIBS="$LIBS"
CFLAGS="$CFLAGS $GTK_CFLAGS"
LIBS="$GTK_LIBS $LIBS"
dnl
dnl Now check if the installed GTK is sufficiently new. (Also sanity
dnl checks the results of gtk-config to some extent
dnl
rm -f conf.gtktest
AC_TRY_RUN([
#include <gtk/gtk.h>
#include <stdio.h>
#include <stdlib.h>
int
main ()
{
int major, minor, micro;
char *tmp_version;
system ("touch conf.gtktest");
/* HP/UX 9 (%@#!) writes to sscanf strings */
tmp_version = g_strdup("$min_gtk_version");
if (sscanf(tmp_version, "%d.%d.%d", &major, &minor, &micro) != 3) {
printf("%s, bad version string\n", "$min_gtk_version");
exit(1);
}
if ((gtk_major_version != $gtk_config_major_version) ||
(gtk_minor_version != $gtk_config_minor_version) ||
(gtk_micro_version != $gtk_config_micro_version))
{
printf("\n*** 'gtk-config --version' returned %d.%d.%d, but GTK+ (%d.%d.%d)\n",
$gtk_config_major_version, $gtk_config_minor_version, $gtk_config_micro_version,
gtk_major_version, gtk_minor_version, gtk_micro_version);
printf ("*** was found! If gtk-config was correct, then it is best\n");
printf ("*** to remove the old version of GTK+. You may also be able to fix the error\n");
- printf("*** by modifying your LD_LIBRARY_PATH enviroment variable, or by editing\n");
+ printf("*** by modifying your LD_LIBRARY_PATH environment variable, or by editing\n");
printf("*** /etc/ld.so.conf. Make sure you have run ldconfig if that is\n");
printf("*** required on your system.\n");
printf("*** If gtk-config was wrong, set the environment variable GTK_CONFIG\n");
printf("*** to point to the correct copy of gtk-config, and remove the file config.cache\n");
printf("*** before re-running configure\n");
}
#if defined (GTK_MAJOR_VERSION) && defined (GTK_MINOR_VERSION) && defined (GTK_MICRO_VERSION)
else if ((gtk_major_version != GTK_MAJOR_VERSION) ||
(gtk_minor_version != GTK_MINOR_VERSION) ||
(gtk_micro_version != GTK_MICRO_VERSION))
{
printf("*** GTK+ header files (version %d.%d.%d) do not match\n",
GTK_MAJOR_VERSION, GTK_MINOR_VERSION, GTK_MICRO_VERSION);
printf("*** library (version %d.%d.%d)\n",
gtk_major_version, gtk_minor_version, gtk_micro_version);
}
#endif /* defined (GTK_MAJOR_VERSION) ... */
else
{
if ((gtk_major_version > major) ||
((gtk_major_version == major) && (gtk_minor_version > minor)) ||
((gtk_major_version == major) && (gtk_minor_version == minor) && (gtk_micro_version >= micro)))
{
return 0;
}
else
{
printf("\n*** An old version of GTK+ (%d.%d.%d) was found.\n",
gtk_major_version, gtk_minor_version, gtk_micro_version);
printf("*** You need a version of GTK+ newer than %d.%d.%d. The latest version of\n",
major, minor, micro);
printf("*** GTK+ is always available from ftp://ftp.gtk.org.\n");
printf("***\n");
printf("*** If you have already installed a sufficiently new version, this error\n");
printf("*** probably means that the wrong copy of the gtk-config shell script is\n");
printf("*** being found. The easiest way to fix this is to remove the old version\n");
printf("*** of GTK+, but you can also set the GTK_CONFIG environment to point to the\n");
printf("*** correct copy of gtk-config. (In this case, you will have to\n");
- printf("*** modify your LD_LIBRARY_PATH enviroment variable, or edit /etc/ld.so.conf\n");
+ printf("*** modify your LD_LIBRARY_PATH environment variable, or edit /etc/ld.so.conf\n");
printf("*** so that the correct libraries are found at run-time))\n");
}
}
return 1;
}
],, no_gtk=yes,[echo $ac_n "cross compiling; assumed OK... $ac_c"])
CFLAGS="$ac_save_CFLAGS"
LIBS="$ac_save_LIBS"
fi
fi
if test "x$no_gtk" = x ; then
AC_MSG_RESULT(yes)
ifelse([$2], , :, [$2])
else
AC_MSG_RESULT(no)
if test "$GTK_CONFIG" = "no" ; then
echo "*** The gtk-config script installed by GTK could not be found"
echo "*** If GTK was installed in PREFIX, make sure PREFIX/bin is in"
echo "*** your path, or set the GTK_CONFIG environment variable to the"
echo "*** full path to gtk-config."
else
if test -f conf.gtktest ; then
:
else
echo "*** Could not run GTK test program, checking why..."
CFLAGS="$CFLAGS $GTK_CFLAGS"
LIBS="$LIBS $GTK_LIBS"
AC_TRY_LINK([
#include <gtk/gtk.h>
#include <stdio.h>
], [ return ((gtk_major_version) || (gtk_minor_version) || (gtk_micro_version)); ],
[ echo "*** The test program compiled, but did not run. This usually means"
echo "*** that the run-time linker is not finding GTK or finding the wrong"
echo "*** version of GTK. If it is not finding GTK, you'll need to set your"
echo "*** LD_LIBRARY_PATH environment variable, or edit /etc/ld.so.conf to point"
echo "*** to the installed location Also, make sure you have run ldconfig if that"
echo "*** is required on your system"
echo "***"
echo "*** If you have an old version installed, it is best to remove it, although"
echo "*** you may also be able to get things to work by modifying LD_LIBRARY_PATH"
echo "***"
echo "*** If you have a RedHat 5.0 system, you should remove the GTK package that"
echo "*** came with the system with the command"
echo "***"
echo "*** rpm --erase --nodeps gtk gtk-devel" ],
[ echo "*** The test program failed to compile or link. See the file config.log for the"
- echo "*** exact error that occured. This usually means GTK was incorrectly installed"
+ echo "*** exact error that occurred. This usually means GTK was incorrectly installed"
echo "*** or that you have moved GTK since it was installed. In the latter case, you"
echo "*** may want to edit the gtk-config script: $GTK_CONFIG" ])
CFLAGS="$ac_save_CFLAGS"
LIBS="$ac_save_LIBS"
fi
fi
GTK_CFLAGS=""
GTK_LIBS=""
ifelse([$3], , :, [$3])
fi
AC_SUBST(GTK_CFLAGS)
AC_SUBST(GTK_LIBS)
rm -f conf.gtktest
])
diff --git a/pinentry/argparse.c b/pinentry/argparse.c
index e31b67e..4f586d4 100644
--- a/pinentry/argparse.c
+++ b/pinentry/argparse.c
@@ -1,1607 +1,1607 @@
/* [argparse.c wk 17.06.97] Argument Parser for option handling
* Copyright (C) 1998-2001, 2006-2008, 2012 Free Software Foundation, Inc.
* Copyright (C) 1997-2001, 2006-2008, 2013-2015 Werner Koch
*
* This file is part of JNLIB, which is a subsystem of GnuPG.
*
* JNLIB is free software; you can redistribute it and/or modify it
* under the terms of either
*
* - the GNU Lesser General Public License as published by the Free
* Software Foundation; either version 3 of the License, or (at
* your option) any later version.
*
* or
*
* - 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.
*
* or both in parallel, as here.
*
* JNLIB 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 copies of the GNU General Public License
* and the GNU Lesser General Public License along with this program;
* if not, see <http://www.gnu.org/licenses/>.
*/
/* This file may be used as part of GnuPG or standalone. A GnuPG
build is detected by the presence of the macro GNUPG_MAJOR_VERSION.
Some feature are only availalbe in the GnuPG build mode.
*/
#ifdef HAVE_CONFIG_H
#include <config.h>
#endif
#include <stdio.h>
#include <stdlib.h>
#include <ctype.h>
#include <string.h>
#include <stdarg.h>
#include <limits.h>
#include <errno.h>
#ifdef GNUPG_MAJOR_VERSION
# include "libjnlib-config.h"
# include "mischelp.h"
# include "stringhelp.h"
# include "logging.h"
# ifdef JNLIB_NEED_UTF8CONV
# include "utf8conv.h"
# endif
#endif /*GNUPG_MAJOR_VERSION*/
#include "argparse.h"
/* GnuPG uses GPLv3+ but a standalone version of this defaults to
GPLv2+ because that is the license of this file. Change this if
you include it in a program which uses GPLv3. If you don't want to
set a a copyright string for your usage() you may also hardcode it
here. */
#ifndef GNUPG_MAJOR_VERSION
# define ARGPARSE_GPL_VERSION 2
# define ARGPARSE_CRIGHT_STR "Copyright (C) YEAR NAME"
#else /* Used by GnuPG */
# define ARGPARSE_GPL_VERSION 3
# define ARGPARSE_CRIGHT_STR "Copyright (C) 2015 Free Software Foundation, Inc."
#endif /*GNUPG_MAJOR_VERSION*/
/* Replacements for standalone builds. */
#ifndef GNUPG_MAJOR_VERSION
# ifndef _
# define _(a) (a)
# endif
# ifndef DIM
# define DIM(v) (sizeof(v)/sizeof((v)[0]))
# endif
# define jnlib_malloc(a) malloc ((a))
# define jnlib_realloc(a,b) realloc ((a), (b))
# define jnlib_strdup(a) strdup ((a))
# define jnlib_free(a) free ((a))
# define jnlib_log_error my_log_error
# define jnlib_log_bug my_log_bug
# define trim_spaces(a) my_trim_spaces ((a))
# define map_static_macro_string(a) (a)
#endif /*!GNUPG_MAJOR_VERSION*/
#define ARGPARSE_STR(v) #v
#define ARGPARSE_STR2(v) ARGPARSE_STR(v)
/* Replacements for standalone builds. */
#ifndef GNUPG_MAJOR_VERSION
static void
my_log_error (const char *fmt, ...)
{
va_list arg_ptr ;
va_start (arg_ptr, fmt);
fprintf (stderr, "%s: ", strusage (11));
vfprintf (stderr, fmt, arg_ptr);
va_end (arg_ptr);
}
static void
my_log_bug (const char *fmt, ...)
{
va_list arg_ptr ;
va_start (arg_ptr, fmt);
fprintf (stderr, "%s: Ohhhh jeeee: ", strusage (11));
vfprintf (stderr, fmt, arg_ptr);
va_end (arg_ptr);
abort ();
}
static char *
my_trim_spaces (char *str)
{
char *string, *p, *mark;
string = str;
/* Find first non space character. */
for (p=string; *p && isspace (*(unsigned char*)p) ; p++)
;
/* Move characters. */
for ((mark = NULL); (*string = *p); string++, p++)
if (isspace (*(unsigned char*)p))
{
if (!mark)
mark = string;
}
else
mark = NULL;
if (mark)
*mark = '\0' ; /* Remove trailing spaces. */
return str ;
}
#endif /*!GNUPG_MAJOR_VERSION*/
/*********************************
* @Summary arg_parse
* #include "argparse.h"
*
* typedef struct {
* char *argc; pointer to argc (value subject to change)
* char ***argv; pointer to argv (value subject to change)
* unsigned flags; Global flags (DO NOT CHANGE)
* int err; print error about last option
* 1 = warning, 2 = abort
* int r_opt; return option
* int r_type; type of return value (0 = no argument found)
* union {
* int ret_int;
* long ret_long
* ulong ret_ulong;
* char *ret_str;
* } r; Return values
* struct {
* int idx;
* const char *last;
* void *aliases;
* } internal; DO NOT CHANGE
* } ARGPARSE_ARGS;
*
* typedef struct {
* int short_opt;
* const char *long_opt;
* unsigned flags;
* } ARGPARSE_OPTS;
*
* int arg_parse( ARGPARSE_ARGS *arg, ARGPARSE_OPTS *opts );
*
* @Description
* This is my replacement for getopt(). See the example for a typical usage.
* Global flags are:
* Bit 0 : Do not remove options form argv
* Bit 1 : Do not stop at last option but return other args
* with r_opt set to -1.
* Bit 2 : Assume options and real args are mixed.
* Bit 3 : Do not use -- to stop option processing.
* Bit 4 : Do not skip the first arg.
* Bit 5 : allow usage of long option with only one dash
* Bit 6 : ignore --version
* all other bits must be set to zero, this value is modified by the
* function, so assume this is write only.
* Local flags (for each option):
* Bit 2-0 : 0 = does not take an argument
* 1 = takes int argument
* 2 = takes string argument
* 3 = takes long argument
* 4 = takes ulong argument
* Bit 3 : argument is optional (r_type will the be set to 0)
* Bit 4 : allow 0x etc. prefixed values.
* Bit 6 : Ignore this option
* Bit 7 : This is a command and not an option
* You stop the option processing by setting opts to NULL, the function will
* then return 0.
* @Return Value
* Returns the args.r_opt or 0 if ready
* r_opt may be -2/-7 to indicate an unknown option/command.
* @See Also
* ArgExpand
* @Notes
* You do not need to process the options 'h', '--help' or '--version'
* because this function includes standard help processing; but if you
* specify '-h', '--help' or '--version' you have to do it yourself.
* The option '--' stops argument processing; if bit 1 is set the function
* continues to return normal arguments.
* To process float args or unsigned args you must use a string args and do
* the conversion yourself.
* @Example
*
* ARGPARSE_OPTS opts[] = {
* { 'v', "verbose", 0 },
* { 'd', "debug", 0 },
* { 'o', "output", 2 },
* { 'c', "cross-ref", 2|8 },
* { 'm', "my-option", 1|8 },
* { 300, "ignored-long-option, ARGPARSE_OP_IGNORE},
* { 500, "have-no-short-option-for-this-long-option", 0 },
* {0} };
* ARGPARSE_ARGS pargs = { &argc, &argv, 0 }
*
* while( ArgParse( &pargs, &opts) ) {
* switch( pargs.r_opt ) {
* case 'v': opt.verbose++; break;
* case 'd': opt.debug++; break;
* case 'o': opt.outfile = pargs.r.ret_str; break;
* case 'c': opt.crf = pargs.r_type? pargs.r.ret_str:"a.crf"; break;
* case 'm': opt.myopt = pargs.r_type? pargs.r.ret_int : 1; break;
* case 500: opt.a_long_one++; break
* default : pargs.err = 1; break; -- force warning output --
* }
* }
* if( argc > 1 )
* log_fatal( "Too many args");
*
*/
typedef struct alias_def_s *ALIAS_DEF;
struct alias_def_s {
ALIAS_DEF next;
char *name; /* malloced buffer with name, \0, value */
const char *value; /* ptr into name */
};
/* Object to store the names for the --ignore-invalid-option option.
This is a simple linked list. */
typedef struct iio_item_def_s *IIO_ITEM_DEF;
struct iio_item_def_s
{
IIO_ITEM_DEF next;
char name[1]; /* String with the long option name. */
};
static const char *(*strusage_handler)( int ) = NULL;
static int (*custom_outfnc) (int, const char *);
static int set_opt_arg(ARGPARSE_ARGS *arg, unsigned flags, char *s);
static void show_help(ARGPARSE_OPTS *opts, unsigned flags);
static void show_version(void);
static int writestrings (int is_error, const char *string, ...)
#if __GNUC__ >= 4
__attribute__ ((sentinel(0)))
#endif
;
void
argparse_register_outfnc (int (*fnc)(int, const char *))
{
custom_outfnc = fnc;
}
/* Write STRING and all following const char * arguments either to
stdout or, if IS_ERROR is set, to stderr. The list of strings must
be terminated by a NULL. */
static int
writestrings (int is_error, const char *string, ...)
{
va_list arg_ptr;
const char *s;
int count = 0;
if (string)
{
s = string;
va_start (arg_ptr, string);
do
{
if (custom_outfnc)
custom_outfnc (is_error? 2:1, s);
else
fputs (s, is_error? stderr : stdout);
count += strlen (s);
}
while ((s = va_arg (arg_ptr, const char *)));
va_end (arg_ptr);
}
return count;
}
static void
flushstrings (int is_error)
{
if (custom_outfnc)
custom_outfnc (is_error? 2:1, NULL);
else
fflush (is_error? stderr : stdout);
}
static void
initialize( ARGPARSE_ARGS *arg, const char *filename, unsigned *lineno )
{
if( !(arg->flags & (1<<15)) )
{
/* Initialize this instance. */
arg->internal.idx = 0;
arg->internal.last = NULL;
arg->internal.inarg = 0;
arg->internal.stopped = 0;
arg->internal.aliases = NULL;
arg->internal.cur_alias = NULL;
arg->internal.iio_list = NULL;
arg->err = 0;
arg->flags |= 1<<15; /* Mark as initialized. */
if ( *arg->argc < 0 )
jnlib_log_bug ("invalid argument for arg_parse\n");
}
if (arg->err)
{
/* Last option was erroneous. */
const char *s;
if (filename)
{
if ( arg->r_opt == ARGPARSE_UNEXPECTED_ARG )
s = _("argument not expected");
else if ( arg->r_opt == ARGPARSE_READ_ERROR )
s = _("read error");
else if ( arg->r_opt == ARGPARSE_KEYWORD_TOO_LONG )
s = _("keyword too long");
else if ( arg->r_opt == ARGPARSE_MISSING_ARG )
s = _("missing argument");
else if ( arg->r_opt == ARGPARSE_INVALID_ARG )
s = _("invalid argument");
else if ( arg->r_opt == ARGPARSE_INVALID_COMMAND )
s = _("invalid command");
else if ( arg->r_opt == ARGPARSE_INVALID_ALIAS )
s = _("invalid alias definition");
else if ( arg->r_opt == ARGPARSE_OUT_OF_CORE )
s = _("out of core");
else
s = _("invalid option");
jnlib_log_error ("%s:%u: %s\n", filename, *lineno, s);
}
else
{
s = arg->internal.last? arg->internal.last:"[??]";
if ( arg->r_opt == ARGPARSE_MISSING_ARG )
jnlib_log_error (_("missing argument for option \"%.50s\"\n"), s);
else if ( arg->r_opt == ARGPARSE_INVALID_ARG )
jnlib_log_error (_("invalid argument for option \"%.50s\"\n"), s);
else if ( arg->r_opt == ARGPARSE_UNEXPECTED_ARG )
jnlib_log_error (_("option \"%.50s\" does not expect an "
"argument\n"), s );
else if ( arg->r_opt == ARGPARSE_INVALID_COMMAND )
jnlib_log_error (_("invalid command \"%.50s\"\n"), s);
else if ( arg->r_opt == ARGPARSE_AMBIGUOUS_OPTION )
jnlib_log_error (_("option \"%.50s\" is ambiguous\n"), s);
else if ( arg->r_opt == ARGPARSE_AMBIGUOUS_COMMAND )
jnlib_log_error (_("command \"%.50s\" is ambiguous\n"),s );
else if ( arg->r_opt == ARGPARSE_OUT_OF_CORE )
jnlib_log_error ("%s\n", _("out of core\n"));
else
jnlib_log_error (_("invalid option \"%.50s\"\n"), s);
}
if (arg->err != ARGPARSE_PRINT_WARNING)
exit (2);
arg->err = 0;
}
/* Zero out the return value union. */
arg->r.ret_str = NULL;
arg->r.ret_long = 0;
}
static void
store_alias( ARGPARSE_ARGS *arg, char *name, char *value )
{
/* TODO: replace this dummy function with a rea one
* and fix the probelms IRIX has with (ALIAS_DEV)arg..
* used as lvalue
*/
(void)arg;
(void)name;
(void)value;
#if 0
ALIAS_DEF a = jnlib_xmalloc( sizeof *a );
a->name = name;
a->value = value;
a->next = (ALIAS_DEF)arg->internal.aliases;
(ALIAS_DEF)arg->internal.aliases = a;
#endif
}
/* Return true if KEYWORD is in the ignore-invalid-option list. */
static int
ignore_invalid_option_p (ARGPARSE_ARGS *arg, const char *keyword)
{
IIO_ITEM_DEF item = arg->internal.iio_list;
for (; item; item = item->next)
if (!strcmp (item->name, keyword))
return 1;
return 0;
}
/* Add the keywords up to the next LF to the list of to be ignored
options. After returning FP will either be at EOF or the next
character read wll be the first of a new line. The function
returns 0 on success or true on malloc failure. */
static int
ignore_invalid_option_add (ARGPARSE_ARGS *arg, FILE *fp)
{
IIO_ITEM_DEF item;
int c;
char name[100];
int namelen = 0;
int ready = 0;
enum { skipWS, collectNAME, skipNAME, addNAME} state = skipWS;
while (!ready)
{
c = getc (fp);
if (c == '\n')
ready = 1;
else if (c == EOF)
{
c = '\n';
ready = 1;
}
again:
switch (state)
{
case skipWS:
if (!isascii (c) || !isspace(c))
{
namelen = 0;
state = collectNAME;
goto again;
}
break;
case collectNAME:
if (isspace (c))
{
state = addNAME;
goto again;
}
else if (namelen < DIM(name)-1)
name[namelen++] = c;
else /* Too long. */
state = skipNAME;
break;
case skipNAME:
if (isspace (c))
{
state = skipWS;
goto again;
}
break;
case addNAME:
name[namelen] = 0;
if (!ignore_invalid_option_p (arg, name))
{
item = jnlib_malloc (sizeof *item + namelen);
if (!item)
return 1;
strcpy (item->name, name);
item->next = (IIO_ITEM_DEF)arg->internal.iio_list;
arg->internal.iio_list = item;
}
state = skipWS;
goto again;
}
}
return 0;
}
/* Clear the entire ignore-invalid-option list. */
static void
ignore_invalid_option_clear (ARGPARSE_ARGS *arg)
{
IIO_ITEM_DEF item, tmpitem;
for (item = arg->internal.iio_list; item; item = tmpitem)
{
tmpitem = item->next;
jnlib_free (item);
}
arg->internal.iio_list = NULL;
}
/****************
* Get options from a file.
* Lines starting with '#' are comment lines.
* Syntax is simply a keyword and the argument.
* Valid keywords are all keywords from the long_opt list without
* the leading dashes. The special keywords "help", "warranty" and "version"
* are not valid here.
* The special keyword "alias" may be used to store alias definitions,
* which are later expanded like long options.
* The option
* ignore-invalid-option OPTIONNAMEs
* is recognized and updates a list of option which should be ignored if they
* are not defined.
* Caller must free returned strings.
* If called with FP set to NULL command line args are parse instead.
*
* Q: Should we allow the syntax
* keyword = value
* and accept for boolean options a value of 1/0, yes/no or true/false?
* Note: Abbreviation of options is here not allowed.
*/
int
optfile_parse (FILE *fp, const char *filename, unsigned *lineno,
ARGPARSE_ARGS *arg, ARGPARSE_OPTS *opts)
{
int state, i, c;
int idx=0;
char keyword[100];
char *buffer = NULL;
size_t buflen = 0;
int in_alias=0;
if (!fp) /* Divert to to arg_parse() in this case. */
return arg_parse (arg, opts);
initialize (arg, filename, lineno);
/* Find the next keyword. */
state = i = 0;
for (;;)
{
c = getc (fp);
if (c == '\n' || c== EOF )
{
if ( c != EOF )
++*lineno;
if (state == -1)
break;
else if (state == 2)
{
keyword[i] = 0;
for (i=0; opts[i].short_opt; i++ )
{
if (opts[i].long_opt && !strcmp (opts[i].long_opt, keyword))
break;
}
idx = i;
arg->r_opt = opts[idx].short_opt;
if ((opts[idx].flags & ARGPARSE_OPT_IGNORE))
{
state = i = 0;
continue;
}
else if (!opts[idx].short_opt )
{
if (!strcmp (keyword, "ignore-invalid-option"))
{
/* No argument - ignore this meta option. */
state = i = 0;
continue;
}
else if (ignore_invalid_option_p (arg, keyword))
{
/* This invalid option is in the iio list. */
state = i = 0;
continue;
}
arg->r_opt = ((opts[idx].flags & ARGPARSE_OPT_COMMAND)
? ARGPARSE_INVALID_COMMAND
: ARGPARSE_INVALID_OPTION);
}
else if (!(opts[idx].flags & ARGPARSE_TYPE_MASK))
arg->r_type = 0; /* Does not take an arg. */
else if ((opts[idx].flags & ARGPARSE_OPT_OPTIONAL) )
arg->r_type = 0; /* Arg is optional. */
else
arg->r_opt = ARGPARSE_MISSING_ARG;
break;
}
else if (state == 3)
{
/* No argument found. */
if (in_alias)
arg->r_opt = ARGPARSE_MISSING_ARG;
else if (!(opts[idx].flags & ARGPARSE_TYPE_MASK))
arg->r_type = 0; /* Does not take an arg. */
else if ((opts[idx].flags & ARGPARSE_OPT_OPTIONAL))
arg->r_type = 0; /* No optional argument. */
else
arg->r_opt = ARGPARSE_MISSING_ARG;
break;
}
else if (state == 4)
{
/* Has an argument. */
if (in_alias)
{
if (!buffer)
arg->r_opt = ARGPARSE_UNEXPECTED_ARG;
else
{
char *p;
buffer[i] = 0;
p = strpbrk (buffer, " \t");
if (p)
{
*p++ = 0;
trim_spaces (p);
}
if (!p || !*p)
{
jnlib_free (buffer);
arg->r_opt = ARGPARSE_INVALID_ALIAS;
}
else
{
store_alias (arg, buffer, p);
}
}
}
else if (!(opts[idx].flags & ARGPARSE_TYPE_MASK))
arg->r_opt = ARGPARSE_UNEXPECTED_ARG;
else
{
char *p;
if (!buffer)
{
keyword[i] = 0;
buffer = jnlib_strdup (keyword);
if (!buffer)
arg->r_opt = ARGPARSE_OUT_OF_CORE;
}
else
buffer[i] = 0;
if (buffer)
{
trim_spaces (buffer);
p = buffer;
if (*p == '"')
{
/* Remove quotes. */
p++;
if (*p && p[strlen(p)-1] == '\"' )
p[strlen(p)-1] = 0;
}
if (!set_opt_arg (arg, opts[idx].flags, p))
jnlib_free(buffer);
}
}
break;
}
else if (c == EOF)
{
ignore_invalid_option_clear (arg);
if (ferror (fp))
arg->r_opt = ARGPARSE_READ_ERROR;
else
arg->r_opt = 0; /* EOF. */
break;
}
state = 0;
i = 0;
}
else if (state == -1)
; /* Skip. */
else if (state == 0 && isascii (c) && isspace(c))
; /* Skip leading white space. */
else if (state == 0 && c == '#' )
state = 1; /* Start of a comment. */
else if (state == 1)
; /* Skip comments. */
else if (state == 2 && isascii (c) && isspace(c))
{
/* Check keyword. */
keyword[i] = 0;
for (i=0; opts[i].short_opt; i++ )
if (opts[i].long_opt && !strcmp (opts[i].long_opt, keyword))
break;
idx = i;
arg->r_opt = opts[idx].short_opt;
if ((opts[idx].flags & ARGPARSE_OPT_IGNORE))
{
state = 1; /* Process like a comment. */
}
else if (!opts[idx].short_opt)
{
if (!strcmp (keyword, "alias"))
{
in_alias = 1;
state = 3;
}
else if (!strcmp (keyword, "ignore-invalid-option"))
{
if (ignore_invalid_option_add (arg, fp))
{
arg->r_opt = ARGPARSE_OUT_OF_CORE;
break;
}
state = i = 0;
++*lineno;
}
else if (ignore_invalid_option_p (arg, keyword))
state = 1; /* Process like a comment. */
else
{
arg->r_opt = ((opts[idx].flags & ARGPARSE_OPT_COMMAND)
? ARGPARSE_INVALID_COMMAND
: ARGPARSE_INVALID_OPTION);
state = -1; /* Skip rest of line and leave. */
}
}
else
state = 3;
}
else if (state == 3)
{
/* Skip leading spaces of the argument. */
if (!isascii (c) || !isspace(c))
{
i = 0;
keyword[i++] = c;
state = 4;
}
}
else if (state == 4)
{
/* Collect the argument. */
if (buffer)
{
if (i < buflen-1)
buffer[i++] = c;
else
{
char *tmp;
size_t tmplen = buflen + 50;
tmp = jnlib_realloc (buffer, tmplen);
if (tmp)
{
buflen = tmplen;
buffer = tmp;
buffer[i++] = c;
}
else
{
jnlib_free (buffer);
arg->r_opt = ARGPARSE_OUT_OF_CORE;
break;
}
}
}
else if (i < DIM(keyword)-1)
keyword[i++] = c;
else
{
size_t tmplen = DIM(keyword) + 50;
buffer = jnlib_malloc (tmplen);
if (buffer)
{
buflen = tmplen;
memcpy(buffer, keyword, i);
buffer[i++] = c;
}
else
{
arg->r_opt = ARGPARSE_OUT_OF_CORE;
break;
}
}
}
else if (i >= DIM(keyword)-1)
{
arg->r_opt = ARGPARSE_KEYWORD_TOO_LONG;
state = -1; /* Skip rest of line and leave. */
}
else
{
keyword[i++] = c;
state = 2;
}
}
return arg->r_opt;
}
static int
find_long_option( ARGPARSE_ARGS *arg,
ARGPARSE_OPTS *opts, const char *keyword )
{
int i;
size_t n;
(void)arg;
/* Would be better if we can do a binary search, but it is not
possible to reorder our option table because we would mess
up our help strings - What we can do is: Build a nice option
- lookup table wehn this function is first invoked */
+ lookup table when this function is first invoked */
if( !*keyword )
return -1;
for(i=0; opts[i].short_opt; i++ )
if( opts[i].long_opt && !strcmp( opts[i].long_opt, keyword) )
return i;
#if 0
{
ALIAS_DEF a;
/* see whether it is an alias */
for( a = args->internal.aliases; a; a = a->next ) {
if( !strcmp( a->name, keyword) ) {
/* todo: must parse the alias here */
args->internal.cur_alias = a;
return -3; /* alias available */
}
}
}
#endif
/* not found, see whether it is an abbreviation */
/* aliases may not be abbreviated */
n = strlen( keyword );
for(i=0; opts[i].short_opt; i++ ) {
if( opts[i].long_opt && !strncmp( opts[i].long_opt, keyword, n ) ) {
int j;
for(j=i+1; opts[j].short_opt; j++ ) {
if( opts[j].long_opt
&& !strncmp( opts[j].long_opt, keyword, n ) )
return -2; /* abbreviation is ambiguous */
}
return i;
}
}
return -1; /* Not found. */
}
int
arg_parse( ARGPARSE_ARGS *arg, ARGPARSE_OPTS *opts)
{
int idx;
int argc;
char **argv;
char *s, *s2;
int i;
initialize( arg, NULL, NULL );
argc = *arg->argc;
argv = *arg->argv;
idx = arg->internal.idx;
if (!idx && argc && !(arg->flags & ARGPARSE_FLAG_ARG0))
{
/* Skip the first argument. */
argc--; argv++; idx++;
}
next_one:
if (!argc)
{
/* No more args. */
arg->r_opt = 0;
goto leave; /* Ready. */
}
s = *argv;
arg->internal.last = s;
if (arg->internal.stopped && (arg->flags & ARGPARSE_FLAG_ALL))
{
arg->r_opt = ARGPARSE_IS_ARG; /* Not an option but an argument. */
arg->r_type = 2;
arg->r.ret_str = s;
argc--; argv++; idx++; /* set to next one */
}
else if( arg->internal.stopped )
{
arg->r_opt = 0;
goto leave; /* Ready. */
}
else if ( *s == '-' && s[1] == '-' )
{
/* Long option. */
char *argpos;
arg->internal.inarg = 0;
if (!s[2] && !(arg->flags & ARGPARSE_FLAG_NOSTOP))
{
/* Stop option processing. */
arg->internal.stopped = 1;
arg->flags |= ARGPARSE_FLAG_STOP_SEEN;
argc--; argv++; idx++;
goto next_one;
}
argpos = strchr( s+2, '=' );
if ( argpos )
*argpos = 0;
i = find_long_option ( arg, opts, s+2 );
if ( argpos )
*argpos = '=';
if ( i < 0 && !strcmp ( "help", s+2) )
show_help (opts, arg->flags);
else if ( i < 0 && !strcmp ( "version", s+2) )
{
if (!(arg->flags & ARGPARSE_FLAG_NOVERSION))
{
show_version ();
exit(0);
}
}
else if ( i < 0 && !strcmp( "warranty", s+2))
{
writestrings (0, strusage (16), "\n", NULL);
exit (0);
}
else if ( i < 0 && !strcmp( "dump-options", s+2) )
{
for (i=0; opts[i].short_opt; i++ )
{
if (opts[i].long_opt && !(opts[i].flags & ARGPARSE_OPT_IGNORE))
writestrings (0, "--", opts[i].long_opt, "\n", NULL);
}
writestrings (0, "--dump-options\n--help\n--version\n--warranty\n",
NULL);
exit (0);
}
if ( i == -2 )
arg->r_opt = ARGPARSE_AMBIGUOUS_OPTION;
else if ( i == -1 )
{
arg->r_opt = ARGPARSE_INVALID_OPTION;
arg->r.ret_str = s+2;
}
else
arg->r_opt = opts[i].short_opt;
if ( i < 0 )
;
else if ( (opts[i].flags & ARGPARSE_TYPE_MASK) )
{
if ( argpos )
{
s2 = argpos+1;
if ( !*s2 )
s2 = NULL;
}
else
s2 = argv[1];
if ( !s2 && (opts[i].flags & ARGPARSE_OPT_OPTIONAL) )
{
arg->r_type = ARGPARSE_TYPE_NONE; /* Argument is optional. */
}
else if ( !s2 )
{
arg->r_opt = ARGPARSE_MISSING_ARG;
}
else if ( !argpos && *s2 == '-'
&& (opts[i].flags & ARGPARSE_OPT_OPTIONAL) )
{
/* The argument is optional and the next seems to be an
option. We do not check this possible option but
assume no argument */
arg->r_type = ARGPARSE_TYPE_NONE;
}
else
{
set_opt_arg (arg, opts[i].flags, s2);
if ( !argpos )
{
argc--; argv++; idx++; /* Skip one. */
}
}
}
else
{
/* Does not take an argument. */
if ( argpos )
arg->r_type = ARGPARSE_UNEXPECTED_ARG;
else
arg->r_type = 0;
}
argc--; argv++; idx++; /* Set to next one. */
}
else if ( (*s == '-' && s[1]) || arg->internal.inarg )
{
/* Short option. */
int dash_kludge = 0;
i = 0;
if ( !arg->internal.inarg )
{
arg->internal.inarg++;
if ( (arg->flags & ARGPARSE_FLAG_ONEDASH) )
{
for (i=0; opts[i].short_opt; i++ )
if ( opts[i].long_opt && !strcmp (opts[i].long_opt, s+1))
{
dash_kludge = 1;
break;
}
}
}
s += arg->internal.inarg;
if (!dash_kludge )
{
for (i=0; opts[i].short_opt; i++ )
if ( opts[i].short_opt == *s )
break;
}
if ( !opts[i].short_opt && ( *s == 'h' || *s == '?' ) )
show_help (opts, arg->flags);
arg->r_opt = opts[i].short_opt;
if (!opts[i].short_opt )
{
arg->r_opt = (opts[i].flags & ARGPARSE_OPT_COMMAND)?
ARGPARSE_INVALID_COMMAND:ARGPARSE_INVALID_OPTION;
arg->internal.inarg++; /* Point to the next arg. */
arg->r.ret_str = s;
}
else if ( (opts[i].flags & ARGPARSE_TYPE_MASK) )
{
if ( s[1] && !dash_kludge )
{
s2 = s+1;
set_opt_arg (arg, opts[i].flags, s2);
}
else
{
s2 = argv[1];
if ( !s2 && (opts[i].flags & ARGPARSE_OPT_OPTIONAL) )
{
arg->r_type = ARGPARSE_TYPE_NONE;
}
else if ( !s2 )
{
arg->r_opt = ARGPARSE_MISSING_ARG;
}
else if ( *s2 == '-' && s2[1]
&& (opts[i].flags & ARGPARSE_OPT_OPTIONAL) )
{
/* The argument is optional and the next seems to
be an option. We do not check this possible
option but assume no argument. */
arg->r_type = ARGPARSE_TYPE_NONE;
}
else
{
set_opt_arg (arg, opts[i].flags, s2);
argc--; argv++; idx++; /* Skip one. */
}
}
s = "x"; /* This is so that !s[1] yields false. */
}
else
{
/* Does not take an argument. */
arg->r_type = ARGPARSE_TYPE_NONE;
arg->internal.inarg++; /* Point to the next arg. */
}
if ( !s[1] || dash_kludge )
{
/* No more concatenated short options. */
arg->internal.inarg = 0;
argc--; argv++; idx++;
}
}
else if ( arg->flags & ARGPARSE_FLAG_MIXED )
{
arg->r_opt = ARGPARSE_IS_ARG;
arg->r_type = 2;
arg->r.ret_str = s;
argc--; argv++; idx++; /* Set to next one. */
}
else
{
arg->internal.stopped = 1; /* Stop option processing. */
goto next_one;
}
leave:
*arg->argc = argc;
*arg->argv = argv;
arg->internal.idx = idx;
return arg->r_opt;
}
/* Returns: -1 on error, 0 for an integer type and 1 for a non integer
type argument. */
static int
set_opt_arg (ARGPARSE_ARGS *arg, unsigned flags, char *s)
{
int base = (flags & ARGPARSE_OPT_PREFIX)? 0 : 10;
long l;
switch ( (arg->r_type = (flags & ARGPARSE_TYPE_MASK)) )
{
case ARGPARSE_TYPE_LONG:
case ARGPARSE_TYPE_INT:
errno = 0;
l = strtol (s, NULL, base);
if ((l == LONG_MIN || l == LONG_MAX) && errno == ERANGE)
{
arg->r_opt = ARGPARSE_INVALID_ARG;
return -1;
}
if (arg->r_type == ARGPARSE_TYPE_LONG)
arg->r.ret_long = l;
else if ( (l < 0 && l < INT_MIN) || l > INT_MAX )
{
arg->r_opt = ARGPARSE_INVALID_ARG;
return -1;
}
else
arg->r.ret_int = (int)l;
return 0;
case ARGPARSE_TYPE_ULONG:
while (isascii (*s) && isspace(*s))
s++;
if (*s == '-')
{
arg->r.ret_ulong = 0;
arg->r_opt = ARGPARSE_INVALID_ARG;
return -1;
}
errno = 0;
arg->r.ret_ulong = strtoul (s, NULL, base);
if (arg->r.ret_ulong == ULONG_MAX && errno == ERANGE)
{
arg->r_opt = ARGPARSE_INVALID_ARG;
return -1;
}
return 0;
case ARGPARSE_TYPE_STRING:
default:
arg->r.ret_str = s;
return 1;
}
}
static size_t
long_opt_strlen( ARGPARSE_OPTS *o )
{
size_t n = strlen (o->long_opt);
if ( o->description && *o->description == '|' )
{
const char *s;
#ifdef JNLIB_NEED_UTF8CONV
int is_utf8 = is_native_utf8 ();
#endif
s=o->description+1;
if ( *s != '=' )
n++;
/* For a (mostly) correct length calculation we exclude
continuation bytes (10xxxxxx) if we are on a native utf8
terminal. */
for (; *s && *s != '|'; s++ )
#ifdef JNLIB_NEED_UTF8CONV
if ( is_utf8 && (*s&0xc0) != 0x80 )
#endif
n++;
}
return n;
}
/****************
* Print formatted help. The description string has some special
* meanings:
* - A description string which is "@" suppresses help output for
* this option
* - a description,ine which starts with a '@' and is followed by
* any other characters is printed as is; this may be used for examples
* ans such.
* - A description which starts with a '|' outputs the string between this
* bar and the next one as arguments of the long option.
*/
static void
show_help (ARGPARSE_OPTS *opts, unsigned int flags)
{
const char *s;
char tmp[2];
show_version ();
writestrings (0, "\n", NULL);
s = strusage (42);
if (s && *s == '1')
{
s = strusage (40);
writestrings (1, s, NULL);
if (*s && s[strlen(s)] != '\n')
writestrings (1, "\n", NULL);
}
s = strusage(41);
writestrings (0, s, "\n", NULL);
if ( opts[0].description )
{
/* Auto format the option description. */
int i,j, indent;
/* Get max. length of long options. */
for (i=indent=0; opts[i].short_opt; i++ )
{
if ( opts[i].long_opt )
if ( !opts[i].description || *opts[i].description != '@' )
if ( (j=long_opt_strlen(opts+i)) > indent && j < 35 )
indent = j;
}
/* Example: " -v, --verbose Viele Sachen ausgeben" */
indent += 10;
if ( *opts[0].description != '@' )
writestrings (0, "Options:", "\n", NULL);
for (i=0; opts[i].short_opt; i++ )
{
s = map_static_macro_string (_( opts[i].description ));
if ( s && *s== '@' && !s[1] ) /* Hide this line. */
continue;
if ( s && *s == '@' ) /* Unindented comment only line. */
{
for (s++; *s; s++ )
{
if ( *s == '\n' )
{
if( s[1] )
writestrings (0, "\n", NULL);
}
else
{
tmp[0] = *s;
tmp[1] = 0;
writestrings (0, tmp, NULL);
}
}
writestrings (0, "\n", NULL);
continue;
}
j = 3;
if ( opts[i].short_opt < 256 )
{
tmp[0] = opts[i].short_opt;
tmp[1] = 0;
writestrings (0, " -", tmp, NULL );
if ( !opts[i].long_opt )
{
if (s && *s == '|' )
{
writestrings (0, " ", NULL); j++;
for (s++ ; *s && *s != '|'; s++, j++ )
{
tmp[0] = *s;
tmp[1] = 0;
writestrings (0, tmp, NULL);
}
if ( *s )
s++;
}
}
}
else
writestrings (0, " ", NULL);
if ( opts[i].long_opt )
{
tmp[0] = opts[i].short_opt < 256?',':' ';
tmp[1] = 0;
j += writestrings (0, tmp, " --", opts[i].long_opt, NULL);
if (s && *s == '|' )
{
if ( *++s != '=' )
{
writestrings (0, " ", NULL);
j++;
}
for ( ; *s && *s != '|'; s++, j++ )
{
tmp[0] = *s;
tmp[1] = 0;
writestrings (0, tmp, NULL);
}
if ( *s )
s++;
}
writestrings (0, " ", NULL);
j += 3;
}
for (;j < indent; j++ )
writestrings (0, " ", NULL);
if ( s )
{
if ( *s && j > indent )
{
writestrings (0, "\n", NULL);
for (j=0;j < indent; j++ )
writestrings (0, " ", NULL);
}
for (; *s; s++ )
{
if ( *s == '\n' )
{
if ( s[1] )
{
writestrings (0, "\n", NULL);
for (j=0; j < indent; j++ )
writestrings (0, " ", NULL);
}
}
else
{
tmp[0] = *s;
tmp[1] = 0;
writestrings (0, tmp, NULL);
}
}
}
writestrings (0, "\n", NULL);
}
if ( (flags & ARGPARSE_FLAG_ONEDASH) )
writestrings (0, "\n(A single dash may be used "
"instead of the double ones)\n", NULL);
}
if ( (s=strusage(19)) )
{
writestrings (0, "\n", NULL);
writestrings (0, s, NULL);
}
flushstrings (0);
exit(0);
}
static void
show_version ()
{
const char *s;
int i;
/* Version line. */
writestrings (0, strusage (11), NULL);
if ((s=strusage (12)))
writestrings (0, " (", s, ")", NULL);
writestrings (0, " ", strusage (13), "\n", NULL);
/* Additional version lines. */
for (i=20; i < 30; i++)
if ((s=strusage (i)))
writestrings (0, s, "\n", NULL);
/* Copyright string. */
if ((s=strusage (14)))
writestrings (0, s, "\n", NULL);
/* Licence string. */
if( (s=strusage (10)) )
writestrings (0, s, "\n", NULL);
/* Copying conditions. */
if ( (s=strusage(15)) )
writestrings (0, s, NULL);
/* Thanks. */
if ((s=strusage(18)))
writestrings (0, s, NULL);
/* Additional program info. */
for (i=30; i < 40; i++ )
if ( (s=strusage (i)) )
writestrings (0, s, NULL);
flushstrings (0);
}
void
usage (int level)
{
const char *p;
if (!level)
{
writestrings (1, strusage(11), " ", strusage(13), "; ",
strusage (14), "\n", NULL);
flushstrings (1);
}
else if (level == 1)
{
p = strusage (40);
writestrings (1, p, NULL);
if (*p && p[strlen(p)] != '\n')
writestrings (1, "\n", NULL);
exit (2);
}
else if (level == 2)
{
p = strusage (42);
if (p && *p == '1')
{
p = strusage (40);
writestrings (1, p, NULL);
if (*p && p[strlen(p)] != '\n')
writestrings (1, "\n", NULL);
}
writestrings (0, strusage(41), "\n", NULL);
exit (0);
}
}
/* Level
* 0: Print copyright string to stderr
* 1: Print a short usage hint to stderr and terminate
* 2: Print a long usage hint to stdout and terminate
* 10: Return license info string
* 11: Return the name of the program
* 12: Return optional name of package which includes this program.
* 13: version string
* 14: copyright string
* 15: Short copying conditions (with LFs)
* 16: Long copying conditions (with LFs)
* 17: Optional printable OS name
* 18: Optional thanks list (with LFs)
* 19: Bug report info
*20..29: Additional lib version strings.
*30..39: Additional program info (with LFs)
* 40: short usage note (with LF)
* 41: long usage note (with LF)
* 42: Flag string:
* First char is '1':
* The short usage notes needs to be printed
* before the long usage note.
*/
const char *
strusage( int level )
{
const char *p = strusage_handler? strusage_handler(level) : NULL;
if ( p )
return map_static_macro_string (p);
switch ( level )
{
case 10:
#if ARGPARSE_GPL_VERSION == 3
p = ("License GPLv3+: GNU GPL version 3 or later "
"<http://gnu.org/licenses/gpl.html>");
#else
p = ("License GPLv2+: GNU GPL version 2 or later "
"<http://gnu.org/licenses/>");
#endif
break;
case 11: p = "foo"; break;
case 13: p = "0.0"; break;
case 14: p = ARGPARSE_CRIGHT_STR; break;
case 15: p =
"This is free software: you are free to change and redistribute it.\n"
"There is NO WARRANTY, to the extent permitted by law.\n";
break;
case 16: p =
"This is free software; you can redistribute it and/or modify\n"
"it under the terms of the GNU General Public License as published by\n"
"the Free Software Foundation; either version "
ARGPARSE_STR2(ARGPARSE_GPL_VERSION)
" of the License, or\n"
"(at your option) any later version.\n\n"
"It is distributed in the hope that it will be useful,\n"
"but WITHOUT ANY WARRANTY; without even the implied warranty of\n"
"MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the\n"
"GNU General Public License for more details.\n\n"
"You should have received a copy of the GNU General Public License\n"
"along with this software. If not, see <http://www.gnu.org/licenses/>.\n";
break;
case 40: /* short and long usage */
case 41: p = ""; break;
}
return p;
}
/* Set the usage handler. This function is basically a constructor. */
void
set_strusage ( const char *(*f)( int ) )
{
strusage_handler = f;
}
#ifdef TEST
static struct {
int verbose;
int debug;
char *outfile;
char *crf;
int myopt;
int echo;
int a_long_one;
} opt;
int
main(int argc, char **argv)
{
ARGPARSE_OPTS opts[] = {
ARGPARSE_x('v', "verbose", NONE, 0, "Laut sein"),
ARGPARSE_s_n('e', "echo" , ("Zeile ausgeben, damit wir sehen, "
"was wir eingegeben haben")),
ARGPARSE_s_n('d', "debug", "Debug\nfalls mal etwas\nschief geht"),
ARGPARSE_s_s('o', "output", 0 ),
ARGPARSE_o_s('c', "cross-ref", "cross-reference erzeugen\n" ),
/* Note that on a non-utf8 terminal the ß might garble the output. */
ARGPARSE_s_n('s', "street","|Straße|set the name of the street to Straße"),
ARGPARSE_o_i('m', "my-option", 0),
ARGPARSE_s_n(500, "a-long-option", 0 ),
ARGPARSE_end()
};
ARGPARSE_ARGS pargs = { &argc, &argv, (ARGPARSE_FLAG_ALL
| ARGPARSE_FLAG_MIXED
| ARGPARSE_FLAG_ONEDASH) };
int i;
while (arg_parse (&pargs, opts))
{
switch (pargs.r_opt)
{
case ARGPARSE_IS_ARG :
printf ("arg='%s'\n", pargs.r.ret_str);
break;
case 'v': opt.verbose++; break;
case 'e': opt.echo++; break;
case 'd': opt.debug++; break;
case 'o': opt.outfile = pargs.r.ret_str; break;
case 'c': opt.crf = pargs.r_type? pargs.r.ret_str:"a.crf"; break;
case 'm': opt.myopt = pargs.r_type? pargs.r.ret_int : 1; break;
case 500: opt.a_long_one++; break;
default : pargs.err = ARGPARSE_PRINT_WARNING; break;
}
}
for (i=0; i < argc; i++ )
printf ("%3d -> (%s)\n", i, argv[i] );
puts ("Options:");
if (opt.verbose)
printf (" verbose=%d\n", opt.verbose );
if (opt.debug)
printf (" debug=%d\n", opt.debug );
if (opt.outfile)
printf (" outfile='%s'\n", opt.outfile );
if (opt.crf)
printf (" crffile='%s'\n", opt.crf );
if (opt.myopt)
printf (" myopt=%d\n", opt.myopt );
if (opt.a_long_one)
printf (" a-long-one=%d\n", opt.a_long_one );
if (opt.echo)
printf (" echo=%d\n", opt.echo );
return 0;
}
#endif /*TEST*/
/**** bottom of file ****/
diff --git a/pinentry/pinentry.c b/pinentry/pinentry.c
index 3131b1e..4a0f859 100644
--- a/pinentry/pinentry.c
+++ b/pinentry/pinentry.c
@@ -1,1590 +1,1590 @@
/* pinentry.c - The PIN entry support library
Copyright (C) 2002, 2003, 2007, 2008, 2010, 2015 g10 Code GmbH
This file is part of PINENTRY.
PINENTRY 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.
PINENTRY is distributed in the hope that it will be useful, but
WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program; if not, see <http://www.gnu.org/licenses/>.
*/
#ifdef HAVE_CONFIG_H
#include <config.h>
#endif
#ifndef HAVE_W32CE_SYSTEM
# include <errno.h>
#endif
#include <stdlib.h>
#include <string.h>
#include <unistd.h>
#include <assert.h>
#ifndef HAVE_W32CE_SYSTEM
# include <locale.h>
#endif
#ifdef HAVE_LANGINFO_H
#include <langinfo.h>
#endif
#include <limits.h>
#ifdef HAVE_W32CE_SYSTEM
# include <windows.h>
#endif
#if defined FALLBACK_CURSES || defined PINENTRY_CURSES || defined PINENTRY_GTK
#include <iconv.h>
#endif
#include <assuan.h>
#include "memory.h"
#include "secmem-util.h"
#include "argparse.h"
#include "pinentry.h"
#include "password-cache.h"
#ifdef INSIDE_EMACS
#include "pinentry-emacs.h"
#endif
#ifdef HAVE_W32CE_SYSTEM
#define getpid() GetCurrentProcessId ()
#endif
/* Keep the name of our program here. */
static char this_pgmname[50];
struct pinentry pinentry;
static void
pinentry_reset (int use_defaults)
{
/* GPG Agent sets these options once when it starts the pinentry.
Don't reset them. */
int grab = pinentry.grab;
char *ttyname = pinentry.ttyname;
char *ttytype = pinentry.ttytype;
char *lc_ctype = pinentry.lc_ctype;
char *lc_messages = pinentry.lc_messages;
int allow_external_password_cache = pinentry.allow_external_password_cache;
char *default_ok = pinentry.default_ok;
char *default_cancel = pinentry.default_cancel;
char *default_prompt = pinentry.default_prompt;
char *default_pwmngr = pinentry.default_pwmngr;
char *default_cf_visi = pinentry.default_cf_visi;
char *default_tt_visi = pinentry.default_tt_visi;
char *default_tt_hide = pinentry.default_tt_hide;
char *touch_file = pinentry.touch_file;
/* These options are set from the command line. Don't reset
them. */
int debug = pinentry.debug;
char *display = pinentry.display;
int parent_wid = pinentry.parent_wid;
pinentry_color_t color_fg = pinentry.color_fg;
int color_fg_bright = pinentry.color_fg_bright;
pinentry_color_t color_bg = pinentry.color_bg;
pinentry_color_t color_so = pinentry.color_so;
int color_so_bright = pinentry.color_so_bright;
- int timout = pinentry.timeout;
+ int timeout = pinentry.timeout;
char *invisible_char = pinentry.invisible_char;
/* Free any allocated memory. */
if (use_defaults)
{
free (pinentry.ttyname);
free (pinentry.ttytype);
free (pinentry.lc_ctype);
free (pinentry.lc_messages);
free (pinentry.default_ok);
free (pinentry.default_cancel);
free (pinentry.default_prompt);
free (pinentry.default_pwmngr);
free (pinentry.default_cf_visi);
free (pinentry.default_tt_visi);
free (pinentry.default_tt_hide);
free (pinentry.touch_file);
free (pinentry.display);
}
free (pinentry.title);
free (pinentry.description);
free (pinentry.error);
free (pinentry.prompt);
free (pinentry.ok);
free (pinentry.notok);
free (pinentry.cancel);
secmem_free (pinentry.pin);
free (pinentry.repeat_passphrase);
free (pinentry.repeat_error_string);
free (pinentry.quality_bar);
free (pinentry.quality_bar_tt);
free (pinentry.keyinfo);
free (pinentry.specific_err_info);
/* Reset the pinentry structure. */
memset (&pinentry, 0, sizeof (pinentry));
/* Restore options without a default we want to preserve. */
pinentry.invisible_char = invisible_char;
/* Restore other options or set defaults. */
if (use_defaults)
{
/* Pinentry timeout in seconds. */
pinentry.timeout = 60;
/* Global grab. */
pinentry.grab = 1;
pinentry.color_fg = PINENTRY_COLOR_DEFAULT;
pinentry.color_fg_bright = 0;
pinentry.color_bg = PINENTRY_COLOR_DEFAULT;
pinentry.color_so = PINENTRY_COLOR_DEFAULT;
pinentry.color_so_bright = 0;
}
else
/* Restore the options. */
{
pinentry.grab = grab;
pinentry.ttyname = ttyname;
pinentry.ttytype = ttytype;
pinentry.lc_ctype = lc_ctype;
pinentry.lc_messages = lc_messages;
pinentry.allow_external_password_cache = allow_external_password_cache;
pinentry.default_ok = default_ok;
pinentry.default_cancel = default_cancel;
pinentry.default_prompt = default_prompt;
pinentry.default_pwmngr = default_pwmngr;
pinentry.default_cf_visi = default_cf_visi;
pinentry.default_tt_visi = default_tt_visi;
pinentry.default_tt_hide = default_tt_hide;
pinentry.touch_file = touch_file;
pinentry.debug = debug;
pinentry.display = display;
pinentry.parent_wid = parent_wid;
pinentry.color_fg = color_fg;
pinentry.color_fg_bright = color_fg_bright;
pinentry.color_bg = color_bg;
pinentry.color_so = color_so;
pinentry.color_so_bright = color_so_bright;
- pinentry.timeout = timout;
+ pinentry.timeout = timeout;
}
}
static gpg_error_t
pinentry_assuan_reset_handler (assuan_context_t ctx, char *line)
{
(void)ctx;
(void)line;
pinentry_reset (0);
return 0;
}
static int lc_ctype_unknown_warning = 0;
#if defined FALLBACK_CURSES || defined PINENTRY_CURSES || defined PINENTRY_GTK
char *
pinentry_utf8_to_local (const char *lc_ctype, const char *text)
{
iconv_t cd;
const char *input = text;
size_t input_len = strlen (text) + 1;
char *output;
size_t output_len;
char *output_buf;
size_t processed;
char *old_ctype;
char *target_encoding;
/* If no locale setting could be determined, simply copy the
string. */
if (!lc_ctype)
{
if (! lc_ctype_unknown_warning)
{
fprintf (stderr, "%s: no LC_CTYPE known - assuming UTF-8\n",
this_pgmname);
lc_ctype_unknown_warning = 1;
}
return strdup (text);
}
old_ctype = strdup (setlocale (LC_CTYPE, NULL));
if (!old_ctype)
return NULL;
setlocale (LC_CTYPE, lc_ctype);
target_encoding = nl_langinfo (CODESET);
if (!target_encoding)
target_encoding = "?";
setlocale (LC_CTYPE, old_ctype);
free (old_ctype);
/* This is overkill, but simplifies the iconv invocation greatly. */
output_len = input_len * MB_LEN_MAX;
output_buf = output = malloc (output_len);
if (!output)
return NULL;
cd = iconv_open (target_encoding, "UTF-8");
if (cd == (iconv_t) -1)
{
fprintf (stderr, "%s: can't convert from UTF-8 to %s: %s\n",
this_pgmname, target_encoding, strerror (errno));
free (output_buf);
return NULL;
}
processed = iconv (cd, (ICONV_CONST char **)&input, &input_len,
&output, &output_len);
iconv_close (cd);
if (processed == (size_t) -1 || input_len)
{
fprintf (stderr, "%s: error converting from UTF-8 to %s: %s\n",
this_pgmname, target_encoding, strerror (errno));
free (output_buf);
return NULL;
}
return output_buf;
}
/* Convert TEXT which is encoded according to LC_CTYPE to UTF-8. With
SECURE set to true, use secure memory for the returned buffer.
Return NULL on error. */
char *
pinentry_local_to_utf8 (char *lc_ctype, char *text, int secure)
{
char *old_ctype;
char *source_encoding;
iconv_t cd;
const char *input = text;
size_t input_len = strlen (text) + 1;
char *output;
size_t output_len;
char *output_buf;
size_t processed;
/* If no locale setting could be determined, simply copy the
string. */
if (!lc_ctype)
{
if (! lc_ctype_unknown_warning)
{
fprintf (stderr, "%s: no LC_CTYPE known - assuming UTF-8\n",
this_pgmname);
lc_ctype_unknown_warning = 1;
}
output_buf = secure? secmem_malloc (input_len) : malloc (input_len);
if (output_buf)
strcpy (output_buf, input);
return output_buf;
}
old_ctype = strdup (setlocale (LC_CTYPE, NULL));
if (!old_ctype)
return NULL;
setlocale (LC_CTYPE, lc_ctype);
source_encoding = nl_langinfo (CODESET);
setlocale (LC_CTYPE, old_ctype);
free (old_ctype);
/* This is overkill, but simplifies the iconv invocation greatly. */
output_len = input_len * MB_LEN_MAX;
output_buf = output = secure? secmem_malloc (output_len):malloc (output_len);
if (!output)
return NULL;
cd = iconv_open ("UTF-8", source_encoding);
if (cd == (iconv_t) -1)
{
fprintf (stderr, "%s: can't convert from %s to UTF-8: %s\n",
this_pgmname, source_encoding? source_encoding : "?",
strerror (errno));
if (secure)
secmem_free (output_buf);
else
free (output_buf);
return NULL;
}
processed = iconv (cd, (ICONV_CONST char **)&input, &input_len,
&output, &output_len);
iconv_close (cd);
if (processed == (size_t) -1 || input_len)
{
fprintf (stderr, "%s: error converting from %s to UTF-8: %s\n",
this_pgmname, source_encoding? source_encoding : "?",
strerror (errno));
if (secure)
secmem_free (output_buf);
else
free (output_buf);
return NULL;
}
return output_buf;
}
#endif
/* Copy TEXT or TEXTLEN to BUFFER and escape as required. Return a
pointer to the end of the new buffer. Note that BUFFER must be
large enough to keep the entire text; allocataing it 3 times of
TEXTLEN is sufficient. */
static char *
copy_and_escape (char *buffer, const void *text, size_t textlen)
{
int i;
const unsigned char *s = (unsigned char *)text;
char *p = buffer;
for (i=0; i < textlen; i++)
{
if (s[i] < ' ' || s[i] == '+')
{
snprintf (p, 4, "%%%02X", s[i]);
p += 3;
}
else if (s[i] == ' ')
*p++ = '+';
else
*p++ = s[i];
}
return p;
}
/* Run a quality inquiry for PASSPHRASE of LENGTH. (We need LENGTH
because not all backends might be able to return a proper
C-string.). Returns: A value between -100 and 100 to give an
estimate of the passphrase's quality. Negative values are use if
the caller won't even accept that passphrase. Note that we expect
just one data line which should not be escaped in any represent a
numeric signed decimal value. Extra data is currently ignored but
should not be send at all. */
int
pinentry_inq_quality (pinentry_t pin, const char *passphrase, size_t length)
{
assuan_context_t ctx = pin->ctx_assuan;
const char prefix[] = "INQUIRE QUALITY ";
char *command;
char *line;
size_t linelen;
int gotvalue = 0;
int value = 0;
int rc;
if (!ctx)
return 0; /* Can't run the callback. */
if (length > 300)
length = 300; /* Limit so that it definitely fits into an Assuan
line. */
command = secmem_malloc (strlen (prefix) + 3*length + 1);
if (!command)
return 0;
strcpy (command, prefix);
copy_and_escape (command + strlen(command), passphrase, length);
rc = assuan_write_line (ctx, command);
secmem_free (command);
if (rc)
{
fprintf (stderr, "ASSUAN WRITE LINE failed: rc=%d\n", rc);
return 0;
}
for (;;)
{
do
{
rc = assuan_read_line (ctx, &line, &linelen);
if (rc)
{
fprintf (stderr, "ASSUAN READ LINE failed: rc=%d\n", rc);
return 0;
}
}
while (*line == '#' || !linelen);
if (line[0] == 'E' && line[1] == 'N' && line[2] == 'D'
&& (!line[3] || line[3] == ' '))
break; /* END command received*/
if (line[0] == 'C' && line[1] == 'A' && line[2] == 'N'
&& (!line[3] || line[3] == ' '))
break; /* CAN command received*/
if (line[0] == 'E' && line[1] == 'R' && line[2] == 'R'
&& (!line[3] || line[3] == ' '))
break; /* ERR command received*/
if (line[0] != 'D' || line[1] != ' ' || linelen < 3 || gotvalue)
continue;
gotvalue = 1;
value = atoi (line+2);
}
if (value < -100)
value = -100;
else if (value > 100)
value = 100;
return value;
}
/* Try to make room for at least LEN bytes in the pinentry. Returns
new buffer on success and 0 on failure or when the old buffer is
sufficient. */
char *
pinentry_setbufferlen (pinentry_t pin, int len)
{
char *newp;
if (pin->pin_len)
assert (pin->pin);
else
assert (!pin->pin);
if (len < 2048)
len = 2048;
if (len <= pin->pin_len)
return pin->pin;
newp = secmem_realloc (pin->pin, len);
if (newp)
{
pin->pin = newp;
pin->pin_len = len;
}
else
{
secmem_free (pin->pin);
pin->pin = 0;
pin->pin_len = 0;
}
return newp;
}
static void
pinentry_setbuffer_clear (pinentry_t pin)
{
if (! pin->pin)
{
assert (pin->pin_len == 0);
return;
}
assert (pin->pin_len > 0);
secmem_free (pin->pin);
pin->pin = NULL;
pin->pin_len = 0;
}
static void
pinentry_setbuffer_init (pinentry_t pin)
{
pinentry_setbuffer_clear (pin);
pinentry_setbufferlen (pin, 0);
}
/* passphrase better be alloced with secmem_alloc. */
void
pinentry_setbuffer_use (pinentry_t pin, char *passphrase, int len)
{
if (! passphrase)
{
assert (len == 0);
pinentry_setbuffer_clear (pin);
return;
}
if (passphrase && len == 0)
len = strlen (passphrase) + 1;
if (pin->pin)
secmem_free (pin->pin);
pin->pin = passphrase;
pin->pin_len = len;
}
static struct assuan_malloc_hooks assuan_malloc_hooks = {
secmem_malloc, secmem_realloc, secmem_free
};
/* Initialize the secure memory subsystem, drop privileges and return.
Must be called early. */
void
pinentry_init (const char *pgmname)
{
/* Store away our name. */
if (strlen (pgmname) > sizeof this_pgmname - 2)
abort ();
strcpy (this_pgmname, pgmname);
gpgrt_check_version (NULL);
/* Initialize secure memory. 1 is too small, so the default size
will be used. */
secmem_init (1);
secmem_set_flags (SECMEM_WARN);
drop_privs ();
if (atexit (secmem_term))
{
/* FIXME: Could not register at-exit function, bail out. */
}
assuan_set_malloc_hooks (&assuan_malloc_hooks);
}
/* Simple test to check whether DISPLAY is set or the option --display
was given. Used to decide whether the GUI or curses should be
initialized. */
int
pinentry_have_display (int argc, char **argv)
{
#ifndef HAVE_W32CE_SYSTEM
const char *s;
s = getenv ("DISPLAY");
if (s && *s)
return 1;
#endif
for (; argc; argc--, argv++)
if (!strcmp (*argv, "--display") || !strncmp (*argv, "--display=", 10))
return 1;
return 0;
}
/* Print usage information and and provide strings for help. */
static const char *
my_strusage( int level )
{
const char *p;
switch (level)
{
case 11: p = this_pgmname; break;
case 12: p = "pinentry"; break;
case 13: p = PACKAGE_VERSION; break;
case 14: p = "Copyright (C) 2015 g10 Code GmbH"; break;
case 19: p = "Please report bugs to <" PACKAGE_BUGREPORT ">.\n"; break;
case 1:
case 40:
{
static char *str;
if (!str)
{
size_t n = 50 + strlen (this_pgmname);
str = malloc (n);
if (str)
snprintf (str, n, "Usage: %s [options] (-h for help)",
this_pgmname);
}
p = str;
}
break;
case 41:
p = "Ask securely for a secret and print it to stdout.";
break;
case 42:
p = "1"; /* Flag print 40 as part of 41. */
break;
default: p = NULL; break;
}
return p;
}
char *
parse_color (char *arg, pinentry_color_t *color_p, int *bright_p)
{
static struct
{
const char *name;
pinentry_color_t color;
} colors[] = { { "none", PINENTRY_COLOR_NONE },
{ "default", PINENTRY_COLOR_DEFAULT },
{ "black", PINENTRY_COLOR_BLACK },
{ "red", PINENTRY_COLOR_RED },
{ "green", PINENTRY_COLOR_GREEN },
{ "yellow", PINENTRY_COLOR_YELLOW },
{ "blue", PINENTRY_COLOR_BLUE },
{ "magenta", PINENTRY_COLOR_MAGENTA },
{ "cyan", PINENTRY_COLOR_CYAN },
{ "white", PINENTRY_COLOR_WHITE } };
int i;
char *new_arg;
pinentry_color_t color = PINENTRY_COLOR_DEFAULT;
if (!arg)
return NULL;
new_arg = strchr (arg, ',');
if (new_arg)
new_arg++;
if (bright_p)
{
const char *bname[] = { "bright-", "bright", "bold-", "bold" };
*bright_p = 0;
for (i = 0; i < sizeof (bname) / sizeof (bname[0]); i++)
if (!strncasecmp (arg, bname[i], strlen (bname[i])))
{
*bright_p = 1;
arg += strlen (bname[i]);
}
}
for (i = 0; i < sizeof (colors) / sizeof (colors[0]); i++)
if (!strncasecmp (arg, colors[i].name, strlen (colors[i].name)))
color = colors[i].color;
*color_p = color;
return new_arg;
}
/* Parse the command line options. May exit the program if only help
or version output is requested. */
void
pinentry_parse_opts (int argc, char *argv[])
{
static ARGPARSE_OPTS opts[] = {
ARGPARSE_s_n('d', "debug", "Turn on debugging output"),
ARGPARSE_s_s('D', "display", "|DISPLAY|Set the X display"),
ARGPARSE_s_s('T', "ttyname", "|FILE|Set the tty terminal node name"),
ARGPARSE_s_s('N', "ttytype", "|NAME|Set the tty terminal type"),
ARGPARSE_s_s('C', "lc-ctype", "|STRING|Set the tty LC_CTYPE value"),
ARGPARSE_s_s('M', "lc-messages", "|STRING|Set the tty LC_MESSAGES value"),
ARGPARSE_s_i('o', "timeout",
"|SECS|Timeout waiting for input after this many seconds"),
ARGPARSE_s_n('g', "no-global-grab",
"Grab keyboard only while window is focused"),
ARGPARSE_s_u('W', "parent-wid", "Parent window ID (for positioning)"),
ARGPARSE_s_s('c', "colors", "|STRING|Set custom colors for ncurses"),
ARGPARSE_end()
};
ARGPARSE_ARGS pargs = { &argc, &argv, 0 };
set_strusage (my_strusage);
pinentry_reset (1);
while (arg_parse (&pargs, opts))
{
switch (pargs.r_opt)
{
case 'd':
pinentry.debug = 1;
break;
case 'g':
pinentry.grab = 0;
break;
case 'D':
/* Note, this is currently not used because the GUI engine
has already been initialized when parsing these options. */
pinentry.display = strdup (pargs.r.ret_str);
if (!pinentry.display)
{
#ifndef HAVE_W32CE_SYSTEM
fprintf (stderr, "%s: %s\n", this_pgmname, strerror (errno));
#endif
exit (EXIT_FAILURE);
}
break;
case 'T':
pinentry.ttyname = strdup (pargs.r.ret_str);
if (!pinentry.ttyname)
{
#ifndef HAVE_W32CE_SYSTEM
fprintf (stderr, "%s: %s\n", this_pgmname, strerror (errno));
#endif
exit (EXIT_FAILURE);
}
break;
case 'N':
pinentry.ttytype = strdup (pargs.r.ret_str);
if (!pinentry.ttytype)
{
#ifndef HAVE_W32CE_SYSTEM
fprintf (stderr, "%s: %s\n", this_pgmname, strerror (errno));
#endif
exit (EXIT_FAILURE);
}
break;
case 'C':
pinentry.lc_ctype = strdup (pargs.r.ret_str);
if (!pinentry.lc_ctype)
{
#ifndef HAVE_W32CE_SYSTEM
fprintf (stderr, "%s: %s\n", this_pgmname, strerror (errno));
#endif
exit (EXIT_FAILURE);
}
break;
case 'M':
pinentry.lc_messages = strdup (pargs.r.ret_str);
if (!pinentry.lc_messages)
{
#ifndef HAVE_W32CE_SYSTEM
fprintf (stderr, "%s: %s\n", this_pgmname, strerror (errno));
#endif
exit (EXIT_FAILURE);
}
break;
case 'W':
pinentry.parent_wid = pargs.r.ret_ulong;
break;
case 'c':
{
char *tmpstr = pargs.r.ret_str;
tmpstr = parse_color (tmpstr, &pinentry.color_fg,
&pinentry.color_fg_bright);
tmpstr = parse_color (tmpstr, &pinentry.color_bg, NULL);
tmpstr = parse_color (tmpstr, &pinentry.color_so,
&pinentry.color_so_bright);
}
break;
case 'o':
pinentry.timeout = pargs.r.ret_int;
break;
default:
pargs.err = ARGPARSE_PRINT_WARNING;
break;
}
}
}
static gpg_error_t
option_handler (assuan_context_t ctx, const char *key, const char *value)
{
(void)ctx;
if (!strcmp (key, "no-grab") && !*value)
pinentry.grab = 0;
else if (!strcmp (key, "grab") && !*value)
pinentry.grab = 1;
else if (!strcmp (key, "debug-wait"))
{
#ifndef HAVE_W32CE_SYSTEM
fprintf (stderr, "%s: waiting for debugger - my pid is %u ...\n",
this_pgmname, (unsigned int) getpid());
sleep (*value?atoi (value):5);
fprintf (stderr, "%s: ... okay\n", this_pgmname);
#endif
}
else if (!strcmp (key, "display"))
{
if (pinentry.display)
free (pinentry.display);
pinentry.display = strdup (value);
if (!pinentry.display)
return gpg_error_from_syserror ();
}
else if (!strcmp (key, "ttyname"))
{
if (pinentry.ttyname)
free (pinentry.ttyname);
pinentry.ttyname = strdup (value);
if (!pinentry.ttyname)
return gpg_error_from_syserror ();
}
else if (!strcmp (key, "ttytype"))
{
if (pinentry.ttytype)
free (pinentry.ttytype);
pinentry.ttytype = strdup (value);
if (!pinentry.ttytype)
return gpg_error_from_syserror ();
}
else if (!strcmp (key, "lc-ctype"))
{
if (pinentry.lc_ctype)
free (pinentry.lc_ctype);
pinentry.lc_ctype = strdup (value);
if (!pinentry.lc_ctype)
return gpg_error_from_syserror ();
}
else if (!strcmp (key, "lc-messages"))
{
if (pinentry.lc_messages)
free (pinentry.lc_messages);
pinentry.lc_messages = strdup (value);
if (!pinentry.lc_messages)
return gpg_error_from_syserror ();
}
else if (!strcmp (key, "parent-wid"))
{
pinentry.parent_wid = atoi (value);
/* FIXME: Use strtol and add some error handling. */
}
else if (!strcmp (key, "touch-file"))
{
if (pinentry.touch_file)
free (pinentry.touch_file);
pinentry.touch_file = strdup (value);
if (!pinentry.touch_file)
return gpg_error_from_syserror ();
}
else if (!strcmp (key, "default-ok"))
{
pinentry.default_ok = strdup (value);
if (!pinentry.default_ok)
return gpg_error_from_syserror ();
}
else if (!strcmp (key, "default-cancel"))
{
pinentry.default_cancel = strdup (value);
if (!pinentry.default_cancel)
return gpg_error_from_syserror ();
}
else if (!strcmp (key, "default-prompt"))
{
pinentry.default_prompt = strdup (value);
if (!pinentry.default_prompt)
return gpg_error_from_syserror ();
}
else if (!strcmp (key, "default-pwmngr"))
{
pinentry.default_pwmngr = strdup (value);
if (!pinentry.default_pwmngr)
return gpg_error_from_syserror ();
}
else if (!strcmp (key, "default-cf-visi"))
{
pinentry.default_cf_visi = strdup (value);
if (!pinentry.default_cf_visi)
return gpg_error_from_syserror ();
}
else if (!strcmp (key, "default-tt-visi"))
{
pinentry.default_tt_visi = strdup (value);
if (!pinentry.default_tt_visi)
return gpg_error_from_syserror ();
}
else if (!strcmp (key, "default-tt-hide"))
{
pinentry.default_tt_hide = strdup (value);
if (!pinentry.default_tt_hide)
return gpg_error_from_syserror ();
}
else if (!strcmp (key, "allow-external-password-cache") && !*value)
{
pinentry.allow_external_password_cache = 1;
pinentry.tried_password_cache = 0;
}
else if (!strcmp (key, "allow-emacs-prompt") && !*value)
{
#ifdef INSIDE_EMACS
pinentry_enable_emacs_cmd_handler ();
#endif
}
else if (!strcmp (key, "invisible-char"))
{
if (pinentry.invisible_char)
free (pinentry.invisible_char);
pinentry.invisible_char = strdup (value);
if (!pinentry.invisible_char)
return gpg_error_from_syserror ();
}
else
return gpg_error (GPG_ERR_UNKNOWN_OPTION);
return 0;
}
/* Note, that it is sufficient to allocate the target string D as
long as the source string S, i.e.: strlen(s)+1; */
static void
strcpy_escaped (char *d, const char *s)
{
while (*s)
{
if (*s == '%' && s[1] && s[2])
{
s++;
*d++ = xtoi_2 ( s);
s += 2;
}
else
*d++ = *s++;
}
*d = 0;
}
static void
write_status_error (assuan_context_t ctx, pinentry_t pe)
{
char buf[500];
const char *pgm;
pgm = strchr (this_pgmname, '-');
if (pgm && pgm[1])
pgm++;
else
pgm = this_pgmname;
snprintf (buf, sizeof buf, "%s.%s %d %s",
pgm,
pe->specific_err_loc? pe->specific_err_loc : "?",
pe->specific_err,
pe->specific_err_info? pe->specific_err_info : "");
assuan_write_status (ctx, "ERROR", buf);
}
static gpg_error_t
cmd_setdesc (assuan_context_t ctx, char *line)
{
char *newd;
(void)ctx;
newd = malloc (strlen (line) + 1);
if (!newd)
return gpg_error_from_syserror ();
strcpy_escaped (newd, line);
if (pinentry.description)
free (pinentry.description);
pinentry.description = newd;
return 0;
}
static gpg_error_t
cmd_setprompt (assuan_context_t ctx, char *line)
{
char *newp;
(void)ctx;
newp = malloc (strlen (line) + 1);
if (!newp)
return gpg_error_from_syserror ();
strcpy_escaped (newp, line);
if (pinentry.prompt)
free (pinentry.prompt);
pinentry.prompt = newp;
return 0;
}
/* The data provided at LINE may be used by pinentry implementations
to identify a key for caching strategies of its own. The empty
string and --clear mean that the key does not have a stable
identifier. */
static gpg_error_t
cmd_setkeyinfo (assuan_context_t ctx, char *line)
{
(void)ctx;
if (pinentry.keyinfo)
free (pinentry.keyinfo);
if (*line && strcmp(line, "--clear") != 0)
pinentry.keyinfo = strdup (line);
else
pinentry.keyinfo = NULL;
return 0;
}
static gpg_error_t
cmd_setrepeat (assuan_context_t ctx, char *line)
{
char *p;
(void)ctx;
p = malloc (strlen (line) + 1);
if (!p)
return gpg_error_from_syserror ();
strcpy_escaped (p, line);
free (pinentry.repeat_passphrase);
pinentry.repeat_passphrase = p;
return 0;
}
static gpg_error_t
cmd_setrepeaterror (assuan_context_t ctx, char *line)
{
char *p;
(void)ctx;
p = malloc (strlen (line) + 1);
if (!p)
return gpg_error_from_syserror ();
strcpy_escaped (p, line);
free (pinentry.repeat_error_string);
pinentry.repeat_error_string = p;
return 0;
}
static gpg_error_t
cmd_seterror (assuan_context_t ctx, char *line)
{
char *newe;
(void)ctx;
newe = malloc (strlen (line) + 1);
if (!newe)
return gpg_error_from_syserror ();
strcpy_escaped (newe, line);
if (pinentry.error)
free (pinentry.error);
pinentry.error = newe;
return 0;
}
static gpg_error_t
cmd_setok (assuan_context_t ctx, char *line)
{
char *newo;
(void)ctx;
newo = malloc (strlen (line) + 1);
if (!newo)
return gpg_error_from_syserror ();
strcpy_escaped (newo, line);
if (pinentry.ok)
free (pinentry.ok);
pinentry.ok = newo;
return 0;
}
static gpg_error_t
cmd_setnotok (assuan_context_t ctx, char *line)
{
char *newo;
(void)ctx;
newo = malloc (strlen (line) + 1);
if (!newo)
return gpg_error_from_syserror ();
strcpy_escaped (newo, line);
if (pinentry.notok)
free (pinentry.notok);
pinentry.notok = newo;
return 0;
}
static gpg_error_t
cmd_setcancel (assuan_context_t ctx, char *line)
{
char *newc;
(void)ctx;
newc = malloc (strlen (line) + 1);
if (!newc)
return gpg_error_from_syserror ();
strcpy_escaped (newc, line);
if (pinentry.cancel)
free (pinentry.cancel);
pinentry.cancel = newc;
return 0;
}
static gpg_error_t
cmd_settimeout (assuan_context_t ctx, char *line)
{
(void)ctx;
if (line && *line)
pinentry.timeout = atoi (line);
return 0;
}
static gpg_error_t
cmd_settitle (assuan_context_t ctx, char *line)
{
char *newt;
(void)ctx;
newt = malloc (strlen (line) + 1);
if (!newt)
return gpg_error_from_syserror ();
strcpy_escaped (newt, line);
if (pinentry.title)
free (pinentry.title);
pinentry.title = newt;
return 0;
}
static gpg_error_t
cmd_setqualitybar (assuan_context_t ctx, char *line)
{
char *newval;
(void)ctx;
if (!*line)
line = "Quality:";
newval = malloc (strlen (line) + 1);
if (!newval)
return gpg_error_from_syserror ();
strcpy_escaped (newval, line);
if (pinentry.quality_bar)
free (pinentry.quality_bar);
pinentry.quality_bar = newval;
return 0;
}
/* Set the tooltip to be used for a quality bar. */
static gpg_error_t
cmd_setqualitybar_tt (assuan_context_t ctx, char *line)
{
char *newval;
(void)ctx;
if (*line)
{
newval = malloc (strlen (line) + 1);
if (!newval)
return gpg_error_from_syserror ();
strcpy_escaped (newval, line);
}
else
newval = NULL;
if (pinentry.quality_bar_tt)
free (pinentry.quality_bar_tt);
pinentry.quality_bar_tt = newval;
return 0;
}
static gpg_error_t
cmd_getpin (assuan_context_t ctx, char *line)
{
int result;
int set_prompt = 0;
int just_read_password_from_cache = 0;
(void)line;
pinentry_setbuffer_init (&pinentry);
if (!pinentry.pin)
return gpg_error (GPG_ERR_ENOMEM);
/* Try reading from the password cache. */
if (/* If repeat passphrase is set, then we don't want to read from
the cache. */
! pinentry.repeat_passphrase
/* Are we allowed to read from the cache? */
&& pinentry.allow_external_password_cache
&& pinentry.keyinfo
/* Only read from the cache if we haven't already tried it. */
&& ! pinentry.tried_password_cache
/* If the last read resulted in an error, then don't read from
the cache. */
&& ! pinentry.error)
{
char *password;
pinentry.tried_password_cache = 1;
password = password_cache_lookup (pinentry.keyinfo);
if (password)
/* There is a cached password. Try it. */
{
int len = strlen(password) + 1;
if (len > pinentry.pin_len)
len = pinentry.pin_len;
memcpy (pinentry.pin, password, len);
pinentry.pin[len] = '\0';
secmem_free (password);
pinentry.pin_from_cache = 1;
assuan_write_status (ctx, "PASSWORD_FROM_CACHE", "");
/* Result is the length of the password not including the
NUL terminator. */
result = len - 1;
just_read_password_from_cache = 1;
goto out;
}
}
/* The password was not cached (or we are not allowed to / cannot
use the cache). Prompt the user. */
pinentry.pin_from_cache = 0;
if (!pinentry.prompt)
{
pinentry.prompt = pinentry.default_prompt?pinentry.default_prompt:"PIN:";
set_prompt = 1;
}
pinentry.locale_err = 0;
pinentry.specific_err = 0;
pinentry.specific_err_loc = NULL;
free (pinentry.specific_err_info);
pinentry.specific_err_info = NULL;
pinentry.close_button = 0;
pinentry.repeat_okay = 0;
pinentry.one_button = 0;
pinentry.ctx_assuan = ctx;
result = (*pinentry_cmd_handler) (&pinentry);
pinentry.ctx_assuan = NULL;
if (pinentry.error)
{
free (pinentry.error);
pinentry.error = NULL;
}
if (pinentry.repeat_passphrase)
{
free (pinentry.repeat_passphrase);
pinentry.repeat_passphrase = NULL;
}
if (set_prompt)
pinentry.prompt = NULL;
pinentry.quality_bar = 0; /* Reset it after the command. */
if (pinentry.close_button)
assuan_write_status (ctx, "BUTTON_INFO", "close");
if (result < 0)
{
pinentry_setbuffer_clear (&pinentry);
if (pinentry.specific_err)
{
write_status_error (ctx, &pinentry);
return pinentry.specific_err;
}
return (pinentry.locale_err
? gpg_error (GPG_ERR_LOCALE_PROBLEM)
: gpg_error (GPG_ERR_CANCELED));
}
out:
if (result)
{
if (pinentry.repeat_okay)
assuan_write_status (ctx, "PIN_REPEATED", "");
result = assuan_send_data (ctx, pinentry.pin, strlen(pinentry.pin));
if (!result)
result = assuan_send_data (ctx, NULL, 0);
if (/* GPG Agent says it's okay. */
pinentry.allow_external_password_cache && pinentry.keyinfo
/* We didn't just read it from the cache. */
&& ! just_read_password_from_cache
/* And the user said it's okay. */
&& pinentry.may_cache_password)
/* Cache the password. */
password_cache_save (pinentry.keyinfo, pinentry.pin);
}
pinentry_setbuffer_clear (&pinentry);
return result;
}
/* Note that the option --one-button is a hack to allow the use of old
pinentries while the caller is ignoring the result. Given that
options have never been used or flagged as an error the new option
is an easy way to enable the messsage mode while not requiring to
update pinentry or to have the caller test for the message
command. New applications which are free to require an updated
pinentry should use MESSAGE instead. */
static gpg_error_t
cmd_confirm (assuan_context_t ctx, char *line)
{
int result;
pinentry.one_button = !!strstr (line, "--one-button");
pinentry.quality_bar = 0;
pinentry.close_button = 0;
pinentry.locale_err = 0;
pinentry.specific_err = 0;
pinentry.specific_err_loc = NULL;
free (pinentry.specific_err_info);
pinentry.specific_err_info = NULL;
pinentry.canceled = 0;
pinentry_setbuffer_clear (&pinentry);
result = (*pinentry_cmd_handler) (&pinentry);
if (pinentry.error)
{
free (pinentry.error);
pinentry.error = NULL;
}
if (pinentry.close_button)
assuan_write_status (ctx, "BUTTON_INFO", "close");
if (result)
return 0;
if (pinentry.specific_err)
{
write_status_error (ctx, &pinentry);
return pinentry.specific_err;
}
if (pinentry.locale_err)
return gpg_error (GPG_ERR_LOCALE_PROBLEM);
if (pinentry.one_button)
return 0;
if (pinentry.canceled)
return gpg_error (GPG_ERR_CANCELED);
return gpg_error (GPG_ERR_NOT_CONFIRMED);
}
static gpg_error_t
cmd_message (assuan_context_t ctx, char *line)
{
(void)line;
return cmd_confirm (ctx, "--one-button");
}
/* GETINFO <what>
Multipurpose function to return a variety of information.
Supported values for WHAT are:
version - Return the version of the program.
pid - Return the process id of the server.
*/
static gpg_error_t
cmd_getinfo (assuan_context_t ctx, char *line)
{
int rc;
if (!strcmp (line, "version"))
{
const char *s = VERSION;
rc = assuan_send_data (ctx, s, strlen (s));
}
else if (!strcmp (line, "pid"))
{
char numbuf[50];
snprintf (numbuf, sizeof numbuf, "%lu", (unsigned long)getpid ());
rc = assuan_send_data (ctx, numbuf, strlen (numbuf));
}
else
rc = gpg_error (GPG_ERR_ASS_PARAMETER);
return rc;
}
/* CLEARPASSPHRASE <cacheid>
Clear the cache passphrase associated with the key identified by
cacheid.
*/
static gpg_error_t
cmd_clear_passphrase (assuan_context_t ctx, char *line)
{
(void)ctx;
if (! line)
return gpg_error (GPG_ERR_ASS_INV_VALUE);
/* Remove leading and trailing white space. */
while (*line == ' ')
line ++;
while (line[strlen (line) - 1] == ' ')
line[strlen (line) - 1] = 0;
switch (password_cache_clear (line))
{
case 1: return 0;
case 0: return gpg_error (GPG_ERR_ASS_INV_VALUE);
default: return gpg_error (GPG_ERR_ASS_GENERAL);
}
}
/* Tell the assuan library about our commands. */
static gpg_error_t
register_commands (assuan_context_t ctx)
{
static struct
{
const char *name;
gpg_error_t (*handler) (assuan_context_t, char *line);
} table[] =
{
{ "SETDESC", cmd_setdesc },
{ "SETPROMPT", cmd_setprompt },
{ "SETKEYINFO", cmd_setkeyinfo },
{ "SETREPEAT", cmd_setrepeat },
{ "SETREPEATERROR", cmd_setrepeaterror },
{ "SETERROR", cmd_seterror },
{ "SETOK", cmd_setok },
{ "SETNOTOK", cmd_setnotok },
{ "SETCANCEL", cmd_setcancel },
{ "GETPIN", cmd_getpin },
{ "CONFIRM", cmd_confirm },
{ "MESSAGE", cmd_message },
{ "SETQUALITYBAR", cmd_setqualitybar },
{ "SETQUALITYBAR_TT", cmd_setqualitybar_tt },
{ "GETINFO", cmd_getinfo },
{ "SETTITLE", cmd_settitle },
{ "SETTIMEOUT", cmd_settimeout },
{ "CLEARPASSPHRASE", cmd_clear_passphrase },
{ NULL }
};
int i, j;
gpg_error_t rc;
for (i = j = 0; table[i].name; i++)
{
rc = assuan_register_command (ctx, table[i].name, table[i].handler, NULL);
if (rc)
return rc;
}
return 0;
}
int
pinentry_loop2 (int infd, int outfd)
{
gpg_error_t rc;
assuan_fd_t filedes[2];
assuan_context_t ctx;
/* Extra check to make sure we have dropped privs. */
#ifndef HAVE_DOSISH_SYSTEM
if (getuid() != geteuid())
abort ();
#endif
rc = assuan_new (&ctx);
if (rc)
{
fprintf (stderr, "server context creation failed: %s\n",
gpg_strerror (rc));
return -1;
}
/* For now we use a simple pipe based server so that we can work
from scripts. We will later add options to run as a daemon and
wait for requests on a Unix domain socket. */
filedes[0] = assuan_fdopen (infd);
filedes[1] = assuan_fdopen (outfd);
rc = assuan_init_pipe_server (ctx, filedes);
if (rc)
{
fprintf (stderr, "%s: failed to initialize the server: %s\n",
this_pgmname, gpg_strerror (rc));
return -1;
}
rc = register_commands (ctx);
if (rc)
{
fprintf (stderr, "%s: failed to the register commands with Assuan: %s\n",
this_pgmname, gpg_strerror (rc));
return -1;
}
assuan_register_option_handler (ctx, option_handler);
#if 0
assuan_set_log_stream (ctx, stderr);
#endif
assuan_register_reset_notify (ctx, pinentry_assuan_reset_handler);
for (;;)
{
rc = assuan_accept (ctx);
if (rc == -1)
break;
else if (rc)
{
fprintf (stderr, "%s: Assuan accept problem: %s\n",
this_pgmname, gpg_strerror (rc));
break;
}
rc = assuan_process (ctx);
if (rc)
{
fprintf (stderr, "%s: Assuan processing failed: %s\n",
this_pgmname, gpg_strerror (rc));
continue;
}
}
assuan_release (ctx);
return 0;
}
/* Start the pinentry event loop. The program will start to process
Assuan commands until it is finished or an error occurs. If an
error occurs, -1 is returned. Otherwise, 0 is returned. */
int
pinentry_loop (void)
{
return pinentry_loop2 (STDIN_FILENO, STDOUT_FILENO);
}
diff --git a/pinentry/pinentry.h b/pinentry/pinentry.h
index 50d5f98..c6e12e4 100644
--- a/pinentry/pinentry.h
+++ b/pinentry/pinentry.h
@@ -1,301 +1,301 @@
/* pinentry.h - The interface for the PIN entry support library.
Copyright (C) 2002, 2003, 2010, 2015 g10 Code GmbH
This file is part of PINENTRY.
PINENTRY 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.
PINENTRY is distributed in the hope that it will be useful, but
WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program; if not, see <http://www.gnu.org/licenses/>.
*/
#ifndef PINENTRY_H
#define PINENTRY_H
#ifdef __cplusplus
extern "C" {
#if 0
}
#endif
#endif
typedef enum {
PINENTRY_COLOR_NONE, PINENTRY_COLOR_DEFAULT,
PINENTRY_COLOR_BLACK, PINENTRY_COLOR_RED,
PINENTRY_COLOR_GREEN, PINENTRY_COLOR_YELLOW,
PINENTRY_COLOR_BLUE, PINENTRY_COLOR_MAGENTA,
PINENTRY_COLOR_CYAN, PINENTRY_COLOR_WHITE
} pinentry_color_t;
struct pinentry
{
/* The window title, or NULL. (Assuan: "SETTITLE TITLE".) */
char *title;
/* The description to display, or NULL. (Assuan: "SETDESC
DESC".) */
char *description;
/* The error message to display, or NULL. (Assuan: "SETERROR
MESSAGE".) */
char *error;
/* The prompt to display, or NULL. (Assuan: "SETPROMPT
prompt".) */
char *prompt;
/* The OK button text to display, or NULL. (Assuan: "SETOK
OK".) */
char *ok;
/* The Not-OK button text to display, or NULL. This is the text for
the alternative option shown by the third button. (Assuan:
"SETNOTOK NOTOK".) */
char *notok;
/* The Cancel button text to display, or NULL. (Assuan: "SETCANCEL
CANCEL".) */
char *cancel;
/* The buffer to store the secret into. */
char *pin;
/* The length of the buffer. */
int pin_len;
/* Whether the pin was read from an external cache (1) or entered by
the user (0). */
int pin_from_cache;
/* The name of the X display to use if X is available and supported.
(Assuan: "OPTION display DISPLAY".) */
char *display;
/* The name of the terminal node to open if X not available or
supported. (Assuan: "OPTION ttyname TTYNAME".) */
char *ttyname;
/* The type of the terminal. (Assuan: "OPTION ttytype TTYTYPE".) */
char *ttytype;
/* The LC_CTYPE value for the terminal. (Assuan: "OPTION lc-ctype
LC_CTYPE".) */
char *lc_ctype;
/* The LC_MESSAGES value for the terminal. (Assuan: "OPTION
lc-messages LC_MESSAGES".) */
char *lc_messages;
/* True if debug mode is requested. */
int debug;
/* The number of seconds before giving up while waiting for user input. */
int timeout;
/* True if caller should grab the keyboard. (Assuan: "OPTION grab"
or "OPTION no-grab".) */
int grab;
/* The window ID of the parent window over which the pinentry window
should be displayed. (Assuan: "OPTION parent-wid WID".) */
int parent_wid;
/* The name of an optional file which will be touched after a curses
entry has been displayed. (Assuan: "OPTION touch-file
FILENAME".) */
char *touch_file;
/* The frontend should set this to -1 if the user canceled the
request, and to the length of the PIN stored in pin
otherwise. */
int result;
/* The frontend should set this if the NOTOK button was pressed. */
int canceled;
/* The frontend should set this to true if an error with the local
- conversion occured. */
+ conversion occurred. */
int locale_err;
/* The frontend should set this to a gpg-error so that commands are
able to return specific error codes. This is an ugly hack due to
the fact that pinentry_cmd_handler_t returns the length of the
passphrase or a negative error code. */
int specific_err;
/* The frontend may store a string with the error location here. */
const char *specific_err_loc;
/* The frontend may store a malloced string here to emit an ERROR
* status code with this extra info along with SPECIFIC_ERR. */
char *specific_err_info;
/* The frontend should set this to true if the window close button
has been used. This flag is used in addition to a regular return
value. */
int close_button;
/* The caller should set this to true if only one button is
required. This is useful for notification dialogs where only a
dismiss button is required. */
int one_button;
/* If true a second prompt for the passphrase is shown and the user
is expected to enter the same passphrase again. Pinentry checks
that both match. (Assuan: "SETREPEAT".) */
char *repeat_passphrase;
/* The string to show if a repeated passphrase does not match.
(Assuan: "SETREPEATERROR ERROR".) */
char *repeat_error_string;
/* Set to true if the passphrase has been entered a second time and
matches the first passphrase. */
int repeat_okay;
/* If this is not NULL, a passphrase quality indicator is shown.
There will also be an inquiry back to the caller to get an
indication of the quality for the passphrase entered so far. The
string is used as a label for the quality bar. (Assuan:
"SETQUALITYBAR LABEL".) */
char *quality_bar;
/* The tooltip to be show for the qualitybar. Malloced or NULL.
(Assuan: "SETQUALITYBAR_TT TOOLTIP".) */
char *quality_bar_tt;
/* For the curses pinentry, the color of error messages. */
pinentry_color_t color_fg;
int color_fg_bright;
pinentry_color_t color_bg;
pinentry_color_t color_so;
int color_so_bright;
/* Malloced and i18ned default strings or NULL. These strings may
include an underscore character to indicate an accelerator key.
A double underscore represents a plain one. */
/* (Assuan: "OPTION default-ok OK"). */
char *default_ok;
/* (Assuan: "OPTION default-cancel CANCEL"). */
char *default_cancel;
/* (Assuan: "OPTION default-prompt PROMPT"). */
char *default_prompt;
/* (Assuan: "OPTION default-pwmngr
SAVE_PASSWORD_WITH_PASSWORD_MANAGER?"). */
char *default_pwmngr;
/* (Assuan: "OPTION default-cf-visi
Do you really want to make your passphrase visible?"). */
char *default_cf_visi;
/* (Assuan: "OPTION default-tt-visi
Make passphrase visible?"). */
char *default_tt_visi;
/* (Assuan: "OPTION default-tt-hide
Hide passphrase"). */
char *default_tt_hide;
/* Whether we are allowed to read the password from an external
cache. (Assuan: "OPTION allow-external-password-cache") */
int allow_external_password_cache;
/* We only try the cache once. */
int tried_password_cache;
/* A stable identifier for the key. (Assuan: "SETKEYINFO
KEYINFO".) */
char *keyinfo;
/* Whether we may cache the password (according to the user). */
int may_cache_password;
/* NOTE: If you add any additional fields to this structure, be sure
to update the initializer in pinentry/pinentry.c!!! */
/* For the quality indicator we need to do an inquiry. Thus we need
to save the assuan ctx. */
void *ctx_assuan;
/* An UTF-8 string with an invisible character used to override the
default in some pinentries. Only the first character is
used. */
char *invisible_char;
};
typedef struct pinentry *pinentry_t;
/* The pinentry command handler type processes the pinentry request
PIN. If PIN->pin is zero, request a confirmation, otherwise a PIN
entry. On confirmation, the function should return TRUE if
confirmed, and FALSE otherwise. On PIN entry, the function should
- return -1 if an error occured or the user cancelled the operation
+ return -1 if an error occurred or the user cancelled the operation
and 1 otherwise. */
typedef int (*pinentry_cmd_handler_t) (pinentry_t pin);
/* Start the pinentry event loop. The program will start to process
Assuan commands until it is finished or an error occurs. If an
error occurs, -1 is returned and errno indicates the type of an
error. Otherwise, 0 is returned. */
int pinentry_loop (void);
/* The same as above but allows to specify the i/o descriptors.
* infd and outfd will be duplicated in this function so the caller
* still has to close them if necessary.
*/
int pinentry_loop2 (int infd, int outfd);
/* Convert the UTF-8 encoded string TEXT to the encoding given in
LC_CTYPE. Return NULL on error. */
char *pinentry_utf8_to_local (const char *lc_ctype, const char *text);
/* Convert TEXT which is encoded according to LC_CTYPE to UTF-8. With
SECURE set to true, use secure memory for the returned buffer.
Return NULL on error. */
char *pinentry_local_to_utf8 (char *lc_ctype, char *text, int secure);
/* Run a quality inquiry for PASSPHRASE of LENGTH. */
int pinentry_inq_quality (pinentry_t pin,
const char *passphrase, size_t length);
/* Try to make room for at least LEN bytes for the pin in the pinentry
PIN. Returns new buffer on success and 0 on failure. */
char *pinentry_setbufferlen (pinentry_t pin, int len);
/* Use the buffer at BUFFER for PIN->PIN. BUFFER must be NULL or
allocated using secmem_alloc. LEN is the size of the buffer. If
it is unknown, but BUFFER is a NUL terminated string, you pass 0 to
just use strlen(buffer)+1. */
void pinentry_setbuffer_use (pinentry_t pin, char *buffer, int len);
/* Initialize the secure memory subsystem, drop privileges and
return. Must be called early. */
void pinentry_init (const char *pgmname);
/* Return true if either DISPLAY is set or ARGV contains the string
"--display". */
int pinentry_have_display (int argc, char **argv);
/* Parse the command line options. May exit the program if only help
or version output is requested. */
void pinentry_parse_opts (int argc, char *argv[]);
/* The caller must define this variable to process assuan commands. */
extern pinentry_cmd_handler_t pinentry_cmd_handler;
#ifdef HAVE_W32_SYSTEM
/* Windows declares sleep as obsolete, but provides a definition for
_sleep but non for the still existing sleep. */
#define sleep(a) _sleep ((a))
#endif /*HAVE_W32_SYSTEM*/
#if 0
{
#endif
#ifdef __cplusplus
}
#endif
#endif /* PINENTRY_H */
diff --git a/qt/pinentrydialog.cpp b/qt/pinentrydialog.cpp
index 12f7718..f9dd700 100644
--- a/qt/pinentrydialog.cpp
+++ b/qt/pinentrydialog.cpp
@@ -1,470 +1,470 @@
/* pinentrydialog.cpp - A (not yet) secure Qt 4 dialog for PIN entry.
Copyright (C) 2002, 2008 Klarälvdalens Datakonsult AB (KDAB)
Copyright 2007 Ingo Klöcker
Copyright 2016 Intevation GmbH
Written by Steffen Hansen <steffen@klaralvdalens-datakonsult.se>.
Modified by Andre Heinecke <aheinecke@intevation.de>
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 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., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
*/
#include "pinentrydialog.h"
#include <QGridLayout>
#include <QProgressBar>
#include <QApplication>
#include <QStyle>
#include <QPainter>
#include <QPushButton>
#include <QDialogButtonBox>
#include <QKeyEvent>
#include <QLabel>
#include <QPalette>
#include <QLineEdit>
#include <QAction>
#include <QCheckBox>
#ifdef Q_OS_WIN
#include <windows.h>
#endif
/* I [wk] have no idea for what this code was supposed to do.
Foregrounding a window is heavily restricted by modern Windows
versions. This is the reason why gpg-agent employs its
AllowSetForegroundWindow callback machinery to ask the supposed to
be be calling process to allow a pinentry to go into the
foreground.
[ah] This is a Hack to workaround the fact that Foregrounding
a Window is so restricted that it AllowSetForegroundWindow
does not always work (e.g. when the ForegroundWindow timeout
has not expired.
*/
#ifdef Q_OS_WIN
WINBOOL SetForegroundWindowEx(HWND hWnd)
{
//Attach foreground window thread to our thread
const DWORD ForeGroundID = GetWindowThreadProcessId(::GetForegroundWindow(), NULL);
const DWORD CurrentID = GetCurrentThreadId();
WINBOOL retval;
AttachThreadInput(ForeGroundID, CurrentID, TRUE);
//Do our stuff here
HWND hLastActivePopupWnd = GetLastActivePopup(hWnd);
retval = SetForegroundWindow(hLastActivePopupWnd);
//Detach the attached thread
AttachThreadInput(ForeGroundID, CurrentID, FALSE);
return retval;
}// End SetForegroundWindowEx
#endif
void raiseWindow(QWidget *w)
{
- /* Maybe Qt will become agressive enough one day that
+ /* Maybe Qt will become aggressive enough one day that
* this is enough on windows too*/
w->raise();
#ifdef Q_OS_WIN
HWND wid = (HWND)w->effectiveWinId();
/* In the meantime we do our own attention grabbing */
if (!SetForegroundWindow(wid) && !SetForegroundWindowEx(wid)) {
OutputDebugString("SetForegroundWindow (ex) failed");
/* Yet another fallback which will not work on some
* versions and is not recommended by msdn */
if (!ShowWindow(wid, SW_SHOWNORMAL)) {
OutputDebugString("ShowWindow failed.");
}
}
/* Even if SetForgeoundWindow / SetForegroundWinowEx don't fail
* we sometimes are still not in the foreground. So we try yet
* another hack by using SetWindowPos */
if (!SetWindowPos(wid, HWND_TOPMOST, 0, 0, 0, 0,
SWP_NOMOVE | SWP_NOSIZE | SWP_SHOWWINDOW)) {
OutputDebugString("SetWindowPos failed.");
} else {
/* Without moving back to NOTOPMOST we just stay on top.
* Even if the user changes focus. */
SetWindowPos(wid, HWND_NOTOPMOST, 0, 0, 0, 0,
SWP_NOMOVE | SWP_NOSIZE | SWP_SHOWWINDOW);
}
#endif
}
QPixmap icon(QStyle::StandardPixmap which)
{
QPixmap pm = qApp->windowIcon().pixmap(48, 48);
if (which != QStyle::SP_CustomBase) {
const QIcon ic = qApp->style()->standardIcon(which);
QPainter painter(&pm);
const int emblemSize = 22;
painter.drawPixmap(pm.width() - emblemSize, 0,
ic.pixmap(emblemSize, emblemSize));
}
return pm;
}
void PinEntryDialog::slotTimeout()
{
reject();
}
PinEntryDialog::PinEntryDialog(QWidget *parent, const char *name,
int timeout, bool modal, bool enable_quality_bar,
const QString &repeatString,
const QString &visibilityTT,
const QString &hideTT)
: QDialog(parent, Qt::WindowStaysOnTopHint),
mRepeat(NULL),
_grabbed(false),
mVisibilityTT(visibilityTT),
mHideTT(hideTT),
mVisiActionEdit(NULL),
mVisiCB(NULL)
{
setWindowFlags(windowFlags() & ~Qt::WindowContextHelpButtonHint);
if (modal) {
setWindowModality(Qt::ApplicationModal);
}
_icon = new QLabel(this);
_icon->setPixmap(icon());
_error = new QLabel(this);
QPalette pal;
pal.setColor(QPalette::WindowText, Qt::red);
_error->setPalette(pal);
_error->hide();
_desc = new QLabel(this);
_desc->hide();
_prompt = new QLabel(this);
_prompt->hide();
_edit = new QLineEdit(this);
_edit->setMaxLength(256);
_edit->setEchoMode(QLineEdit::Password);
_prompt->setBuddy(_edit);
if (enable_quality_bar) {
_quality_bar_label = new QLabel(this);
_quality_bar_label->setAlignment(Qt::AlignVCenter);
_quality_bar = new QProgressBar(this);
_quality_bar->setAlignment(Qt::AlignCenter);
_have_quality_bar = true;
} else {
_have_quality_bar = false;
}
QDialogButtonBox *const buttons = new QDialogButtonBox(this);
buttons->setStandardButtons(QDialogButtonBox::Ok | QDialogButtonBox::Cancel);
_ok = buttons->button(QDialogButtonBox::Ok);
_cancel = buttons->button(QDialogButtonBox::Cancel);
_ok->setDefault(true);
if (style()->styleHint(QStyle::SH_DialogButtonBox_ButtonsHaveIcons)) {
_ok->setIcon(style()->standardIcon(QStyle::SP_DialogOkButton));
_cancel->setIcon(style()->standardIcon(QStyle::SP_DialogCancelButton));
}
if (timeout > 0) {
_timer = new QTimer(this);
connect(_timer, SIGNAL(timeout()), this, SLOT(slotTimeout()));
_timer->start(timeout * 1000);
} else {
_timer = NULL;
}
connect(buttons, SIGNAL(accepted()), this, SLOT(accept()));
connect(buttons, SIGNAL(rejected()), this, SLOT(reject()));
connect(_edit, SIGNAL(textChanged(QString)),
this, SLOT(updateQuality(QString)));
connect(_edit, SIGNAL(textChanged(QString)),
this, SLOT(textChanged(QString)));
_edit->setFocus();
QGridLayout *const grid = new QGridLayout(this);
int row = 1;
grid->addWidget(_error, row++, 1, 1, 2);
grid->addWidget(_desc, row++, 1, 1, 2);
//grid->addItem( new QSpacerItem( 0, _edit->height() / 10, QSizePolicy::Minimum, QSizePolicy::Fixed ), 1, 1 );
grid->addWidget(_prompt, row, 1);
grid->addWidget(_edit, row++, 2);
if (!repeatString.isNull()) {
mRepeat = new QLineEdit;
mRepeat->setMaxLength(256);
mRepeat->setEchoMode(QLineEdit::Password);
connect(mRepeat, SIGNAL(textChanged(QString)),
this, SLOT(textChanged(QString)));
QLabel *repeatLabel = new QLabel(repeatString);
repeatLabel->setBuddy(mRepeat);
grid->addWidget(repeatLabel, row, 1);
grid->addWidget(mRepeat, row++, 2);
setTabOrder(_edit, mRepeat);
setTabOrder(mRepeat, _ok);
}
if (enable_quality_bar) {
grid->addWidget(_quality_bar_label, row, 1);
grid->addWidget(_quality_bar, row++, 2);
}
/* Set up the show password action */
const QIcon visibilityIcon = QIcon::fromTheme(QLatin1String("visibility"));
const QIcon hideIcon = QIcon::fromTheme(QLatin1String("hint"));
#if QT_VERSION >= 0x050200
if (!visibilityIcon.isNull() && !hideIcon.isNull()) {
mVisiActionEdit = _edit->addAction(visibilityIcon, QLineEdit::TrailingPosition);
mVisiActionEdit->setVisible(false);
mVisiActionEdit->setToolTip(mVisibilityTT);
connect(mVisiActionEdit, SIGNAL(triggered()), this, SLOT(toggleVisibility()));
} else
#endif
{
if (!mVisibilityTT.isNull()) {
mVisiCB = new QCheckBox(mVisibilityTT);
connect(mVisiCB, SIGNAL(toggled(bool)), this, SLOT(toggleVisibility()));
grid->addWidget(mVisiCB, row++, 1, 1, 2, Qt::AlignLeft);
}
}
grid->addWidget(buttons, ++row, 0, 1, 3);
grid->addWidget(_icon, 0, 0, row - 1, 1, Qt::AlignVCenter | Qt::AlignLeft);
grid->setSizeConstraint(QLayout::SetFixedSize);
connect(qApp, SIGNAL(focusChanged(QWidget *, QWidget *)),
this, SLOT(focusChanged(QWidget *, QWidget *)));
}
void PinEntryDialog::showEvent(QShowEvent *event)
{
QDialog::showEvent(event);
raiseWindow(this);
}
void PinEntryDialog::setDescription(const QString &txt)
{
_desc->setVisible(!txt.isEmpty());
_desc->setText(txt);
#ifndef QT_NO_ACCESSIBILITY
_desc->setAccessibleDescription(txt);
#endif
_icon->setPixmap(icon());
setError(QString::null);
}
QString PinEntryDialog::description() const
{
return _desc->text();
}
void PinEntryDialog::setError(const QString &txt)
{
if (!txt.isNull()) {
_icon->setPixmap(icon(QStyle::SP_MessageBoxCritical));
}
_error->setText(txt);
#ifndef QT_NO_ACCESSIBILITY
_error->setAccessibleDescription(txt);
#endif
_error->setVisible(!txt.isEmpty());
}
QString PinEntryDialog::error() const
{
return _error->text();
}
void PinEntryDialog::setPin(const QString &txt)
{
_edit->setText(txt);
}
QString PinEntryDialog::pin() const
{
return _edit->text();
}
void PinEntryDialog::setPrompt(const QString &txt)
{
_prompt->setText(txt);
_prompt->setVisible(!txt.isEmpty());
}
QString PinEntryDialog::prompt() const
{
return _prompt->text();
}
void PinEntryDialog::setOkText(const QString &txt)
{
_ok->setText(txt);
#ifndef QT_NO_ACCESSIBILITY
_ok->setAccessibleDescription(txt);
#endif
_ok->setVisible(!txt.isEmpty());
}
void PinEntryDialog::setCancelText(const QString &txt)
{
_cancel->setText(txt);
#ifndef QT_NO_ACCESSIBILITY
_cancel->setAccessibleDescription(txt);
#endif
_cancel->setVisible(!txt.isEmpty());
}
void PinEntryDialog::setQualityBar(const QString &txt)
{
if (_have_quality_bar) {
_quality_bar_label->setText(txt);
#ifndef QT_NO_ACCESSIBILITY
_quality_bar_label->setAccessibleDescription(txt);
#endif
}
}
void PinEntryDialog::setQualityBarTT(const QString &txt)
{
if (_have_quality_bar) {
_quality_bar->setToolTip(txt);
}
}
void PinEntryDialog::updateQuality(const QString &txt)
{
int length;
int percent;
QPalette pal;
if (_timer) {
_timer->stop();
}
if (!_have_quality_bar || !_pinentry_info) {
return;
}
const QByteArray utf8_pin = txt.toUtf8();
const char *pin = utf8_pin.constData();
length = strlen(pin);
percent = length ? pinentry_inq_quality(_pinentry_info, pin, length) : 0;
if (!length) {
_quality_bar->reset();
} else {
pal = _quality_bar->palette();
if (percent < 0) {
pal.setColor(QPalette::Highlight, QColor("red"));
percent = -percent;
} else {
pal.setColor(QPalette::Highlight, QColor("green"));
}
_quality_bar->setPalette(pal);
_quality_bar->setValue(percent);
}
}
void PinEntryDialog::setPinentryInfo(pinentry_t peinfo)
{
_pinentry_info = peinfo;
}
void PinEntryDialog::focusChanged(QWidget *old, QWidget *now)
{
// Grab keyboard. It might be a little weird to do it here, but it works!
// Previously this code was in showEvent, but that did not work in Qt4.
if (!_pinentry_info || _pinentry_info->grab) {
if (_grabbed && old && (old == _edit || old == mRepeat)) {
old->releaseKeyboard();
_grabbed = false;
}
if (!_grabbed && now && (now == _edit || now == mRepeat)) {
now->grabKeyboard();
_grabbed = true;
}
}
}
void PinEntryDialog::textChanged(const QString &text)
{
Q_UNUSED(text);
if (mRepeat && mRepeat->text() == _edit->text()) {
_ok->setEnabled(true);
_ok->setToolTip(QString());
} else if (mRepeat) {
_ok->setEnabled(false);
_ok->setToolTip(mRepeatError);
}
if (mVisiActionEdit && sender() == _edit) {
mVisiActionEdit->setVisible(!_edit->text().isEmpty());
}
}
void PinEntryDialog::toggleVisibility()
{
if (sender() == mVisiActionEdit) {
if (_edit->echoMode() == QLineEdit::Password) {
mVisiActionEdit->setIcon(QIcon::fromTheme(QLatin1String("hint")));
mVisiActionEdit->setToolTip(mHideTT);
_edit->setEchoMode(QLineEdit::Normal);
if (mRepeat) {
mRepeat->setEchoMode(QLineEdit::Normal);
}
} else {
mVisiActionEdit->setIcon(QIcon::fromTheme(QLatin1String("visibility")));
mVisiActionEdit->setToolTip(mVisibilityTT);
_edit->setEchoMode(QLineEdit::Password);
if (mRepeat) {
mRepeat->setEchoMode(QLineEdit::Password);
}
}
}
if (sender() == mVisiCB) {
if (mVisiCB->isChecked()) {
if (mRepeat) {
mRepeat->setEchoMode(QLineEdit::Normal);
}
_edit->setEchoMode(QLineEdit::Normal);
} else {
if (mRepeat) {
mRepeat->setEchoMode(QLineEdit::Password);
}
_edit->setEchoMode(QLineEdit::Password);
}
}
}
QString PinEntryDialog::repeatedPin() const
{
if (mRepeat) {
return mRepeat->text();
}
return QString();
}
void PinEntryDialog::setRepeatErrorText(const QString &err)
{
mRepeatError = err;
}
#include "pinentrydialog.moc"

File Metadata

Mime Type
text/x-diff
Expires
Thu, Jul 17, 1:19 AM (7 h, 51 m)
Storage Engine
local-disk
Storage Format
Raw Data
Storage Handle
2a/8d/cea80f1442c89dfb4b62d694235b

Event Timeline