Page MenuHome GnuPG
Files F36623134

No OneTemporary
Actions

View Options

This file is larger than 256 KB, so syntax highlighting was skipped.
diff --git a/ChangeLog b/ChangeLog
index 70605fd..c9888b4 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,116 +1,131 @@
+2006-11-21 Marcus Brinkmann <marcus@g10code.de>
+
+ * configure.ac (AC_CONFIG_FILES): Add doc/manual/Makefile.
+ * doc/Makefile.am (SUBDIRS): New variable.
+ * doc/manual/Makefile, doc/manual/scute.texi, doc/manual/gpl.texi,
+ doc/manual/mdate-sh, doc/manual/texinfo.tex,
+ doc/manual/firefox-cm.png, doc/manual/firefox-cm-view-detail.png,
+ doc/manual/firefox-cm-view.png,
+ doc/manual/firefox-dm-load-after.png,
+ doc/manual/firefox-dm-load-before.png,
+ doc/manual/firefox-dm-load.png,
+ doc/manual/firefox-dm-token-present.png,
+ doc/manual/firefox-pref.png, doc/manual/firefox-pref-view.png: New
+ files.
+
2006-11-11 Marcus Brinkmann <marcus@g10code.de>
* Released version 1.0.0.
2006-11-11 Marcus Brinkmann <marcus@g10code.de>
* src/error-mapping.c (scute_gpg_err_to_ck): Report error on debug
stream.
* src/slots.c (add_object): New function.
(slot_init): Rewritten using add_object.
* src/gpgsm.c: Include "gpgsm.h".
(struct search): Replace members ATTRP, ATTR_COUNTP, PRV_ATTRP,
PRV_ATTR_COUNTP by CERT_GET_CB, HOOK.
(search_cb): Rewritten to add all certificates for a certain key
(scute_gpgsm_get_cert): Take a callback function instead of
attribute pointers.
* gpgsm.h (scute_gpgsm_get_cert): Adjust prototype.
and also the certificate chain.
* src/cert.h (scute_gpgsm_search_certs_by_fpr): New prototype.
(struct cert): New member chain_id.
* src/cert-gpgsm.c (struct search_ctx_by_grip): Rename to ...
(struct search_ctx): ... this. Rename field GRIP to PATTERN, and
add new field FIELD.
(search_certs_by_grip): Rename function to ...
(search_certs): ... this.
(scute_gpgsm_search_certs_by_fpr): Change user of the above.
(scute_gpgsm_search_certs_by_fpr): New function.
(search_certs_line): Store chain ID.
2006-11-07 Marcus Brinkmann <marcus@g10code.de>
* src/p11-gettokeninfo.c (C_GetTokenInfo): Fix assignment.
2006-11-06 Marcus Brinkmann <marcus@g10code.de>
* doc/website/format/web.css: Set margin to 0.
* configure.ac (AC_CONFIG_FILES): Add doc/Makefile.
Makefile.am (SUBDIRS): Add doc/.
doc/, doc/website/, doc/website/format: New directories.
doc/Makefile.am, doc/website/index.xhtml,
doc/website/documentation.xhtml, doc/website/download.xhtml,
doc/website/contact.xhtml, doc/website/format/web.css,
doc/website/format/scute-logo.svg,
doc/website/format/scute-border.jpg: New files.
2006-10-30 Marcus Brinkmann <marcus@g10code.de>
* configure.ac (PACKAGE, VERSION): Define these variables.
* src/Makefile.am (libscute_la_CPPFLAGS): Fix include path for
VPATH builds.
* configure.ac (VERSION_MAJOR, VERSION_MINOR): New variable
definitions.
* src/gpgsm.h: Fix syntax error.
* src/table.c, src/table.h: Rewritten.
* src/slots.c: Reorganized and rewritten to use the new table
interface. Include gpgsm.h.
* src/p11-signinit.c, src/p11-sign.c, src/p11-getslotlist.c,
src/p11-getsessioninfo.c, src/p11-getmechanismlist.c,
src/p11-getattributevalue.c, src/p11-findobjectsinit.c,
src/p11-findobjectsfinal.c, src/p11-findobjects.c,
src/p11-closesession.c: Adjust callers.
2006-10-26 Marcus Brinkmann <marcus@g10code.de>
* src/settings.h: Require VERSION_MAJOR and VERSION_MINOR to be
defined.
* src/p11-getinfo.c (CRYPTOKI_VERSION_MAJOR,
CRYPTOKI_VERSION_MINOR): Move macros to ...
* src/cryptoki.h (CRYPTOKI_VERSION_MAJOR, CRYPTOKI_VERSION_MINOR):
... here
* src/gpgsm.c (gpgsm_get_cert): Rename to ...
(scute_gpgsm_get_cert): ... this.
* src/gpgsm.h: New file.
* src/Makefile.am (libscute_la_SOURCES): Add gpgsm.h.
* src/slots.c (gpgsm_get_cert): Remove prototype.
(slots_update_slot): Call scute_gpgsm_get_cert, not gpgsm_get_cert.
* src/error-mapping.c (scute_gpg_err_to_ck): Map GPG_ERR_NO_AGENT
to CKR_GENERAL_ERROR.
* src/agent.h (struct agent_card_info_s): Remove member error.
Prefix external functions with scute_ if not already done so, also
for callers.
* src/agent.c: Include <stdarg.h>. Replace one-letter variable
names by proper ones. Replace assuan_error_t with gpg_error_t
everywhere.
(agent_connect, agent_simple_cmd, agent_configure): New functions.
(scute_agent_initialize): Use these new functions.
ing of common code. Lots of cleanups.
(scute_agent_finalize): Clear agent_ctx. Add debug message.
2006-10-13 Werner Koch <wk@g10code.com>
* src/cert-gpgsm.c (MAX_LINE_LEN): Increased.
* src/locking.c: Include string.h for memset declaration.
* configure.ac: New option --disable-optimization.
2006-08-02 Marcus Brinkmann <marcus@g10code.de>
* Initial version.
Copyright 2006 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/configure.ac b/configure.ac
index 62fa7cd..4d09aee 100644
--- a/configure.ac
+++ b/configure.ac
@@ -1,276 +1,277 @@
# -*- Autoconf -*-
# Process this file with autoconf to produce a configure script.
AC_PREREQ(2.59)
min_automake_version="1.9.3"
# Version number: Remember to change it immediately *after* a release.
# Make sure to run "svn up" before a "make dist".
# See below for the LT versions.
#
# The CVS version is usually the next intended release version with
# the string "-cvs" appended. The reason for this is that tests for a
# specific feature can already be done under the assumption that the
# CVS version is the most recent one in a branch. To disable the CVS
# version for the real release, just comment out the my_iscvs macro.
# Note, that we are now using Subversion instead of CVS and append the
# SVN revision number to the "cvs" suffix. To make this most useful
# for snapshot releases please do an "svn up" right before recreating
# the configure script, so that a proper revision number for all files
# is available when running a "make distcheck".
m4_define(my_version, [1.0.0])
-#m4_define(my_iscvs, yes)
+m4_define(my_iscvs, yes)
AC_INIT([scute], my_version[]m4_ifdef([my_iscvs], [-cvs[]m4_translit(
[$Revision: 1179 $],[Ra-z $:])]),
[marcus@g10code.com])
# LT Version numbers, remember to change them just *before* a release.
# (Code changed: REVISION++)
# (Interfaces added/removed/changed: CURRENT++, REVISION=0)
# (Interfaces added: AGE++)
# (Interfaces removed/changed: AGE=0)
#
LIBSCUTE_LT_CURRENT=0
LIBSCUTE_LT_AGE=0
LIBSCUTE_LT_REVISION=0
# Version numbers reported by the PKCS #11 module to its users.
VERSION_MAJOR=1
VERSION_MINOR=0
NEED_GPG_ERROR_VERSION=0.7
NEED_LIBASSUAN_VERSION=0.6.10
NEED_GPGSM_VERSION=1.9.6
# Some status variables to give feedback at the end of a configure run.
have_gpg_error=no
have_libassuan=no
PACKAGE=$PACKAGE_NAME
VERSION=$PACKAGE_VERSION
AC_CONFIG_SRCDIR([src/cryptoki.h])
AC_CONFIG_HEADER([config.h])
dnl FIXME: Enable this with autoconf 2.59.
dnl AC_CONFIG_MACRO_DIR(m4)
AM_INIT_AUTOMAKE($PACKAGE, $VERSION)
AM_MAINTAINER_MODE
AC_CANONICAL_HOST
# Enable GNU extensions on systems that have them.
AC_GNU_SOURCE
AH_VERBATIM([_REENTRANT],
[/* To allow the use of scute in multithreaded programs we have to use
special features from the library. */
#ifndef _REENTRANT
# define _REENTRANT 1
#endif])
# Checks for programs.
AC_PROG_CC
# FIXME: We should support setting the CFLAGS by the user. But
# -std=gnu99 is required.
CFLAGS="-std=gnu99 -Wall -Wno-pointer-sign -g -O3"
AC_ARG_ENABLE(optimization,
AC_HELP_STRING([--disable-optimization],
[disable compiler optimization]),
[if test $enableval = no ; then
CFLAGS=`echo $CFLAGS | sed 's/-O[[0-9]]//'`
fi])
AC_SUBST(LIBSCUTE_LT_CURRENT)
AC_SUBST(LIBSCUTE_LT_AGE)
AC_SUBST(LIBSCUTE_LT_REVISION)
AC_DEFINE_UNQUOTED(NEED_GPGSM_VERSION, "$NEED_GPGSM_VERSION",
[Min. needed GPGSM version.])
AC_SUBST(PACKAGE)
AC_SUBST(VERSION)
AC_DEFINE_UNQUOTED(PACKAGE, "$PACKAGE", [Name of this package])
AC_DEFINE_UNQUOTED(VERSION, "$VERSION", [Version of this package])
AC_DEFINE_UNQUOTED(VERSION_MAJOR, $VERSION_MAJOR, [Major version number])
AC_DEFINE_UNQUOTED(VERSION_MINOR, $VERSION_MINOR, [Minor version number])
# Don't default to build static libs.
AC_DISABLE_STATIC
AC_LIBTOOL_WIN32_DLL
AC_LIBTOOL_RC
AC_PROG_LIBTOOL
# For now we hardcode the use of version scripts. It would be better
# to write a test for this or even implement this within libtool.
have_ld_version_script=no
case "${host}" in
*-*-linux*)
have_ld_version_script=yes
;;
*-*-gnu*)
have_ld_version_script=yes
;;
esac
AM_CONDITIONAL(HAVE_LD_VERSION_SCRIPT, test "$have_ld_version_script" = "yes")
# Checks for libraries.
# The error code library. Error codes are sent over the IPC layer and
# have to be interpreted.
AM_PATH_GPG_ERROR("$NEED_GPG_ERROR_VERSION",
have_gpg_error=yes, have_gpg_error=no)
# The IPC library.
AM_PATH_LIBASSUAN("$NEED_LIBASSUAN_VERSION",
have_libassuan=yes, have_libassuan=no)
# GPGSM
GPGSM_DEFAULT=no
NO_OVERRIDE=no
AC_ARG_WITH(gpgsm,
AC_HELP_STRING([--with-gpgsm=PATH], [use GpgSM binary at PATH]),
GPGSM=$withval, NO_OVERRIDE=yes)
if test "$NO_OVERRIDE" = "yes" || test "$GPGSM" = "yes"; then
GPGSM=
NO_OVERRIDE=yes
if test "$cross_compiling" != "yes"; then
AC_PATH_PROG(GPGSM, gpgsm)
fi
if test -z "$GPGSM"; then
GPGSM="$GPGSM_DEFAULT"
fi
fi
if test "$GPGSM" = no; then
if test "$NO_OVERRIDE" = "yes"; then
if test "$cross_compiling" != "yes"; then
AC_MSG_WARN([
***
*** Could not find GpgSM, install GpgSM or use --with-gpgsm=PATH to enable it
***])
else
AC_MSG_ERROR([
***
*** Can not determine path to GpgSM when cross-compiling, use --with-gpgsm=PATH
***])
fi
fi
else
AC_DEFINE_UNQUOTED(GPGSM_PATH, "$GPGSM", [Path to the GPGSM binary.])
AC_DEFINE(ENABLE_GPGSM,1,[Whether GPGSM support is enabled])
fi
AM_CONDITIONAL(HAVE_GPGSM, test "$GPGSM" != "no")
dnl Check for GPGSM version requirement.
GPGSM_VERSION=unknown
ok=maybe
if test -z "$GPGSM" -o "x$GPGSM" = "xno"; then
ok=no
else
if test "$cross_compiling" = "yes"; then
AC_MSG_WARN([GPGSM version can not be checked when cross compiling])
ok=no
else
if test ! -x "$GPGSM"; then
AC_MSG_WARN([GPGSM not executable, version check disabled])
ok=no
fi
fi
fi
if test "$ok" = "maybe"; then
AC_MSG_CHECKING(for GPGSM >= $NEED_GPGSM_VERSION)
req_major=`echo $NEED_GPGSM_VERSION | \
sed 's/\([[0-9]]*\)\.\([[0-9]]*\)\.\([[0-9]]*\)/\1/'`
req_minor=`echo $NEED_GPGSM_VERSION | \
sed 's/\([[0-9]]*\)\.\([[0-9]]*\)\.\([[0-9]]*\)/\2/'`
req_micro=`echo $NEED_GPGSM_VERSION | \
sed 's/\([[0-9]]*\)\.\([[0-9]]*\)\.\([[0-9]]*\)/\3/'`
gpgsm_version=`$GPGSM --version | grep ^gpgsm`
major=`echo $gpgsm_version | \
sed 's/^gpgsm (GnuPG) \([[0-9]]*\)\.\([[0-9]]*\)\.\([[0-9]]*\).*/\1/'`
minor=`echo $gpgsm_version | \
sed 's/^gpgsm (GnuPG) \([[0-9]]*\)\.\([[0-9]]*\)\.\([[0-9]]*\).*/\2/'`
micro=`echo $gpgsm_version | \
sed 's/^gpgsm (GnuPG) \([[0-9]]*\)\.\([[0-9]]*\)\.\([[0-9]]*\).*/\3/'`
GPGSM_VERSION=`echo $gpgsm_version | sed 's/^gpgsm (GnuPG) //'`
if test "$major" -gt "$req_major"; then
ok=yes
else
if test "$major" -eq "$req_major"; then
if test "$minor" -gt "$req_minor"; then
ok=yes
else
if test "$minor" -eq "$req_minor"; then
if test "$micro" -ge "$req_micro"; then
ok=yes
fi
fi
fi
fi
fi
if test "$ok" = "yes"; then
AC_MSG_RESULT(yes)
else
AC_MSG_RESULT(no)
AC_MSG_WARN([GPGSM must be at least version $NEED_GPGSM_VERSION])
fi
fi
gpgsm_ok="$ok"
# Checks for header files.
AC_HEADER_STDC
AC_CHECK_HEADERS([stdlib.h string.h])
# Checks for typedefs, structures, and compiler characteristics.
AC_HEADER_STDBOOL
AC_C_INLINE
# Checks for library functions.
AC_FUNC_REALLOC
# Test if tests can be run
ok=yes
AM_CONDITIONAL(RUN_TESTS, test "$ok" = "yes")
# Print errors here so that they are visible all
# together and the user can acquire them all together.
die=no
if test "$have_gpg_error" = "no"; then
die=yes
AC_MSG_NOTICE([[
***
*** You need libgpg-error to build this program.
** This library is for example available at
*** ftp://ftp.gnupg.org/pub/gcrypt/libgpg-error
*** (at least version $NEED_GPG_ERROR_VERSION is required.)
***]])
fi
if test "$have_libassuan" = "no"; then
die=yes
AC_MSG_NOTICE([[
***
*** You need libassuan to build this program.
*** This library is for example available at
*** ftp://ftp.gnupg.org/pub/gcrypt/alpha/libassuan/
*** (at least version $NEED_LIBASSUAN_VERSION is required).
***]])
fi
if test "$die" = "yes"; then
AC_MSG_ERROR([[
***
*** Required libraries not found. Please consult the above messages
*** and install them before running configure again.
***]])
fi
AC_CONFIG_FILES([Makefile
include/Makefile
src/Makefile
tests/Makefile
+ doc/manual/Makefile
doc/Makefile])
AC_OUTPUT
diff --git a/doc/Makefile.am b/doc/Makefile.am
index 6f62966..83f49cf 100644
--- a/doc/Makefile.am
+++ b/doc/Makefile.am
@@ -1,38 +1,40 @@
# Makefile.am
# Copyright (C) 2006 g10 Code GmbH
#
# This file is part of scute[1].
#
# [1] Derived from the RSA Security Inc. PKCS #11 Cryptographic Token
# Interface (Cryptoki).
#
# Scute 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.
#
# Scute 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 Scute; if not, write to the Free Software Foundation,
# Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
#
# In addition, as a special exception, g10 Code GmbH gives permission
# to link this library: with the Mozilla Foundation's code for
# Mozilla (or with modified versions of it that use the same license
# as the "Mozilla" code), and distribute the linked executables. You
# must obey the GNU General Public License in all respects for all of
# the code used other than "Mozilla". If you modify this file, you
# may extend this exception to your version of the file, but you are
# not obligated to do so. If you do not wish to do so, delete this
# exception statement from your version.
## Process this file with automake to produce Makefile.in
website = index.xhtml documentation.xhtml download.xhtml contact.xhtml \
format/web.css format/scute-logo.svg format/scute-border.jpg
EXTRA_DIST = $(addprefix website/, $(website))
+
+SUBDIRS = manual
\ No newline at end of file
diff --git a/doc/Makefile.am b/doc/manual/Makefile.am
similarity index 88%
copy from doc/Makefile.am
copy to doc/manual/Makefile.am
index 6f62966..7acda8f 100644
--- a/doc/Makefile.am
+++ b/doc/manual/Makefile.am
@@ -1,38 +1,38 @@
-# Makefile.am
+# Makefile.am - Doc directory Makefile for scute.
# Copyright (C) 2006 g10 Code GmbH
#
# This file is part of scute[1].
#
# [1] Derived from the RSA Security Inc. PKCS #11 Cryptographic Token
# Interface (Cryptoki).
#
# Scute 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.
#
# Scute 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 Scute; if not, write to the Free Software Foundation,
# Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
#
# In addition, as a special exception, g10 Code GmbH gives permission
# to link this library: with the Mozilla Foundation's code for
# Mozilla (or with modified versions of it that use the same license
# as the "Mozilla" code), and distribute the linked executables. You
# must obey the GNU General Public License in all respects for all of
# the code used other than "Mozilla". If you modify this file, you
# may extend this exception to your version of the file, but you are
# not obligated to do so. If you do not wish to do so, delete this
# exception statement from your version.
## Process this file with automake to produce Makefile.in
-website = index.xhtml documentation.xhtml download.xhtml contact.xhtml \
- format/web.css format/scute-logo.svg format/scute-border.jpg
+DISTCLEANFILES = scute.tmp
-EXTRA_DIST = $(addprefix website/, $(website))
+info_TEXINFOS = scute.texi
+scute_TEXINFOS = gpl.texi
diff --git a/doc/manual/firefox-cm-view-detail.png b/doc/manual/firefox-cm-view-detail.png
new file mode 100644
index 0000000..dbfa446
Binary files /dev/null and b/doc/manual/firefox-cm-view-detail.png differ
diff --git a/doc/manual/firefox-cm-view.png b/doc/manual/firefox-cm-view.png
new file mode 100644
index 0000000..3c6ed0a
Binary files /dev/null and b/doc/manual/firefox-cm-view.png differ
diff --git a/doc/manual/firefox-cm.png b/doc/manual/firefox-cm.png
new file mode 100644
index 0000000..881d0f7
Binary files /dev/null and b/doc/manual/firefox-cm.png differ
diff --git a/doc/manual/firefox-dm-load-after.png b/doc/manual/firefox-dm-load-after.png
new file mode 100644
index 0000000..04fc2b6
Binary files /dev/null and b/doc/manual/firefox-dm-load-after.png differ
diff --git a/doc/manual/firefox-dm-load-before.png b/doc/manual/firefox-dm-load-before.png
new file mode 100644
index 0000000..3d406bd
Binary files /dev/null and b/doc/manual/firefox-dm-load-before.png differ
diff --git a/doc/manual/firefox-dm-load.png b/doc/manual/firefox-dm-load.png
new file mode 100644
index 0000000..df8fc0d
Binary files /dev/null and b/doc/manual/firefox-dm-load.png differ
diff --git a/doc/manual/firefox-dm-token-present.png b/doc/manual/firefox-dm-token-present.png
new file mode 100644
index 0000000..0813139
Binary files /dev/null and b/doc/manual/firefox-dm-token-present.png differ
diff --git a/doc/manual/firefox-pref-view.png b/doc/manual/firefox-pref-view.png
new file mode 100644
index 0000000..14d8e82
Binary files /dev/null and b/doc/manual/firefox-pref-view.png differ
diff --git a/doc/manual/firefox-pref.png b/doc/manual/firefox-pref.png
new file mode 100644
index 0000000..1361680
Binary files /dev/null and b/doc/manual/firefox-pref.png differ
diff --git a/doc/manual/gpl.texi b/doc/manual/gpl.texi
new file mode 100644
index 0000000..ca0508f
--- /dev/null
+++ b/doc/manual/gpl.texi
@@ -0,0 +1,397 @@
+@node Copying
+@appendix GNU GENERAL PUBLIC LICENSE
+
+@cindex GPL, GNU General Public License
+@center Version 2, June 1991
+
+@display
+Copyright @copyright{} 1989, 1991 Free Software Foundation, Inc.
+59 Temple Place -- Suite 330, Boston, MA 02111-1307, USA
+
+Everyone is permitted to copy and distribute verbatim copies
+of this license document, but changing it is not allowed.
+@end display
+
+@appendixsubsec Preamble
+
+ The licenses for most software are designed to take away your
+freedom to share and change it. By contrast, the GNU General Public
+License is intended to guarantee your freedom to share and change free
+software---to make sure the software is free for all its users. This
+General Public License applies to most of the Free Software
+Foundation's software and to any other program whose authors commit to
+using it. (Some other Free Software Foundation software is covered by
+the GNU Library General Public License instead.) You can apply it to
+your programs, too.
+
+ When we speak of free software, we are referring to freedom, not
+price. Our General Public Licenses are designed to make sure that you
+have the freedom to distribute copies of free software (and charge for
+this service if you wish), that you receive source code or can get it
+if you want it, that you can change the software or use pieces of it
+in new free programs; and that you know you can do these things.
+
+ To protect your rights, we need to make restrictions that forbid
+anyone to deny you these rights or to ask you to surrender the rights.
+These restrictions translate to certain responsibilities for you if you
+distribute copies of the software, or if you modify it.
+
+ For example, if you distribute copies of such a program, whether
+gratis or for a fee, you must give the recipients all the rights that
+you have. You must make sure that they, too, receive or can get the
+source code. And you must show them these terms so they know their
+rights.
+
+ We protect your rights with two steps: (1) copyright the software, and
+(2) offer you this license which gives you legal permission to copy,
+distribute and/or modify the software.
+
+ Also, for each author's protection and ours, we want to make certain
+that everyone understands that there is no warranty for this free
+software. If the software is modified by someone else and passed on, we
+want its recipients to know that what they have is not the original, so
+that any problems introduced by others will not reflect on the original
+authors' reputations.
+
+ Finally, any free program is threatened constantly by software
+patents. We wish to avoid the danger that redistributors of a free
+program will individually obtain patent licenses, in effect making the
+program proprietary. To prevent this, we have made it clear that any
+patent must be licensed for everyone's free use or not licensed at all.
+
+ The precise terms and conditions for copying, distribution and
+modification follow.
+
+@iftex
+@appendixsubsec TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION
+@end iftex
+@ifinfo
+@center TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION
+@end ifinfo
+
+@enumerate
+@item
+This License applies to any program or other work which contains
+a notice placed by the copyright holder saying it may be distributed
+under the terms of this General Public License. The ``Program'', below,
+refers to any such program or work, and a ``work based on the Program''
+means either the Program or any derivative work under copyright law:
+that is to say, a work containing the Program or a portion of it,
+either verbatim or with modifications and/or translated into another
+language. (Hereinafter, translation is included without limitation in
+the term ``modification''.) Each licensee is addressed as ``you''.
+
+Activities other than copying, distribution and modification are not
+covered by this License; they are outside its scope. The act of
+running the Program is not restricted, and the output from the Program
+is covered only if its contents constitute a work based on the
+Program (independent of having been made by running the Program).
+Whether that is true depends on what the Program does.
+
+@item
+You may copy and distribute verbatim copies of the Program's
+source code as you receive it, in any medium, provided that you
+conspicuously and appropriately publish on each copy an appropriate
+copyright notice and disclaimer of warranty; keep intact all the
+notices that refer to this License and to the absence of any warranty;
+and give any other recipients of the Program a copy of this License
+along with the Program.
+
+You may charge a fee for the physical act of transferring a copy, and
+you may at your option offer warranty protection in exchange for a fee.
+
+@item
+You may modify your copy or copies of the Program or any portion
+of it, thus forming a work based on the Program, and copy and
+distribute such modifications or work under the terms of Section 1
+above, provided that you also meet all of these conditions:
+
+@enumerate a
+@item
+You must cause the modified files to carry prominent notices
+stating that you changed the files and the date of any change.
+
+@item
+You must cause any work that you distribute or publish, that in
+whole or in part contains or is derived from the Program or any
+part thereof, to be licensed as a whole at no charge to all third
+parties under the terms of this License.
+
+@item
+If the modified program normally reads commands interactively
+when run, you must cause it, when started running for such
+interactive use in the most ordinary way, to print or display an
+announcement including an appropriate copyright notice and a
+notice that there is no warranty (or else, saying that you provide
+a warranty) and that users may redistribute the program under
+these conditions, and telling the user how to view a copy of this
+License. (Exception: if the Program itself is interactive but
+does not normally print such an announcement, your work based on
+the Program is not required to print an announcement.)
+@end enumerate
+
+These requirements apply to the modified work as a whole. If
+identifiable sections of that work are not derived from the Program,
+and can be reasonably considered independent and separate works in
+themselves, then this License, and its terms, do not apply to those
+sections when you distribute them as separate works. But when you
+distribute the same sections as part of a whole which is a work based
+on the Program, the distribution of the whole must be on the terms of
+this License, whose permissions for other licensees extend to the
+entire whole, and thus to each and every part regardless of who wrote it.
+
+Thus, it is not the intent of this section to claim rights or contest
+your rights to work written entirely by you; rather, the intent is to
+exercise the right to control the distribution of derivative or
+collective works based on the Program.
+
+In addition, mere aggregation of another work not based on the Program
+with the Program (or with a work based on the Program) on a volume of
+a storage or distribution medium does not bring the other work under
+the scope of this License.
+
+@item
+You may copy and distribute the Program (or a work based on it,
+under Section 2) in object code or executable form under the terms of
+Sections 1 and 2 above provided that you also do one of the following:
+
+@enumerate a
+@item
+Accompany it with the complete corresponding machine-readable
+source code, which must be distributed under the terms of Sections
+1 and 2 above on a medium customarily used for software interchange; or,
+
+@item
+Accompany it with a written offer, valid for at least three
+years, to give any third party, for a charge no more than your
+cost of physically performing source distribution, a complete
+machine-readable copy of the corresponding source code, to be
+distributed under the terms of Sections 1 and 2 above on a medium
+customarily used for software interchange; or,
+
+@item
+Accompany it with the information you received as to the offer
+to distribute corresponding source code. (This alternative is
+allowed only for noncommercial distribution and only if you
+received the program in object code or executable form with such
+an offer, in accord with Subsection b above.)
+@end enumerate
+
+The source code for a work means the preferred form of the work for
+making modifications to it. For an executable work, complete source
+code means all the source code for all modules it contains, plus any
+associated interface definition files, plus the scripts used to
+control compilation and installation of the executable. However, as a
+special exception, the source code distributed need not include
+anything that is normally distributed (in either source or binary
+form) with the major components (compiler, kernel, and so on) of the
+operating system on which the executable runs, unless that component
+itself accompanies the executable.
+
+If distribution of executable or object code is made by offering
+access to copy from a designated place, then offering equivalent
+access to copy the source code from the same place counts as
+distribution of the source code, even though third parties are not
+compelled to copy the source along with the object code.
+
+@item
+You may not copy, modify, sublicense, or distribute the Program
+except as expressly provided under this License. Any attempt
+otherwise to copy, modify, sublicense or distribute the Program is
+void, and will automatically terminate your rights under this License.
+However, parties who have received copies, or rights, from you under
+this License will not have their licenses terminated so long as such
+parties remain in full compliance.
+
+@item
+You are not required to accept this License, since you have not
+signed it. However, nothing else grants you permission to modify or
+distribute the Program or its derivative works. These actions are
+prohibited by law if you do not accept this License. Therefore, by
+modifying or distributing the Program (or any work based on the
+Program), you indicate your acceptance of this License to do so, and
+all its terms and conditions for copying, distributing or modifying
+the Program or works based on it.
+
+@item
+Each time you redistribute the Program (or any work based on the
+Program), the recipient automatically receives a license from the
+original licensor to copy, distribute or modify the Program subject to
+these terms and conditions. You may not impose any further
+restrictions on the recipients' exercise of the rights granted herein.
+You are not responsible for enforcing compliance by third parties to
+this License.
+
+@item
+If, as a consequence of a court judgment or allegation of patent
+infringement or for any other reason (not limited to patent issues),
+conditions are imposed on you (whether by court order, agreement or
+otherwise) that contradict the conditions of this License, they do not
+excuse you from the conditions of this License. If you cannot
+distribute so as to satisfy simultaneously your obligations under this
+License and any other pertinent obligations, then as a consequence you
+may not distribute the Program at all. For example, if a patent
+license would not permit royalty-free redistribution of the Program by
+all those who receive copies directly or indirectly through you, then
+the only way you could satisfy both it and this License would be to
+refrain entirely from distribution of the Program.
+
+If any portion of this section is held invalid or unenforceable under
+any particular circumstance, the balance of the section is intended to
+apply and the section as a whole is intended to apply in other
+circumstances.
+
+It is not the purpose of this section to induce you to infringe any
+patents or other property right claims or to contest validity of any
+such claims; this section has the sole purpose of protecting the
+integrity of the free software distribution system, which is
+implemented by public license practices. Many people have made
+generous contributions to the wide range of software distributed
+through that system in reliance on consistent application of that
+system; it is up to the author/donor to decide if he or she is willing
+to distribute software through any other system and a licensee cannot
+impose that choice.
+
+This section is intended to make thoroughly clear what is believed to
+be a consequence of the rest of this License.
+
+@item
+If the distribution and/or use of the Program is restricted in
+certain countries either by patents or by copyrighted interfaces, the
+original copyright holder who places the Program under this License
+may add an explicit geographical distribution limitation excluding
+those countries, so that distribution is permitted only in or among
+countries not thus excluded. In such case, this License incorporates
+the limitation as if written in the body of this License.
+
+@item
+The Free Software Foundation may publish revised and/or new versions
+of the General Public License from time to time. Such new versions will
+be similar in spirit to the present version, but may differ in detail to
+address new problems or concerns.
+
+Each version is given a distinguishing version number. If the Program
+specifies a version number of this License which applies to it and ``any
+later version'', you have the option of following the terms and conditions
+either of that version or of any later version published by the Free
+Software Foundation. If the Program does not specify a version number of
+this License, you may choose any version ever published by the Free Software
+Foundation.
+
+@item
+If you wish to incorporate parts of the Program into other free
+programs whose distribution conditions are different, write to the author
+to ask for permission. For software which is copyrighted by the Free
+Software Foundation, write to the Free Software Foundation; we sometimes
+make exceptions for this. Our decision will be guided by the two goals
+of preserving the free status of all derivatives of our free software and
+of promoting the sharing and reuse of software generally.
+
+@iftex
+@heading NO WARRANTY
+@end iftex
+@ifinfo
+@center NO WARRANTY
+@end ifinfo
+
+@item
+BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY
+FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT WHEN
+OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES
+PROVIDE THE PROGRAM ``AS IS'' WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED
+OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
+MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS
+TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE
+PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING,
+REPAIR OR CORRECTION.
+
+@item
+IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING
+WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY AND/OR
+REDISTRIBUTE THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES,
+INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING
+OUT OF THE USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED
+TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY
+YOU OR THIRD PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER
+PROGRAMS), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGES.
+@end enumerate
+
+@iftex
+@heading END OF TERMS AND CONDITIONS
+@end iftex
+@ifinfo
+@center END OF TERMS AND CONDITIONS
+@end ifinfo
+
+@page
+@unnumberedsec How to Apply These Terms to Your New Programs
+
+ If you develop a new program, and you want it to be of the greatest
+possible use to the public, the best way to achieve this is to make it
+free software which everyone can redistribute and change under these terms.
+
+ To do so, attach the following notices to the program. It is safest
+to attach them to the start of each source file to most effectively
+convey the exclusion of warranty; and each file should have at least
+the ``copyright'' line and a pointer to where the full notice is found.
+
+@smallexample
+@var{one line to give the program's name and an idea of what it does.}
+Copyright (C) 19@var{yy} @var{name of author}
+
+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.,
+59 Temple Place, Suite 330, Boston, MA 02111-1307, USA.
+@end smallexample
+
+Also add information on how to contact you by electronic and paper mail.
+
+If the program is interactive, make it output a short notice like this
+when it starts in an interactive mode:
+
+@smallexample
+Gnomovision version 69, Copyright (C) 19@var{yy} @var{name of author}
+Gnomovision comes with ABSOLUTELY NO WARRANTY; for details
+type `show w'. This is free software, and you are welcome
+to redistribute it under certain conditions; type `show c'
+for details.
+@end smallexample
+
+The hypothetical commands @samp{show w} and @samp{show c} should show
+the appropriate parts of the General Public License. Of course, the
+commands you use may be called something other than @samp{show w} and
+@samp{show c}; they could even be mouse-clicks or menu items---whatever
+suits your program.
+
+You should also get your employer (if you work as a programmer) or your
+school, if any, to sign a ``copyright disclaimer'' for the program, if
+necessary. Here is a sample; alter the names:
+
+@smallexample
+@group
+Yoyodyne, Inc., hereby disclaims all copyright
+interest in the program `Gnomovision'
+(which makes passes at compilers) written
+by James Hacker.
+
+@var{signature of Ty Coon}, 1 April 1989
+Ty Coon, President of Vice
+@end group
+@end smallexample
+
+This General Public License does not permit incorporating your program into
+proprietary programs. If your program is a subroutine library, you may
+consider it more useful to permit linking proprietary applications with the
+library. If this is what you want to do, use the GNU Library General
+Public License instead of this License.
diff --git a/doc/manual/mdate-sh b/doc/manual/mdate-sh
new file mode 100755
index 0000000..b610b47
--- /dev/null
+++ b/doc/manual/mdate-sh
@@ -0,0 +1,133 @@
+#!/bin/sh
+# Get modification time of a file or directory and pretty-print it.
+# Copyright (C) 1995, 1996, 1997, 2003 Free Software Foundation, Inc.
+# written by Ulrich Drepper <drepper@gnu.ai.mit.edu>, June 1995
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 2, or (at your option)
+# any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program; if not, write to the Free Software Foundation,
+# Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+
+# As a special exception to the GNU General Public License, if you
+# distribute this file as part of a program that contains a
+# configuration script generated by Autoconf, you may include it under
+# the same distribution terms that you use for the rest of that program.
+
+# Prevent date giving response in another language.
+LANG=C
+export LANG
+LC_ALL=C
+export LC_ALL
+LC_TIME=C
+export LC_TIME
+
+save_arg1="$1"
+
+# Find out how to get the extended ls output of a file or directory.
+if ls -L /dev/null 1>/dev/null 2>&1; then
+ ls_command='ls -L -l -d'
+else
+ ls_command='ls -l -d'
+fi
+
+# A `ls -l' line looks as follows on OS/2.
+# drwxrwx--- 0 Aug 11 2001 foo
+# This differs from Unix, which adds ownership information.
+# drwxrwx--- 2 root root 4096 Aug 11 2001 foo
+#
+# To find the date, we split the line on spaces and iterate on words
+# until we find a month. This cannot work with files whose owner is a
+# user named `Jan', or `Feb', etc. However, it's unlikely that `/'
+# will be owned by a user whose name is a month. So we first look at
+# the extended ls output of the root directory to decide how many
+# words should be skipped to get the date.
+
+# On HPUX /bin/sh, "set" interprets "-rw-r--r--" as options, so the "x" below.
+set - x`$ls_command /`
+
+# Find which argument is the month.
+month=
+command=
+until test $month
+do
+ shift
+ # Add another shift to the command.
+ command="$command shift;"
+ case $1 in
+ Jan) month=January; nummonth=1;;
+ Feb) month=February; nummonth=2;;
+ Mar) month=March; nummonth=3;;
+ Apr) month=April; nummonth=4;;
+ May) month=May; nummonth=5;;
+ Jun) month=June; nummonth=6;;
+ Jul) month=July; nummonth=7;;
+ Aug) month=August; nummonth=8;;
+ Sep) month=September; nummonth=9;;
+ Oct) month=October; nummonth=10;;
+ Nov) month=November; nummonth=11;;
+ Dec) month=December; nummonth=12;;
+ esac
+done
+
+# Get the extended ls output of the file or directory.
+set - x`eval "$ls_command \"\$save_arg1\""`
+
+# Remove all preceding arguments
+eval $command
+
+# Get the month. Next argument is day, followed by the year or time.
+case $1 in
+ Jan) month=January; nummonth=1;;
+ Feb) month=February; nummonth=2;;
+ Mar) month=March; nummonth=3;;
+ Apr) month=April; nummonth=4;;
+ May) month=May; nummonth=5;;
+ Jun) month=June; nummonth=6;;
+ Jul) month=July; nummonth=7;;
+ Aug) month=August; nummonth=8;;
+ Sep) month=September; nummonth=9;;
+ Oct) month=October; nummonth=10;;
+ Nov) month=November; nummonth=11;;
+ Dec) month=December; nummonth=12;;
+esac
+
+day=$2
+
+# Here we have to deal with the problem that the ls output gives either
+# the time of day or the year.
+case $3 in
+ *:*) set `date`; eval year=\$$#
+ case $2 in
+ Jan) nummonthtod=1;;
+ Feb) nummonthtod=2;;
+ Mar) nummonthtod=3;;
+ Apr) nummonthtod=4;;
+ May) nummonthtod=5;;
+ Jun) nummonthtod=6;;
+ Jul) nummonthtod=7;;
+ Aug) nummonthtod=8;;
+ Sep) nummonthtod=9;;
+ Oct) nummonthtod=10;;
+ Nov) nummonthtod=11;;
+ Dec) nummonthtod=12;;
+ esac
+ # For the first six month of the year the time notation can also
+ # be used for files modified in the last year.
+ if (expr $nummonth \> $nummonthtod) > /dev/null;
+ then
+ year=`expr $year - 1`
+ fi;;
+ *) year=$3;;
+esac
+
+# The result.
+echo $day $month $year
diff --git a/doc/manual/scute.texi b/doc/manual/scute.texi
new file mode 100644
index 0000000..999dc4b
--- /dev/null
+++ b/doc/manual/scute.texi
@@ -0,0 +1,782 @@
+\input texinfo @c -*- Texinfo -*-
+@setfilename scute.info
+@settitle The Scute Manual
+
+@dircategory GNU Utilities
+@direntry
+* Scute: (scute). PKCS #11 module for the OpenPGP Card
+@end direntry
+
+@include version.texi
+
+@c Unify some of the indices.
+@syncodeindex tp fn
+@syncodeindex pg fn
+
+@ifinfo
+This file documents the Scute module.
+
+This is Edition @value{EDITION}, last updated @value{UPDATED}, of
+@cite{The Scute Manual}, for Version @value{VERSION}.
+
+@c NOTE: Don't forget to update the year for the TeX version, too.
+Copyright @copyright{} 2002, 2003, 2004, 2005, 2006 g10 Code GmbH.
+
+The Scute Manual 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.
+
+The Scute Manual 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 Lesser General Public License
+along with this program; if not, write to the Free Software Foundation,
+Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA
+
+@end ifinfo
+
+@iftex
+@shorttitlepage The Scute Manual
+@end iftex
+@titlepage
+@center @titlefont{The Scute Manual}
+@sp 6
+@center Edition @value{EDITION}
+@sp 1
+@center last updated @value{UPDATED}
+@sp 1
+@center for version @value{VERSION}
+@page
+@vskip 0pt plus 1filll
+Copyright @copyright{} 2002, 2003, 2004, 2005, 2006 g10 Code GmbH.
+
+
+The Scute Manual 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.
+
+The Scute Manual 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 Lesser General Public License
+along with this program; if not, write to the Free Software Foundation,
+Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA
+@end titlepage
+@page
+
+@ifnottex
+@node Top
+@top Main Menu
+This is Edition @value{EDITION}, last updated @value{UPDATED}, of
+@cite{The Scute Manual}, for Version @value{VERSION} of the Scute
+module.
+@end ifnottex
+
+@menu
+* Introduction:: How to use this manual.
+* Preparation:: What you should do before using Scute.
+* Client Authentication:: How to use Scute for client authentication.
+* Troubleshooting:: What to do when things go wrong.
+* Internals:: Technical details about Scute.
+
+Appendices
+
+* Copying:: The GNU General Public License says
+ how you can copy and share Scute.
+
+Indices
+
+* Concept Index:: Index of concepts and programs.
+
+
+@detailmenu
+ --- The Detailed Node Listing ---
+
+Introduction
+
+* Getting Started:: Purpose of the manual, and how to use it.
+* Features:: Reasons to install and use Scute.
+* Overview:: Basic architecture of the Scute module.
+
+Preparation
+
+* Prerequisites:: Dependencies of Scute.
+* Building the Source:: How to build Scute.
+* Certificate Preparation:: Preparing certificates for use with Scute.
+
+Client Authentication
+
+* Application Configuration:: Preparing the application for use with Scute.
+* Authentication With Service:: Using Scute for client authentication.
+
+Troubleshooting
+
+Internals
+
+* Features and Limitations:: Details about the interfaces implemented.
+* Developing Scute:: How to understand or improve Scute.
+* Mozilla Compatibility:: Interoperability notes for Mozilla NSS.
+
+
+@end detailmenu
+@end menu
+
+@node Introduction
+@chapter Introduction
+
+This is a PKCS #11 implementation for the GnuPG Agent using the GnuPG
+Smart Card Daemon. Currently, only the OpenPGP card is supported.
+
+Scute enables use of the OpenPGP smart card in applications supporting
+PKCS #11 compliant security tokens. The main application at this time
+is client authentication in Mozilla-based web browsers. In the future,
+other applications will be supported.
+
+
+@menu
+* Getting Started:: Purpose of the manual, and how to use it.
+* Features:: Reasons to install and use Scute.
+* Overview:: Basic architecture of the Scute module.
+@end menu
+
+
+@node Getting Started
+@section Getting Started
+
+This manual documents the Scute module, how it can be used for common
+applications supported by it, and how it can be extended and improved by
+programmers. It is thus a user manual as well as a developer manual.
+
+The reader is assumed to possess basic knowledge about cryptography in
+general, and public key cryptography in particular. The underlying
+cryptographic engines that are used by the library are not explained,
+but where necessary, special features or requirements are provided.
+
+This manual can be used in several ways. If read from the beginning to
+the end, it gives a good introduction into the module and how it can be
+used in an application. Forward references are included where
+necessary. Later on, the manual can be used as a reference manual to
+get just the information needed about any particular application of the
+module.
+
+
+@node Features
+@section Features
+
+Scute is currently the only implementation of PKCS #11 for the OpenPGP
+smart card. Apart from that, it offers a couple of other benefits:
+
+@table @asis
+@item it's free software
+Anybody can use, modify, and redistribute it under the terms of the GNU
+General Public License (@pxref{Copying}).
+
+@item it's built to grow
+Although Scute currently only provides a single function, client
+authentication using OpenPGP smart cards in Mozilla-based web browsers,
+it was built with the intention of supporting other applications as well
+in the future.
+
+@item it's easy
+Building and installing Scute is easy, and preparing smart cards for use
+with Scute is a snatch using the GnuPG 2.0 framework. The integration
+of Scute into the application is seamless.
+@end table
+
+
+@node Overview
+@section Overview
+
+Scute is a security device that implements the PKCS #11 interface for
+security tokens. Applications which know how to use the PKCS #11
+interface to access security tokens for crytpographic operations can use
+Scute to access the OpenPGP smart card. An important example of such an
+application is the Firefox web browser by the Mozilla project, which
+uses the Mozilla Network Security Services library (NSS).
+
+Scute itself does not include a driver for the smart card itself.
+Instead, it uses the GnuPG 2.0 framework to access the smart cards and
+associated data like certificates. Scute acts as the glue between the
+application and GnuPG 2.0.
+
+Currently, only client authentication over HTTPS with Firefox using the
+OpenPGP card is supported. In this configuration, Scute allows users to
+authenticate themselves to a remote web service without entering their
+log-in information.
+
+
+@node Preparation
+@chapter Preparation
+
+To use Scute, you first have to install the software. You also have to
+prepare each card you want to use with Scute before it can be used.
+Furthermore, you need to configure the application to make use of Scute
+for cryptographic operations. This chapter explains each of these steps
+in detail.
+
+@menu
+* Prerequisites:: Dependencies of Scute.
+* Building the Source:: How to build Scute.
+* Certificate Preparation:: Preparing certificates for use with Scute.
+@end menu
+
+
+@node Prerequisites
+@section Prerequisites
+
+There are two types of dependencies for Scute: compile-time dependencies
+and run-time dependencies. The compile-time dependencies only need to
+be fulfilled when Scute is compiled and installed. The run-time
+dependencies need to be fulfilled when Scute is used in an application.
+
+Scute depends, in addition to the essential build utilities, on the
+following packages at build time:
+
+@table @code
+@item libgpg-error
+Scute uses the GnuPG 2.0 framework for error handling, so it depends on
+the GPG error library. The minimum version required is 0.7.
+
+@item libassuan
+Scute uses the GnuPG 2.0 framework for communication with the GPG Agent,
+so it depends on the Assuan library. The minimum version required is
+0.6.10.
+@end table
+
+At run-time, in addition to the run-time versions of the above
+libraries, you also need the following packages installed and
+configured:
+
+@table @asis
+@item GnuPG
+Scute uses the GnuPG 2.0 framework to access the OpenPGP card and for
+certificate management. The minimum version required is 2.0.0.
+
+@strong{Caution:} A functional installation of GnuPG 2.0 requires a
+running GPG Agent process, which must be advertised to the applications
+via the @code{GPG_AGENT_INFO} environment variable. Please make sure
+that you fulfill this requirement before using Scute in an application,
+running the Scute test suite, or preparing certificates as described in
+@ref{Certificate Preparation}. @xref{Invoking GPG-AGENT, , , gnupg,
+Using the GNU Privacy Guard}, for details on how to run the GPG Agent.
+
+@item Pinentry
+Pinentry is a dependency of GnuPG 2.0, so it also needs to be installed
+with it.
+
+@item Firefox et al.
+Firefox is the first application supported by Scute. In the future,
+other applications may be supported. The applications are not
+dependencies of Scute, but Scute can not be used stand-alone, so you can
+not experience it without an application.
+@end table
+
+
+@node Building the Source
+@section Building the Source
+
+Scute does comply to the GNU coding standards and thus can be compiled
+and installed according to the generic installation instructions found
+in the source package in the file @code{INSTALL}. There are no Scute
+specific options to the configure script.
+
+After installation, the @code{libscute.so} module file can be found in
+the library directory of the installation path.
+
+
+@node Certificate Preparation
+@section Certificate Preparation
+
+To use an OpenPGP card with Scute, it first has to be initialized by
+generating or loading a key on the card, see
+@uref{http://www.gnupg.org/(en)/howtos/card-howto/en/smartcard-howto.html,
+the OpenPGP Card How-To}. Then a certificate has to be created and
+imported into GPGSM. This task involves three steps: First, a
+certificate signing request (@acronym{CSR}) has to be created that
+matches the key on the card. This certificate signing request then has
+to be submitted to a certificate authority (@acronym{CA}), which will
+create the certificate and send it back to you. At last, the
+certificate has to be imported into GPGSM. This section will explain
+all of these steps in detail.
+
+Before you start, make sure that the GPG Agent is running, see
+@ref{Prerequisites}. There is no need to configure GPGSM, so you can
+create a CSR with the command:
+
+@example
+$ gpgsm-gencert.sh > floppy-head.p10
+Key type
+ [1] RSA
+ [2] Existing key
+ [3] Direct from card
+Your selection: 3
+You selected: Direct from card
+@end example
+
+As we create a certificate for the OpenPGP Card, the option ``@code{[3]
+Direct from card}'' should be selected.
+
+@example
+Card with S/N D27600012401010100010000051B0000 found
+gpg-agent uses OPENPGP.3 as ssh key
+Select key
+ [1] OPENPGP.1
+ [2] OPENPGP.2
+ [3] OPENPGP.3
+ [4] back
+Your selection: 3
+You selected: OPENPGP.3
+Key usage
+ [1] sign, encrypt
+ [2] sign
+ [3] encrypt
+Your selection: 2
+You selected: sign
+@end example
+
+The only operation currently supported is client authentication. For
+this, the authentication key has to be selected. This is the third key
+on the card, so the options ``@code{[3] OPENPGP.3}'' and ``@code{[2]
+sign}'' should be chosen. Note that the key usage is only advisory, and
+the CA may assign different capabilities.
+
+@example
+Name (DN)
+> CN=Floppy Head,OU="Webserver Team",O="Snake Oil, Ltd",L="Snake Town",ST="Snake Desert",C=XY
+E-Mail addresses (end with an empty line)
+> floppy.head@@example.com
+E-Mail addresses (end with an empty line)
+>
+DNS Names (optional; end with an empty line)
+>
+URIs (optional; end with an empty line)
+>
+@end example
+
+As a last step, the common name and e-mail address of the key owner need
+to be specified by you. The above are only an example for a fictious
+person working at a fictious company. DNS names are only meaningful for
+server certificates and thus should be left empty.
+
+We have now entered all required information and gpgsm will display what
+it has gathered and ask whether to create the certificate request:
+
+@example
+Parameters for certificate request to create:
+ 1 Key-Type: card:OPENPGP.3
+ 2 Key-Length:
+ 3 Key-Usage: sign
+ 4 Name-DN: CN=Floppy Head,OU="Webserver Team",O="Snake Oil, Ltd",L="Snake Town",ST="Snake Desert",C=XY
+ 5 Name-Email: floppy.head@@example.com
+
+Really create such a CSR?
+ [1] yes
+ [2] no
+Your selection: 1
+You selected: yes
+@end example
+
+GPGSM will now start working on creating the request. During this time
+you will be asked once for a passphrase to unprotect the authentication
+key on the card. A pop up window will appear to ask for it.
+
+When it is ready, you should see the final notice:
+
+@example
+ gpgsm: certificate request created
+@end example
+
+Now, you may look at the created request:
+
+@example
+$ cat floppy-head.p10
+-----BEGIN CERTIFICATE REQUEST-----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=
+-----END CERTIFICATE REQUEST-----
+$
+@end example
+
+The next step is to submit this certificate request to the CA, which can
+then create a certificate and send it back to you.
+
+If, for example, you use the CA @uref{http://www.cacert.org, CAcert},
+then you can log into your account at the CAcert website, choose
+``FIXME'', check ``FIXME'', paste the above request block into the text
+field and click on ``Submit''. If everything works correctly, a
+certificate will be shown, which you can cut and paste into a new file
+@file{floppy-head.crt}.
+
+Alternatively if, for example, you set up your own CA with OpenSSL, then
+you can create your own certificate by issueing a command similar
+@code{openssl ca -in floppy-head.p10 -cert snakeoil-ca-rsa.crt -keyfile
+snakeoil-ca-rsa.key -out floppy-head.crt}. Please see the OpenSSL
+documentation for more details on how to set up and administrate a
+certificate authority infrastructure.
+
+In any way you should end up with a certificate file
+@file{floppy-head.crt}, which you then have to import into GPGSM. It is
+also recommended that you import the root certificate of the CA first in
+the same fashion.
+
+@example
+$ gpgsm --import floppy-head.crt
+gpgsm: certificate imported
+
+gpgsm: total number processed: 1
+gpgsm: imported: 1
+@end example
+
+gpgsm tells you that it has imported the certificate. It is now
+associated with the key you used when creating the request. To see the
+content of your certificate, you may now enter:
+
+@example
+$ gpgsm -K Floppy
+/home/foo/.gnupg/pubring.kbx
+---------------------------
+Serial number: 10
+ Issuer: /CN=Snake Oil CA/OU=Certificate Authority/O=Snake Oil, Ltd/L=Snake Town/ST=Snake Desert/C=XY/EMail=ca@@snakeoil.dom
+ Subject: /CN=Floppy Head/OU=Webserver Team/O=Snake Oil, Ltd/ST=Snake Desert/C=XY
+ validity: 2006-11-11 14:09:12 through 2007-11-11 14:09:12
+ key type: 1024 bit RSA
+ fingerprint: EC:93:A2:55:C6:58:7F:C9:9E:96:DB:12:6E:64:99:54:BB:E1:94:68
+@end example
+
+The option ``@code{-K}'' is used above because this will only list
+certificates for which a private key is available. To see more details,
+you may use ``@code{--dump-secret-keys}'' instead of ``@code{-K}''.
+
+
+@node Client Authentication
+@chapter Client Authentication
+
+@menu
+* Application Configuration:: Preparing the application for use with Scute.
+* Authentication With Service:: Using Scute for client authentication.
+@end menu
+
+Scute allows you to authenticate yourself to a website securely without
+entering a username or password by simply using your OpenPGP card.
+Currently, only Mozilla-based browsers like Firefox are supported,
+although other applications using Mozilla NSS or supporting PKCS #11
+modules may work.
+
+
+@node Application Configuration
+@section Application Configuration
+
+To prepare your application for use with Scute, you have to load the
+Scute module as a PKCS #11 module into the application. With Firefox,
+this can be done by choosing @code{Edit->Preferences} in the menu. In
+the preferences configuration dialog, you should select the
+@code{Advanced} configuration section, then the @code{Security} tab, and
+then select @code{Security Devices} in the category @code{Certificates}.
+
+@image{firefox-pref}
+
+In the devices manager dialog, you can select @code{Load} to load a new
+PKCS #11 device.
+
+@image{firefox-dm-load-before}
+
+In the pop-up dialog that follows, you can give a module name
+(e.g. ``@code{Scute}'') and a module filename. The latter should
+correspond to the full file name of the installed Scute module file
+@file{libscute.so}. The default installation path is
+@file{/usr/local/lib}, which would mean that you have to provide the
+file name @file{/usr/local/lib/libscute.so}. If you or your system
+administrator installed Scute in a different location, you have to
+adjust the file name correspondingly.
+
+@image{firefox-dm-load}
+
+After confirming installation of the security device, a pop-up window
+should confirm that the module was successfully loaded, and an entry for
+the security device should appear in the device manager list of
+@code{Security Modules and Devices}.
+
+@image{firefox-dm-load-after}
+
+When you insert the OpenPGP card for which you generated and imported a
+certificate earlier (see @ref{Certificate Preparation}), the device
+manager should detect this security token and display some information
+about it in the @code{Details} list when you select it from the module
+list.
+
+@image{firefox-dm-token-present}
+
+FIXME: Add some documentation and screenshots for the certificate
+manager.
+
+@node Authentication With Service
+@section Authentication With Service
+
+Before you access a web service which requires client authentication,
+for instance a fictious web service @indicateurl{https://example.com},
+the OpenPGP card should be present. In this case, a pop-up window will
+appear that requests you to enter the PIN number protecting the
+authentication key on the OpenPGP card. After entering the PIN number,
+your browser will be authenticated to the server. If the server accepts
+your request and certificate, this is all which is required. You should
+leave the card in the reader as long as the connection persists.
+Depending on how aggressively GPG Agent caches your PIN number, you may
+have to enter the PIN number again later to keep up the connection to
+the server.
+
+If the card is not present, or you enter the wrong PIN, or the server
+does not admit your certificate, you will get an error message. This
+error message is generated by the application and Scute can not
+influence it. Unfortunately, in Firefox 1, this error message is not
+very user friendly: ``@code{example.com has received an incorrect or
+unexpected message. Error code: -12227}''.
+
+FIXME: Screenshot.
+
+FIXME: Document possible error codes.
+
+
+@node Troubleshooting
+@chapter Troubleshooting
+
+@strong{Symptom:} Loading the Scute security device in the security
+device manager of Firefox fails with "Unable to load module".
+
+@strong{Solution:} Make sure that Scute is correctly installed, and that
+all libraries and executables are available. Make sure that the GPG
+Agent is running and can be found via the environment variable
+@code{GPG_AGENT_INFO}. @xref{Invoking GPG-AGENT, , , gnupg, Using the
+GNU Privacy Guard}, for details on how to run the GPG Agent.
+
+
+@strong{Symptom:} Client authentication fails with "<example.com> has
+received an incorrect or unexpected message. Error code: -12227".
+
+@strong{Solution:} Make sure that the correct OpenPGP card is inserted
+and the certificate available in GPGSM. Check that the OpenPGP card is
+detected correctly in the security device manager and the corresponding
+certificate is displayed in the certificate manager of Firefox.
+@xref{Authentication With Service}.
+
+
+@strong{Symptom:} The OpenPGP card is detected and displayed in the
+security device manager in Firefox, but no corresponding certificate is
+displayed in the certificate manager of Firefox.
+
+@strong{Solution:} Make sure that the corresponding certificate is
+imported in GPGSM.
+
+FIXME: Can this really happen???
+
+
+@node Internals
+@chapter Internals
+
+The following notes are intended for people interested in more technical
+details about Scute and its implementation. They give an overview about
+its scope and potential compatibility issues with applications.
+
+@menu
+* Features and Limitations:: Details about the interfaces implemented.
+* Developing Scute:: How to understand or improve Scute.
+* Mozilla Compatibility:: Interoperability notes for Mozilla NSS.
+@end menu
+
+
+@node Features and Limitations
+@section Features and Limitations
+
+Scute implements version 2.20 of the
+@uref{http://www.rsasecurity.com/rsalabs/node.asp?id=2133, PKCS #11}
+specification.
+
+The @uref{http://www.g10code.com/p-card.html,OpenPGP smart card}
+application is supported in read-only mode.
+
+The following functions are not supported:
+
+@table @code
+@item C_Initialize
+No support for native thread package. Locking callbacks must be
+provided if multi-threaded operation is desired.
+
+@item C_WaitForSlotEvent
+Not implemented. The interface as specified by PKCS #11 is broken
+anyway, as the function can not safely be canceled. Thus, we require
+polling.
+
+@item C_GetOperationState
+@itemx C_SetOperationState
+Not supported.
+
+@item C_InitToken
+@itemx
+
+@itemx C_InitPIN
+@itemx C_SetPIN
+Not supported. No write operations are allowed. To configure the
+token, please use the tools accompanying the GnuPG software suite.
+
+@item C_Login
+@itemx C_Logout
+Not supported. No login into the token by the software is required.
+Passphrase queries are implemented by the use of GPG Agent and Pinentry.
+
+@item C_EncryptInit
+@itemx C_Encrypt
+@itemx C_EncryptUpdate
+@itemx C_EncryptFinal
+@itemx C_DigestInit
+@itemx C_Digest
+@itemx C_DigestUpdate
+@itemx C_DigestKey
+@itemx C_DigestFinal
+@itemx C_VerifyInit
+@itemx C_Verify
+@itemx C_VerifyUpdate
+@itemx C_VerifyFinal
+@itemx C_VerifyRecoverInit
+@itemx C_VerifyRec
+Not supported. Only secret key operations are supported.
+
+@item C_SignInit
+@itemx C_Sign
+Currently, only signing 36 bytes (MD5+SHA1) hashes is supported (used
+for client authentication).
+
+@item C_DecryptInit
+@itemx C_Decrypt
+Not yet supported, but will be in the future.
+
+@item C_SignUpdate
+@itemx C_SignFinal
+@itemx C_DecryptUpdate
+@itemx C_DecryptFinal
+No progressive crypto-operations are supported.
+
+@item C_SignRecoverInit
+@itemx C_SignRecover
+Not supported.
+
+@item C_DigestEncryptUpdate
+@itemx C_DecryptDigestUpdate
+@itemx C_SignEncryptUpdate
+@itemx C_DecryptVerifyUpdate
+Dual-purpose cryptographic functions are not supported.
+
+@item C_GenerateKey
+@itemx C_GenerateKeyPair
+@itemx C_WrapKey
+@itemx C_UnwrapKey
+@itemx C_DeriveKey
+Key management functions are not supported. Please use the tools
+accompanying the GnuPG software suite to generate and import keys for
+use with the token.
+
+@item C_SeedRandom
+@itemx C_GenerateRandom
+Not supported at this point. @code{C_GenerateRandom} may be supported
+in the future, though.
+
+@item C_CreateObject
+@itemx C_CopyObject
+@itemx C_DestroyObject
+@itemx C_SetAttributeValue:
+Only read-only operations are supported on objects.
+
+@item C_GetObjectSize
+Not supported.
+
+@item CKO_CERTIFICATE
+The label specifies the key on the card used (e.g. @code{OPENPGP.3}).
+The ID is the fingerprint.
+
+@item CKO_PRIVATE_KEY:
+The @code{CKA_LOCAL} attribute can not be supported by the OpenPGP card.
+It is always set to false (as the key on the card may be copied to the
+card from an external source).
+@end table
+
+@node Developing Scute
+@section Developing Scute
+
+Scute is single-threaded. There is a global lock that is taken in all
+entry points of Scute, except for @code{C_Initialize},
+@code{C_Finalize}, @code{C_GetFunctionList}, and stubs.
+
+Here are a couple of hints on how to develop PKCS #11 modules for
+Mozilla:
+
+@code{libopensc2} ships with a @code{pkcs11-spy} library that can be
+loaded as a wrapper around the PKCS #11 library you want to use to log
+all functions invoked by Mozilla. Here is how to use it:
+
+Set the @code{PKCS11SPY_OUTPUT} environment variable to a filename.
+@code{pkcs11-spy} appends its log messages at the end of this file. Set
+the @code{PKCS11SPY} environment variable to the filename of the PKCS
+#11 module you actually want to use. Start Mozilla within this
+environment.
+
+There is a different, probably more powerful way to debug Mozilla PKCS
+#11 libraries. However, to be able to use it, you need to configure and
+compile the Mozilla NSS sources with @code{--enable-debug}.
+Instructions can be found at:
+@uref{http://www.mozilla.org/projects/security/pki/nss/tech-notes/tn2.html}
+
+Here are a couple of links to more information about implementing a
+PKCS #11 module for Mozilla:
+
+@table @uref
+@item http://docs.sun.com/source/816-6150-10/index.htm
+@itemx http://docs.sun.com/source/816-6150-10/pkcs.htm
+Implementing PKCS #11 for the Netscape Security Library (Caution: The
+content may be out of date)
+
+@item http://www.mozilla.org/projects/security/pki/pkcs11/netscape/problems.html
+Common PKCS #11 Implementation Problems
+
+@item http://www.mozilla.org/projects/security/pki/pkcs11/
+PKCS #11 Conformance Testing
+
+@item http://www.mozilla.org/projects/security/pki/nss/
+The Mozilla NSS web page
+@end table
+
+
+@node Mozilla Compatibility
+@section Mozilla Compatibility
+
+Mozilla has a bug that causes the wrong security device to be unloaded
+when unloading a security device. Also, the displayed list becomes
+corrupt. When closing and reopening the security device manager, the
+list displayed is correct, but in anyway the wrong security module is
+unloaded.
+
+
+@include gpl.texi
+
+
+@node Concept Index
+@unnumbered Concept Index
+
+@printindex cp
+
+
+@summarycontents
+@contents
+@bye
diff --git a/doc/manual/texinfo.tex b/doc/manual/texinfo.tex
new file mode 100644
index 0000000..e9293f3
--- /dev/null
+++ b/doc/manual/texinfo.tex
@@ -0,0 +1,6773 @@
+% 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{2003-05-04.08}
+%
+% Copyright (C) 1985, 1986, 1988, 1990, 1991, 1992, 1993, 1994, 1995,
+% 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003 Free Software Foundation, Inc.
+%
+% This texinfo.tex file is free software; you can redistribute it and/or
+% modify it under the terms of the GNU General Public License as
+% published by the Free Software Foundation; either version 2, or (at
+% your option) any later version.
+%
+% This texinfo.tex file is distributed in the hope that it will be
+% useful, but WITHOUT ANY WARRANTY; without even the implied warranty
+% of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+% General Public License for more details.
+%
+% You should have received a copy of the GNU General Public License
+% along with this texinfo.tex file; see the file COPYING. If not, write
+% to the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
+% Boston, MA 02111-1307, USA.
+%
+% In other words, you are welcome to use, share and improve this program.
+% You are forbidden to forbid anyone else to use, share and improve
+% what you give them. Help stamp out software-hoarding!
+%
+% Please try the latest version of texinfo.tex before submitting bug
+% reports; you can get the latest version from:
+% ftp://ftp.gnu.org/gnu/texinfo/texinfo.tex
+% (and all GNU mirrors, see http://www.gnu.org/order/ftp.html)
+% ftp://tug.org/tex/texinfo.tex
+% (and all CTAN mirrors, see http://www.ctan.org),
+% and /home/gd/gnu/doc/texinfo.tex on the GNU machines.
+%
+% The GNU Texinfo home page is http://www.gnu.org/software/texinfo.
+%
+% The texinfo.tex in any given Texinfo 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.
+
+\message{Loading texinfo [version \texinfoversion]:}
+
+% If in a .fmt file, print the version number
+% and turn on active characters that we couldn't do earlier because
+% they might have appeared in the input file name.
+\everyjob{\message{[Texinfo version \texinfoversion]}%
+ \catcode`+=\active \catcode`\_=\active}
+
+\message{Basics,}
+\chardef\other=12
+
+% We never want plain's \outer definition of \+ in Texinfo.
+% For @tex, we can use \tabalign.
+\let\+ = \relax
+
+% Save some plain tex macros whose names we will redefine.
+\let\ptexb=\b
+\let\ptexbullet=\bullet
+\let\ptexc=\c
+\let\ptexcomma=\,
+\let\ptexdot=\.
+\let\ptexdots=\dots
+\let\ptexend=\end
+\let\ptexequiv=\equiv
+\let\ptexexclam=\!
+\let\ptexgtr=>
+\let\ptexhat=^
+\let\ptexi=\i
+\let\ptexindent=\indent
+\let\ptexlbrace=\{
+\let\ptexless=<
+\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
+
+% 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\putwordDeftypevar\undefined\gdef\putwordDeftypevar{Variable}\fi
+\ifx\putwordDeffunc\undefined \gdef\putwordDeffunc{Function}\fi
+\ifx\putwordDeftypefun\undefined\gdef\putwordDeftypefun{Function}\fi
+
+% In some macros, we cannot use the `\? notation---the left quote is
+% in some cases the escape char.
+\chardef\colonChar = `\:
+\chardef\commaChar = `\,
+\chardef\dotChar = `\.
+\chardef\equalChar = `\=
+\chardef\exclamChar= `\!
+\chardef\questChar = `\?
+\chardef\semiChar = `\;
+\chardef\spaceChar = `\ %
+\chardef\underChar = `\_
+
+% Ignore a token.
+%
+\def\gobble#1{}
+
+% True if #1 is the empty string, i.e., called like `\ifempty{}'.
+%
+\def\ifempty#1{\ifemptyx #1\emptymarkA\emptymarkB}%
+\def\ifemptyx#1#2\emptymarkB{\ifx #1\emptymarkA}%
+
+% Hyphenation fixes.
+\hyphenation{ap-pen-dix}
+\hyphenation{eshell}
+\hyphenation{mini-buf-fer mini-buf-fers}
+\hyphenation{time-stamp}
+\hyphenation{white-space}
+
+% Margin to add to right of even pages, to left of odd pages.
+\newdimen\bindingoffset
+\newdimen\normaloffset
+\newdimen\pagewidth \newdimen\pageheight
+
+% 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
+ \errorcontextlines\maxdimen
+}%
+
+% add check for \lastpenalty to plain's definitions. If the last thing
+% we did was a \nobreak, we don't want to insert more space.
+%
+\def\smallbreak{\ifnum\lastpenalty<10000\par\ifdim\lastskip<\smallskipamount
+ \removelastskip\penalty-50\smallskip\fi\fi}
+\def\medbreak{\ifnum\lastpenalty<10000\par\ifdim\lastskip<\medskipamount
+ \removelastskip\penalty-100\medskip\fi\fi}
+\def\bigbreak{\ifnum\lastpenalty<10000\par\ifdim\lastskip<\bigskipamount
+ \removelastskip\penalty-200\bigskip\fi\fi}
+
+% For @cropmarks command.
+% Do @cropmarks to get crop marks.
+%
+\newif\ifcropmarks
+\let\cropmarks = \cropmarkstrue
+%
+% Dimensions to add cropmarks at corners.
+% Added by P. A. MacKay, 12 Nov. 1986
+%
+\newdimen\outerhsize \newdimen\outervsize % set by the paper size routines
+\newdimen\cornerlong \cornerlong=1pc
+\newdimen\cornerthick \cornerthick=.3pt
+\newdimen\topandbottommargin \topandbottommargin=.75in
+
+% Main output routine.
+\chardef\PAGE = 255
+\output = {\onepageout{\pagecontents\PAGE}}
+
+\newbox\headlinebox
+\newbox\footlinebox
+
+% \onepageout takes a vbox as an argument. Note that \pagecontents
+% does insertions, but you have to call it yourself.
+\def\onepageout#1{%
+ \ifcropmarks \hoffset=0pt \else \hoffset=\normaloffset \fi
+ %
+ \ifodd\pageno \advance\hoffset by \bindingoffset
+ \else \advance\hoffset by -\bindingoffset\fi
+ %
+ % Do this outside of the \shipout so @code etc. will be expanded in
+ % the headline as they should be, not taken literally (outputting ''code).
+ \setbox\headlinebox = \vbox{\let\hsize=\pagewidth \makeheadline}%
+ \setbox\footlinebox = \vbox{\let\hsize=\pagewidth \makefootline}%
+ %
+ {%
+ % Have to do this stuff outside the \shipout because we want it to
+ % take effect in \write's, yet the group defined by the \vbox ends
+ % before the \shipout runs.
+ %
+ \escapechar = `\\ % use backslash in output files.
+ \indexdummies % don't expand commands in the output.
+ \normalturnoffactive % \ in index entries must not stay \, e.g., if
+ % the page break happens to be in the middle of an example.
+ \shipout\vbox{%
+ % Do this early so pdf references go to the beginning of the page.
+ \ifpdfmakepagedest \pdfmkdest{\the\pageno} \fi
+ %
+ \ifcropmarks \vbox to \outervsize\bgroup
+ \hsize = \outerhsize
+ \vskip-\topandbottommargin
+ \vtop to0pt{%
+ \line{\ewtop\hfil\ewtop}%
+ \nointerlineskip
+ \line{%
+ \vbox{\moveleft\cornerthick\nstop}%
+ \hfill
+ \vbox{\moveright\cornerthick\nstop}%
+ }%
+ \vss}%
+ \vskip\topandbottommargin
+ \line\bgroup
+ \hfil % center the page within the outer (page) hsize.
+ \ifodd\pageno\hskip\bindingoffset\fi
+ \vbox\bgroup
+ \fi
+ %
+ \unvbox\headlinebox
+ \pagebody{#1}%
+ \ifdim\ht\footlinebox > 0pt
+ % Only leave this space if the footline is nonempty.
+ % (We lessened \vsize for it in \oddfootingxxx.)
+ % The \baselineskip=24pt in plain's \makefootline has no effect.
+ \vskip 2\baselineskip
+ \unvbox\footlinebox
+ \fi
+ %
+ \ifcropmarks
+ \egroup % end of \vbox\bgroup
+ \hfil\egroup % end of (centering) \line\bgroup
+ \vskip\topandbottommargin plus1fill minus1fill
+ \boxmaxdepth = \cornerthick
+ \vbox to0pt{\vss
+ \line{%
+ \vbox{\moveleft\cornerthick\nsbot}%
+ \hfill
+ \vbox{\moveright\cornerthick\nsbot}%
+ }%
+ \nointerlineskip
+ \line{\ewbot\hfil\ewbot}%
+ }%
+ \egroup % \vbox from first cropmarks clause
+ \fi
+ }% end of \shipout\vbox
+ }% end of group with \normalturnoffactive
+ \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#1{%
+ \let\next = #1%
+ \begingroup
+ \obeylines
+ \futurelet\temp\parseargx
+}
+
+% If the next token is an obeyed space (from an @example environment or
+% the like), remove it and recurse. Otherwise, we're done.
+\def\parseargx{%
+ % \obeyedspace is defined far below, after the definition of \sepspaces.
+ \ifx\obeyedspace\temp
+ \expandafter\parseargdiscardspace
+ \else
+ \expandafter\parseargline
+ \fi
+}
+
+% Remove a single space (as the delimiter token to the macro call).
+{\obeyspaces %
+ \gdef\parseargdiscardspace {\futurelet\temp\parseargx}}
+
+{\obeylines %
+ \gdef\parseargline#1^^M{%
+ \endgroup % End of the group started in \parsearg.
+ %
+ % First remove any @c comment, then any @comment.
+ % Result of each macro is put in \toks0.
+ \argremovec #1\c\relax %
+ \expandafter\argremovecomment \the\toks0 \comment\relax %
+ %
+ % Call the caller's macro, saved as \next in \parsearg.
+ \expandafter\next\expandafter{\the\toks0}%
+ }%
+}
+
+% Since all \c{,omment} does is throw away the argument, we can let TeX
+% do that for us. The \relax here is matched by the \relax in the call
+% in \parseargline; it could be more or less anything, its purpose is
+% just to delimit the argument to the \c.
+\def\argremovec#1\c#2\relax{\toks0 = {#1}}
+\def\argremovecomment#1\comment#2\relax{\toks0 = {#1}}
+
+% \argremovec{,omment} might leave us with trailing spaces, though; e.g.,
+% @end itemize @c foo
+% will have two active spaces as part of the argument with the
+% `itemize'. Here we remove all active spaces from #1, and assign the
+% result to \toks0.
+%
+% This loses if there are any *other* active characters besides spaces
+% in the argument -- _ ^ +, for example -- since they get expanded.
+% Fortunately, Texinfo does not define any such commands. (If it ever
+% does, the catcode of the characters in questionwill have to be changed
+% here.) But this means we cannot call \removeactivespaces as part of
+% \argremovec{,omment}, since @c uses \parsearg, and thus the argument
+% that \parsearg gets might well have any character at all in it.
+%
+\def\removeactivespaces#1{%
+ \begingroup
+ \ignoreactivespaces
+ \edef\temp{#1}%
+ \global\toks0 = \expandafter{\temp}%
+ \endgroup
+}
+
+% Change the active space to expand to nothing.
+%
+\begingroup
+ \obeyspaces
+ \gdef\ignoreactivespaces{\obeyspaces\let =\empty}
+\endgroup
+
+
+\def\flushcr{\ifx\par\lisppar \def\next##1{}\else \let\next=\relax \fi \next}
+
+%% These are used to keep @begin/@end levels from running away
+%% Call \inENV within environments (after a \begingroup)
+\newif\ifENV \ENVfalse \def\inENV{\ifENV\relax\else\ENVtrue\fi}
+\def\ENVcheck{%
+\ifENV\errmessage{Still within an environment; press RETURN to continue}
+\endgroup\fi} % This is not perfect, but it should reduce lossage
+
+% @begin foo is the same as @foo, for now.
+\newhelp\EMsimple{Press RETURN to continue.}
+
+\outer\def\begin{\parsearg\beginxxx}
+
+\def\beginxxx #1{%
+\expandafter\ifx\csname #1\endcsname\relax
+{\errhelp=\EMsimple \errmessage{Undefined command @begin #1}}\else
+\csname #1\endcsname\fi}
+
+% @end foo executes the definition of \Efoo.
+%
+\def\end{\parsearg\endxxx}
+\def\endxxx #1{%
+ \removeactivespaces{#1}%
+ \edef\endthing{\the\toks0}%
+ %
+ \expandafter\ifx\csname E\endthing\endcsname\relax
+ \expandafter\ifx\csname \endthing\endcsname\relax
+ % There's no \foo, i.e., no ``environment'' foo.
+ \errhelp = \EMsimple
+ \errmessage{Undefined command `@end \endthing'}%
+ \else
+ \unmatchedenderror\endthing
+ \fi
+ \else
+ % Everything's ok; the right environment has been started.
+ \csname E\endthing\endcsname
+ \fi
+}
+
+% There is an environment #1, but it hasn't been started. Give an error.
+%
+\def\unmatchedenderror#1{%
+ \errhelp = \EMsimple
+ \errmessage{This `@end #1' doesn't have a matching `@#1'}%
+}
+
+% Define the control sequence \E#1 to give an unmatched @end error.
+%
+\def\defineunmatchedend#1{%
+ \expandafter\def\csname E#1\endcsname{\unmatchedenderror{#1}}%
+}
+
+
+%% Simple single-character @ commands
+
+% @@ prints an @
+% Kludge this until the fonts are right (grr).
+\def\@{{\tt\char64}}
+
+% This is turned off because it was never documented
+% and you can use @w{...} around a quote to suppress ligatures.
+%% Define @` and @' to be the same as ` and '
+%% but suppressing ligatures.
+%\def\`{{`}}
+%\def\'{{'}}
+
+% Used to generate quoted braces.
+\def\mylbrace {{\tt\char123}}
+\def\myrbrace {{\tt\char125}}
+\let\{=\mylbrace
+\let\}=\myrbrace
+\begingroup
+ % Definitions to produce \{ and \} commands for indices,
+ % and @{ and @} for the aux file.
+ \catcode`\{ = \other \catcode`\} = \other
+ \catcode`\[ = 1 \catcode`\] = 2
+ \catcode`\! = 0 \catcode`\\ = \other
+ !gdef!lbracecmd[\{]%
+ !gdef!rbracecmd[\}]%
+ !gdef!lbraceatcmd[@{]%
+ !gdef!rbraceatcmd[@}]%
+!endgroup
+
+% 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
+% Plain TeX defines: @AA @AE @O @OE @L (plus lowercase versions) @ss.
+\def\questiondown{?`}
+\def\exclamdown{!`}
+
+% 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
+}
+
+% Be sure we're in horizontal mode when doing a tie, since we make space
+% equivalent to this in @example-like environments. Otherwise, a space
+% at the beginning of a line will start with \penalty -- and
+% since \penalty is valid in vertical mode, we'd end up putting the
+% penalty on the vertical list instead of in the new paragraph.
+{\catcode`@ = 11
+ % Avoid using \@M directly, because that causes trouble
+ % if the definition is written into an index file.
+ \global\let\tiepenalty = \@M
+ \gdef\tie{\leavevmode\penalty\tiepenalty\ }
+}
+
+% @: forces normal size whitespace following.
+\def\:{\spacefactor=1000 }
+
+% @* forces a line break.
+\def\*{\hfil\break\hbox{}\ignorespaces}
+
+% @/ allows a line break.
+\let\/=\allowbreak
+
+% @. is an end-of-sentence period.
+\def\.{.\spacefactor=3000 }
+
+% @! is an end-of-sentence bang.
+\def\!{!\spacefactor=3000 }
+
+% @? is an end-of-sentence query.
+\def\?{?\spacefactor=3000 }
+
+% @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}
+%
+\def\group{\begingroup
+ \ifnum\catcode13=\active \else
+ \errhelp = \groupinvalidhelp
+ \errmessage{@group invalid in context where filling is enabled}%
+ \fi
+ %
+ % The \vtop we start below 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. (See p.82 of
+ % the TeXbook.) Thus, space below is not quite equal to space
+ % above. But it's pretty close.
+ \def\Egroup{%
+ \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
+ \copy\groupbox
+ \endgroup % End the \group.
+ }%
+ %
+ \setbox\groupbox = \vtop\bgroup
+ % We have to put a strut on the last line in case the @group is in
+ % the midst of an example, rather than completely enclosing it.
+ % Otherwise, the interline space between the last line of the group
+ % and the first line afterwards is too small. But we can't put the
+ % strut in \Egroup, since there it would be on a line by itself.
+ % Hence this just inserts a strut at the beginning of each line.
+ \everypar = {\strut}%
+ %
+ % Since we have a strut on every line, we don't need any of TeX's
+ % normal interline spacing.
+ \offinterlineskip
+ %
+ % OK, but now we have to do something about blank
+ % lines in the input in @example-like environments, which normally
+ % just turn into \lisppar, which will insert no space now that we've
+ % turned off the interline space. Simplest is to make them be an
+ % empty paragraph.
+ \ifx\par\lisppar
+ \edef\par{\leavevmode \par}%
+ %
+ % Reset ^^M's definition to new definition of \par.
+ \obeylines
+ \fi
+ %
+ % 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
+}
+%
+% 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
+
+\def\need{\parsearg\needx}
+
+% Old definition--didn't work.
+%\def\needx #1{\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
+%}}
+
+\def\needx#1{%
+ % 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
+
+\let\br = \par
+
+% @dots{} output an ellipsis using the current font.
+% We do .5em per period so that it has the same spacing in a typewriter
+% font as three actual period characters.
+%
+\def\dots{%
+ \leavevmode
+ \hbox to 1.5em{%
+ \hskip 0pt plus 0.25fil minus 0.25fil
+ .\hss.\hss.%
+ \hskip 0pt plus 0.5fil minus 0.5fil
+ }%
+}
+
+% @enddots{} is an end-of-sentence ellipsis.
+%
+\def\enddots{%
+ \leavevmode
+ \hbox to 2em{%
+ \hskip 0pt plus 0.25fil minus 0.25fil
+ .\hss.\hss.\hss.%
+ \hskip 0pt plus 0.5fil minus 0.5fil
+ }%
+ \spacefactor=3000
+}
+
+% @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.
+\def\exdent{\parsearg\exdentyyy}
+\def\exdentyyy #1{{\hfil\break\hbox{\kern -\exdentamount{\rm#1}}\hfil\break}}
+
+% This defn is used inside nofill environments such as @example.
+\def\nofillexdent{\parsearg\nofillexdentyyy}
+\def\nofillexdentyyy #1{{\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.
+% Allow normal characters that we make active in the argument (a file name).
+\def\include{\begingroup
+ \catcode`\\=\other
+ \catcode`~=\other
+ \catcode`^=\other
+ \catcode`_=\other
+ \catcode`|=\other
+ \catcode`<=\other
+ \catcode`>=\other
+ \catcode`+=\other
+ \parsearg\includezzz}
+% Restore active chars for included file.
+\def\includezzz#1{\endgroup\begingroup
+ % Read the included file in a group so nested @include's work.
+ \def\thisfile{#1}%
+ \let\value=\expandablevalue
+ \input\thisfile
+\endgroup}
+
+\def\thisfile{}
+
+% @center line
+% outputs that line, centered.
+%
+\def\center{\parsearg\docenter}
+\def\docenter#1{{%
+ \ifhmode \hfil\break \fi
+ \advance\hsize by -\leftskip
+ \advance\hsize by -\rightskip
+ \line{\hfil \ignorespaces#1\unskip \hfil}%
+ \ifhmode \break \fi
+}}
+
+% @sp n outputs n lines of vertical space
+
+\def\sp{\parsearg\spxxx}
+\def\spxxx #1{\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}
+%
+\def\paragraphindent{\parsearg\doparagraphindent}
+\def\doparagraphindent#1{%
+ \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.
+\def\exampleindent{\parsearg\doexampleindent}
+\def\doexampleindent#1{%
+ \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 indentat 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}
+\newdimen\currentparindent
+%
+\def\insertword{insert}
+%
+\def\firstparagraphindent{\parsearg\dofirstparagraphindent}
+\def\dofirstparagraphindent#1{%
+ \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{%
+ \global\let\indent=\ptexindent
+ \global\everypar = {}%
+ }%
+ \global\everypar = {%
+ \kern-\parindent
+ \global\let\indent=\ptexindent
+ \global\everypar = {}%
+ }%
+}%
+
+
+% @asis just yields its argument. Used with @table, for example.
+%
+\def\asis#1{#1}
+
+% @math outputs its argument in math mode.
+% We don't use $'s directly in the definition of \math because we need
+% to set catcodes according to plain TeX first, to allow for subscripts,
+% superscripts, special math chars, etc.
+%
+\let\implicitmath = $%$ font-lock fix
+%
+% One complication: _ usually means subscripts, but it could also mean
+% an actual _ character, as in @math{@var{some_variable} + 1}. So make
+% _ within @math be active (mathcode "8000), and distinguish by seeing
+% if the current family is \slfam, which is what @var uses.
+%
+{\catcode\underChar = \active
+\gdef\mathunderscore{%
+ \catcode\underChar=\active
+ \def_{\ifnum\fam=\slfam \_\else\sb\fi}%
+}}
+%
+% Another complication: we want \\ (and @\) to output a \ character.
+% FYI, plain.tex uses \\ as a temporary control sequence (why?), but
+% this is not advertised and we don't care. Texinfo does not
+% otherwise define @\.
+%
+% The \mathchar is class=0=ordinary, family=7=ttfam, position=5C=\.
+\def\mathbackslash{\ifnum\fam=\ttfam \mathchar"075C \else\backslash \fi}
+%
+\def\math{%
+ \tex
+ \mathcode`\_="8000 \mathunderscore
+ \let\\ = \mathbackslash
+ \mathactive
+ \implicitmath\finishmath}
+\def\finishmath#1{#1\implicitmath\Etex}
+
+% 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 set 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{\implicitmath\ptexbullet\implicitmath}
+\def\minus{\implicitmath-\implicitmath}
+
+% @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{%
+ \iflinks
+ \readauxfile
+ \fi % \openindices needs to do some work in any case.
+ \openindices
+ \fixbackslash % Turn off hack to swallow `\input texinfo'.
+ \global\let\setfilename=\comment % Ignore extra @setfilename cmds.
+ %
+ % If texinfo.cnf is present on the system, read it.
+ % Useful for site-wide @afourpaper, etc.
+ % Just to be on the safe side, close the input stream before the \input.
+ \openin 1 texinfo.cnf
+ \ifeof1 \let\temp=\relax \else \def\temp{\input texinfo.cnf }\fi
+ \closein1
+ \temp
+ %
+ \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
+
+\ifx\pdfoutput\undefined
+ \pdffalse
+ \let\pdfmkdest = \gobble
+ \let\pdfurl = \gobble
+ \let\endlink = \relax
+ \let\linkcolor = \relax
+ \let\pdfmakeoutlines = \relax
+\else
+ \pdftrue
+ \pdfoutput = 1
+ \input pdfcolor
+ \def\dopdfimage#1#2#3{%
+ \def\imagewidth{#2}%
+ \def\imageheight{#3}%
+ % without \immediate, pdftex seg faults when the same image is
+ % included twice. (Version 3.14159-pre-1.0-unofficial-20010704.)
+ \ifnum\pdftexversion < 14
+ \immediate\pdfimage
+ \else
+ \immediate\pdfximage
+ \fi
+ \ifx\empty\imagewidth\else width \imagewidth \fi
+ \ifx\empty\imageheight\else height \imageheight \fi
+ \ifnum\pdftexversion<13
+ #1.pdf%
+ \else
+ {#1.pdf}%
+ \fi
+ \ifnum\pdftexversion < 14 \else
+ \pdfrefximage \pdflastximage
+ \fi}
+ \def\pdfmkdest#1{{\normalturnoffactive \pdfdest name{#1} xyz}}
+ \def\pdfmkpgn#1{#1}
+ \let\linkcolor = \Blue % was Cyan, but that seems light?
+ \def\endlink{\Black\pdfendlink}
+ % Adding outlines to PDF; macros for calculating structure of outlines
+ % come from Petr Olsak
+ \def\expnumber#1{\expandafter\ifx\csname#1\endcsname\relax 0%
+ \else \csname#1\endcsname \fi}
+ \def\advancenumber#1{\tempnum=\expnumber{#1}\relax
+ \advance\tempnum by1
+ \expandafter\xdef\csname#1\endcsname{\the\tempnum}}
+ \def\pdfmakeoutlines{{%
+ \openin 1 \jobname.toc
+ \ifeof 1\else\begingroup
+ \closein 1
+ % Thanh's hack / proper braces in bookmarks
+ \edef\mylbrace{\iftrue \string{\else}\fi}\let\{=\mylbrace
+ \edef\myrbrace{\iffalse{\else\string}\fi}\let\}=\myrbrace
+ %
+ \def\chapentry ##1##2##3{}
+ \def\secentry ##1##2##3##4{\advancenumber{chap##2}}
+ \def\subsecentry ##1##2##3##4##5{\advancenumber{sec##2.##3}}
+ \def\subsubsecentry ##1##2##3##4##5##6{\advancenumber{subsec##2.##3.##4}}
+ \let\appendixentry = \chapentry
+ \let\unnumbchapentry = \chapentry
+ \let\unnumbsecentry = \secentry
+ \let\unnumbsubsecentry = \subsecentry
+ \let\unnumbsubsubsecentry = \subsubsecentry
+ \input \jobname.toc
+ \def\chapentry ##1##2##3{%
+ \pdfoutline goto name{\pdfmkpgn{##3}}count-\expnumber{chap##2}{##1}}
+ \def\secentry ##1##2##3##4{%
+ \pdfoutline goto name{\pdfmkpgn{##4}}count-\expnumber{sec##2.##3}{##1}}
+ \def\subsecentry ##1##2##3##4##5{%
+ \pdfoutline goto name{\pdfmkpgn{##5}}count-\expnumber{subsec##2.##3.##4}{##1}}
+ \def\subsubsecentry ##1##2##3##4##5##6{%
+ \pdfoutline goto name{\pdfmkpgn{##6}}{##1}}
+ \let\appendixentry = \chapentry
+ \let\unnumbchapentry = \chapentry
+ \let\unnumbsecentry = \secentry
+ \let\unnumbsubsecentry = \subsecentry
+ \let\unnumbsubsubsecentry = \subsubsecentry
+ %
+ % Make special characters normal for writing to the pdf file.
+ %
+ \indexnofonts
+ \let\tt=\relax
+ \turnoffactive
+ \input \jobname.toc
+ \endgroup\fi
+ }}
+ \def\makelinks #1,{%
+ \def\params{#1}\def\E{END}%
+ \ifx\params\E
+ \let\nextmakelinks=\relax
+ \else
+ \let\nextmakelinks=\makelinks
+ \ifnum\lnkcount>0,\fi
+ \picknum{#1}%
+ \startlink attr{/Border [0 0 0]}
+ goto name{\pdfmkpgn{\the\pgn}}%
+ \linkcolor #1%
+ \advance\lnkcount by 1%
+ \endlink
+ \fi
+ \nextmakelinks
+ }
+ \def\picknum#1{\expandafter\pn#1}
+ \def\pn#1{%
+ \def\p{#1}%
+ \ifx\p\lbrace
+ \let\nextpn=\ppn
+ \else
+ \let\nextpn=\ppnn
+ \def\first{#1}
+ \fi
+ \nextpn
+ }
+ \def\ppn#1{\pgn=#1\gobble}
+ \def\ppnn{\pgn=\first}
+ \def\pdfmklnk#1{\lnkcount=0\makelinks #1,END,}
+ \def\addtokens#1#2{\edef\addtoks{\noexpand#1={\the#1#2}}\addtoks}
+ \def\skipspaces#1{\def\PP{#1}\def\D{|}%
+ \ifx\PP\D\let\nextsp\relax
+ \else\let\nextsp\skipspaces
+ \ifx\p\space\else\addtokens{\filename}{\PP}%
+ \advance\filenamelength by 1
+ \fi
+ \fi
+ \nextsp}
+ \def\getfilename#1{\filenamelength=0\expandafter\skipspaces#1|\relax}
+ \ifnum\pdftexversion < 14
+ \let \startlink \pdfannotlink
+ \else
+ \let \startlink \pdfstartlink
+ \fi
+ \def\pdfurl#1{%
+ \begingroup
+ \normalturnoffactive\def\@{@}%
+ \let\value=\expandablevalue
+ \leavevmode\Red
+ \startlink attr{/Border [0 0 0]}%
+ user{/Subtype /Link /A << /S /URI /URI (#1) >>}%
+ % #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|
+ \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}
+\fi % \ifx\pdfoutput
+
+
+\message{fonts,}
+% Font-change commands.
+
+% Texinfo sort of supports the sans serif font style, which plain TeX does not.
+% So we set up a \sf analogous to plain's \rm, etc.
+\newfam\sffam
+\def\sf{\fam=\sffam \tensf}
+\let\li = \sf % Sometimes we call it \li, not \sf.
+
+% We don't need math for this one.
+\def\ttsl{\tenttsl}
+
+% Default leading.
+\newdimen\textleading \textleading = 13.2pt
+
+% Set the baselineskip to #1, and the lineskip and strut size
+% correspondingly. There is no deep meaning behind these magic numbers
+% used as factors; they just match (closely enough) what Knuth defined.
+%
+\def\lineskipfactor{.08333}
+\def\strutheightpercent{.70833}
+\def\strutdepthpercent {.29167}
+%
+\def\setleading#1{%
+ \normalbaselineskip = #1\relax
+ \normallineskip = \lineskipfactor\normalbaselineskip
+ \normalbaselines
+ \setbox\strutbox =\hbox{%
+ \vrule width0pt height\strutheightpercent\baselineskip
+ depth \strutdepthpercent \baselineskip
+ }%
+}
+
+% Set the font macro #1 to the font named #2, adding on the
+% specified font prefix (normally `cm').
+% #3 is the font's design size, #4 is a scale factor
+\def\setfont#1#2#3#4{\font#1=\fontprefix#2#3 scaled #4}
+
+% Use cm as the default font prefix.
+% To specify the font prefix, you must define \fontprefix
+% before you read in texinfo.tex.
+\ifx\fontprefix\undefined
+\def\fontprefix{cm}
+\fi
+% Support font families that don't use the same naming scheme as CM.
+\def\rmshape{r}
+\def\rmbshape{bx} %where the normal face is bold
+\def\bfshape{b}
+\def\bxshape{bx}
+\def\ttshape{tt}
+\def\ttbshape{tt}
+\def\ttslshape{sltt}
+\def\itshape{ti}
+\def\itbshape{bxti}
+\def\slshape{sl}
+\def\slbshape{bxsl}
+\def\sfshape{ss}
+\def\sfbshape{ss}
+\def\scshape{csc}
+\def\scbshape{csc}
+
+\newcount\mainmagstep
+\ifx\bigger\relax
+ % not really supported.
+ \mainmagstep=\magstep1
+ \setfont\textrm\rmshape{12}{1000}
+ \setfont\texttt\ttshape{12}{1000}
+\else
+ \mainmagstep=\magstephalf
+ \setfont\textrm\rmshape{10}{\mainmagstep}
+ \setfont\texttt\ttshape{10}{\mainmagstep}
+\fi
+% Instead of cmb10, you may want to use cmbx10.
+% cmbx10 is a prettier font on its own, but cmb10
+% looks better when embedded in a line with cmr10
+% (in Bob's opinion).
+\setfont\textbf\bfshape{10}{\mainmagstep}
+\setfont\textit\itshape{10}{\mainmagstep}
+\setfont\textsl\slshape{10}{\mainmagstep}
+\setfont\textsf\sfshape{10}{\mainmagstep}
+\setfont\textsc\scshape{10}{\mainmagstep}
+\setfont\textttsl\ttslshape{10}{\mainmagstep}
+\font\texti=cmmi10 scaled \mainmagstep
+\font\textsy=cmsy10 scaled \mainmagstep
+
+% A few fonts for @defun, etc.
+\setfont\defbf\bxshape{10}{\magstep1} %was 1314
+\setfont\deftt\ttshape{10}{\magstep1}
+\def\df{\let\tentt=\deftt \let\tenbf = \defbf \bf}
+
+% Fonts for indices, footnotes, small examples (9pt).
+\setfont\smallrm\rmshape{9}{1000}
+\setfont\smalltt\ttshape{9}{1000}
+\setfont\smallbf\bfshape{10}{900}
+\setfont\smallit\itshape{9}{1000}
+\setfont\smallsl\slshape{9}{1000}
+\setfont\smallsf\sfshape{9}{1000}
+\setfont\smallsc\scshape{10}{900}
+\setfont\smallttsl\ttslshape{10}{900}
+\font\smalli=cmmi9
+\font\smallsy=cmsy9
+
+% Fonts for small examples (8pt).
+\setfont\smallerrm\rmshape{8}{1000}
+\setfont\smallertt\ttshape{8}{1000}
+\setfont\smallerbf\bfshape{10}{800}
+\setfont\smallerit\itshape{8}{1000}
+\setfont\smallersl\slshape{8}{1000}
+\setfont\smallersf\sfshape{8}{1000}
+\setfont\smallersc\scshape{10}{800}
+\setfont\smallerttsl\ttslshape{10}{800}
+\font\smalleri=cmmi8
+\font\smallersy=cmsy8
+
+% Fonts for title page:
+\setfont\titlerm\rmbshape{12}{\magstep3}
+\setfont\titleit\itbshape{10}{\magstep4}
+\setfont\titlesl\slbshape{10}{\magstep4}
+\setfont\titlett\ttbshape{12}{\magstep3}
+\setfont\titlettsl\ttslshape{10}{\magstep4}
+\setfont\titlesf\sfbshape{17}{\magstep1}
+\let\titlebf=\titlerm
+\setfont\titlesc\scbshape{10}{\magstep4}
+\font\titlei=cmmi12 scaled \magstep3
+\font\titlesy=cmsy10 scaled \magstep4
+\def\authorrm{\secrm}
+\def\authortt{\sectt}
+
+% Chapter (and unnumbered) fonts (17.28pt).
+\setfont\chaprm\rmbshape{12}{\magstep2}
+\setfont\chapit\itbshape{10}{\magstep3}
+\setfont\chapsl\slbshape{10}{\magstep3}
+\setfont\chaptt\ttbshape{12}{\magstep2}
+\setfont\chapttsl\ttslshape{10}{\magstep3}
+\setfont\chapsf\sfbshape{17}{1000}
+\let\chapbf=\chaprm
+\setfont\chapsc\scbshape{10}{\magstep3}
+\font\chapi=cmmi12 scaled \magstep2
+\font\chapsy=cmsy10 scaled \magstep3
+
+% Section fonts (14.4pt).
+\setfont\secrm\rmbshape{12}{\magstep1}
+\setfont\secit\itbshape{10}{\magstep2}
+\setfont\secsl\slbshape{10}{\magstep2}
+\setfont\sectt\ttbshape{12}{\magstep1}
+\setfont\secttsl\ttslshape{10}{\magstep2}
+\setfont\secsf\sfbshape{12}{\magstep1}
+\let\secbf\secrm
+\setfont\secsc\scbshape{10}{\magstep2}
+\font\seci=cmmi12 scaled \magstep1
+\font\secsy=cmsy10 scaled \magstep2
+
+% Subsection fonts (13.15pt).
+\setfont\ssecrm\rmbshape{12}{\magstephalf}
+\setfont\ssecit\itbshape{10}{1315}
+\setfont\ssecsl\slbshape{10}{1315}
+\setfont\ssectt\ttbshape{12}{\magstephalf}
+\setfont\ssecttsl\ttslshape{10}{1315}
+\setfont\ssecsf\sfbshape{12}{\magstephalf}
+\let\ssecbf\ssecrm
+\setfont\ssecsc\scbshape{10}{\magstep1}
+\font\sseci=cmmi12 scaled \magstephalf
+\font\ssecsy=cmsy10 scaled 1315
+% The smallcaps and symbol fonts should actually be scaled \magstep1.5,
+% but that is not a standard magnification.
+
+% 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 so that font changes will continue to work
+% in math mode, where it is the current \fam that is relevant in most
+% cases, not the current font. Plain TeX does \def\bf{\fam=\bffam
+% \tenbf}, for example. By redefining \tenbf, we obviate the need to
+% redefine \bf itself.
+\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
+ \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
+ \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
+ \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
+ \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
+ \resetmathfonts \setleading{15pt}}
+\let\subsubsecfonts = \subsecfonts % Maybe make sssec fonts scaled magstephalf?
+\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
+ \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
+ \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 \smallerfonts (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 we used A4 paper on this side of the Atlantic.
+%
+% --karl, 24jan03.
+
+
+% Set up the default fonts, so we can use them for creating boxes.
+%
+\textfonts
+
+% Define these so they can be easily changed for other fonts.
+\def\angleleft{$\langle$}
+\def\angleright{$\rangle$}
+
+% Count depth in font-changes, for error checks
+\newcount\fontdepth \fontdepth=0
+
+% Fonts for short table of contents.
+\setfont\shortcontrm\rmshape{12}{1000}
+\setfont\shortcontbf\bxshape{12}{1000}
+\setfont\shortcontsl\slshape{12}{1000}
+\setfont\shortconttt\ttshape{12}{1000}
+
+%% Add scribe-like font environments, plus @l for inline lisp (usually sans
+%% serif) and @ii for TeX italic
+
+% \smartitalic{ARG} outputs arg in italics, followed by an italic correction
+% unless the following character is such as not to need one.
+\def\smartitalicx{\ifx\next,\else\ifx\next-\else\ifx\next.\else
+ \ptexslash\fi\fi\fi}
+\def\smartslanted#1{{\ifusingtt\ttsl\sl #1}\futurelet\next\smartitalicx}
+\def\smartitalic#1{{\ifusingtt\ttsl\it #1}\futurelet\next\smartitalicx}
+
+\let\i=\smartitalic
+\let\var=\smartslanted
+\let\dfn=\smartslanted
+\let\emph=\smartitalic
+\let\cite=\smartslanted
+
+\def\b#1{{\bf #1}}
+\let\strong=\b
+
+% We can't just use \exhyphenpenalty, because that only has effect at
+% the end of a paragraph. Restore normal hyphenation at the end of the
+% group within which \nohyphenation is presumably called.
+%
+\def\nohyphenation{\hyphenchar\font = -1 \aftergroup\restorehyphenation}
+\def\restorehyphenation{\hyphenchar\font = `- }
+
+% Set sfcode to normal for the chars that usually have another value.
+% Can't use plain's \frenchspacing because it uses the `\x notation, and
+% sometimes \x has an active definition that messes things up.
+%
+\catcode`@=11
+ \def\frenchspacing{%
+ \sfcode\dotChar =\@m \sfcode\questChar=\@m \sfcode\exclamChar=\@m
+ \sfcode\colonChar=\@m \sfcode\semiChar =\@m \sfcode\commaChar =\@m
+ }
+\catcode`@=\other
+
+\def\t#1{%
+ {\tt \rawbackslash \frenchspacing #1}%
+ \null
+}
+\let\ttfont=\t
+\def\samp#1{`\tclose{#1}'\null}
+\setfont\keyrm\rmshape{8}{1000}
+\font\keysy=cmsy9
+\def\key#1{{\keyrm\textfont2=\keysy \leavevmode\hbox{%
+ \raise0.4pt\hbox{\angleleft}\kern-.08em\vtop{%
+ \vbox{\hrule\kern-0.4pt
+ \hbox{\raise0.4pt\hbox{\vphantom{\angleleft}}#1}}%
+ \kern-0.4pt\hrule}%
+ \kern-.06em\raise0.4pt\hbox{\angleright}}}}
+% The old definition, with no lozenge:
+%\def\key #1{{\ttsl \nohyphenation \uppercase{#1}}\null}
+\def\ctrl #1{{\tt \rawbackslash \hat}#1}
+
+% @file, @option are the same as @samp.
+\let\file=\samp
+\let\option=\samp
+
+% @code is a modification of @t,
+% which makes spaces the same size as normal in the surrounding text.
+\def\tclose#1{%
+ {%
+ % Change normal interword space to be same as for the current font.
+ \spaceskip = \fontdimen2\font
+ %
+ % Switch to typewriter.
+ \tt
+ %
+ % But `\ ' produces the large typewriter interword space.
+ \def\ {{\spaceskip = 0pt{} }}%
+ %
+ % Turn off hyphenation.
+ \nohyphenation
+ %
+ \rawbackslash
+ \frenchspacing
+ #1%
+ }%
+ \null
+}
+
+% We *must* turn on hyphenation at `-' and `_' in \code.
+% Otherwise, it is too hard to avoid overfull hboxes
+% in the Emacs manual, the Library manual, etc.
+
+% Unfortunately, TeX uses one parameter (\hyphenchar) to control
+% both hyphenation at - and hyphenation within words.
+% We must therefore turn them both off (\tclose does that)
+% and arrange explicitly to hyphenate at a dash.
+% -- rms.
+{
+ \catcode`\-=\active
+ \catcode`\_=\active
+ %
+ \global\def\code{\begingroup
+ \catcode`\-=\active \let-\codedash
+ \catcode`\_=\active \let_\codeunder
+ \codex
+ }
+ %
+ % If we end up with any active - characters when handling the index,
+ % just treat them as a normal -.
+ \global\def\indexbreaks{\catcode`\-=\active \let-\realdash}
+}
+
+\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}
+
+% @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).
+\def\kbdinputstyle{\parsearg\kbdinputstylexxx}
+\def\kbdinputstylexxx#1{%
+ \def\arg{#1}%
+ \ifx\arg\worddistinct
+ \gdef\kbdexamplefont{\ttsl}\gdef\kbdfont{\ttsl}%
+ \else\ifx\arg\wordexample
+ \gdef\kbdexamplefont{\ttsl}\gdef\kbdfont{\tt}%
+ \else\ifx\arg\wordcode
+ \gdef\kbdexamplefont{\tt}\gdef\kbdfont{\tt}%
+ \else
+ \errhelp = \EMsimple
+ \errmessage{Unknown @kbdinputstyle option `\arg'}%
+ \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 @url, @env, @command quotes seem unnecessary, so use \code.
+\let\url=\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}
+
+% 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 downcases the argument and prints in smallcaps.
+\def\acronym#1{{\smallcaps \lowercase{#1}}}
+
+% @pounds{} is a sterling sign.
+\def\pounds{{\it\$}}
+
+% @registeredsymbol - R in a circle. For now, only works in text size;
+% we'd have to redo the font mechanism to change the \scriptstyle and
+% \scriptscriptstyle font sizes to make it look right in headings.
+% Adapted from the plain.tex definition of \copyright.
+%
+\def\registeredsymbol{%
+ $^{{\ooalign{\hfil\raise.07ex\hbox{$\scriptstyle\rm R$}\hfil\crcr\Orb}}%
+ }$%
+}
+
+
+\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
+
+\def\shorttitlepage{\parsearg\shorttitlepagezzz}
+\def\shorttitlepagezzz #1{\begingroup\hbox{}\vskip 1.5in \chaprm \centerline{#1}%
+ \endgroup\page\hbox{}\page}
+
+\def\titlepage{\begingroup \parindent=0pt \textfonts
+ \let\subtitlerm=\tenrm
+ \def\subtitlefont{\subtitlerm \normalbaselineskip = 13pt \normalbaselines}%
+ %
+ \def\authorfont{\authorrm \normalbaselineskip = 16pt \normalbaselines
+ \let\tt=\authortt}%
+ %
+ % Leave some space at the very top of the page.
+ \vglue\titlepagetopglue
+ %
+ % Now you can print the title using @title.
+ \def\title{\parsearg\titlezzz}%
+ \def\titlezzz##1{\leftline{\titlefonts\rm ##1}
+ % print a rule at the page bottom also.
+ \finishedtitlepagefalse
+ \vskip4pt \hrule height 4pt width \hsize \vskip4pt}%
+ % No rule at page bottom unless we print one at the top with @title.
+ \finishedtitlepagetrue
+ %
+ % Now you can put text using @subtitle.
+ \def\subtitle{\parsearg\subtitlezzz}%
+ \def\subtitlezzz##1{{\subtitlefont \rightline{##1}}}%
+ %
+ % @author should come last, but may come many times.
+ \def\author{\parsearg\authorzzz}%
+ \def\authorzzz##1{\ifseenauthor\else\vskip 0pt plus 1filll\seenauthortrue\fi
+ {\authorfont \leftline{##1}}}%
+ %
+ % 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
+ \oldpage
+ \let\page = \oldpage
+ \hbox{}}%
+% \def\page{\oldpage \hbox{}}
+}
+
+\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
+}
+
+%%% 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\oddheading{\parsearg\oddheadingxxx}
+\def\everyheading{\parsearg\everyheadingxxx}
+
+\def\evenfooting{\parsearg\evenfootingxxx}
+\def\oddfooting{\parsearg\oddfootingxxx}
+\def\everyfooting{\parsearg\everyfootingxxx}
+
+{\catcode`\@=0 %
+
+\gdef\evenheadingxxx #1{\evenheadingyyy #1@|@|@|@|\finish}
+\gdef\evenheadingyyy #1@|#2@|#3@|#4\finish{%
+\global\evenheadline={\rlap{\centerline{#2}}\line{#1\hfil#3}}}
+
+\gdef\oddheadingxxx #1{\oddheadingyyy #1@|@|@|@|\finish}
+\gdef\oddheadingyyy #1@|#2@|#3@|#4\finish{%
+\global\oddheadline={\rlap{\centerline{#2}}\line{#1\hfil#3}}}
+
+\gdef\everyheadingxxx#1{\oddheadingxxx{#1}\evenheadingxxx{#1}}%
+
+\gdef\evenfootingxxx #1{\evenfootingyyy #1@|@|@|@|\finish}
+\gdef\evenfootingyyy #1@|#2@|#3@|#4\finish{%
+\global\evenfootline={\rlap{\centerline{#2}}\line{#1\hfil#3}}}
+
+\gdef\oddfootingxxx #1{\oddfootingyyy #1@|@|@|@|\finish}
+\gdef\oddfootingyyy #1@|#2@|#3@|#4\finish{%
+ \global\oddfootline = {\rlap{\centerline{#2}}\line{#1\hfil#3}}%
+ %
+ % Leave some space for the footline. Hopefully ok to assume
+ % @evenfooting will not be used by itself.
+ \global\advance\pageheight by -\baselineskip
+ \global\advance\vsize by -\baselineskip
+}
+
+\gdef\everyfootingxxx#1{\oddfootingxxx{#1}\evenfootingxxx{#1}}
+%
+}% unbind the catcode of @.
+
+% @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\settitlezzz}
+\def\settitlezzz #1{\gdef\thistitle{#1}}
+
+
+\message{tables,}
+% Tables -- @table, @ftable, @vtable, @item(x), @kitem(x), @xitem(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, @vtable, 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\internalBxitem "#1"{\def\xitemsubtopix{#1} \smallbreak \parsearg\xitemzzz}
+\def\internalBxitemx "#1"{\def\xitemsubtopix{#1} \itemxpar \parsearg\xitemzzz}
+
+\def\internalBkitem{\smallbreak \parsearg\kitemzzz}
+\def\internalBkitemx{\itemxpar \parsearg\kitemzzz}
+
+\def\kitemzzz #1{\dosubind {kw}{\code{#1}}{for {\bf \lastfunction}}%
+ \itemzzz {#1}}
+
+\def\xitemzzz #1{\dosubind {kw}{\code{#1}}{for {\bf \xitemsubtopic}}%
+ \itemzzz {#1}}
+
+\def\itemzzz #1{\begingroup %
+ \advance\hsize by -\rightskip
+ \advance\hsize by -\tableindent
+ \setbox0=\hbox{\itemfont{#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. (Unfortunately
+ % we can't prevent a possible page break at the following
+ % \baselineskip glue.) However, if what follows is an environment
+ % such as @example, there will be no \parskip glue; then
+ % the negative vskip we just 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.
+ % (Possibly there are other commands that could be followed by
+ % @example which need the same treatment, but not section titles; or
+ % maybe section titles are the only special case and they should be
+ % penalty 10001...)
+ \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 table}}
+\def\itemx{\errmessage{@itemx while not in a table}}
+\def\kitem{\errmessage{@kitem while not in a table}}
+\def\kitemx{\errmessage{@kitemx while not in a table}}
+\def\xitem{\errmessage{@xitem while not in a table}}
+\def\xitemx{\errmessage{@xitemx while not in a table}}
+
+% Contains a kludge to get @end[description] to work.
+\def\description{\tablez{\dontindex}{1}{}{}{}{}}
+
+% @table, @ftable, @vtable.
+\def\table{\begingroup\inENV\obeylines\obeyspaces\tablex}
+{\obeylines\obeyspaces%
+\gdef\tablex #1^^M{%
+\tabley\dontindex#1 \endtabley}}
+
+\def\ftable{\begingroup\inENV\obeylines\obeyspaces\ftablex}
+{\obeylines\obeyspaces%
+\gdef\ftablex #1^^M{%
+\tabley\fnitemindex#1 \endtabley
+\def\Eftable{\endgraf\afterenvbreak\endgroup}%
+\let\Etable=\relax}}
+
+\def\vtable{\begingroup\inENV\obeylines\obeyspaces\vtablex}
+{\obeylines\obeyspaces%
+\gdef\vtablex #1^^M{%
+\tabley\vritemindex#1 \endtabley
+\def\Evtable{\endgraf\afterenvbreak\endgroup}%
+\let\Etable=\relax}}
+
+\def\dontindex #1{}
+\def\fnitemindex #1{\doind {fn}{\code{#1}}}%
+\def\vritemindex #1{\doind {vr}{\code{#1}}}%
+
+{\obeyspaces %
+\gdef\tabley#1#2 #3 #4 #5 #6 #7\endtabley{\endgroup%
+\tablez{#1}{#2}{#3}{#4}{#5}{#6}}}
+
+\def\tablez #1#2#3#4#5#6{%
+\aboveenvbreak %
+\begingroup %
+\def\Edescription{\Etable}% Necessary kludge.
+\let\itemindex=#1%
+\ifnum 0#3>0 \advance \leftskip by #3\mil \fi %
+\ifnum 0#4>0 \tableindent=#4\mil \fi %
+\ifnum 0#5>0 \advance \rightskip by #5\mil \fi %
+\def\itemfont{#2}%
+\itemmax=\tableindent %
+\advance \itemmax by -\itemmargin %
+\advance \leftskip by \tableindent %
+\exdentamount=\tableindent
+\parindent = 0pt
+\parskip = \smallskipamount
+\ifdim \parskip=0pt \parskip=2pt \fi%
+\def\Etable{\endgraf\afterenvbreak\endgroup}%
+\let\item = \internalBitem %
+\let\itemx = \internalBitemx %
+\let\kitem = \internalBkitem %
+\let\kitemx = \internalBkitemx %
+\let\xitem = \internalBxitem %
+\let\xitemx = \internalBxitemx %
+}
+
+% This is the counter used by @enumerate, which is really @itemize
+
+\newcount \itemno
+
+\def\itemize{\parsearg\itemizezzz}
+
+\def\itemizezzz #1{%
+ \begingroup % ended by the @end itemize
+ \itemizey {#1}{\Eitemize}
+}
+
+\def\itemizey#1#2{%
+ \aboveenvbreak
+ \itemmax=\itemindent
+ \advance\itemmax by -\itemmargin
+ \advance\leftskip by \itemindent
+ \exdentamount=\itemindent
+ \parindent=0pt
+ \parskip=\smallskipamount
+ \ifdim\parskip=0pt \parskip=2pt \fi
+ \def#2{\endgraf\afterenvbreak\endgroup}%
+ \def\itemcontents{#1}%
+ % @itemize with no arg is equivalent to @itemize @bullet.
+ \ifx\itemcontents\empty\def\itemcontents{\bullet}\fi
+ \let\item=\itemizeitem
+}
+
+% \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'.
+%
+\def\enumerate{\parsearg\enumeratezzz}
+\def\enumeratezzz #1{\enumeratey #1 \endenumeratey}
+\def\enumeratey #1 #2\endenumeratey{%
+ \begingroup % ended by the @end enumerate
+ %
+ % 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 itemizey, 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
+ \itemizey{#1.}\Eenumerate\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}
+
+% Definition of @item while inside @itemize.
+
+\def\itemizeitem{%
+\advance\itemno by 1
+{\let\par=\endgraf \smallbreak}%
+\ifhmode \errmessage{In hmode at itemizeitem}\fi
+{\parskip=0in \hskip 0pt
+\hbox to 0pt{\hss \itemcontents\hskip \itemmargin}%
+\vadjust{\penalty 1200}}%
+\flushcr}
+
+% @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.
+%
+% For those who want to use more than one line's worth of words in
+% the preamble, break the line within one argument and it
+% will parse correctly, i.e.,
+%
+% @multitable {Column 1 template} {Column 2 template} {Column 3
+% template}
+% Not:
+% @multitable {Column 1 template} {Column 2 template}
+% {Column 3 template}
+
+% 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, @multitable or @end multitable 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 part of the @columnfraction before the decimal point, which
+% is presumably either 0 or the empty string (but we don't check, we
+% just throw it away). #2 is the decimal part, which we use as the
+% percent of \hsize for this column.
+\def\pickupwholefraction#1.#2 {%
+ \global\advance\colcount by 1
+ \expandafter\xdef\csname col\the\colcount\endcsname{.#2\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 ... @end multitable definitions:
+%
+\def\multitable{\parsearg\dotable}
+\def\dotable#1{\bgroup
+ \vskip\parskip
+ \let\item=\crcrwithfootnotes
+ % 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.
+ \let\tab=&%
+ \let\startfootins=\startsavedfootnote
+ \tolerance=9500
+ \hbadness=9500
+ \setmultitablespacing
+ \parskip=\multitableparskip
+ \parindent=\multitableparindent
+ \overfullrule=0pt
+ \global\colcount=0
+ \def\Emultitable{%
+ \global\setpercentfalse
+ \crcrwithfootnotes\crcr
+ \egroup\egroup
+ }%
+ %
+ % To parse everything between @multitable and @item:
+ \setuptable#1 \endsetuptable
+ %
+ % \everycr will reset column counter, \colcount, at the end of
+ % each line. Every column entry will cause \colcount to advance by one.
+ % The table preamble
+ % looks at the current \colcount to find the correct column width.
+ \everycr{\noalign{%
+ %
+ % \filbreak%% keeps underfull box messages off when table breaks over pages.
+ % 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.
+ \global\colcount=0\relax}}%
+ %
+ % 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\relax
+ \multistrut\vtop{\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\setmultitablespacing{% test to see if user has set \multitablelinespace.
+% If so, do nothing. If not, give it an appropriate dimension based on
+% current baselineskip.
+\ifdim\multitablelinespace=0pt
+\setbox0=\vbox{X}\global\multitablelinespace=\the\baselineskip
+\global\advance\multitablelinespace by-\ht0
+%% strut to put in table in case some entry doesn't have descenders,
+%% to keep lines equally spaced
+\let\multistrut = \strut
+\else
+%% FIXME: what is \box0 supposed to be?
+\gdef\multistrut{\vrule height\multitablelinespace depth\dp0
+width0pt\relax} \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}
+
+% In case 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. Otherwise, the insertion is lost, it never migrates to the
+% main vertical list. --kasal, 22jan03.
+%
+\newbox\savedfootnotes
+%
+% \dotable \let's \startfootins to this, so that \dofootnote will call
+% it instead of starting the insertion right away.
+\def\startsavedfootnote{%
+ \global\setbox\savedfootnotes = \vbox\bgroup
+ \unvbox\savedfootnotes
+}
+\def\crcrwithfootnotes{%
+ \crcr
+ \ifvoid\savedfootnotes \else
+ \noalign{\insert\footins{\box\savedfootnotes}}%
+ \fi
+}
+
+\message{conditionals,}
+% Prevent errors for section commands.
+% Used in @ignore and in failing conditionals.
+\def\ignoresections{%
+ \let\chapter=\relax
+ \let\unnumbered=\relax
+ \let\top=\relax
+ \let\unnumberedsec=\relax
+ \let\unnumberedsection=\relax
+ \let\unnumberedsubsec=\relax
+ \let\unnumberedsubsection=\relax
+ \let\unnumberedsubsubsec=\relax
+ \let\unnumberedsubsubsection=\relax
+ \let\section=\relax
+ \let\subsec=\relax
+ \let\subsubsec=\relax
+ \let\subsection=\relax
+ \let\subsubsection=\relax
+ \let\appendix=\relax
+ \let\appendixsec=\relax
+ \let\appendixsection=\relax
+ \let\appendixsubsec=\relax
+ \let\appendixsubsection=\relax
+ \let\appendixsubsubsec=\relax
+ \let\appendixsubsubsection=\relax
+ \let\contents=\relax
+ \let\smallbook=\relax
+ \let\titlepage=\relax
+}
+
+% Used in nested conditionals, where we have to parse the Texinfo source
+% and so want to turn off most commands, in case they are used
+% incorrectly.
+%
+% We use \empty instead of \relax for the @def... commands, so that \end
+% doesn't throw an error. For instance:
+% @ignore
+% @deffn ...
+% @end deffn
+% @end ignore
+%
+% The @end deffn is going to get expanded, because we're trying to allow
+% nested conditionals. But we don't want to expand the actual @deffn,
+% since it might be syntactically correct and intended to be ignored.
+% Since \end checks for \relax, using \empty does not cause an error.
+%
+\def\ignoremorecommands{%
+ \let\defcodeindex = \relax
+ \let\defcv = \empty
+ \let\defcvx = \empty
+ \let\Edefcv = \empty
+ \let\deffn = \empty
+ \let\deffnx = \empty
+ \let\Edeffn = \empty
+ \let\defindex = \relax
+ \let\defivar = \empty
+ \let\defivarx = \empty
+ \let\Edefivar = \empty
+ \let\defmac = \empty
+ \let\defmacx = \empty
+ \let\Edefmac = \empty
+ \let\defmethod = \empty
+ \let\defmethodx = \empty
+ \let\Edefmethod = \empty
+ \let\defop = \empty
+ \let\defopx = \empty
+ \let\Edefop = \empty
+ \let\defopt = \empty
+ \let\defoptx = \empty
+ \let\Edefopt = \empty
+ \let\defspec = \empty
+ \let\defspecx = \empty
+ \let\Edefspec = \empty
+ \let\deftp = \empty
+ \let\deftpx = \empty
+ \let\Edeftp = \empty
+ \let\deftypefn = \empty
+ \let\deftypefnx = \empty
+ \let\Edeftypefn = \empty
+ \let\deftypefun = \empty
+ \let\deftypefunx = \empty
+ \let\Edeftypefun = \empty
+ \let\deftypeivar = \empty
+ \let\deftypeivarx = \empty
+ \let\Edeftypeivar = \empty
+ \let\deftypemethod = \empty
+ \let\deftypemethodx = \empty
+ \let\Edeftypemethod = \empty
+ \let\deftypeop = \empty
+ \let\deftypeopx = \empty
+ \let\Edeftypeop = \empty
+ \let\deftypevar = \empty
+ \let\deftypevarx = \empty
+ \let\Edeftypevar = \empty
+ \let\deftypevr = \empty
+ \let\deftypevrx = \empty
+ \let\Edeftypevr = \empty
+ \let\defun = \empty
+ \let\defunx = \empty
+ \let\Edefun = \empty
+ \let\defvar = \empty
+ \let\defvarx = \empty
+ \let\Edefvar = \empty
+ \let\defvr = \empty
+ \let\defvrx = \empty
+ \let\Edefvr = \empty
+ \let\clear = \relax
+ \let\down = \relax
+ \let\evenfooting = \relax
+ \let\evenheading = \relax
+ \let\everyfooting = \relax
+ \let\everyheading = \relax
+ \let\headings = \relax
+ \let\include = \relax
+ \let\item = \relax
+ \let\lowersections = \relax
+ \let\oddfooting = \relax
+ \let\oddheading = \relax
+ \let\printindex = \relax
+ \let\pxref = \relax
+ \let\raisesections = \relax
+ \let\ref = \relax
+ \let\set = \relax
+ \let\setchapternewpage = \relax
+ \let\setchapterstyle = \relax
+ \let\settitle = \relax
+ \let\up = \relax
+ \let\verbatiminclude = \relax
+ \let\xref = \relax
+}
+
+% Ignore @ignore, @ifhtml, @ifinfo, and the like.
+%
+\def\direntry{\doignore{direntry}}
+\def\documentdescriptionword{documentdescription}
+\def\documentdescription{\doignore{documentdescription}}
+\def\html{\doignore{html}}
+\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}}
+
+% @dircategory CATEGORY -- specify a category of the dir file
+% which this file should belong to. Ignore this in TeX.
+\let\dircategory = \comment
+
+% Ignore text until a line `@end #1'.
+%
+\def\doignore#1{\begingroup
+ % Don't complain about control sequences we have declared \outer.
+ \ignoresections
+ %
+ % Define a command to swallow text until we reach `@end #1'.
+ % This @ is a catcode 12 token (that is the normal catcode of @ in
+ % this texinfo.tex file). We change the catcode of @ below to match.
+ \long\def\doignoretext##1@end #1{\enddoignore}%
+ %
+ % Make sure that spaces turn into tokens that match what \doignoretext wants.
+ \catcode\spaceChar = 10
+ %
+ % Ignore braces, too, so mismatched braces don't cause trouble.
+ \catcode`\{ = 9
+ \catcode`\} = 9
+ %
+ % We must not have @c interpreted as a control sequence.
+ \catcode`\@ = 12
+ %
+ \def\ignoreword{#1}%
+ \ifx\ignoreword\documentdescriptionword
+ % The c kludge breaks documentdescription, since
+ % `documentdescription' contains a `c'. Means not everything will
+ % be ignored inside @documentdescription, but oh well...
+ \else
+ % Make the letter c a comment character so that the rest of the line
+ % will be ignored. This way, the document can have (for example)
+ % @c @end ifinfo
+ % and the @end ifinfo will be properly ignored.
+ % (We've just changed @ to catcode 12.)
+ \catcode`\c = 14
+ \fi
+ %
+ % And now expand the command defined above.
+ \doignoretext
+}
+
+% What we do to finish off ignored text.
+%
+\def\enddoignore{\endgroup\ignorespaces}%
+
+\newif\ifwarnedobs\warnedobsfalse
+\def\obstexwarn{%
+ \ifwarnedobs\relax\else
+ % We need to warn folks that they may have trouble with TeX 3.0.
+ % This uses \immediate\write16 rather than \message to get newlines.
+ \immediate\write16{}
+ \immediate\write16{WARNING: for users of Unix TeX 3.0!}
+ \immediate\write16{This manual trips a bug in TeX version 3.0 (tex hangs).}
+ \immediate\write16{If you are running another version of TeX, relax.}
+ \immediate\write16{If you are running Unix TeX 3.0, kill this TeX process.}
+ \immediate\write16{ Then upgrade your TeX installation if you can.}
+ \immediate\write16{ (See ftp://ftp.gnu.org/non-gnu/TeX.README.)}
+ \immediate\write16{If you are stuck with version 3.0, run the}
+ \immediate\write16{ script ``tex3patch'' from the Texinfo distribution}
+ \immediate\write16{ to use a workaround.}
+ \immediate\write16{}
+ \global\warnedobstrue
+ \fi
+}
+
+% **In TeX 3.0, setting text in \nullfont hangs tex. For a
+% workaround (which requires the file ``dummy.tfm'' to be installed),
+% uncomment the following line:
+%%%%%\font\nullfont=dummy\let\obstexwarn=\relax
+
+% Ignore text, except that we keep track of conditional commands for
+% purposes of nesting, up to an `@end #1' command.
+%
+\def\nestedignore#1{%
+ \obstexwarn
+ % We must actually expand the ignored text to look for the @end
+ % command, so that nested ignore constructs work. Thus, we put the
+ % text into a \vbox and then do nothing with the result. To minimize
+ % the chance of memory overflow, we follow the approach outlined on
+ % page 401 of the TeXbook.
+ %
+ \setbox0 = \vbox\bgroup
+ % Don't complain about control sequences we have declared \outer.
+ \ignoresections
+ %
+ % Define `@end #1' to end the box, which will in turn undefine the
+ % @end command again.
+ \expandafter\def\csname E#1\endcsname{\egroup\ignorespaces}%
+ %
+ % We are going to be parsing Texinfo commands. Most cause no
+ % trouble when they are used incorrectly, but some commands do
+ % complicated argument parsing or otherwise get confused, so we
+ % undefine them.
+ %
+ % We can't do anything about stray @-signs, unfortunately;
+ % they'll produce `undefined control sequence' errors.
+ \ignoremorecommands
+ %
+ % Set the current font to be \nullfont, a TeX primitive, and define
+ % all the font commands to also use \nullfont. We don't use
+ % dummy.tfm, as suggested in the TeXbook, because some sites
+ % might not have that installed. Therefore, math mode will still
+ % produce output, but that should be an extremely small amount of
+ % stuff compared to the main input.
+ %
+ \nullfont
+ \let\tenrm=\nullfont \let\tenit=\nullfont \let\tensl=\nullfont
+ \let\tenbf=\nullfont \let\tentt=\nullfont \let\smallcaps=\nullfont
+ \let\tensf=\nullfont
+ % Similarly for index fonts.
+ \let\smallrm=\nullfont \let\smallit=\nullfont \let\smallsl=\nullfont
+ \let\smallbf=\nullfont \let\smalltt=\nullfont \let\smallsc=\nullfont
+ \let\smallsf=\nullfont
+ % Similarly for smallexample fonts.
+ \let\smallerrm=\nullfont \let\smallerit=\nullfont \let\smallersl=\nullfont
+ \let\smallerbf=\nullfont \let\smallertt=\nullfont \let\smallersc=\nullfont
+ \let\smallersf=\nullfont
+ %
+ % Don't complain when characters are missing from the fonts.
+ \tracinglostchars = 0
+ %
+ % Don't bother to do space factor calculations.
+ \frenchspacing
+ %
+ % Don't report underfull hboxes.
+ \hbadness = 10000
+ %
+ % Do minimal line-breaking.
+ \pretolerance = 10000
+ %
+ % Do not execute instructions in @tex.
+ \def\tex{\doignore{tex}}%
+ % Do not execute macro definitions.
+ % `c' is a comment character, so the word `macro' will get cut off.
+ \def\macro{\doignore{ma}}%
+}
+
+% @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. Make sure the catcode of space is correct to avoid
+% losing inside @example, for instance.
+%
+\def\set{\begingroup\catcode` =10
+ \catcode`\-=12 \catcode`\_=12 % Allow - and _ in VAR.
+ \parsearg\setxxx}
+\def\setxxx#1{\setyyy#1 \endsetyyy}
+\def\setyyy#1 #2\endsetyyy{%
+ \def\temp{#2}%
+ \ifx\temp\empty \global\expandafter\let\csname SET#1\endcsname = \empty
+ \else \setzzz{#1}#2\endsetzzz % Remove the trailing space \setxxx inserted.
+ \fi
+ \endgroup
+}
+% Can't use \xdef to pre-expand #2 and save some time, since \temp or
+% \next or other control sequences that we've defined might get us into
+% an infinite loop. Consider `@set foo @cite{bar}'.
+\def\setzzz#1#2 \endsetzzz{\expandafter\gdef\csname SET#1\endcsname{#2}}
+
+% @clear VAR clears (i.e., unsets) the variable VAR.
+%
+\def\clear{\parsearg\clearxxx}
+\def\clearxxx#1{\global\expandafter\let\csname SET#1\endcsname=\relax}
+
+% @value{foo} gets the text saved in variable foo.
+{
+ \catcode`\_ = \active
+ %
+ % We might end up with active _ or - characters in the argument if
+ % we're called from @code, as @code{@value{foo-bar_}}. So \let any
+ % such active characters to their normal equivalents.
+ \gdef\value{\begingroup
+ \catcode`\-=\other \catcode`\_=\other
+ \indexbreaks \let_\normalunderscore
+ \valuexxx}
+}
+\def\valuexxx#1{\expandablevalue{#1}\endgroup}
+
+% We have this subroutine so that we can handle at least some @value's
+% properly in indexes (we \let\value to this in \indexdummies). Ones
+% whose names contain - or _ still won't work, but we can't do anything
+% about that. 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.
+%
+\def\ifset{\parsearg\doifset}
+\def\doifset#1{%
+ \expandafter\ifx\csname SET#1\endcsname\relax
+ \let\next=\ifsetfail
+ \else
+ \let\next=\ifsetsucceed
+ \fi
+ \next
+}
+\def\ifsetsucceed{\conditionalsucceed{ifset}}
+\def\ifsetfail{\nestedignore{ifset}}
+\defineunmatchedend{ifset}
+
+% @ifclear VAR ... @end ifclear reads the `...' iff VAR has never been
+% defined with @set, or has been undefined with @clear.
+%
+\def\ifclear{\parsearg\doifclear}
+\def\doifclear#1{%
+ \expandafter\ifx\csname SET#1\endcsname\relax
+ \let\next=\ifclearsucceed
+ \else
+ \let\next=\ifclearfail
+ \fi
+ \next
+}
+\def\ifclearsucceed{\conditionalsucceed{ifclear}}
+\def\ifclearfail{\nestedignore{ifclear}}
+\defineunmatchedend{ifclear}
+
+% @iftex, @ifnothtml, @ifnotinfo, @ifnotplaintext always succeed; we
+% read the text following, through the first @end iftex (etc.). Make
+% `@end iftex' (etc.) valid only after an @iftex.
+%
+\def\iftex{\conditionalsucceed{iftex}}
+\def\ifnothtml{\conditionalsucceed{ifnothtml}}
+\def\ifnotinfo{\conditionalsucceed{ifnotinfo}}
+\def\ifnotplaintext{\conditionalsucceed{ifnotplaintext}}
+\defineunmatchedend{iftex}
+\defineunmatchedend{ifnothtml}
+\defineunmatchedend{ifnotinfo}
+\defineunmatchedend{ifnotplaintext}
+
+% True conditional. Since \set globally defines its variables, we can
+% just start and end a group (to keep the @end definition undefined at
+% the outer level).
+%
+\def\conditionalsucceed#1{\begingroup
+ \expandafter\def\csname E#1\endcsname{\endgroup}%
+}
+
+% @defininfoenclose.
+\let\definfoenclose=\comment
+
+
+\message{indexing,}
+% Index generation facilities
+
+% Define \newwrite to be identical to plain tex's \newwrite
+% except not \outer, so it can be used within \newindex.
+{\catcode`\@=11
+\gdef\newwrite{\alloc@7\write\chardef\sixt@@n}}
+
+% \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{%
+ \def\@{@}% change to @@ when we switch to @ as escape char in index files.
+ \def\ {\realbackslash\space }%
+ % Need these in case \tex is in effect and \{ is a \delimiter again.
+ % But can't use \lbracecmd and \rbracecmd because texindex assumes
+ % braces and backslashes are used only as delimiters.
+ \let\{ = \mylbrace
+ \let\} = \myrbrace
+ %
+ % \definedummyword defines \#1 as \realbackslash #1\space, thus
+ % effectively preventing its expansion. This is used only for control
+ % words, not control letters, because the \space would be incorrect
+ % for control characters, but is needed to separate the control word
+ % from whatever follows.
+ %
+ % For control letters, we have \definedummyletter, which omits the
+ % space.
+ %
+ % These can be used both for control words that take an argument and
+ % those that do not. If it is followed by {arg} in the input, then
+ % that will dutifully get written to the index (or wherever).
+ %
+ \def\definedummyword##1{%
+ \expandafter\def\csname ##1\endcsname{\realbackslash ##1\space}%
+ }%
+ \def\definedummyletter##1{%
+ \expandafter\def\csname ##1\endcsname{\realbackslash ##1}%
+ }%
+ %
+ % Do the redefinitions.
+ \commondummies
+}
+
+% For the aux file, @ is the escape character. So we want to redefine
+% everything using @ instead of \realbackslash. When everything uses
+% @, this will be simpler.
+%
+\def\atdummies{%
+ \def\@{@@}%
+ \def\ {@ }%
+ \let\{ = \lbraceatcmd
+ \let\} = \rbraceatcmd
+ %
+ % (See comments in \indexdummies.)
+ \def\definedummyword##1{%
+ \expandafter\def\csname ##1\endcsname{@##1\space}%
+ }%
+ \def\definedummyletter##1{%
+ \expandafter\def\csname ##1\endcsname{@##1}%
+ }%
+ %
+ % Do the redefinitions.
+ \commondummies
+}
+
+% Called from \indexdummies and \atdummies. \definedummyword and
+% \definedummyletter must be defined first.
+%
+\def\commondummies{%
+ %
+ \normalturnoffactive
+ %
+ % Control letters and accents.
+ \definedummyletter{_}%
+ \definedummyletter{,}%
+ \definedummyletter{"}%
+ \definedummyletter{`}%
+ \definedummyletter{'}%
+ \definedummyletter{^}%
+ \definedummyletter{~}%
+ \definedummyletter{=}%
+ \definedummyword{u}%
+ \definedummyword{v}%
+ \definedummyword{H}%
+ \definedummyword{dotaccent}%
+ \definedummyword{ringaccent}%
+ \definedummyword{tieaccent}%
+ \definedummyword{ubaraccent}%
+ \definedummyword{udotaccent}%
+ \definedummyword{dotless}%
+ %
+ % Other 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}%
+ %
+ % 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}%
+ %
+ % Texinfo font commands.
+ \definedummyword{b}%
+ \definedummyword{i}%
+ \definedummyword{r}%
+ \definedummyword{sc}%
+ \definedummyword{t}%
+ %
+ \definedummyword{TeX}%
+ \definedummyword{acronym}%
+ \definedummyword{cite}%
+ \definedummyword{code}%
+ \definedummyword{command}%
+ \definedummyword{dfn}%
+ \definedummyword{dots}%
+ \definedummyword{emph}%
+ \definedummyword{env}%
+ \definedummyword{file}%
+ \definedummyword{kbd}%
+ \definedummyword{key}%
+ \definedummyword{math}%
+ \definedummyword{option}%
+ \definedummyword{samp}%
+ \definedummyword{strong}%
+ \definedummyword{uref}%
+ \definedummyword{url}%
+ \definedummyword{var}%
+ \definedummyword{w}%
+ %
+ % Assorted special characters.
+ \definedummyword{bullet}%
+ \definedummyword{copyright}%
+ \definedummyword{dots}%
+ \definedummyword{enddots}%
+ \definedummyword{equiv}%
+ \definedummyword{error}%
+ \definedummyword{expansion}%
+ \definedummyword{minus}%
+ \definedummyword{pounds}%
+ \definedummyword{point}%
+ \definedummyword{print}%
+ \definedummyword{result}%
+ %
+ % Handle some cases of @value -- where the variable name does not
+ % contain - or _, and the value does not contain any
+ % (non-fully-expandable) commands.
+ \let\value = \expandablevalue
+ %
+ % Normal spaces, not active ones.
+ \unsepspaces
+ %
+ % No macro expansion.
+ \turnoffmacros
+}
+
+% 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 \ ).
+{\obeyspaces
+ \gdef\unsepspaces{\obeyspaces\let =\space}}
+
+
+% \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\indexdummytex{TeX}
+\def\indexdummydots{...}
+%
+\def\indexnofonts{%
+ \def\ { }%
+ \def\@{@}%
+ % how to handle braces?
+ \def\_{\normalunderscore}%
+ %
+ \let\,=\asis
+ \let\"=\asis
+ \let\`=\asis
+ \let\'=\asis
+ \let\^=\asis
+ \let\~=\asis
+ \let\==\asis
+ \let\u=\asis
+ \let\v=\asis
+ \let\H=\asis
+ \let\dotaccent=\asis
+ \let\ringaccent=\asis
+ \let\tieaccent=\asis
+ \let\ubaraccent=\asis
+ \let\udotaccent=\asis
+ \let\dotless=\asis
+ %
+ % Other 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{?}%
+ %
+ % 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
+ %
+ % Texinfo font commands.
+ \let\b=\asis
+ \let\i=\asis
+ \let\r=\asis
+ \let\sc=\asis
+ \let\t=\asis
+ %
+ \let\TeX=\indexdummytex
+ \let\acronym=\asis
+ \let\cite=\asis
+ \let\code=\asis
+ \let\command=\asis
+ \let\dfn=\asis
+ \let\dots=\indexdummydots
+ \let\emph=\asis
+ \let\env=\asis
+ \let\file=\asis
+ \let\kbd=\asis
+ \let\key=\asis
+ \let\math=\asis
+ \let\option=\asis
+ \let\samp=\asis
+ \let\strong=\asis
+ \let\uref=\asis
+ \let\url=\asis
+ \let\var=\asis
+ \let\w=\asis
+}
+
+\let\indexbackslash=0 %overridden during \printindex.
+\let\SETmarginindex=\relax % put index entries in margin (undocumented)?
+
+% For \ifx comparisons.
+\def\emptymacro{\empty}
+
+% Most index entries go through here, but \dosubind is the general case.
+%
+\def\doind#1#2{\dosubind{#1}{#2}\empty}
+
+% 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 defuns, which call us directly.
+%
+\def\dosubind#1#2#3{%
+ % Put the index entry in the margin if desired.
+ \ifx\SETmarginindex\relax\else
+ \insert\margin{\hbox{\vrule height8pt depth3pt width0pt #2}}%
+ \fi
+ {%
+ \count255=\lastpenalty
+ {%
+ \indexdummies % Must do this here, since \bf, etc expand at this stage
+ \escapechar=`\\
+ {%
+ \let\folio = 0% We will expand all macros now EXCEPT \folio.
+ \def\rawbackslashxx{\indexbackslash}% \indexbackslash isn't defined now
+ % so it will be output as is; and it will print as backslash.
+ %
+ % The main index entry text.
+ \toks0 = {#2}%
+ %
+ % If third arg is present, precede it with space in sort key.
+ \def\thirdarg{#3}%
+ \ifx\thirdarg\emptymacro \else
+ % If the third (subentry) arg is present, add it to the index
+ % line to write.
+ \toks0 = \expandafter{\the\toks0 \space #3}%
+ \fi
+ %
+ % 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\csname#1indfile\endcsname{%
+ \realbackslash entry{\indexsorttmp}{\folio}{\the\toks0}}%
+ }%
+ %
+ % If a skip is the last thing on the list now, preserve it
+ % by backing up by \lastskip, doing the \write, then inserting
+ % the skip again. Otherwise, the whatsit generated by the
+ % \write will make \lastskip zero. The result is that sequences
+ % like this:
+ % @end defun
+ % @tindex whatever
+ % @defun ...
+ % will have extra space inserted, because the \medbreak in the
+ % start of the @defun won't see the skip inserted by the @end of
+ % the previous defun.
+ %
+ % But don't do any of this if we're not in vertical mode. We
+ % don't want to do a \vskip and prematurely end a paragraph.
+ %
+ % Avoid page breaks due to these extra skips, too.
+ %
+ \iflinks
+ \ifvmode
+ \skip0 = \lastskip
+ \ifdim\lastskip = 0pt \else \nobreak\vskip-\skip0 \fi
+ \fi
+ %
+ \temp % do the write
+ %
+ \ifvmode \ifdim\skip0 = 0pt \else \nobreak\vskip\skip0 \fi \fi
+ \fi
+ }%
+ }%
+ \penalty\count255
+ }%
+}
+
+% 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).
+%
+\def\printindex{\parsearg\doprintindex}
+\def\doprintindex#1{\begingroup
+ \dobreak \chapheadingskip{10000}%
+ %
+ \smallfonts \rm
+ \tolerance = 9500
+ \everypar = {}% don't want the \kern\-parindent from indentation suppression.
+ \indexbreaks
+ %
+ % 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{\rawbackslashxx}%
+ \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.
+ \penalty -300
+ %
+ % 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}%
+ \vskip .33\baselineskip plus .1\baselineskip
+ %
+ % Do our best not to break after the initial.
+ \nobreak
+}}
+
+% This typesets a paragraph consisting of #1, dot leaders, and then #2
+% flush to the right margin. It is used for index and table of contents
+% entries. The paragraph is indented by \leftskip.
+%
+\def\entry#1#2{\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
+ %
+ % Start a ``paragraph'' for the index entry so the line breaking
+ % parameters we've set above will have an effect.
+ \noindent
+ %
+ % Insert the text of the index entry. TeX will do line-breaking on it.
+ #1%
+ % 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{#2}%
+ \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#2.\ \the\toksA % The page number ends the paragraph.
+ \else
+ \ #2% The page number ends the paragraph.
+ \fi
+ \fi%
+ \par
+\endgroup}
+
+% Like \dotfill except takes at least 1 em.
+\def\indexdotfill{\cleaders
+ \hbox{$\mathsurround=0pt \mkern1.5mu ${\it .}$ \mkern1.5mu$}\hskip 1em plus 1fill}
+
+\def\primary #1{\line{#1\hfil}}
+
+\newskip\secondaryindent \secondaryindent=0.5cm
+\def\secondary#1#2{{%
+ \parfillskip=0in
+ \parskip=0in
+ \hangindent=1in
+ \hangafter=1
+ \noindent\hskip\secondaryindent\hbox{#1}\indexdotfill
+ \ifpdf
+ \pdfgettoks#2.\ \the\toksA % The page number ends the paragraph.
+ \else
+ #2
+ \fi
+ \par
+}}
+
+% Define two-column mode, which we use to typeset indexes.
+% Adapted from the TeXbook, page 416, which is to say,
+% the manmac.tex format used to print the TeXbook itself.
+\catcode`\@=11
+
+\newbox\partialpage
+\newdimen\doublecolumnhsize
+
+\def\begindoublecolumns{\begingroup % ended by \enddoublecolumns
+ % Grab any single-column material above us.
+ \output = {%
+ %
+ % Here is a possibility not foreseen in manmac: if we accumulate a
+ % whole lot of material, we might end up calling this \output
+ % routine twice in a row (see the doublecol-lose test, which is
+ % essentially a couple of indexes with @setchapternewpage off). In
+ % that case we just ship out what is in \partialpage with the normal
+ % output routine. Generally, \partialpage will be empty when this
+ % runs and this will be a no-op. See the indexspread.tex test case.
+ \ifvoid\partialpage \else
+ \onepageout{\pagecontents\partialpage}%
+ \fi
+ %
+ \global\setbox\partialpage = \vbox{%
+ % Unvbox the main output page.
+ \unvbox\PAGE
+ \kern-\topskip \kern\baselineskip
+ }%
+ }%
+ \eject % run that output routine to set \partialpage
+ %
+ % Use the double-column output routine for subsequent pages.
+ \output = {\doublecolumnout}%
+ %
+ % Change the page size parameters. We could do this once outside this
+ % routine, in each of @smallbook, @afourpaper, and the default 8.5x11
+ % format, but then we repeat the same computation. Repeating a couple
+ % of assignments once per index is clearly meaningless for the
+ % execution time, so we may as well do it in one place.
+ %
+ % First we halve the line length, less a little for the gutter between
+ % the columns. We compute the gutter based on the line length, so it
+ % changes automatically with the paper format. The magic constant
+ % below is chosen so that the gutter has the same value (well, +-<1pt)
+ % as it did when we hard-coded it.
+ %
+ % We put the result in a separate register, \doublecolumhsize, so we
+ % can restore it in \pagesofar, after \hsize itself has (potentially)
+ % been clobbered.
+ %
+ \doublecolumnhsize = \hsize
+ \advance\doublecolumnhsize by -.04154\hsize
+ \divide\doublecolumnhsize by 2
+ \hsize = \doublecolumnhsize
+ %
+ % Double the \vsize as well. (We don't need a separate register here,
+ % since nobody clobbers \vsize.)
+ \vsize = 2\vsize
+}
+
+% The double-column output routine for all double-column pages except
+% the last.
+%
+\def\doublecolumnout{%
+ \splittopskip=\topskip \splitmaxdepth=\maxdepth
+ % Get the available space for the double columns -- the normal
+ % (undoubled) page height minus any material left over from the
+ % previous page.
+ \dimen@ = \vsize
+ \divide\dimen@ by 2
+ \advance\dimen@ by -\ht\partialpage
+ %
+ % box0 will be the left-hand column, box2 the right.
+ \setbox0=\vsplit255 to\dimen@ \setbox2=\vsplit255 to\dimen@
+ \onepageout\pagesofar
+ \unvbox255
+ \penalty\outputpenalty
+}
+%
+% Re-output the contents of the output page -- any previous material,
+% followed by the two boxes we just split, in box0 and box2.
+\def\pagesofar{%
+ \unvbox\partialpage
+ %
+ \hsize = \doublecolumnhsize
+ \wd0=\hsize \wd2=\hsize
+ \hbox to\pagewidth{\box0\hfil\box2}%
+}
+%
+% All done with double columns.
+\def\enddoublecolumns{%
+ \output = {%
+ % Split the last of the double-column material. Leave it on the
+ % current page, no automatic page break.
+ \balancecolumns
+ %
+ % If we end up splitting too much material for the current page,
+ % though, there will be another page break right after this \output
+ % invocation ends. Having called \balancecolumns once, we do not
+ % want to call it again. Therefore, reset \output to its normal
+ % definition right away. (We hope \balancecolumns will never be
+ % called on to balance too much material, but if it is, this makes
+ % the output somewhat more palatable.)
+ \global\output = {\onepageout{\pagecontents\PAGE}}%
+ }%
+ \eject
+ \endgroup % started in \begindoublecolumns
+ %
+ % \pagegoal was set to the doubled \vsize above, since we restarted
+ % the current page. We're now back to normal single-column
+ % typesetting, so reset \pagegoal to the normal \vsize (after the
+ % \endgroup where \vsize got restored).
+ \pagegoal = \vsize
+}
+%
+% Called at the end of the double column material.
+\def\balancecolumns{%
+ \setbox0 = \vbox{\unvbox255}% like \box255 but more efficient, see p.120.
+ \dimen@ = \ht0
+ \advance\dimen@ by \topskip
+ \advance\dimen@ by-\baselineskip
+ \divide\dimen@ by 2 % target to split to
+ %debug\message{final 2-column material height=\the\ht0, target=\the\dimen@.}%
+ \splittopskip = \topskip
+ % Loop until we get a decent breakpoint.
+ {%
+ \vbadness = 10000
+ \loop
+ \global\setbox3 = \copy0
+ \global\setbox1 = \vsplit3 to \dimen@
+ \ifdim\ht3>\dimen@
+ \global\advance\dimen@ by 1pt
+ \repeat
+ }%
+ %debug\message{split to \the\dimen@, column heights: \the\ht1, \the\ht3.}%
+ \setbox0=\vbox to\dimen@{\unvbox1}%
+ \setbox2=\vbox to\dimen@{\unvbox3}%
+ %
+ \pagesofar
+}
+\catcode`\@ = \other
+
+
+\message{sectioning,}
+% Chapters, sections, etc.
+
+\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 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.
+\def\thischapter{}
+\def\thissection{}
+
+\newcount\absseclevel % used to calculate proper heading level
+\newcount\secbase\secbase=0 % @raise/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
+
+% Choose a numbered-heading macro
+% #1 is heading level if unmodified by @raisesections or @lowersections
+% #2 is text for heading
+\def\numhead#1#2{\absseclevel=\secbase\advance\absseclevel by #1
+\ifcase\absseclevel
+ \chapterzzz{#2}
+\or
+ \seczzz{#2}
+\or
+ \numberedsubseczzz{#2}
+\or
+ \numberedsubsubseczzz{#2}
+\else
+ \ifnum \absseclevel<0
+ \chapterzzz{#2}
+ \else
+ \numberedsubsubseczzz{#2}
+ \fi
+\fi
+\suppressfirstparagraphindent
+}
+
+% like \numhead, but chooses appendix heading levels
+\def\apphead#1#2{\absseclevel=\secbase\advance\absseclevel by #1
+\ifcase\absseclevel
+ \appendixzzz{#2}
+\or
+ \appendixsectionzzz{#2}
+\or
+ \appendixsubseczzz{#2}
+\or
+ \appendixsubsubseczzz{#2}
+\else
+ \ifnum \absseclevel<0
+ \appendixzzz{#2}
+ \else
+ \appendixsubsubseczzz{#2}
+ \fi
+\fi
+\suppressfirstparagraphindent
+}
+
+% like \numhead, but chooses numberless heading levels
+\def\unnmhead#1#2{\absseclevel=\secbase\advance\absseclevel by #1
+\ifcase\absseclevel
+ \unnumberedzzz{#2}
+\or
+ \unnumberedseczzz{#2}
+\or
+ \unnumberedsubseczzz{#2}
+\or
+ \unnumberedsubsubseczzz{#2}
+\else
+ \ifnum \absseclevel<0
+ \unnumberedzzz{#2}
+ \else
+ \unnumberedsubsubseczzz{#2}
+ \fi
+\fi
+\suppressfirstparagraphindent
+}
+
+% @chapter, @appendix, @unnumbered.
+\def\thischaptername{No Chapter Title}
+\outer\def\chapter{\parsearg\chapteryyy}
+\def\chapteryyy #1{\numhead0{#1}} % normally numhead0 calls chapterzzz
+\def\chapterzzz #1{%
+ \secno=0 \subsecno=0 \subsubsecno=0
+ \global\advance \chapno by 1 \message{\putwordChapter\space \the\chapno}%
+ \chapmacro {#1}{\the\chapno}%
+ \gdef\thissection{#1}%
+ \gdef\thischaptername{#1}%
+ % We don't substitute the actual chapter name into \thischapter
+ % because we don't want its macros evaluated now.
+ \xdef\thischapter{\putwordChapter{} \the\chapno: \noexpand\thischaptername}%
+ \writetocentry{chap}{#1}{{\the\chapno}}
+ \donoderef
+ \global\let\section = \numberedsec
+ \global\let\subsection = \numberedsubsec
+ \global\let\subsubsection = \numberedsubsubsec
+}
+
+% we use \chapno to avoid indenting back
+\def\appendixbox#1{%
+ \setbox0 = \hbox{\putwordAppendix{} \the\chapno}%
+ \hbox to \wd0{#1\hss}}
+
+\outer\def\appendix{\parsearg\appendixyyy}
+\def\appendixyyy #1{\apphead0{#1}} % normally apphead0 calls appendixzzz
+\def\appendixzzz #1{%
+ \secno=0 \subsecno=0 \subsubsecno=0
+ \global\advance \appendixno by 1
+ \message{\putwordAppendix\space \appendixletter}%
+ \chapmacro {#1}{\appendixbox{\putwordAppendix{} \appendixletter}}%
+ \gdef\thissection{#1}%
+ \gdef\thischaptername{#1}%
+ \xdef\thischapter{\putwordAppendix{} \appendixletter: \noexpand\thischaptername}%
+ \writetocentry{appendix}{#1}{{\appendixletter}}
+ \appendixnoderef
+ \global\let\section = \appendixsec
+ \global\let\subsection = \appendixsubsec
+ \global\let\subsubsection = \appendixsubsubsec
+}
+
+% @centerchap is like @unnumbered, but the heading is centered.
+\outer\def\centerchap{\parsearg\centerchapyyy}
+\def\centerchapyyy #1{{\let\unnumbchapmacro=\centerchapmacro \unnumberedyyy{#1}}}
+
+% @top is like @unnumbered.
+\outer\def\top{\parsearg\unnumberedyyy}
+
+\outer\def\unnumbered{\parsearg\unnumberedyyy}
+\def\unnumberedyyy #1{\unnmhead0{#1}} % normally unnmhead0 calls unnumberedzzz
+\def\unnumberedzzz #1{%
+ \secno=0 \subsecno=0 \subsubsecno=0
+ %
+ % 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)}%
+ %
+ \unnumbchapmacro {#1}%
+ \gdef\thischapter{#1}\gdef\thissection{#1}%
+ \writetocentry{unnumbchap}{#1}{{\the\chapno}}
+ \unnumbnoderef
+ \global\let\section = \unnumberedsec
+ \global\let\subsection = \unnumberedsubsec
+ \global\let\subsubsection = \unnumberedsubsubsec
+}
+
+% Sections.
+\outer\def\numberedsec{\parsearg\secyyy}
+\def\secyyy #1{\numhead1{#1}} % normally calls seczzz
+\def\seczzz #1{%
+ \subsecno=0 \subsubsecno=0 \global\advance \secno by 1 %
+ \gdef\thissection{#1}\secheading {#1}{\the\chapno}{\the\secno}%
+ \writetocentry{sec}{#1}{{\the\chapno}{\the\secno}}
+ \donoderef
+ \nobreak
+}
+
+\outer\def\appendixsection{\parsearg\appendixsecyyy}
+\outer\def\appendixsec{\parsearg\appendixsecyyy}
+\def\appendixsecyyy #1{\apphead1{#1}} % normally calls appendixsectionzzz
+\def\appendixsectionzzz #1{%
+ \subsecno=0 \subsubsecno=0 \global\advance \secno by 1 %
+ \gdef\thissection{#1}\secheading {#1}{\appendixletter}{\the\secno}%
+ \writetocentry{sec}{#1}{{\appendixletter}{\the\secno}}
+ \appendixnoderef
+ \nobreak
+}
+
+\outer\def\unnumberedsec{\parsearg\unnumberedsecyyy}
+\def\unnumberedsecyyy #1{\unnmhead1{#1}} % normally calls unnumberedseczzz
+\def\unnumberedseczzz #1{%
+ \plainsecheading {#1}\gdef\thissection{#1}%
+ \writetocentry{unnumbsec}{#1}{{\the\chapno}{\the\secno}}
+ \unnumbnoderef
+ \nobreak
+}
+
+% Subsections.
+\outer\def\numberedsubsec{\parsearg\numberedsubsecyyy}
+\def\numberedsubsecyyy #1{\numhead2{#1}} % normally calls numberedsubseczzz
+\def\numberedsubseczzz #1{%
+ \gdef\thissection{#1}\subsubsecno=0 \global\advance \subsecno by 1 %
+ \subsecheading {#1}{\the\chapno}{\the\secno}{\the\subsecno}%
+ \writetocentry{subsec}{#1}{{\the\chapno}{\the\secno}{\the\subsecno}}
+ \donoderef
+ \nobreak
+}
+
+\outer\def\appendixsubsec{\parsearg\appendixsubsecyyy}
+\def\appendixsubsecyyy #1{\apphead2{#1}} % normally calls appendixsubseczzz
+\def\appendixsubseczzz #1{%
+ \gdef\thissection{#1}\subsubsecno=0 \global\advance \subsecno by 1 %
+ \subsecheading {#1}{\appendixletter}{\the\secno}{\the\subsecno}%
+ \writetocentry{subsec}{#1}{{\appendixletter}{\the\secno}{\the\subsecno}}
+ \appendixnoderef
+ \nobreak
+}
+
+\outer\def\unnumberedsubsec{\parsearg\unnumberedsubsecyyy}
+\def\unnumberedsubsecyyy #1{\unnmhead2{#1}} %normally calls unnumberedsubseczzz
+\def\unnumberedsubseczzz #1{%
+ \plainsubsecheading {#1}\gdef\thissection{#1}%
+ \writetocentry{unnumbsubsec}{#1}{{\the\chapno}{\the\secno}{\the\subsecno}}
+ \unnumbnoderef
+ \nobreak
+}
+
+% Subsubsections.
+\outer\def\numberedsubsubsec{\parsearg\numberedsubsubsecyyy}
+\def\numberedsubsubsecyyy #1{\numhead3{#1}} % normally numberedsubsubseczzz
+\def\numberedsubsubseczzz #1{%
+ \gdef\thissection{#1}\global\advance \subsubsecno by 1 %
+ \subsubsecheading {#1}
+ {\the\chapno}{\the\secno}{\the\subsecno}{\the\subsubsecno}%
+ \writetocentry{subsubsec}{#1}{{\the\chapno}{\the\secno}{\the\subsecno}{\the\subsubsecno}}
+ \donoderef
+ \nobreak
+}
+
+\outer\def\appendixsubsubsec{\parsearg\appendixsubsubsecyyy}
+\def\appendixsubsubsecyyy #1{\apphead3{#1}} % normally appendixsubsubseczzz
+\def\appendixsubsubseczzz #1{%
+ \gdef\thissection{#1}\global\advance \subsubsecno by 1 %
+ \subsubsecheading {#1}
+ {\appendixletter}{\the\secno}{\the\subsecno}{\the\subsubsecno}%
+ \writetocentry{subsubsec}{#1}{{\appendixletter}{\the\secno}{\the\subsecno}{\the\subsubsecno}}
+ \appendixnoderef
+ \nobreak
+}
+
+\outer\def\unnumberedsubsubsec{\parsearg\unnumberedsubsubsecyyy}
+\def\unnumberedsubsubsecyyy #1{\unnmhead3{#1}} %normally unnumberedsubsubseczzz
+\def\unnumberedsubsubseczzz #1{%
+ \plainsubsubsecheading {#1}\gdef\thissection{#1}%
+ \writetocentry{unnumbsubsubsec}{#1}{{\the\chapno}{\the\secno}{\the\subsecno}{\the\subsubsecno}}
+ \unnumbnoderef
+ \nobreak
+}
+
+% These are variants which are not "outer", so they can appear in @ifinfo.
+% Actually, they should now be obsolete; ordinary section commands should work.
+\def\infotop{\parsearg\unnumberedzzz}
+\def\infounnumbered{\parsearg\unnumberedzzz}
+\def\infounnumberedsec{\parsearg\unnumberedseczzz}
+\def\infounnumberedsubsec{\parsearg\unnumberedsubseczzz}
+\def\infounnumberedsubsubsec{\parsearg\unnumberedsubsubseczzz}
+
+\def\infoappendix{\parsearg\appendixzzz}
+\def\infoappendixsec{\parsearg\appendixseczzz}
+\def\infoappendixsubsec{\parsearg\appendixsubseczzz}
+\def\infoappendixsubsubsec{\parsearg\appendixsubsubseczzz}
+
+\def\infochapter{\parsearg\chapterzzz}
+\def\infosection{\parsearg\sectionzzz}
+\def\infosubsection{\parsearg\subsectionzzz}
+\def\infosubsubsection{\parsearg\subsubsectionzzz}
+
+% 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.
+\global\let\section = \numberedsec
+\global\let\subsection = \numberedsubsec
+\global\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{\parsearg\majorheadingzzz}
+\def\majorheadingzzz #1{%
+ {\advance\chapheadingskip by 10pt \chapbreak }%
+ {\chapfonts \vbox{\hyphenpenalty=10000\tolerance=5000
+ \parindent=0pt\raggedright
+ \rm #1\hfill}}\bigskip \par\penalty 200}
+
+\def\chapheading{\parsearg\chapheadingzzz}
+\def\chapheadingzzz #1{\chapbreak %
+ {\chapfonts \vbox{\hyphenpenalty=10000\tolerance=5000
+ \parindent=0pt\raggedright
+ \rm #1\hfill}}\bigskip \par\penalty 200}
+
+% @heading, @subheading, @subsubheading.
+\def\heading{\parsearg\plainsecheading}
+\def\subheading{\parsearg\plainsubsecheading}
+\def\subsubheading{\parsearg\plainsubsubsecheading}
+
+% 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}
+
+\def\setchapterstyle #1 {\csname CHAPF#1\endcsname}
+
+%%% 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
+
+\def\CHAPFplain{
+\global\let\chapmacro=\chfplain
+\global\let\unnumbchapmacro=\unnchfplain
+\global\let\centerchapmacro=\centerchfplain}
+
+% Plain chapter opening.
+% #1 is the text, #2 the chapter number or empty if unnumbered.
+\def\chfplain#1#2{%
+ \pchapsepmacro
+ {%
+ \chapfonts \rm
+ \def\chapnum{#2}%
+ \setbox0 = \hbox{#2\ifx\chapnum\empty\else\enspace\fi}%
+ \vbox{\hyphenpenalty=10000 \tolerance=5000 \parindent=0pt \raggedright
+ \hangindent = \wd0 \centerparametersmaybe
+ \unhbox0 #1\par}%
+ }%
+ \nobreak\bigskip % no page break after a chapter title
+ \nobreak
+}
+
+% Plain opening for unnumbered.
+\def\unnchfplain#1{\chfplain{#1}{}}
+
+% @centerchap -- centered and unnumbered.
+\let\centerparametersmaybe = \relax
+\def\centerchfplain#1{{%
+ \def\centerparametersmaybe{%
+ \advance\rightskip by 3\rightskip
+ \leftskip = \rightskip
+ \parfillskip = 0pt
+ }%
+ \chfplain{#1}{}%
+}}
+
+\CHAPFplain % The default
+
+\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\unnumbchapmacro=\unnchfopen
+\global\let\centerchapmacro=\centerchfopen}
+
+
+% Section titles.
+\newskip\secheadingskip
+\def\secheadingbreak{\dobreak \secheadingskip {-1000}}
+\def\secheading#1#2#3{\sectionheading{sec}{#2.#3}{#1}}
+\def\plainsecheading#1{\sectionheading{sec}{}{#1}}
+
+% Subsection titles.
+\newskip \subsecheadingskip
+\def\subsecheadingbreak{\dobreak \subsecheadingskip {-500}}
+\def\subsecheading#1#2#3#4{\sectionheading{subsec}{#2.#3.#4}{#1}}
+\def\plainsubsecheading#1{\sectionheading{subsec}{}{#1}}
+
+% Subsubsection titles.
+\let\subsubsecheadingskip = \subsecheadingskip
+\let\subsubsecheadingbreak = \subsecheadingbreak
+\def\subsubsecheading#1#2#3#4#5{\sectionheading{subsubsec}{#2.#3.#4.#5}{#1}}
+\def\plainsubsubsecheading#1{\sectionheading{subsubsec}{}{#1}}
+
+
+% Print any size section title.
+%
+% #1 is the section type (sec/subsec/subsubsec), #2 is the section
+% number (maybe empty), #3 the text.
+\def\sectionheading#1#2#3{%
+ {%
+ \expandafter\advance\csname #1headingskip\endcsname by \parskip
+ \csname #1headingbreak\endcsname
+ }%
+ {%
+ % Switch to the right set of fonts.
+ \csname #1fonts\endcsname \rm
+ %
+ % Only insert the separating space if we have a section number.
+ \def\secnum{#2}%
+ \setbox0 = \hbox{#2\ifx\secnum\empty\else\enspace\fi}%
+ %
+ \vbox{\hyphenpenalty=10000 \tolerance=5000 \parindent=0pt \raggedright
+ \hangindent = \wd0 % zero if no section number
+ \unhbox0 #3}%
+ }%
+ % Add extra space after the heading -- either a line space or a
+ % paragraph space, whichever is more. (Some people like to set
+ % \parskip to large values for some reason.) Don't allow stretch, though.
+ \nobreak
+ \ifdim\parskip>\normalbaselineskip
+ \kern\parskip
+ \else
+ \kern\normalbaselineskip
+ \fi
+ \nobreak
+}
+
+
+\message{toc,}
+% Table of contents.
+\newwrite\tocfile
+
+% Write an entry to the toc file, opening it if necessary.
+% Called from @chapter, etc. We supply {\folio} at the end of the
+% argument, which will end up as the last argument to the \...entry macro.
+%
+% Usage: \writetocentry{chap}{The Name of The Game}{{\the\chapno}}
+% 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.
+%
+\newif\iftocfileopened
+\def\writetocentry#1#2#3{%
+ \iftocfileopened\else
+ \immediate\openout\tocfile = \jobname.toc
+ \global\tocfileopenedtrue
+ \fi
+ %
+ \iflinks
+ \toks0 = {#2}%
+ \edef\temp{\write\tocfile{\realbackslash #1entry{\the\toks0}#3{\folio}}}%
+ \temp
+ \fi
+ %
+ % Tell \shipout to create a page destination if we're doing pdf, which
+ % will be the target of the links in the table of contents. We can't
+ % just do it 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 \pdfmakepagedesttrue \fi
+}
+
+\newskip\contentsrightmargin \contentsrightmargin=1in
+\newcount\savepageno
+\newcount\lastnegativepageno \lastnegativepageno = -1
+
+% Finish up the main text and 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.
+ \unnumbchapmacro{#1}\def\thischapter{}%
+ \savepageno = \pageno
+ \begingroup % Set up to handle contents files properly.
+ \catcode`\\=0 \catcode`\{=1 \catcode`\}=2 \catcode`\@=11
+ % We can't do this, because then an actual ^ in a section
+ % title fails, e.g., @chapter ^ -- exponentiation. --karl, 9jul97.
+ %\catcode`\^=7 % to see ^^e4 as \"a etc. juha@piuha.ydi.vtt.fi
+ \raggedbottom % Worry more about breakpoints than the bottom.
+ \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
+ \closein 1
+ \input \jobname.toc
+ \fi
+ \vfill \eject
+ \contentsalignmacro % in case @setchapternewpage odd is in effect
+ \pdfmakeoutlines
+ \endgroup
+ \lastnegativepageno = \pageno
+ \global\pageno = \savepageno
+}
+
+% And just the chapters.
+\def\summarycontents{%
+ \startcontents{\putwordShortTOC}%
+ %
+ \let\chapentry = \shortchapentry
+ \let\appendixentry = \shortappendixentry
+ \let\unnumbchapentry = \shortunnumberedentry
+ % 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\secentry ##1##2##3##4{}
+ \def\subsecentry ##1##2##3##4##5{}
+ \def\subsubsecentry ##1##2##3##4##5##6{}
+ \let\unnumbsecentry = \secentry
+ \let\unnumbsubsecentry = \subsecentry
+ \let\unnumbsubsubsecentry = \subsubsecentry
+ \openin 1 \jobname.toc
+ \ifeof 1 \else
+ \closein 1
+ \input \jobname.toc
+ \fi
+ \vfill \eject
+ \contentsalignmacro % in case @setchapternewpage odd is in effect
+ \endgroup
+ \lastnegativepageno = \pageno
+ \global\pageno = \savepageno
+}
+\let\shortcontents = \summarycontents
+
+\ifpdf
+ \pdfcatalog{/PageMode /UseOutlines}%
+\fi
+
+% 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\chapentry#1#2#3{\dochapentry{#2\labelspace#1}{#3}}
+%
+% Chapters, in the short toc.
+% See comments in \dochapentry re vbox and related settings.
+\def\shortchapentry#1#2#3{%
+ \tocentry{\shortchaplabel{#2}\labelspace #1}{\doshortpageno\bgroup#3\egroup}%
+}
+
+% Appendices, in the main contents.
+\def\appendixentry#1#2#3{%
+ \dochapentry{\appendixbox{\putwordAppendix{} #2}\labelspace#1}{#3}}
+%
+% Appendices, in the short toc.
+\let\shortappendixentry = \shortchapentry
+
+% Typeset the label for a chapter or appendix for the short contents.
+% The arg is, e.g., `Appendix A' for an appendix, or `3' for a chapter.
+% We could simplify the code here by writing out an \appendixentry
+% command in the toc file for appendices, instead of using \chapentry
+% for both, but it doesn't seem worth it.
+%
+\newdimen\shortappendixwidth
+%
+\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.)
+ \dimen0 = 1em
+ \hbox to \dimen0{#1\hss}%
+}
+
+% Unnumbered chapters.
+\def\unnumbchapentry#1#2#3{\dochapentry{#1}{#3}}
+\def\shortunnumberedentry#1#2#3{\tocentry{#1}{\doshortpageno\bgroup#3\egroup}}
+
+% Sections.
+\def\secentry#1#2#3#4{\dosecentry{#2.#3\labelspace#1}{#4}}
+\def\unnumbsecentry#1#2#3#4{\dosecentry{#1}{#4}}
+
+% Subsections.
+\def\subsecentry#1#2#3#4#5{\dosubsecentry{#2.#3.#4\labelspace#1}{#5}}
+\def\unnumbsubsecentry#1#2#3#4#5{\dosubsecentry{#1}{#5}}
+
+% And subsubsections.
+\def\subsubsecentry#1#2#3#4#5#6{%
+ \dosubsubsecentry{#2.#3.#4.#5\labelspace#1}{#6}}
+\def\unnumbsubsubsecentry#1#2#3#4#5#6{\dosubsubsecentry{#1}{#6}}
+
+% This parameter controls the indentation of the various levels.
+\newdimen\tocindent \tocindent = 3pc
+
+% 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}
+
+% Final typesetting of a toc entry; we use the same \entry macro as for
+% the index entries, but we want to suppress hyphenation here. (We
+% can't do that in the \entry macro, since index entries might consist
+% of hyphenated-identifiers-that-do-not-fit-on-a-line-and-nothing-else.)
+\def\tocentry#1#2{\begingroup
+ \vskip 0pt plus1pt % allow a little stretch for the sake of nice page breaks
+ % Do not use \turnoffactive in these arguments. Since the toc is
+ % typeset in cmr, characters such as _ would come out wrong; we
+ % have to do the usual translation tricks.
+ \entry{#1}{#2}%
+\endgroup}
+
+% 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}
+\let\subsecentryfonts = \textfonts
+\let\subsubsecentryfonts = \textfonts
+
+
+\message{environments,}
+% @foo ... @end foo.
+
+% @point{}, @result{}, @expansion{}, @print{}, @equiv{}.
+%
+% Since these characters are used in examples, it should be an even number of
+% \tt widths. Each \tt character is 1en, so two makes it 1em.
+%
+\def\point{$\star$}
+\def\result{\leavevmode\raise.15ex\hbox to 1em{\hfil$\Rightarrow$\hfil}}
+\def\expansion{\leavevmode\raise.1ex\hbox to 1em{\hfil$\mapsto$\hfil}}
+\def\print{\leavevmode\lower.1ex\hbox to 1em{\hfil$\dashv$\hfil}}
+\def\equiv{\leavevmode\lower.1ex\hbox to 1em{\hfil$\ptexequiv$\hfil}}
+
+% The @error{} command.
+% Adapted from the TeXbook's \boxit.
+%
+\newbox\errorbox
+%
+{\tentt \global\dimen0 = 3em}% Width of the box.
+\dimen2 = .55pt % Thickness of rules
+% The text. (`r' is open on the right, `e' somewhat less so on the left.)
+\setbox0 = \hbox{\kern-.75pt \tensf error\kern-1.5pt}
+%
+\global\setbox\errorbox=\hbox to \dimen0{\hfil
+ \hsize = \dimen0 \advance\hsize by -5.8pt % Space to left+right.
+ \advance\hsize by -2\dimen2 % Rules.
+ \vbox{
+ \hrule height\dimen2
+ \hbox{\vrule width\dimen2 \kern3pt % Space to left of text.
+ \vtop{\kern2.4pt \box0 \kern2.4pt}% Space above/below.
+ \kern3pt\vrule width\dimen2}% Space to right.
+ \hrule height\dimen2}
+ \hfil}
+%
+\def\error{\leavevmode\lower.7ex\copy\errorbox}
+
+% @tex ... @end tex escapes into raw Tex temporarily.
+% One exception: @ is still an escape character, so that @end tex works.
+% But \@ or @@ will get a plain tex @ character.
+
+\def\tex{\begingroup
+ \catcode `\\=0 \catcode `\{=1 \catcode `\}=2
+ \catcode `\$=3 \catcode `\&=4 \catcode `\#=6
+ \catcode `\^=7 \catcode `\_=8 \catcode `\~=\active \let~=\tie
+ \catcode `\%=14
+ \catcode `\+=\other
+ \catcode `\"=\other
+ \catcode `\==\other
+ \catcode `\|=\other
+ \catcode `\<=\other
+ \catcode `\>=\other
+ \escapechar=`\\
+ %
+ \let\b=\ptexb
+ \let\bullet=\ptexbullet
+ \let\c=\ptexc
+ \let\,=\ptexcomma
+ \let\.=\ptexdot
+ \let\dots=\ptexdots
+ \let\equiv=\ptexequiv
+ \let\!=\ptexexclam
+ \let\i=\ptexi
+ \let\indent=\ptexindent
+ \let\{=\ptexlbrace
+ \let\+=\tabalign
+ \let\}=\ptexrbrace
+ \let\/=\ptexslash
+ \let\*=\ptexstar
+ \let\t=\ptext
+ %
+ \def\endldots{\mathinner{\ldots\ldots\ldots\ldots}}%
+ \def\enddots{\relax\ifmmode\endldots\else$\mathsurround=0pt \endldots\,$\fi}%
+ \def\@{@}%
+\let\Etex=\endgroup}
+
+% Define @lisp ... @end lisp.
+% @lisp does a \begingroup 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}
+
+% 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.
+%
+{\obeyspaces %
+\gdef\sepspaces{\obeyspaces\let =\tie}}
+
+% Define \obeyedspace to be our active space, whatever it is. This is
+% for use in \parsearg.
+{\sepspaces%
+\global\let\obeyedspace= }
+
+% 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, 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 \else \penalty-50 \fi
+ \vskip\envskipamount
+ \fi
+ \fi
+}}
+
+\let\afterenvbreak = \aboveenvbreak
+
+% \nonarrowing is a flag. If "set", @lisp etc don't narrow margins.
+\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
+
+\def\cartouche{%
+\par % can't be in the midst of a paragraph.
+\begingroup
+ \lskip=\leftskip \rskip=\rightskip
+ \leftskip=0pt\rightskip=0pt %we want these *outside*.
+ \cartinner=\hsize \advance\cartinner by-\lskip
+ \advance\cartinner by-\rskip
+ \cartouter=\hsize
+ \advance\cartouter by 18.4pt % allow for 3pt kerns on either
+% side, and for 6pt waste from
+% each corner char, and rule thickness
+ \normbskip=\baselineskip \normpskip=\parskip \normlskip=\lineskip
+ % Flag to tell @lisp, etc., not to narrow margin.
+ \let\nonarrowing=\comment
+ \vbox\bgroup
+ \baselineskip=0pt\parskip=0pt\lineskip=0pt
+ \carttop
+ \hbox\bgroup
+ \hskip\lskip
+ \vrule\kern3pt
+ \vbox\bgroup
+ \hsize=\cartinner
+ \kern3pt
+ \begingroup
+ \baselineskip=\normbskip
+ \lineskip=\normlskip
+ \parskip=\normpskip
+ \vskip -\parskip
+\def\Ecartouche{%
+ \endgroup
+ \kern3pt
+ \egroup
+ \kern3pt\vrule
+ \hskip\rskip
+ \egroup
+ \cartbot
+ \egroup
+\endgroup
+}}
+
+
+% This macro is called at the beginning of all the @example variants,
+% inside a group.
+\def\nonfillstart{%
+ \aboveenvbreak
+ \inENV % This group ends at the end of the body
+ \hfuzz = 12pt % Don't be fussy
+ \sepspaces % Make spaces be word-separators rather than space tokens.
+ \let\par = \lisppar % don't ignore blank lines
+ \obeylines % each line of input is a line of output
+ \parskip = 0pt
+ \parindent = 0pt
+ \emergencystretch = 0pt % don't try to avoid overfull boxes
+ % @cartouche defines \nonarrowing to inhibit narrowing
+ % at next level down.
+ \ifx\nonarrowing\relax
+ \advance \leftskip by \lispnarrowing
+ \exdentamount=\lispnarrowing
+ \let\exdent=\nofillexdent
+ \let\nonarrowing=\relax
+ \fi
+}
+
+% Define the \E... control sequence only if we are inside the particular
+% environment, so the error checking in \end will work.
+%
+% To end an @example-like environment, we first end the paragraph (via
+% \afterenvbreak's vertical glue), and then the group. That way we keep
+% the zero \parskip that the environments set -- \parskip glue will be
+% inserted at the beginning of the next paragraph in the document, after
+% the environment.
+%
+\def\nonfillfinish{\afterenvbreak\endgroup}
+
+% @lisp: indented, narrowed, typewriter font.
+\def\lisp{\begingroup
+ \nonfillstart
+ \let\Elisp = \nonfillfinish
+ \tt
+ \let\kbdfont = \kbdexamplefont % Allow @kbd to do something special.
+ \gobble % eat return
+}
+
+% @example: Same as @lisp.
+\def\example{\begingroup \def\Eexample{\nonfillfinish\endgroup}\lisp}
+
+% @smallexample and @smalllisp: use smaller fonts.
+% Originally contributed by Pavel@xerox.
+\def\smalllisp{\begingroup
+ \def\Esmalllisp{\nonfillfinish\endgroup}%
+ \def\Esmallexample{\nonfillfinish\endgroup}%
+ \smallexamplefonts
+ \lisp
+}
+\let\smallexample = \smalllisp
+
+
+% @display: same as @lisp except keep current font.
+%
+\def\display{\begingroup
+ \nonfillstart
+ \let\Edisplay = \nonfillfinish
+ \gobble
+}
+%
+% @smalldisplay: @display plus smaller fonts.
+%
+\def\smalldisplay{\begingroup
+ \def\Esmalldisplay{\nonfillfinish\endgroup}%
+ \smallexamplefonts \rm
+ \display
+}
+
+% @format: same as @display except don't narrow margins.
+%
+\def\format{\begingroup
+ \let\nonarrowing = t
+ \nonfillstart
+ \let\Eformat = \nonfillfinish
+ \gobble
+}
+%
+% @smallformat: @format plus smaller fonts.
+%
+\def\smallformat{\begingroup
+ \def\Esmallformat{\nonfillfinish\endgroup}%
+ \smallexamplefonts \rm
+ \format
+}
+
+% @flushleft (same as @format).
+%
+\def\flushleft{\begingroup \def\Eflushleft{\nonfillfinish\endgroup}\format}
+
+% @flushright.
+%
+\def\flushright{\begingroup
+ \let\nonarrowing = t
+ \nonfillstart
+ \let\Eflushright = \nonfillfinish
+ \advance\leftskip by 0pt plus 1fill
+ \gobble
+}
+
+
+% @quotation does normal linebreaking (hence we can't use \nonfillstart)
+% and narrows the margins.
+%
+\def\quotation{%
+ \begingroup\inENV %This group ends at the end of the @quotation body
+ {\parskip=0pt \aboveenvbreak}% because \aboveenvbreak inserts \parskip
+ \parindent=0pt
+ % We have retained a nonzero parskip for the environment, since we're
+ % doing normal filling. So to avoid extra space below the environment...
+ \def\Equotation{\parskip = 0pt \nonfillfinish}%
+ %
+ % @cartouche defines \nonarrowing to inhibit narrowing at next level down.
+ \ifx\nonarrowing\relax
+ \advance\leftskip by \lispnarrowing
+ \advance\rightskip by \lispnarrowing
+ \exdentamount = \lispnarrowing
+ \let\nonarrowing = \relax
+ \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=12}\dospecials}
+%
+% [Knuth] pp. 380,381,391
+% Disable Spanish ligatures ?` and !` of \tt font
+\begingroup
+ \catcode`\`=\active\gdef`{\relax\lq}
+\endgroup
+%
+% Setup for the @verb command.
+%
+% Eight spaces for a tab
+\begingroup
+ \catcode`\^^I=\active
+ \gdef\tabeightspaces{\catcode`\^^I=\active\def^^I{\ \ \ \ \ \ \ \ }}
+\endgroup
+%
+\def\setupverb{%
+ \tt % easiest (and conventionally used) font for verbatim
+ \def\par{\leavevmode\endgraf}%
+ \catcode`\`=\active
+ \tabeightspaces
+ % Respect line breaks,
+ % print special symbols as themselves, and
+ % make each space count
+ % must do in this order:
+ \obeylines \uncatcodespecials \sepspaces
+}
+
+% Setup for the @verbatim environment
+%
+% Real tab expansion
+\newdimen\tabw \setbox0=\hbox{\tt\space} \tabw=8\wd0 % tab amount
+%
+\def\starttabbox{\setbox0=\hbox\bgroup}
+\begingroup
+ \catcode`\^^I=\active
+ \gdef\tabexpand{%
+ \catcode`\^^I=\active
+ \def^^I{\leavevmode\egroup
+ \dimen0=\wd0 % the width so far, or since the previous tab
+ \divide\dimen0 by\tabw
+ \multiply\dimen0 by\tabw % compute previous multiple of \tabw
+ \advance\dimen0 by\tabw % advance to next multiple of \tabw
+ \wd0=\dimen0 \box0 \starttabbox
+ }%
+ }
+\endgroup
+\def\setupverbatim{%
+ % Easiest (and conventionally used) font for verbatim
+ \tt
+ \def\par{\leavevmode\egroup\box0\endgraf}%
+ \catcode`\`=\active
+ \tabexpand
+ % Respect line breaks,
+ % print special symbols as themselves, and
+ % make each space count
+ % must do in this order:
+ \obeylines \uncatcodespecials \sepspaces
+ \everypar{\starttabbox}%
+}
+
+% Do the @verb magic: verbatim text is quoted by unique
+% delimiter characters. Before first delimiter expect a
+% right brace, after last delimiter expect closing brace:
+%
+% \def\doverb'{'<char>#1<char>'}'{#1}
+%
+% [Knuth] p. 382; only eat outer {}
+\begingroup
+ \catcode`[=1\catcode`]=2\catcode`\{=12\catcode`\}=12
+ \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]
+%% Include LaTeX hack for completeness -- never know
+%% \begingroup
+%% \catcode`|=0 \catcode`[=1
+%% \catcode`]=2\catcode`\{=12\catcode`\}=12\catcode`\ =\active
+%% \catcode`\\=12|gdef|doverbatim#1@end verbatim[
+%% #1|endgroup|def|Everbatim[]|end[verbatim]]
+%% |endgroup
+%
+\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.
+ \gdef\doverbatim#1^^M#2@end verbatim{#2\end{verbatim}}%
+\endgroup
+%
+\def\verbatim{%
+ \def\Everbatim{\nonfillfinish\endgroup}%
+ \begingroup
+ \nonfillstart
+ \advance\leftskip by -\defbodyindent
+ \begingroup\setupverbatim\doverbatim
+}
+
+% @verbatiminclude FILE - insert text of file in verbatim environment.
+%
+% Allow normal characters that we make active in the argument (a file name).
+\def\verbatiminclude{%
+ \begingroup
+ \catcode`\\=\other
+ \catcode`~=\other
+ \catcode`^=\other
+ \catcode`_=\other
+ \catcode`|=\other
+ \catcode`<=\other
+ \catcode`>=\other
+ \catcode`+=\other
+ \parsearg\doverbatiminclude
+}
+\def\setupverbatiminclude{%
+ \begingroup
+ \nonfillstart
+ \advance\leftskip by -\defbodyindent
+ \begingroup\setupverbatim
+}
+%
+\def\doverbatiminclude#1{%
+ % Restore active chars for included file.
+ \endgroup
+ \begingroup
+ \let\value=\expandablevalue
+ \def\thisfile{#1}%
+ \expandafter\expandafter\setupverbatiminclude\input\thisfile
+ \endgroup
+ \nonfillfinish
+ \endgroup
+}
+
+% @copying ... @end copying.
+% Save the text away for @insertcopying later. Many commands won't be
+% allowed in this context, but that's ok.
+%
+% 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{\begingroup
+ % Define a command to swallow text until we reach `@end copying'.
+ % \ is the escape char in this texinfo.tex file, so it is the
+ % delimiter for the command; @ will be the escape char when we read
+ % it, but that doesn't matter.
+ \long\def\docopying##1\end copying{\gdef\copyingtext{##1}\enddocopying}%
+ %
+ % We must preserve ^^M's in the input file; see \insertcopying below.
+ \catcode`\^^M = \active
+ \docopying
+}
+
+% What we do to finish off the copying text.
+%
+\def\enddocopying{\endgroup\ignorespaces}
+
+% @insertcopying. Here we must play games with ^^M's. On the one hand,
+% we need them to delimit commands such as `@end quotation', so they
+% must be active. On the other hand, we certainly don't want every
+% end-of-line to be a \par, as would happen with the normal active
+% definition of ^^M. On the third hand, two ^^M's in a row should still
+% generate a \par.
+%
+% Our approach is to make ^^M insert a space and a penalty1 normally;
+% then it can also check if \lastpenalty=1. If it does, then manually
+% do \par.
+%
+% This messes up the normal definitions of @c[omment], so we redefine
+% it. Similarly for @ignore. (These commands are used in the gcc
+% manual for man page generation.)
+%
+% Seems pretty fragile, most line-oriented commands will presumably
+% fail, but for the limited use of getting the copying text (which
+% should be quite simple) inserted, we can hope it's ok.
+%
+{\catcode`\^^M=\active %
+\gdef\insertcopying{\begingroup %
+ \parindent = 0pt % looks wrong on title page
+ \def^^M{%
+ \ifnum \lastpenalty=1 %
+ \par %
+ \else %
+ \space \penalty 1 %
+ \fi %
+ }%
+ %
+ % Fix @c[omment] for catcode 13 ^^M's.
+ \def\c##1^^M{\ignorespaces}%
+ \let\comment = \c %
+ %
+ % Don't bother jumping through all the hoops that \doignore does, it
+ % would be very hard since the catcodes are already set.
+ \long\def\ignore##1\end ignore{\ignorespaces}%
+ %
+ \copyingtext %
+\endgroup}%
+}
+
+\message{defuns,}
+% @defun etc.
+
+% Allow user to change definition object font (\df) internally
+\def\setdeffont#1 {\csname DEF#1\endcsname}
+
+\newskip\defbodyindent \defbodyindent=.4in
+\newskip\defargsindent \defargsindent=50pt
+\newskip\deflastargmargin \deflastargmargin=18pt
+
+\newcount\parencount
+
+% 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 = )
+
+{\activeparens % Now, smart parens don't turn on until &foo (see \amprm)
+
+% 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.
+\global\let(=\lparen \global\let)=\rparen
+\global\let[=\lbrack \global\let]=\rbrack
+
+\gdef\functionparens{\boldbrax\let&=\amprm\parencount=0 }
+\gdef\boldbrax{\let(=\opnr\let)=\clnr\let[=\lbrb\let]=\rbrb}
+% This is used to turn on special parens
+% but make & act ordinary (given that it's active).
+\gdef\boldbraxnoamp{\let(=\opnr\let)=\clnr\let[=\lbrb\let]=\rbrb\let&=\ampnr}
+
+% Definitions of (, ) and & used in args for functions.
+% This is the definition of ( outside of all parentheses.
+\gdef\oprm#1 {{\rm\char`\(}#1 \bf \let(=\opnested
+ \global\advance\parencount by 1
+}
+%
+% This is the definition of ( when already inside a level of parens.
+\gdef\opnested{\char`\(\global\advance\parencount by 1 }
+%
+\gdef\clrm{% Print a paren in roman if it is taking us back to depth of 0.
+ % also in that case restore the outer-level definition of (.
+ \ifnum \parencount=1 {\rm \char `\)}\sl \let(=\oprm \else \char `\) \fi
+ \global\advance \parencount by -1 }
+% If we encounter &foo, then turn on ()-hacking afterwards
+\gdef\amprm#1 {{\rm\&#1}\let(=\oprm \let)=\clrm\ }
+%
+\gdef\normalparens{\boldbrax\let&=\ampnr}
+} % End of definition inside \activeparens
+%% These parens (in \boldbrax) actually are a little bolder than the
+%% contained text. This is especially needed for [ and ]
+\def\opnr{{\sf\char`\(}\global\advance\parencount by 1 }
+\def\clnr{{\sf\char`\)}\global\advance\parencount by -1 }
+\let\ampnr = \&
+\def\lbrb{{\bf\char`\[}}
+\def\rbrb{{\bf\char`\]}}
+
+% Active &'s sneak into the index arguments, so make sure it's defined.
+{
+ \catcode`& = \active
+ \global\let& = \ampnr
+}
+
+% \defname, which formats the name of the @def (not the args).
+% #1 is the function name.
+% #2 is the type of definition, such as "Function".
+%
+\def\defname#1#2{%
+ % How we'll output 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.
+ \ifempty{#2}%
+ \def\defnametype{}%
+ \else
+ \def\defnametype{[\rm #2]}%
+ \fi
+ %
+ % Get the values of \leftskip and \rightskip as they were outside the @def...
+ \dimen2=\leftskip
+ \advance\dimen2 by -\defbodyindent
+ %
+ % Figure out values for the paragraph shape.
+ \setbox0=\hbox{\hskip \deflastargmargin{\defnametype}}%
+ \dimen0=\hsize \advance \dimen0 by -\wd0 % compute size for first line
+ \dimen1=\hsize \advance \dimen1 by -\defargsindent % size for continuations
+ \parshape 2 0in \dimen0 \defargsindent \dimen1
+ %
+ % Output arg 2 ("Function" or some such) but stuck inside a box of
+ % width 0 so it does not interfere with linebreaking.
+ \noindent
+ %
+ {% Adjust \hsize to exclude the ambient margins,
+ % so that \rightline will obey them.
+ \advance \hsize by -\dimen2
+ \dimen3 = 0pt % was -1.25pc
+ \rlap{\rightline{\defnametype\kern\dimen3}}%
+ }%
+ %
+ % Allow all lines to be underfull without complaint:
+ \tolerance=10000 \hbadness=10000
+ \advance\leftskip by -\defbodyindent
+ \exdentamount=\defbodyindent
+ {\df #1}\enskip % output function name
+ % \defunargs will be called next to output the arguments, if any.
+}
+
+% Common pieces to start any @def...
+% #1 is the \E... control sequence to end the definition (which we define).
+% #2 is the \...x control sequence (which our caller defines).
+% #3 is the control sequence to process the header, such as \defunheader.
+%
+\def\parsebodycommon#1#2#3{%
+ \begingroup\inENV
+ % 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 want to allow a
+ % break after all. Check for penalty 10002 (inserted by
+ % \defargscommonending) instead of 10000, since the sectioning
+ % commands insert a \penalty10000, and we don't want to allow a break
+ % between a section heading and a defun.
+ \ifnum\lastpenalty=10002 \penalty0 \fi
+ \medbreak
+ %
+ % Define the \E... end token that this defining construct specifies
+ % so that it will exit this group.
+ \def#1{\endgraf\endgroup\medbreak}%
+ %
+ \parindent=0in
+ \advance\leftskip by \defbodyindent
+ \exdentamount=\defbodyindent
+}
+
+% Common part of the \...x definitions.
+%
+\def\defxbodycommon{%
+ % As with \parsebodycommon above, allow line break if we have multiple
+ % x headers in a row. It's not a great place, though.
+ \ifnum\lastpenalty=10000 \penalty1000 \fi
+ %
+ \begingroup\obeylines
+}
+
+% Process body of @defun, @deffn, @defmac, etc.
+%
+\def\defparsebody#1#2#3{%
+ \parsebodycommon{#1}{#2}{#3}%
+ \def#2{\defxbodycommon \activeparens \spacesplit#3}%
+ \catcode\equalChar=\active
+ \begingroup\obeylines\activeparens
+ \spacesplit#3%
+}
+
+% #1, #2, #3 are the common arguments (see \parsebodycommon above).
+% #4, delimited by the space, is the class name.
+%
+\def\defmethparsebody#1#2#3#4 {%
+ \parsebodycommon{#1}{#2}{#3}%
+ \def#2##1 {\defxbodycommon \activeparens \spacesplit{#3{##1}}}%
+ \begingroup\obeylines\activeparens
+ % The \empty here prevents misinterpretation of a construct such as
+ % @deffn {whatever} {Enharmonic comma}
+ % See comments at \deftpparsebody, although in our case we don't have
+ % to remove the \empty afterwards, since it is empty.
+ \spacesplit{#3{#4}}\empty
+}
+
+% Used for @deftypemethod and @deftypeivar.
+% #1, #2, #3 are the common arguments (see \defparsebody).
+% #4, delimited by a space, is the class name.
+% #5 is the method's return type.
+%
+\def\deftypemethparsebody#1#2#3#4 #5 {%
+ \parsebodycommon{#1}{#2}{#3}%
+ \def#2##1 ##2 {\defxbodycommon \activeparens \spacesplit{#3{##1}{##2}}}%
+ \begingroup\obeylines\activeparens
+ \spacesplit{#3{#4}{#5}}%
+}
+
+% Used for @deftypeop. The change from \deftypemethparsebody is an
+% extra argument at the beginning which is the `category', instead of it
+% being the hardwired string `Method' or `Instance Variable'. We have
+% to account for this both in the \...x definition and in parsing the
+% input at hand. Thus also need a control sequence (passed as #5) for
+% the \E... definition to assign the category name to.
+%
+\def\deftypeopparsebody#1#2#3#4#5 #6 {%
+ \parsebodycommon{#1}{#2}{#3}%
+ \def#2##1 ##2 ##3 {\def#4{##1}%
+ \defxbodycommon \activeparens \spacesplit{#3{##2}{##3}}}%
+ \begingroup\obeylines\activeparens
+ \spacesplit{#3{#5}{#6}}%
+}
+
+% For @defop.
+\def\defopparsebody #1#2#3#4#5 {%
+ \parsebodycommon{#1}{#2}{#3}%
+ \def#2##1 ##2 {\def#4{##1}%
+ \defxbodycommon \activeparens \spacesplit{#3{##2}}}%
+ \begingroup\obeylines\activeparens
+ \spacesplit{#3{#5}}%
+}
+
+% These parsing functions are similar to the preceding ones
+% except that they do not make parens into active characters.
+% These are used for "variables" since they have no arguments.
+%
+\def\defvarparsebody #1#2#3{%
+ \parsebodycommon{#1}{#2}{#3}%
+ \def#2{\defxbodycommon \spacesplit#3}%
+ \catcode\equalChar=\active
+ \begingroup\obeylines
+ \spacesplit#3%
+}
+
+% @defopvar.
+\def\defopvarparsebody #1#2#3#4#5 {%
+ \parsebodycommon{#1}{#2}{#3}%
+ \def#2##1 ##2 {\def#4{##1}%
+ \defxbodycommon \spacesplit{#3{##2}}}%
+ \begingroup\obeylines
+ \spacesplit{#3{#5}}%
+}
+
+\def\defvrparsebody#1#2#3#4 {%
+ \parsebodycommon{#1}{#2}{#3}%
+ \def#2##1 {\defxbodycommon \spacesplit{#3{##1}}}%
+ \begingroup\obeylines
+ \spacesplit{#3{#4}}%
+}
+
+% This loses on `@deftp {Data Type} {struct termios}' -- it thinks the
+% type is just `struct', because we lose the braces in `{struct
+% termios}' when \spacesplit reads its undelimited argument. Sigh.
+% \let\deftpparsebody=\defvrparsebody
+%
+% So, to get around this, we put \empty in with the type name. That
+% way, TeX won't find exactly `{...}' as an undelimited argument, and
+% won't strip off the braces.
+%
+\def\deftpparsebody #1#2#3#4 {%
+ \parsebodycommon{#1}{#2}{#3}%
+ \def#2##1 {\defxbodycommon \spacesplit{#3{##1}}}%
+ \begingroup\obeylines
+ \spacesplit{\parsetpheaderline{#3{#4}}}\empty
+}
+
+% Fine, but then we have to eventually remove the \empty *and* the
+% braces (if any). That's what this does.
+%
+\def\removeemptybraces\empty#1\relax{#1}
+
+% After \spacesplit has done its work, this is called -- #1 is the final
+% thing to call, #2 the type name (which starts with \empty), and #3
+% (which might be empty) the arguments.
+%
+\def\parsetpheaderline#1#2#3{%
+ #1{\removeemptybraces#2\relax}{#3}%
+}%
+
+% Split up #2 (the rest of the input line) at the first space token.
+% call #1 with two arguments:
+% the first is all of #2 before the space token,
+% the second is all of #2 after that space token.
+% If #2 contains no space token, all of it is passed as the first arg
+% and the second is passed as empty.
+%
+{\obeylines %
+ \gdef\spacesplit#1#2^^M{\endgroup\spacesplitx{#1}#2 \relax\spacesplitx}%
+ \long\gdef\spacesplitx#1#2 #3#4\spacesplitx{%
+ \ifx\relax #3%
+ #1{#2}{}%
+ \else %
+ #1{#2}{#3#4}%
+ \fi}%
+}
+
+% Define @defun.
+
+% This is called to end the arguments processing for all the @def... commands.
+%
+\def\defargscommonending{%
+ \interlinepenalty = 10000
+ \advance\rightskip by 0pt plus 1fil
+ \endgraf
+ \nobreak\vskip -\parskip
+ \penalty 10002 % signal to \parsebodycommon.
+}
+
+% This expands the args and terminates the paragraph they comprise.
+%
+\def\defunargs#1{\functionparens \sl
+% Expand, preventing hyphenation at `-' chars.
+% Note that groups don't affect changes in \hyphenchar.
+% Set the font temporarily and use \font in case \setfont made \tensl a macro.
+{\tensl\hyphenchar\font=0}%
+#1%
+{\tensl\hyphenchar\font=45}%
+\ifnum\parencount=0 \else \errmessage{Unbalanced parentheses in @def}\fi%
+ \defargscommonending
+}
+
+\def\deftypefunargs #1{%
+% Expand, preventing hyphenation at `-' chars.
+% Note that groups don't affect changes in \hyphenchar.
+% Use \boldbraxnoamp, not \functionparens, so that & is not special.
+\boldbraxnoamp
+\tclose{#1}% avoid \code because of side effects on active chars
+ \defargscommonending
+}
+
+% Do complete processing of one @defun or @defunx line already parsed.
+
+% @deffn Command forward-char nchars
+
+\def\deffn{\defmethparsebody\Edeffn\deffnx\deffnheader}
+
+\def\deffnheader #1#2#3{\doind {fn}{\code{#2}}%
+\begingroup\defname {#2}{#1}\defunargs{#3}\endgroup %
+\catcode\equalChar=\other % Turn off change made in \defparsebody
+}
+
+% @defun == @deffn Function
+
+\def\defun{\defparsebody\Edefun\defunx\defunheader}
+
+\def\defunheader #1#2{\doind {fn}{\code{#1}}% Make entry in function index
+\begingroup\defname {#1}{\putwordDeffunc}%
+\defunargs {#2}\endgroup %
+\catcode\equalChar=\other % Turn off change made in \defparsebody
+}
+
+% @deftypefun int foobar (int @var{foo}, float @var{bar})
+
+\def\deftypefun{\defparsebody\Edeftypefun\deftypefunx\deftypefunheader}
+
+% #1 is the data type. #2 is the name and args.
+\def\deftypefunheader #1#2{\deftypefunheaderx{#1}#2 \relax}
+% #1 is the data type, #2 the name, #3 the args.
+\def\deftypefunheaderx #1#2 #3\relax{%
+\doind {fn}{\code{#2}}% Make entry in function index
+\begingroup\defname {\defheaderxcond#1\relax$.$#2}{\putwordDeftypefun}%
+\deftypefunargs {#3}\endgroup %
+\catcode\equalChar=\other % Turn off change made in \defparsebody
+}
+
+% @deftypefn {Library Function} int foobar (int @var{foo}, float @var{bar})
+
+\def\deftypefn{\defmethparsebody\Edeftypefn\deftypefnx\deftypefnheader}
+
+% \defheaderxcond#1\relax$.$
+% puts #1 in @code, followed by a space, but does nothing if #1 is null.
+\def\defheaderxcond#1#2$.${\ifx#1\relax\else\code{#1#2} \fi}
+
+% #1 is the classification. #2 is the data type. #3 is the name and args.
+\def\deftypefnheader #1#2#3{\deftypefnheaderx{#1}{#2}#3 \relax}
+% #1 is the classification, #2 the data type, #3 the name, #4 the args.
+\def\deftypefnheaderx #1#2#3 #4\relax{%
+\doind {fn}{\code{#3}}% Make entry in function index
+\begingroup
+\normalparens % notably, turn off `&' magic, which prevents
+% at least some C++ text from working
+\defname {\defheaderxcond#2\relax$.$#3}{#1}%
+\deftypefunargs {#4}\endgroup %
+\catcode\equalChar=\other % Turn off change made in \defparsebody
+}
+
+% @defmac == @deffn Macro
+
+\def\defmac{\defparsebody\Edefmac\defmacx\defmacheader}
+
+\def\defmacheader #1#2{\doind {fn}{\code{#1}}% Make entry in function index
+\begingroup\defname {#1}{\putwordDefmac}%
+\defunargs {#2}\endgroup %
+\catcode\equalChar=\other % Turn off change made in \defparsebody
+}
+
+% @defspec == @deffn Special Form
+
+\def\defspec{\defparsebody\Edefspec\defspecx\defspecheader}
+
+\def\defspecheader #1#2{\doind {fn}{\code{#1}}% Make entry in function index
+\begingroup\defname {#1}{\putwordDefspec}%
+\defunargs {#2}\endgroup %
+\catcode\equalChar=\other % Turn off change made in \defparsebody
+}
+
+% @defop CATEGORY CLASS OPERATION ARG...
+%
+\def\defop #1 {\def\defoptype{#1}%
+\defopparsebody\Edefop\defopx\defopheader\defoptype}
+%
+\def\defopheader#1#2#3{%
+ \dosubind{fn}{\code{#2}}{\putwordon\ \code{#1}}% function index entry
+ \begingroup
+ \defname{#2}{\defoptype\ \putwordon\ #1}%
+ \defunargs{#3}%
+ \endgroup
+}
+
+% @deftypeop CATEGORY CLASS TYPE OPERATION ARG...
+%
+\def\deftypeop #1 {\def\deftypeopcategory{#1}%
+ \deftypeopparsebody\Edeftypeop\deftypeopx\deftypeopheader
+ \deftypeopcategory}
+%
+% #1 is the class name, #2 the data type, #3 the operation name, #4 the args.
+\def\deftypeopheader#1#2#3#4{%
+ \dosubind{fn}{\code{#3}}{\putwordon\ \code{#1}}% entry in function index
+ \begingroup
+ \defname{\defheaderxcond#2\relax$.$#3}
+ {\deftypeopcategory\ \putwordon\ \code{#1}}%
+ \deftypefunargs{#4}%
+ \endgroup
+}
+
+% @deftypemethod CLASS TYPE METHOD ARG...
+%
+\def\deftypemethod{%
+ \deftypemethparsebody\Edeftypemethod\deftypemethodx\deftypemethodheader}
+%
+% #1 is the class name, #2 the data type, #3 the method name, #4 the args.
+\def\deftypemethodheader#1#2#3#4{%
+ \dosubind{fn}{\code{#3}}{\putwordon\ \code{#1}}% entry in function index
+ \begingroup
+ \defname{\defheaderxcond#2\relax$.$#3}{\putwordMethodon\ \code{#1}}%
+ \deftypefunargs{#4}%
+ \endgroup
+}
+
+% @deftypeivar CLASS TYPE VARNAME
+%
+\def\deftypeivar{%
+ \deftypemethparsebody\Edeftypeivar\deftypeivarx\deftypeivarheader}
+%
+% #1 is the class name, #2 the data type, #3 the variable name.
+\def\deftypeivarheader#1#2#3{%
+ \dosubind{vr}{\code{#3}}{\putwordof\ \code{#1}}% entry in variable index
+ \begingroup
+ \defname{\defheaderxcond#2\relax$.$#3}
+ {\putwordInstanceVariableof\ \code{#1}}%
+ \defvarargs{#3}%
+ \endgroup
+}
+
+% @defmethod == @defop Method
+%
+\def\defmethod{\defmethparsebody\Edefmethod\defmethodx\defmethodheader}
+%
+% #1 is the class name, #2 the method name, #3 the args.
+\def\defmethodheader#1#2#3{%
+ \dosubind{fn}{\code{#2}}{\putwordon\ \code{#1}}% entry in function index
+ \begingroup
+ \defname{#2}{\putwordMethodon\ \code{#1}}%
+ \defunargs{#3}%
+ \endgroup
+}
+
+% @defcv {Class Option} foo-class foo-flag
+
+\def\defcv #1 {\def\defcvtype{#1}%
+\defopvarparsebody\Edefcv\defcvx\defcvarheader\defcvtype}
+
+\def\defcvarheader #1#2#3{%
+ \dosubind{vr}{\code{#2}}{\putwordof\ \code{#1}}% variable index entry
+ \begingroup
+ \defname{#2}{\defcvtype\ \putwordof\ #1}%
+ \defvarargs{#3}%
+ \endgroup
+}
+
+% @defivar CLASS VARNAME == @defcv {Instance Variable} CLASS VARNAME
+%
+\def\defivar{\defvrparsebody\Edefivar\defivarx\defivarheader}
+%
+\def\defivarheader#1#2#3{%
+ \dosubind{vr}{\code{#2}}{\putwordof\ \code{#1}}% entry in var index
+ \begingroup
+ \defname{#2}{\putwordInstanceVariableof\ #1}%
+ \defvarargs{#3}%
+ \endgroup
+}
+
+% @defvar
+% First, define the processing that is wanted for arguments of @defvar.
+% This is actually simple: just print them in roman.
+% This must expand the args and terminate the paragraph they make up
+\def\defvarargs #1{\normalparens #1%
+ \defargscommonending
+}
+
+% @defvr Counter foo-count
+
+\def\defvr{\defvrparsebody\Edefvr\defvrx\defvrheader}
+
+\def\defvrheader #1#2#3{\doind {vr}{\code{#2}}%
+\begingroup\defname {#2}{#1}\defvarargs{#3}\endgroup}
+
+% @defvar == @defvr Variable
+
+\def\defvar{\defvarparsebody\Edefvar\defvarx\defvarheader}
+
+\def\defvarheader #1#2{\doind {vr}{\code{#1}}% Make entry in var index
+\begingroup\defname {#1}{\putwordDefvar}%
+\defvarargs {#2}\endgroup %
+}
+
+% @defopt == @defvr {User Option}
+
+\def\defopt{\defvarparsebody\Edefopt\defoptx\defoptheader}
+
+\def\defoptheader #1#2{\doind {vr}{\code{#1}}% Make entry in var index
+\begingroup\defname {#1}{\putwordDefopt}%
+\defvarargs {#2}\endgroup %
+}
+
+% @deftypevar int foobar
+
+\def\deftypevar{\defvarparsebody\Edeftypevar\deftypevarx\deftypevarheader}
+
+% #1 is the data type. #2 is the name, perhaps followed by text that
+% is actually part of the data type, which should not be put into the index.
+\def\deftypevarheader #1#2{%
+\dovarind#2 \relax% Make entry in variables index
+\begingroup\defname {\defheaderxcond#1\relax$.$#2}{\putwordDeftypevar}%
+ \defargscommonending
+\endgroup}
+\def\dovarind#1 #2\relax{\doind{vr}{\code{#1}}}
+
+% @deftypevr {Global Flag} int enable
+
+\def\deftypevr{\defvrparsebody\Edeftypevr\deftypevrx\deftypevrheader}
+
+\def\deftypevrheader #1#2#3{\dovarind#3 \relax%
+\begingroup\defname {\defheaderxcond#2\relax$.$#3}{#1}
+ \defargscommonending
+\endgroup}
+
+% Now define @deftp
+% Args are printed in bold, a slight difference from @defvar.
+
+\def\deftpargs #1{\bf \defvarargs{#1}}
+
+% @deftp Class window height width ...
+
+\def\deftp{\deftpparsebody\Edeftp\deftpx\deftpheader}
+
+\def\deftpheader #1#2#3{\doind {tp}{\code{#2}}%
+\begingroup\defname {#2}{#1}\deftpargs{#3}\endgroup}
+
+% These definitions are used if you use @defunx (etc.)
+% anywhere other than immediately after a @defun or @defunx.
+%
+\def\defcvx#1 {\errmessage{@defcvx in invalid context}}
+\def\deffnx#1 {\errmessage{@deffnx in invalid context}}
+\def\defivarx#1 {\errmessage{@defivarx in invalid context}}
+\def\defmacx#1 {\errmessage{@defmacx in invalid context}}
+\def\defmethodx#1 {\errmessage{@defmethodx in invalid context}}
+\def\defoptx #1 {\errmessage{@defoptx in invalid context}}
+\def\defopx#1 {\errmessage{@defopx in invalid context}}
+\def\defspecx#1 {\errmessage{@defspecx in invalid context}}
+\def\deftpx#1 {\errmessage{@deftpx in invalid context}}
+\def\deftypefnx#1 {\errmessage{@deftypefnx in invalid context}}
+\def\deftypefunx#1 {\errmessage{@deftypefunx in invalid context}}
+\def\deftypeivarx#1 {\errmessage{@deftypeivarx in invalid context}}
+\def\deftypemethodx#1 {\errmessage{@deftypemethodx in invalid context}}
+\def\deftypeopx#1 {\errmessage{@deftypeopx in invalid context}}
+\def\deftypevarx#1 {\errmessage{@deftypevarx in invalid context}}
+\def\deftypevrx#1 {\errmessage{@deftypevrx in invalid context}}
+\def\defunx#1 {\errmessage{@defunx in invalid context}}
+\def\defvarx#1 {\errmessage{@defvarx in invalid context}}
+\def\defvrx#1 {\errmessage{@defvrx in invalid context}}
+
+
+\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\scanmacro#1{%
+ \begingroup \newlinechar`\^^M
+ % Undo catcode changes of \startcontents and \doprintindex
+ \catcode`\@=0 \catcode`\\=\other \escapechar=`\@
+ % Append \endinput to make sure that TeX does not see the ending newline.
+ \toks0={#1\endinput}%
+ \immediate\openout\macscribble=\jobname.tmp
+ \immediate\write\macscribble{\the\toks0}%
+ \immediate\closeout\macscribble
+ \let\xeatspaces\eatspaces
+ \input \jobname.tmp
+ \endgroup
+}
+\else
+\def\scanmacro#1{%
+\begingroup \newlinechar`\^^M
+% Undo catcode changes of \startcontents and \doprintindex
+\catcode`\@=0 \catcode`\\=\other \escapechar=`\@
+\let\xeatspaces\eatspaces\scantokens{#1\endinput}\endgroup}
+\fi
+
+\newcount\paramno % Count of parameters
+\newtoks\macname % Macro name
+\newif\ifrecursive % Is it recursive?
+\def\macrolist{} % List of all defined macros in the form
+ % \do\macro1\do\macro2...
+
+% Utility routines.
+% Thisdoes \let #1 = #2, except with \csnames.
+\def\cslet#1#2{%
+\expandafter\expandafter
+\expandafter\let
+\expandafter\expandafter
+\csname#1\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\macrobodyctxt{%
+ \catcode`\~=\other
+ \catcode`\^=\other
+ \catcode`\_=\other
+ \catcode`\|=\other
+ \catcode`\<=\other
+ \catcode`\>=\other
+ \catcode`\+=\other
+ \catcode`\{=\other
+ \catcode`\}=\other
+ \catcode`\@=\other
+ \catcode`\^^M=\other
+ \usembodybackslash}
+
+\def\macroargctxt{%
+ \catcode`\~=\other
+ \catcode`\^=\other
+ \catcode`\_=\other
+ \catcode`\|=\other
+ \catcode`\<=\other
+ \catcode`\>=\other
+ \catcode`\+=\other
+ \catcode`\@=\other
+ \catcode`\\=\other}
+
+% \mbodybackslash is the definition of \ in @macro bodies.
+% It maps \foo\ => \csname macarg.foo\endcsname => #N
+% where N is the macro parameter number.
+% We define \csname macarg.\endcsname to be \realbackslash, so
+% \\ in macro replacement text gets you a backslash.
+
+{\catcode`@=0 @catcode`@\=@active
+ @gdef@usembodybackslash{@let\=@mbodybackslash}
+ @gdef@mbodybackslash#1\{@csname macarg.#1@endcsname}
+}
+\expandafter\def\csname macarg.\endcsname{\realbackslash}
+
+\def\macro{\recursivefalse\parsearg\macroxxx}
+\def\rmacro{\recursivetrue\parsearg\macroxxx}
+
+\def\macroxxx#1{%
+ \getargs{#1}% now \macname is the macname and \argl the arglist
+ \ifx\argl\empty % no arguments
+ \paramno=0%
+ \else
+ \expandafter\parsemargdef \argl;%
+ \fi
+ \if1\csname ismacro.\the\macname\endcsname
+ \message{Warning: redefining \the\macname}%
+ \else
+ \expandafter\ifx\csname \the\macname\endcsname \relax
+ \else \errmessage{Macro name \the\macname\space already defined}\fi
+ \global\cslet{macsave.\the\macname}{\the\macname}%
+ \global\expandafter\let\csname ismacro.\the\macname\endcsname=1%
+ % Add the macroname to \macrolist
+ \toks0 = \expandafter{\macrolist\do}%
+ \xdef\macrolist{\the\toks0
+ \expandafter\noexpand\csname\the\macname\endcsname}%
+ \fi
+ \begingroup \macrobodyctxt
+ \ifrecursive \expandafter\parsermacbody
+ \else \expandafter\parsemacbody
+ \fi}
+
+\def\unmacro{\parsearg\dounmacro}
+\def\dounmacro#1{%
+ \if1\csname ismacro.#1\endcsname
+ \global\cslet{#1}{macsave.#1}%
+ \global\expandafter\let \csname ismacro.#1\endcsname=0%
+ % Remove the macro name from \macrolist:
+ \begingroup
+ \expandafter\let\csname#1\endcsname \relax
+ \let\do\unmacrodo
+ \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\do \noexpand #1%
+ \fi
+}
+
+% This makes use of the obscure feature that if the last token of a
+% <parameter list> is #, then the preceding argument is delimited by
+% an opening brace, and that opening brace is not consumed.
+\def\getargs#1{\getargsxxx#1{}}
+\def\getargsxxx#1#{\getmacname #1 \relax\getmacargs}
+\def\getmacname #1 #2\relax{\macname={#1}}
+\def\getmacargs#1{\def\argl{#1}}
+
+% Parse the optional {params} list. Set up \paramno and \paramlist
+% so \defmacro knows what to do. Define \macarg.blah for each blah
+% in the params list, to be ##N where N is the position in that list.
+% That gets used by \mbodybackslash (above).
+
+% We need to get `macro parameter char #' into several definitions.
+% The technique used is stolen from LaTeX: let \hash be something
+% unexpandable, insert that wherever you need a #, and then redefine
+% it to # just before using the token list produced.
+%
+% The same technique is used to protect \eatspaces till just before
+% the macro is used.
+
+\def\parsemargdef#1;{\paramno=0\def\paramlist{}%
+ \let\hash\relax\let\xeatspaces\relax\parsemargdefxxx#1,;,}
+\def\parsemargdefxxx#1,{%
+ \if#1;\let\next=\relax
+ \else \let\next=\parsemargdefxxx
+ \advance\paramno by 1%
+ \expandafter\edef\csname macarg.\eatspaces{#1}\endcsname
+ {\xeatspaces{\hash\the\paramno}}%
+ \edef\paramlist{\paramlist\hash\the\paramno,}%
+ \fi\next}
+
+% These two commands read recursive and nonrecursive macro bodies.
+% (They're different since rec and nonrec macros end differently.)
+
+\long\def\parsemacbody#1@end macro%
+{\xdef\temp{\eatcr{#1}}\endgroup\defmacro}%
+\long\def\parsermacbody#1@end rmacro%
+{\xdef\temp{\eatcr{#1}}\endgroup\defmacro}%
+
+% This defines the macro itself. There are six cases: recursive and
+% nonrecursive macros of zero, one, and many arguments.
+% Much magic with \expandafter here.
+% \xdef is used so that macro definitions will survive the file
+% they're defined in; @include reads the file inside a group.
+\def\defmacro{%
+ \let\hash=##% convert placeholders to macro parameter chars
+ \ifrecursive
+ \ifcase\paramno
+ % 0
+ \expandafter\xdef\csname\the\macname\endcsname{%
+ \noexpand\scanmacro{\temp}}%
+ \or % 1
+ \expandafter\xdef\csname\the\macname\endcsname{%
+ \bgroup\noexpand\macroargctxt
+ \noexpand\braceorline
+ \expandafter\noexpand\csname\the\macname xxx\endcsname}%
+ \expandafter\xdef\csname\the\macname xxx\endcsname##1{%
+ \egroup\noexpand\scanmacro{\temp}}%
+ \else % many
+ \expandafter\xdef\csname\the\macname\endcsname{%
+ \bgroup\noexpand\macroargctxt
+ \noexpand\csname\the\macname xx\endcsname}%
+ \expandafter\xdef\csname\the\macname xx\endcsname##1{%
+ \expandafter\noexpand\csname\the\macname xxx\endcsname ##1,}%
+ \expandafter\expandafter
+ \expandafter\xdef
+ \expandafter\expandafter
+ \csname\the\macname xxx\endcsname
+ \paramlist{\egroup\noexpand\scanmacro{\temp}}%
+ \fi
+ \else
+ \ifcase\paramno
+ % 0
+ \expandafter\xdef\csname\the\macname\endcsname{%
+ \noexpand\norecurse{\the\macname}%
+ \noexpand\scanmacro{\temp}\egroup}%
+ \or % 1
+ \expandafter\xdef\csname\the\macname\endcsname{%
+ \bgroup\noexpand\macroargctxt
+ \noexpand\braceorline
+ \expandafter\noexpand\csname\the\macname xxx\endcsname}%
+ \expandafter\xdef\csname\the\macname xxx\endcsname##1{%
+ \egroup
+ \noexpand\norecurse{\the\macname}%
+ \noexpand\scanmacro{\temp}\egroup}%
+ \else % many
+ \expandafter\xdef\csname\the\macname\endcsname{%
+ \bgroup\noexpand\macroargctxt
+ \expandafter\noexpand\csname\the\macname xx\endcsname}%
+ \expandafter\xdef\csname\the\macname xx\endcsname##1{%
+ \expandafter\noexpand\csname\the\macname xxx\endcsname ##1,}%
+ \expandafter\expandafter
+ \expandafter\xdef
+ \expandafter\expandafter
+ \csname\the\macname xxx\endcsname
+ \paramlist{%
+ \egroup
+ \noexpand\norecurse{\the\macname}%
+ \noexpand\scanmacro{\temp}\egroup}%
+ \fi
+ \fi}
+
+\def\norecurse#1{\bgroup\cslet{#1}{macsave.#1}}
+
+% \braceorline decides whether the next nonwhitespace character is a
+% {. If so it reads up to the closing }, if not, it reads the whole
+% line. Whatever was read is then fed to the next control sequence
+% as an argument (by \parsebrace or \parsearg)
+\def\braceorline#1{\let\next=#1\futurelet\nchar\braceorlinexxx}
+\def\braceorlinexxx{%
+ \ifx\nchar\bgroup\else
+ \expandafter\parsearg
+ \fi \next}
+
+% We mant to disable all macros during \shipout so that they are not
+% expanded by \write.
+\def\turnoffmacros{\begingroup \def\do##1{\let\noexpand##1=\relax}%
+ \edef\next{\macrolist}\expandafter\endgroup\next}
+
+
+% @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{\begingroup\obeyspaces\parsearg\aliasxxx}
+\def\aliasxxx #1{\aliasyyy#1\relax}
+\def\aliasyyy #1=#2\relax{\ignoreactivespaces
+\edef\next{\global\let\expandafter\noexpand\csname#1\endcsname=%
+ \expandafter\noexpand\csname#2\endcsname}%
+\expandafter\endgroup\next}
+
+
+\message{cross references,}
+% @xref etc.
+
+\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 job is to define \lastnode.
+\def\node{\ENVcheck\parsearg\nodezzz}
+\def\nodezzz#1{\nodexxx #1,\finishnodeparse}
+\def\nodexxx#1,#2\finishnodeparse{\gdef\lastnode{#1}}
+\let\nwnode=\node
+\let\lastnode=\relax
+
+% The sectioning commands (@chapter, etc.) call these.
+\def\donoderef{%
+ \ifx\lastnode\relax\else
+ \expandafter\expandafter\expandafter\setref{\lastnode}%
+ {Ysectionnumberandtype}%
+ \global\let\lastnode=\relax
+ \fi
+}
+\def\unnumbnoderef{%
+ \ifx\lastnode\relax\else
+ \expandafter\expandafter\expandafter\setref{\lastnode}{Ynothing}%
+ \global\let\lastnode=\relax
+ \fi
+}
+\def\appendixnoderef{%
+ \ifx\lastnode\relax\else
+ \expandafter\expandafter\expandafter\setref{\lastnode}%
+ {Yappendixletterandtype}%
+ \global\let\lastnode=\relax
+ \fi
+}
+
+
+% @anchor{NAME} -- define xref target at arbitrary point.
+%
+\newcount\savesfregister
+\gdef\savesf{\relax \ifhmode \savesfregister=\spacefactor \fi}
+\gdef\restoresf{\relax \ifhmode \spacefactor=\savesfregister \fi}
+\gdef\anchor#1{\savesf \setref{#1}{Ynothing}\restoresf \ignorespaces}
+
+% \setref{NAME}{SNT} defines a cross-reference point NAME (a node or an
+% anchor), namely NAME-title (the corresponding @chapter/etc. name),
+% NAME-pg (the page number), and NAME-snt (section number and type).
+% Called from \foonoderef.
+%
+% We have to set \indexdummies so commands such as @code in a section
+% title aren't expanded. It would be nicer not to expand the titles in
+% the first place, but there's so many layers that that is hard to do.
+%
+% Likewise, use \turnoffactive so that punctuation chars such as underscore
+% and backslash work in node names.
+%
+\def\setref#1#2{{%
+ \atdummies
+ \pdfmkdest{#1}%
+ %
+ \turnoffactive
+ \dosetq{#1-title}{Ytitle}%
+ \dosetq{#1-pg}{Ypagenumber}%
+ \dosetq{#1-snt}{#2}%
+}}
+
+% @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\printednodename{\ignorespaces #3}%
+ \setbox1=\hbox{\printedmanual}%
+ \setbox0=\hbox{\printednodename}%
+ \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\printednodename{\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\printednodename{\ignorespaces #1}%
+ \else
+ \ifhavexrefs
+ % We know the real title if we have the xref values.
+ \def\printednodename{\refx{#1-title}{}}%
+ \else
+ % Otherwise just copy the Info node name.
+ \def\printednodename{\ignorespaces #1}%
+ \fi%
+ \fi
+ \fi
+ \fi
+ %
+ % 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.
+ \ifpdf
+ \leavevmode
+ \getfilename{#4}%
+ {\turnoffactive \otherbackslash
+ \ifnum\filenamelength>0
+ \startlink attr{/Border [0 0 0]}%
+ goto file{\the\filename.pdf} name{#1}%
+ \else
+ \startlink attr{/Border [0 0 0]}%
+ goto name{#1}%
+ \fi
+ }%
+ \linkcolor
+ \fi
+ %
+ \ifdim \wd1 > 0pt
+ \putwordsection{} ``\printednodename'' \putwordin{} \cite{\printedmanual}%
+ \else
+ % _ (for example) has to be the character _ for the purposes of the
+ % control sequence corresponding to the node, but it has to expand
+ % into the usual \leavevmode...\vrule stuff for purposes of
+ % printing. So we \turnoffactive for the \refx-snt, back on for the
+ % printing, back off for the \refx-pg.
+ {\turnoffactive \otherbackslash
+ % 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.
+ \xrefprintnodename\printednodename
+ %
+ % But we always want a comma and a space:
+ ,\space
+ %
+ % output the `page 3'.
+ \turnoffactive \otherbackslash \putwordpage\tie\refx{#1-pg}{}%
+ \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 not square brackets don't work in some documents. Particularly
+% one that Bob is working on :).
+%
+\def\xrefprintnodename#1{[#1]}
+
+% \dosetq is called from \setref to do the actual \write (\iflinks).
+%
+\def\dosetq#1#2{%
+ {\let\folio=0%
+ \edef\next{\write\auxfile{\internalsetq{#1}{#2}}}%
+ \iflinks \next \fi
+ }%
+}
+
+% \internalsetq{foo}{page} expands into
+% CHARACTERS @xrdef{foo}{...expansion of \page...}
+\def\internalsetq#1#2{@xrdef{#1}{\csname #2\endcsname}}
+
+% Things to be expanded by \internalsetq.
+%
+\def\Ypagenumber{\folio}
+\def\Ytitle{\thissection}
+\def\Ynothing{}
+\def\Ysectionnumberandtype{%
+ \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\Yappendixletterandtype{%
+ \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
+}
+
+% 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{\the\inputlineno:\space}
+\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 X#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.
+%
+\def\xrdef#1{\expandafter\gdef\csname X#1\endcsname}
+
+% Read the last existing aux file, if any. No error if none exists.
+\def\readauxfile{\begingroup
+ \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
+ %
+ % Make the characters 128-255 be printing characters
+ {%
+ \count 1=128
+ \def\loop{%
+ \catcode\count 1=\other
+ \advance\count 1 by 1
+ \ifnum \count 1<256 \loop \fi
+ }%
+ }%
+ %
+ % Turn off \ as an escape so we do not lose on
+ % entries which were dumped with control sequences in their names.
+ % For example, @xrdef{$\leq $-fun}{page ...} made by @defun ^^
+ % Reference to such entries still does not work the way one would wish,
+ % but at least they do not bomb out when the aux file is read in.
+ \catcode`\\=\other
+ %
+ % @ is our escape character in .aux files.
+ \catcode`\{=1
+ \catcode`\}=2
+ \catcode`\@=0
+ %
+ \openin 1 \jobname.aux
+ \ifeof 1 \else
+ \closein 1
+ \input \jobname.aux
+ \global\havexrefstrue
+ \global\warnedobstrue
+ \fi
+ % Open the new aux file. TeX will close it automatically at exit.
+ \openout\auxfile=\jobname.aux
+\endgroup}
+
+
+% Footnotes.
+
+\newcount \footnoteno
+
+% The trailing space in the following definition for supereject is
+% vital for proper filling; pages come out unaligned when you do a
+% pagealignmacro call if that space before the closing brace is
+% removed. (Generally, numeric constants should always be followed by a
+% space to prevent strange expansion errors.)
+\def\supereject{\par\penalty -20000\footnoteno =0 }
+
+% @footnotestyle is meaningful for info output only.
+\let\footnotestyle=\comment
+
+\let\ptexfootnote=\footnote
+
+{\catcode `\@=11
+%
+% Auto-number footnotes. Otherwise like plain.
+\gdef\footnote{%
+ \let\indent=\ptexindent
+ \global\advance\footnoteno by \@ne
+ \edef\thisfootno{$^{\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 fail inside footnotes because the tokens are fixed when
+% the footnote is read. --karl, 16nov96.
+%
+% The start of the footnote looks usually like this:
+\gdef\startfootins{\insert\footins\bgroup}
+%
+% ... but this macro is redefined inside @multitable.
+%
+\gdef\dofootnote{%
+ \startfootins
+ % 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
+
+% @| 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
+ }%
+ }%
+}
+
+% 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}
+
+% @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
+ \closein 1
+ % 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
+%
+% We will only complain once about lack of epsf.tex.
+\newif\ifwarnednoepsf
+\newhelp\noepsfhelp{epsf.tex must be installed for images to
+ work. It is also included in the Texinfo distribution, or you can get
+ it from ftp://tug.org/tex/epsf.tex.}
+%
+\def\image#1{%
+ \ifx\epsfbox\undefined
+ \ifwarnednoepsf \else
+ \errhelp = \noepsfhelp
+ \errmessage{epsf.tex not found, images will be ignored}%
+ \global\warnednoepsftrue
+ \fi
+ \else
+ \imagexxx #1,,,,,\finish
+ \fi
+}
+%
+% Arguments to @image:
+% #1 is (mandatory) image filename; we tack on .eps extension.
+% #2 is (optional) width, #3 is (optional) height.
+% #4 is (ignored optional) html alt text.
+% #5 is (ignored optional) extension.
+% #6 is just the usual extra ignored arg for parsing this stuff.
+\newif\ifimagevmode
+\def\imagexxx#1,#2,#3,#4,#5,#6\finish{\begingroup
+ \catcode`\^^M = 5 % in case we're inside an example
+ \normalturnoffactive % allow _ et al. in names
+ % If the image is by itself, center it.
+ \ifvmode
+ \imagevmodetrue
+ \nobreak\bigskip
+ % Usually we'll have text after the image which will insert
+ % \parskip glue, so insert it here too to equalize the space
+ % above and below.
+ \nobreak\vskip\parskip
+ \nobreak
+ \line\bgroup\hss
+ \fi
+ %
+ % Output the image.
+ \ifpdf
+ \dopdfimage{#1}{#2}{#3}%
+ \else
+ % \epsfbox itself resets \epsf?size at each figure.
+ \setbox0 = \hbox{\ignorespaces #2}\ifdim\wd0 > 0pt \epsfxsize=#2\relax \fi
+ \setbox0 = \hbox{\ignorespaces #3}\ifdim\wd0 > 0pt \epsfysize=#3\relax \fi
+ \epsfbox{#1.eps}%
+ \fi
+ %
+ \ifimagevmode \hss \egroup \bigbreak \fi % space after the image
+\endgroup}
+
+
+\message{localization,}
+% and i18n.
+
+% @documentlanguage is usually given very early, just after
+% @setfilename. If done too late, it may not override everything
+% properly. Single argument is the language abbreviation.
+% It would be nice if we could set up a hyphenation file here.
+%
+\def\documentlanguage{\parsearg\dodocumentlanguage}
+\def\dodocumentlanguage#1{%
+ \tex % read txi-??.tex file in plain TeX.
+ % Read the file if it exists.
+ \openin 1 txi-#1.tex
+ \ifeof1
+ \errhelp = \nolanghelp
+ \errmessage{Cannot read language file txi-#1.tex}%
+ \let\temp = \relax
+ \else
+ \def\temp{\input txi-#1.tex }%
+ \fi
+ \temp
+ \endgroup
+}
+\newhelp\nolanghelp{The given language definition file cannot be found or
+is empty. Maybe you need to install it? In the current directory
+should work if nowhere else does.}
+
+
+% @documentencoding should change something in TeX eventually, most
+% likely, but for now just recognize it.
+\let\documentencoding = \comment
+
+
+% Page size parameters.
+%
+\newdimen\defaultparindent \defaultparindent = 15pt
+
+\chapheadingskip = 15pt plus 4pt minus 2pt
+\secheadingskip = 12pt plus 3pt minus 2pt
+\subsecheadingskip = 9pt plus 2pt minus 2pt
+
+% Prevent underfull vbox error messages.
+\vbadness = 10000
+
+% Don't be so finicky about underfull hboxes, either.
+\hbadness = 2000
+
+% Following George Bush, just get rid of widows and orphans.
+\widowpenalty=10000
+\clubpenalty=10000
+
+% Use TeX 3.0's \emergencystretch to help line breaking, but if we're
+% using an old version of TeX, don't do anything. We want the amount of
+% stretch added to depend on the line length, hence the dependence on
+% \hsize. We call this whenever the paper size is set.
+%
+\def\setemergencystretch{%
+ \ifx\emergencystretch\thisisundefined
+ % Allow us to assign to \emergencystretch anyway.
+ \def\emergencystretch{\dimen0}%
+ \else
+ \emergencystretch = .15\hsize
+ \fi
+}
+
+% Parameters in order: 1) textheight; 2) textwidth; 3) voffset;
+% 4) hoffset; 5) binding offset; 6) topskip; 7) physical page height; 8)
+% physical page width.
+%
+% We also call \setleading{\textleading}, so the caller should define
+% \textleading. The caller should also set \parskip.
+%
+\def\internalpagesizes#1#2#3#4#5#6#7#8{%
+ \voffset = #3\relax
+ \topskip = #6\relax
+ \splittopskip = \topskip
+ %
+ \vsize = #1\relax
+ \advance\vsize by \topskip
+ \outervsize = \vsize
+ \advance\outervsize by 2\topandbottommargin
+ \pageheight = \vsize
+ %
+ \hsize = #2\relax
+ \outerhsize = \hsize
+ \advance\outerhsize by 0.5in
+ \pagewidth = \hsize
+ %
+ \normaloffset = #4\relax
+ \bindingoffset = #5\relax
+ %
+ \ifpdf
+ \pdfpageheight #7\relax
+ \pdfpagewidth #8\relax
+ \fi
+ %
+ \setleading{\textleading}
+ %
+ \parindent = \defaultparindent
+ \setemergencystretch
+}
+
+% @letterpaper (the default).
+\def\letterpaper{{\globaldefs = 1
+ \parskip = 3pt plus 2pt minus 1pt
+ \textleading = 13.2pt
+ %
+ % If page is nothing but text, make it come out even.
+ \internalpagesizes{46\baselineskip}{6in}%
+ {\voffset}{.25in}%
+ {\bindingoffset}{36pt}%
+ {11in}{8.5in}%
+}}
+
+% Use @smallbook to reset parameters for 7x9.5 (or so) format.
+\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 @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.
+%
+\def\pagesizes{\parsearg\pagesizesxxx}
+\def\pagesizesxxx#1{\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 ttfont
+% where it can probably just be output, 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\_}
+% 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
+
+% Set up an active definition for =, but don't enable it most of the time.
+{\catcode`\==\active
+\global\def={{\tt \char 61}}}
+
+\catcode`+=\active
+\catcode`\_=\active
+
+% If a .fmt file is being used, characters that might appear in a file
+% name cannot be active until we have parsed the command line.
+% So turn them off again, and have \everyjob (or @setfilename) turn them on.
+% \otherifyactive is called near the end of this file.
+\def\otherifyactive{\catcode`+=\other \catcode`\_=\other}
+
+\catcode`\@=0
+
+% \rawbackslashxx outputs one backslash character in current font,
+% as in \char`\\.
+\global\chardef\rawbackslashxx=`\\
+
+% \rawbackslash defines an active \ to do \rawbackslashxx.
+% \otherbackslash defines an active \ to be a literal `\' character with
+% catcode other.
+{\catcode`\\=\active
+ @gdef@rawbackslash{@let\=@rawbackslashxx}
+ @gdef@otherbackslash{@let\=@realbackslash}
+}
+
+% \realbackslash is an actual character `\' with catcode other.
+{\catcode`\\=\other @gdef@realbackslash{\}}
+
+% \normalbackslash outputs one backslash in fixed width font.
+\def\normalbackslash{{\tt\rawbackslashxx}}
+
+\catcode`\\=\active
+
+% Used sometimes to turn off (effectively) the active characters
+% even after parsing them.
+@def@turnoffactive{%
+ @let"=@normaldoublequote
+ @let\=@realbackslash
+ @let~=@normaltilde
+ @let^=@normalcaret
+ @let_=@normalunderscore
+ @let|=@normalverticalbar
+ @let<=@normalless
+ @let>=@normalgreater
+ @let+=@normalplus
+ @let$=@normaldollar %$ font-lock fix
+}
+
+% Same as @turnoffactive except outputs \ as {\tt\char`\\} instead of
+% the literal character `\'. (Thus, \ is not expandable when this is in
+% effect.)
+%
+@def@normalturnoffactive{@turnoffactive @let\=@normalbackslash}
+
+% Make _ and + \other characters, temporarily.
+% This is canceled by @fixbackslash.
+@otherifyactive
+
+% If a .fmt file is being used, we don't want the `\input texinfo' to show up.
+% That is what \eatinput is for; after that, the `\' should revert to printing
+% a backslash.
+%
+@gdef@eatinput input texinfo{@fixbackslash}
+@global@let\ = @eatinput
+
+% On the other hand, perhaps the file did not have a `\input texinfo'. Then
+% the first `\{ in the file would cause an error. This macro tries to fix
+% that, assuming it is called before the first `\' could plausibly occur.
+% Also back turn on active characters that might appear in the input
+% 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 Set initial fonts.
+@textfonts
+@rm
+
+
+@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:

File Metadata

Mime Type
text/x-diff
Expires
Thu, Feb 26, 6:40 PM (16 h, 3 m)
Storage Engine
local-disk
Storage Format
Raw Data
Storage Handle
cf/3e/27ae93040361de1ddc588f4b6f9e

Event Timeline