release/4.1.1

13 files changed, 0 insertions, 11014 deletions

diff --git a/contrib/awk/doc/Makefile.in b/contrib/awk/doc/Makefile.in
deleted file mode 100644
index edcf404656021..0000000000000
--- a/contrib/awk/doc/Makefile.in
+++ /dev/null
@@ -1,129 +0,0 @@
-# Makefile for GNU Awk documentation.
-#
-# Copyright (C) 1993-2000 the Free Software Foundation, Inc.
-# 
-# This file is part of GAWK, the GNU implementation of the
-# AWK Programming Language.
-# 
-# GAWK 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.
-# 
-# GAWK 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
-
-SHELL = /bin/sh
-
-srcdir = @srcdir@
-VPATH = @srcdir@
-
-INSTALL = @INSTALL@
-INSTALL_PROGRAM = @INSTALL_PROGRAM@
-INSTALL_DATA = @INSTALL_DATA@
-
-prefix = @prefix@
-exec_prefix = @exec_prefix@
-binprefix =
-manprefix =
-
-bindir = @bindir@
-libdir = @libdir@
-mandir = @mandir@/man1
-manext = .1
-infodir = @infodir@
-datadir = @datadir@/awk
-
-TEXI2DVI = texi2dvi
-TEX = tex
-MAKEINFO = makeinfo --no-split
-TROFF = groff -t -Tps
-SEDME = sed -e "s/^level0 restore/level0 restore flashme 100 72 moveto (Copyright `date '+%m-%d-%y %T'`, FSF, Inc. (all)) show/" \
-		-e "s/^\/level0 save def/\/level0 save def 30 -48 translate/"
-
-DOCS= gawk.1 igawk.1 gawk.texi
-
-TEXFILES= gawk.aux gawk.cp gawk.cps gawk.fn gawk.fns gawk.ky gawk.kys \
-	   gawk.pg gawk.pgs gawk.toc gawk.tp gawk.tps gawk.vr gawk.vrs
-
-ALLDOC= gawk.dvi $(TEXFILES) gawk.log awkcard.tr
-
-CARDSRC = $(srcdir)/macros $(srcdir)/cardfonts $(srcdir)/colors awkcard.tr
-CARDSRC_N = $(srcdir)/macros $(srcdir)/cardfonts $(srcdir)/no.colors awkcard.tr
-CARDFILES= $(CARDSRC) ad.block awkcard.in setter.outline
-
-# Use this if your troff can correctly handle macros from 'colors' file
-AWKCARD = awkcard.ps
-
-# Uncomment the following definition of AWKCARD if your troff can produce
-# Postscript but still has troubles with macros from 'colors'.  As this
-# is not groff you will have to change TROFF macro as well.  Do not forget
-# to ensure that awkcard.tr is processed by tbl.
-#AWKCARD = awkcard.nc
-
-all: $(DOCS) info
-
-install: $(mandir)/gawk$(manext) $(mandir)/igawk$(manext) $(infodir)/gawk.info
-
-$(infodir)/gawk.info::
-	-if test -f gawk.info; then d=.; \
-	else d=$(srcdir); fi; \
-	if    [ -f $(infodir)/dir -a -f $(infodir)/gawk.info ] \
-	   && cmp $$d/gawk.info $(infodir)/gawk.info > /dev/null \
-	   && grep '(gawk)' $(infodir)/dir  > /dev/null; then \
-		exit 0; \
-	fi; \
-	$(INSTALL_DATA) $$d/gawk.info $(infodir)/gawk.info ; \
-	if $(SHELL) -c 'install-info --version' > /dev/null 2>&1 ; \
-	then install-info --info-dir=$(infodir) gawk.info ; \
-	else true ; fi; exit 0
-
-$(mandir)/gawk$(manext):: gawk.1
-	$(INSTALL_DATA) $(srcdir)/gawk.1 $(mandir)/gawk$(manext)
-
-$(mandir)/igawk$(manext):: igawk.1
-	$(INSTALL_DATA) $(srcdir)/igawk.1 $(mandir)/igawk$(manext)
-
-uninstall:
-	rm -f $(mandir)/gawk$(manext) $(mandir)/igawk$(manext) $(infodir)/gawk.info*
-
-dvi:	gawk.dvi
-
-gawk.dvi: gawk.texi
-	-TEXINPUTS=$(srcdir):$$TEXINPUTS $(TEXI2DVI) $(srcdir)/gawk.texi
-
-info:	gawk.info
-
-gawk.info: gawk.texi
-	$(MAKEINFO) $(srcdir)/gawk.texi
-
-postscript: dvi gawk.1 igawk.1 $(AWKCARD)
-	-groff -man $(srcdir)/gawk.1 > gawk.1.ps
-	-groff -man $(srcdir)/igawk.1 > igawk.1.ps
-	dvips -o gawk.ps gawk.dvi
-
-awkcard.tr: awkcard.in
-	sed 's:SRCDIR:$(srcdir):' < $(srcdir)/awkcard.in > awkcard.tr
-
-awkcard.ps: $(CARDFILES)
-	$(TROFF) $(CARDSRC) | $(SEDME) | cat $(srcdir)/setter.outline - > awkcard.ps
-
-awkcard.nc: $(CARDFILES)
-	$(TROFF) $(CARDSRC_N) | $(SEDME) | cat $(srcdir)/setter.outline - > awkcard.ps && touch awkcard.nc
-
-clean:
-	rm -f *.ps $(ALLDOC) *~ awkcard.nc
-
-distclean: clean
-	rm -f Makefile
-
-maintainer-clean: distclean
-	@echo "This command is intended for maintainers to use; it"
-	@echo "deletes files that may require special tools to rebuild."
-	rm -f gawk.info
diff --git a/contrib/awk/doc/README.card b/contrib/awk/doc/README.card
deleted file mode 100644
index ef77cdacec7b0..0000000000000
--- a/contrib/awk/doc/README.card
+++ /dev/null
@@ -1,19 +0,0 @@
-Mon Dec  9 12:45:48 EST 1996
-
-The AWK reference card included here requires a modern version of troff
-(ditroff). GNU Troff (groff) is known to work.
-
-If your troff is able to produce Postscript but does not know how to
-properly use the macros from `colors' file then try to uncomment in
-Makefile the defintion which sets AWKCARD to awkcard.nc (no colors).
-This will definitely require changes to the TROFF macro and you have to
-ensure that the tbl preprocessor is called.  For example, the following
-modifications on NeXT:
-
-TROFF = tbl
-SEDME = ptroff -t | sed -e \
-	"s/^level0 restore/level0 restore flashme 100 72 moveto\
- (Copyright `date`, FSF, Inc. (all)) show/" \
-	-e "s/^\/level0 save def/\/level0 save def 30 -48 translate/"
-
-will produce a correctly formatted, albeit monochromatic, reference card.
diff --git a/contrib/awk/doc/ad.block b/contrib/awk/doc/ad.block
deleted file mode 100644
index d31f5d50a0437..0000000000000
--- a/contrib/awk/doc/ad.block
+++ /dev/null
@@ -1,49 +0,0 @@
-.\" AWK Reference Card --- Arnold Robbins, arnold@gnu.org
-.\" This file is the Ad block (included in cover)
-.\"
-.\" Copyright (C) 1996, 98 Free Software Foundation, Inc.
-.\" 
-.\" Permission is granted to make and distribute verbatim copies of
-.\" this reference card provided the copyright notice and this permission
-.\" notice are preserved on all copies.
-.\" 
-.\" Permission is granted to process this file through troff and print the
-.\" results, provided the printed document carries copying permission
-.\" notice identical to this one except for the removal of this paragraph
-.\" (this paragraph not being relevant to the printed reference card).
-.\" 
-.\" Permission is granted to copy and distribute modified versions of this
-.\" reference card under the conditions for verbatim copying, provided that
-.\" the entire resulting derived work is distributed under the terms of a
-.\" permission notice identical to this one.
-.\" 
-.\" Permission is granted to copy and distribute translations of this
-.\" reference card into another language, under the above conditions for
-.\" modified versions, except that this permission notice may be stated in
-.\" a translation approved by the Foundation.
-.\"
-.ft HB
-.ps 10
-.vs 12
-.ES
-.nf
-.ce 7
-\*(CBFree Software Foundation, Inc.
-.ft H
-59 Temple Place \(em Suite 330
-Boston, MA  02111-1307 USA
-Phone: +1-617-542-5942
-Fax (including Japan): +1-617-542-2652
-E-mail: gnu@gnu.org
-URL: http://www.gnu.org
-
-.ce 7
-.ft HB
-\*(CGFree Software
-Source Distributions on CD-ROM
-Deluxe Distributions
-Emacs, Gawk, Make and GDB Manuals
-Emacs and GDB References\*(CX
-.EB "\f(HBOTHER FSF PRODUCTS:\*(FR"
-.ps
-.vs
diff --git a/contrib/awk/doc/awkcard.in b/contrib/awk/doc/awkcard.in
deleted file mode 100644
index ac1e8e56c74c8..0000000000000
--- a/contrib/awk/doc/awkcard.in
+++ /dev/null
@@ -1,1550 +0,0 @@
-.\" AWK Reference Card --- Arnold Robbins, arnold@gnu.org
-.\"
-.\" Copyright (C) 1996, 1997, 1998, 1999, 2000 Free Software Foundation, Inc.
-.\" 
-.\" Permission is granted to make and distribute verbatim copies of
-.\" this reference card provided the copyright notice and this permission
-.\" notice are preserved on all copies.
-.\" 
-.\" Permission is granted to process this file through troff and print the
-.\" results, provided the printed document carries copying permission
-.\" notice identical to this one except for the removal of this paragraph
-.\" (this paragraph not being relevant to the printed reference card).
-.\" 
-.\" Permission is granted to copy and distribute modified versions of this
-.\" reference card under the conditions for verbatim copying, provided that
-.\" the entire resulting derived work is distributed under the terms of a
-.\" permission notice identical to this one.
-.\" 
-.\" Permission is granted to copy and distribute translations of this
-.\" reference card into another language, under the above conditions for
-.\" modified versions, except that this permission notice may be stated in
-.\" a translation approved by the Foundation.
-.\"
-.\" Strings to save typing
-.ds AK \*(FCawk\*(FR
-.ds GK \*(FCgawk\*(FR
-.ds NK Bell Labs \*(FCawk\*(FR
-.ds MK \*(FCmawk\*(FR
-.\"
-.\"
-.de TD\" tab defaults
-.ta .2i .78i 1i 1.2i 1.4i 1.7i
-..
-.de TE
-.TD
-..
-
-.sp
-.ce
-\*(CD\f(HB\s+8AWK REFERENCE\s0\*(FR
-.sp
-.\" --- Table Of Contents
-.ta 2.4i 2.6iR
-.lc .
-.ES
-.in +.2i
-.nf
-\*(FRAWK Program Execution
	4
-Action Statements
	7
-Arrays
	9
-Bug Reports
	15
-Command Line Arguments (standard)
	2
-Command Line Arguments (\*(GK)
	3
-Command Line Arguments (\*(MK)
	4
-Conversions And Comparisons
	10
-Copying Permissions
	16
-Definitions
	2
-Environment Variables
	16
-Escape Sequences
	7
-Expressions
	9
-Fields
	6
-FTP Information
	16
-Historical Features (\*(GK)
	16
-Input Control
	11
-Lines And Statements
	5
-.ig
-Localization
	10
-..
-Numeric Functions
	13
-Output Control
	11
-Pattern Elements
	7
-POSIX Character Classes (\*(GK)
	6
-Printf Formats
	12
-Records
	6
-Regular Expressions
	5
-Special Filenames
	13
-String Functions
	14
-Time Functions (\*(GK)
	15
-User-defined Functions
	15
-Variables
	8\*(CX
-.in -.2i
-.EB "\s+2\f(HBCONTENTS\*(FR\s0"
-.sp
-.TD
-.fi
-\*(CD\*(FRThis reference card was written by Arnold Robbins.
-Brian Kernighan and Michael Brennan reviewed it; we thank them
-for their help.
-.sp
-.SL
-.sp
-.so SRCDIR/ad.block
-.\" a subtlety here; this line changes color. We rely on it
-.\" also to provide a blank line.
-\*(CD
-.SL
-.nf
-\*(FR\(co Copyright 1996-2000, Free Software Foundation
-59 Temple Place \(em Suite 330
-Boston, MA  02111-1307 USA
-.nf
-.BT
-
-
-.\"
-.\"
-.\" --- Definitions
-.fi
-.ES
-\*(CDThis card describes POSIX AWK, as well as the three
-freely available \*(AK implementations
-(see \fHFTP Information\fP below).
-\*(CLCommon extensions (in two or more versions) are printed in light blue.
-\*(CBFeatures specific to just one version\(emusually GNU AWK (\*(GK)\(emare
-printed in dark blue.
-\*(CRExceptions and deprecated features are printed in red.
-\*(CDFeatures mandated by POSIX are printed in black.
-.sp .5
-Several type faces are used to clarify the meaning:
-.br
-.nr IN \w'\(bu '
-\(bu \*(FC\*(CN\fP is used for computer input.
-.br
-.fi
-.in +\n(INu
-.ti -\n(INu
-\(bu\|\^\*(FI\*(IN\fP is used to indicate user input and for syntactic
-placeholders, such as \*(FIvariable\fP or \*(FIaction\fP.
-.in -\n(INu
-.br
-\(bu \*(RN is used for explanatory text.
-.sp .5
-\*(FInumber\fP \- a floating point number as in ANSI C, such as
-\*(FC3\*(FR,
-\*(FC2.3\*(FR,
-\*(FC.4\*(FR,
-\*(FC1.4e2\*(FR
-or
-\*(FC4.1E5\*(FR.
-.sp .5
-\*(FIescape sequences\fP \- a special sequence of characters beginning
-with a backslash, used to describe otherwise unprintable characters.
-(See \fHEscape Sequences\fP below.)
-.sp .5
-\*(FIstring\fP \- a group of characters enclosed in double quotes.
-Strings may contain \*(FIescape sequences\*(FR.
-.sp .5
-\*(FIregexp\fP \- a regular expression, either a regexp constant
-enclosed in forward slashes, or a dynamic regexp computed at run-time.
-Regexp constants may contain \*(FIescape sequences\*(FR.
-.sp .5
-\*(FIname\fP \- a variable, array or function name.
-.sp .5
-\*(FIentry\fP(\*(FIN\fP) \- entry \*(FIentry\fP in section \*(FIN\fP of the
-UNIX reference manual.
-.sp .5
-\*(FIpattern\fP \- an expression describing an input record to be matched.
-.sp .5
-\*(FIaction\fP \- statements to execute when an input record is matched.
-.sp .5
-\*(FIrule\fP \- a pattern-action pair, where the pattern or action may
-be missing.\*(CX
-.EB \s+2\f(HBDEFINITIONS\*(FR\s0
-
-.\" --- Command Line Arguments
-.ES
-.fi
-\*(CDCommand line arguments control setting the field separator,
-setting variables before the \*(FCBEGIN\fP rule is run, and
-the location of AWK program source code.
-Implementation-specific command line arguments change
-the behavior of the running interpreter.
-.sp .5
-.TS
-expand;
-l lw(2.2i).
-\*(FC\-F \*(FIfs\*(FR	use \*(FIfs\fP for the input field separator.
-\*(FC\-v\*(FI var\*(FC\^=\^\*(FIval\*(FR	T{
-assign the value \*(FIval\*(FR, to the variable \*(FIvar\*(FR,
-before execution of the program begins. Such
-variable values are available to the \*(FCBEGIN\fP rule.
-T}
-\*(FC\-f \*(FIprog-file\*(FR	T{
-read the AWK program source from the file
-\*(FIprog-file\*(FR, instead of from the first command
-line argument. Multiple \*(FC\-f\*(FR options may be used.
-T}
-\*(FC\-\^\-\*(FR	signal the end of options.
-.TE
-.sp .5
-.fi
-\*(CLThe following options are accepted by both \*(NK and \*(GK
-\*(CR(ignored by \*(GK, not in \*(MK).\*(CL
-.sp .5
-.nf
-.TS
-expand, tab(%);
-l lw(2.2i).
-\*(FC\-mf \*(FIval\*(FR%set the maximum number of fields to \*(FIval\fP
-\*(FC\-mr \*(FIval\*(FR%set the maximum record size to \*(FIval\fP\*(CX
-.TE
-.EB "\s+2\f(HBCOMMAND LINE ARGUMENTS (standard)\*(FR\s0"
-
-.BT
-
-.ES
-.fi
-\*(CDThe following options are specific to \*(GK. The \*(FC\-W\*(FR
-forms are for full POSIX compliance.
-.sp .5
-.ig
-.\" This option is left undocumented, on purpose.
-\*(FC\-\^\-nostalgia\*(FR
-\*(FC\-W nostalgia\*(FR%T{
-provide a moment of nostalgia for
-long time \*(AK users.
-T}
-..
-.TS
-expand, tab(%);
-ls
-l lw(1.8i).
-\*(FC\-\^\-field-separator \*(FIfs\*(FR
-%just like \*(FC\-F\fP
-\*(FC\-\^\-assign \*(FIvar\*(FC\^=\^\*(FIval\*(FR%just like \*(FC\-v\fP
-\*(FC\-\^\-file \*(FIprog-file%\*(FRjust like \*(FC\-f\fP
-\*(FC\-\^\-traditional\*(FR
-\*(FC\-\^\-compat\*(FR
-\*(FC\-W compat\*(FR
-\*(FC\-W traditional\*(FR%T{
-turn off \*(GK-specific extensions
-(\*(FC\-\^\-traditional\*(FR preferred).
-T}
-\*(FC\-\^\-copyleft\*(FR
-\*(FC\-\^\-copyright\*(FR
-\*(FC\-W copyleft\*(FR
-\*(FC\-W copyright\*(FR%T{
-print the short version of the GNU
-copyright information on \*(FCstdout\*(FR.
-T}
-\*(FC\-\^\-help\*(FR
-\*(FC\-\^\-usage\*(FR
-\*(FC\-W help\*(FR
-\*(FC\-W usage\*(FR%T{
-print a short summary of the available
-options on \*(FCstdout\*(FR, then exit zero.
-T}
-\*(FC\-\^\-lint\*(FR
-\*(FC\-W lint\*(FR%T{
-warn about constructs that are dubious
-or non-portable to other \*(AKs.
-T}
-\*(FC\-\^\-lint\-old\*(FR
-\*(FC\-W lint\-old\*(FR%T{
-warn about constructs that are not
-portable to the original version of
-Unix \*(AK.
-T}
-\*(FC\-\^\-posix\*(FR
-\*(FC\-W posix\*(FR%T{
-disable common and GNU extensions.
-Enable \*(FIinterval expressions\*(FR in regular
-expression matching (see \fHRegular
-Expressions\fP below).
-T}
-\*(FC\-\^\-re\-interval\*(FR
-\*(FC\-W re\-interval\*(FR%T{
-enable \*(FIinterval expressions\*(FR in regular
-expression matching (see \fHRegular
-Expressions\fP below). Useful if
-\*(FC\-\^\-posix\*(FR is not specified.
-T}
-\*(FC\-\^\-source '\*(FItext\*(FC'\*(FR
-\*(FC\-W source '\*(FItext\*(FC'\*(FR%use \*(FItext\*(FR as AWK program source code.
-\*(FC\-\^\-version\*(FR
-\*(FC\-W version\*(FR%T{
-print version information on \*(FCstdout\fP
-and exit zero.
-T}
-.TE
-.sp .5
-.fi
-In compatibility mode,
-any other options are flagged as illegal, but are otherwise ignored.
-In normal operation, as long as program text has been supplied, unknown
-options are passed on to the AWK program in
-\*(FCARGV\*(FR
-for processing. This is most useful for running AWK
-programs via the \*(FC#!\*(FR executable interpreter mechanism.\*(CB
-.EB "\s+2\f(HBCOMMAND LINE ARGUMENTS (\*(GK\f(HB)\*(FR\s0"
-
-.BT
-
-.ES
-.fi
-\*(CDThe following options are specific to \*(MK.
-.sp .5
-.fi
-.TS
-expand;
-l lw(1.8i).
-\*(FC\-W dump\*(FR	T{
-print an assembly listing of the program to
-\*(FCstdout\fP and exit zero.
-T}
-\*(FC\-W exec \*(FIfile\*(FR	T{
-read program text from \*(FIfile\fP. No other
-options are processed. Useful with \*(FC#!\fP.
-T}
-\*(FC\-W interactive\*(FR	T{
-unbuffer \*(FCstdout\fP and line buffer \*(FCstdin\fP.
-Lines are always records, ignoring \*(FCRS\fP
-T}
-\*(FC\-W posix_space\*(FR	T{
-\*(FC\en\*(FR separates fields when \*(FCRS = "\^"\fP.
-T}
-\*(FC\-W sprintf=\*(FInum\*(FR	T{
-adjust the size of \*(MK's internal
-\*(FCsprintf\*(FR buffer.
-T}
-\*(FC\-W version\*(FR	T{
-print version and copyright on
-\*(FCstdout\fP and limit information on \*(FCstderr\fP
-and exit zero.
-T}
-.TE
-.sp .5
-.fi
-The options may be abbreviated using just the first letter, e.g.,
-\*(FC\-We\*(FR,
-\*(FC\-Wv\*(FR
-and so on.\*(CB
-.EB "\s+2\f(HBCOMMAND LINE ARGUMENTS (\*(MK\f(HB)\*(FR\s0"
-
-.\" --- Awk Program Execution
-.ES
-.fi
-\*(CDAWK programs are a sequence of pattern-action statements
-and optional function definitions.
-.sp .5
-	\*(FIpattern\*(FC	{ \*(FIaction statements\*(FC }\*(FR
-.br
-	\*(FCfunction \*(FIname\*(FC(\*(FIparameter list\*(FC) { \*(FIstatements\*(FC }\*(FR
-.sp .5
-\*(AK first reads the program source from the
-\*(FIprog-file\*(FR(s), if specified,
-\*(CBfrom arguments to \*(FC\-\^\-source\*(FR,\*(CD
-or from the first non-option argument on the command line.
-The program text is read as if all the \*(FIprog-file\*(FR(s)
-\*(CBand command line
-source texts\*(CD had been concatenated.
-.sp .5
-AWK programs execute in the following order.
-First, all variable assignments specified via the \*(FC\-v\fP
-option are performed.
-Next, \*(AK executes the code in the
-\*(FCBEGIN\fP rules(s), if any, and then proceeds to read
-the files \*(FC1\fP through \*(FCARGC \- 1\fP in the \*(FCARGV\fP array.
-(Adjusting \*(FCARGC\fP and \*(FCARGV\fP thus provides control over
-the input files that will be processed.)
-If there are no files named on the command line,
-\*(AK reads the standard input.
-.sp .5
-If a command line argument has the form
-\*(FIvar\*(FC=\*(FIval\*(FR,
-it is treated as a variable assignment. The variable
-\*(FIvar\fP will be assigned the value \*(FIval\*(FR.
-(This happens after any \*(FCBEGIN\fP rule(s) have been run.)
-... delete this paragraph if no space
-Command line variable assignment
-is most useful for dynamically assigning values to the variables
-\*(AK uses to control how input is broken into fields and records. It
-is also useful for controlling state if multiple passes are needed over
-a single data file.
-.sp .5
-If the value of a particular element of \*(FCARGV\fP is empty
-(\*(FC"\^"\*(FR), \*(AK skips over it.
-.sp .5
-For each record in the input, \*(AK tests to see if it matches any
-\*(FIpattern\fP in the AWK program.
-For each pattern that the record matches, the associated
-\*(FIaction\fP is executed.
-The patterns are tested in the order they occur in the program.
-.sp .5
-Finally, after all the input is exhausted,
-\*(AK executes the code in the \*(FCEND\fP rule(s), if any.
-.sp .5
-If a program only has a \*(FCBEGIN\fP rule, no input files are processed.
-If a program only has an \*(FCEND\fP rule, the input will be read.
-\*(CX
-.EB "\s+2\f(HBAWK PROGRAM EXECUTION\*(FR\s0"
-
-
-.BT
-
-.\" --- Lines And Statements
-.ES
-.fi
-\*(CDAWK is a line oriented language. The pattern comes first, and then the
-action. Action statements are enclosed in \*(FC{\fP and \*(FC}\*(FR.
-Either the pattern  or the action may be missing, but
-not both. If the pattern is missing, the action will be
-executed for every input record.
-A missing action is equivalent to
-.sp .5
-	\*(FC{ print }\fP
-.sp .5
-which prints the entire record.
-.sp .5
-Comments begin with the \*(FC#\*(FR character, and continue until the
-end of the line.
-Normally, a statement ends with a newline, but lines ending in
-a ``,'',
-\*(FC{\*(FR,
-\*(CB\*(FC?\*(FR,
-\*(FC:\*(FR,\*(CD
-\*(FC&&\*(FR
-or
-\*(FC||\*(FR
-are automatically continued.
-Lines ending in \*(FCdo\fP or \*(FCelse\fP
-also have their statements automatically continued on the following line.
-In other cases, a line can be continued by ending it with a ``\e'',
-in which case the newline will be ignored. However, a ``\e'' after a
-\*(FC#\*(FR is not special.
-.sp .5
-Multiple statements may be put on one line by separating them with a ``;''.
-This applies to both the statements within the action part of a
-pattern-action pair (the usual case)
-and to the pattern-action statements themselves.\*(CX
-.EB "\s+2\f(HBLINES AND STATEMENTS\*(FR\s0"
-
-
-
-.\" --- Regular Expressions
-.ES
-.fi
-\*(CDRegular expressions are the extended kind originally defined by
-\*(FCegrep\fP.
-\*(CBAdditional GNU regexp operators are supported by \*(GK.
-A \*(FIword-constituent\fP character is a letter, digit, or
-underscore (\*(FC_\fP).\*(CD
-.sp .5
-.TS
-center, tab(~);
-cp8 sp8
-cp8 sp8
-lp8|lp8.
-.\" .vs 10
-_
-Summary of Regular Expressions
-In Decreasing Precedence
-_
-\*(FC(\^\*(FIr\*(FC)\*(FR~regular expression (for grouping)
-\*(FIc\*(FR~if non-special char, matches itself
-\*(FC\e\*(FI\^c\*(FR~turn off special meaning of \*(FIc\fP
-\*(FC^\*(FR~beginning of string (note: \*(FInot\fP line)
-\*(FC$\*(FR~end of string (note: \*(FInot\fP line)
-\*(FC.\*(FR~any single character, including newline
-\*(FC[\*(FR...\*(FC]\*(FR~any one character in ... or range
-\*(FC[^\*(FR...\*(FC]\*(FR~any one character not in ... or range
-\*(CB\*(FC\ey\*(FR~word boundary
-\*(FC\eB\*(FR~middle of a word
-\*(FC\e<\*(FR~beginning of a word
-\*(FC\e>\*(FR~end of a word
-\*(FC\ew\*(FR~any word-constituent character
-\*(FC\eW\*(FR~any non-word-constituent character
-\*(FC\e`\*(FR~beginning of a buffer (string)
-\*(FC\e'\*(FR~end of a buffer (string)\*(CD
-\*(FIr\*(FC*\*(FR~zero or more occurrences of \*(FIr\*(FR
-\*(FIr\*(FC+\*(FR~one or more occurrences of \*(FIr\*(FR
-\*(FIr\*(FC?\*(FR~zero or one occurrences of \*(FIr\*(FR
-\*(FIr\*(FC{\*(FIn\*(FC,\*(FIm\*(FC}\*(FR~\*(FIn\fP to \*(FIm\fP occurrences of \*(FIr\*(FR \*(CR(POSIX: see note below)\*(CD
-\*(FIr1\*(FC|\|\*(FIr2\*(FR~\*(FIr1\*(FR or \*(FIr2\*(FR
-.TE
-.sp .5
-.fi
-\*(CRThe \*(FIr\*(FC{\*(FIn\*(FC,\*(FIm\*(FC}\*(FR notation is called an
-\*(FIinterval expression\fP. POSIX mandates it for AWK regexps, but
-most \*(AKs don't implement it. \*(CBUse \*(FC\-\^\-re\-interval\*(FR
-or \*(FC\-\^\-posix\*(FR to enable
-this feature in \*(GK.\*(CX
-.EB "\s+2\f(HBREGULAR EXPRESSIONS\*(FR\s0"
-
-
-.BT
-
-.\" --- POSIX Character Classes (gawk)
-.ES
-.fi
-\*(CDIn regular expressions, within character ranges
-(\*(FC[\*(FR...\*(FC]\*(FR),
-the notation \*(FC[[:\*(FIclass\*(FC:]]\*(FR defines characters classes:
-.sp .5
-.TS
-center, tab(~);
-lp8 lp8 lp8 lp8.
-\*(FCalnum\*(FR~alphanumeric~\*(FClower\*(FR~lower-case
-\*(FCalpha\*(FR~alphabetic~\*(FCprint\*(FR~printable
-\*(FCblank\*(FR~space or tab~\*(FCpunct\*(FR~punctuation
-\*(FCcntrl\*(FR~control~\*(FCspace\*(FR~whitespace
-\*(FCdigit\*(FR~decimal~\*(FCupper\*(FR~upper-case
-\*(FCgraph\*(FR~non-spaces~\*(FCxdigit\*(FR~hexadecimal\*(CB
-.TE
-.fi
-.EB "\s+2\f(HBPOSIX CHARACTER CLASSES (\*(GK\f(HB)\*(FR\s0"
-
-.\" --- Records
-.ES
-.fi
-\*(CDNormally, records are separated by newline characters.
-Assigning values to the built-in variable \*(FCRS\*(FR
-controls how records are separated.
-If \*(FCRS\fP is any single character, that character separates records.
-\*(CLOtherwise, \*(FCRS\fP is a regular expression.
-\*(CR(Not \*(NK.)\*(CL
-Text in the input that matches this
-regular expression will separate the record.
-\*(CB\*(GK sets \*(FCRT\*(FR to the value of the
-input text that matched the regular expression.
-The value of \*(FCIGNORECASE\fP
-will also affect how records are separated when
-\*(FCRS\fP is a regular expression.\*(CD
-If \*(FCRS\fP is set to the null string,
-then records are separated by one or more blank lines.
-When \*(FCRS\fP is set to the null string,
-the newline character always acts as
-a field separator, in addition to whatever value
-\*(FCFS\fP may have.
-\*(CB\*(MK does not apply exceptional rules to \*(FCFS\fP
-when \*(FCRS = "\^"\fP.\*(CX
-.EB \s+2\f(HBRECORDS\*(FR\s0
-
-.\" --- Fields
-.ES
-.fi
-\*(CDAs each input record is read, \*(AK splits the record into
-\*(FIfields\*(FR, using the value of the \*(FCFS\fP
-variable as the field separator.
-If \*(FCFS\fP is a single character,
-fields are separated by that character.
-\*(CLIf \*(FCFS\fP is the null string,
-then each individual character becomes a separate field.\*(CD
-Otherwise, \*(FCFS\fP is expected to be a full regular expression.
-In the special case that \*(FCFS\fP
-is a single space, fields are separated
-by runs of spaces and/or tabs
-\*(CLand/or newlines\*(CD.
-Leading and trailing whitespace are ignored.
-\*(CBThe value of \*(FCIGNORECASE\fP
-will also affect how fields are split when
-\*(FCFS\fP is a regular expression.\*(CD
-.sp .5
-\*(CBIf the \*(FCFIELDWIDTHS\fP
-variable is set to a space separated list of numbers, each field is
-expected to have a fixed width, and \*(GK
-will split up the record using the specified widths.
-The value of \*(FCFS\fP is ignored.
-Assigning a new value to \*(FCFS\fP
-overrides the use of \*(FCFIELDWIDTHS\*(FR,
-and restores the default behavior.\*(CD
-.sp .5
-Each field in the input record may be referenced by its position,
-\*(FC$1\*(FR, \*(FC$2\*(FR and so on.
-\*(FC$0\fP is the whole record.
-Fields may also be assigned new values.
-.sp .5
-The variable \*(FCNF\fP
-is set to the total number of fields in the input record.
-.sp .5
-References to non-existent fields (i.e., fields after \*(FC$NF\*(FR)
-produce the null-string. However, assigning to a non-existent field
-(e.g., \*(FC$(NF+2) = 5\*(FR) will increase the value of
-\*(FCNF\*(FR, create any intervening fields with the null string as their value,
-and cause the value of \*(FC$0\fP
-to be recomputed with the fields being separated by the
-value of \*(FCOFS\*(FR.
-References to negative numbered fields cause a fatal error.
-Decreasing the value of \*(FCNF\fP causes the trailing fields to be lost
-\*(CR(not \*(NK).\*(CX
-.EB \s+2\f(HBFIELDS\*(FR\s0
-
-.BT
-
-.\" --- Pattern Elements
-.ES
-.fi
-\*(CDAWK patterns may be one of the following.
-.sp .5
-.nf
-	\*(FCBEGIN
-	END
-	\*(FIexpression	
-	pat1\*(FC,\*(FIpat2\*(FR
-.sp .5
-.fi
-\*(FCBEGIN\fP and \*(FCEND\fP are special patterns that provide start-up
-and clean-up actions respectively. They must have actions. There can
-be multiple \*(FCBEGIN\fP and \*(FCEND\fP rules; they are merged and
-executed as if there had just been one large rule. They may occur anywhere
-in a program, including different source files.
-.sp .5
-Expression patterns can be any expression, as described
-under \fHExpressions\fP.
-.sp .5
-The \*(FIpat1\*(FC,\*(FIpat2\*(FR pattern
-is called a \*(FIrange pattern\*(FR.
-It matches all input records starting with a record that matches
-\*(FIpat1\*(FR, and continuing until a record that matches
-\*(FIpat2\*(FR, inclusive.
-It does not combine with any other pattern expression.\*(CX
-.EB "\s+2\f(HBPATTERN ELEMENTS\*(FR\s0"
-
-
-.\" --- Action Statements
-.ES
-.nf
-\*(CD\*(FCif (\*(FIcondition\*(FC) \*(FIstatement\*(FR [ \*(FCelse\*(FI statement \*(FR]
-\*(FCwhile (\*(FIcondition\*(FC) \*(FIstatement \*(FR
-\*(FCdo \*(FIstatement  \*(FCwhile (\*(FIcondition\*(FC)\*(FR
-\*(FCfor (\*(FIexpr1\*(FC; \*(FIexpr2\*(FC; \*(FIexpr3\*(FC) \*(FIstatement\*(FR
-\*(FCfor (\*(FIvar \*(FCin\*(FI array\*(FC) \*(FIstatement\*(FR
-.ig
-\*(CB\*(FCabort\*(FR [ \*(FIexpression\*(FR ]\*(CD
-..
-\*(FCbreak\*(FR
-\*(FCcontinue\*(FR
-\*(FCdelete \*(FIarray\^\*(FC[\^\*(FIindex\^\*(FC]\*(FR
-\*(CL\*(FCdelete \*(FIarray\^\*(FR\*(CD
-\*(FCexit\*(FR [ \*(FIexpression\*(FR ]
-\*(FCnext\*(FR
-\*(CL\*(FCnextfile\*(FR \*(CR(not \*(MK)\*(CD
-\*(FC{ \*(FIstatements \*(FC}\*(CX
-.EB "\s+2\f(HBACTION STATEMENTS\*(FR\s0"
-
-
-
-.\" --- Escape Sequences
-.ES
-.fi
-\*(CDWithin strings constants (\*(FC"..."\fP) and regexp
-constants (\*(FC/.../\fP), escape sequences may be used to
-generate otherwise unprintable characters. This table lists
-the available escape sequences.
-.sp .5
-.ig
-\*(CB\*(FCPROCINFO\fP	T{
-elements of this array provide access to info
-about the running AWK program. See
-\*(AM for details.\*(CD
-T}
-..
-.TS
-center, tab(~);
-lp8 lp8 lp8 lp8.
-\*(FC\ea\fP~alert (bell)~\*(FC\er\fP~carriage return
-\*(FC\eb\fP~backspace~\*(FC\et\fP~horizontal tab
-\*(FC\ef\fP~form feed~\*(FC\ev\fP~vertical tab
-\*(FC\en\fP~newline~\*(FC\e\e\fP~backslash
-\*(FC\e\*(FIddd\*(FR~octal value \*(FIddd\fP~\*(CL\*(FC\ex\*(FIhh\*(FR~hex value \*(FIhh\fP\*(CD
-\*(FC\e"\fP~double quote~\*(FC\e/\fP~forward slash\*(CX
-.TE
-.EB "\s+2\f(HBESCAPE SEQUENCES\*(FR\s0"
-
-
-.BT
-
-.\" --- Variables
-.ES
-.fi
-.TS
-expand;
-l lw(2i).
-\*(FCARGC\fP	T{
-number of command line arguments.
-T}
-\*(CB\*(FCARGIND\fP	T{
-index in \*(FCARGV\fP of current data file.\*(CD
-T}
-\*(FCARGV\fP	T{
-array of command line arguments. Indexed from
-0 to \*(FCARGC\fP \- 1. Dynamically changing the
-contents of \*(FCARGV\fP can control the files used
-for data.
-T}
-\*(FCCONVFMT\fP	T{
-conversion format for numbers, default value
-is \*(FC"%.6g"\*(FR.
-T}
-\*(FCENVIRON\fP	T{
-array containing the the current environment.
-The array is indexed by the environment
-variables, each element being the value of
-that variable.
-T}
-\*(CB\*(FCERRNO\fP	T{
-contains a string describing the error when a
-redirection or read for \*(FCgetline\*(FR fails, or if
-\*(FCclose()\*(FR fails.
-T}
-\*(FCFIELDWIDTHS\fP	T{
-white-space separated list of fieldwidths. Used
-to parse the input into fields of fixed width,
-instead of the value of \*(FCFS\fP.\*(CD
-T}
-\*(FCFILENAME\fP	T{
-name of the current input file. If no files given
-on the command line, \*(FCFILENAME\fP is ``\-''.
-\*(FCFILENAME\fP is undefined inside the \*(FCBEGIN\fP rule
-(unless set by \*(FCgetline\fP).
-T}
-\*(FCFNR\fP	T{
-number of the input record in current input file.
-T}
-\*(FCFS\fP	T{
-input field separator, a space by default
-(see \fHFields\fP above).
-T}
-\*(CB\*(FCIGNORECASE\fP	T{
-if non-zero, all regular expression and string
-operations ignore case. \*(CRIn versions of \*(GK
-prior to 3.0, \*(FCIGNORECASE\fP only affected
-regular expression operations and \*(FCindex()\*(FR.\*(CD
-T}
-\*(FCNF\fP	T{
-number of fields in the current input record.
-T}
-\*(FCNR\fP	T{
-total number of input records seen so far.
-T}
-\*(FCOFMT\fP	T{
-output format for numbers, \*(FC"%.6g"\*(FR, by default.
-\*(CROld versions of \*(AK also used this for number
-to string conversion instead of \*(FCCONVFMT\fP.\*(CD
-T}
-\*(FCOFS\fP	T{
-output field separator, a space by default.
-T}
-\*(FCORS\fP	T{
-output record separator, a newline by default.
-T}
-\*(FCRS\fP	T{
-input record separator, a newline by default
-(see \fHRecords\fP above).
-T}
-\*(CB\*(FCRT\fP	T{
-record terminator. \*(GK sets \*(FCRT\fP to the input
-text that matched the character or regular
-expression specified by \*(FCRS\*(FR.\*(CD
-T}
-\*(FCRSTART\fP	T{
-index of the first character matched by
-\*(FCmatch()\*(FR; 0 if no match.
-T}
-\*(FCRLENGTH\fP	T{
-length of the string matched by \*(FCmatch()\*(FR;
-\-1 if no match.
-T}
-\*(FCSUBSEP\fP	T{
-character(s) used to separate multiple subscripts
-in array elements, by default \*(FC"\e034"\*(FR. (see
-\fHArrays\fP below).\*(CX
-T}
-.TE
-.EB \s+2\f(HBVARIABLES\*(FR\s0
-
-.BT
-
-.\" --- Arrays
-.ES
-.fi
-\*(CDAn arrays subscript is an expression between square brackets
-(\*(FC[ \*(FRand \*(FC]\*(FR).
-If the expression is a list
-\*(FC(\*(FIexpr\*(FC, \*(FIexpr \*(FC...)\*(FR,
-then the subscript is a string consisting of the
-concatenation of the (string) value of each expression,
-separated by the value of the \*(FCSUBSEP\fP variable.
-This simulates multi-dimensional
-arrays. For example:
-.nf
-.sp .5
-	\*(FCi = "A";\^ j = "B";\^ k = "C"
-	x[i, j, k] = "hello, world\en"\*(FR
-.sp .5
-.fi
-assigns \*(FC"hello, world\en"\*(FR to the element of the array
-\*(FCx\fP
-indexed by the string \*(FC"A\e034B\e034C"\*(FR. All arrays in AWK
-are associative, i.e., indexed by string values.
-.sp .5
-Use the special operator \*(FCin\fP in an \*(FCif\fP
-or \*(FCwhile\fP statement to see if a particular value is
-an array index.
-.sp .5
-.nf
-	\*(FCif (val in array)
-		print array[val]\*(FR
-.sp .5
-.fi
-If the array has multiple subscripts, use
-\*(FC(i, j) in array\*(FR.
-.sp .5
-Use the \*(FCin\fP construct in a \*(FCfor\fP
-loop to iterate over all the elements of an array.
-.sp .5
-Use the \*(FCdelete\fP statement to delete an
-element from an array.
-\*(CLSpecifying just the array name without a subscript in
-the \*(FCdelete\fP
-statement deletes the entire contents of an array.\*(CX
-.EB \s+2\f(HBARRAYS\*(FR\s0
-
-.\" --- Expressions
-.ES
-.fi
-\*(CDExpressions are used as patterns, for controlling conditional action
-statements, and to produce parameter values when calling functions.
-Expressions may also be used as simple statements,
-particularly if they have side-effects such as assignment.
-Expressions mix \*(FIoperands\fP and \*(FIoperators\fP. Operands are
-constants, fields, variables, array elements, and the return
-values from function calls (both built-in and user-defined).
-.sp .5
-Regexp constants (\*(FC/\*(FIpat\*(FC/\*(FR), when used as simple expressions,
-i.e., not used on the right-hand side of
-\*(FC~\fP and \*(FC!~\fP, or as arguments to the
-\*(CB\*(FCgensub()\fP,\*(CD
-\*(FCgsub()\fP,
-\*(FCmatch()\fP,
-\*(FCsplit()\fP,
-and
-\*(FCsub()\fP,
-functions, mean \*(FC$0 ~ /\*(FIpat\*(FC/\*(FR.
-.sp .5
-The AWK operators, in order of decreasing precedence, are
-.sp .5
-.fi
-.TS
-expand;
-l lw(1.8i).
-\*(FC(\&...)\*(FR	grouping
-\*(FC$\fP	field reference
-\*(FC++ \-\^\-\fP	T{
-increment and decrement,
-prefix and postfix
-T}
-\*(FC^\fP \*(CL\*(FC**\*(FR\*(CD	exponentiation
-\*(FC+ \- !\fP	unary plus, unary minus, and logical negation
-\*(FC* / %\fP	multiplication, division, and modulus
-\*(FC+ \-\fP	addition and subtraction
-\*(FIspace\fP	string concatenation
-\*(FC< >\fP	less than, greater than
-\*(FC<= >=\fP	less than or equal, greater than or equal
-\*(FC!= ==\fP	not equal, equal
-\*(FC~ !~\fP	regular expression match, negated match
-\*(FCin\fP	array membership
-\*(FC&&\fP	logical AND, short circuit
-\*(FC||\fP	logical OR, short circuit
-\*(FC?\^:\fP	in-line conditional expression
-.T&
-l s
-l lw(1.8i).
-\*(FC=\0+=\0\-=\0*=\0/=\0%=\0^=\0\*(CL**=\*(CD\fP
-	assignment operators\*(CX
-.TE
-.EB \s+2\f(HBEXPRESSIONS\*(FR\s0
-
-
-.BT
-
-.\" --- Conversions and Comparisons
-.ES
-.fi
-\*(CDVariables and fields may be (floating point) numbers, strings or both.
-Context determines how the value of a variable is interpreted. If used in
-a numeric expression, it will be treated as a number, if used as a string
-it will be treated as a string.
-.sp .5
-To force a variable to be treated as a number, add 0 to it; to force it
-to be treated as a string, concatenate it with the null string.
-.sp .5
-When a string must be converted to a number, the conversion is accomplished
-using \*(FIatof\*(FR(3).
-A number is converted to a string by using the value of \*(FCCONVFMT\fP
-as a format string for \*(FIsprintf\*(FR(3),
-with the numeric value of the variable as the argument.
-However, even though all numbers in AWK are floating-point,
-integral values are \*(FIalways\fP converted as integers.
-.sp .5
-Comparisons are performed as follows:
-If two variables are numeric, they are compared numerically.
-If one value is numeric and the other has a string value that is a
-``numeric string,'' then comparisons are also done numerically.
-Otherwise, the numeric value is converted to a string, and a string
-comparison is performed.
-Two strings are compared, of course, as strings.
-\*(CRAccording to the POSIX standard, even if two strings are
-numeric strings, a numeric comparison is performed. However, this is
-clearly incorrect, and none of the three free \*(AK\*(FRs do this.\*(CD
-.sp .5
-Note that string constants, such as \*(FC"57"\fP, are \*(FInot\fP
-numeric strings, they are string constants. The idea of ``numeric string''
-only applies to fields, \*(FCgetline\fP input,
-\*(FCFILENAME\*(FR, \*(FCARGV\fP elements, \*(FCENVIRON\fP
-elements and the elements of an array created by
-\*(FCsplit()\fP that are numeric strings.
-The basic idea is that \*(FIuser input\*(FR,
-and only user input, that looks numeric,
-should be treated that way.
-.sp .5
-Uninitialized variables have the numeric value 0 and the string value
-\*(FC"\^"\fP
-(the null, or empty, string).\*(CX
-.EB "\s+2\f(HBCONVERSIONS AND COMPARISONS\*(FR\s0"
-
-.ig
-.\" --- Localization
-.ES
-.nf
-.ce 100
-\*(CDThis
-section
-is
-under
-construction.
-.sp .5
-This
-section
-is
-under
-construction.\*(CB
-.ce 0
-.EB "\s+2\f(HBLOCALIZATION\*(FR\s0"
-..
-
-.ig
-.ps +2
-.ce 1
-\*(CD\fHISBN: 0-916151-97-2\*(FR
-.ps -2
-..
-
-.BT
-
-
-.\" --- Input Control
-.ES
-.fi
-.TS
-expand;
-l lw(1.8i).
-\*(CD\*(FCclose(\*(FIfile\*(FC)\*(FR	close input file or pipe.
-\*(FCgetline\fP	T{
-set \*(FC$0\fP from next input record;
-set \*(FCNF\*(FR, \*(FCNR\*(FR, \*(FCFNR\*(FR.
-T}
-\*(FCgetline < \*(FIfile\*(FR	set \*(FC$0\fP from next record of \*(FIfile\*(FR; set \*(FCNF\*(FR.
-\*(FCgetline \*(FIv\*(FR	T{
-set \*(FIv\fP from next input record;
-set \*(FCNR\*(FR, \*(FCFNR\*(FR.
-T}
-\*(FCgetline \*(FIv \*(FC< \*(FIfile\*(FR	set \*(FIv\fP from next record of \*(FIfile\*(FR.
-\*(FIcmd \*(FC| getline\*(FR	pipe into \*(FCgetline\*(FR; set \*(FC$0\*(FR, \*(FCNF\*(FR.
-\*(FIcmd \*(FC| getline \*(FIv\*(FR	pipe into \*(FCgetline\*(FR; set \*(FIv\*(FR.
-.TE
-.fi
-.in +.2i
-.ti -.2i
-\*(FCnext\fP
-.br
-stop processing the current input
-record. Read next input record and
-start over with the first pattern in the
-program. Upon end of the input data,
-execute any \*(FCEND\fP rule(s).
-.br
-.ti -.2i
-\*(CL\*(FCnextfile\fP
-.br
-stop processing the current input file.
-The next input record comes from the
-next input file. \*(FCFILENAME\fP \*(CBand
-\*(FCARGIND\fP\*(CL are updated, \*(FCFNR\fP is reset to 1,
-and processing starts over with the first
-pattern in the AWK program. Upon end
-of input data, execute any \*(FCEND\fP rule(s).
-\*(CREarlier versions of \*(GK used
-\*(FCnext file\*(FR, as two words. This
-generates a warning message and will
-eventually be removed. \*(CR\*(MK does not
-currently support \*(FCnextfile\*(FR.\*(CD
-.in -.2i
-.sp .5
-.fi
-\*(FCgetline\*(FR returns 0 on end of file, and \-1 on an
-error.\*(CX
-.EB "\s+2\f(HBINPUT CONTROL\*(FR\s0"
-
-.\" --- Output Control
-.ES
-.fi
-.in +.2i
-.ti -.2i
-\*(CD\*(FCclose(\*(FIfile\*(FC)\*(FR
-.br
-close output file or pipe.
-.ti -.2i
-\*(CL\*(FCfflush(\*(FR[\*(FIfile\^\*(FR]\*(FC)\*(FR
-.br
-flush any buffers associated
-with the open output file or pipe \*(FIfile\*(FR.\*(CD
-\*(CBIf \*(FIfile\fP is missing, then standard output is flushed.
-If \*(FIfile\fP is the null string, then all open output files and pipes
-are flushed \*(CR(not \*(NK)\*(CD.
-.ti -.2i
-\*(FCprint\fP
-.br
-print the current record. The output record is terminated
-with the value of \*(FCORS\fP.
-.ti -.2i
-\*(FCprint \*(FIexpr-list\*(FR
-.br
-print expressions. Each expression is separated
-by the value of \*(FCOFS\fP. The output record is
-terminated with the value of \*(FCORS\fP.
-.ti -.2i
-\*(FCprintf \*(FIfmt\*(FC, \*(FIexpr-list\*(FR
-.br
-format and print (see \fHPrintf Formats\fP below).
-.ti -.2i
-\*(FCsystem(\*(FIcmd\*(FC)\*(FR
-.br
-execute the command \*(FIcmd\*(FR,
-and return the exit status
-\*(CR(may not be available on non-POSIX systems)\*(CD.
-.sp .5
-.in -.2i
-I/O redirections may be used with both \*(FCprint\fP and \*(FCprintf\fP.
-.sp .5
-.in +.2i
-.ti -.2i
-\*(CD\*(FCprint "hello" > \*(FIfile\*(FR
-.br
-Print data to \*(FIfile\fP. The first time the file is written to, it
-will be truncated. Subsequent commands append data.
-.ti -.2i
-\*(FCprint "hello" >> \*(FIfile\*(FR
-.br
-Append data to \*(FIfile\fP. The previous contents of the file are not lost.
-.ti -.2i
-\*(FCprint "hello" | \*(FIcmd\*(FR
-.br
-Print data down a pipeline to \*(FIcmd\*(FR.\*(CX
-.in -.2i
-.EB "\s+2\f(HBOUTPUT CONTROL\*(FR\s0"
-
-.BT
-
-
-.\" --- Printf Formats
-.ES
-.fi
-\*(CDThe \*(FCprintf\fP statement and
-\*(FCsprintf()\fP function
-accept the following conversion specification formats:
-.sp .5
-.nf
-\*(FC%c\fP		an \s-1ASCII\s+1 character
-\*(FC%d\fP		a decimal number (the integer part)
-\*(FC%i\fP		a decimal number (the integer part)
-\*(FC%e\fP		a floating point number of the form
-		\*(FC[\-]d.dddddde[+\^\-]dd\*(FR
-\*(FC%E\fP		like \*(FC%e\fP, but use \*(FCE\fP instead of \*(FCe\*(FR
-\*(FC%f\fP		a floating point number of the form
-		\*(FC[\-]ddd.dddddd\*(FR
-\*(FC%g\fP		use \*(FC%e\fP or \*(FC%f\fP, whichever is shorter, with
-		nonsignificant zeros suppressed
-\*(FC%G\fP		like \*(FC%g\fP, but use \*(FC%E\fP instead of \*(FC%e\*(FR
-\*(FC%o\fP		an unsigned octal integer
-\*(FC%u\fP		an unsigned decimal integer
-\*(FC%s\fP		a character string
-\*(FC%x\fP		an unsigned hexadecimal integer
-\*(FC%X\fP		like \*(FC%x\fP, but use \*(FCABCDEF\fP for 10\(en15
-\*(FC%%\fP		A literal \*(FC%\fP; no argument is converted
-.sp .5
-.fi
-Optional, additional parameters may lie between the \*(FC%\fP
-and the control letter:
-.sp .5
-.TS
-expand;
-l lw(2.2i).
-\*(FC\-\fP	T{
-left-justify the expression within its field.
-T}
-\*(FIspace\fP	T{
-for numeric conversions, prefix positive values
-with a space and negative values with a
-minus sign.
-T}
-\*(FC+\fP	T{
-used before the \*(FIwidth\fP modifier means to always
-supply a sign for numeric conversions, even if
-the data to be formatted is positive. The \*(FC+\fP
-overrides the space modifier.
-T}
-\*(FC#\fP	T{
-use an ``alternate form'' for some control letters.
-T}
- \*(FC%o\*(FR	T{
-supply a leading zero.
-T}
- \*(FC%x\*(FR, \*(FC%X\*(FR	T{
-supply a leading \*(FC0x\*(FR or \*(FC0X\*(FR for a nonzero result.
-T}
- \*(FC%e\*(FR, \*(FC%E\*(FR, \*(FC%f\*(FR	T{
-the result always has a decimal point.
-T}
- \*(FC%g\*(FR, \*(FC%G\*(FR	T{
-trailing zeros are not removed.
-T}
-\*(FC0\fP	T{
-a leading zero acts as a flag, indicating output
-should be padded with zeroes instead of spaces.
-This applies even to non-numeric output formats.
-Only has an effect when the field width is wider
-than the value to be printed.
-T}
-\*(FIwidth\fP	T{
-pad the field to this width. The field is normally
-padded with spaces. If the \*(FC0\fP flag has been used,
-pad with zeroes.
-T}
-\*(FC.\fP\*(FIprec\fP	T{
-precision.
-The meaning varies by control letter:
-T}
- \*(FC%d\*(FR, \*(FC%o\*(FR, \*(FC%i\*(FR,
- \*(FC%u\*(FR, \*(FC%x\*(FR, \*(FC%X\fP	T{
-the minimum number of digits to print.
-T}
- \*(FC%e\*(FR, \*(FC%E\*(FR, \*(FC%f\*(FR	T{
-the number of digits to print to the right of the decimal point.
-T}
- \*(FC%g\*(FR, \*(FC%G\fP	T{
-the maximum number of significant digits.
-T}
- \*(FC%s\fP	T{
-the maximum number of characters to print.
-T}
-.TE
-.sp .5
-.fi
-The dynamic \*(FIwidth\fP and \*(FIprec\fP capabilities of the ANSI C
-\*(FCprintf()\fP routines are supported.
-A \*(FC*\fP in place of either the \*(FIwidth\fP or \*(FIprec\fP
-specifications will cause their values to be taken from
-the argument list to \*(FCprintf\fP or \*(FCsprintf()\*(FR.\*(CX
-.EB "\s+2\f(HBPRINTF FORMATS\*(FR\s0"
-
-
-
-.BT
-
-.\" --- Special Filenames
-.ES
-.fi
-\*(CDWhen doing I/O redirection from either \*(FCprint\fP
-or \*(FCprintf\fP into a file or via \*(FCgetline\fP
-from a file, all three implementations of \*(FCawk\fP
-recognize certain special filenames internally. These filenames
-allow access to open file descriptors inherited from the
-parent process (usually the shell).
-These filenames may also be used on the command line to name data files.
-The filenames are:
-.sp .5
-.TS
-expand;
-l lw(2i).
-\*(FC"\-"\fP	standard input
-\*(FC/dev/stdin\fP	standard input \*(CR(not \*(MK)\*(CD
-\*(FC/dev/stdout\fP	standard output
-\*(FC/dev/stderr\fP	standard error output
-.TE
-.sp .5
-.fi
-\*(CBThe following names are specific to \*(GK.
-.sp .5
-.TS
-expand;
-l lw(2i).
-\*(FC/dev/fd/\^\*(FIn\*(FR	T{
-file associated with the open file descriptor \*(FIn\*(FR
-T}
-.TE
-.sp .5
-.fi
-Other special filenames provide access to information about the running
-\*(FCgawk\fP process.
-Reading from these files returns a single record.
-The filenames and what they return are:\*(FR
-.sp .5
-.TS
-expand;
-l lw(2i).
-\*(FC/dev/pid\fP	process ID of current process
-\*(FC/dev/ppid\fP	parent process ID of current process
-\*(FC/dev/pgrpid\fP	process group ID of current process
-\*(FC/dev/user\fP	T{
-.nf
-a single newline-terminated record.
-The fields are separated with spaces.
-\*(FC$1\fP is the return value of \*(FIgetuid\*(FR(2),
-\*(FC$2\fP is the return value of \*(FIgeteuid\*(FR(2),
-\*(FC$3\fP is the return value of \*(FIgetgid\*(FR(2) , and
-\*(FC$4\fP is the return value of \*(FIgetegid\*(FR(2).
-.fi
-Any additional fields are the group IDs returned
-by \*(FIgetgroups\*(FR(2). Multiple groups may not be
-supported on all systems.
-T}
-.TE
-.sp .5
-.fi
-.ig
-\*(CRThese filenames are now obsolete.
-Use the \*(FCPROCINFO\fP array to obtain the information they provide.\*(CL
-..
-.\" BEGIN FOR 3.0.x
-\*(CRThese filenames will become obsolete in \*(GK 3.1.
-Be aware that you will have to change your programs.\*(CL
-.\" END FOR 3.0.x
-.EB "\s+2\f(HBSPECIAL FILENAMES\*(FR\s0"
-
-
-
-
-.\" --- Builtin Numeric Functions
-.ES
-.fi
-.TS
-expand;
-l lw(2i).
-\*(CD\*(FCatan2(\*(FIy\*(FC, \*(FIx\*(FC)\*(FR	the arctangent of \*(FIy/x\fP in radians.
-\*(FCcos(\*(FIexpr\*(FC)\*(FR	the cosine of \*(FIexpr\fP, which is in radians.
-\*(FCexp(\*(FIexpr\*(FC)\*(FR	the exponential function (\*(FIe \*(FC^ \*(FIx\*(FR).
-\*(FCint(\*(FIexpr\*(FC)\*(FR	truncates to integer.
-\*(FClog(\*(FIexpr\*(FC)\*(FR	the natural logarithm function (base \*(FIe\^\*(FR).
-\*(FCrand()\fP	a random number between 0 and 1.
-\*(FCsin(\*(FIexpr\*(FC)\*(FR	the sine of \*(FIexpr\fP, which is in radians.
-\*(FCsqrt(\*(FIexpr\*(FC)\*(FR	the square root function.
-\&\*(FCsrand(\*(FR[\*(FIexpr\^\*(FR]\*(FC)\*(FR	T{
-uses \*(FIexpr\fP as a new seed for the random number
-generator. If no \*(FIexpr\fP, the time of day is used.
-Returns previous seed for the random number
-generator.\*(CX
-T}
-.TE
-.EB "\s+2\f(HBNUMERIC FUNCTIONS\*(FR\s0"
-
-
-.BT
-
-
-.\" --- Builtin String Functions
-.ES
-.fi
-.in +.2i
-.ti -.2i
-\*(CB\*(FCgensub(\*(FIr\*(FC, \*(FIs\*(FC, \*(FIh \*(FR[\*(FC, \*(FIt\*(FR]\*(FC)\*(FR
-.br
-search the target string
-\*(FIt\fP for matches of the regular expression \*(FIr\*(FR. If
-\*(FIh\fP is a string beginning with \*(FCg\fP or \*(FCG\*(FR,
-replace all matches of \*(FIr\fP with \*(FIs\*(FR. Otherwise, \*(FIh\fP
-is a number indicating which match of \*(FIr\fP to replace. If no
-\*(FIt\fP is supplied, \*(FC$0\fP is used instead. Within the
-replacement text \*(FIs\*(FR, the sequence \*(FC\e\*(FIn\*(FR,
-where \*(FIn\fP is a digit from 1 to 9, may be used to indicate just
-the text that matched the \*(FIn\*(FRth parenthesized subexpression.
-The sequence \*(FC\e0\fP represents the entire matched text, as does
-the character \*(FC&\*(FR. Unlike \*(FCsub()\fP and \*(FCgsub()\*(FR,
-the modified string is returned as the result of the function,
-and the original target string is \*(FInot\fP changed.\*(CD
-.ti -.2i
-\*(FCgsub(\*(FIr\*(FC, \*(FIs \*(FR[\*(FC, \*(FIt\*(FR]\*(FC)\*(FR
-.br
-for each substring matching the
-regular expression \*(FIr\fP in the string \*(FIt\*(FR, substitute the
-string \*(FIs\*(FR, and return the number of substitutions. If
-\*(FIt\fP is not supplied, use \*(FC$0\*(FR. An \*(FC&\fP in the
-replacement text is replaced with the text that was actually matched.
-Use \*(FC\e&\fP to get a literal \*(FC&\*(FR. See \*(AM
-for a fuller discussion of the rules for \*(FC&\*(FR's and backslashes
-in the replacement text of \*(CB\*(FCgensub()\*(FR,\*(CD \*(FCsub()\*(FR
-and \*(FCgsub()\*(FR
-.ti -.2i
-\*(FCindex(\*(FIs\*(FC, \*(FIt\*(FC)\*(FR
-.br
-returns the index of the string
-\*(FIt\fP in the string \*(FIs\*(FR, or 0 if \*(FIt\fP is not present.
-.ti -.2i
-\*(FClength(\*(FR[\*(FIs\*(FR]\*(FC)\*(FR
-.br
-returns the length of the string
-\*(FIs\*(FR, or the length of \*(FC$0\fP if \*(FIs\fP is not supplied.
-.ti -.2i
-\*(FCmatch(\*(FIs\*(FC, \*(FIr\*(FC)\*(FR
-.br
-returns the position in
-\*(FIs\fP where the regular expression \*(FIr\fP occurs, or 0 if
-\*(FIr\fP is not present, and sets the values of variables
-\*(FCRSTART\fP
-and \*(FCRLENGTH\*(FR.
-.ti -.2i
-\*(FCsplit(\*(FIs\*(FC, \*(FIa \*(FR[\*(FC, \*(FIr\*(FR]\*(FC)\*(FR
-.br
-splits the string
-\*(FIs\fP into the array \*(FIa\fP using the regular expression \*(FIr\*(FR,
-and returns the number of fields. If \*(FIr\fP is omitted, \*(FCFS\fP
-is used instead. The array \*(FIa\fP is cleared first.
-Splitting behaves identically to field splitting.
-(See \fHFields\fP, above.)
-.ti -.2i
-\*(FCsprintf(\*(FIfmt\*(FC, \*(FIexpr-list\*(FC)\*(FR
-.br
-prints \*(FIexpr-list\fP
-according to \*(FIfmt\*(FR, and returns the resulting string.
-.ti -.2i
-\*(FCsub(\*(FIr\*(FC, \*(FIs \*(FR[\*(FC, \*(FIt\*(FR]\*(FC)\*(FR
-.br
-just like
-\*(FCgsub()\*(FR, but only the first matching substring is replaced.
-.ti -.2i
-\*(FCsubstr(\*(FIs\*(FC, \*(FIi \*(FR[\*(FC, \*(FIn\*(FR]\*(FC)\*(FR
-.br
-returns the at most
-\*(FIn\*(FR-character substring of \*(FIs\fP starting at \*(FIi\*(FR.
-If \*(FIn\fP is omitted, the rest of \*(FIs\fP is used.
-.ti -.2i
-\*(FCtolower(\*(FIstr\*(FC)\*(FR
-.br
-returns a copy of the string \*(FIstr\*(FR,
-with all the upper-case characters in \*(FIstr\fP translated to their
-corresponding lower-case counterparts. Non-alphabetic characters are
-left unchanged.
-.ti -.2i
-\*(FCtoupper(\*(FIstr\*(FC)\*(FR
-.br
-returns a copy of the string \*(FIstr\*(FR,
-with all the lower-case characters in \*(FIstr\fP translated to their
-corresponding upper-case counterparts. Non-alphabetic characters are
-left unchanged.\*(CX
-.in -.2i
-.EB "\s+2\f(HBSTRING FUNCTIONS\*(FR\s0"
-
-
-
-.BT
-
-
-.\" --- Builtin Time Functions
-.ES
-.fi
-\*(CD\*(GK
-provides the following functions for obtaining time stamps and
-formatting them.
-.sp .5
-.fi
-.in +.2i
-.ig
-.ti -.2i
-\*(FCmktime(\*(FIdatespec\*(FC)\*(FR
-.br
-turns \*(FIdatespec\fP into a time
-stamp of the same form as returned by \*(FCsystime()\*(FR.
-The \*(FIdatespec\fP is a string of the form
-\*(FC"\*(FIYYYY MM DD HH MM SS\*(FC"\*(FR.
-..
-.ti -.2i
-\*(FCstrftime(\*(FR[\*(FIformat \*(FR[\*(FC, \*(FItimestamp\*(FR]]\*(FC)\*(FR
-.br
-formats \*(FItimestamp\fP
-according to the specification in \*(FIformat\*(FR. The
-\*(FItimestamp\fP should be of the same form as returned by
-\*(FCsystime()\*(FR.
-If \*(FItimestamp\fP is missing, the current time of day is used. If
-\*(FIformat\fP is missing, a default format equivalent to the output
-of \*(FIdate\*(FR(1) will be used.
-.ti -.2i
-\*(FCsystime()\fP
-.br
-returns the current time of day as the number of
-seconds since the Epoch.\*(CB
-.in -.2i
-.EB "\s+2\f(HBTIME FUNCTIONS (\*(GK\f(HB)\*(FR\s0"
-
-
-
-.\" --- User-defined Functions
-.ES
-.fi
-\*(CDFunctions in AWK are defined as follows:
-.sp .5
-.nf
-	\*(FCfunction \*(FIname\*(FC(\*(FIparameter list\*(FC)
-	{
-		\*(FIstatements
-	\*(FC}\*(FR
-.sp .5
-.fi
-Functions are executed when they are called from within expressions
-in either patterns or actions. Actual parameters supplied in the function
-call instantiate the formal parameters declared in the function.
-Arrays are passed by reference, other variables are passed by value.
-.sp .5
-Local variables are declared as extra parameters
-in the parameter list. The convention is to separate local variables from
-real parameters by extra spaces in the parameter list. For example:
-.sp .5
-.nf
-	\*(FC# a & b are local
-	function f(p, q,     a, b)
-	{
-		\&.....
-	}
-.sp .3
-	/abc/	{ ... ; f(1, 2) ; ... }\*(FR
-.fi
-.sp .5
-The left parenthesis in a function call is required
-to immediately follow the function name
-without any intervening white space.
-This is to avoid a syntactic ambiguity with the concatenation operator.
-This restriction does not apply to the built-in functions.
-.sp .5
-Functions may call each other and may be recursive.
-Function parameters used as local variables are initialized
-to the null string and the number zero upon function invocation.
-.sp .5
-Use \*(FCreturn\fP to return a value from a function. The return value
-is undefined if no value is provided, or if the function returns by
-``falling off'' the end.
-.sp .5
-\*(CLThe word
-\*(FCfunc\fP
-may be used in place of
-\*(FCfunction\*(FR.
-\*(CRNote: This usage is deprecated.\*(CX
-.EB "\s+2\f(HBUSER-DEFINED FUNCTIONS\*(FR\s0"
-
-
-
-.\" --- Bug Reports
-.ES
-.fi
-\*(CDIf you find a bug in this reference card, please report it via electronic
-mail to \*(FCarnold@gnu.org\*(FR.\*(CX
-.EB "\s+2\f(HBBUG REPORTS\*(FR\s0"
-
-.BT
-
-.\" --- Environment Variables
-.ES
-.fi
-\*(CDThe environment variable \*(FCAWKPATH\fP specifies a search path to use
-when finding source files named with the \*(FC\-f\fP
-option.
-The default path is
-\*(FC".:/usr/local/share/awk"\*(FR,
-if this variable does not exist.
-(The actual directory may vary,
-depending upon how \*(GK was built and installed.)
-If a file name given to the \*(FC\-f\fP option contains a ``/'' character,
-no path search is performed.
-.sp .5
-If \*(FCPOSIXLY_CORRECT\fP exists in the environment, then \*(GK
-behaves exactly as if \*(FC\-\^\-posix\fP had been specified on the
-command line.\*(CB
-.EB "\s+2\f(HBENVIRONMENT VARIABLES (\*(GK\f(HB)\*(FR\s0"
-
-.\" --- Historical Features
-.ES
-.fi
-\*(CD\*(GK supports two features of historical AWK implementations.
-First, it is possible to call the \*(FClength()\fP
-built-in function not only with no argument, but even without parentheses.
-This feature is marked as ``deprecated'' in the POSIX standard, and \*(GK
-will issue a warning about its use if \*(FC\-\^\-lint\fP
-is specified on the command line.
-.sp .5
-The other feature is the use of \*(FCcontinue\fP
-or \*(FCbreak\fP statements outside the body of a
-\*(FCwhile\*(FR, \*(FCfor\*(FR, or \*(FCdo\fP loop.
-Historical AWK implementations have treated such usage as
-equivalent to the \*(FCnext\fP statement.
-\*(GK will support this usage if \*(FC\-\^\-traditional\fP
-has been specified.\*(CB
-.EB "\s+2\f(HBHISTORICAL FEATURES (\*(GK\f(HB)\*(FR\s0"
-
-
-.\" --- FTP Information
-.ES
-.nf
-\*(CDHost: \*(FCgnudist.gnu.org\*(FR
-File: \*(FC/gnu/gawk/gawk-3.0.6.tar.gz\fP
-.in +.2i
-.fi
-GNU \*(AK (\*(GK). There may be a later version.
-.in -.2i
-.nf
-.sp .5
-Host: \*(FCnetlib.bell-labs.com\*(FR
-File: \*(FC/netlib/research/awk.bundle.gz\fP
-.in +.2i
-.fi
-\*(NK. This version requires an ANSI C compiler;
-GCC (the GNU C compiler) works well.
-.in -.2i
-.nf
-.sp .5
-Host: \*(FCftp.whidbey.net\*(FR
-File: \*(FC/pub/brennan/mawk1.3.3.tar.gz\fP
-.in +.2i
-.fi
-Michael Brennan's \*(MK. There may be a newer version.\*(CX
-.in -.2i
-.EB "\s+2\f(HBFTP INFORMATION\*(FR\s0"
-
-.\" --- Copying Permissions
-.ES
-.fi
-\*(CDCopyright \(co 1996-2000 Free Software Foundation, Inc.
-.sp .5   
-Permission is granted to make and distribute verbatim copies of this
-reference card provided the copyright notice and this permission notice
-are preserved on all copies.
-.sp .5
-Permission is granted to copy and distribute modified versions of this
-reference card under the conditions for verbatim copying, provided that
-the entire resulting derived work is distributed under the terms of a
-permission notice identical to this one.
-.sp .5
-Permission is granted to copy and distribute translations of this
-reference card into another language, under the above conditions for
-modified versions, except that this permission notice may be stated in a
-translation approved by the Foundation.\*(CX
-.EB "\s+2\f(HBCOPYING PERMISSIONS\*(FR\s0"
-.BT
diff --git a/contrib/awk/doc/awkforai.txt b/contrib/awk/doc/awkforai.txt
deleted file mode 100644
index 3fca3204324f1..0000000000000
--- a/contrib/awk/doc/awkforai.txt
+++ /dev/null
@@ -1,150 +0,0 @@
-Draft for ACM SIGPLAN Patterns (Language Trends)
-
-1996
-
-Why GAWK for AI?
-
-Ronald P. Loui
-
-Most people are surprised when I tell them what language we use in our
-undergraduate AI programming class.  That's understandable.  We use
-GAWK.  GAWK, Gnu's version of Aho, Weinberger, and Kernighan's old
-pattern scanning language isn't even viewed as a programming language by
-most people.  Like PERL and TCL, most prefer to view it as a "scripting
-language."  It has no objects; it is not functional; it does no built-in
-logic programming.  Their surprise turns to puzzlement when I confide
-that (a) while the students are allowed to use any language they want;
-(b) with a single exception, the best work consistently results from
-those working in GAWK.  (footnote:  The exception was a PASCAL
-programmer who is now an NSF graduate fellow getting a Ph.D. in
-mathematics at Harvard.) Programmers in C, C++, and LISP haven't even
-been close (we have not seen work in PROLOG or JAVA).
-
-Why GAWK?
-
-There are some quick answers that have to do with the pragmatics of
-undergraduate programming.  Then there are more instructive answers that
-might be valuable to those who debate programming paradigms or to those
-who study the history of AI languages.  And there are some deep
-philosophical answers that expose the nature of reasoning and symbolic
-AI.  I think the answers, especially the last ones, can be even more
-surprising than the observed effectiveness of GAWK for AI.
-
-First it must be confessed that PERL programmers can cobble together AI
-projects well, too.  Most of GAWK's attractiveness is reproduced in
-PERL, and the success of PERL forebodes some of the success of GAWK.
-Both are powerful string-processing languages that allow the programmer
-to exploit many of the features of a UNIX environment.  Both provide
-powerful constructions for manipulating a wide variety of data in
-reasonably efficient ways.  Both are interpreted, which can reduce
-development time.  Both have short learning curves.  The GAWK manual can
-be consumed in a single lab session and the language can be mastered by
-the next morning by the average student.  GAWK's automatic
-initialization, implicit coercion, I/O support and lack of pointers
-forgive many of the mistakes that young programmers are likely to make.
-Those who have seen C but not mastered it are happy to see that GAWK
-retains some of the same sensibilities while adding what must be
-regarded as spoonsful of syntactic sugar.  Some will argue that
-PERL has superior functionality, but for quick AI applications, the
-additional functionality is rarely missed.  In fact, PERL's terse syntax
-is not friendly when regular expressions begin to proliferate and
-strings contain fragments of HTML, WWW addresses, or shell commands.
-PERL provides new ways of doing things, but not necessarily ways of
-doing new things.
-
-In the end, despite minor difference, both PERL and GAWK minimize
-programmer time.  Neither really provides the programmer the setting in
-which to worry about minimizing run-time.
-
-There are further simple answers.  Probably the best is the fact that
-increasingly, undergraduate AI programming is involving the Web.  Oren
-Etzioni (University of Washington, Seattle) has for a while been arguing
-that the "softbot" is replacing the mechanical engineers' robot as the
-most glamorous AI testbed.  If the artifact whose behavior needs to be
-controlled in an intelligent way is the software agent, then a language
-that is well-suited to controlling the software environment is the
-appropriate language.  That would imply a scripting language.  If the
-robot is KAREL, then the right language is "turn left; turn right." If
-the robot is Netscape, then the right language is something that can
-generate "netscape -remote 'openURL(http://cs.wustl.edu/~loui)'" with
-elan.
-
-Of course, there are deeper answers.  Jon Bentley found two pearls in
-GAWK:  its regular expressions and its associative arrays.  GAWK asks
-the programmer to use the file system for data organization and the
-operating system for debugging tools and subroutine libraries.  There is
-no issue of user-interface.  This forces the programmer to return to the
-question of what the program does, not how it looks.  There is no time
-spent programming a binsort when the data can be shipped to /bin/sort
-in no time.  (footnote:  I am reminded of my IBM colleague Ben Grosof's
-advice for Palo Alto:  Don't worry about whether it's highway 101 or 280.
-Don't worry if you have to head south for an entrance to go north.  Just
-get on the highway as quickly as possible.)
-
-There are some similarities between GAWK and LISP that are illuminating.
-Both provided a powerful uniform data structure (the associative array
-implemented as a hash table for GAWK and the S-expression, or list of
-lists, for LISP).  Both were well-supported in their environments (GAWK
-being a child of UNIX, and LISP being the heart of lisp machines).  Both
-have trivial syntax and find their power in the programmer's willingness
-to use the simple blocks to build a complex approach.
-
-Deeper still, is the nature of AI programming.  AI is about
-functionality and exploratory programming.  It is about bottom-up design
-and the building of ambitions as greater behaviors can be demonstrated.
-Woe be to the top-down AI programmer who finds that the bottom-level
-refinements, "this subroutine parses the sentence," cannot actually be
-implemented.  Woe be to the programmer who perfects the data structures
-for that heapsort when the whole approach to the high-level problem
-needs to be rethought, and the code is sent to the junkheap the next day.
-
-AI programming requires high-level thinking.  There have always been a few
-gifted programmers who can write high-level programs in assembly language.
-Most however need the ambient abstraction to have a higher floor.  
-
-Now for the surprising philosophical answers.  First, AI has discovered
-that brute-force combinatorics, as an approach to generating intelligent
-behavior, does not often provide the solution.  Chess, neural nets, and
-genetic programming show the limits of brute computation.  The
-alternative is clever program organization.  (footnote: One might add
-that the former are the AI approaches that work, but that is easily
-dismissed:  those are the AI approaches that work in general, precisely
-because cleverness is problem-specific.)  So AI programmers always want
-to maximize the content of their program, not optimize the efficiency
-of an approach.  They want minds, not insects.  Instead of enumerating
-large search spaces, they define ways of reducing search, ways of
-bringing different knowledge to the task.  A language that maximizes
-what the programmer can attempt rather than one that provides tremendous
-control over how to attempt it, will be the AI choice in the end.
-
-Second, inference is merely the expansion of notation.  No matter whether
-the logic that underlies an AI program is fuzzy, probabilistic, deontic,
-defeasible, or deductive, the logic merely defines how strings can be 
-transformed into other strings.  A language that provides the best
-support for string processing in the end provides the best support for
-logic, for the exploration of various logics, and for most forms of
-symbolic processing that AI might choose to call "reasoning" instead of
-"logic."  The implication is that PROLOG, which saves the AI programmer
-from having to write a unifier, saves perhaps two dozen lines of GAWK
-code at the expense of strongly biasing the logic and representational
-expressiveness of any approach.
-
-I view these last two points as news not only to the programming language
-community, but also to much of the AI community that has not reflected on
-the past decade's lessons.
-
-In the puny language, GAWK, which Aho, Weinberger, and Kernighan thought
-not much more important than grep or sed, I find lessons in AI's trends,
-AI's history, and the foundations of AI.  What I have found not only
-surprising but also hopeful, is that when I have approached the AI
-people who still enjoy programming, some of them are not the least bit
-surprised.
-
-
-R. Loui (loui@ai.wustl.edu) is Associate Professor of Computer Science,
-at Washington University in St. Louis.  He has published in AI Journal,
-Computational Intelligence, ACM SIGART, AI Magazine, AI and Law, the ACM
-Computing Surveys Symposium on AI, Cognitive Science, Minds and
-Machines, Journal of Philosophy, and is on this year's program
-committees for AAAI (National AI conference) and KR (Knowledge
-Representation and Reasoning).
diff --git a/contrib/awk/doc/cardfonts b/contrib/awk/doc/cardfonts
deleted file mode 100644
index 5529ba9891aa9..0000000000000
--- a/contrib/awk/doc/cardfonts
+++ /dev/null
@@ -1,37 +0,0 @@
-.\" AWK Reference Card --- Arnold Robbins, arnold@gnu.org
-.\" cardfonts --- this file sets the fonts to use for the reference card
-.\"
-.\" Copyright (C) 1996 Free Software Foundation, Inc.
-.\" 
-.\" Permission is granted to make and distribute verbatim copies of
-.\" this reference card provided the copyright notice and this permission
-.\" notice are preserved on all copies.
-.\" 
-.\" Permission is granted to process this file through troff and print the
-.\" results, provided the printed document carries copying permission
-.\" notice identical to this one except for the removal of this paragraph
-.\" (this paragraph not being relevant to the printed reference card).
-.\" 
-.\" Permission is granted to copy and distribute modified versions of this
-.\" reference card under the conditions for verbatim copying, provided that
-.\" the entire resulting derived work is distributed under the terms of a
-.\" permission notice identical to this one.
-.\" 
-.\" Permission is granted to copy and distribute translations of this
-.\" reference card into another language, under the above conditions for
-.\" modified versions, except that this permission notice may be stated in
-.\" a translation approved by the Foundation.
-.\"
-.ig
-Strings for inline font change.
-FR - font roman
-FI - font italic
-FC - font courier
-..
-.ds FR \fR
-.ds FI \fI
-.ds FC \f(CB
-.ds RN Times Roman
-.ds IN Times Italic
-.ds CN Courier Bold
-.ds AM \fIThe GNU Awk User's Guide\fP
diff --git a/contrib/awk/doc/colors b/contrib/awk/doc/colors
deleted file mode 100644
index 933d25efedfcb..0000000000000
--- a/contrib/awk/doc/colors
+++ /dev/null
@@ -1,39 +0,0 @@
-.\" AWK Reference Card --- Arnold Robbins, arnold@gnu.org
-.\" This file sets the colors to use.
-.\"
-.\" Copyright (C) 1996,97,99 Free Software Foundation, Inc.
-.\" 
-.\" Permission is granted to make and distribute verbatim copies of
-.\" this reference card provided the copyright notice and this permission
-.\" notice are preserved on all copies.
-.\" 
-.\" Permission is granted to process this file through troff and print the
-.\" results, provided the printed document carries copying permission
-.\" notice identical to this one except for the removal of this paragraph
-.\" (this paragraph not being relevant to the printed reference card).
-.\" 
-.\" Permission is granted to copy and distribute modified versions of this
-.\" reference card under the conditions for verbatim copying, provided that
-.\" the entire resulting derived work is distributed under the terms of a
-.\" permission notice identical to this one.
-.\" 
-.\" Permission is granted to copy and distribute translations of this
-.\" reference card into another language, under the above conditions for
-.\" modified versions, except that this permission notice may be stated in
-.\" a translation approved by the Foundation.
-.\"
-.ig
-Strings for inline color change.
-CR - color red
-CG - color green
-CL - color light blue
-CB - color blue
-CD - color dark, i.e. black
-CX - color boX, i.e. for the surrounding boxes (red for now)
-..
-.ds CR \X'ps: exec 0 .96 .65 0 setcmykcolor'
-.ds CG \X'ps: exec 1.0 0 .51 .43 setcmykcolor'
-.ds CL \X'ps: exec .69 .34 0 0 setcmykcolor'
-.ds CB \X'ps: exec 1 .72 0 .06 setcmykcolor'
-.ds CD \X'ps: exec 1 1 1 1 setcmykcolor'
-.ds CX \*(CG
diff --git a/contrib/awk/doc/gawk.1 b/contrib/awk/doc/gawk.1
deleted file mode 100644
index 6f07cfab136e8..0000000000000
--- a/contrib/awk/doc/gawk.1
+++ /dev/null
@@ -1,2657 +0,0 @@
-.ds PX \s-1POSIX\s+1
-.ds UX \s-1UNIX\s+1
-.ds AN \s-1ANSI\s+1
-.ds GN \s-1GNU\s+1
-.ds AK \s-1AWK\s+1
-.if !\n(.g \{\
-.	if !\w|\*(lq| \{\
-.		ds lq ``
-.		if \w'\(lq' .ds lq "\(lq
-.	\}
-.	if !\w|\*(rq| \{\
-.		ds rq ''
-.		if \w'\(rq' .ds rq "\(rq
-.	\}
-.\}
-.TH GAWK 1 "May 17 2000" "Free Software Foundation" "Utility Commands"
-.SH NAME
-gawk \- pattern scanning and processing language
-.SH SYNOPSIS
-.B gawk
-[ \*(PX or \*(GN style options ]
-.B \-f
-.I program-file
-[
-.B \-\^\-
-] file .\|.\|.
-.br
-.B gawk
-[ \*(PX or \*(GN style options ]
-[
-.B \-\^\-
-]
-.I program-text
-file .\|.\|.
-.SH DESCRIPTION
-.I Gawk
-is the \*(GN Project's implementation of the \*(AK programming language.
-It conforms to the definition of the language in
-the \*(PX 1003.2 Command Language And Utilities Standard.
-This version in turn is based on the description in
-.IR "The AWK Programming Language" ,
-by Aho, Kernighan, and Weinberger,
-with the additional features found in the System V Release 4 version
-of \*(UX
-.IR awk .
-.I Gawk
-also provides more recent Bell Labs
-.I awk
-extensions, and some \*(GN-specific extensions.
-.PP
-The command line consists of options to
-.I gawk
-itself, the \*(AK program text (if not supplied via the
-.B \-f
-or
-.B \-\^\-file
-options), and values to be made
-available in the
-.B ARGC
-and
-.B ARGV
-pre-defined \*(AK variables.
-.SH OPTION FORMAT
-.PP
-.I Gawk
-options may be either the traditional \*(PX one letter options,
-or the \*(GN style long options.  \*(PX options start with a single \*(lq\-\*(rq,
-while long options start with \*(lq\-\^\-\*(rq.
-Long options are provided for both \*(GN-specific features and
-for \*(PX mandated features.
-.PP
-Following the \*(PX standard,
-.IR gawk -specific
-options are supplied via arguments to the
-.B \-W
-option.  Multiple
-.B \-W
-options may be supplied
-Each
-.B \-W
-option has a corresponding long option, as detailed below.
-Arguments to long options are either joined with the option
-by an
-.B =
-sign, with no intervening spaces, or they may be provided in the
-next command line argument.
-Long options may be abbreviated, as long as the abbreviation
-remains unique.
-.SH OPTIONS
-.PP
-.I Gawk
-accepts the following options.
-.TP
-.PD 0
-.BI \-F " fs"
-.TP
-.PD
-.BI \-\^\-field-separator " fs"
-Use
-.I fs
-for the input field separator (the value of the
-.B FS
-predefined
-variable).
-.TP
-.PD 0
-\fB\-v\fI var\fB\^=\^\fIval\fR
-.TP
-.PD
-\fB\-\^\-assign \fIvar\fB\^=\^\fIval\fR
-Assign the value
-.IR val ,
-to the variable
-.IR var ,
-before execution of the program begins.
-Such variable values are available to the
-.B BEGIN
-block of an \*(AK program.
-.TP
-.PD 0
-.BI \-f " program-file"
-.TP
-.PD
-.BI \-\^\-file " program-file"
-Read the \*(AK program source from the file
-.IR program-file ,
-instead of from the first command line argument.
-Multiple
-.B \-f
-(or
-.BR \-\^\-file )
-options may be used.
-.TP
-.PD 0
-.BI \-mf " NNN"
-.TP
-.PD
-.BI \-mr " NNN"
-Set various memory limits to the value
-.IR NNN .
-The
-.B f
-flag sets the maximum number of fields, and the
-.B r
-flag sets the maximum record size.  These two flags and the
-.B \-m
-option are from the Bell Labs research version of \*(UX
-.IR awk .
-They are ignored by
-.IR gawk ,
-since
-.I gawk
-has no pre-defined limits.
-.TP
-.PD 0
-.B "\-W traditional"
-.TP
-.PD 0
-.B "\-W compat"
-.TP
-.PD 0
-.B \-\^\-traditional
-.TP
-.PD
-.B \-\^\-compat
-Run in
-.I compatibility
-mode.  In compatibility mode,
-.I gawk
-behaves identically to \*(UX
-.IR awk ;
-none of the \*(GN-specific extensions are recognized.
-The use of
-.B \-\^\-traditional
-is preferred over the other forms of this option.
-See
-.BR "GNU EXTENSIONS" ,
-below, for more information.
-.TP
-.PD 0
-.B "\-W copyleft"
-.TP
-.PD 0
-.B "\-W copyright"
-.TP
-.PD 0
-.B \-\^\-copyleft
-.TP
-.PD
-.B \-\^\-copyright
-Print the short version of the \*(GN copyright information message on
-the standard output, and exits successfully.
-.TP
-.PD 0
-.B "\-W help"
-.TP
-.PD 0
-.B "\-W usage"
-.TP
-.PD 0
-.B \-\^\-help
-.TP
-.PD
-.B \-\^\-usage
-Print a relatively short summary of the available options on
-the standard output.
-(Per the
-.IR "GNU Coding Standards" ,
-these options cause an immediate, successful exit.)
-.TP
-.PD 0
-.B "\-W lint"
-.TP
-.PD
-.B \-\^\-lint
-Provide warnings about constructs that are
-dubious or non-portable to other \*(AK implementations.
-.TP
-.PD 0
-.B "\-W lint\-old"
-.TP
-.PD
-.B \-\^\-lint\-old
-Provide warnings about constructs that are
-not portable to the original version of Unix
-.IR awk .
-.ig
-.\" This option is left undocumented, on purpose.
-.TP
-.PD 0
-.B "\-W nostalgia"
-.TP
-.PD
-.B \-\^\-nostalgia
-Provide a moment of nostalgia for long time
-.I awk
-users.
-..
-.TP
-.PD 0
-.B "\-W posix"
-.TP
-.PD
-.B \-\^\-posix
-This turns on
-.I compatibility
-mode, with the following additional restrictions:
-.RS
-.TP \w'\(bu'u+1n
-\(bu
-.B \ex
-escape sequences are not recognized.
-.TP
-\(bu
-Only space and tab act as field separators when
-.B FS
-is set to a single space, newline does not.
-.TP
-\(bu
-The synonym
-.B func
-for the keyword
-.B function
-is not recognized.
-.TP
-\(bu
-The operators
-.B **
-and
-.B **=
-cannot be used in place of
-.B ^
-and
-.BR ^= .
-.TP
-\(bu
-The
-.B fflush()
-function is not available.
-.RE
-.TP
-.PD 0
-.B "\-W re\-interval"
-.TP
-.PD
-.B \-\^\-re\-interval
-Enable the use of
-.I "interval expressions"
-in regular expression matching
-(see
-.BR "Regular Expressions" ,
-below).
-Interval expressions were not traditionally available in the
-\*(AK language.  The \*(PX standard added them, to make
-.I awk
-and
-.I egrep
-consistent with each other.
-However, their use is likely
-to break old \*(AK programs, so
-.I gawk
-only provides them if they are requested with this option, or when
-.B \-\^\-posix
-is specified.
-.TP
-.PD 0
-.BI "\-W source " program-text
-.TP
-.PD
-.BI \-\^\-source " program-text"
-Use
-.I program-text
-as \*(AK program source code.
-This option allows the easy intermixing of library functions (used via the
-.B \-f
-and
-.B \-\^\-file
-options) with source code entered on the command line.
-It is intended primarily for medium to large \*(AK programs used
-in shell scripts.
-.TP
-.PD 0
-.B "\-W version"
-.TP
-.PD
-.B \-\^\-version
-Print version information for this particular copy of
-.I gawk
-on the standard output.
-This is useful mainly for knowing if the current copy of
-.I gawk
-on your system
-is up to date with respect to whatever the Free Software Foundation
-is distributing.
-This is also useful when reporting bugs.
-(Per the
-.IR "GNU Coding Standards" ,
-these options cause an immediate, successful exit.)
-.TP
-.B \-\^\-
-Signal the end of options.  This is useful to allow further arguments to the
-\*(AK program itself to start with a \*(lq\-\*(rq.
-This is mainly for consistency with the argument parsing convention used
-by most other \*(PX programs.
-.PP
-In compatibility mode,
-any other options are flagged as illegal, but are otherwise ignored.
-In normal operation, as long as program text has been supplied, unknown
-options are passed on to the \*(AK program in the
-.B ARGV
-array for processing.  This is particularly useful for running \*(AK
-programs via the \*(lq#!\*(rq executable interpreter mechanism.
-.SH AWK PROGRAM EXECUTION
-.PP
-An \*(AK program consists of a sequence of pattern-action statements
-and optional function definitions.
-.RS
-.PP
-\fIpattern\fB	{ \fIaction statements\fB }\fR
-.br
-\fBfunction \fIname\fB(\fIparameter list\fB) { \fIstatements\fB }\fR
-.RE
-.PP
-.I Gawk
-first reads the program source from the
-.IR program-file (s)
-if specified,
-from arguments to
-.BR \-\^\-source ,
-or from the first non-option argument on the command line.
-The
-.B \-f
-and
-.B \-\^\-source
-options may be used multiple times on the command line.
-.I Gawk
-will read the program text as if all the
-.IR program-file s
-and command line source texts
-had been concatenated together.  This is useful for building libraries
-of \*(AK functions, without having to include them in each new \*(AK
-program that uses them.  It also provides the ability to mix library
-functions with command line programs.
-.PP
-The environment variable
-.B AWKPATH
-specifies a search path to use when finding source files named with
-the
-.B \-f
-option.  If this variable does not exist, the default path is
-\fB".:/usr/local/share/awk"\fR.
-(The actual directory may vary, depending upon how
-.I gawk
-was built and installed.)
-If a file name given to the
-.B \-f
-option contains a \*(lq/\*(rq character, no path search is performed.
-.PP
-.I Gawk
-executes \*(AK programs in the following order.
-First,
-all variable assignments specified via the
-.B \-v
-option are performed.
-Next,
-.I gawk
-compiles the program into an internal form.
-Then,
-.I gawk
-executes the code in the
-.B BEGIN
-block(s) (if any),
-and then proceeds to read
-each file named in the
-.B ARGV
-array.
-If there are no files named on the command line,
-.I gawk
-reads the standard input.
-.PP
-If a filename on the command line has the form
-.IB var = val
-it is treated as a variable assignment.  The variable
-.I var
-will be assigned the value
-.IR val .
-(This happens after any
-.B BEGIN
-block(s) have been run.)
-Command line variable assignment
-is most useful for dynamically assigning values to the variables
-\*(AK uses to control how input is broken into fields and records.
-It is also useful for controlling state if multiple passes are needed over
-a single data file.
-.PP
-If the value of a particular element of
-.B ARGV
-is empty (\fB""\fR),
-.I gawk
-skips over it.
-.PP
-For each record in the input,
-.I gawk
-tests to see if it matches any
-.I pattern
-in the \*(AK program.
-For each pattern that the record matches, the associated
-.I action
-is executed.
-The patterns are tested in the order they occur in the program.
-.PP
-Finally, after all the input is exhausted,
-.I gawk
-executes the code in the
-.B END
-block(s) (if any).
-.SH VARIABLES, RECORDS AND FIELDS
-\*(AK variables are dynamic; they come into existence when they are
-first used.  Their values are either floating-point numbers or strings,
-or both,
-depending upon how they are used.  \*(AK also has one dimensional
-arrays; arrays with multiple dimensions may be simulated.
-Several pre-defined variables are set as a program
-runs; these will be described as needed and summarized below.
-.SS Records
-Normally, records are separated by newline characters.  You can control how
-records are separated by assigning values to the built-in variable
-.BR RS .
-If
-.B RS
-is any single character, that character separates records.
-Otherwise,
-.B RS
-is a regular expression.  Text in the input that matches this
-regular expression will separate the record.
-However, in compatibility mode,
-only the first character of its string
-value is used for separating records.
-If
-.B RS
-is set to the null string, then records are separated by
-blank lines.
-When
-.B RS
-is set to the null string, the newline character always acts as
-a field separator, in addition to whatever value
-.B FS
-may have.
-.SS Fields
-.PP
-As each input record is read,
-.I gawk
-splits the record into
-.IR fields ,
-using the value of the
-.B FS
-variable as the field separator.
-If
-.B FS
-is a single character, fields are separated by that character.
-If
-.B FS
-is the null string, then each individual character becomes a
-separate field.
-Otherwise,
-.B FS
-is expected to be a full regular expression.
-In the special case that
-.B FS
-is a single space, fields are separated
-by runs of spaces and/or tabs and/or newlines.
-(But see the discussion of
-.BR \-\^\-posix ,
-below).
-Note that the value of
-.B IGNORECASE
-(see below) will also affect how fields are split when
-.B FS
-is a regular expression, and how records are separated when
-.B RS
-is a regular expression.
-.PP
-If the
-.B FIELDWIDTHS
-variable is set to a space separated list of numbers, each field is
-expected to have fixed width, and
-.I gawk
-will split up the record using the specified widths.  The value of
-.B FS
-is ignored.
-Assigning a new value to
-.B FS
-overrides the use of
-.BR FIELDWIDTHS ,
-and restores the default behavior.
-.PP
-Each field in the input record may be referenced by its position,
-.BR $1 ,
-.BR $2 ,
-and so on.
-.B $0
-is the whole record.  The value of a field may be assigned to as well.
-Fields need not be referenced by constants:
-.RS
-.PP
-.ft B
-n = 5
-.br
-print $n
-.ft R
-.RE
-.PP
-prints the fifth field in the input record.
-The variable
-.B NF
-is set to the total number of fields in the input record.
-.PP
-References to non-existent fields (i.e. fields after
-.BR $NF )
-produce the null-string.  However, assigning to a non-existent field
-(e.g.,
-.BR "$(NF+2) = 5" )
-will increase the value of
-.BR NF ,
-create any intervening fields with the null string as their value, and
-cause the value of
-.B $0
-to be recomputed, with the fields being separated by the value of
-.BR OFS .
-References to negative numbered fields cause a fatal error.
-Decrementing
-.B NF
-causes the values of fields past the new value to be lost, and the value of
-.B $0
-to be recomputed, with the fields being separated by the value of
-.BR OFS .
-.SS Built-in Variables
-.PP
-.IR Gawk 's
-built-in variables are:
-.PP
-.TP \w'\fBFIELDWIDTHS\fR'u+1n
-.B ARGC
-The number of command line arguments (does not include options to
-.IR gawk ,
-or the program source).
-.TP
-.B ARGIND
-The index in
-.B ARGV
-of the current file being processed.
-.TP
-.B ARGV
-Array of command line arguments.  The array is indexed from
-0 to
-.B ARGC
-\- 1.
-Dynamically changing the contents of
-.B ARGV
-can control the files used for data.
-.TP
-.B CONVFMT
-The conversion format for numbers, \fB"%.6g"\fR, by default.
-.TP
-.B ENVIRON
-An array containing the values of the current environment.
-The array is indexed by the environment variables, each element being
-the value of that variable (e.g., \fBENVIRON["HOME"]\fP might be
-.BR /home/arnold ).
-Changing this array does not affect the environment seen by programs which
-.I gawk
-spawns via redirection or the
-.B system()
-function.
-(This may change in a future version of
-.IR gawk .)
-.\" but don't hold your breath...
-.TP
-.B ERRNO
-If a system error occurs either doing a redirection for
-.BR getline ,
-during a read for
-.BR getline ,
-or during a
-.BR close() ,
-then
-.B ERRNO
-will contain
-a string describing the error.
-.TP
-.B FIELDWIDTHS
-A white-space separated list of fieldwidths.  When set,
-.I gawk
-parses the input into fields of fixed width, instead of using the
-value of the
-.B FS
-variable as the field separator.
-The fixed field width facility is still experimental; the
-semantics may change as
-.I gawk
-evolves over time.
-.TP
-.B FILENAME
-The name of the current input file.
-If no files are specified on the command line, the value of
-.B FILENAME
-is \*(lq\-\*(rq.
-However,
-.B FILENAME
-is undefined inside the
-.B BEGIN
-block.
-.TP
-.B FNR
-The input record number in the current input file.
-.TP
-.B FS
-The input field separator, a space by default.  See
-.BR Fields ,
-above.
-.TP
-.B IGNORECASE
-Controls the case-sensitivity of all regular expression
-and string operations.  If
-.B IGNORECASE
-has a non-zero value, then string comparisons and
-pattern matching in rules,
-field splitting with
-.BR FS ,
-record separating with
-.BR RS ,
-regular expression
-matching with
-.B ~
-and
-.BR !~ ,
-and the
-.BR gensub() ,
-.BR gsub() ,
-.BR index() ,
-.BR match() ,
-.BR split() ,
-and
-.B sub()
-pre-defined functions will all ignore case when doing regular expression
-operations.  Thus, if
-.B IGNORECASE
-is not equal to zero,
-.B /aB/
-matches all of the strings \fB"ab"\fP, \fB"aB"\fP, \fB"Ab"\fP,
-and \fB"AB"\fP.
-As with all \*(AK variables, the initial value of
-.B IGNORECASE
-is zero, so all regular expression and string
-operations are normally case-sensitive.
-Under Unix, the full ISO 8859-1 Latin-1 character set is used
-when ignoring case.
-.B NOTE:
-In versions of
-.I gawk
-prior to 3.0,
-.B IGNORECASE
-only affected regular expression operations.  It now affects string
-comparisons as well.
-.TP
-.B NF
-The number of fields in the current input record.
-.TP
-.B NR
-The total number of input records seen so far.
-.TP
-.B OFMT
-The output format for numbers, \fB"%.6g"\fR, by default.
-.TP
-.B OFS
-The output field separator, a space by default.
-.TP
-.B ORS
-The output record separator, by default a newline.
-.TP
-.B RS
-The input record separator, by default a newline.
-.TP
-.B RT
-The record terminator.
-.I Gawk
-sets
-.B RT
-to the input text that matched the character or regular expression
-specified by
-.BR RS .
-.TP
-.B RSTART
-The index of the first character matched by
-.BR match() ;
-0 if no match.
-.TP
-.B RLENGTH
-The length of the string matched by
-.BR match() ;
-\-1 if no match.
-.TP
-.B SUBSEP
-The character used to separate multiple subscripts in array
-elements, by default \fB"\e034"\fR.
-.SS Arrays
-.PP
-Arrays are subscripted with an expression between square brackets
-.RB ( [ " and " ] ).
-If the expression is an expression list
-.RI ( expr ", " expr " .\|.\|.)"
-then the array subscript is a string consisting of the
-concatenation of the (string) value of each expression,
-separated by the value of the
-.B SUBSEP
-variable.
-This facility is used to simulate multiply dimensioned
-arrays.  For example:
-.PP
-.RS
-.ft B
-i = "A";\^ j = "B";\^ k = "C"
-.br
-x[i, j, k] = "hello, world\en"
-.ft R
-.RE
-.PP
-assigns the string \fB"hello, world\en"\fR to the element of the array
-.B x
-which is indexed by the string \fB"A\e034B\e034C"\fR.  All arrays in \*(AK
-are associative, i.e. indexed by string values.
-.PP
-The special operator
-.B in
-may be used in an
-.B if
-or
-.B while
-statement to see if an array has an index consisting of a particular
-value.
-.PP
-.RS
-.ft B
-.nf
-if (val in array)
-	print array[val]
-.fi
-.ft
-.RE
-.PP
-If the array has multiple subscripts, use
-.BR "(i, j) in array" .
-.PP
-The
-.B in
-construct may also be used in a
-.B for
-loop to iterate over all the elements of an array.
-.PP
-An element may be deleted from an array using the
-.B delete
-statement.
-The
-.B delete
-statement may also be used to delete the entire contents of an array,
-just by specifying the array name without a subscript.
-.SS Variable Typing And Conversion
-.PP
-Variables and fields
-may be (floating point) numbers, or strings, or both.  How the
-value of a variable is interpreted depends upon its context.  If used in
-a numeric expression, it will be treated as a number, if used as a string
-it will be treated as a string.
-.PP
-To force a variable to be treated as a number, add 0 to it; to force it
-to be treated as a string, concatenate it with the null string.
-.PP
-When a string must be converted to a number, the conversion is accomplished
-using
-.IR atof (3).
-A number is converted to a string by using the value of
-.B CONVFMT
-as a format string for
-.IR sprintf (3),
-with the numeric value of the variable as the argument.
-However, even though all numbers in \*(AK are floating-point,
-integral values are
-.I always
-converted as integers.  Thus, given
-.PP
-.RS
-.ft B
-.nf
-CONVFMT = "%2.2f"
-a = 12
-b = a ""
-.fi
-.ft R
-.RE
-.PP
-the variable
-.B b
-has a string value of \fB"12"\fR and not \fB"12.00"\fR.
-.PP
-.I Gawk
-performs comparisons as follows:
-If two variables are numeric, they are compared numerically.
-If one value is numeric and the other has a string value that is a
-\*(lqnumeric string,\*(rq then comparisons are also done numerically.
-Otherwise, the numeric value is converted to a string and a string
-comparison is performed.
-Two strings are compared, of course, as strings.
-According to the \*(PX standard, even if two strings are
-numeric strings, a numeric comparison is performed.  However, this is
-clearly incorrect, and
-.I gawk
-does not do this.
-.PP
-Note that string constants, such as \fB"57"\fP, are
-.I not
-numeric strings, they are string constants.
-The idea of \*(lqnumeric string\*(rq
-only applies to fields,
-.B getline
-input,
-.BR FILENAME ,
-.B ARGV
-elements,
-.B ENVIRON
-elements and the elements of an array created by
-.B split()
-that are numeric strings.
-The basic idea is that
-.IR "user input" ,
-and only user input, that looks numeric,
-should be treated that way.
-.PP
-Uninitialized variables have the numeric value 0 and the string value ""
-(the null, or empty, string).
-.SH PATTERNS AND ACTIONS
-\*(AK is a line-oriented language.  The pattern comes first, and then the
-action.  Action statements are enclosed in
-.B {
-and
-.BR } .
-Either the pattern may be missing, or the action may be missing, but,
-of course, not both.  If the pattern is missing, the action will be
-executed for every single record of input.
-A missing action is equivalent to
-.RS
-.PP
-.B "{ print }"
-.RE
-.PP
-which prints the entire record.
-.PP
-Comments begin with the \*(lq#\*(rq character, and continue until the
-end of the line.
-Blank lines may be used to separate statements.
-Normally, a statement ends with a newline, however, this is not the
-case for lines ending in
-a \*(lq,\*(rq,
-.BR { ,
-.BR ? ,
-.BR : ,
-.BR && ,
-or
-.BR || .
-Lines ending in
-.B do
-or
-.B else
-also have their statements automatically continued on the following line.
-In other cases, a line can be continued by ending it with a \*(lq\e\*(rq,
-in which case the newline will be ignored.
-.PP
-Multiple statements may
-be put on one line by separating them with a \*(lq;\*(rq.
-This applies to both the statements within the action part of a
-pattern-action pair (the usual case),
-and to the pattern-action statements themselves.
-.SS Patterns
-\*(AK patterns may be one of the following:
-.PP
-.RS
-.nf
-.B BEGIN
-.B END
-.BI / "regular expression" /
-.I "relational expression"
-.IB pattern " && " pattern
-.IB pattern " || " pattern
-.IB pattern " ? " pattern " : " pattern
-.BI ( pattern )
-.BI ! " pattern"
-.IB pattern1 ", " pattern2
-.fi
-.RE
-.PP
-.B BEGIN
-and
-.B END
-are two special kinds of patterns which are not tested against
-the input.
-The action parts of all
-.B BEGIN
-patterns are merged as if all the statements had
-been written in a single
-.B BEGIN
-block.  They are executed before any
-of the input is read.  Similarly, all the
-.B END
-blocks are merged,
-and executed when all the input is exhausted (or when an
-.B exit
-statement is executed).
-.B BEGIN
-and
-.B END
-patterns cannot be combined with other patterns in pattern expressions.
-.B BEGIN
-and
-.B END
-patterns cannot have missing action parts.
-.PP
-For
-.BI / "regular expression" /
-patterns, the associated statement is executed for each input record that matches
-the regular expression.
-Regular expressions are the same as those in
-.IR egrep (1),
-and are summarized below.
-.PP
-A
-.I "relational expression"
-may use any of the operators defined below in the section on actions.
-These generally test whether certain fields match certain regular expressions.
-.PP
-The
-.BR && ,
-.BR || ,
-and
-.B !
-operators are logical AND, logical OR, and logical NOT, respectively, as in C.
-They do short-circuit evaluation, also as in C, and are used for combining
-more primitive pattern expressions.  As in most languages, parentheses
-may be used to change the order of evaluation.
-.PP
-The
-.B ?\^:
-operator is like the same operator in C.  If the first pattern is true
-then the pattern used for testing is the second pattern, otherwise it is
-the third.  Only one of the second and third patterns is evaluated.
-.PP
-The
-.IB pattern1 ", " pattern2
-form of an expression is called a
-.IR "range pattern" .
-It matches all input records starting with a record that matches
-.IR pattern1 ,
-and continuing until a record that matches
-.IR pattern2 ,
-inclusive.  It does not combine with any other sort of pattern expression.
-.SS Regular Expressions
-Regular expressions are the extended kind found in
-.IR egrep .
-They are composed of characters as follows:
-.TP \w'\fB[^\fIabc.\|.\|.\fB]\fR'u+2n
-.I c
-matches the non-metacharacter
-.IR c .
-.TP
-.I \ec
-matches the literal character
-.IR c .
-.TP
-.B .
-matches any character
-.I including
-newline.
-.TP
-.B ^
-matches the beginning of a string.
-.TP
-.B $
-matches the end of a string.
-.TP
-.BI [ abc.\|.\|. ]
-character list, matches any of the characters
-.IR abc.\|.\|. .
-.TP
-.BI [^ abc.\|.\|. ]
-negated character list, matches any character except
-.IR abc.\|.\|. .
-.TP
-.IB r1 | r2
-alternation: matches either
-.I r1
-or
-.IR r2 .
-.TP
-.I r1r2
-concatenation: matches
-.IR r1 ,
-and then
-.IR r2 .
-.TP
-.IB r +
-matches one or more
-.IR r 's.
-.TP
-.IB r *
-matches zero or more
-.IR r 's.
-.TP
-.IB r ?
-matches zero or one
-.IR r 's.
-.TP
-.BI ( r )
-grouping: matches
-.IR r .
-.TP
-.PD 0
-.IB r { n }
-.TP
-.PD 0
-.IB r { n ,}
-.TP
-.PD
-.IB r { n , m }
-One or two numbers inside braces denote an
-.IR "interval expression" .
-If there is one number in the braces, the preceding regexp
-.I r
-is repeated
-.I n
-times.  If there are two numbers separated by a comma,
-.I r
-is repeated
-.I n
-to
-.I m
-times.
-If there is one number followed by a comma, then
-.I r
-is repeated at least
-.I n
-times.
-.sp .5
-Interval expressions are only available if either
-.B \-\^\-posix
-or
-.B \-\^\-re\-interval
-is specified on the command line.
-.TP
-.B \ey
-matches the empty string at either the beginning or the
-end of a word.
-.TP
-.B \eB
-matches the empty string within a word.
-.TP
-.B \e<
-matches the empty string at the beginning of a word.
-.TP
-.B \e>
-matches the empty string at the end of a word.
-.TP
-.B \ew
-matches any word-constituent character (letter, digit, or underscore).
-.TP
-.B \eW
-matches any character that is not word-constituent.
-.TP
-.B \e`
-matches the empty string at the beginning of a buffer (string).
-.TP
-.B \e'
-matches the empty string at the end of a buffer.
-.PP
-The escape sequences that are valid in string constants (see below)
-are also legal in regular expressions.
-.PP
-.I "Character classes"
-are a new feature introduced in the \*(PX standard.
-A character class is a special notation for describing
-lists of characters that have a specific attribute, but where the
-actual characters themselves can vary from country to country and/or
-from character set to character set.  For example, the notion of what
-is an alphabetic character differs in the USA and in France.
-.PP
-A character class is only valid in a regexp
-.I inside
-the brackets of a character list.  Character classes consist of
-.BR [: ,
-a keyword denoting the class, and
-.BR :] .
-Here are the character
-classes defined by the \*(PX standard.
-.TP
-.B [:alnum:]
-Alphanumeric characters.
-.TP
-.B [:alpha:]
-Alphabetic characters.
-.TP
-.B [:blank:]
-Space or tab characters.
-.TP
-.B [:cntrl:]
-Control characters.
-.TP
-.B [:digit:]
-Numeric characters.
-.TP
-.B [:graph:]
-Characters that are both printable and visible.
-(A space is printable, but not visible, while an
-.B a
-is both.)
-.TP
-.B [:lower:]
-Lower-case alphabetic characters.
-.TP
-.B [:print:]
-Printable characters (characters that are not control characters.)
-.TP
-.B [:punct:]
-Punctuation characters (characters that are not letter, digits,
-control characters, or space characters).
-.TP
-.B [:space:]
-Space characters (such as space, tab, and formfeed, to name a few).
-.TP
-.B [:upper:]
-Upper-case alphabetic characters.
-.TP
-.B [:xdigit:]
-Characters that are hexadecimal digits.
-.PP
-For example, before the \*(PX standard, to match alphanumeric
-characters, you would have had to write
-.BR /[A\-Za\-z0\-9]/ .
-If your character set had other alphabetic characters in it, this would not
-match them.  With the \*(PX character classes, you can write
-.BR /[[:alnum:]]/ ,
-and this will match
-.I all
-the alphabetic and numeric characters in your character set.
-.PP
-Two additional special sequences can appear in character lists.
-These apply to non-ASCII character sets, which can have single symbols
-(called
-.IR "collating elements" )
-that are represented with more than one
-character, as well as several characters that are equivalent for
-.IR collating ,
-or sorting, purposes.  (E.g., in French, a plain \*(lqe\*(rq
-and a grave-accented e\` are equivalent.)
-.TP
-Collating Symbols
-A collating symbols is a multi-character collating element enclosed in
-.B [.
-and
-.BR .] .
-For example, if
-.B ch
-is a collating element, then
-.B [[.ch.]]
-is a regexp that matches this collating element, while
-.B [ch]
-is a regexp that matches either
-.B c
-or
-.BR h .
-.TP
-Equivalence Classes
-An equivalence class is a locale-specific name for a list of
-characters that are equivalent.  The name is enclosed in
-.B [=
-and
-.BR =] .
-For example, the name
-.B e
-might be used to represent all of
-\*(lqe,\*(rq \*(lqe\`,\*(rq and \*(lqe\`.\*(rq
-In this case,
-.B [[=e]]
-is a regexp
-that matches any of
- .BR e ,
- .BR e\' ,
-or
- .BR e\` .
-.PP
-These features are very valuable in non-English speaking locales.
-The library functions that
-.I gawk
-uses for regular expression matching
-currently only recognize \*(PX character classes; they do not recognize
-collating symbols or equivalence classes.
-.PP
-The
-.BR \ey ,
-.BR \eB ,
-.BR \e< ,
-.BR \e> ,
-.BR \ew ,
-.BR \eW ,
-.BR \e` ,
-and
-.B \e'
-operators are specific to
-.IR gawk ;
-they are extensions based on facilities in the \*(GN regexp libraries.
-.PP
-The various command line options
-control how
-.I gawk
-interprets characters in regexps.
-.TP
-No options
-In the default case,
-.I gawk
-provide all the facilities of
-\*(PX regexps and the \*(GN regexp operators described above.
-However, interval expressions are not supported.
-.TP
-.B \-\^\-posix
-Only \*(PX regexps are supported, the \*(GN operators are not special.
-(E.g.,
-.B \ew
-matches a literal
-.BR w ).
-Interval expressions are allowed.
-.TP
-.B \-\^\-traditional
-Traditional Unix
-.I awk
-regexps are matched.  The \*(GN operators
-are not special, interval expressions are not available, and neither
-are the \*(PX character classes
-.RB ( [[:alnum:]]
-and so on).
-Characters described by octal and hexadecimal escape sequences are
-treated literally, even if they represent regexp metacharacters.
-.TP
-.B \-\^\-re\-interval
-Allow interval expressions in regexps, even if
-.B \-\^\-traditional
-has been provided.
-.SS Actions
-Action statements are enclosed in braces,
-.B {
-and
-.BR } .
-Action statements consist of the usual assignment, conditional, and looping
-statements found in most languages.  The operators, control statements,
-and input/output statements
-available are patterned after those in C.
-.SS Operators
-.PP
-The operators in \*(AK, in order of decreasing precedence, are
-.PP
-.TP "\w'\fB*= /= %= ^=\fR'u+1n"
-.BR ( \&.\|.\|. )
-Grouping
-.TP
-.B $
-Field reference.
-.TP
-.B "++ \-\^\-"
-Increment and decrement, both prefix and postfix.
-.TP
-.B ^
-Exponentiation (\fB**\fR may also be used, and \fB**=\fR for
-the assignment operator).
-.TP
-.B "+ \- !"
-Unary plus, unary minus, and logical negation.
-.TP
-.B "* / %"
-Multiplication, division, and modulus.
-.TP
-.B "+ \-"
-Addition and subtraction.
-.TP
-.I space
-String concatenation.
-.TP
-.PD 0
-.B "< >"
-.TP
-.PD 0
-.B "<= >="
-.TP
-.PD
-.B "!= =="
-The regular relational operators.
-.TP
-.B "~ !~"
-Regular expression match, negated match.
-.B NOTE:
-Do not use a constant regular expression
-.RB ( /foo/ )
-on the left-hand side of a
-.B ~
-or
-.BR !~ .
-Only use one on the right-hand side.  The expression
-.BI "/foo/ ~ " exp
-has the same meaning as \fB(($0 ~ /foo/) ~ \fIexp\fB)\fR.
-This is usually
-.I not
-what was intended.
-.TP
-.B in
-Array membership.
-.TP
-.B &&
-Logical AND.
-.TP
-.B ||
-Logical OR.
-.TP
-.B ?:
-The C conditional expression.  This has the form
-.IB expr1 " ? " expr2 " : " expr3\c
-\&.
-If
-.I expr1
-is true, the value of the expression is
-.IR expr2 ,
-otherwise it is
-.IR expr3 .
-Only one of
-.I expr2
-and
-.I expr3
-is evaluated.
-.TP
-.PD 0
-.B "= += \-="
-.TP
-.PD
-.B "*= /= %= ^="
-Assignment.  Both absolute assignment
-.BI ( var " = " value )
-and operator-assignment (the other forms) are supported.
-.SS Control Statements
-.PP
-The control statements are
-as follows:
-.PP
-.RS
-.nf
-\fBif (\fIcondition\fB) \fIstatement\fR [ \fBelse\fI statement \fR]
-\fBwhile (\fIcondition\fB) \fIstatement \fR
-\fBdo \fIstatement \fBwhile (\fIcondition\fB)\fR
-\fBfor (\fIexpr1\fB; \fIexpr2\fB; \fIexpr3\fB) \fIstatement\fR
-\fBfor (\fIvar \fBin\fI array\fB) \fIstatement\fR
-\fBbreak\fR
-\fBcontinue\fR
-\fBdelete \fIarray\^\fB[\^\fIindex\^\fB]\fR
-\fBdelete \fIarray\^\fR
-\fBexit\fR [ \fIexpression\fR ]
-\fB{ \fIstatements \fB}
-.fi
-.RE
-.SS "I/O Statements"
-.PP
-The input/output statements are as follows:
-.PP
-.TP "\w'\fBprintf \fIfmt, expr-list\fR'u+1n"
-.BI close( file )
-Close file (or pipe, see below).
-.TP
-.B getline
-Set
-.B $0
-from next input record; set
-.BR NF ,
-.BR NR ,
-.BR FNR .
-.TP
-.BI "getline <" file
-Set
-.B $0
-from next record of
-.IR file ;
-set
-.BR NF .
-.TP
-.BI getline " var"
-Set
-.I var
-from next input record; set
-.BR NR ,
-.BR FNR .
-.TP
-.BI getline " var" " <" file
-Set
-.I var
-from next record of
-.IR file .
-.TP
-.B next
-Stop processing the current input record.  The next input record
-is read and processing starts over with the first pattern in the
-\*(AK program.  If the end of the input data is reached, the
-.B END
-block(s), if any, are executed.
-.TP
-.B "nextfile"
-Stop processing the current input file.  The next input record read
-comes from the next input file.
-.B FILENAME
-and
-.B ARGIND
-are updated,
-.B FNR
-is reset to 1, and processing starts over with the first pattern in the
-\*(AK program.  If the end of the input data is reached, the
-.B END
-block(s), if any, are executed.
-.B NOTE:
-Earlier versions of gawk used
-.BR "next file" ,
-as two words.  While this usage is still recognized, it generates a
-warning message and will eventually be removed.
-.TP
-.B print
-Prints the current record.
-The output record is terminated with the value of the
-.B ORS
-variable.
-.TP
-.BI print " expr-list"
-Prints expressions.
-Each expression is separated by the value of the
-.B OFS
-variable.
-The output record is terminated with the value of the
-.B ORS
-variable.
-.TP
-.BI print " expr-list" " >" file
-Prints expressions on
-.IR file .
-Each expression is separated by the value of the
-.B OFS
-variable.  The output record is terminated with the value of the
-.B ORS
-variable.
-.TP
-.BI printf " fmt, expr-list"
-Format and print.
-.TP
-.BI printf " fmt, expr-list" " >" file
-Format and print on
-.IR file .
-.TP
-.BI system( cmd-line )
-Execute the command
-.IR cmd-line ,
-and return the exit status.
-(This may not be available on non-\*(PX systems.)
-.TP
-\&\fBfflush(\fR[\fIfile\^\fR]\fB)\fR
-Flush any buffers associated with the open output file or pipe
-.IR file .
-If
-.I file
-is missing, then standard output is flushed.
-If
-.I file
-is the null string,
-then all open output files and pipes
-have their buffers flushed.
-.PP
-Other input/output redirections are also allowed.  For
-.B print
-and
-.BR printf ,
-.BI >> " file"
-appends output to the
-.IR file ,
-while
-.BI | " command"
-writes on a pipe.
-In a similar fashion,
-.IB command " | getline"
-pipes into
-.BR getline .
-The
-.BR getline
-command will return 0 on end of file, and \-1 on an error.
-.PP
-NOTE: If using a pipe to
-.BR getline ,
-or from
-.B print
-or
-.BR printf
-within a loop, you
-.I must
-use
-.B close()
-to create new instances of the command.
-AWK does not automatically close pipes when
-they return EOF.
-.SS The \fIprintf\fP\^ Statement
-.PP
-The \*(AK versions of the
-.B printf
-statement and
-.B sprintf()
-function
-(see below)
-accept the following conversion specification formats:
-.TP
-.B %c
-An \s-1ASCII\s+1 character.
-If the argument used for
-.B %c
-is numeric, it is treated as a character and printed.
-Otherwise, the argument is assumed to be a string, and the only first
-character of that string is printed.
-.TP
-.PD 0
-.B %d
-.TP
-.PD
-.B %i
-A decimal number (the integer part).
-.TP
-.PD 0
-.B %e
-.TP
-.PD
-.B %E
-A floating point number of the form
-.BR [\-]d.dddddde[+\^\-]dd .
-The
-.B %E
-format uses
-.B E
-instead of
-.BR e .
-.TP
-.B %f
-A floating point number of the form
-.BR [\-]ddd.dddddd .
-.TP
-.PD 0
-.B %g
-.TP
-.PD
-.B %G
-Use
-.B %e
-or
-.B %f
-conversion, whichever is shorter, with nonsignificant zeros suppressed.
-The
-.B %G
-format uses
-.B %E
-instead of
-.BR %e .
-.TP
-.B %o
-An unsigned octal number (also an integer).
-.TP
-.PD
-.B %u
-An unsigned decimal number (again, an integer).
-.TP
-.B %s
-A character string.
-.TP
-.PD 0
-.B %x
-.TP
-.PD
-.B %X
-An unsigned hexadecimal number (an integer).
-The
-.B %X
-format uses
-.B ABCDEF
-instead of
-.BR abcdef .
-.TP
-.B %%
-A single
-.B %
-character; no argument is converted.
-.PP
-There are optional, additional parameters that may lie between the
-.B %
-and the control letter:
-.TP
-.B \-
-The expression should be left-justified within its field.
-.TP
-.I space
-For numeric conversions, prefix positive values with a space, and
-negative values with a minus sign.
-.TP
-.B +
-The plus sign, used before the width modifier (see below),
-says to always supply a sign for numeric conversions, even if the data
-to be formatted is positive.  The
-.B +
-overrides the space modifier.
-.TP
-.B #
-Use an \*(lqalternate form\*(rq for certain control letters.
-For
-.BR %o ,
-supply a leading zero.
-For
-.BR %x ,
-and
-.BR %X ,
-supply a leading
-.BR 0x
-or
-.BR 0X
-for
-a nonzero result.
-For
-.BR %e ,
-.BR %E ,
-and
-.BR %f ,
-the result will always contain a
-decimal point.
-For
-.BR %g ,
-and
-.BR %G ,
-trailing zeros are not removed from the result.
-.TP
-.B 0
-A leading
-.B 0
-(zero) acts as a flag, that indicates output should be
-padded with zeroes instead of spaces.
-This applies even to non-numeric output formats.
-This flag only has an effect when the field width is wider than the
-value to be printed.
-.TP
-.I width
-The field should be padded to this width.  The field is normally padded
-with spaces.  If the
-.B 0
-flag has been used, it is padded with zeroes.
-.TP
-.BI \&. prec
-A number that specifies the precision to use when printing.
-For the
-.BR %e ,
-.BR %E ,
-and
-.BR %f
-formats, this specifies the
-number of digits you want printed to the right of the decimal point.
-For the
-.BR %g ,
-and
-.B %G
-formats, it specifies the maximum number
-of significant digits.  For the
-.BR %d ,
-.BR %o ,
-.BR %i ,
-.BR %u ,
-.BR %x ,
-and
-.B %X
-formats, it specifies the minimum number of
-digits to print.  For a string, it specifies the maximum number of
-characters from the string that should be printed.
-.PP
-The dynamic
-.I width
-and
-.I prec
-capabilities of the \*(AN C
-.B printf()
-routines are supported.
-A
-.B *
-in place of either the
-.B width
-or
-.B prec
-specifications will cause their values to be taken from
-the argument list to
-.B printf
-or
-.BR sprintf() .
-.SS Special File Names
-.PP
-When doing I/O redirection from either
-.B print
-or
-.B printf
-into a file,
-or via
-.B getline
-from a file,
-.I gawk
-recognizes certain special filenames internally.  These filenames
-allow access to open file descriptors inherited from
-.IR gawk 's
-parent process (usually the shell).
-Other special filenames provide access to information about the running
-.B gawk
-process.
-The filenames are:
-.TP \w'\fB/dev/stdout\fR'u+1n
-.B /dev/pid
-Reading this file returns the process ID of the current process,
-in decimal, terminated with a newline.
-.TP
-.B /dev/ppid
-Reading this file returns the parent process ID of the current process,
-in decimal, terminated with a newline.
-.TP
-.B /dev/pgrpid
-Reading this file returns the process group ID of the current process,
-in decimal, terminated with a newline.
-.TP
-.B /dev/user
-Reading this file returns a single record terminated with a newline.
-The fields are separated with spaces.
-.B $1
-is the value of the
-.IR getuid (2)
-system call,
-.B $2
-is the value of the
-.IR geteuid (2)
-system call,
-.B $3
-is the value of the
-.IR getgid (2)
-system call, and
-.B $4
-is the value of the
-.IR getegid (2)
-system call.
-If there are any additional fields, they are the group IDs returned by
-.IR getgroups (2).
-Multiple groups may not be supported on all systems.
-.TP
-.B /dev/stdin
-The standard input.
-.TP
-.B /dev/stdout
-The standard output.
-.TP
-.B /dev/stderr
-The standard error output.
-.TP
-.BI /dev/fd/\^ n
-The file associated with the open file descriptor
-.IR n .
-.PP
-These are particularly useful for error messages.  For example:
-.PP
-.RS
-.ft B
-print "You blew it!" > "/dev/stderr"
-.ft R
-.RE
-.PP
-whereas you would otherwise have to use
-.PP
-.RS
-.ft B
-print "You blew it!" | "cat 1>&2"
-.ft R
-.RE
-.PP
-These file names may also be used on the command line to name data files.
-.SS Numeric Functions
-.PP
-\*(AK has the following pre-defined arithmetic functions:
-.PP
-.TP \w'\fBsrand(\fR[\fIexpr\^\fR]\fB)\fR'u+1n
-.BI atan2( y , " x" )
-returns the arctangent of
-.I y/x
-in radians.
-.TP
-.BI cos( expr )
-returns the cosine of
-.IR expr ,
-which is in radians.
-.TP
-.BI exp( expr )
-the exponential function.
-.TP
-.BI int( expr )
-truncates to integer.
-.TP
-.BI log( expr )
-the natural logarithm function.
-.TP
-.B rand()
-returns a random number between 0 and 1.
-.TP
-.BI sin( expr )
-returns the sine of
-.IR expr ,
-which is in radians.
-.TP
-.BI sqrt( expr )
-the square root function.
-.TP
-\&\fBsrand(\fR[\fIexpr\^\fR]\fB)\fR
-uses
-.I expr
-as a new seed for the random number generator.  If no
-.I expr
-is provided, the time of day will be used.
-The return value is the previous seed for the random
-number generator.
-.SS String Functions
-.PP
-.I Gawk
-has the following pre-defined string functions:
-.PP
-.TP "\w'\fBsprintf(\^\fIfmt\fB\^, \fIexpr-list\^\fB)\fR'u+1n"
-\fBgensub(\fIr\fB, \fIs\fB, \fIh \fR[\fB, \fIt\fR]\fB)\fR
-search the target string
-.I t
-for matches of the regular expression
-.IR r .
-If
-.I h
-is a string beginning with
-.B g
-or
-.BR G ,
-then replace all matches of
-.I r
-with
-.IR s .
-Otherwise,
-.I h
-is a number indicating which match of
-.I r
-to replace.
-If no
-.I t
-is supplied,
-.B $0
-is used instead.
-Within the replacement text
-.IR s ,
-the sequence
-.BI \e n\fR,
-where
-.I n
-is a digit from 1 to 9, may be used to indicate just the text that
-matched the
-.IR n 'th
-parenthesized subexpression.  The sequence
-.B \e0
-represents the entire matched text, as does the character
-.BR & .
-Unlike
-.B sub()
-and
-.BR gsub() ,
-the modified string is returned as the result of the function,
-and the original target string is
-.I not
-changed.
-.TP "\w'\fBsprintf(\^\fIfmt\fB\^, \fIexpr-list\^\fB)\fR'u+1n"
-\fBgsub(\fIr\fB, \fIs \fR[\fB, \fIt\fR]\fB)\fR
-for each substring matching the regular expression
-.I r
-in the string
-.IR t ,
-substitute the string
-.IR s ,
-and return the number of substitutions.
-If
-.I t
-is not supplied, use
-.BR $0 .
-An
-.B &
-in the replacement text is replaced with the text that was actually matched.
-Use
-.B \e&
-to get a literal
-.BR & .
-See
-.I "Effective AWK Programming"
-for a fuller discussion of the rules for
-.BR &'s
-and backslashes in the replacement text of
-.BR sub() ,
-.BR gsub() ,
-and
-.BR gensub() .
-.TP
-.BI index( s , " t" )
-returns the index of the string
-.I t
-in the string
-.IR s ,
-or 0 if
-.I t
-is not present.
-.TP
-\fBlength(\fR[\fIs\fR]\fB)
-returns the length of the string
-.IR s ,
-or the length of
-.B $0
-if
-.I s
-is not supplied.
-.TP
-.BI match( s , " r" )
-returns the position in
-.I s
-where the regular expression
-.I r
-occurs, or 0 if
-.I r
-is not present, and sets the values of
-.B RSTART
-and
-.BR RLENGTH .
-.TP
-\fBsplit(\fIs\fB, \fIa \fR[\fB, \fIr\fR]\fB)\fR
-splits the string
-.I s
-into the array
-.I a
-on the regular expression
-.IR r ,
-and returns the number of fields.  If
-.I r
-is omitted,
-.B FS
-is used instead.
-The array
-.I a
-is cleared first.
-Splitting behaves identically to field splitting, described above.
-.TP
-.BI sprintf( fmt , " expr-list" )
-prints
-.I expr-list
-according to
-.IR fmt ,
-and returns the resulting string.
-.TP
-\fBsub(\fIr\fB, \fIs \fR[\fB, \fIt\fR]\fB)\fR
-just like
-.BR gsub() ,
-but only the first matching substring is replaced.
-.TP
-\fBsubstr(\fIs\fB, \fIi \fR[\fB, \fIn\fR]\fB)\fR
-returns the at most
-.IR n -character
-substring of
-.I s
-starting at
-.IR i .
-If
-.I n
-is omitted, the rest of
-.I s
-is used.
-.TP
-.BI tolower( str )
-returns a copy of the string
-.IR str ,
-with all the upper-case characters in
-.I str
-translated to their corresponding lower-case counterparts.
-Non-alphabetic characters are left unchanged.
-.TP
-.BI toupper( str )
-returns a copy of the string
-.IR str ,
-with all the lower-case characters in
-.I str
-translated to their corresponding upper-case counterparts.
-Non-alphabetic characters are left unchanged.
-.SS Time Functions
-.PP
-Since one of the primary uses of \*(AK programs is processing log files
-that contain time stamp information,
-.I gawk
-provides the following two functions for obtaining time stamps and
-formatting them.
-.PP
-.TP "\w'\fBsystime()\fR'u+1n"
-.B systime()
-returns the current time of day as the number of seconds since the Epoch
-(Midnight UTC, January 1, 1970 on \*(PX systems).
-.TP
-\fBstrftime(\fR[\fIformat \fR[\fB, \fItimestamp\fR]]\fB)\fR
-formats
-.I timestamp
-according to the specification in
-.IR format.
-The
-.I timestamp
-should be of the same form as returned by
-.BR systime() .
-If
-.I timestamp
-is missing, the current time of day is used.
-If
-.I format
-is missing, a default format equivalent to the output of
-.IR date (1)
-will be used.
-See the specification for the
-.B strftime()
-function in \*(AN C for the format conversions that are
-guaranteed to be available.
-A public-domain version of
-.IR strftime (3)
-and a man page for it come with
-.IR gawk ;
-if that version was used to build
-.IR gawk ,
-then all of the conversions described in that man page are available to
-.IR gawk.
-.SS String Constants
-.PP
-String constants in \*(AK are sequences of characters enclosed
-between double quotes (\fB"\fR).  Within strings, certain
-.I "escape sequences"
-are recognized, as in C.  These are:
-.PP
-.TP \w'\fB\e\^\fIddd\fR'u+1n
-.B \e\e
-A literal backslash.
-.TP
-.B \ea
-The \*(lqalert\*(rq character; usually the \s-1ASCII\s+1 \s-1BEL\s+1 character.
-.TP
-.B \eb
-backspace.
-.TP
-.B \ef
-form-feed.
-.TP
-.B \en
-newline.
-.TP
-.B \er
-carriage return.
-.TP
-.B \et
-horizontal tab.
-.TP
-.B \ev
-vertical tab.
-.TP
-.BI \ex "\^hex digits"
-The character represented by the string of hexadecimal digits following
-the
-.BR \ex .
-As in \*(AN C, all following hexadecimal digits are considered part of
-the escape sequence.
-(This feature should tell us something about language design by committee.)
-E.g., \fB"\ex1B"\fR is the \s-1ASCII\s+1 \s-1ESC\s+1 (escape) character.
-.TP
-.BI \e ddd
-The character represented by the 1-, 2-, or 3-digit sequence of octal
-digits.
-E.g., \fB"\e033"\fR is the \s-1ASCII\s+1 \s-1ESC\s+1 (escape) character.
-.TP
-.BI \e c
-The literal character
-.IR c\^ .
-.PP
-The escape sequences may also be used inside constant regular expressions
-(e.g.,
-.B "/[\ \et\ef\en\er\ev]/"
-matches whitespace characters).
-.PP
-In compatibility mode, the characters represented by octal and
-hexadecimal escape sequences are treated literally when used in
-regexp constants.  Thus,
-.B /a\e52b/
-is equivalent to
-.BR /a\e*b/ .
-.SH FUNCTIONS
-Functions in \*(AK are defined as follows:
-.PP
-.RS
-\fBfunction \fIname\fB(\fIparameter list\fB) { \fIstatements \fB}\fR
-.RE
-.PP
-Functions are executed when they are called from within expressions
-in either patterns or actions.  Actual parameters supplied in the function
-call are used to instantiate the formal parameters declared in the function.
-Arrays are passed by reference, other variables are passed by value.
-.PP
-Since functions were not originally part of the \*(AK language, the provision
-for local variables is rather clumsy: They are declared as extra parameters
-in the parameter list.  The convention is to separate local variables from
-real parameters by extra spaces in the parameter list.  For example:
-.PP
-.RS
-.ft B
-.nf
-function  f(p, q,     a, b)	# a & b are local
-{
-	\&.\|.\|.
-}
-
-/abc/	{ .\|.\|. ; f(1, 2) ; .\|.\|. }
-.fi
-.ft R
-.RE
-.PP
-The left parenthesis in a function call is required
-to immediately follow the function name,
-without any intervening white space.
-This is to avoid a syntactic ambiguity with the concatenation operator.
-This restriction does not apply to the built-in functions listed above.
-.PP
-Functions may call each other and may be recursive.
-Function parameters used as local variables are initialized
-to the null string and the number zero upon function invocation.
-.PP
-Use
-.BI return " expr"
-to return a value from a function.  The return value is undefined if no
-value is provided, or if the function returns by \*(lqfalling off\*(rq the
-end.
-.PP
-If
-.B \-\^\-lint
-has been provided,
-.I gawk
-will warn about calls to undefined functions at parse time,
-instead of at run time.
-Calling an undefined function at run time is a fatal error.
-.PP
-The word
-.B func
-may be used in place of
-.BR function .
-.SH EXAMPLES
-.nf
-Print and sort the login names of all users:
-
-.ft B
-	BEGIN	{ FS = ":" }
-		{ print $1 | "sort" }
-
-.ft R
-Count lines in a file:
-
-.ft B
-		{ nlines++ }
-	END	{ print nlines }
-
-.ft R
-Precede each line by its number in the file:
-
-.ft B
-	{ print FNR, $0 }
-
-.ft R
-Concatenate and line number (a variation on a theme):
-
-.ft B
-	{ print NR, $0 }
-.ft R
-.fi
-.SH SEE ALSO
-.IR egrep (1),
-.IR getpid (2),
-.IR getppid (2),
-.IR getpgrp (2),
-.IR getuid (2),
-.IR geteuid (2),
-.IR getgid (2),
-.IR getegid (2),
-.IR getgroups (2)
-.PP
-.IR "The AWK Programming Language" ,
-Alfred V. Aho, Brian W. Kernighan, Peter J. Weinberger,
-Addison-Wesley, 1988.  ISBN 0-201-07981-X.
-.PP
-.IR "Effective AWK Programming" ,
-Edition 1.0, published by the Free Software Foundation, 1995.
-.SH POSIX COMPATIBILITY
-A primary goal for
-.I gawk
-is compatibility with the \*(PX standard, as well as with the
-latest version of \*(UX
-.IR awk .
-To this end,
-.I gawk
-incorporates the following user visible
-features which are not described in the \*(AK book,
-but are part of the Bell Labs version of
-.IR awk ,
-and are in the \*(PX standard.
-.PP
-The
-.B \-v
-option for assigning variables before program execution starts is new.
-The book indicates that command line variable assignment happens when
-.I awk
-would otherwise open the argument as a file, which is after the
-.B BEGIN
-block is executed.  However, in earlier implementations, when such an
-assignment appeared before any file names, the assignment would happen
-.I before
-the
-.B BEGIN
-block was run.  Applications came to depend on this \*(lqfeature.\*(rq
-When
-.I awk
-was changed to match its documentation, this option was added to
-accommodate applications that depended upon the old behavior.
-(This feature was agreed upon by both the AT&T and \*(GN developers.)
-.PP
-The
-.B \-W
-option for implementation specific features is from the \*(PX standard.
-.PP
-When processing arguments,
-.I gawk
-uses the special option \*(lq\-\^\-\*(rq to signal the end of
-arguments.
-In compatibility mode, it will warn about, but otherwise ignore,
-undefined options.
-In normal operation, such arguments are passed on to the \*(AK program for
-it to process.
-.PP
-The \*(AK book does not define the return value of
-.BR srand() .
-The \*(PX standard
-has it return the seed it was using, to allow keeping track
-of random number sequences.  Therefore
-.B srand()
-in
-.I gawk
-also returns its current seed.
-.PP
-Other new features are:
-The use of multiple
-.B \-f
-options (from MKS
-.IR awk );
-the
-.B ENVIRON
-array; the
-.BR \ea ,
-and
-.BR \ev
-escape sequences (done originally in
-.I gawk
-and fed back into AT&T's); the
-.B tolower()
-and
-.B toupper()
-built-in functions (from AT&T); and the \*(AN C conversion specifications in
-.B printf
-(done first in AT&T's version).
-.SH GNU EXTENSIONS
-.I Gawk
-has a number of extensions to \*(PX
-.IR awk .
-They are described in this section.  All the extensions described here
-can be disabled by
-invoking
-.I gawk
-with the
-.B \-\^\-traditional
-option.
-.PP
-The following features of
-.I gawk
-are not available in
-\*(PX
-.IR awk .
-.RS
-.TP \w'\(bu'u+1n
-\(bu
-The
-.B \ex
-escape sequence.
-(Disabled with
-.BR \-\^\-posix .)
-.TP \w'\(bu'u+1n
-\(bu
-The
-.B fflush()
-function.
-(Disabled with
-.BR \-\^\-posix .)
-.TP
-\(bu
-The
-.BR systime(),
-.BR strftime(),
-and
-.B gensub()
-functions.
-.TP
-\(bu
-The special file names available for I/O redirection are not recognized.
-.TP
-\(bu
-The
-.BR ARGIND ,
-.BR ERRNO ,
-and
-.B RT
-variables are not special.
-.TP
-\(bu
-The
-.B IGNORECASE
-variable and its side-effects are not available.
-.TP
-\(bu
-The
-.B FIELDWIDTHS
-variable and fixed-width field splitting.
-.TP
-\(bu
-The use of
-.B RS
-as a regular expression.
-.TP
-\(bu
-The ability to split out individual characters using the null string
-as the value of
-.BR FS ,
-and as the third argument to
-.BR split() .
-.TP
-\(bu
-No path search is performed for files named via the
-.B \-f
-option.  Therefore the
-.B AWKPATH
-environment variable is not special.
-.TP
-\(bu
-The use of
-.B "nextfile"
-to abandon processing of the current input file.
-.TP
-\(bu
-The use of
-.BI delete " array"
-to delete the entire contents of an array.
-.RE
-.PP
-The AWK book does not define the return value of the
-.B close()
-function.
-.IR Gawk\^ 's
-.B close()
-returns the value from
-.IR fclose (3),
-or
-.IR pclose (3),
-when closing a file or pipe, respectively.
-.PP
-When
-.I gawk
-is invoked with the
-.B \-\^\-traditional
-option,
-if the
-.I fs
-argument to the
-.B \-F
-option is \*(lqt\*(rq, then
-.B FS
-will be set to the tab character.
-Note that typing
-.B "gawk \-F\et \&.\|.\|."
-simply causes the shell to quote the \*(lqt,\*(rq, and does not pass
-\*(lq\et\*(rq to the
-.B \-F
-option.
-Since this is a rather ugly special case, it is not the default behavior.
-This behavior also does not occur if
-.B \-\^\-posix
-has been specified.
-To really get a tab character as the field separator, it is best to use
-quotes:
-.BR "gawk \-F'\et' \&.\|.\|." .
-.ig
-.PP
-If
-.I gawk
-was compiled for debugging, it will
-accept the following additional options:
-.TP
-.PD 0
-.B \-Wparsedebug
-.TP
-.PD
-.B \-\^\-parsedebug
-Turn on
-.IR yacc (1)
-or
-.IR bison (1)
-debugging output during program parsing.
-This option should only be of interest to the
-.I gawk
-maintainers, and may not even be compiled into
-.IR gawk .
-..
-.SH HISTORICAL FEATURES
-There are two features of historical \*(AK implementations that
-.I gawk
-supports.
-First, it is possible to call the
-.B length()
-built-in function not only with no argument, but even without parentheses!
-Thus,
-.RS
-.PP
-.ft B
-a = length	# Holy Algol 60, Batman!
-.ft R
-.RE
-.PP
-is the same as either of
-.RS
-.PP
-.ft B
-a = length()
-.br
-a = length($0)
-.ft R
-.RE
-.PP
-This feature is marked as \*(lqdeprecated\*(rq in the \*(PX standard, and
-.I gawk
-will issue a warning about its use if
-.B \-\^\-lint
-is specified on the command line.
-.PP
-The other feature is the use of either the
-.B continue
-or the
-.B break
-statements outside the body of a
-.BR while ,
-.BR for ,
-or
-.B do
-loop.  Traditional \*(AK implementations have treated such usage as
-equivalent to the
-.B next
-statement.
-.I Gawk
-will support this usage if
-.B \-\^\-traditional
-has been specified.
-.SH ENVIRONMENT VARIABLES
-If
-.B POSIXLY_CORRECT
-exists in the environment, then
-.I gawk
-behaves exactly as if
-.B \-\^\-posix
-had been specified on the command line.
-If
-.B \-\^\-lint
-has been specified,
-.I gawk
-will issue a warning message to this effect.
-.PP
-The
-.B AWKPATH
-environment variable can be used to provide a list of directories that
-.I gawk
-will search when looking for files named via the
-.B \-f
-and
-.B \-\^\-file
-options.
-.SH BUGS
-The
-.B \-F
-option is not necessary given the command line variable assignment feature;
-it remains only for backwards compatibility.
-.PP
-If your system actually has support for
-.B /dev/fd
-and the associated
-.BR /dev/stdin ,
-.BR /dev/stdout ,
-and
-.B /dev/stderr
-files, you may get different output from
-.I gawk
-than you would get on a system without those files.  When
-.I gawk
-interprets these files internally, it synchronizes output to the standard
-output with output to
-.BR /dev/stdout ,
-while on a system with those files, the output is actually to different
-open files.
-Caveat Emptor.
-.PP
-Syntactically invalid single character programs tend to overflow
-the parse stack, generating a rather unhelpful message.  Such programs
-are surprisingly difficult to diagnose in the completely general case,
-and the effort to do so really is not worth it.
-.SH VERSION INFORMATION
-This man page documents
-.IR gawk ,
-version 3.0.6.
-.SH AUTHORS
-The original version of \*(UX
-.I awk
-was designed and implemented by Alfred Aho,
-Peter Weinberger, and Brian Kernighan of AT&T Bell Labs.  Brian Kernighan
-continues to maintain and enhance it.
-.PP
-Paul Rubin and Jay Fenlason,
-of the Free Software Foundation, wrote
-.IR gawk ,
-to be compatible with the original version of
-.I awk
-distributed in Seventh Edition \*(UX.
-John Woods contributed a number of bug fixes.
-David Trueman, with contributions
-from Arnold Robbins, made
-.I gawk
-compatible with the new version of \*(UX
-.IR awk .
-Arnold Robbins is the current maintainer.
-.PP
-The initial DOS port was done by Conrad Kwok and Scott Garfinkle.
-Scott Deifik is the current DOS maintainer.  Pat Rankin did the
-port to VMS, and Michal Jaegermann did the port to the Atari ST.
-The port to OS/2 was done by Kai Uwe Rommel, with contributions and
-help from Darrel Hankerson.  Fred Fish supplied support for the Amiga.
-.SH BUG REPORTS
-If you find a bug in
-.IR gawk ,
-please send electronic mail to
-.BR bug-gawk@gnu.org .
-Please include your operating system and its revision, the version of
-.IR gawk ,
-what C compiler you used to compile it, and a test program
-and data that are as small as possible for reproducing the problem.
-.PP
-Before sending a bug report, please do two things.  First, verify that
-you have the latest version of
-.IR gawk .
-Many bugs (usually subtle ones) are fixed at each release, and if
-yours is out of date, the problem may already have been solved.
-Second, please read this man page and the reference manual carefully to
-be sure that what you think is a bug really is, instead of just a quirk
-in the language.
-.PP
-Whatever you do, do
-.B NOT
-post a bug report in
-.BR comp.lang.awk .
-While the
-.I gawk
-developers occasionally read this newsgroup, posting bug reports there
-is an unreliable way to report bugs.  Instead, please use the electronic mail
-addresses given above.
-.SH ACKNOWLEDGEMENTS
-Brian Kernighan of Bell Labs
-provided valuable assistance during testing and debugging.
-We thank him.
-.SH COPYING PERMISSIONS
-Copyright \(co 1996\-2000 Free Software Foundation, Inc.
-.PP
-Permission is granted to make and distribute verbatim copies of
-this manual page provided the copyright notice and this permission
-notice are preserved on all copies.
-.ig
-Permission is granted to process this file through troff and print the
-results, provided the printed document carries copying permission
-notice identical to this one except for the removal of this paragraph
-(this paragraph not being relevant to the printed manual page).
-..
-.PP
-Permission is granted to copy and distribute modified versions of this
-manual page under the conditions for verbatim copying, provided that
-the entire resulting derived work is distributed under the terms of a
-permission notice identical to this one.
-.PP
-Permission is granted to copy and distribute translations of this
-manual page into another language, under the above conditions for
-modified versions, except that this permission notice may be stated in
-a translation approved by the Foundation.
diff --git a/contrib/awk/doc/igawk.1 b/contrib/awk/doc/igawk.1
deleted file mode 100644
index 08173ec182ffa..0000000000000
--- a/contrib/awk/doc/igawk.1
+++ /dev/null
@@ -1,73 +0,0 @@
-.TH IGAWK 1 "Nov 3 1999" "Free Software Foundation" "Utility Commands"
-.SH NAME
-igawk \- gawk with include files
-.SH SYNOPSIS
-.B igawk
-[ all
-.I gawk
-options ]
-.B \-f
-.I program-file
-[
-.B \-\^\-
-] file .\^.\^.
-.br
-.B igawk
-[ all
-.I gawk
-options ]
-[
-.B \-\^\-
-]
-.I program-text
-file .\^.\^.
-.SH DESCRIPTION
-.I Igawk
-is a simple shell script that adds the ability to have ``include files'' to
-.IR gawk (1).
-.PP
-AWK programs for
-.I igawk
-are the same as for
-.IR gawk ,
-except that, in addition, you may have lines like
-.RS
-.sp
-.ft B
-@include getopt.awk
-.ft R
-.sp
-.RE
-in your program to include the file
-.B getopt.awk
-from either the current directory or one of the other directories
-in the search path.
-.SH OPTIONS
-See
-.IR gawk (1)
-for a full description of the AWK language and the options that
-.I gawk
-supports.
-.SH EXAMPLES
-.nf
-.ft B
-cat << EOF > test.awk
-@include getopt.awk
-.sp
-BEGIN {
-	while (getopt(ARGC, ARGV, "am:q") != \-1)
-		\&.\^.\^.
-}
-EOF
-.sp
-igawk \-f test.awk
-.ft R
-.fi
-.SH SEE ALSO
-.IR gawk (1)
-.PP
-.IR "Effective AWK Programming" ,
-Edition 1.0, published by the Free Software Foundation, 1995.
-.SH AUTHOR
-Arnold Robbins
-.RB ( arnold@skeeve.com ).
diff --git a/contrib/awk/doc/macros b/contrib/awk/doc/macros
deleted file mode 100644
index bdfc5c8cbf691..0000000000000
--- a/contrib/awk/doc/macros
+++ /dev/null
@@ -1,211 +0,0 @@
-.\" SSC Reference card macros
-.\"
-.\" Copyright (C) 1996, Specialized System Consultants Inc. (SSC)
-.\" 
-.\" These macros are free software; you can redistribute them and/or modify
-.\" them 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.
-.\" 
-.\" These macros are 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
-.\"
-.\"	Generic SSC "card" macros
-.\"     based on lots of other macros
-.\"     Last update: 4-25-91 ph
-.\"     attempting to get margins in the boxes Aug  3 09:43:48 PDT 1994
-.ll 3i		\" length of text line
-.lt 3.2i	\" length of title line 
-.de BT	\" bottom of page trap
-.sp |8.2i		\" go to where we put footer
-.ie \\n(CL=1 \{\
-.	nr CL 2
-.tl ''\\*(CD\\n+(PN''	\" footer is just page number
-.	po 4i		\" go to second column
-.TP			\" print header if any
-\}
-.el \{\
-.	nr CL 1
-.tl ''\\*(CD\\n+(PN''	\" footer is just page number
-.	po .5i		\" go to first column
-.	bp		\" force a new page (which will force header)
-.	TP
-\}
-..
-.de TP	\" top of page
-.\" .sp |.2i
-.sp |0
-.\" put page header stuff here 
-.\" for example: .tl ''WOW!''
-.\".sp
-..
-.\" .wh 8.1i BT 		\" set bottom of column trap
-.nf			\" don't fill lines
-.nh			\" no hyphenation
-.nr CL 1		\" start with column = 1
-.po .5i			\" offset for first column
-.vs 9			\" line spacing
-.ps 8			\" point size
-.de ST	\" set tabs to normal places
-.ta .2i .78i 1.2i 1.7i	\" set tabs 
-..
-.ig
-    From: bryang@chinet.chi.il.us (Bryan Glennon)
-    Box macro.  Do a .mk z where the box is to start, and a .eb
-    where it is to end.  Optional argument is a title to be centered
-    within the top box line.
-   
-    Usage:
-   
-     .mk z
-     Text, etc to be boxed...
-     .eb "Optional title goes here"
-    
-     ~or~
-   
-     .mk z
-     Text, etc to be boxed...
-     .eb 
-   
-   
-    Some explanation:
-     The macro name is eb <.de eb>.  First thing we do is to go up on line
-     <.sp -1> and turn off fill mode <.nf>.  Now it gets interesting: the
-    .ie is the if/else construct.  We check the number of arguments provided
-    to the macro <\\n(.$> and if it is greater than 0 (meaning we have a title),
-    we do the rest of the .ie line, as follows:
-   
-                         \h'-.5n'     - move left one-half of an n
-                         \L'|\\nzu-1' - draw a vertical line <\L> to the 
-                                        absolute position (|) given by \\nzu-1,
-                                        which is the position set with the .mk 
-                                        command into register z <\\nz> in base 
-                                        units <u> minus 1.
-    \l'(\\n(.lu+1n-\w'\\$1'u/2u)\(ul' - Draw a horizontal line <\l> with length
-                                        equal to the current line length 
-                                        <\\n(.l> in base units <u> plus the 
-                                        space required for an 'n' <1n>, minus 
-                                        the width <\w> of the title string 
-                                        <\\$1> in base units <u> divided by 2 
-                                        <in base units <u>>.  Draw the line 
-                                        using the underline character, <\(ul>.
-             \v'.3m'\|\\$1\|\v'-.3m' -  Move down the page 3/10 of an m,
-                                        <\v'.3m'>, output a 1/6 of an m space
-                                        <\|>, output the title <\\$1>, another
-                                        1/6 of an m space <\|>, and then move
-                                        up the page 3/10 of an m <\v'-.3m'>.
-                         \l'...\(ul' -  Draw the second part of the line, just
-                                        like the corresponding left half done
-                                        before.
-                       \L'-|\\nzu+1' -  Draw a verticle line <\L> going down
-                                        the absolute distance <-|> from where 
-                                        the macro was given to where the start 
-                                        point was marked <\\nz> in base units
-                                        <u> plus one line <+1>
-                     \l'|0u-.5n\(ul' -  Draw a horizontal line to the absolute
-                                        position (|0) when the macro was 
-                                        invoked, minus half an n <-.5n> using
-                                        the underline character <\(ul>.
-   
-       The .el beings the else part, which is identical to the above, except
-       the string dosen't get printed.  This makes the printing of the top
-       line much easier: just draw a line <\l> with width equal to the
-       current line plus the witdh of an n <\\n(.l+1n> using the underline
-       character <.\(ul>.
-..
-.de ES	\" start "text in a box"
-.mk z
-.in +.5n
-.ll -.5n
-.sp 1.3
-..
-.de EB	\" end "text in a box" -- optional box title as argument
-.sp -.6
-.nf
-.in -.5n
-.ll +.5n
-.ie \\n(.$\
-\L'|\\nzu'\
-\l'(\\n(.lu-\w'\\$1'u)/2u-.33m\(ul'\
-\v'.3m'\|\\$1\|\v'-.3m'\
-\l'(\\n(.lu-\w'\\$1'u)/2u\(ul'\
-\L'-|\\nzu'\
-\l'|0u\(ul'
-.el \h'-.5n'\L'|\\nzu-1'\l'\\n(.lu+1n\(ul'\L'-|\\nzu+1'\l'|0u-.5n\(ul'
-.in 0
-..
-.de SL	\" draw single line (works in non-fill mode only)
-.sp -.8
-.ti 0
-\l'\\n(.lu\(ul'
-..
-.de Hl \" draw horizontal line
-.br
-.ti 0
-\l'\\n(.lu-\\n(.iu'
-.br
-..
-.de DL	\" draw double line (works in non-fill mode only)
-.sp -.8
-.ti 0
-\l'\\n(.lu\(ul'
-.sp -.8
-.ti 0
-\l'\\n(.lu\(ul'
-..
-.ST
-.nr PN 0 1	\" sets starting page number and auto-increment
-.\" must define page header (if any) before here
-.TP
-.ds 3) \|\v'3p'\s+5\z\(sq\s0\v'-3p'\h'1.25p'\v'-.5p'3\v'.5p'\h'2p'
-.\" old one .ds 2) \h'-1.5p'\v'1p'\s+4\z\(ci\s0\v'-1p'\h'3.25p'2
-.ds 2) \|\v'-2.4p'\D'c.095id'\h'-5.15p'\v'2.4p'2\h'1.9p'
-.ds dC \v'1p'\s+5\(bu\s0\v'-1p'\" for development commands
-.ds tC \s+2\(dm\s0\" (for DWB) should be a triangle
-.ds tP \s+2\(dm\s0\" (for other text processing) should be a triangle
-.\" various trademark symbols
-.ds Tm \v'-0.5m'\s8TM\s0\v'0.5m'
-.ds Ts \v'-0.5m'\s4TM\s0\v'0.5m'
-.ig ++
-.\" mount Serifa fonts
-.fp 5 SR
-.fp 6 SB
-.fp 4 Si
-.++
-.\" other assorted junk
-.lg 0
-.\" Fl requires extended version of troff
-.de Fl \" draw fat horizontal line
-.br
-.ti 0
-.ruw 1.5p
-\l'\\n(.lu-\\n(.iu'
-.br
-.ruw
-..
-.de Bx \" box for keys in text
-\\$3\&\|\&\c
-\s-3\(br\|\fH\v'.18n'\\$1\v'-.18n\fP\|\(br\l'|0\(rn'\l'|0\(ul'\&\s0\|\\$2
-..
-.de Fn	\" function name - left justified, gray background
-.\" bold with gray for function name
-.ns
-.br
-\ 
-.ns
-.br
-\!! gsave ( ) stringwidth neg 0 rmoveto
-\!! /Serifa-Bold findfont 8 scalefont setfont
-\!! (\\$1) dup stringwidth pop 6 gsave dup 0 exch rlineto neg exch 0 rlineto
-\!! 0 exch rlineto closepath .9 setgray fill grestore show
-\!! grestore
-.nf
-.rs
-..
-.rs
diff --git a/contrib/awk/doc/no.colors b/contrib/awk/doc/no.colors
deleted file mode 100644
index 974f985b349db..0000000000000
--- a/contrib/awk/doc/no.colors
+++ /dev/null
@@ -1,31 +0,0 @@
-.\" AWK Reference Card --- Arnold Robbins, arnold@gnu.org
-.\" This file is for troff which does not know what to do
-.\" with a literal Poscript and cannot use macros from 'colors'.
-.\"
-.\" Copyright (C) 1996 Free Software Foundation, Inc.
-.\" 
-.\" Permission is granted to make and distribute verbatim copies of
-.\" this reference card provided the copyright notice and this permission
-.\" notice are preserved on all copies.
-.\" 
-.\" Permission is granted to process this file through troff and print the
-.\" results, provided the printed document carries copying permission
-.\" notice identical to this one except for the removal of this paragraph
-.\" (this paragraph not being relevant to the printed reference card).
-.\" 
-.\" Permission is granted to copy and distribute modified versions of this
-.\" reference card under the conditions for verbatim copying, provided that
-.\" the entire resulting derived work is distributed under the terms of a
-.\" permission notice identical to this one.
-.\" 
-.\" Permission is granted to copy and distribute translations of this
-.\" reference card into another language, under the above conditions for
-.\" modified versions, except that this permission notice may be stated in
-.\" a translation approved by the Foundation.
-.\"
-.ds CR
-.ds CG
-.ds CL
-.ds CB
-.ds CD
-.ds CX
diff --git a/contrib/awk/doc/setter.outline b/contrib/awk/doc/setter.outline
deleted file mode 100644
index 67ade73a178a6..0000000000000
--- a/contrib/awk/doc/setter.outline
+++ /dev/null
@@ -1,77 +0,0 @@
-%!PS-Adobe-3.0
-% SSC Reference card typesetter outline / cut marks
-%
-% Copyright (C) 1996, Specialized System Consultants Inc. (SSC)
-% 
-% This file is free software; you can redistribute it and/or modify
-% it under the terms of the GNU General Public License as published by
-% the Free Software Foundation; either version 2 of the License, or
-% (at your option) any later version.
-% 
-% This 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 program; if not, write to the Free Software
-% Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA  02111-1307, USA
-%
-%! page cut marks and stuff for Pocket References - 10-26-88 - ph
-%! modified to move the cut marks onto the page
-%! center a string
-/inch {72 mul} def
-/cshow  % stk: string 
-	% center string in space (space us variable)
-	{
-	dup stringwidth pop	% get length of string
-	space exch sub 2 div	% compute initial space needed
-	0 rmoveto		% move over
-	show
-	} def
-
-/flashme
-{	/space 612 def
-	0 0 0 setrgbcolor	% always print this stuff
-
-	/Helvetica findfont 12 scalefont setfont
-	gsave
-% for groff (I hope)
-	-6 -6 translate
-	0.2 setlinewidth
-
-	0.25 inch 10.5 inch moveto
-	0.5 inch 10.5 inch lineto
-	.75 inch 10.75 inch moveto
-	.75 inch 11 inch lineto
-	stroke
-
-	0.25 inch 2 inch moveto
-	0.5 inch 2 inch lineto
-	.75 inch 1.75 inch moveto
-	.75 inch 1.50 inch lineto
-	stroke
-	4.25 inch 11 inch moveto
-	4.25 inch 10.75 inch lineto
-	stroke
-
-	4.25 inch 1.75 inch moveto	
-	4.25 inch 1.5 inch lineto
-	stroke
-
-	7.75 inch 1.5 inch moveto
-	7.75 inch 1.75 inch lineto
-	8 inch 2 inch moveto
-	8.25 inch 2 inch lineto
-	stroke
-
-	7.75 inch 11 inch moveto
-	7.75 inch 10.75 inch lineto
-	8 inch 10.5 inch moveto
-	8.25 inch 10.5 inch lineto
-	stroke
-	grestore
-	} def
-
-% actually do something
-
diff --git a/contrib/awk/doc/texinfo.tex b/contrib/awk/doc/texinfo.tex
deleted file mode 100644
index ebf58d8e8345e..0000000000000
--- a/contrib/awk/doc/texinfo.tex
+++ /dev/null
@@ -1,5992 +0,0 @@
-% 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{1999-10-01.07}
-%
-% Copyright (C) 1985, 86, 88, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99
-% 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.tex
-%   (and all GNU mirrors, see http://www.gnu.org/order/ftp.html)
-%   ftp://texinfo.org/tex/texinfo.tex
-%   ftp://us.ctan.org/macros/texinfo/texinfo.tex
-%   (and all CTAN mirrors, finger ctan@us.ctan.org for a list).
-%   /home/gd/gnu/doc/texinfo.tex on the GNU machines.
-% The texinfo.tex in any given Texinfo distribution could well be out
-% of date, so if that's what you're using, please check.
-% Texinfo has a small home page at http://texinfo.org/.
-%
-% 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, to process the dvi file; this makes foo.ps.
-% The extra runs of TeX 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.  You can get
-% the existing language-specific files from ftp://ftp.gnu.org/gnu/texinfo/.
-
-\message{Loading texinfo [version \texinfoversion]:}
-
-% If in a .fmt file, print the version number
-% and turn on active characters that we couldn't do earlier because
-% they might have appeared in the input file name.
-\everyjob{\message{[Texinfo version \texinfoversion]}%
-  \catcode`+=\active \catcode`\_=\active}
-
-% Save some parts of plain tex 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\ptexi=\i
-\let\ptexlbrace=\{
-\let\ptexrbrace=\}
-\let\ptexstar=\*
-\let\ptext=\t
-
-% We never want plain's outer \+ definition in Texinfo.
-% For @tex, we can use \tabalign.
-\let\+ = \relax
-
-\message{Basics,}
-\chardef\other=12
-
-% 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
-
-% Ignore a token.
-%
-\def\gobble#1{}
-
-\hyphenation{ap-pen-dix}
-\hyphenation{mini-buf-fer mini-buf-fers}
-\hyphenation{eshell}
-\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.
-%
-\def\gloggingall{\begingroup \globaldefs = 1 \loggingall \endgroup}%
-\ifx\eTeXversion\undefined
-\def\loggingall{\tracingcommands2 \tracingstats2
-   \tracingpages1 \tracingoutput1 \tracinglostchars1
-   \tracingmacros2 \tracingparagraphs1 \tracingrestores1
-   \showboxbreadth\maxdimen\showboxdepth\maxdimen
-}%
-\else
-\def\loggingall{\tracingcommands3 \tracingstats2
-   \tracingpages1 \tracingoutput1 \tracinglostchars1
-   \tracingmacros2 \tracingparagraphs1 \tracingrestores1
-   \tracingscantokens1 \tracingassigns1 \tracingifs1
-   \tracinggroups1 \tracingnesting2
-   \showboxbreadth\maxdimen\showboxdepth\maxdimen
-}%
-\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{%
-      \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
-      %
-      \ifpdfmakepagedest \pdfmkdest{\the\pageno} \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 \turnoffactive
-  \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}}%
-}
-
-
-% Single-spacing is done by various environments (specifically, in
-% \nonfillstart and \quotations).
-\newskip\singlespaceskip \singlespaceskip = 12.5pt
-\def\singlespace{%
-  % Why was this kern here?  It messes up equalizing space above and below
-  % environments.  --karl, 6may93
-  %{\advance \baselineskip by -\singlespaceskip
-  %\kern \baselineskip}%
-  \setleading \singlespaceskip
-}
-
-%% 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 actual \{ & \} command in an index.
-  \catcode`\{ = 12 \catcode`\} = 12
-  \catcode`\[ = 1 \catcode`\] = 2
-  \catcode`\@ = 0 \catcode`\\ = 12
-  @gdef@lbracecmd[\{]%
-  @gdef@rbracecmd[\}]%
-@endgroup
-
-% Accents: @, @dotaccent @ringaccent @ubaraccent @udotaccent
-% Others are defined by plain TeX: @` @' @" @^ @~ @= @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 (and 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}
-
-% @. 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.
-%
-\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.
-    \endgroup         % End the \group.
-  }%
-  %
-  \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{TEXT} puts TEXT in the margin next to the current paragraph.
-
-\def\inmargin#1{%
-\strut\vadjust{\nobreak\kern-\strutdepth
-  \vtop to \strutdepth{\baselineskip\strutdepth\vss
-  \llap{\rightskip=\inmarginspacing \vbox{\noindent #1}}\null}}}
-\newskip\inmarginspacing \inmarginspacing=1cm
-\def\strutdepth{\dp\strutbox}
-
-%\hbox{{\rm#1}}\hfil\break}}
-
-% @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`\\=12
-  \catcode`~=12
-  \catcode`^=12
-  \catcode`_=12
-  \catcode`|=12
-  \catcode`<=12
-  \catcode`>=12
-  \catcode`+=12
-  \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}%
-  \input\thisfile
-\endgroup}
-
-\def\thisfile{}
-
-% @center line   outputs that line, centered
-
-\def\center{\parsearg\centerzzz}
-\def\centerzzz #1{{\advance\hsize by -\leftskip
-\advance\hsize by -\rightskip
-\centerline{#1}}}
-
-% @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.
-% We cannot 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
-}
-
-% @asis just yields its argument.  Used with @table, for example.
-%
-\def\asis#1{#1}
-
-% @math means output in math mode.
-% We don't use $'s directly in the definition of \math because control
-% sequences like \math are expanded when the toc file is written.  Then,
-% we read the toc file back, the $'s will be normal characters (as they
-% should be, according to the definition of Texinfo).  So we must use a
-% control sequence to switch into and out of math mode.
-%
-% This isn't quite enough for @math to work properly in indices, but it
-% seems unlikely it will ever be needed there.
-%
-\let\implicitmath = $
-\def\math#1{\implicitmath #1\implicitmath}
-
-% @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}%
-    \ifnum\pdftexversion < 14
-      \pdfimage
-    \else
-      \pdfximage
-    \fi
-      \ifx\empty\imagewidth\else width \imagewidth \fi
-      \ifx\empty\imageheight\else height \imageheight \fi
-      {#1.pdf}%
-    \ifnum\pdftexversion < 14 \else
-      \pdfrefximage \pdflastximage
-    \fi}
-  \def\pdfmkdest#1{\pdfdest name{#1@} xyz}
-  \def\pdfmkpgn#1{#1@}
-  \let\linkcolor = \Cyan
-  \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\bgroup
-      \closein 1 
-      \indexnofonts
-      \def\tt{}
-      % 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\unnumbchapentry ##1##2{}
-      \def\secentry ##1##2##3##4{\advancenumber{chap##2}}
-      \def\unnumbsecentry ##1##2{}
-      \def\subsecentry ##1##2##3##4##5{\advancenumber{sec##2.##3}}
-      \def\unnumbsubsecentry ##1##2{}
-      \def\subsubsecentry ##1##2##3##4##5##6{\advancenumber{subsec##2.##3.##4}}
-      \def\unnumbsubsubsecentry ##1##2{}
-      \input \jobname.toc
-      \def\chapentry ##1##2##3{%
-        \pdfoutline goto name{\pdfmkpgn{##3}}count-\expnumber{chap##2}{##1}}
-      \def\unnumbchapentry ##1##2{%
-        \pdfoutline goto name{\pdfmkpgn{##2}}{##1}}
-      \def\secentry ##1##2##3##4{%
-        \pdfoutline goto name{\pdfmkpgn{##4}}count-\expnumber{sec##2.##3}{##1}}
-      \def\unnumbsecentry ##1##2{%
-        \pdfoutline goto name{\pdfmkpgn{##2}}{##1}}
-      \def\subsecentry ##1##2##3##4##5{%
-        \pdfoutline goto name{\pdfmkpgn{##5}}count-\expnumber{subsec##2.##3.##4}{##1}}
-      \def\unnumbsubsecentry ##1##2{%
-        \pdfoutline goto name{\pdfmkpgn{##2}}{##1}}
-      \def\subsubsecentry ##1##2##3##4##5##6{%
-        \pdfoutline goto name{\pdfmkpgn{##6}}{##1}}
-      \def\unnumbsubsubsecentry ##1##2{%
-        \pdfoutline goto name{\pdfmkpgn{##2}}{##1}}
-      \input \jobname.toc
-    \egroup\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\@{@}%
-      \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{\mkpgn{#1}}
-    \linkcolor #1\endlink}
-  \def\mkpgn#1{#1@} 
-  \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}
-
-% Use Computer Modern fonts at \magstephalf (11pt).
-\newcount\mainmagstep
-\mainmagstep=\magstephalf
-
-% 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}
-
-\ifx\bigger\relax
-\let\mainmagstep=\magstep1
-\setfont\textrm\rmshape{12}{1000}
-\setfont\texttt\ttshape{12}{1000}
-\else
-\setfont\textrm\rmshape{10}{\mainmagstep}
-\setfont\texttt\ttshape{10}{\mainmagstep}
-\fi
-% Instead of cmb10, you many want to use cmbx10.
-% cmbx10 is a prettier font on its own, but cmb10
-% looks better when embedded in a line with cmr10.
-\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 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}
-
-% 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
-
-% \setfont\ssecrm\bxshape{10}{\magstep1}    % This size an font looked bad.
-% \setfont\ssecit\itshape{10}{\magstep1}    % The letters were too crowded.
-% \setfont\ssecsl\slshape{10}{\magstep1}
-% \setfont\ssectt\ttshape{10}{\magstep1}
-% \setfont\ssecsf\sfshape{10}{\magstep1}
-
-%\setfont\ssecrm\bfshape{10}{1315}      % Note the use of cmb rather than cmbx.
-%\setfont\ssecit\itshape{10}{1315}      % Also, the size is a little larger than
-%\setfont\ssecsl\slshape{10}{1315}      % being scaled magstep1.
-%\setfont\ssectt\ttshape{10}{1315}
-%\setfont\ssecsf\sfshape{10}{1315}
-
-%\let\ssecbf=\ssecrm
-
-% 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, 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}
-\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{11pt}}
-
-% 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}
-
-%% 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\/\fi\fi\fi}
-\def\smartslanted#1{{\sl #1}\futurelet\next\smartitalicx}
-\def\smartitalic#1{{\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 = `- }
-
-\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{\ifusingtt{\normalunderscore\discretionary{}{}{}}{\_}}
-\def\codex #1{\tclose{#1}\endgroup}
-
-%\let\exp=\tclose  %Was temporary
-
-% @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}%
-  \fi\fi\fi
-}
-\def\worddistinct{distinct}
-\def\wordexample{example}
-\def\wordcode{code}
-
-% Default is kbdinputdistinct.  (Too much of a hassle to call the macro,
-% the catcodes are wrong for parsearg to work.)
-\gdef\kbdexamplefont{\ttsl}\gdef\kbdfont{\ttsl}
-
-\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\$}}
-
-
-\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}%
-   %
-   % 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
-   %
-   % 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
-   %
-   \ifpdf \pdfmakepagedesttrue \fi
-   %
-   \HEADINGSon
-}
-
-\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
-% Produces Day Month Year style of output.
-\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}
-
-% @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.
-    \nobreak
-    \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}%
-\let\item=\itemizeitem}
-
-% Set sfcode to normal for the chars that usually have another value.
-% These are `.?!:;,'
-\def\frenchspacing{\sfcode46=1000 \sfcode63=1000 \sfcode33=1000
-  \sfcode58=1000 \sfcode59=1000 \sfcode44=1000 }
-
-% \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 }% 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
-}
-
-% This used to have \hskip1sp.  But then the space in a template line is
-% not enough.  That is bad.  So let's go back to just & until we
-% encounter the problem it was intended to solve again.
-% --karl, nathan@acm.org, 20apr99.
-\def\tab{&}
-
-% @multitable ... @end multitable definitions:
-%
-\def\multitable{\parsearg\dotable}
-\def\dotable#1{\bgroup
-  \vskip\parskip
-  \let\item\crcr
-  \tolerance=9500
-  \hbadness=9500
-  \setmultitablespacing
-  \parskip=\multitableparskip
-  \parindent=\multitableparindent
-  \overfullrule=0pt
-  \global\colcount=0
-  \def\Emultitable{\global\setpercentfalse\cr\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}
-
-
-\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.
-%
-\def\ignoremorecommands{%
-  \let\defcodeindex = \relax
-  \let\defcv = \relax
-  \let\deffn = \relax
-  \let\deffnx = \relax
-  \let\defindex = \relax
-  \let\defivar = \relax
-  \let\defmac = \relax
-  \let\defmethod = \relax
-  \let\defop = \relax
-  \let\defopt = \relax
-  \let\defspec = \relax
-  \let\deftp = \relax
-  \let\deftypefn = \relax
-  \let\deftypefun = \relax
-  \let\deftypeivar = \relax
-  \let\deftypeop = \relax
-  \let\deftypevar = \relax
-  \let\deftypevr = \relax
-  \let\defun = \relax
-  \let\defvar = \relax
-  \let\defvr = \relax
-  \let\ref = \relax
-  \let\xref = \relax
-  \let\printindex = \relax
-  \let\pxref = \relax
-  \let\settitle = \relax
-  \let\setchapternewpage = \relax
-  \let\setchapterstyle = \relax
-  \let\everyheading = \relax
-  \let\evenheading = \relax
-  \let\oddheading = \relax
-  \let\everyfooting = \relax
-  \let\evenfooting = \relax
-  \let\oddfooting = \relax
-  \let\headings = \relax
-  \let\include = \relax
-  \let\lowersections = \relax
-  \let\down = \relax
-  \let\raisesections = \relax
-  \let\up = \relax
-  \let\set = \relax
-  \let\clear = \relax
-  \let\item = \relax
-}
-
-% Ignore @ignore ... @end ignore.
-%
-\def\ignore{\doignore{ignore}}
-
-% Ignore @ifinfo, @ifhtml, @ifnottex, @html, @menu, and @direntry text.
-%
-\def\ifinfo{\doignore{ifinfo}}
-\def\ifhtml{\doignore{ifhtml}}
-\def\ifnottex{\doignore{ifnottex}}
-\def\html{\doignore{html}}
-\def\menu{\doignore{menu}}
-\def\direntry{\doignore{direntry}}
-
-% @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.
-  \catcode32 = 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
-  %
-  % 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
-  %
-  % And now expand that command.
-  \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/pub/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 change of memory overflow, we follow the approach outlined on
-  % page 401 of the TeXbook: make the current font be a dummy font.
-  %
-  \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 not all sites
-    % might 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 (mostly for their use in smallexample).
-    \let\smallrm=\nullfont \let\smallit=\nullfont \let\smallsl=\nullfont
-    \let\smallbf=\nullfont \let\smalltt=\nullfont \let\smallsc=\nullfont
-    \let\smallsf=\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`\-=12 \catcode`\_=12
-    \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, 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'']}%
-  \else
-    \csname SET#1\endcsname
-  \fi
-}
-
-% @ifset VAR ... @end ifset reads the `...' iff VAR has been defined
-% with @set.
-%
-\def\ifset{\parsearg\ifsetxxx}
-\def\ifsetxxx #1{%
-  \expandafter\ifx\csname SET#1\endcsname\relax
-    \expandafter\ifsetfail
-  \else
-    \expandafter\ifsetsucceed
-  \fi
-}
-\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\ifclearxxx}
-\def\ifclearxxx #1{%
-  \expandafter\ifx\csname SET#1\endcsname\relax
-    \expandafter\ifclearsucceed
-  \else
-    \expandafter\ifclearfail
-  \fi
-}
-\def\ifclearsucceed{\conditionalsucceed{ifclear}}
-\def\ifclearfail{\nestedignore{ifclear}}
-\defineunmatchedend{ifclear}
-
-% @iftex, @ifnothtml, @ifnotinfo 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}}
-\defineunmatchedend{iftex}
-\defineunmatchedend{ifnothtml}
-\defineunmatchedend{ifnotinfo}
-
-% We can't just want to start a group at @iftex (for example) and end it
-% at @end iftex, since then @set commands inside the conditional have no
-% effect (they'd get reverted at the end of the group).  So we must
-% define \Eiftex to redefine itself to be its previous value.  (We can't
-% just define it to fail again with an ``unmatched end'' error, since
-% the @ifset might be nested.)
-%
-\def\conditionalsucceed#1{%
-  \edef\temp{%
-    % Remember the current value of \E#1.
-    \let\nece{prevE#1} = \nece{E#1}%
-    %
-    % At the `@end #1', redefine \E#1 to be its previous value.
-    \def\nece{E#1}{\let\nece{E#1} = \nece{prevE#1}}%
-  }%
-  \temp
-}
-
-% We need to expand lots of \csname's, but we don't want to expand the
-% control sequences after we've constructed them.
-%
-\def\nece#1{\expandafter\noexpand\csname#1\endcsname}
-
-% @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\newcodeindex#1{%
-  \iflinks
-    \expandafter\newwrite \csname#1indfile\endcsname
-    \openout \csname#1indfile\endcsname \jobname.#1
-  \fi
-  \expandafter\xdef\csname#1index\endcsname{%
-    \noexpand\docodeindex{#1}}
-}
-
-\def\defcodeindex{\parsearg\newcodeindex}
-
-% @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.
-% The \closeout helps reduce unnecessary open files; the limit on the
-% Acorn RISC OS is a mere 16 files.
-\def\synindex#1 #2 {%
-  \expandafter\let\expandafter\synindexfoo\expandafter=\csname#2indfile\endcsname
-  \expandafter\closeout\csname#1indfile\endcsname
-  \expandafter\let\csname#1indfile\endcsname=\synindexfoo
-  \expandafter\xdef\csname#1index\endcsname{% define \xxxindex
-    \noexpand\doindex{#2}}%
-}
-
-% @syncodeindex foo bar   similar, but put all entries made for index foo
-% inside @code.
-\def\syncodeindex#1 #2 {%
-  \expandafter\let\expandafter\synindexfoo\expandafter=\csname#2indfile\endcsname
-  \expandafter\closeout\csname#1indfile\endcsname
-  \expandafter\let\csname#1indfile\endcsname=\synindexfoo
-  \expandafter\xdef\csname#1index\endcsname{% define \xxxindex
-    \noexpand\docodeindex{#2}}%
-}
-
-% 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}}}
-
-\def\indexdummies{%
-\def\ { }%
-% Take care of the plain tex accent commands.
-\def\"{\realbackslash "}%
-\def\`{\realbackslash `}%
-\def\'{\realbackslash '}%
-\def\^{\realbackslash ^}%
-\def\~{\realbackslash ~}%
-\def\={\realbackslash =}%
-\def\b{\realbackslash b}%
-\def\c{\realbackslash c}%
-\def\d{\realbackslash d}%
-\def\u{\realbackslash u}%
-\def\v{\realbackslash v}%
-\def\H{\realbackslash H}%
-% Take care of the plain tex special European modified letters.
-\def\oe{\realbackslash oe}%
-\def\ae{\realbackslash ae}%
-\def\aa{\realbackslash aa}%
-\def\OE{\realbackslash OE}%
-\def\AE{\realbackslash AE}%
-\def\AA{\realbackslash AA}%
-\def\o{\realbackslash o}%
-\def\O{\realbackslash O}%
-\def\l{\realbackslash l}%
-\def\L{\realbackslash L}%
-\def\ss{\realbackslash ss}%
-% Take care of texinfo commands likely to appear in an index entry.
-% (Must be a way to avoid doing expansion at all, and thus not have to
-% laboriously list every single command here.)
-\def\@{@}% will be @@ when we switch to @ as escape char.
-% 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
-\def\_{{\realbackslash _}}%
-\def\w{\realbackslash w }%
-\def\bf{\realbackslash bf }%
-%\def\rm{\realbackslash rm }%
-\def\sl{\realbackslash sl }%
-\def\sf{\realbackslash sf}%
-\def\tt{\realbackslash tt}%
-\def\gtr{\realbackslash gtr}%
-\def\less{\realbackslash less}%
-\def\hat{\realbackslash hat}%
-\def\TeX{\realbackslash TeX}%
-\def\dots{\realbackslash dots }%
-\def\result{\realbackslash result}%
-\def\equiv{\realbackslash equiv}%
-\def\expansion{\realbackslash expansion}%
-\def\print{\realbackslash print}%
-\def\error{\realbackslash error}%
-\def\point{\realbackslash point}%
-\def\copyright{\realbackslash copyright}%
-\def\tclose##1{\realbackslash tclose {##1}}%
-\def\code##1{\realbackslash code {##1}}%
-\def\uref##1{\realbackslash uref {##1}}%
-\def\url##1{\realbackslash url {##1}}%
-\def\env##1{\realbackslash env {##1}}%
-\def\command##1{\realbackslash command {##1}}%
-\def\option##1{\realbackslash option {##1}}%
-\def\dotless##1{\realbackslash dotless {##1}}%
-\def\samp##1{\realbackslash samp {##1}}%
-\def\,##1{\realbackslash ,{##1}}%
-\def\t##1{\realbackslash t {##1}}%
-\def\r##1{\realbackslash r {##1}}%
-\def\i##1{\realbackslash i {##1}}%
-\def\b##1{\realbackslash b {##1}}%
-\def\sc##1{\realbackslash sc {##1}}%
-\def\cite##1{\realbackslash cite {##1}}%
-\def\key##1{\realbackslash key {##1}}%
-\def\file##1{\realbackslash file {##1}}%
-\def\var##1{\realbackslash var {##1}}%
-\def\kbd##1{\realbackslash kbd {##1}}%
-\def\dfn##1{\realbackslash dfn {##1}}%
-\def\emph##1{\realbackslash emph {##1}}%
-\def\acronym##1{\realbackslash acronym {##1}}%
-%
-% 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
-%
-\unsepspaces
-% Turn off 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 no-ops all font-change commands.
-% This is used when outputting the strings to sort the index by.
-\def\indexdummyfont#1{#1}
-\def\indexdummytex{TeX}
-\def\indexdummydots{...}
-
-\def\indexnofonts{%
-% Just ignore accents.
-\let\,=\indexdummyfont
-\let\"=\indexdummyfont
-\let\`=\indexdummyfont
-\let\'=\indexdummyfont
-\let\^=\indexdummyfont
-\let\~=\indexdummyfont
-\let\==\indexdummyfont
-\let\b=\indexdummyfont
-\let\c=\indexdummyfont
-\let\d=\indexdummyfont
-\let\u=\indexdummyfont
-\let\v=\indexdummyfont
-\let\H=\indexdummyfont
-\let\dotless=\indexdummyfont
-% Take care of the plain tex special European modified letters.
-\def\oe{oe}%
-\def\ae{ae}%
-\def\aa{aa}%
-\def\OE{OE}%
-\def\AE{AE}%
-\def\AA{AA}%
-\def\o{o}%
-\def\O{O}%
-\def\l{l}%
-\def\L{L}%
-\def\ss{ss}%
-\let\w=\indexdummyfont
-\let\t=\indexdummyfont
-\let\r=\indexdummyfont
-\let\i=\indexdummyfont
-\let\b=\indexdummyfont
-\let\emph=\indexdummyfont
-\let\strong=\indexdummyfont
-\let\cite=\indexdummyfont
-\let\sc=\indexdummyfont
-%Don't no-op \tt, since it isn't a user-level command
-% and is used in the definitions of the active chars like <, >, |...
-%\let\tt=\indexdummyfont
-\let\tclose=\indexdummyfont
-\let\code=\indexdummyfont
-\let\url=\indexdummyfont
-\let\uref=\indexdummyfont
-\let\env=\indexdummyfont
-\let\acronym=\indexdummyfont
-\let\command=\indexdummyfont
-\let\option=\indexdummyfont
-\let\file=\indexdummyfont
-\let\samp=\indexdummyfont
-\let\kbd=\indexdummyfont
-\let\key=\indexdummyfont
-\let\var=\indexdummyfont
-\let\TeX=\indexdummytex
-\let\dots=\indexdummydots
-\def\@{@}%
-}
-
-% To define \realbackslash, we must make \ not be an escape.
-% We must first make another character (@) an escape
-% so we do not become unable to do a definition.
-
-{\catcode`\@=0 \catcode`\\=\other
- @gdef@realbackslash{\}}
-
-\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.
-        %
-        \def\thirdarg{#3}%
-        %
-        % If third arg is present, precede it with space in sort key.
-        \ifx\thirdarg\emptymacro
-          \let\subentry = \empty
-        \else
-          \def\subentry{ #3}%
-        \fi
-        %
-        % First process the index entry with all font commands turned
-        % off to get the string to sort by.
-        {\indexnofonts \xdef\indexsorttmp{#2\subentry}}%
-        %
-        % Now the real index entry with the fonts.
-        \toks0 = {#2}%
-        %
-        % If third (subentry) arg is present, add it to the index
-        % string.  And include a space.
-        \ifx\thirdarg\emptymacro \else
-          \toks0 = \expandafter{\the\toks0 \space #3}%
-        \fi
-        %
-        % 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, 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-\lastskip \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
-  \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 #2\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.)
-  \advance\vsize by -\ht\partialpage
-  \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
-  %
-  % box0 will be the left-hand column, box2 the right.
-  \setbox0=\vsplit255 to\dimen@ \setbox2=\vsplit255 to\dimen@
-  \onepageout\pagesofar
-  \unvbox255
-  \penalty\outputpenalty
-}
-\def\pagesofar{%
-  % Re-output the contents of the output page -- any previous material,
-  % followed by the two boxes we just split, in box0 and box2.
-  \unvbox\partialpage
-  %
-  \hsize = \doublecolumnhsize
-  \wd0=\hsize \wd2=\hsize
-  \hbox to\pagewidth{\box0\hfil\box2}%
-}
-\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
-}
-\def\balancecolumns{%
-  % Called at the end of the double column material.
-  \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
-}
-
-% 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
-}
-
-% 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
-}
-
-% @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}%
-\toks0 = {#1}%
-\edef\temp{\noexpand\writetocentry{\realbackslash chapentry{\the\toks0}%
-                                  {\the\chapno}}}%
-\temp
-\donoderef
-\global\let\section = \numberedsec
-\global\let\subsection = \numberedsubsec
-\global\let\subsubsection = \numberedsubsubsec
-}
-
-\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}{\putwordAppendix{} \appendixletter}%
-\gdef\thissection{#1}%
-\gdef\thischaptername{#1}%
-\xdef\thischapter{\putwordAppendix{} \appendixletter: \noexpand\thischaptername}%
-\toks0 = {#1}%
-\edef\temp{\noexpand\writetocentry{\realbackslash chapentry{\the\toks0}%
-                       {\putwordAppendix{} \appendixletter}}}%
-\temp
-\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}%
-\toks0 = {#1}%
-\edef\temp{\noexpand\writetocentry{\realbackslash unnumbchapentry{\the\toks0}}}%
-\temp
-\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}%
-\toks0 = {#1}%
-\edef\temp{\noexpand\writetocentry{\realbackslash secentry{\the\toks0}%
-                                  {\the\chapno}{\the\secno}}}%
-\temp
-\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}%
-\toks0 = {#1}%
-\edef\temp{\noexpand\writetocentry{\realbackslash secentry{\the\toks0}%
-                                  {\appendixletter}{\the\secno}}}%
-\temp
-\appendixnoderef
-\nobreak
-}
-
-\outer\def\unnumberedsec{\parsearg\unnumberedsecyyy}
-\def\unnumberedsecyyy #1{\unnmhead1{#1}} % normally calls unnumberedseczzz
-\def\unnumberedseczzz #1{%
-\plainsecheading {#1}\gdef\thissection{#1}%
-\toks0 = {#1}%
-\edef\temp{\noexpand\writetocentry{\realbackslash unnumbsecentry{\the\toks0}}}%
-\temp
-\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}%
-\toks0 = {#1}%
-\edef\temp{\noexpand\writetocentry{\realbackslash subsecentry{\the\toks0}%
-                                    {\the\chapno}{\the\secno}{\the\subsecno}}}%
-\temp
-\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}%
-\toks0 = {#1}%
-\edef\temp{\noexpand\writetocentry{\realbackslash subsecentry{\the\toks0}%
-                                {\appendixletter}{\the\secno}{\the\subsecno}}}%
-\temp
-\appendixnoderef
-\nobreak
-}
-
-\outer\def\unnumberedsubsec{\parsearg\unnumberedsubsecyyy}
-\def\unnumberedsubsecyyy #1{\unnmhead2{#1}} %normally calls unnumberedsubseczzz
-\def\unnumberedsubseczzz #1{%
-\plainsubsecheading {#1}\gdef\thissection{#1}%
-\toks0 = {#1}%
-\edef\temp{\noexpand\writetocentry{\realbackslash unnumbsubsecentry%
-                                    {\the\toks0}}}%
-\temp
-\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}%
-\toks0 = {#1}%
-\edef\temp{\noexpand\writetocentry{\realbackslash subsubsecentry{\the\toks0}%
-  {\the\chapno}{\the\secno}{\the\subsecno}{\the\subsubsecno}}}%
-\temp
-\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}%
-\toks0 = {#1}%
-\edef\temp{\noexpand\writetocentry{\realbackslash subsubsecentry{\the\toks0}%
-  {\appendixletter}{\the\secno}{\the\subsecno}{\the\subsubsecno}}}%
-\temp
-\appendixnoderef
-\nobreak
-}
-
-\outer\def\unnumberedsubsubsec{\parsearg\unnumberedsubsubsecyyy}
-\def\unnumberedsubsubsecyyy #1{\unnmhead3{#1}} %normally unnumberedsubsubseczzz
-\def\unnumberedsubsubseczzz #1{%
-\plainsubsubsecheading {#1}\gdef\thissection{#1}%
-\toks0 = {#1}%
-\edef\temp{\noexpand\writetocentry{\realbackslash unnumbsubsubsecentry%
-                                    {\the\toks0}}}%
-\temp
-\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}%
-  }%
-  \ifdim\parskip<10pt \nobreak\kern10pt\nobreak\kern-\parskip\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.
-%
-% We open the .toc file here instead of at @setfilename or any other
-% given time so that @contents can be put in the document anywhere.
-%
-\newif\iftocfileopened
-\def\writetocentry#1{%
-  \iftocfileopened\else
-    \immediate\openout\tocfile = \jobname.toc
-    \global\tocfileopenedtrue
-  \fi
-  \iflinks \write\tocfile{#1{\folio}}\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 \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
-   \pageno = \savepageno
-}
-
-% And just the chapters.
-\def\summarycontents{%
-   \startcontents{\putwordShortTOC}%
-      %
-      \let\chapentry = \shortchapentry
-      \let\unnumbchapentry = \shortunnumberedentry
-      % We want a true roman here for the page numbers.
-      \secfonts
-      \let\rm=\shortcontrm \let\bf=\shortcontbf \let\sl=\shortcontsl
-      \rm
-      \hyphenpenalty = 10000
-      \advance\baselineskip by 1pt % Open it up a little.
-      \def\secentry ##1##2##3##4{}
-      \def\unnumbsecentry ##1##2{}
-      \def\subsecentry ##1##2##3##4##5{}
-      \def\unnumbsubsecentry ##1##2{}
-      \def\subsubsecentry ##1##2##3##4##5##6{}
-      \def\unnumbsubsubsecentry ##1##2{}
-      \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
-   \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, ...
-
-% Chapter-level things, for both the long and short contents.
-\def\chapentry#1#2#3{\dochapentry{#2\labelspace#1}{#3}}
-
-% See comments in \dochapentry re vbox and related settings
-\def\shortchapentry#1#2#3{%
-  \tocentry{\shortchaplabel{#2}\labelspace #1}{\doshortpageno\bgroup#3\egroup}%
-}
-
-% 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{%
-  % Compute width of word "Appendix", may change with language.
-  \setbox0 = \hbox{\shortcontrm \putwordAppendix}%
-  \shortappendixwidth = \wd0
-  %
-  % We typeset #1 in a box of constant width, regardless of the text of
-  % #1, so the chapter titles will come out aligned.
-  \setbox0 = \hbox{#1}%
-  \dimen0 = \ifdim\wd0 > \shortappendixwidth \shortappendixwidth \else 0pt \fi
-  %
-  % This space should be plenty, since a single number is .5em, and the
-  % widest letter (M) is 1em, at least in the Computer Modern fonts.
-  % (This space doesn't include the extra space that gets added after
-  % the label; that gets put in by \shortchapentry above.)
-  \advance\dimen0 by 1.1em
-  \hbox to \dimen0{#1\hfil}%
-}
-
-\def\unnumbchapentry#1#2{\dochapentry{#1}{#2}}
-\def\shortunnumberedentry#1#2{\tocentry{#1}{\doshortpageno\bgroup#2\egroup}}
-
-% Sections.
-\def\secentry#1#2#3#4{\dosecentry{#2.#3\labelspace#1}{#4}}
-\def\unnumbsecentry#1#2{\dosecentry{#1}{#2}}
-
-% Subsections.
-\def\subsecentry#1#2#3#4#5{\dosubsecentry{#2.#3.#4\labelspace#1}{#5}}
-\def\unnumbsubsecentry#1#2{\dosubsecentry{#1}{#2}}
-
-% And subsubsections.
-\def\subsubsecentry#1#2#3#4#5#6{%
-  \dosubsubsecentry{#2.#3.#4.#5\labelspace#1}{#6}}
-\def\unnumbsubsubsecentry#1#2{\dosubsubsecentry{#1}{#2}}
-
-% 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, so 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.
-
-% 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.
-% Furthermore, these definitions must come after we define our fonts.
-\newbox\dblarrowbox    \newbox\longdblarrowbox
-\newbox\pushcharbox    \newbox\bullbox
-\newbox\equivbox       \newbox\errorbox
-
-%{\tentt
-%\global\setbox\dblarrowbox = \hbox to 1em{\hfil$\Rightarrow$\hfil}
-%\global\setbox\longdblarrowbox = \hbox to 1em{\hfil$\mapsto$\hfil}
-%\global\setbox\pushcharbox = \hbox to 1em{\hfil$\dashv$\hfil}
-%\global\setbox\equivbox = \hbox to 1em{\hfil$\ptexequiv$\hfil}
-% Adapted from the manmac format (p.420 of TeXbook)
-%\global\setbox\bullbox = \hbox to 1em{\kern.15em\vrule height .75ex width .85ex
-%                                      depth .1ex\hfil}
-%}
-
-% @point{}, @result{}, @expansion{}, @print{}, @equiv{}.
-\def\point{$\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$\hfil}}
-
-% Adapted from the TeXbook's \boxit.
-{\tentt \global\dimen0 = 3em}% Width of the box.
-\dimen2 = .55pt % Thickness of rules
-% The text. (`r' is open on the right, `e' somewhat less so on the left.)
-\setbox0 = \hbox{\kern-.75pt \tensf error\kern-1.5pt}
-
-\global\setbox\errorbox=\hbox to \dimen0{\hfil
-   \hsize = \dimen0 \advance\hsize by -5.8pt % Space to left+right.
-   \advance\hsize by -2\dimen2 % Rules.
-   \vbox{
-      \hrule height\dimen2
-      \hbox{\vrule width\dimen2 \kern3pt          % Space to left of text.
-         \vtop{\kern2.4pt \box0 \kern2.4pt}% Space above/below.
-         \kern3pt\vrule width\dimen2}% Space to right.
-      \hrule height\dimen2}
-    \hfil}
-
-% The @error{} command.
-\def\error{\leavevmode\lower.7ex\copy\errorbox}
-
-% @tex ... @end tex    escapes into raw Tex temporarily.
-% One exception: @ is still an escape character, so that @end tex works.
-% But \@ or @@ will get a plain tex @ character.
-
-\def\tex{\begingroup
-  \catcode `\\=0 \catcode `\{=1 \catcode `\}=2
-  \catcode `\$=3 \catcode `\&=4 \catcode `\#=6
-  \catcode `\^=7 \catcode `\_=8 \catcode `\~=13 \let~=\tie
-  \catcode `\%=14
-  \catcode 43=12 % plus
-  \catcode`\"=12
-  \catcode`\==12
-  \catcode`\|=12
-  \catcode`\<=12
-  \catcode`\>=12
-  \escapechar=`\\
-  %
-  \let\b=\ptexb
-  \let\bullet=\ptexbullet
-  \let\c=\ptexc
-  \let\,=\ptexcomma
-  \let\.=\ptexdot
-  \let\dots=\ptexdots
-  \let\equiv=\ptexequiv
-  \let\!=\ptexexclam
-  \let\i=\ptexi
-  \let\{=\ptexlbrace
-  \let\+=\tabalign
-  \let\}=\ptexrbrace
-  \let\*=\ptexstar
-  \let\t=\ptext
-  %
-  \def\endldots{\mathinner{\ldots\ldots\ldots\ldots}}%
-  \def\enddots{\relax\ifmmode\endldots\else$\mathsurround=0pt \endldots\,$\fi}%
-  \def\@{@}%
-\let\Etex=\endgroup}
-
-% Define @lisp ... @endlisp.
-% @lisp does a \begingroup so it can rebind things,
-% including the definition of @endlisp (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{{\advance\envskipamount by \parskip
-\endgraf \ifdim\lastskip<\envskipamount
-\removelastskip \penalty-50 \vskip\envskipamount \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
-
-\long\def\cartouche{%
-\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.
-  \singlespace
-  \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}
-
-% @small... is usually equivalent to the non-small (@smallbook
-% redefines).  We must call \example (or whatever) last in the
-% definition, since it reads the return following the @example (or
-% whatever) command.
-%
-% This actually allows (for example) @end display inside an
-% @smalldisplay.  Too bad, but makeinfo will catch the error anyway.
-%
-\def\smalldisplay{\begingroup\def\Esmalldisplay{\nonfillfinish\endgroup}\display}
-\def\smallexample{\begingroup\def\Esmallexample{\nonfillfinish\endgroup}\lisp}
-\def\smallformat{\begingroup\def\Esmallformat{\nonfillfinish\endgroup}\format}
-\def\smalllisp{\begingroup\def\Esmalllisp{\nonfillfinish\endgroup}\lisp}
-
-% Real @smallexample and @smalllisp (when @smallbook): use smaller fonts.
-% Originally contributed by Pavel@xerox.
-\def\smalllispx{\begingroup
-  \def\Esmalllisp{\nonfillfinish\endgroup}%
-  \def\Esmallexample{\nonfillfinish\endgroup}%
-  \smallfonts
-  \lisp
-}
-
-% @display: same as @lisp except keep current font.
-%
-\def\display{\begingroup
-  \nonfillstart
-  \let\Edisplay = \nonfillfinish
-  \gobble
-}
-
-% @smalldisplay (when @smallbook): @display plus smaller fonts.
-%
-\def\smalldisplayx{\begingroup
-  \def\Esmalldisplay{\nonfillfinish\endgroup}%
-  \smallfonts \rm
-  \display
-}
-
-% @format: same as @display except don't narrow margins.
-%
-\def\format{\begingroup
-  \let\nonarrowing = t
-  \nonfillstart
-  \let\Eformat = \nonfillfinish
-  \gobble
-}
-
-% @smallformat (when @smallbook): @format plus smaller fonts.
-%
-\def\smallformatx{\begingroup
-  \def\Esmallformat{\nonfillfinish\endgroup}%
-  \smallfonts \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
-  \singlespace
-  \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
-}
-
-
-\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\deftypemargin \deftypemargin=12pt
-\newskip\deflastargmargin \deflastargmargin=18pt
-
-\newcount\parencount
-% define \functionparens, which makes ( and ) and & do special things.
-% \functionparens affects the group it is contained in.
-\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`& = 13
-  \global\let& = \ampnr
-}
-
-% First, defname, which formats the header line itself.
-% #1 should be the function name.
-% #2 should be the type of definition, such as "Function".
-
-\def\defname #1#2{%
-% Get the values of \leftskip and \rightskip as they were
-% outside the @def...
-\dimen2=\leftskip
-\advance\dimen2 by -\defbodyindent
-\noindent
-\setbox0=\hbox{\hskip \deflastargmargin{\rm #2}\hskip \deftypemargin}%
-\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
-% Now output arg 2 ("Function" or some such)
-% ending at \deftypemargin from the right margin,
-% but stuck inside a box of width 0 so it does not interfere with linebreaking
-{% Adjust \hsize to exclude the ambient margins,
-% so that \rightline will obey them.
-\advance \hsize by -\dimen2
-\rlap{\rightline{{\rm #2}\hskip -1.25pc }}}%
-% Make all lines underfull and no complaints:
-\tolerance=10000 \hbadness=10000
-\advance\leftskip by -\defbodyindent
-\exdentamount=\defbodyindent
-{\df #1}\enskip        % Generate function name
-}
-
-% Actually process the body of a definition
-% #1 should be the terminating control sequence, such as \Edefun.
-% #2 should be the "another name" control sequence, such as \defunx.
-% #3 should be the control sequence that actually processes the header,
-%    such as \defunheader.
-
-\def\defparsebody #1#2#3{\begingroup\inENV% Environment for definitionbody
-\medbreak %
-% Define the end token that this defining construct specifies
-% so that it will exit this group.
-\def#1{\endgraf\endgroup\medbreak}%
-\def#2{\begingroup\obeylines\activeparens\spacesplit#3}%
-\parindent=0in
-\advance\leftskip by \defbodyindent
-\exdentamount=\defbodyindent
-\begingroup %
-\catcode 61=\active % 61 is `='
-\obeylines\activeparens\spacesplit#3}
-
-% #1 is the \E... control sequence to end the definition (which we define).
-% #2 is the \...x control sequence for consecutive fns (which we define).
-% #3 is the control sequence to call to resume processing.
-% #4, delimited by the space, is the class name.
-%
-\def\defmethparsebody#1#2#3#4 {\begingroup\inENV %
-\medbreak %
-% Define the end token that this defining construct specifies
-% so that it will exit this group.
-\def#1{\endgraf\endgroup\medbreak}%
-\def#2##1 {\begingroup\obeylines\activeparens\spacesplit{#3{##1}}}%
-\parindent=0in
-\advance\leftskip by \defbodyindent
-\exdentamount=\defbodyindent
-\begingroup\obeylines\activeparens\spacesplit{#3{#4}}}
-
-% Used for @deftypemethod and @deftypeivar.
-% #1 is the \E... control sequence to end the definition (which we define).
-% #2 is the \...x control sequence for consecutive fns (which we define).
-% #3 is the control sequence to call to resume processing.
-% #4, delimited by a space, is the class name.
-% #5 is the method's return type.
-%
-\def\deftypemethparsebody#1#2#3#4 #5 {\begingroup\inENV
-  \medbreak
-  \def#1{\endgraf\endgroup\medbreak}%
-  \def#2##1 ##2 {\begingroup\obeylines\activeparens\spacesplit{#3{##1}{##2}}}%
-  \parindent=0in
-  \advance\leftskip by \defbodyindent
-  \exdentamount=\defbodyindent
-  \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 {\begingroup\inENV
-  \medbreak
-  \def#1{\endgraf\endgroup\medbreak}%
-  \def#2##1 ##2 ##3 {%
-    \def#4{##1}%
-    \begingroup\obeylines\activeparens\spacesplit{#3{##2}{##3}}}%
-  \parindent=0in
-  \advance\leftskip by \defbodyindent
-  \exdentamount=\defbodyindent
-  \begingroup\obeylines\activeparens\spacesplit{#3{#5}{#6}}}
-
-\def\defopparsebody #1#2#3#4#5 {\begingroup\inENV %
-\medbreak %
-% Define the end token that this defining construct specifies
-% so that it will exit this group.
-\def#1{\endgraf\endgroup\medbreak}%
-\def#2##1 ##2 {\def#4{##1}%
-\begingroup\obeylines\activeparens\spacesplit{#3{##2}}}%
-\parindent=0in
-\advance\leftskip by \defbodyindent
-\exdentamount=\defbodyindent
-\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{\begingroup\inENV% Environment for definitionbody
-\medbreak %
-% Define the end token that this defining construct specifies
-% so that it will exit this group.
-\def#1{\endgraf\endgroup\medbreak}%
-\def#2{\begingroup\obeylines\spacesplit#3}%
-\parindent=0in
-\advance\leftskip by \defbodyindent
-\exdentamount=\defbodyindent
-\begingroup %
-\catcode 61=\active %
-\obeylines\spacesplit#3}
-
-% This is used for \def{tp,vr}parsebody.  It could probably be used for
-% some of the others, too, with some judicious conditionals.
-%
-\def\parsebodycommon#1#2#3{%
-  \begingroup\inENV %
-  \medbreak %
-  % Define the end token that this defining construct specifies
-  % so that it will exit this group.
-  \def#1{\endgraf\endgroup\medbreak}%
-  \def#2##1 {\begingroup\obeylines\spacesplit{#3{##1}}}%
-  \parindent=0in
-  \advance\leftskip by \defbodyindent
-  \exdentamount=\defbodyindent
-  \begingroup\obeylines
-}
-
-\def\defvrparsebody#1#2#3#4 {%
-  \parsebodycommon{#1}{#2}{#3}%
-  \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}%
-  \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}%
-}%
-
-\def\defopvarparsebody #1#2#3#4#5 {\begingroup\inENV %
-\medbreak %
-% Define the end token that this defining construct specifies
-% so that it will exit this group.
-\def#1{\endgraf\endgroup\medbreak}%
-\def#2##1 ##2 {\def#4{##1}%
-\begingroup\obeylines\spacesplit{#3{##2}}}%
-\parindent=0in
-\advance\leftskip by \defbodyindent
-\exdentamount=\defbodyindent
-\begingroup\obeylines\spacesplit{#3{#5}}}
-
-% Split up #2 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\spacesplitfoo{#1}#2 \relax\spacesplitfoo}%
-\long\gdef\spacesplitfoo#1#2 #3#4\spacesplitfoo{%
-\ifx\relax #3%
-#1{#2}{}\else #1{#2}{#3#4}\fi}}
-
-% So much for the things common to all kinds of definitions.
-
-% Define @defun.
-
-% First, define the processing that is wanted for arguments of \defun
-% Use this to expand the args and terminate the paragraph they make up
-
-\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%
-\interlinepenalty=10000
-\advance\rightskip by 0pt plus 1fil
-\endgraf\nobreak\vskip -\parskip\nobreak
-}
-
-\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
-\interlinepenalty=10000
-\advance\rightskip by 0pt plus 1fil
-\endgraf\nobreak\vskip -\parskip\nobreak
-}
-
-% 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 61=\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 61=\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$$$#2}{\putwordDeftypefun}%
-\deftypefunargs {#3}\endgroup %
-\catcode 61=\other % Turn off change made in \defparsebody
-}
-
-% @deftypefn {Library Function} int foobar (int @var{foo}, float @var{bar})
-
-\def\deftypefn{\defmethparsebody\Edeftypefn\deftypefnx\deftypefnheader}
-
-% \defheaderxcond#1\relax$$$
-% 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 61=\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 61=\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 61=\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\ #1}% Make entry in function index
-\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{#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\ #1}% Make entry in var index
-\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\ #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%
-\interlinepenalty=10000
-\endgraf\nobreak\vskip -\parskip\nobreak}
-
-% @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$$$#2}{\putwordDeftypevar}%
-\interlinepenalty=10000
-\endgraf\nobreak\vskip -\parskip\nobreak
-\endgroup}
-\def\dovarind#1 #2\relax{\doind{vr}{\code{#1}}}
-
-% @deftypevr {Global Flag} int enable
-
-\def\deftypevr{\defvrparsebody\Edeftypevr\deftypevrx\deftypevrheader}
-
-\def\deftypevrheader #1#2#3{\dovarind#3 \relax%
-\begingroup\defname {\defheaderxcond#2\relax$$$#3}{#1}
-\interlinepenalty=10000
-\endgraf\nobreak\vskip -\parskip\nobreak
-\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`\\=12 \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`\\=12 \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=12\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`\~=12
-  \catcode`\^=12
-  \catcode`\_=12
-  \catcode`\|=12
-  \catcode`\<=12
-  \catcode`\>=12
-  \catcode`\+=12
-  \catcode`\{=12
-  \catcode`\}=12
-  \catcode`\@=12
-  \catcode`\^^M=12
-  \usembodybackslash}
-
-\def\macroargctxt{%
-  \catcode`\~=12
-  \catcode`\^=12
-  \catcode`\_=12
-  \catcode`\|=12
-  \catcode`\<=12
-  \catcode`\>=12
-  \catcode`\+=12
-  \catcode`\@=12
-  \catcode`\\=12}
-
-% \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{The name \the\macname\space is reserved}\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\unmacroxxx}
-\def\unmacroxxx#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
-      \edef\tempa{\expandafter\noexpand\csname#1\endcsname}%
-      \def\do##1{%
-        \def\tempb{##1}%
-        \ifx\tempa\tempb
-          % remove this
-        \else
-          \toks0 = \expandafter{\newmacrolist\do}%
-          \edef\newmacrolist{\the\toks0\expandafter\noexpand\tempa}%
-        \fi}%
-      \def\newmacrolist{}%
-      % Execute macro list to define \newmacrolist
-      \macrolist
-      \global\let\macrolist\newmacrolist
-    \endgroup
-  \else
-    \errmessage{Macro #1 not defined}%
-  \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,]}
-\def\nodexxx[#1,#2]{\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, namely
-% NAME-title, NAME-pg, and NAME-SNT.  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.
-%
-\def\setref#1#2{{%
-  \indexdummies
-  \pdfmkdest{#1}%
-  \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}%
-    \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.
-    {\normalturnoffactive
-     % 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
-    }%
-    % [mynode],
-    [\printednodename],\space
-    % page 3
-    \turnoffactive \putwordpage\tie\refx{#1-pg}{}%
-  \fi
-  \endlink
-\endgroup}
-
-% \dosetq is the interface for calls from other macros
-
-% Use \normalturnoffactive so that punctuation chars such as underscore
-% and backslash work in node names.  (\turnoffactive doesn't do \.)
-\def\dosetq#1#2{%
-  {\let\folio=0%
-   \normalturnoffactive
-   \edef\next{\write\auxfile{\internalsetq{#1}{#2}}}%
-   \iflinks
-     \next
-   \fi
-  }%
-}
-
-% \internalsetq {foo}{page} expands into
-% CHARACTERS 'xrdef {foo}{...expansion of \Ypage...}
-% When the aux file is read, ' is the escape character
-
-\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\xreftie\the\chapno %
-\else \ifnum \subsecno=0 \putwordSection\xreftie\the\chapno.\the\secno %
-\else \ifnum \subsubsecno=0 %
-\putwordSection\xreftie\the\chapno.\the\secno.\the\subsecno %
-\else %
-\putwordSection\xreftie\the\chapno.\the\secno.\the\subsecno.\the\subsubsecno %
-\fi \fi \fi }
-
-\def\Yappendixletterandtype{%
-\ifnum\secno=0 \putwordAppendix\xreftie'char\the\appendixno{}%
-\else \ifnum \subsecno=0 \putwordSection\xreftie'char\the\appendixno.\the\secno %
-\else \ifnum \subsubsecno=0 %
-\putwordSection\xreftie'char\the\appendixno.\the\secno.\the\subsecno %
-\else %
-\putwordSection\xreftie'char\the\appendixno.\the\secno.\the\subsecno.\the\subsubsecno %
-\fi \fi \fi }
-
-\gdef\xreftie{'tie}
-
-% 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 % Non-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{%
-  \expandafter\ifx\csname X#1\endcsname\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.
-    \csname X#1\endcsname
-  \fi
-  #2% Output the suffix in any case.
-}
-
-% This is the macro invoked by entries in the aux file.
-%
-\def\xrdef#1{\begingroup
-  % Reenable \ as an escape while reading the second argument.
-  \catcode`\\ = 0
-  \afterassignment\endgroup
-  \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
-  \catcode`\@=\other
-  \catcode`\^=\other
-  % It was suggested to define this as 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
-  \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
-    }%
-  }%
-  % The aux file uses ' as the escape (for now).
-  % 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 $-fun}{page ...} made by @defun ^^
-  % Reference to such entries still does not work the way one would wish,
-  % but at least they do not bomb out when the aux file is read in.
-  \catcode`\{=1
-  \catcode`\}=2
-  \catcode`\%=\other
-  \catcode`\'=0
-  \catcode`\\=\other
-  %
-  \openin 1 \jobname.aux
-  \ifeof 1 \else
-    \closein 1
-    \input \jobname.aux
-    \global\havexrefstrue
-    \global\warnedobstrue
-  \fi
-  % Open the new aux file.  TeX will close it automatically at exit.
-  \openout\auxfile=\jobname.aux
-\endgroup}
-
-
-% Footnotes.
-
-\newcount \footnoteno
-
-% The trailing space in the following definition for supereject is
-% vital for proper filling; pages come out unaligned when you do a
-% pagealignmacro call if that space before the closing brace is
-% removed. (Generally, numeric constants should always be followed by a
-% space to prevent strange expansion errors.)
-\def\supereject{\par\penalty -20000\footnoteno =0 }
-
-% @footnotestyle is meaningful for info output only.
-\let\footnotestyle=\comment
-
-\let\ptexfootnote=\footnote
-
-{\catcode `\@=11
-%
-% Auto-number footnotes.  Otherwise like plain.
-\gdef\footnote{%
-  \global\advance\footnoteno by \@ne
-  \edef\thisfootno{$^{\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}\/\fi
-  %
-  % Remove inadvertent blank space before typesetting the footnote number.
-  \unskip
-  \thisfootno\@sf
-  \footnotezzz
-}%
-
-% 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.
-%
-\long\gdef\footnotezzz{\insert\footins\bgroup
-  % We want to typeset this text as a normal paragraph, even if the
-  % footnote reference occurs in (for example) a display environment.
-  % So reset some parameters.
-  \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
-  %
-  % Hang the footnote text off the number.
-  \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
-}
-\def\fo@t{\ifcat\bgroup\noexpand\next \let\next\f@@t
-  \else\let\next\f@t\fi \next}
-\def\f@@t{\bgroup\aftergroup\@foot\let\next}
-\def\f@t#1{#1\@foot}
-\def\@foot{\strut\par\egroup}
-
-}%end \catcode `\@=11
-
-% 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
-  }%
-}
-
-% @| 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 post-v2.7 epsf.tex (available in
-  % doc/epsf.tex until it shows up 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 just the usual extra ignored arg for parsing this stuff.
-\def\imagexxx#1,#2,#3,#4\finish{%
-  \ifpdf
-    \centerline{\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
-    \begingroup
-      \catcode`\^^M = 5 % in case we're inside an example
-      % If the image is by itself, center it.
-      \ifvmode
-        \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
-        \centerline{\epsfbox{#1.eps}}%
-        \bigbreak
-      \else
-        % In the middle of a paragraph, no extra space.
-        \epsfbox{#1.eps}%
-      \fi
-    \endgroup
-  \fi
-}
-
-
-\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.  Then whoever calls us can
-% set \parskip and call \setleading for \baselineskip.
-%
-\def\internalpagesizes#1#2#3#4#5#6{%
-  \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
-  %
-  \parindent = \defaultparindent
-  \setemergencystretch
-}
-
-% @letterpaper (the default).
-\def\letterpaper{{\globaldefs = 1
-  \parskip = 3pt plus 2pt minus 1pt
-  \setleading{13.2pt}%
-  %
-  % If page is nothing but text, make it come out even.
-  \internalpagesizes{46\baselineskip}{6in}{\voffset}{.25in}{\bindingoffset}{36pt}%
-}}
-
-% Use @smallbook to reset parameters for 7x9.5 (or so) format.
-\def\smallbook{{\globaldefs = 1
-  \parskip = 2pt plus 1pt
-  \setleading{12pt}%
-  %
-  \internalpagesizes{7.5in}{5.in}{\voffset}{.25in}{\bindingoffset}{16pt}%
-  %
-  \lispnarrowing = 0.3in
-  \tolerance = 700
-  \hfuzz = 1pt
-  \contentsrightmargin = 0pt
-  \deftypemargin = 0pt
-  \defbodyindent = .5cm
-  %
-  \let\smalldisplay = \smalldisplayx
-  \let\smallexample = \smalllispx
-  \let\smallformat = \smallformatx
-  \let\smalllisp = \smalllispx
-}}
-
-% Use @afourpaper to print on European A4 paper.
-\def\afourpaper{{\globaldefs = 1
-  \setleading{12pt}%
-  \parskip = 3pt plus 2pt minus 1pt
-  %
-  \internalpagesizes{53\baselineskip}{160mm}{\voffset}{4mm}{\bindingoffset}{44pt}%
-  %
-  \tolerance = 700
-  \hfuzz = 1pt
-}}
-
-% A specific text layout, 24x15cm overall, intended for A4 paper.  Top margin
-% 29mm, hence bottom margin 28mm, nominal side margin 3cm.
-\def\afourlatex{{\globaldefs = 1
-  \setleading{13.6pt}%
-  %
-  \afourpaper
-  \internalpagesizes{237mm}{150mm}{3.6mm}{3.6mm}{3mm}{7mm}%
-  %
-  \globaldefs = 0
-}}
-
-% Use @afourwide to print on European A4 paper in wide format.
-\def\afourwide{%
-  \afourpaper
-  \internalpagesizes{6.5in}{9.5in}{\hoffset}{\normaloffset}{\bindingoffset}{7mm}%
-  %
-  \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{13.2pt}%
-  %
-  \internalpagesizes{#1}{\hsize}{\voffset}{\normaloffset}{\bindingoffset}{44pt}%
-}}
-
-% 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{$}
-
-% 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.06em \vbox{\hrule width.3em height.1ex}}
-
-\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${\ifusingit{{\sl\$}}\normaldollar}
-%\catcode 27=\active
-%\def^^[{$\diamondsuit$}
-
-% Set up an active definition for =, but don't enable it most of the time.
-{\catcode`\==\active
-\global\def={{\tt \char 61}}}
-
-\catcode`+=\active
-\catcode`\_=\active
-
-% If a .fmt file is being used, characters that might appear in a file
-% name cannot be active until we have parsed the command line.
-% So turn them off again, and have \everyjob (or @setfilename) turn them on.
-% \otherifyactive is called near the end of this file.
-\def\otherifyactive{\catcode`+=\other \catcode`\_=\other}
-
-\catcode`\@=0
-
-% \rawbackslashxx output one backslash character in current font
-\global\chardef\rawbackslashxx=`\\
-%{\catcode`\\=\other
-%@gdef@rawbackslashxx{\}}
-
-% \rawbackslash redefines \ as input to do \rawbackslashxx.
-{\catcode`\\=\active
-@gdef@rawbackslash{@let\=@rawbackslashxx }}
-
-% \normalbackslash outputs one backslash in fixed width font.
-\def\normalbackslash{{\tt\rawbackslashxx}}
-
-% \catcode 17=0   % Define control-q
-\catcode`\\=\active
-
-% Used sometimes to turn off (effectively) the active characters
-% even after parsing them.
-@def@turnoffactive{@let"=@normaldoublequote
-@let\=@realbackslash
-@let~=@normaltilde
-@let^=@normalcaret
-@let_=@normalunderscore
-@let|=@normalverticalbar
-@let<=@normalless
-@let>=@normalgreater
-@let+=@normalplus
-@let$=@normaldollar}
-
-@def@normalturnoffactive{@let"=@normaldoublequote
-@let\=@normalbackslash
-@let~=@normaltilde
-@let^=@normalcaret
-@let_=@normalunderscore
-@let|=@normalverticalbar
-@let<=@normalless
-@let>=@normalgreater
-@let+=@normalplus
-@let$=@normaldollar}
-
-% 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: