Return to texinfo.tex CVS log

Up to [GnuMach] / Gnu-Mach / build-aux

Annotation of Gnu-Mach/build-aux/texinfo.tex, revision 1.1

1.1     ! root        1: % texinfo.tex -- TeX macros to handle Texinfo files.
        !             2: % 
        !             3: % Load plain if necessary, i.e., if running under initex.
        !             4: \expandafter\ifx\csname fmtname\endcsname\relax\input plain\fi
        !             5: %
        !             6: \def\texinfoversion{2013-02-01.11}
        !             7: %
        !             8: % Copyright 1985, 1986, 1988, 1990, 1991, 1992, 1993, 1994, 1995,
        !             9: % 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006,
        !            10: % 2007, 2008, 2009, 2010, 2011, 2012, 2013 Free Software Foundation, Inc.
        !            11: %
        !            12: % This texinfo.tex file is free software: you can redistribute it and/or
        !            13: % modify it under the terms of the GNU General Public License as
        !            14: % published by the Free Software Foundation, either version 3 of the
        !            15: % License, or (at your option) any later version.
        !            16: %
        !            17: % This texinfo.tex file is distributed in the hope that it will be
        !            18: % useful, but WITHOUT ANY WARRANTY; without even the implied warranty
        !            19: % of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
        !            20: % General Public License for more details.
        !            21: %
        !            22: % You should have received a copy of the GNU General Public License
        !            23: % along with this program.  If not, see <http://www.gnu.org/licenses/>.
        !            24: %
        !            25: % As a special exception, when this file is read by TeX when processing
        !            26: % a Texinfo source document, you may use the result without
        !            27: % restriction. This Exception is an additional permission under section 7
        !            28: % of the GNU General Public License, version 3 ("GPLv3").
        !            29: %
        !            30: % Please try the latest version of texinfo.tex before submitting bug
        !            31: % reports; you can get the latest version from:
        !            32: %   http://ftp.gnu.org/gnu/texinfo/ (the Texinfo release area), or
        !            33: %   http://ftpmirror.gnu.org/texinfo/ (same, via a mirror), or
        !            34: %   http://www.gnu.org/software/texinfo/ (the Texinfo home page)
        !            35: % The texinfo.tex in any given distribution could well be out
        !            36: % of date, so if that's what you're using, please check.
        !            37: %
        !            38: % Send bug reports to [email protected].  Please include including a
        !            39: % complete document in each bug report with which we can reproduce the
        !            40: % problem.  Patches are, of course, greatly appreciated.
        !            41: %
        !            42: % To process a Texinfo manual with TeX, it's most reliable to use the
        !            43: % texi2dvi shell script that comes with the distribution.  For a simple
        !            44: % manual foo.texi, however, you can get away with this:
        !            45: %   tex foo.texi
        !            46: %   texindex foo.??
        !            47: %   tex foo.texi
        !            48: %   tex foo.texi
        !            49: %   dvips foo.dvi -o  # or whatever; this makes foo.ps.
        !            50: % The extra TeX runs get the cross-reference information correct.
        !            51: % Sometimes one run after texindex suffices, and sometimes you need more
        !            52: % than two; texi2dvi does it as many times as necessary.
        !            53: %
        !            54: % It is possible to adapt texinfo.tex for other languages, to some
        !            55: % extent.  You can get the existing language-specific files from the
        !            56: % full Texinfo distribution.
        !            57: %
        !            58: % The GNU Texinfo home page is http://www.gnu.org/software/texinfo.
        !            59: 
        !            60: 
        !            61: \message{Loading texinfo [version \texinfoversion]:}
        !            62: 
        !            63: % If in a .fmt file, print the version number
        !            64: % and turn on active characters that we couldn't do earlier because
        !            65: % they might have appeared in the input file name.
        !            66: \everyjob{\message{[Texinfo version \texinfoversion]}%
        !            67:   \catcode`+=\active \catcode`\_=\active}
        !            68: 
        !            69: \chardef\other=12
        !            70: 
        !            71: % We never want plain's \outer definition of \+ in Texinfo.
        !            72: % For @tex, we can use \tabalign.
        !            73: \let\+ = \relax
        !            74: 
        !            75: % Save some plain tex macros whose names we will redefine.
        !            76: \let\ptexb=\b
        !            77: \let\ptexbullet=\bullet
        !            78: \let\ptexc=\c
        !            79: \let\ptexcomma=\,
        !            80: \let\ptexdot=\.
        !            81: \let\ptexdots=\dots
        !            82: \let\ptexend=\end
        !            83: \let\ptexequiv=\equiv
        !            84: \let\ptexexclam=\!
        !            85: \let\ptexfootnote=\footnote
        !            86: \let\ptexgtr=>
        !            87: \let\ptexhat=^
        !            88: \let\ptexi=\i
        !            89: \let\ptexindent=\indent
        !            90: \let\ptexinsert=\insert
        !            91: \let\ptexlbrace=\{
        !            92: \let\ptexless=<
        !            93: \let\ptexnewwrite\newwrite
        !            94: \let\ptexnoindent=\noindent
        !            95: \let\ptexplus=+
        !            96: \let\ptexraggedright=\raggedright
        !            97: \let\ptexrbrace=\}
        !            98: \let\ptexslash=\/
        !            99: \let\ptexstar=\*
        !           100: \let\ptext=\t
        !           101: \let\ptextop=\top
        !           102: {\catcode`\'=\active \global\let\ptexquoteright'}% active in plain's math mode
        !           103: 
        !           104: % If this character appears in an error message or help string, it
        !           105: % starts a new line in the output.
        !           106: \newlinechar = `^^J
        !           107: 
        !           108: % Use TeX 3.0's \inputlineno to get the line number, for better error
        !           109: % messages, but if we're using an old version of TeX, don't do anything.
        !           110: %
        !           111: \ifx\inputlineno\thisisundefined
        !           112:   \let\linenumber = \empty % Pre-3.0.
        !           113: \else
        !           114:   \def\linenumber{l.\the\inputlineno:\space}
        !           115: \fi
        !           116: 
        !           117: % Set up fixed words for English if not already set.
        !           118: \ifx\putwordAppendix\undefined  \gdef\putwordAppendix{Appendix}\fi
        !           119: \ifx\putwordChapter\undefined   \gdef\putwordChapter{Chapter}\fi
        !           120: \ifx\putworderror\undefined     \gdef\putworderror{error}\fi
        !           121: \ifx\putwordfile\undefined      \gdef\putwordfile{file}\fi
        !           122: \ifx\putwordin\undefined        \gdef\putwordin{in}\fi
        !           123: \ifx\putwordIndexIsEmpty\undefined       \gdef\putwordIndexIsEmpty{(Index is empty)}\fi
        !           124: \ifx\putwordIndexNonexistent\undefined   \gdef\putwordIndexNonexistent{(Index is nonexistent)}\fi
        !           125: \ifx\putwordInfo\undefined      \gdef\putwordInfo{Info}\fi
        !           126: \ifx\putwordInstanceVariableof\undefined \gdef\putwordInstanceVariableof{Instance Variable of}\fi
        !           127: \ifx\putwordMethodon\undefined  \gdef\putwordMethodon{Method on}\fi
        !           128: \ifx\putwordNoTitle\undefined   \gdef\putwordNoTitle{No Title}\fi
        !           129: \ifx\putwordof\undefined        \gdef\putwordof{of}\fi
        !           130: \ifx\putwordon\undefined        \gdef\putwordon{on}\fi
        !           131: \ifx\putwordpage\undefined      \gdef\putwordpage{page}\fi
        !           132: \ifx\putwordsection\undefined   \gdef\putwordsection{section}\fi
        !           133: \ifx\putwordSection\undefined   \gdef\putwordSection{Section}\fi
        !           134: \ifx\putwordsee\undefined       \gdef\putwordsee{see}\fi
        !           135: \ifx\putwordSee\undefined       \gdef\putwordSee{See}\fi
        !           136: \ifx\putwordShortTOC\undefined  \gdef\putwordShortTOC{Short Contents}\fi
        !           137: \ifx\putwordTOC\undefined       \gdef\putwordTOC{Table of Contents}\fi
        !           138: %
        !           139: \ifx\putwordMJan\undefined \gdef\putwordMJan{January}\fi
        !           140: \ifx\putwordMFeb\undefined \gdef\putwordMFeb{February}\fi
        !           141: \ifx\putwordMMar\undefined \gdef\putwordMMar{March}\fi
        !           142: \ifx\putwordMApr\undefined \gdef\putwordMApr{April}\fi
        !           143: \ifx\putwordMMay\undefined \gdef\putwordMMay{May}\fi
        !           144: \ifx\putwordMJun\undefined \gdef\putwordMJun{June}\fi
        !           145: \ifx\putwordMJul\undefined \gdef\putwordMJul{July}\fi
        !           146: \ifx\putwordMAug\undefined \gdef\putwordMAug{August}\fi
        !           147: \ifx\putwordMSep\undefined \gdef\putwordMSep{September}\fi
        !           148: \ifx\putwordMOct\undefined \gdef\putwordMOct{October}\fi
        !           149: \ifx\putwordMNov\undefined \gdef\putwordMNov{November}\fi
        !           150: \ifx\putwordMDec\undefined \gdef\putwordMDec{December}\fi
        !           151: %
        !           152: \ifx\putwordDefmac\undefined    \gdef\putwordDefmac{Macro}\fi
        !           153: \ifx\putwordDefspec\undefined   \gdef\putwordDefspec{Special Form}\fi
        !           154: \ifx\putwordDefvar\undefined    \gdef\putwordDefvar{Variable}\fi
        !           155: \ifx\putwordDefopt\undefined    \gdef\putwordDefopt{User Option}\fi
        !           156: \ifx\putwordDeffunc\undefined   \gdef\putwordDeffunc{Function}\fi
        !           157: 
        !           158: % Since the category of space is not known, we have to be careful.
        !           159: \chardef\spacecat = 10
        !           160: \def\spaceisspace{\catcode`\ =\spacecat}
        !           161: 
        !           162: % sometimes characters are active, so we need control sequences.
        !           163: \chardef\ampChar   = `\&
        !           164: \chardef\colonChar = `\:
        !           165: \chardef\commaChar = `\,
        !           166: \chardef\dashChar  = `\-
        !           167: \chardef\dotChar   = `\.
        !           168: \chardef\exclamChar= `\!
        !           169: \chardef\hashChar  = `\#
        !           170: \chardef\lquoteChar= `\`
        !           171: \chardef\questChar = `\?
        !           172: \chardef\rquoteChar= `\'
        !           173: \chardef\semiChar  = `\;
        !           174: \chardef\slashChar = `\/
        !           175: \chardef\underChar = `\_
        !           176: 
        !           177: % Ignore a token.
        !           178: %
        !           179: \def\gobble#1{}
        !           180: 
        !           181: % The following is used inside several \edef's.
        !           182: \def\makecsname#1{\expandafter\noexpand\csname#1\endcsname}
        !           183: 
        !           184: % Hyphenation fixes.
        !           185: \hyphenation{
        !           186:   Flor-i-da Ghost-script Ghost-view Mac-OS Post-Script
        !           187:   ap-pen-dix bit-map bit-maps
        !           188:   data-base data-bases eshell fall-ing half-way long-est man-u-script
        !           189:   man-u-scripts mini-buf-fer mini-buf-fers over-view par-a-digm
        !           190:   par-a-digms rath-er rec-tan-gu-lar ro-bot-ics se-vere-ly set-up spa-ces
        !           191:   spell-ing spell-ings
        !           192:   stand-alone strong-est time-stamp time-stamps which-ever white-space
        !           193:   wide-spread wrap-around
        !           194: }
        !           195: 
        !           196: % Margin to add to right of even pages, to left of odd pages.
        !           197: \newdimen\bindingoffset
        !           198: \newdimen\normaloffset
        !           199: \newdimen\pagewidth \newdimen\pageheight
        !           200: 
        !           201: % For a final copy, take out the rectangles
        !           202: % that mark overfull boxes (in case you have decided
        !           203: % that the text looks ok even though it passes the margin).
        !           204: %
        !           205: \def\finalout{\overfullrule=0pt }
        !           206: 
        !           207: % Sometimes it is convenient to have everything in the transcript file
        !           208: % and nothing on the terminal.  We don't just call \tracingall here,
        !           209: % since that produces some useless output on the terminal.  We also make
        !           210: % some effort to order the tracing commands to reduce output in the log
        !           211: % file; cf. trace.sty in LaTeX.
        !           212: %
        !           213: \def\gloggingall{\begingroup \globaldefs = 1 \loggingall \endgroup}%
        !           214: \def\loggingall{%
        !           215:   \tracingstats2
        !           216:   \tracingpages1
        !           217:   \tracinglostchars2  % 2 gives us more in etex
        !           218:   \tracingparagraphs1
        !           219:   \tracingoutput1
        !           220:   \tracingmacros2
        !           221:   \tracingrestores1
        !           222:   \showboxbreadth\maxdimen \showboxdepth\maxdimen
        !           223:   \ifx\eTeXversion\thisisundefined\else % etex gives us more logging
        !           224:     \tracingscantokens1
        !           225:     \tracingifs1
        !           226:     \tracinggroups1
        !           227:     \tracingnesting2
        !           228:     \tracingassigns1
        !           229:   \fi
        !           230:   \tracingcommands3  % 3 gives us more in etex
        !           231:   \errorcontextlines16
        !           232: }%
        !           233: 
        !           234: % @errormsg{MSG}.  Do the index-like expansions on MSG, but if things
        !           235: % aren't perfect, it's not the end of the world, being an error message,
        !           236: % after all.
        !           237: % 
        !           238: \def\errormsg{\begingroup \indexnofonts \doerrormsg}
        !           239: \def\doerrormsg#1{\errmessage{#1}}
        !           240: 
        !           241: % add check for \lastpenalty to plain's definitions.  If the last thing
        !           242: % we did was a \nobreak, we don't want to insert more space.
        !           243: %
        !           244: \def\smallbreak{\ifnum\lastpenalty<10000\par\ifdim\lastskip<\smallskipamount
        !           245:   \removelastskip\penalty-50\smallskip\fi\fi}
        !           246: \def\medbreak{\ifnum\lastpenalty<10000\par\ifdim\lastskip<\medskipamount
        !           247:   \removelastskip\penalty-100\medskip\fi\fi}
        !           248: \def\bigbreak{\ifnum\lastpenalty<10000\par\ifdim\lastskip<\bigskipamount
        !           249:   \removelastskip\penalty-200\bigskip\fi\fi}
        !           250: 
        !           251: % Do @cropmarks to get crop marks.
        !           252: %
        !           253: \newif\ifcropmarks
        !           254: \let\cropmarks = \cropmarkstrue
        !           255: %
        !           256: % Dimensions to add cropmarks at corners.
        !           257: % Added by P. A. MacKay, 12 Nov. 1986
        !           258: %
        !           259: \newdimen\outerhsize \newdimen\outervsize % set by the paper size routines
        !           260: \newdimen\cornerlong  \cornerlong=1pc
        !           261: \newdimen\cornerthick \cornerthick=.3pt
        !           262: \newdimen\topandbottommargin \topandbottommargin=.75in
        !           263: 
        !           264: % Output a mark which sets \thischapter, \thissection and \thiscolor.
        !           265: % We dump everything together because we only have one kind of mark.
        !           266: % This works because we only use \botmark / \topmark, not \firstmark.
        !           267: %
        !           268: % A mark contains a subexpression of the \ifcase ... \fi construct.
        !           269: % \get*marks macros below extract the needed part using \ifcase.
        !           270: %
        !           271: % Another complication is to let the user choose whether \thischapter
        !           272: % (\thissection) refers to the chapter (section) in effect at the top
        !           273: % of a page, or that at the bottom of a page.  The solution is
        !           274: % described on page 260 of The TeXbook.  It involves outputting two
        !           275: % marks for the sectioning macros, one before the section break, and
        !           276: % one after.  I won't pretend I can describe this better than DEK...
        !           277: \def\domark{%
        !           278:   \toks0=\expandafter{\lastchapterdefs}%
        !           279:   \toks2=\expandafter{\lastsectiondefs}%
        !           280:   \toks4=\expandafter{\prevchapterdefs}%
        !           281:   \toks6=\expandafter{\prevsectiondefs}%
        !           282:   \toks8=\expandafter{\lastcolordefs}%
        !           283:   \mark{%
        !           284:                    \the\toks0 \the\toks2
        !           285:       \noexpand\or \the\toks4 \the\toks6
        !           286:     \noexpand\else \the\toks8
        !           287:   }%
        !           288: }
        !           289: % \topmark doesn't work for the very first chapter (after the title
        !           290: % page or the contents), so we use \firstmark there -- this gets us
        !           291: % the mark with the chapter defs, unless the user sneaks in, e.g.,
        !           292: % @setcolor (or @url, or @link, etc.) between @contents and the very
        !           293: % first @chapter.
        !           294: \def\gettopheadingmarks{%
        !           295:   \ifcase0\topmark\fi
        !           296:   \ifx\thischapter\empty \ifcase0\firstmark\fi \fi
        !           297: }
        !           298: \def\getbottomheadingmarks{\ifcase1\botmark\fi}
        !           299: \def\getcolormarks{\ifcase2\topmark\fi}
        !           300: 
        !           301: % Avoid "undefined control sequence" errors.
        !           302: \def\lastchapterdefs{}
        !           303: \def\lastsectiondefs{}
        !           304: \def\prevchapterdefs{}
        !           305: \def\prevsectiondefs{}
        !           306: \def\lastcolordefs{}
        !           307: 
        !           308: % Main output routine.
        !           309: \chardef\PAGE = 255
        !           310: \output = {\onepageout{\pagecontents\PAGE}}
        !           311: 
        !           312: \newbox\headlinebox
        !           313: \newbox\footlinebox
        !           314: 
        !           315: % \onepageout takes a vbox as an argument.  Note that \pagecontents
        !           316: % does insertions, but you have to call it yourself.
        !           317: \def\onepageout#1{%
        !           318:   \ifcropmarks \hoffset=0pt \else \hoffset=\normaloffset \fi
        !           319:   %
        !           320:   \ifodd\pageno  \advance\hoffset by \bindingoffset
        !           321:   \else \advance\hoffset by -\bindingoffset\fi
        !           322:   %
        !           323:   % Do this outside of the \shipout so @code etc. will be expanded in
        !           324:   % the headline as they should be, not taken literally (outputting ''code).
        !           325:   \ifodd\pageno \getoddheadingmarks \else \getevenheadingmarks \fi
        !           326:   \setbox\headlinebox = \vbox{\let\hsize=\pagewidth \makeheadline}%
        !           327:   \ifodd\pageno \getoddfootingmarks \else \getevenfootingmarks \fi
        !           328:   \setbox\footlinebox = \vbox{\let\hsize=\pagewidth \makefootline}%
        !           329:   %
        !           330:   {%
        !           331:     % Have to do this stuff outside the \shipout because we want it to
        !           332:     % take effect in \write's, yet the group defined by the \vbox ends
        !           333:     % before the \shipout runs.
        !           334:     %
        !           335:     \indexdummies         % don't expand commands in the output.
        !           336:     \normalturnoffactive  % \ in index entries must not stay \, e.g., if
        !           337:                % the page break happens to be in the middle of an example.
        !           338:                % We don't want .vr (or whatever) entries like this:
        !           339:                % \entry{{\tt \indexbackslash }acronym}{32}{\code {\acronym}}
        !           340:                % "\acronym" won't work when it's read back in;
        !           341:                % it needs to be
        !           342:                % {\code {{\tt \backslashcurfont }acronym}
        !           343:     \shipout\vbox{%
        !           344:       % Do this early so pdf references go to the beginning of the page.
        !           345:       \ifpdfmakepagedest \pdfdest name{\the\pageno} xyz\fi
        !           346:       %
        !           347:       \ifcropmarks \vbox to \outervsize\bgroup
        !           348:         \hsize = \outerhsize
        !           349:         \vskip-\topandbottommargin
        !           350:         \vtop to0pt{%
        !           351:           \line{\ewtop\hfil\ewtop}%
        !           352:           \nointerlineskip
        !           353:           \line{%
        !           354:             \vbox{\moveleft\cornerthick\nstop}%
        !           355:             \hfill
        !           356:             \vbox{\moveright\cornerthick\nstop}%
        !           357:           }%
        !           358:           \vss}%
        !           359:         \vskip\topandbottommargin
        !           360:         \line\bgroup
        !           361:           \hfil % center the page within the outer (page) hsize.
        !           362:           \ifodd\pageno\hskip\bindingoffset\fi
        !           363:           \vbox\bgroup
        !           364:       \fi
        !           365:       %
        !           366:       \unvbox\headlinebox
        !           367:       \pagebody{#1}%
        !           368:       \ifdim\ht\footlinebox > 0pt
        !           369:         % Only leave this space if the footline is nonempty.
        !           370:         % (We lessened \vsize for it in \oddfootingyyy.)
        !           371:         % The \baselineskip=24pt in plain's \makefootline has no effect.
        !           372:         \vskip 24pt
        !           373:         \unvbox\footlinebox
        !           374:       \fi
        !           375:       %
        !           376:       \ifcropmarks
        !           377:           \egroup % end of \vbox\bgroup
        !           378:         \hfil\egroup % end of (centering) \line\bgroup
        !           379:         \vskip\topandbottommargin plus1fill minus1fill
        !           380:         \boxmaxdepth = \cornerthick
        !           381:         \vbox to0pt{\vss
        !           382:           \line{%
        !           383:             \vbox{\moveleft\cornerthick\nsbot}%
        !           384:             \hfill
        !           385:             \vbox{\moveright\cornerthick\nsbot}%
        !           386:           }%
        !           387:           \nointerlineskip
        !           388:           \line{\ewbot\hfil\ewbot}%
        !           389:         }%
        !           390:       \egroup % \vbox from first cropmarks clause
        !           391:       \fi
        !           392:     }% end of \shipout\vbox
        !           393:   }% end of group with \indexdummies
        !           394:   \advancepageno
        !           395:   \ifnum\outputpenalty>-20000 \else\dosupereject\fi
        !           396: }
        !           397: 
        !           398: \newinsert\margin \dimen\margin=\maxdimen
        !           399: 
        !           400: \def\pagebody#1{\vbox to\pageheight{\boxmaxdepth=\maxdepth #1}}
        !           401: {\catcode`\@ =11
        !           402: \gdef\pagecontents#1{\ifvoid\topins\else\unvbox\topins\fi
        !           403: % marginal hacks, [email protected] (Juha Takala)
        !           404: \ifvoid\margin\else % marginal info is present
        !           405:   \rlap{\kern\hsize\vbox to\z@{\kern1pt\box\margin \vss}}\fi
        !           406: \dimen@=\dp#1\relax \unvbox#1\relax
        !           407: \ifvoid\footins\else\vskip\skip\footins\footnoterule \unvbox\footins\fi
        !           408: \ifr@ggedbottom \kern-\dimen@ \vfil \fi}
        !           409: }
        !           410: 
        !           411: % Here are the rules for the cropmarks.  Note that they are
        !           412: % offset so that the space between them is truly \outerhsize or \outervsize
        !           413: % (P. A. MacKay, 12 November, 1986)
        !           414: %
        !           415: \def\ewtop{\vrule height\cornerthick depth0pt width\cornerlong}
        !           416: \def\nstop{\vbox
        !           417:   {\hrule height\cornerthick depth\cornerlong width\cornerthick}}
        !           418: \def\ewbot{\vrule height0pt depth\cornerthick width\cornerlong}
        !           419: \def\nsbot{\vbox
        !           420:   {\hrule height\cornerlong depth\cornerthick width\cornerthick}}
        !           421: 
        !           422: % Parse an argument, then pass it to #1.  The argument is the rest of
        !           423: % the input line (except we remove a trailing comment).  #1 should be a
        !           424: % macro which expects an ordinary undelimited TeX argument.
        !           425: %
        !           426: \def\parsearg{\parseargusing{}}
        !           427: \def\parseargusing#1#2{%
        !           428:   \def\argtorun{#2}%
        !           429:   \begingroup
        !           430:     \obeylines
        !           431:     \spaceisspace
        !           432:     #1%
        !           433:     \parseargline\empty% Insert the \empty token, see \finishparsearg below.
        !           434: }
        !           435: 
        !           436: {\obeylines %
        !           437:   \gdef\parseargline#1^^M{%
        !           438:     \endgroup % End of the group started in \parsearg.
        !           439:     \argremovecomment #1\comment\ArgTerm%
        !           440:   }%
        !           441: }
        !           442: 
        !           443: % First remove any @comment, then any @c comment.
        !           444: \def\argremovecomment#1\comment#2\ArgTerm{\argremovec #1\c\ArgTerm}
        !           445: \def\argremovec#1\c#2\ArgTerm{\argcheckspaces#1\^^M\ArgTerm}
        !           446: 
        !           447: % Each occurrence of `\^^M' or `<space>\^^M' is replaced by a single space.
        !           448: %
        !           449: % \argremovec might leave us with trailing space, e.g.,
        !           450: %    @end itemize  @c foo
        !           451: % This space token undergoes the same procedure and is eventually removed
        !           452: % by \finishparsearg.
        !           453: %
        !           454: \def\argcheckspaces#1\^^M{\argcheckspacesX#1\^^M \^^M}
        !           455: \def\argcheckspacesX#1 \^^M{\argcheckspacesY#1\^^M}
        !           456: \def\argcheckspacesY#1\^^M#2\^^M#3\ArgTerm{%
        !           457:   \def\temp{#3}%
        !           458:   \ifx\temp\empty
        !           459:     % Do not use \next, perhaps the caller of \parsearg uses it; reuse \temp:
        !           460:     \let\temp\finishparsearg
        !           461:   \else
        !           462:     \let\temp\argcheckspaces
        !           463:   \fi
        !           464:   % Put the space token in:
        !           465:   \temp#1 #3\ArgTerm
        !           466: }
        !           467: 
        !           468: % If a _delimited_ argument is enclosed in braces, they get stripped; so
        !           469: % to get _exactly_ the rest of the line, we had to prevent such situation.
        !           470: % We prepended an \empty token at the very beginning and we expand it now,
        !           471: % just before passing the control to \argtorun.
        !           472: % (Similarly, we have to think about #3 of \argcheckspacesY above: it is
        !           473: % either the null string, or it ends with \^^M---thus there is no danger
        !           474: % that a pair of braces would be stripped.
        !           475: %
        !           476: % But first, we have to remove the trailing space token.
        !           477: %
        !           478: \def\finishparsearg#1 \ArgTerm{\expandafter\argtorun\expandafter{#1}}
        !           479: 
        !           480: % \parseargdef\foo{...}
        !           481: %      is roughly equivalent to
        !           482: % \def\foo{\parsearg\Xfoo}
        !           483: % \def\Xfoo#1{...}
        !           484: %
        !           485: % Actually, I use \csname\string\foo\endcsname, ie. \\foo, as it is my
        !           486: % favourite TeX trick.  --kasal, 16nov03
        !           487: 
        !           488: \def\parseargdef#1{%
        !           489:   \expandafter \doparseargdef \csname\string#1\endcsname #1%
        !           490: }
        !           491: \def\doparseargdef#1#2{%
        !           492:   \def#2{\parsearg#1}%
        !           493:   \def#1##1%
        !           494: }
        !           495: 
        !           496: % Several utility definitions with active space:
        !           497: {
        !           498:   \obeyspaces
        !           499:   \gdef\obeyedspace{ }
        !           500: 
        !           501:   % Make each space character in the input produce a normal interword
        !           502:   % space in the output.  Don't allow a line break at this space, as this
        !           503:   % is used only in environments like @example, where each line of input
        !           504:   % should produce a line of output anyway.
        !           505:   %
        !           506:   \gdef\sepspaces{\obeyspaces\let =\tie}
        !           507: 
        !           508:   % If an index command is used in an @example environment, any spaces
        !           509:   % therein should become regular spaces in the raw index file, not the
        !           510:   % expansion of \tie (\leavevmode \penalty \@M \ ).
        !           511:   \gdef\unsepspaces{\let =\space}
        !           512: }
        !           513: 
        !           514: 
        !           515: \def\flushcr{\ifx\par\lisppar \def\next##1{}\else \let\next=\relax \fi \next}
        !           516: 
        !           517: % Define the framework for environments in texinfo.tex.  It's used like this:
        !           518: %
        !           519: %   \envdef\foo{...}
        !           520: %   \def\Efoo{...}
        !           521: %
        !           522: % It's the responsibility of \envdef to insert \begingroup before the
        !           523: % actual body; @end closes the group after calling \Efoo.  \envdef also
        !           524: % defines \thisenv, so the current environment is known; @end checks
        !           525: % whether the environment name matches.  The \checkenv macro can also be
        !           526: % used to check whether the current environment is the one expected.
        !           527: %
        !           528: % Non-false conditionals (@iftex, @ifset) don't fit into this, so they
        !           529: % are not treated as environments; they don't open a group.  (The
        !           530: % implementation of @end takes care not to call \endgroup in this
        !           531: % special case.)
        !           532: 
        !           533: 
        !           534: % At run-time, environments start with this:
        !           535: \def\startenvironment#1{\begingroup\def\thisenv{#1}}
        !           536: % initialize
        !           537: \let\thisenv\empty
        !           538: 
        !           539: % ... but they get defined via ``\envdef\foo{...}'':
        !           540: \long\def\envdef#1#2{\def#1{\startenvironment#1#2}}
        !           541: \def\envparseargdef#1#2{\parseargdef#1{\startenvironment#1#2}}
        !           542: 
        !           543: % Check whether we're in the right environment:
        !           544: \def\checkenv#1{%
        !           545:   \def\temp{#1}%
        !           546:   \ifx\thisenv\temp
        !           547:   \else
        !           548:     \badenverr
        !           549:   \fi
        !           550: }
        !           551: 
        !           552: % Environment mismatch, #1 expected:
        !           553: \def\badenverr{%
        !           554:   \errhelp = \EMsimple
        !           555:   \errmessage{This command can appear only \inenvironment\temp,
        !           556:     not \inenvironment\thisenv}%
        !           557: }
        !           558: \def\inenvironment#1{%
        !           559:   \ifx#1\empty
        !           560:     outside of any environment%
        !           561:   \else
        !           562:     in environment \expandafter\string#1%
        !           563:   \fi
        !           564: }
        !           565: 
        !           566: % @end foo executes the definition of \Efoo.
        !           567: % But first, it executes a specialized version of \checkenv
        !           568: %
        !           569: \parseargdef\end{%
        !           570:   \if 1\csname iscond.#1\endcsname
        !           571:   \else
        !           572:     % The general wording of \badenverr may not be ideal.
        !           573:     \expandafter\checkenv\csname#1\endcsname
        !           574:     \csname E#1\endcsname
        !           575:     \endgroup
        !           576:   \fi
        !           577: }
        !           578: 
        !           579: \newhelp\EMsimple{Press RETURN to continue.}
        !           580: 
        !           581: 
        !           582: % Be sure we're in horizontal mode when doing a tie, since we make space
        !           583: % equivalent to this in @example-like environments. Otherwise, a space
        !           584: % at the beginning of a line will start with \penalty -- and
        !           585: % since \penalty is valid in vertical mode, we'd end up putting the
        !           586: % penalty on the vertical list instead of in the new paragraph.
        !           587: {\catcode`@ = 11
        !           588:  % Avoid using \@M directly, because that causes trouble
        !           589:  % if the definition is written into an index file.
        !           590:  \global\let\tiepenalty = \@M
        !           591:  \gdef\tie{\leavevmode\penalty\tiepenalty\ }
        !           592: }
        !           593: 
        !           594: % @: forces normal size whitespace following.
        !           595: \def\:{\spacefactor=1000 }
        !           596: 
        !           597: % @* forces a line break.
        !           598: \def\*{\unskip\hfil\break\hbox{}\ignorespaces}
        !           599: 
        !           600: % @/ allows a line break.
        !           601: \let\/=\allowbreak
        !           602: 
        !           603: % @. is an end-of-sentence period.
        !           604: \def\.{.\spacefactor=\endofsentencespacefactor\space}
        !           605: 
        !           606: % @! is an end-of-sentence bang.
        !           607: \def\!{!\spacefactor=\endofsentencespacefactor\space}
        !           608: 
        !           609: % @? is an end-of-sentence query.
        !           610: \def\?{?\spacefactor=\endofsentencespacefactor\space}
        !           611: 
        !           612: % @frenchspacing on|off  says whether to put extra space after punctuation.
        !           613: %
        !           614: \def\onword{on}
        !           615: \def\offword{off}
        !           616: %
        !           617: \parseargdef\frenchspacing{%
        !           618:   \def\temp{#1}%
        !           619:   \ifx\temp\onword \plainfrenchspacing
        !           620:   \else\ifx\temp\offword \plainnonfrenchspacing
        !           621:   \else
        !           622:     \errhelp = \EMsimple
        !           623:     \errmessage{Unknown @frenchspacing option `\temp', must be on|off}%
        !           624:   \fi\fi
        !           625: }
        !           626: 
        !           627: % @w prevents a word break.  Without the \leavevmode, @w at the
        !           628: % beginning of a paragraph, when TeX is still in vertical mode, would
        !           629: % produce a whole line of output instead of starting the paragraph.
        !           630: \def\w#1{\leavevmode\hbox{#1}}
        !           631: 
        !           632: % @group ... @end group forces ... to be all on one page, by enclosing
        !           633: % it in a TeX vbox.  We use \vtop instead of \vbox to construct the box
        !           634: % to keep its height that of a normal line.  According to the rules for
        !           635: % \topskip (p.114 of the TeXbook), the glue inserted is
        !           636: % max (\topskip - \ht (first item), 0).  If that height is large,
        !           637: % therefore, no glue is inserted, and the space between the headline and
        !           638: % the text is small, which looks bad.
        !           639: %
        !           640: % Another complication is that the group might be very large.  This can
        !           641: % cause the glue on the previous page to be unduly stretched, because it
        !           642: % does not have much material.  In this case, it's better to add an
        !           643: % explicit \vfill so that the extra space is at the bottom.  The
        !           644: % threshold for doing this is if the group is more than \vfilllimit
        !           645: % percent of a page (\vfilllimit can be changed inside of @tex).
        !           646: %
        !           647: \newbox\groupbox
        !           648: \def\vfilllimit{0.7}
        !           649: %
        !           650: \envdef\group{%
        !           651:   \ifnum\catcode`\^^M=\active \else
        !           652:     \errhelp = \groupinvalidhelp
        !           653:     \errmessage{@group invalid in context where filling is enabled}%
        !           654:   \fi
        !           655:   \startsavinginserts
        !           656:   %
        !           657:   \setbox\groupbox = \vtop\bgroup
        !           658:     % Do @comment since we are called inside an environment such as
        !           659:     % @example, where each end-of-line in the input causes an
        !           660:     % end-of-line in the output.  We don't want the end-of-line after
        !           661:     % the `@group' to put extra space in the output.  Since @group
        !           662:     % should appear on a line by itself (according to the Texinfo
        !           663:     % manual), we don't worry about eating any user text.
        !           664:     \comment
        !           665: }
        !           666: %
        !           667: % The \vtop produces a box with normal height and large depth; thus, TeX puts
        !           668: % \baselineskip glue before it, and (when the next line of text is done)
        !           669: % \lineskip glue after it.  Thus, space below is not quite equal to space
        !           670: % above.  But it's pretty close.
        !           671: \def\Egroup{%
        !           672:     % To get correct interline space between the last line of the group
        !           673:     % and the first line afterwards, we have to propagate \prevdepth.
        !           674:     \endgraf % Not \par, as it may have been set to \lisppar.
        !           675:     \global\dimen1 = \prevdepth
        !           676:   \egroup           % End the \vtop.
        !           677:   % \dimen0 is the vertical size of the group's box.
        !           678:   \dimen0 = \ht\groupbox  \advance\dimen0 by \dp\groupbox
        !           679:   % \dimen2 is how much space is left on the page (more or less).
        !           680:   \dimen2 = \pageheight   \advance\dimen2 by -\pagetotal
        !           681:   % if the group doesn't fit on the current page, and it's a big big
        !           682:   % group, force a page break.
        !           683:   \ifdim \dimen0 > \dimen2
        !           684:     \ifdim \pagetotal < \vfilllimit\pageheight
        !           685:       \page
        !           686:     \fi
        !           687:   \fi
        !           688:   \box\groupbox
        !           689:   \prevdepth = \dimen1
        !           690:   \checkinserts
        !           691: }
        !           692: %
        !           693: % TeX puts in an \escapechar (i.e., `@') at the beginning of the help
        !           694: % message, so this ends up printing `@group can only ...'.
        !           695: %
        !           696: \newhelp\groupinvalidhelp{%
        !           697: group can only be used in environments such as @example,^^J%
        !           698: where each line of input produces a line of output.}
        !           699: 
        !           700: % @need space-in-mils
        !           701: % forces a page break if there is not space-in-mils remaining.
        !           702: 
        !           703: \newdimen\mil  \mil=0.001in
        !           704: 
        !           705: \parseargdef\need{%
        !           706:   % Ensure vertical mode, so we don't make a big box in the middle of a
        !           707:   % paragraph.
        !           708:   \par
        !           709:   %
        !           710:   % If the @need value is less than one line space, it's useless.
        !           711:   \dimen0 = #1\mil
        !           712:   \dimen2 = \ht\strutbox
        !           713:   \advance\dimen2 by \dp\strutbox
        !           714:   \ifdim\dimen0 > \dimen2
        !           715:     %
        !           716:     % Do a \strut just to make the height of this box be normal, so the
        !           717:     % normal leading is inserted relative to the preceding line.
        !           718:     % And a page break here is fine.
        !           719:     \vtop to #1\mil{\strut\vfil}%
        !           720:     %
        !           721:     % TeX does not even consider page breaks if a penalty added to the
        !           722:     % main vertical list is 10000 or more.  But in order to see if the
        !           723:     % empty box we just added fits on the page, we must make it consider
        !           724:     % page breaks.  On the other hand, we don't want to actually break the
        !           725:     % page after the empty box.  So we use a penalty of 9999.
        !           726:     %
        !           727:     % There is an extremely small chance that TeX will actually break the
        !           728:     % page at this \penalty, if there are no other feasible breakpoints in
        !           729:     % sight.  (If the user is using lots of big @group commands, which
        !           730:     % almost-but-not-quite fill up a page, TeX will have a hard time doing
        !           731:     % good page breaking, for example.)  However, I could not construct an
        !           732:     % example where a page broke at this \penalty; if it happens in a real
        !           733:     % document, then we can reconsider our strategy.
        !           734:     \penalty9999
        !           735:     %
        !           736:     % Back up by the size of the box, whether we did a page break or not.
        !           737:     \kern -#1\mil
        !           738:     %
        !           739:     % Do not allow a page break right after this kern.
        !           740:     \nobreak
        !           741:   \fi
        !           742: }
        !           743: 
        !           744: % @br   forces paragraph break (and is undocumented).
        !           745: 
        !           746: \let\br = \par
        !           747: 
        !           748: % @page forces the start of a new page.
        !           749: %
        !           750: \def\page{\par\vfill\supereject}
        !           751: 
        !           752: % @exdent text....
        !           753: % outputs text on separate line in roman font, starting at standard page margin
        !           754: 
        !           755: % This records the amount of indent in the innermost environment.
        !           756: % That's how much \exdent should take out.
        !           757: \newskip\exdentamount
        !           758: 
        !           759: % This defn is used inside fill environments such as @defun.
        !           760: \parseargdef\exdent{\hfil\break\hbox{\kern -\exdentamount{\rm#1}}\hfil\break}
        !           761: 
        !           762: % This defn is used inside nofill environments such as @example.
        !           763: \parseargdef\nofillexdent{{\advance \leftskip by -\exdentamount
        !           764:   \leftline{\hskip\leftskip{\rm#1}}}}
        !           765: 
        !           766: % @inmargin{WHICH}{TEXT} puts TEXT in the WHICH margin next to the current
        !           767: % paragraph.  For more general purposes, use the \margin insertion
        !           768: % class.  WHICH is `l' or `r'.  Not documented, written for gawk manual.
        !           769: %
        !           770: \newskip\inmarginspacing \inmarginspacing=1cm
        !           771: \def\strutdepth{\dp\strutbox}
        !           772: %
        !           773: \def\doinmargin#1#2{\strut\vadjust{%
        !           774:   \nobreak
        !           775:   \kern-\strutdepth
        !           776:   \vtop to \strutdepth{%
        !           777:     \baselineskip=\strutdepth
        !           778:     \vss
        !           779:     % if you have multiple lines of stuff to put here, you'll need to
        !           780:     % make the vbox yourself of the appropriate size.
        !           781:     \ifx#1l%
        !           782:       \llap{\ignorespaces #2\hskip\inmarginspacing}%
        !           783:     \else
        !           784:       \rlap{\hskip\hsize \hskip\inmarginspacing \ignorespaces #2}%
        !           785:     \fi
        !           786:     \null
        !           787:   }%
        !           788: }}
        !           789: \def\inleftmargin{\doinmargin l}
        !           790: \def\inrightmargin{\doinmargin r}
        !           791: %
        !           792: % @inmargin{TEXT [, RIGHT-TEXT]}
        !           793: % (if RIGHT-TEXT is given, use TEXT for left page, RIGHT-TEXT for right;
        !           794: % else use TEXT for both).
        !           795: %
        !           796: \def\inmargin#1{\parseinmargin #1,,\finish}
        !           797: \def\parseinmargin#1,#2,#3\finish{% not perfect, but better than nothing.
        !           798:   \setbox0 = \hbox{\ignorespaces #2}%
        !           799:   \ifdim\wd0 > 0pt
        !           800:     \def\lefttext{#1}%  have both texts
        !           801:     \def\righttext{#2}%
        !           802:   \else
        !           803:     \def\lefttext{#1}%  have only one text
        !           804:     \def\righttext{#1}%
        !           805:   \fi
        !           806:   %
        !           807:   \ifodd\pageno
        !           808:     \def\temp{\inrightmargin\righttext}% odd page -> outside is right margin
        !           809:   \else
        !           810:     \def\temp{\inleftmargin\lefttext}%
        !           811:   \fi
        !           812:   \temp
        !           813: }
        !           814: 
        !           815: % @| inserts a changebar to the left of the current line.  It should
        !           816: % surround any changed text.  This approach does *not* work if the
        !           817: % change spans more than two lines of output.  To handle that, we would
        !           818: % have adopt a much more difficult approach (putting marks into the main
        !           819: % vertical list for the beginning and end of each change).  This command
        !           820: % is not documented, not supported, and doesn't work.
        !           821: %
        !           822: \def\|{%
        !           823:   % \vadjust can only be used in horizontal mode.
        !           824:   \leavevmode
        !           825:   %
        !           826:   % Append this vertical mode material after the current line in the output.
        !           827:   \vadjust{%
        !           828:     % We want to insert a rule with the height and depth of the current
        !           829:     % leading; that is exactly what \strutbox is supposed to record.
        !           830:     \vskip-\baselineskip
        !           831:     %
        !           832:     % \vadjust-items are inserted at the left edge of the type.  So
        !           833:     % the \llap here moves out into the left-hand margin.
        !           834:     \llap{%
        !           835:       %
        !           836:       % For a thicker or thinner bar, change the `1pt'.
        !           837:       \vrule height\baselineskip width1pt
        !           838:       %
        !           839:       % This is the space between the bar and the text.
        !           840:       \hskip 12pt
        !           841:     }%
        !           842:   }%
        !           843: }
        !           844: 
        !           845: % @include FILE -- \input text of FILE.
        !           846: %
        !           847: \def\include{\parseargusing\filenamecatcodes\includezzz}
        !           848: \def\includezzz#1{%
        !           849:   \pushthisfilestack
        !           850:   \def\thisfile{#1}%
        !           851:   {%
        !           852:     \makevalueexpandable  % we want to expand any @value in FILE.
        !           853:     \turnoffactive        % and allow special characters in the expansion
        !           854:     \indexnofonts         % Allow `@@' and other weird things in file names.
        !           855:     \wlog{texinfo.tex: doing @include of #1^^J}%
        !           856:     \edef\temp{\noexpand\input #1 }%
        !           857:     %
        !           858:     % This trickery is to read FILE outside of a group, in case it makes
        !           859:     % definitions, etc.
        !           860:     \expandafter
        !           861:   }\temp
        !           862:   \popthisfilestack
        !           863: }
        !           864: \def\filenamecatcodes{%
        !           865:   \catcode`\\=\other
        !           866:   \catcode`~=\other
        !           867:   \catcode`^=\other
        !           868:   \catcode`_=\other
        !           869:   \catcode`|=\other
        !           870:   \catcode`<=\other
        !           871:   \catcode`>=\other
        !           872:   \catcode`+=\other
        !           873:   \catcode`-=\other
        !           874:   \catcode`\`=\other
        !           875:   \catcode`\'=\other
        !           876: }
        !           877: 
        !           878: \def\pushthisfilestack{%
        !           879:   \expandafter\pushthisfilestackX\popthisfilestack\StackTerm
        !           880: }
        !           881: \def\pushthisfilestackX{%
        !           882:   \expandafter\pushthisfilestackY\thisfile\StackTerm
        !           883: }
        !           884: \def\pushthisfilestackY #1\StackTerm #2\StackTerm {%
        !           885:   \gdef\popthisfilestack{\gdef\thisfile{#1}\gdef\popthisfilestack{#2}}%
        !           886: }
        !           887: 
        !           888: \def\popthisfilestack{\errthisfilestackempty}
        !           889: \def\errthisfilestackempty{\errmessage{Internal error:
        !           890:   the stack of filenames is empty.}}
        !           891: %
        !           892: \def\thisfile{}
        !           893: 
        !           894: % @center line
        !           895: % outputs that line, centered.
        !           896: %
        !           897: \parseargdef\center{%
        !           898:   \ifhmode
        !           899:     \let\centersub\centerH
        !           900:   \else
        !           901:     \let\centersub\centerV
        !           902:   \fi
        !           903:   \centersub{\hfil \ignorespaces#1\unskip \hfil}%
        !           904:   \let\centersub\relax % don't let the definition persist, just in case
        !           905: }
        !           906: \def\centerH#1{{%
        !           907:   \hfil\break
        !           908:   \advance\hsize by -\leftskip
        !           909:   \advance\hsize by -\rightskip
        !           910:   \line{#1}%
        !           911:   \break
        !           912: }}
        !           913: %
        !           914: \newcount\centerpenalty
        !           915: \def\centerV#1{%
        !           916:   % The idea here is the same as in \startdefun, \cartouche, etc.: if
        !           917:   % @center is the first thing after a section heading, we need to wipe
        !           918:   % out the negative parskip inserted by \sectionheading, but still
        !           919:   % prevent a page break here.
        !           920:   \centerpenalty = \lastpenalty
        !           921:   \ifnum\centerpenalty>10000 \vskip\parskip \fi
        !           922:   \ifnum\centerpenalty>9999 \penalty\centerpenalty \fi
        !           923:   \line{\kern\leftskip #1\kern\rightskip}%
        !           924: }
        !           925: 
        !           926: % @sp n   outputs n lines of vertical space
        !           927: %
        !           928: \parseargdef\sp{\vskip #1\baselineskip}
        !           929: 
        !           930: % @comment ...line which is ignored...
        !           931: % @c is the same as @comment
        !           932: % @ignore ... @end ignore  is another way to write a comment
        !           933: %
        !           934: \def\comment{\begingroup \catcode`\^^M=\other%
        !           935: \catcode`\@=\other \catcode`\{=\other \catcode`\}=\other%
        !           936: \commentxxx}
        !           937: {\catcode`\^^M=\other \gdef\commentxxx#1^^M{\endgroup}}
        !           938: %
        !           939: \let\c=\comment
        !           940: 
        !           941: % @paragraphindent NCHARS
        !           942: % We'll use ems for NCHARS, close enough.
        !           943: % NCHARS can also be the word `asis' or `none'.
        !           944: % We cannot feasibly implement @paragraphindent asis, though.
        !           945: %
        !           946: \def\asisword{asis} % no translation, these are keywords
        !           947: \def\noneword{none}
        !           948: %
        !           949: \parseargdef\paragraphindent{%
        !           950:   \def\temp{#1}%
        !           951:   \ifx\temp\asisword
        !           952:   \else
        !           953:     \ifx\temp\noneword
        !           954:       \defaultparindent = 0pt
        !           955:     \else
        !           956:       \defaultparindent = #1em
        !           957:     \fi
        !           958:   \fi
        !           959:   \parindent = \defaultparindent
        !           960: }
        !           961: 
        !           962: % @exampleindent NCHARS
        !           963: % We'll use ems for NCHARS like @paragraphindent.
        !           964: % It seems @exampleindent asis isn't necessary, but
        !           965: % I preserve it to make it similar to @paragraphindent.
        !           966: \parseargdef\exampleindent{%
        !           967:   \def\temp{#1}%
        !           968:   \ifx\temp\asisword
        !           969:   \else
        !           970:     \ifx\temp\noneword
        !           971:       \lispnarrowing = 0pt
        !           972:     \else
        !           973:       \lispnarrowing = #1em
        !           974:     \fi
        !           975:   \fi
        !           976: }
        !           977: 
        !           978: % @firstparagraphindent WORD
        !           979: % If WORD is `none', then suppress indentation of the first paragraph
        !           980: % after a section heading.  If WORD is `insert', then do indent at such
        !           981: % paragraphs.
        !           982: %
        !           983: % The paragraph indentation is suppressed or not by calling
        !           984: % \suppressfirstparagraphindent, which the sectioning commands do.
        !           985: % We switch the definition of this back and forth according to WORD.
        !           986: % By default, we suppress indentation.
        !           987: %
        !           988: \def\suppressfirstparagraphindent{\dosuppressfirstparagraphindent}
        !           989: \def\insertword{insert}
        !           990: %
        !           991: \parseargdef\firstparagraphindent{%
        !           992:   \def\temp{#1}%
        !           993:   \ifx\temp\noneword
        !           994:     \let\suppressfirstparagraphindent = \dosuppressfirstparagraphindent
        !           995:   \else\ifx\temp\insertword
        !           996:     \let\suppressfirstparagraphindent = \relax
        !           997:   \else
        !           998:     \errhelp = \EMsimple
        !           999:     \errmessage{Unknown @firstparagraphindent option `\temp'}%
        !          1000:   \fi\fi
        !          1001: }
        !          1002: 
        !          1003: % Here is how we actually suppress indentation.  Redefine \everypar to
        !          1004: % \kern backwards by \parindent, and then reset itself to empty.
        !          1005: %
        !          1006: % We also make \indent itself not actually do anything until the next
        !          1007: % paragraph.
        !          1008: %
        !          1009: \gdef\dosuppressfirstparagraphindent{%
        !          1010:   \gdef\indent{%
        !          1011:     \restorefirstparagraphindent
        !          1012:     \indent
        !          1013:   }%
        !          1014:   \gdef\noindent{%
        !          1015:     \restorefirstparagraphindent
        !          1016:     \noindent
        !          1017:   }%
        !          1018:   \global\everypar = {%
        !          1019:     \kern -\parindent
        !          1020:     \restorefirstparagraphindent
        !          1021:   }%
        !          1022: }
        !          1023: 
        !          1024: \gdef\restorefirstparagraphindent{%
        !          1025:   \global \let \indent = \ptexindent
        !          1026:   \global \let \noindent = \ptexnoindent
        !          1027:   \global \everypar = {}%
        !          1028: }
        !          1029: 
        !          1030: 
        !          1031: % @refill is a no-op.
        !          1032: \let\refill=\relax
        !          1033: 
        !          1034: % If working on a large document in chapters, it is convenient to
        !          1035: % be able to disable indexing, cross-referencing, and contents, for test runs.
        !          1036: % This is done with @novalidate (before @setfilename).
        !          1037: %
        !          1038: \newif\iflinks \linkstrue % by default we want the aux files.
        !          1039: \let\novalidate = \linksfalse
        !          1040: 
        !          1041: % @setfilename is done at the beginning of every texinfo file.
        !          1042: % So open here the files we need to have open while reading the input.
        !          1043: % This makes it possible to make a .fmt file for texinfo.
        !          1044: \def\setfilename{%
        !          1045:    \fixbackslash  % Turn off hack to swallow `\input texinfo'.
        !          1046:    \iflinks
        !          1047:      \tryauxfile
        !          1048:      % Open the new aux file.  TeX will close it automatically at exit.
        !          1049:      \immediate\openout\auxfile=\jobname.aux
        !          1050:    \fi % \openindices needs to do some work in any case.
        !          1051:    \openindices
        !          1052:    \let\setfilename=\comment % Ignore extra @setfilename cmds.
        !          1053:    %
        !          1054:    % If texinfo.cnf is present on the system, read it.
        !          1055:    % Useful for site-wide @afourpaper, etc.
        !          1056:    \openin 1 texinfo.cnf
        !          1057:    \ifeof 1 \else \input texinfo.cnf \fi
        !          1058:    \closein 1
        !          1059:    %
        !          1060:    \comment % Ignore the actual filename.
        !          1061: }
        !          1062: 
        !          1063: % Called from \setfilename.
        !          1064: %
        !          1065: \def\openindices{%
        !          1066:   \newindex{cp}%
        !          1067:   \newcodeindex{fn}%
        !          1068:   \newcodeindex{vr}%
        !          1069:   \newcodeindex{tp}%
        !          1070:   \newcodeindex{ky}%
        !          1071:   \newcodeindex{pg}%
        !          1072: }
        !          1073: 
        !          1074: % @bye.
        !          1075: \outer\def\bye{\pagealignmacro\tracingstats=1\ptexend}
        !          1076: 
        !          1077: 
        !          1078: \message{pdf,}
        !          1079: % adobe `portable' document format
        !          1080: \newcount\tempnum
        !          1081: \newcount\lnkcount
        !          1082: \newtoks\filename
        !          1083: \newcount\filenamelength
        !          1084: \newcount\pgn
        !          1085: \newtoks\toksA
        !          1086: \newtoks\toksB
        !          1087: \newtoks\toksC
        !          1088: \newtoks\toksD
        !          1089: \newbox\boxA
        !          1090: \newcount\countA
        !          1091: \newif\ifpdf
        !          1092: \newif\ifpdfmakepagedest
        !          1093: 
        !          1094: % when pdftex is run in dvi mode, \pdfoutput is defined (so \pdfoutput=1
        !          1095: % can be set).  So we test for \relax and 0 as well as being undefined.
        !          1096: \ifx\pdfoutput\thisisundefined
        !          1097: \else
        !          1098:   \ifx\pdfoutput\relax
        !          1099:   \else
        !          1100:     \ifcase\pdfoutput
        !          1101:     \else
        !          1102:       \pdftrue
        !          1103:     \fi
        !          1104:   \fi
        !          1105: \fi
        !          1106: 
        !          1107: % PDF uses PostScript string constants for the names of xref targets,
        !          1108: % for display in the outlines, and in other places.  Thus, we have to
        !          1109: % double any backslashes.  Otherwise, a name like "\node" will be
        !          1110: % interpreted as a newline (\n), followed by o, d, e.  Not good.
        !          1111: % 
        !          1112: % See http://www.ntg.nl/pipermail/ntg-pdftex/2004-July/000654.html and
        !          1113: % related messages.  The final outcome is that it is up to the TeX user
        !          1114: % to double the backslashes and otherwise make the string valid, so
        !          1115: % that's what we do.  pdftex 1.30.0 (ca.2005) introduced a primitive to
        !          1116: % do this reliably, so we use it.
        !          1117: 
        !          1118: % #1 is a control sequence in which to do the replacements,
        !          1119: % which we \xdef.
        !          1120: \def\txiescapepdf#1{%
        !          1121:   \ifx\pdfescapestring\thisisundefined
        !          1122:     % No primitive available; should we give a warning or log?
        !          1123:     % Many times it won't matter.
        !          1124:   \else
        !          1125:     % The expandable \pdfescapestring primitive escapes parentheses,
        !          1126:     % backslashes, and other special chars.
        !          1127:     \xdef#1{\pdfescapestring{#1}}%
        !          1128:   \fi
        !          1129: }
        !          1130: 
        !          1131: \newhelp\nopdfimagehelp{Texinfo supports .png, .jpg, .jpeg, and .pdf images
        !          1132: with PDF output, and none of those formats could be found.  (.eps cannot
        !          1133: be supported due to the design of the PDF format; use regular TeX (DVI
        !          1134: output) for that.)}
        !          1135: 
        !          1136: \ifpdf
        !          1137:   %
        !          1138:   % Color manipulation macros based on pdfcolor.tex,
        !          1139:   % except using rgb instead of cmyk; the latter is said to render as a
        !          1140:   % very dark gray on-screen and a very dark halftone in print, instead
        !          1141:   % of actual black.
        !          1142:   \def\rgbDarkRed{0.50 0.09 0.12}
        !          1143:   \def\rgbBlack{0 0 0}
        !          1144:   %
        !          1145:   % k sets the color for filling (usual text, etc.);
        !          1146:   % K sets the color for stroking (thin rules, e.g., normal _'s).
        !          1147:   \def\pdfsetcolor#1{\pdfliteral{#1 rg  #1 RG}}
        !          1148:   %
        !          1149:   % Set color, and create a mark which defines \thiscolor accordingly,
        !          1150:   % so that \makeheadline knows which color to restore.
        !          1151:   \def\setcolor#1{%
        !          1152:     \xdef\lastcolordefs{\gdef\noexpand\thiscolor{#1}}%
        !          1153:     \domark
        !          1154:     \pdfsetcolor{#1}%
        !          1155:   }
        !          1156:   %
        !          1157:   \def\maincolor{\rgbBlack}
        !          1158:   \pdfsetcolor{\maincolor}
        !          1159:   \edef\thiscolor{\maincolor}
        !          1160:   \def\lastcolordefs{}
        !          1161:   %
        !          1162:   \def\makefootline{%
        !          1163:     \baselineskip24pt
        !          1164:     \line{\pdfsetcolor{\maincolor}\the\footline}%
        !          1165:   }
        !          1166:   %
        !          1167:   \def\makeheadline{%
        !          1168:     \vbox to 0pt{%
        !          1169:       \vskip-22.5pt
        !          1170:       \line{%
        !          1171:         \vbox to8.5pt{}%
        !          1172:         % Extract \thiscolor definition from the marks.
        !          1173:         \getcolormarks
        !          1174:         % Typeset the headline with \maincolor, then restore the color.
        !          1175:         \pdfsetcolor{\maincolor}\the\headline\pdfsetcolor{\thiscolor}%
        !          1176:       }%
        !          1177:       \vss
        !          1178:     }%
        !          1179:     \nointerlineskip
        !          1180:   }
        !          1181:   %
        !          1182:   %
        !          1183:   \pdfcatalog{/PageMode /UseOutlines}
        !          1184:   %
        !          1185:   % #1 is image name, #2 width (might be empty/whitespace), #3 height (ditto).
        !          1186:   \def\dopdfimage#1#2#3{%
        !          1187:     \def\pdfimagewidth{#2}\setbox0 = \hbox{\ignorespaces #2}%
        !          1188:     \def\pdfimageheight{#3}\setbox2 = \hbox{\ignorespaces #3}%
        !          1189:     %
        !          1190:     % pdftex (and the PDF format) support .pdf, .png, .jpg (among
        !          1191:     % others).  Let's try in that order, PDF first since if
        !          1192:     % someone has a scalable image, presumably better to use that than a
        !          1193:     % bitmap.
        !          1194:     \let\pdfimgext=\empty
        !          1195:     \begingroup
        !          1196:       \openin 1 #1.pdf \ifeof 1
        !          1197:         \openin 1 #1.PDF \ifeof 1
        !          1198:           \openin 1 #1.png \ifeof 1
        !          1199:             \openin 1 #1.jpg \ifeof 1
        !          1200:               \openin 1 #1.jpeg \ifeof 1
        !          1201:                 \openin 1 #1.JPG \ifeof 1
        !          1202:                   \errhelp = \nopdfimagehelp
        !          1203:                   \errmessage{Could not find image file #1 for pdf}%
        !          1204:                 \else \gdef\pdfimgext{JPG}%
        !          1205:                 \fi
        !          1206:               \else \gdef\pdfimgext{jpeg}%
        !          1207:               \fi
        !          1208:             \else \gdef\pdfimgext{jpg}%
        !          1209:             \fi
        !          1210:           \else \gdef\pdfimgext{png}%
        !          1211:           \fi
        !          1212:         \else \gdef\pdfimgext{PDF}%
        !          1213:         \fi
        !          1214:       \else \gdef\pdfimgext{pdf}%
        !          1215:       \fi
        !          1216:       \closein 1
        !          1217:     \endgroup
        !          1218:     %
        !          1219:     % without \immediate, ancient pdftex seg faults when the same image is
        !          1220:     % included twice.  (Version 3.14159-pre-1.0-unofficial-20010704.)
        !          1221:     \ifnum\pdftexversion < 14
        !          1222:       \immediate\pdfimage
        !          1223:     \else
        !          1224:       \immediate\pdfximage
        !          1225:     \fi
        !          1226:       \ifdim \wd0 >0pt width \pdfimagewidth \fi
        !          1227:       \ifdim \wd2 >0pt height \pdfimageheight \fi
        !          1228:       \ifnum\pdftexversion<13
        !          1229:          #1.\pdfimgext
        !          1230:        \else
        !          1231:          {#1.\pdfimgext}%
        !          1232:        \fi
        !          1233:     \ifnum\pdftexversion < 14 \else
        !          1234:       \pdfrefximage \pdflastximage
        !          1235:     \fi}
        !          1236:   %
        !          1237:   \def\pdfmkdest#1{{%
        !          1238:     % We have to set dummies so commands such as @code, and characters
        !          1239:     % such as \, aren't expanded when present in a section title.
        !          1240:     \indexnofonts
        !          1241:     \turnoffactive
        !          1242:     \makevalueexpandable
        !          1243:     \def\pdfdestname{#1}%
        !          1244:     \txiescapepdf\pdfdestname
        !          1245:     \safewhatsit{\pdfdest name{\pdfdestname} xyz}%
        !          1246:   }}
        !          1247:   %
        !          1248:   % used to mark target names; must be expandable.
        !          1249:   \def\pdfmkpgn#1{#1}
        !          1250:   %
        !          1251:   % by default, use a color that is dark enough to print on paper as
        !          1252:   % nearly black, but still distinguishable for online viewing.
        !          1253:   \def\urlcolor{\rgbDarkRed}
        !          1254:   \def\linkcolor{\rgbDarkRed}
        !          1255:   \def\endlink{\setcolor{\maincolor}\pdfendlink}
        !          1256:   %
        !          1257:   % Adding outlines to PDF; macros for calculating structure of outlines
        !          1258:   % come from Petr Olsak
        !          1259:   \def\expnumber#1{\expandafter\ifx\csname#1\endcsname\relax 0%
        !          1260:     \else \csname#1\endcsname \fi}
        !          1261:   \def\advancenumber#1{\tempnum=\expnumber{#1}\relax
        !          1262:     \advance\tempnum by 1
        !          1263:     \expandafter\xdef\csname#1\endcsname{\the\tempnum}}
        !          1264:   %
        !          1265:   % #1 is the section text, which is what will be displayed in the
        !          1266:   % outline by the pdf viewer.  #2 is the pdf expression for the number
        !          1267:   % of subentries (or empty, for subsubsections).  #3 is the node text,
        !          1268:   % which might be empty if this toc entry had no corresponding node.
        !          1269:   % #4 is the page number
        !          1270:   %
        !          1271:   \def\dopdfoutline#1#2#3#4{%
        !          1272:     % Generate a link to the node text if that exists; else, use the
        !          1273:     % page number.  We could generate a destination for the section
        !          1274:     % text in the case where a section has no node, but it doesn't
        !          1275:     % seem worth the trouble, since most documents are normally structured.
        !          1276:     \edef\pdfoutlinedest{#3}%
        !          1277:     \ifx\pdfoutlinedest\empty
        !          1278:       \def\pdfoutlinedest{#4}%
        !          1279:     \else
        !          1280:       \txiescapepdf\pdfoutlinedest
        !          1281:     \fi
        !          1282:     %
        !          1283:     % Also escape PDF chars in the display string.
        !          1284:     \edef\pdfoutlinetext{#1}%
        !          1285:     \txiescapepdf\pdfoutlinetext
        !          1286:     %
        !          1287:     \pdfoutline goto name{\pdfmkpgn{\pdfoutlinedest}}#2{\pdfoutlinetext}%
        !          1288:   }
        !          1289:   %
        !          1290:   \def\pdfmakeoutlines{%
        !          1291:     \begingroup
        !          1292:       % Read toc silently, to get counts of subentries for \pdfoutline.
        !          1293:       \def\partentry##1##2##3##4{}% ignore parts in the outlines
        !          1294:       \def\numchapentry##1##2##3##4{%
        !          1295:        \def\thischapnum{##2}%
        !          1296:        \def\thissecnum{0}%
        !          1297:        \def\thissubsecnum{0}%
        !          1298:       }%
        !          1299:       \def\numsecentry##1##2##3##4{%
        !          1300:        \advancenumber{chap\thischapnum}%
        !          1301:        \def\thissecnum{##2}%
        !          1302:        \def\thissubsecnum{0}%
        !          1303:       }%
        !          1304:       \def\numsubsecentry##1##2##3##4{%
        !          1305:        \advancenumber{sec\thissecnum}%
        !          1306:        \def\thissubsecnum{##2}%
        !          1307:       }%
        !          1308:       \def\numsubsubsecentry##1##2##3##4{%
        !          1309:        \advancenumber{subsec\thissubsecnum}%
        !          1310:       }%
        !          1311:       \def\thischapnum{0}%
        !          1312:       \def\thissecnum{0}%
        !          1313:       \def\thissubsecnum{0}%
        !          1314:       %
        !          1315:       % use \def rather than \let here because we redefine \chapentry et
        !          1316:       % al. a second time, below.
        !          1317:       \def\appentry{\numchapentry}%
        !          1318:       \def\appsecentry{\numsecentry}%
        !          1319:       \def\appsubsecentry{\numsubsecentry}%
        !          1320:       \def\appsubsubsecentry{\numsubsubsecentry}%
        !          1321:       \def\unnchapentry{\numchapentry}%
        !          1322:       \def\unnsecentry{\numsecentry}%
        !          1323:       \def\unnsubsecentry{\numsubsecentry}%
        !          1324:       \def\unnsubsubsecentry{\numsubsubsecentry}%
        !          1325:       \readdatafile{toc}%
        !          1326:       %
        !          1327:       % Read toc second time, this time actually producing the outlines.
        !          1328:       % The `-' means take the \expnumber as the absolute number of
        !          1329:       % subentries, which we calculated on our first read of the .toc above.
        !          1330:       %
        !          1331:       % We use the node names as the destinations.
        !          1332:       \def\numchapentry##1##2##3##4{%
        !          1333:         \dopdfoutline{##1}{count-\expnumber{chap##2}}{##3}{##4}}%
        !          1334:       \def\numsecentry##1##2##3##4{%
        !          1335:         \dopdfoutline{##1}{count-\expnumber{sec##2}}{##3}{##4}}%
        !          1336:       \def\numsubsecentry##1##2##3##4{%
        !          1337:         \dopdfoutline{##1}{count-\expnumber{subsec##2}}{##3}{##4}}%
        !          1338:       \def\numsubsubsecentry##1##2##3##4{% count is always zero
        !          1339:         \dopdfoutline{##1}{}{##3}{##4}}%
        !          1340:       %
        !          1341:       % PDF outlines are displayed using system fonts, instead of
        !          1342:       % document fonts.  Therefore we cannot use special characters,
        !          1343:       % since the encoding is unknown.  For example, the eogonek from
        !          1344:       % Latin 2 (0xea) gets translated to a | character.  Info from
        !          1345:       % Staszek Wawrykiewicz, 19 Jan 2004 04:09:24 +0100.
        !          1346:       %
        !          1347:       % TODO this right, we have to translate 8-bit characters to
        !          1348:       % their "best" equivalent, based on the @documentencoding.  Too
        !          1349:       % much work for too little return.  Just use the ASCII equivalents
        !          1350:       % we use for the index sort strings.
        !          1351:       % 
        !          1352:       \indexnofonts
        !          1353:       \setupdatafile
        !          1354:       % We can have normal brace characters in the PDF outlines, unlike
        !          1355:       % Texinfo index files.  So set that up.
        !          1356:       \def\{{\lbracecharliteral}%
        !          1357:       \def\}{\rbracecharliteral}%
        !          1358:       \catcode`\\=\active \otherbackslash
        !          1359:       \input \tocreadfilename
        !          1360:     \endgroup
        !          1361:   }
        !          1362:   {\catcode`[=1 \catcode`]=2
        !          1363:    \catcode`{=\other \catcode`}=\other
        !          1364:    \gdef\lbracecharliteral[{]%
        !          1365:    \gdef\rbracecharliteral[}]%
        !          1366:   ]
        !          1367:   %
        !          1368:   \def\skipspaces#1{\def\PP{#1}\def\D{|}%
        !          1369:     \ifx\PP\D\let\nextsp\relax
        !          1370:     \else\let\nextsp\skipspaces
        !          1371:       \addtokens{\filename}{\PP}%
        !          1372:       \advance\filenamelength by 1
        !          1373:     \fi
        !          1374:     \nextsp}
        !          1375:   \def\getfilename#1{%
        !          1376:     \filenamelength=0
        !          1377:     % If we don't expand the argument now, \skipspaces will get
        !          1378:     % snagged on things like "@value{foo}".
        !          1379:     \edef\temp{#1}%
        !          1380:     \expandafter\skipspaces\temp|\relax
        !          1381:   }
        !          1382:   \ifnum\pdftexversion < 14
        !          1383:     \let \startlink \pdfannotlink
        !          1384:   \else
        !          1385:     \let \startlink \pdfstartlink
        !          1386:   \fi
        !          1387:   % make a live url in pdf output.
        !          1388:   \def\pdfurl#1{%
        !          1389:     \begingroup
        !          1390:       % it seems we really need yet another set of dummies; have not
        !          1391:       % tried to figure out what each command should do in the context
        !          1392:       % of @url.  for now, just make @/ a no-op, that's the only one
        !          1393:       % people have actually reported a problem with.
        !          1394:       %
        !          1395:       \normalturnoffactive
        !          1396:       \def\@{@}%
        !          1397:       \let\/=\empty
        !          1398:       \makevalueexpandable
        !          1399:       % do we want to go so far as to use \indexnofonts instead of just
        !          1400:       % special-casing \var here?
        !          1401:       \def\var##1{##1}%
        !          1402:       %
        !          1403:       \leavevmode\setcolor{\urlcolor}%
        !          1404:       \startlink attr{/Border [0 0 0]}%
        !          1405:         user{/Subtype /Link /A << /S /URI /URI (#1) >>}%
        !          1406:     \endgroup}
        !          1407:   \def\pdfgettoks#1.{\setbox\boxA=\hbox{\toksA={#1.}\toksB={}\maketoks}}
        !          1408:   \def\addtokens#1#2{\edef\addtoks{\noexpand#1={\the#1#2}}\addtoks}
        !          1409:   \def\adn#1{\addtokens{\toksC}{#1}\global\countA=1\let\next=\maketoks}
        !          1410:   \def\poptoks#1#2|ENDTOKS|{\let\first=#1\toksD={#1}\toksA={#2}}
        !          1411:   \def\maketoks{%
        !          1412:     \expandafter\poptoks\the\toksA|ENDTOKS|\relax
        !          1413:     \ifx\first0\adn0
        !          1414:     \else\ifx\first1\adn1 \else\ifx\first2\adn2 \else\ifx\first3\adn3
        !          1415:     \else\ifx\first4\adn4 \else\ifx\first5\adn5 \else\ifx\first6\adn6
        !          1416:     \else\ifx\first7\adn7 \else\ifx\first8\adn8 \else\ifx\first9\adn9
        !          1417:     \else
        !          1418:       \ifnum0=\countA\else\makelink\fi
        !          1419:       \ifx\first.\let\next=\done\else
        !          1420:         \let\next=\maketoks
        !          1421:         \addtokens{\toksB}{\the\toksD}
        !          1422:         \ifx\first,\addtokens{\toksB}{\space}\fi
        !          1423:       \fi
        !          1424:     \fi\fi\fi\fi\fi\fi\fi\fi\fi\fi
        !          1425:     \next}
        !          1426:   \def\makelink{\addtokens{\toksB}%
        !          1427:     {\noexpand\pdflink{\the\toksC}}\toksC={}\global\countA=0}
        !          1428:   \def\pdflink#1{%
        !          1429:     \startlink attr{/Border [0 0 0]} goto name{\pdfmkpgn{#1}}
        !          1430:     \setcolor{\linkcolor}#1\endlink}
        !          1431:   \def\done{\edef\st{\global\noexpand\toksA={\the\toksB}}\st}
        !          1432: \else
        !          1433:   % non-pdf mode
        !          1434:   \let\pdfmkdest = \gobble
        !          1435:   \let\pdfurl = \gobble
        !          1436:   \let\endlink = \relax
        !          1437:   \let\setcolor = \gobble
        !          1438:   \let\pdfsetcolor = \gobble
        !          1439:   \let\pdfmakeoutlines = \relax
        !          1440: \fi  % \ifx\pdfoutput
        !          1441: 
        !          1442: 
        !          1443: \message{fonts,}
        !          1444: 
        !          1445: % Change the current font style to #1, remembering it in \curfontstyle.
        !          1446: % For now, we do not accumulate font styles: @b{@i{foo}} prints foo in
        !          1447: % italics, not bold italics.
        !          1448: %
        !          1449: \def\setfontstyle#1{%
        !          1450:   \def\curfontstyle{#1}% not as a control sequence, because we are \edef'd.
        !          1451:   \csname ten#1\endcsname  % change the current font
        !          1452: }
        !          1453: 
        !          1454: % Select #1 fonts with the current style.
        !          1455: %
        !          1456: \def\selectfonts#1{\csname #1fonts\endcsname \csname\curfontstyle\endcsname}
        !          1457: 
        !          1458: \def\rm{\fam=0 \setfontstyle{rm}}
        !          1459: \def\it{\fam=\itfam \setfontstyle{it}}
        !          1460: \def\sl{\fam=\slfam \setfontstyle{sl}}
        !          1461: \def\bf{\fam=\bffam \setfontstyle{bf}}\def\bfstylename{bf}
        !          1462: \def\tt{\fam=\ttfam \setfontstyle{tt}}
        !          1463: 
        !          1464: % Unfortunately, we have to override this for titles and the like, since
        !          1465: % in those cases "rm" is bold.  Sigh.
        !          1466: \def\rmisbold{\rm\def\curfontstyle{bf}}
        !          1467: 
        !          1468: % Texinfo sort of supports the sans serif font style, which plain TeX does not.
        !          1469: % So we set up a \sf.
        !          1470: \newfam\sffam
        !          1471: \def\sf{\fam=\sffam \setfontstyle{sf}}
        !          1472: \let\li = \sf % Sometimes we call it \li, not \sf.
        !          1473: 
        !          1474: % We don't need math for this font style.
        !          1475: \def\ttsl{\setfontstyle{ttsl}}
        !          1476: 
        !          1477: 
        !          1478: % Set the baselineskip to #1, and the lineskip and strut size
        !          1479: % correspondingly.  There is no deep meaning behind these magic numbers
        !          1480: % used as factors; they just match (closely enough) what Knuth defined.
        !          1481: %
        !          1482: \def\lineskipfactor{.08333}
        !          1483: \def\strutheightpercent{.70833}
        !          1484: \def\strutdepthpercent {.29167}
        !          1485: %
        !          1486: % can get a sort of poor man's double spacing by redefining this.
        !          1487: \def\baselinefactor{1}
        !          1488: %
        !          1489: \newdimen\textleading
        !          1490: \def\setleading#1{%
        !          1491:   \dimen0 = #1\relax
        !          1492:   \normalbaselineskip = \baselinefactor\dimen0
        !          1493:   \normallineskip = \lineskipfactor\normalbaselineskip
        !          1494:   \normalbaselines
        !          1495:   \setbox\strutbox =\hbox{%
        !          1496:     \vrule width0pt height\strutheightpercent\baselineskip
        !          1497:                     depth \strutdepthpercent \baselineskip
        !          1498:   }%
        !          1499: }
        !          1500: 
        !          1501: % PDF CMaps.  See also LaTeX's t1.cmap.
        !          1502: %
        !          1503: % do nothing with this by default.
        !          1504: \expandafter\let\csname cmapOT1\endcsname\gobble
        !          1505: \expandafter\let\csname cmapOT1IT\endcsname\gobble
        !          1506: \expandafter\let\csname cmapOT1TT\endcsname\gobble
        !          1507: 
        !          1508: % if we are producing pdf, and we have \pdffontattr, then define cmaps.
        !          1509: % (\pdffontattr was introduced many years ago, but people still run
        !          1510: % older pdftex's; it's easy to conditionalize, so we do.)
        !          1511: \ifpdf \ifx\pdffontattr\thisisundefined \else
        !          1512:   \begingroup
        !          1513:     \catcode`\^^M=\active \def^^M{^^J}% Output line endings as the ^^J char.
        !          1514:     \catcode`\%=12 \immediate\pdfobj stream {%!PS-Adobe-3.0 Resource-CMap
        !          1515: %%DocumentNeededResources: ProcSet (CIDInit)
        !          1516: %%IncludeResource: ProcSet (CIDInit)
        !          1517: %%BeginResource: CMap (TeX-OT1-0)
        !          1518: %%Title: (TeX-OT1-0 TeX OT1 0)
        !          1519: %%Version: 1.000
        !          1520: %%EndComments
        !          1521: /CIDInit /ProcSet findresource begin
        !          1522: 12 dict begin
        !          1523: begincmap
        !          1524: /CIDSystemInfo
        !          1525: << /Registry (TeX)
        !          1526: /Ordering (OT1)
        !          1527: /Supplement 0
        !          1528: >> def
        !          1529: /CMapName /TeX-OT1-0 def
        !          1530: /CMapType 2 def
        !          1531: 1 begincodespacerange
        !          1532: <00> <7F>
        !          1533: endcodespacerange
        !          1534: 8 beginbfrange
        !          1535: <00> <01> <0393>
        !          1536: <09> <0A> <03A8>
        !          1537: <23> <26> <0023>
        !          1538: <28> <3B> <0028>
        !          1539: <3F> <5B> <003F>
        !          1540: <5D> <5E> <005D>
        !          1541: <61> <7A> <0061>
        !          1542: <7B> <7C> <2013>
        !          1543: endbfrange
        !          1544: 40 beginbfchar
        !          1545: <02> <0398>
        !          1546: <03> <039B>
        !          1547: <04> <039E>
        !          1548: <05> <03A0>
        !          1549: <06> <03A3>
        !          1550: <07> <03D2>
        !          1551: <08> <03A6>
        !          1552: <0B> <00660066>
        !          1553: <0C> <00660069>
        !          1554: <0D> <0066006C>
        !          1555: <0E> <006600660069>
        !          1556: <0F> <00660066006C>
        !          1557: <10> <0131>
        !          1558: <11> <0237>
        !          1559: <12> <0060>
        !          1560: <13> <00B4>
        !          1561: <14> <02C7>
        !          1562: <15> <02D8>
        !          1563: <16> <00AF>
        !          1564: <17> <02DA>
        !          1565: <18> <00B8>
        !          1566: <19> <00DF>
        !          1567: <1A> <00E6>
        !          1568: <1B> <0153>
        !          1569: <1C> <00F8>
        !          1570: <1D> <00C6>
        !          1571: <1E> <0152>
        !          1572: <1F> <00D8>
        !          1573: <21> <0021>
        !          1574: <22> <201D>
        !          1575: <27> <2019>
        !          1576: <3C> <00A1>
        !          1577: <3D> <003D>
        !          1578: <3E> <00BF>
        !          1579: <5C> <201C>
        !          1580: <5F> <02D9>
        !          1581: <60> <2018>
        !          1582: <7D> <02DD>
        !          1583: <7E> <007E>
        !          1584: <7F> <00A8>
        !          1585: endbfchar
        !          1586: endcmap
        !          1587: CMapName currentdict /CMap defineresource pop
        !          1588: end
        !          1589: end
        !          1590: %%EndResource
        !          1591: %%EOF
        !          1592:     }\endgroup
        !          1593:   \expandafter\edef\csname cmapOT1\endcsname#1{%
        !          1594:     \pdffontattr#1{/ToUnicode \the\pdflastobj\space 0 R}%
        !          1595:   }%
        !          1596: %
        !          1597: % \cmapOT1IT
        !          1598:   \begingroup
        !          1599:     \catcode`\^^M=\active \def^^M{^^J}% Output line endings as the ^^J char.
        !          1600:     \catcode`\%=12 \immediate\pdfobj stream {%!PS-Adobe-3.0 Resource-CMap
        !          1601: %%DocumentNeededResources: ProcSet (CIDInit)
        !          1602: %%IncludeResource: ProcSet (CIDInit)
        !          1603: %%BeginResource: CMap (TeX-OT1IT-0)
        !          1604: %%Title: (TeX-OT1IT-0 TeX OT1IT 0)
        !          1605: %%Version: 1.000
        !          1606: %%EndComments
        !          1607: /CIDInit /ProcSet findresource begin
        !          1608: 12 dict begin
        !          1609: begincmap
        !          1610: /CIDSystemInfo
        !          1611: << /Registry (TeX)
        !          1612: /Ordering (OT1IT)
        !          1613: /Supplement 0
        !          1614: >> def
        !          1615: /CMapName /TeX-OT1IT-0 def
        !          1616: /CMapType 2 def
        !          1617: 1 begincodespacerange
        !          1618: <00> <7F>
        !          1619: endcodespacerange
        !          1620: 8 beginbfrange
        !          1621: <00> <01> <0393>
        !          1622: <09> <0A> <03A8>
        !          1623: <25> <26> <0025>
        !          1624: <28> <3B> <0028>
        !          1625: <3F> <5B> <003F>
        !          1626: <5D> <5E> <005D>
        !          1627: <61> <7A> <0061>
        !          1628: <7B> <7C> <2013>
        !          1629: endbfrange
        !          1630: 42 beginbfchar
        !          1631: <02> <0398>
        !          1632: <03> <039B>
        !          1633: <04> <039E>
        !          1634: <05> <03A0>
        !          1635: <06> <03A3>
        !          1636: <07> <03D2>
        !          1637: <08> <03A6>
        !          1638: <0B> <00660066>
        !          1639: <0C> <00660069>
        !          1640: <0D> <0066006C>
        !          1641: <0E> <006600660069>
        !          1642: <0F> <00660066006C>
        !          1643: <10> <0131>
        !          1644: <11> <0237>
        !          1645: <12> <0060>
        !          1646: <13> <00B4>
        !          1647: <14> <02C7>
        !          1648: <15> <02D8>
        !          1649: <16> <00AF>
        !          1650: <17> <02DA>
        !          1651: <18> <00B8>
        !          1652: <19> <00DF>
        !          1653: <1A> <00E6>
        !          1654: <1B> <0153>
        !          1655: <1C> <00F8>
        !          1656: <1D> <00C6>
        !          1657: <1E> <0152>
        !          1658: <1F> <00D8>
        !          1659: <21> <0021>
        !          1660: <22> <201D>
        !          1661: <23> <0023>
        !          1662: <24> <00A3>
        !          1663: <27> <2019>
        !          1664: <3C> <00A1>
        !          1665: <3D> <003D>
        !          1666: <3E> <00BF>
        !          1667: <5C> <201C>
        !          1668: <5F> <02D9>
        !          1669: <60> <2018>
        !          1670: <7D> <02DD>
        !          1671: <7E> <007E>
        !          1672: <7F> <00A8>
        !          1673: endbfchar
        !          1674: endcmap
        !          1675: CMapName currentdict /CMap defineresource pop
        !          1676: end
        !          1677: end
        !          1678: %%EndResource
        !          1679: %%EOF
        !          1680:     }\endgroup
        !          1681:   \expandafter\edef\csname cmapOT1IT\endcsname#1{%
        !          1682:     \pdffontattr#1{/ToUnicode \the\pdflastobj\space 0 R}%
        !          1683:   }%
        !          1684: %
        !          1685: % \cmapOT1TT
        !          1686:   \begingroup
        !          1687:     \catcode`\^^M=\active \def^^M{^^J}% Output line endings as the ^^J char.
        !          1688:     \catcode`\%=12 \immediate\pdfobj stream {%!PS-Adobe-3.0 Resource-CMap
        !          1689: %%DocumentNeededResources: ProcSet (CIDInit)
        !          1690: %%IncludeResource: ProcSet (CIDInit)
        !          1691: %%BeginResource: CMap (TeX-OT1TT-0)
        !          1692: %%Title: (TeX-OT1TT-0 TeX OT1TT 0)
        !          1693: %%Version: 1.000
        !          1694: %%EndComments
        !          1695: /CIDInit /ProcSet findresource begin
        !          1696: 12 dict begin
        !          1697: begincmap
        !          1698: /CIDSystemInfo
        !          1699: << /Registry (TeX)
        !          1700: /Ordering (OT1TT)
        !          1701: /Supplement 0
        !          1702: >> def
        !          1703: /CMapName /TeX-OT1TT-0 def
        !          1704: /CMapType 2 def
        !          1705: 1 begincodespacerange
        !          1706: <00> <7F>
        !          1707: endcodespacerange
        !          1708: 5 beginbfrange
        !          1709: <00> <01> <0393>
        !          1710: <09> <0A> <03A8>
        !          1711: <21> <26> <0021>
        !          1712: <28> <5F> <0028>
        !          1713: <61> <7E> <0061>
        !          1714: endbfrange
        !          1715: 32 beginbfchar
        !          1716: <02> <0398>
        !          1717: <03> <039B>
        !          1718: <04> <039E>
        !          1719: <05> <03A0>
        !          1720: <06> <03A3>
        !          1721: <07> <03D2>
        !          1722: <08> <03A6>
        !          1723: <0B> <2191>
        !          1724: <0C> <2193>
        !          1725: <0D> <0027>
        !          1726: <0E> <00A1>
        !          1727: <0F> <00BF>
        !          1728: <10> <0131>
        !          1729: <11> <0237>
        !          1730: <12> <0060>
        !          1731: <13> <00B4>
        !          1732: <14> <02C7>
        !          1733: <15> <02D8>
        !          1734: <16> <00AF>
        !          1735: <17> <02DA>
        !          1736: <18> <00B8>
        !          1737: <19> <00DF>
        !          1738: <1A> <00E6>
        !          1739: <1B> <0153>
        !          1740: <1C> <00F8>
        !          1741: <1D> <00C6>
        !          1742: <1E> <0152>
        !          1743: <1F> <00D8>
        !          1744: <20> <2423>
        !          1745: <27> <2019>
        !          1746: <60> <2018>
        !          1747: <7F> <00A8>
        !          1748: endbfchar
        !          1749: endcmap
        !          1750: CMapName currentdict /CMap defineresource pop
        !          1751: end
        !          1752: end
        !          1753: %%EndResource
        !          1754: %%EOF
        !          1755:     }\endgroup
        !          1756:   \expandafter\edef\csname cmapOT1TT\endcsname#1{%
        !          1757:     \pdffontattr#1{/ToUnicode \the\pdflastobj\space 0 R}%
        !          1758:   }%
        !          1759: \fi\fi
        !          1760: 
        !          1761: 
        !          1762: % Set the font macro #1 to the font named \fontprefix#2.
        !          1763: % #3 is the font's design size, #4 is a scale factor, #5 is the CMap
        !          1764: % encoding (only OT1, OT1IT and OT1TT are allowed, or empty to omit).
        !          1765: % Example:
        !          1766: % #1 = \textrm
        !          1767: % #2 = \rmshape
        !          1768: % #3 = 10
        !          1769: % #4 = \mainmagstep
        !          1770: % #5 = OT1
        !          1771: %
        !          1772: \def\setfont#1#2#3#4#5{%
        !          1773:   \font#1=\fontprefix#2#3 scaled #4
        !          1774:   \csname cmap#5\endcsname#1%
        !          1775: }
        !          1776: % This is what gets called when #5 of \setfont is empty.
        !          1777: \let\cmap\gobble
        !          1778: %
        !          1779: % (end of cmaps)
        !          1780: 
        !          1781: % Use cm as the default font prefix.
        !          1782: % To specify the font prefix, you must define \fontprefix
        !          1783: % before you read in texinfo.tex.
        !          1784: \ifx\fontprefix\thisisundefined
        !          1785: \def\fontprefix{cm}
        !          1786: \fi
        !          1787: % Support font families that don't use the same naming scheme as CM.
        !          1788: \def\rmshape{r}
        !          1789: \def\rmbshape{bx}               % where the normal face is bold
        !          1790: \def\bfshape{b}
        !          1791: \def\bxshape{bx}
        !          1792: \def\ttshape{tt}
        !          1793: \def\ttbshape{tt}
        !          1794: \def\ttslshape{sltt}
        !          1795: \def\itshape{ti}
        !          1796: \def\itbshape{bxti}
        !          1797: \def\slshape{sl}
        !          1798: \def\slbshape{bxsl}
        !          1799: \def\sfshape{ss}
        !          1800: \def\sfbshape{ss}
        !          1801: \def\scshape{csc}
        !          1802: \def\scbshape{csc}
        !          1803: 
        !          1804: % Definitions for a main text size of 11pt.  (The default in Texinfo.)
        !          1805: %
        !          1806: \def\definetextfontsizexi{%
        !          1807: % Text fonts (11.2pt, magstep1).
        !          1808: \def\textnominalsize{11pt}
        !          1809: \edef\mainmagstep{\magstephalf}
        !          1810: \setfont\textrm\rmshape{10}{\mainmagstep}{OT1}
        !          1811: \setfont\texttt\ttshape{10}{\mainmagstep}{OT1TT}
        !          1812: \setfont\textbf\bfshape{10}{\mainmagstep}{OT1}
        !          1813: \setfont\textit\itshape{10}{\mainmagstep}{OT1IT}
        !          1814: \setfont\textsl\slshape{10}{\mainmagstep}{OT1}
        !          1815: \setfont\textsf\sfshape{10}{\mainmagstep}{OT1}
        !          1816: \setfont\textsc\scshape{10}{\mainmagstep}{OT1}
        !          1817: \setfont\textttsl\ttslshape{10}{\mainmagstep}{OT1TT}
        !          1818: \font\texti=cmmi10 scaled \mainmagstep
        !          1819: \font\textsy=cmsy10 scaled \mainmagstep
        !          1820: \def\textecsize{1095}
        !          1821: 
        !          1822: % A few fonts for @defun names and args.
        !          1823: \setfont\defbf\bfshape{10}{\magstep1}{OT1}
        !          1824: \setfont\deftt\ttshape{10}{\magstep1}{OT1TT}
        !          1825: \setfont\defttsl\ttslshape{10}{\magstep1}{OT1TT}
        !          1826: \def\df{\let\tentt=\deftt \let\tenbf = \defbf \let\tenttsl=\defttsl \bf}
        !          1827: 
        !          1828: % Fonts for indices, footnotes, small examples (9pt).
        !          1829: \def\smallnominalsize{9pt}
        !          1830: \setfont\smallrm\rmshape{9}{1000}{OT1}
        !          1831: \setfont\smalltt\ttshape{9}{1000}{OT1TT}
        !          1832: \setfont\smallbf\bfshape{10}{900}{OT1}
        !          1833: \setfont\smallit\itshape{9}{1000}{OT1IT}
        !          1834: \setfont\smallsl\slshape{9}{1000}{OT1}
        !          1835: \setfont\smallsf\sfshape{9}{1000}{OT1}
        !          1836: \setfont\smallsc\scshape{10}{900}{OT1}
        !          1837: \setfont\smallttsl\ttslshape{10}{900}{OT1TT}
        !          1838: \font\smalli=cmmi9
        !          1839: \font\smallsy=cmsy9
        !          1840: \def\smallecsize{0900}
        !          1841: 
        !          1842: % Fonts for small examples (8pt).
        !          1843: \def\smallernominalsize{8pt}
        !          1844: \setfont\smallerrm\rmshape{8}{1000}{OT1}
        !          1845: \setfont\smallertt\ttshape{8}{1000}{OT1TT}
        !          1846: \setfont\smallerbf\bfshape{10}{800}{OT1}
        !          1847: \setfont\smallerit\itshape{8}{1000}{OT1IT}
        !          1848: \setfont\smallersl\slshape{8}{1000}{OT1}
        !          1849: \setfont\smallersf\sfshape{8}{1000}{OT1}
        !          1850: \setfont\smallersc\scshape{10}{800}{OT1}
        !          1851: \setfont\smallerttsl\ttslshape{10}{800}{OT1TT}
        !          1852: \font\smalleri=cmmi8
        !          1853: \font\smallersy=cmsy8
        !          1854: \def\smallerecsize{0800}
        !          1855: 
        !          1856: % Fonts for title page (20.4pt):
        !          1857: \def\titlenominalsize{20pt}
        !          1858: \setfont\titlerm\rmbshape{12}{\magstep3}{OT1}
        !          1859: \setfont\titleit\itbshape{10}{\magstep4}{OT1IT}
        !          1860: \setfont\titlesl\slbshape{10}{\magstep4}{OT1}
        !          1861: \setfont\titlett\ttbshape{12}{\magstep3}{OT1TT}
        !          1862: \setfont\titlettsl\ttslshape{10}{\magstep4}{OT1TT}
        !          1863: \setfont\titlesf\sfbshape{17}{\magstep1}{OT1}
        !          1864: \let\titlebf=\titlerm
        !          1865: \setfont\titlesc\scbshape{10}{\magstep4}{OT1}
        !          1866: \font\titlei=cmmi12 scaled \magstep3
        !          1867: \font\titlesy=cmsy10 scaled \magstep4
        !          1868: \def\titleecsize{2074}
        !          1869: 
        !          1870: % Chapter (and unnumbered) fonts (17.28pt).
        !          1871: \def\chapnominalsize{17pt}
        !          1872: \setfont\chaprm\rmbshape{12}{\magstep2}{OT1}
        !          1873: \setfont\chapit\itbshape{10}{\magstep3}{OT1IT}
        !          1874: \setfont\chapsl\slbshape{10}{\magstep3}{OT1}
        !          1875: \setfont\chaptt\ttbshape{12}{\magstep2}{OT1TT}
        !          1876: \setfont\chapttsl\ttslshape{10}{\magstep3}{OT1TT}
        !          1877: \setfont\chapsf\sfbshape{17}{1000}{OT1}
        !          1878: \let\chapbf=\chaprm
        !          1879: \setfont\chapsc\scbshape{10}{\magstep3}{OT1}
        !          1880: \font\chapi=cmmi12 scaled \magstep2
        !          1881: \font\chapsy=cmsy10 scaled \magstep3
        !          1882: \def\chapecsize{1728}
        !          1883: 
        !          1884: % Section fonts (14.4pt).
        !          1885: \def\secnominalsize{14pt}
        !          1886: \setfont\secrm\rmbshape{12}{\magstep1}{OT1}
        !          1887: \setfont\secit\itbshape{10}{\magstep2}{OT1IT}
        !          1888: \setfont\secsl\slbshape{10}{\magstep2}{OT1}
        !          1889: \setfont\sectt\ttbshape{12}{\magstep1}{OT1TT}
        !          1890: \setfont\secttsl\ttslshape{10}{\magstep2}{OT1TT}
        !          1891: \setfont\secsf\sfbshape{12}{\magstep1}{OT1}
        !          1892: \let\secbf\secrm
        !          1893: \setfont\secsc\scbshape{10}{\magstep2}{OT1}
        !          1894: \font\seci=cmmi12 scaled \magstep1
        !          1895: \font\secsy=cmsy10 scaled \magstep2
        !          1896: \def\sececsize{1440}
        !          1897: 
        !          1898: % Subsection fonts (13.15pt).
        !          1899: \def\ssecnominalsize{13pt}
        !          1900: \setfont\ssecrm\rmbshape{12}{\magstephalf}{OT1}
        !          1901: \setfont\ssecit\itbshape{10}{1315}{OT1IT}
        !          1902: \setfont\ssecsl\slbshape{10}{1315}{OT1}
        !          1903: \setfont\ssectt\ttbshape{12}{\magstephalf}{OT1TT}
        !          1904: \setfont\ssecttsl\ttslshape{10}{1315}{OT1TT}
        !          1905: \setfont\ssecsf\sfbshape{12}{\magstephalf}{OT1}
        !          1906: \let\ssecbf\ssecrm
        !          1907: \setfont\ssecsc\scbshape{10}{1315}{OT1}
        !          1908: \font\sseci=cmmi12 scaled \magstephalf
        !          1909: \font\ssecsy=cmsy10 scaled 1315
        !          1910: \def\ssececsize{1200}
        !          1911: 
        !          1912: % Reduced fonts for @acro in text (10pt).
        !          1913: \def\reducednominalsize{10pt}
        !          1914: \setfont\reducedrm\rmshape{10}{1000}{OT1}
        !          1915: \setfont\reducedtt\ttshape{10}{1000}{OT1TT}
        !          1916: \setfont\reducedbf\bfshape{10}{1000}{OT1}
        !          1917: \setfont\reducedit\itshape{10}{1000}{OT1IT}
        !          1918: \setfont\reducedsl\slshape{10}{1000}{OT1}
        !          1919: \setfont\reducedsf\sfshape{10}{1000}{OT1}
        !          1920: \setfont\reducedsc\scshape{10}{1000}{OT1}
        !          1921: \setfont\reducedttsl\ttslshape{10}{1000}{OT1TT}
        !          1922: \font\reducedi=cmmi10
        !          1923: \font\reducedsy=cmsy10
        !          1924: \def\reducedecsize{1000}
        !          1925: 
        !          1926: \textleading = 13.2pt % line spacing for 11pt CM
        !          1927: \textfonts            % reset the current fonts
        !          1928: \rm
        !          1929: } % end of 11pt text font size definitions, \definetextfontsizexi
        !          1930: 
        !          1931: 
        !          1932: % Definitions to make the main text be 10pt Computer Modern, with
        !          1933: % section, chapter, etc., sizes following suit.  This is for the GNU
        !          1934: % Press printing of the Emacs 22 manual.  Maybe other manuals in the
        !          1935: % future.  Used with @smallbook, which sets the leading to 12pt.
        !          1936: %
        !          1937: \def\definetextfontsizex{%
        !          1938: % Text fonts (10pt).
        !          1939: \def\textnominalsize{10pt}
        !          1940: \edef\mainmagstep{1000}
        !          1941: \setfont\textrm\rmshape{10}{\mainmagstep}{OT1}
        !          1942: \setfont\texttt\ttshape{10}{\mainmagstep}{OT1TT}
        !          1943: \setfont\textbf\bfshape{10}{\mainmagstep}{OT1}
        !          1944: \setfont\textit\itshape{10}{\mainmagstep}{OT1IT}
        !          1945: \setfont\textsl\slshape{10}{\mainmagstep}{OT1}
        !          1946: \setfont\textsf\sfshape{10}{\mainmagstep}{OT1}
        !          1947: \setfont\textsc\scshape{10}{\mainmagstep}{OT1}
        !          1948: \setfont\textttsl\ttslshape{10}{\mainmagstep}{OT1TT}
        !          1949: \font\texti=cmmi10 scaled \mainmagstep
        !          1950: \font\textsy=cmsy10 scaled \mainmagstep
        !          1951: \def\textecsize{1000}
        !          1952: 
        !          1953: % A few fonts for @defun names and args.
        !          1954: \setfont\defbf\bfshape{10}{\magstephalf}{OT1}
        !          1955: \setfont\deftt\ttshape{10}{\magstephalf}{OT1TT}
        !          1956: \setfont\defttsl\ttslshape{10}{\magstephalf}{OT1TT}
        !          1957: \def\df{\let\tentt=\deftt \let\tenbf = \defbf \let\tenttsl=\defttsl \bf}
        !          1958: 
        !          1959: % Fonts for indices, footnotes, small examples (9pt).
        !          1960: \def\smallnominalsize{9pt}
        !          1961: \setfont\smallrm\rmshape{9}{1000}{OT1}
        !          1962: \setfont\smalltt\ttshape{9}{1000}{OT1TT}
        !          1963: \setfont\smallbf\bfshape{10}{900}{OT1}
        !          1964: \setfont\smallit\itshape{9}{1000}{OT1IT}
        !          1965: \setfont\smallsl\slshape{9}{1000}{OT1}
        !          1966: \setfont\smallsf\sfshape{9}{1000}{OT1}
        !          1967: \setfont\smallsc\scshape{10}{900}{OT1}
        !          1968: \setfont\smallttsl\ttslshape{10}{900}{OT1TT}
        !          1969: \font\smalli=cmmi9
        !          1970: \font\smallsy=cmsy9
        !          1971: \def\smallecsize{0900}
        !          1972: 
        !          1973: % Fonts for small examples (8pt).
        !          1974: \def\smallernominalsize{8pt}
        !          1975: \setfont\smallerrm\rmshape{8}{1000}{OT1}
        !          1976: \setfont\smallertt\ttshape{8}{1000}{OT1TT}
        !          1977: \setfont\smallerbf\bfshape{10}{800}{OT1}
        !          1978: \setfont\smallerit\itshape{8}{1000}{OT1IT}
        !          1979: \setfont\smallersl\slshape{8}{1000}{OT1}
        !          1980: \setfont\smallersf\sfshape{8}{1000}{OT1}
        !          1981: \setfont\smallersc\scshape{10}{800}{OT1}
        !          1982: \setfont\smallerttsl\ttslshape{10}{800}{OT1TT}
        !          1983: \font\smalleri=cmmi8
        !          1984: \font\smallersy=cmsy8
        !          1985: \def\smallerecsize{0800}
        !          1986: 
        !          1987: % Fonts for title page (20.4pt):
        !          1988: \def\titlenominalsize{20pt}
        !          1989: \setfont\titlerm\rmbshape{12}{\magstep3}{OT1}
        !          1990: \setfont\titleit\itbshape{10}{\magstep4}{OT1IT}
        !          1991: \setfont\titlesl\slbshape{10}{\magstep4}{OT1}
        !          1992: \setfont\titlett\ttbshape{12}{\magstep3}{OT1TT}
        !          1993: \setfont\titlettsl\ttslshape{10}{\magstep4}{OT1TT}
        !          1994: \setfont\titlesf\sfbshape{17}{\magstep1}{OT1}
        !          1995: \let\titlebf=\titlerm
        !          1996: \setfont\titlesc\scbshape{10}{\magstep4}{OT1}
        !          1997: \font\titlei=cmmi12 scaled \magstep3
        !          1998: \font\titlesy=cmsy10 scaled \magstep4
        !          1999: \def\titleecsize{2074}
        !          2000: 
        !          2001: % Chapter fonts (14.4pt).
        !          2002: \def\chapnominalsize{14pt}
        !          2003: \setfont\chaprm\rmbshape{12}{\magstep1}{OT1}
        !          2004: \setfont\chapit\itbshape{10}{\magstep2}{OT1IT}
        !          2005: \setfont\chapsl\slbshape{10}{\magstep2}{OT1}
        !          2006: \setfont\chaptt\ttbshape{12}{\magstep1}{OT1TT}
        !          2007: \setfont\chapttsl\ttslshape{10}{\magstep2}{OT1TT}
        !          2008: \setfont\chapsf\sfbshape{12}{\magstep1}{OT1}
        !          2009: \let\chapbf\chaprm
        !          2010: \setfont\chapsc\scbshape{10}{\magstep2}{OT1}
        !          2011: \font\chapi=cmmi12 scaled \magstep1
        !          2012: \font\chapsy=cmsy10 scaled \magstep2
        !          2013: \def\chapecsize{1440}
        !          2014: 
        !          2015: % Section fonts (12pt).
        !          2016: \def\secnominalsize{12pt}
        !          2017: \setfont\secrm\rmbshape{12}{1000}{OT1}
        !          2018: \setfont\secit\itbshape{10}{\magstep1}{OT1IT}
        !          2019: \setfont\secsl\slbshape{10}{\magstep1}{OT1}
        !          2020: \setfont\sectt\ttbshape{12}{1000}{OT1TT}
        !          2021: \setfont\secttsl\ttslshape{10}{\magstep1}{OT1TT}
        !          2022: \setfont\secsf\sfbshape{12}{1000}{OT1}
        !          2023: \let\secbf\secrm
        !          2024: \setfont\secsc\scbshape{10}{\magstep1}{OT1}
        !          2025: \font\seci=cmmi12
        !          2026: \font\secsy=cmsy10 scaled \magstep1
        !          2027: \def\sececsize{1200}
        !          2028: 
        !          2029: % Subsection fonts (10pt).
        !          2030: \def\ssecnominalsize{10pt}
        !          2031: \setfont\ssecrm\rmbshape{10}{1000}{OT1}
        !          2032: \setfont\ssecit\itbshape{10}{1000}{OT1IT}
        !          2033: \setfont\ssecsl\slbshape{10}{1000}{OT1}
        !          2034: \setfont\ssectt\ttbshape{10}{1000}{OT1TT}
        !          2035: \setfont\ssecttsl\ttslshape{10}{1000}{OT1TT}
        !          2036: \setfont\ssecsf\sfbshape{10}{1000}{OT1}
        !          2037: \let\ssecbf\ssecrm
        !          2038: \setfont\ssecsc\scbshape{10}{1000}{OT1}
        !          2039: \font\sseci=cmmi10
        !          2040: \font\ssecsy=cmsy10
        !          2041: \def\ssececsize{1000}
        !          2042: 
        !          2043: % Reduced fonts for @acro in text (9pt).
        !          2044: \def\reducednominalsize{9pt}
        !          2045: \setfont\reducedrm\rmshape{9}{1000}{OT1}
        !          2046: \setfont\reducedtt\ttshape{9}{1000}{OT1TT}
        !          2047: \setfont\reducedbf\bfshape{10}{900}{OT1}
        !          2048: \setfont\reducedit\itshape{9}{1000}{OT1IT}
        !          2049: \setfont\reducedsl\slshape{9}{1000}{OT1}
        !          2050: \setfont\reducedsf\sfshape{9}{1000}{OT1}
        !          2051: \setfont\reducedsc\scshape{10}{900}{OT1}
        !          2052: \setfont\reducedttsl\ttslshape{10}{900}{OT1TT}
        !          2053: \font\reducedi=cmmi9
        !          2054: \font\reducedsy=cmsy9
        !          2055: \def\reducedecsize{0900}
        !          2056: 
        !          2057: \divide\parskip by 2  % reduce space between paragraphs
        !          2058: \textleading = 12pt   % line spacing for 10pt CM
        !          2059: \textfonts            % reset the current fonts
        !          2060: \rm
        !          2061: } % end of 10pt text font size definitions, \definetextfontsizex
        !          2062: 
        !          2063: 
        !          2064: % We provide the user-level command
        !          2065: %   @fonttextsize 10
        !          2066: % (or 11) to redefine the text font size.  pt is assumed.
        !          2067: %
        !          2068: \def\xiword{11}
        !          2069: \def\xword{10}
        !          2070: \def\xwordpt{10pt}
        !          2071: %
        !          2072: \parseargdef\fonttextsize{%
        !          2073:   \def\textsizearg{#1}%
        !          2074:   %\wlog{doing @fonttextsize \textsizearg}%
        !          2075:   %
        !          2076:   % Set \globaldefs so that documents can use this inside @tex, since
        !          2077:   % makeinfo 4.8 does not support it, but we need it nonetheless.
        !          2078:   %
        !          2079:  \begingroup \globaldefs=1
        !          2080:   \ifx\textsizearg\xword \definetextfontsizex
        !          2081:   \else \ifx\textsizearg\xiword \definetextfontsizexi
        !          2082:   \else
        !          2083:     \errhelp=\EMsimple
        !          2084:     \errmessage{@fonttextsize only supports `10' or `11', not `\textsizearg'}
        !          2085:   \fi\fi
        !          2086:  \endgroup
        !          2087: }
        !          2088: 
        !          2089: 
        !          2090: % In order for the font changes to affect most math symbols and letters,
        !          2091: % we have to define the \textfont of the standard families.  Since
        !          2092: % texinfo doesn't allow for producing subscripts and superscripts except
        !          2093: % in the main text, we don't bother to reset \scriptfont and
        !          2094: % \scriptscriptfont (which would also require loading a lot more fonts).
        !          2095: %
        !          2096: \def\resetmathfonts{%
        !          2097:   \textfont0=\tenrm \textfont1=\teni \textfont2=\tensy
        !          2098:   \textfont\itfam=\tenit \textfont\slfam=\tensl \textfont\bffam=\tenbf
        !          2099:   \textfont\ttfam=\tentt \textfont\sffam=\tensf
        !          2100: }
        !          2101: 
        !          2102: % The font-changing commands redefine the meanings of \tenSTYLE, instead
        !          2103: % of just \STYLE.  We do this because \STYLE needs to also set the
        !          2104: % current \fam for math mode.  Our \STYLE (e.g., \rm) commands hardwire
        !          2105: % \tenSTYLE to set the current font.
        !          2106: %
        !          2107: % Each font-changing command also sets the names \lsize (one size lower)
        !          2108: % and \lllsize (three sizes lower).  These relative commands are used in
        !          2109: % the LaTeX logo and acronyms.
        !          2110: %
        !          2111: % This all needs generalizing, badly.
        !          2112: %
        !          2113: \def\textfonts{%
        !          2114:   \let\tenrm=\textrm \let\tenit=\textit \let\tensl=\textsl
        !          2115:   \let\tenbf=\textbf \let\tentt=\texttt \let\smallcaps=\textsc
        !          2116:   \let\tensf=\textsf \let\teni=\texti \let\tensy=\textsy
        !          2117:   \let\tenttsl=\textttsl
        !          2118:   \def\curfontsize{text}%
        !          2119:   \def\lsize{reduced}\def\lllsize{smaller}%
        !          2120:   \resetmathfonts \setleading{\textleading}}
        !          2121: \def\titlefonts{%
        !          2122:   \let\tenrm=\titlerm \let\tenit=\titleit \let\tensl=\titlesl
        !          2123:   \let\tenbf=\titlebf \let\tentt=\titlett \let\smallcaps=\titlesc
        !          2124:   \let\tensf=\titlesf \let\teni=\titlei \let\tensy=\titlesy
        !          2125:   \let\tenttsl=\titlettsl
        !          2126:   \def\curfontsize{title}%
        !          2127:   \def\lsize{chap}\def\lllsize{subsec}%
        !          2128:   \resetmathfonts \setleading{27pt}}
        !          2129: \def\titlefont#1{{\titlefonts\rmisbold #1}}
        !          2130: \def\chapfonts{%
        !          2131:   \let\tenrm=\chaprm \let\tenit=\chapit \let\tensl=\chapsl
        !          2132:   \let\tenbf=\chapbf \let\tentt=\chaptt \let\smallcaps=\chapsc
        !          2133:   \let\tensf=\chapsf \let\teni=\chapi \let\tensy=\chapsy
        !          2134:   \let\tenttsl=\chapttsl
        !          2135:   \def\curfontsize{chap}%
        !          2136:   \def\lsize{sec}\def\lllsize{text}%
        !          2137:   \resetmathfonts \setleading{19pt}}
        !          2138: \def\secfonts{%
        !          2139:   \let\tenrm=\secrm \let\tenit=\secit \let\tensl=\secsl
        !          2140:   \let\tenbf=\secbf \let\tentt=\sectt \let\smallcaps=\secsc
        !          2141:   \let\tensf=\secsf \let\teni=\seci \let\tensy=\secsy
        !          2142:   \let\tenttsl=\secttsl
        !          2143:   \def\curfontsize{sec}%
        !          2144:   \def\lsize{subsec}\def\lllsize{reduced}%
        !          2145:   \resetmathfonts \setleading{16pt}}
        !          2146: \def\subsecfonts{%
        !          2147:   \let\tenrm=\ssecrm \let\tenit=\ssecit \let\tensl=\ssecsl
        !          2148:   \let\tenbf=\ssecbf \let\tentt=\ssectt \let\smallcaps=\ssecsc
        !          2149:   \let\tensf=\ssecsf \let\teni=\sseci \let\tensy=\ssecsy
        !          2150:   \let\tenttsl=\ssecttsl
        !          2151:   \def\curfontsize{ssec}%
        !          2152:   \def\lsize{text}\def\lllsize{small}%
        !          2153:   \resetmathfonts \setleading{15pt}}
        !          2154: \let\subsubsecfonts = \subsecfonts
        !          2155: \def\reducedfonts{%
        !          2156:   \let\tenrm=\reducedrm \let\tenit=\reducedit \let\tensl=\reducedsl
        !          2157:   \let\tenbf=\reducedbf \let\tentt=\reducedtt \let\reducedcaps=\reducedsc
        !          2158:   \let\tensf=\reducedsf \let\teni=\reducedi \let\tensy=\reducedsy
        !          2159:   \let\tenttsl=\reducedttsl
        !          2160:   \def\curfontsize{reduced}%
        !          2161:   \def\lsize{small}\def\lllsize{smaller}%
        !          2162:   \resetmathfonts \setleading{10.5pt}}
        !          2163: \def\smallfonts{%
        !          2164:   \let\tenrm=\smallrm \let\tenit=\smallit \let\tensl=\smallsl
        !          2165:   \let\tenbf=\smallbf \let\tentt=\smalltt \let\smallcaps=\smallsc
        !          2166:   \let\tensf=\smallsf \let\teni=\smalli \let\tensy=\smallsy
        !          2167:   \let\tenttsl=\smallttsl
        !          2168:   \def\curfontsize{small}%
        !          2169:   \def\lsize{smaller}\def\lllsize{smaller}%
        !          2170:   \resetmathfonts \setleading{10.5pt}}
        !          2171: \def\smallerfonts{%
        !          2172:   \let\tenrm=\smallerrm \let\tenit=\smallerit \let\tensl=\smallersl
        !          2173:   \let\tenbf=\smallerbf \let\tentt=\smallertt \let\smallcaps=\smallersc
        !          2174:   \let\tensf=\smallersf \let\teni=\smalleri \let\tensy=\smallersy
        !          2175:   \let\tenttsl=\smallerttsl
        !          2176:   \def\curfontsize{smaller}%
        !          2177:   \def\lsize{smaller}\def\lllsize{smaller}%
        !          2178:   \resetmathfonts \setleading{9.5pt}}
        !          2179: 
        !          2180: % Fonts for short table of contents.
        !          2181: \setfont\shortcontrm\rmshape{12}{1000}{OT1}
        !          2182: \setfont\shortcontbf\bfshape{10}{\magstep1}{OT1}  % no cmb12
        !          2183: \setfont\shortcontsl\slshape{12}{1000}{OT1}
        !          2184: \setfont\shortconttt\ttshape{12}{1000}{OT1TT}
        !          2185: 
        !          2186: % Define these just so they can be easily changed for other fonts.
        !          2187: \def\angleleft{$\langle$}
        !          2188: \def\angleright{$\rangle$}
        !          2189: 
        !          2190: % Set the fonts to use with the @small... environments.
        !          2191: \let\smallexamplefonts = \smallfonts
        !          2192: 
        !          2193: % About \smallexamplefonts.  If we use \smallfonts (9pt), @smallexample
        !          2194: % can fit this many characters:
        !          2195: %   8.5x11=86   smallbook=72  a4=90  a5=69
        !          2196: % If we use \scriptfonts (8pt), then we can fit this many characters:
        !          2197: %   8.5x11=90+  smallbook=80  a4=90+  a5=77
        !          2198: % For me, subjectively, the few extra characters that fit aren't worth
        !          2199: % the additional smallness of 8pt.  So I'm making the default 9pt.
        !          2200: %
        !          2201: % By the way, for comparison, here's what fits with @example (10pt):
        !          2202: %   8.5x11=71  smallbook=60  a4=75  a5=58
        !          2203: % --karl, 24jan03.
        !          2204: 
        !          2205: % Set up the default fonts, so we can use them for creating boxes.
        !          2206: %
        !          2207: \definetextfontsizexi
        !          2208: 
        !          2209: 
        !          2210: \message{markup,}
        !          2211: 
        !          2212: % Check if we are currently using a typewriter font.  Since all the
        !          2213: % Computer Modern typewriter fonts have zero interword stretch (and
        !          2214: % shrink), and it is reasonable to expect all typewriter fonts to have
        !          2215: % this property, we can check that font parameter.
        !          2216: %
        !          2217: \def\ifmonospace{\ifdim\fontdimen3\font=0pt }
        !          2218: 
        !          2219: % Markup style infrastructure.  \defmarkupstylesetup\INITMACRO will
        !          2220: % define and register \INITMACRO to be called on markup style changes.
        !          2221: % \INITMACRO can check \currentmarkupstyle for the innermost
        !          2222: % style and the set of \ifmarkupSTYLE switches for all styles
        !          2223: % currently in effect.
        !          2224: \newif\ifmarkupvar
        !          2225: \newif\ifmarkupsamp
        !          2226: \newif\ifmarkupkey
        !          2227: %\newif\ifmarkupfile % @file == @samp.
        !          2228: %\newif\ifmarkupoption % @option == @samp.
        !          2229: \newif\ifmarkupcode
        !          2230: \newif\ifmarkupkbd
        !          2231: %\newif\ifmarkupenv % @env == @code.
        !          2232: %\newif\ifmarkupcommand % @command == @code.
        !          2233: \newif\ifmarkuptex % @tex (and part of @math, for now).
        !          2234: \newif\ifmarkupexample
        !          2235: \newif\ifmarkupverb
        !          2236: \newif\ifmarkupverbatim
        !          2237: 
        !          2238: \let\currentmarkupstyle\empty
        !          2239: 
        !          2240: \def\setupmarkupstyle#1{%
        !          2241:   \csname markup#1true\endcsname
        !          2242:   \def\currentmarkupstyle{#1}%
        !          2243:   \markupstylesetup
        !          2244: }
        !          2245: 
        !          2246: \let\markupstylesetup\empty
        !          2247: 
        !          2248: \def\defmarkupstylesetup#1{%
        !          2249:   \expandafter\def\expandafter\markupstylesetup
        !          2250:     \expandafter{\markupstylesetup #1}%
        !          2251:   \def#1%
        !          2252: }
        !          2253: 
        !          2254: % Markup style setup for left and right quotes.
        !          2255: \defmarkupstylesetup\markupsetuplq{%
        !          2256:   \expandafter\let\expandafter \temp
        !          2257:     \csname markupsetuplq\currentmarkupstyle\endcsname
        !          2258:   \ifx\temp\relax \markupsetuplqdefault \else \temp \fi
        !          2259: }
        !          2260: 
        !          2261: \defmarkupstylesetup\markupsetuprq{%
        !          2262:   \expandafter\let\expandafter \temp
        !          2263:     \csname markupsetuprq\currentmarkupstyle\endcsname
        !          2264:   \ifx\temp\relax \markupsetuprqdefault \else \temp \fi
        !          2265: }
        !          2266: 
        !          2267: {
        !          2268: \catcode`\'=\active
        !          2269: \catcode`\`=\active
        !          2270: 
        !          2271: \gdef\markupsetuplqdefault{\let`\lq}
        !          2272: \gdef\markupsetuprqdefault{\let'\rq}
        !          2273: 
        !          2274: \gdef\markupsetcodequoteleft{\let`\codequoteleft}
        !          2275: \gdef\markupsetcodequoteright{\let'\codequoteright}
        !          2276: }
        !          2277: 
        !          2278: \let\markupsetuplqcode \markupsetcodequoteleft
        !          2279: \let\markupsetuprqcode \markupsetcodequoteright
        !          2280: %
        !          2281: \let\markupsetuplqexample \markupsetcodequoteleft
        !          2282: \let\markupsetuprqexample \markupsetcodequoteright
        !          2283: %
        !          2284: \let\markupsetuplqkbd     \markupsetcodequoteleft
        !          2285: \let\markupsetuprqkbd     \markupsetcodequoteright
        !          2286: %
        !          2287: \let\markupsetuplqsamp \markupsetcodequoteleft
        !          2288: \let\markupsetuprqsamp \markupsetcodequoteright
        !          2289: %
        !          2290: \let\markupsetuplqverb \markupsetcodequoteleft
        !          2291: \let\markupsetuprqverb \markupsetcodequoteright
        !          2292: %
        !          2293: \let\markupsetuplqverbatim \markupsetcodequoteleft
        !          2294: \let\markupsetuprqverbatim \markupsetcodequoteright
        !          2295: 
        !          2296: % Allow an option to not use regular directed right quote/apostrophe
        !          2297: % (char 0x27), but instead the undirected quote from cmtt (char 0x0d).
        !          2298: % The undirected quote is ugly, so don't make it the default, but it
        !          2299: % works for pasting with more pdf viewers (at least evince), the
        !          2300: % lilypond developers report.  xpdf does work with the regular 0x27.
        !          2301: %
        !          2302: \def\codequoteright{%
        !          2303:   \expandafter\ifx\csname SETtxicodequoteundirected\endcsname\relax
        !          2304:     \expandafter\ifx\csname SETcodequoteundirected\endcsname\relax
        !          2305:       '%
        !          2306:     \else \char'15 \fi
        !          2307:   \else \char'15 \fi
        !          2308: }
        !          2309: %
        !          2310: % and a similar option for the left quote char vs. a grave accent.
        !          2311: % Modern fonts display ASCII 0x60 as a grave accent, so some people like
        !          2312: % the code environments to do likewise.
        !          2313: %
        !          2314: \def\codequoteleft{%
        !          2315:   \expandafter\ifx\csname SETtxicodequotebacktick\endcsname\relax
        !          2316:     \expandafter\ifx\csname SETcodequotebacktick\endcsname\relax
        !          2317:       % [Knuth] pp. 380,381,391
        !          2318:       % \relax disables Spanish ligatures ?` and !` of \tt font.
        !          2319:       \relax`%
        !          2320:     \else \char'22 \fi
        !          2321:   \else \char'22 \fi
        !          2322: }
        !          2323: 
        !          2324: % Commands to set the quote options.
        !          2325: % 
        !          2326: \parseargdef\codequoteundirected{%
        !          2327:   \def\temp{#1}%
        !          2328:   \ifx\temp\onword
        !          2329:     \expandafter\let\csname SETtxicodequoteundirected\endcsname
        !          2330:       = t%
        !          2331:   \else\ifx\temp\offword
        !          2332:     \expandafter\let\csname SETtxicodequoteundirected\endcsname
        !          2333:       = \relax
        !          2334:   \else
        !          2335:     \errhelp = \EMsimple
        !          2336:     \errmessage{Unknown @codequoteundirected value `\temp', must be on|off}%
        !          2337:   \fi\fi
        !          2338: }
        !          2339: %
        !          2340: \parseargdef\codequotebacktick{%
        !          2341:   \def\temp{#1}%
        !          2342:   \ifx\temp\onword
        !          2343:     \expandafter\let\csname SETtxicodequotebacktick\endcsname
        !          2344:       = t%
        !          2345:   \else\ifx\temp\offword
        !          2346:     \expandafter\let\csname SETtxicodequotebacktick\endcsname
        !          2347:       = \relax
        !          2348:   \else
        !          2349:     \errhelp = \EMsimple
        !          2350:     \errmessage{Unknown @codequotebacktick value `\temp', must be on|off}%
        !          2351:   \fi\fi
        !          2352: }
        !          2353: 
        !          2354: % [Knuth] pp. 380,381,391, disable Spanish ligatures ?` and !` of \tt font.
        !          2355: \def\noligaturesquoteleft{\relax\lq}
        !          2356: 
        !          2357: % Count depth in font-changes, for error checks
        !          2358: \newcount\fontdepth \fontdepth=0
        !          2359: 
        !          2360: % Font commands.
        !          2361: 
        !          2362: % #1 is the font command (\sl or \it), #2 is the text to slant.
        !          2363: % If we are in a monospaced environment, however, 1) always use \ttsl,
        !          2364: % and 2) do not add an italic correction.
        !          2365: \def\dosmartslant#1#2{%
        !          2366:   \ifusingtt 
        !          2367:     {{\ttsl #2}\let\next=\relax}%
        !          2368:     {\def\next{{#1#2}\futurelet\next\smartitaliccorrection}}%
        !          2369:   \next
        !          2370: }
        !          2371: \def\smartslanted{\dosmartslant\sl}
        !          2372: \def\smartitalic{\dosmartslant\it}
        !          2373: 
        !          2374: % Output an italic correction unless \next (presumed to be the following
        !          2375: % character) is such as not to need one.
        !          2376: \def\smartitaliccorrection{%
        !          2377:   \ifx\next,%
        !          2378:   \else\ifx\next-%
        !          2379:   \else\ifx\next.%
        !          2380:   \else\ptexslash
        !          2381:   \fi\fi\fi
        !          2382:   \aftersmartic
        !          2383: }
        !          2384: 
        !          2385: % Unconditional use \ttsl, and no ic.  @var is set to this for defuns.
        !          2386: \def\ttslanted#1{{\ttsl #1}}
        !          2387: 
        !          2388: % @cite is like \smartslanted except unconditionally use \sl.  We never want
        !          2389: % ttsl for book titles, do we?
        !          2390: \def\cite#1{{\sl #1}\futurelet\next\smartitaliccorrection}
        !          2391: 
        !          2392: \def\aftersmartic{}
        !          2393: \def\var#1{%
        !          2394:   \let\saveaftersmartic = \aftersmartic
        !          2395:   \def\aftersmartic{\null\let\aftersmartic=\saveaftersmartic}%
        !          2396:   \smartslanted{#1}%
        !          2397: }
        !          2398: 
        !          2399: \let\i=\smartitalic
        !          2400: \let\slanted=\smartslanted
        !          2401: \let\dfn=\smartslanted
        !          2402: \let\emph=\smartitalic
        !          2403: 
        !          2404: % Explicit font changes: @r, @sc, undocumented @ii.
        !          2405: \def\r#1{{\rm #1}}              % roman font
        !          2406: \def\sc#1{{\smallcaps#1}}       % smallcaps font
        !          2407: \def\ii#1{{\it #1}}             % italic font
        !          2408: 
        !          2409: % @b, explicit bold.  Also @strong.
        !          2410: \def\b#1{{\bf #1}}
        !          2411: \let\strong=\b
        !          2412: 
        !          2413: % @sansserif, explicit sans.
        !          2414: \def\sansserif#1{{\sf #1}}
        !          2415: 
        !          2416: % We can't just use \exhyphenpenalty, because that only has effect at
        !          2417: % the end of a paragraph.  Restore normal hyphenation at the end of the
        !          2418: % group within which \nohyphenation is presumably called.
        !          2419: %
        !          2420: \def\nohyphenation{\hyphenchar\font = -1  \aftergroup\restorehyphenation}
        !          2421: \def\restorehyphenation{\hyphenchar\font = `- }
        !          2422: 
        !          2423: % Set sfcode to normal for the chars that usually have another value.
        !          2424: % Can't use plain's \frenchspacing because it uses the `\x notation, and
        !          2425: % sometimes \x has an active definition that messes things up.
        !          2426: %
        !          2427: \catcode`@=11
        !          2428:   \def\plainfrenchspacing{%
        !          2429:     \sfcode\dotChar  =\@m \sfcode\questChar=\@m \sfcode\exclamChar=\@m
        !          2430:     \sfcode\colonChar=\@m \sfcode\semiChar =\@m \sfcode\commaChar =\@m
        !          2431:     \def\endofsentencespacefactor{1000}% for @. and friends
        !          2432:   }
        !          2433:   \def\plainnonfrenchspacing{%
        !          2434:     \sfcode`\.3000\sfcode`\?3000\sfcode`\!3000
        !          2435:     \sfcode`\:2000\sfcode`\;1500\sfcode`\,1250
        !          2436:     \def\endofsentencespacefactor{3000}% for @. and friends
        !          2437:   }
        !          2438: \catcode`@=\other
        !          2439: \def\endofsentencespacefactor{3000}% default
        !          2440: 
        !          2441: % @t, explicit typewriter.
        !          2442: \def\t#1{%
        !          2443:   {\tt \rawbackslash \plainfrenchspacing #1}%
        !          2444:   \null
        !          2445: }
        !          2446: 
        !          2447: % @samp.
        !          2448: \def\samp#1{{\setupmarkupstyle{samp}\lq\tclose{#1}\rq\null}}
        !          2449: 
        !          2450: % @indicateurl is \samp, that is, with quotes.
        !          2451: \let\indicateurl=\samp
        !          2452: 
        !          2453: % @code (and similar) prints in typewriter, but with spaces the same
        !          2454: % size as normal in the surrounding text, without hyphenation, etc.
        !          2455: % This is a subroutine for that.
        !          2456: \def\tclose#1{%
        !          2457:   {%
        !          2458:     % Change normal interword space to be same as for the current font.
        !          2459:     \spaceskip = \fontdimen2\font
        !          2460:     %
        !          2461:     % Switch to typewriter.
        !          2462:     \tt
        !          2463:     %
        !          2464:     % But `\ ' produces the large typewriter interword space.
        !          2465:     \def\ {{\spaceskip = 0pt{} }}%
        !          2466:     %
        !          2467:     % Turn off hyphenation.
        !          2468:     \nohyphenation
        !          2469:     %
        !          2470:     \rawbackslash
        !          2471:     \plainfrenchspacing
        !          2472:     #1%
        !          2473:   }%
        !          2474:   \null % reset spacefactor to 1000
        !          2475: }
        !          2476: 
        !          2477: % We *must* turn on hyphenation at `-' and `_' in @code.
        !          2478: % Otherwise, it is too hard to avoid overfull hboxes
        !          2479: % in the Emacs manual, the Library manual, etc.
        !          2480: %
        !          2481: % Unfortunately, TeX uses one parameter (\hyphenchar) to control
        !          2482: % both hyphenation at - and hyphenation within words.
        !          2483: % We must therefore turn them both off (\tclose does that)
        !          2484: % and arrange explicitly to hyphenate at a dash.
        !          2485: %  -- rms.
        !          2486: {
        !          2487:   \catcode`\-=\active \catcode`\_=\active
        !          2488:   \catcode`\'=\active \catcode`\`=\active
        !          2489:   \global\let'=\rq \global\let`=\lq  % default definitions
        !          2490:   %
        !          2491:   \global\def\code{\begingroup
        !          2492:     \setupmarkupstyle{code}%
        !          2493:     % The following should really be moved into \setupmarkupstyle handlers.
        !          2494:     \catcode\dashChar=\active  \catcode\underChar=\active
        !          2495:     \ifallowcodebreaks
        !          2496:      \let-\codedash
        !          2497:      \let_\codeunder
        !          2498:     \else
        !          2499:      \let-\normaldash
        !          2500:      \let_\realunder
        !          2501:     \fi
        !          2502:     \codex
        !          2503:   }
        !          2504: }
        !          2505: 
        !          2506: \def\codex #1{\tclose{#1}\endgroup}
        !          2507: 
        !          2508: \def\normaldash{-}
        !          2509: \def\codedash{-\discretionary{}{}{}}
        !          2510: \def\codeunder{%
        !          2511:   % this is all so @math{@code{var_name}+1} can work.  In math mode, _
        !          2512:   % is "active" (mathcode"8000) and \normalunderscore (or \char95, etc.)
        !          2513:   % will therefore expand the active definition of _, which is us
        !          2514:   % (inside @code that is), therefore an endless loop.
        !          2515:   \ifusingtt{\ifmmode
        !          2516:                \mathchar"075F % class 0=ordinary, family 7=ttfam, pos 0x5F=_.
        !          2517:              \else\normalunderscore \fi
        !          2518:              \discretionary{}{}{}}%
        !          2519:             {\_}%
        !          2520: }
        !          2521: 
        !          2522: % An additional complication: the above will allow breaks after, e.g.,
        !          2523: % each of the four underscores in __typeof__.  This is bad.
        !          2524: % @allowcodebreaks provides a document-level way to turn breaking at -
        !          2525: % and _ on and off.
        !          2526: %
        !          2527: \newif\ifallowcodebreaks  \allowcodebreakstrue
        !          2528: 
        !          2529: \def\keywordtrue{true}
        !          2530: \def\keywordfalse{false}
        !          2531: 
        !          2532: \parseargdef\allowcodebreaks{%
        !          2533:   \def\txiarg{#1}%
        !          2534:   \ifx\txiarg\keywordtrue
        !          2535:     \allowcodebreakstrue
        !          2536:   \else\ifx\txiarg\keywordfalse
        !          2537:     \allowcodebreaksfalse
        !          2538:   \else
        !          2539:     \errhelp = \EMsimple
        !          2540:     \errmessage{Unknown @allowcodebreaks option `\txiarg', must be true|false}%
        !          2541:   \fi\fi
        !          2542: }
        !          2543: 
        !          2544: % For @command, @env, @file, @option quotes seem unnecessary,
        !          2545: % so use \code rather than \samp.
        !          2546: \let\command=\code
        !          2547: \let\env=\code
        !          2548: \let\file=\code
        !          2549: \let\option=\code
        !          2550: 
        !          2551: % @uref (abbreviation for `urlref') takes an optional (comma-separated)
        !          2552: % second argument specifying the text to display and an optional third
        !          2553: % arg as text to display instead of (rather than in addition to) the url
        !          2554: % itself.  First (mandatory) arg is the url.
        !          2555: % (This \urefnobreak definition isn't used now, leaving it for a while
        !          2556: % for comparison.)
        !          2557: \def\urefnobreak#1{\dourefnobreak #1,,,\finish}
        !          2558: \def\dourefnobreak#1,#2,#3,#4\finish{\begingroup
        !          2559:   \unsepspaces
        !          2560:   \pdfurl{#1}%
        !          2561:   \setbox0 = \hbox{\ignorespaces #3}%
        !          2562:   \ifdim\wd0 > 0pt
        !          2563:     \unhbox0 % third arg given, show only that
        !          2564:   \else
        !          2565:     \setbox0 = \hbox{\ignorespaces #2}%
        !          2566:     \ifdim\wd0 > 0pt
        !          2567:       \ifpdf
        !          2568:         \unhbox0             % PDF: 2nd arg given, show only it
        !          2569:       \else
        !          2570:         \unhbox0\ (\code{#1})% DVI: 2nd arg given, show both it and url
        !          2571:       \fi
        !          2572:     \else
        !          2573:       \code{#1}% only url given, so show it
        !          2574:     \fi
        !          2575:   \fi
        !          2576:   \endlink
        !          2577: \endgroup}
        !          2578: 
        !          2579: % This \urefbreak definition is the active one.
        !          2580: \def\urefbreak{\begingroup \urefcatcodes \dourefbreak}
        !          2581: \let\uref=\urefbreak
        !          2582: \def\dourefbreak#1{\urefbreakfinish #1,,,\finish}
        !          2583: \def\urefbreakfinish#1,#2,#3,#4\finish{% doesn't work in @example
        !          2584:   \unsepspaces
        !          2585:   \pdfurl{#1}%
        !          2586:   \setbox0 = \hbox{\ignorespaces #3}%
        !          2587:   \ifdim\wd0 > 0pt
        !          2588:     \unhbox0 % third arg given, show only that
        !          2589:   \else
        !          2590:     \setbox0 = \hbox{\ignorespaces #2}%
        !          2591:     \ifdim\wd0 > 0pt
        !          2592:       \ifpdf
        !          2593:         \unhbox0             % PDF: 2nd arg given, show only it
        !          2594:       \else
        !          2595:         \unhbox0\ (\urefcode{#1})% DVI: 2nd arg given, show both it and url
        !          2596:       \fi
        !          2597:     \else
        !          2598:       \urefcode{#1}% only url given, so show it
        !          2599:     \fi
        !          2600:   \fi
        !          2601:   \endlink
        !          2602: \endgroup}
        !          2603: 
        !          2604: % Allow line breaks around only a few characters (only).
        !          2605: \def\urefcatcodes{%
        !          2606:   \catcode\ampChar=\active   \catcode\dotChar=\active
        !          2607:   \catcode\hashChar=\active  \catcode\questChar=\active
        !          2608:   \catcode\slashChar=\active
        !          2609: }
        !          2610: {
        !          2611:   \urefcatcodes
        !          2612:   %
        !          2613:   \global\def\urefcode{\begingroup
        !          2614:     \setupmarkupstyle{code}%
        !          2615:     \urefcatcodes
        !          2616:     \let&\urefcodeamp
        !          2617:     \let.\urefcodedot
        !          2618:     \let#\urefcodehash
        !          2619:     \let?\urefcodequest
        !          2620:     \let/\urefcodeslash
        !          2621:     \codex
        !          2622:   }
        !          2623:   %
        !          2624:   % By default, they are just regular characters.
        !          2625:   \global\def&{\normalamp}
        !          2626:   \global\def.{\normaldot}
        !          2627:   \global\def#{\normalhash}
        !          2628:   \global\def?{\normalquest}
        !          2629:   \global\def/{\normalslash}
        !          2630: }
        !          2631: 
        !          2632: % we put a little stretch before and after the breakable chars, to help
        !          2633: % line breaking of long url's.  The unequal skips make look better in
        !          2634: % cmtt at least, especially for dots.
        !          2635: \def\urefprestretch{\urefprebreak \hskip0pt plus.13em }
        !          2636: \def\urefpoststretch{\urefpostbreak \hskip0pt plus.1em }
        !          2637: %
        !          2638: \def\urefcodeamp{\urefprestretch \&\urefpoststretch}
        !          2639: \def\urefcodedot{\urefprestretch .\urefpoststretch}
        !          2640: \def\urefcodehash{\urefprestretch \#\urefpoststretch}
        !          2641: \def\urefcodequest{\urefprestretch ?\urefpoststretch}
        !          2642: \def\urefcodeslash{\futurelet\next\urefcodeslashfinish}
        !          2643: {
        !          2644:   \catcode`\/=\active
        !          2645:   \global\def\urefcodeslashfinish{%
        !          2646:     \urefprestretch \slashChar
        !          2647:     % Allow line break only after the final / in a sequence of
        !          2648:     % slashes, to avoid line break between the slashes in http://.
        !          2649:     \ifx\next/\else \urefpoststretch \fi
        !          2650:   }
        !          2651: }
        !          2652: 
        !          2653: % One more complication: by default we'll break after the special
        !          2654: % characters, but some people like to break before the special chars, so
        !          2655: % allow that.  Also allow no breaking at all, for manual control.
        !          2656: % 
        !          2657: \parseargdef\urefbreakstyle{%
        !          2658:   \def\txiarg{#1}%
        !          2659:   \ifx\txiarg\wordnone
        !          2660:     \def\urefprebreak{\nobreak}\def\urefpostbreak{\nobreak}
        !          2661:   \else\ifx\txiarg\wordbefore
        !          2662:     \def\urefprebreak{\allowbreak}\def\urefpostbreak{\nobreak}
        !          2663:   \else\ifx\txiarg\wordafter
        !          2664:     \def\urefprebreak{\nobreak}\def\urefpostbreak{\allowbreak}
        !          2665:   \else
        !          2666:     \errhelp = \EMsimple
        !          2667:     \errmessage{Unknown @urefbreakstyle setting `\txiarg'}%
        !          2668:   \fi\fi\fi
        !          2669: }
        !          2670: \def\wordafter{after}
        !          2671: \def\wordbefore{before}
        !          2672: \def\wordnone{none}
        !          2673: 
        !          2674: \urefbreakstyle after
        !          2675: 
        !          2676: % @url synonym for @uref, since that's how everyone uses it.
        !          2677: %
        !          2678: \let\url=\uref
        !          2679: 
        !          2680: % rms does not like angle brackets --karl, 17may97.
        !          2681: % So now @email is just like @uref, unless we are pdf.
        !          2682: %
        !          2683: %\def\email#1{\angleleft{\tt #1}\angleright}
        !          2684: \ifpdf
        !          2685:   \def\email#1{\doemail#1,,\finish}
        !          2686:   \def\doemail#1,#2,#3\finish{\begingroup
        !          2687:     \unsepspaces
        !          2688:     \pdfurl{mailto:#1}%
        !          2689:     \setbox0 = \hbox{\ignorespaces #2}%
        !          2690:     \ifdim\wd0>0pt\unhbox0\else\code{#1}\fi
        !          2691:     \endlink
        !          2692:   \endgroup}
        !          2693: \else
        !          2694:   \let\email=\uref
        !          2695: \fi
        !          2696: 
        !          2697: % @kbdinputstyle -- arg is `distinct' (@kbd uses slanted tty font always),
        !          2698: %   `example' (@kbd uses ttsl only inside of @example and friends),
        !          2699: %   or `code' (@kbd uses normal tty font always).
        !          2700: \parseargdef\kbdinputstyle{%
        !          2701:   \def\txiarg{#1}%
        !          2702:   \ifx\txiarg\worddistinct
        !          2703:     \gdef\kbdexamplefont{\ttsl}\gdef\kbdfont{\ttsl}%
        !          2704:   \else\ifx\txiarg\wordexample
        !          2705:     \gdef\kbdexamplefont{\ttsl}\gdef\kbdfont{\tt}%
        !          2706:   \else\ifx\txiarg\wordcode
        !          2707:     \gdef\kbdexamplefont{\tt}\gdef\kbdfont{\tt}%
        !          2708:   \else
        !          2709:     \errhelp = \EMsimple
        !          2710:     \errmessage{Unknown @kbdinputstyle setting `\txiarg'}%
        !          2711:   \fi\fi\fi
        !          2712: }
        !          2713: \def\worddistinct{distinct}
        !          2714: \def\wordexample{example}
        !          2715: \def\wordcode{code}
        !          2716: 
        !          2717: % Default is `distinct'.
        !          2718: \kbdinputstyle distinct
        !          2719: 
        !          2720: % @kbd is like @code, except that if the argument is just one @key command,
        !          2721: % then @kbd has no effect.
        !          2722: \def\kbd#1{{\def\look{#1}\expandafter\kbdsub\look??\par}}
        !          2723: 
        !          2724: \def\xkey{\key}
        !          2725: \def\kbdsub#1#2#3\par{%
        !          2726:   \def\one{#1}\def\three{#3}\def\threex{??}%
        !          2727:   \ifx\one\xkey\ifx\threex\three \key{#2}%
        !          2728:   \else{\tclose{\kbdfont\setupmarkupstyle{kbd}\look}}\fi
        !          2729:   \else{\tclose{\kbdfont\setupmarkupstyle{kbd}\look}}\fi
        !          2730: }
        !          2731: 
        !          2732: % definition of @key that produces a lozenge.  Doesn't adjust to text size.
        !          2733: %\setfont\keyrm\rmshape{8}{1000}{OT1}
        !          2734: %\font\keysy=cmsy9
        !          2735: %\def\key#1{{\keyrm\textfont2=\keysy \leavevmode\hbox{%
        !          2736: %  \raise0.4pt\hbox{\angleleft}\kern-.08em\vtop{%
        !          2737: %    \vbox{\hrule\kern-0.4pt
        !          2738: %     \hbox{\raise0.4pt\hbox{\vphantom{\angleleft}}#1}}%
        !          2739: %    \kern-0.4pt\hrule}%
        !          2740: %  \kern-.06em\raise0.4pt\hbox{\angleright}}}}
        !          2741: 
        !          2742: % definition of @key with no lozenge.  If the current font is already
        !          2743: % monospace, don't change it; that way, we respect @kbdinputstyle.  But
        !          2744: % if it isn't monospace, then use \tt.
        !          2745: %
        !          2746: \def\key#1{{\setupmarkupstyle{key}%
        !          2747:   \nohyphenation
        !          2748:   \ifmonospace\else\tt\fi
        !          2749:   #1}\null}
        !          2750: 
        !          2751: % @clicksequence{File @click{} Open ...}
        !          2752: \def\clicksequence#1{\begingroup #1\endgroup}
        !          2753: 
        !          2754: % @clickstyle @arrow   (by default)
        !          2755: \parseargdef\clickstyle{\def\click{#1}}
        !          2756: \def\click{\arrow}
        !          2757: 
        !          2758: % Typeset a dimension, e.g., `in' or `pt'.  The only reason for the
        !          2759: % argument is to make the input look right: @dmn{pt} instead of @dmn{}pt.
        !          2760: %
        !          2761: \def\dmn#1{\thinspace #1}
        !          2762: 
        !          2763: % @l was never documented to mean ``switch to the Lisp font'',
        !          2764: % and it is not used as such in any manual I can find.  We need it for
        !          2765: % Polish suppressed-l.  --karl, 22sep96.
        !          2766: %\def\l#1{{\li #1}\null}
        !          2767: 
        !          2768: % @acronym for "FBI", "NATO", and the like.
        !          2769: % We print this one point size smaller, since it's intended for
        !          2770: % all-uppercase.
        !          2771: %
        !          2772: \def\acronym#1{\doacronym #1,,\finish}
        !          2773: \def\doacronym#1,#2,#3\finish{%
        !          2774:   {\selectfonts\lsize #1}%
        !          2775:   \def\temp{#2}%
        !          2776:   \ifx\temp\empty \else
        !          2777:     \space ({\unsepspaces \ignorespaces \temp \unskip})%
        !          2778:   \fi
        !          2779:   \null % reset \spacefactor=1000
        !          2780: }
        !          2781: 
        !          2782: % @abbr for "Comput. J." and the like.
        !          2783: % No font change, but don't do end-of-sentence spacing.
        !          2784: %
        !          2785: \def\abbr#1{\doabbr #1,,\finish}
        !          2786: \def\doabbr#1,#2,#3\finish{%
        !          2787:   {\plainfrenchspacing #1}%
        !          2788:   \def\temp{#2}%
        !          2789:   \ifx\temp\empty \else
        !          2790:     \space ({\unsepspaces \ignorespaces \temp \unskip})%
        !          2791:   \fi
        !          2792:   \null % reset \spacefactor=1000
        !          2793: }
        !          2794: 
        !          2795: % @asis just yields its argument.  Used with @table, for example.
        !          2796: %
        !          2797: \def\asis#1{#1}
        !          2798: 
        !          2799: % @math outputs its argument in math mode.
        !          2800: %
        !          2801: % One complication: _ usually means subscripts, but it could also mean
        !          2802: % an actual _ character, as in @math{@var{some_variable} + 1}.  So make
        !          2803: % _ active, and distinguish by seeing if the current family is \slfam,
        !          2804: % which is what @var uses.
        !          2805: {
        !          2806:   \catcode`\_ = \active
        !          2807:   \gdef\mathunderscore{%
        !          2808:     \catcode`\_=\active
        !          2809:     \def_{\ifnum\fam=\slfam \_\else\sb\fi}%
        !          2810:   }
        !          2811: }
        !          2812: % Another complication: we want \\ (and @\) to output a math (or tt) \.
        !          2813: % FYI, plain.tex uses \\ as a temporary control sequence (for no
        !          2814: % particular reason), but this is not advertised and we don't care.
        !          2815: %
        !          2816: % The \mathchar is class=0=ordinary, family=7=ttfam, position=5C=\.
        !          2817: \def\mathbackslash{\ifnum\fam=\ttfam \mathchar"075C \else\backslash \fi}
        !          2818: %
        !          2819: \def\math{%
        !          2820:   \tex
        !          2821:   \mathunderscore
        !          2822:   \let\\ = \mathbackslash
        !          2823:   \mathactive
        !          2824:   % make the texinfo accent commands work in math mode
        !          2825:   \let\"=\ddot
        !          2826:   \let\'=\acute
        !          2827:   \let\==\bar
        !          2828:   \let\^=\hat
        !          2829:   \let\`=\grave
        !          2830:   \let\u=\breve
        !          2831:   \let\v=\check
        !          2832:   \let\~=\tilde
        !          2833:   \let\dotaccent=\dot
        !          2834:   $\finishmath
        !          2835: }
        !          2836: \def\finishmath#1{#1$\endgroup}  % Close the group opened by \tex.
        !          2837: 
        !          2838: % Some active characters (such as <) are spaced differently in math.
        !          2839: % We have to reset their definitions in case the @math was an argument
        !          2840: % to a command which sets the catcodes (such as @item or @section).
        !          2841: %
        !          2842: {
        !          2843:   \catcode`^ = \active
        !          2844:   \catcode`< = \active
        !          2845:   \catcode`> = \active
        !          2846:   \catcode`+ = \active
        !          2847:   \catcode`' = \active
        !          2848:   \gdef\mathactive{%
        !          2849:     \let^ = \ptexhat
        !          2850:     \let< = \ptexless
        !          2851:     \let> = \ptexgtr
        !          2852:     \let+ = \ptexplus
        !          2853:     \let' = \ptexquoteright
        !          2854:   }
        !          2855: }
        !          2856: 
        !          2857: % ctrl is no longer a Texinfo command, but leave this definition for fun.
        !          2858: \def\ctrl #1{{\tt \rawbackslash \hat}#1}
        !          2859: 
        !          2860: % @inlinefmt{FMTNAME,PROCESSED-TEXT} and @inlineraw{FMTNAME,RAW-TEXT}.
        !          2861: % Ignore unless FMTNAME == tex; then it is like @iftex and @tex,
        !          2862: % except specified as a normal braced arg, so no newlines to worry about.
        !          2863: % 
        !          2864: \def\outfmtnametex{tex}
        !          2865: %
        !          2866: \long\def\inlinefmt#1{\doinlinefmt #1,\finish}
        !          2867: \long\def\doinlinefmt#1,#2,\finish{%
        !          2868:   \def\inlinefmtname{#1}%
        !          2869:   \ifx\inlinefmtname\outfmtnametex \ignorespaces #2\fi
        !          2870: }
        !          2871: % For raw, must switch into @tex before parsing the argument, to avoid
        !          2872: % setting catcodes prematurely.  Doing it this way means that, for
        !          2873: % example, @inlineraw{html, foo{bar} gets a parse error instead of being
        !          2874: % ignored.  But this isn't important because if people want a literal
        !          2875: % *right* brace they would have to use a command anyway, so they may as
        !          2876: % well use a command to get a left brace too.  We could re-use the
        !          2877: % delimiter character idea from \verb, but it seems like overkill.
        !          2878: % 
        !          2879: \long\def\inlineraw{\tex \doinlineraw}
        !          2880: \long\def\doinlineraw#1{\doinlinerawtwo #1,\finish}
        !          2881: \def\doinlinerawtwo#1,#2,\finish{%
        !          2882:   \def\inlinerawname{#1}%
        !          2883:   \ifx\inlinerawname\outfmtnametex \ignorespaces #2\fi
        !          2884:   \endgroup % close group opened by \tex.
        !          2885: }
        !          2886: 
        !          2887: 
        !          2888: \message{glyphs,}
        !          2889: % and logos.
        !          2890: 
        !          2891: % @@ prints an @, as does @atchar{}.
        !          2892: \def\@{\char64 }
        !          2893: \let\atchar=\@
        !          2894: 
        !          2895: % @{ @} @lbracechar{} @rbracechar{} all generate brace characters.
        !          2896: % Unless we're in typewriter, use \ecfont because the CM text fonts do
        !          2897: % not have braces, and we don't want to switch into math.
        !          2898: \def\mylbrace{{\ifmonospace\else\ecfont\fi \char123}}
        !          2899: \def\myrbrace{{\ifmonospace\else\ecfont\fi \char125}}
        !          2900: \let\{=\mylbrace \let\lbracechar=\{
        !          2901: \let\}=\myrbrace \let\rbracechar=\}
        !          2902: \begingroup
        !          2903:   % Definitions to produce \{ and \} commands for indices,
        !          2904:   % and @{ and @} for the aux/toc files.
        !          2905:   \catcode`\{ = \other \catcode`\} = \other
        !          2906:   \catcode`\[ = 1 \catcode`\] = 2
        !          2907:   \catcode`\! = 0 \catcode`\\ = \other
        !          2908:   !gdef!lbracecmd[\{]%
        !          2909:   !gdef!rbracecmd[\}]%
        !          2910:   !gdef!lbraceatcmd[@{]%
        !          2911:   !gdef!rbraceatcmd[@}]%
        !          2912: !endgroup
        !          2913: 
        !          2914: % @comma{} to avoid , parsing problems.
        !          2915: \let\comma = ,
        !          2916: 
        !          2917: % Accents: @, @dotaccent @ringaccent @ubaraccent @udotaccent
        !          2918: % Others are defined by plain TeX: @` @' @" @^ @~ @= @u @v @H.
        !          2919: \let\, = \ptexc
        !          2920: \let\dotaccent = \ptexdot
        !          2921: \def\ringaccent#1{{\accent23 #1}}
        !          2922: \let\tieaccent = \ptext
        !          2923: \let\ubaraccent = \ptexb
        !          2924: \let\udotaccent = \d
        !          2925: 
        !          2926: % Other special characters: @questiondown @exclamdown @ordf @ordm
        !          2927: % Plain TeX defines: @AA @AE @O @OE @L (plus lowercase versions) @ss.
        !          2928: \def\questiondown{?`}
        !          2929: \def\exclamdown{!`}
        !          2930: \def\ordf{\leavevmode\raise1ex\hbox{\selectfonts\lllsize \underbar{a}}}
        !          2931: \def\ordm{\leavevmode\raise1ex\hbox{\selectfonts\lllsize \underbar{o}}}
        !          2932: 
        !          2933: % Dotless i and dotless j, used for accents.
        !          2934: \def\imacro{i}
        !          2935: \def\jmacro{j}
        !          2936: \def\dotless#1{%
        !          2937:   \def\temp{#1}%
        !          2938:   \ifx\temp\imacro \ifmmode\imath \else\ptexi \fi
        !          2939:   \else\ifx\temp\jmacro \ifmmode\jmath \else\j \fi
        !          2940:   \else \errmessage{@dotless can be used only with i or j}%
        !          2941:   \fi\fi
        !          2942: }
        !          2943: 
        !          2944: % The \TeX{} logo, as in plain, but resetting the spacing so that a
        !          2945: % period following counts as ending a sentence.  (Idea found in latex.)
        !          2946: %
        !          2947: \edef\TeX{\TeX \spacefactor=1000 }
        !          2948: 
        !          2949: % @LaTeX{} logo.  Not quite the same results as the definition in
        !          2950: % latex.ltx, since we use a different font for the raised A; it's most
        !          2951: % convenient for us to use an explicitly smaller font, rather than using
        !          2952: % the \scriptstyle font (since we don't reset \scriptstyle and
        !          2953: % \scriptscriptstyle).
        !          2954: %
        !          2955: \def\LaTeX{%
        !          2956:   L\kern-.36em
        !          2957:   {\setbox0=\hbox{T}%
        !          2958:    \vbox to \ht0{\hbox{%
        !          2959:      \ifx\textnominalsize\xwordpt
        !          2960:        % for 10pt running text, \lllsize (8pt) is too small for the A in LaTeX.
        !          2961:        % Revert to plain's \scriptsize, which is 7pt.
        !          2962:        \count255=\the\fam $\fam\count255 \scriptstyle A$%
        !          2963:      \else
        !          2964:        % For 11pt, we can use our lllsize.
        !          2965:        \selectfonts\lllsize A%
        !          2966:      \fi
        !          2967:      }%
        !          2968:      \vss
        !          2969:   }}%
        !          2970:   \kern-.15em
        !          2971:   \TeX
        !          2972: }
        !          2973: 
        !          2974: % Some math mode symbols.
        !          2975: \def\bullet{$\ptexbullet$}
        !          2976: \def\geq{\ifmmode \ge\else $\ge$\fi}
        !          2977: \def\leq{\ifmmode \le\else $\le$\fi}
        !          2978: \def\minus{\ifmmode -\else $-$\fi}
        !          2979: 
        !          2980: % @dots{} outputs an ellipsis using the current font.
        !          2981: % We do .5em per period so that it has the same spacing in the cm
        !          2982: % typewriter fonts as three actual period characters; on the other hand,
        !          2983: % in other typewriter fonts three periods are wider than 1.5em.  So do
        !          2984: % whichever is larger.
        !          2985: %
        !          2986: \def\dots{%
        !          2987:   \leavevmode
        !          2988:   \setbox0=\hbox{...}% get width of three periods
        !          2989:   \ifdim\wd0 > 1.5em
        !          2990:     \dimen0 = \wd0
        !          2991:   \else
        !          2992:     \dimen0 = 1.5em
        !          2993:   \fi
        !          2994:   \hbox to \dimen0{%
        !          2995:     \hskip 0pt plus.25fil
        !          2996:     .\hskip 0pt plus1fil
        !          2997:     .\hskip 0pt plus1fil
        !          2998:     .\hskip 0pt plus.5fil
        !          2999:   }%
        !          3000: }
        !          3001: 
        !          3002: % @enddots{} is an end-of-sentence ellipsis.
        !          3003: %
        !          3004: \def\enddots{%
        !          3005:   \dots
        !          3006:   \spacefactor=\endofsentencespacefactor
        !          3007: }
        !          3008: 
        !          3009: % @point{}, @result{}, @expansion{}, @print{}, @equiv{}.
        !          3010: %
        !          3011: % Since these characters are used in examples, they should be an even number of
        !          3012: % \tt widths. Each \tt character is 1en, so two makes it 1em.
        !          3013: %
        !          3014: \def\point{$\star$}
        !          3015: \def\arrow{\leavevmode\raise.05ex\hbox to 1em{\hfil$\rightarrow$\hfil}}
        !          3016: \def\result{\leavevmode\raise.05ex\hbox to 1em{\hfil$\Rightarrow$\hfil}}
        !          3017: \def\expansion{\leavevmode\hbox to 1em{\hfil$\mapsto$\hfil}}
        !          3018: \def\print{\leavevmode\lower.1ex\hbox to 1em{\hfil$\dashv$\hfil}}
        !          3019: \def\equiv{\leavevmode\hbox to 1em{\hfil$\ptexequiv$\hfil}}
        !          3020: 
        !          3021: % The @error{} command.
        !          3022: % Adapted from the TeXbook's \boxit.
        !          3023: %
        !          3024: \newbox\errorbox
        !          3025: %
        !          3026: {\tentt \global\dimen0 = 3em}% Width of the box.
        !          3027: \dimen2 = .55pt % Thickness of rules
        !          3028: % The text. (`r' is open on the right, `e' somewhat less so on the left.)
        !          3029: \setbox0 = \hbox{\kern-.75pt \reducedsf \putworderror\kern-1.5pt}
        !          3030: %
        !          3031: \setbox\errorbox=\hbox to \dimen0{\hfil
        !          3032:    \hsize = \dimen0 \advance\hsize by -5.8pt % Space to left+right.
        !          3033:    \advance\hsize by -2\dimen2 % Rules.
        !          3034:    \vbox{%
        !          3035:       \hrule height\dimen2
        !          3036:       \hbox{\vrule width\dimen2 \kern3pt          % Space to left of text.
        !          3037:          \vtop{\kern2.4pt \box0 \kern2.4pt}% Space above/below.
        !          3038:          \kern3pt\vrule width\dimen2}% Space to right.
        !          3039:       \hrule height\dimen2}
        !          3040:     \hfil}
        !          3041: %
        !          3042: \def\error{\leavevmode\lower.7ex\copy\errorbox}
        !          3043: 
        !          3044: % @pounds{} is a sterling sign, which Knuth put in the CM italic font.
        !          3045: %
        !          3046: \def\pounds{{\it\$}}
        !          3047: 
        !          3048: % @euro{} comes from a separate font, depending on the current style.
        !          3049: % We use the free feym* fonts from the eurosym package by Henrik
        !          3050: % Theiling, which support regular, slanted, bold and bold slanted (and
        !          3051: % "outlined" (blackboard board, sort of) versions, which we don't need).
        !          3052: % It is available from http://www.ctan.org/tex-archive/fonts/eurosym.
        !          3053: %
        !          3054: % Although only regular is the truly official Euro symbol, we ignore
        !          3055: % that.  The Euro is designed to be slightly taller than the regular
        !          3056: % font height.
        !          3057: %
        !          3058: % feymr - regular
        !          3059: % feymo - slanted
        !          3060: % feybr - bold
        !          3061: % feybo - bold slanted
        !          3062: %
        !          3063: % There is no good (free) typewriter version, to my knowledge.
        !          3064: % A feymr10 euro is ~7.3pt wide, while a normal cmtt10 char is ~5.25pt wide.
        !          3065: % Hmm.
        !          3066: %
        !          3067: % Also doesn't work in math.  Do we need to do math with euro symbols?
        !          3068: % Hope not.
        !          3069: %
        !          3070: %
        !          3071: \def\euro{{\eurofont e}}
        !          3072: \def\eurofont{%
        !          3073:   % We set the font at each command, rather than predefining it in
        !          3074:   % \textfonts and the other font-switching commands, so that
        !          3075:   % installations which never need the symbol don't have to have the
        !          3076:   % font installed.
        !          3077:   %
        !          3078:   % There is only one designed size (nominal 10pt), so we always scale
        !          3079:   % that to the current nominal size.
        !          3080:   %
        !          3081:   % By the way, simply using "at 1em" works for cmr10 and the like, but
        !          3082:   % does not work for cmbx10 and other extended/shrunken fonts.
        !          3083:   %
        !          3084:   \def\eurosize{\csname\curfontsize nominalsize\endcsname}%
        !          3085:   %
        !          3086:   \ifx\curfontstyle\bfstylename
        !          3087:     % bold:
        !          3088:     \font\thiseurofont = \ifusingit{feybo10}{feybr10} at \eurosize
        !          3089:   \else
        !          3090:     % regular:
        !          3091:     \font\thiseurofont = \ifusingit{feymo10}{feymr10} at \eurosize
        !          3092:   \fi
        !          3093:   \thiseurofont
        !          3094: }
        !          3095: 
        !          3096: % Glyphs from the EC fonts.  We don't use \let for the aliases, because
        !          3097: % sometimes we redefine the original macro, and the alias should reflect
        !          3098: % the redefinition.
        !          3099: %
        !          3100: % Use LaTeX names for the Icelandic letters.
        !          3101: \def\DH{{\ecfont \char"D0}} % Eth
        !          3102: \def\dh{{\ecfont \char"F0}} % eth
        !          3103: \def\TH{{\ecfont \char"DE}} % Thorn
        !          3104: \def\th{{\ecfont \char"FE}} % thorn
        !          3105: %
        !          3106: \def\guillemetleft{{\ecfont \char"13}}
        !          3107: \def\guillemotleft{\guillemetleft}
        !          3108: \def\guillemetright{{\ecfont \char"14}}
        !          3109: \def\guillemotright{\guillemetright}
        !          3110: \def\guilsinglleft{{\ecfont \char"0E}}
        !          3111: \def\guilsinglright{{\ecfont \char"0F}}
        !          3112: \def\quotedblbase{{\ecfont \char"12}}
        !          3113: \def\quotesinglbase{{\ecfont \char"0D}}
        !          3114: %
        !          3115: % This positioning is not perfect (see the ogonek LaTeX package), but
        !          3116: % we have the precomposed glyphs for the most common cases.  We put the
        !          3117: % tests to use those glyphs in the single \ogonek macro so we have fewer
        !          3118: % dummy definitions to worry about for index entries, etc.
        !          3119: %
        !          3120: % ogonek is also used with other letters in Lithuanian (IOU), but using
        !          3121: % the precomposed glyphs for those is not so easy since they aren't in
        !          3122: % the same EC font.
        !          3123: \def\ogonek#1{{%
        !          3124:   \def\temp{#1}%
        !          3125:   \ifx\temp\macrocharA\Aogonek
        !          3126:   \else\ifx\temp\macrochara\aogonek
        !          3127:   \else\ifx\temp\macrocharE\Eogonek
        !          3128:   \else\ifx\temp\macrochare\eogonek
        !          3129:   \else
        !          3130:     \ecfont \setbox0=\hbox{#1}%
        !          3131:     \ifdim\ht0=1ex\accent"0C #1%
        !          3132:     \else\ooalign{\unhbox0\crcr\hidewidth\char"0C \hidewidth}%
        !          3133:     \fi
        !          3134:   \fi\fi\fi\fi
        !          3135:   }%
        !          3136: }
        !          3137: \def\Aogonek{{\ecfont \char"81}}\def\macrocharA{A}
        !          3138: \def\aogonek{{\ecfont \char"A1}}\def\macrochara{a}
        !          3139: \def\Eogonek{{\ecfont \char"86}}\def\macrocharE{E}
        !          3140: \def\eogonek{{\ecfont \char"A6}}\def\macrochare{e}
        !          3141: %
        !          3142: % Use the ec* fonts (cm-super in outline format) for non-CM glyphs.
        !          3143: \def\ecfont{%
        !          3144:   % We can't distinguish serif/sans and italic/slanted, but this
        !          3145:   % is used for crude hacks anyway (like adding French and German
        !          3146:   % quotes to documents typeset with CM, where we lose kerning), so
        !          3147:   % hopefully nobody will notice/care.
        !          3148:   \edef\ecsize{\csname\curfontsize ecsize\endcsname}%
        !          3149:   \edef\nominalsize{\csname\curfontsize nominalsize\endcsname}%
        !          3150:   \ifmonospace
        !          3151:     % typewriter:
        !          3152:     \font\thisecfont = ectt\ecsize \space at \nominalsize
        !          3153:   \else
        !          3154:     \ifx\curfontstyle\bfstylename
        !          3155:       % bold:
        !          3156:       \font\thisecfont = ecb\ifusingit{i}{x}\ecsize \space at \nominalsize
        !          3157:     \else
        !          3158:       % regular:
        !          3159:       \font\thisecfont = ec\ifusingit{ti}{rm}\ecsize \space at \nominalsize
        !          3160:     \fi
        !          3161:   \fi
        !          3162:   \thisecfont
        !          3163: }
        !          3164: 
        !          3165: % @registeredsymbol - R in a circle.  The font for the R should really
        !          3166: % be smaller yet, but lllsize is the best we can do for now.
        !          3167: % Adapted from the plain.tex definition of \copyright.
        !          3168: %
        !          3169: \def\registeredsymbol{%
        !          3170:   $^{{\ooalign{\hfil\raise.07ex\hbox{\selectfonts\lllsize R}%
        !          3171:                \hfil\crcr\Orb}}%
        !          3172:     }$%
        !          3173: }
        !          3174: 
        !          3175: % @textdegree - the normal degrees sign.
        !          3176: %
        !          3177: \def\textdegree{$^\circ$}
        !          3178: 
        !          3179: % Laurent Siebenmann reports \Orb undefined with:
        !          3180: %  Textures 1.7.7 (preloaded format=plain 93.10.14)  (68K)  16 APR 2004 02:38
        !          3181: % so we'll define it if necessary.
        !          3182: %
        !          3183: \ifx\Orb\thisisundefined
        !          3184: \def\Orb{\mathhexbox20D}
        !          3185: \fi
        !          3186: 
        !          3187: % Quotes.
        !          3188: \chardef\quotedblleft="5C
        !          3189: \chardef\quotedblright=`\"
        !          3190: \chardef\quoteleft=`\`
        !          3191: \chardef\quoteright=`\'
        !          3192: 
        !          3193: 
        !          3194: \message{page headings,}
        !          3195: 
        !          3196: \newskip\titlepagetopglue \titlepagetopglue = 1.5in
        !          3197: \newskip\titlepagebottomglue \titlepagebottomglue = 2pc
        !          3198: 
        !          3199: % First the title page.  Must do @settitle before @titlepage.
        !          3200: \newif\ifseenauthor
        !          3201: \newif\iffinishedtitlepage
        !          3202: 
        !          3203: % Do an implicit @contents or @shortcontents after @end titlepage if the
        !          3204: % user says @setcontentsaftertitlepage or @setshortcontentsaftertitlepage.
        !          3205: %
        !          3206: \newif\ifsetcontentsaftertitlepage
        !          3207:  \let\setcontentsaftertitlepage = \setcontentsaftertitlepagetrue
        !          3208: \newif\ifsetshortcontentsaftertitlepage
        !          3209:  \let\setshortcontentsaftertitlepage = \setshortcontentsaftertitlepagetrue
        !          3210: 
        !          3211: \parseargdef\shorttitlepage{%
        !          3212:   \begingroup \hbox{}\vskip 1.5in \chaprm \centerline{#1}%
        !          3213:   \endgroup\page\hbox{}\page}
        !          3214: 
        !          3215: \envdef\titlepage{%
        !          3216:   % Open one extra group, as we want to close it in the middle of \Etitlepage.
        !          3217:   \begingroup
        !          3218:     \parindent=0pt \textfonts
        !          3219:     % Leave some space at the very top of the page.
        !          3220:     \vglue\titlepagetopglue
        !          3221:     % No rule at page bottom unless we print one at the top with @title.
        !          3222:     \finishedtitlepagetrue
        !          3223:     %
        !          3224:     % Most title ``pages'' are actually two pages long, with space
        !          3225:     % at the top of the second.  We don't want the ragged left on the second.
        !          3226:     \let\oldpage = \page
        !          3227:     \def\page{%
        !          3228:       \iffinishedtitlepage\else
        !          3229:         \finishtitlepage
        !          3230:       \fi
        !          3231:       \let\page = \oldpage
        !          3232:       \page
        !          3233:       \null
        !          3234:     }%
        !          3235: }
        !          3236: 
        !          3237: \def\Etitlepage{%
        !          3238:     \iffinishedtitlepage\else
        !          3239:        \finishtitlepage
        !          3240:     \fi
        !          3241:     % It is important to do the page break before ending the group,
        !          3242:     % because the headline and footline are only empty inside the group.
        !          3243:     % If we use the new definition of \page, we always get a blank page
        !          3244:     % after the title page, which we certainly don't want.
        !          3245:     \oldpage
        !          3246:   \endgroup
        !          3247:   %
        !          3248:   % Need this before the \...aftertitlepage checks so that if they are
        !          3249:   % in effect the toc pages will come out with page numbers.
        !          3250:   \HEADINGSon
        !          3251:   %
        !          3252:   % If they want short, they certainly want long too.
        !          3253:   \ifsetshortcontentsaftertitlepage
        !          3254:     \shortcontents
        !          3255:     \contents
        !          3256:     \global\let\shortcontents = \relax
        !          3257:     \global\let\contents = \relax
        !          3258:   \fi
        !          3259:   %
        !          3260:   \ifsetcontentsaftertitlepage
        !          3261:     \contents
        !          3262:     \global\let\contents = \relax
        !          3263:     \global\let\shortcontents = \relax
        !          3264:   \fi
        !          3265: }
        !          3266: 
        !          3267: \def\finishtitlepage{%
        !          3268:   \vskip4pt \hrule height 2pt width \hsize
        !          3269:   \vskip\titlepagebottomglue
        !          3270:   \finishedtitlepagetrue
        !          3271: }
        !          3272: 
        !          3273: % Settings used for typesetting titles: no hyphenation, no indentation,
        !          3274: % don't worry much about spacing, ragged right.  This should be used
        !          3275: % inside a \vbox, and fonts need to be set appropriately first.  Because
        !          3276: % it is always used for titles, nothing else, we call \rmisbold.  \par
        !          3277: % should be specified before the end of the \vbox, since a vbox is a group.
        !          3278: % 
        !          3279: \def\raggedtitlesettings{%
        !          3280:   \rmisbold
        !          3281:   \hyphenpenalty=10000
        !          3282:   \parindent=0pt
        !          3283:   \tolerance=5000
        !          3284:   \ptexraggedright
        !          3285: }
        !          3286: 
        !          3287: % Macros to be used within @titlepage:
        !          3288: 
        !          3289: \let\subtitlerm=\tenrm
        !          3290: \def\subtitlefont{\subtitlerm \normalbaselineskip = 13pt \normalbaselines}
        !          3291: 
        !          3292: \parseargdef\title{%
        !          3293:   \checkenv\titlepage
        !          3294:   \vbox{\titlefonts \raggedtitlesettings #1\par}%
        !          3295:   % print a rule at the page bottom also.
        !          3296:   \finishedtitlepagefalse
        !          3297:   \vskip4pt \hrule height 4pt width \hsize \vskip4pt
        !          3298: }
        !          3299: 
        !          3300: \parseargdef\subtitle{%
        !          3301:   \checkenv\titlepage
        !          3302:   {\subtitlefont \rightline{#1}}%
        !          3303: }
        !          3304: 
        !          3305: % @author should come last, but may come many times.
        !          3306: % It can also be used inside @quotation.
        !          3307: %
        !          3308: \parseargdef\author{%
        !          3309:   \def\temp{\quotation}%
        !          3310:   \ifx\thisenv\temp
        !          3311:     \def\quotationauthor{#1}% printed in \Equotation.
        !          3312:   \else
        !          3313:     \checkenv\titlepage
        !          3314:     \ifseenauthor\else \vskip 0pt plus 1filll \seenauthortrue \fi
        !          3315:     {\secfonts\rmisbold \leftline{#1}}%
        !          3316:   \fi
        !          3317: }
        !          3318: 
        !          3319: 
        !          3320: % Set up page headings and footings.
        !          3321: 
        !          3322: \let\thispage=\folio
        !          3323: 
        !          3324: \newtoks\evenheadline    % headline on even pages
        !          3325: \newtoks\oddheadline     % headline on odd pages
        !          3326: \newtoks\evenfootline    % footline on even pages
        !          3327: \newtoks\oddfootline     % footline on odd pages
        !          3328: 
        !          3329: % Now make TeX use those variables
        !          3330: \headline={{\textfonts\rm \ifodd\pageno \the\oddheadline
        !          3331:                             \else \the\evenheadline \fi}}
        !          3332: \footline={{\textfonts\rm \ifodd\pageno \the\oddfootline
        !          3333:                             \else \the\evenfootline \fi}\HEADINGShook}
        !          3334: \let\HEADINGShook=\relax
        !          3335: 
        !          3336: % Commands to set those variables.
        !          3337: % For example, this is what  @headings on  does
        !          3338: % @evenheading @thistitle|@thispage|@thischapter
        !          3339: % @oddheading @thischapter|@thispage|@thistitle
        !          3340: % @evenfooting @thisfile||
        !          3341: % @oddfooting ||@thisfile
        !          3342: 
        !          3343: 
        !          3344: \def\evenheading{\parsearg\evenheadingxxx}
        !          3345: \def\evenheadingxxx #1{\evenheadingyyy #1\|\|\|\|\finish}
        !          3346: \def\evenheadingyyy #1\|#2\|#3\|#4\finish{%
        !          3347: \global\evenheadline={\rlap{\centerline{#2}}\line{#1\hfil#3}}}
        !          3348: 
        !          3349: \def\oddheading{\parsearg\oddheadingxxx}
        !          3350: \def\oddheadingxxx #1{\oddheadingyyy #1\|\|\|\|\finish}
        !          3351: \def\oddheadingyyy #1\|#2\|#3\|#4\finish{%
        !          3352: \global\oddheadline={\rlap{\centerline{#2}}\line{#1\hfil#3}}}
        !          3353: 
        !          3354: \parseargdef\everyheading{\oddheadingxxx{#1}\evenheadingxxx{#1}}%
        !          3355: 
        !          3356: \def\evenfooting{\parsearg\evenfootingxxx}
        !          3357: \def\evenfootingxxx #1{\evenfootingyyy #1\|\|\|\|\finish}
        !          3358: \def\evenfootingyyy #1\|#2\|#3\|#4\finish{%
        !          3359: \global\evenfootline={\rlap{\centerline{#2}}\line{#1\hfil#3}}}
        !          3360: 
        !          3361: \def\oddfooting{\parsearg\oddfootingxxx}
        !          3362: \def\oddfootingxxx #1{\oddfootingyyy #1\|\|\|\|\finish}
        !          3363: \def\oddfootingyyy #1\|#2\|#3\|#4\finish{%
        !          3364:   \global\oddfootline = {\rlap{\centerline{#2}}\line{#1\hfil#3}}%
        !          3365:   %
        !          3366:   % Leave some space for the footline.  Hopefully ok to assume
        !          3367:   % @evenfooting will not be used by itself.
        !          3368:   \global\advance\pageheight by -12pt
        !          3369:   \global\advance\vsize by -12pt
        !          3370: }
        !          3371: 
        !          3372: \parseargdef\everyfooting{\oddfootingxxx{#1}\evenfootingxxx{#1}}
        !          3373: 
        !          3374: % @evenheadingmarks top     \thischapter <- chapter at the top of a page
        !          3375: % @evenheadingmarks bottom  \thischapter <- chapter at the bottom of a page
        !          3376: %
        !          3377: % The same set of arguments for:
        !          3378: %
        !          3379: % @oddheadingmarks
        !          3380: % @evenfootingmarks
        !          3381: % @oddfootingmarks
        !          3382: % @everyheadingmarks
        !          3383: % @everyfootingmarks
        !          3384: 
        !          3385: \def\evenheadingmarks{\headingmarks{even}{heading}}
        !          3386: \def\oddheadingmarks{\headingmarks{odd}{heading}}
        !          3387: \def\evenfootingmarks{\headingmarks{even}{footing}}
        !          3388: \def\oddfootingmarks{\headingmarks{odd}{footing}}
        !          3389: \def\everyheadingmarks#1 {\headingmarks{even}{heading}{#1}
        !          3390:                           \headingmarks{odd}{heading}{#1} }
        !          3391: \def\everyfootingmarks#1 {\headingmarks{even}{footing}{#1}
        !          3392:                           \headingmarks{odd}{footing}{#1} }
        !          3393: % #1 = even/odd, #2 = heading/footing, #3 = top/bottom.
        !          3394: \def\headingmarks#1#2#3 {%
        !          3395:   \expandafter\let\expandafter\temp \csname get#3headingmarks\endcsname
        !          3396:   \global\expandafter\let\csname get#1#2marks\endcsname \temp
        !          3397: }
        !          3398: 
        !          3399: \everyheadingmarks bottom
        !          3400: \everyfootingmarks bottom
        !          3401: 
        !          3402: % @headings double      turns headings on for double-sided printing.
        !          3403: % @headings single      turns headings on for single-sided printing.
        !          3404: % @headings off         turns them off.
        !          3405: % @headings on          same as @headings double, retained for compatibility.
        !          3406: % @headings after       turns on double-sided headings after this page.
        !          3407: % @headings doubleafter turns on double-sided headings after this page.
        !          3408: % @headings singleafter turns on single-sided headings after this page.
        !          3409: % By default, they are off at the start of a document,
        !          3410: % and turned `on' after @end titlepage.
        !          3411: 
        !          3412: \def\headings #1 {\csname HEADINGS#1\endcsname}
        !          3413: 
        !          3414: \def\headingsoff{% non-global headings elimination
        !          3415:   \evenheadline={\hfil}\evenfootline={\hfil}%
        !          3416:    \oddheadline={\hfil}\oddfootline={\hfil}%
        !          3417: }
        !          3418: 
        !          3419: \def\HEADINGSoff{{\globaldefs=1 \headingsoff}} % global setting
        !          3420: \HEADINGSoff  % it's the default
        !          3421: 
        !          3422: % When we turn headings on, set the page number to 1.
        !          3423: % For double-sided printing, put current file name in lower left corner,
        !          3424: % chapter name on inside top of right hand pages, document
        !          3425: % title on inside top of left hand pages, and page numbers on outside top
        !          3426: % edge of all pages.
        !          3427: \def\HEADINGSdouble{%
        !          3428: \global\pageno=1
        !          3429: \global\evenfootline={\hfil}
        !          3430: \global\oddfootline={\hfil}
        !          3431: \global\evenheadline={\line{\folio\hfil\thistitle}}
        !          3432: \global\oddheadline={\line{\thischapter\hfil\folio}}
        !          3433: \global\let\contentsalignmacro = \chapoddpage
        !          3434: }
        !          3435: \let\contentsalignmacro = \chappager
        !          3436: 
        !          3437: % For single-sided printing, chapter title goes across top left of page,
        !          3438: % page number on top right.
        !          3439: \def\HEADINGSsingle{%
        !          3440: \global\pageno=1
        !          3441: \global\evenfootline={\hfil}
        !          3442: \global\oddfootline={\hfil}
        !          3443: \global\evenheadline={\line{\thischapter\hfil\folio}}
        !          3444: \global\oddheadline={\line{\thischapter\hfil\folio}}
        !          3445: \global\let\contentsalignmacro = \chappager
        !          3446: }
        !          3447: \def\HEADINGSon{\HEADINGSdouble}
        !          3448: 
        !          3449: \def\HEADINGSafter{\let\HEADINGShook=\HEADINGSdoublex}
        !          3450: \let\HEADINGSdoubleafter=\HEADINGSafter
        !          3451: \def\HEADINGSdoublex{%
        !          3452: \global\evenfootline={\hfil}
        !          3453: \global\oddfootline={\hfil}
        !          3454: \global\evenheadline={\line{\folio\hfil\thistitle}}
        !          3455: \global\oddheadline={\line{\thischapter\hfil\folio}}
        !          3456: \global\let\contentsalignmacro = \chapoddpage
        !          3457: }
        !          3458: 
        !          3459: \def\HEADINGSsingleafter{\let\HEADINGShook=\HEADINGSsinglex}
        !          3460: \def\HEADINGSsinglex{%
        !          3461: \global\evenfootline={\hfil}
        !          3462: \global\oddfootline={\hfil}
        !          3463: \global\evenheadline={\line{\thischapter\hfil\folio}}
        !          3464: \global\oddheadline={\line{\thischapter\hfil\folio}}
        !          3465: \global\let\contentsalignmacro = \chappager
        !          3466: }
        !          3467: 
        !          3468: % Subroutines used in generating headings
        !          3469: % This produces Day Month Year style of output.
        !          3470: % Only define if not already defined, in case a txi-??.tex file has set
        !          3471: % up a different format (e.g., txi-cs.tex does this).
        !          3472: \ifx\today\thisisundefined
        !          3473: \def\today{%
        !          3474:   \number\day\space
        !          3475:   \ifcase\month
        !          3476:   \or\putwordMJan\or\putwordMFeb\or\putwordMMar\or\putwordMApr
        !          3477:   \or\putwordMMay\or\putwordMJun\or\putwordMJul\or\putwordMAug
        !          3478:   \or\putwordMSep\or\putwordMOct\or\putwordMNov\or\putwordMDec
        !          3479:   \fi
        !          3480:   \space\number\year}
        !          3481: \fi
        !          3482: 
        !          3483: % @settitle line...  specifies the title of the document, for headings.
        !          3484: % It generates no output of its own.
        !          3485: \def\thistitle{\putwordNoTitle}
        !          3486: \def\settitle{\parsearg{\gdef\thistitle}}
        !          3487: 
        !          3488: 
        !          3489: \message{tables,}
        !          3490: % Tables -- @table, @ftable, @vtable, @item(x).
        !          3491: 
        !          3492: % default indentation of table text
        !          3493: \newdimen\tableindent \tableindent=.8in
        !          3494: % default indentation of @itemize and @enumerate text
        !          3495: \newdimen\itemindent  \itemindent=.3in
        !          3496: % margin between end of table item and start of table text.
        !          3497: \newdimen\itemmargin  \itemmargin=.1in
        !          3498: 
        !          3499: % used internally for \itemindent minus \itemmargin
        !          3500: \newdimen\itemmax
        !          3501: 
        !          3502: % Note @table, @ftable, and @vtable define @item, @itemx, etc., with
        !          3503: % these defs.
        !          3504: % They also define \itemindex
        !          3505: % to index the item name in whatever manner is desired (perhaps none).
        !          3506: 
        !          3507: \newif\ifitemxneedsnegativevskip
        !          3508: 
        !          3509: \def\itemxpar{\par\ifitemxneedsnegativevskip\nobreak\vskip-\parskip\nobreak\fi}
        !          3510: 
        !          3511: \def\internalBitem{\smallbreak \parsearg\itemzzz}
        !          3512: \def\internalBitemx{\itemxpar \parsearg\itemzzz}
        !          3513: 
        !          3514: \def\itemzzz #1{\begingroup %
        !          3515:   \advance\hsize by -\rightskip
        !          3516:   \advance\hsize by -\tableindent
        !          3517:   \setbox0=\hbox{\itemindicate{#1}}%
        !          3518:   \itemindex{#1}%
        !          3519:   \nobreak % This prevents a break before @itemx.
        !          3520:   %
        !          3521:   % If the item text does not fit in the space we have, put it on a line
        !          3522:   % by itself, and do not allow a page break either before or after that
        !          3523:   % line.  We do not start a paragraph here because then if the next
        !          3524:   % command is, e.g., @kindex, the whatsit would get put into the
        !          3525:   % horizontal list on a line by itself, resulting in extra blank space.
        !          3526:   \ifdim \wd0>\itemmax
        !          3527:     %
        !          3528:     % Make this a paragraph so we get the \parskip glue and wrapping,
        !          3529:     % but leave it ragged-right.
        !          3530:     \begingroup
        !          3531:       \advance\leftskip by-\tableindent
        !          3532:       \advance\hsize by\tableindent
        !          3533:       \advance\rightskip by0pt plus1fil\relax
        !          3534:       \leavevmode\unhbox0\par
        !          3535:     \endgroup
        !          3536:     %
        !          3537:     % We're going to be starting a paragraph, but we don't want the
        !          3538:     % \parskip glue -- logically it's part of the @item we just started.
        !          3539:     \nobreak \vskip-\parskip
        !          3540:     %
        !          3541:     % Stop a page break at the \parskip glue coming up.  However, if
        !          3542:     % what follows is an environment such as @example, there will be no
        !          3543:     % \parskip glue; then the negative vskip we just inserted would
        !          3544:     % cause the example and the item to crash together.  So we use this
        !          3545:     % bizarre value of 10001 as a signal to \aboveenvbreak to insert
        !          3546:     % \parskip glue after all.  Section titles are handled this way also.
        !          3547:     %
        !          3548:     \penalty 10001
        !          3549:     \endgroup
        !          3550:     \itemxneedsnegativevskipfalse
        !          3551:   \else
        !          3552:     % The item text fits into the space.  Start a paragraph, so that the
        !          3553:     % following text (if any) will end up on the same line.
        !          3554:     \noindent
        !          3555:     % Do this with kerns and \unhbox so that if there is a footnote in
        !          3556:     % the item text, it can migrate to the main vertical list and
        !          3557:     % eventually be printed.
        !          3558:     \nobreak\kern-\tableindent
        !          3559:     \dimen0 = \itemmax  \advance\dimen0 by \itemmargin \advance\dimen0 by -\wd0
        !          3560:     \unhbox0
        !          3561:     \nobreak\kern\dimen0
        !          3562:     \endgroup
        !          3563:     \itemxneedsnegativevskiptrue
        !          3564:   \fi
        !          3565: }
        !          3566: 
        !          3567: \def\item{\errmessage{@item while not in a list environment}}
        !          3568: \def\itemx{\errmessage{@itemx while not in a list environment}}
        !          3569: 
        !          3570: % @table, @ftable, @vtable.
        !          3571: \envdef\table{%
        !          3572:   \let\itemindex\gobble
        !          3573:   \tablecheck{table}%
        !          3574: }
        !          3575: \envdef\ftable{%
        !          3576:   \def\itemindex ##1{\doind {fn}{\code{##1}}}%
        !          3577:   \tablecheck{ftable}%
        !          3578: }
        !          3579: \envdef\vtable{%
        !          3580:   \def\itemindex ##1{\doind {vr}{\code{##1}}}%
        !          3581:   \tablecheck{vtable}%
        !          3582: }
        !          3583: \def\tablecheck#1{%
        !          3584:   \ifnum \the\catcode`\^^M=\active
        !          3585:     \endgroup
        !          3586:     \errmessage{This command won't work in this context; perhaps the problem is
        !          3587:       that we are \inenvironment\thisenv}%
        !          3588:     \def\next{\doignore{#1}}%
        !          3589:   \else
        !          3590:     \let\next\tablex
        !          3591:   \fi
        !          3592:   \next
        !          3593: }
        !          3594: \def\tablex#1{%
        !          3595:   \def\itemindicate{#1}%
        !          3596:   \parsearg\tabley
        !          3597: }
        !          3598: \def\tabley#1{%
        !          3599:   {%
        !          3600:     \makevalueexpandable
        !          3601:     \edef\temp{\noexpand\tablez #1\space\space\space}%
        !          3602:     \expandafter
        !          3603:   }\temp \endtablez
        !          3604: }
        !          3605: \def\tablez #1 #2 #3 #4\endtablez{%
        !          3606:   \aboveenvbreak
        !          3607:   \ifnum 0#1>0 \advance \leftskip by #1\mil \fi
        !          3608:   \ifnum 0#2>0 \tableindent=#2\mil \fi
        !          3609:   \ifnum 0#3>0 \advance \rightskip by #3\mil \fi
        !          3610:   \itemmax=\tableindent
        !          3611:   \advance \itemmax by -\itemmargin
        !          3612:   \advance \leftskip by \tableindent
        !          3613:   \exdentamount=\tableindent
        !          3614:   \parindent = 0pt
        !          3615:   \parskip = \smallskipamount
        !          3616:   \ifdim \parskip=0pt \parskip=2pt \fi
        !          3617:   \let\item = \internalBitem
        !          3618:   \let\itemx = \internalBitemx
        !          3619: }
        !          3620: \def\Etable{\endgraf\afterenvbreak}
        !          3621: \let\Eftable\Etable
        !          3622: \let\Evtable\Etable
        !          3623: \let\Eitemize\Etable
        !          3624: \let\Eenumerate\Etable
        !          3625: 
        !          3626: % This is the counter used by @enumerate, which is really @itemize
        !          3627: 
        !          3628: \newcount \itemno
        !          3629: 
        !          3630: \envdef\itemize{\parsearg\doitemize}
        !          3631: 
        !          3632: \def\doitemize#1{%
        !          3633:   \aboveenvbreak
        !          3634:   \itemmax=\itemindent
        !          3635:   \advance\itemmax by -\itemmargin
        !          3636:   \advance\leftskip by \itemindent
        !          3637:   \exdentamount=\itemindent
        !          3638:   \parindent=0pt
        !          3639:   \parskip=\smallskipamount
        !          3640:   \ifdim\parskip=0pt \parskip=2pt \fi
        !          3641:   %
        !          3642:   % Try typesetting the item mark that if the document erroneously says
        !          3643:   % something like @itemize @samp (intending @table), there's an error
        !          3644:   % right away at the @itemize.  It's not the best error message in the
        !          3645:   % world, but it's better than leaving it to the @item.  This means if
        !          3646:   % the user wants an empty mark, they have to say @w{} not just @w.
        !          3647:   \def\itemcontents{#1}%
        !          3648:   \setbox0 = \hbox{\itemcontents}%
        !          3649:   %
        !          3650:   % @itemize with no arg is equivalent to @itemize @bullet.
        !          3651:   \ifx\itemcontents\empty\def\itemcontents{\bullet}\fi
        !          3652:   %
        !          3653:   \let\item=\itemizeitem
        !          3654: }
        !          3655: 
        !          3656: % Definition of @item while inside @itemize and @enumerate.
        !          3657: %
        !          3658: \def\itemizeitem{%
        !          3659:   \advance\itemno by 1  % for enumerations
        !          3660:   {\let\par=\endgraf \smallbreak}% reasonable place to break
        !          3661:   {%
        !          3662:    % If the document has an @itemize directly after a section title, a
        !          3663:    % \nobreak will be last on the list, and \sectionheading will have
        !          3664:    % done a \vskip-\parskip.  In that case, we don't want to zero
        !          3665:    % parskip, or the item text will crash with the heading.  On the
        !          3666:    % other hand, when there is normal text preceding the item (as there
        !          3667:    % usually is), we do want to zero parskip, or there would be too much
        !          3668:    % space.  In that case, we won't have a \nobreak before.  At least
        !          3669:    % that's the theory.
        !          3670:    \ifnum\lastpenalty<10000 \parskip=0in \fi
        !          3671:    \noindent
        !          3672:    \hbox to 0pt{\hss \itemcontents \kern\itemmargin}%
        !          3673:    %
        !          3674:    \vadjust{\penalty 1200}}% not good to break after first line of item.
        !          3675:   \flushcr
        !          3676: }
        !          3677: 
        !          3678: % \splitoff TOKENS\endmark defines \first to be the first token in
        !          3679: % TOKENS, and \rest to be the remainder.
        !          3680: %
        !          3681: \def\splitoff#1#2\endmark{\def\first{#1}\def\rest{#2}}%
        !          3682: 
        !          3683: % Allow an optional argument of an uppercase letter, lowercase letter,
        !          3684: % or number, to specify the first label in the enumerated list.  No
        !          3685: % argument is the same as `1'.
        !          3686: %
        !          3687: \envparseargdef\enumerate{\enumeratey #1  \endenumeratey}
        !          3688: \def\enumeratey #1 #2\endenumeratey{%
        !          3689:   % If we were given no argument, pretend we were given `1'.
        !          3690:   \def\thearg{#1}%
        !          3691:   \ifx\thearg\empty \def\thearg{1}\fi
        !          3692:   %
        !          3693:   % Detect if the argument is a single token.  If so, it might be a
        !          3694:   % letter.  Otherwise, the only valid thing it can be is a number.
        !          3695:   % (We will always have one token, because of the test we just made.
        !          3696:   % This is a good thing, since \splitoff doesn't work given nothing at
        !          3697:   % all -- the first parameter is undelimited.)
        !          3698:   \expandafter\splitoff\thearg\endmark
        !          3699:   \ifx\rest\empty
        !          3700:     % Only one token in the argument.  It could still be anything.
        !          3701:     % A ``lowercase letter'' is one whose \lccode is nonzero.
        !          3702:     % An ``uppercase letter'' is one whose \lccode is both nonzero, and
        !          3703:     %   not equal to itself.
        !          3704:     % Otherwise, we assume it's a number.
        !          3705:     %
        !          3706:     % We need the \relax at the end of the \ifnum lines to stop TeX from
        !          3707:     % continuing to look for a <number>.
        !          3708:     %
        !          3709:     \ifnum\lccode\expandafter`\thearg=0\relax
        !          3710:       \numericenumerate % a number (we hope)
        !          3711:     \else
        !          3712:       % It's a letter.
        !          3713:       \ifnum\lccode\expandafter`\thearg=\expandafter`\thearg\relax
        !          3714:         \lowercaseenumerate % lowercase letter
        !          3715:       \else
        !          3716:         \uppercaseenumerate % uppercase letter
        !          3717:       \fi
        !          3718:     \fi
        !          3719:   \else
        !          3720:     % Multiple tokens in the argument.  We hope it's a number.
        !          3721:     \numericenumerate
        !          3722:   \fi
        !          3723: }
        !          3724: 
        !          3725: % An @enumerate whose labels are integers.  The starting integer is
        !          3726: % given in \thearg.
        !          3727: %
        !          3728: \def\numericenumerate{%
        !          3729:   \itemno = \thearg
        !          3730:   \startenumeration{\the\itemno}%
        !          3731: }
        !          3732: 
        !          3733: % The starting (lowercase) letter is in \thearg.
        !          3734: \def\lowercaseenumerate{%
        !          3735:   \itemno = \expandafter`\thearg
        !          3736:   \startenumeration{%
        !          3737:     % Be sure we're not beyond the end of the alphabet.
        !          3738:     \ifnum\itemno=0
        !          3739:       \errmessage{No more lowercase letters in @enumerate; get a bigger
        !          3740:                   alphabet}%
        !          3741:     \fi
        !          3742:     \char\lccode\itemno
        !          3743:   }%
        !          3744: }
        !          3745: 
        !          3746: % The starting (uppercase) letter is in \thearg.
        !          3747: \def\uppercaseenumerate{%
        !          3748:   \itemno = \expandafter`\thearg
        !          3749:   \startenumeration{%
        !          3750:     % Be sure we're not beyond the end of the alphabet.
        !          3751:     \ifnum\itemno=0
        !          3752:       \errmessage{No more uppercase letters in @enumerate; get a bigger
        !          3753:                   alphabet}
        !          3754:     \fi
        !          3755:     \char\uccode\itemno
        !          3756:   }%
        !          3757: }
        !          3758: 
        !          3759: % Call \doitemize, adding a period to the first argument and supplying the
        !          3760: % common last two arguments.  Also subtract one from the initial value in
        !          3761: % \itemno, since @item increments \itemno.
        !          3762: %
        !          3763: \def\startenumeration#1{%
        !          3764:   \advance\itemno by -1
        !          3765:   \doitemize{#1.}\flushcr
        !          3766: }
        !          3767: 
        !          3768: % @alphaenumerate and @capsenumerate are abbreviations for giving an arg
        !          3769: % to @enumerate.
        !          3770: %
        !          3771: \def\alphaenumerate{\enumerate{a}}
        !          3772: \def\capsenumerate{\enumerate{A}}
        !          3773: \def\Ealphaenumerate{\Eenumerate}
        !          3774: \def\Ecapsenumerate{\Eenumerate}
        !          3775: 
        !          3776: 
        !          3777: % @multitable macros
        !          3778: % Amy Hendrickson, 8/18/94, 3/6/96
        !          3779: %
        !          3780: % @multitable ... @end multitable will make as many columns as desired.
        !          3781: % Contents of each column will wrap at width given in preamble.  Width
        !          3782: % can be specified either with sample text given in a template line,
        !          3783: % or in percent of \hsize, the current width of text on page.
        !          3784: 
        !          3785: % Table can continue over pages but will only break between lines.
        !          3786: 
        !          3787: % To make preamble:
        !          3788: %
        !          3789: % Either define widths of columns in terms of percent of \hsize:
        !          3790: %   @multitable @columnfractions .25 .3 .45
        !          3791: %   @item ...
        !          3792: %
        !          3793: %   Numbers following @columnfractions are the percent of the total
        !          3794: %   current hsize to be used for each column. You may use as many
        !          3795: %   columns as desired.
        !          3796: 
        !          3797: 
        !          3798: % Or use a template:
        !          3799: %   @multitable {Column 1 template} {Column 2 template} {Column 3 template}
        !          3800: %   @item ...
        !          3801: %   using the widest term desired in each column.
        !          3802: 
        !          3803: % Each new table line starts with @item, each subsequent new column
        !          3804: % starts with @tab. Empty columns may be produced by supplying @tab's
        !          3805: % with nothing between them for as many times as empty columns are needed,
        !          3806: % ie, @tab@tab@tab will produce two empty columns.
        !          3807: 
        !          3808: % @item, @tab do not need to be on their own lines, but it will not hurt
        !          3809: % if they are.
        !          3810: 
        !          3811: % Sample multitable:
        !          3812: 
        !          3813: %   @multitable {Column 1 template} {Column 2 template} {Column 3 template}
        !          3814: %   @item first col stuff @tab second col stuff @tab third col
        !          3815: %   @item
        !          3816: %   first col stuff
        !          3817: %   @tab
        !          3818: %   second col stuff
        !          3819: %   @tab
        !          3820: %   third col
        !          3821: %   @item first col stuff @tab second col stuff
        !          3822: %   @tab Many paragraphs of text may be used in any column.
        !          3823: %
        !          3824: %         They will wrap at the width determined by the template.
        !          3825: %   @item@tab@tab This will be in third column.
        !          3826: %   @end multitable
        !          3827: 
        !          3828: % Default dimensions may be reset by user.
        !          3829: % @multitableparskip is vertical space between paragraphs in table.
        !          3830: % @multitableparindent is paragraph indent in table.
        !          3831: % @multitablecolmargin is horizontal space to be left between columns.
        !          3832: % @multitablelinespace is space to leave between table items, baseline
        !          3833: %                                                            to baseline.
        !          3834: %   0pt means it depends on current normal line spacing.
        !          3835: %
        !          3836: \newskip\multitableparskip
        !          3837: \newskip\multitableparindent
        !          3838: \newdimen\multitablecolspace
        !          3839: \newskip\multitablelinespace
        !          3840: \multitableparskip=0pt
        !          3841: \multitableparindent=6pt
        !          3842: \multitablecolspace=12pt
        !          3843: \multitablelinespace=0pt
        !          3844: 
        !          3845: % Macros used to set up halign preamble:
        !          3846: %
        !          3847: \let\endsetuptable\relax
        !          3848: \def\xendsetuptable{\endsetuptable}
        !          3849: \let\columnfractions\relax
        !          3850: \def\xcolumnfractions{\columnfractions}
        !          3851: \newif\ifsetpercent
        !          3852: 
        !          3853: % #1 is the @columnfraction, usually a decimal number like .5, but might
        !          3854: % be just 1.  We just use it, whatever it is.
        !          3855: %
        !          3856: \def\pickupwholefraction#1 {%
        !          3857:   \global\advance\colcount by 1
        !          3858:   \expandafter\xdef\csname col\the\colcount\endcsname{#1\hsize}%
        !          3859:   \setuptable
        !          3860: }
        !          3861: 
        !          3862: \newcount\colcount
        !          3863: \def\setuptable#1{%
        !          3864:   \def\firstarg{#1}%
        !          3865:   \ifx\firstarg\xendsetuptable
        !          3866:     \let\go = \relax
        !          3867:   \else
        !          3868:     \ifx\firstarg\xcolumnfractions
        !          3869:       \global\setpercenttrue
        !          3870:     \else
        !          3871:       \ifsetpercent
        !          3872:          \let\go\pickupwholefraction
        !          3873:       \else
        !          3874:          \global\advance\colcount by 1
        !          3875:          \setbox0=\hbox{#1\unskip\space}% Add a normal word space as a
        !          3876:                    % separator; typically that is always in the input, anyway.
        !          3877:          \expandafter\xdef\csname col\the\colcount\endcsname{\the\wd0}%
        !          3878:       \fi
        !          3879:     \fi
        !          3880:     \ifx\go\pickupwholefraction
        !          3881:       % Put the argument back for the \pickupwholefraction call, so
        !          3882:       % we'll always have a period there to be parsed.
        !          3883:       \def\go{\pickupwholefraction#1}%
        !          3884:     \else
        !          3885:       \let\go = \setuptable
        !          3886:     \fi%
        !          3887:   \fi
        !          3888:   \go
        !          3889: }
        !          3890: 
        !          3891: % multitable-only commands.
        !          3892: %
        !          3893: % @headitem starts a heading row, which we typeset in bold.
        !          3894: % Assignments have to be global since we are inside the implicit group
        !          3895: % of an alignment entry.  \everycr resets \everytab so we don't have to
        !          3896: % undo it ourselves.
        !          3897: \def\headitemfont{\b}% for people to use in the template row; not changeable
        !          3898: \def\headitem{%
        !          3899:   \checkenv\multitable
        !          3900:   \crcr
        !          3901:   \global\everytab={\bf}% can't use \headitemfont since the parsing differs
        !          3902:   \the\everytab % for the first item
        !          3903: }%
        !          3904: %
        !          3905: % A \tab used to include \hskip1sp.  But then the space in a template
        !          3906: % line is not enough.  That is bad.  So let's go back to just `&' until
        !          3907: % we again encounter the problem the 1sp was intended to solve.
        !          3908: %                                      --karl, [email protected], 20apr99.
        !          3909: \def\tab{\checkenv\multitable &\the\everytab}%
        !          3910: 
        !          3911: % @multitable ... @end multitable definitions:
        !          3912: %
        !          3913: \newtoks\everytab  % insert after every tab.
        !          3914: %
        !          3915: \envdef\multitable{%
        !          3916:   \vskip\parskip
        !          3917:   \startsavinginserts
        !          3918:   %
        !          3919:   % @item within a multitable starts a normal row.
        !          3920:   % We use \def instead of \let so that if one of the multitable entries
        !          3921:   % contains an @itemize, we don't choke on the \item (seen as \crcr aka
        !          3922:   % \endtemplate) expanding \doitemize.
        !          3923:   \def\item{\crcr}%
        !          3924:   %
        !          3925:   \tolerance=9500
        !          3926:   \hbadness=9500
        !          3927:   \setmultitablespacing
        !          3928:   \parskip=\multitableparskip
        !          3929:   \parindent=\multitableparindent
        !          3930:   \overfullrule=0pt
        !          3931:   \global\colcount=0
        !          3932:   %
        !          3933:   \everycr = {%
        !          3934:     \noalign{%
        !          3935:       \global\everytab={}%
        !          3936:       \global\colcount=0 % Reset the column counter.
        !          3937:       % Check for saved footnotes, etc.
        !          3938:       \checkinserts
        !          3939:       % Keeps underfull box messages off when table breaks over pages.
        !          3940:       %\filbreak
        !          3941:        % Maybe so, but it also creates really weird page breaks when the
        !          3942:        % table breaks over pages. Wouldn't \vfil be better?  Wait until the
        !          3943:        % problem manifests itself, so it can be fixed for real --karl.
        !          3944:     }%
        !          3945:   }%
        !          3946:   %
        !          3947:   \parsearg\domultitable
        !          3948: }
        !          3949: \def\domultitable#1{%
        !          3950:   % To parse everything between @multitable and @item:
        !          3951:   \setuptable#1 \endsetuptable
        !          3952:   %
        !          3953:   % This preamble sets up a generic column definition, which will
        !          3954:   % be used as many times as user calls for columns.
        !          3955:   % \vtop will set a single line and will also let text wrap and
        !          3956:   % continue for many paragraphs if desired.
        !          3957:   \halign\bgroup &%
        !          3958:     \global\advance\colcount by 1
        !          3959:     \multistrut
        !          3960:     \vtop{%
        !          3961:       % Use the current \colcount to find the correct column width:
        !          3962:       \hsize=\expandafter\csname col\the\colcount\endcsname
        !          3963:       %
        !          3964:       % In order to keep entries from bumping into each other
        !          3965:       % we will add a \leftskip of \multitablecolspace to all columns after
        !          3966:       % the first one.
        !          3967:       %
        !          3968:       % If a template has been used, we will add \multitablecolspace
        !          3969:       % to the width of each template entry.
        !          3970:       %
        !          3971:       % If the user has set preamble in terms of percent of \hsize we will
        !          3972:       % use that dimension as the width of the column, and the \leftskip
        !          3973:       % will keep entries from bumping into each other.  Table will start at
        !          3974:       % left margin and final column will justify at right margin.
        !          3975:       %
        !          3976:       % Make sure we don't inherit \rightskip from the outer environment.
        !          3977:       \rightskip=0pt
        !          3978:       \ifnum\colcount=1
        !          3979:        % The first column will be indented with the surrounding text.
        !          3980:        \advance\hsize by\leftskip
        !          3981:       \else
        !          3982:        \ifsetpercent \else
        !          3983:          % If user has not set preamble in terms of percent of \hsize
        !          3984:          % we will advance \hsize by \multitablecolspace.
        !          3985:          \advance\hsize by \multitablecolspace
        !          3986:        \fi
        !          3987:        % In either case we will make \leftskip=\multitablecolspace:
        !          3988:       \leftskip=\multitablecolspace
        !          3989:       \fi
        !          3990:       % Ignoring space at the beginning and end avoids an occasional spurious
        !          3991:       % blank line, when TeX decides to break the line at the space before the
        !          3992:       % box from the multistrut, so the strut ends up on a line by itself.
        !          3993:       % For example:
        !          3994:       % @multitable @columnfractions .11 .89
        !          3995:       % @item @code{#}
        !          3996:       % @tab Legal holiday which is valid in major parts of the whole country.
        !          3997:       % Is automatically provided with highlighting sequences respectively
        !          3998:       % marking characters.
        !          3999:       \noindent\ignorespaces##\unskip\multistrut
        !          4000:     }\cr
        !          4001: }
        !          4002: \def\Emultitable{%
        !          4003:   \crcr
        !          4004:   \egroup % end the \halign
        !          4005:   \global\setpercentfalse
        !          4006: }
        !          4007: 
        !          4008: \def\setmultitablespacing{%
        !          4009:   \def\multistrut{\strut}% just use the standard line spacing
        !          4010:   %
        !          4011:   % Compute \multitablelinespace (if not defined by user) for use in
        !          4012:   % \multitableparskip calculation.  We used define \multistrut based on
        !          4013:   % this, but (ironically) that caused the spacing to be off.
        !          4014:   % See bug-texinfo report from Werner Lemberg, 31 Oct 2004 12:52:20 +0100.
        !          4015: \ifdim\multitablelinespace=0pt
        !          4016: \setbox0=\vbox{X}\global\multitablelinespace=\the\baselineskip
        !          4017: \global\advance\multitablelinespace by-\ht0
        !          4018: \fi
        !          4019: % Test to see if parskip is larger than space between lines of
        !          4020: % table. If not, do nothing.
        !          4021: %        If so, set to same dimension as multitablelinespace.
        !          4022: \ifdim\multitableparskip>\multitablelinespace
        !          4023: \global\multitableparskip=\multitablelinespace
        !          4024: \global\advance\multitableparskip-7pt % to keep parskip somewhat smaller
        !          4025:                                       % than skip between lines in the table.
        !          4026: \fi%
        !          4027: \ifdim\multitableparskip=0pt
        !          4028: \global\multitableparskip=\multitablelinespace
        !          4029: \global\advance\multitableparskip-7pt % to keep parskip somewhat smaller
        !          4030:                                       % than skip between lines in the table.
        !          4031: \fi}
        !          4032: 
        !          4033: 
        !          4034: \message{conditionals,}
        !          4035: 
        !          4036: % @iftex, @ifnotdocbook, @ifnothtml, @ifnotinfo, @ifnotplaintext,
        !          4037: % @ifnotxml always succeed.  They currently do nothing; we don't
        !          4038: % attempt to check whether the conditionals are properly nested.  But we
        !          4039: % have to remember that they are conditionals, so that @end doesn't
        !          4040: % attempt to close an environment group.
        !          4041: %
        !          4042: \def\makecond#1{%
        !          4043:   \expandafter\let\csname #1\endcsname = \relax
        !          4044:   \expandafter\let\csname iscond.#1\endcsname = 1
        !          4045: }
        !          4046: \makecond{iftex}
        !          4047: \makecond{ifnotdocbook}
        !          4048: \makecond{ifnothtml}
        !          4049: \makecond{ifnotinfo}
        !          4050: \makecond{ifnotplaintext}
        !          4051: \makecond{ifnotxml}
        !          4052: 
        !          4053: % Ignore @ignore, @ifhtml, @ifinfo, and the like.
        !          4054: %
        !          4055: \def\direntry{\doignore{direntry}}
        !          4056: \def\documentdescription{\doignore{documentdescription}}
        !          4057: \def\docbook{\doignore{docbook}}
        !          4058: \def\html{\doignore{html}}
        !          4059: \def\ifdocbook{\doignore{ifdocbook}}
        !          4060: \def\ifhtml{\doignore{ifhtml}}
        !          4061: \def\ifinfo{\doignore{ifinfo}}
        !          4062: \def\ifnottex{\doignore{ifnottex}}
        !          4063: \def\ifplaintext{\doignore{ifplaintext}}
        !          4064: \def\ifxml{\doignore{ifxml}}
        !          4065: \def\ignore{\doignore{ignore}}
        !          4066: \def\menu{\doignore{menu}}
        !          4067: \def\xml{\doignore{xml}}
        !          4068: 
        !          4069: % Ignore text until a line `@end #1', keeping track of nested conditionals.
        !          4070: %
        !          4071: % A count to remember the depth of nesting.
        !          4072: \newcount\doignorecount
        !          4073: 
        !          4074: \def\doignore#1{\begingroup
        !          4075:   % Scan in ``verbatim'' mode:
        !          4076:   \obeylines
        !          4077:   \catcode`\@ = \other
        !          4078:   \catcode`\{ = \other
        !          4079:   \catcode`\} = \other
        !          4080:   %
        !          4081:   % Make sure that spaces turn into tokens that match what \doignoretext wants.
        !          4082:   \spaceisspace
        !          4083:   %
        !          4084:   % Count number of #1's that we've seen.
        !          4085:   \doignorecount = 0
        !          4086:   %
        !          4087:   % Swallow text until we reach the matching `@end #1'.
        !          4088:   \dodoignore{#1}%
        !          4089: }
        !          4090: 
        !          4091: { \catcode`_=11 % We want to use \_STOP_ which cannot appear in texinfo source.
        !          4092:   \obeylines %
        !          4093:   %
        !          4094:   \gdef\dodoignore#1{%
        !          4095:     % #1 contains the command name as a string, e.g., `ifinfo'.
        !          4096:     %
        !          4097:     % Define a command to find the next `@end #1'.
        !          4098:     \long\def\doignoretext##1^^M@end #1{%
        !          4099:       \doignoretextyyy##1^^M@#1\_STOP_}%
        !          4100:     %
        !          4101:     % And this command to find another #1 command, at the beginning of a
        !          4102:     % line.  (Otherwise, we would consider a line `@c @ifset', for
        !          4103:     % example, to count as an @ifset for nesting.)
        !          4104:     \long\def\doignoretextyyy##1^^M@#1##2\_STOP_{\doignoreyyy{##2}\_STOP_}%
        !          4105:     %
        !          4106:     % And now expand that command.
        !          4107:     \doignoretext ^^M%
        !          4108:   }%
        !          4109: }
        !          4110: 
        !          4111: \def\doignoreyyy#1{%
        !          4112:   \def\temp{#1}%
        !          4113:   \ifx\temp\empty                      % Nothing found.
        !          4114:     \let\next\doignoretextzzz
        !          4115:   \else                                        % Found a nested condition, ...
        !          4116:     \advance\doignorecount by 1
        !          4117:     \let\next\doignoretextyyy          % ..., look for another.
        !          4118:     % If we're here, #1 ends with ^^M\ifinfo (for example).
        !          4119:   \fi
        !          4120:   \next #1% the token \_STOP_ is present just after this macro.
        !          4121: }
        !          4122: 
        !          4123: % We have to swallow the remaining "\_STOP_".
        !          4124: %
        !          4125: \def\doignoretextzzz#1{%
        !          4126:   \ifnum\doignorecount = 0     % We have just found the outermost @end.
        !          4127:     \let\next\enddoignore
        !          4128:   \else                                % Still inside a nested condition.
        !          4129:     \advance\doignorecount by -1
        !          4130:     \let\next\doignoretext      % Look for the next @end.
        !          4131:   \fi
        !          4132:   \next
        !          4133: }
        !          4134: 
        !          4135: % Finish off ignored text.
        !          4136: { \obeylines%
        !          4137:   % Ignore anything after the last `@end #1'; this matters in verbatim
        !          4138:   % environments, where otherwise the newline after an ignored conditional
        !          4139:   % would result in a blank line in the output.
        !          4140:   \gdef\enddoignore#1^^M{\endgroup\ignorespaces}%
        !          4141: }
        !          4142: 
        !          4143: 
        !          4144: % @set VAR sets the variable VAR to an empty value.
        !          4145: % @set VAR REST-OF-LINE sets VAR to the value REST-OF-LINE.
        !          4146: %
        !          4147: % Since we want to separate VAR from REST-OF-LINE (which might be
        !          4148: % empty), we can't just use \parsearg; we have to insert a space of our
        !          4149: % own to delimit the rest of the line, and then take it out again if we
        !          4150: % didn't need it.
        !          4151: % We rely on the fact that \parsearg sets \catcode`\ =10.
        !          4152: %
        !          4153: \parseargdef\set{\setyyy#1 \endsetyyy}
        !          4154: \def\setyyy#1 #2\endsetyyy{%
        !          4155:   {%
        !          4156:     \makevalueexpandable
        !          4157:     \def\temp{#2}%
        !          4158:     \edef\next{\gdef\makecsname{SET#1}}%
        !          4159:     \ifx\temp\empty
        !          4160:       \next{}%
        !          4161:     \else
        !          4162:       \setzzz#2\endsetzzz
        !          4163:     \fi
        !          4164:   }%
        !          4165: }
        !          4166: % Remove the trailing space \setxxx inserted.
        !          4167: \def\setzzz#1 \endsetzzz{\next{#1}}
        !          4168: 
        !          4169: % @clear VAR clears (i.e., unsets) the variable VAR.
        !          4170: %
        !          4171: \parseargdef\clear{%
        !          4172:   {%
        !          4173:     \makevalueexpandable
        !          4174:     \global\expandafter\let\csname SET#1\endcsname=\relax
        !          4175:   }%
        !          4176: }
        !          4177: 
        !          4178: % @value{foo} gets the text saved in variable foo.
        !          4179: \def\value{\begingroup\makevalueexpandable\valuexxx}
        !          4180: \def\valuexxx#1{\expandablevalue{#1}\endgroup}
        !          4181: {
        !          4182:   \catcode`\- = \active \catcode`\_ = \active
        !          4183:   %
        !          4184:   \gdef\makevalueexpandable{%
        !          4185:     \let\value = \expandablevalue
        !          4186:     % We don't want these characters active, ...
        !          4187:     \catcode`\-=\other \catcode`\_=\other
        !          4188:     % ..., but we might end up with active ones in the argument if
        !          4189:     % we're called from @code, as @code{@value{foo-bar_}}, though.
        !          4190:     % So \let them to their normal equivalents.
        !          4191:     \let-\normaldash \let_\normalunderscore
        !          4192:   }
        !          4193: }
        !          4194: 
        !          4195: % We have this subroutine so that we can handle at least some @value's
        !          4196: % properly in indexes (we call \makevalueexpandable in \indexdummies).
        !          4197: % The command has to be fully expandable (if the variable is set), since
        !          4198: % the result winds up in the index file.  This means that if the
        !          4199: % variable's value contains other Texinfo commands, it's almost certain
        !          4200: % it will fail (although perhaps we could fix that with sufficient work
        !          4201: % to do a one-level expansion on the result, instead of complete).
        !          4202: %
        !          4203: \def\expandablevalue#1{%
        !          4204:   \expandafter\ifx\csname SET#1\endcsname\relax
        !          4205:     {[No value for ``#1'']}%
        !          4206:     \message{Variable `#1', used in @value, is not set.}%
        !          4207:   \else
        !          4208:     \csname SET#1\endcsname
        !          4209:   \fi
        !          4210: }
        !          4211: 
        !          4212: % @ifset VAR ... @end ifset reads the `...' iff VAR has been defined
        !          4213: % with @set.
        !          4214: %
        !          4215: % To get special treatment of `@end ifset,' call \makeond and the redefine.
        !          4216: %
        !          4217: \makecond{ifset}
        !          4218: \def\ifset{\parsearg{\doifset{\let\next=\ifsetfail}}}
        !          4219: \def\doifset#1#2{%
        !          4220:   {%
        !          4221:     \makevalueexpandable
        !          4222:     \let\next=\empty
        !          4223:     \expandafter\ifx\csname SET#2\endcsname\relax
        !          4224:       #1% If not set, redefine \next.
        !          4225:     \fi
        !          4226:     \expandafter
        !          4227:   }\next
        !          4228: }
        !          4229: \def\ifsetfail{\doignore{ifset}}
        !          4230: 
        !          4231: % @ifclear VAR ... @end executes the `...' iff VAR has never been
        !          4232: % defined with @set, or has been undefined with @clear.
        !          4233: %
        !          4234: % The `\else' inside the `\doifset' parameter is a trick to reuse the
        !          4235: % above code: if the variable is not set, do nothing, if it is set,
        !          4236: % then redefine \next to \ifclearfail.
        !          4237: %
        !          4238: \makecond{ifclear}
        !          4239: \def\ifclear{\parsearg{\doifset{\else \let\next=\ifclearfail}}}
        !          4240: \def\ifclearfail{\doignore{ifclear}}
        !          4241: 
        !          4242: % @ifcommandisdefined CMD ... @end executes the `...' if CMD (written
        !          4243: % without the @) is in fact defined.  We can only feasibly check at the
        !          4244: % TeX level, so something like `mathcode' is going to considered
        !          4245: % defined even though it is not a Texinfo command.
        !          4246: % 
        !          4247: \makecond{ifcommanddefined}
        !          4248: \def\ifcommanddefined{\parsearg{\doifcmddefined{\let\next=\ifcmddefinedfail}}}
        !          4249: %
        !          4250: \def\doifcmddefined#1#2{{%
        !          4251:     \makevalueexpandable
        !          4252:     \let\next=\empty
        !          4253:     \expandafter\ifx\csname #2\endcsname\relax
        !          4254:       #1% If not defined, \let\next as above.
        !          4255:     \fi
        !          4256:     \expandafter
        !          4257:   }\next
        !          4258: }
        !          4259: \def\ifcmddefinedfail{\doignore{ifcommanddefined}}
        !          4260: 
        !          4261: % @ifcommandnotdefined CMD ... handled similar to @ifclear above.
        !          4262: \makecond{ifcommandnotdefined}
        !          4263: \def\ifcommandnotdefined{%
        !          4264:   \parsearg{\doifcmddefined{\else \let\next=\ifcmdnotdefinedfail}}}
        !          4265: \def\ifcmdnotdefinedfail{\doignore{ifcommandnotdefined}}
        !          4266: 
        !          4267: % Set the `txicommandconditionals' variable, so documents have a way to
        !          4268: % test if the @ifcommand...defined conditionals are available.
        !          4269: \set txicommandconditionals
        !          4270: 
        !          4271: % @dircategory CATEGORY  -- specify a category of the dir file
        !          4272: % which this file should belong to.  Ignore this in TeX.
        !          4273: \let\dircategory=\comment
        !          4274: 
        !          4275: % @defininfoenclose.
        !          4276: \let\definfoenclose=\comment
        !          4277: 
        !          4278: 
        !          4279: \message{indexing,}
        !          4280: % Index generation facilities
        !          4281: 
        !          4282: % Define \newwrite to be identical to plain tex's \newwrite
        !          4283: % except not \outer, so it can be used within macros and \if's.
        !          4284: \edef\newwrite{\makecsname{ptexnewwrite}}
        !          4285: 
        !          4286: % \newindex {foo} defines an index named foo.
        !          4287: % It automatically defines \fooindex such that
        !          4288: % \fooindex ...rest of line... puts an entry in the index foo.
        !          4289: % It also defines \fooindfile to be the number of the output channel for
        !          4290: % the file that accumulates this index.  The file's extension is foo.
        !          4291: % The name of an index should be no more than 2 characters long
        !          4292: % for the sake of vms.
        !          4293: %
        !          4294: \def\newindex#1{%
        !          4295:   \iflinks
        !          4296:     \expandafter\newwrite \csname#1indfile\endcsname
        !          4297:     \openout \csname#1indfile\endcsname \jobname.#1 % Open the file
        !          4298:   \fi
        !          4299:   \expandafter\xdef\csname#1index\endcsname{%     % Define @#1index
        !          4300:     \noexpand\doindex{#1}}
        !          4301: }
        !          4302: 
        !          4303: % @defindex foo  ==  \newindex{foo}
        !          4304: %
        !          4305: \def\defindex{\parsearg\newindex}
        !          4306: 
        !          4307: % Define @defcodeindex, like @defindex except put all entries in @code.
        !          4308: %
        !          4309: \def\defcodeindex{\parsearg\newcodeindex}
        !          4310: %
        !          4311: \def\newcodeindex#1{%
        !          4312:   \iflinks
        !          4313:     \expandafter\newwrite \csname#1indfile\endcsname
        !          4314:     \openout \csname#1indfile\endcsname \jobname.#1
        !          4315:   \fi
        !          4316:   \expandafter\xdef\csname#1index\endcsname{%
        !          4317:     \noexpand\docodeindex{#1}}%
        !          4318: }
        !          4319: 
        !          4320: 
        !          4321: % @synindex foo bar    makes index foo feed into index bar.
        !          4322: % Do this instead of @defindex foo if you don't want it as a separate index.
        !          4323: %
        !          4324: % @syncodeindex foo bar   similar, but put all entries made for index foo
        !          4325: % inside @code.
        !          4326: %
        !          4327: \def\synindex#1 #2 {\dosynindex\doindex{#1}{#2}}
        !          4328: \def\syncodeindex#1 #2 {\dosynindex\docodeindex{#1}{#2}}
        !          4329: 
        !          4330: % #1 is \doindex or \docodeindex, #2 the index getting redefined (foo),
        !          4331: % #3 the target index (bar).
        !          4332: \def\dosynindex#1#2#3{%
        !          4333:   % Only do \closeout if we haven't already done it, else we'll end up
        !          4334:   % closing the target index.
        !          4335:   \expandafter \ifx\csname donesynindex#2\endcsname \relax
        !          4336:     % The \closeout helps reduce unnecessary open files; the limit on the
        !          4337:     % Acorn RISC OS is a mere 16 files.
        !          4338:     \expandafter\closeout\csname#2indfile\endcsname
        !          4339:     \expandafter\let\csname donesynindex#2\endcsname = 1
        !          4340:   \fi
        !          4341:   % redefine \fooindfile:
        !          4342:   \expandafter\let\expandafter\temp\expandafter=\csname#3indfile\endcsname
        !          4343:   \expandafter\let\csname#2indfile\endcsname=\temp
        !          4344:   % redefine \fooindex:
        !          4345:   \expandafter\xdef\csname#2index\endcsname{\noexpand#1{#3}}%
        !          4346: }
        !          4347: 
        !          4348: % Define \doindex, the driver for all \fooindex macros.
        !          4349: % Argument #1 is generated by the calling \fooindex macro,
        !          4350: %  and it is "foo", the name of the index.
        !          4351: 
        !          4352: % \doindex just uses \parsearg; it calls \doind for the actual work.
        !          4353: % This is because \doind is more useful to call from other macros.
        !          4354: 
        !          4355: % There is also \dosubind {index}{topic}{subtopic}
        !          4356: % which makes an entry in a two-level index such as the operation index.
        !          4357: 
        !          4358: \def\doindex#1{\edef\indexname{#1}\parsearg\singleindexer}
        !          4359: \def\singleindexer #1{\doind{\indexname}{#1}}
        !          4360: 
        !          4361: % like the previous two, but they put @code around the argument.
        !          4362: \def\docodeindex#1{\edef\indexname{#1}\parsearg\singlecodeindexer}
        !          4363: \def\singlecodeindexer #1{\doind{\indexname}{\code{#1}}}
        !          4364: 
        !          4365: % Take care of Texinfo commands that can appear in an index entry.
        !          4366: % Since there are some commands we want to expand, and others we don't,
        !          4367: % we have to laboriously prevent expansion for those that we don't.
        !          4368: %
        !          4369: \def\indexdummies{%
        !          4370:   \escapechar = `\\     % use backslash in output files.
        !          4371:   \def\@{@}% change to @@ when we switch to @ as escape char in index files.
        !          4372:   \def\ {\realbackslash\space }%
        !          4373:   %
        !          4374:   % Need these unexpandable (because we define \tt as a dummy)
        !          4375:   % definitions when @{ or @} appear in index entry text.  Also, more
        !          4376:   % complicated, when \tex is in effect and \{ is a \delimiter again.
        !          4377:   % We can't use \lbracecmd and \rbracecmd because texindex assumes
        !          4378:   % braces and backslashes are used only as delimiters.  Perhaps we
        !          4379:   % should define @lbrace and @rbrace commands a la @comma.
        !          4380:   \def\{{{\tt\char123}}%
        !          4381:   \def\}{{\tt\char125}}%
        !          4382:   %
        !          4383:   % I don't entirely understand this, but when an index entry is
        !          4384:   % generated from a macro call, the \endinput which \scanmacro inserts
        !          4385:   % causes processing to be prematurely terminated.  This is,
        !          4386:   % apparently, because \indexsorttmp is fully expanded, and \endinput
        !          4387:   % is an expandable command.  The redefinition below makes \endinput
        !          4388:   % disappear altogether for that purpose -- although logging shows that
        !          4389:   % processing continues to some further point.  On the other hand, it
        !          4390:   % seems \endinput does not hurt in the printed index arg, since that
        !          4391:   % is still getting written without apparent harm.
        !          4392:   %
        !          4393:   % Sample source (mac-idx3.tex, reported by Graham Percival to
        !          4394:   % help-texinfo, 22may06):
        !          4395:   % @macro funindex {WORD}
        !          4396:   % @findex xyz
        !          4397:   % @end macro
        !          4398:   % ...
        !          4399:   % @funindex commtest
        !          4400:   %
        !          4401:   % The above is not enough to reproduce the bug, but it gives the flavor.
        !          4402:   %
        !          4403:   % Sample whatsit resulting:
        !          4404:   % .@write3{\entry{xyz}{@folio }{@code {xyz@endinput }}}
        !          4405:   %
        !          4406:   % So:
        !          4407:   \let\endinput = \empty
        !          4408:   %
        !          4409:   % Do the redefinitions.
        !          4410:   \commondummies
        !          4411: }
        !          4412: 
        !          4413: % For the aux and toc files, @ is the escape character.  So we want to
        !          4414: % redefine everything using @ as the escape character (instead of
        !          4415: % \realbackslash, still used for index files).  When everything uses @,
        !          4416: % this will be simpler.
        !          4417: %
        !          4418: \def\atdummies{%
        !          4419:   \def\@{@@}%
        !          4420:   \def\ {@ }%
        !          4421:   \let\{ = \lbraceatcmd
        !          4422:   \let\} = \rbraceatcmd
        !          4423:   %
        !          4424:   % Do the redefinitions.
        !          4425:   \commondummies
        !          4426:   \otherbackslash
        !          4427: }
        !          4428: 
        !          4429: % Called from \indexdummies and \atdummies.
        !          4430: %
        !          4431: \def\commondummies{%
        !          4432:   %
        !          4433:   % \definedummyword defines \#1 as \string\#1\space, thus effectively
        !          4434:   % preventing its expansion.  This is used only for control words,
        !          4435:   % not control letters, because the \space would be incorrect for
        !          4436:   % control characters, but is needed to separate the control word
        !          4437:   % from whatever follows.
        !          4438:   %
        !          4439:   % For control letters, we have \definedummyletter, which omits the
        !          4440:   % space.
        !          4441:   %
        !          4442:   % These can be used both for control words that take an argument and
        !          4443:   % those that do not.  If it is followed by {arg} in the input, then
        !          4444:   % that will dutifully get written to the index (or wherever).
        !          4445:   %
        !          4446:   \def\definedummyword  ##1{\def##1{\string##1\space}}%
        !          4447:   \def\definedummyletter##1{\def##1{\string##1}}%
        !          4448:   \let\definedummyaccent\definedummyletter
        !          4449:   %
        !          4450:   \commondummiesnofonts
        !          4451:   %
        !          4452:   \definedummyletter\_%
        !          4453:   \definedummyletter\-%
        !          4454:   %
        !          4455:   % Non-English letters.
        !          4456:   \definedummyword\AA
        !          4457:   \definedummyword\AE
        !          4458:   \definedummyword\DH
        !          4459:   \definedummyword\L
        !          4460:   \definedummyword\O
        !          4461:   \definedummyword\OE
        !          4462:   \definedummyword\TH
        !          4463:   \definedummyword\aa
        !          4464:   \definedummyword\ae
        !          4465:   \definedummyword\dh
        !          4466:   \definedummyword\exclamdown
        !          4467:   \definedummyword\l
        !          4468:   \definedummyword\o
        !          4469:   \definedummyword\oe
        !          4470:   \definedummyword\ordf
        !          4471:   \definedummyword\ordm
        !          4472:   \definedummyword\questiondown
        !          4473:   \definedummyword\ss
        !          4474:   \definedummyword\th
        !          4475:   %
        !          4476:   % Although these internal commands shouldn't show up, sometimes they do.
        !          4477:   \definedummyword\bf
        !          4478:   \definedummyword\gtr
        !          4479:   \definedummyword\hat
        !          4480:   \definedummyword\less
        !          4481:   \definedummyword\sf
        !          4482:   \definedummyword\sl
        !          4483:   \definedummyword\tclose
        !          4484:   \definedummyword\tt
        !          4485:   %
        !          4486:   \definedummyword\LaTeX
        !          4487:   \definedummyword\TeX
        !          4488:   %
        !          4489:   % Assorted special characters.
        !          4490:   \definedummyword\arrow
        !          4491:   \definedummyword\bullet
        !          4492:   \definedummyword\comma
        !          4493:   \definedummyword\copyright
        !          4494:   \definedummyword\registeredsymbol
        !          4495:   \definedummyword\dots
        !          4496:   \definedummyword\enddots
        !          4497:   \definedummyword\entrybreak
        !          4498:   \definedummyword\equiv
        !          4499:   \definedummyword\error
        !          4500:   \definedummyword\euro
        !          4501:   \definedummyword\expansion
        !          4502:   \definedummyword\geq
        !          4503:   \definedummyword\guillemetleft
        !          4504:   \definedummyword\guillemetright
        !          4505:   \definedummyword\guilsinglleft
        !          4506:   \definedummyword\guilsinglright
        !          4507:   \definedummyword\lbracechar
        !          4508:   \definedummyword\leq
        !          4509:   \definedummyword\minus
        !          4510:   \definedummyword\ogonek
        !          4511:   \definedummyword\pounds
        !          4512:   \definedummyword\point
        !          4513:   \definedummyword\print
        !          4514:   \definedummyword\quotedblbase
        !          4515:   \definedummyword\quotedblleft
        !          4516:   \definedummyword\quotedblright
        !          4517:   \definedummyword\quoteleft
        !          4518:   \definedummyword\quoteright
        !          4519:   \definedummyword\quotesinglbase
        !          4520:   \definedummyword\rbracechar
        !          4521:   \definedummyword\result
        !          4522:   \definedummyword\textdegree
        !          4523:   %
        !          4524:   % We want to disable all macros so that they are not expanded by \write.
        !          4525:   \macrolist
        !          4526:   %
        !          4527:   \normalturnoffactive
        !          4528:   %
        !          4529:   % Handle some cases of @value -- where it does not contain any
        !          4530:   % (non-fully-expandable) commands.
        !          4531:   \makevalueexpandable
        !          4532: }
        !          4533: 
        !          4534: % \commondummiesnofonts: common to \commondummies and \indexnofonts.
        !          4535: %
        !          4536: \def\commondummiesnofonts{%
        !          4537:   % Control letters and accents.
        !          4538:   \definedummyletter\!%
        !          4539:   \definedummyaccent\"%
        !          4540:   \definedummyaccent\'%
        !          4541:   \definedummyletter\*%
        !          4542:   \definedummyaccent\,%
        !          4543:   \definedummyletter\.%
        !          4544:   \definedummyletter\/%
        !          4545:   \definedummyletter\:%
        !          4546:   \definedummyaccent\=%
        !          4547:   \definedummyletter\?%
        !          4548:   \definedummyaccent\^%
        !          4549:   \definedummyaccent\`%
        !          4550:   \definedummyaccent\~%
        !          4551:   \definedummyword\u
        !          4552:   \definedummyword\v
        !          4553:   \definedummyword\H
        !          4554:   \definedummyword\dotaccent
        !          4555:   \definedummyword\ogonek
        !          4556:   \definedummyword\ringaccent
        !          4557:   \definedummyword\tieaccent
        !          4558:   \definedummyword\ubaraccent
        !          4559:   \definedummyword\udotaccent
        !          4560:   \definedummyword\dotless
        !          4561:   %
        !          4562:   % Texinfo font commands.
        !          4563:   \definedummyword\b
        !          4564:   \definedummyword\i
        !          4565:   \definedummyword\r
        !          4566:   \definedummyword\sansserif
        !          4567:   \definedummyword\sc
        !          4568:   \definedummyword\slanted
        !          4569:   \definedummyword\t
        !          4570:   %
        !          4571:   % Commands that take arguments.
        !          4572:   \definedummyword\abbr
        !          4573:   \definedummyword\acronym
        !          4574:   \definedummyword\anchor
        !          4575:   \definedummyword\cite
        !          4576:   \definedummyword\code
        !          4577:   \definedummyword\command
        !          4578:   \definedummyword\dfn
        !          4579:   \definedummyword\dmn
        !          4580:   \definedummyword\email
        !          4581:   \definedummyword\emph
        !          4582:   \definedummyword\env
        !          4583:   \definedummyword\file
        !          4584:   \definedummyword\image
        !          4585:   \definedummyword\indicateurl
        !          4586:   \definedummyword\inforef
        !          4587:   \definedummyword\kbd
        !          4588:   \definedummyword\key
        !          4589:   \definedummyword\math
        !          4590:   \definedummyword\option
        !          4591:   \definedummyword\pxref
        !          4592:   \definedummyword\ref
        !          4593:   \definedummyword\samp
        !          4594:   \definedummyword\strong
        !          4595:   \definedummyword\tie
        !          4596:   \definedummyword\uref
        !          4597:   \definedummyword\url
        !          4598:   \definedummyword\var
        !          4599:   \definedummyword\verb
        !          4600:   \definedummyword\w
        !          4601:   \definedummyword\xref
        !          4602: }
        !          4603: 
        !          4604: % \indexnofonts is used when outputting the strings to sort the index
        !          4605: % by, and when constructing control sequence names.  It eliminates all
        !          4606: % control sequences and just writes whatever the best ASCII sort string
        !          4607: % would be for a given command (usually its argument).
        !          4608: %
        !          4609: \def\indexnofonts{%
        !          4610:   % Accent commands should become @asis.
        !          4611:   \def\definedummyaccent##1{\let##1\asis}%
        !          4612:   % We can just ignore other control letters.
        !          4613:   \def\definedummyletter##1{\let##1\empty}%
        !          4614:   % All control words become @asis by default; overrides below.
        !          4615:   \let\definedummyword\definedummyaccent
        !          4616:   %
        !          4617:   \commondummiesnofonts
        !          4618:   %
        !          4619:   % Don't no-op \tt, since it isn't a user-level command
        !          4620:   % and is used in the definitions of the active chars like <, >, |, etc.
        !          4621:   % Likewise with the other plain tex font commands.
        !          4622:   %\let\tt=\asis
        !          4623:   %
        !          4624:   \def\ { }%
        !          4625:   \def\@{@}%
        !          4626:   \def\_{\normalunderscore}%
        !          4627:   \def\-{}% @- shouldn't affect sorting
        !          4628:   %
        !          4629:   % Unfortunately, texindex is not prepared to handle braces in the
        !          4630:   % content at all.  So for index sorting, we map @{ and @} to strings
        !          4631:   % starting with |, since that ASCII character is between ASCII { and }.
        !          4632:   \def\{{|a}%
        !          4633:   \def\lbracechar{|a}%
        !          4634:   %
        !          4635:   \def\}{|b}%
        !          4636:   \def\rbracechar{|b}%
        !          4637:   %
        !          4638:   % Non-English letters.
        !          4639:   \def\AA{AA}%
        !          4640:   \def\AE{AE}%
        !          4641:   \def\DH{DZZ}%
        !          4642:   \def\L{L}%
        !          4643:   \def\OE{OE}%
        !          4644:   \def\O{O}%
        !          4645:   \def\TH{ZZZ}%
        !          4646:   \def\aa{aa}%
        !          4647:   \def\ae{ae}%
        !          4648:   \def\dh{dzz}%
        !          4649:   \def\exclamdown{!}%
        !          4650:   \def\l{l}%
        !          4651:   \def\oe{oe}%
        !          4652:   \def\ordf{a}%
        !          4653:   \def\ordm{o}%
        !          4654:   \def\o{o}%
        !          4655:   \def\questiondown{?}%
        !          4656:   \def\ss{ss}%
        !          4657:   \def\th{zzz}%
        !          4658:   %
        !          4659:   \def\LaTeX{LaTeX}%
        !          4660:   \def\TeX{TeX}%
        !          4661:   %
        !          4662:   % Assorted special characters.
        !          4663:   % (The following {} will end up in the sort string, but that's ok.)
        !          4664:   \def\arrow{->}%
        !          4665:   \def\bullet{bullet}%
        !          4666:   \def\comma{,}%
        !          4667:   \def\copyright{copyright}%
        !          4668:   \def\dots{...}%
        !          4669:   \def\enddots{...}%
        !          4670:   \def\equiv{==}%
        !          4671:   \def\error{error}%
        !          4672:   \def\euro{euro}%
        !          4673:   \def\expansion{==>}%
        !          4674:   \def\geq{>=}%
        !          4675:   \def\guillemetleft{<<}%
        !          4676:   \def\guillemetright{>>}%
        !          4677:   \def\guilsinglleft{<}%
        !          4678:   \def\guilsinglright{>}%
        !          4679:   \def\leq{<=}%
        !          4680:   \def\minus{-}%
        !          4681:   \def\point{.}%
        !          4682:   \def\pounds{pounds}%
        !          4683:   \def\print{-|}%
        !          4684:   \def\quotedblbase{"}%
        !          4685:   \def\quotedblleft{"}%
        !          4686:   \def\quotedblright{"}%
        !          4687:   \def\quoteleft{`}%
        !          4688:   \def\quoteright{'}%
        !          4689:   \def\quotesinglbase{,}%
        !          4690:   \def\registeredsymbol{R}%
        !          4691:   \def\result{=>}%
        !          4692:   \def\textdegree{o}%
        !          4693:   %
        !          4694:   \expandafter\ifx\csname SETtxiindexlquoteignore\endcsname\relax
        !          4695:   \else \indexlquoteignore \fi
        !          4696:   %
        !          4697:   % We need to get rid of all macros, leaving only the arguments (if present).
        !          4698:   % Of course this is not nearly correct, but it is the best we can do for now.
        !          4699:   % makeinfo does not expand macros in the argument to @deffn, which ends up
        !          4700:   % writing an index entry, and texindex isn't prepared for an index sort entry
        !          4701:   % that starts with \.
        !          4702:   %
        !          4703:   % Since macro invocations are followed by braces, we can just redefine them
        !          4704:   % to take a single TeX argument.  The case of a macro invocation that
        !          4705:   % goes to end-of-line is not handled.
        !          4706:   %
        !          4707:   \macrolist
        !          4708: }
        !          4709: 
        !          4710: % Undocumented (for FSFS 2nd ed.): @set txiindexlquoteignore makes us
        !          4711: % ignore left quotes in the sort term.
        !          4712: {\catcode`\`=\active
        !          4713:  \gdef\indexlquoteignore{\let`=\empty}}
        !          4714: 
        !          4715: \let\indexbackslash=0  %overridden during \printindex.
        !          4716: \let\SETmarginindex=\relax % put index entries in margin (undocumented)?
        !          4717: 
        !          4718: % Most index entries go through here, but \dosubind is the general case.
        !          4719: % #1 is the index name, #2 is the entry text.
        !          4720: \def\doind#1#2{\dosubind{#1}{#2}{}}
        !          4721: 
        !          4722: % Workhorse for all \fooindexes.
        !          4723: % #1 is name of index, #2 is stuff to put there, #3 is subentry --
        !          4724: % empty if called from \doind, as we usually are (the main exception
        !          4725: % is with most defuns, which call us directly).
        !          4726: %
        !          4727: \def\dosubind#1#2#3{%
        !          4728:   \iflinks
        !          4729:   {%
        !          4730:     % Store the main index entry text (including the third arg).
        !          4731:     \toks0 = {#2}%
        !          4732:     % If third arg is present, precede it with a space.
        !          4733:     \def\thirdarg{#3}%
        !          4734:     \ifx\thirdarg\empty \else
        !          4735:       \toks0 = \expandafter{\the\toks0 \space #3}%
        !          4736:     \fi
        !          4737:     %
        !          4738:     \edef\writeto{\csname#1indfile\endcsname}%
        !          4739:     %
        !          4740:     \safewhatsit\dosubindwrite
        !          4741:   }%
        !          4742:   \fi
        !          4743: }
        !          4744: 
        !          4745: % Write the entry in \toks0 to the index file:
        !          4746: %
        !          4747: \def\dosubindwrite{%
        !          4748:   % Put the index entry in the margin if desired.
        !          4749:   \ifx\SETmarginindex\relax\else
        !          4750:     \insert\margin{\hbox{\vrule height8pt depth3pt width0pt \the\toks0}}%
        !          4751:   \fi
        !          4752:   %
        !          4753:   % Remember, we are within a group.
        !          4754:   \indexdummies % Must do this here, since \bf, etc expand at this stage
        !          4755:   \def\backslashcurfont{\indexbackslash}% \indexbackslash isn't defined now
        !          4756:       % so it will be output as is; and it will print as backslash.
        !          4757:   %
        !          4758:   % Process the index entry with all font commands turned off, to
        !          4759:   % get the string to sort by.
        !          4760:   {\indexnofonts
        !          4761:    \edef\temp{\the\toks0}% need full expansion
        !          4762:    \xdef\indexsorttmp{\temp}%
        !          4763:   }%
        !          4764:   %
        !          4765:   % Set up the complete index entry, with both the sort key and
        !          4766:   % the original text, including any font commands.  We write
        !          4767:   % three arguments to \entry to the .?? file (four in the
        !          4768:   % subentry case), texindex reduces to two when writing the .??s
        !          4769:   % sorted result.
        !          4770:   \edef\temp{%
        !          4771:     \write\writeto{%
        !          4772:       \string\entry{\indexsorttmp}{\noexpand\folio}{\the\toks0}}%
        !          4773:   }%
        !          4774:   \temp
        !          4775: }
        !          4776: 
        !          4777: % Take care of unwanted page breaks/skips around a whatsit:
        !          4778: %
        !          4779: % If a skip is the last thing on the list now, preserve it
        !          4780: % by backing up by \lastskip, doing the \write, then inserting
        !          4781: % the skip again.  Otherwise, the whatsit generated by the
        !          4782: % \write or \pdfdest will make \lastskip zero.  The result is that
        !          4783: % sequences like this:
        !          4784: % @end defun
        !          4785: % @tindex whatever
        !          4786: % @defun ...
        !          4787: % will have extra space inserted, because the \medbreak in the
        !          4788: % start of the @defun won't see the skip inserted by the @end of
        !          4789: % the previous defun.
        !          4790: %
        !          4791: % But don't do any of this if we're not in vertical mode.  We
        !          4792: % don't want to do a \vskip and prematurely end a paragraph.
        !          4793: %
        !          4794: % Avoid page breaks due to these extra skips, too.
        !          4795: %
        !          4796: % But wait, there is a catch there:
        !          4797: % We'll have to check whether \lastskip is zero skip.  \ifdim is not
        !          4798: % sufficient for this purpose, as it ignores stretch and shrink parts
        !          4799: % of the skip.  The only way seems to be to check the textual
        !          4800: % representation of the skip.
        !          4801: %
        !          4802: % The following is almost like \def\zeroskipmacro{0.0pt} except that
        !          4803: % the ``p'' and ``t'' characters have catcode \other, not 11 (letter).
        !          4804: %
        !          4805: \edef\zeroskipmacro{\expandafter\the\csname z@skip\endcsname}
        !          4806: %
        !          4807: \newskip\whatsitskip
        !          4808: \newcount\whatsitpenalty
        !          4809: %
        !          4810: % ..., ready, GO:
        !          4811: %
        !          4812: \def\safewhatsit#1{\ifhmode
        !          4813:   #1%
        !          4814:  \else
        !          4815:   % \lastskip and \lastpenalty cannot both be nonzero simultaneously.
        !          4816:   \whatsitskip = \lastskip
        !          4817:   \edef\lastskipmacro{\the\lastskip}%
        !          4818:   \whatsitpenalty = \lastpenalty
        !          4819:   %
        !          4820:   % If \lastskip is nonzero, that means the last item was a
        !          4821:   % skip.  And since a skip is discardable, that means this
        !          4822:   % -\whatsitskip glue we're inserting is preceded by a
        !          4823:   % non-discardable item, therefore it is not a potential
        !          4824:   % breakpoint, therefore no \nobreak needed.
        !          4825:   \ifx\lastskipmacro\zeroskipmacro
        !          4826:   \else
        !          4827:     \vskip-\whatsitskip
        !          4828:   \fi
        !          4829:   %
        !          4830:   #1%
        !          4831:   %
        !          4832:   \ifx\lastskipmacro\zeroskipmacro
        !          4833:     % If \lastskip was zero, perhaps the last item was a penalty, and
        !          4834:     % perhaps it was >=10000, e.g., a \nobreak.  In that case, we want
        !          4835:     % to re-insert the same penalty (values >10000 are used for various
        !          4836:     % signals); since we just inserted a non-discardable item, any
        !          4837:     % following glue (such as a \parskip) would be a breakpoint.  For example:
        !          4838:     %   @deffn deffn-whatever
        !          4839:     %   @vindex index-whatever
        !          4840:     %   Description.
        !          4841:     % would allow a break between the index-whatever whatsit
        !          4842:     % and the "Description." paragraph.
        !          4843:     \ifnum\whatsitpenalty>9999 \penalty\whatsitpenalty \fi
        !          4844:   \else
        !          4845:     % On the other hand, if we had a nonzero \lastskip,
        !          4846:     % this make-up glue would be preceded by a non-discardable item
        !          4847:     % (the whatsit from the \write), so we must insert a \nobreak.
        !          4848:     \nobreak\vskip\whatsitskip
        !          4849:   \fi
        !          4850: \fi}
        !          4851: 
        !          4852: % The index entry written in the file actually looks like
        !          4853: %  \entry {sortstring}{page}{topic}
        !          4854: % or
        !          4855: %  \entry {sortstring}{page}{topic}{subtopic}
        !          4856: % The texindex program reads in these files and writes files
        !          4857: % containing these kinds of lines:
        !          4858: %  \initial {c}
        !          4859: %     before the first topic whose initial is c
        !          4860: %  \entry {topic}{pagelist}
        !          4861: %     for a topic that is used without subtopics
        !          4862: %  \primary {topic}
        !          4863: %     for the beginning of a topic that is used with subtopics
        !          4864: %  \secondary {subtopic}{pagelist}
        !          4865: %     for each subtopic.
        !          4866: 
        !          4867: % Define the user-accessible indexing commands
        !          4868: % @findex, @vindex, @kindex, @cindex.
        !          4869: 
        !          4870: \def\findex {\fnindex}
        !          4871: \def\kindex {\kyindex}
        !          4872: \def\cindex {\cpindex}
        !          4873: \def\vindex {\vrindex}
        !          4874: \def\tindex {\tpindex}
        !          4875: \def\pindex {\pgindex}
        !          4876: 
        !          4877: \def\cindexsub {\begingroup\obeylines\cindexsub}
        !          4878: {\obeylines %
        !          4879: \gdef\cindexsub "#1" #2^^M{\endgroup %
        !          4880: \dosubind{cp}{#2}{#1}}}
        !          4881: 
        !          4882: % Define the macros used in formatting output of the sorted index material.
        !          4883: 
        !          4884: % @printindex causes a particular index (the ??s file) to get printed.
        !          4885: % It does not print any chapter heading (usually an @unnumbered).
        !          4886: %
        !          4887: \parseargdef\printindex{\begingroup
        !          4888:   \dobreak \chapheadingskip{10000}%
        !          4889:   %
        !          4890:   \smallfonts \rm
        !          4891:   \tolerance = 9500
        !          4892:   \plainfrenchspacing
        !          4893:   \everypar = {}% don't want the \kern\-parindent from indentation suppression.
        !          4894:   %
        !          4895:   % See if the index file exists and is nonempty.
        !          4896:   % Change catcode of @ here so that if the index file contains
        !          4897:   % \initial {@}
        !          4898:   % as its first line, TeX doesn't complain about mismatched braces
        !          4899:   % (because it thinks @} is a control sequence).
        !          4900:   \catcode`\@ = 11
        !          4901:   \openin 1 \jobname.#1s
        !          4902:   \ifeof 1
        !          4903:     % \enddoublecolumns gets confused if there is no text in the index,
        !          4904:     % and it loses the chapter title and the aux file entries for the
        !          4905:     % index.  The easiest way to prevent this problem is to make sure
        !          4906:     % there is some text.
        !          4907:     \putwordIndexNonexistent
        !          4908:   \else
        !          4909:     %
        !          4910:     % If the index file exists but is empty, then \openin leaves \ifeof
        !          4911:     % false.  We have to make TeX try to read something from the file, so
        !          4912:     % it can discover if there is anything in it.
        !          4913:     \read 1 to \temp
        !          4914:     \ifeof 1
        !          4915:       \putwordIndexIsEmpty
        !          4916:     \else
        !          4917:       % Index files are almost Texinfo source, but we use \ as the escape
        !          4918:       % character.  It would be better to use @, but that's too big a change
        !          4919:       % to make right now.
        !          4920:       \def\indexbackslash{\backslashcurfont}%
        !          4921:       \catcode`\\ = 0
        !          4922:       \escapechar = `\\
        !          4923:       \begindoublecolumns
        !          4924:       \input \jobname.#1s
        !          4925:       \enddoublecolumns
        !          4926:     \fi
        !          4927:   \fi
        !          4928:   \closein 1
        !          4929: \endgroup}
        !          4930: 
        !          4931: % These macros are used by the sorted index file itself.
        !          4932: % Change them to control the appearance of the index.
        !          4933: 
        !          4934: \def\initial#1{{%
        !          4935:   % Some minor font changes for the special characters.
        !          4936:   \let\tentt=\sectt \let\tt=\sectt \let\sf=\sectt
        !          4937:   %
        !          4938:   % Remove any glue we may have, we'll be inserting our own.
        !          4939:   \removelastskip
        !          4940:   %
        !          4941:   % We like breaks before the index initials, so insert a bonus.
        !          4942:   \nobreak
        !          4943:   \vskip 0pt plus 3\baselineskip
        !          4944:   \penalty 0
        !          4945:   \vskip 0pt plus -3\baselineskip
        !          4946:   %
        !          4947:   % Typeset the initial.  Making this add up to a whole number of
        !          4948:   % baselineskips increases the chance of the dots lining up from column
        !          4949:   % to column.  It still won't often be perfect, because of the stretch
        !          4950:   % we need before each entry, but it's better.
        !          4951:   %
        !          4952:   % No shrink because it confuses \balancecolumns.
        !          4953:   \vskip 1.67\baselineskip plus .5\baselineskip
        !          4954:   \leftline{\secbf #1}%
        !          4955:   % Do our best not to break after the initial.
        !          4956:   \nobreak
        !          4957:   \vskip .33\baselineskip plus .1\baselineskip
        !          4958: }}
        !          4959: 
        !          4960: % \entry typesets a paragraph consisting of the text (#1), dot leaders, and
        !          4961: % then page number (#2) flushed to the right margin.  It is used for index
        !          4962: % and table of contents entries.  The paragraph is indented by \leftskip.
        !          4963: %
        !          4964: % A straightforward implementation would start like this:
        !          4965: %      \def\entry#1#2{...
        !          4966: % But this freezes the catcodes in the argument, and can cause problems to
        !          4967: % @code, which sets - active.  This problem was fixed by a kludge---
        !          4968: % ``-'' was active throughout whole index, but this isn't really right.
        !          4969: % The right solution is to prevent \entry from swallowing the whole text.
        !          4970: %                                 --kasal, 21nov03
        !          4971: \def\entry{%
        !          4972:   \begingroup
        !          4973:     %
        !          4974:     % Start a new paragraph if necessary, so our assignments below can't
        !          4975:     % affect previous text.
        !          4976:     \par
        !          4977:     %
        !          4978:     % Do not fill out the last line with white space.
        !          4979:     \parfillskip = 0in
        !          4980:     %
        !          4981:     % No extra space above this paragraph.
        !          4982:     \parskip = 0in
        !          4983:     %
        !          4984:     % Do not prefer a separate line ending with a hyphen to fewer lines.
        !          4985:     \finalhyphendemerits = 0
        !          4986:     %
        !          4987:     % \hangindent is only relevant when the entry text and page number
        !          4988:     % don't both fit on one line.  In that case, bob suggests starting the
        !          4989:     % dots pretty far over on the line.  Unfortunately, a large
        !          4990:     % indentation looks wrong when the entry text itself is broken across
        !          4991:     % lines.  So we use a small indentation and put up with long leaders.
        !          4992:     %
        !          4993:     % \hangafter is reset to 1 (which is the value we want) at the start
        !          4994:     % of each paragraph, so we need not do anything with that.
        !          4995:     \hangindent = 2em
        !          4996:     %
        !          4997:     % When the entry text needs to be broken, just fill out the first line
        !          4998:     % with blank space.
        !          4999:     \rightskip = 0pt plus1fil
        !          5000:     %
        !          5001:     % A bit of stretch before each entry for the benefit of balancing
        !          5002:     % columns.
        !          5003:     \vskip 0pt plus1pt
        !          5004:     %
        !          5005:     % When reading the text of entry, convert explicit line breaks
        !          5006:     % from @* into spaces.  The user might give these in long section
        !          5007:     % titles, for instance.
        !          5008:     \def\*{\unskip\space\ignorespaces}%
        !          5009:     \def\entrybreak{\hfil\break}%
        !          5010:     %
        !          5011:     % Swallow the left brace of the text (first parameter):
        !          5012:     \afterassignment\doentry
        !          5013:     \let\temp =
        !          5014: }
        !          5015: \def\entrybreak{\unskip\space\ignorespaces}%
        !          5016: \def\doentry{%
        !          5017:     \bgroup % Instead of the swallowed brace.
        !          5018:       \noindent
        !          5019:       \aftergroup\finishentry
        !          5020:       % And now comes the text of the entry.
        !          5021: }
        !          5022: \def\finishentry#1{%
        !          5023:     % #1 is the page number.
        !          5024:     %
        !          5025:     % The following is kludged to not output a line of dots in the index if
        !          5026:     % there are no page numbers.  The next person who breaks this will be
        !          5027:     % cursed by a Unix daemon.
        !          5028:     \setbox\boxA = \hbox{#1}%
        !          5029:     \ifdim\wd\boxA = 0pt
        !          5030:       \ %
        !          5031:     \else
        !          5032:       %
        !          5033:       % If we must, put the page number on a line of its own, and fill out
        !          5034:       % this line with blank space.  (The \hfil is overwhelmed with the
        !          5035:       % fill leaders glue in \indexdotfill if the page number does fit.)
        !          5036:       \hfil\penalty50
        !          5037:       \null\nobreak\indexdotfill % Have leaders before the page number.
        !          5038:       %
        !          5039:       % The `\ ' here is removed by the implicit \unskip that TeX does as
        !          5040:       % part of (the primitive) \par.  Without it, a spurious underfull
        !          5041:       % \hbox ensues.
        !          5042:       \ifpdf
        !          5043:        \pdfgettoks#1.%
        !          5044:        \ \the\toksA
        !          5045:       \else
        !          5046:        \ #1%
        !          5047:       \fi
        !          5048:     \fi
        !          5049:     \par
        !          5050:   \endgroup
        !          5051: }
        !          5052: 
        !          5053: % Like plain.tex's \dotfill, except uses up at least 1 em.
        !          5054: \def\indexdotfill{\cleaders
        !          5055:   \hbox{$\mathsurround=0pt \mkern1.5mu.\mkern1.5mu$}\hskip 1em plus 1fill}
        !          5056: 
        !          5057: \def\primary #1{\line{#1\hfil}}
        !          5058: 
        !          5059: \newskip\secondaryindent \secondaryindent=0.5cm
        !          5060: \def\secondary#1#2{{%
        !          5061:   \parfillskip=0in
        !          5062:   \parskip=0in
        !          5063:   \hangindent=1in
        !          5064:   \hangafter=1
        !          5065:   \noindent\hskip\secondaryindent\hbox{#1}\indexdotfill
        !          5066:   \ifpdf
        !          5067:     \pdfgettoks#2.\ \the\toksA % The page number ends the paragraph.
        !          5068:   \else
        !          5069:     #2
        !          5070:   \fi
        !          5071:   \par
        !          5072: }}
        !          5073: 
        !          5074: % Define two-column mode, which we use to typeset indexes.
        !          5075: % Adapted from the TeXbook, page 416, which is to say,
        !          5076: % the manmac.tex format used to print the TeXbook itself.
        !          5077: \catcode`\@=11
        !          5078: 
        !          5079: \newbox\partialpage
        !          5080: \newdimen\doublecolumnhsize
        !          5081: 
        !          5082: \def\begindoublecolumns{\begingroup % ended by \enddoublecolumns
        !          5083:   % Grab any single-column material above us.
        !          5084:   \output = {%
        !          5085:     %
        !          5086:     % Here is a possibility not foreseen in manmac: if we accumulate a
        !          5087:     % whole lot of material, we might end up calling this \output
        !          5088:     % routine twice in a row (see the doublecol-lose test, which is
        !          5089:     % essentially a couple of indexes with @setchapternewpage off).  In
        !          5090:     % that case we just ship out what is in \partialpage with the normal
        !          5091:     % output routine.  Generally, \partialpage will be empty when this
        !          5092:     % runs and this will be a no-op.  See the indexspread.tex test case.
        !          5093:     \ifvoid\partialpage \else
        !          5094:       \onepageout{\pagecontents\partialpage}%
        !          5095:     \fi
        !          5096:     %
        !          5097:     \global\setbox\partialpage = \vbox{%
        !          5098:       % Unvbox the main output page.
        !          5099:       \unvbox\PAGE
        !          5100:       \kern-\topskip \kern\baselineskip
        !          5101:     }%
        !          5102:   }%
        !          5103:   \eject % run that output routine to set \partialpage
        !          5104:   %
        !          5105:   % Use the double-column output routine for subsequent pages.
        !          5106:   \output = {\doublecolumnout}%
        !          5107:   %
        !          5108:   % Change the page size parameters.  We could do this once outside this
        !          5109:   % routine, in each of @smallbook, @afourpaper, and the default 8.5x11
        !          5110:   % format, but then we repeat the same computation.  Repeating a couple
        !          5111:   % of assignments once per index is clearly meaningless for the
        !          5112:   % execution time, so we may as well do it in one place.
        !          5113:   %
        !          5114:   % First we halve the line length, less a little for the gutter between
        !          5115:   % the columns.  We compute the gutter based on the line length, so it
        !          5116:   % changes automatically with the paper format.  The magic constant
        !          5117:   % below is chosen so that the gutter has the same value (well, +-<1pt)
        !          5118:   % as it did when we hard-coded it.
        !          5119:   %
        !          5120:   % We put the result in a separate register, \doublecolumhsize, so we
        !          5121:   % can restore it in \pagesofar, after \hsize itself has (potentially)
        !          5122:   % been clobbered.
        !          5123:   %
        !          5124:   \doublecolumnhsize = \hsize
        !          5125:     \advance\doublecolumnhsize by -.04154\hsize
        !          5126:     \divide\doublecolumnhsize by 2
        !          5127:   \hsize = \doublecolumnhsize
        !          5128:   %
        !          5129:   % Double the \vsize as well.  (We don't need a separate register here,
        !          5130:   % since nobody clobbers \vsize.)
        !          5131:   \vsize = 2\vsize
        !          5132: }
        !          5133: 
        !          5134: % The double-column output routine for all double-column pages except
        !          5135: % the last.
        !          5136: %
        !          5137: \def\doublecolumnout{%
        !          5138:   \splittopskip=\topskip \splitmaxdepth=\maxdepth
        !          5139:   % Get the available space for the double columns -- the normal
        !          5140:   % (undoubled) page height minus any material left over from the
        !          5141:   % previous page.
        !          5142:   \dimen@ = \vsize
        !          5143:   \divide\dimen@ by 2
        !          5144:   \advance\dimen@ by -\ht\partialpage
        !          5145:   %
        !          5146:   % box0 will be the left-hand column, box2 the right.
        !          5147:   \setbox0=\vsplit255 to\dimen@ \setbox2=\vsplit255 to\dimen@
        !          5148:   \onepageout\pagesofar
        !          5149:   \unvbox255
        !          5150:   \penalty\outputpenalty
        !          5151: }
        !          5152: %
        !          5153: % Re-output the contents of the output page -- any previous material,
        !          5154: % followed by the two boxes we just split, in box0 and box2.
        !          5155: \def\pagesofar{%
        !          5156:   \unvbox\partialpage
        !          5157:   %
        !          5158:   \hsize = \doublecolumnhsize
        !          5159:   \wd0=\hsize \wd2=\hsize
        !          5160:   \hbox to\pagewidth{\box0\hfil\box2}%
        !          5161: }
        !          5162: %
        !          5163: % All done with double columns.
        !          5164: \def\enddoublecolumns{%
        !          5165:   % The following penalty ensures that the page builder is exercised
        !          5166:   % _before_ we change the output routine.  This is necessary in the
        !          5167:   % following situation:
        !          5168:   %
        !          5169:   % The last section of the index consists only of a single entry.
        !          5170:   % Before this section, \pagetotal is less than \pagegoal, so no
        !          5171:   % break occurs before the last section starts.  However, the last
        !          5172:   % section, consisting of \initial and the single \entry, does not
        !          5173:   % fit on the page and has to be broken off.  Without the following
        !          5174:   % penalty the page builder will not be exercised until \eject
        !          5175:   % below, and by that time we'll already have changed the output
        !          5176:   % routine to the \balancecolumns version, so the next-to-last
        !          5177:   % double-column page will be processed with \balancecolumns, which
        !          5178:   % is wrong:  The two columns will go to the main vertical list, with
        !          5179:   % the broken-off section in the recent contributions.  As soon as
        !          5180:   % the output routine finishes, TeX starts reconsidering the page
        !          5181:   % break.  The two columns and the broken-off section both fit on the
        !          5182:   % page, because the two columns now take up only half of the page
        !          5183:   % goal.  When TeX sees \eject from below which follows the final
        !          5184:   % section, it invokes the new output routine that we've set after
        !          5185:   % \balancecolumns below; \onepageout will try to fit the two columns
        !          5186:   % and the final section into the vbox of \pageheight (see
        !          5187:   % \pagebody), causing an overfull box.
        !          5188:   %
        !          5189:   % Note that glue won't work here, because glue does not exercise the
        !          5190:   % page builder, unlike penalties (see The TeXbook, pp. 280-281).
        !          5191:   \penalty0
        !          5192:   %
        !          5193:   \output = {%
        !          5194:     % Split the last of the double-column material.  Leave it on the
        !          5195:     % current page, no automatic page break.
        !          5196:     \balancecolumns
        !          5197:     %
        !          5198:     % If we end up splitting too much material for the current page,
        !          5199:     % though, there will be another page break right after this \output
        !          5200:     % invocation ends.  Having called \balancecolumns once, we do not
        !          5201:     % want to call it again.  Therefore, reset \output to its normal
        !          5202:     % definition right away.  (We hope \balancecolumns will never be
        !          5203:     % called on to balance too much material, but if it is, this makes
        !          5204:     % the output somewhat more palatable.)
        !          5205:     \global\output = {\onepageout{\pagecontents\PAGE}}%
        !          5206:   }%
        !          5207:   \eject
        !          5208:   \endgroup % started in \begindoublecolumns
        !          5209:   %
        !          5210:   % \pagegoal was set to the doubled \vsize above, since we restarted
        !          5211:   % the current page.  We're now back to normal single-column
        !          5212:   % typesetting, so reset \pagegoal to the normal \vsize (after the
        !          5213:   % \endgroup where \vsize got restored).
        !          5214:   \pagegoal = \vsize
        !          5215: }
        !          5216: %
        !          5217: % Called at the end of the double column material.
        !          5218: \def\balancecolumns{%
        !          5219:   \setbox0 = \vbox{\unvbox255}% like \box255 but more efficient, see p.120.
        !          5220:   \dimen@ = \ht0
        !          5221:   \advance\dimen@ by \topskip
        !          5222:   \advance\dimen@ by-\baselineskip
        !          5223:   \divide\dimen@ by 2 % target to split to
        !          5224:   %debug\message{final 2-column material height=\the\ht0, target=\the\dimen@.}%
        !          5225:   \splittopskip = \topskip
        !          5226:   % Loop until we get a decent breakpoint.
        !          5227:   {%
        !          5228:     \vbadness = 10000
        !          5229:     \loop
        !          5230:       \global\setbox3 = \copy0
        !          5231:       \global\setbox1 = \vsplit3 to \dimen@
        !          5232:     \ifdim\ht3>\dimen@
        !          5233:       \global\advance\dimen@ by 1pt
        !          5234:     \repeat
        !          5235:   }%
        !          5236:   %debug\message{split to \the\dimen@, column heights: \the\ht1, \the\ht3.}%
        !          5237:   \setbox0=\vbox to\dimen@{\unvbox1}%
        !          5238:   \setbox2=\vbox to\dimen@{\unvbox3}%
        !          5239:   %
        !          5240:   \pagesofar
        !          5241: }
        !          5242: \catcode`\@ = \other
        !          5243: 
        !          5244: 
        !          5245: \message{sectioning,}
        !          5246: % Chapters, sections, etc.
        !          5247: 
        !          5248: % Let's start with @part.
        !          5249: \outer\parseargdef\part{\partzzz{#1}}
        !          5250: \def\partzzz#1{%
        !          5251:   \chapoddpage
        !          5252:   \null
        !          5253:   \vskip.3\vsize  % move it down on the page a bit
        !          5254:   \begingroup
        !          5255:     \noindent \titlefonts\rmisbold #1\par % the text
        !          5256:     \let\lastnode=\empty      % no node to associate with
        !          5257:     \writetocentry{part}{#1}{}% but put it in the toc
        !          5258:     \headingsoff              % no headline or footline on the part page
        !          5259:     \chapoddpage
        !          5260:   \endgroup
        !          5261: }
        !          5262: 
        !          5263: % \unnumberedno is an oxymoron.  But we count the unnumbered
        !          5264: % sections so that we can refer to them unambiguously in the pdf
        !          5265: % outlines by their "section number".  We avoid collisions with chapter
        !          5266: % numbers by starting them at 10000.  (If a document ever has 10000
        !          5267: % chapters, we're in trouble anyway, I'm sure.)
        !          5268: \newcount\unnumberedno \unnumberedno = 10000
        !          5269: \newcount\chapno
        !          5270: \newcount\secno        \secno=0
        !          5271: \newcount\subsecno     \subsecno=0
        !          5272: \newcount\subsubsecno  \subsubsecno=0
        !          5273: 
        !          5274: % This counter is funny since it counts through charcodes of letters A, B, ...
        !          5275: \newcount\appendixno  \appendixno = `\@
        !          5276: %
        !          5277: % \def\appendixletter{\char\the\appendixno}
        !          5278: % We do the following ugly conditional instead of the above simple
        !          5279: % construct for the sake of pdftex, which needs the actual
        !          5280: % letter in the expansion, not just typeset.
        !          5281: %
        !          5282: \def\appendixletter{%
        !          5283:   \ifnum\appendixno=`A A%
        !          5284:   \else\ifnum\appendixno=`B B%
        !          5285:   \else\ifnum\appendixno=`C C%
        !          5286:   \else\ifnum\appendixno=`D D%
        !          5287:   \else\ifnum\appendixno=`E E%
        !          5288:   \else\ifnum\appendixno=`F F%
        !          5289:   \else\ifnum\appendixno=`G G%
        !          5290:   \else\ifnum\appendixno=`H H%
        !          5291:   \else\ifnum\appendixno=`I I%
        !          5292:   \else\ifnum\appendixno=`J J%
        !          5293:   \else\ifnum\appendixno=`K K%
        !          5294:   \else\ifnum\appendixno=`L L%
        !          5295:   \else\ifnum\appendixno=`M M%
        !          5296:   \else\ifnum\appendixno=`N N%
        !          5297:   \else\ifnum\appendixno=`O O%
        !          5298:   \else\ifnum\appendixno=`P P%
        !          5299:   \else\ifnum\appendixno=`Q Q%
        !          5300:   \else\ifnum\appendixno=`R R%
        !          5301:   \else\ifnum\appendixno=`S S%
        !          5302:   \else\ifnum\appendixno=`T T%
        !          5303:   \else\ifnum\appendixno=`U U%
        !          5304:   \else\ifnum\appendixno=`V V%
        !          5305:   \else\ifnum\appendixno=`W W%
        !          5306:   \else\ifnum\appendixno=`X X%
        !          5307:   \else\ifnum\appendixno=`Y Y%
        !          5308:   \else\ifnum\appendixno=`Z Z%
        !          5309:   % The \the is necessary, despite appearances, because \appendixletter is
        !          5310:   % expanded while writing the .toc file.  \char\appendixno is not
        !          5311:   % expandable, thus it is written literally, thus all appendixes come out
        !          5312:   % with the same letter (or @) in the toc without it.
        !          5313:   \else\char\the\appendixno
        !          5314:   \fi\fi\fi\fi\fi\fi\fi\fi\fi\fi\fi\fi\fi
        !          5315:   \fi\fi\fi\fi\fi\fi\fi\fi\fi\fi\fi\fi\fi}
        !          5316: 
        !          5317: % Each @chapter defines these (using marks) as the number+name, number
        !          5318: % and name of the chapter.  Page headings and footings can use
        !          5319: % these.  @section does likewise.
        !          5320: \def\thischapter{}
        !          5321: \def\thischapternum{}
        !          5322: \def\thischaptername{}
        !          5323: \def\thissection{}
        !          5324: \def\thissectionnum{}
        !          5325: \def\thissectionname{}
        !          5326: 
        !          5327: \newcount\absseclevel % used to calculate proper heading level
        !          5328: \newcount\secbase\secbase=0 % @raisesections/@lowersections modify this count
        !          5329: 
        !          5330: % @raisesections: treat @section as chapter, @subsection as section, etc.
        !          5331: \def\raisesections{\global\advance\secbase by -1}
        !          5332: \let\up=\raisesections % original BFox name
        !          5333: 
        !          5334: % @lowersections: treat @chapter as section, @section as subsection, etc.
        !          5335: \def\lowersections{\global\advance\secbase by 1}
        !          5336: \let\down=\lowersections % original BFox name
        !          5337: 
        !          5338: % we only have subsub.
        !          5339: \chardef\maxseclevel = 3
        !          5340: %
        !          5341: % A numbered section within an unnumbered changes to unnumbered too.
        !          5342: % To achieve this, remember the "biggest" unnum. sec. we are currently in:
        !          5343: \chardef\unnlevel = \maxseclevel
        !          5344: %
        !          5345: % Trace whether the current chapter is an appendix or not:
        !          5346: % \chapheadtype is "N" or "A", unnumbered chapters are ignored.
        !          5347: \def\chapheadtype{N}
        !          5348: 
        !          5349: % Choose a heading macro
        !          5350: % #1 is heading type
        !          5351: % #2 is heading level
        !          5352: % #3 is text for heading
        !          5353: \def\genhead#1#2#3{%
        !          5354:   % Compute the abs. sec. level:
        !          5355:   \absseclevel=#2
        !          5356:   \advance\absseclevel by \secbase
        !          5357:   % Make sure \absseclevel doesn't fall outside the range:
        !          5358:   \ifnum \absseclevel < 0
        !          5359:     \absseclevel = 0
        !          5360:   \else
        !          5361:     \ifnum \absseclevel > 3
        !          5362:       \absseclevel = 3
        !          5363:     \fi
        !          5364:   \fi
        !          5365:   % The heading type:
        !          5366:   \def\headtype{#1}%
        !          5367:   \if \headtype U%
        !          5368:     \ifnum \absseclevel < \unnlevel
        !          5369:       \chardef\unnlevel = \absseclevel
        !          5370:     \fi
        !          5371:   \else
        !          5372:     % Check for appendix sections:
        !          5373:     \ifnum \absseclevel = 0
        !          5374:       \edef\chapheadtype{\headtype}%
        !          5375:     \else
        !          5376:       \if \headtype A\if \chapheadtype N%
        !          5377:        \errmessage{@appendix... within a non-appendix chapter}%
        !          5378:       \fi\fi
        !          5379:     \fi
        !          5380:     % Check for numbered within unnumbered:
        !          5381:     \ifnum \absseclevel > \unnlevel
        !          5382:       \def\headtype{U}%
        !          5383:     \else
        !          5384:       \chardef\unnlevel = 3
        !          5385:     \fi
        !          5386:   \fi
        !          5387:   % Now print the heading:
        !          5388:   \if \headtype U%
        !          5389:     \ifcase\absseclevel
        !          5390:        \unnumberedzzz{#3}%
        !          5391:     \or \unnumberedseczzz{#3}%
        !          5392:     \or \unnumberedsubseczzz{#3}%
        !          5393:     \or \unnumberedsubsubseczzz{#3}%
        !          5394:     \fi
        !          5395:   \else
        !          5396:     \if \headtype A%
        !          5397:       \ifcase\absseclevel
        !          5398:          \appendixzzz{#3}%
        !          5399:       \or \appendixsectionzzz{#3}%
        !          5400:       \or \appendixsubseczzz{#3}%
        !          5401:       \or \appendixsubsubseczzz{#3}%
        !          5402:       \fi
        !          5403:     \else
        !          5404:       \ifcase\absseclevel
        !          5405:          \chapterzzz{#3}%
        !          5406:       \or \seczzz{#3}%
        !          5407:       \or \numberedsubseczzz{#3}%
        !          5408:       \or \numberedsubsubseczzz{#3}%
        !          5409:       \fi
        !          5410:     \fi
        !          5411:   \fi
        !          5412:   \suppressfirstparagraphindent
        !          5413: }
        !          5414: 
        !          5415: % an interface:
        !          5416: \def\numhead{\genhead N}
        !          5417: \def\apphead{\genhead A}
        !          5418: \def\unnmhead{\genhead U}
        !          5419: 
        !          5420: % @chapter, @appendix, @unnumbered.  Increment top-level counter, reset
        !          5421: % all lower-level sectioning counters to zero.
        !          5422: %
        !          5423: % Also set \chaplevelprefix, which we prepend to @float sequence numbers
        !          5424: % (e.g., figures), q.v.  By default (before any chapter), that is empty.
        !          5425: \let\chaplevelprefix = \empty
        !          5426: %
        !          5427: \outer\parseargdef\chapter{\numhead0{#1}} % normally numhead0 calls chapterzzz
        !          5428: \def\chapterzzz#1{%
        !          5429:   % section resetting is \global in case the chapter is in a group, such
        !          5430:   % as an @include file.
        !          5431:   \global\secno=0 \global\subsecno=0 \global\subsubsecno=0
        !          5432:     \global\advance\chapno by 1
        !          5433:   %
        !          5434:   % Used for \float.
        !          5435:   \gdef\chaplevelprefix{\the\chapno.}%
        !          5436:   \resetallfloatnos
        !          5437:   %
        !          5438:   % \putwordChapter can contain complex things in translations.
        !          5439:   \toks0=\expandafter{\putwordChapter}%
        !          5440:   \message{\the\toks0 \space \the\chapno}%
        !          5441:   %
        !          5442:   % Write the actual heading.
        !          5443:   \chapmacro{#1}{Ynumbered}{\the\chapno}%
        !          5444:   %
        !          5445:   % So @section and the like are numbered underneath this chapter.
        !          5446:   \global\let\section = \numberedsec
        !          5447:   \global\let\subsection = \numberedsubsec
        !          5448:   \global\let\subsubsection = \numberedsubsubsec
        !          5449: }
        !          5450: 
        !          5451: \outer\parseargdef\appendix{\apphead0{#1}} % normally calls appendixzzz
        !          5452: %
        !          5453: \def\appendixzzz#1{%
        !          5454:   \global\secno=0 \global\subsecno=0 \global\subsubsecno=0
        !          5455:     \global\advance\appendixno by 1
        !          5456:   \gdef\chaplevelprefix{\appendixletter.}%
        !          5457:   \resetallfloatnos
        !          5458:   %
        !          5459:   % \putwordAppendix can contain complex things in translations.
        !          5460:   \toks0=\expandafter{\putwordAppendix}%
        !          5461:   \message{\the\toks0 \space \appendixletter}%
        !          5462:   %
        !          5463:   \chapmacro{#1}{Yappendix}{\appendixletter}%
        !          5464:   %
        !          5465:   \global\let\section = \appendixsec
        !          5466:   \global\let\subsection = \appendixsubsec
        !          5467:   \global\let\subsubsection = \appendixsubsubsec
        !          5468: }
        !          5469: 
        !          5470: % normally unnmhead0 calls unnumberedzzz:
        !          5471: \outer\parseargdef\unnumbered{\unnmhead0{#1}}
        !          5472: \def\unnumberedzzz#1{%
        !          5473:   \global\secno=0 \global\subsecno=0 \global\subsubsecno=0
        !          5474:     \global\advance\unnumberedno by 1
        !          5475:   %
        !          5476:   % Since an unnumbered has no number, no prefix for figures.
        !          5477:   \global\let\chaplevelprefix = \empty
        !          5478:   \resetallfloatnos
        !          5479:   %
        !          5480:   % This used to be simply \message{#1}, but TeX fully expands the
        !          5481:   % argument to \message.  Therefore, if #1 contained @-commands, TeX
        !          5482:   % expanded them.  For example, in `@unnumbered The @cite{Book}', TeX
        !          5483:   % expanded @cite (which turns out to cause errors because \cite is meant
        !          5484:   % to be executed, not expanded).
        !          5485:   %
        !          5486:   % Anyway, we don't want the fully-expanded definition of @cite to appear
        !          5487:   % as a result of the \message, we just want `@cite' itself.  We use
        !          5488:   % \the<toks register> to achieve this: TeX expands \the<toks> only once,
        !          5489:   % simply yielding the contents of <toks register>.  (We also do this for
        !          5490:   % the toc entries.)
        !          5491:   \toks0 = {#1}%
        !          5492:   \message{(\the\toks0)}%
        !          5493:   %
        !          5494:   \chapmacro{#1}{Ynothing}{\the\unnumberedno}%
        !          5495:   %
        !          5496:   \global\let\section = \unnumberedsec
        !          5497:   \global\let\subsection = \unnumberedsubsec
        !          5498:   \global\let\subsubsection = \unnumberedsubsubsec
        !          5499: }
        !          5500: 
        !          5501: % @centerchap is like @unnumbered, but the heading is centered.
        !          5502: \outer\parseargdef\centerchap{%
        !          5503:   % Well, we could do the following in a group, but that would break
        !          5504:   % an assumption that \chapmacro is called at the outermost level.
        !          5505:   % Thus we are safer this way:                --kasal, 24feb04
        !          5506:   \let\centerparametersmaybe = \centerparameters
        !          5507:   \unnmhead0{#1}%
        !          5508:   \let\centerparametersmaybe = \relax
        !          5509: }
        !          5510: 
        !          5511: % @top is like @unnumbered.
        !          5512: \let\top\unnumbered
        !          5513: 
        !          5514: % Sections.
        !          5515: % 
        !          5516: \outer\parseargdef\numberedsec{\numhead1{#1}} % normally calls seczzz
        !          5517: \def\seczzz#1{%
        !          5518:   \global\subsecno=0 \global\subsubsecno=0  \global\advance\secno by 1
        !          5519:   \sectionheading{#1}{sec}{Ynumbered}{\the\chapno.\the\secno}%
        !          5520: }
        !          5521: 
        !          5522: % normally calls appendixsectionzzz:
        !          5523: \outer\parseargdef\appendixsection{\apphead1{#1}}
        !          5524: \def\appendixsectionzzz#1{%
        !          5525:   \global\subsecno=0 \global\subsubsecno=0  \global\advance\secno by 1
        !          5526:   \sectionheading{#1}{sec}{Yappendix}{\appendixletter.\the\secno}%
        !          5527: }
        !          5528: \let\appendixsec\appendixsection
        !          5529: 
        !          5530: % normally calls unnumberedseczzz:
        !          5531: \outer\parseargdef\unnumberedsec{\unnmhead1{#1}}
        !          5532: \def\unnumberedseczzz#1{%
        !          5533:   \global\subsecno=0 \global\subsubsecno=0  \global\advance\secno by 1
        !          5534:   \sectionheading{#1}{sec}{Ynothing}{\the\unnumberedno.\the\secno}%
        !          5535: }
        !          5536: 
        !          5537: % Subsections.
        !          5538: % 
        !          5539: % normally calls numberedsubseczzz:
        !          5540: \outer\parseargdef\numberedsubsec{\numhead2{#1}}
        !          5541: \def\numberedsubseczzz#1{%
        !          5542:   \global\subsubsecno=0  \global\advance\subsecno by 1
        !          5543:   \sectionheading{#1}{subsec}{Ynumbered}{\the\chapno.\the\secno.\the\subsecno}%
        !          5544: }
        !          5545: 
        !          5546: % normally calls appendixsubseczzz:
        !          5547: \outer\parseargdef\appendixsubsec{\apphead2{#1}}
        !          5548: \def\appendixsubseczzz#1{%
        !          5549:   \global\subsubsecno=0  \global\advance\subsecno by 1
        !          5550:   \sectionheading{#1}{subsec}{Yappendix}%
        !          5551:                  {\appendixletter.\the\secno.\the\subsecno}%
        !          5552: }
        !          5553: 
        !          5554: % normally calls unnumberedsubseczzz:
        !          5555: \outer\parseargdef\unnumberedsubsec{\unnmhead2{#1}}
        !          5556: \def\unnumberedsubseczzz#1{%
        !          5557:   \global\subsubsecno=0  \global\advance\subsecno by 1
        !          5558:   \sectionheading{#1}{subsec}{Ynothing}%
        !          5559:                  {\the\unnumberedno.\the\secno.\the\subsecno}%
        !          5560: }
        !          5561: 
        !          5562: % Subsubsections.
        !          5563: % 
        !          5564: % normally numberedsubsubseczzz:
        !          5565: \outer\parseargdef\numberedsubsubsec{\numhead3{#1}}
        !          5566: \def\numberedsubsubseczzz#1{%
        !          5567:   \global\advance\subsubsecno by 1
        !          5568:   \sectionheading{#1}{subsubsec}{Ynumbered}%
        !          5569:                  {\the\chapno.\the\secno.\the\subsecno.\the\subsubsecno}%
        !          5570: }
        !          5571: 
        !          5572: % normally appendixsubsubseczzz:
        !          5573: \outer\parseargdef\appendixsubsubsec{\apphead3{#1}}
        !          5574: \def\appendixsubsubseczzz#1{%
        !          5575:   \global\advance\subsubsecno by 1
        !          5576:   \sectionheading{#1}{subsubsec}{Yappendix}%
        !          5577:                  {\appendixletter.\the\secno.\the\subsecno.\the\subsubsecno}%
        !          5578: }
        !          5579: 
        !          5580: % normally unnumberedsubsubseczzz:
        !          5581: \outer\parseargdef\unnumberedsubsubsec{\unnmhead3{#1}}
        !          5582: \def\unnumberedsubsubseczzz#1{%
        !          5583:   \global\advance\subsubsecno by 1
        !          5584:   \sectionheading{#1}{subsubsec}{Ynothing}%
        !          5585:                  {\the\unnumberedno.\the\secno.\the\subsecno.\the\subsubsecno}%
        !          5586: }
        !          5587: 
        !          5588: % These macros control what the section commands do, according
        !          5589: % to what kind of chapter we are in (ordinary, appendix, or unnumbered).
        !          5590: % Define them by default for a numbered chapter.
        !          5591: \let\section = \numberedsec
        !          5592: \let\subsection = \numberedsubsec
        !          5593: \let\subsubsection = \numberedsubsubsec
        !          5594: 
        !          5595: % Define @majorheading, @heading and @subheading
        !          5596: 
        !          5597: \def\majorheading{%
        !          5598:   {\advance\chapheadingskip by 10pt \chapbreak }%
        !          5599:   \parsearg\chapheadingzzz
        !          5600: }
        !          5601: 
        !          5602: \def\chapheading{\chapbreak \parsearg\chapheadingzzz}
        !          5603: \def\chapheadingzzz#1{%
        !          5604:   \vbox{\chapfonts \raggedtitlesettings #1\par}%
        !          5605:   \nobreak\bigskip \nobreak
        !          5606:   \suppressfirstparagraphindent
        !          5607: }
        !          5608: 
        !          5609: % @heading, @subheading, @subsubheading.
        !          5610: \parseargdef\heading{\sectionheading{#1}{sec}{Yomitfromtoc}{}
        !          5611:   \suppressfirstparagraphindent}
        !          5612: \parseargdef\subheading{\sectionheading{#1}{subsec}{Yomitfromtoc}{}
        !          5613:   \suppressfirstparagraphindent}
        !          5614: \parseargdef\subsubheading{\sectionheading{#1}{subsubsec}{Yomitfromtoc}{}
        !          5615:   \suppressfirstparagraphindent}
        !          5616: 
        !          5617: % These macros generate a chapter, section, etc. heading only
        !          5618: % (including whitespace, linebreaking, etc. around it),
        !          5619: % given all the information in convenient, parsed form.
        !          5620: 
        !          5621: % Args are the skip and penalty (usually negative)
        !          5622: \def\dobreak#1#2{\par\ifdim\lastskip<#1\removelastskip\penalty#2\vskip#1\fi}
        !          5623: 
        !          5624: % Parameter controlling skip before chapter headings (if needed)
        !          5625: \newskip\chapheadingskip
        !          5626: 
        !          5627: % Define plain chapter starts, and page on/off switching for it.
        !          5628: \def\chapbreak{\dobreak \chapheadingskip {-4000}}
        !          5629: \def\chappager{\par\vfill\supereject}
        !          5630: % Because \domark is called before \chapoddpage, the filler page will
        !          5631: % get the headings for the next chapter, which is wrong.  But we don't
        !          5632: % care -- we just disable all headings on the filler page.
        !          5633: \def\chapoddpage{%
        !          5634:   \chappager
        !          5635:   \ifodd\pageno \else
        !          5636:     \begingroup
        !          5637:       \headingsoff
        !          5638:       \null
        !          5639:       \chappager
        !          5640:     \endgroup
        !          5641:   \fi
        !          5642: }
        !          5643: 
        !          5644: \def\setchapternewpage #1 {\csname CHAPPAG#1\endcsname}
        !          5645: 
        !          5646: \def\CHAPPAGoff{%
        !          5647: \global\let\contentsalignmacro = \chappager
        !          5648: \global\let\pchapsepmacro=\chapbreak
        !          5649: \global\let\pagealignmacro=\chappager}
        !          5650: 
        !          5651: \def\CHAPPAGon{%
        !          5652: \global\let\contentsalignmacro = \chappager
        !          5653: \global\let\pchapsepmacro=\chappager
        !          5654: \global\let\pagealignmacro=\chappager
        !          5655: \global\def\HEADINGSon{\HEADINGSsingle}}
        !          5656: 
        !          5657: \def\CHAPPAGodd{%
        !          5658: \global\let\contentsalignmacro = \chapoddpage
        !          5659: \global\let\pchapsepmacro=\chapoddpage
        !          5660: \global\let\pagealignmacro=\chapoddpage
        !          5661: \global\def\HEADINGSon{\HEADINGSdouble}}
        !          5662: 
        !          5663: \CHAPPAGon
        !          5664: 
        !          5665: % Chapter opening.
        !          5666: %
        !          5667: % #1 is the text, #2 is the section type (Ynumbered, Ynothing,
        !          5668: % Yappendix, Yomitfromtoc), #3 the chapter number.
        !          5669: %
        !          5670: % To test against our argument.
        !          5671: \def\Ynothingkeyword{Ynothing}
        !          5672: \def\Yomitfromtockeyword{Yomitfromtoc}
        !          5673: \def\Yappendixkeyword{Yappendix}
        !          5674: %
        !          5675: \def\chapmacro#1#2#3{%
        !          5676:   % Insert the first mark before the heading break (see notes for \domark).
        !          5677:   \let\prevchapterdefs=\lastchapterdefs
        !          5678:   \let\prevsectiondefs=\lastsectiondefs
        !          5679:   \gdef\lastsectiondefs{\gdef\thissectionname{}\gdef\thissectionnum{}%
        !          5680:                         \gdef\thissection{}}%
        !          5681:   %
        !          5682:   \def\temptype{#2}%
        !          5683:   \ifx\temptype\Ynothingkeyword
        !          5684:     \gdef\lastchapterdefs{\gdef\thischaptername{#1}\gdef\thischapternum{}%
        !          5685:                           \gdef\thischapter{\thischaptername}}%
        !          5686:   \else\ifx\temptype\Yomitfromtockeyword
        !          5687:     \gdef\lastchapterdefs{\gdef\thischaptername{#1}\gdef\thischapternum{}%
        !          5688:                           \gdef\thischapter{}}%
        !          5689:   \else\ifx\temptype\Yappendixkeyword
        !          5690:     \toks0={#1}%
        !          5691:     \xdef\lastchapterdefs{%
        !          5692:       \gdef\noexpand\thischaptername{\the\toks0}%
        !          5693:       \gdef\noexpand\thischapternum{\appendixletter}%
        !          5694:       % \noexpand\putwordAppendix avoids expanding indigestible
        !          5695:       % commands in some of the translations.
        !          5696:       \gdef\noexpand\thischapter{\noexpand\putwordAppendix{}
        !          5697:                                  \noexpand\thischapternum:
        !          5698:                                  \noexpand\thischaptername}%
        !          5699:     }%
        !          5700:   \else
        !          5701:     \toks0={#1}%
        !          5702:     \xdef\lastchapterdefs{%
        !          5703:       \gdef\noexpand\thischaptername{\the\toks0}%
        !          5704:       \gdef\noexpand\thischapternum{\the\chapno}%
        !          5705:       % \noexpand\putwordChapter avoids expanding indigestible
        !          5706:       % commands in some of the translations.
        !          5707:       \gdef\noexpand\thischapter{\noexpand\putwordChapter{}
        !          5708:                                  \noexpand\thischapternum:
        !          5709:                                  \noexpand\thischaptername}%
        !          5710:     }%
        !          5711:   \fi\fi\fi
        !          5712:   %
        !          5713:   % Output the mark.  Pass it through \safewhatsit, to take care of
        !          5714:   % the preceding space.
        !          5715:   \safewhatsit\domark
        !          5716:   %
        !          5717:   % Insert the chapter heading break.
        !          5718:   \pchapsepmacro
        !          5719:   %
        !          5720:   % Now the second mark, after the heading break.  No break points
        !          5721:   % between here and the heading.
        !          5722:   \let\prevchapterdefs=\lastchapterdefs
        !          5723:   \let\prevsectiondefs=\lastsectiondefs
        !          5724:   \domark
        !          5725:   %
        !          5726:   {%
        !          5727:     \chapfonts \rmisbold
        !          5728:     %
        !          5729:     % Have to define \lastsection before calling \donoderef, because the
        !          5730:     % xref code eventually uses it.  On the other hand, it has to be called
        !          5731:     % after \pchapsepmacro, or the headline will change too soon.
        !          5732:     \gdef\lastsection{#1}%
        !          5733:     %
        !          5734:     % Only insert the separating space if we have a chapter/appendix
        !          5735:     % number, and don't print the unnumbered ``number''.
        !          5736:     \ifx\temptype\Ynothingkeyword
        !          5737:       \setbox0 = \hbox{}%
        !          5738:       \def\toctype{unnchap}%
        !          5739:     \else\ifx\temptype\Yomitfromtockeyword
        !          5740:       \setbox0 = \hbox{}% contents like unnumbered, but no toc entry
        !          5741:       \def\toctype{omit}%
        !          5742:     \else\ifx\temptype\Yappendixkeyword
        !          5743:       \setbox0 = \hbox{\putwordAppendix{} #3\enspace}%
        !          5744:       \def\toctype{app}%
        !          5745:     \else
        !          5746:       \setbox0 = \hbox{#3\enspace}%
        !          5747:       \def\toctype{numchap}%
        !          5748:     \fi\fi\fi
        !          5749:     %
        !          5750:     % Write the toc entry for this chapter.  Must come before the
        !          5751:     % \donoderef, because we include the current node name in the toc
        !          5752:     % entry, and \donoderef resets it to empty.
        !          5753:     \writetocentry{\toctype}{#1}{#3}%
        !          5754:     %
        !          5755:     % For pdftex, we have to write out the node definition (aka, make
        !          5756:     % the pdfdest) after any page break, but before the actual text has
        !          5757:     % been typeset.  If the destination for the pdf outline is after the
        !          5758:     % text, then jumping from the outline may wind up with the text not
        !          5759:     % being visible, for instance under high magnification.
        !          5760:     \donoderef{#2}%
        !          5761:     %
        !          5762:     % Typeset the actual heading.
        !          5763:     \nobreak % Avoid page breaks at the interline glue.
        !          5764:     \vbox{\raggedtitlesettings \hangindent=\wd0 \centerparametersmaybe
        !          5765:           \unhbox0 #1\par}%
        !          5766:   }%
        !          5767:   \nobreak\bigskip % no page break after a chapter title
        !          5768:   \nobreak
        !          5769: }
        !          5770: 
        !          5771: % @centerchap -- centered and unnumbered.
        !          5772: \let\centerparametersmaybe = \relax
        !          5773: \def\centerparameters{%
        !          5774:   \advance\rightskip by 3\rightskip
        !          5775:   \leftskip = \rightskip
        !          5776:   \parfillskip = 0pt
        !          5777: }
        !          5778: 
        !          5779: 
        !          5780: % I don't think this chapter style is supported any more, so I'm not
        !          5781: % updating it with the new noderef stuff.  We'll see.  --karl, 11aug03.
        !          5782: %
        !          5783: \def\setchapterstyle #1 {\csname CHAPF#1\endcsname}
        !          5784: %
        !          5785: \def\unnchfopen #1{%
        !          5786:   \chapoddpage
        !          5787:   \vbox{\chapfonts \raggedtitlesettings #1\par}%
        !          5788:   \nobreak\bigskip\nobreak
        !          5789: }
        !          5790: \def\chfopen #1#2{\chapoddpage {\chapfonts
        !          5791: \vbox to 3in{\vfil \hbox to\hsize{\hfil #2} \hbox to\hsize{\hfil #1} \vfil}}%
        !          5792: \par\penalty 5000 %
        !          5793: }
        !          5794: \def\centerchfopen #1{%
        !          5795:   \chapoddpage
        !          5796:   \vbox{\chapfonts \raggedtitlesettings \hfill #1\hfill}%
        !          5797:   \nobreak\bigskip \nobreak
        !          5798: }
        !          5799: \def\CHAPFopen{%
        !          5800:   \global\let\chapmacro=\chfopen
        !          5801:   \global\let\centerchapmacro=\centerchfopen}
        !          5802: 
        !          5803: 
        !          5804: % Section titles.  These macros combine the section number parts and
        !          5805: % call the generic \sectionheading to do the printing.
        !          5806: %
        !          5807: \newskip\secheadingskip
        !          5808: \def\secheadingbreak{\dobreak \secheadingskip{-1000}}
        !          5809: 
        !          5810: % Subsection titles.
        !          5811: \newskip\subsecheadingskip
        !          5812: \def\subsecheadingbreak{\dobreak \subsecheadingskip{-500}}
        !          5813: 
        !          5814: % Subsubsection titles.
        !          5815: \def\subsubsecheadingskip{\subsecheadingskip}
        !          5816: \def\subsubsecheadingbreak{\subsecheadingbreak}
        !          5817: 
        !          5818: 
        !          5819: % Print any size, any type, section title.
        !          5820: %
        !          5821: % #1 is the text, #2 is the section level (sec/subsec/subsubsec), #3 is
        !          5822: % the section type for xrefs (Ynumbered, Ynothing, Yappendix), #4 is the
        !          5823: % section number.
        !          5824: %
        !          5825: \def\seckeyword{sec}
        !          5826: %
        !          5827: \def\sectionheading#1#2#3#4{%
        !          5828:   {%
        !          5829:     \checkenv{}% should not be in an environment.
        !          5830:     %
        !          5831:     % Switch to the right set of fonts.
        !          5832:     \csname #2fonts\endcsname \rmisbold
        !          5833:     %
        !          5834:     \def\sectionlevel{#2}%
        !          5835:     \def\temptype{#3}%
        !          5836:     %
        !          5837:     % Insert first mark before the heading break (see notes for \domark).
        !          5838:     \let\prevsectiondefs=\lastsectiondefs
        !          5839:     \ifx\temptype\Ynothingkeyword
        !          5840:       \ifx\sectionlevel\seckeyword
        !          5841:         \gdef\lastsectiondefs{\gdef\thissectionname{#1}\gdef\thissectionnum{}%
        !          5842:                               \gdef\thissection{\thissectionname}}%
        !          5843:       \fi
        !          5844:     \else\ifx\temptype\Yomitfromtockeyword
        !          5845:       % Don't redefine \thissection.
        !          5846:     \else\ifx\temptype\Yappendixkeyword
        !          5847:       \ifx\sectionlevel\seckeyword
        !          5848:         \toks0={#1}%
        !          5849:         \xdef\lastsectiondefs{%
        !          5850:           \gdef\noexpand\thissectionname{\the\toks0}%
        !          5851:           \gdef\noexpand\thissectionnum{#4}%
        !          5852:           % \noexpand\putwordSection avoids expanding indigestible
        !          5853:           % commands in some of the translations.
        !          5854:           \gdef\noexpand\thissection{\noexpand\putwordSection{}
        !          5855:                                      \noexpand\thissectionnum:
        !          5856:                                      \noexpand\thissectionname}%
        !          5857:         }%
        !          5858:       \fi
        !          5859:     \else
        !          5860:       \ifx\sectionlevel\seckeyword
        !          5861:         \toks0={#1}%
        !          5862:         \xdef\lastsectiondefs{%
        !          5863:           \gdef\noexpand\thissectionname{\the\toks0}%
        !          5864:           \gdef\noexpand\thissectionnum{#4}%
        !          5865:           % \noexpand\putwordSection avoids expanding indigestible
        !          5866:           % commands in some of the translations.
        !          5867:           \gdef\noexpand\thissection{\noexpand\putwordSection{}
        !          5868:                                      \noexpand\thissectionnum:
        !          5869:                                      \noexpand\thissectionname}%
        !          5870:         }%
        !          5871:       \fi
        !          5872:     \fi\fi\fi
        !          5873:     %
        !          5874:     % Go into vertical mode.  Usually we'll already be there, but we
        !          5875:     % don't want the following whatsit to end up in a preceding paragraph
        !          5876:     % if the document didn't happen to have a blank line.
        !          5877:     \par
        !          5878:     %
        !          5879:     % Output the mark.  Pass it through \safewhatsit, to take care of
        !          5880:     % the preceding space.
        !          5881:     \safewhatsit\domark
        !          5882:     %
        !          5883:     % Insert space above the heading.
        !          5884:     \csname #2headingbreak\endcsname
        !          5885:     %
        !          5886:     % Now the second mark, after the heading break.  No break points
        !          5887:     % between here and the heading.
        !          5888:     \let\prevsectiondefs=\lastsectiondefs
        !          5889:     \domark
        !          5890:     %
        !          5891:     % Only insert the space after the number if we have a section number.
        !          5892:     \ifx\temptype\Ynothingkeyword
        !          5893:       \setbox0 = \hbox{}%
        !          5894:       \def\toctype{unn}%
        !          5895:       \gdef\lastsection{#1}%
        !          5896:     \else\ifx\temptype\Yomitfromtockeyword
        !          5897:       % for @headings -- no section number, don't include in toc,
        !          5898:       % and don't redefine \lastsection.
        !          5899:       \setbox0 = \hbox{}%
        !          5900:       \def\toctype{omit}%
        !          5901:       \let\sectionlevel=\empty
        !          5902:     \else\ifx\temptype\Yappendixkeyword
        !          5903:       \setbox0 = \hbox{#4\enspace}%
        !          5904:       \def\toctype{app}%
        !          5905:       \gdef\lastsection{#1}%
        !          5906:     \else
        !          5907:       \setbox0 = \hbox{#4\enspace}%
        !          5908:       \def\toctype{num}%
        !          5909:       \gdef\lastsection{#1}%
        !          5910:     \fi\fi\fi
        !          5911:     %
        !          5912:     % Write the toc entry (before \donoderef).  See comments in \chapmacro.
        !          5913:     \writetocentry{\toctype\sectionlevel}{#1}{#4}%
        !          5914:     %
        !          5915:     % Write the node reference (= pdf destination for pdftex).
        !          5916:     % Again, see comments in \chapmacro.
        !          5917:     \donoderef{#3}%
        !          5918:     %
        !          5919:     % Interline glue will be inserted when the vbox is completed.
        !          5920:     % That glue will be a valid breakpoint for the page, since it'll be
        !          5921:     % preceded by a whatsit (usually from the \donoderef, or from the
        !          5922:     % \writetocentry if there was no node).  We don't want to allow that
        !          5923:     % break, since then the whatsits could end up on page n while the
        !          5924:     % section is on page n+1, thus toc/etc. are wrong.  Debian bug 276000.
        !          5925:     \nobreak
        !          5926:     %
        !          5927:     % Output the actual section heading.
        !          5928:     \vbox{\hyphenpenalty=10000 \tolerance=5000 \parindent=0pt \ptexraggedright
        !          5929:           \hangindent=\wd0  % zero if no section number
        !          5930:           \unhbox0 #1}%
        !          5931:   }%
        !          5932:   % Add extra space after the heading -- half of whatever came above it.
        !          5933:   % Don't allow stretch, though.
        !          5934:   \kern .5 \csname #2headingskip\endcsname
        !          5935:   %
        !          5936:   % Do not let the kern be a potential breakpoint, as it would be if it
        !          5937:   % was followed by glue.
        !          5938:   \nobreak
        !          5939:   %
        !          5940:   % We'll almost certainly start a paragraph next, so don't let that
        !          5941:   % glue accumulate.  (Not a breakpoint because it's preceded by a
        !          5942:   % discardable item.)  However, when a paragraph is not started next
        !          5943:   % (\startdefun, \cartouche, \center, etc.), this needs to be wiped out
        !          5944:   % or the negative glue will cause weirdly wrong output, typically
        !          5945:   % obscuring the section heading with something else.
        !          5946:   \vskip-\parskip
        !          5947:   %
        !          5948:   % This is so the last item on the main vertical list is a known
        !          5949:   % \penalty > 10000, so \startdefun, etc., can recognize the situation
        !          5950:   % and do the needful.
        !          5951:   \penalty 10001
        !          5952: }
        !          5953: 
        !          5954: 
        !          5955: \message{toc,}
        !          5956: % Table of contents.
        !          5957: \newwrite\tocfile
        !          5958: 
        !          5959: % Write an entry to the toc file, opening it if necessary.
        !          5960: % Called from @chapter, etc.
        !          5961: %
        !          5962: % Example usage: \writetocentry{sec}{Section Name}{\the\chapno.\the\secno}
        !          5963: % We append the current node name (if any) and page number as additional
        !          5964: % arguments for the \{chap,sec,...}entry macros which will eventually
        !          5965: % read this.  The node name is used in the pdf outlines as the
        !          5966: % destination to jump to.
        !          5967: %
        !          5968: % We open the .toc file for writing here instead of at @setfilename (or
        !          5969: % any other fixed time) so that @contents can be anywhere in the document.
        !          5970: % But if #1 is `omit', then we don't do anything.  This is used for the
        !          5971: % table of contents chapter openings themselves.
        !          5972: %
        !          5973: \newif\iftocfileopened
        !          5974: \def\omitkeyword{omit}%
        !          5975: %
        !          5976: \def\writetocentry#1#2#3{%
        !          5977:   \edef\writetoctype{#1}%
        !          5978:   \ifx\writetoctype\omitkeyword \else
        !          5979:     \iftocfileopened\else
        !          5980:       \immediate\openout\tocfile = \jobname.toc
        !          5981:       \global\tocfileopenedtrue
        !          5982:     \fi
        !          5983:     %
        !          5984:     \iflinks
        !          5985:       {\atdummies
        !          5986:        \edef\temp{%
        !          5987:          \write\tocfile{@#1entry{#2}{#3}{\lastnode}{\noexpand\folio}}}%
        !          5988:        \temp
        !          5989:       }%
        !          5990:     \fi
        !          5991:   \fi
        !          5992:   %
        !          5993:   % Tell \shipout to create a pdf destination on each page, if we're
        !          5994:   % writing pdf.  These are used in the table of contents.  We can't
        !          5995:   % just write one on every page because the title pages are numbered
        !          5996:   % 1 and 2 (the page numbers aren't printed), and so are the first
        !          5997:   % two pages of the document.  Thus, we'd have two destinations named
        !          5998:   % `1', and two named `2'.
        !          5999:   \ifpdf \global\pdfmakepagedesttrue \fi
        !          6000: }
        !          6001: 
        !          6002: 
        !          6003: % These characters do not print properly in the Computer Modern roman
        !          6004: % fonts, so we must take special care.  This is more or less redundant
        !          6005: % with the Texinfo input format setup at the end of this file.
        !          6006: %
        !          6007: \def\activecatcodes{%
        !          6008:   \catcode`\"=\active
        !          6009:   \catcode`\$=\active
        !          6010:   \catcode`\<=\active
        !          6011:   \catcode`\>=\active
        !          6012:   \catcode`\\=\active
        !          6013:   \catcode`\^=\active
        !          6014:   \catcode`\_=\active
        !          6015:   \catcode`\|=\active
        !          6016:   \catcode`\~=\active
        !          6017: }
        !          6018: 
        !          6019: 
        !          6020: % Read the toc file, which is essentially Texinfo input.
        !          6021: \def\readtocfile{%
        !          6022:   \setupdatafile
        !          6023:   \activecatcodes
        !          6024:   \input \tocreadfilename
        !          6025: }
        !          6026: 
        !          6027: \newskip\contentsrightmargin \contentsrightmargin=1in
        !          6028: \newcount\savepageno
        !          6029: \newcount\lastnegativepageno \lastnegativepageno = -1
        !          6030: 
        !          6031: % Prepare to read what we've written to \tocfile.
        !          6032: %
        !          6033: \def\startcontents#1{%
        !          6034:   % If @setchapternewpage on, and @headings double, the contents should
        !          6035:   % start on an odd page, unlike chapters.  Thus, we maintain
        !          6036:   % \contentsalignmacro in parallel with \pagealignmacro.
        !          6037:   % From: Torbjorn Granlund <[email protected]>
        !          6038:   \contentsalignmacro
        !          6039:   \immediate\closeout\tocfile
        !          6040:   %
        !          6041:   % Don't need to put `Contents' or `Short Contents' in the headline.
        !          6042:   % It is abundantly clear what they are.
        !          6043:   \chapmacro{#1}{Yomitfromtoc}{}%
        !          6044:   %
        !          6045:   \savepageno = \pageno
        !          6046:   \begingroup                  % Set up to handle contents files properly.
        !          6047:     \raggedbottom              % Worry more about breakpoints than the bottom.
        !          6048:     \advance\hsize by -\contentsrightmargin % Don't use the full line length.
        !          6049:     %
        !          6050:     % Roman numerals for page numbers.
        !          6051:     \ifnum \pageno>0 \global\pageno = \lastnegativepageno \fi
        !          6052: }
        !          6053: 
        !          6054: % redefined for the two-volume lispref.  We always output on
        !          6055: % \jobname.toc even if this is redefined.
        !          6056: %
        !          6057: \def\tocreadfilename{\jobname.toc}
        !          6058: 
        !          6059: % Normal (long) toc.
        !          6060: %
        !          6061: \def\contents{%
        !          6062:   \startcontents{\putwordTOC}%
        !          6063:     \openin 1 \tocreadfilename\space
        !          6064:     \ifeof 1 \else
        !          6065:       \readtocfile
        !          6066:     \fi
        !          6067:     \vfill \eject
        !          6068:     \contentsalignmacro % in case @setchapternewpage odd is in effect
        !          6069:     \ifeof 1 \else
        !          6070:       \pdfmakeoutlines
        !          6071:     \fi
        !          6072:     \closein 1
        !          6073:   \endgroup
        !          6074:   \lastnegativepageno = \pageno
        !          6075:   \global\pageno = \savepageno
        !          6076: }
        !          6077: 
        !          6078: % And just the chapters.
        !          6079: \def\summarycontents{%
        !          6080:   \startcontents{\putwordShortTOC}%
        !          6081:     %
        !          6082:     \let\partentry = \shortpartentry
        !          6083:     \let\numchapentry = \shortchapentry
        !          6084:     \let\appentry = \shortchapentry
        !          6085:     \let\unnchapentry = \shortunnchapentry
        !          6086:     % We want a true roman here for the page numbers.
        !          6087:     \secfonts
        !          6088:     \let\rm=\shortcontrm \let\bf=\shortcontbf
        !          6089:     \let\sl=\shortcontsl \let\tt=\shortconttt
        !          6090:     \rm
        !          6091:     \hyphenpenalty = 10000
        !          6092:     \advance\baselineskip by 1pt % Open it up a little.
        !          6093:     \def\numsecentry##1##2##3##4{}
        !          6094:     \let\appsecentry = \numsecentry
        !          6095:     \let\unnsecentry = \numsecentry
        !          6096:     \let\numsubsecentry = \numsecentry
        !          6097:     \let\appsubsecentry = \numsecentry
        !          6098:     \let\unnsubsecentry = \numsecentry
        !          6099:     \let\numsubsubsecentry = \numsecentry
        !          6100:     \let\appsubsubsecentry = \numsecentry
        !          6101:     \let\unnsubsubsecentry = \numsecentry
        !          6102:     \openin 1 \tocreadfilename\space
        !          6103:     \ifeof 1 \else
        !          6104:       \readtocfile
        !          6105:     \fi
        !          6106:     \closein 1
        !          6107:     \vfill \eject
        !          6108:     \contentsalignmacro % in case @setchapternewpage odd is in effect
        !          6109:   \endgroup
        !          6110:   \lastnegativepageno = \pageno
        !          6111:   \global\pageno = \savepageno
        !          6112: }
        !          6113: \let\shortcontents = \summarycontents
        !          6114: 
        !          6115: % Typeset the label for a chapter or appendix for the short contents.
        !          6116: % The arg is, e.g., `A' for an appendix, or `3' for a chapter.
        !          6117: %
        !          6118: \def\shortchaplabel#1{%
        !          6119:   % This space should be enough, since a single number is .5em, and the
        !          6120:   % widest letter (M) is 1em, at least in the Computer Modern fonts.
        !          6121:   % But use \hss just in case.
        !          6122:   % (This space doesn't include the extra space that gets added after
        !          6123:   % the label; that gets put in by \shortchapentry above.)
        !          6124:   %
        !          6125:   % We'd like to right-justify chapter numbers, but that looks strange
        !          6126:   % with appendix letters.  And right-justifying numbers and
        !          6127:   % left-justifying letters looks strange when there is less than 10
        !          6128:   % chapters.  Have to read the whole toc once to know how many chapters
        !          6129:   % there are before deciding ...
        !          6130:   \hbox to 1em{#1\hss}%
        !          6131: }
        !          6132: 
        !          6133: % These macros generate individual entries in the table of contents.
        !          6134: % The first argument is the chapter or section name.
        !          6135: % The last argument is the page number.
        !          6136: % The arguments in between are the chapter number, section number, ...
        !          6137: 
        !          6138: % Parts, in the main contents.  Replace the part number, which doesn't
        !          6139: % exist, with an empty box.  Let's hope all the numbers have the same width.
        !          6140: % Also ignore the page number, which is conventionally not printed.
        !          6141: \def\numeralbox{\setbox0=\hbox{8}\hbox to \wd0{\hfil}}
        !          6142: \def\partentry#1#2#3#4{\dochapentry{\numeralbox\labelspace#1}{}}
        !          6143: %
        !          6144: % Parts, in the short toc.
        !          6145: \def\shortpartentry#1#2#3#4{%
        !          6146:   \penalty-300
        !          6147:   \vskip.5\baselineskip plus.15\baselineskip minus.1\baselineskip
        !          6148:   \shortchapentry{{\bf #1}}{\numeralbox}{}{}%
        !          6149: }
        !          6150: 
        !          6151: % Chapters, in the main contents.
        !          6152: \def\numchapentry#1#2#3#4{\dochapentry{#2\labelspace#1}{#4}}
        !          6153: %
        !          6154: % Chapters, in the short toc.
        !          6155: % See comments in \dochapentry re vbox and related settings.
        !          6156: \def\shortchapentry#1#2#3#4{%
        !          6157:   \tocentry{\shortchaplabel{#2}\labelspace #1}{\doshortpageno\bgroup#4\egroup}%
        !          6158: }
        !          6159: 
        !          6160: % Appendices, in the main contents.
        !          6161: % Need the word Appendix, and a fixed-size box.
        !          6162: %
        !          6163: \def\appendixbox#1{%
        !          6164:   % We use M since it's probably the widest letter.
        !          6165:   \setbox0 = \hbox{\putwordAppendix{} M}%
        !          6166:   \hbox to \wd0{\putwordAppendix{} #1\hss}}
        !          6167: %
        !          6168: \def\appentry#1#2#3#4{\dochapentry{\appendixbox{#2}\labelspace#1}{#4}}
        !          6169: 
        !          6170: % Unnumbered chapters.
        !          6171: \def\unnchapentry#1#2#3#4{\dochapentry{#1}{#4}}
        !          6172: \def\shortunnchapentry#1#2#3#4{\tocentry{#1}{\doshortpageno\bgroup#4\egroup}}
        !          6173: 
        !          6174: % Sections.
        !          6175: \def\numsecentry#1#2#3#4{\dosecentry{#2\labelspace#1}{#4}}
        !          6176: \let\appsecentry=\numsecentry
        !          6177: \def\unnsecentry#1#2#3#4{\dosecentry{#1}{#4}}
        !          6178: 
        !          6179: % Subsections.
        !          6180: \def\numsubsecentry#1#2#3#4{\dosubsecentry{#2\labelspace#1}{#4}}
        !          6181: \let\appsubsecentry=\numsubsecentry
        !          6182: \def\unnsubsecentry#1#2#3#4{\dosubsecentry{#1}{#4}}
        !          6183: 
        !          6184: % And subsubsections.
        !          6185: \def\numsubsubsecentry#1#2#3#4{\dosubsubsecentry{#2\labelspace#1}{#4}}
        !          6186: \let\appsubsubsecentry=\numsubsubsecentry
        !          6187: \def\unnsubsubsecentry#1#2#3#4{\dosubsubsecentry{#1}{#4}}
        !          6188: 
        !          6189: % This parameter controls the indentation of the various levels.
        !          6190: % Same as \defaultparindent.
        !          6191: \newdimen\tocindent \tocindent = 15pt
        !          6192: 
        !          6193: % Now for the actual typesetting. In all these, #1 is the text and #2 is the
        !          6194: % page number.
        !          6195: %
        !          6196: % If the toc has to be broken over pages, we want it to be at chapters
        !          6197: % if at all possible; hence the \penalty.
        !          6198: \def\dochapentry#1#2{%
        !          6199:    \penalty-300 \vskip1\baselineskip plus.33\baselineskip minus.25\baselineskip
        !          6200:    \begingroup
        !          6201:      \chapentryfonts
        !          6202:      \tocentry{#1}{\dopageno\bgroup#2\egroup}%
        !          6203:    \endgroup
        !          6204:    \nobreak\vskip .25\baselineskip plus.1\baselineskip
        !          6205: }
        !          6206: 
        !          6207: \def\dosecentry#1#2{\begingroup
        !          6208:   \secentryfonts \leftskip=\tocindent
        !          6209:   \tocentry{#1}{\dopageno\bgroup#2\egroup}%
        !          6210: \endgroup}
        !          6211: 
        !          6212: \def\dosubsecentry#1#2{\begingroup
        !          6213:   \subsecentryfonts \leftskip=2\tocindent
        !          6214:   \tocentry{#1}{\dopageno\bgroup#2\egroup}%
        !          6215: \endgroup}
        !          6216: 
        !          6217: \def\dosubsubsecentry#1#2{\begingroup
        !          6218:   \subsubsecentryfonts \leftskip=3\tocindent
        !          6219:   \tocentry{#1}{\dopageno\bgroup#2\egroup}%
        !          6220: \endgroup}
        !          6221: 
        !          6222: % We use the same \entry macro as for the index entries.
        !          6223: \let\tocentry = \entry
        !          6224: 
        !          6225: % Space between chapter (or whatever) number and the title.
        !          6226: \def\labelspace{\hskip1em \relax}
        !          6227: 
        !          6228: \def\dopageno#1{{\rm #1}}
        !          6229: \def\doshortpageno#1{{\rm #1}}
        !          6230: 
        !          6231: \def\chapentryfonts{\secfonts \rm}
        !          6232: \def\secentryfonts{\textfonts}
        !          6233: \def\subsecentryfonts{\textfonts}
        !          6234: \def\subsubsecentryfonts{\textfonts}
        !          6235: 
        !          6236: 
        !          6237: \message{environments,}
        !          6238: % @foo ... @end foo.
        !          6239: 
        !          6240: % @tex ... @end tex    escapes into raw TeX temporarily.
        !          6241: % One exception: @ is still an escape character, so that @end tex works.
        !          6242: % But \@ or @@ will get a plain @ character.
        !          6243: 
        !          6244: \envdef\tex{%
        !          6245:   \setupmarkupstyle{tex}%
        !          6246:   \catcode `\\=0 \catcode `\{=1 \catcode `\}=2
        !          6247:   \catcode `\$=3 \catcode `\&=4 \catcode `\#=6
        !          6248:   \catcode `\^=7 \catcode `\_=8 \catcode `\~=\active \let~=\tie
        !          6249:   \catcode `\%=14
        !          6250:   \catcode `\+=\other
        !          6251:   \catcode `\"=\other
        !          6252:   \catcode `\|=\other
        !          6253:   \catcode `\<=\other
        !          6254:   \catcode `\>=\other
        !          6255:   \catcode`\`=\other
        !          6256:   \catcode`\'=\other
        !          6257:   \escapechar=`\\
        !          6258:   %
        !          6259:   % ' is active in math mode (mathcode"8000).  So reset it, and all our
        !          6260:   % other math active characters (just in case), to plain's definitions.
        !          6261:   \mathactive
        !          6262:   %
        !          6263:   \let\b=\ptexb
        !          6264:   \let\bullet=\ptexbullet
        !          6265:   \let\c=\ptexc
        !          6266:   \let\,=\ptexcomma
        !          6267:   \let\.=\ptexdot
        !          6268:   \let\dots=\ptexdots
        !          6269:   \let\equiv=\ptexequiv
        !          6270:   \let\!=\ptexexclam
        !          6271:   \let\i=\ptexi
        !          6272:   \let\indent=\ptexindent
        !          6273:   \let\noindent=\ptexnoindent
        !          6274:   \let\{=\ptexlbrace
        !          6275:   \let\+=\tabalign
        !          6276:   \let\}=\ptexrbrace
        !          6277:   \let\/=\ptexslash
        !          6278:   \let\*=\ptexstar
        !          6279:   \let\t=\ptext
        !          6280:   \expandafter \let\csname top\endcsname=\ptextop  % outer
        !          6281:   \let\frenchspacing=\plainfrenchspacing
        !          6282:   %
        !          6283:   \def\endldots{\mathinner{\ldots\ldots\ldots\ldots}}%
        !          6284:   \def\enddots{\relax\ifmmode\endldots\else$\mathsurround=0pt \endldots\,$\fi}%
        !          6285:   \def\@{@}%
        !          6286: }
        !          6287: % There is no need to define \Etex.
        !          6288: 
        !          6289: % Define @lisp ... @end lisp.
        !          6290: % @lisp environment forms a group so it can rebind things,
        !          6291: % including the definition of @end lisp (which normally is erroneous).
        !          6292: 
        !          6293: % Amount to narrow the margins by for @lisp.
        !          6294: \newskip\lispnarrowing \lispnarrowing=0.4in
        !          6295: 
        !          6296: % This is the definition that ^^M gets inside @lisp, @example, and other
        !          6297: % such environments.  \null is better than a space, since it doesn't
        !          6298: % have any width.
        !          6299: \def\lisppar{\null\endgraf}
        !          6300: 
        !          6301: % This space is always present above and below environments.
        !          6302: \newskip\envskipamount \envskipamount = 0pt
        !          6303: 
        !          6304: % Make spacing and below environment symmetrical.  We use \parskip here
        !          6305: % to help in doing that, since in @example-like environments \parskip
        !          6306: % is reset to zero; thus the \afterenvbreak inserts no space -- but the
        !          6307: % start of the next paragraph will insert \parskip.
        !          6308: %
        !          6309: \def\aboveenvbreak{{%
        !          6310:   % =10000 instead of <10000 because of a special case in \itemzzz and
        !          6311:   % \sectionheading, q.v.
        !          6312:   \ifnum \lastpenalty=10000 \else
        !          6313:     \advance\envskipamount by \parskip
        !          6314:     \endgraf
        !          6315:     \ifdim\lastskip<\envskipamount
        !          6316:       \removelastskip
        !          6317:       % it's not a good place to break if the last penalty was \nobreak
        !          6318:       % or better ...
        !          6319:       \ifnum\lastpenalty<10000 \penalty-50 \fi
        !          6320:       \vskip\envskipamount
        !          6321:     \fi
        !          6322:   \fi
        !          6323: }}
        !          6324: 
        !          6325: \let\afterenvbreak = \aboveenvbreak
        !          6326: 
        !          6327: % \nonarrowing is a flag.  If "set", @lisp etc don't narrow margins; it will
        !          6328: % also clear it, so that its embedded environments do the narrowing again.
        !          6329: \let\nonarrowing=\relax
        !          6330: 
        !          6331: % @cartouche ... @end cartouche: draw rectangle w/rounded corners around
        !          6332: % environment contents.
        !          6333: \font\circle=lcircle10
        !          6334: \newdimen\circthick
        !          6335: \newdimen\cartouter\newdimen\cartinner
        !          6336: \newskip\normbskip\newskip\normpskip\newskip\normlskip
        !          6337: \circthick=\fontdimen8\circle
        !          6338: %
        !          6339: \def\ctl{{\circle\char'013\hskip -6pt}}% 6pt from pl file: 1/2charwidth
        !          6340: \def\ctr{{\hskip 6pt\circle\char'010}}
        !          6341: \def\cbl{{\circle\char'012\hskip -6pt}}
        !          6342: \def\cbr{{\hskip 6pt\circle\char'011}}
        !          6343: \def\carttop{\hbox to \cartouter{\hskip\lskip
        !          6344:         \ctl\leaders\hrule height\circthick\hfil\ctr
        !          6345:         \hskip\rskip}}
        !          6346: \def\cartbot{\hbox to \cartouter{\hskip\lskip
        !          6347:         \cbl\leaders\hrule height\circthick\hfil\cbr
        !          6348:         \hskip\rskip}}
        !          6349: %
        !          6350: \newskip\lskip\newskip\rskip
        !          6351: 
        !          6352: \envdef\cartouche{%
        !          6353:   \ifhmode\par\fi  % can't be in the midst of a paragraph.
        !          6354:   \startsavinginserts
        !          6355:   \lskip=\leftskip \rskip=\rightskip
        !          6356:   \leftskip=0pt\rightskip=0pt % we want these *outside*.
        !          6357:   \cartinner=\hsize \advance\cartinner by-\lskip
        !          6358:   \advance\cartinner by-\rskip
        !          6359:   \cartouter=\hsize
        !          6360:   \advance\cartouter by 18.4pt % allow for 3pt kerns on either
        !          6361:                                % side, and for 6pt waste from
        !          6362:                                % each corner char, and rule thickness
        !          6363:   \normbskip=\baselineskip \normpskip=\parskip \normlskip=\lineskip
        !          6364:   % Flag to tell @lisp, etc., not to narrow margin.
        !          6365:   \let\nonarrowing = t%
        !          6366:   %
        !          6367:   % If this cartouche directly follows a sectioning command, we need the
        !          6368:   % \parskip glue (backspaced over by default) or the cartouche can
        !          6369:   % collide with the section heading.
        !          6370:   \ifnum\lastpenalty>10000 \vskip\parskip \penalty\lastpenalty \fi
        !          6371:   %
        !          6372:   \vbox\bgroup
        !          6373:       \baselineskip=0pt\parskip=0pt\lineskip=0pt
        !          6374:       \carttop
        !          6375:       \hbox\bgroup
        !          6376:          \hskip\lskip
        !          6377:          \vrule\kern3pt
        !          6378:          \vbox\bgroup
        !          6379:              \kern3pt
        !          6380:              \hsize=\cartinner
        !          6381:              \baselineskip=\normbskip
        !          6382:              \lineskip=\normlskip
        !          6383:              \parskip=\normpskip
        !          6384:              \vskip -\parskip
        !          6385:              \comment % For explanation, see the end of def\group.
        !          6386: }
        !          6387: \def\Ecartouche{%
        !          6388:               \ifhmode\par\fi
        !          6389:              \kern3pt
        !          6390:          \egroup
        !          6391:          \kern3pt\vrule
        !          6392:          \hskip\rskip
        !          6393:       \egroup
        !          6394:       \cartbot
        !          6395:   \egroup
        !          6396:   \checkinserts
        !          6397: }
        !          6398: 
        !          6399: 
        !          6400: % This macro is called at the beginning of all the @example variants,
        !          6401: % inside a group.
        !          6402: \newdimen\nonfillparindent
        !          6403: \def\nonfillstart{%
        !          6404:   \aboveenvbreak
        !          6405:   \hfuzz = 12pt % Don't be fussy
        !          6406:   \sepspaces % Make spaces be word-separators rather than space tokens.
        !          6407:   \let\par = \lisppar % don't ignore blank lines
        !          6408:   \obeylines % each line of input is a line of output
        !          6409:   \parskip = 0pt
        !          6410:   % Turn off paragraph indentation but redefine \indent to emulate
        !          6411:   % the normal \indent.
        !          6412:   \nonfillparindent=\parindent
        !          6413:   \parindent = 0pt
        !          6414:   \let\indent\nonfillindent
        !          6415:   %
        !          6416:   \emergencystretch = 0pt % don't try to avoid overfull boxes
        !          6417:   \ifx\nonarrowing\relax
        !          6418:     \advance \leftskip by \lispnarrowing
        !          6419:     \exdentamount=\lispnarrowing
        !          6420:   \else
        !          6421:     \let\nonarrowing = \relax
        !          6422:   \fi
        !          6423:   \let\exdent=\nofillexdent
        !          6424: }
        !          6425: 
        !          6426: \begingroup
        !          6427: \obeyspaces
        !          6428: % We want to swallow spaces (but not other tokens) after the fake
        !          6429: % @indent in our nonfill-environments, where spaces are normally
        !          6430: % active and set to @tie, resulting in them not being ignored after
        !          6431: % @indent.
        !          6432: \gdef\nonfillindent{\futurelet\temp\nonfillindentcheck}%
        !          6433: \gdef\nonfillindentcheck{%
        !          6434: \ifx\temp %
        !          6435: \expandafter\nonfillindentgobble%
        !          6436: \else%
        !          6437: \leavevmode\nonfillindentbox%
        !          6438: \fi%
        !          6439: }%
        !          6440: \endgroup
        !          6441: \def\nonfillindentgobble#1{\nonfillindent}
        !          6442: \def\nonfillindentbox{\hbox to \nonfillparindent{\hss}}
        !          6443: 
        !          6444: % If you want all examples etc. small: @set dispenvsize small.
        !          6445: % If you want even small examples the full size: @set dispenvsize nosmall.
        !          6446: % This affects the following displayed environments:
        !          6447: %    @example, @display, @format, @lisp
        !          6448: %
        !          6449: \def\smallword{small}
        !          6450: \def\nosmallword{nosmall}
        !          6451: \let\SETdispenvsize\relax
        !          6452: \def\setnormaldispenv{%
        !          6453:   \ifx\SETdispenvsize\smallword
        !          6454:     % end paragraph for sake of leading, in case document has no blank
        !          6455:     % line.  This is redundant with what happens in \aboveenvbreak, but
        !          6456:     % we need to do it before changing the fonts, and it's inconvenient
        !          6457:     % to change the fonts afterward.
        !          6458:     \ifnum \lastpenalty=10000 \else \endgraf \fi
        !          6459:     \smallexamplefonts \rm
        !          6460:   \fi
        !          6461: }
        !          6462: \def\setsmalldispenv{%
        !          6463:   \ifx\SETdispenvsize\nosmallword
        !          6464:   \else
        !          6465:     \ifnum \lastpenalty=10000 \else \endgraf \fi
        !          6466:     \smallexamplefonts \rm
        !          6467:   \fi
        !          6468: }
        !          6469: 
        !          6470: % We often define two environments, @foo and @smallfoo.
        !          6471: % Let's do it in one command.  #1 is the env name, #2 the definition.
        !          6472: \def\makedispenvdef#1#2{%
        !          6473:   \expandafter\envdef\csname#1\endcsname {\setnormaldispenv #2}%
        !          6474:   \expandafter\envdef\csname small#1\endcsname {\setsmalldispenv #2}%
        !          6475:   \expandafter\let\csname E#1\endcsname \afterenvbreak
        !          6476:   \expandafter\let\csname Esmall#1\endcsname \afterenvbreak
        !          6477: }
        !          6478: 
        !          6479: % Define two environment synonyms (#1 and #2) for an environment.
        !          6480: \def\maketwodispenvdef#1#2#3{%
        !          6481:   \makedispenvdef{#1}{#3}%
        !          6482:   \makedispenvdef{#2}{#3}%
        !          6483: }
        !          6484: %
        !          6485: % @lisp: indented, narrowed, typewriter font;
        !          6486: % @example: same as @lisp.
        !          6487: %
        !          6488: % @smallexample and @smalllisp: use smaller fonts.
        !          6489: % Originally contributed by Pavel@xerox.
        !          6490: %
        !          6491: \maketwodispenvdef{lisp}{example}{%
        !          6492:   \nonfillstart
        !          6493:   \tt\setupmarkupstyle{example}%
        !          6494:   \let\kbdfont = \kbdexamplefont % Allow @kbd to do something special.
        !          6495:   \gobble % eat return
        !          6496: }
        !          6497: % @display/@smalldisplay: same as @lisp except keep current font.
        !          6498: %
        !          6499: \makedispenvdef{display}{%
        !          6500:   \nonfillstart
        !          6501:   \gobble
        !          6502: }
        !          6503: 
        !          6504: % @format/@smallformat: same as @display except don't narrow margins.
        !          6505: %
        !          6506: \makedispenvdef{format}{%
        !          6507:   \let\nonarrowing = t%
        !          6508:   \nonfillstart
        !          6509:   \gobble
        !          6510: }
        !          6511: 
        !          6512: % @flushleft: same as @format, but doesn't obey \SETdispenvsize.
        !          6513: \envdef\flushleft{%
        !          6514:   \let\nonarrowing = t%
        !          6515:   \nonfillstart
        !          6516:   \gobble
        !          6517: }
        !          6518: \let\Eflushleft = \afterenvbreak
        !          6519: 
        !          6520: % @flushright.
        !          6521: %
        !          6522: \envdef\flushright{%
        !          6523:   \let\nonarrowing = t%
        !          6524:   \nonfillstart
        !          6525:   \advance\leftskip by 0pt plus 1fill\relax
        !          6526:   \gobble
        !          6527: }
        !          6528: \let\Eflushright = \afterenvbreak
        !          6529: 
        !          6530: 
        !          6531: % @raggedright does more-or-less normal line breaking but no right
        !          6532: % justification.  From plain.tex.
        !          6533: \envdef\raggedright{%
        !          6534:   \rightskip0pt plus2em \spaceskip.3333em \xspaceskip.5em\relax
        !          6535: }
        !          6536: \let\Eraggedright\par
        !          6537: 
        !          6538: \envdef\raggedleft{%
        !          6539:   \parindent=0pt \leftskip0pt plus2em
        !          6540:   \spaceskip.3333em \xspaceskip.5em \parfillskip=0pt
        !          6541:   \hbadness=10000 % Last line will usually be underfull, so turn off
        !          6542:                   % badness reporting.
        !          6543: }
        !          6544: \let\Eraggedleft\par
        !          6545: 
        !          6546: \envdef\raggedcenter{%
        !          6547:   \parindent=0pt \rightskip0pt plus1em \leftskip0pt plus1em
        !          6548:   \spaceskip.3333em \xspaceskip.5em \parfillskip=0pt
        !          6549:   \hbadness=10000 % Last line will usually be underfull, so turn off
        !          6550:                   % badness reporting.
        !          6551: }
        !          6552: \let\Eraggedcenter\par
        !          6553: 
        !          6554: 
        !          6555: % @quotation does normal linebreaking (hence we can't use \nonfillstart)
        !          6556: % and narrows the margins.  We keep \parskip nonzero in general, since
        !          6557: % we're doing normal filling.  So, when using \aboveenvbreak and
        !          6558: % \afterenvbreak, temporarily make \parskip 0.
        !          6559: %
        !          6560: \makedispenvdef{quotation}{\quotationstart}
        !          6561: %
        !          6562: \def\quotationstart{%
        !          6563:   \indentedblockstart % same as \indentedblock, but increase right margin too.
        !          6564:   \ifx\nonarrowing\relax
        !          6565:     \advance\rightskip by \lispnarrowing
        !          6566:   \fi
        !          6567:   \parsearg\quotationlabel
        !          6568: }
        !          6569: 
        !          6570: % We have retained a nonzero parskip for the environment, since we're
        !          6571: % doing normal filling.
        !          6572: %
        !          6573: \def\Equotation{%
        !          6574:   \par
        !          6575:   \ifx\quotationauthor\thisisundefined\else
        !          6576:     % indent a bit.
        !          6577:     \leftline{\kern 2\leftskip \sl ---\quotationauthor}%
        !          6578:   \fi
        !          6579:   {\parskip=0pt \afterenvbreak}%
        !          6580: }
        !          6581: \def\Esmallquotation{\Equotation}
        !          6582: 
        !          6583: % If we're given an argument, typeset it in bold with a colon after.
        !          6584: \def\quotationlabel#1{%
        !          6585:   \def\temp{#1}%
        !          6586:   \ifx\temp\empty \else
        !          6587:     {\bf #1: }%
        !          6588:   \fi
        !          6589: }
        !          6590: 
        !          6591: % @indentedblock is like @quotation, but indents only on the left and
        !          6592: % has no optional argument.
        !          6593: % 
        !          6594: \makedispenvdef{indentedblock}{\indentedblockstart}
        !          6595: %
        !          6596: \def\indentedblockstart{%
        !          6597:   {\parskip=0pt \aboveenvbreak}% because \aboveenvbreak inserts \parskip
        !          6598:   \parindent=0pt
        !          6599:   %
        !          6600:   % @cartouche defines \nonarrowing to inhibit narrowing at next level down.
        !          6601:   \ifx\nonarrowing\relax
        !          6602:     \advance\leftskip by \lispnarrowing
        !          6603:     \exdentamount = \lispnarrowing
        !          6604:   \else
        !          6605:     \let\nonarrowing = \relax
        !          6606:   \fi
        !          6607: }
        !          6608: 
        !          6609: % Keep a nonzero parskip for the environment, since we're doing normal filling.
        !          6610: %
        !          6611: \def\Eindentedblock{%
        !          6612:   \par
        !          6613:   {\parskip=0pt \afterenvbreak}%
        !          6614: }
        !          6615: \def\Esmallindentedblock{\Eindentedblock}
        !          6616: 
        !          6617: 
        !          6618: % LaTeX-like @verbatim...@end verbatim and @verb{<char>...<char>}
        !          6619: % If we want to allow any <char> as delimiter,
        !          6620: % we need the curly braces so that makeinfo sees the @verb command, eg:
        !          6621: % `@verbx...x' would look like the '@verbx' command.  [email protected]
        !          6622: %
        !          6623: % [Knuth]: Donald Ervin Knuth, 1996.  The TeXbook.
        !          6624: %
        !          6625: % [Knuth] p.344; only we need to do the other characters Texinfo sets
        !          6626: % active too.  Otherwise, they get lost as the first character on a
        !          6627: % verbatim line.
        !          6628: \def\dospecials{%
        !          6629:   \do\ \do\\\do\{\do\}\do\$\do\&%
        !          6630:   \do\#\do\^\do\^^K\do\_\do\^^A\do\%\do\~%
        !          6631:   \do\<\do\>\do\|\do\@\do+\do\"%
        !          6632:   % Don't do the quotes -- if we do, @set txicodequoteundirected and
        !          6633:   % @set txicodequotebacktick will not have effect on @verb and
        !          6634:   % @verbatim, and ?` and !` ligatures won't get disabled.
        !          6635:   %\do\`\do\'%
        !          6636: }
        !          6637: %
        !          6638: % [Knuth] p. 380
        !          6639: \def\uncatcodespecials{%
        !          6640:   \def\do##1{\catcode`##1=\other}\dospecials}
        !          6641: %
        !          6642: % Setup for the @verb command.
        !          6643: %
        !          6644: % Eight spaces for a tab
        !          6645: \begingroup
        !          6646:   \catcode`\^^I=\active
        !          6647:   \gdef\tabeightspaces{\catcode`\^^I=\active\def^^I{\ \ \ \ \ \ \ \ }}
        !          6648: \endgroup
        !          6649: %
        !          6650: \def\setupverb{%
        !          6651:   \tt  % easiest (and conventionally used) font for verbatim
        !          6652:   \def\par{\leavevmode\endgraf}%
        !          6653:   \setupmarkupstyle{verb}%
        !          6654:   \tabeightspaces
        !          6655:   % Respect line breaks,
        !          6656:   % print special symbols as themselves, and
        !          6657:   % make each space count
        !          6658:   % must do in this order:
        !          6659:   \obeylines \uncatcodespecials \sepspaces
        !          6660: }
        !          6661: 
        !          6662: % Setup for the @verbatim environment
        !          6663: %
        !          6664: % Real tab expansion.
        !          6665: \newdimen\tabw \setbox0=\hbox{\tt\space} \tabw=8\wd0 % tab amount
        !          6666: %
        !          6667: % We typeset each line of the verbatim in an \hbox, so we can handle
        !          6668: % tabs.  The \global is in case the verbatim line starts with an accent,
        !          6669: % or some other command that starts with a begin-group.  Otherwise, the
        !          6670: % entire \verbbox would disappear at the corresponding end-group, before
        !          6671: % it is typeset.  Meanwhile, we can't have nested verbatim commands
        !          6672: % (can we?), so the \global won't be overwriting itself.
        !          6673: \newbox\verbbox
        !          6674: \def\starttabbox{\global\setbox\verbbox=\hbox\bgroup}
        !          6675: %
        !          6676: \begingroup
        !          6677:   \catcode`\^^I=\active
        !          6678:   \gdef\tabexpand{%
        !          6679:     \catcode`\^^I=\active
        !          6680:     \def^^I{\leavevmode\egroup
        !          6681:       \dimen\verbbox=\wd\verbbox % the width so far, or since the previous tab
        !          6682:       \divide\dimen\verbbox by\tabw
        !          6683:       \multiply\dimen\verbbox by\tabw % compute previous multiple of \tabw
        !          6684:       \advance\dimen\verbbox by\tabw  % advance to next multiple of \tabw
        !          6685:       \wd\verbbox=\dimen\verbbox \box\verbbox \starttabbox
        !          6686:     }%
        !          6687:   }
        !          6688: \endgroup
        !          6689: 
        !          6690: % start the verbatim environment.
        !          6691: \def\setupverbatim{%
        !          6692:   \let\nonarrowing = t%
        !          6693:   \nonfillstart
        !          6694:   \tt % easiest (and conventionally used) font for verbatim
        !          6695:   % The \leavevmode here is for blank lines.  Otherwise, we would
        !          6696:   % never \starttabox and the \egroup would end verbatim mode.
        !          6697:   \def\par{\leavevmode\egroup\box\verbbox\endgraf}%
        !          6698:   \tabexpand
        !          6699:   \setupmarkupstyle{verbatim}%
        !          6700:   % Respect line breaks,
        !          6701:   % print special symbols as themselves, and
        !          6702:   % make each space count.
        !          6703:   % Must do in this order:
        !          6704:   \obeylines \uncatcodespecials \sepspaces
        !          6705:   \everypar{\starttabbox}%
        !          6706: }
        !          6707: 
        !          6708: % Do the @verb magic: verbatim text is quoted by unique
        !          6709: % delimiter characters.  Before first delimiter expect a
        !          6710: % right brace, after last delimiter expect closing brace:
        !          6711: %
        !          6712: %    \def\doverb'{'<char>#1<char>'}'{#1}
        !          6713: %
        !          6714: % [Knuth] p. 382; only eat outer {}
        !          6715: \begingroup
        !          6716:   \catcode`[=1\catcode`]=2\catcode`\{=\other\catcode`\}=\other
        !          6717:   \gdef\doverb{#1[\def\next##1#1}[##1\endgroup]\next]
        !          6718: \endgroup
        !          6719: %
        !          6720: \def\verb{\begingroup\setupverb\doverb}
        !          6721: %
        !          6722: %
        !          6723: % Do the @verbatim magic: define the macro \doverbatim so that
        !          6724: % the (first) argument ends when '@end verbatim' is reached, ie:
        !          6725: %
        !          6726: %     \def\doverbatim#1@end verbatim{#1}
        !          6727: %
        !          6728: % For Texinfo it's a lot easier than for LaTeX,
        !          6729: % because texinfo's \verbatim doesn't stop at '\end{verbatim}':
        !          6730: % we need not redefine '\', '{' and '}'.
        !          6731: %
        !          6732: % Inspired by LaTeX's verbatim command set [latex.ltx]
        !          6733: %
        !          6734: \begingroup
        !          6735:   \catcode`\ =\active
        !          6736:   \obeylines %
        !          6737:   % ignore everything up to the first ^^M, that's the newline at the end
        !          6738:   % of the @verbatim input line itself.  Otherwise we get an extra blank
        !          6739:   % line in the output.
        !          6740:   \xdef\doverbatim#1^^M#2@end verbatim{#2\noexpand\end\gobble verbatim}%
        !          6741:   % We really want {...\end verbatim} in the body of the macro, but
        !          6742:   % without the active space; thus we have to use \xdef and \gobble.
        !          6743: \endgroup
        !          6744: %
        !          6745: \envdef\verbatim{%
        !          6746:     \setupverbatim\doverbatim
        !          6747: }
        !          6748: \let\Everbatim = \afterenvbreak
        !          6749: 
        !          6750: 
        !          6751: % @verbatiminclude FILE - insert text of file in verbatim environment.
        !          6752: %
        !          6753: \def\verbatiminclude{\parseargusing\filenamecatcodes\doverbatiminclude}
        !          6754: %
        !          6755: \def\doverbatiminclude#1{%
        !          6756:   {%
        !          6757:     \makevalueexpandable
        !          6758:     \setupverbatim
        !          6759:     \indexnofonts       % Allow `@@' and other weird things in file names.
        !          6760:     \wlog{texinfo.tex: doing @verbatiminclude of #1^^J}%
        !          6761:     \input #1
        !          6762:     \afterenvbreak
        !          6763:   }%
        !          6764: }
        !          6765: 
        !          6766: % @copying ... @end copying.
        !          6767: % Save the text away for @insertcopying later.
        !          6768: %
        !          6769: % We save the uninterpreted tokens, rather than creating a box.
        !          6770: % Saving the text in a box would be much easier, but then all the
        !          6771: % typesetting commands (@smallbook, font changes, etc.) have to be done
        !          6772: % beforehand -- and a) we want @copying to be done first in the source
        !          6773: % file; b) letting users define the frontmatter in as flexible order as
        !          6774: % possible is very desirable.
        !          6775: %
        !          6776: \def\copying{\checkenv{}\begingroup\scanargctxt\docopying}
        !          6777: \def\docopying#1@end copying{\endgroup\def\copyingtext{#1}}
        !          6778: %
        !          6779: \def\insertcopying{%
        !          6780:   \begingroup
        !          6781:     \parindent = 0pt  % paragraph indentation looks wrong on title page
        !          6782:     \scanexp\copyingtext
        !          6783:   \endgroup
        !          6784: }
        !          6785: 
        !          6786: 
        !          6787: \message{defuns,}
        !          6788: % @defun etc.
        !          6789: 
        !          6790: \newskip\defbodyindent \defbodyindent=.4in
        !          6791: \newskip\defargsindent \defargsindent=50pt
        !          6792: \newskip\deflastargmargin \deflastargmargin=18pt
        !          6793: \newcount\defunpenalty
        !          6794: 
        !          6795: % Start the processing of @deffn:
        !          6796: \def\startdefun{%
        !          6797:   \ifnum\lastpenalty<10000
        !          6798:     \medbreak
        !          6799:     \defunpenalty=10003 % Will keep this @deffn together with the
        !          6800:                         % following @def command, see below.
        !          6801:   \else
        !          6802:     % If there are two @def commands in a row, we'll have a \nobreak,
        !          6803:     % which is there to keep the function description together with its
        !          6804:     % header.  But if there's nothing but headers, we need to allow a
        !          6805:     % break somewhere.  Check specifically for penalty 10002, inserted
        !          6806:     % by \printdefunline, instead of 10000, since the sectioning
        !          6807:     % commands also insert a nobreak penalty, and we don't want to allow
        !          6808:     % a break between a section heading and a defun.
        !          6809:     %
        !          6810:     % As a further refinement, we avoid "club" headers by signalling
        !          6811:     % with penalty of 10003 after the very first @deffn in the
        !          6812:     % sequence (see above), and penalty of 10002 after any following
        !          6813:     % @def command.
        !          6814:     \ifnum\lastpenalty=10002 \penalty2000 \else \defunpenalty=10002 \fi
        !          6815:     %
        !          6816:     % Similarly, after a section heading, do not allow a break.
        !          6817:     % But do insert the glue.
        !          6818:     \medskip  % preceded by discardable penalty, so not a breakpoint
        !          6819:   \fi
        !          6820:   %
        !          6821:   \parindent=0in
        !          6822:   \advance\leftskip by \defbodyindent
        !          6823:   \exdentamount=\defbodyindent
        !          6824: }
        !          6825: 
        !          6826: \def\dodefunx#1{%
        !          6827:   % First, check whether we are in the right environment:
        !          6828:   \checkenv#1%
        !          6829:   %
        !          6830:   % As above, allow line break if we have multiple x headers in a row.
        !          6831:   % It's not a great place, though.
        !          6832:   \ifnum\lastpenalty=10002 \penalty3000 \else \defunpenalty=10002 \fi
        !          6833:   %
        !          6834:   % And now, it's time to reuse the body of the original defun:
        !          6835:   \expandafter\gobbledefun#1%
        !          6836: }
        !          6837: \def\gobbledefun#1\startdefun{}
        !          6838: 
        !          6839: % \printdefunline \deffnheader{text}
        !          6840: %
        !          6841: \def\printdefunline#1#2{%
        !          6842:   \begingroup
        !          6843:     % call \deffnheader:
        !          6844:     #1#2 \endheader
        !          6845:     % common ending:
        !          6846:     \interlinepenalty = 10000
        !          6847:     \advance\rightskip by 0pt plus 1fil\relax
        !          6848:     \endgraf
        !          6849:     \nobreak\vskip -\parskip
        !          6850:     \penalty\defunpenalty  % signal to \startdefun and \dodefunx
        !          6851:     % Some of the @defun-type tags do not enable magic parentheses,
        !          6852:     % rendering the following check redundant.  But we don't optimize.
        !          6853:     \checkparencounts
        !          6854:   \endgroup
        !          6855: }
        !          6856: 
        !          6857: \def\Edefun{\endgraf\medbreak}
        !          6858: 
        !          6859: % \makedefun{deffn} creates \deffn, \deffnx and \Edeffn;
        !          6860: % the only thing remaining is to define \deffnheader.
        !          6861: %
        !          6862: \def\makedefun#1{%
        !          6863:   \expandafter\let\csname E#1\endcsname = \Edefun
        !          6864:   \edef\temp{\noexpand\domakedefun
        !          6865:     \makecsname{#1}\makecsname{#1x}\makecsname{#1header}}%
        !          6866:   \temp
        !          6867: }
        !          6868: 
        !          6869: % \domakedefun \deffn \deffnx \deffnheader
        !          6870: %
        !          6871: % Define \deffn and \deffnx, without parameters.
        !          6872: % \deffnheader has to be defined explicitly.
        !          6873: %
        !          6874: \def\domakedefun#1#2#3{%
        !          6875:   \envdef#1{%
        !          6876:     \startdefun
        !          6877:     \doingtypefnfalse    % distinguish typed functions from all else
        !          6878:     \parseargusing\activeparens{\printdefunline#3}%
        !          6879:   }%
        !          6880:   \def#2{\dodefunx#1}%
        !          6881:   \def#3%
        !          6882: }
        !          6883: 
        !          6884: \newif\ifdoingtypefn       % doing typed function?
        !          6885: \newif\ifrettypeownline    % typeset return type on its own line?
        !          6886: 
        !          6887: % @deftypefnnewline on|off says whether the return type of typed functions
        !          6888: % are printed on their own line.  This affects @deftypefn, @deftypefun,
        !          6889: % @deftypeop, and @deftypemethod.
        !          6890: % 
        !          6891: \parseargdef\deftypefnnewline{%
        !          6892:   \def\temp{#1}%
        !          6893:   \ifx\temp\onword
        !          6894:     \expandafter\let\csname SETtxideftypefnnl\endcsname
        !          6895:       = \empty
        !          6896:   \else\ifx\temp\offword
        !          6897:     \expandafter\let\csname SETtxideftypefnnl\endcsname
        !          6898:       = \relax
        !          6899:   \else
        !          6900:     \errhelp = \EMsimple
        !          6901:     \errmessage{Unknown @txideftypefnnl value `\temp',
        !          6902:                 must be on|off}%
        !          6903:   \fi\fi
        !          6904: }
        !          6905: 
        !          6906: % Untyped functions:
        !          6907: 
        !          6908: % @deffn category name args
        !          6909: \makedefun{deffn}{\deffngeneral{}}
        !          6910: 
        !          6911: % @deffn category class name args
        !          6912: \makedefun{defop}#1 {\defopon{#1\ \putwordon}}
        !          6913: 
        !          6914: % \defopon {category on}class name args
        !          6915: \def\defopon#1#2 {\deffngeneral{\putwordon\ \code{#2}}{#1\ \code{#2}} }
        !          6916: 
        !          6917: % \deffngeneral {subind}category name args
        !          6918: %
        !          6919: \def\deffngeneral#1#2 #3 #4\endheader{%
        !          6920:   % Remember that \dosubind{fn}{foo}{} is equivalent to \doind{fn}{foo}.
        !          6921:   \dosubind{fn}{\code{#3}}{#1}%
        !          6922:   \defname{#2}{}{#3}\magicamp\defunargs{#4\unskip}%
        !          6923: }
        !          6924: 
        !          6925: % Typed functions:
        !          6926: 
        !          6927: % @deftypefn category type name args
        !          6928: \makedefun{deftypefn}{\deftypefngeneral{}}
        !          6929: 
        !          6930: % @deftypeop category class type name args
        !          6931: \makedefun{deftypeop}#1 {\deftypeopon{#1\ \putwordon}}
        !          6932: 
        !          6933: % \deftypeopon {category on}class type name args
        !          6934: \def\deftypeopon#1#2 {\deftypefngeneral{\putwordon\ \code{#2}}{#1\ \code{#2}} }
        !          6935: 
        !          6936: % \deftypefngeneral {subind}category type name args
        !          6937: %
        !          6938: \def\deftypefngeneral#1#2 #3 #4 #5\endheader{%
        !          6939:   \dosubind{fn}{\code{#4}}{#1}%
        !          6940:   \doingtypefntrue
        !          6941:   \defname{#2}{#3}{#4}\defunargs{#5\unskip}%
        !          6942: }
        !          6943: 
        !          6944: % Typed variables:
        !          6945: 
        !          6946: % @deftypevr category type var args
        !          6947: \makedefun{deftypevr}{\deftypecvgeneral{}}
        !          6948: 
        !          6949: % @deftypecv category class type var args
        !          6950: \makedefun{deftypecv}#1 {\deftypecvof{#1\ \putwordof}}
        !          6951: 
        !          6952: % \deftypecvof {category of}class type var args
        !          6953: \def\deftypecvof#1#2 {\deftypecvgeneral{\putwordof\ \code{#2}}{#1\ \code{#2}} }
        !          6954: 
        !          6955: % \deftypecvgeneral {subind}category type var args
        !          6956: %
        !          6957: \def\deftypecvgeneral#1#2 #3 #4 #5\endheader{%
        !          6958:   \dosubind{vr}{\code{#4}}{#1}%
        !          6959:   \defname{#2}{#3}{#4}\defunargs{#5\unskip}%
        !          6960: }
        !          6961: 
        !          6962: % Untyped variables:
        !          6963: 
        !          6964: % @defvr category var args
        !          6965: \makedefun{defvr}#1 {\deftypevrheader{#1} {} }
        !          6966: 
        !          6967: % @defcv category class var args
        !          6968: \makedefun{defcv}#1 {\defcvof{#1\ \putwordof}}
        !          6969: 
        !          6970: % \defcvof {category of}class var args
        !          6971: \def\defcvof#1#2 {\deftypecvof{#1}#2 {} }
        !          6972: 
        !          6973: % Types:
        !          6974: 
        !          6975: % @deftp category name args
        !          6976: \makedefun{deftp}#1 #2 #3\endheader{%
        !          6977:   \doind{tp}{\code{#2}}%
        !          6978:   \defname{#1}{}{#2}\defunargs{#3\unskip}%
        !          6979: }
        !          6980: 
        !          6981: % Remaining @defun-like shortcuts:
        !          6982: \makedefun{defun}{\deffnheader{\putwordDeffunc} }
        !          6983: \makedefun{defmac}{\deffnheader{\putwordDefmac} }
        !          6984: \makedefun{defspec}{\deffnheader{\putwordDefspec} }
        !          6985: \makedefun{deftypefun}{\deftypefnheader{\putwordDeffunc} }
        !          6986: \makedefun{defvar}{\defvrheader{\putwordDefvar} }
        !          6987: \makedefun{defopt}{\defvrheader{\putwordDefopt} }
        !          6988: \makedefun{deftypevar}{\deftypevrheader{\putwordDefvar} }
        !          6989: \makedefun{defmethod}{\defopon\putwordMethodon}
        !          6990: \makedefun{deftypemethod}{\deftypeopon\putwordMethodon}
        !          6991: \makedefun{defivar}{\defcvof\putwordInstanceVariableof}
        !          6992: \makedefun{deftypeivar}{\deftypecvof\putwordInstanceVariableof}
        !          6993: 
        !          6994: % \defname, which formats the name of the @def (not the args).
        !          6995: % #1 is the category, such as "Function".
        !          6996: % #2 is the return type, if any.
        !          6997: % #3 is the function name.
        !          6998: %
        !          6999: % We are followed by (but not passed) the arguments, if any.
        !          7000: %
        !          7001: \def\defname#1#2#3{%
        !          7002:   \par
        !          7003:   % Get the values of \leftskip and \rightskip as they were outside the @def...
        !          7004:   \advance\leftskip by -\defbodyindent
        !          7005:   %
        !          7006:   % Determine if we are typesetting the return type of a typed function
        !          7007:   % on a line by itself.
        !          7008:   \rettypeownlinefalse
        !          7009:   \ifdoingtypefn  % doing a typed function specifically?
        !          7010:     % then check user option for putting return type on its own line:
        !          7011:     \expandafter\ifx\csname SETtxideftypefnnl\endcsname\relax \else
        !          7012:       \rettypeownlinetrue
        !          7013:     \fi
        !          7014:   \fi
        !          7015:   %
        !          7016:   % How we'll format the category name.  Putting it in brackets helps
        !          7017:   % distinguish it from the body text that may end up on the next line
        !          7018:   % just below it.
        !          7019:   \def\temp{#1}%
        !          7020:   \setbox0=\hbox{\kern\deflastargmargin \ifx\temp\empty\else [\rm\temp]\fi}
        !          7021:   %
        !          7022:   % Figure out line sizes for the paragraph shape.  We'll always have at
        !          7023:   % least two.
        !          7024:   \tempnum = 2
        !          7025:   %
        !          7026:   % The first line needs space for \box0; but if \rightskip is nonzero,
        !          7027:   % we need only space for the part of \box0 which exceeds it:
        !          7028:   \dimen0=\hsize  \advance\dimen0 by -\wd0  \advance\dimen0 by \rightskip
        !          7029:   %
        !          7030:   % If doing a return type on its own line, we'll have another line.
        !          7031:   \ifrettypeownline
        !          7032:     \advance\tempnum by 1
        !          7033:     \def\maybeshapeline{0in \hsize}%
        !          7034:   \else
        !          7035:     \def\maybeshapeline{}%
        !          7036:   \fi
        !          7037:   %
        !          7038:   % The continuations:
        !          7039:   \dimen2=\hsize  \advance\dimen2 by -\defargsindent
        !          7040:   %
        !          7041:   % The final paragraph shape:
        !          7042:   \parshape \tempnum  0in \dimen0  \maybeshapeline  \defargsindent \dimen2
        !          7043:   %
        !          7044:   % Put the category name at the right margin.
        !          7045:   \noindent
        !          7046:   \hbox to 0pt{%
        !          7047:     \hfil\box0 \kern-\hsize
        !          7048:     % \hsize has to be shortened this way:
        !          7049:     \kern\leftskip
        !          7050:     % Intentionally do not respect \rightskip, since we need the space.
        !          7051:   }%
        !          7052:   %
        !          7053:   % Allow all lines to be underfull without complaint:
        !          7054:   \tolerance=10000 \hbadness=10000
        !          7055:   \exdentamount=\defbodyindent
        !          7056:   {%
        !          7057:     % defun fonts. We use typewriter by default (used to be bold) because:
        !          7058:     % . we're printing identifiers, they should be in tt in principle.
        !          7059:     % . in languages with many accents, such as Czech or French, it's
        !          7060:     %   common to leave accents off identifiers.  The result looks ok in
        !          7061:     %   tt, but exceedingly strange in rm.
        !          7062:     % . we don't want -- and --- to be treated as ligatures.
        !          7063:     % . this still does not fix the ?` and !` ligatures, but so far no
        !          7064:     %   one has made identifiers using them :).
        !          7065:     \df \tt
        !          7066:     \def\temp{#2}% text of the return type
        !          7067:     \ifx\temp\empty\else
        !          7068:       \tclose{\temp}% typeset the return type
        !          7069:       \ifrettypeownline
        !          7070:         % put return type on its own line; prohibit line break following:
        !          7071:         \hfil\vadjust{\nobreak}\break  
        !          7072:       \else
        !          7073:         \space  % type on same line, so just followed by a space
        !          7074:       \fi
        !          7075:     \fi           % no return type
        !          7076:     #3% output function name
        !          7077:   }%
        !          7078:   {\rm\enskip}% hskip 0.5 em of \tenrm
        !          7079:   %
        !          7080:   \boldbrax
        !          7081:   % arguments will be output next, if any.
        !          7082: }
        !          7083: 
        !          7084: % Print arguments in slanted roman (not ttsl), inconsistently with using
        !          7085: % tt for the name.  This is because literal text is sometimes needed in
        !          7086: % the argument list (groff manual), and ttsl and tt are not very
        !          7087: % distinguishable.  Prevent hyphenation at `-' chars.
        !          7088: %
        !          7089: \def\defunargs#1{%
        !          7090:   % use sl by default (not ttsl),
        !          7091:   % tt for the names.
        !          7092:   \df \sl \hyphenchar\font=0
        !          7093:   %
        !          7094:   % On the other hand, if an argument has two dashes (for instance), we
        !          7095:   % want a way to get ttsl.  We used to recommend @var for that, so
        !          7096:   % leave the code in, but it's strange for @var to lead to typewriter.
        !          7097:   % Nowadays we recommend @code, since the difference between a ttsl hyphen
        !          7098:   % and a tt hyphen is pretty tiny.  @code also disables ?` !`.
        !          7099:   \def\var##1{{\setupmarkupstyle{var}\ttslanted{##1}}}%
        !          7100:   #1%
        !          7101:   \sl\hyphenchar\font=45
        !          7102: }
        !          7103: 
        !          7104: % We want ()&[] to print specially on the defun line.
        !          7105: %
        !          7106: \def\activeparens{%
        !          7107:   \catcode`\(=\active \catcode`\)=\active
        !          7108:   \catcode`\[=\active \catcode`\]=\active
        !          7109:   \catcode`\&=\active
        !          7110: }
        !          7111: 
        !          7112: % Make control sequences which act like normal parenthesis chars.
        !          7113: \let\lparen = ( \let\rparen = )
        !          7114: 
        !          7115: % Be sure that we always have a definition for `(', etc.  For example,
        !          7116: % if the fn name has parens in it, \boldbrax will not be in effect yet,
        !          7117: % so TeX would otherwise complain about undefined control sequence.
        !          7118: {
        !          7119:   \activeparens
        !          7120:   \global\let(=\lparen \global\let)=\rparen
        !          7121:   \global\let[=\lbrack \global\let]=\rbrack
        !          7122:   \global\let& = \&
        !          7123: 
        !          7124:   \gdef\boldbrax{\let(=\opnr\let)=\clnr\let[=\lbrb\let]=\rbrb}
        !          7125:   \gdef\magicamp{\let&=\amprm}
        !          7126: }
        !          7127: 
        !          7128: \newcount\parencount
        !          7129: 
        !          7130: % If we encounter &foo, then turn on ()-hacking afterwards
        !          7131: \newif\ifampseen
        !          7132: \def\amprm#1 {\ampseentrue{\bf\&#1 }}
        !          7133: 
        !          7134: \def\parenfont{%
        !          7135:   \ifampseen
        !          7136:     % At the first level, print parens in roman,
        !          7137:     % otherwise use the default font.
        !          7138:     \ifnum \parencount=1 \rm \fi
        !          7139:   \else
        !          7140:     % The \sf parens (in \boldbrax) actually are a little bolder than
        !          7141:     % the contained text.  This is especially needed for [ and ] .
        !          7142:     \sf
        !          7143:   \fi
        !          7144: }
        !          7145: \def\infirstlevel#1{%
        !          7146:   \ifampseen
        !          7147:     \ifnum\parencount=1
        !          7148:       #1%
        !          7149:     \fi
        !          7150:   \fi
        !          7151: }
        !          7152: \def\bfafterword#1 {#1 \bf}
        !          7153: 
        !          7154: \def\opnr{%
        !          7155:   \global\advance\parencount by 1
        !          7156:   {\parenfont(}%
        !          7157:   \infirstlevel \bfafterword
        !          7158: }
        !          7159: \def\clnr{%
        !          7160:   {\parenfont)}%
        !          7161:   \infirstlevel \sl
        !          7162:   \global\advance\parencount by -1
        !          7163: }
        !          7164: 
        !          7165: \newcount\brackcount
        !          7166: \def\lbrb{%
        !          7167:   \global\advance\brackcount by 1
        !          7168:   {\bf[}%
        !          7169: }
        !          7170: \def\rbrb{%
        !          7171:   {\bf]}%
        !          7172:   \global\advance\brackcount by -1
        !          7173: }
        !          7174: 
        !          7175: \def\checkparencounts{%
        !          7176:   \ifnum\parencount=0 \else \badparencount \fi
        !          7177:   \ifnum\brackcount=0 \else \badbrackcount \fi
        !          7178: }
        !          7179: % these should not use \errmessage; the glibc manual, at least, actually
        !          7180: % has such constructs (when documenting function pointers).
        !          7181: \def\badparencount{%
        !          7182:   \message{Warning: unbalanced parentheses in @def...}%
        !          7183:   \global\parencount=0
        !          7184: }
        !          7185: \def\badbrackcount{%
        !          7186:   \message{Warning: unbalanced square brackets in @def...}%
        !          7187:   \global\brackcount=0
        !          7188: }
        !          7189: 
        !          7190: 
        !          7191: \message{macros,}
        !          7192: % @macro.
        !          7193: 
        !          7194: % To do this right we need a feature of e-TeX, \scantokens,
        !          7195: % which we arrange to emulate with a temporary file in ordinary TeX.
        !          7196: \ifx\eTeXversion\thisisundefined
        !          7197:   \newwrite\macscribble
        !          7198:   \def\scantokens#1{%
        !          7199:     \toks0={#1}%
        !          7200:     \immediate\openout\macscribble=\jobname.tmp
        !          7201:     \immediate\write\macscribble{\the\toks0}%
        !          7202:     \immediate\closeout\macscribble
        !          7203:     \input \jobname.tmp
        !          7204:   }
        !          7205: \fi
        !          7206: 
        !          7207: \def\scanmacro#1{\begingroup
        !          7208:   \newlinechar`\^^M
        !          7209:   \let\xeatspaces\eatspaces
        !          7210:   %
        !          7211:   % Undo catcode changes of \startcontents and \doprintindex
        !          7212:   % When called from @insertcopying or (short)caption, we need active
        !          7213:   % backslash to get it printed correctly.  Previously, we had
        !          7214:   % \catcode`\\=\other instead.  We'll see whether a problem appears
        !          7215:   % with macro expansion.                              --kasal, 19aug04
        !          7216:   \catcode`\@=0 \catcode`\\=\active \escapechar=`\@
        !          7217:   %
        !          7218:   % ... and for \example:
        !          7219:   \spaceisspace
        !          7220:   %
        !          7221:   % The \empty here causes a following catcode 5 newline to be eaten as
        !          7222:   % part of reading whitespace after a control sequence.  It does not
        !          7223:   % eat a catcode 13 newline.  There's no good way to handle the two
        !          7224:   % cases (untried: maybe e-TeX's \everyeof could help, though plain TeX
        !          7225:   % would then have different behavior).  See the Macro Details node in
        !          7226:   % the manual for the workaround we recommend for macros and
        !          7227:   % line-oriented commands.
        !          7228:   % 
        !          7229:   \scantokens{#1\empty}%
        !          7230: \endgroup}
        !          7231: 
        !          7232: \def\scanexp#1{%
        !          7233:   \edef\temp{\noexpand\scanmacro{#1}}%
        !          7234:   \temp
        !          7235: }
        !          7236: 
        !          7237: \newcount\paramno   % Count of parameters
        !          7238: \newtoks\macname    % Macro name
        !          7239: \newif\ifrecursive  % Is it recursive?
        !          7240: 
        !          7241: % List of all defined macros in the form
        !          7242: %    \definedummyword\macro1\definedummyword\macro2...
        !          7243: % Currently is also contains all @aliases; the list can be split
        !          7244: % if there is a need.
        !          7245: \def\macrolist{}
        !          7246: 
        !          7247: % Add the macro to \macrolist
        !          7248: \def\addtomacrolist#1{\expandafter \addtomacrolistxxx \csname#1\endcsname}
        !          7249: \def\addtomacrolistxxx#1{%
        !          7250:      \toks0 = \expandafter{\macrolist\definedummyword#1}%
        !          7251:      \xdef\macrolist{\the\toks0}%
        !          7252: }
        !          7253: 
        !          7254: % Utility routines.
        !          7255: % This does \let #1 = #2, with \csnames; that is,
        !          7256: %   \let \csname#1\endcsname = \csname#2\endcsname
        !          7257: % (except of course we have to play expansion games).
        !          7258: %
        !          7259: \def\cslet#1#2{%
        !          7260:   \expandafter\let
        !          7261:   \csname#1\expandafter\endcsname
        !          7262:   \csname#2\endcsname
        !          7263: }
        !          7264: 
        !          7265: % Trim leading and trailing spaces off a string.
        !          7266: % Concepts from aro-bend problem 15 (see CTAN).
        !          7267: {\catcode`\@=11
        !          7268: \gdef\eatspaces #1{\expandafter\trim@\expandafter{#1 }}
        !          7269: \gdef\trim@ #1{\trim@@ @#1 @ #1 @ @@}
        !          7270: \gdef\trim@@ #1@ #2@ #3@@{\trim@@@\empty #2 @}
        !          7271: \def\unbrace#1{#1}
        !          7272: \unbrace{\gdef\trim@@@ #1 } #2@{#1}
        !          7273: }
        !          7274: 
        !          7275: % Trim a single trailing ^^M off a string.
        !          7276: {\catcode`\^^M=\other \catcode`\Q=3%
        !          7277: \gdef\eatcr #1{\eatcra #1Q^^MQ}%
        !          7278: \gdef\eatcra#1^^MQ{\eatcrb#1Q}%
        !          7279: \gdef\eatcrb#1Q#2Q{#1}%
        !          7280: }
        !          7281: 
        !          7282: % Macro bodies are absorbed as an argument in a context where
        !          7283: % all characters are catcode 10, 11 or 12, except \ which is active
        !          7284: % (as in normal texinfo). It is necessary to change the definition of \
        !          7285: % to recognize macro arguments; this is the job of \mbodybackslash.
        !          7286: %
        !          7287: % Non-ASCII encodings make 8-bit characters active, so un-activate
        !          7288: % them to avoid their expansion.  Must do this non-globally, to
        !          7289: % confine the change to the current group.
        !          7290: %
        !          7291: % It's necessary to have hard CRs when the macro is executed. This is
        !          7292: % done by making ^^M (\endlinechar) catcode 12 when reading the macro
        !          7293: % body, and then making it the \newlinechar in \scanmacro.
        !          7294: %
        !          7295: \def\scanctxt{% used as subroutine
        !          7296:   \catcode`\"=\other
        !          7297:   \catcode`\+=\other
        !          7298:   \catcode`\<=\other
        !          7299:   \catcode`\>=\other
        !          7300:   \catcode`\@=\other
        !          7301:   \catcode`\^=\other
        !          7302:   \catcode`\_=\other
        !          7303:   \catcode`\|=\other
        !          7304:   \catcode`\~=\other
        !          7305:   \ifx\declaredencoding\ascii \else \setnonasciicharscatcodenonglobal\other \fi
        !          7306: }
        !          7307: 
        !          7308: \def\scanargctxt{% used for copying and captions, not macros.
        !          7309:   \scanctxt
        !          7310:   \catcode`\\=\other
        !          7311:   \catcode`\^^M=\other
        !          7312: }
        !          7313: 
        !          7314: \def\macrobodyctxt{% used for @macro definitions
        !          7315:   \scanctxt
        !          7316:   \catcode`\{=\other
        !          7317:   \catcode`\}=\other
        !          7318:   \catcode`\^^M=\other
        !          7319:   \usembodybackslash
        !          7320: }
        !          7321: 
        !          7322: \def\macroargctxt{% used when scanning invocations
        !          7323:   \scanctxt
        !          7324:   \catcode`\\=0
        !          7325: }
        !          7326: % why catcode 0 for \ in the above?  To recognize \\ \{ \} as "escapes"
        !          7327: % for the single characters \ { }.  Thus, we end up with the "commands"
        !          7328: % that would be written @\ @{ @} in a Texinfo document.
        !          7329: % 
        !          7330: % We already have @{ and @}.  For @\, we define it here, and only for
        !          7331: % this purpose, to produce a typewriter backslash (so, the @\ that we
        !          7332: % define for @math can't be used with @macro calls):
        !          7333: %
        !          7334: \def\\{\normalbackslash}%
        !          7335: % 
        !          7336: % We would like to do this for \, too, since that is what makeinfo does.
        !          7337: % But it is not possible, because Texinfo already has a command @, for a
        !          7338: % cedilla accent.  Documents must use @comma{} instead.
        !          7339: %
        !          7340: % \anythingelse will almost certainly be an error of some kind.
        !          7341: 
        !          7342: 
        !          7343: % \mbodybackslash is the definition of \ in @macro bodies.
        !          7344: % It maps \foo\ => \csname macarg.foo\endcsname => #N
        !          7345: % where N is the macro parameter number.
        !          7346: % We define \csname macarg.\endcsname to be \realbackslash, so
        !          7347: % \\ in macro replacement text gets you a backslash.
        !          7348: %
        !          7349: {\catcode`@=0 @catcode`@\=@active
        !          7350:  @gdef@usembodybackslash{@let\=@mbodybackslash}
        !          7351:  @gdef@mbodybackslash#1\{@csname macarg.#1@endcsname}
        !          7352: }
        !          7353: \expandafter\def\csname macarg.\endcsname{\realbackslash}
        !          7354: 
        !          7355: \def\margbackslash#1{\char`\#1 }
        !          7356: 
        !          7357: \def\macro{\recursivefalse\parsearg\macroxxx}
        !          7358: \def\rmacro{\recursivetrue\parsearg\macroxxx}
        !          7359: 
        !          7360: \def\macroxxx#1{%
        !          7361:   \getargs{#1}% now \macname is the macname and \argl the arglist
        !          7362:   \ifx\argl\empty       % no arguments
        !          7363:      \paramno=0\relax
        !          7364:   \else
        !          7365:      \expandafter\parsemargdef \argl;%
        !          7366:      \if\paramno>256\relax
        !          7367:        \ifx\eTeXversion\thisisundefined
        !          7368:          \errhelp = \EMsimple
        !          7369:          \errmessage{You need eTeX to compile a file with macros with more than 256 arguments}
        !          7370:        \fi
        !          7371:      \fi
        !          7372:   \fi
        !          7373:   \if1\csname ismacro.\the\macname\endcsname
        !          7374:      \message{Warning: redefining \the\macname}%
        !          7375:   \else
        !          7376:      \expandafter\ifx\csname \the\macname\endcsname \relax
        !          7377:      \else \errmessage{Macro name \the\macname\space already defined}\fi
        !          7378:      \global\cslet{macsave.\the\macname}{\the\macname}%
        !          7379:      \global\expandafter\let\csname ismacro.\the\macname\endcsname=1%
        !          7380:      \addtomacrolist{\the\macname}%
        !          7381:   \fi
        !          7382:   \begingroup \macrobodyctxt
        !          7383:   \ifrecursive \expandafter\parsermacbody
        !          7384:   \else \expandafter\parsemacbody
        !          7385:   \fi}
        !          7386: 
        !          7387: \parseargdef\unmacro{%
        !          7388:   \if1\csname ismacro.#1\endcsname
        !          7389:     \global\cslet{#1}{macsave.#1}%
        !          7390:     \global\expandafter\let \csname ismacro.#1\endcsname=0%
        !          7391:     % Remove the macro name from \macrolist:
        !          7392:     \begingroup
        !          7393:       \expandafter\let\csname#1\endcsname \relax
        !          7394:       \let\definedummyword\unmacrodo
        !          7395:       \xdef\macrolist{\macrolist}%
        !          7396:     \endgroup
        !          7397:   \else
        !          7398:     \errmessage{Macro #1 not defined}%
        !          7399:   \fi
        !          7400: }
        !          7401: 
        !          7402: % Called by \do from \dounmacro on each macro.  The idea is to omit any
        !          7403: % macro definitions that have been changed to \relax.
        !          7404: %
        !          7405: \def\unmacrodo#1{%
        !          7406:   \ifx #1\relax
        !          7407:     % remove this
        !          7408:   \else
        !          7409:     \noexpand\definedummyword \noexpand#1%
        !          7410:   \fi
        !          7411: }
        !          7412: 
        !          7413: % This makes use of the obscure feature that if the last token of a
        !          7414: % <parameter list> is #, then the preceding argument is delimited by
        !          7415: % an opening brace, and that opening brace is not consumed.
        !          7416: \def\getargs#1{\getargsxxx#1{}}
        !          7417: \def\getargsxxx#1#{\getmacname #1 \relax\getmacargs}
        !          7418: \def\getmacname#1 #2\relax{\macname={#1}}
        !          7419: \def\getmacargs#1{\def\argl{#1}}
        !          7420: 
        !          7421: % For macro processing make @ a letter so that we can make Texinfo private macro names.
        !          7422: \edef\texiatcatcode{\the\catcode`\@}
        !          7423: \catcode `@=11\relax
        !          7424: 
        !          7425: % Parse the optional {params} list.  Set up \paramno and \paramlist
        !          7426: % so \defmacro knows what to do.  Define \macarg.BLAH for each BLAH
        !          7427: % in the params list to some hook where the argument si to be expanded.  If
        !          7428: % there are less than 10 arguments that hook is to be replaced by ##N where N
        !          7429: % is the position in that list, that is to say the macro arguments are to be
        !          7430: % defined `a la TeX in the macro body.  
        !          7431: %
        !          7432: % That gets used by \mbodybackslash (above).
        !          7433: %
        !          7434: % We need to get `macro parameter char #' into several definitions.
        !          7435: % The technique used is stolen from LaTeX: let \hash be something
        !          7436: % unexpandable, insert that wherever you need a #, and then redefine
        !          7437: % it to # just before using the token list produced.
        !          7438: %
        !          7439: % The same technique is used to protect \eatspaces till just before
        !          7440: % the macro is used.
        !          7441: %
        !          7442: % If there are 10 or more arguments, a different technique is used, where the
        !          7443: % hook remains in the body, and when macro is to be expanded the body is
        !          7444: % processed again to replace the arguments.
        !          7445: %
        !          7446: % In that case, the hook is \the\toks N-1, and we simply set \toks N-1 to the
        !          7447: % argument N value and then \edef  the body (nothing else will expand because of
        !          7448: % the catcode regime underwhich the body was input).
        !          7449: %
        !          7450: % If you compile with TeX (not eTeX), and you have macros with 10 or more
        !          7451: % arguments, you need that no macro has more than 256 arguments, otherwise an
        !          7452: % error is produced.
        !          7453: \def\parsemargdef#1;{%
        !          7454:   \paramno=0\def\paramlist{}%
        !          7455:   \let\hash\relax
        !          7456:   \let\xeatspaces\relax
        !          7457:   \parsemargdefxxx#1,;,%
        !          7458:   % In case that there are 10 or more arguments we parse again the arguments
        !          7459:   % list to set new definitions for the \macarg.BLAH macros corresponding to
        !          7460:   % each BLAH argument. It was anyhow needed to parse already once this list
        !          7461:   % in order to count the arguments, and as macros with at most 9 arguments
        !          7462:   % are by far more frequent than macro with 10 or more arguments, defining
        !          7463:   % twice the \macarg.BLAH macros does not cost too much processing power.
        !          7464:   \ifnum\paramno<10\relax\else
        !          7465:     \paramno0\relax
        !          7466:     \parsemmanyargdef@@#1,;,% 10 or more arguments
        !          7467:   \fi
        !          7468: }
        !          7469: \def\parsemargdefxxx#1,{%
        !          7470:   \if#1;\let\next=\relax
        !          7471:   \else \let\next=\parsemargdefxxx
        !          7472:     \advance\paramno by 1
        !          7473:     \expandafter\edef\csname macarg.\eatspaces{#1}\endcsname
        !          7474:         {\xeatspaces{\hash\the\paramno}}%
        !          7475:     \edef\paramlist{\paramlist\hash\the\paramno,}%
        !          7476:   \fi\next}
        !          7477: 
        !          7478: \def\parsemmanyargdef@@#1,{%
        !          7479:   \if#1;\let\next=\relax
        !          7480:   \else 
        !          7481:     \let\next=\parsemmanyargdef@@
        !          7482:     \edef\tempb{\eatspaces{#1}}%
        !          7483:     \expandafter\def\expandafter\tempa
        !          7484:        \expandafter{\csname macarg.\tempb\endcsname}%
        !          7485:     % Note that we need some extra \noexpand\noexpand, this is because we
        !          7486:     % don't want \the  to be expanded in the \parsermacbody  as it uses an
        !          7487:     % \xdef .
        !          7488:     \expandafter\edef\tempa
        !          7489:       {\noexpand\noexpand\noexpand\the\toks\the\paramno}%
        !          7490:     \advance\paramno by 1\relax
        !          7491:   \fi\next}
        !          7492: 
        !          7493: % These two commands read recursive and nonrecursive macro bodies.
        !          7494: % (They're different since rec and nonrec macros end differently.)
        !          7495: %
        !          7496: 
        !          7497: \catcode `\@\texiatcatcode
        !          7498: \long\def\parsemacbody#1@end macro%
        !          7499: {\xdef\temp{\eatcr{#1}}\endgroup\defmacro}%
        !          7500: \long\def\parsermacbody#1@end rmacro%
        !          7501: {\xdef\temp{\eatcr{#1}}\endgroup\defmacro}%
        !          7502: \catcode `\@=11\relax
        !          7503: 
        !          7504: \let\endargs@\relax
        !          7505: \let\nil@\relax
        !          7506: \def\nilm@{\nil@}%
        !          7507: \long\def\nillm@{\nil@}%
        !          7508: 
        !          7509: % This macro is expanded during the Texinfo macro expansion, not during its
        !          7510: % definition.  It gets all the arguments values and assigns them to macros
        !          7511: % macarg.ARGNAME
        !          7512: %
        !          7513: % #1 is the macro name
        !          7514: % #2 is the list of argument names
        !          7515: % #3 is the list of argument values
        !          7516: \def\getargvals@#1#2#3{%
        !          7517:   \def\macargdeflist@{}%
        !          7518:   \def\saveparamlist@{#2}% Need to keep a copy for parameter expansion.
        !          7519:   \def\paramlist{#2,\nil@}%
        !          7520:   \def\macroname{#1}%
        !          7521:   \begingroup
        !          7522:   \macroargctxt
        !          7523:   \def\argvaluelist{#3,\nil@}%
        !          7524:   \def\@tempa{#3}%
        !          7525:   \ifx\@tempa\empty
        !          7526:     \setemptyargvalues@
        !          7527:   \else
        !          7528:     \getargvals@@
        !          7529:   \fi
        !          7530: }
        !          7531: 
        !          7532: % 
        !          7533: \def\getargvals@@{%
        !          7534:   \ifx\paramlist\nilm@
        !          7535:       % Some sanity check needed here that \argvaluelist is also empty.
        !          7536:       \ifx\argvaluelist\nillm@
        !          7537:       \else
        !          7538:         \errhelp = \EMsimple
        !          7539:         \errmessage{Too many arguments in macro `\macroname'!}%
        !          7540:       \fi
        !          7541:       \let\next\macargexpandinbody@
        !          7542:   \else
        !          7543:     \ifx\argvaluelist\nillm@
        !          7544:        % No more arguments values passed to macro.  Set remaining named-arg
        !          7545:        % macros to empty.
        !          7546:        \let\next\setemptyargvalues@
        !          7547:     \else
        !          7548:       % pop current arg name into \@tempb
        !          7549:       \def\@tempa##1{\pop@{\@tempb}{\paramlist}##1\endargs@}%
        !          7550:       \expandafter\@tempa\expandafter{\paramlist}%
        !          7551:        % pop current argument value into \@tempc
        !          7552:       \def\@tempa##1{\longpop@{\@tempc}{\argvaluelist}##1\endargs@}%
        !          7553:       \expandafter\@tempa\expandafter{\argvaluelist}%
        !          7554:        % Here \@tempb is the current arg name and \@tempc is the current arg value.
        !          7555:        % First place the new argument macro definition into \@tempd
        !          7556:        \expandafter\macname\expandafter{\@tempc}%
        !          7557:        \expandafter\let\csname macarg.\@tempb\endcsname\relax
        !          7558:        \expandafter\def\expandafter\@tempe\expandafter{%
        !          7559:          \csname macarg.\@tempb\endcsname}%
        !          7560:        \edef\@tempd{\long\def\@tempe{\the\macname}}%
        !          7561:        \push@\@tempd\macargdeflist@
        !          7562:        \let\next\getargvals@@
        !          7563:     \fi
        !          7564:   \fi
        !          7565:   \next
        !          7566: }
        !          7567: 
        !          7568: \def\push@#1#2{%
        !          7569:   \expandafter\expandafter\expandafter\def
        !          7570:   \expandafter\expandafter\expandafter#2%
        !          7571:   \expandafter\expandafter\expandafter{%
        !          7572:   \expandafter#1#2}%
        !          7573: }
        !          7574: 
        !          7575: % Replace arguments by their values in the macro body, and place the result
        !          7576: % in macro \@tempa
        !          7577: \def\macvalstoargs@{%
        !          7578:   %  To do this we use the property that token registers that are \the'ed
        !          7579:   % within an \edef  expand only once. So we are going to place all argument
        !          7580:   % values into respective token registers.
        !          7581:   %
        !          7582:   % First we save the token context, and initialize argument numbering.
        !          7583:   \begingroup
        !          7584:     \paramno0\relax
        !          7585:     % Then, for each argument number #N, we place the corresponding argument
        !          7586:     % value into a new token list register \toks#N
        !          7587:     \expandafter\putargsintokens@\saveparamlist@,;,%
        !          7588:     % Then, we expand the body so that argument are replaced by their
        !          7589:     % values. The trick for values not to be expanded themselves is that they
        !          7590:     % are within tokens and that tokens expand only once in an \edef .
        !          7591:     \edef\@tempc{\csname mac.\macroname .body\endcsname}%
        !          7592:     % Now we restore the token stack pointer to free the token list registers
        !          7593:     % which we have used, but we make sure that expanded body is saved after
        !          7594:     % group.
        !          7595:     \expandafter
        !          7596:   \endgroup
        !          7597:   \expandafter\def\expandafter\@tempa\expandafter{\@tempc}%
        !          7598:   }
        !          7599: 
        !          7600: \def\macargexpandinbody@{% 
        !          7601:   %% Define the named-macro outside of this group and then close this group. 
        !          7602:   \expandafter
        !          7603:   \endgroup
        !          7604:   \macargdeflist@
        !          7605:   % First the replace in body the macro arguments by their values, the result
        !          7606:   % is in \@tempa .
        !          7607:   \macvalstoargs@
        !          7608:   % Then we point at the \norecurse or \gobble (for recursive) macro value
        !          7609:   % with \@tempb .
        !          7610:   \expandafter\let\expandafter\@tempb\csname mac.\macroname .recurse\endcsname
        !          7611:   % Depending on whether it is recursive or not, we need some tailing
        !          7612:   % \egroup .
        !          7613:   \ifx\@tempb\gobble
        !          7614:      \let\@tempc\relax
        !          7615:   \else
        !          7616:      \let\@tempc\egroup
        !          7617:   \fi
        !          7618:   % And now we do the real job:
        !          7619:   \edef\@tempd{\noexpand\@tempb{\macroname}\noexpand\scanmacro{\@tempa}\@tempc}%
        !          7620:   \@tempd
        !          7621: }
        !          7622: 
        !          7623: \def\putargsintokens@#1,{%
        !          7624:   \if#1;\let\next\relax
        !          7625:   \else
        !          7626:     \let\next\putargsintokens@
        !          7627:     % First we allocate the new token list register, and give it a temporary
        !          7628:     % alias \@tempb .
        !          7629:     \toksdef\@tempb\the\paramno
        !          7630:     % Then we place the argument value into that token list register.
        !          7631:     \expandafter\let\expandafter\@tempa\csname macarg.#1\endcsname
        !          7632:     \expandafter\@tempb\expandafter{\@tempa}%
        !          7633:     \advance\paramno by 1\relax
        !          7634:   \fi
        !          7635:   \next
        !          7636: }
        !          7637: 
        !          7638: % Save the token stack pointer into macro #1
        !          7639: \def\texisavetoksstackpoint#1{\edef#1{\the\@cclvi}}
        !          7640: % Restore the token stack pointer from number in macro #1
        !          7641: \def\texirestoretoksstackpoint#1{\expandafter\mathchardef\expandafter\@cclvi#1\relax}
        !          7642: % newtoks that can be used non \outer .
        !          7643: \def\texinonouternewtoks{\alloc@ 5\toks \toksdef \@cclvi}
        !          7644: 
        !          7645: % Tailing missing arguments are set to empty
        !          7646: \def\setemptyargvalues@{%
        !          7647:   \ifx\paramlist\nilm@
        !          7648:     \let\next\macargexpandinbody@
        !          7649:   \else
        !          7650:     \expandafter\setemptyargvaluesparser@\paramlist\endargs@
        !          7651:     \let\next\setemptyargvalues@
        !          7652:   \fi
        !          7653:   \next
        !          7654: }
        !          7655: 
        !          7656: \def\setemptyargvaluesparser@#1,#2\endargs@{%
        !          7657:   \expandafter\def\expandafter\@tempa\expandafter{%
        !          7658:     \expandafter\def\csname macarg.#1\endcsname{}}%
        !          7659:   \push@\@tempa\macargdeflist@
        !          7660:   \def\paramlist{#2}%
        !          7661: }
        !          7662: 
        !          7663: % #1 is the element target macro
        !          7664: % #2 is the list macro
        !          7665: % #3,#4\endargs@ is the list value
        !          7666: \def\pop@#1#2#3,#4\endargs@{%
        !          7667:    \def#1{#3}%
        !          7668:    \def#2{#4}%
        !          7669: }
        !          7670: \long\def\longpop@#1#2#3,#4\endargs@{%
        !          7671:    \long\def#1{#3}%
        !          7672:    \long\def#2{#4}%
        !          7673: }
        !          7674: 
        !          7675: % This defines a Texinfo @macro. There are eight cases: recursive and
        !          7676: % nonrecursive macros of zero, one, up to nine, and many arguments.
        !          7677: % Much magic with \expandafter here.
        !          7678: % \xdef is used so that macro definitions will survive the file
        !          7679: % they're defined in; @include reads the file inside a group.
        !          7680: %
        !          7681: \def\defmacro{%
        !          7682:   \let\hash=##% convert placeholders to macro parameter chars
        !          7683:   \ifrecursive
        !          7684:     \ifcase\paramno
        !          7685:     % 0
        !          7686:       \expandafter\xdef\csname\the\macname\endcsname{%
        !          7687:         \noexpand\scanmacro{\temp}}%
        !          7688:     \or % 1
        !          7689:       \expandafter\xdef\csname\the\macname\endcsname{%
        !          7690:          \bgroup\noexpand\macroargctxt
        !          7691:          \noexpand\braceorline
        !          7692:          \expandafter\noexpand\csname\the\macname xxx\endcsname}%
        !          7693:       \expandafter\xdef\csname\the\macname xxx\endcsname##1{%
        !          7694:          \egroup\noexpand\scanmacro{\temp}}%
        !          7695:     \else
        !          7696:       \ifnum\paramno<10\relax % at most 9
        !          7697:         \expandafter\xdef\csname\the\macname\endcsname{%
        !          7698:            \bgroup\noexpand\macroargctxt
        !          7699:            \noexpand\csname\the\macname xx\endcsname}%
        !          7700:         \expandafter\xdef\csname\the\macname xx\endcsname##1{%
        !          7701:             \expandafter\noexpand\csname\the\macname xxx\endcsname ##1,}%
        !          7702:         \expandafter\expandafter
        !          7703:         \expandafter\xdef
        !          7704:         \expandafter\expandafter
        !          7705:           \csname\the\macname xxx\endcsname
        !          7706:             \paramlist{\egroup\noexpand\scanmacro{\temp}}%
        !          7707:       \else % 10 or more
        !          7708:         \expandafter\xdef\csname\the\macname\endcsname{%
        !          7709:           \noexpand\getargvals@{\the\macname}{\argl}%
        !          7710:         }%    
        !          7711:         \global\expandafter\let\csname mac.\the\macname .body\endcsname\temp
        !          7712:         \global\expandafter\let\csname mac.\the\macname .recurse\endcsname\gobble
        !          7713:       \fi
        !          7714:     \fi
        !          7715:   \else
        !          7716:     \ifcase\paramno
        !          7717:     % 0
        !          7718:       \expandafter\xdef\csname\the\macname\endcsname{%
        !          7719:         \noexpand\norecurse{\the\macname}%
        !          7720:         \noexpand\scanmacro{\temp}\egroup}%
        !          7721:     \or % 1
        !          7722:       \expandafter\xdef\csname\the\macname\endcsname{%
        !          7723:          \bgroup\noexpand\macroargctxt
        !          7724:          \noexpand\braceorline
        !          7725:          \expandafter\noexpand\csname\the\macname xxx\endcsname}%
        !          7726:       \expandafter\xdef\csname\the\macname xxx\endcsname##1{%
        !          7727:         \egroup
        !          7728:         \noexpand\norecurse{\the\macname}%
        !          7729:         \noexpand\scanmacro{\temp}\egroup}%
        !          7730:     \else % at most 9
        !          7731:       \ifnum\paramno<10\relax
        !          7732:         \expandafter\xdef\csname\the\macname\endcsname{%
        !          7733:            \bgroup\noexpand\macroargctxt
        !          7734:            \expandafter\noexpand\csname\the\macname xx\endcsname}%
        !          7735:         \expandafter\xdef\csname\the\macname xx\endcsname##1{%
        !          7736:             \expandafter\noexpand\csname\the\macname xxx\endcsname ##1,}%
        !          7737:         \expandafter\expandafter
        !          7738:         \expandafter\xdef
        !          7739:         \expandafter\expandafter
        !          7740:         \csname\the\macname xxx\endcsname
        !          7741:         \paramlist{%
        !          7742:             \egroup
        !          7743:             \noexpand\norecurse{\the\macname}%
        !          7744:             \noexpand\scanmacro{\temp}\egroup}%
        !          7745:       \else % 10 or more:
        !          7746:         \expandafter\xdef\csname\the\macname\endcsname{%
        !          7747:           \noexpand\getargvals@{\the\macname}{\argl}%
        !          7748:         }%
        !          7749:         \global\expandafter\let\csname mac.\the\macname .body\endcsname\temp
        !          7750:         \global\expandafter\let\csname mac.\the\macname .recurse\endcsname\norecurse
        !          7751:       \fi
        !          7752:     \fi
        !          7753:   \fi}
        !          7754: 
        !          7755: \catcode `\@\texiatcatcode\relax
        !          7756: 
        !          7757: \def\norecurse#1{\bgroup\cslet{#1}{macsave.#1}}
        !          7758: 
        !          7759: % \braceorline decides whether the next nonwhitespace character is a
        !          7760: % {.  If so it reads up to the closing }, if not, it reads the whole
        !          7761: % line.  Whatever was read is then fed to the next control sequence
        !          7762: % as an argument (by \parsebrace or \parsearg).
        !          7763: % 
        !          7764: \def\braceorline#1{\let\macnamexxx=#1\futurelet\nchar\braceorlinexxx}
        !          7765: \def\braceorlinexxx{%
        !          7766:   \ifx\nchar\bgroup\else
        !          7767:     \expandafter\parsearg
        !          7768:   \fi \macnamexxx}
        !          7769: 
        !          7770: 
        !          7771: % @alias.
        !          7772: % We need some trickery to remove the optional spaces around the equal
        !          7773: % sign.  Make them active and then expand them all to nothing.
        !          7774: %
        !          7775: \def\alias{\parseargusing\obeyspaces\aliasxxx}
        !          7776: \def\aliasxxx #1{\aliasyyy#1\relax}
        !          7777: \def\aliasyyy #1=#2\relax{%
        !          7778:   {%
        !          7779:     \expandafter\let\obeyedspace=\empty
        !          7780:     \addtomacrolist{#1}%
        !          7781:     \xdef\next{\global\let\makecsname{#1}=\makecsname{#2}}%
        !          7782:   }%
        !          7783:   \next
        !          7784: }
        !          7785: 
        !          7786: 
        !          7787: \message{cross references,}
        !          7788: 
        !          7789: \newwrite\auxfile
        !          7790: \newif\ifhavexrefs    % True if xref values are known.
        !          7791: \newif\ifwarnedxrefs  % True if we warned once that they aren't known.
        !          7792: 
        !          7793: % @inforef is relatively simple.
        !          7794: \def\inforef #1{\inforefzzz #1,,,,**}
        !          7795: \def\inforefzzz #1,#2,#3,#4**{%
        !          7796:   \putwordSee{} \putwordInfo{} \putwordfile{} \file{\ignorespaces #3{}},
        !          7797:   node \samp{\ignorespaces#1{}}}
        !          7798: 
        !          7799: % @node's only job in TeX is to define \lastnode, which is used in
        !          7800: % cross-references.  The @node line might or might not have commas, and
        !          7801: % might or might not have spaces before the first comma, like:
        !          7802: % @node foo , bar , ...
        !          7803: % We don't want such trailing spaces in the node name.
        !          7804: %
        !          7805: \parseargdef\node{\checkenv{}\donode #1 ,\finishnodeparse}
        !          7806: %
        !          7807: % also remove a trailing comma, in case of something like this:
        !          7808: % @node Help-Cross,  ,  , Cross-refs
        !          7809: \def\donode#1 ,#2\finishnodeparse{\dodonode #1,\finishnodeparse}
        !          7810: \def\dodonode#1,#2\finishnodeparse{\gdef\lastnode{#1}}
        !          7811: 
        !          7812: \let\nwnode=\node
        !          7813: \let\lastnode=\empty
        !          7814: 
        !          7815: % Write a cross-reference definition for the current node.  #1 is the
        !          7816: % type (Ynumbered, Yappendix, Ynothing).
        !          7817: %
        !          7818: \def\donoderef#1{%
        !          7819:   \ifx\lastnode\empty\else
        !          7820:     \setref{\lastnode}{#1}%
        !          7821:     \global\let\lastnode=\empty
        !          7822:   \fi
        !          7823: }
        !          7824: 
        !          7825: % @anchor{NAME} -- define xref target at arbitrary point.
        !          7826: %
        !          7827: \newcount\savesfregister
        !          7828: %
        !          7829: \def\savesf{\relax \ifhmode \savesfregister=\spacefactor \fi}
        !          7830: \def\restoresf{\relax \ifhmode \spacefactor=\savesfregister \fi}
        !          7831: \def\anchor#1{\savesf \setref{#1}{Ynothing}\restoresf \ignorespaces}
        !          7832: 
        !          7833: % \setref{NAME}{SNT} defines a cross-reference point NAME (a node or an
        !          7834: % anchor), which consists of three parts:
        !          7835: % 1) NAME-title - the current sectioning name taken from \lastsection,
        !          7836: %                 or the anchor name.
        !          7837: % 2) NAME-snt   - section number and type, passed as the SNT arg, or
        !          7838: %                 empty for anchors.
        !          7839: % 3) NAME-pg    - the page number.
        !          7840: %
        !          7841: % This is called from \donoderef, \anchor, and \dofloat.  In the case of
        !          7842: % floats, there is an additional part, which is not written here:
        !          7843: % 4) NAME-lof   - the text as it should appear in a @listoffloats.
        !          7844: %
        !          7845: \def\setref#1#2{%
        !          7846:   \pdfmkdest{#1}%
        !          7847:   \iflinks
        !          7848:     {%
        !          7849:       \atdummies  % preserve commands, but don't expand them
        !          7850:       \edef\writexrdef##1##2{%
        !          7851:        \write\auxfile{@xrdef{#1-% #1 of \setref, expanded by the \edef
        !          7852:          ##1}{##2}}% these are parameters of \writexrdef
        !          7853:       }%
        !          7854:       \toks0 = \expandafter{\lastsection}%
        !          7855:       \immediate \writexrdef{title}{\the\toks0 }%
        !          7856:       \immediate \writexrdef{snt}{\csname #2\endcsname}% \Ynumbered etc.
        !          7857:       \safewhatsit{\writexrdef{pg}{\folio}}% will be written later, at \shipout
        !          7858:     }%
        !          7859:   \fi
        !          7860: }
        !          7861: 
        !          7862: % @xrefautosectiontitle on|off says whether @section(ing) names are used
        !          7863: % automatically in xrefs, if the third arg is not explicitly specified.
        !          7864: % This was provided as a "secret" @set xref-automatic-section-title
        !          7865: % variable, now it's official.
        !          7866: % 
        !          7867: \parseargdef\xrefautomaticsectiontitle{%
        !          7868:   \def\temp{#1}%
        !          7869:   \ifx\temp\onword
        !          7870:     \expandafter\let\csname SETxref-automatic-section-title\endcsname
        !          7871:       = \empty
        !          7872:   \else\ifx\temp\offword
        !          7873:     \expandafter\let\csname SETxref-automatic-section-title\endcsname
        !          7874:       = \relax
        !          7875:   \else
        !          7876:     \errhelp = \EMsimple
        !          7877:     \errmessage{Unknown @xrefautomaticsectiontitle value `\temp',
        !          7878:                 must be on|off}%
        !          7879:   \fi\fi
        !          7880: }
        !          7881: 
        !          7882: % 
        !          7883: % @xref, @pxref, and @ref generate cross-references.  For \xrefX, #1 is
        !          7884: % the node name, #2 the name of the Info cross-reference, #3 the printed
        !          7885: % node name, #4 the name of the Info file, #5 the name of the printed
        !          7886: % manual.  All but the node name can be omitted.
        !          7887: %
        !          7888: \def\pxref#1{\putwordsee{} \xrefX[#1,,,,,,,]}
        !          7889: \def\xref#1{\putwordSee{} \xrefX[#1,,,,,,,]}
        !          7890: \def\ref#1{\xrefX[#1,,,,,,,]}
        !          7891: %
        !          7892: \newbox\toprefbox
        !          7893: \newbox\printedrefnamebox
        !          7894: \newbox\infofilenamebox
        !          7895: \newbox\printedmanualbox
        !          7896: %
        !          7897: \def\xrefX[#1,#2,#3,#4,#5,#6]{\begingroup
        !          7898:   \unsepspaces
        !          7899:   %
        !          7900:   % Get args without leading/trailing spaces.
        !          7901:   \def\printedrefname{\ignorespaces #3}%
        !          7902:   \setbox\printedrefnamebox = \hbox{\printedrefname\unskip}%
        !          7903:   %
        !          7904:   \def\infofilename{\ignorespaces #4}%
        !          7905:   \setbox\infofilenamebox = \hbox{\infofilename\unskip}%
        !          7906:   %
        !          7907:   \def\printedmanual{\ignorespaces #5}%
        !          7908:   \setbox\printedmanualbox  = \hbox{\printedmanual\unskip}%
        !          7909:   %
        !          7910:   % If the printed reference name (arg #3) was not explicitly given in
        !          7911:   % the @xref, figure out what we want to use.
        !          7912:   \ifdim \wd\printedrefnamebox = 0pt
        !          7913:     % No printed node name was explicitly given.
        !          7914:     \expandafter\ifx\csname SETxref-automatic-section-title\endcsname \relax
        !          7915:       % Not auto section-title: use node name inside the square brackets.
        !          7916:       \def\printedrefname{\ignorespaces #1}%
        !          7917:     \else
        !          7918:       % Auto section-title: use chapter/section title inside
        !          7919:       % the square brackets if we have it.
        !          7920:       \ifdim \wd\printedmanualbox > 0pt
        !          7921:         % It is in another manual, so we don't have it; use node name.
        !          7922:         \def\printedrefname{\ignorespaces #1}%
        !          7923:       \else
        !          7924:         \ifhavexrefs
        !          7925:           % We (should) know the real title if we have the xref values.
        !          7926:           \def\printedrefname{\refx{#1-title}{}}%
        !          7927:         \else
        !          7928:           % Otherwise just copy the Info node name.
        !          7929:           \def\printedrefname{\ignorespaces #1}%
        !          7930:         \fi%
        !          7931:       \fi
        !          7932:     \fi
        !          7933:   \fi
        !          7934:   %
        !          7935:   % Make link in pdf output.
        !          7936:   \ifpdf
        !          7937:     {\indexnofonts
        !          7938:      \turnoffactive
        !          7939:      \makevalueexpandable
        !          7940:      % This expands tokens, so do it after making catcode changes, so _
        !          7941:      % etc. don't get their TeX definitions.  This ignores all spaces in
        !          7942:      % #4, including (wrongly) those in the middle of the filename.
        !          7943:      \getfilename{#4}%
        !          7944:      %
        !          7945:      % This (wrongly) does not take account of leading or trailing
        !          7946:      % spaces in #1, which should be ignored.
        !          7947:      \edef\pdfxrefdest{#1}%
        !          7948:      \ifx\pdfxrefdest\empty
        !          7949:        \def\pdfxrefdest{Top}% no empty targets
        !          7950:      \else
        !          7951:        \txiescapepdf\pdfxrefdest  % escape PDF special chars
        !          7952:      \fi
        !          7953:      %
        !          7954:      \leavevmode
        !          7955:      \startlink attr{/Border [0 0 0]}%
        !          7956:      \ifnum\filenamelength>0
        !          7957:        goto file{\the\filename.pdf} name{\pdfxrefdest}%
        !          7958:      \else
        !          7959:        goto name{\pdfmkpgn{\pdfxrefdest}}%
        !          7960:      \fi
        !          7961:     }%
        !          7962:     \setcolor{\linkcolor}%
        !          7963:   \fi
        !          7964:   %
        !          7965:   % Float references are printed completely differently: "Figure 1.2"
        !          7966:   % instead of "[somenode], p.3".  We distinguish them by the
        !          7967:   % LABEL-title being set to a magic string.
        !          7968:   {%
        !          7969:     % Have to otherify everything special to allow the \csname to
        !          7970:     % include an _ in the xref name, etc.
        !          7971:     \indexnofonts
        !          7972:     \turnoffactive
        !          7973:     \expandafter\global\expandafter\let\expandafter\Xthisreftitle
        !          7974:       \csname XR#1-title\endcsname
        !          7975:   }%
        !          7976:   \iffloat\Xthisreftitle
        !          7977:     % If the user specified the print name (third arg) to the ref,
        !          7978:     % print it instead of our usual "Figure 1.2".
        !          7979:     \ifdim\wd\printedrefnamebox = 0pt
        !          7980:       \refx{#1-snt}{}%
        !          7981:     \else
        !          7982:       \printedrefname
        !          7983:     \fi
        !          7984:     %
        !          7985:     % If the user also gave the printed manual name (fifth arg), append
        !          7986:     % "in MANUALNAME".
        !          7987:     \ifdim \wd\printedmanualbox > 0pt
        !          7988:       \space \putwordin{} \cite{\printedmanual}%
        !          7989:     \fi
        !          7990:   \else
        !          7991:     % node/anchor (non-float) references.
        !          7992:     % 
        !          7993:     % If we use \unhbox to print the node names, TeX does not insert
        !          7994:     % empty discretionaries after hyphens, which means that it will not
        !          7995:     % find a line break at a hyphen in a node names.  Since some manuals
        !          7996:     % are best written with fairly long node names, containing hyphens,
        !          7997:     % this is a loss.  Therefore, we give the text of the node name
        !          7998:     % again, so it is as if TeX is seeing it for the first time.
        !          7999:     % 
        !          8000:     \ifdim \wd\printedmanualbox > 0pt
        !          8001:       % Cross-manual reference with a printed manual name.
        !          8002:       % 
        !          8003:       \crossmanualxref{\cite{\printedmanual\unskip}}%
        !          8004:     %
        !          8005:     \else\ifdim \wd\infofilenamebox > 0pt
        !          8006:       % Cross-manual reference with only an info filename (arg 4), no
        !          8007:       % printed manual name (arg 5).  This is essentially the same as
        !          8008:       % the case above; we output the filename, since we have nothing else.
        !          8009:       % 
        !          8010:       \crossmanualxref{\code{\infofilename\unskip}}%
        !          8011:     %
        !          8012:     \else
        !          8013:       % Reference within this manual.
        !          8014:       %
        !          8015:       % _ (for example) has to be the character _ for the purposes of the
        !          8016:       % control sequence corresponding to the node, but it has to expand
        !          8017:       % into the usual \leavevmode...\vrule stuff for purposes of
        !          8018:       % printing. So we \turnoffactive for the \refx-snt, back on for the
        !          8019:       % printing, back off for the \refx-pg.
        !          8020:       {\turnoffactive
        !          8021:        % Only output a following space if the -snt ref is nonempty; for
        !          8022:        % @unnumbered and @anchor, it won't be.
        !          8023:        \setbox2 = \hbox{\ignorespaces \refx{#1-snt}{}}%
        !          8024:        \ifdim \wd2 > 0pt \refx{#1-snt}\space\fi
        !          8025:       }%
        !          8026:       % output the `[mynode]' via the macro below so it can be overridden.
        !          8027:       \xrefprintnodename\printedrefname
        !          8028:       %
        !          8029:       % But we always want a comma and a space:
        !          8030:       ,\space
        !          8031:       %
        !          8032:       % output the `page 3'.
        !          8033:       \turnoffactive \putwordpage\tie\refx{#1-pg}{}%
        !          8034:     \fi\fi
        !          8035:   \fi
        !          8036:   \endlink
        !          8037: \endgroup}
        !          8038: 
        !          8039: % Output a cross-manual xref to #1.  Used just above (twice).
        !          8040: % 
        !          8041: % Only include the text "Section ``foo'' in" if the foo is neither
        !          8042: % missing or Top.  Thus, @xref{,,,foo,The Foo Manual} outputs simply
        !          8043: % "see The Foo Manual", the idea being to refer to the whole manual.
        !          8044: % 
        !          8045: % But, this being TeX, we can't easily compare our node name against the
        !          8046: % string "Top" while ignoring the possible spaces before and after in
        !          8047: % the input.  By adding the arbitrary 7sp below, we make it much less
        !          8048: % likely that a real node name would have the same width as "Top" (e.g.,
        !          8049: % in a monospaced font).  Hopefully it will never happen in practice.
        !          8050: % 
        !          8051: % For the same basic reason, we retypeset the "Top" at every
        !          8052: % reference, since the current font is indeterminate.
        !          8053: % 
        !          8054: \def\crossmanualxref#1{%
        !          8055:   \setbox\toprefbox = \hbox{Top\kern7sp}%
        !          8056:   \setbox2 = \hbox{\ignorespaces \printedrefname \unskip \kern7sp}%
        !          8057:   \ifdim \wd2 > 7sp  % nonempty?
        !          8058:     \ifdim \wd2 = \wd\toprefbox \else  % same as Top?
        !          8059:       \putwordSection{} ``\printedrefname'' \putwordin{}\space
        !          8060:     \fi
        !          8061:   \fi
        !          8062:   #1%
        !          8063: }
        !          8064: 
        !          8065: % This macro is called from \xrefX for the `[nodename]' part of xref
        !          8066: % output.  It's a separate macro only so it can be changed more easily,
        !          8067: % since square brackets don't work well in some documents.  Particularly
        !          8068: % one that Bob is working on :).
        !          8069: %
        !          8070: \def\xrefprintnodename#1{[#1]}
        !          8071: 
        !          8072: % Things referred to by \setref.
        !          8073: %
        !          8074: \def\Ynothing{}
        !          8075: \def\Yomitfromtoc{}
        !          8076: \def\Ynumbered{%
        !          8077:   \ifnum\secno=0
        !          8078:     \putwordChapter@tie \the\chapno
        !          8079:   \else \ifnum\subsecno=0
        !          8080:     \putwordSection@tie \the\chapno.\the\secno
        !          8081:   \else \ifnum\subsubsecno=0
        !          8082:     \putwordSection@tie \the\chapno.\the\secno.\the\subsecno
        !          8083:   \else
        !          8084:     \putwordSection@tie \the\chapno.\the\secno.\the\subsecno.\the\subsubsecno
        !          8085:   \fi\fi\fi
        !          8086: }
        !          8087: \def\Yappendix{%
        !          8088:   \ifnum\secno=0
        !          8089:      \putwordAppendix@tie @char\the\appendixno{}%
        !          8090:   \else \ifnum\subsecno=0
        !          8091:      \putwordSection@tie @char\the\appendixno.\the\secno
        !          8092:   \else \ifnum\subsubsecno=0
        !          8093:     \putwordSection@tie @char\the\appendixno.\the\secno.\the\subsecno
        !          8094:   \else
        !          8095:     \putwordSection@tie
        !          8096:       @char\the\appendixno.\the\secno.\the\subsecno.\the\subsubsecno
        !          8097:   \fi\fi\fi
        !          8098: }
        !          8099: 
        !          8100: % Define \refx{NAME}{SUFFIX} to reference a cross-reference string named NAME.
        !          8101: % If its value is nonempty, SUFFIX is output afterward.
        !          8102: %
        !          8103: \def\refx#1#2{%
        !          8104:   {%
        !          8105:     \indexnofonts
        !          8106:     \otherbackslash
        !          8107:     \expandafter\global\expandafter\let\expandafter\thisrefX
        !          8108:       \csname XR#1\endcsname
        !          8109:   }%
        !          8110:   \ifx\thisrefX\relax
        !          8111:     % If not defined, say something at least.
        !          8112:     \angleleft un\-de\-fined\angleright
        !          8113:     \iflinks
        !          8114:       \ifhavexrefs
        !          8115:         {\toks0 = {#1}% avoid expansion of possibly-complex value
        !          8116:          \message{\linenumber Undefined cross reference `\the\toks0'.}}%
        !          8117:       \else
        !          8118:         \ifwarnedxrefs\else
        !          8119:           \global\warnedxrefstrue
        !          8120:           \message{Cross reference values unknown; you must run TeX again.}%
        !          8121:         \fi
        !          8122:       \fi
        !          8123:     \fi
        !          8124:   \else
        !          8125:     % It's defined, so just use it.
        !          8126:     \thisrefX
        !          8127:   \fi
        !          8128:   #2% Output the suffix in any case.
        !          8129: }
        !          8130: 
        !          8131: % This is the macro invoked by entries in the aux file.  Usually it's
        !          8132: % just a \def (we prepend XR to the control sequence name to avoid
        !          8133: % collisions).  But if this is a float type, we have more work to do.
        !          8134: %
        !          8135: \def\xrdef#1#2{%
        !          8136:   {% The node name might contain 8-bit characters, which in our current
        !          8137:    % implementation are changed to commands like @'e.  Don't let these
        !          8138:    % mess up the control sequence name.
        !          8139:     \indexnofonts
        !          8140:     \turnoffactive
        !          8141:     \xdef\safexrefname{#1}%
        !          8142:   }%
        !          8143:   %
        !          8144:   \expandafter\gdef\csname XR\safexrefname\endcsname{#2}% remember this xref
        !          8145:   %
        !          8146:   % Was that xref control sequence that we just defined for a float?
        !          8147:   \expandafter\iffloat\csname XR\safexrefname\endcsname
        !          8148:     % it was a float, and we have the (safe) float type in \iffloattype.
        !          8149:     \expandafter\let\expandafter\floatlist
        !          8150:       \csname floatlist\iffloattype\endcsname
        !          8151:     %
        !          8152:     % Is this the first time we've seen this float type?
        !          8153:     \expandafter\ifx\floatlist\relax
        !          8154:       \toks0 = {\do}% yes, so just \do
        !          8155:     \else
        !          8156:       % had it before, so preserve previous elements in list.
        !          8157:       \toks0 = \expandafter{\floatlist\do}%
        !          8158:     \fi
        !          8159:     %
        !          8160:     % Remember this xref in the control sequence \floatlistFLOATTYPE,
        !          8161:     % for later use in \listoffloats.
        !          8162:     \expandafter\xdef\csname floatlist\iffloattype\endcsname{\the\toks0
        !          8163:       {\safexrefname}}%
        !          8164:   \fi
        !          8165: }
        !          8166: 
        !          8167: % Read the last existing aux file, if any.  No error if none exists.
        !          8168: %
        !          8169: \def\tryauxfile{%
        !          8170:   \openin 1 \jobname.aux
        !          8171:   \ifeof 1 \else
        !          8172:     \readdatafile{aux}%
        !          8173:     \global\havexrefstrue
        !          8174:   \fi
        !          8175:   \closein 1
        !          8176: }
        !          8177: 
        !          8178: \def\setupdatafile{%
        !          8179:   \catcode`\^^@=\other
        !          8180:   \catcode`\^^A=\other
        !          8181:   \catcode`\^^B=\other
        !          8182:   \catcode`\^^C=\other
        !          8183:   \catcode`\^^D=\other
        !          8184:   \catcode`\^^E=\other
        !          8185:   \catcode`\^^F=\other
        !          8186:   \catcode`\^^G=\other
        !          8187:   \catcode`\^^H=\other
        !          8188:   \catcode`\^^K=\other
        !          8189:   \catcode`\^^L=\other
        !          8190:   \catcode`\^^N=\other
        !          8191:   \catcode`\^^P=\other
        !          8192:   \catcode`\^^Q=\other
        !          8193:   \catcode`\^^R=\other
        !          8194:   \catcode`\^^S=\other
        !          8195:   \catcode`\^^T=\other
        !          8196:   \catcode`\^^U=\other
        !          8197:   \catcode`\^^V=\other
        !          8198:   \catcode`\^^W=\other
        !          8199:   \catcode`\^^X=\other
        !          8200:   \catcode`\^^Z=\other
        !          8201:   \catcode`\^^[=\other
        !          8202:   \catcode`\^^\=\other
        !          8203:   \catcode`\^^]=\other
        !          8204:   \catcode`\^^^=\other
        !          8205:   \catcode`\^^_=\other
        !          8206:   % It was suggested to set the catcode of ^ to 7, which would allow ^^e4 etc.
        !          8207:   % in xref tags, i.e., node names.  But since ^^e4 notation isn't
        !          8208:   % supported in the main text, it doesn't seem desirable.  Furthermore,
        !          8209:   % that is not enough: for node names that actually contain a ^
        !          8210:   % character, we would end up writing a line like this: 'xrdef {'hat
        !          8211:   % b-title}{'hat b} and \xrdef does a \csname...\endcsname on the first
        !          8212:   % argument, and \hat is not an expandable control sequence.  It could
        !          8213:   % all be worked out, but why?  Either we support ^^ or we don't.
        !          8214:   %
        !          8215:   % The other change necessary for this was to define \auxhat:
        !          8216:   % \def\auxhat{\def^{'hat }}% extra space so ok if followed by letter
        !          8217:   % and then to call \auxhat in \setq.
        !          8218:   %
        !          8219:   \catcode`\^=\other
        !          8220:   %
        !          8221:   % Special characters.  Should be turned off anyway, but...
        !          8222:   \catcode`\~=\other
        !          8223:   \catcode`\[=\other
        !          8224:   \catcode`\]=\other
        !          8225:   \catcode`\"=\other
        !          8226:   \catcode`\_=\other
        !          8227:   \catcode`\|=\other
        !          8228:   \catcode`\<=\other
        !          8229:   \catcode`\>=\other
        !          8230:   \catcode`\$=\other
        !          8231:   \catcode`\#=\other
        !          8232:   \catcode`\&=\other
        !          8233:   \catcode`\%=\other
        !          8234:   \catcode`+=\other % avoid \+ for paranoia even though we've turned it off
        !          8235:   %
        !          8236:   % This is to support \ in node names and titles, since the \
        !          8237:   % characters end up in a \csname.  It's easier than
        !          8238:   % leaving it active and making its active definition an actual \
        !          8239:   % character.  What I don't understand is why it works in the *value*
        !          8240:   % of the xrdef.  Seems like it should be a catcode12 \, and that
        !          8241:   % should not typeset properly.  But it works, so I'm moving on for
        !          8242:   % now.  --karl, 15jan04.
        !          8243:   \catcode`\\=\other
        !          8244:   %
        !          8245:   % Make the characters 128-255 be printing characters.
        !          8246:   {%
        !          8247:     \count1=128
        !          8248:     \def\loop{%
        !          8249:       \catcode\count1=\other
        !          8250:       \advance\count1 by 1
        !          8251:       \ifnum \count1<256 \loop \fi
        !          8252:     }%
        !          8253:   }%
        !          8254:   %
        !          8255:   % @ is our escape character in .aux files, and we need braces.
        !          8256:   \catcode`\{=1
        !          8257:   \catcode`\}=2
        !          8258:   \catcode`\@=0
        !          8259: }
        !          8260: 
        !          8261: \def\readdatafile#1{%
        !          8262: \begingroup
        !          8263:   \setupdatafile
        !          8264:   \input\jobname.#1
        !          8265: \endgroup}
        !          8266: 
        !          8267: 
        !          8268: \message{insertions,}
        !          8269: % including footnotes.
        !          8270: 
        !          8271: \newcount \footnoteno
        !          8272: 
        !          8273: % The trailing space in the following definition for supereject is
        !          8274: % vital for proper filling; pages come out unaligned when you do a
        !          8275: % pagealignmacro call if that space before the closing brace is
        !          8276: % removed. (Generally, numeric constants should always be followed by a
        !          8277: % space to prevent strange expansion errors.)
        !          8278: \def\supereject{\par\penalty -20000\footnoteno =0 }
        !          8279: 
        !          8280: % @footnotestyle is meaningful for Info output only.
        !          8281: \let\footnotestyle=\comment
        !          8282: 
        !          8283: {\catcode `\@=11
        !          8284: %
        !          8285: % Auto-number footnotes.  Otherwise like plain.
        !          8286: \gdef\footnote{%
        !          8287:   \let\indent=\ptexindent
        !          8288:   \let\noindent=\ptexnoindent
        !          8289:   \global\advance\footnoteno by \@ne
        !          8290:   \edef\thisfootno{$^{\the\footnoteno}$}%
        !          8291:   %
        !          8292:   % In case the footnote comes at the end of a sentence, preserve the
        !          8293:   % extra spacing after we do the footnote number.
        !          8294:   \let\@sf\empty
        !          8295:   \ifhmode\edef\@sf{\spacefactor\the\spacefactor}\ptexslash\fi
        !          8296:   %
        !          8297:   % Remove inadvertent blank space before typesetting the footnote number.
        !          8298:   \unskip
        !          8299:   \thisfootno\@sf
        !          8300:   \dofootnote
        !          8301: }%
        !          8302: 
        !          8303: % Don't bother with the trickery in plain.tex to not require the
        !          8304: % footnote text as a parameter.  Our footnotes don't need to be so general.
        !          8305: %
        !          8306: % Oh yes, they do; otherwise, @ifset (and anything else that uses
        !          8307: % \parseargline) fails inside footnotes because the tokens are fixed when
        !          8308: % the footnote is read.  --karl, 16nov96.
        !          8309: %
        !          8310: \gdef\dofootnote{%
        !          8311:   \insert\footins\bgroup
        !          8312:   % We want to typeset this text as a normal paragraph, even if the
        !          8313:   % footnote reference occurs in (for example) a display environment.
        !          8314:   % So reset some parameters.
        !          8315:   \hsize=\pagewidth
        !          8316:   \interlinepenalty\interfootnotelinepenalty
        !          8317:   \splittopskip\ht\strutbox % top baseline for broken footnotes
        !          8318:   \splitmaxdepth\dp\strutbox
        !          8319:   \floatingpenalty\@MM
        !          8320:   \leftskip\z@skip
        !          8321:   \rightskip\z@skip
        !          8322:   \spaceskip\z@skip
        !          8323:   \xspaceskip\z@skip
        !          8324:   \parindent\defaultparindent
        !          8325:   %
        !          8326:   \smallfonts \rm
        !          8327:   %
        !          8328:   % Because we use hanging indentation in footnotes, a @noindent appears
        !          8329:   % to exdent this text, so make it be a no-op.  makeinfo does not use
        !          8330:   % hanging indentation so @noindent can still be needed within footnote
        !          8331:   % text after an @example or the like (not that this is good style).
        !          8332:   \let\noindent = \relax
        !          8333:   %
        !          8334:   % Hang the footnote text off the number.  Use \everypar in case the
        !          8335:   % footnote extends for more than one paragraph.
        !          8336:   \everypar = {\hang}%
        !          8337:   \textindent{\thisfootno}%
        !          8338:   %
        !          8339:   % Don't crash into the line above the footnote text.  Since this
        !          8340:   % expands into a box, it must come within the paragraph, lest it
        !          8341:   % provide a place where TeX can split the footnote.
        !          8342:   \footstrut
        !          8343:   %
        !          8344:   % Invoke rest of plain TeX footnote routine.
        !          8345:   \futurelet\next\fo@t
        !          8346: }
        !          8347: }%end \catcode `\@=11
        !          8348: 
        !          8349: % In case a @footnote appears in a vbox, save the footnote text and create
        !          8350: % the real \insert just after the vbox finished.  Otherwise, the insertion
        !          8351: % would be lost.
        !          8352: % Similarly, if a @footnote appears inside an alignment, save the footnote
        !          8353: % text to a box and make the \insert when a row of the table is finished.
        !          8354: % And the same can be done for other insert classes.  --kasal, 16nov03.
        !          8355: 
        !          8356: % Replace the \insert primitive by a cheating macro.
        !          8357: % Deeper inside, just make sure that the saved insertions are not spilled
        !          8358: % out prematurely.
        !          8359: %
        !          8360: \def\startsavinginserts{%
        !          8361:   \ifx \insert\ptexinsert
        !          8362:     \let\insert\saveinsert
        !          8363:   \else
        !          8364:     \let\checkinserts\relax
        !          8365:   \fi
        !          8366: }
        !          8367: 
        !          8368: % This \insert replacement works for both \insert\footins{foo} and
        !          8369: % \insert\footins\bgroup foo\egroup, but it doesn't work for \insert27{foo}.
        !          8370: %
        !          8371: \def\saveinsert#1{%
        !          8372:   \edef\next{\noexpand\savetobox \makeSAVEname#1}%
        !          8373:   \afterassignment\next
        !          8374:   % swallow the left brace
        !          8375:   \let\temp =
        !          8376: }
        !          8377: \def\makeSAVEname#1{\makecsname{SAVE\expandafter\gobble\string#1}}
        !          8378: \def\savetobox#1{\global\setbox#1 = \vbox\bgroup \unvbox#1}
        !          8379: 
        !          8380: \def\checksaveins#1{\ifvoid#1\else \placesaveins#1\fi}
        !          8381: 
        !          8382: \def\placesaveins#1{%
        !          8383:   \ptexinsert \csname\expandafter\gobblesave\string#1\endcsname
        !          8384:     {\box#1}%
        !          8385: }
        !          8386: 
        !          8387: % eat @SAVE -- beware, all of them have catcode \other:
        !          8388: {
        !          8389:   \def\dospecials{\do S\do A\do V\do E} \uncatcodespecials  %  ;-)
        !          8390:   \gdef\gobblesave @SAVE{}
        !          8391: }
        !          8392: 
        !          8393: % initialization:
        !          8394: \def\newsaveins #1{%
        !          8395:   \edef\next{\noexpand\newsaveinsX \makeSAVEname#1}%
        !          8396:   \next
        !          8397: }
        !          8398: \def\newsaveinsX #1{%
        !          8399:   \csname newbox\endcsname #1%
        !          8400:   \expandafter\def\expandafter\checkinserts\expandafter{\checkinserts
        !          8401:     \checksaveins #1}%
        !          8402: }
        !          8403: 
        !          8404: % initialize:
        !          8405: \let\checkinserts\empty
        !          8406: \newsaveins\footins
        !          8407: \newsaveins\margin
        !          8408: 
        !          8409: 
        !          8410: % @image.  We use the macros from epsf.tex to support this.
        !          8411: % If epsf.tex is not installed and @image is used, we complain.
        !          8412: %
        !          8413: % Check for and read epsf.tex up front.  If we read it only at @image
        !          8414: % time, we might be inside a group, and then its definitions would get
        !          8415: % undone and the next image would fail.
        !          8416: \openin 1 = epsf.tex
        !          8417: \ifeof 1 \else
        !          8418:   % Do not bother showing banner with epsf.tex v2.7k (available in
        !          8419:   % doc/epsf.tex and on ctan).
        !          8420:   \def\epsfannounce{\toks0 = }%
        !          8421:   \input epsf.tex
        !          8422: \fi
        !          8423: \closein 1
        !          8424: %
        !          8425: % We will only complain once about lack of epsf.tex.
        !          8426: \newif\ifwarnednoepsf
        !          8427: \newhelp\noepsfhelp{epsf.tex must be installed for images to
        !          8428:   work.  It is also included in the Texinfo distribution, or you can get
        !          8429:   it from ftp://tug.org/tex/epsf.tex.}
        !          8430: %
        !          8431: \def\image#1{%
        !          8432:   \ifx\epsfbox\thisisundefined
        !          8433:     \ifwarnednoepsf \else
        !          8434:       \errhelp = \noepsfhelp
        !          8435:       \errmessage{epsf.tex not found, images will be ignored}%
        !          8436:       \global\warnednoepsftrue
        !          8437:     \fi
        !          8438:   \else
        !          8439:     \imagexxx #1,,,,,\finish
        !          8440:   \fi
        !          8441: }
        !          8442: %
        !          8443: % Arguments to @image:
        !          8444: % #1 is (mandatory) image filename; we tack on .eps extension.
        !          8445: % #2 is (optional) width, #3 is (optional) height.
        !          8446: % #4 is (ignored optional) html alt text.
        !          8447: % #5 is (ignored optional) extension.
        !          8448: % #6 is just the usual extra ignored arg for parsing stuff.
        !          8449: \newif\ifimagevmode
        !          8450: \def\imagexxx#1,#2,#3,#4,#5,#6\finish{\begingroup
        !          8451:   \catcode`\^^M = 5     % in case we're inside an example
        !          8452:   \normalturnoffactive  % allow _ et al. in names
        !          8453:   % If the image is by itself, center it.
        !          8454:   \ifvmode
        !          8455:     \imagevmodetrue
        !          8456:   \else \ifx\centersub\centerV
        !          8457:     % for @center @image, we need a vbox so we can have our vertical space
        !          8458:     \imagevmodetrue
        !          8459:     \vbox\bgroup % vbox has better behavior than vtop herev
        !          8460:   \fi\fi
        !          8461:   %
        !          8462:   \ifimagevmode
        !          8463:     \nobreak\medskip
        !          8464:     % Usually we'll have text after the image which will insert
        !          8465:     % \parskip glue, so insert it here too to equalize the space
        !          8466:     % above and below.
        !          8467:     \nobreak\vskip\parskip
        !          8468:     \nobreak
        !          8469:   \fi
        !          8470:   %
        !          8471:   % Leave vertical mode so that indentation from an enclosing
        !          8472:   %  environment such as @quotation is respected.
        !          8473:   % However, if we're at the top level, we don't want the
        !          8474:   %  normal paragraph indentation.
        !          8475:   % On the other hand, if we are in the case of @center @image, we don't
        !          8476:   %  want to start a paragraph, which will create a hsize-width box and
        !          8477:   %  eradicate the centering.
        !          8478:   \ifx\centersub\centerV\else \noindent \fi
        !          8479:   %
        !          8480:   % Output the image.
        !          8481:   \ifpdf
        !          8482:     \dopdfimage{#1}{#2}{#3}%
        !          8483:   \else
        !          8484:     % \epsfbox itself resets \epsf?size at each figure.
        !          8485:     \setbox0 = \hbox{\ignorespaces #2}\ifdim\wd0 > 0pt \epsfxsize=#2\relax \fi
        !          8486:     \setbox0 = \hbox{\ignorespaces #3}\ifdim\wd0 > 0pt \epsfysize=#3\relax \fi
        !          8487:     \epsfbox{#1.eps}%
        !          8488:   \fi
        !          8489:   %
        !          8490:   \ifimagevmode
        !          8491:     \medskip  % space after a standalone image
        !          8492:   \fi  
        !          8493:   \ifx\centersub\centerV \egroup \fi
        !          8494: \endgroup}
        !          8495: 
        !          8496: 
        !          8497: % @float FLOATTYPE,LABEL,LOC ... @end float for displayed figures, tables,
        !          8498: % etc.  We don't actually implement floating yet, we always include the
        !          8499: % float "here".  But it seemed the best name for the future.
        !          8500: %
        !          8501: \envparseargdef\float{\eatcommaspace\eatcommaspace\dofloat#1, , ,\finish}
        !          8502: 
        !          8503: % There may be a space before second and/or third parameter; delete it.
        !          8504: \def\eatcommaspace#1, {#1,}
        !          8505: 
        !          8506: % #1 is the optional FLOATTYPE, the text label for this float, typically
        !          8507: % "Figure", "Table", "Example", etc.  Can't contain commas.  If omitted,
        !          8508: % this float will not be numbered and cannot be referred to.
        !          8509: %
        !          8510: % #2 is the optional xref label.  Also must be present for the float to
        !          8511: % be referable.
        !          8512: %
        !          8513: % #3 is the optional positioning argument; for now, it is ignored.  It
        !          8514: % will somehow specify the positions allowed to float to (here, top, bottom).
        !          8515: %
        !          8516: % We keep a separate counter for each FLOATTYPE, which we reset at each
        !          8517: % chapter-level command.
        !          8518: \let\resetallfloatnos=\empty
        !          8519: %
        !          8520: \def\dofloat#1,#2,#3,#4\finish{%
        !          8521:   \let\thiscaption=\empty
        !          8522:   \let\thisshortcaption=\empty
        !          8523:   %
        !          8524:   % don't lose footnotes inside @float.
        !          8525:   %
        !          8526:   % BEWARE: when the floats start float, we have to issue warning whenever an
        !          8527:   % insert appears inside a float which could possibly float. --kasal, 26may04
        !          8528:   %
        !          8529:   \startsavinginserts
        !          8530:   %
        !          8531:   % We can't be used inside a paragraph.
        !          8532:   \par
        !          8533:   %
        !          8534:   \vtop\bgroup
        !          8535:     \def\floattype{#1}%
        !          8536:     \def\floatlabel{#2}%
        !          8537:     \def\floatloc{#3}% we do nothing with this yet.
        !          8538:     %
        !          8539:     \ifx\floattype\empty
        !          8540:       \let\safefloattype=\empty
        !          8541:     \else
        !          8542:       {%
        !          8543:         % the floattype might have accents or other special characters,
        !          8544:         % but we need to use it in a control sequence name.
        !          8545:         \indexnofonts
        !          8546:         \turnoffactive
        !          8547:         \xdef\safefloattype{\floattype}%
        !          8548:       }%
        !          8549:     \fi
        !          8550:     %
        !          8551:     % If label is given but no type, we handle that as the empty type.
        !          8552:     \ifx\floatlabel\empty \else
        !          8553:       % We want each FLOATTYPE to be numbered separately (Figure 1,
        !          8554:       % Table 1, Figure 2, ...).  (And if no label, no number.)
        !          8555:       %
        !          8556:       \expandafter\getfloatno\csname\safefloattype floatno\endcsname
        !          8557:       \global\advance\floatno by 1
        !          8558:       %
        !          8559:       {%
        !          8560:         % This magic value for \lastsection is output by \setref as the
        !          8561:         % XREFLABEL-title value.  \xrefX uses it to distinguish float
        !          8562:         % labels (which have a completely different output format) from
        !          8563:         % node and anchor labels.  And \xrdef uses it to construct the
        !          8564:         % lists of floats.
        !          8565:         %
        !          8566:         \edef\lastsection{\floatmagic=\safefloattype}%
        !          8567:         \setref{\floatlabel}{Yfloat}%
        !          8568:       }%
        !          8569:     \fi
        !          8570:     %
        !          8571:     % start with \parskip glue, I guess.
        !          8572:     \vskip\parskip
        !          8573:     %
        !          8574:     % Don't suppress indentation if a float happens to start a section.
        !          8575:     \restorefirstparagraphindent
        !          8576: }
        !          8577: 
        !          8578: % we have these possibilities:
        !          8579: % @float Foo,lbl & @caption{Cap}: Foo 1.1: Cap
        !          8580: % @float Foo,lbl & no caption:    Foo 1.1
        !          8581: % @float Foo & @caption{Cap}:     Foo: Cap
        !          8582: % @float Foo & no caption:        Foo
        !          8583: % @float ,lbl & Caption{Cap}:     1.1: Cap
        !          8584: % @float ,lbl & no caption:       1.1
        !          8585: % @float & @caption{Cap}:         Cap
        !          8586: % @float & no caption:
        !          8587: %
        !          8588: \def\Efloat{%
        !          8589:     \let\floatident = \empty
        !          8590:     %
        !          8591:     % In all cases, if we have a float type, it comes first.
        !          8592:     \ifx\floattype\empty \else \def\floatident{\floattype}\fi
        !          8593:     %
        !          8594:     % If we have an xref label, the number comes next.
        !          8595:     \ifx\floatlabel\empty \else
        !          8596:       \ifx\floattype\empty \else % if also had float type, need tie first.
        !          8597:         \appendtomacro\floatident{\tie}%
        !          8598:       \fi
        !          8599:       % the number.
        !          8600:       \appendtomacro\floatident{\chaplevelprefix\the\floatno}%
        !          8601:     \fi
        !          8602:     %
        !          8603:     % Start the printed caption with what we've constructed in
        !          8604:     % \floatident, but keep it separate; we need \floatident again.
        !          8605:     \let\captionline = \floatident
        !          8606:     %
        !          8607:     \ifx\thiscaption\empty \else
        !          8608:       \ifx\floatident\empty \else
        !          8609:        \appendtomacro\captionline{: }% had ident, so need a colon between
        !          8610:       \fi
        !          8611:       %
        !          8612:       % caption text.
        !          8613:       \appendtomacro\captionline{\scanexp\thiscaption}%
        !          8614:     \fi
        !          8615:     %
        !          8616:     % If we have anything to print, print it, with space before.
        !          8617:     % Eventually this needs to become an \insert.
        !          8618:     \ifx\captionline\empty \else
        !          8619:       \vskip.5\parskip
        !          8620:       \captionline
        !          8621:       %
        !          8622:       % Space below caption.
        !          8623:       \vskip\parskip
        !          8624:     \fi
        !          8625:     %
        !          8626:     % If have an xref label, write the list of floats info.  Do this
        !          8627:     % after the caption, to avoid chance of it being a breakpoint.
        !          8628:     \ifx\floatlabel\empty \else
        !          8629:       % Write the text that goes in the lof to the aux file as
        !          8630:       % \floatlabel-lof.  Besides \floatident, we include the short
        !          8631:       % caption if specified, else the full caption if specified, else nothing.
        !          8632:       {%
        !          8633:         \atdummies
        !          8634:         %
        !          8635:         % since we read the caption text in the macro world, where ^^M
        !          8636:         % is turned into a normal character, we have to scan it back, so
        !          8637:         % we don't write the literal three characters "^^M" into the aux file.
        !          8638:        \scanexp{%
        !          8639:          \xdef\noexpand\gtemp{%
        !          8640:            \ifx\thisshortcaption\empty
        !          8641:              \thiscaption
        !          8642:            \else
        !          8643:              \thisshortcaption
        !          8644:            \fi
        !          8645:          }%
        !          8646:        }%
        !          8647:         \immediate\write\auxfile{@xrdef{\floatlabel-lof}{\floatident
        !          8648:          \ifx\gtemp\empty \else : \gtemp \fi}}%
        !          8649:       }%
        !          8650:     \fi
        !          8651:   \egroup  % end of \vtop
        !          8652:   %
        !          8653:   % place the captured inserts
        !          8654:   %
        !          8655:   % BEWARE: when the floats start floating, we have to issue warning
        !          8656:   % whenever an insert appears inside a float which could possibly
        !          8657:   % float. --kasal, 26may04
        !          8658:   %
        !          8659:   \checkinserts
        !          8660: }
        !          8661: 
        !          8662: % Append the tokens #2 to the definition of macro #1, not expanding either.
        !          8663: %
        !          8664: \def\appendtomacro#1#2{%
        !          8665:   \expandafter\def\expandafter#1\expandafter{#1#2}%
        !          8666: }
        !          8667: 
        !          8668: % @caption, @shortcaption
        !          8669: %
        !          8670: \def\caption{\docaption\thiscaption}
        !          8671: \def\shortcaption{\docaption\thisshortcaption}
        !          8672: \def\docaption{\checkenv\float \bgroup\scanargctxt\defcaption}
        !          8673: \def\defcaption#1#2{\egroup \def#1{#2}}
        !          8674: 
        !          8675: % The parameter is the control sequence identifying the counter we are
        !          8676: % going to use.  Create it if it doesn't exist and assign it to \floatno.
        !          8677: \def\getfloatno#1{%
        !          8678:   \ifx#1\relax
        !          8679:       % Haven't seen this figure type before.
        !          8680:       \csname newcount\endcsname #1%
        !          8681:       %
        !          8682:       % Remember to reset this floatno at the next chap.
        !          8683:       \expandafter\gdef\expandafter\resetallfloatnos
        !          8684:         \expandafter{\resetallfloatnos #1=0 }%
        !          8685:   \fi
        !          8686:   \let\floatno#1%
        !          8687: }
        !          8688: 
        !          8689: % \setref calls this to get the XREFLABEL-snt value.  We want an @xref
        !          8690: % to the FLOATLABEL to expand to "Figure 3.1".  We call \setref when we
        !          8691: % first read the @float command.
        !          8692: %
        !          8693: \def\Yfloat{\floattype@tie \chaplevelprefix\the\floatno}%
        !          8694: 
        !          8695: % Magic string used for the XREFLABEL-title value, so \xrefX can
        !          8696: % distinguish floats from other xref types.
        !          8697: \def\floatmagic{!!float!!}
        !          8698: 
        !          8699: % #1 is the control sequence we are passed; we expand into a conditional
        !          8700: % which is true if #1 represents a float ref.  That is, the magic
        !          8701: % \lastsection value which we \setref above.
        !          8702: %
        !          8703: \def\iffloat#1{\expandafter\doiffloat#1==\finish}
        !          8704: %
        !          8705: % #1 is (maybe) the \floatmagic string.  If so, #2 will be the
        !          8706: % (safe) float type for this float.  We set \iffloattype to #2.
        !          8707: %
        !          8708: \def\doiffloat#1=#2=#3\finish{%
        !          8709:   \def\temp{#1}%
        !          8710:   \def\iffloattype{#2}%
        !          8711:   \ifx\temp\floatmagic
        !          8712: }
        !          8713: 
        !          8714: % @listoffloats FLOATTYPE - print a list of floats like a table of contents.
        !          8715: %
        !          8716: \parseargdef\listoffloats{%
        !          8717:   \def\floattype{#1}% floattype
        !          8718:   {%
        !          8719:     % the floattype might have accents or other special characters,
        !          8720:     % but we need to use it in a control sequence name.
        !          8721:     \indexnofonts
        !          8722:     \turnoffactive
        !          8723:     \xdef\safefloattype{\floattype}%
        !          8724:   }%
        !          8725:   %
        !          8726:   % \xrdef saves the floats as a \do-list in \floatlistSAFEFLOATTYPE.
        !          8727:   \expandafter\ifx\csname floatlist\safefloattype\endcsname \relax
        !          8728:     \ifhavexrefs
        !          8729:       % if the user said @listoffloats foo but never @float foo.
        !          8730:       \message{\linenumber No `\safefloattype' floats to list.}%
        !          8731:     \fi
        !          8732:   \else
        !          8733:     \begingroup
        !          8734:       \leftskip=\tocindent  % indent these entries like a toc
        !          8735:       \let\do=\listoffloatsdo
        !          8736:       \csname floatlist\safefloattype\endcsname
        !          8737:     \endgroup
        !          8738:   \fi
        !          8739: }
        !          8740: 
        !          8741: % This is called on each entry in a list of floats.  We're passed the
        !          8742: % xref label, in the form LABEL-title, which is how we save it in the
        !          8743: % aux file.  We strip off the -title and look up \XRLABEL-lof, which
        !          8744: % has the text we're supposed to typeset here.
        !          8745: %
        !          8746: % Figures without xref labels will not be included in the list (since
        !          8747: % they won't appear in the aux file).
        !          8748: %
        !          8749: \def\listoffloatsdo#1{\listoffloatsdoentry#1\finish}
        !          8750: \def\listoffloatsdoentry#1-title\finish{{%
        !          8751:   % Can't fully expand XR#1-lof because it can contain anything.  Just
        !          8752:   % pass the control sequence.  On the other hand, XR#1-pg is just the
        !          8753:   % page number, and we want to fully expand that so we can get a link
        !          8754:   % in pdf output.
        !          8755:   \toksA = \expandafter{\csname XR#1-lof\endcsname}%
        !          8756:   %
        !          8757:   % use the same \entry macro we use to generate the TOC and index.
        !          8758:   \edef\writeentry{\noexpand\entry{\the\toksA}{\csname XR#1-pg\endcsname}}%
        !          8759:   \writeentry
        !          8760: }}
        !          8761: 
        !          8762: 
        !          8763: \message{localization,}
        !          8764: 
        !          8765: % For single-language documents, @documentlanguage is usually given very
        !          8766: % early, just after @documentencoding.  Single argument is the language
        !          8767: % (de) or locale (de_DE) abbreviation.
        !          8768: %
        !          8769: {
        !          8770:   \catcode`\_ = \active
        !          8771:   \globaldefs=1
        !          8772: \parseargdef\documentlanguage{\begingroup
        !          8773:   \let_=\normalunderscore  % normal _ character for filenames
        !          8774:   \tex % read txi-??.tex file in plain TeX.
        !          8775:     % Read the file by the name they passed if it exists.
        !          8776:     \openin 1 txi-#1.tex
        !          8777:     \ifeof 1
        !          8778:       \documentlanguagetrywithoutunderscore{#1_\finish}%
        !          8779:     \else
        !          8780:       \globaldefs = 1  % everything in the txi-LL files needs to persist
        !          8781:       \input txi-#1.tex
        !          8782:     \fi
        !          8783:     \closein 1
        !          8784:   \endgroup % end raw TeX
        !          8785: \endgroup}
        !          8786: %
        !          8787: % If they passed de_DE, and txi-de_DE.tex doesn't exist,
        !          8788: % try txi-de.tex.
        !          8789: %
        !          8790: \gdef\documentlanguagetrywithoutunderscore#1_#2\finish{%
        !          8791:   \openin 1 txi-#1.tex
        !          8792:   \ifeof 1
        !          8793:     \errhelp = \nolanghelp
        !          8794:     \errmessage{Cannot read language file txi-#1.tex}%
        !          8795:   \else
        !          8796:     \globaldefs = 1  % everything in the txi-LL files needs to persist
        !          8797:     \input txi-#1.tex
        !          8798:   \fi
        !          8799:   \closein 1
        !          8800: }
        !          8801: }% end of special _ catcode
        !          8802: %
        !          8803: \newhelp\nolanghelp{The given language definition file cannot be found or
        !          8804: is empty.  Maybe you need to install it?  Putting it in the current
        !          8805: directory should work if nowhere else does.}
        !          8806: 
        !          8807: % This macro is called from txi-??.tex files; the first argument is the
        !          8808: % \language name to set (without the "\lang@" prefix), the second and
        !          8809: % third args are \{left,right}hyphenmin.
        !          8810: %
        !          8811: % The language names to pass are determined when the format is built.
        !          8812: % See the etex.log file created at that time, e.g.,
        !          8813: % /usr/local/texlive/2008/texmf-var/web2c/pdftex/etex.log.
        !          8814: %
        !          8815: % With TeX Live 2008, etex now includes hyphenation patterns for all
        !          8816: % available languages.  This means we can support hyphenation in
        !          8817: % Texinfo, at least to some extent.  (This still doesn't solve the
        !          8818: % accented characters problem.)
        !          8819: %
        !          8820: \catcode`@=11
        !          8821: \def\txisetlanguage#1#2#3{%
        !          8822:   % do not set the language if the name is undefined in the current TeX.
        !          8823:   \expandafter\ifx\csname lang@#1\endcsname \relax
        !          8824:     \message{no patterns for #1}%
        !          8825:   \else
        !          8826:     \global\language = \csname lang@#1\endcsname
        !          8827:   \fi
        !          8828:   % but there is no harm in adjusting the hyphenmin values regardless.
        !          8829:   \global\lefthyphenmin = #2\relax
        !          8830:   \global\righthyphenmin = #3\relax
        !          8831: }
        !          8832: 
        !          8833: % Helpers for encodings.
        !          8834: % Set the catcode of characters 128 through 255 to the specified number.
        !          8835: %
        !          8836: \def\setnonasciicharscatcode#1{%
        !          8837:    \count255=128
        !          8838:    \loop\ifnum\count255<256
        !          8839:       \global\catcode\count255=#1\relax
        !          8840:       \advance\count255 by 1
        !          8841:    \repeat
        !          8842: }
        !          8843: 
        !          8844: \def\setnonasciicharscatcodenonglobal#1{%
        !          8845:    \count255=128
        !          8846:    \loop\ifnum\count255<256
        !          8847:       \catcode\count255=#1\relax
        !          8848:       \advance\count255 by 1
        !          8849:    \repeat
        !          8850: }
        !          8851: 
        !          8852: % @documentencoding sets the definition of non-ASCII characters
        !          8853: % according to the specified encoding.
        !          8854: %
        !          8855: \parseargdef\documentencoding{%
        !          8856:   % Encoding being declared for the document.
        !          8857:   \def\declaredencoding{\csname #1.enc\endcsname}%
        !          8858:   %
        !          8859:   % Supported encodings: names converted to tokens in order to be able
        !          8860:   % to compare them with \ifx.
        !          8861:   \def\ascii{\csname US-ASCII.enc\endcsname}%
        !          8862:   \def\latnine{\csname ISO-8859-15.enc\endcsname}%
        !          8863:   \def\latone{\csname ISO-8859-1.enc\endcsname}%
        !          8864:   \def\lattwo{\csname ISO-8859-2.enc\endcsname}%
        !          8865:   \def\utfeight{\csname UTF-8.enc\endcsname}%
        !          8866:   %
        !          8867:   \ifx \declaredencoding \ascii
        !          8868:      \asciichardefs
        !          8869:   %
        !          8870:   \else \ifx \declaredencoding \lattwo
        !          8871:      \setnonasciicharscatcode\active
        !          8872:      \lattwochardefs
        !          8873:   %
        !          8874:   \else \ifx \declaredencoding \latone
        !          8875:      \setnonasciicharscatcode\active
        !          8876:      \latonechardefs
        !          8877:   %
        !          8878:   \else \ifx \declaredencoding \latnine
        !          8879:      \setnonasciicharscatcode\active
        !          8880:      \latninechardefs
        !          8881:   %
        !          8882:   \else \ifx \declaredencoding \utfeight
        !          8883:      \setnonasciicharscatcode\active
        !          8884:      \utfeightchardefs
        !          8885:   %
        !          8886:   \else
        !          8887:     \message{Unknown document encoding #1, ignoring.}%
        !          8888:   %
        !          8889:   \fi % utfeight
        !          8890:   \fi % latnine
        !          8891:   \fi % latone
        !          8892:   \fi % lattwo
        !          8893:   \fi % ascii
        !          8894: }
        !          8895: 
        !          8896: % A message to be logged when using a character that isn't available
        !          8897: % the default font encoding (OT1).
        !          8898: %
        !          8899: \def\missingcharmsg#1{\message{Character missing in OT1 encoding: #1.}}
        !          8900: 
        !          8901: % Take account of \c (plain) vs. \, (Texinfo) difference.
        !          8902: \def\cedilla#1{\ifx\c\ptexc\c{#1}\else\,{#1}\fi}
        !          8903: 
        !          8904: % First, make active non-ASCII characters in order for them to be
        !          8905: % correctly categorized when TeX reads the replacement text of
        !          8906: % macros containing the character definitions.
        !          8907: \setnonasciicharscatcode\active
        !          8908: %
        !          8909: % Latin1 (ISO-8859-1) character definitions.
        !          8910: \def\latonechardefs{%
        !          8911:   \gdef^^a0{\tie}
        !          8912:   \gdef^^a1{\exclamdown}
        !          8913:   \gdef^^a2{\missingcharmsg{CENT SIGN}}
        !          8914:   \gdef^^a3{{\pounds}}
        !          8915:   \gdef^^a4{\missingcharmsg{CURRENCY SIGN}}
        !          8916:   \gdef^^a5{\missingcharmsg{YEN SIGN}}
        !          8917:   \gdef^^a6{\missingcharmsg{BROKEN BAR}}
        !          8918:   \gdef^^a7{\S}
        !          8919:   \gdef^^a8{\"{}}
        !          8920:   \gdef^^a9{\copyright}
        !          8921:   \gdef^^aa{\ordf}
        !          8922:   \gdef^^ab{\guillemetleft}
        !          8923:   \gdef^^ac{$\lnot$}
        !          8924:   \gdef^^ad{\-}
        !          8925:   \gdef^^ae{\registeredsymbol}
        !          8926:   \gdef^^af{\={}}
        !          8927:   %
        !          8928:   \gdef^^b0{\textdegree}
        !          8929:   \gdef^^b1{$\pm$}
        !          8930:   \gdef^^b2{$^2$}
        !          8931:   \gdef^^b3{$^3$}
        !          8932:   \gdef^^b4{\'{}}
        !          8933:   \gdef^^b5{$\mu$}
        !          8934:   \gdef^^b6{\P}
        !          8935:   %
        !          8936:   \gdef^^b7{$^.$}
        !          8937:   \gdef^^b8{\cedilla\ }
        !          8938:   \gdef^^b9{$^1$}
        !          8939:   \gdef^^ba{\ordm}
        !          8940:   %
        !          8941:   \gdef^^bb{\guillemetright}
        !          8942:   \gdef^^bc{$1\over4$}
        !          8943:   \gdef^^bd{$1\over2$}
        !          8944:   \gdef^^be{$3\over4$}
        !          8945:   \gdef^^bf{\questiondown}
        !          8946:   %
        !          8947:   \gdef^^c0{\`A}
        !          8948:   \gdef^^c1{\'A}
        !          8949:   \gdef^^c2{\^A}
        !          8950:   \gdef^^c3{\~A}
        !          8951:   \gdef^^c4{\"A}
        !          8952:   \gdef^^c5{\ringaccent A}
        !          8953:   \gdef^^c6{\AE}
        !          8954:   \gdef^^c7{\cedilla C}
        !          8955:   \gdef^^c8{\`E}
        !          8956:   \gdef^^c9{\'E}
        !          8957:   \gdef^^ca{\^E}
        !          8958:   \gdef^^cb{\"E}
        !          8959:   \gdef^^cc{\`I}
        !          8960:   \gdef^^cd{\'I}
        !          8961:   \gdef^^ce{\^I}
        !          8962:   \gdef^^cf{\"I}
        !          8963:   %
        !          8964:   \gdef^^d0{\DH}
        !          8965:   \gdef^^d1{\~N}
        !          8966:   \gdef^^d2{\`O}
        !          8967:   \gdef^^d3{\'O}
        !          8968:   \gdef^^d4{\^O}
        !          8969:   \gdef^^d5{\~O}
        !          8970:   \gdef^^d6{\"O}
        !          8971:   \gdef^^d7{$\times$}
        !          8972:   \gdef^^d8{\O}
        !          8973:   \gdef^^d9{\`U}
        !          8974:   \gdef^^da{\'U}
        !          8975:   \gdef^^db{\^U}
        !          8976:   \gdef^^dc{\"U}
        !          8977:   \gdef^^dd{\'Y}
        !          8978:   \gdef^^de{\TH}
        !          8979:   \gdef^^df{\ss}
        !          8980:   %
        !          8981:   \gdef^^e0{\`a}
        !          8982:   \gdef^^e1{\'a}
        !          8983:   \gdef^^e2{\^a}
        !          8984:   \gdef^^e3{\~a}
        !          8985:   \gdef^^e4{\"a}
        !          8986:   \gdef^^e5{\ringaccent a}
        !          8987:   \gdef^^e6{\ae}
        !          8988:   \gdef^^e7{\cedilla c}
        !          8989:   \gdef^^e8{\`e}
        !          8990:   \gdef^^e9{\'e}
        !          8991:   \gdef^^ea{\^e}
        !          8992:   \gdef^^eb{\"e}
        !          8993:   \gdef^^ec{\`{\dotless i}}
        !          8994:   \gdef^^ed{\'{\dotless i}}
        !          8995:   \gdef^^ee{\^{\dotless i}}
        !          8996:   \gdef^^ef{\"{\dotless i}}
        !          8997:   %
        !          8998:   \gdef^^f0{\dh}
        !          8999:   \gdef^^f1{\~n}
        !          9000:   \gdef^^f2{\`o}
        !          9001:   \gdef^^f3{\'o}
        !          9002:   \gdef^^f4{\^o}
        !          9003:   \gdef^^f5{\~o}
        !          9004:   \gdef^^f6{\"o}
        !          9005:   \gdef^^f7{$\div$}
        !          9006:   \gdef^^f8{\o}
        !          9007:   \gdef^^f9{\`u}
        !          9008:   \gdef^^fa{\'u}
        !          9009:   \gdef^^fb{\^u}
        !          9010:   \gdef^^fc{\"u}
        !          9011:   \gdef^^fd{\'y}
        !          9012:   \gdef^^fe{\th}
        !          9013:   \gdef^^ff{\"y}
        !          9014: }
        !          9015: 
        !          9016: % Latin9 (ISO-8859-15) encoding character definitions.
        !          9017: \def\latninechardefs{%
        !          9018:   % Encoding is almost identical to Latin1.
        !          9019:   \latonechardefs
        !          9020:   %
        !          9021:   \gdef^^a4{\euro}
        !          9022:   \gdef^^a6{\v S}
        !          9023:   \gdef^^a8{\v s}
        !          9024:   \gdef^^b4{\v Z}
        !          9025:   \gdef^^b8{\v z}
        !          9026:   \gdef^^bc{\OE}
        !          9027:   \gdef^^bd{\oe}
        !          9028:   \gdef^^be{\"Y}
        !          9029: }
        !          9030: 
        !          9031: % Latin2 (ISO-8859-2) character definitions.
        !          9032: \def\lattwochardefs{%
        !          9033:   \gdef^^a0{\tie}
        !          9034:   \gdef^^a1{\ogonek{A}}
        !          9035:   \gdef^^a2{\u{}}
        !          9036:   \gdef^^a3{\L}
        !          9037:   \gdef^^a4{\missingcharmsg{CURRENCY SIGN}}
        !          9038:   \gdef^^a5{\v L}
        !          9039:   \gdef^^a6{\'S}
        !          9040:   \gdef^^a7{\S}
        !          9041:   \gdef^^a8{\"{}}
        !          9042:   \gdef^^a9{\v S}
        !          9043:   \gdef^^aa{\cedilla S}
        !          9044:   \gdef^^ab{\v T}
        !          9045:   \gdef^^ac{\'Z}
        !          9046:   \gdef^^ad{\-}
        !          9047:   \gdef^^ae{\v Z}
        !          9048:   \gdef^^af{\dotaccent Z}
        !          9049:   %
        !          9050:   \gdef^^b0{\textdegree}
        !          9051:   \gdef^^b1{\ogonek{a}}
        !          9052:   \gdef^^b2{\ogonek{ }}
        !          9053:   \gdef^^b3{\l}
        !          9054:   \gdef^^b4{\'{}}
        !          9055:   \gdef^^b5{\v l}
        !          9056:   \gdef^^b6{\'s}
        !          9057:   \gdef^^b7{\v{}}
        !          9058:   \gdef^^b8{\cedilla\ }
        !          9059:   \gdef^^b9{\v s}
        !          9060:   \gdef^^ba{\cedilla s}
        !          9061:   \gdef^^bb{\v t}
        !          9062:   \gdef^^bc{\'z}
        !          9063:   \gdef^^bd{\H{}}
        !          9064:   \gdef^^be{\v z}
        !          9065:   \gdef^^bf{\dotaccent z}
        !          9066:   %
        !          9067:   \gdef^^c0{\'R}
        !          9068:   \gdef^^c1{\'A}
        !          9069:   \gdef^^c2{\^A}
        !          9070:   \gdef^^c3{\u A}
        !          9071:   \gdef^^c4{\"A}
        !          9072:   \gdef^^c5{\'L}
        !          9073:   \gdef^^c6{\'C}
        !          9074:   \gdef^^c7{\cedilla C}
        !          9075:   \gdef^^c8{\v C}
        !          9076:   \gdef^^c9{\'E}
        !          9077:   \gdef^^ca{\ogonek{E}}
        !          9078:   \gdef^^cb{\"E}
        !          9079:   \gdef^^cc{\v E}
        !          9080:   \gdef^^cd{\'I}
        !          9081:   \gdef^^ce{\^I}
        !          9082:   \gdef^^cf{\v D}
        !          9083:   %
        !          9084:   \gdef^^d0{\DH}
        !          9085:   \gdef^^d1{\'N}
        !          9086:   \gdef^^d2{\v N}
        !          9087:   \gdef^^d3{\'O}
        !          9088:   \gdef^^d4{\^O}
        !          9089:   \gdef^^d5{\H O}
        !          9090:   \gdef^^d6{\"O}
        !          9091:   \gdef^^d7{$\times$}
        !          9092:   \gdef^^d8{\v R}
        !          9093:   \gdef^^d9{\ringaccent U}
        !          9094:   \gdef^^da{\'U}
        !          9095:   \gdef^^db{\H U}
        !          9096:   \gdef^^dc{\"U}
        !          9097:   \gdef^^dd{\'Y}
        !          9098:   \gdef^^de{\cedilla T}
        !          9099:   \gdef^^df{\ss}
        !          9100:   %
        !          9101:   \gdef^^e0{\'r}
        !          9102:   \gdef^^e1{\'a}
        !          9103:   \gdef^^e2{\^a}
        !          9104:   \gdef^^e3{\u a}
        !          9105:   \gdef^^e4{\"a}
        !          9106:   \gdef^^e5{\'l}
        !          9107:   \gdef^^e6{\'c}
        !          9108:   \gdef^^e7{\cedilla c}
        !          9109:   \gdef^^e8{\v c}
        !          9110:   \gdef^^e9{\'e}
        !          9111:   \gdef^^ea{\ogonek{e}}
        !          9112:   \gdef^^eb{\"e}
        !          9113:   \gdef^^ec{\v e}
        !          9114:   \gdef^^ed{\'{\dotless{i}}}
        !          9115:   \gdef^^ee{\^{\dotless{i}}}
        !          9116:   \gdef^^ef{\v d}
        !          9117:   %
        !          9118:   \gdef^^f0{\dh}
        !          9119:   \gdef^^f1{\'n}
        !          9120:   \gdef^^f2{\v n}
        !          9121:   \gdef^^f3{\'o}
        !          9122:   \gdef^^f4{\^o}
        !          9123:   \gdef^^f5{\H o}
        !          9124:   \gdef^^f6{\"o}
        !          9125:   \gdef^^f7{$\div$}
        !          9126:   \gdef^^f8{\v r}
        !          9127:   \gdef^^f9{\ringaccent u}
        !          9128:   \gdef^^fa{\'u}
        !          9129:   \gdef^^fb{\H u}
        !          9130:   \gdef^^fc{\"u}
        !          9131:   \gdef^^fd{\'y}
        !          9132:   \gdef^^fe{\cedilla t}
        !          9133:   \gdef^^ff{\dotaccent{}}
        !          9134: }
        !          9135: 
        !          9136: % UTF-8 character definitions.
        !          9137: %
        !          9138: % This code to support UTF-8 is based on LaTeX's utf8.def, with some
        !          9139: % changes for Texinfo conventions.  It is included here under the GPL by
        !          9140: % permission from Frank Mittelbach and the LaTeX team.
        !          9141: %
        !          9142: \newcount\countUTFx
        !          9143: \newcount\countUTFy
        !          9144: \newcount\countUTFz
        !          9145: 
        !          9146: \gdef\UTFviiiTwoOctets#1#2{\expandafter
        !          9147:    \UTFviiiDefined\csname u8:#1\string #2\endcsname}
        !          9148: %
        !          9149: \gdef\UTFviiiThreeOctets#1#2#3{\expandafter
        !          9150:    \UTFviiiDefined\csname u8:#1\string #2\string #3\endcsname}
        !          9151: %
        !          9152: \gdef\UTFviiiFourOctets#1#2#3#4{\expandafter
        !          9153:    \UTFviiiDefined\csname u8:#1\string #2\string #3\string #4\endcsname}
        !          9154: 
        !          9155: \gdef\UTFviiiDefined#1{%
        !          9156:   \ifx #1\relax
        !          9157:     \message{\linenumber Unicode char \string #1 not defined for Texinfo}%
        !          9158:   \else
        !          9159:     \expandafter #1%
        !          9160:   \fi
        !          9161: }
        !          9162: 
        !          9163: \begingroup
        !          9164:   \catcode`\~13
        !          9165:   \catcode`\"12
        !          9166: 
        !          9167:   \def\UTFviiiLoop{%
        !          9168:     \global\catcode\countUTFx\active
        !          9169:     \uccode`\~\countUTFx
        !          9170:     \uppercase\expandafter{\UTFviiiTmp}%
        !          9171:     \advance\countUTFx by 1
        !          9172:     \ifnum\countUTFx < \countUTFy
        !          9173:       \expandafter\UTFviiiLoop
        !          9174:     \fi}
        !          9175: 
        !          9176:   \countUTFx = "C2
        !          9177:   \countUTFy = "E0
        !          9178:   \def\UTFviiiTmp{%
        !          9179:     \xdef~{\noexpand\UTFviiiTwoOctets\string~}}
        !          9180:   \UTFviiiLoop
        !          9181: 
        !          9182:   \countUTFx = "E0
        !          9183:   \countUTFy = "F0
        !          9184:   \def\UTFviiiTmp{%
        !          9185:     \xdef~{\noexpand\UTFviiiThreeOctets\string~}}
        !          9186:   \UTFviiiLoop
        !          9187: 
        !          9188:   \countUTFx = "F0
        !          9189:   \countUTFy = "F4
        !          9190:   \def\UTFviiiTmp{%
        !          9191:     \xdef~{\noexpand\UTFviiiFourOctets\string~}}
        !          9192:   \UTFviiiLoop
        !          9193: \endgroup
        !          9194: 
        !          9195: \begingroup
        !          9196:   \catcode`\"=12
        !          9197:   \catcode`\<=12
        !          9198:   \catcode`\.=12
        !          9199:   \catcode`\,=12
        !          9200:   \catcode`\;=12
        !          9201:   \catcode`\!=12
        !          9202:   \catcode`\~=13
        !          9203: 
        !          9204:   \gdef\DeclareUnicodeCharacter#1#2{%
        !          9205:     \countUTFz = "#1\relax
        !          9206:     %\wlog{\space\space defining Unicode char U+#1 (decimal \the\countUTFz)}%
        !          9207:     \begingroup
        !          9208:       \parseXMLCharref
        !          9209:       \def\UTFviiiTwoOctets##1##2{%
        !          9210:         \csname u8:##1\string ##2\endcsname}%
        !          9211:       \def\UTFviiiThreeOctets##1##2##3{%
        !          9212:         \csname u8:##1\string ##2\string ##3\endcsname}%
        !          9213:       \def\UTFviiiFourOctets##1##2##3##4{%
        !          9214:         \csname u8:##1\string ##2\string ##3\string ##4\endcsname}%
        !          9215:       \expandafter\expandafter\expandafter\expandafter
        !          9216:        \expandafter\expandafter\expandafter
        !          9217:        \gdef\UTFviiiTmp{#2}%
        !          9218:     \endgroup}
        !          9219: 
        !          9220:   \gdef\parseXMLCharref{%
        !          9221:     \ifnum\countUTFz < "A0\relax
        !          9222:       \errhelp = \EMsimple
        !          9223:       \errmessage{Cannot define Unicode char value < 00A0}%
        !          9224:     \else\ifnum\countUTFz < "800\relax
        !          9225:       \parseUTFviiiA,%
        !          9226:       \parseUTFviiiB C\UTFviiiTwoOctets.,%
        !          9227:     \else\ifnum\countUTFz < "10000\relax
        !          9228:       \parseUTFviiiA;%
        !          9229:       \parseUTFviiiA,%
        !          9230:       \parseUTFviiiB E\UTFviiiThreeOctets.{,;}%
        !          9231:     \else
        !          9232:       \parseUTFviiiA;%
        !          9233:       \parseUTFviiiA,%
        !          9234:       \parseUTFviiiA!%
        !          9235:       \parseUTFviiiB F\UTFviiiFourOctets.{!,;}%
        !          9236:     \fi\fi\fi
        !          9237:   }
        !          9238: 
        !          9239:   \gdef\parseUTFviiiA#1{%
        !          9240:     \countUTFx = \countUTFz
        !          9241:     \divide\countUTFz by 64
        !          9242:     \countUTFy = \countUTFz
        !          9243:     \multiply\countUTFz by 64
        !          9244:     \advance\countUTFx by -\countUTFz
        !          9245:     \advance\countUTFx by 128
        !          9246:     \uccode `#1\countUTFx
        !          9247:     \countUTFz = \countUTFy}
        !          9248: 
        !          9249:   \gdef\parseUTFviiiB#1#2#3#4{%
        !          9250:     \advance\countUTFz by "#10\relax
        !          9251:     \uccode `#3\countUTFz
        !          9252:     \uppercase{\gdef\UTFviiiTmp{#2#3#4}}}
        !          9253: \endgroup
        !          9254: 
        !          9255: \def\utfeightchardefs{%
        !          9256:   \DeclareUnicodeCharacter{00A0}{\tie}
        !          9257:   \DeclareUnicodeCharacter{00A1}{\exclamdown}
        !          9258:   \DeclareUnicodeCharacter{00A3}{\pounds}
        !          9259:   \DeclareUnicodeCharacter{00A8}{\"{ }}
        !          9260:   \DeclareUnicodeCharacter{00A9}{\copyright}
        !          9261:   \DeclareUnicodeCharacter{00AA}{\ordf}
        !          9262:   \DeclareUnicodeCharacter{00AB}{\guillemetleft}
        !          9263:   \DeclareUnicodeCharacter{00AD}{\-}
        !          9264:   \DeclareUnicodeCharacter{00AE}{\registeredsymbol}
        !          9265:   \DeclareUnicodeCharacter{00AF}{\={ }}
        !          9266: 
        !          9267:   \DeclareUnicodeCharacter{00B0}{\ringaccent{ }}
        !          9268:   \DeclareUnicodeCharacter{00B4}{\'{ }}
        !          9269:   \DeclareUnicodeCharacter{00B8}{\cedilla{ }}
        !          9270:   \DeclareUnicodeCharacter{00BA}{\ordm}
        !          9271:   \DeclareUnicodeCharacter{00BB}{\guillemetright}
        !          9272:   \DeclareUnicodeCharacter{00BF}{\questiondown}
        !          9273: 
        !          9274:   \DeclareUnicodeCharacter{00C0}{\`A}
        !          9275:   \DeclareUnicodeCharacter{00C1}{\'A}
        !          9276:   \DeclareUnicodeCharacter{00C2}{\^A}
        !          9277:   \DeclareUnicodeCharacter{00C3}{\~A}
        !          9278:   \DeclareUnicodeCharacter{00C4}{\"A}
        !          9279:   \DeclareUnicodeCharacter{00C5}{\AA}
        !          9280:   \DeclareUnicodeCharacter{00C6}{\AE}
        !          9281:   \DeclareUnicodeCharacter{00C7}{\cedilla{C}}
        !          9282:   \DeclareUnicodeCharacter{00C8}{\`E}
        !          9283:   \DeclareUnicodeCharacter{00C9}{\'E}
        !          9284:   \DeclareUnicodeCharacter{00CA}{\^E}
        !          9285:   \DeclareUnicodeCharacter{00CB}{\"E}
        !          9286:   \DeclareUnicodeCharacter{00CC}{\`I}
        !          9287:   \DeclareUnicodeCharacter{00CD}{\'I}
        !          9288:   \DeclareUnicodeCharacter{00CE}{\^I}
        !          9289:   \DeclareUnicodeCharacter{00CF}{\"I}
        !          9290: 
        !          9291:   \DeclareUnicodeCharacter{00D0}{\DH}
        !          9292:   \DeclareUnicodeCharacter{00D1}{\~N}
        !          9293:   \DeclareUnicodeCharacter{00D2}{\`O}
        !          9294:   \DeclareUnicodeCharacter{00D3}{\'O}
        !          9295:   \DeclareUnicodeCharacter{00D4}{\^O}
        !          9296:   \DeclareUnicodeCharacter{00D5}{\~O}
        !          9297:   \DeclareUnicodeCharacter{00D6}{\"O}
        !          9298:   \DeclareUnicodeCharacter{00D8}{\O}
        !          9299:   \DeclareUnicodeCharacter{00D9}{\`U}
        !          9300:   \DeclareUnicodeCharacter{00DA}{\'U}
        !          9301:   \DeclareUnicodeCharacter{00DB}{\^U}
        !          9302:   \DeclareUnicodeCharacter{00DC}{\"U}
        !          9303:   \DeclareUnicodeCharacter{00DD}{\'Y}
        !          9304:   \DeclareUnicodeCharacter{00DE}{\TH}
        !          9305:   \DeclareUnicodeCharacter{00DF}{\ss}
        !          9306: 
        !          9307:   \DeclareUnicodeCharacter{00E0}{\`a}
        !          9308:   \DeclareUnicodeCharacter{00E1}{\'a}
        !          9309:   \DeclareUnicodeCharacter{00E2}{\^a}
        !          9310:   \DeclareUnicodeCharacter{00E3}{\~a}
        !          9311:   \DeclareUnicodeCharacter{00E4}{\"a}
        !          9312:   \DeclareUnicodeCharacter{00E5}{\aa}
        !          9313:   \DeclareUnicodeCharacter{00E6}{\ae}
        !          9314:   \DeclareUnicodeCharacter{00E7}{\cedilla{c}}
        !          9315:   \DeclareUnicodeCharacter{00E8}{\`e}
        !          9316:   \DeclareUnicodeCharacter{00E9}{\'e}
        !          9317:   \DeclareUnicodeCharacter{00EA}{\^e}
        !          9318:   \DeclareUnicodeCharacter{00EB}{\"e}
        !          9319:   \DeclareUnicodeCharacter{00EC}{\`{\dotless{i}}}
        !          9320:   \DeclareUnicodeCharacter{00ED}{\'{\dotless{i}}}
        !          9321:   \DeclareUnicodeCharacter{00EE}{\^{\dotless{i}}}
        !          9322:   \DeclareUnicodeCharacter{00EF}{\"{\dotless{i}}}
        !          9323: 
        !          9324:   \DeclareUnicodeCharacter{00F0}{\dh}
        !          9325:   \DeclareUnicodeCharacter{00F1}{\~n}
        !          9326:   \DeclareUnicodeCharacter{00F2}{\`o}
        !          9327:   \DeclareUnicodeCharacter{00F3}{\'o}
        !          9328:   \DeclareUnicodeCharacter{00F4}{\^o}
        !          9329:   \DeclareUnicodeCharacter{00F5}{\~o}
        !          9330:   \DeclareUnicodeCharacter{00F6}{\"o}
        !          9331:   \DeclareUnicodeCharacter{00F8}{\o}
        !          9332:   \DeclareUnicodeCharacter{00F9}{\`u}
        !          9333:   \DeclareUnicodeCharacter{00FA}{\'u}
        !          9334:   \DeclareUnicodeCharacter{00FB}{\^u}
        !          9335:   \DeclareUnicodeCharacter{00FC}{\"u}
        !          9336:   \DeclareUnicodeCharacter{00FD}{\'y}
        !          9337:   \DeclareUnicodeCharacter{00FE}{\th}
        !          9338:   \DeclareUnicodeCharacter{00FF}{\"y}
        !          9339: 
        !          9340:   \DeclareUnicodeCharacter{0100}{\=A}
        !          9341:   \DeclareUnicodeCharacter{0101}{\=a}
        !          9342:   \DeclareUnicodeCharacter{0102}{\u{A}}
        !          9343:   \DeclareUnicodeCharacter{0103}{\u{a}}
        !          9344:   \DeclareUnicodeCharacter{0104}{\ogonek{A}}
        !          9345:   \DeclareUnicodeCharacter{0105}{\ogonek{a}}
        !          9346:   \DeclareUnicodeCharacter{0106}{\'C}
        !          9347:   \DeclareUnicodeCharacter{0107}{\'c}
        !          9348:   \DeclareUnicodeCharacter{0108}{\^C}
        !          9349:   \DeclareUnicodeCharacter{0109}{\^c}
        !          9350:   \DeclareUnicodeCharacter{0118}{\ogonek{E}}
        !          9351:   \DeclareUnicodeCharacter{0119}{\ogonek{e}}
        !          9352:   \DeclareUnicodeCharacter{010A}{\dotaccent{C}}
        !          9353:   \DeclareUnicodeCharacter{010B}{\dotaccent{c}}
        !          9354:   \DeclareUnicodeCharacter{010C}{\v{C}}
        !          9355:   \DeclareUnicodeCharacter{010D}{\v{c}}
        !          9356:   \DeclareUnicodeCharacter{010E}{\v{D}}
        !          9357: 
        !          9358:   \DeclareUnicodeCharacter{0112}{\=E}
        !          9359:   \DeclareUnicodeCharacter{0113}{\=e}
        !          9360:   \DeclareUnicodeCharacter{0114}{\u{E}}
        !          9361:   \DeclareUnicodeCharacter{0115}{\u{e}}
        !          9362:   \DeclareUnicodeCharacter{0116}{\dotaccent{E}}
        !          9363:   \DeclareUnicodeCharacter{0117}{\dotaccent{e}}
        !          9364:   \DeclareUnicodeCharacter{011A}{\v{E}}
        !          9365:   \DeclareUnicodeCharacter{011B}{\v{e}}
        !          9366:   \DeclareUnicodeCharacter{011C}{\^G}
        !          9367:   \DeclareUnicodeCharacter{011D}{\^g}
        !          9368:   \DeclareUnicodeCharacter{011E}{\u{G}}
        !          9369:   \DeclareUnicodeCharacter{011F}{\u{g}}
        !          9370: 
        !          9371:   \DeclareUnicodeCharacter{0120}{\dotaccent{G}}
        !          9372:   \DeclareUnicodeCharacter{0121}{\dotaccent{g}}
        !          9373:   \DeclareUnicodeCharacter{0124}{\^H}
        !          9374:   \DeclareUnicodeCharacter{0125}{\^h}
        !          9375:   \DeclareUnicodeCharacter{0128}{\~I}
        !          9376:   \DeclareUnicodeCharacter{0129}{\~{\dotless{i}}}
        !          9377:   \DeclareUnicodeCharacter{012A}{\=I}
        !          9378:   \DeclareUnicodeCharacter{012B}{\={\dotless{i}}}
        !          9379:   \DeclareUnicodeCharacter{012C}{\u{I}}
        !          9380:   \DeclareUnicodeCharacter{012D}{\u{\dotless{i}}}
        !          9381: 
        !          9382:   \DeclareUnicodeCharacter{0130}{\dotaccent{I}}
        !          9383:   \DeclareUnicodeCharacter{0131}{\dotless{i}}
        !          9384:   \DeclareUnicodeCharacter{0132}{IJ}
        !          9385:   \DeclareUnicodeCharacter{0133}{ij}
        !          9386:   \DeclareUnicodeCharacter{0134}{\^J}
        !          9387:   \DeclareUnicodeCharacter{0135}{\^{\dotless{j}}}
        !          9388:   \DeclareUnicodeCharacter{0139}{\'L}
        !          9389:   \DeclareUnicodeCharacter{013A}{\'l}
        !          9390: 
        !          9391:   \DeclareUnicodeCharacter{0141}{\L}
        !          9392:   \DeclareUnicodeCharacter{0142}{\l}
        !          9393:   \DeclareUnicodeCharacter{0143}{\'N}
        !          9394:   \DeclareUnicodeCharacter{0144}{\'n}
        !          9395:   \DeclareUnicodeCharacter{0147}{\v{N}}
        !          9396:   \DeclareUnicodeCharacter{0148}{\v{n}}
        !          9397:   \DeclareUnicodeCharacter{014C}{\=O}
        !          9398:   \DeclareUnicodeCharacter{014D}{\=o}
        !          9399:   \DeclareUnicodeCharacter{014E}{\u{O}}
        !          9400:   \DeclareUnicodeCharacter{014F}{\u{o}}
        !          9401: 
        !          9402:   \DeclareUnicodeCharacter{0150}{\H{O}}
        !          9403:   \DeclareUnicodeCharacter{0151}{\H{o}}
        !          9404:   \DeclareUnicodeCharacter{0152}{\OE}
        !          9405:   \DeclareUnicodeCharacter{0153}{\oe}
        !          9406:   \DeclareUnicodeCharacter{0154}{\'R}
        !          9407:   \DeclareUnicodeCharacter{0155}{\'r}
        !          9408:   \DeclareUnicodeCharacter{0158}{\v{R}}
        !          9409:   \DeclareUnicodeCharacter{0159}{\v{r}}
        !          9410:   \DeclareUnicodeCharacter{015A}{\'S}
        !          9411:   \DeclareUnicodeCharacter{015B}{\'s}
        !          9412:   \DeclareUnicodeCharacter{015C}{\^S}
        !          9413:   \DeclareUnicodeCharacter{015D}{\^s}
        !          9414:   \DeclareUnicodeCharacter{015E}{\cedilla{S}}
        !          9415:   \DeclareUnicodeCharacter{015F}{\cedilla{s}}
        !          9416: 
        !          9417:   \DeclareUnicodeCharacter{0160}{\v{S}}
        !          9418:   \DeclareUnicodeCharacter{0161}{\v{s}}
        !          9419:   \DeclareUnicodeCharacter{0162}{\cedilla{t}}
        !          9420:   \DeclareUnicodeCharacter{0163}{\cedilla{T}}
        !          9421:   \DeclareUnicodeCharacter{0164}{\v{T}}
        !          9422: 
        !          9423:   \DeclareUnicodeCharacter{0168}{\~U}
        !          9424:   \DeclareUnicodeCharacter{0169}{\~u}
        !          9425:   \DeclareUnicodeCharacter{016A}{\=U}
        !          9426:   \DeclareUnicodeCharacter{016B}{\=u}
        !          9427:   \DeclareUnicodeCharacter{016C}{\u{U}}
        !          9428:   \DeclareUnicodeCharacter{016D}{\u{u}}
        !          9429:   \DeclareUnicodeCharacter{016E}{\ringaccent{U}}
        !          9430:   \DeclareUnicodeCharacter{016F}{\ringaccent{u}}
        !          9431: 
        !          9432:   \DeclareUnicodeCharacter{0170}{\H{U}}
        !          9433:   \DeclareUnicodeCharacter{0171}{\H{u}}
        !          9434:   \DeclareUnicodeCharacter{0174}{\^W}
        !          9435:   \DeclareUnicodeCharacter{0175}{\^w}
        !          9436:   \DeclareUnicodeCharacter{0176}{\^Y}
        !          9437:   \DeclareUnicodeCharacter{0177}{\^y}
        !          9438:   \DeclareUnicodeCharacter{0178}{\"Y}
        !          9439:   \DeclareUnicodeCharacter{0179}{\'Z}
        !          9440:   \DeclareUnicodeCharacter{017A}{\'z}
        !          9441:   \DeclareUnicodeCharacter{017B}{\dotaccent{Z}}
        !          9442:   \DeclareUnicodeCharacter{017C}{\dotaccent{z}}
        !          9443:   \DeclareUnicodeCharacter{017D}{\v{Z}}
        !          9444:   \DeclareUnicodeCharacter{017E}{\v{z}}
        !          9445: 
        !          9446:   \DeclareUnicodeCharacter{01C4}{D\v{Z}}
        !          9447:   \DeclareUnicodeCharacter{01C5}{D\v{z}}
        !          9448:   \DeclareUnicodeCharacter{01C6}{d\v{z}}
        !          9449:   \DeclareUnicodeCharacter{01C7}{LJ}
        !          9450:   \DeclareUnicodeCharacter{01C8}{Lj}
        !          9451:   \DeclareUnicodeCharacter{01C9}{lj}
        !          9452:   \DeclareUnicodeCharacter{01CA}{NJ}
        !          9453:   \DeclareUnicodeCharacter{01CB}{Nj}
        !          9454:   \DeclareUnicodeCharacter{01CC}{nj}
        !          9455:   \DeclareUnicodeCharacter{01CD}{\v{A}}
        !          9456:   \DeclareUnicodeCharacter{01CE}{\v{a}}
        !          9457:   \DeclareUnicodeCharacter{01CF}{\v{I}}
        !          9458: 
        !          9459:   \DeclareUnicodeCharacter{01D0}{\v{\dotless{i}}}
        !          9460:   \DeclareUnicodeCharacter{01D1}{\v{O}}
        !          9461:   \DeclareUnicodeCharacter{01D2}{\v{o}}
        !          9462:   \DeclareUnicodeCharacter{01D3}{\v{U}}
        !          9463:   \DeclareUnicodeCharacter{01D4}{\v{u}}
        !          9464: 
        !          9465:   \DeclareUnicodeCharacter{01E2}{\={\AE}}
        !          9466:   \DeclareUnicodeCharacter{01E3}{\={\ae}}
        !          9467:   \DeclareUnicodeCharacter{01E6}{\v{G}}
        !          9468:   \DeclareUnicodeCharacter{01E7}{\v{g}}
        !          9469:   \DeclareUnicodeCharacter{01E8}{\v{K}}
        !          9470:   \DeclareUnicodeCharacter{01E9}{\v{k}}
        !          9471: 
        !          9472:   \DeclareUnicodeCharacter{01F0}{\v{\dotless{j}}}
        !          9473:   \DeclareUnicodeCharacter{01F1}{DZ}
        !          9474:   \DeclareUnicodeCharacter{01F2}{Dz}
        !          9475:   \DeclareUnicodeCharacter{01F3}{dz}
        !          9476:   \DeclareUnicodeCharacter{01F4}{\'G}
        !          9477:   \DeclareUnicodeCharacter{01F5}{\'g}
        !          9478:   \DeclareUnicodeCharacter{01F8}{\`N}
        !          9479:   \DeclareUnicodeCharacter{01F9}{\`n}
        !          9480:   \DeclareUnicodeCharacter{01FC}{\'{\AE}}
        !          9481:   \DeclareUnicodeCharacter{01FD}{\'{\ae}}
        !          9482:   \DeclareUnicodeCharacter{01FE}{\'{\O}}
        !          9483:   \DeclareUnicodeCharacter{01FF}{\'{\o}}
        !          9484: 
        !          9485:   \DeclareUnicodeCharacter{021E}{\v{H}}
        !          9486:   \DeclareUnicodeCharacter{021F}{\v{h}}
        !          9487: 
        !          9488:   \DeclareUnicodeCharacter{0226}{\dotaccent{A}}
        !          9489:   \DeclareUnicodeCharacter{0227}{\dotaccent{a}}
        !          9490:   \DeclareUnicodeCharacter{0228}{\cedilla{E}}
        !          9491:   \DeclareUnicodeCharacter{0229}{\cedilla{e}}
        !          9492:   \DeclareUnicodeCharacter{022E}{\dotaccent{O}}
        !          9493:   \DeclareUnicodeCharacter{022F}{\dotaccent{o}}
        !          9494: 
        !          9495:   \DeclareUnicodeCharacter{0232}{\=Y}
        !          9496:   \DeclareUnicodeCharacter{0233}{\=y}
        !          9497:   \DeclareUnicodeCharacter{0237}{\dotless{j}}
        !          9498: 
        !          9499:   \DeclareUnicodeCharacter{02DB}{\ogonek{ }}
        !          9500: 
        !          9501:   \DeclareUnicodeCharacter{1E02}{\dotaccent{B}}
        !          9502:   \DeclareUnicodeCharacter{1E03}{\dotaccent{b}}
        !          9503:   \DeclareUnicodeCharacter{1E04}{\udotaccent{B}}
        !          9504:   \DeclareUnicodeCharacter{1E05}{\udotaccent{b}}
        !          9505:   \DeclareUnicodeCharacter{1E06}{\ubaraccent{B}}
        !          9506:   \DeclareUnicodeCharacter{1E07}{\ubaraccent{b}}
        !          9507:   \DeclareUnicodeCharacter{1E0A}{\dotaccent{D}}
        !          9508:   \DeclareUnicodeCharacter{1E0B}{\dotaccent{d}}
        !          9509:   \DeclareUnicodeCharacter{1E0C}{\udotaccent{D}}
        !          9510:   \DeclareUnicodeCharacter{1E0D}{\udotaccent{d}}
        !          9511:   \DeclareUnicodeCharacter{1E0E}{\ubaraccent{D}}
        !          9512:   \DeclareUnicodeCharacter{1E0F}{\ubaraccent{d}}
        !          9513: 
        !          9514:   \DeclareUnicodeCharacter{1E1E}{\dotaccent{F}}
        !          9515:   \DeclareUnicodeCharacter{1E1F}{\dotaccent{f}}
        !          9516: 
        !          9517:   \DeclareUnicodeCharacter{1E20}{\=G}
        !          9518:   \DeclareUnicodeCharacter{1E21}{\=g}
        !          9519:   \DeclareUnicodeCharacter{1E22}{\dotaccent{H}}
        !          9520:   \DeclareUnicodeCharacter{1E23}{\dotaccent{h}}
        !          9521:   \DeclareUnicodeCharacter{1E24}{\udotaccent{H}}
        !          9522:   \DeclareUnicodeCharacter{1E25}{\udotaccent{h}}
        !          9523:   \DeclareUnicodeCharacter{1E26}{\"H}
        !          9524:   \DeclareUnicodeCharacter{1E27}{\"h}
        !          9525: 
        !          9526:   \DeclareUnicodeCharacter{1E30}{\'K}
        !          9527:   \DeclareUnicodeCharacter{1E31}{\'k}
        !          9528:   \DeclareUnicodeCharacter{1E32}{\udotaccent{K}}
        !          9529:   \DeclareUnicodeCharacter{1E33}{\udotaccent{k}}
        !          9530:   \DeclareUnicodeCharacter{1E34}{\ubaraccent{K}}
        !          9531:   \DeclareUnicodeCharacter{1E35}{\ubaraccent{k}}
        !          9532:   \DeclareUnicodeCharacter{1E36}{\udotaccent{L}}
        !          9533:   \DeclareUnicodeCharacter{1E37}{\udotaccent{l}}
        !          9534:   \DeclareUnicodeCharacter{1E3A}{\ubaraccent{L}}
        !          9535:   \DeclareUnicodeCharacter{1E3B}{\ubaraccent{l}}
        !          9536:   \DeclareUnicodeCharacter{1E3E}{\'M}
        !          9537:   \DeclareUnicodeCharacter{1E3F}{\'m}
        !          9538: 
        !          9539:   \DeclareUnicodeCharacter{1E40}{\dotaccent{M}}
        !          9540:   \DeclareUnicodeCharacter{1E41}{\dotaccent{m}}
        !          9541:   \DeclareUnicodeCharacter{1E42}{\udotaccent{M}}
        !          9542:   \DeclareUnicodeCharacter{1E43}{\udotaccent{m}}
        !          9543:   \DeclareUnicodeCharacter{1E44}{\dotaccent{N}}
        !          9544:   \DeclareUnicodeCharacter{1E45}{\dotaccent{n}}
        !          9545:   \DeclareUnicodeCharacter{1E46}{\udotaccent{N}}
        !          9546:   \DeclareUnicodeCharacter{1E47}{\udotaccent{n}}
        !          9547:   \DeclareUnicodeCharacter{1E48}{\ubaraccent{N}}
        !          9548:   \DeclareUnicodeCharacter{1E49}{\ubaraccent{n}}
        !          9549: 
        !          9550:   \DeclareUnicodeCharacter{1E54}{\'P}
        !          9551:   \DeclareUnicodeCharacter{1E55}{\'p}
        !          9552:   \DeclareUnicodeCharacter{1E56}{\dotaccent{P}}
        !          9553:   \DeclareUnicodeCharacter{1E57}{\dotaccent{p}}
        !          9554:   \DeclareUnicodeCharacter{1E58}{\dotaccent{R}}
        !          9555:   \DeclareUnicodeCharacter{1E59}{\dotaccent{r}}
        !          9556:   \DeclareUnicodeCharacter{1E5A}{\udotaccent{R}}
        !          9557:   \DeclareUnicodeCharacter{1E5B}{\udotaccent{r}}
        !          9558:   \DeclareUnicodeCharacter{1E5E}{\ubaraccent{R}}
        !          9559:   \DeclareUnicodeCharacter{1E5F}{\ubaraccent{r}}
        !          9560: 
        !          9561:   \DeclareUnicodeCharacter{1E60}{\dotaccent{S}}
        !          9562:   \DeclareUnicodeCharacter{1E61}{\dotaccent{s}}
        !          9563:   \DeclareUnicodeCharacter{1E62}{\udotaccent{S}}
        !          9564:   \DeclareUnicodeCharacter{1E63}{\udotaccent{s}}
        !          9565:   \DeclareUnicodeCharacter{1E6A}{\dotaccent{T}}
        !          9566:   \DeclareUnicodeCharacter{1E6B}{\dotaccent{t}}
        !          9567:   \DeclareUnicodeCharacter{1E6C}{\udotaccent{T}}
        !          9568:   \DeclareUnicodeCharacter{1E6D}{\udotaccent{t}}
        !          9569:   \DeclareUnicodeCharacter{1E6E}{\ubaraccent{T}}
        !          9570:   \DeclareUnicodeCharacter{1E6F}{\ubaraccent{t}}
        !          9571: 
        !          9572:   \DeclareUnicodeCharacter{1E7C}{\~V}
        !          9573:   \DeclareUnicodeCharacter{1E7D}{\~v}
        !          9574:   \DeclareUnicodeCharacter{1E7E}{\udotaccent{V}}
        !          9575:   \DeclareUnicodeCharacter{1E7F}{\udotaccent{v}}
        !          9576: 
        !          9577:   \DeclareUnicodeCharacter{1E80}{\`W}
        !          9578:   \DeclareUnicodeCharacter{1E81}{\`w}
        !          9579:   \DeclareUnicodeCharacter{1E82}{\'W}
        !          9580:   \DeclareUnicodeCharacter{1E83}{\'w}
        !          9581:   \DeclareUnicodeCharacter{1E84}{\"W}
        !          9582:   \DeclareUnicodeCharacter{1E85}{\"w}
        !          9583:   \DeclareUnicodeCharacter{1E86}{\dotaccent{W}}
        !          9584:   \DeclareUnicodeCharacter{1E87}{\dotaccent{w}}
        !          9585:   \DeclareUnicodeCharacter{1E88}{\udotaccent{W}}
        !          9586:   \DeclareUnicodeCharacter{1E89}{\udotaccent{w}}
        !          9587:   \DeclareUnicodeCharacter{1E8A}{\dotaccent{X}}
        !          9588:   \DeclareUnicodeCharacter{1E8B}{\dotaccent{x}}
        !          9589:   \DeclareUnicodeCharacter{1E8C}{\"X}
        !          9590:   \DeclareUnicodeCharacter{1E8D}{\"x}
        !          9591:   \DeclareUnicodeCharacter{1E8E}{\dotaccent{Y}}
        !          9592:   \DeclareUnicodeCharacter{1E8F}{\dotaccent{y}}
        !          9593: 
        !          9594:   \DeclareUnicodeCharacter{1E90}{\^Z}
        !          9595:   \DeclareUnicodeCharacter{1E91}{\^z}
        !          9596:   \DeclareUnicodeCharacter{1E92}{\udotaccent{Z}}
        !          9597:   \DeclareUnicodeCharacter{1E93}{\udotaccent{z}}
        !          9598:   \DeclareUnicodeCharacter{1E94}{\ubaraccent{Z}}
        !          9599:   \DeclareUnicodeCharacter{1E95}{\ubaraccent{z}}
        !          9600:   \DeclareUnicodeCharacter{1E96}{\ubaraccent{h}}
        !          9601:   \DeclareUnicodeCharacter{1E97}{\"t}
        !          9602:   \DeclareUnicodeCharacter{1E98}{\ringaccent{w}}
        !          9603:   \DeclareUnicodeCharacter{1E99}{\ringaccent{y}}
        !          9604: 
        !          9605:   \DeclareUnicodeCharacter{1EA0}{\udotaccent{A}}
        !          9606:   \DeclareUnicodeCharacter{1EA1}{\udotaccent{a}}
        !          9607: 
        !          9608:   \DeclareUnicodeCharacter{1EB8}{\udotaccent{E}}
        !          9609:   \DeclareUnicodeCharacter{1EB9}{\udotaccent{e}}
        !          9610:   \DeclareUnicodeCharacter{1EBC}{\~E}
        !          9611:   \DeclareUnicodeCharacter{1EBD}{\~e}
        !          9612: 
        !          9613:   \DeclareUnicodeCharacter{1ECA}{\udotaccent{I}}
        !          9614:   \DeclareUnicodeCharacter{1ECB}{\udotaccent{i}}
        !          9615:   \DeclareUnicodeCharacter{1ECC}{\udotaccent{O}}
        !          9616:   \DeclareUnicodeCharacter{1ECD}{\udotaccent{o}}
        !          9617: 
        !          9618:   \DeclareUnicodeCharacter{1EE4}{\udotaccent{U}}
        !          9619:   \DeclareUnicodeCharacter{1EE5}{\udotaccent{u}}
        !          9620: 
        !          9621:   \DeclareUnicodeCharacter{1EF2}{\`Y}
        !          9622:   \DeclareUnicodeCharacter{1EF3}{\`y}
        !          9623:   \DeclareUnicodeCharacter{1EF4}{\udotaccent{Y}}
        !          9624: 
        !          9625:   \DeclareUnicodeCharacter{1EF8}{\~Y}
        !          9626:   \DeclareUnicodeCharacter{1EF9}{\~y}
        !          9627: 
        !          9628:   \DeclareUnicodeCharacter{2013}{--}
        !          9629:   \DeclareUnicodeCharacter{2014}{---}
        !          9630:   \DeclareUnicodeCharacter{2018}{\quoteleft}
        !          9631:   \DeclareUnicodeCharacter{2019}{\quoteright}
        !          9632:   \DeclareUnicodeCharacter{201A}{\quotesinglbase}
        !          9633:   \DeclareUnicodeCharacter{201C}{\quotedblleft}
        !          9634:   \DeclareUnicodeCharacter{201D}{\quotedblright}
        !          9635:   \DeclareUnicodeCharacter{201E}{\quotedblbase}
        !          9636:   \DeclareUnicodeCharacter{2022}{\bullet}
        !          9637:   \DeclareUnicodeCharacter{2026}{\dots}
        !          9638:   \DeclareUnicodeCharacter{2039}{\guilsinglleft}
        !          9639:   \DeclareUnicodeCharacter{203A}{\guilsinglright}
        !          9640:   \DeclareUnicodeCharacter{20AC}{\euro}
        !          9641: 
        !          9642:   \DeclareUnicodeCharacter{2192}{\expansion}
        !          9643:   \DeclareUnicodeCharacter{21D2}{\result}
        !          9644: 
        !          9645:   \DeclareUnicodeCharacter{2212}{\minus}
        !          9646:   \DeclareUnicodeCharacter{2217}{\point}
        !          9647:   \DeclareUnicodeCharacter{2261}{\equiv}
        !          9648: }% end of \utfeightchardefs
        !          9649: 
        !          9650: 
        !          9651: % US-ASCII character definitions.
        !          9652: \def\asciichardefs{% nothing need be done
        !          9653:    \relax
        !          9654: }
        !          9655: 
        !          9656: % Make non-ASCII characters printable again for compatibility with
        !          9657: % existing Texinfo documents that may use them, even without declaring a
        !          9658: % document encoding.
        !          9659: %
        !          9660: \setnonasciicharscatcode \other
        !          9661: 
        !          9662: 
        !          9663: \message{formatting,}
        !          9664: 
        !          9665: \newdimen\defaultparindent \defaultparindent = 15pt
        !          9666: 
        !          9667: \chapheadingskip = 15pt plus 4pt minus 2pt
        !          9668: \secheadingskip = 12pt plus 3pt minus 2pt
        !          9669: \subsecheadingskip = 9pt plus 2pt minus 2pt
        !          9670: 
        !          9671: % Prevent underfull vbox error messages.
        !          9672: \vbadness = 10000
        !          9673: 
        !          9674: % Don't be very finicky about underfull hboxes, either.
        !          9675: \hbadness = 6666
        !          9676: 
        !          9677: % Following George Bush, get rid of widows and orphans.
        !          9678: \widowpenalty=10000
        !          9679: \clubpenalty=10000
        !          9680: 
        !          9681: % Use TeX 3.0's \emergencystretch to help line breaking, but if we're
        !          9682: % using an old version of TeX, don't do anything.  We want the amount of
        !          9683: % stretch added to depend on the line length, hence the dependence on
        !          9684: % \hsize.  We call this whenever the paper size is set.
        !          9685: %
        !          9686: \def\setemergencystretch{%
        !          9687:   \ifx\emergencystretch\thisisundefined
        !          9688:     % Allow us to assign to \emergencystretch anyway.
        !          9689:     \def\emergencystretch{\dimen0}%
        !          9690:   \else
        !          9691:     \emergencystretch = .15\hsize
        !          9692:   \fi
        !          9693: }
        !          9694: 
        !          9695: % Parameters in order: 1) textheight; 2) textwidth;
        !          9696: % 3) voffset; 4) hoffset; 5) binding offset; 6) topskip;
        !          9697: % 7) physical page height; 8) physical page width.
        !          9698: %
        !          9699: % We also call \setleading{\textleading}, so the caller should define
        !          9700: % \textleading.  The caller should also set \parskip.
        !          9701: %
        !          9702: \def\internalpagesizes#1#2#3#4#5#6#7#8{%
        !          9703:   \voffset = #3\relax
        !          9704:   \topskip = #6\relax
        !          9705:   \splittopskip = \topskip
        !          9706:   %
        !          9707:   \vsize = #1\relax
        !          9708:   \advance\vsize by \topskip
        !          9709:   \outervsize = \vsize
        !          9710:   \advance\outervsize by 2\topandbottommargin
        !          9711:   \pageheight = \vsize
        !          9712:   %
        !          9713:   \hsize = #2\relax
        !          9714:   \outerhsize = \hsize
        !          9715:   \advance\outerhsize by 0.5in
        !          9716:   \pagewidth = \hsize
        !          9717:   %
        !          9718:   \normaloffset = #4\relax
        !          9719:   \bindingoffset = #5\relax
        !          9720:   %
        !          9721:   \ifpdf
        !          9722:     \pdfpageheight #7\relax
        !          9723:     \pdfpagewidth #8\relax
        !          9724:     % if we don't reset these, they will remain at "1 true in" of
        !          9725:     % whatever layout pdftex was dumped with.
        !          9726:     \pdfhorigin = 1 true in
        !          9727:     \pdfvorigin = 1 true in
        !          9728:   \fi
        !          9729:   %
        !          9730:   \setleading{\textleading}
        !          9731:   %
        !          9732:   \parindent = \defaultparindent
        !          9733:   \setemergencystretch
        !          9734: }
        !          9735: 
        !          9736: % @letterpaper (the default).
        !          9737: \def\letterpaper{{\globaldefs = 1
        !          9738:   \parskip = 3pt plus 2pt minus 1pt
        !          9739:   \textleading = 13.2pt
        !          9740:   %
        !          9741:   % If page is nothing but text, make it come out even.
        !          9742:   \internalpagesizes{607.2pt}{6in}% that's 46 lines
        !          9743:                     {\voffset}{.25in}%
        !          9744:                     {\bindingoffset}{36pt}%
        !          9745:                     {11in}{8.5in}%
        !          9746: }}
        !          9747: 
        !          9748: % Use @smallbook to reset parameters for 7x9.25 trim size.
        !          9749: \def\smallbook{{\globaldefs = 1
        !          9750:   \parskip = 2pt plus 1pt
        !          9751:   \textleading = 12pt
        !          9752:   %
        !          9753:   \internalpagesizes{7.5in}{5in}%
        !          9754:                     {-.2in}{0in}%
        !          9755:                     {\bindingoffset}{16pt}%
        !          9756:                     {9.25in}{7in}%
        !          9757:   %
        !          9758:   \lispnarrowing = 0.3in
        !          9759:   \tolerance = 700
        !          9760:   \hfuzz = 1pt
        !          9761:   \contentsrightmargin = 0pt
        !          9762:   \defbodyindent = .5cm
        !          9763: }}
        !          9764: 
        !          9765: % Use @smallerbook to reset parameters for 6x9 trim size.
        !          9766: % (Just testing, parameters still in flux.)
        !          9767: \def\smallerbook{{\globaldefs = 1
        !          9768:   \parskip = 1.5pt plus 1pt
        !          9769:   \textleading = 12pt
        !          9770:   %
        !          9771:   \internalpagesizes{7.4in}{4.8in}%
        !          9772:                     {-.2in}{-.4in}%
        !          9773:                     {0pt}{14pt}%
        !          9774:                     {9in}{6in}%
        !          9775:   %
        !          9776:   \lispnarrowing = 0.25in
        !          9777:   \tolerance = 700
        !          9778:   \hfuzz = 1pt
        !          9779:   \contentsrightmargin = 0pt
        !          9780:   \defbodyindent = .4cm
        !          9781: }}
        !          9782: 
        !          9783: % Use @afourpaper to print on European A4 paper.
        !          9784: \def\afourpaper{{\globaldefs = 1
        !          9785:   \parskip = 3pt plus 2pt minus 1pt
        !          9786:   \textleading = 13.2pt
        !          9787:   %
        !          9788:   % Double-side printing via postscript on Laserjet 4050
        !          9789:   % prints double-sided nicely when \bindingoffset=10mm and \hoffset=-6mm.
        !          9790:   % To change the settings for a different printer or situation, adjust
        !          9791:   % \normaloffset until the front-side and back-side texts align.  Then
        !          9792:   % do the same for \bindingoffset.  You can set these for testing in
        !          9793:   % your texinfo source file like this:
        !          9794:   % @tex
        !          9795:   % \global\normaloffset = -6mm
        !          9796:   % \global\bindingoffset = 10mm
        !          9797:   % @end tex
        !          9798:   \internalpagesizes{673.2pt}{160mm}% that's 51 lines
        !          9799:                     {\voffset}{\hoffset}%
        !          9800:                     {\bindingoffset}{44pt}%
        !          9801:                     {297mm}{210mm}%
        !          9802:   %
        !          9803:   \tolerance = 700
        !          9804:   \hfuzz = 1pt
        !          9805:   \contentsrightmargin = 0pt
        !          9806:   \defbodyindent = 5mm
        !          9807: }}
        !          9808: 
        !          9809: % Use @afivepaper to print on European A5 paper.
        !          9810: % From [email protected], 2 July 2000.
        !          9811: % He also recommends making @example and @lisp be small.
        !          9812: \def\afivepaper{{\globaldefs = 1
        !          9813:   \parskip = 2pt plus 1pt minus 0.1pt
        !          9814:   \textleading = 12.5pt
        !          9815:   %
        !          9816:   \internalpagesizes{160mm}{120mm}%
        !          9817:                     {\voffset}{\hoffset}%
        !          9818:                     {\bindingoffset}{8pt}%
        !          9819:                     {210mm}{148mm}%
        !          9820:   %
        !          9821:   \lispnarrowing = 0.2in
        !          9822:   \tolerance = 800
        !          9823:   \hfuzz = 1.2pt
        !          9824:   \contentsrightmargin = 0pt
        !          9825:   \defbodyindent = 2mm
        !          9826:   \tableindent = 12mm
        !          9827: }}
        !          9828: 
        !          9829: % A specific text layout, 24x15cm overall, intended for A4 paper.
        !          9830: \def\afourlatex{{\globaldefs = 1
        !          9831:   \afourpaper
        !          9832:   \internalpagesizes{237mm}{150mm}%
        !          9833:                     {\voffset}{4.6mm}%
        !          9834:                     {\bindingoffset}{7mm}%
        !          9835:                     {297mm}{210mm}%
        !          9836:   %
        !          9837:   % Must explicitly reset to 0 because we call \afourpaper.
        !          9838:   \globaldefs = 0
        !          9839: }}
        !          9840: 
        !          9841: % Use @afourwide to print on A4 paper in landscape format.
        !          9842: \def\afourwide{{\globaldefs = 1
        !          9843:   \afourpaper
        !          9844:   \internalpagesizes{241mm}{165mm}%
        !          9845:                     {\voffset}{-2.95mm}%
        !          9846:                     {\bindingoffset}{7mm}%
        !          9847:                     {297mm}{210mm}%
        !          9848:   \globaldefs = 0
        !          9849: }}
        !          9850: 
        !          9851: % @pagesizes TEXTHEIGHT[,TEXTWIDTH]
        !          9852: % Perhaps we should allow setting the margins, \topskip, \parskip,
        !          9853: % and/or leading, also. Or perhaps we should compute them somehow.
        !          9854: %
        !          9855: \parseargdef\pagesizes{\pagesizesyyy #1,,\finish}
        !          9856: \def\pagesizesyyy#1,#2,#3\finish{{%
        !          9857:   \setbox0 = \hbox{\ignorespaces #2}\ifdim\wd0 > 0pt \hsize=#2\relax \fi
        !          9858:   \globaldefs = 1
        !          9859:   %
        !          9860:   \parskip = 3pt plus 2pt minus 1pt
        !          9861:   \setleading{\textleading}%
        !          9862:   %
        !          9863:   \dimen0 = #1\relax
        !          9864:   \advance\dimen0 by \voffset
        !          9865:   %
        !          9866:   \dimen2 = \hsize
        !          9867:   \advance\dimen2 by \normaloffset
        !          9868:   %
        !          9869:   \internalpagesizes{#1}{\hsize}%
        !          9870:                     {\voffset}{\normaloffset}%
        !          9871:                     {\bindingoffset}{44pt}%
        !          9872:                     {\dimen0}{\dimen2}%
        !          9873: }}
        !          9874: 
        !          9875: % Set default to letter.
        !          9876: %
        !          9877: \letterpaper
        !          9878: 
        !          9879: 
        !          9880: \message{and turning on texinfo input format.}
        !          9881: 
        !          9882: \def^^L{\par} % remove \outer, so ^L can appear in an @comment
        !          9883: 
        !          9884: % DEL is a comment character, in case @c does not suffice.
        !          9885: \catcode`\^^? = 14
        !          9886: 
        !          9887: % Define macros to output various characters with catcode for normal text.
        !          9888: \catcode`\"=\other \def\normaldoublequote{"}
        !          9889: \catcode`\$=\other \def\normaldollar{$}%$ font-lock fix
        !          9890: \catcode`\+=\other \def\normalplus{+}
        !          9891: \catcode`\<=\other \def\normalless{<}
        !          9892: \catcode`\>=\other \def\normalgreater{>}
        !          9893: \catcode`\^=\other \def\normalcaret{^}
        !          9894: \catcode`\_=\other \def\normalunderscore{_}
        !          9895: \catcode`\|=\other \def\normalverticalbar{|}
        !          9896: \catcode`\~=\other \def\normaltilde{~}
        !          9897: 
        !          9898: % This macro is used to make a character print one way in \tt
        !          9899: % (where it can probably be output as-is), and another way in other fonts,
        !          9900: % where something hairier probably needs to be done.
        !          9901: %
        !          9902: % #1 is what to print if we are indeed using \tt; #2 is what to print
        !          9903: % otherwise.  Since all the Computer Modern typewriter fonts have zero
        !          9904: % interword stretch (and shrink), and it is reasonable to expect all
        !          9905: % typewriter fonts to have this, we can check that font parameter.
        !          9906: %
        !          9907: \def\ifusingtt#1#2{\ifdim \fontdimen3\font=0pt #1\else #2\fi}
        !          9908: 
        !          9909: % Same as above, but check for italic font.  Actually this also catches
        !          9910: % non-italic slanted fonts since it is impossible to distinguish them from
        !          9911: % italic fonts.  But since this is only used by $ and it uses \sl anyway
        !          9912: % this is not a problem.
        !          9913: \def\ifusingit#1#2{\ifdim \fontdimen1\font>0pt #1\else #2\fi}
        !          9914: 
        !          9915: % Turn off all special characters except @
        !          9916: % (and those which the user can use as if they were ordinary).
        !          9917: % Most of these we simply print from the \tt font, but for some, we can
        !          9918: % use math or other variants that look better in normal text.
        !          9919: 
        !          9920: \catcode`\"=\active
        !          9921: \def\activedoublequote{{\tt\char34}}
        !          9922: \let"=\activedoublequote
        !          9923: \catcode`\~=\active
        !          9924: \def~{{\tt\char126}}
        !          9925: \chardef\hat=`\^
        !          9926: \catcode`\^=\active
        !          9927: \def^{{\tt \hat}}
        !          9928: 
        !          9929: \catcode`\_=\active
        !          9930: \def_{\ifusingtt\normalunderscore\_}
        !          9931: \let\realunder=_
        !          9932: % Subroutine for the previous macro.
        !          9933: \def\_{\leavevmode \kern.07em \vbox{\hrule width.3em height.1ex}\kern .07em }
        !          9934: 
        !          9935: \catcode`\|=\active
        !          9936: \def|{{\tt\char124}}
        !          9937: \chardef \less=`\<
        !          9938: \catcode`\<=\active
        !          9939: \def<{{\tt \less}}
        !          9940: \chardef \gtr=`\>
        !          9941: \catcode`\>=\active
        !          9942: \def>{{\tt \gtr}}
        !          9943: \catcode`\+=\active
        !          9944: \def+{{\tt \char 43}}
        !          9945: \catcode`\$=\active
        !          9946: \def${\ifusingit{{\sl\$}}\normaldollar}%$ font-lock fix
        !          9947: 
        !          9948: % If a .fmt file is being used, characters that might appear in a file
        !          9949: % name cannot be active until we have parsed the command line.
        !          9950: % So turn them off again, and have \everyjob (or @setfilename) turn them on.
        !          9951: % \otherifyactive is called near the end of this file.
        !          9952: \def\otherifyactive{\catcode`+=\other \catcode`\_=\other}
        !          9953: 
        !          9954: % Used sometimes to turn off (effectively) the active characters even after
        !          9955: % parsing them.
        !          9956: \def\turnoffactive{%
        !          9957:   \normalturnoffactive
        !          9958:   \otherbackslash
        !          9959: }
        !          9960: 
        !          9961: \catcode`\@=0
        !          9962: 
        !          9963: % \backslashcurfont outputs one backslash character in current font,
        !          9964: % as in \char`\\.
        !          9965: \global\chardef\backslashcurfont=`\\
        !          9966: \global\let\rawbackslashxx=\backslashcurfont  % let existing .??s files work
        !          9967: 
        !          9968: % \realbackslash is an actual character `\' with catcode other, and
        !          9969: % \doublebackslash is two of them (for the pdf outlines).
        !          9970: {\catcode`\\=\other @gdef@realbackslash{\} @gdef@doublebackslash{\\}}
        !          9971: 
        !          9972: % In texinfo, backslash is an active character; it prints the backslash
        !          9973: % in fixed width font.
        !          9974: \catcode`\\=\active  % @ for escape char from now on.
        !          9975: 
        !          9976: % The story here is that in math mode, the \char of \backslashcurfont
        !          9977: % ends up printing the roman \ from the math symbol font (because \char
        !          9978: % in math mode uses the \mathcode, and plain.tex sets
        !          9979: % \mathcode`\\="026E).  It seems better for @backslashchar{} to always
        !          9980: % print a typewriter backslash, hence we use an explicit \mathchar,
        !          9981: % which is the decimal equivalent of "715c (class 7, e.g., use \fam;
        !          9982: % ignored family value; char position "5C).  We can't use " for the
        !          9983: % usual hex value because it has already been made active.
        !          9984: @def@normalbackslash{{@tt @ifmmode @mathchar29020 @else @backslashcurfont @fi}}
        !          9985: @let@backslashchar = @normalbackslash % @backslashchar{} is for user documents.
        !          9986: 
        !          9987: % On startup, @fixbackslash assigns:
        !          9988: %  @let \ = @normalbackslash
        !          9989: % \rawbackslash defines an active \ to do \backslashcurfont.
        !          9990: % \otherbackslash defines an active \ to be a literal `\' character with
        !          9991: % catcode other.  We switch back and forth between these.
        !          9992: @gdef@rawbackslash{@let\=@backslashcurfont}
        !          9993: @gdef@otherbackslash{@let\=@realbackslash}
        !          9994: 
        !          9995: % Same as @turnoffactive except outputs \ as {\tt\char`\\} instead of
        !          9996: % the literal character `\'.  Also revert - to its normal character, in
        !          9997: % case the active - from code has slipped in.
        !          9998: %
        !          9999: {@catcode`- = @active
        !          10000:  @gdef@normalturnoffactive{%
        !          10001:    @let-=@normaldash
        !          10002:    @let"=@normaldoublequote
        !          10003:    @let$=@normaldollar %$ font-lock fix
        !          10004:    @let+=@normalplus
        !          10005:    @let<=@normalless
        !          10006:    @let>=@normalgreater
        !          10007:    @let\=@normalbackslash
        !          10008:    @let^=@normalcaret
        !          10009:    @let_=@normalunderscore
        !          10010:    @let|=@normalverticalbar
        !          10011:    @let~=@normaltilde
        !          10012:    @markupsetuplqdefault
        !          10013:    @markupsetuprqdefault
        !          10014:    @unsepspaces
        !          10015:  }
        !          10016: }
        !          10017: 
        !          10018: % Make _ and + \other characters, temporarily.
        !          10019: % This is canceled by @fixbackslash.
        !          10020: @otherifyactive
        !          10021: 
        !          10022: % If a .fmt file is being used, we don't want the `\input texinfo' to show up.
        !          10023: % That is what \eatinput is for; after that, the `\' should revert to printing
        !          10024: % a backslash.
        !          10025: %
        !          10026: @gdef@eatinput input texinfo{@fixbackslash}
        !          10027: @global@let\ = @eatinput
        !          10028: 
        !          10029: % On the other hand, perhaps the file did not have a `\input texinfo'. Then
        !          10030: % the first `\' in the file would cause an error. This macro tries to fix
        !          10031: % that, assuming it is called before the first `\' could plausibly occur.
        !          10032: % Also turn back on active characters that might appear in the input
        !          10033: % file name, in case not using a pre-dumped format.
        !          10034: %
        !          10035: @gdef@fixbackslash{%
        !          10036:   @ifx\@eatinput @let\ = @normalbackslash @fi
        !          10037:   @catcode`+=@active
        !          10038:   @catcode`@_=@active
        !          10039: }
        !          10040: 
        !          10041: % Say @foo, not \foo, in error messages.
        !          10042: @escapechar = `@@
        !          10043: 
        !          10044: % These (along with & and #) are made active for url-breaking, so need
        !          10045: % active definitions as the normal characters.
        !          10046: @def@normaldot{.}
        !          10047: @def@normalquest{?}
        !          10048: @def@normalslash{/}
        !          10049: 
        !          10050: % These look ok in all fonts, so just make them not special.
        !          10051: % @hashchar{} gets its own user-level command, because of #line.
        !          10052: @catcode`@& = @other @def@normalamp{&}
        !          10053: @catcode`@# = @other @def@normalhash{#}
        !          10054: @catcode`@% = @other @def@normalpercent{%}
        !          10055: 
        !          10056: @let @hashchar = @normalhash
        !          10057: 
        !          10058: @c Finally, make ` and ' active, so that txicodequoteundirected and
        !          10059: @c txicodequotebacktick work right in, e.g., @w{@code{`foo'}}.  If we
        !          10060: @c don't make ` and ' active, @code will not get them as active chars.
        !          10061: @c Do this last of all since we use ` in the previous @catcode assignments.
        !          10062: @catcode`@'=@active
        !          10063: @catcode`@`=@active
        !          10064: @markupsetuplqdefault
        !          10065: @markupsetuprqdefault
        !          10066: 
        !          10067: @c Local variables:
        !          10068: @c eval: (add-hook 'write-file-hooks 'time-stamp)
        !          10069: @c page-delimiter: "^\\\\message"
        !          10070: @c time-stamp-start: "def\\\\texinfoversion{"
        !          10071: @c time-stamp-format: "%:y-%02m-%02d.%02H"
        !          10072: @c time-stamp-end: "}"
        !          10073: @c End:
        !          10074: 
        !          10075: @c vim:sw=2:
        !          10076: 
        !          10077: @ignore
        !          10078:    arch-tag: e1b36e32-c96e-4135-a41a-0b2efa2ea115
        !          10079: @end ignore