summaryrefslogtreecommitdiff
path: root/Master/texmf-dist
diff options
context:
space:
mode:
authorKarl Berry <karl@freefriends.org>2006-01-11 23:54:06 +0000
committerKarl Berry <karl@freefriends.org>2006-01-11 23:54:06 +0000
commite89704e0b30a35f981c03c7d7bda4b7cea848c81 (patch)
tree247b30dafc3ac9afb6d1557f801d3c683b9829fe /Master/texmf-dist
parent7333dfd3878a1762d52a045af3a07c2541c37608 (diff)
trunk/Master/texmf-dist/source/latex/index
git-svn-id: svn://tug.org/texlive/trunk@330 c570f23f-e606-0410-a88d-b1316a301751
Diffstat (limited to 'Master/texmf-dist')
-rw-r--r--Master/texmf-dist/source/latex/index/index.dtx1398
-rw-r--r--Master/texmf-dist/source/latex/index/index.ins8
2 files changed, 1406 insertions, 0 deletions
diff --git a/Master/texmf-dist/source/latex/index/index.dtx b/Master/texmf-dist/source/latex/index/index.dtx
new file mode 100644
index 00000000000..4a7ecb77bb4
--- /dev/null
+++ b/Master/texmf-dist/source/latex/index/index.dtx
@@ -0,0 +1,1398 @@
+% \iffalse
+% ====================================================================
+% @LaTeX-style-file{
+% filename = "index.dtx",
+% version = "4.02beta",
+% date = "20 January 2004",
+% time = "21:15:11 EST",
+% author = "David M. Jones",
+% address = "MIT Laboratory for Computer Science
+% Room NE43-316
+% 545 Technology Square
+% Cambridge, MA 02139
+% USA",
+% telephone = "(617) 253-5936",
+% FAX = "(617) 253-3480",
+% checksum = "39539 1398 6018 50774",
+% email = "dmjones@theory.lcs.mit.edu",
+% codetable = "ISO/ASCII",
+% keywords = "LaTeX, index",
+% supported = "yes",
+% docstring = "This is a reimplementation of LaTeX's
+% indexing macros to provide better support
+% for indexing in LaTeX. For example, it
+% supports multiple indexes in a single
+% document and provides a more robust \index
+% command.
+%
+% The checksum field above contains a CRC-16
+% checksum as the first value, followed by
+% the equivalent of the standard UNIX wc
+% (word count) utility output of lines,
+% words, and characters. This is produced
+% by Robert Solovay's checksum utility.",
+%
+% }
+% ====================================================================
+%
+% This file may be distributed and/or modified under the
+% conditions of the LaTeX Project Public License, either version 1.2
+% of this license or (at your option) any later version.
+% The latest version of this license is in
+% http://www.latex-project.org/lppl.txt
+% and version 1.2 or later is part of all distributions of LaTeX
+% version 1999/12/01 or later.
+%
+% CAUTION: Use only as directed. Do not take internally. May cause
+% rash if applied directly to skin. Federal law prohibits distributing
+% without a proscription.
+%
+% \fi
+%
+%% \CheckSum{755}
+%% \CharacterTable
+%% {Upper-case \A\B\C\D\E\F\G\H\I\J\K\L\M\N\O\P\Q\R\S\T\U\V\W\X\Y\Z
+%% Lower-case \a\b\c\d\e\f\g\h\i\j\k\l\m\n\o\p\q\r\s\t\u\v\w\x\y\z
+%% Digits \0\1\2\3\4\5\6\7\8\9
+%% Exclamation \! Double quote \" Hash (number) \#
+%% Dollar \$ Percent \% Ampersand \&
+%% Acute accent \' Left paren \( Right paren \)
+%% Asterisk \* Plus \+ Comma \,
+%% Minus \- Point \. Solidus \/
+%% Colon \: Semicolon \; Less than \<
+%% Equals \= Greater than \> Question mark \?
+%% Commercial at \@ Left bracket \[ Backslash \\
+%% Right bracket \] Circumflex \^ Underscore \_
+%% Grave accent \` Left brace \{ Vertical bar \|
+%% Right brace \} Tilde \~}
+%
+% \iffalse
+% \begin{macrocode}
+%<*driver>
+\ProvidesFile{index.dtx}[1995/09/28 v4.1beta Improved index support (dmj)]
+%</driver>
+% \end{macrocode}
+%
+% \begin{macrocode}
+%<*driver>
+\documentclass{ltxdoc}
+\def\docdate {7 March 1994}
+% \end{macrocode}
+% We don't want everything to appear in the index
+% \begin{macrocode}
+\DoNotIndex{\!,\/,\?,\@,\^,\_}
+\DoNotIndex{\@@par,\@M,\@auxout,\@bsphack,\@esphack,\@depth,\@ehc}
+\DoNotIndex{\@for,\@flushglue,\@gobble,\@gobbletwo,\@height,\@idxitem}
+\DoNotIndex{\@ifnextchar,\@ifstar,\@ifundefined,\@input,\@latexerr}
+\DoNotIndex{\@makeschapterhead,\@namedef,\@nameuse,\@nil}
+\DoNotIndex{\@nobreakfalse,\@restonecolfalse,\@restonecoltrue}
+\DoNotIndex{\@tempa,\@tempf,\@temptokena,\@themark,\@width}
+\DoNotIndex{\active,\aindex,\baselineskip,\begin,\begingroup,\box}
+\DoNotIndex{\c@page,\catcode,\chapter,\char,\chardef,\closeout}
+\DoNotIndex{\CodelineIndex,\sp,\sb,\label,\leavevmode,\mark}
+\DoNotIndex{\mark,\newinsert,\newwrite,\newtoks,\xdef}
+\DoNotIndex{\columnsep,\columnseprule,\columnwidth,\csname,\def}
+\DoNotIndex{\dimen,\do,\DocInput,\documentstyle,\edef,\em}
+\DoNotIndex{\EnableCrossrefs,\end,\endcsname,\endgroup,\endinput}
+\DoNotIndex{\everypar,\expandafter,\filedate,\fileversion}
+\DoNotIndex{\footnotesize,\gdef,\global,\glossary,\hangindent}
+\DoNotIndex{\if@filesw,\else,\fi}
+\DoNotIndex{\if@nobreak,\if@twocolumn,\if@twoside,\fi,\fi,\fi}
+\DoNotIndex{\hsize,\hskip}
+\DoNotIndex{\ifhmode,\ifmmode,\ifodd,\ifvmode,\ifx,\fi,\fi,\fi,\fi,\fi}
+\DoNotIndex{\immediate,\insert,\item,\jobname,\long}
+\DoNotIndex{\let,\lineskip,\marginparsep,\marginparwidth,\maxdimen}
+\DoNotIndex{\makeatletter,\noexpand,\openout,\protect,\rlap}
+\DoNotIndex{\min,\newpage,\nobreak,\normalbaselineskip}
+\DoNotIndex{\normallineskip,\p@,\par,\parfillskip,\parindent,\parskip}
+\DoNotIndex{\penalty,\relax,\section,\sin,\sloppy,\space,\string}
+\DoNotIndex{\tableofcontents,\the,\thepage,\thispagestyle,\toks,\tt}
+\DoNotIndex{\twocolumn,\uppercase,\vbox,\vrule,\vskip,\vss}
+\DoNotIndex{\write,\z@,\z@skip}
+% \end{macrocode}
+% Some useful macros and parameter settings:
+% \begin{macrocode}
+
+\setcounter{StandardModuleDepth}{1}
+
+\GetFileInfo{index.dtx}
+
+\newcommand*{\email}[1]{$\langle$\texttt{#1}$\rangle$}
+\newcommand*{\vdate}[1]{$\langle$#1$\rangle$}
+\newcommand*{\bundle}[1]{\texttt{#1}}
+\newcommand*{\program}[1]{\textsf{#1}}
+\newcommand*{\Ltag}[1]{\texttt{\bslash#1}}
+\newcommand*{\Lopt}[1]{\textsf {#1}}
+\newcommand*{\Lenv}[1]{\texttt {#1}}
+\newcommand*{\cls}[1]{\texttt {#1}}
+\newcommand*{\pck}[1]{\texttt {#1}}
+\newcommand*{\file}[1]{\texttt {#1}}
+\CodelineIndex
+% \end{macrocode}
+% And the document itself:
+% \begin{macrocode}
+\begin{document}
+\DocInput{index.dtx}
+\PrintIndex
+% ^^A\PrintChanges
+\end{document}
+%</driver>
+% \end{macrocode}
+%
+% \fi
+%
+% \title{A new implementation of \LaTeX's indexing
+% commands\thanks{This file has version number \fileversion,
+% last revised \filedate, documentation dated \docdate. The
+% definitive version of this file is at
+% \file{ftp://theory.lcs.mit.edu/pub/tex/index/}.}}
+%
+% \author{David M. Jones}
+%
+% \date{\filedate}
+%
+% \maketitle
+%
+% \section{Introduction}
+%
+% This style file reimplements \LaTeX's indexing macros to provide
+% better and more robust support for indexes. In particular, it
+% provides the following features:\footnote{Earlier versions of
+% this package provided a ``shortindexing'' feature (see below for
+% description). This feature is now deprecated and will be removed
+% in a future release of this package.}
+% \begin{enumerate}
+%
+% \item
+% Support for multiple indexes.
+%
+% \item
+% Indexing of items by counters other than the page number.
+%
+% \item
+% A $*$-variant of the \cs{index} command that, in addition to
+% putting it's argument in the index, also typesets it in the
+% running text.
+%
+% \item
+% The \bundle{showidx} style option has been merged into this file.
+% The command \cs{proofmodetrue} can be used to enable the printing
+% of index entries in the margin of pages. The size and style of
+% font can be controlled with the \cs{indexproofstyle} command.
+%
+% \item
+% A two-stage process, similar to that used to create tables of
+% contents, for creating the raw index files. This means that when
+% processing a portion of a document using the \cs{includeonly}
+% command, the index entries from the rest of the document are not
+% lost.
+%
+% \item
+% A more robust \cs{index} command. In particular, it no longer
+% depends on \cs{catcode} changes to work properly, so the new
+% \cs{index} command can be used in places that the original
+% couldn't, such as inside the arguments of other macros.
+%
+% \end{enumerate}
+%
+%
+% \section{Creating an index with \LaTeX}
+%
+% Conceptually, there are four stages to creating an index. First,
+% \LaTeX\ must be informed of your intention to include an index in
+% your document. Second, you must add appropriate markup commands
+% to your document to tell \LaTeX\ what to put in the index.
+% Third, after \LaTeX\ has been run on your document, the raw index
+% information must be processed and turned into a form that \LaTeX\
+% can process to typeset the index. Finally, the finished index
+% must be inserted at the appropriate point in your document.
+%
+% In \LaTeX, these steps are accomplished with the commands
+% \cs{makeindex}, \cs{index}, \cs{printindex}, and (typically) with
+% the auxiliary program \program{MakeIndex}. For example, assuming
+% that your main file is called \file{foo.tex}, \cs{makeindex}
+% opens the file \file{foo.idx} and initializes it for holding the
+% raw index entries, and \cs{index} is used to add raw index
+% entries into \file{foo.idx}. Then the raw index file is
+% processed by \program{MakeIndex}, which puts the finished index
+% in \file{foo.ind}. Finally, the \cs{printindex} command is used
+% in your \LaTeX\ document to indicate where the file
+% \file{foo.idx} should be inserted, i.e., where the index should
+% appear in your document.
+%
+% The \bundle{index} package modifies the \cs{makeindex},
+% \cs{index}, and \cs{printindex} commands, as described below.
+%
+%
+% \section{The user interface}
+%
+% There are four pieces of information associated with each index:
+% \begin{enumerate}
+%
+% \item
+% A short, unique tag that identifies the index.
+%
+% \item
+% The extension of the output file where the raw index information
+% will be put by \LaTeX.
+%
+% \item
+% The extension of the input file where the processed information
+% created by \program{MakeIndex} will be stored to be read in later
+% by \LaTeX.
+%
+% \item
+% The title of the index.
+%
+% \end{enumerate}
+%
+% \DescribeMacro{\newindex}
+% Correspondingly, the \cs{newindex} command has four required
+% arguments. For example, to declare an author index, you might
+% use the following:
+% \begin{verbatim}
+% \newindex{aut}{adx}{and}{Name Index}\end{verbatim}
+% Here, \texttt{aut} is the tag used to identify the author index,
+% and ``Name Index'' is the title of the index. If the name of
+% your main file is \file{root.tex}, then \LaTeX\ will write the
+% raw index entries to the file \file{root.adx}, and you will
+% execute the following \program{MakeIndex} command to process the
+% author index:
+% \begin{verbatim}
+% makeindex -o root.and root.adx\end{verbatim}
+%
+% By default, the \cs{index} tags its argument with the page number
+% (i.e., the value of \cs{thepage}), but occasionaly you may want
+% to index items according to a different counter. For example,
+% you may want an index that contains figure numbers instead of
+% page numbers. To accomodate, this, the \cs{newindex} command
+% takes an optional argument, which is the name of the command that
+% generates the number that should be included in the index. For
+% instance, to include the number of a figure, you might say
+% \begin{verbatim}
+% \newindex[thefigure]{fig}{fdx}{fnd}{Figures}\end{verbatim}
+%
+% However, this introduces a new technicality: When creating an
+% index with page numbers, the choice of which page number is to be
+% written to the \texttt{aux} file should be deferred until the
+% page containing the entry is shipped out to the \texttt{dvi}
+% file, otherwise the wrong number will sometimes be chosen.
+% However, when using counters other than the page counter, one
+% normally wants the opposite behaviour: the number written to the
+% \texttt{aux} file should be chosen immediately, otherwise every
+% item on a given page will be tagged with the number of the last
+% item on that page. So, when a counter is specified using the
+% optional argument of \cs{newindex}, it is assumed that the
+% counter should be evaluated immediately. If for some reason you
+% need the choice to be deferred until the page is written to the
+% \texttt{dvi} file, you can force this behaviour by putting a $*$
+% {\em after\/} the optional argument:
+% \begin{verbatim}
+% \newindex[thefigure]*{fig}{fdx}{fnd}{Figures}\end{verbatim}
+% (One consequence of this scheme is that if, for some reason, you
+% need the choice of page number to be made immediately instead of
+% being deferred until a page is shipped out to the \texttt{dvi}
+% file, you can acomplish this by beginning your index declaration
+% with
+% \begin{verbatim}
+% \newindex[thepage]*\end{verbatim}
+%
+%
+% \DescribeMacro{\renewindex}
+% The \cs{renewindex} command takes the same arguments as the
+% \cs{newindex} command and can be used to redefine indexes that
+% have been previously declared.
+%
+%
+% \DescribeMacro{\makeindex}
+% For backwards compatibility, the \cs{makeindex} command is
+% redefined to use \cs{newindex}. It is essentially equivalent to
+% \begin{verbatim}
+% \newindex{default}{idx}{ind}{Index}\end{verbatim}
+% The index labeled \texttt{default} is special: it is the one that
+% will be used by \cs{index} and \cs{printindex} unless another
+% index is specified (see below).
+%
+%
+% \DescribeMacro{\printindex}
+% The \cs{printindex} command is modified by the addition of an
+% optional argument, which is the tag of the index that should be
+% printed.
+%
+%
+% \DescribeMacro{\index}
+% The \cs{index} command is modified in two ways. First, there is
+% a $*$-variant of the command that, in addition to putting its
+% argument into an index, also typesets it on the page. Second,
+% \cs{index} now takes an optional argument to indicate which index
+% the new entry should be added to. If given, the optional
+% argument should be the identifying tag of a previously-defined
+% index. If no such tag is supplied, the \texttt{default} index
+% (such as that opened by \cs{makeindex} above) is used.
+%
+%
+% \DescribeMacro{\shortindexingon}
+% \DescribeMacro{\shortindexingoff}
+% Perhaps the most dubious feature of \bundle{index.sty} is that it
+% allows you to define the characters |^| and |_| to be
+% abbreviations for \cs{index*} and \cs{index} outside of math
+% mode. These abbreviations are enabled by the
+% \cs{shortindexingon} command and disabled by the
+% \cs{shortindexingoff} command. The scope of both of these latter
+% commands is local to the current group. (This might be useful,
+% for example, if you wanted the abbreviations turned on throughout
+% most of the documentation, but turned off in one particular
+% environment.) In addition,
+% \DescribeEnv{shortindexingon}\Lenv{shortindexingon} can be used
+% as an environment if that seems appropriate. \textbf{Warning:
+% This feature is deprecated and will disappear in a future release
+% of this package.}
+%
+%
+% \DescribeMacro{\proofmodetrue}
+% \DescribeMacro{\proofmodefalse}
+% \DescribeMacro{\indexproofstyle}
+% As mentioned above, the \bundle{showidx} document-style option
+% has been merged into \bundle{index.sty}. It can be turned on
+% with \cs{proofmodetrue} and turned off with \cs{proofmodefalse}.
+% When it is turned on, all index entries\footnote{Well, most, at
+% least. There are some circumstances under which the index
+% entries won't show up in the proofs, although they will show up
+% in the index.} will be put in the margin of the page where they
+% appear. By default, they appear in the typewriter font at
+% \cs{footnotesize}, but the user can override this with the
+% \cs{indexproofstyle} command; for example,
+% \begin{verbatim}
+% \indexproofstyle{\footnotesize\it}\end{verbatim}
+% will cause them to be put in italics instead.
+%
+%
+% \DescribeMacro{\disableindex}
+% There are some circumstances where it might be helpful to
+% suppress the writing of a particular index. The
+% \cs{disableindex} command is provided for this purpose. It takes
+% one argument, a comma-separated list of tags of the indexes that
+% should be disabled. This command should come {\em before\/} the
+% declarations for the indexes that are being
+% disabled\footnote{This limits its usefulness somewhat, but since
+% the output file for an index is opened when the index is
+% declared, the damage has already been done. We could close the
+% file, but we can't prevent a new output stream from being
+% allocated and we can't keep the old file from being truncated.}.
+% One situation where the \cs{disableindex} command might be useful
+% is if there are so many indexes that you are exhausting \TeX's
+% supply of output streams\footnote{\TeX\ only has 16 output
+% streams, which are allocated with the {\tt\string\newwrite}
+% command. The standard \LaTeX\ styles use from 3 to 7 of these,
+% which should leave room for up to 9 indexes. Of course, if you
+% have extra output files, then there will be fewer output streams
+% left for indexes.}. For example, suppose you have 10 indexes,
+% but only 5 output streams available for indexes. Then you could
+% add a \cs{disableindex} command to the top of your file to
+% suppress the writing of all but 5 of the indexes. (Note that the
+% index entries would still get written to the \texttt{aux} file;
+% they just wouldn't get copied to the individual raw index files
+% at the end of the run.) At the end of the run, you could then
+% re-run your main file a couple of times with different indexes
+% disabled until you had created all of the raw index files. This
+% is somewhat clumsy, but safer than any alternative I've come up
+% with\footnote{A less clumsy (for the user, at least) solution
+% would be to read the \texttt{aux} file multiple times at the end
+% of the run, each time writing just one of the raw index files.
+% The main disadvantage of this scheme at present is that it would
+% require a modification of {\tt\string\enddocument}.}.
+%
+%
+% \section{Caveats}
+%
+% In order to implement this style file, it's been necessary to
+% modify a number of \LaTeX\ commands seemingly unrelated to
+% indexing, namely, \cs{@starttoc}, \cs{raggedbottom},
+% \cs{flushbottom}, \cs{addcontents}, \cs{markboth}, and
+% \cs{markright}. Naturally, this could cause incompatibilities
+% between \bundle{index.sty} and any style files that either
+% redefine these same commands or make specific assumptions about
+% how they operate. See Section~\ref{sec:thecode} for explanations
+% of why these various commands needed modification.
+%
+% The redefinition of \cs{@starttoc} is particularly bad, since it
+% introduces an incompatibility with the AMS document classes.
+% This will be addressed soon.
+%
+% Unfortunately, it's also been necessary to modify the
+% \Lenv{theindex} environment, so if you don't like the default
+% \LaTeX\ definition, you'll need copy the definition of
+% \Lenv{theindex} from this file and modify it appropriately.
+%
+% In the current implementation, \bundle{index.sty} uses one output
+% stream for each index. Since there are a limited number of
+% output indexes, this means that there is a limit on the number of
+% indexes you can have in a document. See the description of
+% \cs{disableindex} for a fuller discussion of this problem and one
+% way around it.
+%
+%
+% \section{To do's}
+%
+% It might be nice if the \cs{index*} command parsed its argument
+% so that, for example, instead of writing
+% `|\index{sin@$\sin$}$\sin$|', one could write
+% `|index*{sin@$\sin$}|'. However, this is fraught with numerous
+% dangers, and I'm both too lazy and too cowardly to undertake it
+% now.
+%
+% It would be reasonable to add support for \cs{makeglossary} and
+% similar things, if they were well-defined enough to decide what
+% the general syntax for defining them should be.
+%
+% The documentation should be carefully read, edited, and finished,
+% especially since it's still based on the 2.09 version, even
+% though a few substantial changes have been made for the \LaTeXe\
+% version.
+%
+% For some truly outlandish ideas, see the file \file{TODO} in the
+% distribution.
+%
+% \StopEventually{}
+%
+% \section{The code}
+% \label{sec:thecode}
+%
+% As is customary, identify this as a \LaTeXe\ package.
+% \begin{macrocode}
+%<*style>
+\NeedsTeXFormat{LaTeX2e}[1995/06/01]
+
+\ProvidesPackage{index}[2004/01/20 v4.2beta Improved index support (dmj)]
+% \end{macrocode}
+%
+% \begin{macro}{\disableindex}
+% The \cs{disableindex} should come before the declarations of the
+% indexes it refers to. (Question: If an index has been disabled,
+% should it show up in index proofs? Maybe there should be a
+% separate command to disable index proofs on and index-by-index
+% basis.)
+% \begin{macrocode}
+\def\disableindex#1{%
+ \@for\@tempa:=#1\do{%
+ \@namedef{disable@\@tempa}{}%
+ \@ifundefined{tf@\@tempa}{}{%
+ \PackageWarningNoLine{index}{It's too late to disable
+ the `\@tempa' index;\MessageBreak
+ \jobname.\@tempa\space has already
+ been opened for output. You \MessageBreak
+ should put the \string\disableindex\space command
+ before\MessageBreak
+ the declaration of the `\@tempa' index}%
+ }%
+ }%
+}
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\if@newindex}
+% \begin{macro}{\newindex}
+% \begin{macro}{\renewindex}
+% The \cs{newindex} and \cs{renewindex} commands are defined on
+% analogy with the \cs{[re]newcommand} macros. Each index is
+% identified by a unique tag, which is specified in the first
+% required argument of \cs{newindex}. Much of the information
+% about the index labeled \meta{tag} is kept in the macro
+% \cs{idx@}\meta{tag}, so we can check to see if a particular index
+% has already been defined by checking whether \cs{idx@}\meta{tag}
+% is defined. \cs{newindex} and \cs{renewindex} both check to see
+% if their first argument is already associated with an index and
+% then either issue an appropriate error message or call
+% \cs{def@index}.
+%
+% The \cs{if@newindex} flag will be used to keep \cs{renewindex}
+% from re-allocating \cs{write} and \cs{toks} registers later. The
+% \cs{if@tempswa} switch will be used to determine whether the
+% \cs{write}s for this index should be done \cs{immediate}ly or
+% not.
+% \begin{macrocode}
+\newif\if@newindex
+
+\def\newindex{%
+ \@tempswafalse
+ \@ifnextchar[{\@tempswatrue\x@newindex}{\x@newindex[thepage]}%
+}
+
+\def\x@newindex[#1]{%
+ \@ifstar {\@tempswafalse\y@newindex{#1}}
+ {\y@newindex{#1}}%
+}
+
+\def\y@newindex#1#2{%
+ \@ifundefined{idx@#2}%
+ {\@newindextrue\def@index{#1}{#2}}%
+ {%
+ \@latexerr{Index type `\string#2' already defined}\@ehc
+ \expandafter\@gobble\@gobbletwo
+ }%
+}
+
+\def\renewindex{%
+ \@tempswafalse
+ \@ifnextchar[{\@tempswatrue\x@renewindex}{\x@renewindex[thepage]}%
+}
+
+\def\x@renewindex[#1]{%
+ \@ifstar {\@tempswafalse\y@renewindex{#1}}
+ {\y@renewindex{#1}}%
+}
+
+\def\y@renewindex#1#2{%
+ \@ifundefined{idx@#2}%
+ {%
+ \@newindextrue
+ \@latexerr{Index type `\string#2' not defined}\@ehc
+ }%
+ {\@newindexfalse}%
+ \def@index{#1}{#2}%
+}
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\@preamblecmds}
+% Neither \cs{newindex}, \cs{renewindex}, nor \cs{disableindex}
+% should be used anywhere except inside style files or in the
+% preamble of a document, so we add them to the \cs{@preamblecmds}
+% list.
+% \begin{macrocode}
+\@onlypreamble\newindex
+\@onlypreamble\renewindex
+\@onlypreamble\disableindex
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\def@index}
+% \cs{def@index} does most of the work. First, it picks up the
+% first three arguments of the \cs{[re]newindex} command and stores
+% the second two in an appropriate \cs{idx@} macro. The title of
+% the index is treated differently, however, since it is
+% potentially fragile in a particularly odd way. To prevent
+% mishaps, it is stored in a token register. In addition to
+% stashing away the information about the index, \cs{def@index}
+% also opens an appropriate output file if we are writing auxiliary
+% files (i.e., unless \cs{nofiles} is in effect).
+%
+% \begin{macrocode}
+\def\def@index#1#2#3#4{%
+ \@namedef{idx@#2}{#3:#4:#1}%
+ \expandafter\let\csname if@immediate@#2\endcsname\if@tempswa
+ \if@filesw
+ \if@newindex
+ \expandafter\newtoks\csname idxtitle@#2\endcsname
+ \fi
+ \@ifundefined{disable@#2}{%
+ \if@newindex
+ \expandafter\newwrite\csname tf@#2\endcsname
+ \else
+ \immediate\closeout\@nameuse{tf@#2}%
+ \fi
+ \immediate\openout\@nameuse{tf@#2}\jobname.#3 %
+ \PackageInfo{index}{Writing index file \jobname.#3}%
+ }
+ {\PackageInfo{index}{Index `#2' disabled -- not opening
+ \jobname.#3}}%
+ \fi
+ \expandafter\csname idxtitle@#2\endcsname
+}
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@second}
+% \begin{macro}{\@third}
+% These are useful macros for retrieving the second and third field
+% of an index specification.
+% \begin{macrocode}
+\def\@second#1:#2:#3\@nil{#2}
+
+\def\@third#1:#2:#3\@nil{#3}
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\@nearverbatim}
+% |\@nearverbatim\foo| is much like |\meaning\foo|,
+% except that it suppresses the ``\texttt{macro ->}'' string
+% produced when \cs{meaning} expands a macro. It is used by
+% \cs{@wrindex} to produce an ``almost verbatim'' copy of their
+% arguments. This method replaces the use of \cs{@sanitize} from
+% latex.tex and allows indexing macros to be used in places (such
+% as inside macro arguments) where the original \cs{index} command
+% could not. Thanks to Donald Arseneau
+% \email{asnd@erich.triumf.ca} for pointing out this trick to me.
+% (For more information on this trick, see Dirty Trick \#3 of the
+% \TeX book, page 382).
+%
+% As defined, \@nearverbatim only works on macros. It would be
+% nice if it could work with other tokens, but it's more important
+% that it work only by expansion, which means we can't put in tests
+% to see what the next token is.
+% \begin{macrocode}
+\def\@nearverbatim{\expandafter\strip@prefix\meaning}
+% \end{macrocode}
+% \end{macro}
+%
+% Now we define the \cs{index} macro itself. The following
+% definitions are adapted from \bundle{latex.tex} v2.09 \vdate{25
+% March 1992}.
+%
+% \begin{macro}{\makeindex}
+% First we redefine \cs{makeindex} to define the default index
+% using \cs{newindex}. We use \cs{edef} to make sure that
+% \cs{indexname} gets expanded here. Otherwise we'll get into an
+% infinite loop later on when we try to redefine \cs{indexname}
+% inside the \cs{theindex} environment.
+%
+% Unfortunately, this means that if the user changes \cs{indexname}
+% in the preamble, the index will come out with the wrong heading.
+% \begin{macrocode}
+\edef\makeindex{%
+ \noexpand\newindex{default}{idx}{ind}{\indexname}%
+}
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\if@silentindex}
+% \begin{macro}{\if@addtoindex}
+% \begin{macro}{\if@proofmode}
+% We need three new flags. The first, \cs{if@silentindex},
+% indicates whether the entry should be typeset in running text, as
+% well as written out to the index; this is used to implement the
+% \cs{index*} command. The second, \cs{if@addtoindex}, indicates
+% whether entries should be written to the index; this is used to
+% disable the \cs{index} command inside of page headings and tables
+% of contents. The third, \cs{ifproofmode}, indicates whether
+% index entries should be put in the margin of the page for
+% proofing purposes.
+% \begin{macrocode}
+\newif\if@silentindex\@silentindextrue
+
+\newif\if@addtoindex\@addtoindextrue
+
+\newif\ifproofmode\proofmodefalse
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\index}
+% \begin{macro}{\p@index}
+% \begin{macro}{\x@index}
+% \cs{index} will be made self-protecting (a la \cs{em}, etc.) so
+% it can be used inside, for example, sectioning commands.
+% Unfortunately, to really make \cs{index} robust, we have to
+% redefine some of \LaTeX's commands for dealing with tables of
+% contents and page headings. (See below.) $*$sigh$*$
+% \begin{macrocode}
+\def\index{\protect\p@index}
+
+\def\p@index{%
+ \if@silentindex\@bsphack\fi
+ \@ifstar{\@silentindexfalse\@xindex}{\@silentindextrue\@xindex}%
+}
+
+\def\@xindex{\@ifnextchar[{\@index}{\@index[default]}}
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\@index}
+% \begin{macro}{\@@index}
+% \begin{macro}{\@wrindex}
+% The following is much more complicated than it should have to be.
+% First, note the check to see if \cs{index} is equal to
+% \cs{@gobble}. This is so I don't have to redefine
+% \cs{@outputpage}, which temporarily disables \cs{label},
+% \cs{index}, and \cs{glossary} by \cs{let}'ing them equal to
+% \cs{@gobble}. (For this reason, we have to be very careful to
+% make sure that \cs{index} has expanded to \cs{p@index} before it
+% gets to \cs{@outputpage}.) Second, note that if
+% \cs{if@addtoindex} is false, we don't complain about undefined
+% index types. This is because if your page headings, for example,
+% are being typeset in all uppercase, you might end up with
+% something like \cs{index[AUT]{...}} instead of
+% \cs{index[aut]{...}}.
+% \begin{macrocode}
+\def\@index[#1]{%
+ \ifx\index\@gobble
+ \@addtoindexfalse
+ \fi
+ \def\@tempf{%
+ \begingroup
+ \@sanitize
+ \@@index{#1}%
+ }%
+ \if@addtoindex
+ \@ifundefined{idx@#1}%
+ {%
+ \def\@tempf{%
+ \@latexerr{Index type `\string#1' undefined}%
+ \@ehc
+ \@silentindextrue
+ \@gobble
+ }%
+ }%
+ {}%
+ \fi
+ \@tempf
+}
+
+\def\@@index#1#2{%
+ \endgroup
+ \if@addtoindex
+ \if@filesw\@wrindex{#1}{#2}\fi
+ \ifproofmode\@showidx{#2}\fi
+ \fi
+ \if@silentindex
+ \expandafter\@esphack
+ \else
+ \@silentindextrue#2%
+ \fi
+}
+
+\def\@wrindex#1#2{%
+ \begingroup
+ \def\@tempa{#2}%
+ \edef\@tempb{\@nameuse{idx@#1}}%
+ \edef\@tempb{\expandafter\@third\@tempb\@nil}%
+ \csname if@immediate@#1\endcsname \else
+ \expandafter\let\csname\@tempb\endcsname\relax
+ \fi
+ \edef\@tempa{%
+ \write\@auxout{%
+ \string\@writefile{#1}{%
+ \string\indexentry{\@nearverbatim\@tempa}%
+ {\@nameuse{\@tempb}}%
+ }%
+ }%
+ }%
+ \expandafter\endgroup\@tempa
+ \if@nobreak\ifvmode\nobreak\fi\fi
+}
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\seename}
+% \begin{macro}{\see}
+% \begin{macro}{\printindex}
+% \begin{macro}{\@printindex}
+% The following are adapted from \bundle{makeidx.sty}, v2.09
+% \vdate{21 Oct 91}. \cs{index@prologue} adapted from
+% \bundle{doc.dtx}. \Lenv{theindex} based on version from
+% \bundle{classes.dtx}, v1.3g, 26 June 1995.
+% \begin{macrocode}
+\providecommand{\seename}{see}
+
+\providecommand*{\see}[2]{\emph{\seename} #1}
+
+\@ifclassloaded{article}{%
+
+ \renewenvironment{theindex}{%
+ \edef\indexname{\the\@nameuse{idxtitle@\@indextype}}%
+ \if@twocolumn
+ \@restonecolfalse
+ \else
+ \@restonecoltrue
+ \fi
+ \columnseprule \z@
+ \columnsep 35\p@
+ \twocolumn[%
+ \section*{\indexname}%
+ \ifx\index@prologue\@empty\else
+ \index@prologue
+ \bigskip
+ \fi
+ ]%
+ \@mkboth{\MakeUppercase\indexname}%
+ {\MakeUppercase\indexname}%
+ \thispagestyle{plain}%
+ \parindent\z@
+ \parskip\z@ \@plus .3\p@\relax
+ \let\item\@idxitem
+ }{%
+ \if@restonecol
+ \onecolumn
+ \else
+ \clearpage
+ \fi
+ }
+}{%
+ \renewenvironment{theindex}{%
+ \edef\indexname{\the\@nameuse{idxtitle@\@indextype}}%
+ \if@twocolumn
+ \@restonecolfalse
+ \else
+ \@restonecoltrue
+ \fi
+ \columnseprule \z@
+ \columnsep 35\p@
+ \twocolumn[%
+ \@makeschapterhead{\indexname}%
+ \ifx\index@prologue\@empty\else
+ \index@prologue
+ \bigskip
+ \fi
+ ]%
+ \@mkboth{\MakeUppercase\indexname}%
+ {\MakeUppercase\indexname}%
+ \thispagestyle{plain}%
+ \parindent\z@
+ \parskip\z@ \@plus .3\p@\relax
+ \let\item\@idxitem
+ }{%
+ \if@restonecol
+ \onecolumn
+ \else
+ \clearpage
+ \fi
+ }
+}
+
+\def\printindex{\@ifnextchar[{\@printindex}{\@printindex[default]}}
+
+\def\@printindex[#1]{%
+ \@ifnextchar[{\@print@index[#1]}{\@print@index[#1][]}%
+}
+
+\long\def\@print@index[#1][#2]{%
+ \def\@indextype{#1}%
+ \long\def\index@prologue{#2}%
+ \@ifundefined{idx@#1}%
+ {\@latexerr{Index type `\string#1' undefined}\@ehc}%
+ {%
+ \edef\@tempa{\@nameuse{idx@#1}}%
+ \edef\@tempa{%
+ \noexpand\@input@{\jobname.\expandafter\@second\@tempa\@nil}%
+ }%
+ \@tempa
+ }%
+}
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\@indexstar@}
+% Now we set things up for
+% \cs{shortindexing}.\footnote{\textbf{Warning:} This feature is
+% deprecated and will be removed entirely in a future release of
+% this package.} First, we define a one-token shorthand for
+% \cs{index*}. This will be needed in the definition of
+% \cs{idx@activehat}.
+% \begin{macrocode}
+\def\@indexstar@{\index*}
+% \end{macrocode}
+% \end{macro}
+% \begin{macro}{\idx@activehat}
+% \begin{macro}{\idx@activebar}
+% Next, we define the values that |^| and |_| will have when
+% shortindexing is turned on.
+% \begin{macrocode}
+\def\idx@activehat{%
+ \relax
+ \ifmmode\expandafter\sp\else\expandafter\@indexstar@\fi
+}
+
+\def\idx@activebar{%
+ \relax
+ \ifmmode\expandafter\sb\else\expandafter\index\fi
+}
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\shortindexingon}
+% \begin{macro}{\shortindexingoff}
+% Now we define the \cs{shortindexingon} and \cs{shortindexinoff}
+% commands to turn shortindexing on and off (surprise!).
+% \cs{shortindexingon} saves the old definitions and \cs{catcode}'s
+% of |^| and |_| so they can later be restored by
+% \cs{shortindexingoff}. Both of these make their changes local to
+% any enclosing group, so they can be used as declarations to
+% disable or enable shortindexing temporarily. In addition,
+% \Lenv{shortindexingon} can also be used as an environment.
+%
+% This is potentially very confusing. My basic rationale (if it
+% can be described as such) was that under normal circumstances,
+% one would put \cs{shortindexingon} in the preamble of one's
+% document, and never want to turn it off. \cs{shortindexingoff}
+% is an attempt to make allowance for the contingency that someone
+% might want to turn shortindexing off, either permanently or
+% temporarily.
+% \begin{macrocode}
+\newif\if@shortindexing
+
+\begingroup
+
+ \catcode`\^\active
+ \catcode`\_\active
+
+ \gdef\shortindexingon{%
+ \@shortindexingtrue
+ \chardef\old@idxhatcode\catcode`\^\relax
+ \chardef\old@idxbarcode\catcode`\_\relax
+ \catcode`\^\active
+ \catcode`\_\active
+ \let\old@idxhat ^%
+ \let\old@idxbar _%
+ \let^\idx@activehat
+ \let_\idx@activebar
+ }
+
+ \gdef\shortindexingoff{%
+ \if@shortindexing
+ \@shortindexingfalse
+ \let^\old@idxhat
+ \let_\old@idxbar
+ \catcode`\^\old@idxhatcode
+ \catcode`\_\old@idxbarcode
+ \fi
+ }
+
+\endgroup
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% Now we take some code from \bundle{showidx.sty} and merge it into
+% our new system. There are four reasons for redefining the
+% commands here rather than just inputting \bundle{showidx.sty} (or
+% requiring the user to do so). First, \bundle{showidx.sty} ends
+% with a call to \cs{flushbottom}, which I want to avoid. Second,
+% the instructions for successfully using \bundle{showidx.sty}
+% along with \bundle{index.sty} would be somewhat tricky. This
+% way, I can just tell users not to use \bundle{showidx.sty} at
+% all. Third, I need to make some alterations to \cs{@showidx}
+% anyway. In particular, (a) I need to add the \cs{@sanitizeat}
+% command so this works correctly with AMS-\LaTeX\ and (b) I want
+% to add the \cs{indexproofstyle} command so the user can customize
+% the size and font used for the index proofs. Finally,
+% \bundle{showidx.sty} has at least two annoying bugs in it. See
+% the edit-history for version 2.01 for a description.
+%
+% \begin{macro}{\@indexbox}
+% This code is adapted from \bundle{showidx.sty}, v2.09 \vdate{16
+% Jun 1991}.
+% \begin{macrocode}
+\newinsert\@indexbox
+
+\dimen\@indexbox\maxdimen
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@sanitizeat}
+% The definition of \cs{@sanitizeat} is slightly tricky, since we
+% need |@| to be active when this macro is defined, but we also
+% need it to be part of the control sequence name.
+% \begin{macrocode}
+\begingroup
+ \catcode`\@\active
+ \expandafter\gdef\csname\string @sanitizeat\endcsname
+ {\def @{\char`\@}}
+\endgroup
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\indexproofstyle}
+% \begin{macro}{\@showidx}
+% \begin{macro}{\@leftidx}
+% \begin{macro}{\@rightidx}
+% \begin{macro}{\@mkidx}
+% \begin{macro}{\raggedbottom}
+% \begin{macro}{\flushbottom}
+% \begin{macro}{\@texttop}
+% \begin{macrocode}
+\newtoks\indexproofstyle
+
+\indexproofstyle{\footnotesize\reset@font\ttfamily}
+
+\def\@showidx#1{%
+ \insert\@indexbox{%
+ \@sanitizeat
+ \the\indexproofstyle
+ \hsize\marginparwidth
+ \hangindent\marginparsep \parindent\z@
+ \everypar{}\let\par\@@par \parfillskip\@flushglue
+ \lineskip\normallineskip
+ \baselineskip .8\normalbaselineskip\sloppy
+ \raggedright \leavevmode
+ \vrule \@height .7\normalbaselineskip \@width \z@\relax#1\relax
+ \vrule \@height\z@ \@depth.3\normalbaselineskip \@width\z@\relax
+ }%
+ \ifhmode\penalty\@M \hskip\z@skip\fi
+}
+
+\def\@leftidx{\hskip-\marginparsep \hskip-\marginparwidth}
+
+\def\@rightidx{\hskip\columnwidth \hskip\marginparsep}
+
+\def\@mkidx{%
+ \vbox to \z@{%
+ \rlap{%
+ \if@twocolumn
+ \if@firstcolumn \@leftidx \else \@rightidx \fi
+ \else
+ \if@twoside
+ \ifodd\c@page \@rightidx \else \@leftidx \fi
+ \else
+ \@rightidx
+ \fi
+ \fi
+ \box\@indexbox
+ }%
+ \vss
+ }%
+}
+
+\def\raggedbottom{%
+ \def\@textbottom{\vskip\z@ plus.0001fil}%
+ \let\@texttop\@mkidx
+}
+
+\def\flushbottom{\let\@textbottom\relax \let\@texttop\@mkidx}
+
+\let\@texttop\@mkidx
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% Now, this next bit really gets up my nose. The only way to make
+% sure that the \cs{index} command gets handled correctly when used
+% inside of sectioning commands is to redefine a bunch of \LaTeX's
+% table of contents and running-heads macros. $*$blech$*$ Fragility
+% rears its ugly head again.
+%
+% These are based on \bundle{latex.tex} 2.09 \vdate{25 March 1992}.
+%
+% \begin{macro}{\addcontentsline}
+% We need to redefine \cs{addcontentsline} to keep it from
+% expanding \cs{index} commands too far. In particular, we have
+% removed \cs{index} from the list of macros that are set equal to
+% \cs{@gobble} and we substitute \cs{@vwritefile} for
+% \cs{@writefile}. This latter change also means that we can
+% simplify the definition of \cs{protect} somewhat.
+% \begin{macrocode}
+\CheckCommand\addtocontents[2]{%
+ \protected@write\@auxout
+ {\let\label\@gobble \let\index\@gobble \let\glossary\@gobble}%
+ {\string\@writefile{#1}{#2}}%
+}
+
+\renewcommand{\addtocontents}[2]{%
+ \protected@write\@auxout
+ {\let\label\@gobble \let\glossary\@gobble}%
+ {\string\@writefile{#1}{#2}}%
+}
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@starttoc}
+% We need to redefine \cs{@starttoc} to \cs{@addtoindexfalse} so
+% that items don't get written to the index from within tables of
+% contents. The only change here is the addition of
+% \cs{@addtoindexfalse}.
+%
+% Unfortunately, this will break pretty badly with the AMS document
+% classes, since they redefine \cs{@starttoc} to take two arguments
+% rather than one. This must be addressed.
+%
+% \begin{macrocode}
+\let\old@starttoc\@starttoc
+
+\renewcommand{\@starttoc}[1]{%
+ \begingroup
+ \@addtoindexfalse
+ \old@starttoc{#1}%
+ \endgroup
+}
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\markboth}
+% \begin{macro}{\markright}
+% Finally, we have to redefine \cs{markboth} and \cs{markright} to
+% keep them from disabling the expansion of \cs{index} while
+% putting section heads into the \cs{mark}. Otherwise, we'd end up
+% with ``\cs{index}'' in the mark, which would cause problems when
+% \cs{@outputpage} redefines \cs{index} to be equal to
+% \cs{@gobble}. Instead, we want \cs{index} to expand to
+% \cs{p@index} in the \cs{mark}, so we retain control over what
+% happens in \cs{@outputpage}.
+%
+% This time, the only change is to remove \cs{index} from the list
+% of macros that are \cs{let} equal to \cs{relax}.
+% \begin{macrocode}
+\CheckCommand*{\markboth}[2]{%
+ \begingroup
+ \let\label\relax \let\index\relax \let\glossary\relax
+ \unrestored@protected@xdef\@themark {{#1}{#2}}%
+ \@temptokena \expandafter{\@themark}%
+ \mark{\the\@temptokena}%
+ \endgroup
+ \if@nobreak\ifvmode\nobreak\fi\fi}
+\CheckCommand*{\markright}[1]{%
+ \begingroup
+ \let\label\relax \let\index\relax \let\glossary\relax
+ \expandafter\@markright\@themark {#1}%
+ \@temptokena \expandafter{\@themark}%
+ \mark{\the\@temptokena}%
+ \endgroup
+ \if@nobreak\ifvmode\nobreak\fi\fi}
+
+\renewcommand{\markboth}[2]{%
+ \begingroup
+ \let\label\relax \let\glossary\relax
+ \unrestored@protected@xdef\@themark {{#1}{#2}}%
+ \@temptokena \expandafter{\@themark}%
+ \mark{\the\@temptokena}%
+ \endgroup
+ \if@nobreak\ifvmode\nobreak\fi\fi}
+
+\renewcommand{\markright}[1]{%
+ \begingroup
+ \let\label\relax \let\glossary\relax
+ \expandafter\@markright\@themark {#1}%
+ \@temptokena \expandafter{\@themark}%
+ \mark{\the\@temptokena}%
+ \endgroup
+ \if@nobreak\ifvmode\nobreak\fi\fi}
+%</style>
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \section{Edit history}
+%
+% \begin{description}
+%
+% \item[v1.00 (4 Mar 1993)]
+% initial version, posted to comp.text.tex.
+%
+% \item[v1.01 (4 Mar 1993)]
+% added \cs{renewindex} command and checking to make sure index is
+% (or is not) defined in \cs{newindex}, \cs{index} and
+% \cs{printindex}. Also tightened up the code in various places
+% and added check to make sure file is only loaded once.
+%
+% \item[v2.00 (24 Mar 1993)]
+% added support for \cs{index*}, proofmode, \cs{shortindexingon}
+% and \cs{shortindexingoff}.
+%
+% \item[v2.01 (24 Jun 1993)]
+% Fixed 3 bugs. (1) If proofmode was turned on, then something
+% like ``\cs{index{WORD}WORD}'' would suppress the hyphenation of
+% WORD. This was fixed by adding ``|\penalty\@M\hskip\z@skip|'' to
+% the end of \cs{@showidx}. (This is just the definition of
+% \cs{allowhyphens} borrowed from \bundle{german.sty}, v2 \vdate{4
+% Nov 1988}). (2) The \cs{hbox} in \cs{@mkidx} was being set at
+% its natural width, which had a tendency to interfere with the
+% width of the page. The \cs{hbox} is now replaced by \cs{rlap}.
+% (3) If the title of an index (i.e., the fourth argument of
+% \cs{newindex}) contained a particularly fragile command
+% like~\cs{d}, havoc would ensue when \cs{theindex} tried to
+% extract the title. Titles are now kept in token registers to
+% prevent such unpleasantness. Bugs (2) and (3) were reported by
+% Dominik Wujastyk \email{D.Wujastyk@ucl.ac.uk} on 24 June 1993.
+% Note that bugs (1) and (2) are actually bugs in showidx.sty,
+% v2.09 \vdate{16 Jun 1991}.
+%
+% \item[v2.02 (25 Jun 1993)]
+% Rewrote the code that implements the short indexing commands (|^|
+% and |_|) to make index.sty compatible with other style files that
+% need to make |^| and |^| active in some contexts. See the code
+% for more details.
+%
+% \item[v2.03 (30 Jun 1993)]
+% Once again rewrote the code that implements the short indexing
+% commands. Dumped the shortindexing environment and rewrote the
+% \cs{shortindexingon} and \cs{shortindxingoff} commands to save
+% and restore the \cs{catcode}'s and meanings of |^| and |^| in the
+% safest possible (I hope) order. Also added the
+% \cs{if@shortindexing} flag to keep \cs{shortindexingoff} from
+% doing anything if it is called outside of the scope of a
+% \cs{shortindexingon} command. (Question: Should
+% \cs{shortindexingon} check that flag before doing anything?)
+%
+% \item[v2.04 (beta) (14 Jul 1993)]
+% Added \cs{disableindex} command. Added \cs{newindex} and
+% \cs{renewindex} to \cs{@preamblecmds}. Added \cs{if@newindex}
+% flag to \cs{@newindex} to prevent \cs{renewindex} from
+% re-allocating new \cs{write} and \cs{toks} registers. Rewrote
+% using \bundle{doc.sty} and \program{DocStrip}. Also cleaned up
+% the code somewhat.
+%
+% \item[v3.00 (15 Jul 1993)]
+% Made further minor tweaks to code and internal documentation.
+% Booted version number up to 3.00 and released on the world.
+%
+% \item[v3.01 (19 Jul 1993)]
+% Fixed \program{DocStrip} CheckSum.
+%
+% \item[v3.02 (15 Sep 1993)]
+% Corrected spelling of \cs{@shortindexingfalse} in definition of
+% \cs{shortindexingoff}. Thanks to Hendrik G. Seliger
+% \email{hank@Blimp.automat.uni-essen.de} for this bug report.
+% Also added redefinitions of \cs{@leftmark} and \cs{@rightmark} to
+% fix a bug reported by Dominik Wujastyk
+% \email{D.Wujastyk@ucl.ac.uk}.
+%
+% \item[v3.03 (beta) (20 Feb 1994)]
+% Added \cs{long} to the definition of \cs{@ifundefined} to cover
+% the unlikely contingency that someone wanted to use, for example,
+% |\string\par| in the middle of a control sequence name. Added an
+% optional argument to \cs{newindex} to specify which counter to
+% use in place of \cs{thepage}. The first change was suggested by
+% Martin Schr\"oder \email{l15d@zfn.uni-bremen.de}; the second was
+% suggested independently by Schr\"oder and Stefan Heinrich
+% H\"oning \email{hoening@pool.informatik.rwth-aachen.de}. The
+% \cs{@newindex} command was renamed \cs{def@index}. Also fixed
+% the \cs{disableindex} command.
+%
+% \item[v3.04 (7 Mar 1994)]
+% Rewrote the user documentation (Sections 1--5) and released on
+% the world. Also deleted some extraneous spaces that had crept
+% into some macros.
+%
+% \item[v4.00beta, (20 Feb 1995)]
+% Preliminary conversion to a native \LaTeXe\ package. Fixed
+% \cs{@printindex} to work under \LaTeXe\ (bug reported by Carsten
+% Folkertsma \email{cai@butler.fee.uva.nl}). Removed much code
+% that had been put in to work around various ancient versions of
+% \LaTeX~2.09. Added \cs{index@prologue} support (modelled on
+% \bundle{doc.sty}) at suggestion of Nick Higham
+% \email{higham@ma.man.ac.uk}.
+%
+% \item[v4.01beta (28 Sep 1995)]
+% Rewrote as a \LaTeXe\ package (finally!). Changes too numerous
+% to list, but in general deleted some now-superfluous code,
+% replaced some tricks by tricks from the \LaTeXe\ kernel, and
+% added some bullet-proofing. Much still remains to be done, but
+% this should be good enough for testing.
+%
+% Changed definition of \cs{protect} in \cs{markright} and
+% \cs{markboth} to fix bug reported by Dominik Wujastyk.
+%
+% \item[??? (5 Jan 2004)]
+%
+%
+% \end{description}
+%
+% \DisableCrossrefs
+%
+% \section{The sample file}
+%
+% \begin{macrocode}
+%<*sample>
+%% latex sample.tex
+%% makeindex sample
+%% makeindex -o sample.and sample.adx
+%% makeindex -o sample.nnd sample.ndx
+%% makeindex -o sample.lnd sample.ldx
+%% latex sample.tex
+
+\documentclass{book}
+\usepackage{index}
+
+\listfiles
+
+\makeindex
+\newindex{aut}{adx}{and}{Name Index}
+\newindex{not}{ndx}{nnd}{List of Notation}
+
+\newindex[theenumi]{list}{ldx}{lnd}{Items}
+
+\shortindexingon
+
+\proofmodetrue
+
+\def\aindex{\index*[aut]}
+
+\begin{document}
+
+\tableofcontents
+
+\newpage
+
+\chapter{Here is a ^[aut]{chapter} title}
+
+\section{Section header\index[aut]{section}}
+
+Here is some text.\index{subject}
+
+Here is \index[not]{notation}some more \index[not]{sin@$\sin$}
+text.
+
+\newpage
+
+Here is some ^{more} _[not]{notation} text.
+
+Here is yet more \aindex{text}.
+
+\section{Another Section header _[aut]{section2}}
+
+And here is some math: $x^1_b$.
+
+Here is an ^[aut]{index} entry \fbox{inside an
+\index[not]{min@$\min$}fbox}
+
+\fbox{Here is an ^[aut]{entry} in a box.}
+
+\section{An indexed list environment}
+
+\begin{enumerate}
+
+\item
+First item
+
+\item
+Second item\index[list]{second item}
+
+\item
+Third item
+
+\newpage
+
+\item
+Fourth item
+
+\item
+Fifth item\index[list]{fifth item}
+
+\item
+Sixth item
+
+\end{enumerate}
+
+\printindex[not]
+
+\printindex[aut][Here is a prologue for the author index.
+Note that it is set in a single column at the top of the
+first page of the index.]
+
+\printindex[list]
+
+\printindex
+
+\end{document}
+%</sample>
+% \end{macrocode}
+%
+% \Finale
+%
+\endinput
diff --git a/Master/texmf-dist/source/latex/index/index.ins b/Master/texmf-dist/source/latex/index/index.ins
new file mode 100644
index 00000000000..84d221a5b7a
--- /dev/null
+++ b/Master/texmf-dist/source/latex/index/index.ins
@@ -0,0 +1,8 @@
+\def\batchfile{index.ins}
+\input docstrip.tex
+
+\generateFile{index.sty}{t}{\from{index.dtx}{style}}
+
+\generateFile{sample.tex}{t}{\from{index.dtx}{sample}}
+
+\endinput