path: root/macros/latex/contrib/sauerj/collect.dtx

% \iffalse meta-comment
% Line endings: UNIX
% Tab size:     4
%
% Copyright 2004 Jonathan Sauer
%
% This work may be distributed and/or modified under the
% conditions of the LaTeX Project Public License, either version 1.3
% 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 1.3 or later is part of all distributions of LaTeX
% version 2003/12/01 or later.
%
% This work has the LPPL maintenance status "maintained".
%
% The Current Maintainer of this work is Jonathan Sauer
% (<jonathan.sauer@gmx.de>).
%
% This work consists of the files collect.dtx and collect.ins
% and the derived file collect.sty.
%
% \fi
%
% \iffalse
%
%<*driver>
\documentclass{ltxdoc}
\usepackage{collect}
\EnableCrossrefs
\CodelineIndex
\RecordChanges
% Modification of verbatim for tabs in listings
\makeatletter
{\catcode`\ =\active%
\catcode`\^^I=\active%
\gdef\@vobeyspaces{%
\catcode`\ \active\let \@xobeysp%
\catcode`\^^I\active\def^^I{~~}%
}}%
\makeatother
\begin{document}
	\DocInput{\jobname.dtx}
\end{document}
%</driver>
%
%<package>\NeedsTeXFormat{LaTeX2e}
%<package>\ProvidesPackage{collect}
%<package>   [2004/09/12 v0.9 collect the contents of an environment]
%
% \fi
%
% \CheckSum{256}
% \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         \~}
%
%
% \DoNotIndex{\@currenvir,\@ehc,\@ifundefined,\@input}
% \DoNotIndex{\@tempa,\@tempb,\bgroup,\catcode,\closeout,\csname}
% \DoNotIndex{\def,\edef,\egroup,\else,\end,\endcsname,\expandafter}
% \DoNotIndex{\fi,\futurelet,\gdef,\global,\ifx,\immediate,\jobname,\let,\long}
% \DoNotIndex{\meaning,\newcommand,\newenvironment,\newif,\newwrite}
% \DoNotIndex{\openout,\PackageError,\par,\strip@prefix,\the,\toks@}
% \DoNotIndex{\write}
%
% \GetFileInfo{\jobname.sty}
%
% \title{The \textsf{collect} package\thanks{This document
% corresponds to \textsf{\filename}~\fileversion, dated \filedate.}}
%
% \author{Jonathan Sauer \\ \texttt{jonathan.sauer@gmx.de}}
%
% \date{\filedate}
%
% \maketitle
%
% \begin{abstract}
% This file describes the \textsf{collect} package that
% makes it possible to collect text for later use.
%
% \end{abstract}
%
% \tableofcontents
%
% \section{Introduction}
%
% Suppose you are writing the specification of a programming language.
% Then you will surely insert definitions of the grammar (i.e. using
% the \textsf{syntax} package). Most likely you will insert the
% grammar for i.e. loops, conditions et cetera in the appropriate
% chapter, but for easy implementation of a parser, you will want to
% include the complete grammar as an appendix, so that one does not
% have to collect the complete grammar from the bits in each chapter.
%
% Of course you could do this by hand using the copy-and-paste
% facility of your text editor. But this is cumbersome and errorprone
% if you modify the grammar afterwards, i.e. for a second version.
%
% This package provides the |collect| environment that typesets
% its contents and collects it for later use as well.
%
% In other situations you will want to save the contents of an
% environment in a macro. Then the |collectinmacro| environment
% should do the trick.
%
%
%
% \section{Usage}
%
% \paragraph{General note} If an environment exists in a normal and a
% starred form (i.e. |collect| and |collect*|), then the former does
% not typeset the environment contents, while the latter typesets it.
% Note, too, that the normal and the starred version of an environment
% often have a different number of parameters.
%
% \paragraph{Stripping of spaces} All environments strip any leading
% space before the start of the text and any trailing space after the
% end of the text. Trailing |\par|s (i.e. resulting from a blank line
% at the end of the environment) are \emph{not} removed.
%
%
%
%
% \subsection{\texttt{collect}}
%
% You can define several \emph{collections}. Each collection can be
% used independently to collect text across the document. You define a
% collection using |\definecollection|, use it in the |collect| and
% |collect*| environment and typeset it using |\includecollection|.
%
% \paragraph{Note} The name of the collection is used as the suffix of
% the file that stores the collection (the complete name is
% \meta{jobname}|.|\meta{name)}. So you should not name a collection
% |tex|, |log|, |dvi|, |pdf| and so on.
%
% \paragraph{Note} You can only include a collection after completely
% collecting it. If you include it and continue to collect afterwards,
% the collection will be cleared after its inclusion.
%
% This can be used to recycle a collection: Use it in the first part
% of your document to collect text, typeset this text using
% |\includecollection| and then reuse the collection in the second
% part of your document to collect new text which will finally be
% typeset using |\includecollection| a second time.
%
%
% \DescribeMacro{\definecollection} Usage: |\definecollection|
% \marg{name}.
%
% Defines a collection of the name \meta{name}.
%
%
% \DescribeEnv{collect*} Usage: |\begin{collect*}|
% \marg{name} \marg{before} \marg{after} \marg{beforecol}
% \marg{aftercol} \ldots\ |\end{collect*}|.
%
% Collects the text inside the environment in the collection
% \meta{name}. Then typesets the text. \meta{name} must have been
% defined using |\definecollection|.
%
% \meta{before} and \meta{after} are tokens inserted before and after
% the environment contents in the collection as well as the typeset
% text.
%
% \meta{beforecol} and \meta{aftercol} are tokens inserted before and
% after the environment contents only in the collection. This can be
% used to insert a section heading (\meta{beforecol}) and a backref to
% the section where the text has been collected (\meta{aftercol}).
%
% \paragraph{Note} The |collect*| environment uses a temporary file
% \meta{jobname}|.tmp| to temporarily store the environment contents
% and include it afterwards.\footnote{This way, catcode changes inside
% the environment are honored.}
%
%
% \DescribeEnv{collect} Usage: |\begin{collect}| \marg{name}
% \marg{beforecol} \marg{aftercol} \ldots\ |\end{collect}|.
%
% Collects the text inside the environment in the collection
% \meta{name}. \emph{Does not} typeset the text (therefore, the two
% parameters \meta{before} and \meta{after} are missing); except for
% that, this environment is identical to |collect*|.
%
%
% \DescribeMacro{\includecollection} Usage: |\includecollection|
% \marg{name}. Includes the collection \meta{name}. Afterwards, the
% collection is reset.
%
% You can include the collection \emph{before} it is actually
% collected in the same way you can include a table of contents at the
% beginning of your document. Simply say |\includecollection| and then
% use the |collect*| or |collect| environment to collect text. Then
% run \LaTeX\ on your file \emph{twice}. (the first time to collect
% the text, the second time to put it in the document using
% |\includecollection|)
%
%
%
% \subsection{Other environments}
%
% \DescribeEnv{collectinmacro} Usage: |\begin{collectinmacro}|
% \marg{macro} \marg{before} \marg{after} \ldots\
% |\end{collectinmacro}|.
%
% Collects the contents of an environment inside a macro \meta{macro}
% without typesetting it. If \meta{macro} has been defined prior to
% using this environment, its previous meaning is lost. The new
% definition is global.
%
% \meta{before} and \meta{after} are tokens inserted before and after
% the environment contents in the macro.
%
% \paragraph{Important note} This environment differs slightly from
% |collect| and |collect*|: Some macros, i.e. |\verb|, do not work
% correctly when \meta{macro} is used later on.\footnote{As the
% catcodes have already been set when collecting the contents and
% cannot be changed afterwards, at least not without e\TeX.}
%
%
%
%
% \section{Examples}
%
% \subsection{\texttt{collect}}
%
% \begin{verbatim}
% \begin{collect*}{tst}{Before\par}{\par After}{Before file\par}{\par After file}
% This is a test
% \end{collect*}
% \end{verbatim}
%
% \noindent This results in the following text typeset directly:
%
% \begin{verbatim}
% Before
% This is a test
% After
% \end{verbatim}
%
% \noindent And following text is typeset when the collection |tst| is
% included using |\includecollection{tst}|:
%
% \begin{verbatim}
% Before file
% Before
% This is a test
% After
% After file
% \end{verbatim}
%
%
%
%
% \subsection{Other environments}
%
% \begin{verbatim}
% \begin{collectinmacro}{\collectedtext}{Before\par}{\par After}
% This is a test
% \end{collectinmacro}
% \end{verbatim}
%
% \noindent This results in no text typeset directly and the following
% text stored in the macro |\collectedtext|:
%
% \begin{verbatim}
% Before\par This is a test\par After
% \end{verbatim}
%
%
%
%
% \section{Creating your own environment}
%
% You can create your own environment based on any of the environments
% in this package, i.e. to create a |grammarpart| environment to
% typeset part of a grammar.
%
% You can for example say:
%
% \begin{verbatim}
% \definecollection{gra}
%
% \newenvironment{grammarpart}[1]{%
% 	\@nameuse{collect*}{gra}{%
% 		\emph{Start of grammar `#1'}\par%
% 	}{%
% 		\par\emph{End of grammar}\par%
% 	}{%
% 		\subsection{#1}%
% 	}{%
% 		\par See some section in the text.
% 	}%
% }{%
% 	\@nameuse{endcollect*}%
% }
% \end{verbatim}
%
% \noindent This will create a |grammarpart| environment with one
% parameter, the part of the grammar defined. (i.e. `Identifiers') At
% the start of the environment the text `\emph{Start of grammar
% \meta{part}}' will be typeset, followed by the grammar (the contents
% of the environment), and finally a line `\emph{End of grammar}' will
% be added.
%
% When including the collected grammar parts using
% |\includecollection| each grammar part will be prefixed by a
% subsection heading bearing the grammar part as its title and
% suffixed by `See some section in the text.' (this could be extended
% to include a backref using |\ref|\meta{label})
%
% \paragraph{Note} If you use any of the environments inside a custom
% environment to afterwards collect text with this custom environment
% (as in the example of the |grammarpart| environment above), you
% \emph{must not} begin and end it using |\begin| and |\end|, or the
% environment contents will not be collected correctly. You must use
% |\@nameuse{|\meta{environment}|}| and
% |\@nameuse{end|\meta{environment}|}| as in the example above, or, if
% the environment is not starred, |\|\meta{environment} and
% |\end|\meta{environment}.
%
% \paragraph{Note} You still can use the environments of this package
% inside other environments (i.e. a |itemize| environment) without any
% problems.
%
%
%
%
% \section{Notes}
%
% \begin{itemize}
%	\item	If you use any of the environments inside your own
%			environment, note that the end of the environment is
%			executed, but not included in the collected text! (that's
%			why the the environments |collect*| and |collectinmacro|
%			have two parameters for including text before and after
%			the environment, \meta{before} and \meta{after}).
%
% \end{itemize}
%
%
%
%
% \StopEventually{}
%
% \section{Implementation}
%
% \subsection{Main environments and macros}
%
% \subsubsection{\texttt{collect}}
%
% \begin{macro}{\definecollection}
%
% Usage: |\definecollection| \marg{name}
%
% Defines a collection \meta{name}.
%
% This macro allocates a |\newwrite| |CE@@|\meta{name}|@out| and
% defines a |\newif| |CE@@|\meta{name}|open|.
%
%    \begin{macrocode}
\newcommand{\definecollection}[1]{%
	\@ifundefined{CE@@#1@out}{%
		\expandafter\newwrite\csname CE@@#1@out\endcsname%
		\expandafter\newif\csname ifCE@@#1@open\endcsname%
		\csname CE@@#1@openfalse\endcsname%
	}{%
		\PackageError{collect}{Collection `#1' has already %
			been defined}{\@ehc}%
	}%
}
%    \end{macrocode}
% \end{macro}
%
%
% \begin{environment}{collect*}
%
% Usage: |\begin{collect*}| \marg{name} \marg{before}
% \marg{after} \marg{beforecol} \marg{aftercol} \ldots\
% |\end{collect*}|.
%
%    \begin{macrocode}
\newenvironment{collect*}[5]{%
	\global\toks@{}%
	\def\CE@file{#1}%
	\def\CE@preenv{#2}%
	\def\CE@postenv{#3}%
	\def\CE@prefileenv{#4}%
	\def\CE@postfileenv{#5}%
	\CE@get@env@body@start%
}{%
%    \end{macrocode}
%
% What are we doing? We make sure the collection is open, then we save
% \meta{beforecol}, \meta{before}, the collected environment contents,
% \meta{after} and \meta{aftercol} in the collection file.
%
%    \begin{macrocode}
	\CE@ensure@opened{\CE@file}%
	\edef\@tempa{\csname CE@@\CE@file @out\endcsname}%
	\immediate\write\@tempa{\CE@meaning\CE@prefileenv}%
	\immediate\write\@tempa{\CE@meaning\CE@preenv}%
	\immediate\write\@tempa{\the\toks@}%
	\immediate\write\@tempa{\CE@meaning\CE@postenv}%
	\immediate\write\@tempa{\CE@meaning\CE@postfileenv}%
%    \end{macrocode}
%
% Now we repeat the same thing, but with the temporary file and
% writing only \meta{before}, the collected environment contents and
% \meta{after}. Then we include the temporary file.
%
% Why so complicated, as we have the contents of the environment in
% |\toks@|? Because the catcodes might not be correct, i.e. if |\verb|
% is used inside the environment. So we have to read the environment
% contents again, which without e\TeX\ is only possible using a
% temporary file.
%
%    \begin{macrocode}
	\immediate\openout\CE@tmp@out=\jobname.tmp%
	\immediate\write\CE@tmp@out{\CE@meaning\CE@preenv}%
	\immediate\write\CE@tmp@out{\the\toks@}%
	\immediate\write\CE@tmp@out{\CE@meaning\CE@postenv}%
	\immediate\closeout\CE@tmp@out%
	\@input{\jobname.tmp}%
	\par%
}
%    \end{macrocode}
% \end{environment}
%
%
% \begin{environment}{collect}
%
% Usage: |\begin{collect}| \marg{name} \marg{beforecol}
% \marg{aftercol} \ldots\ |\end{collect}|.
%
%    \begin{macrocode}
\newenvironment{collect}[3]{%
	\global\toks@{}%
	\def\CE@file{#1}%
	\def\CE@prefileenv{#2}%
	\def\CE@postfileenv{#3}%
	\CE@get@env@body@start%
}{%
%    \end{macrocode}
%
% As this environment, contrary to |collect*|, does not typeset its
% contents, we only write to the collection file:
%
%    \begin{macrocode}
	\CE@ensure@opened{\CE@file}%
	\edef\@tempa{\csname CE@@\CE@file @out\endcsname}%
	\immediate\write\@tempa{\CE@meaning\CE@prefileenv}%
	\immediate\write\@tempa{\the\toks@}%
	\immediate\write\@tempa{\CE@meaning\CE@postfileenv}%
}
%    \end{macrocode}
% \end{environment}
%
%
% \begin{macro}{\includecollection}
%
% Usage: |\includecollection| \marg{name}.
%
% Includes the collection \meta{name} which must have been defined
% previously using |\definecollection|. Afterwards, the collection is
% cleared.
%
%    \begin{macrocode}
\newcommand{\includecollection}[1]{%
	\CE@ensure@closed{#1}%
	\@input{\jobname.#1}%
}
%    \end{macrocode}
% \end{macro}
%
%
%
% \subsubsection{Other environments}
%
% \begin{environment}{collectinmacro}
%
% Usage: |\begin{collectinmacro}| \marg{macro} \marg{before}
% \marg{after} \ldots\ |\end{collectinmacro}|.
%
% Collects the contents of an environment inside a macro.
%
%    \begin{macrocode}
\newenvironment{collectinmacro}[3]{%
	\def\CE@destmacro{#1}%
	\def\CE@postenv{#3}%
%    \end{macrocode}
%
% We initialize the result with \meta{before}:
%
%    \begin{macrocode}
	\toks@{#2}%
	\CE@get@env@body@start%
}{%
%    \end{macrocode}
%
% We add \meta{after}:
%
%    \begin{macrocode}
	\toks@\expandafter\expandafter\expandafter{%
		\expandafter\the\expandafter\toks@\CE@postenv}%
%    \end{macrocode}
%
% Finally we globally define \meta{macro} to contain the collected
% contents:
%
%    \begin{macrocode}
	\expandafter\expandafter\expandafter%
		\gdef\expandafter\CE@destmacro\expandafter{\the\toks@}%
	\toks@{}%
}
%    \end{macrocode}
% \end{environment}
%
%
%
%
% \subsection{Internal environments and macros}
%
% We allocate a new |\newwrite| for the processing of a temporary
% file:
%
%    \begin{macrocode}
\newwrite\CE@tmp@out
%    \end{macrocode}
%
%
% \begin{macro}{\CE@get@env@body@start}
%
% Starts the collecting of the contents of an environment. (the
% environment starts immediately after the macro)
%
%    \begin{macrocode}
\def\CE@get@env@body@start{%
	\let\@tempa\CE@get@env@body%
%    \end{macrocode}
%
% We may have to gobble leading spaces, therefore we check the first
% character in the environment:
%
%    \begin{macrocode}
	\futurelet\@tempb\CE@get@env@body@start@%
}
%    \end{macrocode}
% \end{macro}
%
%
% \begin{macro}{\CE@get@env@body@start@}
%
% Support macro for |\CE@get@env@body@start|. Checks if the next token
% is a space, then calls |\CE@get@env@body@start@@|. Otherwise, the
% collecting of the environment contents is started.
%
%    \begin{macrocode}
\def\CE@get@env@body@start@{%
%    \end{macrocode}
% 
% |\@sptoken| contains a single space and is defined in
% \textsf{ltdefns.dtx}:
% 
%    \begin{macrocode}
	\ifx\@tempb\@sptoken%
		\expandafter\CE@get@env@body@start@@%
	\else%
		\expandafter\CE@get@env@body%
	\fi%
}
%    \end{macrocode}
% \end{macro}
%
%
% \begin{macro}{\CE@get@env@body@start@@}
%
% Support macro for |\CE@get@env@body@start@|. Gobbles up any space
% following the macro, then start the collecting of the environment
% contents using |\CE@get@env@body|.
%
%    \begin{macrocode}
\def\CE@get@env@body@start@@{%
	\afterassignment\CE@get@env@body%
	\let\@tempb= %
}
%    \end{macrocode}
% \end{macro}
%
%
% \begin{macro}{\CE@get@env@body}
%
% Usage: |\CE@get@env@body| \marg{body} |\end| \marg{envname}.
%
% To ensure proper initialization, this macro should not be called
% directly; instead |\CE@get@env@body@start| should be called.
%
% First we change the catcode of |Q| to 3 (math switch) in order to
% have a really unique character for parameter matching later
% on.\footnote{Taken from `Around the bend 15'} We do this in a group
% in order to easily restore the catcode later on and make all macro
% definitions global:
%
%    \begin{macrocode}
\bgroup
\catcode`\Q=3
%    \end{macrocode}
%
% Now we begin the macro:
%
%    \begin{macrocode}
\long\gdef\CE@get@env@body#1\end#2{%
%    \end{macrocode}
%
% Right at the beginning of the macro, we are at an |\end| (and |#1|
% contains the contents prior to it). We check if it ends the current
% environment (|#2| contains the name of the environment ended):
%
%    \begin{macrocode}
	\def\@tempb{#2}%
	\ifx\@tempb\@currenvir%
%    \end{macrocode}
%
% \paragraph{Yes, it ends the current environment.} We add the
% contents to |\toks@| and are done. But we do not add the contents
% directly as there may be a trailing space left (multiple spaces have
% been collapsed into one space by \TeX). So we call
% |\CE@get@env@body@| using delimited parameters (note that |Q| has
% catcode 3, therefore a normal |Q| is not matched).
%
% What exactly is going on here? Suppose we have the text
%
% \begin{verbatim}
% Hello World_
% \end{verbatim}
% %
% \noindent (the |_| denotes the trailing space). Then
% |\CE@get@env@body@| is called like this:
%
% \begin{verbatim}
% \CE@get@env@body@Hello World Q Q
% \end{verbatim}
% %
% \noindent |\CE@get@env@body@| matches the parameters like this: |#1|
% is `Hello World'; leaving |_Q| (added by the call from
% |\CE@get@env@body|) in the input. |\CE@get@env@body@| then calls
% |\CE@get@env@body@@| using |#1| and |Q|, resulting in the call:
%
% \begin{verbatim}
% \CE@get@env@body@@Hello WorldQ Q
% \end{verbatim}
% %
% \noindent |\CE@get@env@body@@| matches its parameter like this: |#1|
% is `Hello World' again, and |#2| is the second |Q|.
%
% But what happened to the space between the two |Q|s? As |#2| is not
% a delimited parameter, \TeX\ skips spaces after matching the first
% |Q| until it reaches the second |Q|, thus gobbling up the space
% inbetween.\footnote{c.f. \TeX book chapter 20.}
%
% \addvspace{\baselineskip}\noindent Now suppose we have the text
% \begin{verbatim}
% Hello World
% \end{verbatim}
% %
% \noindent without any trailing space. Then |\CE@get@env@body@| is
% called like this:
%
% \begin{verbatim}
% \CE@get@env@body@Hello WorldQ Q
% \end{verbatim}
% %
% \noindent |\CE@get@env@body@| matches its parameters like this: |#1|
% is `Hello WorldQ'. |\CE@get@env@body@| then calls
% |\CE@get@env@body@@| using |#1| and |Q|, resulting in the call:
%
% \begin{verbatim}
% \CE@get@env@body@@Hello WorldQQ
% \end{verbatim}
% %
% \noindent |\CE@get@env@body@@| matches its parameter like this: |#1|
% is `Hello World' without the trailing `Q', and |#2| is the second
% |Q|. The only difference to the situation described above is the
% missing space between the two |Q|s.
%
% Remember that the `Q' is always `Q' with catcode 3, thus no `Q' in
% the environment contents is matched.
%
%    \begin{macrocode}
		\CE@get@env@body@#1Q Q%
		\def\@tempa{\end{#2}}%
	\else
%    \end{macrocode}
%
% \paragraph{No, it ends another environment.} We add the contents as
% well as the |\end|\meta{environment} to |\toks@|. Then we continue
% collecting:
%
%    \begin{macrocode}
		\toks@\expandafter{\the\toks@#1\end{#2}}%
	\fi%
	\@tempa%
}
%    \end{macrocode}
% \end{macro}
%
%
% \begin{macro}{\CE@get@env@body@}
%
% Support macro for |\CE@get@env@body|. See |\CE@get@env@body| for
% explanations.
%
%    \begin{macrocode}
\long\gdef\CE@get@env@body@#1 Q{%
	\CE@get@env@body@@#1Q%
}
%    \end{macrocode}
% \end{macro}
%

% \begin{macro}{\CE@get@env@body@@}
%
% Support macro for |\CE@get@env@body@|. See |\CE@get@env@body| for
% explanations.
%
%    \begin{macrocode}
\long\gdef\CE@get@env@body@@#1Q#2{%
	\toks@\expandafter{\the\toks@#1}%
}
%    \end{macrocode}
% \end{macro}
%
% Finally we end the group, thus restoring the catcode of |Q|:
%
%    \begin{macrocode}
\egroup
%    \end{macrocode}
%
%
% \begin{macro}{\CE@meaning}
%
% Usage: |\CE@meaning| \marg{macro}.
%
% Expands to the meaning of \meta{macro}.
%
%    \begin{macrocode}
\long\def\CE@meaning#1{%
	\expandafter\strip@prefix\meaning#1%
}
%    \end{macrocode}
% \end{macro}
%
%
% \begin{macro}{\CE@ensure@opened}
%
% Usage: |\CE@ensure@opened| \marg{name}.
%
% Ensures that the file for collection \meta{name} is opened.
%
%    \begin{macrocode}
\def\CE@ensure@opened#1{%
	\@ifundefined{ifCE@@#1@open}{%
		\PackageError{collect}{Collection `#1' has not been defined}{\@ehc}%
	}{%
		\csname ifCE@@#1@open\endcsname\else%
			\expandafter\immediate\expandafter\openout%
				\csname CE@@#1@out\endcsname=\jobname.#1%
			\expandafter\global\csname CE@@#1@opentrue\endcsname%
		\fi%
	}%
}
%    \end{macrocode}
% \end{macro}
%
%
% \begin{macro}{\CE@ensure@closed}
%
% Usage: |\CE@ensure@closed| \marg{name}.
%
% Ensures that the file for collection \meta{name} is closed.
%
%    \begin{macrocode}
\def\CE@ensure@closed#1{%
	\@ifundefined{ifCE@@#1@open}{%
		\PackageError{collect}{Collection `#1' has not been defined}{\@ehc}%
	}{%
		\csname ifCE@@#1@open\endcsname%
			\expandafter\immediate\expandafter\closeout%
				\csname CE@@#1@out\endcsname%
			\expandafter\global\csname CE@@#1@openfalse\endcsname%
		\fi%
	}%
}
%    \end{macrocode}
% \end{macro}
%
%
% \Finale
% \PrintChanges
% \PrintIndex
\endinput