summaryrefslogtreecommitdiff
path: root/texmf/source/latex/revtex
diff options
context:
space:
mode:
authorottjk <joshott16@gmail.com>2023-12-31 15:01:25 -0500
committerottjk <joshott16@gmail.com>2023-12-31 15:01:25 -0500
commit99baacb1f33acaea45b00e6661c29302b01562a6 (patch)
treee0d6d6b87436ee15e745a89bcaf58f6daf6101e4 /texmf/source/latex/revtex
parentfa8aa006e1628b47a2b34b0b800f71bd3beece2d (diff)
downloaddotfiles-99baacb1f33acaea45b00e6661c29302b01562a6.tar.gz
dotfiles-99baacb1f33acaea45b00e6661c29302b01562a6.zip
sway and tex
Diffstat (limited to 'texmf/source/latex/revtex')
-rw-r--r--texmf/source/latex/revtex/aip4-2.dtx4233
-rw-r--r--texmf/source/latex/revtex/ltxdocext.dtx1259
-rw-r--r--texmf/source/latex/revtex/ltxfront.dtx3763
-rw-r--r--texmf/source/latex/revtex/ltxgrid.dtx6849
-rw-r--r--texmf/source/latex/revtex/ltxutil.dtx4449
-rw-r--r--texmf/source/latex/revtex/revtex-bst.dtx3873
-rw-r--r--texmf/source/latex/revtex/revtex4-2.dtx8383
7 files changed, 32809 insertions, 0 deletions
diff --git a/texmf/source/latex/revtex/aip4-2.dtx b/texmf/source/latex/revtex/aip4-2.dtx
new file mode 100644
index 0000000..2c40254
--- /dev/null
+++ b/texmf/source/latex/revtex/aip4-2.dtx
@@ -0,0 +1,4233 @@
+% \iffalse meta-comment
+% aip4-2.dtx: package to change page grid, MVL.
+% Copyright (c) 2019 American Institute of Physics
+% mailto:tex@aip.org
+%
+% Disclaimer
+% This file is distributed WITHOUT ANY WARRANTY;
+% without even the implied warranty of
+% MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
+% License
+% You may distribute this file under the conditions of the
+% LaTeX Project Public License 1.3c or later
+% (http://www.latex-project.org/lppl.txt).
+% ReadMe
+% For the documentation and more detailed instructions for
+% installation, typeset this document with \LaTeX.
+% Maintenance Status
+% This work has the LPPL maintenance status "maintained";
+% Current Maintainer of this work is Arthur Ogawa
+% changes for version 4.2d and 4.2e by Phelype Oleinik.
+%
+% Version (4.2e)
+% Modified by Aptara
+%
+% This work consists of the main source file aip4-2.dtx
+% and the derived files
+% aip.rtx, aip.pdf, aip.ins, aip.drv.
+% Distribution:
+% CTAN:macros/latex/contrib/aip/
+%
+% Unpacking:
+% tex aip4-2.dtx
+%
+% Documentation:
+% latex aip4-2.dtx; ...
+%
+% Programm calls to get the documentation (example):
+% pdflatex aip4-2.dtx
+% makeindex -s gind.ist aip
+% makeindex -s gglo.ist -o aip.gls aip.glo
+% pdflatex aip4-2.dtx
+% makeindex -s gind.ist aip
+% pdflatex aip4-2.dtx
+%
+% Installation:
+% TDS:bibtex/bst/aip/
+% TDS:doc/latex/aip/
+% TDS:source/latex/aip/
+% TDS:tex/latex/aip/
+%
+% Thanks, Heiko!
+% This method of letting a single .dtx file serve as both
+% documentation (via latex) and installer (via tex) follows
+% the example of Heiko Oberdiek. Thanks!
+%<*ignore>
+\begingroup
+ \def\x{LaTeX2e}%
+\expandafter\endgroup
+\ifcase
+ 0\expandafter\ifx\csname processbatchFile\endcsname\relax\else1\fi\ifx\fmtname\x\else 1\fi
+ \relax
+\else
+ \csname fi\endcsname
+%</ignore>
+%<*install>
+%% This file will generate documentation and runtime files
+%% from aip4-2.dtx when run through LaTeX or TeX.
+%% This file requires docstrip version 2.4 or higher available from
+%% ftp://ctan.tug.org/tex-archive/macros/latex/unpacked/docstrip.tex
+\input docstrip
+\preamble
+
+This is a generated file;
+altering it directly is inadvisable;
+instead, modify the original source file.
+See the URL in the file README-AIP.
+
+Copyright (c) 2019--2020 American Institute of Physics.
+mailto:tex@aip.org
+
+Maintained by Arthur Ogawa (mailto:arthur_ogawa at sbcglobal.net)
+under contract to American Institute of Physics
+
+Version (4.2c)
+Modified by Aptara
+under contract to American Institute of Physics
+
+Version (4.2d,4.2e)
+Modified by Phelype Oleinik for the American Physical Society (mailto:phelype.oleinik at latex-project.org)
+
+License
+ You may distribute this file under the conditions of the
+ LaTeX Project Public License 1.3c or later
+ (http://www.latex-project.org/lppl.txt).
+
+ This file is distributed WITHOUT ANY WARRANTY;
+ without even the implied warranty of MERCHANTABILITY
+ or FITNESS FOR A PARTICULAR PURPOSE.
+
+\endpreamble
+\askforoverwritefalse
+\keepsilent
+ \generate{%
+ %{ignore}
+% \file{aip.ins}{\from{aip4-2.dtx}{install}}%
+% \file{aip.drv}{\from{aip4-2.dtx}{driver}}%
+% \usedir{tex/latex/aip}%
+ \file{aip4-2.rtx}{\from{aip4-2.dtx}{package}}%
+ \file{aapm4-2.rtx}{\from{aip4-2.dtx}{aapm}}%
+ \file{sor4-2.rtx}{\from{aip4-2.dtx}{sor}}%
+ }%
+\ifToplevel{
+\Msg{***********************************************************}
+\Msg{*}
+\Msg{* To finish the installation, please move}
+\Msg{* aip.rtx}
+\Msg{* into a directory searched by TeX;}
+\Msg{* in a TDS-compliant installation:}
+\Msg{* texmf/tex/macros/latex/aip/.}
+\Msg{*}
+\Msg{* To produce the documentation,
+ run aip4-2.dtx through LaTeX.}
+\Msg{*}
+\Msg{* Happy TeXing}
+\Msg{***********************************************************}
+}
+\endbatchfile
+%</install>
+%<*ignore>
+\fi
+%</ignore>
+% \fi
+%
+% \GetFileInfo{aip4-2.dtx}
+%% \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 \~}
+%%
+%
+% \StopEventually{}
+% \iffalse ltxdoc klootch
+%<*package>
+%%% @LaTeX-file{
+%%% filename = "aip4-2.dtx",
+%%% version = "4.2e",
+%%% date = "2020/10/03",
+%%% author = "Arthur Ogawa (mailto:arthur_ogawa at sbcglobal.net),
+%%% Phelype Oleinik (mailto:phelype.oleinik at latex-project.org),
+%%% commissioned by the American Institute of Physics.
+%%% ",
+%%% copyright = "Copyright (C) 2019--2020 American Institute of Physics,
+%%% distributed under the terms of the
+%%% LaTeX Project Public License 1.3c, see
+%%% ftp://ctan.tug.org/macros/latex/base/lppl.txt
+%%% ",
+%%% address = "AIP Journal Program
+%%% American Institute of Physics,
+%%% Suite 1NO1, 2 Huntington Quadrangle,
+%%% Melville, NY 11747 USA",
+%%% telephone = "",
+%%% FAX = "",
+%%% email = "mailto colon tex at aip.org",
+%%% codetable = "ISO/ASCII",
+%%% keywords = "latex, page grid, main vertical list",
+%%% supported = "yes",
+%%% abstract = "aip substyle for REVTeX",
+%%% }
+%</package>
+% \fi
+%
+% \iffalse ltxdoc klootch
+% The following references the \file{README-AIP} file,
+% which contains basic information about this package.
+% The contents of this file are generated when
+% you typeset the programmer's documentation.
+% Search on "{filecontents*}{README-AIP}" to locate it.
+% \fi\input{README-AIP}%
+%
+% \subsection{Bill of Materials}
+%
+% Following is a list of the files in this distribution arranged
+% according to provenance.
+%
+% \subsubsection{Primary Source}%
+% One single file generates all.
+%\begin{verbatim}
+%aip4-2.dtx
+%\end{verbatim}
+%
+% \subsubsection{Generated by \texttt{latex aip4-2.dtx}}%
+% Typesetting the source file under \LaTeX\
+% generates the readme and the installer.
+%\begin{verbatim}
+%README aip.ins
+%\end{verbatim}
+%
+% \subsubsection{Generated by \texttt{tex aip4-2.dtx}}%
+% Typesetting the installer generates
+% the package files.
+%\begin{verbatim}
+%aip.rtx
+%\end{verbatim}
+%
+% \subsubsection{Documentation}%
+% The following are the online documentation:
+% \begin{verbatim}
+%aip.pdf
+% \end{verbatim}
+%
+% \subsubsection{Auxiliary}%
+% The following are auxiliary files generated
+% in the course of running \LaTeX:
+% \begin{verbatim}
+%aip.aux aip.idx aip.ind aip.log aip.toc
+% \end{verbatim}
+%
+% \section{Code common to all modules}%
+%
+% The following may look a bit klootchy, but we
+% want to require only one place in this file
+% where the version number is stated,
+% and we also want to ensure that the version
+% number is embedded into every generated file.
+%
+% Now we declare that
+% these files can only be used with \LaTeXe.
+% An appropriate message is displayed if
+% a different \TeX{} format is used.
+% \begin{macrocode}
+%<*driver|package|aapm|sor>
+\NeedsTeXFormat{LaTeX2e}[1996/12/01]%
+%</driver|package|aapm|sor>
+% \end{macrocode}
+% As desired, the following modules all
+% take common version information:
+% \begin{macrocode}
+%<package>\ProvidesFile{aip4-2.rtx}%
+%<aapm>\ProvidesFile{aapm4-2.rtx}%
+%<sor>\ProvidesFile{sor4-2.rtx}%
+%<*driver>
+\expandafter\ProvidesFile\expandafter{\jobname.dtx}%
+%</driver>
+% \end{macrocode}
+%
+% The following line contains, for once and for all,
+% the version and date information.
+% By various means, this information is reproduced
+% consistently in all generated files and in the
+% typeset documentation.
+% \begin{macrocode}
+%<*driver|package>
+%<version>
+ [2020/10/03 4.2e AIP substyle for REVTeX]% \fileversion
+%</driver|package>
+% \end{macrocode}
+%
+%
+% \section{The driver module \texttt{driver}}
+%
+% This module, consisting of the present section,
+% typesets the programmer's documentation,
+% generating the \file{.ins} installer and \file{README-AIP} as required.
+%
+% Because the only uncommented-out lines of code at the beginning of
+% this file constitute the \file{driver} module itself,
+% we can simply typeset the \file{.dtx} file directly,
+% and there is thus rarely any need to
+% generate the ``driver'' {\sc docstrip} module.
+% Module delimiters are nonetheless required so that
+% this code does not find its way into the other modules.
+%
+% The \enve{document} command concludes the typesetting run.
+%
+% \begin{macrocode}
+%<*driver>
+% \end{macrocode}
+%
+% \subsection{The Preamble}
+% The programmers documentation is formatted
+% with the \classname{ltxdoc} class with local customizations,
+% and with the usual code line indexing.
+% \begin{macrocode}
+\documentclass{ltxdoc}
+\RequirePackage{ltxdocext}%
+\let\url\undefined
+\RequirePackage[colorlinks=true,linkcolor=blue]{hyperref}%
+\gdef\lquote{`}\gdef\rquote{'}%
+\CodelineIndex\EnableCrossrefs % makeindex -s gind.ist aip
+\RecordChanges % makeindex -s gglo.ist -o aip.gls aip.glo
+% \end{macrocode}
+%
+% \subsubsection{Docstrip and info directives}
+% We use so many {\sc docstrip} modules that we set the
+% \texttt{StandardModuleDepth} counter to 1.
+% \begin{macrocode}
+\setcounter{StandardModuleDepth}{1}
+% \end{macrocode}
+% The following command retrieves the date and version information
+% from this file.
+% \begin{macrocode}
+\expandafter\GetFileInfo\expandafter{\jobname.dtx}%
+% \end{macrocode}
+%
+% \subsection{The ``Read Me'' File}
+% As promised above, here is the contents of the
+% ``Read Me'' file. That file serves a double purpose,
+% since it also constitutes the beginining of the
+% programmer's documentation. What better thing, after
+% all, to have appear at the beginning of the
+% typeset documentation?
+%
+% A good discussion of how to write a ReadMe file can be found in
+% Engst, Tonya, ``Writing a ReadMe File? Read This''
+% \emph{MacTech} October 1998, p. 58.
+%
+% Note the appearance of the
+% \cmd\StopEventually\ command, which marks the
+% dividing line between the user documentation
+% and the programmer documentation.
+%
+% The usual user will not be asked to
+% do a full build, not to speak
+% of the bootstrap.
+% Instructions for carrying these processes
+% begin the programmer's manual.
+%
+%\changes{4.1b}{2008/08/06}{Beta candidate b}
+% \begin{macrocode}
+\begin{filecontents*}{README-AIP}
+\title{%
+ The \classname{aip} substyle for REV\TeX%
+ \protect\thanks{%
+ \fileversion\ \copyright 2009 American Institute of Physics
+ % For version number and date,
+ % search on "\fileversion" in the .dtx file,
+ % or see the end of the README file.
+ }%
+}%
+
+\author{%
+Arthur Ogawa\thanks{mailto:arthur\_ogawa at sbcglobal.net}%
+}%
+\date{Version \fileversion, dated \filedate}%
+\maketitle
+
+This file embodies the implementation of the
+AIP journal substyles for APS's \revtex\ 4.2 document class
+for electronic submissions to AIP journals.
+
+The distribution point for this work is
+\url{http://publishing.aip.org/authors/preparing-your-manuscript}.
+The archive contains ready-to-Install files, documentation, and full source.
+
+This package is also distributed via CTAN:
+\url{http://www.ctan.org/pub/tex-archive/macros/latex/contrib/aip}, and as a ready-to-install image as
+\url{http://www.ctan.org/pub/tex-archive/install/macros/latex/contrib/aip.tds.zip}.
+
+The \classname{aip} package was commissioned by the American Institute of Physics
+and is distributed under the terms of the \LaTeX\ Project Public License,
+the same license under which all the portions of \LaTeX\ itself is distributed.
+Please see \url{http://ctan.tug.org/macros/latex/base/lppl.txt} for details.
+
+To use this document class, you must have a working
+\TeX\ installation equipped with \LaTeXe\
+and possibly pdftex and Adobe Acrobat Reader or equivalent.
+
+To install, retrieve the distribution,
+unpack it into a directory on the target computer,
+and move the file \file{aip.rtx}
+into a location in your filesystem where it will be found by \LaTeX.
+
+To use, read the user documentation \file{auguide.pdf}.
+
+\tableofcontents
+
+\section{Overview}%
+\revtex\ is a \LaTeXe\ document class, somewhat like a hybrid of
+the standard \LaTeX\ \classname{book} and \classname{article} classes.
+
+This document class implements the AIP society and journal substyles:
+the journals represent a set of mutually exclusive class options that, in this case, allow the
+document class to address multiple journals within the AIP family.
+This society is obtained with class option \classoption{aip}.
+
+\section{Processing Instructions}
+
+The package file \file{aip.rtx}
+is generated from this file, \file{aip4-2.dtx},
+using the {\sc docstrip} facility of \LaTeX
+via |tex aip4-2.dtx|.
+The typeset documentation that you are now reading is generated from
+the same file by typesetting it with \LaTeX\ or pdftex
+via |latex aip4-2.dtx| or |pdflatex aip4-2.dtx|.
+
+\subsection{Build Instructions}
+
+You may bootstrap this suite of files solely from \file{aip4-2.dtx}.
+Prepare by installing \LaTeXe\ (and either tex or pdftex) on your computer,
+then carry out the following steps:
+\begin{enumerate}
+\item
+Within an otherwise empty directory,
+typeset \file{aip4-2.dtx} with \LaTeX\ or pdflatex;
+you will obtain the typeset documentation you are now reading,
+along with
+the installer \file{aip.ins},
+and the file \file{README-AIP}.
+
+Note: you will have to run \LaTeX\ twice, then \file{makeindex}, then
+\LaTeX\ again in order to obtain a valid index and table of contents.
+\item
+Now typeset \file{aip4-2.dtx} with \TeX (not \LaTeX),
+thereby generating the package file \file{aip.rtx}.
+\item
+Install \classname{aip.rtx}
+by moving it to a location
+in your filesystem where it will be found by \LaTeX,
+like \file{tex/latex/aip}.
+\item
+Install \classname{aip.pdf}
+by moving it to
+\file{doc/latex/aip}.
+\item
+Using the \file{.dbj} information herein,
+generate the needed \file{.bst} files and install them in
+\file{bibtex/bst/aip}.
+\end{enumerate}
+\end{filecontents*}
+% \end{macrocode}
+%
+% \subsection{The Document Body}
+%
+% Here is the document body, containing only a
+% \cmd\DocInput\ directive---referring to this very file.
+% This very cute self-reference is a common \classname{ltxdoc} idiom.
+% \begin{macrocode}
+\begin{document}%
+\newcommand\revtex{REV\TeX}%
+\expandafter\DocInput\expandafter{\jobname.dtx}%
+% ^^A\PrintChanges
+\end{document}
+% \end{macrocode}
+%
+% \begin{macrocode}
+%</driver>
+% \end{macrocode}
+%
+% \section{AIP: A set of journal-specifc extensions to \revtex}
+%
+% To create a journal substyle, you create
+% a \file{.rtx} file, in our case \file{aip.rtx}.
+% Within that file, you override procedures and parameter assignments as you see fit.
+% Ideally they will be generally applicable to all of that society's
+% journals (see the file \file{aps.rtx} for a realization of this scheme).
+% Also within that file, you include a
+% section of code for each journal, that for JMP looks like:
+% \begin{verbatim}
+% \@ifx{\@journal\journal@jmp}{%
+% <code specific to the JMP>
+% }{}%
+% \end{verbatim}
+%
+% \section{The \classoption{aip} class option: the \texttt{aip} module}
+% The file \file{aip.rtx} is read in by the \classname{revtex4} document class
+% if \cmd\@society\ has the value \classoption{aip}.
+%
+% Here, code specific to AIP journals appears.
+%
+% We first give some text entities (amounting to journal abbreviations),
+% then some AIP-specific initialisations,
+% then code for particular AIP journals.
+% In the latter case, the choice is
+% keyed off the macro \cmd\@journal.
+%
+% Note on \cmd\AtEndOfClass:
+% this file, like all journal substyles, is read in at \cmd\AtEndOfClass\ time,
+% so you should not use this command in this file.
+% Note for \LaTeX\ developers: It would be an improvement in \LaTeX\ to
+% \cmd\let\cmd\AtEndOfClass\ to something like \cmd\@firstofone.
+% This change would be effected in \cmd\@onefilewithoptions.
+%
+% \begin{macrocode}
+%<*package>
+% \end{macrocode}
+%
+% \subsection{Defend Against Forseeable Errors}%
+% Protect this file from being read in by anything but \revtex.
+% \begin{macrocode}
+\ifx\undefined\substyle@ext
+ \def\@tempa{%
+ \endinput
+ \GenericWarning{I must be read in by REVTeX! (Bailing out)}%
+ }%
+ \expandafter\else
+ \def\@tempa{}%
+ \expandafter\fi\@tempa
+ \class@info{RevTeX society AIP selected}%
+% \end{macrocode}
+%
+% \subsection{Substyle Options}%
+%
+% Here are the class options relating to the AIP:
+% \begin{macrocode}
+\DeclareOption{jcp}{\change@journal{jcp}}%
+\DeclareOption{pop}{\change@journal{pop}}%
+\DeclareOption{rsi}{\change@journal{rsi}}%
+\DeclareOption{jap}{\change@journal{jap}}%
+\DeclareOption{apl}{\change@journal{apl}}%
+\DeclareOption{apm}{\change@journal{apm}}%
+\DeclareOption{cha}{\change@journal{cha}}%
+\DeclareOption{pof}{\change@journal{pof}}%
+\DeclareOption{bmf}{\change@journal{bmf}}%
+\DeclareOption{rse}{\change@journal{rse}}%
+\DeclareOption{jmp}{\change@journal{jmp}}%
+\DeclareOption{adv}{\change@journal{adv}}%
+\DeclareOption{sd}{\change@journal{sd}}%
+\DeclareOption{jor}{\change@journal{jor}}%
+\DeclareOption{cp}{\change@journal{cp}}%
+% \end{macrocode}
+%
+% \subsection{Abbreviations}%
+% \changes{4.1p}{2010/02/24}{(AO, 584) Per MD, remove trailing space character from each journal abbreviation: it had caused an extraneous space in the .bbl}
+% The following macros constitute typing shortcuts for
+% certain journal names.
+% \begin{macrocode}
+\def\adv{AIP Advances}%
+\def\ao{Appl.\ Opt.}%
+\def\ap{Appl.\ Phys.}%
+\def\apl{Appl.\ Phys.\ Lett.}%
+\def\apm{Appl.\ Phys.\ Lett.\ Mater.}%
+\def\apj{Astrophys.\ J.}%
+\def\bell{Bell Syst.\ Tech.\ J.}%
+\def\bmf{Biomicrofluidics}%
+\def\cha{Chaos}%
+\def\jqe{IEEE J.\ Quantum Electron.}%
+\def\assp{IEEE Trans.\ Acoust.\ Speech Signal Process.}%
+\def\aprop{IEEE Trans.\ Antennas Propag.}%
+\def\mtt{IEEE Trans.\ Microwave Theory Tech.}%
+\def\iovs{Invest.\ Ophthalmol.\ Vis.\ Sci.}%
+\def\jcp{J.\ Chem.\ Phys.}%
+\def\jap{J.\ Appl.\ Phys.}%
+\def\jmp{J.\ Math.\ Phys.}%
+\def\jmo{J.\ Mod.\ Opt.}%
+\def\josa{J.\ Opt.\ Soc.\ Am.}%
+\def\josaa{J.\ Opt.\ Soc.\ Am.\ A}%
+\def\josab{J.\ Opt.\ Soc.\ Am.\ B}%
+\def\jpp{J.\ Phys.\ (Paris)}%
+\def\jpr{J.\ Phys.\ Chem.\ Ref.\ Data}%
+\def\ltp{Low.\ Temp.\ Phys.}%
+\def\nat{Nature (London)}%
+\def\oc{Opt.\ Commun.}%
+\def\ol{Opt.\ Lett.}%
+\def\pl{Phys.\ Lett.}%
+\def\pop{Phys.\ Plasmas}%
+\def\pof{Phys.\ Fluids}%
+\def\pra{Phys.\ Rev.\ A}%
+\def\prb{Phys.\ Rev.\ B}%
+\def\prc{Phys.\ Rev.\ C}%
+\def\prd{Phys.\ Rev.\ D}%
+\def\pre{Phys.\ Rev.\ E}%
+\def\prl{Phys.\ Rev.\ Lett.}%
+\def\rmp{Rev.\ Mod.\ Phys.}%
+\def\rsi{Rev.\ Sci.\ Instrum.}%
+\def\rse{J.\ Renewable Sustainable Energy}%
+\def\pspie{Proc.\ Soc.\ Photo-Opt.\ Instrum.\ Eng.}%
+\def\sjqe{Sov.\ J.\ Quantum Electron.}%
+\def\vr{Vision Res.}%
+\def\sd{Structural\ Dynamics}%
+\def\jor{J.\ Rheol.}%
+\def\cp{AIP\ Conference\ Proceedings}%
+% \end{macrocode}
+%
+% Option allowing for ``numerical author-year'' bibliography.
+% \begin{macrocode}
+\DeclareOption{author-numerical}{%
+ \@booleantrue\authoryear@sw
+ \@booleantrue\authornum@sw
+}%
+% \end{macrocode}
+%
+% Optionally allow article title in bibliography entries.
+% A new switch is defined here: \cmd\aip@jtitx@sw\ signifies that
+% the numbered style bibliography is to be prepared without article titles in @ARTICLE entries.
+% Otherwise, such titles are incorporated if present in the \file{.bib} entry.
+% By default, exclude the article titles.
+% \begin{macrocode}
+\DeclareOption{article-title}{%
+ \@booleanfalse\aip@jtitx@sw
+}%
+\@booleantrue \aip@jtitx@sw
+% \end{macrocode}
+%
+% Establish default options for this society.
+%
+% \begin{macrocode}
+\@booleanfalse\authoryear@sw
+\@booleanfalse\authornum@sw
+\@booleanfalse\onecolumn@sw
+\@booleanfalse\newonecolumn@sw
+% \end{macrocode}
+%
+% \begin{macro}{\@bibdataout@init}
+% \begin{macro}{\@bibdataout@aip}
+% \changes{4.1o}{2010/02/12}{(AO, 580) Control .bst at run time.}%
+% Procedure \cmd\@bibdataout@aip\ has the job of writing the control record into the
+% job's \cmd\jobname\filename{.bib} file, where it will adjust the options to \filename{.bst} processing.
+% It is installed into the initialization procedure \cmd\@bibdataout@init, and its meaning
+% is set by the society (AIP) and journal.
+%
+% \begin{macrocode}
+\appdef\@bibdataout@rev{\@bibdataout@aip}%
+\def\@bibdataout@aip{%
+ \immediate\write\@bibdataout{%
+% \end{macrocode}
+% An entry that controls processing of the \filename{.bst} file has entry type \texttt{@CONTROL}.
+% \begin{macrocode}
+ @CONTROL{%
+ aip41Control%
+% \end{macrocode}
+% Say whether we want long bibliography style (the default), or the abbreviated style.
+% \begin{macrocode}
+ \longbibliography@sw{\true@sw}{\aip@jtitx@sw{\false@sw}{\true@sw}}%
+ {%
+ ,pages="1",title="0"%
+ }{%
+ ,pages="0",title=""%
+ }%
+ }%
+ }%
+% \end{macrocode}
+% Place a \cmd\citation\ into the auxiliary file corresponding to this entry.
+% \begin{macrocode}
+ \if@filesw
+ \immediate\write\@auxout{\string\citation{aip41Control}}%
+ \fi
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% For AIP, the preprint style \classoption{preprint} is the default.
+% \begin{macrocode}
+\@booleantrue\preprintsty@sw
+% \end{macrocode}
+%
+% \begin{macro}{\showPACS@sw}
+% \begin{macro}{\showKEYS@sw}
+% If \cmd\showPACS@sw\ is true, print the PACS information in the title block,
+% otherwise not.
+% Similarly for \cmd\showKEYS@sw\ and the keywords.
+% \begin{macrocode}
+\@booleantrue\showPACS@sw
+\@booleantrue\showKEYS@sw
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\refname}
+%
+% In reprint style only, we eliminate the head above the bibliography.
+% \begin{macrocode}
+% TeXSupport
+\let\old@refname\refname
+\appdef\setup@hook{%
+ \preprintsty@sw{}{%
+ \let\refname\@empty
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \subsection{AIP Setup}%
+% Here we define the default procedures for AIP journals.
+% Individual AIP journals may override these definitions.
+%
+% \subsubsection{Running header and footer}%
+%
+% Page style for all AIP journals.
+% We assert our own page style only if nobody else has done so.
+% Users wishing to customize their documents will be able to invoke a \cmd\pagestyle\
+% command anywhere in the preamble.
+%FIXME: \cmd\preprintsty@sw\ is defined?
+% \begin{macrocode}
+\appdef\setup@hook{%
+ \preprintsty@sw{%
+ \ps@preprint
+ }{%
+ \ps@article
+ }%
+}%
+% \end{macrocode}
+%
+% The spc says: ``Running title must be provided on title page'',
+% and that journal classes 1b and 2 have no running title.
+% \begin{macrocode}
+\def\ps@preprint{%
+ \def\@oddhead{\@runningtitle\hfil}%
+ \def\@evenhead{\@runningtitle\hfil}%
+ \def\@oddfoot{\hfil\thepage\quad\checkindate\hfil}%
+ \def\@evenfoot{\hfil\thepage\quad\checkindate\hfil}%
+ \let\@mkboth\@gobbletwo
+ \let\sectionmark\@gobble
+ \let\subsectionmark\@gobble
+}%
+\def\ps@article{%
+ \def\@evenhead{\let\\\heading@cr\thepage\quad\checkindate\hfil\@runningtitle}%
+ \def\@oddhead{\let\\\heading@cr\@runningtitle\hfil\checkindate\quad\thepage}%
+ \def\@oddfoot{}%
+ \def\@evenfoot{}%
+ \let\@mkboth\@gobbletwo
+ \let\sectionmark\@gobble
+ \let\subsectionmark\@gobble
+}%
+\def\@runningtitle{\@shorttitle}%
+% \end{macrocode}
+%
+% \subsubsection{Title block}%
+%
+% The specifics of the title block.
+% Apply to all AIP journals; individual journals may override these settings.
+%
+% \begin{macro}{\titlepage}
+% \begin{macrocode}
+\renewenvironment{titlepage}{%
+ \let\wastwocol@sw\twocolumn@sw
+ \onecolumngrid
+ \newpage
+ \thispagestyle{titlepage}%
+ \c@page\z@% article sets this to one not zero???
+}{%
+ \wastwocol@sw{\twocolumngrid}{\newpage}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@fnsymbol}
+% On the title page, footnotes are lettered with right parenthesis: ``a)''.
+% \begin{macrocode}
+\let\@fnsymbol@latex\@fnsymbol
+\let\@fnsymbol\@alph
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\adjust@abstractwidth}
+% This procedure is used by many titlepage elements to indent on the left by a particular
+% amount. The author list uses \cmd\frontmatter@authorformat\ to accomplish the same.
+% \begin{macrocode}
+\def\adjust@abstractwidth{%
+ \parindent1em\relax
+ \advance\leftskip.5in\relax
+ \@totalleftmargin\leftskip
+ \@afterheading\@afterindentfalse
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@abstractheading}
+% AIP Journals all set the abstract head the same way, with no head.
+% \begin{macrocode}
+\def\frontmatter@abstractheading{}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@abstractfont}
+% All AIP journals set the abstract body the same way.
+% \begin{macrocode}
+\def\frontmatter@abstractfont{%
+%\small
+ \adjust@abstractwidth
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% All AIP journal preprints use separate titlepage and full-width abstract.
+% \begin{macrocode}
+\appdef\setup@hook{%
+ \preprintsty@sw{%
+ \@booleantrue\titlepage@sw
+% \def\frontmatter@affiliationfont{\it}%
+ \let\section\section@preprintsty
+% \let\@hangfrom@section\@hangfrom@section@preprintsty
+ \let\subsection\subsection@preprintsty
+ \let\subsubsection\subsubsection@preprintsty
+ }{}%
+}%
+% \end{macrocode}
+%
+% \begin{macro}{\frontmatter@authorformat}
+% All AIP journals set the author list the same.
+% The leading is 11.5 points, and
+% there is 11.5 points of extra space above the first author line
+% (which amounts to the same thing as 11.5 points extra below the title)
+% for a total of 23 points base-to-base.
+%
+% \begin{macrocode}
+\def\frontmatter@@indent{%
+ \skip@\@flushglue
+ \@flushglue\z@ plus.3\hsize\relax
+ \raggedright
+ \advance\leftskip.5in\relax
+ \@totalleftmargin\leftskip
+ \@flushglue\skip@
+}%
+\def\frontmatter@authorformat{%
+ \frontmatter@@indent
+ \sffamily
+%\advance\baselineskip\p@
+%\parskip11.5\p@\relax
+}%
+\renewcommand*\email[1][Electronic mail: ]{\begingroup\sanitize@url\@email{#1}}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@above@affilgroup}
+% The default amount of space above affiliation.
+% \begin{macrocode}
+\def\frontmatter@above@affilgroup{%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@above@affiliation}
+% \begin{macro}{\frontmatter@above@affiliation@script}
+% The default amount of space above affiliation.
+% \begin{macrocode}
+\def\frontmatter@above@affiliation@script{%
+ \frontmatter@@indent
+%\addvspace{3.5\p@}%
+}%
+\def\frontmatter@above@affiliation{%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@affiliationfont}
+% All AIP journals set the affiliation the same.
+% Like the author, but in italic.
+% \begin{macrocode}
+\def\frontmatter@affiliationfont{%
+ \frontmatter@@indent
+ \preprintsty@sw{}{\small}%
+ \it
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@collaboration@above}
+% \begin{macrocode}
+\def\frontmatter@collaboration@above{%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% Set up the default AIP style for title block authors and affiliations.
+%
+% \begin{macro}{\frontmatter@setup}
+% All AIP journals set the title page using the same font and size.
+% However, justification varies for the title block elements, so
+% we assert none here.
+% \begin{macrocode}
+\def\frontmatter@setup{%
+ \normalfont
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@title@above}
+% \begin{macro}{\frontmatter@title@format}
+% \begin{macro}{\frontmatter@title@below}
+% All AIP journals set the article title 12 point Times Roman, bold, flush left.
+%
+% \begin{macrocode}
+\def\frontmatter@title@above{\addvspace{6\p@}}%
+\def\frontmatter@title@format{%
+ \preprintsty@sw{}{\Large}%
+ \sffamily
+ \bfseries
+ \raggedright
+ \parskip\z@skip
+}%
+\def\frontmatter@title@below{\addvspace{3\p@}}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@makefnmark}
+% All AIP journals share this procedure for setting the titlepage footnote text:
+% superior lowercase letter, with right parenthesis.
+% \begin{macrocode}
+\def\@author@parskip{3\p@}%
+\@booleantrue\altaffilletter@sw
+\def\frontmatter@makefnmark{%
+ \@textsuperscript{%
+ \normalfont\@thefnmark%(
+ )%
+ }%
+}%
+\def\frontmatter@authorbelow{%
+ \addvspace{3\p@}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\affil@cutoff}
+% We override \cmd\affil@cutoff\ to enforce the rule that,
+% if a single affiliation appears on the title page, then
+% no affiliation superscript at all is produced.
+% \begin{macrocode}
+\let\affil@cutoff\tw@
+% \end{macrocode}
+% De awa.
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@RRAP@format}
+% All AIP journals use the same format for the ``Received, Revised, etc.'' block on the title page.
+%
+% Change note: 11.5 points b-b from author/affiliation down to date.
+% \begin{macrocode}
+\def\frontmatter@RRAP@format{%
+ \addvspace{5\p@}%
+ \small
+ \raggedright
+ \advance\leftskip.5in\relax
+ \@totalleftmargin\leftskip
+% \preprintsty@sw{}{\parskip.5ex\relax}%
+ \everypar{%
+ \hbox\bgroup(\@gobble@leavemode@uppercase%)
+ }%
+ \def\par{%
+ \@ifvmode{}{%(
+ \unskip)\egroup\@@par
+ }%
+ }%
+}%
+\def\punct@RRAP{;\egroup\ \hbox\bgroup}%
+\def\@gobble@leavemode@uppercase#1#2{\expandafter\MakeTextUppercase}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@PACS@format}
+%
+% \begin{macrocode}
+\def\frontmatter@PACS@format{%
+ \addvspace{11\p@}%
+% \footnotesize
+ \adjust@abstractwidth
+ \parskip\z@skip
+ \samepage
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@keys@format}
+%
+% \begin{macrocode}
+\def\frontmatter@keys@format{%
+% \footnotesize
+ \adjust@abstractwidth
+ \samepage
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\ps@titlepage}
+% Title page style. Currently empty except for preprint header,
+% which consists of all the \cmd\preprint\ arguments,
+% stacked flush right at the right margin.
+% \begin{macrocode}
+\def\ps@titlepage{%
+ \def\@oddhead{%
+ \@runningtitle
+ \hfill
+ \produce@preprints\@preprint
+ }%
+ \let\@evenhead\@oddhead
+ \def\@oddfoot{%
+ \hb@xt@\z@{\byrevtex\hss}%
+ \hfil
+ \preprintsty@sw{\thepage}{}%
+ \quad\checkindate
+ \hfil
+ }%
+ \let\@evenfoot\@oddfoot
+}%
+\def\byrevtex{\byrevtex@sw{Typeset by REV\TeX and AIP}{}}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\produce@preprints}
+% \begin{macrocode}
+\def\produce@preprints#1{%
+ \preprint@sw{%
+ \vtop to \z@{%
+ \def\baselinestretch{1}%
+ \small
+ \let\preprint\preprint@count
+ \count@\z@#1\@ifnum{\count@>\tw@}{%
+ \hbox{%
+ \let\preprint\preprint@hlist
+ #1\setbox\z@\lastbox
+ }%
+ }{%
+ \let\preprint\preprint@cr
+ \halign{\hfil##\cr#1\crcr}%
+ \par
+ \vss
+ }%
+ }%
+ }{}%
+}%
+\def\preprint@cr#1{#1\cr}%
+\def\preprint@count#1{\advance\count@\@ne}%
+\def\preprint@hlist#1{#1\hbox{, }}%
+% \end{macrocode}
+% \end{macro}
+%
+% \subsubsection{Lead-in paragraph}%
+%
+% The \env{Lead-in paragraph} environment formats the AIP ``lead paragraph''.
+% To avoid introducing new syntax, we take over the \env{quotation} environment
+% until the first \cmd\section\ (or other sectioning command) appears.
+%
+% \begin{macrocode}
+\newenvironment{Lead@inParagraph}{%
+ \par
+ \bfseries
+ \@afterheading\@afterindentfalse
+}{%
+ \par
+ \hb@xt@\hsize{\hfil\leaders\hrule\hfil\leaders\hrule\hfil\hfil}%
+}%
+% \end{macrocode}
+% At the beginning of the document temporarily change the meaning of
+% the \env{quotation} environment, restoring it once the first sectioning
+% command is given.
+% \begin{macrocode}
+\appdef\frontmatter@init{%
+ \let@environment{quotation@ltx}{quotation}%
+ \let@environment{quotation}{Lead@inParagraph}%
+}%
+\appdef\@startsection@hook{%
+ \let@environment{quotation}{quotation@ltx}%
+}%
+% \end{macrocode}
+%
+% \subsubsection{Stacked Heads}%
+% All AIP journals put a period (.), followed by quad space, after the section number.
+% Also, no hanging section number.
+% \begin{macrocode}
+\def\@seccntformat#1{\csname the#1\endcsname.\quad}%
+\def\@hang@from#1#2#3{#1#2#3}%
+% \end{macrocode}
+%
+% Note that in the following, we wish to set the section head uppercase, so we use
+% David Carlisle's \cmd\MakeTextUppercase. However, because this procedure effectively
+% parses its argument (looking for things to \emph{not} translate), it has to be invoked
+% in such a way that the argument of the \cmd\section\ command is passed to it as its
+% own argument.
+%
+% To accomplish this, we use the \cmd\@hangfrom@\ hook, which was developed for this
+% purpose.
+%
+% \begin{macrocode}
+\def\section{%
+ \@startsection
+ {section}%
+ {1}%
+ {\z@}%
+ {0.8cm \@plus1ex \@minus .2ex}%
+ {0.5cm}%
+ {%
+ \normalfont
+ \small
+ \sffamily
+ \bfseries
+ \raggedright
+ }%
+}%
+\def\@hangfrom@section#1#2#3{\@hangfrom{#1#2}\MakeTextUppercase{#3}}%
+\def\@hangfroms@section#1#2{#1\MakeTextUppercase{#2}}%
+% \end{macrocode}
+%
+% See, e.g., BMF\_044101\_1 (1) or BMF\_044103\_1 (1)
+% \begin{macrocode}
+\def\subsection{%
+ \@startsection
+ {subsection}%
+ {2}%
+ {\z@}%
+ {.8cm \@plus1ex \@minus .2ex}%
+ {.5cm}%
+ {%
+ \normalfont
+ \small
+ \sffamily
+ \bfseries
+ \raggedright
+ }%
+}%
+% \end{macrocode}
+%
+% See, e.g., BMF\_044104\_1 (1), JMP\_123520\_1 (48),
+% \begin{macrocode}
+\def\subsubsection{%
+ \@startsection
+ {subsubsection}%
+ {3}%
+ {\z@}%
+ {.8cm \@plus1ex \@minus .2ex}%
+ {.5cm}%
+ {%
+ \normalfont
+ \small
+ \sffamily
+ \bfseries
+ \itshape
+ \raggedright
+ }%
+}%
+% \end{macrocode}
+%
+% \subsubsection{Runin Heads}%
+% \begin{macrocode}
+\def\paragraph{%
+ \@startsection
+ {paragraph}%
+ {4}%
+ {\parindent}%
+ {\z@}%
+ {-1em}%
+ {\normalfont\normalsize\itshape}%
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\subparagraph{%
+ \@startsection
+ {subparagraph}%
+ {5}%
+ {\parindent}%
+ {3.25ex \@plus1ex \@minus .2ex}%
+ {-1em}%
+ {\normalfont\normalsize\bfseries}%
+}%
+% \end{macrocode}
+%
+% \begin{macro}{\section@preprintsty}
+% \begin{macro}{\subsection@preprintsty}
+% \begin{macro}{\subsubsection@preprintsty}
+% Here are the formatting procedures specific to the preprint style;
+% the only difference is that the heads are flush left instead of centered.
+%
+% \begin{macrocode}
+\def\section@preprintsty{%
+ \@startsection
+ {section}%
+ {1}%
+ {\z@}%
+ {0.8cm \@plus1ex \@minus .2ex}%
+ {0.5cm}%
+ {%
+ \normalfont
+ \bfseries
+ \raggedright
+ }%
+}%
+%\def\@hangfrom@section@preprintsty#1#2#3{\@hangfrom{#1#2}\MakeTextUppercase{#3}}%
+% \end{macrocode}
+% \begin{macrocode}
+\def\subsection@preprintsty{%
+ \@startsection
+ {subsection}%
+ {2}%
+ {\z@}%
+ {.8cm \@plus1ex \@minus .2ex}%
+ {.5cm}%
+ {%
+ \normalfont
+ \bfseries
+ \raggedright
+ }%
+}%
+% \end{macrocode}
+% \begin{macrocode}
+\def\subsubsection@preprintsty{%
+ \@startsection
+ {subsubsection}%
+ {3}%
+ {\z@}%
+ {.8cm \@plus1ex \@minus .2ex}%
+ {.5cm}%
+ {%
+ \normalfont
+% \small
+ \itshape\bfseries
+ \raggedright
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% For examples of
+% \env{theorem}, \env{proposition}, \env{lemma}, \env{remark}, \env{corollary}, \env{example},
+% and \env{proof} (with optional title),
+% using independent numbering for each class,
+% and with numbered and roman lists therein,
+% see JMP\_122901\_1 (48).
+%
+% For theorems, etc, numbered by section, (and with theorems in the appendix)
+% but equations numbered throughout,
+% see JMP\_123301\_1 (48).
+%
+% For a (roman) list in text, theorem, and proof,
+% see JMP\_123514\_1 (48).
+%
+% For numbered list in text,
+% and equations numbered by section,
+% see JMP\_123518\_1 (48).
+%
+% All AIP journals use frontmatter footnotes by default.
+% \begin{macrocode}
+\let\frontmatter@footnote@produce\frontmatter@footnote@produce@footnote
+% \end{macrocode}
+%
+%
+% \subsubsection{Table of Contents}%
+% The toc will itself make an entry in the toc,
+% but we temporarily turn off toc formatting for the duration.
+% \begin{macrocode}
+\def\@pnumwidth{1.55em}
+\def\@tocrmarg {2.55em}
+\def\@dotsep{2}
+\def\ltxu@dotsep{4.5pt}
+\setcounter{tocdepth}{3}
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\tableofcontents{%
+ \addtocontents{toc}{\string\tocdepth@munge}%
+ \print@toc{toc}%
+ \addtocontents{toc}{\string\tocdepth@restore}%
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\tocdepth@munge{%
+ \let\l@section@saved\l@section
+ \let\l@section\@gobble@tw@
+}%
+\def\@gobble@tw@#1#2{}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\tocdepth@restore{%
+ \let\l@section\l@section@saved
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\l@part#1#2{\addpenalty{\@secpenalty}%
+ \begingroup
+ \set@tocdim@pagenum{#2}%
+% \@tempdima 3em %
+ \parindent \z@
+ \rightskip\tocleft@pagenum plus 1fil\relax
+ \skip@\parfillskip\parfillskip\z@
+ \addvspace{2.25em plus\p@}%
+ \large \bf %
+ \leavevmode\ignorespaces#1\unskip\nobreak\hskip\skip@
+ \hb@xt@\rightskip{\hfil\unhbox\z@}\hskip-\rightskip\hskip\z@skip
+ \par
+ \nobreak %
+ \endgroup
+}%
+% \end{macrocode}
+%
+% \begin{macro}{\l@section}
+% Determine which TOC elements are automatically indented.
+% \begin{macrocode}
+\def\tocleft@{\z@}%
+\def\tocdim@min{5\p@}%
+\def\l@section{%
+ \l@@sections{}{section}% Implicit #3#4
+}%
+\def\l@f@section{%
+ \addpenalty{\@secpenalty}%
+ \addvspace{1.0em plus\p@}%
+ \bf
+}%
+\def\l@subsection{%
+ \l@@sections{section}{subsection}% Implicit #3#4
+}%
+\def\l@subsubsection{%
+ \l@@sections{subsection}{subsubsection}% Implicit #3#4
+}%
+\def\l@paragraph#1#2{}%
+\def\l@subparagraph#1#2{}%
+% \end{macrocode}
+% \end{macro}
+%
+% Activate the auto TOC processing.
+% \begin{macrocode}
+\let\toc@pre\toc@pre@auto
+\let\toc@post\toc@post@auto
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\listoffigures{\print@toc{lof}}%
+\def\l@figure{\@dottedtocline{1}{1.5em}{2.3em}}
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\listoftables{\print@toc{lot}}%
+\let\l@table\l@figure
+% \end{macrocode}
+%
+% \subsubsection{Default column bottom}%
+% All AIP journal styles have flush bottoms.
+%
+% \begin{macrocode}
+\@booleanfalse\raggedcolumn@sw
+% \end{macrocode}
+%
+%
+% \subsubsection{Table alignment style}%
+%
+% \begin{macro}{\tableft@skip@float}
+% \begin{macro}{\tabmid@skip@float}
+% \begin{macro}{\tabright@skip@float}
+% \begin{macro}{\array@row@pre@float}
+% \begin{macro}{\array@row@pst@float}
+% All AIP publications have the same table specification:
+% Scotch rules above and below, centered in column.
+% \begin{macrocode}
+\def\tableft@skip@float{\z@ plus\hsize}%
+\def\tabmid@skip@float{\@flushglue}%
+\def\tabright@skip@float{\z@ plus\hsize}%
+\def\array@row@pre@float{\hline\hline\noalign{\vskip\doublerulesep}}%
+\def\array@row@pst@float{\noalign{\vskip\doublerulesep}\hline\hline}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+%
+% \subsubsection{Footnote formatting}%
+% We customize the formatting of footnotes for all AIP journals.
+% \begin{macro}{\@makefntext}
+% \begin{macrocode}
+\def\@makefntext#1{%
+ \def\baselinestretch{1}%
+ \leftskip1em%
+ \parindent1em%
+ \noindent
+ \nobreak\hskip-\leftskip
+ \hb@xt@\leftskip{%
+ \hss\@makefnmark\ %
+ }%
+ #1%
+ \par
+}%
+% \end{macrocode}
+% \end{macro}
+%
+%
+% \subsubsection{Appendix}%
+% \begin{macro}{\appendix}
+% \begin{macro}{\@hangfrom@appendix}
+% \begin{macro}{\@hangfroms@appendix}
+% \begin{macro}{\@appendixcntformat}
+% \begin{macrocode}
+\prepdef\appendix{%
+ \par
+ \let\@hangfrom@section\@hangfrom@appendix
+%\let\@hangfroms@section\@hangfroms@appendix
+ \let\@sectioncntformat\@appendixcntformat
+}%
+\def\@hangfrom@appendix#1#2#3{%
+ #1%
+ \@if@empty{#2}{%
+ #3%
+ }{%
+ #2\@if@empty{#3}{}{:\ #3}%
+ }%
+}%
+\def\@hangfroms@appendix#1#2{%
+ #1#2%
+}%
+\def\@appendixcntformat#1{\appendixname\ \csname the#1\endcsname}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \subsubsection{Bibliography}%
+%
+% Customize the \revtex\ for the AIP society.
+% this task requires three components:
+% the \BibTeX\ \file{aipnum.bst} and \file{aipauth.bst} style files,
+% customizing code for \file{natbib}, and
+% customizations of the \env{thebibliography} environment.
+%
+% \begin{macro}{\@bibstyle}
+% Define the argument of the \cmd\bibliographystyle\ command.
+% The user must have installed a \file{.bst} file of the corresponding name.
+% This file will then be used by \BibTeX\ when compiling the document's \file{.bbl} file.
+%
+% The default bibliography style for the AIP journal substyles are \file{aipnum.bst} and \file{aipauth.bst}.
+% To generate them, use \classname{custom-bib} version 4.21 or later.
+% Run the \file{.bst} generator, \file{makebst.tex}, and
+% accept all defaults, with the following exceptions:
+%
+% \begin{enumerate}
+% \item
+% ORDERING OF REFERENCES: c: seq-no (references in order of Citation);
+% \item
+% AUTHOR NAMES: i: nm-init,ed-au (Initials + surname);
+% \item
+% NUMBER OF AUTHORS: l: max 12, min 12 (there will be three prompts total);
+% \item
+% TYPEFACE FOR AUTHORS IN LIST OF REFERENCES: u: nmft,nmft-def (User defined author font);
+% \item
+% FONT FOR FIRST NAMES: u: fnm-def (First names in user defined font);
+% \item
+% EDITOR NAMES IN INCOLLECTION ETC: a: nmfted (Editors incollection like authors);
+% \item
+% DATE FORMAT: p: yr-par (Date in parentheses);
+% \item
+% SUPPRESS MONTH: x: xmth (Date is year only);
+% \item
+% TITLE OF ARTICLE: i: tit-it (Title italic)
+% \item
+% ARTICLE TITLE PRESENT: x: jtit-x (No article title);
+% \item
+% JOURNAL NAME FONT: r: jttl-rm (Journal title normal);
+% \item
+% TECHNICAL REPORT TITLE: b: trtit-b (Tech. report title like books);
+% \item
+% JOURNAL VOLUME: b: vol-bf (Volume bold);
+% \item
+% VOLUME PUNCTUATION: c: volp-com (Volume with comma);
+% \item
+% PAGE NUMBERS: f: jpg-1 (Only start page number);
+% \item
+% JOURNAL NAME PUNCTUATION: x: jnm-x (Space after journal);
+% \item
+% PUBLISHER IN PARENTHESES: d: pub-date (Publisher and date in parentheses);
+% \item
+% PUBLISHER POSITION: p: pre-pub (Publisher before chapter, pages);
+% \item
+% ISBN NUMBER: isbn: (Include ISBN for books, booklets)
+% \item
+% ISSN NUMBER: issn: (Include ISSN for periodicals)
+% \item
+% EDITOR IN COLLECTIONS: b: edby (Booktitle, edited by \dots);
+% \item
+% PUNCTUATION BETWEEN SECTIONS (BLOCKS): c: blk-com (Comma between blocks);
+% \item
+% ABBREVIATE WORD `PAGES': a: pp (`Page' abbreviated);
+% \item
+% ABBREVIATE WORD `EDITORS': a: ed (`Editor' abbreviated);
+% \item
+% OTHER ABBREVIATIONS: a: abr (Abbreviations);
+% \item
+% ABBREVIATION FOR `EDITION': a: ednx (`Edition' abbreviated as `ed');
+% \item
+% EDITION NUMBERS: n: ord (Numerical editions);
+% \item
+% STORED JOURNAL NAMES: a: jabr (Abbreviated journal names);
+% \item
+% FONT OF `ET AL': i: etal-it (Italic et al);
+% \item
+% ADDITIONAL REVTeX DATA FIELDS: r: revdata, eprint, url, url-blk (Include REVTeX data fields collaboration, eid, eprint, numpages, url)
+% \item
+% NEW FONT SELECTION SCHEME: n: nfss (NFSS);
+% \item
+% ADDITIONAL REVTeX DATA FIELDS: y: revdata (additional data fields);
+% \item
+% REFERENCE COMPONENT TAGS: y: reference component tags;
+% \item
+% URL ADDRESS: n: URL as note;
+% \end{enumerate}
+%
+% A file \file{aipnum.dbj} file equivalent to the following should result:
+% \begin{verbatim}
+%\input docstrip
+%\preamble
+%----------------------------------------
+%*** REVTeX-compatible aipnum4-1.bst 2009-10-08 ***
+%\endpreamble
+%\postamble
+%End of customized bst file
+%\endpostamble
+%\keepsilent
+%\askforoverwritefalse
+%\def\MBopts{\from{merlin.mbs}{%
+% head,\MBopta}
+%\from{physjour.mbs}{\MBopta}
+%\from{geojour.mbs}{\MBopta}
+%\from{photjour.mbs}{\MBopta}
+%\from{merlin.mbs}{tail,\MBopta}}
+%\def\MBopta{%
+% lang,%: Use language field to switch hyphenation patterns for title
+% pres,pres-bf,%: Presentation, speaker bold face
+% seq-no,%: Citation order (unsorted, like unsrt.bst)
+% vonx,%: Sort without von part (de la Maire after Mahone)
+% nm-init,ed-au,%: Initials + surname (J. F. Smith)
+% nmft,nmft-def,%: User defined author font (\bibnamefont)
+% fnm-def,%: First names in user defined font (\bibfnamefont)
+% nmfted,%: Editors incollection like authors font
+% nmand-rm,%: `And' in normal font (JONES and JAMES)
+% lab,lab-def,%: User defined citation font (\citenamefont)
+% and-rm,%: Cited `and' in normal font
+% yr-par,%: Date in parentheses as (May 1993)
+% date-nil-x,%: If date is empty, then do not produce the surrounding punctuation (parens, brackets, colon, comma)
+% tit-qq,%: Title and punctuation in double quotes (``Title,'' ..)
+% inproceedings-chapter,%: produce pages after chapter, just as in InBook
+% jtit-x,%: Title is ignored
+% inproceedings-chapter,%: produce pages after chapter just as in InBook
+% article-booktitle,%: format booktitle
+% article-series,%: article can has series
+% jttl-rm,%: Journal name normal font
+% journal-address,%: Include address field (in parentheses) along with journal name
+% book-bt,%: Field `booktitle', or if absent field `title', is book title
+% thesis-title-o,%: Title is optional: no warning issued if empty
+% techreport-institution-par,%: format tech report institution like book publisher
+% vol-bf,%: Volume bold as {\bf vol}(num)
+% vnum-x,%: Journal vol, without number as 34
+% volp-com,%: Volume with comma as vol(num), ppp
+% jpg-1,%: Only start page number
+% book-editor-booktitle,%: Book permits empty author, produces title before editor in this case
+% inbook-editor-booktitle,%: Allow using both title/booktitle, both author/editor
+% bookaddress,%: Italic booktitle followed by bookaddress in roman
+% num-xser,%: Allows number without series and suppresses word "number"
+% number-cap,%: Capitalize word `number' as: "Number 123"
+% chapter-cap,%: Capitalize word `chapter' as: `Chapter 42'
+% series-number,%: Series number as: `Springer Lecture Notes No. 125'
+% numser-booktitle,%: After book title and conference address, and before editors
+% ser-vol,%: Series, vol. 23
+% ser-rm,%: format series roman , even when used with volume
+% volume-cap,%: Capitalize word `volume', as: `Volume 7 in Lecture Series'
+% ser-ed,%: Series and volume after booktitle and before editors
+% jnm-x,%: Space after journal name
+% pg-bk,book-chapter-pages,%: As chapter and page: chapter 42, page 345
+% pub-date,%: Publisher with address and date in parentheses (Oxford, 1994)
+% ay-empty-pub-parens-x,%: eliminate parentheses altogether if nothing inside
+% pre-pub,%: Publisher before volume, chapter, pages
+% pre-edn,%: Edition before publisher
+% pre-pub,pre-edn,%: Edition, publisher, volume, chapter, pages
+% isbn,%: Include ISBN for books, booklets, etc.
+% issn,%: Include ISSN for periodicals
+% doi-link,doi,%: Doi forms a link to the publication, anchored to the volume or title
+% edby,%: In booktitle, edited by .. (where .. is names)
+% blk-com,%: Comma between blocks
+% fin-endbibitem,%: Command at end instead of period
+% pp,%: `Page' abbreviated as p. or pp.
+% ed,%: `Editor' abbreviated as ed. or eds.
+% abr,%: Abbreviations of such words
+% ednx,%: `Edition' abbreviated as `ed'
+% ord,%: Numerical editions as 1st, 2nd, 3rd, etc
+% jabr,%: Abbreviated journal names
+% etal-it,%: Italic et al
+% revdata,eprint,url,url-blk,translation,%: Include REVTeX data fields collaboration, eid, eprint, archive, url, translation
+% SLACcitation,%: Produce SLACcitation field
+% numpages-x,%: Do not include numpages field
+% url,url-prefix-x,%: URL without prefix (default: `URL ')
+% bibinfo,%: Reference component tags like \bibinfo in the content of \bibitem
+% bibfield,%: Element tags like \bibfield in the content of \bibitem
+% nfss,%: Use LaTeX commands which may not work with Plain TeX
+%,{%
+% }}
+%\generate{\file{aipnum4-1.bst}{\MBopts}}
+%\endbatchfile
+% \end{verbatim}
+%
+% Between the two files \file{aipnum.dbj} and \file{aipauth.dbj},
+% the differences are limited:
+% \file{aipnum.dbj} has two lines lacking in \file{aipauth.dbj}:
+% \begin{verbatim}
+% seq-no,%: Citation order (unsorted, like unsrt.bst)
+% nm-init,ed-au,%: Initials + surname (J. F. Smith)
+% \end{verbatim}
+% thus, the ``numbered citation'' bibliography is sorted by citation order, and
+% the names are give first inital, then last name.
+%
+% Likewise, \file{aipauth.dbj} has one line lacking in \file{aipnum.dbj}:
+% \begin{verbatim}
+% nm-rev,%: Surname + comma + initials (Smith, J. F.)
+% \end{verbatim}
+% Thus, the ``author-year'' bibliography is sorted by author name, and
+% names are given last name first, followed by initials.
+%
+% We ensure that the journal substyle has the first word
+% in the matter by installing the (default) AIP code
+% later on (see Section~\ref{sec:aip-defaults}).
+% \end{macro}
+%
+% \begin{macro}{\pre@bibdata}
+%
+% Set up to write endnotes to a .bib file; its data will be incorporated into the bibliography.
+% \begin{macrocode}
+ \def\pre@bibdata{\jobname\bibdata@app}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\bibsection}
+% We define the sectioning command to use when starting the bibliography
+% (we use \cmd\refname).
+% \begin{macrocode}
+\def\refname{References}%
+\def\rtx@bibsection{%
+ \@ifx@empty\refname{%
+ \par\vspace{6\p@ plus 6\p@}%
+ }{%
+% \end{macrocode}
+% The following line has been commented out:
+% \begin{verbatim}
+% \let\@hangfroms@section\@hang@froms
+% \end{verbatim}
+% \begin{macrocode}
+ \expandafter\section\expandafter*\expandafter{\refname}%
+ \@nobreaktrue
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\bibpreamble}
+% \begin{macro}{\bibsep}
+% \begin{macro}{\newblock}
+% \begin{macrocode}
+\let\bibpreamble\@empty
+\appdef\setup@hook{%
+ \bibsep\z@\relax
+}%
+\def\newblock{\ }%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\bibfont}
+% We define the font switch that applies to the body of the bibliography.
+%
+% \begin{macrocode}
+\appdef\setup@hook{%
+ \def\bibfont{%
+ \preprintsty@sw{}{\footnotesize}%
+ \@clubpenalty\clubpenalty
+ \labelsep\z@
+ }%
+}%
+\let\place@bibnumber\place@bibnumber@sup
+% \end{macrocode}
+% \end{macro}
+%
+%
+% \subsubsection{Index}%
+% \begin{macrocode}
+\newenvironment{theindex}{%
+ \columnseprule \z@
+ \columnsep 35\p@
+ \c@secnumdepth-\maxdimen
+ \onecolumngrid@push
+ \section{\indexname}%
+ \thispagestyle{plain}%
+ \parindent\z@
+ \parskip\z@ plus.3\p@\relax
+ \let\item\@idxitem
+ \onecolumngrid@pop
+}{%
+}%
+%
+\def\@idxitem{\par\hangindent 40\p@}
+%
+\def\subitem{\par\hangindent 40\p@ \hspace*{20\p@}}
+%
+\def\subsubitem{\par\hangindent 40\p@ \hspace*{30\p@}}
+%
+\def\indexspace{\par \vskip 10\p@ plus5\p@ minus3\p@\relax}
+% \end{macrocode}
+%
+%
+% \subsubsection{Typesize Processing}%
+% The formatting specification for the preprint option is:
+% Times Roman 12 pt, double spacing, with 1-inch margins on letter paper.
+% Paragraphs indented ``five spaces''.
+% Display math on standard indent, with equation number flush right in parenthesis, with subequations roman.
+%
+% That for reprint option is to format similar to the printed journal; with most journals
+% set (approximately analyzed) 10-point, 2-column grid, letter paper.
+% \begin{macrocode}
+\expandafter\def\csname rtx@aip10pt\endcsname{%
+ \let\@currname@class\@currname
+ \def\@currname{aps10pt\substyle@post}%
+ \class@info{Reading file \@currname.\substyle@ext}%
+ \input{\@currname.\substyle@ext}%
+ \let\@currname\@currname@class
+ \class@info{Overriding 10pt}%
+ \aipreprint
+}%
+\expandafter\def\csname rtx@aip11pt\endcsname{\csname rtx@aip12pt\endcsname}%
+\expandafter\def\csname rtx@aip12pt\endcsname{%
+ \let\@currname@class\@currname
+ \def\@currname{aps12pt\substyle@post}%
+ \class@info{Reading file \@currname.\substyle@ext}%
+ \input{\@currname.\substyle@ext}%
+ \let\@currname\@currname@class
+ \class@info{Overriding 12pt}%
+ \aippreprint
+}%
+% \end{macrocode}
+%
+% \begin{macro}{\today}
+% Procedure \cmd\today\ is used in the article class, but not in
+% this document class.
+% \begin{macrocode}
+\def\today{%
+ \number\day\space
+ \ifcase\month
+ \or January\or February\or March\or April\or May\or June%
+ \or July\or August\or September\or October\or November\or December%
+ \fi\space
+ \number\year
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \subsection{A. Running Title}
+% User-provided running title \cmd\@shorttitle\ to be set.
+%
+% \subsection{B. Equation numbering, figure and table numbering}
+% Equations can be numbered throughout, or by section, with appendix equations prepended by the appendix label.
+% Figures and tables are numbered throughout.
+%
+%
+% \subsection{C. Title}
+% Title page may break above abstract, within author list.
+% Title notes are signified by superior letter followed by right parenthesis.
+%
+% \subsection{D. Author footnote}
+% Set short line above these footnotes.
+%
+% \subsection{E. Author}
+% Class option \classoption{superscriptaddress} is standard.
+% \begin{macrocode}
+ \clo@superscriptaddress
+% \end{macrocode}
+%
+% \subsection{F. Affiliations}
+% Set italic.
+%
+% \subsection{G. Received date}
+%
+%
+% \subsection{H. Abstract}
+% Single paragraph, no indent.
+%
+% \subsection{J. Lead Paragraph}
+% Set boldface, flush left before main text, single paragraph.
+% Separated from text by a short centered rule.
+%
+% \subsection{K. Headings}
+% Use labels: ROMAN, LETTER, arabic, letter.
+% Set ALLCAPS, boldface; Initial cap, boldface; Initial cap, bold-italic; Initial cap, italic.
+%
+% Theorem, proof title followed by colon. Follow author.
+%
+% \subsection{L. Text Footnotes}
+% Not permitted; use endnotes.
+%
+% \subsection{M. Citations and Bibliography}
+% Numerical \filename{aipnum.bst}, author-year \filename{aipauth.bst}, and numbered author-year \filename{aipnumauth.bst} are the available choices.
+% Numerical is standard, with all styles permitted in journal classes 3a and 4b.
+%
+%
+% \subsection{N. References}
+%
+% Article Title usage:
+%
+% Article title required in all journals with "unpublished", "to be published", "in press" and "submitted" refs.
+%
+% Exceptions:
+%
+% Journal classes 1a,1b,2,4a:
+%
+% Article title not allowed in published references, except in the case of "in press" or submitted" (see sample references below).
+%
+% Journal class 3a:
+%
+% Article title allowed but not required in author-year references (use must be consistent within a single manuscript).
+%
+% Journal class 3b:
+%
+% Article title required in all journal references and report references.
+%
+% Journal class 4b:
+%
+% Article title allowed in journal references in author-year mode.
+%
+% \subsection{O. Examples of Numerical References}
+%
+%
+% \subsection{P. References in Author-year mode}
+%
+%
+% \subsection{Q. Numbered Author-year References}
+%
+%
+% \subsection{R. Tables}
+% Placed within text (floated), labeled with Roman numeral.
+%
+% Table caption placed above table, followed by space,
+% two double lines, space, table column headings, space,
+% single line, table contents, two double lines.
+%
+% Footnotes labeled with superior lower-case letter, set below table contents.
+%
+% \subsection{S. Figures}
+% Placed within text (floated), labeled with arabic numbers.
+%
+% Caption set below figure. A citation to be set inline, not superscripted.
+%
+% \subsection{Society defaults}%
+%
+% After this society file is read in, we will process the
+% \cmd\@journal- and \cmd\@pointsize-specific code.
+% Here we define the defaults.
+%
+% We select \textbf{Chaos} as the default journal substyle, because it is
+% the most permissive in terms of bibliography and citation style,
+% and it is formatted in two column in the reprint style.
+%
+% \begin{macrocode}
+\def\@journal@default{cha}%
+\def\@pointsize@default{12}%
+% \end{macrocode}
+%
+% \subsection{Journal-Specific Code}%
+%
+% For AIP journals, we supply code specific to JCP, POP, RSI, JAP, APL, CHA, POF, BMF, RSE, JMP.
+%
+% \subsubsection{\classoption{jcp}}%
+%A member of the journal class 1a.
+% \begin{macrocode}
+\def\rtx@aipjcp{%
+ \typeout{Using journal substyle \@journal.}%
+% \end{macrocode}
+% Journal class 1a uses the (superscript) numerical citation style by default.
+% \begin{macrocode}
+ \@booleanfalse\authoryear@sw
+% \end{macrocode}
+%
+% End of \classoption{jcp} code.
+% \begin{macrocode}
+}%
+% \end{macrocode}
+%
+% \subsubsection{\classoption{pop}}%
+%A member of the journal class 1a.
+% \begin{macrocode}
+\def\rtx@aippop{%
+ \typeout{Using journal substyle \@journal.}%
+% \end{macrocode}
+% Journal class 1a uses the (superscript) numerical citation style by default.
+% \begin{macrocode}
+ \@booleanfalse\authoryear@sw
+% \end{macrocode}
+%
+% End of \classoption{pop} code.
+% \begin{macrocode}
+}%
+% \end{macrocode}
+%
+% \subsubsection{\classoption{rsi}}%
+%A member of the journal class 1a.
+% \begin{macrocode}
+\def\rtx@aiprsi{%
+ \typeout{Using journal substyle \@journal.}%
+% \end{macrocode}
+% Journal class 1a uses the (superscript) numerical citation style by default.
+% \begin{macrocode}
+ \@booleanfalse\authoryear@sw
+% \end{macrocode}
+%
+% End of \classoption{rsi} code.
+% \begin{macrocode}
+}%
+% \end{macrocode}
+%
+% \subsubsection{\classoption{jap}}%
+%Sole member of the journal class 1b.
+%
+% No running title.
+% \begin{macrocode}
+\def\rtx@aipjap{%
+ \typeout{Using journal substyle \@journal.}%
+% \end{macrocode}
+% Journal class 1b uses the (superscript) numerical citation style by default.
+% \begin{macrocode}
+ \@booleanfalse\authoryear@sw
+ \let\@runningtitle\@empty
+% \end{macrocode}
+%
+% End of \classoption{jap} code.
+% \begin{macrocode}
+}%
+% \end{macrocode}
+%
+% \subsubsection{\classoption{apl}}%
+%Sole member of the journal class 2.
+%
+% No running title.
+% \begin{macrocode}
+\def\rtx@aipapl{%
+ \typeout{Using journal substyle \@journal.}%
+% \end{macrocode}
+% Journal class 2 uses the (superscript) numerical citation style by default.
+% \begin{macrocode}
+ \@booleanfalse\authoryear@sw
+ \let\@runningtitle\@empty
+% \end{macrocode}
+%
+% End of \classoption{apl} code.
+% \begin{macrocode}
+}%
+% \end{macrocode}
+%
+% \subsubsection{\classoption{apm}}%
+%
+% \begin{macrocode}
+\def\rtx@aipapm{%
+ \typeout{Using journal substyle \@journal.}%
+ \@booleanfalse\authoryear@sw
+ \let\@runningtitle\@empty
+}%
+% \end{macrocode}
+%
+%
+% \subsubsection{\classoption{cha}}%
+%Sole member of the journal class 3a.
+% \begin{macrocode}
+\def\rtx@aipcha{%
+ \typeout{Using journal substyle \@journal.}%
+% \end{macrocode}
+% Article titles are allowed.
+% \begin{macrocode}
+ \@booleanfalse\aip@jtitx@sw
+% \end{macrocode}
+%
+% End of \classoption{cha} code.
+% \begin{macrocode}
+}%
+% \end{macrocode}
+%
+% \subsubsection{\classoption{pof}}%
+%Sole member of the journal class 3b.
+% \begin{macrocode}
+\def\rtx@aippof{%
+ \typeout{Using journal substyle \@journal.}%
+% \end{macrocode}
+% Journal class 3b uses the (superscript) numerical citation style by default.
+% Article titles are required, so we include them if available,
+% but if absent, we can do no more than make a warning in the \file{.blg}.
+% \begin{macrocode}
+ \@booleanfalse\authoryear@sw
+ \@booleanfalse\aip@jtitx@sw
+% \end{macrocode}
+%
+% End of \classoption{pof} code.
+% \begin{macrocode}
+}%
+% \end{macrocode}
+%
+% \subsubsection{\classoption{bmf}}%
+% A member of the journal class 4a,
+% the Biomicrofluidics journal is single column.
+% \begin{macrocode}
+\def\rtx@aipbmf{%
+ \typeout{Using journal substyle \@journal.}%
+% \end{macrocode}
+% Journal class 4a uses the (superscript) numerical citation style by default.
+% Journal is formatted in a single column.
+% \begin{macrocode}
+ \@booleanfalse\authoryear@sw
+% \end{macrocode}
+%
+% End of \classoption{bmf} code.
+% \begin{macrocode}
+}%
+% \end{macrocode}
+%
+%
+% \subsubsection{\classoption{sd}}%
+% \begin{macrocode}
+\def\rtx@aipsd{%
+ \typeout{Using journal substyle \@journal.}%
+ \@booleanfalse\authoryear@sw
+}%
+% \end{macrocode}
+%
+% \subsubsection{\classoption{cp}}%
+% \begin{macrocode}
+\def\rtx@aipcp{%
+ \typeout{Using journal substyle \@journal.}%
+ \@booleanfalse\authoryear@sw%
+% \@booleantrue\authoryear@sw
+ \@booleanfalse\twocolumn@sw%
+% \@booleantrue\longbibliography@sw%
+% \@booleanfalse\aip@jtitx@sw
+}%
+% \end{macrocode}
+%
+% \subsubsection{\classoption{jor}}%
+% \begin{macrocode}
+\def\rtx@aipjor{%
+ \typeout{Using journal substyle \@journal.}%
+ \@booleantrue\authoryear@sw%
+ \@booleantrue\twoside@sw\@mparswitchfalse%
+}%
+% \end{macrocode}
+%
+% \subsubsection{\classoption{rse}}%
+%A member of the journal class 4a.
+% \begin{macrocode}
+\def\rtx@aiprse{%
+ \typeout{Using journal substyle \@journal.}%
+% \end{macrocode}
+% Journal class 4a uses the (superscript) numerical citation style by default.
+% \begin{macrocode}
+ \@booleanfalse\authoryear@sw
+% \end{macrocode}
+%
+% End of \classoption{rse} code.
+% \begin{macrocode}
+}%
+% \end{macrocode}
+%
+% \subsubsection{\classoption{jmp}}%
+% Sole member of the journal class 4b, Journal of Mathematical Physics is single column.
+% For an example of \env{itemize}, see JMP\_122901\_1 (48).
+% \begin{macrocode}
+\def\rtx@aipjmp{%
+ \typeout{Using journal substyle \@journal.}%
+% \end{macrocode}
+% Article titles are allowed.
+% Journal is formatted in a single column.
+% \begin{macrocode}
+ \@booleanfalse\aip@jtitx@sw
+ \@booleanfalse\twocolumn@sw
+ \@booleantrue\onecolumn@sw
+% \end{macrocode}
+%
+% End of \classoption{jmp} code.
+% \begin{macrocode}
+}%
+% \end{macrocode}
+%
+%
+% \subsubsection{\classoption{adv}}%
+% \begin{macrocode}
+\def\rtx@aipadv{%
+ \typeout{Using journal substyle \@journal.}%
+ \@booleanfalse\authoryear@sw
+ \let\@runningtitle\@empty
+}%
+% \end{macrocode}
+%
+%
+% \subsection{Establish AIP Defaults\label{sec:aip-defaults}}
+%
+% \begin{macro}{\footinbib@sw}
+% All AIP journals invoke the \classoption{footinbib} option.
+% \begin{macrocode}
+\@booleantrue\footinbib@sw
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\place@bibnumber}
+% \begin{macro}{\@bibstyle}
+% \begin{macro}{\bibpunct}
+% We install code that will
+% govern the style in which \cmd\cite\ commands are
+% formatted,
+% select the presentation for \cmd\bibitem s
+% and control the \BibTeX\ processing.
+%
+% Note that a journal substyle may override these settings.
+% Likewise, document preamble may itself invoke \cmd\bibpunct\ or \cmd\bibliographystyle,
+% thereby overriding these settings and those of the journal substyle.
+%
+% The numbered citations of \file{aipnum} and \file{aipauthnum}
+% are compatible with \classoption{footinbib} and
+% the compression and coalescing features of \classname{natbib},
+% while \file{aipauth}'s author-year citations are not.
+% Therefore, we de-select such options if we are selecting author-year citations.
+%
+% Note on \classname{natbib} presets:
+% \file{aipnum} and \file{aipauthnum} uses the Chicago \cmd\bibpunct\ style;
+% while \file{aipauth} uses that of Nature.
+%
+% Note on \cmd\NAT@mcite: if not using numerical citations, we set \cmd\NAT@mcite\
+% to a lower value, to turn off the mcite semantics of \classname{natbib}.
+% \begin{macrocode}
+\let\old@place@bibnumber\place@bibnumber
+\let\place@bibnumber\place@bibnumber@sup
+\appdef\setup@hook{%
+ \authoryear@sw{%
+ \aip@jtitx@sw{%
+ \def\@bibstyle{aipauth\substyle@post}%
+ }{%
+ \def\@bibstyle{aipauth\substyle@post}%
+ }%
+ \authornum@sw{%
+ \bibpunct{}{}{,}{s}{}{\textsuperscript{,}}%
+ \let\onlinecite\rev@citealpnum
+ }{%
+ \bibhang10\p@
+ \bibpunct{(%)
+ }{%(
+ )}{; }{a}{,}{,}%
+ \@booleanfalse\footinbib@sw
+ \let\NAT@mcite\@ne
+ \let\NAT@sort\z@
+ \def\NAT@cmprs{\z@}%
+ \let\NAT@def@citea\rtx@def@citea
+ \let\NAT@def@citea@close\rtx@def@citea@close
+ }%
+ }{%
+ \aip@jtitx@sw{%
+ \def\@bibstyle{aipnum\substyle@post}%
+ }{%
+ \def\@bibstyle{aipnum\substyle@post}%
+ }%
+ \bibpunct{}{}{,}{s}{}{\textsuperscript{,}}%
+ \let\onlinecite\rev@citealpnum
+ }%
+}%
+\def\make@footnote@endnote{%
+ \footinbib@sw{%
+ \authoryear@sw{\authornum@sw{\false@sw}{\true@sw}}{\false@sw}%
+ {}{%
+ \ltx@footnote@push
+ \def\thempfn{Note\thefootnote}%
+ \let\ltx@footmark\rev@citemark
+ \let\ltx@foottext\rev@endtext
+ \appdef\class@enddocumenthook{\auto@bib}%
+ \let\printendnotes\relax
+ }%
+ }{}%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\aipreprint}%
+% We want to override \filename{aps10pt.rtx}.
+% \begin{macrocode}
+\def\aipreprint{%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\aippreprint}%
+% We want to override \filename{aps12pt.rtx}.
+% \begin{macrocode}
+\def\aippreprint{%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \subsubsection{Fixing the bug for single author/affiliation}%
+% In case of single author/affiliation no need to produce the affiliation number after the author name.
+%
+% \begin{macrocode}
+%%
+\xdef\t@talAU{0}% TeXSupport
+\def\frontmatter@author@produce@script{%
+ \begingroup
+ \let\@author@present\@author@present@script
+ \frontmatterverbose@sw{\typeout{\string\frontmatter@author@produce@script:}\say\@AAC@list\say\@AFF@list\say\@AFG@list}{}%
+ \let\AU@temp\@empty
+ \@tempcnta\z@
+ \let\AF@opr \@gobble
+ \def\AU@opr{\@author@count\@tempcnta}%
+ \def\CO@opr{\@collaboration@count\AU@temp\@tempcnta}%
+ \@AAC@list%
+ \xdef\t@talAU{\the\@tempcnta}% TeXSupport
+ \expandafter\CO@opr\@author@cleared
+ \begingroup
+ \frontmatter@authorformat
+ \let\AF@opr \@affilID@def
+ \let\AU@opr \@author@present
+ \def\CO@opr{\@collaboration@present\AU@temp}%
+ \set@listcomma@list\AU@temp
+ \@AAC@list
+ \unskip\unskip
+ \par
+ \endgroup
+ \begingroup
+ \frontmatter@above@affiliation@script
+ \let\AFF@opr \@affil@script
+ \@AFF@list
+ \frontmatter@footnote@produce
+ \par
+ \endgroup
+ \endgroup
+}%
+\def\doauthor#1#2#3{% TeXSupport
+ \ignorespaces#1\unskip\@listcomma
+ \begingroup
+ \ifnum\t@talAU=1\else\ifnum\c@affil=\@ne\relax\else#3\fi\fi% TeXSupport
+ \@if@empty{#2}{\endgroup{}{}}{\endgroup{\ifnum\t@talAU=1\else\ifnum\c@affil=\@ne\relax\else\comma@space\fi\fi}{}\frontmatter@footnote{#2}}% TeXSupport
+ \space \@listand
+}%
+%%
+% \end{macrocode}
+%
+% \subsubsection{\classoption{Page restriction for APL (removed)}}%
+%
+% This macro was installed to calculate page-lenght for APL articles.
+%% Appended by Ms. Sehar Tahir (Aptara) for American Institute of Physics.
+%% To impose page restrictions for APL journal at MSP stage.
+%
+% \begin{macrocode}
+%%
+%% HISTORY
+%% Revised page restriction to 4 pages: 05/10/2011
+%% Tweaked the page over length message: 31/10/2011
+%% Removed unused page restriction code: 30/09/2020
+%%
+\DeclareOption{no-pagerestrict}{%
+ \class@warn{Deprecated pagerestrict code. Option 'no-pagerestrict' was ignored.}}%
+% \end{macrocode}
+% End of \classoption{apl} pagerestriction code.
+%
+% \subsubsection{Style updates for \texttt{jmp} and \texttt{rse}}%
+%
+% JMP, BMF and RSE journal's to be as one column at Reprint stage
+%
+%\begin{macrocode}
+%%
+%% JMP, BMF and RSE journal's to be as one column at Reprint stage
+%%
+\appdef\rtx@require@packages{%
+\def\jnl@rse{rse}%
+\def\jnl@jmp{jmp}%
+\preprintsty@sw{}%
+ {%\ifx\@journal\jnl@jmp%
+ \onecolumn@sw{%
+ \typeout{AIP Info: \@journal\space journal style Single column, 2013/07/30}%
+ \@booleanfalse\twocolumn@sw%
+ \appdef\setup@hook{%
+ \twoside@sw{%
+ \oddsidemargin 28pt
+ \evensidemargin 0pt
+ \marginparwidth 60pt
+ }{%
+ \oddsidemargin 28pt
+ \evensidemargin 0pt
+ \marginparwidth 44pt
+ }%
+ }%
+ \marginparsep 10pt
+ \topmargin -17pt
+ \headheight 12pt
+ \headsep 25pt
+ \topskip 10pt
+ \splittopskip\topskip
+ \footskip 30pt
+ \textheight=53.5pc
+ \textwidth 33pc
+ \columnsep 10pt
+ \def\title@column#1{%
+ \minipagefootnote@init
+ \begingroup
+ \let\@footnotetext\frontmatter@footnotetext
+ \ltx@no@footnote
+ #1%
+ \endgroup
+ \minipagefootnote@foot
+ }%
+}{}%
+% \fi% Curly brace
+}
+% \end{macrocode}
+%
+% \subsubsection{Style updates for \texttt{bmf} and addition of new substyle \texttt{sd}}%
+%
+% BMF journal to match with final layout (textheight and other page setup dimensions).
+% Addition of new journal SD: Structural Dynamics.
+%
+% \begin{macrocode}
+%% BMF journal to match with final layout (textheight)
+%% Addition of new journals SD: Structural Dynamics
+%
+\def\jnl@bmf{bmf}%
+\def\jnl@sd{sd}%
+\preprintsty@sw{}%
+ {%\ifx\@journal\jnl@bmf%
+ \newonecolumn@sw{%
+ \typeout{AIP Info: \@journal\space journal style Single column, 2011/08/11}%
+ \@booleanfalse\twocolumn@sw%
+ \appdef\setup@hook{%
+ \twoside@sw{%
+ \oddsidemargin 28pt
+ \evensidemargin 0pt
+ \marginparwidth 60pt
+ }{%
+ \oddsidemargin 28pt
+ \evensidemargin 0pt
+ \marginparwidth 44pt
+ }%
+ }%
+ \marginparsep 10pt
+ \topmargin -17pt
+ \headheight 12pt
+ \headsep 25pt% 25pt
+ \topskip 10pt
+ \splittopskip\topskip
+ \footskip 30pt
+% \textheight=53.5pc
+ \textheight=54.5pc%
+ \textwidth 33pc
+ \columnsep 10pt
+ \def\title@column#1{%
+ \minipagefootnote@init
+ \begingroup
+ \let\@footnotetext\frontmatter@footnotetext
+ \ltx@no@footnote
+ #1%
+ \endgroup
+ \minipagefootnote@foot
+ }%
+\def\adjust@abstractwidth{%
+ \parindent1em\relax
+% \advance\leftskip.5in\relax
+ \advance\leftskip2.5pc\relax
+ \advance\rightskip2.5pc\relax
+ \@totalleftmargin\leftskip
+ \@afterheading\@afterindentfalse
+}%
+% \fi
+}{}%
+}
+% \end{macrocode}
+%
+% \subsubsection{Addition of new substyle \texttt{jor}}%
+%
+% Addition of new journal JOR: Journal of Rheology.
+%
+% \begin{macrocode}
+%%
+%% Journal of Rheology (jor)
+%%
+\def\jnl@jor{jor}%
+\preprintsty@sw{}%
+ {\ifx\@journal\jnl@jor%
+ \typeout{AIP Info: \@journal\space journal style Single column, 2013/10/24}%
+ \@booleanfalse\twocolumn@sw%
+ \appdef\setup@hook{%
+ \twoside@sw{%
+ \oddsidemargin 28pt
+ \evensidemargin 0pt
+ \marginparwidth 60pt
+ }{%
+ \oddsidemargin 28pt
+ \evensidemargin 0pt
+ \marginparwidth 44pt
+ }%
+ }%
+ \marginparsep 10pt
+ \topmargin -17pt
+ \headheight 12pt
+ \headsep 25pt
+ \topskip 10pt
+ \splittopskip\topskip
+ \footskip 30pt
+ \textheight=53.5pc
+ \textwidth 33pc
+ \columnsep 10pt
+ \def\title@column#1{%
+ \minipagefootnote@init
+ \begingroup
+ \let\@footnotetext\frontmatter@footnotetext
+ \ltx@no@footnote
+ #1%
+ \endgroup
+ \minipagefootnote@foot
+ }%
+\def\frontmatter@title@format{%
+ \preprintsty@sw{}{\Large}%
+ \sffamily%
+ \bfseries%
+ \leftskip0pt plus1fill%
+ \rightskip0pt plus1fill%
+ \parindent\z@%
+%\raggedright%
+ \parskip\z@skip%
+}%
+\def\frontmatter@@indent{%
+ \skip@\@flushglue
+ \@flushglue\z@ plus.3\hsize\relax
+ \leftskip0pt plus1fill%
+ \rightskip0pt plus1fill%
+ \parindent\z@%
+%\raggedright
+%\advance\leftskip.5in\relax
+ \@totalleftmargin\leftskip
+ \@flushglue\skip@
+}%
+%\def\frontmatter@authorformat{%
+% \frontmatter@@indent
+% \sffamily
+%}%
+\def\frontmatter@RRAP@format{%
+ \addvspace{5\p@}%
+ \small
+ \leftskip0pt plus1fill%
+ \rightskip0pt plus1fill%
+ \parindent\z@%
+%%\raggedright
+%%\advance\leftskip.5in\relax
+ \@totalleftmargin\leftskip
+ \everypar{%
+ \hbox\bgroup(\@gobble@leavemode@uppercase%)
+ }%
+ \def\par{%
+ \@ifvmode{}{%(
+ \unskip)\egroup\@@par
+ }%
+ }%
+}%
+\def\frontmatter@abstractfont{}%
+\def\frontmatter@abstractwidth{\textwidth}
+\def\abstractname{Synopsis}
+\def\frontmatter@abstractheading{%
+ \begingroup
+ \centering\large
+ {\bfseries\abstractname}
+ \par\vskip.25\baselineskip
+ \endgroup
+}%
+\appdef\setup@hook{%
+ \preprintsty@sw{}{%
+ \let\refname\old@refname%\@empty
+ }%
+}%
+%
+\def\ps@article{%
+ \def\@evenhead{\let\\\heading@cr\sffamily\thepage\quad\checkindate\hfil\@runningtitle\hfil}%
+ \def\@oddhead{\let\\\heading@cr\hfil\sffamily\@runningtitle\hfil\checkindate\quad\thepage}%
+ \def\@oddfoot{}%
+ \def\@evenfoot{}%
+ \let\@mkboth\@gobbletwo
+ \let\sectionmark\@gobble
+ \let\subsectionmark\@gobble
+}%
+\def\@make@capt@title#1#2{%
+ \@ifx@empty\float@link{\@firstofone}{\expandafter\href\expandafter{\float@link}}%
+ {{\bfseries#1}}\@caption@fignum@sep#2%
+}%
+\def\@caption@fignum@sep{{\bfseries.} }%
+ \fi}
+% \end{macrocode}
+%
+% \subsubsection{Addition of new substyle \texttt{cp}}%
+%
+% Addition of new journal AIP-CP: AIP Conference Proceedings.
+%
+% \begin{macrocode}
+%%
+%% AIP Conference Proceedings
+%%
+\def\jnl@cp{cp}%
+\preprintsty@sw{}%
+ {\ifx\@journal\jnl@cp%
+ \typeout{AIP Info: \@journal\space journal style Single column, 2013/10/30}%
+ \@booleanfalse\twocolumn@sw%
+ \appdef\setup@hook{%
+ \twoside@sw{%
+ \oddsidemargin 0pt
+ \evensidemargin 0pt
+ \marginparwidth 60pt
+ }{%
+ \oddsidemargin 0pt
+ \evensidemargin 0pt
+ \marginparwidth 44pt
+ }%
+ }%
+ \marginparsep 10pt
+ \topmargin -29.5pt
+ \headheight 12pt
+ \headsep 25pt
+ \topskip 10pt
+ \splittopskip\topskip
+ \footskip 30pt
+ \textheight=53.5pc
+ \textwidth 39pc
+ \columnsep 10pt
+%%
+ \def\title@column#1{%
+ \minipagefootnote@init
+ \begingroup
+ \let\@footnotetext\frontmatter@footnotetext
+ \ltx@no@footnote
+ #1%
+ \endgroup
+ \minipagefootnote@foot
+ }%
+%%
+\def\frontmatter@title@format{%
+ \preprintsty@sw{}{\LARGE}% \Large
+%\sffamily%
+ \bfseries%
+ \leftskip0pt plus1fill%
+ \rightskip0pt plus1fill%
+ \parindent\z@%
+%\raggedright%
+ \parskip\z@skip%
+}%
+%%
+\def\frontmatter@@indent{%
+ \skip@\@flushglue
+ \@flushglue\z@ plus.3\hsize\relax
+ \leftskip0pt plus1fill%
+ \rightskip0pt plus1fill%
+ \parindent\z@%
+%\raggedright
+%\advance\leftskip.5in\relax
+ \@totalleftmargin\leftskip
+ \@flushglue\skip@
+}%
+%%
+\def\frontmatter@RRAP@format{%
+ \addvspace{5\p@}%
+ \small
+ \leftskip0pt plus1fill%
+ \rightskip0pt plus1fill%
+ \parindent\z@%
+%%\raggedright
+%%\advance\leftskip.5in\relax
+ \@totalleftmargin\leftskip
+ \everypar{%
+ \hbox\bgroup(\@gobble@leavemode@uppercase%)
+ }%
+ \def\par{%
+ \@ifvmode{}{%(
+ \unskip)\egroup\@@par
+ }%
+ }%
+}%
+\def\frontmatter@abstractfont{\small}%
+\def\frontmatter@abstractwidth{\textwidth}
+\def\@keys@name{{\small\bfseries Keywords:} }%
+\def\@pacs@name{{\small\bfseries PACS:} }%
+\def\abstractname{Abstract.}
+%%
+\def\frontmatter@abstractheading{%
+ \begingroup
+% \centering\large
+ \small%
+ {\bfseries\abstractname}
+% \par\vskip.25\baselineskip
+ \endgroup
+}%
+%%
+\def\adjust@abstractwidth{%
+ \parindent1em\relax
+% \advance\leftskip.5in\relax
+ \@totalleftmargin\leftskip
+ \@afterheading\@afterindentfalse
+ \small%
+}%
+%%
+\def\frontmatter@PACS@format{%
+% \addvspace{11\p@}%
+ \adjust@abstractwidth
+% \parskip\z@skip
+ \samepage
+}%
+\def\frontmatter@keys@format{%
+ \addvspace{11\p@}%
+ \adjust@abstractwidth
+ \parskip\z@skip
+ \samepage
+}%
+%%
+\def\frontmatter@authorformat{%
+ \frontmatter@@indent
+ \Large%
+% \sffamily
+ \vskip.5\baselineskip%
+}%
+%%
+%%
+\def\titleblock@produce{%
+ \begingroup
+ \ltx@footnote@pop
+ \def\@mpfn{mpfootnote}%
+ \def\thempfn{\thempfootnote}%
+ \c@mpfootnote\z@
+ \let\@makefnmark\frontmatter@makefnmark
+ \frontmatter@setup
+ \thispagestyle{titlepage}\label{FirstPage}%
+ \frontmatter@title@produce
+ \groupauthors@sw{%
+ \frontmatter@author@produce@group
+ }{%
+ \frontmatter@author@produce@script
+ }%
+ \frontmatter@RRAPformat{%
+ \expandafter\produce@RRAP\expandafter{\@date}%
+ \expandafter\produce@RRAP\expandafter{\@received}%
+ \expandafter\produce@RRAP\expandafter{\@revised}%
+ \expandafter\produce@RRAP\expandafter{\@accepted}%
+ \expandafter\produce@RRAP\expandafter{\@published}%
+ }%
+ \frontmatter@abstract@produce
+ \@ifx@empty\@keywords{}{%
+ \@keywords@produce\@keywords
+ }%
+ \@ifx@empty\@pacs{}{%
+ \@pacs@produce\@pacs
+ }%
+ \par
+ \frontmatter@finalspace
+ \endgroup
+}%
+%%
+\appdef\setup@hook{%
+ \preprintsty@sw{}{%
+ \let\refname\old@refname%\@empty
+ }%
+}%
+%%
+\def\ps@article{%
+% \def\@evenhead{\let\\\heading@cr\thepage\quad\checkindate\hfil{\leftmark}}%
+% \def\@oddhead{\let\\\heading@cr{\rightmark}\hfil\checkindate\quad\thepage}%
+ \def\@evenhead{}%
+ \def\@oddhead{}%
+ \def\@oddfoot{}%
+ \def\@evenfoot{}%
+ \let\@mkboth\markboth
+ \let\sectionmark\@gobble
+ \let\subsectionmark\@gobble
+}%
+%%
+\def\@make@capt@title#1#2{%
+ \@ifx@empty\float@link{\@firstofone}{\expandafter\href\expandafter{\float@link}}%
+ {{\bfseries#1}}\@caption@fignum@sep#2%
+}%
+\def\@caption@fignum@sep{{\bfseries.} }%
+\setcounter{secnumdepth}{0}
+\def\section{\@startsection{section}{1}{\z@}{-0.8cm \@plus-1ex \@minus -.2ex}{0.5cm}{\normalfont\large\bfseries\centering}}% \sffamily\small
+\def\subsection{\@startsection{subsection}{2}{\z@}{-.8cm \@plus-1ex \@minus -.2ex}{.5cm}{\normalfont\large\bfseries\centering}}% \sffamily\small
+\def\subsubsection{\@startsection{subsubsection}{3}{\z@}{-.8cm \@plus-1ex \@minus -.2ex}{.5cm}{\normalfont\large\bfseries\itshape\centering}}% \sffamily\small
+%
+\def\BIBsection{\@startsection{section}{1}{\z@}{-0.8cm \@plus-1ex \@minus -.2ex}{0.5cm}{\normalfont\large\bfseries\centering}}% \sffamily\small
+\def\figurename{FIGURE}
+%%
+\newcommand\rtx@myciteonline[3]{%
+ \ifNAT@swa
+ \leavevmode
+ \unskip
+% \textsuperscript{\normalfont#1}% TeXSupport
+ \space[{\normalfont#1}]%
+ \if*#3*\else\ (#3)\fi
+ \else
+ #1%
+ \fi
+ \endgroup
+}%
+%%
+\appdef\setup@hook{%
+ \authoryear@sw{}{%
+ \def\NAT@cmprs{\z@}
+% \let\rtx@citesuper\rtx@mycitesuper%
+ \let\NAT@citesuper\rtx@myciteonline%\rtx@citesuper
+ \let\place@bibnumber\old@place@bibnumber
+ \gdef\place@bibnumber@online#1{{#1}.\hskip5pt}%
+ \let\place@bibnumber\place@bibnumber@online
+ \def\NAT@spacechar{}%
+ \def\NAT@conj{and\ }%
+ \bibpunct{}{}{, }{s}{}{\textsuperscript{,}}% TeXSupport
+ }%
+}%
+%%
+ \def\rtx@mybibsection{%
+ \@ifx@empty\refname{%
+ \par\vspace{6\p@ plus 6\p@}%
+ }{%
+ \expandafter\BIBsection\expandafter*\expandafter{\refname}%
+ \@nobreaktrue
+ }%
+}%
+\let\bibsection\rtx@mybibsection%
+ \fi}
+%
+% \end{macrocode}
+%
+% \subsubsection{\classoption{pof}}%
+% This macro modifies the reprint output of POF journal to single column.
+% \begin{macrocode}
+%% POF journal to be one column at Reprint stage
+%%
+%% HISTORY
+%% Further corrected the text area and margins to closely match Print output: 23/04/2012
+\def\jnl@pof{pof}%
+\preprintsty@sw{}%
+ {\ifx\@journal\jnl@pof%
+ \typeout{AIP Info: \@journal\space journal style Single column, 2011/08/11}%
+ \@booleanfalse\twocolumn@sw%
+ \appdef\setup@hook{%
+ \twoside@sw{%
+ \oddsidemargin 28pt
+ \evensidemargin 0pt
+ \marginparwidth 60pt
+ }{%
+ \oddsidemargin 28pt
+ \evensidemargin 0pt
+ \marginparwidth 44pt
+ }%
+ }%
+ \marginparsep 10pt
+ \topmargin -17pt
+ \headheight 12pt
+ \headsep 25pt
+ \topskip 10pt
+ \splittopskip\topskip
+ \footskip 30pt
+ \textheight=53.5pc
+ \textwidth 33pc
+ \columnsep 10pt
+ \def\title@column#1{%
+ \minipagefootnote@init
+ \begingroup
+ \let\@footnotetext\frontmatter@footnotetext
+ \ltx@no@footnote
+ #1%
+ \endgroup
+ \minipagefootnote@foot
+ }%
+ \fi}
+}%
+% \end{macrocode}
+% End of \classoption{pof} code.
+%
+% \begin{macrocode}
+%</package>
+% \end{macrocode}
+%
+%
+%
+% \section{AAPM: The Socity Journals Style (\texttt{mph} substyle)}
+%
+% The file \file{aapm.rtx} is read in by the \classname{revtex4} document class
+% for applying the journal specific changes. User should use \texttt{[aapm,mph]}
+% for applying the required updates.
+%
+%
+% \begin{macrocode}
+%<*aapm>
+% \end{macrocode}
+%
+% Protect this file from being read in by anything but \revtex.
+% \begin{macrocode}
+\ifx\undefined\substyle@ext
+ \def\@tempa{%
+ \endinput
+ \GenericWarning{I must be read in by REVTeX! (Bailing out)}%
+ }%
+ \expandafter\else
+ \def\@tempa{}%
+ \expandafter\fi\@tempa
+ \class@info{RevTeX society AAPM selected}%
+\DeclareOption{mph}{\change@journal{mph}}%
+%%
+\@booleantrue\longbibliography@sw
+\@booleanfalse\authoryear@sw
+\def\@bibstyle{aapmrev\substyle@post}%
+\appdef\@bibdataout@rev{\@bibdataout@aapm}%
+\def\@bibdataout@aapm{%
+ \immediate\write\@bibdataout{%
+ @CONTROL{%
+ aapm41Control%
+ \longbibliography@sw{%
+ ,pages="1",title="0"%
+ }{%
+ ,pages="0",title=""%
+ }%
+ }%
+ }%
+ \if@filesw
+ \immediate\write\@auxout{\string\citation{aapm41Control}}%
+ \fi
+}%
+\appdef\setup@hook{%
+ \lengthcheck@sw{%
+ \RequirePackage{times}%
+ \frenchspacing%
+ }{}%
+}
+\@booleantrue\preprintsty@sw
+\@booleantrue\showPACS@sw
+\@booleantrue\showKEYS@sw
+\appdef\setup@hook{%
+ \preprintsty@sw{}{%
+ \let\refname\@empty
+ }%
+}%
+\appdef\setup@hook{%
+ \preprintsty@sw{%
+ \ps@preprint
+ }{%
+ \ps@article
+ }%
+}%
+\def\ps@preprint{%
+ \def\@oddhead{\@runningtitle\hfil}%
+ \def\@evenhead{\@runningtitle\hfil}%
+ \def\@oddfoot{\hfil\thepage\quad\checkindate\hfil}%
+ \def\@evenfoot{\hfil\thepage\quad\checkindate\hfil}%
+ \let\@mkboth\@gobbletwo
+ \let\sectionmark\@gobble
+ \let\subsectionmark\@gobble
+}%
+\def\ps@article{%
+ \def\@evenhead{\let\\\heading@cr\thepage\quad\checkindate\hfil\@runningtitle}%
+ \def\@oddhead{\let\\\heading@cr\@runningtitle\hfil\checkindate\quad\thepage}%
+ \def\@oddfoot{}%
+ \def\@evenfoot{}%
+ \let\@mkboth\@gobbletwo
+ \let\sectionmark\@gobble
+ \let\subsectionmark\@gobble
+}%
+\def\@runningtitle{\@shorttitle}%
+\renewenvironment{titlepage}{%
+ \let\wastwocol@sw\twocolumn@sw
+ \onecolumngrid
+ \newpage
+ \thispagestyle{titlepage}%
+ \c@page\z@% article sets this to one not zero???
+}{%
+ \wastwocol@sw{\twocolumngrid}{\newpage}%
+}%
+\let\@fnsymbol@latex\@fnsymbol
+\let\@fnsymbol\@alph
+\def\adjust@abstractwidth{%
+ \parindent1em\relax
+ \advance\leftskip.5in\relax
+ \@totalleftmargin\leftskip
+ \preprintsty@sw{}{\rightskip.14\hsize\relax}
+ \@afterheading\@afterindentfalse
+}%
+\def\frontmatter@abstractheading{}%
+\def\frontmatter@abstractfont{%
+ \adjust@abstractwidth
+}%
+\def\frontmatter@postabstractspace{1\baselineskip}
+\def\frontmatter@finalspace{\addvspace{28\p@}}
+\appdef\setup@hook{%
+ \preprintsty@sw{%
+ \@booleantrue\titlepage@sw
+ \let\section\section@preprintsty
+ \let\subsection\subsection@preprintsty
+ \let\subsubsection\subsubsection@preprintsty
+ }{}%
+}%
+\def\frontmatter@@indent{%
+ \skip@\@flushglue
+ \preprintsty@sw{\@flushglue\z@ plus.3\hsize\relax}%
+ {\@flushglue.14\hsize\relax}
+ \raggedright
+ \advance\leftskip.5in\relax
+ \@totalleftmargin\leftskip
+ \@flushglue\skip@
+}%
+\def\frontmatter@authorformat{%
+ \frontmatter@@indent
+ \sffamily
+}%
+\renewcommand*\email[1][Electronic mail: ]{\begingroup\sanitize@url\@email{#1}}%
+\def\frontmatter@above@affilgroup{\par\addvspace{6\p@}%
+}%
+\def\frontmatter@above@affiliation@script{%
+ \frontmatter@@indent
+}%
+\def\frontmatter@above@affiliation{%
+}%
+\def\frontmatter@affiliationfont{%
+ \frontmatter@@indent
+ \preprintsty@sw{}{\small}%
+ \it
+}%
+\def\frontmatter@collaboration@above{%
+}%
+\def\frontmatter@setup{%
+ \normalfont
+}%
+\def\frontmatter@title@above{\addvspace{6\p@}}%
+\def\frontmatter@title@format{%
+ \preprintsty@sw{}{\Large}%
+ \sffamily
+ \bfseries
+ \raggedright
+ \parskip\z@skip
+}%
+\def\frontmatter@title@below{\addvspace{7.5\p@}}%
+\def\@author@parskip{3\p@}%
+\@booleanfalse\altaffilletter@sw
+\def\frontmatter@makefnmark{%
+ \@textsuperscript{%
+ \normalfont\@thefnmark%(
+ )%
+ }%
+}%
+\def\frontmatter@authorbelow{%
+\addvspace{3\p@}%
+}%
+\let\affil@cutoff\tw@
+\def\frontmatter@RRAP@format{%
+ \addvspace{2\p@}%
+ \raggedright
+ \advance\leftskip.5in\relax
+ \@totalleftmargin\leftskip
+ \preprintsty@sw{}{\advance\rightskip.14\hsize\relax}
+ \everypar{%
+ \hbox\bgroup(\@gobble@leavemode@uppercase%)
+ }%
+ \def\par{%
+ \@ifvmode{}{%(
+ \unskip)\egroup\@@par
+ }%
+ }%
+}%
+\def\punct@RRAP{;\egroup\ \hbox\bgroup}%
+\def\@gobble@leavemode@uppercase#1#2{\expandafter\MakeTextUppercase}%
+\def\frontmatter@PACS@format{%
+ \addvspace{11\p@}%
+ \adjust@abstractwidth
+ \parskip\z@skip
+ \samepage
+}%
+\def\frontmatter@keys@format{%
+ \adjust@abstractwidth
+ \samepage
+}%
+\def\ps@titlepage{%
+ \def\@oddhead{%
+ \@runningtitle
+ \hfill
+ \produce@preprints\@preprint
+ }%
+ \let\@evenhead\@oddhead
+ \def\@oddfoot{%
+ \hb@xt@\z@{\byrevtex\hss}%
+ \hfil
+ \preprintsty@sw{\thepage}{}%
+ \quad\checkindate
+ \hfil
+ }%
+ \let\@evenfoot\@oddfoot
+}%
+\def\byrevtex{\byrevtex@sw{Typeset by REV\TeX and AAPM}{}}%
+\def\produce@preprints#1{%
+ \preprint@sw{%
+ \vtop to \z@{%
+ \def\baselinestretch{1}%
+ \small
+ \let\preprint\preprint@count
+ \count@\z@#1\@ifnum{\count@>\tw@}{%
+ \hbox{%
+ \let\preprint\preprint@hlist
+ #1\setbox\z@\lastbox
+ }%
+ }{%
+ \let\preprint\preprint@cr
+ \halign{\hfil##\cr#1\crcr}%
+ \par
+ \vss
+ }%
+ }%
+ }{}%
+}%
+\def\preprint@cr#1{#1\cr}%
+\def\preprint@count#1{\advance\count@\@ne}%
+\def\preprint@hlist#1{#1\hbox{, }}%
+\newenvironment{Lead@inParagraph}{%
+ \par
+ \bfseries
+ \@afterheading\@afterindentfalse
+}{%
+ \par
+ \hb@xt@\hsize{\hfil\leaders\hrule\hfil\leaders\hrule\hfil\hfil}%
+}%
+\appdef\frontmatter@init{%
+ \let@environment{quotation@ltx}{quotation}%
+ \let@environment{quotation}{Lead@inParagraph}%
+}%
+\appdef\@startsection@hook{%
+ \let@environment{quotation}{quotation@ltx}%
+}%
+\def\secnums@rtx{%
+ \@ifxundefined\thepart{%
+ \def\thepart {\Roman{part}}%
+ }{}%
+ \@ifxundefined\thesection{%
+ \def\thesection {\Roman{section}}%
+ \def\p@section {}%
+ }{}%
+ \@ifxundefined\thesubsection{%
+ \def\thesubsection {\thesection.\Alph{subsection}}%
+ \def\p@subsection {}%
+ }{}%
+ \@ifxundefined\thesubsubsection{%
+ \def\thesubsubsection {\thesubsection.\arabic{subsubsection}}%
+ \def\p@subsubsection {}%
+ }{}%
+ \@ifxundefined\theparagraph{%
+ \def\theparagraph {\thesubsubsection.\alph{paragraph}}%
+ \def\p@paragraph {}%
+ }{}%
+ \@ifxundefined\thesubparagraph{%
+ \def\thesubparagraph {\theparagraph.\arabic{subparagraph}}%
+ \def\p@subparagraph {}%
+ }{}%
+}%
+\def\@seccntformat#1{\csname the#1\endcsname.\hskip0.5em\relax}%
+\def\@hang@from#1#2#3{#1#2#3}%
+\def\section{%
+ \@startsection
+ {section}%
+ {1}%
+ {\z@}%
+ {1.5\baselineskip \@plus1ex \@minus .2ex}%
+ {.5\baselineskip}%
+ {%
+ \normalfont
+ \sffamily
+ \bfseries
+ \raggedright
+ }%
+}%
+\def\@hangfrom@section#1#2#3{\@hangfrom{#1#2}\MakeTextUppercase{#3}}%
+\def\@hangfroms@section#1#2{#1\MakeTextUppercase{#2}}%
+\def\subsection{%
+ \@startsection
+ {subsection}%
+ {2}%
+ {\z@}%
+ {1.5\baselineskip \@plus1ex \@minus .2ex}%
+ {.5\baselineskip}%
+ {%
+ \normalfont
+ \fontsize{9.5}{12}
+ \sffamily
+ \bfseries
+ \raggedright
+ }%
+}%
+\def\subsubsection{%
+ \@startsection
+ {subsubsection}%
+ {3}%
+ {\z@}%
+ {1.5\baselineskip \@plus1ex \@minus .2ex}%
+ {.5\baselineskip}%
+ {%
+ \normalfont
+ \fontsize{9.5}{12}
+ \sffamily
+ \bfseries
+ \itshape
+ \raggedright
+ }%
+}%
+\def\paragraph{%
+ \@startsection
+ {paragraph}%
+ {4}%
+ {\parindent}%
+ {\z@}%
+ {-1em}%
+ {\normalfont\normalsize\itshape}%
+}%
+\def\subparagraph{%
+ \@startsection
+ {subparagraph}%
+ {5}%
+ {\parindent}%
+ {3.25ex \@plus1ex \@minus .2ex}%
+ {-1em}%
+ {\normalfont\normalsize\bfseries}%
+}%
+\def\section@preprintsty{%
+ \@startsection
+ {section}%
+ {1}%
+ {\z@}%
+ {0.8cm \@plus1ex \@minus .2ex}%
+ {0.5cm}%
+ {%
+ \normalfont
+ \bfseries
+ \raggedright
+ }%
+}%
+\def\subsection@preprintsty{%
+ \@startsection
+ {subsection}%
+ {2}%
+ {\z@}%
+ {.8cm \@plus1ex \@minus .2ex}%
+ {.5cm}%
+ {%
+ \normalfont
+ \bfseries
+ \raggedright
+ }%
+}%
+\def\subsubsection@preprintsty{%
+ \@startsection
+ {subsubsection}%
+ {3}%
+ {\z@}%
+ {.8cm \@plus1ex \@minus .2ex}%
+ {.5cm}%
+ {%
+ \normalfont
+ \itshape\bfseries
+ \raggedright
+ }%
+}%
+\let\frontmatter@footnote@produce\frontmatter@footnote@produce@endnote
+\def\@pnumwidth{1.55em}
+\def\@tocrmarg {2.55em}
+\def\@dotsep{2}
+\def\ltxu@dotsep{4.5pt}
+\setcounter{tocdepth}{3}
+\def\tableofcontents{%
+ \addtocontents{toc}{\string\tocdepth@munge}%
+ \print@toc{toc}%
+ \addtocontents{toc}{\string\tocdepth@restore}%
+}%
+\def\tocdepth@munge{%
+ \let\l@section@saved\l@section
+ \let\l@section\@gobble@tw@
+}%
+\def\@gobble@tw@#1#2{}%
+\def\tocdepth@restore{%
+ \let\l@section\l@section@saved
+}%
+\def\l@part#1#2{\addpenalty{\@secpenalty}%
+ \begingroup
+ \set@tocdim@pagenum{#2}%
+ \parindent \z@
+ \rightskip\tocleft@pagenum plus 1fil\relax
+ \skip@\parfillskip\parfillskip\z@
+ \addvspace{2.25em plus\p@}%
+ \large \bf %
+ \leavevmode\ignorespaces#1\unskip\nobreak\hskip\skip@
+ \hb@xt@\rightskip{\hfil\unhbox\z@}\hskip-\rightskip\hskip\z@skip
+ \par
+ \nobreak %
+ \endgroup
+}%
+\def\tocleft@{\z@}%
+\def\tocdim@min{5\p@}%
+\def\l@section{%
+ \l@@sections{}{section}% Implicit #3#4
+}%
+\def\l@f@section{%
+ \addpenalty{\@secpenalty}%
+ \addvspace{1.0em plus\p@}%
+ \bf
+}%
+\def\l@subsection{%
+ \l@@sections{section}{subsection}% Implicit #3#4
+}%
+\def\l@subsubsection{%
+ \l@@sections{subsection}{subsubsection}% Implicit #3#4
+}%
+\def\l@paragraph#1#2{}%
+\def\l@subparagraph#1#2{}%
+\let\toc@pre\toc@pre@auto
+\let\toc@post\toc@post@auto
+\def\listoffigures{\print@toc{lof}}%
+\def\l@figure{\@dottedtocline{1}{1.5em}{2.3em}}
+\def\listoftables{\print@toc{lot}}%
+\let\l@table\l@figure
+\def\figurename{\textsc{Fig.}}
+\def\tablename{\textsc{Table}}
+\long\def\@makecaption#1#2{%
+ \par
+ \vskip\abovecaptionskip
+ \begingroup
+ \preprintsty@sw{\small}{\footnotesize}\rmfamily
+ \sbox\@tempboxa{%
+ \let\\\heading@cr
+ \@make@capt@title{#1}{#2}%
+ }%
+ \@ifdim{\wd\@tempboxa >\hsize}{%
+ \begingroup
+ \samepage
+ \flushing
+ \let\footnote\@footnotemark@gobble
+ \@make@capt@title{#1}{#2}\par
+ \endgroup
+ }{%
+ \global \@minipagefalse
+ \hb@xt@\hsize{\hfil\unhbox\@tempboxa\hfil}%
+ }%
+ \endgroup
+ \vskip\belowcaptionskip
+}%
+\def\@caption@fignum@sep{\nobreak\hskip.5em plus.2em\ignorespaces}%
+\@booleanfalse\raggedcolumn@sw
+\def\table@hook{\preprintsty@sw{\small}{\footnotesize}}%
+\def\tableft@skip@float{\z@ plus\hsize}%
+\def\tabmid@skip@float{\@flushglue}%
+\def\tabright@skip@float{\z@ plus\hsize}%
+\def\array@row@pre@float{\hline\hline\noalign{\vskip\doublerulesep}}%
+\def\array@row@pst@float{\noalign{\vskip\doublerulesep}\hline\hline}%
+\def\@makefntext#1{%
+ \def\baselinestretch{1}%
+ \leftskip1em%
+ \parindent1em%
+ \noindent
+ \nobreak\hskip-\leftskip
+ \hb@xt@\leftskip{%
+ \hss\@makefnmark\ %
+ }%
+ #1%
+ \par
+}%
+\prepdef\appendix{%
+ \par
+ \let\@hangfrom@section\@hangfrom@appendix
+ \let\@sectioncntformat\@appendixcntformat
+}%
+\def\@hangfrom@appendix#1#2#3{%
+ #1%
+ \@if@empty{#2}{%
+ #3%
+ }{%
+ #2\@if@empty{#3}{}{:\ #3}%
+ }%
+}%
+\def\@hangfroms@appendix#1#2{%
+ #1#2%
+}%
+\def\@appendixcntformat#1{\appendixname\ \csname the#1\endcsname}%
+ \def\pre@bibdata{\jobname\bibdata@app}%
+\def\refname{References}%
+\def\rtx@bibsection{%
+ \@ifx@empty\refname{%
+ \par\vspace{6\p@ plus 6\p@}%
+ }{%
+ \expandafter\section\expandafter*\expandafter{\refname}%
+ \@nobreaktrue
+ }%
+}%
+\let\bibpreamble\@empty
+\appdef\setup@hook{%
+ \bibsep\z@\relax
+}%
+\def\newblock{\ }%
+\appdef\setup@hook{%
+ \def\bibfont{%
+ \preprintsty@sw{}{\footnotesize}%
+ \@clubpenalty\clubpenalty
+ \labelsep\z@
+ }%
+}%
+\let\place@bibnumber\place@bibnumber@sup
+\newenvironment{theindex}{%
+ \columnseprule \z@
+ \columnsep 35\p@
+ \c@secnumdepth-\maxdimen
+ \onecolumngrid@push
+ \section{\indexname}%
+ \thispagestyle{plain}%
+ \parindent\z@
+ \parskip\z@ plus.3\p@\relax
+ \let\item\@idxitem
+ \onecolumngrid@pop
+}{%
+}%
+\def\@idxitem{\par\hangindent 40\p@}
+\def\subitem{\par\hangindent 40\p@ \hspace*{20\p@}}
+\def\subsubitem{\par\hangindent 40\p@ \hspace*{30\p@}}
+\def\indexspace{\par \vskip 10\p@ plus5\p@ minus3\p@\relax}
+\expandafter\def\csname rtx@aapm10pt\endcsname{%
+ \let\@currname@class\@currname
+ \def\@currname{aps10pt\substyle@post}%
+ \class@info{Reading file \@currname.\substyle@ext}%
+ \input{\@currname.\substyle@ext}%
+ \let\@currname\@currname@class
+ \class@info{Overriding 10pt}%
+ \aapmreprint
+}%
+\expandafter\def\csname rtx@aapm11pt\endcsname{\csname rtx@aapm12pt\endcsname}%
+\expandafter\def\csname rtx@aapm12pt\endcsname{%
+ \let\@currname@class\@currname
+ \def\@currname{aps12pt\substyle@post}%
+ \class@info{Reading file \@currname.\substyle@ext}%
+ \input{\@currname.\substyle@ext}%
+ \let\@currname\@currname@class
+ \class@info{Overriding 12pt}%
+ \aapmpreprint
+}%
+\def\today{%
+ \number\day\space
+ \ifcase\month
+ \or January\or February\or March\or April\or May\or June%
+ \or July\or August\or September\or October\or November\or December%
+ \fi\space
+ \number\year
+}%
+\@booleantrue\groupauthors@sw
+\@booleanfalse\@affils@sw
+\@booleantrue\runinaddress@sw
+\def\@journal@default{mph}%
+\def\@pointsize@default{12}%
+ \appdef\setup@hook{%
+ \preprintsty@sw{}{%
+ \def\normalsize{%
+ \@setfontsize\normalsize\@xpt{12}%
+ \abovedisplayskip 6\p@ plus2\p@ minus5\p@
+ \belowdisplayskip \abovedisplayskip
+ \abovedisplayshortskip \abovedisplayskip
+ \belowdisplayshortskip \abovedisplayskip
+ \let\@listi\@listI
+ }%
+ \def\small{%
+ \@setfontsize\small\@ixpt{11}%
+ \abovedisplayskip 8.5\p@ \@plus3\p@ \@minus4\p@
+ \belowdisplayskip \abovedisplayskip
+ \abovedisplayshortskip \z@ \@plus2\p@
+ \belowdisplayshortskip 4\p@ \@plus2\p@ \@minus2\p@
+ \def\@listi{%
+ \leftmargin\leftmargini
+ \topsep 4\p@ \@plus2\p@ \@minus2\p@
+ \parsep 2\p@ \@plus\p@ \@minus\p@
+ \itemsep \parsep
+ }%
+ }%
+ \def\Large{%
+ \@setfontsize\Large\@xivpt{16pt}%
+ }%
+ \def\@listI{%
+ \leftmargin\leftmargini
+ \parsep 4\p@ plus2\p@ minus\p@
+ \topsep 8\p@ plus2\p@ minus4\p@
+ \itemsep\z@
+ }%
+ \textheight = 694.0\p@
+ }%
+ }%
+\def\rtx@aapmmph{%
+ \typeout{Using journal substyle \@journal.}%
+ \@booleanfalse\authoryear@sw%
+ \input{fleqn.clo}%
+ \PassOptionsToPackage{fleqn}{amsmath}%
+ \AtBeginDocument{\mathindent12pt\relax}%
+}%
+\appdef\setup@hook{%
+ \bibpunct{}{}{,}{s}{}{\textsuperscript{,}}%
+ \let\onlinecite\rev@citealpnum
+}%
+\@booleantrue\footinbib@sw
+\let\place@bibnumber\place@bibnumber@sup
+\appdef\setup@hook{%
+ \footinbib@sw{}{%
+ \class@warn{Citations are superscript numbers: footnotes must be endnotes; changing to that configuration}%
+ \@booleantrue\footinbib@sw
+ }%
+}%
+\def\present@bibnote#1#2{%
+ \item[%
+ \textsuperscript{%
+ \normalfont%
+ \Hy@raisedlink{\hyper@anchorstart{frontmatter.#1}\hyper@anchorend}%
+ \begingroup%
+ \csname c@\@mpfn\endcsname#1\relax%
+ \frontmatter@thefootnote%
+ )\endgroup%
+ }%
+ ]#2\par%
+}%
+\def\make@footnote@endnote{%
+ \footinbib@sw{%
+ \ltx@footnote@push
+ \def\thempfn{Note\thefootnote}%
+ \let\ltx@footmark\rev@citemark
+ \let\ltx@foottext\rev@endtext
+ \appdef\class@enddocumenthook{\auto@bib}%
+ \let\printendnotes\relax
+ }{}%
+}%
+\def\aapmreprint{%
+}%
+\def\aapmpreprint{%
+}%
+% \end{macrocode}
+%
+%
+% \begin{macrocode}
+%</aapm>
+% \end{macrocode}
+%
+%
+%
+% \section{SOR: The Socity Journals Style (\texttt{jor} substyle)}
+%
+% The file \file{sor.rtx} is read in by the \classname{revtex4} document class
+% for applying the journal specific changes. User should use \texttt{[sor,jor]}
+% for applying the required updates.
+%
+%
+% \begin{macrocode}
+%<*sor>
+% \end{macrocode}
+%
+% Protect this file from being read in by anything but \revtex.
+% \begin{macrocode}
+\ifx\undefined\substyle@ext
+ \def\@tempa{%
+ \endinput
+ \GenericWarning{I must be read in by REVTeX! (Bailing out)}%
+ }%
+ \expandafter\else
+ \def\@tempa{}%
+ \expandafter\fi\@tempa
+ \class@info{RevTeX society AIP selected}%
+%%
+\DeclareOption{jor}{\change@journal{jor}}%
+\def\adv{AIP Advances}%
+\def\ao{Appl.\ Opt.}%
+\def\ap{Appl.\ Phys.}%
+\def\apl{Appl.\ Phys.\ Lett.}%
+\def\apm{Appl.\ Phys.\ Lett.\ Mater.}%
+\def\apj{Astrophys.\ J.}%
+\def\bell{Bell Syst.\ Tech.\ J.}%
+\def\bmf{Biomicrofluidics}%
+\def\cha{Chaos}%
+\def\jqe{IEEE J.\ Quantum Electron.}%
+\def\assp{IEEE Trans.\ Acoust.\ Speech Signal Process.}%
+\def\aprop{IEEE Trans.\ Antennas Propag.}%
+\def\mtt{IEEE Trans.\ Microwave Theory Tech.}%
+\def\iovs{Invest.\ Ophthalmol.\ Vis.\ Sci.}%
+\def\jcp{J.\ Chem.\ Phys.}%
+\def\jap{J.\Appl.\Phys.}%
+\def\jmp{j.\Math.\Phys.}%
+\def\jmo{J.\ Mod.\ Opt.}%
+\def\josa{J.\ Opt.\ Soc.\ Am.}%
+\def\josaa{J.\ Opt.\ Soc.\ Am.\ A}%
+\def\josab{J.\ Opt.\ Soc.\ Am.\ B}%
+\def\jpp{J.\ Phys.\ (Paris)}%
+\def\jpr{j.\Phys.\Chem.\Ref.\Data}%
+\def\ltp{Low.\Temp.\Phys.}%
+\def\nat{Nature (London)}%
+\def\oc{Opt.\ Commun.}%
+\def\ol{Opt.\ Lett.}%
+\def\pl{Phys.\ Lett.}%
+\def\pop{Phys.\Plasmas}%
+\def\pof{Phys.\Fluids}%
+\def\pra{Phys.\ Rev.\ A}%
+\def\prb{Phys.\ Rev.\ B}%
+\def\prc{Phys.\ Rev.\ C}%
+\def\prd{Phys.\ Rev.\ D}%
+\def\pre{Phys.\ Rev.\ E}%
+\def\prl{Phys.\ Rev.\ Lett.}%
+\def\rmp{Rev.\ Mod.\ Phys.}%
+\def\rsi{Rev.\Sci.\Instrum.}%
+\def\rse{J. \Renewable Sustainable Energy}%
+\def\pspie{Proc.\ Soc.\ Photo-Opt.\ Instrum.\ Eng.}%
+\def\sjqe{Sov.\ J.\ Quantum Electron.}%
+\def\vr{Vision Res.}%
+\def\sd{Structural Dynamics}%
+\def\jor{J.\ Rheol.}%
+\def\cp{AIP\ Conference\ Proceedings}%
+%%
+\DeclareOption{author-numerical}{%
+ \@booleantrue\authoryear@sw
+ \@booleantrue\authornum@sw
+}%
+\DeclareOption{article-title}{%
+ \@booleanfalse\sor@jtitx@sw
+}%
+\@booleantrue \sor@jtitx@sw
+\@booleanfalse\authoryear@sw
+\@booleanfalse\authornum@sw
+\@booleanfalse\onecolumn@sw
+\@booleanfalse\newonecolumn@sw
+\appdef\@bibdataout@rev{\@bibdataout@sor}%
+\def\@bibdataout@sor{%
+ \immediate\write\@bibdataout{%
+ @CONTROL{%
+ sor41Control%% TeXSupport: aip
+ \longbibliography@sw{\true@sw}{\sor@jtitx@sw{\false@sw}{\true@sw}}%
+ {%
+ ,pages="1",title="0"%
+ }{%
+ ,pages="0",title=""%
+ }%
+ }%
+ }%
+ \if@filesw
+ \immediate\write\@auxout{\string\citation{sor41Control}}%% TeXSupport: aip
+ \fi
+}%
+\@booleantrue\preprintsty@sw
+\@booleantrue\showPACS@sw
+\@booleantrue\showKEYS@sw
+%% TeXSupport
+\let\old@refname\refname
+\appdef\setup@hook{%
+ \preprintsty@sw{}{%
+ \let\refname\@empty
+ }%
+}%
+\appdef\setup@hook{%
+ \preprintsty@sw{%
+ \ps@preprint
+ }{%
+ \ps@article
+ }%
+}%
+\def\ps@preprint{%
+ \def\@oddhead{\@runningtitle\hfil}%
+ \def\@evenhead{\@runningtitle\hfil}%
+ \def\@oddfoot{\hfil\thepage\quad\checkindate\hfil}%
+ \def\@evenfoot{\hfil\thepage\quad\checkindate\hfil}%
+ \let\@mkboth\@gobbletwo
+ \let\sectionmark\@gobble
+ \let\subsectionmark\@gobble
+}%
+\def\ps@article{%
+ \def\@evenhead{\let\\\heading@cr\thepage\quad\checkindate\hfil\@runningtitle}%
+ \def\@oddhead{\let\\\heading@cr\@runningtitle\hfil\checkindate\quad\thepage}%
+ \def\@oddfoot{}%
+ \def\@evenfoot{}%
+ \let\@mkboth\@gobbletwo
+ \let\sectionmark\@gobble
+ \let\subsectionmark\@gobble
+}%
+\def\@runningtitle{\@shorttitle}%
+\renewenvironment{titlepage}{%
+ \let\wastwocol@sw\twocolumn@sw
+ \onecolumngrid
+ \newpage
+ \thispagestyle{titlepage}%
+ \c@page\z@% article sets this to one not zero???
+}{%
+ \wastwocol@sw{\twocolumngrid}{\newpage}%
+}%
+\let\@fnsymbol@latex\@fnsymbol
+\let\@fnsymbol\@alph
+\def\adjust@abstractwidth{%
+ \parindent1em\relax
+ \advance\leftskip.5in\relax
+ \@totalleftmargin\leftskip
+ \@afterheading\@afterindentfalse
+}%
+\def\frontmatter@abstractheading{}%
+\def\frontmatter@abstractfont{%
+ \adjust@abstractwidth
+}%
+\appdef\setup@hook{%
+ \preprintsty@sw{%
+ \@booleantrue\titlepage@sw
+ \let\section\section@preprintsty
+ \let\subsection\subsection@preprintsty
+ \let\subsubsection\subsubsection@preprintsty
+ }{}%
+}%
+\def\frontmatter@@indent{%
+ \skip@\@flushglue
+ \@flushglue\z@ plus.3\hsize\relax
+ \raggedright
+ \advance\leftskip.5in\relax
+ \@totalleftmargin\leftskip
+ \@flushglue\skip@
+}%
+\def\frontmatter@authorformat{%
+ \frontmatter@@indent
+ \sffamily
+}%
+\renewcommand*\email[1][Electronic mail: ]{\begingroup\sanitize@url\@email{#1}}%
+\def\frontmatter@above@affilgroup{%
+}%
+\def\frontmatter@above@affiliation@script{%
+ \frontmatter@@indent
+}%
+\def\frontmatter@above@affiliation{%
+}%
+\def\frontmatter@affiliationfont{%
+ \frontmatter@@indent
+ \preprintsty@sw{}{\small}%
+ \it
+}%
+\def\frontmatter@collaboration@above{%
+}%
+\def\frontmatter@setup{%
+ \normalfont
+}%
+\def\frontmatter@title@above{\addvspace{6\p@}}%
+\def\frontmatter@title@format{%
+ \preprintsty@sw{}{\Large}%
+ \sffamily
+ \bfseries
+ \raggedright
+ \parskip\z@skip
+}%
+\def\frontmatter@title@below{\addvspace{3\p@}}%
+\def\@author@parskip{3\p@}%
+\@booleantrue\altaffilletter@sw
+\def\frontmatter@makefnmark{%
+ \@textsuperscript{%
+ \normalfont\@thefnmark%(
+ )%
+ }%
+}%
+\def\frontmatter@authorbelow{%
+\addvspace{3\p@}%
+}%
+\let\affil@cutoff\tw@
+\def\frontmatter@RRAP@format{%
+ \addvspace{5\p@}%
+ \small
+ \raggedright
+ \advance\leftskip.5in\relax
+ \@totalleftmargin\leftskip
+ \everypar{%
+ \hbox\bgroup(\@gobble@leavemode@uppercase%)
+ }%
+ \def\par{%
+ \@ifvmode{}{%(
+ \unskip)\egroup\@@par
+ }%
+ }%
+}%
+\def\punct@RRAP{;\egroup\ \hbox\bgroup}%
+\def\@gobble@leavemode@uppercase#1#2{\expandafter\MakeTextUppercase}%
+\def\frontmatter@PACS@format{%
+ \addvspace{11\p@}%
+ \adjust@abstractwidth
+ \parskip\z@skip
+ \samepage
+}%
+\def\frontmatter@keys@format{%
+ \adjust@abstractwidth
+ \samepage
+}%
+\def\ps@titlepage{%
+ \def\@oddhead{%
+ \@runningtitle
+ \hfill
+ \produce@preprints\@preprint
+ }%
+ \let\@evenhead\@oddhead
+ \def\@oddfoot{%
+ \hb@xt@\z@{\byrevtex\hss}%
+ \hfil
+ \preprintsty@sw{\thepage}{}%
+ \quad\checkindate
+ \hfil
+ }%
+ \let\@evenfoot\@oddfoot
+}%
+\def\byrevtex{\byrevtex@sw{Typeset by REV\TeX and AIP}{}}%
+\def\produce@preprints#1{%
+ \preprint@sw{%
+ \vtop to \z@{%
+ \def\baselinestretch{1}%
+ \small
+ \let\preprint\preprint@count
+ \count@\z@#1\@ifnum{\count@>\tw@}{%
+ \hbox{%
+ \let\preprint\preprint@hlist
+ #1\setbox\z@\lastbox
+ }%
+ }{%
+ \let\preprint\preprint@cr
+ \halign{\hfil##\cr#1\crcr}%
+ \par
+ \vss
+ }%
+ }%
+ }{}%
+}%
+\def\preprint@cr#1{#1\cr}%
+\def\preprint@count#1{\advance\count@\@ne}%
+\def\preprint@hlist#1{#1\hbox{, }}%
+\newenvironment{Lead@inParagraph}{%
+ \par
+ \bfseries
+ \@afterheading\@afterindentfalse
+}{%
+ \par
+ \hb@xt@\hsize{\hfil\leaders\hrule\hfil\leaders\hrule\hfil\hfil}%
+}%
+\appdef\frontmatter@init{%
+ \let@environment{quotation@ltx}{quotation}%
+ \let@environment{quotation}{Lead@inParagraph}%
+}%
+\appdef\@startsection@hook{%
+ \let@environment{quotation}{quotation@ltx}%
+}%
+\def\@seccntformat#1{\csname the#1\endcsname.\quad}%
+\def\@hang@from#1#2#3{#1#2#3}%
+\def\section{%
+ \@startsection
+ {section}%
+ {1}%
+ {\z@}%
+ {0.8cm \@plus1ex \@minus .2ex}%
+ {0.5cm}%
+ {%
+ \normalfont
+ \small
+ \sffamily
+ \bfseries
+ \raggedright
+ }%
+}%
+\def\@hangfrom@section#1#2#3{\@hangfrom{#1#2}\MakeTextUppercase{#3}}%
+\def\@hangfroms@section#1#2{#1\MakeTextUppercase{#2}}%
+\def\subsection{%
+ \@startsection
+ {subsection}%
+ {2}%
+ {\z@}%
+ {.8cm \@plus1ex \@minus .2ex}%
+ {.5cm}%
+ {%
+ \normalfont
+ \small
+ \sffamily
+ \bfseries
+ \raggedright
+ }%
+}%
+\def\subsubsection{%
+ \@startsection
+ {subsubsection}%
+ {3}%
+ {\z@}%
+ {.8cm \@plus1ex \@minus .2ex}%
+ {.5cm}%
+ {%
+ \normalfont
+ \small
+ \sffamily
+ \bfseries
+ \itshape
+ \raggedright
+ }%
+}%
+\def\paragraph{%
+ \@startsection
+ {paragraph}%
+ {4}%
+ {\parindent}%
+ {\z@}%
+ {-1em}%
+ {\normalfont\normalsize\itshape}%
+}%
+\def\subparagraph{%
+ \@startsection
+ {subparagraph}%
+ {5}%
+ {\parindent}%
+ {3.25ex \@plus1ex \@minus .2ex}%
+ {-1em}%
+ {\normalfont\normalsize\bfseries}%
+}%
+\def\section@preprintsty{%
+ \@startsection
+ {section}%
+ {1}%
+ {\z@}%
+ {0.8cm \@plus1ex \@minus .2ex}%
+ {0.5cm}%
+ {%
+ \normalfont
+ \bfseries
+ \raggedright
+ }%
+}%
+\def\subsection@preprintsty{%
+ \@startsection
+ {subsection}%
+ {2}%
+ {\z@}%
+ {.8cm \@plus1ex \@minus .2ex}%
+ {.5cm}%
+ {%
+ \normalfont
+ \bfseries
+ \raggedright
+ }%
+}%
+\def\subsubsection@preprintsty{%
+ \@startsection
+ {subsubsection}%
+ {3}%
+ {\z@}%
+ {.8cm \@plus1ex \@minus .2ex}%
+ {.5cm}%
+ {%
+ \normalfont
+ \itshape\bfseries
+ \raggedright
+ }%
+}%
+\let\frontmatter@footnote@produce\frontmatter@footnote@produce@footnote
+\def\@pnumwidth{1.55em}
+\def\@tocrmarg {2.55em}
+\def\@dotsep{2}
+\def\ltxu@dotsep{4.5pt}
+\setcounter{tocdepth}{3}
+\def\tableofcontents{%
+ \addtocontents{toc}{\string\tocdepth@munge}%
+ \print@toc{toc}%
+ \addtocontents{toc}{\string\tocdepth@restore}%
+}%
+\def\tocdepth@munge{%
+ \let\l@section@saved\l@section
+ \let\l@section\@gobble@tw@
+}%
+\def\@gobble@tw@#1#2{}%
+\def\tocdepth@restore{%
+ \let\l@section\l@section@saved
+}%
+\def\l@part#1#2{\addpenalty{\@secpenalty}%
+ \begingroup
+ \set@tocdim@pagenum{#2}%
+ \parindent \z@
+ \rightskip\tocleft@pagenum plus 1fil\relax
+ \skip@\parfillskip\parfillskip\z@
+ \addvspace{2.25em plus\p@}%
+ \large \bf %
+ \leavevmode\ignorespaces#1\unskip\nobreak\hskip\skip@
+ \hb@xt@\rightskip{\hfil\unhbox\z@}\hskip-\rightskip\hskip\z@skip
+ \par
+ \nobreak %
+ \endgroup
+}%
+\def\tocleft@{\z@}%
+\def\tocdim@min{5\p@}%
+\def\l@section{%
+ \l@@sections{}{section}% Implicit #3#4
+}%
+\def\l@f@section{%
+ \addpenalty{\@secpenalty}%
+ \addvspace{1.0em plus\p@}%
+ \bf
+}%
+\def\l@subsection{%
+ \l@@sections{section}{subsection}% Implicit #3#4
+}%
+\def\l@subsubsection{%
+ \l@@sections{subsection}{subsubsection}% Implicit #3#4
+}%
+\def\l@paragraph#1#2{}%
+\def\l@subparagraph#1#2{}%
+\let\toc@pre\toc@pre@auto
+\let\toc@post\toc@post@auto
+\def\listoffigures{\print@toc{lof}}%
+\def\l@figure{\@dottedtocline{1}{1.5em}{2.3em}}
+\def\listoftables{\print@toc{lot}}%
+\let\l@table\l@figure
+\@booleanfalse\raggedcolumn@sw
+\def\tableft@skip@float{\z@ plus\hsize}%
+\def\tabmid@skip@float{\@flushglue}%
+\def\tabright@skip@float{\z@ plus\hsize}%
+\def\array@row@pre@float{\hline\hline\noalign{\vskip\doublerulesep}}%
+\def\array@row@pst@float{\noalign{\vskip\doublerulesep}\hline\hline}%
+\def\@makefntext#1{%
+ \def\baselinestretch{1}%
+ \leftskip1em%
+ \parindent1em%
+ \noindent
+ \nobreak\hskip-\leftskip
+ \hb@xt@\leftskip{%
+ \hss\@makefnmark\ %
+ }%
+ #1%
+ \par
+}%
+\prepdef\appendix{%
+ \par
+ \let\@hangfrom@section\@hangfrom@appendix
+ \let\@sectioncntformat\@appendixcntformat
+}%
+\def\@hangfrom@appendix#1#2#3{%
+ #1%
+ \@if@empty{#2}{%
+ #3%
+ }{%
+ #2\@if@empty{#3}{}{:\ #3}%
+ }%
+}%
+\def\@hangfroms@appendix#1#2{%
+ #1#2%
+}%
+\def\@appendixcntformat#1{\appendixname\ \csname the#1\endcsname}%
+ \def\pre@bibdata{\jobname\bibdata@app}%
+\def\refname{References}%
+\def\rtx@bibsection{%
+ \@ifx@empty\refname{%
+ \par\vspace{6\p@ plus 6\p@}%
+ }{%
+ \expandafter\section\expandafter*\expandafter{\refname}%
+ \@nobreaktrue
+ }%
+}%
+\let\bibpreamble\@empty
+\appdef\setup@hook{%
+ \bibsep\z@\relax
+}%
+\def\newblock{\ }%
+\appdef\setup@hook{%
+ \def\bibfont{%
+ \preprintsty@sw{}{\footnotesize}%
+ \@clubpenalty\clubpenalty
+ \labelsep\z@
+ }%
+}%
+\let\place@bibnumber\place@bibnumber@sup
+\newenvironment{theindex}{%
+ \columnseprule \z@
+ \columnsep 35\p@
+ \c@secnumdepth-\maxdimen
+ \onecolumngrid@push
+ \section{\indexname}%
+ \thispagestyle{plain}%
+ \parindent\z@
+ \parskip\z@ plus.3\p@\relax
+ \let\item\@idxitem
+ \onecolumngrid@pop
+}{%
+}%
+\def\@idxitem{\par\hangindent 40\p@}
+\def\subitem{\par\hangindent 40\p@ \hspace*{20\p@}}
+\def\subsubitem{\par\hangindent 40\p@ \hspace*{30\p@}}
+\def\indexspace{\par \vskip 10\p@ plus5\p@ minus3\p@\relax}
+\expandafter\def\csname rtx@sor10pt\endcsname{%% TeXSupport: aip
+ \let\@currname@class\@currname
+ \def\@currname{aps10pt\substyle@post}%
+ \class@info{Reading file \@currname.\substyle@ext}%
+ \input{\@currname.\substyle@ext}%
+ \let\@currname\@currname@class
+ \class@info{Overriding 10pt}%
+ \sorreprint
+}%
+\expandafter\def\csname rtx@sor11pt\endcsname{\csname rtx@sor12pt\endcsname}%
+\expandafter\def\csname rtx@sor12pt\endcsname{%
+ \let\@currname@class\@currname
+ \def\@currname{aps12pt\substyle@post}%
+ \class@info{Reading file \@currname.\substyle@ext}%
+ \input{\@currname.\substyle@ext}%
+ \let\@currname\@currname@class
+ \class@info{Overriding 12pt}%
+ \sorpreprint
+}%
+\def\today{%
+ \number\day\space
+ \ifcase\month
+ \or January\or February\or March\or April\or May\or June%
+ \or July\or August\or September\or October\or November\or December%
+ \fi\space
+ \number\year
+}%
+ \clo@superscriptaddress
+\def\@journal@default{cha}%
+\def\@pointsize@default{12}%
+\@booleanfalse\pagerestrict@sw%
+%%
+\def\rtx@sorjor{%
+ \typeout{Using journal substyle \@journal.}%
+ \@booleantrue\authoryear@sw%
+ \@booleantrue\twoside@sw\@mparswitchfalse%
+}%
+%%
+\@booleantrue\footinbib@sw
+\let\old@place@bibnumber\place@bibnumber
+\let\place@bibnumber\place@bibnumber@sup
+\appdef\setup@hook{%
+ \authoryear@sw{%
+ \sor@jtitx@sw{%
+ \def\@bibstyle{aipauth\substyle@post}%
+ }{%
+ \def\@bibstyle{aipauth\substyle@post}%
+ }%
+ \authornum@sw{%
+ \bibpunct{}{}{,}{s}{}{\textsuperscript{,}}%
+ \let\onlinecite\rev@citealpnum
+ }{%
+ \bibhang10\p@
+ \bibpunct{(%)
+ }{%(
+ )}{; }{a}{,}{,}%
+ \@booleanfalse\footinbib@sw
+ \let\NAT@mcite\@ne
+ \let\NAT@sort\z@
+ \def\NAT@cmprs{\z@}%
+ \let\NAT@def@citea\rtx@def@citea
+ \let\NAT@def@citea@close\rtx@def@citea@close
+ }%
+ }{%
+ \sor@jtitx@sw{%
+ \def\@bibstyle{sornum\substyle@post}%
+ }{%
+ \def\@bibstyle{sornum\substyle@post}%
+ }%
+ \bibpunct{}{}{,}{s}{}{\textsuperscript{,}}%
+ \let\onlinecite\rev@citealpnum
+ }%
+}%
+\def\make@footnote@endnote{%
+ \footinbib@sw{%
+ \authoryear@sw{\authornum@sw{\false@sw}{\true@sw}}{\false@sw}%
+ {}{%
+ \ltx@footnote@push
+ \def\thempfn{Note\thefootnote}%
+ \let\ltx@footmark\rev@citemark
+ \let\ltx@foottext\rev@endtext
+ \appdef\class@enddocumenthook{\auto@bib}%
+ \let\printendnotes\relax
+ }%
+ }{}%
+}%
+\def\sorreprint{%
+}%
+\def\sorpreprint{%
+}%
+%%
+\xdef\t@talAU{0}% TeXSupport
+\def\frontmatter@author@produce@script{%
+ \begingroup
+ \let\@author@present\@author@present@script
+ \frontmatterverbose@sw{\typeout{\string\frontmatter@author@produce@script:}\say\@AAC@list\say\@AFF@list\say\@AFG@list}{}%
+ \let\AU@temp\@empty
+ \@tempcnta\z@
+ \let\AF@opr \@gobble
+ \def\AU@opr{\@author@count\@tempcnta}%
+ \def\CO@opr{\@collaboration@count\AU@temp\@tempcnta}%
+ \@AAC@list%
+ \xdef\t@talAU{\the\@tempcnta}% TeXSupport
+ \expandafter\CO@opr\@author@cleared
+ \begingroup
+ \frontmatter@authorformat
+ \let\AF@opr \@affilID@def
+ \let\AU@opr \@author@present
+ \def\CO@opr{\@collaboration@present\AU@temp}%
+ \set@listcomma@list\AU@temp
+ \@AAC@list
+ \unskip\unskip
+ \par
+ \endgroup
+ \begingroup
+ \frontmatter@above@affiliation@script
+ \let\AFF@opr \@affil@script
+ \@AFF@list
+ \frontmatter@footnote@produce
+ \par
+ \endgroup
+ \endgroup
+}%
+\def\doauthor#1#2#3{% TeXSupport
+ \ignorespaces#1\unskip\@listcomma
+ \begingroup
+ \ifnum\t@talAU=1\else\ifnum\c@affil=\@ne\relax\else#3\fi\fi% TeXSupport
+ \@if@empty{#2}{\endgroup{}{}}{\endgroup{\ifnum\t@talAU=1\else\ifnum\c@affil=\@ne\relax\else\comma@space\fi\fi}{}\frontmatter@footnote{#2}}% TeXSupport
+ \space \@listand
+}%
+%%
+\appdef\rtx@require@packages{%
+%%
+%%
+%% Journal of Rheology (jor)
+%%
+\def\jnl@jor{jor}%
+\preprintsty@sw{}%
+ {\ifx\@journal\jnl@jor%
+ \typeout{SOR Info: \@journal\space journal style Single column, 2013/10/24}%
+ \@booleanfalse\twocolumn@sw%
+ \appdef\setup@hook{%
+ \twoside@sw{%
+ \oddsidemargin 28pt%
+ \evensidemargin 28pt%
+ \marginparwidth 30pt%
+ }{%
+ \oddsidemargin 28pt
+ \evensidemargin 0pt
+ \marginparwidth 44pt
+ }%
+ }%
+ \marginparsep 10pt
+ \topmargin -17pt
+ \headheight 12pt
+ \headsep 25pt
+ \topskip 10pt
+ \splittopskip\topskip
+ \footskip 30pt
+ \textheight=53.5pc
+ \textwidth 33pc
+ \columnsep 10pt
+ \def\title@column#1{%
+ \minipagefootnote@init
+ \begingroup
+ \let\@footnotetext\frontmatter@footnotetext
+ \ltx@no@footnote
+ #1%
+ \endgroup
+ \minipagefootnote@foot
+ }%
+\def\frontmatter@title@format{%
+ \preprintsty@sw{}{\Large}%
+ \sffamily%
+ \bfseries%
+ \leftskip0pt plus1fill%
+ \rightskip0pt plus1fill%
+ \parindent\z@%
+%% \raggedright%
+ \parskip\z@skip%
+}%
+\def\frontmatter@@indent{%
+ \skip@\@flushglue
+ \@flushglue\z@ plus.3\hsize\relax
+ \leftskip0pt plus1fill%
+ \rightskip0pt plus1fill%
+ \parindent\z@%
+%% \raggedright
+%% \advance\leftskip.5in\relax
+ \@totalleftmargin\leftskip
+ \@flushglue\skip@
+}%
+%% \def\frontmatter@authorformat{%
+%% \frontmatter@@indent
+%% \sffamily
+%% }%
+\def\frontmatter@RRAP@format{%
+ \addvspace{5\p@}%
+ \small
+ \leftskip0pt plus1fill%
+ \rightskip0pt plus1fill%
+ \parindent\z@%
+%%\raggedright
+%%\advance\leftskip.5in\relax
+ \@totalleftmargin\leftskip
+ \everypar{%
+ \hbox\bgroup(\@gobble@leavemode@uppercase%)
+ }%
+ \def\par{%
+ \@ifvmode{}{%(
+ \unskip)\egroup\@@par
+ }%
+ }%
+}%
+\def\frontmatter@abstractfont{}%
+\def\frontmatter@abstractwidth{\textwidth}
+\def\abstractname{Synopsis}
+\def\frontmatter@abstractheading{%
+ \begingroup
+ \centering\large
+ {\bfseries\abstractname}
+ \par\vskip.25\baselineskip
+ \endgroup
+}%
+\appdef\setup@hook{%
+ \preprintsty@sw{}{%
+ \let\refname\old@refname%\@empty
+ }%
+}%
+%
+\def\ps@article{%
+ \def\@evenhead{\let\\\heading@cr\sffamily\thepage\quad\checkindate\hfil\@runningtitle\hfil}%
+ \def\@oddhead{\let\\\heading@cr\hfil\sffamily\@runningtitle\hfil\checkindate\quad\thepage}%
+ \def\@oddfoot{}%
+ \def\@evenfoot{}%
+ \let\@mkboth\@gobbletwo
+ \let\sectionmark\@gobble
+ \let\subsectionmark\@gobble
+}%
+\def\@make@capt@title#1#2{%
+ \@ifx@empty\float@link{\@firstofone}{\expandafter\href\expandafter{\float@link}}%
+ {{\bfseries#1}}\@caption@fignum@sep#2%
+}%
+\def\@caption@fignum@sep{{\bfseries.} }%
+%%
+ \fi%
+}
+}%
+% \end{macrocode}
+%
+%
+% \begin{macrocode}
+%</sor>
+% \end{macrocode}
+%
+%
+%
+%
+% \Finale
+% %Here ends the programmer's documentation.
+% \endinput
+%
+\endinput
diff --git a/texmf/source/latex/revtex/ltxdocext.dtx b/texmf/source/latex/revtex/ltxdocext.dtx
new file mode 100644
index 0000000..d3d8b77
--- /dev/null
+++ b/texmf/source/latex/revtex/ltxdocext.dtx
@@ -0,0 +1,1259 @@
+% \iffalse meta-comment balanced on line 108
+% ltxdocext.dtx: package to extend the ltxdoc class of standard LaTeX
+% Copyright (c) 2008 Arthur Ogawa
+%
+% Disclaimer
+% This file is distributed WITHOUT ANY WARRANTY;
+% without even the implied warranty of
+% MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
+% License
+% You may distribute this file under the conditions of the
+% LaTeX Project Public License 1.3c or later
+% (http://www.latex-project.org/lppl.txt).
+% ReadMe
+% For the documentation and more detailed instructions for
+% installation, typeset this document with \LaTeX.
+% Maintenance Status
+% This work has the LPPL maintenance status "maintained";
+% Current Maintainer of this work is Arthur Ogawa.
+%
+% This work consists of the main source file ltxdocext.dtx
+% and the derived files
+% ltxdocext.sty, ltxdocext.pdf, ltxdocext.ins, ltxdocext.drv.
+% Distribution:
+% CTAN:macros/latex/contrib/revtex/
+%
+% Unpacking:
+% tex ltxdocext.dtx
+%
+% Documentation:
+% latex ltxdocext.dtx; ...
+%
+% Program calls to get the documentation (example):
+% pdflatex ltxdocext.dtx
+% makeindex -s gind.ist ltxdocext.idx
+% makeindex -s gglo.ist -o ltxdocext.gls ltxdocext.glo
+% pdflatex ltxdocext.dtx
+% makeindex -s gind.ist ltxdocext.idx
+% pdflatex ltxdocext.dtx
+%
+% Installation:
+% TDS:doc/latex/revtex/
+% TDS:source/latex/revtex/
+% TDS:tex/latex/revtex/
+%
+% Thanks, Heiko!
+% This method of letting a single .dtx file serve as both
+% documentation (via latex) and installer (via tex) follows
+% the example of Heiko Oberdiek. Thanks!
+%<*ignore>
+\begingroup
+ \def\x{LaTeX2e}%
+\expandafter\endgroup
+\ifcase
+ 0\expandafter\ifx\csname processbatchFile\endcsname\relax\else1\fi\ifx\fmtname\x\else 1\fi
+ \relax
+\else
+ \csname fi\endcsname
+%</ignore>
+%<*install>
+\input docstrip
+\preamble
+
+This is a generated file;
+altering it directly is inadvisable;
+instead, modify the original source file.
+See the URL in the file README.
+
+License
+ You may distribute this file under the conditions of the
+ LaTeX Project Public License 1.3c or later
+ (http://www.latex-project.org/lppl.txt).
+
+ This file is distributed WITHOUT ANY WARRANTY;
+ without even the implied warranty of MERCHANTABILITY
+ or FITNESS FOR A PARTICULAR PURPOSE.
+
+\endpreamble
+\askforoverwritefalse
+\keepsilent
+ \generate{%
+ %{ignore}
+%\usedir{tex/latex/revtex}%
+ \file{ltxdocext.sty}{%
+ \from{ltxdocext.dtx}{package,kernel}%
+ }%
+ % \file{acrofont.sty}{\from{ltxdocext.dtx}{fonts}}%
+ }%
+\ifToplevel{
+\Msg{***********************************************************}
+\Msg{*}
+\Msg{* To finish the installation, please move}
+\Msg{* ltxdocext.sty}
+\Msg{* into a directory searched by TeX;}
+\Msg{* in a TDS-compliant installation:}
+\Msg{* texmf/tex/macros/latex/revtex/.}
+\Msg{*}
+\Msg{* To produce the documentation,
+ run ltxdocext.dtx through LaTeX.}
+\Msg{*}
+\Msg{* Happy TeXing}
+\Msg{***********************************************************}
+}
+\endbatchfile
+%</install>
+%<*ignore>
+\fi
+%</ignore>
+% \fi
+%
+% \GetFileInfo{ltxdocext.dtx}
+%
+% \iffalse ltxdoc klootch
+%<*ltxdocext>
+%%% @LaTeX-file{
+%%% filename = "ltxdocext.dtx",
+%%% version = "1.0a",
+%%% date = "2018/12/26",
+%%% author = "Arthur Ogawa (mailto:arthur_ogawa at sbcglobal.net),
+%%% commissioned by the American Physical Society.
+%%% ",
+%%% copyright = "Copyright (C) 1999, 2008 Arthur Ogawa,
+%%% distributed under the terms of the
+%%% LaTeX Project Public License, see
+%%% ftp://ctan.tug.org/macros/latex/base/lppl.txt
+%%% ",
+%%% address = "Arthur Ogawa,
+%%% USA",
+%%% telephone = "",
+%%% FAX = "",
+%%% email = "mailto colon arthur_ogawa at sbcglobal.net",
+%%% codetable = "ISO/ASCII",
+%%% keywords = "latex, ltxdoc",
+%%% supported = "yes",
+%%% abstract = "extensions to the ltxdoc class",
+%%% }
+%</ltxdocext>
+% \fi
+%
+% \iffalse ltxdoc klootch
+% The following references the \file{README-LTXDOCEXT} file,
+% which contains basic information about this package.
+% The contents of this file are generated when
+% you typeset the programmer's documentation.
+% Search on "{filecontents*}{README-LTXDOCEXT}" to locate it.
+% \fi\input{README-LTXDOCEXT}%
+%
+% \subsection{Bill of Materials}
+%
+% Following is a list of the files in this distribution arranged
+% according to provenance.
+%
+% \subsubsection{Primary Source}%
+% One single file generates all.
+%\begin{verbatim}
+%ltxdocext.dtx
+%\end{verbatim}
+%
+% \subsubsection{Generated by \texttt{latex ltxdocext.dtx}}%
+% Typesetting the source file under pdflatex
+% generates the readme and the documentation.
+%\begin{verbatim}
+%README ltxdocext.pdf
+%\end{verbatim}
+%
+% \subsubsection{Generated by \texttt{tex ltxdocext.ins}}%
+% Typesetting this file with \TeX\ generates
+% the package file.
+%\begin{verbatim}
+%ltxdocext.sty acrofont.sty
+%\end{verbatim}
+%
+% \subsubsection{Documentation}%
+% The following are the online documentation:
+% \begin{verbatim}
+%ltxdocext.pdf
+% \end{verbatim}
+%
+% \subsubsection{Auxiliary}%
+% The following are auxiliary files generated
+% in the course of running \LaTeX:
+% \begin{verbatim}
+%ltxdocext.aux ltxdocext.idx ltxdocext.ind ltxdocext.log ltxdocext.toc
+% \end{verbatim}
+%
+% \section{Code common to all modules}%
+%
+% We want to require only one place in this file
+% where the version number is stated,
+% and we also want to ensure that the version
+% number is embedded into every generated file.
+%
+% Now we declare that
+% these files can only be used with \LaTeXe.
+% An appropriate message is displayed if
+% a different \TeX{} format is used.
+% \begin{macrocode}
+%<*driver|package|fonts>
+\NeedsTeXFormat{LaTeX2e}[1995/12/01]%
+%</driver|package|fonts>
+% \end{macrocode}
+% As desired, the following modules all
+% take common version information:
+% \begin{macrocode}
+%<package>\ProvidesFile{ltxdocext.sty}%
+%<fonts>\ProvidesFile{acrofont.sty}%
+%<*driver>
+\expandafter\ProvidesFile\expandafter{\jobname.dtx}%
+%</driver>
+% \end{macrocode}
+%
+% The following line contains, for once and for all,
+% the version and date information.
+% By various means, this information is reproduced
+% consistently in all generated files and in the
+% typeset documentation.
+% \begin{macrocode}
+%<version>
+ [2018/12/26 1.0a ltxdoc extensions package]% \fileversion
+% \end{macrocode}
+%
+%
+% \section{The driver module \texttt{doc}}
+%
+% This module, consisting of the present section,
+% typesets the programmer's documentation,
+% generating the \file{README-LTXDOCEXT} as required.
+%
+% Because the only uncommented-out lines of code at the beginning of
+% this file constitute the \file{doc} module itself,
+% we can simply typeset the \file{.dtx} file directly,
+% and there is thus rarely any need to
+% generate the ``doc'' {\sc docstrip} module.
+% Module delimiters are nonetheless required so that
+% this code does not find its way into the other modules.
+%
+% The \enve{document} command concludes the typesetting run.
+%
+% \begin{macrocode}
+%<*driver>
+% \end{macrocode}
+%
+% \subsection{The Preamble}
+% The programmers documentation is formatted
+% with the \classname{ltxdoc} document class, with local customizations,
+% and with the usual code line indexing.
+% \begin{macrocode}
+\documentclass[draft]{ltxdoc}
+\RequirePackage{ltxdocext}%
+\RequirePackage[colorlinks=true,linkcolor=blue]{hyperref}%
+%\expandafter\ifx\csname package@font\endcsname\@undefined\else
+% \expandafter\RequirePackage\expandafter{\csname package@font\endcsname}%
+%\fi
+\CodelineIndex\EnableCrossrefs % makeindex -s gind.ist ltxdocext
+\RecordChanges % makeindex -s gglo.ist -o ltxdocext.gls ltxdocext.glo
+% \end{macrocode}
+%
+% \subsubsection{Docstrip and info directives}
+% We use so many {\sc docstrip} modules that we set the
+% \texttt{StandardModuleDepth} counter to 1.
+% \begin{macrocode}
+\setcounter{StandardModuleDepth}{1}
+% \end{macrocode}
+% The following command retrieves the date and version information
+% from this file.
+% \begin{macrocode}
+\expandafter\GetFileInfo\expandafter{\jobname.dtx}%
+% \end{macrocode}
+%
+% \subsection{The ``Read Me'' File}
+% As promised above, here is the contents of the
+% ``Read Me'' file. That file serves a double purpose,
+% since it also constitutes the beginining of the
+% programmer's documentation. What better thing, after
+% all, to have appear at the beginning of the
+% typeset documentation?
+%
+% A good discussion of how to write a ReadMe file can be found in
+% Engst, Tonya, ``Writing a ReadMe File? Read This''
+% \emph{MacTech} October 1998, p. 58.
+%
+% Note the appearance of the
+% \cmd\StopEventually\ command, which marks the
+% dividing line between the user documentation
+% and the programmer documentation.
+%
+% The usual user will not be asked to
+% do a full build, not to speak
+% of the bootstrap.
+% Instructions for carrying these processes
+% begin the programmer's manual.
+%
+% \begin{macrocode}
+\begin{filecontents*}{README-LTXDOCEXT}
+\title{%
+ Extensions to the \classname{ltxdoc} class%
+ \thanks{%
+ This file has version number \fileversion,
+ last revised \filedate.%
+ }%
+ \thanks{%
+ Version \fileversion\ \copyright\ 2019 American Physical Society
+ }%
+}%
+\author{%
+Arthur Ogawa%
+ \thanks{\texttt{mailto:arthur\_ogawa at sbcglobal.net}}%
+}%
+%\iffalse
+% For version number and date,
+% search on "\fileversion" in the .dtx file,
+% or see the end of the README file.
+%\fi
+
+\maketitle
+
+This file embodies the \classname{ltxdocext} package,
+the implementation and its user documentation.
+
+The distribution point for this work is
+\url{journals.aps.org/revtex},
+which contains prebuilt runtime files, documentation, and full source,
+ready to add to a TDS-compliant \TeX\ installation.
+
+The \classname{ltxdocext} package was commissioned by the American Physical Society
+and is distributed under the terms of the \LaTeX\ Project Public License,
+the same license under which all the portions of \LaTeX\ itself are distributed.
+Please see \url{http://ctan.tug.org/macros/latex/base/lppl.txt} for details.
+
+To use this document class, you must have a working
+\TeX\ installation equipped with \LaTeXe\
+and possibly pdftex and Adobe Acrobat Reader or equivalent.
+
+To install, retrieve the distribution,
+unpack it into a directory on the target computer,
+and move the files \file{ltxdocext.sty} and \file{acrofont.sty}
+into a location in your filesystem where they will be found by \LaTeX.
+
+If you will be using the \classname{acrofont} package, you must
+also install the virtual fonts
+\file{zpsynocmrv}, \file{zptmnocmr},
+\file{zptmnocmrm}, and \file{zpzcnocmry}.
+The corresponding \file{.tfm}, \file{.vf}, and \file{.vpl}
+files are part of this distribution.
+
+To use, read the user documentation \file{ltxdocext.pdf}.
+The \file{.dtx} file, \file{ltxdocext.dtx}, constitutes
+in itself an instance of use of the \classname{ltxdocext}
+package and the \classname{acrofont} package.
+
+\tableofcontents
+
+\section{Processing Instructions}
+
+The package files \file{ltxdocext.sty} and \file{acrofont.sty}
+are generated from this file, \file{ltxdocext.dtx},
+using the {\sc docstrip} facility of \LaTeX
+via |tex ltxdocext.dtx| (Note: do \emph{not} use \LaTeX\ for this step).
+The typeset documentation that you are now reading is generated from
+the same file by typesetting it with \LaTeX\ or pdftex
+via |latex ltxdocext.dtx| or |pdflatex ltxdocext.dtx|.
+
+\subsection{Build Instructions}
+
+You may bootstrap this suite of files solely from \file{ltxdocext.dtx}.
+Prepare by installing \LaTeXe\ (and either tex or pdftex) on your computer,
+then carry out the following steps:
+\begin{enumerate}
+\item
+Within an otherwise empty directory,
+typeset \file{ltxdocext.dtx} with \TeX\ or pdftex;
+thereby generating the package file \file{ltxdocext.sty},
+and the package file \file{acrofont.sty}.
+Make sure that {\sc docstrip} receives permission
+to overwrite existing versions of these packages.
+\item
+Now
+typeset \file{ltxdocext.dtx} with \LaTeX\ or pdflatex;
+you will obtain the typeset documentation you are now reading,
+along with
+the file \file{README}.
+
+Note: you will have to run \LaTeX, then
+\file{makeindex} \texttt{-s gind.ist ltxdocext.idx}, then
+\file{makeindex} \texttt{-s gglo.ist -o ltxdocext.gls ltxdocext.glo}, then
+\LaTeX\ again in order to obtain a valid index and table of contents.
+\item
+Install the following files into indicated locations within your
+TDS-compliant \texttt{texmf} tree (you may need root access):
+\begin{itemize}
+\item
+\file{$TEXMF/}\file{tex/}\file{latex/}\file{revtex/}\classname{ltxdocext.sty} and
+\file{$TEXMF/}\file{tex/}\file{latex/}\file{revtex/}\classname{acrofont.sty}
+\item
+\file{$TEXMF/}\file{source/}\file{latex/}\file{revtex/}\classname{ltxdocext.dtx}
+\item
+\file{$TEXMF/}\file{doc/}\file{latex/}\file{revtex/}\classname{ltxdocext.pdf}
+\end{itemize}
+where \file{$TEXMF/} stands for \file{texmf-local/}, or some other \texttt{texmf} tree
+in your installation.
+\item
+Run \texttt{mktexlsr} on \file{$TEXMF/} (you may need root access).
+\item
+Build and installation are now complete;
+now put a \cmd\usepackage\texttt{\{ltxutil\}} in your document preamble!
+\end{enumerate}
+
+\subsection{Change Log}
+\changes{1.0a}{2018/12/12}{(MD) Updated name of README file and use standard fonts when typesetting}%
+
+
+\end{filecontents*}
+% \end{macrocode}
+%
+% \subsection{The Document Body}
+%
+% Here is the document body, containing only a
+% \cmd\DocInput\ directive---referring to this very file.
+% This very cute self-reference is a common \classname{ltxdoc} idiom.
+% \begin{macrocode}
+\begin{document}%
+\expandafter\DocInput\expandafter{\jobname.dtx}%
+\PrintChanges
+\end{document}
+%</driver>
+% \end{macrocode}
+%
+% \section{Using the \classname{ltxdoc} and \classname{acrofont} packages}%
+% These packages are an adjunct to the
+% standard \LaTeX\ \classname{ltxdoc} class and may be
+% simply invoked as follows:
+%\begin{verbatim}
+%\documentclass[draft]{ltxdoc}
+%\RequirePackage{ltxdocext}%
+%\RequirePackage{acrofont}%
+%\CodelineIndex\EnableCrossrefs
+%\end{verbatim}
+%
+% Your document should simply cleave to the standards of the
+% \classname{ltxdoc} class, with extensions and alterations as noted.
+%
+% \subsection{Extensions to the \classname{ltxdoc} class}%
+%
+% \subsubsection{Extensions to the \env{verbatim} environment and \cs{verb} command}%
+%
+% The delimiters \verb$<<$ and \verb$>>$ within the scope of the verbatim environment
+% or within the argument of the \cmd\verb\ command produce
+% italics surrounded by angle brackets.
+% This typographic convention usually indicates
+% \emph{metalanguage}, i.e., a placeholder.
+%
+% To obtain the angle bracket character per se,
+% double the character, viz., ``\verb$the delimiter \verb+<<<<+$''.
+%
+% \subsubsection{The -\texttt{matter} Commands Work}
+% The sectioning commands \cmd\frontmatter, \cmd\mainmatter, and \cmd\backmatter
+% of the standard \LaTeX\ \classname{book} class are operative in the
+% \classname{ltxdoc} class.
+%
+% \subsubsection{The \cs{GetFileInfo} command}\label{sec:GetFileInfo}
+% You can use the \cmd\GetFileInfo\ command to extract
+% the date, version, and file info of a file which has registered itself
+% via the \cmd\ProvidesFile\ or \cmd\ProvidesClass\ command
+% (employing the optional argument thereto).
+%
+% For instance, if your document contains the following:
+%\begin{verbatim}
+%\RequirePackage{ltxdocext}%
+%\GetFileInfo{ltxdocext.sty}%
+%\end{verbatim}
+% then the following control sequence names will have
+% a value corresponding to that package's \cmd\ProvidesFile\
+% command:
+% \cmd\filedate: the file's date,
+% \cmd\fileversion: the file's version, and
+% \cmd\fileinfo: the file's info.
+%
+% \subsubsection{Self-Indexing Commands}%
+% Certain commands automatically produce an index entry
+% (or several related entries) according to the meaning.
+%
+% \begin{unnumtable}
+% \begin{tabular}{ll}
+% meta-text &\cmd\marg\arg{text}\\
+% command &\cmd\cmd\cmd\csname\\
+% environment name &\cmd\env\arg{name}\\
+% \cmd\begin\verb`{foo}`&\cmd\envb\arg{foo}\\
+% \cmd\end\verb`{foo}` &\cmd\enve\arg{foo}\\
+% argument &\cmd\arg\arg{name}\\
+% optional &\cmd\oarg\arg{name}\\
+% filename &\cmd\file\arg{name}\\
+% url &\cmd\url\arg{name}\\
+% document class &\cmd\classname\arg{name}\\
+% document substyle &\cmd\substyle\arg{name}\\
+% class option &\cmd\classoption\arg{name}
+% \end{tabular}
+% \end{unnumtable}
+%
+% \subsubsection{Unnumbered Tables}%
+%
+% When your documentation requires the use of an unnumbered table,
+% use the \env{unnumtable} environment:
+% \begin{verbatim}
+%\begin{unnumtable}
+%\begin{tabular}{ll}
+%<table rows>
+%\end{tabular}
+%\end{unnumtable}
+% \end{verbatim}
+%
+% \subsubsection{Structuring Tables}%
+% The commands \cmd\toprule, \cmd\colrule, and \cmd\botrule
+% allow you to mark the beginning of the column heads
+% the beginning of the table body, and the end of
+% the table body, respectively.
+% In context,
+% \begin{verbatim}
+%\begin{tabular}{ll}
+%\toprule
+%<table head rows>
+%\colrule
+%<table rows>
+%\botrule
+%\end{tabular}
+% \end{verbatim}
+%
+% \subsubsection{A Sectioning Command Below \cs{subsection}}%
+% The \cmd\subsubsection\ command is defined.
+%
+%
+% \subsection{Alterations to the \classname{ltxdoc} class}%
+% The following involve no new markup, but they
+% do change the appearance of your formatted documentation:
+% \begin{enumerate}
+% \item
+% Using the \classname{acrofont} package causes your
+% document to be formatted using the standard
+% Acrobat fonts to the greatest extent possible.
+% This means that for most documents, Computer Modern
+% is not used at all.
+% Math that unavoidable must use CM still exists, however.
+% \item
+% An index will be produced at the end of the document
+% without your needing to explicitly mark it up, and
+% it will have an entry in the TOC.
+% \item
+% The \env{quote} environment has a slightly smaller left margin.
+% \item
+% Array columns are set tight by default.
+% \item
+% A host of \cmd\DoNotIndex\ directives are invoked.
+% I intend this list to grow to encompass
+% even more commands. Send me your suggestions.
+% \end{enumerate}
+%
+%
+%
+%\StopEventually{}
+%
+% \section{Extensions to the ltxdoc class}
+% The \file{package} {\sc docstrip} module comprises the
+% package \classname{ltxdocext.sty}, which provides extensions
+% to the standard \LaTeX\ \classname{ltxdoc} class.
+%
+% \subsection{Beginning of the \file{package} {\sc docstrip} module}
+% This portion of code is only present in the \LaTeX\ package (\texttt{.sty} file),
+% not in the kernel portion.
+% \begin{macrocode}
+%<*package>
+\def\class@name{ltxdocext}%
+\expandafter\PackageInfo\expandafter{\class@name}{%
+ An extension to the \protect\LaTeXe\space ltxdoc class
+ by A. Ogawa (arthur\_ogawa sbcglobal.net)%
+}%
+%</package>
+% \end{macrocode}
+%
+% \subsection{Beginning of the \file{kernel} {\sc docstrip} module}
+%
+% The bulk of the code is the kernel portion; a brief tail of \file{package} code then follows.
+% \begin{macrocode}
+%<*kernel>
+% \end{macrocode}
+%
+% \subsection{Incorporate \classname{ltxguide.cls} extensions}
+%
+% Code extracted from \classname{ltxguide.cls}, by Alan Jeffrey.
+% ``This code stolen from \classname{ltxguide.cls}:
+% Some hacks with verbatim... NB: this would be better done with the
+% verbatim package, but this document has to run on any \LaTeX
+% installation.''
+% \begin{macrocode}
+\RequirePackage{verbatim}%
+\let\o@verbatim\verbatim
+\def\verbatim{%
+ \ifhmode\unskip\par\fi
+% \nopagebreak % Overridden by list penalty
+ \ifx\@currsize\normalsize
+ \small
+ \fi
+ \o@verbatim
+}%
+% \end{macrocode}
+%
+% Here we extend the font-setting command to include making \texttt{<>} active
+% (i.e., adjusting the input encoding).
+% \begin{macrocode}
+\renewcommand \verbatim@font {%
+ \normalfont \ttfamily
+ \catcode`\<=\active
+ \catcode`\>=\active
+}%
+% \end{macrocode}
+%
+% Make \verb$|...|$ a synonym for \cmd\verb\verb$|...|$.
+% \begin{macrocode}
+\RequirePackage{shortvrb}
+\AtBeginDocument{%
+ \MakeShortVerb{\|}%
+}%
+% \end{macrocode}
+%
+% Make active bracket characters produce italics surrounded by angle brackets
+% (used in \env{verbatim} and \cmd\verb).
+% \verb|<<| produces a less-than, and \verb|>>| produces a greater-than.
+% \begin{macrocode}
+\begingroup
+ \catcode`\<=\active
+ \catcode`\>=\active
+ \gdef<{\@ifnextchar<\@lt\@meta}
+ \gdef>{\@ifnextchar>\@gt\@gtr@err}
+ \gdef\@meta#1>{\marg{#1}}
+ \gdef\@lt<{\char`\<}
+ \gdef\@gt>{\char`\>}
+\endgroup
+\def\@gtr@err{%
+ \ClassError{ltxguide}{%
+ Isolated \protect>%
+ }{%
+ In this document class, \protect<...\protect>
+ is used to indicate a parameter.\MessageBreak
+ I've just found a \protect> on its own.
+ Perhaps you meant to type \protect>\protect>?
+ }%
+}
+\def\verbatim@nolig@list{\do\`\do\,\do\'\do\-}
+% \end{macrocode}
+% End of code stolen from \file{ltxguide.cls}. Thanks, Alan.
+%
+% Add functionality from doc.dtx:
+% (code stolen from doc.dtx):
+% \begin{macrocode}
+\def\GetFileInfo#1{%
+ \def\filename{#1}%
+ \def\@tempb##1 ##2 ##3\relax##4\relax{%
+ \def\filedate{##1}%
+ \def\fileversion{##2}%
+ \def\fileinfo{##3}}%
+ \edef\@tempa{\csname ver@#1\endcsname}%
+ \expandafter\@tempb\@tempa\relax? ? \relax\relax}
+% \end{macrocode}
+% (end of code stolen from doc.dtx. Thanks FMi.)
+%
+% Various forms of self-indexing commands:
+% \begin{macrocode}
+\DeclareRobustCommand{\marg}[1]{%
+ \meta{#1}%
+ \index{#1=\string\meta{#1} placeholder}\index{placeholder>#1=\string\meta{#1}}%
+}%
+\DeclareRobustCommand\meta[1]{%
+ \mbox{\LANGLE\itshape#1\/\RANGLE}%
+}%
+\def\LANGLE{$\langle$}%
+\def\RANGLE{$\rangle$}%
+\DeclareRobustCommand{\arg}[1]{%
+ {\ttfamily\string{}\meta{#1}{\ttfamily\string}}%
+ \index{#1=\string\ttt{#1}, argument}\index{argument>#1=\string\ttt{#1}}%
+}%
+\let\oarg\undefined
+\DeclareRobustCommand{\oarg}[1]{%
+ {\ttfamily[%]
+ }\meta{#1}{\ttfamily%[
+ ]}%
+ \index{#1=\string\ttt{#1}, optional argument}%
+ \index{argument, optional>#1=\string\ttt{#1}}%
+}%
+\DeclareRobustCommand\cmd{\begingroup\makeatletter\@cmd}%
+\long\def\@cmd#1{%
+ \endgroup
+ \cs{\expandafter\cmd@to@cs\string#1}%
+ \expandafter\cmd@to@index\string#1\@nil
+}%
+\def\cmd@to@cs#1#2{\char\number`#2\relax}%
+\def\cmd@to@index#1#2\@nil{%
+ \index{#2=\string\cmd#1#2}%\index{command>#2=\string\cmd#1#2}%
+}%
+\DeclareRobustCommand\cs[1]{{\ttfamily\char`\\#1}}%
+\def\scmd#1{%
+ \cs{\expandafter\cmd@to@cs\string#1}%
+ \expandafter\scmd@to@index\string#1\@nil
+}%
+\def\scmd@to@index#1#2\@nil#3{%
+ \index{\string$#3=\string\cmd#1#2---#3}%
+%\index{command>\string$#3=\string\cmd#1#2---#3}%
+}%
+\DeclareRobustCommand\env{\name@idx{environment}}%
+\DeclareRobustCommand\envb[1]{%
+ {\ttfamily\string\begin\string{}\env{#1}{\ttfamily\string}}%
+}%
+\DeclareRobustCommand\enve[1]{{\ttfamily\string\end\string{}\env{#1}{\ttfamily\string}}}%
+\DeclareRobustCommand{\file}{\begingroup\@sanitize\@file}%
+\long\def\@file#1{\endgroup
+ {\ttfamily#1}%
+ \index{#1=\string\ttt{#1}}\index{file>#1=\string\ttt{#1}}%
+}%
+\DeclareRobustCommand\substyle{\name@idx{document substyle}}%
+\DeclareRobustCommand\classoption{\name@idx{document class option}}%
+\DeclareRobustCommand\classname{\name@idx{document class}}%
+\def\name@idx#1#2{%
+ {\ttfamily#2}%
+ \index{#2\space#1=\string\ttt{#2}\space#1}\index{#1>#2=\string\ttt{#2}}%
+}%
+\DeclareRobustCommand\url@ltxdocext{\begingroup\catcode`\/\active\catcode`\.\active\catcode`\:\active\@url}%
+\AtBeginDocument{%
+ \ifx\url\undefined\let\url\url@ltxdocext\fi
+}%
+\def\@url#1{%
+ \url@break{\ttfamily#1}%
+ \url@char\edef\@tempa{#1=\string\url{#1}}%
+ \expandafter\index\expandafter{\@tempa}%
+ \expandafter\index\expandafter{\expandafter u\expandafter r\expandafter l\expandafter >\@tempa}%
+ \endgroup
+}%
+{\catcode`\:\active\aftergroup\def\aftergroup:}{\active@colon}%
+\def\colon@break{\colon@char\allowbreak}%
+\def\colon@char{:}%
+{\catcode`\/\active\aftergroup\def\aftergroup/}{\active@slash}%
+\def\slash@break{\slash@char\allowbreak}%
+\def\slash@char{/}%
+{\catcode`\.\active\aftergroup\def\aftergroup.}{\active@dot}%
+\def\dot@break{\dot@char\allowbreak}%
+\def\dot@char{.}%
+\def\url@break{\let\active@slash\slash@break\let\active@dot\dot@break\let\active@colon\colon@break}%
+\def\url@char{\let\active@slash\slash@char\let\active@dot\dot@char\let\active@colon\colon@char}%
+% \end{macrocode}
+%
+% \subsection{Changes to the base class of the ltxdoc class}
+% Modify \env{theindex} environment so that it produces a TOC entry
+% \begin{macrocode}
+\renewenvironment{theindex}
+ {\if@twocolumn
+ \@restonecolfalse
+ \else
+ \@restonecoltrue
+ \fi
+ \columnseprule \z@
+ \columnsep 35\p@
+\def\see##1##2{\textit{See} ##1}%
+\def\seealso##1##2{\textit{See also} ##1}%
+\long\def\cmd##1{\cs{\expandafter\cmd@to@cs\string##1}}%
+\def\@url##1{\url@break\ttt{##1}\endgroup}%
+\def\ttt{\begingroup\@sanitize\ttfamily\@ttt}%
+\def\@ttt##1{##1\endgroup}%
+\mathchardef\save@secnumdepth\c@secnumdepth
+\c@secnumdepth\m@ne
+ \twocolumn[\section{\indexname}]%
+% \@mkboth{\MakeUppercase\indexname}%
+% {\MakeUppercase\indexname}%
+\c@secnumdepth\save@secnumdepth
+ \thispagestyle{plain}\parindent\z@
+ \parskip\z@ \@plus .3\p@\relax
+ \let\item\@idxitem}
+ {\if@restonecol\onecolumn\else\clearpage\fi}
+\renewenvironment{quote}
+ {\list{}{%
+ \leftmargin1em\relax
+ \rightmargin\leftmargin
+ }%
+ \item\relax}
+ {\endlist}
+% \end{macrocode}
+%
+% \subsection{Extensions to the base class of \classname{ltxdoc.cls}}
+%
+% Matter commands from \classname{book.cls}
+% \begin{macrocode}
+\newif\if@mainmatter
+\newif\if@openright
+\@openrighttrue
+\DeclareRobustCommand\frontmatter{%
+ \cleartorecto
+ \@mainmatterfalse
+ \pagenumbering{roman}%
+}%
+\DeclareRobustCommand\mainmatter{%
+ \cleartorecto
+ \@mainmattertrue
+ \pagenumbering{arabic}%
+}%
+\DeclareRobustCommand\backmatter{%
+ \if@openright
+ \cleartorecto
+ \else
+ \clearpage
+ \fi
+ \@mainmatterfalse
+}%
+\ifx\undefined\cleartorecto
+ \def\cleartorecto{\cleardoublepage}%
+\fi
+% \end{macrocode}
+%
+% Unnumbered tables
+%
+% \begin{environment}{unnumtable}
+% An unnumbered table does not float.
+% \begin{macrocode}
+\def\@to{to}%
+\newenvironment{unnumtable}{%
+ \par
+ \addpenalty\predisplaypenalty
+ \addvspace\abovedisplayskip
+ \hbox\@to\hsize\bgroup\hfil\ignorespaces
+ \let\@Hline\@empty
+}{%
+ \unskip\hfil\egroup
+ \penalty\postdisplaypenalty
+ \vskip\belowdisplayskip
+ \aftergroup\ignorespaces
+ \@endpetrue
+}%
+% \end{macrocode}
+% \end{environment}
+%
+% Emulate \cmd\toprule\ and friends
+% \begin{macrocode}
+\providecommand\toprule{\hline\hline}%
+\providecommand\colrule{\\\hline}%
+\providecommand\botrule{\\\hline\hline}%
+% \end{macrocode}
+%
+% Define sectioning command below \cmd\subsubsection.
+% \begin{macrocode}
+\DeclareRobustCommand\subsubsubsection{%
+ \@startsection{subsubsection}{4}%
+ {\z@}{-15\p@\@plus-5\p@\@minus-2\p@}%
+ {5\p@}{\normalfont\normalsize\itshape}%
+}%
+% \end{macrocode}
+%
+% \subsection{In lieu of \file{ltxdoc.cfg}}
+% We don't want everything to appear in the index
+% \begin{macrocode}
+\DoNotIndex{\',\.,\@M,\@@input,\@Alph,\@alph,\@addtoreset,\@arabic}
+\DoNotIndex{\@badmath,\@centercr,\@cite}
+\DoNotIndex{\@dotsep,\@empty,\@float,\@gobble,\@gobbletwo,\@ignoretrue}
+\DoNotIndex{\@input,\@ixpt,\@m,\@minus,\@mkboth}
+\DoNotIndex{\@ne,\@nil,\@nomath,\@plus,\roman,\@set@topoint}
+\DoNotIndex{\@tempboxa,\@tempcnta,\@tempdima,\@tempdimb}
+\DoNotIndex{\@tempswafalse,\@tempswatrue,\@viipt,\@viiipt,\@vipt}
+\DoNotIndex{\@vpt,\@warning,\@xiipt,\@xipt,\@xivpt,\@xpt,\@xviipt}
+\DoNotIndex{\@xxpt,\@xxvpt,\\,\ ,\addpenalty,\addtolength,\addvspace}
+\DoNotIndex{\advance,\ast,\begin,\begingroup,\bfseries,\bgroup,\box}
+\DoNotIndex{\bullet}
+\DoNotIndex{\cdot,\cite,\CodelineIndex,\cr,\day,\DeclareOption}
+\DoNotIndex{\def,\DisableCrossrefs,\divide,\DocInput,\documentclass}
+\DoNotIndex{\DoNotIndex,\egroup,\ifdim,\else,\fi,\em,\endtrivlist}
+\DoNotIndex{\EnableCrossrefs,\end,\end@dblfloat,\end@float,\endgroup}
+\DoNotIndex{\endlist,\everycr,\everypar,\ExecuteOptions,\expandafter}
+\DoNotIndex{\fbox}
+\DoNotIndex{\filedate,\filename,\fileversion,\fontsize,\framebox,\gdef}
+\DoNotIndex{\global,\halign,\hangindent,\hbox,\hfil,\hfill,\hrule}
+\DoNotIndex{\hsize,\hskip,\hspace,\hss,\if@tempswa,\ifcase,\or,\fi,\fi}
+\DoNotIndex{\ifhmode,\ifvmode,\ifnum,\iftrue,\ifx,\fi,\fi,\fi,\fi,\fi}
+\DoNotIndex{\input}
+\DoNotIndex{\jobname,\kern,\leavevmode,\let,\leftmark}
+\DoNotIndex{\list,\llap,\long,\m@ne,\m@th,\mark,\markboth,\markright}
+\DoNotIndex{\month,\newcommand,\newcounter,\newenvironment}
+\DoNotIndex{\NeedsTeXFormat,\newdimen}
+\DoNotIndex{\newlength,\newpage,\nobreak,\noindent,\null,\number}
+\DoNotIndex{\numberline,\OldMakeindex,\OnlyDescription,\p@}
+\DoNotIndex{\pagestyle,\par,\paragraph,\paragraphmark,\parfillskip}
+\DoNotIndex{\penalty,\PrintChanges,\PrintIndex,\ProcessOptions}
+\DoNotIndex{\protect,\ProvidesClass,\raggedbottom,\raggedright}
+\DoNotIndex{\refstepcounter,\relax,\renewcommand}
+\DoNotIndex{\rightmargin,\rightmark,\rightskip,\rlap,\rmfamily}
+\DoNotIndex{\secdef,\selectfont,\setbox,\setcounter,\setlength}
+\DoNotIndex{\settowidth,\sfcode,\skip,\sloppy,\slshape,\space}
+\DoNotIndex{\symbol,\the,\trivlist,\typeout,\tw@,\undefined,\uppercase}
+\DoNotIndex{\usecounter,\usefont,\usepackage,\vfil,\vfill,\viiipt}
+\DoNotIndex{\viipt,\vipt,\vskip,\vspace}
+\DoNotIndex{\wd,\xiipt,\year,\z@}
+\DoNotIndex{\next}
+% \end{macrocode}
+%
+% Direct \classname{ltxdoc} to produce an index.
+% \begin{macrocode}
+\AtEndDocument{\PrintIndex\PrintChanges}%
+% \end{macrocode}
+%
+% \subsection{Extension to \LaTeX's \env{filecontents} Environment}
+% We want to
+% coax the version number into \env{filecontents}-generated files.
+% Note that we expect \cmd\fileversion\ and \cmd\filedate\ to
+% hold the needed information. For this to be the case,
+% your document should execute the \cmd\GetFileInfo\ command
+% (as documented in section~\ref{sec:GetFileInfo}) before
+% any instances of \env{filecontents}.
+% \begin{macrocode}
+\makeatletter
+\def\endfilecontents{%
+ \immediate\write\reserved@c{%
+ \string\iffalse\space ltxdoc klootch^^J%
+ \ifx\undefined\fileversion\else
+ \ifx\undefined\filedate\else
+ This file has version number \fileversion, last revised \filedate.%
+ \fi\fi
+ \string\fi
+ }%
+ \immediate\closeout\reserved@c
+ \def\T##1##2##3{%
+ \ifx##1\@undefined\else
+ \@latex@warning@no@line{##2 has been converted to Blank ##3e}%
+ \fi
+ }%
+ \T\L{Form Feed}{Lin}%
+ \T\I{Tab}{Spac}%
+ \immediate\write\@unused{}%
+}%
+\expandafter\let\csname endfilecontents*\endcsname\endfilecontents
+\makeatother
+% \end{macrocode}
+%
+% Alter formatting in arrays; set them tight.
+% \begin{macrocode}
+\setlength\arraycolsep{0pt}%
+% \end{macrocode}
+%
+% \subsection{End of the \file{kernel} {\sc docstrip} module}
+%
+% \begin{macrocode}
+%</kernel>
+% \end{macrocode}
+%
+% \subsection{Tail of the \file{package} {\sc docstrip} module}
+% Here is the remainder of the package code.
+% \begin{macrocode}
+%<*package>
+% \end{macrocode}
+% Currently, there is little.
+% \begin{macrocode}
+%</package>
+% \end{macrocode}
+%
+%
+% \section{Font Package for Acrobat Compatability}%
+% The package \classname{acrofont}
+% substitutes Acrobat-standard fonts for Computer Modern where possible,
+% even in math mode.
+% Documents typeset with this package in effect will require as little
+% downloaded font data as possible, but will not be exemplars of
+% fine math typesetting.
+%
+% \subsection{Beginning of the \file{fonts} {\sc docstrip} module}
+% The document class module comprises this and the next
+% four sections.
+% \begin{macro}{\class@base}
+% We define in exactly one spot the base class.
+% Typically that class will be one of \classname{book},
+% \classname{article}, or \classname{report}.
+% The base class effectively defines the use and the markup scheme
+% of the class of documents to be handled by this class.
+%
+% This class is a variant of the standard \LaTeX\ book class:
+% \url{ftp://ctan.tug.org/tex-archive/macros/latex/unpacked}.
+% \begin{macrocode}
+%<*fonts>
+\def\class@name{ltxdocext}%
+\expandafter\ClassInfo\expandafter{\class@name}{%
+ Written for \protect\LaTeXe\space
+ by A. Ogawa (arthur_ogawa at sbcglobal.net)%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \subsection{Variants on psfonts packages}%
+% The following uses \file{times.sty} from \url{/packages/psnfss/psfonts.dtx}
+% \begin{macrocode}
+\RequirePackage{times}%
+% \end{macrocode}
+%
+% The following uses \file{mathptm.sty} from \url{/packages/psnfss/psfonts.dtx}
+% \begin{macrocode}
+\RequirePackage{mathptm}%
+% \end{macrocode}
+%
+% The following is a customization of \file{ot1ptmcm.fd}.
+% The virtual font referred to here \file{zptmnocmr} is
+% a variant of Sebastian Rahtz's \file{zptmcmr}, but with
+% even more glyphs moved from CM to Acrobat-standard fonts.
+% \begin{macrocode}
+\DeclareFontFamily{OT1}{ptmcm}{}
+\DeclareFontShape{OT1}{ptmcm}{m}{n}{
+ <-> zptmnocmr
+}{}
+\DeclareFontShape{OT1}{ptmcm}{l}{n}{<->ssub * ptmnocm/m/n}{}
+% \end{macrocode}
+%
+% The following is a customization of \file{omlptmcm.fd}
+% The virtual font referred to here \file{zptmnocmrm} is
+% a variant of Sebastian Rahtz's \file{zptmcmrm}, but with
+% even more glyphs moved from CM to Acrobat-standard fonts.
+% \begin{macrocode}
+\DeclareFontFamily{OML}{ptmcm}{\skewchar \font =127}
+\DeclareFontShape{OML}{ptmcm}{m}{it}{
+ <-> zptmnocmrm
+}{}
+\DeclareFontShape{OML}{ptmcm}{l}{it}{<->ssub * ptmcm/m/it}{}
+\DeclareFontShape{OML}{ptmcm}{m}{sl}{<->ssub * ptmcm/m/it}{}
+\DeclareFontShape{OML}{ptmcm}{l}{sl}{<->ssub * ptmcm/m/sl}{}
+% \end{macrocode}
+%
+% The following is a customization of \file{omspzccm.fd}
+% The virtual font referred to here \file{zpzcnocmry} is
+% a variant of Sebastian Rahtz's \file{zpzccmry}, but with
+% even more glyphs moved from CM to Acrobat-standard fonts.
+% \begin{macrocode}
+\DeclareFontFamily{OMS}{pzccm}{}
+\DeclareFontShape{OMS}{pzccm}{m}{n}{
+ <-> zpzcnocmry
+}{}% cmsy10 Symbol Zapf Chancery Medium-Italic Times-Roman
+\DeclareFontShape{OMS}{pzccm}{l}{n}{<->ssub * pzccm/m/n}{}
+% \end{macrocode}
+%
+% The following is a customization of \file{omxpsycm.fd}
+% The virtual font referred to here \file{zpsynocmrv} is
+% a variant of Sebastian Rahtz's \file{zpsycmrv}, but with
+% even more glyphs moved from CM to Acrobat-standard fonts.
+% \begin{macrocode}
+\DeclareFontFamily{OMX}{psycm}{}
+\DeclareFontShape{OMX}{psycm}{m}{n}{
+ <-> zpsynocmrv
+}{}
+\DeclareFontShape{OMX}{psycm}{l}{n}{<->ssub * psycm/m/n}{}
+%
+\DeclareFontEncoding{8r}{}{}% from file: 8renc.def
+\DeclareFontFamily{8r}{cmr}{\hyphenchar\font45 }% from file: 8rcmr.fd
+\DeclareFontShape{8r}{cmr}{m}{n}{
+ <-> cmr10
+}{}
+% \end{macrocode}
+%
+% \subsection{Font definition files}%
+%
+% The following forces \LaTeX\ to do now what it would do anyway:
+% load the `font definition' information for the fonts that we
+% use. In this way, we prepare for faster processing through
+% the \cmd\dump\ of a preformatted macro package that will not
+% need to read in any packages or font definitions from disk.
+% \begin{macrocode}
+\input{8rphv.fd}%
+\input{8rptm.fd}%
+\input{ot1phv.fd}%
+\input{ot1ptm.fd}%
+\input{t1ptm.fd}%
+% \end{macrocode}
+%
+% \subsection{More math substitutions}%
+%
+% The following definitions arrange to get certain glyphs from the
+% text font instead of out of math pi fonts.
+% In particular, the copyright and registered symbols
+% are single glyphs instead of composites involving
+% the big circle from the \file{cmsy} font.
+% \begin{macrocode}
+ \def\eightRChar#1{{\def\encodingdefault{8r}\fontencoding\encodingdefault\selectfont\char"#1}}%
+ \def\LANGLE{$<$}%{\eightRChar{8B}}%
+ \def\RANGLE{$>$}%{\eightRChar{9B}}%
+%\def\ASTER{\eightRChar{2A}}%
+%\def\DAGGER{\eightRChar{86}}%
+%\def\DDAGGER{\eightRChar{87}}%
+ \def\BULLET{\eightRChar{95}}%
+%\def\SECTION{\eightRChar{A7}}%
+%\def\PARAGRAPH{\eightRChar{B6}}%
+ \def\VERTBAR{\eightRChar{7C}}%
+ \def\COPYRIGHT{\eightRChar{A9}}%
+ \def\REGISTERED{\eightRChar{AE}}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+ \def\textbar{\VERTBAR}%
+ \def\textbullet{\BULLET}%
+ \def\textcopyright{\COPYRIGHT}%
+ \def\textregistered{\REGISTERED}%
+% \end{macrocode}
+%
+% I have removed \cmd\ensuremath\ from the following definition, and
+% all commands like \cmd\mathsection have been converted to
+% e.g., \cmd\textsection.
+% \begin{macrocode}
+\def\@makefnmark{\@thefnmark}%
+\def\@fnsymbol#1{{\ifcase#1\or *\or \dagger\or \ddagger\or
+ \textsection\or \textparagraph\or \|\or **\or \dagger\dagger
+ \or \ddagger\ddagger \else\@ctrerr\fi}}
+% \end{macrocode}
+%
+% \subsection{End of the \file{fonts} {\sc docstrip} module}
+% Here ends the module.
+% \begin{macrocode}
+%</fonts>
+% \end{macrocode}
+%
+%
+% \section{Programming Conventions}%
+% In writing the above code, I cleave to certain conventions, noted
+% here.
+% My goal in explaining them is to encourage others maintaining this
+% body of code to consider following them as well.
+% The benefits are twofold:
+% Some of the coding conventions aim to avoid programming pitfalls;
+% following them reduces maintenance costs.
+% Others make the code easier to read; following these eases the
+% process of understanding how the code works.
+%
+% And, for my part, I prefer to read code of this type.
+%
+% \subsection{Whitespace Conventions}%
+% Exactly where code lines break and indent, and where additional
+% whitespace is inserted is explained here.
+% \begin{itemize}
+% \item
+% Each new macro definition or register assignment begins a new line.
+% Therefore, \cmd\def, \cmd\newcommand, and their ilk will start in column~1.
+% \item
+% Code is indented one space for each level of nesting within braces
+% \verb|{}|.
+% \item
+% Likewise, if possible, for \cmd\if\dots and matching \cmd\fi.
+% \item
+% However, the closing brace or \cmd\fi\ is outdented by one so that it
+% falls
+% at the same level of indentation as its matching brace or \cmd\if,
+% and it appears alone on its line.
+% \item
+% Use of the |tab| character is deprecated
+% (tabs are not standardized across all applications and operating
+% systems).
+% \item
+% Lines of code are limited to 72 characters.
+% I follow this convention mostly to ease the transmission of files
+% via email (a deprecated practice) and to accomodate people with
+% small monitors.
+% But \classname{ltxdoc} output looks better with the shorter lines, too.
+% \item
+% Extraneous whitespace in the replacement part of a macro definition
+% is avoided by using the comment character |%|.
+% In most cases, if falling at the end of a line of code,
+% a brace will be immediately followed by a comment character,
+% as will the macro parameter |#1| and any one-letter control sequence,
+% like |\\|.
+% \item
+% Extraneous whitespace in the package file is also avoided.
+% When \TeX\ reads in the .sty file, it will process \cmd\def s, and other commands,
+% but will not process blank spaces.
+% This practice is simply a discipline. You don't need to do this.
+% But sometimes \TeX\ has to read in a file while it is in horizontal mode,
+% at which point this practice is essential.
+% \end{itemize}
+% These conventions taken together are illustrated by the following:
+% \begin{verbatim}
+%\def\prepdef#1#2{%
+% \@ifxundefined#1{\toks@{}}{\toks@\expandafter{#1}}%
+% \toks@ii{#2}%
+% \edef#1{\the\toks@ii\the\toks@}%
+%}%
+% \end{verbatim}
+% In the above, the definition of \cmd\prepdef\ would not fit on a single
+% line,
+% and required breaking. The first and last lines have matching braces,
+% and are a the same level of indentation, with the last line containing
+% a single brace.
+%
+% Each of the three intervening lines has balanced braces and is
+% indented by one space. Each line that would otherwise end in a single
+% brace character is terminated by a comment character.
+%
+% Some coders rely on the fact that a space character seen by \TeX's
+% scanner while in vertical mode is a no-op.
+% Be that as it may, I eliminate them unless actually intentional.
+%
+% \subsection{Conventions For Procedures}%
+% Here are some of my preferences when writing procedures:
+% \begin{itemize}
+%
+% \item
+% I dislike defining a macro within another macro, especially when the
+% pattern part is non-nil.
+% You know, you are not saving much space in |mem| when you do this,
+% right?
+% You do save space in the hash table and the string pool, though.
+% On the other hand, we are not dealing with small \TeX\ engines
+% anymore; Team \LaTeX\ has made sure of this.
+%
+% \item
+% If two or more macros have very similar replacement parts, consider
+% layering.
+%
+% \item
+% Macros may perform parsing, may maintain tokens or registers, or may
+% set type (produce marks). I try to avoid combining the three functions
+% in a single macro.
+%
+% \item
+% When a procedure both does assignments and sets type, I try to have a
+% clean separations between the two activities. Try to avoid:
+% \begin{verbatim}
+% \vskip10pt
+% \parindent=0pt
+% \leavevmode
+% \end{verbatim}
+%
+% \item
+% The Boolean calculus (cf. \cmd\@ifx)
+% is very useful and produces code that executes nicely.
+% Using it also helps avoid your having to debug problems where
+% \cmd\if\dots\ and \cmd\fi\ are not properly balanced
+% (a nightmare, in case you have not already experienced it).
+%
+% \end{itemize}
+%
+% \subsection{Conventions For \LaTeX}%
+% Team \LaTeX\ make certain recommendations in \file{clsguide.tex}.
+% Ones that I particularly pay attention to are:
+% \begin{itemize}
+%
+% \item
+% For the sake of ``color safety'',
+% use \cmd\sbox\ rather than \cmd\setbox, \cmd\mbox\ rather than \cmd\hbox, and
+% \cmd\parbox\ or \env{minipage} rather than \cmd\vbox.
+%
+% \item
+% Use \cmd\newcommand\ and \cmd\newenvironment\ to declare user-level commands
+% and environments. Avoid the idiom \cmd\def\cmd\foo, \cmd\def\cmd\endfoo\ to define
+% an environment.
+% Ideally, all user-level markup could be extracted from the
+% document class by grepping on \cmd\newcommand\ and \cmd\newenvironment.
+%
+% \item
+% Prefer to use \cmd\setlength\ to assign registers.
+%
+% \end{itemize}
+% I cannot help but notice that much of the code of \LaTeX\ itself fails
+% to comply with many of the coding recommendations of Team \LaTeX.
+%
+% \Finale
+% \iffalse Here ends the programmer's documentation.\fi
+% \endinput
+%
+\endinput
+%%EOF
diff --git a/texmf/source/latex/revtex/ltxfront.dtx b/texmf/source/latex/revtex/ltxfront.dtx
new file mode 100644
index 0000000..86dc0f3
--- /dev/null
+++ b/texmf/source/latex/revtex/ltxfront.dtx
@@ -0,0 +1,3763 @@
+% \iffalse meta-comment balanced on line 108
+% ltxfront.dtx: package to change page grid, MVL.
+% Copyright (c) 2009 Arthur Ogawa
+%
+% Disclaimer
+% This file is distributed WITHOUT ANY WARRANTY;
+% without even the implied warranty of
+% MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
+% License
+% You may distribute this file under the conditions of the
+% LaTeX Project Public License 1.3c or later
+% (http://www.latex-project.org/lppl.txt).
+% ReadMe
+% For the documentation and more detailed instructions for
+% installation, typeset this document with \LaTeX.
+% Maintenance Status
+% This work has the LPPL maintenance status "maintained";
+% Current Maintainer of this work is Arthur Ogawa.
+%
+% This work consists of the main source file ltxfront.dtx
+% and the derived files
+% ltxfront.sty, ltxfront.pdf
+% Distribution:
+% CTAN:macros/latex/contrib/revtex/
+%
+% Unpacking:
+% tex ltxfront.dtx
+%
+% Documentation:
+% latex ltxfront.dtx; ...
+%
+% Programm calls to get the documentation (example):
+% pdflatex ltxfront.dtx
+% makeindex -s gind.ist ltxfront.idx
+% makeindex -s gglo.ist -o ltxfront.gls ltxfront.glo
+% pdflatex ltxfront.dtx
+% makeindex -s gind.ist ltxfront.idx
+% pdflatex ltxfront.dtx
+%
+% Installation:
+% TDS:doc/latex/revtex/
+% TDS:source/latex/revtex/
+% TDS:tex/latex/revtex/
+%
+% Thanks, Heiko!
+% This method of letting a single .dtx file serve as both
+% documentation (via latex) and installer (via tex) follows
+% the example of Heiko Oberdiek. Thanks!
+%<*ignore>
+\begingroup
+ \def\x{LaTeX2e}%
+\expandafter\endgroup
+\ifcase
+ 0\expandafter\ifx\csname processbatchFile\endcsname\relax\else1\fi\ifx\fmtname\x\else 1\fi
+ \relax
+\else
+ \csname fi\endcsname
+%</ignore>
+%<*install>
+%% This file will generate documentation and runtime files
+%% from ltxfront.dtx when run through LaTeX or TeX.
+\input docstrip
+\preamble
+
+This is a generated file;
+altering it directly is inadvisable;
+instead, modify the original source file.
+See the URL in the file README-LTXFRONT.tex.
+
+License
+ You may distribute this file under the conditions of the
+ LaTeX Project Public License 1.3c or later
+ (http://www.latex-project.org/lppl.txt).
+
+ This file is distributed WITHOUT ANY WARRANTY;
+ without even the implied warranty of MERCHANTABILITY
+ or FITNESS FOR A PARTICULAR PURPOSE.
+
+\endpreamble
+\askforoverwritefalse
+\keepsilent
+ \generate{%
+ %{ignore}
+% \usedir{tex/latex/revtex}%
+ \file{ltxfront.sty}{%
+ \from{ltxfront.dtx}{package,options,kernel}%
+ }%
+ }%
+\ifToplevel{
+\Msg{***********************************************************}
+\Msg{*}
+\Msg{* To finish the installation, please move}
+\Msg{* ltxfront.sty}
+\Msg{* into a directory searched by TeX;}
+\Msg{* in a TDS-compliant installation:}
+\Msg{* texmf/tex/macros/latex/revtex/.}
+\Msg{*}
+\Msg{* To produce the documentation,
+ run ltxfront.dtx through LaTeX.}
+\Msg{*}
+\Msg{* Happy TeXing}
+\Msg{***********************************************************}
+}
+\endbatchfile
+%</install>
+%<*ignore>
+\fi
+%</ignore>
+% \fi
+%
+% \GetFileInfo{ltxfront.dtx}
+%
+% \iffalse ltxdoc klootch
+%<*package>
+%%% @LaTeX-file{
+%%% filename = "ltxfront.dtx",
+%%% version = "4.2e",
+%%% date = "2020/10/03",
+%%% author = "Arthur Ogawa (mailto:arthur_ogawa at sbcglobal.net),
+%%% commissioned by the American Physical Society.
+%%% ",
+%%% copyright = "Copyright (C) 1999, 2009 Arthur Ogawa,
+%%% distributed under the terms of the
+%%% LaTeX Project Public License 1.3c, see
+%%% ftp://ctan.tug.org/macros/latex/base/lppl.txt
+%%% ",
+%%% address = "Arthur Ogawa,
+%%% USA",
+%%% telephone = "",
+%%% FAX = "",
+%%% email = "mailto colon arthur_ogawa at sbcglobal.net",
+%%% codetable = "ISO/ASCII",
+%%% keywords = "latex, page grid, main vertical list",
+%%% supported = "yes",
+%%% abstract = "package to change page grid, MVL",
+%%% }
+%</package>
+% \fi
+%
+% \iffalse ltxdoc klootch
+% The following references the \file{README-LTXFRONT} file,
+% which contains basic information about this package.
+% The contents of this file are generated when
+% you typeset the programmer's documentation.
+% Search on "{filecontents*}{README-LTXFRONT.tex}" to locate it.
+% \fi\input{README-LTXFRONT}%
+%
+% \subsection{Bill of Materials}
+%
+% Following is a list of the files in this distribution arranged
+% according to provenance.
+%
+% \subsubsection{Primary Source}%
+% One single file generates all.
+%\begin{verbatim}ltxfront.dtx\end{verbatim}
+%
+% \subsubsection{Generated by \texttt{latex ltxfront.dtx}}%
+% Typesetting the source file under \LaTeX\
+% generates the readme and the documentation.
+%\begin{verbatim}README-LTXFRONT\end{verbatim}
+%
+% \subsubsection{Generated by \texttt{tex ltxfront.dtx}}%
+% Typesetting the installer generates
+% the package files.
+%\begin{verbatim}ltxfront.sty\end{verbatim}
+%
+% \subsubsection{Documentation}%
+% The following are the online documentation:
+% \begin{verbatim}ltxfront.pdf\end{verbatim}
+%
+% \subsubsection{Auxiliary}%
+% The following are auxiliary files generated
+% in the course of running \LaTeX:
+% \begin{verbatim}ltxfront.aux ltxfront.idx ltxfront.ind ltxfront.log ltxfront.toc\end{verbatim}
+%
+% \section{Code common to all modules}%
+%
+% We want to require only one place in this file
+% where the version number is stated,
+% and we also want to ensure that the version
+% number is embedded into every generated file.
+%
+% Now we declare that
+% these files can only be used with \LaTeXe.
+% An appropriate message is displayed if
+% a different \TeX{} format is used.
+% \begin{macrocode}
+%<*doc|package>
+\NeedsTeXFormat{LaTeX2e}[1995/12/01]%
+%</doc|package>
+% \end{macrocode}
+% As desired, the following modules all
+% take common version information:
+% \begin{macrocode}
+%<kernel&!package&!doc>\typeout{%
+%<*package|doc>
+\ProvidesFile{%
+%</package|doc>
+%<*kernel|package|doc>
+ltxfront%
+%</kernel|package|doc>
+%<*doc>
+.dtx%
+%</doc>
+%<package>.sty%
+%<*package|doc>
+}%
+%</package|doc>
+% \end{macrocode}
+%
+% The following line contains, for once and for all,
+% the version and date information.
+% By various means, this information is reproduced
+% consistently in all generated files and in the
+% typeset documentation.
+% \begin{macrocode}
+%<*doc|package|kernel>
+%<version>
+ [2020/10/03 4.2e frontmatter package (AO,DPC,MD)]% \fileversion
+%</doc|package|kernel>
+%<kernel&!package&!doc>}%
+% \end{macrocode}
+%
+%
+% \section{The doc module \texttt{doc}}
+%
+% This module, consisting of the present section,
+% typesets the programmer's documentation,
+% generating the \file{README-LTXFRONT} and sample document as needed.
+%
+% Because the only uncommented-out lines of code at the beginning of
+% this file constitute the \file{doc} module itself,
+% we can simply typeset the \file{.dtx} file directly,
+% and there is thus rarely any need to
+% generate the ``doc'' {\sc docstrip} module.
+% Module delimiters are nonetheless required so that
+% this code does not find its way into the other modules.
+%
+% The \enve{document} command concludes the typesetting run.
+%
+% \begin{macrocode}
+%<*doc>
+% \end{macrocode}
+%
+% \subsection{The Preamble}
+% The programmers documentation is formatted
+% with the \classname{ltxdoc} class with local customizations,
+% and with the usual code line indexing.
+% \begin{macrocode}
+\documentclass{ltxdoc}
+\RequirePackage{ltxdocext}%
+\let\url\undefined
+\RequirePackage[colorlinks=true,linkcolor=blue]{hyperref}%
+%\expandafter\ifx\csname package@font\endcsname\@undefined\else
+% \expandafter\RequirePackage\expandafter{\csname package@font\endcsname}%
+%\fi
+\CodelineIndex\EnableCrossrefs % makeindex -s gind.ist ltxfront
+\RecordChanges % makeindex -s gglo.ist -o ltxfront.gls ltxfront.glo
+% \end{macrocode}
+%
+% \subsubsection{Docstrip and info directives}
+% We use so many {\sc docstrip} modules that we set the
+% \texttt{StandardModuleDepth} counter to 1.
+% \begin{macrocode}
+\setcounter{StandardModuleDepth}{1}
+% \end{macrocode}
+% The following command retrieves the date and version information
+% from this file.
+% \begin{macrocode}
+\expandafter\GetFileInfo\expandafter{\jobname.dtx}%
+% \end{macrocode}
+%
+% \subsection{The ``Read Me'' File}
+% As promised above, here is the contents of the
+% ``Read Me'' file. That file serves a double purpose,
+% since it also constitutes the beginining of the
+% programmer's documentation. What better thing, after
+% all, to have appear at the beginning of the
+% typeset documentation?
+%
+% A good discussion of how to write a ReadMe file can be found in
+% Engst, Tonya, ``Writing a ReadMe File? Read This''
+% \emph{MacTech} October 1998, p. 58.
+%
+% Note the appearance of the
+% \cmd\StopEventually\ command, which marks the
+% dividing line between the user documentation
+% and the programmer documentation.
+%
+% The usual user will not be asked to
+% do a full build, not to speak
+% of the bootstrap.
+% Instructions for carrying these processes
+% begin the programmer's manual.
+%
+% \begin{macrocode}
+\begin{filecontents*}{README-LTXFRONT.tex}
+\title{%
+ A \LaTeX\ Package for\\the frontmatter and title page%
+ \thanks{%
+ This file has version number \fileversion,
+ last revised \filedate.%
+ }%
+ \thanks{%
+ Version \fileversion\ \copyright\ 2019 American Physical Society
+ }%
+}%
+
+\author{%
+ Arthur Ogawa%
+ \thanks{\texttt{mailto:arthur\_ogawa at sbcglobal.net}}%
+}%
+%\iffalse
+% For version number and date,
+% search on "\fileversion" in the .dtx file,
+% or see the end of the README-LTXFRONT file.
+%\fi
+\maketitle
+
+This file embodies the \classname{ltxfront} package,
+the implementation and its user documentation.
+
+The distribution point for this work is
+\url{https://journals.aps.org/revtex},
+which contains fully unpacked, prebuilt runtime files and documentation.
+
+The \classname{ltxfront} package was commissioned by the American Physical Society
+and is distributed under the terms of the \LaTeX\ Project Public License 1.3c,
+the same license under which all the portions of \LaTeX\ itself is distributed.
+Please see \url{http://ctan.tug.org/macros/latex/base/lppl.txt} for details.
+
+To use this document class, you must have a working
+\TeX\ installation equipped with \LaTeXe\
+and possibly pdftex and Adobe Acrobat Reader or equivalent.
+
+To install, retrieve the distribution,
+unpack it into a directory on the target computer,
+and move the file \file{ltxfront.sty}
+into a location in your filesystem where it will be found by \LaTeX.
+
+To use, read the user documentation \file{ltxfront.pdf}.
+
+\tableofcontents
+
+\section{Processing Instructions}
+
+The package file \file{ltxfront.sty}
+is generated from this file, \file{ltxfront.dtx},
+using the {\sc docstrip} facility of \LaTeX
+via |tex ltxfront.dtx|.
+The typeset documentation that you are now reading is generated from
+the same file by typesetting it with \LaTeX\ or pdftex
+via |latex ltxfront.dtx| or |pdflatex ltxfront.dtx|.
+
+\subsection{Build Instructions}
+
+You may bootstrap this suite of files solely from \file{ltxfront.dtx}.
+Prepare by installing \LaTeXe\ (and either tex or pdftex) on your computer,
+then carry out the following steps:
+\begin{enumerate}
+
+\item
+Within an otherwise empty directory,
+typeset \file{ltxfront.dtx} with \TeX\ or pdftex,
+thereby generating the package file \file{ltxfront.sty}.
+
+\item
+Next typeset \file{ltxfront.dtx} with \LaTeX\ or pdflatex;
+you will obtain the typeset documentation you are now reading,
+along with
+the sample document \file{ltxfront-test.tex}
+and the file \file{README-LTXFRONT}.
+
+Note: you will have to run \LaTeX, then
+\file{makeindex} \texttt{-s gind.ist ltxfront.idx}, then
+\file{makeindex} \texttt{-s gglo.ist -o ltxfront.gls ltxfront.glo}, then
+\LaTeX\ again in order to obtain a valid index and table of contents.
+\item
+Now typeset \file{ltxfront.dtx} with \TeX (not \LaTeX),
+thereby generating the package file \file{ltxfront.sty}.
+\item
+Install the following files into indicated locations within your
+TDS-compliant \texttt{texmf} tree (you may need root access):
+\begin{itemize}
+\item
+\file{$TEXMF/}\file{tex/}\file{latex/}\file{revtex/}\classname{ltxfront.sty}
+\item
+\file{$TEXMF/}\file{source/}\file{latex/}\file{revtex/}\classname{ltxfront.dtx}
+\item
+\file{$TEXMF/}\file{doc/}\file{latex/}\file{revtex/}\classname{ltxfront.pdf}
+\end{itemize}
+where \file{$TEXMF/} stands for \file{texmf-local/}, or some other \texttt{texmf} tree
+in your installation.
+\item
+Run \texttt{mktexlsr} on \file{$TEXMF/} (you may need root access).
+\item
+Build and installation are now complete;
+now put a \cmd\usepackage\texttt{\{ltxfront\}} in your document preamble!
+\end{enumerate}
+
+\subsection{Change Log}
+\changes{4.0b}{1999/06/20}{Change handling of \cs{@author} to prevent parsing problems. (Helps in error recovery when syntax errors exist)}
+\changes{4.0b}{1999/06/20}{Gathered all code for the abstract environment together and abstracted out the formatting details for journals to override.}
+\changes{4.0b}{1999/06/20}{Separate option now controls production}
+\changes{4.0b}{1999/06/20}{Set \cs{footnote} to \LaTeX\ standard version for this scope.}
+\changes{4.0c}{1999/11/13}{grid changes with ltxgrid}
+\changes{4.0d}{1998/01/31}{\cs{comma@space}}
+\changes{4.0d}{1998/01/31}{change name from \cs{tempaffiliation}}
+\changes{4.0d}{1998/01/31}{hookify}
+\changes{4.0d}{1998/01/31}{If empty \cs{abstractname} omit spacing}
+\changes{4.0d}{1998/01/31}{PACS boxed to width of abstract}
+\changes{4.0e}{2000/09/20}{Keywords on same footing as PACS}
+\changes{4.0e}{2000/09/20}{Warn if no production}
+\changes{4.0f}{2001/02/13}{Actually append \cs{AU@grp} to \cs{@AAClist}.}
+\changes{4.0g}{1998/06/10}{multiple preprint commands}
+\changes{4.0h}{1998/06/25}{Changes to support groups of affiliations}
+\changes{4.0h}{1998/06/25}{new hook}
+\changes{4.0l}{1998/09/01}{add optional argument handling back}
+\changes{4.0l}{1998/09/01}{Changes to add collaboration processing, which now means doing comma/and processing on each sublist}
+\changes{4.0l}{1998/09/01}{macro added}
+\changes{4.1a}{2008-06-30}{(AO, 443) create an entry in the PDF bookmarks for the abstract via the toc}
+\changes{4.1a}{2008-07-01}{(AO, 496) remove default prependation to \cs{email}.}
+\changes{4.1a}{2008-07-01}{(AO, 496) remove default prependation to \cs{homepage}.}
+\changes{4.1a}{2008/06/16}{removed code that had been commented out}
+\changes{4.1b}{2008/06/29}{(AO, 455) provide a \texttt{description} environment for the abstract}
+\changes{4.1b}{2008/08/02}{(AO, 486) \cs{collaboration} to work with \texttt{groupedaddress}}
+\changes{4.1b}{2008/08/02}{Change csname \cs{AFS@opr} to \cs{AFG@opr}}
+\changes{4.1b}{2008/08/02}{Make the syntax of this package switch optional}%
+\changes{4.1b}{2008/08/02}{Provide for diagnostics via \cs{frontmatterverbose@sw}}
+\changes{4.1b}{2008/08/02}{Use \cs{@affil@def} for all access to \cs{@affil}}
+\changes{4.1b}{2008/08/02}{Use \cs{frontmatter@init} as the vehicle for all initialization}
+\changes{4.1b}{2008/08/04}{(AO, 443) PDF Bookmark for title page}%
+\changes{4.1b}{2008/08/04}{Provide facility \cs{affil@cutoff}: if there is only one affiliation on the title page, no superscript}%
+\changes{4.1c}{2008/08/04}{Hyperref compatibility: take \cs{maketitle} back.}
+\changes{4.1c}{2008/08/04}{Parametrize the production after the address}%
+\changes{4.1c}{2008/08/04}{Produce PACS, keywords only if specified}%
+\changes{4.1d}{2008/10/17}{(AO, 410) compare \cs{@tempa} to \cs{blankaffiliation}.}%
+\changes{4.1f}{2009/07/07}{(AO, 515) Use \cs{set@footnotefont}, which is defined in ltxutil}
+\changes{4.1g}{2009/10/05}{(AO, 535) Robustify \cs{footnote} while \cs{class@warn} is expanding its argument}
+\changes{4.1g}{2009/10/05}{(AO, 539) Robustify \cs{class@warn} against TOC processing}%
+\changes{4.1i}{2009/10/18}{(AO, 540) More runtime diagnostics via \cs{frontmatterverbose@sw}}
+\changes{4.1i}{2009/10/23}{(AO, 540) Rationalize code that appends to \cs{AF@grp}}
+\changes{4.1i}{2009/10/23}{(AO, 540) Optimize boolean logic }
+\changes{4.1i}{2009/10/23}{(AO, 544) Class options runinaddress and unsortedaddress stopped working}
+\changes{4.1m}{2009/11/20}{(AO, 563) \cs{collaboration} to work with both \texttt{groupedaddress} and \texttt{superscriptaddress}}
+\changes{4.1m}{2009/11/20}{(AO, 563) More complete runtime diagnostics via \cs{frontmatterverbose@sw}}
+\changes{4.1m}{2009/11/20}{(AO, 563) Let \cs{@collaboration@gobble} parse all three arguments of \cs{CO@opr}}
+\changes{4.1n}{2009/11/30}{(AO, 567) Handle \cs{noaffiliation} case}
+\changes{4.1n}{2009/11/30}{(AO, 567) Handle punctuation for \classoption{superscriptaddress} style authors' extra information, such as \cs{email}}
+\changes{4.1n}{2009/12/03}{(AO) remove http:// from \cs{href} call}
+\changes{4.1n}{2010/01/02}{(AO, 571) Interface \cs{set@footnotewidth} for determining the set width of footnotes}%
+\changes{4.1n}{2010/01/02}{(AO, 571) allow split after last line of footnote}%
+\changes{4.1n}{2010/01/02}{(AO, 572) Independent footnote counter for title block. Abstract footnote counter shared with body.}%
+\changes{4.2a}{2018/12/12}{(MD) Updated name of README file and use standard fonts when typesetting}%
+
+\end{filecontents*}
+% \end{macrocode}
+%
+% \subsection{A Gnuly Created Sample Document}
+%
+% \begin{macrocode}
+\begin{filecontents*}{ltxfront-test.tex}
+\documentclass{article}% use any combination of titlepage, twocolumn
+\usepackage[]{ltxfront}% use inactive to turn off features
+\begin{document}
+
+\title{Gnu Veldt Cuisine}
+
+\author{G. Picking}
+\thanks{Supported by a grant from the GSF.}
+\affiliation{Acme Kitchen Products}
+
+\author{R. Dillo}
+\thanks{On leave during 1985.}
+\affiliation{Cordon Puce School}
+
+\date{24 July 1984}%
+\revised{5 January 1985}%
+
+\maketitle
+
+\section{Introduction}%
+We reproduce most features of \LaTeX\ manual, Figure C.4, p. 183.
+\end{document}
+\end{filecontents*}
+% \end{macrocode}
+%
+% \subsection{The Document Body}
+%
+% Here is the document body, containing only a
+% \cmd\DocInput\ directive---referring to this very file.
+% This very cute self-reference is a common \classname{ltxdoc} idiom.
+% \begin{macrocode}
+\begin{document}%
+\expandafter\DocInput\expandafter{\jobname.dtx}%
+% ^^A\PrintChanges
+\end{document}
+% \end{macrocode}
+%
+% \begin{macrocode}
+%</doc>
+% \end{macrocode}
+%
+% \section{Using this package}
+% Once this package is installed on your filesystem, you can employ it in
+% adding functionality to \LaTeX\ by invoking it in your document or document class.
+%
+% \subsection{Invoking the package}
+% In your document, you can simply call it up in your preamble:
+% \begin{verbatim}
+%\documentclass{book}%
+%\usepackage{ltxfront}%
+%\begin{document}
+%<your document here>
+%\end{document}\end{verbatim}
+% However, the preferred way is to invoke this package from within your
+% customized document class:
+% \begin{verbatim}
+%\NeedsTeXFormat{LaTeX2e}[1995/12/01]%
+%\ProvidesClass{myclass}%
+%\RequirePackage{ltxfront}%
+%\LoadClass{book}%
+%<class customization commands>
+%\endinput\end{verbatim}
+%
+% Once loaded, the package gives you acccess to certain procedures,
+% usually to be invoked by a \LaTeX\ command or environment.
+%
+% \section{Frontmatter commands}
+%
+% A document using this package will use commands like \cmd\title, \cmd\author, \cmd\thanks,
+% the \texttt{abstract} environment
+% and \cmd\maketitle.
+% These are part of \LaTeX's standard document classes.
+% But this package supplements \LaTeX\ syntax with additional commands, such as
+% \cmd\affiliation, \cmd\collaboration, \cmd\email, \cmd\homepage, and others.
+%
+% Furthermore, you can have multiple \cmd\author, \cmd\affiliation, and \cmd\collaboration\ commands,
+% each with their own set of \cmd\email, \cmd\homepage, etc. commands.
+%
+% \section{The Front Matter}\label{sec:front}
+%
+% This section describes the macros
+% this package provides for formatting the front matter of an
+% article. The behavior and usage of these macros can be quite
+% different from those provided in \LaTeXe.
+%
+% \subsection{Setting the title}
+%
+% The title of the manuscript is simply specified by using the
+% \cmd\title\arg{title} macro. A \verb+\\+ may be used to put a line
+% break in a long title.
+% The \cmd\title\ commnd takes an optional argument, which you will use to
+% give an abbreviated title for use in the running header.
+%
+% \subsection{Specifying a date}%
+%
+% The \cmd\date\marg{date} command outputs the date on the manuscript.
+% Using \cmd\today\ will cause \LaTeX{} to insert the
+% current date whenever the file is run:
+% \begin{verbatim}\date{\today}\end{verbatim}
+%
+% \subsection{Specifying authors and affiliations}
+%
+% Commands specifying author and affilliation have been improved to save
+% labor for authors and in production. Authors and affiliations are
+% arranged into groupings called, appropriately enough, \emph{author
+% groups}. Each author group is a set of authors who share the same set
+% of affiliations. Author names are specified with the \cmd\author\
+% macro while affiliations (or addresses) are specified with the
+% \cmd\affiliation\ macro. Author groups are specified by sequences of
+% \cmd\author\ macros followed by \cmd\affiliation\ macros. An
+% \cmd\affiliation\ macro applies to all previously specified
+% \cmd\author\ macros which don't already have an affiliation supplied.
+%
+% For example, if Bugs Bunny and Roger Rabbit are both at Looney Tune
+% Studios, while Mickey Mouse is at Disney World, the markup would be:
+% \begin{verbatim}
+% \author{Bugs Bunny}
+% \author{Roger Rabbit}
+% \affiliation{Looney Tune Studios}
+% \author{Mickey Mouse}
+% \affiliation{Disney World}\end{verbatim}
+% The default is to display this as
+% \begin{center}
+% Bugs Bunny and Roger Rabbit\\
+% \emph{Looney Tune Studios}\\
+% Mickey Mouse\\
+% \emph{Disney World}\\
+% \end{center}
+% This layout style for displaying authors and their affiliations is
+% chosen by selecting the class option
+% \classoption{groupedaddress}. This option is the default for all APS
+% journal styles, so it does not need to be specified explicitly.
+% The other major way of displaying this
+% information is to use superscripts on the authors and
+% affiliations. This can be accomplished by selecting the class option
+% \classoption{superscriptaddress}. To achieve the display
+% \begin{center}
+% Bugs Bunny,$^{1}$ Roger Rabbit,$^{1,2}$ and Mickey Mouse$^{2}$\\
+% \emph{$^{1}$Looney Tune Studios}\\
+% \emph{$^{2}$Disney World}\\
+% \end{center}
+% one would use the markup
+% \begin{verbatim}
+% \author{Bugs Bunny}
+% \affiliation{Looney Tune Studios}
+% \author{Roger Rabbit}
+% \affiliation{Looney Tune Studios}
+% \affiliation{Disney World}
+% \author{Mickey Mouse}
+% \affiliation{Disney World}\end{verbatim}
+%
+% Note that this package takes care of any commas and \emph{and}'s that join
+% the author names together and font selection, as well as any
+% superscript numbering. Only the author names and affiliations should
+% be given within their respective macros.
+%
+% There is a third class option, \classoption{unsortedaddress}, for
+% controlling author/affiliation display. The default
+% \classoption{groupedaddress} will actually sort authors into the
+% approriate author groups if one chooses to specify an affiliation for
+% each author. The markup:
+% \begin{verbatim}
+% \author{Bugs Bunny}
+% \affiliation{Looney Tune Studios}
+% \author{Mickey Mouse}
+% \affiliation{Disney World}
+% \author{Roger Rabbit}
+% \affiliation{Looney Tune Studios}\end{verbatim}
+% will result in the same display as for the first case given
+% above even though Roger Rabbit is specified after Mickey Mouse. To
+% avoid Roger Rabbit being moved into the same author group as Bugs
+% Bunny, use the
+% \classoption{unsortedaddress} option instead. In general, it is safest
+% to list authors in the order they should appear and specify
+% affiliations for multiple authors rather than one at a time. This will
+% afford the most independence for choosing the display option. Finally,
+% it should be mentioned that the affiliations for the
+% \classoption{superscriptaddress} are presented and numbered
+% in the order that they are encountered. This means that the order
+% will usually follow the order of the authors. An alternative ordering
+% can be forced by including a list of \cmd\affiliation\ commands before
+% the first \cmd{\author} in the desired order. Then use the exact same
+% text for each affilation when specifying them for each author.
+%
+% If an author doesn't have an affiliation, the \cmd\noaffiliation\
+% macro may be used in the place of an \cmd\affiliation\ macro.
+%
+%
+% \subsubsection{Collaborations}
+%
+% A collaboration name can be specified with the \cmd\collaboration\
+% command. This is very similar to the \cmd\author\ command.
+% It should appear at the end of an author group.
+% The collaboration name will appear centered in parentheses.
+% You may have multiple author groups, each with its own \cmd\collaboration.
+% Because collaborations don't normally have affiliations, one needs to follow the
+% \cmd\collaboration\ with \cmd\noaffiliation.
+%
+% \subsubsection{Footnotes for authors, collaborations, affiliations or title}\label{sec:footau}
+%
+% Often one wants to specify additional information associated with an
+% author, collaboration, or affiliation such an e-mail address, an
+% alternate affiliation, or some other ancillary information.
+% This package introduces several new macros just for this purpose. They
+% are:
+% \begin{itemize}
+% \item\cmd\email\oarg{optional text}\arg{e-mail address}
+% \item\cmd\homepage\oarg{optional text}\arg{URL}
+% \item\cmd\altaffiliation\oarg{optional text}\arg{affiliation}
+% \item\cmd\thanks\arg{miscellaneous text}
+% \end{itemize}
+% In the first three, the \emph{optional text} will be prepended before the
+% actual information specified in the required argument. \cmd\email\ and
+% \cmd\homepage\ each have a default text for their optional arguments
+% (`Electronic address:' and `URL:' respectively). The \cmd\thanks\
+% macro should only be used if one of the other three do not apply. Any
+% author name can have multiple instances of these four commands.
+% Note that unlike the
+% \cmd\affiliation\ macro, these macros only apply to the \cmd\author\
+% that directly precedes it. Any \cmd\affiliation\ \emph{must} follow
+% the other author-specific macros. A typical usage might be as follows:
+% \begin{verbatim}
+% \author{Bugs Bunny}
+% \email[E-mail me at: ]{bugs@looney.com}
+% \homepage[Visit: ]{http://looney.com/}
+% \altaffiliation[Permanent address: ]
+% {Warner Brothers}
+% \affiliation{Looney Tunes}\end{verbatim}
+% This would result in the footnote ``E-mail me at: \texttt{bugs@looney.com},
+% Visit: \texttt{http://looney.com/}, Permanent address: Warner
+% Brothers'' being attached to Bugs Bunny. Note that:
+% \begin{itemize}
+% \item Only an e-mail address, URL, or affiliation should go in the
+% required argument in the curly braces.
+% \item The font is automatically taken care of.
+% \item An explicit space is needed at the end of the optional text if one is
+% desired in the output.
+% \item Use the optional arguments to provide customized
+% text only if there is a good reason to.
+% \end{itemize}
+%
+% The \cmd\collaboration\ , \cmd\affiliation, or even \cmd\title\ can
+% also have footnotes attached via these commands. If any ancillary data
+% (\cmd\thanks, \cmd\email, \cmd\homepage, or
+% \cmd\altaffiliation) are given in the wrong context (e.g., before any
+% \cmd\title, \cmd\author, \cmd\collaboration, or \cmd\affiliation\
+% command has been given), then a warning is given in the \TeX\ log, and
+% the command is ignored.
+%
+% Duplicate sets of ancillary data are merged, giving rise to a single
+% shared footnote. However, this only applies if the ancillary data are
+% identical: even the order of the commands specifying the data must be
+% identical. Thus, for example, two authors can share a single footnote
+% indicating a group e-mail address.
+%
+% Duplicate \cmd\affiliation\ commands may be given in the course of the
+% front matter, without the danger of producing extraneous affiliations
+% on the title page. However, ancillary data should be specified for
+% only the first instance of any particular institution's
+% \cmd\affiliation\ command; a later instance with different ancillary
+% data will result in a warning in the \TeX\ log.
+%
+% It is preferable to arrange authors into
+% sets. Within each set all the authors share the same group of
+% affiliations. For each author, give the \cmd\author\ (and appropriate
+% ancillary data), then follow this author group with the needed group
+% of \cmd\affiliation\ commands.
+%
+% If affiliations have been listed before the first
+% \cmd\author\ macro to ensure a particular ordering, be sure
+% that any later \cmd\affiliation\ command for the given institution is
+% an exact copy of the first, and also ensure that no ancillary data is
+% given in these later instances.
+%
+%
+% A journal may determine the placement of these
+% ancillary information footnotes. One journal may put all
+% such footnotes at the start of the bibliography while another
+% journal may display them on the first page, as a footnote. One can override a
+% journal style's default behavior by specifying explicitly the class
+% option
+% \classoption{bibnotes} (puts the footnotes at the start of the
+% bibliography) or \classoption{nobibnotes} (puts them on the first page).
+%
+% \subsubsection{Specifying first names and surnames}
+%
+% Many authors have names in which either the surname appears first
+% or in which the surname is made up of more than one name. To ensure
+% that such names are accurately captured for indexing and other
+% purposes, the \cmd\surname\ macro should be used to indicate which portion
+% of a name is the surname. Similarly, there is a \cmd\firstname\ macro
+% as well, although usage of \cmd\surname\ should be sufficient. If an
+% author's surname is a single name and written last, it is not
+% necessary to use these macros. These macros do nothing but indicate
+% how a name should be indexed. Here are some examples;
+% \begin{verbatim}
+% \author{Andrew \surname{Lloyd Weber}}
+% \author{\surname{Mao} Tse-Tung}\end{verbatim}
+%
+% \subsection{The abstract}
+% An abstract for a paper is specified by using the \env{abstract}
+% environment:
+% \begin{verbatim}
+% \begin{abstract}
+% Text of abstract
+% \end{abstract}\end{verbatim}
+% Note that with this package, the abstract must be specified before the
+% \cmd\maketitle\ command, and there is no need to embed it in an explicit
+% minipage environment.
+%
+% \subsection{PACS codes}
+% Some authors are asked to supply suggested PACS codes with their
+% submissions. The \cmd\pacs\ macro is provided as a way to do this:
+% \begin{verbatim}\pacs{23.23.+x, 56.65.Dy}\end{verbatim}
+% The actual display of the PACS numbers below the abstract is
+% controlled by two class options: \classoption{showpacs} and
+% \classoption{noshowpacs}. In particular, this is now independent of
+% the \classoption{preprint} option. \classoption{showpacs} must be
+% explicitly included in the class options to display the PACS codes.
+%
+% \subsection{Keywords}
+% A \cmd\keywords\ macro may also be used to indicate keywords for the
+% article.
+% \begin{verbatim}\keywords{nuclear form; yrast level}\end{verbatim}
+% This will be displayed below the abstract and PACS (if supplied). Like
+% PACS codes, the actual display of the the keywords is controlled by
+% two classoptions: \classoption{showkeys} and
+% \classoption{noshowkeys}. An explicit \classoption{showkeys} must be
+% included in the \cmd\documentclass\ line to display the keywords.
+%
+% \subsection{Institutional report numbers}
+% Institutional report numbers can be specified using the \cmd\preprint\
+% macro. These will be displayed in the upper lefthand corner of the
+% first page. Multiple \cmd\preprint\ macros may be supplied (space is
+% limited though, so only three or less may actually fit).
+%
+% \subsection{maketitle}
+% After specifying the title, authors, affiliations, abstract, PACS
+% codes, and report numbers, the final step for formatting the front
+% matter of the manuscript is to issue the \cmd\maketitle\ command:
+% \begin{verbatim}\maketitle\end{verbatim}
+% This command must follow all of the macros listed above.
+% The macro will format the front matter in accordance with the various
+% class options that were specified in the
+% \cmd\documentclass\ line (either implicitly through defaults or
+% explicitly).
+%
+%
+% \section{Compatability with \LaTeX's Required Packages}
+% Certain packages, usually ones written by members of the
+% \LaTeX\ Project itself, have been designated ``required'' and
+% are distributed as part of standard \LaTeX.
+% These packages have been placed in a priviledged position
+% vis \'a vis the \LaTeX\ kernel in that they override the definitions of certain kernel macros.
+%
+% The \classname{ltxfront} package will be incompatible with any package that
+% redefines any of the kernel macros that \classname{ltxfront} patches---if that
+% package is loaded \emph{after} \classname{ltxfront}.
+%
+% Hereinafter follows some notes on specific \LaTeX\ packages.
+%
+%\StopEventually{}
+%
+% \section{Implementation of package}
+%
+% \subsection{Beginning of the \file{package} {\sc docstrip} module}
+% \begin{macrocode}
+%<*package>
+\def\package@name{ltxfront}%
+\expandafter\PackageInfo\expandafter{\package@name}{%
+ Title page macros for \protect\LaTeXe,
+ by A. Ogawa (arthur_ogawa at sbcglobal.net)%
+}%
+% \end{macrocode}
+%
+% \subsection{Prerequisites}%
+%
+% This package requires the use of the \classname{ltxutil} package.
+% \begin{macrocode}
+\let\class@name\package@name
+\RequirePackage{ltxutil}%
+%</package>
+% \end{macrocode}
+% End of the preliminary portion of the package code.
+% \begin{macrocode}
+% \end{macrocode}
+%
+% \subsection{Options}%
+% Package options are treated in this portion of the dtx file.
+% This portion is guarded by the \texttt{options} \classname{docstrip} code so that it can be
+% merged into a larger package gracefully.
+% \begin{macrocode}
+%<*options>
+% \end{macrocode}
+%
+% Package option turns on diagnostics that trace its operation.
+% \cmd\frontmatterverbose@sw\ is set false by default: we do not elect to monitor
+% the workings of these procedures.
+% \begin{macrocode}
+\DeclareOption{frontmatterverbose}{\@booleantrue\frontmatterverbose@sw}%
+\@booleanfalse\frontmatterverbose@sw
+% \end{macrocode}
+%
+% Package option inactivates its syntax, for testing purposes.
+% \begin{macrocode}
+\DeclareOption{inactive}{\@booleanfalse\frontmatter@syntax@sw}%
+\@booleantrue\frontmatter@syntax@sw
+% \end{macrocode}
+%
+% \begin{macro}{\runinaddress@sw}
+% The flag \cmd\runinaddress@sw\ signifies that author addresses are to be set on the same
+% line with the author.
+%
+% The \classoption{runinaddress} option is the only one that sets \cmd\runinaddress@sw\ to true.
+%
+% FIXME: this option pertains to cases where \cmd\groupauthors@sw\ is true, but is independant of \cmd\@affils@sw.
+% So, it should be possible to assert both \classoption{runinaddress} and \classoption{unsortedaddress} as well as
+% each separately.
+% \begin{macrocode}
+\@booleanfalse\runinaddress@sw
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@affils@sw}
+% If \cmd\@affils@sw\ is false, an address is never recognised as ``new''
+% and is therefore always entered into the affiliation list,
+% stopping groups of authors at the same address being consolidated
+% into the same list:
+% the address will be printed the same number of times it is entered.
+%
+% The \classoption{unsortedaddress} option is the only one that sets \cmd\@affils@sw\ to false.
+% \begin{macrocode}
+\@booleantrue\@affils@sw
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\groupauthors@sw}
+% The flag \cmd\groupauthors@sw\ signifies that authors are to be grouped,
+% affecting the meaning of \cmd\@author@present.
+%
+% Clients will want to set \cmd\groupauthors@sw\ true if they want grouped addresses
+% or either of the following variants:
+% for unsorted addresses: set \cmd\@affils@sw\ false;
+% for run-in addresses, set \cmd\runinaddress@sw\ true.
+% \begin{macrocode}
+\@booleanfalse\groupauthors@sw
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{groupedaddress}
+% If we take the \classoption{groupedaddress} option as the default, then we can look upon
+% the \classoption{superscriptaddress} option as simply turning \cmd\groupauthors@sw\ to \cmd\false@sw.
+%
+% \classoption{groupedaddress}, the default in APS journals, groups authors above a common address.
+% \begin{macrocode}
+\DeclareOption{groupedaddress}{\clo@groupedaddress}%
+\def\clo@groupedaddress{%
+ \@booleantrue\groupauthors@sw
+ \@booleantrue\@affils@sw
+ \@booleanfalse\runinaddress@sw
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{unsortedaddress}
+% \classoption{unsortedaddress} is similar to \classoption{groupedaddress},
+% but turns off \cmd\@affils@sw,
+% with the result that each address that is entered will be printed.
+% \begin{macrocode}
+\DeclareOption{unsortedaddress}{\clo@unsortedaddress}%
+\def\clo@unsortedaddress{%
+ \@booleantrue\groupauthors@sw
+ \@booleanfalse\@affils@sw
+ \@booleanfalse\runinaddress@sw
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{runinaddress}
+%
+% \classoption{runinaddress} is similar to \classoption{groupedaddress},
+% with the addition of the Boolean \cmd\runinaddress@sw,
+% which causes the authors associated with a particular address group
+% to be formatted in a paragraph instead of on separate lines.
+% \begin{macrocode}
+\DeclareOption{runinaddress}{\clo@runinaddress}%
+\def\clo@runinaddress{%
+ \@booleantrue\groupauthors@sw
+ \@booleantrue\@affils@sw
+ \@booleantrue\runinaddress@sw
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{superscriptaddress}
+% \classoption{superscriptaddress} presents author affiliations
+% as superscripts. Authors with like affiliations share the same
+% superscript.
+%
+% FIXME: if \cmd\groupauthors@sw\ is false, would not \cmd\@affils@sw\ be a don't care?
+% \begin{macrocode}
+\DeclareOption{superscriptaddress}{\clo@superscriptaddress}%
+\def\clo@superscriptaddress{%
+ \@booleanfalse\groupauthors@sw
+ \@booleantrue\@affils@sw
+ \@booleanfalse\runinaddress@sw
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macrocode}
+%</options>
+% \end{macrocode}
+%
+% \subsection{Process package options}%
+% \begin{macrocode}
+%<*package>
+\DeclareOption*{\OptionNotUsed}%
+\ProcessOptions*
+%</package>
+% \end{macrocode}
+%
+% \section{Kernel of Front Matter}
+% Here begins the kernel of the \classname{frontmatter} package.
+% \begin{macrocode}
+%<*kernel>
+% \end{macrocode}
+%
+% This package implements a new syntax for the title page of an article.
+%
+% Special acknowledgment: this package uses concepts pioneered
+% and first realized by William Baxter (mailto:web at superscript.com)
+% in his SuperScript line of commercial typesetting tools, and
+% which are used here with his permission.
+%
+% These macros were first coded by David P. Carlisle for the American Physical
+% society in the late 1990s as part of the development of REV\TeX4.
+% That development work was taken over by Arthur Ogawa, who is the author of
+% this package.
+%
+% \subsection{Initialization}%
+%
+% \begin{macro}{\frontmatter@init}
+% The \cmd\frontmatter@init\ procedure sets up all the registers for
+% processing the title page.
+% At present, this is done once, at \cmd\class@documenthook\ time.
+% If multiple title pages are to be processed within a job,
+% the initialization must be done for each.
+% \begin{macrocode}
+\appdef\class@documenthook{\frontmatter@init}%
+\let\frontmatter@init\@empty
+% \end{macrocode}
+% \end{macro}
+%
+% \subsection{The title command}%
+%
+% \begin{macro}{\title}
+% \begin{macro}{\do@title}
+% The author uses \cmd\title\ to specify the title. As in the AMS
+% classes, \cmd\title\ has an optional argument specifying a short form
+% for use in running heads.
+%
+% Once the \cmd\title\ is given, you can specify any combination of
+% \cmd\thanks, \cmd\email, \cmd\homepage, and \cmd\altaffiliation\ commands,
+% applying to the title of the document.
+%
+% Compatibility note: as with all of the standard \LaTeX\ commands that we override,
+% we make here a private definition and later switch it in if so required.
+% \changes{4.1b}{2008/08/02}{Make the syntax of this package switch optional}%
+% \begin{macrocode}
+\newcommand\frontmatter@title[2][]{%
+ \def\@title{#2}%
+ \def\@shorttitle{#1}%
+ \let\@AF@join\@title@join
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\@title}
+% \begin{macro}{\@shorttitle}
+% The arguments to \cmd\title\ are saved in these internal
+% macros, which are set up to produce a warning if \cmd\title\ has not
+% been used before \cmd\maketitle.
+% \changes{4.1b}{2008/08/02}{Use \cs{frontmatter@init} as the vehicle for all initialization}
+%
+% Extensibility note: by using \cmd\frontmatter@init\ as the initialization procedure,
+% we open the possibility of more than one title page per document.
+% \begin{macrocode}
+\appdef\frontmatter@init{%
+ \def\@title{\class@warn{No title}}%
+ \let\@shorttitle\@empty
+ \let\@title@aux\@title@aux@cleared
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\@title@join}
+% \begin{macro}{\@title@join@}
+% \begin{macro}{\@title@aux@cleared}
+%
+% \begin{macrocode}
+\def\@title@join{\expandafter\@title@join@\@title@aux}%
+\def\@title@join@#1#2{%
+ \def\@title@aux{{\@join{\@separator}{#1}{#2}}}%
+}%
+\def\@title@aux@cleared{{}}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+%
+% \subsection{The author, affiliation, and related commands}%
+%
+% \begin{macro}{\c@affil}
+% \begin{macro}{\c@collab}
+% \changes{4.1b}{2008/08/02}{(AO, 486) \cs{collaboration} to work with \texttt{groupedaddress}}
+%
+% \cmd\c@affil\ stores the affiliation numbers used for the superscript marks.
+% The \cmd\newcounter\ command defines \cmd\theaffil, which we never use:
+% this counter is evaluated solely via \cmd\the.
+%
+% \cmd\c@collab\ stores the collaboration numbers.
+% The \cmd\newcounter\ command defines \cmd\thecollab.
+% As with \cmd\c@affil, we never use \cmd\thecollab; only \cmd\the\ \cmd\c@collab.
+%
+% Each counter must be cleared before frontmatter can be processed.
+% \begin{macrocode}
+\newcounter{affil}%
+\newcounter{collab}%
+\appdef\frontmatter@init{%
+ \c@affil\z@
+ \c@collab\z@
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \subsubsection{Commands for author or collaboration}%
+%
+% \begin{macro}{\author}
+% Each author is given in a \emph{separate} \cmd\author\ command.
+% This is similar to the AMS classes, but the AMS also try to support
+% the \classname{article} class's \cmd\and\ command for putting multiple
+% authors in one \cmd\author\ command.
+% This package, however, does not support the \cmd\and\ command.
+%
+% When the \cmd\author\ command is encountered,
+% the previous author, author group, affiliation, and affiliation group are at an end.
+% This procedure moves the previous author to the author group (\cmd\move@AU),
+% moves the previous affiliation to the affiliation group (\cmd\move@AF),
+% and terminates the previous author/affiliation group (\cmd\move@AUAF) as the case may require.
+%
+% Compatibility note: as with all of the standard \LaTeX\ commands that we override,
+% we make here a private definition and later switch it in if so required.
+% \changes{4.1b}{2008/08/02}{Make the syntax of this package switch optional}%
+% \begin{macrocode}
+\newcommand\frontmatter@author{% implicit #1
+ \@author@def{}% implicit #2
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\collaboration}
+% Specify the collaboration (given after a group of authors).
+%
+% This procedure does exactly what \cmd\author\ does, and sets
+% a flag signifying that the \cmd\collaboration\ command was given.
+%
+% Note: earlier versions attempted to ignore the collaboration command if it was considered inappropriate.
+% We must not do this, however, because it throws the state machine out of kilter.
+% \changes{4.1b}{2008/08/02}{(AO, 486) \cs{collaboration} to work with \texttt{groupedaddress}}
+% \begin{macrocode}
+\def\collaboration{% implicit #1
+ \@author@def{\@booleantrue\collaboration@sw}% implicit #2
+}%
+\appdef\frontmatter@init{%
+ \@booleanfalse\collaboration@sw
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@author@init}
+% \begin{macro}{\@author@cleared}
+% \begin{macro}{\@authorclear@sw}
+% \begin{macro}{\@author@def}
+% \begin{macro}{\@author@join@}
+% \begin{macro}{\@author@join}
+% \changes{4.0b}{1999/06/20}{Change handling of \cs{@author} to prevent parsing problems. (Helps in error recovery when syntax errors exist)}
+% \changes{4.1b}{2008/08/02}{(AO, 486) \cs{collaboration} to work with \texttt{groupedaddress}}
+% \changes{4.1b}{2008/08/02}{Use \cs{frontmatter@init} as the vehicle for all initialization}
+%
+% Here are all the access procedures for the author data structure.
+%
+% We define a flag value for \cmd\@author (private),
+% a procedure \cmd\@author@gobble\ to read author data without any effect,
+% a procedure for setting \cmd\@author\ to the flag value (\cmd\@author@init),
+% a Boolean to test against the flag value (\cmd\@authorclear@sw),
+% and a procedure to add new ancillary data to the author.
+%
+% Note that expanding the functionality of the \cmd\collaboration\ command entailed changing
+% the data structures, because now all authors must remember the collaboration they belong to.
+% This change affects procedures \cmd\@author@cleared\ and \cmd\@author@gobble.
+% \begin{macrocode}
+\def\@author@cleared{{}{}{}}%
+\def\@author@gobble#1#2#3{}%
+\def\@author@init{%
+ \let\@author\@author@cleared
+ \@booleanfalse\collaboration@sw
+}%
+\def\@authorclear@sw{\@ifx{\@author\@author@cleared}}%
+\appdef\frontmatter@init{%
+ \@author@init
+}%
+% \end{macrocode}
+% \changes{4.1i}{2009/10/18}{(AO, 540) More runtime diagnostics via \cs{frontmatterverbose@sw}}
+% \begin{macrocode}
+\def\@author@def#1#2{%
+ \frontmatterverbose@sw{\typeout{\string\author\space\string\collaboration}}{}%
+ \move@AU\move@AF\move@AUAF
+ \let\@AF@join\@author@join
+ #1%
+ \def\@author{{#2}{}}%
+}%
+\def\@author@join@#1#2#3{%
+ \def\@author{{#1}{\@join{\@separator}{#2}{#3}}}%
+}%
+\def\@author@join{\expandafter\@author@join@\@author}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\move@AU}
+% \begin{macro}{\add@AUCO@grp}
+% \changes{4.1b}{2008/08/02}{(AO, 486) \cs{collaboration} to work with \texttt{groupedaddress}}
+% The \cmd\move@AU\ procedure moves the most recent author to the current author group \cmd\AU@grp,
+% or the most recent collaboration to the collaboration group \cmd\CO@grp.
+% If we recently did a \cmd\@author@init, this is a no-op.
+%
+% To accommodate the expanded syntax of \cmd\collaboration, we add a new field to
+% those associated with an author (collaborator), namely the value of the
+% collaboration counter.
+% \begin{macrocode}
+\def\move@AU{%
+ \@authorclear@sw{}{%
+ \collaboration@sw{%
+ \advance\c@collab\@ne
+ \@argswap{\CO@grp\CO@opr}%
+ }{%
+ \@argswap{\AU@grp\AU@opr}%
+ }%
+ {%
+ \expandafter\@argswap@val
+ \expandafter{\@author}%
+ {\expandafter\@argswap@val\expandafter{\the\c@collab}{\add@AUCO@grp}}%
+ }%
+ }%
+ \@author@init
+}%
+% \end{macrocode}
+% \changes{4.1b}{2008/08/02}{Provide for diagnostics via \cs{frontmatterverbose@sw}}
+% The procedure \cmd\add@AUCO@grp\ adds to the specified group \#3 using the
+% given intercollated operator \#4, with arguments \{\#1\} and \#2,
+% where the first argument is the collaboration number,
+% the second is the author name, and the third the joined attributes of the author.
+% For example, \cmd\move@AU\ could finally execute:
+% \cmd\appdef\ \cmd\AU@grp\ \{\cmd\AU@opr\{1\}\{A. Author\}\{\}\}.
+% \begin{macrocode}
+\def\add@AUCO@grp#1#2#3#4{%
+ \appdef#3{#4{#1}#2}%
+ \frontmatterverbose@sw{\say#3}{}%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\@author@finish}
+% We define a procedure to terminate author/affiliation parsing just before the title block is formatted.
+% We detect the case where an author group is not followed by an affiliation group.
+% \changes{4.0f}{2001/02/13}{Actually append \cs{AU@grp} to \cs{@AAClist}.}
+% \changes{4.1b}{2008/08/02}{(AO, 486) \cs{collaboration} to work with \texttt{groupedaddress}}
+% \changes{4.1b}{2008/08/02}{Use \cs{@affil@def} for all access to \cs{@affil}}
+% \changes{4.1b}{2008/08/02}{Provide for diagnostics via \cs{frontmatterverbose@sw}}
+% \changes{4.1g}{2009/10/05}{(AO, 535) Robustify \cs{footnote} while \cs{class@warn} is expanding its argument}
+% \changes{4.1i}{2009/10/18}{(AO, 540) More runtime diagnostics via \cs{frontmatterverbose@sw}}
+% \changes{4.1m}{2009/11/20}{(AO, 563) \cs{collaboration} to work with both \texttt{groupedaddress} and \texttt{superscriptaddress}}
+% \changes{4.1n}{2009/11/30}{(AO, 567) Handle \cs{noaffiliation} case}
+%
+% We first finish off any author- or affiliation groups in progress.
+% Next, we detect the case where there is an author group or a collaboration group in progress, but no affiliation group.
+% If so, we emit a diagnositic message and act as if a \cmd\noaffiliation\ command had been given.
+% In any case, we finish off any author/affiliation pair that may remain.
+% Formatting of the title block may now proceed.
+% \begin{macrocode}
+\def\@author@finish{%
+ \frontmatterverbose@sw{\typeout{\string\@author@finish}}{}%
+ \move@AU\move@AF
+ \@ifx{\AU@grp\@empty}{%
+ \@ifx{\CO@grp\@empty}%
+ }{%
+ \false@sw
+ }%
+ {}{%
+ \@ifx{\AF@grp\@empty}{%
+ \begingroup
+ \let\href\@secondoftwo
+ \let\AU@opr\@secondofthree
+ \let\CO@opr\@secondofthree
+ \let\footnote\@gobble
+ \@ifx{\CO@grp\@empty}{%
+ \class@warn{Assuming \string\noaffiliation\space for authors}%
+ \frontmatterverbose@sw{\say\AU@grp}%
+ }{%
+ \class@warn{Assuming \string\noaffiliation\space for collaboration}%
+ \frontmatterverbose@sw{\say\CO@grp}{}%
+ }%
+ \endgroup
+ \@affil@none\move@AF
+ }{}%
+ }%
+ \move@AUAF
+}%
+\def\@secondofthree#1#2#3{#2}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@join}
+% The procedure \cmd\@join\ inserts
+% a separator between two tokens, or, if the first token is
+% nil, elides both that token and the separator.
+% \begin{macrocode}
+\def\@join#1#2#3{%
+ \@if@empty{#2}{#3}{#2#1#3}%
+}%
+\def\@separator{;\space}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\surname}
+% \begin{macro}{\firstname}
+% No-op to allow better post-processing of the file.
+% \begin{macrocode}
+\let\surname\@firstofone
+\let\firstname\@firstofone
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\and}
+% The original \LaTeX\ idea of using one \cmd\author\ command to capture
+% all authors and their address just doesn't work with multiple authors
+% possibly sharing addresses, so in this class disable \cmd\and.
+% \begin{macrocode}
+\newcommand\frontmatter@and{\class@err{\protect\and\space is not supported}}
+% \end{macrocode}
+% \end{macro}
+%
+%
+% \subsubsection{Commands for affiliation}%
+%
+% \begin{macro}{\affiliation}
+% \changes{4.1b}{2008/08/02}{Use \cs{@affil@def} for all access to \cs{@affil}}
+% After each group of authors with the same address,
+% give that address in \cmd\affiliation.
+% If later in the list you have an author
+% with the address of an earlier author, repeat the \cmd\affiliation\
+% command: the system will detect the equivalence and (if using the \classoption{superscriptaddress} option)
+% only print the affiliation once, reusing the superscript marker.
+%
+% When the \cmd\affiliation\ command is encountered, the current author and author group (if any) are at an end.
+% Also, the current affiliation is at an end.
+%
+% \changes{4.1i}{2009/10/18}{(AO, 540) More runtime diagnostics via \cs{frontmatterverbose@sw}}
+% FIXME: changes to catcode required elsewhere now?
+% \begin{macrocode}
+\def\cat@comma@active{\catcode`\,\active}%
+{\cat@comma@active\gdef,{\active@comma}}%
+\def\active@comma{,\penalty-300\relax}%
+\newcommand\affiliation{%
+ \frontmatterverbose@sw{\typeout{\string\affiliation}}{}%
+ \move@AU\move@AF
+ \begingroup
+ \cat@comma@active
+ \@affiliation
+}%
+\def\@affiliation#1{%
+ \endgroup
+ \let\@AF@join\@affil@join
+ \@affil@def{#1}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\noaffiliation}
+% \changes{4.1b}{2008/08/02}{Use \cs{@affil@def} for all access to \cs{@affil}}
+% \changes{4.1i}{2009/10/18}{(AO, 540) More runtime diagnostics via \cs{frontmatterverbose@sw}}
+% \changes{4.1n}{2009/11/30}{(AO, 567) Handle \cs{noaffiliation} case}
+% User-level command \cmd\noaffiliation\ signifies that the current group of authors, or the current collaboration, has \emph{no} affiliation.
+%
+% This is implemented by acting as if the \cmd\affiliation\ command has been given, and using a
+% flag value of \{\cmd\relax\}, which adds an element to the affiliation group of this author,
+% albeit a nil one (\cmd\AFF@opr\ \{\cmd\relax\} \{\}).
+%
+% Note that this procedure does substantially the same as \cmd\@author@finish. (the latter, If it does not bail out altogether, differs chiefly in providing diagnostics).
+% \begin{macrocode}
+\newcommand\frontmatter@noaffiliation{%
+ \frontmatterverbose@sw{\typeout{\string\noaffiliation}}{}%
+ \move@AU\move@AF
+% \end{macrocode}
+% In \cmd\@author@finish, there is code testing if an effective
+% \cmd\noaffiliation\ should be executed now. Here, we do so in any case.
+% \begin{macrocode}
+ \@affil@none\move@AF
+ \move@AUAF
+}%
+\def\blankaffiliation{{}}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@affil@cleared}
+% \begin{macro}{\@affil@nil}
+% \begin{macro}{\@affil@init}
+% \begin{macro}{\@affil@none}
+% \begin{macro}{\@affilclear@sw}
+% \begin{macro}{\@affil@def}
+% \begin{macro}{\@affil@join@}
+% \begin{macro}{\@affil@join}
+% \changes{4.1b}{2008/08/02}{Use \cs{@affil@def} for all access to \cs{@affil}}
+% \changes{4.1n}{2009/11/30}{(AO, 567) Handle \cs{noaffiliation} case}
+%
+% Here are all the access procedures for the affiliation data structure.
+% Note the similarity to those of the \cmd\@author\ data structure.
+%
+% We define a flag value for \cmd\@affil\ (private),
+% a procedure for setting \cmd\@affil\ to the flag value (\cmd\@affil@init),
+% a Boolean to test against the flag value (\cmd\@affilclear@sw),
+% a procedure \cmd\@affil@def\ to load a value into the structure,
+% and a procedure to add new ancillary data to the affiliation.
+% \begin{macrocode}
+\def\@affil@cleared{{{}}{}}%
+\def\@affil@nil{{\relax}{}}%
+\appdef\frontmatter@init{%
+ \@affil@init
+}%
+\def\@affil@none{%
+ \let\@affil\@affil@nil
+}%
+\def\@affil@init{%
+ \let\@affil\@affil@cleared
+}%
+\def\@affilclear@sw{\@ifx{\@affil\@affil@cleared}}%
+\def\@affil@def#1{%
+ \def\@affil{{#1}{}}%
+}%
+\def\@affil@join@#1#2#3{%
+ \def\@affil{{#1}{\@join{\@separator}{#2}{#3}}}%
+}%
+\def\@affil@join{\expandafter\@affil@join@\@affil}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\move@AF}
+% \changes{4.1b}{2008/08/02}{(AO, 486) \cs{collaboration} to work with \texttt{groupedaddress}}
+% \changes{4.1i}{2009/10/23}{(AO, 540) Rationalize code that appends to \cs{AF@grp}}
+% \changes{4.1n}{2009/11/30}{(AO, 567) Handle \cs{noaffiliation} case}
+% Move the most recent affiliation to the affiliation group.
+% If we recently did a \cmd\@affil@init, this is a no-op.
+%
+% Set the \cmd\temp@sw\ to false, then execute the affiliation list (\cmd\@AFF@list)
+% with the intercollated operator set to \cmd\@affil@match\
+% and with \cmd\@tempa\ holding the first component of \cmd\@affil, the current affiliation, which is to be matched.
+%
+% If the current affiliation has been seen before, then by side effect
+% \cmd\temp@sw\ will be set to true,
+% \cmd\@tempc\ will be set to the matching affiliation number,
+% and \cmd\@tempd\ will be set to the affiliation's ancillary data.
+% The Boolean \cmd\@affils@sw\ being false prevents this test from ever returning a ``true'' result.
+%
+% This procedure uses \cmd\@tempa\ and sets \cmd\@tempc, \cmd\@tempd, and \cmd\temp@sw.
+% \begin{macrocode}
+\def\move@AF{%
+ \@affilclear@sw{}{%
+ \@booleanfalse\temp@sw
+ \let\@tempd\@empty
+ \@affils@sw{%
+ \expandafter\@affil@addr@def\expandafter\@tempa\@affil
+ \def\AFF@opr{\@affil@match\@tempa}%
+ \@AFF@list
+ }{}\temp@sw
+% \end{macrocode}
+% True clause: This affiliation has been seen before.
+% If ancillary data for the affiliation have been given, but are not identical to those seen before, give a warning.
+% \begin{macrocode}
+ {%
+ \expandafter\@affil@aux@def\expandafter\@tempb\@affil
+ \@ifx{\@tempb\@empty}{}{%
+ \@ifx{\@tempb\@tempd}{}{%
+ \class@warn{%
+ Ancillary information for \@tempa\space must not be different!
+ Please put all of it on the first instance%
+ }%
+ }%
+ }%
+ }%
+% \end{macrocode}
+% False clause: This affiliation is a new one:
+% increment the affiliation counter to get a unique affiliation ID
+% and append the new ID, collaboration number, address, and auxiliary information
+% to the list \cmd\@AFF@list.
+%
+% Note that, with the expanded syntax of \cmd\collaboration, we need to label each address with
+% its associated collaboration.
+%
+% Note also that the \cmd\noaffiliation\ case is handled by recognizing the flag value \cmd\@affil@nil.
+% \begin{macrocode}
+ {%
+ \@ifx{\@affil\@affil@nil}{%
+ \def\@tempc{0}%
+ \@argswap@val{0}%
+ }{%
+ \advance\c@affil\@ne
+ \expandafter\def\expandafter\@tempc\expandafter{\the\c@affil}%
+ \expandafter\@argswap@val\expandafter{\the\c@affil}%
+ }%
+ {%
+ \expandafter\@argswap@val\expandafter{\the\c@collab}{%
+ \expandafter\@argswap@val\expandafter{\@affil}{%
+ \add@list@val@val@val\@AFF@list\AFF@opr
+ }%
+ }%
+ }%
+ }%
+% \end{macrocode}
+% In either case, append this affiliation number to the \cmd\AF@grp, the affiliation group of the current author group.
+% (It will later be considered for the \cmd\@AFG@list.)
+% Then assign the flag value to \cmd\@affil\ that signifies it has been cleared.
+% \changes{4.1i}{2009/10/18}{(AO, 540) Rationalize code that appends to \cs{AF@grp}}
+% \begin{macrocode}
+ \appdef@eval\AF@grp\@tempc
+ \frontmatterverbose@sw{\say\AF@grp}{}%
+ \@affil@init
+ }%
+}%
+\def\@affil@addr@def#1#2#3{%
+ \def#1{#2}%
+}%
+\def\@affil@aux@def#1#2#3{%
+ \def#1{#3}%
+}%
+% \end{macrocode}
+% \changes{4.1b}{2008/08/02}{Provide for diagnostics via \cs{frontmatterverbose@sw}}
+% Procedure \cmd\add@list@val@val@val\ appends to the given list \#1 using the given intercollated operator \#2,
+% with arguments \#5 (affiliation ID) and \#4 (collaboration ID) delimited by braces, and \#3 bare.
+% In the end, \cmd\move@AF\ may execute something like
+% \cmd\appdef\ \cmd\@AFF@list\ \cmd\AFF@opr\ \{1\} \{3\} \{My Institution\} \{thanks\}.
+% \begin{macrocode}
+\def\add@list@val@val@val#1#2#3#4#5{%
+ \appdef#1{#2{#5}{#4}#3}%
+ \frontmatterverbose@sw{\say#1}{}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@affil@match}
+% \changes{4.1b}{2008/08/02}{(AO, 486) \cs{collaboration} to work with \texttt{groupedaddress}}
+% The procedure \cmd\@affil@match\ is the definition for \cmd\AFF@opr,
+% the intercollated operator for \cmd\@AFF@list,
+% when adding a new affiliation: it seeks a match with any earlier affiliation.
+% When the affiliation list is executed, this procedure tests each entry against
+% the `new' affiliation (which has been stored in \cmd\@tempa) and its associated
+% collaboration number (in \cmd\@tempb).
+%
+% If \cmd\groupauthors@sw\ is true, then we require the collaboration ID to match,
+% otherwise there is no such requirement,
+% as is appropriate for the \classoption{superscriptaddress} class option.
+%
+% Uses \cmd\temp@sw, \cmd\@tempc, and \cmd\@tempd\ to communicate back:
+% if it returns \cmd\temp@sw\ true, a match was found.
+% Then
+% \cmd\@tempc\ will be the affiliation ID of the matching entry
+% and \cmd\@tempd\ will be the auxiliary information of the matching entry.
+% \begin{macrocode}
+\def\@affil@match#1#2#3#4#5{%
+ \temp@sw{}{%
+ \def\@tempifx{#4}%
+ \@ifx{\@tempifx#1}{%
+ \groupauthors@sw{%
+ \@ifnum{#3=\c@collab}{%
+ \true@sw
+ }{%
+ \false@sw
+ }%
+ }{%
+ \true@sw
+ }%
+ }{%
+ \false@sw
+ }%
+ {%
+ \@booleantrue\temp@sw
+ \def\@tempc{#2}%
+ \def\@tempd{#5}%
+ }{%
+ }%
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\move@AUAF}
+% \changes{4.0h}{1998/06/25}{Changes to support groups of affiliations}
+% \changes{4.1b}{2008/08/02}{(AO, 486) \cs{collaboration} to work with \texttt{groupedaddress}}
+% \changes{4.1b}{2008/08/02}{Change csname \cs{AFS@opr} to \cs{AFG@opr}}
+% \changes{4.1b}{2008/08/02}{Provide for diagnostics via \cs{frontmatterverbose@sw}}
+% Append \cmd\AF@grp, \cmd\AU@grp, and \cmd\CO@grp\ to \cmd\@AAC@list,
+% append \cmd\CO@grp\ to \cmd\@AFG@list,
+% and merge any new \cmd\AF@grp\ to \cmd\@AFG@list.
+%
+% The entire procedure is predicated on something non-trivial to move.
+%
+% If both author group \cmd\AU@grp\ and affiliation group \cmd\AF@grp\ are nil, bail out.
+%
+% No, try that again:
+%
+% If we have seen any affiliations (\cmd\AF@grp\ is not nil),
+% then it is time to move the current affiliation group and author group to \cmd\@AAC@list.
+% If not, we are picking up authors into an author group, and we should bail out.
+%
+% To extend this scheme to the new \cmd\collaboration\ abilities, we must append an affiliation \cmd\AF@opr\ to \cmd\@AAC@list\
+% also in the case where there was no affiliation specified (e.g., \cmd\noaffiliation). This affiliation will be a nil affiliation:
+% \cmd\AF@opr\verb+{{0}}+.
+%
+%FIXME: only one of \cmd\CO@grp\ or \cmd\AU@grp\ is non-empty at this point,
+% but this code does not appear to recognize this fact!
+% \changes{4.1i}{2009/10/18}{(AO, 540) More runtime diagnostics via \cs{frontmatterverbose@sw}}
+% \begin{macrocode}
+\def\move@AUAF{%
+ \frontmatterverbose@sw{\say\AU@grp\say\AF@grp\say\CO@grp}{}%
+ \@ifx{\AF@grp\@empty}{%
+ \@ifx{\@empty\CO@grp}{%
+% \end{macrocode}
+% Both \cmd\AF@grp\ and \cmd\CO@grp\ are empty: nothing to do. But what if \cmd\AU@grp\ is not empty?
+% \begin{macrocode}
+ }{%
+% \end{macrocode}
+% Append the nil \cmd\AF@opr\ to \cmd\@AAC@list.
+% \begin{macrocode}
+ \appdef \@AAC@list{\AF@opr{{0}}}%
+% \end{macrocode}
+% Append the collaboration group to the \cmd\@AAC@list; no intercollated operator need be added,
+% because we already have \cmd\CO@opr.
+% \begin{macrocode}
+ \appdef@e \@AAC@list{\CO@grp}%
+ \appdef@e \@AFG@list{\CO@grp}%
+ \let\CO@grp\@empty
+ }%
+ }{%
+% \end{macrocode}
+% Append current affiliation group (expansion of \cmd\AF@grp)
+% to the author/affiliation list (\cmd\@AAC@list),
+% using \cmd\AF@opr\ as the intercollated operator;
+% also append the current authors (\cmd\AU@grp).
+% \begin{macrocode}
+ \appdef \@AAC@list{\AF@opr}%
+ \appdef@eval\@AAC@list{\AF@grp}%
+ \appdef@e \@AAC@list{\AU@grp}%
+% \end{macrocode}
+% Determine if either \cmd\AU@grp\ or \cmd\CO@grp\ is non-empty.
+% \begin{macrocode}
+ \@ifx{\@empty\AU@grp}{%
+ \@ifx{\@empty\CO@grp}%
+ }{%
+ \false@sw
+ }%
+% \end{macrocode}
+% At this point, there is a Boolean in \TeX's scanner;
+% it will parse the following two brace-delimited sequences and select one of them.
+% \begin{macrocode}
+ {%
+% \end{macrocode}
+% This branch is executed if and only if both \cmd\AU@grp\ and \cmd\CO@grp\ are empty.
+% \begin{macrocode}
+ }{%
+% \end{macrocode}
+% There was either a \cmd\author\ or a \cmd\collaboration\ command.
+% Append the current affiliation group to \cmd\@AFG@list,
+% (only if it is not already there)
+% using \cmd\AFG@opr\ as the intercollated operator.
+%
+% Note that \cmd\@AFG@list\ is a list of \emph{unique} affiliation \emph{groups},
+% and building it entails an \(N^2\) computation.
+% \begin{macrocode}
+ \@booleanfalse\temp@sw
+ \def\AFG@opr{\x@match\AF@grp}%
+ \let\CO@opr\@author@gobble
+ \@AFG@list
+ \temp@sw{}{%
+ \appdef \@AFG@list{\AFG@opr}%
+ \appdef@eval\@AFG@list{\AF@grp}%
+ }%
+% \end{macrocode}
+% If the collaboration \cmd\CO@grp\ is not empty, we append it to the \cmd\@AFG@list\ and to the \cmd\@AAC@list.
+% \changes{4.1m}{2009/11/20}{(AO, 563) \cs{collaboration} to work with both \texttt{groupedaddress} and \texttt{superscriptaddress}}
+% \begin{macrocode}
+ \@ifx{\@empty\CO@grp}{}{%
+ \appdef@e \@AAC@list{\CO@grp}%
+ \appdef@e \@AFG@list{\CO@grp}%
+ \let\CO@grp\@empty
+ }%
+ }%
+% \end{macrocode}
+% Finally, null out the macros holding the author group \cmd\AU@grp, affiliation group \cmd\AF@grp, and collaboration group \cmd\CO@grp.
+% \begin{macrocode}
+ \let\CO@grp\@empty
+ \let\AU@grp\@empty
+ \let\AF@grp\@empty
+ }%
+ \frontmatterverbose@sw{\say\@AAC@list\say\@AFG@list}{}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\AU@grp}
+% \begin{macro}{\AF@grp}
+% \begin{macro}{\@AAC@list}
+% \begin{macro}{\@AFG@list}
+% \begin{macro}{\@AFF@list}
+%
+% The control sequence name \cmd\AU@grp\ accumulates
+% consecutive \cmd\author\ entries (with \cmd\AU@opr\ as the intercollated operator)
+% and \cmd\collaboration\ entries (with \cmd\CO@opr\ as the intercollated operator).
+% Ultimately, its contents are appended to the author/affiliation list \cmd\@AAC@list\
+% and to the list of affiliation groups, \cmd\@AFG@list.
+% It must be initialized to \cmd\@empty.
+%
+% Note on \cmd\AU@opr\ and \cmd\CO@opr: it is essential to treat these two operators to the greatest extent possible on an equal footing.
+% Therefore we invariably assign values to the pair of them within the same procedure.
+%
+% The macro \cmd\AF@grp\ accumulates affiliation IDs into an affiliation group.
+% It must be empty at the beginning of the frontmatter.
+%
+% The \cmd\@AAC@list\ macro accumulates
+% authors (using \cmd\AU@opr\ as the intercollated operator),
+% collaborations (using \cmd\CO@opr\ as the intercollated operator),
+% and
+% affiliations (using \cmd\AF@opr\ as the intercollated operator).
+%
+% It must be empty at the beginning of the frontmatter.
+%
+% The macro \cmd\@AFG@list\ accumulates
+% affiliation groups, with \cmd\AFG@opr\ as the intercollated operator,
+% and collaborations, with no intercollated operator.
+% This token list is employed when and only when \classoption{groupedaddress} is in effect.
+%
+% The macro \cmd\@AFF@list\ accumulates the list of affiliations
+% with \cmd\AFF@opr\ as the intercollated operator.
+% It must be empty at the beginning of the frontmatter.
+%
+% Each \cmd\affiliation\ command in the document contributes
+% to this list, as long as the argument of that instance of the command
+% is ``new''.
+%
+% Each entry in this list consists of the \cmd\AFF@opr\ intercollated operator
+% followed by three brace-delimited tokens, representing, in order:
+% \begin{enumerate}
+% \item
+% the affiliation's unique ID, a number increasing monotonically for
+% each new entry,
+% \item
+% the optional argument of the \cmd\affiliation\ command,
+% the footnote to the affiliation, and
+% \item
+% the text of the affiliation.
+% \end{enumerate}
+%
+% At various junctures in the code, the control sequence name
+% \cmd\AFF@opr\ assumes the meaning of
+% \cmd\@affil@group, \cmd\affil@script,
+% or is defined to expand to \cmd\@affil@match\cmd\@tempa,
+% and the list \cmd\@AFF@list\ is expanded.
+%
+% For each of these lists, here is the syntax of an element and the usage of the associated operator:
+% \begin{verbatim}
+%\@AFF@list \AFF@opr{<affiliation ID>}{<collaboration ID>}{<address>}{<auxiliary info>} \frontmatter@author@produce@script \affils@present@group
+%\@AFG@list \AFG@opr{{<affiliation ID>}..{<affiliation ID>}} \frontmatter@author@produce@group
+%\@AAC@list \AF@opr{{<affiliation ID>}..{<affiliation ID>}} (\AU@opr|\CO@opr){<collaboration ID>}{<author>}{<auxilliary info>} \frontmatter@author@produce@script \frontmatter@author@produce@group \affils@present@group
+%
+%\AFF@opr \@affil@match\@tempa \@affil@script \@affil@group
+%\AFG@opr \affils@present@group \x@match\AF@grp
+%\AF@opr \@gobble \@affilID@def \@affilID@count\AF@temp\@tempcnta \@affilID@match\AF@temp
+%\AU@opr \@secondofthree \@author@count\@tempcnta \@author@present \@author@gobble \@author@count#2 \@author@present
+%\CO@opr \@secondofthree \@collaboration@count\AU@temp\@tempcnta \@collaboration@present\AU@temp \@collaboration@present\AU@temp \@collaboration@gobble
+%
+%\CO@opr\@author@cleared
+% \end{verbatim}
+% \begin{macrocode}
+\appdef\frontmatter@init{%
+ \let\AU@grp\@empty
+ \let\CO@grp\@empty
+ \let\AF@grp\@empty
+ \let\@AAC@list\@empty
+ \let\@AFG@list\@empty
+ \let\@AFF@list\@empty
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\@AF@join}
+% The \cmd\csname\ \cmd\@AF@join\ is a procedure used within
+% \cmd\email, \cmd\homepage, \cmd\thanks, and \cmd\altaffiliation.
+% When such a command appears after an \cmd\author, \cmd\collaboration, or \cmd\affiliation
+% command, the \cmd\@AF@join\ procedure appends the argument given to the appropriate
+% macro.
+% Its default is to give an error message, since these commands are legal only within
+% the particular context mentioned.
+% \begin{macrocode}
+\appdef\frontmatter@init{%
+ \let\@AF@join\@AF@join@error
+}%
+\def\@AF@join@error#1{%
+ \class@warn{%
+ \string\email, \string\homepage, \string\thanks, or \string\altaffiliation\space
+ appears in wrong context.
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+%
+% \subsubsection{Commands for auxiliary information}%
+%
+% \begin{macro}{\email}
+% Just tacks the email address on to the current author or affiliation.
+% \changes{4.1a}{2008-07-01}{(AO, 496) remove default prependation to \cs{email}.}
+% \begin{macrocode}
+\def\sanitize@url{%
+ \@makeother\%%
+ \@makeother\~%
+ \@makeother\_%
+}%
+\newcommand*\email[1][]{\begingroup\sanitize@url\@email{#1}}%
+\def\@email#1#2{%
+ \endgroup
+ \@AF@join{#1\href{mailto:#2}{#2}}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\homepage}
+% Just tacks the URL on to the current author or affiliation.
+% Note: group opened in \cmd\homepage\ is closed in \cmd\@homepage.
+% \changes{4.1a}{2008-07-01}{(AO, 496) remove default prependation to \cs{homepage}.}
+% \changes{4.1n}{2009/12/03}{(AO) remove http:// from \cs{href} call}
+% \begin{macrocode}
+\newcommand*\homepage[1][]{\begingroup\sanitize@url\@homepage{#1}}%
+\def\@homepage#1#2{%
+ \endgroup
+ \@AF@join{#1\href{#2}{#2}}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\href}
+% The document may load a package that defines either of these commands;
+% if not, we give a default meaning.
+% \begin{macrocode}
+\appdef\class@documenthook{%
+ \providecommand\href[1]{}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\thanks}
+% \changes{4.1b}{2008/08/02}{Make the syntax of this package switch optional}%
+%
+% The operative version of \cmd\thanks\ appends an item to \cmd\@author, or \cmd\@affil's auxiliary data.
+% \begin{macrocode}
+\def\frontmatter@thanks{% implicit #1
+ \@AF@join
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\altaffiliation}
+% \changes{4.0d}{1998/01/31}{change name from \cs{tempaffiliation}}
+% \changes{4.0l}{1998/09/01}{add optional argument handling back}
+% Implemented more or less like \cmd\thanks\ but shares the affiliation
+% counter.
+% Optional argument may be used to give explanatory text eg
+% `currently staying at'
+% This will be placed before the address, if used in a footnote.
+% \begin{macrocode}
+\newcommand*\altaffiliation[2][]{%
+ \@AF@join{#1#2}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+%
+% \subsubsection{Procedures for author, collaboration, and affiliation}%
+%
+% Macros that list off, say, authors, and which require punctuation, like ``A, B, and C'', will use
+% \cmd\set@listcomma@list, \cmd\set@listcomma@count, \cmd\@listand, and \cmd\@listcomma.
+% These macros use the \cmd\@tempcnta\ register to keep track of how many items remain to be listed off.
+%
+% At present, \cmd\@author@present\ will decrement that counter.
+%
+% \begin{macro}{\set@listcomma@list}
+% \begin{macro}{\set@listcomma@count}
+% \changes{4.0l}{1998/09/01}{macro added}
+% Pop the author count for this collaboration.
+%
+% Note: by side effect, it assigns \cmd\@listcomma:
+% for a list of length two, suppress comma addition.
+% \begin{macrocode}
+\def\set@listcomma@list#1{%
+ \expandafter\@reset@ac\expandafter#1#1{0}\@reset@ac{%
+ \let\@listcomma\relax
+ }{%
+ \let\@listcomma\@listcomma@comma
+ }%
+}%
+\def\set@listcomma@count#1{%
+ \@ifnum{#1=\tw@}{%
+ \let\@listcomma\relax
+ }{%
+ \let\@listcomma\@listcomma@comma
+ }%
+}%
+% \end{macrocode}
+% Does the actual pop operation, then generates a Boolean which selects one of the two assignments for \cmd\@listcomma.
+%
+% Note: this procedure sets \cmd\@tempcnta\ to reflect the number of items to list off.
+% \begin{macrocode}
+\def\@reset@ac#1#2#3\@reset@ac{%
+ \def#1{#3}%
+ \@tempcnta#2\relax
+ \@ifnum{#2=\tw@}%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\listand}
+% Might need extending with penalties etc.
+%
+% Note: this procedure expects \cmd\@tempcnta\ to reflect the current number of items remaining to list off.
+% \begin{macrocode}
+\def\@listand{\@ifnum{\@tempcnta=\tw@}{\andname\space}{}}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@listcomma}
+% This definition, with \cmd\@ne\ puts a comma before and.
+%
+% David Carlisle says: In UK English (at least) would have \cmd\tw@\ here,
+% which would then implement the convention of indicating conjunction with comma or ``and'', but not both.
+%
+% Note: this procedure expects \cmd\@tempcnta\ to reflect the current number of items remaining to list off.
+% \begin{macrocode}
+\def\@listcomma@comma{\@ifnum{\@tempcnta>\@ne}{,}{}}%
+\def\@listcomma@comma@UK{\@ifnum{\@tempcnta>\tw@}{,}{}}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@collaboration@gobble}
+% This command will be the synonym of \cmd\CO@grp\ during the expansion of \cmd\@AAC@list,
+% within \cmd\affils@present@group execution (in a \classoption{groupedaddress} situation).
+%
+% \changes{4.1m}{2009/11/20}{(AO, 563) Let \cs{@collaboration@gobble} parse all three arguments of \cs{CO@opr}}
+% \begin{macrocode}
+\def\@collaboration@gobble#1#2#3{}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\doauthor}
+% \changes{4.1b}{2008/08/04}{Provide facility \cs{affil@cutoff}: if there is only one affiliation on the title page, no superscript}%
+% \changes{4.1n}{2009/11/30}{(AO, 567) Handle punctuation for \classoption{superscriptaddress} style authors' extra information, such as \cs{email}}
+% Main control over how authors are typeset.
+%
+% |#1| is loaded by \cmd\author
+%
+% |#2| is loaded by \cmd\email, \cmd\homepage, \cmd\altaffiliation, or \cmd\thanks
+%
+% |#3| is the superscript affiliation, if at all used.
+%
+% First, the author name is formatted, followed by a comma,
+% then come any marks relating to affiliation (if present),
+% then come the homepage URL and email address, if any, with
+% appropriate punctuation.
+%
+% Here, as in \cmd\@affil@script, the parameter \cmd\affil@cutoff\
+% controls whether the argument \#3 is produced as a superscript label
+% connecting this author with its affiliation. (If there is only one
+% affiliation on the title page, it makes sense to remove the superscript.)
+%
+% Note that argument \#3 must effectively execute either
+% \cmd\aftergroup\cmd\true@sw\ or \cmd\aftergroup\cmd\false@sw\ (exactly once!).
+% In the case of \classoption{superscriptaddress}, this is done when
+% \cmd\@author@present@script\ invokes \cmd\doauthor\ with its third argument containing
+% \cmd\@affil@present@script.
+% Otherwise (e.g., \classoption{groupedaddress}), \cmd\doauthor\ is invoked by
+% \cmd\@author@present@group\ with its third argument containing \cmd\@affil@present@group.
+% \begin{macrocode}
+\def\doauthor#1#2#3{%
+ \ignorespaces#1\unskip\@listcomma
+ \begingroup
+ #3%
+% \end{macrocode}
+% At this point, we must have queued up a Boolean (either \cmd\true@sw\ or \cmd\false@sw).
+% \begin{macrocode}
+ \@if@empty{#2}{\endgroup{}{}}{\endgroup{\comma@space}{}\frontmatter@footnote{#2}}%
+ \space \@listand
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\x@match}
+% Procedure \cmd\x@match\ is used as an alias \cmd\AFG@opr,
+% probing the list of affiliation groups
+% and setting a Boolean if a match is found to \#1 (usually the current affiliation group).
+%
+% Procedure \cmd\y@match\ is used as an alias to \cmd\@TBN@opr,
+% probing the list of unique title block footnotes for a match to the footnote text:
+% when the first match is found it sets a Boolean,
+% defining \cmd\@tempb\ to the number of that footnote.
+% \begin{macrocode}
+\def\x@match#1#2{%
+ \temp@sw{}{%
+ \def\@tempifx{#2}%
+ \@ifx{\@tempifx#1}{%
+ \@booleantrue\temp@sw
+ }{%
+ }%
+ }%
+}%
+\def\y@match#1#2#3{%
+ \temp@sw{}{%
+ \def\@tempifx{#3}%
+ \@ifx{\@tempifx#1}{%
+ \@booleantrue\temp@sw
+ \def\@tempb{#2}%
+ }{%
+ }%
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@footnote}
+% Used when typesetting a title block footnote.
+% (Item 290: merge duplicates.)
+%
+% Maintains a list of frontmatter footnotes along with associated device.
+% Uniquifies each footnote by traversing the list and looking for a match to the footnote text.
+% If found, define \cmd\@tempb\ to the corresponding footnote device, and typeset it with \cmd\@footnotemark.
+% If never found, manually increment footnote counter,
+% determine the corresponding footnote device, and
+% add footnote text and device to the list of frontmatter footnotes.
+%
+% The list \cmd\@FMN@list\ is processed later in the title block production.
+%
+% Note that this method of making footnotes runs afoul of \classname{hyperref}'s patches of the \LaTeX\ kernel.
+% Therefore, we avoid trouble by refraining from using \cmd\@footnotemark\ and by creating our own hypertext links.
+% \begin{macrocode}
+\def\frontmatter@footnote#1{%
+ \begingroup
+ \@booleanfalse\temp@sw
+ \def\@tempa{#1}%
+ \let\@tempb\@empty
+ \def\@TBN@opr{\y@match\@tempa}%
+ \@FMN@list
+% \end{macrocode}
+% By side effect, \cmd\@tempb\ holds the value of the footnote counter if
+% This note has been seen before. Otherwise \cmd\temp@sw\ is false.
+% \begin{macrocode}
+ \temp@sw{%
+ \expandafter\frontmatter@footnotemark
+ \expandafter{\@tempb}%
+ }{%
+% \end{macrocode}
+% Execute \cmd\frontmatter@foot@mark, feeding it the value of the current footnote counter.
+% This call requires two levels of pre-expansion to accomplish, the first to expand \cmd\csname, the second to evaluate \cmd\the.
+% \begin{macrocode}
+ \stepcounter\@mpfn
+ \expandafter\expandafter
+ \expandafter\frontmatter@foot@mark
+ \expandafter\expandafter
+ \expandafter{%
+ \expandafter \the\csname c@\@mpfn\endcsname
+ }{#1}%
+ }%
+ \endgroup
+}%
+% \end{macrocode}
+% Procedure \cmd\frontmatter@foot@mark\ lays down the footnote device (mark) and takes care of the footnote text.
+% The latter merely involves queueing that text (along with its associated device) into \cmd\@FMN@list.
+% \begin{macrocode}
+\def\frontmatter@foot@mark#1#2{%
+ \frontmatter@footnotemark{#1}%
+ \g@addto@macro\@FMN@list{\@TBN@opr{#1}{#2}}%
+}%
+\appdef\frontmatter@init{%
+ \global\let\@FMN@list\@empty
+}%
+% \end{macrocode}
+% \changes{4.1n}{2010/01/02}{(AO, 572) Independent footnote counter for title block. Abstract footnote counter shared with body.}%
+% \begin{macrocode}
+\def\frontmatter@footnotemark#1{%
+ \leavevmode
+ \ifhmode\edef\@x@sf{\the\spacefactor}\nobreak\fi
+ \begingroup
+ \hyper@linkstart {link}{frontmatter.#1}%
+ \csname c@\@mpfn\endcsname#1\relax
+ \def\@thefnmark{\frontmatter@thefootnote}%
+ \@makefnmark
+ \hyper@linkend
+ \endgroup
+ \ifhmode\spacefactor\@x@sf\fi
+ \relax
+}%
+% \end{macrocode}
+% \end{macro}
+%
+%
+%
+% \subsection{The keywords command}%
+% \begin{macro}{\keywords}
+% \begin{macro}{\@keywords}
+% Usual game, save text in a macro for processing by \cmd\maketitle.
+% \begin{macrocode}
+\def\keywords#1{%
+ \aftermaketitle@chk{\keywords}%
+ \gdef\@keywords{#1}%
+}%
+\appdef\frontmatter@init{%
+ \let\@keywords\@empty
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \subsection{The \cs{date} command and related commands}%
+% \begin{macro}{\date}
+% \begin{macro}{\received}
+% \begin{macro}{\revised}
+% \begin{macro}{\accepted}
+% \begin{macro}{\published}
+% \begin{macro}{\@date}
+% \begin{macro}{\@received}
+% \begin{macro}{\@revised}
+% \begin{macro}{\@accepted}
+% \begin{macro}{\@published}
+% \changes{4.1b}{2008/08/02}{Make the syntax of this package switch optional}%
+% Use the \cmd\date\ command to specify the document date,
+% the \cmd\received\ command to specify the date received,
+% \cmd\revised\ for date revised,
+% \cmd\accepted\ for date accepted, and
+% \cmd\published\ for date published.
+% Normally only \cmd\date\ to be used by author,
+% the remainder used only by editors.
+%
+% DPC: As for \cmd\keywords, but this time don't flag a warning if there is no
+% revision date specified.
+%
+% In each case the user-level command defines the value of a
+% macro which buffers the data entered by the user.
+% For instance, \cmd\received\ causes \cmd\@received\ to
+% acquire a value.
+% The optional argument allows the user to override the
+% text that will be typeset along with the date;
+% the default value of that text is itself
+% a localized macro.
+% \begin{macrocode}
+\newcommand*\frontmatter@date[2][\Dated@name]{\def\@date{#1#2}}%
+\def\@date{}%
+\newcommand*\received[2][\Received@name]{\def\@received{#1#2}}%
+\def\@received{}%
+\newcommand*\revised[2][\Revised@name]{\def\@revised{#1#2}}%
+\def\@revised{}%
+\newcommand*\accepted[2][\Accepted@name]{\def\@accepted{#1#2}}%
+\def\@accepted{}%
+\newcommand*\published[2][\Published@name]{\def\@published{#1#2}}%
+\def\@published{}%
+% \end{macrocode}
+%
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \subsection{The pacs command}%
+% PACS, the Physics and Astronomy Classification Scheme.
+% \begin{macro}{\pacs}
+% \begin{macro}{\@pacs}
+% As for \cmd\keywords\
+% \begin{macrocode}
+\def\pacs#1{%
+ \aftermaketitle@chk{\pacs}%
+ \gdef\@pacs{#1}%
+}%
+\appdef\frontmatter@init{%
+ \let\@pacs\@empty
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \subsection{The \cs{preprint} command}
+%
+% \changes{4.0g}{1998/06/10}{multiple preprint commands}
+% \begin{macrocode}
+\def\preprint#1{\gappdef\@preprint{\preprint{#1}}}%
+\appdef\frontmatter@init{%
+ \let\@preprint\@empty
+}%
+% \end{macrocode}
+%
+% \subsection{The \env{abstract} environment}%
+%
+% \changes{4.0b}{1999/06/20}{Gathered all code for the abstract environment together and abstracted out the formatting details for journals to override.}
+%
+% \begin{macro}{\absbox}
+%
+% \begin{macrocode}
+\newbox\absbox
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{environment}{abstract}
+% Abstract, as in AMS classes, must be specified \emph{before}
+% \cmd\maketitle. It just saves everything up in \cmd\absbox.
+% \changes{4.0d}{1998/01/31}{If empty \cs{abstractname} omit spacing}
+% \changes{4.0d}{1998/01/31}{hookify}
+% Note that the specifics of how the abstract is to be
+% formatted are set by \cmd\frontmatter@abstractwidth,
+% \cmd\frontmatter@abstractheading, and \cmd\frontmatter@abstractfont.
+%
+% Here we wish to set the abstract into type but save it away in a box, much like
+% the \cmd\minipage\ command does.
+%
+% Note that the \cmd\endabstract\ portion of the environment
+% begins with code that mostly duplicates \cmd\endminipage, but without the \cmd\@iiiparbox.
+% At the end, we simply transfer the contents of the \cmd\minipage\
+% into a box of our own.
+%
+% However, we also have to end the gratuitous
+% paragraph begun by \cmd\minipage.
+% We ensure that no trace of that paragraph is left,
+% by absorbing the \cmd\parindent\ box.
+% Doing so destroys the paragraph entirely, except
+% for the \cmd\parskip\ glue, and that morsel is pruned, because we are at the top
+% of a page.
+%
+% This mechanism is vulnerable, however, because anything placed into the horizontal list
+% after the \cmd\parindent\ box will confound it.
+% And that is exactly what happens under Mik\TeX:
+% There seems to be an extension to the \TeX\ standard operating here which inserts a \cmd\special \ at the beginning of every paragraph.
+% (The mechnism is \emph{not} based on \cmd\everypar.)
+%
+% To accomodate this state of affairs,
+% we have a new mechanism that offers an even more complete rewrite of \env{minipage}.
+%
+% Note that an entry is made in the toc for the abstract; this has the side effect
+% of creating a PDF bookmark for this element if the \classname{hyperref} package is in effect.
+% \changes{4.1a}{2008-06-30}{(AO, 443) create an entry in the PDF bookmarks for the abstract via the toc}
+% \begin{macrocode}
+\def\toclevel@abstract{1}%
+\def\addcontents@abstract{%
+ \phantomsection
+ \expandafter\def\csname Parent0\endcsname{section*.2}%
+ \expandafter\@argswap@val\expandafter{\abstractname}{\addcontentsline{toc}{abstract}}%
+}%
+\newenvironment{frontmatter@abstract}{%
+ \aftermaketitle@chk{\begin{abstract}}%
+ \global\setbox\absbox\vbox\bgroup
+ \color@begingroup
+ \columnwidth\textwidth
+ \hsize\columnwidth
+ \@parboxrestore
+% \end{macrocode}
+% The following line switches footnotes to the \texttt{mpfoootnote} series.
+% This action should rather be taken by \cmd\frontmatter@abstractfont, which can be customized.
+% There is also an interaction with \cmd\mini@note.
+% \begin{macrocode}
+ \def\@mpfn{mpfootnote}\def\thempfn{\thempfootnote}\c@mpfootnote\z@
+ \let\@footnotetext\frontmatter@footnotetext
+ \minipagefootnote@init
+ \let\set@listindent\set@listindent@
+ \let\@listdepth\@mplistdepth \@mplistdepth\z@
+ \let@environment{description}{frontmatter@description}%
+ \@minipagerestore
+ \@setminipage
+ \frontmatter@abstractheading
+ \frontmatter@abstractfont
+ \let\footnote\mini@note
+ \expandafter\everypar\expandafter{\the\everypar\addcontents@abstract\everypar{}}%
+}{%
+ \par
+ \unskip
+ \minipagefootnote@here
+ \@minipagefalse %% added 24 May 89
+ \color@endgroup
+ \egroup
+}%
+% \end{macrocode}
+% \end{environment}
+%
+% \begin{macro}{\frontmatter@footnotetext}
+% We reimplement \cmd\@mpfootnotetext\ under a new name (so as not to be overridden by the \classname{hyperref} package)
+% and extend it to accomodate hyperrefs.
+%
+% Note that this procedure is very like \classname{ltxutil}'s \cmd\mp@footnotetext, except that it
+% uses \cmd\frontmatter@makefntext\ instead of \cmd\@makefntext.
+%
+% \changes{4.1f}{2009/07/07}{(AO, 515) Use \cs{set@footnotefont}, which is defined in ltxutil}
+% \changes{4.1n}{2010/01/02}{(AO, 571) allow split after last line of footnote}%
+% \changes{4.1n}{2010/01/02}{(AO, 572) Independent footnote counter for title block. Abstract footnote counter shared with body.}%
+% \begin{macrocode}
+\long\def\frontmatter@footnotetext#1{%
+ \minipagefootnote@pick
+ \set@footnotefont
+ \set@footnotewidth
+ \@parboxrestore
+ \protected@edef\@currentlabel{\csname p@\@mpfn\endcsname\@thefnmark}%
+ \color@begingroup
+ \frontmatter@makefntext{%
+ \rule\z@\footnotesep\ignorespaces#1\@finalstrut\strutbox\vadjust{\vskip\z@skip}%
+ }%
+ \color@endgroup
+ \minipagefootnote@drop
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\ltx@no@footwarn}
+%
+% \begin{macrocode}
+\def\ltx@no@footnote{%
+ \let\ltx@xfootnote\ltx@no@xfootnote\let\ltx@yfootnote\ltx@no@yfootnote
+ \let\ltx@xfootmark\ltx@no@xfootmark\let\ltx@yfootmark\ltx@no@yfootmark
+ \let\ltx@xfoottext\ltx@no@xfoottext\let\ltx@yfoottext\ltx@no@yfoottext
+}%
+\def\ltx@no@xfootnote[#1]#2{\ltx@no@footwarn\footnote}%
+\def\ltx@no@yfootnote#1{\ltx@no@footwarn\footnote}%
+\def\ltx@no@xfootmark[#1]{\ltx@no@footwarn\footnotemark}%
+\def\ltx@no@yfootmark{\ltx@no@footwarn\footnotemark}%
+\def\ltx@no@xfoottext[#1]#2{\ltx@no@footwarn\footnotetext}%
+\def\ltx@no@yfoottext#1{\ltx@no@footwarn\footnotetext}%
+\def\ltx@no@footwarn#1{%
+ \class@warn{%
+ The \string#1\space command is not legal on the title page;
+ using \string\thanks\space instead might suit you: consult the manual for details%
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@abstractheading}
+% The default abstract head; journals will override this procedure.
+% \begin{macrocode}
+\def\frontmatter@abstractheading{%
+ \begingroup
+ \centering\large
+ \abstractname
+ \par
+ \endgroup
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@abstractfont}
+% The default type specification for the body of the abstract.
+% Journals will override this setting.
+% \begin{macrocode}
+\def\frontmatter@abstractfont{}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{frontmatter@description}
+% \changes{4.1b}{2008/06/29}{(AO, 455) provide a \texttt{description} environment for the abstract}
+% Within the abstract, the description environment is defined as follows:
+% \begin{macrocode}
+\newenvironment{frontmatter@description}{%
+ \list{}{%
+ \leftmargin\z@
+ \labelwidth\z@
+% \end{macrocode}
+% \begin{verbatim}
+% \itemindent-\leftmargin
+% \end{verbatim}
+% \begin{macrocode}
+ \itemindent\z@
+ \let\makelabel\frontmatter@descriptionlabel
+ }%
+}{%
+ \endlist
+}%
+\def\frontmatter@descriptionlabel#1{%
+ \hspace\labelsep
+ \normalfont\bfseries
+ #1:%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@abstractwidth}
+% The default setting is the full text width;
+% journals can override this setting.
+% \begin{macrocode}
+\def\frontmatter@abstractwidth{\textwidth}
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@abstract@produce}
+% This procedure determines how the abstract is incorporated
+% into the title block. We split this out in anticipation
+% of the need to lift the limitation that the abstract not break over pages.
+%
+% If we are in preprint style, we provide an easy pagebreak point immediately above the abstract.
+% This means that the abstract will either fit entirely on the same page as the title block,
+% or it will start a page of its own (which may itself break onto a subsequent page if necessary).
+% \begin{macrocode}
+\def\frontmatter@abstract@produce{%
+ \par
+ \preprintsty@sw{%
+ \do@output@MVL{%
+ \vskip\frontmatter@preabstractspace
+ \vskip200\p@\@plus1fil
+ \penalty-200\relax
+ \vskip-200\p@\@plus-1fil
+ }%
+ }{%
+ \addvspace{\frontmatter@preabstractspace}%
+ }%
+ \begingroup
+ \dimen@\baselineskip
+ \setbox\z@\vtop{\unvcopy\absbox}%
+ \advance\dimen@-\ht\z@\advance\dimen@-\prevdepth
+ \@ifdim{\dimen@>\z@}{\vskip\dimen@}{}%
+ \endgroup
+ \begingroup
+ \prep@absbox
+ \unvbox\absbox
+ \post@absbox
+ \endgroup
+ \@ifx{\@empty\mini@notes}{}{\mini@notes\par}%
+ \addvspace\frontmatter@postabstractspace
+}%
+\appdef\frontmatter@init{\let\mini@notes\@empty}%
+\let\prep@absbox\@empty
+\let\post@absbox\@empty
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@preabstractspace}
+% \begin{macro}{\frontmatter@postabstractspace}
+% Space above and space below abstract in title block
+% \begin{macrocode}
+\def\frontmatter@preabstractspace{.5\baselineskip}
+\def\frontmatter@postabstractspace{.5\baselineskip}
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+%
+%
+% \subsection{Formatting the title block}%
+%
+% \begin{environment}{titlepage}
+% This is \LaTeXe's \classname{article} class version,
+% with modifications.
+% \changes{4.0c}{1999/11/13}{grid changes with ltxgrid}
+% \changes{4.1b}{2008/08/02}{Make the syntax of this package switch optional}%
+% \begin{macrocode}
+\newenvironment{frontmatter@titlepage}{%
+ \twocolumn@sw{\onecolumngrid}{\newpage}%
+ \thispagestyle{titlepage}%
+ \setcounter{page}\@ne
+}{%
+ \twocolumn@sw{\twocolumngrid}{\newpage}%
+ \twoside@sw{}{%
+ \setcounter{page}\@ne
+ }%
+}%
+% \end{macrocode}
+% \end{environment}
+%
+% \begin{macro}{\maketitle}
+% \changes{4.0d}{1998/01/31}{hookify}
+% Put it all together to format the title block.
+%
+% Gotcha!
+% If you expand \cmd\csname undefined\cmd\endcsname, you change the meaning of \cmd\undefined\ from
+% ``Undefined'' to \cmd\relax.
+% Watchout!
+% \begin{macrocode}
+\def\frontmatter@maketitle{%
+ \@author@finish
+ \title@column\titleblock@produce
+ \suppressfloats[t]%
+% \end{macrocode}
+% Now save some memory.
+% \begin{macrocode}
+ \let\and\relax
+ \let\affiliation\@gobble
+ \let\author\@gobble
+ \let\@AAC@list\@empty
+ \let\@AFF@list\@empty
+ \let\@AFG@list\@empty
+ \let\@AF@join\@AF@join@error
+ \let\email\@gobble
+ \let\@address\@empty
+ \let\maketitle\relax
+ \let\thanks\@gobble
+ \let\abstract\@undefined\let\endabstract\@undefined
+ \titlepage@sw{%
+ \vfil
+ \clearpage
+ }{}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\maketitle@Hy}
+% \changes{4.1c}{2008/08/04}{Hyperref compatibility: take \cs{maketitle} back.}
+% We provide the means to take back the definition of \cmd\maketitle\ from \classname{hyperref}:
+% we do not need its help in providing hypertext services in the title page.
+% \begin{macrocode}
+\def\maketitle@Hy{%
+ \let\Hy@saved@footnotemark\@footnotemark
+ \let\Hy@saved@footnotetext\@footnotetext
+ \let\@footnotemark\H@@footnotemark
+ \let\@footnotetext\H@@footnotetext
+ \@ifnextchar[%]
+ \Hy@maketitle@optarg
+ {%
+ \HyOrg@maketitle
+ \Hy@maketitle@end
+ }%
+}%
+\appdef\class@documenthook{%
+ \@ifx{\maketitle\maketitle@Hy}{%
+ \class@info{Taking \string\maketitle\space back from hyperref}%
+ \let\maketitle\frontmatter@maketitle
+ }{%
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\titleblock@produce}
+% This procedure produces the title block.
+%
+% It effectively executes inside a group
+% and always returns us to vertical mode.
+%
+% Note: we assume that it is OK to set the footnote counter to zero at this point.
+% Is this a safe assumption?
+% \begin{macrocode}
+\def\titleblock@produce{%
+ \begingroup
+% \end{macrocode}
+% \changes{4.0b}{1999/06/20}{Set \cs{footnote} to \LaTeX\ standard version for this scope.}
+% First, we restore the footnote mechanism to its default state,
+% then we customize the way the footnote mark is produced and how it is formatted.
+% \changes{4.1n}{2010/01/02}{(AO, 572) Independent footnote counter for title block. Abstract footnote counter shared with body.}%
+% \begin{macrocode}
+ \ltx@footnote@pop
+% \end{macrocode}
+% The following three lines establish an independent footnote counter for use in the title block.
+% \changes{4.1n}{2010/01/02}{(AO, 572) Independent footnote counter for title block. Abstract footnote counter shared with body.}%
+% \begin{macrocode}
+ \def\@mpfn{mpfootnote}%
+ \def\thempfn{\thempfootnote}%
+ \c@mpfootnote\z@
+ \let\@makefnmark\frontmatter@makefnmark
+ \frontmatter@setup
+ \thispagestyle{titlepage}\label{FirstPage}%
+% \end{macrocode}
+% Produce the title:
+% \begin{macrocode}
+ \frontmatter@title@produce
+% \end{macrocode}
+% Produce the author list:
+% \begin{macrocode}
+ \groupauthors@sw{%
+ \frontmatter@author@produce@group
+ }{%
+ \frontmatter@author@produce@script
+ }%
+% \end{macrocode}
+% Produce the dates:
+% \begin{macrocode}
+ \frontmatter@RRAPformat{%
+ \expandafter\produce@RRAP\expandafter{\@date}%
+ \expandafter\produce@RRAP\expandafter{\@received}%
+ \expandafter\produce@RRAP\expandafter{\@revised}%
+ \expandafter\produce@RRAP\expandafter{\@accepted}%
+ \expandafter\produce@RRAP\expandafter{\@published}%
+ }%
+% \end{macrocode}
+% Produce the abstract, PACS, and keywords, and end any paragraph.
+% \changes{4.1c}{2008/08/04}{Produce PACS, keywords only if specified}%
+% \begin{macrocode}
+ \frontmatter@abstract@produce
+ \@ifx@empty\@pacs{}{%
+ \@pacs@produce\@pacs
+ }%
+ \@ifx@empty\@keywords{}{%
+ \@keywords@produce\@keywords
+ }%
+ \par
+ \frontmatter@finalspace
+ \endgroup
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@title@produce}
+% The default formatting procedure for the article title.
+% This procedure should take care of the vertical space below the title.
+%
+% Clients may override this procedure, but it is more likely that they will
+% use the hooks \cmd\frontmatter@title@above, \cmd\frontmatter@title@format, and \cmd\frontmatter@title@below.
+% \changes{4.1b}{2008/08/04}{(AO, 443) PDF Bookmark for title page}%
+% \begin{macrocode}
+\def\toclevel@title{0}%
+\def\frontmatter@title@produce{%
+ \begingroup
+ \frontmatter@title@above
+ \frontmatter@title@format
+ \@title
+ \unskip
+ \phantomsection\expandafter\@argswap@val\expandafter{\@title}{\addcontentsline{toc}{title}}%
+ \@ifx{\@title@aux\@title@aux@cleared}{}{%
+ \expandafter\frontmatter@footnote\expandafter{\@title@aux}%
+ }%
+ \par
+ \frontmatter@title@below
+ \endgroup
+}%
+\appdef\let@mark{\let\\\relax}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@title@above}
+% \begin{macro}{\frontmatter@title@format}
+% \begin{macro}{\frontmatter@title@below}
+% The default values for formatting specifications for the article title.
+%
+% The procedure \cmd\frontmatter@title@above\ should take care of the vertical space above the title;
+% \cmd\frontmatter@title@below\ below.
+% The procedure \cmd\frontmatter@title@format\ should invoke any font switches, etc.
+% that may apply to the title.
+%
+% Journals will override this procedure.
+% \begin{macrocode}
+\def\frontmatter@title@above{}%
+\def\frontmatter@title@format{}%
+\def\frontmatter@title@below{\addvspace{\baselineskip}}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \subsubsection{Authors and affiliations in superscriptaddress style}
+%
+% \begin{macro}{\frontmatter@author@produce@script}
+% \changes{4.1b}{2008/08/02}{Provide for diagnostics via \cs{frontmatterverbose@sw}}
+% \cmd\frontmatter@author@produce@script\ is an alias for \cmd\frontmatter@author@produce,
+% the procedure that formats the author/affiliation list.
+%
+% In this case, the authors affiliations are being superscripted
+% (class option \classoption{superscriptaddress}).
+%
+% This procedure must ensure that any paragraph that it starts gets ended finally.
+% \changes{4.1i}{2009/10/18}{(AO, 540) More runtime diagnostics via \cs{frontmatterverbose@sw}}
+% \changes{4.1m}{2009/11/20}{(AO, 563) More complete runtime diagnostics via \cs{frontmatterverbose@sw}}
+% \begin{macrocode}
+\def\frontmatter@author@produce@script{%
+ \begingroup
+ \let\@author@present\@author@present@script
+ \frontmatterverbose@sw{\typeout{\string\frontmatter@author@produce@script:}\say\@AAC@list\say\@AFF@list\say\@AFG@list}{}%
+% \end{macrocode}
+% \changes{4.0l}{1998/09/01}{Changes to add collaboration processing, which now means doing comma/and processing on each sublist}
+% When the author/affiliation list \cmd\@AAC@list\ is executed,
+% \cmd\@tempcnt\ counts each author,
+% and
+% \cmd\@tempa\ stores a list of author indices for the current collaboration.
+%
+% Note: this procedure uses \cmd\AU@temp\ to hold a list of author counts for each collaboration.
+% Note: also uses \cmd\@tempcnta\ to communicate between procedures.
+% This usage ot \cmd\@tempcnta\ is OK, because the expansion of \cmd\@AAC@list\ will trigger only the execution of
+% our own procedures.
+% \begin{macrocode}
+ \let\AU@temp\@empty
+ \@tempcnta\z@
+% \end{macrocode}
+%
+% We wish to know how many authors are in each collaboration, for the purposes of \cmd\listcomma\ and \cmd\listand.
+%
+% Here we assign values for the intercollated operators appearing within \cmd\@AAC@list,
+% then execute that macro, registering the authors in each collaboration.
+% Afterwards, clean up by emulating an empty collaboration.
+%
+% The result, stored in \cmd\AU@temp, is a list of brace-delmited tokens, each a number
+% representing the number of authors in that collaboration.
+% \begin{macrocode}
+ \let\AF@opr \@gobble
+ \def\AU@opr{\@author@count\@tempcnta}%
+ \def\CO@opr{\@collaboration@count\AU@temp\@tempcnta}%
+ \@AAC@list
+ \expandafter\CO@opr\@author@cleared
+% \end{macrocode}
+% Next, present the authors and their affiliations, with the collaborations interleaved.
+%
+% We assign values for the intercollated operators appearing within \cmd\@AAC@list.
+% Next, assign \cmd\@listcomma\ based on the number of authors in the first collaboration.
+% Then expand \cmd\@AAC@list.
+%
+% Upon completion, there will be two space tokens following the last author because of \cmd\listand\ processing;
+% remove them, then end the last paragraph.
+%
+% FIXME: this juncture would be a good time to effectively eliminate
+% the unused affiliations in \cmd\@AFF@list.
+% \begin{macrocode}
+ \begingroup
+ \frontmatter@authorformat
+ \let\AF@opr \@affilID@def
+ \let\AU@opr \@author@present
+ \def\CO@opr{\@collaboration@present\AU@temp}%
+ \set@listcomma@list\AU@temp
+ \@AAC@list
+ \unskip\unskip
+ \par
+ \endgroup
+% \end{macrocode}
+% We now list out the affiliations in the order they appeared.
+% \begin{macrocode}
+ \begingroup
+ \frontmatter@above@affiliation@script
+ \let\AFF@opr \@affil@script
+ \@AFF@list
+ \frontmatter@footnote@produce
+ \par
+ \endgroup
+ \endgroup
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@author@count}
+% This version of \cmd\AU@opr\ counts the number of authors it processes.
+% \begin{macrocode}
+\def\@author@count#1{%
+ \advance#1\@ne
+ \@author@gobble
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@collaboration@present}
+% Format a collaboration.
+% Note that we immediately end the paragraph thus begun, because we only support
+% \classoption{superscriptaddress}-style processing.
+%
+% Note also that the execution of the production procedures takes place inside of a group;
+% for reasons of protective programming.
+%
+% Bug note: originally, the number of authors in each collaboration was stored in \cmd\@tempa,
+% but this exposed us to procedures in our production that modified the meaning of \cmd\@tempa.
+%
+% The last action done in this procedure is to assign \cmd\@listcomma\ based on the number of authors in the next collaboration.
+% \begin{macrocode}
+\def\@collaboration@present#1#2#3#4{%
+ \par
+ \begingroup
+ \frontmatter@collaboration@above
+ \@affilID@def{}%
+ \@tempcnta\z@
+ \@author@present{}{(\ignorespaces#3\unskip)}{#4}%
+ \par
+ \endgroup
+ \set@listcomma@list#1%
+}%
+\def\frontmatter@collaboration@above{}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@collaboration@count}
+% Simply register each author in this collaboration.
+% Note: \#1 is a \cmd\csname\ to hold the value, \#2 is the \cmd\count\ register holding the count of interest.
+% \begin{macrocode}
+\def\@collaboration@count#1#2{%
+ \appdef@eval#1{\the#2}#2\z@
+ \@author@gobble
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@affilID@def}
+%
+% The \cmd\@affilID@def\ procedure, an alias of \cmd\AF@opr\ used during \cmd\frontmatter@author@produce@script\ processing,
+% memorizes the affiliation in \cmd\@affilID@temp.
+% \begin{macrocode}
+\def\@affilID@def{\def\@affilID@temp}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@affilID@temp}
+%
+% The macro \cmd\@affilID@temp\ is used to communicate between
+% \cmd\doauthor\ and instances of \cmd\AF@opr\ within the author list.
+% \begin{macrocode}
+\let\@affilID@temp\@empty
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\affils@present@script}
+% An alias of \cmd\AFG@opr, this procedure is applied to elements in \cmd\@AFG@list.
+% It builds \cmd\@AFU@list, a list of unique affiliations found within affiliation groups.
+%
+% \changes{4.1a}{2008/06/16}{removed code that had been commented out}
+% \begin{verbatim}
+%\def\affils@present@script#1{%
+% \get@affil#1{}%
+%}%
+%\def\get@affil#1{%
+% \def\@tempa{#1}%
+% \@ifx{\@empty\@tempa}{}{%
+% \@booleanfalse\temp@sw
+% \def\AF@opr{\x@match\@tempa}%
+% \@AFU@list
+% \temp@sw{}{%
+% \appdef\@AFU@list{\AF@opr{#1}}%
+% }%
+% \get@affil
+% }%
+%}%
+% \end{verbatim}
+% \end{macro}
+%
+% \begin{macro}{\affil@present@script}
+% An alias of \cmd\AF@opr, this procedure is applied to \cmd\@AFU@list\ when formatting the affiliations
+% under \classoption{superscriptaddress}.
+% It in turn applies \cmd\affil@script\ to \cmd\@AFF@list, thereby formatting the affiliation
+% associated with this affiliation index.
+%
+% Note: it traverses the \cmd\@AFF@list\ for each entry in the \cmd\@AFU@list, thereby making this
+% portion of code execute in $N^{2}$ time.
+% \changes{4.1a}{2008/06/16}{removed code that had been commented out}
+% \begin{verbatim}
+%\def\affil@present@script#1{%
+% \def\@tempa{#1}%
+% \begingroup
+% \frontmatter@affiliationfont
+% \let\AFF@opr \affil@script
+% \@AFF@list
+% \endgroup
+%}%
+% \end{verbatim}
+% \end{macro}
+%
+% \begin{macro}{\affil@script}
+% Alias of \cmd\AFF@opr, this procedure is applied to \cmd\@AFF@list\ when formatting the affiliations
+% under \classoption{superscriptaddress}.
+% It rejects all entries not matching the affiliation index held in \cmd\@tempa.
+% \begin{macrocode}
+\def\affil@script#1#2#3{%
+ \def\@tempifx{#1}\@ifx{\@tempifx\@tempa}{%
+ \@if@empty{#2}{}{%
+ \par
+ \begingroup
+ \def\@thefnmark{#1}\@makefnmark\ignorespaces
+ #2%
+ \@if@empty{#3}{}{\frontmatter@footnote{#3}}%
+ \par
+ \endgroup
+ }%
+ }{}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@affil@script}
+% Alias of \cmd\AFF@opr:
+% The affiliations are being formatted in the case
+% where affiliations are being superscripted:
+% make a list out of the affiliations with the numbers printed.
+% Society-specific code can change the formatting
+% by overriding the definition of \cmd\frontmatter@affiliationfont.
+%
+% A client may choose to set \cmd\affil@cutoff\ to either \cmd\@ne\ (the default)
+% or \cmd\tw@. The latter setting will enforce a rule to the effect that
+% if a single affiliation appears on the title page,
+% then no affiliation superscript at all is produced.
+%
+% |#1|---Affiliation ID: a number; zero signifies a \cmd\noaffiliation.
+%
+% |#2|---Collaboration ID
+%
+% |#3|---Affiliation address
+%
+% |#4|---Auxiliary info
+% \begin{macrocode}
+\def\@affil@script#1#2#3#4{%
+% \end{macrocode}
+% If the affiliation is the \cmd\noaffiliation, then we do nothing.
+% \begin{macrocode}
+ \@ifnum{#1=\z@}{}{%
+ \par
+ \begingroup
+ \frontmatter@affiliationfont
+ \@ifnum{\c@affil<\affil@cutoff}{}{%
+ \def\@thefnmark{#1}\@makefnmark
+ }%
+ \ignorespaces#3%
+ \@if@empty{#4}{}{\frontmatter@footnote{#4}}%
+ \par
+ \endgroup
+ }%
+}%
+\let\affil@cutoff\@ne
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@author@present@script}
+% \changes{4.0h}{1998/06/25}{Changes to support groups of affiliations}
+% \changes{4.1n}{2009/11/30}{(AO, 567) Handle \cs{noaffiliation} case}
+% This version of \cmd\AU@opr\ applies to the \classoption{superscriptaddress} class option.
+% Need to add commas between groups
+% of address numbers, which are passed in the macro \cmd\@affilID@temp\ in the form
+% |{3}{4}{7}| if this set of authors is related to addresses, 3, 4 and 7.
+% \begin{macrocode}
+\def\@author@present@script#1#2#3{%
+ \begingroup
+ \gdef\comma@space{\textsuperscript{,\,}}%
+ \doauthor{#2}{#3}{\@affil@present@script}%
+ \endgroup
+ \advance\@tempcnta\m@ne
+}%
+\def\@affilcomma#1#2{%
+ \@ifx{\z@#1}{%
+ \@ifx{\relax#2}{}{%
+ \@affilcomma{#2}%
+ }%
+ }{%
+ #1%
+ \@ifx{\relax#2}{}{%
+ \@ifx{\z@#2}{%
+ \@affilcomma
+ }{%
+ ,\,\@affilcomma{#2}%
+ }%
+ }%
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@affil@present@script}
+% \begin{macro}{\@affil@present@script@}
+% \changes{4.1n}{2009/11/30}{(AO, 567) Handle punctuation for \classoption{superscriptaddress} style authors' extra information, such as \cs{email}}
+% The \cmd\@affil@present@script\ procedure presents the reference to the affiliations in the case of superscript addresses.
+%
+% The affiliations are presented to \cmd\@affil@present@script@ as a list of brace-delimited tokens, each containing the affiliation ID (a number).
+% The list is terminated by the token \cmd\relax.
+% The macro recurses (via tail recursion) until the terminating token is encountered.
+% Note that an affiliation ID of zero signifies a \cmd\noaffiliation:
+% references to the \cmd\noaffiliation\ are passed over.
+%
+% Upon termination, the affiliations, if any, are produced via \cmd\textsuperscript\ and the appropriate boolean is queued via \cmd\aftergroup.
+% The boolean signifies that a non-trivial affiliation was produced, so that the proper punctuation can be produced.
+% The group in question is that opened and closed by the \cmd\doauthor\ procedure.
+%
+% This procedure uses \cmd\@tempa\ as a scratch register.
+% \begin{macrocode}
+\def\@affil@present@script{%
+ \let\@tempa\@empty
+ \expandafter\@affil@present@script@\@affilID@temp\relax
+}%
+\def\@affil@present@script@#1{%
+ \@ifx{\relax#1}{%
+ \@ifx{\@tempa\@empty}{%
+ \aftergroup\false@sw
+ }{%
+ \textsuperscript{\expandafter\@affilcomma\@tempa\relax\relax}%
+ \aftergroup\true@sw
+ }%
+ }{%
+ \@ifnum{#1=\z@}{}{\appdef\@tempa{{#1}}}%
+ \@affil@present@script@
+ }%
+}%
+\@provide\@author@parskip{\z@skip}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+%
+% \subsubsection{Authors and affiliations in groupedaddress style}
+%
+% \begin{macro}{\frontmatter@author@produce@group}
+% \changes{4.1b}{2008/08/02}{Provide for diagnostics via \cs{frontmatterverbose@sw}}
+% Alias for \cmd\frontmatter@author@produce when \classoption{groupedaddress} is in effect,
+% the procedure that formats the author/affiliation list.
+%
+% In this case, the authors are being grouped above their shared addresses.
+% How it works: the \cmd\@AFG@list, a list of all unique affiliation groups, is expanded.
+% For each such group, all authors belonging to that group are formatted above the affiliation group.
+% Collaborations are silently passed over.
+%
+% This procedure must ensure that any paragraph that it starts gets ended finally.
+% \changes{4.1i}{2009/10/18}{(AO, 540) More runtime diagnostics via \cs{frontmatterverbose@sw}}
+% \changes{4.1m}{2009/11/20}{(AO, 563) More complete runtime diagnostics via \cs{frontmatterverbose@sw}}
+% \begin{macrocode}
+\def\frontmatter@author@produce@group{%
+ \begingroup
+ \let\@author@present\@author@present@group
+ \frontmatter@authorformat
+ \frontmatterverbose@sw{\typeout{\string\frontmatter@author@produce@group:}\say\@AAC@list\say\@AFF@list\say\@AFG@list}{}%
+ \let\AU@temp\@empty
+ \set@listcomma@list\AU@temp
+ \def\CO@opr{\@collaboration@present\AU@temp}%
+ \let\AFG@opr \affils@present@group
+ \let\@listcomma\relax
+ \@AFG@list
+ \frontmatter@footnote@produce
+ \par
+ \endgroup
+ \frontmatter@authorbelow
+}%
+\@provide\frontmatter@authorbelow{}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\affils@present@group}
+% \changes{4.1b}{2008/08/02}{Provide for diagnostics via \cs{frontmatterverbose@sw}}
+% Alias of \cmd\AFG@opr, this procedure is used on every element of
+% \cmd\@AFG@list, the list of unique affiliation groups,
+% when formatting the author/affiliation block:
+% It presents all authors sharing this affiliation group.
+%
+% This procedure is called via the procedure
+% \cmd\frontmatter@author@produce@group, an alias for \cmd\frontmatter@author@produce,
+%
+% Its use pertains to the class options
+% \classoption{groupedaddress}, \classoption{unsortedaddress}, and \classoption{runinaddress}
+% (that is, \cmd\groupauthors@sw\ is true).
+%
+% For each invocation, the entire \cmd\@AAC@list\ is run through (twice),
+% making this process execute in $N^{2}$ time. (Is this really true?)
+% \begin{macrocode}
+\def\affils@present@group#1{%
+% \end{macrocode}
+% Count (using \cmd\@tempcnta) the authors in the author list whose affiliation group
+% matches the argument |#1|.
+%
+% Note that \cmd\AF@temp\ is used as a storage register for the argument
+% of the procedure, for purposes of comparison using \cmd\ifx.
+%
+% Note: here we use \cmd\@tempcnta\ to accumulate the number of authors in the current affiliation set.
+%
+% QUERY: do the value of \cmd\AF@temp\ and \cmd\AF@opr\ need to persist?
+%
+% QUERY: what value does \cmd\AU@opr\ have at this point?
+% ANSWER: the \cmd\@affilID@count\ and \cmd\@affilID@match\ procedures
+% assign meanings to \cmd\AU@opr\ and \cmd\CO@opr\ dynamically.
+% We initialize the two operators to \cmd\@undefined\ so as to catch the
+% pathological cases.
+% \begin{macrocode}
+ \begingroup
+ \def\AF@temp{#1}%
+ \@tempcnta\z@
+ \let\AU@opr \@undefined
+ \let\CO@opr \@undefined
+ \def\AF@opr{\@affilID@count\AF@temp\@tempcnta}%
+ \@AAC@list
+% \end{macrocode}
+%
+% If there are no authors using this affiliation group, skip it.
+% \begin{macrocode}
+ \@ifnum{\@tempcnta=\z@}{}{%
+% \end{macrocode}
+%
+% DPC: If we have a list of length two, need to locally zap a comma.
+% \begin{macrocode}
+ \begingroup
+ \frontmatter@above@affilgroup
+ \set@listcomma@count\@tempcnta
+% \end{macrocode}
+% Run through the author list, presenting (with \cmd\@author@present)
+% those authors whose affiliation matched the given one.
+% \begin{macrocode}
+ \let\AU@opr \@undefined
+ \let\CO@opr \@undefined
+ \def\AF@opr{\@affilID@match\AF@temp}%
+ \@AAC@list
+ \endgroup
+% \end{macrocode}
+% Now that the authors have all been presented, present the affiliations, grouped.
+%
+% Note: \cmd\@tempcnta\ is the number of affiliations for this set of authors.
+% \begin{macrocode}
+ \begingroup
+ \par
+ \frontmatter@above@affiliation
+ \frontmatter@affiliationfont
+ \let\\\frontmatter@addressnewline
+ \@tempcnta\z@
+ \@tfor\AF@temp:=#1\do{%
+ \expandafter\@ifx\expandafter{\expandafter\z@\AF@temp}{}{%
+ \advance\@tempcnta\@ne
+ }%
+ }%
+ \@ifnum{\@tempcnta=\tw@}{%
+ \let\@listcomma\relax
+ }{}%
+ \def@after@address
+ \runinaddress@sw{%
+ }{%
+ \tightenlines@sw{}{%
+% \end{macrocode}
+% The following line determines the space between affilations in grouped mode
+% \begin{macrocode}
+ \parskip\z@
+ }%
+ \appdef\after@address\par
+ }%
+ \let\AFF@opr \@affil@group
+ \do@affil@fromgroup\@AFF@list#1\relax
+ \endgroup
+% \end{macrocode}
+% End of branch handling authors.
+% \begin{macrocode}
+ }%
+% \end{macrocode}
+%
+% \begin{macrocode}
+ \par
+ \endgroup
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\def@after@address}
+% \begin{macro}{\def@after@address@empty}
+% \changes{4.1c}{2008/08/04}{Parametrize the production after the address}%
+%
+% Each of these procedures assigns a meaning to \cmd\after@address, the first
+% a useful default, the second a nil.
+% A society or journal substyle may define it according to its own requirements.
+% \begin{macrocode}
+\def\def@after@address{\def\after@address{\@listcomma\ \@listand}}%
+\def\def@after@address@empty{\let\after@address\@empty}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\@affilID@count}
+% \begin{macro}{\@affilID@match}
+%
+% The control sequence name \cmd\AF@opr\ is placed in the
+% author/affiliation list (\cmd\@AAC@list) for each (new) affiliation, with that affiliation as its argument.
+%
+% The control sequence name \cmd\AF@opr\ is an alias of a procedure
+% to count the number of authors at a given affiliation (\cmd\@affilID@count)
+% or to present the given authors (\cmd\@affilID@match).
+% These two procedures are executed when formatting under the \classoption{groupedaddress} class option.
+% \begin{macrocode}
+\def\@affilID@count#1#2#3{%
+ \def\@tempifx{#3}%
+ \@ifx{\@tempifx#1}{%
+ \def\AU@opr{\@author@count#2}%
+ }{%
+ \let\AU@opr \@author@gobble
+ }%
+ \let\CO@opr \@collaboration@gobble
+}%
+\def\@affilID@match#1#2{%
+ \def\@tempifx{#2}%
+ \@ifx{\@tempifx#1}{%
+ \let\AU@opr \@author@present
+ }{%
+ \let\AU@opr \@author@gobble
+ }%
+ \let\CO@opr \@collaboration@gobble
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\do@affil@fromgroup}
+% Executes iteratively:
+% selects the next address ID to print,
+% and then re-execute the list of addresses to print the text of that address.
+%
+% Note: an argument of \cmd\relax\ breaks out of this iteration,
+% while an argument of \cmd\z@\ (\cmd\noaffiliation) is a no-op.
+%
+% Note: we have created an $N^{2}$ computing process.
+% FIXME: can use hashing instead, you know.
+% \begin{macrocode}
+\def\do@affil@fromgroup#1#2{%
+ \@ifx{\relax#2}{}{%
+ \count@#2\relax
+ \@ifnum{\z@=\count@}{}{#1}%
+ \do@affil@fromgroup#1%
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@affil@group}
+% The alias of \cmd\AFF@opr\ when the list \cmd\@AFF@list\
+% is executed by \cmd\do@affil@fromgroup\ (\classoption{groupedaddress}):
+% Print the address if its number matches \cmd\count@\ (value set by \cmd\do@affil@fromgroup).
+%
+% Note that we detect a \cmd\noaffiliation\ here by the affiliation address being
+% the same as the expansion of \cmd\blankaffiliation.
+%
+% \changes{4.1d}{2008/10/17}{(AO, 410) compare \cs{@tempa} to \cs{blankaffiliation}.}%
+%
+% I made a mistake when coding the comparison to \cmd\blankaffiliation:
+% it originally read \cmd\@ifx\{\#3\cmd\blankaffiliation\}.
+% This is obviously wrong once you think twice.
+% \begin{macrocode}
+\def\@affil@group#1#2#3#4{%
+ \@ifnum{#1=\count@}{%
+ \def\@tempa{#3}%
+ \@ifx{\@tempa\blankaffiliation}{}{%
+ #3%
+ \@if@empty{#4}{}{%
+ \frontmatter@footnote{#4}%
+ }%
+ \after@address
+ }%
+ \advance\@tempcnta\m@ne
+ }{}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@author@present@group}
+% \begin{macro}{\@affil@present@group}
+% \changes{4.0d}{1998/01/31}{\cs{comma@space}}
+% This version of \cmd\AU@opr\ applies to the \classoption{groupedaddress} class option and its ilk.
+% Traverse the list calling the \cmd\doauthor\ hook on each author.
+%
+% The third argument of \cmd\doauthor\ fulfills the requirement that it
+% queue up a boolean via \cmd\aftergroup.
+% \begin{macrocode}
+\def\@author@present@group#1#2#3{%
+ \gdef\comma@space{\gdef\comma@space{\textsuperscript{,\,}}}%
+ \doauthor{#2}{#3}{\@affil@present@group}%
+ \advance\@tempcnta\m@ne
+}%
+\def\@affil@present@group{%
+ \aftergroup\false@sw
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+%
+% \begin{macro}{\@pacs@produce}
+% PACS, keywords and dates.
+% \changes{4.0d}{1998/01/31}{PACS boxed to width of abstract}
+% \changes{4.0b}{1999/06/20}{Separate option now controls production}
+% \changes{4.0e}{2000/09/20}{Warn if no production}
+% \begin{macrocode}
+\def\@pacs@produce#1{%
+ \showPACS@sw{%
+ \begingroup
+ \frontmatter@PACS@format
+ \@pacs@name#1\par
+ \endgroup
+ }{%
+ \@if@empty{#1}{}{%
+ \class@warn{\PACS@warn}%
+ }%
+ }%
+}%
+\def\PACS@warn{If you want your PACS to appear in your output, use document class option showpacs}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@keywords@produce}
+% \changes{4.0e}{2000/09/20}{Keywords on same footing as PACS}
+% \begin{macrocode}
+\def\@keywords@produce#1{%
+ \showKEYS@sw{%
+ \begingroup
+ \frontmatter@keys@format
+ \@keys@name#1\par
+ \endgroup
+ }{%
+ \@if@empty{#1}{}{%
+ \class@warn{If you want your keywords to appear in your output, use document class option showkeys}%
+ }%
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@footnote@produce@footnote}
+% \begin{macro}{\frontmatter@footnote@produce@endnote}
+% Produce any footnotes to appear at the bottom of the title page.
+%
+% If frontmatter footnotes have been redirected to the bibliography, this will be a no-op.
+% ref.: \cmd\present@bibnote.
+% \begin{macrocode}
+\def\frontmatter@footnote@produce@footnote{%
+ \let\@TBN@opr\present@FM@footnote
+ \@FMN@list
+ \global\let\@FMN@list\@empty
+}%
+% \end{macrocode}
+% \changes{4.1n}{2010/01/02}{(AO, 572) Independent footnote counter for title block. Abstract footnote counter shared with body.}%
+% \begin{macrocode}
+\def\present@FM@footnote#1#2{%
+ \begingroup
+ \csname c@\@mpfn\endcsname#1\relax
+ \def\@thefnmark{\frontmatter@thefootnote}%
+ \frontmatter@footnotetext{#2}%
+ \endgroup
+}%
+\def\frontmatter@footnote@produce@endnote{%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \section{Package defaults}
+%
+% \begin{macro}{\title@column}
+% \begin{macro}{\preprintsty@sw}
+% \begin{macro}{\collaboration@sw}
+% \begin{macro}{\frontmatter@footnote@produce}
+% \begin{macro}{\do@output@MVL}
+% \begin{macro}{\comma@space}
+% We give meanings to the Booleans and other macros that a client document class would
+% set on our behalf.
+% \begin{macrocode}
+\appdef\frontmatter@init{%
+ \@ifxundefined\title@column {\let\title@column\@empty}{}%
+ \@ifxundefined\preprintsty@sw {\@booleanfalse\preprintsty@sw}{}%
+ \@ifxundefined\frontmatter@footnote@produce{\let\frontmatter@footnote@produce\frontmatter@footnote@produce@footnote}{}%
+ \@ifxundefined\do@output@MVL {\let\do@output@MVL\@firstofone}{}%
+ \@ifxundefined\comma@space {\let\comma@space\@empty}{}%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@thefootnote}
+% \changes{4.1n}{2010/01/02}{(AO, 572) Independent footnote counter for title block. Abstract footnote counter shared with body.}%
+% In the title block, the footnote symbol (used for alternate affiliation and sundry)
+% is a \cmd\@fnsymbol\ or a lowercase letter, as selected by a boolean.
+% The society or journal may override this definition.
+% \begin{macrocode}
+\def\frontmatter@thefootnote{%
+ \altaffilletter@sw{\@alph}{\@fnsymbol}{\csname c@\@mpfn\endcsname}%
+}%
+\@ifx{\altaffilletter@sw\@undefined}{\@booleantrue\altaffilletter@sw}{}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@makefnmark}
+% The default procedure for setting the footnote mark within the frontmatter.
+% A client may override this procedure.
+% \begin{macrocode}
+\def\frontmatter@makefnmark{%
+ \@textsuperscript{%
+ \normalfont\@thefnmark
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@makefntext}
+% This procedure is used when setting footnotes within the title block.
+%
+% Note that we are \classname{hyperref} friendly: we invoke \cmd\Hy@raisedlink\
+% so that this footnote becomes a hypertext anchor. If \classname{hyperref} is
+% not loaded, then \classname{ltxutil} has our act covered via some stubs.
+%
+% \changes{4.1n}{2010/01/02}{(AO, 572) Independent footnote counter for title block. Abstract footnote counter shared with body.}%
+% \begin{macrocode}
+\long\def\frontmatter@makefntext#1{%
+ \parindent 1em
+ \noindent
+ \Hy@raisedlink{\hyper@anchorstart{frontmatter.\expandafter\the\csname c@\@mpfn\endcsname}\hyper@anchorend}%
+ \@makefnmark
+ #1%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@setup}
+% The default type specification for the title page.
+% Clients will override this definition.
+% \begin{macrocode}
+\def\frontmatter@setup{}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@RRAPformat}
+% \begin{macro}{\punct@RRAP}
+% \begin{macro}{\produce@RRAP}
+% The procedure \cmd\frontmatter@RRAPformat\ is the formatting procedure
+% for the ``Received, Revised, etc.'' block on the title page.
+% The macro \cmd\punct@RRAP\ produces the punctuation between the entries.
+% Clients will override these two definitions.
+%
+% The procedure \cmd\produce@RRAP\ is fairly general, and should be used by most
+% journal substyles.
+%
+% The procedure \cmd\frontmatter@RRAP@format\ will be overridden by the journal.
+% \begin{macrocode}
+\def\frontmatter@RRAPformat#1{%
+ \removelastskip
+ \begingroup
+ \frontmatter@RRAP@format
+ #1\par
+ \endgroup
+}%
+\def\punct@RRAP{; }%
+\def\produce@RRAP#1{%
+ \@if@empty{#1}{}{%
+ \@ifvmode{\leavevmode}{\unskip\punct@RRAP\ignorespaces}%
+ #1%
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@authorformat}
+% The default type specification for the author list.
+% Clients will override this definition.
+% For one alternative, see the \classname{revtex4} document class.
+% \begin{macrocode}
+\def\frontmatter@authorformat{}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@above@affilgroup}
+% The default amount of space above an affiliation group
+% (in a \cmd\affils@present@group\ production).
+% Clients will override this definition.
+% \begin{macrocode}
+\def\frontmatter@above@affilgroup{}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@above@affiliation}
+% \begin{macro}{\frontmatter@above@affiliation@script}
+% The \cmd\frontmatter@above@affiliation\ is the amount of space above affiliations in the \classoption{groupedaddress}
+% style, and \cmd\frontmatter@above@affiliation@script\ is that for \classoption{superscriptaddress}.
+% Clients will override this definition.
+% For one alternative, see the \classname{revtex4} document class.
+% \begin{macrocode}
+\def\frontmatter@above@affiliation{}%
+\def\frontmatter@above@affiliation@script{}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@affiliationfont}
+% The default type specification for the affiliation.
+% Clients will override this definition.
+% For one alternative, see the \classname{revtex4} document class, where
+% several alternative definitions are made.
+% \begin{macrocode}
+\def\frontmatter@affiliationfont{\itshape\selectfont}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@RRAP@format}
+% The default type specification for the dates.
+% Clients will override this definition.
+% \begin{macrocode}
+\def\frontmatter@RRAP@format{}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@PACS@format}
+% Specify the formatting of the title page PACS statement.
+%
+% Journals will override.
+% \begin{macrocode}
+\def\frontmatter@PACS@format{}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@keys@format}
+% Specify the formatting of the title page keywords statement.
+%
+% Journals will override.
+% \begin{macrocode}
+\def\frontmatter@keys@format{}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@finalspace}
+%
+% \begin{macrocode}
+\def\frontmatter@finalspace{\addvspace{18\p@}}
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@addressnewline}
+% \changes{4.0h}{1998/06/25}{new hook}
+% The definition of \cmd\\\ for address handling. Default puts all the `lines'
+% on a run-in line, separated by comma and space.
+% DPC: was \newline space between lines of addresss.
+% \begin{macrocode}
+\def\frontmatter@addressnewline{%
+ \@ifhmode{\skip@\lastskip\unskip\unpenalty\break\hskip\skip@}{}%
+ % was: \vskip-.5ex
+}%
+% \end{macrocode}
+% \end{macro}
+%
+%
+% \begin{macro}{\frontmatter@preabstractspace}
+% \begin{macro}{\frontmatter@postabstractspace}
+% Space above and space below abstract in title block
+% \begin{macrocode}
+\def\frontmatter@preabstractspace{5.5\p@}
+\def\frontmatter@postabstractspace{6.5\p@}
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\aftermaketitle@chk}
+% Error-checking code.
+% \begin{macrocode}
+\def\aftermaketitle@chk#1{%
+ \@ifx{\maketitle\relax}{%
+ \class@err{\protect#1 must be used before \protect\maketitle}%
+ }{}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\ps@titlepage}
+% Default page style for title page.
+% Journals will override this procedure.
+% \begin{macrocode}
+\def\ps@titlepage{\ps@empty}%
+% \end{macrocode}
+% \end{macro}
+%
+%FIXME: there is a limitation to the default meaning of
+% \cmd\@startpage: the label ``FirstPage'' is only
+% defined if the document has a \cmd\maketitle\ command.
+%
+% \begin{macrocode}
+\def\volumeyear#1{\gdef\@volumeyear{#1}}%
+\def\@volumeyear{}%
+\def\volumenumber#1{\gdef\@volumenumber{#1}}%
+\def\@volumenumber{}%
+\def\issuenumber#1{\gdef\@issuenumber{#1}}%
+\def\@issuenumber{}%
+\def\eid#1{\gdef\@eid{#1}}%
+\def\@eid{}%
+%
+\def\startpage#1{\gdef\@startpage{#1}\c@page#1\relax}%
+\def\@startpage{\pageref{FirstPage}}%
+\def\endpage#1{\gdef\@endpage{#1}}%
+\def\@endpage{\pageref{LastPage}}%
+% \end{macrocode}
+%
+%
+% \subsection{Printing out the ``list-of'' elements}%
+%
+% FIXME: The \cmd\appendix@toc\ procedure should change the meaning
+% of \cmd\l@section\ so that the \cmd\section s can be appropriately formatted,
+% reflecting their status as appendices.
+%
+% \begin{macrocode}
+\def\print@toc#1{%
+ \begingroup
+ \expandafter\section
+ \expandafter*%
+ \expandafter{%
+ \csname#1name\endcsname
+ }%
+ \let\appendix\appendix@toc
+ \@starttoc{#1}%
+ \endgroup
+}%
+\def\appendix@toc{}%
+% \end{macrocode}
+%
+% \begin{macro}{\Dated@name}
+% \begin{macro}{\Received@name}
+% \begin{macro}{\Revised@name}
+% \begin{macro}{\Accepted@name}
+% \begin{macro}{\Published@name}
+% These strings are used in the \cmd\date, et al. commands.
+% \begin{macrocode}
+\def\Dated@name{Dated }%
+\def\Received@name{Received }%
+\def\Revised@name{Revised }%
+\def\Accepted@name{Accepted }%
+\def\Published@name{Published }%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \changes{4.1g}{2009/10/05}{(AO, 539) Robustify \cs{class@warn} against TOC processing}%
+% Two commands require robustifying against harsh treatment when TOC entries are written out.
+% Note that \cmd\class@warn\ is not user-level markup, but crops up when
+% the \cmd\title\ command is missing from the document.
+% \begin{macrocode}
+\appdef\robustify@contents{%
+ \let\thanks\@gobble\let\class@warn\@gobble
+ \def\begin{\string\begin}\def\end{\string\end}%
+}%
+% \end{macrocode}
+%
+% \subsection{Syntax switch}%
+% \changes{4.1b}{2008/08/02}{Make the syntax of this package switch optional}%
+%
+% Shall we take over the syntax of \LaTeX?
+% By default, Yes, unless the Boolean below has already been defined.
+% In the package version, this is controlled by an option;
+% in the kernel version, the Boolean remains undefined.
+%
+% \cmd\maketitle\ presents a complication. If this code will execute under
+% \classname{ltxgrid}, then its version of the procedure will do fine.
+%
+% As a package, under \classname{article}, things are different.
+% In this case, what should be done depends upon \cmd\if@titlepage.
+% If true, then just execute \cmd\maketitle\ as it stands.
+% If false, then the \classname{article} class has defined its wrapper
+% procedure, and we should take over the meaning of \cmd\@maketitle\ instead.
+%
+% Here is a list of the commands defined by this package that potentially override
+% those of standard \LaTeX:
+% \cmd\frontmatter@title,
+% \cmd\frontmatter@author,
+% \cmd\frontmatter@and,
+% \cmd\frontmatter@thanks,
+% \cmd\frontmatter@date,
+% \env{frontmatter@abstract},
+% \env{frontmatter@titlepage},
+% \cmd\frontmatter@maketitle.
+%
+% The following code will perform that override, given the appropriate state of the Boolean.
+% \begin{macrocode}
+\@ifxundefined\frontmatter@syntax@sw{\@booleantrue\frontmatter@syntax@sw}{}%
+\frontmatter@syntax@sw{%
+ \let\title \frontmatter@title
+ \let\author \frontmatter@author
+ \let\date \frontmatter@date
+ \@ifxundefined\@maketitle{%
+ \let\maketitle \frontmatter@maketitle
+ \@booleantrue \titlepage@sw
+ }{%
+ \let\@maketitle \frontmatter@maketitle
+ \prepdef\maketitle\@author@finish
+ }%
+ \let\noaffiliation \frontmatter@noaffiliation
+ \let\thanks@latex \thanks
+ \let\thanks \frontmatter@thanks
+ \let\and@latex \and
+ \let\and \frontmatter@and
+ \let@environment{titlepage}{frontmatter@titlepage}%
+ \let@environment{abstract}{frontmatter@abstract}%
+}{%
+ \let\noaffiliation\@empty
+}%
+% \end{macrocode}
+%
+% \subsection{End of the \file{ltxfront} kernel}
+% The kernel portion of the \classname{ltxfront} package is complete.
+% \begin{macrocode}
+%</kernel>
+% \end{macrocode}
+%
+% \subsection{Remainder of the \classname{ltxfront} package}
+%
+% Include any code here that ought to be incorporated into the package,
+% but should not be part of the kernel.
+%
+% \begin{macrocode}
+%<*package>
+% \end{macrocode}
+%
+% \begin{macrocode}
+% \end{macrocode}
+% Define formatting as like as possible to the article class.
+% Comands that prepare for \cmd\maketitle: \cmd\author, \cmd\title, \cmd\date.
+%
+% Argument of the above three may contain \cmd\\ and \cmd\thanks; argument of \cmd\author may contain \cmd\and.
+% The \cmd\\ may still work fine; \cmd\and\ will be ignored (warning);
+% \cmd\thanks\ even though it should not reside in the argument, will work.
+%
+% An explicit \cmd\footnote should be treated as if it were \cmd\thanks.
+%
+% \begin{verbatim}
+% \@maketitle: \frontmatter@footnote@produce
+% \end{verbatim}
+% \changes{4.1n}{2010/01/02}{(AO, 572) Independent footnote counter for title block. Abstract footnote counter shared with body.}%
+%
+% \begin{macrocode}
+\def\thanks@latex#1{%
+ \footnotemark
+ \expandafter\expandafter
+ \expandafter\gappdef
+ \expandafter\expandafter
+ \expandafter\@thanks
+ \expandafter\expandafter
+ \expandafter{%
+ \expandafter\expandafter
+ \expandafter\footnotetext
+ \expandafter\expandafter
+ \expandafter[%
+ \expandafter\the\csname c@\@mpfn\endcsname]{#1}}%
+}%
+\@booleanfalse\altaffilletter@sw
+\@if@sw\if@titlepage\fi{\@booleantrue}{\@booleanfalse}\titlepage@sw
+\def\frontmatter@title@above{\newpage\null\vskip2em\relax}%
+\def\frontmatter@title@format{\centering\LARGE\let\thanks\thanks@latex}%
+\def\frontmatter@title@below{\vskip1.5em\relax}%
+\def\frontmatter@authorformat{\centering\large\advance\baselineskip\p@\parskip11.5\p@\let\thanks\thanks@latex\let\and\and@space}%
+\def\frontmatter@authorbelow{\vskip 1em\relax}%
+\def\frontmatter@above@affiliation{}%
+\def\frontmatter@above@affiliation@script{}%
+\def\frontmatter@affiliationfont{\centering\itshape}%
+\def\frontmatter@RRAP@format{\centering\large}%
+\def\frontmatter@preabstractspace{1.5em}%
+% \end{macrocode}
+% \changes{4.1n}{2010/01/02}{(AO, 572) Independent footnote counter for title block. Abstract footnote counter shared with body.}%
+% \begin{macrocode}
+\long\def\frontmatter@footnotetext{%
+ \expandafter\expandafter
+ \expandafter\footnotetext
+ \expandafter\expandafter
+ \expandafter[%
+ \expandafter\the\csname c@\@mpfn\endcsname]%
+}%
+\def\and@space{\\}%
+\def\andname{and}%
+% \end{macrocode}
+% Implement the feature of the article class whereby there is a default \cmd\@date, left over
+% from the \LaTeX\ kernel.
+% This package does not have a default date. The user must put in an explicit command, like
+% \cmd\date\arg{\cmd\today}, which will retore \LaTeX's default behavior.
+%
+% What if the abstract environment falls after the \cmd\maketitle\ command?
+% The environment will be undefined; the user will have to deal with this.
+% \begin{macrocode}
+% \end{macrocode}
+% End of the package.
+% \begin{macrocode}
+%</package>
+% \end{macrocode}
+%
+% \Finale
+% %Here ends the programmer's documentation.
+% \endinput
+%
+\endinput
+%%EOF
diff --git a/texmf/source/latex/revtex/ltxgrid.dtx b/texmf/source/latex/revtex/ltxgrid.dtx
new file mode 100644
index 0000000..6d0c659
--- /dev/null
+++ b/texmf/source/latex/revtex/ltxgrid.dtx
@@ -0,0 +1,6849 @@
+% \iffalse meta-comment balanced on line 109
+% ltxgrid.dtx: package to change page grid, MVL.
+% Copyright (c) 2009 Arthur Ogawa
+%
+% Disclaimer
+% This file is distributed WITHOUT ANY WARRANTY;
+% without even the implied warranty of
+% MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
+% License
+% You may distribute this file under the conditions of the
+% LaTeX Project Public License 1.3c or later
+% (http://www.latex-project.org/lppl.txt).
+% ReadMe
+% For the documentation and more detailed instructions for
+% installation, typeset this document with \LaTeX.
+% Maintenance Status
+% This work has the LPPL maintenance status "maintained";
+% Current Maintainer of this work is Arthur Ogawa
+% changes for version 4.2d and 4.2e by Phelype Oleinik.
+%
+% This work consists of the main source file ltxgrid.dtx
+% and the derived files
+% ltxgrid.sty, ltxgrid.pdf
+% Distribution:
+% CTAN:macros/latex/contrib/revtex/
+%
+% Unpacking:
+% tex ltxgrid.dtx
+%
+% Documentation:
+% latex ltxgrid.dtx; ...
+%
+% Programm calls to get the documentation (example):
+% pdflatex ltxgrid.dtx
+% makeindex -s gind.ist ltxgrid
+% makeindex -s gglo.ist -o ltxgrid.gls ltxgrid.glo
+% pdflatex ltxgrid.dtx
+% makeindex -s gind.ist ltxgrid.idx
+% pdflatex ltxgrid.dtx
+%
+% Installation:
+% TDS:doc/latex/revtex/
+% TDS:tex/latex/revtex/
+% TDS:source/latex/revtex/
+%
+% Thanks, Heiko!
+% This method of letting a single .dtx file serve as both
+% documentation (via latex) and installer (via tex) follows
+% the example of Heiko Oberdiek. Thanks!
+%<*ignore>
+\begingroup
+ \def\x{LaTeX2e}%
+\expandafter\endgroup
+\ifcase
+ 0\expandafter\ifx\csname processbatchFile\endcsname\relax\else1\fi\ifx\fmtname\x\else 1\fi
+ \relax
+\else
+ \csname fi\endcsname
+%</ignore>
+%<*install>
+%% This file will generate documentation and runtime files
+%% from ltxgrid.dtx when run through LaTeX or TeX.
+\input docstrip
+\preamble
+
+This is a generated file;
+altering it directly is inadvisable;
+instead, modify the original source file.
+See the URL in the file README-LTXGRID.tex.
+
+License
+ You may distribute this file under the conditions of the
+ LaTeX Project Public License 1.3c or later
+ (http://www.latex-project.org/lppl.txt).
+
+ This file is distributed WITHOUT ANY WARRANTY;
+ without even the implied warranty of MERCHANTABILITY
+ or FITNESS FOR A PARTICULAR PURPOSE.
+
+\endpreamble
+\askforoverwritefalse
+\keepsilent
+ \generate{%
+ %{ignore}
+% \usedir{tex/latex/revtex}%
+ \file{ltxgrid.sty}{%
+ \from{ltxgrid.dtx}{package,kernel}%
+ }%
+ }%
+\ifToplevel{
+\Msg{***********************************************************}
+\Msg{*}
+\Msg{* To finish the installation, please move}
+\Msg{* ltxgrid.sty}
+\Msg{* into a directory searched by TeX;}
+\Msg{* in a TDS-compliant installation:}
+\Msg{* texmf/tex/macros/latex/revtex/.}
+\Msg{*}
+\Msg{* To produce the documentation,
+ run ltxgrid.dtx through LaTeX.}
+\Msg{*}
+\Msg{* Happy TeXing}
+\Msg{***********************************************************}
+}
+\endbatchfile
+%</install>
+%<*ignore>
+\fi
+%</ignore>
+% \fi
+%
+% \GetFileInfo{ltxgrid.dtx}
+%
+% \iffalse ltxdoc klootch
+%<*package>
+%%% @LaTeX-file{
+%%% filename = "ltxgrid.dtx",
+%%% version = "4.2e",
+%%% date = "2020/10/03",
+%%% author = "Arthur Ogawa (mailto:arthur_ogawa at sbcglobal.net),
+%%% Phelype Oleinik (mailto:phelype.oleinik at latex-project.org),
+%%% commissioned by the American Physical Society.
+%%% ",
+%%% copyright = "Copyright (C) 1999, 2009 Arthur Ogawa,
+%%% distributed under the terms of the
+%%% LaTeX Project Public License 1.3c, see
+%%% ftp://ctan.tug.org/macros/latex/base/lppl.txt
+%%% ",
+%%% address = "Arthur Ogawa,
+%%% USA",
+%%% telephone = "",
+%%% FAX = "",
+%%% email = "mailto colon arthur_ogawa at sbcglobal.net",
+%%% codetable = "ISO/ASCII",
+%%% keywords = "latex, page grid, main vertical list",
+%%% supported = "yes",
+%%% abstract = "package to change page grid, MVL",
+%%% }
+%</package>
+% \fi
+%
+% \iffalse ltxdoc klootch
+% The following references the \file{README-LTXGRID} file,
+% which contains basic information about this package.
+% The contents of this file are generated when
+% you typeset the programmer's documentation.
+% Search on "{filecontents*}{README-LTXGRID}" to locate it.
+% \fi\input{README-LTXGRID}%
+%
+% \subsection{Bill of Materials}
+%
+% Following is a list of the files in this distribution arranged
+% according to provenance.
+%
+% \subsubsection{Primary Source}%
+% One single file generates all.
+%\begin{verbatim}
+%ltxgrid.dtx
+%\end{verbatim}
+%
+% \subsubsection{Generated by \texttt{latex ltxgrid.dtx}}%
+% Typesetting the source file under \LaTeX\
+% generates the readme and the installer.
+%\begin{verbatim}
+%README-LTXGRID ltxgrid.ins
+%\end{verbatim}
+%
+% \subsubsection{Generated by \texttt{tex ltxgrid.ins}}%
+% Typesetting the installer generates
+% the package files.
+%\begin{verbatim}
+%ltxgrid.sty
+%\end{verbatim}
+%
+% \subsubsection{Documentation}%
+% The following are the online documentation:
+% \begin{verbatim}
+%ltxgrid.pdf
+% \end{verbatim}
+%
+% \subsubsection{Auxiliary}%
+% The following are auxiliary files generated
+% in the course of running \LaTeX:
+% \begin{verbatim}
+%ltxgrid.aux ltxgrid.idx ltxgrid.ind ltxgrid.log ltxgrid.toc
+% \end{verbatim}
+%
+% \section{Code common to all modules}%
+%
+% The following may look a bit klootchy, but we
+% want to require only one place in this file
+% where the version number is stated,
+% and we also want to ensure that the version
+% number is embedded into every generated file.
+%
+% Now we declare that
+% these files can only be used with \LaTeXe.
+% An appropriate message is displayed if
+% a different \TeX{} format is used.
+% \begin{macrocode}
+%<*driver|package>
+\NeedsTeXFormat{LaTeX2e}[1995/12/01]%
+%</driver|package>
+% \end{macrocode}
+% As desired, the following modules all
+% take common version information:
+% \begin{macrocode}
+%<kernel&!package&!doc>\typeout{%
+%<*package|doc>
+\ProvidesFile{%
+%</package|doc>
+%<*kernel|package|doc>
+ltxgrid%
+%</kernel|package|doc>
+%<*doc>
+.dtx%
+%</doc>
+%<package>.sty%
+%<*package|doc>
+}%
+%</package|doc>
+% \end{macrocode}
+%
+% The following line contains, for once and for all,
+% the version and date information.
+% By various means, this information is reproduced
+% consistently in all generated files and in the
+% typeset documentation.
+% Give credit where due.
+% \begin{macrocode}
+%<*doc|package|kernel>
+%<version>
+ [2020/10/03 4.2e page grid package (portions licensed from W. E. Baxter web at superscript.com)]% \fileversion
+%</doc|package|kernel>
+%<kernel&!package&!doc>}%
+% \end{macrocode}
+%
+%
+% \section{The driver module \texttt{doc}}
+%
+% This module, consisting of the present section,
+% typesets the programmer's documentation,
+% generating the \file{.ins} installer and \file{README-LTXGRID} as required.
+%
+% Because the only uncommented-out lines of code at the beginning of
+% this file constitute the \file{doc} module itself,
+% we can simply typeset the \file{.dtx} file directly,
+% and there is thus rarely any need to
+% generate the ``doc'' {\sc docstrip} module.
+% Module delimiters are nonetheless required so that
+% this code does not find its way into the other modules.
+%
+% The \enve{document} command concludes the typesetting run.
+%
+% \begin{macrocode}
+%<*driver>
+% \end{macrocode}
+%
+% \subsection{The Preamble}
+% The programmers documentation is formatted
+% with the \classname{ltxdoc} class with local customizations,
+% and with the usual code line indexing.
+% \begin{macrocode}
+\documentclass{ltxdoc}
+\RequirePackage{ltxdocext}%
+\RequirePackage[colorlinks=true,linkcolor=blue]{hyperref}%
+%\ifx\package@font\@undefined\else
+% \expandafter\expandafter
+% \expandafter\RequirePackage
+% \expandafter\expandafter
+% \expandafter{%
+% \csname package@font\endcsname
+% }%
+%\fi
+\CodelineIndex\EnableCrossrefs % makeindex -s gind.ist ltxgrid
+\RecordChanges % makeindex -s gglo.ist -o ltxgrid.gls ltxgrid.glo
+% \end{macrocode}
+%
+% \subsubsection{Docstrip and info directives}
+% We use so many {\sc docstrip} modules that we set the
+% \texttt{StandardModuleDepth} counter to 1.
+% \begin{macrocode}
+\setcounter{StandardModuleDepth}{1}
+% \end{macrocode}
+% The following command retrieves the date and version information
+% from this file.
+% \begin{macrocode}
+\expandafter\GetFileInfo\expandafter{\jobname.dtx}%
+% \end{macrocode}
+%
+% \subsection{The ``Read Me'' File}
+% As promised above, here is the contents of the
+% ``Read Me'' file. That file serves a double purpose,
+% since it also constitutes the beginining of the
+% programmer's documentation. What better thing, after
+% all, to have appear at the beginning of the
+% typeset documentation?
+%
+% A good discussion of how to write a ReadMe file can be found in
+% Engst, Tonya, ``Writing a ReadMe File? Read This''
+% \emph{MacTech} October 1998, p. 58.
+%
+% Note the appearance of the
+% \cmd\StopEventually\ command, which marks the
+% dividing line between the user documentation
+% and the programmer documentation.
+%
+% The usual user will not be asked to
+% do a full build, not to speak
+% of the bootstrap.
+% Instructions for carrying these processes
+% begin the programmer's manual.
+%
+% \begin{macrocode}
+\begin{filecontents*}{README-LTXGRID}
+\title{%
+ A \LaTeX\ Package for changing the page grid and MVL%
+ \thanks{%
+ This file has version number \fileversion,
+ last revised \filedate.%
+ }%
+ \thanks{%
+ Version \fileversion\ \copyright\ 2019--2020 American Physical Society
+ }%
+}%
+\author{%
+ Arthur Ogawa%
+ \thanks{\texttt{mailto:arthur\_ogawa at sbcglobal.net}}%
+}%
+%\iffalse
+% For version number and date,
+% search on "\fileversion" in the .dtx file,
+% or see the end of the README-LTXGRID file.
+%\fi
+\maketitle
+
+This file embodies the \classname{ltxgrid} package,
+the implementation and its user documentation.
+
+The distribution point for this work is
+\url{journals.aps.org/revtex},
+which contains the REV\TeX\ package, and includes source and documentation for this package.
+
+The \classname{ltxgrid} package was commissioned by the American Physical Society
+and is distributed under the terms of the \LaTeX\ Project Public License 1.3c,
+the same license under which all the portions of \LaTeX\ itself is distributed.
+Please see \url{http://ctan.tug.org/macros/latex/base/lppl.txt} for details.
+
+To use this document class, you must have a working
+\TeX\ installation equipped with \LaTeXe\
+and possibly pdftex and Adobe Acrobat Reader or equivalent.
+
+To install, retrieve the distribution,
+unpack it into a directory on the target computer,
+into a location in your filesystem where it will be found by \LaTeX;
+in a TDS-compliant installation this would be:
+\file{texmf/tex/macros/latex/revtex/.}
+
+To use, read the user documentation \file{src/ltxgrid.pdf}.
+
+\tableofcontents
+
+\section{Processing Instructions}
+
+The package file \file{ltxgrid.sty}
+is generated from this file, \file{ltxgrid.dtx},
+using the {\sc docstrip} facility of \LaTeX
+via |tex ltxgrid.dtx|.
+The typeset documentation that you are now reading is generated from
+this same file by typesetting it with \LaTeX\ or pdftex
+via |latex ltxgrid.dtx| or |pdflatex ltxgrid.dtx|.
+
+\subsection{Build Instructions}
+
+You may bootstrap this suite of files solely from \file{ltxgrid.dtx}.
+Prepare by installing \LaTeXe\ (and either tex or pdftex) on your computer,
+then carry out the following steps:
+\begin{enumerate}
+\item
+Within an otherwise empty directory,
+typeset \file{ltxgrid.dtx} with \TeX\ or pdftex;
+thereby generating the package file \file{ltxgrid.sty}.
+
+\item
+Now typeset \file{ltxgrid.dtx} with \LaTeX\ or pdflatex;
+you will obtain the typeset documentation you are now reading,
+along with the file \file{README-LTXGRID}.
+
+Note: you will have to run \LaTeX\ twice, then \file{makeindex}, then
+\LaTeX\ again in order to obtain a valid index and table of contents.
+
+\item
+Install the following files into indicated locations within your
+TDS-compliant \texttt{texmf} tree (you may need root access):
+\begin{itemize}
+\item
+\file{$TEXMF/}\file{tex/}\file{latex/}\file{revtex/}\classname{ltxgrid.sty}
+\item
+\file{$TEXMF/}\file{source/}\file{latex/}\file{revtex/}\classname{ltxgrid.dtx}
+\item
+\file{$TEXMF/}\file{doc/}\file{latex/}\file{revtex/}\classname{ltxgrid.pdf}
+\end{itemize}
+where \file{TEXMF/} stands for \file{texmf-local/}, or some other \texttt{texmf} tree in your installation.
+\item
+Run \texttt{mktexlsr} on directory \file{$TEXMF/} (you may need root access).
+\item
+Build and installation are now complete;
+now put a \cmd\usepackage\texttt{\{ltxgrid\}} in your document preamble!
+(Note: \texttt{ltxgrid} requires package \texttt{ltxutil}.)
+\end{enumerate}
+
+\subsection{Change Log}
+\changes{4.0a}{2001/06/18}{Introduce \cs{marry@height} }
+\changes{4.0a}{2001/06/18}{Introduce \cs{set@marry@height} }
+\changes{4.0a}{2008/06/26 }{\cs{@yfloat}: de-fang \cs{set@footnotewidth} (see ltxutil.dtx): we have already done its job. }
+\changes{4.1a}{2008/06/29}{Change \cs{LT@array@new}: restore \cs{@tabularcr} and \cs{@xtabularcr}}
+\changes{4.1a}{2008/06/29}{Change \cs{LT@array@new}: set \cs{LT@LL@FM@cr} to \cs{@arraycr@array} instead of \cs{@arraycr}}
+\changes{4.1a}{2008/06/29}{Repair error in \cs{endlongtable@new} involving \cs{@ifx}: argument not delimited.}
+\changes{4.1b}{2008/08/04}{Get rid of the \cs{reserved@a} idiom}
+\changes{4.1b}{2008/08/04}{Turn off the \cs{set@footnotewidth} mechanism; a float `knows' its proper typesetting context}
+\changes{4.1b}{2008/08/04}{(AO, 452) Support length checking: show size of shipped out text.}
+\changes{4.1b}{2008/08/04}{(AO, 456) Compatibility with other packages that override the output routine, following suggestion by David Kastrup.}
+\changes{4.1b}{2008/08/04}{}
+\changes{4.1b}{2008/08/04}{Box \cs{footbox} changed to box \cs{footsofar}}
+\changes{4.1b}{2008/08/04}{Change \cs{@combinepage} to \cs{@combinepage} with argument}
+\changes{4.1b}{2008/08/04}{Change \cs{@makecol} to \cs{@makecolumn} with argument}
+\changes{4.1b}{2008/08/04}{Change \cs{set@colroom} to \cs{set@colht}}
+\changes{4.1b}{2008/08/04}{New procedure \cs{@iffpsbit} replaces \cs{@getfpsbit}}
+\changes{4.1b}{2008/08/04}{New procedure \cs{@output@combined@page}}
+\changes{4.1b}{2008/08/04}{New procedure for showing a box contents, \cs{trace@box}}
+\changes{4.1b}{2008/08/04}{Procedure \cs{@outputpage@head} headpatches \cs{@outputpage}}%
+\changes{4.1b}{2008/08/04}{Procedure \cs{@outputpage@tail} tailpatches \cs{@outputpage}}%
+\changes{4.1b}{2008/08/04}{Procedure \cs{balance@2} defined more transparently}%
+\changes{4.1b}{2008/08/04}{Tally the height of the float}
+\changes{4.1b}{2008/08/04}{Use \cs{document@inithook} instead of \cs{AtBeginDocument}}
+\changes{4.1b}{2008/08/04}{Use \cs{trace@box} instead of \cs{showbox}}
+\changes{4.1f}{2009/07/07}{(AO, 515) Prevent line numbering within a footnote}
+\changes{4.1f}{2009/07/10}{(AO, 518) Tally register overflow when locument is long}
+\changes{4.1f}{2009/07/14}{(AO, 519) \cs{footins} content must be preserved and reintegrated}
+\changes{4.1f}{2009/07/15}{(AO, 519) Preserve footnotes that are in \cs{footsofar} across a page grid change}
+\changes{4.1g}{2009/10/06}{(AO, 531) Fix package \classname{float} }
+\changes{4.1n}{2009/12/02}{Restore the \cs{lastbox} if it is not a footnote}
+\changes{4.1n}{2009/12/02}{More diagnostics of column balancing}
+\changes{4.1n}{2009/12/18}{(AO, 571) Deconstruct balanced footnotes when needed}
+\changes{4.1n}{2010/01/02}{(AO, 571) Interface \cs{set@footnotewidth} for determining the set width of footnotes}%
+\changes{4.1n}{2010/01/02}{(AO, 571) Footnotes, when columns are balanced or when they are composed with their column}%
+\changes{4.1n}{2010/01/02}{(AO, 571) Abandon \cs{recover@footins} in favor of \cs{recover@column}}%
+\changes{4.1n}{2010/01/02}{(AO, 571) Use procedures \cs{output@do@prep} and \cs{output@column@do} as dispatchers}
+\changes{4.1n}{2010/01/02}{(AO, 571) coding convention: use \cs{bgroup}, \cs{egroup} (instead of braces) when a box is being built}%
+\changes{4.1n}{2010/01/02}{(AO, 571) calling sequence of \cs{combine@foot@inserts} and \cs{grid@column} to expose box registers; that of \cs{append@column@} its column counters}%
+\changes{4.1n}{2010/01/02}{(AO, 571) footnote rule is leaders, so that it may be removed by \cs{vsplit}; mechanism of \cs{kern} signals to indicate footnote height}%
+\changes{4.1o}{2010/02/02}{(AO, 576) Allow \classname{lscape} to act on \cs{@outputbox} at the right time}%
+\changes{4.1p}{2010/02/24}{(AO, 583) Provide setup code also for footnotes in a one-column document}%
+\changes{4.2a}{2018/12/12}{(MD) Updated name of README file and use standard fonts when typesetting}%
+\changes{4.2d}{2020/09/19}{(PHO) Adapt \cs{document} and \cs{enddocument} hooks to the 2020-10-01 \LaTeX{} release.}%
+\changes{4.2d}{2020/09/30}{(PHO) Correct \cs{@normalcr} patch for \LaTeX{} 2020-02-02 and later.}%
+\end{filecontents*}
+% \end{macrocode}
+%
+% \subsection{The Document Body}
+%
+% Here is the document body, containing only a
+% \cmd\DocInput\ directive---referring to this very file.
+% This very cute self-reference is a common \classname{ltxdoc} idiom.
+% \begin{macrocode}
+\begin{document}%
+\def\revtex{REV\TeX}%
+\expandafter\DocInput\expandafter{\jobname.dtx}%
+\end{document}
+%</driver>
+% \end{macrocode}
+%
+% \section{Using this package}
+% Once this package is installed on your filesystem, you can employ it in
+% adding functionality to \LaTeX\ by invoking it in your document or document class.
+%
+% \subsection{Invoking the package}
+% In your document, you can simply call it up in your preamble:
+% \begin{verbatim}
+%\documentclass{book}%
+%\usepackage{ltxgrid}%
+%\begin{document}
+%<your document here>
+%\end{document}
+% \end{verbatim}
+% However, the preferred way is to invoke this package from within your
+% customized document class:
+% \begin{verbatim}
+%\NeedsTeXFormat{LaTeX2e}[1995/12/01]%
+%\ProvidesClass{myclass}%
+%\LoadClass{book}%
+%\RequirePackage{ltxgrid}%
+%<class customization commands>
+%\endinput
+% \end{verbatim}
+%
+% Note that this package requires the features of the \classname{ltxutil} package,
+% available at
+% \url{publish.aps.org/revtex}.
+%
+% Once loaded, the package gives you acccess to certain procedures,
+% usually to be invoked by a \LaTeX\ command or environment, but not at the document level.
+%
+% \subsection{Changing the page grid}%
+% This package provides two procedures, \cmd\onecolumngrid, \cmd\twocolumngrid,
+% that change the page grid (it can be extended to more columns and to other page grids).
+%
+% They differ from standard \LaTeX's \cmd\onecolumn\ and \cmd\twocolumn\ commands in that
+% they do not force a page break. Also, upon leaving a multiple-column grid, the columns are balanced.
+% In other respects they work same.
+%
+% They differ from the grid-changing commands of Frank Mittelbach's \classname{multicol} package
+% in that they allow floats of all types (single- and double column floats, that is) and
+% preserve compatability with the \classname{longtable} package.
+%
+% These commands must be issued in vertical mode (conceivably via a \cmd\vadjust) such that
+% they are ultimately present in the MVL, where they can do their work.
+% Because they do not work in \LaTeX's left-right mode, they are unsuitable at the
+% document level.
+% Furthermore, packaging a grid command in a \cmd\vadjust, although possible, will probably
+% not acheive satisfactory page layout.
+%
+% Page grid commands are not intended to be issued unnecessarily: only the first of
+% two successive \cmd\onecolumngrid\ commands is effective; the second will be silently ignored.
+%
+% \DescribeMacro\onecolumngrid
+% You command \LaTeX\ to return to the one-column grid with the
+% \cmd\onecolumngrid\ command. If you are already in the one-column grid, this
+% is a no-op. The one-column grid is considered special of all page grids, in that
+% no portion of the page is held back (in \cmd\pagesofar); all items that might go
+% on the current page (with the exception of floats and footnotes) are on the MVL.
+%
+% \DescribeMacro\twocolumngrid
+% You command \LaTeX\ to return to the two-column grid with the
+% \cmd\twocolumngrid\ command. If you are already in the two-column grid, this
+% is a no-op.
+%
+% These two commands should be issued by a macro procedure that can ensure that
+% \TeX\ is in outer vertical mode.
+%
+% \subsection{Changing the MVL}%
+%
+% This package also provides commands to modify the main vertical list (MVL) in a safe way.
+% The scheme here is to structure, insofar possible, \TeX's MVL as follows:
+%\begin{quotation}
+%box or boxes\\
+%penalty\\
+%glue
+%\end{quotation}
+% This should be a familiar sequence. It is the prototype sequence for a vertical list,
+% and is followed when \TeX\ breaks paragraphs into lines, and when \TeX\ generates
+% a display math equation.
+%
+% If you (as a macro programmer) wish to modify the value of the penalty or glue item,
+% you can use one of the MVL-altering commands to do so. Certain operations are implemented
+% here; you can make up your own.
+%
+% Note that these commands must be issued in vertical mode, perhaps via a \cmd\vadjust\ or a \cmd\noalign.
+% They can work directly if you are in inner mode (say within a parbox or a minipage).
+%
+% \DescribeMacro\removestuff
+% You instruct \LaTeX\ to remove both the penalty and the glue item with this command.
+%
+% \DescribeMacro\addstuff
+% You issue the \cmd\addstuff\arg{penalty}\arg{glue} command to add a penalty, glue, or both.
+% If you do not wish to add one or the other, the corresponding argument should be nil.
+% Note that the effect of \cmd\addstuff\ is to stack the penalties and glue items.
+% Therefore, the lesser of the two penalties takes effect,
+% and the two glue items add together.
+%
+% \cmd\addstuff\ is limited because once applied, it cannot be applied again with correct results.
+%
+% \DescribeMacro\replacestuff
+% The \cmd\replacestuff\ command is syntactically the same as \cmd\addstuff, but works
+% differently: the existing penalty and glue are replaced or modified.
+%
+% The specified penalty is not inserted if the existing penalty is greater than 10000
+% (that is, in case of a \cmd\nobreak), otherwise, the lower (non-zero) of the two penalties is inserted.
+%
+% If the specified glue has a larger natural component than the existing glue, we replace the glue.
+% However, if the specified glue's natural component is negative, then the existing glue's natural component is
+% changed by that amount.
+%
+% \cmd\replacestuff\ can be applied mutiple times bceause it retains the list structure in the canonical form.
+%
+% Note that we treat two penalties specially (as does \TeX): a penalty of 10000 is considered
+% a garbage value, to be replaced if found. This is the signal value that \TeX\ inserts on
+% the MVL replacing the penalty that caused the page break (if the page break occurred at a penalty).
+% Also, a penalty of zero is indistinguishable from no penalty at all, so it will always
+% be replaced by the given value.
+%
+% Therefore, it is highly recommended to never set any of \TeX's penalty parameters to
+% zero (a value of, say, 1, is practically the same), nor should a skip parameter be set to zero
+% (instead, use, say, 1sp). Also, to prevent a pagebreak, do not use a penalty of 10000, use, say
+% 10001 instead.
+%
+% You can define your own construct that modifies the MVL:
+% Define a command, say, \cmd\myadjust, as follows:
+%\begin{verbatim}
+%\def\myadjust#1{\noexpand\do@main@vlist{\noexpand\@myadjust{#1}}\@tempa}%
+%\end{verbatim}
+% that is, \cmd\myadjust\ invokes \cmd\do@main@vlist, passing it the procedure name
+% \cmd\@myadjust\ along with the arguments thereof pre-expanded.
+% Next, define the procedure \cmd\@myadjust:
+%\begin{verbatim}
+%\def\@myadjust#1{<meddle with the MVL>}%
+%\end{verbatim}
+% when \cmd\@myadjust\ executes, you will be in the output routine (in inner vertical mode)
+% and the MVL will be that very vertical list.
+%
+%
+%
+%
+% \section{Compatability with \LaTeX's Required Packages}
+% Certain packages, usually ones written by members of the
+% \LaTeX\ Project itself, have been designated ``required'' and
+% are distributed as part of standard \LaTeX.
+% These packages have been placed in a priviledged position
+% vis \'a vis the \LaTeX\ kernel in that they override the definitions of certain kernel macros.
+%
+% Compatability between \classname{ltxgrid} and these packages is complicated
+% by a number of factors. First is that \classname{ltxgrid} alters the meaning of some of the same
+% kernel macros as certain of the ``required'' packages.
+% Second is that fact that certain of the ``required'' packages of \LaTeX\ are incompatible with
+% each other.
+%
+% Examples of the first kind are the \classname{ftnright}, \classname{multicol}, and \classname{longtable}
+% packages.
+% The \classname{ltxgrid} package is not compatible with \classname{multicol},
+% but if you are using \classname{ltxgrid}, you do not need to use \classname{ftnright} or \classname{multicol}
+% anyway. The \classname{ltxgrid} package does however attempt to be compatible with \classname{longtable}.
+%
+% Among the ``required'' packages that are mutually incompatible are \classname{multicol} and \classname{longtable},
+% the incompatibility arising because both packages replace \LaTeX's output routine:
+% if one package is active, the other must not be so.
+% This state of affairs has remained essentially unchanged since the introduction of the two as \LaTeX2.09 packages in the late 1980s.
+%
+% The reason that \classname{ltxgrid} can remain compatible with \classname{longtable} is due to the
+% introduction of a more modern architecture, the ``output routine dispatcher'', which allows all macro packages access to the
+% safe processing environment of the output routine, on an equal footing.
+% The relevant portions of the \classname{longtable} package are reimplemented in \classname{ltxgrid}
+% to take advantage of this mechanism.
+%
+% Timing is critical:
+% the \classname{ltxgrid} package will be incompatible with any package that
+% redefines any of the kernel macros that \classname{ltxgrid} patches---if that
+% package is loaded \emph{after} \classname{ltxgrid}.
+%
+% Hereinafter follows some notes on specific \LaTeX\ packages.
+%
+% \subsection{ftnright}
+% Frank Mittelbach's \classname{ftnright} package effects a change to \LaTeX's
+% \cmd\twocolumn\ mode such that footnotes are set at the bottom of the right-hand
+% column instead of at the foot of each of the two columns.
+%
+% Note that it overwrites three \LaTeX\ kernel macros: \cmd\@outputdblcol, \cmd\@startcolumn, and \cmd\@makecolumn.
+% Fortunately none of the three are patched by \classname{ltxgrid}, so that compatability
+% is not excluded on this basis.
+%
+% At the same time, it changes the meaning of \cmd\footnotesize, the macro that is automatically
+% invoked when setting a document's footnote into type.
+% One might well argue that it is an error for the meaning of \cmd\footnotesize\ to be determined by
+% a package such as \classname{ftnright}, that indeed such a choice should be made in the
+% document class, or in a file such as \file{bk10.clo}.
+%
+% To avoid being tripped up by this misfeature in \classname{ftnright}, it is only necessary to
+% reassert our meaning for \cmd\footnotesize\ later on, after \classname{ftnright} has been loaded.
+%
+% Note that \classname{ftnright} inserts code that demands that \LaTeX's flag \cmd\if@twocolumn\
+% is true, that is, it will complain if deployed in a \cmd\onecolumn\ document.
+% It is therefore necessary for any other multicolumn package to assert that flag in order to
+% avoid this package's complaint. It is an interesting question exactly why this package has
+% this limitation. After all, a one-column page grid is just a degenerate case of the
+% two column.
+%
+% \subsection{longtable}
+% David Carlisle's \classname{longtable} package sets tables that can be so long as to break over pages.
+% According to its author, it uses the same override of \LaTeX's output routine as
+% Frank Mittelbach's \classname{multicol} package. By implication, then, it has a hard
+% incompatability with the latter.
+%
+% The \classname{longtable} package also performs a check of whether the document is in
+% \cmd\twocolumn\ mode, and declines to work if this is the case. It is not clear, however,
+% that there is any true incompatability present if so. It's just that David did not see any reason
+% anyone would want to set such long tables in a multicolumn document, hence the check.
+%
+% There does not appear to be any indication that \classname{longtable} would work less
+% well under \classname{ltxgrid} than under standard \LaTeX's \cmd\twocolumn\ mode.
+% Therefore, this \classname{ltxgrid} patches \classname{longtable} (if loaded) so as to provide
+% compatability. In the course of which, \classname{longtable} becomes more robust
+% (\classname{longtable} has mumerous bugs and incompatabilities of long standing,
+% some of which are repaired by \classname{ltxgrid}).
+%
+% One problem remains, namely that, if a \env{longtable} environment breaks over columns
+% and thereby inserts its special headers and footers at that break, and those columms are then
+% balanced (due to a return to the one-column page grid), then those inserted rows
+% will remain, and may no longer fall at the column break. This will, of course look
+% wrong.
+%
+% The only way to fix this problem is to avoid doing column balancing in the way
+% I have implemented here; such an enhancement to this package is possible.
+%
+% \subsection{multicol}
+% Frank Mittelbach's \classname{multicol} package provides a page grid with many columns,
+% albeit denies the placement of floats in individual columns.
+%
+% It establishes its own \cmd\output\ routine, which is the reason it runs afoul of
+% the \classname{longtable} package. On the other hand, \classname{ltxgrid} specifically
+% allows for the case where a package installs its own \cmd\output\ routine, so
+% there is no incompatability on that basis.
+%
+% Still, it is pointless to use \classname{multicol} if you are using \classname{ltxgrid},
+% since both packages provide multicolumn page layouts.
+% Therefore, \classname{multicol} is not supported by \classname{ltxgrid}.
+%
+% \subsection{ltxgrid}
+% It has been pointed out that one of the disadvantages of adopting the \classname{ltxgrid} package is that
+% it does alter the \LaTeX\ kernel.
+% Any package that itself alters the \LaTeX\ kernel may be incompatible with \classname{ltxgrid}, and new packages
+% (destined perhaps to become part of the successor to \LaTeXe) may break \classname{ltxgrid}.
+%
+% The consequence is that packages introduced in future, and future changes to \LaTeX\ may be incompatible
+% with \classname{ltxgrid}.
+% This is, of course, true.
+% The development plan for \classname{ltxgrid} is that when such packages and \LaTeX\ kernel changes come about,
+% the burden will be on \classname{ltxgrid} to change in a way that provides for continued compatability with
+% those packages and \LaTeX\ kernel changes.
+%
+%
+% \section{How \classname{ltxgrid} places footnotes}
+%
+% In conventional multicolumn layouts, a footnote will appear at the bottom of the column in which it is called out.
+% The \classname{ltxgrid} package implements this conventional layout choice by default.
+% However, other choices are possible (a la \classname{ftnright}, whose compatability with \classname{ltxgrid} has not been tested).
+%
+% One unusual feature of \classname{ltxgrid}'s default implementation must be mentioned, though,
+% namely the case in a two-column page grid, where a footnote is followed by a temporary change to the one-column page grid
+% (e.g., for a wide equation).
+% In such a case, the material above the wide material is split into two columns, and a footnote whose callout
+% appears in the right-hand column will nonetheless be set at the base of the left column.
+%
+% This arrangement was chosen because it ensures that the footnotes at the bottom of any page will appear in
+% numerical order. It can be argued that this choice is ``incorrect'', but be that as it may,
+% the \classname{ltxgrid} package does not foreclose on other arrangements for the footnotes.
+% The package can be adapted to accomodate any page design desired.
+%
+% \section{Limitations in \classname{ltxgrid}'s default column balancing method}%
+%
+% In a multicolumn page grid, when encountering a page that is not completely full,
+% it is customary to set the material in balanced columns (typically with the last column no longer than any of the others).
+% Such a case also crops up when temporarily interrupting the multicolumn grid to set material on the full width of the page:
+% the material on the page above the break is customarily set in balanced columns.
+%
+% An awkward case arises when we have already set one or more complete columns of type before encountering the need to
+% balance columns. In this subset of cases, the default in \classname{ltxgrid} is to
+% do an operation I call ``re-balancing'':
+% the material on the page so far is pasted back together into a single column, and new, balanced column breaks are
+% calculated.
+%
+% This scheme typically works fine, but it has a significant vulnerability:
+% any discardable items trimmed at the original column break are lost, never to be retrieved.
+% Consequently, after re-balancing, an element like, say, a section head can fail to have the correct amount of whitespace above.
+%
+% This problem is due to an unfortunate optimization in \TeX, wherein a certain class of nodes is trimmed from the
+% top of main vertical list upon returning from the output routine:
+% any penalty, glue, or leader node falls in to this class of discardable nodes,
+% and trimming proceeds until a non-discardable node (such as a box, or rule) is encountered.
+% It gets better: a third class of nodes is transparent to this trimming process;
+% they are neither discarded nor do they halt the process of trimming:
+% mark nodes and all whatsits fall into this class of transparent nodes;
+% they are quietly passed over during trimming.
+%
+% An alternative approach for \TeX\ to take would have been,
+% rather than discarding the node entirely, to simply \emph{mark} it as discarded.
+% (Implementors of extended \TeX, please note!)
+% Then, upon shipping out, such nodes would not make it into the DVI.
+% \TeX's optimization, driven by the small computer architectures current when it was developed,
+% does save mem, but at the cost of revisiting page breaks in a reliable way.
+%
+% FIXME: how to fix a column break in the above case? Widetext?
+%
+%\StopEventually{}
+%
+% \section{Implementation of package}
+%
+% Special acknowledgment: this package uses concepts pioneered
+% and first realized by William Baxter (mailto:web at superscript.com)
+% in his SuperScript line of commercial typesetting tools, and
+% which are used here with his permission. His thorough understanding
+% of \TeX's output routine underpins the entire \classname{ltxgrid}
+% package.
+%
+% \subsection{Beginning of the \file{ltxgrid} {\sc docstrip} module}
+% Requires the underpinnings of the \classname{ltxkrnext} package.
+% \begin{macrocode}
+%<*package>
+\def\package@name{ltxgrid}%
+\expandafter\PackageInfo\expandafter{\package@name}{%
+ Page grid for \protect\LaTeXe,
+ by A. Ogawa (arthur_ogawa at sbcglobal.net)%
+}%
+\RequirePackage{ltxutil}%
+%</package>
+% \end{macrocode}
+%
+% \subsection{Banner}%
+% \begin{macrocode}
+%<*kernel>
+% \end{macrocode}
+%
+% \subsection{Sundry}%
+% Here are assorted macro definitions.
+% \begin{macro}{\lineloop}
+% \begin{macro}{\linefoot}
+% The (document-level) command \cmd\lineloop\ sets numbered lines until the
+% specified count is reached.
+% The command \cmd\linefoot\ sets a single, automatically numbered line, but with a footnote (with the specified label);
+% it automatically increments the line counter.
+% These commands are typically used to construct test documents.
+%
+% Because the counter is globally advanced and never reset, successive
+% calls to \cmd\lineloop\ should have an argument ever larger.
+% The formatted output will have each line labeled with its ordinal number.
+% \begin{macrocode}
+\newcounter{linecount}
+\def\loop@line#1#2{%
+ \par
+ \hb@xt@\hsize{%
+ \global\advance#1\@ne
+ \edef\@tempa{\@ifnum{100>#1}{0}{}\@ifnum{10>#1}{0}{}\number#1}%
+ \@tempa\edef\@tempa{\special{line:\@tempa}}\@tempa
+ \vrule depth2.5\p@#2\leaders\hrule\hfil
+ }%
+}%
+\def\lineloop#1{%
+ \loopwhile{\loop@line\c@linecount{}\@ifnum{#1>\c@linecount}}%
+}%
+\def\linefoot#1{%
+ \loop@line\c@linecount{%
+ \footnote{%
+ #1\special{foot:#1}\vrule depth2.5\p@\leaders\hrule\hfill
+ }%
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+%
+% \subsection{Mark Components}%
+%
+% Override LaTeX's mark macros to allow more components.
+%
+% We remain bound by the weakness of LaTeX's scheme in that
+% one cannot emulate the action of \TeX\ whereby
+% material with marks can be inserted in the middle of
+% a vertical list such that the marks are reliably calculated.
+% If we did that, \cmd\@themark\ would no longer be utilized.
+%
+% A more robust scheme involves placing all marks (component and value)
+% into a list (using global scoping, i.e., \cmd\gdef),
+% and using \cmd\@@mark to place an index on that list into the MVL.
+% Then, e.g., \cmd\@@botmark signifies the place where that list is to be cut,
+% and the \cmd\botmark\ of any component is
+% the value of the last element of the cut
+% list having the given component. The \cmd\firstmark\ and \cmd\topmark\
+% can likewise be defined relative to \cmd\@@firstmark\ and \cmd\@@topmark,
+% except in the latter case, we want the first following the cut instead of the last
+% preceding the cut.
+%
+% The limitation of this scheme is its demands upon \TeX's mem.
+% The list of marks would need to be trimmed back to, effectively,
+% \cmd\topmark\ at the beginning of every page.
+%
+% This approach is not yet part of the extended LaTeX kernel.
+%
+% \begin{macro}{\@@mark}
+% \begin{macro}{\@@topmark}
+% \begin{macro}{\@@firstmark}
+% \begin{macro}{\@@botmark}
+% \begin{macro}{\@@splitfirstmark}
+% \begin{macro}{\@@splitbotmark}
+% Remember primitives under a new set of names.
+% \begin{macrocode}
+\let\@@mark\mark
+\let\@@topmark\topmark
+\let\@@firstmark\firstmark
+\let\@@botmark\botmark
+\let\@@splitfirstmark\splitfirstmark
+\let\@@splitbotmark\splitbotmark
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \subsubsection{Procedures that expose the component data structure}%
+% This portion of the code exposes the internal representation of
+% the mark components. If we wish to add more components, we will have to revise
+% these macro definitions:
+% \cmd\@themark,
+% \cmd\nul@mark,
+% \cmd\set@mark@netw@,
+% \cmd\set@marktw@,
+% \cmd\set@markthr@@,
+% \cmd\get@mark@@ne,
+% \cmd\get@mark@tw@,
+% \cmd\get@mark@thr@@,
+% \cmd\get@mark@f@ur.
+%
+% \begin{macro}{\@themark}
+%FIXME: is it safer to eliminate \cmd\@themark\ in favor of a message that evaluates \cmd\@@botmark?
+%
+% Note: these definitions expose the data structure of mark components.
+% \begin{macrocode}
+\def\@themark{{}{}{}{}}%
+\def\nul@mark{{}{}{}{}\@@nul}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\set@mark@netw@}
+% \begin{macro}{\set@marktw@}
+% \begin{macro}{\set@markthr@@}
+% These procedures insert the new value of a particular mark component into the given argument.
+% They expose the data structure of mark components.
+%
+% \begin{macrocode}
+\def\set@mark@netw@#1#2#3#4#5#6#7{\gdef#1{{#6}{#7}{#4}{#5}}\do@mark}%
+\def\set@marktw@#1#2#3#4#5#6{\gdef#1{{#2}{#6}{#4}{#5}}\do@mark}%
+\def\set@markthr@@#1#2#3#4#5#6{\gdef#1{{#2}{#3}{#6}{#5}}\do@mark}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\get@mark@@ne}
+% \begin{macro}{\get@mark@tw@}
+% \begin{macro}{\get@mark@thr@@}
+% \begin{macro}{\get@mark@f@ur}
+% These procedures retreive the value of a particular mark component.
+% They expose the data structure of mark components.
+% \begin{macrocode}
+\def\get@mark@@ne#1#2#3#4#5\@@nul{#1}%
+\def\get@mark@tw@#1#2#3#4#5\@@nul{#2}%
+\def\get@mark@thr@@#1#2#3#4#5\@@nul{#3}%
+\def\get@mark@f@ur#1#2#3#4#5\@@nul{#4}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+%
+% \subsubsection{Procedures that do not expose the component data structure}%
+%
+% \begin{macro}{\mark@netw@}
+% \begin{macro}{\marktw@}
+% \begin{macro}{\markthr@@}
+% These procedures insert the new value of a particular mark component into \cmd\@themark,
+% then execute \cmd\do@mark.
+% They constitute the implementation layer for mark components one, two, and three.
+% An analogous procedure for component four could be defined; call it \cmd\markf@ur.
+%
+% \begin{macrocode}
+\def\mark@netw@{\expandafter\set@mark@netw@\expandafter\@themark\@themark}%
+\def\marktw@{\expandafter\set@marktw@\expandafter\@themark\@themark}%
+\def\markthr@@{\expandafter\set@markthr@@\expandafter\@themark\@themark}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\do@mark}
+% \begin{macro}{\do@@mark}
+% Access procedures \cmd\mark (AKA \cmd\@@mark).
+% The \cmd\do@mark\ procedure is used when a mark is being put down into the MVL;
+% \cmd\do@@mark\ when this happens in the output routine.
+% \begin{macrocode}
+\def\do@mark{\do@@mark\@themark\nobreak@mark}%
+\def\do@@mark#1{%
+ \begingroup
+ \let@mark
+ \@@mark{#1}%
+ \endgroup
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\let@mark}
+% \begin{macro}{\nobreak@mark}%
+% The procedure that makes \cmd\csname s robust within a mark.
+% Use \cmd\appdef\ and \cmd\robust@\ to extend the list.
+% \begin{macrocode}
+\def\let@mark{%
+ \let\protect\@unexpandable@protect
+ \let\label\relax
+ \let\index\relax
+ \let\glossary\relax
+}%
+\def\nobreak@mark{%
+ \@if@sw\if@nobreak\fi{\@ifvmode{\nobreak}{}}{}%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+%
+% \subsubsection{Using mark components}%
+%
+% These procedures use the component mark mechanism to implement
+% a mark component that remembers the current environment (used in page makeup)
+% and the the two mark components left over from the original \LaTeX.
+% The fourth component is presently unused.
+%
+% \begin{macro}{\mark@envir}
+% The third mark component's access procedures.
+% The \cmd\mark@envir\ and \cmd\bot@envir\ commands are a good model of how to write
+% access procedures for a new mark component.
+% \begin{macrocode}
+\def\mark@envir{\markthr@@}%
+\def\bot@envir{%
+ \expandafter\expandafter
+ \expandafter\get@mark@thr@@
+ \expandafter\@@botmark
+ \nul@mark
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\markboth}
+% \begin{macro}{\markright}
+% \begin{macro}{\leftmark}
+% \begin{macro}{\rightmark}
+% Set procedures for legacy components.
+% \begin{macrocode}
+\def\markboth{\mark@netw@}%
+\def\markright{\marktw@}%
+% \end{macrocode}
+%
+% Retrieval procedures for legacy mark components.
+% The procedure for retrieving the first component from \cmd\botmark\
+% and the second component from \cmd\firstmark have names in \LaTeX;
+% they are called, respectively, \cmd\leftmark\ and \cmd\rightmark.
+%
+% It is possible to retrieve the components of \cmd\topmark\
+% as well: use \cmd\saved@@topmark.
+% \begin{macrocode}
+\def\leftmark{%
+ \expandafter\expandafter
+ \expandafter\get@mark@@ne
+ \expandafter\saved@@botmark
+ \nul@mark
+}%
+\def\rightmark{%
+ \expandafter\expandafter
+ \expandafter\get@mark@tw@
+ \expandafter\saved@@firstmark
+ \nul@mark
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+%
+%
+% \subsection{Output Super-routine}%
+%
+% We want to change \LaTeX's output routine, but do not wish to remain vulnerable
+% to interference from such ``required'' packages as
+% \classname{multicol} (authored by Frank Mittelbach)
+% and \classname{longtable} (authored by David P. Carlisle), which
+% swap in their own output routines when the respective package is active.
+%
+% The better mechanism, used here, is due to William Baxter (web at superscript.com),
+% who has allowed his several ideas to be used in this package.
+%
+% In what follows, we effectively wrap up the old \LaTeX\ output routine inside
+% a new, more flexible ``super routine''. When the output routine is called,
+% the ``super routine'' acts as a dispatcher. If the old routine is needed, it is called.
+%
+% If a package attempts to substitute in their own output routine, they will effectively
+% be modifying a token register by the name of \cmd\output.
+% The primitive \cmd\output\ is now known by a different name, which should no longer be
+% necessary to use.
+%
+% Usage note: to make a visit to the output routine employing the dispatcher, enter
+% with a value of \cmd\outputpenalty\ that corresponds to a macro. Defining as follows:
+%\begin{verbatim}
+%\@namedef{output@10000}{<your code here>}%
+%\end{verbatim}
+% by convention, your output routine should void out \cmd\box\cmd\@cclv.
+%
+% In rewriting \LaTeX's output dispatcher
+% in a much simpler form, we also avoid the sin of multiple \cmd\shipout s
+% within a single visit to the output routine.
+%
+% Conceptually, we divide visits to the output routine into two classes.
+% The first involves natural page breaks
+% (at a \cmd\newpage\ or when \cmd\pagetotal $>$ \cmd\pagegoal)
+% and usually resulting in \cmd\box\cmd\@cclv\ either being shipped out or
+% salted away (e.g., each column in a multicolumn layout).
+% We might call this class the ``natural output routines''; the \cmd\outputpenalty\
+% will never be less than $-10000$.
+% Furthermore, we ensure that \cmd\holdinginserts\ is cleared when
+% calling such routines.
+%
+% The other class involves a forced visit to the output routine
+% via a large negative penalty ($< -10000$). They do not generally
+% result in a \cmd\shipout\ of \cmd\box\cmd\@cclv: they may be dead cycles.
+% We provide a mechanism (call it a ``one-off'' output routine) that allows
+% us to specify certain processing to be done when \TeX\ reaches
+% the current position on the page.
+%
+% One-off output routines themselves fall into two divisions, ones
+% that process \cmd\box\cmd\@cclv, and ones that work on the main vertical list (MVL).
+% The former are typified by changes to the page grid, perhaps
+% even column balancing.
+% The latter involve the insertion of penalties or glue and the processing of floats.
+%
+% The natural output routine is a single procedure. We have not introduced multiple
+% natural output routines based on the \cmd\outputpenalty\ because \TeX\ does not
+% support such a thing: \TeX\ sometimes lays down a penalty whose value is the sum
+% of other penalties. Because of this, we cannot depend on the value of \cmd\outputpenalty\
+% in such areas.
+%
+% We do introduce flexibility in the form of a mechanism for patching into the
+% natural output routine. Three hooks are offered, allowing a procedure
+% to prepare for the upcoming visit to the output routine,
+% access to \cmd\box\cmd\@cclv, and after doing \cmd\shipout\ (or otherwise
+% committing the material to the page).
+%
+% Environments, commands, and even packages can install their
+% own procedures into these hooks.
+% For instance, if the longtable package is loaded, it will install
+% its procedures, but those procedures will punt if the page break
+% being processed does not actually fall within a longtable environment.
+%
+% \begin{macro}{\primitive@output}
+% Here we remember the \TeX\ primitive \cmd\output\ and its value,
+% and then proceed to take over the \cmd\csname\ of \cmd\output,
+% making it a \cmd\toks\ register and installing the old value of
+% the output routine.
+% \begin{macrocode}
+\let\primitive@output\output
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\output@latex}
+% \begin{macro}{\output}
+% Grab the tokens in \cmd\the\cmd\output\ (but without the extra set of braces).
+% The value of \cmd\toks@\ must remain untouched until loaded into the appropriate token
+% register; this is done a few lines below.
+% \begin{macrocode}
+\long\def\@tempa#1\@@nil{#1}%
+ \toks@
+\expandafter\expandafter
+\expandafter{%
+\expandafter \@tempa
+ \the\primitive@output
+ \@@nil
+ }%
+\newtoks\output@latex
+\output@latex\expandafter{\the\toks@}%
+\let\output\output@latex
+% \end{macrocode}
+%
+% A comment on compatibility with other packages that co-opt the output routine.
+%
+% Somewhere on the LaTeX-L list, David Kastrup has urged macro writers to take over
+% the output routine in such a way that others can do likewise. How is this to be
+% accomplished?
+%
+% Consider what the \classname{lineno} package does when it loads.
+% \begin{enumerate}
+% \item
+% It does \cmd\let\ cmd\@tempa\ \cmd\output.
+% This has the effect of identifying \cmd\@tempa\ with the
+% \cmd\toks\ register we created above to hold the old output routine of \LaTeX.
+% Let us say that was \cmd\toks14.
+% \item
+% \classname{lineno} itself effectively does \cmd\newtoks\ \cmd\@LN@output,
+% which assigns that \cmd\csname\ to \cmd\toks 15.
+% \item
+% It loads \cmd\@LN@output\ with the contents of \cmd\@tempa
+% (that is, \cmd\toks14, our copy of \LaTeX's output routine).
+% \item
+% Then it loads \cmd\@tempa\ with its own desired procedure,
+% to be executed at \cmd\output\ time, thereby taking over what it thinks is the output routine,
+% but which is in reality the procedure \revtex\ executes when it wants to pass control
+% to \LaTeX's original output routine.
+% \item
+% It then does \cmd\let\ \cmd\output\ \cmd\@LN@output,
+% which now identifies \cmd\output\ with \cmd\toks15, the output routine of \classname{lineno}.
+% \item
+% When the \cmd\output\ routine is triggered,
+% the primitive output routine \cmd\primitive@output\ is executed, and if
+% appropriate, control is passed to \cmd\output@latex, which \revtex\ had loaded with the old
+% \LaTeX\ output routine, but which is presently loaded with that of \classname{lineno}.
+% \item
+% The output routine of \classname{lineno} is executed, and if appropriate control is passed
+% to \cmd\@LN@output, the old output routine of \LaTeX.
+% \item
+% Furthermore, the \cmd\csname\ \cmd\output\ now points to \cmd\@LN@output\ (\cmd\toks15).
+% This means that someone coming in after \classname{lineno} to take over the output routine
+% will actually get executed after that of \classname{lineno}, but before \LaTeX.
+% \end{enumerate}
+% As you can see, the process of taking over the output routine may continue until all of the
+% \cmd\toks\ registers have been allocated.
+% If, say, \classname{newpackage} would itself like to take over the output routine,
+% and if it uses the above set of steps, then when the output routine is triggered, the order
+% of execution is \revtex, then \classname{lineno}, then \classname{newpackage}, then \LaTeX.
+% Each new package inserts itself on front of \LaTeX.
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\dispatch@output}
+% \changes{4.1b}{2008/08/04}{(AO, 456) Compatibility with other packages that override the output routine, following suggestion by David Kastrup.}
+% We now install our own output routine in place of the original
+% output routine of \LaTeX, which is still available as \cmd\the\ \cmd\output.
+%
+% The output routine is simply the procedure \cmd\dispatch@output.
+% It either dispatches to a procedure based on a particular value of
+% \cmd\outputpenalty\ or it executes \cmd\the\cmd\output@latex\ tokens.
+% \begin{macrocode}
+\primitive@output{\dispatch@output}%
+\def\dispatch@output{%
+ \let\par\@@par
+% \end{macrocode}
+% Try to interpret \cmd\outputpenalty\ as a dispatcher to a message handler,
+% its value is, e.g., \cmd\do@startpage@pen.
+% \begin{macrocode}
+ \expandafter\let\expandafter\output@procedure\csname output@\the\outputpenalty\endcsname
+% \end{macrocode}
+% If we have failed to find a dispatcher, then settle for \cmd\output@latex.
+% \begin{macrocode}
+ \@ifnotrelax\output@procedure{}{%
+ \expandafter\def\expandafter\output@procedure\expandafter{\the\output@latex}%
+ }%
+% \end{macrocode}
+% Now test if the dispatcher is the special case of \cmd\execute@message@pen, in which case execute the \cmd\@message@saved.
+% \changes{4.1n}{2009/12/02}{More diagnostics of column balancing}
+% \begin{macrocode}
+ \expandafter\@ifx\expandafter{\csname output@-\the\execute@message@pen\endcsname\output@procedure}{%
+ \let\output@procedure\@message@saved
+ }{}%
+ \ltxgrid@info@sw{\class@info{\string\dispatch@output}\say\output@procedure\saythe\holdinginserts}{}%
+ \outputdebug@sw{\output@debug}{}%
+ \output@procedure
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\set@output@procedure#1#2{%
+ \count@\outputpenalty\advance\count@-#2%
+ \expandafter\let\expandafter#1\csname output@\the\count@\endcsname
+}%
+% \end{macrocode}
+%
+% \changes{4.1n}{2009/12/02}{More diagnostics of column balancing}
+% The following procedure is executed at the beginning of each visit to the output routine,
+% contingent on the level of diagnostics specified.
+% However, it bails out when the visit is part of a tight sequence of visits to the output routine.
+% \begin{macrocode}
+\def\output@debug{%
+ \def\@tempa{\save@message}%
+ \@ifx{\output@procedure\@tempa}{%
+ \true@sw
+ }{%
+ \@ifnum{\outputpenalty=-\save@column@insert@pen}{%
+ \@ifnum{\holdinginserts>\z@}%
+ }{%
+ \false@sw
+ }%
+ }%
+ {}{\output@debug@}%
+}%
+\def\output@debug@{%
+%<ignore> \saythe\inputlineno
+ \saythe\outputpenalty
+ \saythe\interlinepenalty
+ \saythe\brokenpenalty
+ \saythe\clubpenalty
+ \saythe\widowpenalty
+ \saythe\displaywidowpenalty
+ \saythe\predisplaypenalty
+ \saythe\interdisplaylinepenalty
+ \saythe\postdisplaypenalty
+ \saythe\badness
+ \say\thepagegrid
+ \saythe\pagegrid@col
+ \saythe\pagegrid@cur
+%<ignore> \say\bot@envir
+ \saythe\insertpenalties
+%<ignore> \say\@@topmark
+%<ignore> \say\saved@@topmark
+%<ignore> \say\@@firstmark
+%<ignore> \say\saved@@firstmark
+ \say\@@botmark
+%<ignore> \say\saved@@botmark
+ \saythe\pagegoal
+ \saythe\pagetotal
+ \saythe{\badness\@cclv}%
+ \say\@toplist
+ \say\@botlist
+ \say\@dbltoplist
+ \say\@deferlist
+ \trace@scroll{%
+ \showbox\@cclv
+ \showbox\@cclv@saved
+ \showbox\pagesofar
+% \end{macrocode}
+% Klootch! The following line provides only for two-column page grid; if debugging more columns, you must
+% add more statements here.
+% \begin{macrocode}
+ \showbox\csname col@1\endcsname
+ \showbox\footsofar
+ \showbox\footins
+ \showbox\footins@saved
+ \showlists
+ }%
+}%
+\@ifxundefined{\outputdebug@sw}{%
+ \@booleanfalse\outputdebug@sw
+}{}%
+\def\trace@scroll#1{\begingroup\showboxbreadth\maxdimen\showboxdepth\maxdimen\scrollmode#1\endgroup}%
+\def\trace@box#1{\trace@scroll{\showbox#1}}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@outputpage}
+% \begin{macro}{\@outputpage@head}
+% \begin{macro}{\@outputpage@tail}
+% \changes{4.1b}{2008/08/04}{Procedure \cs{@outputpage@tail} tailpatches \cs{@outputpage}}%
+% \changes{4.1b}{2008/08/04}{Procedure \cs{@outputpage@head} headpatches \cs{@outputpage}}%
+% The procedure \cmd\@outputpage\ of standard \LaTeX\ is the sole place where a \cmd\shipout\ is carried out.
+% The procedures that build \cmd\@outputbox\ just before a page is shipped out by \cmd\@outputpage\ are:
+% \cmd\@makecolumn,
+% \cmd\@combinepage, and
+% \cmd\@combinedblfloats.
+%
+% We need to head- and tailpatch this procedure, so we perform here
+% the only modifications to that procedure that are essential.
+% Elsewhere, we will build up the meanings of \cmd\@outputpage@head\ and \cmd\@outputpage@tail.
+% \begin{macrocode}
+\prepdef\@outputpage{\@outputpage@head}%
+\let\@outputpage@head\@empty
+\appdef\@outputpage{\@outputpage@tail}%
+\let\@outputpage@tail\@empty
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\show@box@size}
+% \begin{macro}{\show@text@box@size}
+% \begin{macro}{\show@pagesofar@size}
+% \begin{macro}{\show@box@size@sw}
+% \begin{macro}{\total@text}
+% \changes{4.1b}{2008/08/04}{(AO, 452) Support length checking: show size of shipped out text.}
+% Procedure \cmd\show@box@size\ is a diagnostic for the sizes of boxes;
+% the boolean \cmd\show@box@size@sw\ turns it on and off.
+% \begin{macrocode}
+\def\show@box@size#1#2{%
+ \show@box@size@sw{%
+ \begingroup
+ \setbox\z@\vbox{\unvcopy#2\hrule}%
+ \class@info{Show box size: #1^^J%
+ (\the\ht\z@\space X \the\wd\z@)
+ \the\c@page\space\space\the\pagegrid@cur\space\the\pagegrid@col
+ }%
+ \endgroup
+ }{}%
+}%
+% \end{macrocode}
+% Procedure \cmd\show@text@box@size\ tallies the size of the indicated column.
+% If \cmd\box\ \cmd\pagesofar\ is a factor, then its height has been memorized
+% in the depth of the tally box.
+%
+% \changes{4.1f}{2009/07/10}{(AO, 518) Tally register overflow when locument is long}
+% \begin{macrocode}
+\def\show@text@box@size{%
+ \show@box@size{Text column}\@outputbox
+ \tally@box@size@sw{%
+ \@ifdim{\wd\@outputbox>\z@}{%
+ \dimen@\ht\@outputbox\divide\dimen@\@twopowerfourteen
+ \advance\dimen@-\dp\csname box@size@\the\pagegrid@col\endcsname
+ \@ifdim{\dimen@>\z@}{%
+ \advance\dimen@ \ht\csname box@size@\the\pagegrid@col\endcsname
+ \global\ht\csname box@size@\the\pagegrid@col\endcsname\dimen@
+ \show@box@size@sw{%
+ \class@info{Column: \the\dimen@}%
+ }{}%
+ }{}%
+ }{}%
+ \global\dp\csname box@size@\the\pagegrid@col\endcsname\z@
+ }{}%
+}%
+% \end{macrocode}
+% Take the height of \cmd\box\ \cmd\pagesofar\ into account.
+% \begin{macrocode}
+\def\show@pagesofar@size{%
+ \show@box@size{Page so far}\pagesofar
+ \dimen@\ht\pagesofar\divide\dimen@\@twopowerfourteen
+ \global\dp\csname box@size@1\endcsname\dimen@
+ \show@box@size@sw{%
+ \class@info{Pagesofar: \the\dimen@}%
+ }{}%
+}%
+\@booleanfalse\tally@box@size@sw
+\@booleanfalse\show@box@size@sw
+\expandafter\newbox\csname box@size@1\endcsname
+\expandafter\setbox\csname box@size@1\endcsname\hbox{}%
+\expandafter\newbox\csname box@size@2\endcsname
+\expandafter\setbox\csname box@size@2\endcsname\hbox{}%
+\def\total@text{%
+ \@tempdima\the\ht\csname box@size@2\endcsname\divide\@tempdima\@twopowertwo\@tempcnta\@tempdima
+ \@tempdimb\the\ht\csname box@size@1\endcsname\divide\@tempdimb\@twopowertwo\@tempcntb\@tempdimb
+ \class@info{Total text: Column(\the\@tempcnta pt), Page(\the\@tempcntb pt)}%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+%
+% \subsection{Further thoughts about inserts}
+%
+% The only safe way to deal with inserts is to either set \cmd\holdininserts\ or
+% to commit to using whatever insert comes your way: you cannot change your mind
+% once you see a non-void \cmd\box\cmd\footins, say.
+%
+% Therefore all output routine processing must proceed with \cmd\holdinginserts\ set
+% until you are sure of the material to be committed to the page. At that point, you
+% can clear \cmd\holdinginserts, spew \cmd\box\cmd\@cclv, put down the appropriate penalty,
+% and exit, with the knowledge that \TeX\ will re-find the same pagebreak, this time
+% visiting the output routine with everything, including inserts, in their proper
+% place.
+% This technique applies to split elements (screens, longtable, index) as well as to
+% manufactured pages (float pages and clearpage pages).
+%
+% Therefore, the output routine must not make assumptions about whether \cmd\holdinginserts\
+% should be cleared; instead this must be left to the one-off output routines or the natural output routine.
+%
+% If we are manufacturing pages (``float page processing''), and if \cmd\pagegoal\ is not equal to
+% \cmd\vsize, then inserts are at hand, and our criterion should take into account the insert
+% material, even though we cannot measure its height based on the size of \cmd\box\cmd\footins\
+% (because \cmd\holdinginserts\ is set, you see).
+%
+% It would be better to take the complement of \cmd\floatpagefraction\ and use that
+% as a standard for the looseness of the page. Since \cmd\pagegoal\ reflects the inserted material,
+% the criterion becomes the difference of the aggregate height of the floats and the \cmd\pagegoal\
+% versus this "page looseness" standard.
+%
+% As a check, consider what happens if we bail out: \cmd\@deferlist\ has never been touched, so it
+% requires no attention. Also, \cmd\holdinginserts\ has never been cleared, so inserts require
+% no attention. So we only have to ensure that marks are preserved, which is already taken
+% care of by the message handler mechanism.
+%
+% If we are doing ordinary page cutting, then the scheme would be to detect whether we are within
+% a screen (or longtable as may be), do the adjustment to the page height, and return, but this time
+% with \cmd\holdinginserts\ cleared. Upon reentering the output routine, we may or may not be within
+% the screen environment, but we are now sure to have a final page break, and we can commit this
+% material (by shipping out or by saving it out as a full column).
+%
+% In the above, the first of the two visits to the output routine is a dead cycle and requires
+% propagation of marks, but nothing else.
+%
+% \subsection{The difference between inserts and floats}
+%
+% While revisiting this package in 2008, I needed to clarify under what circumstances inserts
+% would be added to the \cmd\pagesofar. My conclusion is that I had been treating them
+% exactly the same as floats, but that was a mistake.
+%
+% Floats can be committed at the top of a column, in the middle, or at the bottom.
+% Footnotes (the only \cmd\insert\ that is used in \LaTeX) may only be committed at the bottom
+% of a column. So, it was necessary to provide two versions of \cmd\@combinepage, one
+% that committed \cmd\insert s, and the other that did not, the former used only when
+% a column of text was committed. Note that even after a column is committed, we could change our minds:
+% for instance if in multicolumn grid and we decide to balance the columns.
+%
+%
+% \subsection{The natural output routine}
+%
+% Here is the portion of the output routine that fields cases not handled by
+% the dispatcher.
+%
+% The default is to ship out a page and then look around for more material
+% that might constitute a ``float page''. However, because \cmd\holdinginserts\
+% is normally set, this output routine must first have a dead cycle and
+% come back again with \cmd\holdinginserts\ cleared.
+% Then, after shipping out, it puts down a message that
+% will manufacture zero or more float pages, finally terminating
+% with a procedure that commits floats to a new unfinished page.
+%
+% To accomodate special processing, we execute hooks whose name is based
+% on the value of the "envir" mark component. The default is "document",
+% ensured by an initial mark of that value; the associated procedures
+% are all nil. Any unknown envir value will "\cmd\relax\ out".
+%
+% The test made by \cmd\toggle@insert\ tells whether we are on our first visit to
+% the output routine (with \cmd\holdinginserts\ still positive), or our second
+% (with \cmd\holdinginserts\ zeroed). The output routine will toggle the setting.
+%
+% The commands \cmd\hold@insertions\ and \cmd\move@insertions\ respectively
+% clear and set \cmd\holdinginserts, so this procedure effectively
+% clears \cmd\holdinginserts\ just long enough to pick up the insertions.
+% Important: any output routine that clears \cmd\holdinginserts\
+% must guarentee that it is restored on the subsequent visit to the output routine.
+% Or, to put it another way, if an output routine detects that \cmd\holdinginserts\
+% is cleared, it should take it upon itself to restore it to a positive value before exiting.
+%
+% The branch with \cmd\holdinginserts\ set is executed first; the other
+% branch follows on practically immediately thereafter. In the first branch,
+% we simply execute the appropriate hook and then execute a dead cycle.
+%
+% In the branch with \cmd\holdinginserts\ cleared, the procedure
+% builds up the current column, which is now complete, with \cmd\@makecolumn, then
+% dispatches to the shipout routine associated with the current page grid, \cmd\output@column@.
+% At the end, it triggers the execution of an output routine to prepare the next column (or page).
+%
+% \subsection{Natural output routine}%
+%
+% \begin{macro}{\natural@output}
+% \begin{macro}{\output}
+% Here is the output routine that handles natural pagebreaks:
+% we now have page that needs to be shipped out or a portion of a page that is ready to be committed to the page grid.
+% Processing is of necessity divided into phases, \cmd\output@holding\ is executed upon first encountering the natural page-breaking point, while inserts are being held.
+% The second phase, \cmd\output@moving, is set in motion by the first: here the same material (in most cases) will be processed with \cmd\holdinginserts\ cleared,
+% and the insertions (e.g., footnotes) are split off into their assigned box registers.
+% \begin{macrocode}
+\def\natural@output{\toggle@insert{\output@holding}{\output@moving}}%
+\output@latex{\natural@output}%
+% \end{macrocode}
+%
+% In accordance with the scheme suggested by David Kastrup for allowing another output routine
+% to slip itself into ours, we use a token register called \cmd\output.
+% However, we reserve the ability to restore things if we so desire.
+% This we must do in the case of the \filename{lineno.sty} package, because its functionality
+% is best served by being integrated into our own dispatcher-based output routine.
+%
+% To restore our own output routine, we can repeat the above assignment,
+% \begin{verbatim}
+%\output@latex{\natural@output}%
+% \end{verbatim}
+% some time before the document begins.
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\output@holding}
+% \begin{macro}{\@if@exceed@pagegoal}
+% The procedure \cmd\output@holding\
+% is our first cycle through the output routine; \cmd\holdinginserts\ is still set.
+% We give the current environment a heads up
+% (it is through this means that \classname{longtable} sets its running header and footer),
+% then we execute a dead cycle, which should propagate marks.
+%
+% One corner case that can crop up is the presence of a single unbreakable chunk whose size is larger
+% than \cmd\vsize.
+% Doing a dead cycle under such circumstances will not find the same breakpoint as this time
+% (remember we threw in a \cmd\mark\ node).
+% Instead, we attempt to remove the excess height of the material, so we can continue to propagate marks.
+%
+% The corner case is at hand if the natural size of \cmd\box\cmd\@cclv\ exceeds \cmd\pagegoal\ and
+% the contents cannot be shrunk to fit.
+%
+% \changes{4.1b}{2008/08/04}{New procedure for showing a box contents, \cs{trace@box}}
+% \begin{macrocode}
+\def\output@holding{%
+ \csname output@init@\bot@envir\endcsname
+ \@if@exceed@pagegoal{\unvcopy\@cclv}{%
+ \setbox\z@\vbox{\unvcopy\@cclv}%
+ \outputdebug@sw{\trace@box\z@}{}%
+ \dimen@\ht\@cclv\advance\dimen@-\ht\z@
+ \dead@cycle@repair\dimen@
+ }{%
+ \dead@cycle
+ }%
+}%
+\def\@if@exceed@pagegoal#1{%
+ \begingroup
+ \setbox\z@\vbox{#1}%
+ \dimen@\ht\z@\advance\dimen@\dp\z@
+ \outputdebug@sw{\saythe\dimen@}{}%
+ \@ifdim{\dimen@>\pagegoal}{%
+ \setbox\z@\vbox{\@@mark{}\unvbox\z@}%
+ \splittopskip\topskip
+ \splitmaxdepth\maxdepth
+ \vbadness\@M
+ \vfuzz\maxdimen
+ \setbox\tw@\vsplit\z@ to\pagegoal
+ \outputdebug@sw{\trace@scroll{\showbox\tw@\showbox\z@}}{}%
+ \setbox\tw@\vbox{\unvbox\tw@}%
+ \@ifdim{\ht\tw@=\z@}{%
+ \ltxgrid@info{Found overly large chunk while preparing to move insertions. Attempting repairs}%
+ \aftergroup\true@sw
+ }{%
+ \aftergroup\false@sw
+ }%
+ }{%
+ \aftergroup\false@sw
+ }%
+ \endgroup
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\output@moving}
+% \begin{macro}{\@cclv@nontrivial@sw}
+% The procedure \cmd\output@moving\
+% is our second cycle through the output routine; \cmd\holdinginserts\ is now cleared,
+% and \cmd\insert s will have been split off into their respective box registers, like \cmd\footins.
+%
+% \begin{enumerate}
+% \item
+% Set the values of \cmd\topmark\ and \cmd\firstmark.
+% \item
+% If we got here because of a \cmd\clearpage\ command, remove the protection box that this mechanism has left on the MVL.
+% \item
+% If the contents of \cmd\box\cmd\@cclv\ are non-trivial, commit it to the current page (as a column) or ship it out, as the case may call for.
+% \item
+% If not, discard it (we are at the end of \cmd\clearpage\ processing).
+% \item
+% Set various values, including the available space for setting type on the next column (\cmd\@colroom).
+% \end{enumerate}
+%
+% The processing for a non-trivial \cmd\box\cmd\@cclv\ are:
+% \begin{enumerate}
+% \item
+% Execute the head procedure for the current environment.
+% \item
+% Make up a column and ship it out (or commit it to the current page) via a procedure keyed to the current page grid.
+% \item
+% Put down an interrupt for \cmd\do@startcolumn@pen: this will force a visit to the output routine for the
+% purpose of committing floats to the next column.
+% \item
+% Possibly put down an interrupt to continue \cmd\clearpage\ proccessing.
+% \item
+% Execute the tail procedure for the current environment.
+% \end{enumerate}
+%
+%
+% The processing for a trivial \cmd\box\cmd\@cclv\ are:
+% \begin{enumerate}
+% \item
+% Void out \cmd\box\cmd\@cclv\ and give appropriate warning messages and diagnostics.
+% \item
+% Put down the same interrupts as for the non-trivial case above.
+% \end{enumerate}
+%
+% This instance of \cmd\@makecolumn\ is followed by \cmd\output@column@,
+% that is, it builds a column for \cmd\shipout\ rather than for adding to \cmd\pagesofar.
+%
+% We need to handle cases where the \cmd\output@pre@, \cmd\output@column@, or \cmd\output@post@\
+% dispatchers come up \cmd\relax ed out: the default is to execute the corresponding
+% procedures from the \env{docuemnt} environment and the one-column grid respectively.
+%
+% One such case comes up with frequency: at the end of the document, where the \cmd\botmark\ is
+% now empty.
+% \changes{4.1b}{2008/08/04}{Change \cs{@makecol} to \cs{@makecolumn} with argument}
+% \changes{4.1b}{2008/08/04}{Change \cs{set@colroom} to \cs{set@colht}}
+% \changes{4.1n}{2010/01/02}{(AO, 571) Use procedures \cs{output@do@prep} and \cs{output@column@do} as dispatchers}
+% \begin{macrocode}
+\def\output@moving{%
+ \set@top@firstmark
+ \@ifnum{\outputpenalty=\do@newpage@pen}{%
+ \setbox\@cclv\vbox{%
+ \unvbox\@cclv
+ \remove@lastbox
+ \@ifdim{\ht\z@=\ht\@protection@box}{\box\lastbox}{\unskip}%
+ }%
+ }{}%
+ \@cclv@nontrivial@sw{%
+ \expandafter\output@do@prep\csname output@prep@\bot@envir \endcsname
+ \@makecolumn\true@sw
+ \expandafter\output@column@do\csname output@column@\thepagegrid\endcsname
+ \protect@penalty\do@startcolumn@pen
+ \clearpage@sw{%
+ \protect@penalty\do@endpage@pen
+ }{}%
+ \expandafter\let\expandafter\output@post@\csname output@post@\bot@envir \endcsname
+ \outputdebug@sw{\say\output@post@}{}%
+ \@ifx{\output@post@\relax}{\output@post@document}{\output@post@}%
+ }{%
+ \void@cclv
+ }%
+ \set@colht
+ \global\@mparbottom\z@
+ \global\@textfloatsheight\z@
+}%
+% \end{macrocode}
+% Procedure \cmd\output@do@prep\ dispatches to the proper procedure to prepare page.
+% \begin{macrocode}
+\def\output@do@prep#1{%
+ \outputdebug@sw{\class@info{Prep: \string#1}}{}%
+ \@ifx{#1\relax}{\output@prep@document}{#1}%
+}%
+% \end{macrocode}
+% Procedure \cmd\output@column@do\ dispatches to the proper procedure to output column or page.
+% \begin{macrocode}
+\def\output@column@do#1{%
+ \outputdebug@sw{\class@info{Output column: \string#1}}{}%
+ \@ifx{#1\relax}{\output@column@one}{#1}%
+}%
+\def\void@cclv{\begingroup\setbox\z@\box\@cclv\endgroup}%
+\def\remove@lastbox{\setbox\z@\lastbox}%
+% \end{macrocode}
+%
+% The procedure \cmd\@cclv@nontrivial@sw\ determines if this visit to \cmd\output@moving\
+% is a trivial one, which happens at the end of \cmd\clearpage\ processing and under some pathological circumstances.
+% It emits a Boolean, so it is syntactically like \cmd\true@sw, albeit does not execute solely via expansion.
+%
+% Note: the case where \cmd\box\cmd\@cclv\ is void comes up at the very beginning of the job, when
+% typesetting a (full-page-width) title block in a two-column layout.
+%
+% Note: the code that removes the last box and skip from the output is intended to detect the case
+% where the output has whatsit nodes followed by topskip and a protection box.
+% This is what happens under normal circumstances at the end of \cmd\clearpage\ processing.
+% \changes{4.1b}{2008/08/04}{New procedure for showing a box contents, \cs{trace@box}}
+% \begin{macrocode}
+\def\@cclv@nontrivial@sw{%
+ \@ifx@empty\@toplist{%
+ \@ifx@empty\@botlist{%
+ \@ifvoid\footins{%
+ \@ifvoid\@cclv{%
+ \false@sw
+ }{%
+ \setbox\z@\vbox{\unvcopy\@cclv}%
+ \@ifdim{\ht\z@=\topskip}{%
+ \setbox\z@\vbox\bgroup
+ \unvbox\z@
+ \remove@lastbox
+ \dimen@\lastskip\unskip
+ \@ifdim{\ht\z@=\ht\@protection@box}{%
+ \advance\dimen@\ht\z@
+ \@ifdim{\dimen@=\topskip}{%
+ \aftergroup\true@sw
+ }{%
+ \aftergroup\false@sw
+ }%
+ }{%
+ \aftergroup\false@sw
+ }%
+% \end{macrocode}
+% End of \cmd\box\cmd\z@.
+% \begin{macrocode}
+ \egroup
+ {%
+% \end{macrocode}
+% Normal for \clearpage
+% \begin{macrocode}
+ \false@sw
+ }{%
+ \true@sw
+ }%
+ }{%
+ \@ifdim{\ht\z@=\z@}{%
+ \ltxgrid@info{Found trivial column. Discarding it}%
+ \outputdebug@sw{\trace@box\@cclv}{}%
+ \false@sw
+ }{%
+ \true@sw
+ }%
+ }%
+ }%
+ }{%
+ \true@sw
+ }%
+ }{%
+ \true@sw
+ }%
+ }{%
+ \true@sw
+ }%
+}%
+% \end{macrocode}
+%
+%
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\protect@penalty}
+% The procedure \cmd\protect@penalty\ is the utility procedure for invoking a
+% one-off output routine. Such a routine can expect to find the protection box
+% above it in \cmd\box\cmd\@cclv: it should remove that box.
+%
+% Note that \cmd\execute@message\ does the same thing as \cmd\protect@penalty, but
+% in a slightly different way.
+%
+% We create a specially formulated box that will be universally used when a protection box is needed.
+% In this way, we can always recognize when \cmd\box\cmd\@cclv\ is trivial:
+% it will consist of whatsits followed by \cmd\topskip\ glue and the \cmd\@protection@box.
+% \begin{macrocode}
+\def\protect@penalty#1{\protection@box\penalty-#1\relax}%
+\newbox\@protection@box
+\setbox\@protection@box\vbox to1986sp{\vfil}%
+\def\protection@box{\nointerlineskip\copy\@protection@box}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\dead@cycle}
+% \begin{macro}{\dead@cycle@repair}
+% The procedure \cmd\dead@cycle\ is defined separately as a utility which can be used by
+% any output processing routine to emulate what takes place in the standard output routine.
+%
+% Here, we have entered the output routine with \cmd\holdinginserts\ enabled, which means that we
+% are not yet ready to ship out material, because the \cmd\insert\ registers are being held.
+% We want to clear \cmd\holdinginserts\ and come back here with the same page break as before, whereupon
+% we may properly proceed with page makeup.
+%
+% To do this, we
+% propagate marks, then spew the contents of \cmd\box\cmd\@cclv\ followed by the
+% original output penalty that landed us here (but only if it is not 10000,
+% the flag value for a pagebreak not at a penalty).
+%
+% However, the natural output routine should do this only if \cmd\box\cmd\@cclv\ is nontrivial.
+% A pathological case exists wherein a box of height greater than \cmd\textheight\ would cause an infinite loop involving the output routine.
+% The procedure \cmd\dead@cycle@repair, attempts to catch this case and avoid the loop.
+%
+% The test of the height of \cmd\box\cmd\@cclv\ is not the correct one, because this test will run afoul in
+% the case where \cmd\box\cmd\@cclv\ contains nothing but an \cmd\insert\ node. What to do?
+%
+% It is possible that the pathological case can be detected by looking at \cmd\pagetotal. If that quantity is
+% zero, then \cmd\box\cmd\@cclv\ really is trivial.
+%
+% In the procedure \cmd\dead@cycle@repair, if \cmd\box\cmd\@cclv\ is nontrivial, we execute \cmd\dead@cycle,
+% otherwise it contains nothing but a mark, so we dispense with propagating marks
+% and we simply spew out \cmd\box\cmd\@cclv\ without an accompanying mark.
+% This has the effect of failing to propagate marks, but this problem is preferrable to the infinite loop,
+% which in principle could crash even a robust operating system by filling up the file system.
+%
+% If a document has such a large chunk, it should be fixed, so we give a message in the log.
+%
+% You ask, ``In what way does this infinite loop come about?'' Good question!
+%
+% The setup is a chunk in the MVL that is taller than \cmd\textheight.
+% (Yes, it's that simple.)
+% As soon as the previous page ships out, the MVL will contain a mark (propagated from the previous page) followed
+% by that large chunk (call it the `big bad box', albeit does not need to be a single box).
+% The next visit to the output routine will be a natural page break, but
+% \TeX\ will select the juncture between the mark and the big bad box as the least-cost page break.
+% Unless the test in \cmd\dead@cycle\ is done, the cycle is perpetuated when the macro
+% reinserts the mark.
+%
+% The crux matter is achieving, in a robust way, the goal of going from a \cmd\holdinginserts\ state to one
+% where the insertions are moving.
+%
+% \begin{macrocode}
+\def\dead@cycle@repair#1{%
+ \expandafter\do@@mark
+ \expandafter{%
+ \@@botmark
+ }%
+ \unvbox\@cclv
+ \nointerlineskip
+ \vbox to#1{\vss}%
+ \@ifnum{\outputpenalty<\@M}{\penalty\outputpenalty}{}%
+}%
+\def\dead@cycle@repair@protected#1{%
+ \expandafter\do@@mark
+ \expandafter{%
+ \@@botmark
+ }%
+ \begingroup
+ \unvbox\@cclv
+% \end{macrocode}
+% Remove the protection box
+% \begin{macrocode}
+ \remove@lastbox
+ \nointerlineskip
+ \advance#1-\ht\@protection@box
+ \vbox to#1{\vss}%
+ \protection@box % Reinsert protection box
+ \@ifnum{\outputpenalty<\@M}{\penalty\outputpenalty}{}%
+ \endgroup
+}%
+\def\dead@cycle{%
+ \expandafter\do@@mark
+ \expandafter{%
+ \@@botmark
+ }%
+ \unvbox\@cclv
+ \@ifnum{\outputpenalty<\@M}{\penalty\outputpenalty}{}%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\output@init@document}
+% \begin{macro}{\output@prep@document}
+% \begin{macro}{\output@post@document}
+% The default processing simply provides for insertion of held-over footnotes.
+% At a natural page break, we are either at the bottom of a column or at the bottom
+% of a page. In either case, the \cmd\output@init@\ processing adjusts for the height
+% of the held-over footnotes and bails out.
+% Upon our return, at \cmd\output@prep@\ time, the page break will accomodate the material;
+% it is now actually inserted by concatenating it with the contents of \cmd\footins.
+% The default processing for \cmd\output@post@\ is nil.
+% \changes{4.1b}{2008/08/04}{Box \cs{footbox} changed to box \cs{footsofar}}
+% \changes{4.1n}{2009/12/02}{More diagnostics of column balancing}
+% \begin{macrocode}
+\def\output@init@document{%
+ \ltxgrid@info@sw{\class@info{\string\output@init@document}}{}%
+ \global\vsize\vsize
+}%
+% \end{macrocode}
+% QUERY: the following procedure is very like \cmd\combine@foot@inserts. Should it be the same?
+% Answer: no, the two differ: this procedure makes a local assignment of \cmd\footins;
+% the latter makes a global assignment of \cmd\footsofar.
+%
+% Note: In a multicolumn document, footnotes must \emph{not} be balanced at this point.
+% \changes{4.1n}{2009/12/02}{More diagnostics of column balancing}
+% \begin{macrocode}
+\def\output@prep@document{%
+ \ltxgrid@foot@info@sw{\class@info{\string\output@prep@document}\trace@scroll{\showbox\footins\showbox\footsofar}}{}%
+ \@ifvoid\footsofar{%
+ }{%
+ \global\setbox\footins\vbox\bgroup
+ \unvbox\footsofar
+ \@ifvoid\footins{}{%
+ \marry@baselines
+ \unvbox\footins
+ }%
+ \egroup
+ \ltxgrid@foot@info@sw{\trace@box\footins}{}%
+ }%
+}%
+\def\output@post@document{}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\@opcol}
+% The standard \LaTeX\ procedure \cmd\@opcol\ is now completely obsoleted.
+% \begin{macrocode}
+\let\@opcol\@undefined
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@makecolumn}
+% \changes{4.1b}{2008/08/04}{Change \cs{@makecol} to \cs{@makecolumn} with argument}
+% The procedure \cmd\@makecolumn\ packages up a page along with all its insertions and floats.
+% Therefore it is essential that it be executed with \cmd\holdininserts\ cleared.
+%
+% Note that there is a corner case when in a multi-column grid, where the change back to
+% one-column grid occurs just after a complete page ships out. We want to detect when
+% \cmd\@cclv\ contains nothing but a \cmd\mark, but this is a \TeX\ impossibility.
+%
+% Note on \cmd\@kludgeins: we have removed this mechanism from \LaTeX, because the implementation
+% of \cmd\enlargethispage\ no longer requires it.
+% Here, for consistency sake, we remove \cmd\@makespecialcolbox.
+%
+% The argument of \cmd\@makecolumn\ is a Boolean and determines if we combine the footnote material
+% into the present column. If the procedure is building a column for shipping out, then we will
+% combine the footnote material, if not, we return with the \cmd\footins\ box unchanged.
+%
+% \changes{4.1f}{2009/07/14}{(AO, 519) \cs{footins} content must be preserved and reintegrated}
+% \changes{4.1n}{2009/12/02}{More diagnostics of column balancing}
+% \changes{4.1n}{2010/01/02}{(AO, 571) coding convention: use \cs{bgroup}, \cs{egroup} (instead of braces) when a box is being built}%
+% \changes{4.1n}{2010/01/02}{(AO, 571) change calling sequence of \cs{combine@foot@inserts} to expose box registers}%
+% I changed the behavior of this procedure in the case where the argument is \cmd\false@sw:
+% send the unused footnote material to \cmd\footsofar.
+% \begin{macrocode}
+\def\@makecolumn#1{%
+ \ltxgrid@foot@info@sw{\class@info{\string\@makecolumn\string#1}}{}%
+ \setbox\@outputbox\vbox\bgroup
+ \boxmaxdepth\@maxdepth
+ \@tempdima\dp\@cclv
+ \unvbox\@cclv
+ \vskip-\@tempdima
+ \egroup
+ \xdef\@freelist{\@freelist\@midlist}\global\let\@midlist\@empty
+ \show@text@box@size
+ \@combinefloats
+ #1{%
+ \@combineinserts\@outputbox\footins
+ }{%
+ \combine@foot@inserts\footsofar\footins
+ }%
+ \set@adj@colht\dimen@
+ \count@\vbadness
+ \vbadness\@M
+ \setbox\@outputbox\vbox to\dimen@\bgroup
+ \@texttop
+ \dimen@\dp\@outputbox
+ \unvbox\@outputbox
+ \vskip-\dimen@
+ \@textbottom
+ \egroup
+ \vbadness\count@
+ \global\maxdepth\@maxdepth
+}%
+\let\@makespecialcolbox\@undefined
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@combineinserts}
+% The procedure to add the specified insertions to the packaged-up page.
+% All other classes of insertions should also be dealt with at this time.
+%
+% Note that the second argument must be a \cmd\newinsert\ register: we access the \cmd\box\ along with the \cmd\skip.
+% \changes{4.1n}{2009/12/02}{More diagnostics of column balancing}
+% \changes{4.1n}{2010/01/02}{(AO, 571) coding convention: use \cs{bgroup}, \cs{egroup} (instead of braces) when a box is being built}%
+% \changes{4.1n}{2010/01/02}{(AO, 571) footnote rule is leaders, so that it may be removed by \cs{vsplit}; mechanism of \cs{kern} signals to indicate footnote height}%
+% \begin{macrocode}
+\def\@combineinserts#1#2{%
+ \ltxgrid@foot@info@sw{\class@info{\string\@combineinserts\string#1\string#2}\trace@box#2}{}%
+ \setbox#1\vbox\bgroup
+ \unvbox#1%
+ \@ifvoid{#2}{}{%
+ \dimen@\ht#2\advance\dimen@\dp#2\advance\dimen@\skip#2%
+ \show@box@size{Combining inserts}#2%
+ \vskip\skip#2%
+% \end{macrocode}
+% The footnote rule is created as leaders, so that it may be removed automatically (via \cmd\vsplit)
+% in the event the footnote is recovered from this column.
+% Note that if \cmd\color@begingroup\ or \cmd\normalcolor\ produce marks, this technique will be confounded.
+% \begin{macrocode}
+ \setbox\z@\vbox{\footnoterule}\dimen@i\ht\z@
+ \color@begingroup
+ \normalcolor
+ \cleaders\box\z@\vskip\dimen@i\kern-\dimen@i
+ \csname combine@insert@\the\pagegrid@col\endcsname#2%
+ \color@endgroup
+% \end{macrocode}
+% The following tells \cmd\recover@column\ the size of the footnotes added here, including the skip glue above.
+% \begin{macrocode}
+ \kern-\dimen@\kern\dimen@
+ }%
+ \egroup
+ \ltxgrid@foot@info@sw{\trace@box#1}{}%
+}%
+% \end{macrocode}
+% We provide for a layer of abstraction for the laying down of footnotes at the bottom of this column or page.
+% \end{macro}
+%
+% \begin{macro}{\combine@insert@tw@}
+% \begin{macro}{\combine@insert@@ne}
+% \begin{macro}{\twocolumn@grid@setup}
+% \begin{macro}{\onecolumn@grid@setup}
+% \begin{macro}{\columngrid@setup}
+% \changes{4.1p}{2010/02/24}{(AO, 583) Provide setup code also for footnotes in a one-column document}%
+% The following two definitions cover the cases of a two-column document (with footnotes set on a single-column width),
+% and a one-column document.
+% However, the case of a two-column document with footnotes set on full text width is not covered.
+%
+% For a document in an overall two-column page grid, execute the commands
+% \cmd\twocolumn@grid@setup\ followed by \cmd\open@twocolumn;
+% if on the full page width (one-column grid), the command \cmd\onecolumn@grid@setup.
+%
+% The following is the way REVTeX does the initialization.
+% The procedure \cmd\select@column@grid\ is executed at \cmd\AtBeginDocument\ time;
+% the boolean \cmd\twocolumn@sw\ selects between the two alternatives.
+% \begin{verbatim}
+%\def\select@column@grid{%
+% \twocolumn@sw{%
+% \twocolumn@grid@setup
+% \open@twocolumn
+% }{%
+% \onecolumn@grid@setup
+% }%
+%}%
+%\appdef\class@documenthook{%
+% \select@column@grid
+%}%
+% \end{verbatim}
+%
+% \begin{macrocode}
+\def\combine@insert@tw@#1{%
+ \compose@footnotes@two#1\@ifvbox{#1}{\unvbox}{\box}#1%
+}%
+\def\combine@insert@@ne#1{%
+ \compose@footnotes@one#1\@ifvbox{#1}{\unvbox}{\box}#1%
+}%
+\def\twocolumn@grid@setup{%
+ \expandafter\let\csname combine@insert@1\endcsname\combine@insert@tw@
+ \expandafter\let\csname combine@insert@2\endcsname\combine@insert@@ne
+}%
+\def\onecolumn@grid@setup{%
+ \expandafter\let\csname combine@insert@1\endcsname\combine@insert@@ne
+ \expandafter\let\csname combine@insert@2\endcsname\combine@insert@@ne
+}%
+\let\columngrid@setup\onecolumn@grid@setup
+\columngrid@setup
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\@floatplacement}
+% In standard \LaTeX, someone (DPC?) makes the assumption that \cmd\@fpmin\ can be assigned
+% locally. This is no longer true now that we ship no more than one page per visit to the output routine.
+% We apply a bandaid.
+% \begin{macrocode}
+\appdef\@floatplacement{%
+ \global\@fpmin\@fpmin
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\pagebreak@pen}
+% While we are in the way of registering certain penalty values,
+% let us register the smallest one that will force a visit to the output routine.
+% However, this penalty will not have an assciated macro: we wish to execute the
+% natural output routine instead.
+%
+% Note that this penalty is invoked by \cmd\clearpage\ and \cmd\newpage.
+% \begin{macrocode}
+\mathchardef\pagebreak@pen=\@M
+\expandafter\let\csname output@-\the\pagebreak@pen\endcsname\relax
+% \end{macrocode}
+% \end{macro}
+%
+%
+% \subsection{Float placement}%
+%
+% \begin{macro}{\do@startcolumn@pen}
+% The procedure \cmd\do@startcolumn@pen\ is executed as a one-off output routine
+% just after a page is shipped out (or, in a multicolumn page grid, a column is salted away).
+%
+% Its job is to either generate a ``float page'' (in reality a column) for shipping out,
+% or to commit deferred floats to the fresh column, concluding with a dead cycle.
+% In the former case, we accomodate split footnotes and other insertions (by comparing \cmd\vsize\ and \cmd\pagegoal):
+% the floats are spewed onto the page, whereupon \LaTeX's output routine will place the footnotes and ship out,
+% iterating the process once again.
+%
+% Note that when this procedure is invoked, \cmd\box\cmd\@cclv\ still has within it the protection box, so we
+% start by removing it. Note also that if there was a split insertion held over from the previous page, the
+% insert node will be present in \cmd\box\cmd\@cclv, \emph{prior to} the protection box. For this reason, we cannot
+% just throw away that box, as we might be tempted to do.
+%
+% FIXME: where else do we possibly inappropriately discard \cmd\box\cmd\@cclv?
+%
+% Note that, because a column or page page had previously just been completed,
+% we can assume that there is nothing of importance on the page,
+% and because no message is being passed, we can preserve marks in a simple way.
+%
+% A Note on terminology:
+% In a single-column page grid, you might expect that we would execute the procedure \cmd\do@startpage.
+% But this is not so.
+% \LaTeX\ has a confustion of long standing,
+% in which the procedures that handle full-page width floats in a two-column page grid all have in their names
+% the string `dbl', which erroneously suggests having something to do with ``double''. It does not:
+% when you see `dbl', think ``full page width''.
+% \changes{4.1b}{2008/08/04}{Change \cs{set@colroom} to \cs{set@colht}}
+% \begin{macrocode}
+\mathchardef\do@startcolumn@pen=10005
+\@namedef{output@-\the\do@startcolumn@pen}{\do@startcolumn}%
+\def\do@startcolumn{%
+ \setbox\@cclv\vbox{\unvbox\@cclv\remove@lastbox\unskip}%
+ \clearpage@sw{\@clearfloatplacement}{\@floatplacement}%
+ \set@colht
+ \@booleanfalse\pfloat@avail@sw
+ \begingroup
+ \@colht\@colroom
+ \@booleanfalse\float@avail@sw
+ \@tryfcolumn\test@colfloat
+ \float@avail@sw{\aftergroup\@booleantrue\aftergroup\pfloat@avail@sw}{}%
+ \endgroup
+ \fcolmade@sw{%
+ \setbox\@cclv\vbox{\unvbox\@outputbox\unvbox\@cclv}%
+% \end{macrocode}
+% Now ask for a return visit, this time with insertions and all.
+% \begin{macrocode}
+ \outputpenalty-\pagebreak@pen
+ \dead@cycle
+ }{%
+ \begingroup
+ \let\@elt\@scolelt
+ \let\reserved@b\@deferlist\global\let\@deferlist\@empty\reserved@b
+ \endgroup
+ \clearpage@sw{%
+ \outputpenalty\@M
+ }{%
+ \outputpenalty\do@newpage@pen
+ }%
+ \dead@cycle
+ }%
+ \check@deferlist@stuck\do@startcolumn
+ \set@vsize
+}%
+\def\@scolelt#1{\def\@currbox{#1}\@addtonextcol}%
+\def\test@colfloat#1{%
+ \csname @floatselect@sw@\thepagegrid\endcsname#1{}{\@testtrue}%
+ \@if@sw\if@test\fi{}{\aftergroup\@booleantrue\aftergroup\float@avail@sw}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@addtonextcol}
+% We must adjust \cmd\@addtonextcol\ to take held-over inserts into account.
+% Now that all deferred floats are queued up together (in order), we must have a way of
+% differentiating them; this is done by the page grid-dependent procedure \cmd\@floatselect@sw@.
+% \begin{macrocode}
+\def\@addtonextcol{%
+ \begingroup
+ \@insertfalse
+ \@setfloattypecounts
+ \csname @floatselect@sw@\thepagegrid\endcsname\@currbox{%
+ \@ifnum{\@fpstype=8 }{}{%
+ \@ifnum{\@fpstype=24 }{}{%
+ \@flsettextmin
+ \@reqcolroom \ht\@currbox
+ \advance \@reqcolroom \@textmin
+ \advance \@reqcolroom \vsize % take into account split insertions
+ \advance \@reqcolroom -\pagegoal
+ \@ifdim{\@colroom>\@reqcolroom}{%
+ \@flsetnum \@colnum
+ \@ifnum{\@colnum>\z@}{%
+ \@bitor\@currtype\@deferlist
+ \@if@sw\if@test\fi{}{%
+ \@addtotoporbot
+ }%
+ }{}%
+ }{}%
+ }%
+ }%
+ }{}%
+ \@if@sw\if@insert\fi{}{%
+ \@cons\@deferlist\@currbox
+ }%
+ \endgroup
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\do@startpage@pen}
+% \begin{macro}{\forcefloats@sw}
+% \begin{macro}{\@output@combined@page}
+% \begin{macro}{\@sdblcolelt}
+% \begin{macro}{\test@dblfloat}
+% \begin{macro}{\@if@notdblfloat}
+% Similar to \cmd\do@startcolumn,
+% the procedure \cmd\do@startpage\ starts up a new page (not column) in a multi-column page grid.
+% It is invoked after a page is shipped out in a multi-column page grid, and
+% it commits full-page-width floats to the fresh page, possibly resulting in a float page.
+% In implementation, it is similar to \cmd\do@startcolumn, except that
+% it commits effectively via \cmd\@addtodblcol\ instead of \cmd\@addtonextcol.
+% Note that this procedure will inevitably be followed by \cmd\do@startcolumn.
+%
+% Some details of the procedure:
+%
+% We begin by removing the protection box from \cmd\box\cmd\@cclv, then setting the values of the
+% float placement parameters appropriately, and resetting \cmd\@colht, \cmd\@colroom, and \cmd\vsize\ to base values.
+%
+% Next we attempt to compose a float page, a page consisting entirely of floats. If successful,
+% we ship out the float page and lay down an interrupt that will send us back here for another try.
+%
+% If no float page is formed, we attempt to commit full-page-width floats to the text page, and return with a dead cycle.
+% We are now ready to compose columns of text.
+%
+% Note that all floats (both column floats and full-page-width floats) move through a single queue.
+% To differentiate between the two, the width of the float is compared to \cmd\textwidth.
+% This comparison is encapsulated in the macro \cmd\@if@notdblfloat, which should be used whenever
+% such a determination must be made. This procedure returns a Boolean.
+%
+% \changes{4.1b}{2008/08/04}{Change \cs{set@colroom} to \cs{set@colht}}
+% \begin{macrocode}
+\mathchardef\do@startpage@pen=10006
+\@namedef{output@-\the\do@startpage@pen}{\do@startpage}%
+\def\do@startpage{%
+ \setbox\@cclv\vbox{\unvbox\@cclv\remove@lastbox\unskip}%
+ \clearpage@sw{\@clearfloatplacement}{\@dblfloatplacement}%
+ \set@colht
+ \@booleanfalse\pfloat@avail@sw
+ \begingroup
+ \@booleanfalse\float@avail@sw
+ \@tryfcolumn\test@dblfloat
+ \float@avail@sw{\aftergroup\@booleantrue\aftergroup\pfloat@avail@sw}{}%
+ \endgroup
+ \fcolmade@sw{%
+ \global\setbox\pagesofar\vbox{\unvbox\pagesofar\unvbox\@outputbox}%
+ \@output@combined@page
+ }{%
+ \begingroup
+ \@booleanfalse\float@avail@sw
+ \let\@elt\@sdblcolelt
+ \let\reserved@b\@deferlist\global\let\@deferlist\@empty\reserved@b
+ \endgroup
+ \@ifdim{\@colht=\textheight}{% No luck...
+ \pfloat@avail@sw{% ...but a float *was* available!
+ \forcefloats@sw{%
+ \ltxgrid@warn{Forced dequeueing of floats stalled}%
+ }{%
+ \ltxgrid@warn{Dequeueing of floats stalled}%
+ }%
+ }{}%
+ }{}%
+ \outputpenalty\@M
+ \dead@cycle
+ }%
+ \check@deferlist@stuck\do@startpage
+ \set@colht
+}%
+% \end{macrocode}
+%
+% Procedure \cmd\@output@combined@page\ is a utility that ships out a page consisting of the
+% result of \cmd\@combinepage\ and \cmd\@combinedblfloats, after which it prepares for
+% the process to repeat.
+%
+% It is coincidentally identical to what needs to happen with
+% a float page that has been built by \cmd\@tryfcolumn, in the multi-column page grid,
+% and also handles the case where a page needs to be shipped out when in multicolumn mode.
+% \changes{4.1b}{2008/08/04}{New procedure \cs{@output@combined@page}}
+% \changes{4.1b}{2008/08/04}{Change \cs{@combinepage} to \cs{@combinepage} with argument}
+% \begin{macrocode}
+\def\@output@combined@page{%
+ \@combinepage\true@sw
+ \@combinedblfloats
+ \@outputpage
+ \global\pagegrid@cur\@ne
+ \protect@penalty\do@startpage@pen
+}%
+\def\@sdblcolelt#1{\def\@currbox{#1}\@addtodblcol}%
+\def\test@dblfloat#1{%
+ \@if@notdblfloat{#1}{\@testtrue}{}%
+ \@if@sw\if@test\fi{}{\aftergroup\@booleantrue\aftergroup\float@avail@sw}%
+}%
+\def\@if@notdblfloat#1{\@ifdim{\wd#1<\textwidth}}%
+\@booleanfalse\forcefloats@sw
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\@addtodblcol}
+%
+% The procedure \cmd\@addtodblcol\ is called into play at the beginning of each fresh page
+% and operates on each deferred float, in the hopes of placing one or more such floats
+% at the top of the current page.
+%
+% We alter the procedure of standard \LaTeX\ by putting failed floats into
+% \cmd\@deferlist\ instead of \cmd\@dbldeferlist. Having done so, we must have a means
+% of differentiating full-page-width floats from column-width floats.
+% We assume that the latter will always be narrower than \cmd\textwidth.
+%
+% In aid of detecting a stalled float flushing process, we set a Boolean if we encounter
+% a qualified full-page-width float here. Any that qualify but fail the rest of the tests
+% might still pass when reconsidered on an otherwise blank page.
+% \begin{macrocode}
+\def\@addtodblcol{%
+ \begingroup
+ \@if@notdblfloat{\@currbox}{%
+ \false@sw
+ }{%
+ \@setfloattypecounts
+ \@getfpsbit \tw@
+ \@bitor \@currtype \@deferlist
+ \@if@sw\if@test\fi{%
+ \false@sw
+ }{%
+ \@ifodd\@tempcnta{%
+ \aftergroup\@booleantrue\aftergroup\float@avail@sw
+ \@flsetnum \@dbltopnum
+ \@ifnum{\@dbltopnum>\z@}{%
+ \@ifdim{\@dbltoproom>\ht\@currbox}{%
+ \true@sw
+ }{%
+ \@ifnum{\@fpstype<\sixt@@n}{%
+ \begingroup
+ \advance \@dbltoproom \@textmin
+ \@ifdim{\@dbltoproom>\ht\@currbox}{%
+ \endgroup\true@sw
+ }{%
+ \endgroup\false@sw
+ }%
+ }{%
+ \false@sw
+ }%
+ }%
+ }{%
+ \false@sw
+ }%
+ }{%
+ \false@sw
+ }%
+ }%
+ }%
+ {%
+ \@tempdima -\ht\@currbox
+ \advance\@tempdima
+ -\@ifx{\@dbltoplist\@empty}{\dbltextfloatsep}{\dblfloatsep}%
+ \global \advance \@dbltoproom \@tempdima
+ \global \advance \@colht \@tempdima
+ \global \advance \@dbltopnum \m@ne
+ \@cons \@dbltoplist \@currbox
+ }{%
+ \@cons \@deferlist \@currbox
+ }%
+ \endgroup
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@tryfcolumn}
+% \begin{macro}{\@wtryfc}
+% \begin{macro}{\@xtryfc}
+% \begin{macro}{\@ztryfc}
+% Whenever a page is shipped out, \LaTeX\ automatically tries out a float column:
+% a page containing nothing but floats (and, as we have added here, split footnotes).
+%
+% The following four procedures employ certain macros to communicate between each other:
+%
+% \cmd\fcolmade@sw, a boolean, says whether we were successful in making a float column.
+%
+% \cmd\if@test, a \cmd\newif\ switch, says a float has failed some test.
+%
+% \cmd\@deferlist, is the input to the process, a list, of deferred floats.
+%
+% \cmd\@trylist, a list, stores the deferred floats to be tried out on the float column.
+%
+% \cmd\@failedlist, a list of floats that have failed the selection for the float column.
+%
+% \cmd\@flfail, a list of floats that have failed the second selection for the float column.
+%
+% \cmd\@flsucceed, a list, the floats that have been successfully placed on the float column.
+%
+% \cmd\@freelist, a list, receives any freed floats.
+%
+% \cmd\@colht, a dimen, the available space for the column, including column floats and insertions (footnotes).
+%
+% \cmd\@fpmin, a dimen, the required minimum height for the float column.
+%
+% \cmd\@outputbox, a box, the output of the process.
+%
+% \cmd\@fptop, \cmd\@fpsep, \cmd\@fpbot, glue, placed above, between, and below floats on the float column.
+%
+% \cmd\@currtype, a count, used temporarily for the float's bits.
+%
+% \cmd\@tempcnta, a count, used temporarily for the float's bits.
+%
+% In \cmd\@tryfcolumn, we alter the criterion for a float page, because if footnotes are present at this point
+% (presumably due to a split insertion) then \cmd\@fpmin is no longer the right threshold to apply.
+%
+% Note that we have changed \cmd\@tryfcolumn, \cmd\@xtryfc, and \cmd\@ztryfc\ syntactically so that the procedure
+% to test for the float's being a column float versus a full-page-width float is passed in as an
+% argument.
+%
+% \begin{macrocode}
+\def\@tryfcolumn#1{%
+ \global\@booleanfalse\fcolmade@sw
+ \@ifx@empty\@deferlist{}{%
+ \global\let\@trylist\@deferlist
+ \global\let\@failedlist\@empty
+ \begingroup
+ \dimen@\vsize\advance\dimen@-\pagegoal\@ifdim{\dimen@>\z@}{%
+ \advance\@fpmin-\dimen@
+ }{}%
+ \def\@elt{\@xtryfc#1}\@trylist
+ \endgroup
+ \fcolmade@sw{%
+ \global\setbox\@outputbox\vbox{\vskip \@fptop}%
+ \let \@elt \@wtryfc \@flsucceed
+ \global\setbox\@outputbox\vbox{\unvbox\@outputbox
+ \unskip \vskip \@fpbot
+ }%
+ \let \@elt \relax
+ \xdef\@deferlist{\@failedlist\@flfail}%
+ \xdef\@freelist{\@freelist\@flsucceed}%
+ }{}%
+ }%
+}%
+\def\@wtryfc #1{%
+ \global\setbox\@outputbox\vbox{\unvbox\@outputbox
+ \box #1\vskip\@fpsep
+ }%
+}%
+\def\@xtryfc#1#2{%
+ \@next\reserved@a\@trylist{}{}% trim \@trylist. Ugly!
+ \@currtype \count #2%
+ \divide\@currtype\@xxxii\multiply\@currtype\@xxxii
+ \@bitor \@currtype \@failedlist
+ \@testfp #2%
+ #1#2%
+ \@ifdim{\ht #2>\@colht }{\@testtrue}{}%
+ \@if@sw\if@test\fi{%
+ \@cons\@failedlist #2%
+ }{%
+ \begingroup
+ \gdef\@flsucceed{\@elt #2}%
+ \global\let\@flfail\@empty
+ \@tempdima\ht #2%
+ \def \@elt {\@ztryfc#1}\@trylist
+ \@ifdim{\@tempdima >\@fpmin}{%
+ \global\@booleantrue\fcolmade@sw
+ }{%
+ \@cons\@failedlist #2%
+ }%
+ \endgroup
+ \fcolmade@sw{%
+ \let \@elt \@gobble
+ }{}%
+ }%
+}%
+\def\@ztryfc #1#2{%
+ \@tempcnta \count#2%
+ \divide\@tempcnta\@xxxii\multiply\@tempcnta\@xxxii
+ \@bitor \@tempcnta {\@failedlist \@flfail}%
+ \@testfp #2%
+ #1#2%
+ \@tempdimb\@tempdima
+ \advance\@tempdimb \ht#2\advance\@tempdimb\@fpsep
+ \@ifdim{\@tempdimb >\@colht}{%
+ \@testtrue
+ }{}%
+ \@if@sw\if@test\fi{%
+ \@cons\@flfail #2%
+ }{%
+ \@cons\@flsucceed #2%
+ \@tempdima\@tempdimb
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+%
+% \subsection{Clearing pages}%
+%
+% Clearing the page is an elaboration of ending the page: it entails flushing all floats.
+%
+% This package might make number of float flushing algorithms available,
+% a very simple one that does not try to produce excellent pages,
+% another that tries to make the best use of space,
+% and a more complex one that tries to balance columns.
+%
+% At the beginning of the page-clearing process, by definition all of the paragraph text involved is on the MVL and all floats have been encountered.
+% There may be material in \cmd\pagesofar, and (in a multi-column page grid) any number of columns of the page have been composed.
+% Also, there might be footnote material saved up in \cmd\footsofar.
+%
+% Because we did not want to perform multiple \cmd\shipout s per visit to the output routine,
+% our multi-column page makeup will not compose multiple columns per visit.
+% This implementation detail may not require alteration, but it is not a limitation that is truly necessary:
+% it is only multiple \cmd\shipout s per visit that must be avoided.
+%
+% The crux matter is how to continue with flushing floats even after the material in the MVL is exhausted.
+% At that point, we must, upon completion of the output routine,
+% insert into the MVL an interrupt that triggers the next step in the processing.
+%
+% Therefore, after processing a \cmd\do@startcolumn\ interrupt, we must somehow force the completion of that column.
+% This could be done by inserting a \cmd\do@newpage@pen\ interrupt.
+%
+% And after processing a \cmd\do@startpage@pen\ interrupt, that results in \cmd\@dbltopinsert s,
+% we must ensure that the multiple columns on the page get completed, so that the page itself finally gets shipped out.
+% This part will proceed automatically given that \cmd\do@startcolumn\ processing completes successfully.
+%
+% The process will not be complete until all deferred floats have been placed and shipped out, and all saved-up footnotes have been inserted.
+%
+% Full-page-width floats can get out of order of column floats. This problem can be remedied by holding them all in the same list.
+% We therefore stop using \cmd\@dbldeferlist\ entirely, and all of the procedures that formerly used it have been rewritten to
+% use \cmd\@deferlist\ instead. When traversing the list, we apply a selector on the given box that determines whether it is a column-width or page-width float.
+% This selector is different depending on the page grid.
+%
+% When the \cmd\@deferlist\ is processed (by any means), we have to take care of the case where a float of one category is passed over but we are looking for a float of the other category.
+% Here, we must terminate processing, to avoid disordering the floats. This we do by the usual means.
+%
+% The system has a Boolean that says we are clearing pages: \cmd\clearpage@sw; if it is true,
+% then at the tail of \cmd\do@startcolumn\ processing, we should put down a (\cmd\vfil?) \cmd\do@newpage@pen\ interrupt.
+% This is because the MVL is now empty, so we have to force the columns to complete.
+%
+% One potential very pathological case would be where there is one or more deferred floats that never successfully get placed:
+% placing floats has stalled, and we will ship out blank pages indefinitely. How to detect this case?
+%
+% First, \cmd\do@startpage\ will evidently be stalled if the following are all true:
+% a) \cmd\@tryfcolumn\ and \cmd\@sdblcolelt\ both fail,
+% b) there are deferred floats available for page placement, and
+% c) the \cmd\@colht=\cmd\textheight, that is, the full page height is available for placement of column floats.
+%
+% Second, \cmd\do@startcolumn\ will evidently be stalled if the following are all true:
+% a) tryfcolumn fails,
+% b) there are deferred floats available for column placement, and
+% a) the \cmd\@colroom=\cmd\textheight, that is, the full page height is available for placement of column floats.
+%
+%
+% \begin{macro}{\cleardoublepage}
+% \begin{macro}{\clearpage}
+% \begin{macro}{\newpage}
+% \begin{macro}{\newpage@prep}
+% The function of \cmd\clearpage\ is to end the current page with \cmd\newpage\ and then
+% ship out additional pages until (\footins) inserts and (deferred) floats are exhausted.
+%
+% The method involves setting the float placement parameters to completely permissive values
+% and kicking out the current page (using a non-discardable penalty).
+% A possibly short page will be shipped out, followed
+% by any number of float pages. However these float pages, because using permissive float placement,
+% will exhaust all inserts and deferred floats.
+%
+% Bug Note: in the code for \cmd\clearpage, the first penalty we output is an unprotected \cmd\pagebreak@pen.
+% I tried using a protected \cmd\do@newpage@pen, but that gave rise to a corner case where a blank page
+% was output.
+%
+% At present, the \cmd\clearpage\ procedure does the same as \cmd\newpage,
+% except that \cmd\clearpage@sw\ is turned on,
+% and the (discardable) \cmd\newpage\ is inevitably followed by the same procedures
+% that are executed if a page is shipped out.
+%
+% FIXME: it seems that better than \cmd\pagebreak@pen\ would be an unprotected penalty of a special value that would
+% entail output routine processing consisting of the following steps:
+% 3) \cmd\unvbox\cmd\@cclv,
+% 1) set \cmd\clearpage@sw\ to \cmd\true@sw,
+% 2) put down a protected \cmd\do@startcolumn@pen,
+% 4) take a dead cycle.
+%
+% The effect would be to liberalize float placement options for the current column as well as further columns that may be output as part of \cmd\clearpage\ processing.
+% Of course, it would still be necessary to set \cmd\clearpage@sw\ again via an interrupt.
+%
+% An optimization might be to clear \cmd\clearpage@sw\ as part of the same interrupt,
+% but that would actually not work properly, because it is necessary for \cmd\do@endpage\ to
+% possibly invoke further visits to the output routine before clearpage processing ceases.
+% \begin{macrocode}
+\def\newpage@prep{%
+ \if@noskipsec
+ \ifx \@nodocument\relax
+ \leavevmode
+ \global \@noskipsecfalse
+ \fi
+ \fi
+ \if@inlabel
+ \leavevmode
+ \global \@inlabelfalse
+ \fi
+ \if@nobreak \@nobreakfalse \everypar{}\fi
+ \par
+}%
+\def \newpage {%
+ \newpage@prep
+ \do@output@MVL{%
+ \vfil
+ \penalty-\pagebreak@pen
+ }%
+}%
+\def\clearpage{%
+ \newpage@prep
+ \do@output@MVL{%
+ \vfil
+ \penalty-\pagebreak@pen
+ \global\@booleantrue\clearpage@sw
+ \protect@penalty\do@startcolumn@pen
+ \protect@penalty\do@endpage@pen
+ }%
+ \do@output@MVL{%
+ \global\@booleanfalse\clearpage@sw
+ }%
+}%
+\def\cleardoublepage{%
+ \clearpage
+ \@if@sw\if@twoside\fi{%
+ \@ifodd\c@page{}{%
+ \null\clearpage
+ }%
+ }{}%
+}%
+\@booleanfalse\clearpage@sw
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\do@endpage@pen}
+% The penalty \cmd\do@endpage@pen\ simply dispatches to the page grid procedure that forces an end page.
+% That procedure should test whether there is anything to ship out (say committed floats), then act accordingly.
+% Note that as part of this work, it should \cmd\unvbox\cmd\@cclv, which has been left boxed up so it can be
+% measured.
+% \begin{macrocode}
+\mathchardef\do@endpage@pen=10007
+\@namedef{output@-\the\do@endpage@pen}{\csname end@column@\thepagegrid\endcsname}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\do@newpage@pen}
+% The penalty \cmd\do@newpage@pen\ allows a ``non-discardable \cmd\newpage'' command:
+% a \cmd\newpage\ command that will not disappear at a pagebreak.
+% This visit to the output routine will not be dispatched to an interrupt,
+% rather the natural output routine will be executed, where it
+% will remove the protection box.
+%
+% Call this routine by executing \cmd\protect@penalty\cmd\do@newpage@pen.
+% \begin{macrocode}
+\mathchardef\do@newpage@pen=10001
+\expandafter\let\csname output@-\the\do@newpage@pen\endcsname\relax
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@clearfloatplacement}
+% The procedure \cmd\@clearfloatplacement\ sets the float placement parameters
+% to completely permissive values (except for \cmd\@fpmin). The standard values are:
+%
+% \begin{tabular}{ll}
+%\cmd\@topnum &\cmd\c@topnumber\\
+%\cmd\@toproom &\cmd\topfraction\cmd\@colht\\
+%\cmd\@botnum &\cmd\c@bottomnumber\\
+%\cmd\@botroom &\cmd\bottomfraction\cmd\@colht\\
+%\cmd\@colnum &\cmd\c@totalnumber\\
+%\cmd\@fpmin &\cmd\floatpagefraction\cmd\@colht\\
+%\cmd\@dbltopnum &\cmd\c@dbltopnumber\\
+%\cmd\@dbltoproom &\cmd\dbltopfraction\cmd\@colht\\
+%\cmd\@textmin &\cmd\@colht\cmd\advance\cmd\@textmin -\cmd\@dbltoproom\\
+%\cmd\@fpmin &\cmd\dblfloatpagefraction\cmd\textheight
+% \end{tabular}
+% \begin{macrocode}
+\def\@clearfloatplacement{%
+ \global\@topnum \maxdimen
+ \global\@toproom \maxdimen
+ \global\@botnum \maxdimen
+ \global\@botroom \maxdimen
+ \global\@colnum \maxdimen
+ \global\@dbltopnum \maxdimen
+ \global\@dbltoproom \maxdimen
+ \global\@textmin \z@
+ \global\@fpmin \z@
+ \let\@testfp\@gobble
+ \appdef\@setfloattypecounts{\@fpstype16\advance\@fpstype\m@ne}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@doclearpage}
+% \begin{macro}{\@makefcolumn}
+% \begin{macro}{\@makecol}
+% The \cmd\@doclearpage\ procedure is now obsoleted, as is \cmd\@makefcolumn, which it invoked.
+% We also completely avoid using \cmd\@makecol\ (in favor of \cmd\@makecolumn).
+% \begin{macrocode}
+\let\@doclearpage\@undefined
+\let\@makefcolumn\@undefined
+\let\@makecol\@undefined
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\clr@top@firstmark}
+% \begin{macro}{\set@top@firstmark}
+% \begin{macro}{\@outputpage@tail}
+% We want accurate values of \cmd\topmark\ and \cmd\firstmark, but we must deal with
+% the fact that there are many different ways of contributing material
+% to the page. Only upon the first contribution to the page is the value of \cmd\topmark\
+% accurate. However, with \cmd\firstmark\ we must potentially examine each contribution
+% because the first mark on the page may happen to fall in the last piece of material contributed.
+%
+% To begin, we define the procedure that initializes the macros to appropriate flag values.
+% \begin{macrocode}
+\def\clr@top@firstmark{%
+ \global\let\saved@@topmark\@undefined
+ \global\let\saved@@firstmark\@empty
+ \global\let\saved@@botmark\@empty
+}%
+\clr@top@firstmark
+% \end{macrocode}
+%
+% Note that the flag value for \cmd\saved@@topmark\ is \cmd\@undefined, just as one would
+% expect. But that for \cmd\saved@@firstmark\ and \cmd\saved@@botmark\ is \cmd\@empty.
+%
+% Next, we define procedure \cmd\set@top@firstmark; it will be exercised everywhere material is contributed,
+% capturing the mark values if appropriate.
+% \begin{macrocode}
+\def\set@top@firstmark{%
+ \@ifxundefined\saved@@topmark{\expandafter\gdef\expandafter\saved@@topmark\expandafter{\@@topmark}}{}%
+ \@if@empty\saved@@firstmark{\expandafter\gdef\expandafter\saved@@firstmark\expandafter{\@@firstmark}}{}%
+ \@if@empty\@@botmark{}{\expandafter\gdef\expandafter\saved@@botmark\expandafter{\@@botmark}}%
+}%
+% \end{macrocode}
+% When should \cmd\set@top@firstmark\ be called?
+% A good candidate for a universal procedure for handling contributed material is
+% the natural output routine; are any other calls needed?
+%
+% Yes, in \cmd\save@column\ we must execute \cmd\set@top@firstmark\ because we are about to
+% save away \cmd\box\cmd\@cclv, and we will never see its marks again (unless it is unboxed into the MVL),
+% because \TeX\ lets one access a box's marks only within an output routine that has put that box into \cmd\box\cmd\@cclv.
+%
+% As soon as a page is shipped out, we initialize the two macros that
+% hold the values of \cmd\topmark\ and \cmd\firstmark, respectively.
+% \changes{4.1b}{2008/08/04}{Procedure \cs{@outputpage@tail} tailpatches \cs{@outputpage}}%
+% \begin{macrocode}
+\appdef\@outputpage@tail{%
+ \clr@top@firstmark
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+%
+% \subsection{Other interfaces to \LaTeX}%
+%
+% \begin{macro}{\@float}
+% \begin{macro}{\@dblfloat}
+% \begin{macro}{\@yfloat}
+% \begin{macro}{\fps@}
+% \begin{macro}{\fpsd@}
+% The \LaTeX\ kernel procedures \cmd\@float\ and \cmd\@dblfloat\ are treated on an equal footing.
+% Each now takes environment-specific float placement defaults.
+% If none are defined for the calling environment, we apply a default.
+%
+% A parameter is passed that will set the width of text within the float, normally \cmd\columnwidth,
+% and in the "dbl" version, \cmd\textwidth. However, an environment such as \env{turnpage}
+% may change the meanings of these macros to allow turnpage floats.
+%
+% \changes{4.0a}{2008/06/26 }{\cs{@yfloat}: de-fang \cs{set@footnotewidth} (see ltxutil.dtx): we have already done its job. }
+%
+% Note on \cmd\@xfloat: the optional argument must come to it fully expanded, because the macro
+% does a weird procedure on this argument, involving \cmd\@onelevel@sanitize, which I do not
+% understand, and which does not work if not so expanded.
+% \changes{4.1b}{2008/08/04}{Get rid of the \cs{reserved@a} idiom}
+% \begin{macrocode}
+\def\@float#1{%
+ \@ifnextchar[{%
+% \end{macrocode}
+% \verb+}]{+Brace-matching klootch
+% \begin{macrocode}
+ \@yfloat\width@float{#1}%
+ }{%
+ \@ifxundefined@cs{fps@#1}{}{\expandafter\let\expandafter\fps@\csname fps@#1\endcsname}%
+ \expandafter\@argswap\expandafter{\expandafter[\fps@]}{\@yfloat\width@float{#1}}%
+ }%
+}%
+\def\@dblfloat#1{%
+ \@ifnum{\pagegrid@col=\@ne}{%
+ \@float{#1}%
+ }{%
+ \@ifnextchar[{%
+% \end{macrocode}
+% \verb+}]{+Brace-matching klootch
+% \begin{macrocode}
+ \@yfloat\widthd@float{#1}%
+ }{%
+ \@ifxundefined@cs{fpsd@#1}{}{\expandafter\let\expandafter\fpsd@\csname fpsd@#1\endcsname}%
+ \expandafter\@argswap\expandafter{\expandafter[\fpsd@]}{\@yfloat\widthd@float{#1}}%
+ }%
+ }%
+}%
+% \end{macrocode}
+% \changes{4.1b}{2008/08/04}{Turn off the \cs{set@footnotewidth} mechanism; a float `knows' its proper typesetting context}
+% \changes{4.1n}{2010/01/02}{(AO, 571) Interface \cs{set@footnotewidth} for determining the set width of footnotes}%
+% \cmd\@yfloat\ is the go-to procdure for creating the proper environment for the content of a float.
+% Argument \verb+#1+ is the width of the float environment (we disable \cmd\set@footnotewidth), and
+% we establish a self-contained (minipage) environment for footnotes.
+% \begin{macrocode}
+\def\@yfloat#1#2[#3]{%
+ \@xfloat{#2}[#3]%
+ \hsize#1\linewidth\hsize
+ \let\set@footnotewidth\@empty
+ \minipagefootnote@init
+}%
+\def\fps@{tbp}%
+\def\fpsd@{tp}%
+\def\width@float{\columnwidth}%
+\def\widthd@float{\textwidth}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\end@float}
+% \begin{macro}{\end@dblfloat}
+% \begin{macro}{\end@@float}
+% \begin{macro}{\check@currbox@count}
+% \begin{macro}{\minipagefootnote@init}
+% \begin{macro}{\minipagefootnote@here}
+% \LaTeX\ kernel procedures \cmd\end@float\ and \cmd\end@dblfloat\
+% have been changed to work alike; in particular, floats of both classes
+% are deferred into the same queue.
+% This measure ensures that they will be placed in their original order,
+% an aspect in which \LaTeX\ is broken.
+%
+% Note: when retrieving floats from the queues, we can differentiate those of the two categories
+% by the width of the box.
+%
+% Floats are processed via an output routine message, and are checked for
+% sanity in re the float placement options. In the case of full-page-width floats,
+% we ensure that the h and b float placement options are never asserted, because they
+% make no sense.
+%
+% Note that if we get to the end of the float box and still have pending
+% footnotes, we put then out.
+%
+% LaTeX Bug note: if a user types \cmd\begin{table*}[h], the float will never succeed in being placed!
+% we try to catch such cases.
+%
+% Note that the macro \cmd\check@currbox@count\ tries to catch cases where the float placement options
+% are such that the float can never be placed.
+%
+% The calls to \cmd\@iffpsbit\ are part of a procedure to deny certain of the float placement
+% parameters: ``h'' and ``b'' are not possible, the former because the \cmd\marginpar\ mechanism
+% cannot place a full-page-width float within a multicolumn page grid, the latter because nobody
+% has yet written the code to do so (pretty bad reason, I know).
+% \changes{4.1b}{2008/08/04}{New procedure \cs{@iffpsbit} replaces \cs{@getfpsbit}}
+% \begin{macrocode}
+\def\end@float{%
+ \end@@float{%
+ \check@currbox@count
+ }%
+}%
+\def\end@dblfloat{%
+ \@ifnum{\pagegrid@col=\@ne}{%
+ \end@float
+ }{%
+ \end@@float{%
+ \@iffpsbit\@ne{\global\advance\count\@currbox\m@ne}{}%
+ \@iffpsbit\f@ur{\global\advance\count\@currbox-4\relax}{}%
+ \global\wd\@currbox\textwidth % Klootch
+ \check@currbox@count
+ }%
+ }%
+}%
+\def\end@@float#1{%
+ \minipagefootnote@here
+ \@endfloatbox
+ #1%
+ \@ifnum{\@floatpenalty <\z@}{%
+ \@largefloatcheck
+ \@cons\@currlist\@currbox
+ \@ifnum{\@floatpenalty <-\@Mii}{%
+ \do@output@cclv{\@add@float}%
+ }{%
+ \vadjust{\do@output@cclv{\@add@float}}%
+ \@Esphack
+ }%
+ }{}%
+}%
+% \end{macrocode}
+% The \classname{float} package of Anselm Lingnau fails when used under \classname{ltxgrid}, but
+% we can fix things.
+% We also repair a bug in that package.
+%
+% \changes{4.1g}{2009/10/06}{(AO, 531) Fix package \classname{float} }
+% \begin{macrocode}
+\newcommand\float@end@float{%
+ \@endfloatbox
+ \global\setbox\@currbox\float@makebox\columnwidth
+ \let\@endfloatbox\relax
+ \end@float
+}%
+\newcommand\float@end@ltx{%
+ \end@@float{%
+ \global\setbox\@currbox\float@makebox\columnwidth
+ \check@currbox@count
+ }%
+}%
+\newcommand\newfloat@float[3]{%
+ \@namedef{ext@#1}{#3} %!
+ \let\float@do=\relax
+ \xdef\@tempa{\noexpand\float@exts{\the\float@exts \float@do{#3}}}%
+ \@tempa
+ \floatplacement{#1}{#2}%
+ \@ifundefined{fname@#1}{\floatname{#1}{#1}}{} %!
+ \expandafter\edef\csname ftype@#1\endcsname{\value{float@type}}%
+ \addtocounter{float@type}{\value{float@type}} %!
+ \restylefloat{#1}%
+ \expandafter\edef\csname fnum@#1\endcsname{%
+ \expandafter\noexpand\csname fname@#1\endcsname{} %!
+ \expandafter\noexpand\csname the#1\endcsname
+ }
+ \@ifnextchar[%]
+ {%
+ \float@newx{#1}%
+ }{%
+ \@ifundefined{c@#1}{\newcounter{#1}\@namedef{the#1}{\arabic{#1}}}{}%
+ }%
+}%
+\newcommand\newfloat@ltx[3]{%
+ \@namedef{ext@#1}{#3}%
+ \let\float@do=\relax
+ \xdef\@tempa{\noexpand\float@exts{\the\float@exts \float@do{#3}}}%
+ \@tempa
+ \floatplacement{#1}{#2}%
+ \@ifundefined{fname@#1}{\floatname{#1}{#1}}{}%
+ \expandafter\edef\csname ftype@#1\expandafter\endcsname\expandafter{\the\c@float@type}%
+ \addtocounter{float@type}{\value{float@type}}%
+ \restylefloat{#1}%
+ \expandafter\edef\csname fnum@#1\endcsname{%
+ \expandafter\noexpand\csname fname@#1\endcsname{}%
+ \expandafter\noexpand\csname the#1\endcsname
+ }
+ \@ifnextchar[%]
+ {%
+ \float@newx{#1}%
+ }{%
+ \@ifundefined{c@#1}{\newcounter{#1}\@namedef{the#1}{\arabic{#1}}}{}%
+ }%
+}%
+\appdef\document@inithook{%
+ \@ifxundefined\newfloat{}{%
+ \@ifx{\float@end\float@end@float}{%
+ \@ifx{\newfloat\newfloat@float}{\true@sw}{\false@sw}%
+ }{\false@sw}%
+ {%
+ \class@warn{Repair the float package}%
+ \let\float@end\float@end@ltx
+ \let\newfloat\newfloat@ltx
+ }{%
+ \class@warn{Failed to patch the float package}%
+ }%
+ }%
+}%
+% \end{macrocode}
+%
+% Boolean procedure \cmd\@iffpsbit\ is similar to the \cmd\@getfpsbit\ of \LaTeX, except
+% that we do not expose the scratch count register or even change its value.
+% \begin{macrocode}
+\def\@iffpsbit#1{%
+ \begingroup
+ \@tempcnta\count\@currbox
+ \divide\@tempcnta#1\relax
+ \@ifodd\@tempcnta{\aftergroup\true@sw}{\aftergroup\false@sw}%
+ \endgroup
+}%
+% \end{macrocode}
+%
+% In procedure \cmd\check@currbox@count, we calculate the net float placement
+% directive (encoded into \cmd\count\ \cmd\@currbox's least significant four bits).
+% If zero, issue a warning.
+% \changes{4.1b}{2008/08/04}{Tally the height of the float}
+% \begin{macrocode}
+\def\check@currbox@count{%
+ \@ifnum{\count\@currbox>\z@}{%
+ \count@\count\@currbox\divide\count@\sixt@@n\multiply\count@\sixt@@n
+ \@tempcnta\count\@currbox\advance\@tempcnta-\count@
+ \@ifnum{\@tempcnta=\z@}{%
+ \ltxgrid@warn{Float cannot be placed}%
+ }{}%
+ \expandafter\tally@float\expandafter{\@captype}%
+ }{%
+% \end{macrocode}
+% In this case, the float is a \cmd\marginpar.
+% \begin{macrocode}
+ }%
+}%
+\providecommand\minipagefootnote@init{}%
+\providecommand\minipagefootnote@here{}%
+\providecommand\tally@float[1]{}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\@specialoutput}
+% The \cmd\@add@float\ procedure used to reside in standard \LaTeX's \cmd\@specialoutput,
+% which is no more.
+%
+% Historical Note: \cmd\@specialoutput\ and Lamport's method of an output routine dispatcher
+% is the genesis of our more powerful and refined way of using \TeX's output routine to
+% safely accomplish page makeup tasks. To it and to him we owe acknowledgement and thanks.
+% \begin{macrocode}
+\let\@specialoutput\@undefined
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@add@float}
+% In the following, we do not need to execute \cmd\@reinserts, which was wrong anyway, as you cannot
+% reliably recover insertions when they split (unless you have a way of reinserting the captured insertion
+% ahead of the split-off part).
+%
+% Now that full-page-width floats are being processed the same as column floats, we
+% have to nip in here and cause them always to be deferred.
+%
+% At the very end, the \cmd\vsize\ is adjusted for any newly committed float.
+% \begin{macrocode}
+\def\@add@float{%
+ \@pageht\ht\@cclv\@pagedp\dp\@cclv
+ \unvbox\@cclv
+ \@next\@currbox\@currlist{%
+ \csname @floatselect@sw@\thepagegrid\endcsname\@currbox{%
+ \@ifnum{\count\@currbox>\z@}{%
+ \advance \@pageht \@pagedp
+% \end{macrocode}
+% Do not assume \cmd\holdinginserts is cleared:
+% \begin{macrocode}
+ \advance \@pageht \vsize \advance \@pageht -\pagegoal
+% \end{macrocode}
+% Commit an `h' float:
+% \begin{macrocode}
+ \@addtocurcol
+ }{%
+ \@addmarginpar
+ }%
+ }{%
+ \@resethfps
+ \@cons\@deferlist\@currbox
+ }%
+ }{\@latexbug}%
+ \@ifnum{\outputpenalty<\z@}{%
+ \@if@sw\if@nobreak\fi{%
+ \nobreak
+ }{%
+ \addpenalty \interlinepenalty
+ }%
+ }{}%
+ \set@vsize
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@reinserts}
+% The \cmd\@reinserts\ procedure of standard \LaTeX\ is now obsoleted (it had been erroneous anyway).
+% \begin{macrocode}
+\let\@reinserts\@undefined
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@addtocurcol}
+% We modify the \cmd\@addtocurcol\ procedure of standard \LaTeX\
+% so that a float placed ``here'' may break over pages.
+% \begin{macrocode}
+\def \@addtocurcol {%
+ \@insertfalse
+ \@setfloattypecounts
+ \ifnum \@fpstype=8
+ \else
+ \ifnum \@fpstype=24
+ \else
+ \@flsettextmin
+ \advance \@textmin \@textfloatsheight
+ \@reqcolroom \@pageht
+ \ifdim \@textmin>\@reqcolroom
+ \@reqcolroom \@textmin
+ \fi
+ \advance \@reqcolroom \ht\@currbox
+ \ifdim \@colroom>\@reqcolroom
+ \@flsetnum \@colnum
+ \ifnum \@colnum>\z@
+ \@bitor\@currtype\@deferlist
+ \if@test
+ \else
+ \@bitor\@currtype\@botlist
+ \if@test
+ \@addtobot
+ \else
+ \ifodd \count\@currbox
+ \advance \@reqcolroom \intextsep
+ \ifdim \@colroom>\@reqcolroom
+ \global \advance \@colnum \m@ne
+ \global \advance \@textfloatsheight \ht\@currbox
+ \global \advance \@textfloatsheight 2\intextsep
+ \@cons \@midlist \@currbox
+ \if@nobreak
+ \nobreak
+ \@nobreakfalse
+ \everypar{}%
+ \else
+ \addpenalty \interlinepenalty
+ \fi
+ \vskip \intextsep
+ \unvbox\@currbox %AO
+ \penalty\interlinepenalty
+ \vskip\intextsep
+ \ifnum\outputpenalty <-\@Mii \vskip -\parskip\fi
+ \outputpenalty \z@
+ \@inserttrue
+ \fi
+ \fi
+ \if@insert
+ \else
+ \@addtotoporbot
+ \fi
+ \fi
+ \fi
+ \fi
+ \fi
+ \fi
+ \fi
+ \if@insert
+ \else
+ \@resethfps
+ \@cons\@deferlist\@currbox
+ \fi
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\if@twocolumn}
+% The \cmd\newif\ switch \cmd\if@twocolumn\ is entirely unused. However its access words are invoked by
+% \LaTeX's \cmd\document\ procedure, so we de-fang it.
+% \begin{macrocode}
+\@twocolumnfalse
+\let\@twocolumntrue\@twocolumnfalse
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@addmarginpar}
+% The procedure \cmd\@addmarginpar\ used to access \cmd\if@twocolumn, but that switch is not reliable;
+% the better way is to use \cmd\thepagegrid. We establish a convention for a page-grid-oriented
+% procedure, e.g., \cmd\@addmarginpar@one, that emits a boolean, telling this procedure
+% whether to set the marginpar on the left or right.
+% \begin{macrocode}
+\def\@addmarginpar{%
+ \@next\@marbox\@currlist{%
+ \@cons\@freelist\@marbox\@cons\@freelist\@currbox
+ }\@latexbug
+ \setbox\@marbox\hb@xt@\columnwidth{%
+ \csname @addmarginpar@\thepagegrid\endcsname{%
+ \hskip-\marginparsep\hskip-\marginparwidth
+ \box\@currbox
+ }{%
+ \hskip\columnwidth\hskip\marginparsep
+ \box\@marbox
+ }%
+ \hss
+ }%
+ \setbox\z@\box\@currbox
+ \@tempdima\@mparbottom
+ \advance\@tempdima -\@pageht
+ \advance\@tempdima\ht\@marbox
+ \@ifdim{\@tempdima >\z@}{%
+ \@latex@warning@no@line {Marginpar on page \thepage\space moved}%
+ }{%
+ \@tempdima\z@
+ }%
+ \global\@mparbottom\@pageht
+ \global\advance\@mparbottom\@tempdima
+ \global\advance\@mparbottom\dp\@marbox
+ \global\advance\@mparbottom\marginparpush
+ \advance\@tempdima -\ht\@marbox
+ \global\setbox \@marbox
+ \vbox {\vskip \@tempdima
+ \box \@marbox}%
+ \global \ht\@marbox \z@
+ \global \dp\@marbox \z@
+ \kern -\@pagedp
+ \nointerlineskip
+ \box\@marbox
+ \nointerlineskip
+ \hbox{\vrule \@height\z@ \@width\z@ \@depth\@pagedp}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{environment}{turnpage}
+% Any float (viz., \env{figure} or \env{table}) within the scope of this environment
+% will be a turnpage float: It will be assumed to occupy an entire page (constitute a float page),
+% the width will be \cmd\textheight, the height \cmd\textwidth, and the entire float will be presented
+% rotated 90 degrees.
+%
+% The implementation requires the services of the \cmd\rotatebox\ command, so we supply a dummy definition
+% that explains things to the user.
+% \begin{macrocode}
+\newenvironment{turnpage}{%
+ \def\width@float{\textheight}%
+ \def\widthd@float{\textheight}%
+ \appdef\@endfloatbox{%
+ \@ifxundefined\@currbox{%
+ \ltxgrid@warn{Cannot rotate! Not a float}%
+ }{%
+ \setbox\@currbox\vbox to\textwidth{\vfil\unvbox\@currbox\vfil}%
+ \global\setbox\@currbox\vbox{\rotatebox{90}{\box\@currbox}}%
+ }%
+ }%
+}{%
+}%
+\def\rotatebox@dummy#1#2{%
+ \ltxgrid@warn{You must load the graphics or graphicx package in order to use the turnpage environment}%
+ #2%
+}%
+% \end{macrocode}
+% \changes{4.1b}{2008/08/04}{Use \cs{document@inithook} instead of \cs{AtBeginDocument}}
+% \begin{macrocode}
+\appdef\document@inithook{%
+ \@ifxundefined\rotatebox{\let\rotatebox\rotatebox@dummy}{}%
+}%
+% \end{macrocode}
+% \end{environment}
+%
+%
+% \subsection{One-off output routines}
+%
+% These procedures are executed in lieu of \cmd\the\cmd\output\ when the output penalty has the associated flag value.
+%
+% \begin{macro}{output@-1073741824}
+% The first one-off output routine handles the end of the job, wherein
+% \LaTeX\ executes \cmd\@@end, and breaks to the output with a penalty of
+% $"40000000 = 2^{32}/4 = 1073741824$.
+% We simply discard \cmd\box\cmd\@cclv\ and leave.
+% This means that \LaTeX\ is obligated to do \cmd\clearpage\ as part of
+% its \enve{document} processing, otherwise material will be lost.
+% \begin{macrocode}
+\@namedef{output@-1073741824}{%
+ \deadcycles\z@
+% \end{macrocode}
+% \begin{verbatim}
+%\showbox\@cclv
+% \end{verbatim}
+% \begin{macrocode}
+ \void@cclv
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\save@column@pen}
+% The one-off output routine associated with \cmd\penalty\cmd\save@column@pen\
+% will be called within a sequence of three such routines by \cmd\execute@message
+% or its companion routine \cmd\execute@message@insert.
+% This procedure must save away any the current page and preserve marks.
+% \begin{macrocode}
+\mathchardef\save@column@pen=10016
+\@namedef{output@-\the\save@column@pen}{\save@column}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@cclv@saved}
+% We take over the \cmd\@holdpg\ box register. Hereafter,
+% we no longer use the \cmd\@holdpg\ box register, so let the world know.
+% This should decisively break packages that assume standard \LaTeX.
+% Breaking decisively is preferred to quietly proceeding erroneously.
+% \begin{macrocode}
+\let \@cclv@saved \@holdpg
+\let \@holdpg \@undefined
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\save@column}
+% The procedure \cmd\save@column\ does the actual work of saving away the material
+% on the page. It is invoked both by \cmd\save@column@pen\ and by \cmd\save@column@insert@pen.
+% We save \cmd\box\cmd\@cclv\ and the primitive \cmd\@@topmark.
+% \begin{macrocode}
+\def\save@column{%
+ \@ifvoid\@cclv@saved{%
+ \set@top@firstmark
+ \global\@topmark@saved\expandafter{\@@topmark}%
+ }{}%
+ \global\setbox\@cclv@saved\vbox{%
+ \@ifvoid\@cclv@saved{}{%
+ \unvbox\@cclv@saved
+ \marry@baselines
+ }%
+ \unvbox\@cclv
+ \lose@breaks
+ \remove@lastbox
+ }%
+}%
+\newtoks\@topmark@saved
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\prep@cclv}
+% The procedure \cmd\prep@cclv\ is used by message handlers to set up their environment
+% to ape that of the usual output routine, with the boxed-up page in \cmd\box\cmd\@cclv.
+% Here, we retrieve the material from \cmd\@cclv@saved, where it was saved away by
+% the one-off output routine associated with \cmd\save@column@pen.
+% \begin{macrocode}
+\def\prep@cclv{%
+ \void@cclv
+ \setbox\@cclv\box\@cclv@saved
+ \vbadness\@M
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\save@column@insert@pen}
+% The one-off output routine associated with \cmd\penalty\cmd\save@column@insert@pen\
+% is similar to that of \cmd\save@column@pen\ augmented with the processing of insertions.
+% It is called by \cmd\execute@message@insert\ (i.e., at a grid change)
+% and saves away the current page and preserves marks.
+% In addition, it saves away any insertions that fall on the current page.
+% As with the natural output routine, it executes in two phases,
+% first with \cmd\holdinginserts\ set, then cleared.
+% \begin{macrocode}
+\mathchardef\save@column@insert@pen=10017
+\@namedef{output@-\the\save@column@insert@pen}{\toggle@insert{\savecolumn@holding}{\savecolumn@moving}}%
+% \end{macrocode}
+% The procedure \cmd\savecolumn@holding\ is the first phase of saving a column with its inserts.
+% This phase must detect and remedy the one circumstance that will confound our efforts to propagate marks.
+% It is similar to \cmd\output@holding, except that we have to deal with the protection box, which must
+% remain, because the messaging mechanism is being used.
+%
+% If it appears that we have the pathological ``Big Bad Box'' case at hand, we use the \cmd\dead@cycle@repair@protected\
+% procedure instead of \cmd\dead@cycle\ to do our dead cycle.
+% \changes{4.1b}{2008/08/04}{Use \cs{trace@box} instead of \cs{showbox}}
+% \begin{macrocode}
+\def\savecolumn@holding{%
+ \@if@exceed@pagegoal{\unvcopy\@cclv\remove@lastbox}{%
+ \setbox\z@\vbox{\unvcopy\@cclv\remove@lastbox}%
+ \outputdebug@sw{\trace@box\z@}{}%
+ \dimen@\ht\@cclv\advance\dimen@-\ht\z@
+ \dead@cycle@repair@protected\dimen@
+ }{%
+ \dead@cycle
+ }%
+}%
+% \end{macrocode}
+% \changes{4.1b}{2008/08/04}{New procedure for showing a box contents, \cs{trace@box}}
+% \changes{4.1n}{2009/12/02}{More diagnostics of column balancing}
+% The procedure \cmd\save@column@moving\ is the second phase of saving a column with its inserts.
+% Now that \cmd\holdinginserts\ is cleared, we can look in the various \cmd\insert\ registers for
+% our inserts (at present there is only one, \cmd\footins, along with \cmd\footins@saved).
+% if anything is there, we save it away and ask for another cycle (because it may have split).
+%
+% Note that the message that is about to be executed had better deal properly with the contents of
+% the \cmd\footins@saved\ box.
+%
+% \begin{macrocode}
+\def\savecolumn@moving{%
+ \ltxgrid@info@sw{\class@info{\string\savecolumn@moving}}{}%
+ \@cclv@nontrivial@sw{%
+ \save@column
+ }{%
+ \void@cclv
+ }%
+ \@ifvoid\footins{}{%
+ \ltxgrid@foot@info@sw{\class@info{\string\savecolumn@moving}\trace@scroll{\showbox\footins@saved\showbox\footins}}{}%
+% \end{macrocode}
+% Save all away in \cmd\footins@saved. Note that if \cmd\footins\ is void, then \cmd\footins@saved\ remains untouched.
+% \begin{macrocode}
+ \@ifvoid\footins@saved{%
+ \global\setbox\footins@saved\box\footins
+ }{%
+ \global\setbox\footins@saved\vbox\bgroup
+ \unvbox\footins@saved
+ \marry@baselines
+ \unvbox\footins
+ \egroup
+ }%
+ \ltxgrid@foot@info@sw{\trace@box\footins@saved}{}%
+ \protect@penalty\save@column@insert@pen
+ }%
+}%
+\newbox\footins@saved
+\newbox\footins@recovered
+\newbox\column@recovered
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\save@message@pen}
+% The one-off output routine associated with \cmd\penalty\cmd\save@message@pen\
+% saves away the message that has been passed.
+% This procedure is penultimate in a sequence of one-off output routine calls;
+% earlier ones have saved away the MVL and preserved marks, the last executes the message.
+%
+% Note that we are passing tokens to \TeX's primitive \cmd\mark\ mechanism, so we must ensure
+% that they are not inappropriately expanded. We use the same mechanism for all such cases,
+% namely \cmd\let@mark.
+%
+% Note: we expect that \cmd\box\cmd\@cclv's contents are well known:
+% \cmd\topskip, protection box, and a \cmd\mark, the latter containing the message.
+% But if we came here via \cmd\penalty 10017, there might be an \cmd\insert\ node present as well,
+% because a footnote may have split.
+% Because this procedure simply voids out \cmd\box\cmd\@cclv, such material would be lost.
+% Perhaps we can repair things by manipulating the \cmd\insert \ mechanism temporarily.
+% \begin{macrocode}
+\mathchardef\save@message@pen=10018
+\@namedef{output@-\the\save@message@pen}{\save@message}%
+\def\save@message{%
+ \void@cclv
+% \end{macrocode}
+%FIXME: what if \cmd\box\cmd\@cclv is not empty?
+% \begin{macrocode}
+ \toks@\expandafter{\@@firstmark}%
+ \expandafter\gdef\expandafter\@message@saved\expandafter{\the\toks@}%
+ \expandafter\do@@mark\expandafter{\the\@topmark@saved}%
+}%
+\gdef\@message@saved{}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\execute@message@pen}
+% The one-off output routine associated with \cmd\execute@message@pen\
+% simply executes the given message. It is last in a sequence of one-off output routine calls;
+% earlier ones have saved all that require saving.
+% \begin{macrocode}
+\mathchardef\execute@message@pen=10019
+\@namedef{output@-\the\execute@message@pen}{\@message@saved}%
+% \end{macrocode}
+% \end{macro}
+%
+%
+% \subsection{Output messages}
+%
+% Message handlers are procedures that execute output messages, tokens that are passed to
+% the output routine for execution in an environment appropriate to page makeup.
+%
+% How it works. We put down three large negative penalties, each of which will be handled by
+% the output dispatcher (\emph{not} the natural output routine), each penalty being protected by
+% a removable, non-discardable item (i.e., a box).
+% Either three or four invocations of one-off output routines are involved per message.
+%
+% We make the last of the three protection boxes have a depth equal to the value of \cmd\prevdepth\
+% that was current when the procedure is called. This effectively restores \cmd\prevdepth.
+%
+% In each case, the one-off output routine will remove the extraneous box we have inserted.
+% And the second and third one-off routines will simply void \cmd\box\cmd\@cclv, because its contents
+% are entirely artificial.
+%
+% FIXME: not so! If \cmd\holdinginserts\ is cleared, that box may have an insert node; it must be preserved, too.
+%
+% The first routine saves away the current column contents and remembers the \cmd\topmark\ for later
+% use. There is a variant routine that first clears \cmd\holdinginserts, so that the message can
+% handle any inserts present in the boxed-up page; this of course entails yet another
+% visit to the output routine.
+%
+% The penultimate routine saves away the tokens transmitted in via the \cmd\@@mark: the argument of
+% the macro. These tokens are of course the very thing we wish to execute within the safety
+% of the output routine. It also puts down a mark containing the \cmd\topmark\ tokens saved
+% by the first routine. By this means, the mark, which we have clobbered, is restored.
+%
+% The last routine simply executes the given tokens.
+% In the course of doing this, it must take care of \cmd\box\cmd\@cclv, either by shipping it out,
+% or by \cmd\unvbox ing it onto the MVL.
+%
+% \begin{macro}{\execute@message}
+% The procedure \cmd\execute@message\ simply calls the utility procedure \cmd\@execute@message\
+% with a penalty value for the standard treatment.
+% \begin{macrocode}
+\def\execute@message{%
+ \@execute@message\save@column@pen
+% \end{macrocode}
+% Implicit second argument
+% \begin{macrocode}
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\execute@message@insert}
+% The procedure \cmd\execute@message@insert\ is like \cmd\execute@message\ in all respects
+% except that the penalty value is \cmd\save@column@insert@pen, which arranges for the
+% message handler involved to deal with the page's insertions.
+% At the same time, we prepare the \cmd\footins\ box so that these insertions can be dealt with.
+%
+% Note:
+% If more insertions are added to \LaTeX\ (presumably via \cmd\newinsert), then they
+% must be dealt with in a way entirely analogous to \cmd\footins.
+% \changes{4.1n}{2009/12/02}{More diagnostics of column balancing}
+% \begin{macrocode}
+\def\execute@message@insert#1{%
+ \@execute@message\save@column@insert@pen{%
+ \setbox \footins \box \footins@saved
+ \ltxgrid@foot@info@sw{\class@info{\string\execute@message@insert}\trace@box\footins}{}%
+ #1%
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@execute@message}
+% The utility procedure \cmd\@execute@message\ is called by \cmd\execute@message\ and \cmd\execute@message@insert.
+% We prepare by creating a \cmd\vbox\ containing all the needed nodes and proceed by simply
+% \cmd\unvbox ing that box onto the MVL.
+% We ensure that \cmd\box\cmd\@cclv\ is properly set up for the output message handler
+% by always inserting \cmd\prep@cclv\ in advance of the argument.
+%
+% Note that each one-off output routine is invoked effectively the same as
+% \cmd\protect@penalty, except that the second invocation involves an additional
+% \cmd\mark\ node, and the third a specially prepared protection box.
+%
+% Note also that \TeX's primitive \cmd\mark\ is called here without any expansion protection.
+% This is the only place where it is called that way, but it's OK because those tokens
+% have have been pre-expanded by procedures that call \cmd\execute@message.
+% FIXME: all procedures calling \cmd\execute@message\ must pre-expand their tokens!
+% \begin{macrocode}
+\long\def\@execute@message#1#2{%
+ \begingroup
+ \dimen@\prevdepth\@ifdim{\dimen@<\z@}{\dimen@\z@}{}%
+ \setbox\z@\vbox{%
+ \protect@penalty#1%
+ \protection@box
+ \toks@{\prep@cclv#2}%
+ \@@mark{\the\toks@}%
+ \penalty-\save@message@pen
+% \end{macrocode}
+%\begin{verbatim}
+% \hbox{\vrule\@height\z@\@width\z@\@depth\dimen@}%
+%\end{verbatim}
+% \begin{macrocode}
+ \setbox\z@\null\dp\z@\dimen@\ht\z@-\dimen@
+ \nointerlineskip\box\z@
+ \penalty-\execute@message@pen
+ }\unvbox\z@
+ \endgroup
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\do@output@cclv}
+% The procedure \cmd\do@output@cclv\ provides access to message handlers at their simplest.
+% The message will execute in the usual environment of the output routine, with
+% the boxed-up page in \cmd\box\cmd\@cclv, and we assume that \cmd\holdinginserts\ remains set.
+% This procedure must be invoked within main vertical mode;
+% it is the obligation of the macro writer to ensure that this is the case.
+% \begin{macrocode}
+\def\do@output@cclv{\execute@message}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\do@output@MVL}
+% The procedure \cmd\do@output@MVL, like \cmd\do@output@cclv, is an interface for messages,
+% but provides two additional services: the command may also be invoked in horizontal mode, and
+% the message handler will execute with the MVL unboxed.
+% \begin{macrocode}
+\def\do@output@MVL#1{%
+ \@ifvmode{%
+ \begingroup\execute@message{\unvbox\@cclv#1}\endgroup
+ }{%
+ \@ifhmode{%
+ \vadjust{\execute@message{\unvbox\@cclv#1}}%
+ }{%
+ \@latexerr{\string\do@output@MVL\space cannot be executed in this mode!}\@eha
+ }%
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\lose@breaks}
+% The purpose of this procedure is to get rid of all the extraneous
+% \cmd\penalty\cmd\@M\ nodes that tend to build up in the MVL.
+% \begin{macrocode}
+\def\lose@breaks{%
+ \loopwhile{%
+ \count@\lastpenalty
+ \@ifnum{\count@=\@M}{%
+% \end{macrocode}
+% Note: 10000 is a TeX magic number!
+% \begin{macrocode}
+ \unpenalty\true@sw
+ }{%
+ \false@sw
+ }%
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\removestuff}
+% \cmd\removestuff\ is a document-level command that removes the bottom skip glue item
+% from the MVL.
+% \begin{macrocode}
+\def\removestuff{\do@output@MVL{\unskip\unpenalty}}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\removephantombox}
+% The procedure \cmd\removephantombox\ is a special-purpose message handler exclusively for
+% preventing incorrect spacing above display math. It must be issued in
+% horizontal mode within the phantom paragraph generated when display math starts up in
+% vertical mode.
+% \begin{macrocode}
+\def\removephantombox{%
+ \vadjust{%
+ \execute@message{%
+ \unvbox\@cclv
+ \remove@lastbox
+ \unskip
+ \unskip
+ \unpenalty
+ \penalty\predisplaypenalty
+ \vskip\abovedisplayskip
+ }%
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\addstuff}
+% \cmd\addstuff\ is a document-level command that adds penalty, glue, or both to the
+% MVL. The penalty and glue items are rearranged so that all penalties nodes precede all the glue nodes,
+% which is the canonical arrangement.
+% \begin{macrocode}
+\def\addstuff#1#2{\edef\@tempa{\noexpand\do@output@MVL{\noexpand\@addstuff{#1}{#2}}}\@tempa}%
+\def\@addstuff#1#2{%
+ \skip@\lastskip\unskip
+ \count@\lastpenalty\unpenalty
+ \@if@empty{#1}{}{\penalty#1\relax}%
+ \@ifnum{\count@=\z@}{}{\penalty\count@}%
+ \vskip\skip@
+ \@if@empty{#2}{}{\vskip#2\relax}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\replacestuff}
+% \cmd\replacestuff\ is a document-level command similar to \cmd\addstuff;
+% but it replaces penalty, glue, or both in the MVL.
+% The penalty and glue items are rearranged so that all penalties nodes precede all the glue nodes,
+% which is the canonical arrangement.
+% \begin{macrocode}
+\def\replacestuff#1#2{\edef\@tempa{\noexpand\do@output@MVL{\noexpand\@replacestuff{#1}{#2}}}\@tempa}%
+\def\@replacestuff#1#2{%
+ \skip@\lastskip\unskip
+ \count@\lastpenalty\unpenalty
+ \@if@empty{#1}{}{%
+ \@ifnum{\count@>\@M}{}{%
+ \@ifnum{\count@=\z@}{\count@=#1\relax}{%
+ \@ifnum{\count@<#1\relax}{}{%
+ \count@=#1\relax
+ }%
+ }%
+ }%
+ }%
+ \@ifnum{\count@=\z@}{}{\penalty\count@}%
+ \@if@empty{#2}{}{%
+ \@tempskipa#2\relax
+ \@ifdim{\z@>\@tempskipa}{%
+ \advance\skip@-\@tempskipa
+ }{%
+ \@ifdim{\skip@>\@tempskipa}{}{%
+ \skip@\@tempskipa
+ }%
+ }%
+ }%
+ \vskip\skip@
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\move@insertions}
+% \begin{macro}{\hold@insertions}
+% In order to avoid bolluxing up \cmd\insert\ registers by our one-off
+% output routines, we set \cmd\holdinginserts\ to zero by default and only
+% clear it (briefly) while we handle cases where we want inserts to show up.
+% \begin{macrocode}
+\def\move@insertions{\global\holdinginserts\z@}%
+\def\hold@insertions{\global\holdinginserts\@ne}%
+\hold@insertions
+\def\toggle@insert#1#2{%
+ \@ifnum{\holdinginserts>\z@}{\move@insertions#1}{\hold@insertions#2}%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+%
+% \subsection{Messages to alter the page grid}%
+%
+% Here is the implementation of the grid-switching procedures.
+% We perform two checks when changing the page grid; first to ensure that
+% the target page grid is known (defensive programming), second to ensure that
+% the switch is a non-trivial one. The latter check must be performed within
+% the safety of the output routine, so requires using an output message.
+% Thus, a grid change requires two messages, for a total of six visits to the output routine.
+%
+% \begin{macro}{\do@columngrid}
+% Utility procedure \cmd\do@columngrid\ changes the page grid.
+% Note that this command forces an end to the current paragraph. This is necessary, because
+% a page grid change makes no sense unless we can alter the \cmd\hsize\ before commencing to
+% typeset the following paragraph. So the command should never be executed in horizontal mode
+% anyway.
+% \begin{macrocode}
+\def\do@columngrid#1#2{%
+ \par
+ \expandafter\let\expandafter\@tempa\csname open@column@#1\endcsname
+ \@ifx{\relax\@tempa}{%
+ \ltxgrid@warn{Unknown page grid #1. No action taken}%
+ }{%
+ \do@output@MVL{\start@column{#1}{#2}}%
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\start@column}
+% Procedure \cmd\start@column\ lays down the interrupts to switch the page grid.
+% If the change to the page grid would have been trivial, it bails out.
+% It seems a reasonable tradeoff of processing versus security: once we commit
+% to changing the page grid, we clear \cmd\holdinginserts, so there is no turning back.
+%
+% Note that the second argument to the macro allows us to pass an argument to the
+% page grid that is starting up. This can be handy, because a single procedure can
+% handle multiple page grids, differing only by the value of a parameter.
+%
+% FIXME: this means that you cannot switch between mlt page grids in a single step.
+% But do we want to do this, at all, at all?
+% \begin{macrocode}
+\def\start@column#1#2{%
+ \def\@tempa{#1}\@ifx{\@tempa\thepagegrid}{%
+ \ltxgrid@info{Already in page grid \thepagegrid. No action taken}%
+ }{%
+ \expandafter\execute@message@insert
+ \expandafter{%
+ \csname shut@column@\thepagegrid\expandafter\endcsname
+ \csname open@column@#1\endcsname{#2}%
+ \set@vsize
+ }%
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\thepagegrid}
+% The macro \cmd\thepagegrid\ tracks what kind of page grid we are in.
+%
+% Note: Access \cmd\thepagegrid\ only within the safety of the output routine.
+%
+% Warning: The page grid should be changed only within the safety of the output routine.
+% People who write multicol page grid mechanisms appear not to understand the matter, so they
+% should particularly heed this warning. Think about it: obviously Lamport did so, which is why
+% his \cmd\twocolumn\ command forced a pagebreak, which is limiting, but safe.
+% \begin{macrocode}
+\def\thepagegrid{one}%
+% \end{macrocode}
+% \end{macro}
+%
+%
+% \subsection{Application Note: implementing a page grid}
+%
+% If you want to create a new page grid for \LaTeX,
+% you must define five procedures with specific names:
+% \cmd\open@column@ name, \cmd\shut@column@ name, \cmd\end@column@ name,
+% \cmd\output@column@ name, and \cmd\@addmarginpar@ name, where ``name'' is the
+% name of your page grid.
+%
+% The procedure \cmd\open@column@ name starts the new page grid. It should define \cmd\thepagegrid,
+% deal with \cmd\box\cmd\pagesofar\ and \cmd\box\cmd\footsofar\ (perhaps by leaving them alone),
+% and it should set the values of \LaTeX's page layout parameters for the column size and height.
+%
+% The procedure \cmd\shut@column@ name should expect to be called with \cmd\holdinginserts\
+% cleared (it can assume that \cmd\holdinginserts\ will automatically be restored).
+% It should properly deal with insertions (like footnotes); calling \cmd\@makecolumn\
+% with an argument of \cmd\false@sw\ will do this.
+% It should know that the page grid is being terminated
+% in the middle of a page, so it should make arrangements to carry the footnotes down to the
+% bottom of the column or page, and it should possibly salt away the material for later
+% incorporation into the page. The box registers \cmd\footsofar\ and \cmd\pagesofar\ are customarily used
+% for this purpose.
+%
+% The procedure \cmd\end@column@ name should kick out a possibly short page containing all the
+% floats committed to the page. It will be invoked during \cmd\clearpage\ processing.
+% After that, it should \cmd\unvbox\cmd\@cclv.
+%
+% The procedure \cmd\output@column@ name should ship out or commit the current \cmd\@outputbox.
+% In a one-column layout, you ship out; in a multicolumn layout, you commit the box as the contents
+% of a particular column, and if that column is the last, you ship out.
+%
+% The procedure \cmd\@addmarginpar@ name should return a boolean (either \cmd\true@sw\ or \cmd\false@sw\ or an equivalent)
+% to tell the marginpar mechanism to place the marginal material to the right or left, respectively.
+%
+% You can use the existing page grids ``one'' and ``mlt'' as a point of departure for creating others.
+% The former can be the basis for, say, a single-column page grid with a side column.
+%
+% \begin{macro}{\pagesofar}
+% \begin{macro}{\footsofar}
+% The box register \cmd\pagesofar\ holds the portion of the (full-width) page that is already composed into columns.
+% This, plus the finished columns, each with its floats, plus \cmd\box255 constitute the full galley.
+%
+% The box register \cmd\footsofar\ holds all of the footnotes associated with \cmd\pagesofar.
+% \begin{macrocode}
+\newbox\pagesofar
+\newbox\footsofar
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\combine@foot@inserts}
+% \changes{4.1n}{2010/01/02}{(AO, 571) calling sequence of \cs{combine@foot@inserts} and \cs{grid@column} to expose box registers; that of \cs{append@column@} its column counters}%
+% \changes{4.1n}{2010/01/02}{(AO, 571) coding convention: use \cs{bgroup}, \cs{egroup} (instead of braces) when a box is being built}%
+% The procedure \cmd\combine@foot@inserts\ is for the purpose of merging
+% the recently contributed footnotes (usually \cmd\box\cmd\footins)
+% with those saved from earlier on the page (usually \cmd\box\cmd\footsofar).
+%
+% It is employed in a number of circumstances.
+%
+% \cmd\@makecolumn (when its argument is \cmd\false@sw):
+% we are not shipping out, so we need to salt away any footnotes there may be.
+%
+% \cmd\shut@column@one: we are leaving the one-column page grid, so recover the footnotes from that material
+% and combine them with those of \cmd\pagesofar.
+%
+% \expandafter\cmd\csname balance@2\endcsname: two columns of type have been balanced, so now balance the footnotes.
+% The \cmd\combine@foot@inserts\ procedure is first used to gather footnotes from the columns balanced with those of \cmd\pagesofar.
+%
+% Bug 571 note: if balancing a two-column page grid, and there had been footnotes in the \cmd\pagesofar,
+% those footnotes will have been balanced into a page-width box, \cmd\box\cmd\footsofar.
+% We need to now re-cast them into a single, column-width galley, and only then
+% combine them with those in \cmd\box\cmd\footins.
+%
+% \begin{macrocode}
+\def\combine@foot@inserts#1#2{%
+ \ltxgrid@info@sw{\class@info{\string\combine@foot@inserts\string#1\string#2}}{}%
+ \@ifvoid#1{%
+ \ltxgrid@foot@info@sw{\trace@box#2}{}\global\setbox#1\box#2%
+ }{%
+ \global\setbox#1\vbox\bgroup
+ \ltxgrid@foot@info@sw{\trace@box#1}{}\unvbox#1%
+ \@ifvoid#2{}{%
+ \marry@baselines
+ \ltxgrid@foot@info@sw{\trace@box#2}{}\unvbox#2%
+ }%
+ \egroup
+ }%
+ \ltxgrid@foot@info@sw{\trace@scroll{\showbox#1\showbox#2}}{}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+%
+% \subsubsection{One-column page grid}
+%
+% \begin{macro}{\onecolumngrid}
+% \begin{macro}{\open@column@one}
+% \begin{macro}{\shut@column@one}
+% \begin{macro}{\float@column@one}
+% \begin{macro}{\end@column@one}
+% \begin{macro}{\output@column@one}
+% \begin{macro}{\@addmarginpar@one}
+% Here are all the procedures necessary for the standard page grid named ``one'':
+% a single column layout. It is, of course, \LaTeX's familiar \cmd\onecolumn\ layout.
+% We begin with the procedure exposed to the style writer.
+% This is, however, not a \LaTeX\ command; users should not change the page grid.
+% \begin{macrocode}
+\newcommand\onecolumngrid{\do@columngrid{one}{\@ne}}%
+% \end{macrocode}
+%
+% Note that a document class that issues the command \cmd\onecolumn\ will break. This includes
+% \LaTeX's standard classes.dtx-based classes: if your class descends from one of these, you
+% must expunge it of all such commands.
+% \begin{macrocode}
+\let\onecolumn\@undefined
+% \end{macrocode}
+%
+% The procedure \cmd\open@column@one\ takes advantage of the special nature of the one-column
+% page grid to deal with \cmd\box\cmd\pagesofar, therefore it must also reset \cmd\@colroom.
+% \changes{4.1b}{2008/08/04}{Change \cs{set@colroom} to \cs{set@colht}}
+% \changes{4.1n}{2009/12/02}{More diagnostics of column balancing}
+% \begin{macrocode}
+\def\open@column@one#1{%
+ \ltxgrid@info@sw{\class@info{\string\open@column@one\string#1}}{}%
+% \end{macrocode}
+% Throw the \cmd\pagesofar\ back onto the Main Vertical List.
+% At this point, we must also \cmd\insert\ the footnotes back into the MVL.
+% \changes{4.1n}{2010/01/02}{(AO, 571) coding convention: use \cs{bgroup}, \cs{egroup} (instead of braces) when a box is being built}%
+% \begin{macrocode}
+ \unvbox\pagesofar
+ \@ifvoid{\footsofar}{}{%
+ \insert\footins\bgroup\unvbox\footsofar\egroup
+ \penalty\z@
+ }%
+% \end{macrocode}
+% Record which page grid we are using. Then calculate the set width (\cmd\hsize) and the goal height (\cmd\vsize).
+%
+% Klootch: we set the \cmd\count\cmd\footins\ to a magic number. This is only correct in the case of a two-column document.
+% \begin{macrocode}
+ \gdef\thepagegrid{one}%
+ \global\pagegrid@col#1%
+ \global\pagegrid@cur\@ne
+ \global\count\footins\@m
+ \global\divide\count\footins\tw@
+ \set@column@hsize\pagegrid@col
+ \set@colht
+}%
+% \end{macrocode}
+%
+% The procedure \cmd\shut@column@one\ saves away the one-column material into the box register \cmd\pagesofar.
+% Because it is called from a message handler, we are assured that marks are properly taken care of.
+%
+% This instance of \cmd\@makecolumn\ is building a column for saving into \cmd\pagesofar.
+% \changes{4.1b}{2008/08/04}{Change \cs{@makecol} to \cs{@makecolumn} with argument}
+% \changes{4.1b}{2008/08/04}{Change \cs{set@colroom} to \cs{set@colht}}
+% \changes{4.1f}{2009/07/15}{(AO, 519) Preserve footnotes that are in \cs{footsofar} across a page grid change}
+% \changes{4.1n}{2009/12/02}{More diagnostics of column balancing}
+% \changes{4.1n}{2010/01/02}{(AO, 571) calling sequence of \cs{combine@foot@inserts} and \cs{grid@column} to expose box registers; that of \cs{append@column@} its column counters}%
+% We recover the footnotes into \cmd\footsofar\ (globally) and the column into \cmd\pagesofar\ (also globally), voiding \cmd\@outputbox\ by side effect.
+% \begin{macrocode}
+\def\shut@column@one{%
+ \ltxgrid@info@sw{\class@info{\string\shut@column@one}}{}%
+ \@makecolumn\false@sw
+% \end{macrocode}
+% Split text portion of \cmd\@outputbox\ into \cmd\pagesofar,
+% and add its footnote portion to \cmd\footsofar.
+% Then void out \cmd\@outputbox.
+% \begin{macrocode}
+ \global\setbox\pagesofar\vbox\bgroup
+ \recover@column\@outputbox\footsofar\column@recovered\footins@recovered
+ \egroup
+ \begingroup\setbox\z@\box\@outputbox\endgroup
+% \end{macrocode}
+% FIXME: is \cmd\combine@foot@inserts\ needed?
+% Also: if this procedure is immediately followed by \cmd\open@column@ \texttt{grid}, then \cmd\set@colht\ will be unneeded.
+% \begin{macrocode}
+ \combine@foot@inserts\footsofar\footins
+ \set@colht
+}%
+% \end{macrocode}
+%
+%FIXME: the first line of a footnote should have an up-strut, and the last line a down-strut, so that they can marry baselines.
+% The latter is the case; how about the former?
+%
+% The procedure \cmd\float@column@one\ takes care of a float column that has been built by
+% \cmd\@tryfcolumn, in the single-column page grid.
+%
+% This instance of \cmd\@makecolumn\ is followed by \cmd\@outputpage: it is building
+% a column for \cmd\shipout, rather than for saving into \cmd\pagesofar.
+% \changes{4.1b}{2008/08/04}{Change \cs{@makecol} to \cs{@makecolumn} with argument}
+% \begin{macrocode}
+\def\float@column@one{%
+ \@makecolumn\true@sw
+ \@outputpage
+}%
+% \end{macrocode}
+%
+% The procedure \cmd\end@column@one\ is executed at the end of \cmd\clearpage\ processing,
+% if we were in a one-column page grid, once all permissive float pages have been shipped out.
+% At this point, one could perhaps
+% assume that nothing more need be done, but let us anyway test for committed floats and force a shipout.
+%
+% FIXME: this procedure does the same as \cmd\end@column@mlt
+% (except for the test of \cmd\@ifx@empty\cmd\@dbltoplist):
+% the two could almost be the same procedure.
+%
+% I have changed this procedure to avoid the testing it once did: it simply puts down interrupts,
+% upon which it relies to correctly do what \cmd\clearpage\ requires.
+% \begin{macrocode}
+\def\end@column@one{%
+ \unvbox\@cclv\remove@lastbox
+ \protect@penalty\do@newpage@pen
+}%
+% \end{macrocode}
+%
+% The procedure \cmd\output@column@one\ is dispatched from the output routine when
+% we have completed a page (that is, a column in a one-column page grid);
+% it ships out the page using the \cmd\@outputpage.
+% It will be followed up with an output routine message to prepare a new column.
+%
+% Query: by what mechanism do the footnotes get placed onto such a page?
+% \begin{macrocode}
+\def\output@column@one{%
+ \@outputpage
+}%
+% \end{macrocode}
+%
+% The following procedure determines which side of the page a marginpar will appear.
+% It reproduces the behavior of standard \LaTeX.
+% \begin{macrocode}
+\def\@addmarginpar@one{%
+ \@if@sw\if@mparswitch\fi{%
+ \@ifodd\c@page{\false@sw}{\true@sw}%
+ }{\false@sw}{%
+ \@if@sw\if@reversemargin\fi{\false@sw}{\true@sw}%
+ }{%
+ \@if@sw\if@reversemargin\fi{\true@sw}{\false@sw}%
+ }%
+}%
+% \end{macrocode}
+%
+% The following procedure yields a Boolean value; it determines whether a float in the deferred queue
+% is appropriate for placing. In the one-column grid, all floats are so.
+% \begin{macrocode}
+\def\@floatselect@sw@one#1{\true@sw}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\onecolumngrid@push{%
+ \do@output@MVL{%
+ \@ifnum{\pagegrid@col=\@ne}{%
+ \global\let\restorecolumngrid\@empty
+ }{%
+ \xdef\restorecolumngrid{%
+ \noexpand\start@column{\thepagegrid}{\the\pagegrid@col}%
+ }%
+ \start@column{one}{\@ne}%
+ }%
+ }%
+}%
+\def\onecolumngrid@pop{%
+ \do@output@MVL{\restorecolumngrid}%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+%
+% \subsubsection{Two-column page grid}
+%
+% \begin{macro}{\twocolumngrid}
+% \begin{macro}{\open@column@mlt}
+% \begin{macro}{\shut@column@mlt}
+% \begin{macro}{\end@column@mlt}
+% \begin{macro}{\output@column@mlt}
+% \begin{macro}{\@addmarginpar@mlt}
+% \begin{macro}{\set@footnotewidth@mlt}
+% \begin{macro}{\set@footnotewidth@two}
+% \begin{macro}{\compose@footnotes@two}
+% Here are all the procedures necessary for the standard page grid named ``mlt'':
+% the multi-column page grid. With an argument of "2", it is,
+% of course, \LaTeX's familiar \cmd\twocolumn\ layout.
+%
+% We start with the procedure to switch to the two-column page grid.
+% \begin{macrocode}
+\newcommand\twocolumngrid{\do@columngrid{mlt}{\tw@}}%
+% \end{macrocode}
+%
+% The corresponding command of \LaTeX\ is obsolete.
+% \begin{macrocode}
+\let\twocolumn\@undefined
+% \end{macrocode}
+%
+% Of course, \cmd\@topnewpage\ is also obsolete. Just do
+%\begin{quote}
+% \cmd\clearpage\cmd\onecolumngrid<vertical mode material>\cmd\twocolumngrid.
+%\end{quote}
+% \begin{macrocode}
+\let\@topnewpage\@undefined
+% \end{macrocode}
+%
+% If your document class descends from one of \LaTeX's standard classes.dtx-derived
+% classes, it will break. You must expunge from it all such commands.
+%
+% \changes{4.1b}{2008/08/04}{Change \cs{set@colroom} to \cs{set@colht}}
+%
+% Bug 571 note: it is not enough to have the \cmd\pagesofar, we must also deal with the \cmd\footsofar.
+% At this juncture, we should treat the case where the document has an essentially two-column page grid,
+% with occasional excursions into the one-column grid.
+% If a footnote is set within the latter grid, its set width should be that of the two-column grid.
+%
+% When a page is shipped out, if we are currently in a one-column grid, we will compose the footnotes onto the page
+% in the form of balanced columns. This is only one way to handle footnotes: \classname{multicol}
+% appears to set footnotes on the full text width.
+% \begin{macrocode}
+\def\open@column@mlt#1{%
+ \ltxgrid@info@sw{\class@info{\string\open@column@mlt\string#1}}{}%
+% \end{macrocode}
+% At this point, we must \cmd\insert\ the footnotes back into the Main Vertical List.
+% \changes{4.1n}{2010/01/02}{(AO, 571) coding convention: use \cs{bgroup}, \cs{egroup} (instead of braces) when a box is being built}%
+% \begin{macrocode}
+ \@ifvoid{\footsofar}{}{%
+ \insert\footins\bgroup\unvbox\footsofar\egroup
+ }%
+% \end{macrocode}
+% Record which page grid we are using. Then calculate the set width (\cmd\hsize) and the goal height (\cmd\vsize).
+%
+% Klootch: we set the \cmd\count\cmd\footins\ to a magic number. This value is valid whether footnotes are being
+% set on the column width or the full text width.
+% \begin{macrocode}
+ \gdef\thepagegrid{mlt}%
+ \global\pagegrid@col#1%
+ \global\pagegrid@cur\@ne
+ \global\count\footins\@m
+ \set@column@hsize\pagegrid@col
+ \set@colht
+}%
+% \end{macrocode}
+%
+% The procedure \cmd\shut@column@mlt\ ends the current column, balances the columns, and
+% salts away all in \cmd\pagesofar. Because it is called in a message handler,
+% we are assured that marks are handled properly.
+% Attention: because this procedure balances columns, all footnotes are
+% held aside in \cmd\footsofar\ for placement at the bottom of the page.
+%
+% Bug note: the last macro executed by this procedure is \cmd\set@colht,
+% but had been erroneously \cmd\set@colroom.
+% I now believe that the latter should be changed pretty much everywhere to the former.
+%
+% This instance of \cmd\@makecolumn\ is building material for \cmd\pagesofar, rather than
+% for \cmd\shipout.
+% \changes{4.1b}{2008/08/04}{Change \cs{@makecol} to \cs{@makecolumn} with argument}
+% \changes{4.1b}{2008/08/04}{Change \cs{@combinepage} to \cs{@combinepage} with argument}
+% \changes{4.1b}{2008/08/04}{Change \cs{set@colroom} to \cs{set@colht}}
+% \changes{4.1n}{2009/12/02}{More diagnostics of column balancing}
+% \changes{4.1n}{2010/01/02}{(AO, 571) calling sequence of \cs{combine@foot@inserts} and \cs{grid@column} to expose box registers; that of \cs{append@column@} its column counters}%
+% \begin{macrocode}
+\def\shut@column@mlt{%
+ \ltxgrid@info@sw{\class@info{\string\shut@column@mlt}}{}%
+ \@cclv@nontrivial@sw{%
+ \@makecolumn\false@sw
+ \@ifnum{\pagegrid@cur<\pagegrid@col}{%
+ \expandafter\global\expandafter\setbox\csname col@\the\pagegrid@cur\endcsname\box\@outputbox
+ \global\advance\pagegrid@cur\@ne
+ }{}%
+ }{%
+ \void@cclv
+ }%
+ \@ifnum{\pagegrid@cur>\@ne}{%
+ \csname balance@\the\pagegrid@col\endcsname
+ \grid@column\@outputbox{}%
+ \@combinepage\false@sw
+ \@combinedblfloats
+ \global\setbox\pagesofar\box\@outputbox
+ \show@pagesofar@size
+ }{}%
+ \set@colht
+}%
+% \end{macrocode}
+%
+% The procedure \cmd\float@column@mlt\ takes care of a float page that has been built by \cmd\@tryfcolumn,
+% in the multi-column page grid. It is coincidentally identical to what happens in
+% \cmd\do@startpage\ when a page needs to be shipped out.
+% \changes{4.1b}{2008/08/04}{New procedure \cs{@output@combined@page}}
+% \begin{macrocode}
+\def\float@column@mlt{%
+ \@output@combined@page
+}%
+% \end{macrocode}
+%
+% The procedure \cmd\end@column@mlt\ is executed at the end of \cmd\clearpage\ processing,
+% if we were in a multi-column page grid, once all permissive float pages have been shipped out.
+% If no floats are committed and if no columns are yet filled, we have nothing to do.
+% Otherwise, we kick out a column and try again.
+%
+% Note that in our code to kick out a column, we must deal properly with the case where the column
+% is trivial: it will have nothing but \cmd\topskip\ glue plus a protection box. We substitute an ordinary
+% \cmd\null\ for the protection box.
+%
+% \begin{macrocode}
+\def\end@column@mlt{%
+ \@ifx@empty\@toplist{%
+ \@ifx@empty\@botlist{%
+ \@ifx@empty\@dbltoplist{%
+ \@ifx@empty\@deferlist{%
+ \@ifnum{\pagegrid@cur=\@ne}{%
+ \false@sw
+ }{%
+ \true@sw
+ }%
+ }{%
+ \true@sw
+ }%
+ }{%
+ \true@sw
+ }%
+ }{%
+ \true@sw
+ }%
+ }{%
+ \true@sw
+ }%
+ % true = kick out a column and try again
+ {%
+ \@cclv@nontrivial@sw{%
+ \unvbox\@cclv\remove@lastbox
+ }{%
+ \unvbox\@cclv\remove@lastbox\unskip\null
+ }%
+ \protect@penalty\do@newpage@pen
+ \protect@penalty\do@endpage@pen
+ }{%
+ \unvbox\@cclv\remove@lastbox
+ }%
+}%
+% \end{macrocode}
+%
+% The procedure \cmd\output@column@mlt (cf. \cmd\output@column@one)
+% is dispatched from the output routine when
+% we have completed a column in a multi-column page grid).
+% (It replaces the \cmd\@outputdblcol\ of standard \LaTeX.)
+% If a complete set of columns is at hand, it ships out the page and
+% lays down an interrupt for \cmd\do@startpage@pen, which will commit the
+% full-page-width floats to the next page.
+% Like \cmd\output@column@mlt, this is followed by
+% an output routine message to prepare a new column.
+%
+% If a page needs to be shipped out, it uses the same mechanism as \cmd\do@startpage.
+% \changes{4.1b}{2008/08/04}{New procedure \cs{@output@combined@page}}
+% \changes{4.1n}{2010/01/02}{(AO, 571) calling sequence of \cs{combine@foot@inserts} and \cs{grid@column} to expose box registers; that of \cs{append@column@} its column counters}%
+% \begin{macrocode}
+\def\output@column@mlt{%
+ \@ifnum{\pagegrid@cur<\pagegrid@col}{%
+ \expandafter\global\expandafter\setbox\csname col@\the\pagegrid@cur\endcsname\box\@outputbox
+ \global\advance\pagegrid@cur\@ne
+ }{%
+ \set@adj@colht\dimen@
+ \grid@column\@outputbox{}%
+ \@output@combined@page
+ }%
+}%
+% \end{macrocode}
+% The procedure \cmd\output@column@mlt\ obsoletes \LaTeX's \cmd\@outputdblcol
+% \begin{macrocode}
+\let\@outputdblcol\@undefined
+% \end{macrocode}
+%
+% The following procedure yields a Boolean value; it determines whether a float in the deferred queue
+% is appropriate for placement in the column. In the multi-column grid, only those narrower than \cmd\textwidth\ are so.
+% \begin{macrocode}
+\def\@floatselect@sw@mlt#1{\@if@notdblfloat{#1}}%
+% \end{macrocode}
+%
+% The following procedure determines which side of the page a marginpar will appear.
+% It reproduces the behavior of standard \LaTeX.
+% \begin{macrocode}
+\def\@addmarginpar@mlt{% emits a boolean
+ \@ifnum{\pagegrid@cur=\@ne}%
+}%
+% \end{macrocode}
+%
+% \cmd\set@footnotewidth@one\ sets the width of type within footnotes to span the full text width;
+% \cmd\set@footnotewidth@two\ to span a single column of the two-column grid, and
+% more generally \cmd\set@footnotewidth@mlt\ for a multi-column page grid.
+% \begin{macrocode}
+\def\set@footnotewidth@one{%
+ \hsize\columnwidth
+ \linewidth\hsize
+}%
+\def\set@footnotewidth@two{\set@footnotewidth@mlt\tw@}%
+\def\set@footnotewidth@mlt#1{%
+ \hsize\textwidth
+ \advance\hsize\columnsep
+ \divide\hsize#1%
+ \advance\hsize-\columnsep
+ \linewidth\hsize
+}%
+% \end{macrocode}
+%
+% \cmd\compose@footnotes\ is the procedure for arranging the footnotes for
+% placement at the bottom of the page or column.
+% In the former case, the material will be shipped out; in the latter, we must allow
+% the column to possibly be balanced later on.
+%
+% \cmd\compose@footnotes@one\ is a no-op, because the footnotes require no rearrangement.
+% In a scheme where footnotes are set on the full text width, this would be the procedure called.
+%
+% \cmd\compose@footnotes@two\ implements the case where a two-column document has been interrupted
+% with full-page-width text (e.g., the widetext environment or the end of the document),
+% and a natural page break appears.
+%
+% In either case, we assume that argument \verb+#1+ is an \cmd\insert\ register and must be assigned globally,
+% so that when it is accessed with \cmd\box\ or \cmd\unvbox, it will be voided globally as well.
+%
+% To extend this scheme to a three-column page grid \cmd\compose@footnotes@thr@@\ would be created:
+% it would balance the saved up footnotes into three columns.
+%
+% \begin{macrocode}
+\def\compose@footnotes@one#1{%
+ \ltxgrid@foot@info@sw{\class@info{\string\compose@footnotes@one\string#1}\trace@box#1}{}%
+}%
+\let\compose@footnotes\compose@footnotes@one
+\def\compose@footnotes@two#1{%
+ \ltxgrid@foot@info@sw{\class@info{\string\compose@footnotes@two\string#1}\trace@box#1}{}%
+ \setbox\z@\box\@tempboxa
+ \let\recover@column\recover@column@null
+ \let\marry@baselines\@empty
+ \balance@two#1\@tempboxa
+ \global\setbox#1\hbox to\textwidth{\box#1\hfil\box\@tempboxa}%
+ \ltxgrid@foot@info@sw{\trace@box#1}{}%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+%
+% \subsubsection{Page grid utility procedures}
+%
+% \begin{macro}{\pagegrid@cur}
+% \begin{macro}{\pagegrid@col}
+% \begin{macro}{\pagegrid@init}
+% We take over \LaTeX's \cmd\col@number, and \cmd\@leftcolumn, which are obsolete (\cmd\@holdpg\ could also be taken over).
+% We create two counters to hold the columns in the page grid and the current column within.
+% We also create the first of a set of box registers to hold the committted columns.
+% \begin{macrocode}
+\let\pagegrid@cur\col@number
+\let\col@number\@undefined
+\newcount\pagegrid@col
+\pagegrid@cur\@ne
+\expandafter\let\csname col@\the\pagegrid@cur\endcsname\@leftcolumn
+\let\@leftcolumn\@undefined
+% \end{macrocode}
+%
+% The default is for maximum two columns.
+% If your class will require more columns, assign that number to \cmd\pagegrid@col\
+% before \envb{document} time.
+% \begin{macrocode}
+\pagegrid@col\tw@
+% \end{macrocode}
+%
+% The procedure \cmd\pagegrid@init\ is a loop, exercising \cmd\newbox\ sufficiently to create the
+% boxes for holding the columns in the page grid; these have names like \expandafter\cmd\csname col@1\endcsname, etc.
+% \begin{macrocode}
+\def\pagegrid@init{%
+ \advance\pagegrid@cur\@ne
+ \@ifnum{\pagegrid@cur<\pagegrid@col}{%
+ \csname newbox\expandafter\endcsname\csname col@\the\pagegrid@cur\endcsname
+ \pagegrid@init
+ }{%
+ }%
+}%
+\appdef\class@documenthook{%
+ \pagegrid@init
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\grid@column}
+% \changes{4.1n}{2010/01/02}{(AO, 571) calling sequence of \cs{combine@foot@inserts} and \cs{grid@column} to expose box registers; that of \cs{append@column@} its column counters}%
+% The procedure \cmd\grid@column\ knows how to lay up the columns in a multi-column page grid.
+% It uses utility procedures \cmd\append@column@\ and \cmd\box@column.
+%
+% The first argument is the box register to create, usually \cmd\@outputbox, and provides both input and output.
+% The second argument a dimension, allowing us to strut down the depth of the box we create.
+%
+% \begin{macrocode}
+\def\grid@column#1#2{%
+ \ltxgrid@info@sw{\class@info{\string\grid@column\string#1}}{}%
+ \global\setbox#1\vbox\bgroup
+ \hb@xt@\textwidth\bgroup
+ \vrule\@height\z@\@width\z@\@if@empty{#2}{}{\@depth#2}%
+ \pagegrid@cur\@ne
+ \@ifnum{\pagegrid@cur<\pagegrid@col}{\loopwhile{\append@column@\pagegrid@cur\pagegrid@col}}{}%
+ \box@column#1%
+ \egroup
+% \end{macrocode}
+% FIXME: page depth!
+% \begin{macrocode}
+ \vskip\z@skip
+ \egroup
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\append@column@}
+% \begin{macro}{\box@column}
+% \begin{macro}{\marry@baselines}
+% The procedure \cmd\append@column@\ appends columns for \cmd\grid@column,
+% \cmd\box@column\ builds the columns for \cmd\append@column@,
+% and \cmd\marry@baselines\ pastes vertical things back together.
+% \changes{4.0a}{2001/06/18}{Introduce \cs{marry@height} }
+%
+% Note that \cmd\box@column\ makes an attempt to prevent excessive \cmd\topskip\
+% or \cmd\baselineskip\ glue
+% from being applied by \TeX\ when \cmd\@outputbox\ is contributed to the MVL.
+% If this is not done, it is possible to get into an infinite loop in the corner case,
+% wherein the page grid is changed to one column and the balanced-up columns are
+% already sufficient to fill the page.
+%
+% Note (AO 0920): I have changed the dimension involved with \cmd\box@column\ from
+% \cmd\vsize\ to \cmd\textheight, because the former is certainly not the correct value
+% to use: it will change if floats have been placed in the last column of the page.
+% I believe \cmd\textheight\ is the correct parameter to use here.
+%
+% A REVTeX4 user, Sergey Strelkov (strelkov@maik.rssi.ru), wants the option
+% of ragged-bottom columns. Implementing this feature properly means reboxing the
+% columns to their natural height only if \cmd\raggedcolumn@sw\ is true.
+% Otherwise, they get reboxed to their common height (\cmd\@colht?).
+%
+% Note that the default has hereby changed from ragged to flush.
+% It's not clear that anyone but Sergey will notice.
+%
+% The macro \cmd\marry@skip\ addresses (in a limited way)
+% the fact that neither the value of \cmd\baselineskip\ nor that of \cmd\topskip\
+% can be relied upon for the purpose of marrying the baselines of two split columns.
+% (Because there might have been a local change to their values at the point where
+% the output routine got triggered.)
+%
+% For best results, your document class should call for grid changes only when in basal text
+% settings. The \cmd\marry@baselines\ procedure will use the values appropriate to that point
+% when attempting to put the columns back together.
+%
+% In any case, we are not attempting to solve the more general problem of how to marry baselines
+% where the leading can change arbitrarily within the galley or where glue could have been trimmed
+% at a page top.
+%
+% \changes{4.1n}{2010/01/02}{(AO, 571) calling sequence of \cs{combine@foot@inserts} and \cs{grid@column} to expose box registers; that of \cs{append@column@} its column counters}%
+% Procedure \cmd\append@column@\ composes a column onto the horizontal list along with its \cmd\columnseprule.
+% Its arguments are: \verb+#1+---\cmd\pagegrid@cur, and \verb+#2+---\cmd\pagegrid@col
+% \begin{macrocode}
+\def\append@column@#1#2{%
+ \expandafter\box@column\csname col@\the#1\endcsname
+ \hfil\vrule\@width\columnseprule\hfil
+ \advance#1\@ne
+% \end{macrocode}
+% This procedure is the argument of \cmd\loopwhile, so it must leave a Boolean (e.g., \cmd\true@sw) in \TeX's scanner.
+% \begin{macrocode}
+ \@ifnum{#1<#2}%
+}%
+% \end{macrocode}
+% \changes{4.1n}{2010/01/02}{(AO, 571) calling sequence of \cs{combine@foot@inserts} and \cs{grid@column} to expose box registers; that of \cs{append@column@} its column counters}%
+% \changes{4.1n}{2009/12/02}{More diagnostics of column balancing}
+% \changes{4.1n}{2010/01/02}{(AO, 571) coding convention: use \cs{bgroup}, \cs{egroup} (instead of braces) when a box is being built}%
+% Procdure \cmd\box@column, used by \cmd\append@column@, puts down a box containing the specified column.
+% Its height is adjusted down to \cmd\@colht, if needed; likewise, the width is set to \cmd\columnwidth.
+% The rag at the bottom is controlled by \cmd\raggedcolumn@skip.
+% \begin{macrocode}
+\def\box@column#1{%
+ \ltxgrid@info@sw{\class@info{\string\box@column\string#1}}{}%
+ \raise\topskip
+ \hb@xt@\columnwidth\bgroup
+ \dimen@\ht#1\@ifdim{\dimen@>\@colht}{\dimen@\@colht}{}%
+ \count@\vbadness\vbadness\@M
+ \dimen@ii\vfuzz\vfuzz\maxdimen
+ \ltxgrid@info@sw{\saythe\@colht\saythe\dimen@}{}%
+ \vtop to\dimen@\bgroup
+ \hrule\@height\z@
+ \unvbox#1%
+ \raggedcolumn@skip
+ \egroup
+ \vfuzz\dimen@ii
+ \vbadness\count@
+ \hss
+ \egroup
+}%
+% \end{macrocode}
+% The purpose of procedure \cmd\marry@baselines\ is to ensure that the baseline spacing is correct;
+% it does this by making adjustments to the previous line, compensating for its depth, and by adding in
+% skip glue in an amount that assumes the added material has \cmd\topskip\ glue above.
+% \begin{macrocode}
+\def\marry@baselines{%
+ \begingroup
+ \setbox\z@\lastbox
+ \@ifvoid{\z@}{%
+ \endgroup
+ }{%
+ \aftergroup\kern
+ \aftergroup-%
+ \expandafter\box\expandafter\z@\expandafter\endgroup\the\dp\z@\relax
+ }%
+ \vskip\marry@skip\relax
+}%
+\gdef\marry@skip{\z@skip}%
+\def\set@marry@skip{%
+ \begingroup
+ \skip@\baselineskip\advance\skip@-\topskip
+ \@ifdim{\skip@>\z@}{%
+ \xdef\marry@skip{\the\skip@}%
+ }{}%
+ \endgroup
+}%
+% \end{macrocode}
+% \changes{4.1b}{2008/08/04}{Use \cs{document@inithook} instead of \cs{AtBeginDocument}}
+% \begin{macrocode}
+\appdef\document@inithook{%
+ \@ifxundefined\raggedcolumn@sw{\@booleanfalse\raggedcolumn@sw}{}%
+}%
+\def\raggedcolumn@skip{%
+ \vskip\z@\raggedcolumn@sw{\@plus.0001fil\@minus.0001fil}{}\relax
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\@combinepage}
+% The procedure \cmd\@combinepage\ prepends the stored page (\cmd\pagesofar) to \cmd\@outputbox\ and
+% employs \cmd\@combineinserts\ to lay down the footnotes.
+% The next event will usually be shipping out the made-up page, but not always.
+% Therefore the argument of \cmd\@combinepage, which must be a Boolean, determines
+% if the footnotes are to be combined into this page.
+%
+% \changes{4.1b}{2008/08/04}{Change \cs{@combinepage} to \cs{@combinepage} with argument}
+% \changes{4.1b}{2008/08/04}{(AO, 452) Support length checking: show size of shipped out text.}
+% \changes{4.1n}{2009/12/02}{More diagnostics of column balancing}
+%
+% QUERY: In the following, if \cmd\box\cmd\footins\ is not void, its contents are lost. Can this ever happen?
+% \begin{macrocode}
+\def\@combinepage#1{%
+ \ltxgrid@foot@info@sw{\class@info{\string\@combinepage\string#1}}{}%
+ \@ifvoid\pagesofar{}{%
+ \setbox\@outputbox\vbox{%
+ \unvbox\pagesofar
+ \marry@baselines
+ \unvbox\@outputbox
+ }%
+ }%
+ #1{%
+ \@ifvoid\footsofar{}{%
+% \end{macrocode}
+% At this point, \cmd\footins\ is empty; all of the footnotes have been combined into \cmd\footsofar.
+% \begin{macrocode}
+ \show@box@size{Combining page footnotes}\footsofar
+ \setbox\footins\box\footsofar
+% \end{macrocode}
+% Depending on the page grid, we compose the footnotes for placement on the page.
+% \begin{macrocode}
+ \compose@footnotes
+ \@combineinserts\@outputbox\footins
+ }%
+ }{%
+% \end{macrocode}
+%QUERY: The following line was removed, probably to fix a bug. When was this done?
+% \begin{verbatim}
+% \global\setbox\footins\box\footsofar
+% \end{verbatim}
+% \begin{macrocode}
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@cflt}
+% \begin{macro}{\@cflb}
+% \changes{4.1b}{2008/08/04}{}
+% We modify \LaTeX's \cmd\@cflt\ and \cmd\@cflb\ to remove the unwanted glue with \cmd\unskip.
+% \begin{macrocode}
+\def \@cflt{%
+ \let \@elt \@comflelt
+ \setbox\@tempboxa \vbox{}%
+ \@toplist
+ \setbox\@outputbox \vbox{%
+ \boxmaxdepth \maxdepth
+ \unvbox\@tempboxa\unskip
+ \topfigrule\vskip \textfloatsep
+ \unvbox\@outputbox
+ }%
+ \let\@elt\relax
+ \xdef\@freelist{\@freelist\@toplist}%
+ \global\let\@toplist\@empty
+}%
+\def \@cflb {%
+ \let\@elt\@comflelt
+ \setbox\@tempboxa \vbox{}%
+ \@botlist
+ \setbox\@outputbox \vbox{%
+ \unvbox\@outputbox
+ \vskip \textfloatsep\botfigrule
+ \unvbox\@tempboxa\unskip
+ }%
+ \let\@elt\relax
+ \xdef\@freelist{\@freelist\@botlist}%
+ \global \let \@botlist\@empty
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\@combinedblfloats}
+% We modify \LaTeX's \cmd\@combinedblfloats\ to be more appropriate for incremental page building:
+% we \cmd\unvbox\ the \cmd\@outputbox.
+% \begin{macrocode}
+\def\@combinedblfloats{%
+ \@ifx@empty\@dbltoplist{}{%
+ \setbox\@tempboxa\vbox{}%
+ \let\@elt\@comdblflelt\@dbltoplist
+ \let\@elt\relax\xdef\@freelist{\@freelist\@dbltoplist}%
+ \global\let\@dbltoplist\@empty
+ \setbox\@outputbox\vbox{%
+ %\boxmaxdepth\maxdepth %% probably not needed, CAR
+ \unvbox\@tempboxa\unskip
+ \@ifnum{\@dbltopnum>\m@ne}{\dblfigrule}{}%FIXME: how is \@dbltopnum maintained?
+ \vskip\dbltextfloatsep
+ \unvbox\@outputbox
+ }%
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\set@column@hsize}%
+% The procedure \cmd\set@column@hsize\ takes care of setting up the horizontal dimensions
+% for the current page grid. The present routine will certainly not be adequate for more
+% complex page layouts (e.g., with a side column), but works for the common ones.
+% \changes{4.0a}{2001/06/18}{Introduce \cs{set@marry@height} }
+% \begin{macrocode}
+\def\set@column@hsize#1{%
+ \pagegrid@col#1%
+ \global\columnwidth\textwidth
+ \global\advance\columnwidth\columnsep
+ \global\divide\columnwidth\pagegrid@col
+ \global\advance\columnwidth-\columnsep
+ \global\hsize\columnwidth
+ \global\linewidth\columnwidth
+ \skip@\baselineskip\advance\skip@-\topskip
+ \@ifnum{\pagegrid@col>\@ne}{\set@marry@skip}{}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\set@colht}%
+% \begin{macro}{\set@colroom}%
+% \begin{macro}{\set@vsize}%
+% \begin{macro}{\set@adj@colht}%
+% The story of \cmd\textheight, \cmd\@colht, \cmd\@colroom, and \cmd\vsize.
+%
+% \cmd\textheight---height of the text column. Not a running parameter, however, each time a page is
+% shipped out, the \cmd\textheight\ could in principle be altered. This must be done before
+%
+% \cmd\@colht---\cmd\textheight\ minus the height of any full-page-width floats. The latter are committed
+% only just after shipping out, and only if we are in a multicolumn page grid.
+% Therefore, \cmd\@colht\ should be set after a \cmd\shipout\ (by \cmd\@outputpage) and
+% will be adjusted when full-page-width floats are committed to the fresh page by \cmd\do@startpage.
+%
+% \cmd\@colroom---\cmd\@colht\ (adjusted by \cmd\pagesofar) minus the height of any column-width floats.
+% The latter are committed anywhere on the page, at which point \cmd\@colroom\ must be adjusted.
+% Therefore, \cmd\@colroom\ should be set (by \cmd\set@colroom) whenever a column is prepared (by ).
+%FIXME: committed (by \cmd\output@column@) and
+% will be adjusted (by \cmd\@add@float\ or \cmd\do@startcolumn) whenever a float is committted to the column.
+%
+% \cmd\vsize---\cmd\@colroom.
+% Therefore, \cmd\vsize\ should be set (by \cmd\set@vsize) whenever
+% the \cmd\@colroom\ is set (by \cmd\set@colroom) or adjusted (by \cmd\@add@float\ or \cmd\do@startcolumn)
+%FIXME: or when the \cmd\pagesofar\ box is changed (after invoking \cmd\open@column@).
+%
+% Question: what if there are committed floats? Footnotes?
+% Answer: full-page-width floats are only committed at top, and they are already reckoned with in \cmd\@colht.
+% Column-width committed floats are incorporated by \cmd\@makecolumn.
+%
+% As to footnotes, our scheme is to keep the \cmd\footins\ insert register up to date, and to use the insert mechanism
+% to ensure room for footnotes.
+% When a change is made to the page grid, the footnotes will need to be propagated back into the MVL.
+%
+% Note: FIXME: adjusting for \cmd\pagesofar\ is done at not quite the right time. I need to reexamine \cmd\set@colht,
+% because \cmd\@dbltoplist\ and \cmd\pagesofar\ really should be on the same footing.
+% Perhaps \cmd\@colht\ and \cmd\@colroom\ should both deal with their respective ``lists'' in the same way?
+%
+% These concerns will be particularly germane if we ever extend this package to deal with full-page-width floats
+% placed at the bottom of the page, or committed on the same page as called out.
+%
+% It occurs to me that we should ditch \cmd\set@colroom\ and only ever execute \cmd\set@colht,
+% which sets \cmd\@colroom\ as a side effect.
+% If so, we can make \cmd\@colht\ take \cmd\pagesofar\ into account, as it should. Then \cmd\@colht\ will return to its
+% original significance as the value that \cmd\@colroom\ is set to after a column is committed.
+%
+% On the other hand, why not simply forget all this caching and (re-)calculate \cmd\vsize\ as late as possible?
+% Particularly, \cmd\@colht\ is an artifact of the old way of doing things, where once it was set, it would never change.
+%
+% \changes{4.1n}{2009/12/02}{More diagnostics of column balancing}
+% \begin{macrocode}
+\def\set@colht{%
+ \set@adj@textheight\@colht
+ \global\let\enlarge@colroom\@empty
+ \set@colroom
+}%
+\def\set@adj@textheight#1{%
+ \ltxgrid@info@sw{\class@info{\string\set@adj@textheight\string#1}\saythe\textheight}{}%
+ #1\textheight
+ \def\@elt{\adj@page#1}%
+ \@booleantrue\firsttime@sw\@dbltoplist
+ \let\@elt\relax
+ \global#1#1\relax
+ \ltxgrid@info@sw{\saythe#1}{}%
+}%
+\def\set@colroom{%
+ \ltxgrid@info@sw{\class@info{\string\set@colroom}}{}%
+ \set@adj@colht\@colroom
+ \@if@empty\enlarge@colroom{}{%
+ \global\advance\@colroom\enlarge@colroom\relax
+ \ltxgrid@info@sw{\saythe\@colroom}{}%
+ }%
+ \@ifdim{\@colroom>\topskip}{}{%
+ \ltxgrid@info{Not enough room: \string\@colroom=\the\@colroom; increasing to \the\topskip}%
+ \@colroom\topskip
+ }%
+ \global\@colroom\@colroom
+%<ignore> \ltxgrid@info@sw{\class@info{\string\set@colroom\string\vsize=\string\colroom}\saythe\vsize}{}%
+ \set@vsize
+}%
+%
+\def\set@vsize{%
+ \global\vsize\@colroom
+ \ltxgrid@info@sw{\class@info{\string\set@vsize\string\vsize=\string\colroom}\saythe\vsize}{}%
+}%
+% \end{macrocode}
+% \changes{4.1n}{2009/12/02}{More diagnostics of column balancing}
+% \begin{macrocode}
+\def\set@adj@colht#1{%
+ #1\@colht
+ \ltxgrid@info@sw{\class@info{\string\set@adj@colht\string#1-\string\pagesofar}\saythe#1}{}%
+ \@ifvoid\pagesofar{}{%
+ \advance#1-\ht\pagesofar\advance#1-\dp\pagesofar
+ \ltxgrid@info@sw{\class@info{\string\pagesofar}\saythe#1}{}%
+ }%
+ \def\@elt{\adj@column#1}%
+ \@booleantrue\firsttime@sw\@toplist
+ \@booleantrue\firsttime@sw\@botlist
+ \let\@elt\relax
+}%
+\def\adj@column#1#2{%
+ \advance#1-\ht#2%
+ \advance#1-\firsttime@sw{\textfloatsep\@booleanfalse\firsttime@sw}{\floatsep}%
+ \ltxgrid@info@sw{\class@info{\string\adj@column\string#1-\string#2}\saythe#1}{}%
+}%
+\def\adj@page#1#2{%
+ \advance#1-\ht#2%
+ \advance#1-\firsttime@sw{\dbltextfloatsep\@booleanfalse\firsttime@sw}{\dblfloatsep}%
+ \ltxgrid@info@sw{\class@info{\string\adj@page\string#1-\string#2}\saythe#1}{}%
+}%
+\def\set@adj@box#1#2{%
+ \@ifvoid#2{}{%
+ \advance#1-\ht#2\advance#1-\dp#2%
+ \@booleantrue\temp@sw
+ \ltxgrid@foot@info@sw{\class@info{\string\set@adj@box\string#2}\saythe#1}{}%
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\@outputpage@tail}%
+% \changes{4.1b}{2008/08/04}{Procedure \cs{@outputpage@tail} tailpatches \cs{@outputpage}}%
+% In \cmd\@outputpage@tail, we set \cmd\@colht\ and the float placement parameters
+% (this is the one point where it is appropriate to set \cmd\@colht).
+% At \cmd\do@startpage\ time, we adjust \cmd\@colht's value to reflect committed
+% full-page-width floats.
+%
+% Note: with a correctly written output routine, a call to \cmd\@outputpage\ will inevitably be
+% followed by a call to \cmd\do@startpage, so these procedure calls would be unneeded.
+% \begin{macrocode}
+\appdef\@outputpage@tail{%
+ \set@colht % FIXME: needed?
+ \@floatplacement % FIXME: needed?
+ \@dblfloatplacement % FIXME: needed?
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{balance@2}
+% \changes{4.1b}{2008/08/04}{Procedure \cs{balance@2} defined more transparently}%
+% We define procedures for balancing columns in a multicolumn layout.
+% For now, we define only one: a procedure for the two-column grid.
+% All others will simply \cmd\relax\ out.
+%
+% The following code defines \verb+\balance@2+ without all the clunky \cmd\csname\ commands
+% in the replacement part,
+% which appears on the right-hand side of the assignment to \cmd\toks@.
+%
+% The method is straightforward: balance the two columns of text, and balance the footnotes.
+% Later on, \cmd\@combineinserts\ will be called to place the footnotes after the now-balanced columns.
+%
+% \changes{4.1f}{2009/07/14}{(AO, 519) \cs{footins} content must be preserved and reintegrated}
+% It was necessary to deal with the case where \cmd\box\cmd\footsofar\ was not empty
+% upon execution of this balancing code. We store it away in \cmd\box\cmd\footins\ and
+% add it back in afterwards.
+%
+% Here is a conundrum: if we switch between single-, two-, and three-column page grids:
+% On what measure should the footnotes be set?
+% \begin{macrocode}
+\begingroup
+ \catcode`\1=\cat@letter
+ \catcode`\2=\cat@letter
+% \end{macrocode}
+% \cmd\toks@\ contains the repacement part for an effective \cmd\def\cs{balance@2}.
+% \begin{macrocode}
+ \toks@{%
+% \end{macrocode}
+% \cmd\balance@two, by side effect, strips footnotes into \cmd\box\cmd\footins.
+% \begin{macrocode}
+ \setbox\footins\box\footsofar
+ \balance@two\col@1\@outputbox
+% \end{macrocode}
+% We ensure that the box assignments are global.
+% \begin{macrocode}
+ \global\setbox\col@1\box\col@1
+ \global\setbox\@outputbox\box\@outputbox
+% \end{macrocode}
+% The following line puts all footnotes into the footnote galley, \cmd\footsofar.
+% \begin{macrocode}
+ \combine@foot@inserts\footsofar\footins
+ }%
+ \aftergroup\def\aftergroup\balance@2\expandafter
+\endgroup\expandafter{\the\toks@}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\balance@two}
+% The procedure \cmd\balance@two\ takes two columns and balances them; in the process it removes
+% any footnotes that may be present to a place of safety \cmd\footsofar,
+% for later placement at the foot of the shipped-out page.
+% The box register \cmd\box\cmd\@ne\ is the aggregate of all columns.
+% The box register \cmd\box\cmd \z@\ is the last column.
+% The box register \cmd\box\cmd\tw@\ is the first column.
+% The \cmd\dimen\ register \cmd\dimen@\ is the trial value to \cmd\vsplit\ to,
+% initially half the height of \cmd\box\cmd\@ne.
+% The \cmd\dimen\ register \cmd\dimen@i\ is the increment for the next trial;
+% its initial value is equal to the initial value of \cmd\dimen@.
+% The \cmd\dimen\ register \cmd\dimen@ii\ is the difference of the heights of the two columns.
+%
+% The procedure uses a binary search for that value of \cmd\dimen@\ which is stable to within .5\cmd\p@\ and which
+% makes the last column be shorter than the others.
+%
+% This procedure can be extended to multiple columns simply by changing it to execute \cmd\vsplit\ multiple
+% times (one less than the total number of columns in the page layout) and to calculating \cmd\dimen@ii\
+% to be the difference of the heights of last column and the \cmd\dimen@.
+% Upon termination of the search, one would excute the \cmd\vsplit s once again, this time
+% using the actual \cmd\col@\ box registers to store the
+% balanced columns, thereby clobbering their former contents.
+%
+% Bug Note:
+% as originally written, this macro had a bug, which is well worth avoiding under similar circumstances anywhere.
+% So, learn from the mistakes of others, as they say.
+% In trying to remove the depth of the boxes created via \cmd\vsplit\ within the \cmd\loopwhile\ control,
+% I originally coded
+% \cmd\unvbox
+% \cmd\z@\
+% \cmd\setbox
+% \cmd\z@
+% \cmd\lastbox\
+% \cmd\dimen@
+% \cmd\dp
+% \cmd\z@\
+% \cmd\box
+% \cmd\z@\
+% \cmd\vskip-%
+% \cmd\dimen@.
+% The error here is that the (horizontal) shift of the last box in the vertical list will be lost in the process.
+% Simply put, \cmd\setbox\cmd\z@\cmd\lastbox\ fails to retain the shift of the box node in the vertical list,
+% and when it is put down again via \cmd\box\cmd\z@, it will no longer have the correct shift.
+%
+% This bug affected things placed in the MVL with \cmd\moveleft, \cmd\moveright, \cmd\parshape, and
+% \cmd\hangindent, as well as things shifted by \TeX's primitive mechanisms.
+%
+% A superior strategy for removing the depth of the last line of the list is more expensive, but safer:
+% make a separate copy of the list, measure the depth of the last box as above, but then discard
+% the list, retaining only the value of the dimension.
+%
+% Note that this procedure will not work if the material within is excessively chunky.
+% A particular failure mode exists where none of the material is allocated to the last (right) column.
+% We detect this case and revert to unbalanced columns.
+%
+% Another failure mode is where a large chunk occurs at the beginning of the composite box.
+% In this case, the left column may fill up even when \cmd\dimen@\ is very small.
+% If this configuration leaves the left column longer than the right, then we are done,
+% but \cmd\dimen@\ by no means represents the height of either finished box.
+%
+% Therefore the last step in the process is to rebox the two columns to a common height determined
+% independently of the balancing process.
+%
+% The dimension involved is checked against the current \cmd\@colroom\ to guard against the case where
+% excessive material happens to fall in either column.
+% \changes{4.1b}{2008/08/04}{Change \cs{set@colroom} to \cs{set@colht}}
+% \changes{4.1n}{2009/12/02}{More diagnostics of column balancing}
+% \changes{4.1n}{2010/01/02}{(AO, 571) coding convention: use \cs{bgroup}, \cs{egroup} (instead of braces) when a box is being built}%
+% \changes{4.1n}{2010/01/02}{(AO, 571) Footnotes, when columns are balanced or when they are composed with their column}%
+% \changes{4.1n}{2010/01/02}{(AO, 571) Change \cs{balance@two}'s balancing algorithm to more successfully balance extremely short columns.}
+% \begin{macrocode}
+\def\balance@two#1#2{%
+ \ltxgrid@info@sw{\class@info{\string\balance@two\string#1\string#2}}{}%
+ \outputdebug@sw{\trace@scroll{\showbox#1\showbox#2}}{}%
+% \end{macrocode}
+% The first step is to recover the footnotes from the bottoms of the two columns (globally, into \cmd\footsofar)
+% and to combine the text into \cmd\box\cmd\@ne, but without voiding either of the argument boxes.
+% \begin{macrocode}
+ \setbox\thr@@\copy\footsofar
+ \setbox\@ne\vbox\bgroup
+ \@ifvoid{#1}{}{%
+ \recover@column#1\footsofar\column@recovered\footins@recovered
+ \@ifvoid{#2}{}{\marry@baselines}%
+ }%
+ \@ifvoid{#2}{}{%
+ \recover@column#2\footsofar\column@recovered\footins@recovered
+ }%
+ \egroup
+ \outputdebug@sw{\trace@scroll{\showbox\@ne}}{}%
+ \ltxgrid@foot@info@sw{\trace@scroll{\showbox\footsofar}}{}%
+% \end{macrocode}
+% Hereunder, \cmd\dimen@\ is the split value. We adjust it until the step size is small enough, while the split is acceptable.
+% Also, \cmd\dimen@i\ is the step size. Once this value is greater than a half point, we must iterate.
+% \begin{macrocode}
+ \dimen@\ht\@ne\divide\dimen@\tw@
+ \dimen@i\dimen@
+ \vbadness\@M
+ \vfuzz\maxdimen
+ \splittopskip\topskip
+ \loopwhile{%
+ \setbox\z@\copy\@ne\setbox\tw@\vsplit\z@ to\dimen@
+ \remove@depth\z@\remove@depth\tw@
+% \end{macrocode}
+% The following line would provide a diagnostic of the iterations of column balancing, were we to use it.
+% \begin{verbatim}
+% \outputdebug@sw{\trace@scroll{\showbox\tw@\showbox\z@}}{}%
+% \end{verbatim}
+% Hereunder, \cmd\dimen@ii\ is used to reckon the difference in height between the left box and the right.
+% \begin{macrocode}
+ \dimen@ii\ht\tw@\advance\dimen@ii-\ht\z@
+ \dimen@i=.5\dimen@i
+ \ltxgrid@info@sw{\saythe\dimen@\saythe\dimen@i\saythe\dimen@ii}{}%
+% \end{macrocode}
+% If the columns are within a half-point of each other,
+% \begin{macrocode}
+ \@ifdim{\dimen@ii<.5\p@}{%
+ \@ifdim{\dimen@ii>-.5\p@}%
+ }{%
+ \false@sw
+ }%
+% \end{macrocode}
+% The above results in a Boolean, which now chooses between the following two brace-delimited clauses.
+% If the step size is less than a half-point, then terminate the loop.
+% \begin{macrocode}
+ {%
+ \true@sw
+ }{%
+ \@ifdim{\dimen@i<.5\p@}%
+ }%
+% \end{macrocode}
+% The above results in a Boolean, which now chooses between the following two brace-delimited clauses.
+% The true-part terminates the loop, otherwise iterate.
+% \begin{macrocode}
+ {%
+ \false@sw
+ }%
+ {%
+% \end{macrocode}
+% For the next iteration, the candidate split dimension \cmd\dimen@\ will be one step larger
+% if the height of the left box is less than that of the right box.
+% Otherwise it will be one step smaller.
+% \begin{macrocode}
+ \advance\dimen@\@ifdim{\dimen@ii<\z@}{}{-}\dimen@i
+ \true@sw
+ }%
+ }%
+% \end{macrocode}
+% The loop has terminated.
+% \begin{macrocode}
+ \ltxgrid@info@sw{\saythe\dimen@\saythe\dimen@i\saythe\dimen@ii}{}%
+% \end{macrocode}
+% The algorithm has failed to find a satisfactory result if
+% the left column is of non-zero height and the right column is of zero height.
+% \begin{macrocode}
+ \@ifdim{\ht\z@=\z@}{%
+ \@ifdim{\ht\tw@=\z@}%
+ }{%
+ \true@sw
+ }%
+% \end{macrocode}
+% The \cmd\false@sw\ branch is executed if the algorithm has failed.
+% We restore the original boxes.
+% \begin{macrocode}
+ {%
+ }{%
+ \ltxgrid@info{Unsatifactorily balanced columns: giving up}%
+ \setbox\tw@\box#1%
+ \setbox\z@ \box#2%
+ \global\setbox\footsofar\box\thr@@
+ }%
+ \setbox\tw@\vbox{\unvbox\tw@\vskip\z@skip}%
+ \setbox\z@ \vbox{\unvbox\z@ \vskip\z@skip}%
+ \set@colht
+ \dimen@\ht\z@\@ifdim{\dimen@<\ht\tw@}{\dimen@\ht\tw@}{}%
+ \@ifdim{\dimen@>\@colroom}{\dimen@\@colroom}{}%
+ \ltxgrid@info@sw{\saythe{\ht\z@}\saythe{\ht\tw@}\saythe\@colroom\saythe\dimen@}{}%
+ \setbox#1\vbox to\dimen@{\unvbox\tw@\unskip\raggedcolumn@skip}%
+ \setbox#2\vbox to\dimen@{\unvbox\z@ \unskip\raggedcolumn@skip}%
+ \outputdebug@sw{\trace@scroll{\showbox#1\showbox#2}}{}%
+}%
+% \end{macrocode}
+% Procedure \cmd\remove@depth\ rearranges the given (vertical) box register so that it has zero depth.
+% \begin{macrocode}
+\def\remove@depth#1{%
+ \setbox#1\vbox\bgroup
+ \unvcopy#1%
+ \setbox\z@\vbox\bgroup
+ \unvbox#1%
+ \setbox\z@\lastbox
+ \aftergroup\kern\aftergroup-\expandafter
+ \egroup
+ \the\dp\z@\relax
+ \egroup
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \changes{4.1n}{2010/01/02}{(AO, 571) Abandon \cs{recover@footins} in favor of \cs{recover@column}}%
+% Procedure \cmd\recover@column\ is a utility to separate a column box into text and footnotes;
+% the former being contributed to the current (vertical) list, the latter appended to the given register, usually \cmd\footsofar.
+%
+% Argument \verb+#1+ is the input: it should be a \cmd\vbox, and it remains unaltered.
+% Argument \verb+#2+ is the box into which to (globally) add the footnotes, usually \cmd\footsofar.
+% Arguments \verb+#3+ and \verb+#4+ are scratch box registers to use in this calculation.
+% As a side effect, \verb+#3+ will be unboxed into whatever vertical mode we are in at the moment (should be a \cmd\vbox).
+% \begin{macrocode}
+\def\recover@column#1#2#3#4{%
+ \ltxgrid@info@sw{\class@info{\string\recover@column\string#1\string#2\string#3\string#4}}{}%
+ \setbox#4\vbox{\unvcopy#1}%
+ \ltxgrid@foot@info@sw{\trace@scroll{\showbox#4}}{}%
+ \dimen@\ht#4%
+ \ltxgrid@foot@info@sw{\saythe\dimen@}{}%
+ \setbox#4\vbox\bgroup
+ \unvbox#4\unskip
+% \end{macrocode}
+% We now strip the footnotes from the bottom of this box, adding them to \cmd\footsofar.
+% The method relies on a signal, consisting of a complementary pair of kerns, placed at the bottom of the box by \cmd\@combineinserts.
+% \begin{macrocode}
+ \dimen@i\lastkern\unkern\advance\dimen@i\lastkern
+ \@ifdim{\dimen@i=\z@}{%
+ \dimen@i\lastkern\unkern
+ \ltxgrid@foot@info@sw{\saythe\dimen@i}{}%
+ \aftergroup\dimen@i
+ \expandafter\egroup\the\dimen@i\relax
+ }{%
+ \egroup
+ }%
+% \end{macrocode}
+% Split the column into \verb+#3+ and the footnote into \verb+#4+.
+% Append the footnote to \verb+#2+.
+% \begin{macrocode}
+ \@ifdim{\dimen@i<\z@}{%
+ \advance\dimen@\dimen@i
+ \ltxgrid@foot@info@sw{\saythe\dimen@i\saythe\dimen@}{}%
+ \splittopskip\z@skip
+ \global\setbox#3\vsplit#4 to\dimen@
+ \global\setbox#4\vbox{\unvbox#4}%
+ \ltxgrid@foot@info@sw{\trace@scroll{\showbox#1\showbox#2\showbox#3\showbox#4}}{}%
+ \global\setbox#2\vbox\bgroup\unvbox#2\vskip\z@skip\unvbox#4\egroup
+ }{%
+% \end{macrocode}
+% What if \cmd\dimen@i\ is zero?
+% In that case, \cmd\setbox\#3\cmd\box\#4, and do not touch \cmd\box\#2.
+% \begin{macrocode}
+ \setbox#3\box#4%
+ \ltxgrid@foot@info@sw{\trace@scroll{\showbox#1\showbox#2\showbox#3\showbox#4}}{}%
+ }%
+ \unvbox#3%
+ \loopwhile{\dimen@\lastskip\@ifdim{\dimen@>\z@}{\unskip\true@sw}{\false@sw}}%
+}%
+\def\recover@column@null#1#2#3#4{%
+ \unvcopy#1%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@begindocumenthook}
+% Initialization:
+% we initialize to the page grid named ``one''.
+% If the class decides to initially set type in a different grid, it
+% should execute these same commands, but changing the first to the appropriate procedure.
+%
+% Note that the point where this sequence is executed would be an excellent place to arrange for
+% floats to be committed to the first page of a document.
+% That is, we execute \cmd\do@startpage, which triggers \cmd\do@startcolumn.
+%
+% FIXME: it should be the job of the page grid to determine the procedure to execute at
+% the start of the job. Make this a hook.
+% \begin{macrocode}
+\rvtx@ifformat@geq{2020-10-01}%
+ {%
+ \AddToHook{begindocument}{%
+ \open@column@one\@ne
+ \set@colht
+ \@floatplacement
+ \@dblfloatplacement
+ }%
+ }{%
+ \prepdef\@begindocumenthook{%
+ \open@column@one\@ne
+ \set@colht
+ \@floatplacement
+ \@dblfloatplacement
+ }%
+ }
+% \end{macrocode}
+% \end{macro}
+%
+% Comment: our technique of balancing columns is severely limited, because it cannot properly work
+% with \env{longtable}, which places material at the bottom and top of the column break.
+%
+% The proper way to handle a grid change in the middle of the page is to accumulate all the material for
+% an entire article (or chapter) and then assemble finished pages therefrom. This approach is fundamentally
+% superior for complex layouts: it corresponds to real-world workflows.
+% Such a scheme is an excellent subject for another \LaTeX\ package.
+%
+%
+% \subsection{Patches for the longtable package}%
+%
+% \LaTeX's ``required'' package \classname{longtable} (written by David P. Carlilsle),
+% which is part of /latex/required/tools, is incmpatible with both
+% \LaTeX's ``required'' package \classname{multicol} and with
+% \LaTeX's native \cmd\twocolumn\ capability. There is no essential reason
+% for this incompatability, aside from implementation details, and the
+% \classname{ltxgrid} package gives us the ability to lift them.
+%
+% Only four of \classname{longtable}'s procedures require rewriting:
+% \cmd\longtable,
+% \cmd\endlongtable,
+% \cmd\LT@start, and
+% \cmd\LT@end@hd@ft.
+% The procedure \cmd\switch@longtable\ checks against their expected meanings
+% and, if all is as expected, applies the patches.
+% In the process, we simplify things considerably and also make them more
+% secure.
+%
+% Why does \classname{longtable} need to access the output routine, anyway?
+% What it comes down to, is what happens when a pagebreak falls within a
+% long table. If this happens, we would like to append a row at the bottom of
+% the broken table and add a row at the top of the next page.
+%
+% These things can be accomodated easily by the \classname{ltxgrid} output
+% routine hooks.
+%
+% \begin{macro}{\longtable}
+% \begin{macrocode}
+\def\longtable@longtable{%
+ \par
+ \ifx\multicols\@undefined\else\ifnum\col@number>\@ne\@twocolumntrue\fi\fi
+ \if@twocolumn\LT@err{longtable not in 1-column mode}\@ehc\fi
+ \begingroup
+ \@ifnextchar[\LT@array{\LT@array[x]}%
+}%
+\def\longtable@new{%
+ \par
+ \@ifnextchar[\LT@array{\LT@array[x]}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\endlongtable}
+% \begin{macrocode}
+\def\endlongtable@longtable{%
+ \crcr
+ \noalign{%
+ \let\LT@entry\LT@entry@chop
+ \xdef\LT@save@row{\LT@save@row}}%
+ \LT@echunk
+ \LT@start
+ \unvbox\z@
+ \LT@get@widths
+ \if@filesw
+ {\let\LT@entry\LT@entry@write\immediate\write\@auxout{%
+ \gdef\expandafter\noexpand
+ \csname LT@\romannumeral\c@LT@tables\endcsname
+ {\LT@save@row}}}%
+ \fi
+ \ifx\LT@save@row\LT@@save@row
+ \else
+ \LT@warn{Column \@width s have changed\MessageBreak
+ in table \thetable}%
+ \LT@final@warn
+ \fi
+ \endgraf\penalty -\LT@end@pen
+ \endgroup
+ \global\@mparbottom\z@
+ \pagegoal\vsize
+ \endgraf\penalty\z@\addvspace\LTpost
+ \ifvoid\footins\else\insert\footins{}\fi
+}%
+% \end{macrocode}
+% \changes{4.1a}{2008/06/29}{Repair error in \cs{endlongtable@new} involving \cs{@ifx}: argument not delimited.}
+% \begin{macrocode}
+\def\endlongtable@new{%
+ \crcr
+ \noalign{%
+ \let\LT@entry\LT@entry@chop
+ \xdef\LT@save@row{\LT@save@row}%
+ }%
+ \LT@echunk
+ \LT@start
+ \unvbox\z@
+ \LT@get@widths
+ \@if@sw\if@filesw\fi{%
+ {%
+ \let\LT@entry\LT@entry@write
+ \immediate\write\@auxout{%
+ \gdef\expandafter\noexpand\csname LT@\romannumeral\c@LT@tables\endcsname
+ {\LT@save@row}%
+ }%
+ }%
+ }{}%
+ \@ifx{\LT@save@row\LT@@save@row}{}{%
+ \LT@warn{%
+ Column \@width s have changed\MessageBreak in table \thetable
+ }\LT@final@warn
+ }%
+ \endgraf
+ \nobreak
+ \box\@ifvoid\LT@lastfoot{\LT@foot}{\LT@lastfoot}%
+ \global\@mparbottom\z@
+ \endgraf
+ \LT@post
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\LT@start}
+% \begin{macrocode}
+\def\LT@start@longtable{%
+ \let\LT@start\endgraf
+ \endgraf\penalty\z@\vskip\LTpre
+ \dimen@\pagetotal
+ \advance\dimen@ \ht\ifvoid\LT@firsthead\LT@head\else\LT@firsthead\fi
+ \advance\dimen@ \dp\ifvoid\LT@firsthead\LT@head\else\LT@firsthead\fi
+ \advance\dimen@ \ht\LT@foot
+ \dimen@ii\vfuzz
+ \vfuzz\maxdimen
+ \setbox\tw@\copy\z@
+ \setbox\tw@\vsplit\tw@ to \ht\@arstrutbox
+ \setbox\tw@\vbox{\unvbox\tw@}%
+ \vfuzz\dimen@ii
+ \advance\dimen@ \ht
+ \ifdim\ht\@arstrutbox>\ht\tw@\@arstrutbox\else\tw@\fi
+ \advance\dimen@\dp
+ \ifdim\dp\@arstrutbox>\dp\tw@\@arstrutbox\else\tw@\fi
+ \advance\dimen@ -\pagegoal
+ \ifdim \dimen@>\z@\vfil\break\fi
+ \global\@colroom\@colht
+ \ifvoid\LT@foot\else
+ \advance\vsize-\ht\LT@foot
+ \global\advance\@colroom-\ht\LT@foot
+ \dimen@\pagegoal\advance\dimen@-\ht\LT@foot\pagegoal\dimen@
+ \maxdepth\z@
+ \fi
+ \ifvoid\LT@firsthead\copy\LT@head\else\box\LT@firsthead\fi
+% \end{macrocode}
+% At some point before version 4.11, the \cmd\nobreak\ was added.
+% \begin{macrocode}
+\nobreak
+% \end{macrocode}
+%
+% \begin{macrocode}
+ \output{\LT@output}%
+}%
+\def\LT@start@new{%
+ \let\LT@start\endgraf
+ \endgraf
+ \markthr@@{}%
+ \LT@pre
+ \@ifvoid\LT@firsthead{\LT@top}{\box\LT@firsthead\nobreak}%
+ \mark@envir{longtable}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\LT@end}
+% \begin{macrocode}
+\def\LT@end@hd@ft@longtable#1{%
+ \LT@echunk
+ \ifx\LT@start\endgraf
+ \LT@err{Longtable head or foot not at start of table}{Increase LTchunksize}%
+ \fi
+ \setbox#1\box\z@
+ \LT@get@widths\LT@bchunk
+}%
+\def\LT@end@hd@ft@new#1{%
+ \LT@echunk
+ \@ifx{\LT@start\endgraf}{%
+ \LT@err{Longtable head or foot not at start of table}{Increase LTchunksize}%
+ }%
+ \global\setbox#1\box\z@
+ \LT@get@widths
+ \LT@bchunk
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\LT@array}
+%
+%
+% \begin{macrocode}
+\def\LT@array@longtable[#1]#2{%
+ \refstepcounter{table}\stepcounter{LT@tables}%
+ \if l#1%
+ \LTleft\z@ \LTright\fill
+ \else\if r#1%
+ \LTleft\fill \LTright\z@
+ \else\if c#1%
+ \LTleft\fill \LTright\fill
+ \fi\fi\fi
+ \let\LT@mcol\multicolumn
+ \let\LT@@tabarray\@tabarray
+ \let\LT@@hl\hline
+ \def\@tabarray{%
+ \let\hline\LT@@hl
+ \LT@@tabarray}%
+ \let\\\LT@tabularcr\let\tabularnewline\\%
+ \def\newpage{\noalign{\break}}%
+ \def\pagebreak{\noalign{\ifnum`}=0\fi\@testopt{\LT@no@pgbk-}4}%
+ \def\nopagebreak{\noalign{\ifnum`}=0\fi\@testopt\LT@no@pgbk4}%
+ \let\hline\LT@hline \let\kill\LT@kill\let\caption\LT@caption
+ \@tempdima\ht\strutbox
+ \let\@endpbox\LT@endpbox
+ \ifx\extrarowheight\@undefined
+ \let\@acol\@tabacol
+ \let\@classz\@tabclassz \let\@classiv\@tabclassiv
+ \def\@startpbox{\vtop\LT@startpbox}%
+ \let\@@startpbox\@startpbox
+ \let\@@endpbox\@endpbox
+ \let\LT@LL@FM@cr\@tabularcr
+ \else
+ \advance\@tempdima\extrarowheight
+ \col@sep\tabcolsep
+ \let\@startpbox\LT@startpbox\let\LT@LL@FM@cr\@arraycr
+ \fi
+ \setbox\@arstrutbox\hbox{\vrule
+ \@height \arraystretch \@tempdima
+ \@depth \arraystretch \dp \strutbox
+ \@width \z@}%
+ \let\@sharp##\let\protect\relax
+ \begingroup
+ \@mkpream{#2}%
+ \xdef\LT@bchunk{%
+ \global\advance\c@LT@chunks\@ne
+ \global\LT@rows\z@\setbox\z@\vbox\bgroup
+ \LT@setprevdepth
+% \end{macrocode}
+% At some point before version 4.11, the \cmd\noexpand\ was added.
+% We need not change our own version, because we did it right, back in 1998 (using \cmd\appdef).
+% \begin{macrocode}
+ \tabskip\LTleft \noexpand\halign to\hsize\bgroup
+ \tabskip\z@ \@arstrut \@preamble \tabskip\LTright \cr}%
+ \endgroup
+ \expandafter\LT@nofcols\LT@bchunk&\LT@nofcols
+ \LT@make@row
+ \m@th\let\par\@empty
+ \everycr{}\lineskip\z@\baselineskip\z@
+ \LT@bchunk}%
+\def\LT@LR@l{\LTleft\z@ \LTright\fill}%
+\def\LT@LR@r{\LTleft\fill \LTright\z@ }%
+\def\LT@LR@c{\LTleft\fill \LTright\fill}%
+\def\LT@array@new[#1]#2{%
+ \refstepcounter{table}\stepcounter{LT@tables}%
+ \table@hook
+ \LTleft\fill \LTright\fill
+ \csname LT@LR@#1\endcsname
+ \let\LT@mcol\multicolumn
+ \let\LT@@hl\hline
+ \prepdef\@tabarray{\let\hline\LT@@hl}%
+ \let\\\LT@tabularcr
+ \let\tabularnewline\\%
+ \def\newpage{\noalign{\break}}%
+ \def\pagebreak{\noalign{\ifnum`}=0\fi\@testopt{\LT@no@pgbk-}4}%
+ \def\nopagebreak{\noalign{\ifnum`}=0\fi\@testopt\LT@no@pgbk4}%
+ \let\hline\LT@hline
+ \let\kill\LT@kill
+ \let\caption\LT@caption
+ \@tempdima\ht\strutbox
+ \let\@endpbox\LT@endpbox
+ \@ifxundefined\extrarowheight{%
+ \let\@acol\@tabacol
+ \let\@classz\@tabclassz
+ \let\@classiv\@tabclassiv
+ \def\@startpbox{\vtop\LT@startpbox}%
+ \let\@@startpbox\@startpbox
+ \let\@@endpbox\@endpbox
+% \end{macrocode}
+% Because \classname{ltxutil} patches \LaTeX's \cmd\@tabularcr and \cmd\@xtabularcr,
+% we must restore these procedures in the scope of \env{longtable}.
+% Ironically, the patches in \classname{ltxutil} were for the purpose of extending
+% the \env{tabular} environment to prevent pagebreaks with the *-form of \cmd\\,
+% just the same as is being done here. But the two mechanisms conflict.
+% \changes{4.1a}{2008/06/29}{Change \cs{LT@array@new}: restore \cs{@tabularcr} and \cs{@xtabularcr}}
+% \begin{macrocode}
+ \let\LT@LL@FM@cr\@tabularcr@LaTeX
+ \let\@xtabularcr\@xtabularcr@LaTeX
+ }{%
+ \advance\@tempdima\extrarowheight
+ \col@sep\tabcolsep
+ \let\@startpbox\LT@startpbox
+% \end{macrocode}
+% \changes{4.1a}{2008/06/29}{Change \cs{LT@array@new}: set \cs{LT@LL@FM@cr} to \cs{@arraycr@array} instead of \cs{@arraycr}}
+% \begin{macrocode}
+ \let\LT@LL@FM@cr\@arraycr@array
+ }%
+%
+ \let\@acoll\@tabacoll
+ \let\@acolr\@tabacolr
+ \let\@acol\@tabacol
+%
+ \setbox\@arstrutbox\hbox{%
+ \vrule
+ \@height \arraystretch \@tempdima
+ \@depth \arraystretch \dp \strutbox
+ \@width \z@
+ }%
+ \let\@sharp##%
+ \let\protect\relax
+ \begingroup
+ \@mkpream{#2}%
+ \@mkpream@relax
+ \edef\@preamble{\@preamble}%
+ \prepdef\@preamble{%
+ \global\advance\c@LT@chunks\@ne
+ \global\LT@rows\z@
+ \setbox\z@\vbox\bgroup
+ \LT@setprevdepth
+ \tabskip\LTleft
+ \halign to\hsize\bgroup
+ \tabskip\z@
+ \@arstrut
+ }%
+ \appdef\@preamble{%
+ \tabskip\LTright
+ \cr
+ }%
+ \global\let\LT@bchunk\@preamble
+ \endgroup
+ \expandafter\LT@nofcols\LT@bchunk&\LT@nofcols
+ \LT@make@row
+ \m@th
+ \let\par\@empty
+ \everycr{}%
+ \lineskip\z@
+ \baselineskip\z@
+ \LT@bchunk
+}%
+\appdef\table@hook{}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\switch@longtable}
+%
+% Here is the switch from standard \classname{longtable} to the new, \classname{ltxgrid}-compatible values.
+%
+% At this point, we extend \env{longtable} with a \env{longtable*} form, which signifies that we want to
+% use the full page width for setting the table.
+% You can think this way: \env{longtable*} is to \env{longtable} as \env{table*} is to \env{table}.
+%
+% \begin{macrocode}
+\def\switch@longtable{%
+ \@ifpackageloaded{longtable}{%
+ \@ifx{\longtable\longtable@longtable}{%
+ \@ifx{\endlongtable\endlongtable@longtable}{%
+ \@ifx{\LT@start\LT@start@longtable}{%
+ \@ifx{\LT@end@hd@ft\LT@end@hd@ft@longtable}{%
+ \@ifx{\LT@array\LT@array@longtable}{%
+ \true@sw
+ }{\false@sw}%
+ }{\false@sw}%
+ }{\false@sw}%
+ }{\false@sw}%
+ }{\false@sw}%
+ {%
+ \class@info{Patching longtable package}%
+ }{%
+ \class@info{Patching unrecognized longtable package. (Proceeding with fingers crossed)}%
+ }%
+ \let\longtable\longtable@new
+ \let\endlongtable\endlongtable@new
+ \let\LT@start\LT@start@new
+ \let\LT@end@hd@ft\LT@end@hd@ft@new
+ \let\LT@array\LT@array@new
+ \newenvironment{longtable*}{%
+ \onecolumngrid@push
+ \longtable
+ }{%
+ \endlongtable
+ \onecolumngrid@pop
+ }%
+% \end{macrocode}
+% Removed obsolete code.
+% \begin{macrocode}
+ }{}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\LT@pre}
+% \begin{macro}{\LT@bot}
+% \begin{macro}{\LT@top}
+% \begin{macro}{\LT@post}
+% \begin{macro}{\LT@adj}
+% Note that at the end of the longtable environment, we reestablish the \cmd\mark@envir\ of the
+% containing environment. We have left \cmd\curr@envir\ alone, so this will work.
+% \begin{macrocode}
+\def\LT@pre{\penalty\z@\vskip\LTpre}%
+\def\LT@bot{\nobreak\copy\LT@foot\vfil}%
+\def\LT@top{\copy\LT@head\nobreak}%
+\def\LT@post{\penalty\z@\addvspace\LTpost\mark@envir{\curr@envir}}%
+\def\LT@adj{%
+ \setbox\z@\vbox{\null}\dimen@-\ht\z@
+ \setbox\z@\vbox{\unvbox\z@\LT@bot}\advance\dimen@\ht\z@
+ \global\advance\vsize-\dimen@
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{output@init}
+% \begin{macro}{output@prep}
+% \begin{macro}{output@post}
+% \begin{macrocode}
+\def\output@init@longtable{\LT@adj}%
+\def\output@prep@longtable{\setbox\@cclv\vbox{\unvbox\@cclv\LT@bot}}%
+\def\output@post@longtable{\LT@top}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+%
+% \subsection{Patches for index processing}%
+%
+% Another feature that uses the output routine hooks occurs within
+% an index, where one wishes to apply a ``continue head'' when a
+% column breaks within a primary index entry.
+% Some book designs call for the continue head to only be applied
+% at a turnpage break.
+%
+% In any case, it is easy enough for \cmd\output@post@theindex\
+% to do this in conjunction with component marks.
+% Only the bare outlines are shown here.
+%
+% \begin{macro}{\output@init}
+% \begin{macro}{\output@prep}
+% \begin{macro}{\output@post}
+% \begin{macrocode}
+\let\output@init@theindex\@empty
+\let\output@prep@theindex\@empty
+\def\output@post@theindex{%
+ \@ifodd\c@page{}{%
+ \@ifnum{\pagegrid@cur=\@ne}{%
+% \end{macrocode}
+% We have the leftmost column of a verso page:
+% Insert the current top-level continued head.
+% \begin{macrocode}
+ }%
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+%
+% \subsection{Checking the auxiliary file}%
+%
+% We relegate the checking of the auxiliary file to the output routine.
+% This task must wait until the last page is shipped out, because otherwise
+% the stream might get closed before the last page is shipped out.
+% Obviously, we must use \cmd\do@output@MVL\ for the job.
+%
+% \begin{macro}{\check@aux}
+% \begin{macrocode}
+\def\check@aux{\do@output@MVL{\do@check@aux}}%
+% \end{macrocode}
+% \end{macro}
+%
+%
+%
+% \subsection{Dealing with stuck floats and stalled float dequeueing}%
+%
+% \LaTeX's float placement mechanism is fundamentally flawed, as evidenced by
+% its warning message ``too many unprocessed floats'', which users understandably find frustrating.
+% The \classname{ltxgrid} package provides tools for ameliorating the situation somewhat.
+%
+% Two cases require detection and rectification:
+% \begin{enumerate}
+% \item
+% A float is ``stuck'' in the \cmd\@deferlist: for whatever reason, the float fails to be committed,
+% even at the start of a fresh page.
+% Once this condition prevails, following floats can never be committed, subsequently all of \LaTeX's
+% float registers are used up.
+%
+% If this condition is detected, we reconsider float dequeueing under permissive (\cmd\clearpage-style) processing.
+%
+% \item
+% The \cmd\@freelist\ is exhausted:
+% a large concentration of floats, say, uses up all of \LaTeX's float registers all at once.
+% This condition commonly occurs when the user collects floats at the end of the document, for some reason.
+%
+% When a float is encountered, \LaTeX\ uses a float register (allocated from a pool of free registers) to contain it until it can be placed.
+% However, no further action is taken until the pagebuilder is visited, so floats can accumulate.
+% Also, even after the pagebuilder is visited, deferred floats can accumulate, and these are not committed
+% until a column (or page) of text is completed.
+%
+% Once the last free float register is used, action should be taken that will commit some of the deferred floats,
+% even if this might require ending the page right where we are (resulting in a short page).
+%
+% Perhaps, committed floats should be stored using some mechanism other than a list, as is currently done.
+% A feasible alternative storage method would be to use a \cmd\box\ register in place of
+% \cmd\@toplist,
+% \cmd\@botlist, and
+% \cmd\@dbltoplist.
+% This is probably just fine, since such committed floats are not reconsidered (I think).
+%
+% \end{enumerate}
+%
+% The emergency processing implemented here immediately ends the current page and begins to output float pages under (\cmd\clearpage-style) rules.
+% It proceeds until all deferred floats have been flushed.
+%
+% Users should expect non-optimal page makeup under these circumstances.
+%
+% Note that there is a weakness in our approach that we have not attempted to repair: if floats are being
+% added as part of a paragraph, we will not be able to take these remedial steps until the paragraph ends.
+% This means that the approach implemented here cannot fix all \LaTeX\ documents. Users can still construct
+% documents that exhaust \LaTeX's pool of float registers!
+%
+% \begin{macro}{\check@deferlist@stuck}
+% \begin{macro}{\@outputpage@tail}
+% \changes{4.1b}{2008/08/04}{Procedure \cs{@outputpage@tail} tailpatches \cs{@outputpage}}%
+% We detect the case where, at the start of a fresh page, there are deferred floats, but none are
+% committed. We memorize the \cmd\@deferlist\ at \cmd\shipout\ time, then examine it at the point where
+% our efforts to commit floats to the new page are complete.
+% If it has not changed, the first float must be stuck, and we
+% attempt to fix things via \cmd\force@deferlist@stuck.
+%
+% This simple approach is comp[letely effective in for typical documents.
+%
+% Note that we try to avoid an infinite loop by examining the value of \cmd\clearpage@sw:
+% if we come here with that boolean true, we are in a loop.
+% \begin{macrocode}
+\def\check@deferlist@stuck#1{%
+ \@ifx{\@deferlist@postshipout\@empty}{}{%
+ \@ifx{\@deferlist@postshipout\@deferlist}{%
+ \@fltstk
+ \clearpage@sw{%
+ \ltxgrid@warn{Deferred float stuck during \string\clearpage\space processing}%
+ }{%
+ \force@deferlist@stuck#1%
+ }%
+ }{%
+% \end{macrocode}
+% We have successfully committed float(s)
+% \begin{macrocode}
+ }%
+ \global\let\@deferlist@postshipout\@empty
+ }%
+}%
+\def\@fltstk{%
+ \@latex@warning{A float is stuck (cannot be placed without \string\clearpage)}%
+}%
+\appdef\@outputpage@tail{%
+ \global\let\@deferlist@postshipout\@deferlist
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\@next}
+% \begin{macro}{\@xnext}
+% We rewrite the \LaTeX\ kernel macros that dequeue float registers from, e.g., \cmd\@deferlist,
+% providing a test for the condition where the pool of free registers is about to underflow.
+%
+% In this case, we attempt to fix things via \cmd\force@deferlist@empty.
+% \begin{macrocode}
+\def\@next#1#2{%
+ \@ifx{#2\@empty}{\false@sw}{%
+ \expandafter\@xnext#2\@@#1#2%
+ \true@sw
+ }%
+}%
+\def\@xnext\@elt#1#2\@@#3#4{%
+ \def#3{#1}%
+ \gdef#4{#2}%
+ \def\@tempa{#4}\def\@tempb{\@freelist}%
+ \@ifx{\@tempa\@tempb}{%
+ \@ifx{#4\@empty}{%
+ \force@deferlist@empty%{Float register pool exhausted}%
+ }{}%
+ }{}%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\force@deferlist@stuck}
+% \begin{macro}{\force@deferlist@empty}
+% \begin{macro}{\force@deferlist@sw}
+% \begin{macro}{\do@forcecolumn@pen}
+% \begin{macro}{\do@forcecolumn}
+% The procedure \cmd\force@deferlist@empty\ is an attempt to rectify a situation where \LaTeX's float placement mechanism
+% may fail (``too many unprocessed floats'').
+%
+% We put down interrupts that call for the float placement to be redone, but under permissive conditions,
+% just the same as if \cmd\clearpage\ had been invoked.
+%
+% Note that the attempt to rectify the error is contingent on the setting of \cmd\force@deferlist@sw,
+% default false. A document class using this package that wishes to enable this error recovery mechanism should
+% set this boolean to true.
+%
+% The interrupt \cmd\do@forcecolumn@pen, which invokes the procedure \cmd\do@forcecolumn,
+% does the same as \cmd\do@startcolumn, except under permissive conditions:
+% we are trying to empty out the float registers completely.
+%
+% In order to properly with the case where there is material in \cmd\box\cmd\@cclv,
+% \cmd\@toplist, \cmd\@botlist, \cmd\@dbltoplist, etc,
+% we do what amounts to \cmd\newpage\ to get things rolling.
+%
+% In \cmd\force@deferlist@stuck, we take advantage of already being in the output routine:
+% simply reinvoke \cmd\do@startcolumn\ under permissive conditions.
+%
+% \begin{macrocode}
+\def\force@deferlist@stuck#1{%
+ \force@deferlist@sw{%
+ \@booleantrue\clearpage@sw
+ \@booleantrue\forcefloats@sw
+ #1%
+ }{%
+ }%
+}%
+\def\force@deferlist@empty{%
+ \force@deferlist@sw{%
+ \penalty-\pagebreak@pen
+ \protect@penalty\do@forcecolumn@pen
+ }{%
+ }%
+}%
+\@booleanfalse\force@deferlist@sw
+\mathchardef\do@forcecolumn@pen=10009
+\@namedef{output@-\the\do@forcecolumn@pen}{\do@forcecolumn}%
+\def\do@forcecolumn{%
+ \@booleantrue\clearpage@sw
+ \@booleantrue\forcefloats@sw
+% \end{macrocode}
+% \begin{verbatim}
+%\unvbox\@cclv
+%\vfil
+%\penalty-\pagebreak@pen
+% \end{verbatim}
+% \begin{macrocode}
+ \do@startcolumn
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% A more thorough revision of \LaTeX's float placement mechanism would involve substituting a single \cmd\box\
+% register for the \cmd\@deferlist. This way, \LaTeX's ability to have latent floats would be limited by
+% box memory alone.
+%
+% Because only the \cmd\box\ and \cmd\count\ components of the float box register are actually used by \LaTeX,
+% our scheme can be accomplished if we can find a way to encode the information held in the \cmd\count\ component.
+%
+% A first-in, first-out mechanism exists, wherein a box-penalty pair is dequeued by \cmd\lastbox\cmd\lastpenalty\cmd\unpenalty\ and enqueued
+% by \cmd\setbox\cmd\foo=\cmd\hbox\cmd\bgroup\cmd\penalty\cmd\floatpenalty\cmd\box\cmd\floatbox\cmd\unhbox\cmd\foo\cmd\egroup.
+%
+% Note that this scheme is made possible by our change to \LaTeX's float placement mechanism,
+% wherein we consolidated the two \cmd\@deferlist s into one.
+%
+% \section{Support for legacy \LaTeX\ commands}
+%
+% We provide support for the \cmd\enlargethispage\ command.
+%
+% Note: using a command of this sort does not automatically enlarge both pages of a spread, which would be the convention in page composition.
+%
+% Timing Note: In a multicolumn page grid, the user should issue the \cmd\enlargethispage\ command
+% while the first column of the page is being typeset.
+% We provide a helpful message if the timing is wrong.
+%
+% This code can serve as a model for introducing commands that need to execute within the safety of the output routine.
+% We ensure that the arguments are fully expanded, then execute \cmd\do@output@MVL\ to cause an output procedure,
+% \cmd\@@enlargethispage, to execute. When it does execute, the MVL will be exposed.
+%
+% The \cmd\@@enlargethispage\ procedure simply adjusts the vertical dimensions of the page.
+% The adjustment will persist until the column is committed, at which point the page dimension
+% will revert to its standard value.
+% \begin{macrocode}
+\def\enlargethispage{%
+ \@ifstar{%
+ \@enlargethispage{}%
+ }{%
+ \@enlargethispage{}%
+ }%
+}%
+\def\@enlargethispage#1#2{%
+ \begingroup
+ \dimen@#2\relax
+ \edef\@tempa{#1}%
+ \edef\@tempa{\noexpand\@@enlargethispage{\@tempa}{\the\dimen@}}%
+ \expandafter\do@output@MVL\expandafter{\@tempa}%
+ \endgroup
+}%
+\def\@@enlargethispage#1#2{%
+ \def\@tempa{one}%
+ \@ifx{\thepagegrid\@tempa}{%
+ \true@sw
+ }{%
+ \def\@tempa{mlt}%
+ \@ifx{\thepagegrid\@tempa}{%
+ \@ifnum{\pagegrid@cur=\@ne}{%
+% \end{macrocode}
+% OK to adjust this page
+% \begin{macrocode}
+ \gdef\enlarge@colroom{#2}%
+ \true@sw
+ }{%
+% \end{macrocode}
+% Can only adjust this column; give up
+% \begin{macrocode}
+ \ltxgrid@warn{Too late to enlarge this page; move the command to the first column.}%
+ \false@sw
+ }%
+ }{%
+% \end{macrocode}
+% Unknown page grid
+% \begin{macrocode}
+ \ltxgrid@warn{Unable to enlarge a page of this kind.}%
+ \false@sw
+ }%
+ }%
+ {%
+ \class@info{Enlarging page \thepage\space by #2}%
+ \global\advance\@colroom#2\relax
+ \set@vsize
+ }{%
+% \end{macrocode}
+% Could not adjust this page
+% \begin{macrocode}
+ }%
+}%
+\let\enlarge@colroom\@empty
+% \end{macrocode}
+% The \cmd\@kludgeins\ insert register is now unneeded.
+% Ensure that packages using this mechanism break (preferrable to subtle bugs).
+% \begin{macrocode}
+\let\@kludgeins\@undefined
+% \end{macrocode}
+%
+% \subsubsection{Building the page for shipout}
+%
+% \begin{macro}{\@outputpage@head}
+% We set \cmd\@outputpage@head\ to make the \cmd\@outputbox\ be of fixed height.
+% \changes{4.1b}{2008/08/04}{Procedure \cs{@outputpage@head} headpatches \cs{@outputpage}}%
+% \begin{macrocode}
+\@booleantrue\textheight@sw
+\prepdef\@outputpage@head{%
+ \textheight@sw{%
+ \count@\vbadness\vbadness\@M
+ \dimen@\vfuzz\vfuzz\maxdimen
+ \setbox\@outputbox\vbox to\textheight{\unvbox\@outputbox}%
+ \vfuzz\dimen@
+ \vbadness\count@
+ }{}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@outputpage@head}
+% For compatibility with David Carlisle's \classname{lscape} package, we need to allow the
+% \cmd\LS@rot\ procedure to mung \cmd\@outputbox.
+%
+% Implementation note: the \classname{lscape} package effectively tailpatches two \LaTeX\ internals to accomplish its purpose,
+% an approach that is not robust. It is more robust to headpatch \cmd\@outputpage, which is what we do here.
+% \changes{4.1o}{2010/02/02}{(AO, 576) Allow \classname{lscape} to act on \cs{@outputbox} at the right time}%
+% \begin{macrocode}
+\appdef\@outputpage@head{%
+ \@ifx{\LS@rot\@undefined}{}{\LS@rot}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \subsubsection{Warning message}
+%
+% \begin{macro}{\ltxgrid@info}
+% \begin{macro}{\ltxgrid@warn}
+% Something has happened that the user might be interested in.
+% Print a message to the log, but only if the user selected the verbose option.
+% \begin{macrocode}
+\def\ltxgrid@info{%
+ \ltxgrid@info@sw{\class@info}{\@gobble}%
+}%
+\@booleanfalse\ltxgrid@info@sw
+\def\ltxgrid@warn{%
+ \ltxgrid@warn@sw{\class@warn}{\@gobble}%
+}%
+\@booleantrue\ltxgrid@warn@sw
+\@booleanfalse\ltxgrid@foot@info@sw
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \section{Line-wise processing}%
+% Sometimes we wish to process each line of type that will be placed into the galley,
+% for example, applying line numbering to a document.
+% To accomplish the task, we have to force a visit to the output routine after each such line, whereupon
+% we can process it accordingly (in the case of line numbering, we could do as \filename{lineno.sty} and
+% append an appropriately formed box to the MVL).
+%
+% In implementing such a scheme, we will have to instantiate interrupts for the following cases:
+% \begin{description}
+%
+% \item[\cmd\interlinepenalty\ and friends] These include \cmd\clubpenalty, \cmd\widowpenalty, \cmd\displaywidowpenalty, and \cmd\brokenpenalty.
+%
+%
+% \item[Display math penalties] Includes \cmd\predisplaypenalty, \cmd\postdisplaypenalty, and \cmd\interdisplaylinepenalty.
+%
+%
+% \item[\cs{par}] The penalty following the last line of the paragraph.
+%
+%
+% \item[\cmd\vadjust] A trap for any \cmd\vadjust\ command that falls in the paragraph.
+%
+% \end{description}
+%
+% \begin{macro}{\def@next@handler}
+% \begin{macro}{\def@line@handler}
+% Utility procedures \cmd\def@next@handler\ and \cmd\def@line@handler\ help in the creation of
+% interrupt handlers.
+%
+% \cmd\def@next@handler\ increments the scratch count register (argument 1),
+% using this value to \cmd\mathchardef\ its second argument
+% as the negative of the flag value to be used as a penalty for exciting the interrupt (argument 3).
+% As a byproduct, it leaves the given scratch counter incremented.
+% \begin{macrocode}
+\def\def@next@handler#1#2#3{%
+ \advance#1\@ne\mathchardef#2\the#1%
+ \expandafter\def\csname output@-\the#1\endcsname{#3}%
+% \end{macrocode}
+% The following line is for diagnostic purposes.
+% \begin{verbatim}
+% \typeout{\string#2(\expandafter\string\csname output@\the#1\endcsname:\expandafter\meaning\csname output@\the#1\endcsname)}%
+% \end{verbatim}
+% \begin{macrocode}
+}%
+% \end{macrocode}
+% \cmd\def@line@handler\ uses \cmd\int@parpenalty\ as a base.
+% The interrupt is the sum of that base with the first argument,
+% and the handler is the second argument.
+% \begin{macrocode}
+\def\def@line@handler#1#2{%
+ \begingroup
+ \@tempcnta\int@parpenalty
+ \advance\@tempcnta-#1%
+% \end{macrocode}
+% The following line is for diagnostic purposes.
+% \begin{verbatim}
+% \typeout{Defining: \expandafter\string\csname output@\the\linenopenalty\endcsname}%
+% \end{verbatim}
+% \begin{macrocode}
+ \aftergroup\def
+ \expandafter\aftergroup\csname output@-\the\@tempcnta\endcsname
+ \endgroup{#2}%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\int@parpenalty}
+% \begin{macro}{\@handle@line@ltx}
+% \begin{macro}{\@@handle@line@ltx}
+% We first set \cmd\int@parpenalty\ to our chosen base value${}\le -11012$.
+% We then define all the handlers for lines within a paragraph, of which there are 12 different cases.
+% \begin{macrocode}
+\mathchardef\int@parpenalty11012
+\def@line@handler\z@{\@handle@line@ltx{}{}{}}%
+\def@line@handler\@ne{\@handle@line@ltx{}{}{\brokenpenalty@ltx}}%
+\def@line@handler\tw@{\@handle@line@ltx{}{\clubpenalty@ltx}{}}%
+\def@line@handler\thr@@{\@handle@line@ltx{\clubpenalty@ltx}{}{\brokenpenalty@ltx}}%
+\def@line@handler\f@ur{\@handle@line@ltx{\widowpenalty@ltx}{}{}}%
+\def@line@handler{5}{\@handle@line@ltx{\widowpenalty@ltx}{}{\brokenpenalty@ltx}}%
+\def@line@handler{6}{\@handle@line@ltx{\widowpenalty@ltx}{\clubpenalty@ltx}{}}%
+\def@line@handler{7}{\@handle@line@ltx{\widowpenalty@ltx}{\clubpenalty@ltx}{\brokenpenalty@ltx}}%
+\def@line@handler{8}{\@handle@line@ltx{\displaywidowpenalty@ltx}{}{}}%
+\def@line@handler{9}{\@handle@line@ltx{\displaywidowpenalty@ltx}{}{\brokenpenalty@ltx}}%
+\def@line@handler{10}{\@handle@line@ltx{\displaywidowpenalty@ltx}{\clubpenalty@ltx}{}}%
+\def@line@handler{11}{\@handle@line@ltx{\displaywidowpenalty@ltx}{\clubpenalty@ltx}{\brokenpenalty@ltx}}%
+% \end{macrocode}
+%
+% The default handler for lines within a paragraph simply
+% restores the value of the \cmd\penalty\ to the normal value.
+% If something more useful needs to be done, we can change the definition of \cmd\@@handle@line@ltx.
+% \begin{macrocode}
+\def\@handle@line@ltx#1#2#3{%
+ \@@handle@line@ltx
+ \@tempcnta\lastpenalty
+ \@tempcntb\interlinepenalty@ltx\relax
+ \@if@empty{#1}{}{\advance\@tempcntb#1\relax}%
+ \@if@empty{#2}{}{\advance\@tempcntb#2\relax}%
+ \@if@empty{#3}{}{\advance\@tempcntb#3\relax}%
+ \penalty\@ifnum{\@tempcnta<\@tempcntb}{\@tempcntb}{\@tempcnta}%
+}%
+\let\@@handle@line@ltx\@empty
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\int@postparpenalty}
+% \begin{macro}{\int@vadjustpenalty}
+% \begin{macro}{\int@whatsitpenalty}
+% \begin{macro}{\int@predisplaypenalty}
+% \begin{macro}{\int@interdisplaylinepenalty}
+% \begin{macro}{\int@postdisplaypenalty}
+% \begin{macro}{\@handle@display@ltx}
+% \begin{macro}{\@@handle@display@ltx}
+% \begin{macro}{\handle@par@ltx}
+% We herewith define all the handlers for cases relating to display math:
+% last line before a display math,
+% last line of a display math, and a line within a display math.
+% We also handle the last line of a paragraph, a whatsit node, and a \cmd\vadjust.
+% \begin{macrocode}
+\@tempcnta\int@parpenalty
+\def@next@handler\@tempcnta\int@postparpenalty{\reset@queues@ltx\handle@par@ltx}%
+\def@next@handler\@tempcnta\int@vadjustpenalty{\handle@vadjust@ltx}%
+\def@next@handler\@tempcnta\int@whatsitpenalty{\handle@whatsit@ltx}%
+\def@next@handler\@tempcnta\int@predisplaypenalty{\reset@queues@ltx\@handle@display@ltx{\predisplaypenalty@ltx}}%
+\def@next@handler\@tempcnta\int@interdisplaylinepenalty{\@handle@display@ltx{\interdisplaylinepenalty@ltx}}%
+\def@next@handler\@tempcnta\int@postdisplaypenalty{\@handle@display@ltx{\postdisplaypenalty@ltx}}%
+% \end{macrocode}
+% The default handler for display math lines simply
+% restores the value of the \cmd\penalty\ to the normal value.
+% If something more useful needs to be done, we can change the definition of \cmd\@@handle@display@ltx.
+% \begin{macrocode}
+\def\@handle@display@ltx#1{%
+ \@@handle@display@ltx
+ \@tempcnta\lastpenalty
+ \@tempcntb#1%
+ \penalty\@ifnum{\@tempcnta<\@tempcntb}{\@tempcntb}{\@tempcnta}%
+}%
+\let\@@handle@display@ltx\@empty
+% \end{macrocode}
+%
+% We provide stub definitions for the handlers for the last line of a paragraph, a \cmd\vadjust, and a whatsit node (e.g., \cmd\write, \cmd\special).
+% There is no canonical penalty for such cases.
+% \begin{macrocode}
+\def\handle@par@ltx{}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% Note that a whatsit needs to be handled differently from a \cmd\vadjust:
+% a whatsit node does not affect the (crucial) depth of \cmd\box\cmd\@cclv, while the more
+% general \cmd\vadjust\ may cause any kind of vertical mode material to be interposed just below the line
+% we are trying to trap, in particular \cmd\vskip s and \cmd\penalty s.
+%
+% \begin{macro}{\set@linepenalties}
+% \begin{macro}{\restore@linepenalties}
+% \begin{macro}{\set@displaypenalties}
+% Now we define utility procedures that set up for a paragraph to be broken into lines,
+% restoring the penalties afterwards.
+%
+% Utility procedure \cmd\set@linepenalties\ systematically sets the penalties of paragraph breaking
+% to flag values, meanwhile storing away the normal values for access by the output routine.
+% \begin{macrocode}
+\def\set@linepenalties{%
+ \expandafter\def\expandafter\interlinepenalty@ltx\expandafter{\the\interlinepenalty}%
+ \interlinepenalty-\int@parpenalty
+ \expandafter\def\expandafter\brokenpenalty@ltx\expandafter{\the\brokenpenalty}%
+ \brokenpenalty\@ne
+ \expandafter\def\expandafter\clubpenalty@ltx\expandafter{\the\clubpenalty}%
+ \clubpenalty\tw@
+ \expandafter\def\expandafter\widowpenalty@ltx\expandafter{\the\widowpenalty}%
+ \widowpenalty\f@ur
+ \expandafter\def\expandafter\displaywidowpenalty@ltx\expandafter{\the\displaywidowpenalty}%
+ \displaywidowpenalty8\relax
+}%
+% \end{macrocode}
+%
+% Utility procedure \cmd\restore@linepenalties\ restores the values of the penalty parameters that were
+% modified by \cmd\set@linepenalties.
+% \begin{macrocode}
+\def\restore@linepenalties{%
+ \interlinepenalty\interlinepenalty@ltx
+ \brokenpenalty\brokenpenalty@ltx
+ \clubpenalty\clubpenalty@ltx
+ \widowpenalty\widowpenalty@ltx
+ \displaywidowpenalty\displaywidowpenalty@ltx
+ \relax
+}%
+% \end{macrocode}
+%
+% In the following, the first argument should be a boolean (either \cmd\true@sw\ or \cmd\false@sw).
+% \begin{macrocode}
+\def\set@displaypenalties#1{%
+ \expandafter\def\expandafter\predisplaypenalty@ltx\expandafter{\the\predisplaypenalty}%
+ \expandafter\def\expandafter\interdisplaylinepenalty@ltx\expandafter{\the\interdisplaylinepenalty}%
+ \expandafter\def\expandafter\postdisplaypenalty@ltx\expandafter{\the\postdisplaypenalty}%
+ \@ifhmode{\predisplaypenalty-\int@predisplaypenalty\relax}{}%
+ #1{\interdisplaylinepenalty-\int@interdisplaylinepenalty\relax}{}%
+ #1{\postdisplaypenalty-\int@postdisplaypenalty\relax}{}%
+}%
+% \end{macrocode}
+% We provide no procedure to restore the respective penalties, because they are altered within a group:
+% \TeX 's context stack will automatically restore things.
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\enqueue@whatsit@ltx}%
+% \begin{macro}{\handle@whatsit@ltx}%
+% \begin{macro}{\do@whatsit}%
+% \begin{macro}{\@g@pop@ltx}%
+% Here is a facility for dealing with whatsit nodes while we are trapping paragraph lines.
+% We simply enqueue a macro that will create the desired whatsit node, dequeueing it in the output routine.
+% \begin{macrocode}
+\def\enqueue@whatsit@ltx#1{%
+ \gappdef\g@whatsit@queue{{#1}}%
+ \vadjust{\penalty-\int@whatsitpenalty}%
+}%
+\def\handle@whatsit@ltx{%
+ \unvbox\@cclv
+ \g@pop@ltx\g@whatsit@queue\@tempa
+ \expandafter\do@whatsit\expandafter{\@tempa}%
+}%
+\def\do@whatsit#1{}%
+\def\g@pop@ltx#1#2{%
+ \expandafter\@g@pop@ltx#1{}{}\@@#1#2%
+}%
+\def\@g@pop@ltx#1#2\@@#3#4{%
+ \gdef#3{#2}%
+ \def#4{#1}%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\vspace}%
+% \begin{macro}{\pagebreak}%
+% \begin{macro}{\nopagebreak}%
+% \begin{macro}{\\}%
+% \begin{macro}{\@arrayparboxrestore}%
+% We wish to prevent \filename{lineno.sty} from patching \cmd\vspace\ and \cmd\pagebreak,
+% because that package does it through global assignments, which is prone to failure.
+%
+% We also wish to prevent that package from patching \cmd\@arrayparboxrestore, because
+% it prevents us from \cmd\unvbox ing vertical mode material into the MVL and numbering those
+% lines.
+%
+% We start by retaining the original definitions of these commands, so we can restore them
+% if \filename{lineno.sty} does get loaded.
+% \begin{macrocode}
+\let\vspace@ltx\vspace
+\let\pagebreak@ltx\pagebreak
+\let\nopagebreak@ltx\nopagebreak
+\let\endline@ltx\\
+\let\@arrayparboxrestore@ltx\@arrayparboxrestore
+% \end{macrocode}
+%
+% Next, we provide for line-wise processing by patching the procedures associated with these same three commands.
+%
+% Depending on the running version of \LaTeX, these commands use \cs{vskip} (before 2020-10-01), or
+% \cs{@vspace@calcify} to add \textsf{calc}-like support in the kernel. To avoid duplicating the tests,
+% define a macro \cs{@tempa} that takes the right form of the primitive spacing command and defines \LaTeX's
+% spacing macros accordingly.
+% \begin{macrocode}
+\def\@tempa#1{%
+% \end{macrocode}
+%
+% There are exactly four core \LaTeX\ procedures that use \cmd\vadjust\ to insert vertical mode material
+% into the main vertical list: \cmd\vspace, \cmd\pagebreak, \cmd\nopagebreak, and \cmd\\.
+% Other commands may use \cmd\vadjust, but they are inserting an interrupt (via a penalty${}<10000$),
+% and such a thing does not mask the depth of \cmd\box\cmd\@cclv, hence is permissible.
+%
+% In each case, we replace the core \LaTeX\ procedure with one that itself replaces \cmd\vadjust\ with
+% \cmd\ex@vadjust@ltx. The meaning of this procedure can be left as \cmd\vadjust, or it can be changed to
+% one that accomplishes the equivalent without masking the depth of \cmd\box\cmd\@cclv.
+%
+% The first procedure is \cmd\@vspace, here shown in original form and in the patched alternative form.
+% This procedure and \cmd\@vspacer\ implement the \cmd\vspace\ command.
+%
+% \begin{macrocode}
+\def\@vspace@org ##1{%
+ \ifvmode
+ #1% \vskip #1
+ \vskip\z@skip
+ \else
+ \@bsphack
+ \vadjust{\@restorepar
+ #1% \vskip #1
+ \vskip\z@skip
+ }%
+ \@esphack
+ \fi
+}%
+\def\@vspace@ltx##1{%
+ \@ifvmode{%
+ #1% \vskip #1
+ \vskip\z@skip
+ }{%
+ \@bsphack
+ \ex@vadjust@ltx{%
+ \@restorepar
+ \nobreak
+ #1% \vskip #1
+ \vskip\z@skip
+ }%
+ \@esphack
+ }%
+}%
+% \end{macrocode}
+%
+% The second procedure is \cmd\@vspacer.
+% \begin{macrocode}
+\def\@vspacer@org##1{%
+ \ifvmode
+ \dimen@\prevdepth
+ \hrule \@height\z@
+ \nobreak
+ #1%\vskip #1
+ \vskip\z@skip
+ \prevdepth\dimen@
+ \else
+ \@bsphack
+ \vadjust{\@restorepar
+ \hrule \@height\z@
+ \nobreak
+ #1%\vskip #1
+ \vskip\z@skip}%
+ \@esphack
+\fi
+}%
+\def\@vspacer@ltx##1{%
+ \@ifvmode{%
+ \dimen@\prevdepth
+ \hrule\@height\z@
+ \nobreak
+ #1%\vskip#1
+ \vskip\z@skip
+ \prevdepth\dimen@
+ }{%
+ \@bsphack
+ \ex@vadjust@ltx{%
+ \@restorepar
+ \hrule\@height\z@
+ \nobreak
+ #1%\vskip#1
+ \vskip\z@skip
+ }%
+ \@esphack
+ }%
+}%
+% \end{macrocode}
+%
+% Now define the macros above conditionally depending on the \LaTeX version.
+% \begin{macrocode}
+}
+\rvtx@ifformat@geq{2020/10/01}%
+ {\@tempa{\@vspace@calcify{#1}}}%
+ {\@tempa{\vskip #1 }}%
+% \end{macrocode}
+%
+% The procedure \cmd\@no@pgbk\ implements both \cmd\pagebreak\ and \cmd\nopagebreak.
+% \begin{macrocode}
+\def\@no@pgbk@org #1[#2]{%
+ \ifvmode
+ \penalty #1\@getpen{#2}%
+ \else
+ \@bsphack
+ \vadjust{\penalty #1\@getpen{#2}}%
+ \@esphack
+ \fi
+}%
+\def\@no@pgbk@ltx#1[#2]{%
+ \@ifvmode{%
+ \penalty#1\@getpen{#2}%
+ }{%
+ \@bsphack
+ \ex@vadjust@ltx{%
+ \penalty#1\@getpen{#2}%
+ }%
+ \@esphack
+ }%
+}%
+% \end{macrocode}
+%
+% The command to end a line of type, \cmd\\, is defined via \cmd\DeclareRobustCommand,
+% so we must proceed carefully:
+% A procedure is defined whose \cmd\long\cmd\csname\ is contructed via the incantation:
+% \cmd\csname\cmd\expandafter\cmd\@gobble\cmd\string\cmd\\ \cmd\endcsname.
+% Note the non-trivial space character after the \cmd\\: it is incorporated into
+% the \cmd\csname.
+%
+% Here is the original core \LaTeX\ definition for
+% the procedure involved, along with our revised version.
+% \changes{2020/09/30}{4.2d}{Conditionally use \cs{protected} or \cs{long} to track \LaTeX{} changes.}
+% \begin{macrocode}
+\rvtx@ifformat@geq{2020/02/02}%
+{\protected}{\long}\def\end@line@org{%
+ \let\reserved@e\relax
+ \let\reserved@f\relax
+ \@ifstar{%
+ \let\reserved@e\vadjust
+ \let\reserved@f\nobreak
+ \@xnewline
+ }%
+ \@xnewline
+}%
+\rvtx@ifformat@geq{2020/02/02}%
+{\protected}{\long}\def\end@line@ltx{%
+ \let\reserved@e\relax
+ \let\reserved@f\relax
+ \@ifstar{%
+ \let\reserved@e\ex@vadjust@ltx
+ \let\reserved@f\nobreak
+ \@xnewline
+ }{%
+ \@xnewline
+ }%
+}%
+% \end{macrocode}
+% An additional procedure requiring patching has the following original core \LaTeX\
+% definition; we modify it correspondingly.
+% \begin{macrocode}
+\def\@tempa#1{%
+ \def\@newline@org[##1]{%
+ \let\reserved@e\vadjust
+ \@gnewline{#1}% \vskip#1
+ }%
+ \def\@newline@ltx[##1]{%
+ \let\reserved@e\ex@vadjust@ltx
+ \@gnewline{#1}% \vskip#1
+ }%
+}
+\rvtx@ifformat@geq{2020/10/01}%
+ {\@tempa{\@vspace@calcify{#1}}}%
+ {\@tempa{\vskip #1}}%
+% \end{macrocode}
+% We now install our patches.
+% If some package overrides these macros, we will detect and complain.
+%
+% \begin{macrocode}
+ \@ifx{\@vspace\@vspace@org}{%
+ \@ifx{\@vspacer\@vspacer@org}{%
+ \@ifx{\@no@pgbk\@no@pgbk@org}{%
+ \@ifx{\@newline\@newline@org}{%
+ \expandafter\@ifx\expandafter{%
+ \csname\rvtx@ifformat@geq{2020/02/02}%
+ {\expandafter\@gobble\string\\}%
+ {\expandafter\@gobble\string\\ }\endcsname
+ \end@line@org
+ }{%
+ \true@sw
+ }{\false@sw}%
+ }{\false@sw}%
+ }{\false@sw}%
+ }{\false@sw}%
+ }{\false@sw}%
+ {%
+ \class@info{Overriding \string\@vspace, \string\@vspacer, \string\@no@pgbk, \string\@newline, and \string\\ }%
+ \let\@normalcr\end@line@ltx
+ \expandafter\let
+ \csname\rvtx@ifformat@geq{2020/02/02}%
+ {\expandafter\@gobble\string\\}%
+ {\expandafter\@gobble\string\\ }\endcsname\@normalcr
+ \let\@newline\@newline@ltx
+ \let\@vspace\@vspace@ltx
+ \let\@vspacer\@vspacer@ltx
+ \let\@no@pgbk\@no@pgbk@ltx
+ }{%
+ \class@warn{%
+ Failed to recognize \string\@vspace, \string\@vspacer, \string\@no@pgbk, \string\@newline, and \string\\;
+ no patches applied. Please get a more up-to-date class,
+ }%
+ }%
+% \end{macrocode}
+% Note that we have assigned the same meaning to \cmd\@normalcr, which is necessary to \LaTeX.
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\ex@vadjust@ltx}
+% \begin{macro}{\enqueue@vadjust@ltx}
+% \begin{macro}{\handle@vadjust@ltx}
+% \begin{macro}{\g@vadjust@line}
+% \begin{macro}{\reset@queues@ltx}
+% Here we give the default definition for \cmd\ex@vadjust@ltx\ along with
+% the definitions for the alternative version and its the associated handler.
+% \begin{macrocode}
+\let\ex@vadjust@ltx\vadjust
+\def\enqueue@vadjust@ltx#1{%
+ \gappdef\g@vadjust@queue{{#1}}%
+ \vadjust{\penalty-\int@vadjustpenalty}%
+}%
+\def\handle@vadjust@ltx{%
+ \unvbox\@cclv
+ \g@pop@ltx\g@vadjust@queue\@tempa
+ \expandafter\gappdef\expandafter\g@vadjust@line\expandafter{\@tempa}%
+}%
+\let\g@vadjust@line\@empty
+% \end{macrocode}
+% Procedure \cmd\reset@queues@ltx\ resets the whatsit queue and the \cmd\vadjust\ queues
+% to their empty state. This should be done whenever we leave horizontal mode and
+% complete the processing of these queues: upon executing, effectively, primitive \cmd\par\
+% or interrupting a paragraph with display math.
+% \begin{macrocode}
+\def\reset@queues@ltx{%
+ \global\let\g@whatsit@queue\@empty
+ \global\let\g@vadjust@queue\@empty
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \section{Patching the lineno.sty package}
+%
+% \filename{lineno.sty} is a \LaTeX\ package that applies line numbering to a document.
+% The basic method is to give \cmd\interlinepenalty\ and like penalties such a value as
+% to force a visit to the output routine, where the line of type is given its number.
+% In order to properly measure the depth of \cmd\box\cmd\@cclv, it defers \cmd\vadjust\
+% commands that may insert \cmd\vskip\ or \cmd\penalty\ nodes.
+%
+% The implementation of that package, however, manipulates \cmd\holdinginserts\ in a dangerous
+% way: outside the safety of the output routine. It also alters the meaning of \cmd\vadjust\
+% using global assignments. We patch its code to avoid these problems. The \filename{ltxgrid}
+% package already has the needed mechanisms in place to do these jobs correctly.
+%
+% The methods we use can accomodate any values of penalties like \cmd\clubpenalty, etc:
+% we do not make assumptions about the range of values these penalty parameters could take.
+%
+%
+% \begin{macro}{\linenomathWithnumbers}
+% \begin{macro}{\linenomathNonumbers}
+% \begin{macro}{\endlinenomath}
+% \begin{macro}{\linenumberpar}
+% Here are the definitions of procedures in \filename{lineno.sty}
+% that alter \cmd\holdinginserts. They are current as of version v4.41, 2005/11/02.
+% We patch them to avoid doing this: in ltxgrid-based classes like REVTeX, the output routine properly manages \cmd\holdinginserts, so packages should not attempt to do so.
+% Also, we will want \cmd\linenumberpar\ to set \cmd\interlinepenaly\ to dispatch to \cmd\MakeLineNo.
+%
+% \begin{macrocode}
+\newcommand\linenomathWithnumbers@LN{%
+ \ifLineNumbers
+ \ifnum\interlinepenalty>-\linenopenaltypar
+ \global\holdinginserts\thr@@
+ \advance\interlinepenalty \linenopenalty
+ \ifhmode
+ \advance\predisplaypenalty \linenopenalty
+ \fi
+ \advance\postdisplaypenalty \linenopenalty
+ \advance\interdisplaylinepenalty \linenopenalty
+ \fi
+ \fi
+ \ignorespaces
+}%
+\newcommand\linenomathNonumbers@LN{%
+ \ifLineNumbers
+ \ifnum\interlinepenalty>-\linenopenaltypar
+ \global\holdinginserts\thr@@
+ \advance\interlinepenalty \linenopenalty
+ \ifhmode
+ \advance\predisplaypenalty \linenopenalty
+ \fi
+ \fi
+ \fi
+ \ignorespaces
+}%
+\def\endlinenomath@LN{%
+ \ifLineNumbers
+ \global\holdinginserts\@LN@outer@holdins
+ \fi
+ \global\@ignoretrue
+}
+\def\linenumberpar@LN{%
+ \ifvmode \@@@par \else
+ \ifinner \@@@par \else
+ \xdef\@LN@outer@holdins{\the\holdinginserts}%
+ \advance \interlinepenalty \linenopenalty
+ \linenoprevgraf \prevgraf
+ \global \holdinginserts \thr@@
+ \@@@par
+ \ifnum\prevgraf>\linenoprevgraf
+ \penalty-\linenopenaltypar
+ \fi
+ \@LN@parpgbrk
+ \global\holdinginserts\@LN@outer@holdins
+ \advance\interlinepenalty -\linenopenalty
+ \fi
+ \fi
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\class@documenthook}
+% We patch only if we recognize the definitions of all the procedures we are to patch.
+% \begin{macrocode}
+\appdef\class@documenthook{%
+ \@ifpackageloaded{lineno}{%
+ \@ifx{\linenomathWithnumbers\linenomathWithnumbers@LN}{%
+ \@ifx{\linenomathNonumbers\linenomathNonumbers@LN}{%
+ \@ifx{\endlinenomath\endlinenomath@LN}{%
+ \@ifx{\linenumberpar\linenumberpar@LN}{%
+ \true@sw
+ }{\false@sw}%
+ }{\false@sw}%
+ }{\false@sw}%
+ }{\false@sw}%
+ {%
+ \class@info{Overriding lineo.sty, restoring output routine,}%
+% \end{macrocode}
+% We commence overriding the procedures of \filename{lineno.sty}.
+% \begin{macrocode}
+ \let\linenumberpar\linenumberpar@ltx
+ \let\endlinenomath\endlinenomath@ltx
+ \expandafter\let\csname endlinenomath*\endcsname\endlinenomath@ltx
+ \let\linenomathWithnumbers\linenomathWithnumbers@ltx
+ \let\linenomathNonumbers\linenomathNonumbers@ltx
+% \end{macrocode}
+% Override \filename{lineno.sty}'s equipment for \cmd\vadjust\ and \cmd\linelabel:
+% we have existing interrupts and handlers for these purposes.
+% \begin{macrocode}
+ \let\ex@vadjust@ltx\ex@vadjust@line
+ \let\@LN@postlabel\enqueue@whatsit@ltx
+ \let\do@whatsit\write@linelabel
+% \end{macrocode}
+% Redirect handlers to those provided by \filename{lineno.sty}, and
+% give an appropriate meaning to the respective headpatch within the handlers.
+% \begin{macrocode}
+ \let\handle@par@ltx\handle@par@LN
+ \let\@@handle@line@ltx\Make@LineNo@ltx
+ \let\@@handle@display@ltx\Make@LineNo@ltx
+% \end{macrocode}
+% Next, we undo the action taken by \filename{lineno.sty} wherein it
+% took over the output routine. Instead, we service \filename{lineno.sty}
+% existing equipment of \filename{ltxgrid}.
+% We also revert the core \LaTeX\ definitions of
+% \cmd\vspace, \cmd\pagebreak, \cmd\nopagebreak, and \cmd\\,
+% which that package takes over (we have our own ways of doing these things).
+% \begin{macrocode}
+ \output@latex{\natural@output}%
+ \let\vspace\vspace@ltx
+ \let\pagebreak\pagebreak@ltx
+ \let\nopagebreak\nopagebreak@ltx
+ \let\@arrayparboxrestore\@arrayparboxrestore@ltx
+ \let\\\endline@ltx
+% \end{macrocode}
+% \changes{4.1f}{2009/07/07}{(AO, 515) Prevent line numbering within a footnote}
+% When line numbering is in effect, we must avoid any attempt to number the lines of a footnote.
+% \begin{macrocode}
+ \appdef\set@footnotefont{%
+ \let\par\@@@par
+ \let\@@par\@@@par
+ }%
+% \end{macrocode}
+% At last, we detect if the \cmd\linenumbers\ command has already been given;
+% if so, we do its assignments again, because we have changed the meaning of \cmd\linenumberpar.
+% \begin{macrocode}
+ \@if@sw\ifLineNumbers\fi{%
+ \class@info{Reinvoke \string\linenumbers}%
+ \let\@@par\linenumberpar
+ \@ifx{\@par\linenumberpar@LN}{\let\@par\linenumberpar}{}%
+ \@ifx{\par\linenumberpar@LN}{\let\par\linenumberpar}{}%
+ }{%
+ \class@info{Line numbering not turned on yet}%
+ }%
+% \end{macrocode}
+% Here ends the ``true branch'' of the patch code.
+% \begin{macrocode}
+ }{%
+% \end{macrocode}
+% If the \filename{lineno.sty} package is loaded, but we fail to patch it, notify the user.
+% \begin{macrocode}
+ \class@warn{Failed to recognize lineno.sty procedures; no patches applied. Please get a more up-to-date class.}%
+ }%
+ }{%
+% \end{macrocode}
+% \filename{lineno.sty} is not loaded, so no patches are needed.
+% \begin{macrocode}
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\linenumberpar}
+% \begin{macro}{\@linenumberpar}
+% Procedure \cmd\linenumberpar\ takes the place of \cmd\par\ when line numbering is in effect;
+% It executes the \cmd\par\ primitive if we are in vertical mode.
+% Otherwise we are in horizontal mode in the MVL and wish to end the current paragraph,
+% or we have \cmd\unvbox ed material onto the MVL.
+%
+% \begin{macrocode}
+\def\linenumberpar@ltx{\@ifvmode{\@@@par}{\@linenumberpar}}%
+% \end{macrocode}
+%
+% Procedure \cmd\@linenumberpar
+% \begin{macrocode}
+\def\@linenumberpar{%
+% \end{macrocode}
+% Prepare for our trip into the output routine by
+% saving away the current value of \cmd\prevgraf.
+% \begin{macrocode}
+ \linenoprevgraf\prevgraf
+% \end{macrocode}
+% The following will be used in the output routine dispatcher
+% to sense that we came from here.
+% \begin{macrocode}
+ \set@linepenalties
+% \end{macrocode}
+% Finally, call primitive \cmd\par\ with the signal value of \cmd\interlinepenalty\ and friends.
+% \begin{macrocode}
+ \@@@par
+% \end{macrocode}
+% We are now in vertical mode.
+% If lines of type were contributed to the MVL (non-trivial paragraph),
+% we must force another trip into the output routine to apply
+% line numbering to the last line of the paragraph.
+% \begin{macrocode}
+ \@ifnum{\prevgraf>\linenoprevgraf}{
+ \penalty-\int@postparpenalty
+ }{}%
+% \end{macrocode}
+% Execute procedure \cmd\@LN@parpgbrk, which has been set up in the
+% output routine for us to invoke here.
+%
+% \begin{macrocode}
+ \@LN@parpgbrk
+% \end{macrocode}
+% To wrap things up, we restore the original value of \cmd\interlinepenalty\ and friends.
+%
+% Query: why not employ \TeX's context stack to do the restore?
+% Would there be something wrong with executing primitive \cmd\par\ within a group?
+% \begin{macrocode}
+ \restore@linepenalties
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\linenomathWithnumbers}
+% \begin{macro}{\linenomathNonumbers}
+% Here are the patched definitions for the commands enabling
+% line numbering in display math.
+% \begin{macrocode}
+\newcommand\linenomathWithnumbers@ltx{\@linenomathnumbers@ltx\true@sw}%
+\newcommand\linenomathNonumbers@ltx{\@linenomathnumbers@ltx\false@sw}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\@linenomathnumbers}
+% \begin{macro}{\endlinenomath}
+% We have just begun a display math, and any paragraph we are setting will now end.
+% We set all relevant penalties to interrupt values; in the visit to the output routine,
+% we will replace the penalty with its normal value.
+% \begin{macrocode}
+\def\@linenomathnumbers@ltx#1{%
+ \@if@sw\ifLineNumbers\fi{%
+ \set@linepenalties
+ \set@displaypenalties#1%
+ }{}%
+ \ignorespaces
+}%
+\def\endlinenomath@ltx{%
+ \global\@ignoretrue
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% We provide a handler for the last line of a paragraph.
+% \begin{macrocode}
+\def\handle@par@LN{%
+ \Make@LineNo@ltx
+% \end{macrocode}
+% After setting the line number, we arrange for an appropriate penalty to
+% be laid down after this visit to the output routine ends.
+%
+% Query: why not contribute the penalty right here in the visit to the output routine?
+% \begin{macrocode}
+ \@tempcnta\lastpenalty
+ \@ifnum{\@tempcnta=\z@}{}{%
+ \expandafter\gdef
+ \expandafter\@LN@parpgbrk
+ \expandafter{%
+ \expandafter\penalty
+ \the\@tempcnta
+% \end{macrocode}
+% When \cmd\@LN@parpgbrk\ is executed, it resets itself to the default value,
+% \cmd\@LN@screenoff@pen.
+%
+% Query: \cmd\@LN@screenoff@pen\ appears to try to restore the depth of the last box:
+% why is this being done outside the safety of the output routine?
+% \begin{macrocode}
+ \global\let\@LN@parpgbrk\@LN@screenoff@pen
+ }%
+ }%
+}%
+% \end{macrocode}
+%
+% \begin{macro}{\Make@LineNo}
+% The procedure \cmd\Make@LineNo\ sets the box containing the line number itself.
+% \begin{macrocode}
+\def\Make@LineNo@ltx{%
+ \@LN@maybe@normalLineNumber
+% \end{macrocode}
+% We measure the depth of \cmd\box\cmd\@cclv\ and unbox it.
+% At this point, it is crucial that that box have the same depth as that of the last
+% box within it.
+%
+% In the simple case, \cmd\box\cmd\@cclv\ is
+% a \cmd\vbox\ containing as its last box the \cmd\hbox\ of the paragraph we are processing.
+%
+% Query: under what circumstances will this \emph{not} be the case?
+% \begin{macrocode}
+ \boxmaxdepth\maxdimen\setbox\z@\vbox{\unvbox\@cclv}%
+ \@tempdima\dp\z@
+ \unvbox\z@
+% \end{macrocode}
+% Then we create the box with the line number, setting its height to zero.
+% \begin{macrocode}
+ \sbox\@tempboxa{\hb@xt@\z@{\makeLineNumber}}%
+ \ht\@tempboxa\z@
+% \end{macrocode}
+% With these preparations, we invoke \cmd\@LN@depthbox, which
+% lays that box down (with its depth appropriately set):
+% this procedure depends on our having set \cmd\@tempdima\ and \cmd\@tempboxa
+% (kinda kludgy way of passing arguments, really).
+% \begin{macrocode}
+ \@LN@depthbox
+% \end{macrocode}
+% Now increment the line number. I have relocated this token past \cmd\@LN@depthbox:
+% this may induce a bug, but I am going to hereby force the issue:
+% why split up the procedure that lays down boxes with a procedure that sets a register value?
+% \begin{macrocode}
+ \stepLineNumber
+% \end{macrocode}
+% Finally, execute the \cmd\vadjust s that fell within the line that we just handled.
+%
+% Note that \cmd\enqueue@vadjust@ltx\ had queued up all the \cmd\vadjust\ commands for the paragraph
+% into \cmd\g@vadjust@queue, laying down an (\cmd\int@vadjustpenalty) interrupt in each ones' place.
+% The interrupts associated with this line of the paragraph have now moved the tokens to \cmd\g@vadjust@line,
+% which we now expand and execute.
+% \begin{macrocode}
+ \g@vadjust@line
+ \global\let\g@vadjust@line\@empty
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\write@linelabel#1{%
+ \protected@write\@auxout{}{%
+ \string\newlabel{#1}{{\theLineNumber}{\thepage}{}{}{}}%
+ }%
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\ex@vadjust@line{%
+ \@if@sw\ifLineNumbers\fi{\enqueue@vadjust@ltx}{\vadjust}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% Note that the \cmd\linelabel\ commands use a mechanism different from \cmd\vadjust,
+% embodied in the procedure \cmd\enqueue@vadjust@ltx, wherein the
+% \cmd\write\ primitives are enqueued while the paragraph is being processed, each replaced with
+% an interrupt, then dequeued and executed by the interrupt handler, leaving a \cmd\write\ node
+% in place of the interrupt (just where the \cmd\vadjust 's vertical mode material would had been)
+% just below the box containing the line of type. This \cmd\write, like all whatsits,
+% does not affect the depth of \cmd\box\cmd\@cclv, unlike the case of general vertical mode material,
+% which could have interfered.
+%
+% \section{End of the \file{ltxgrid} {\sc docstrip} module}
+% Here ends the module.
+% \begin{macrocode}
+%</kernel>
+% \end{macrocode}
+%
+% \Finale
+% %Here ends the programmer's documentation.
+% \endinput
+%
+\endinput
+%%EOF
diff --git a/texmf/source/latex/revtex/ltxutil.dtx b/texmf/source/latex/revtex/ltxutil.dtx
new file mode 100644
index 0000000..4ce6464
--- /dev/null
+++ b/texmf/source/latex/revtex/ltxutil.dtx
@@ -0,0 +1,4449 @@
+% \iffalse meta-comment balanced on line 107
+% ltxutil.dtx: utilities package
+% Copyright (c) 2009 Arthur Ogawa
+%
+% Disclaimer
+% This file is distributed WITHOUT ANY WARRANTY;
+% without even the implied warranty of
+% MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
+% License
+% You may distribute this file under the conditions of the
+% LaTeX Project Public License 1.3c or later
+% (http://www.latex-project.org/lppl.txt).
+% ReadMe
+% For the documentation and more detailed instructions for
+% installation, typeset this document with \LaTeX.
+% Maintenance Status
+% This work has the LPPL maintenance status "maintained";
+% Current Maintainer of this work is Arthur Ogawa
+% changes for version 4.2d and 4.2e by Phelype Oleinik.
+%
+% This work consists of the main source file ltxutil.dtx
+% and the derived files
+% ltxutil.sty, ltxutil.pdf
+% Distribution:
+% CTAN:macros/latex/contrib/revtex/
+%
+% Unpacking:
+% tex ltxutil.dtx
+%
+% Documentation:
+% latex ltxutil.dtx; ...
+%
+% Program calls to get the documentation (example):
+% pdflatex ltxutil.dtx
+% makeindex -s gind.ist ltxutil.idx
+% makeindex -s gglo.ist -o ltxutil.gls ltxutil.glo
+% pdflatex ltxutil.dtx
+% makeindex -s gind.ist ltxutil.idx
+% pdflatex ltxutil.dtx
+%
+% Installation:
+% TDS:doc/latex/revtex/
+% TDS:source/latex/revtex/
+% TDS:tex/latex/revtex/
+%
+% Thanks, Heiko!
+% This method of letting a single .dtx file serve as both
+% documentation (via latex) and installer (via tex) follows
+% the example of Heiko Oberdiek. Thanks!
+%<*ignore>
+\begingroup
+ \def\x{LaTeX2e}%
+\expandafter\endgroup
+\ifcase
+ 0\expandafter\ifx\csname processbatchFile\endcsname\relax\else1\fi\ifx\fmtname\x\else 1\fi
+ \relax
+\else
+ \csname fi\endcsname
+%</ignore>
+%<*install>
+\input docstrip
+\preamble
+
+This is a generated file;
+altering it directly is inadvisable;
+instead, modify the original source file.
+See the URL in the file README-LTXUTIL.tex.
+
+License
+ You may distribute this file under the conditions of the
+ LaTeX Project Public License 1.3c or later
+ (http://www.latex-project.org/lppl.txt).
+
+ This file is distributed WITHOUT ANY WARRANTY;
+ without even the implied warranty of MERCHANTABILITY
+ or FITNESS FOR A PARTICULAR PURPOSE.
+
+\endpreamble
+\askforoverwritefalse
+\keepsilent
+ \generate{%
+ %{ignore}
+ \file{ltxutil.sty}{%
+ \from{ltxutil.dtx}{package,kernel}%
+ }%
+ \file{ltxutil.krn}{%
+ \from{ltxutil.dtx}{kernel}%
+ }%
+ }%
+\ifToplevel{
+\Msg{***********************************************************}
+\Msg{*}
+\Msg{* To finish the installation, please move}
+\Msg{* ltxutil.sty}
+\Msg{* into a directory searched by TeX;}
+\Msg{* in a TDS-compliant installation:}
+\Msg{* texmf/tex/macros/latex/revtex/.}
+\Msg{*}
+\Msg{* To produce the documentation,
+ run ltxutil.dtx through LaTeX.}
+\Msg{*}
+\Msg{* Happy TeXing}
+\Msg{***********************************************************}
+}
+\endbatchfile
+%</install>
+%<*ignore>
+\fi
+%</ignore>
+% \fi
+%
+% \GetFileInfo{ltxutil.dtx}
+%
+% \iffalse ltxdoc klootch
+%<*package>
+%%% @LaTeX-file{
+%%% filename = "ltxutil.dtx",
+%%% version = "4.2e",
+%%% date = "2020/10/03",
+%%% author = "Arthur Ogawa (mailto:arthur_ogawa at sbcglobal.net),
+%%% Phelype Oleinik (mailto:phelype.oleinik at latex-project.org),
+%%% commissioned by the American Physical Society. Minor changes by Mark Doyle for version 4.2a-c.
+%%% ",
+%%% copyright = "Copyright (C) 1999, 2009 Arthur Ogawa,
+%%% distributed under the terms of the
+%%% LaTeX Project Public License 1.3c, see
+%%% ftp://ctan.tug.org/macros/latex/base/lppl.txt
+%%% ",
+%%% address = "Arthur Ogawa,
+%%% USA",
+%%% telephone = "",
+%%% FAX = "",
+%%% email = "mailto colon arthur_ogawa at sbcglobal.net",
+%%% codetable = "ISO/ASCII",
+%%% keywords = "latex, page grid, main vertical list",
+%%% supported = "yes",
+%%% abstract = "utilities package",
+%%% }
+%</package>
+% \fi
+%
+% \iffalse ltxdoc klootch
+% The following references the \file{README-LTXUTIL} file,
+% which contains basic information about this package.
+% The contents of this file are generated when
+% you typeset the programmer's documentation.
+% Search on "{filecontents*}{README-LTXUTIL}" to locate it.
+% \fi\input{README-LTXUTIL}%
+%
+% \subsection{Bill of Materials}
+%
+% Following is a list of the files in this distribution arranged
+% according to provenance.
+%
+% \subsubsection{Primary Source}%
+% One single file generates all.
+%\begin{verbatim}
+%ltxutil.dtx
+%\end{verbatim}
+%
+% \subsubsection{Generated by \texttt{latex ltxutil.dtx}}%
+% Typesetting the source file under pdflatex
+% generates the readme and the documentation.
+%\begin{verbatim}
+%README-LTXUTIL ltxutil.pdf
+%\end{verbatim}
+%
+% \subsubsection{Generated by \texttt{tex ltxutil.dtx}}%
+% Typesetting this file with \TeX\ generates
+% the package file.
+%\begin{verbatim}
+%ltxutil.sty
+%\end{verbatim}
+%
+% \subsubsection{Auxiliary}%
+% The following are auxiliary files generated
+% in the course of running \LaTeX:
+% \begin{verbatim}
+%ltxutil.aux ltxutil.idx ltxutil.ind ltxutil.log ltxutil.toc
+% \end{verbatim}
+%
+% \section{Code common to all modules}%
+%
+% We want to require only one place in this file
+% where the version number is stated,
+% and we also want to ensure that the version
+% number is embedded into every generated file.
+%
+% Now we declare that
+% these files can only be used with \LaTeXe.
+% An appropriate message is displayed if
+% a different \TeX{} format is used.
+% \begin{macrocode}
+%<*doc|package>
+\NeedsTeXFormat{LaTeX2e}[1995/12/01]%
+%</doc|package>
+% \end{macrocode}
+% As desired, the following modules all
+% take common version information:
+% \begin{macrocode}
+%<kernel&!package&!doc>\typeout{%
+%<*package|doc>
+\ProvidesFile{%
+%</package|doc>
+%<*kernel|package|doc>
+ltxutil%
+%</kernel|package|doc>
+%<*doc>
+.dtx%
+%</doc>
+%<package>.sty%
+%<*package|doc>
+}%
+%</package|doc>
+% \end{macrocode}
+%
+% The following line contains, for once and for all,
+% the version and date information.
+% By various means, this information is reproduced
+% consistently in all generated files and in the
+% typeset documentation.
+% Give credit where due.
+% \begin{macrocode}
+%<*doc|package|kernel>
+%<version>
+ [2020/10/03 4.2e utilities package (portions licensed from W. E. Baxter web at superscript.com)]% \fileversion
+%</doc|package|kernel>
+%<kernel&!package&!doc>}%
+% \end{macrocode}
+%
+%
+% \section{The driver module \texttt{doc}}
+%
+% This module, consisting of the present section,
+% typesets the programmer's documentation,
+% generating the \file{README-LTXUTIL} as required.
+%
+% Because the only uncommented-out lines of code at the beginning of
+% this file constitute the \file{doc} module itself,
+% we can simply typeset the \file{.dtx} file directly,
+% and there is thus rarely any need to
+% generate the ``doc'' {\sc docstrip} module.
+% Module delimiters are nonetheless required so that
+% this code does not find its way into the other modules.
+%
+% The \enve{document} command concludes the typesetting run.
+%
+% \begin{macrocode}
+%<*doc>
+% \end{macrocode}
+%
+% \subsection{The Preamble}
+% The programmers documentation is formatted
+% with the \classname{ltxdoc} class with local customizations,
+% and with the usual code line indexing.
+% \begin{macrocode}
+\documentclass{ltxdoc}
+\RequirePackage{ltxdocext}%
+\let\url\undefined
+\RequirePackage[colorlinks=true,linkcolor=blue]{hyperref}%
+\pdfstringdefDisableCommands{%
+ \let\file\relax
+ \let\sc\relax
+}
+%\expandafter\ifx\csname package@font\endcsname\@undefined\else
+% \expandafter\RequirePackage\expandafter{\csname package@font\endcsname}%
+%\fi
+\CodelineIndex\EnableCrossrefs % makeindex -s gind.ist ltxutil
+\RecordChanges % makeindex -s gglo.ist -o ltxutil.gls ltxutil.glo
+% \end{macrocode}
+%
+% \subsubsection{Docstrip and info directives}
+% We use so many {\sc docstrip} modules that we set the
+% \texttt{StandardModuleDepth} counter to 1.
+% \begin{macrocode}
+\setcounter{StandardModuleDepth}{1}
+% \end{macrocode}
+% The following command retrieves the date and version information
+% from this file.
+% \begin{macrocode}
+\expandafter\GetFileInfo\expandafter{\jobname.dtx}%
+% \end{macrocode}
+%
+% \subsection{The ``Read Me'' File}
+% As promised above, here is the contents of the
+% ``Read Me'' file. That file serves a double purpose,
+% since it also constitutes the beginining of the
+% programmer's documentation. What better thing, after
+% all, to have appear at the beginning of the
+% typeset documentation?
+%
+% A good discussion of how to write a ReadMe file can be found in
+% Engst, Tonya, ``Writing a ReadMe File? Read This''
+% \emph{MacTech} October 1998, p. 58.
+%
+% Note the appearance of the
+% \cmd\StopEventually\ command, which marks the
+% dividing line between the user documentation
+% and the programmer documentation.
+%
+% The usual user will not be asked to
+% do a full build, not to speak
+% of the bootstrap.
+% Instructions for carrying out these procedures
+% begin the programmer's manual.
+%
+% \begin{macrocode}
+\begin{filecontents*}{README-LTXUTIL}
+\title{%
+ A \LaTeX\ Package of utility macros%
+ \thanks{%
+ This file has version number \fileversion,
+ last revised \filedate.%
+ }%
+ \thanks{%
+ Version \fileversion\ \copyright\ 2019--2020 American Physical Society
+ }%
+}%
+\author{%
+ Arthur Ogawa%
+ \thanks{\texttt{mailto:arthur\_ogawa at sbcglobal.net}}%
+}%
+%\iffalse
+% For version number and date,
+% search on "\fileversion" in the .dtx file,
+% or see the end of the README-LTXUTIL file.
+%\fi
+\maketitle
+
+This file embodies the \classname{ltxutil} package,
+the implementation and its user documentation.
+
+The distribution point for this work is
+\url{journals.aps.org/revtex},
+which contains prebuilt runtime files, documentation, and full source,
+ready to add to a TDS-compliant \TeX\ installation.
+
+The \classname{ltxutil} package was commissioned by the American Physical Society
+and is distributed under the terms of the \LaTeX\ Project Public License 1.3c,
+the same license under which all the portions of \LaTeX\ itself are distributed.
+Please see \url{http://ctan.tug.org/macros/latex/base/lppl.txt} for details.
+
+To use this document class, you must have a working
+\TeX\ installation equipped with \LaTeXe\
+and possibly pdftex and Adobe Acrobat Reader or equivalent.
+
+To install, retrieve the distribution,
+unpack it into a directory on the target computer,
+and move the file \file{ltxutil.sty}
+into a location in your filesystem where it will be found by \LaTeX.
+
+To use, read the user documentation \file{ltxutil.pdf}.
+
+\tableofcontents
+
+\section{Processing Instructions}
+
+The package file \file{ltxutil.sty}
+is generated from this file, \file{ltxutil.dtx},
+using the {\sc docstrip} facility of \LaTeX
+via |tex ltxutil.dtx| (Note: do \emph{not} use \LaTeX\ for this task).
+The typeset documentation that you are now reading is generated from
+the same file by typesetting it with \LaTeX\ or pdftex
+via |latex ltxutil.dtx| or |pdflatex ltxutil.dtx|.
+
+\subsection{Build Instructions}
+
+You may bootstrap this suite of files solely from \file{ltxutil.dtx}.
+Prepare by installing \LaTeXe\ (and either tex or pdftex) on your computer,
+then carry out the following steps:
+\begin{enumerate}
+\item
+Within an otherwise empty directory,
+typeset \file{ltxutil.dtx} with \LaTeX\ or pdflatex;
+you will obtain the typeset documentation you are now reading,
+along with the file \file{README-LTXUTIL}.
+
+Note: you will have to run \LaTeX, then
+\file{makeindex} \texttt{-s gind.ist ltxutil.idx}, then
+\file{makeindex} \texttt{-s gglo.ist -o ltxutil.gls ltxutil.glo}, then
+\LaTeX\ again in order to obtain a valid index and table of contents.
+\item
+Now typeset \file{ltxutil.dtx} with \TeX (not \LaTeX),
+thereby generating the package file \file{ltxutil.sty}.
+\item
+Install the following files into indicated locations within your
+TDS-compliant \texttt{texmf} tree (you may need root access):
+\begin{itemize}
+\item
+\file{$TEXMF/}\file{tex/}\file{latex/}\file{revtex/}\classname{ltxutil.sty}
+\item
+\file{$TEXMF/}\file{source/}\file{latex/}\file{revtex/}\classname{ltxutil.dtx}
+\item
+\file{$TEXMF/}\file{doc/}\file{latex/}\file{revtex/}\classname{ltxutil.pdf}
+\end{itemize}
+where \file{$TEXMF/} stands for \file{texmf-local/}, or some other \texttt{texmf} tree
+in your installation.
+\item
+Run \texttt{mktexlsr} on \file{$TEXMF/} (you may need root access).
+\item
+Build and installation are now complete;
+now put a \cmd\usepackage\texttt{\{ltxutil\}} in your document preamble!
+\end{enumerate}
+
+\subsection{Change Log}
+\changes{4.0b}{1999/06/20}{AO: Fixed spurious \texttt{CR} and (return) characters in output file. Also, if the document did not have the \cs{end}\texttt{figure} on a line of its own, the macro wouldn't work. Fixed.}
+\changes{4.0b}{1999/06/20}{AO: Removed superfluous \cs{def}s, changed to using \cs{floats@sw} as the flag. Also stopped using DPC's \cs{if@twocolumn} flag: using \cs{floats@sw} instead. Also added \cs{par}\cs{vskip}\cs{z@skip} after the \cs{minipagefootnotes} so that the float box would have zero depth like the kernel one. }
+\changes{4.0b}{1999/06/20}{only execute if there really were floats of the given type}
+\changes{4.0b}{1999/06/20}{Support the hack with \cs{prepdef}, and delay until \cs{AtBeginDocument} time, since \classname{hyperref} clobbers \cs{caption}.}
+\changes{4.0c}{1999/11/13}{(AO, 110) Install hooks for endfloats processing}
+\changes{4.0c}{1999/11/13}{(AO, 116) Hyperref compatibility}
+\changes{4.0c}{1999/11/13}{(AO, 130) Interference from array package}
+\changes{4.0c}{1999/11/13}{*-form mandates pagebreak at each float; only print section head if there is something there.}
+\changes{4.0d}{2000/04/10}{(AO, 127) Floats placed [h] to allow page breaks}
+\changes{4.0d}{2000/04/10}{(AO, 174) kernel fix}
+\changes{4.0d}{2000/05/19}{(AO, 224) Hyperref compatibility.}
+\changes{4.0d}{2000/05/23}{Allow things to break over pages by setting array@default.}
+\changes{4.0e}{2000/11/16}{(AO, 221) Remove samepage command from @xfloat@prep: If the float can break over pages, we want better control.}
+\changes{4.0f}{2001/07/13}{(AO, 404) Hyperref compatibility}
+\changes{4.1a}{2008/01/19}{(AO, 459) do not assume \cs{class@name} is defined}%
+\changes{4.1a}{2008/01/19}{(AO, 461) Change the csname from \cs{@dotsep} to \cs{ltxu@dotsep}. The former is understood in mu. (What we wanted was a dimension.)}%
+\changes{4.1a}{2008/01/19}{(AO, 475) I had not properly reproduced the LaTeX macro \cs{eqnarray}.}%
+\changes{4.1a}{2008/01/19}{(AO, 479) Per: Dylan Thurston<dpt at math.harvard.edu>}%
+\changes{4.1a}{2008/06/30}{(AO) Make \cs{addtocontents} a \cs{long} \cs{def}; gobble up \cs{footnote}}%
+\changes{4.1a}{2008/06/30}{(AO) Remove code that avoided changes to \cs{@xfootnotemark}}%
+\changes{4.1a}{2008/06/30}{(AO, 438) Complete rewrite of footnote macros.}
+\changes{4.1a}{2008/07/07}{\cs{@xfloat@prep} calls \cs{ltx@footnote@pop} to restore the original \cs{ltx@footmark} and \cs{ltx@foottext} procedures, in case footnote processing has switched.}
+\changes{4.1a}{2008/08/12}{\cs{class@documenthook} is the last \cs{AtBeginDocument} token now}
+\changes{4.1a}{2008/08/12}{Class extension mechanism \cs{@pushfilename@ltx} and \cs{@p@pfilename@ltx}.}
+\changes{4.1a}{2008/08/12}{Class extension mechanism \cs{class@extension}, \cs{class@extensionfile}, and \cs{class@ext@hook}.}
+\changes{4.1a}{2008/08/12}{Get rid of \cs{set@typesize@hook} \cs{set@pica@hook} and the \cs{normalsize} directive}%
+\changes{4.1b}{2008/08/12}{(AO, 487) Support for video figures and the \cs{setfloatlink} command}%
+\changes{4.1b}{2008/08/12}{(AO, 505) try to accommodate \classname{colortbl}.}
+\changes{4.1b}{2008/08/12}{Acquire \classname{hyperref} savoire}
+\changes{4.1b}{2008/08/12}{Default assignment of \cs{float@sw} now, not at \cs{AtBeginDocument} time.}%
+\changes{4.1b}{2008/08/12}{If class option \classoption{lengthcheck} is in effect, log the height of this float class.}
+\changes{4.1b}{2008/08/12}{No need to protect against undefined \cs{float@sw}}
+\changes{4.1b}{2008/08/12}{Patch the array package even later: after all package patches go in.}
+\changes{4.1b}{2008/08/12}{Refine toc processing: provide default.}%
+\changes{4.1b}{2008/08/12}{Tally and log the height of a float class}
+\changes{4.1d}{2009/03/27}{(AO, 511) Compatability with lineno.sty's erroneous way of detecting fleqn.clo}%
+\changes{4.1f}{2009/07/07}{(AO, 515) Hook for setting the font of a footnote}
+\changes{4.1f}{2009/07/10}{(AO, 518) Tally register overflow when locument is long}
+\changes{4.1g}{2009/10/06}{(AO, 532) Both arguments of \cs{href} get sanitized}%
+\changes{4.1g}{2009/10/07}{(AO, 525) Remove phantom paragraph above display math that is given in vertical mode}%
+\changes{4.1g}{2009/10/07}{(AO, 539) Use of double-backslash in argument of \cs{section} gives error. The \classname{textcase} package is involved.}%
+\changes{4.1n}{2009/12/05}{(AO, 569) Use of \classname{hyperref} interferes with column balancing of last page}%
+\changes{4.1n}{2009/12/06}{(AO) Incorporate change to ltmiscen.dtx v1.1i 2000/05/19}%
+\changes{4.1n}{2009/12/09}{(AO, 569) execute \classname{atveryend}'s \cs{Call@AfterLastShipout} at the proper time}%
+\changes{4.1n}{2009/12/13}{(AO, 574) protect against \classname{lineno.sty}, which forces a visit to the output routine, which appears to destroy the value of \cs{@tempdima}}%
+\changes{4.1n}{2010/01/02}{(AO, 571) Interface \cs{set@footnotewidth} for determining the set width of footnotes}%
+\changes{4.1n}{2010/01/02}{(AO, 571) allow split after last line of footnote}%
+\changes{4.1n}{2010/01/06}{(AO, 572) title block footnotes numbered independently from body footnotes}%
+\changes{4.1p}{2010/02/24}{(AO, 582) A patch of \classname{hyperref.sty} to provide backward compatibility to \TeX Live 2007's version 6.75r}%
+\changes{4.2a}{2017/11/21}{(MD) Use updated best practice to use https and doi.org}%
+\changes{4.2a}{2018/12/12}{(MD) Updated name of README file and use standard fonts when typesetting}%
+\changes{4.2d}{2020/09/19}{(PHO) Adapt \cs{document} and \cs{enddocument} hooks to the 2020-10-01 \LaTeX{} release.}%
+
+\end{filecontents*}
+% \end{macrocode}
+%
+% \subsection{The Document Body}
+%
+% Here is the document body, containing only a
+% \cmd\DocInput\ directive---referring to this very file.
+% This very cute self-reference is a common \classname{ltxdoc} idiom.
+% \begin{macrocode}
+\begin{document}%
+\expandafter\DocInput\expandafter{\jobname.dtx}%
+\end{document}
+% \end{macrocode}
+%
+% \begin{macrocode}
+%</doc>
+% \end{macrocode}
+%
+% \section{Using this package}
+% Once this package is installed on your filesystem, you can employ it in
+% adding functionality to \LaTeX\ by invoking it in your document or document class.
+%
+% \subsection{Invoking the package}
+% In your document, you can simply call it up in your preamble:
+% \begin{verbatim}
+%\documentclass{book}%
+%\usepackage{ltxutil}%
+%\begin{document}
+%<your document here>
+%\end{document}\end{verbatim}
+% However, the preferred way is to invoke this package from within your
+% customized document class:
+% \begin{verbatim}
+%\NeedsTeXFormat{LaTeX2e}[1995/12/01]%
+%\ProvidesClass{myclass}%
+%\RequirePackage{ltxutil}%
+%\LoadClass{book}%
+%<class customization commands>
+%\endinput\end{verbatim}
+%
+% Once loaded, the package gives you acccess to certain procedures,
+% usually to be invoked by a \LaTeX\ command or environment, but not at the document level.
+%
+%
+% \section{Compatibility with \LaTeX's Required Packages}
+% Certain packages, usually ones written by members of the
+% \LaTeX\ Project itself, have been designated ``required'' and
+% are distributed as part of standard \LaTeX.
+% These packages have been placed in a priviledged position
+% vis \'a vis the \LaTeX\ kernel in that they override the definitions of certain kernel macros.
+%
+% The \classname{ltxutil} package will be incompatible with any package that
+% redefines any of the kernel macros that \classname{ltxutil} patches---if that
+% package is loaded \emph{after} \classname{ltxutil}. This means that for
+% greatest compatibility, \classname{ltxutil} should be loaded \emph{after},
+% say, \classname{ftnright}, which overwrites \LaTeX's kernel
+% procedures \cmd\@outputdblcol, \cmd\@startcolumn, and \cmd\@makecol.
+%
+% Hereinafter follows some notes on specific \LaTeX\ packages.
+%
+% \subsection{array}
+% This package alters the way tabular environments are done,
+% therefore it could run afoul of the \LaTeX\ ``required'' package \classname{array} or any
+% package that calls for it to be loaded.
+% However, this package has provisions for remaining compatible with \classname{array}.
+% So long as the version of \classname{array} that is used with this package has the appropriate
+% meanings for the procedures it overwrites, all should be well.
+%
+% \subsection{longtable}
+% David Carlisle's \classname{longtable} package modifies both the \LaTeX\ kernel and the
+% \classname{array} package. This package must therefore alter \cmd\LT@array.
+% For now, that job is handled by \classname{ltxgrid}.
+%
+%
+%\StopEventually{}
+%
+% \section{Implementation of package}
+%
+% Special acknowledgment: this package uses concepts pioneered
+% and first realized by William Baxter (mailto:web at superscript.com)
+% in his SuperScript line of commercial typesetting tools, and
+% which are used here with his permission.
+%
+% \subsection{Beginning of the \file{package} {\sc docstrip} module}
+% \begin{macrocode}
+%<*package>
+\def\package@name{ltxutil}%
+\expandafter\PackageInfo\expandafter{\package@name}{%
+ Utility macros for \protect\LaTeXe,
+ by A. Ogawa (arthur_ogawa at sbcglobal.net)%
+}%
+%</package>
+% \end{macrocode}
+%
+% \subsection{Banner and beginning of the \file{kernel} {\sc docstrip} module}%
+% \begin{macrocode}
+%<*kernel>
+% \end{macrocode}
+%
+% \subsection{Errors and warnings}
+%
+% \begin{macro}{\class@err}
+% \begin{macro}{\class@warn}
+% \begin{macro}{\class@info}
+% A few shorthands for Class messages.
+% Your document class should define \cmd\class@name.
+% \begin{macrocode}
+\def\class@err#1{\ClassError{\class@name}{#1}\@eha}%
+\def\class@warn#1{\ClassWarningNoLine{\class@name}{#1}}%
+\def\class@info#1{\ClassInfo{\class@name}{#1}}%
+\def\obsolete@command#1{%
+ \class@warn@end{Command \string#1\space is obsolete.^^JPlease remove from your document}%
+ \global\let#1\@empty
+ #1%
+}%
+\def\replace@command#1#2{%
+ \class@warn@end{Command \string#1\space is obsolete;^^JUse \string#2\space instead}%
+ \global\let#1#2%
+ #1%
+}%
+\def\replace@environment#1#2{%
+ \class@warn@end{Environment #1 is obsolete;^^JUse #2 instead}%
+ \glet@environment{#1}{#2}%
+ \@nameuse{#1}%
+}%
+\def\incompatible@package#1{%
+ \@ifpackageloaded{#1}{%
+ \def\@tempa{I cannot continue. You must remove the \string\usepackage\ statement that caused that package to be loaded.}%
+ \ClassError{\class@name}{The #1 package cannot be used with \class@name}%
+ \@tempa\stop
+ }{%
+ \class@info{#1 was not loaded (OK!)}%
+ }%
+}%
+\def\class@warn@end#1{%
+ \gappdef\class@enddocumenthook{\class@warn{#1}}%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \changes{4.1a}{2008/01/19}{(AO, 459) do not assume \cs{class@name} is defined}%
+% Give \cmd\class@name\ a meaning if it does not already have one.
+% \begin{macrocode}
+\ifx\undefined\class@name
+ \def\class@name{ltxutil}%
+ \class@warn{You should define the class name before reading in this package. Using default}%
+\fi
+% \end{macrocode}
+%
+% \subsection{New Tools}%
+%
+% \begin{macro}{\t@}
+% \begin{macrocode}
+\def\t@{to}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\dimen@iii}
+% \begin{macrocode}
+\dimendef\dimen@iii\thr@@
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\halignt@}
+% \begin{macrocode}
+\def\halignt@{\halign\t@}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\f@ur}
+% Analogous to \cmd\@ne, \cmd\tw@, and \cmd\thr@@.
+% \begin{macrocode}
+\chardef\f@ur=4\relax
+\chardef\cat@letter=11\relax
+\chardef\other=12\relax
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\let@environment}
+% \begin{macro}{\glet@environment}
+% The directive \cmd\let@environment\ takes care of a common programming
+% idiom whereby one environment is made a synonym for another.
+% \begin{macrocode}
+\def\let@environment#1#2{%
+ \expandafter\let
+ \csname#1\expandafter\endcsname\csname#2\endcsname
+ \expandafter\let
+ \csname end#1\expandafter\endcsname\csname end#2\endcsname
+}%
+\def\glet@environment#1#2{%
+ \global\expandafter\let
+ \csname#1\expandafter\endcsname\csname#2\endcsname
+ \global\expandafter\let
+ \csname end#1\expandafter\endcsname\csname end#2\endcsname
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\tracingplain}
+% The command \cmd\tracingplain\ causes \TeX's tracing parameters to
+% return to the values set by default. This command is sometimes
+% useful when you have said \cmd\tracingall\ somewhere and want to
+% restore.
+% The \cmd\traceoutput\ command causes \cmd\tracingoutput\ diagnostics
+% upon \cmd\shipout.
+% \begin{macrocode}
+\newcommand\tracingplain{%
+ \tracingonline\z@\tracingcommands\z@\tracingstats\z@
+ \tracingpages\z@\tracingoutput\z@\tracinglostchars\@ne
+ \tracingmacros\z@\tracingparagraphs\z@\tracingrestores\z@
+ \showboxbreadth5\showboxdepth3\relax %\errorstopmode
+ }%
+\newcommand\traceoutput{%
+ \appdef\@resetactivechars{\showoutput}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\say}
+% \begin{macro}{\saythe}
+% The commands \cmd\say\ and \cmd\saythe\ cause diagnostic messages in the
+% \TeX\ log that give the value of a control sequence name or a register
+% respectively.
+% \begin{macrocode}
+\newcommand\say[1]{\typeout{<\noexpand#1=\meaning#1>}}%
+\newcommand\saythe[1]{\typeout{<\noexpand#1=\the#1>}}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\fullinterlineskip}
+% Resets the \cmd\prevdepth\ so that the full amount of \cmd\baselineskip\ glue will be inserted by
+% the \cmd\baselinesklip\ mechanism.
+% Can be invoked just after a \cmd\hrule\ to undo its default suppression of base line skip.
+% \begin{macrocode}
+\def\fullinterlineskip{\prevdepth\z@}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\count@i}
+% \begin{macro}{\count@ii}
+%
+% \begin{macrocode}
+\countdef\count@i\@ne
+\countdef\count@ii\tw@
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+%
+% \subsection{Boolean Control}%
+% We introduce just enough of the Boolean calculus for \TeX.
+% Alan Jeffrey was the pioneer here, with an article in TUGboat
+% (Vol. 11, No. 2, page 237).
+% This implementation owes a debt to
+% William Baxter (web at superscript.com).
+% See articles by Baxter and Ogawa in the proceedings of the
+% 1994 TUG meeting, TUGboat Vol.~15, No.~3.
+%
+% \begin{macro}{\prepdef}
+% \begin{macro}{\appdef}
+% \begin{macro}{\gappdef}
+%
+% Provide the capability of performing head- and tail patches.
+% The procedure \cmd\prepdef\ prepends to the given macro
+% the tokens specified in its second argument.
+% Likewise for \cmd\appdef, except that it appends.
+% Note that the first 10 toks registers are utility registers,
+% and we simply make a control sequence name, \cmd\toks@ii, for one of
+% them.
+% \begin{macrocode}
+\long\def\prepdef#1#2{%
+ \@ifxundefined#1{\toks@{}}{\toks@\expandafter{#1}}%
+ \toks@ii{#2}%
+ \edef#1{\the\toks@ii\the\toks@}%
+}%
+\long\def\appdef#1#2{%
+ \@ifxundefined#1{\toks@{}}{\toks@\expandafter{#1}}%
+ \toks@ii{#2}%
+ \edef#1{\the\toks@\the\toks@ii}%
+}%
+\long\def\gappdef#1#2{%
+ \@ifxundefined#1{\toks@{}}{\toks@\expandafter{#1}}%
+ \toks@ii{#2}%
+ \global\edef#1{\the\toks@\the\toks@ii}%
+}%
+\long\def\appdef@val#1#2{%
+ \appdef#1{{#2}}%
+}%
+\long\def\appdef@e#1#2{%
+ \expandafter\appdef
+ \expandafter#1%
+ \expandafter{#2}%
+}%
+\long\def\appdef@eval#1#2{%
+ \expandafter\appdef@val
+ \expandafter#1%
+ \expandafter{#2}%
+}%
+\toksdef\toks@ii=\tw@
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\@ifxundefined}
+% \begin{macro}{\@ifnotrelax}
+% \begin{macro}{\@argswap}
+% \begin{macro}{\@argswap@val}
+%
+% Certain utility procedures use \cmd\@ifxundefined,
+% which is defined here in terms of \cmd\@ifx.
+% Others use \cmd\@ifnotrelax, namely when
+% the control sequence name is manufactured by
+% the use of \cmd\csname.
+%
+% The procedures \cmd\@argswap and \cmd\@argswap@val
+% are used to facilitate control of expansion.
+%
+% \begin{macrocode}
+\long\def\@ifxundefined#1{\@ifx{\undefined#1}}%
+\long\def\@ifnotrelax#1#2#3{\@ifx{\relax#1}{#3}{#2}}%
+\long\def\@argswap#1#2{#2#1}%
+\long\def\@argswap@val#1#2{#2{#1}}%
+\def\@ifxundefined@cs#1{\expandafter\@ifx\expandafter{\csname#1\endcsname\relax}}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\rvtx@ifformat@geq}
+% Some changes in the \LaTeX{} kernel requires us to conditionally
+% define some macros depending on the version of the kernel.
+% \cmd\rvtx@ifformat@geq{} will check if the release date of the
+% currently-running \LaTeXe{} kernel is greater or equal to the
+% argument (the argument should be in the format \texttt{yyyy-mm-dd}).
+% \changes{4.2d}{2020/09/17}{(PHO) Add \cs{rvtx@ifformat@geq}.}%
+% \begin{macrocode}
+\ifx\IfFormatAtLeastTF\undefined
+ \def\rvtx@ifformat@geq{\@ifl@t@r\fmtversion}%
+\else
+ \let\rvtx@ifformat@geq\IfFormatAtLeastTF
+\fi
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@boolean}
+% \begin{macro}{\@boole@def}
+% In order to define \cmd\@ifx, we first must create the
+% ``defining word'' (term taken form our Forth vocabulary)
+% \cmd\@boole@def, which employs \cmd\@boolean\ to do its job.
+% \begin{macrocode}
+\def\@boolean#1#2{%
+ \long\def#1{%
+ #2% \if<something>
+ \expandafter\true@sw
+ \else
+ \expandafter\false@sw
+ \fi
+ }%
+}%
+\def\@boole@def#1#{\@boolean{#1}}% Implicit #2
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\@booleantrue}
+% \begin{macro}{\@booleanfalse}
+% The procedures \cmd\@booleantrue\ and
+% \cmd\@booleanfalse\ are assignment operators
+% for Boolean flags.
+% \begin{macrocode}
+\def\@booleantrue#1{\let#1\true@sw}%
+\def\@booleanfalse#1{\let#1\false@sw}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\@ifx}
+% \begin{macro}{\@ifx@empty}
+% \begin{macro}{\@if@empty}
+% \begin{macro}{\@ifcat}%
+% \begin{macro}{\@ifdim}%
+% \begin{macro}{\@ifeof}%
+% \begin{macro}{\@ifhbox}%
+% \begin{macro}{\@ifhmode}%
+% \begin{macro}{\@ifinner}%
+% \begin{macro}{\@ifmmode}%
+% \begin{macro}{\@ifnum}%
+% \begin{macro}{\@ifodd}%
+% \begin{macro}{\@ifvbox}%
+% \begin{macro}{\@ifvmode}%
+% \begin{macro}{\@ifvoid}%
+% We can now invoke the defining word to create
+% the procedures \cmd\@ifx\ and friends.
+%
+% Compatibility Note: earlier versions of this package
+% defined a procedure \cmd\@ifempty. However, for compatibility with AMS\LaTeX,
+% we must avoid the following three names:
+% \cmd\@ifempty, \cmd\@xifempty, and \cmd\@ifnotempty.
+%
+% \begin{macrocode}
+\@boole@def\@ifx#1{\ifx#1}%
+\@boole@def\@ifx@empty#1{\ifx\@empty#1}%
+\@boole@def\@if@empty#1{\if!#1!}%
+%\@boole@def\@if@sw#1{\csname if#1\endcsname}%
+\def\@if@sw#1#2{#1\expandafter\true@sw\else\expandafter\false@sw#2}%
+\@boole@def\@ifdim#1{\ifdim#1}%
+\@boole@def\@ifeof#1{\ifeof#1}%
+\@boole@def\@ifhbox#1{\ifhbox#1}%
+\@boole@def\@ifhmode{\ifhmode}%
+\@boole@def\@ifinner{\ifinner}%
+\@boole@def\@ifmmode{\ifmmode}%
+\@boole@def\@ifnum#1{\ifnum#1}%
+\@boole@def\@ifodd#1{\ifodd#1}%
+\@boole@def\@ifvbox#1{\ifvbox#1}%
+\@boole@def\@ifvmode{\ifvmode}%
+\@boole@def\@ifvoid#1{\ifvoid#1}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\true@sw}
+% \begin{macro}{\false@sw}
+%
+% Note that when a Boolean operator expands, it
+% employs two macros that act as selectors, defined here.
+%
+% \begin{macrocode}
+\long\def\true@sw#1#2{#1}%
+\long\def\false@sw#1#2{#2}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\loopuntil}
+% \begin{macro}{\loopwhile}
+%
+% Loop control using the Boolean idiom.
+% Superior to \cmd\loop\dots\cmd\repeat\ because these can be nested.
+% The tail of the argument must have a Boolean predicate.
+%
+% \begin{macrocode}
+\long\def\loopuntil#1{#1{}{\loopuntil{#1}}}%
+\long\def\loopwhile#1{#1{\loopwhile{#1}}{}}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\@provide}
+%
+% A defining word that refuses to clobber a prior meaning.
+%
+% \begin{macrocode}
+\def\@provide#1{%
+ \@ifx{\undefined#1}{\true@sw}{\@ifx{\relax#1}{\true@sw}{\false@sw}}%
+ {\def#1}{\def\j@nk}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+%
+% \subsection{Begin Document Structure}
+% The standard \LaTeX\ mechanism \cmd\AtBeginDocument\
+% is inadequate because the \cmd\vsize\ is bound much too early.
+% We supply here a mechanism whereby decisions about the
+% page layout can be deferred until \cmd\AtBeginDocument\ time.
+%
+% The problem we are working around is that the \cmd\AtBeginDocument\
+% hook in \cmd\document\ appears long after the calculation of
+% \cmd\vsize\ and \cmd\hsize, that is, \LaTeX\ provides no mechanism
+% for deferring the decision about the page grid until \cmd\AtBeginDocument\ time.
+% We fix things by prepending a hook at the very beginning of \cmd\document.
+%
+% As it turns out, though, it appears feasible to simply invoke the desired
+% column grid command at \cmd\AtBeginDocument\ time, since the MVL has nothing in it
+% at that time that would be problematical.
+%
+% \begin{macro}{\document}
+% We begin by installing hooks into \cmd\document\ that
+% we will manage ourselves.
+%
+% The 2020-10-01 \LaTeX{} release got a new hook management system and
+% several new hooks (several previously provided by \textsf{etoolbox}).
+% The one we want here is \texttt{begindocument/before}, the first thing
+% executed by \cmd\document{}, right after ending the group started by
+% \cmd\begin{}.
+%
+% Thus, if the \LaTeX{} kernel date is 2020-10-01 we just add to that
+% hook, otherwise resort to the old method, patching \cmd\document:
+% end the group started by \cmd\begin, apply our hook, and
+% conclude our shenanigans by absorbing
+% the first token of the expansion of \cmd\document, which
+% we assume to be \cmd\endgroup{} (true until the aforementioned release).
+% \changes{4.1a}{2008/08/12}{Get rid of \cs{set@typesize@hook} \cs{set@pica@hook} and the \cs{normalsize} directive}%
+% \changes{4.2d}{2020/09/17}{(PHO) Use \LaTeX's hook management system, if possible.}%
+% \begin{macrocode}
+\rvtx@ifformat@geq{2020-10-01}%
+ {%
+ \AddToHook{begindocument/before}{\document@inithook}%
+ }{%
+ \prepdef\document{%
+ \endgroup
+ \document@inithook
+ \true@sw{}%
+ }%
+ }
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\document@inithook}
+% To use, simply \cmd\appdef\cmd\document@inithook\arg{your tokens here}.
+% \begin{macrocode}
+\let\document@inithook\@empty
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\class@documenthook}
+% \begin{macro}{\class@enddocumenthook}
+% \changes{4.1a}{2008/08/12}{\cs{class@documenthook} is the last \cs{AtBeginDocument} token now}
+% We install the last \cmd\AtBeginDocument\ hook, namely the
+% procedure \cmd\class@documenthook. Within the document class,
+% we will use this hook exclusively, so as to avoid interference from other packages.
+% Similarly with \cmd\class@enddocumenthook, installed via \cmd\AtEndDocument.
+%
+% A document class using this package should do as this package does and just say,
+% \cmd\appdef\ \cmd\class@documenthook\ instead of \cmd\AtBeginDocument,
+% and \cmd\appdef\ \cmd\class@enddocumenthook\ instead of \cmd\AtEndDocument.
+% \begin{macrocode}
+\appdef\document@inithook{%
+ \AtBeginDocument{\class@documenthook}%
+}%
+\AtEndDocument{%
+ \class@enddocumenthook
+}%
+\let\class@documenthook\@empty
+\let\class@enddocumenthook\@empty
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\enddocument}
+% \begin{macro}{\check@aux}
+% \begin{macro}{\do@check@aux}
+% \changes{4.1n}{2009/12/05}{(AO, 569) Use of \classname{hyperref} interferes with column balancing of last page}%
+% The standard \LaTeX\ \enve{document} processing is a potential problem,
+% particularly when the output routine has been changed by \classname{ltxgrid}.
+% We separate out the procedure that checks the auxiliary file at the end of
+% the job so that later it can be called from the safety of the output
+% routine.
+% We will do this to ensure that the \cmd\@mainaux\ stream is not closed until
+% the last page of the job is shipped out, and that can only be done by coordinating
+% with the output routine.
+%
+% \changes{4.2d}{2020/09/17}{(PHO) Only redefine \cs{enddocument} in older versions.}%
+% This approach, however, will only be done for older versions of the
+% \LaTeX{} kernel:
+% \begin{macrocode}
+\rvtx@ifformat@geq{2020-10-01}{%
+ % <definitions for newer LaTeX later>
+}{%
+ % <definitions for older LaTeX>
+\def\enddocument{%
+% \end{macrocode}
+% \changes{4.1n}{2009/12/06}{(AO) Incorporate change to ltmiscen.dtx v1.1i 2000/05/19}%
+% The following line from \filename{ltmiscen.dtx} `resets \cmd\AtEndDocument for latex/3060'.
+% \begin{macrocode}
+ \let\AtEndDocument\@firstofone
+% \end{macrocode}
+% \begin{macrocode}
+ \@enddocumenthook
+ \@checkend{document}%
+% \end{macrocode}
+% The \cmd\clear@document\ statement ends the current page (we must guarantee no further shipouts),
+% then executes all cleanup procedures that must occur only after the last shipout.
+% Clients will queue up their procedures via \cmd\AfterLastShipout, if it exists, otherwise by doing \cmd\appdef\cmd\clear@document.
+% \begin{macrocode}
+ \clear@document
+% \end{macrocode}
+% We are very close to ending the \TeX\ run, now.
+% \begin{macrocode}
+ \check@aux
+ \deadcycles\z@
+ \@@end
+}%
+\def\check@aux{\do@check@aux}%
+\def\do@check@aux{%
+ \@if@sw\if@filesw\fi{%
+ \immediate\closeout\@mainaux
+ \let\@setckpt\@gobbletwo
+ \let\@newl@bel\@testdef
+ \@tempswafalse
+ \makeatletter
+ \input\jobname.aux\relax
+ }{}%
+ \@dofilelist
+ \@ifdim{\font@submax >\fontsubfuzz\relax}{%
+ \@font@warning{%
+ Size substitutions with differences\MessageBreak
+ up to \font@submax\space have occured.\@gobbletwo
+ }%
+ }{}%
+ \@defaultsubs
+ \@refundefined
+ \@if@sw\if@filesw\fi{%
+ \@ifx{\@multiplelabels\relax}{%
+ \@if@sw\if@tempswa\fi{%
+ \@latex@warning@no@line{%
+ Label(s) may have changed.
+ Rerun to get cross-references right%
+ }%
+ }{}%
+ }{%
+ \@multiplelabels
+ }%
+ }{}%
+}%
+}
+% \end{macrocode}
+%
+% \changes{4.2d}{2020/09/17}{(PHO) Patch \cs{enddocument} at runtime in newer versions.}%
+% \begin{macro}{\rvtx@enddocument@patch}
+% For newer \LaTeX{} we'll try to be a bit more future-proof
+% (no miracle though). The code for \cmd\enddocument{}
+% (in pre-2020-10-01 \LaTeX) is roughly:
+% \begin{verbatim}
+% \def\enddocument{%
+% <hooks and bookkeeping>
+% \clearpage
+% <read main .aux and final checks>
+% \@@end
+% }
+% \end{verbatim}
+% and the patches above replace the \cmd\clearpage{} by its own
+% \cmd\clear@document, and \verb|<read main .aux and final checks>| by
+% \cmd\do@check@aux, which it can later control the timing.
+%
+% Now we will apply the same changes, but this time without redefining
+% \cmd\enddocument: we will instead replace tokens on-the-fly, when
+% \cmd\enddocument{} is expanded. This will grant us a slightly safer
+% approach that won't depend so much on the internals of
+% \cmd\enddocument.
+%
+% This entire patch should work with the previous definition of
+% \cmd\enddocument{} as well (except it cannot be used in the hook),
+% but for now leave previous versions untouched.
+%
+% The entire patching will reside in the \texttt{enddocument} hook:
+% \begin{macrocode}
+\rvtx@ifformat@geq{2020-10-01}{%
+ \AddToHook{enddocument}{\rvtx@enddocument@patch{}}%
+}{}
+% \end{macrocode}
+%
+% This macro will be executed after \cmd\enddocument{} has expanded,
+% so all its tokens are now exposed. Here we will assume that
+% \cmd\enddocument{} contains the tokens \verb|\@checkend{document}|
+% and \cmd\endgroup, and use them as delimiters:
+% \begin{macrocode}
+\protected\long\def\rvtx@enddocument@patch#1#2\@checkend#3{%
+ \begingroup
+ \edef\x{\detokenize{#3}}%
+ \edef\y{\detokenize{document}}%
+ \expandafter\endgroup
+ \ifx\x\y
+ \expandafter\rvtx@enddocument@patch@end
+ \else
+ \expandafter\rvtx@enddocument@patch@more
+ \fi
+ {#1#2}{#3}}
+\def\rvtx@enddocument@patch@more#1#2{%
+ \rvtx@enddocument@patch{#1\@checkend{#2}}}
+% \end{macrocode}
+%
+% When the \verb|\@checkend{document}| is reached, use \cmd\clearpage{}
+% and \cmd\enddocument{} as delimiters for the
+% \verb|<read main .aux and final checks>| part, and save it in
+% \cmd\do@check@aux{}:
+% \begin{macrocode}
+\long\def\rvtx@enddocument@patch@end#1#2\clearpage#3\endgroup{%
+ \def\do@check@aux{#3\endgroup}%
+% \end{macrocode}
+% Then execute the code consumed in the previous step:
+% \begin{macrocode}
+ #1%
+ \@checkend{#2}%
+% \end{macrocode}
+% Do \cmd\clear@document{} instead of \cmd\clearpage{} and
+% \cmd\check@aux{} instead of the code grabbed.
+% \begin{macrocode}
+ \clear@document
+ \check@aux}
+\def\check@aux{\do@check@aux}%
+% \end{macrocode}
+% \end{macro}
+%
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\clear@document}
+% \changes{4.1n}{2009/12/05}{(AO, 569) Use of \classname{hyperref} interferes with column balancing of last page}%
+% The procedure \cmd\clear@document\ is responsible for flushing out the last page of the document,
+% if not already done.
+% The procedure then executes those procedures that must wait for execution until
+% after the last page is shipped out.
+% Clients of \classname{ltxutil}, such as \classname{ltxgrid} and \classname{revtex4}
+% will queue these procedures up via \cmd\AfterLastShipout, if it exists, otherwise by doing \cmd\appdef\cmd\clear@document.
+%
+% The command \cmd\Call@AfterLastShipout\ is provided by Heiko Oberdiek's \classname{atveryend} package.
+% This package is compatible with \classname{ltxutil}.
+%
+% Note on compatibility with \classname{atveryend}:
+% we arrange for \cmd\Call@AfterLastShipout\ to be called from the safety of the output routine,
+% thereby ensuring that all of the procedures queued up by that package's
+% \cmd\AfterLastShipout\ are executed at the right time.
+% We also ensure that \cmd\Call@AfterLastShipout\ has a default definition, in case the package was never loaded.
+% \changes{4.1n}{2009/12/09}{(AO, 569) execute \classname{atveryend}'s \cs{Call@AfterLastShipout} at the proper time}%
+% \begin{macrocode}
+\def\clear@document{%
+ \clearpage
+ \do@output@cclv{%
+ \Call@AfterLastShipout
+ }%
+}%
+\appdef\class@documenthook{%
+ \providecommand\Call@AfterLastShipout{}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \subsection{Class Extensions}%
+% \changes{4.1a}{2008/08/12}{Class extension mechanism \cs{class@extension}, \cs{class@extensionfile}, and \cs{class@ext@hook}.}
+% The \LaTeX\ procedure \cmd\@onefilewithoptions\ is the vehicle for reading in
+% a \LaTeX\ class or package.
+% The APS RevTeX class implements the use of what are called ``substyles'',
+% actually extensions to the class itself.
+% Any document class can do likewise.
+%
+% \begin{macro}{\class@extension}
+% \begin{macro}{\class@extensionfile}
+% \begin{macro}{\class@ext@hook}
+% A procedure similar to \LaTeX's \cmd\@onefilewithoptions, but as an extension to the
+% current document class.
+%
+% Read in the given file as if it were a document class file.
+% Usage: \cmd\class@extensionfile\ \arg{class} \cmd\@extension,
+% where \marg{class} is a file (similar to \file{aps.rtx}) and where \cmd\@extension\ is
+% the file extension for \marg{class}.
+% For instance, to read in the file \file{aps.rtx},
+% do \cmd\class@extensionfile\ \texttt{\{aps\}} \cmd\substyle@ext,
+% where the latter has been define to expand to \file{.rtx}.
+%
+% Features supported include
+% passing existing class options on to the class extension,
+% \cmd\AtEndOfClass\ processing,
+% a stack that restores \cmd\@currname, \cmd\@currext, \cmd\@clsextension, and the \cmd\catcode\ of `@',
+% fall-back to a control sequence name (with leading `rtx@') if no file exists.
+%
+% Note that \cmd\LoadClass\ gives one the ability to write a class that calls in
+% another class as a (sort of) module: this scheme is like \cmd\LoadClass, but
+% turned inside out.
+% \begin{macrocode}
+\def\class@extension#1#2{%
+ \IfFileExists{#1.#2}{%
+ \expandafter\class@extensionfile\csname ver@\@currname.\@currext\endcsname{#1}#2%
+ }{%
+ \csname rtx@#1\endcsname
+ }%
+}%
+\def\class@extensionfile#1#2#3{%
+ \@pass@ptions#3\@unusedoptionlist{#2}%
+ \global\let\@unusedoptionlist\@empty
+ \expandafter\class@ext@hook\csname#2.#3-h@@k\endcsname#1{#2}#3%
+}%
+\def\class@ext@hook#1#2#3#4{%
+ \@pushfilename@ltx
+ \makeatletter
+ \let\CurrentOption\@empty
+ \@reset@ptions
+ \let#1\@empty
+ \xdef\@currname{#3}%
+ \global\let\@currext#4%
+ \global\let\@clsextension\@currext
+ \input{#3.#4}%
+ \@ifl@ter#4{#3}#2{%
+ \class@info{Class extension later than: #2}%
+ }{%
+ \class@info{Class extension earlier: #2}%
+ \@@end
+ }%
+ #1%
+ \let#1\@undefined
+ \expandafter\@p@pfilename@ltx\@currnamestack@ltx\@nil
+ \@reset@ptions
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\@pushfilename}
+% \begin{macro}{\@p@pfilename}
+% \changes{4.1a}{2008/08/12}{Class extension mechanism \cs{@pushfilename@ltx} and \cs{@p@pfilename@ltx}.}
+% But! \LaTeX\ does not provide for a class extension other than \file{.cls}, therefore we
+% must extend \LaTeX's file name stack with the file extension of a class extension.
+% This way, procedures like
+% \cmd\ProvidesPackage, \cmd\OptionNotUsed, \cmd\ProcessOptions, \cmd\@reset@ptions\
+% will still work properly.
+% \begin{macrocode}
+\def\@pushfilename@ltx{%
+ \xdef\@currnamestack@ltx{%
+ {\@currname}%
+ {\@currext}%
+ {\@clsextension}%
+ {\the\catcode`\@}%
+ \@currnamestack@ltx
+ }%
+}%
+\def\@p@pfilename@ltx#1#2#3#4#5\@nil{%
+ \gdef\@currname{#1}%
+ \gdef\@currext{#2}%
+ \gdef\@clsextension{#3}%
+ \catcode`\@#4\relax
+ \gdef\@currnamestack@ltx{#5}%
+}%
+\global\let\@currnamestack@ltx\@empty
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% We carefully patch \LaTeX\ so that the current value of \cmd\@clsextension\ can be
+% restored after reading in a class file.
+%
+% \subsection{Type Tools}%
+%
+% \begin{macro}{\flushing}
+% Undoes \cmd\centering. Should also undo \cmd\raggedleft\ and \cmd\raggedright.
+% \begin{macrocode}
+\def\flushing{%
+ \let\\\@normalcr
+ \leftskip\z@skip
+ \rightskip\z@skip
+ \@rightskip\z@skip
+ \parfillskip\@flushglue
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \changes{4.1g}{2009/10/07}{(AO, 539) Use of double-backslash in argument of \cs{section} gives error. The \classname{textcase} package is involved.}%
+% \begin{macro}{\@centercr}
+% The \cmd\@centercr\ command is the replacement for \cmd\@normalcr\ when setting type centered or ragged.
+% Normally, the meaning of \cmd\\ is \cmd\@normalcr, which \LaTeX\ defines via \cmd\DeclareRobustCommand.
+% In centered or ragged typesetting, the meaning of \cmd\\ is \cmd\@centercr,
+% therefore it ought to to be defined via \cmd\DeclareRobustCommand\ (but unfortunately is not).
+% The fact that it is not is yet another of \LaTeX's early failures that will never get fixed.
+%
+% The following exemplar fails under \LaTeX\ version 2005/12/01, package \classname{textcase} 2004/10/07 v0.07:
+% \begin{verbatim}
+%\documentclass{article}%
+%\usepackage[overload]{textcase}
+%\begin{document}
+%\centering
+%\section{\MakeTextUppercase{Section\\title}}
+%Text
+%\end{document}
+% \end{verbatim}
+% The solution is to promote \cmd\@centercr\ to a robust command, just the same as \cmd\\.
+% We do that here without needing to know the meaning of the command.
+% \begin{macrocode}
+\expandafter\DeclareRobustCommand\expandafter\@centercr\expandafter{\@centercr}%
+% \end{macrocode}
+% \end{macro}
+%
+%
+% \subsection{Display Math}%
+%
+% \begin{macro}{\eqnarray@LaTeX}
+% \begin{macro}{\eqnarray@fleqn@fixed}
+% Team \LaTeX\ has stated they will never repair Leslie's broken definition of \env{eqnarray}.
+% Let us be bold\dots.
+%
+% Note on \classname{hyperref} package compatibility: that package overrides
+% \cmd\eqnarray\ by wrapping it up in a larger procedure, so its changes
+% are compatible with this package's changes.
+%
+% \changes{4.1a}{2008/01/19}{(AO, 475) I had not properly reproduced the LaTeX macro \cs{eqnarray}.}%
+% \begin{macrocode}
+\def\eqnarray@LaTeX{%
+ \stepcounter{equation}%
+ \def\@currentlabel{\p@equation\theequation}%
+ \global\@eqnswtrue
+ \m@th
+ \global\@eqcnt\z@
+ \tabskip\@centering
+ \let\\\@eqncr
+ $$\everycr{}\halign to\displaywidth\bgroup
+ \hskip\@centering$\displaystyle\tabskip\z@skip{##}$\@eqnsel
+ &\global\@eqcnt\@ne\hskip \tw@\arraycolsep \hfil${##}$\hfil
+ &\global\@eqcnt\tw@ \hskip \tw@\arraycolsep
+ $\displaystyle{##}$\hfil\tabskip\@centering
+ &\global\@eqcnt\thr@@ \hb@xt@\z@\bgroup\hss##\egroup
+ \tabskip\z@skip
+ \cr
+}
+\long\def\eqnarray@fleqn@fixed{%
+ \stepcounter{equation}\def\@currentlabel{\p@equation\theequation}%
+ \global\@eqnswtrue\m@th\global\@eqcnt\z@
+ \tabskip\ltx@mathindent
+ \let\\=\@eqncr
+ \setlength\abovedisplayskip{\topsep}%
+ \ifvmode\addtolength\abovedisplayskip{\partopsep}\fi
+ \addtolength\abovedisplayskip{\parskip}%
+ \setlength\belowdisplayskip{\abovedisplayskip}%
+ \setlength\belowdisplayshortskip{\abovedisplayskip}%
+ \setlength\abovedisplayshortskip{\abovedisplayskip}%
+ $$%
+ \everycr{}%
+ \halignt@\linewidth\bgroup
+ \hskip\@centering$\displaystyle\tabskip\z@skip{##}$\@eqnsel
+ &\global\@eqcnt\@ne
+ \hskip\tw@\eqncolsep
+ \hfil${{}##{}}$\hfil
+ &\global\@eqcnt\tw@
+ \hskip\tw@\eqncolsep
+ $\displaystyle{##}$\hfil\tabskip\@centering
+ &\global\@eqcnt\thr@@\hb@xt@\z@\bgroup\hss##\egroup
+ \tabskip\z@skip
+ \cr
+}%
+\@ifx{\eqnarray\eqnarray@LaTeX}{%
+ \class@info{Repairing broken LaTeX eqnarray}%
+ \let\eqnarray\eqnarray@fleqn@fixed
+ \newlength\eqncolsep
+ \setlength\eqncolsep\z@
+ \let\eqnarray@LaTeX\relax
+ \let\eqnarray@fleqn@fixed\relax
+}{}%
+% \end{macrocode}
+% The macro \cmd\ltx@mathindent\ is assigned to the \cmd\tabskip\ glue just
+% before the alignment preamble is expanded, the value therefore applying at the
+% left of the first column.
+%
+% The below value specifies the display math to be set centered, as is common practice.
+% Alternatively, \cmd\tabskip\ can be set to a different glue value, accomplishing
+% flush-left display math.
+%
+% Note that the \filename{fleqn.clo} package provides its own meaning for
+% the eqnarray environment, which is also broken. We do not patch that package, however.
+%
+% \changes{4.1d}{2009/03/27}{(AO, 511) Compatability with lineno.sty's erroneous way of detecting fleqn.clo}%
+% Bug note: The \filename{lineno.sty} package detects \filename{fleqn.clo}
+% by testing whether \cmd\mathindent\ is defined, instead of using correct \LaTeXe\ means.
+% Even though our eqnarray environment is modelled after \filename{fleqn.clo},
+% we must program defensively here.
+% \begin{macrocode}
+\def\ltx@mathindent{\@centering}%
+\def\set@eqnarray@skips{}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \changes{4.1g}{2009/10/07}{(AO, 525) Remove phantom paragraph above display math that is given in vertical mode}%
+%\begin{macro}{\prep@math}
+%\begin{macro}{\prep@math@patch}
+% If we are in vertical mode when display math mode is entered (via \verb+$$+),
+% \TeX\ will first enter horizontal mode, then display math mode; this results in a phantom paragraph
+% containing a single \cmd\hbox\ consisting of the \cmd\parindent\ box followed by the \cmd\parskipfillskip\ glue.
+% Of course, that \cmd\hbox\ is accompanied by \cmd\parskip\ glue and \cmd\baselineskip\ glue.
+%
+% The \cmd\prep@math\ procedure removes the \cmd\parindent\ box, thereby (magically) eliminating the phantom paragraph.
+% The \cmd\prep@math@patch\ procedure headpatches the \env{equation} and \env{eqnarray} environments
+% to accomplish this removal of the phantom paragraph.
+%
+% Note that there are three remaining ways to enter display math mode that we do not treat:
+% the \env{displaymath} environment (equivalent to \cmd\[/\cmd\]), and the primitive the \verb+$$+ markup.
+% I refrain from treating the first case because \env{displaymath} already detects the case
+% where it is entered from vertical mode: I do not wish to engage in the dubious enterprise of attempting to correct a procedure that is ill conceived from the outset.
+% As to the primitive \verb+$$+, there is no help for users who insist upon employing procedural markup in their documents.
+% in their documents.
+% \begin{macrocode}
+\def\prep@math{%
+ \@ifvmode{\everypar{{\setbox\z@\lastbox}}}{}%
+}%
+\def\prep@math@patch{%
+ \prepdef\equation{\prep@math}%
+ \prepdef\eqnarray{\prep@math}%
+}%
+% \end{macrocode}
+% A document class may invoke \cmd\prep@math@patch\ at any point it wishes to prevent the appearance
+% of the phantom paragraph: it may be a global declaration or a local one.
+% \end{macro}
+% \end{macro}
+%
+% We fail to patch \cmd\[, \cmd\equation, however.
+%
+% \subsection{Footnotes}
+%
+% \changes{4.0d}{2000/04/10}{(AO, 174) kernel fix}
+%
+% \begin{macro}{\footnotemark}
+% \begin{macro}{\footnotetest}
+% \begin{macro}{\ltx@xfootnote}
+% \begin{macro}{\ltx@footmark}
+% \begin{macro}{\ltx@foottext}
+% \begin{macro}{\ltx@make@current@footnote}
+% We repair an error in the \LaTeX\ kernel (see \file{ltfloat.dtx}) involving footnotes.
+% The symptom is that the \cmd\footnotemark\ command does not work properly within a \env{minipage} environment.
+% The source of the problem is in the way the \cmd\footnotemark\ and \cmd\@xfootnotemark\ procedures are defined:
+% they do not share the method, used by \cmd\footnote\ and other procedures, that allows a context switch to
+% change the way footnotes behave within a minipage environment.
+% This is a \LaTeX\ bug of long standing; our fix dates to 1987.
+%
+% While we are at it, we rewrite both the \cmd\footnote, \cmd\footnotemark\ and \cmd\footnotetext\ procedures,
+% achieving a cleaner separation of syntax and semantics.
+% Note that the \cmd\@footnotetext\ procedure is not involved in context switching;
+% \classname{hyperref} will take over that procedure, substituting its own processing around
+% its argument and passing this to \cmd\H@@footnotetext.
+% We anticipate this, and do our context switching on \cmd\H@@footnotetext.
+%
+% The \cmd\@makefnmark\ continues as the method of formatting the footnote mark.
+%
+% A note about the context switch mentioned above:
+% the \env{minipage} environment executes the following in order to alter the way footnotes
+% behave:\begin{verbatim}
+%\def\@mpfn{mpfootnote}%
+%\def\thempfn{\thempfootnote}%
+%\let\@footnotetext\@mpfootnotetext
+%\let\@makefnmark\@mpmakefnmark
+%\c@mpfootnote\z@\end{verbatim}
+% This code changes the counter used in autonumbered footnotes, the choice of footnote marker,
+% and the procedure used on the footnote text. Changing the counter is needed because minipage
+% footnotes are in their own sequence, and the footnote marker is customarily different within
+% a minipage. The procedure that works on the footnote text must be different because the footnotes
+% are placed at the bottom of the minipage, not the bottom of the text column.
+%
+% Note that \LaTeX\ initially defines \cmd\@mpfn\ as \texttt{footnote} and \cmd\thempfn\ as \cmd\thefootnote,
+% so we are initially doing general footnotes.
+%
+% Any procedure that establishes a minipage-like context (e.g., floats) can do the same as the mimipage context switch
+% illustrated above.
+% \changes{4.1a}{2008/06/30}{(AO, 438) Complete rewrite of footnote macros.}
+%
+% Three user-level command, \cmd\footnote, \cmd\footnotemark, and \cmd\footnotetext\ are defined (see the \LaTeX\ manual for user-level details).
+% \begin{macro}{\footnote}
+% The first user-level command is \cmd\footnote.
+% A simple way to look at this command is to think of it as
+% \cmd\footnotemark\ \oarg{number} \cmd\footnotetext\ \oarg{number} \arg{text},
+% where the optional argument is the same in both calls.
+% We also define a syntactical helper procedure \cmd\ltx@xfootnote.
+%
+% We employ the procedures \cmd\ltx@stp@footproc\ and \cmd\ltx@def@footproc,
+% passing in the procedure to execute, in this case \cmd\ltx@footmark,
+% which sets the footnote mark.
+% In any case, we end on the procedure \cmd\ltx@foottext, which sets the footnote text.
+% \begin{macrocode}
+\def\footnote{\@ifnextchar[\ltx@xfootnote\ltx@yfootnote}%
+\def\ltx@xfootnote[#1]{%
+ \ltx@def@footproc\ltx@footmark[#1]%
+ \expandafter\ltx@foottext\expandafter{\the\csname c@\@mpfn\endcsname}%
+}%
+\def\ltx@yfootnote{%
+ \ltx@stp@footproc\ltx@footmark
+ \expandafter\ltx@foottext\expandafter{\the\csname c@\@mpfn\endcsname}%
+}%
+% \end{macrocode}
+% The \cmd\footmark\ user-level command is next.
+% Here we use the procedures \cmd\ltx@stp@footproc\ and \cmd\ltx@def@footproc\ again,
+% but unlike \cmd\footnote, we do not set the footnote text.
+% \begin{macrocode}
+\def\footnotemark{\@ifnextchar[\ltx@xfootmark\ltx@yfootmark}%
+\def\ltx@xfootmark{\ltx@def@footproc\ltx@footmark}%
+\def\ltx@yfootmark{\ltx@stp@footproc\ltx@footmark}%
+\def\ltx@footmark#1{%
+ \leavevmode
+ \ifhmode\edef\@x@sf{\the\spacefactor}\nobreak\fi
+ \begingroup
+ \expandafter\ltx@make@current@footnote\expandafter{\@mpfn}{#1}%
+ \expandafter\@argswap@val\expandafter{\Hy@footnote@currentHref}{\hyper@linkstart {link}}%
+ \@makefnmark
+ \hyper@linkend
+ \endgroup
+ \ifhmode\spacefactor\@x@sf\fi
+ \relax
+}%
+% \end{macrocode}
+% The third user-level command is \cmd\footnotetext.
+% As with \cmd\footnotemark, we use the procedures \cmd\ltx@stp@footproc\ and \cmd\ltx@def@footproc,
+% this time passing in the procedure \cmd\ltx@foottext, which sets the footnote text.
+% \begin{macrocode}
+\def\footnotetext{\@ifnextchar[\ltx@xfoottext\ltx@yfoottext}%
+\def\ltx@xfoottext{\ltx@def@footproc\ltx@foottext}%
+\def\ltx@yfoottext{\ltx@stp@footproc\ltx@foottext}%
+\long\def\ltx@foottext#1#2{%
+ \begingroup
+ \expandafter\ltx@make@current@footnote\expandafter{\@mpfn}{#1}%
+ \@footnotetext{#2}%
+ \endgroup
+}%
+% \end{macrocode}
+% Here are the definitions of the procedures \cmd\ltx@stp@footproc\ and \cmd\ltx@def@footproc.
+% The require argument is the procedure to execute afterwards, and
+% \cmd\ltx@def@footproc\ parses a bracket-delimited argument (it is not optional).
+% In each case the given procedure is executed with an argument prepared for it:
+% the value of the footnote counter.
+% \begin{macrocode}
+\def\ltx@def@footproc#1[#2]{%
+ \begingroup
+ \csname c@\@mpfn\endcsname #2\relax
+ \unrestored@protected@xdef\@thefnmark{\thempfn}%
+ \expandafter\endgroup
+ \expandafter#1%
+ \expandafter{\the\csname c@\@mpfn\endcsname}%
+}%
+\def\ltx@stp@footproc#1{%
+ \expandafter\stepcounter\expandafter{\@mpfn}%
+ \protected@xdef\@thefnmark{\thempfn}%
+ \expandafter#1%
+ \expandafter{\the\csname c@\@mpfn\endcsname}%
+}%
+% \end{macrocode}
+% Here we provide for our good friend \classname{hyperref}
+% to enter in like a bull in a china shop. If it is not loaded,
+% we do what it would have done, but gentlier and without hypertext functionality.
+% \begin{macrocode}
+\appdef\class@documenthook{%
+ \let\footnote@latex\footnote
+ \@ifpackageloaded{hyperref}{}{%
+ \let\H@@footnotetext\@footnotetext
+ \def\@footnotetext{\H@@footnotetext}%
+ \let\H@@mpfootnotetext\@mpfootnotetext
+ \def\@mpfootnotetext{\H@@mpfootnotetext}%
+ }%
+}%
+% \end{macrocode}
+% In the following, we must use \LaTeX's rococco equipment in the form of \cmd\protected@edef,
+% because of the presence of a font switch in the meaning of \cmd\thempfootnote.
+% But, really, isn't this a sloppy conflation of semantics and presentation?
+% \begin{macrocode}
+\def\ltx@make@current@footnote#1#2{%
+ \csname c@#1\endcsname#2\relax
+ \protected@edef\Hy@footnote@currentHref{\@currentHref-#1.\csname the#1\endcsname}%
+}%
+\def\thempfootnote@latex{{\itshape \@alph \c@mpfootnote }}%
+\def\ltx@thempfootnote{\@alph\c@mpfootnote}%
+\@ifx{\thempfootnote\thempfootnote@latex}{%
+ \class@info{Repairing hyperref-unfriendly LaTeX definition of \string\mpfootnote}%
+ \let\thempfootnote\ltx@thempfootnote
+}{}%
+% \end{macrocode}
+%
+% Note on \classname{hyperref} compatibility:
+% In its ``Automated \LaTeX\ hypertext cross-references'',
+% the \classname{hyperref} package alters footnote processing,
+% but it does nothing to address the several issues of concern to us.
+%
+% The \classname{hyperref} package takes over the \cmd\@mpfootnotetext\ and
+% \cmd\@footnotetext\ procedures, wrapping the argument in its own code.
+% It also rewrites \cmd\@footnotemark, making it a hyperlink, and
+% \cmd\@xfootnotenext, removing from it all hypertext capabilities.
+%
+% However, if the \cmd\footnotemark\ command has been supplied with an optional argument,
+% \classname{hyperref}'s changes do not apply: it punts in this case.
+%
+% At the same time, it attempts to turn off its changes during
+% \cmd\maketitle\ processing, destroying one of the capabilities we desire.
+%
+% We make ourself \classname{hyperref} savvy:
+% we re-implement footnote processing, using \classname{hyperref} capabilities if that
+% package has been loaded.
+%
+% Any other package that rewrites \LaTeX's footnote macros will be incompatible
+% with this package.
+% \changes{4.1a}{2008/06/30}{(AO) Remove code that avoided changes to \cs{@xfootnotemark}}%
+%
+% Two thoughts about \classname{hyperref}: what for does it define \cmd\realfootnote?
+% Apparently even SR himself cannot remember.
+%
+% Also: a document class that desires high hypertext capabilities might
+% well wish to reimplement \cmd\maketitle\ so that footnotes called out from there
+% are hypertext links: the \classname{hyperref} package's
+% ``Automated \LaTeX\ hypertext cross-references''
+% does not do any of this:
+%\begin{quotation}
+% But the special footnotes in |\maketitle| are much too hard to deal with properly.
+% Let them revert to plain behaviour.
+%\end{quotation}
+% Note that the document class, in reimplementing \cmd\maketitle, must ensure
+% that the \classname{hyperref} package does not clobber its own definition!
+%
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\@footnotetext}
+% \begin{macro}{\@mpfootnotetext}
+% \begin{macro}{\@tpfootnotetext}
+% \begin{macro}{\make@footnotetext}
+% \begin{macro}{\set@footnotewidth}
+% The two procedures \cmd\@footnotetext\ and \cmd\@mpfootnotetext\ share code.
+% We make that explicit here.
+%
+% Note that the procedure calling \cmd\make@footnotetext\ will open a group
+% with \cmd\bgroup\ which is then closed by \cmd\minipagefootnote@drop.
+%
+% Difference from \LaTeX: here we do not set \cmd\floatingpenalty\ to infinity.
+% Doing this must date back to a time when \LaTeX\ could not accomodate split insertions (footnotes).
+% I cannot think of any other reason to do have done this.
+% At any rate, with the \classname{ltxgrid} package, split insertions are properly
+% taken care of, so we allow it.
+%
+% We provide the hook \cmd\set@footnotewidth\
+% that sets the footnote on a particular measure.
+% Some page grids are such as to set a footnote in a context where \cmd\columnwidth
+% is not the right parameter to use for the set width of a footnote.
+% In such a case, for the applicable scope, you should define
+% \cmd\set@footnotewidth\ to perform this job correctly.
+%
+% If we are setting type on multiple page grids, we must still ensure that all footnotes that
+% find their way into the \cmd\footins\ insert register are set on the same width.
+% This implies the need for a document to have an ``overall'' page grid, which determines the set width of all footnotes
+% with the exception of minipage footnotes.
+%
+% In general, remember that footnotes, like all insertions (including floats),
+% are a step outside of the galley context, and all aspects of insertions need to be
+% properly handled, including the set width.
+% \begin{macrocode}
+\def\@makefnmark{%
+ \hbox{%
+ \@textsuperscript{%
+ \normalfont\itshape\@thefnmark
+ }%
+ }%
+}%
+% \end{macrocode}
+% \begin{macrocode}
+\long\def\@footnotetext{%
+ \insert\footins\bgroup
+ \make@footnotetext
+}%
+% \end{macrocode}
+% \begin{macrocode}
+\long\def\@mpfootnotetext{%
+ \minipagefootnote@pick
+ \make@footnotetext
+}%
+% \end{macrocode}
+% Procedure \cmd\make@footnotetext\ sets the footnote \verb+#1+ into type, with the proper
+% font, color, leading, width, and label in effect.
+% It also establishes a strut and null glue at the end of the last paragraph of the footnote;
+% The strut helps compensate for the lack of \cmd\interlineskip\ glue between \cmd\insert s;
+% the glue establishes a feasible \cmd\vsplit\ point between footnotes.
+%
+% Note that in the title block (\classname{ltxfront}), the alternative definition, under the name \cmd\frontmatter@footnotetext, is used.
+% The only material difference there is the reference to \cmd\frontmatter@makefntext\ instead of \cmd\@makefntext.
+%
+% Dependency note: the \cmd\@makefntext\ procedure is used to further process the footnote text
+% and to execute the \cmd\@makefnmark\ procedure to produce the footnote mark.
+% The definition of the former is customarily found in the document class (hereunder that of \filename{article.cls}),
+% the latter in \filename{latex.ltx}. They are as follows:
+% \begin{verbatim}
+%\newcommand\@makefntext[1]{%
+% \parindent 1em\noindent
+% \hb@xt@1.8em{\hss\@makefnmark}%
+% #1%
+%}%
+%\def\@makefnmark{%
+% \hbox{\@textsuperscript{\normalfont\@thefnmark}}%
+%}%
+% \end{verbatim}
+%
+% \changes{4.1n}{2010/01/06}{(AO, 572) title block footnotes numbered independently from body footnotes}%
+% \begin{macrocode}
+\long\def\make@footnotetext#1{%
+ \set@footnotefont
+% \end{macrocode}
+% As noted above, we do \emph{not} do \cmd\floatingpenalty\ \cmd\@MM, as in standard \LaTeX.
+% \begin{macrocode}
+ \set@footnotewidth
+ \@parboxrestore
+ \protected@edef\@currentlabel{%
+% \end{macrocode}
+% Note that we employ \cmd\@mpfn\ as a level of redirection for the footnotecounter.
+% \begin{macrocode}
+ \csname p@\@mpfn\endcsname\@thefnmark
+ }%
+ \color@begingroup
+ \@makefntext{%
+ \rule\z@\footnotesep\ignorespaces#1%
+% \end{macrocode}
+% \changes{4.1n}{2010/01/02}{(AO, 571) allow split after last line of footnote}%
+% The following strut and glue are for spacing and splitting, as mentioned above.
+% \begin{macrocode}
+ \@finalstrut\strutbox\vadjust{\vskip\z@skip}%
+ }%
+ \color@endgroup
+ \minipagefootnote@drop
+}%
+% \end{macrocode}
+% \changes{4.1f}{2009/07/07}{(AO, 515) Hook for setting the font of a footnote}
+% \cmd\set@footnotefont\ is the procedure for setting the font of a footnote.
+% Other aspects of the environment may be set using this hook.
+% \begin{macrocode}
+\def\set@footnotefont{%
+ \reset@font\footnotesize
+ \interlinepenalty\interfootnotelinepenalty
+ \splittopskip\footnotesep
+ \splitmaxdepth\dp\strutbox
+}%
+% \end{macrocode}
+% \cmd\set@footnotewidth\ is the procedure for setting the width of a footnote.
+% The default page grid, a single, full-width column, sets footnotes on the width of the text.
+% \changes{4.1n}{2010/01/02}{(AO, 571) Interface \cs{set@footnotewidth} for determining the set width of footnotes}%
+% \begin{macrocode}
+\def\set@footnotewidth{\set@footnotewidth@one}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+%
+% \subsection{Floats}
+%
+% \subsubsection{Usage notes}%
+% We extend the \LaTeX\ kernel for three purposes:
+% \begin{enumerate}
+%
+% \item
+% When the \cmd\footnote\ command is used within the
+% scope of a float, we do as \env{minipage} does.
+%
+% \item
+% We provide a mechanism to write floats out to an external
+% stream for temporary storage (deferred floats).
+%
+% \item
+% We provide mechanism for placing a float \texttt{here}
+% invariably, that is, floats are unfloated.
+% This mechanism is used to read the external stream mentioned above.
+%
+% \end{enumerate}
+%
+% To use these mechanisms, the document class should
+% define a float, say, \env{figure} as per usual, and in addition:
+% \begin{enumerate}
+%
+% \item
+% Optionally define an alternative, say \env{figure@write} as follows:
+% \begin{verbatim}\newenvironment{figure@write}{%
+% \write@float{figure}%
+%}{%
+% \endwrite@float
+%}\end{verbatim}
+% That is, the alternative environment executes \cmd\write@float\
+% instead of \cmd\@float.
+% Note that this step is not needed if the float environment
+% is defined in the simple way of \file{classes.dtx}.
+% However, an environment like \env{longtable} will require it.
+%
+% \item
+% Install into \cmd\AtBeginDocument\ a call to \cmd\do@if@floats,
+% with the float name and an appropriate file extension as its arguments.
+% \begin{verbatim}\appdef\class@documenthook{\do@if@floats{figure}{.fgx}}\end{verbatim}
+%
+% \item
+% Optionally define a text entity \cmd\figuresname\ that will
+% be the text of the head that is set over the
+% deferred floats.
+% If not defined, there will be no head.
+%
+% \item
+% Optionally define a user-level command to allow
+% the document to determine where the figures are printed out
+% (default is to print at end of document). E.g.,
+% \begin{verbatim}\newcommand\printfigures{\print@float{figure}}\end{verbatim}
+% \item
+% Install into \cmd\appdef\cmd\class@enddocumenthook\ a call to \cmd\printfigures,
+% or, if the latter is not defined, as follows:
+% \begin{verbatim}\appdef\class@enddocumenthook{\print@float{figure}}\end{verbatim}
+% Note that installing this command into \cmd\AtBeginDocument
+% is best done earlier than calls that assume the last page of
+% the document is at hand.
+%
+% \end{enumerate}
+%
+% \subsubsection{Robustifying fragile commands}%
+% Certain of \LaTeX's commands cannot be written out to a file or appear within a \cmd\mark\ command argument
+% because they do calculations during expansion.
+% We provide for a little help, but without changing the meanings of these commands.
+%
+% \begin{macro}{\addtocontents}
+% \begin{macro}{\robustify@contents}
+%
+% \changes{4.1a}{2008/06/30}{(AO) Make \cs{addtocontents} a \cs{long} \cs{def}; gobble up \cs{footnote}}%
+% \begin{macrocode}
+\def\robustify@contents{%
+ \let \label \@gobble
+ \let \index \@gobble
+ \let \glossary \@gobble
+ \let\footnote \@gobble
+ \def\({\string\(}%
+ \def\){\string\)}%
+ \def\\{\string\\}%
+}%
+\long\def\addtocontents#1#2{%
+ \protected@write\@auxout{\robustify@contents}{\string \@writefile {#1}{#2}}%
+}%
+% \end{macrocode}
+%
+% \end{macro}
+% \end{macro}
+%
+% \subsubsection{Preparing for the \classname{hyperref package}}%
+%
+% \begin{macro}{\addcontentsline}
+% \begin{macro}{\label}
+% \begin{macro}{\ltx@contentsline}
+% \changes{4.1b}{2008/08/12}{Refine toc processing: provide default.}%
+% The \classname{hyperref} package assumes that the \cmd\contentsline\ command will be given four arguments.
+% Therefore it cannot successfully process a \filename{.toc} file that had been written by standard \LaTeX.
+% We fix things up by always writing that fourth argument and by supplying a \cmd\contentsline\ command that
+% can read them.
+%
+% We also give the \cmd\newlabel\ command's second argument five tokens.
+%
+% Finally, we wrap \LaTeX's \cmd\contentsline\ command with code to detect the case
+% where the expected procedure is not defined, and we give it a syntax with no semantics.
+%
+% We switch over to this new definition only after \classname{hyperref} has loaded.
+% \begin{macrocode}
+\def\addcontentsline#1#2#3{%
+ \addtocontents{#1}{%
+ \protect\contentsline{#2}{#3}{\thepage}{}%
+ }%
+}%
+\def\label#1{%
+ \@bsphack
+ \protected@write\@auxout{}{%
+ \string\newlabel{#1}{{\@currentlabel}{\thepage}{}{}{}}%
+ }%
+ \@esphack
+}%
+\def\ltx@contentsline#1{%
+ \expandafter\@ifnotrelax\csname l@#1\endcsname{}{%
+ \expandafter\let\csname l@#1\endcsname\@gobbletwo
+ }%
+ \contentsline@latex{#1}%
+}%
+\appdef\document@inithook{%
+ \let\contentsline@latex\contentsline
+ \let\contentsline\ltx@contentsline
+}%
+% \end{macrocode}
+%
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \subsubsection{Footnotes within floats, unfloating floats, float font}%
+%
+% \begin{macro}{\caption}
+% DPC: Er a bit of a hack, but seems best way of supporting normal
+% \LaTeX\ syntax at this point: If a caption is used below a table,
+% then put out the footnotes before the caption.
+% \changes{4.0b}{1999/06/20}{Support the hack with \cs{prepdef}, and delay until \cs{AtBeginDocument} time, since \classname{hyperref} clobbers \cs{caption}.}
+% \begin{macrocode}
+\appdef\class@documenthook{%
+ \prepdef\caption{\minipagefootnote@here}%
+}%
+% \end{macrocode}
+%
+% Note on \classname{hyperref} compatibility:
+% this change to the \cmd\caption\ command is compatible with the
+% ``Automated \LaTeX\ hypertext cross-references'' patches of that package.
+%
+% All the same, I think Sebastian's changes to \cmd\caption\ and \cmd\@caption\
+% could bear with some improvement.
+% The following implementation requires knowing only the pattern part of the
+% \cmd\@caption\ macro:
+%\begin{verbatim}
+%\def\caption{%
+% \H@refstepcounter\@captype
+% \hyper@makecurrent{\@captype}%
+% \@dblarg{\H@caption\@captype}%
+%}%
+%\def\H@caption#1[#2]#3{%
+% \@caption{#1}[#2]{%
+% \ifHy@nesting
+% \hyper@@anchor{\@currentHref}{#3}%
+% \else
+% \hyper@@anchor{\@currentHref}{\relax}#3%
+% \fi
+% }%
+%}\end{verbatim}
+%
+% \end{macro}
+%
+% \begin{macro}{\minipagefootnote@init}
+% \begin{macro}{\minipagefootnote@here}
+% \begin{macro}{\minipagefootnote@foot}
+% \begin{macro}{\minipagefootnote@pick}
+% \begin{macro}{\minipagefootnote@drop}
+% Procedure to deal with footnotes accumulated within a minipage environment.
+% These procedures encapsulate all uses of the \cmd\@mpfootins\ box.
+%
+%
+% Note: \cmd\minipagefootnote@here\ must \emph{not} be executed within the MVL!
+% \begin{macrocode}
+\def\minipagefootnote@init{%
+ \setbox\@mpfootins\box\voidb@x
+}%
+\def\minipagefootnote@pick{%
+ \global\setbox\@mpfootins\vbox\bgroup
+ \unvbox\@mpfootins
+}%
+\def\minipagefootnote@drop{%
+ \egroup
+}%
+\def\minipagefootnote@here{%
+ \par
+ \@ifvoid\@mpfootins{}{%
+ \vskip\skip\@mpfootins
+ \fullinterlineskip
+ \@ifinner{%
+ \vtop{\unvcopy\@mpfootins}%
+ {\setbox\z@\lastbox}%
+ }{}%
+ \unvbox\@mpfootins
+ }%
+}%
+\def\minipagefootnote@foot{%
+ \@ifvoid\@mpfootins{}{%
+ \insert\footins\bgroup\unvbox\@mpfootins\egroup
+ }%
+}%
+\def\endminipage{%
+ \par
+ \unskip
+ \minipagefootnote@here
+ \@minipagefalse %% added 24 May 89
+ \color@endgroup
+ \egroup
+ \expandafter\@iiiparbox\@mpargs{\unvbox\@tempboxa}%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\floats@sw}
+% \changes{4.1b}{2008/08/12}{Default assignment of \cs{float@sw} now, not at \cs{AtBeginDocument} time.}%
+% The Boolean \cmd\floats@sw\ signifies that floats are to be floated;
+% if false, that floats are to be deferred to the end of the document.
+% Note that the assignment of this Boolean is to be overridden by
+% the document class in response to user-selected options.
+% \begin{macrocode}
+\@booleantrue\floats@sw
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@xfloat}
+% \begin{macro}{\@mpmakefntext}
+% The float start-code is redefined to set up footnotes in the style of minipage.
+% Also, the \cmd\floats@sw\ Boolean informs us that floats are to be
+% all placed \texttt{here}.
+% Note that, to protect against the Boolean being undefined
+% at this late hour, we default it globally to true.
+%
+% \changes{4.0b}{1999/06/20}{AO: Removed superfluous \cs{def}s, changed to using \cs{floats@sw} as the flag. Also stopped using DPC's \cs{if@twocolumn} flag: using \cs{floats@sw} instead. Also added \cs{par}\cs{vskip}\cs{z@skip} after the \cs{minipagefootnotes} so that the float box would have zero depth like the kernel one. }
+% \changes{4.0c}{1999/11/13}{(AO, 110) Install hooks for endfloats processing}
+% \changes{4.0d}{2000/04/10}{(AO, 127) Floats placed [h] to allow page breaks}
+% \changes{4.0d}{2000/05/19}{(AO, 224) Hyperref compatibility.}
+% \changes{4.0e}{2000/11/16}{(AO, 221) Remove samepage command from @xfloat@prep: If the float can break over pages, we want better control.}
+% \changes{4.1a}{2008/07/07}{\cs{@xfloat@prep} calls \cs{ltx@footnote@pop} to restore the original \cs{ltx@footmark} and \cs{ltx@foottext} procedures, in case footnote processing has switched.}
+% \changes{4.1b}{2008/08/12}{No need to protect against undefined \cs{float@sw}}
+%
+% \begin{macrocode}
+\let\@xfloat@LaTeX\@xfloat
+\def\@xfloat#1[#2]{%
+ \@xfloat@prep
+ \@nameuse{fp@proc@#2}%
+ \floats@sw{\@xfloat@LaTeX{#1}[#2]}{\@xfloat@anchored{#1}[]}%
+}%
+\def\@xfloat@prep{%
+ \ltx@footnote@pop
+ \def\@mpfn{mpfootnote}%
+ \def\thempfn{\thempfootnote}%
+ \c@mpfootnote\z@
+ \let\H@@footnotetext\H@@mpfootnotetext
+}%
+\let\ltx@footnote@pop\@empty
+\def\@xfloat@anchored#1[#2]{%
+ \def\@captype{#1}%
+ \begin@float@pagebreak
+ \let\end@float\end@float@anchored
+ \let\end@dblfloat\end@float@anchored
+ \hsize\columnwidth
+ \@parboxrestore
+ \@floatboxreset
+ \minipagefootnote@init
+}%
+\def\end@float@anchored{%
+ \minipagefootnote@here
+ \par\vskip\z@skip
+ \par
+ \end@float@pagebreak
+}%
+\def\begin@float@pagebreak{\par\addvspace\intextsep}%
+\def\end@float@pagebreak{\par\addvspace\intextsep}%
+\def\@mpmakefntext#1{%
+ \parindent=1em
+ \noindent
+ \hb@xt@1em{\hss\@makefnmark}%
+ #1%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+%
+% \subsubsection{Writing floats out to a file}%
+%
+% \begin{macro}{\do@if@floats}
+% \changes{4.1b}{2008/08/12}{No need to protect against undefined \cs{float@sw}}
+% The procedure \cmd\do@if@floats\ should be executed at
+% \cmd\class@documenthook\ time: it arranges to write out
+% the floats of the given class to a temporary file, to be
+% read back later (deferred floats),
+% given that \cmd\floats@sw\ is false.
+% Note that, to protect against the Boolean being undefined
+% at this late hour, we default it globally to true.
+% \begin{macrocode}
+\def\do@if@floats#1#2{%
+ \floats@sw{}{%
+% \end{macrocode}
+% Open the stream to save out the document's floats of this class.
+% \begin{macrocode}
+ \expandafter\newwrite
+ \csname#1write\endcsname
+ \expandafter\def
+ \csname#1@stream\endcsname{\jobname#2}%
+ \expandafter\immediate
+ \expandafter\openout
+ \csname#1write\endcsname
+ \csname#1@stream\endcsname\relax
+% \end{macrocode}
+% Swap environments.
+% If the class writer has defined, e.g., \env{figure@write},
+% then we use this as the procedure to execute for writing
+% the float out to the external stream.
+% Otherwise, the replacement of \cmd\@float\ by \cmd\write@float\
+% should do the right thing for float environments defined
+% in the simple way of \classname{classes.dtx}.
+% \begin{macrocode}
+ \@ifxundefined\@float@LaTeX{%
+ \let\@float@LaTeX\@float
+ \let\@dblfloat@LaTeX\@dblfloat
+ \let\@float\write@float
+ \let\@dblfloat\write@floats
+ }{}%
+ \let@environment{#1@float}{#1}%
+ \let@environment{#1@floats}{#1*}%
+ \@ifxundefined@cs{#1@write}{}{%
+ \let@environment{#1}{#1@write}%
+ }%
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\print@float}
+% The procedure \cmd\print@float\ prints out the
+% deferred floats.
+% \changes{4.0b}{1999/06/20}{only execute if there really were floats of the given type}
+% \changes{4.0c}{1999/11/13}{*-form mandates pagebreak at each float; only print section head if there is something there.}
+% \changes{4.0d}{2000/05/23}{Allow things to break over pages by setting array@default.}
+% \changes{4.1b}{2008/08/12}{If class option \classoption{lengthcheck} is in effect, log the height of this float class.}
+%
+% Here, we make use of the \cmd\floats@sw\ Boolean to select
+% the non-floating type of processing.
+% \begin{macrocode}
+\def\triggerpar{\leavevmode\@@par}%
+\def\oneapage{\def\begin@float@pagebreak{\newpage}\def\end@float@pagebreak{\newpage}}%
+\def\print@float#1#2{%
+ \lengthcheck@sw{%
+ \total@float{#1}%
+ }{}%
+ \@ifxundefined@cs{#1write}{}{%
+ \begingroup
+ \@booleanfalse\floats@sw
+ #2%
+ \raggedbottom
+ \def\array@default{v}% floats must
+ \let\@float\@float@LaTeX
+ \let\@dblfloat\@dblfloat@LaTeX
+ \let\trigger@float@par\triggerpar
+ \let@environment{#1}{#1@float}%
+ \let@environment{#1*}{#1@floats}%
+ \expandafter\prepdef\csname#1\endcsname{\trigger@float@par}%
+ \expandafter\prepdef\csname#1*\endcsname{\trigger@float@par}%
+ \@namedef{fps@#1}{h!}%
+ \expandafter\immediate
+ \expandafter\closeout
+ \csname#1write\endcsname
+ \everypar{%
+ \global\let\trigger@float@par\relax
+ \global\everypar{}\setbox\z@\lastbox
+ \@ifxundefined@cs{#1sname}{}{%
+ \begin@float@pagebreak
+ \expandafter\section
+ \expandafter*%
+ \expandafter{%
+ \csname#1sname\endcsname
+ }%
+ }%
+ }%
+ \input{\csname#1@stream\endcsname}%
+ \endgroup
+ \global\expandafter\let\csname#1write\endcsname\relax
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\tally@float}
+% \begin{macro}{\total@float}
+% \changes{4.1b}{2008/08/12}{Tally and log the height of a float class}
+% If we are tallying column inches, \cmd\tally@float\ tallies a contribution to \cmd\ftype@\ \cmd\@captype,
+% depending upon the width of \cmd\@currbox.
+% In effect, each float class is tallied in two sections, one for narrow, one for wide floats.
+%
+% If statistics are wanted, \cmd\total@float\ logs the tally for the given float class.
+% \changes{4.1f}{2009/07/10}{(AO, 518) Tally register overflow when locument is long}
+% The quantity \cmd\@twopowerfourteen\ is $2^{14}$, \cmd\@twopowertwo\ is $2^{2}$.
+% \begin{macrocode}
+\chardef\@xvi=16\relax
+\mathchardef\@twopowerfourteen="4000
+\mathchardef\@twopowertwo="4
+\def\tally@float#1{%
+ \begingroup
+% \end{macrocode}
+% We strip all but the least significant 5 bits from \cmd\count\ \cmd\@currbox,
+% and put them into \cmd\@tempcnta. We then subtract 16 from \cmd\count\ \cmd\@currbox
+% (unless this would make it negative), effectively reversing the process
+% carried out in \cmd\@float.
+% \begin{macrocode}
+ \@tempcnta\count\@currbox
+ \divide\@tempcnta\@xxxii
+ \multiply\@tempcnta\@xxxii
+ \advance\count\@currbox-\@tempcnta
+ \divide\@tempcnta\@xxxii
+ \@ifnum{\count\@currbox>\@xvi}{%
+ \advance\count\@currbox-\@xvi\@booleantrue\@temp@sw
+ }{%
+ \@booleanfalse\@temp@sw
+ }%
+% \end{macrocode}
+% If so desired, we log the characteristics of this float object:
+% float class and float placement parameters, height, depth, and width.
+% \begin{macrocode}
+ \show@box@size@sw{%
+ \class@info{Float #1
+ (\the\@tempcnta)[\@temp@sw{16+}{}\the\count\@currbox]^^J%
+ (\the\ht\@currbox+\the\dp\@currbox)X\the\wd\@currbox
+ }%
+ }{}%
+ \endgroup
+% \end{macrocode}
+% Here we tally the height of this float object.
+% \begin{macrocode}
+ \expandafter\let
+ \expandafter\@tempa
+ \csname fbox@\csname ftype@#1\endcsname\endcsname
+ \@ifnotrelax\@tempa{%
+ \@ifhbox\@tempa{%
+ \setbox\@tempboxa\vbox{\unvcopy\@currbox\hrule}%
+ \dimen@\ht\@tempboxa
+ \divide\dimen@\@twopowerfourteen
+ \@ifdim{\wd\@tempboxa<\textwidth}{%
+ \advance\dimen@\ht\@tempa
+ \global\ht\@tempa\dimen@
+ }{%
+ \advance\dimen@\dp\@tempa
+ \global\dp\@tempa\dimen@
+ }%
+ }{}%
+ }{}%
+}%
+\def\total@float#1{%
+ \expandafter\let
+ \expandafter\@tempa
+ \csname fbox@\csname ftype@#1\endcsname\endcsname
+ \@ifnotrelax\@tempa{%
+ \@ifhbox\@tempa{%
+ \@tempdima\the\ht\@tempa\divide\@tempdima\@twopowertwo\@tempcnta\@tempdima
+ \@tempdimb\the\dp\@tempa\divide\@tempdimb\@twopowertwo\@tempcntb\@tempdimb
+ \class@info{Total #1: Column(\the\@tempcnta pt), Page(\the\@tempcnta pt)}%
+ }{}%
+ }{}%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\write@float}
+% \begin{macro}{\write@floats}
+% \begin{macro}{\write@@float}
+% Handles the case where the name of the float is the same as
+% that of the stream. Note that
+% \env{longtable} does \emph{not} fit this case.
+% Note also: \cmd\write@float\ is \emph{not} a user-level environment,
+% therefore it is properly not defined with \cmd\newenvironment.
+% \begin{macrocode}
+\def\write@float#1{\write@@float{#1}{#1}}%
+\def\endwrite@float{\@Esphack}%
+\def\write@floats#1{\write@@float{#1*}{#1}}%
+\def\endwrite@floats{\@Esphack}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\write@@float}
+% \changes{4.0b}{1999/06/20}{AO: Fixed spurious \texttt{CR} and (return) characters in output file. Also, if the document did not have the \cs{end}\texttt{figure} on a line of its own, the macro wouldn't work. Fixed.}
+% \begin{macrocode}
+\def\write@@float#1#2{%
+ \ifhmode
+ \@bsphack
+ \fi
+ \chardef\@tempc\csname#2write\endcsname
+ \toks@{\begin{#1}}%
+ \def\@tempb{#1}%
+ \expandafter\let\csname end#1\endcsname\endwrite@float
+ \catcode`\^^M\active
+ \@makeother\{\@makeother\}\@makeother\%
+ \write@floatline
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\write@floatline}
+% \begin{macro}{\@write@floatline}
+% \begin{macro}{\float@end@tag}
+% The procedure \cmd\write@floatline\ only parses;
+% it passes its result to \cmd\@write@floatline, which
+% writes the line to output, then tests the line
+% for the \cmd\end\arg{float} tokens with
+% aid of the \cmd\float@end@tag\ procedure.
+% \begin{macrocode}
+\begingroup
+ \catcode`\[\the\catcode`\{\catcode`\]\the\catcode`\}\@makeother\{\@makeother\}%
+ \gdef\float@end@tag#1\end{#2}#3\@nul[%
+ \def\@tempa[#2]%
+ \@ifx[\@tempa\@tempb][\end[#2]][\write@floatline]%
+ ]%
+ \obeylines%
+ \gdef\write@floatline#1^^M[%
+ \begingroup%
+ \newlinechar`\^^M%
+ \toks@\expandafter[\the\toks@#1]\immediate\write\@tempc[\the\toks@]%
+ \endgroup%
+ \toks@[]%
+ \float@end@tag#1\end{}\@nul%
+ ]%
+\endgroup
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+%
+%
+% \subsection{Counters}
+% The following definitions override those of the \LaTeX\ kernel,
+% providing for a greater range of inputs.
+% \begin{macrocode}
+\def\@alph#1{\ifcase#1\or a\or b\or c\or d\else\@ialph{#1}\fi}
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\@ialph#1{\ifcase#1\or \or \or \or \or e\or f\or g\or h\or i\or j\or
+ k\or l\or m\or n\or o\or p\or q\or r\or s\or t\or u\or v\or w\or x\or
+ y\or z\or aa\or bb\or cc\or dd\or ee\or ff\or gg\or hh\or ii\or jj\or
+ kk\or ll\or mm\or nn\or oo\or pp\or qq\or rr\or ss\or tt\or uu\or
+ vv\or ww\or xx\or yy\or zz\else\@ctrerr\fi}
+% \end{macrocode}
+%
+%
+% \subsection{Customization of Sections}%
+%
+% Patch the standard \LaTeX\ sectioning procedure to:
+%\begin{itemize}
+%\item
+% Allow a sectioning command to trigger the title page, or more generally
+% to recognize that it is the first object in the document,
+% so we headpatch \cmd\@startsection.
+%
+%\item
+% Allow a tail command in |#6| to uppercase the title, so we retain
+% DPC's braces.
+%
+%\item
+% Allow each type of sectioning command to format its number differently,
+% so we generalize \cmd\@seccntformat.
+%
+%\item
+% Allow each type of sectioning command to format its argument differently,
+% so we generalize \cmd\@hangfrom.
+%
+%\item
+% Allow the starred form of the command to
+% mark (the running head) and
+% make an entry in the TOC,
+% so we put \cmd\@ssect\ on the same footing as \cmd\@sect.
+%
+% Note that the tokens passed to the TOC now are \emph{not}
+% the optional argument of the command, but the required.
+% This means that the user can no longer use the former
+% to put variant content in to the TOC as the Manual says.
+%
+% Instead, the optional argument is used to put an alternative
+% title into the running headers, a better choice.
+%
+%\end{itemize}
+%
+% \begin{macro}{\@startsection}
+% Patch a head hook into the basic sectioning command.
+% Treat \cmd\@sect\ and \cmd\@ssect\ on an equal footing:
+% now their pattern parts are identical.
+% \begin{macrocode}
+\def\@startsection#1#2#3#4#5#6{%
+ \@startsection@hook
+ \if@noskipsec \leavevmode \fi
+ \par
+ \@tempskipa #4\relax
+ \@afterindenttrue
+ \ifdim \@tempskipa <\z@
+ \@tempskipa -\@tempskipa \@afterindentfalse
+ \fi
+ \if@nobreak
+ \everypar{}%
+ \else
+ \addpenalty\@secpenalty\addvspace\@tempskipa
+ \fi
+ \@ifstar
+ {\@dblarg{\@ssect@ltx{#1}{#2}{#3}{#4}{#5}{#6}}}%
+ {\@dblarg{\@sect@ltx {#1}{#2}{#3}{#4}{#5}{#6}}}%
+}%
+\def\@startsection@hook{}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@sect}
+% When defining \cmd\@svsec, do not expand \cmd\@seccntformat.
+% Put brace characters back where they were before David Carlisle got at them
+% (i.e., as if \cmd\@hangfrom\ had two arguments).
+% Protect the mark mechanism from an undefined meaning.
+% Pass |#8| to the TOC instead of |#7|.
+% Remove \cmd\relax\ from the replacement part of \cmd\@svsec.
+%
+% The procedure \cmd\@hangfrom\ and \cmd\@runin@to\ can
+% be used to process the argument of the head.
+% The head can define, e.g., \cmd\@hangfrom@section, to
+% do its own processing.
+%
+% In using \cmd\H@refstepcounter\ in place of \cmd\refstepcounter\ we rely on
+% either loading before any package that patches the latter, or
+% the convention that the former is the original \LaTeX\ procedure.
+%
+% \begin{macrocode}
+\class@info{Repairing broken LateX \string\@sect}%
+\def\@sect@ltx#1#2#3#4#5#6[#7]#8{%
+ \@ifnum{#2>\c@secnumdepth}{%
+ \def\H@svsec{\phantomsection}%
+ \let\@svsec\@empty
+ }{%
+ \H@refstepcounter{#1}%
+ \def\H@svsec{%
+ \phantomsection
+ }%
+ \protected@edef\@svsec{{#1}}%
+ \@ifundefined{@#1cntformat}{%
+ \prepdef\@svsec\@seccntformat
+ }{%
+ \expandafter\prepdef
+ \expandafter\@svsec
+ \csname @#1cntformat\endcsname
+ }%
+ }%
+ \@tempskipa #5\relax
+ \@ifdim{\@tempskipa>\z@}{%
+ \begingroup
+ \interlinepenalty \@M
+ #6{%
+ \@ifundefined{@hangfrom@#1}{\@hang@from}{\csname @hangfrom@#1\endcsname}%
+ {\hskip#3\relax\H@svsec}{\@svsec}{#8}%
+ }%
+ \@@par
+ \endgroup
+ \@ifundefined{#1mark}{\@gobble}{\csname #1mark\endcsname}{#7}%
+ \addcontentsline{toc}{#1}{%
+ \@ifnum{#2>\c@secnumdepth}{%
+ \protect\numberline{}%
+ }{%
+ \protect\numberline{\csname the#1\endcsname}%
+ }%
+ #8}%
+ }{%
+ \def\@svsechd{%
+ #6{%
+ \@ifundefined{@runin@to@#1}{\@runin@to}{\csname @runin@to@#1\endcsname}%
+ {\hskip#3\relax\H@svsec}{\@svsec}{#8}%
+ }%
+ \@ifundefined{#1mark}{\@gobble}{\csname #1mark\endcsname}{#7}%
+ \addcontentsline{toc}{#1}{%
+ \@ifnum{#2>\c@secnumdepth}{%
+ \protect\numberline{}%
+ }{%
+ \protect\numberline{\csname the#1\endcsname}%
+ }%
+ #8}%
+ }%
+ }%
+ \@xsect{#5}%
+}%
+\def\@hang@from#1#2#3{\@hangfrom{#1#2}#3}%
+\def\@runin@to #1#2#3{#1#2#3}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@ssect}
+% Put brace characters back where they were before David Carlisle got at them
+% (as if \cmd\@hangfrom\ has two arguments).
+% Possibly set a mark.
+% Make a TOC entry.
+%
+% Note that, for compatibility with the \classname{hyperref} package, we
+% need to provide the interface required by that package
+% (actually required by \file{pdfmark.def} and \file{nameref.sty}),
+% namely
+% the definition of \cmd\@currentlabelname\ (but now removed),
+% the insertion of the procedure \cmd\Sectionformat\ (but why is this needed?), and
+% the call to \cmd\phantomsection\ (which must precede the call to \cmd\addcontentsline).
+% We also have to sidestep the patch to \cmd\@ssect\ in that same file, therefore
+% we use a different control sequence name in the call from \cmd\@startsection.
+% \changes{4.0c}{1999/11/13}{(AO, 116) Hyperref compatibility}
+% \changes{4.0f}{2001/07/13}{(AO, 404) Hyperref compatibility}
+% \begin{macrocode}
+\def\@ssect@ltx#1#2#3#4#5#6[#7]#8{%
+% \end{macrocode}
+% Removed |\def\@currentlabelname{#8}|
+% \begin{macrocode}
+ \def\H@svsec{\phantomsection}%
+ \@tempskipa #5\relax
+ \@ifdim{\@tempskipa>\z@}{%
+ \begingroup
+ \interlinepenalty \@M
+ #6{%
+ \@ifundefined{@hangfroms@#1}{\@hang@froms}{\csname @hangfroms@#1\endcsname}%
+% \end{macrocode}
+% Removed |{\hskip#3\relax\H@svsec}{\Sectionformat{#8}{#1}}|
+% \begin{macrocode}
+ {\hskip#3\relax\H@svsec}{#8}%
+ }%
+ \@@par
+ \endgroup
+ \@ifundefined{#1smark}{\@gobble}{\csname #1smark\endcsname}{#7}%
+ \addcontentsline{toc}{#1}{\protect\numberline{}#8}%
+ }{%
+ \def\@svsechd{%
+ #6{%
+ \@ifundefined{@runin@tos@#1}{\@runin@tos}{\csname @runin@tos@#1\endcsname}%
+% \end{macrocode}
+% Removed |{\hskip#3\relax\H@svsec}{\Sectionformat{#8}{#1}}|
+% \begin{macrocode}
+ {\hskip#3\relax\H@svsec}{#8}%
+ }%
+ \@ifundefined{#1smark}{\@gobble}{\csname #1smark\endcsname}{#7}%
+ \addcontentsline{toc}{#1}{\protect\numberline{}#8}%
+ }%
+ }%
+ \@xsect{#5}%
+}%
+\def\@hang@froms#1#2{#1#2}%
+\def\@runin@tos #1#2{#1#2}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\init@hyperref}
+% \changes{4.1b}{2008/08/12}{Acquire \classname{hyperref} savoire}
+% Document classes that incorporate this package will be \classname{hyperref}-savvy.
+% (To accomplish this, we ensure that \cmd\hyperanchor\ and \cmd\hyper@last\ are both defined.)
+% Being \classname{hyperref}-savvy levels some requirements on us, but the benefits are many.
+%
+% One is that the TOC will not get amnesia and require a full set of three typesetting runs before its formatting is stable.
+% Instead, only two runs are required: the first updates the auxiliary file, the second the TOC.
+% However, the formatting of the document does not change.
+%
+% Another aspect of being \classname{hyperref}-savvy is that the syntax of commands in the \file{.aux} file will not change
+% if \classname{hyperref} is turned on or off.
+%
+% Note that \cmd\hyper@anchorstart\ and \cmd\hyper@anchorend\ constitute the programming interface
+% for a hypertext anchor (the target of a hypertext link); \cmd\hyper@linkstart\ and \cmd\hyper@linkend\
+% are the interface for a hypertext link.
+% \begin{macrocode}
+\def\init@hyperref{%
+ \providecommand\phantomsection{}%
+ \providecommand\hyper@makecurrent[1]{}%
+ \providecommand\Hy@raisedlink[1]{}%
+ \providecommand\hyper@anchorstart[1]{}%
+ \providecommand\hyper@anchorend{}%
+ \providecommand\hyper@linkstart[2]{}%
+ \providecommand\hyper@linkend{}%
+ \providecommand\@currentHref{}%
+}%
+\let\H@refstepcounter\refstepcounter
+\appdef\document@inithook{%
+ \init@hyperref
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\sec@upcase}
+% Upper case for sections (optional upper case items). These are
+% created so that some headings can be toggled between mixed case and
+% upper case readily.
+% Headings that might be changed can be wrapped in the style file in
+% \cmd\sec@upcase\arg{text} constructs;
+% the expansion of \cmd\sec@upcase\ is
+% controlled here. It is \cmd\relax\ by default (mixed case heads), and
+% can easily be changed to \cmd\uppercase\ if desired.
+% If mixed-case headings are wanted by the editor, authors {\em must}
+% supply mixed case text, although this is what authors should be doing
+% anyway.
+% (Mixed can be converted to upper,
+% but the reverse transformation cannot be automated.)
+%
+% The following setting gives the \LaTeX\ default.
+% \begin{macrocode}
+\def\sec@upcase#1{\relax{#1}}%
+% \end{macrocode}
+% \end{macro}
+%
+%
+% \subsection{Patch the \env{tabular} and \env{array} Environments}
+%
+% \begin{macro}{\endtabular}
+% \begin{macro}{\endarray}
+% \changes{4.1b}{2008/08/12}{Patch the array package even later: after all package patches go in.}
+% We headpatch the begin processing and tailpatch the end processing
+% of the \env{tabular} and \env{array} environments.
+% A document class can define these hooks as needed.
+%
+% We proceed with care to make further patches to
+% support tabulars that break over pages.
+% Our patches will not necessarily be effective for
+% other packages that replace the \LaTeX\ \env{array} and \env{tabular}
+% environments. I know of none that do so.
+% \begin{macrocode}
+\appdef\document@inithook{%
+ \@ifpackageloaded{array}{\switch@array}{\switch@tabular}%
+ \prepdef\endtabular{\endtabular@hook}%
+ \@provide\endtabular@hook{}%
+ \prepdef\endarray{\endarray@hook}%
+ \@provide\endarray@hook{}%
+ \providecommand\array@hook{}%
+% \end{macrocode}
+% Install, effectively, a head patch to \cmd\tabular.
+% In order to avoid interference from, e.g., the \classname{array} package,
+% we must perform this patch only \emph{after} packages load.
+% \changes{4.0c}{1999/11/13}{(AO, 130) Interference from array package}
+% \begin{macrocode}
+ \prepdef\@tabular{\tabular@hook}%
+ \@provide\tabular@hook{}%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\switch@tabular}
+% \begin{macro}{\switch@array}
+% The two procedures \cmd\switch@tabular\ and \cmd\switch@array\
+% apply needed patches to the various tabular procedures,
+% the former applying to the \LaTeX\ kernel, the latter to the
+% required \classname{array} package (and to the number of other
+% required packages that load it).
+%
+% \begin{macrocode}
+\def\switch@tabular{%
+ \let\@array@sw\@array@sw@array
+ \@ifx{\@array\@array@LaTeX}{%
+ \@ifx{\multicolumn\multicolumn@LaTeX}{%
+ \@ifx{\@tabular\@tabular@LaTeX}{%
+ \@ifx{\@tabarray\@tabarray@LaTeX}{%
+ \@ifx{\array\array@LaTeX}{%
+ \@ifx{\endarray\endarray@LaTeX}{%
+ \@ifx{\endtabular\endtabular@LaTeX}{%
+ \@ifx{\@mkpream\@mkpream@LaTeX}{%
+ \@ifx{\@addamp\@addamp@LaTeX}{%
+ \@ifx{\@arrayacol\@arrayacol@LaTeX}{%
+ \@ifx{\@tabacol\@tabacol@LaTeX}{%
+ \@ifx{\@arrayclassz\@arrayclassz@LaTeX}{%
+ \@ifx{\@tabclassiv\@tabclassiv@LaTeX}{%
+ \@ifx{\@arrayclassiv\@arrayclassiv@LaTeX}{%
+ \@ifx{\@tabclassz\@tabclassz@LaTeX}{%
+ \@ifx{\@classv\@classv@LaTeX}{%
+ \@ifx{\hline\hline@LaTeX}{%
+ \@ifx{\@tabularcr\@tabularcr@LaTeX}{%
+ \@ifx{\@xtabularcr\@xtabularcr@LaTeX}{%
+ \@ifx{\@xargarraycr\@xargarraycr@LaTeX}{%
+ \@ifx{\@yargarraycr\@yargarraycr@LaTeX}{%
+ \true@sw
+ }{%
+ \false@sw
+ }%
+ }{%
+ \false@sw
+ }%
+ }{%
+ \false@sw
+ }%
+ }{%
+ \false@sw
+ }%
+ }{%
+ \false@sw
+ }%
+ }{%
+ \false@sw
+ }%
+ }{%
+ \false@sw
+ }%
+ }{%
+ \false@sw
+ }%
+ }{%
+ \false@sw
+ }%
+ }{%
+ \false@sw
+ }%
+ }{%
+ \false@sw
+ }%
+ }{%
+ \false@sw
+ }%
+ }{%
+ \false@sw
+ }%
+ }{%
+ \false@sw
+ }%
+ }{%
+ \false@sw
+ }%
+ }{%
+ \false@sw
+ }%
+ }{%
+ \false@sw
+ }%
+ }{%
+ \false@sw
+ }%
+ }{%
+ \false@sw
+ }%
+ }{%
+ \false@sw
+ }%
+ }{%
+ \false@sw
+ }%
+ {%
+ \class@info{Patching LaTeX tabular.}%
+ }{%
+ \class@info{Unrecognized LaTeX tabular. Please update this document class! (Proceeding with fingers crossed.)}%
+ }%
+ \let\@array\@array@ltx
+ \let\multicolumn\multicolumn@ltx
+ \let\@tabular\@tabular@ltx
+ \let\@tabarray\@tabarray@ltx
+ \let\array\array@ltx
+ \let\endarray\endarray@ltx
+ \let\endtabular\endtabular@ltx
+ \let\@mkpream\@mkpream@ltx
+ \let\@addamp\@addamp@ltx
+ \let\@arrayacol\@arrayacol@ltx
+ \let\@tabacol\@tabacol@ltx
+ \let\@arrayclassz\@arrayclassz@ltx
+ \let\@tabclassiv\@tabclassiv@ltx
+ \let\@arrayclassiv\@arrayclassiv@ltx
+ \let\@tabclassz\@tabclassz@ltx
+ \let\@classv\@classv@ltx
+ \let\hline\hline@ltx
+ \let\@tabularcr\@tabularcr@ltx
+ \let\@xtabularcr\@xtabularcr@ltx
+ \let\@xargarraycr\@xargarraycr@ltx
+ \let\@yargarraycr\@yargarraycr@ltx
+}%
+% \end{macrocode}
+% \changes{4.1b}{2008/08/12}{(AO, 505) Try to accommodate \classname{colortbl}.}
+% \begin{macrocode}
+\def\switch@array{%
+ \@ifpackageloaded{colortbl}{\let\switch@array@info\colortbl@message}{\let\switch@array@info\array@message}%
+ \let\@array@sw\@array@sw@LaTeX
+ \@ifx{\@array\@array@array}{%
+ \@ifx{\@tabular\@tabular@array}{%
+ \@ifx{\@tabarray\@tabarray@array}{%
+ \@ifx{\array\array@array}{%
+ \@ifx{\endarray\endarray@array}{%
+ \@ifx{\endtabular\endtabular@array}{%
+ \@ifx{\@mkpream\@mkpream@array}{%
+ \@ifx{\@classx\@classx@array}{%
+ \@ifx{\insert@column\insert@column@array}{%
+ \@ifx{\@arraycr\@arraycr@array}{%
+ \@ifx{\@xarraycr\@xarraycr@array}{%
+ \@ifx{\@xargarraycr\@xargarraycr@array}{%
+ \@ifx{\@yargarraycr\@yargarraycr@array}{%
+ \true@sw
+ }{%
+ \false@sw
+ }%
+ }{%
+ \false@sw
+ }%
+ }{%
+ \false@sw
+ }%
+ }{%
+ \false@sw
+ }%
+ }{%
+ \false@sw
+ }%
+ }{%
+ \false@sw
+ }%
+ }{%
+ \false@sw
+ }%
+ }{%
+ \false@sw
+ }%
+ }{%
+ \false@sw
+ }%
+ }{%
+ \false@sw
+ }%
+ }{%
+ \false@sw
+ }%
+ }{%
+ \false@sw
+ }%
+ }{%
+ \false@sw
+ }{%
+ \class@info{Patching array package.}%
+ }{%
+ \switch@array@info
+ }%
+ \let\@array \@array@array@new
+ \let\@@array \@array % Cosi fan tutti
+ \let\@tabular \@tabular@array@new
+ \let\@tabarray \@tabarray@array@new
+ \let\array \array@array@new
+ \let\endarray \endarray@array@new
+ \let\endtabular\endtabular@array@new
+ \let\@mkpream \@mkpream@array@new
+ \let\@classx \@classx@array@new
+ \let\@arrayacol\@arrayacol@ltx
+ \let\@tabacol \@tabacol@ltx
+ \let\insert@column\insert@column@array@new
+ \expandafter\let\csname endtabular*\endcsname\endtabular % Cosi fan tutti
+ \let\@arraycr \@arraycr@new
+ \let\@xarraycr \@xarraycr@new
+ \let\@xargarraycr\@xargarraycr@new
+ \let\@yargarraycr\@yargarraycr@new
+}%
+\def\array@message{%
+ \class@info{Unrecognized array package. Please update this document class! (Proceeding with fingers crossed.)}%
+}%
+\def\colortbl@message{%
+ \class@info{colortbl package is loaded. (Proceeding with fingers crossed.)}%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\@array@sw}
+% The Boolean \cmd\@array@sw\ must be different depending on
+% whether the \classname{array} package is loaded.
+% \begin{macrocode}
+\def\@array@sw@LaTeX{\@ifx{\\\@tabularcr}}%
+\def\@array@sw@array{\@ifx{\d@llarbegin\begingroup}}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@tabular}
+% \changes{4.1b}{2008/08/12}{(AO, 505) try to accommodate \classname{colortbl}.}
+% We provide the old versions of \cmd\@tabular\ along with the respective new versions.
+% The change here is to avoid committing to LR mode. That will be done later (as late as possible, naturally).
+%
+% Compatibility note: I had done \cmd\let\ \cmd\col@sep\ \cmd\@undefined\ here, but
+% this was not compatible with \classname{colortbl}. I have removed that statement.
+% \begin{macrocode}
+\def\@tabular@LaTeX{%
+ \leavevmode
+ \hbox\bgroup$%
+ \let\@acol\@tabacol
+ \let\@classz\@tabclassz
+ \let\@classiv\@tabclassiv
+ \let\\\@tabularcr
+ \@tabarray
+}%
+\def\@tabular@ltx{%
+ \let\@acoll\@tabacoll
+ \let\@acolr\@tabacolr
+ \let\@acol\@tabacol
+ \let\@classz\@tabclassz
+ \let\@classiv\@tabclassiv
+ \let\\\@tabularcr
+ \@tabarray
+}%
+\def\@tabular@array{%
+ \leavevmode
+ \hbox\bgroup$%
+ \col@sep\tabcolsep
+ \let\d@llarbegin\begingroup
+ \let\d@llarend\endgroup
+ \@tabarray
+}%
+\def\@tabular@array@new{%
+ \let\@acoll\@tabacoll
+ \let\@acolr\@tabacolr
+ \let\@acol\@tabacol
+% \end{macrocode}
+% \let\col@sep\@undefined
+% \begin{macrocode}
+ \let\d@llarbegin\begingroup
+ \let\d@llarend\endgroup
+ \@tabarray
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@tabarray}
+% Here we provide old and new versions of the \cmd\@tabarray\ procedure.
+% The change here is to parametrize the default vertical alignment,
+% which is 'c' in standard \LaTeX.
+% Under some circumstances, we want to change this to, say, 'v'.
+%
+% \changes{4.1b}{2008/08/12}{(AO, 505) try to accommodate \classname{colortbl}.}
+% FIXME: must decouple \env{array} and \env{tabular}. Done (it seems).
+%
+% Note on \classname{colortbl}: this package head-patches \cmd\@tabarray
+% with its own command \cmd\CT@start, and tails onto \cmd\endarray\ with \cmd\CT@end.
+% It fortuitously does the former at \cmd\AtBeginDocument\ time, and, fortuitously,
+% we do not patch \cmd\endarray, which it overwrites.
+% \begin{macrocode}
+\def\@tabarray@LaTeX{%
+ \m@th\@ifnextchar[\@array{\@array[c]}%
+}%
+\def\@tabarray@ltx{%
+ \m@th\@ifnextchar[\@array{\expandafter\@array\expandafter[\array@default]}%
+}%
+\def\@tabarray@array{%
+ \@ifnextchar[{\@@array}{\@@array[c]}%
+}%
+\def\@tabarray@array@new{%
+ \@ifnextchar[{\@@array}{\expandafter\@@array\expandafter[\array@default]}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@tabularcr}
+% \begin{macro}{\@tbpen}
+% \begin{macro}{\@tabularcr}
+% \begin{macro}{\@xtabularcr}
+% \begin{macro}{\@xargarraycr}
+% \begin{macro}{\@yargarraycr}
+% \begin{macro}{\@arraycr}
+% \begin{macro}{\@xarraycr}
+% We provide for the \cmd\\ command within \env{tabular} to provide control over page breaking, just the same as
+% that of \env{eqnarray}.
+%
+% The count register \cmd\intertabularlinepenalty\ is similar to \cmd\interdisplaylinepenalty: it is the penalty
+% associated with each row of a tabular. When it is set to \cmd\@M, the tabular will cleave together.
+%
+% The count register \cmd\@tbpen\ is similar to \cmd\@eqpen: it memorizes the penalty to use after the current tabular row.
+% If the \cmd\\ command is in its star form, then \cmd\@eqpen\ is set to \cmd\@M.
+%
+% We append code to \cmd\samepage\ so that a tabular within its scope will cleave together.
+%
+% We keep the standard definition of \cmd\@tabularcr\ in \cmd\@tabularcr@LaTeX\ for reference,
+% and provide a new definition that works like \cmd\@eqncr: it sets \cmd\@tbpen\ to \cmd\@M\ if the star was given.
+%
+% We also provide new versions of \cmd\@xtabularcr, \cmd\@xargarraycr, and \cmd\@yargarraycr, all of which invoke \cmd\@tbpen.
+%
+% The \cmd\switch@tabular\ procedure switches in the new definitions.
+% \begin{macrocode}
+\newcount\intertabularlinepenalty
+\intertabularlinepenalty=100
+\newcount\@tbpen
+\appdef\samepage{\intertabularlinepenalty\@M}%
+\def\@tabularcr@LaTeX{{\ifnum 0=`}\fi \@ifstar \@xtabularcr \@xtabularcr}%
+\def\@tabularcr@ltx{{\ifnum 0=`}\fi \@ifstar {\global \@tbpen \@M \@xtabularcr }{\global \@tbpen \intertabularlinepenalty \@xtabularcr }}%
+\def\@xtabularcr@LaTeX{\@ifnextchar [\@argtabularcr {\ifnum 0=`{\fi }\cr }}%
+\def\@xtabularcr@ltx{\@ifnextchar [\@argtabularcr {\ifnum 0=`{\fi }\cr \noalign {\penalty \@tbpen }}}%
+\def\@xargarraycr@LaTeX#1{\@tempdima #1\advance \@tempdima \dp \@arstrutbox \vrule \@height \z@ \@depth \@tempdima \@width \z@ \cr}%
+\def\@xargarraycr@ltx#1{\@tempdima #1\advance \@tempdima \dp \@arstrutbox \vrule \@height \z@ \@depth \@tempdima \@width \z@ \cr \noalign {\penalty \@tbpen }}%
+\def\@yargarraycr@LaTeX#1{\cr \noalign {\vskip #1}}%
+\def\@yargarraycr@ltx#1{\cr \noalign {\penalty \@tbpen \vskip #1}}%
+% \end{macrocode}
+%
+% If the \classname{array} package has been loaded, we must alter the meanings of
+% \cmd\@arraycr, \cmd\@xarraycr, \cmd\@xargarraycr, and \cmd\@yargarraycr.
+% In this case, it is \cmd\switch@array\ that switches in the new definitions.
+% \begin{macrocode}
+\def\@arraycr@array{%
+ \relax
+ \iffalse{\fi\ifnum 0=`}\fi
+ \@ifstar \@xarraycr \@xarraycr
+}%
+\def\@arraycr@new{%
+ \relax
+ \iffalse{\fi\ifnum 0=`}\fi
+ \@ifstar {\global \@tbpen \@M \@xarraycr }{\global \@tbpen \intertabularlinepenalty \@xarraycr }%
+}%
+\def\@xarraycr@array{%
+ \@ifnextchar [%]
+ \@argarraycr {\ifnum 0=`{}\fi\cr}%
+}%
+\def\@xarraycr@new{%
+ \@ifnextchar [%]
+ \@argarraycr {\ifnum 0=`{}\fi\cr \noalign {\penalty \@tbpen }}%
+}%
+\def\@xargarraycr@array#1{%
+ \unskip
+ \@tempdima #1\advance\@tempdima \dp\@arstrutbox
+ \vrule \@depth\@tempdima \@width\z@
+ \cr
+}%
+\def\@xargarraycr@new#1{%
+ \unskip
+ \@tempdima #1\advance\@tempdima \dp\@arstrutbox
+ \vrule \@depth\@tempdima \@width\z@
+ \cr
+ \noalign {\penalty \@tbpen }%
+}%
+\def\@yargarraycr@array#1{%
+ \cr
+ \noalign{\vskip #1}%
+}%
+\def\@yargarraycr@new#1{%
+ \cr
+ \noalign{\penalty \@tbpen \vskip #1}%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\array}
+% We provide old and new versions of the \cmd\array\ procedure for both \LaTeX\ and the \classname{array} package.
+% The change here is to accomodate the new procedures that will be called for the array boundaries, even
+% though at present they are not special.
+% A thought: here is where matrices can be readily accomodated.
+% \begin{macrocode}
+\def\array@LaTeX{%
+ \let\@acol\@arrayacol
+ \let\@classz\@arrayclassz
+ \let\@classiv\@arrayclassiv
+ \let\\\@arraycr
+ \let\@halignto\@empty
+ \@tabarray
+}%
+\def\array@ltx{%
+ \@ifmmode{}{\@badmath$}%
+ \let\@acoll\@arrayacol
+ \let\@acolr\@arrayacol
+ \let\@acol\@arrayacol
+ \let\@classz\@arrayclassz
+ \let\@classiv\@arrayclassiv
+ \let\\\@arraycr
+ \let\@halignto\@empty
+ \@tabarray
+}%
+\def\array@array{%
+ \col@sep\arraycolsep
+ \def\d@llarbegin{$}\let\d@llarend\d@llarbegin\gdef\@halignto{}%
+ \@tabarray
+}
+\def\array@array@new{%
+ \@ifmmode{}{\@badmath$}%
+ \let\@acoll\@arrayacol
+ \let\@acolr\@arrayacol
+ \let\@acol\@arrayacol
+% \end{macrocode}
+% \changes{4.1b}{2008/08/12}{(AO, 505) try to accommodate \classname{colortbl}.}
+% Removed: \verb+\let\col@sep\@undefined+
+% \begin{macrocode}
+ \def\d@llarbegin{$}%
+ \let\d@llarend\d@llarbegin
+ \gdef\@halignto{}%
+ \@tabarray
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@array}
+% Here we provide old and new versions of \cmd\@array.
+% The change here is to provide a convenient, flexible, and extensible
+% mechanism for new vertical alignment options.
+%
+% Instead of testing the optional argument with \cmd\if, we
+% use a dispatcher based on \cmd\csname.
+%
+% We also refrain from using \cmd\ialign, which would set
+% the \cmd\tabskip\ to the wrong value.
+%
+% Finally, the procedure to set the \cmd\@arstrutbox\
+% is broken out so that it can be patched.
+% \begin{macrocode}
+\def\@array@LaTeX[#1]#2{%
+ \if #1t\vtop \else \if#1b\vbox \else \vcenter \fi\fi
+ \bgroup
+ \setbox\@arstrutbox\hbox{%
+ \vrule \@height\arraystretch\ht\strutbox
+ \@depth\arraystretch \dp\strutbox
+ \@width\z@}%
+ \@mkpream{#2}%
+ \edef\@preamble{%
+ \ialign \noexpand\@halignto
+ \bgroup \@arstrut \@preamble \tabskip\z@skip \cr}%
+ \let\@startpbox\@@startpbox \let\@endpbox\@@endpbox
+ \let\tabularnewline\\%
+ \let\par\@empty
+ \let\@sharp##%
+ \set@typeset@protect
+ \lineskip\z@skip\baselineskip\z@skip
+ \ifhmode \@preamerr\z@ \@@par\fi
+ \@preamble
+}%
+\def\@array@ltx[#1]#2{%
+ \@nameuse{@array@align@#1}%
+ \set@arstrutbox
+ \@mkpream{#2}%
+ \prepdef\@preamble{%
+ \tabskip\tabmid@skip
+ \@arstrut
+ }%
+ \appdef\@preamble{%
+ \tabskip\tabright@skip
+ \cr
+ \array@row@pre
+ }%
+% \let\@startpbox\@@startpbox
+% \let\@endpbox\@@endpbox
+ \let\tabularnewline\\%
+ \let\par\@empty
+ \let\@sharp##%
+ \set@typeset@protect
+ \lineskip\z@skip\baselineskip\z@skip
+ \tabskip\tableft@skip\relax
+ \ifhmode \@preamerr\z@ \@@par\fi
+ \everycr{}%
+ \expandafter\halign\expandafter\@halignto\expandafter\bgroup\@preamble
+}%
+%
+\def\set@arstrutbox{%
+ \setbox\@arstrutbox\hbox{%
+ \vrule \@height\arraystretch\ht\strutbox
+ \@depth\arraystretch \dp\strutbox
+ \@width\z@
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@array@array}
+%
+% \begin{macrocode}
+\def\@array@array[#1]#2{%
+ \@tempdima \ht \strutbox
+ \advance \@tempdima by\extrarowheight
+ \setbox \@arstrutbox \hbox{\vrule
+ \@height \arraystretch \@tempdima
+ \@depth \arraystretch \dp \strutbox
+ \@width \z@}%
+ \begingroup
+ \@mkpream{#2}%
+ \xdef\@preamble{\noexpand \ialign \@halignto
+ \bgroup \@arstrut \@preamble
+ \tabskip \z@ \cr}%
+ \endgroup
+ \@arrayleft
+ \if #1t\vtop \else \if#1b\vbox \else \vcenter \fi \fi
+ \bgroup
+ \let \@sharp ##\let \protect \relax
+ \lineskip \z@
+ \baselineskip \z@
+ \m@th
+ \let\\\@arraycr \let\tabularnewline\\\let\par\@empty \@preamble
+}%
+\def\@array@array@new[#1]#2{%
+ \@tempdima\ht\strutbox
+ \advance\@tempdima by\extrarowheight
+ \setbox\@arstrutbox\hbox{%
+ \vrule \@height\arraystretch\@tempdima
+ \@depth \arraystretch\dp\strutbox
+ \@width \z@
+ }%
+ \begingroup
+ \@mkpream{#2}%
+ \xdef\@preamble{\@preamble}%
+ \endgroup
+ \prepdef\@preamble{%
+ \tabskip\tabmid@skip
+ \@arstrut
+ }%
+ \appdef\@preamble{%
+ \tabskip\tabright@skip
+ \cr
+ \array@row@pre
+ }%
+ \@arrayleft
+ \@nameuse{@array@align@#1}%
+ \m@th
+ \let\\\@arraycr
+ \let\tabularnewline\\%
+ \let\par\@empty
+ \let\@sharp##%
+ \set@typeset@protect
+ \lineskip\z@\baselineskip\z@
+ \tabskip\tableft@skip
+ \everycr{}%
+ \expandafter\halign\expandafter\@halignto\expandafter\bgroup\@preamble
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\endarray}
+% Here we provide old and new versions of \cmd\endarray.
+% The change here is to use a single procedure to close
+% out any array-like structure, namely \cmd\endarray@ltx.
+% It merely closes out the \cmd\halign.
+% \begin{macrocode}
+\def\endarray@LaTeX{%
+ \crcr\egroup\egroup
+}%
+\def\endarray@ltx{%
+ \crcr\array@row@pst\egroup\egroup
+}%
+\def\endarray@array{%
+ \crcr \egroup \egroup \@arrayright \gdef\@preamble{}%
+}%
+\def\endarray@array@new{%
+ \crcr\array@row@pst\egroup\egroup % Same as \endarray@ltx
+ \@arrayright
+ \global\let\@preamble\@empty
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\endtabular}
+%
+% \begin{macrocode}
+\def\endtabular@LaTeX{%
+ \crcr\egroup\egroup $\egroup
+}%
+\def\endtabular@ltx{%
+ \endarray
+}%
+\def\endtabular@array{%
+ \endarray $\egroup
+}%
+\def\endtabular@array@new{%
+ \endarray
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{endtabular*}
+% Here we provide a proper definition for the star-form of \enve{endtabular}.
+% It is one of the enduring curiosities that the \LaTeX\ kernel continues to use
+% dangerously and inappropriately ``optimized'' definitions for such commands.
+% \begin{macrocode}
+\@namedef{endtabular*}{\endtabular}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\multicolumn}
+%
+% \begin{macrocode}
+\long\def\multicolumn@LaTeX#1#2#3{%
+ \multispan{#1}\begingroup
+ \@mkpream{#2}%
+ \def\@sharp{#3}\set@typeset@protect
+ \let\@startpbox\@@startpbox\let\@endpbox\@@endpbox
+ \@arstrut \@preamble\hbox{}\endgroup\ignorespaces
+}%
+\long\def\multicolumn@ltx#1#2#3{%
+ \multispan{#1}%
+ \begingroup
+ \@mkpream{#2}%
+ \def\@sharp{#3}%
+ \set@typeset@protect
+ %\let\@startpbox\@@startpbox\let\@endpbox\@@endpbox
+ \@arstrut
+ \@preamble
+ \hbox{}%
+ \endgroup
+ \ignorespaces
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@array@align@}
+% \begin{macro}{\array@default}
+% Here are the various procedures for the vertical alignment options.
+% The change from standard \LaTeX\ is that we do not go into math mode
+% in every case: only when required by \cmd\vcenter.
+% Also, we use \cmd\aftergroup\ to close out the boxes and modes we have started.
+% It requires only that each procedure issue exactly one unmatched \cmd\bgroup.
+%
+% We establish here the default vertical alignment.
+% \begin{macrocode}
+\def\@array@align@t{\leavevmode\vtop\bgroup}%
+\def\@array@align@b{\leavevmode\vbox\bgroup}%
+\def\@array@align@c{\leavevmode\@ifmmode{\vcenter\bgroup}{$\vcenter\bgroup\aftergroup$\aftergroup\relax}}%
+\def\@array@align@v{%
+ \@ifmmode{%
+ \@badmath
+ \vcenter\bgroup
+ }{%
+ \@ifinner{%
+ $\vcenter\bgroup\aftergroup$
+ }{%
+ \@@par\bgroup
+ }%
+ }%
+}%
+\def\array@default{c}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\array@row@pre}
+% \begin{macro}{\array@row@pst}
+% \begin{macro}{\array@row@rst}
+% The procedure \cmd\array@row@rst\ reestablishes a default context for
+% an alignment, so that they can be nested.
+% Any environment or procedure that alters the way alignments are formatted
+% must patch this procedure to restore from that alteration.
+% To start things off, we equate \cmd\@array@align@v\ to \cmd\@array@align@c,
+% because it does not make sense to do the former in any context other
+% than the MVL or in a list that will be unboxed onto the MVL.
+% \begin{macrocode}
+\def\array@row@rst{%
+ \let\@array@align@v\@array@align@c
+}%
+\def\array@row@pre{}%
+\def\array@row@pst{}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\toprule}
+% \begin{macro}{\colrule}
+% \begin{macro}{\botrule}
+% Default definitions for \cmd\toprule, \cmd\colrule, \cmd\botrule
+% \begin{macrocode}
+\newcommand\toprule{\tab@rule{\column@font}{\column@fil}{\frstrut}}%
+\newcommand\colrule{\unskip\lrstrut\\\tab@rule{\body@font}{}{\frstrut}}%
+\newcommand\botrule{\unskip\lrstrut\\\noalign{\hline@rule}{}}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\hline}
+% \begin{macrocode}
+\def\hline@LaTeX{%
+ \noalign{\ifnum0=`}\fi\hrule \@height \arrayrulewidth \futurelet
+ \reserved@a\@xhline
+}%
+\def\hline@ltx{%
+ \noalign{%
+ \ifnum0=`}\fi
+ \hline@rule
+ \futurelet\reserved@a\@xhline
+ % \noalign ended in \@xhline
+}%
+\def\@xhline@unneeded{%
+ \say\reserved@a
+ \ifx\reserved@a\hline
+ \vskip\doublerulesep
+ \vskip-\arrayrulewidth
+ \fi
+ \ifnum0=`{\fi}%
+}%
+\def\tab@rule#1#2#3{%
+ \crcr
+ \noalign{%
+ \hline@rule
+ \gdef\@arstrut@hook{%
+ \global\let\@arstrut@hook\@empty
+ #3%
+ }%
+ \gdef\cell@font{#1}%
+ \gdef\cell@fil{#2}%
+ }%
+}%
+\def\column@font{}%
+\def\column@fil{}%
+\def\body@font{}%
+\def\cell@font{}%
+\def\frstrut{}%
+\def\lrstrut{}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@arstrut@hline}
+% \begin{macro}{\@arstrut@org}
+% \begin{macro}{\@arstrut@hook}
+% \begin{macro}{\@arstrutbox@hline}
+% \begin{macro}{\set@arstrutbox}
+% \begin{macro}{\hline@rule}
+% The procedure \cmd\@arstrut@hline\ is substantially the same as
+% \cmd\@arstrut, except the strut copied in is \cmd\@arstrutbox@hline
+% instead of \cmd\@arstrutbox.
+%
+% The procedure \cmd\@arstrut@hook\ is redefined in \cmd\tab@rule!
+%
+% The register \cmd\@arstrutbox@hline.
+%
+% We append to \cmd\set@arstrutbox\ the code necessary to set a strut following an \cmd\hline.
+%
+% The procedure \cmd\hline@rule\ lays down a rule, and changes the meaning of \cmd\@arstrut\
+% so that the next line will be correctly strutted.
+%
+% The \cmd\@arstrut@hline@clnc\ is a klootch, a magic number.
+% \begin{macrocode}
+\def\@arstrut@hline{%
+ \relax
+ \@ifmmode{\copy}{\unhcopy}\@arstrutbox@hline
+ \@arstrut@hook
+}%
+%
+\let\@arstrut@org\@arstrut
+\def\@arstrut@hook{%
+ \global\let\@arstrut\@arstrut@org
+}%
+%
+\newbox\@arstrutbox@hline
+\appdef\set@arstrutbox{%
+ \setbox\@arstrutbox@hline\hbox{%
+ \setbox\z@\hbox{$0^{0}_{}$}%
+ \dimen@\ht\z@\advance\dimen@\@arstrut@hline@clnc
+ \@ifdim{\dimen@<\arraystretch\ht\strutbox}{\dimen@=\arraystretch\ht\strutbox}{}%
+ \vrule \@height\dimen@
+ \@depth\arraystretch \dp\strutbox
+ \@width\z@
+ }%
+}%
+%
+\def\hline@rule{%
+ \hrule \@height \arrayrulewidth
+ \global\let\@arstrut\@arstrut@hline
+}%
+\def\@arstrut@hline@clnc{2\p@}% % Klootch: magic number
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\tableft@skip}
+% \begin{macrocode}
+\def\tableft@skip{\z@skip}%
+\def\tabmid@skip{\z@skip}%\@flushglue
+\def\tabright@skip{\z@skip}%
+\def\tableftsep{\tabcolsep}%
+\def\tabmidsep{\tabcolsep}%
+\def\tabrightsep{\tabcolsep}%
+\def\cell@fil{}%
+\def\pbox@hook{}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@arstrut}
+% \begin{macrocode}
+\appdef\@arstrut{\@arstrut@hook}%
+\let\@arstrut@hook\@empty
+\def\@addtopreamble{\appdef\@preamble}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@mkpream}
+% \begin{macrocode}
+\def\@mkpream@LaTeX#1{%
+ \@firstamptrue\@lastchclass6
+ \let\@preamble\@empty
+ \let\protect\@unexpandable@protect
+ \let\@sharp\relax
+ \let\@startpbox\relax\let\@endpbox\relax
+ \@expast{#1}%
+ \expandafter\@tfor \expandafter
+ \@nextchar \expandafter:\expandafter=\reserved@a\do
+ {\@testpach\@nextchar
+ \ifcase \@chclass \@classz \or \@classi \or \@classii \or \@classiii
+ \or \@classiv \or\@classv \fi\@lastchclass\@chclass}%
+ \ifcase \@lastchclass \@acol
+ \or \or \@preamerr \@ne\or \@preamerr \tw@\or \or \@acol \fi
+}%
+\def\@mkpream@ltx#1{%
+ \@firstamptrue
+ \@lastchclass6
+ \let\@preamble\@empty
+ \let\protect\@unexpandable@protect
+ \let\@sharp\relax
+%\let\@startpbox\relax\let\@endpbox\relax
+ \@expast{#1}%
+ \expandafter\@tfor\expandafter\@nextchar\expandafter:\expandafter=\reserved@a
+ \do{%
+ \expandafter\@testpach\expandafter{\@nextchar}%
+ \ifcase\@chclass
+ \@classz
+ \or
+ \@classi
+ \or
+ \@classii
+ \or
+ \@classiii
+ \or
+ \@classiv
+ \or
+ \@classv
+ \fi
+ \@lastchclass\@chclass
+ }%
+ \ifcase\@lastchclass
+ \@acolr % right-hand column
+ \or
+ \or
+ \@preamerr\@ne
+ \or
+ \@preamerr\tw@
+ \or
+ \or
+ \@acolr % right-hand column
+ \fi
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\insert@column}
+% \begin{macrocode}
+\def\insert@column@array{%
+ \the@toks \the \@tempcnta
+ \ignorespaces \@sharp \unskip
+ \the@toks \the \count@ \relax
+}%
+\def\insert@column@array@new{%
+ \the@toks\the\@tempcnta
+ \array@row@rst\cell@font
+ \ignorespaces\@sharp\unskip
+ \the@toks\the\count@
+ \relax
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@mkpream@relax}
+% The procedure \cmd\@mkpream@relax\ participates in a strange and wonderful
+% method of binding the alignment procedure---but only certain parts thereof.
+%
+% Here is how it works: in \LaTeX, the \classname{array} package, and in the
+% \classname{longtable} package alike, there is a need to create an alignment
+% preamble (using \cmd\@mkpream) for use by the upcoming \cmd\halign.
+% Then, in both \classname{array} and \classname{longtable}, \TeX's \cmd\edef\
+% is used to `compile in place' that alignment preamble.
+%
+% In the case of \classname{array},
+% the operation is done in order to pre-expand the use of \texttt{*};
+% in \classname{longtable}, it is to set the widths of the columns.
+%
+% Now, during this \cmd\edef, certain control sequence names must \emph{not}
+% be expanded, and those are robustified by \cmd\@mkpream@relax.
+%
+% \begin{macrocode}
+\def\@mkpream@relax{%
+ \let\tableftsep \relax
+ \let\tabmidsep \relax
+ \let\tabrightsep \relax
+ \let\array@row@rst\relax
+ \let\cell@font \relax
+ \let\@startpbox \relax
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@mkpream}
+% We insert \cmd\@mkpream@relax\ at the head of the procedure.
+% The robustifying of \cmd\@startpbox\ and \cmd\@endpbox\ is taken over by this mechanism.
+% We also invoke \cmd\@acolr\ instead of \cmd\@acol\ when a right-hand column is at hand.
+%
+% \changes{4.1b}{2008/08/12}{(AO, 505) try to accommodate \classname{colortbl}.}
+% Note on \classname{colortbl}: this package head-patches \cmd\@mkpream\ to robustify
+% a number of its commands during the construction of the alignment preamble.
+% The best we can do is to supplement the \cmd\@mkpream@relax\ procedure to perform this action.
+% \begin{macrocode}
+\def\@mkpream@array#1{%
+ \gdef\@preamble{}\@lastchclass 4 \@firstamptrue
+ \let\@sharp\relax \let\@startpbox\relax \let\@endpbox\relax
+ \@temptokena{#1}\@tempswatrue
+ \@whilesw\if@tempswa\fi{\@tempswafalse\the\NC@list}%
+ \count@\m@ne
+ \let\the@toks\relax
+ \prepnext@tok
+ \expandafter \@tfor \expandafter \@nextchar
+ \expandafter :\expandafter =\the\@temptokena \do
+ {\@testpach
+ \ifcase \@chclass \@classz \or \@classi \or \@classii
+ \or \save@decl \or \or \@classv \or \@classvi
+ \or \@classvii \or \@classviii
+ \or \@classx
+ \or \@classx \fi
+ \@lastchclass\@chclass}%
+ \ifcase\@lastchclass
+ \@acol \or
+ \or
+ \@acol \or
+ \@preamerr \thr@@ \or
+ \@preamerr \tw@ \@addtopreamble\@sharp \or
+ \or
+ \else \@preamerr \@ne \fi
+ \def\the@toks{\the\toks}%
+}%
+\def\@mkpream@array@new#1{%
+ \gdef\@preamble{}%
+ \@lastchclass\f@ur
+ \@firstamptrue
+ \let\@sharp\relax
+ \@mkpream@relax
+%\let\@startpbox\relax\let\@endpbox\relax
+ \@temptokena{#1}\@tempswatrue
+ \@whilesw\if@tempswa\fi{\@tempswafalse\the\NC@list}%
+ \count@\m@ne
+ \let\the@toks\relax
+ \prepnext@tok
+ \expandafter\@tfor\expandafter\@nextchar\expandafter:\expandafter=\the\@temptokena
+ \do{%
+ \@testpach
+ \ifcase\@chclass
+ \@classz
+ \or
+ \@classi
+ \or
+ \@classii
+ \or
+ \save@decl
+ \or
+ \or
+ \@classv
+ \or
+ \@classvi
+ \or
+ \@classvii
+ \or
+ \@classviii
+ \or
+ \@classx
+ \or
+ \@classx
+ \fi
+ \@lastchclass\@chclass
+ }%
+ \ifcase\@lastchclass
+ \@acolr % right-hand column
+ \or
+ \or
+ \@acolr % right-hand column
+ \or
+ \@preamerr\thr@@
+ \or
+ \@preamerr\tw@\@addtopreamble\@sharp
+ \or
+ \or
+ \else
+ \@preamerr\@ne
+ \fi
+ \def\the@toks{\the\toks}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@mkpream@relax}
+% \changes{4.1b}{2008/08/12}{(AO, 505) try to accommodate \classname{colortbl}.}
+% David P. Carlisle's \classname{colortbl} package headpatches \cmd\@mkpream\ in place
+% during package loading, so it does not know whom it is working on.
+% Let us try to accomodate this package by doing what it would have liked to have done.
+%
+% Note: it would be far better to break out this mechanism in the \classname{array} package.
+% \begin{macrocode}
+\appdef\@mkpream@relax{%
+ \let\CT@setup \relax
+ \let\CT@color \relax
+ \let\CT@do@color \relax
+ \let\color \relax
+ \let\CT@column@color\relax
+ \let\CT@row@color \relax
+ \let\CT@cell@color \relax
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@addamp}
+% \begin{macrocode}
+\def\@addamp@LaTeX{%
+ \if@firstamp\@firstampfalse\else\edef\@preamble{\@preamble &}\fi
+}%
+\def\@addamp@ltx{%
+ \if@firstamp\@firstampfalse\else\@addtopreamble{&}\fi
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@arrayacol}
+% \begin{macrocode}
+\def\@arrayacol@LaTeX{%
+ \edef\@preamble{\@preamble \hskip \arraycolsep}%
+}%
+\def\@arrayacol@ltx{%
+ \@addtopreamble{\hskip\arraycolsep}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@tabacol}
+% \begin{macrocode}
+\def\@tabacoll{%
+ \@addtopreamble{\hskip\tableftsep\relax}%
+}%
+\def\@tabacol@LaTeX{%
+ \edef\@preamble{\@preamble \hskip \tabcolsep}%
+}%
+\def\@tabacol@ltx{%
+ \@addtopreamble{\hskip\tabmidsep\relax}%
+}%
+\def\@tabacolr{%
+ \@addtopreamble{\hskip\tabrightsep\relax}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@arrayclassz}
+% \begin{macrocode}
+\def\@arrayclassz@LaTeX{%
+ \ifcase \@lastchclass \@acolampacol \or \@ampacol \or
+ \or \or \@addamp \or
+ \@acolampacol \or \@firstampfalse \@acol \fi
+ \edef\@preamble{\@preamble
+ \ifcase \@chnum
+ \hfil$\relax\@sharp$\hfil \or $\relax\@sharp$\hfil
+ \or \hfil$\relax\@sharp$\fi}%
+}%
+\def\@arrayclassz@ltx{%
+ \ifcase\@lastchclass
+ \@acolampacol
+ \or
+ \@ampacol
+ \or
+ \or
+ \or
+ \@addamp
+ \or
+ \@acolampacol
+ \or
+ \@firstampfalse\@acoll
+ \fi
+ \ifcase\@chnum
+ \@addtopreamble{%
+ \hfil\array@row@rst$\relax\@sharp$\hfil
+ }%
+ \or
+ \@addtopreamble{%
+ \array@row@rst$\relax\@sharp$\hfil
+ }%
+ \or
+ \@addtopreamble{%
+ \hfil\array@row@rst$\relax\@sharp$%
+ }%
+ \fi
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@tabclassz}
+% \begin{macrocode}
+\def\@tabclassz@LaTeX{%
+ \ifcase\@lastchclass
+ \@acolampacol
+ \or
+ \@ampacol
+ \or
+ \or
+ \or
+ \@addamp
+ \or
+ \@acolampacol
+ \or
+ \@firstampfalse\@acol
+ \fi
+ \edef\@preamble{%
+ \@preamble{%
+ \ifcase\@chnum
+ \hfil\ignorespaces\@sharp\unskip\hfil
+ \or
+ \hskip1sp\ignorespaces\@sharp\unskip\hfil
+ \or
+ \hfil\hskip1sp\ignorespaces\@sharp\unskip
+ \fi}}%
+}%
+\def\@tabclassz@ltx{%
+ \ifcase\@lastchclass
+ \@acolampacol
+ \or
+ \@ampacol
+ \or
+ \or
+ \or
+ \@addamp
+ \or
+ \@acolampacol
+ \or
+ \@firstampfalse\@acoll
+ \fi
+ \ifcase\@chnum
+ \@addtopreamble{%
+ {\hfil\array@row@rst\cell@font\ignorespaces\@sharp\unskip\hfil}%
+ }%
+ \or
+ \@addtopreamble{%
+ {\cell@fil\hskip1sp\array@row@rst\cell@font\ignorespaces\@sharp\unskip\hfil}%
+ }%
+ \or
+ \@addtopreamble{%
+ {\hfil\hskip1sp\array@row@rst\cell@font\ignorespaces\@sharp\unskip\cell@fil}%
+ }%
+ \fi
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@tabclassiv}
+% \begin{macrocode}
+\def\@tabclassiv@LaTeX{%
+ \@addtopreamble\@nextchar
+}%
+\def\@tabclassiv@ltx{%
+ \expandafter\@addtopreamble\expandafter{\@nextchar}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@arrayclassiv}
+% \begin{macrocode}
+\def\@arrayclassiv@LaTeX{%
+ \@addtopreamble{$\@nextchar$}%
+}%
+\def\@arrayclassiv@ltx{%
+ \expandafter\@addtopreamble\expandafter{\expandafter$\@nextchar$}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@classv}
+% \begin{macrocode}
+\def\@classv@LaTeX{%
+ \@addtopreamble{\@startpbox{\@nextchar}\ignorespaces
+ \@sharp\@endpbox}%
+}%
+\def\@classv@ltx{%
+ \expandafter\@addtopreamble
+ \expandafter{%
+ \expandafter \@startpbox
+ \expandafter {\@nextchar}%
+ \pbox@hook\array@row@rst\cell@font\ignorespaces\@sharp\@endpbox
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@classx}
+% \begin{macrocode}
+\def\@classx@array{%
+ \ifcase \@lastchclass
+ \@acolampacol \or
+ \@addamp \@acol \or
+ \@acolampacol \or
+ \or
+ \@acol \@firstampfalse \or
+ \@addamp
+ \fi
+}%
+\def\@classx@array@new{%
+ \ifcase \@lastchclass
+ \@acolampacol
+ \or
+ \@addamp \@acol
+ \or
+ \@acolampacol
+ \or
+ \or
+ \@firstampfalse\@acoll
+ \or
+ \@addamp
+ \fi
+}%
+% \end{macrocode}
+% \end{macro}
+%
+%
+% \subsection{Repair other broken parts of \LaTeX}
+%
+% \begin{macro}{\@xbitor}
+% Expansion part has extraneous space token. Removed.
+% \begin{macrocode}
+\def\@xbitor@LaTeX #1{\@tempcntb \count#1
+ \ifnum \@tempcnta =\z@
+ \else
+ \divide\@tempcntb\@tempcnta
+ \ifodd\@tempcntb \@testtrue\fi
+ \fi}%
+\def\@xbitor@ltx#1{%
+ \@tempcntb\count#1\relax
+ \@ifnum{\@tempcnta=\z@}{}{%
+ \divide\@tempcntb\@tempcnta
+ \@ifodd\@tempcntb{\@testtrue}{}%
+ }%
+}%
+\@ifx{\@xbitor\@xbitor@LaTeX}{%
+ \class@info{Repairing broken LaTeX \string\@xbitor}%
+}{%
+ \class@info{Unrecognized LaTeX \string\@xbitor. Please update this document class! (Proceeding with fingers crossed.)}%
+}%
+\let\@xbitor\@xbitor@ltx
+% \end{macrocode}
+% \end{macro}
+%
+%
+% \subsection{Syntax}
+% \begin{macro}{\@gobble@opt@one}
+% The \cmd\@gobble@opt@one\ command eats up an optional argument
+% and one required argument.
+% \begin{macrocode}
+\newcommand*\@gobble@opt@one[2][]{}%
+% \end{macrocode}
+% \end{macro}
+%
+% \subsection{Auto-indented Contents}
+% Facility to automatically determine the proper indentation of
+% the TOC entries.
+%
+% Note on \classname{hyperref} compatibility:
+% We must respect that
+% \cmd\contentsline\ now has a fourth argument.
+% So, instead of trying to override the meaning of \cmd\contentsline,
+% we use the aux file to remember max values from one run to the next.
+%
+% In this respect, this package retains compatibility with
+% \classname{hyperref}.
+%
+% \begin{macro}{\@starttoc}
+% Install hooks at beginning and end of the TOC processing.
+% \begin{macrocode}
+\def\@starttoc#1{%
+ \begingroup
+ \toc@pre
+ \makeatletter
+ \@input{\jobname.#1}%
+ \if@filesw
+ \expandafter\newwrite\csname tf@#1\endcsname
+ \immediate\openout \csname tf@#1\endcsname \jobname.#1\relax
+ \fi
+ \@nobreakfalse
+ \toc@post
+ \endgroup
+}%
+\def\toc@pre{}%
+\def\toc@post{}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\toc@@font}
+% Interface for setting the formatting characteristics of this part
+% of the TOC.
+%
+% Note: \cmd\toc@@font\ is the common font for all auto-sizing toc commands,
+% although this, too, could become a dispatcher.
+% \changes{4.1a}{2008/01/19}{(AO, 461) Change the csname from \cs{@dotsep} to \cs{ltxu@dotsep}. The former is understood in mu. (What we wanted was a dimension.)}%
+% \begin{macrocode}
+\def\toc@@font{}%
+\def\ltxu@dotsep{\z@}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\l@section}
+% Interface for determining which TOC elements are automatically indented.
+%
+% All of the \cmd\l@\dots\ commands simply go through the
+% utility procedure \cmd\l@@sections. The calling convention is
+% to pass the name of self and the name of parent.
+% If you want to exclude any of these from the indentation
+% scheme, simply leave the \cmd\l@\dots\ command undefined.
+%
+% Note that the parent of ``section'' is nil, so we have to define a stub.
+% \begin{verbatim}\def\l@section{\l@@sections{}{section}}% Implicit #3#4\end{verbatim}
+% \begin{verbatim}\def\tocleft@{\z@}%\end{verbatim}
+% \begin{verbatim}\def\l@subsection{\l@@sections{section}{subsection}}% Implicit #3#4\end{verbatim}
+% \begin{verbatim}\def\l@subsubsection{\l@@sections{subsection}{subsubsection}}% Implicit #3#4\end{verbatim}
+% \begin{verbatim}\def\l@paragraph{\l@@sections{subsubsection}{paragraph}}% Implicit #3#4\end{verbatim}
+% \begin{verbatim}\def\l@subparagraph#1#2{\l@@sections{paragraph}{subparagraph}}% Implicit #3#4\end{verbatim}
+% \end{macro}
+%
+% Glom some \cmd\dimen\ registers.
+% \begin{macrocode}
+\let\tocdim@section \leftmargini
+\let\tocdim@subsection \leftmarginii
+\let\tocdim@subsubsection \leftmarginiii
+\let\tocdim@paragraph \leftmarginiv
+\let\tocdim@appendix \leftmarginv
+\let\tocdim@pagenum \leftmarginvi
+% \end{macrocode}
+%
+% \begin{macro}{\toc@pre@auto}
+% \begin{macro}{\toc@post@auto}
+% We patch \cmd\@starttoc\ to:
+% 1) before TOC processing,
+% initialize the max registers and
+% set the needed dimensions from
+% the values stored in the auxiliary file, and
+% 2) after TOC processing,
+% store out those max register values into the auxiliary file.
+%
+% Note that the font is set here: all other TOC entries must
+% override these font settings.
+%
+% To activate this override of the standard \LaTeX\ processing,
+% the substyle does: \cmd\let\cmd\toc@pre\cmd\toc@pre@auto\
+% and \cmd\let\cmd\toc@post\cmd\toc@post@auto.
+% \begin{macrocode}
+\def\toc@pre@auto{%
+ \toc@@font
+ \@tempdima\z@
+ \toc@setindent\@tempdima{section}%
+ \toc@setindent\@tempdima{subsection}%
+ \toc@setindent\@tempdima{subsubsection}%
+ \toc@setindent\@tempdima{paragraph}%
+ \toc@letdimen{appendix}%
+ \toc@letdimen{pagenum}%
+}%
+\def\toc@post@auto{%
+ \if@filesw
+ \begingroup
+ \toc@writedimen{section}%
+ \toc@writedimen{subsection}%
+ \toc@writedimen{subsubsection}%
+ \toc@writedimen{paragraph}%
+ \toc@writedimen{appendix}%
+ \toc@writedimen{pagenum}%
+ \endgroup
+ \fi
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\toc@setindent}
+% \begin{macrocode}
+\def\toc@setindent#1#2{%
+ \csname tocdim@#2\endcsname\tocdim@min\relax
+ \@ifundefined{tocmax@#2}{\@namedef{tocmax@#2}{\z@}}{}%
+ \advance#1\@nameuse{tocmax@#2}\relax
+ \expandafter\edef\csname tocleft@#2\endcsname{\the#1}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\toc@letdimen}
+% \begin{macrocode}
+\def\toc@letdimen#1{%
+ \csname tocdim@#1\endcsname\tocdim@min\relax
+ \@ifundefined{tocmax@#1}{\@namedef{tocmax@#1}{\z@}}{}%
+ \expandafter\let\csname tocleft@#1\expandafter\endcsname\csname tocmax@#1\endcsname
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\toc@writedimen}
+% \begin{macrocode}
+\def\toc@writedimen#1{%
+ \immediate\write\@auxout{%
+ \gdef\expandafter\string\csname tocmax@#1\endcsname{%
+ \expandafter\the\csname tocdim@#1\endcsname
+ }%
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\l@@sections}
+% The procedure for formatting the indented TOC entries.
+% We use control sequence names such as \cmd\tocmax@section\ and
+% \cmd\tocleft@section, the former being written to the auxiliary file
+% and the latter only defined for the duration of the TOC processing.
+%
+% Note that the assignment of \cmd\box\cmd\@tempboxa\ by \cmd\set@tocdim@pagenum\
+% must endure over the invocation of |#3|: it contains the
+% page number which will be set just before the \cmd\par.
+%
+% The arguments:\begin{enumerate}
+% \item[\#1] superior section
+% \item[\#2] this section
+% \item[\#3] content, including possible \cmd\numberline
+% \item[\#4] page number\end{enumerate}
+% \begin{macrocode}
+\def\l@@sections#1#2#3#4{%
+ \begingroup
+ \everypar{}%
+ \set@tocdim@pagenum\@tempboxa{#4}%
+ \global\@tempdima\csname tocdim@#2\endcsname
+ \leftskip\csname tocleft@#2\endcsname\relax
+ \dimen@\csname tocleft@#1\endcsname\relax
+ \parindent-\leftskip\advance\parindent\dimen@
+ \rightskip\tocleft@pagenum plus 1fil\relax
+ \skip@\parfillskip\parfillskip\z@
+ \let\numberline\numberline@@sections
+ \@nameuse{l@f@#2}%
+ \ignorespaces#3\unskip\nobreak\hskip\skip@
+ \hb@xt@\rightskip{\hfil\unhbox\@tempboxa}\hskip-\rightskip\hskip\z@skip
+% \end{macrocode}
+% \changes{4.1n}{2009/12/13}{(AO, 574) protect against \classname{lineno.sty}, which forces a visit to the output routine, which appears to destroy the value of \cs{@tempdima}}%
+%
+% By side effect, set the value of, e.g., \cmd\tocdim@section.
+%
+% Note that the \cmd\par\ must not be executed before the value of \cmd\@tempdima\ is expanded (outside the current group).
+% Otherwise, the \classname{lineno.sty} package may interfere (it unfortunately does a global assignment of \cmd\@tempdima).
+% \begin{macrocode}
+ \expandafter\par
+ \expandafter\aftergroup\csname tocdim@#2%
+ \expandafter\endcsname
+ \expandafter\endgroup
+ \the\@tempdima\relax
+}%
+% \end{macrocode}
+% \changes{4.1a}{2008/01/19}{(AO, 479) Per: Dylan Thurston<dpt at math.harvard.edu>}%
+% In the call to \cmd\set@tocdim@pagenum, I am now exposing the use of the particular box register.
+% \begin{macrocode}
+\def\set@tocdim@pagenum#1#2{%
+ \setbox#1\hbox{\ignorespaces#2}%
+ \@ifdim{\tocdim@pagenum<\wd#1}{\global\tocdim@pagenum\wd#1}{}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\numberline@@sections}
+% \changes{4.1a}{2008/01/19}{(AO, 461) Change the csname from \cs{@dotsep} to \cs{ltxu@dotsep}. The former is understood in mu. (What we wanted was a dimension.)}%
+% The utility procedure for all \cmd\numberline\ processing in indented TOC entries.
+% The first argument is self.
+%
+% We use \cmd\@tempdima\ to pass a value around (via global assignment) because
+% \cmd\numberline\ executes inside a group if the
+% \classname{hyperref} package is loaded.
+% Would that it were not so!
+% \begin{macrocode}
+\def\numberline@@sections#1{%
+ \leavevmode\hb@xt@-\parindent{%
+ \hfil
+ \@if@empty{#1}{}{%
+ \setbox\z@\hbox{#1.\kern\ltxu@dotsep}%
+ \@ifdim{\@tempdima<\wd\z@}{\global\@tempdima\wd\z@}{}%
+ \unhbox\z@
+ }%
+ }%
+ \ignorespaces
+}%
+\def\tocdim@min{\z@}%
+% \end{macrocode}
+% \end{macro}
+%
+%
+% \subsection{Lists}
+% \begin{macro}{\list}
+% Using \cmd\parshape\ to implement lists was always suspect
+% (can you get behind \cmd\parshape\cmd\@ne?) and we now see that
+% it was a mistake all along. Why? Because \cmd\parshape, like
+% \cmd\hangindent, achieves its effect via ``shifting'' the \cmd\hbox es
+% in a paragraph
+% instead of using \cmd\leftskip\ and \cmd\parindent, which is
+% robust during column balancing.
+%
+% We introduce the alternative method with a hook into
+% the \LaTeX\ kernel procedure \cmd\list, which is
+% the implementation of all lists.
+%
+% \begin{macrocode}
+\def\list#1#2{%
+ \ifnum \@listdepth >5\relax
+ \@toodeep
+ \else
+ \global\advance\@listdepth\@ne
+ \fi
+ \rightmargin\z@
+ \listparindent\z@
+ \itemindent\z@
+ \csname @list\romannumeral\the\@listdepth\endcsname
+ \def\@itemlabel{#1}%
+ \let\makelabel\@mklab
+ \@nmbrlistfalse
+ #2\relax
+ \@trivlist
+ \parskip\parsep
+ \set@listindent
+ \ignorespaces
+}%
+\def\set@listindent@parshape{%
+ \parindent\listparindent
+ \advance\@totalleftmargin\leftmargin
+ \advance\linewidth-\rightmargin
+ \advance\linewidth-\leftmargin
+ \parshape\@ne\@totalleftmargin\linewidth
+}%
+\def\set@listindent@{%
+ \parindent\listparindent
+ \advance\@totalleftmargin\leftmargin
+ \advance\rightskip\rightmargin
+ \advance\leftskip\@totalleftmargin
+}%
+\let\set@listindent\set@listindent@parshape
+% \end{macrocode}
+% \end{macro}
+%
+%
+% \subsection{Hypertext capabilities}
+%
+% \begin{macro}{\href}
+% \begin{macro}{\url}
+% \begin{macro}{\URL@prefix}
+% \begin{macro}{\doi}
+% \begin{macro}{\doibase}
+% \changes{4.1b}{2008/08/12}{(AO, 487) Support for video figures and the \cs{setfloatlink} command}%
+% We provide support for the \cmd\href, \cmd\url, and \cmd\doi\ commands.
+% Packages, like \classname{hyperref}, may override these definitions
+% and provide better semantics.
+% \changes{4.1g}{2009/10/06}{(AO, 532) Both arguments of \cs{href} get sanitized}%
+% \changes{4.1j}{2009/10/24}{(AO, 545) Provide definition for \cs{doi} that does hypertext}%
+% \begin{macrocode}
+\providecommand\href[0]{\begingroup\@sanitize@url\@href}%
+\def\@href#1{\@@startlink{#1}\endgroup\@@href}%
+\def\@@href#1{#1\@@endlink}%
+\providecommand \url [0]{\begingroup\@sanitize@url \@url }%
+\def \@url #1{\endgroup\@href {#1}{\URL@prefix#1}}%
+\providecommand \URL@prefix [0]{URL }%
+\providecommand\doi[0]{\begingroup\@sanitize@url\@doi}%
+\def\@doi#1{\endgroup\@@startlink{\doibase#1}doi:\discretionary {}{}{}#1\@@endlink }%
+%changes{4.2a}{2017/11/21}{(MD) Use updated best practice to use https and doi.org}%
+\providecommand \doibase [0]{https://doi.org/}%
+\providecommand \@sanitize@url[0]{\chardef\cat@space\the\catcode`\ \@sanitize\catcode`\ \cat@space}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\@@startlink}
+% \begin{macro}{\@@endlink}
+% \begin{macro}{\pdfstartlink@attr}
+% \begin{macro}{\hypertext@enable@ltx}
+% How we define \cmd\@@startlink\ and \cmd\@@endlink\ will depend on
+% whether we are running under \textsc{pdflatex}.
+% If so, and if PDF output is requested, then we use its primitives
+% to implement hypertext,
+% breaking out the link attributes in \cmd\pdfstartlink@attr\
+% and using the \classname{hyperref} defaults;
+% \cmd\pdfstartlink@attr\ can be redefined by a client package.
+% Otherwise we fall back the Hyper\TeX\ standard and leave things to the DVI translator.
+% \changes{4.1j}{2009/10/25}{(AO, 545) hypertext capabilities off by default; enable with \classoption{hypertext}}
+%
+% A class or package that wishes to employ hypertext capabilities should
+% execute the \cmd\hypertext@enable@ltx\ procedure.
+% \begin{macrocode}
+\def\@@startlink#1{}%
+\def\@@endlink{}%
+\@ifxundefined \pdfoutput {\true@sw}{\@ifnum{\z@=\pdfoutput}{\true@sw}{\false@sw}}%
+{%
+ \def\@@startlink@hypertext#1{\leavevmode\special{html:<a href="#1">}}%
+ \def\@@endlink@hypertext{\special{html:</a>}}%
+}{%
+ \def\@@startlink@hypertext#1{%
+ \leavevmode
+ \pdfstartlink\pdfstartlink@attr
+ user{/Subtype/Link/A<</Type/Action/S/URI/URI(#1)>>}%
+ \relax
+ }%
+ \def\@@endlink@hypertext{\pdfendlink}%
+ \def\pdfstartlink@attr{attr{/Border[0 0 1 ]/H/I/C[0 1 1]}}%
+}%
+\def\hypertext@enable@ltx{%
+ \let\@@startlink\@@startlink@hypertext
+ \let\@@endlink\@@endlink@hypertext
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\href}
+% \changes{4.1p}{2010/02/24}{(AO, 582) A patch of \classname{hyperref.sty} to provide backward compatibility to \TeX Live 2007's version 6.75r}%
+% The \cmd\href\ command of \classname{hyperref} was extend somewhere
+% between versions 6.75r and 6.80e. We apply a repair to the earlier
+% version (if present) so that it works like the later version.
+%
+% The issue is the presence of whitespace, either following the \cmd\href\ token
+% or following the first argument's closing brace character.
+%
+% \begin{macrocode}
+\def\href@Hy{\hyper@normalise \href@ }%
+\def\href@Hy@ltx{\@ifnextchar\bgroup\Hy@href{\hyper@normalise\href@}}%
+\def\Hy@href#{\hyper@normalise\href@}%
+\begingroup
+ \endlinechar=-1 %
+ \catcode`\^^A=14 %
+ \catcode`\^^M\active
+ \catcode`\%\active
+ \catcode`\#\active
+ \catcode`\_\active
+ \catcode`\$\active
+ \catcode`\&\active
+ \gdef\hyper@normalise@ltx{^^A
+ \begingroup
+ \catcode`\^^M\active
+ \def^^M{ }^^A
+ \catcode`\%\active
+ \let%\@percentchar
+ \let\%\@percentchar
+ \catcode`\#\active
+ \def#{\hyper@hash}^^A
+ \def\#{\hyper@hash}^^A
+ \@makeother\&^^A
+ \edef&{\string&}^^A
+ \edef\&{\string&}^^A
+ \edef\textunderscore{\string_}^^A
+ \let\_\textunderscore
+ \catcode`\_\active
+ \let_\textunderscore
+ \let~\hyper@tilde
+ \let\~\hyper@tilde
+ \let\textasciitilde\hyper@tilde
+ \let\\\@backslashchar
+ \edef${\string$}^^A
+ \Hy@safe@activestrue
+ \hyper@n@rmalise
+ }^^A
+ \catcode`\#=6 ^^A
+ \gdef\Hy@ActiveCarriageReturn@ltx{^^M}^^A
+ \gdef\hyper@n@rmalise@ltx#1#2{^^A
+ \def\Hy@tempa{#2}^^A
+ \ifx\Hy@tempa\Hy@ActiveCarriageReturn
+ \Hy@ReturnAfterElseFi{^^A
+ \hyper@@normalise{#1}^^A
+ }^^A
+ \else
+ \Hy@ReturnAfterFi{^^A
+ \hyper@@normalise{#1}{#2}^^A
+ }^^A
+ \fi
+ }^^A
+ \gdef\hyper@@normalise@ltx#1#2{^^A
+ \edef\Hy@tempa{^^A
+ \endgroup
+ \noexpand#1{\Hy@RemovePercentCr#2%^^M\@nil}^^A
+ }^^A
+ \Hy@tempa
+ }^^A
+ \gdef\Hy@RemovePercentCr@ltx#1%^^M#2\@nil{^^A
+ #1^^A
+ \ifx\limits#2\limits
+ \else
+ \Hy@ReturnAfterFi{^^A
+ \Hy@RemovePercentCr #2\@nil
+ }^^A
+ \fi
+ }^^A
+\endgroup
+\def\switch@hyperref@href{%
+ \expandafter\@ifx\expandafter{\csname href \endcsname\href@Hy}{
+ \class@info{Repairing hyperref 6.75r \string\href}%
+ \let\hyper@normalise\hyper@normalise@ltx
+ \let\hyper@@normalise\hyper@@normalise@ltx
+ \let\hyper@n@rmalise\hyper@n@rmalise@ltx
+ \let\Hy@ActiveCarriageReturn\Hy@ActiveCarriageReturn@ltx
+ \let\Hy@RemovePercentCr\Hy@RemovePercentCr@ltx
+ \let\href\href@Hy@ltx
+ }{}%
+}%
+\appdef\document@inithook{\switch@hyperref@href}%
+% \end{macrocode}
+% \end{macro}
+%
+%
+% \begin{macro}{\typeout}
+% We make the \cmd\typeout\ procedure of \LaTeX\ be \cmd\long,
+% because sometimes we are talking about \cmd\par.
+% \begin{macrocode}
+\def\typeout@org#1{%
+ \begingroup
+ \set@display@protect
+ \immediate\write\@unused{#1}%
+ \endgroup
+}%
+\long\def\typeout@ltx#1{%
+ \begingroup
+ \set@display@protect
+ \immediate\write\@unused{#1}%
+ \endgroup
+}%
+\@ifx{\typeout\typeout@org}{%
+ \let\typeout\typeout@ltx
+ \true@sw
+}{%
+ \rvtx@ifformat@geq{2020-10-01}%
+ {\true@sw}{\false@sw}%
+}%
+ {\class@info{Making \string\typeout\space \string\long}}%
+ {}%
+% \end{macrocode}
+% \end{macro}
+%
+% \subsection{End of the \file{kernel} {\sc docstrip} module}
+% Here ends the module.
+% \begin{macrocode}
+%</kernel>
+% \end{macrocode}
+%
+% \Finale
+% \iffalse Here ends the programmer's documentation.\fi
+% \endinput
+%
+\endinput
+%%EOF
diff --git a/texmf/source/latex/revtex/revtex-bst.dtx b/texmf/source/latex/revtex/revtex-bst.dtx
new file mode 100644
index 0000000..1b5c843
--- /dev/null
+++ b/texmf/source/latex/revtex/revtex-bst.dtx
@@ -0,0 +1,3873 @@
+% \iffalse meta-comment balanced on line 122
+% revtex-bst.dtx: Package to extract .bst files for RevTeX
+% Copyright (c) 2020 American Physical Society.
+% https://journals.aps.org/revtex/
+% mailto:revtex@aps.org
+%
+% Disclaimer
+% This file is distributed WITHOUT ANY WARRANTY;
+% without even the implied warranty of
+% MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
+% License
+% You may distribute this file under the conditions of the
+% LaTeX Project Public License 1.3c or later
+% (http://www.latex-project.org/lppl.txt).
+% ReadMe
+% For the documentation and more detailed instructions for
+% installation, typeset this document with \LaTeX.
+% Maintenance Status
+% This work has the LPPL maintenance status "maintained";
+% Current Maintainer of this work is Phelype Oleinik.
+%
+% This work consists of the main source file revtex-bst.dtx
+% and the derived files
+% aapmrev4-2.bst, aipauth4-2.bst, aipnum4-2.bst, apsrev4-2.bst, apsrmp4-2.bst
+% Distribution:
+% CTAN:macros/latex/contrib/revtex/
+%
+% Unpacking:
+% tex revtex-bst.dtx
+%
+% Installation:
+% TDS:bibtex/bst/revtex/
+%
+%<*ignore>
+\begingroup
+ \def\x{LaTeX2e}%
+\expandafter\endgroup
+\ifcase
+ 0\expandafter\ifx\csname processbatchFile\endcsname\relax\else1\fi\ifx\fmtname\x\else 1\fi
+ \relax
+\else
+ \csname fi\endcsname
+%</ignore>
+%<*install>
+%% This file will generate documentation and runtime files
+%% from revtex4-2.dtx when run through TeX.
+\input docstrip
+\nopreamble
+\edef\defaultpostamble{%
+ \MetaPrefix\space End of customized bst file^^J%
+ \MetaPrefix ^^J%
+ \MetaPrefix\space End of file `\outFileName'.%
+}
+\askforoverwritefalse
+\keepsilent
+\generate{%
+ \file{aapmrev4-2.bst}{\from{revtex-bst.dtx}{aapmrev}}%
+ \file{aipauth4-2.bst}{\from{revtex-bst.dtx}{aipauth}}%
+ \file{aipnum4-2.bst}{\from{revtex-bst.dtx}{aipnum}}%
+ \file{apsrev4-2.bst}{\from{revtex-bst.dtx}{apsrev}}%
+ \file{apsrmp4-2.bst}{\from{revtex-bst.dtx}{apsrmp}}%
+}%
+\ifToplevel{
+\Msg{***********************************************************}
+\Msg{*}
+\Msg{* To finish the installation, please move}
+\Msg{* *.bst }
+\Msg{* into a directory searched by TeX;}
+\Msg{* in a TDS-compliant installation:}
+\Msg{* texmf/bibtex/bst/revtex/.}
+\Msg{*}
+\Msg{* Happy BibTeXing}
+\Msg{***********************************************************}
+}
+\endbatchfile
+%</install>
+%<*ignore>
+\fi
+%</ignore>
+% \fi
+%
+% \GetFileInfo{revtex-bst.dtx}
+%
+% \StopEventually{}
+% \iffalse ltxdoc klootch
+%<*package>
+%%% @LaTeX-file{
+%%% filename = "revtex-bst.dtx",
+%%% version = "4.2e",
+%%% date = "2020/10/03",
+%%% author = "Phelype Oleinik (mailto: phelype.oleinik at latex-project.org),
+%%% commissioned by the American Physical Society.
+%%% ",
+%%% copyright = "Copyright (C) 2020 American Physical Society
+%%% distributed under the terms of the
+%%% LaTeX Project Public License 1.3c, see
+%%% ftp://ctan.tug.org/macros/latex/base/lppl.txt
+%%% ",
+%%% email = "mailto colon revtex at aps.org",
+%%% codetable = "ISO/ASCII",
+%%% keywords = "",
+%%% supported = "yes",
+%%% abstract = "",
+%%% }
+%</package>
+% \fi
+%
+% \subsection{Bill of Materials}
+%
+% Following is a list of the files in this distribution arranged
+% according to provenance.
+%
+% \subsubsection{Primary Source}%
+% One single file generates all.
+%\begin{verbatim}
+%revtex-bst.dtx
+%\end{verbatim}
+%
+% \subsubsection{Generated by \texttt{tex revtex-bst.dtx}}%
+% Typesetting ths file under \TeX\ itself runs the installer,
+% which generates the bibliography style files.
+%\begin{verbatim}
+%aapmrev4-2.bst, aipauth4-2.bst, aipnum4-2.bst,
+%apsrev4-2.bst, apsrmp4-2.bst
+%\end{verbatim}
+%
+% \section{Code common to all modules}%
+%
+% \begin{macrocode}
+%<*aapmrev|aipauth|aipnum|apsrev|apsrmp>
+%%
+%<aapmrev>%% This is file `aapmrev4-2.bst', and is a hand-edited version of aapmrev4-1.bst
+%<aipauth>%% This is file `aipauth4-2.bst', and is a hand-edited version of aipauth4-1.bst
+%<aipnum>%% This is file `aipnum4-2.bst', and is a hand-edited version of aipnum4-1.bst
+%<apsrev>%% This is file `apsrev4-2.bst', and is a hand-edited version of apsrev4-1.bst
+%<apsrmp>%% This is file `apsrmp4-2.bst', and is a hand-edited version of apsrmp4-1.bst
+ % ===============================================================
+ % IMPORTANT NOTICE:
+ % This file can be redistributed and/or modified under the terms
+ % of the LaTeX Project Public License Distributed from CTAN
+ % archives in directory macros/latex/base/lppl.txt; either
+ % version 1 of the License, or any later version.
+ % ===============================================================
+ %-------------------------------------------------------------------
+ % This bibliography style file is intended for texts in ENGLISH
+ % This is an author-year citation style bibliography. As such, it is
+ % non-standard LaTeX, and requires a special package file to function properly.
+ % Such a package is natbib.sty by Patrick W. Daly
+ % The form of the \bibitem entries is
+ % \bibitem[Jones et al.(1990)]{key}...
+ % \bibitem[Jones et al.(1990)Jones, Baker, and Smith]{key}...
+ % The essential feature is that the label (the part in brackets) consists
+ % of the author names, as they should appear in the citation, with the year
+ % in parentheses following. There must be no space before the opening
+ % parenthesis!
+ % With natbib v5.3, a full list of authors may also follow the year.
+ % In natbib.sty, it is possible to define the type of enclosures that is
+ % really wanted (brackets or parentheses), but in either case, there must
+ % be parentheses in the label.
+ % The \cite command functions as follows:
+ % \citet{key} ==>> Jones et al. (1990)
+ % \citet*{key} ==>> Jones, Baker, and Smith (1990)
+ % \citep{key} ==>> (Jones et al., 1990)
+ % \citep*{key} ==>> (Jones, Baker, and Smith, 1990)
+ % \citep[chap. 2]{key} ==>> (Jones et al., 1990, chap. 2)
+ % \citep[e.g.][]{key} ==>> (e.g. Jones et al., 1990)
+ % \citep[e.g.][p. 32]{key} ==>> (e.g. Jones et al., p. 32)
+ % \citeauthor{key} ==>> Jones et al.
+ % \citeauthor*{key} ==>> Jones, Baker, and Smith
+ % \citeyear{key} ==>> 1990
+ %---------------------------------------------------------------------
+%</aapmrev|aipauth|aipnum|apsrev|apsrmp>
+%
+% \subsection{The Frontmatter File}
+% As promised above, here is the contents of the frontmatter file.
+% This file doesn't have any documentation, so just stop early.
+% \begin{macrocode}
+%<*ignore>
+\stop
+\changes{4.2e}{2020/10/03}{(PHO) Initial version---Put all .bst files in a single .dtx}
+%</ignore>
+% \end{macrocode}
+%
+% \begin{macrocode}
+%<*aapmrev|aipauth|aipnum|apsrev|apsrmp>
+% \end{macrocode}
+%
+% \begin{macrocode}
+
+%<aapmrev>FUNCTION {id.bst} {"aapmrev4-2.bst 2019-01-14 (MD) hand-edited version of aapmrev4-1.bst"}
+%<aipauth>FUNCTION {id.bst} {"aipnauth4-2.bst 2018-12-27 (MD) hand-edited version of apsauth4-1.bst"}
+%<aipnum>FUNCTION {id.bst} {"aipnum4-2.bst 2019-01-14 (MD) hand-edited version of apsrev4-1.bst"}
+%<apsrev>FUNCTION {id.bst} {"apsrev4-2.bst 2019-01-14 (MD) hand-edited version of apsrev4-1.bst"}
+%<apsrmp>FUNCTION {id.bst} {"apsrmp4-2.bst 2018-12-27 (MD) hand-edited version of apsrmp4-1.bst"}
+ENTRY
+{
+ address
+ archive
+ archivePrefix
+ author
+ bookaddress
+ booktitle
+ chapter
+ collaboration
+ doi
+ edition
+ editor
+ eid
+ eprint
+ howpublished
+ institution
+ isbn
+ issn
+ journal
+ key
+ language
+ month
+ note
+ number
+ organization
+ pages
+ primaryClass
+ publisher
+ school
+ SLACcitation
+ series
+ title
+ translation
+ type
+ url
+ volume
+ year
+}{
+}{
+ label
+ extra.label sort.label
+ short.list
+}
+
+INTEGERS
+{
+ output.state before.all
+ after.word after.punctuation
+ after.sentence after.block
+}
+
+INTEGERS
+{
+ punctuation.state punctuation.no punctuation.space punctuation.yes
+}
+
+STRINGS { bibfield output.bibfield }
+
+FUNCTION {not}
+{ { #0 }
+ { #1 }
+ if$
+}
+
+FUNCTION {and}
+{ 'skip$
+ { pop$ #0 }
+ if$
+}
+
+FUNCTION {or}
+{ { pop$ #1 }
+ 'skip$
+ if$
+}
+
+FUNCTION {non.stop}
+{ duplicate$
+ "}" * add.period$
+ #-1 #1 substring$ "." =
+}
+
+INTEGERS { arith.mulitplier arith.multiplicand }
+
+FUNCTION {multiply}
+{
+ 'arith.multiplicand :=
+ 'arith.mulitplier :=
+ #0
+ { arith.mulitplier #0 > }
+ { arith.multiplicand +
+ arith.mulitplier #1 - 'arith.mulitplier :=
+ }
+ while$
+}
+
+FUNCTION {chr.to.hex}
+{
+ chr.to.int$
+ duplicate$ "0" chr.to.int$ -
+ duplicate$ duplicate$ #0 < swap$ #9 > or not
+ { swap$ pop$ }
+ { pop$
+ duplicate$ "A" chr.to.int$ -
+ duplicate$ duplicate$ #0 < swap$ #5 > or not
+ { swap$ pop$ #10 + }
+ { pop$
+ duplicate$ "a" chr.to.int$ -
+ duplicate$ duplicate$ #0 < swap$ #5 > or not
+ { swap$ pop$ #10 + }
+ { pop$
+ pop$ #-1
+ }
+ if$
+ }
+ if$
+ }
+ if$
+}
+
+INTEGERS { arith.accumulator }
+
+FUNCTION {str.to.hex}
+{ #0 'arith.accumulator :=
+ { duplicate$ empty$ not }
+ { duplicate$ #1 #1 substring$ chr.to.hex
+ duplicate$ #0 <
+ { pop$ pop$ ""
+ }
+ { arith.accumulator #16 multiply + 'arith.accumulator :=
+ #2 global.max$ substring$
+ }
+ if$
+ }
+ while$
+ pop$ arith.accumulator
+}
+
+FUNCTION {diagn.cmntlog}
+{
+ duplicate$ top$ "%" swap$ * write$ newline$
+}
+
+INTEGERS { control.key control.author control.editor control.title control.pages control.eprint control.year }
+
+INTEGERS { control.author.jnrlst control.author.dotless control.author.nospace control.author.initials control.author.nocomma control.author.first control.author.reversed }
+
+FUNCTION { control.init }
+{
+ #0
+ 'control.key :=
+ #0
+%<aipauth> #1 +
+%<apsrmp> #1 +
+%<apsrmp> #2 +
+ #8 +
+%<apsrev> #64 +
+%<apsrmp> #64 +
+ 'control.author :=
+%<aapmrev> #0
+%<aipauth> #0
+%<aipnum> #0
+%<apsrev> #1
+%<apsrmp> #1
+ 'control.author.jnrlst :=
+ #0
+ 'control.author.dotless :=
+ #0
+ 'control.author.nospace :=
+ #1
+ 'control.author.initials :=
+ #0
+ 'control.author.nocomma :=
+%<!apsrmp> #0
+%<apsrmp> #1
+ 'control.author.first :=
+%<aapmrev> #0
+%<aipauth> #1
+%<aipnum> #0
+%<apsrev> #0
+%<apsrmp> #1
+ 'control.author.reversed :=
+%<aapmrev> #1
+%<aipauth> #0
+%<aipnum> #1
+%<apsrev> #1
+%<apsrmp> #0
+ 'control.editor :=
+ #-1
+ 'control.title :=
+ #0
+ 'control.pages :=
+ #0
+ 'control.eprint :=
+ #1
+ 'control.year :=
+}
+
+FUNCTION {warning.dependency}
+{
+ " (dependency: " * swap$ * ") set " * swap$ int.to.str$ * warning$
+}
+
+FUNCTION {control.check}
+{
+ control.editor
+ {
+ "editor formatted same as author"
+ control.author.reversed {
+ duplicate$ #0 swap$ "reversed" warning.dependency
+ #0 'control.author.reversed :=
+ } 'skip$ if$
+ control.author.first {
+ duplicate$ #0 swap$ "first" warning.dependency
+ #0 'control.author.first :=
+ } 'skip$ if$
+ control.author.nocomma {
+ duplicate$ #0 swap$ "nocomma" warning.dependency
+ #0 'control.author.nocomma :=
+ } 'skip$ if$
+ pop$
+ } 'skip$ if$
+ control.author.reversed 'skip$
+ {
+ "not reversed"
+ control.author.nospace {
+ duplicate$ #0 swap$ "nospace" warning.dependency
+ #0 'control.author.nospace :=
+ } 'skip$ if$
+ control.author.jnrlst 'skip$ {
+ duplicate$ #1 swap$ "jnrlst" warning.dependency
+ #1 'control.author.jnrlst :=
+ } if$
+ control.author.initials {
+ duplicate$ ", initials" *
+ control.author.dotless {
+ duplicate$ #0 swap$ "dotless" warning.dependency
+ #0 'control.author.dotless :=
+ } 'skip$ if$
+ pop$
+ } 'skip$ if$
+ pop$
+ }
+ if$
+ control.author.initials 'skip$ {
+ "not initials"
+ control.author.nocomma {
+ duplicate$ #0 swap$ "nocomma" warning.dependency
+ #0 'control.author.nocomma :=
+ } 'skip$ if$
+ control.author.nospace {
+ duplicate$ #0 swap$ "nospace" warning.dependency
+ #0 'control.author.nospace :=
+ } 'skip$ if$
+ control.author.dotless 'skip$ {
+ duplicate$ #1 swap$ "dotless" warning.dependency
+ #1 'control.author.dotless :=
+ } if$
+ pop$
+ } if$
+}
+
+FUNCTION {control.parse}
+{
+ duplicate$ duplicate$ missing$
+ {
+ pop$ pop$ pop$
+ }
+ { empty$
+ {
+ pop$ #-1
+ }{
+ str.to.hex
+ }
+ if$
+ swap$ :=
+ }
+ if$
+}
+
+FUNCTION {control.dump}
+{
+ duplicate$ missing$ { pop$ "N/A" } 'skip$ if$
+ "{" swap$ * "}, " *
+ *
+}
+
+INTEGERS { decode.threshold }
+
+FUNCTION {control.decode}
+{
+ - duplicate$
+ #0 <
+ {
+ skip$ pop$ swap$ #0
+ }
+ {
+ swap$ pop$ swap$ #1
+ }
+ if$
+ swap$ :=
+}
+
+FUNCTION {control.author.decode}
+{
+ control.author
+ duplicate$ duplicate$ #0 < swap$ #128 < not or
+ {
+ int.to.str$ "(" swap$ * ")" *
+ "Control cannot interpret author " swap$ *
+ warning$
+ }{
+ 'control.author.jnrlst swap$ duplicate$ #64 control.decode
+ 'control.author.dotless swap$ duplicate$ #32 control.decode
+ 'control.author.nospace swap$ duplicate$ #16 control.decode
+ 'control.author.initials swap$ duplicate$ #8 control.decode
+ 'control.author.nocomma swap$ duplicate$ #4 control.decode
+ 'control.author.first swap$ duplicate$ #2 control.decode
+ 'control.author.reversed swap$ duplicate$ #1 control.decode
+ duplicate$ #0 =
+ 'skip$
+ {
+ "Control: residue of author"
+ "(" swap$ * ")" * *
+ warning$
+ }
+ if$
+ pop$
+ }
+ if$
+}
+
+FUNCTION {control.setup}
+{
+ type$ cite$ "{" swap$ * "}, " * *
+ "control.key" key control.dump *
+ "control.author" author control.dump *
+ "control.editor" editor control.dump *
+ "control.title" title control.dump *
+ "control.pages" pages control.dump *
+ "control.year" year control.dump *
+ "control.eprint" eprint control.dump *
+ top$
+ 'control.key key control.parse
+ 'control.author author control.parse
+ 'control.editor editor control.parse
+ 'control.title title control.parse
+ 'control.pages pages control.parse
+ 'control.year year control.parse
+ 'control.eprint eprint control.parse
+ control.author.decode
+}
+
+FUNCTION {control.pass}
+{ type$ "control" = 'control.setup 'skip$ if$
+}
+
+FUNCTION {control.presort}
+{
+}
+
+FUNCTION {control.forward}
+{
+}
+
+FUNCTION {control.reverse}
+{
+}
+
+FUNCTION {control.sort}
+{
+}
+
+FUNCTION {control.longest.label}
+{
+}
+
+FUNCTION {control.key.bib}
+{
+ "Control: key "
+ control.key
+ duplicate$ "(" swap$ int.to.str$ * ") " * swap$
+ pop$ *
+ diagn.cmntlog
+}
+
+FUNCTION {control.author.bib}
+{
+ "Control: author "
+ control.author "(" swap$ int.to.str$ * ")" * *
+ control.author.reversed { " reversed" * }{} if$
+ control.author.first { " first" * }{} if$
+ control.author.nocomma { " nocomma" * }{} if$
+ control.author.initials { " initials" * }{} if$
+ control.author.nospace { " nospace" * }{} if$
+ control.author.dotless { " dotless" * }{} if$
+ control.author.jnrlst { " jnrlst" * }{} if$
+ diagn.cmntlog
+}
+
+FUNCTION {control.editor.bib}
+{
+ "Control: editor formatted "
+ control.editor
+ duplicate$ "(" swap$ int.to.str$ * ") " * swap$
+ duplicate$ #0 < {
+ pop$
+ "disabled!"
+ } {
+ #0 > {
+ "identically to author"
+ } {
+ "differently from author"
+ } if$
+ } if$ * *
+ diagn.cmntlog
+}
+
+FUNCTION {control.title.bib}
+{
+ "Control: production of article title "
+ control.title
+ duplicate$ "(" swap$ int.to.str$ * ") " * swap$
+ duplicate$ #0 < {
+ pop$
+ "disabled"
+ } {
+ #0 > {
+ "required"
+ } {
+ "allowed"
+ } if$
+ } if$ * *
+ diagn.cmntlog
+}
+
+FUNCTION {control.pages.bib}
+{
+ "Control: page "
+ control.pages
+ duplicate$ "(" swap$ int.to.str$ * ") " * swap$
+ duplicate$ #0 < {
+ pop$
+ "none"
+ } {
+ #0 > {
+ "range"
+ } {
+ "single"
+ } if$
+ } if$ * *
+ diagn.cmntlog
+}
+
+FUNCTION {control.year.bib}
+{
+ "Control: year "
+ control.year
+ duplicate$ "(" swap$ int.to.str$ * ") " * swap$
+ duplicate$ #0 < {
+ pop$
+ "disabled!"
+ } {
+ #0 > {
+ "truncated"
+ } {
+ "verbatim"
+ } if$
+ } if$ * *
+ diagn.cmntlog
+}
+
+FUNCTION {control.eprint.bib}
+{
+ "Control: production of eprint "
+ control.eprint
+ duplicate$ "(" swap$ int.to.str$ * ") " * swap$
+ #0 < { "disabled" } { "enabled" } if$ * *
+ diagn.cmntlog
+}
+
+FUNCTION {control.bib}
+{
+ control.key.bib
+ control.author.bib
+ control.editor.bib
+ control.title.bib
+ control.pages.bib
+ control.year.bib
+ control.eprint.bib
+}
+
+FUNCTION {init.state.consts}
+{
+ #0 'before.all :=
+ #1 'after.word :=
+ #2 'after.punctuation :=
+ #3 'after.sentence :=
+ #4 'after.block :=
+ #0 'punctuation.no :=
+ #1 'punctuation.space :=
+ #2 'punctuation.yes :=
+ "" 'bibfield :=
+ "" 'output.bibfield :=
+}
+
+STRINGS { s t}
+FUNCTION {block.punctuation}
+{ ""
+ "," *
+}
+
+FUNCTION {word.space}
+{
+ "\ "
+}
+
+FUNCTION {show.stackstring.one}{
+ "(" *
+ output.state int.to.str$ *
+ "," * punctuation.state int.to.str$ *
+ ")" * top$
+ duplicate$ "1(" swap$ * ")" * top$
+}
+
+FUNCTION {show.stackstring.two}{
+ "(" *
+ output.state int.to.str$ *
+ "," * punctuation.state int.to.str$ *
+ ")" * top$
+ swap$
+ duplicate$ "1(" swap$ * ")" * top$
+ swap$
+ duplicate$ "2(" swap$ * ")" * top$
+}
+
+FUNCTION {bibfield.command}{ "\bibfield "}
+
+FUNCTION {output.nonnull}
+{
+ swap$
+ output.state after.word =
+ {
+ block.punctuation *
+ word.space *
+ }
+ {
+ output.state after.punctuation =
+ {
+ word.space *
+ }
+ {
+ output.state after.block = output.state after.sentence = or
+ {
+ add.period$
+ "\EOS\ " *
+ }{
+ }
+ if$
+ }
+ if$
+ }
+ if$
+ output.bibfield duplicate$ empty$ 'pop$
+ {
+ bibfield.command
+ " {" * swap$ * "} {" * swap$ * "}" *
+ }
+ if$
+ write$
+ bibfield 'output.bibfield := "" 'bibfield :=
+ output.state after.block =
+ {
+ newline$
+ "\newblock " write$
+ }
+ 'skip$
+ if$
+ punctuation.state duplicate$
+ punctuation.yes 'punctuation.state :=
+ punctuation.no =
+ { pop$ before.all }
+ { punctuation.yes = { after.word }{ after.punctuation } if$ }
+ if$
+ 'output.state :=
+}
+
+FUNCTION {output}
+{ duplicate$ empty$
+ {
+ pop$
+ "" 'bibfield :=
+ }
+ 'output.nonnull
+ if$
+}
+
+FUNCTION {output.nopunct}
+{
+ punctuation.no 'punctuation.state :=
+ output.nonnull
+}
+
+FUNCTION {output.check}
+{ swap$
+ duplicate$ empty$
+ { pop$ "empty " swap$ * " in " * cite$ * warning$ }
+ { swap$ pop$ output.nonnull }
+ if$
+}
+
+FUNCTION {bbl.open} { "\BibitemOpen " }
+
+FUNCTION {bbl.shut} { "\BibitemShut " }
+
+FUNCTION {bibitem.shut.stop} { bbl.shut "{Stop}%" * }
+
+FUNCTION {bibitem.shut.nostop} { bbl.shut "{NoStop}%" * }
+
+FUNCTION {bibitem.shut}
+{
+ non.stop
+ {
+ bibitem.shut.nostop *
+ }{
+ bibitem.shut.stop *
+ }
+ if$
+}
+
+FUNCTION {html.itag} {
+ "p"
+}
+
+FUNCTION {html.ltag} {
+ ""
+}
+
+FUNCTION {output.SLACcitation}
+{ SLACcitation empty$
+ 'skip$
+ {
+ newline$
+ SLACcitation write$
+ }
+ if$
+}
+
+FUNCTION {fin.entry}
+{
+ bibitem.shut
+ write$
+ output.SLACcitation
+}
+
+FUNCTION {new.block}
+{ output.state before.all =
+ 'skip$
+ { after.block 'output.state := }
+ if$
+}
+
+FUNCTION {new.block.comma}
+{
+}
+
+FUNCTION {new.sentence}
+{ output.state after.block = output.state before.all = or
+ 'skip$
+ { after.sentence 'output.state := }
+ if$
+}
+
+FUNCTION {new.sentence.comma}
+{
+}
+
+FUNCTION {sentence.or.colon}
+{
+ new.sentence
+}
+
+FUNCTION {add.blank}
+{
+ word.space *
+ before.all 'output.state :=
+}
+
+FUNCTION {no.blank.or.punct}
+{
+ "\hspace {0pt}" *
+ before.all 'output.state :=
+}
+
+FUNCTION {date.block}
+{
+ new.block.comma
+ skip$
+}
+
+STRINGS {z}
+FUNCTION {remove.dots}
+{
+ control.author.dotless {
+ 'z :=
+ ""
+ { z empty$ not }
+ { z #1 #1 substring$
+ z #2 global.max$ substring$ 'z :=
+ duplicate$ "." = 'pop$
+ { * }
+ if$
+ }
+ while$
+ } 'skip$ if$
+}
+
+FUNCTION {new.block.checkb}
+{ empty$
+ swap$ empty$
+ and
+ 'skip$
+ 'new.block
+ if$
+}
+
+FUNCTION {field.or.null}
+{ duplicate$ empty$
+ { pop$ "" }
+ 'skip$
+ if$
+}
+
+FUNCTION {emphasize}
+{
+ duplicate$ empty$
+ { pop$ "" }
+ {
+ "\emph {" swap$ * "}" *
+ }
+ if$
+}
+
+FUNCTION {bolden}
+{ duplicate$ empty$
+ { pop$ "" }
+ { "\textbf {" swap$ * "}" * }
+ if$
+}
+
+FUNCTION {bib.name.font}
+{
+ duplicate$ empty$
+ { pop$ "" }
+ {
+ "\bibnamefont {" swap$ * "}" *
+ }
+ if$
+}
+
+FUNCTION {bib.fname.font}
+{
+ duplicate$ empty$
+ { pop$ "" }
+ {
+ "\bibfnamefont {" swap$ * "}" *
+ }
+ if$
+}
+
+FUNCTION {cite.name.font}
+{
+ duplicate$ empty$
+ { pop$ "" }
+ {
+ "\citenamefont {" swap$ * "}" *
+ }
+ if$
+}
+
+FUNCTION {tie.or.space.prefix}
+{ duplicate$ text.length$ #3 <
+ { "~" }
+ { word.space }
+ if$
+ swap$
+}
+
+FUNCTION {capitalize}
+{
+ "u" change.case$ "t" change.case$
+}
+
+FUNCTION {space.word}
+{ word.space swap$ * word.space * }
+
+ % Here are the language-specific definitions for explicit words.
+ % Each function has a name bbl.xxx where xxx is the English word.
+ % The language selected here is ENGLISH
+
+FUNCTION {bbl.and}
+{
+ "and"
+}
+
+FUNCTION {bbl.etal}
+{
+ "et~al."
+}
+
+FUNCTION {bbl.editors}
+{
+ "eds."
+}
+
+FUNCTION {bbl.editor}
+{
+ "ed."
+}
+
+FUNCTION {bbl.edby}
+{ "edited by" }
+
+FUNCTION {bbl.edition}
+{
+ "ed."
+}
+
+FUNCTION {bbl.volume}
+{
+ "vol."
+}
+
+FUNCTION {bbl.of}
+{ "of" }
+
+FUNCTION {bbl.number}
+{
+ "no."
+}
+
+FUNCTION {bbl.nr}
+{ "no." }
+
+FUNCTION {bbl.in}
+{ "in" }
+
+FUNCTION {bbl.pages}
+{
+ "pp."
+}
+
+FUNCTION {bbl.page}
+{
+ "p."
+}
+
+FUNCTION {bbl.eidpp}
+{ "pages" }
+
+FUNCTION {bbl.chapter}
+{
+ "chap."
+}
+
+FUNCTION {bbl.techrep}
+{
+ "Tech. Rep."
+}
+
+FUNCTION {bbl.mthesis}
+{ "Master's thesis" }
+
+FUNCTION {bbl.phdthesis}
+{ "Ph.D. thesis" }
+
+FUNCTION {bbl.first}
+{
+ "1st"
+}
+
+FUNCTION {bbl.second}
+{
+ "2nd"
+}
+
+FUNCTION {bbl.third}
+{
+ "3rd"
+}
+
+FUNCTION {bbl.fourth}
+{
+ "4th"
+}
+
+FUNCTION {bbl.fifth}
+{
+ "5th"
+}
+
+FUNCTION {bbl.st}
+{ "st" }
+
+FUNCTION {bbl.nd}
+{ "nd" }
+
+FUNCTION {bbl.rd}
+{ "rd" }
+
+FUNCTION {bbl.th}
+{ "th" }
+
+MACRO {jan} {"Jan."}
+
+MACRO {feb} {"Feb."}
+
+MACRO {mar} {"Mar."}
+
+MACRO {apr} {"Apr."}
+
+MACRO {may} {"May"}
+
+MACRO {jun} {"Jun."}
+
+MACRO {jul} {"Jul."}
+
+MACRO {aug} {"Aug."}
+
+MACRO {sep} {"Sep."}
+
+MACRO {oct} {"Oct."}
+
+MACRO {nov} {"Nov."}
+
+MACRO {dec} {"Dec."}
+
+FUNCTION {bbl.url.prefix}
+{
+ "\urlprefix "
+}
+
+FUNCTION {eng.ord}
+{ duplicate$ "1" swap$ *
+ #-2 #1 substring$ "1" =
+ { bbl.th * }
+ { duplicate$ #-1 #1 substring$
+ duplicate$ "1" =
+ { pop$ bbl.st * }
+ { duplicate$ "2" =
+ { pop$ bbl.nd * }
+ { "3" =
+ { bbl.rd * }
+ { bbl.th * }
+ if$
+ }
+ if$
+ }
+ if$
+ }
+ if$
+}
+
+ %-------------------------------------------------------------------
+ % Begin module:
+ % \ProvidesFile{physjour.mbs}[2002/01/14 2.2 (PWD)]
+MACRO {aa}{"Astron. \& Astrophys."}
+MACRO {aasup}{"Astron. \& Astrophys. Suppl. Ser."}
+MACRO {aj} {"Astron. J."}
+MACRO {aph} {"Acta Phys."}
+MACRO {advp} {"Adv. Phys."}
+MACRO {ajp} {"Amer. J. Phys."}
+MACRO {ajm} {"Amer. J. Math."}
+MACRO {amsci} {"Amer. Sci."}
+MACRO {anofd} {"Ann. Fluid Dyn."}
+MACRO {am} {"Ann. Math."}
+MACRO {ap} {"Ann. Phys. (NY)"}
+MACRO {adp} {"Ann. Phys. (Leipzig)"}
+MACRO {ao} {"Appl. Opt."}
+MACRO {apl} {"Appl. Phys. Lett."}
+MACRO {app} {"Astroparticle Phys."}
+MACRO {apj} {"Astrophys. J."}
+MACRO {apjsup} {"Astrophys. J. Suppl."}
+MACRO {apss} {"Astrophys. Space Sci."}
+MACRO {araa} {"Ann. Rev. Astron. Astrophys."}
+MACRO {baas} {"Bull. Amer. Astron. Soc."}
+MACRO {baps} {"Bull. Amer. Phys. Soc."}
+MACRO {cmp} {"Comm. Math. Phys."}
+MACRO {cpam} {"Commun. Pure Appl. Math."}
+MACRO {cppcf} {"Comm. Plasma Phys. \& Controlled Fusion"}
+MACRO {cpc} {"Comp. Phys. Comm."}
+MACRO {cqg} {"Class. Quant. Grav."}
+MACRO {cra} {"C. R. Acad. Sci. A"}
+MACRO {fed} {"Fusion Eng. \& Design"}
+MACRO {ft} {"Fusion Tech."}
+MACRO {grg} {"Gen. Relativ. Gravit."}
+MACRO {ieeens} {"IEEE Trans. Nucl. Sci."}
+MACRO {ieeeps} {"IEEE Trans. Plasma Sci."}
+MACRO {ijimw} {"Interntl. J. Infrared \& Millimeter Waves"}
+MACRO {ip} {"Infrared Phys."}
+MACRO {irp} {"Infrared Phys."}
+MACRO {jap} {"J. Appl. Phys."}
+MACRO {jasa} {"J. Acoust. Soc. America"}
+MACRO {jcp} {"J. Comp. Phys."}
+MACRO {jetp} {"Sov. Phys.--JETP"}
+MACRO {jfe} {"J. Fusion Energy"}
+MACRO {jfm} {"J. Fluid Mech."}
+MACRO {jmp} {"J. Math. Phys."}
+MACRO {jne} {"J. Nucl. Energy"}
+MACRO {jnec} {"J. Nucl. Energy, C: Plasma Phys., Accelerators, Thermonucl. Res."}
+MACRO {jnm} {"J. Nucl. Mat."}
+MACRO {jpc} {"J. Phys. Chem."}
+MACRO {jpp} {"J. Plasma Phys."}
+MACRO {jpsj} {"J. Phys. Soc. Japan"}
+MACRO {jsi} {"J. Sci. Instrum."}
+MACRO {jvst} {"J. Vac. Sci. \& Tech."}
+MACRO {nat} {"Nature"}
+MACRO {nature} {"Nature"}
+MACRO {nedf} {"Nucl. Eng. \& Design/Fusion"}
+MACRO {nf} {"Nucl. Fusion"}
+MACRO {nim} {"Nucl. Inst. \& Meth."}
+MACRO {nimpr} {"Nucl. Inst. \& Meth. in Phys. Res."}
+MACRO {np} {"Nucl. Phys."}
+MACRO {npb} {"Nucl. Phys. B"}
+MACRO {nt/f} {"Nucl. Tech./Fusion"}
+MACRO {npbpc} {"Nucl. Phys. B (Proc. Suppl.)"}
+MACRO {inc} {"Nuovo Cimento"}
+MACRO {nc} {"Nuovo Cimento"}
+MACRO {pf} {"Phys. Fluids"}
+MACRO {pfa} {"Phys. Fluids A: Fluid Dyn."}
+MACRO {pfb} {"Phys. Fluids B: Plasma Phys."}
+MACRO {pl} {"Phys. Lett."}
+MACRO {pla} {"Phys. Lett. A"}
+MACRO {plb} {"Phys. Lett. B"}
+MACRO {prep} {"Phys. Rep."}
+MACRO {pnas} {"Proc. Nat. Acad. Sci. USA"}
+MACRO {pp} {"Phys. Plasmas"}
+MACRO {ppcf} {"Plasma Phys. \& Controlled Fusion"}
+MACRO {phitrsl} {"Philos. Trans. Roy. Soc. London"}
+MACRO {prl} {"Phys. Rev. Lett."}
+MACRO {pr} {"Phys. Rev."}
+MACRO {physrev} {"Phys. Rev."}
+MACRO {pra} {"Phys. Rev. A"}
+MACRO {prb} {"Phys. Rev. B"}
+MACRO {prc} {"Phys. Rev. C"}
+MACRO {prd} {"Phys. Rev. D"}
+MACRO {pre} {"Phys. Rev. E"}
+MACRO {ps} {"Phys. Scripta"}
+MACRO {procrsl} {"Proc. Roy. Soc. London"}
+MACRO {rmp} {"Rev. Mod. Phys."}
+MACRO {rsi} {"Rev. Sci. Inst."}
+MACRO {science} {"Science"}
+MACRO {sciam} {"Sci. Am."}
+MACRO {sam} {"Stud. Appl. Math."}
+MACRO {sjpp} {"Sov. J. Plasma Phys."}
+MACRO {spd} {"Sov. Phys.--Doklady"}
+MACRO {sptp} {"Sov. Phys.--Tech. Phys."}
+MACRO {spu} {"Sov. Phys.--Uspeki"}
+MACRO {st} {"Sky and Telesc."}
+ % End module: physjour.mbs
+ %-------------------------------------------------------------------
+ % Begin module:
+ % \ProvidesFile{geojour.mbs}[2002/07/10 2.0h (PWD)]
+MACRO {aisr} {"Adv. Space Res."}
+MACRO {ag} {"Ann. Geophys."}
+MACRO {anigeo} {"Ann. Geofis."}
+MACRO {angl} {"Ann. Glaciol."}
+MACRO {andmet} {"Ann. d. Meteor."}
+MACRO {andgeo} {"Ann. d. Geophys."}
+MACRO {andphy} {"Ann. Phys.-Paris"}
+MACRO {afmgb} {"Arch. Meteor. Geophys. Bioklimatol."}
+MACRO {atph} {"Atm\'osphera"}
+MACRO {aao} {"Atmos. Ocean"}
+MACRO {ass}{"Astrophys. Space Sci."}
+MACRO {atenv} {"Atmos. Environ."}
+MACRO {aujag} {"Aust. J. Agr. Res."}
+MACRO {aumet} {"Aust. Meteorol. Mag."}
+MACRO {blmet} {"Bound.-Lay. Meteorol."}
+MACRO {bams} {"Bull. Amer. Meteorol. Soc."}
+MACRO {cch} {"Clim. Change"}
+MACRO {cdyn} {"Clim. Dynam."}
+MACRO {cbul} {"Climatol. Bull."}
+MACRO {cap} {"Contrib. Atmos. Phys."}
+MACRO {dsr} {"Deep-Sea Res."}
+MACRO {dhz} {"Dtsch. Hydrogr. Z."}
+MACRO {dao} {"Dynam. Atmos. Oceans"}
+MACRO {eco} {"Ecology"}
+MACRO {empl}{"Earth, Moon and Planets"}
+MACRO {envres} {"Environ. Res."}
+MACRO {envst} {"Environ. Sci. Technol."}
+MACRO {ecms} {"Estuarine Coastal Mar. Sci."}
+MACRO {expa}{"Exper. Astron."}
+MACRO {geoint} {"Geofis. Int."}
+MACRO {geopub} {"Geofys. Publ."}
+MACRO {geogeo} {"Geol. Geofiz."}
+MACRO {gafd} {"Geophys. Astrophys. Fluid Dyn."}
+MACRO {gfd} {"Geophys. Fluid Dyn."}
+MACRO {geomag} {"Geophys. Mag."}
+MACRO {georl} {"Geophys. Res. Lett."}
+MACRO {grl} {"Geophys. Res. Lett."}
+MACRO {ga} {"Geophysica"}
+MACRO {gs} {"Geophysics"}
+MACRO {ieeetap} {"IEEE Trans. Antenn. Propag."}
+MACRO {ijawp} {"Int. J. Air Water Pollut."}
+MACRO {ijc} {"Int. J. Climatol."}
+MACRO {ijrs} {"Int. J. Remote Sens."}
+MACRO {jam} {"J. Appl. Meteorol."}
+MACRO {jaot} {"J. Atmos. Ocean. Technol."}
+MACRO {jatp} {"J. Atmos. Terr. Phys."}
+MACRO {jastp} {"J. Atmos. Solar-Terr. Phys."}
+MACRO {jce} {"J. Climate"}
+MACRO {jcam} {"J. Climate Appl. Meteor."}
+MACRO {jcm} {"J. Climate Meteor."}
+MACRO {jcy} {"J. Climatol."}
+MACRO {jgr} {"J. Geophys. Res."}
+MACRO {jga} {"J. Glaciol."}
+MACRO {jh} {"J. Hydrol."}
+MACRO {jmr} {"J. Mar. Res."}
+MACRO {jmrj} {"J. Meteor. Res. Japan"}
+MACRO {jm} {"J. Meteor."}
+MACRO {jpo} {"J. Phys. Oceanogr."}
+MACRO {jra} {"J. Rech. Atmos."}
+MACRO {jaes} {"J. Aeronaut. Sci."}
+MACRO {japca} {"J. Air Pollut. Control Assoc."}
+MACRO {jas} {"J. Atmos. Sci."}
+MACRO {jmts} {"J. Mar. Technol. Soc."}
+MACRO {jmsj} {"J. Meteorol. Soc. Japan"}
+MACRO {josj} {"J. Oceanogr. Soc. Japan"}
+MACRO {jwm} {"J. Wea. Mod."}
+MACRO {lao} {"Limnol. Oceanogr."}
+MACRO {mwl} {"Mar. Wea. Log"}
+MACRO {mau} {"Mausam"}
+MACRO {meteor} {"``Meteor'' Forschungsergeb."}
+MACRO {map} {"Meteorol. Atmos. Phys."}
+MACRO {metmag} {"Meteor. Mag."}
+MACRO {metmon} {"Meteor. Monogr."}
+MACRO {metrun} {"Meteor. Rundsch."}
+MACRO {metzeit} {"Meteor. Z."}
+MACRO {metgid} {"Meteor. Gidrol."}
+MACRO {mwr} {"Mon. Weather Rev."}
+MACRO {nwd} {"Natl. Weather Dig."}
+MACRO {nzjmfr} {"New Zeal. J. Mar. Freshwater Res."}
+MACRO {npg} {"Nonlin. Proc. Geophys."}
+MACRO {om} {"Oceanogr. Meteorol."}
+MACRO {ocac} {"Oceanol. Acta"}
+MACRO {oceanus} {"Oceanus"}
+MACRO {paleoc} {"Paleoceanography"}
+MACRO {pce} {"Phys. Chem. Earth"}
+MACRO {pmg} {"Pap. Meteor. Geophys."}
+MACRO {ppom} {"Pap. Phys. Oceanogr. Meteor."}
+MACRO {physzeit} {"Phys. Z."}
+MACRO {pps} {"Planet. Space Sci."}
+MACRO {pss} {"Planet. Space Sci."}
+MACRO {pag} {"Pure Appl. Geophys."}
+MACRO {qjrms} {"Quart. J. Roy. Meteorol. Soc."}
+MACRO {quatres} {"Quat. Res."}
+MACRO {rsci} {"Radio Sci."}
+MACRO {rse} {"Remote Sens. Environ."}
+MACRO {rgeo} {"Rev. Geophys."}
+MACRO {rgsp} {"Rev. Geophys. Space Phys."}
+MACRO {rdgeo} {"Rev. Geofis."}
+MACRO {revmeta} {"Rev. Meteorol."}
+MACRO {sgp}{"Surveys in Geophys."}
+MACRO {sp} {"Solar Phys."}
+MACRO {ssr} {"Space Sci. Rev."}
+MACRO {tellus} {"Tellus"}
+MACRO {tac} {"Theor. Appl. Climatol."}
+MACRO {tagu} {"Trans. Am. Geophys. Union (EOS)"}
+MACRO {wrr} {"Water Resour. Res."}
+MACRO {weather} {"Weather"}
+MACRO {wafc} {"Weather Forecast."}
+MACRO {ww} {"Weatherwise"}
+MACRO {wmob} {"WMO Bull."}
+MACRO {zeitmet} {"Z. Meteorol."}
+ % End module: geojour.mbs
+ %-------------------------------------------------------------------
+ % Begin module:
+ % \ProvidesFile{photjour.mbs}[1999/02/24 2.0b (PWD)]
+
+MACRO {appopt} {"Appl. Opt."}
+MACRO {bell} {"Bell Syst. Tech. J."}
+MACRO {ell} {"Electron. Lett."}
+MACRO {jasp} {"J. Appl. Spectr."}
+MACRO {jqe} {"IEEE J. Quantum Electron."}
+MACRO {jlwt} {"J. Lightwave Technol."}
+MACRO {jmo} {"J. Mod. Opt."}
+MACRO {josa} {"J. Opt. Soc. America"}
+MACRO {josaa} {"J. Opt. Soc. Amer.~A"}
+MACRO {josab} {"J. Opt. Soc. Amer.~B"}
+MACRO {jdp} {"J. Phys. (Paris)"}
+MACRO {oc} {"Opt. Commun."}
+MACRO {ol} {"Opt. Lett."}
+MACRO {phtl} {"IEEE Photon. Technol. Lett."}
+MACRO {pspie} {"Proc. Soc. Photo-Opt. Instrum. Eng."}
+MACRO {sse} {"Solid-State Electron."}
+MACRO {sjot} {"Sov. J. Opt. Technol."}
+MACRO {sjqe} {"Sov. J. Quantum Electron."}
+MACRO {sleb} {"Sov. Phys.--Leb. Inst. Rep."}
+MACRO {stph} {"Sov. Phys.--Techn. Phys."}
+MACRO {stphl} {"Sov. Techn. Phys. Lett."}
+MACRO {vr} {"Vision Res."}
+MACRO {zph} {"Z. f. Physik"}
+MACRO {zphb} {"Z. f. Physik~B"}
+MACRO {zphd} {"Z. f. Physik~D"}
+
+MACRO {CLEO} {"CLEO"}
+MACRO {ASSL} {"Adv. Sol.-State Lasers"}
+MACRO {OSA} {"OSA"}
+ % End module: photjour.mbs
+%% Copyright 1994-2007 Patrick W Daly
+MACRO {acmcs} {"ACM Comput. Surv."}
+MACRO {acta} {"Acta Inf."}
+MACRO {cacm} {"Commun. ACM"}
+MACRO {ibmjrd} {"IBM J. Res. Dev."}
+MACRO {ibmsj} {"IBM Syst.~J."}
+MACRO {ieeese} {"IEEE Trans. Software Eng."}
+MACRO {ieeetc} {"IEEE Trans. Comput."}
+MACRO {ieeetcad}
+ {"IEEE Trans. Comput. Aid. Des."}
+MACRO {ipl} {"Inf. Process. Lett."}
+MACRO {jacm} {"J.~ACM"}
+MACRO {jcss} {"J.~Comput. Syst. Sci."}
+MACRO {scp} {"Sci. Comput. Program."}
+MACRO {sicomp} {"SIAM J. Comput."}
+MACRO {tocs} {"ACM Trans. Comput. Syst."}
+MACRO {tods} {"ACM Trans. Database Syst."}
+MACRO {tog} {"ACM Trans. Graphic."}
+MACRO {toms} {"ACM Trans. Math. Software"}
+MACRO {toois} {"ACM Trans. Office Inf. Syst."}
+MACRO {toplas} {"ACM Trans. Progr. Lang. Syst."}
+MACRO {tcs} {"Theor. Comput. Sci."}
+
+MACRO {jhep} {"J. High Energy Phys."}
+MACRO {jcap} {"J. Cosmol. Astropart. Phys."}
+MACRO {jinst} {"J. Instrum."}
+MACRO {jstat} {"J. Stat. Mech.: Theory Exp."}
+
+FUNCTION {bibinfo.command} { "\bibinfo " }
+
+FUNCTION {bibinfo.check}
+{ swap$
+ duplicate$ missing$
+ {
+ pop$
+ pop$ ""
+ }{
+ duplicate$ empty$
+ {
+ swap$ pop$
+ }{
+ swap$
+ bibinfo.command "{" * swap$ * "} {" * swap$ * "}" *
+ }
+ if$
+ }
+ if$
+}
+
+FUNCTION {bibinfo.warn}
+{ swap$
+ duplicate$ missing$
+ {
+ swap$ "missing " swap$ * " in " * cite$ * warning$ pop$
+ ""
+ }{
+ duplicate$ empty$
+ {
+ swap$ "empty " swap$ * " in " * cite$ * warning$
+ }{
+ swap$
+ bibinfo.command " {" * swap$ * "} {" * swap$ * "}" *
+ }
+ if$
+ }
+ if$
+}
+
+FUNCTION {archiv.base}
+{
+ "https://arxiv.org/abs"
+}
+
+FUNCTION {archiv.prefix.base}
+{
+ "arXiv"
+}
+
+FUNCTION {eprint.command}
+{
+ "\Eprint "
+}
+
+FUNCTION {format.eprint.controlled}
+{
+ eprint duplicate$ empty$
+ control.eprint #1 =
+ or
+ { pop$ "" }
+ {
+ duplicate$
+ ""
+ archive duplicate$ empty$ { pop$ archiv.base } 'skip$ if$ *
+ "/" *
+ swap$ *
+ "{" swap$ * "} " *
+ swap$
+ ""
+ archivePrefix duplicate$ empty$ { pop$ "" } { ":" * } if$ *
+ swap$ *
+ primaryClass duplicate$ empty$ { pop$ "" } { " [" swap$ * "]" * } if$ *
+ "{" swap$ * "} " *
+ *
+ eprint.command swap$ *
+ }
+ if$
+}
+
+FUNCTION {format.eprint}
+{
+ eprint duplicate$ empty$
+ { pop$ "" }
+ {
+ duplicate$
+ ""
+ archive duplicate$ empty$ { pop$ archiv.base } 'skip$ if$ *
+ "/" *
+ swap$ *
+ "{" swap$ * "} " *
+ swap$
+ ""
+ archivePrefix duplicate$ empty$ { pop$ "" } { ":" * } if$ *
+ swap$ *
+ primaryClass duplicate$ empty$ { pop$ "" } { " [" swap$ * "]" * } if$ *
+ "{" swap$ * "} " *
+ *
+ eprint.command swap$ *
+ }
+ if$
+}
+
+FUNCTION {format.translation}
+{ translation duplicate$ empty$
+ 'skip$
+ { ""
+ "\translation{" * swap$ * "}" *
+ punctuation.space 'punctuation.state :=
+ }
+ if$
+}
+
+FUNCTION {format.url}
+{
+ url duplicate$ empty$
+ { pop$ "" }
+ {
+ "\url "
+ "{" * swap$ * "}" *
+ }
+ if$
+}
+
+INTEGERS { nameptr namesleft numnames }
+
+FUNCTION {check.speaker}
+{ key empty$ 'skip$
+ { key nameptr int.to.str$ =
+ {
+ bolden
+ }
+ 'skip$
+ if$
+ }
+ if$
+}
+
+STRINGS { bibinfo}
+
+FUNCTION {format.names.fname}
+{
+ control.author.initials {
+ control.author.dotless {
+ control.author.nospace {
+ "f{}"
+ } {
+ "f{~}"
+ } if$
+ } {
+ control.author.nospace {
+ "f{.}."
+ } {
+ "f."
+ } if$
+ } if$
+ } {
+ "ff"
+ } if$
+}
+
+FUNCTION {bracify}
+{
+ "{" swap$ * "}" *
+}
+
+FUNCTION {name.comma}
+{
+ control.author.nocomma 'skip$ { "," swap$ * } if$
+}
+
+FUNCTION {format.names.format.onefont}
+{
+ "{vv~}{ll}"
+ nameptr #1 >
+ control.author.first
+ and
+ control.author.reversed not
+ or
+ {
+ control.author.initials {
+ "f"
+ control.author.dotless 'skip$ {
+ "." *
+ } if$
+ "~" *
+ } {
+ "ff"
+ } if$
+ bracify
+ swap$
+ } {
+ format.names.fname
+ " " swap$ *
+ name.comma
+ bracify
+ }
+ if$
+ "jj"
+ " " swap$ *
+ name.comma
+ bracify
+ control.author.jnrlst 'skip$ 'swap$ if$
+ * *
+}
+
+FUNCTION {format.names.onefont}
+{
+ s nameptr format.names.format.onefont format.name$
+ remove.dots
+ bib.name.font
+}
+
+FUNCTION {format.names.morfont}
+{ s nameptr
+ "{vv~}{ll}" format.name$ bib.name.font
+ nameptr #1 >
+ control.author.first
+ and
+ control.author.reversed not
+ or
+ {
+ s nameptr
+ control.author.initials {
+ "f" % default: name + surname + comma junior
+ } {
+ "ff"
+ } if$
+ control.author.dotless 'skip$ {
+ "." * % nm-init % Initials. + surname (J. F. Smith) control.author.initials
+ } if$
+ bracify
+ format.name$ duplicate$ empty$ 'skip$
+ { tie.or.space.prefix bib.fname.font swap$ * }
+ if$
+ swap$
+ *
+ s nameptr
+ "{jj}" format.name$ duplicate$ empty$ 'skip$
+ { bib.fname.font ", " swap$ * }
+ if$
+ } {
+ "," *
+ s nameptr
+ format.names.fname
+ "jj"
+ " "
+ name.comma
+ control.author.jnrlst {
+ swap$ * skip$
+ } {
+ skip$ * swap$
+ } if$
+ bracify swap$ bracify swap$
+ *
+ format.name$
+ remove.dots
+ duplicate$ empty$ 'skip$
+ { bib.fname.font " " swap$ * }
+ if$
+ } if$
+ *
+}
+
+FUNCTION {names.punctuate}
+{
+ "," *
+ " " *
+}
+%<*!aipauth&!apsrmp>
+
+FUNCTION {names.comma}
+{
+ "," *
+}
+%</!aipauth&!apsrmp>
+
+FUNCTION {format.names}
+{ 'bibinfo :=
+ duplicate$ empty$ { pop$ "" } {
+ duplicate$ num.names$
+ duplicate$ 'numnames :=
+ 'namesleft :=
+ 's :=
+ #1 'nameptr :=
+ ""
+ { namesleft #0 > }
+ {
+ format.names.morfont
+ bibinfo bibinfo.check
+ type$ "presentation" =
+ 'check.speaker
+ 'skip$
+ if$
+ 't :=
+ nameptr #1 > not
+ {
+ t *
+ } {
+ namesleft #1 >
+ {
+ names.punctuate
+ t *
+ } {
+ s nameptr "{ll}" format.name$ duplicate$ "others" =
+ { 't := }
+ { pop$ }
+ if$
+%<*!apsrmp>
+ numnames #2 >
+%<!aipauth> 'names.comma
+%<aipauth> 'names.punctuate
+ 'skip$
+ if$
+%</!apsrmp>
+%<apsrmp> names.punctuate
+ t "others" =
+ {
+ " " *
+ bbl.etal
+ emphasize
+ *
+ } {
+ bbl.and
+%<aapmrev> space.word *
+%<aipauth> word.space * *
+%<aipnum> space.word *
+%<apsrev> space.word *
+%<apsrmp> word.space * *
+ t *
+ }
+ if$
+ }
+ if$
+ }
+ if$
+ nameptr #1 + 'nameptr :=
+ namesleft #1 - 'namesleft :=
+ }
+ while$
+ } if$
+}
+
+FUNCTION {format.names.ed.onefont}
+{
+ s nameptr
+ control.author.initials {
+ control.author.dotless {
+ control.author.nospace {
+ "{f{}~}{vv~}{ll}{ jj}" % nm-rvx|nm-rvcx
+ } {
+ "{f{~}~}{vv~}{ll}{ jj}" % nm-rv
+ } if$
+ } {
+ control.author.nospace {
+ "{f{.}.~}{vv~}{ll}{ jj}" % nm-rvv|nm-rvvc
+ }{
+ "{f.~}{vv~}{ll}{, jj}" % nm-init|nm-rev|nm-rev1
+ } if$
+ } if$
+ } {
+ "{ff~}{vv~}{ll}{, jj}"
+ } if$
+ format.name$
+ remove.dots
+ bib.name.font
+}
+
+FUNCTION {format.names.ed.morfont}
+{
+ control.author.reversed { %
+ control.author.initials { %
+ control.author.dotless { %
+ s nameptr
+ control.author.nospace { % nm-rvx nm-rvcx
+ "{f{}}"
+ } { % nm-rv
+ "{f{~}}"
+ } if$
+ format.name$ duplicate$ empty$ 'skip$
+ { tie.or.space.prefix bib.fname.font swap$ * }
+ if$
+ s nameptr
+ "{vv~}{ll}" format.name$ bib.name.font *
+ s nameptr
+ "{jj}" format.name$
+ remove.dots
+ duplicate$ empty$ 'skip$
+ { bib.fname.font " " swap$ * }
+ if$
+ } { % !control.author.dotless
+ s nameptr
+ control.author.nospace { % nm-rvv
+ "{ff}"
+ } { % nm-rev nm-rev1
+ "{f.}"
+ } if$
+ format.name$ duplicate$ empty$ 'skip$
+ { tie.or.space.prefix bib.fname.font swap$ * }
+ if$
+ s nameptr
+ "{vv~}{ll}" format.name$ bib.name.font *
+ s nameptr
+ "{jj}" format.name$
+ duplicate$ empty$ 'skip$
+ { bib.fname.font ", " swap$ * }
+ if$
+ } if$
+ } { % Full names !control.author.initials nm-revf nm-revv1
+ s nameptr
+ "{ff}"
+ format.name$ duplicate$ empty$ 'skip$
+ { tie.or.space.prefix bib.fname.font swap$ * }
+ if$
+ s nameptr
+ "{vv~}{ll}" format.name$ bib.name.font *
+ s nameptr
+ "{jj}" format.name$
+ duplicate$ empty$ 'skip$
+ { bib.fname.font ", " swap$ * }
+ if$
+ } if$
+ } { % !control.author.reversed nm-init
+ s nameptr
+ "{f.}"
+ format.name$ duplicate$ empty$ 'skip$
+ { tie.or.space.prefix bib.fname.font swap$ * }
+ if$
+ s nameptr
+ "{vv~}{ll}" format.name$ bib.name.font *
+ s nameptr
+ "{jj}" format.name$
+ duplicate$ empty$ 'skip$
+ { bib.fname.font ", " swap$ * }
+ if$
+ } if$
+ *
+}
+
+FUNCTION {format.names.ed}
+{
+ control.editor #0 > {
+ format.names
+ } {
+ 'bibinfo :=
+ duplicate$ empty$ 'skip$ {
+ 's :=
+ "" 't :=
+ #1 'nameptr :=
+ s num.names$ 'numnames :=
+ numnames 'namesleft :=
+ { namesleft #0 > }
+ {
+ format.names.ed.morfont
+ bibinfo bibinfo.check
+ 't :=
+ nameptr #1 >
+ {
+ namesleft #1 >
+ {
+ names.punctuate
+ t *
+ }{
+ s nameptr "{ll}" format.name$ duplicate$ "others" =
+ { 't := }
+ { pop$ }
+ if$
+ numnames #2 >
+%<aapmrev> 'names.comma
+%<aipauth> 'names.punctuate
+%<aipnum> 'names.comma
+%<apsrev> 'names.comma
+%<apsrmp> 'names.punctuate
+ 'skip$
+ if$
+ t "others" =
+ {
+ " " * bbl.etal emphasize *
+ }{
+ bbl.and
+ space.word * t *
+ }
+ if$
+ }
+ if$
+ }
+ 't
+ if$
+ nameptr #1 + 'nameptr :=
+ namesleft #1 - 'namesleft :=
+ }
+ while$
+ } if$
+ } if$
+}
+
+FUNCTION {format.key}
+{ empty$
+ { key field.or.null }
+ { "" }
+ if$
+}
+
+FUNCTION {format.authors}
+{ author "author" format.names
+ duplicate$ empty$ 'skip$
+ { collaboration "collaboration" bibinfo.check
+ duplicate$ empty$ 'skip$
+ { " (" swap$ * ")" * }
+ if$
+ *
+ }
+ if$
+ "author" 'bibfield :=
+}
+
+FUNCTION {get.bbl.editor}
+{ editor num.names$ #1 > 'bbl.editors 'bbl.editor if$
+}
+
+FUNCTION {format.editors}
+{ editor "editor" format.names duplicate$ empty$ 'skip$
+ {
+ "," *
+ word.space *
+ get.bbl.editor
+%<apsrmp> capitalize
+ *
+ }
+ if$
+}
+
+FUNCTION {format.isbn.output}
+{
+}
+
+FUNCTION {format.issn.output}
+{
+}
+
+FUNCTION {doi.base}
+{
+ "https://doi.org/"
+}
+
+FUNCTION {doi.base.command}
+{
+ "\doibase "
+}
+
+FUNCTION {noop.command}
+{
+ "\href@noop "
+}
+
+FUNCTION {href.command}
+{
+ "\href "
+}
+
+FUNCTION {link.tag.open}
+{
+ doi duplicate$ empty$
+ {
+ pop$
+ url duplicate$ empty$
+ {
+ pop$ "" noop.command
+ }{
+ href.command
+ }
+ if$
+ }
+ {
+ "https://doi.org/" swap$ *
+ href.command
+ }
+ if$
+ "{" * swap$ * "} {" *
+}
+
+FUNCTION {link.tag.shut}
+{
+ "}"
+}
+
+FUNCTION {link.open}
+{
+ link.tag.open output.nopunct
+}
+
+FUNCTION {link.shut}
+{
+ link.tag.shut *
+}
+
+FUNCTION {add.doi}
+{
+ link.tag.open swap$ * link.tag.shut *
+}
+
+FUNCTION {select.language}
+{ duplicate$ empty$
+ 'pop$
+ { language empty$
+ 'skip$
+ { "{\selectlanguage {" language * "}" * swap$ * "}" * }
+ if$
+ }
+ if$
+}
+
+FUNCTION {format.note}
+{
+ note empty$
+ { "" }
+ { note #1 #1 substring$
+ duplicate$ "{" =
+ 'skip$
+ {
+ output.state after.word = output.state after.punctuation = or
+ { "l" }
+ { "u" }
+ if$ change.case$
+ }
+ if$
+ note #2 global.max$ substring$ * "note" bibinfo.check
+ }
+ if$
+}
+
+FUNCTION {bbl.enquote}
+{
+ "\enquote "
+}
+
+FUNCTION {string.enquote}
+{
+ punctuation.no 'punctuation.state :=
+ non.stop {
+ block.punctuation
+ } { "" } if$
+ swap$ pop$
+ *
+ bbl.enquote "{" * swap$ * "}" *
+ word.space *
+}
+
+FUNCTION {format.title}
+{ title
+ duplicate$ empty$ 'skip$ { "t" change.case$ } if$
+ duplicate$ "title" bibinfo.check swap$
+ duplicate$ empty$ 'pop$
+ {
+ punctuation.yes 'punctuation.state :=
+%<!apsrev> string.enquote
+%<apsrev> pop$
+ select.language
+ }
+ if$
+}
+
+FUNCTION {end.quote.title}
+{ title empty$
+ 'skip$
+ { before.all 'output.state := }
+ if$
+}
+
+FUNCTION {format.name.apply}
+{
+ s nameptr
+ "{vv~}{ll}"
+ format.name$
+ cite.name.font
+}
+
+FUNCTION {format.full.names}
+{
+ 's :=
+ "" 't :=
+ #1 'nameptr :=
+ s num.names$ 'numnames :=
+ numnames 'namesleft :=
+ { namesleft #0 > }
+ { format.name.apply
+ 't :=
+ nameptr #1 >
+ {
+ namesleft #1 >
+ { ", " * t * }
+ {
+ s nameptr "{ll}" format.name$ duplicate$ "others" =
+ {
+ 't :=
+ }
+ 'pop$
+ if$
+ t "others" =
+ {
+ " " * bbl.etal
+ emphasize *
+ }{
+ numnames #2 > { "," * }{ skip$ } if$
+ bbl.and
+ space.word * t *
+ }
+ if$
+ }
+ if$
+ }
+ 't
+ if$
+ nameptr #1 + 'nameptr :=
+ namesleft #1 - 'namesleft :=
+ }
+ while$
+}
+
+FUNCTION {make.full.names}
+{
+ key editor author
+ type$ "proceedings" =
+ type$ "book" =
+ type$ "inbook" =
+ or { pop$ }{ { pop$ "" }{ swap$ pop$ "" swap$ } if$ } if$
+ duplicate$ empty$
+ { pop$
+ duplicate$ empty$
+ { pop$
+ duplicate$ empty$
+ { pop$
+ cite$ #1 #3 substring$
+ }{
+ skip$
+ }
+ if$
+ }
+ { swap$ pop$ format.full.names }
+ if$
+ }
+ { swap$ pop$ swap$ pop$ format.full.names }
+ if$
+}
+
+FUNCTION {year.bibitem}
+{
+ year duplicate$ empty$
+ { pop$ ""
+%<apsrmp> "????" *
+ }{
+ skip$
+ } if$
+ extra.label *
+}
+
+FUNCTION {output.bibitem}
+{
+ newline$
+ ""
+ label
+ * ")" *
+ make.full.names duplicate$ short.list =
+ { pop$ }{ * } if$
+ bracify
+ "[" swap$ * "]" *
+ cite$ bracify "%" *
+ *
+ "\bibitem "
+ swap$ *
+ write$ newline$
+ " "
+ duplicate$ bbl.open * write$ newline$
+ before.all 'output.state :=
+ punctuation.yes 'punctuation.state :=
+}
+
+FUNCTION {n.dashify}
+{
+ 't :=
+ ""
+ { t empty$ not }
+ { t #1 #1 substring$ "-" =
+ { t #1 #2 substring$ "--" = not
+ { "--" *
+ t #2 global.max$ substring$ 't :=
+ }{
+ { t #1 #1 substring$ "-" = }
+ { "-" *
+ t #2 global.max$ substring$ 't :=
+ }
+ while$
+ }
+ if$
+ }{
+ t #1 #1 substring$ *
+ t #2 global.max$ substring$ 't :=
+ }
+ if$
+ }
+ while$
+}
+
+FUNCTION {word.in}
+{
+ bbl.in
+ word.space *
+}
+
+%<!apsrev>FUNCTION {date.encapsulate}
+%<apsrev>FUNCTION {parenthesize}
+{
+ duplicate$ empty$
+ 'skip$
+ {
+ before.all 'output.state :=
+ " (" swap$ * ")" *
+ }
+ if$
+}
+
+FUNCTION {format.date}
+{
+ year "year" bibinfo.check duplicate$ empty$
+ {
+%<*apsrmp>
+ "empty year in " cite$ * "; set to ????" * warning$
+ pop$ "????"
+%</apsrmp>
+ }
+ 'skip$
+ if$
+ extra.label *
+%<!apsrev> date.encapsulate
+%<apsrev> parenthesize
+}
+
+FUNCTION {format.date.output.check}
+{
+ format.date
+ "year" output.check
+}
+
+FUNCTION {format.date.output}
+{
+ format.date.output.check
+}
+
+FUNCTION {format.btitle}
+{
+ booktitle duplicate$ empty$ { pop$
+ title
+ } 'skip$ if$
+ "title" bibinfo.check
+ duplicate$ empty$ 'skip$
+ {
+ emphasize
+ select.language
+ }
+ if$
+}
+
+FUNCTION {either.or.check}
+{ empty$
+ 'pop$
+ { "can't use both " swap$ * " fields in " * cite$ * warning$ }
+ if$
+}
+
+FUNCTION {editor.check.book}
+{ editor empty$ 'skip$
+ {
+ "can't use both author and editor fields in " cite$ *
+ ": try using @inbook instead" *
+ warning$
+ }
+ if$
+}
+
+FUNCTION {format.bvolume}
+{ volume duplicate$ empty$
+ { pop$ "" }
+ {
+ "volume and number" number either.or.check
+ bbl.volume
+ capitalize
+ swap$
+ tie.or.space.prefix "volume" bibinfo.check * *
+ series "series" bibinfo.check duplicate$ empty$ 'pop$
+ {
+ ", "
+ * swap$ *
+ }
+ if$
+ }
+ if$
+}
+
+FUNCTION {format.number}
+{
+ bbl.number
+ output.state after.word = output.state after.punctuation = or
+ #1 or
+ #0 and
+ 'skip$
+ { capitalize }
+ if$
+ number tie.or.space.prefix "number" bibinfo.check * *
+}
+
+FUNCTION {format.number.series}
+{ volume empty$
+ { number empty$
+ { series field.or.null }
+ {
+ series empty$
+ {
+ number "number" bibinfo.check
+ }{
+ format.number
+ series "series" bibinfo.check
+ word.space * swap$ *
+ }
+ if$
+ }
+ if$
+ }
+ { "" }
+ if$
+}
+
+FUNCTION {is.num}
+{ chr.to.int$
+ duplicate$ "0" chr.to.int$ < not
+ swap$ "9" chr.to.int$ > not and
+}
+
+FUNCTION {extract.num}
+{ duplicate$ 't :=
+ "" 's :=
+ { t empty$ not }
+ { t #1 #1 substring$
+ t #2 global.max$ substring$ 't :=
+ duplicate$ is.num
+ { s swap$ * 's := }
+ { pop$ "" 't := }
+ if$
+ }
+ while$
+ s empty$
+ 'skip$
+ { pop$ s }
+ if$
+}
+
+FUNCTION {convert.edition}
+{ extract.num "l" change.case$ 's :=
+ s "first" = s "1" = or
+ { bbl.first 't := }
+ { s "second" = s "2" = or
+ { bbl.second 't := }
+ { s "third" = s "3" = or
+ { bbl.third 't := }
+ { s "fourth" = s "4" = or
+ { bbl.fourth 't := }
+ { s "fifth" = s "5" = or
+ { bbl.fifth 't := }
+ { s #1 #1 substring$ is.num
+ { s
+ eng.ord
+ 't := }
+ { edition 't := }
+ if$
+ }
+ if$
+ }
+ if$
+ }
+ if$
+ }
+ if$
+ }
+ if$
+ t
+}
+
+FUNCTION {format.edition}
+{ edition duplicate$ empty$ 'skip$
+ {
+ convert.edition
+ output.state after.word = output.state after.punctuation = or
+ { "l" }
+ { "t" }
+ if$ change.case$
+ "edition" bibinfo.check
+ word.space * bbl.edition *
+ }
+ if$
+}
+
+INTEGERS { multiresult }
+FUNCTION {multi.page.check}
+{ 't :=
+ #0 'multiresult :=
+ { multiresult not
+ t empty$ not
+ and
+ }
+ { t #1 #1 substring$
+ duplicate$ "-" =
+ swap$ duplicate$ "," =
+ swap$ "+" =
+ or or
+ { #1 'multiresult := }
+ { t #2 global.max$ substring$ 't := }
+ if$
+ }
+ while$
+ multiresult
+}
+
+FUNCTION {format.pages}
+{ pages duplicate$ empty$
+ 'skip$
+ { duplicate$ multi.page.check
+ {
+ bbl.pages swap$
+ n.dashify
+ }{
+ bbl.page swap$
+ }
+ if$
+ tie.or.space.prefix
+ "pages" bibinfo.check
+ * *
+ }
+ if$
+}
+
+FUNCTION {first.page}
+{ 't :=
+ ""
+ { t empty$ not t #1 #1 substring$ "-" = not and }
+ { t #1 #1 substring$ *
+ t #2 global.max$ substring$ 't :=
+ }
+ while$
+}
+
+FUNCTION {format.book.pages}
+{
+ pages duplicate$ empty$ 'skip$
+ {
+ "pages" bibinfo.check word.space bbl.pages * *
+ }
+ if$
+}
+
+FUNCTION {volnum.punct}
+{
+ ","
+ word.space *
+}
+
+FUNCTION {format.journal.pages}
+{ pages duplicate$ empty$ 'pop$
+ { swap$ duplicate$ empty$
+ { pop$ pop$ format.pages }
+ { volnum.punct *
+ swap$
+ control.pages duplicate$ #0 < {
+ pop$ pop$
+ }{
+ #0 >
+ {
+ n.dashify
+ }{
+ first.page
+ } if$
+ } if$
+ "pages" bibinfo.check
+ *
+ }
+ if$
+ }
+ if$
+}
+
+FUNCTION {format.journal.eid}
+{ eid "eid" bibinfo.check
+ duplicate$ empty$ 'pop$
+ { swap$ duplicate$ empty$ 'skip$
+ { volnum.punct * }
+ if$
+ swap$ *
+ }
+ if$
+}
+
+FUNCTION {eid.or.pages}
+{
+ eid empty$
+ { format.journal.pages }
+ { format.journal.eid }
+ if$
+}
+
+FUNCTION {format.ser.vol.num}
+{
+ series "series" bibinfo.check output
+ volume field.or.null
+ duplicate$ empty$ 'skip$
+ {
+ "volume" bibinfo.check
+ }
+ if$
+ bolden
+%<*apsrmp>
+ number "number" bibinfo.check duplicate$ empty$ 'skip$
+ {
+ swap$ duplicate$ empty$
+ { "there's a number but no volume in " cite$ * warning$ }
+ 'skip$
+ if$
+ swap$
+ "~(" swap$ * ")" *
+ }
+ if$ *
+%</apsrmp>
+}
+
+%<*apsrev>
+FUNCTION {format.vol.num}
+{
+ volume field.or.null
+ duplicate$ empty$ 'skip$
+ {
+ "volume" bibinfo.check
+ }
+ if$
+ bolden
+}
+
+%</apsrev>
+FUNCTION {format.chapter.pages}
+{ chapter empty$
+ {
+%<!apsrmp> format.pages
+%<apsrmp> ""
+ }
+ { type empty$
+ {
+ bbl.chapter
+ capitalize
+ }{
+ type
+ capitalize
+ "type" bibinfo.check
+ }
+ if$
+ chapter tie.or.space.prefix
+ "chapter" bibinfo.check
+ * *
+%<*!apsrmp>
+ pages empty$
+ 'skip$
+ { ", " * format.pages * }
+ if$
+%</!apsrmp>
+ }
+ if$
+}
+
+FUNCTION {format.booktitle}
+{
+ booktitle duplicate$ "booktitle" bibinfo.check swap$
+ duplicate$ empty$ 'pop$
+ {
+ punctuation.yes 'punctuation.state :=
+ pop$ emphasize
+ select.language
+ }
+ if$
+}
+
+FUNCTION {format.editor.in}
+{
+ editor "editor" format.names.ed duplicate$ empty$ 'skip$
+ {
+ bbl.edby
+ word.space * swap$ *
+ }
+ if$
+}
+
+FUNCTION {output.article.booktitle}
+{
+ format.booktitle
+ "booktitle" 'bibfield :=
+ output
+ bookaddress "address" bibinfo.check duplicate$ empty$ 'pop$
+ {
+ "address" 'bibfield :=
+ output.nonnull after.punctuation 'output.state :=
+ }
+ if$
+}
+
+FUNCTION {format.in.ed.booktitle}
+{
+ format.booktitle duplicate$ empty$ 'pop$
+ {
+ add.doi
+ word.in swap$ * output.nonnull
+ bookaddress "address" bibinfo.check output
+ format.number.series "series and number" bibinfo.check output
+ format.bvolume output
+ format.editor.in "editor" bibinfo.check output
+ }
+ if$
+}
+
+FUNCTION {format.in.ed.booktitle.inbook}
+{
+ format.booktitle duplicate$ empty$ 'pop$
+ {
+ add.doi
+ word.in swap$ * output.nonnull
+ bookaddress "address" bibinfo.check output
+ format.number.series "series and number" bibinfo.check output
+ format.bvolume output
+ author empty$ 'skip$
+ { format.editor.in "editor" bibinfo.check output }
+ if$
+ }
+ if$
+}
+
+FUNCTION {format.thesis.type}
+{ type duplicate$ empty$
+ 'pop$
+ { swap$ pop$
+ "t" change.case$ "type" bibinfo.check
+ }
+ if$
+}
+
+FUNCTION {format.tr.number}
+{ number "number" bibinfo.check
+ type duplicate$ empty$
+ { pop$ bbl.techrep }
+ 'skip$
+ if$
+ "type" bibinfo.check
+ swap$ duplicate$ empty$
+ { pop$ "t" change.case$ }
+ { tie.or.space.prefix * * }
+ if$
+}
+
+FUNCTION {format.article.crossref}
+{
+ word.in
+ " \cite{" * crossref * "}" *
+}
+
+FUNCTION {format.book.crossref}
+{ volume duplicate$ empty$
+ { "empty volume in " cite$ * "'s crossref of " * crossref * warning$
+ pop$ word.in
+ }
+ { bbl.volume
+ swap$ tie.or.space.prefix "volume" bibinfo.check * * bbl.of space.word *
+ }
+ if$
+ " \cite{" * crossref * "}" *
+}
+
+FUNCTION {format.incoll.inproc.crossref}
+{
+ word.in
+ " \cite{" * crossref * "}" *
+}
+
+FUNCTION {format.org.or.pub}
+{ 't :=
+ ""
+ address "address" bibinfo.check
+ duplicate$ empty$
+ { pop$ t }
+ { t duplicate$ empty$
+ { pop$ }
+ {
+ "," word.space *
+ * swap$ *
+ }
+ if$
+ }
+ if$
+ *
+%<*!apsrmp>
+ year duplicate$ empty$
+ {
+ "empty year in " cite$ *
+ warning$
+ pop$ ""
+ }
+ 'skip$
+ if$
+ duplicate$ empty$
+ { pop$ }
+ {
+ "year" bibinfo.check
+ swap$
+ duplicate$ empty$
+ { pop$ }
+ {
+ "," *
+ word.space *
+ swap$ *
+ }
+ if$
+ }
+ if$
+%</!apsrmp>
+ duplicate$ empty$ 'skip$ {
+ ""
+ "(" * swap$ * ")" *
+ after.punctuation 'output.state :=
+ punctuation.space 'punctuation.state :=
+ } if$
+}
+
+FUNCTION {format.publisher.address}
+{ publisher "publisher" bibinfo.warn format.org.or.pub
+}
+
+FUNCTION {format.organization.address}
+{ organization "organization" bibinfo.check format.org.or.pub
+}
+
+FUNCTION {format.organization.publisher.address}
+{
+ publisher empty$
+ { format.organization.address }
+ { organization "organization" bibinfo.check output
+ format.publisher.address
+ }
+ if$
+}
+
+FUNCTION {format.school.address.output}
+{
+ school "school" bibinfo.warn
+ address "address" bibinfo.check
+ duplicate$ empty$ 'skip$
+ {
+ swap$
+ duplicate$ empty$ 'skip$
+ {
+ ", " *
+ }
+ if$
+ swap$
+ }
+ if$
+ *
+%<apsrmp> duplicate$ empty$ 'skip$ { "(" swap$ * ")" * } if$ after.punctuation 'output.state :=
+ output
+}
+
+FUNCTION {article.title.produce}
+{
+ control.title duplicate$ #0 <
+ { pop$
+ }{
+ format.title
+ "title" 'bibfield :=
+ swap$ #0 >
+ {
+ "title" output.check
+ }{
+ output
+ } if$
+ new.block.comma
+ } if$
+}
+
+FUNCTION {control}
+{
+}
+
+FUNCTION {article}
+{ output.bibitem
+ format.authors
+ booktitle empty$ {
+ "author" output.check
+ }{ output } if$
+ author format.key output
+%<!apsrmp> new.block.comma
+%<apsrmp> format.date.output.check date.block
+ article.title.produce
+ output.article.booktitle
+ crossref missing$
+%<apsrev>%% Handle special case of SISSA journals which require an issue number for unique citations and use volume/year interchangably
+ {
+%<*!apsrev>
+ link.open
+ journal
+ "journal" bibinfo.warn
+ "journal" 'bibfield :=
+ output
+ add.blank
+ format.ser.vol.num
+ output
+ eid.or.pages
+%<!apsrmp> format.date.output.check
+ pages empty$ {
+ doi output
+ } 'skip$ if$
+ link.shut
+%</!apsrev>
+%<*apsrev>
+ journal "J. High Energy Phys." =
+ journal "J. High Energ. Phys." = or
+ journal "JHEP" = or
+ journal "Journal of Cosmology and Astroparticle Physics" = or
+ journal "J. Cosmol. Astropart. Phys." = or
+ journal "JCAP" = or
+ journal "Journal of Instrumentation" = or
+ journal "J. Instrum." = or
+ journal "JINST" = or
+ journal "Journal of Statistical Mechanics" = or
+ journal "J. of Stat. Mech." = or
+ journal "J. Stat. Mech.: Theory Exp." = or
+ journal "JSTAT" = or
+ {
+ link.open
+ journal
+ "journal" bibinfo.warn
+ "journal" 'bibfield :=
+ output
+ add.blank
+ format.vol.num output
+ number parenthesize
+ "number" bibinfo.warn
+ "number" 'bibfield :=
+ output
+ eid.or.pages
+ link.shut
+ }
+ {
+ eid missing$
+ pages missing$ and
+ doi missing$ not and
+ {
+ journal
+ "journal" bibinfo.warn
+ "journal" 'bibfield :=
+ output
+ add.blank
+ format.ser.vol.num
+ output
+ link.open
+ doi output
+ link.shut
+ format.date.output.check
+ }
+ {
+ link.open
+ journal
+ "journal" bibinfo.warn
+ "journal" 'bibfield :=
+ output
+ add.blank
+ format.ser.vol.num
+ output
+ eid.or.pages
+ format.date.output.check
+ link.shut
+ } if$
+ } if$
+%</apsrev>
+ }{
+ format.article.crossref output.nonnull
+%<!apsrmp> format.pages output
+%<apsrmp> eid.or.pages
+ }
+ if$
+%<!apsrev&!apsrmp> format.issn.output
+ new.block.comma
+ new.sentence.comma
+ format.note output
+ format.eprint.controlled output
+ format.translation output
+ fin.entry
+}
+
+FUNCTION {book}
+{ output.bibitem
+ author empty$
+ {
+ format.editors "author and editor" output.check
+ editor format.key output
+ }{
+ format.authors output.nonnull
+%% crossref missing$ { editor.check.book } 'skip$ if$
+ }
+ if$
+%<!apsrmp> new.block.comma
+%<apsrmp> format.date.output.check date.block
+ link.open
+ format.btitle
+ "title" output.check
+ link.shut
+ format.edition output
+ author empty$
+ {
+ }
+ {
+ format.editor.in output
+ editor format.key output
+ }
+ if$
+ format.number.series output
+ crossref missing$
+ {
+ format.bvolume output
+ new.block.comma
+ format.publisher.address output
+ }{
+ new.block.comma
+ format.book.crossref output.nonnull
+%<!apsrmp> format.date.output.check
+ }
+ if$
+ format.isbn.output
+ format.chapter.pages
+ output
+ new.block.comma
+ new.sentence.comma
+ format.note output
+ format.eprint output
+ fin.entry
+}
+
+FUNCTION {booklet}
+{ output.bibitem
+ format.authors output
+ author format.key output
+%<!apsrmp> new.block.comma
+%<apsrmp> format.date.output date.block
+ link.open
+ format.title
+ "title" output.check
+ link.shut
+ new.block.comma
+ howpublished "howpublished" bibinfo.check output
+ address "address" bibinfo.check output
+%<!apsrmp> format.date.output
+ format.isbn.output
+ format.book.pages output
+ new.block.comma
+ new.sentence.comma
+ format.note output
+ format.eprint output
+ fin.entry
+}
+
+FUNCTION {footnote}
+{ output.bibitem
+ format.note output
+ fin.entry
+}
+
+FUNCTION {inbook}
+{ output.bibitem
+ author empty$
+ {
+ format.editors "editor" output.check
+ editor format.key output
+ }{
+ format.authors output.nonnull
+ author format.key output
+ }
+ if$
+%<!apsrmp> new.block.comma
+%<apsrmp> format.date.output.check date.block
+
+ format.title output
+ new.block.comma
+
+ crossref missing$
+ {
+ format.in.ed.booktitle.inbook
+%<!apsrmp> format.publisher.address output
+ format.chapter.pages
+%<!apsrmp> "chapter and pages"
+%<apsrmp> "chapter"
+ output.check
+ new.block.comma
+ format.edition output
+ new.block.comma
+%<apsrmp> format.publisher.address output
+ }{
+ format.chapter.pages
+%<!apsrmp> "chapter and pages"
+%<apsrmp> "chapter"
+ output.check
+ new.block.comma
+ format.book.crossref output.nonnull
+%<!apsrmp> format.date.output.check
+ }
+ if$
+ crossref missing$
+ { format.isbn.output }
+ 'skip$
+ if$
+%<apsrmp> format.pages "pages" output.check
+ new.block.comma
+ new.sentence.comma
+ format.note output
+ format.eprint output
+ fin.entry
+}
+
+FUNCTION {incollection}
+{ output.bibitem
+ format.authors "author" output.check
+ author format.key output
+%<!apsrmp> new.block.comma
+%<apsrmp> format.date.output.check date.block
+ article.title.produce
+ crossref missing$
+ {
+ format.in.ed.booktitle
+%<!apsrmp> format.publisher.address output
+ format.edition output
+ format.chapter.pages output
+%<apsrmp> format.publisher.address output
+ format.isbn.output
+ }{
+ format.incoll.inproc.crossref output.nonnull
+ format.chapter.pages output
+ }
+ if$
+%<apsrmp> format.pages "pages" output.check
+ new.block.comma
+ new.sentence.comma
+ format.note output
+ format.eprint output
+ fin.entry
+}
+
+FUNCTION {inproceedings}
+{ output.bibitem
+ format.authors "author" output.check
+ author format.key output
+%<!apsrmp> new.block.comma
+%<apsrmp> format.date.output.check date.block
+ article.title.produce
+ crossref missing$
+ {
+ format.in.ed.booktitle
+ format.organization.publisher.address output
+%<!apsrmp> format.chapter.pages output
+ format.isbn.output
+ format.issn.output
+ }{
+ format.incoll.inproc.crossref output.nonnull
+%<!apsrmp> format.chapter.pages output
+ }
+ if$
+%<apsrmp> format.chapter.pages "chapter and pages" output.check
+%<apsrmp> format.pages "pages" output.check
+ new.block.comma
+ new.sentence.comma
+ format.note output
+ format.eprint output
+ fin.entry
+}
+
+FUNCTION {conference} { inproceedings }
+FUNCTION {manual}
+{ output.bibitem
+ format.authors output
+ author format.key output
+%<!apsrmp> new.block.comma
+%<apsrmp> format.date.output date.block
+ link.open
+ format.btitle
+ "title" output.check
+ link.shut
+ organization "organization" bibinfo.check output
+ address "address" bibinfo.check output
+ format.edition output
+%<!apsrmp> format.date.output
+ new.block.comma
+ new.sentence.comma
+ format.note output
+ format.eprint output
+ fin.entry
+}
+
+FUNCTION {mastersthesis}
+{ output.bibitem
+ format.authors "author" output.check
+ author format.key output
+%<!apsrmp> new.block.comma
+%<apsrmp> format.date.output.check date.block
+ format.btitle
+ output
+ new.block.comma
+ link.open
+ bbl.mthesis
+ format.thesis.type
+ output.nonnull
+ link.shut
+ format.school.address.output
+%<!apsrmp> format.date.output.check
+ new.block.comma
+ new.sentence.comma
+ format.note output
+ format.eprint output
+ fin.entry
+}
+
+FUNCTION {misc}
+{ output.bibitem
+ format.authors output
+ author format.key output
+%<apsrmp> format.date.output
+%<!apsrmp> new.block.comma
+%<apsrmp> date.block
+ link.open
+ format.title
+ output
+ link.shut
+ new.block.comma
+ howpublished "howpublished" bibinfo.check output
+%<!apsrmp> format.date.output
+ new.block.comma
+ new.sentence.comma
+ format.note output
+ format.eprint output
+ fin.entry
+}
+
+FUNCTION {phdthesis}
+{ output.bibitem
+ format.authors "author" output.check
+ author format.key output
+%<!apsrmp> new.block.comma
+%<apsrmp> format.date.output.check date.block
+ format.btitle
+ output
+ new.block.comma
+ link.open
+ bbl.phdthesis
+ format.thesis.type
+ output.nonnull
+ link.shut
+ format.school.address.output
+%<!apsrmp> format.date.output.check
+ new.block.comma
+ new.sentence.comma
+ format.note output
+ format.eprint output
+ fin.entry
+}
+
+FUNCTION {presentation}
+{ output.bibitem
+ format.authors output
+ author format.key output
+ new.block.comma
+ link.open
+ format.title
+ output
+ link.shut
+ new.block.comma
+ format.organization.address "organization and address" output.check
+ month "month" output.check
+ year "year" output.check
+ new.block.comma
+ new.sentence.comma
+ format.note output
+ new.sentence
+ type missing$ 'skip$
+ {"(" type capitalize * ")" * output}
+ if$
+ fin.entry
+}
+
+FUNCTION {proceedings}
+{ output.bibitem
+ format.editors output
+ editor format.key output
+%<!apsrmp> new.block.comma
+%<apsrmp> format.date.output.check date.block
+ link.open
+ format.btitle
+ "title" output.check
+ link.shut
+ bookaddress "address" bibinfo.check output
+ format.number.series output
+ format.bvolume output
+ format.organization.publisher.address output
+ format.isbn.output
+ format.issn.output
+ new.block.comma
+ new.sentence.comma
+ format.note output
+ format.eprint output
+ fin.entry
+}
+
+FUNCTION {techreport}
+{ output.bibitem
+ format.authors "author" output.check
+ author format.key output
+%<!apsrmp> new.block.comma
+%<apsrmp> format.date.output.check date.block
+ link.open
+%<!apsrev&!apsrmp> format.title
+%<apsrev|apsrmp> format.btitle
+ "title" output.check
+ link.shut
+ new.block.comma
+ format.tr.number
+ output.nonnull
+ institution "institution" bibinfo.warn
+ format.org.or.pub output
+ new.block.comma
+ new.sentence.comma
+ format.note output
+ format.eprint output
+ fin.entry
+}
+
+FUNCTION {unpublished}
+{ output.bibitem
+ format.authors "author" output.check
+ author format.key output
+%<*!apsrev>
+%<!apsrmp> new.block.comma
+%<apsrmp> format.date.output date.block
+ link.open
+ format.title
+ "title" output.check
+ link.shut
+%</!apsrev>
+%<*apsrev>
+ article.title.produce
+ format.eprint output
+%</apsrev>
+%<!apsrmp> format.date.output
+ new.block.comma
+ new.sentence.comma
+ format.note "note" output.check
+%<!apsrev> format.eprint output
+ fin.entry
+}
+
+FUNCTION {dataset}
+{
+ output.bibitem
+ format.authors "author" output.check
+ author format.key output
+%<apsrmp> format.date.output date.block
+ article.title.produce
+ doi missing$
+ url missing$ not and
+ {
+ link.open
+ url "url" output.check
+ link.shut
+ }
+ {
+ link.open
+ doi "doi" output.check
+ link.shut
+ } if$
+%<!apsrmp> format.date.output
+ format.note output
+ fin.entry
+}
+
+FUNCTION {default.type} { misc }
+
+READ
+
+EXECUTE {control.init}
+
+ITERATE {control.pass}
+
+EXECUTE {control.check}
+
+FUNCTION {sortify}
+{ purify$
+ "l" change.case$
+}
+
+INTEGERS { len }
+
+FUNCTION {chop.word}
+{ 's :=
+ 'len :=
+ s #1 len substring$ =
+ { s len #1 + global.max$ substring$ }
+ 's
+ if$
+}
+
+FUNCTION {cite.name.font.apply}
+{
+ word.space * bbl.etal
+ emphasize
+ *
+}
+
+%<*!apsrev&!apsrmp>
+FUNCTION {format.lab.names}
+{ 's :=
+ "" 't :=
+ #1 'nameptr :=
+ s num.names$ 'numnames :=
+ numnames 'namesleft :=
+ { namesleft #0 > }
+ {
+ format.name.apply
+ 't :=
+ nameptr #1 >
+ {
+ nameptr
+ #2
+ =
+ numnames
+ #3
+ > and
+ {
+ "others" 't :=
+ #1 'namesleft :=
+ }
+ 'skip$
+ if$
+ namesleft #1 >
+ { ", " * t * }
+ {
+ s nameptr "{ll}" format.name$ duplicate$ "others" =
+ { 't := }
+ { pop$ }
+ if$
+ t "others" =
+ {
+ cite.name.font.apply
+ }{
+ numnames #2 > { "," * } 'skip$ if$
+ bbl.and
+ space.word * t *
+ }
+ if$
+ }
+ if$
+ }
+ 't
+ if$
+ nameptr #1 + 'nameptr :=
+ namesleft #1 - 'namesleft :=
+ }
+ while$
+}
+
+%</!apsrev&!apsrmp>
+%<*apsrev|apsrmp>
+FUNCTION {format.lab.names}
+{ 's :=
+ "" 't :=
+ #1 'nameptr :=
+ format.name.apply
+ s num.names$ duplicate$
+ #2 >
+ { pop$
+ cite.name.font.apply
+ }{
+ #2 <
+ 'skip$
+ {
+ s #2 "{ff }{vv }{ll}{ jj}" format.name$ "others" =
+ {
+ cite.name.font.apply
+ }{
+ bbl.and space.word *
+ s #2 "{vv~}{ll}" format.name$
+ cite.name.font
+ *
+ }
+ if$
+ }
+ if$
+ }
+ if$
+}
+
+%</apsrev|apsrmp>
+FUNCTION {author.key.label}
+{ author empty$
+ { key empty$
+ { cite$ #1 #3 substring$ }
+ 'key
+ if$
+ }
+ { author format.lab.names }
+ if$
+}
+
+FUNCTION {author.editor.key.label}
+{ author empty$
+ { editor empty$
+ { key empty$
+ { cite$ #1 #3 substring$ }
+ 'key
+ if$
+ }
+ { editor format.lab.names }
+ if$
+ }
+ { author format.lab.names }
+ if$
+}
+
+FUNCTION {editor.key.label}
+{ editor empty$
+ { key empty$
+ { cite$ #1 #3 substring$ }
+ 'key
+ if$
+ }
+ { editor format.lab.names }
+ if$
+}
+
+FUNCTION {calc.short.authors}
+{ type$ "book" =
+ type$ "inbook" =
+ or
+ 'author.editor.key.label
+ { type$ "proceedings" =
+ 'editor.key.label
+ 'author.key.label
+ if$
+ }
+ if$
+ 'short.list :=
+}
+
+FUNCTION {calc.label}
+{
+ calc.short.authors
+ short.list
+ year duplicate$ empty$
+%<apsrev|apsrmp> short.list key field.or.null = or
+ {
+ pop$ ""
+%<apsrmp> "????" *
+ }{
+ control.year #0 > { purify$ #-1 #4 substring$ } 'skip$ if$
+ }
+ if$
+ "(" swap$ *
+ * 'label :=
+}
+
+FUNCTION {sort.format.names}
+{ 's :=
+ #1 'nameptr :=
+ ""
+ s num.names$ 'numnames :=
+ numnames 'namesleft :=
+ { namesleft #0 > }
+ { s nameptr
+ "{ll{ }}"
+ control.author.initials {
+ "{ f{ }}" *
+ }{
+ "{ ff{ }}" *
+ } if$
+ "{ jj{ }}" *
+ format.name$ 't :=
+ nameptr #1 >
+ {
+ " " *
+ namesleft #1 = t "others" = and
+ { "zzzzz" * }
+ {
+ t sortify *
+ }
+ if$
+ }
+ { t sortify * }
+ if$
+ nameptr #1 + 'nameptr :=
+ namesleft #1 - 'namesleft :=
+ }
+ while$
+}
+
+FUNCTION {sort.format.title}
+{ 't :=
+ "A " #2
+ "An " #3
+ "The " #4 t chop.word
+ chop.word
+ chop.word
+ sortify
+ #1 global.max$ substring$
+}
+
+FUNCTION {author.sort}
+{ author empty$
+ { key empty$
+ { "to sort, need author or key in " cite$ * warning$
+ ""
+ }
+ { key sortify }
+ if$
+ }
+ { author sort.format.names }
+ if$
+}
+
+FUNCTION {author.editor.sort}
+{ author empty$
+ { editor empty$
+ { key empty$
+ { "to sort, need author, editor, or key in " cite$ * warning$
+ ""
+ }
+ { key sortify }
+ if$
+ }
+ { editor sort.format.names }
+ if$
+ }
+ { author sort.format.names }
+ if$
+}
+
+FUNCTION {year.sort.key}
+{
+ year
+}
+
+FUNCTION {editor.sort}
+{ editor empty$
+ { key empty$
+ { "to sort, need editor or key in " cite$ * warning$
+ ""
+ }
+ { key sortify }
+ if$
+ }
+ { editor sort.format.names }
+ if$
+}
+
+INTEGERS { seq.num }
+
+FUNCTION {init.seq}
+{ #0 'seq.num :=}
+
+EXECUTE {init.seq}
+
+FUNCTION {int.to.fix}
+{ "000000000" swap$ int.to.str$ *
+ #-1 #10 substring$
+}
+
+FUNCTION {label.presort}
+{
+ calc.label
+ label sortify
+ " "
+ *
+%<*!aipauth&!apsrmp>
+ seq.num #1 + 'seq.num :=
+ seq.num int.to.fix
+%</!aipauth&!apsrmp>
+%<*aipauth|apsrmp>
+ type$ "book" =
+ type$ "inbook" =
+ or
+ 'author.editor.sort
+ { type$ "proceedings" =
+ 'editor.sort
+ 'author.sort
+ if$
+ }
+ if$
+ #1 entry.max$ substring$
+%</aipauth|apsrmp>
+ 'sort.label :=
+ sort.label
+ *
+ " "
+ *
+ title field.or.null sort.format.title
+ *
+ #1 entry.max$ substring$
+ 'sort.key$ :=
+}
+
+FUNCTION {presort.pass}
+{ type$ "control" = 'control.presort 'label.presort if$
+}
+
+ITERATE {presort.pass}
+
+SORT
+
+STRINGS { last.label next.extra }
+
+INTEGERS { last.extra.num number.label }
+
+FUNCTION {initialize.extra.label.stuff}
+{ #0 int.to.chr$ 'last.label :=
+ "" 'next.extra :=
+ #0 'last.extra.num :=
+ #0 'number.label :=
+}
+
+FUNCTION {label.forward}
+{
+ last.label label =
+ { last.extra.num #1 + 'last.extra.num :=
+ last.extra.num int.to.chr$ 'extra.label :=
+ }
+ { "a" chr.to.int$ 'last.extra.num :=
+ "" 'extra.label :=
+ label 'last.label :=
+ }
+ if$
+ number.label #1 + 'number.label :=
+}
+
+FUNCTION {label.reverse}
+{ next.extra "b" =
+ { "a" 'extra.label := }
+ 'skip$
+ if$
+ extra.label 'next.extra :=
+ extra.label
+ duplicate$ empty$
+ 'skip$
+ { "{\natexlab{" swap$ * "}}" * }
+ if$
+ 'extra.label :=
+ label extra.label * 'label :=
+}
+
+EXECUTE {initialize.extra.label.stuff}
+
+FUNCTION {forward.pass}
+{ type$ "control" = 'control.forward 'label.forward if$
+}
+
+ITERATE {forward.pass}
+
+FUNCTION {reverse.pass}
+{ type$ "control" = 'control.reverse 'label.reverse if$
+}
+
+REVERSE {reverse.pass}
+
+FUNCTION {sortkey.sort}
+{ sort.label
+ " "
+ *
+ year.sort.key
+ field.or.null sortify
+ *
+ " "
+ *
+ title field.or.null sort.format.title
+ *
+ #1 entry.max$ substring$
+ 'sort.key$ :=
+}
+
+FUNCTION {bib.sort.pass}
+{ type$ "control" = 'control.sort 'sortkey.sort if$
+}
+
+ITERATE {bib.sort.pass}
+
+SORT
+
+FUNCTION {init.bib.eprint}
+{
+ "\texttt {"
+ pop$
+ "\providecommand \url [0]{\begingroup\@sanitize@url \@url }%" write$ newline$
+ "\providecommand \@url [1]{\endgroup\@href {#1}{" "}}%" bbl.url.prefix swap$ * * write$ newline$
+ "\providecommand " " [0]{URL }%" bbl.url.prefix swap$ * * write$ newline$
+ eprint.command "\providecommand " swap$ * "[0]{\href }%" * write$ newline$
+}
+
+FUNCTION {init.bib.doi}
+{
+ "\providecommand \doibase [0]{" doi.base "}%" * * write$ newline$
+}
+
+FUNCTION {init.bib.hypertex}
+{
+ "\providecommand " noop.command "[0]{\@secondoftwo}%" * * write$ newline$
+ "\providecommand " href.command "[0]{\begingroup \@sanitize@url \@href}%" * * write$ newline$
+ "\providecommand \@href[1]{\@@startlink{#1}\@@href}%" write$ newline$
+ "\providecommand \@@href[1]{\endgroup#1\@@endlink}%" write$ newline$
+ "\providecommand \@sanitize@url [0]{\catcode `\\12\catcode `\$12\catcode `\&12\catcode `\#12\catcode `\^12\catcode `\_12\catcode `\%12\relax}%" write$ newline$
+ "\providecommand \@@startlink[1]{}%" write$ newline$
+ "\providecommand \@@endlink[0]{}%" write$ newline$
+}
+
+FUNCTION {init.bib.namefont}
+{
+ "\providecommand \bibnamefont [1]{#1}%" write$ newline$
+ "\providecommand \bibfnamefont [1]{#1}%" write$ newline$
+ "\providecommand \citenamefont [1]{#1}%" write$ newline$
+}
+
+FUNCTION {init.bib.quote}
+{
+ "\providecommand " bbl.enquote * " [1]{" *
+ "``" "''"
+ "#1" swap$ "}%" * * * * write$ newline$
+}
+
+FUNCTION {init.bib.ay}
+{
+ "\providecommand \natexlab [1]{#1}%"
+ write$ newline$
+}
+
+FUNCTION {init.bib.bibinfo}
+{
+ bibinfo.command "\providecommand " swap$ * " [0]{\@secondoftwo}%" * write$ newline$
+ bibfield.command "\providecommand " swap$ * " [0]{\@secondoftwo}%" * write$ newline$
+}
+
+FUNCTION {init.bib.lang}
+{
+ "\providecommand \selectlanguage [0]{\@gobble}%" write$ newline$
+}
+
+FUNCTION {init.bib.endbibitem}
+{
+ "\providecommand " bbl.open * "[0]{}%" * write$ newline$
+ "\providecommand \bibitemStop [0]{}%" write$ newline$
+ "\providecommand \bibitemNoStop [0]{.\EOS\space}%" write$ newline$
+ "\providecommand \EOS [0]{\spacefactor3000\relax}%" write$ newline$
+}
+
+FUNCTION {init.bib.translation}
+{
+ "\providecommand \translation [1]{[#1]}%" write$ newline$
+}
+
+FUNCTION {warn.bib}
+{
+}
+
+FUNCTION {init.bib}
+{
+ warn.bib
+ "\makeatletter" write$ newline$
+ "\providecommand \@ifxundefined [1]{%" write$ newline$
+ " \@ifx{#1\undefined}" write$ newline$
+ "}%" write$ newline$
+ "\providecommand \@ifnum [1]{%" write$ newline$
+ " \ifnum #1\expandafter \@firstoftwo" write$ newline$
+ " \else \expandafter \@secondoftwo" write$ newline$
+ " \fi" write$ newline$
+ "}%" write$ newline$
+ "\providecommand \@ifx [1]{%" write$ newline$
+ " \ifx #1\expandafter \@firstoftwo" write$ newline$
+ " \else \expandafter \@secondoftwo" write$ newline$
+ " \fi" write$ newline$
+ "}%" write$ newline$
+ init.bib.ay
+ init.bib.quote
+ init.bib.namefont
+ init.bib.hypertex
+ init.bib.eprint
+ init.bib.doi
+ init.bib.lang
+ init.bib.bibinfo
+ init.bib.translation
+ init.bib.endbibitem
+ "\providecommand " bbl.shut * " [1]{\csname bibitem#1\endcsname}%" * write$ newline$
+ "\let\auto@bib@innerbib\@empty" write$ newline$
+ "%</preamble>" write$
+}
+
+FUNCTION {begin.bib}
+{
+ id.bst diagn.cmntlog
+ control.bib
+ preamble$ empty$
+ 'skip$
+ { preamble$ write$ newline$ }
+ if$
+ "\begin{thebibliography}{"
+ number.label int.to.str$
+ * "}%" *
+ write$ newline$
+ init.bib
+}
+
+EXECUTE {begin.bib}
+
+EXECUTE {init.state.consts}
+
+ITERATE {call.type$}
+
+FUNCTION {end.bib}
+{ newline$
+ "\end{thebibliography}%"
+ write$ newline$
+}
+
+EXECUTE {end.bib}
+
+% \end{macrocode}
+%
+% \begin{macrocode}
+%</aapmrev|aipauth|aipnum|apsrev|apsrmp>
+% \end{macrocode}
+%
+% \endinput
+%
+% \Finale
diff --git a/texmf/source/latex/revtex/revtex4-2.dtx b/texmf/source/latex/revtex/revtex4-2.dtx
new file mode 100644
index 0000000..f6e0b1c
--- /dev/null
+++ b/texmf/source/latex/revtex/revtex4-2.dtx
@@ -0,0 +1,8383 @@
+% \iffalse meta-comment balanced on line 122
+% revtex4-2.dtx: package to typeset APS, AIP, SOR, ASA, etc. journal articles
+% Copyright (c) 2019 American Physical Society.
+% https://journals.aps.org/revtex/
+% mailto:revtex@aps.org
+%
+% Disclaimer
+% This file is distributed WITHOUT ANY WARRANTY;
+% without even the implied warranty of
+% MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
+% License
+% You may distribute this file under the conditions of the
+% LaTeX Project Public License 1.3c or later
+% (http://www.latex-project.org/lppl.txt).
+% ReadMe
+% For the documentation and more detailed instructions for
+% installation, typeset this document with \LaTeX.
+% Maintenance Status
+% This work has the LPPL maintenance status "maintained";
+% Current Maintainer of this work is Arthur Ogawa.
+%
+% This work consists of the main source file revtex4-2.dtx
+% and the derived files
+% revtex4-2.cls, revtex4-2.pdf
+% Distribution:
+% CTAN:macros/latex/contrib/revtex/
+%
+% Unpacking:
+% tex revtex4-2.dtx
+%
+% Documentation:
+% latex revtex4-2.dtx; ...
+%
+% Programm calls to get the documentation (example):
+% pdflatex revtex4-2.dtx
+% makeindex -s gind.ist revtex4-2.idx
+% makeindex -s gglo.ist -o revtex4-2.gls revtex4-2.glo
+% pdflatex revtex4-2.dtx
+% makeindex -s gind.ist revtex4-2.idx
+% pdflatex revtex4-2.dtx
+%
+% Installation:
+% TDS:bibtex/bst/revtex/
+% TDS:doc/latex/revtex/
+% TDS:tex/latex/revtex/
+% TDS:source/latex/revtex/
+%
+% Thanks, Heiko!
+% This method of letting a single .dtx file serve as both
+% documentation (via latex) and installer (via tex) follows
+% the example of Heiko Oberdiek. Thanks!
+%<*ignore>
+\begingroup
+ \def\x{LaTeX2e}%
+\expandafter\endgroup
+\ifcase
+ 0\expandafter\ifx\csname processbatchFile\endcsname\relax\else1\fi\ifx\fmtname\x\else 1\fi
+ \relax
+\else
+ \csname fi\endcsname
+%</ignore>
+%<*install>
+%% This file will generate documentation and runtime files
+%% from revtex4-2.dtx when run through TeX.
+\input docstrip
+\preamble
+
+This file is part of the APS files in the REVTeX 4 distribution.
+For the version number, search on the string %FileInfo
+
+Original version by David Carlisle
+Modified by Arthur Ogawa (mailto:arthur_ogawa at sbcglobal dot net)
+
+Version (4.2a, unreleased)
+Modified by Aptara on behalf of American Physical Society and American Institute of Physics
+
+Version (4.2b,4.2c)
+Modified by Mark Doyle, American Physical Society (mailto:revtex at aps.org)
+
+Version (4.2d,4.2e)
+Modified by Phelype Oleinik for the American Physical Society (mailto:phelype.oleinik at latex-project.org)
+
+Copyright (c) 2019--2020 American Physical Society.
+https://journals.aps.org/revtex/
+mailto:revtex@aps.org
+
+See the REVTeX 4.2 README-REVTEX file for restrictions and more information.
+
+\endpreamble
+\askforoverwritefalse
+\keepsilent
+\generate{%
+ \file{revtex4-2.cls}{%
+ \from{revtex4-2.dtx}{kernel}%
+ \from{ltxutil.dtx}{kernel}%
+ \from{ltxfront.dtx}{kernel}%
+ \from{ltxgrid.dtx}{kernel}%
+ \from{revtex4-2.dtx}{options}%
+ \from{ltxutil.dtx}{options}%
+ \from{ltxfront.dtx}{options}%
+ \from{ltxgrid.dtx}{options}%
+ \from{revtex4-2.dtx}{package}%
+ }%
+ \file{aps4-2.rtx}{\from{revtex4-2.dtx}{aps}}%
+ \file{apsrmp4-2.rtx}{\from{revtex4-2.dtx}{rmp}}%
+ \file{aps10pt4-2.rtx}{\from{revtex4-2.dtx}{10pt}}%
+ \file{aps11pt4-2.rtx}{\from{revtex4-2.dtx}{11pt}}%
+ \file{aps12pt4-2.rtx}{\from{revtex4-2.dtx}{12pt}}%
+ \file{revsymb4-2.sty}{\from{revtex4-2.dtx}{revsymb}}%
+}%
+\ifToplevel{
+\Msg{***********************************************************}
+\Msg{*}
+\Msg{* To finish the installation, please move}
+\Msg{* *.cls, *.rtx, and *.sty }
+\Msg{* into a directory searched by TeX;}
+\Msg{* in a TDS-compliant installation:}
+\Msg{* texmf/tex/macros/latex/revtex/.}
+\Msg{*}
+\Msg{* To produce the documentation,
+ run revtex4-2.dtx through LaTeX.}
+\Msg{*}
+\Msg{* Happy TeXing}
+\Msg{***********************************************************}
+}
+\endbatchfile
+%</install>
+%<*ignore>
+\fi
+%</ignore>
+% \fi
+%
+% \GetFileInfo{revtex4-2.dtx}
+%
+% \StopEventually{}
+% \iffalse ltxdoc klootch
+%<*package>
+%%% @LaTeX-file{
+%%% filename = "revtex4-2.dtx",
+%%% version = "4.2e",
+%%% date = "2020/10/03",
+%%% author = "Mark Doyle (mailto: revtex at aps.org), Arthur Ogawa (mailto:arthur_ogawa at sbcglobal.net),
+%%% commissioned by the American Physical Society.
+%%% ",
+%%% copyright = "Copyright (C) 1999, 2009 Arthur Ogawa, Version 4.2 Copyright (C) 2019 American Physical Society
+%%% distributed under the terms of the
+%%% LaTeX Project Public License 1.3c, see
+%%% ftp://ctan.tug.org/macros/latex/base/lppl.txt
+%%% ",
+%%% address = "Mark Doyle,
+%%% USA",
+%%% telephone = "",
+%%% FAX = "",
+%%% email = "mailto colon revtex at aps.org",
+%%% codetable = "ISO/ASCII",
+%%% keywords = "",
+%%% supported = "yes",
+%%% abstract = "",
+%%% }
+%</package>
+% \fi
+%
+% \iffalse ltxdoc klootch
+% The following references the \file{README-REVTEX.tex} file,
+% which contains the frontmatter for the programmer's docs.
+% The contents of this file are generated when
+% you typeset this file with LaTeX.
+% Search on "{filecontents*}{README-REVTEX.tex}" to locate it.
+% \fi\input{README-REVTEX.tex}%
+%
+% \subsection{Bill of Materials}
+%
+% Following is a list of the files in this distribution arranged
+% according to provenance.
+%
+% \subsubsection{Primary Source}%
+% One single file generates all.
+%\begin{verbatim}
+%revtex4-2.dtx
+%\end{verbatim}
+%
+% \subsubsection{Generated by \texttt{tex revtex4-2.dtx}}%
+% Typesetting ths file under \TeX\ itself runs the installer,
+% which generates the package files.
+%\begin{verbatim}
+%revtex4-2.cls, revtex4.ins, revtex4.drv, aps4-2.rtx,
+%aps10pt4-2.rtx, aps11pt4-2.rtx, aps12pt4-2.rtx, revsymp.sty
+%\end{verbatim}
+%
+% \subsubsection{Generated by \texttt{pdflatex revtex4-2.dtx}}%
+% Typesetting the source file under \LaTeX\
+% generates the documentation.
+%\begin{verbatim}
+%revtex4.pdf,
+%\end{verbatim}
+%
+% \subsubsection{Auxiliary}%
+% The following are auxiliary files generated
+% in the course of running \LaTeX:
+% \begin{verbatim}
+%revtex4.aux revtex4.idx revtex4.ind revtex4.log revtex4.toc
+% \end{verbatim}
+%
+% \section{Code common to all modules}%
+%
+% The following may look a bit klootchy, but we
+% want to require only one place in this file
+% where the version number is stated,
+% and we also want to ensure that the version
+% number is embedded into every generated file.
+%
+% Now we declare that
+% these files can only be used with \LaTeXe.
+% An appropriate message is displayed if
+% a different \TeX{} format is used.
+% \begin{macrocode}
+%<*doc|kernel|aps|rmp|revsymb>
+\NeedsTeXFormat{LaTeX2e}[1996/12/01]%
+%</doc|kernel|aps|rmp|revsymb>
+%<kernel>\ProvidesClass{revtex4-2}
+%<aps>\ProvidesFile{aps4-2}
+%<rmp>\ProvidesFile{apsrmp4-2}
+%<10pt>\ProvidesFile{aps10pt4-2}
+%<11pt>\ProvidesFile{aps11pt4-2}
+%<12pt>\ProvidesFile{aps12pt4-2}
+%<revsymb>\ProvidesPackage{revsymb4-2}
+%<*doc>
+\ProvidesFile{revtex4-2.dtx}
+%</doc>
+%<*!package&!options>
+%<version>
+ [2020/10/03 4.2e (https://journals.aps.org/revtex/ for documentation)]% \fileversion
+%</!package&!options>
+% \end{macrocode}
+% The current class name is remembered in \cmd\class@name.
+% This is something of a klootch, relying as it does on knowledge of the implementation of \cmd\ProvidesPackage.
+% \begin{macrocode}
+%<kernel>\let\class@name\@gtempa
+% \end{macrocode}
+%
+% \section{The driver module \texttt{driver}}
+%
+% This module, consisting of the present section,
+% typesets the programmer's documentation,
+% generating the \file{README-REVTEX.txt} and sample document as needed.
+%
+% Because the only uncommented-out lines of code at the beginning of
+% this file constitute the \file{driver} module itself,
+% we can simply typeset the \file{.dtx} file directly,
+% and there is thus rarely any need to
+% generate the ``driver'' {\sc docstrip} module.
+% Module delimiters are nonetheless required so that
+% this code does not find its way into the other modules.
+%
+% The \enve{document} command concludes the typesetting run.
+%
+% \begin{macrocode}
+%<*doc>
+% \end{macrocode}
+% The driver uses packages \file{ltxdoc.sty}, \file{ltxdocext.sty},
+% \file{hyperref.sty}, and whatever font package has been selected.
+% \begin{macrocode}
+\documentclass{ltxdoc}
+\RequirePackage{ltxdocext}%
+\let\url\undefined
+\RequirePackage[colorlinks=true,linkcolor=blue]{hyperref}%
+% \end{macrocode}
+% We ask for the usual indices and glossaries.
+% \begin{macrocode}
+\CodelineIndex\EnableCrossrefs % makeindex -s gind.ist revtex4
+\RecordChanges % makeindex -s gglo.ist -o revtex4.gls revtex4.glo
+% \end{macrocode}
+%
+% \subsubsection{Docstrip and info directives}
+% We use so many {\sc docstrip} modules that we set the
+% \texttt{StandardModuleDepth} counter to 1.
+% \begin{macrocode}
+\setcounter{StandardModuleDepth}{1}
+% \end{macrocode}
+% The following command retrieves the date and version information
+% from this file.
+% \begin{macrocode}
+\expandafter\GetFileInfo\expandafter{\jobname.dtx}%
+% \end{macrocode}
+%
+% \subsection{The Frontmatter File}
+% As promised above, here is the contents of the frontmatter file.
+% \begin{macrocode}
+\begin{filecontents*}{README-REVTEX.tex}
+\title{%
+ The \classname{revtex4-2} document class of the American Physical Society%
+ \protect\thanks{Work under hire to American Physical Society. Version \fileversion\ \copyright\ 2019 American Physical Society}%
+}%
+\author{Arthur Ogawa and Mark Doyle%
+ \protect\thanks{First revision of REV\TeX4.0 (unreleased) by David Carlisle, all released versions of 4.0 and 4.1 by Art Ogawa, 4.2a (unreleased) by Aptara, 4.2b,c by Mark Doyle}%
+}%
+\date{Version \fileversion, dated \filedate}%
+\newcommand\revtex{REV\TeX}
+
+\maketitle
+
+This file embodies the implementation of the APS \revtex\ 4.2 document class
+for electronic submissions to journals.
+
+The distribution point for this work is
+\url{https://journals.aps.org/revtex/},
+which contains fully unpacked, prebuilt runtime files and documentation.
+
+\tableofcontents
+
+\section{Using \protect\revtex}
+
+The file \file{README} has retrieval and installation information.
+
+User documentation is presented separately in \file{auguide.tex}.
+
+The file \file{template.aps} is a boilerplate file.
+
+\changes{4.0a}{1998/01/16}{Initial version}
+\changes{4.0a}{1998/01/31}{Move after process options, so \cs{clearpage} not in scope of twocolumn}
+\changes{4.0a}{1998/01/31}{Rearrange the ordering so numerical ones come first. AO: David, what does this mean?}
+\changes{4.0a}{1998/01/31}{use font-dependent spacing}
+\changes{4.0a}{1998/01/31}{4.0d had twoside option setting twoside switch to false}
+\changes{4.0a}{1998/01/31}{Move after process options, so the following test works}
+\changes{4.0a}{1998/01/31}{print homepage}
+\changes{4.0a}{1998/01/31}{protect against hyperref revtex kludges which are not needed now}
+\changes{4.0a}{1998/06/10}{multiple preprint commands}
+\changes{4.0a}{1998/06/10}{comma not space between email and homepage}
+\changes{4.0a}{1998/06/10}{single space footnotes}
+\changes{4.0b}{1999/06/20}{First modifications by Arthur Ogawa (mailto:arthur\_ogawa at sbcglobal dot net)}
+\changes{4.0b}{1999/06/20}{Added localization of \cs{figuresname}}
+\changes{4.0b}{1999/06/20}{Added localization of \cs{tablesname}}
+\changes{4.0b}{1999/06/20}{AO: all code for \protect\classoption{10pt} is in this module.}
+\changes{4.0b}{1999/06/20}{AO: all code for \protect\classoption{11pt} is in this module.}
+\changes{4.0b}{1999/06/20}{AO: all code for \protect\classoption{12pt} is in this module.}
+\changes{4.0b}{1999/06/20}{AO: made aps.rtx part of revtex4.dtx}
+\changes{4.0b}{1999/06/20}{AO: remove duplicates}
+\changes{4.0b}{1999/06/20}{call \cs{print@floats}}
+\changes{4.0b}{1999/06/20}{Defer assignment until \cs{AtBeginDocument} time.}
+\changes{4.0b}{1999/06/20}{Defer decision until \cs{AtBeginDocument} time}
+\changes{4.0b}{1999/06/20}{Define three separate environments, defer assignment to \cs{AtBeginDocument} time.}
+\changes{4.0b}{1999/06/20}{Frank Mittelbach, has stated in \protect\classname{multicol}: ``The kernel command \cs{@footnotetext} should not be modified.'' Thus, I have removed David Carlisle's redefinition of that command. Note, however, that later versions of \protect\classname{multicol} do not require this workaround. Belt and suspenders.}%
+\changes{4.0b}{1999/06/20}{Move this ``complex'' option to the front, where it can be overridden by ``simple'' options.}
+\changes{4.0b}{1999/06/20}{New option}
+\changes{4.0b}{1999/06/20}{One-line caption sets flush left.}
+\changes{4.0b}{1999/06/20}{only execute if appropriate}
+\changes{4.0b}{1999/06/20}{Processing delayed to \cs{AtBeginDocument} time}
+\changes{4.0b}{1999/06/20}{Removed invocation of nonexistent class option \protect\classoption{groupauthors} and all other class options that should only be invoked by the document. (Otherwise precedence of class options does not work.)}
+\changes{4.0b}{1999/06/20}{Restore all media size class option of \protect\file{classes.dtx}}
+\changes{4.0b}{1999/06/20}{Stack \cs{preprint} args flush right at right margin.}
+\changes{4.0c}{1999/11/13}{(AO, 115) If three or more preprints specified, set on single line, with commas.}
+\changes{4.0c}{1999/11/13}{(AO, 129) section* within appendix was producing appendixname}
+\changes{4.0c}{1999/11/13}{*-form mandates pagebreak}
+\changes{4.0c}{1999/11/13}{also spelled ``acknowledgements''.}
+\changes{4.0c}{1999/11/13}{Do not put by REVTeX in every page foot}
+\changes{4.0c}{1999/11/13}{grid changes via ltxgrid procedures}
+\changes{4.0c}{1999/11/13}{grid changes with ltxgrid}
+\changes{4.0c}{1999/11/13}{Insert procedure \cs{checkindate}}
+\changes{4.0c}{1999/11/13}{Lose compatability mode.}
+\changes{4.0c}{1999/11/13}{New ltxgrid-based code, other bug fixes}
+\changes{4.0c}{1999/11/13}{New option ``checkin''}
+\changes{4.0c}{1999/11/13}{Prevent an inner footnote from performing twice}
+\changes{4.0d}{2000/04/10}{Also alter how lists get indented.}
+\changes{4.0d}{2000/04/10}{eprint takes an optional argument, syntactical only in this case.}
+\changes{4.0d}{2000/04/10}{New option}
+\changes{4.0d}{2000/05/10}{More features and bug fixes: compatability with longtable and array packages. Now certainly incompatible with multicol.}
+\changes{4.0d}{2000/05/17}{make longtable trigger the head, too}
+\changes{4.0d}{2000/05/18}{But alternative spelling is deprecated.}
+\changes{4.0e}{2000/09/20}{New option showkeys}
+\changes{4.0e}{2000/11/14}{Bug fixes and minor new features: title block affiliations can have ancillary data, just like authors; clearpage processing revamped, with floats staying in order; widetext ornaments.}
+\changes{4.0e}{2000/11/21}{adornments above and below.}
+\changes{4.0f}{2001/02/13}{Last bug fixes before release.}
+\changes{4.0rc1}{2001/06/17}{Running headers always as if two-sided}
+\changes{4.0rc1}{2001/06/18}{grid changes with push and pop}
+\changes{4.0rc1}{2001/06/18}{grid changes with push and pop}
+\changes{4.0rc4}{2001/07/23}{hyperref is no longer loaded via class option: use a usepackage statement instead}
+\changes{4.1a}{2008/01/18}{(AO, 457) Endnotes to be sorted in with numerical citations.}%
+\changes{4.1a}{2008/01/18}{(AO, 451) ``Cannot have more than 256 cites in a document''}%
+\changes{4.1a}{2008/01/18}{(AO, 457) Endnotes to be sorted in with numerical citations.}%
+\changes{4.1a}{2008/01/18}{(AO, 460) ``Proper style is "FIG. 1. ..." (no colon)''}%
+\changes{4.1a}{2008/01/18}{(AO, 478) \cs{ds@letterpaper}, so that ``letterpaper really is the default''}%
+\changes{4.1a}{2008/01/18}{(AO, 488) Change processing of options to allow an unused option to specify society and journal}%
+\changes{4.1a}{2008/01/19}{(AO, 461) Change the csname revtex uses from @dotsep to ltxu@dotsep. The former is understood in mu. (What we wanted was a dimension.)}%
+\changes{4.1a}{2008/01/19}{For natbib versions before 8.21, \cs{NAT@sort} was consulted only as natbib was being read in. Now it is fully dynamic.}
+\changes{4.1b}{2008/05/29}{The csname substyle@ext is now defined without a dot (.), to be compatible with \LaTeX usage (see @clsextension and @pkgextension).}
+\changes{4.1b}{2008/06/01}{(AO) Implement bibnotes through \cs{frontmatter@footnote@produce} instead of \cs{bibnotes@sw}}%
+\changes{4.1b}{2008/06/01}{Add option reprint, opposite of preprint, and preferred alternative to twocolumn}
+\changes{4.1b}{2008/06/29}{(AO, 455) Be nice to a list within the abstract (assign \cs{@totalleftmargin}).}
+\changes{4.1b}{2008/06/30}{(AO) Structure the Abstract using the \texttt{bibliography} environment}
+\changes{4.1b}{2008/07/01}{(AO) coordinate \cs{if@twoside} with \cs{twoside@sw}}
+\changes{4.1b}{2008/07/01}{(AO) make settings at class time instead of deferring them to later.}
+\changes{4.1b}{2008/07/01}{(AO) No longer need to test \cs{chapter} as of \texttt{natbib} version 8.2}
+\changes{4.1b}{2008/07/01}{(AO) No longer use \cs{secnumarabic@sw}, instead use \cs{setup@secnums}}
+\changes{4.1b}{2008/07/01}{(AO) Provide more diagnostics when \cs{@society} is assigned.}
+\changes{4.1b}{2008/07/01}{(AO) provide option longbibliography}
+\changes{4.1b}{2008/07/01}{Add \cs{@hangfroms@section}}
+\changes{4.1b}{2008/07/01}{Break out \cs{@caption@fignum@sep}}
+\changes{4.1b}{2008/07/01}{Class option galley sets \cs{preprintsty@sw} to false}
+\changes{4.1b}{2008/07/01}{Code relating to new syntax for frontmatter has been placed in \file{ltxfront.dtx}}
+\changes{4.1b}{2008/07/01}{Package textcase is now simply a required package}
+\changes{4.1b}{2008/07/01}{Procedures \cs{@parse@class@options@society} and \cs{@parse@class@options@journal} and friends}
+\changes{4.1b}{2008/07/01}{Read in all required packages together}
+\changes{4.1b}{2008/07/01}{Remove options newabstract and oldabstract}
+\changes{4.1b}{2008/08/01}{Section numbering via procedures \cs{secnums@rtx} and \cs{secnums@arabic}.}
+\changes{4.1b}{2008/08/04}{As with author formatting, rag the right more, and assign \cs{@totalleftmargin}. Also neutralize \cs{def@after@address}.}%
+\changes{4.1b}{2008/08/04}{Rag the right even more: .8\cs{hsize}. Also, assign \cs{@totalleftmargin}.}%
+\changes{4.1b}{2008/08/04}{The \texttt{rmp} journal substyle selects \texttt{groupedaddress} by default.}%
+\changes{4.1b}{2008/08/04}{Use \cs{setup@hook} to initialize all.}
+\changes{4.1c}{2008/08/15}{Document class option longbibliography via \cs{substyle@post}}
+\changes{4.1d}{2009/03/27}{Definition of \cs{ @fnsymbol} follows fixltx2e.sty}
+\changes{4.1e}{2008/06/29}{(AO, 455) be nice to a list within the abstract}
+\changes{4.1f}{2009/07/07}{(AO, 513) Add class option linenumbers: number the lines a la \classname{lineno}}
+\changes{4.1f}{2009/07/07}{(AO, 516) Merged references are separated with a semicolon}
+\changes{4.1f}{2009/07/10}{(AO, 520) Automatically produce \cs{bibliography} command when needed}%
+\changes{4.1f}{2009/07/11}{(AO, 521) Lonely bibliography head}%
+\changes{4.1f}{2009/07/11}{(AO, 522) Warn if software is expired}%
+\changes{4.1f}{2009/07/15}{(AO, 523) Add class option nomerge, to turn off new natbib 8.3 syntax}
+\changes{4.1f}{2009/07/20}{(AO, 524) Makes no sense if citations are superscript numbers and so are footnotes}
+\changes{4.1f}{2009/10/05}{(AO, 530) \cs{@fnsymbol}: Failed to import fixltx2e.sty technology. Return to LaTeX core.}
+\changes{4.1g}{2009/10/07}{(AO, 525) Remove phantom paragraph above display math that is given in vertical mode}%
+\changes{4.1g}{2009/10/07}{(AO, 538) \cs{MakeTextUppercase} inappropriately expands the double backslash}
+\changes{4.1h}{2009/10/09}{(AO) Remove expiry code in the release software}%
+\changes{4.1i}{2009/10/23}{(AO, 541) Defer assignment of \cs{cite} until after natbib loads}
+\changes{4.1j}{2009/10/24}{(AO, 549) Repairing natbib's \cs{BibitemShut} and \cs{bibAnnote}}
+\changes{4.1j}{2009/10/25}{(AO, 545) hypertext capabilities off by default; enable with \classoption{hypertext}}
+\changes{4.1j}{2009/10/25}{(AO, 552) Repair spacing in \cs{onlinecite}}
+\changes{4.1k}{2009/11/06}{(AO, 554) give the \cs{newlabel} command syntax appropriate to the hyperref package}
+\changes{4.1n}{2009/11/06}{(AO, 565) restore 4.0 behavior: invoking class option preprint implies class option preprintnumbers}
+\changes{4.1n}{2009/11/30}{(AO, 566) restore 4.0 behavior: flush column bottoms}
+\changes{4.1n}{2009/12/05}{(AO, 569) Use of \classname{hyperref} interferes with column balancing of last page}%
+\changes{4.1n}{2009/12/09}{(AO, 569) execute the after-last-shipout procedures from within the safety of the output routine}%
+\changes{4.1n}{2010/01/02}{(AO, 571) Interface \cs{set@footnotewidth} for determining the set width of footnotes}%
+\changes{4.1n}{2010/01/02}{(AO, 572) Independent footnote counter for title block. Abstract footnote counter shared with body.}%
+\changes{4.1n}{2009/12/13}{(AO, 573) arrange to load \classname{lineno} after any other packages.}%
+\changes{4.1n}{2010/01/04}{(AO, 575) the default for journal prstper is longbibliography}%
+\changes{4.1n}{2010/01/04}{(AO, 576) In .bst files, remove support for the annote field}%
+\changes{4.1n}{2010/01/02}{(AO) fine-tune spacing above and below widetext}%
+\changes{4.1n}{2010/01/02}{(AO, 571) class file must set \cs{splittopskip}; fine tune \cs{skip}\cs{footins}; \cs{footnoterule} defined in terms of \cs{skip}\cs{footins}}%
+\changes{4.1n}{2010/01/02}{(AO, 572) \cs{@makefntext} and \cs{frontmatter@makefntext} must be defined harmoniously}%
+\changes{4.1o}{2010/02/02}{(AO, 575) Automatically incorporate the (Bib\TeX-generated) .bbl into an explicit \env{thebibliography}}%
+\changes{4.1o}{2010/02/05}{(AO, 549) Remove patch to natbib, which is now at version 8.31a}
+\changes{4.1o}{2010/02/07}{(AO, 578) accommodate the possible space character preceding \cs{BibitemShut}.}
+\changes{4.1o}{2010/02/05}{(AO, 579) Endnote shall comprise their own Bib\TeX\ entry type: @FOOTNOTE.}
+\changes{4.1o}{2010/02/10}{(AO, 580) Provide a document class option to turn off production of eprint field in bibliography.}
+\changes{4.1o}{2010/02/12}{(AO, 580) Control .bst at run time.}%
+\changes{4.1o}{2010/02/09}{(AO, 581) Handle case: merged references, with first ending in a stop character.}
+\changes{4.1p}{2010/02/24}{(AO, 583) Provide interface to \classname{ltxgrid} \cs{onecolumn@grid@setup} and \cs{twocolumn@grid@setup}}
+\changes{4.1p}{2010/02/24}{(AO, 584) Per MD, remove trailing space character from each journal abbreviation: it had caused an extraneous space in the .bbl}
+\changes{4.1q}{2010/04/01}{(AO, 586) When .bbl is pasted into the document, prevent automatic bibliography inclusion.}%
+\changes{4.1q}{2010/04/13}{(AO, 588) Only write \revtex-specific BibTeX .bib data if the .bst style is set by REVTeX.}%
+\changes{4.1r}{2010/06/22}{(AO, 595) Provide \cs{lovname} along with other List of Videos definitions.}%
+\changes{4.2a}{2014/12/31}{(Aptara, MD) Added initial support for SOR and AAPM journals, additional journals for APS, and additional journals and proceedings for AIP, unreleased.}%
+\changes{4.2a}{2014/12/31}{(Aptara) Make prb style to follow other Phys. Rev. journals.}%
+\changes{4.2a}{2014/12/31}{(Aptara) Corrected indentation for tableofcontents appearing along with listoffigure/listoftable.}%
+\changes{4.2a}{2017/11/21}{(MD) Make long bibliography style the default now.}%
+\changes{4.2a}{2017/11/28}{(MD) Add call to normalsize to be a good citizen and allow booktabs.sty to work properly}.%
+\changes{4.2b}{2018/12/26}{(MD) Make titles in bibliography default, prb style to follow other Phys. Rev. journals, add a unified physrev option as well as prx and prapplied options. Corrected indentation for tableofcontents appearing along with listoffigure/listoftable.}%
+\changes{4.2b}{2017/11/21}{(MD) Update options for new titles without "Special Topics" and make prper match style of other journal options}%
+\changes{4.2b}{2017/11/21}{(MD) Add options for new APS journals and a generic physrev option for future-proofing}%
+\changes{4.2b}{2017/11/22}{(MD) Change default to not use a title page - it seems antiquated}%
+\changes{4.2b}{2017/11/22}{(MD) MD - not sure why these parameters were different previously. Made them match except for title.}%
+\changes{4.2b}{2017/11/22}{(MD) PACS are obsolete altogether now}%
+\changes{4.2b}{2018/12/26}{(MD) Improve control over display of e-print ids in bibliography.}%
+
+\end{filecontents*}
+% \end{macrocode}
+%
+% \subsection{The Document Body}
+%
+% Here is the document body, containing only a
+% \cmd\DocInput\ directive---referring to this very file.
+% This very cute self-reference is a common \classname{ltxdoc} idiom.
+% \begin{macrocode}
+%%\newcommand\revtex{REV\TeX}% TeXSupport
+\begin{document}%
+\expandafter\DocInput\expandafter{\jobname.dtx}%
+\PrintChanges
+\end{document}
+% \end{macrocode}
+% And that is the end of the driver for the programmer's documentation.
+% \begin{macrocode}
+%</doc>
+% \end{macrocode}
+%
+% \section{Overview}%
+% \revtex\ is a \LaTeXe\ document class, somewhat like a hybrid of
+% the standard \LaTeX\ \classname{book} and \classname{article} classes.
+%
+% Certain packages are (should be) loaded by this class in any case:
+% \classname{amssymb}, \classname{amsmath}, \classname{bm}, \classname{natbib}.
+%
+% Certain packages are automatically loaded by this class
+% when a corresponding class option has been invoked:
+%
+% \begin{tabular}{ll}
+% \textbf{\revtex\ option}&\textbf{package}\\
+% \classoption{amsfonts}&\classname{amsfonts}\\
+% \classoption{amssymb}&\classname{amssymb}\\
+% \classoption{aps}&\classname{overcite}\\
+% \end{tabular}
+%
+% Certain other packages are to be loaded by the
+% document through explicit use of \cmd\usepackage.
+% Some mentioned in the user documentation are
+% \classname{graphicx}, \classname{longtable}, \classname{hyperref},
+% and \classname{bm}.
+%
+% Certain commonly used packages are known to be incompatible with
+% \revtex, among them \classname{multicol} and \classname{cite}.
+% If such a package is found to be loaded, \revtex\ issues an error message
+% and halts the job.
+% Halting might be considered severe punishment for loading an incompatible
+% package, but if we were to proceed, an even weirder error might be encountered
+% further down the road.
+%
+% This document class implements the substyle: a set of mutually
+% exclusive class options that, in this case, allow the
+% document class to address multiple societies.
+% It also implements a sub-substyle, giving the
+% journal of the given society.
+% Any society may create a substyle;
+% this file generates one for \classoption{aps}.
+%
+% FIXME: should always load the \classname{graphicx} package. No, allow user to load whichever graphics package is desired.
+%
+% QUERY: since \classname{amsfonts} and \classname{amssymb} extend
+% syntax, why not load them in any case?
+%
+% Certain processing occurs at the endgame for reading in \revtex, thereby establishing
+% precedence for assignments to \LaTeX's (and \revtex's) parameters:
+% \begin{enumerate}
+% \item
+% Figure out which society is operative and read in the indicated \file{.rtx} file.
+% \item
+% Figure out which journal option is operative and execute the indicated journal command. This may lead to reading in a journal substyle \file{.rtx} file.
+% \item
+% Figure out which pointsize is operative and execute the indicateded pointsize command. This may lead to reading in a pointsize \file{.rtx} file.
+% \item
+% Execute all of the document's options, in the order declared within the document.
+% \item
+% Read in all required packages (like \classname{natbib}, \classname{amsfonts}, \classname{amssymb}), that were determined by class, society, and journal.
+% \item
+% The last required package, if existing, is the document's style file, the \file{.rty} file.
+% Note that the \file{.rty} file can override the assignments of \revtex, society, journal, and required packages, and even load its own packages.
+% It can also, via appendations to \cmd\setup@hook, override the setup code itself.
+% \item
+% Execute all of the setup code accumulated. Such code can be queued by \revtex\ itself, by the society, by the journal, or by the pointsize.
+% \item
+% At this point, \revtex\ has completed the process of inputting itself,
+% and \LaTeX\ will now execute the \cmd\AtEndOfClass\ procedures.
+% \end{enumerate}
+%
+% \revtex\ will have enqueued code to execute at \cmd\AtBeginDocument\ time,
+% in two different queues.
+% \cmd\document@inithook\ executes immediately upon encountering the \cmd\begin\env{document} statement,
+% \cmd\class@documenthook\ at the end of all the code enqueued via \cmd\AtBeginDocument.
+% \begin{enumerate}
+% \item
+% Install procedures to execute at the very end of the class's
+% \cmd\AtBeginDocument\ processing, such as
+% \begin{enumerate}
+% \item
+% closing out the page grid
+% \item
+% putting out the LastPage label.
+% \item
+% issuing a \cmd\bibliographystyle\ command, based on the value set
+% by the society substyle.
+% \item
+% setting default values for parameters used in the document.
+%FIXME: differentiate between class's parameters that can wait until
+% they are used in the document, and parameters that are used at \cmd\AtBeginDocument\ time.
+% \end{enumerate}
+% \item
+% Install procedures to execute the very last at \cmd\AtEndDocument\ time,
+% such as the \cmd\clearpage\ processing.
+% \end{enumerate}
+%
+% Certain events are optionally scheduled for \cmd\AtBeginDocument\ time:
+% \begin{enumerate}
+% \item
+% Setting default values for the Booleans and for other procedures used in formatting.
+% \item
+% In response to class options options, adjusting parameters and procedures used in formatting.
+% \item
+% Implementing the \classoption{eqsecnum} option, if required.
+% \item
+% Setting the state engine for data commands.
+% \item
+% Memorizing procedures for later use.
+% \item
+% Setting type size and area, for use by later calculations.
+% \end{enumerate}
+%
+% Certain events are scheduled for \cmd\class@enddocumenthook\ time:
+% \begin{enumerate}
+% \item
+% Print out the migrated floats or the end notes, if needed.
+% \item
+% Close out the page grid.
+% \item
+% Label the last page of the document
+% \item
+% (\classname{natbib}) prepare to read in the \file{.aux} file.
+% \end{enumerate}
+%
+% \section{Writing journal-specifc extensions to \revtex}
+%
+% With this version of \revtex, we introduce a somewhat different
+% scheme for adapting \revtex\ to the needs of a specific journal.
+%
+% To create a journal substyle, you create new class options in
+% \revtex\ for the society, say \classoption{osa}, and any of that society's journals,
+% one of which is, say, \classoption{josaa},
+% using the code for the APS as a guide.
+% In particular, each of your new options should separately define \cmd\@society\
+% and \cmd\@journal. That for the former will be the same for
+% all options relating to a particular society.
+%
+% Then, for the society, you create a corresponding \file{.rtx}
+% file, in our case \file{osa.rtx}.
+% Within that file, you override procedures and parameter assignments as you see fit.
+% Ideally they will be generally applicable to all of that society's
+% journals (see the file \file{aps.rtx} for a realization of this scheme).
+% Also within that file, you include a
+% section of code for each journal, that for josaa looks like:
+% \begin{verbatim}
+% \@ifx{\@journal\journal@josaa}{%
+% <code specific to the josaa>
+% }{}%
+% \end{verbatim}
+%
+% Thus far, the scheme is similar to that used in \revtex\ 3.1.
+% However, the new scheme does differ from the old in that
+% the \file{.rtx} file should define no syntactical extensions
+% to \revtex.
+%
+% \section{The \classname{revtex4} Document Class}
+% Above, we took
+% advantage of 1) the \LaTeX\ definition of \cmd\ProvidesPackage\ and
+% 2) that the line of code immediately afterwards follows the \cmd\ProvidesClass\ statement above.
+% \begin{macrocode}
+%<*kernel>
+% \end{macrocode}
+%
+% Print a banner in the log:
+% \begin{macrocode}
+\GenericInfo{}{\space
+ Copyright (c) 2019 American Physical Society.^^J
+ mailto:revtex@aps.org^^J
+ Licensed under the LPPL:^^Jhttp://www.ctan.org/tex-archive/macros/latex/base/lppl.txt^^J
+ Arthur Ogawa <arthur_ogawa at sbcglobal dot net>^^J
+ Based on work by David Carlisle <david at dcarlisle.demon.co.uk>^^J
+ Version (4.2d,4.2e): Modified by Mark Doyle and Phelype Oleinik^^J
+ \@gobble
+}%
+% \end{macrocode}
+%
+% \subsection{Compatibility Processing}
+%
+% If the document has \cmd\documentstyle|{revtex4}|,
+% then, instead of attempting to run in compatability mode,
+% just complain and exit.
+% \changes{4.0c}{1999/11/13}{Lose compatability mode.}
+% \begin{macrocode}
+\if@compatibility
+ \edef\reserved@a{\errhelp{%
+ Change your \string\documentstyle\space statement to
+ \string\documentclass\space and rerun.
+ }}\reserved@a
+ \errmessage{You cannot run \class@name\space in compatability mode}%
+ \expandafter\@@end
+\fi
+% \end{macrocode}
+%
+%
+% \section{Extensions to the \LaTeX\ Kernel}
+%
+% \begin{macrocode}
+%</kernel>
+% \end{macrocode}
+%
+% Here, we incorporate the utility, frontmatter, and page grid packages.
+% The \classname{ltxutil}, \classname{ltxfront}, and \classname{ltxgrid} source are distributed with
+% \revtex.
+%
+% Here begins the \file{options} \textsc{docstrip} module.
+% \begin{macrocode}
+%<*options>
+% \end{macrocode}
+%
+% \subsection{Hooks}
+%
+% \begin{macro}{\setup@hook}
+% The procedure
+% \cmd\setup@hook\
+% serves as the vehicle for all code that
+% gives values to the class's parameters once all the
+% society, journal, options, and packages have been processed.
+%
+% Arrange for journal substyles to set their own default values.
+% \begin{macrocode}
+\let\setup@hook\@empty
+% \end{macrocode}
+% \end{macro}
+%
+% After preamble processing is complete,
+% detect whether package \classname{longtable} has been loaded and patch it.
+% \begin{macrocode}
+\appdef\document@inithook{%
+ \switch@longtable
+ \let\LT@makecaption\LT@makecaption@rtx
+}%
+% \end{macrocode}
+%
+% \begin{macro}{\LT@makecaption}
+% We override the caption processing method of the \classname{longtable} package:
+% space below the caption is created via strut instead of whitespace.
+% \begin{macrocode}
+\def\LT@makecaption@rtx#1#2#3{%
+ \LT@mcol\LT@cols c{%
+ \hbox to\z@{%
+ \hss
+ \parbox[t]\LTcapwidth{%
+ \sbox\@tempboxa{#1{#2: }#3\unskip\nobreak\vrule\@width\z@\@height\z@\@depth .5\baselineskip}%
+ \ifdim\wd\@tempboxa>\hsize
+ #1{#2: }#3\unskip\nobreak\vrule\@width\z@\@height\z@\@depth .5\baselineskip
+ \else
+ \hbox to\hsize{\hfil\box\@tempboxa\hfil}%
+ \fi
+ \endgraf
+ }%
+ \hss
+ }%
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\robust@boldmath}
+% Robustify the \cmd\boldmath command.
+% If Team \LaTeX\ (or any package) ever gets around to fixing this problem,
+% we will bow out. (This fix relates to bug \#394.)
+% \begin{macrocode}
+\def\protectdef@boldmath{%
+ \expandafter\@ifnotrelax\csname boldmath \endcsname{}{%
+ \class@info{Robustifying \string\LaTeX's \string\boldmath\space command}%
+ \expandafter\let\csname boldmath \endcsname\boldmath
+ \expandafter\def\expandafter\boldmath\expandafter{\expandafter\protect\csname boldmath \endcsname}%
+ }%
+}%
+\appdef\document@inithook{%
+ \protectdef@boldmath
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \section{Compatability with the geometry package}
+%
+% The \classname{geometry} package of Hideo Umeki provides a way to
+% specify the metrics of the media and page layout.
+% We want to ensure that \revtex\ does not foreclose on the use of
+% this or any other such package, nor to interfere with
+% explicit assignments of such metric parameters within the
+% document preamble.
+%
+% The \LaTeX\ parameters (resp. \TeX\ primitives) set by \classname{geometry} are:
+% \cmd\paperwidth, \cmd\paperheight,
+% \cmd\textwidth, \cmd\textheight,
+% \cmd\topmargin, \cmd\headheight, \cmd\headsep,
+% \cmd\footskip, \cmd\skip\cmd\footins,
+% \cmd\marginparwidth, \cmd\marginparsep,
+% \cmd\oddsidemargin, \cmd\evensidemargin,
+% \cmd\columnsep,
+% \cmd\hoffset, \cmd\voffset, \cmd\mag,
+% \cmd\if@twocolumn, \cmd\if@twoside, \cmd\if@mparswitch, \cmd\if@reversemargin,
+%
+% \revtex's assignments can be overridden by
+% any package or other statement in the document preamble,
+% so it should be compatible.
+%
+%FIXME: one parameter is rendered semantically void (by \file{ltxgrid.dtx}):
+% \cmd\if@twocolumn.
+%
+% \section{Options}\label{sec:options}%
+%
+% \subsection{Define Booleans Used in Options}%
+% The following Booleans are used within the document class to
+% allow the document or the substyle to make selections
+% of formatting.
+% An explicit document class option overrides an assignment made by a substyle.
+%
+% \subsection{Declare Options}%
+%
+% \subsubsection{Checkin: for Editorial Use}%
+% A document class option declaring that the document is being processed by the
+% editorial staff.
+%
+% This option should:
+% \begin{itemize}
+% \item
+% put date in footer along with folio
+% \item
+% Have the effect of selecting the \classoption{preprint} class option.
+% \item
+% Have the effect of selecting the \classoption{showpacs} class option.
+% \item
+% specify that when a float is placed h or H, it will be allowed to break over pages.
+% (Note: be sure that if the enclosed \env{tabular} has an optional argument, you change it to [v], or remove it entirely.)
+% \end{itemize}
+% \changes{4.0c}{1999/11/13}{New option ``checkin''}
+% \begin{macrocode}
+\DeclareOption{checkin}{%
+ \@booleantrue\dateinRH@sw
+ \@booleantrue\preprintsty@sw
+ \def\@pointsize{12}%
+ \@booleantrue\showPACS@sw
+ \@booleantrue\showKEYS@sw
+ \def\fp@proc@h{\allow@breaking@tables}%
+ \def\fp@proc@H{\allow@breaking@tables}%
+}%
+\@booleanfalse\dateinRH@sw
+\def\checkindate{\dateinRH@sw{{\tiny(\today)}}{}}%
+\def\allow@breaking@tables{%
+ \def\array@default{v}% tabular can break over pages
+ \@booleanfalse\floats@sw % table can break over pages
+}%
+% \end{macrocode}
+%
+% \subsubsection{Preprint Style}%
+%
+% \begin{macro}{\preprintsty@sw}
+% The boolean \cmd\preprintsty@sw\ signifies that
+% the document is to be formatted in preprint style.
+% \changes{4.0a}{1998/01/31}{Rearrange the ordering so numerical ones come first. AO: David, what does this mean?}
+% \changes{4.0a}{1998/06/10}{comma not space between email and homepage}
+% \changes{4.0b}{1999/06/20}{Move this ``complex'' option to the front, where it can be overridden by ``simple'' options.}
+% \changes{4.1b}{2008/06/01}{Add option reprint, opposite of preprint, and preferred alternative to twocolumn}
+% \begin{macrocode}
+\DeclareOption{preprint}{%
+ \@booleantrue\preprintsty@sw
+ \ExecuteOptions{12pt}%
+}%
+\DeclareOption{reprint}{%
+\@booleanfalse\preprintsty@sw
+\@booleantrue\twocolumn@sw
+ \ExecuteOptions{10pt}%
+}%
+\DeclareOption{manuscript}{%
+ \class@warn{Document class option manuscript is obsolete; use preprint instead}%
+ \ExecuteOptions{preprint}%
+}%
+\@booleanfalse\preprintsty@sw
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\showPACS@sw}
+% \begin{macro}{\showKEYS@sw}
+% If \cmd\showPACS@sw\ is true, print the PACS information in the title block,
+% otherwise not.
+% Similarly for \cmd\showKEYS@sw\ and the keywords.
+% \subsubsection{Showing PACS and keywords}%
+% \changes{4.0b}{1999/06/20}{New option}
+% \changes{4.0e}{2000/09/20}{New option showkeys}
+% \begin{macrocode}
+\DeclareOption{showpacs}{%
+ \@booleantrue\showPACS@sw
+}%
+\DeclareOption{noshowpacs}{%
+ \@booleanfalse\showPACS@sw
+}%
+\DeclareOption{showkeys}{%
+ \@booleantrue\showKEYS@sw
+}%
+\DeclareOption{noshowkeys}{%
+ \@booleanfalse\showKEYS@sw
+}%
+\@booleanfalse\showPACS@sw
+\@booleanfalse\showKEYS@sw
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \subsubsection{Balance the last page when in two-column page grid}%
+%
+% \begin{macro}{\balancelastpage@sw}
+% \changes{4.1b}{2008/07/01}{(AO) make settings at class time instead of deferring them to later.}
+% If we are in a two-column page grid, we may wish to balance the columns
+% of the last page. This will be done automatically if the
+% \classoption{twocolumn} document class option is chosen.
+% This action will be turned off by the \classoption{nobalancelastpage}.
+% A complementary class option, \classoption{balancelastpage} is also provided.
+% \begin{macrocode}
+\DeclareOption{balancelastpage}{%
+ \@booleantrue\balancelastpage@sw
+}%
+\DeclareOption{nobalancelastpage}{%
+ \@booleanfalse\balancelastpage@sw
+}%
+\@booleantrue\balancelastpage@sw
+% \end{macrocode}
+% \end{macro}
+%
+% \subsubsection{Showing preprint numbers}%
+%
+% \begin{macro}{\preprint@sw}
+% \changes{4.0d}{2000/04/10}{New option}
+% \changes{4.1b}{2008/07/01}{(AO) make settings at class time instead of deferring them to later.}
+% \changes{4.1n}{2009/11/06}{(AO, 565) restore 4.0 behavior: default value of \cs{preprint@sw} follows \cs{preprintsty@sw}}
+% The boolean \cmd\preprint@sw\ signifies that
+% the preprints (cf. \cmd\preprint) are to be formatted (usually on the title page).
+% The option \classoption{preprintnumbers} declares to do so,
+% \classoption{nopreprintnumbers} declares not to;
+% the default is to follow \cmd\preprintsty@sw.
+% \begin{macrocode}
+\DeclareOption{nopreprintnumbers}{\@booleanfalse\preprint@sw}%
+\DeclareOption{preprintnumbers}{\@booleantrue\preprint@sw}%
+\appdef\setup@hook{%
+ \@ifxundefined\preprint@sw{\let\preprint@sw\preprintsty@sw}{}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \subsubsection{Hypertext Option}%
+%
+% \changes{4.0a}{1998/01/31}{protect against hyperref revtex kludges which are not needed now}
+% \changes{4.0b}{1999/06/20}{Move this ``complex'' option to the front, where it can be overridden by ``simple'' options.}
+% \changes{4.0rc4}{2001/07/23}{hyperref is no longer loaded via class option: use a usepackage statement instead}
+% \begin{macrocode}
+\DeclareOption{hyperref}{%
+% \end{macrocode}
+% The following code had been commented out, it is now truly a comment:
+% \begin{verbatim}
+%\AtEndOfClass{%
+% \begingroup
+% \edef\@tempa{%
+% \let
+% \noexpand\@clsextension
+% \noexpand\@empty
+% \noexpand\RequirePackage{hyperref}%
+% \def\noexpand
+% \@clsextension{\@clsextension}%
+% }%
+% \expandafter
+% \endgroup
+% \@tempa
+%}%
+% \def\@pointsize{10}%
+% \end{verbatim}
+%
+% If you have a hyper-foo enabled browser you may prefer this format
+% which does not print the URL for the home page, but just makes the
+% name a link, but by default print it so it works on paper.
+% \changes{4.0a}{1998/01/31}{print homepage}
+%
+% \begin{macrocode}
+ \class@warn{Class option "hyperref" is no longer supported.^^JEmploy \string\usepackage{hyperref} instead}%
+}%
+% \end{macrocode}
+%
+% \subsubsection{Type Size}%
+% \changes{4.1b}{2008/07/01}{(AO) make settings at class time instead of deferring them to later.}
+% Use \cmd\@pointsize|=10| rather than \cmd\@ptsize|=0| to allow easy extensions
+% to 9pt or whatever.
+% Note: the three alternatives are mutually exclusive.
+%
+% At this point, the parameter \cmd\@pointsize\ is set to \cmd\undefined:
+% a society \emph{must} give it a definition.
+% \begin{macrocode}
+\DeclareOption{10pt}{\def\@pointsize{10}}%
+\DeclareOption{11pt}{\def\@pointsize{11}}%
+\DeclareOption{12pt}{\def\@pointsize{12}}%
+\let\@pointsize\@undefined
+% \end{macrocode}
+%
+% \subsubsection{Media Size}%
+%
+% \begin{macro}{\paperheight}
+% \begin{macro}{\paperwidth}
+% \changes{4.0b}{1999/06/20}{Restore all media size class option of \protect\file{classes.dtx}}
+% \begin{macrocode}
+\DeclareOption{a4paper}{%
+ \setlength\paperheight {297mm}%
+ \setlength\paperwidth {210mm}%
+}%
+\DeclareOption{a5paper}{%
+ \setlength\paperheight {210mm}%
+ \setlength\paperwidth {148mm}%
+}%
+\DeclareOption{b5paper}{%
+ \setlength\paperheight {250mm}%
+ \setlength\paperwidth {176mm}%
+}%
+\DeclareOption{letterpaper}{%
+ \setlength\paperheight {11in}%
+ \setlength\paperwidth {8.5in}%
+}%
+\DeclareOption{legalpaper}{%
+ \setlength\paperheight {14in}%
+ \setlength\paperwidth {8.5in}%
+}%
+\DeclareOption{executivepaper}{%
+ \setlength\paperheight {10.5in}%
+ \setlength\paperwidth {7.25in}%
+}%
+\DeclareOption{landscape}{%
+ \setlength\@tempdima {\paperheight}%
+ \setlength\paperheight {\paperwidth}%
+ \setlength\paperwidth {\@tempdima}%
+}%
+% \end{macrocode}
+%
+% \changes{4.1a}{2008/01/18}{(AO, 478) \cs{ds@letterpaper}, so that ``letterpaper really is the default''}%
+% Effectively select \classoption{letterpaper}.
+% \begin{macrocode}
+\ds@letterpaper
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \subsubsection{Bibnotes}%
+%
+% \begin{macro}{\frontmatter@footnote@produce}
+% Frontmatter footnotes result from frontmatter commands like
+% \cmd\email, \cmd\homepage, \cmd\altaffiliation, and \cmd\thanks.
+% The default for \cmd\frontmatter@footnote@produce\ is
+% \cmd\frontmatter@footnote@produce@footnote, which formats the frontmatter footnotes
+% at the foot of the title page. The \classoption{bibnotes} class option
+% defers them to the bibliography.
+% \changes{4.1b}{2008/06/01}{(AO) Implement bibnotes through \cs{frontmatter@footnote@produce} instead of \cs{bibnotes@sw}}%
+% \begin{macrocode}
+\DeclareOption{bibnotes}{\let\frontmatter@footnote@produce\frontmatter@footnote@produce@endnote}%
+\DeclareOption{nobibnotes}{\let\frontmatter@footnote@produce\frontmatter@footnote@produce@footnote}%
+\let\frontmatter@footnote@produce\frontmatter@footnote@produce@footnote
+\appdef\class@enddocumenthook{\auto@bib}%
+% \end{macrocode}
+% \end{macro}
+%
+% \subsubsection{Footinbib}%
+%
+% \begin{macro}{\footinbib@sw}
+% The boolean \cmd\footinbib@sw\ signifies that
+% text footnotes are to be set in the bibliography, as endnotes.
+%
+% The document may set the value one way or the other via
+% the following two class options.
+% \begin{macrocode}
+\DeclareOption{footinbib}{\@booleantrue\footinbib@sw}
+\DeclareOption{nofootinbib}{\@booleanfalse\footinbib@sw}
+% \end{macrocode}
+% The default value is \cmd\false@sw, and the society or journal
+% may override the default.
+% \changes{4.1b}{2008/07/01}{(AO) make settings at class time instead of deferring them to later.}
+% \begin{macrocode}
+\@booleanfalse\footinbib@sw
+% \end{macrocode}
+% \end{macro}
+%
+% \subsubsection{altaffilletter}%
+%
+% \begin{macro}{\altaffilletter@sw}
+% \changes{4.1b}{2008/07/01}{(AO) make settings at class time instead of deferring them to later.}
+% Determine the procedure \cmd\thefootnote\ used in frontmatter:
+% the footnote symbol used in titlepage footnotes.
+% \begin{macrocode}
+\DeclareOption{altaffilletter}{\@booleantrue\altaffilletter@sw}%
+\DeclareOption{altaffilsymbol}{\@booleanfalse\altaffilletter@sw}%
+\@booleanfalse\altaffilletter@sw
+% \end{macrocode}
+% \end{macro}
+%
+% \subsubsection{superbib}%
+%
+% \begin{macro}{\place@bibnumber}
+% \changes{4.1b}{2008/07/01}{(AO) make settings at class time instead of deferring them to later.}
+% The procedure \cmd\place@bibnumber\ produces the number
+% at the head of the \cmd\bibitem, in the bibliography.
+% By default, it has the \cmd\bibnumfmt\ meaning assigned by the \classname{natbib} package.
+% It may be overridden by society, journal, or by the document options.
+%
+% \begin{macrocode}
+\DeclareOption{superbib}{%
+ \let\place@bibnumber\place@bibnumber@sup
+}%
+\def\place@bibnumber{\NATx@bibnumfmt}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\place@bibnumber@sup}
+% \begin{macro}{\place@bibnumber@inl}
+% For producing the \cmd\bibitem\ device,
+% we define two procedures to select from.
+%
+% Note that we could have used \classname{natbib}'s \cmd\ifNAT@super\
+% switch, but it does not allow for altering the meaning of \cmd\bibnumfmt.
+% \begin{macrocode}
+\def\place@bibnumber@sup#1{\textsuperscript{#1}}%
+\def\place@bibnumber@inl#1{[#1]}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \subsubsection{citeautoscript}%
+% \begin{macro}{\citeautoscript@sw}
+% This class option allows you to automatically accomodate a change from non-superscripted, numbered references
+% to superscripted, bunmbered references.
+%
+% Note: you should always mark up your document with the assumption that references are \emph{not} going to be superscripted.
+% Otherwise this option has no hope of working properly.
+%
+% \begin{macrocode}
+\DeclareOption{citeautoscript}{\@booleantrue\citeautoscript@sw}%
+\@booleanfalse\citeautoscript@sw
+% \end{macrocode}
+% \end{macro}
+%
+% \subsubsection{Variants on the Bibliography Style}%
+% \changes{4.1b}{2008/07/01}{(AO) provide option longbibliography}
+%
+% \revtex\ anticipates that a society or journal will provide two related Bib\TeX bibliography style variants,
+% designating one as the default.
+% A pair of document class options \classoption{longbibliography} and \classoption{nolongbibliography}
+% allows the user to explicitly select between them.
+% \begin{macro}{\longbibliography@sw}
+%
+%\changes{4.2a}{2017/11/21}{(MD) Make long bibliography style the default now.}
+% \revtex's default for \cmd\longbibliography@sw\ is \cmd\true@sw.
+%
+% \begin{macrocode}
+\DeclareOption{longbibliography}{\@booleantrue\longbibliography@sw}%
+\DeclareOption{nolongbibliography}{\@booleanfalse\longbibliography@sw}%
+\@booleantrue\longbibliography@sw
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\eprint@enable@sw}
+% \changes{4.1o}{2010/02/10}{(AO, 580) Provide a document class option to turn off production of eprint field in bibliography.}
+%
+% The document class options \classoption{eprint} and \classoption{noeprint}
+% give the user the ability to turn off production of the \texttt{eprint} field in the bibliography.
+% \begin{macrocode}
+\DeclareOption{eprint}{\@booleantrue\eprint@enable@sw}%
+\DeclareOption{noeprint}{\@booleanfalse\eprint@enable@sw}%
+\@booleantrue\eprint@enable@sw
+% \end{macrocode}
+% \end{macro}
+%
+% \subsubsection{Simplex/Duplex Pages}%
+%
+% \begin{macro}{\twoside@sw}
+% \changes{4.1b}{2008/07/01}{(AO) coordinate \cs{if@twoside} with \cs{twoside@sw}}
+% The flag \cmd\twoside@sw\ signifies that the document is to be
+% formatted for duplex printing.
+% After the preamble is processed, we align the value
+% of the kernel \cmd\newif\ switch \cmd\if@twoside\
+% to that of \cmd\twoside@sw.
+% \cmd\if@twoside\ itself is used in the kernel's
+% \cmd\cleardoublepage\ and \cmd\@outputpage\ procedures.
+% \begin{macrocode}
+\@booleanfalse\twoside@sw
+\appdef\document@inithook{%
+ \twoside@sw{\@twosidetrue}{\@twosidefalse}%
+}%
+% \end{macrocode}
+% \changes{4.0a}{1998/01/31}{4.d had twoside option setting twoside switch to false}
+% The complementary options \classoption{twoside} and \classoption{oneside}
+% assert formatting for duplex or simplex printing, respectively.
+% At the same time, we arrange for the selection of the page grid
+% with respect to the marginal column:
+% Because \cmd\if@reversemargin\ remains default (false),
+% if duplex printing, this column will always be on the (right),
+% if simplex printing, it will always be on the (outside).
+% QUERY: correct choice?
+%FIXME: assign \cmd\if@mparswitch\ later (and protect the assignment, too).
+% \begin{macrocode}
+\DeclareOption{twoside}{\@booleantrue \twoside@sw\@mparswitchfalse}%
+\DeclareOption{oneside}{\@booleanfalse\twoside@sw\@mparswitchtrue}%
+% \end{macrocode}
+% \end{macro}
+%
+% \subsubsection{Two-Column Page Grid}%
+%
+% \begin{macro}{\twocolumn@sw}
+% \changes{4.1b}{2008/07/01}{(AO) make settings at class time instead of deferring them to later.}
+% \changes{4.1p}{2010/02/24}{(AO, 583) Provide interface to \classname{ltxgrid} \cs{onecolumn@grid@setup} and \cs{twocolumn@grid@setup}}
+% The flag \cmd\twocolumn@sw\ signifies that the document is to be
+% formatted in the two-column page grid.
+%
+% If no options relating to page grid are invoked by
+% \cmd\AtBeginDcoument\ time, we set default values.
+% Up to that point, the class can check if \cmd\twocolumn@sw\
+% is \cmd\undefined\ to see if any related options have been invoked.
+%
+% \classoption{onecolumn} specifies one-column page grid.
+% The \classoption{twocolumn} class option employs the standard mechanism for
+% changing the column grid: the \classname{ltxgrid} package.
+% \begin{macrocode}
+\DeclareOption{onecolumn}{\@booleanfalse\twocolumn@sw}%
+\DeclareOption{twocolumn}{\@booleantrue\twocolumn@sw}%
+\@booleanfalse\twocolumn@sw
+% \end{macrocode}
+% The column grid is determined by the state of the switch \cmd\twocolumn@sw
+% and is effected at \cmd\class@documenthook\ time.
+% The society or journal file may re-define \cmd\select@column@grid\
+% to accomodate, e.g., more than two choices for the page grid.
+%
+% Note that \cmd\open@column@two\ adds items to the Main Vertical List,
+% so constitutes the true beginning of the document.
+%
+% Note also that if the selected column grid is a one-column grid,
+% there is nothing to do, because \classname{ltxgrid} has already
+% set that up via \cmd\@begindocumenthook.
+% \begin{macrocode}
+\def\select@column@grid{%
+ \twocolumn@sw{%
+ \twocolumn@grid@setup
+ \open@twocolumn
+ }{%
+ \onecolumn@grid@setup
+% \end{macrocode}
+% \begin{verbatim}
+% \open@onecolumn
+% \end{verbatim}
+% \begin{macrocode}
+ }%
+}%
+\appdef\class@documenthook{%
+ \select@column@grid
+}%
+% \end{macrocode}
+%
+% \end{macro}
+%
+% \begin{macro}{\clear@document}
+% \changes{4.1n}{2009/12/05}{(AO, 569) Use of \classname{hyperref} interferes with column balancing of last page}%
+%
+% We install into \cmd\class@enddocumenthook\
+% a trap for the procedure \cmd\clearpage\ that attempts to end the current page.
+% This procedure needs to be headpatched with \cmd\close@column\ to end the current page grid.
+%
+% This procedure is executed after all typesetting is complete
+% i.e., after items like \cmd\printtables, as well as all commands queued up by \cmd\AtEndDocument.
+%
+% \begin{macrocode}
+\appdef\setup@hook{%
+ \let\clearpage@ltx\clearpage
+ \prepdef\clear@document{\let\clearpage\clearpage@ltx\let\clear@document\@empty\close@column}%
+ \appdef\class@documenthook{%
+ \appdef\class@enddocumenthook{%
+ \let\clearpage\clear@document
+ }%
+ }%
+}%
+% \end{macrocode}
+%
+% \end{macro}
+%
+% \begin{macro}{\authoryear@sw}
+% \changes{4.1a}{2008/01/18}{(AO, 457) Endnotes to be sorted in with numerical citations.}%
+% The boolean \cmd\authoryear@sw\ signifies that we are to use author-year citations rather than numerical citations.
+%
+% The \classoption{author-year} class option selects ``'author-year'' citations;
+% \classoption{numerical} selectes ``numerical'' citations.
+% The former is the default.
+% \begin{macrocode}
+\DeclareOption{author-year}{\@booleantrue\authoryear@sw}%
+\DeclareOption{numerical}{\@booleanfalse\authoryear@sw}%
+\@booleanfalse\authoryear@sw
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\galley@sw}
+% The boolean \cmd\galley@sw\ signifies that
+% the document is to be formatted in galley style.
+%
+% Asserting both \cmd\galley@sw\ and \cmd\preprintsty@sw\
+% may produce strange formatting results, but it is not illegal.
+% However, it \emph{is} illegal to assert galley and any
+% twocolumn option.
+% \begin{macrocode}
+% \end{macrocode}
+% \classoption{galley} emulates setting the galleys of
+% a two-column journal.
+% CHANGED: this option should effectively set \cmd\preprintsty@sw\ false.
+% NOTE: it makes no sense to assert both galley and twocolumn.
+% \changes{4.1b}{2008/07/01}{Class option galley sets \cs{preprintsty@sw} to false}
+%
+% \begin{macrocode}
+\DeclareOption{galley}{%
+ \ExecuteOptions{onecolumn}%
+ \@booleantrue\galley@sw
+ \@booleanfalse\preprintsty@sw
+ \appdef\setup@hook{%
+ \advance\textwidth-\columnsep
+ \textwidth.5\textwidth
+ }%
+}%
+\@booleanfalse\galley@sw
+% \end{macrocode}
+% \end{macro}
+%
+% \subsubsection{raggedbottom or flushbottom}%
+% \begin{macro}{\raggedcolumn@sw}
+% \changes{4.1b}{2008/07/01}{(AO) make settings at class time instead of deferring them to later.}
+% The class options \classoption{raggedbottom} and \classoption{flushbottom} determine whether the columns (page)
+% are ragged bottom or flush bottom.
+% Note that we do not select a default here; that is done by the journal substyle.
+% \begin{macrocode}
+\DeclareOption{raggedbottom}{\@booleantrue\raggedcolumn@sw}
+\DeclareOption{flushbottom}{\@booleanfalse\raggedcolumn@sw}
+\@booleanfalse\raggedcolumn@sw
+\appdef\setup@hook{%
+ \raggedcolumn@sw{\raggedbottom}{\flushbottom}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \subsubsection{tightenlines}%
+%
+% This class option specifies that standard leading
+% is to be used to set the type.
+% If lacking, the leading will be loose.
+%
+% \begin{macro}{\tightenlines@sw}
+% \changes{4.1b}{2008/07/01}{(AO) make settings at class time instead of deferring them to later.}
+% The boolean \cmd\tightenlines@sw\ signifies that
+% the leading is to be made standard amount.
+% If false, it means that the leading is to be set extra open.
+% Has no effect on 10pt size option.
+%
+% \begin{macrocode}
+\DeclareOption{tightenlines}{\@booleantrue\tightenlines@sw}
+\@booleanfalse\tightenlines@sw
+% \end{macrocode}
+% \end{macro}
+%
+% \subsubsection{lengthcheck}%
+%
+% \begin{macro}{\lengthcheck@sw}
+% \changes{4.1b}{2008/07/01}{(AO) make settings at class time instead of deferring them to later.}
+% The flag \cmd\lengthcheck@sw\ signifies that
+% the length checking is in effect.
+% It is up to the individual journal substyle
+% to alter its formatting accordingly.
+% \begin{macrocode}
+\@booleanfalse\lengthcheck@sw
+% \end{macrocode}
+% This class option specifies that the formatted document
+% should approach as closely as possible the
+% formatting of an actual journal article
+% to facilitate the author's performance of a length check.
+%
+%FIXME: society or journal may have its own definition of this option.
+% \begin{macrocode}
+\DeclareOption{lengthcheck}{%
+ \@booleantrue\lengthcheck@sw
+ \ExecuteOptions{reprint}%
+}%
+% \end{macrocode}
+% In addition, if length checking is in effect, we will enable
+% the tally of text length.
+% \begin{macrocode}
+\appdef\setup@hook{%
+ \lengthcheck@sw{\@booleantrue\tally@box@size@sw}{}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \subsubsection{Draft and Final}%
+%
+% \begin{macro}{\draft@sw}
+% \changes{4.1b}{2008/07/01}{(AO) make settings at class time instead of deferring them to later.}
+% The flag \cmd\draft@sw\ signifies that the document is to be
+% formatted in draft mode.
+%
+% \begin{macrocode}
+\appdef\setup@hook{%
+ \draft@sw{\overfullrule 5\p@}{\overfullrule\z@}%
+}%
+% \end{macrocode}
+%
+% Certain packages may pay attention to the
+% class option \classoption{draft} that sets this Boolean.
+% \begin{macrocode}
+\DeclareOption{draft}{\@booleantrue\draft@sw}%
+\DeclareOption{final}{\@booleanfalse\draft@sw}%
+\@booleanfalse\draft@sw
+% \end{macrocode}
+%
+% \end{macro}
+%
+% \subsubsection{eqsecnum}%
+%
+% \begin{macro}{\eqsecnum@sw}
+% The flag \cmd\eqsecnum@sw\ signifies that equations are to be
+% numbered with the section, e.g., ``Eq. (2.13)''.
+% \begin{macrocode}
+\appdef\setup@hook{%
+ \eqsecnum@sw{%
+ \@addtoreset{equation}{section}%
+ \def\theequation@prefix{\arabic{section}.}%
+ }{}%
+}%
+% \end{macrocode}
+%
+% The \classoption{eqsecnum} class option signifies that equations are
+% to be numbered within sections.
+% \begin{macrocode}
+\DeclareOption{eqsecnum}{\@booleantrue\eqsecnum@sw}%
+\@booleanfalse\eqsecnum@sw
+% \end{macrocode}
+% \end{macro}
+%
+% \subsubsection{secnumarabic}%
+% \changes{4.1b}{2008/07/01}{(AO) No longer use \cs{secnumarabic@sw}, instead use \cs{setup@secnums}}
+% The \classoption{secnumarabic} class option signifies that sectioning commands are
+% to be numbered arabic:
+% the procedure \cmd\secnums@arabic\ is executed as the default.
+% Otherwise, the procedure \cmd\secnums@rtx\ determines things.
+% The society or journal may redefine either procedure,
+% and may change the definition of \cmd\setup@secnums\ itself,
+% thereby establishing a different default.
+% \begin{macrocode}
+\appdef\setup@hook{%
+ \setup@secnums
+}%
+\DeclareOption{secnumarabic}{%
+ \def\setup@secnums{\secnums@arabic}%
+}%
+\def\setup@secnums{\secnums@rtx}%
+% \end{macrocode}
+% The code that defines \cmd\secnums@rtx\ and \cmd\secnums@arabic\ appears
+% in Section~\ref{sec:secnumarabic}.
+%
+% \begin{macro}{fleqn}
+% FIXME: model fleqn after amsfonts. I no longer understand why I said this.
+% \file{fleqn.clo} is not a package, so it can simply be \cmd\input.
+% \begin{macrocode}
+\DeclareOption{fleqn}{%
+ \input{fleqn.clo}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \subsubsection{floats/endfloats}%
+%
+% \begin{macro}{\floats@sw}
+% \begin{macro}{\floatp@sw}
+% \changes{4.1b}{2008/07/01}{(AO) make settings at class time instead of deferring them to later.}
+% The Boolean \cmd\floats@sw\ signifies that floats are to be floated;
+% if false, that floats are to be deferred to the end of the document.
+% By default, the former.
+% Note that the state of this Boolean is to be changed by
+% the document class in response to user-selected options.
+%
+% This boolean and the assignment of its default value is done by
+% the \classname{ltxutil} package.
+%
+% The Boolean \cmd\floatp@sw\ signifies that endfloats are to be set one per page;
+% if false, that endfloats are to be set with multiple floats per page permitted.
+% By default, the latter.
+% Note that the state of this Boolean is to be changed by
+% the document class in response to user-selected options.
+% The default is established here.
+%
+% These options control, via the Boolean \cmd\floats@sw,
+% whether floats are to be migrated to the end of the document.
+% \begin{macrocode}
+\DeclareOption{floats}{\@booleantrue\floats@sw\@booleanfalse\floatp@sw}
+\DeclareOption{endfloats}{\@booleanfalse\floats@sw\@booleanfalse\floatp@sw}
+\DeclareOption{endfloats*}{\@booleanfalse\floats@sw\@booleantrue\floatp@sw}
+\@booleantrue\floats@sw
+\@booleantrue\floatp@sw
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \subsubsection{titlepage/notitlepage}%
+% These options control, via \cmd\titlepage@sw,
+% whether the title block is to be set on a separate page.
+%
+% \begin{macro}{\titlepage@sw}
+% \changes{4.1b}{2008/07/01}{(AO) make settings at class time instead of deferring them to later.}
+% The flag \cmd\titlepage@sw\ signifies that a forced
+% page break is to follow the title page: the article
+% title appears on a page by itself.
+%
+% \begin{macrocode}
+\DeclareOption{titlepage}{\@booleantrue\titlepage@sw}
+\DeclareOption{notitlepage}{\@booleanfalse\titlepage@sw}
+\@booleanfalse\titlepage@sw
+% \end{macrocode}
+% \end{macro}
+%
+% \subsubsection{Substyle and Sub-substyle}%
+%
+% \begin{macro}{\change@society}
+% \begin{macro}{\change@journal}
+% \changes{4.1b}{2008/07/01}{(AO) Provide more diagnostics when \cs{@society} is assigned.}
+% If the society or, resp., journal has already been assigned,
+% notify user whether it is being overridden.
+%
+% \begin{macrocode}
+\def\change@society#1{%
+ \def\@tempa{#1}%
+ \@ifxundefined\@society{%
+ \class@info{Selecting society \@tempa}%
+ \let\@society\@tempa
+ }{%
+ \@ifx{\@tempa\@society}{}{%
+ \class@warn{Conflicting society \@tempa<>\@society; not selected}%
+ }%
+ }%
+}%
+\def\change@journal#1{%
+ \def\@tempa{#1}%
+ \@ifxundefined\@journal{%
+ \class@info{Selecting journal \@tempa}%
+ \let\@journal\@tempa
+ }{%
+ \@ifx{\@tempa\@journal}{}{%
+ \class@warn{Conflicting journal \@tempa<>\@journal; not selected}%
+ }%
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% Here had been the class options relating to the APS.
+% Now that all societies are on an equal footing, this code is in the
+% respective \file{.rtx} file.
+% \begin{macrocode}
+% \end{macrocode}
+%
+% \subsubsection{Optical Society of America}%
+% Here are the class options relating to the Optical Society of America.
+%
+% Note: as of 2008, the only OSA module being distributed by ctan is \file{osajnl.rtx}.
+% The class options declared here are, I think, unused.
+% \begin{macrocode}
+\DeclareOption{osa}{\change@society{osa}\let\@journal\@undefined}%
+\DeclareOption{osameet}{\change@society{osa}\def\@journal{osameet}}%
+\DeclareOption{opex}{\change@society{osa}\def\@journal{opex}}%
+\DeclareOption{tops}{\change@society{osa}\def\@journal{tops}}%
+\DeclareOption{josa}{\change@society{osa}\def\@journal{josa}}%
+% \end{macrocode}
+%
+% \begin{macro}{\rtx@require@packages}
+% \changes{4.1b}{2008/07/01}{Read in all required packages together}
+% The procedure \cmd\rtx@require@packages\ accumulates all \cmd\RequirePackage\ statements
+% in the course of loading the document class.
+% Carrying out these operations at that time is needed:
+% \cmd\ProcessOptions\ must be executed first.
+% \begin{macrocode}
+\let\rtx@require@packages\@empty
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\MakeUppercase}
+% \begin{macro}{\MakeLowercase}
+% \changes{4.1b}{2008/07/01}{Package textcase is now simply a required package}
+% \changes{4.1b}{2008/07/01}{Read in all required packages together}
+% \changes{4.1g}{2009/10/07}{(AO, 538) \cs{MakeTextUppercase} inappropriately expands the double backslash}
+% We load the \classname{textcase} package of David Carlisle.
+% Now that its bug of long standing has been repaired, we no longer need to doctor it up.
+% And, because its loading has been deferred until \cmd\rtx@require@packages\ time,
+% we no longer override \LaTeX\ here. Instead, the \classname{textcase} package
+% will be asked to do that.
+% \begin{macrocode}
+\appdef\rtx@require@packages{%
+ \RequirePackage[overload]{textcase}%
+}%
+% \end{macrocode}
+% The following code used to
+% let the \classname{textcase} commands override those of \LaTeX:
+% \begin{verbatim}
+%\appdef\setup@hook{%
+% \expandafter
+% \let\csname MakeUppercase \expandafter\endcsname
+% \csname MakeTextUppercase \endcsname
+% \expandafter
+% \let\csname MakeLowercase \expandafter\endcsname
+% \csname MakeTextLowercase \endcsname
+% }%
+% \end{verbatim}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{amsfonts}
+% \changes{4.1b}{2008/07/01}{Read in all required packages together}
+% The class option \classoption{amsfonts} has the
+% same effect as if the document preamble contained
+% a \cmd\usepackage\verb${amsfonts}$ statement.
+% \begin{macrocode}
+\DeclareOption{amsfonts}{%
+ \def\class@amsfonts{\RequirePackage{amsfonts}}%
+}%
+\DeclareOption{noamsfonts}{%
+ \let\class@amsfonts\@empty
+}%
+\appdef\rtx@require@packages{%
+ \@ifxundefined\class@amsfonts{}{\class@amsfonts}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{amssymb}
+% \changes{4.1b}{2008/07/01}{Read in all required packages together}
+% The class option \classoption{amssymb} has the
+% same effect as if the document preamble contained
+% a \cmd\usepackage\verb${amssymb}$ statement.
+% \begin{macrocode}
+\DeclareOption{amssymb}{%
+ \def\class@amssymb{\RequirePackage{amssymb}}%
+}%
+\DeclareOption{noamssymb}{%
+ \let\class@amssymb\@empty
+}%
+\appdef\rtx@require@packages{%
+ \@ifxundefined\class@amssymb{}{\class@amssymb}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{amsmath}
+% \changes{4.1b}{2008/07/01}{Read in all required packages together}
+% The class option \classoption{amsmath} has the
+% same effect as if the document preamble contained
+% a \cmd\usepackage\verb${amsmath}$ statement.
+%
+% We require version 1.2 (datestamped 1997/03/20) or later.
+% The \expandafter\cmd\csname ver@amsmath.sty\endcsname, will
+%
+% \LaTeX\ note:
+% Certain \LaTeX\ procedures have an arbitrary and pointless restriction that they may be used only within the preamble.
+% We get around this by preserving the procedures in private \cmd\csname s.
+%
+% FIXME note: it is difficult to ensure that an error summary will be printed on the console at the very end,
+% but \classname{ltxgrid} allows accomplishing this via an interrupt, put down at \cmd\AtEndDocument\ time.
+%
+% \begin{macrocode}
+\DeclareOption{amsmath}{%
+ \def\class@amsmath{\RequirePackage{amsmath}[\ver@amsmath@prefer]}%
+}%
+\DeclareOption{noamsmath}{%
+ \let\class@amsmath\@empty
+}%
+\appdef\rtx@require@packages{%
+ \preserve@LaTeX
+ \@ifxundefined\class@amsmath{}{\class@amsmath}%
+ \appdef\class@enddocumenthook{\test@amsmath@ver}%
+}%
+\appdef\preserve@LaTeX{%
+ \let\@ifl@aded@LaTeX\@ifl@aded
+ \let\@ifpackageloaded@LaTeX\@ifpackageloaded
+ \let\@pkgextension@LaTeX\@pkgextension
+ \let\@ifpackagelater@LaTeX\@ifpackagelater
+ \let\@ifl@ter@LaTeX\@ifl@ter
+ \let\@ifl@t@r@LaTeX\@ifl@t@r
+ \let\@parse@version@LaTeX\@parse@version
+}%
+\appdef\restore@LaTeX{%
+ \let\@ifl@aded\@ifl@aded@LaTeX
+ \let\@ifpackageloaded\@ifpackageloaded@LaTeX
+ \let\@pkgextension\@pkgextension@LaTeX
+ \let\@ifpackagelater\@ifpackagelater@LaTeX
+ \let\@ifl@ter\@ifl@ter@LaTeX
+ \let\@ifl@t@r\@ifl@t@r@LaTeX
+ \let\@parse@version\@parse@version@LaTeX
+}%
+\def\test@amsmath@ver{%
+ \begingroup
+ \restore@LaTeX
+ \@ifpackageloaded{amsmath}{%
+ \@ifpackagelater{amsmath}{\ver@amsmath@prefer}{}{%
+ \class@warn{%
+ You have loaded amsmath, version "\csname ver@amsmath.sty\endcsname",\MessageBreak
+ but this class requires version "\ver@amsmath@prefer", or later.\MessageBreak
+ Please update your LaTeX installation.
+ }%
+ }%
+ }{%
+ }%
+ \endgroup
+}%
+\def\ver@amsmath@prefer{2000/01/15 v2.05 AMS math features}%
+% \end{macrocode}
+% \end{macro}
+%
+% \subsubsection{Presenting Authors and Their Affiliations}%
+% \changes{4.1b}{2008/07/01}{Code relating to new syntax for frontmatter has been placed in \file{ltxfront.dtx}}
+% Class options for presenting authors and their affiliations are now defined in
+% \file{ltxfront.dtx}.
+%
+% \begin{macrocode}
+% \end{macrocode}
+%
+% \subsubsection{Typeset by \revtex}%
+%
+% \begin{macro}{\byrevtex@sw}
+% \changes{4.1b}{2008/07/01}{(AO) make settings at class time instead of deferring them to later.}
+% The flag \cmd\byrevtex@sw\ signifies that the document should
+% bear an imprint to the effect that it was formatted by this document class.
+%
+% The class option \classoption{byrevtex} signifies that you want the
+% ``Typeset by \revtex'' byline to appear on your formatted output.
+% By default, no such byline appears.
+% \begin{macrocode}
+\DeclareOption{byrevtex}{\@booleantrue\byrevtex@sw}%
+\@booleanfalse\byrevtex@sw
+% \end{macrocode}
+% \end{macro}
+%
+% \subsection{Attempt to fix float placement failure}%
+%
+% \begin{macro}{\force@deferlist@sw}
+% \changes{4.1b}{2008/07/01}{(AO) make settings at class time instead of deferring them to later.}
+% \revtex\ uses the \classname{ltxgrid} package, which provides the ability to attempt repairs when
+% \LaTeX's float placement mechanism is about to fail, but that facility is turned off by default.
+% Users should invoke the \classoption{floatfix} document class option to enable this \LaTeX\ extension.
+% If not, a helpful message is printed in the log, indicating how to work around the difficulty.
+% \begin{macrocode}
+\DeclareOption{floatfix}{\@booleantrue\force@deferlist@sw}%
+\DeclareOption{nofloatfix}{\@booleanfalse\force@deferlist@sw}%
+\@booleanfalse\force@deferlist@sw
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@fltovf}
+% \begin{macro}{\@fltstk}
+% The \LaTeX\ kernel error message \cmd\@fltovf\ may now be a bit more helpful to the user;
+% likewise for the \cmd\@fltstk\ message of \classname{ltxgrid}.
+% \begin{macrocode}
+\gdef\@fltovf{%
+ \@latex@error{%
+ Too many unprocessed floats%
+ \force@deferlist@sw{}{; try class option [floatfix]}%
+ }\@ehb
+}%
+\def\@fltstk{%
+ \@latex@warning{%
+ A float is stuck (cannot be placed)%
+ \force@deferlist@sw{}{; try class option [floatfix]}%
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\ltxgrid@info@sw}
+% \begin{macro}{\outputdebug@sw}
+% The two options \classoption{ltxgridinfo} and \classoption{outputdebug} turn on informative diagnostics within the
+% package \classname{ltxgrid}. Only people who really want to see this output will select these class options.
+% Consult documentation for the \classname{ltxgrid} package to see what output the related switches enable.
+% \begin{macrocode}
+\DeclareOption{ltxgridinfo}{%
+ \@booleantrue\ltxgrid@info@sw
+%\@booleantrue\ltxgrid@foot@info@sw
+}%
+\DeclareOption{outputdebug}{%
+ \@booleantrue\outputdebug@sw
+ \@booleantrue\ltxgrid@info@sw
+ \@booleantrue\ltxgrid@foot@info@sw
+ \traceoutput
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \subsection{Option to relax page height}%
+%
+% \begin{macro}{\textheight@sw}
+% The \classname{ltxgrid} package can set text pages to their natural height or force them to the full
+% text height; the latter is the default.
+% If setting the pages with a variable length,
+% the running foot will move up or down with the natural length of the text column.
+% While I recommend against doing so, this option will turn that switch to the latter setting.
+% \begin{macrocode}
+\DeclareOption{raggedfooter}{\@booleanfalse\textheight@sw}%
+\DeclareOption{noraggedfooter}{\@booleantrue\textheight@sw}%
+% \end{macrocode}
+% \end{macro}
+%
+% \subsection{Selecting procedure for processing abstract}%
+% \changes{4.1b}{2008/07/01}{Remove options newabstract and oldabstract}
+% Code defining options \classoption{newabstract} and \classoption{oldabstract} has been removed.
+%
+% \subsection{Option to turn on diagnostics in the frontmatter}%
+%
+% \begin{macro}{\frontmatterverbose@sw}
+% \changes{4.1b}{2008/07/01}{Code relating to new syntax for frontmatter has been placed in \file{ltxfront.dtx}}
+% A diagnostic option, not for the average enduser, which reveals
+% the workings of the frontmatter.
+% This code interfaces to that of \file{ltxfront.dtx}.
+% \begin{macrocode}
+\DeclareOption{frontmatterverbose}{\@booleantrue\frontmatterverbose@sw}%
+\@booleanfalse\frontmatterverbose@sw
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\linenumbers@sw}
+% \changes{4.1f}{2009/07/07}{(AO, 513) Add class option linenumbers: number the lines a la \classname{lineno}}
+% \changes{4.1n}{2009/12/13}{(AO, 573) arrange to load \classname{lineno} after any other packages.}%
+% An option to number the lines of type in the output in the manner of \classname{lineno}.
+%
+% At present, we use that very package to implement this functionality.
+% This means that users may modify the workings of that package per its documentation (which see).
+%
+% However, compatibility with \classname{amsmath} requires that \classname{lineno} be loaded afterwards.
+% Therefore, we defer loading of this package until after the preamble is completed.
+% \begin{macrocode}
+\DeclareOption{linenumbers}{%
+ \appdef
+ \class@documenthook{%
+ \RequirePackage{lineno}[2005/11/02 v4.41]%
+ \linenumbersep4pt\relax
+ \linenumbers\relax
+ }%
+}%
+% \end{macrocode}
+% By default, line numbering is off.
+% \end{macro}
+%
+% \begin{macro}{\NAT@merge}
+% \changes{4.1f}{2009/07/15}{(AO, 523) Add class option nomerge, to turn off new natbib 8.3 syntax}
+% Add class option nomerge, to turn off natbib 8.3 syntax for citation key.
+% The default value of REVTeX 4.1 for \cmd\NAT@merge is \cmd\thr@@, which turns on
+% the new syntax along with its semantics.
+% Legacy documents that would be incompatible with the new syntax can be successfully processed
+% with class option \classoption{nomerge}.
+% \begin{macrocode}
+\DeclareOption{nomerge}{%
+ \appdef\setup@hook{%
+ \@ifnum{\NAT@merge>\z@}{\let\NAT@merge\z@}{}%
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \subsection{Default Option, Society, Journal, and pointsize}%
+%
+% \changes{4.1a}{2008/01/18}{(AO, 488) Change processing of options to allow an unused option to specify society and journal}%
+% This change will not break OSA documents because that society is still built in to revtex4.
+%
+% \begin{macro}{\@parse@class@options@society}
+% \begin{macro}{\@parse@class@options@}
+% \changes{4.1b}{2008/07/01}{Procedures \cs{@parse@class@options@society} and \cs{@parse@class@options@journal} and friends}
+% The procedure \cmd\@parse@class@options@society\ parses the options passed to this document class
+% for the \cmd\@society.
+% It is like \cmd\ProcessOptions*\ in that it accesses
+% \cmd\@ptionlist\{\cmd\@currname.\cmd\@currext\}.
+% Any undefined option is considered: if there is a corresponding \file{.rtx} file,
+% it will change the society accordingly
+% and define a placeholder class option for the society thus found
+% (thus preventing a spurious "option not found" message).
+%
+% The procedure \cmd\@parse@class@options@\ parses the document's options
+% for any that set the \cmd\csname\ provided.
+% \begin{macrocode}
+\def\@parse@class@options@society{%
+ \edef\@tempa{\@ptionlist{\@currname.\@currext}}%
+ \expandafter\@for\expandafter\CurrentOption\expandafter:\expandafter=\@tempa\do{%
+ \expandafter\@ifnotrelax\csname ds@\CurrentOption\endcsname{}{%
+ \IfFileExists{\CurrentOption\substyle@post.\substyle@ext}{%
+ \expandafter\change@society\expandafter{\CurrentOption}%
+ \expandafter\let\csname ds@\CurrentOption\endcsname\@empty
+ }{}%
+ }%
+ }%
+}%
+\def\@parse@class@options@#1{%
+ \edef\@tempa{\@ptionlist{\@currname.\@currext}}%
+ \expandafter\@for\expandafter\CurrentOption\expandafter:\expandafter=\@tempa\do{%
+ \expandafter\@ifnotrelax\csname ds@\CurrentOption\endcsname{%
+ \begingroup\csname ds@\CurrentOption\endcsname
+ \@ifxundefined#1{%
+ \endgroup
+ }{%
+ \expandafter\endgroup\expandafter\def\expandafter#1\expandafter{#1}%
+ }%
+ }{}%
+ }%
+}%
+\def\@parse@class@options@journal{%
+ \edef\@tempa{\@ptionlist{\@currname.\@currext}}%
+ \expandafter\@for\expandafter\CurrentOption\expandafter:\expandafter=\@tempa\do{%
+ \expandafter\@ifnotrelax\csname ds@\CurrentOption\endcsname{%
+ \begingroup
+ \csname ds@\CurrentOption\endcsname
+ \@ifxundefined\@journal{%
+ \endgroup
+ }{%
+ \expandafter\endgroup\expandafter\def\expandafter\@journal\expandafter{\@journal}%
+ }%
+ }{}%
+ }%
+}%
+\def\@parse@class@options{%
+ \edef\@tempa{\@ptionlist{\@currname.\@currext}}%
+ \expandafter\@for\expandafter\CurrentOption\expandafter:\expandafter=\@tempa\do{%
+ \expandafter\@ifnotrelax\csname ds@\CurrentOption\endcsname{%
+ \begingroup
+ \csname ds@\CurrentOption\endcsname
+ \@ifxundefined\@pointsize{%
+ \endgroup
+ }{%
+ \expandafter\endgroup\expandafter\def\expandafter\@pointsize\expandafter{\@pointsize}%
+ }%
+ }{%
+ \IfFileExists{\CurrentOption\substyle@post.\substyle@ext}{%
+ \expandafter\change@society\expandafter{\CurrentOption}%
+ \expandafter\let\csname ds@\CurrentOption\endcsname\@empty
+ }{}%
+ }%
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \changes{4.1j}{2009/10/25}{(AO, 545) hypertext capabilities off by default; enable with \classoption{hypertext}}
+% The class option \classoption{hypertext} enables the built-in hypertext capabilities,
+% which coincide with those of \classname{custom-bib}-generated BIBTeX styles
+% using the guard code \texttt{hypertext}.
+%
+% Note that APS has these capabilities turned off by default;
+% Loading the hyperref package turns them on.
+% \begin{macrocode}
+\DeclareOption{hypertext}{\hypertext@enable@ltx}%
+\appdef\document@inithook{\@ifpackageloaded{hyperref}{\hypertext@enable@ltx}{}}%
+% \end{macrocode}
+%
+% The default handling for a document class option depends upon whether the \cmd\@society\ is
+% defined.
+%
+% If not, then hunt for a \file{.rtx} file with that name.
+% If it exists, then we will take this option as the name of the society,
+% otherwise, declare the option as not used.
+%
+% (This behavior is similar to the \LaTeX2.09 handling, where
+% one looked for a \file{.sty} file, except that
+% in this case, we must provide for journal substyles that may be
+% defined in the society file, or have their own journal substyle file.)
+%
+% At the point where the class file is finished loading,
+% we then read in the society file.
+% That file can define further class options, such as the journal substyle.
+%
+% For users, this will mean that they can specify the society and journal simply
+% by specifying first the former and then the latter among their document class options.
+% The society \emph{must} have a corresponding \file{.rtx}.
+%
+% \changes{4.1b}{2008/05/29}{The csname substyle@ext is now defined without a dot (.), to be compatible with \LaTeX usage (see @clsextension and @pkgextension).}
+%
+% \begin{macrocode}
+%</options>
+% \end{macrocode}
+%
+% \begin{macrocode}
+%<*package>
+% \end{macrocode}
+%
+% \begin{macro}{\@process@society}
+% \begin{macro}{\@process@journal}
+% \begin{macro}{\@process@pointsize}
+% \begin{macrocode}
+\def\substyle@post{4-2}%
+\def\substyle@ext{rtx}%
+\DeclareOption*{\OptionNotUsed}%
+% \end{macrocode}
+% A society substyle may define its own options, via \cmd\DeclareOption.
+%
+% At the end of this document class, we process the society file,
+% using \file{aps.rtx} if none has been specified in the document.
+% \begin{macrocode}
+\def\@process@society#1{%
+ \@ifxundefined\@society{%
+ \class@warn{No Society specified, using default society #1}%
+ \def\@society{#1}\let\@journal\@undefined
+ }{}%
+ \expandafter\input\expandafter{\@society\substyle@post.\substyle@ext}%
+}%
+% \end{macrocode}
+%
+% A society substyle can encompass any number of journal substyles;
+% we use the following procedure to invoke the proper one.
+% \begin{macrocode}
+\def\@process@journal#1{%
+ \@ifxundefined\@journal{%
+ \class@warn{No journal specified, using default #1}%
+ \def\@journal{#1}%
+ }{}%
+ \expandafter\expandafter
+ \expandafter\rtx@do@substyle
+ \expandafter\expandafter
+ \expandafter{\expandafter\@society\@journal}%
+}%
+\def\rtx@do@substyle#1{%
+ \InputIfFileExists{#1\substyle@post.\substyle@ext}{}{\csname rtx@#1\endcsname}%
+}%
+% \end{macrocode}
+%
+% Document class options \classoption{10pt}, \classoption{11pt}, and \classoption{12pt}
+% are implemented by \revtex\ itself and determine \cmd\@pointsize.
+% These provide formatting settings appropriate to the society's journals.
+%
+% If not specified by the document, a value \cmd\@pointsize@default\ is used.
+% This default can be set by the journal. Here, the society sets its default.
+% \begin{macrocode}
+\def\@process@pointsize#1{%
+ \@ifxundefined\@pointsize{%
+ \def\@pointsize{#1}%
+ \class@warn{No type size specified, using default \@pointsize}%
+ }{}%
+ \expandafter\expandafter
+ \expandafter\rtx@do@substyle
+ \expandafter\expandafter
+ \expandafter{\expandafter\@society\@pointsize pt}%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \subsection{Class-Asserted Options}%
+% \changes{4.0b}{1999/06/20}{Removed invocation of nonexistent class option \protect\classoption{groupauthors} and all other class options that should only be invoked by the document. (Otherwise precedence of class options does not work.)}
+% Here we establish the default document class options.
+% Those of the document itself will override these.
+%
+% \section{Procedures Dependent Upon Options}
+%
+% Here we introduce \classname{classes.dtx} definitions for
+% the page styles that people will expect to be able to use.
+% \begin{macro}{\ps@headings}
+% \begin{macro}{\ps@myheadings}
+% \begin{macrocode}
+ \def\ps@headings{%
+ \let\@oddfoot\@empty\let\@evenfoot\@empty
+ \def\@evenhead{\thepage\hfil\slshape\leftmark}%
+ \def\@oddhead{{\slshape\rightmark}\hfil\thepage}%
+ \let\@mkboth\markboth
+ \def\sectionmark##1{%
+ \markboth {\MakeUppercase{%
+ \ifnum \c@secnumdepth >\z@
+ \thesection\quad
+ \fi
+ ##1}}{}}%
+ \def\subsectionmark##1{%
+ \markright {%
+ \ifnum \c@secnumdepth >\@ne
+ \thesubsection\quad
+ \fi
+ ##1}}}%
+\def\ps@myheadings{%
+ \let\@oddfoot\@empty\let\@evenfoot\@empty
+ \def\@evenhead{\thepage\hfil\slshape\leftmark}%
+ \def\@oddhead{{\slshape\rightmark}\hfil\thepage}%
+ \let\@mkboth\@gobbletwo
+ \let\sectionmark\@gobble
+ \let\subsectionmark\@gobble
+ }%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\ps@article}
+% \begin{macro}{\ps@article@final}
+% \begin{macro}{\ps@preprint}
+% \changes{4.0a}{1998/01/31}{Move after process options, so the following test works}
+% \changes{4.0b}{1999/06/20}{Defer decision until \cs{AtBeginDocument} time}
+% \changes{4.0c}{1999/11/13}{Insert procedure \cs{checkindate}}
+% \changes{4.0rc1}{2001/06/17}{Running headers always as if two-sided}
+% \begin{macrocode}
+\def\ps@article{%
+ \def\@evenhead{\let\\\heading@cr\thepage\quad\checkindate\hfil{\leftmark}}%
+ \def\@oddhead{\let\\\heading@cr{\rightmark}\hfil\checkindate\quad\thepage}%
+ \def\@oddfoot{}%
+ \def\@evenfoot{}%
+ \let\@mkboth\markboth
+ \let\sectionmark\@gobble
+ \let\subsectionmark\@gobble
+}%
+\def\ps@article@final{%
+ \def\@evenhead{\let\\\heading@cr\thepage\quad\checkindate\hfil{\leftmark}}%
+ \def\@oddhead{\let\\\heading@cr{\rightmark}\hfil\checkindate\quad\thepage}%
+ \def\@oddfoot{}%
+ \def\@evenfoot{}%
+ \let\@mkboth\markboth
+ \def\sectionmark##1{%
+ \markboth{%
+ \MakeTextUppercase{%
+ \@ifnum{\c@secnumdepth >\z@}{\thesection\hskip 1em\relax}{}%
+ ##1%
+ }%
+ }{}%
+ }%
+ \def\subsectionmark##1{%
+ \markright {%
+ \@ifnum{\c@secnumdepth >\@ne}{\thesubsection\hskip 1em\relax}{}%
+ ##1%
+ }%
+ }%
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\heading@cr{\unskip\space\ignorespaces}%
+% \end{macrocode}
+%
+% \changes{4.0c}{1999/11/13}{Do not put by REVTeX in every page foot}
+% \changes{4.0c}{1999/11/13}{Insert procedure \cs{checkindate}}
+% \begin{macrocode}
+\def\ps@preprint{%
+ \def\@oddfoot{\hfil\thepage\quad\checkindate\hfil}%
+ \def\@evenfoot{\hfil\thepage\quad\checkindate\hfil}%
+ \def\@oddhead{}%
+ \def\@evenhead{}%
+ \let\@mkboth\@gobbletwo
+ \let\sectionmark\@gobble
+ \let\subsectionmark\@gobble
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\let\@oddhead\@empty
+\let\@evenhead\@empty
+\let\@oddfoot\@empty
+\let\@evenfoot\@empty
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\lastpage@putlabel}
+% Support the default meaning of \cmd\@endpage.
+% \changes{4.0a}{1998/01/31}{Move after process options, so \cs{clearpage} not in scope of twocolumn}
+% \changes{4.1k}{2009/11/06}{(AO, 554) give the \cs{newlabel} command syntax appropriate to the hyperref package}
+% Name of this macro (and the \cmd\label\ key) taken from
+% \url{CTAN:/macros/latex/contrib/other/lastpage}%
+% with code optimised slightly.
+% \begin{macrocode}
+\def\lastpage@putlabel{%
+ \if@filesw
+ \begingroup
+ \advance\c@page\m@ne
+ \immediate\write\@auxout{\string\newlabel{LastPage}{{}{\thepage}{}{}{}}}%
+ \endgroup
+ \fi
+}%
+% \end{macrocode}
+% Install a procedure into document endgame processing that
+% labels the last page of the document.
+% This is done just before the \file{.aux} file
+% is closed, and does not require a \cmd\shipout, because it writes
+% directly to the \file{.aux} file.
+% Note that we assume no further \cmd\shipout s will be done past this point.
+% \changes{4.1n}{2009/12/09}{(AO, 569) execute the after-last-shipout procedures from within the safety of the output routine}%
+% \begin{macrocode}
+\appdef\clear@document{%
+ \do@output@cclv{%
+ \lastpage@putlabel
+ \tally@box@size@sw{\total@text}{}%
+ }%
+}%
+\providecommand\write@column@totals{}%
+% \end{macrocode}
+% \end{macro}
+%
+%
+% \section{Required Packages}
+% \url{CTAN:macros/latex/contrib/other/misc/url.sty}%
+% \begin{macrocode}
+\appdef\rtx@require@packages{%
+ \RequirePackage{url}%
+}%
+% \end{macrocode}
+%
+% \section{Incompatible Packages}
+%
+% We wait until after the preamble is processed, then check for any packages
+% that might have been loaded which we know to be incompatible with \revtex.
+%
+% The \classname{multicol} package is incompatible with \classname{ltxgrid}, which replaces it.
+% The \classname{cite} package is incompatible with \classname{natbib}, which replaces its functionality.
+% The functionality of the \classname{mcite} package is provided by \classname{natbib}.
+% \begin{macrocode}
+\appdef\document@inithook{%
+ \incompatible@package{cite}%
+ \incompatible@package{mcite}%
+ \incompatible@package{multicol}%
+}%
+% \end{macrocode}
+%
+% \section{Society- and Journal-Specific Code}
+%
+% \begin{macro}{\@journal}
+% Journal test helper, used as
+% \begin{verbatim}
+%\@ifx{\@journal\journal@pra}{%
+% <journal-specific setup>
+%\fi
+% \end{verbatim}
+% Journal code might like to further specify (if as yet undefined) or distinguish on
+% the following Booleans.
+%
+% Note: the journal substyle code should only alter the value of one of these Booleans if
+% the Boolean is \cmd\undefined. This convention is what makes the document's
+% options take precedence over the values set by the journal.
+%
+% FIXME: make this table an exhaustive listing of all the parameters set by
+% the class options.
+%
+% \begin{tabular}{ll}
+% \cmd\@pointsize &(10|11|12), depending on the type size\\
+% \cmd\footinbib@sw &true if footnotes are to be formatted in the bibliography\\
+% \cmd\preprintsty@sw &true for preprint and hyperpreprint\\
+% \cmd\eqsecnum@sw &true means that equations are numbered within sections\\
+% \cmd\groupauthors@sw &true means authors listed separately for each address\\
+% \cmd\preprint@sw &true means to produce the preprint numbers as part of the title block\\
+% \cmd\showPACS@sw &true means to produce the PACS as part of the title block\\
+% \cmd\showKEYS@sw &true means to produce the keywords as part of the title block\\
+% \cmd\@affils@sw &true means each affiliation is printed, for each author\\
+% \cmd\runinaddress@sw &true means author addresses are printed run-in\\
+% \cmd\draft@sw &true implies that PACS will be printed\\
+% \cmd\tightenlines@sw &true if preprint single spaced\\
+% \cmd\lengthcheck@sw &true if length checking is in effect\\
+% \cmd\byrevtex@sw &true means to announce ``typeset by \revtex''\\
+% \cmd\titlepage@sw &true for title is to be set on a separate page\\
+% \cmd\twocolumn@sw &true if two-column page grid\\
+% \cmd\twocolumn@sw &true if we are to automatically balance the columns of the last page\\
+% \cmd\twoside@sw &true means to format pages for duplex printing\\
+% \cmd\floats@sw &false means floats are migrated to end of document\\
+% \cmd\floatp@sw &true means endfloats are set one to a page\\
+% \cmd\class@amsfonts &if \cmd\@empty, means that \classname{amsfonts} will \emph{not} be loaded\\
+% \cmd\class@amssymb &if \cmd\@empty, means that \classname{amssymb} will \emph{not} be loaded\\
+% \cmd\frontmatter@footnote &if \cmd\undefined, means that the default (\cmd\footnote) will be used\\
+% \cmd\place@bibnumber &if \cmd\undefined, means that the default (inline) will be used\\
+% \end{tabular}
+%
+% Note: if \cmd\twocolumn@sw\ and \cmd\preprintsty@sw\ are both false,
+% then `galley' style is in effect. The \classoption{galley} option invokes
+% \classoption{onecolumn}, but does not affect the \cmd\preprintsty@sw.
+%
+% Note: \cmd\paperwidth\ and \cmd\paperheight\ are not integrated into this scheme,
+% and should be selected by the document alone.
+% \end{macro}
+%
+% \section{Body}
+%
+% \subsection{counters}
+% The following definitions are probably identical to those in
+% \file{classes.dtx}%
+% \begin{macrocode}
+\def\labelenumi{\theenumi.}
+\def\theenumi{\arabic{enumi}}
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\labelenumii{(\theenumii)}
+\def\theenumii{\alph{enumii}}
+\def\p@enumii{\theenumi}
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\labelenumiii{\theenumiii.}
+\def\theenumiii{\roman{enumiii}}
+\def\p@enumiii{\theenumi(\theenumii)}
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\labelenumiv{\theenumiv.}
+\def\theenumiv{\Alph{enumiv}}
+\def\p@enumiv{\p@enumiii\theenumiii}
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\labelitemi{\textbullet}
+\def\labelitemii{\normalfont\bfseries\textendash}
+\def\labelitemiii{\textasteriskcentered}
+\def\labelitemiv{\textperiodcentered}
+% \end{macrocode}
+%
+% \begin{macrocode}
+\pagenumbering{arabic}
+% \end{macrocode}
+%
+% \subsection{float parameters}
+%
+% from the old aps.sty. (DPC: same as article I think)
+% AO: here, \LaTeX's standard classes fail very poorly
+% (the price of backward compatability): the values for
+% \cmd\floatpagefraction\ and \cmd\dblfloatpagefraction\
+% need to be raised to avoid creating extremely short
+% float pages.
+% \begin{macrocode}
+\setcounter{topnumber}{2}
+\def\topfraction{.9}
+\setcounter{bottomnumber}{1}
+\def\bottomfraction{.9}
+\setcounter{totalnumber}{3}
+\def\textfraction{.1}
+\def\floatpagefraction{.9}
+\setcounter{dbltopnumber}{2}
+\def\dbltopfraction{.9}
+\def\dblfloatpagefraction{.9}
+% \end{macrocode}
+%
+% \subsection{List Environments}
+%
+% \begin{macrocode}
+\newenvironment{verse}{%
+ \let\\=\@centercr
+ \list{}{%
+ \itemsep\z@ \itemindent -1.5em\listparindent \itemindent
+ \rightmargin\leftmargin\advance\leftmargin 1.5em}\item[]%
+}{%
+ \endlist
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\newenvironment{quotation}{%
+ \list{}{%
+ \listparindent 1.5em
+ \itemindent\listparindent
+ \rightmargin\leftmargin \parsep \z@ \@plus\p@}\item[]%
+}{%
+ \endlist
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\newenvironment{quote}{%
+ \list{}{%
+ \rightmargin\leftmargin}\item[]%
+}{%
+ \endlist
+}%
+% \end{macrocode}
+%
+% \changes{4.1b}{2008/06/30}{(AO) Structure the Abstract using the \texttt{bibliography} environment}
+% \begin{macrocode}
+\def\descriptionlabel#1{%
+ \hspace\labelsep \normalfont\bfseries #1\unskip:%
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\newenvironment{description}{%
+ \list{}{%
+ \labelwidth\z@ \itemindent-\leftmargin
+ \let\makelabel\descriptionlabel
+ }%
+}{%
+ \endlist
+}%
+% \end{macrocode}
+%
+% \subsection{Sectioning Commands\label{sec:secnumarabic}}
+%
+% \subsubsection{Sectioning Commands and Their Productions}%
+% The following counters are defined by LaTeX's standard document classes.
+% We do likewise, then assign flag values to the productions, awaiting
+% overrides.
+% \begin{macrocode}
+\newcounter{part}%
+\let\thepart\@undefined
+\newcounter{section}%
+\let\thesection\@undefined
+\newcounter{subsection}[section]%
+\let\thesubsection\@undefined
+\newcounter{subsubsection}[subsection]%
+\let\thesubsubsection\@undefined
+\newcounter{paragraph}[subsubsection]%
+\let\theparagraph\@undefined
+\newcounter{subparagraph}[paragraph]%
+\let\thesubparagraph\@undefined
+% \end{macrocode}
+% The procedure invoked by \cmd\setup@secnums\ provides meanings for these productions.
+%
+% \begin{macro}{\secnums@rtx}
+% \begin{macro}{\secnums@arabic}
+% These two procedures define the meanings of each of the productions
+% of the counters of the sectioning commands,
+% but only if nothing else has defined it.
+% \begin{macrocode}
+\def\secnums@rtx{%
+ \@ifxundefined\thepart{%
+ \def\thepart{\Roman{part}}%
+ }{}%
+ \@ifxundefined\thesection{%
+ \def\thesection {\Roman{section}}%
+ \def\p@section {}%
+ }{}%
+ \@ifxundefined\thesubsection{%
+ \def\thesubsection {\Alph{subsection}}%
+ \def\p@subsection {\thesection\,}%
+ }{}%
+ \@ifxundefined\thesubsubsection{%
+ \def\thesubsubsection {\arabic{subsubsection}}%
+ \def\p@subsubsection {\thesection\,\thesubsection\,}%
+ }{}%
+ \@ifxundefined\theparagraph{%
+ \def\theparagraph {\alph{paragraph}}%
+ \def\p@paragraph {\thesection\,\thesubsection\,\thesubsubsection\,}%
+ }{}%
+ \@ifxundefined\thesubparagraph{%
+ \def\thesubparagraph {\arabic{subparagraph}}%
+ \def\p@subparagraph {\thesection\,\thesubsection\,\thesubsubsection\,\theparagraph\,}%
+ }{}%
+}%
+\def\secnums@arabic{%
+ \@ifxundefined\thepart{%
+ \def\thepart {\Roman{part}}%
+ }{}%
+ \@ifxundefined\thesection{%
+ \def\thesection {\Roman{section}}%
+ \def\p@section {}%
+ }{}%
+ \@ifxundefined\thesubsection{%
+ \def\thesubsection {\thesection.\arabic{subsection}}%
+ \def\p@subsection {}%
+ }{}%
+ \@ifxundefined\thesubsubsection{%
+ \def\thesubsubsection {\thesubsection.\arabic{subsubsection}}%
+ \def\p@subsubsection {}%
+ }{}%
+ \@ifxundefined\theparagraph{%
+ \def\theparagraph {\thesubsubsection.\arabic{paragraph}}%
+ \def\p@paragraph {}%
+ }{}%
+ \@ifxundefined\thesubparagraph{%
+ \def\thesubparagraph {\theparagraph.\arabic{subparagraph}}%
+ \def\p@subparagraph {}%
+ }{}%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \subsubsection{The Acknowledgments Environment}%
+% This user-level markup produces a head introducing
+% the acknowledgments, and acts as a wrapper for the text.
+% In this implementation, it is an unnumbered section,
+% but appears within the toc.
+%
+% For compatiability's sake, we implement it under the alternative
+% spelling \env{acknowledgements}.
+% \changes{4.0c}{1999/11/13}{also spelled ``acknowledgements''.}
+% \changes{4.0d}{2000/05/18}{But alternative spelling is deprecated.}
+% \begin{macrocode}
+\newenvironment{acknowledgments}{%
+ \acknowledgments@sw{%
+ \expandafter\section\expandafter*\expandafter{\acknowledgmentsname}%
+ }{%
+ \par
+ \phantomsection
+ \addcontentsline{toc}{section}{\protect\numberline{}\acknowledgmentsname}%
+ }%
+}{%
+ \par
+}%
+\@booleantrue\acknowledgments@sw
+\newenvironment{acknowledgements}{%
+ \replace@environment{acknowledgements}{acknowledgments}%
+}{%
+ \endacknowledgments
+}%
+% \end{macrocode}
+%
+% \subsubsection{Part Opener}%
+% section setup copied verbatim from revtex3 aps/osa.
+% Does not explicitly depend on pointsize options.
+% \begin{macrocode}
+\def\part{\par
+ \addvspace{4ex}%
+ \@afterindentfalse
+ \secdef\@part\@spart}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\@part[#1]#2{%
+ \@ifnum{\c@secnumdepth >\m@ne}{%
+ \refstepcounter{part}%
+ \addcontentsline{toc}{part}{\thepart\hspace{1em}#1}%
+ }{%
+ \addcontentsline{toc}{part}{#1}%
+ }%
+ \begingroup
+ \parindent \z@ \raggedright
+ \interlinepenalty\@M
+ \@ifnum{\c@secnumdepth >\m@ne}{%
+ \Large \bf \partname~\thepart%
+ \par\nobreak
+ }{}%
+ \huge \bf
+ #2%
+ \markboth{}{}\par
+ \endgroup
+ \nobreak
+ \vskip 3ex
+ \@afterheading
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\@spart#1{{\parindent \z@ \raggedright
+ \interlinepenalty\@M
+ \huge \bf
+ #1\par}
+ \nobreak
+ \vskip 3ex
+ \@afterheading}
+% \end{macrocode}
+%
+% \subsubsection{Stacked Heads}%
+% Here are the class default definitions for sectioning commands.
+% A society or a journal substyle will likely override these definitions.
+%
+% In doing so, you can customize the formatting for a particular level by defining,
+% e.g., \cmd\@hangfrom@section\ or \cmd\@subsectioncntformat.
+% \begin{macrocode}
+\def\section{%
+ \@startsection
+ {section}%
+ {1}%
+ {\z@}%
+ {0.8cm \@plus1ex \@minus .2ex}%
+ {0.5cm}%
+ {\normalfont\small\bfseries}%
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\subsection{%
+ \@startsection
+ {subsection}%
+ {2}%
+ {\z@}%
+ {.8cm \@plus1ex \@minus .2ex}%
+ {.5cm}%
+ {\normalfont\small\bfseries}%
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\subsubsection{%
+ \@startsection
+ {subsubsection}%
+ {3}%
+ {\z@}%
+ {.8cm \@plus1ex \@minus .2ex}%
+ {.5cm}%
+ {\normalfont\small\itshape}%
+}%
+% \end{macrocode}
+%
+% \subsubsection{Runin Heads}%
+% \begin{macrocode}
+\def\paragraph{%
+ \@startsection
+ {paragraph}%
+ {4}%
+ {\parindent}%
+ {\z@}%
+ {-1em}%
+ {\normalfont\normalsize\itshape}%
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\subparagraph{%
+ \@startsection
+ {subparagraph}%
+ {5}%
+ {\parindent}%
+ {3.25ex \@plus1ex \@minus .2ex}%
+ {-1em}%
+ {\normalfont\normalsize\bfseries}%
+}%
+% \end{macrocode}
+%
+%
+% \subsection{Math}
+%
+% \begin{macro}{\theequation}
+% We change the production of the equation counter so that
+% we can accomodate the \classoption{eqsecnum} option.
+% \begin{macrocode}
+\def\theequation{%
+ \theequation@prefix\arabic{equation}%
+}%
+\def\theequation@prefix{}%
+% \end{macrocode}
+% \end{macro}
+%
+%
+% \subsection{Type Size-Dependent Settings}
+%
+% \subsection{All Point Sizes}
+% \begin{macrocode}
+\setcounter{secnumdepth}{4}
+% \end{macrocode}
+%
+% \begin{macrocode}
+\lineskip 1pt
+\normallineskip 1pt
+\def\baselinestretch{1}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\@lowpenalty 51
+\@medpenalty 151
+\@highpenalty 301
+% \end{macrocode}
+%
+% \begin{macrocode}
+\@beginparpenalty -\@lowpenalty
+\@endparpenalty -\@lowpenalty
+\@itempenalty -\@lowpenalty
+% \end{macrocode}
+%
+% \begin{macrocode}
+\arraycolsep 3pt
+\tabcolsep 2pt
+\arrayrulewidth .4pt
+\doublerulesep 2pt
+% \end{macrocode}
+%
+% \begin{macrocode}
+\skip\@mpfootins = 0pt
+% \end{macrocode}
+%
+% \begin{macrocode}
+\fboxsep = 3.0pt
+\fboxrule = 0.4pt
+% \end{macrocode}
+%
+% \subsection{Figures}
+% \changes{4.0b}{1999/06/20}{Processing delayed to \cs{AtBeginDocument} time}
+%
+% \begin{environment}{figure}
+% We define the \env{figure} environment.
+% Later, we will horse around with its meaning
+% in order to accomodate \cmd\floats@sw.
+% \begin{macrocode}
+\newenvironment{figure}
+ {\@float{figure}}
+ {\end@float}
+\newenvironment{figure*}
+ {\@dblfloat{figure}}
+ {\end@dblfloat}
+% \end{macrocode}
+% \end{environment}
+%
+% \begin{macrocode}
+\def\listoffigures{\print@toc{lof}}%
+\def\l@figure{\@dottedtocline{1}{1.5em}{2.3em}}%
+% \end{macrocode}
+%
+% \begin{macro}{\@makecaption}
+% If caption is one line long, to be centered; if lines turn, then set justified.
+% \begin{macrocode}
+\newlength\abovecaptionskip
+\newlength\belowcaptionskip
+\setlength\abovecaptionskip{10\p@}
+\setlength\belowcaptionskip{2\p@}
+% \end{macrocode}
+%
+% \changes{4.0b}{1999/06/20}{One-line caption sets flush left.}
+% \changes{4.0c}{1999/11/13}{Prevent an inner footnote from performing twice}
+% \changes{4.1a}{2008/01/18}{(AO, 460) ``Proper style is "FIG. 1. ..." (no colon)''}%
+% There is a hook \cmd\@caption@fignum@sep\ for determining
+% the separator following the float number, e.g., ``Fig.1''.
+% Formerly, we had defined it to be ``: '', now the colon has been replace by a period (full stop).
+% \begin{macrocode}
+\long\def\@makecaption#1#2{%
+ \par
+% \nobreak
+ \vskip\abovecaptionskip
+ \begingroup
+ \small\rmfamily
+ \sbox\@tempboxa{%
+ \let\\\heading@cr
+ \@make@capt@title{#1}{#2}%
+ }%
+ \@ifdim{\wd\@tempboxa >\hsize}{%
+ \begingroup
+ \samepage
+ \flushing
+ \let\footnote\@footnotemark@gobble
+ \@make@capt@title{#1}{#2}\par
+ \endgroup
+ }{%
+ \global \@minipagefalse
+ \hb@xt@\hsize{\hfil\unhbox\@tempboxa\hfil}%
+ }%
+ \endgroup
+ \vskip\belowcaptionskip
+}%
+\def\@make@capt@title#1#2{%
+ \@ifx@empty\float@link{\@firstofone}{\expandafter\href\expandafter{\float@link}}%
+ {#1}\@caption@fignum@sep#2%
+}%
+\def\@footnotemark@gobble{%
+ \@footnotemark
+ \@ifnextchar[{\@gobble@opt@i}{\@gobble}%
+}%
+\def\@gobble@opt@i[#1]#2{}%
+\def\@mpmakefntext#1{%
+ \flushing
+ \parindent=1em
+ \noindent
+ \hb@xt@1em{\hss\@makefnmark}%
+ #1%
+}%
+\def\@caption@fignum@sep{. }%
+\def\setfloatlink{\def\float@link}%
+\let\float@link\@empty
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\thefigure}
+% The figure counter and float placement defaults.
+% \begin{macrocode}
+\newcounter{figure}
+\renewcommand \thefigure {\@arabic\c@figure}
+% \end{macrocode}
+% Note that we give the `!' modifier by default.
+% This is an effort to avoid the syndrome wherein
+% a deferred float finds itself unqualified for
+% placement, thereby getting carried until \cmd\clearpage.
+% \begin{macrocode}
+\def\fps@figure{tbp}
+\def\ftype@figure{1}
+\def\ext@figure{lof}
+\def\fnum@figure{\figurename~\thefigure}
+% \end{macrocode}
+% We allocate a box register for use in tallying the column inches of floats of this type.
+% \begin{macrocode}
+\expandafter\newbox\csname fbox@\ftype@figure\endcsname
+\expandafter\setbox\csname fbox@\ftype@figure\endcsname\hbox{}%
+% \end{macrocode}
+% \end{macro}
+%
+% \subsubsection{Deferring \env{figure} Floats}%
+%
+% We determine if figures are to float or be deferred until \cmd\printfigures\ time.
+% If so, we open the stream that will receive the deferred document portions.
+% \begin{macrocode}
+\appdef\class@documenthook{%
+ \do@if@floats{figure}{.fgx}%
+}%
+\appdef\class@enddocumenthook{%
+ \printfigures\relax
+}%
+% \end{macrocode}
+%
+% \begin{macro}{\printfigures}
+% The user-level command \cmd\printfigures\ determines
+% where the figures are to appear in a document in which
+% \cmd\floats@sw\ is false.
+% If the user invokes the \classoption{endfloats} class option and fails to
+% insert a \cmd\printfigures\ command,
+% the figures will be printed at the end of the document.
+% If the command is given, but floats are not being deferred,
+% it amounts to a no-op.
+% \changes{4.0b}{1999/06/20}{call \cs{print@floats}}
+% \changes{4.0c}{1999/11/13}{*-form mandates pagebreak}
+% \begin{macrocode}
+\newcommand\printfigures{%
+ \@ifstar{\true@sw}{\floatp@sw{\true@sw}{\false@sw}}%
+ {%
+ \print@float{figure}{\oneapage}%
+ }{%
+ \print@float{figure}{}%
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+%
+% \begin{macro}{\@xfloat@prep}
+% We patch into the procedure \cmd\@xfloat@prep.
+% This patch applies to all floats (not \env{figure} alone)
+% and makes the type center.
+% \begin{macrocode}
+\appdef\@xfloat@prep{%
+ \appdef\@parboxrestore{\centering}%
+%\let\@makefnmark\@makefnmark@latex
+}%
+% \end{macrocode}
+% \end{macro}
+%
+%
+% \subsection{Tables}
+% \changes{4.0b}{1999/06/20}{Processing delayed to \cs{AtBeginDocument} time}
+%
+% DPC: More or less taken from \file{revtex2} \file{aps.sty},
+% but using dcolumn for decimal alignment.
+% \begin{environment}{table}
+% We define the \env{table} environment.
+% Later, we will horse around with its meaning
+% in order to accomodate \cmd\floats@sw.
+% \begin{macrocode}
+\newenvironment{table}
+ {\@float{table}}
+ {\end@float}
+\newenvironment{table*}
+ {\@dblfloat{table}}
+ {\end@dblfloat}
+% \end{macrocode}
+% \end{environment}
+%
+% \begin{macro}{\thetable}
+% Table counter and default float placement declarations.
+% \begin{macrocode}
+\newcounter{table}
+\renewcommand\thetable{\@Roman\c@table}
+% \end{macrocode}
+% Note that we give the `!' modifier by default.
+% This is an effort to avoid the syndrome wherein
+% a deferred float finds itself unqualified for
+% placement, thereby getting carried until \cmd\clearpage.
+% \begin{macrocode}
+\def\fps@table{tbp}
+\def\ftype@table{2}
+\def\ext@table{lot}
+\def\fnum@table{\tablename~\thetable}
+% \end{macrocode}
+% We allocate a box register for use in tallying the column inches of floats of this type.
+% \begin{macrocode}
+\expandafter\newbox\csname fbox@\ftype@table\endcsname
+\expandafter\setbox\csname fbox@\ftype@table\endcsname\hbox{}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macrocode}
+\def\listoftables{\print@toc{lot}}%
+\let\l@table\l@figure
+% \end{macrocode}
+%
+% \begin{macro}{\table@hook}
+% \begin{macro}{\squeezetable}
+% Assign a meaning to the hook installed into float processing.
+%
+% By default floats are \cmd\small. The \cmd\squeezetable\ declaration
+% makes them smaller (\cmd\scriptsize). In general you can locally
+% redefine \cmd\table@hook\ to be whatever you like.
+% (DPC: |\Huge\color{magenta}|\ldots?)
+% \begin{macrocode}
+\def\table@hook{\small}%
+\def\squeezetable{\def\table@hook{\scriptsize}}%
+\appdef\@floatboxreset{\table@hook}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \subsubsection{Deferring \env{table} Floats}%
+%
+% After all packages are loaded, we decide
+% if tables will float or will be deferred until \cmd\printtables\ time.
+%
+% We also deal with the possibility of \env{longtable} environments.
+%
+% \begin{macrocode}
+\def\set@table@environments{%
+ \floats@sw{}{%
+ \let@environment{longtable@float}{longtable}%
+ \let@environment{longtable}{longtable@write}%
+ \let@environment{longtable*@float}{longtable*}%
+ \let@environment{longtable*}{longtable*@write}%
+ \let@environment{turnpage@float}{turnpage}%
+ \let@environment{turnpage}{turnpage@write}%
+ }%
+ \do@if@floats{table}{.tbx}%
+}%
+\appdef\document@inithook{%
+ \set@table@environments
+}%
+\appdef\class@enddocumenthook{%
+ \printtables\relax
+}%
+\newenvironment{longtable@write}{%
+ \write@@float{longtable}{table}%
+}{%
+ \endwrite@float
+}%
+\newenvironment{longtable*@write}{%
+ \write@@float{longtable*}{table}%
+}{%
+ \endwrite@float
+}%
+\newenvironment{turnpage@write}{%
+ \immediate\write\tablewrite{\string\begin{turnpage}}%
+}{%
+ \immediate\write\tablewrite{\string\end{turnpage}}%
+}%
+% \end{macrocode}
+%
+% \begin{macro}{\printtables}
+% The user-level command \cmd\printtables\ determines
+% where the tables are to appear in a document in which
+% \cmd\floats@sw\ is false.
+% If the user invokes the \classoption{nofloats} and fails to
+% insert a \cmd\printtables\ command,
+% the tables will be printed at the end of the document.
+% If the command is given, but floats are not being deferred,
+% it amounts to a no-op.
+% \changes{4.0b}{1999/06/20}{only execute if appropriate}
+% \changes{4.0c}{1999/11/13}{*-form mandates pagebreak}
+% \changes{4.0d}{2000/05/17}{make longtable trigger the head, too}
+% \begin{macrocode}
+\newcommand\printtables{%
+ \begingroup
+ \let@environment{longtable}{longtable@float}%
+ \let@environment{longtable*}{longtable*@float}%
+ \let@environment{turnpage}{turnpage@anchored}%
+ \prepdef\longtable{\trigger@float@par}%
+ \expandafter\prepdef\csname longtable*\endcsname{\trigger@float@par}%
+ \expandafter\prepdef\csname table@floats\endcsname{%
+ \onecolumngrid@push
+ }%
+ \expandafter\appdef\csname endtable@floats\endcsname{%
+ \onecolumngrid@pop
+ }%
+ \@ifstar{\true@sw}{\floatp@sw{\true@sw}{\false@sw}}%
+ {%
+ \print@float{table}{\oneapage}%
+ }{%
+ \print@float{table}{}%
+ }%
+ \endgroup
+}%
+\newenvironment{turnpage@anchored}{%
+ \onecolumngrid@push
+ \setbox\z@\vbox to\textwidth\bgroup
+ \columnwidth\textheight
+}{%
+ \vfil
+ \egroup
+ \rotatebox{90}{\box\z@}%
+ \onecolumngrid@pop
+}%
+% \end{macrocode}
+% \end{macro}
+%
+%
+% \subsection{Videos}
+%\changes{4.1r}{2010/06/22}{(AO, 595) Provide \cs{lovname} along with other List of Videos definitions.}%
+%
+% \begin{environment}{video}
+% We define the \env{video} environment analogously to the
+% \env{figure} and \env{table} environments; it is intended to contain a video.
+%
+% \begin{macrocode}
+\newenvironment{video}
+ {\@float{video}}
+ {\end@float}%
+\newenvironment{video*}
+ {\@dblfloat{video}}
+ {\end@dblfloat}%
+% \end{macrocode}
+% \end{environment}
+%
+% \begin{macro}{\thevideo}
+% The video counter, float placement defaults, strings.
+% \begin{macrocode}
+\newcounter{video}
+\renewcommand \thevideo {\@arabic\c@video}
+% \end{macrocode}
+%
+% File extension and localizable strings.
+% \begin{macrocode}
+\def\ext@video{lov}%
+\def\fname@video{Video}%
+\def\lovname{List of Videos}%
+% \end{macrocode}
+%
+% Float type and default placement.
+% \begin{macrocode}
+\def\fps@video{tbp}%
+\def\ftype@video{4}%
+\def\fnum@video{\fname@video~\thevideo}%
+\appdef\document@inithook{%
+ \@ifxundefined\c@float@type{}{%
+ \global\setcounter{float@type}{8}%
+ }%
+}%
+% \end{macrocode}
+% We allocate a box register for use in tallying the column inches of floats of this type.
+% \begin{macrocode}
+\expandafter\newbox\csname fbox@\ftype@video\endcsname
+\expandafter\setbox\csname fbox@\ftype@video\endcsname\hbox{}%
+% \end{macrocode}
+% The documentation for the \classname{hyperref} package, \file{hyperref.dtx} states:
+% ``classes or package which introduce new elements need to define
+% an equivalent \verb+\theH<name>+ for every \verb+\the<name>+''
+% We do accordingly here.
+% \begin{macrocode}
+\let\theHvideo\thevideo
+% \end{macrocode}
+% But \file{hyperref.dtx} goes on to say,
+% ``We do make a trap to make \verb+\theH<name>+ be the same as \verb+\arabic{<name>}+,
+% if \verb+\theH<name>+ is not defined\ldots''
+% However, it's not doing that right now (as of 6.77u),
+% and I cannot find any such code in there anymore.
+% \end{macro}
+%
+% \begin{macrocode}
+\def\listofvideos{\print@toc{lov}}%
+\let\l@video\l@figure
+% \end{macrocode}
+%
+% \subsubsection{Deferring \env{video} Floats}%
+%
+% We determine if videos are to float or be deferred until \cmd\printvideos\ time.
+% If so, we open the stream that will receive the deferred document portions.
+% \begin{macrocode}
+\appdef\class@documenthook{%
+ \do@if@floats{video}{.vdx}%
+}%
+\appdef\class@enddocumenthook{%
+ \printvideos\relax
+}%
+% \end{macrocode}
+%
+% \begin{macro}{\printvideos}
+% The user-level command \cmd\printvideos\ determines
+% where the videos are to appear in a document in which
+% \cmd\floats@sw\ is false.
+% If the user invokes the \classoption{endfloats} class option and fails to
+% insert a \cmd\printvideos\ command,
+% the videos will be printed at the end of the document.
+% If the command is given, but floats are not being deferred,
+% it amounts to a no-op.
+% \begin{macrocode}
+\newcommand\printvideos{%
+ \@ifstar{\true@sw}{\floatp@sw{\true@sw}{\false@sw}}%
+ {%
+ \print@float{video}{\oneapage}%
+ }{%
+ \print@float{video}{}%
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+%
+% \section{Tabular}
+% Every APS tabular has a double (Scotch) rule above and below.
+% The column specifier ``d'' is implemented using the \classname{dcolumn}
+% package, if available.
+% FIXME: always load dcolumn!
+% \begin{macro}{\tabular@hook}
+% \begin{macro}{\endtabular@hook}
+% \begin{environment}{ruledtabular}
+% \begin{macrocode}
+\def\endtabular@hook{}%
+% \end{macrocode}
+% \begin{verbatim}
+%\RequirePackage{dcolumn}%
+% \end{verbatim}
+% \begin{macrocode}
+\appdef\document@inithook{%
+ \@ifpackageloaded{dcolumn}{%
+ \expandafter\@ifnotrelax\csname NC@find@d\endcsname{}{%
+ \newcolumntype{d}{D{.}{.}{-1}}%
+ }%
+ }{}%
+}%
+\def\toprule{\hline\hline}%
+\def\colrule{\hline}%
+\def\botrule{\hline\hline}%
+\newenvironment{ruledtabular}{%
+ \def\array@default{v}%
+ \appdef\tabular@hook{\def\@halignto{to\hsize}}%
+ \let\tableft@skip@default\tableft@skip
+ \let\tableft@skip\tableft@skip@float
+ \let\tabmid@skip@default\tabmid@skip
+ \let\tabmid@skip\tabmid@skip@float
+ \let\tabright@skip@default\tabright@skip
+ \let\tabright@skip\tabright@skip@float
+ \let\array@row@pre@default\array@row@pre
+ \let\array@row@pre\array@row@pre@float
+ \let\array@row@pst@default\array@row@pst
+ \let\array@row@pst\array@row@pst@float
+ \appdef\array@row@rst{%
+ \let\array@row@pre\array@row@pre@default
+ \let\array@row@pst\array@row@pst@default
+ \let\tableft@skip\tableft@skip@default
+ \let\tabmid@skip\tabmid@skip@default
+ \let\tabright@skip\tabright@skip@default
+ \appdef\tabular@hook{\let\@halignto\@empty}%
+ }%
+}{%
+}%
+% \end{macrocode}
+% \end{environment}
+% \end{macro}
+% \end{macro}
+%
+% \section{Footnote Text}
+%
+% \changes{4.0a}{1998/06/10}{single space footnotes}
+% \changes{4.0b}{1999/06/20}{Frank Mittelbach, has stated in \protect\classname{multicol}: ``The kernel command \cs{@footnotetext} should not be modified.'' Thus, I have removed David Carlisle's redefinition of that command. Note, however, that later versions of \protect\classname{multicol} do not require this workaround. Belt and suspenders.}%
+%
+% \begin{macro}{\@makefntext}
+% \begin{macro}{\@makefnmark}
+% We customize the presentation of the footnote mark: it will not be italic.
+% \begin{macrocode}
+\def\@makefntext#1{%
+ \def\baselinestretch{1}%
+ \parindent1em%
+ \noindent
+ \hb@xt@1.8em{%
+ \hss\@makefnmark
+ }%
+ #1%
+ \par
+}%
+\def\@makefnmark{%
+ \hbox{%
+ \@textsuperscript{%
+ \normalfont\@thefnmark
+ }%
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+%
+% \subsection{Citations, Bibliography, Endnotes}
+%
+% \subsubsection{Bibliography}
+%
+% Load Patrick Daly's \classname{natbib} package,
+% \url{ftp://ctan.tug.org/macros/latex/contrib/supported/natbib}
+%
+% Note that \classname{natbib} assumes that it loads
+% over a document class, such as the \classname{article} class,
+% that has already defined \env{thebibliography} and \cmd\@listi.
+%
+% Note also that \classname{natbib} also installs a command
+% \cmd\NAT@set@cites\ into \cmd\AtBeginDocument\ which presumes
+% that the proper \cmd\bibpunct\ command has been issued.
+%
+% Note that the macro \cmd\NAT@sort\ controls
+% whether citations are left alone (\cmd\NAT@sort=0), sorted (\cmd\NAT@sort=1), or sorted and compressed (\cmd\NAT@sort=2).
+% Since we give \classname{natbib} the \classoption{sort\&compress} option,
+% if you prefer \classoption{sort}, you need only \cmd\let \cmd\NAT@sort\ to be \cmd\@ne.
+% However, if you prefer the effect of having neither \classoption{sort} nor \classoption{sort\&compress},
+% you must \cmd\let \cmd\NAT@sort\ to be \cmd\z@ and you must also define
+% \cmd\let\cmd\NAT@cmprs\ to be \cmd\z@.
+%
+% As of version 8.2, \classname{natbib} now no longer binds at the point where it is read in.
+% This means that we can freely change \cmd\NAT@sort, \cmd\NAT@cmprs, and the new
+% \cmd\NAT@merge. Henceforth, we require that this later version be used.
+%
+% For other \classname{natbib} customizations, you may proceed as if you were going to use
+% the \file{natbib.cfg} file: anything that you can modify by this means is fair game.
+% Once \revtex\ is finished loading, you can assert any definitions for \classname{natbib} that you wish.
+%
+% \begin{macro}{\rev@citet}
+% \begin{macro}{\rev@citealp}
+% \begin{macro}{\rev@citealpnum}
+% We define variants on \classname{natbib}'s commands \cmd\citet, \cmd\citealp, and \cmd\citealpnum.
+% \cmd\rev@citealpnum\ uses a numerical citation.
+% \cmd\rev@citealp\ and \cmd\rev@citealpnum\ are the aliases of \cmd\onlinecite,
+% \cmd\rev@citet\ that of \cmd\textcite.
+%
+% In each case, we invoke \cmd\rtx@swap@citea\ to effect different productions between
+% multiple arguments to the \cmd\cite\ command.
+%
+% \cmd\rev@citealpnum\ provides textual citations where superscript citations are the default.
+% These should be accessible via the \cmd\citet\ command.
+%
+% Therefore we remember how to do a numerical citation
+% even when the superscript citation has been selected.
+% \begin{macrocode}
+\expandafter\DeclareRobustCommand
+\expandafter\rev@citet
+\expandafter{%
+ \expandafter\begingroup
+ \expandafter\rtx@swap@citea
+ \expandafter\g@bblefirsttoken
+ \csname citet \endcsname
+}%
+\expandafter\DeclareRobustCommand
+\expandafter\rev@citealp
+\expandafter{%
+ \expandafter\begingroup
+ \expandafter\rtx@swap@citea
+ \expandafter\g@bblefirsttoken
+ \csname citealp \endcsname
+}%
+\expandafter\DeclareRobustCommand
+\expandafter\rev@citealpnum
+\expandafter{%
+ \expandafter\begingroup
+ \expandafter\rtx@swap@citenum
+ \expandafter\g@bblefirsttoken
+ \csname citealp \endcsname
+}%
+\def\rtx@swap@citenum{%
+ \rtx@swap@citea
+ \let\@cite\NAT@citenum
+ \let\NAT@mbox\mbox
+ \let\citeyear\NAT@citeyear
+ \let\NAT@space\NAT@spacechar
+}%
+\def\g@bblefirsttoken{%
+ \expandafter\true@sw
+ \expandafter\@empty
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\rtx@citesuper}
+% We prepare to redefine \classname{natbib}'s procedure \cmd\NAT@citesuper,
+% which is executed when setting a superscript citation.
+% The \cmd\hspace\ is removed: in any case, it
+% should really be \cmd\hspace*, to prevent an unwanted pagebreak.
+% \begin{macrocode}
+\newcommand\rtx@citesuper[3]{%
+ \ifNAT@swa
+ \leavevmode
+ \unskip
+% \hspace{1\p@}%
+ \textsuperscript{\normalfont#1}%
+ \if*#3*\else\ (#3)\fi
+ \else
+ #1%
+ \fi
+ \endgroup
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@makefnmark@cite}
+% We define a procedure that will set a footnote mark the same way that
+% a citation is set. If footnotes are put in the bibliography with
+% \cmd\footinbib@sw, then the corresponding mark should look the same
+% as the result of a \cmd\cite. This is how we do it.
+% \begin{macrocode}
+\def\@makefnmark@cite{\begingroup\NAT@swatrue\@cite{{\@thefnmark}}{}{}}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\rtx@bibsection}
+% Prepare to override \classname{natbib}'s definition of \cmd\bibsection.
+% \begin{macrocode}
+\def\rtx@bibsection{%
+ \@ifx@empty\refname{%
+ \par
+ }{%
+% \let\@hangfroms@section\@hang@froms
+ \expandafter\section\expandafter*\expandafter{\refname}%
+ \@nobreaktrue
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\rtx@swap@citea}
+% \begin{macro}{\rtx@def@citea}
+% \begin{macro}{\rtx@def@citea@close}
+% \begin{macro}{\rtx@def@citea@box}
+% The procedures |\rtx@def@citea|, |\rtx@def@citea@close|, and |\rtx@def@citea@box|
+% can take over the management of \classname{natbib}'s |\@citea| macro to effect more sophisticated
+% behavior of the punctuation between textual citations. The switch is performed by \cmd\rtx@swap@citea.
+%
+% \changes{4.1j}{2009/10/25}{(AO, 552) Repair spacing in \cs{onlinecite}}
+% In these procedures, we use \cmd\count@\ to count the number of arguments of the \cmd\cite\ command,
+% and we use \cmd\c@NAT@ctr\ to keep track of which argument we are processing.
+% The latter counter is created by \classname{natbib} and used there solely in bibliography processing,
+% where it keeps track of the reference number.
+% We take over its use in these macros, but only locally; therefore these procedures should work properly, even within the bibliography.
+%FIXME: check whether this is true!
+%
+% Because we are using a scratch counter \cmd\count@, we are vulnerable to other \TeX\ programmers
+% who patch in to \classname{natbib}'s processing and who might use that counter at the same time we are doing so.
+% This is a potential source of trouble for us.
+%FIXME: store the value of \cmd\count@\ in a private \cmd\csname!
+%
+% Note that \cmd\rtx@def@citea\ begins the same as \cmd\NAT@def@citea, which it replaces,
+% then makes further decisions based on the values of the counters.
+%
+% Note also that, in \classname{natbib}, the replacement part of \cmd\NAT@def@citea@close\ could be rewritten as
+% \cmd\NAT@def@citea\cmd\prepdef\cmd\@citea\verb+{+\cmd\NAT@@close\verb+}+, which
+% would them obviate the need for us to override its meaning.
+%
+% Note, too, the effect of \cmd\rtx@def@citea@box, which replaces \cmd\NAT@def@citea@box,
+% is almost the same as the latter, except the entire \cmd\@citea\ is given as the argument
+% of \cmd\NAT@mbox.
+%
+% Finally, bear in mind that the English (and some American editors) do not place a comma
+% before the ``and''; our procedures do (but they could be rewritten with that convention).
+% \begin{macrocode}
+\def\rtx@swap@citea{%
+ \let\NAT@def@citea\rtx@def@citea
+ \let\NAT@def@citea@close\rtx@def@citea@close
+ \let\NAT@def@citea@box\rtx@def@citea@box
+}%
+\def\rtx@def@citea{%
+ \def\@citea{\NAT@separator\NAT@space}%
+ \advance\c@NAT@ctr\@ne
+ \@ifnum{\count@>\tw@}{%
+ \@ifnum{\c@NAT@ctr=\count@}{\appdef\@citea{\NAT@conj\NAT@space}}{}%
+ }{%
+ \def\@citea{\NAT@space\NAT@conj\NAT@space}%
+ }%
+}%
+\def\rtx@def@citea@close{%
+ \rtx@def@citea
+ \prepdef\@citea{\NAT@@close}%
+}%
+\def\rtx@def@citea@box{%
+ \rtx@def@citea@close
+ \expandafter\def\expandafter\@citea\expandafter{\expandafter\NAT@mbox\expandafter{\@citea}}%
+}%
+\def\NAT@conj{and}%
+% \end{macrocode}
+%
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\BibitemShut}
+% \begin{macro}{\bibAnnote}
+% \changes{4.1j}{2009/10/24}{(AO, 549) Repairing natbib's \cs{BibitemShut} and \cs{bibAnnote}}
+% \changes{4.1o}{2010/02/05}{(AO, 549) Remove patch to natbib, which is now at version 8.31a}
+% \changes{4.1o}{2010/02/07}{(AO, 578) accommodate the possible space character preceding \cs{BibitemShut}.}
+% We remember a temporary patch to \classname{natbib}'s definition of \cmd\BibitemShut.
+% \begin{macrocode}
+\def\NAT@BibitemShut#1{%
+ \def\@bibstop{#1}%
+ \let\bibitem@Stop\bibitemStop
+ \let\bibitem@NoStop\bibitemNoStop
+ \@ifx{\bibitemShut\relax}{\let\@bibitemShut\@empty}{%
+ \expandafter\def\expandafter\@bibitemShut\expandafter{\bibitemShut}%
+ }%
+}%
+% \end{macrocode}
+% The following is a bug fix to \classname{natbib} version 8.31b.
+% \begin{macrocode}
+\def\BibitemShut@ltx#1{%
+ \unskip
+ \def\@bibstop{#1}%
+ \let\bibitem@Stop\bibitemStop
+ \let\bibitem@NoStop\bibitemNoStop
+ \@ifx{\bibitemShut\relax}{\let\@bibitemShut\@empty}{%
+ \expandafter\def\expandafter\@bibitemShut\expandafter{\bibitemShut}%
+ }%
+}%
+% \end{macrocode}
+%
+% \begin{verbatim}
+%\providecommand{\bibAnnote}[3]{%
+% \BibitemShut{#1}%
+% \def\@tempa{#3}\@ifx{\@tempa\@empty}{}{%
+% \begin{quotation}\noindent
+% \textsc{Key:}\ #2\\\textsc{Annotation:}\ \@tempa
+% \end{quotation}%
+% }%
+% \ignorespaces
+%}%
+%\def\@bibitemShut{}%
+% \end{verbatim}
+% \end{macro}
+% \end{macro}
+%
+%
+% \begin{macrocode}
+\newenvironment{thebibliography}{}{}%
+\let\@listi\@empty
+\appdef\rtx@require@packages{%
+ \RequirePackage[sort&compress]{natbib}[2009/11/07 8.31a (PWD, AO)]%
+ \let@environment{NAT@thebibliography}{thebibliography}%
+ \let@environment{thebibliography}{rtx@thebibliography}%
+ \let\bibliographystyle@latex\bibliographystyle
+ \let\NAT@citesuper\rtx@citesuper
+% \end{macrocode}
+%
+% \begin{macro}{\NAT@bibsetnum}
+% \begin{macro}{\NAT@bibsetup}
+% \begin{macro}{\bibpreamble}
+% \begin{macro}{\newblock}
+% \begin{macro}{\bibnumfmt}
+% \begin{macro}{\NAT@merge}
+% We define the sectioning command to use when starting the bibliography
+% and gently coax \classname{natbib} into using
+% the formatting procedures that \emph{we} want it to use.
+%
+% This way of setting up \env{thebibliography}
+% automatically sets the label width based on
+% the largest number used within the bibliography.
+% This scheme will not work properly using the
+% author/year style of bib entry, though.
+%
+% We define \cmd\bibnumfmt\ to be \cmd\place@bibnumber, which is a macro
+% managed by \revtex. If the document defines \cmd\bibnumfmt, then that
+% definition will be used instead, which is what the \classname{natbib}
+% package gives as its programming interface.
+%
+% We set \cmd\NAT@merge\ to \cmd\tw@, which turns on \classname{natbib}'s
+% \classname{mcite} capabilities. This is the default setting.
+% If numerical citations are not to be used, then \cmd\NAT@merge\
+% should be set to \cmd\@ne\ (syntax is still enabled, but semantics are turned off).
+% \begin{macrocode}
+\let\bibsection\rtx@bibsection
+\let\NATx@bibsetnum\NAT@bibsetnum
+\def\NAT@bibsetnum#1{%
+ \setlength{\topsep}{\z@}%
+ \NATx@bibsetnum{\ref{LastBibItem}}%
+}%
+\let\NATx@bibsetup\NAT@bibsetup
+\def\NAT@bibsetup{%
+ \setlength{\labelwidth}{\z@}%
+ \setlength{\labelsep}{\z@}%
+ \setlength{\itemindent}{\z@}%
+ \setlength{\listparindent}{\z@}%
+ \setlength{\topsep}{\z@}%
+ \setlength{\parsep}{\z@}%
+ \NATx@bibsetup
+}%
+\let\bibpreamble\@empty
+\def\newblock{\ }%
+\let\NATx@bibnumfmt\bibnumfmt
+\def\bibnumfmt{\place@bibnumber}%
+\let\NAT@merge\thr@@
+\let\NAT@citeyear\citeyear
+\let\onlinecite\rev@citealp
+\let\textcite\rev@citet
+% \end{macrocode}
+% The following is needed until \classname{natbib} is at 8.31b.
+% \begin{macrocode}
+\@ifx{\BibitemShut\NAT@BibitemShut}{%
+ \class@info{Repairing natbib's \string\BibitemShut}%
+ \let\BibitemShut\BibitemShut@ltx
+}{}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\bibliographystyle}
+% \changes{4.1q}{2010/04/01}{(AO, 586) When .bbl is pasted into the document, prevent automatic bibliography inclusion.}%
+% \changes{4.1q}{2010/04/13}{(AO, 588) Only write \revtex-specific BibTeX .bib data if the .bst style is set by REVTeX.}%
+% We arrange for the selection of bibliography style
+% to occur either due to the document's explicit
+% \cmd\bibliographystyle\ statement or
+% via the journal substyle.
+%
+% Note that \revtex\ is incompatible with
+% any package that patches \cmd\bibliographystyle.
+% Since \classname{natbib} does this, we need a fix.
+%
+% The Boolean \cmd\bibliographystyle@sw\ signifies that the document contains explicit
+% \cmd\bibliographystyle\ markup. If, on the contrary, the bibliography style is set by
+% the the society or the journal, then no explicit \cmd\bibliographystyle\ command appears in the document instance.
+% In this case \cmd\bibliographystyle@sw\ will be \cmd\false@sw.
+% \begin{macrocode}
+\let\bibliographystyle@latex\bibliographystyle
+\def\bibliographystyle{\@booleantrue\bibliographystyle@sw\def\@bibstyle}%
+\@booleanfalse\bibliographystyle@sw
+% \end{macrocode}
+% \end{macro}
+%
+% \changes{4.1j}{2009/11/03}{(AO, 551) Punctuation at end of reference when optional arguments to the cite key are given.}
+% The following had been bug fixes to \classname{natbib} version 8.31a.
+% \begin{verbatim}
+%\def\bibitemStop{\@bibitemShut}%
+%\def\NAT@bibitem@cont{%
+% \let\bibitem@Stop\bibitemContinue
+% \let\bibitem@NoStop\bibitemContinue
+%}%
+% \end{verbatim}
+% The following are alterations to \classname{natbib} version 8.31a to
+% accommodate the possible space character preceding \cmd\BibitemShut,
+% and to handle the case of merged references, where the first ends with a stop character.
+% \changes{4.1o}{2010/02/07}{(AO, 578) accommodate the possible space character preceding \cs{BibitemShut}.}
+% \changes{4.1o}{2010/02/09}{(AO, 581) Handle case: merged references, with first ending in a stop character.}
+% \begin{macrocode}
+\def\NAT@bibitem@cont{%
+ \let\bibitem@Stop\bibitemContinue@Stop
+ \let\bibitem@NoStop\bibitemContinue
+}%
+% \end{macrocode}
+% \begin{macrocode}
+\def\bibitemNoStop{%
+ \@ifx@empty\@bibitemShut{.\spacefactor\@mmm\space}{\@bibitemShut}%
+}%
+% \end{macrocode}
+% \begin{macrocode}
+\def\bibitemContinue{%
+ \@ifx@empty\@bibitemShut{;\spacefactor\@mmm\space}{\@bibitemShut}%
+}%
+% \end{macrocode}
+% \begin{macrocode}
+\def\bibitemContinue@Stop{%
+ \@ifx@empty\@bibitemShut{\spacefactor\@mmm\space}{\@bibitemShut}%
+}%
+% \end{macrocode}
+%
+% \changes{4.1k}{2009/11/11}{(AO, 561) remove dead code relating to \cs{bibitemContinue}}
+% We used to customize one of the productions of \classname{natbib}, but no longer.
+% \begin{verbatim}
+%\let\bibitemContinue\bibitemContinue@rtx
+% \end{verbatim}
+%
+% Here ends the code to be executed at \cmd\rtx@require@packages\ time.
+% \begin{macrocode}
+}%
+% \end{macrocode}
+% \changes{4.1f}{2009/07/07}{(AO, 516) Merged references are separated with a semicolon}
+% Redefine a macro of \classname{natbib} so that merged references are separated with a semicolon.
+% \changes{4.1k}{2009/11/11}{(AO, 561) remove dead code relating to \cs{bibitemContinue}}
+% \begin{verbatim}
+% \def\bibitemContinue@rtx{;\spacefactor\@mmm\space}%
+% \end{verbatim}
+%
+% \begin{macro}{\onlinecite}
+% \begin{macro}{\textcite}
+% We extend \classname{natbib}'s syntax with two commands to set a citation
+% on the baseline (as opposed to superscripted) and as text (rather than parenthetical), respectively.
+% A journal substyle that makes citations be superscripted or parenthetical as the case may be, should ensure
+% that the author has continued access to these two styles.
+%
+% Note that the society or journal substyle override the meanings of \cmd\@onlinecite\ or
+% \cmd\@textcite\ given here.
+%
+% \begin{macrocode}
+\DeclareRobustCommand\onlinecite{\@onlinecite}%
+\DeclareRobustCommand\textcite{\@textcite}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\bibliography}
+% \changes{4.1a}{2008/01/18}{(AO, 457) Endnotes to be sorted in with numerical citations.}%
+% Provide a hook for supplying Bib\TeX\ a bibliographic database that may contain, say,
+% footnotes.
+%
+% Note that Bib\TeX\ chokes if the argument of the \cmd\bibdata\ command has null fields,
+% hence these tests.
+% \begin{macrocode}
+\let\bibliography@latex\bibliography
+\def\bibliography#1{%
+ \auto@bib@empty
+ \begingroup
+ \let\auto@bib@innerbib\@empty
+ \@ifx@empty{\pre@bibdata}{%
+ \bibliography@latex{#1}%
+ }{%
+ \@if@empty{#1}{%
+ \expandafter\bibliography@latex\expandafter{\pre@bibdata}%
+ }{%
+ \expandafter\bibliography@latex\expandafter{\pre@bibdata,#1}%
+ }%
+ }%
+ \endgroup
+}%
+\let\pre@bibdata\@empty
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{environment}{rtx@thebibliography}%
+% \begin{macro}{\present@bibnote}%
+% \changes{4.1f}{2009/07/11}{(AO, 521) Lonely bibliography head}%
+% \changes{4.1n}{2010/01/02}{(AO, 572) Independent footnote counter for title block. Abstract footnote counter shared with body.}%
+% \changes{4.1o}{2010/02/02}{(AO, 575) Automatically incorporate the (Bib\TeX-generated) .bbl into an explicit \env{thebibliography}}%
+% \changes{4.1q}{2010/04/01}{(AO, 586) When .bbl is pasted into the document, prevent automatic bibliography inclusion.}%
+% We put a tail patch into \cmd\thebibliogrphy\ and
+% a headpatch into \cmd\endthebibliography.
+%
+% Here we provide a default treatment for frontmatter notes
+% deferred to the bibliography; a journal substyle
+% might want to override the definition of \cmd\present@bibnote.
+%
+% We make provisions for the case where there are no \cmd\bibitem s
+% for the bibliography: we produce no bibliography head at all.
+%
+% \begin{macrocode}
+\newenvironment{rtx@thebibliography}[1]{%
+ \NAT@thebibliography{#1}%
+ \let\@TBN@opr\present@bibnote
+ \@FMN@list
+}{%
+% \end{macrocode}
+% The following line was commented out:
+% \begin{verbatim}
+%\@endnotesinbib
+% \end{verbatim}
+% The \cmd\auto@bib@innerbib\ directive has been moved from the begin processing to the end processing.
+% This means that the content of the \env{thebibliography} environment can itself prevent the
+% automatic reading in of the .bbl file. This would be needed when the user has pasted in the
+% content of the .bbl file into the document itself, something required by APS and AIP editorial direction.
+% \begin{macrocode}
+ \auto@bib@innerbib
+ \edef\@currentlabel{\arabic{NAT@ctr}}%
+ \label{LastBibItem}%
+ \endNAT@thebibliography
+ \aftergroup\auto@bib@empty
+}%
+\def\present@bibnote#1#2{%
+ \item[%
+ \textsuperscript{%
+ \normalfont
+ \Hy@raisedlink{\hyper@anchorstart{frontmatter.#1}\hyper@anchorend}%
+ \begingroup
+ \csname c@\@mpfn\endcsname#1\relax
+ \frontmatter@thefootnote
+ \endgroup
+ }%
+ ]#2\par
+% \end{macrocode}
+% The following line was commented out:
+% \begin{verbatim}
+%\global\let\NAT@bibitem@first@sw\@secondoftwo
+% \end{verbatim}
+% \begin{macrocode}
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{environment}
+%
+% \begin{macro}{write@bibliographystyle}
+% \changes{4.1q}{2010/04/13}{(AO, 588) Only write \revtex-specific BibTeX .bib data if the .bst style is set by REVTeX.}%
+% We wish to delay committing the \cmd\bibliographystyle\ until as late as possible.
+% The journal substyle will define a default bibliography style, and
+% the document's explicit \cmd\bibliographystyle\ command, if any, will override that default.
+%
+% The \cmd\bibstyle\ command is allowed appear quite late in the \file{.aux} file.
+% We now delay the automatic writing of the \cmd\bibstyle\ command to the end of the job.
+%
+% The procedure \cmd\write@bibliographystyle\ tests whether a \cmd\bibliographystyle\ command has
+% already been given.
+% If not, it effectively executes the needed \cmd\bibliographystyle\ command, then neutralizes itself
+% (we only need to do this once per job).
+%
+% If the document lacks explicit \cmd\bibliographystyle\ markup, we execute
+% \cmd\@bibdataout@rev, a hook for \revtex-aware processing.
+% \begin{macrocode}
+\def\write@bibliographystyle{%
+ \@ifxundefined\@bibstyle{}{%
+ \expandafter\bibliographystyle@latex\expandafter{\@bibstyle}%
+ \bibliographystyle@sw{}{\@bibdataout@rev}%
+ }%
+ \global\let\write@bibliographystyle\relax
+}%
+\AtEndDocument{\write@bibliographystyle}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\rtx@@citetp}
+% \begin{macro}{\rtx@citex}
+% \begin{macro}{\super@cite@let}
+% \begin{macro}{\super@cite@end}
+% \begin{macro}{\super@cite@swap}
+% We wish to extend \classname{natbib} to move spaces and citations around
+% a superscript-style citation, imitating Donald Arseneau's \classname{cite} package
+% with the \classoption{super}.
+%
+% The \cmd\rtx@@citetp\ procedure is substituted for \cmd\NAT@@citetp;
+% it then calls the \cmd\rtx@citex\ procedure and implements the features of the
+% \classoption{citeautoscript} class option.
+% In the end, \cmd\@citex\ is called with its customary parameters.
+%
+% The document should be marked up as if citations were \emph{not} superscripted,
+% and then if you select a journal substyle that has superscripted citations, \revtex\ will do its best
+% to alter the formatting of the \cmd\cite s to accomodate superscript style.
+%
+% Only citations set as superscript are affected by this procedure,
+% because we check \cmd\@cite\ against \cmd\NAT@citesuper.
+%
+% Here's a subtle point: when is the argument of \cmd\super@cite@swap\ not the same as the token \cmd\@let@token?
+% Answer: when the latter is \cmd\@sptoken! This case has to be handled separately.
+%
+% Note that whether a punctuation is movable is determined by the definition
+% of a particular control sequence name.
+% A society or journal can alter things: to remove a character from the set,
+% do, say, \cmd\expandafter \cmd\let \cmd\csname \texttt{rtx@automove;} \cmd\endcsname \cmd\relax.
+% To add a character to the set, do, say, \cmd\expandafter \cmd\let \cmd\csname \texttt{rtx@automove;} \cmd\endcsname \cmd\@empty.
+%
+% Implementation note: due to a \TeX\ peculiarity, we have to check for the case
+% where \cmd\@let@token\ is a space token \emph{before} we parse forward.
+% At issue is the corner case where an end of file is at hand. If we
+% were to let \cmd\super@cite@swap\ parse forward, we would encounter a \TeX\ end-of-file
+% error. Note that the test will be true in many distinct cases:
+% the file ends, the next character is a line terminator, the next character is a space.
+% \begin{macrocode}
+\def\rtx@@citetp[#1]{\@ifnextchar[{\rtx@citex[#1]}{\rtx@citex[][#1]}}%
+\def\rtx@citex[#1][#2]#3{%
+ \begingroup
+ \def\@tempa{[#1][#2]{#3}}%
+ \@ifx{\@cite\NAT@citesuper}{%
+ \leavevmode
+ \skip@\lastskip
+ \unskip
+ \super@cite@let
+ }{%
+ \super@cite@end
+ }%
+}%
+\def\super@cite@let{%
+ \futurelet\@let@token\super@cite@check
+}%
+\def\super@cite@end{%
+ \aftergroup\@citex\expandafter\endgroup\@tempa
+}%
+\def\super@cite@check{%
+ \@ifx{\@let@token\@sptoken}{%
+ \super@cite@end
+ }{%
+ \super@cite@swap
+ }%
+}%
+\long\def\super@cite@swap#1{%
+ \expandafter\@ifx\expandafter{\csname rtx@automove#1\endcsname\@empty}{%
+ #1%
+ \super@cite@let
+ }{%
+ \super@cite@end
+ #1%
+ }%
+}%
+\expandafter\let\csname rtx@automove.\endcsname\@empty
+\expandafter\let\csname rtx@automove,\endcsname\@empty
+\expandafter\let\csname rtx@automove:\endcsname\@empty
+\expandafter\let\csname rtx@automove;\endcsname\@empty
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% The following must execute only after \classname{natbib} is loaded and has set
+% up its parameters (which it does at \cmd\AtBeginDocument\ time).
+% If superscript citations have been selected, and
+% if the \classoption{citeautoscript} class option has been selected,
+% we patch into \classname{natbib}'s mechanism to migrate punctuation around the
+% citation, as in class \classname{cite} with the \classoption{superscript} option.
+% \begin{macrocode}
+\appdef\class@documenthook{%
+ \citeautoscript@sw{%
+ \@ifx{\@cite\NAT@citesuper}{%
+ \let\NAT@@citetp\rtx@@citetp
+ }{}%
+ }{}%
+}%
+% \end{macrocode}
+%
+% Resolve an incompatability between \classname{natbib} and \classname{listings}.
+% The latter package tests \cmd\chapter (which has now been \cmd\let\ to \cmd\relax\ as a side effect \classname{natbib}'s use of \LaTeX's \cmd\@ifundefined).
+%
+% We couch our fix in such terms that will not be disruptive if \cmd\chapter\ is actually defined at this point.
+%
+% \changes{4.1b}{2008/07/01}{(AO) No longer need to test \cs{chapter} as of \texttt{natbib} version 8.2}
+% \begin{verbatim}
+%\@ifx{\chapter\relax}{\let\chapter\@undefined}{}%
+% \end{verbatim}
+%
+%
+% \subsubsection{\cs{endnote}s and \cs{rtx@bibnote}s}
+%
+% \begin{macro}{\mini@note}
+% \begin{macro}{\save@note}
+% \changes{4.1n}{2010/01/02}{(AO, 572) Independent footnote counter for title block. Abstract footnote counter shared with body.}%
+% QUERY: how do footnotes get thrown to the bibliography. \cmd\footinbib@sw\ appears to be irrelevant.
+% \begin{macrocode}
+\def\mini@note{\save@note\mini@notes}%Implicit #2
+\def\save@note#1#2{%
+ \stepcounter\@mpfn
+ \protected@xdef\@thefnmark{\thempfn}%
+ \@footnotemark
+ \expandafter\g@addto@macro
+ \expandafter#1%
+ \expandafter{%
+ \expandafter \@@footnotetext
+ \expandafter {\@thefnmark}{#2}%
+ }%
+}%
+\long\def\@@footnotetext#1{\def\@thefnmark{#1}\@footnotetext}%
+\let\mini@notes\@empty
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\endnote}
+% A version of footnote that appears in the bibliography, or where \cmd\printendnotes\ appears.
+% \begin{verbatim}
+%\def\@endnote{%
+% \begingroup
+% \aftergroup\@footnotemark
+% \aftergroup\@endnotetext
+% \@ifnextchar[{%
+% \@xendnote
+% }{%
+% \stepcounter{footnote}%
+% \protected@xdef\@tempa{\thefootnote}%
+% \expandafter\@xendnote\expandafter[\the\c@footnote]%
+% }%
+%}%
+% \end{verbatim}
+% \end{macro}
+%
+% \begin{macro}{\@xendnote}
+% \changes{4.1a}{2008/01/18}{(AO, 457) Endnotes to be sorted in with numerical citations.}%
+% \begin{verbatim}
+%\def\unused@xendnote[#1]{%
+% \begingroup
+% \c@footnote#1\relax
+% \end{macrocode}
+% New for 4.1
+% \begin{macrocode}
+% \unrestored@protected@xdef\@endnotelabel{Note\thefootnote}%
+% \authoryear@sw{%
+% \unrestored@protected@xdef\@thefnmark{\noexpand\ref{\@endnotelabel}}%
+% }{%
+% \unrestored@protected@xdef\@thefnmark{\@endnotelabel}%
+% }%
+% \end{macrocode}
+% Was:
+% \unrestored@protected@xdef\@thefnmark{endnote\thefootnote}%
+% End 4.1 changes
+% \begin{macrocode}
+% \endgroup
+% \endgroup
+%}%
+%\def\@endnotemark{%
+% \expandafter\cite\expandafter{\@thefnmark}%
+%}%
+% \end{verbatim}
+% \begin{macrocode}
+\def\rev@citemark#1{%
+ \expandafter\cite\expandafter{\@thefnmark}%
+}%
+\def\rev@endtext#1{%
+ \let\@endnotelabel\@thefnmark
+ \@endnotetext
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\endnote@ext}
+% \begin{macro}{\bibdata@app}
+% \begin{macro}{\bibdata@ext}
+% The macro \cmd\endnote@ext\ is the file extension for the auxiliary file holding footnotes.
+% The \cmd\bibdata@app\ and \cmd\bibdata@ext\ macros are used to form the name of a
+% Bib\TeX\ database file holding footnotes.
+% \begin{macrocode}
+\def\endnote@ext{.end}%
+\def\bibdata@app{Notes}%
+\def\bibdata@ext{bib}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\@endnotetext}
+% \changes{4.1a}{2008/01/18}{(AO, 457) Endnotes to be sorted in with numerical citations.}%
+% The procedure \cmd\@endnotetext\ writes a Bib\TeX\ .bib file for the purpose
+% of insering a footnote into the (numbered, unsorted) bibliography.
+%
+% We need to define \cmd\pre@bibdata\ to be
+% \cmd\jobname\cmd\endnote@ext, and we probably should define \cmd\endnote@ext\ to
+% be something like ``Notes.bib''.
+%
+% In each case, the material to be written out requires robustification, provided by \cmd\endnote@relax.
+% The commands \cmd\label, \cmd\index, and \cmd\glossary,
+% which are robustified for \cmd\markright\ and \cmd\addcontentsline,
+% are likewise robustified here.
+%
+% Procedure \cmd\@endnotetext@note\ is the alias for \cmd\@endnotetext\ when the endnotes are to be
+% processed separately from the bibliography (generally true when citations are not sorted).
+% \begin{verbatim}
+%\long\def\unused@endnotetext@note#1{%
+% \@ifxundefined\@endnoteout{%
+% \newwrite\@endnoteout
+% \gdef\endnote@stream{\jobname\endnote@ext}%
+% \immediate\openout\@endnoteout\endnote@stream\relax
+% }{}%
+% \begingroup
+% \endnote@relax
+% \immediate\write\@endnoteout{\string\@doendnote{\@endnotelabel}{#1}}%
+% \endgroup
+%}%
+% \end{verbatim}
+%
+% \cmd\@doendnote\ is obsolete.
+% \begin{verbatim}
+%\def\@doendnote#1#2{\bibitem{#1}#2}%
+% \end{verbatim}
+%
+% \changes{4.1o}{2010/02/05}{(AO, 579) Endnote shall comprise their own Bib\TeX\ entry type: @FOOTNOTE.}
+% Procedure \cmd\@endnotetext\ is the operative procedure\ when the endnotes are to be
+% collated in with the other references, typically true when numerical citations are being used.
+% The technique involves writing a .bib file (\cmd\@bibdataout) with each endnote typed as
+% a \texttt{@FOOTNOTE} entry.
+%
+% Timing note: doing \cmd\openout\ should be deferred until the beginning of the document,
+% as is done here. This allows one to make a format (\filename{.fmt}) file out of this class.
+% \begin{macrocode}
+\long\def\@endnotetext#1{%
+ \begingroup
+ \endnote@relax
+ \immediate\write\@bibdataout{%
+ @FOOTNOTE{%
+ \@endnotelabel,%
+% \end{macrocode}
+% The \texttt{key} field is recommended in cases where there is no author (see \filename{btxdoc}).
+% \begin{macrocode}
+ key="\@endnotelabel",%
+% \end{macrocode}
+% The \texttt{note} field is simply the content of the footnote.
+% \begin{macrocode}
+ note="#1"%
+ }%
+ }%
+ \endgroup
+}%
+\newwrite\@bibdataout
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\endnote@relax}
+% At \cmd\AtBeginDocument\ time, we open the job's \filename{.bib} file.
+%
+% Procedure \cmd\endnote@relax\ robustifies commands that ought not to be expanded when
+% the endnote is written out.
+% Note the similarity between \cmd\endnote@relax\ and \cmd\protected@write.
+% \begin{macrocode}
+\def\endnote@relax{%
+ \let\label\relax \let\index\relax \let\glossary\relax
+ \let\cite \relax \let\ref \relax \let\pageref \relax
+ \let\( \relax \let\) \relax \let\\ \relax
+ \let~\relax
+% \end{macrocode}
+% \begin{verbatim}
+%\let\protect\noexpand
+% \end{verbatim}
+% \begin{macrocode}
+ \let \protect \@unexpandable@protect
+ \newlinechar`\^^M%
+% \end{macrocode}
+% \begin{verbatim}
+%\newlinechar`\ %
+% \end{verbatim}
+% \begin{macrocode}
+ \let\begin\relax \let\end\relax
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@bibdataout@init}
+% \begin{macro}{\@bibdataout@aps}
+% \changes{4.1o}{2010/02/10}{(AO, 580) Provide a document class option to turn off production of eprint field in bibliography.}
+% \changes{4.1o}{2010/02/12}{(AO, 580) Control .bst at run time.}%
+% \changes{4.1q}{2010/04/13}{(AO, 588) Only write \revtex-specific BibTeX .bib data if the .bst style is set by REVTeX.}%
+%\changes{4.2b}{2018/12/26}{(MD) Improve control over display of e-print ids in bibliography.}%
+% At \cmd\AtBeginDocument\ time, we open the job's \filename{.bib} file.
+% The hook is available for use by a society to place its own @CONTROL record in the \cmd\@bibdataout\ stream.
+% \begin{macrocode}
+\appdef\class@documenthook{\@bibdataout@init}%
+\def\@bibdataout@init{%
+ \immediate\openout\@bibdataout\pre@bibdata.\bibdata@ext\relax
+}%
+\def\@bibdataout@rev{%
+ \immediate\write\@bibdataout{%
+% \end{macrocode}
+% The entry that controls processing of the \filename{.bst} file has entry type \texttt{@CONTROL}.
+% The citation key (REVTEX42Control) is effectively a version number,
+% which the \filename{.bst} can use to interpret the bib entry.
+% \begin{macrocode}
+ @CONTROL{%
+ REVTEX42Control%
+% \end{macrocode}
+% Say if we want the \texttt{eprint} field disabled. Otherwise accept the default of the \filename{.bst}.
+% \begin{macrocode}
+ \eprint@enable@sw{}{,eprint="1"}%
+ }%
+ }%
+% \end{macrocode}
+% Place a \cmd\citation\ into the auxiliary file corresponding to this entry.
+% \begin{macrocode}
+ \if@filesw
+ \immediate\write\@auxout{\string\citation{REVTEX42Control}}%
+ \fi
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\printendnotes}
+% \changes{4.1f}{2009/07/10}{(AO, 520) Automatically produce \cs{bibliography} command when needed}%
+% We have removed the endnotes facility from REVTeX, so the \cmd\printendnotes\ command now does nothing.
+%
+% Moving footnotes to the bibliogrphy is now accomplished through
+% the automatic generation of a job BiB\TeX\ database (called \cmd\pre@bibdata) containing the footnotes.
+% \begin{macrocode}
+\def\printendnotes{%
+ \class@warn{The \string\printendnotes\space command no longer serves any function. Please remove it from your document.}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@endnotesinbib}
+% \begin{macro}{\@endnotesinbibliography}
+% \changes{4.1a}{2008/01/18}{(AO, 457) Endnotes to be sorted in with numerical citations.}%
+% We define a function \cmd\@endnotesinbib,
+% and a variant \cmd\@endnotesinbibliography.
+% The former is invoked at the start of
+% the end processing for \enve{thebibliography};
+% the latter is a synonym.
+%
+% The procedure typesets the footnotes that are to
+% appear in the bibliography; the default is to
+% simply arrange for the footnote counter to be
+% reset at the start of the document.
+%
+% Note that this code make the assumption that
+% the counter used in \env{thebibliography} is \cmd\c@NAT@ctr.
+% \changes{4.1f}{2009/07/10}{(AO, 520) Automatically produce \cs{bibliography} command when needed}%
+%
+% Here is the sole place where \cmd\footinbib@sw\ has an effect, other code simple assigning its value.
+% If it is false, or \cmd\authoryear@sw\ is true, then footnotes are handled by the default mechanism.
+% \begin{macrocode}
+\def\make@footnote@endnote{%
+ \footinbib@sw{%
+ \authoryear@sw{}{%
+ \ltx@footnote@push
+ \def\thempfn{Note\thefootnote}%
+ \let\ltx@footmark\rev@citemark
+ \let\ltx@foottext\rev@endtext
+% \end{macrocode}
+% The endnotes facility has been removed.
+% Also, there is no need to queue up \cmd\auto@bib\ here, since it is always queued up elsewhere.
+% \changes{4.1f}{2009/07/10}{(AO, 520) Automatically produce \cs{bibliography} command when needed}%
+% \begin{verbatim}
+% \appdef\class@enddocumenthook{\auto@bib}%
+% \let\printendnotes\relax
+% \end{verbatim}
+% \begin{macrocode}
+ }%
+ }{}%
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\ltx@footnote@push{%
+ \let\ltx@footmark@latex\ltx@footmark
+ \let\ltx@foottext@latex\ltx@foottext
+ \let\thempfn@latex\thempfn
+ \def\ltx@footnote@pop{%
+ \let\ltx@footmark\ltx@footmark@latex
+ \let\ltx@foottext\ltx@foottext@latex
+ \let\thempfn\thempfn@latex
+ }%
+}%
+% \end{macrocode}
+%
+% The switchover to setting footnotes in the bibliography
+% changes the meaning of \cmd\footnote\ and
+% substitutes the synonym for \cmd\@endnotesinbib.
+%
+% We arrange for the procedure \cmd\make@footnote@endnote\ to be executed
+% at \cmd\class@documenthook\ time (we mustn't do this earlier because
+% the meaning of \cmd\@footnotemark\ must not be changed before then,
+% for the sake of \file{ltxutil.dtx}).
+% \begin{macrocode}
+\appdef\class@documenthook{%
+ \make@footnote@endnote
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\auto@bib}
+% \begin{macro}{\auto@bib@empty}
+% \begin{macro}{\test@bbl@sw}
+% \begin{macro}{\bibitem@set}
+% \begin{macro}{\auto@bib@innerbib}
+% \begin{macro}{\thebibliography@nogroup}
+% \changes{4.1f}{2009/07/10}{(AO, 520) Automatically produce \cs{bibliography} command when needed}%
+% \changes{4.1n}{2010/01/02}{(AO, 572) Independent footnote counter for title block. Abstract footnote counter shared with body.}%
+% \changes{4.1o}{2010/02/02}{(AO, 575) Automatically incorporate the (Bib\TeX-generated) .bbl into an explicit \env{thebibliography}}%
+% Under some circumstances, we must typeset the bibliography automatically.
+% If the document requires footnotes to be set in the bibliography (effectively, class option \texttt{footinbib}),
+% or that frontmatter footnotes be set in the bibliography (effectively, class option \texttt{bibnotes}),
+% but contains no explicit \cmd\bibliography\ statement.
+%
+% Note that this facility is not able to work more than once per document.
+% If multiple bibliographys are required (e.g., per article), it will be the responsibility of the journal style
+% to restore \cmd\auto@bib\ to its original meaning so it can be re-invoked.
+%
+% In procedure \cmd\auto@bib, we first test for the presence of frontmatter footnotes deferred to the bibliography.
+% If none, we further test for the presence of \cmd\bibitem\ commands in the job's \filename{.bbl} file.
+% If either condition is met, we ask for a bibliography.
+% We know that the document itself lacks a \cmd\bibliography\ statement,
+% so we know the argument of the \cmd\bibliography\ that we will issue.
+% \begin{macrocode}
+\def\auto@bib{%
+ \@ifx@empty\@FMN@list{%
+ \footinbib@sw{%
+ \@ifnum{\csname c@\@mpfn\endcsname>\z@}{%
+ \true@sw
+ }{%
+ \test@bbl@sw
+ }%
+ }{%
+ \test@bbl@sw
+ }%
+ }{%
+ \true@sw
+ }%
+ {%
+ \bibliography{}%
+ }{}%
+}%
+\def\auto@bib@empty{%
+ \let\auto@bib\@empty
+}%
+% \end{macrocode}
+% Testing the \filename{.bbl} file involves defanging all expected commands
+% and processing that file inside a box register (that will be simply discarded).
+% We provide a new meaning for the \cmd\bibitem\ command: it queues a Boolean.
+% \begin{macrocode}
+\def\test@bbl@sw{%
+ \setbox\z@\vbox\bgroup
+ \let\providecommand\providecommand@j@nk
+ \let\bibfield\@gobbletwo
+ \let\bibinfo\@gobbletwo
+ \let\translation\@gobble
+ \let\BibitemOpen\@empty
+ \let\bibitemStop\@empty
+ \let\bibitemNoStop\@empty
+ \let\EOS\@empty
+ \let\BibitemShut\@gobble
+ \let\bibAnnoteFile\@gobbletwo
+ \let\bibAnnote\@gobblethree
+ \let\textbf\@gobble
+ \let\emph\@gobble
+ \@booleanfalse\bibitem@sw
+ \let\bibitem\bibitem@set
+ \auto@bib@innerbib
+ \bibitem@sw{\aftergroup\true@sw}{\aftergroup\false@sw}%
+ \egroup
+}%
+% \end{macrocode}
+% The \cmd\bibitem@set\ is an alias for \cmd\bibitem\ for the purpose of detecting a non-trivial bibliography.
+% \begin{macrocode}
+\newcommand\bibitem@set[1][]{%
+ \bibitem@sw{}{%
+ \@booleantrue\bibitem@sw
+ \aftergroup\@booleantrue\aftergroup\bibitem@sw
+ }%
+}%
+% \end{macrocode}
+% The \cmd\auto@bib@innerbib\ procedure reads in the \filename{.bbl} file (if it exists)
+% within a context where its \env{thebibliography} environment does nothing, not even establishing a group.
+% \begin{macrocode}
+\def\auto@bib@innerbib{%
+ \begingroup
+ \let@environment{thebibliography}{thebibliography@nogroup}%
+ \bibliography{}%
+ \endgroup
+}%
+% \end{macrocode}
+% Environment \env{thebibliography@nogroup} is an alias of the \env{thebibliography} environment that cancels itself.
+% It assumes that it is called within a \env{thebibliography} environment.
+% \begin{macrocode}
+\def\thebibliography@nogroup#1{%
+ \endgroup
+ \def\@currenvir{thebibliography}%
+}%
+\def\endthebibliography@nogroup{\begingroup}%
+% \end{macrocode}
+% The following should be part of \filename{ltxutil}.
+% \begin{macrocode}
+\long\def \@gobblethree #1#2#3{}%
+\def\providecommand@j@nk#1[#2]{%
+ \@ifnum{#2=\z@}{\def\j@nk}{%
+ \@ifnum{#2=\@ne}{\def\j@nk##1}{%
+ \@ifnum{#2=\tw@}{\def\j@nk##1##2}{%
+ \@ifnum{#2=\thr@@}{\def\j@nk##1##2##3}{%
+ }%
+ }%
+ }%
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \section{Initial setup}
+%
+% The standard LaTeX document classes execute certain commands
+% that are best deferred until \cmd\class@documenthook\ time.
+% Here, we effectively split \cmd\pagenumbering\ into two halves,
+% with a default definition for \cmd\thepage
+% and an initialization of \cmd\c@page\ at \cmd\class@documenthook\ time.
+%
+% The meaning of \cmd\thepage\ can be overridden by society, journal, or
+% anywhere within the document pramble, and the counter itself will be
+% preset at the beginning of the document.
+% \begin{macrocode}
+\def\thepage{\@arabic\c@page}%
+% \end{macrocode}
+% \changes{4.0b}{1999/06/20}{Defer assignment until \cs{AtBeginDocument} time.}
+% Note that this code is executed at \cmd\setup@hook\ time
+% to allow for the possibility of overrides by packages like \classname{geometry}.
+% \begin{macrocode}
+\appdef\setup@hook{%
+ \tabbingsep \labelsep
+ \leftmargin\leftmargini
+ \labelwidth\leftmargin\advance\labelwidth-\labelsep
+ \let\@listi\@listI
+ \@listi
+}%
+% \begin{macrocode}
+%
+% We ensure that the ``environment'' component mark (implemented by \file{ltxgrid.dtx})
+% is initialized properly (via a hook, itself defined via \file{ltxutil.dtx}).
+% \begin{macrocode}
+\appdef\class@documenthook{%
+ \global\c@page\@ne
+ \def\curr@envir{document}%
+ \mark@envir{\curr@envir}%
+}%
+% \end{macrocode}
+%
+% \begin{macro}{\open@onecolumn}%
+% \begin{macro}{\open@column@two}%
+% \changes{4.0c}{1999/11/13}{Grid changes with ltxgrid}
+% \changes{4.0d}{2000/04/10}{Also alter how lists get indented.}
+% When setting the column grid, we have to override the procedure
+% for formatting lists.
+% Because \cmd\twocolumngrid\ requires rebalancing columns at some points,
+% typesetting must employ only the manipulation of
+% \cmd\leftskip\ and \cmd\rightskip, and must avoid the use of
+% \cmd\moveleft, \cmd\moveright, and \cmd\parshape.
+%
+% It is one of the stranger features of \TeX\ that these two separate
+% mechanisms exist. The latter three have the effect of adding things to the
+% Main Vertical List that cannot be removed and later added back with all their
+% properties intact.
+%
+% In detail, \cmd\moveleft, say, adds a box to the MVL with
+% its reference point shifted horizontally by some amount relative to the
+% reference point of the enclosing list. If that box is removed from the
+% MVL (via a \cmd\lastbox\ operation in the output routine), and later
+% thrown back to the MVL, the shift of the box will have been ``forgotten'' by \TeX.
+% This is a bug, but not one ``acceptible to D. E. Knuth'', so it will never be fixed.
+%
+% \begin{macrocode}
+\def\open@onecolumn{%
+ \open@column@one\@ne
+ \set@colht
+ \@floatplacement
+ \@dblfloatplacement
+}%
+\def\open@twocolumn{%
+ \open@column@mlt\tw@
+ \set@colht
+ \@floatplacement
+ \@dblfloatplacement
+ \sloppy
+ \let\set@listindent\set@listindent@
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \section{\cs{appendix}}
+%
+% \begin{macrocode}
+%\newif\ifappendixon
+% \end{macrocode}
+% Note that, within appendices,
+% equations are numbered within sections (appendices).
+% \begin{macrocode}
+\def\appendix{%
+ \par
+%\appendixontrue
+ \setcounter{section}\z@
+ \setcounter{subsection}\z@
+ \setcounter{subsubsection}\z@
+ \def\thesubsection{\arabic{subsection}}%
+ \def\thesubsubsection{\alph{subsubsection}}%
+ \@addtoreset{equation}{section}%
+ \def\theequation@prefix{\thesection}%
+ \addtocontents{toc}{\protect\appendix}%
+ \@ifstar{%
+ \def\thesection{\unskip}%
+ \def\theequation@prefix{A.}%
+ }{%
+ \def\thesection{\Alph{section}}%
+ }%
+}%
+% \end{macrocode}
+%
+% \section{Changing the page grid}
+% \changes{4.0b}{1999/06/20}{Define three separate environments, defer assignment to \cs{AtBeginDocument} time.}
+%
+% \subsection{Avoiding Grid Changes}%
+%
+% In preprint styles, ``wide text'' is a no-op, and the title page
+% processing involves no grid change.
+%
+% \begin{macro}{\title@column}%
+% \begin{macro}{\close@column}%
+% Provide default meanings for \cmd\title@column\ and \cmd\close@column,
+% in case they were never defined.
+% Note that the society or journal substyle may define
+% \cmd\title@column\ or \cmd\close@column: this code will not override.
+% \begin{macrocode}
+\def\title@column#1{%
+ \minipagefootnote@init
+ #1%
+ \minipagefootnote@foot
+}%
+\def\close@column{%
+ \newpage
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+%
+% \subsection{Galley Style: Margin Changes}%
+% A variant of preprint processing. Emulate journal appearance somewhat.
+%
+% \begin{environment}{widetext@galley}
+% DPC: We're in galley style so do a lob sided display environment.
+%
+% QUERY: How can we be sure that we are in galley style?
+% ANSWER: as noted elsewhere, require that both \cmd\twocolumn@sw\ and \cmd\preprintsty@sw\ be false.
+% \begin{macrocode}
+\def\galley@outdent{\rightmargin-\columnwidth\advance\rightmargin-\columnsep}%
+\let\widetext@outdent\@empty
+\newenvironment{widetext@galley}{%
+ \list{}{%
+ \topsep \z@skip
+ \listparindent \parindent
+ \itemindent \parindent
+ \leftmargin \z@
+ \parsep \z@\@plus\p@
+ \widetext@outdent
+ \relax
+ }%
+ \item\relax
+}{
+ \endlist
+}%
+% \end{macrocode}
+% \end{environment}
+%
+%
+% \subsection{Grid Changing Via \classname{ltxgrid}}%
+%
+% In case \classoption{twocolumngrid} has been invoked,
+% switch column grid using the column grid-changing commands.
+% Supply stub definitions of those commands here.
+% \changes{4.0c}{1999/11/13}{grid changes with ltxgrid}
+%
+% \begin{macro}{\title@column@grid}%
+% \begin{macro}{\close@column@grid}%
+% The title block always starts at the top of a new page.
+% \changes{4.0c}{1999/11/13}{grid changes with ltxgrid}
+%
+% Note that, for the procedure \cmd\close@column@grid,
+% we balance columns by switching to the one-column page grid.
+% \begin{macrocode}
+\def\title@column@grid#1{%
+ \minipagefootnote@init
+ \onecolumngrid
+ \begingroup
+ \let\@footnotetext\frontmatter@footnotetext
+%<ignore> \let\set@footnotewidth\set@footnotewidth@two
+ \ltx@no@footnote
+ #1%
+ \endgroup
+ \twocolumngrid
+ \minipagefootnote@foot
+}%
+\def\close@column@grid{%
+ \balancelastpage@sw{%
+ \onecolumngrid
+%<ignore> \twocolumngrid
+ }{}%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{environment}{widetext@grid}
+% \changes{4.0c}{1999/11/13}{grid changes via ltxgrid procedures}
+% \changes{4.0e}{2000/11/21}{adornments above and below.}
+% \changes{4.1g}{2009/10/07}{(AO, 525) Remove phantom paragraph above display math that is given in vertical mode}%
+% We slip into the one-column page grid
+% within the scope of this environment.
+%
+% Note that we set adornments above and below the \env{widettext}.
+% These are set as leaders, so they will disappear at a page break.
+% \changes{4.1n}{2010/01/02}{(AO) fine-tune spacing above and below widetext}%
+% \begin{macrocode}
+\newenvironment{widetext@grid}{%
+ \par\ignorespaces
+ \setbox\widetext@top\vbox{%
+%<ignore> \vskip15\p@
+ \hb@xt@\hsize{%
+ \leaders\hrule\hfil
+ \vrule\@height6\p@
+ }%
+%<ignore> \vskip6\p@
+ }%
+ \setbox\widetext@bot\hb@xt@\hsize{%
+ \vrule\@depth6\p@
+ \leaders\hrule\hfil
+ }%
+ \onecolumngrid
+ \vskip10\p@
+ \dimen@\ht\widetext@top\advance\dimen@\dp\widetext@top
+ \cleaders\box\widetext@top\vskip\dimen@
+%<ignore> \let\set@footnotewidth\set@footnotewidth@two
+ \vskip6\p@
+ \prep@math@patch
+}{%
+ \par
+ \vskip6\p@
+ \setbox\widetext@bot\vbox{%
+ \hb@xt@\hsize{\hfil\box\widetext@bot}%
+%<ignore> \vskip14\p@
+ }%
+ \dimen@\ht\widetext@bot\advance\dimen@\dp\widetext@bot
+ \cleaders\box\widetext@bot\vskip\dimen@
+ \vskip8.5\p@
+ \twocolumngrid\global\@ignoretrue
+ \@endpetrue
+}%
+\newbox\widetext@top
+\newbox\widetext@bot
+% \end{macrocode}
+% \end{environment}
+%
+%
+% Decide, finally, how the page grid is to be manipulated.
+% \begin{macrocode}
+\def\set@page@grid{%
+ \twocolumn@sw{%
+% \end{macrocode}
+% The following two assignments determine what procedures are to be executed when
+% the footnote set width is calculated, and how footnotes are to be composed at the bottom of the page.
+% A society or journal wishing to do otherwise will override this code.
+% \begin{macrocode}
+ \let\set@footnotewidth\set@footnotewidth@two
+ \let\compose@footnotes\compose@footnotes@two
+ \let@environment{widetext}{widetext@grid}%
+ \let\title@column\title@column@grid
+ \let\close@column\close@column@grid
+ }{%
+ \let@environment{widetext}{widetext@galley}%
+ \preprintsty@sw{%
+% \end{macrocode}
+% Change the page grid not at all.
+% \begin{macrocode}
+ }{%
+% \end{macrocode}
+% If we are galley style, change the page margin only.
+% \begin{macrocode}
+ \galley@sw{%
+ \let\widetext@outdent\galley@outdent
+ }{}%
+ }%
+ }%
+}%
+\appdef\setup@hook{\set@page@grid}%
+% \end{macrocode}
+%
+% \section{Old font commands}
+%
+% \begin{macrocode}
+\DeclareOldFontCommand{\rm}{\normalfont\rmfamily}{\mathrm}
+\DeclareOldFontCommand{\sf}{\normalfont\sffamily}{\mathsf}
+\DeclareOldFontCommand{\tt}{\normalfont\ttfamily}{\mathtt}
+\DeclareOldFontCommand{\bf}{\normalfont\bfseries}{\mathbf}
+\DeclareOldFontCommand{\it}{\normalfont\itshape}{\mathit}
+\DeclareOldFontCommand{\sl}{\normalfont\slshape}{\@nomath\sl}
+\DeclareOldFontCommand{\sc}{\normalfont\scshape}{\@nomath\sc}
+\DeclareRobustCommand*\cal{\@fontswitch\relax\mathcal}
+\DeclareRobustCommand*\mit{\@fontswitch\relax\mathnormal}
+% \end{macrocode}
+%
+% \section{English-Language Texts}
+% As this class is just for English language journals, we
+% could hardwire these texts,
+% but to make it easier to use this as a
+% basis for the code for similar journal styles, separate out all the
+% fixed text strings into babel-style macros of the form
+% |\|\ldots|name|
+%
+% Note: for babel compatability, use version 1999/05/05 v3.6x or later.
+%
+% Some of these might need changing in the society-specific code.
+%
+% \begin{macro}{\today}
+% Procedure \cmd\today\ is used in the article class, but not in
+% this document class.
+% \begin{macrocode}
+\def\today{\ifcase\month\or
+ January\or February\or March\or April\or May\or June\or
+ July\or August\or September\or October\or November\or December\fi
+ \space\number\day, \number\year}
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\notesname}
+% Text entity \cmd\notesname\ had been used in \cmd\printendnotes.
+% \changes{4.1f}{2009/07/10}{(AO, 520) Automatically produce \cs{bibliography} command when needed}%
+% However, we have removed the endnotes facility from REVTeX.
+%
+% \begin{verbatim}
+%\def\notesname{Notes}
+% \end{verbatim}
+% \end{macro}
+%
+% \begin{macro}{\partname}
+% Text entity \cmd\partname\ is used in \cmd\@part.
+% \begin{macrocode}
+\def\partname{Part}
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\tocname}
+% Text entity \cmd\tocname\ is used in \cmd\tableofcontents,
+% as defined in the standard \LaTeX\ book class.
+% \begin{macrocode}
+\def\tocname{Contents}
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\lofname}
+% Text entity \cmd\lofname\ is used in \cmd\listoffigures,
+% as defined in the standard \LaTeX\ book class.
+% \begin{macrocode}
+\def\lofname{List of Figures}
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\lotname}
+% Text entity \cmd\lotname\ is used in \cmd\listoftables,
+% as defined in the standard \LaTeX\ book class.
+% \begin{macrocode}
+\def\lotname{List of Tables}
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\refname}
+% Text entity \cmd\refname\ is used in \env{thebibliography}.
+% \begin{macrocode}
+\def\refname{References}
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\indexname}
+% Text entity \cmd\indexname\ is used in \env{theindex},
+% as defined in the standard \LaTeX\ book class.
+% \begin{macrocode}
+\def\indexname{Index}
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\figurename}
+% \changes{4.0b}{1999/06/20}{Added localization of \cs{figuresname}}
+% Text entity \cmd\figurename\ is used in \env{figure},
+% \cmd\figuresname\ in \cmd\printfigures.
+% \begin{macrocode}
+\def\figurename{FIG.}
+\def\figuresname{Figures}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\tablename}
+% \changes{4.0b}{1999/06/20}{Added localization of \cs{tablesname}}
+% Text entity \cmd\tablename\ is used in \env{table},
+% \cmd\tablesname\ in \cmd\printtables.
+% \begin{macrocode}
+\def\tablename{TABLE}
+\def\tablesname{Tables}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\abstractname}
+% Text entity \cmd\abstractname\ is used in \env{abstract}.
+% \begin{macrocode}
+\def\abstractname{Abstract}
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\appendixesname}
+% \begin{macro}{\appendixname}
+% Text entity \cmd\appendixesname\ is used in TOC.
+% \begin{macrocode}
+\def\appendixesname{Appendixes}%
+\def\appendixname{Appendix}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\acknowledgmentsname}
+% Text entity \cmd\acknowledgmentsname\ is used in \env{acknowledgments}.
+% \begin{macrocode}
+\def\acknowledgmentsname{Acknowledgments}
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\journalname}
+% This should be set by the society journal options, eg `pra'.
+% \begin{macrocode}
+\def\journalname{??}
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\copyrightname}
+% Default layout does not assign copyright, but a journal that wants
+% to might use this.
+% \begin{macrocode}
+\def\copyrightname{??}
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\andname}
+% The text string ``and'' for use in author lists.
+% \begin{macrocode}
+\def\andname{and}
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@pacs@name}
+% \begin{macro}{\@keys@name}
+% The text string prepended to PACS numbers, resp. to keywords.
+% \begin{macrocode}
+\def\@pacs@name{PACS numbers: }%
+\def\@keys@name{Keywords: }%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\ppname}
+% The text string ``pp'' for use in page ranges.
+% \begin{macrocode}
+\def\ppname{pp}
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\numbername}
+% The text string ``number'' for use in article reference.
+% \begin{macrocode}
+\def\numbername{number}
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\volumename}
+% The text string ``volume'' for use in article reference.
+% \begin{macrocode}
+\def\volumename{volume}
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\Dated@name}
+% \begin{macro}{\Received@name}
+% \begin{macro}{\Revised@name}
+% \begin{macro}{\Accepted@name}
+% \begin{macro}{\Published@name}
+% These texts are used in the \cmd\date, et al. commands.
+% \begin{macrocode}
+\def\Dated@name{Dated: }%
+\def\Received@name{Received }%
+\def\Revised@name{Revised }%
+\def\Accepted@name{Accepted }%
+\def\Published@name{Published }%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \section{Legacy Commands}
+% We define some commands left over from version 3.1, or give default meanings.
+% Some definitions can be overridden
+% in the document preamble or in included packages.
+%
+% Note on the namespace: command names like \cmd\REV@\emph{name} are used here,
+% because it is not clear that any of this code is generally useful.
+% \begin{macrocode}
+\def\address{\replace@command\address\affiliation}%
+\def\altaddress{\replace@command\altaddress\altaffiliation}%
+\newenvironment{references}{%
+ \class@warn@end{The references environment is not supported; use thebibliography instead.}
+ \gdef\references{\thebibliography{}}\references
+}{%
+ \endthebibliography
+}%
+\def\draft{%
+ \class@warn@end{Command \string\draft\space is obsolete;^^JInvoke option draft instead.}%
+ \@booleantrue\draft@sw
+}%
+\def\tighten{%
+ \class@warn@end{Command \string\tighten\space is obsolete;^^JInvoke option tightenlines instead.}%
+ \@booleantrue\tightenlines@sw
+}%
+\def\tableline{%
+ \noalign{%
+ \class@warn@end{Command \string\tableline\space is obsolete;^^JUse \string\colrule\space instead.}%
+ \global\let\tableline\colrule
+ }%
+ \tableline
+}%
+\def\case{\replace@command\case\frac}%
+\def\slantfrac{\replace@command\slantfrac\frac}%
+\def\tablenote{\replace@command\tablenote\footnote}%
+\def\tablenotemark{\replace@command\tablenotemark\footnotemark}%
+\def\tablenotetext{\replace@command\tablenotetext\footnotetext}%
+% Lose the following definition:
+\DeclareRobustCommand\REV@text[1]{%
+ \relax
+ \ifmmode
+ \mathchoice
+ {\hbox{{\everymath{\displaystyle }#1}}}%
+ {\hbox{{\everymath{\textstyle }#1}}}%
+ {\hbox{{\everymath{\scriptstyle }\let\f@size\sf@size\selectfont#1}}}%
+ {\hbox{{\everymath{\scriptscriptstyle}\let\f@size\ssf@size\selectfont#1}}}%
+ \glb@settings
+ \else
+ \mbox{#1}%
+ \fi
+}%
+% Lose the following definition:
+\DeclareRobustCommand\REV@bbox[1]{%
+ \relax
+ \ifmmode
+ \mathchoice
+ {\hbox{{\everymath{\displaystyle }\boldmath$#1$}}}%
+ {\hbox{{\everymath{\textstyle }\boldmath$#1$}}}%
+ {\hbox{{\everymath{\scriptstyle }\boldmath$#1$}}}%
+ {\hbox{{\everymath{\scriptscriptstyle}\boldmath$#1$}}}%
+ \glb@settings
+ \else
+ \mbox{#1}%
+ \fi
+}%
+\DeclareRobustCommand\REV@bm[1]{%
+ \class@warn@end{To use \string\bm, please load the bm package!}%
+ \global\let\bm\relax
+}%
+\def\FL{\obsolete@command\FL}%
+\def\FR{\obsolete@command\FR}%
+\def\narrowtext{\obsolete@command\narrowtext}%
+\def\mediumtext{\obsolete@command\mediumtext}%
+\newenvironment{quasitable}{%
+ \let@environment{tabular}{longtable}%
+}{%
+}%
+% \end{macrocode}
+%
+% \begin{macro}{\text}
+% \begin{macro}{\bm}
+% \begin{macro}{\bibinfo}
+% \begin{macro}{\eprint}
+% \begin{macro}{\url}
+%
+% If not otherwise defined, give default meanings to certain commands.
+% \changes{4.0d}{2000/04/10}{eprint takes an optional argument, syntactical only in this case.}
+%FIXME: \cmd\bibinfo?
+% \begin{macrocode}
+\let\text\REV@text
+\let\bm\REV@bm
+\appdef\setup@hook{%
+ \providecommand\bibinfo[2]{#2}%
+ \providecommand\eprint[2][]{#2}%
+%\providecommand\url[1]{#1}%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\bbox}
+%
+% \begin{macrocode}
+\def\bbox#1{%
+ \class@warn@end{\string\bbox\space is obsolete,^^Jload the bm package and use \string\bm\space instead.}%
+ \global\let\bbox\relax
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\mathletters}
+%
+% \begin{macrocode}
+\newenvironment{mathletters}{%
+ \class@warn@end{Environment {mathletters} is obsolete;^^Jload the amsmath package and use {subequations}!}%
+ \global\let\mathletters\@empty
+}{%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\eqnum}
+%
+% \begin{macrocode}
+\def\eqnum#1{%
+ \class@warn@end{\string\eqnum\space is obsolete, load the amsmath package and use \string\tag!}%
+ \global\let\eqnum\@gobble
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% We read in the symbol definitions.
+% \begin{macrocode}
+\appdef\rtx@require@packages{%
+ \RequirePackage{revsymb4-2}%
+}%
+\appdef\class@documenthook{\revsymb@inithook}%
+% \end{macrocode}
+%
+%
+% \section{Corrected Indentation for \texttt{tableofcontents}}
+% \changes{4.2a}{2014/12/31}{(MD, Aptara) Make titles in bibliography default, prb style to follow other Phys. Rev. journals, add a unified physrev option as well as prx, prapplied, prmaterials, prfluids, prab, and prper options. Corrected indentation for tableofcontents appearing along with listoffigure/listoftable.}%
+%
+% Corrected indentation for \texttt{tableofcontents}, when appearing with \texttt{listoffigure} or \texttt{listoftable}.
+%
+% \begin{macrocode}
+%%
+\def\@startflt#1{%
+ \begingroup
+ %\toc@pre
+ \makeatletter
+ \@input{\jobname.#1}%
+ \if@filesw
+ \expandafter\newwrite\csname tf@#1\endcsname
+ \immediate\openout \csname tf@#1\endcsname \jobname.#1\relax
+ \fi
+ \@nobreakfalse
+ %\toc@post
+ \endgroup
+}%
+\def\att@TOC{toc}%
+\def\print@toc#1{%
+ \begingroup
+ \expandafter\section
+ \expandafter*%
+ \expandafter{%
+ \csname#1name\endcsname
+ }%
+ \let\appendix\appendix@toc
+ \def\tempa{#1}%
+ \ifx\tempa\att@TOC%%
+ \@starttoc{#1}%
+ \else%%
+ \@startflt{#1}%%
+ \fi%%
+ \endgroup
+}%
+%%
+% \end{macrocode}
+%
+%
+% \section{Patches for lineno.sty}
+%
+% The \classname{lineno} package detects the case where the package has been loaded
+% and the document invokes \cmd\linelabel, but the \cmd\linenumbers\ command has not been
+% issued: it treats this case as an error.
+%
+% It is wrong for validity of document syntax to be dependent upon package semantics:
+% we make the condition a warning rather than an error.
+%
+% \begin{macrocode}
+\def\@LN@LLerror@org{%
+ \PackageError{lineno}{%
+ \string\linelabel\space without \string\linenumbers
+ }{%
+ Just see documentation. (New feature v4.11)%
+ }%
+ \@gobble
+}%
+\def\@LN@LLerror@ltx{%
+ \PackageWarning{lineno}{%
+ To make the \string\linelabel\space command work, you must issue the \string\linenumbers\ command
+ }%
+ \@gobble
+}%
+% \end{macrocode}
+%
+% If appropriate, enable line numbering within the abstract.
+%
+% This mechanism applies generally:
+% Create the box in a context in which the meaning of \cmd\par\ has been patched by \classname{lineno},
+% then \cmd\unvbox\ the box in a context where \cmd\set@linepenalties\ has been executed, and
+% follow up with \cmd\@linenumberpar, which forces a visit to the output routine just there.
+% Note that here, we have to de-fang \cmd\@LN@parpgbrk, which would otherwise causes the appearance
+% of a box with depth -1000 points. Go figure.
+% \begin{macrocode}
+\appdef\class@documenthook{%
+ \@ifx{\@LN@LLerror\@LN@LLerror@org}{%
+ \class@info{Overriding \string\@LN@LLerror}%
+ \let\@LN@LLerror\@LN@LLerror@ltx
+ }{}%
+ \@ifpackageloaded{lineno}{%
+ \@ifxundefined{\set@linepenalties}{}{%
+ \def\prep@absbox{\set@linepenalties}%
+ \def\post@absbox{\let\@LN@parpgbrk\@empty\@linenumberpar}%
+ }%
+ }{}%
+}%
+% \end{macrocode}
+%
+% One may well ask: how to obtain line numbering within an alignment in a float?
+% This objective, along with line numbering within footnotes, would require extraordinary measures.
+% The float would have to be thrown onto the MVL in order to acquire its line numbers, but
+% that fragment of MVL would then have to be protected from being shipped out.
+% The question of how to coordinate those lines' numbers with those of lines in the MVL would
+% also require dealing with.
+%
+% \section{Endgame for the Document Class}
+%
+% We provide for a ``job macro package'' that can override
+% definitions and assignments made by the class or any other packages it loads.
+%
+% \subsection{Job Macro Package}
+% You can create a ``job macro package'' for your document
+% that will be read in automatically every time
+% your document is processed.
+% Thus, if your job is a file called \file{myarticle.tex},
+% then the file \file{myarticle.rty} will be read in
+% just the same as if you had placed a
+% \cmd\usepackage|{myarticle.rty}| statement
+% immediately following your \cmd\documentclass\ statement.
+%
+% Within your \file{.rty} file,
+% you can define and use control sequence names that use the |@| character
+% and you can override any of the definitions or assignments made
+% by the \revtex\ document class or the selected journal substyle.
+% That is, you have the power to really mess things up badly.
+%
+% If you choose to have a job macro package, you are well
+% advised to read the \LaTeX\ guide to document classes,
+% \file{clsguide.tex} or read up on the subject
+% in a book like the \LaTeX\ Companion.
+%
+% The file \file{template.rty} contains a template for
+% creating your own job macro package.
+%
+% \begin{macrocode}
+\appdef\rtx@require@packages{%
+ \InputIfFileExists{\jobname.rty}{}{}%
+}%
+% \end{macrocode}
+%
+% \subsection{Endgame Processing for the Document Class}
+%
+% The remaining steps in processing the document class involve
+% determining the needed society, journal, and pointsize from
+% the document's class options and inputting the needed files or
+% executing the indicated procedures.
+%
+% Note that the society file is expected to declare options that will
+% allow us to determine the journal involved, and the society and journal
+% themselves determine the which pointsize options are declared, along with
+% their meanings.
+%
+% Note also that required packages are read in only after the document options
+% have been processed, because the latter can affect the former.
+%
+% Finally, the setup code is executed: this is code that depends on the
+% meanings of the switches we define and on the code within the packages we load.
+%
+% Note that there are other hooks in use: \cmd\document@inithook, which is
+% executed right at the beginning of the document, and \cmd\class@documenthook,
+% which serves as a vehicle for any \cmd\AtBeginDocument\ code we might wish to
+% execute.
+%
+%FIXME: use \cmd\class@documenthook\ only for things that bear on the MVL;
+% use \cmd\document@inithook\ for all patches to procedures defined within the preamble.
+%
+% Remember that \classname{natbib} changes its state at \cmd\AtBeginDocument\ time,
+% so we have to install our own code at a later point in the processing.
+%
+% We determine the proper \cmd\@society\ by examining the document's class options.
+% \begin{macrocode}
+\@parse@class@options@society
+% \end{macrocode}
+%
+% Then, we input the society's substyle
+% (which may in turn lead to loading a journal substyle or a pointsize substyle).
+% The substyle should not assume the value of any class option:
+% instead, it should install code into \cmd\setup@hook.
+% \begin{macrocode}
+\@process@society{aps}%
+% \end{macrocode}
+%
+% Now that the society has defined the class options relating to journals,
+% and has defined \cmd\@journal@default, we can process the journal substyle.
+% We parse the options for one that sets \cmd\@journal.
+% \begin{macrocode}
+\@parse@class@options@\@journal
+% \end{macrocode}
+% And we process the journal.
+% Note that \emph{it is an error} for a society file to fail to define
+% \cmd\@journal@default.
+%
+% \begin{macrocode}
+\expandafter\@process@journal\expandafter{\@journal@default}%
+% \end{macrocode}
+%
+% Now that the society and journal have finished defining any options relating to
+% point size, we process the class options for any that set \cmd\@pointsize.
+% \begin{macrocode}
+\@parse@class@options@\@pointsize
+% \end{macrocode}
+% And we process the pointsize.
+% Note that it is an error for the society and journal to leave \cmd\@pointsize@default\
+% undefined at this point, however, the journal may have overriden the assignment of the society.
+%
+% \begin{macrocode}
+\expandafter\@process@pointsize\expandafter{\@pointsize@default}%
+% \end{macrocode}
+%
+% Next, we process the class options for once and all.
+% Doing so sets values for some of the Booleans that were introduced along with
+% the \cmd\DeclareOption\ statements above.
+%
+%CHANGE: We process the options in the order declared in the document; this
+% gives the document greater control.
+%
+% \begin{macrocode}
+\@options
+% \end{macrocode}
+%
+% Now that the class options have been processed, we can load all the
+% packages that we know need loading.
+% \begin{macrocode}
+\rtx@require@packages
+% \end{macrocode}
+%
+% At this point, the society substyle, the journal substyle, and the
+% pointsize have all been processed, along with the document class options.
+% Some of these have left things for later; we do these now.
+%
+% \begin{macro}{\setup@hook}
+%
+% This portion of the code for this class file
+% \emph{must} appear at the very end:
+% The procedure
+% \cmd\setup@hook\
+% should be executed at the very end of the class file.
+% Any code that relies on the value of any of the \texttt{@sw} switches
+% or will patch the code of one of the required packages
+% should be executed here.
+%\changes{4.2a}{2017/11/28}{(MD) Add call to \normalsize to be a good citizen and allow booktabs.sty to work properly}
+%
+% \begin{macrocode}
+\appdef\setup@hook{\normalsize}%
+\setup@hook
+% \end{macrocode}
+% \end{macro}
+%
+%
+% Warn if past maturation date. This code to be enabled only in beta software.
+% \changes{4.1f}{2009/07/11}{(AO, 522) Warn if software is expired}%
+% \changes{4.1h}{2009/10/09}{(AO) Remove expiry code in the release software}%
+% \begin{macrocode}
+%<*ignore>
+ \def\rtx@fin@year{2010}%
+ \def\rtx@fin@month{01}%
+ \def\rtx@fin@day{01}%
+ \def\rtx@fin@warn{%
+ \@ifnum{\rtx@fin@year>\the\year\relax}{\true@sw}{%
+ \@ifnum{\rtx@fin@month>\the\month\relax}{\true@sw}{%
+ \@ifnum{\rtx@fin@day>\the\day\relax}{\true@sw}{%
+ \false@sw
+ }%
+ }%
+ }%
+ {%
+ \class@info{Beta software expires \rtx@fin@year-\rtx@fin@month-\rtx@fin@day; updates available at https://journals.aps.org/revtex/}%
+ }{%
+ \class@warn{Outdated software expired \rtx@fin@year-\rtx@fin@month-\rtx@fin@day; please retrieve an update at https://journals.aps.org/revtex/}%
+ }%
+ }%
+%</ignore>
+% \end{macrocode}
+% In shipping (non-beta) software, the following line should be commented out.
+% \changes{4.1f}{2009/07/11}{(AO, 522) Warn if software is expired}%
+% \begin{verbatim}
+%\appdef\class@enddocumenthook{\rtx@fin@warn}%
+% \end{verbatim}
+%
+% End of the class file.
+% \begin{macrocode}
+%</package>
+% \end{macrocode}
+%
+%
+% \section{Symbols: the \texttt{revsymb} module}
+% We immediately define a utility command: this module's warning.
+% \begin{macrocode}
+%<*revsymb>
+\def\REVSYMB@warn#1{\PackageWarningNoLine{revsymb}{#1}}%
+% \end{macrocode}
+%
+% \begin{macro}{\lambdabar}
+% \begin{macrocode}
+\DeclareRobustCommand\lambdabar{%
+ \bgroup
+ \def\@tempa{%
+ \hbox{%
+ \raise.73\ht\z@
+ \hb@xt@\z@{%
+ \kern.25\wd\z@
+ \vrule \@width.5\wd\z@\@height.1\p@\@depth.1\p@
+ \hss
+ }%
+ \box\z@
+ }%
+ }%
+ \mathchoice
+ {\setbox\z@\hbox{$\displaystyle \lambda$}\@tempa}%
+ {\setbox\z@\hbox{$\textstyle \lambda$}\@tempa}%
+ {\setbox\z@\hbox{$\scriptstyle \lambda$}\@tempa}%
+ {\setbox\z@\hbox{$\scriptscriptstyle\lambda$}\@tempa}%
+ \egroup
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\openone}
+% DPC: Really should use a font that includes this glyph.
+% Unfortunately not in AMS ones, but is in bbold, cmbb.
+% (I think, must check),
+% \changes{4.0a}{1998/01/31}{use font-dependent spacing}
+% FIXME: check for bbold.
+% \begin{macrocode}
+\DeclareRobustCommand\openone{\leavevmode\hbox{\small1\normalsize\kern-.33em1}}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\corresponds}
+% \begin{macro}{\overdots}
+% \begin{macro}{\overcirc}
+% J\"org Knappen suggests the replacements:
+% replace \cmd\corresponds\ with \cmd\triangleq, source \classname{amssymb};
+% replace \cmd\overcirc\ with \cmd\mathring, source \classname{latex2e};
+% replace \cmd\overdots\ with \cmd\dddot, source \classname{amsmath}.
+%
+% Any use of any of these commands will result in a warning message at the end
+% of the log file. If the corresponding package is not loaded, a
+% definition will quietly be provided.
+% \begin{macrocode}
+\DeclareRobustCommand\corresponds{\replace@command\corresponds\triangleq}%
+\DeclareRobustCommand\overcirc{\replace@command\overcirc\mathring}%
+\DeclareRobustCommand\overdots{\replace@command\overdots\dddot}%
+\DeclareRobustCommand\REV@triangleq{%
+ {\lower.2ex\hbox{=}}{\kern-.75em^\triangle}%
+}%
+\DeclareRobustCommand\REV@dddot[1]{%
+ \@ontopof{#1}{\cdots}{1.0}\mathord{\box2}%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\succsim}
+% \begin{macro}{\precsim}
+% \begin{macro}{\lesssim}
+% \begin{macro}{\gtrsim}
+% \begin{macro}{\alt}
+% \begin{macro}{\agt}
+% These version 3.1 commands are always supplied,
+% but the definitions in \classname{amssymb} are preferred.
+% \begin{macrocode}
+\DeclareRobustCommand\altsuccsim{\succ\kern-.9em_\sim\kern.3em}%
+\DeclareRobustCommand\altprecsim{\prec\kern-1em_\sim\kern.3em}%
+\let\REV@succsim\altsuccsim
+\let\REV@precsim\altprecsim
+\DeclareRobustCommand\REV@lesssim{\mathrel{\mathpalette\vereq{<}}}%
+\DeclareRobustCommand\REV@gtrsim{\mathrel{\mathpalette\vereq{>}}}%
+\DeclareRobustCommand\alt{\lesssim}
+\DeclareRobustCommand\agt{\gtrsim}
+\def\vereq#1#2{%
+ \lower3\p@\vbox{%
+ \baselineskip1.5\p@
+ \lineskip1.5\p@
+ \ialign{$\m@th#1\hfill##\hfil$\crcr#2\crcr\sim\crcr}%
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\tensor}
+% \begin{macro}{\overstar}
+% \begin{macro}{\loarrow}
+% \begin{macro}{\roarrow}
+% \changes{4.0b}{1999/06/20}{AO: remove duplicates}
+% \begin{macrocode}
+\DeclareRobustCommand\tensor[1]{\@ontopof{#1}{\leftrightarrow}{1.15}\mathord{\box2}}
+\DeclareRobustCommand\overstar[1]{\@ontopof{#1}{\ast}{1.15}\mathord{\box2}}
+\DeclareRobustCommand\loarrow[1]{\@ontopof{#1}{\leftarrow}{1.15}\mathord{\box2}}
+\DeclareRobustCommand\roarrow[1]{\@ontopof{#1}{\rightarrow}{1.15}\mathord{\box2}}
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\@ontopof}
+% \begin{macrocode}
+\def\@ontopof#1#2#3{%
+ {%
+ \mathchoice
+ {\@@ontopof{#1}{#2}{#3}\displaystyle \scriptstyle }%
+ {\@@ontopof{#1}{#2}{#3}\textstyle \scriptstyle }%
+ {\@@ontopof{#1}{#2}{#3}\scriptstyle \scriptscriptstyle}%
+ {\@@ontopof{#1}{#2}{#3}\scriptscriptstyle\scriptscriptstyle}%
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@@ontopof}
+% Same as \revtex3, more or less.
+% \begin{macrocode}
+\def\@@ontopof#1#2#3#4#5{%
+ \setbox\z@\hbox{$#4#1$}%
+ \setbox\f@ur\hbox{$#5#2$}%
+ \setbox\tw@\null\ht\tw@\ht\z@ \dp\tw@\dp\z@
+ \@ifdim{\wd\z@>\wd\f@ur}{%
+ \setbox\f@ur\hb@xt@\wd\z@{\hss\box\f@ur\hss}%
+ \mathord{\rlap{\raise#3\ht\z@\box\f@ur}\box\z@}%
+ }{%
+ \setbox\f@ur\hb@xt@.9\wd\f@ur{\hss\box\f@ur\hss}%
+ \setbox\z@\hb@xt@\wd\f@ur{\hss$#4\relax#1$\hss}%
+ \mathord{\rlap{\copy\z@}\raise#3\ht\z@\box\f@ur}%
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frak}
+% Deal with legacy \cmd\frak:
+% if \classname{amsfonts} not loaded, defined in such a way as to ask for that package.
+% Also, says to use \cmd\mathfrak\ instead.
+% \begin{macrocode}
+\DeclareRobustCommand\frak{%
+ \REVSYMB@warn{%
+ Command \string\frak\space unsupported:^^J%
+ please use \string\mathfrak\space instead.%
+ }%
+ \global\let\frak\mathfrak
+ \frak
+}%
+\DeclareRobustCommand\REV@mathfrak{%
+ \REVSYMB@warn{%
+ Command \string\mathfrak\space undefined:^^J%
+ please specify the amsfonts or amssymb option!%
+ }%
+ \global\let\mathfrak\@firstofone
+ \mathfrak
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\Bbb}
+% Deal with legacy \cmd\Bbb:
+% if \classname{amsfonts} not loaded, defined in such a way as to ask for that package.
+% Also, says to use \cmd\mathbb\ instead.
+% \begin{macrocode}
+\DeclareRobustCommand\Bbb{%
+ \REVSYMB@warn{%
+ Command \string\Bbb\space unsupported:^^J%
+ please use \string\mathbb\space instead.%
+ }%
+ \global\let\Bbb\mathbb
+ \Bbb
+}%
+\DeclareRobustCommand\REV@mathfrak{%
+ \REVSYMB@warn{%
+ Command \string\mathbb\space undefined:^^J%
+ please specify the amsfonts or amssymb option!%
+ }%
+ \global\let\mathbb\@firstofone
+ \mathbb
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\Bigglb}
+% Deal with legacy bold delimiters.
+% Each of the following takes an implicit argument consisting of
+% the delimiter to be made big and bold.
+% FIXME: \cmd\DeclareBoldMathCommand\ is not the right tool!
+% \begin{macrocode}
+\def\Bigglb{\REV@boldopen \Bigg}%
+\def\Biglb {\REV@boldopen \Big }%
+\def\bigglb{\REV@boldopen \bigg}%
+\def\biglb {\REV@boldopen \big }%
+\def\Biggrb{\REV@boldclose\Bigg}%
+\def\Bigrb {\REV@boldclose\Big }%
+\def\biggrb{\REV@boldclose\bigg}%
+\def\bigrb {\REV@boldclose\big }%
+\def\REV@pmb#1{%
+ \hbox{%
+ \setbox\z@=\hbox{#1}%
+ \kern-.02em\copy\z@\kern-\wd\z@
+ \kern .04em\copy\z@\kern-\wd\z@
+ \kern-.02em
+ \raise.04em\copy\z@
+ }%
+}%
+\def\REV@boldopen #1#2{\mathopen {\REV@pmb{$#1#2$}}}%
+\def\REV@boldclose#1#2{\mathclose{\REV@pmb{$#1#2$}}}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\revsymb@inithook}
+% \changes{4.1b}{2008/07/01}{(AO) make settings at class time instead of deferring them to later.}
+% Package dependencies are taken care of at \cmd\setup@hook\ time.
+% \begin{macrocode}
+\def\revsymb@inithook{%
+ \@ifxundefined\dddot{\let\dddot\REV@dddot}{}%
+ \@ifxundefined\triangleq{\let\triangleq\REV@triangleq}{}%
+ \@ifxundefined\succsim{\let\succsim\altsuccsim}{}%
+ \@ifxundefined\precsim{\let\precsim\altprecsim}{}%
+ \@ifxundefined\lesssim{\let\lesssim\REV@lesssim}{}%
+ \@ifxundefined\gtrsim {\let\gtrsim \REV@gtrsim }{}%
+ \@ifxundefined\mathfrak{\let\mathfrak\REV@mathfrak}{}%
+ \@ifxundefined\mathbb{\let\mathbb\REV@mathbb}{}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macrocode}
+%</revsymb>
+% \end{macrocode}
+%
+% \section{The \classoption{10pt} class option: the \texttt{10pt} module}
+% \changes{4.0b}{1999/06/20}{AO: all code for \protect\classoption{10pt} is in this module.}
+% The file \file{aps10pt.rtx} is read in by the \classname{revtex4} document class
+% if \cmd\@pointsize\ has the value 10.
+%
+% \begin{macrocode}
+%<*10pt>
+% \end{macrocode}
+%
+% \subsection{Defend Against Forseeable Errors}%
+% Protect this file from being read in by anything but \revtex.
+% \begin{macrocode}
+\ifx\undefined\substyle@ext
+ \def\@tempa{%
+ \endinput
+ \GenericWarning{I must be read in by REVTeX! (Bailing out)}%
+ }%
+ \expandafter\else
+ \def\@tempa{}%
+ \expandafter\fi\@tempa
+ \class@info{RevTeX pointsize 10pt selected}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\normalsize{%
+ \@setfontsize\normalsize\@xpt{11.5}%
+ \abovedisplayskip 10\p@ \@plus2\p@ \@minus5\p@
+ \belowdisplayskip \abovedisplayskip
+ \abovedisplayshortskip \abovedisplayskip
+ \belowdisplayshortskip \abovedisplayskip
+ \let\@listi\@listI
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\small{%
+ \@setfontsize\small\@ixpt{10.5}%
+ \abovedisplayskip 8.5\p@ \@plus3\p@ \@minus4\p@
+ \belowdisplayskip \abovedisplayskip
+ \abovedisplayshortskip \z@ \@plus2\p@
+ \belowdisplayshortskip 4\p@ \@plus2\p@ \@minus2\p@
+ \def\@listi{%
+ \leftmargin\leftmargini
+ \topsep 4\p@ \@plus2\p@ \@minus2\p@
+ \parsep 2\p@ \@plus\p@ \@minus\p@
+ \itemsep \parsep
+ }%
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\footnotesize{%
+ \@setfontsize\footnotesize\@viiipt{9.5pt}%
+ \abovedisplayskip 6\p@ \@plus2\p@ \@minus4\p@
+ \belowdisplayskip \abovedisplayskip
+ \abovedisplayshortskip \z@ \@plus\p@
+ \belowdisplayshortskip 3\p@ \@plus\p@ \@minus2\p@
+ \def\@listi{%
+ \leftmargin\leftmargini
+ \topsep 3\p@ \@plus\p@ \@minus\p@
+ \parsep 2\p@ \@plus\p@ \@minus\p@
+ \itemsep \parsep
+ }%
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\scriptsize{%
+ \@setfontsize\scriptsize\@viipt\@viiipt
+}%
+\def\tiny{%
+ \@setfontsize\tiny\@vpt\@vipt
+}%
+\def\large{%
+ \@setfontsize\large\@xiipt{14pt}%
+}%
+\def\Large{%
+ \@setfontsize\Large\@xivpt{18pt}%
+}%
+\def\LARGE{%
+ \@setfontsize\LARGE\@xviipt{22pt}%
+}%
+\def\huge{%
+ \@setfontsize\huge\@xxpt{25pt}%
+}%
+\def\Huge{%
+ \@setfontsize\Huge\@xxvpt{30pt}%
+}%
+% \end{macrocode}
+%
+% The values of these margin parameters are dependent upon
+% \cmd\twoside@sw; any society or journal that
+% has its own preferences should override these assignments
+% by doing \cmd\appdef\cmd\setup@hook.
+% \begin{macrocode}
+\appdef\setup@hook{%
+ \twoside@sw{%
+% \oddsidemargin -.1in
+% \evensidemargin -.4in
+ \oddsidemargin -20pt
+ \evensidemargin -20pt
+ \marginparwidth 107pt
+ }{%
+ \oddsidemargin -.25in
+ \evensidemargin -.25in
+ \marginparwidth 30pt
+ }%
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\marginparsep 6pt
+% \end{macrocode}
+%
+% \begin{macrocode}
+\topmargin -61pt
+% \end{macrocode}
+%
+% \begin{macrocode}
+\headheight 25pt
+\headsep 16pt
+% \end{macrocode}
+%
+% \changes{4.1n}{2010/01/02}{(AO, 571) class file must set \cs{splittopskip}; fine tune \cs{skip}\cs{footins}; \cs{footnoterule} defined in terms of \cs{skip}\cs{footins}}%
+% \begin{macrocode}
+\topskip 10pt
+\splittopskip\topskip
+% \end{macrocode}
+%
+% \begin{macrocode}
+\footskip 30pt
+% \end{macrocode}
+%
+% \begin{macrocode}
+ \textheight = 56pc
+% \end{macrocode}
+%
+% \begin{macrocode}
+\textwidth42.5pc
+% \end{macrocode}
+%
+% \begin{macrocode}
+\columnsep 1.5pc
+\columnseprule 0pt
+% \end{macrocode}
+%
+% \changes{4.1n}{2010/01/02}{(AO, 571) class file must set \cs{splittopskip}; fine tune \cs{skip}\cs{footins}; \cs{footnoterule} defined in terms of \cs{skip}\cs{footins}}%
+% \begin{macrocode}
+\footnotesep 1pt
+\skip\footins 39pt plus 4pt minus 12pt
+\def\footnoterule{%
+ \dimen@\skip\footins\divide\dimen@\tw@
+ \kern-\dimen@\hrule width.5in\kern\dimen@
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\floatsep 12pt plus 2pt minus 2pt
+\textfloatsep 20pt plus 2pt minus 4pt
+\intextsep 12pt plus 2pt minus 2pt
+% \end{macrocode}
+%
+% \begin{macrocode}
+\dblfloatsep 12pt plus 2pt minus 2pt
+\dbltextfloatsep 20pt plus 2pt minus 4pt
+% \end{macrocode}
+%
+% \begin{macrocode}
+\@fptop 0pt plus 1fil
+\@fpsep 8pt plus 2fil
+\@fpbot 0pt plus 1fil
+\@dblfptop 0pt plus 1fil
+\@dblfpsep 8pt plus 2fil
+\@dblfpbot 0pt plus 1fil
+% \end{macrocode}
+%
+% \begin{macrocode}
+\marginparpush 5pt
+% \end{macrocode}
+%
+% \begin{macrocode}
+\parskip 0pt plus 1pt
+\parindent 10pt
+\emergencystretch8\p@
+% \end{macrocode}
+%
+% \begin{macrocode}
+\partopsep 2pt plus 1pt minus 1pt
+% \end{macrocode}
+%
+% \begin{macrocode}
+\leftmargini 25pt
+\leftmarginii 22pt
+\leftmarginiii 18.7pt
+\leftmarginiv 17pt
+\leftmarginv 10pt
+\leftmarginvi 10pt
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\@listI{%
+ \leftmargin\leftmargini
+ \parsep 4\p@ plus2\p@ minus\p@
+ \topsep 8\p@ plus2\p@ minus4\p@
+ \itemsep 4\p@ plus2\p@ minus\p@
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\labelsep 4pt
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\@listii{%
+ \leftmargin\leftmarginii
+ \labelwidth\leftmarginii
+ \advance\labelwidth-\labelsep
+ \topsep 4\p@ plus2\p@ minus\p@
+ \parsep 2\p@ plus\p@ minus\p@
+ \itemsep \parsep
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\@listiii{%
+ \leftmargin\leftmarginiii
+ \labelwidth\leftmarginiii
+ \advance\labelwidth-\labelsep
+ \topsep 2\p@ plus\p@ minus\p@
+ \parsep \z@
+ \partopsep \p@ plus\z@ minus\p@
+ \itemsep \topsep
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\@listiv{%
+ \leftmargin\leftmarginiv
+ \labelwidth\leftmarginiv
+ \advance\labelwidth-\labelsep
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\@listv{%
+ \leftmargin\leftmarginv
+ \labelwidth\leftmarginv
+ \advance\labelwidth-\labelsep
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\@listvi{%
+ \leftmargin\leftmarginvi
+ \labelwidth\leftmarginvi
+ \advance\labelwidth-\labelsep
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+%</10pt>
+% \end{macrocode}
+%
+% \section{The \classoption{11pt} class option: the \texttt{11pt} module}
+% \changes{4.0b}{1999/06/20}{AO: all code for \protect\classoption{11pt} is in this module.}
+% The file \file{11pt.rtx} is read in by the \classname{revtex4} document class
+% if \cmd\@pointsize\ has the value 11.
+%
+% \begin{macrocode}
+%<*11pt>
+% \end{macrocode}
+%
+% \subsection{Defend Against Forseeable Errors}%
+% Protect this file from being read in by anything but \revtex.
+% \begin{macrocode}
+\ifx\undefined\substyle@ext
+ \def\@tempa{%
+ \endinput
+ \GenericWarning{I must be read in by REVTeX! (Bailing out)}%
+ }%
+ \expandafter\else
+ \def\@tempa{}%
+ \expandafter\fi\@tempa
+ \class@info{RevTeX pointsize 11pt selected}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\normalsize{%
+ \@setfontsize\normalsize\@xipt{13.6}%
+ \abovedisplayskip 11\p@ \@plus3\p@ \@minus6\p@
+ \belowdisplayskip \abovedisplayskip
+ \abovedisplayshortskip \abovedisplayskip
+ \belowdisplayshortskip \abovedisplayskip
+ \let\@listi\@listI
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\small{%
+ \@setfontsize\small\@xpt\@xiipt
+ \abovedisplayskip 10\p@ \@plus2\p@ \@minus5\p@
+ \abovedisplayshortskip \z@ \@plus3\p@
+ \belowdisplayshortskip 6\p@ \@plus3\p@ \@minus3\p@
+ \def\@listi{\leftmargin\leftmargini
+ \topsep 6\p@ \@plus2\p@ \@minus2\p@
+ \parsep 3\p@ \@plus2\p@ \@minus\p@
+ \itemsep \parsep
+ }%
+ \belowdisplayskip \abovedisplayskip
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\footnotesize{%
+ \@setfontsize\footnotesize\@ixpt{11}%
+ \abovedisplayskip 8\p@ \@plus2\p@ \@minus4\p@
+ \abovedisplayshortskip \z@ \@plus\p@
+ \belowdisplayshortskip 4\p@ \@plus2\p@ \@minus2\p@
+ \def\@listi{\leftmargin\leftmargini
+ \topsep 4\p@ \@plus2\p@ \@minus2\p@
+ \parsep 2\p@ \@plus\p@ \@minus\p@
+ \itemsep \parsep
+ }%
+ \belowdisplayskip \abovedisplayskip
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\scriptsize{%
+ \@setfontsize\scriptsize\@viiipt{9.5}%
+}%
+\def\tiny{%
+ \@setfontsize\tiny\@vipt\@viipt
+}%
+\def\large{%
+ \@setfontsize\large\@xiipt{14}%
+}%
+\def\Large{%
+ \@setfontsize\Large\@xivpt{18}%
+}%
+\def\LARGE{%
+ \@setfontsize\LARGE\@xviipt{22}%
+}%
+\def\huge{%
+ \@setfontsize\huge\@xxpt{25pt}%
+}%
+\def\Huge{%
+ \@setfontsize\Huge\@xxvpt{30pt}%
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+%</11pt>
+% \end{macrocode}
+%
+% \section{The \classoption{12pt} class option: the \texttt{12pt} module}
+% \changes{4.0b}{1999/06/20}{AO: all code for \protect\classoption{12pt} is in this module.}
+% The file \file{12pt.rtx} is read in by the \classname{revtex4} document class
+% if \cmd\@pointsize\ has the value 12.
+%
+% \begin{macrocode}
+%<*12pt>
+% \end{macrocode}
+%
+% \subsection{Defend Against Forseeable Errors}%
+% Protect this file from being read in by anything but \revtex.
+% \begin{macrocode}
+\ifx\undefined\substyle@ext
+ \def\@tempa{%
+ \endinput
+ \GenericWarning{I must be read in by REVTeX! (Bailing out)}%
+ }%
+ \expandafter\else
+ \def\@tempa{}%
+ \expandafter\fi\@tempa
+ \class@info{RevTeX pointsize 12pt selected}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\normalsize{%
+ \@setfontsize\normalsize\@xiipt{14pt}%
+ \abovedisplayskip 12\p@ \@plus3\p@ \@minus7\p@
+ \belowdisplayskip \abovedisplayskip
+ \abovedisplayshortskip \z@ plus3\p@
+ \belowdisplayshortskip 6.5\p@ \@plus3.5\p@ \@minus3\p@
+ \let\@listi\@listI
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+ \def\small{%
+ \@setfontsize\small\@xipt{14.5pt}%
+ \abovedisplayskip 8\p@ \@plus3\p@ \@minus6\p@
+ \belowdisplayskip \abovedisplayskip
+ \abovedisplayshortskip \z@ \@plus3\p@
+ \belowdisplayshortskip 6.5\p@ \@plus3.5\p@ \@minus3\p@
+ \def\@listi{%
+ \leftmargin\leftmargini
+ \topsep 9\p@ \@plus3\p@ \@minus5\p@
+ \parsep 4.5\p@ \@plus2\p@ \@minus\p@
+ \itemsep \parsep
+ }%
+}%
+% \end{macrocode}
+%
+% Same baselineskip as \cmd\small\ ?
+% \begin{macrocode}
+ \def\footnotesize{%
+ \@setfontsize\footnotesize\@xpt{14.5pt}%
+ \abovedisplayskip 10\p@ \@plus2\p@ \@minus5\p@
+ \belowdisplayskip \abovedisplayskip
+ \abovedisplayshortskip \z@ \@plus3\p@
+ \belowdisplayshortskip 6\p@ \@plus3\p@ \@minus3\p@
+ \def\@listi{%
+ \leftmargin\leftmargini
+ \topsep 6\p@ \@plus2\p@ \@minus2\p@
+ \parsep 3\p@ \@plus2\p@ \@minus\p@
+ \itemsep \parsep
+ }%
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\scriptsize{%
+ \@setfontsize\scriptsize\@viiipt{9.5pt}%
+}%
+\def\tiny{%
+ \@setfontsize\tiny\@vipt{7pt}%
+}%
+\def\large{%
+ \@setfontsize\large\@xivpt{18pt}%
+}%
+\def\Large{%
+ \@setfontsize\Large\@xviipt{22pt}%
+}%
+\def\LARGE{%
+ \@setfontsize\LARGE\@xxpt{25pt}%
+}%
+\def\huge{%
+ \@setfontsize\huge\@xxvpt{30pt}%
+}%
+\let\Huge=\huge
+% \end{macrocode}
+%
+% \begin{macrocode}
+%</12pt>
+% \end{macrocode}
+%
+% \section{Page parameters}%
+% This code is common to both \classoption{11pt} and \classoption{12pt}.
+%
+% \begin{macrocode}
+%<*11pt|12pt>
+% \end{macrocode}
+%
+% \begin{macrocode}
+\appdef\setup@hook{%
+ \twoside@sw{%
+ \oddsidemargin 0pt
+ \evensidemargin 0pt
+ \marginparwidth 60pt
+ }{%
+ \oddsidemargin 0pt
+ \evensidemargin 0pt
+ \marginparwidth 44pt
+ }%
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\marginparsep 10pt
+% \end{macrocode}
+%
+% \begin{macrocode}
+\topmargin -37pt
+% \end{macrocode}
+%
+% \begin{macrocode}
+\headheight 12pt
+\headsep 25pt
+% \end{macrocode}
+%
+% \changes{4.1n}{2010/01/02}{(AO, 571) class file must set \cs{splittopskip}; fine tune \cs{skip}\cs{footins}; \cs{footnoterule} defined in terms of \cs{skip}\cs{footins}}%
+% \begin{macrocode}
+\topskip 10pt
+\splittopskip\topskip
+% \end{macrocode}
+%
+% \begin{macrocode}
+\footskip 30pt
+% \end{macrocode}
+%
+% \begin{macrocode}
+\textheight=665.5\p@
+% \end{macrocode}
+%
+% \begin{macrocode}
+\appdef\setup@hook{%
+ \tightenlines@sw{%
+ \def\baselinestretch{1}%
+ }{%
+ \def\baselinestretch{1.5}%
+ }%
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\textwidth 468pt
+% \end{macrocode}
+%
+% \begin{macrocode}
+\columnsep 10pt
+\columnseprule 0pt
+% \end{macrocode}
+%
+% \changes{4.1n}{2010/01/02}{(AO, 571) class file must set \cs{splittopskip}; fine tune \cs{skip}\cs{footins}; \cs{footnoterule} defined in terms of \cs{skip}\cs{footins}}%
+% \begin{macrocode}
+\footnotesep 1pt
+\skip\footins 25.25pt plus 4pt minus 12pt
+\def\footnoterule{%
+ \dimen@\skip\footins\divide\dimen@\f@ur
+ \kern-\dimen@\hrule width.5in\kern\dimen@
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\floatsep 14pt plus 2pt minus 4pt
+\textfloatsep 20pt plus 2pt minus 4pt
+\intextsep 14pt plus 4pt minus 4pt
+% \end{macrocode}
+%
+% \begin{macrocode}
+\dblfloatsep 14pt plus 2pt minus 4pt
+\dbltextfloatsep 20pt plus 2pt minus 4pt
+% \end{macrocode}
+%
+% \begin{macrocode}
+\@fptop 0pt plus 1fil
+\@fpsep 10pt plus 2fil
+\@fpbot 0pt plus 1fil
+\@dblfptop 0pt plus 1fil
+\@dblfpsep 10pt plus 2fil%
+\@dblfpbot 0pt plus 1fil
+% \end{macrocode}
+%
+% \begin{macrocode}
+\marginparpush 7pt
+% \end{macrocode}
+%
+% \begin{macrocode}
+\parskip 0pt plus 1pt
+\parindent 15pt
+\emergencystretch8\p@
+\partopsep 3pt plus 2pt minus 2pt
+% \end{macrocode}
+%
+% \begin{macrocode}
+\leftmargini 30pt
+\leftmarginii 26pt
+\leftmarginiii 22pt
+\leftmarginiv 20pt
+\leftmarginv 12pt
+\leftmarginvi 12pt
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\@listI{\leftmargin\leftmargini \parsep 5\p@ plus2.5\p@ minus\p@
+ \topsep 10\p@ plus4\p@ minus6\p@
+ \itemsep 5\p@ plus2.5\p@ minus\p@
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\labelsep 6pt
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\@listii{\leftmargin\leftmarginii
+ \labelwidth\leftmarginii\advance\labelwidth-\labelsep
+ \topsep 5\p@ plus2.5\p@ minus\p@
+ \parsep 2.5\p@ plus\p@ minus\p@
+ \itemsep \parsep
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\@listiii{\leftmargin\leftmarginiii
+ \labelwidth\leftmarginiii\advance\labelwidth-\labelsep
+ \topsep 2.5\p@ plus\p@ minus\p@
+ \parsep \z@ \partopsep \p@ plus\z@ minus\p@
+ \itemsep \topsep
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\@listiv{\leftmargin\leftmarginiv
+ \labelwidth\leftmarginiv\advance\labelwidth-\labelsep
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\@listv{\leftmargin\leftmarginv
+ \labelwidth\leftmarginv\advance\labelwidth-\labelsep
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\@listvi{\leftmargin\leftmarginvi
+ \labelwidth\leftmarginvi\advance\labelwidth-\labelsep
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+%</11pt|12pt>
+% \end{macrocode}
+%
+% \section{The \classoption{aps} class extension: the \texttt{aps} module}
+% \changes{4.0b}{1999/06/20}{AO: made aps.rtx part of revtex4.dtx}
+% The file \file{aps.rtx} is read in by the \classname{revtex4} document class
+% if \cmd\@society\ has the value \classoption{aps}.
+%
+% Here, code specific to APS journals is separated out from the \revtex\ document class.
+% (Other societies can customize \revtex\ by supplying their own \file{.rtx} file.)
+%
+% This class extension file is a model for a class extension you might write yourself.
+%
+% First, incorporate a \cmd\ProvidesFile\ command with an optional argument giving
+% the version information, e.g.,
+% \begin{verbatim}
+% \ProvidesFile{foo}[2001/09/11 v1.1 Docinfo]%
+% \end{verbatim}
+%
+% Within the society substyle, there are two things we must do as well:
+% define the default journal,
+% \begin{verbatim}
+% \def\@journal@default{pra}%
+% \end{verbatim}
+% And do likewise for the point size:
+% \begin{verbatim}
+% \def\@pointsize@default{10}%
+% \end{verbatim}
+%
+% We first define some text entities (amounting to journal abbreviations),
+% then some APS-specific initializations,
+% then code for particular APS journals.
+% In the latter case, the choice is
+% keyed off the macro \cmd\@journal.
+%
+% \begin{macrocode}
+%<*aps>
+% \end{macrocode}
+%
+% \subsection{Defend Against Forseeable Errors}%
+% Protect this file from being read in by anything but \revtex.
+% \begin{macrocode}
+\ifx\undefined\substyle@ext
+ \def\@tempa{%
+ \endinput
+ \GenericWarning{I must be read in by REVTeX! (Bailing out)}%
+ }%
+ \expandafter\else
+ \def\@tempa{}%
+ \expandafter\fi\@tempa
+ \class@info{RevTeX society APS selected}%
+% \end{macrocode}
+%
+% Here are the class options relating to the APS:
+% \begin{macrocode}
+\DeclareOption{pra}{\change@journal{pra}}%
+\DeclareOption{prb}{\change@journal{prb}}%
+\DeclareOption{prc}{\change@journal{prc}}%
+\DeclareOption{prd}{\change@journal{prd}}%
+\DeclareOption{pre}{\change@journal{pre}}%
+\DeclareOption{prl}{\change@journal{prl}}%
+%\changes{4.2b}{2017/11/21}{(MD) Update options for new titles without "Special Topics" and make prper match style of other journal options}
+\DeclareOption{prab}{\change@journal{prab}}%
+\DeclareOption{prper}{\change@journal{prper}}%
+\DeclareOption{rmp}{\change@journal{rmp}}%
+%\changes{4.2b}{2017/11/21}{(MD) Add options for new APS journals and a generic physrev option for future-proofing}
+\DeclareOption{prx}{\change@journal{prx}}%
+\DeclareOption{prapplied}{\change@journal{prapplied}}%
+\DeclareOption{prmaterials}{\change@journal{prmaterials}}%
+\DeclareOption{prfluids}{\change@journal{prfluids}}%
+\DeclareOption{physrev}{\change@journal{physrev}}%
+% \end{macrocode}
+%
+% \subsection{Abbreviations}%
+% \changes{4.1p}{2010/02/24}{(AO, 584) Per MD, remove trailing space character from each journal abbreviation: it had caused an extraneous space in the .bbl}
+% The following macros constitute typing shortcuts for
+% certain journal names.
+% \begin{macrocode}
+\def\adv{AIP Advances}%
+\def\ao{Appl.\ Opt.}%
+\def\ap{Appl.\ Phys.}%
+\def\apl{Appl.\ Phys.\ Lett.}%
+\def\apm{Appl.\ Phys.\ Lett.\ Mater.}%
+\def\apj{Astrophys.\ J.}%
+\def\bell{Bell Syst.\ Tech.\ J.}%
+\def\bmf{Biomicrofluidics}%
+\def\cha{Chaos}%
+\def\jqe{IEEE J.\ Quantum Electron.}%
+\def\assp{IEEE Trans.\ Acoust.\ Speech Signal Process.}%
+\def\aprop{IEEE Trans.\ Antennas Propag.}%
+\def\mtt{IEEE Trans.\ Microwave Theory Tech.}%
+\def\iovs{Invest.\ Ophthalmol.\ Vis.\ Sci.}%
+\def\jcp{J.\ Chem.\ Phys.}%
+\def\jap{J.\ Appl.\ Phys.}%
+\def\jmp{J.\ Math.\ Phys.}%
+\def\jmo{J.\ Mod.\ Opt.}%
+\def\josa{J.\ Opt.\ Soc.\ Am.}%
+\def\josaa{J.\ Opt.\ Soc.\ Am.\ A}%
+\def\josab{J.\ Opt.\ Soc.\ Am.\ B}%
+\def\jpp{J.\ Phys.\ (Paris)}%
+\def\jpr{J.\ Phys.\ Chem.\ Ref.\ Data}%
+\def\ltp{Low.\ Temp.\ Phys.}%
+\def\nat{Nature (London)}%
+\def\oc{Opt.\ Commun.}%
+\def\ol{Opt.\ Lett.}%
+\def\pl{Phys.\ Lett.}%
+\def\pop{Phys.\ Plasmas}%
+\def\pof{Phys.\ Fluids}%
+\def\pra{Phys.\ Rev.\ A}%
+\def\prb{Phys.\ Rev.\ B}%
+\def\prc{Phys.\ Rev.\ C}%
+\def\prd{Phys.\ Rev.\ D}%
+\def\pre{Phys.\ Rev.\ E}%
+\def\prl{Phys.\ Rev.\ Lett.}%
+\def\rmp{Rev.\ Mod.\ Phys.}%
+\def\rsi{Rev.\ Sci.\ Instrum.}%
+\def\rse{J.\ Renewable Sustainable Energy}%
+\def\pspie{Proc.\ Soc.\ Photo-Opt.\ Instrum.\ Eng.}%
+\def\sjqe{Sov.\ J.\ Quantum Electron.}%
+\def\vr{Vision Res.}%
+\def\sd{Structural\ Dynamics}%
+\def\jor{J.\ Rheol.}%
+\def\cp{AIP\ Conference\ Proceedings}%
+% \end{macrocode}
+%
+% \subsection{APS Setup}%
+% Here we define the default procedures for APS journals.
+% Individual APS journals may override these definitions.
+%
+% \subsubsection{Title block}%
+%
+% The specifics of the title block.
+% Apply to all APS journals; individual journals may override these settings.
+%
+% \begin{macro}{\@fnsymbol}
+% The \LaTeX\ kernel definition of \cmd\@fnsymbol\ is overridden.
+% The definition in \filename{fixltx2e.sty} serves as a guide to the new way to symbol,
+% working in both text- and math modes.
+%
+% \filename{fixltx2e.sty} duplicates some features of \filename{ltxgrid} and \filename{ltxutil},
+% however, so it may be incompatible with \revtex.
+% In case it is not loaded, we must provide a meaning for \cmd\TextOrMath,
+% which that package makes robust.
+% I believe that it is \cmd\@fnsymbol\ itself that ought to be robustified.
+% e\TeX\ further complicates matters; we do not especially accomodate it.
+%
+% \changes{4.1d}{2009/03/27}{Definition of \cs{ @fnsymbol} follows fixltx2e.sty}
+%
+% Not! \cmd\TextOrMath\ must be made robust in any case (Bug 530). I return things to follow
+% core \LaTeXe\ (\filename{latex.ltx}).
+% \changes{4.1f}{2009/10/05}{(AO, 530) \cs{@fnsymbol}: Failed to import fixltx2e.sty technology. Return to LaTeX core.}
+% \begin{macrocode}
+\def\@fnsymbol#1{%
+ \ensuremath{%
+ \ifcase#1\or
+ *\or
+ \dagger\or
+ \ddagger\or
+ \mathsection\or
+ \mathparagraph\or
+% \|\or
+ **\or
+ \dagger\dagger\or
+ \ddagger\ddagger\else
+ \mathsection\mathsection\or
+ \mathparagraph\mathparagraph\or
+ ***\or
+ \dagger\dagger\dagger\or
+ \ddagger\ddagger\ddagger\else
+ \mathsection\mathsection\mathsection\or
+ \mathparagraph\mathparagraph\mathparagraph\or
+%
+ \@ctrerr
+ \fi
+ }%
+}%
+\appdef\document@inithook{%
+ \@ifxundefined\TextOrMath{%
+ \DeclareRobustCommand\TextOrMath{\@ifmmode{\false@sw}{\true@sw}}%
+ }{}%
+}%
+\let\thefootnote@latex\thefootnote
+% \end{macrocode}
+% \end{macro}
+%
+% We assign the default titlepage style for APS; a journal or document instance
+% may override by invoking one of the other \cmd\clo@... procedures defined in \revtex.
+% \begin{macrocode}
+\clo@groupedaddress
+% \end{macrocode}
+%
+% \begin{macro}{\titlepage}
+% \begin{macrocode}
+\renewenvironment{titlepage}{%
+ \let\wastwocol@sw\twocolumn@sw
+ \onecolumngrid
+ \newpage
+ \thispagestyle{titlepage}%
+ \c@page\z@
+% \end{macrocode}
+% A comment: ``article.cls sets this to one not zero?''
+% \begin{macrocode}
+}{%
+ \wastwocol@sw{\twocolumngrid}{\newpage}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@abstractheading}
+% APS Journals all set the abstract head the same way, with no head.
+% However, if the user has specified the \classoption{preprint} class option,
+% then the abstract will have a head.
+% \begin{macrocode}
+\def\frontmatter@abstractheading{%
+ \preprintsty@sw{%
+ \begingroup
+ \centering\large
+ \abstractname
+ \par
+ \endgroup
+ }{}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@abstractwidth}
+% All APS journals set the abstract to the same width.
+% \begin{macrocode}
+\def\frontmatter@abstractwidth{400\p@}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@abstractfont}
+% All APS journals set the abstract body the same way.
+% \changes{4.0c}{1999/11/13}{(AO, 123) add parskip to the abstract.}
+% \changes{4.1e}{2008/06/29}{(AO, 455) be nice to a list within the abstract}
+% \begin{macrocode}
+\def\frontmatter@abstractfont{%
+ \small
+ \parindent1em\relax
+ \adjust@abstractwidth
+}%
+\def\adjust@abstractwidth{%
+ \dimen@\textwidth\advance\dimen@-\frontmatter@abstractwidth
+ \divide\dimen@\tw@
+ \galley@sw{%
+ \advance\rightskip\tw@\dimen@
+ }{%
+ \advance\leftskip\dimen@
+ \advance\rightskip\dimen@
+ }%
+ \@totalleftmargin\leftskip
+}%
+% \end{macrocode}
+% \end{macro}
+%
+%\changes{4.2b}{2017/11/22}{(MD) Change default to not use a title page - it seems antiquated}
+% All APS journal preprints use separate titlepage and full-width abstract.
+%
+% In effect, we establish a society default value for \cmd\preprintsty@sw,
+% and for \cmd\titlepage@sw.
+% \begin{macrocode}
+\@booleanfalse\preprintsty@sw
+\@booleanfalse\titlepage@sw
+% \end{macrocode}
+%
+% We choose the page style for all APS journals.
+% The journal may override by inserting its own code in \cmd\setup@hook.
+% Users wishing to customize their documents will be able to invoke a \cmd\pagestyle\
+% command anywhere in the preamble; it will override the assignments here.
+%
+% Here is the big switch for APS preprints. Note that \cmd\preprintsty@sw
+% is also consulted in various procedures, but we assume its value does
+% not change after \cmd\setup@hook\ time.
+% \begin{macrocode}
+\appdef\setup@hook{%
+ \preprintsty@sw{%
+ \ps@preprint
+ \def\frontmatter@abstractwidth{\textwidth}%
+ \def\frontmatter@affiliationfont{\it}%
+ \let\section\section@preprintsty
+% \end{macrocode}
+% The following line of code had been commented out at this point.
+% \begin{verbatim}
+% \let\@hangfrom@section\@hangfrom@section@preprintsty
+% \end{verbatim}
+% \begin{macrocode}
+ \let\subsection\subsection@preprintsty
+ \let\subsubsection\subsubsection@preprintsty
+ }{%
+ \ps@article
+ }%
+}%
+% \end{macrocode}
+%
+% \begin{macro}{\frontmatter@authorformat}
+% All APS journals set the author list the same.
+% The leading is 11.5 points, and
+% there is 11.5 points of extra space above the first author line
+% (which amounts to the same thing as 11.5 points extra below the title)
+% for a total of 23 points base-to-base.
+%
+% \begin{macrocode}
+\def\frontmatter@authorformat{%
+ \skip@\@flushglue
+ \@flushglue\z@ plus.3\hsize\relax
+ \centering
+ \advance\baselineskip\p@
+ \parskip11.5\p@\relax
+ \@flushglue\skip@
+% \end{macrocode}
+% The following line of code had been commented out at this point.
+% \begin{verbatim}
+%\preprintsty@sw{}{%
+% \addvspace{0\p@}%
+%}%
+% \end{verbatim}
+% \begin{macrocode}
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@above@affilgroup}
+% The default amount of space above affiliation.
+% APS Journals have 24 points b-b above an affiliation group.
+% \begin{macrocode}
+\def\frontmatter@above@affilgroup{%
+% \end{macrocode}
+% The following line of code had been commented out at this point.
+% \begin{verbatim}
+%\preprintsty@sw{}{%
+% \addvspace{11\p@}%
+%}%
+% \end{verbatim}
+% \begin{macrocode}
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@above@affiliation}
+% \begin{macro}{\frontmatter@above@affiliation@script}
+% The default amount of space above affiliation.
+% APS Journals have no extra space between author group down to common affiliation.
+% \begin{macrocode}
+\def\frontmatter@above@affiliation@script{%
+ \skip@\@flushglue
+ \@flushglue\z@ plus.3\hsize\relax
+ \centering
+ \@flushglue\skip@
+ \addvspace{3.5\p@}%
+}%
+\def\frontmatter@above@affiliation{%
+ \preprintsty@sw{}{%
+% \end{macrocode}
+% The following line of code had been commented out at this point.
+% \begin{verbatim}
+% \addvspace{12\p@}%
+% \end{verbatim}
+% \begin{macrocode}
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@affiliationfont}
+% All APS journals set the affiliation the same.
+% \begin{macrocode}
+\def\frontmatter@affiliationfont{%
+ \small\it
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@collaboration@above}
+% PRL: 1.5 points extra: 13 points base-to-base above.
+% \begin{macrocode}
+\def\frontmatter@collaboration@above{%
+ \preprintsty@sw{%
+ }{%
+ \parskip1.5\p@\relax
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@setup}
+% All APS journals set the title page using the same font and size.
+% However, justification varies for the title block elements, so
+% we assert none here.
+% \begin{macrocode}
+\def\frontmatter@setup{%
+ \normalfont
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@title@above}
+% \begin{macro}{\frontmatter@title@format}
+% \begin{macro}{\frontmatter@title@below}
+% All APS journals set the article title the same.
+%
+% Note: Spacing from title to author is 23 points base-to-base.
+% \begin{macrocode}
+\def\frontmatter@title@above{\addvspace{6\p@}}%
+\def\frontmatter@title@format{\large\bfseries\centering\parskip\z@skip}%
+\def\frontmatter@title@below{}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@makefnmark}
+% All APS journals share this procedure for setting the titlepage footnote text.
+% \begin{macrocode}
+\def\@author@parskip{3\p@}%
+\def\frontmatter@makefnmark{%
+ \@textsuperscript{%
+ \normalfont\@thefnmark
+ }%
+}%
+\def\frontmatter@authorbelow{%
+ \addvspace{3\p@}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@RRAP@format}
+% All APS journals use the same format for the ``Received, Revised, etc.'' block on the title page.
+%
+% Change note: 11.5 points b-b from author/affiliation down to date.
+% \begin{macrocode}
+\def\frontmatter@RRAP@format{%
+ \small
+ \centering
+% \end{macrocode}
+% The following line of code had been commented out at this point.
+% \begin{verbatim}
+% \preprintsty@sw{}{\parskip.5ex\relax}%
+% \end{verbatim}
+% \begin{macrocode}
+ \everypar{\hbox\bgroup(\@gobble@leavemode@uppercase}%
+ \def\par{\@ifvmode{}{\unskip)\egroup\@@par}}%
+}%
+\def\punct@RRAP{;\egroup\ \hbox\bgroup}%
+\def\@gobble@leavemode@uppercase#1#2{\expandafter\MakeTextUppercase}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@PACS@format}
+%
+% \begin{macrocode}
+\def\frontmatter@PACS@format{%
+ \addvspace{11\p@}%
+ \footnotesize
+ \adjust@abstractwidth
+ \parindent\z@
+ \parskip\z@skip
+ \samepage
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@keys@format}
+%
+% \begin{macrocode}
+\def\frontmatter@keys@format{%
+ \footnotesize
+ \adjust@abstractwidth
+ \parindent\z@
+ \samepage
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\ps@titlepage}
+% \changes{4.0a}{1998/06/10}{multiple preprint commands}
+% \changes{4.0b}{1999/06/20}{Stack \cs{preprint} args flush right at right margin.}
+% Title page style. Currently empty except for preprint header,
+% which consists of all the \cmd\preprint\ arguments,
+% stacked flush right at the right margin.
+% \changes{4.0c}{1999/11/13}{Insert procedure \cs{checkindate}}
+% \begin{macrocode}
+\def\ps@titlepage{%
+ \def\@oddhead{%
+ \hfill
+ \preprint@sw{%
+ \expandafter\produce@preprints\expandafter{\@preprint}%
+ }{}%
+ }%
+ \let\@evenhead\@oddhead
+ \def\@oddfoot{%
+ \hb@xt@\z@{\byrevtex\hss}%
+ \hfil
+ \preprintsty@sw{\thepage}{}%
+ \quad\checkindate
+ \hfil
+ }%
+ \let\@evenfoot\@oddfoot
+}%
+\def\byrevtex{\byrevtex@sw{Typeset by REV\TeX}{}}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\produce@preprints}
+% \changes{4.0c}{1999/11/13}{(AO, 115) If three or more preprints specified, set on single line, with commas.}
+% \begin{macrocode}
+\def\produce@preprints#1{%
+ \vtop to \z@{%
+ \def\baselinestretch{1}%
+ \small
+ \let\preprint\preprint@count
+ \count@\z@
+ #1%
+ \@ifnum{\count@>\tw@}{%
+ \hbox{%
+ \let\preprint\preprint@hlist
+ #1%
+ \setbox\z@\lastbox
+ }%
+ }{%
+ \let\preprint\preprint@cr
+ \halign{\hfil##\cr#1\crcr}%
+ \par
+ \vss
+ }%
+ }%
+}%
+\def\preprint@cr#1{#1\cr}%
+\def\preprint@count#1{\advance\count@\@ne}%
+\def\preprint@hlist#1{#1\hbox{, }}%
+% \end{macrocode}
+% \end{macro}
+%
+% \subsubsection{Stacked Heads}%
+% All APS journals put a period (.), followed by quad space, after the section number.
+% Also, no hanging section number.
+% \begin{macrocode}
+\def\@seccntformat#1{\csname the#1\endcsname.\quad}%
+\def\@hang@from#1#2#3{#1#2#3}%
+% \end{macrocode}
+%
+% Note that in the following, we wish to set the section head uppercase, so we use
+% David Carlisle's \cmd\MakeTextUppercase. However, because this procedure effectively
+% parses its argument (looking for things to \emph{not} translate), it has to be invoked
+% in such a way that the argument of the \cmd\section\ command is passed to it as its
+% own argument.
+%
+% To accomplish this, we use the \cmd\@hangfrom@\ hook, which was developed for this
+% purpose.
+%
+% \begin{macrocode}
+\def\section{%
+ \@startsection
+ {section}%
+ {1}%
+ {\z@}%
+ {0.8cm \@plus1ex \@minus .2ex}%
+ {0.5cm}%
+ {%
+ \normalfont\small\bfseries
+ \centering
+ }%
+}%
+\def\@hangfrom@section#1#2#3{\@hangfrom{#1#2}\MakeTextUppercase{#3}}%
+\def\@hangfroms@section#1#2{#1\MakeTextUppercase{#2}}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\subsection{%
+ \@startsection
+ {subsection}%
+ {2}%
+ {\z@}%
+ {.8cm \@plus1ex \@minus .2ex}%
+ {.5cm}%
+ {%
+ \normalfont\small\bfseries
+ \centering
+ }%
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\subsubsection{%
+ \@startsection
+ {subsubsection}%
+ {3}%
+ {\z@}%
+ {.8cm \@plus1ex \@minus .2ex}%
+ {.5cm}%
+ {%
+ \normalfont\small\itshape
+ \centering
+ }%
+}%
+% \end{macrocode}
+%
+% \subsubsection{Runin Heads}%
+% \begin{macrocode}
+\def\paragraph{%
+ \@startsection
+ {paragraph}%
+ {4}%
+ {\parindent}%
+ {\z@}%
+ {-1em}%
+ {\normalfont\normalsize\itshape}%
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\subparagraph{%
+ \@startsection
+ {subparagraph}%
+ {5}%
+ {\parindent}%
+ {3.25ex \@plus1ex \@minus .2ex}%
+ {-1em}%
+ {\normalfont\normalsize\bfseries}%
+}%
+% \end{macrocode}
+%
+%
+%
+% \begin{macro}{\section@preprintsty}
+% \begin{macro}{\subsection@preprintsty}
+% \begin{macro}{\subsubsection@preprintsty}
+% Here are the formatting procedures specific to the preprint style;
+% the only difference is that the heads are flush left instead of centered.
+%
+% \begin{macrocode}
+\def\section@preprintsty{%
+ \@startsection
+ {section}%
+ {1}%
+ {\z@}%
+ {0.8cm \@plus1ex \@minus .2ex}%
+ {0.5cm}%
+ {%
+ \normalfont\small\bfseries
+% \centering
+ }%
+}%
+%\def\@hangfrom@section@preprintsty#1#2#3{\@hangfrom{#1#2}\MakeTextUppercase{#3}}%
+% \end{macrocode}
+% \begin{macrocode}
+\def\subsection@preprintsty{%
+ \@startsection
+ {subsection}%
+ {2}%
+ {\z@}%
+ {.8cm \@plus1ex \@minus .2ex}%
+ {.5cm}%
+ {%
+ \normalfont\small\bfseries
+% \centering
+ }%
+}%
+% \end{macrocode}
+% \begin{macrocode}
+\def\subsubsection@preprintsty{%
+ \@startsection
+ {subsubsection}%
+ {3}%
+ {\z@}%
+ {.8cm \@plus1ex \@minus .2ex}%
+ {.5cm}%
+ {%
+ \normalfont\small\itshape
+% \centering
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% By default, APS journals set titlepage notes as footnotes.
+% \begin{verbatim}
+\let\frontmatter@footnote@produce\frontmatter@footnote@produce@footnote
+% \end{verbatim}
+%
+% \subsubsection{Table of Contents}%
+% The toc will itself make an entry in the toc,
+% but we temporarily turn off toc formatting for the duration.
+% \changes{4.1a}{2008/01/19}{(AO, 461) Change the csname revtex uses from @dotsep to ltxu@dotsep. The former is understood in mu. (What we wanted was a dimension.)}%
+% \begin{macrocode}
+\def\@pnumwidth{1.55em}%
+\def\@tocrmarg {2.55em}%
+\def\@dotsep{2}%
+\def\ltxu@dotsep{4.5pt}%
+\setcounter{tocdepth}{3}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\tableofcontents{%
+ \addtocontents{toc}{\string\tocdepth@munge}%
+ \print@toc{toc}%
+ \addtocontents{toc}{\string\tocdepth@restore}%
+}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\tocdepth@munge{%
+ \let\l@section@saved\l@section
+ \let\l@section\@gobble@tw@
+}%
+\def\@gobble@tw@#1#2{}%
+% \end{macrocode}
+%
+% \begin{macrocode}
+\def\tocdepth@restore{%
+ \let\l@section\l@section@saved
+}%
+% \end{macrocode}
+%
+% The following definition of \cmd\l@part\ is a variant on
+% the definition of \cmd\l@@sections\ in \file{ltxutil.dtx}.
+% \begin{macrocode}
+\def\l@part#1#2{\addpenalty{\@secpenalty}%
+ \begingroup
+ \set@tocdim@pagenum\@tempboxa{#2}%
+% \@tempdima 3em %
+ \parindent \z@
+ \rightskip\tocleft@pagenum plus 1fil\relax
+ \skip@\parfillskip\parfillskip\z@
+ \addvspace{2.25em plus\p@}%
+ \large \bf %
+ \leavevmode\ignorespaces#1\unskip\nobreak\hskip\skip@
+ \hb@xt@\rightskip{\hfil\unhbox\@tempboxa}\hskip-\rightskip\hskip\z@skip
+ \par
+ \nobreak %
+ \endgroup
+}%
+% \end{macrocode}
+%
+% \begin{macro}{\l@section}
+% Determine which TOC elements are automatically indented.
+%
+% We set the TOC to the standard of RMP.
+% If APS has its own specification, we will code it, and RMP must override.
+% \begin{macrocode}
+\def\tocleft@{\z@}%
+\def\tocdim@min{5\p@}%
+\def\l@section{%
+ \l@@sections{}{section}% Implicit #3#4
+}%
+\def\l@f@section{%
+ \addpenalty{\@secpenalty}%
+ \addvspace{1.0em plus\p@}%
+ %\bf
+}%
+\def\l@subsection{%
+ \l@@sections{section}{subsection}% Implicit #3#4
+}%
+\def\l@subsubsection{%
+ \l@@sections{subsection}{subsubsection}% Implicit #3#4
+}%
+\def\l@paragraph#1#2{}%
+\def\l@subparagraph#1#2{}%
+% \end{macrocode}
+% \end{macro}
+%
+% Activate the auto TOC processing.
+% \begin{macrocode}
+\let\toc@pre\toc@pre@auto
+\let\toc@post\toc@post@auto
+% \end{macrocode}
+%
+% \subsubsection{Default column bottom}%
+% All APS journal styles have flush bottoms.
+%\changes{4.1n}{2009/11/30}{(AO, 566) restore 4.0 behavior: flush column bottoms}
+% \begin{macrocode}
+\@booleanfalse\raggedcolumn@sw
+% \end{macrocode}
+%
+%
+% \subsubsection{Table alignment style}%
+%
+% \begin{macro}{\tableft@skip@float}
+% \begin{macro}{\tabmid@skip@float}
+% \begin{macro}{\tabright@skip@float}
+% \begin{macro}{\array@row@pre@float}
+% \begin{macro}{\array@row@pst@float}
+% All APS publications have the same table specification:
+% Scotch rules above and below, centered in column.
+% \begin{macrocode}
+\def\tableft@skip@float{\z@ plus\hsize}%
+\def\tabmid@skip@float{\@flushglue}%
+\def\tabright@skip@float{\z@ plus\hsize}%
+\def\array@row@pre@float{\hline\hline\noalign{\vskip\doublerulesep}}%
+\def\array@row@pst@float{\noalign{\vskip\doublerulesep}\hline\hline}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+%
+% \subsubsection{Footnote formatting}%
+% \changes{4.1n}{2010/01/02}{(AO, 572) \cs{@makefntext} and \cs{frontmatter@makefntext} must be defined harmoniously}%
+% We customize the formatting of footnotes for all APS journals.
+% \begin{macro}{\@makefntext}
+% \begin{macrocode}
+\long\def\@makefntext#1{%
+ \def\baselinestretch{1}%
+ \leftskip1em%
+ \parindent1em%
+ \noindent
+ \nobreak\hskip-\leftskip
+ \hb@xt@\leftskip{%
+ \hss\@makefnmark\ %
+ }%
+ #1%
+ \par
+}%
+% \end{macrocode}
+% \end{macro}
+% \begin{macro}{\frontmatter@makefntext}
+% We ensure that frontmatter footnotes format similarly to body footnotes.
+% But we provide our own hypertext anchor, otherwise not provided.
+% \changes{4.1n}{2010/01/02}{(AO, 572) Independent footnote counter for title block. Abstract footnote counter shared with body.}%
+% \begin{macrocode}
+\long\def\frontmatter@makefntext#1{%
+ \def\baselinestretch{1}%
+ \leftskip1em%
+ \parindent1em%
+ \noindent
+ \nobreak\hskip-\leftskip
+ \Hy@raisedlink{\hyper@anchorstart{frontmatter.\expandafter\the\csname c@\@mpfn\endcsname}\hyper@anchorend}%
+ \hb@xt@\leftskip{%
+ \hss\@makefnmark\ %
+ }%
+ #1%
+ \par
+}%
+% \end{macrocode}
+% \end{macro}
+%
+%
+% \subsubsection{Appendix}%
+% \begin{macro}{\appendix}
+% \begin{macro}{\@hangfrom@appendix}
+% \begin{macro}{\@hangfroms@appendix}
+% \begin{macro}{\@appendixcntformat}
+% \changes{4.0c}{1999/11/13}{(AO, 129) section* within appendix was producing appendixname}
+% \begin{macrocode}
+\prepdef\appendix{%
+ \par
+ \let\@hangfrom@section\@hangfrom@appendix
+%\let\@hangfroms@section\@hangfroms@appendix
+ \let\@sectioncntformat\@appendixcntformat
+}%
+\def\@hangfrom@appendix#1#2#3{%
+ #1%
+ \@if@empty{#2}{%
+ #3%
+ }{%
+ #2\@if@empty{#3}{}{:\ #3}%
+ }%
+}%
+\def\@hangfroms@appendix#1#2{%
+ #1#2%
+}%
+\def\@appendixcntformat#1{\appendixname\ \csname the#1\endcsname}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \subsubsection{Bibliography}%
+% \changes{4.1n}{2010/01/04}{(AO, 576) In .bst files, remove support for the annote field}%
+%
+% Customize \revtex\ for the journal substyle;
+% this task requires three components:
+% the \BibTeX\ \file{apsrev.bst} and \file{apsrmp.bst} style files, and
+% customizations of the \env{thebibliography} environment.
+%
+% \begin{macro}{\@bibstyle}
+% Define the argument of the \cmd\bibliographystyle\ command
+% (if the document does not do so).
+% The user must have installed a \file{.bst} file of the corresponding name.
+% This file will then be used by \BibTeX\ when compiling the document's \file{.bbl} file.
+%
+% To generate \file{apsrev.bst}, use \classname{custom-bib} version 4.21 or later.
+% Run the \file{.bst} generator, \file{makebst.tex}, and
+% accept all defaults, with the following exceptions:
+%
+% \def\QandA#1#2#3{\item#1: \textbf{#2}: \texttt{#3}---}%
+% \def\CandE#1{\item\relax\texttt{#1}---}%
+% \begin{enumerate}
+% \QandA{LANGUAGE FIELD}{l}{lang}Use language field to switch hyphenation patterns for title
+% \QandA{PRESENTATIONS}{b}{pres,pres-bf}Presentation, speaker bold face
+% \QandA{ORDERING OF REFERENCES}{c}{seq-no}Citation order (unsorted, like unsrt.bst)
+% \QandA{ORDER ON VON PART}{x}{vonx}Sort without von part (de la Maire after Mahone)
+% \QandA{AUTHOR NAMES}{i}{nm-init,ed-au}Initials + surname (J. F. Smith)
+% \QandA{POSITION OF JUNIOR}{*}{jnrlst}Junior comes last as Smith, John, Jr.
+% \QandA{TYPEFACE FOR AUTHORS IN LIST OF REFERENCES}{u}{nmft,nmft-def}User defined author font (\cmd\bibnamefont)
+% \QandA{FONT FOR FIRST NAMES}{u}{fnm-def}First names in user defined font (\cmd\bibfnamefont)
+% \QandA{EDITOR NAMES IN INCOLLECTION ETC}{a}{nmfted}Editors incollection like authors font
+% \QandA{FONT FOR `AND' IN LIST}{r}{nmand-rm}`And' in normal font (JONES and JAMES)
+% \QandA{FONT OF CITATION LABELS IN TEXT}{u}{lab,lab-def}User defined citation font (\cmd\citenamefont)
+% \QandA{FONT FOR `AND' IN CITATIONS}{r}{and-rm}Cited `and' in normal font
+% \QandA{DATE FORMAT}{*}{yr-par}Date in parentheses as (May 1993)
+% \QandA{DATE EMPTY}{-}{date-nil-x}If date is empty, then do not produce the surrounding punctuation (parens, brackets, colon, comma)
+% \QandA{TITLE OF ARTICLE}{d}{tit-qq}Title and punctuation in double quotes (``Title,'' ..)
+% \QandA{INPROCEEDINGS CHAPTER AND PAGES, LIKE INBOOK}{-}{inproceedings-chapter}produce pages after chapter, just as in InBook
+% \QandA{TITLE PRESENT IN ARTICLE, INCOLLECTION, AND INPROCEEDINGS}{x}{jtit-x}Title is ignored
+% \QandA{INPROCEEDINGS CHAPTER AND PAGES, LIKE INBOOK}{y}{inproceedings-chapter}produce pages after chapter, just as in InBook
+% \QandA{ARTICLE BOOKTITLE PRESENT}{}{article-booktitle}format booktitle
+% \QandA{ARTICLE SERIES PRESENT}{}{article-series}article can has series
+% \QandA{JOURNAL NAME FONT}{r}{jttl-rm}Journal name normal font
+% \QandA{JOURNAL NAME WITH ADDRESS}{y}{journal-address}Include address field (in parentheses) along with journal name
+% \QandA{BOOK TITLE FIELDS}{y}{book-bt}Field `booktitle', or if absent field `title', is book title
+% \QandA{THESIS TITLE OPTIONAL}{}{thesis-title-o}Title is optional: no warning issued if empty
+% \QandA{TECHNICAL REPORT TITLE}{b}{trtit-b}Tech. report title like books
+% \QandA{TECHNICAL REPORT INSTITUTION}{}{techreport-institution-par}format tech report institution like book publisher
+% \QandA{JOURNAL VOLUME}{b}{vol-bf}Volume bold as {\bf vol}(num)
+% \QandA{JOURNAL VOL AND NUMBER}{x}{vnum-x}Journal vol, without number as 34
+% \QandA{VOLUME PUNCTUATION}{c}{volp-com}Volume with comma as vol(num), ppp
+% \QandA{PAGE NUMBERS}{f}{jpg-1}Only start page number
+% \QandA{BOOK EDITOR W/O AUTHOR}{}{book-editor-booktitle}Book permits empty author, produces title before editor in this case
+% \QandA{INBOOK PERMITS TITLE, BOOKTITLE, AUTHOR, EDITOR}{a}{inbook-editor-booktitle}Allow using both title/booktitle, both author/editor
+% \QandA{CONFERENCE ADDRESS FOR BOOK, INBOOK, INCOLLECTION, INPROCEEDINGS, PROCEEDINGS}{a}{bookaddress}Italic booktitle followed by bookaddress in roman
+% \QandA{NUMBER AND SERIES FOR BOOK, INBOOK, INCOLLECTION, INPROCEEDINGS, PROCEEDINGS}{*}{num-xser}Allows number without series and suppresses word "number"
+% \QandA{WORD NUMBER CAPITALIZED FOR NUMBER AND SERIES}{c}{number-cap}Capitalize word `number' as: "Number 123"
+% \QandA{WORD CHAPTER CAPITALIZED}{c}{chapter-cap}Capitalize word `chapter' as: `Chapter 42'
+% \QandA{COMBINING NUMBER AND SERIES}{x}{series-number}Series number as: `Springer Lecture Notes No. 125'
+% \QandA{POSITION OF NUMBER AND SERIES}{b}{numser-booktitle}After book title and conference address, and before editors
+% \QandA{VOLUME AND SERIES FOR BOOKS/COLLECTIONS}{s}{ser-vol}Series, vol. 23
+% \QandA{VOLUME AND SERIES FORMATTING}{y}{ser-rm}format series roman , even when used with volume
+% \QandA{WORD VOLUME CAPITALIZED FOR VOLUME AND SERIES}{y}{volume-cap}Capitalize word `volume', as: `Volume 7 in Lecture Series'
+% \QandA{POSITION OF VOLUME AND SERIES FOR INCOLLECTION, INBOOK, AND INPROCEEDINGS}{e}{ser-ed}Series and volume after booktitle and before editors
+% \QandA{JOURNAL NAME PUNCTUATION}{x}{jnm-x}Space after journal name
+% \QandA{PAGES IN BOOK}{*}{pg-bk,book-chapter-pages}As chapter and page: chapter 42, page 345
+% \QandA{PUBLISHER IN PARENTHESES}{d}{pub-date}Publisher with address and date in parentheses (Oxford, 1994)
+% \QandA{EMPTY PUBLISHER PARENTHESES}{y}{ay-empty-pub-parens-x}eliminate parentheses altogether if nothing inside
+% \QandA{PUBLISHER POSITION}{}{pre-pub}Publisher before volume, chapter, pages
+% \QandA{}{}{pre-edn}Edition before publisher
+% \QandA{}{p}{pre-pub,pre-edn}Edition, publisher, volume, chapter, pages
+% \QandA{ISBN NUMBER}{*}{isbn}Include ISBN for books, booklets, etc.
+% \QandA{ISSN NUMBER}{*}{issn}Include ISSN for periodicals
+% \QandA{DOI NUMBER}{a}{doi-link,doi}Doi forms a link to the publication, anchored to the volume or title
+% \QandA{EDITOR IN COLLECTIONS}{b}{edby}In booktitle, edited by .. (where .. is names)
+% \QandA{PUNCTUATION BETWEEN SECTIONS (BLOCKS)}{c}{blk-com}Comma between blocks
+% \QandA{FINAL PUNCTUATION}{c}{fin-endbibitem}Command at end instead of period
+% \QandA{ABBREVIATE WORD `PAGES'}{a}{pp}`Page' abbreviated as p. or pp.
+% \QandA{ABBREVIATE WORD `EDITORS'}{a}{ed}`Editor' abbreviated as ed. or eds.
+% \QandA{OTHER ABBREVIATIONS}{a}{abr}Abbreviations of such words
+% \QandA{ABBREVIATION FOR `EDITION'}{a}{ednx}`Edition' abbreviated as `ed'
+% \QandA{EDITION NUMBERS}{n}{ord}Numerical editions as 1st, 2nd, 3rd, etc
+% \QandA{STORED JOURNAL NAMES}{a}{jabr}Abbreviated journal names
+% \QandA{FONT OF `ET AL'}{i}{etal-it}Italic et al
+% \QandA{ADDITIONAL REVTeX DATA FIELDS}{r}{revdata,eprint,url,url-blk,translation}Include REVTeX data fields collaboration, eid, eprint, archive, url, translation
+% \QandA{SLACcitation FIELD}{}{SLACcitation}Produce SLACcitation field
+% \QandA{NUMPAGES DATA FIELD}{*}{numpages-x}Do not include numpages field
+% \QandA{URL ADDRESS}{*}{url,url-prefix-x}URL without prefix (default: `URL ')
+% \QandA{REFERENCE COMPONENT TAGS}{b}{bibinfo}Reference component tags like \cmd\bibinfo in the content of \cmd\bibitem
+% \QandA{ELEMENT TAGS}{b}{bibfield}Element tags like \cmd\bibfield in the content of \cmd\bibitem
+% \QandA{COMPATIBILITY WITH PLAIN TEX}{*}{nfss}Use LaTeX commands which may not work with Plain TeX
+% \end{enumerate}
+%
+% A file \file{apsrev.dbj} file equivalent to the following should result:
+% \begin{verbatim}
+%\input docstrip
+%\preamble
+%----------------------------------------
+%*** REVTeX-compatible Phys Rev 2010-02-12 ***
+%\endpreamble
+%\postamble
+%End of customized bst file
+%\endpostamble
+%\keepsilent
+%\askforoverwritefalse
+%\def\MBopts{\from{merlin.mbs}{%
+% head,\MBopta}
+%\from{physjour.mbs}{\MBopta}
+%\from{geojour.mbs}{\MBopta}
+%\from{photjour.mbs}{\MBopta}
+%\from{merlin.mbs}{tail,\MBopta}}
+%\def\MBopta{%
+% ay,%: Author-year with some non-standard interface
+% nat,%: Natbib for use with natbib v5.3 or later
+% lang,%: Use language field to switch hyphenation patterns for title
+% pres,pres-bf,%: Presentation, speaker bold face
+% seq-no,%: Citation order (unsorted, only meaningful for numericals)
+% vonx,%: Sort without von part (de la Maire after Mahone)
+% nm-init,ed-au,%: Initials + surname (J. F. Smith)
+% jnrlst,%: Junior comes last as Smith, John, Jr.
+% nmft,nmft-def,%: User defined author font (\bibnamefont)
+% fnm-def,%: First names in user defined font (\bibfnamefont)
+% nmfted,%: Editors incollection like authors font
+% nmand-rm,%: `And' in normal font (JONES and JAMES)
+% lab,lab-def,%: User defined citation font (\citenamefont)
+% and-rm,%: Cited `and' in normal font
+% keyxyr,%: Year blank when KEY replaces missing author (for natbib 7.0)
+% blkyear,%: Missing date left blank
+% yr-par,%: Year in parentheses as (1993)
+% dtrev,%: Date as year month
+% date-nil-x,%: If date is empty, then do not produce the surrounding punctuation (parens, brackets, colon, comma)
+% tit-qq,%: Title and punctuation in double quotes (``Title,'' ..)
+% inproceedings-chapter,%: produce pages after chapter, just as in InBook
+% jtit-x,%: Title is ignored
+% inproceedings-chapter,%: produce pages after chapter just as in InBook
+% article-booktitle,%: format booktitle
+% article-series,%: article can has series
+% jttl-rm,%: Journal name normal font
+% journal-address,%: Include address field (in parentheses) along with journal name
+% book-bt,%: Field `booktitle', or if absent field `title', is book title
+% thesis-title-o,%: Title is optional: no warning issued if empty
+% trtit-b,%: Tech. report title like books
+% techreport-institution-par,%: format tech report institution like book publisher
+% vol-bf,%: Volume bold as {\bf vol}(num)
+% vnum-x,%: Journal vol, without number as 34
+% volp-com,%: Volume with comma as vol(num), ppp
+% jpg-1,%: Only start page number
+% book-editor-booktitle,%: Book permits empty author, produces title before editor in this case
+% inbook-editor-booktitle,%: Allow using both title/booktitle, both author/editor
+% bookaddress,%: Italic booktitle followed by bookaddress in roman
+% num-xser,%: Allows number without series and suppresses word "number"
+% number-cap,%: Capitalize word `number' as: "Number 123"
+% chapter-cap,%: Capitalize word `chapter' as: `Chapter 42'
+% series-number,%: Series number as: `Springer Lecture Notes No. 125'
+% numser-booktitle,%: After book title and conference address, and before editors
+% ser-vol,%: Series, vol. 23
+% ser-rm,%: format series roman , even when used with volume
+% volume-cap,%: Capitalize word `volume', as: `Volume 7 in Lecture Series'
+% ser-ed,%: Series and volume after booktitle and before editors
+% jnm-x,%: Space after journal name
+% pg-bk,book-chapter-pages,%: As chapter and page: chapter 42, page 345
+% pub-date,%: Publisher with address and date in parentheses (Oxford, 1994)
+% ay-empty-pub-parens-x,%: eliminate parentheses altogether if nothing inside
+% pre-pub,pre-edn,%: Edition, publisher, volume, chapter, pages
+% isbn,%: Include ISBN for books, booklets, etc.
+% issn,%: Include ISSN for periodicals
+% doi-link,doi,%: Doi forms a link to the publication, anchored to the volume or title
+% edby,%: In booktitle, edited by .. (where .. is names)
+% blk-com,%: Comma between blocks
+% fin-endbibitem,%: Command at end instead of period
+% pp,%: `Page' abbreviated as p. or pp.
+% ed,%: `Editor' abbreviated as ed. or eds.
+% abr,%: Abbreviations of such words
+% ednx,%: `Edition' abbreviated as `ed'
+% ord,%: Numerical editions as 1st, 2nd, 3rd, etc
+% jabr,%: Abbreviated journal names
+% etal-it,%: Italic et al
+% revdata,eprint,url,url-blk,translation,%: Include REVTeX data fields collaboration, eid, eprint, archive, url, translation
+% SLACcitation,%: Produce SLACcitation field
+% numpages-x,%: Do not include numpages field
+% url,url-prefix-x,%: URL without prefix (default: `URL ')
+% bibinfo,%: Reference component tags like \bibinfo in the content of \bibitem
+% bibfield,%: Element tags like \bibfield in the content of \bibitem
+% nfss,%: Use LaTeX commands which may not work with Plain TeX
+%,{%
+% }}
+%\generate{\file{apsrev4-2.bst}{\MBopts}}
+%\endbatchfile
+% \end{verbatim}
+%
+% \subsubsection{Comparing apsrev.bst and apsrmp.bst}\label{sec:CompRevRmp}
+% These two bibliographic styles differ as follows:
+% \file{apsrev.dbj} has the following guard codes, which \file{apsrmp.dbj} does not:
+% \begin{itemize}
+% \CandE{seq-no}%
+% Citation order (unsorted, like unsrt.bst)
+% \CandE{nm-init,ed-au}%
+% Initials + surname (J. F. Smith)
+% \CandE{blkyear}%
+% Missing date left blank
+% \CandE{date-nil-x}%
+% If date is empty, then do not produce the surrounding punctuation (parens, brackets, colon, comma)
+% \CandE{inproceedings-chapter}%
+% produce pages after chapter, just as in InBook
+% \CandE{techreport-institution-par}%
+% format tech report institution like book publisher
+% \CandE{vnum-x}%
+% Journal vol, without number as `34'
+% \CandE{pub-date}%
+% Publisher with address and date in parentheses (Oxford, 1994)
+% \CandE{pre-pub}%
+% Edition, publisher, volume, chapter, pages. Note that both use guard code pre-edn.
+% \end{itemize}
+%
+% \file{apsrmp.dbj} has the following guard codes, which \file{apsrev.dbj} does not:
+% \begin{itemize}
+% \CandE{nm-rev1}%
+% Only first name reversed, initials (AGU style: Smith, J. F., H. K. Jones)
+% \CandE{dt-beg}%
+% Date after authors
+% \CandE{vnum-sp}%
+% Journal vol (num) as `34 (2)'
+% \CandE{pp-last}%
+% Pages at end, but before any notes
+% \CandE{pub-par}%
+% Publisher in parentheses
+% \CandE{school-par}%
+% School/address in parens: `(school, address)'
+% \CandE{bkedcap}%
+% `Name Editor,' as above, editor upper case
+% \CandE{and-com}%
+% Comma even with 2 authors as `Tom, and Harry'
+%
+% \end{itemize}
+%
+% We ensure that the journal substyle has the first word
+% in the matter by installing the (default) APS code
+% later on (see Section~\ref{sec:aps-defaults}).
+% \end{macro}
+%
+% \begin{macro}{\authoryear@sw}
+% \changes{4.1a}{2008/01/18}{(AO, 457) Endnotes to be sorted in with numerical citations.}%
+% Numerical citations: default value of \cmd\authoryear@sw\ is false.
+% \begin{macrocode}
+\@booleanfalse\authoryear@sw
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\bibpunct}
+% The following commands effectively establish the style in which \cmd\cite\ commands are
+% formatted.
+% You can think of them as the second needed component for the bibliography.
+%
+% Set up for APS numerical citations (once the packages are loaded).
+% The journal substyle can override these choices.
+%
+% Note that, prior to \classname{natbib} version 8.21, changing \cmd\NAT@sort\ at this late hour would not be totally effective;
+% you would have to give \classname{natbib} the relevant options at load time.
+% From version 8.21 on, \cmd\NAT@sort\ and \cmd\NAT@cmprs\ are not bound at all.
+% \changes{4.1a}{2008/01/19}{For natbib versions before 8.21, \cs{NAT@sort} was consulted only as natbib was being read in. Now it is fully dynamic.}
+% \begin{macrocode}
+\appdef\setup@hook{%
+ \bibpunct{[}{]}{,}{n}{}{,}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\pre@bibdata}
+%
+% Set up to write endnotes to a .bib file; its data will be incorporated into the bibliography.
+% \begin{macrocode}
+ \def\pre@bibdata{\jobname\bibdata@app}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\bibsection}
+% We define the sectioning command to use when starting the bibliography.
+%
+% \changes{4.0rc1}{2001/06/18}{grid changes with push and pop}
+% \begin{macrocode}
+\appdef\setup@hook{%
+ \def\bibsection{%
+ \par
+ \onecolumngrid@push
+ \begingroup
+ \baselineskip26\p@
+ \bib@device{\textwidth}{245.5\p@}%
+ \endgroup
+ \nobreak\@nobreaktrue
+ \addvspace{19\p@}%
+ \par
+ \onecolumngrid@pop
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\bib@device}
+% \begin{macro}{\bibpreamble}
+% \begin{macro}{\bibsep}
+% \begin{macro}{\newblock}
+% We define the sectioning command to use when starting the bibliography.
+% \begin{macrocode}
+\def\bib@device#1#2{%
+ \hb@xt@\z@{%
+ \hb@xt@#1{%
+ \hfil
+ \phantomsection
+ \addcontentsline {toc}{section}{\protect\numberline{}\refname}%
+% \hyper@anchorstart {\@currentHref }%
+ \hb@xt@#2{%
+ \skip@\z@\@plus-1fil\relax
+ \leaders\hrule height.25 \p@ depth.25 \p@ \hskip\z@\@plus1fil
+ \hskip\skip@
+ \hskip\z@\@plus0.125fil\leaders\hrule height.375\p@ depth.375\p@ \hskip\z@\@plus0.75fil \hskip\z@\@plus0.125fil
+ \hskip\skip@
+ \hskip\z@\@plus0.25 fil\leaders\hrule height.5 \p@ depth.5 \p@ \hskip\z@\@plus0.5 fil \hskip\z@\@plus0.25 fil
+ \hskip\skip@
+ \hskip\z@\@plus0.375fil\leaders\hrule height.625\p@ depth.625\p@ \hskip\z@\@plus0.25fil \hskip\z@\@plus0.375fil
+ % \hskip\skip@
+ % \hfil
+ }%
+% \hyper@anchorend
+ \hfil
+ }%
+ \hss
+ }%
+}%
+\appdef\setup@hook{%
+ \let\bibpreamble\@empty
+ \bibsep\z@\relax
+ \def\newblock{\ }%
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\bibfont}
+% We define the font switch that applies to the body of the bibliography.
+%
+% \begin{macrocode}
+\appdef\setup@hook{%
+ \def\bibfont{%
+ \small
+ \@clubpenalty\clubpenalty
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+%
+% \subsubsection{Index}%
+% FIXME: the following call to \cmd\twocolumn\ appears wrong if we were in two-column grid.
+% \changes{4.0c}{1999/11/13}{grid changes with ltxgrid}
+% \changes{4.0rc1}{2001/06/18}{grid changes with push and pop}
+% \begin{macrocode}
+\newenvironment{theindex}{%
+ \columnseprule \z@
+ \columnsep 35\p@
+ \c@secnumdepth-\maxdimen
+ \onecolumngrid@push
+ \section{\indexname}%
+ \thispagestyle{plain}%
+ \parindent\z@
+ \parskip\z@ plus.3\p@\relax
+ \let\item\@idxitem
+ \onecolumngrid@pop
+}{%
+%\onecolumngrid@pop
+}%
+%
+\def\@idxitem{\par\hangindent 40\p@}%
+%
+\def\subitem{\par\hangindent 40\p@ \hspace*{20\p@}}%
+%
+\def\subsubitem{\par\hangindent 40\p@ \hspace*{30\p@}}%
+%
+\def\indexspace{\par \vskip 10\p@ plus5\p@ minus3\p@\relax}%
+% \end{macrocode}
+%
+%
+% \subsection{Journal- and Pointsize-Specific Code}%
+%
+% After this substyle is read in, we will execute the code specific to the selected
+% journal: execute the society/journal \file{.rtx} file if it exists, or
+% execute the society/journal macro (if the latter is not defined, it will \cmd\relax\ out).
+% Here we define the default journal.
+% \begin{macrocode}
+\def\@journal@default{pra}%
+% \end{macrocode}
+%
+% \subsection{Typesize-Specific Code}%
+%
+% After this society file is read in, we will process the \cmd\@pointsize-specific
+% code. Here we define the default.
+% \begin{macrocode}
+\def\@pointsize@default{10}%
+% \end{macrocode}
+%
+% Note: the convention in \revtex\ and its substyles is that
+% the substyle must not override any explicit class options
+% declared by the document.
+% This means that the various Booleans of Section~\ref{sec:options}
+% may be assigned here only if they are still undefined at this point.
+%
+% For the APS, we supply code specific to journals PRA, PRB, PRC, PRD, PRE, PRL, PRX, PRAPPLIED, PRMATERIALS, PRFLUIDS, PRAB (was PRSTAB), PRPER (was PRSTPER), and RMP.
+% At present, they are identical,
+% with the exception of PRL and RMP.
+% We also introduce a new generic physrev style now that all of the Phys. Rev. journals
+% are identical
+% In 4.2, we make the inclusion of titles in the bibliography the default
+%
+% For most all of the APS journals, the journal-dependent
+% code is relatively meager and is therefore embedded in this file.
+% However, the RMP code is sufficiently extensive that
+% splitting it out into a separate file is more convenient.
+%
+% \subsubsection{\classoption{pra}}%
+% There is no code specific to \classoption{pra}.
+% \begin{macrocode}
+\def\rtx@apspra{%
+ \class@info{APS journal PRA selected}%
+}%
+% \end{macrocode}
+%
+% \subsubsection{\classoption{prb}}%
+% There is no code specific to \classoption{prb}.
+% \begin{macrocode}
+\def\rtx@apsprb{%
+ \class@info{APS journal PRB selected}%
+}%
+% \end{macrocode}
+%
+% \subsubsection{\classoption{prc}}%
+% There is no code specific to \classoption{prc}.
+% \begin{macrocode}
+\def\rtx@apsprc{%
+ \class@info{APS journal PRC selected}%
+}%
+% \end{macrocode}
+%
+% \subsubsection{\classoption{prd}}%
+% There is no code specific to \classoption{prd}.
+% \begin{macrocode}
+\def\rtx@apsprd{%
+ \class@info{APS journal PRD selected}%
+}%
+% \end{macrocode}
+%
+% \subsubsection{\classoption{pre}}%
+% There is no code specific to \classoption{pre}.
+% \begin{macrocode}
+\def\rtx@apspre{%
+ \class@info{APS journal PRE selected}%
+}%
+% \end{macrocode}
+%
+% \subsubsection{\classoption{prl}}%
+% \begin{macrocode}
+\def\rtx@apsprl{%
+ \class@info{APS journal PRL selected}%
+% \end{macrocode}
+% In PRL, the default is the \classoption{bibnotes} option,
+% and the Acknowledgments section has no head.
+%
+% The References head is a device that may be described as a lozenge centered on the baseline,
+% 71 points wide by 2 points thick, with the ends tapering to a half point in thickness.
+% Space above 26 points base to base, below 31 base to base.
+%FIXME: this code may confound \classname{geometry}
+% \begin{macrocode}
+ \let\frontmatter@footnote@produce\frontmatter@footnote@produce@endnote
+ \@booleanfalse\acknowledgments@sw
+ \appdef\setup@hook{%
+ \def\bibsection{%
+ \par
+ \begingroup
+ \baselineskip26\p@
+ \bib@device{\hsize}{72\p@}%
+ \endgroup
+ \nobreak\@nobreaktrue
+ \addvspace{19\p@}%
+ }%
+ }%
+% \end{macrocode}
+% Implement length checking.
+% Use the \classname{times} and \classname{mathtime} packages,
+% plus whatever other processing is required
+% to make the formatted output be true
+% to the metrics of the journal.
+%
+% \begin{macrocode}
+\appdef\setup@hook{%
+ \lengthcheck@sw{%
+ \RequirePackage{times}%
+% \end{macrocode}
+% Wait. Do not use mathtime after all. APS has their own way of doing math pi, involving
+% Adobe Mathematical Pi and other fonts.
+% \begin{verbatim}
+% \RequirePackage{mathtime}%
+% \end{verbatim}
+% \begin{macrocode}
+ }{}%
+}%
+% \end{macrocode}
+% A PRL does not have numbered sections.
+% \begin{macrocode}
+ \c@secnumdepth=-\maxdimen
+% \end{macrocode}
+% Note: we defer this code until after type size file is read in.
+% \begin{macrocode}
+ \appdef\setup@hook{%
+ \@ifnum{\@pointsize=10\relax}{%
+ \lengthcheck@sw{%
+ \def\large{%
+ \@setfontsize\large{12.5}{14\p@}%
+ }%
+ \def\normalsize{%
+ \@setfontsize\normalsize{10.5}\@xiipt
+ \abovedisplayskip 6\p@ \@plus6\p@ \@minus5\p@
+ \belowdisplayskip \abovedisplayskip
+ \abovedisplayshortskip \abovedisplayskip
+ \belowdisplayshortskip \abovedisplayskip
+ \let\@listi\@listI
+ }%
+ \def\small{%
+ \@setfontsize\small{9.5}\@xipt
+ \abovedisplayskip 5\p@ \@plus5\p@ \@minus4\p@
+ \belowdisplayskip \abovedisplayskip
+ \abovedisplayshortskip \abovedisplayskip
+ \belowdisplayshortskip \abovedisplayskip
+ \let\@listi\@listI
+ }%
+ \DeclareMathSizes{12.5}{12.5}{9}{6}%
+ \DeclareMathSizes{10.5}{10.5}{7.5}{5}%
+ \DeclareMathSizes{9.5}{9.5}{7.0}{5}%
+ }{%
+ \def\normalsize{%
+ \@setfontsize\normalsize\@xpt\@xiipt
+ \abovedisplayskip 10\p@ \@plus2\p@ \@minus5\p@
+ \belowdisplayskip \abovedisplayskip
+ \abovedisplayshortskip \abovedisplayskip
+ \belowdisplayshortskip \abovedisplayskip
+ \let\@listi\@listI
+ }%
+ }%
+ }{}%
+ }%
+ \textheight = 694.0\p@
+% \end{macrocode}
+%
+% End of \classoption{prl} code.
+% \begin{macrocode}
+}%
+% \end{macrocode}
+%
+% \subsubsection{\classoption{prper}}%
+% There is no code specific to \classoption{prper}
+% \begin{macrocode}
+\def\rtx@apsprper{%
+ \class@info{APS journal PRPER selected}%
+}%
+% \end{macrocode}
+%
+% \subsubsection{\classoption{prab}}%
+% There is no code specific to \classoption{prab}.
+% \begin{macrocode}
+\def\rtx@apsprab{%
+ \class@info{APS journal PRAB selected}%
+}%
+% \end{macrocode}
+%
+% \subsubsection{\classoption{prx}}%
+% There is no code specific to \classoption{prx}.
+% \begin{macrocode}
+\def\rtx@apsprx{%
+ \class@info{APS journal PRX selected}%
+}%
+% \end{macrocode}
+%
+% \subsubsection{\classoption{prapplied}}%
+% There is no code specific to \classoption{prapplied}.
+% \begin{macrocode}
+\def\rtx@apsprapplied{%
+ \class@info{APS journal PRApplied selected}%
+}%
+% \end{macrocode}
+%
+% \subsubsection{\classoption{prmaterials}}%
+% There is no code specific to \classoption{prmaterials}.
+% \begin{macrocode}
+\def\rtx@apsprmaterials{%
+ \class@info{APS journal PRMaterials selected}%
+}%
+% \end{macrocode}
+%
+% \subsubsection{\classoption{prfluids}}%
+% PRFluids uses a one-column format for journal format, but if authors want this, they should use the onecolumn option and not the reprint option.
+% Parsing of documentclass options is rather involved and hard to control precisely enough to have the correct behavior using the reprint option.
+% \begin{macrocode}
+\def\rtx@apsprfluids{%
+ \class@info{APS journal PRFluids selected}%
+ \@booleanfalse\titlepage@sw
+}%
+% \end{macrocode}
+%
+% \subsubsection{\classoption{physrev}}%
+% There is no code specific to \classoption{physrev}.
+% \begin{macrocode}
+\def\rtx@apsphysrev{%
+ \class@info{APS unified Physical Review journal style selected}%
+% \begin{macrocode}
+}%
+% \end{macrocode}
+%
+% \subsubsection{\classoption{rmp}}%
+% If this option has been selected, we will read in the needed code from the file \file{apsrmp.rtx}.
+%
+% \subsection{Establish APS Defaults}\label{sec:aps-defaults}
+%
+% \begin{macro}{\footinbib@sw}
+% All APS journals except RMP effectively invoke the \classoption{footinbib} option.
+% We rely on the RMP-specific code to override this assignment.
+% \begin{macrocode}
+\@booleantrue\footinbib@sw
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@bibdataout@init}
+% \begin{macro}{\@bibdataout@aps}
+% \changes{4.1o}{2010/02/12}{(AO, 580) Control .bst at run time.}%
+% Procedure \cmd\@bibdataout@aps\ has the job of writing the control record into the
+% job's \cmd\jobname\filename{.bib} file, where it will adjust the options to \filename{.bst} processing.
+% It is installed into the initialization procedure \cmd\@bibdataout@init, and its meaning
+% is set by the society (APS) and journal.
+% For all but RMP, we select the Physical Review style.
+% For the latter case, we change the meaning, per the code in \file{apsrmp.rtx}.
+%
+% \begin{macrocode}
+\appdef\@bibdataout@rev{\@bibdataout@aps}%
+\def\@bibdataout@aps{%
+ \immediate\write\@bibdataout{%
+% \end{macrocode}
+% An entry that controls processing of the \filename{.bst} file has entry type \texttt{@CONTROL}.
+% \begin{macrocode}
+ @CONTROL{%
+ apsrev42Control%
+% \end{macrocode}
+% Say whether we want long bibliography style (the default), or the abbreviated style.
+% \changes{4.2b}{2017/11/22}{(MD) MD - not sure why these parameters were different previously. Made them match except for title.}%
+% Use binary flags on control.* flags in bst file to set appropriate parameters
+% author = 08 corresponds to initials, jrnlst
+% editor = 1 corresponds to format identical to authors
+% title = 0 means to include title in journal references if present; title ="" means omit the title even if present (this should be the only difference between long and short bib styles)
+% year = 1 corresponds to truncate
+% page = 0 corresponds to using single page number rather than a range
+% \begin{macrocode}
+ \longbibliography@sw{%
+ ,author="08",editor="1",pages="0",title="0",year="1"%
+ }{%
+ ,author="08",editor="1",pages="0",title="",year="1"%
+ }%
+ }%
+ }%
+% \end{macrocode}
+% Place a \cmd\citation\ into the auxiliary file corresponding to this entry.
+% \begin{macrocode}
+ \if@filesw
+ \immediate\write\@auxout{\string\citation{apsrev42Control}}%
+ \fi
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\place@bibnumber}
+% \begin{macro}{\@bibstyle}
+% We install code that will select the presentation for \cmd\bibitem s
+% and govern the \BibTeX\ processing.
+% \begin{macrocode}
+\let\place@bibnumber\place@bibnumber@inl
+\def\@bibstyle{apsrev\substyle@post}%
+% \end{macrocode}
+% \begin{verbatim}
+%\appdef\setup@hook{%
+% \longbibliography@sw{%
+% \appdef\@bibstyle{long}%
+% }{}%
+%}%
+% \end{verbatim}
+% \end{macro}
+% \end{macro}
+%
+% \subsection{APS Sanity Checking}\label{sec:aps-sanity}
+% \changes{4.1f}{2009/07/20}{(AO, 524) Makes no sense if citations are superscript numbers and so are footnotes}
+%
+% Rule: if \cmd\place@bibnumber\ is \cmd\place@bibnumber@sup\
+% (citations are numbered and set superscript),
+% then it makes no sense for \cmd\footinbib@sw\ to be \cmd\false@sw\
+% (footnotes set in the bibliography, as endnotes).
+% If both conditions prevailed, then the document would have footnotes
+% and citations both as superscript arabic numbers, but independently numbered,
+% which would be confusing.
+%
+% Any society that provides for both superscript numbered citations as well as for
+% numbered footnotes should check for this same condition, and deal with it.
+%
+% Note: an alternative would be for footnotes to use the same sequence of footnote devices
+% that are used by the frontmatter footnotes
+% (\cmd\frontmatter@thefootnote\ instead of arabic numbers).
+%
+% In this case, we would want to refrain from resetting \cmd\c@footnote\ at the end of
+% the title page formatting.
+% We would also want to treat body footnotes identically to frontmatter footnotes:
+% the assignments in \cmd\titleblock@produce\ would persist throughout the document.
+%
+% But APS do not choose to go that route.
+% \begin{macrocode}
+\appdef\setup@hook{%
+ \@ifx{\place@bibnumber\place@bibnumber@sup}{%
+ \footinbib@sw{}{%
+ \class@warn{Citations are superscript numbers: footnotes must be endnotes; changing to that configuration}%
+ \@booleantrue\footinbib@sw
+ }%
+ }{}%
+}%
+% \end{macrocode}
+%
+% Here ends the substyle for society APS.
+% \begin{macrocode}
+%</aps>
+% \end{macrocode}
+%
+%
+% \section{The \classoption{rmp} journal substyle: the \texttt{rmp} module}
+%
+% The file \file{apsrmp.rtx} is read in by the \classname{revtex4} document class
+% if \cmd\@society\ has the value \classoption{aps}
+% and \cmd\@journal\ has the value \classoption{rmp}.
+%
+% It is read at the end of the \file{aps.rtx}, so all definitions
+% and assignments in that file are operative unless overridden here.
+%
+%
+% \begin{macrocode}
+%<*rmp>
+% \end{macrocode}
+%
+% Protect this file from being read in by anything but \revtex.
+% \begin{macrocode}
+\ifx\undefined\substyle@ext
+ \def\@tempa{%
+ \endinput
+ \GenericWarning{I must be read in by REVTeX! (Bailing out)}%
+ }%
+ \expandafter\else
+ \def\@tempa{}%
+ \expandafter\fi\@tempa
+% \end{macrocode}
+%
+% Protect this file from being read in as a society instead of a journal.
+% In such a case, \cmd\@journal\ will be undefined.
+% \begin{macrocode}
+\@ifxundefined\@journal{%
+% \end{macrocode}
+% One alternative: abort the document.
+% Another alternative: try to recover: force load the aps society file
+% \begin{macrocode}
+ \class@warn{Please specify the REVTeX options [aps,rmp]!}%
+ \@@end
+}{}%
+% \end{macrocode}
+% Log the journal substyle.
+% \begin{macrocode}
+ \class@info{APS journal RMP selected}%
+% \end{macrocode}
+%
+% \subsection{Frontmatter}%
+%
+% We assign the titlepage style for RMP; a document instance
+% may override by invoking one of the class options of \revtex.
+% \begin{macrocode}
+\clo@groupedaddress
+% \end{macrocode}
+%
+% \begin{macro}{\frontmatter@setup}
+%
+% \begin{macrocode}
+\def\frontmatter@setup{%
+ \normalfont\sffamily\raggedright
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\PACS@warn}
+% Per Mark Doyle, RMP never displays the PACS, so they don't want the `use showpacs' warning spit out.
+%\changes{4.2b}{2017/11/22}{(MD) PACS are obsolete altogether now}%
+% \begin{macrocode}
+\def\PACS@warn{RMP documents do not display PACS and PACS are obsolete. Your \string\pacs\space will be ignored}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@title@above}
+% \begin{macro}{\frontmatter@title@format}
+% \begin{macro}{\frontmatter@title@below}
+%
+% \begin{macrocode}
+\def\frontmatter@title@above{}%
+\def\frontmatter@title@format{\Large\bfseries\raggedright}% HelveticaNeue-Medium(Italic) 14pt.
+\def\frontmatter@title@below{\addvspace{12\p@}}% 24pt b-b down to first author
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@authorformat}
+% \changes{4.1b}{2008/08/04}{Rag the right even more: .8\cs{hsize}. Also, assign \cs{@totalleftmargin}.}%
+% \changes{4.1b}{2008/06/29}{(AO, 455) Be nice to a list within the abstract (assign \cs{@totalleftmargin}).}
+% Set the rag to a milder value, because we want to do true ragged right
+% typesetting, as opposed to the \LaTeX\ default, which gives very poor results.
+%
+% Note: author font is 9.8bp. 19.2bp/14.3bp above/below.
+% \begin{macrocode}
+\def\frontmatter@authorformat{%
+ \preprintsty@sw{\vskip0.5pc\relax}{}%
+ \@tempskipa\@flushglue
+ \@flushglue\z@ plus.8\hsize
+ \raggedright\advance\leftskip.5in\relax
+ \@flushglue\@tempskipa
+ \parskip\z@skip
+ \@totalleftmargin\leftskip
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@affiliationfont}
+% \changes{4.1b}{2008/08/04}{As with author formatting, rag the right more, and assign \cs{@totalleftmargin}. Also neutralize \cs{def@after@address}.}%
+% \changes{4.1b}{2008/06/29}{(AO, 455) Be nice to a list within the abstract (assign \cs{@totalleftmargin}).}
+% The hook \cmd\frontmatter@affiliationfont\ controls the formatting of affiliations and affiliation groups.
+% The hook \cmd\frontmatter@above@affilgroup\ is invoked just before proceeding with author/affiliation processing.
+% The \cmd\frontmatter@above@affiliation\ is the amount of space above affiliations in the \classoption{groupedaddress} style,
+% and \cmd\frontmatter@above@affiliation@script\ is that for \classoption{superscriptaddress}.
+%
+% Note: affiliation font is 9.03/10.4bp, 14.3bp/19.2bp b-b above/below
+% \begin{macrocode}
+\def\frontmatter@affiliationfont{% Helvetica 9/10.2
+ \small\slshape\selectfont\baselineskip10.5\p@\relax
+ \@tempskipa\@flushglue
+ \@flushglue\z@ plus.8\hsize
+ \raggedright\advance\leftskip.5in\relax
+ \@flushglue\@tempskipa
+ \@totalleftmargin\leftskip
+ \let\def@after@address\def@after@address@empty
+}%
+\def\frontmatter@above@affilgroup{\addvspace{7.2\p@}}% additional leading above an author
+\def\frontmatter@above@affiliation{\addvspace{5.3\p@}}%
+\def\frontmatter@above@affiliation@script{}%
+% \end{macrocode}
+% \end{macro}
+%
+% Set up the default RMP style for title block authors and affiliations.
+% We assign the titlepage style for RMP; a document instance
+% may override by invoking one of the class options of \revtex.
+% \changes{4.1b}{2008/08/04}{The \texttt{rmp} journal substyle selects \texttt{groupedaddress} by default.}%
+% \changes{4.1b}{2008/07/01}{(AO) make settings at class time instead of deferring them to later.}
+%
+% This command should override the effect of the
+% corresponding command in the society substyle,
+% and any document class option bearing on same will in turn override.
+% \begin{macrocode}
+\clo@groupedaddress
+% \end{macrocode}
+%
+% \begin{macro}{\frontmatter@RRAP@format}
+% \changes{4.1b}{2008/06/29}{(AO, 455) Be nice to a list within the abstract (assign \cs{@totalleftmargin}).}
+% Note: in RMP, if we are not in preprint mode, the date will not be produced.
+%
+% Note: Helvetica C/lc, 8.98bp, space above: 16.3bp b-b.
+% \begin{macrocode}
+\def\frontmatter@RRAP@format{%
+ \addvspace{7.3\p@}%
+ \small
+ \raggedright\advance\leftskip.5in\relax
+ \@totalleftmargin\leftskip
+}%
+\def\produce@RRAP#1{%
+ \@if@empty{#1}{}{%
+ \@ifvmode{\leavevmode}{}%
+ \unskip(\ignorespaces#1\unskip)\quad
+ }%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@abstractheading}
+% Space above 21.8bp b-b.
+% \begin{macrocode}
+\def\frontmatter@abstractheading{%
+ \preprintsty@sw{%
+ \begingroup
+ \centering\large
+ \abstractname
+ \par
+ \endgroup
+ \vspace{.5pc}%
+ }{}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@abstractfont}
+% TimesTen 8.93bp/9.6bp X 360bp, indented 36bp, with 21.9/37.6bp b-b above/below
+% \changes{4.1b}{2008/06/29}{(AO, 455) Be nice to a list within the abstract (assign \cs{@totalleftmargin}).}
+% \begin{macrocode}
+\def\frontmatter@abstractfont{%
+ \footnotesize
+ \hsize360\p@
+ \leftskip=0.5in
+ \parindent\z@
+ \@totalleftmargin\leftskip
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\frontmatter@preabstractspace}
+% \begin{macro}{\frontmatter@postabstractspace}
+% Space above and space below abstract in title block.
+% Should be 22/36 points base-to-base.
+% \begin{macrocode}
+\def\frontmatter@preabstractspace{7.7\p@}%
+\def\frontmatter@postabstractspace{24.6\p@}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% FIXME: Not done: PACS.
+% FIXME: TOC: Head is same as \section: HelveticaNeue 8.98. 32/22bp b-b above/below,
+% Body: TimesTen 8/10.5.
+%
+% \subsection{General Text}%
+%
+% \changes{4.1b}{2008/07/01}{(AO) make settings at class time instead of deferring them to later.}
+% If not in preprint mode, RMP sets the type size to 10/12 point. Note: s/b 11.6bp leading
+%FIXME: define \cmd\normalsize\ only if nobody else has done so.
+% \begin{macrocode}
+ \appdef\setup@hook{%
+ \preprintsty@sw{}{%
+ \def\normalsize{%
+ \@setsize\normalsize{12pt}\xpt\@xpt
+ \abovedisplayskip 10\p@ plus2\p@ minus5\p@
+ \belowdisplayskip \abovedisplayskip
+ \abovedisplayshortskip \abovedisplayskip
+ \belowdisplayshortskip \abovedisplayskip
+ \let\@listi\@listI
+ }%
+ }%
+ }%
+% \end{macrocode}
+%
+% Footnote mods:
+% \changes{4.1n}{2010/01/02}{(AO, 571) class file must set \cs{splittopskip}; fine tune \cs{skip}\cs{footins}; \cs{footnoterule} defined in terms of \cs{skip}\cs{footins}}%
+% \begin{macrocode}
+ \footnotesep 9.25pt
+ \skip\footins 36pt plus 4pt minus 12pt
+ \def\footnoterule{%
+ \dimen@\skip\footins\divide\dimen@\thr@@
+ \kern-\dimen@\hrule width.5in\kern\dimen@
+ }%
+% \end{macrocode}
+%
+%
+% \subsection{Sectioning}
+% \changes{4.1b}{2008/08/04}{Use \cs{setup@hook} to initialize all.}
+% \changes{4.1b}{2008/08/01}{Section numbering via procedures \cs{secnums@rtx} and \cs{secnums@arabic}.}
+%
+% We override the meaning of \cmd\secnums@rtx.
+% The class option \classoption{secnumarabic} will continue to work.
+% \begin{macrocode}
+\def\secnums@rtx{%
+ \@ifxundefined\thepart{%
+ \def\thepart{\Roman{part}}%
+ }{}%
+ \@ifxundefined\thesection{%
+ \def\thesection {\Roman{section}}%
+ \def\p@section {}%
+ }{}%
+ \@ifxundefined\thesubsection{%
+ \def\thesubsection {\Alph{subsection}}%
+ \def\p@subsection {\thesection.}%
+ }{}%
+ \@ifxundefined\thesubsubsection{%
+ \def\thesubsubsection {\arabic{subsubsection}}%
+ \def\p@subsubsection {\thesection.\thesubsection.}%
+ }{}%
+ \@ifxundefined\theparagraph{%
+ \def\theparagraph {\alph{paragraph}}%
+ \def\p@paragraph {\thesection.\thesubsection.\thesubsubsection.}%
+ }{}%
+ \@ifxundefined\thesubparagraph{%
+ \def\thesubparagraph {\arabic{subparagraph}}%
+ \def\p@subparagraph {\thesection.\thesubsection.\thesubsubsection.\theparagraph.}%
+ }{}%
+}%
+% \end{macrocode}
+%
+% In RMP, put a period (.), followed by `nut space', after the section number.
+% Also, hang the section number (the \LaTeX\ default).
+% \begin{macrocode}
+\def\@seccntformat#1{\csname the#1\endcsname.\hskip0.5em\relax}%
+% \end{macrocode}
+%
+% Note that we wish to set the section head uppercase, so we use
+% David Carlisle's \cmd\MakeTextUppercase. However, because this procedure effectively
+% parses its argument (looking for things to \emph{not} translate), it has to be invoked
+% in such a way that the argument of the \cmd\section\ command is passed to it as its
+% own argument.
+%
+% To accomplish this, we use the \cmd\@hangfrom@\ hook, which was developed for this
+% purpose.
+% \changes{4.1b}{2008/07/01}{Add \cs{@hangfroms@section}}
+% \begin{macrocode}
+ \def\section{%
+ \@startsection{section}{1}{\z@}{0.8cm plus1ex minus.2ex}{0.4cm}%
+ {%
+ \small\sffamily\bfseries\selectfont
+ \raggedright
+ \parindent\z@
+ }%
+ }%
+ \def\@hangfrom@section#1#2#3{\@hangfrom{#1#2}\MakeTextUppercase{#3}}%
+ \def\@hangfroms@section#1#2{#1\MakeTextUppercase{#2}}%
+ \def\subsection{%
+ \@startsection{subsection}{2}{\z@}{0.8cm plus1ex minus.2ex}{0.4cm}%
+ {%
+ \small\sffamily\bfseries
+ \raggedright
+ \parindent\z@
+ }%
+ }%
+ \def\subsubsection{%
+ \@startsection{subsubsection}{3}{\z@}{.8cm plus1ex minus.2ex}{0.4cm}%
+ {%
+ \small\sffamily\selectfont
+ \raggedright
+ \parindent\z@
+ }%
+ }%
+ \def\paragraph{%
+ \@startsection{paragraph}{4}{\z@}{.8cm plus1ex minus.2ex}{-1em}%
+ {%
+ \small\slshape\selectfont
+ \raggedright
+ \parindent\z@
+ }%
+ }%
+ \def\subparagraph{%
+ \@startsection{subparagraph}{4}{\parindent}{3.25ex plus1ex minus.2ex}{-1em}%
+ {\normalsize\bfseries\selectfont}%
+ }%
+%
+ \setcounter{tocdepth}{4}% FIXME: has no effect
+% \end{macrocode}
+%
+% \begin{macro}{\appendix}
+% \begin{macro}{\@hangfrom@appendix}
+% \begin{macro}{\@hangfroms@appendix}
+% \begin{macro}{\@appendixcntformat}
+% \changes{4.1b}{2008/07/01}{Add \cs{@hangfroms@section}}
+% \begin{macrocode}
+\appdef\appendix{%
+ \let\@hangfrom@section\@hangfrom@appendix
+ \let\@sectioncntformat\@appendixcntformat
+}%
+\def\@hangfrom@appendix#1#2#3{%
+ #1%
+ \@if@empty{#2}{%
+ #3%
+ }{%
+ #2\@if@empty{#3}{}{:\ #3}%
+ }%
+}%
+\def\@hangfroms@appendix#1#2{%
+ #1\appendixname\@if@empty{#2}{}{:\ #2}%
+}%
+\def\@appendixcntformat#1{\appendixname\ \csname the#1\endcsname}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+%
+% \subsection{Figure and Table Caption Formatting}
+% \begin{macro}{\@makecaption}
+% \changes{4.1b}{2008/07/01}{Break out \cs{@caption@fignum@sep}}
+%
+% \begin{macrocode}
+\setlength\belowcaptionskip{2\p@}
+\long\def\@makecaption#1#2{%
+ \vskip\abovecaptionskip
+ \vbox{%
+ \flushing
+ \small\rmfamily
+ \noindent
+ #1\@caption@fignum@sep#2\par
+ }%
+ \vskip\belowcaptionskip
+}%
+\def\@caption@fignum@sep{\nobreak\hskip.5em plus.2em\ignorespaces}%
+% \end{macrocode}
+% \end{macro}
+%
+% \subsection{Citations and Bibliography}%
+%
+% Customize \revtex\ for the journal substyle;
+% this task requires three components:
+% a \BibTeX\ \file{.bst} style file,
+% customizing code for \file{natbib}, and
+% customizations of the \env{thebibliography} environment.
+%
+% \begin{macro}{\@bibstyle}
+% Define the argument of the \cmd\bibliographystyle\ command
+% (if the document does not do so).
+%
+% The user must have installed a \file{.bst} file of the corresponding name.
+% This file will then be used by \BibTeX\ when compiling the document's \file{.bbl} file.
+%
+% To generate \file{apsrmp.bst}, use \classname{custom-bib} version 3.89d1 or later.
+% Run the \file{.bst} generator, \file{makebst.tex}, with the following options:
+%
+% \def\QandA#1#2#3{\item#1: \textbf{#2}: \texttt{#3}---}%
+% \def\CandE#1{\item\relax\texttt{#1}---}%
+% \begin{enumerate}
+% \QandA{STYLE OF CITATIONS}{a}{ay}Author-year with some non-standard interface
+% \QandA{AUTHOR}{*}{nat}Natbib for use with natbib v5.3 or later
+% \QandA{LANGUAGE FIELD}{l}{lang}Use language field to switch hyphenation patterns for title
+% \QandA{PRESENTATIONS}{b}{pres,pres-bf}Presentation, speaker bold face
+% \QandA{ORDER ON VON PART }{x}{vonx}Sort without von part (de la Maire after Mahone)
+% \QandA{AUTHOR NAMES}{a}{nm-rev1}Only first name reversed, initials (AGU style: Smith, J. F., H. K. Jones)
+% \QandA{POSITION OF JUNIOR }{*}{jnrlst}Junior comes last as Smith, John, Jr.
+% \QandA{TYPEFACE FOR AUTHORS IN LIST OF REFERENCES}{u}{nmft,nmft-def}User defined author font (\cmd\bibnamefont)
+% \QandA{FONT FOR FIRST NAMES }{u}{fnm-def}First names in user defined font (\cmd\bibfnamefont)
+% \QandA{EDITOR NAMES IN INCOLLECTION ETC}{a}{nmfted}Editors incollection like authors font
+% \QandA{FONT FOR `AND' IN LIST}{r}{nmand-rm}`And' in normal font (JONES and JAMES)
+% \QandA{FONT OF CITATION LABELS IN TEXT }{u}{lab,lab-def}User defined citation font (\cmd\citenamefont)
+% \QandA{FONT FOR `AND' IN CITATIONS }{r}{and-rm}Cited `and' in normal font
+% \QandA{LABEL WHEN AUTHORS MISSING }{*}{keyxyr}Year blank when KEY replaces missing author (for natbib 7.0)
+% \QandA{DATE POSITION}{b}{dt-beg}Date after authors
+% \QandA{DATE FORMAT }{m}{yr-com}Date preceded by comma as `, 1993'
+% \QandA{INCLUDE MONTHS}{m}{aymth}Include month in date
+% \QandA{REVERSED DATE }{r}{dtrev}Date as year month
+% \QandA{TRUNCATE YEAR }{*}{note-yr}Year text full as 1990--1993 or `in press'
+% \QandA{TITLE OF ARTICLE}{d}{tit-qq}Title and punctuation in double quotes (``Title,'' ..)
+% \QandA{TITLE PRESENT IN ARTICLE, INCOLLECTION, AND INPROCEEDINGS}{x}{jtit-x}Title is ignored
+% \QandA{INPROCEEDINGS CHAPTER AND PAGES, LIKE INBOOK}{y}{inproceedings-chapter}produce pages after chapter just as in InBook
+% \QandA{ARTICLE BOOKTITLE PRESENT}{?}{article-booktitle}format booktitle
+% \QandA{ARTICLE SERIES PRESENT}{?}{article-series}article can has series
+% \QandA{JOURNAL NAME FONT}{r}{jttl-rm}Journal name normal font
+% \QandA{JOURNAL NAME WITH ADDRESS}{y}{journal-address}Include address field (in parentheses) along with journal name
+% \QandA{BOOK TITLE FIELDS}{y}{book-bt}Field `booktitle', or if absent field `title', is book title
+% \QandA{THESIS TITLE OPTIONAL}{?}{thesis-title-o}Title is optional: no warning issued if empty
+% \QandA{TECHNICAL REPORT TITLE}{b}{trtit-b}Tech. report title like books
+% \QandA{JOURNAL VOLUME}{b}{vol-bf}Volume bold as {\bf vol}(num)
+% \QandA{JOURNAL VOL AND NUMBER}{s}{vnum-sp}Journal vol (num) as 34 (2)
+% \QandA{VOLUME PUNCTUATION}{c}{volp-com}Volume with comma as vol(num), ppp
+% \QandA{PAGE NUMBERS}{f}{jpg-1}Only start page number
+% \QandA{POSITION OF PAGES}{e}{pp-last}Pages at end but before any notes
+% \QandA{BOOK EDITOR W/O AUTHOR}{}{book-editor-booktitle}Book permits empty author, produces title before editor in this case
+% \QandA{INBOOK PERMITS TITLE, BOOKTITLE, AUTHOR, EDITOR}{a}{inbook-editor-booktitle}Allow using both title/booktitle, both author/editor
+% \QandA{CONFERENCE ADDRESS FOR BOOK, INBOOK, INCOLLECTION, INPROCEEDINGS, PROCEEDINGS}{a}{bookaddress}Italic booktitle followed by bookaddress in roman
+% \QandA{NUMBER AND SERIES FOR BOOK, INBOOK, INCOLLECTION, INPROCEEDINGS, PROCEEDINGS}{*}{num-xser}Allows number without series and suppresses word "number"
+% \QandA{WORD NUMBER CAPITALIZED FOR NUMBER AND SERIES}{c}{number-cap}Capitalize word `number' as: "Number 123"
+% \QandA{WORD CHAPTER CAPITALIZED}{c}{chapter-cap}Capitalize word `chapter' as: `Chapter 42'
+% \QandA{COMBINING NUMBER AND SERIES}{x}{series-number}Series number as: `Springer Lecture Notes No. 125'
+% \QandA{POSITION OF NUMBER AND SERIES}{b}{numser-booktitle}After book title and conference address, and before editors
+% \QandA{VOLUME AND SERIES FOR BOOKS}{s}{ser-vol}Series, vol. 23
+% \QandA{VOLUME AND SERIES FORMATTING}{y}{ser-rm}format series roman , even when used with volume
+% \QandA{WORD VOLUME CAPITALIZED FOR VOLUME AND SERIES}{y}{volume-cap}Capitalize word `volume', as: `Volume 7 in Lecture Series'
+% \QandA{POSITION OF VOLUME AND SERIES FOR INCOLLECTION, INBOOK, AND INPROCEEDINGS}{e}{ser-ed}Series and volume after booktitle and before editors
+% \QandA{JOURNAL NAME PUNCTUATION}{x}{jnm-x}Space after journal name
+% \QandA{PAGES IN BOOK}{*}{pg-bk,book-chapter-pages}As chapter and page: chapter 42, page 345
+% \QandA{PUBLISHER IN PARENTHESES}{p}{pub-par}Publisher in parentheses
+% \QandA{EMPTY PUBLISHER PARENTHESES}{y}{ay-empty-pub-parens-x}eliminate parentheses altogether if nothing inside
+% \QandA{PUBLISHER POSITION}{e}{pre-edn}Edition before publisher
+% \QandA{SCHOOL}{p}{school-par}school/address in parens: `(school, address)'
+% \QandA{ISBN NUMBER}{*}{isbn}Include ISBN for books, booklets, etc.
+% \QandA{ISSN NUMBER}{*}{issn}Include ISSN for periodicals
+% \QandA{DOI NUMBER}{a}{doi-link,doi}Doi forms a link to the publication, anchored to the volume or title
+% \QandA{`EDITOR' AFTER NAMES}{a}{bkedcap}`Name Editor,' as above, editor upper case
+% \QandA{EDITOR IN COLLECTIONS}{b}{edby}In booktitle, edited by .. (where .. is names)
+% \QandA{PUNCTUATION BETWEEN SECTIONS }{c}{blk-com}Comma between blocks
+% \QandA{FINAL PUNCTUATION}{c}{fin-endbibitem}Command at end instead of period
+% \QandA{ABBREVIATE WORD `PAGES' }{a}{pp}`Page' abbreviated as p. or pp.
+% \QandA{ABBREVIATE WORD `EDITORS'}{a}{ed}`Editor' abbreviated as ed. or eds.
+% \QandA{OTHER ABBREVIATIONS}{a}{abr}Abbreviations of such words
+% \QandA{ABBREVIATION FOR `EDITION' }{a}{ednx}`Edition' abbreviated as `ed'
+% \QandA{EDITION NUMBERS}{n}{ord}Numerical editions as 1st, 2nd, 3rd, etc
+% \QandA{STORED JOURNAL NAMES}{a}{jabr}Abbreviated journal names
+% \QandA{COMMA BEFORE `AND'}{c}{and-com}Comma even with 2 authors as `Tom, and Harry'
+% \QandA{FONT OF `ET AL'}{i}{etal-it}Italic et al
+% \QandA{ADDITIONAL REVTeX DATA FIELDS}{r}{revdata,eprint,url,url-blk,translation}Include REVTeX data fields collaboration, eid, eprint, archive, url, translation
+% \QandA{SLACcitation FIELD}{?}{SLACcitation}Produce SLACcitation field
+% \QandA{NUMPAGES DATA FIELD}{*}{numpages-x}Do not include numpages field
+% \QandA{REFERENCE COMPONENT TAGS}{b}{bibinfo}Reference component tags like \cmd\bibinfo\ in the content of \cmd\bibitem
+% \QandA{ELEMENT TAGS}{b}{bibfield}Element tags like \cmd\bibfield\ in the content of \cmd\bibitem
+% \QandA{COMPATIBILITY WITH PLAIN TEX}{*}{nfss}Use LaTeX commands which may not work with Plain TeX
+% \end{enumerate}
+%
+% A file \file{apsrmp.dbj} file equivalent to the following should result:
+% \begin{verbatim}
+%\input docstrip
+%\preamble
+%----------------------------------------
+%*** REVTeX-compatible RMP 2010-02-12 ***
+%\endpreamble
+%\postamble
+%End of customized bst file
+%\endpostamble
+%\keepsilent
+%\askforoverwritefalse
+%\def\MBopts{\from{merlin.mbs}{%
+% head,\MBopta}
+%\from{physjour.mbs}{\MBopta}
+%\from{geojour.mbs}{\MBopta}
+%\from{photjour.mbs}{\MBopta}
+%\from{merlin.mbs}{tail,\MBopta}}
+%\def\MBopta{%
+% ay,%: Author-year with some non-standard interface
+% nat,%: Natbib for use with natbib v5.3 or later
+% lang,%: Use language field to switch hyphenation patterns for title
+% pres,pres-bf,%: Presentation, speaker bold face
+% vonx,%: Sort without von part (de la Maire after Mahone)
+% nm-rev1,%: Only first name reversed, initials (AGU style: Smith, J. F., H. K. Jones)
+% jnrlst,%: Junior comes last as Smith, John, Jr.
+% nmft,nmft-def,%: User defined author font (\bibnamefont)
+% fnm-def,%: First names in user defined font (\bibfnamefont)
+% nmfted,%: Editors incollection like authors font
+% nmand-rm,%: `And' in normal font (JONES and JAMES)
+% lab,lab-def,%: User defined citation font (\citenamefont)
+% and-rm,%: Cited `and' in normal font
+% keyxyr,%: Year blank when KEY replaces missing author (for natbib 7.0)
+% dt-beg,%: Date after authors
+% yr-par,%: Year in parentheses as (1993)
+% dtrev,%: Date as year month
+% date-nil-x,%: If date is empty, then do not produce the surrounding punctuation (parens, brackets, colon, comma)
+% tit-qq,%: Title and punctuation in double quotes (``Title,'' ..)
+% inproceedings-chapter,%: produce pages after chapter, just as in InBook
+% jtit-x,%: Title is ignored
+% inproceedings-chapter,%: produce pages after chapter just as in InBook
+% article-booktitle,%: format booktitle
+% article-series,%: article can has series
+% jttl-rm,%: Journal name normal font
+% journal-address,%: Include address field (in parentheses) along with journal name
+% book-bt,%: Field `booktitle', or if absent field `title', is book title
+% thesis-title-o,%: Title is optional: no warning issued if empty
+% trtit-b,%: Tech. report title like books
+% techreport-institution-par,%: format tech report institution like book publisher
+% vol-bf,%: Volume bold as {\bf vol}(num)
+% vnum-sp,%: Journal vol (num) as 34 (2)
+% volp-com,%: Volume with comma as vol(num), ppp
+% jpg-1,%: Only start page number
+% pp-last,%: Pages at end but before any notes
+% book-editor-booktitle,%: Book permits empty author, produces title before editor in this case
+% inbook-editor-booktitle,%: Allow using both title/booktitle, both author/editor
+% bookaddress,%: Italic booktitle followed by bookaddress in roman
+% num-xser,%: Allows number without series and suppresses word "number"
+% number-cap,%: Capitalize word `number' as: "Number 123"
+% chapter-cap,%: Capitalize word `chapter' as: `Chapter 42'
+% series-number,%: Series number as: `Springer Lecture Notes No. 125'
+% numser-booktitle,%: After book title and conference address, and before editors
+% ser-vol,%: Series, vol. 23
+% ser-rm,%: format series roman , even when used with volume
+% volume-cap,%: Capitalize word `volume', as: `Volume 7 in Lecture Series'
+% ser-ed,%: Series and volume after booktitle and before editors
+% jnm-x,%: Space after journal name
+% pg-bk,book-chapter-pages,%: As chapter and page: chapter 42, page 345
+% pub-par,%: Publisher in parentheses
+% ay-empty-pub-parens-x,%: eliminate parentheses altogether if nothing inside
+% pre-edn,%: Edition before publisher
+% school-par,%: school/address in parens: `(school, address)'
+% isbn,%: Include ISBN for books, booklets, etc.
+% issn,%: Include ISSN for periodicals
+% doi-link,doi,%: Doi forms a link to the publication, anchored to the volume or title
+% bkedcap,%: `Name Editor,' as above, editor upper case
+% edby,%: In booktitle, edited by .. (where .. is names)
+% blk-com,%: Comma between blocks
+% fin-endbibitem,%: Command at end instead of period
+% pp,%: `Page' abbreviated as p. or pp.
+% ed,%: `Editor' abbreviated as ed. or eds.
+% abr,%: Abbreviations of such words
+% ednx,%: `Edition' abbreviated as `ed'
+% ord,%: Numerical editions as 1st, 2nd, 3rd, etc
+% jabr,%: Abbreviated journal names
+% and-com,%: Comma even with 2 authors as `Tom, and Harry'
+% etal-it,%: Italic et al
+% revdata,eprint,url,url-blk,translation,%: Include REVTeX data fields collaboration, eid, eprint, archive, url, translation
+% SLACcitation,%: Produce SLACcitation field
+% numpages-x,%: Do not include numpages field
+% url,url-prefix-x,%: URL without prefix (default: `URL ')
+% bibinfo,%: Reference component tags like \bibinfo in the content of \bibitem
+% bibfield,%: Element tags like \bibfield in the content of \bibitem
+% nfss,%: Use LaTeX commands which may not work with Plain TeX
+%,{%
+% }}
+%\generate{\file{apsrmp4-2.bst}{\MBopts}}
+%\endbatchfile
+% \end{verbatim}
+%
+% For a comparison between apsrmp.bst and apsrev.bst, see Section~\ref{sec:CompRevRmp}.
+%
+% \changes{4.1c}{2008/08/15}{Document class option longbibliography via \cs{substyle@post}}
+% \begin{macrocode}
+\def\@bibstyle{apsrmp\substyle@post}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\authoryear@sw}
+% \changes{4.1a}{2008/01/18}{(AO, 457) Endnotes to be sorted in with numerical citations.}%
+% Author-year citations: default value of \cmd\authoryear@sw\ is true.
+% \begin{macrocode}
+\@booleantrue\authoryear@sw
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\@bibdataout@rmp}
+% \changes{4.1o}{2010/02/12}{(AO, 580) Control .bst at run time.}%
+% When the journal is RMP, the meaning of the procedure \cmd\@bibdataout@aps\ needs to be different
+% because of the way the author names are formatted. In other respects, it is the same.
+% \begin{macrocode}
+\def\@bibdataout@aps{%
+ \immediate\write\@bibdataout{%
+% \end{macrocode}
+% An entry that controls processing of the \filename{.bst} file has entry type \texttt{@CONTROL}.
+% This entry's cite key is \texttt{apsrmp41Control}, which serves as a version number.
+% \begin{macrocode}
+ @CONTROL{%
+ apsrmp41Control%
+% \end{macrocode}
+% Say whether we want long bibliography style (the default), or the abbreviated style.
+% \begin{macrocode}
+ \longbibliography@sw{%
+ ,author="03",editor="0",pages="1",title="0",year="0"%
+ }{%
+ ,author="0B",editor="0",pages="0",title="0",year="1"% TeXSupport
+ }%
+ }%
+ }%
+% \end{macrocode}
+% Place a \cmd\citation\ into the auxiliary file corresponding to this entry.
+% \begin{macrocode}
+ \if@filesw
+ \immediate\write\@auxout{\string\citation{apsrmp41Control}}%
+ \fi
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\bibpunct}
+% \begin{macro}{\bibsection}
+% \begin{macro}{\bibpreamble}
+% \begin{macro}{\newblock}
+% \begin{macro}{\bibhang}
+% \begin{macro}{\bibsep}
+% \begin{macro}{\cite}
+% The following commands effectively establish the style in which \cmd\cite\ commands are
+% formatted.
+% You can think of them as the second needed component for the bibliography.
+%
+% Set up for author-year citations:
+% when \cmd\NAT@set@cites\ executes (at \envb{document} time), the \cmd\@biblabel\ will
+% be set to \cmd\NAT@biblabel.
+%
+% Per Karie Friedman (friedman@phys.washington.edu), multiple citations are separated by semicolons, e.g.,
+% (Jones, 1999; Abbott and Smith, 2000; Wortley, 2001a),
+% and
+% multiple citations by the same author by commas, e.g.,
+% Abela et al. (1995, 1997a, 1997b).
+% The third argument of \cmd\bibpunct\ handles the former.
+%
+% The fifth argument puts a comma after the author when the year is not in parens:
+% (Lee et al., 1996).
+%
+% Incidently, this \cmd\bibpunct\ command specifies the \classname{natbib} default values.
+%
+% We define the sectioning command to use when starting the bibliography.
+%
+% We change \classname{natbib}'s \cmd\NAT@def@citea\ procedure to effect more elaborate
+% punctuation for RMP: see item 473:
+% \cmd\cite\ order punctuation: ``If possible, \cmd\textcite should put the word 'and' between two citations and before the last citation in a list of 3 or more.''
+%
+% \begin{macrocode}
+\appdef\setup@hook{%
+% \end{macrocode}
+%
+% We define the punctuation to use in the \cmd\cite\ command's production.
+%
+% \begin{macrocode}
+ \bibpunct{(%)
+ }{%(
+ )}{;}{a}{,}{,}%
+% \end{macrocode}
+%
+% We define the sectioning command to use when starting the bibliography.
+%
+% \begin{macrocode}
+ \def\bibsection{%
+ \expandafter\section\expandafter*\expandafter{\refname}%
+ \@nobreaktrue
+ }%
+ \let\bibpreamble\@empty
+ \def\newblock{\ }%
+ \bibhang10\p@
+ \bibsep\z@
+% \end{macrocode}
+%
+% Per Mark Doyle, \cmd\cite\ is mapped to \cmd\citep\ in RMP.
+% \changes{4.1i}{2009/10/23}{(AO, 541) Defer assignment of \cs{cite} until after natbib loads}
+% \begin{macrocode}
+ \let\cite\citep
+% \end{macrocode}
+%
+% End of code to be delayed until after \classname{natbib} loads.
+% \begin{macrocode}
+}%
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\footinbib@sw}
+% Footnotes in bibliography are consistent only with numbered citations,
+% and are particularly nasty under \classname{natbib}:
+% the packcage will automatically change to numbered references if
+% any \cmd\bibitem\ commands lack the optional argument.
+% Therefore, we must uninvoke it now, even if invoked by the document.
+% The same is quietly done with
+% \classname{natbib}'s \classoption{mcite} and \classoption{compress} options.
+%
+% \changes{4.1f}{2009/07/15}{(AO, 523) Add class option nomerge, to turn off new natbib 8.3 syntax}
+% (AO 523) I changed the code that alters \cmd\NAT@merge\ so that it will not override
+% when \cmd\NAT@merge\ has been set to \cmd\z@.
+% \begin{macrocode}
+\@booleanfalse\footinbib@sw
+\appdef\setup@hook{%
+ \footinbib@sw{%
+ \class@warn{%
+ Footnotes in bibliography are incompatible with RMP.^^J%
+ Undoing the footinbib option.
+ }%
+ \@booleanfalse\footinbib@sw
+ }{}%
+ \@ifnum{\NAT@merge>\@ne}{\let\NAT@merge\@ne}{}%
+ \def\NAT@cmprs{\z@}%
+}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\eprint}
+% RMP requires the \cmd\eprint\ field in the bib entry to be set off with the word ``eprint''.
+% \begin{macrocode}
+\def\eprint#1{eprint #1}%
+% \end{macrocode}
+% \end{macro}
+%
+% \subsection{Table of Contents}
+%
+% We set up for auto-sizing of certain TOC elements.
+%
+% To do this, we override the definitions for
+% the default TOC font (\cmd\toc@@font),
+% and define
+% formatting for the needed elements (\cmd\l@\dots).
+% Finally, we activate the autosizing by
+% assigning \cmd\toc@pre\ and \cmd\toc@post.
+%
+% \begin{macro}{\toc@@font}
+% \changes{4.1a}{2008/01/19}{(AO, 461) Change the csname revtex uses from @dotsep to ltxu@dotsep. The former is understood in mu. (What we wanted was a dimension.)}%
+% Set the formatting characteristics of the auto-indenting part
+% of the TOC.
+% \begin{macrocode}
+\def\toc@@font{%
+ \footnotesize\rmfamily
+ \def\\{\space\ignorespaces}%
+}%
+\def\ltxu@dotsep{5.5pt}%
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\l@section}
+% Determine which TOC elements are automatically indented.
+% \begin{macrocode}
+\def\tocleft@{\z@}%
+\def\tocdim@min{5\p@}%
+\def\l@section{%
+ \l@@sections{}{section}% Implicit #3#4
+}%
+\def\l@subsection{%
+ \l@@sections{section}{subsection}% Implicit #3#4
+}%
+\def\l@subsubsection{%
+ \l@@sections{subsection}{subsubsection}% Implicit #3#4
+}%
+%\def\l@subsubsection#1#2{}%
+\def\l@paragraph#1#2{}%
+\def\l@subparagraph#1#2{}%
+% \end{macrocode}
+% \end{macro}
+%
+% Activate the TOC processing.
+% \begin{macrocode}
+\let\toc@pre\toc@pre@auto
+\let\toc@post\toc@post@auto
+% \end{macrocode}
+%
+%
+% \begin{macrocode}
+%</rmp>
+% \end{macrocode}
+%
+%
+% \Finale
+% %Here ends the programmer's documentation.
+% \endinput
+%
+\endinput