summaryrefslogtreecommitdiff
path: root/Master
diff options
context:
space:
mode:
Diffstat (limited to 'Master')
-rw-r--r--Master/texmf-dist/source/latex/cjw/cjw-latex.dtx2333
-rw-r--r--Master/texmf-dist/source/latex/cjw/cjw-latex.ins44
-rw-r--r--Master/texmf-dist/source/latex/cjw/cjw-ltr.dtx821
-rw-r--r--Master/texmf-dist/source/latex/cjw/cjw-ltr.ins18
-rw-r--r--Master/texmf-dist/source/latex/cjw/cjwoutl.dtx656
-rw-r--r--Master/texmf-dist/source/latex/cjw/cjwoutl.ins18
-rw-r--r--Master/texmf-dist/source/latex/cjw/cjwplain.dtx1007
-rw-r--r--Master/texmf-dist/source/latex/cjw/cjwplain.ins20
8 files changed, 4917 insertions, 0 deletions
diff --git a/Master/texmf-dist/source/latex/cjw/cjw-latex.dtx b/Master/texmf-dist/source/latex/cjw/cjw-latex.dtx
new file mode 100644
index 00000000000..515dd7d3d0b
--- /dev/null
+++ b/Master/texmf-dist/source/latex/cjw/cjw-latex.dtx
@@ -0,0 +1,2333 @@
+% \iffalse -*- LaTeX -*-
+%
+% This is file `cjw-latex.dtx'. You can run this file through
+% LaTeX2e to produce a DVI file of documentation. The file
+% `cjw-latex.ins' should have come with this file. Run it through
+% (La)TeX to extract the bundled macro files.
+%
+% \fi
+\def\RCSinfo{$Id: cjw-latex.dtx,v 0.13 1998/09/01 15:54:20 cwynne Exp $}
+
+\def\RCSsplit $#1: #2,v #3 #4 #5 #6 #7${
+ \gdef\filename {#2}
+ \gdef\fileversion{#3}
+ \gdef\filedate {#4}
+ \gdef\filetime {#5}
+ \gdef\fileauthor {#6}
+ \gdef\filelocker {#7}}
+
+\expandafter\RCSsplit\RCSinfo
+%%
+%% \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 \~}
+%%
+% \CheckSum{0}
+%
+% \iffalse
+%<*driver>
+
+\NeedsTeXFormat{LaTeX2e}
+\ProvidesFile{cjw-latex.dtx}
+ [\filedate\space v\fileversion\space
+ Personal macros for LaTeX2e---CJW]
+
+\documentclass{ltxdoc}
+ \usepackage{cjwmacro}
+ % We need math---call it with all options to make sure they work.
+ \usepackage[derivs,integrals,physics]{cjwmath}
+
+\GetFileInfo{cjw-ltr.dtx}
+\EnableCrossrefs
+% \DisableCrossrefs % Say \DisableCrossrefs if index is ready
+% \OnlyDescription % Comment out for implementation details
+\RecordChanges
+
+\begin{document}
+ \DocInput{cjw-latex.dtx}
+\end{document}
+%</driver>
+% \fi
+%
+% \DeclareRobustCommand{\cseq}[1]{\texttt{\bslash#1}}
+% \DeclareRobustCommand{\pkg} [1]{\textsf{#1}}
+% \DeclareRobustCommand{\env} [1]{\textsf{#1}}
+%
+% \renewcommand{\thefootnote}{\fnsymbol{footnote}}
+%
+% \title{The \pkg{cjw-latex} Macro Collection\thanks{%
+% This file has version \fileversion{} as of \filedate.}}
+% \author{Colin J.~Wynne\thanks{E-Mail at:
+% \texttt{cwynne@mts.jhu.edu}, \texttt{cwynne@jhu.edu}\,.}}
+% \date{\filedate}
+%
+% \maketitle
+%
+% \setcounter{StandardModuleDepth}{1}
+%
+% {\parskip 0pt ^^A % This is the hack used by |doc.dtx|.
+% ^^A % (bug in \LaTeX?)
+% \tableofcontents
+% }
+%
+% \section*{Introduction}
+%
+% I have been a \TeX{} user for quite a long time now. It was in my
+% junior year in college, in 1992, that one of my friends and one of
+% my math professors decided to warp my perception of reality and
+% introduce me to Dr.~Knuth's wonderful creation. In those days,
+% Plain-\TeX{} was the tool of choice, we coded everything ourselves
+% from primitives on up, logical markup was unknown, and \LaTeX{} was
+% known as Lame-\TeX. During my senior year I wrote an honors thesis
+% in mathematics which required quite a lot of things not present in
+% standard Plain-\TeX. My macro files grew, and grew, and\dots
+%
+% In the year following my graduation, I converted myself to \LaTeX{},
+% mostly because of the convergence my own personal modified
+% ePlain/NFSS1/Plain format was having towards \LaTeX{} in terms of
+% logical markup. Most of macros were easily converted into
+% \LaTeX{}-ese. They got more complicated from there.
+%
+% So now I regularly write up papers, letters, mathematical problem
+% sets, and just about anything else that uses the English language,
+% in \LaTeX. The macros have evolved quite a bit. More has been
+% added. I took to using the |dtx| format for most of my input files,
+% and finally decided to wrap my big three up into a single documented
+% source file.
+%
+% I hope that these macros will prove useful to somebody out there,
+% and if they do, feel free to buy me a beer next time you see me. I
+% have other |dtx| files available, including a modified \pkg{letter}
+% class which also does German formal letters, and a package for doing
+% outlines. Any package of mine should be identifiable on your
+% friendly neighborhood CTAN site with a name like
+% \verb+cjw*.(dtx|ins)+.
+%
+% \section{General macros}
+%
+% \iffalse
+%
+\NeedsTeXFormat{LaTeX2e}
+%<general>\ProvidesPackage{cjwmacro}
+%<math>\ProvidesPackage{cjwmath}
+%<deriv>\ProvidesFile{cjwderiv.tex}
+%<integ>\ProvidesFile{cjwinteg.tex}
+%<phys>\ProvidesFile{cjwphys.tex}
+%<units>\ProvidesFile{cjwunits}
+ [\filedate\space v\fileversion\space
+%<general> Personal macros for LaTeX2e---CJW]
+%<math> Math macros for LaTeX2e---CJW]
+%<deriv> Derivative macros for cjwmath.sty---CJW]
+%<integ> Integration macros for cjwmath.sty---CJW]
+%<phys> Physics macros for cjwmath.sty---CJW]
+%<units> Typesetting units in LaTeX2e---CJW]
+%
+%<*general>
+%
+%\fi
+%
+% I tend to organize my package files as follows---first come flow
+% control structures for the package itself, usually in the form of
+% new conditionals; then come the options; then comes the meat of the
+% package in some sort of vaguely thought out order. This package is
+% no exception.
+%
+% \subsection{Package initialization}
+%
+% Conditionals are usually used in conjunction with package options to
+% provide conditional inclusion of certain code, either \emph{via} an
+% \cseq{if}\dots{}\cseq{endif} block or using class options. For this
+% package, the subsystem in question is the inclusion of verbatim
+% typeset files.
+% \begin{macrocode}
+\newif\if@verbext \@verbextfalse
+% \end{macrocode}
+% Not surprisingly, this conditional is used directly by an option.
+% \begin{macrocode}
+\DeclareOption{verbext}{\@verbexttrue}
+% \end{macrocode}
+% I used to use options for the loading of additional packages. In
+% particular, when I used \pkg{psfig.new} which could not be handled
+% as a \LaTeXe{} package, I did this. Now that I use the \pkg{epsfig}
+% package that comes with \LaTeXe, I simply issue a warning to
+% include the package separately.
+% \begin{macrocode}
+\DeclareOption{psfig}{%
+ \PackageWarning{cjwmacro}%
+ {Obsolete option \CurrentOption. Use package `epsfig' instead.}}
+% \end{macrocode}
+% Since, however, using \pkg{pstricks} requires several files, I still
+% use a package option to take care of all of that. This option
+% checks for the existence of \emph{both} files before including
+% either, hence the nested calls to \cseq{InputIfFileExists}. A
+% similar option is used for \pkg{pst-plot.tex}, since is not
+% implemented as a package file.
+% \begin{macrocode}
+\DeclareOption{pstricks}{%
+ \InputIfFileExists{pstricks.sty}{%
+ \InputIfFileExists{pst-node.tex}{}{%
+ \PackageError{cjwmacro}{File `pst-node.sty' not found.}{}}}%
+ {\PackageError{cjwmacro}{File `pstricks.sty' not found.}{}}}
+
+\DeclareOption{psplot}{\InputIfFileExists{pst-plot.tex}{}{%
+ \PackageError{cjwmacro}{File `pst-plot.tex' not found.}{}}}
+% \end{macrocode}
+%
+% The next two options are used to change behavior of some macros on
+% draft as opposed to final copies. Currently, only \cseq{ssbreak}
+% has such a dependency, and in final form it uses the PS-Tricks
+% package to typeset a nice section delimiter. Note the use of
+% \cseq{ExecuteOption} by the \pkg{final} option to make sure that
+% PS-Tricks is, indeed, available.
+% \begin{macrocode}
+% What to do for draft vs. final copy.
+\DeclareOption{draft}{%
+ \def\ssbreakbar{\hbox to 2in{\hrulefill}}}
+\DeclareOption{final}{%
+ \ExecuteOptions{pstricks}
+ \def\ssbreakbar{%
+ \psset{linewidth=0.4pt,unit=1in}%
+ \pspicture(-2.5,-0.15)(2.5,0.15)%
+ \qdisk(0,0){0.04}%
+ \qdisk(0.33,0){0.02}%
+ \qdisk(-0.33,0){0.02}%
+ \pspolygon*(0.33,-0.02)(0.33,0.02)(1.75,0)%
+ \pspolygon*(-0.33,-0.02)(-0.33,0.02)(-1.75,0)%
+ \endpspicture}}
+% \end{macrocode}
+%
+% To finish off option handling, we declare a default (warn about
+% unknown options), execute defaults, and process the passed option
+% list.
+% \begin{macrocode}
+\DeclareOption*{%
+ \PackageWarning{cjwmacro}{Unknown option `\CurrentOption'}}
+\ExecuteOptions{draft}
+\ProcessOptions
+% \end{macrocode}
+%
+% \subsection{General definitions}
+%
+% These general definitions set up some `meta-macros', to be used by
+% other commands.
+%
+% One of the things I liked a lot about \TeX{} was the use of
+% \cseq{let} to make aliases for existing commands without
+% sacrificing much of the control sequence space. In the spirit of
+% \LaTeXe, however, I have implemented this thrice with name
+% checking. Analogously to the \cseq{newcommand}-like macros, we
+% offer the following three:
+% \begin{macrocode}
+\newcommand{\alias} [2]{\@ifdefinable #1{\let #1 #2}}
+\alias\realias\let
+\newcommand{\providealias}[2]{\@ifundefined #1{\let #1 #2}}
+% \end{macrocode}
+% Usage is, for example,
+% \begin{display}{l}
+% \cseq{alias}\cseq{foo}\cseq{bar}
+% \end{display}
+% which makes \cseq{foo} an alias for \cseq{bar}. As expected,
+% \cseq{alias} only works if the new name is currently undefined and
+% \cseq{providealias} does nothing if its first argument is already
+% defined. Somewhat more lax than its counterpart
+% \cseq{renewcommand}, \cseq{realias} does not care if its first
+% argument is defined or not. In essence, that command is used to
+% unconditionally alias a command. This is why, oddly enough,
+% \cseq{realias} is itself just an alias of \cseq{let}. Is this
+% getting confusing yet?
+%
+% Next we input wholesale a few useful packages. These are still in
+% the spirit of meta-macros which define this section. The first
+% package, \pkg{amstext}, provides the \cseq{text} command, which
+% basically puts its argument in text mode inside a box, but in the
+% current style (textstyle, scriptstyle, etc.). This is used later
+% on. The \pkg{xspace} package is used for control sequences which
+% would encounter `the space problem' when expanded as is.
+% \begin{macrocode}
+\RequirePackage{amstext}
+\RequirePackage{xspace}
+% \end{macrocode}
+% The command \cseq{intertext} from the \pkg{amsmath} package is quite
+% useful, but I do not want to include that entire package unless it
+% is necessary. Therefore, I make sure that command is defined one
+% way or another. I also give it the alias \cseq{rem} since I am
+% somewhat nostalgic about ancient forms of \textsf{\small BASIC}\dots{}
+% \begin{macrocode}
+\providecommand{\intertext}[1]{\noalign{%
+ \penalty\postdisplaypenalty\addvspace{ 0.5\belowdisplayskip}
+ \vbox{\normalbaselines\noindent#1}%
+ \penalty\predisplaypenalty\addvspace{0.5\abovedisplayskip}}}
+\alias\rem\intertext
+% \end{macrocode}
+%
+% Next we define some font style names which will be used in some
+% contexts later. This is done to avoid hard-coding of certain styles
+% and to allow as much customization as possible.
+% \begin{macrocode}
+\providecommand{\pagenofont} {\normalfont}
+\providecommand{\declarefont} {\normalfont\bfseries\mathversion{bold}}
+\providecommand{\altdeclarefont}{\normalfont\itshape}
+\providecommand{\captionfont} {\normalfont\itshape}
+\providecommand{\examplefont} {\normalfont}
+\providecommand{\altexamplefont}{\normalfont\itshape}
+\providecommand{\labelfont} {\normalfont\bfseries\mathversion{bold}}
+\providecommand{\timelinefont} {\normalfont}
+\providecommand{\titlefont} {\normalfont\bfseries\Large\mathversion{bold}}
+\providecommand{\verbatimfont} {\normalfont\ttfamily}
+% \end{macrocode}
+%
+% The next few commands are for programming convenience. First we
+% want to be able to swap the definitions of two control sequences.
+% \begin{macrocode}
+\newcommand{\swapdef}[2]{{%
+ \let \@tempa #1\relax
+ \global\let #1 #2\relax
+ \global\let #2 \@tempa}}
+% \end{macrocode}
+% We also want to be able to do the same for lengths (or glue or
+% whathaveyou).
+% \begin{macrocode}
+\newcommand{\swapdim}[2]{{%
+ \@tempdima #1\relax
+ \global #1 #2\relax
+ \global #2 \@tempdima}}
+% \end{macrocode}
+% Next is a macro constructed from an exercise in \emph{The
+% \TeX{}book}, which takes three control sequences and expands them in
+% reverse order.
+% \begin{macrocode}
+\newcommand{\expandthree}[2]{%
+ \expandafter\expandafter\expandafter #1\expandafter #2}
+% \end{macrocode}
+%
+% This next macro is modified from code I received in the
+% |comp.text.tex| newsgroup. According to the e-mail in which I
+% received it, the original source is a set of macros for
+% \emph{TUGboat}. It turns a number into an ordinal, finding the
+% correct ordinal label which is set as a superscript.
+% \begin{macrocode}
+\newcommand{\nth}[1]{{%
+ \@tempcnta = #1\relax
+ \ifnum \@tempcnta < 0\relax % Make sure our number is
+ \@tempcnta = -\@tempcnta % non-negative.
+ \fi
+ \ifnum \@tempcnta < 14\relax % Deal first with the
+ \ifnum \@tempcnta > 10\relax % exceptions for
+ \def\@tempa{th} % 11, 12, and 13.
+ \fi
+ \else
+ \loop \ifnum\@tempcnta > 9\relax % Loop until the recursive
+ \@tempcntb = \@tempcnta % remainder (mod 10) is
+ \divide \@tempcntb by 10\relax % a single digit in order
+ \multiply\@tempcntb by 10\relax % to successfully satisfy
+ \advance \@tempcnta by -\@tempcntb% the ordinality test.
+ \repeat
+ \def\@tempa{\ifcase\@tempcnta % Figure the proper label:
+ th% % 0th
+ \or st% % 1st
+ \or nd% % 2nd
+ \or rd% % 3rd
+ \else th% % nth
+ \fi}
+ \fi
+ #1\ensuremath{^{\text{\@tempa}}}}} % Superscript the label in
+ % math mode.
+% \end{macrocode}
+%
+% Continuing in the vein of superscripts, we define two macros which
+% put their arguments as sub- and superscripts in script-script
+% style. This was motivated by such things as derivative indices
+% which look just plain ugly in script style.
+% \begin{macrocode}
+\alias\sst\scriptscriptstyle
+\newcommand{\ssp}[1]{^{\sst#1}}
+\newcommand{\ssb}[1]{_{\sst#1}}
+% \end{macrocode}
+%
+% We now come to some very important and necessary macros, namely the
+% creation of typeset sideways \textsf{\small ASCII} smiley
+% faces. \smiley Since I like to be as general as possible, I have
+% also written an \cseq{emote} macro for indicating
+% emotions. \emote{smirk}
+% \begin{macrocode}
+\newcommand{\smiley}[1][\@smiley]{%
+ \edef\@sf{\spacefactor=\the\spacefactor}%
+ \unskip\spacefactor=1000\relax\space #1\@sf\xspace}
+\newcommand{\@smiley}{%
+ {\ttfamily\raise 0.078em\hbox{:}\kern-0.1em{-}\kern-0.1em{)}}}
+\newcommand{\emote}[1]{%
+ \smiley[\ensuremath{\langle}\emph{#1}\ensuremath{\rangle}]}
+% \end{macrocode}
+%
+% Since I learned the good habit of doing so at Washington and Lee, I
+% often append pledges to my assignments. The generic pledge is
+% implemented as an environment. It formerly took an argument, the
+% date, but I decided that was superfluous, seeing as how the
+% assignment headers set the date once. Why risk inconsistency? Much
+% to my surprise, I found out that \LaTeXe's \cseq{maketitle} command
+% unsets not only the date holder, but also the command which is used
+% to set the date in the first place. Anyhow, this means that the
+% date \emph{does} need to be set, but I have left that to be done by
+% the headers. The \env{pledge} environment issues a warning if the
+% date is not set.
+% \begin{macrocode}
+\newenvironment{pledge}%
+ {\ifx\@empty\@date
+ \PackageWarning{cjwmacro}{Date is not set.}
+ \fi
+ \parskip=2pt \parindent=0pt\relax
+ \null\vfill\begin{flushright}
+ \itshape\small}
+ {\\[5ex]\normalfont\footnotesize
+ \makebox[2in]{\hrulefill}\quad\@date\\
+ \makebox[2in]{Colin J.~Wynne}\quad{\hphantom{\@date}}\\
+ \end{flushright}}
+% \end{macrocode}
+% The old Washington and Lee pledge lives on in my macros\dots{} It
+% requires one argument, namely the type of assignment being pledged.
+% The argument is optional, though, and a paper is assumed by default.
+% \begin{macrocode}
+\newcommand{\wnlpledge}[1][paper]{%
+ \ifx\@empty\@date
+ \PackageWarning{cjwmacro}{Date is not set.}
+ \fi
+ \parskip=2pt \parindent=0pt\relax
+ \null\vfill\begin{flushright}
+ \itshape\small
+ On my honour, I have neither given nor received\\
+ any unacknowledged aid on this #1.\\[5ex]
+ \normalfont\footnotesize
+ \makebox[2in]{\hrulefill}\quad\@date\\
+ \makebox[2in]{Colin J.~Wynne,~'94}\quad{\hphantom{\@date}}\\
+ \end{flushright}}
+% \end{macrocode}
+%
+% As mentioned in the option section, there is a macro used to put
+% fancy section delimiters into, say, a story. The \cseq{ssbreak}
+% command expects the type of delimiter, the \cseq{ssbreakbar}, to be
+% defined. Since either \pkg{draft} or \pkg{final} must be chosen as
+% an option, this should be fine, but I have put a hopefully redundant
+% command in just in case.
+% \begin{macrocode}
+\newcommand{\ssbreak}{\bigskip
+ \centerline{\ssbreakbar}\bigbreak}
+\providecommand{\ssbreakbar}{}
+% \end{macrocode}
+%
+% \subsection{Box formatting}
+%
+% I have written some of my own commands for handling boxes. The
+% first thing I wanted was an analog of \cseq{mbox} or \cseq{hbox} for
+% math mode. The simple version---\cseq{mathbox} puts its argument into an
+% \cseq{hbox}, in math mode, in the current style. The second version
+% is \cseq{Mathbox}, which takes two arguments, the first of which is
+% put in the box and evaluated \emph{before} math mode is entered.
+% This was done for a specific application where I needed to get the
+% contents of the \cseq{mathbox} itself into boldface. Of course,
+% \cseq{boldmath} cannot be evaluated within math mode. Note that the
+% style is chosen by the \cseq{mathpalette} macro, and that the
+% command \cseq{@mathbox} is essentially just a dummy to allow the
+% proper expansion of \cseq{mathpalette}.
+% \begin{macrocode}
+% \mathbox puts its argument into an \hbox, in math mode, with the
+% current \...style.
+\def\mathbox #1{\hbox{$\mathpalette\@mathbox{#1}$}}
+\def\Mathbox #1#2{\hbox{#1$\mathpalette\@mathbox{#2}$}}
+\def\@mathbox#1#2{#1#2}
+% \end{macrocode}
+% Now, there is a reason why these are defined with \cseq{def} and not
+% \cseq{newcommand}. You see, what I really wanted to do was
+% something like
+% \begin{verbatim}
+% \newcommand{\mathbox}[2][]{%
+% \hbox{#1$\mathpalette\@mathbox{#1}$}}
+% \newcommand{\@mathbox}[2]{#1#2}
+% \end{verbatim}
+% in order to get optional arguments to my \cseq{mathbox}es. The
+% problem, though, is that I want to use this command in the context
+% of |\box|$N$|=\mathbox{|\dots{}|}|, and for that to work, the first
+% token in the expansion of \cseq{mathbox} \emph{must} be a
+% |\|$?$|box| command. The overhead imposed by \cseq{newcommand}
+% precludes this. So, I use the cheap hack until I figure out a more
+% workable way of implementing what I really want.
+%
+% A more generically applicable box command is one which does unto
+% width what \cseq{smash} does to height. Hence \cseq{smush}:
+% \begin{macrocode}
+\newcommand{\smush}{\relax
+ \ifmmode
+ \def\next{\mathpalette\math@smush}
+ \else
+ \let\next\make@smush
+ \fi \next}
+\newcommand{\make@smush}[1]{\setbox0=\hbox{#1}\fin@smush}
+\newcommand{\math@smush}[2]{\setbox0=\hbox{$\m@th#1{#2}$}\fin@smush}
+\newcommand{\fin@smush}{\wd0=0pt \box0 }
+% \end{macrocode}
+%
+% And finally, vaguely in the realm of boxes, we have struts. Here I
+% have defined some math struts of various sizes (corresponding to the
+% various delimiter sizes on which they are based).
+% \begin{macrocode}
+\newcommand{\bigmathstrut} {\vphantom{\big()}}
+\newcommand{\biggmathstrut}{\vphantom{\bigg()}}
+\newcommand{\Bigmathstrut} {\vphantom{\Big()}}
+\newcommand{\Biggmathstrut}{\vphantom{\Bigg()}}
+% \end{macrocode}
+%
+% \subsection{Abbreviations, etc.}
+%
+% I have found myself using particular types of abbreviations quite
+% often---often enough that I wanted control sequences for them,
+% whence these first few specimens.
+% \begin{macrocode}
+\newcommand{\ie} {\emph{i.e.}\xspace}
+\newcommand{\eg} {\emph{e.g.}\xspace}
+\newcommand{\heisst}{d.h\null.\xspace} % \dh is taken.
+% \end{macrocode}
+% Note the use of \cseq{xspace} so that explicit space need not be
+% given afterward. I did this mostly because I have never decided
+% whether or not I want to use a comma after either of this.
+%
+% The second type of abbreviation is the initial, or should I say
+% initials. I finally settled on a style I like---two initials should
+% be separated by a thinspace, and will of course need to have the
+% spacefactor adjusted if at the end of a sentence (followed by a
+% period, in particular). Here is the implementation:
+% \begin{macrocode}
+\newcommand{\initials}[2]{%
+ \break@init #2
+ \@ifdefinable #1{%
+ \global\edef#1{%
+ \noexpand\hbox{\@tempa.\noexpand\,\@tempb}%
+ \noexpand\@ifnextchar.{\noexpand\@}{.\noexpand\xspace}}}}
+\def\break@init #1.#2.{%
+ \def\@tempa{#1}\def\@tempb{#2}}
+% \end{macrocode}
+% What happens is this. The \cseq{initials} command is given a
+% control sequence name and the initials to be used. The initials are
+% broken on the periods and returned in the specified tokens. Then,
+% if the control sequence is available for definition, it is defined
+% in such a way to make all the spacing and punctuation work out.
+% Since the tokens need to be expanded back to the separate initials,
+% an \cseq{edef} is required---at the same time, use of
+% \cseq{noexpand} is made to keep things from going kablooie at
+% definition time. Here are some standard initials by way of usage
+% example.
+% \begin{macrocode}
+\initials{\UN}{U.N.}
+\initials{\US}{U.S.}
+\initials{\AI}{A.I.}
+% \end{macrocode}
+%
+% \subsection{Dates}
+%
+% I have had call to do a fair amount of \TeX{} in both English and
+% German. Therefore, in implementing the examples of date macros from
+% \emph{The \TeX{}book}, I have provided for both languages.
+%
+% \begin{macrocode}
+% LaTeX style commands for date-parts, both English and German.
+\providecommand{\theday}{\number\day\relax}
+\providecommand{\themonth}{%
+ \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}
+\providecommand{\themonat}{%
+ \ifcase\month\or Januar\or Februar\or%
+ M\"arz\or April\or Mai\or Juni\or Juli\or August\or%
+ September\or Oktober\or November\or Dezember\fi}
+\providecommand{\theyear}{\number\year\relax}
+% \end{macrocode}
+% Note that \cseq{today} is unconditionally defined by the following
+% underhandedness.
+% \begin{macrocode}
+\providecommand{\today}{}
+\renewcommand{\today}{\theday~\themonth, \theyear\xspace}
+\providecommand{\heute}{}
+\renewcommand{\heute}{den~\theday.\ \themonat\ \theyear\xspace}
+ \alias\gdate\heute
+% \end{macrocode}
+%
+% \subsection{Page styles and titles}
+%
+% We are finally into the realm of more traditional package macros,
+% namely creating some general page appearances. Here I have
+% redefined the \pkg{plain} pagestyle to take advantage of the
+% \cseq{pagenofont} defined above.
+% \begin{macrocode}
+\renewcommand{\ps@plain}{%
+ \let\@mkboth \@gobbletwo
+ \let\@oddhead \@empty
+ \let\@evenhead\@empty
+ \def\@oddfoot{\pagenofont\hfil\thepage\hfil}
+ \let\@evenfoot\@oddfoot}
+% \end{macrocode}
+% The \pkg{topright} pagestyle has page numbers (strangely enough) at the top
+% right of the page.
+% \begin{macrocode}
+\newcommand{\ps@topright}{%
+ \let\@mkboth \@gobbletwo
+ \def\@oddhead{\pagenofont\hfil\thepage}
+ \let\@evenhead\@oddhead
+ \let\@oddfoot \@empty
+ \let\@evenfoot\@empty}
+% \end{macrocode}
+%
+% \subsection{Text formatting}
+%
+% \subsubsection{Timelines}
+%
+% A timeline is a long, running, two-column format used, for example,
+% to do r\'esum\'es or vit\ae (or, if you are in Germany, a
+% \emph{Lebenslauf}). The idea is that a date (or some identifying
+% information) appears at the left, and the content is given in the
+% righthand column. The usage is
+% \begin{display}{l}
+% |\timeline[|\meta{pos}|]{|\meta{date}|}|
+% \end{display}
+% where \meta{pos} is exactly the argument to \cseq{makebox}, the
+% justification of the \meta{date} entry within the lefthand column.
+% That column has length \cseq{timelineskip}, which can off course be
+% set as desired.
+% \begin{macrocode}
+\newlength{\timelineskip}
+\setlength{\timelineskip}{1.75in}
+% \end{macrocode}
+% The actual entries are not considered to be two separate columns.
+% Rather, the first line is padded out to \cseq{timelineskip} with
+% makebox, and following lines use a hanging indentation. The control
+% sequence \cseq{endtimeline} is defined trivially so that a timeline
+% entry may be used as a \env{timeline} environment.
+% \begin{macrocode}
+\newcommand{\timeline}[2][l]{%
+ \noindent\hangindent=\timelineskip
+ \makebox[\timelineskip][#1]{\timelinefont{#2}}\ignorespaces}
+\let\endtimeline\relax
+% \end{macrocode}
+%
+% \subsubsection{Mathematical declarations}
+%
+% In writing up mathematics, one often wishes to declare definitions,
+% theorems, and so forth. I have written generic declaration macros
+% which can be customized for these uses. Since I prefer to have all
+% such things numbered seuqentially, they use a common counter, called
+% \cseq{declare}, strangely enough. They are numbered within sections
+% if sections are being numbered.
+% \begin{macrocode}
+\@ifundefined{c@section}
+ {\newcounter{declare}}
+ {\newcounter{declare}[section]
+ \renewcommand{\thedeclare}{\thesection.\arabic{declare}}}
+% \end{macrocode}
+% When declarations are numbered, it is sometime nice to have the
+% declaration type and number appear uniformly wide throughout. This
+% is done by forcing the declaration to appear in a box of width
+% \cseq{declareindent}.
+% \begin{macrocode}
+\newlength{\declareindent}
+ \setlength{\declareindent}{0pt}
+% \end{macrocode}
+% We use some internal commands to specify exactly how the declaration
+% is typeset. In fact, we will be defining not only declaration, but
+% an alternate declaration form so that two different methods may be
+% used simultaneously in a document---\eg, when theorems and major
+% results are to be italicized, but definitions and so forth are not.
+% \begin{macrocode}
+\newcommand{\@declare} [1]{{\declarefont#1:}\quad}
+\newcommand{\@altdeclare}[1]{{\altdeclarefont#1:}\quad}
+% \end{macrocode}
+% The generic declarations are environments, and are provided in both
+% numbered (normal) and unnumbered (starred) forms. The latter are
+% more simple.
+% \begin{macrocode}
+\newenvironment{declaration*}[1]%
+ {\medbreak\noindent\ignorespaces
+ \@declare{#1}\ignorespaces}%
+ {\kern0pt\nobreak\smallskip}
+\newenvironment{altdeclaration*}[1]%
+ {\medbreak\noindent\ignorespaces
+ \@altdeclare{#1}\ignorespaces}%
+ {\kern0pt\nobreak\smallskip}
+% \end{macrocode}
+% The numbered versions introduce nothing surprising, but are a tad
+% more involved.
+% \begin{macrocode}
+\newenvironment{declaration}[1]%
+ {\medbreak\refstepcounter{declare}
+ \noindent\ignorespaces
+ \ifnum\declareindent = 0\relax%
+ \@declare{\thedeclare\quad #1}
+ \else
+ \makebox[\declareindent]{\@declare{\thedeclare\hss #1}}
+ \fi\ignorespaces}
+ {\kern0pt\nobreak\smallskip}
+\newenvironment{altdeclaration}[1]%
+ {\medbreak\noindent\ignorespaces
+ \refstepcounter{declare}
+ \ifnum\declareindent = 0\relax
+ \@altdeclare{\thedeclare\quad #1}
+ \else
+ \makebox[\declareindent]{\@altdeclare{\thedeclare\hss #1}}
+ \fi\ignorespaces}
+ {\kern0pt\nobreak\smallskip}
+% \end{macrocode}
+% Now, because I am essentially lazy and do not want the extra typing
+% needed for an environment, I have shortcuts, \cseq{declare} and
+% \cseq{altdeclare}, as well as numbered versions \cseq{ndeclare} and
+% \cseq{altndeclare}. The first argument is passed to the
+% corresponding environment, and the following paragraph is the body
+% of the environment.
+% \begin{macrocode}
+\def\declare #1#2\par{%
+ \begin{declaration*}{#1}#2\end{declaration*}\par}
+\def\altdeclare #1#2\par{%
+ \begin{altdeclaration*}{#1}#2\end{altdeclaration*}\par}
+\def\ndeclare #1#2\par{%
+ \begin{declaration}{#1}#2\end{declaration}\par}
+\def\altndeclare#1#2\par{%
+ \begin{altdeclaration}{#1}#2\end{altdeclaration}\par}
+% \end{macrocode}
+% Genreality is all well and good, but there are some stock
+% declarations, given here in both numbered and unnumbered versions.
+% \begin{macrocode}
+\providecommand{\corollary} {\declare{Corollary}}
+\providecommand{\definition} {\declare{Definition}}
+\providecommand{\lemma} {\declare{Lemma}}
+\providecommand{\proposition} {\declare{Proposition}}
+\providecommand{\theorem} {\declare{Theorem}}
+\providecommand{\note} {\altdeclare{Note}}
+
+\providecommand{\ncorollary} {\ndeclare{Corollary}}
+\providecommand{\ndefinition} {\ndeclare{Definition}}
+\providecommand{\nlemma} {\ndeclare{Lemma}}
+\providecommand{\nproposition}{\ndeclare{Proposition}}
+\providecommand{\ntheorem} {\ndeclare{Theorem}}
+\providecommand{\nnote} {\altndeclare{Note}}
+% \end{macrocode}
+% In addition, the following German declaration (meaning a claim) is
+% also defined.
+% \begin{macrocode}
+\providecommand{\behaupt} {\declare{Behauptung}}
+\providecommand{\nbehaupt} {\ndeclare{Behauptung}}
+% \end{macrocode}
+% Finally, since one is not likely to mix numbered and unnumbered,
+% here is a control sequence that will make sure everything is
+% numbered.
+% \begin{macrocode}
+\newcommand{\allndeclares}{%
+ \let\declare \ndeclare
+ \let\altdeclare \altndeclare}
+% \end{macrocode}
+%
+% Now that we have propositions, claims, and theorems (oh my!), we
+% want to be able to prove them. The first step is to define a proof
+% environment. The environment simply sets up a label and some
+% spacing. The label is in |\altdeclarefont|. The label is an
+% optional argument which defaults to `Proof', oddly enough.
+% \begin{macrocode}
+\newenvironment{proof}[1][Proof]%
+ {\smallbreak\noindent{\altdeclarefont#1:}%
+ \quad\ignorespaces}%
+ {\qed}
+% \end{macrocode}
+% Just in case you are prooving in German, we also have the following:
+% \begin{macrocode}
+\newenvironment{beweis}[1][Beweis]%
+ {\smallbreak\noindent{\altdeclarefont#1:}%
+ \quad\ignorespaces}%
+ {\qed}
+% \end{macrocode}
+% Note that the end of a proof has the command |\qed|, which we will
+% unconditionally define here. This is adapted from \emph{The
+% \TeX{}book}. The idea is to right justify |\qedsymbol| on the line
+% where |\qed| is invoked, unless there is not a comfortable amount of
+% room. That amount is given as 2\,em. When this happens, the line
+% is broken and the |\qedsymbol| appears flush right on the following
+% line.
+% \begin{macrocode}
+\providecommand{\qed}{}
+ \renewcommand{\qed}{%
+ {\unskip\nobreak\hfil\penalty 50%
+ \hskip 2em\hbox{}\nobreak\hfil\qedsymbol%
+ \parfillskip=0pt \finalhyphendemerits=0 \par}}
+% \end{macrocode}
+% The standard end-of-proof symbol is a box, but I prefer somewhat
+% less ink. I use \TeX's hollow diamond suit symbol (again
+% unconditionally defined).
+% \begin{macrocode}
+\providecommand{\qedsymbol}{}
+ \renewcommand{\qedsymbol}{\lower 0.35ex\hbox{$\diamondsuit$}}
+% \end{macrocode}
+% In case it is desired, the box symbol is defined as |\qedbox|, and
+% then with a simple alias this can be used to end all proofs.
+% \begin{macrocode}
+ \newcommand{\qedbox}{\vrule height4pt width3pt depth2pt}
+% \end{macrocode}
+% Now we wish to define control sequences for some constructions
+% commonly found in proofs. I often find myself writing out proofs
+% which require cases. There are two types of cases. Often I will
+% want to have two cases, one for each definition of an equivalence
+% for example. In this case, the case delimiters will be something
+% like |\then| and |\when| commands, and should be set in parentheses
+% to mark them clearly. The other type is the more general `Case
+% $n$:' (where $n$, one hopes, will not be \emph{too} large). To
+% cover both of these, we use a fairly typical \LaTeX{} conceit:
+% \begin{display}{l}
+% |\Case*{|\meta{case}|}|
+% \end{display}
+% where the unstarred argument sets \meta{case} inside parentheses and
+% the star supresses the parentheses. A German alias is given.
+% \begin{macrocode}
+\newcommand{\Case}{\@ifstar{\@starCase}{\@Case}}
+\newcommand{\@starCase}[1]{\@@Case{#1}}
+\newcommand{\@Case}[1]{\@@Case{(#1)}}
+\newcommand{\@@Case}[1]{%
+ \noindent{\declarefont#1}\quad\ignorespaces}
+\alias\Fall\Case
+% \end{macrocode}
+% And finally, just in case the proof is by contradiction, we have the
+% following.
+% \begin{macrocode}
+\newcommand{\contra}{\ensuremath{\Rightarrow\Leftarrow}}
+% \end{macrocode}
+%
+%
+% \subsubsection{Problems and examples}
+%
+% In addition to declarations, and for subjects other than
+% mathematics, one might want to provide examples and worked
+% problems. I implement examples as a separate environment (well,
+% \emph{four} separate environments) so that they may have fonts
+% distinct from declarations.
+% \begin{macrocode}
+\newenvironment{example*}%
+ {\@nameuse{declaration*}{Example}\examplefont}
+ {\medbreak}
+\newenvironment{altexample*}%
+ {\@nameuse{declaration*}{Example}\examplefont}
+ {\medbreak}
+\newenvironment{example}%
+ {\declaration{Example}\examplefont}
+ {\medbreak}
+\newenvironment{altexample}%
+ {\declaration{Example}\examplefont}
+ {\medbreak}
+% \end{macrocode}
+%
+% Problems are handled differently. In my experience, problems do not
+% appear in longwinded documents using sectioning, and so the counter
+% need not be embedded.
+% \begin{macrocode}
+\newcounter{problem}
+ \setcounter{problem}{0}
+\renewcommand{\theproblem}{\arabic{problem}}
+\renewcommand{\p@problem}{}
+% \end{macrocode}
+% Often when working a problem, one wishes to include a reference to
+% the source. This is accomplished via the \cseq{Page} macro. Usage
+% is
+% \begin{display}{l}
+% |\Page*[|\meta{author}|]{|\meta{pp}|}{|\meta{problem}|}|
+% \end{display}
+% In the \LaTeX{} tradition, the standard command tries to outguess
+% you by prepending a hash `\#' to the problem number, whereas the
+% starred version omits this. Thus, the standard version produces
+% \begin{display}{l}
+% ([\meta{author},~]p.\,\meta{pp}, \#\meta{problem})
+% \end{display}
+% and \cseq{Page*} leaves off the hash. Notice that the space between
+% author and page number is inserted by the macro. The macro is
+% defined robustly so that it can be used in moving arguments.
+% \begin{macrocode}
+\DeclareRobustCommand{\Page}{%
+ \@ifstar{\@Page{}}{\@Page{\#}}}
+\def\@Page#1{%
+ \@ifnextchar [{\@@Page{#1}}{\@@Page{#1}[]}}
+\def\@@Page#1[#2]#3#4{%
+ \def\@tempa{#2}%
+ \ifx\@empty\@tempa%
+ \let\@tempb\@tempa%
+ \else%
+ \edef\@tempb{\@tempa,~}%
+ \fi%
+ (\@tempb p.\,#3, #1{#4})}
+% \end{macrocode}
+%
+% The statement of a problem is given in an environment, again so that
+% font customization can be easily done. The \env{statement}
+% environment takes a single optional argument, which is typeset at
+% the beginning of the statement in \cseq{altdeclarefont}. The rest
+% of the statement is in \cseq{declarefont}. I use the optional
+% argument to pass a \cseq{Page} reference most often---note, though,
+% that if the optional argument to \cseq{Page} is used, the whole
+% \cseq{Page} command should be put in braces sp that the square
+% brackets of optional arguments do not get confused.
+% \begin{macrocode}
+\newenvironment{statement}[1][\null]%
+ {\def\@tempa{#1}\def\@tempb{\null}%
+ \ifx\@tempa\@tempb%
+ \def\@tempc{\null}%
+ \else%
+ \def\@tempc{\altdeclarefont\@tempa\quad}%
+ \fi%
+ \declarefont{\@tempc}\ignorespaces}
+ {\removelastskip\nopagebreak\smallskip}
+% \end{macrocode}
+% A problem, now, is basically just a wrapper for the statement. It
+% generates and sets the problem number, does some spacing, and the
+% body of the \env{problem} environment becomes the body of the
+% \env{statement}. In essence, a \env{problem} is a numbered
+% \env{statement}. There is a starred version which does not do
+% numbers and references---this just does the correct spacing and
+% calls \env{statement}. Note that the problem number is set using a
+% macro from the \pkg{cjw-outl} package.
+% \begin{macrocode}
+\newenvironment{problem}%
+ {\setcounter{equation}{0}%
+ \gdef\theequation{\theproblem.\arabic{equation}}%
+ \removelastskip\medbreak%
+ \refstepcounter{problem}%
+ \noindent\theoutlabel{\theproblem.}%
+ \statement}
+ {\endstatement}
+\newenvironment{problem*}%
+ {\removelastskip\medbreak%
+ \noindent\statement}
+ {\endstatement}
+% \end{macrocode}
+% Since I have been known to write assignments in German, we provide
+% the aliases to make an \env{aufgabe} environment.
+% \begin{macrocode}
+\alias \aufgabe \problem
+\realias\endaufgabe\endproblem
+% \end{macrocode}
+% For parts and subparts of problems, the aliases are English and the
+% main definitions are German because I wrote these while I was in
+% Germany. So, parts (\emph{Teile}) are numbered within problems and
+% subparts (\emph{Subteile}) within parts. The defaults for
+% cross-referencing (the \cseq{p@} forms) are set, too.
+% \begin{macrocode}
+\newcounter{teil} [problem]
+\newcounter{steil}[teil]
+\renewcommand{\theteil} {(\alph{teil})}
+ \renewcommand{\p@teil} {\theproblem}
+\renewcommand{\thesteil} {(\roman{steil})}
+ \renewcommand{\p@steil}{\p@teil\theteil}
+% \end{macrocode}
+% The environment \env{part}/\env{teil} used to be implemented
+% entirely in terms of the \pkg{cjw-outl} package, but that was a
+% little bit of overkill, and made the numbering more difficult to
+% implement. I ought to one day implement problems/parts/subparts
+% entirely in terms of the outline macros. We'll see.
+%
+% Anyway, the current implementations still rely on some of the
+% definitions from the \pkg{cjw-outl} package. The
+% \env{part}/\env{teil} environments take a single optional argument,
+% namely the outline depth. Level one starts the text flush left at
+% the margin, which is usually where the \env{problem} is, hence a
+% \env{part} should be at level two, and this is the default. A more
+% detailed explanation of the goings-on here can be found in
+% \pkg{cjw-outl.dtx}.
+% \begin{macrocode}
+\newenvironment{teil}[1][2]%
+ {\@tempcnta=#1\advance\@tempcnta by -1\relax
+ \ifnum\@tempcnta < 1\relax
+ \leftskip=0pt\relax
+ \else
+ \leftskip=\@tempcnta\outlindent
+ \fi
+ \refstepcounter{teil}
+ \addvspace{\medskipamount}%
+ \noindent\theoutlabel{\theteil}%
+ \ignorespaces}
+ {\par\smallbreak}
+% \end{macrocode}
+% The default level for \env{ppart} or \env{steil}, the subpart
+% environments, is three.
+% \begin{macrocode}
+\newenvironment{steil}[1][3]%
+ {\@tempcnta=#1\advance\@tempcnta by -1\relax
+ \ifnum\@tempcnta < 1\relax
+ \leftskip=0pt\relax
+ \else
+ \leftskip=\@tempcnta\outlindent
+ \fi
+ \refstepcounter{steil}
+ \addvspace{\medskipamount}%
+ \noindent\theoutlabel{\thesteil}%
+ \ignorespaces}
+ {\par\smallbreak}
+% \end{macrocode}
+% English aliases are, of course, given.
+% \begin{macrocode}
+\realias\part \teil
+\realias\endpart \endteil
+\alias \ppart \steil
+\realias\endppart\endsteil
+% \end{macrocode}
+%
+% \subsubsection{Footnotes}
+%
+% This is a simple modification to a standard \LaTeXe{} internal
+% macro, because I prefer hanging indentation on my footnote text.
+% \begin{macrocode}
+\long\def\@makefntext#1{%
+ \parindent 1em\noindent\hangindent=\parindent%
+ \hb@xt@ 1em{\hss \llap{\@makefnmark} }#1}
+% \end{macrocode}
+%
+% \subsubsection{Text displays}
+%
+% I have turned the \cseq{begindisplay} and \cseq{enddisplay} macro
+% pair from \emph{The \TeX{}book} (page~421) into a \LaTeX{}
+% environment. As with Knuth's macros, local definitions for use
+% within the display can be given, in this case via the
+% \env{display}'s optional argument. Since the environment is
+% implemented through the standard \env{tabular} environment, there is
+% a mandatory argument specifying column layout. Overall, usage is
+% \begin{display}{l}
+% \cseq{begin}|{display}[|\meta{local}|]{|\meta{cols}|}|
+% \end{display}
+% with local definitions \meta{local} and column descriptol
+% \meta{col}. (By the way, the previous display was created with the
+% \env{display} environment\dots)
+%
+% The display's offset from the left margin is specified by
+% \cseq{textdisplay indent}. Default value is equal to
+% \cseq{parindent}, and is therefore set below, after
+% \cseq{parindent}.
+% \begin{macrocode}
+\newlength{\textdisplayindent}
+% \end{macrocode}
+% The actual \env{display} environment uses the spacing and penalties
+% of mathematical displays.
+% \begin{macrocode}
+\newenvironment{display}[2][]
+ {\vadjust{\penalty\predisplaypenalty}
+ \@newline[\abovedisplayskip]%
+ \begingroup%
+ #1%
+ \begin{tabular}{@{\null\hspace{\textdisplayindent}\null}#2}}
+ {\end{tabular}\endgroup
+ \vadjust{\penalty\postdisplaypenalty}
+ \@newline[\belowdisplayskip]\ignorespaces}
+% \end{macrocode}
+%
+% \subsection{Verbatim inclusions}
+%
+% Since I only occassionally need to include verbatim files, the
+% following macros need to be specifically included by a package
+% option, as was mentioned above.
+% \begin{macrocode}
+\if@verbext
+% \end{macrocode}
+% For numbered inclusions, we need a line number counter.
+% \begin{macrocode}
+\newcounter{vfline}
+\renewcommand{\thevfline}{\arabic{vfline}}
+% \end{macrocode}
+% We need the following command from \emph{The \TeX{}book}.
+% \begin{macrocode}
+\providecommand{\uncatcodespecials}{%
+ \def\do##1{\catcode`##1=12 }\dospecials}
+% \end{macrocode}
+% The basic command is \cseq{verbfile} which takes an optional
+% argument specifying the starting line number (default is one), and a
+% mandatory argument which is, of course, the name of the file to
+% include. There is also \cseq{verbfilenolines} which does not number
+% lines, and needs only the one mandatory argument.
+% \begin{macrocode}
+\providecommand{\verbfile}[2][1]{%
+ \par\begingroup\@vf@lines{#1}\input{#2}\relax\endgroup}
+\providecommand{\verbfilenolines}[1]{%
+ \par\begingroup\@vf@nolines\input{#1}\relax\endgroup}
+% \end{macrocode}
+% In the manner of command which need to do \cseq{catcode} trickery,
+% the above are primarily wrappers for the real commands. The one
+% problem with these as currently implemented is that they do not
+% handle leading space in the included file. Oh, well.
+% \begin{macrocode}
+\newcommand{\@vf@lines}[1]{%
+ \verbatimfont
+ \setcounter{vfline}{#1}
+ \addtocounter{vfline}{-1}
+ \setlength{\parindent}{0pt}
+ \setlength{\parskip}{0pt}
+ \def\par{\leavevmode\endgraf}
+ \obeylines \uncatcodespecials \obeyspaces
+ \everypar{\null\stepcounter{vfline}%
+ \llap{\scriptsize\thevfline\quad}\null}}
+\newcommand{\@vf@nolines}{%
+ \verbatimfont
+ \setlength{\parindent}{0pt}
+ \setlength{\parskip}{0pt}
+ \def\par{\leavevmode\endgraf}
+ \obeylines \uncatcodespecials \obeyspaces
+ \everypar{\null}}
+% \end{macrocode}
+% Now we end the inclusion conditional.
+% \begin{macrocode}
+\fi
+% \end{macrocode}
+%
+% \subsection{Initialization}
+%
+% We use the \cseq{AtBeginDocument} command to set up some default
+% values when the document is actually started.
+% \begin{macrocode}
+\AtBeginDocument{%
+ \setlength{\parindent} {20pt}
+ \setlength{\parskip} { 2pt plus 1pt}
+ \setlength{\textdisplayindent}{\parindent}}
+% \end{macrocode}
+%
+% \iffalse
+%
+%</general>
+%
+% \fi
+%
+%
+% \section{Math macros}
+%
+% \iffalse
+%
+%<*math>
+%
+% \fi
+%
+% While some macros useful for typesetting mathematics have already
+% been covered, none of them had to do with mathematical equations or
+% symbols---they were macros for logical flow and delineation. In
+% this package, \pkg{cjwmath}, I have written macros which are
+% specifically for typesetting the actual math---the vast majority of
+% these macros, if not actually all of them, are meant to be used in
+% math mode.
+%
+% \subsection{Package initialization}
+%
+% Since different papers require different types of math, I have again
+% used the introduction of conditionals and package options to control
+% what code is actually loaded. The important one concerns use of the
+% AMS math packages in AMS-\LaTeX. This is included as an option to
+% my package because some of my definitions depend upon whether the
+% AMS macros are being used. There are conditionals for including
+% code for calculus (both derivatives and integrals) and some code for
+% physics.
+% \begin{macrocode}
+\newif \if@amsmath
+\newif \if@derivatives
+\newif \if@integrals
+\newif \if@physics
+% \end{macrocode}
+% There are options corresponding to each conditional.
+% \begin{macrocode}
+\DeclareOption{amsmath} {\@amsmathtrue}
+\DeclareOption{derivs} {\@derivativestrue}
+\DeclareOption{integrals}{\@integralstrue}
+\DeclareOption{physics} {\@physicstrue}
+% \end{macrocode}
+% There used to be another option for typesetting units. While I
+% originally included that code in this package directly, I found
+% several occasions where I wanted units but not the rest of the math
+% code. Therefore, units are in a separate package, and the option
+% now just reminds the user to input that package by itself.
+% \begin{macrocode}
+\DeclareOption{units}{%
+ \PackageWarning{cjwmath}%
+ {Obsolete option \CurrentOption. Use package `cjwunits' instead.}}
+% \end{macrocode}
+% Finally, there is a default option, to warn about unknown options,
+% and the passed option list is processed.
+% \begin{macrocode}
+\DeclareOption*{%
+ \PackageWarning{cjwmath}{Unknown option `\CurrentOption'}}
+\ProcessOptions
+% \end{macrocode}
+%
+% This package depends upon the previous one.
+% \begin{macrocode}
+\RequirePackage{cjwmacro}
+% \end{macrocode}
+% It also uses the AMS fonts, for which we require \pkg{amssymb},
+% which itself requires \pkg{amsfonts}.
+% \begin{macrocode}
+\RequirePackage{amssymb}
+% \end{macrocode}
+% Just in case things get screwy in \pkg{cjwmacro}---which they
+% shouldn't, we explicitly require \pkg{amstext} here, too, for the
+% \cseq{text} command.
+% \begin{macrocode}
+\RequirePackage{amstext}
+% \end{macrocode}
+% I much prefer the following package to AMS's own blackboard bold
+% font.
+% \begin{macrocode}
+\RequirePackage{bbm}
+% \end{macrocode}
+% If the \pkg{amsmath} option is specified, we load the package (which
+% itself brings in a lot of other stuff).
+% \begin{macrocode}
+\if@amsmath
+ \RequirePackage{amsmath}
+\fi
+% \end{macrocode}
+%
+% \subsection{Miscellaneous macros}
+%
+% Here is a package command which I have written to cover the
+% shortcomings of AMS-\LaTeX's |\DeclareNewMathOperator| command. In
+% particular, I would like to be able to set different fonts for some
+% operators. The syntax is
+% \begin{display}{l}
+% |\NewMathOp*[|\meta{font}|]{\cs}{|\meta{text}|}|
+% \end{display}
+% The optional star makes an operator with limits. The \meta{font}
+% is, by default, |\operator@font|. |\cs| is the name of the new
+% mathop. \meta{text} should be the printed version of the operator,
+% but may also include, for example, extra kerning information, as in
+% \begin{display}{l}
+% |\NewMathOp[\mathfrak]{\so}{o\kern 0pt}|
+% \end{display}
+% The command should produce something robust.
+% \begin{macrocode}
+\DeclareRobustCommand{\NewMathOp}{%
+ \@ifstar{\@makenewop{\displaylimits}}
+ {\@makenewop{\nolimits}}}
+% \end{macrocode}
+% The first iteration applies a font if the optional argument is not
+% given.
+% \begin{macrocode}
+\def\@makenewop#1{%
+ \@ifnextchar [{\@@makenewop{#1}}
+ {\@@makenewop{#1}[\operator@font]}}
+% \end{macrocode}
+% Finally, the net operator itself is declared robustly. The
+% arguments are, in order, either |\displaylimits| or |\nolimits|, the
+% font, the control sequence, and the operator text.
+% \begin{macrocode}
+\def\@@makenewop#1[#2]#3#4{%
+ \DeclareRobustCommand{#3}{%
+ \mathop{\kern\z@{#2{#4}}}#1}}
+% \end{macrocode}
+%
+% The next few macros have to do with things not specific to any
+% particular flavor of mathematics. For example, I like some of the
+% alternate Greek characters more than the originals---notice how we
+% cleverly required the \pkg{cjwmacro} package which gives us the
+% \cseq{swapdef} command.
+% \begin{macrocode}
+\swapdef{\epsilon}{\varepsilon}
+% \swapdef{\theta}{\vartheta}
+\swapdef{\rho}{\varrho}
+% \end{macrocode}
+% I also like the empty set symbol from AMS, to which I also assign a
+% German alias.
+% \begin{macrocode}
+\swapdef{\nothing}{\varnothing}
+\alias\leer\nothing
+% \end{macrocode}
+% The standard symbols `$\exists$' and `$\forall$' do not have
+% satisfactory spacing, in my opinion, so I redefine them as
+% relations. Notice the aliasing so that the symbols' redefinition
+% can be carried out regardless of current math fonts.
+% \begin{macrocode}
+\alias\@@exists\exists
+ \renewcommand{\exists}{\mathrel{\@@exists}}
+\alias\@@forall\forall
+ \renewcommand{\forall}{\mathrel{\@@forall}}
+% \end{macrocode}
+% What \LaTeX{} cleverly calls \cseq{ni} (a backwards `$\in$') really
+% ought to mean `such that,' hence I rename it:
+% \begin{macrocode}
+\newcommand{\st}{\mathrel{\ni}}
+% \end{macrocode}
+% Being essentially lazy, I also prefer to make a nice control
+% sequence for some standard abbreviations. One happens to be
+% German (since in German it is more acceptable to use abbreviations
+% of long phrases even in a more formal setting).
+% \begin{macrocode}
+\newcommand{\WLOG}{Without loss of generality\xspace}
+\newcommand{\Wlog}{without loss of generality\xspace}
+\newcommand{\obda}{o.B.d.A.\xspace}
+\newcommand{\fp}{floating-point\xspace}
+% \end{macrocode}
+% The following two commands are simply for phantoms I often find
+% myself using, for example to make alignments in arrays and matrices
+% come out right. The mnemonic is `phantom negative' or `phantom equals'.
+% \begin{macrocode}
+\newcommand{\pneg}{\phantom{-}}
+\newcommand{\peq}{\phantom{=}}
+% \end{macrocode}
+% Going probably too far into the realm of generalization, here are
+% some macros to set their arguments inside matching scaled delimiters
+% of various sorts.
+% \begin{macrocode}
+\newcommand{\anglebrackets}[1]{%
+ \left\langle #1 \right\rangle}
+\newcommand{\curlybrackets}[1]{%
+ \left\{ #1 \right\}}
+\newcommand{\squarebrackets}[1]{%
+ \left[ #1 \right]}
+\newcommand{\vertbrackets}[1]{%
+ \left| #1 \right|}
+\newcommand{\Vertbrackets}[1]{%
+ \left\| #1 \right\|}
+% \end{macrocode}
+% And now for something completely different---sometimes an operand
+% should be left generic, but not in terms of a variable. The usual
+% way of accomplishing this is to place a small dot where the argument
+% would otherwise go. As I consider this to imply `no argument', the
+% command is |\noarg|.
+% \begin{macrocode}
+\newcommand{\noarg}{\,\cdot\,}
+% \end{macrocode}
+% We end with a few things that should be fairly obvious.
+% \begin{macrocode}
+\newcommand{\ee}[1]{\times10^{#1}}
+\newcommand{\half}{\sfrac12}
+\newcommand{\ninfty}{-\infty}
+% \end{macrocode}
+% This is shorthand for function definitions, including an extra
+% control sequence for some backwards compatibility and an alias to
+% German.
+% \begin{macrocode}
+\newcommand{\fcn}[2]{\colon{#1}\rightarrow{#2}}
+ \newcommand{\mapping}[3]{{#1}\fkt{#2}{#3}}
+\alias\fkt\fcn
+% \end{macrocode}
+% Restrictions of functions:
+% \begin{macrocode}
+\newcommand{\restr}[2][\big]{\kern -.1em #1|_{#2}}
+% \end{macrocode}
+%
+%
+% \subsection{Combinatorics}
+%
+% The binomial coefficient $\choose{n}{k}$ is defined, depending on
+% whether or not AMS-\LaTeX{} is being used.
+% \begin{macrocode}
+\if@amsmath
+ \realias\choose\binom
+\else
+ \renewcommand{\choose}[2]{{{#1}\atopwithdelims(){#2}}}
+\fi
+% \end{macrocode}
+% And lastly, we have the combinatorial doohickie which is read as
+% `$n$ multichoose $k$', using doubled parentheses as delimiters. I
+% think this comes out looking right.
+% \begin{macrocode}
+\newcommand{\mchoose}[2]{%
+ \mathchoice%
+ {\left(\kern-0.48em\choose{#1}{#2}\kern-0.48em\right)}
+ {\left(\kern-0.30em
+ \choose{\smash{#1}}{\smash{#2}}\kern-0.30em\right)}
+ {\left(\kern-0.30em
+ \choose{\smash{#1}}{\smash{#2}}\kern-0.30em\right)}
+ {\left(\kern-0.30em
+ \choose{\smash{#1}}{\smash{#2}}\kern-0.30em\right)}
+ }
+% \end{macrocode}
+% There is also the old-fashioned $\Comb{n}{k}$-type notation, as used
+% both in English and in German.
+% \begin{macrocode}
+\newcommand{\Comb}[2]{% % C
+ {}_{#1}{\operator@font C}_{#2}} % #1 #2
+\newcommand{\Komb}[2]{% % #2
+ {\operator@font Ko}_{#1}^{#2}} % Ko
+\newcommand{\Kombun}[2]{\Komb{#1,\neq}{#2}} % #1
+\newcommand{\Perm}[2]{% % #2
+ {\operator@font Pe}_{#1}^{#2}} % Pe
+\newcommand{\Permun}[2]{\Perm{#1,\neq}{#2}} % #1
+% \end{macrocode}
+%
+% \subsection{Sets}
+%
+% The most important macro in this section is named, of course,
+% |\set|. The idea is to make sets which look like
+% \begin{displaymath}
+% \set{x \in \R^2}{\norm{x}_p = 1 \forall p = 1,2,3,\ldots}.
+% \end{displaymath}
+% That is, there should be scaled braces around two halves separated
+% by a scaled logical delimiter, the vertical bar. The problem with
+% this is getting everything the same height, since the $\mid$
+% specifier does not scale and there is no middle-counterpart to
+% |\left|\dots|\right|.
+%
+% So, the command has the form:
+% \begin{display}{l}
+% |\set[|\meta{mid}|]{|\meta{left}|}{|\meta{right}|}|
+% \end{display}
+% The optional \meta{mid} specifies an alternate delimiter to use
+% between the two definition halves of the set. If it is left
+% \emph{empty}, the null delimiter `.' will be assumed. If anything
+% at all appears in the optional argument, though, the first token
+% \emph{must} be a delimiter, as it will immediately be preceded by a
+% sizing macro. For example, if you wish to use a colon to separate
+% the definitions, use `|[.:]|' as the optional argument. The
+% mandatory \meta{left} and \meta{right} are simply the halves of the
+% set definition.
+% \begin{macrocode}
+\newcommand{\set}[3][|]{{%
+ \newdimen\@tempdimd%
+% \end{macrocode}
+% Each half is set in its own box, then the larger of the respective
+% heights and depths are determined.
+% \begin{macrocode}
+ \setbox0=\mathbox{#2}\@tempdima=\ht0 \@tempdimb=\dp0%
+ \setbox0=\mathbox{#3}\@tempdimc=\ht0 \@tempdimd=\dp0%
+ \ifdim\@tempdimc > \@tempdima
+ \@tempdima=\@tempdimc
+ \fi
+ \ifdim\@tempdimd > \@tempdimb
+ \@tempdimb=\@tempdimb
+ \fi
+% \end{macrocode}
+% We create an invisible rule with that height and depth, and make
+% sure we have a valid delimiter if the optional argument is empty.
+% \begin{macrocode}
+ \def\@tempa{\vrule width0pt height\@tempdima depth\@tempdimb}
+ \def\@tempb{#1}
+ \ifx\@empty\@tempb
+ \def\@tempb{.}
+ \fi
+% \end{macrocode}
+% Finally, we use a null left delimiter to balance the middle
+% delimiter, and then a left brace to balance the right brace. The
+% rule is set inside both pairs so that they scale identically. Note
+% the use of |\expandafter| so that when the first |\right| is
+% expanded, it can grab the delimiter in |\@tempb|.
+% \begin{macrocode}
+ \left.\left\{ \@tempa{#2} \,\expandafter\right\@tempb\,{#3} \right\} }}
+% \end{macrocode}
+% For backwards compatibility, I make two aliases for |\set|; the old
+% commands required the user to specify the larger side of the set
+% definition in order to get sizing correct.
+% \begin{macrocode}
+\alias\setl\set
+\alias\setr\set
+% \end{macrocode}
+%
+% Here are some macros for typesetting sets symbolically. First off,
+% we might want to know how to typeset a level set.
+% \begin{macrocode}
+\newcommand{\lvl}[2][\alpha]{\Gamma\ssb{#2}\ssp{(#1)}}
+% \end{macrocode}
+% There are also fuzzy sets, and their corresponding level sets.
+% \begin{macrocode}
+\if@amsmath
+ \newcommand{\fset}[1]{\Tilde{#1}}
+\else
+ \newcommand{\fset}[1]{\tilde{#1}}
+\fi
+\newcommand{\flvl}[2][\alpha]{\lvl[#1]{\fset{#2}}}
+% \end{macrocode}
+% For want of a better font, I will typeset set collections
+% in |\mathcal|.
+% \begin{macrocode}
+\alias\coll\mathcal
+% \end{macrocode}
+%
+% Finally, we deal with some set operators.
+% \emph{The \TeX{}book} points out the difference between |\setminus|
+% and |\backslash|. I prefer to think of them as `set
+% complementation' and `coset', respectively.
+% \begin{macrocode}
+\alias\scomp\setminus
+\alias\coset\backslash
+% \end{macrocode}
+% The next macro attempts to create a symmetric difference operator.
+% I don't like it, but I probably won't do better until I learn to
+% make my own \textsf{M{\small ETAFONT}} characters\ldots
+% \begin{macrocode}
+\newcommand{\symmdiff}{%
+ \mathbin{\text{\footnotesize$\bigtriangleup$}}}
+% \end{macrocode}
+%
+%
+% \subsection{Sequences and series}
+%
+% Just a few macros are required for various sequences and series,
+% mostly for indexing. The best explanation is simply an example.
+% The code
+% \begin{display}{l}
+% |$y \in \seq{x_{ij}}$, where $i\inset{n}$, $j\inrange[0]{m}$|
+% \end{display}
+% produces
+% \begin{display}{l}
+% $y \in \seq{x_{ij}}$, where $i\inset{n}$, $j\inrange[0]{m}$.
+% \end{display}
+%
+% \begin{macrocode}
+\newcommand{\seq} [1] {\curlybrackets{#1}}
+\newcommand{\inset} [2][1]{\in\{ #1,\ldots,#2 \}}
+\newcommand{\inrange}[2][1]{ = #1,\ldots,#2}
+% \end{macrocode}
+%
+%
+% \subsection{Calculus}
+%
+% The calculus macros are relegated to auxiliary files, as I rarely
+% need them.
+%
+% \subsubsection{Derivatives}
+%
+% We load the derivatives in if they are requested.
+% \begin{macrocode}
+\if@derivatives
+ \InputIfFileExists{cjwderiv.tex}{}{%
+ \PackageWarning{cjwmath}{Option `cjwderiv.tex' not found.}
+ \@@derivativesfalse}
+\fi
+% \end{macrocode}
+% This loads both simple and partial derivative macros.
+%\iffalse
+%</math>
+%<*deriv>
+%\fi
+%
+% The derivatives are all variations on the basic |\dd| macro, which
+% should be fairly self explanatory.
+% \begin{macrocode}
+\newcommand{\dd} [2]{\frac{d#1}{d#2}}
+\newcommand{\ddt}[1]{\dd{#1}{t}}
+\newcommand{\ddu}[1]{\dd{#1}{u}}
+\newcommand{\ddv}[1]{\dd{#1}{v}}
+\newcommand{\ddx}[1]{\dd{#1}{x}}
+\newcommand{\ddy}[1]{\dd{#1}{y}}
+
+\newcommand{\sdd} [2]{\frac{d^2#1}{d#2^2}}
+\newcommand{\sddx}[1]{\sdd{#1}{x}}
+\newcommand{\sddy}[1]{\sdd{#1}{y}}
+\newcommand{\sddt}[1]{\sdd{#1}{t}}
+\newcommand{\sddu}[1]{\sdd{#1}{u}}
+\newcommand{\sddv}[1]{\sdd{#1}{v}}
+% \end{macrocode}
+%
+% \subsubsection{Partial derivatives}
+%
+% The partial derivatives are all variations on the theme of |\pard|,
+% which is as |\dd|, replacing the $d$ with $\partial$.
+% \begin{macrocode}
+\newcommand{\pard} [2]{\frac{\partial#1}{\partial#2}}
+\newcommand{\pardx}[1]{\pard{#1}{x}}
+\newcommand{\pardy}[1]{\pard{#1}{y}}
+\newcommand{\pardz}[1]{\pard{#1}{z}}
+\newcommand{\pardu}[1]{\pard{#1}{u}}
+\newcommand{\pardv}[1]{\pard{#1}{v}}
+\newcommand{\pardt}[1]{\pard{#1}{t}}
+
+\newcommand{\spard} [2]{\frac{\partial^2#1}{\partial#2^2}}
+\newcommand{\spardx}[1]{\spard{#1}{x}}
+\newcommand{\spardy}[1]{\spard{#1}{y}}
+\newcommand{\spardz}[1]{\spard{#1}{z}}
+\newcommand{\spardu}[1]{\spard{#1}{u}}
+\newcommand{\spardv}[1]{\spard{#1}{v}}
+\newcommand{\spardt}[1]{\spard{#1}{t}}
+
+\newcommand{\spardxy}[1]{\frac{\partial^2#1}{\partial x\partial y}}
+\newcommand{\spardyx}[1]{\frac{\partial^2#1}{\partial y\partial x}}
+\newcommand{\spardxz}[1]{\frac{\partial^2#1}{\partial x\partial z}}
+\newcommand{\spardzx}[1]{\frac{\partial^2#1}{\partial z\partial x}}
+\newcommand{\spardyz}[1]{\frac{\partial^2#1}{\partial y\partial z}}
+\newcommand{\spardzy}[1]{\frac{\partial^2#1}{\partial z\partial y}}
+% \end{macrocode}
+%\iffalse
+%</deriv>
+%<*math>
+%\fi
+%
+% \subsubsection{Integrals}
+%
+% We load the integrals in if they are requested.
+% \begin{macrocode}
+\if@integrals
+ \InputIfFileExists{cjwinteg.tex}{}{%
+ \PackageWarning{cjwmath}{Option `cjwinteg.tex' not found.}
+ \@@integralsfalse}
+\fi
+% \end{macrocode}
+%\iffalse
+%</math>
+%<*integ>
+%\fi
+%
+% The first macro is simply a variation of |\int| using the |\limits|
+% macro.
+% \begin{macrocode}
+\def\integ{\mathop{\int}\limits}
+% \end{macrocode}
+% Next we have a small shortcut for the differential at the end of an
+% integral. We work around \LaTeX's font encoding macro.
+% \begin{macrocode}
+\alias\latex@d\d
+ \renewcommand{\d}{\,d}
+% \end{macrocode}
+% We have macros for double and triple integrals, with and without
+% |\limits|.
+% \begin{macrocode}
+\newcommand{\dint}{\int\!\!\!\int}
+\newcommand{\dinteg}{\mathop{\int\!\!\!\int}\limits}
+\newcommand{\tint}{\int\!\!\!\int\!\!\!\int}
+\newcommand{\tinteg}{\mathop{\int\!\!\!\int\!\!\!\int}\limits}
+% \end{macrocode}
+% To be honest, I have no idea why I wrote this one. It is probably
+% buried in a homework file of mine somewhere, but I'll be a fiddler
+% crab if I can remember where\dots
+% \begin{macrocode}
+\newcommand{\flushintlim}[1]{{\phantom{#1} #1}}
+% \end{macrocode}
+%\iffalse
+%</integ>
+%<*math>
+%\fi
+%
+% \subsection{Algebra}
+%
+% We define how to typeset an algebra.
+% \begin{macrocode}
+\alias\alg\mathbbm
+% \end{macrocode}
+%
+% \subsubsection{Fields}
+%
+% A field will also be done in blackboard bold.
+% \begin{macrocode}
+\alias\field\mathbbm
+% \end{macrocode}
+% The following fields are defined.
+% \begin{macrocode}
+\newcommand{\C}{\field{C}} % Complex
+\newcommand{\E}{\field{E}} % Euclidean (also Evens)
+% \end{macrocode}
+% Note that $\H$ is a \LaTeX{} accent, so we save it away before
+% redefining it as a field.
+% \begin{macrocode}
+\alias\latex@H\H % Quaternions
+ \renewcommand{\H}{\field{H}} % (Hamiltonian field)
+\newcommand{\N}{\field{N}} % Natural numbers
+\newcommand{\Q}{\field{Q}} % Rationals
+\newcommand{\R}{\field{R}} % Reals
+% \newcommand{\Rn}[1][n]{\R^{#1}}
+\newcommand{\Z}{\field{Z}} % Integers
+\newcommand{\pr}{\field{P}} % Primes
+% \end{macrocode}
+%
+% \subsubsection{Groups}
+%
+% Remember |\NewMathOp|? One use for it is in defining mathematical
+% groups. Here are a bunch.
+% \begin{macrocode}
+% Groups are typeset as operators.
+\NewMathOp {\Aut}{Aut} % Automorphisms
+\NewMathOp {\End}{End} % Endomorphisms
+\NewMathOp {\GL}{GL} % General Linear
+\NewMathOp {\Inn}{Inn} % Inner products
+\NewMathOp {\Pin}{Pin} % Pin
+\NewMathOp {\SL}{SL} % Special Linear
+\NewMathOp {\SO}{SO} % Special Orthogonal
+\NewMathOp {\SU}{SU} % Special Unitary
+\NewMathOp[\mathfrak]{\Sn}{S} % Symmetric
+\NewMathOp {\Spin}{Spin} % Spin
+\NewMathOp {\Sp}{Sp} % Symplectic
+\NewMathOp {\Unit}{U\kern 0pt}% Unitary
+\NewMathOp {\Orth}{O\kern 0pt}% Orthogonal
+\NewMathOp[\mathfrak]{\slin}{sl} % Tangent group to SL
+\NewMathOp[\mathfrak]{\so}{o\kern 0pt} % skew orthogonal
+\NewMathOp[\mathfrak]{\sp}{sp} % skew symplectic
+\NewMathOp[\mathfrak]{\su}{u\kern 0pt} % skew hermitian
+% \end{macrocode}
+%
+% \subsubsection{Linear algebra}
+%
+% If matrices are to be typeset specially, we will use the |\mathcal| font.
+% \begin{macrocode}
+\alias\mtx\mathcal
+% \end{macrocode}
+% I have often seen the letter $\Theta$ used for the matrix of zeros.
+% I like it that way.
+% \begin{macrocode}
+\newcommand{\nullmtx}{\mtx\Theta}
+% \end{macrocode}
+% Taken from Horn and Johnson, a matrix norm can be represented with a
+% triple-bar delimiter.
+% \begin{macrocode}
+\newcommand{\mnorm}[1]{%
+ \left\vert\kern-0.9pt\left\vert\kern-0.9pt\left\vert #1
+ \right\vert\kern-0.9pt\right\vert\kern-0.9pt\right\vert}
+% \end{macrocode}
+% We define the Lie product of two matrices.
+% \begin{macrocode}
+\newcommand{\lie}[1]{\squarebrackets{#1}}
+% \end{macrocode}
+% There is an for the trace (Spur, auf Deutsch) of a matrix\dots
+% \begin{macrocode}
+\NewMathOp{\Spur}{Spur}
+\NewMathOp{\Tr}{Tr}
+% \end{macrocode}
+% \dots as well as for diagonal matrices.
+% \begin{macrocode}
+\NewMathOp{\Diag}{Diag}
+% \end{macrocode}
+%
+% This is a shortcut for putting delimiters around matrices. With
+% AMS-\LaTeX, we use an environment, taking as its two mandatory
+% arguments the left and right delimiters, respectively.
+% \begin{macrocode}
+\if@amsmath
+ \newenvironment{arbmatrix}[2]%
+ {\def\@tempa{#2}\left#1 \matrix}{\endmatrix \right\@tempa}
+% \end{macrocode}
+% Without AMS, we use a command as does standard \LaTeX, and define
+% some basic types.
+% \begin{macrocode}
+\else
+ \newcommand{\arbmatrix}[3]{\left#1 \matrix{#2} \right#3}
+ \providecommand{\bmatrix}[1]{\arbmatrix[{#1}]}
+ \providecommand{\vmatrix}[1]{\arbmatrix|{#1}|}
+\fi
+% \end{macrocode}
+%
+% The next bit of code is used to enter sparse matrices which are
+% often represented in the literature with an oversized zero marking
+% the region of zeros. This takes more than a bit of trickery in
+% \LaTeX. The oversized digit will be put in a box, which in most
+% cases needs horizontal and/or vertical adjustment from the position
+% where it is placed in the matrix by default. The default vertical
+% offset will be called |\numoffset|, and is set by default to the
+% height of a |\Bigmathstrut|.
+% \begin{macrocode}
+\newlength{\numoffset}
+{\setbox0=\hbox{$\Bigmathstrut$}
+ \@tempdima=0.8\ht0\relax
+ \global\numoffset\@tempdima}
+% \end{macrocode}
+% Occasionally, one might wish to use something other than a zero as
+% the oversized digit. We define a generic |\Number| macro to be
+% used. The usage is
+% \begin{display}{l}
+% |\Number|\oarg{raise}\marg{num}
+% \end{display}
+% where \meta{raise} is the amount by which the number is raised and
+% \meta{num} is the number to be used.
+% \begin{macrocode}
+\newcommand{\Number}[2][-\numoffset]{%
+ \@tempdima=#1\relax
+ \smash{\hbox{\raise\@tempdima\@bignumber{#2}}}}
+% \end{macrocode}
+% The macros |\@bignumber| is called to do the dirty work of
+% typesetting the number.
+% \begin{macrocode}
+\newcommand{\@bignumber}[1]{\hbox{\LARGE$#1$}}
+% \end{macrocode}
+% Now, the horizontal adjustment mentioned earlier usually refers to
+% the need to have the large digit straddle two columns in a matrix.
+% This is easily accomplished with |\multicolumn|. Here things start
+% to get ugly, though; |\multicolumn| must be the first thing after
+% the |&| in the array. But the reasonable way to include an optional
+% argument to be passed through to |\Number| is to use the
+% |\newcommand| feature, which ends up putting junk in the way---this
+% is exactly the same problem which arose in writing |\mathbox|
+% earlier. Therefore, there are currently two separate commands, one
+% which takes the optional argument, and one which doesn't. I would
+% really like to remedy this if I ever figure out how.
+% \begin{macrocode}
+\def\bignumber #1{\multicolumn{2}{c}{\Number{#1}}}
+\def\Bignumber[#1]#2{\multicolumn{2}{c}{\Number[#1]{#2}}}
+% \end{macrocode}
+% Here are some specific cases for using zero, as is most commonly the
+% case.
+% \begin{macrocode}
+\newcommand{\Zero}[1][-\numoffset]{\Number[#1]{0}}
+\def\bigzero {\bignumber{0}}
+\def\Bigzero[#1]{\Bignumber[#1]{0}}
+% \end{macrocode}
+%
+% From this we wish to construct various types of matrices. Each
+% variation will have two versions, depending on whether or not
+% AMS-\LaTeX{} has been invoked. Each version, though, has an
+% optional argument which is what will be passed through as
+% \meta{raise} to |\Bigzero|. The names of each matrix are an attempt
+% to indicate the layout. For example, the command |\iidiagi| takes
+% three mandatory arguments which will be the diagonal entries, the
+% first two in the upper left, and the third in the lower right with
+% |\ddots| separating them. Likewise, we have |\idiagii| and
+% |\idiagi|.
+% \begin{macrocode}
+\if@amsmath
+ \newcommand{\iidiagi}[4][-\numoffset]{% % 2 0
+ \begin{bmatrix} % 3
+ #2 & & \Bigzero[#1] \\ % .
+ & #3 & & \\ % .
+ \Bigzero[#1] & \ddots & \\ % 0 4
+ & & & #4
+ \end{bmatrix}}
+ \newcommand{\idiagii}[4][-\numoffset]{% % 2 0
+ \begin{bmatrix} % .
+ #2 & & \Bigzero[#1] \\ % .
+ & \ddots & & \\ % 3
+ \Bigzero[#1] & #3 & \\ % 0 4
+ & & & #4
+ \end{bmatrix}}
+ \newcommand{\idiagi}[3][-1.2pt]{% % 2 0
+ \begin{bmatrix} % .
+ #2 & \Bigzero[#1] \\ % .
+ & \ddots & \\ % .
+ \Bigzero[#1] & #3 % 0 3
+ \end{bmatrix}}
+\else
+ \newcommand{\iidiagi}[4][-\numoffset]{% % 2 0
+ \matrix{% % 3
+ #2 & & \Bigzero[#1] \\ % .
+ & #3 & & \\ % .
+ \Bigzero[#1] & \ddots & \\ % 0 4
+ & & & #4}}
+ \newcommand{\idiagii}[4][-\numoffset]{%
+ \pmatrix{% % 2 0
+ #2 & & \Bigzero[#1] \\ % .
+ & \ddots & & \\ % .
+ \Bigzero[#1] & #3 & \\ % 3
+ & & & #4}} % 0 4
+ %
+ \newcommand{\idiagi}[3][-1.2pt]{% % 2 0
+ \pmatrix{% % .
+ #2 & \Bigzero[#1] \\ % .
+ & \ddots & \\ % .
+ \Bigzero[#1] & #3}} % 0 3
+\fi
+% \end{macrocode}
+%
+% We now wish to typeset the transpose of a matrix. The most general
+% form is
+% \begin{display}{l}
+% |\@trans|\oarg{pre}\marg{post}
+% \end{display}
+% which expands to `|^{|\meta{pre}|t|\meta{post}|}|'. Next is
+% |\trans| which takes a single optional argument for \meta{post} (no
+% \meta{pre}).
+% \begin{macrocode}
+\newcommand{\@trans}[2][]{^{#1\text{\normalfont\textsf{t}}#2}}
+\newcommand{\trans} [1][]{\@trans[]{#1}}
+% \end{macrocode}
+% Why? I don't know---I needed |\trinv|, below, and decided to
+% generalize.
+% \begin{macrocode}
+\newcommand{\trinv} {\@trans[-]{}}
+% \end{macrocode}
+% For backwards compatibility, I have |\ct|\marg{mtx} to represent
+% matrix \meta{mtx} as conjugated and transposed.
+% \begin{macrocode}
+\newcommand{\ct}[1]{\conj{#1}\trans}
+% \end{macrocode}
+%
+% The next topic is vectors. I prefer |\vec| to be logical markup as
+% opposed to a specific accent. Thus, I create an alias |\sarvec|
+% (short arrow vector) for the original |\vec|, and another alias
+% |\arvec| for a long arrow, which is just |\overrightarrow|.
+% \begin{macrocode}
+\alias\sarvec\vec
+\alias\arvec \overrightarrow
+% \end{macrocode}
+% The actual typesetting I prefer for vectors (when I use anything at
+% all) is boldface. This requires the |\Mathbox| command defined
+% earlier so that the argument can be set in a bold math version.
+% \begin{macrocode}
+\renewcommand{\vec}[1]{\Mathbox{\boldmath}{#1}}
+% \end{macrocode}
+% To typeset vectors in long form simply uses the |\matrix|
+% command---but this depends, again, on whether AMS-\LaTeX{} is being
+% used. In either case, a single argument---the contents of the
+% vector---is required. Simply delimit with |\\| for column vectors
+% and |&| for rows.
+% \begin{macrocode}
+\if@amsmath
+ \newcommand{\bvec}[1]{%
+ \begin{bmatrix}#1\end{bmatrix}}
+ \newcommand{\pvec}[1]{%
+ \begin{pmatrix}#1\end{pmatrix}}
+% \end{macrocode}
+% There are also two aliases for row vectors for backwards
+% compatibility.
+% \begin{macrocode}
+ \alias\brvec\bvec
+ \alias\prvec\pvec
+\else
+ \newcommand{\bvec}[1]{\bmatrix{#1}}
+ \newcommand{\pvec}[1]{\pmatrix{#1}}
+% \newcommand{\bvec}[2][r]{%
+% \left[ \begin{array}{#1}#2\end{array} \right]}
+% \newcommand{\pvec}[2][r]{%
+% \left( \begin{array}{#1}#2\end{array} \right)}
+\fi
+% \end{macrocode}
+% The null vector, like the null matrix, is given with $\theta$.
+% \begin{macrocode}
+\newcommand{\nullvec}{\vec{0}}
+% \end{macrocode}
+% Once we have vectors, we need dot products. The simple version just
+% puts its argument inside angled brackets.
+% \begin{macrocode}
+\newcommand{\dotp}[1]{\anglebrackets{#1}}
+% \end{macrocode}
+% If special vector notation is required, the two arguments should be
+% separated by a comma (which should be the case anyway), and then
+% each half is passed to |\vec|.
+% \begin{macrocode}
+\newcommand{\vdotp}[1]{\@vdotp#1@@@}
+\def\@vdotp #1,#2@@@{\dotp{\vec #1,\vec #2}}
+% \end{macrocode}
+%
+% We define some other vector operators to go with the dot product.
+% These are the curl, the divergence, and the Laplacian. These are
+% usually read as `del dot', `del cross', and `del squared', so the
+% first thing to do is rename the |\nabla|(?).
+% \begin{macrocode}
+\newcommand{\del} {\vec\nabla}
+% \end{macrocode}
+% \LaTeX{} defines |\div|, so we rename that and renew the command.
+% \begin{macrocode}
+\alias\@@div\div
+\renewcommand{\div}{\del\dot}
+% \end{macrocode}
+% Finally, we have the other two.
+% \begin{macrocode}
+\newcommand{\curl} {\del\cross}
+\newcommand{\lapl} {\del^2}
+% \end{macrocode}
+%
+% In German texts, the linear hull is usually represented as a list of
+% vectors in square brackets.
+% \begin{macrocode}
+\alias\huelle\squarebrackets
+% \end{macrocode}
+%
+% \subsection{Operators}
+%
+% This section is simply a gathering point for all sorts of
+% mathematical operators---not in the |\NewMathOp| sense, like various
+% groups defined above, but in the sense of mathematical doohickies
+% which take one or two operands and give you something new.
+%
+% \subsubsection{Binary operators}
+%
+% What we have first is a whole bunch of names for the large, `x'-like
+% times symbol. In the case of specifying dimension (as in, a 4-by-3
+% matrix), we declare it a |\mathord| so as not to invoke binary
+% spacing. Then |\mal| is simply German for `times', and |\cross| is
+% the vector space (or group) product using the same symbol.
+% \begin{macrocode}
+\newcommand{\by}{\mathord{\times}}
+\alias\mal \times
+\alias\cross \times
+% \end{macrocode}
+% To indicate isomorphism, I have most often used an equals sign with
+% a tilde above it.
+% \begin{macrocode}
+\alias\iso \simeq
+% \end{macrocode}
+% This next one stretches the usefulness of aliasing by defining an
+% operator for normal subgroups.
+% \begin{macrocode}
+\alias\nsubgrp \trianglelefteq
+% \end{macrocode}
+% Next we specify a congruence symbol.
+% \begin{macrocode}
+\realias\cong \equiv
+% \end{macrocode}
+% In graph theory, we want a symbol for adjacency of nodes.
+% \begin{macrocode}
+\alias\adj\leftrightarrow
+% \end{macrocode}
+% We unconditionally define |\Box| to be a square operator symbol.
+% \begin{macrocode}
+\providecommand{\Box}{}
+\renewcommand{\Box}{\mathbin{\square}}
+% \end{macrocode}
+% We give a number theoretic division relation, in English and
+% German.
+% \begin{macrocode}
+\newcommand{\teilt}{\mathbin{|}}
+ \alias\divides\teilt
+% \end{macrocode}
+% Once upon a time I wanted a `big dot' operator.
+% \begin{macrocode}
+% \newcommand{\bdot}{\mathop{\lower0.33ex\hbox{\LARGE$\cdot$}}}
+% \end{macrocode}
+% We can use \LaTeX's built-in |\stackrel| to create a definition
+% relation.
+% \begin{macrocode}
+\newcommand{\defeq}{\stackrel{\text{def}}{=}}
+% \end{macrocode}
+% The symbol I first saw used to indicate disjoint union was a union
+% `cup' with a bar through the middle. Using a naming convention
+% similar to AMS-\LaTeX's |\Uplus| we get text and display versions.
+% \begin{macrocode}
+\newcommand{\uminus}{%
+ \,\,{\mathbin{\cup\kern-.6em{\raise.05em%
+ \hbox{-\negthinspace-\kern-.25em-}}}}\,\,}
+\newcommand{\Uminus}{%
+ \mathop{\bigcup\kern-0.9em{\raise.05em%
+ \hbox{-\negthinspace-\kern-.25em-}}}}
+% \end{macrocode}
+% We define some handy names for various arrows.
+% \begin{macrocode}
+\providecommand{\implies}{\;\Longrightarrow\;}
+ \alias\then\implies
+\if@amsmath
+ \newcommand{\when}{\DOTSB \;\Longleftarrow \;}
+\else
+ \newcommand{\when}{\;\Longleftarrow \;}
+\fi
+% \end{macrocode}
+% The following can be used when tracing a series of implications
+% through a multiline equation environment, for example.
+% \begin{macrocode}
+\newcommand{\limplies}{\llap{$\implies$}\quad}
+% \end{macrocode}
+% Based on \emph{The \TeX{}book}, I have written a `skewed' fraction,
+% which uses a diagonal separator.
+% \begin{macrocode}
+\newcommand{\sfrac}[2]{%
+ \hbox{\kern 0.1em%
+ \raise 0.5ex\hbox {\scriptsize$#1$}%
+ \kern -0.1em $/$%
+ \kern -0.15em%
+ \lower 0.25ex\hbox {\scriptsize$#2$}}%
+ \kern 0.2em}
+% \end{macrocode}
+% If we are not using AMS-\LaTeX, the following well not yet be
+% defined.
+% \begin{macrocode}
+\providecommand{\tfrac}{\sfrac}
+\providecommand{\dfrac}[2]{{{#1}\over{#2}}}
+% \end{macrocode}
+%
+%
+% \subsubsection{Unary operators}
+%
+% I most often use the longer versions of various math accents, so I
+% redefine them by default to be long. The short ones are saved in a
+% macro with identical name save for a prepended `s' (as we have
+% already seen for |\arvec| and |\sarvec|). The simple ones are bars,
+% tildes, and hats.
+% \begin{macrocode}
+\alias\sbar\bar
+ \renewcommand{\bar}[1]{\overline{#1}}
+\alias\stilde\tilde
+ \alias\retilde\widetilde
+\alias\shat\hat
+ \realias\hat\widehat
+% \end{macrocode}
+% We need something better than the default real- and imaginary-part
+% macros.
+% \begin{macrocode}
+\renewcommand{\Im}{%
+ \mathop{\mathfrak{Im}}}
+\renewcommand{\Re}{%
+ \mathop{\mathfrak{Re}}}
+% \end{macrocode}
+% Complex conjugation is usually denoted by a bar.
+% \begin{macrocode}
+\alias\conj \bar
+% \end{macrocode}
+% Inversion of various types is used often enough to warrant a control
+% sequence.
+% \begin{macrocode}
+\newcommand{\inv}{^{-1}}
+% \end{macrocode}
+% To denote power sets, I have used to different alternatives. The
+% default is the standard |\wp| symbol---I don't know what it is
+% supposed to be used for, but it can be modified for the task. On
+% the other hand, if we have a real math script font (as one might
+% have if using my \pkg{callig} style\dots) we will certainly use
+% that.
+% \begin{macrocode}
+\@ifundefined{mathscript}
+ {\newcommand{\Pow}{\raise 0.4ex\Mathbox{\Large}{\wp}}}
+ {\NewMathOp[\mathscript]{\Pow}{P}}
+% \end{macrocode}
+% And what macro would be complete without a German alias? (Auf
+% Deutsch, die Potenzmenge.)
+% \begin{macrocode}
+\alias\Pot\Pow
+% \end{macrocode}
+% We can leave the letter `\/I\/' to computer scientists who don't
+% know how to write an indicator function. For our purposes, we want
+% the neat-o-keen blackboard bold font.
+% \begin{macrocode}
+\newcommand{\1}{\mathbbm{1}}
+% \end{macrocode}
+%
+% Next we have some trigonometric shortcuts.
+% \begin{macrocode}
+\alias\acos\arccos
+\alias\asin\arcsin
+\alias\atan\arctan
+% \end{macrocode}
+% Using our generic bracketing macros from earlier, we have absolute
+% value, cardinality (order of a set), cyclic generators, and norms.
+% \begin{macrocode}
+\alias\abs \vertbrackets
+\alias\ord \abs
+\alias\cyc \anglebrackets
+\alias\norm\Vertbrackets
+% \end{macrocode}
+%
+% Multiple sums are recurrent enough (no pun intended \smiley) to get
+% macros. We have double sums, triple sums, and $n$-fold sums.
+% \begin{macrocode}
+\newcommand{\dsum}{\mathop{\sum\sum}\limits}
+\newcommand{\tsum}{\mathop{\sum\sum\sum}\limits}
+\newcommand{\nsum}{\mathop{\sum\sum\cdots\sum}\limits}
+% \end{macrocode}
+%
+% The next two macros indicate monotone limits, respectively ascending
+% (mnemonic `limit up') and descending (you guessed it---`limit
+% down').
+% \begin{macrocode}
+\NewMathOp*{\ulim}{lim\raise0.4ex\mathbox{\mathord{\smash{\uparrow}}}}
+\NewMathOp*{\dlim}{lim\raise0.4ex\mathbox{\mathord{\smash{\downarrow}}}}
+% \end{macrocode}
+%
+%
+% Finally we have the miscellany, where we can really go to town with
+% |\NewMathOp|! These should be pretty clear, unless you don't do
+% math in German, in which case some will be pretty odd.
+% \begin{macrocode}
+\NewMathOp*{\argmax}{arg\,max} % arg min
+\NewMathOp*{\argmin}{arg\,min} % arg min
+\NewMathOp {\Aff} {Aff} % Affine hull
+\NewMathOp {\Bild} {Bild} % Bild
+\NewMathOp {\Cone} {Cone} % Cone
+\NewMathOp {\Conv} {Conv} % Convex hull
+\NewMathOp {\Core} {Core} % Fuzzy set core
+\NewMathOp {\diam} {diam} % diameter
+\NewMathOp {\dom} {dom} % Domain
+\NewMathOp {\Epi} {Epi} % Epigraph
+\NewMathOp*{\esssup}{ess\,sup} % Essential supremum
+\NewMathOp {\fl} {fl} % float-point
+\NewMathOp {\ggT} {ggT} % ggT
+\NewMathOp {\Grad} {Grad} % Grad
+\NewMathOp {\Hypo} {Hypo} % Hypograph
+\NewMathOp {\Int} {Int} % Interior
+\NewMathOp {\Kern} {Kern} % Kernel
+\NewMathOp {\kgV} {kgV} % kgV
+\NewMathOp {\Lin} {Lin} % Linear hull
+\NewMathOp {\lcm} {lcm} % LCM
+\NewMathOp {\Ord} {Ord} % order
+\NewMathOp {\proj} {proj} % Projection
+\NewMathOp {\Rang} {Rang} % Rang
+\NewMathOp {\range} {range} % Range
+\NewMathOp {\Rank} {Rank} % Rank
+\NewMathOp {\rot} {rot} % Rotation
+\NewMathOp {\Span} {Span} % Span
+\NewMathOp {\val} {val} % value
+% \end{macrocode}
+%
+%
+% \subsection{Physics}
+%
+% Now, since physics is a subset of mathematics, physics macros are
+% invoked from within the math macro file.
+% \begin{macrocode}
+\if@physics
+ \InputIfFileExists{cjwphys.tex}{}{%
+ \PackageWarning{cjwmath}{Option `cjwphys.tex' not found.}
+ \@@physicsfalse}
+\fi
+% \end{macrocode}
+%\iffalse
+%</math>
+%<*phys>
+%\fi
+%
+% The only thing really specific to physics that I ever used---meaning
+% not applicable in any more general mathematical setting---was the
+% bra/ket notation. Here we have bras (br\ae?), kets, and brakets,
+% the latter of which bear uncoincidental resemblance to the |\set|
+% macro.
+% \begin{macrocode}
+\newcommand{\bra}[1]{\left\langle #1 \right|\,}
+\newcommand{\ket}[1]{\,\left| #1 \right\rangle}
+
+\newcommand{\braket}[2]{%
+ \newdimen\@tempdimd%
+ \setbox0=\mathbox{#1}\@tempdima=\ht0 \@tempdimb=\dp0%
+ \setbox0=\mathbox{#2}\@tempdimc=\ht0 \@tempdimd=\dp0%
+ \ifdim\@tempdimc > \@tempdima
+ \@tempdima=\@tempdimc
+ \fi
+ \ifdim\@tempdimd > \@tempdimb
+ \@tempdimb=\@tempdimb
+ \fi
+ \def\@tempa{\vrule width0pt height\@tempdima depth\@tempdimb}
+ \left.\left\langle \@tempa{#1} \,\right|\,{#2} \right\rangle }
+% \end{macrocode}
+%\iffalse
+%</phys>
+%<*math>
+%\fi
+%
+%
+% \subsection{Probability}
+%
+% Here comes that |\NewMathOp| thingie again. First thing is to
+% define some standard probabilistic operators, which really need to
+% be typeset in an operator font in order not to look horrible.
+% \begin{macrocode}
+\NewMathOp{\Prob} {P} % Probability operator
+\NewMathOp{\Corr} {Corr} % Correlation
+\NewMathOp{\Cov} {Cov} % Covariance
+\NewMathOp{\Expct}{E} % Expectation
+\NewMathOp{\SD} {SD} % Standard Deviation.
+\NewMathOp{\Var} {Var} % Variance
+% \end{macrocode}
+% Here is a macro to put inside of some of those operators where
+% conditional events are being considered.
+% \begin{macrocode}
+\newcommand{\given}{\,|\,}
+% \end{macrocode}
+% Usually a single tilde, which as an operator bears the name of
+% |\sim| in \LaTeX, indicates a distribution. Hence, we make an
+% alias.
+% \begin{macrocode}
+\alias\distrib\sim
+% \end{macrocode}
+% And coming back to |\NewMathOp|, we define some standard
+% distributions\dots
+% \begin{macrocode}
+\NewMathOp{\Bin} {Bin} % Binary dist.
+ \newcommand{\Nbin}{-\!\Bin} % Negative Binom.
+\NewMathOp{\Exp} {Exp} % Exponential dist.
+\NewMathOp{\Geom}{Geom} % Geometric dist.
+\NewMathOp{\Norm}{Norm} % Normal dist.
+\NewMathOp{\Poi} {Poi} % Poisson dist.
+\NewMathOp{\Unif}{Unif} % Uniform dist.
+% \end{macrocode}
+% \dots and the normal density and distribution functions (which might
+% also be represented as $\Phi$ and $\phi$).
+% \begin{macrocode}
+\NewMathOp[\mathfrak]{\Ndens}{N}
+\NewMathOp[\mathfrak]{\Ndist}{n}
+% \end{macrocode}
+% The last thing to do is create some macros for probabilistic modes
+% of convergence. We go, again, from the general to the specific.
+% \begin{macrocode}
+\NewMathOp*{\@mapsto}{\mapstochar\rightarrow}
+\newcommand{\@probconv}[1]{\mathrel{\@mapsto\limits^{1}}}
+ \newcommand{\asconv}{\@probconv{a.s.}} % Almost sure conv.
+ \newcommand{\inprob}{\@probconv{P}} % Conv. in probability
+ \newcommand{\inlaw} {\@probconv{L}} % Conv. in law
+ \newcommand{\vague} {\@probconv{v}} % Vague conv.
+% \end{macrocode}
+%
+% \iffalse
+%
+%</math>
+%
+% \fi
+%
+%
+% \section{Units}
+%
+% \iffalse
+%
+%<*units>
+%
+% \fi
+%
+% The package \pkg{cjwunits} simply standardizes how to typeset units
+% (or dimensions---things such as seconds, meters, and so forth).
+% Most of the work is done by defining a pretty simple macro,
+% |\unit|. The rest is just a collection of standard units which may
+% be invoked (meaning ones I have used, and therefore stuck into the
+% file).
+%
+% \subsection{Package initialization}
+%
+% The initialization does most everything. We first specify a font
+% for the unit types.
+% \begin{macrocode}
+\newcommand{\unitfont}{\operator@font}
+% \end{macrocode}
+% Now the workhorse of the package is defined. It is pretty
+% foolproof, in that it can be invoked in or out of math, may or may
+% not be followed by explicit space, and the font as defined above is
+% pretty customizable.
+% \begin{macrocode}
+\newcommand{\unit}[1]{\ensuremath{\,{\unitfont{#1}\kern\z@}}\xspace}
+% \end{macrocode}
+%
+% Now come the examples.
+%
+% \subsection{Distance}
+% \begin{macrocode}
+\newcommand{\ang} {\unit{\AA}} % angstroms
+ \alias\Ao\ang
+\newcommand{\cm} {\unit{cm}} % centimetres
+\newcommand{\inch}{\unit{in}} % inches
+\newcommand{\km} {\unit{km}} % kilometres
+\newcommand{\mi} {\unit{mi}} % miles
+\newcommand{\m} {\unit{m}} % metres
+% \end{macrocode}
+%
+% \subsection{Electricity and magnetism}
+%
+% \begin{macrocode}
+\newcommand{\Hz} {\unit{Hz}} % herz
+\newcommand{\J} {\unit{J}} % joules
+\newcommand{\V} {\unit{V}} % volts
+\newcommand{\eV} {\unit{eV}} % electron volts
+\newcommand{\erg} {\unit{erg}} % ergs
+% \end{macrocode}
+%
+% \subsection{Mass}
+%
+% \begin{macrocode}
+\newcommand{\amu} {\unit{amu}} % atomic mass units
+\newcommand{\gram}{\unit{g}} % grams
+\newcommand{\kg} {\unit{kg}} % kilograms
+\newcommand{\Ton} {\unit{T}} % tons
+\newcommand{\kT} {\unit{kT}} % kilotons
+\newcommand{\MT} {\unit{MT}} % megatons
+% \end{macrocode}
+%
+% \subsection{Thermodynamics}
+%
+% \begin{macrocode}
+\newcommand{\kelv}{\unit{K}} % kelvins
+% \end{macrocode}
+%
+% \subsection{Time}
+%
+% It seems that |\sec| already means secant, so we need a preservation
+% and renewal here.
+% \begin{macrocode}
+\alias\secant\sec
+\renewcommand{\sec} {\unit{s}} % seconds
+% \end{macrocode}
+%
+% \subsection{Velocity}
+%
+% \begin{macrocode}
+\newcommand{\cee} {\unit{c}} % speed o' light
+% \end{macrocode}
+%
+% \iffalse
+%
+%</units>
+%
+% \fi
diff --git a/Master/texmf-dist/source/latex/cjw/cjw-latex.ins b/Master/texmf-dist/source/latex/cjw/cjw-latex.ins
new file mode 100644
index 00000000000..447e39c97c0
--- /dev/null
+++ b/Master/texmf-dist/source/latex/cjw/cjw-latex.ins
@@ -0,0 +1,44 @@
+% -*- TeX -*- %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+\def\batchfile{cjw-latex.ins}
+\input docstrip.tex
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+\preamble
+
+ Copyright (C) 1997 by: Colin J. Wynne <cwynne@mts.jhu.edu>
+
+This file is part of cjw-latex, a collection of package files for
+LaTeX2e, written by Colin J. Wynne. It may be freely distributed in
+any form---electronic, physical, or otherwise---so long as the
+distributor provides this file in its entirety (including this
+copyright announcement) and charges no fee (except for reasonable
+recompense for the duplication process). Fair 'nough?
+
+I would also prefer that the entire source file `cjw-latex.dtx',
+together with `cjw-latex.ins', be propagated instead of any one
+subfile.
+
+Please feel free to improve on or alter any code herein as suits your
+needs, and by all means notify me of any suggestions, improvements, or
+requests.
+
+This file is beer-ware. If you use it and like it, then you should
+buy me a beer if you ever meet me. :-)
+
+RCS: $Id: cjw-latex.ins,v 1.3 1998/04/10 14:30:18 cwynne Exp $
+
+\endpreamble
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+
+% Don't display progress
+\keepsilent
+
+\generateFile{cjwmacro.sty}{t}{\from{cjw-latex.dtx}{general}}
+\generateFile{cjwmath.sty} {t}{\from{cjw-latex.dtx}{math}}
+\generateFile{cjwderiv.tex}{t}{\from{cjw-latex.dtx}{deriv}}
+\generateFile{cjwinteg.tex}{t}{\from{cjw-latex.dtx}{integ}}
+\generateFile{cjwphys.tex} {t}{\from{cjw-latex.dtx}{phys}}
+\generateFile{cjwunits.sty}{t}{\from{cjw-latex.dtx}{units}}
+
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+\endinput
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
diff --git a/Master/texmf-dist/source/latex/cjw/cjw-ltr.dtx b/Master/texmf-dist/source/latex/cjw/cjw-ltr.dtx
new file mode 100644
index 00000000000..42e80bc36ee
--- /dev/null
+++ b/Master/texmf-dist/source/latex/cjw/cjw-ltr.dtx
@@ -0,0 +1,821 @@
+% \iffalse -*- LaTeX -*-
+%
+% This is file `cjw-ltr.dtx'. You can run this file through LaTeX2e
+% to get a .dvi file of documentation. The file `cjw-ltr.ins' should
+% have come with this file. Run it through LaTeX2e to extract the
+% actual class file.
+%
+% \fi
+\def\fileversion{0.9}
+\def\filedate{1997/01/01}
+%% \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 \~}
+%%
+% \CheckSum{592}
+%
+% \iffalse
+%
+%<*driver>
+
+\NeedsTeXFormat{LaTeX2e}
+\ProvidesFile{cjw-ltr.dtx}
+ [\filedate\space v\fileversion\space
+ Modified Letter Class for LaTeX2e---CJW]
+
+\documentclass{ltxdoc}
+
+\GetFileInfo{cjw-ltr.dtx}
+\EnableCrossrefs
+% \DisableCrossrefs % Say \DisableCrossrefs if index is ready
+% \OnlyDescription % comment out for implementation details
+\RecordChanges
+
+\begin{document}
+ \DocInput{cjw-ltr.dtx}
+\end{document}
+%</driver>
+% \fi
+%
+% \DeclareRobustCommand{\cseq}[1]{\texttt{\bslash#1}}
+% \DeclareRobustCommand{\pkg} [1]{\textsf{#1}}
+% \DeclareRobustCommand{\env} [1]{\textsf{#1}}
+%
+% \providecommand{\ie}{\emph{i.e.},\ }
+%
+% \renewcommand{\thefootnote}{\fnsymbol{footnote}}
+%
+% \hyphenation{white-space}
+%
+% \title{The \pkg{cjw-ltr} Class\thanks{%
+% This file has version \fileversion{} as of \filedate.}}
+% \author{Colin J.~Wynne\thanks{E-Mail at:
+% \texttt{cwynne@mts.jhu.edu}, \texttt{cwynne@jhu.edu}\,.}}
+% \date{\filedate}
+%
+% \maketitle
+%
+% \setcounter{StandardModuleDepth}{1}
+%
+% {\parskip 0pt ^^A % This is the hack used by |doc.dtx|.
+% ^^A % (bug in \LaTeX?)
+% \tableofcontents
+% }
+%
+% \section*{Introduction}
+%
+% I recently made the change to \LaTeXe{} after some years of
+% dedication to \PlainTeX{} and that code which I had already
+% written for it. Among that code were macros for writing letters.
+% Since I am studying in Germany this year, I had also written a set
+% of macros for writing proper letters according to accepted German
+% format.
+%
+% After switching to \LaTeX, I of course gave the \pkg{letter} class a
+% try. I didn't like some of the default behaviour that the class
+% produces. Furthermore, there was no support for writing German
+% letters. I am aware that a style file \pkg{dinbrief} exists for
+% \LaTeX{} 2.09 or \LaTeXe{} in compatibility mode, but I would prefer
+% to use something exclusively \LaTeXe. Since I planned on doing some
+% cosmetic alterations to the basic \pkg{letter} class anyway, I
+% figured it wouldn't be too much extra to include the capability of
+% formatting German letters as well.
+%
+% So, this is what I have come up with. Feel free to e-mail me
+% comments, suggestions, bugs, etc.
+%
+%
+% \section{Creating Letters}
+%
+% \DescribeMacro{\letterskip}
+% The first problem that I had with the \pkg{letter} class was
+% spacing. I do not like having the paragraphs of my letters
+% unindented and separated by white space. This poses a problem since
+% the vertical space between the components of the letter are defined
+% (nay, \emph{hard-wired}) as multiples of \cseq{parskip}. Therefore,
+% all vertical space has been defined in terms of \cseq{letterskip}, a
+% length which is set initially to 0.7\,em, which happens to be the
+% \pkg{letter} class's default value for \cseq{parskip}.
+%
+% \subsection{Standard Letters}
+%
+% \DescribeEnv{letter}
+% \DescribeMacro{fax}
+% American format letters are still delineated by the \env{letter}
+% environment. No changes have been made here; the single mandatory
+% argument (which may be left empty) is still the address of the
+% recipient. I have, however, added a macro \cseq{fax}, analagous to
+% \cseq{telephone}, for including a fax number in the return address.
+% If the recipient's fax is also desired, it should probably just be
+% included in the argument to the \env{letter} environment, perhaps
+% using the \cseq{faxlabel} macro, as well (see below).
+%
+% Now, one of the things that I have always been taught about proper
+% business letters is that the closing and the sender's address should
+% line up, and be as far right as possible. The standard \pkg{letter}
+% class does not do this; instead, the address is set as far right as
+% possible and the closing begins \cseq{indentedwidth} from the right
+% margin. By default, this is exactly in the middle of the page. I
+% can not express satisfactorily just how ugly and unbalanced this
+% seems to me.
+% \DescribeMacro{\addresswidth}
+% So, first I have \cseq{let} the length \cseq{indentedwidth} to the slightly
+% more intuitive \cseq{addresswidth}. When the letter's \cseq{opening} is
+% declared, \cseq{addresswidth} is fixed at the size of the return
+% address. Both the sending address and the closing will be set
+% at in a box of width \cseq{addresswidth}; the box is flush right,
+% the material within the box is flush left.
+%
+% \DescribeMacro{\opening}
+% The \cseq{opening} macro still takes the text of the opening as its one
+% argument, and still begins the letter by typesetting the addresses
+% and date. There are a few differences in the typesetting, though.
+% First, if a \cseq{telephone} command has been entered, the number will
+% be set as the last line of the address. Then comes the skip between
+% the address and the date. This skip is rather smaller that that
+% provided by the \pkg{letter} class. I found this to be
+% preferable. Then come the recipient's address and the greeting.
+% After these are typeset, the \cseq{parindent} is set to 20\,pt and the
+% \cseq{parskip} to 0\,pt plus 1\,pt. The values can be reset after the
+% \cseq{opening} call and before the text of the body if such is desired.
+%
+% \DescribeMacro{\closing}
+% The only change to the closing macro is the substitution of
+% \cseq{letterskip} for the original \cseq{medskipamount}. This means that
+% all vertical space in a letter---between address and date and
+% recipient's address, between address and greeting, etc.---are
+% proportional. I considered defining each of those quantities to be
+% their own macro, such as \cseq{beforedateskip}, \cseq{beforerecipientskip},
+% and so forth, but that seemed a little bit excessive. If, however,
+% such flexibility is desired, feel free to ask for it.
+%
+% \subsection{German Letters}
+%
+% \DescribeEnv{brief}
+% \llap{\footnotemark}\footnotetext{Dies soll wirklich auf Deutsch
+% sein, oder? Das hatte ich betrachtet, aber es schien mir ein
+% bi\ss chen durcheinander, die beiden Sprachen in dieser
+% Dokumentation zu mischen. Wenn jemand die \"Ubersetzung will,
+% k\"onnte er mich fragen. Anderseits ist jeder frei, die
+% \"Ubersetzung selbst zu erledigen.}^^A%
+% The environment for a German letter is, appropriately enough, named
+% \env{brief} (the German word for letter). As for |begin{letter}|,
+% the invocation takes the recipient's address as the one mandatory
+% argument. However, in standard format, the address line containing
+% the \emph{Postleitzahl} (zip code, postal code---whatever) and the
+% city is separated by a bit of whitespace from the rest of the
+% address. Thus, whereas other lines of the address are delineated
+% by the |\\| macro, the sequence |!!| indicates the beginning of the
+% PLZ. Thus, my former address would most likely be entered as:
+% \begin{verbatim}
+% \begin{brief}{%
+% Herrn \\
+% Colin J. Wynne \\
+% Klosterweg 28 / L110 !!
+% 76131 Karlsruhe}
+% \end{verbatim}
+%
+% \DescribeMacro{\totitle}
+% \DescribeMacro{\toPLZ}
+% I have not included any macros for creating mailing labels---other
+% than the standard macros from the \pkg{letter} class. However, if
+% someone feels the need to write such, then the following is
+% significant. For a standard letter, the first line is taken as the
+% recipient's `title'. This can be a proper academic or professional
+% title, or, more commonly, simply a polite \emph{Herrn}, as in the
+% example above. This is stored in the macro \cseq{totitle}. The PLZ
+% information (everything in the argument after |!!|) is stored in
+% \cseq{toPLZ}. These complement the standard \cseq{toname} and \cseq{toaddress}.
+% I would add appropriate macros myself, but I have no facilities for
+% printing mailing labels to check the result. I imagine, though,
+% that setting
+% \begin{verbatim}
+% \edef\toname{\totitle \\ \toname}
+% \edef\toaddress{\toaddress \\ \toPLZ}
+% \end{verbatim}
+% would allow existing labelling functions to suffice. (\cseq{edef} to
+% avoid a recursive definition, of course.)
+%
+% \DescribeMacro{\ort}
+% \DescribeMacro{\datum}
+% \DescribeMacro{\heute}
+% \DescribeMacro{\absender}
+% \DescribeMacro{\telefon}
+% \DescribeMacro{\unterschrift}
+% German equivalents are allowed for the standard information
+% providing macros. \cseq{ort} is equivalent to \cseq{location}, and can be
+% used to provide the place name on the dateline. The date itself can
+% be entered with the \cseq{datum} command, and the \cseq{heute} macro
+% produces the current date in German format. Thus, the input
+% |\ort{Karlsruhe} \datum{\heute}| could produce the dateline
+% `Karlsruhe, den 14. Mai 1995'
+% at the top of the letter. \cseq{absender} is used as \cseq{address} to
+% indicate the sender's address, with \cseq{telefon} to provide a
+% telephone number, and \cseq{fax} is of course the same as in
+% English. Finally, \cseq{unterschrift} provides the signature after
+% the closing.
+% \DescribeMacro{\betreff}
+% \DescribeMacro{\bezug}
+% Commands are also given for entering the optional \emph{Betreff} and
+% \emph{Bezug} information. A note to purists: one should include
+% both or neither of these lines. By default, the labels ``Betreff''
+% and ``Bezug'' are printed at the beginning of their respective
+% lines. The actual label (including an empty label) can be entered
+% as an optional argument to the \cseq{betreff} or \cseq{bezug} commands.
+%
+% \DescribeMacro{\anrede}
+% \DescribeMacro{\gruss}
+% The opening and closing of a letter are called by macros named for
+% the German equivalents to \cseq{opening} and \cseq{closing}, namely
+% \cseq{anrede} and \cseq{gruss}.
+%
+% \DescribeMacro{\anlage}
+% Enclosures are indicated by the \cseq{anlage} command, which calls the
+% standard \cseq{encl} macro. \cseq{anlage} takes an optional argument,
+% namely the \cseq{enclname} macro, which is used to label the enclosures.
+% By default, ``Anlage'' is the label. Since multiple enclosures are
+% fairly standard, one can enter simply `n' (that is,
+% |\anlage[n]|\meta{enclosures}) and produce the expected label
+% ``Anlagen.''
+%
+%
+% \StopEventually{\PrintChanges\PrintIndex}
+%
+% \iffalse
+%<*class>
+
+\NeedsTeXFormat{LaTeX2e}
+\ProvidesClass{cjw-ltr}
+ [\filedate\space v\fileversion\space
+ Modified Letter Class for LaTeX2e---CJW]
+
+% \fi
+%
+% \section{The Code}
+%
+% The very first thing to do is to include the file |letter.cls|,
+% since most of what's in there is very good, and I don't want to
+% have to rewrite it and bloat this file. We make sure as well that
+% it inherits any options.
+% \begin{macrocode}
+\DeclareOption*{\PassOptionsToClass{\CurrentOption}{letter}}
+\ProcessOptions
+
+\LoadClass{letter}
+% \end{macrocode}
+% Now we can start to change things. I think the default \cseq{headsep}
+% is too large; so, we will make it a bit smaller.
+% \begin{macrocode}
+\setlength\headsep {25\p@}
+% \end{macrocode}
+% Next, we provide the length that will be used for vertical space
+% within a letter. We initialise it to a reasonable value.
+% \begin{macrocode}
+\newlength{\letterskip}
+ \setlength{\letterskip}{0.7em}
+% \end{macrocode}
+%
+% \subsection{Standard Letters}
+%
+% \begin{macro}{\addresswidth}
+% We first declare the length \cseq{addresswidth}, which is used to
+% determine how far from the right margin the sender's adress and the
+% closing should be set. Since the \pkg{letter} class uses the
+% \cseq{indentedwidth} length to help place the sender's address, we
+% will simply recycle that length for our own purposes.
+% \begin{macrocode}
+\let\addresswidth\indentedwidth
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\faxlabel}
+% \begin{macro}{\tellabel}
+% If a fax number is included, it should be labeled so as not to be
+% confused with a telephone number. The labels (appropriate to both
+% German and English) will simply be `Tel:' and `Fax:'. But, in the
+% interest of \ae{}sthetics, we want the printed versions to have the
+% same width, so that the actual numbers line up nicely. Since `Fax:'
+% is clearly the wider label, we use it to determine the width we
+% want.
+% \begin{macrocode}
+\newlength\lblwd
+{\setbox0=\hbox{Fax:\enskip}\global\setlength{\lblwd}{\wd0}}
+\newcommand*{\faxlabel}{\hbox to\lblwd{Fax:\enskip\hfil}}
+\newcommand*{\tellabel}{\hbox to\lblwd{Tel:\enskip\hfil}}
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+
+% \begin{macro}{\fax}
+% \begin{macro}{\check@iffax}
+% The \cseq{fax} command simply sets a global placeholder.
+% \begin{macrocode}
+\newcommand*{\fax}[1]{\def\faxnum{#1}}
+ \fax{}
+% \end{macrocode}
+% The \cseq{check@iffax} macro inserts the labels if appropriate.
+% \begin{macrocode}
+\def\check@iffax{%
+ \ifx\@empty\faxnum\else
+ \edef\faxnum{\protect\faxlabel\faxnum}
+ \ifx\@empty\telephonenum\else
+ \edef\telephonenum{\protect\tellabel\telephonenum}
+ \fi\fi}
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\opening}
+% The \cseq{opening} macro will collect all information for the
+% \env{letter} environment and begin the typesetting. First, we
+% calculate the actual width of the given address.
+% \begin{macrocode}
+\def\opening#1{%
+% \end{macrocode}
+% The first step is to determine what actual address information we
+% have. We will figure this out and put it into a placeholder. If a
+% fax number is included, we will want to use labels for telephone and
+% fax.
+% \begin{macrocode}
+ \check@iffax
+ \def\from@address{%
+% \end{macrocode}
+% We see if a return address was supplied. If not, we simply set
+% the date.
+% \begin{macrocode}
+ \ifx\@empty\fromaddress%
+ \@date%
+% \end{macrocode}
+% If so, we typeset the address including a telephone number if one is
+% supplied. Then comes a bit of whitespace and the date.
+% \begin{macrocode}
+ \else% home address
+ \fromaddress%
+ \ifx\@empty\telephonenum\else%
+ \\ \telephonenum%
+ \fi%
+ \ifx\@empty\faxnum\else%
+ \\ \faxnum%
+ \fi%
+ \\*[\letterskip]\@date%
+ \fi}
+ \setbox0=\hbox{%
+ \begin{tabular}{l}\from@address\end{tabular}}
+% \end{macrocode}
+% With these values we set out global value for \cseq{addresswidth}.
+% \begin{macrocode}
+ \global\setlength{\addresswidth}{\wd0}
+% \end{macrocode}
+% With no address, we want the |firstpage| pagestyle.
+% \begin{macrocode}
+ \ifx\@empty\fromaddress
+ \thispagestyle{firstpage}%
+% \end{macrocode}
+% With one, we want the |empty| pagestyle for the first page.
+% \begin{macrocode}
+ \else
+ \thispagestyle{empty}
+ \fi
+% \end{macrocode}
+% In either case, we fill horizontally and dump the box with the
+% address information.
+% \begin{macrocode}
+ \noindent\hfill\box0
+ \par
+% \end{macrocode}
+% After the date comes some more whitespace and the recipient---
+% \begin{macrocode}
+ \vspace{2\letterskip}%
+ {\raggedright\toname\\ \toaddress\par}%
+% \end{macrocode}
+% ---then more whitespace and the text of the greeting. Note that all
+% of the whitespace is given in terms of \cseq{letterskip}.
+% \begin{macrocode}
+ \vspace{2\letterskip}%
+ \noindent#1%
+% \end{macrocode}
+% Finally, we reset the values of \cseq{parindent} and \cseq{parskip} and
+% inhibit a break before the body. (What set of addresses would make
+% \TeX{} \emph{want} to break after the greeting?)
+% \begin{macrocode}
+ \setlength{\parindent}{20\p@}
+ \setlength{\parskip}{0\p@ \@plus\@ne\p@\relax}
+ \par\nobreak}
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\closing}
+% The \cseq{closing} macro is unchanged from the \pkg{letter} class,
+% except for the change from \cseq{parindent} to \cseq{letterskip} and
+% the use of \cseq{addresswidth} in determining the box size.
+% \begin{macrocode}
+\long\def\closing#1{\par\nobreak\vspace{2\letterskip}%
+ \stopbreaks
+ \noindent\hfill
+ \parbox{\addresswidth}{\raggedright%
+ #1\\[6\letterskip]%
+ \ifx\@empty\fromsig
+ \fromname
+ \else
+ \fromsig
+ \fi\strut}
+ \par}
+% \end{macrocode}
+% \end{macro}
+%
+% \subsection{German Letters}
+%
+% First, we \cseq{let} several appropriate macros for German usage.
+% \begin{macrocode}
+\let\unterschrift \signature
+\let\absender \address
+\let\ort \location
+\let\telefon \telephone
+\let\datum \date
+% \end{macrocode}
+% \begin{macro}{\heute}
+% Here are some simple definitions which I use in my own macro files.
+% \begin{macrocode}
+\newcommand{\theday}{\number\day\relax}
+\newcommand{\themonth}{%
+ \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}
+\newcommand{\themonat}{%
+ \ifcase\month\or Januar\or Februar\or%
+ M\"arz\or April\or Mai\or Juni\or Juli\or August\or%
+ September\or Oktober\or November\or Dezember\fi}
+\newcommand{\theyear}{\number\year\relax}
+% \end{macrocode}
+% From these we can construct a German date macro, \cseq{heute}.
+% \begin{macrocode}
+\newcommand{\heute}{den~\theday.\ \themonat\ \theyear}
+% \end{macrocode}
+% These are also useful if one prefers a different (\ie date first)
+% format for the \cseq{today} macro.
+% \end{macro}
+%
+% \begin{macro}{\betreff}
+% \begin{macro}{\bezug}
+% Next come the macros for the additional information which may be
+% included in the headers of a German letter.
+% \changes{v0.5}{1995/06/10}{Fixed a bug (incorrect macro parameter)
+% in \cseq{betreff} and \cseq{bezug}.}
+% \begin{macrocode}
+\newcommand{\betreff}[2][Betreff]{%
+ \def\betreffname{#1}\long\def\@betreff{#2}}
+\newcommand{\bezug}[2][Bezug]{%
+ \def\bezugname{#1}\long\def\@bezug{#2}}
+% \end{macrocode}
+% These are initialised to be empty.
+% \begin{macrocode}
+\betreff[]{}
+\bezug[]{}
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+%
+% \begin{environment}{brief}
+% Now comes the \env{brief} environment itself. German letters are
+% suppose to use unindented and separated paragraphs, so we will set
+% the proper values as soon as the environment is entered.
+% \begin{macrocode}
+\long\def\brief#1{%
+ \newpage
+ \setlength{\parindent}{0pt}
+ \setlength{\parskip}{0.7em}
+% \end{macrocode}
+% Further initialisations are taken from the \cseq{letter} macro.
+% \begin{macrocode}
+ \if@twoside \ifodd\c@page
+ \else\thispagestyle{empty} \hbox{}\newpage\fi
+ \fi
+ \c@page\@ne
+ \interlinepenalty=200 % Smaller than the TeXbook value
+% \end{macrocode}
+% Since the supplied argument has more information in it than that
+% given to a \env{letter} environment, we must have our own equivalent
+% to the \pkg{letter} class's |\@processto|.
+% \begin{macrocode}
+ \@prozessan{\leavevmode\ignorespaces #1}
+% \end{macrocode}
+% Lastly, we make the default date a German one.
+% \begin{macrocode}
+ \date{\heute}}
+% \end{macrocode}
+% \begin{macro}{\@prozessan}
+% \changes{v0.5}{1995/06/10}{The \cseq{@prozessan} macro and its
+% subprocesses completely rewritten, as the previous version (v0.1)
+% was very broken.}
+% The address processing is based on the |\@processto| macro from
+% \pkg{letter.cls}. We call the first subprocess, |\@xproz|, to fetch
+% the PLZ part of the address.
+% \begin{macrocode}
+\long\def\@prozessan#1{%
+ \@xproz #1!!@@@%
+% \end{macrocode}
+% If there is a PLZ, the first test includes an extra occurrence of
+% |!!| in the \cseq{toPLZ} macro. So, we call |\@xproz| again without the
+% extra characters (which were necesary to make sure that we didn't
+% get a `usage doesn't match definition' error the first time around).
+% \begin{macrocode}
+ \ifx\@empty\toPLZ\else\@xproz #1@@@\fi%
+% \end{macrocode}
+% The |\@xproz| macro has now stored all the address parts before the
+% PLZ in |\@tempa|. Now there are up to three different parts that
+% may be in |\@tempa| which we will try to separate out. We use
+% \cseq{expandafter} in order that the tokens within |\@tempa| get
+% properly considered.
+% \begin{macrocode}
+ \expandafter\@yproz \@tempa\\@@@%
+% \end{macrocode}
+% The |\@yproz| splits the input into |\@tempb|, the part before the
+% first linebreak, and |\@tempc|, the rest. If the second part, that
+% in |\@tempc|, is empty, then we want to set the third part,
+% |\@tempd|, to be empty, too.
+% \begin{macrocode}
+ \ifx\@empty\@tempc
+ \let\@tempd\@empty
+% \end{macrocode}
+% Otherwise, we call |\@zproz| to further subdivide the address input.
+% \begin{macrocode}
+ \else
+ \expandafter\@zproz \@tempa\\@@@%
+% \end{macrocode}
+% If there is a third part, |\@tempd|, we have the same problem with
+% extra characters as above in \cseq{toPLZ}. We call |\@zproz| one more
+% time in order to read |\@tempd| correctly.
+% \begin{macrocode}
+ \ifx\@empty\@tempd
+ \else
+ \expandafter\@zproz \@tempa{}@@@%
+ \fi
+ \fi
+% \end{macrocode}
+% Now we need to decide what to call whichever address pieces actually
+% exist. Obviously, if everything back to |\@tempb| was empty, all of
+% the address parts should be likewise.
+% \begin{macrocode}
+ \ifx\@empty\@tempb
+ \let\totitle\@empty \let\toname\@empty \let\toaddress\@empty
+ \else
+% \end{macrocode}
+% Now we have at least one address part. If that is the only one, it
+% is most certainly not a title, so \cseq{totitle} must be set empty.
+% \begin{macrocode}
+ \ifx\@empty\@tempc
+ \let\totitle\@empty
+% \end{macrocode}
+% One line with a PLZ is almost assuredly an address proper. Without
+% a PLZ it is probably a name.
+% \begin{macrocode}
+ \ifx\@empty\toPLZ
+ \let\toname\@tempb \let\toaddress\@empty
+ \else
+ \let\toname\@empty \let\toaddress\@tempb
+ \fi
+% \end{macrocode}
+% If |\@tempb| and |\@tempc| contain something, but |\@tempd| is
+% empty, the two lines should be name and address.
+% \begin{macrocode}
+ \else
+ \ifx\@empty\@tempd
+ \let\totitle\@empty \let\toname\@tempb \let\toaddress\@tempc
+% \end{macrocode}
+% The last option is that title, name and address are all present.
+% \begin{macrocode}
+ \else
+ \let\totitle\@tempb \let\toname\@tempc \let\toaddress\@tempd
+ \fi
+ \fi
+ \fi}
+% \end{macrocode}
+% This is my second implementation of the |\@prozessan| macro. The
+% first was truly ugly and had the added disadvantage of, well, not
+% working. In fact, it failed miserably for a few relatively simple
+% cases. That the current version (as far I can tell) works does not,
+% however, excuse the fact that it is still somewhat ugly and seems to
+% me to be very efficient. I am still open to any suggested
+% improvements.
+%
+% The subproceses to |\@prozessan| are simply designed to be able to
+% make assignments based on the structure of the passed argument.
+% \begin{macrocode}
+\long\def\@xproz #1!!#2@@@{\def\@tempa{#1}\def\toPLZ{#2}}
+\long\def\@yproz #1\\#2@@@{\def\@tempb{#1}\def\@tempc{#2}\def\@tempd{}}
+\long\def\@zproz #1\\#2\\#3@@@{\def\@tempc{#2}\def\@tempd{#3}}
+% \end{macrocode}
+% \end{macro}
+%
+% The end of the \env{brief} environment is identical to that of the
+% \env{letter} environment.
+% \begin{macrocode}
+\let\endbrief\endletter
+% \end{macrocode}
+% \end{environment}
+%
+% \begin{macro}{\anrede}
+% As for the \cseq{opening} macro in the \env{letter} environment, the
+% \cseq{anrede} macro signals the beginning of the letter. We begin by
+% setting the dateline.
+% \changes{v0.5}{1995/06/10}{Use \cseq{@dateline} instead of
+% \cseq{@tempa}, because \cseq{raggedright}, or \emph{something},
+% seemed to mess with \cseq{@tempa}. Also move the conditional
+% outside of the definition of \cseq{@dateline}.}
+% \begin{macrocode}
+\def\anrede#1{%
+ \check@iffax
+ \ifx\@empty\fromlocation
+ \def\@dateline{\@date}
+ \else
+ \def\@dateline{\fromlocation,\space\@date}
+ \fi
+% \end{macrocode}
+% Set with a typewriter, a German letter should leave two blank lines
+% before the return address. Since we are using something a little
+% more capable than a typewriter, we will leave a small amount of
+% whitespace, in the form of a multiple of \cseq{letterskip}.
+% \begin{macrocode}
+ \null\vspace{1.2\letterskip}
+ \noindent\ignorespaces
+% \end{macrocode}
+% If there is no return address (a situation which I don't actually
+% see happening with this format), just set the dateline at the right
+% margin.
+% \begin{macrocode}
+ \ifx\@empty\fromaddress%
+ \pagestyle{firstpage}%
+ {\raggedleft \@dateline}%
+% \end{macrocode}
+% Otherwise, the address is set at the left margin and the dateline
+% right justified at the level of the first line of the sender
+% information. A telephone number and fax, if provided, follow the address.
+% \changes{v0.5}{1995/06/10}{Added code to include a telephone number
+% in the return address information.}
+% \begin{macrocode}
+ \else % Home address
+ \pagestyle{empty}%
+ {\raggedright%
+ \fromname \hfill \@dateline\\
+ \fromaddress}
+ \ifx\@empty\telephonenum
+ \else
+ \\ \telephonenum
+ \fi
+ \ifx\@empty\faxnum
+ \else
+ \\ \faxnum
+ \fi
+ \fi
+ \par
+% \end{macrocode}
+% Next comes whitespace approximately twice as large as the topskip.
+% \begin{macrocode}
+ \vspace{2.2\letterskip}
+% \end{macrocode}
+% We then set the recipient's address. Using the information provided
+% from the |\@prozessan| macro above, we include those pieces which
+% were supplied.
+% \begin{macrocode}
+ {\raggedright
+ \ifx\@empty\totitle
+ \else
+ \totitle \\
+ \fi
+ \toname
+ \ifx\@empty\toaddress
+ \else
+ \\ \toaddress
+ \fi
+ \ifx\@empty\toPLZ
+ \else
+ \\*[0.6\letterskip]\toPLZ
+ \fi
+ \par}
+% \end{macrocode}
+% After the address come, if supplied, the \emph{Betreff} and
+% \emph{Bezug} lines. First, though, we check to see if only one is
+% supplied---if so, we issue a warning.
+% \begin{macrocode}
+ \@checkbetreffbezug
+% \end{macrocode}
+% Now we can typeset those lines. A small amount of whitespace
+% separates them from one another and from the address.
+% \begin{macrocode}
+ \ifx\@empty\@betreff
+ \else
+ \vspace{\letterskip}
+ \ifx\@empty\betreffname
+ \else
+ \betreffname:\quad
+ \fi
+ \@betreff\par
+ \fi
+ \ifx\@empty\@bezug
+ \else
+ \vspace{1\letterskip}
+ \ifx\@empty\bezugname
+ \else
+ \bezugname:\quad
+ \fi
+ \@bezug\par
+ \fi
+% \end{macrocode}
+% A slightly larger amount of whitespace precedes the actual
+% \emph{Anrede}.
+% \begin{macrocode}
+ \vspace{1.4\letterskip}%
+ #1\par\nobreak}
+% \end{macrocode}
+% \end{macro}
+% The |\@checkbetreffbezug| macro emits a warning if only one of those
+% two lines is supplied.
+% \begin{macrocode}
+\def\@checkbetreffbezug{%
+ \ifx\@empty\@betreff
+ \ifx\@empty\@bezug
+ \else
+ \ClassWarningNoLine{cjw-ltr}{%
+ Die \protect\betreff und \protect\bezug Befehlen sollen nur
+ \MessageBreak
+ zusammen verwendet werden. Ihnen fehlt die \MessageBreak
+ Betreff-Zeile.}
+ \fi
+ \else
+ \ifx\@empty\@bezug
+ \ClassWarningNoLine{cjw-ltr}{%
+ Die \protect\betreff und \protect\bezug Befehlen sollen nur
+ \MessageBreak
+ zusammen verwendet werden. Ihnen fehlt die \MessageBreak
+ Bezug-Zeile.}
+ \fi
+ \fi}
+% \end{macrocode}
+%
+% \begin{macro}{\gruss}
+% The command \cseq{gruss} ends a letter by first adding some whitespace
+% and disabling further pagebreaks.
+% \begin{macrocode}
+\long\def\gruss#1{\par\nobreak\vspace{2\letterskip}%
+ \stopbreaks
+ \noindent%
+% \end{macrocode}
+% The closing itself is set, leaving enough whitespace for the actual
+% signature.
+% \begin{macrocode}
+ {\raggedright #1 \\[5\letterskip]
+ \ifx\@empty\fromsig
+ \fromname
+ \else
+ \fromsig
+ \fi}
+ \par
+% \end{macrocode}
+% Finally, a bit of white space separates the bottom of the closing
+% from any postscripts.
+% \begin{macrocode}
+ \vspace{\letterskip}}
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\anlage}
+% Enclosures, \emph{Anlagen}, are specified by the \cseq{anlage} command.
+% As with \cseq{betreff} and \cseq{bezug}, the label is given by an optional
+% argument.
+% \begin{macrocode}
+\newcommand{\anlage}[1][Anlage]{%
+% \end{macrocode}
+% If the optional argument is simply `n', we pluralize the label.
+% \begin{macrocode}
+ \def\@tempa{n} \def\@tempb{#1}
+ \ifx\@tempa\@tempb
+ \renewcommand{\enclname}{Anlagen}
+% \end{macrocode}
+% Otherwise, the optional argument (or the default, if none is given)
+% becomes the label.
+% \begin{macrocode}
+ \else
+ \renewcommand{\enclname}{\@tempb}
+ \fi\encl}
+% \end{macrocode}
+% \end{macro}
+% And with that, we are done.
+%
+% \iffalse
+%</class>
+% \fi
+%
+% \Finale
+%
diff --git a/Master/texmf-dist/source/latex/cjw/cjw-ltr.ins b/Master/texmf-dist/source/latex/cjw/cjw-ltr.ins
new file mode 100644
index 00000000000..4a33db2f3cf
--- /dev/null
+++ b/Master/texmf-dist/source/latex/cjw/cjw-ltr.ins
@@ -0,0 +1,18 @@
+\def\batchfile{cjw-ltr.ins}
+\input docstrip.tex
+
+\preamble
+
+Copyright (C) 1995 by Colin J. Wynne
+
+This file is part of cjw-ltr, a class for creating modified standard
+letters as well as German style letters under LaTeX2e.
+
+\endpreamble
+
+% Don't display progress
+\keepsilent
+
+\generateFile{cjw-ltr.cls}{t}{\from{cjw-ltr.dtx}{class}}
+
+\endinput
diff --git a/Master/texmf-dist/source/latex/cjw/cjwoutl.dtx b/Master/texmf-dist/source/latex/cjw/cjwoutl.dtx
new file mode 100644
index 00000000000..c77c01ad0be
--- /dev/null
+++ b/Master/texmf-dist/source/latex/cjw/cjwoutl.dtx
@@ -0,0 +1,656 @@
+% \iffalse -*- LaTeX -*-
+%
+% This is file `cjwoutl.dtx'. You can run this file through LaTeX2e
+% to produce the .dvi documentation. You should also have received
+% the file `cjwoutl.ins'. Run it through LaTeX2e to extract the
+% actual package file.
+%
+% \fi
+\def\fileversion{0.6}
+\def\filedate{1996/04/26}
+%% \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 \~}
+%%
+% \CheckSum{327}
+%
+% \iffalse
+%
+%<*driver>
+
+\NeedsTeXFormat{LaTeX2e}
+\ProvidesFile{cjwoutl.dtx}
+ [\filedate\space v\fileversion\space
+ Auto-Numbered Outlines--CJW]
+
+\documentclass{ltxdoc}
+\IfFileExists{cjwoutl.sty}
+ {\usepackage{cjwoutl}}
+ {\GenericError{(cjwoutl.dtx)\@spaces\@spaces\@spaces}{Error:
+ You must first LaTeX the file `cjwoutl.ins' \MessageBreak
+ and extract the `cjwoutl.sty' package before \MessageBreak
+ processing the documentation}{}{}}
+\GetFileInfo{cjwoutl.sty}
+\EnableCrossrefs
+% \DisableCrossrefs % Say \DisableCrossrefs if index is ready
+% \OnlyDescription % comment out for implementation details
+\RecordChanges
+
+\begin{document}
+ \DocInput{cjwoutl.dtx}
+\end{document}
+%</driver>
+% \fi
+%
+% \def\pkg#1{\textsf{#1}}
+% \def\env#1{\textsf{#1}}
+% \providecommand{\ie}{\emph{i.e.},\ }
+%
+% ^^A % Change List:
+%
+% \changes{v0.5}{1995/05/13}{Added \texttt{\protect\bslash outlindent} as
+% indentation size.}
+%
+% \title{The \pkg{cjwoutl} Package\thanks{%
+% This file has version \fileversion{} as of \filedate.}}
+% \author{Colin J.~Wynne\thanks{E-Mail at:
+% \texttt{cwynne@brutus.mts.jhu.edu},
+% \texttt{cwynne@jhu.edu}\,.}}
+% \date{\filedate}
+%
+% \maketitle
+%
+% \setcounter{StandardModuleDepth}{1}
+%
+% {\parskip 0pt ^^A % This is the hack used by |doc.dtx|.
+% ^^A % (bug in \LaTeX?)
+% \tableofcontents
+% }
+%
+% \section*{Introduction}
+%
+% This package originated as a simple set of macros for doing outlines
+% under \PlainTeX. The original macros simply provided
+% $n$|\parindent| as a left indentation for a single paragraph.
+% This was almost three years ago. The macros improved somewhat with
+% my knowledge of (and experience with) \TeX. Last summer I finally
+% had the time to add what I had always wanted---automatic level
+% numbering.
+%
+% This year I decided to find out just what \LaTeX{} was all about.
+% After deciding that it wasn't that bad, I rewrote (and improved yet
+% again) the outlining macros. Conveniently enough, a large chunk of
+% the original \PlainTeX{} code dedicated to resetting the proper
+% counters disappeared with the change to \LaTeX.
+%
+% At a basic level, the \env{outline} environment functions like
+% \env{enumeration}. However, one needn't nest the calls of the
+% \env{outline} environment. Rather, the macros used to identify the
+% entries of the outline itself take care of this.
+%
+% I think this code is now both fairly generic and rather useful, and
+% I would be more than happy to know that it is useful to others.
+% Feel free to e-mail me with comments, suggestions, etc. The
+% \PlainTeX{} version is still available by request for anyone who
+% might want it.
+%
+%
+% \section{Making an Outline}
+%
+% \DescribeEnv{outline}
+% An outlined portion of a document is implemented, naturally enough,
+% inside an environment, the \env{outline} environment. There is one
+% optional argument to the |\begin| command. If the argument |new| is
+% given, then the outline counters will be reset. Otherwise, the
+% values of the counters will be exactly as they were at the end of
+% the last \env{outline} environment.
+%
+% \DescribeMacro{\outl}
+% Inside of an \env{outline} environment, there are two main user
+% macros, one for auto-labelled entries and one for user-supplied
+% labels. For labels to be automatically numbered and printed, the
+% user simply enters |\outl|\meta{level}. The \meta{level} is simply
+% an integer which corresponds to the depth within the outline. For
+% example, level one of an outline begins at the left margin (more
+% precisely, the \emph{text} begins at the left margin with the label
+% offset by 1\,en to the left of that) and in a standard outline, and,
+% indeed, according to this package's defaults, a capitalised roman
+% numeral is used as a label. Therefore
+% \begin{verbatim}
+% \begin{outline}
+% \outl{1}This is outline level one.
+% \end{outline}
+% \end{verbatim}
+% will produce
+% \begin{outline}
+% \outl{1}This is outline level one.
+% \end{outline}
+% The package supplies seven defined levels of outline as well as a
+% default `level'. The default is used for any levels over seven.
+% These levels will still receive the correct indentation
+% \DescribeMacro{\outlindent}
+% (specifically, level $n$ is prefaced by |\leftskip = |^^A
+% $(n-1)$|\outlindent| worth of space), but will all receive the
+% default label, normally a bullet. Thus, the user is provided with
+% the following:
+% \begin{outline}[new]
+% \outl{1}\dots
+% \outl{2}\dots
+% \outl{3}\dots
+% \outl{4}\dots
+% \outl{5}\dots
+% \outl{6}\dots
+% \outl{7}\dots
+% \outl{8}\dots
+% \outl{9}\dots
+% \outl{10}\dots
+% \end{outline}
+% Note that levels eight through ten have the default bullet for a
+% label, though the indentation levels are preserved. There are a few
+% special symbols which can be given as the \meta{level} argument. If
+% the user enters a |+|, |-| or |=| then the indentation level of that
+% entry will be respectively one greater, one less or the same as that
+% of the previous entry. Thus,
+% \begin{verbatim}
+% \begin{outline}
+% \outl{1} Level one,
+% \outl{+} Level two,
+% \outl{=} Level two,
+% \outl{-} Level one.
+% \end{outline}
+% \end{verbatim}
+% gives
+% \begin{outline}[new]
+% \outl{1} Level one,
+% \outl{+} Level two,
+% \outl{=} Level two,
+% \outl{-} Level one.
+% \end{outline}
+%
+% \DescribeMacro{\ol}
+% The macro |\ol| is almost identical to the |\outl| macro, except
+% that it takes a second argument: |\ol|\meta{level}\meta{label}. The
+% \meta{level} argument functions identically, and the \meta{label}
+% argument is set as entered 1\,en to the left of the text.
+% \DescribeMacro{\npp}\DescribeMacro{\unpp} Two macros are derived from
+% |\ol|. Both take a \meta{level} argument. The |\npp| macro is
+% essentially |\ol|\meta{level}|{\null}\indent|. The |\unpp| macro
+% does not indent the first line of text. (The macros stand for ``new
+% paragraph'' and ``unindented |\npp|''.)
+%
+% \DescribeMacro{\outlstyle}
+% There is also a macro to allow for some modifications in how an
+% outline entry is typeset. The default style, |plain|, does nothing.
+% \DescribeMacro{firstpar}
+% \DescribeMacro{\firstparfont}
+% An alternate style, called |firstpar|, sets the first paragraph of
+% an outline entry in |\firstparfont|. By default, this is italics,
+% but the font command can, of course, be renewed by the user. To
+% change styles, one simply calls |\outlstyle|\meta{style}; for
+% example,
+% \begin{verbatim}
+% \outlstyle{firstpar}
+% \end{verbatim}
+% calls the |firstpar| outline style.
+%
+% N.B.: There is one `secret' left to tell. Unless one wants to reset
+% the counters by calling |\begin{outline}[new]|, it is really
+% unnecessary to call the environment. Instead, one can simply use
+% the |\ol| or |\outl| commands as desired. The only caveat is that
+% the final outline grouping must be closed. Therefore, the
+% |\endoutline| macro \emph{must} be called. Granted, this doesn't
+% make a whole big difference. I happen to use this `feature' simply
+% because I have used these macros since long before it was written in
+% a \LaTeX-environment format. YMMV.
+%
+% \subsection{Outline Labels}
+%
+% \DescribeMacro{outlN}
+% The counters for the different levels are named |outl|$n$, where $n$
+% is a lowercase roman numeral. Therefore, this package defines
+% |outli|, |outlii|,\ldots,|outlvii|. This is, as one easily sees,
+% similar to the numbering scheme for the \env{enumeration}
+% environment.
+% \DescribeMacro{\theoutlN}
+% \DescribeMacro{\labeloutlN}
+% The numbering style for level $n$ is given by the macro |\theoutlN|.
+% The actual label which is printed for an entry at level $n$ is given
+% (naturally enough) by |\labeloutlN|. The default label is
+% |\labeloutldef|. The example above showed the label at the sixth
+% level as `(a)'. The relevant definitions are:
+% \begin{verbatim}
+% \renewcommand{\theoutlvi}{\alph{outlvi}}
+% \renewcommand{\labeloutlvi}{(\theoutlvi)}
+% \end{verbatim}
+% \DescribeMacro{\labelfont}
+% The labels themselves are typeset in |\labelfont|, which can also be
+% set by the user. The default is bold-extended.
+%
+% \changes{v0.6}{1996/04/26}{Added info about cross-referencing.}
+% \subsection{Cross References}
+%
+% Cross references function as expected for outlines, using the
+% |\label| and |\ref| commands. The |\outl| function defines the
+% current reference string, so the |\label| should be put after the
+% |\outl| to which it refers. For example, we might have the code
+% \begin{verbatim}
+% \begin{outline}[new]
+% \outl{1} Main entry.
+% \outl{+} Next level.
+% \outl{+} The price of tea in China.\label{ol:tea}
+% \end{outline}
+% \end{verbatim}
+% producing
+% \begin{outline}[new]
+% \outl{1} Main entry.
+% \outl{+} Next level.
+% \outl{+} The price of tea in China.\label{ol:tea}
+% \end{outline}
+% whereupon the reference |ol:tea| would refer to tea in
+% China~\ref{ol:tea}.
+%
+% \StopEventually{\PrintChanges\PrintIndex}
+%
+%
+% \section{The Code}
+%
+% \iffalse
+%<*package>
+
+\NeedsTeXFormat{LaTeX2e}
+\ProvidesPackage{cjwoutl}
+ [1995/04/24 v0.5 Auto-Numbered Outlines--CJW]
+
+\DeclareOption*{%
+ \PackageWarning{cjwoutl}{Unknown option `\CurrentOption'}}
+
+\ProcessOptions
+
+% \fi
+%
+% \subsection{Allocations}
+%
+% Since the levels of an outline are defined by a |\leftskip| command,
+% it is necessary to determine when a group needs to be closed.
+% Therefore, we define a test to see if we are in a group.
+% \begin{macrocode}
+\newif\ifoutl@group
+% \end{macrocode}
+% This essentially keeps the first outline entry from closing a
+% non-existent group. Two lengths, |\aboveoutlskip| and
+% |\belowoutlskip|, initially set to the equivalent values for math
+% displays, are used to separate an \env{outline} environment from the
+% surrounding text.
+% \changes{v0.6}{1996/04/26}{Added \texttt{\protect\bslash aboveoutlskip}
+% and \texttt{\protect\bslash belowoutlskip} lengths based on values for
+% displays.}
+% \begin{macrocode}
+\newlength{\aboveoutlskip}
+ \setlength{\aboveoutlskip}{\abovedisplayskip}
+\newlength{\belowoutlskip}
+ \setlength{\belowoutlskip}{\belowdisplayskip}
+% \end{macrocode}
+% The |\outlindent| macro controls the increment between adjacent
+% outline levels. Its default value is |\parindent|.
+% \begin{macrocode}
+\newlength{\outlindent}
+ \setlength{\outlindent}{\parindent}
+% \end{macrocode}
+%
+% Now we define the counters to be used. The first two identify the
+% requested indentation level and the previous level.
+% \begin{macrocode}
+\newcounter {outl@indsize}
+\newcounter {outl@lastind}
+% \end{macrocode}
+% \begin{macro}{outlN}
+% Next come the counters for the levels themselves.
+% \begin{macrocode}
+\newcounter {outli} % I.
+\newcounter {outlii} [outli] % A.
+\newcounter {outliii}[outlii] % 1.
+\newcounter {outliv} [outliii] % a.
+\newcounter {outlv} [outliv] % i.
+\newcounter {outlvi} [outlv] % (a)
+\newcounter {outlvii}[outlvi] % (i)
+% \end{macrocode}
+% There is also a counter definition for the default label. This will
+% be explained below.
+% \begin{macrocode}
+\newcounter {outldef}[outlvii]
+% \end{macrocode}
+% \end{macro}
+% \begin{macro}{\theoutlN}
+% Now we can set the proper level numbering.
+% \begin{macrocode}
+\renewcommand{\theoutli} {\Roman{outli}}
+\renewcommand{\theoutlii} {\Alph{outlii}}
+\renewcommand{\theoutliii} {\arabic{outliii}}
+\renewcommand{\theoutliv} {\alph{outliv}}
+\renewcommand{\theoutlv} {\roman{outlv}}
+\renewcommand{\theoutlvi} {\alph{outlvi}}
+\renewcommand{\theoutlvii} {\roman{outlvii}}
+% \end{macrocode}
+% Again we have the default label, which I \emph{promise} will be
+% explained.
+% \begin{macrocode}
+\renewcommand{\theoutldef} {}
+% \end{macrocode}
+% \end{macro}
+% \begin{macro}{\labeloutlN}
+% Now the label formats will be defined.
+% \begin{macrocode}
+\newcommand{\labeloutli} {\theoutli.}
+\newcommand{\labeloutlii} {\theoutlii.}
+\newcommand{\labeloutliii} {\theoutliii.}
+\newcommand{\labeloutliv} {\theoutliv.}
+\newcommand{\labeloutlv} {\theoutlv.}
+\newcommand{\labeloutlvi} {(\theoutlvi)}
+\newcommand{\labeloutlvii} {(\theoutlvii)}
+\newcommand{\labeloutldef} {\ensuremath{\bullet}}
+% \end{macrocode}
+% \end{macro}
+% This next group of definitions are taken from the standard classes'
+% definitions for the \env{enumeration} environment. I have not
+% plumbed the cross-referencing code deeply enough to see how these
+% macros are applied, but I take it on faith from |classes.dtx| that
+% it works.
+% \changes{v0.6}{1996/04/26}{Tweaked \texttt{\protect\bslash p@}\ldots{}
+% definitions based on information in \emph{The \LaTeX{} Companion}.}
+% \begin{macrocode}
+\renewcommand{\p@outli} {}
+\renewcommand{\p@outlii} {\theoutli.}
+\renewcommand{\p@outliii} {\p@outlii\theoutlii.}
+\renewcommand{\p@outliv} {\p@outliii\theoutliii.}
+\renewcommand{\p@outlv} {\p@outliv\theoutliv.}
+\renewcommand{\p@outlvi} {\p@outlv\labeloutlv.}
+\renewcommand{\p@outlvii} {\p@outlvi\labeloutlvi.}
+% \end{macrocode}
+% And now we have the promised explanation of having a counter for the
+% default labels. Simply put, the |\p@|\ldots{} commands are defined
+% \emph{via} |\setcounter|. To be able to supply cross-references for
+% outline entries above the seven numbered levels it seems necessary
+% to have the relevant counter definition already made. I could be
+% wrong on this, and maybe I could have defined the cross-reference
+% label directly. If so, let me know. Anyway, that was the
+% explanation---probably not worth the wait, was it?
+% \begin{macrocode}
+\renewcommand{\p@outldef} {\p@outlvii\ldots}
+% \end{macrocode}
+%
+% \subsection{The Environment}
+%
+% \begin{environment}{outline}
+% The \env{outline} environment itself has a very simple definition.
+% It has a single optional argument. The default value of the
+% argument is fairly arbitrary, so long as it isn't |new|.
+% \changes{v0.6}{1996/04/26}{Added the new skip values to the beginning
+% and end macros for the environment.}
+% \begin{macrocode}
+\newenvironment{outline}[1][keep]%
+% \end{macrocode}
+% To start the environment, we must simply test to see if the counters
+% should be reset or not.
+% \begin{macrocode}
+ {\def\@tempa{#1} \def\@tempb{new}
+ \ifx\@tempa\@tempb
+ \outl@reset
+ \fi
+ \penalty\predisplaypenalty\vspace{\aboveoutlskip}}%
+% \end{macrocode}
+% To finish an outlined section of a document, we include a |\par| (to
+% make sure the |\leftskip| functions correctly), close the last
+% outline group and let it be known that we are no longer in an
+% outline.
+% \begin{macrocode}
+ {\par\endgroup\global\outl@groupfalse%
+ \penalty\postdisplaypenalty\vspace{\belowoutlskip}}
+% \end{macrocode}
+% The macro to reset the outline counters is about as straightforward
+% as possible.
+% \begin{macrocode}
+\newcommand{\outl@reset}{%
+ \setcounter{outli} {0}
+ \setcounter{outlii} {0}
+ \setcounter{outliii}{0}
+ \setcounter{outliv} {0}
+ \setcounter{outlv} {0}
+ \setcounter{outlvi} {0}
+ \setcounter{outlvii}{0}}
+% \end{macrocode}
+% \end{environment}
+%
+% \subsection{Outline Styles}
+%
+% \begin{macro}{\outlstyle}
+% Outline styles are implemented as a macro call to the desired style
+% just before the text of the outline entry is typeset. \ie in
+% abstract terms, the |\ol| or |\outl| macros will have
+% \begin{tabular}{l}
+% \meta{label}\meta{style-macro}\textit{Text text text\ldots}
+% \end{tabular}
+% The styles are macros named |\ols@|\meta{style}. Thus, the
+% |\outlstyle| macro shecks to see if the style is defined. If not,
+% it emits a warning and selects the plain style.
+% \begin{macrocode}
+\newcommand{\outlstyle}[1]{%
+ \@ifundefined{ols@#1}
+ {\PackageWarning{cjwoutl}{Outline style `#1' is undefined. Using
+ style `plain' instead}%
+ \def\outl@style{\ols@plain}}
+% \end{macrocode}
+% \end{macro}
+% If the requested style exists, we point the macro |\outl@style|
+% towards it. (Note: if anyone can think of a better way to write
+% this, let me know. I wanted to do it as a |\let|, but couldn't
+% figure out the correct sequence of |\expandafter|s to get it to
+% evaluate correctly.)
+% \begin{macrocode}
+ {\def\outl@style{\csname ols@#1\endcsname}}}
+% \end{macrocode}
+% Now we define the two basic styles, |plain| and |firstpar|.
+% \begin{macrocode}
+\def\ols@plain{\@empty}
+% \end{macrocode}
+% The |firstpar| style simply takes the first paragraph of the entry
+% as an argument and sets it in |\firstparfont|.
+% \begin{macrocode}
+\def\ols@firstpar#1\par{%
+ \bgroup\firstparfont #1\par\egroup}
+
+\DeclareOldFontCommand{\firstparfont}{\normalfont\itshape}{\relax}
+% \end{macrocode}
+% Lastly, we select |plain| as the default style.
+% \begin{macrocode}
+\outlstyle{plain}
+% \end{macrocode}
+%
+% \subsection{The Entries}
+%
+% \begin{macro}{\ol}
+% The |\ol| macro first takes care of the outline groupings and then
+% processes the \meta{level} argument given by the user.
+% \begin{macrocode}
+\newcommand{\ol}[2]{%
+ \outl@checkgroups \outl@processlvl{#1}%
+% \end{macrocode}
+% We do some minor error handling---like disallowing negative numbers.
+% If a numerical argument is given, the |\outl@processlvl| macro
+% decrements it by one---thus, the actual |outl@indsize| is not the
+% level number that the user enters, but rather the number of
+% |\outlindent|s for the requested level, and the first entry occurs at
+% zero indents, \ie on the left margin. Thus zero, and not one, is
+% the minimum value which can be assigned to |\outl@indsize|.
+% \begin{macrocode}
+ \ifnum \c@outl@indsize < 1\relax%
+ \setcounter{outl@indsize}{0}%
+ \leftskip = 0pt%
+ \else%
+ \leftskip = \c@outl@indsize\outlindent%
+ \fi%
+% \end{macrocode}
+% We specify that the first paragraph of an entry should not be
+% indented and we make the label.
+% \changes{v0.6}{1996/04/26}{Moved the \texttt{\protect\bslash llap} command
+% into \texttt{\protect\bslash theoutlabel}---otherwise the reference
+% strings were being defined inside groups and being lost.}
+% \begin{macrocode}
+ \noindent\theoutlabel{#2}%
+% \end{macrocode}
+% The |\ignorespaces| macro allows the user to enter |\outl{1} Text|
+% instead of the slightly less readable |\outl{1}Text| in the source
+% file without the extra spaces showing up in the document. Anyway,
+% then we call |\outl@style| which has been defined to be the selected
+% style macro.
+% \begin{macrocode}
+ \ignorespaces\outl@style}
+% \end{macrocode}
+% \end{macro}
+%
+% The |\npp| and |\unpp| macros are straightforward derivatives of the
+% |\ol| macro.
+% \begin{macrocode}
+\newcommand{\npp}[1]{\ol{#1}{\null}\null\indent}
+\newcommand{\unpp}[1]{\ol{#1}{\null}}
+% \end{macrocode}
+%
+% \begin{macro}{\outl}
+% The |\outl| macro is more or less identical to |\ol| above.
+% \begin{macrocode}
+\newcommand{\outl}[1]{%
+ \outl@checkgroups \outl@processlvl{#1}%
+ \ifnum \c@outl@indsize < 1\relax%
+ \setcounter{outl@indsize}{0}%
+ \leftskip = 0pt%
+ \else%
+ \leftskip = \c@outl@indsize\outlindent%
+ \fi%
+% \end{macrocode}
+% The difference is in calling the |\outl@label| macro which takes
+% care of the automatic labels.
+% \begin{macrocode}
+ \noindent\outl@label{\c@outl@indsize}%
+ \ignorespaces\outl@style}
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\outl@checkgroups}
+% Now we define the macro that balances the outline groupings. If we
+% are already in a group, provide a |\par| to get the |\leftskip|
+% right, then close the group and change the group flag.
+% \begin{macrocode}
+\newcommand{\outl@checkgroups}{%
+ \ifoutl@group
+ \par\endgroup\global\outl@groupfalse
+ \smallbreak
+ \fi
+% \end{macrocode}
+% Since we have just finished an outline group, we can set
+% |outl@lastind|, then provide appropriate space (encouraging breaks
+% between enries) and start a new group.
+% \begin{macrocode}
+ \setcounter{outl@lastind}{\c@outl@indsize}
+ \smallbreak
+ \begingroup
+ \global\outl@grouptrue}
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\outl@processlvl}
+% The \meta{level} argument can be a number or one of |+|, |-| and
+% |=|. The check is implemented as a set of nested |\if| statements.
+% \begin{macrocode}
+\newcommand{\outl@processlvl}[1]{%
+ \def\@tempa{+}\def\@tempb{-}\def\@tempc{=}
+ \def\@tempd{#1}
+ \ifx\@tempa\@tempd
+ \addtocounter{outl@indsize}{1}
+ \else\ifx\@tempb\@tempd
+ \addtocounter{outl@indsize}{-1}
+ \else\ifx\@tempc\@tempd
+ % Nothing---stay at same level.
+ \else
+ \setcounter{outl@indsize}{\@tempd}
+% \end{macrocode}
+% If we are setting to a new level number, we decrement by one, as
+% mentioned in the definition of |\ol|, in order to convert the level
+% number to an indentation size.
+% \begin{macrocode}
+ \addtocounter{outl@indsize}{-1}
+ \fi
+ \fi
+ \fi}
+% \end{macrocode}
+% \end{macro}
+%
+% \subsection{Labels}
+%
+% \begin{macro}{\outl@label}
+% The |\outl@label| macro, called by |\outl| with the |outl@indsize|
+% as an argument, simply decides which level to make by way of an
+% |\ifcase| statement.
+% \begin{macrocode}
+\newcommand{\outl@label}[1]{%
+ \ifcase #1
+ \@dolabeli\or\@dolabelii\or\@dolabeliii%
+ \or\@dolabeliv\or\@dolabelv\or\@dolabelvi%
+ \or\@dolabelvii\else\@dolabeldef%
+ \fi}
+% \end{macrocode}
+% \end{macro}
+% The labels are set in |\labelfont|. I have a set of personal macros
+% that define a |\labelfont| for other uses, so the declaration is
+% surrounded by a test.
+% \begin{macrocode}
+\@ifundefined{labelfont}
+ {\DeclareOldFontCommand{\labelfont}%
+ {\normalfont\bfseries\mathversion{bold}}{\mathbf}}
+ {} % Do nothing---|\labelfont| is already defined
+% \end{macrocode}
+% The generic form of the outline labels is given by |\theoutlabel|.
+% \begin{macrocode}
+\newcommand{\theoutlabel}[1]{%
+ \llap{\hbox{\labelfont#1\enskip}}}
+% \end{macrocode}
+% Finally we define the macros which are called for the auto-numbering
+% and label creation for the |\outl| macro.
+% \changes{v0.6}{1996/04/26}{Changed \texttt{\protect\bslash stepcounter}
+% call to \texttt{\protect\bslash refstepcounter} in order to get
+% cross-referencing to work right.}
+% \begin{macrocode}
+\newcommand{\@dolabeli}{%
+ \refstepcounter{outli}\theoutlabel{\labeloutli}}
+\newcommand{\@dolabelii}{%
+ \refstepcounter{outlii}\theoutlabel{\labeloutlii}}
+\newcommand{\@dolabeliii}{%
+ \refstepcounter{outliii}\theoutlabel{\labeloutliii}}
+\newcommand{\@dolabeliv}{%
+ \refstepcounter{outliv}\theoutlabel{\labeloutliv}}
+\newcommand{\@dolabelv}{%
+ \refstepcounter{outlv}\theoutlabel{\labeloutlv}}
+\newcommand{\@dolabelvi}{%
+ \refstepcounter{outlvi}\theoutlabel{\labeloutlvi}}
+\newcommand{\@dolabelvii}{%
+ \refstepcounter{outlvii}\theoutlabel{\labeloutlvii}}
+\newcommand{\@dolabeldef}{%
+ \refstepcounter{outldef}\theoutlabel{\labeloutldef}}
+% \end{macrocode}
+% And with that, we are finished.
+%
+% \iffalse
+%</package>
+% \fi
+%
+% \Finale
diff --git a/Master/texmf-dist/source/latex/cjw/cjwoutl.ins b/Master/texmf-dist/source/latex/cjw/cjwoutl.ins
new file mode 100644
index 00000000000..f93dcfa0587
--- /dev/null
+++ b/Master/texmf-dist/source/latex/cjw/cjwoutl.ins
@@ -0,0 +1,18 @@
+\def\batchfile{cjwoutl.ins}
+\input docstrip.tex
+
+\preamble
+
+Copyright (C) 1995 by Colin J. Wynne
+
+This file is part of cjwoutl, a package
+for generating outlines with LaTeX2e.
+
+\endpreamble
+
+% Don't display progress
+\keepsilent
+
+\generateFile{cjwoutl.sty}{t}{\from{cjwoutl.dtx}{package}}
+
+\endinput
diff --git a/Master/texmf-dist/source/latex/cjw/cjwplain.dtx b/Master/texmf-dist/source/latex/cjw/cjwplain.dtx
new file mode 100644
index 00000000000..494882b5a55
--- /dev/null
+++ b/Master/texmf-dist/source/latex/cjw/cjwplain.dtx
@@ -0,0 +1,1007 @@
+% \iffalse -*- LaTeX -*-
+%
+% This is file `cjwplain.dtx'. You can run this file through
+% LaTeX2e to get a .dvi file of documentation. The file
+% `cjwplain.ins' should have come with this file. Run LaTeX2e on
+% it to produce the actual input files.
+%
+%<+package|tabbing|output> \NeedsTeXFormat{LaTeX2e}
+%<+package> \ProvidesPackage{cjwplain}
+%<+tabbing> \ProvidesFile{cjwpltab.clo}
+%<+output> \ProvidesFile{cjwplout.clo}
+%<+package|tabbing|output> [1995/05/01 v0.5
+%<+package> Plain TeX Macros for LaTeX2e---CJW]
+%<+tabbing> Plain TeX Tabbing Macros for LaTeX2e---CJW]
+%<+output> Plain TeX Output Macros for LaTeX2e---CJW]
+%
+% \fi
+\def\RCSinfo{$Id: cjwplain.dtx,v 1.2 1998/08/31 22:57:40 cwynne Exp $}
+
+\def\RCSsplit $#1: #2,v #3 #4 #5 #6 #7${
+ \gdef\filename {#2}
+ \gdef\fileversion{#3}
+ \gdef\filedate {#4}
+ \gdef\filetime {#5}
+ \gdef\fileauthor {#6}
+ \gdef\filelocker {#7}}
+
+\expandafter\RCSsplit\RCSinfo
+%% \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 \~}
+%%
+% \CheckSum{1208}
+%
+% \iffalse
+%
+%<*driver>
+\NeedsTeXFormat{LaTeX2e}
+\ProvidesFile{cjwplain.dtx}
+ [\filedate\space v\fileversion\space
+ The cjwplain Package for LaTeX2e---CJW]
+\documentclass{ltxdoc}
+ \GetFileInfo{cjwplain.dtx}
+ \RecordChanges
+\begin{document}
+ \DocInput{\filename}
+\end{document}
+%</driver>
+% \fi
+%
+% \DeclareRobustCommand{\cseq}[1]{\texttt{\bslash#1}}
+% \DeclareRobustCommand{\pkg} [1]{\textsf{#1}}
+% \DeclareRobustCommand{\env} [1]{\textsf{#1}}
+%
+% \title{The \textsf{cjwplain} Package\thanks{%
+% This file has version \fileversion.} \\
+% (\PlainTeX{} under \LaTeXe)}
+% \author{Colin J.~Wynne\thanks{E-Mail at:
+% \texttt{cwynne@brutus.mts.jhu.edu},
+% \texttt{cwynne@jhu.edu}\,.}}
+% \date{\filedate}
+%
+% \maketitle
+%
+% \setcounter{StandardModuleDepth}{1}
+%
+% {\parskip 0pt ^^A% This is the hack used by |doc.dtx|.
+% ^^A% (bug in \LaTeX?)
+% \tableofcontents
+% }
+%
+% \section*{Introduction}
+%
+% I first started using \TeX{} some two-and-a-half years ago, having
+% been introduced to it by several \TeX nophiles in my college math
+% department. I was aware from the start that there was a somehow
+% `bastardised' version of this very good program which went by the
+% name `\LaTeX'---invariably referred to by my \TeX{} mentors as
+% `Lame\TeX'. Most of you have probably heard this epithet before.
+%
+% Well, I count it as a good thing that I was discouraged from using
+% \LaTeX{} at first, as I ended up writing quite a lot of \TeX{} code
+% for myself before I ever got around to actually reading \textit{The
+% \TeX book} in its entirety, which I did only a few
+% months ago. My first real project for \TeX{} was writing a large
+% macro set for my undergraduate thesis---table of contents, marks for
+% the running heads, chapter and section delineation and so forth.
+% This growing library was expanded as I decided I wanted a good set
+% of macros for writing outlines and by requirements for various
+% papers, such as using endnotes in lieu of footnotes. This rather
+% haphazard collection of mine underwent a major change when I found
+% the macro package for NFSS (version 1) under \PlainTeX{}, of which I
+% promptly took advantage.
+%
+% This year I finally got my own computer, mainly to run \TeX{}.
+% Given my newly purchased copy of \textit{The \TeX book} and some
+% free time, I began to try to organise that cluster of code. Having
+% learned something in the meanwhile about generic markup, and why
+% it is preferable, I started rewriting for more generalisation. Also
+% in the meanwhile, \LaTeXe{} had come along, greatly enhancing
+% \LaTeX's own use of generic markup. It also standardised the NFSS,
+% which I had so come to appreciate. Basically, between the various
+% chunks of \LaTeXe{} which I had already hacked to work under my
+% custom format and the movement towards increasingly generic code by
+% both myself and the \LaTeX{} folks, that a convergence was taking
+% place, so I finally decided to give \LaTeXe{} a serious looking-at.
+%
+% I started by printing out the documented source code. I liked a lot
+% of what I saw---but there were two problems. Some of my favourite
+% bits of \PlainTeX{} got left by the wayside. For doing a lot of
+% mathematics, I still find |\eqalign| to be the easiest way of
+% aligning a bunch of related equations. First, it involves less
+% typing than a |\begin|\ldots|\end| pair, and I don't often need
+% equation numbers---something not easily done away with under vanilla
+% \LaTeX. The bigger concern was that I had a bunch of source files
+% that were written under \PlainTeX{} (or, rather, under \emph{my}
+% \PlainTeX), and I didn't want to have to make the dozens of minor
+% modifications necessary to get them to work under \LaTeX.
+%
+% So, I decided to learn how \LaTeX{} does things and to do so by
+% writing a package that would, at its simplest, let me add a
+% |\documentclass| and a |\begin{document}| line to one of my existing
+% \PlainTeX{} sourc files and get it to compile under \LaTeX.
+%
+% Thus, I have written my first \LaTeX{} package. I consider the main
+% feature to be the ability to very easily add NFSS commands to a
+% document written under \PlainTeX. Secondly, maybe it will help
+% convince some other \PlainTeX{} die-hards to give \LaTeX{} a try,
+% inasmuch as all of their standard commands will be supported.
+% Finally, it should let those who use \LaTeX{} exclusively to easily
+% deal with \PlainTeX{} files if the need arises.
+%
+% \pagebreak
+% Feel free to let me know if you find this package useful or, of
+% course, if you find any bugs or wish to suggest improvements.
+%
+% \subsection*{This Package}
+%
+% This package is built over the file |ltplain.dtx|, or, more
+% correctly, over those parts of |ltplain.dtx| which were changes to
+% or omissions of the original \PlainTeX{} source. Some parts,
+% specifically font changes, have not been reproduced in their
+% entirety, due basically to the fact that such would be a pointless
+% exercise. See the comments in Section \ref{sec:font} for the
+% explanation.
+%
+% Finally, this document prints with all source code because I feel
+% the source itself, and the modifications to it, are the best
+% documentation.
+%
+%
+% \section{Package Options}
+%
+% According to the documented \LaTeXe{} source file |ltplain.dtx|,
+% \begin{quote}
+% \LaTeX\ includes almost all of the functionality of Knuth's
+% original `Basic Macros' That is, the plain \TeX\ format described
+% in Appendix~B of the \TeX{}Book.
+% \end{quote}
+% It seems to me that removing the qualifying `almost' would be no bad
+% thing.
+%
+% The idea behind the available options is that a given user may need
+% only certain aspects of \PlainTeX{} added back in for a document.
+% Furthermore, the additional code can sometimes be specified in
+% different ways---\emph{i.e.}, either strictly according to the
+% definitions of \PlainTeX{} or in a manner syntactically identical to
+% \PlainTeX{} but functionally grounded in \LaTeX. The overall goal,
+% though, is completeness; I have therefore included everything, in
+% one form or another, even when I can't think of a reason why some
+% things would be necessary.
+%
+% There are nineteen regular and two special options available for the
+% |cjwplain| package. (All are entered in standard \LaTeX{} form, as
+% optional arguments to the |\usepackage| command. I only call two of
+% them `special' so as to draw attention to them.)
+%
+% \subsection{Regular Options}
+%
+% \DescribeMacro{Options}
+% The regular options are:
+% \changes{v0.5}{1995/06/08}{Added \env{plainskips} and
+% \env{plainskipsoff}.}
+% \begin{center}
+% \begin{tabular}{llll}
+% outerallocs & diagnostics & plainskips \\
+% outerallocsoff & diagnosticsoff & plainskipsoff \\
+% strictline & tabbing & strictitem \\
+% strictlineoff & tabbingoff & strictitemoff \\
+% eqalign & magnification & plainoutput \\
+% eqalignoff & magnificationoff & plainoutputoff \\
+% strictfootnotes & altfootnotes & footnotesoff \\
+% \end{tabular}
+% \end{center}
+% Note that most of these options come in
+% \meta{option}/\meta{option}|off| pairs. These are
+% particularly useful in conjunction with the special options
+% (\ref{sec:spec-opts}) or to toggle the default options. By default,
+% the options |diagnostics|, |tabbing| and |eqalign| are active (just
+% what I tend to use\ldots).
+%
+% The actual options will be explained in section \ref{sec:code}.
+% Keep in mind, though, that some options affect others---for example
+% selecting one of |strictfootnotes| and |altfootnotes| will
+% automatically turn the other off; you can, however, disable both
+% forms either with |footnotesoff| or by giving the two separate
+% \ldots|off| commands separately. Also, |plainoutput| requires
+% |strictfootnotes| and |magnification|, but |plainoutputoff| itself
+% does not disable the \PlainTeX{} footnote macros or magnification.
+%
+% \subsection{Special Options}
+% \label{sec:spec-opts}
+%
+% \DescribeMacro{none}\DescribeMacro{all}
+% Two options, called |none| and |all|, are available to allow maximum
+% flexibility. These function because |cjwplain| calls the starred
+% command |\ProcessOptions*| and therefore processes options in the
+% order specified to |\usepackage|, and not the package's internal
+% declaration order. Thus, to make only \PlainTeX's tabbing commands
+% available, one would use the call
+% \begin{verbatim}
+% \usepackage[none,tabbing]{cjwplain}
+% \end{verbatim}
+% and to use everything while leaving \LaTeX's |\item| command alone
+% one would enter the command
+% \begin{verbatim}
+% \usepackage[all,strictitemoff]{cjwplain}
+% \end{verbatim}
+% in the preamble.
+%
+%\StopEventually{\PrintChanges\PrintIndex}
+%
+% \section{The Code}
+% \label{sec:code}
+%
+% \subsection{Declarations}
+%
+% The options are implemented as |\if| statements, as that seemed to
+% me to be the easiest way of including or excluding relatively large
+% sections of code. First we allocate the |\if|s.
+% \changes{v0.5}{1995/06/04}{Added \cseq{if@plainskips} and related macros
+% for the \textsf{plainskips} option.}
+% \begin{macrocode}
+%<*package>
+\newif\if@outerallocs \@outerallocsfalse
+\newif\if@diagnostics \@diagnosticstrue
+\newif\if@plainskips \@plainskipsfalse
+\newif\if@strictline \@strictlinetrue
+\newif\if@tabbing \@tabbingtrue
+\newif\if@strictitem \@strictitemfalse
+\newif\if@eqalign \@eqaligntrue
+\newif\if@strictfootnotes \@strictfootnotesfalse
+\newif\if@altfootnotes \@altfootnotesfalse
+\newif\if@plainoutput \@plainoutputfalse
+\newif\if@magnification \@magnificationfalse
+% \end{macrocode}
+%
+% Now we declare how the options affect these |\if| tests.
+%
+% \begin{macrocode}
+\DeclareOption{outerallocs}{\@outerallocstrue}
+\DeclareOption{outerallocsoff}{\@outerallocsfalse}
+
+\DeclareOption{diagnostics}{\@diagnosticstrue}
+\DeclareOption{diagnosticsoff}{\@diagnosticsfalse}
+
+\DeclareOption{plainskips}{\@plainskipstrue}
+\DeclareOption{planiskipsoff}{\@plainskipsfalse}
+
+\DeclareOption{strictline}{\@strictlinetrue}
+\DeclareOption{strictlineoff}{\@strictlinefalse}
+
+\DeclareOption{tabbing}{\@tabbingtrue}
+\DeclareOption{tabbingoff}{\@tabbingfalse}
+
+\DeclareOption{strictitem}{\@strictitemtrue}
+\DeclareOption{strictitemoff}{\@strictitemfalse}
+
+\DeclareOption{eqalign}{\@eqaligntrue}
+\DeclareOption{eqalignoff}{\@eqalignfalse}
+% \end{macrocode}
+%
+% We will have two possible ways of providing a |\footnote| command.
+% As these are mutually exclusive, we make sure that they cannot both
+% be true.
+% \begin{macrocode}
+\DeclareOption{strictfootnotes}{%
+ \@strictfootnotestrue \@altfootnotesfalse}
+\DeclareOption{altfootnotes}{%
+ \@altfootnotestrue \@strictfootnotesfalse}
+\DeclareOption{footnotesoff}{%
+ \@altfootnotesfalse \@strictfootnotesfalse}
+
+\DeclareOption{magnification}{\@magnificationtrue}
+\DeclareOption{magnificationoff}{\@magnificationfalse}
+% \end{macrocode}
+% To use \PlainTeX's entire output routine will require that
+% magnification code as well as \PlainTeX{} style footnotes be
+% defined.
+% \begin{macrocode}
+\DeclareOption{plainoutput}{%
+ \@plainoutputtrue \@strictfootnotestrue
+ \@altfootnotesfalse \@magnificationtrue}
+\DeclareOption{plainoutputoff}{\@plainoutputfalse}
+% \end{macrocode}
+% The two special options are given.
+% \begin{macrocode}
+\DeclareOption{none}{%
+ \@outerallocsfalse \@eqalignfalse
+ \@diagnosticsfalse \@plainskipsfalse
+ \@strictfootnotesfalse\@strictlinefalse
+ \@altfootnotesfalse \@tabbingfalse
+ \@magnificationfalse \@strictitemfalse
+ \@plainoutputfalse}
+
+\DeclareOption{all}{%
+ \@outerallocstrue \@eqaligntrue
+ \@diagnosticstrue \@plainskipstrue
+ \@strictfootnotestrue\@strictlinetrue
+ \@altfootnotestrue \@tabbingtrue
+ \@magnificationtrue \@strictitemtrue
+ \@plainoutputtrue}
+% \end{macrocode}
+% Finally we define a default option handling routine. I prefer only
+% a warning as opposed to an error.
+% \begin{macrocode}
+\DeclareOption*{%
+ \PackageWarning{cjwplain}{Unknown option `\CurrentOption'}}
+% \end{macrocode}
+% Now that all the options are declared, we process them in the order
+% specified in the package call.
+% \begin{macrocode}
+\ProcessOptions*
+% \end{macrocode}
+%
+% \subsection{Allocation Calls: \env{outerallocs}}
+%
+%\DescribeMacro{\newcount}
+%\DescribeMacro{\newdimen}
+%\DescribeMacro{\newskip}
+%\DescribeMacro{\newbox}
+%\DescribeMacro{\newwrite}
+%\DescribeMacro{\newfam}
+% Originally \PlainTeX{} had all allocation macros (|\newcount|, etc.)
+% defined as |\outer|. \LaTeX{} redefines several of them to be
+% non-outer. Careful consideration has failed to yield to me why
+% these would need to be rewritten as |\outer| in this package---any
+% \PlainTeX{} file which expects |\outer| definitions would not call
+% them in a non-outer position, and any other files would themselves
+% have redefined versions of the macros.
+%
+% Since, however, it is such a small change, we will provide it.
+% \textsc{Note:} Using the |outerallocs| option will break a good
+% deal of standard \LaTeX{} code, namely the standard macros for
+% counters and lengths. This means you probably do \emph{not} want
+% to use it. It is here only for completeness's sake.
+% \begin{macrocode}
+\if@outerallocs
+
+\outer\def\newcount{\alloc@0\count\countdef\insc@unt}
+\outer\def\newdimen{\alloc@1\dimen\dimendef\insc@unt}
+\outer\def\newskip{\alloc@2\skip\skipdef\insc@unt}
+
+\outer\def\newbox{\alloc@4\box\chardef\insc@unt}
+\outer\def\newwrite{\alloc@7\write\chardef\sixt@@n}
+\outer\def\newfam{\alloc@8\fam\chardef\sixt@@n}
+
+\fi
+% \end{macrocode}
+%
+% \subsection{Error Processing: \env{diagnostics}}
+%
+% Any \PlainTeX{} afficianados using this package will feel more
+% comfortable to have the standard values for error processing
+% information. One change, though. \LaTeXe{} uses a \emph{counter}
+% named |errorcontextlines|, and not a count.
+%
+% \begin{macrocode}
+\if@diagnostics
+
+\showboxbreadth=5
+\showboxdepth=3
+\setcounter{errorcontextlines}{5}
+
+\fi
+% \end{macrocode}
+%
+% \subsection{Skips: \env{plainskips}}
+% \changes{v0.5}{1995/06/08}{Added the sub-section `Skips'.}
+%
+% When the |plainskips| option is selected, the three |\|\dots|skip|
+% macros should unconditionally leave horizontal mode and insert a
+% skip, like in \PlainTeX.
+% \begin{macrocode}
+\if@plainskips
+ \def\smallskip{\vskip\smallskipamount}
+ \def\medskip{\vskip\medskipamount}
+ \def\bigskip{\vskip\bigskipamount}
+\fi
+% \end{macrocode}
+%
+% \subsection{Fonts}
+% \label{sec:font}
+%
+% A package already exists whereby oldstyle font commands can be
+% given, namely |oldlfont|. Furthermore, one can use |rawfonts|, if
+% necessary, to load in such specific fonts as |\ninebf|, etc.
+%
+% \begin{macrocode}
+%\font\tenrm=cmr10 % roman text
+% \end{macrocode}
+% \qquad$\vdots$
+% \begin{macrocode}
+%\textfont\ttfam=\tentt
+% \end{macrocode}
+%
+% \subsection{The \cseq{line} Macro: \env{strictline}}
+%
+%\DescribeMacro{\line}
+%\DescribeMacro{\latex@line}
+% Now we get to the first tricky part. The |\line| macro needs to be
+% available to the |picture| environment in \LaTeX, as well as
+% restoring the original \PlainTeX{} definition for our usage here.
+% The good news is that \LaTeX{} \emph{only} uses |\line| inside of
+% the |picture| environment. So we employ the following solution: we
+% keep the definition of |\@@line| as per \LaTeX{} convention, and
+% give in any case a user accessible |\plainline|.
+% \begin{macrocode}
+\let\plainline\@@line
+
+\if@strictline
+
+%\def\@@line{\hbox to\hsize} % Defined in |ltplain.dtx|
+% \end{macrocode}
+% Now we define an internal name for the standard \LaTeX{} macro and
+% restore the \PlainTeX{} definition.
+% \begin{macrocode}
+\let\latex@line\line
+\let\line\@@line
+% \end{macrocode}
+% The definitions of |\leftline|, |\rightline| and |\centerline| can
+% be left as is (though users depending upon personal redefinitions of
+% |\line| for special effects in these macros should simply put their
+% redefinition into the macro |\@@line|).
+% \begin{macrocode}
+%\def\leftline#1{\@@line{#1\hss}}
+%\def\rightline#1{\@@line{\hss#1}}
+%\def\centerline#1{\@@line{\hss#1\hss}}
+% \end{macrocode}
+% Now we make a patch to the definition of |\@picture| (the workhorse
+% macro for the |picture| environment) which will restore the \LaTeX{}
+% definition only within that environment.
+% \begin{macrocode}
+\def\@picture(#1,#2)(#3,#4){%
+ \let\line\latex@line%
+ \@picht#2\unitlength
+ \setbox\@picbox\hbox to#1\unitlength\bgroup
+ \hskip -#3\unitlength
+ \lower #4\unitlength\hbox\bgroup
+ \ignorespaces}
+
+\fi
+% \end{macrocode}
+%
+% \subsection{Tab Alignments: \env{tabbing}}
+%
+% The tabbing macros from \PlainTeX{} use the |\newif| construction,
+% so must occur at an |\outer| level. Thus, they are included in a
+% separate package.
+% \changes{v0.5}{1995/06/04}{Use \cseq{InputIfFileExists} instead of
+% \cseq{RequirePackage} to input \pkg{cjwpltab.clo}; otherwise a
+% warning (error?) is produced, since the latter expects a
+% \pkg{.sty} file.}
+% \begin{macrocode}
+\if@tabbing
+ \InputIfFileExists{cjwpltab.clo}{}{%
+ \PackageWarning{cjwplain}{Option `cjwpltab.clo' not found.}
+ \@tabbingfalse}
+\fi
+%</package>
+%<*tabbing>
+% \end{macrocode}
+%
+%\DescribeMacro{\cleartabs}
+%\DescribeMacro{\settabs}
+%\DescribeMacro{\tabalign}
+%\DescribeMacro{\+}
+% \LaTeX{} may have it's own tabbing environment, but I like
+% \PlainTeX's. The only potential conflict I saw was with the
+% \verb|\+| macro. However, \LaTeX{} only defines \verb|\+| inside of
+% the |tabbing| environment itself, so there should be absolutely no
+% problem.
+% \begin{macrocode}
+\newif\ifus@ \newif\if@cr
+\newbox\tabs \newbox\tabsyet \newbox\tabsdone
+
+\def\cleartabs{\global\setbox\tabsyet\null \setbox\tabs\null}
+\def\settabs{\setbox\tabs\null \futurelet\next\sett@b}
+\let\+=\relax % in case this file is being read in twice
+\def\sett@b{\ifx\next\+\let\next\relax
+\def\next{\afterassignment\s@tt@b\let\next}%
+ \else\let\next\s@tcols\fi\next}
+\def\s@tt@b{\let\next\relax\us@false\m@ketabbox}
+\def\tabalign{\us@true\m@ketabbox} % non-\outer version of \+
+\outer\def\+{\tabalign}
+\def\s@tcols#1\columns{\count@#1 \dimen@\hsize
+ \loop\ifnum\count@>\z@ \@nother \repeat}
+\def\@nother{\dimen@ii\dimen@ \divide\dimen@ii\count@
+ \setbox\tabs\hbox{\hbox to\dimen@ii{}\unhbox\tabs}%
+ \advance\dimen@-\dimen@ii \advance\count@\m@ne}
+
+\def\m@ketabbox{\begingroup
+ \global\setbox\tabsyet\copy\tabs
+ \global\setbox\tabsdone\null
+ \def\cr{\@crtrue\crcr\egroup\egroup
+\ifus@\unvbox\z@\lastbox\fi\endgroup
+\setbox\tabs\hbox{\unhbox\tabsyet\unhbox\tabsdone}}%
+ \setbox\z@\vbox\bgroup\@crfalse
+\ialign\bgroup&\t@bbox##\t@bb@x\crcr}
+
+\def\t@bbox{\setbox\z@\hbox\bgroup}
+\def\t@bb@x{\if@cr\egroup % now \box\z@ holds the column
+ \else\hss\egroup \global\setbox\tabsyet\hbox{\unhbox\tabsyet
+ \global\setbox\@ne\lastbox}% now \box\@ne holds its size
+\ifvoid\@ne\global\setbox\@ne\hbox to\wd\z@{}%
+\else\setbox\z@\hbox to\wd\@ne{\unhbox\z@}\fi
+\global\setbox\tabsdone\hbox{\box\@ne\unhbox\tabsdone}\fi
+ \box\z@}
+%</tabbing>
+%<*package>
+% \end{macrocode}
+%
+% \subsection{Itemising: \env{strictitem}}
+%
+%\DescribeMacro{\@@item}
+%\DescribeMacro{\plainitem}
+% Now we have another problem, namely the |\item| macro. I
+% unfortunately see no way to get around the fact that |\item| is a
+% general macro in \LaTeX{}, and that the formats are completely
+% different: i.e., \PlainTeX{} expects the \meta{label} to be the one
+% mandatory argument, whereas \LaTeX's |\item| macro takes the
+% \meta{label} as an optional argument. Thus, the best I can think of
+% is the following. We redefine \PlainTeX's |\item| after standard
+% \LaTeX{} practice,
+% \begin{macrocode}
+\def\@@item{\par\hang\textindent}
+% \end{macrocode}
+% and we |\let| it to something accessible in normal documents, the
+% command |\plainitem|.
+% \begin{macrocode}
+\let\plainitem\@@item
+% \end{macrocode}
+% The command |\itemitem| can be taken care of directly.
+% \begin{macrocode}
+\def\itemitem{\par\indent \hangindent2\parindent \textindent}
+% \end{macrocode}
+%
+% Now a user will have to replace all occurrences of |\item{foo}| with
+% either |\item[foo]| or |\plainitem{foo}| (I imagine the choice will
+% depend upon one's editor's facilities with regexps\ldots). It's not
+% perfect, but it's the only way I can think of to provide maximum
+% compatibility. Of course, we will still give the option,
+% |strictitem|, of using
+% just the original definition, but that will probably not be terribly
+% convenient for anyone trying to add \LaTeX{} features on top of an
+% existing \PlainTeX{} source. Thus, we will also provide the
+% (slightly longwinded) replacement |\latexitem|.
+% \begin{macrocode}
+\if@strictitem
+ \let\latexitem\item
+ \let\item\@@item
+\fi
+% \end{macrocode}
+%
+% \subsection{Miscellaneous}
+%
+% \subsubsection{Sectioning}
+%
+% I have personally never used the \PlainTeX{} |\beginsection| macro,
+% but somebody might have\ldots
+% \begin{macrocode}
+\outer\def\beginsection#1\par{\vskip\z@ plus.3\vsize\penalty-250
+ \vskip\z@ plus-.3\vsize\bigskip\vskip\parskip
+ \message{#1}\leftline{\bf#1}\nobreak\smallskip\noindent}
+% \end{macrocode}
+%
+% \subsubsection{Proclamations}
+%
+% Once again we will leave \LaTeX's NFSS based redefinition, this time
+% for the |\proclaim| command, in place.
+% \begin{macrocode}
+%\outer\def\proclaim #1. #2\par{\medbreak
+% \noindent{\bfseries#1.\enspace}{\slshape#2\par}%
+% \ifdim\lastskip<\medskipamount \removelastskip\penalty55\medskip\fi}
+% \end{macrocode}
+%
+% \subsubsection{Paragraph Formatting}
+%
+% I have done some simple tests of \LaTeX's |\raggedright| macro, and
+% it seems to me that it mimics the functionality of the \PlainTeX{}
+% macro of the same name. Therefore I see no reason to redefine it as
+% part of this package.
+% \begin{macrocode}
+%\def\raggedright{%
+% \rightskip\z@ plus2em \spaceskip.3333em \xspaceskip.5em\relax}
+% \end{macrocode}
+% Another \LaTeX{} font change will also be left as is.
+% \begin{macrocode}
+%\def\ttraggedright{\reset@font\ttfamily\rightskip\z@ plus2em\relax}
+% \end{macrocode}
+%
+% \subsubsection{Accents and Miscellaneous}
+%
+% These should work as is for \PlainTeX{} documents.
+% \begin{macrocode}
+%\def\_{\leavevmode \kern.06em \vbox{\hrule \@width.3em}}
+%\def\AA{\leavevmode\setbox0\hbox{h}\dimen@\ht0\advance\dimen@-1ex%
+% \rlap{\raise.67\dimen@\hbox{\char'27}}A}
+% \end{macrocode}
+% Nor do I see a reason to change these back to \PlainTeX{}
+% definitions.
+% \begin{macrocode}
+%\def\pd#1{\oalign{#1\crcr\hidewidth\sh@ft{08}.\hidewidth}}
+%\def\d{\protect\pd}
+%
+%\def\pb#1{\oalign{#1\crcr\hidewidth\sh@ft{29}%
+%\vbox to.2ex{\hbox{\char22}\vss}\hidewidth}}
+%\def\b{\protect\pb}
+%
+%\def\pc#1{\setbox\z@\hbox{#1}\ifdim\ht\z@=1ex\accent24 #1%
+% \else{\ooalign{\unhbox\z@\crcr\hidewidth\char24\hidewidth}}\fi}
+%\def\c{\protect\pc}
+%
+%\def\pt#1{{\edef\next{\the\font}\the\textfont1\accent127\next#1}}
+%\def\t{\protect\pt}
+% \end{macrocode}
+%
+% The \LaTeX{} definition of |\ldots| is more or less identical to the
+% \PlainTeX{} macro |\dots|. So we will leave this alone, too.
+% \begin{macrocode}
+%\def\dots{\ldots}
+% \end{macrocode}
+%
+% These changes, as others before, only add functionality without
+% seeming to limit \PlainTeX{} usage, so no change will be made.
+% \begin{macrocode}
+%\def\hrulefill{\leavevmode\leaders\hrule\hfill\kern\z@}
+%\def\dotfill{\leavevmode\cleaders
+% \hbox{$\m@th \mkern1.5mu.\mkern1.5mu$}\hfill\kern\z@}
+%
+%\def\longrightarrow{\protect\@lra}
+% \def\@lra{\relbar\joinrel\rightarrow}
+%\def\longleftarrow{\protect\@lla}
+% \def\@lla{\leftarrow\joinrel\relbar}
+% \end{macrocode}
+%
+% \subsubsection{Ending the Document}
+%
+% We simply add the |\bye| macro back in, though the |\end| should be
+% changed to the \LaTeX{} |\end{document}|.
+% \begin{macrocode}
+\outer\def\bye{\end{document}}
+% \end{macrocode}
+%
+%
+% \subsubsection{Math Commands}
+%
+% Operators and other math-mode font-related changes will be ignored
+% with as other NFSS alterations already mentioned.
+% \begin{macrocode}
+%\def\log{\mathop{\rm log}\nolimits}
+%\def\lg{\mathop{\rm lg}\nolimits}
+%\def\deg{\mathop{\rm deg}\nolimits}
+% \end{macrocode}
+%\qquad$\vdots$
+% \begin{macrocode}
+%\def\bmod{\mskip-\medmuskip\mkern5mu
+% \mathbin{\rm mod}\penalty900\mkern5mu\mskip-\medmuskip}
+%\def\pmod#1{\allowbreak\mkern18mu({\rm mod}\,\,#1)}
+% \end{macrocode}
+% Various |\matrix| type command, including |\bordermatrix| and
+% |\cases| have similarly been rewritten for NFSS commands under
+% \LaTeX.
+%
+% \subsection{Math Alignment: \env{eqalign}}
+%
+%\DescribeMacro{\eqalign}
+%\DescribeMacro{\eqalignno}
+%\DescribeMacro{\leqalignno}
+% If the |eqalign| option has been selected, we add |\eqalign| back
+% in, as well as related macros. Notice that we use the |\@centering|
+% macro provided by \LaTeX, because \LaTeX{} itself reserves
+% |\centering|. We also supply a user macro |\plaincentering| in case
+% such is needed.
+% \begin{macrocode}
+\let\plaincentering\@centering
+
+\if@eqalign
+
+\def\eqalign#1{\null\,\vcenter{\openup\jot\m@th
+ \ialign{\strut\hfil$\displaystyle{##}$&$\displaystyle{{}##}$\hfil
+ \crcr#1\crcr}}\,}
+
+\def\eqalignno#1{\displ@y \tabskip\@centering
+ \halign to\displaywidth{\hfil$\@lign\displaystyle{##}$\tabskip\z@skip
+ &$\@lign\displaystyle{{}##}$\hfil\tabskip\@centering
+ &\llap{$\@lign##$}\tabskip\z@skip\crcr
+ #1\crcr}}
+\def\leqalignno#1{\displ@y \tabskip\@centering
+ \halign to\displaywidth{\hfil$\@lign\displaystyle{##}$\tabskip\z@skip
+ &$\@lign\displaystyle{{}##}$\hfil\tabskip\@centering
+ &\kern-\displaywidth\rlap{$\@lign##$}\tabskip\displaywidth\crcr
+ #1\crcr}}
+
+\else
+% \end{macrocode}
+%\DescribeEnv{eqnarray*}
+% If the user does not choose this option, we will instead define an
+% |eqnarray*| environment which does not number equations.
+% \begin{macrocode}
+\@namedef{eqnarray*}{%
+ \let \\ \cr $$\null\,\vcenter{\openup\jot\m@th
+ \ialign{\strut\hfil$\displaystyle{##}$&$\displaystyle{{}##}$\hfil\crcr}}}
+\@namedef{endeqnarray*}{\crcr}
+
+\fi
+% \end{macrocode}
+%
+% \subsection{Output Routine: \env{plainoutput}}
+%
+% The output routines also involve |\newif| commands, and are
+% therefore also relegated to a separate package.
+% \changes{v0.5}{1995/06/04}{Use \cseq{InputIfFileExists} instead of
+% \cseq{RequirePackage} to input \pkg{cjwplout.clo}; otherwise a
+% warning (error?) is produced, since the latter expects a
+% \pkg{.sty} file.}
+% \begin{macrocode}
+\if@plainoutput
+ \InputIfFileExists{cjwplout.clo}{}{%
+ \PackageWarning{cjwplain}{Option `cjwplout.clo' not found.}
+ \@tabbingfalse}
+\fi
+%</package>
+%<*output>
+% \end{macrocode}
+%
+%\DescribeMacro{\headline}
+%\DescribeMacro{\footline}
+%\DescribeMacro{\pageno}
+%\DescribeMacro{\folio}
+% If the user wishes to use the entire \PlainTeX{} output routine, we
+% first redefine the normal versions of |headline|, |footline| and
+% |pageno|, as well as related macros. We will use NFSS definitions
+% in place of |\tenrm|.
+% \begin{macrocode}
+\countdef\pageno=0 \pageno=1 % first page is number 1
+\newtoks\headline \headline={\hfil} % headline is normally blank
+\newtoks\footline \footline={\hss\reset@font\folio\hss}
+ % footline is normally a centered page number in font \tenrm
+\def\nopagenumbers{\footline{\hfil}} % blank out the footline
+\def\folio{%
+ \ifnum\pageno<\z@ \romannumeral-\pageno \else\number\pageno \fi}
+\def\advancepageno{\ifnum\pageno<\z@ \global\advance\pageno\m@ne
+ \else\global\advance\pageno\@ne \fi} % increase |pageno|
+% \end{macrocode}
+%\DescribeMacro{\raggedbottom}
+%\DescribeMacro{\normalbottom}
+% We also supply the |\raggedbottom| macro and its counterpart,
+% |\normalbottom|.
+% \begin{macrocode}
+\newif\ifr@ggedbottom
+\def\raggedbottom{\topskip 10\p@ plus60\p@ \r@ggedbottomtrue}
+\def\normalbottom{\topskip 10\p@ \r@ggedbottomfalse}
+ % undoes \raggedbottom
+% \end{macrocode}
+%
+%\DescribeMacro{\topinsert}
+%\DescribeMacro{\midinsert}
+%\DescribeMacro{\pageinsert}
+% If the entire output routine is being used, we define the
+% \PlainTeX{} insertion macros as normal.
+% \begin{macrocode}
+\newinsert\topins
+\newif\ifp@ge \newif\if@mid
+\def\topinsert{\@midfalse\p@gefalse\@ins}
+\def\midinsert{\@midtrue\@ins}
+\def\pageinsert{\@midfalse\p@getrue\@ins}
+\skip\topins=\z@skip % no space added when a topinsert is present
+\count\topins=1000 % magnification factor (1 to 1)
+\dimen\topins=\maxdimen % no limit per page
+\def\@ins{\par\begingroup\setbox\z@\vbox\bgroup} % start a \vbox
+\def\endinsert{\egroup % finish the \vbox
+ \if@mid \dimen@\ht\z@ \advance\dimen@\dp\z@ \advance\dimen@12\p@
+ \advance\dimen@\pagetotal \advance\dimen@-\pageshrink
+ \ifdim\dimen@>\pagegoal\@midfalse\p@gefalse\fi\fi
+ \if@mid \bigskip\box\z@\bigbreak
+ \else\insert\topins{\penalty100 % floating insertion
+ \splittopskip\z@skip
+ \splitmaxdepth\maxdimen \floatingpenalty\z@
+ \ifp@ge \dimen@\dp\z@
+ \vbox to\vsize{\unvbox\z@\kern-\dimen@}% depth is zero
+ \else \box\z@\nobreak\bigskip\fi}\fi\endgroup}
+% \end{macrocode}
+%
+%\DescribeMacro{\plainoutput}
+% Now we define the main part of the output routine. We use |\@@line|
+% instead of |\line|, since |\@@line| is guaranteed to have the
+% definition we want.
+% \begin{macrocode}
+\def\plainoutput{\shipout\vbox{\makeheadline\pagebody\makefootline}%
+ \advancepageno
+ \ifnum\outputpenalty>-\@MM \else\dosupereject\fi}
+\def\pagebody{\vbox to\vsize{\boxmaxdepth\maxdepth \pagecontents}}
+\def\makeheadline{\vbox to\z@{\vskip-22.5\p@
+ \@@line{\vbox to8.5\p@{}\the\headline}\vss}\nointerlineskip}
+\def\makefootline{\baselineskip24\p@\@@line{\the\footline}}
+\def\dosupereject{\ifnum\insertpenalties>\z@
+ % something is being held over
+ \@@line{}\kern-\topskip\nobreak\vfill\supereject\fi}
+
+\def\pagecontents{\ifvoid\topins\else\unvbox\topins\fi
+ \dimen@=\dp\@cclv \unvbox\@cclv % open up \box255
+ \ifvoid\footins\else % footnote info is present
+ \vskip\skip\footins
+ \footnoterule
+ \unvbox\footins\fi
+ \ifr@ggedbottom \kern-\dimen@ \vfil \fi}
+% \end{macrocode}
+% Finally, we make the \PlainTeX{} output routines active again.
+% \begin{macrocode}
+ \output{\plainoutput}
+%</output>
+%<*package>
+% \end{macrocode}
+%
+% \subsubsection{Page Numbering, Running Heads and Miscellaneous}
+%
+% We can make the \PlainTeX{} head and foot commands accessible (after
+% a fashion), even if the entire output routine is not being used.
+% \LaTeXe{} provides the commands |\@oddhead| and |\@evenhead|, as
+% well as their footline equivalents. Therefore we can give the
+% following versions.
+% \begin{macrocode}
+\if@plainoutput\else
+
+\def\footline{\@ifnextchar ={\@@footline}{\@@footline=}}
+\def\@@footline=#1{\gdef\@oddfoot{#1} \gdef\@evenfoot{#1}}
+
+\def\headline{\@ifnextchar ={\@@headline}{\@@headline=}}
+\def\@@headline=#1{\gdef\@oddhead{#1} \gdef\@evenhead{#1}}
+
+\def\nopagenumbers{\let\@oddfoot\relax \let\@evenfoot\relax}
+% \end{macrocode}
+% Since the \PlainTeX{} |\headline| and |\footline| macros are
+% actually tokens, we have to allow that the assignments to them can
+% be made with an optional $=$. The use of |\@ifnextchar =| nicely
+% takes care of that. This will alas be a slightly inefficient use of
+% the macros, as many \PlainTeX{} heads and feet already test for odd
+% and even pages---but not all of them. So, we compromise.
+%
+% Now we can simply make |\pageno| an equivalent for the \LaTeX{}
+% counter |\c@page|.
+% \begin{macrocode}
+\let\pageno\c@page
+% \end{macrocode}
+% This means that assignments can either be made in primitive fashion
+% directly to |\pageno| or in \LaTeX fashion to the |page| counter.
+% We would also like to have the \PlainTeX{} |\folio| macro. We are
+% going to expand upon this slightly, though. I find |\folio|
+% generally to be a useful command, so we will write it in such a way
+% as to make it useful with \LaTeX{} counters (as with, say, the
+% |\arabic| macro) as well as with counts.
+% \begin{macrocode}
+\newcommand{\folio}[1][\pageno]{\@folio{#1}}
+\def\@folio#1{%
+ \edef\@tempa{\string#1}%
+ \expandafter\@ifundefined{c@\@tempa}% % Does the counter exist?
+ {% % No such counter.
+ \ifnum #1 <\z@ \romannumeral-#1
+ \else \number #1
+ \fi
+ }%
+ {% % It is a counter.
+ \ifnum\value{#1}<\z@ \roman{#1} \else \arabic{#1} \fi}
+ }
+% \end{macrocode}
+% This is not, unfortunately, perfect, in that it must be used with an
+% optional argument (|\folio[section]|) as opposed to the normal style
+% (|\arabic{section}|). On the other hand, I can't think of many
+% applications for |\folio| \emph{other} than page numbering. At any
+% rate, we finish page numbering with the incrementation command.
+% \begin{macrocode}
+\def\advancepageno{\ifnum\pageno<\z@ \global\advance\pageno\m@ne
+ \else\global\advance\pageno\@ne \fi}
+% \end{macrocode}
+%
+% One more bit from the \PlainTeX{} output routine needs to be dealt
+% with. Although \LaTeX's |\raggedbottom| macro will suffice to
+% simulate \PlainTeX's command of the same name, we need to add a
+% |\let| command to enable \PlainTeX's counterpart, |\normalbottom|.
+% \begin{macrocode}
+\let\normalbottom\flushbottom
+% \end{macrocode}
+%
+% \subsubsection{Insertions}
+%
+% If the \PlainTeX{} output routine is not being used, we simulate the
+% insertions using \LaTeX's |figure| environment.
+% \begin{macrocode}
+\def\topinsert{\begin{figure}[t]}
+\def\pageinsert{\begin{figure}[p]}
+\def\midinsert{\begin{figure}[htpb]}
+\def\endinsert{\end{figure}}
+
+\fi
+% \end{macrocode}
+%
+% \subsubsection{Footnotes: \env{srictfootnotes},
+% \env{altfootnotes}}
+%
+% We define |\@@footnote| as \PlainTeX's footnoting macro.
+% \begin{macrocode}
+%\newinsert\footins
+
+\let\latex@footnote\footnote
+
+\def\@@footnote#1{\let\@sf\empty % parameter #2 (the text) is read later
+ \ifhmode\edef\@sf{\spacefactor\the\spacefactor}\/\fi
+ #1\@sf\vfootnote{#1}}
+\def\vfootnote#1{\insert\footins\bgroup
+ \interlinepenalty\interfootnotelinepenalty
+ \splittopskip\ht\strutbox % top baseline for broken footnotes
+ \splitmaxdepth\dp\strutbox \floatingpenalty\@MM
+ \leftskip\z@skip \rightskip\z@skip \spaceskip\z@skip
+ \xspaceskip\z@skip
+ \textindent{#1}\footstrut\futurelet\next\fo@t}
+\def\fo@t{\ifcat\bgroup\noexpand\next \let\next\f@@t
+ \else\let\next\f@t\fi \next}
+\def\f@@t{\bgroup\aftergroup\@foot\let\next}
+\def\f@t#1{#1\@foot}
+\def\@foot{\strut\egroup}
+\def\footstrut{\vbox to\splittopskip{}}
+% \end{macrocode}
+% \LaTeX{} leaves these initializations for the |\footins| insert.
+% \begin{macrocode}
+%\skip\footins=\bigskipamount % space added when footnote is present
+%\count\footins=1000 % footnote magnification factor (1 to 1)
+%\dimen\footins=8in % maximum footnotes per page
+% \end{macrocode}
+%
+% Now we have several options for how to really deal with footnotes.
+% The easy answer is to do them entirely according to \PlainTeX.
+% \begin{macrocode}
+\if@strictfootnotes
+ \let\footnote\@@footnote
+\fi
+% \end{macrocode}
+% The second option is to just use the \LaTeX{} |\footnote| command.
+% This needs no rewriting, of course. The last option is to rewrite
+% \LaTeX's |\footnote| macro to use the \PlainTeX{} format instead of
+% the \LaTeX{} format, which uses an optional argument.
+% \begin{macrocode}
+\if@altfootnotes
+ \def\footnote#1{\latex@footnote[#1]}
+\fi
+% \end{macrocode}
+% \LaTeX{} keeps \PlainTeX's |\footnoterule| as the default.
+% \begin{macrocode}
+%\def\footnoterule{\kern-3\p@
+% \hrule \@width 2in \kern 2.6\p@} % the \hrule is .4pt high
+% \end{macrocode}
+%
+% \subsubsection{Magnification: \env{magnification}}
+%
+% The last part of \PlainTeX{} for which we need to account is
+% magnification.
+% The magnification macros are easily reinstated, either as part of
+% the overall \PlainTeX{} output routine or standalone. Since the
+% |\mag| primitive is not disabled, it \emph{could} still be used in
+% \LaTeX. However, \LaTeX{} does not itself work with |true| units
+% any usage of |\magnification| could do some strange things to your
+% page layouts.
+% \begin{macrocode}
+\if@magnification
+ \def\magnification{\afterassignment\m@g\count@}
+ \def\m@g{\mag\count@
+ \hsize6.5truein\vsize8.9truein\dimen\footins8truein}
+\fi
+% \end{macrocode}
+%
+% This brings us to the end of the main package.
+% \begin{macrocode}
+%</package>
+% \end{macrocode}
+%\Finale
diff --git a/Master/texmf-dist/source/latex/cjw/cjwplain.ins b/Master/texmf-dist/source/latex/cjw/cjwplain.ins
new file mode 100644
index 00000000000..05ea718657e
--- /dev/null
+++ b/Master/texmf-dist/source/latex/cjw/cjwplain.ins
@@ -0,0 +1,20 @@
+\def\batchfile{cjwplain.ins}
+\input docstrip.tex
+
+\preamble
+
+Copyright (C) 1995 by Colin J. Wynne
+
+This file is part of cjwplain, a style providing
+Plain TeX functionality under LaTeX2e.
+
+\endpreamble
+
+% Don't display progress
+\keepsilent
+
+\generateFile{cjwplain.sty}{t}{\from{cjwplain.dtx}{package}}
+\generateFile{cjwpltab.clo}{t}{\from{cjwplain.dtx}{tabbing}}
+\generateFile{cjwplout.clo}{t}{\from{cjwplain.dtx}{output}}
+
+\endinput
generated by cgit v1.2.3 (git 2.39.1) at 2024-11-15 13:44:39 +0000