Debian Codegraph @

texmf-dist/source/latex/mlist/mlist.dtx

raw
1 % \iffalse
2 %<*internal>
3 \iffalse
4 %</internal>
5 %<*readme>
6 _________________
7 The MLIST package
8 v0.6a
9
10 This package defines the commands \newvect, \newmatr, \newfunc,
11 and \newmset for creating macros to typeset vectors, matrices,
12 functions, and sets. For example, Logical indexing can then be
13 used to refer to elements or arguments without hard-coding the
14 symbols in the document:
15
16 \newvect\myvect{V}[elem=a,b,c]
17 \myvect{1,2,3}
18
19 produces
20
21 \left( V_a, V_b, V_c \right)
22
23 All aspects of the typeset appearance can be adjusted, and new
24 `mlists' can be defined to represent other mathematical constructs.
25
26 Consider this package an exploration of separation of form and content
27 for some mathematical constructions. I hope you find it useful.
28
29 ________________________
30 History of major changes
31
32 v0.6: Added the headhook option to accomodate accents to
33 the head symbol. Perhaps a little experimental.
34
35 v0.5: Introduced \newmset for defining mathematical sets;
36 with pre-defined sets \setR, \setC, \setZ, \setN.
37
38 ______________
39 Will Robertson
40 wspr 81 [at] gmail [dot] com
41
42 Copyright 2007-2008
43 Distributed under the LaTeX Project Public License
44 %</readme>
45 %<*internal>
46 \fi
47 \begingroup
48 %</internal>
49 %<*batchfile>
50 \input docstrip.tex
51 \keepsilent
52 \preamble
53
54 __________________________________
55 Copyright (C) 2007 Will Robertson
56
57 License information appended.
58
59
60 \endpreamble
61 \postamble
62
63 Copyright (C) 2007 by Will Robertson <[email protected]>
64
65 Distributable under the LaTeX Project Public License,
66 version 1.3c or higher (your choice). The latest version of
67 this license is at: http://www.latex-project.org/lppl.txt
68
69 This work is "maintained" (as per LPPL maintenance status)
70 by Will Robertson.
71
72 This work consists of the file mlist.dtx
73 and the derived files mlist.pdf,
74 mlist.sty,
75 mlist.cfg, and
76 mlist.ins.
77
78 \endpostamble
79 \askforoverwritefalse
80 \generate{\file{\jobname.sty}{\from{\jobname.dtx}{package}}}
81 \generate{\file{\jobname.cfg}{\from{\jobname.dtx}{defaults}}}
82 %</batchfile>
83 %<batchfile>\endbatchfile
84 %<*internal>
85 \generate{\file{\jobname.ins}{\from{\jobname.dtx}{batchfile}}}
86 \nopreamble\nopostamble
87 \generate{\file{README.txt}{\from{\jobname.dtx}{readme}}}
88 \generate{\file{dtx-style.sty}{\from{\jobname.dtx}{dtx-style}}}
89 \endgroup
90 \immediate\write18{makeindex -s gind.ist -o \jobname.ind \jobname.idx}
91 \immediate\write18{makeindex -s gglo.ist -o \jobname.gls \jobname.glo}
92 %</internal>
93 %
94 %<*driver>
95 \documentclass{ltxdoc}
96 \EnableCrossrefs
97 \CodelineIndex
98 \RecordChanges
99 %\OnlyDescription
100 \usepackage{dtx-style}
101 \begin{document}
102 \DocInput{\jobname.dtx}
103 \end{document}
104 %</driver>
105 %
106 %
107 % \fi
108 %
109 % \GetFileInfo{\jobname.sty}
110 % \CheckSum{0}
111 % \makeatletter
112 %
113 % \title{Logical markup for mathematical lists}
114 % \author{Will Robertson}
115 % \date{\filedate \qquad \fileversion}
116 %
117 % \maketitle
118 %
119 % \tableofcontents
120 % \clearpage
121 %
122 % \section{Introduction}
123 %
124 % One of \LaTeX's lauded features is its separation of form and
125 % content. When writing a document, they say, you can just focus on
126 % the words. Well, that may be well and good for prose, but
127 % mathematicians have frequently been left out in the cold. A funny
128 % situation, considering \LaTeX's main audience.
129 %
130 % Recently, the \pkg{cool} package appeared, marking the first
131 % large-scale format-independent method of writing maths in
132 % \LaTeX. Need to change the brackets surrounding the arguments of
133 % \cmd\sin\ and \cmd\cos, or switch from $\tan^{-1}$ to $\arctan$
134 % throughout an entire document? No worries.
135 %
136 % This sort of initiative will allow much easier transfer of
137 % mathematics from such software packages as Mathematica and Matlab,
138 % if it becomes popular enough.
139 %
140 % This package, \pkg{\jobname}, provides a method to write various
141 % sorts of mathematical lists without having to worry about
142 % formatting. When I say `lists', I mean things like
143 % $\vect{A}=\vect{A}{1,2,3}$; to me as a mechanical engineer, vectors
144 % and matrices; but the package is more general than that, I hope.
145 %
146 % You can use \pkg{\jobname} straight away with
147 % \codeline \cmd\usepackage\texttt{\char`\{\jobname\char`\}}\\
148 % and
149 % \codeline |$\vect{A}=\vect{A}{1,2,3}$|\\
150 % to get the example in the previous paragraph, but \emph{please}
151 % keep reading to learn its more useful features.
152 %
153 % \section{Basic Functionality}\seclabel{vect}
154 %
155 % \subsection{Prototypical example: vector notation}
156 %
157 % \DescribeMacro{\newvect}
158 % This example demonstrates why I wrote this package. The control
159 % sequence \cmd\dist\ is defined as a vector with any number of elements. The
160 % vector and its elements can be subsequently referred to without
161 % hard-coding any of the mathematical symbols or brackets used.
162 % \begin{example}{}
163 % \newvect\dist{D}[elem={r,\phi,\theta}]
164 % \[ \dist \qquad \dist{2}
165 % \qquad \dist{1,2,3} \]
166 % \end{example}
167 %
168 % \subsection{Matrix notation}
169 %
170 % The basic idea of this package is to separate the formatting and
171 % content of mathematical objects. The above example showed vector
172 % notation, and there are three more `data structures' that are defined
173 % `out of the box'.
174 %
175 % \DescribeMacro{\newmatr} The second example is for matrices, which
176 % one might consider quite similar to vectors; but by default the
177 % formatting internally uses \pkg{amsmath}'s |{bmatrix}|
178 % environment. Note that it's a very good idea to give your maths
179 % objects actual \emph{names} rather than shorthands like |\M|---who
180 % knows if you'll still be calling it `M' in ten years time?
181 % \begin{example}{}
182 % \newmatr\massm{M}[elem={m,0;0,m+M}]
183 % \[ \massm \qquad \massm{4}
184 % \qquad \massm{1,2;3,4} \]
185 % \end{example}
186 %
187 % Again, it is possible then to change the brackets used for matrices
188 % throughout an entire document (or collection of documents) without
189 % changing the fundamental markup of the mathematics itself.
190 %
191 % \subsection{Function and `set' notation}
192 %
193 % \DescribeMacro{\newfunc} Finally, here's something a little
194 % different. Vector and matrices are both quite similar. But it takes
195 % only a slight stretch to extend the syntax of this package to things
196 % completely different---such as functions:\label{func-example}
197 % \begin{example}{}
198 % \newfunc\traj{T}[elem={x_0;x,t},index={1;2,3}]
199 % \[ \traj \qquad \traj{2} \qquad \traj{2,3} \]
200 % \end{example}
201 %
202 % For example, Mathematica users may wish to typeset their arguments
203 % with square brackets in their own documents but not for published
204 % papers. To hammer home the point: this can be now done for an entire
205 % document with a simple switch.
206 %
207 % \DescribeMacro{\newmset} Here's an example to further
208 % demonstrate the flexibility of the package.
209 % \begin{example}{}
210 % \newmset\setT{T}
211 % \[ \setT \qquad \setT{2} \qquad \setT{2,3} \]
212 % \end{example}
213 % The \cmd\setR, \cmd\setC, \cmd\setN, and \cmd\setZ\ `msets' are defined
214 % by default for real, complex, natural, and integer numbers.
215 % \begin{example}{}
216 % \[ \bm{x}\in\setR{2,2} \]
217 % \end{example}
218 %
219 % \subsection{Non-specific commands}
220 % Sometimes you don't always want to define new macros for one-off
221 % maths expressions. Accompanying the \cmd\newvect, \cmd\newmatr,
222 % \cmd\newfunc, and \cmd\newmset\ macros are \cmd\vect, \cmd\matr, \cmd\func,
223 % and \cmd\mset, which can be used for single cases:
224 % \begin{example}{}
225 % \[ \vect{V}{a,b,c} \quad
226 % \matr{M}{a,b;c,d} \quad
227 % \func{f}{x_0;x,t} \quad
228 % \mset{D}{m,n} \]
229 % \end{example}
230 % This usage is generally not recommended except in isolated cases.
231 %
232 % \subsection{An example of some actual maths}
233 %
234 % Here's a practical example using \cmd\matr\ that I think demonstrates
235 % well the advantages of using this package:
236 % \begin{example}{}
237 % \newmatr\state{x}
238 % [elem={x_1;\dot x_1},index={1;2}]
239 % \newmatr\dyn{A}
240 % [elem={0,1;-K/M,-C/M},
241 % index={1,2;3,4}]
242 % \[ \dot{\state{}} = \dyn{}\state{} \]
243 % \[ \state{} = \state \]
244 % \[ \dyn{} = \dyn \]
245 % \[ \frac\partial{\partial t}
246 % \state = \dyn \state \]
247 % \end{example}
248 % There's a lot more semantic meaning in that mathematical \TeX\ source than if it were hard-coded without this package.
249 %
250 % \subsection{Elements and indexing}\seclabel{indexing}
251 %
252 % For context for the following explanations, here's our first example again:
253 % \begin{example}{}
254 % \newvect\dist{D}[elem={r,\phi,\theta}]
255 % \[ \dist \qquad \dist{2} \qquad \dist{1,2,3} \]
256 % \end{example}
257 % \newvect\dist{D}[elem={r,\phi,\theta}]
258 %
259 % Each command created like \cmd\dist\ takes zero or one arguments, as shown
260 % above. Unlike regular \LaTeX\ commands with optional arguments, these still
261 % use curly braces, and an empty argument is equivalent (by default) to omitting the argument altogether:
262 % \codeline |\dist{}|~$\to~\dist{}$\quad$\equiv$\quad|\dist|~$\to~\dist$ .
263 %
264 % While numerical indexing is used to extract \opt{elem} items from
265 % the list, \emph{any} argument can be used instead that is passed
266 % through directly:
267 % \codeline |\dist{x^2}|~$\to~\dist{x^2}$ .
268 %
269 % As we saw for the matrix examples, the input may actually be
270 % `two-dimensional'; as well as being comma-separated, elements (and
271 % sets of elements) may also be separated by semi-colons:
272 % \codeline |\dist{x;y;z}|~$\to~\dist{x;y;z}$ .
273 %
274 % Input elements can also be specified with semi-colon separators to
275 % match up with two-dimensional element indexing. There are two ways
276 % to reference these elements: with a linear index from one to the
277 % number of elements in |elem|; or with `implicit indexing' that
278 % starts from one for every semi-colon list. Implicit indices are
279 % prefix with the |@| symbol.\footnote{The implicit indexing syntax
280 % will work \emph{even if} \cmd\makeatletter\ is in effect.} An
281 % example is useful to show how this works:
282 % \begin{example}{}
283 % \newvect\V{V}[elem={a,b,c;x,y,z}]
284 % \[ \V=\V{1,2,3;4,5,6} \]
285 % \[ \V=\V{4,5,6} \]
286 % \[ \V=\V{1,2,3;@1,@2,@3} \]
287 % \end{example}
288 % See \tabref{index-syntax} for a more detailed example of the indexing system.
289 %
290 % \begin{table}
291 % \topcaption{How two-dimension lists elements are indexed. The first row shows the input list via \opt{elem=\{\}}. The second and third rows respectively show the direct index and implicit index required to reference the elements of the list.}
292 % \tablabel{index-syntax}
293 % \centering
294 % \begin{tabular}{@{}l@{\quad\{~}c@{~,~}c@{~,~}c@{~;~}c@{~,~}c@{~;~}c@{~\}}}
295 % \toprule
296 % List & a & b & c & d & e & f \\
297 % Direct index & 1 & 2 & 3 & 4 & 5 & 6 \\
298 % Implicit index & @1 & @2 & @3 & @1 & @2 & @1 \\
299 % \bottomrule
300 % \end{tabular}
301 % \end{table}
302 %
303 % You can also set the default element list when defining a vector. In this case, an empty list argument is not equivalent to omitting one:
304 % \begin{example}{}
305 % \newvect\LL{L}[elem={i,j,k},index={1,2,3}]
306 % $\LL$ \quad vs.\quad $\LL{}$
307 % \quad vs.\quad $\LL{1,2} $
308 % \end{example}
309 % This was shown originally in the function example on page~\pageref{func-example}.
310 %
311 % \subsection{Shorthand indices and empty elements}
312 %
313 % Shorthands are defined to allow things that
314 % \emph{aren't} elements into the list.
315 % In this example, the semi-colon separation is shown with the
316 % \cmd\dots\ shorthand `|:|' and the \cmd\cdot\ shorthand `|.|'.
317 % \begin{example}{}
318 % \newvect\mm{M}
319 % \[ \mm=\mm{1,2,:,5;a,b,.,d} \]
320 % \end{example}
321 %
322 % More shorthands will be defined in the future via
323 % an extensible mechanism that does not yet exist.
324 %
325 % Empty arguments are ignored (where `empty' $\equiv$ empty or whitespace).
326 % \begin{example}{}
327 % \newvect\nset{N}
328 % \[ \nset=\nset{1,,,3, ,5;;,2,;} \]
329 % \end{example}
330 %
331 % \subsection{\cmd\dots}
332 %
333 % Finally, some macros are defined in order to be able to typeset
334 % unknown ranges ($1,2,\dots,N$) without hard-coding the symbol of the
335 % maximum element. Similarly, it is also useful to denote a
336 % `mid-range' element somewhere inside the ellipses. Note the equivalence in defining the mid and last elements and indexing them.
337 % \begin{example}{}
338 % \newvect\W{W}[elem={i,j,k,:,w}]
339 % \newvect\Y{Y}[elem={i,j,k,:,p,:,y}]
340 % \[ \W=\W{1,2,:,\LAST-1,\LAST} \]
341 % \[ \Y=\Y{1,2,:,\MID,:,\LAST} \]
342 % \end{example}
343 %
344 % \noindent The \cmd\MID\ and \cmd\LAST\ control sequences are `implicit' for two dimensional lists:
345 %
346 % \begin{example}{}
347 % \newvect\X{X}[%
348 % elem={i,j,k,:,m,:,w;ii,jj,kk,:,mm,:,ww}]
349 % \[ \X{1,:,\MID,:,\LAST;@1,:,\MID,:,\LAST} \]
350 % \[ \X{1,2,3;4,5,6} \]
351 % \end{example}
352 %
353 % Finally, note well above that any |{:,X}| pairs for mid/last element definitions do \emph{not} add to the direct index numbering.
354 %
355 %
356 % \section{Generalisation}
357 %
358 % \DescribeMacro{\newmlist}
359 % \DescribeMacro{\renewmlist}
360 % The \cmd\newvect\ command that has been
361 % often shown previously is an example of a macro created with
362 % \cmd\newmlist, which can be considered something like an
363 % \emph{instance generator} for the types of lists we're dealing with.
364 % Use \cmd\renewmlist\ if the command is already in use.
365 %
366 % \cmd\newmlist\cs{<list>}\oarg{list options} creates macros
367 % \cs{<list>} and \cs{new<list>} (and \cs{renew<list>}) that are
368 % analogous to \cmd\vect\ and \cmd\newvect\ seen in
369 % \secref{vect}. \meta{list options} can contain any of those seen so
370 % far in previous \cmd\newvect\ macros, plus more to be introduced
371 % soon.
372 %
373 % In the examples to follow, \cmd\vect\ and \cmd\newvect\ are often
374 % still used to demonstrate various options, overriding the defaults.
375 %
376 % \subsection{Head and element formatting}
377 %
378 % The \opt{headcmd} and \opt{elemcmd} options are used to alter the
379 % formatting of the vector symbol and its elements. They are passed
380 % \emph{macros} that take, respectively, one and two arguments that
381 % define the formatting.
382 %
383 % For \cmd\vect, |headcmd=\mathbf| and |elemcmd=\mlistsub|, where
384 % \codeline \cmd\mlistsub |{#1}{#2}|~$\to$~|#1_{#2}|.\\
385 % \cmd \mlistsup, the
386 % analogous command for creating superscripts, is defined by the
387 % package in case you need it.
388 %
389 % In this example, the vector symbol is formatted with an arrow accent,
390 % and the elements are exactly as specified in \opt{elem}.
391 % \begin{example}{}
392 % \newvect\A{A}[%
393 % headcmd=\vec,
394 % elemcmd=\mlistelem,
395 % elem={a,b,c,d}]
396 %
397 % \[\A \qquad \A{1,2,3,4} \qquad \A{3} \]
398 % \end{example}
399 %
400 % The \cmd\mlistelem\ command (with its friend, \cmd\mlisthead) is defined as follows:
401 % \codeline\cmd\mlistelem|{#1}{#2}|~$\to$~|#2|
402 % \codeline\cmd\mlisthead|{#1}|~$\to$~|#1|
403 %
404 % In \secref{indexing}, we saw that list indexing doesn't \emph{have}
405 % to be numerical. In this example, the use of \cmd\@alph\ allows
406 % alphabetic subscripts but non-numeric indices will produce errors:
407 % \begin{example}{}
408 % \makeatletter
409 % \newcommand\subalph[2]{#1_{\@alph{#2}}}
410 % \makeatother
411 % \newvect\A{A}[%
412 % elemcmd=\subalph]
413 %
414 % \[ \A{1,2,3,4} \qquad \A{3} \]
415 % \end{example}
416 %
417 % \subsection{Head prefixing}
418 %
419 % Applying (things like) accents to \pkg{mlist}s will often give the wrong result:
420 % \begin{example}{}
421 % \newfunc\A{A}
422 % $\A{1}$ \qquad $\bar{\A{1}}$
423 % \end{example}
424 %
425 % The solution for this involves defining a prefix for the head element with the |headhook| option:
426 % \begin{example}{}
427 % \newfunc\A{A}
428 % \newcommand\baraccent[1]{%
429 % {\mlistsetup{headhook=\bar}#1}}
430 % $\A{1}$ \qquad $\baraccent{\A{1}}$
431 % \end{example}
432 %
433 % This method can also be used to apply \emph{post}fix elements as well:
434 % \begin{example}{}
435 % \newfunc\A{A}
436 % \newcommand\postprime[1]{#1'}
437 % \newcommand\primeaccent[1]{%
438 % {\mlistsetup{headhook=\postprime}#1}}
439 % $\A{1}$ \qquad $\A{1}'$ \qquad $\primeaccent{\A{1}}$
440 % \end{example}
441 %
442 % The user interface for these features is rather rough and needs more thought; it might be better if \cmd\mlistsetup\ did not need to be called in the accent definition.
443 %
444 % For these reasons the |headhook| feature should be considered a little experimental and possibly subject to change in the future. It should certainly be accompanied by a variety of `\cmd\baraccent'--like commands, unless \cmd\bar\ can be patched to recognise such situations automatically (this seems unlikely).
445 %
446 % \subsection{Delimiter formatting}
447 %
448 % The \opt{wrap} and \opt{wrapone} options are used to change the way
449 % elements are displayed together. Each take two
450 % arguments to define the opening and closing material.
451 %
452 % By default, if there is more than one element, it is surrounded by
453 % square brackets: |wrap=[]|. A single element is typeset naked:
454 % |wrapone={}{}|. If \opt{wrapone} is called with no `|=|' argument, it
455 % takes the same value as \opt{wrap}.
456 %
457 % In this example, the brackets around the sets are changed to
458 % parentheses, including single elements. The behaviour for a single
459 % element is then changed to use angle brackets:
460 % \begin{example}{firstline=2}
461 % \parskip1ex
462 % \newvect\Q{Q}
463 % $\Q + \Q{1} + \Q{a,b}$
464 %
465 % \newvect\QQ{Q}[wrap=(),wrapone]
466 % $\QQ + \QQ{1} + \QQ{a,b}$
467 %
468 % \newvect\QQQ{Q}[wrapone={\left<}{\right>}]
469 % $\QQQ + \QQQ{1} + \QQQ{a,b}$
470 % \end{example}
471 %
472 % Rather than defining the open/close material around a set of
473 % elements, it can often be necessary to define a macro that defines
474 % the formatting of the set. The \opt{wrapcmd} and \opt{wraponecmd}
475 % options are used for this purpose, and take as argument a single
476 % macro that accepts two arguments: the list head and the
477 % list elements, respectively.
478 %
479 % \begin{example}{firstline=2}
480 % \parskip1ex
481 % \newvect\R{R}
482 % $\R + \R{1} + \R{a,b}$
483 %
484 % \newcommand\mywrap[2]{\langle{\color{red}#2}\rangle_{#1}}
485 % \newvect\RR{R}[wrapcmd=\mywrap,wraponecmd]
486 % $\RR + \RR{1} + \RR{a,b}$
487 %
488 % \newcommand\mywrapone[2]{\{{\color{green}#2}\}^{#1}}
489 % \newvect\RRR{R}[
490 % wrapcmd=\mywrap,
491 % wraponecmd=\mywrapone]
492 % $\RRR + \RRR{1} + \RRR{a,b}$
493 % \end{example}
494 %
495 % \pkg{mlist} provides some example commands for this purpose:\par\nobreak\noindent
496 % \begingroup\ttfamily
497 % \def\mean#1{\string#1 & \expandafter\parse@meaning\meaning#1\@nil}
498 % \def\parse@meaning #1:#2->#3\@nil{\detokenize{#3}\\}
499 % \begin{tabular}{l@{$~\to~$}l}
500 % \mean\mlistnowrap
501 % \mean\mlistparen
502 % \mean\mlistbrack
503 % \mean\mlistbrace
504 % \mean\mlistangle
505 % \mean\mlistheadparen
506 % \mean\mlistheadbrack
507 % \mean\mlistheadbrace
508 % \mean\mlistheadangle
509 % \end{tabular}
510 % \endgroup
511 %
512 % \subsection{Separator formatting}
513 %
514 % The \opt{sep} and \opt{sepsep} options take one argument that is inserted between items in comma and semi-colon lists, respectively. For example, in the \cmd\matr\ list these are defined with |[sep=&, sepsep=\\]|.
515 %
516 % Here's another example:
517 % \begin{example}{}
518 % \newvect\mysum{S}[
519 % sep=+,
520 % wrap={}{},
521 % dots=\cdots]
522 % $\mysum{a,b,c,d,:,p,q,r,s,:,x,y,z}$
523 % \end{example}
524 %
525 % \subsection{Global options definition}
526 %
527 % \DescribeMacro{\mlistsetup} If options aren't specified in
528 % \cmd\newmlist\ they are inherited from the global defaults, which
529 % may be adjusted with \cmd\mlistsetup\marg{mlist options}. The defaults
530 % are shown in \secref[vref]{defaults}.
531 %
532 % \subsection{Redefining \cmd\vect, \cmd\matr, \cmd\func, \cmd\mset}
533 %
534 % This package makes little claim for being imminently usable for most
535 % people \emph{out of the box}. I figure there's just too much
536 % variety; people need to define their own `mlists' with \cmd\newmlist.
537 %
538 % If changes to the mlists provided by default with this package
539 % are required (as they will be if any aspect of their formatting needs
540 % to be adjusted), simply create a local |mlist.cfg| file with different
541 % defintions or even just copy them from \secref[vref]{defaults}
542 % in a \cmd\renewmlist.
543 %
544 % \section{TODO}
545 % \begin{itemize}
546 % \item changing features of (or adding features to, rather) `newvect'
547 % constructions, rather than overwriting them.
548 % \item generalise escaping strings like |:|$\to$|\dots|, etc.
549 % \item accents and appended/prepended material migrating inside the
550 % head (and/or each element) argument.
551 % \item optional arguments for \cmd\MID\ and \cmd\LAST\ (for explicit indexing).
552 % \item coercing one type of list into another
553 % (and retrieving non-wrapped lists as a special case)
554 % \item use ltx3!
555 % \end{itemize}
556 %
557 %\iffalse
558 %<*defaults>
559 %\fi
560 %
561 % \section{The default definitions}
562 % \seclabel{defaults}
563 %
564 % This is the code that appears in the default configuration file \pkg{mlist.cfg} to set up the
565 % default options and mlists. Edit a copy of \pkg{mlist.cfg} in a local location (in a local \texttt{texmf} tree or on a per-document basis) to change these definitions and to create your own mlists.
566 %
567 % \subsection{Global options}
568 % These are inherited by mlists that do not explicitly define their own respective
569 % options. Geared towards \cmd\vect, essentially.
570 % \begin{macrocode}
571 \mlistsetup{%
572 sep={,},
573 sepsep={;},
574 wrapcmd=\mlistparen,
575 wrapone={}{},
576 dots=\dots,
577 dot={\,\cdot\,},
578 elemcmd=\mlistelem,
579 headcmd=\mathbf,
580 index={},
581 headhook={},
582 }
583 % \end{macrocode}
584 %
585 % \subsection{Provided mlists}
586 %
587 % Redefine these with \cmd\renewmlist\ to effect your own formatting
588 % of these mlists. Note that these assume certain global defaults (see
589 % above), so some aspects of their formatting can be changed with
590 % \cmd\mlistsetup.
591 %
592 % \begin{macro}{\vect}
593 % By default |\vect{V}=\vect{V}{a,b,c}| $\to$ $\vect{V}=\vect{V}{a,b,c}$.
594 % \begin{macrocode}
595 \newmlist\vect[
596 wrapcmd=\mlistparen,
597 wraponecmd=\mlistnowrap,
598 elemcmd=\mlistsub,
599 headcmd=\mathbf,
600 ]
601 % \end{macrocode}
602 % \end{macro}
603 %
604 % \begin{macro}{\matr}
605 % By default |\matr{M}=\matr{M}{a,b;c,d}| $\to$ $\matr{M}=\matr{M}{a,b;c,d}$.
606 % \begin{macrocode}
607 \newmlist\matr[
608 sep=&,
609 sepsep=\\,
610 wrap={\begin{bmatrix}}
611 {\end{bmatrix}},
612 ]
613 % \end{macrocode}
614 % \end{macro}
615 %
616 % \begin{macro}{\func}
617 % By default, |\func{f}=\func{f}{x,y,z}| $\to$ $\func{f}=\func{f}{x,y,z}$.
618 % \begin{macrocode}
619 \newmlist\func[
620 headcmd=\mlisthead,
621 wrapcmd=\mlistheadparen,
622 wraponecmd,
623 ]
624 % \end{macrocode}
625 % \end{macro}
626 %
627 % \begin{macro}{\mset}
628 % By default, |\mset{N}{2,3}| $\to$ $\mset{N}{2,3}$.
629 % \begin{macrocode}
630 \newmlist\mset[
631 headcmd=\mathbb,
632 sep=\times,
633 wrapcmd=\mlistsup,
634 wraponecmd,
635 ]
636 % \end{macrocode}
637 % \changes{v0.5}{2007/06/06}{Added.}
638 % \end{macro}
639 % Sets for real, complex, natural, and integer numbers, respectively:
640 % \begin{macrocode}
641 \newmset\setR{R}
642 \newmset\setC{C}
643 \newmset\setN{N}
644 \newmset\setZ{Z}
645 % \end{macrocode}
646 %
647 %\iffalse
648 %</defaults>
649 %\fi
650 %
651 % \section{Prior art}
652 %
653 % With \CTAN\ getting so big these days, it's quite necessary to
654 % undertake extensive literature reviews before writing your own
655 % package. I've been burned before, spending a couple of days playing
656 % with ideas and then realising that someone's already done what I wanted.
657 % This time I looked before I jumped.
658 %
659 % \paragraph{easyvector}
660 % The most similar package to this one (that I could find) is
661 % \pkg{easyvector}.
662 % \note{\url|http://tug.ctan.org/cgi-bin/ctanPackageInformation.py?id=easyvector|}
663 % Superficially, there are a number of similarities, but it didn't
664 % quite do what I wanted. Here's an example demonstrating the creation
665 % of new vector macros:
666 % \begin{example}{}
667 % \newvector[Z,\mathbf{Z}]{X}
668 % \[ \X = (\X[i,j;k]) \qquad
669 % \X[1,2,3] \neq \X![1,2,3] \]
670 % \end{example}
671 %
672 % Note that each macro refers only to a single vector element. This is a package
673 % to simplify input (and abstract formatting) of vectors with complex notation.
674 % \pkg{easyvector} also allows you to customise the form of the
675 % vectors it produces, but this is rather inflexible without a good
676 % deal of work.
677 % \begin{example}{}
678 % \def\myindex[#1,#2,#3]{_{#1_{#2}}^{#3}}
679 % \newcustomvector[\mathtt{b},\mathbf{b}]{bb}\myindex
680 % \[ \bb \qquad \bb[1,2,3] \qquad \bb[3,2,1] \]
681 % \end{example}
682 %
683 % Note also that the main symbol (`$\mathtt{b}$', here) is not
684 % available even in this case, so prepended sub-/superscripts are not possible.
685 %
686 % Has the interesting option to reference matrix row/column sub-vectors: (careful to ensure \cmd\makeatother\ manually)
687 % \makeatother
688 % \begin{example}{}
689 % \newvector(W)[wvec]
690 % \[ \wvec = (\wvec[@,j]) \qquad \wvec[a,b;@] \]
691 % \end{example}
692 % \makeatletter
693 %
694 % I like this idea of `shorthand' symbols.
695 %
696 % \paragraph{\pkg{vector}}
697 % The package \pkg{vector}
698 % \note{\url|http://tug.ctan.org/cgi-bin/ctanPackageInformation.py?id=vector|}
699 % provides a few commands for setting vector and matrix symbols, shown in \tabref{vectorpkg}.
700 % Just a few basic macros to simplify input, but not really to separate form and content.
701 %
702 % \pkg{vector} provided the inspiration to add |\dots| ideas to this package. It also highlights that
703 % something specific should be done with accents in general and provide a content macro for unit vectors.
704 %
705 % \begin{table}
706 % \centering
707 % \begin{tabular}{@{}ccccccc@{}}
708 % \toprule
709 % |\bvec| & |\buvec| & |\svec| & |\suvec| & |\uvec| & |\uuvec| & |\irvec| \\
710 % \bvec{a} & \buvec{a} & \svec{a} & \suvec{a} & \uvec{a} & \uuvec{a} & \irvec{a} \\
711 % \bottomrule
712 % \end{tabular}
713 % \caption{Commands defined by the \pkg{vector} package.
714 % \cmd\uvec\ and \cmd\uuvec\ can be configured to produce an under-tilde instead.
715 % \cmd\irvec\ takes an optional argument for `$n$' and can be configured (globally) to begin from a different index.}
716 % \label{tab:vectorpkg}
717 % \end{table}
718 %
719 % \paragraph{hhtensor}
720 % The hhtensor package
721 % \note{\url|http://tug.ctan.org/cgi-bin/ctanPackageInformation.py?id=hhtensor|}
722 % provides a few content-based macros with global options to specify their appearance.
723 % \cmd\vec, \cmd\matr\, and \cmd\tens\ are provided for vectors, matrices, and tensors, with
724 % mathematical symbols \cmd\dcdot\ to denote double scalar products, and \cmd\trans\ for printing
725 % an upright superscript `T' to denote the transpose operator.
726 %
727 % This package was the inspiration for pre-defined macros for specific meanings.
728 % While I approve of defining the transpose symbol, it's outside the scope of \pkg\jobname.
729 %
730 % \paragraph{Other tensor packages}
731 % There are three main packages for typesetting tensors.
732 % These start to stray from the interest of \pkg\jobname.
733 % The relevant packages are
734 % \pkg{tensind},
735 % \note{\url|http://tug.ctan.org/cgi-bin/ctanPackageInformation.py?id=tensind|}
736 % \pkg{tensor},
737 % \note{\url|http://tug.ctan.org/cgi-bin/ctanPackageInformation.py?id=tensor|}
738 % and \pkg{mattens}.
739 % \note{\url|http://tug.ctan.org/cgi-bin/ctanPackageInformation.py?id=mattens|}
740 % Each focuses on typesetting various forms of tensor notation, which differ from what
741 % this package is trying to do. Integration between this package and those three above are possible,
742 % but has not yet been investigated.
743 %
744 %
745 % \StopEventually{}
746 %
747 % \clearpage
748 % \part{\pkg{\jobname} implementation}
749 %\iffalse
750 %<*package>
751 %\fi
752 %
753 % \section{Setup code}
754 % This is the package.
755 %
756 % \begin{macrocode}
757 \ProvidesPackage{mlist}
758 [2008/08/15 v0.6a Typesetting maths lists]
759 % \end{macrocode}
760 %
761 % \PrintChanges
762 % \changes{v0.1}{27/04/2007}{Code tidy up; first decent version.}
763 % \changes{v0.2}{30/04/2007}{More documentation; list indexing not finalised.}
764 % \changes{v0.3}{2007/05/02}{List indexing decided; empty arguments ignored.}
765 % \changes{v0.4}{2007/05/26}{Tidied things up a little bit.}
766 %
767 % \subsection{User shorthands}
768 %
769 % For |headcmd|:
770 % \begin{macrocode}
771 \let\mlisthead\@firstofone
772 % \end{macrocode}
773 % For |elemcmd|:
774 % \begin{macrocode}
775 \let\mlistelem\@secondoftwo
776 \newcommand\mlistsub [2]{#1_{#2}}
777 \newcommand\mlistsup [2]{#1^{#2}}
778 % \end{macrocode}
779 % For |wrapcmd|:
780 % \begin{macrocode}
781 \let\mlistnowrap\@secondoftwo
782 \newcommand\mlistparen[2]{\left(#2\right)}
783 \newcommand\mlistbrack[2]{\left[#2\right]}
784 \newcommand\mlistbrace[2]{\left\{#2\right\}}
785 \newcommand\mlistangle[2]{\left<#2\right>}
786 \newcommand\mlistheadparen[2]{#1\left(#2\right)}
787 \newcommand\mlistheadbrack[2]{#1\left[#2\right]}
788 \newcommand\mlistheadbrace[2]{#1\left\{#2\right\}}
789 \newcommand\mlistheadangle[2]{#1\left<#2\right>}
790 % \end{macrocode}
791 %
792 %
793 %
794 % \subsection{Packages}
795 % \begin{macrocode}
796 \RequirePackage{xkeyval,ifmtarg}
797 % \end{macrocode}
798 % \subsection{Code we need}
799 % Conditionals and counters and things:
800 % \begin{macrocode}
801 \newif\if@mlist@notfirst@
802 \newif\if@mlist@implicit@
803 \newcount\mlist@elem@N
804 \def\@gobblenil#1\@nil{}
805 % \end{macrocode}
806 % Some specific things:
807 % \begin{macrocode}
808 \newif\if@mlist@warn
809 % \end{macrocode}
810 % Semi-colon delimited iteration (adapted from ltx2e).
811 % \begin{macrocode}
812 \long\def\@sfor#1:=#2\do#3{%
813 \expandafter\def\expandafter\@sfortmp\expandafter{#2}%
814 \ifx\@sfortmp\@empty\else
815 \expandafter\@sforloop#2;\@nil;\@nil\@@#1{#3}%
816 \fi}
817 \long\def\@sforloop#1;#2;#3\@@#4#5{%
818 \def#4{#1}%
819 \ifx #4\@nnil \else
820 #5%
821 \def#4{#2}%
822 \ifx #4\@nnil \else
823 #5%
824 \@siforloop #3\@@#4{#5}%
825 \fi
826 \fi}
827 \long\def\@siforloop#1;#2\@@#3#4{%
828 \def#3{#1}%
829 \ifx #3\@nnil
830 \expandafter\@fornoop
831 \else
832 #4\relax\expandafter\@siforloop
833 \fi
834 #2\@@#3{#4}}
835 % \end{macrocode}
836 %
837 % ltx3-inspired syntax. \cmd\def@c\ still needs arguments to be supplied to it.
838 % \begin{macrocode}
839 \providecommand\let@cc[2]{%
840 \expandafter\let\csname#1\expandafter\endcsname\csname#2\endcsname}
841 \providecommand\def@c[1]{%
842 \expandafter\def\csname#1\endcsname}
843 \providecommand\def@co[2]{%
844 \def@c{#1\expandafter}\expandafter{#2}}
845 % \end{macrocode}
846 % \begin{macro}{\mlist@optarg}
847 % Macro to simplify optional argument parsing.
848 % \begin{macrocode}
849 \newcommand\mlist@optarg[1]{\@ifnextchar[{#1}{#1[]}}%]
850 % \end{macrocode}
851 % \changes{v0.4}{2007/05/26}{Implemented to simplify optional arg processing.}
852 % \changes{v0.6a}{2008/08/15}{Renamed to avoid clashing with \pkg{breqn}'s command of the same name.}
853 % \end{macro}
854 % Shorthand to test for optional brace arguments:
855 % \begin{macrocode}
856 \newcommand\@ifnextbrace{\expandafter\@ifnextchar\bgroup}
857 % \end{macrocode}
858 % Shorthand string definitions for ifx tests:
859 % \begin{macrocode}
860 \def\mlist@colon{:}
861 \def\mlist@period{.}
862 % \end{macrocode}
863 % This is for checking for |@| with an `other' catcode:
864 % \begin{macrocode}
865 \makeatother
866 \expandafter\def\csname mlist@ampersat\endcsname{@}
867 \makeatletter
868 % \end{macrocode}
869 % (`Ampersat' is a name for |@| I find amusing. That symbol doesn't seem to
870 % have a definitive official name.)
871 %
872 % \section{keyval options}
873 %
874 % \begin{macro}{\mlistsetup}
875 % \darg{keyval options}
876 % User command to set global defaults for mlists.
877 % \begin{macrocode}
878 \newcommand\mlistsetup[1]{\setkeys[mlist]{sym}{#1}}
879 % \end{macrocode}
880 % \end{macro}
881 %
882 % \pkg{xkeyval} makes it easy for us to define a whole slew of options
883 % that simply save their argument to a macro.
884 % \begin{macrocode}
885 \define@cmdkeys[mlist]{sym}[mlist@]{%
886 symbol,index,sep,sepsep,elem,
887 elemcmd,headcmd,dot,dots,headhook}
888 % \end{macrocode}
889 %
890 % \subsection{Wrapping}
891 % The wrapcmd options also just save their argument, but wraponecmd
892 % takes wrapcmd as a default:
893 % \begin{macrocode}
894 \define@key[mlist]{sym}{wrapcmd}{\let\mlist@wrap#1}
895 \define@key[mlist]{sym}{wraponecmd}[\mlist@wrap]{\let\mlist@wrapone#1}
896 % \end{macrocode}
897 % The wrap and wrapone options need a bit more logic in them:
898 % \begin{macrocode}
899 \define@key[mlist]{sym}{wrap}{%
900 \def\mlist@wrap##1##2{\@firstoftwo#1##2\@secondoftwo#1}}
901 \define@key[mlist]{sym}{wrapone}[]{%
902 \ifx\relax#1\relax
903 \let\mlist@wrapone\mlist@wrap
904 \else
905 \def\mlist@wrapone##1##2{\@firstoftwo#1##2\@secondoftwo#1}%
906 \fi}
907 % \end{macrocode}
908 %
909 % \subsection{Element definition}
910 %
911 % Takes as input a two dimension list with comma-separated elements
912 % and semicolon-separated lists of elements:
913 % \codeline |#1| $\to$ |{1,2,3;4,5;6}| .
914 %
915 % Elements could be numerical or arbitrary \TeX\ code.
916 % \begin{macrocode}
917 \define@key[mlist]{sym}{elem}{%
918 % \end{macrocode}
919 % It would be easier for all involved if I used specific counters for
920 % the following. For now, they're generic and harder to comprehend in
921 % six months:
922 % \begin{description}[nolistsep]
923 % \item[\cmd\@tempcnta] Number of semicolon list.
924 % \item[\cmd\@tempcntb] Element number of this comma list.
925 % \end{description}
926 % \begin{macrocode}
927 \@tempcnta\z@
928 \@tempcntb\z@
929 \mlist@elem@N\z@
930 % \end{macrocode}
931 % Iterate over every semicolon list. Set \cs{if@tempswa} true only
932 % after every |:| element, for each semicolon list. Maybe we should
933 % use a counter for this, instead?
934 % \begin{macrocode}
935 \@sfor\@jj:=#1\do{%
936 \advance\@tempcnta\@ne
937 \@tempcntb\z@
938 \@tempswafalse
939 % \end{macrocode}
940 % Iterate over every comma list:
941 % \begin{macrocode}
942 \@for\@ii:=\@jj\do{%
943 % \end{macrocode}
944 % If the element is |:|, set the switch for the next comma-iteration.
945 % \begin{macrocode}
946 \ifx\@ii\mlist@colon
947 \@tempswatrue
948 \else
949 % \end{macrocode}
950 % If the previous element was |:|, save the `last' element: (TODO: add
951 % error check for too many |:|). If we've already defined the `last'
952 % element and run into |:| again, that means we actually wanted `mid'
953 % so make the redefinition. This should only happen once anyway so it
954 % can occur every time \cs{if@tempswa} is true.
955 % \begin{macrocode}
956 \if@tempswa
957 \let@cc{mlist@\the\@tempcnta @mid}{mlist@\the\@tempcnta @last}%
958 \def\@tempa{\def@c{mlist@\the\@tempcnta @last}}%
959 \expandafter\@tempa\expandafter{\@ii}%
960 \else
961 % \end{macrocode}
962 % Otherwise, bump up the counters and define the elements. There are
963 % two definitions we used for the element indexing. As an example, if
964 % we're up to index $(3,1)$ of a list like $\{a,b,c;d,e;f\}$ then we define
965 % \codeline |\mlist@3@1| $\to$ elem $f$ , and
966 % \codeline |\mlist@@6| $\to$ elem $f$ . \\
967 % (Index $3,1$ is the sixth in the list.)
968 % \begin{macrocode}
969 \advance\@tempcntb\@ne
970 \advance\mlist@elem@N\@ne
971 \def@co{mlist@\the\@tempcnta @\the\@tempcntb}{\@ii}%
972 \def@co{mlist@@\the\mlist@elem@N}{\@ii}%
973 \fi
974 \fi}%
975 % \end{macrocode}
976 % If swa is true, we are in an element after a |:|, so turn it off.
977 % TODO: I guess no more regular elements can turn up anyway so this is
978 % probably overkill!
979 % \begin{macrocode}
980 \if@tempswa\@tempswafalse\fi}}
981 % \end{macrocode}
982 % \changes{v0.3}{2007/05/09}{Tidied up the \cmd\MID/\cmd\LAST\ element
983 % stuff a bit.}
984 %
985 % \subsection{Shorthand definitions}
986 %
987 % Currently defunct:
988 % \begin{macrocode}
989 \define@key[mlist]{sym}{shorthand}{%
990 \mlist@def@shorthand#1%
991 \def@c{mlist@@\@tempa\expandafter}\expandafter{\@tempb}}
992 % \end{macrocode}
993 %
994 % \begin{macrocode}
995 \def\mlist@def@shorthand#1#{%
996 \def\@tempa{\string#1}
997 \def\@tempb}
998 % \end{macrocode}
999 %
1000 % \section{List indexing}
1001 %
1002 % How do I want indexing to work? Originally, the input was a one
1003 % dimensional list with the output directly following the indexing:
1004 % \begin{verbatim}
1005 % A = {a1,a2,a3,a4}
1006 % A{1,2;3,4} == {a1,a2;a3,a4}
1007 % \end{verbatim}
1008 % I tried a couple of other things, but went back to this idea plus the
1009 % implicit indexing.
1010 %
1011 % \begin{macro}{\@mlist}
1012 % \darg{keyval options}
1013 % This is the \cmd\setkeys\ wrapper, which
1014 % sets macros for us in various ways depending on how it has been
1015 % called and then moves on to do the actual list indexing,
1016 % extraction, and typesetting.
1017 % \begin{macrocode}
1018 \newcommand\@mlist[1]{%
1019 \begingroup
1020 \let\mlist@list@elems\@empty
1021 \@tempcnta\@ne
1022 \mlist@count
1023 % \end{macrocode}
1024 % The \cmd\setkeys\ code to extract our mlist is flanked by some scary
1025 % code to protect ourselves inside things like |{array}|
1026 % environments. Thanks Morten.
1027 % \begin{macrocode}
1028 \iffalse{\fi\ifnum0=`}\fi
1029 \setkeys[mlist]{sym}{#1}%
1030 \ifnum0=`{\fi\iffalse}\fi
1031 % \end{macrocode}
1032 % These macros only exist inside an mlist index:
1033 % \begin{macrocode}
1034 \def\MID{\csname mlist@\the\@tempcnta @mid\endcsname}%
1035 \def\LAST{\csname mlist@\the\@tempcnta @last\endcsname}%
1036 % \end{macrocode}
1037 % If there are no optional arguments, typeset the plain vector symbol.
1038 % \begin{macrocode}
1039 \ifx\@empty\mlist@index\relax
1040 \def\mlist@list{\mlist@headcmd{\mlist@headhook\mlist@symbol}}%
1041 \else
1042 % \end{macrocode}
1043 % Otherwise, iterate over every semicolon-separated list.
1044 % \begin{macrocode}
1045 \@tempcnta\z@
1046 \@tempswafalse
1047 \@sfor\@jj:=\mlist@index\do{%
1048 \let\@jj\@jj
1049 \advance\@tempcnta\@ne
1050 \mlist@count
1051 \expandafter\@ifmtarg\expandafter{\@jj}{}% ignore if empty
1052 {\if@tempswa\mlist@add\mlist@sepsep\fi
1053 \@tempswatrue}%
1054 % \end{macrocode}
1055 % And (sub-)iterate over every comma-separated list:
1056 % \begin{macrocode}
1057 \@tempcntb\z@
1058 \@mlist@notfirst@false
1059 \@for\@ii:=\@jj\do{%
1060 \let\@ii\@ii
1061 \advance\@tempcntb\@ne
1062 % \end{macrocode}
1063 % Now grab the symbol we're up to.
1064 % When it's empty:
1065 % \begin{macrocode}
1066 \expandafter\@ifmtarg\expandafter{\@ii}%
1067 {\advance\@tempcntb\m@ne
1068 \mlist@add{\@gobble}}
1069 % \end{macrocode}
1070 % When it's not empty:
1071 % \begin{macrocode}
1072 {\if@mlist@notfirst@\mlist@add\mlist@sep\fi
1073 \@mlist@notfirst@true
1074 \ifx\@ii\mlist@colon
1075 \mlist@add{\mlist@dots\@gobble}%
1076 \else
1077 \ifx\@ii\mlist@period
1078 \mlist@add{\mlist@dot\@gobble}%
1079 \else
1080 \mlist@add{\mlist@elemcmd{\mlist@headhook\mlist@symbol}}%
1081 \fi
1082 \fi}%
1083 % \end{macrocode}
1084 % Parse the index to see if it's an implicit reference of the form
1085 % |@|$n$. @mlist@implicit@ is true if the index is implicit: (and we assume |@|
1086 % will be only used in an implicit-indexing context. TODO: fix this!)
1087 % \begin{macrocode}
1088 \mlist@parse@implicit
1089 \if@mlist@implicit@
1090 \mlist@add@ifcs{%
1091 mlist@\the\@tempcnta @\expandafter\@gobble\@ii
1092 }{\@ii}%
1093 \else
1094 % \end{macrocode}
1095 % Because \cmd\@ii, \cmd\@jj\ can contain any possible index that
1096 % might be passed through the macro, we use eTeX's \cmd\detokenize\ to
1097 % prevent expansion of any weird argument that might be given.
1098 % \begin{macrocode}
1099 \mlist@add@ifcs{%
1100 mlist@@\expandafter\detokenize\expandafter{\@ii}%
1101 }{\@ii}%
1102 \fi
1103 }%
1104 % \end{macrocode}
1105 % That was the end of comma-separated iteration.
1106 % \begin{macrocode}
1107 }%
1108 % \end{macrocode}
1109 % That was the end of semicolon-separated iteration.
1110 % \begin{macrocode}
1111 \def\mlist@list{%
1112 \mlist@wrap{\mlist@headcmd{\mlist@headhook\mlist@symbol}}
1113 {\mlist@list@elems}}%
1114 % \end{macrocode}
1115 % If only have a single element:
1116 % \begin{macrocode}
1117 \ifnum\@tempcnta=\@ne
1118 \ifnum\@tempcntb=\@ne
1119 \let\mlist@wrap\mlist@wrapone
1120 \fi
1121 \fi
1122 \fi
1123 \mlist@list
1124 \endgroup}
1125 % \end{macrocode}
1126 % TODO: move \cmd\mlist@list\ after the endgroup?
1127 % \changes{v0.4}{2007/05/17}{Eliminated \cmd\global\ no longer
1128 % required since I separated the list creation and typesetting.}
1129 % \changes{v0.4}{2007/05/19}{Simplified `single index' code.}
1130 % \changes{v0.5}{2007/06/06}{Added \cmd\mlist@headcmd\ command to the \cmd\mlist@symbol inside \cmd\mlist@wrap\ (fixed bug with \cmd\mset)}
1131 % \changes{v0.6}{2008/04/02}{Added \cmd\mlist@headhook.}
1132 % \end{macro}
1133 %
1134 % \begin{macro}{\mlist@add}
1135 % Used in the above to build up the \cmd\mlist@list\ list.
1136 % \begin{macrocode}
1137 \newcommand\mlist@add[1]{%
1138 \expandafter\gdef
1139 \expandafter\mlist@list@elems
1140 \expandafter{\mlist@list@elems#1}}
1141 % \end{macrocode}
1142 % \end{macro}
1143 %
1144 % \begin{macro}{\mlist@count}
1145 % This macro is used to increment the counter used in
1146 % \cmd\MID/\cmd\LAST\ during the actual typesetting of the indexed
1147 % elements.
1148 % \begin{macrocode}
1149 \def\mlist@count{%
1150 \expandafter\mlist@add
1151 \expandafter{%
1152 \expandafter\@tempcnta\the\@tempcnta\relax}}
1153 % \end{macrocode}
1154 % \end{macro}
1155 %
1156 % \begin{macro}{\mlist@add@ifcs}
1157 % Pretty specific macro to save some repetition. Uses
1158 % \cmd\mlist@add\ to add the first argument as a csname if it
1159 % exists, otherwise adds the literal second argument.
1160 % \begin{macrocode}
1161 \newcommand\mlist@add@ifcs[2]{%
1162 \ifcsname#1\endcsname
1163 \expandafter\@firstoftwo
1164 \else
1165 \expandafter\@secondoftwo
1166 \fi{\expandafter\mlist@add\expandafter{\expandafter{\csname#1\endcsname}}}
1167 {\expandafter\mlist@add\expandafter{\expandafter{#2}}}}
1168 % \end{macrocode}
1169 % \end{macro}
1170 %
1171 % \begin{macro}{\mlist@parse@implicit}
1172 % Parses the index to see if it's an implicit reference: |@|$n$. The
1173 % idea is to expand out the index, set a conditional if the first
1174 % char is an |@|, and then gobble up the whole thing. We do this
1175 % twice for the two common catcodes that |@| might be.
1176 % \begin{macrocode}
1177 \def\mlist@parse@implicit{%
1178 \@mlist@implicit@false
1179 \def\@tempb{%
1180 \expandafter\@ifnextchar\mlist@ampersat
1181 {\@mlist@implicit@true\@gobblenil}
1182 {\@mlist@implicit@false\@gobblenil}}%
1183 \expandafter\@tempb\@ii\@nil
1184 \unless\if@mlist@implicit@
1185 \def\@tempb{%
1186 \@ifnextchar @
1187 {\@mlist@implicit@true\@gobblenil}
1188 {\@mlist@implicit@false\@gobblenil}}%
1189 \expandafter\@tempb\@ii\@nil
1190 \fi}
1191 % \end{macrocode}
1192 % \changes{v0.3}{2007/05/09}{Added second check to \cmd\@ifnextchar
1193 % |@| for catcode letter.}
1194 % \end{macro}
1195 %
1196 % \section{List creation}
1197 %
1198 % This is the idea behind the user level macros of the package:
1199 % \begin{verbatim}
1200 % \newmlist\vect{all-vect-opts}
1201 % \vect{v}{ind} == \@mlist{symbol={v},all-vect-opts,index={ind}}
1202 %
1203 % \newvect\vv{v}[this-v-options]
1204 % \vv[ind] == \@mlist{symbol={v},all-vect-opts,this-v-options,index={ind}}
1205 % \end{verbatim}
1206 %
1207 % \begin{macro}{\newmlist}
1208 % \begin{macro}{\renewmlist}
1209 % \darg{control sequence}
1210 % \oarg{keyval options}
1211 % Things get pretty hairy because I have to do all the optional
1212 % argument processing manually. I'd be better off using the
1213 % \pkg{suffix} package, but the code works for now. \ltx3 will make
1214 % things much easier.
1215 % \begin{macrocode}
1216 \newcommand\newmlist[1]{%
1217 \@mlist@warntrue
1218 \mlist@optarg{\@defmlist@opt{#1}}}
1219 \newcommand\renewmlist[1]{%
1220 \@mlist@warnfalse
1221 \mlist@optarg{\@defmlist@opt{#1}}}
1222 \def\@defmlist@opt#1[#2]{\@defmlist{#1}{#2}}
1223 % \end{macrocode}
1224 % Processing |#1| = thislist-cmd, |#2| = all-thislist-opts
1225 % \begin{macrocode}
1226 \newcommand\@defmlist[2]{%
1227 \if@mlist@warn
1228 \ifdefined#1
1229 \PackageError{mlist}{Command \string#1 already defined}
1230 {\string#1 will be overwritten}%
1231 \fi
1232 \fi
1233 % \end{macrocode}
1234 % Plain usage: (e.g., |\vect{a}{1,2,3}|)
1235 % \begin{macrocode}
1236 \protected\def#1##1{%
1237 \@ifnextbrace
1238 {\csname\string#1@opt\endcsname{##1}}
1239 {\csname\string#1@opt\endcsname{##1}{}}}%
1240 \def@c{\string#1@opt}##1##2{%
1241 \csname\string#1\endcsname{##1}{##2}}%
1242 \def@c{\string#1}##1##2{%
1243 \@mlist{symbol=##1,#2,index={##2}}}%
1244 % \end{macrocode}
1245 % mlist macro definitions: (e.g., |\newvect{A}[...]|)
1246 % Processing |##1| = this-sym-cmd, |##2| = symbol, |##3| = this-sym-options
1247 % \begin{macrocode}
1248 \def@c{new\expandafter\@gobble\string#1}##1##2{%
1249 \@mlist@warntrue
1250 \mlist@optarg{\csname @new\string#1@opt\endcsname{##1}{##2}}}%
1251 \def@c{renew\expandafter\@gobble\string#1}##1##2{%
1252 \@mlist@warnfalse
1253 \mlist@optarg{\csname @new\string#1@opt\endcsname{##1}{##2}}}%
1254 \def@c{@new\string#1@opt}##1##2[##3]{%
1255 \csname @new\string#1\endcsname{##1}{##2}{##3}}%
1256 \def@c{@new\string#1}##1##2##3{%
1257 \if@mlist@warn
1258 \ifdefined##1
1259 \PackageError{mlist}{Command \string##1 already defined}
1260 {You cannot overwrite previous definition of \string##1}%
1261 \fi
1262 \fi
1263 \protected\def##1{%
1264 \@ifnextbrace
1265 {\csname @new\string##1@opt\endcsname}
1266 {\@mlist{symbol=##2,#2,##3}}}%
1267 \def@c{@new\string##1@opt}####1{%
1268 \@mlist{symbol=##2,#2,##3,index={####1}}}}%
1269 }
1270 % \end{macrocode}
1271 % \end{macro}
1272 % \end{macro}
1273 %
1274 % \begin{example}{firstline=2}
1275 % \raggedright\ttfamily\slshape
1276 % \def\@mlist#1{MLIST: #1\par}
1277 % \newmlist\testlist[opt1=one]
1278 % \newtestlist\testone{S}
1279 % \testone
1280 % \testone{1,2,3}
1281 % \newtestlist\testtwo{T}[opt2=two]
1282 % \testtwo
1283 % \testtwo{1,2,3}
1284 % \end{example}
1285 %
1286 % \begin{macrocode}
1287 \InputIfFileExists{mlist.cfg}{}
1288 {\PackageWarning{mlist}{No configuration file (mlist.cfg) found nor loaded}}
1289 % \end{macrocode}
1290 %
1291 %\iffalse
1292 %</package>
1293 %\fi
1294 %
1295 %
1296 %\iffalse
1297 %<*dtx-style>
1298 % \begin{macrocode}
1299 \ProvidesPackage{dtx-style}
1300
1301 \errorcontextlines=999
1302
1303 \def\@dotsep{1000}
1304 \setcounter{tocdepth}{2}
1305 \setlength\columnseprule{0.1pt}
1306 \renewcommand\tableofcontents{\relax
1307 \begin{multicols}{2}[\section*{\contentsname}]\small
1308 \@starttoc{toc}\relax
1309 \end{multicols}}
1310
1311 \setcounter{IndexColumns}{2}
1312 \renewenvironment{theglossary}
1313 {\small\list{}{}
1314 \item\relax
1315 \glossary@prologue\GlossaryParms
1316 \let\item\@idxitem \ignorespaces
1317 \def\pfill{\hspace*{\fill}}}
1318 {\endlist}
1319
1320 \usepackage[svgnames]{xcolor}
1321 \usepackage{amsmath,array,bm,booktabs,calc,enumitem,fancyvrb,graphicx,ifthen,longtable,refstyle,subfig,topcapt,varioref,mlist}
1322
1323 \usepackage{easyvector,vector,tensind}
1324 \usepackage{url}
1325
1326 \usepackage[sc,osf]{mathpazo}
1327
1328 \linespread{1.069} % A bit more space between lines
1329 \frenchspacing % Remove ugly extra space after punctuation
1330
1331 \definecolor{niceblue}{rgb}{0.2,0.4,0.8}
1332 \newenvironment{example}[1]
1333 {\VerbatimEnvironment
1334 \def\Options{#1}%
1335 \begin{VerbatimOut}[gobble=2]{\examplefilename}}
1336 {\end{VerbatimOut}\relax
1337 \typesetexample}
1338
1339 \fvset{formatcom=\color{niceblue}}
1340 \DefineShortVerb{\|}
1341
1342 \def\theCodelineNo{\textcolor{niceblue}{\sffamily\tiny\arabic{CodelineNo}}}
1343
1344 \let\examplesize\normalsize
1345 \let\auxwidth\relax
1346
1347 \newlength\examplewidth\newlength\verbatimwidth
1348 \newlength\exoutdent \newlength\exverbgap
1349 \setlength\exverbgap{1em}
1350 \setlength\exoutdent{-0.15\textwidth}
1351 \newsavebox\verbatimbox
1352 \edef\examplefilename{\jobname.example}
1353
1354 \newcommand\typesetexample{\relax
1355 \smallskip
1356 \noindent
1357 \begin{minipage}{\linewidth}
1358 \color{niceblue}
1359 \hrulefill\par
1360 \edef\@tempa{[gobble=0,fontsize=\noexpand\small,\Options]}%
1361 \begin{lrbox}{\verbatimbox}\relax
1362 \expandafter\BVerbatimInput\@tempa{\examplefilename}%
1363 \end{lrbox}
1364 \begin{list}{}{\setlength\itemindent{0pt}
1365 \setlength\leftmargin\exoutdent
1366 \setlength\rightmargin{0pt}}\item
1367 \ifx\auxwidth\relax
1368 \setlength\verbatimwidth{\wd\verbatimbox}%
1369 \else
1370 \setlength\verbatimwidth{\auxwidth}%
1371 \fi
1372 \begin{minipage}[c]{\textwidth-\exoutdent-\verbatimwidth-\exverbgap}
1373 \catcode`\%=14\centering\linespread{1.6}\input\examplefilename\relax
1374 \end{minipage}\hfill
1375 \begin{minipage}[c]{\verbatimwidth}
1376 \usebox\verbatimbox
1377 \end{minipage}
1378 \end{list}
1379 \par\noindent\hrulefill
1380 \end{minipage}\par
1381 \smallskip\noindent}
1382
1383 \newcommand*\setverbwidth[1]{\def\auxwidth{#1}}
1384
1385 \newcommand*\name[1]{{#1}}
1386 \newcommand*\pkg[1]{\textsf{#1}}
1387 \newcommand*\feat[1]{\texttt{#1}}
1388 \newcommand*\opt[1]{\texttt{#1}}
1389
1390 \newcommand*\ltx[1]{%
1391 \ifx3#1\relax
1392 \textsc{ltx3}%
1393 \else
1394 \LaTeXe
1395 \fi}
1396
1397 \newcommand*\note[1]{\unskip\footnote{#1}}
1398
1399 \let\latin\textit
1400 \def\eg{\latin{e.g.}}
1401 \def\Eg{\latin{E.g.}}
1402 \def\ie{\latin{i.e.}}
1403 \def\etc{\@ifnextchar.{\latin{etc}}{\latin{etc.}\@}}
1404
1405 \def\STIX{\textsc{stix}}
1406 \def\MacOSX{Mac~OS~X}
1407 \def\ascii{\textsc{ascii}}
1408 \def\OMEGA{Omega}
1409 \def\CTAN{\textsc{ctan}}
1410
1411 \newcounter{argument}
1412 \g@addto@macro\endmacro{\setcounter{argument}{0}}
1413 \newcommand*\darg[1]{%
1414 \stepcounter{argument}%
1415 {\ttfamily\char`\#\theargument~:~}#1\par\noindent\ignorespaces}
1416 \newcommand*\doarg[1]{%
1417 \stepcounter{argument}%
1418 {\ttfamily\makebox[0pt][r]{[}\char`\#\theargument]:~}#1\par\noindent\ignorespaces}
1419
1420 \def\codeline{\par\hspace{\parindent}}% two \indents in total
1421
1422 \newcommand\unichar[2]{\textsc{\MakeLowercase{u+#1: #2}}}
1423
1424 \setlength\parindent{2em}
1425
1426 % \end{macrocode}
1427 %</dtx-style>
1428 %\fi
1429 %
1430 % \typeout{*************************************************************}
1431 % \typeout{*}
1432 % \typeout{* To finish the installation you have to move the following}
1433 % \typeout{* file into a directory searched by XeTeX:}
1434 % \typeout{*}
1435 % \typeout{* \space\space\space mlist.sty}
1436 % \typeout{*}
1437 % \typeout{*************************************************************}
1438 %
1439 \endinput