| 1 |
% \iffalse meta-comment |
| 2 |
% |
| 3 |
% Copyright (C) 2004 by Robert J Lee <[email protected]> |
| 4 |
% -------------------------------------------------------------- |
| 5 |
% |
| 6 |
% This file may be distributed and/or modified under the |
| 7 |
% conditions of the LaTeX Project Public License, either version 1.2 |
| 8 |
% of this license or (at your option) any later version. |
| 9 |
% The latest version of this license is in: |
| 10 |
% |
| 11 |
% http://www.latex-project.org/lppl.txt |
| 12 |
% |
| 13 |
% and version 1.2 or later is part of all distributions of LaTeX |
| 14 |
% version 1999/12/01 or later. |
| 15 |
% |
| 16 |
% \fi |
| 17 |
% |
| 18 |
% \iffalse |
| 19 |
%<*driver> |
| 20 |
\ProvidesFile{rjlpshap.dtx} |
| 21 |
%</driver> |
| 22 |
%<package> \NeedsTeXFormat{LaTeX2e}[1999/12/01] |
| 23 |
%<package> \ProvidesPackage{rjlpshap} |
| 24 |
%<*package> |
| 25 |
[2004/11/05 v1.0 .dtx rjlpshap file] |
| 26 |
%</package> |
| 27 |
% |
| 28 |
%<*driver> |
| 29 |
\documentclass{ltxdoc} |
| 30 |
\EnableCrossrefs |
| 31 |
\CodelineIndex |
| 32 |
\RecordChanges |
| 33 |
\usepackage{lipsum} |
| 34 |
\begin{document} |
| 35 |
\DocInput{rjlpshap.dtx} |
| 36 |
\end{document} |
| 37 |
%</driver> |
| 38 |
% \fi |
| 39 |
% |
| 40 |
% \CheckSum{0} |
| 41 |
% |
| 42 |
% \CharacterTable |
| 43 |
% {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 |
| 44 |
% 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 |
| 45 |
% Digits \0\1\2\3\4\5\6\7\8\9 |
| 46 |
% Exclamation \! Double quote \" Hash (number) \# |
| 47 |
% Dollar \$ Percent \% Ampersand \& |
| 48 |
% Acute accent \' Left paren \( Right paren \) |
| 49 |
% Asterisk \* Plus \+ Comma \, |
| 50 |
% Minus \- Point \. Solidus \/ |
| 51 |
% Colon \: Semicolon \; Less than \< |
| 52 |
% Equals \= Greater than \> Question mark \? |
| 53 |
% Commercial at \@ Left bracket \[ Backslash \\ |
| 54 |
% Right bracket \] Circumflex \^ Underscore \_ |
| 55 |
% Grave accent \` Left brace \{ Vertical bar \| |
| 56 |
% Right brace \} Tilde \~} |
| 57 |
% |
| 58 |
% |
| 59 |
% \changes{v1.0}{2004/11/05}{Initial version} |
| 60 |
% |
| 61 |
% \GetFileInfo{rjlpshap.dtx} |
| 62 |
% |
| 63 |
% \DoNotIndex{\newcommand,\newenvironment} |
| 64 |
% |
| 65 |
% |
| 66 |
% \title{The \textsf{rjlpshap} class\thanks{This document |
| 67 |
% corresponds to \textsf{rjlpshap}~\fileversion, dated \filedate.}} |
| 68 |
% \author{Robert J Lee \\ \texttt{[email protected]}} |
| 69 |
% |
| 70 |
% \maketitle |
| 71 |
% |
| 72 |
% \section{Introduction} |
| 73 |
% |
| 74 |
% This package provides low-level helper macros and environments. It |
| 75 |
% is intended for authors of \LaTeX\ packages, who wish to |
| 76 |
% programmatically change the shape of paragraphs. It overcomes several |
| 77 |
% difficulties with \TeX's \texttt{$\backslash$parshape} register: |
| 78 |
% |
| 79 |
% \begin{enumerate} |
| 80 |
% \item It is not possible to re-use a length register when issuing a |
| 81 |
% \texttt{$\backslash$parshape} command; this makes it difficult to |
| 82 |
% calculate the shape of a paragraph in \LaTeX\ using some |
| 83 |
% algorithms that re-use one or more variables (registers). |
| 84 |
% \item The number of lines must be known \textit{before} the lengths |
| 85 |
% of each line, which can lead to difficult calculations. |
| 86 |
% \item The format of the \texttt{$\backslash$parshape} command is |
| 87 |
% very strict and evaluated early on by \TeX, and so it doesn't |
| 88 |
% easily allow for parameter-replacement, making it difficult to use |
| 89 |
% any calculations at all. |
| 90 |
% \end{enumerate} |
| 91 |
% |
| 92 |
% It differs from the \texttt{shapepar} package, in that it provides |
| 93 |
% no actual calculations for paragraph shapes. Instead, it provides a |
| 94 |
% framework by which paragraph shapes can be calculated in a single |
| 95 |
% run of the document by the author (or package writer). |
| 96 |
% |
| 97 |
% The package provides both environments and commands by which the |
| 98 |
% author can accumulate any number of lengths and then use them in a |
| 99 |
% \texttt{$\backslash$parshape} command injected into the document, |
| 100 |
% without the need to re-run \LaTeX. |
| 101 |
% |
| 102 |
% \pagebreak[4] |
| 103 |
% \section{Usage} |
| 104 |
% |
| 105 |
% There are two modes of operation: environment mode and command |
| 106 |
% mode. Environment mode has the advantage that it does not |
| 107 |
% require the number of lengths to be known in advance, while command |
| 108 |
% mode provides various options for passing pre-calculated lengths. |
| 109 |
% |
| 110 |
% \subsection{Environment Mode} |
| 111 |
% |
| 112 |
% Environment mode uses a single environment, |
| 113 |
% \texttt{parshapecollect}, which starts and ends before the paragraph |
| 114 |
% of text. The contents of the environment are a \LaTeX\ program, |
| 115 |
% supplied by the user, that calculates the lengths of each line. |
| 116 |
% |
| 117 |
% \DescribeEnv{parshapecollect} The \texttt{parshapecollect} |
| 118 |
% environment goes at the start of a paragraph, and the entire |
| 119 |
% environment defines a single \texttt{$\backslash$parshape} command. |
| 120 |
% Any even number of lengths are collected, and at the end of the |
| 121 |
% environment, the \texttt{$\backslash$parshape} command is |
| 122 |
% automatically generated and inserted into the document. |
| 123 |
% |
| 124 |
% \DescribeMacro{\parshapelenout} This macro takes one mandatory |
| 125 |
% parameter, which is the next length in the |
| 126 |
% \texttt{$\backslash$parshape} command. Note that the parameter is |
| 127 |
% expanded: the value of the length at the time |
| 128 |
% \texttt{$\backslash$parshapelenout} is called is the value which is |
| 129 |
% used --- this means the same length command can be re-used |
| 130 |
% repeatedly. The value may be a length register, and the same |
| 131 |
% register may be re-used after changing its contents, for example: |
| 132 |
% |
| 133 |
% \begin{quote} |
| 134 |
% \begin{verbatim} |
| 135 |
% \newlength{\foo} |
| 136 |
% \begin{parshapecollect} |
| 137 |
% \setlength{\foo}{1in}\parshapelenout{\foo} |
| 138 |
% \setlength{\foo}{3in}\parshapelenout{\foo} |
| 139 |
% \setlength{\foo}{0.5in}\parshapelenout{\foo} |
| 140 |
% \setlength{\foo}{3.5in}\parshapelenout{\foo} |
| 141 |
% \setlength{\foo}{0in}\parshapelenout{\foo} |
| 142 |
% \setlength{\foo}{4in}\parshapelenout{\foo} |
| 143 |
% \end{parshapecollect} |
| 144 |
% \lipsum[1] |
| 145 |
% \end{verbatim} |
| 146 |
% Produces: |
| 147 |
% |
| 148 |
% \parshape 3 1in 4in 0.5in 4.5in 0in 5in |
| 149 |
% \lipsum[1] |
| 150 |
% \end{quote} |
| 151 |
|
| 152 |
% |
| 153 |
% \DescribeMacro{\parshapearrlenout} This macro is the same as |
| 154 |
% \texttt{$\backslash$parshapelenout}, but allows the programmer to |
| 155 |
% pass the length as the value from an array (see the |
| 156 |
% \texttt{arrayjob} package). This is intended for situations where |
| 157 |
% pre-calculated lengths are to be used alongside calculations; to |
| 158 |
% pass an array of lengths the various command modes may be |
| 159 |
% easier. The parameters to the macro are the name of the array and |
| 160 |
% the name of a counter containing the index. |
| 161 |
% |
| 162 |
% For example, the above paragraph could have been created |
| 163 |
% using: |
| 164 |
% \begin{quote} |
| 165 |
% \begin{verbatim} |
| 166 |
% \newarray{foo} \newcounter{ctr} |
| 167 |
% \begin{parshapecollect} |
| 168 |
% \foo(1)={1in}\setcounter{ctr}{1}\parshapearrlenout{foo}{ctr} |
| 169 |
% \foo(2)={3in}\setcounter{ctr}{2}\parshapearrlenout{foo}{ctr} |
| 170 |
% \foo(3)={0.5in}\setcounter{ctr}{3}\parshapearrlenout{foo}{ctr} |
| 171 |
% \foo(1)={3.5in}\setcounter{ctr}{1}\parshapearrlenout{foo}{ctr} |
| 172 |
% \foo(2)={0in}\setcounter{ctr}{2}\parshapearrlenout{foo}{ctr} |
| 173 |
% \foo(2)={4in}\parshapearrlenout{foo}{ctr} |
| 174 |
% \end{parshapecollect} |
| 175 |
% \lipsum[1] |
| 176 |
% \end{verbatim} |
| 177 |
% \end{quote} |
| 178 |
% |
| 179 |
% \subsection{Command Mode} |
| 180 |
|
| 181 |
% Command mode enables the author to use calculated shapes by |
| 182 |
% outputting the $\backslash$parshape command as a single operation at |
| 183 |
% the end of the calculation. |
| 184 |
|
| 185 |
% \DescribeMacro{\parshapeary} |
| 186 |
% This macro takes one parameter, being the name of an array. The |
| 187 |
% contents of the array are simply expanded and passed directly to |
| 188 |
% \texttt{$\backslash$parshape} --- the first parameter is the number |
| 189 |
% of lines, then for each line the left margin and line width are |
| 190 |
% supplied in subsequent elements. |
| 191 |
% |
| 192 |
% For example: |
| 193 |
% \begin{quote} |
| 194 |
% \begin{verbatim} |
| 195 |
% \newcommand{\ltest}{50pt} |
| 196 |
% \newlength{\test}\setlength{\test}{300pt} |
| 197 |
% \newarray{sizes}\readarray{sizes}{2&0em&\the\test&\ltest&\test} |
| 198 |
% \parshapeary{sizes} |
| 199 |
% \lipsum[1] |
| 200 |
% \end{verbatim} |
| 201 |
% produces: |
| 202 |
% |
| 203 |
% \parshape 2 0em 300pt 50pt 300pt \lipsum[1] |
| 204 |
% \end{quote} |
| 205 |
% |
| 206 |
% |
| 207 |
% \DescribeMacro{\parshapearray} This macro takes two parameters. The |
| 208 |
% first is also the name of an array. The only difference to |
| 209 |
% \texttt{$\backslash$parshapeary} is that instead of passing the |
| 210 |
% width of the line, the width of the right margin is passed |
| 211 |
% instead. This means the macro needs to know the width of the line. |
| 212 |
% |
| 213 |
% The second parameter is the length of the line. This is usually |
| 214 |
% $\backslash$columnwidth, passing any other value will effectively |
| 215 |
% change the right margin. ($\backslash$columnwidth is often equal to |
| 216 |
% $\backslash$textwidth, but in multi-column mode, |
| 217 |
% $\backslash$columnwidth is more likely to be correct, unless you |
| 218 |
% want to overstrike the text with text in another column --- so |
| 219 |
% $\backslash$columnwidth is the recommended length to use). |
| 220 |
% |
| 221 |
% In a similar vein to the previous example: |
| 222 |
% \begin{quote} |
| 223 |
% \begin{verbatim} |
| 224 |
%\newcommand{\ltest}{50pt} |
| 225 |
%\newlength{\test}\setlength{\test}{75pt} |
| 226 |
%\newlength{\cwidth}\setlength{\cwidth}{\columnwidth} |
| 227 |
%\addtolength{\cwidth}{-50pt} |
| 228 |
% |
| 229 |
%\newarray{sizes}\readarray{sizes}{2&0em&\the\test&\ltest&\test} |
| 230 |
%\parshapearray{sizes}{\cwidth} |
| 231 |
%\lipsum[1] |
| 232 |
% |
| 233 |
% \end{verbatim} |
| 234 |
% which produces: |
| 235 |
% |
| 236 |
% \parshape 2 0.0pt |
| 237 |
%220.0pt |
| 238 |
%50.0pt |
| 239 |
%170.0pt |
| 240 |
% \lipsum[1] |
| 241 |
% \end{quote} |
| 242 |
% |
| 243 |
% \DescribeMacro{\Parshapearray} |
| 244 |
% This macro also takes two parameters, and is used to combine |
| 245 |
% multiple shapes, where the width of the paragraph is not consistent |
| 246 |
% throughout the paragraph. |
| 247 |
% |
| 248 |
% The first parameter is the name of an array, in the same format as |
| 249 |
% the first parameter to \texttt{$\backslash$parshapeary}. |
| 250 |
% |
| 251 |
% The second parameter is the name of a second array. The first |
| 252 |
% element is the number of lines in this second array, and subsequent |
| 253 |
% elements are the widths of lines (excluding margins) in that array |
| 254 |
% --- starting from the first line. As usual, if an insufficient |
| 255 |
% number of lines are passed, the last line will be re-used. |
| 256 |
% |
| 257 |
% The actual width of a line will be the total width excluding margins |
| 258 |
% (as per the second parameter), minus the left margin (taken from the |
| 259 |
% first parameter, even indexes), minus the right margin (as per the |
| 260 |
% first parameter, odd indexes). |
| 261 |
% |
| 262 |
% In this way, it is possible to set a shaped paragraph that allows |
| 263 |
% for non-straight margins or non-rectangular paper --- for example, to |
| 264 |
% typeset a list item in such a way as to allow for a cut-out on the |
| 265 |
% page. |
| 266 |
% |
| 267 |
% For a larger example, consider that the paper has a sloped shape |
| 268 |
% that extends 1em wider for every line for the first 15 lines |
| 269 |
% only. The author wishes to set a text in an quadrilateral, with |
| 270 |
% margins offset $\frac{1}{8}\verb!"!$ to the right for each line, but |
| 271 |
% extending equally to follow the width of the paper. |
| 272 |
|
| 273 |
% \begin{quote} |
| 274 |
% \begin{verbatim} |
| 275 |
% \newarray{quad} |
| 276 |
% \readarray{quad}{17 & 0.000in & 2.000in & 0.125in & 1.875in & |
| 277 |
% 0.250in & 1.750in & 0.375in & 1.625in & 0.500in & 1.500in & |
| 278 |
% 0.625in & 1.375in & 0.750in & 1.250in & 0.875in & 1.125in & |
| 279 |
% 1.000in & 1.000in & 1.125in & 0.875in & 1.250in & 0.750in & |
| 280 |
% 1.375in & 0.675in & 1.500in & 0.500in & 1.625in & 0.375in & |
| 281 |
% 1.750in & 0.250in & 1.875in & 0.125in & 2.000in & 0.000in} |
| 282 |
% \newarray{papershape} |
| 283 |
% \readarray{papershape}{15& 25em & 26em & 27em & 28em & 29em & 30em & 31em |
| 284 |
% & 32em & 33em & 34em & 35em & 36em & 37em & 38em & 39em} |
| 285 |
% \Parshapearray{quad}{papershape} \lipsum[1] |
| 286 |
% \end{verbatim} |
| 287 |
% Produces: |
| 288 |
% |
| 289 |
% \parshape 17 |
| 290 |
% 0.0pt |
| 291 |
% 105.46039pt |
| 292 |
% 9.03374pt |
| 293 |
% 115.46042pt |
| 294 |
% 18.06749pt |
| 295 |
% 125.46043pt |
| 296 |
% 27.10124pt |
| 297 |
% 135.46045pt |
| 298 |
% 36.135pt |
| 299 |
% 145.46045pt |
| 300 |
% 45.16875pt |
| 301 |
% 155.46046pt |
| 302 |
% 54.2025pt |
| 303 |
% 165.46048pt |
| 304 |
% 63.23624pt |
| 305 |
% 175.46051pt |
| 306 |
% 72.26999pt |
| 307 |
% 185.46053pt |
| 308 |
% 81.30374pt |
| 309 |
% 195.46054pt |
| 310 |
% 90.3375pt |
| 311 |
% 205.46054pt |
| 312 |
% 99.37125pt |
| 313 |
% 211.84685pt |
| 314 |
% 108.405pt |
| 315 |
% 225.46057pt |
| 316 |
% 117.43874pt |
| 317 |
% 235.4606pt |
| 318 |
% 126.47249pt |
| 319 |
% 245.46062pt |
| 320 |
% 135.50624pt |
| 321 |
% 235.4606pt |
| 322 |
% 144.54pt |
| 323 |
% 235.46059pt \lipsum[1] |
| 324 |
% \end{quote} |
| 325 |
% |
| 326 |
|
| 327 |
|
| 328 |
% \section{Prerequisites} |
| 329 |
|
| 330 |
% This package requires the \texttt{arrayjob} package from |
| 331 |
% CTAN\footnote{http://tug.ctan.org/cgi-bin/ctanPackageInformation.py?id=arrayjob}. Arrays |
| 332 |
% are used as a convenient mechanism to pass values to functions, |
| 333 |
% although they may not be used in environment mode. |
| 334 |
|
| 335 |
% \section{Limitations and Possible Improvements} |
| 336 |
|
| 337 |
% \begin{itemize} |
| 338 |
% \item Usage of this package can be slow, because two temporary files |
| 339 |
% are used for each $\backslash$parshape. Theoretically, only one |
| 340 |
% should be needed and so it should be possible to improve this. |
| 341 |
% \item There is no support for combining arrays, other than to change |
| 342 |
% the paper size itself. Extra commands could be provided to |
| 343 |
% calculate the intersection of two cutouts, or to add margins |
| 344 |
% together. |
| 345 |
% \item In a future version, it may be possible to avoid temporary |
| 346 |
% files completely by using token registers, or |
| 347 |
% $\backslash$aftergroup, with some form of string processing. |
| 348 |
% \end{itemize} |
| 349 |
|
| 350 |
% \section{Package Name} |
| 351 |
|
| 352 |
% The somewhat obscure name ``rjlpshap'' was chosen following the |
| 353 |
% guidelines given in \textit{\LaTeXe\ for Class and Package |
| 354 |
% Writers}\footnote{\texttt{http://www.tex.ac.uk/tex-archive/macros/latex/doc/clsguide.pdf}}~(section~2.7.3, |
| 355 |
% ``Make it Portable''). The common prefix used by these packages |
| 356 |
% is~''rjl--'', which was chosen as the author's initials as he is |
| 357 |
% not writing these packages as part of any organisation. |
| 358 |
|
| 359 |
% The postfix of ``--pshap'' was chosen as a contraction of |
| 360 |
% ``parshape'', the \TeX\ command to which this package is a front |
| 361 |
% end. The name was contracted due to the 5-character length |
| 362 |
% restriction. It was also considered that that the obvious package |
| 363 |
% name of ``parshape'' would be best avoided to avoid confusion with |
| 364 |
% the existing shapepar\footnote{\texttt{http://www.ctan.org/tex-archive/macros/latex/contrib/shapepar/}} package, and also because parshape is a |
| 365 |
% standard \TeX\ command --- the only requirement that exists is to |
| 366 |
% avoid the name of a standard package, but avoiding the name of a |
| 367 |
% standard macro also seems like a good idea as this leaves the name |
| 368 |
% free for use as a standard package name in the future. |
| 369 |
|
| 370 |
% \section{Other References} |
| 371 |
|
| 372 |
% The following references were essential in producing this package: |
| 373 |
|
| 374 |
% \begin{itemize} |
| 375 |
% \item The \TeX Book\footnote{ISBN:~978-0201134483} |
| 376 |
% \item \TeX Live package |
| 377 |
% contributions\footnote{http://tug.org/texlive/pkgcontrib.html} |
| 378 |
% \item TDS Guidelines\footnote{http://dante.ctan.org/TDS-guidelines.html} |
| 379 |
% \end{itemize} |
| 380 |
|
| 381 |
% \StopEventually{\PrintChanges\PrintIndex} |
| 382 |
% |
| 383 |
% \section{Implementation} |
| 384 |
% |
| 385 |
% Arrayjob: we need to pass arrays of lengths and length-commands, |
| 386 |
% so build on the established array handling functionality. |
| 387 |
% \begin{macrocode} |
| 388 |
\RequirePackage{arrayjob} |
| 389 |
% \end{macrocode} |
| 390 |
|
| 391 |
% Forloop: we need to iterate over the arrays that are passed in |
| 392 |
% \begin{macrocode} |
| 393 |
\RequirePackage{forloop} |
| 394 |
% \end{macrocode} |
| 395 |
% Internal commands use @ as a letter, not a special character, so |
| 396 |
% switch catcodes |
| 397 |
% \begin{macrocode} |
| 398 |
\makeatletter |
| 399 |
% \end{macrocode} |
| 400 |
|
| 401 |
|
| 402 |
% \subsection{Environment mode} |
| 403 |
|
| 404 |
% \begin{environment}{parshapecollect} |
| 405 |
% Environment to handle writing the file, cleaning up etc. |
| 406 |
% |
| 407 |
% Note the importance of the % at the end of each line here: we're |
| 408 |
% in horizontal mode so any space characters get expanded and added |
| 409 |
% into the start of the paragraph. Any spaces here are output before |
| 410 |
% $\backslash$parshape itself, and so will be visible in the output |
| 411 |
% page. The $\backslash$ignorespaces is a last line of defence |
| 412 |
% against any extra spaces inadvertently introduced by the user. |
| 413 |
% |
| 414 |
% \begin{macrocode} |
| 415 |
\newenvironment{parshapecollect}{% |
| 416 |
\immediate\openout\rjlpshap@writer=\jobname.parshape% |
| 417 |
\setcounter{rjlpshap@linecount}{0}% |
| 418 |
\setboolean{rjlpshap@isodd}{false}% |
| 419 |
\ignorespaces |
| 420 |
}{% |
| 421 |
\immediate\closeout\rjlpshap@writer% |
| 422 |
% \end{macrocode} |
| 423 |
% Because the environment forms a group, we need to use aftergroup to |
| 424 |
% delay the $\backslash$parshape command. Otherwise the group will end first and |
| 425 |
% $\backslash$parshape will be discarded. |
| 426 |
% \begin{macrocode} |
| 427 |
\rjlpshap@doparshape% |
| 428 |
\aftergroup\input\aftergroup{\aftergroup\jobname\aftergroup% |
| 429 |
.\aftergroup% |
| 430 |
p\aftergroup% |
| 431 |
a\aftergroup% |
| 432 |
r\aftergroup% |
| 433 |
s\aftergroup% |
| 434 |
h\aftergroup% |
| 435 |
a\aftergroup% |
| 436 |
p\aftergroup% |
| 437 |
e\aftergroup% |
| 438 |
t\aftergroup% |
| 439 |
m\aftergroup% |
| 440 |
p\aftergroup}\ignorespaces |
| 441 |
} |
| 442 |
% \end{macrocode} |
| 443 |
% \end{environment} |
| 444 |
|
| 445 |
% \begin{macro}{\rjlpshap@doparshape} |
| 446 |
% Internal method to handle the final processing of the built |
| 447 |
% up $\backslash$parshape command. The lengths have been written to |
| 448 |
% the first temporary file, so write the entire $\backslash$parshape |
| 449 |
% command to the second one, ready to be input. |
| 450 |
% \begin{macrocode} |
| 451 |
\newcommand{\rjlpshap@doparshape}{% |
| 452 |
\immediate\openout\rjlpshap@finalwriter=\jobname.parshapetmp% |
| 453 |
\immediate\write\rjlpshap@finalwriter{\noexpand\parshape \arabic{rjlpshap@linecount}}% |
| 454 |
\openin\rjlpshap@reader=\jobname.parshape% |
| 455 |
\setboolean{rjlpshap@loop}{true}% |
| 456 |
\whiledo{\boolean{rjlpshap@loop}}{% |
| 457 |
\read\rjlpshap@reader to\rjlpshap@temp% |
| 458 |
% NB: \ifeof matches \loop...\repeat |
| 459 |
\ifeof\rjlpshap@reader% |
| 460 |
\setboolean{rjlpshap@loop}{false}% |
| 461 |
\else |
| 462 |
\immediate\write\rjlpshap@finalwriter{\rjlpshap@temp }% |
| 463 |
\fi |
| 464 |
}% |
| 465 |
\immediate\closeout\rjlpshap@finalwriter% |
| 466 |
\immediate\closein\rjlpshap@reader% |
| 467 |
\let\rjlpshap@temp=\relax% |
| 468 |
} |
| 469 |
% \end{macrocode} |
| 470 |
% \end{macro} |
| 471 |
|
| 472 |
% \begin{macro}{\parshapelenout} |
| 473 |
% Used inside a parshapecollect environment, writes length \#1 to |
| 474 |
% the $\backslash$parshape command file. (If used anywhere else, |
| 475 |
% it'll just write to the console since the file will not be open). |
| 476 |
% \begin{macrocode} |
| 477 |
\newcommand{\parshapelenout}[1]{% |
| 478 |
\setlength{\rjlpshap@tempwidth}{#1}% |
| 479 |
\immediate\write\rjlpshap@writer{\the\rjlpshap@tempwidth}% |
| 480 |
\ifthenelse{\boolean{rjlpshap@isodd}}{% |
| 481 |
\setboolean{rjlpshap@isodd}{false}% |
| 482 |
}{% |
| 483 |
\stepcounter{rjlpshap@linecount}% |
| 484 |
\setboolean{rjlpshap@isodd}{true}% |
| 485 |
}% |
| 486 |
\ignorespaces |
| 487 |
} |
| 488 |
% \end{macrocode} |
| 489 |
% \end{macro} |
| 490 |
|
| 491 |
% \begin{macro}{\parshapearrlenout} |
| 492 |
% Used inside a parshapecollect environment, writes a length to the |
| 493 |
% $\backslash$parshape command file. As $\backslash$parshapelenout, |
| 494 |
% except that the element is passed as the member of an array. |
| 495 |
% |
| 496 |
% Length is taken from array with name \#1, element at index \#2 |
| 497 |
% \begin{macrocode} |
| 498 |
\newcommand{\parshapearrlenout}[2]{% |
| 499 |
\testarray{#1}(\arabic{#2})% |
| 500 |
\parshapelenout{\temp@macro}% |
| 501 |
\ignorespaces |
| 502 |
} |
| 503 |
% \end{macrocode} |
| 504 |
% \end{macro} |
| 505 |
|
| 506 |
% \subsection{Command Mode} |
| 507 |
|
| 508 |
% \begin{macro}{\parshapeary} |
| 509 |
% As $\backslash$parshape, but takes one argument being an array |
| 510 |
% name defined in arrayjob form. The array contains the arguments, |
| 511 |
% starting with the count at index 1, then one length per array |
| 512 |
% element. |
| 513 |
% \begin{macrocode} |
| 514 |
\newcommand{\parshapeary}[1]{% |
| 515 |
\testarray{#1}(1)% |
| 516 |
\edef\rjlpshap@max{\temp@macro}% |
| 517 |
\begin{parshapecollect}% |
| 518 |
\forloop{rjlpshap@ctr}{1}{\not \value{rjlpshap@ctr} > \rjlpshap@max }{% |
| 519 |
\setcounter{rjlpshap@ctr2}{\value{rjlpshap@ctr}}% |
| 520 |
\addtocounter{rjlpshap@ctr2}{\value{rjlpshap@ctr}}% |
| 521 |
\parshapearrlenout{#1}{rjlpshap@ctr2}% |
| 522 |
\stepcounter{rjlpshap@ctr2}% |
| 523 |
\parshapearrlenout{#1}{rjlpshap@ctr2}% |
| 524 |
}% |
| 525 |
\end{parshapecollect}% |
| 526 |
\typeout{parshapeary end}% |
| 527 |
\ignorespaces |
| 528 |
} |
| 529 |
% \end{macrocode} |
| 530 |
% \end{macro} |
| 531 |
|
| 532 |
|
| 533 |
% \begin{macro}{\parshapearray} |
| 534 |
% As $\backslash$parshapeary, but the second parameter is the name |
| 535 |
% of a single-parameter macro taking a number and returning the |
| 536 |
% width of the line on that line |
| 537 |
% \begin{macrocode} |
| 538 |
\newcommand{\parshapearray}[2]{% |
| 539 |
\begin{parshapecollect}% |
| 540 |
\testarray{#1}(1)% |
| 541 |
\edef\rjlpshap@max{\temp@macro}% |
| 542 |
\forloop{rjlpshap@ctr}{1}{\not \value{rjlpshap@ctr} > \rjlpshap@max }{% |
| 543 |
\setcounter{rjlpshap@ctr2}{\value{rjlpshap@ctr}}% |
| 544 |
\addtocounter{rjlpshap@ctr2}{\value{rjlpshap@ctr}}% |
| 545 |
\edef\rjlpshap@pos{\arabic{rjlpshap@ctr2}}% |
| 546 |
\testarray{#1}(\rjlpshap@pos)% |
| 547 |
\setlength{\rjlpshap@tempwidth}{\temp@macro}% |
| 548 |
\setlength{\rjlpshap@templinewidth}{#2}% |
| 549 |
\addtolength{\rjlpshap@templinewidth}{-1\rjlpshap@tempwidth}% |
| 550 |
\parshapelenout{\rjlpshap@tempwidth}% |
| 551 |
\stepcounter{rjlpshap@ctr2}% |
| 552 |
\edef\rjlpshap@pos{\arabic{rjlpshap@ctr2}}% |
| 553 |
\testarray{#1}(\rjlpshap@pos)% |
| 554 |
\setlength{\rjlpshap@tempwidth}{\temp@macro}% |
| 555 |
\addtolength{\rjlpshap@templinewidth}{-1\rjlpshap@tempwidth}% |
| 556 |
\parshapelenout{\rjlpshap@templinewidth}% |
| 557 |
}% |
| 558 |
\end{parshapecollect}% |
| 559 |
} |
| 560 |
% \end{macrocode} |
| 561 |
% \end{macro} |
| 562 |
|
| 563 |
|
| 564 |
% \begin{macro}{\Parshapearray} |
| 565 |
% First parameter is as $\backslash$parshapea\{r,rra\}y, but each pair |
| 566 |
% is left and right margin rather than left margin and text width, |
| 567 |
% while the second tracks the page width. |
| 568 |
% |
| 569 |
% \textit{i.e.} |
| 570 |
% |
| 571 |
% The first parameter (an array) is the shape of the paragraph |
| 572 |
% relative to the page. |
| 573 |
% \begin{itemize} |
| 574 |
% \item first element is the number of pairs of lengths in the array |
| 575 |
% \item subsequent arguments are pairs of lengths, left margin width then |
| 576 |
% right margin width. |
| 577 |
% \end{itemize} |
| 578 |
% |
| 579 |
% The second parameter is the overall text width. This is given in the |
| 580 |
% following format: |
| 581 |
% \begin{itemize} |
| 582 |
% \item first element is the number lengths |
| 583 |
% \item subsequent elements are one width per parameter |
| 584 |
% \end{itemize} |
| 585 |
% \begin{macrocode} |
| 586 |
\newcommand{\Parshapearray}[2]{% |
| 587 |
\testarray{#1}(1)\edef\rjlpshap@maxA{\temp@macro}% |
| 588 |
\testarray{#2}(1)\edef\rjlpshap@maxB{\temp@macro}% |
| 589 |
\ifthenelse{\rjlpshap@maxA > \rjlpshap@maxB} {\edef\rjlpshap@max{\rjlpshap@maxA}}{\edef\rjlpshap@max{\rjlpshap@maxB}}% |
| 590 |
\begin{parshapecollect}% |
| 591 |
\forloop{rjlpshap@ctr}{1}{\not \value{rjlpshap@ctr} > \rjlpshap@max}{% |
| 592 |
\ifthenelse{\value{rjlpshap@ctr} > \rjlpshap@maxA}{% |
| 593 |
\setcounter{rjlpshap@ctr2}{\rjlpshap@maxA}% |
| 594 |
\addtocounter{rjlpshap@ctr2}{\rjlpshap@maxA}% |
| 595 |
}{% |
| 596 |
\setcounter{rjlpshap@ctr2}{\value{rjlpshap@ctr}}% |
| 597 |
\addtocounter{rjlpshap@ctr2}{\value{rjlpshap@ctr}}% |
| 598 |
}% |
| 599 |
% \end{macrocode} |
| 600 |
% left margin from \#1: |
| 601 |
% \begin{macrocode} |
| 602 |
\parshapearrlenout{#1}{rjlpshap@ctr2}% |
| 603 |
% \end{macrocode} |
| 604 |
% get the line width |
| 605 |
% \begin{macrocode} |
| 606 |
\ifthenelse{\value{rjlpshap@ctr} > \rjlpshap@maxB}{% |
| 607 |
\edef\rjlpshap@pos{\rjlpshap@maxB}% |
| 608 |
\setcounter{rjlpshap@ctr3}{\rjlpshap@maxB}% |
| 609 |
\typeout{A pos=\rjlpshap@pos\space of \rjlpshap@maxB}% |
| 610 |
\testarray{#2}(\arabic{rjlpshap@ctr3})% |
| 611 |
\typeout{temp=\temp@macro}% |
| 612 |
\setlength{\rjlpshap@templinewidth}{\temp@macro}% |
| 613 |
}{% |
| 614 |
\stepcounter{rjlpshap@ctr}% |
| 615 |
\edef\rjlpshap@pos{\arabic{rjlpshap@ctr}}% |
| 616 |
\typeout{B pos=\rjlpshap@pos}% |
| 617 |
\addtocounter{rjlpshap@ctr}{-1}% |
| 618 |
\testarray{#2}(\rjlpshap@pos)% |
| 619 |
\setlength{\rjlpshap@templinewidth}{\temp@macro}% |
| 620 |
}% |
| 621 |
\typeout{width=\the\rjlpshap@templinewidth}% |
| 622 |
% \end{macrocode} |
| 623 |
% subtract the left margin from the line width |
| 624 |
% \begin{macrocode} |
| 625 |
\addtolength{\rjlpshap@templinewidth}{-1\rjlpshap@tempwidth}% |
| 626 |
% \end{macrocode} |
| 627 |
% subtract the right margin from the line width |
| 628 |
% \begin{macrocode} |
| 629 |
\stepcounter{rjlpshap@ctr2}% |
| 630 |
\edef\rjlpshap@pos{\arabic{rjlpshap@ctr2}}% |
| 631 |
\testarray{#1}(\rjlpshap@pos)% |
| 632 |
\setlength{\rjlpshap@tempwidth}{\temp@macro}% |
| 633 |
\addtolength{\rjlpshap@templinewidth}{-1\rjlpshap@tempwidth}% |
| 634 |
\parshapelenout{\rjlpshap@templinewidth}% |
| 635 |
}% |
| 636 |
\end{parshapecollect}% |
| 637 |
} |
| 638 |
% \end{macrocode} |
| 639 |
% \end{macro} |
| 640 |
|
| 641 |
% \subsection{Internal Definitions} |
| 642 |
|
| 643 |
% rjlpshap@ctr: outermost loop counter: |
| 644 |
% \begin{macrocode} |
| 645 |
\newcounter{rjlpshap@ctr} |
| 646 |
% \end{macrocode} |
| 647 |
|
| 648 |
% rjlpshap@ctr2: used for index calculations internally: |
| 649 |
% \begin{macrocode} |
| 650 |
\newcounter{rjlpshap@ctr2} |
| 651 |
% \end{macrocode} |
| 652 |
|
| 653 |
% rjlpshap@ctr3: used for index calculations internally: |
| 654 |
% \begin{macrocode} |
| 655 |
\newcounter{rjlpshap@ctr3} |
| 656 |
% \end{macrocode} |
| 657 |
|
| 658 |
% rjlpshap@tempwidth: used as a register to read widths from arrays, |
| 659 |
% and for calculations |
| 660 |
% \begin{macrocode} |
| 661 |
\newlength{\rjlpshap@tempwidth} |
| 662 |
% \end{macrocode} |
| 663 |
|
| 664 |
% rjlpshap@templinewidth: used as a register to calculate the line |
| 665 |
% width when subtracting margins: |
| 666 |
% \begin{macrocode} |
| 667 |
\newlength{\rjlpshap@templinewidth} |
| 668 |
% \end{macrocode} |
| 669 |
|
| 670 |
% boolean used for breaking out of loops |
| 671 |
% \begin{macrocode} |
| 672 |
\newboolean{rjlpshap@loop} |
| 673 |
% \end{macrocode} |
| 674 |
|
| 675 |
% boolean used to track if the number of lengths output is odd or even |
| 676 |
% \begin{macrocode} |
| 677 |
\newboolean{rjlpshap@isodd} |
| 678 |
% \end{macrocode} |
| 679 |
|
| 680 |
% Filehandles needed for reading and writing temporary files |
| 681 |
% \begin{macrocode} |
| 682 |
\newwrite\rjlpshap@writer % holds \parshape lengths |
| 683 |
\newwrite\rjlpshap@finalwriter % holds entire \parshape command |
| 684 |
\newread\rjlpshap@reader % used to read writer to transfer to finalwriter |
| 685 |
% \end{macrocode} |
| 686 |
|
| 687 |
% rjlpshap@linecount: counts the number of lines read in the environment |
| 688 |
% \begin{macrocode} |
| 689 |
\newcounter{rjlpshap@linecount} |
| 690 |
% \end{macrocode} |
| 691 |
|
| 692 |
% Finally, undo the effect of $\backslash$makeatletter, and make @ be |
| 693 |
% a special character again. |
| 694 |
% \begin{macrocode} |
| 695 |
\makeatother |
| 696 |
% \end{macrocode} |
| 697 |
|
| 698 |
% \Finale |
| 699 |
\endinput |