| 1 | % \begin{meta-comment} |
| 2 | % |
| 3 | % $Id: sverb.dtx,v 1.1 2002/02/03 20:49:03 mdw Exp $ |
| 4 | % |
| 5 | % Verbatim typesetting done properly (ahem) |
| 6 | % |
| 7 | % (c) 1996 Mark Wooding |
| 8 | % |
| 9 | %----- Revision history ----------------------------------------------------- |
| 10 | % |
| 11 | % $Log: sverb.dtx,v $ |
| 12 | % Revision 1.1 2002/02/03 20:49:03 mdw |
| 13 | % Checkin for new build system. |
| 14 | % |
| 15 | % Revision 1.3 1996/11/19 21:01:18 mdw |
| 16 | % Entered into RCS |
| 17 | % |
| 18 | % |
| 19 | % \end{meta-comment} |
| 20 | % |
| 21 | % \begin{meta-comment} <general public licence> |
| 22 | %% |
| 23 | %% sverb package -- handling of verbatim text |
| 24 | %% Copyright (c) 1996 Mark Wooding |
| 25 | %% |
| 26 | %% This program is free software; you can redistribute it and/or modify |
| 27 | %% it under the terms of the GNU General Public License as published by |
| 28 | %% the Free Software Foundation; either version 2 of the License, or |
| 29 | %% (at your option) any later version. |
| 30 | %% |
| 31 | %% This program is distributed in the hope that it will be useful, |
| 32 | %% but WITHOUT ANY WARRANTY; without even the implied warranty of |
| 33 | %% MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
| 34 | %% GNU General Public License for more details. |
| 35 | %% |
| 36 | %% You should have received a copy of the GNU General Public License |
| 37 | %% along with this program; if not, write to the Free Software |
| 38 | %% Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. |
| 39 | %% |
| 40 | % \end{meta-comment} |
| 41 | % |
| 42 | % \begin{meta-comment} <Package preamble> |
| 43 | %<+package>\NeedsTeXFormat{LaTeX2e} |
| 44 | %<+package>\ProvidesPackage{sverb} |
| 45 | %<+package> [1996/05/08 1.3 Verbatim typesetting] |
| 46 | % \end{meta-comment} |
| 47 | % |
| 48 | % \CheckSum{651} |
| 49 | %% \CharacterTable |
| 50 | %% {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 |
| 51 | %% 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 |
| 52 | %% Digits \0\1\2\3\4\5\6\7\8\9 |
| 53 | %% Exclamation \! Double quote \" Hash (number) \# |
| 54 | %% Dollar \$ Percent \% Ampersand \& |
| 55 | %% Acute accent \' Left paren \( Right paren \) |
| 56 | %% Asterisk \* Plus \+ Comma \, |
| 57 | %% Minus \- Point \. Solidus \/ |
| 58 | %% Colon \: Semicolon \; Less than \< |
| 59 | %% Equals \= Greater than \> Question mark \? |
| 60 | %% Commercial at \@ Left bracket \[ Backslash \\ |
| 61 | %% Right bracket \] Circumflex \^ Underscore \_ |
| 62 | %% Grave accent \` Left brace \{ Vertical bar \| |
| 63 | %% Right brace \} Tilde \~} |
| 64 | %% |
| 65 | % |
| 66 | % \begin{meta-comment} |
| 67 | % |
| 68 | %<*driver> |
| 69 | \input{mdwtools} |
| 70 | \describespackage{sverb} |
| 71 | \mdwdoc |
| 72 | %</driver> |
| 73 | % |
| 74 | % \end{meta-comment} |
| 75 | % |
| 76 | % \section{User guide} |
| 77 | % |
| 78 | % The \package{sverb} package provides some useful commands and environments |
| 79 | % for doing things with verbatim text. I prefer this code to the standard |
| 80 | % \package{verbatim} package (by Rainer Sch\"opf et al.)\ although I'm |
| 81 | % biased. |
| 82 | % |
| 83 | % The package was written to fulfil a particular purpose: I wanted to be able |
| 84 | % to typeset ARM assembler code, 77~columns wide, on A5~paper, with the |
| 85 | % fields separated by \textit{tab} characters. It's grown up fairly |
| 86 | % organically from that, and I've tidied it when I've seen the code get too |
| 87 | % ugly. |
| 88 | % |
| 89 | % The current features are: |
| 90 | % |
| 91 | % \begin{itemize} |
| 92 | % |
| 93 | % \item A `listing' environment which typesets verbatim text nicely. |
| 94 | % |
| 95 | % \item A command to read verbatim text from an external file. |
| 96 | % |
| 97 | % \item Support for arbitrary-sized chunks of text without overflowing \TeX's |
| 98 | % memory. |
| 99 | % |
| 100 | % \item Support for \textit{tab} characters in the verbatim text. |
| 101 | % |
| 102 | % \item An environment for typesetting demonstrations of \LaTeX\ markup. |
| 103 | % |
| 104 | % \item It all works correctly with the \package{doc} system for documenting |
| 105 | % \LaTeX\ packages. |
| 106 | % |
| 107 | % \item A fairly hairy but quite powerful programmer interface to the yukky |
| 108 | % bits of the package. |
| 109 | % |
| 110 | % \end{itemize} |
| 111 | % |
| 112 | % The interface is described in its own section, so that more timid readers |
| 113 | % can avoid it. That said, some of the stuff in this section gets rather |
| 114 | % technical. |
| 115 | % |
| 116 | % Note that this package doesn't even try to do anything with short bits of |
| 117 | % verbatim text (as handled by the |\verb:...:| command). I have a separate |
| 118 | % package (\package{syntax}) which does all sorts of horrible things along |
| 119 | % those lines. |
| 120 | % |
| 121 | % \subsection{The \env{listing} environment} |
| 122 | % |
| 123 | % \DescribeEnv{listing} |
| 124 | % The main method for typesetting verbatim text is the \env{listing} |
| 125 | % environment. This works pretty much the same as the standard |
| 126 | % \env{verbatim} environment, with some exceptions, which are described |
| 127 | % below. |
| 128 | % |
| 129 | % So that you know exactly what you're getting, here are the rules by which |
| 130 | % \package{sverb} decides what the verbatim text actually is: |
| 131 | % |
| 132 | % \begin{itemize} |
| 133 | % |
| 134 | % \item If there's any text, other than spaces, on the same line as the |
| 135 | % `|\begin{listing}|', then the contents of the environment begins |
| 136 | % immediately after the closing brace (with all leading spaces |
| 137 | % preserved). Otherwise, the text begins on the following line. |
| 138 | % |
| 139 | % \item If there is any text, other than spaces, before the |
| 140 | % `|\end{listing}|', but on the same line, this is considered to be the |
| 141 | % last line of the text; otherwise the text is presumed to have ended |
| 142 | % at the end of the previous line. |
| 143 | % |
| 144 | % \item Any text following the |\end{listing}| on the same line is thrown |
| 145 | % away. There are good reasons for this, but they're technical. |
| 146 | % Essentially there's nothing I can do about it. |
| 147 | % |
| 148 | % \end{itemize} |
| 149 | % |
| 150 | % \begin{figure} |
| 151 | % \begin{demo}[w]{The \env{listing} environment} |
| 152 | %\dots in the following code: |
| 153 | % |
| 154 | %\begin{listing} |
| 155 | %init MOV R0,#200 ;Version 2.00 please |
| 156 | % LDR R1,=&4B534154 ;Magic number (`TASK') |
| 157 | % ADR R2,appName ;Find application name |
| 158 | % SWI Wimp_Initialise ;Register as a WIMP task |
| 159 | %\end{listing} |
| 160 | % |
| 161 | %The next step is to \dots |
| 162 | % \end{demo} |
| 163 | % \end{figure} |
| 164 | % |
| 165 | % Tab characters are supported within the environment: tab stops are set |
| 166 | % every eighth column, although this can be modified. |
| 167 | % |
| 168 | % \subsubsection{Configuring the \env{listing} environment} |
| 169 | % |
| 170 | % The text size used in the \env{listing} environment is set by the |
| 171 | % |\listingsize| command. By default, this is set to |\small|, although you |
| 172 | % can redefine it in the document preamble, or it can be set in the document |
| 173 | % class. |
| 174 | % |
| 175 | % The amount by which the listing text is indented is controlled by the |
| 176 | % |\listingindent| length parameter. This is a fixed length, whose default |
| 177 | % value is 1\,em. |
| 178 | % |
| 179 | % \subsubsection{Choosing a different end-text} |
| 180 | % |
| 181 | % \DescribeEnv{listing*} |
| 182 | % The \env{listing} environment is terminated by the exact character sequence |
| 183 | % `|\end{listing}|'. This isn't too much of a problem, unless you want to |
| 184 | % include this string in the text. This is achieved by the \env{listing$*$} |
| 185 | % environment, which allows you to specify the end-text to find as an |
| 186 | % argument. |
| 187 | % |
| 188 | % For example: |
| 189 | % |
| 190 | % \begin{demo}{The \env{listing$*$} environment} |
| 191 | %Type a listing as follows: |
| 192 | % |
| 193 | %\begin{listing*}{<end-listing*>} |
| 194 | %\begin{listing} |
| 195 | %This is a listing. Yes. |
| 196 | %\end{listing} |
| 197 | %<end-listing*> |
| 198 | %\end{demo} |
| 199 | % |
| 200 | % Don't include `special' characters in your chosen end-text unless you know |
| 201 | % what you're doing. |
| 202 | % |
| 203 | % \subsection{Writing text to a file} |
| 204 | % |
| 205 | % \DescribeEnv{verbwrite} |
| 206 | % You can write verbatim text to a file using the \env{verbwrite} |
| 207 | % environment. The syntax is fairly straightforward: |
| 208 | % |
| 209 | % \begin{quote} |
| 210 | % \syntax{"\\begin{verbwrite}{"<file-name>"}" \dots "\\end{verbwrite}"} |
| 211 | % \end{quote} |
| 212 | % |
| 213 | % The text of the environment is written to the named file. The rules about |
| 214 | % where the text actually starts and ends are the same as for the |
| 215 | % \env{listing} environment. |
| 216 | % |
| 217 | % There is also a $*$-variant, like \env{listing$*$}, which allows you to |
| 218 | % choose the end-text. The end-text is the first argument, the filename |
| 219 | % comes second. |
| 220 | % |
| 221 | % There is a restriction on the characters you can write to the file: they |
| 222 | % must all be considered `printable' by \TeX; otherwise they will be read |
| 223 | % back in as `\syntax{"^^"<chars>}' which isn't too good. Unfortunately, |
| 224 | % this includes tab characters, so you can't write them.\footnote{^^A |
| 225 | % Well, not without doing serious surgery on \TeX\ itself, anyway. } |
| 226 | % |
| 227 | % \iffalse [Example time... Ho hum. There is evilness here.] \fi |
| 228 | %\begin{verbwrite*}{<end-write>}{wrdemo1.tmp} |
| 229 | %\begin{verbwrite}{wrdemo.tmp} |
| 230 | %This is some text written to |
| 231 | %a file near the beginning of |
| 232 | %the file. |
| 233 | %\end{verbwrite} |
| 234 | %<end-write> |
| 235 | % |
| 236 | % For example: \verbinput{wrdemo1.tmp} |
| 237 | % |
| 238 | % \input{wrdemo1.tmp} \iffalse [Now build the file ;-) ] \fi |
| 239 | % |
| 240 | % \subsection{The \cmd\verbinput\ command} |
| 241 | % |
| 242 | % \DescribeMacro{\verbinput} |
| 243 | % You can input a pre-prepared text file exactly as it is in the input using |
| 244 | % the |\verbinput| command. The filename is given as an argument. For |
| 245 | % example: |
| 246 | % |
| 247 | % \begin{demo}{The \cmd\verbinput\ command} |
| 248 | %\verbinput{wrdemo.tmp} |
| 249 | % \end{demo} |
| 250 | % |
| 251 | % \subsection{The \env{demo} environment} |
| 252 | % |
| 253 | % Package authors need to document their packages, and it's common to want |
| 254 | % to display examples showing the original text and the output side-by-side |
| 255 | % (or, when space doesn't permit this, one above the other). Both the |
| 256 | % \LaTeX\ book and \textit{The \LaTeX\ Companion} contain such examples. |
| 257 | % |
| 258 | % The \env{demo} environment allows such displays to be created easily. The |
| 259 | % syntax of the environment is as follows: |
| 260 | % |
| 261 | % \begin{quote} |
| 262 | % \syntax{"\\begin{demo}["<shape>"]{"<title>"}" \dots "\\end{demo}"} |
| 263 | % \end{quote} |
| 264 | % |
| 265 | % The optional \synt{shape} argument can be either `|w|' (wide), or `|n|' |
| 266 | % (narrow). A `wide' shape places the input and output one above the other, |
| 267 | % while the `narrow' shape puts them side-by-side. The default shape is |
| 268 | % `narrow'. An attractive border is drawn around the display to finish it |
| 269 | % off nicely. |
| 270 | % |
| 271 | % An example: |
| 272 | % |
| 273 | %\begin{demo*}{<end-demo>}[w]{The \env{demo} environment} |
| 274 | %\begin{demo}{From the \textit{\TeX book}} |
| 275 | %\[ \sum_{p\;\rm prime} |
| 276 | % f(p) = \int_{t>1} |
| 277 | % f(t)\,{\rm d}\pi(t) \] |
| 278 | %\end{demo} |
| 279 | %<end-demo> |
| 280 | % |
| 281 | % \DescribeEnv{demo*} |
| 282 | % As with the other environments created by this package, there's a |
| 283 | % $*$-variant which takes the end-text as an argument. |
| 284 | % |
| 285 | % |
| 286 | % \section{Programmer interface} |
| 287 | % |
| 288 | % This section describes the publicly available routines provided by the |
| 289 | % \package{sverb} package. Routines not described here are libable to be |
| 290 | % changed or even removed without warning, so don't use them. |
| 291 | % |
| 292 | % \subsection{Environment hooks} |
| 293 | % |
| 294 | % Each of the environments created here works in the same way. For each |
| 295 | % environment \env{foo}, there's a main command responsible for doing the |
| 296 | % work, called |\sv@foo|. This is given all the arguments of the normal |
| 297 | % environment, and two more: |
| 298 | % |
| 299 | % \begin{itemize} |
| 300 | % |
| 301 | % \item The `end-text' to search for, which marks the end of the environment. |
| 302 | % |
| 303 | % \item Some actions to perform after the text has been read and processed. |
| 304 | % This allows the calling macro to do some extra actions, like closing |
| 305 | % boxes, etc. |
| 306 | % |
| 307 | % \end{itemize} |
| 308 | % |
| 309 | % All the environments do is call the main command with appropriate |
| 310 | % arguments. |
| 311 | % |
| 312 | % \subsection{Reading the verbatim text} |
| 313 | % |
| 314 | % \DescribeMacro{\sv@read} |
| 315 | % The main scanning routine is |\sv@read|. It is called with three |
| 316 | % arguments: |
| 317 | % |
| 318 | % \begin{itemize} |
| 319 | % |
| 320 | % \item The end-text marking the end of the environment. |
| 321 | % |
| 322 | % \item The name of a macro (which must be a single token) which is called |
| 323 | % with a line of text as its single argument. This is given each |
| 324 | % line of text which is read from the environment in turn. |
| 325 | % |
| 326 | % \item A macro, or other sort of action, which is to be done when the text |
| 327 | % has been read and processed. |
| 328 | % |
| 329 | % \end{itemize} |
| 330 | % |
| 331 | % The macro |\sv@read| assumes that the caller has already made some |
| 332 | % provision for removing the category codes of the following text, by either |
| 333 | % calling |\@verbatim| or using the construction |
| 334 | % \begin{listing} |
| 335 | %\let\do=\@makeother |
| 336 | %\dospecials |
| 337 | % \end{listing} |
| 338 | % |
| 339 | % \DescribeMacro{\sv@safespc} |
| 340 | % Note that any space characters you read using |\sv@read| will be catcoded |
| 341 | % as |\active|. Normally this is OK because |\obeyspaces| (or |
| 342 | % |\@vobeyspaces|) will be in effect. If you're doing something more exotic, |
| 343 | % like writing text to a file or building a command string, you can call |
| 344 | % |\sv@safespc| which defines the active-space character to be a normal |
| 345 | % whitespace-space when expanded. |
| 346 | % |
| 347 | % \implementation |
| 348 | % |
| 349 | % \section{Implementation} |
| 350 | % |
| 351 | % This section defines several macros and environments which allow verbatim |
| 352 | % typing, with a high degree of configurability. OK, so this sort of |
| 353 | % thing's been done so often before that it isn't true, but I don't really |
| 354 | % care. |
| 355 | % |
| 356 | % \begin{macrocode} |
| 357 | %<*package> |
| 358 | % \end{macrocode} |
| 359 | % |
| 360 | % \subsection{Simple things} |
| 361 | % |
| 362 | % To help us build funny macros which involve strange and different category |
| 363 | % codes, I'll write some simple macros which I can use while building my |
| 364 | % complicated and clever ones. |
| 365 | % |
| 366 | % \begin{macro}{\@cspecials} |
| 367 | % |
| 368 | % This macro is used to assist the definition of some of the environments. |
| 369 | % It makes `|\|', `|{|' and `|}|' into `other' characters, and replaces them |
| 370 | % with `\verb"|"', `|<|' and `|>|' respectively. Note that `|[|' and `|]|' |
| 371 | % aren't used, because they make defining commands which take optional |
| 372 | % arguments awkward. Note that we open a group here. This should be closed |
| 373 | % using \verb"|endgroup" at the end of the special section. |
| 374 | % |
| 375 | % \begin{macrocode} |
| 376 | \def\@cspecials{% |
| 377 | \begingroup% |
| 378 | \catcode`|0% |
| 379 | \catcode`<1% |
| 380 | \catcode`>2% |
| 381 | \catcode`\{12% |
| 382 | \catcode`\}12% |
| 383 | \catcode`\\12% |
| 384 | } |
| 385 | % \end{macrocode} |
| 386 | % \end{macro} |
| 387 | % |
| 388 | % \begin{macro}{\sv@startlisting} |
| 389 | % |
| 390 | % This macro sets everything up nicely for a \env{listing}-type verbatim |
| 391 | % environment. |
| 392 | % |
| 393 | % \begin{macrocode} |
| 394 | \def\sv@startlisting{% |
| 395 | \def\par{\@@par\penalty\interlinepenalty}% |
| 396 | \@@par% |
| 397 | \leftskip\@totalleftmargin% |
| 398 | \obeylines% |
| 399 | \@noligs% |
| 400 | \let\do\@makeother\dospecials% |
| 401 | \verbatim@font% |
| 402 | \frenchspacing% |
| 403 | \@vobeyspaces% |
| 404 | \settabwidth% |
| 405 | \catcode9\active% |
| 406 | \lccode`\~9\lowercase{\let~\sv@vtab}% |
| 407 | \lccode`\~13\lowercase{\let~\vinput@cr}% |
| 408 | \interlinepenalty500% |
| 409 | } |
| 410 | % \end{macrocode} |
| 411 | % |
| 412 | % \end{macro} |
| 413 | % |
| 414 | % \subsection{Tab character handling} |
| 415 | % |
| 416 | % One of the things we want to do here is handle tab characters properly. |
| 417 | % (Here, `properly' means `moving to the next column which is a multiple of |
| 418 | % eight', the way these things were always meant to.) |
| 419 | % |
| 420 | % \begin{macro}{\settabwidth} |
| 421 | % |
| 422 | % The tabs used by our tabbed verbatim environments are set up by this |
| 423 | % routine. It sets the tab width parameter |\svtab| to 8 times the width |
| 424 | % of a |\tt| space. If you really want, you can redefine this macro. |
| 425 | % |
| 426 | % \begin{macrocode} |
| 427 | \newdimen\svtab |
| 428 | \def\settabwidth{\setbox\z@\hbox{\texttt{\space}}\svtab8\wd\z@} |
| 429 | % \end{macrocode} |
| 430 | % |
| 431 | % \end{macro} |
| 432 | % |
| 433 | % \begin{macro}{\sv@vtab} |
| 434 | % |
| 435 | % Here we handle tabs inside verbatim environments. We expect each line to |
| 436 | % be typeset as a box, using something like |
| 437 | % |
| 438 | % \begin{listing} |
| 439 | %\setbox0\hbox{#1} |
| 440 | %\leavevmode |
| 441 | %\box0 |
| 442 | %\par |
| 443 | % \end{listing} |
| 444 | % |
| 445 | % The idea is that you make tab active, and set it to this macro. We stop |
| 446 | % the current box, stretch it to the right width, and start another one |
| 447 | % straight after, so nobody know the difference. The code here is straight |
| 448 | % from Appendix~D of \textit{The \TeX book}. |
| 449 | % |
| 450 | % \begin{macrocode} |
| 451 | \def\sv@vtab{% |
| 452 | \hfill\egroup% |
| 453 | \@tempdima\wd\z@% |
| 454 | \divide\@tempdima\svtab% |
| 455 | \multiply\@tempdima\svtab% |
| 456 | \advance\@tempdima\svtab% |
| 457 | \wd\z@\@tempdima% |
| 458 | \leavevmode\box\z@% |
| 459 | \setbox\z@\hbox\bgroup% |
| 460 | } |
| 461 | % \end{macrocode} |
| 462 | % |
| 463 | % \end{macro} |
| 464 | % |
| 465 | % \begin{macro}{\verbinput} |
| 466 | % |
| 467 | % We allow input from a file, by the |\verbinput| command. We display the |
| 468 | % text pretty much the same as the \env{listing} environment below. |
| 469 | % |
| 470 | % We set tab and return active, and get them to do appropriate things. This |
| 471 | % isn't actually all that hard. |
| 472 | % |
| 473 | % \begin{macrocode} |
| 474 | \def\verbinput#1{% |
| 475 | \begin{listinglist}% |
| 476 | \listingsize% |
| 477 | \sv@startlisting% |
| 478 | \setbox\z@\hbox\bgroup% |
| 479 | \input{#1}% |
| 480 | \sv@stripspc% |
| 481 | \egroup% |
| 482 | \ifdim\wd\z@=\z@% |
| 483 | \ifhmode\par\fi% |
| 484 | \else% |
| 485 | \leavevmode\box\z@\par% |
| 486 | \fi% |
| 487 | \end{listinglist}% |
| 488 | } |
| 489 | % \end{macrocode} |
| 490 | % |
| 491 | % \end{macro} |
| 492 | % |
| 493 | % \begin{macro}{\vinput@cr} |
| 494 | % |
| 495 | % This macro handles return characters while inputting text in |\verbinput|. |
| 496 | % We just output our current box, and start another. |
| 497 | % |
| 498 | % \begin{macrocode} |
| 499 | \def\vinput@cr{% |
| 500 | \egroup% |
| 501 | \leavevmode\box\z@% |
| 502 | \par% |
| 503 | \setbox\z@\hbox\bgroup% |
| 504 | } |
| 505 | % \end{macrocode} |
| 506 | % |
| 507 | % \end{macro} |
| 508 | % |
| 509 | % \subsection{Reading verbatim text} |
| 510 | % |
| 511 | % The traditional way of reading verbatim text is to use a delimited |
| 512 | % argument, as described in the \textit{\TeX book}. This works well-ish if |
| 513 | % the text isn't very long. A better solution would be to pick out the text |
| 514 | % line-by-line and process it like that. So this is what we do. |
| 515 | % |
| 516 | % \begin{macro}{\matcher} |
| 517 | % |
| 518 | % For long verbatim environments, we need to be able to find the end text. |
| 519 | % This is rather tricky. The solution here is rather horrible. The |
| 520 | % environment picks out each line of the text at a time, as an argument, and |
| 521 | % tests to see if it contains the text we're after. We do the test in a |
| 522 | % particularly yukky way: we add the actual target text to the end of the |
| 523 | % line, and inspect the text following the match to see if the match is at |
| 524 | % the end. |
| 525 | % |
| 526 | % The |\matcher| macro creates a `matcher' which will test strings to see if |
| 527 | % they contain something interesting. |
| 528 | % |
| 529 | % To create a matcher, say |
| 530 | % \syntax{"\\matcher{"<cmd-name>"}{"<target>"}{"<process-cmd>"}"}. The |
| 531 | % command \synt{cmd-name} accepts a line of text as an argument and calls |
| 532 | % the \synt{process-cmd} with the text of the line before the match, or the |
| 533 | % whole lot. It also sets |\@ifmatched| appropriately. |
| 534 | % |
| 535 | % (Having spent ages coming up with this cruft myself, I found some very |
| 536 | % similar, but slightly better, code in Appendix~D. So I've changed mine to |
| 537 | % match Donald's. Anyway, credit where it's due: cheers Don.) |
| 538 | % |
| 539 | % \begin{macrocode} |
| 540 | \newif\if@matched |
| 541 | \def\matcher#1#2#3{% |
| 542 | \expandafter\def\csname\string#1$match\endcsname##1#2##2##3\end{% |
| 543 | \ifx##2\relax% |
| 544 | \@matchedfalse% |
| 545 | \else% |
| 546 | \@matchedtrue% |
| 547 | \fi% |
| 548 | #3{##1}% |
| 549 | }% |
| 550 | \expandafter\def\expandafter#1\expandafter##\expandafter1\expandafter{% |
| 551 | \csname\string#1$match\endcsname##1#2\relax\end% |
| 552 | }% |
| 553 | } |
| 554 | % \end{macrocode} |
| 555 | % |
| 556 | % \end{macro} |
| 557 | % |
| 558 | % \begin{macro}{\sv@stripspc} |
| 559 | % |
| 560 | % This macro strips any trailing glue in the current horizontal list. This |
| 561 | % is fairly simple, actually: we just loop while glue is the last item. It's |
| 562 | % slightly complicated by penalties which \TeX\ puts into the list between |
| 563 | % the glue items, but we just remove them too. |
| 564 | % |
| 565 | % \begin{macrocode} |
| 566 | \def\sv@stripspc{% |
| 567 | \unpenalty% |
| 568 | \ifdim\lastskip=\z@\else% |
| 569 | \unskip\expandafter\sv@stripspc% |
| 570 | \fi% |
| 571 | } |
| 572 | % \end{macrocode} |
| 573 | % |
| 574 | % \end{macro} |
| 575 | % |
| 576 | % \begin{macro}{\sv@percent} |
| 577 | % |
| 578 | % This macro strips a single leading percent character if there is one, and |
| 579 | % if the \env{doc} package is loaded. We store the possibly stripped text in |
| 580 | % |\@tempa|. |
| 581 | % |
| 582 | % \begin{macrocode} |
| 583 | \begingroup |
| 584 | \catcode`\%=12 |
| 585 | \gdef\sv@percent#1#2\relax |
| 586 | {\ifx\check@percent\@@undefined |
| 587 | \ifx#1\relax\def\@tempa{}\else |
| 588 | \def\@tempa{#1#2}\fi\else |
| 589 | \ifx#1\relax\def\@tempa{}\else |
| 590 | \ifx#1%\def\@tempa{#2}\else |
| 591 | \def\@tempa{#1#2}\fi\fi\fi} |
| 592 | \endgroup |
| 593 | % \end{macrocode} |
| 594 | % |
| 595 | % \end{macro} |
| 596 | % |
| 597 | % \begin{macro}{\@isspaces} |
| 598 | % |
| 599 | % We want to avoid writing the first and last lines of the environment to the |
| 600 | % file if there's nothing in them. To do this, we need to know whether a |
| 601 | % piece of text contains only space characters. This macro does this, in a |
| 602 | % rather nasty way. See the other macros below for details of how this |
| 603 | % works. |
| 604 | % |
| 605 | % We define |\sv@safespc| at the same time: this makes space active and |
| 606 | % expand to a space character which is not active. Neat, huh? |
| 607 | % |
| 608 | % \begin{macrocode} |
| 609 | \lccode`\~32 |
| 610 | \lccode`\!32 |
| 611 | \lowercase{% |
| 612 | \def\@isspaces#1{% |
| 613 | \ifx#1\relax% |
| 614 | \def\@tempb{\@tempswafalse}% |
| 615 | \else\ifx#1~% |
| 616 | \let\@tempb\@isspaces% |
| 617 | \else% |
| 618 | \def\@tempb##1\relax{}% |
| 619 | \fi\fi% |
| 620 | \@tempb% |
| 621 | } |
| 622 | \def\sv@safespc{% |
| 623 | \catcode32\active% |
| 624 | \def~{ }% |
| 625 | } |
| 626 | } |
| 627 | % \end{macrocode} |
| 628 | % |
| 629 | % \end{macro} |
| 630 | % |
| 631 | % \begin{macro}{\sv@read} |
| 632 | % |
| 633 | % This macro does the main job of reading a chunk of verbatim text. You call |
| 634 | % it like this: |
| 635 | % |
| 636 | % \begin{quote} |
| 637 | % \syntax{"\\sv@read{"<end-text>"}{"<process-line-proc>"}{"<end-proc>"}"} |
| 638 | % \end{quote} |
| 639 | % |
| 640 | % The \synt{end-text} is the text to find at the end of the `environment': we |
| 641 | % stop when we find it. |
| 642 | % |
| 643 | % The \synt{process-line-proc} is a macro which is passed as an argument each |
| 644 | % line which we read from the text. |
| 645 | % |
| 646 | % The \synt{end-proc} is a macro to call once we've finished reading all of |
| 647 | % the text. This can tidy up an environment or close a file or whatever. |
| 648 | % |
| 649 | % We read the text by picking out newlines using a delimited macro. We have |
| 650 | % to be a little clever, because newlines are active in verbatim text. |
| 651 | % |
| 652 | % We will also strip `|%|' signs off the beginning if the \package{doc} |
| 653 | % package is here (\package{doc} tries to play with \LaTeX's verbatim stuff, |
| 654 | % and doesn't understand the way we do things). |
| 655 | % |
| 656 | % \begin{macrocode} |
| 657 | \def\sv@read#1#2#3{% |
| 658 | % \end{macrocode} |
| 659 | % |
| 660 | % This code does all sorts of evil things, so I'll start by opening a group. |
| 661 | % |
| 662 | % \begin{macrocode} |
| 663 | \begingroup% |
| 664 | % \end{macrocode} |
| 665 | % |
| 666 | % So that I can spot the end-text, I'll create a matcher macro. |
| 667 | % |
| 668 | % \begin{macrocode} |
| 669 | \matcher\@match{#1}\sv@read@ii% |
| 670 | % \end{macrocode} |
| 671 | % |
| 672 | % So that I can identify line ends, I'll make them active. I'll also make |
| 673 | % spaces active so that they can expand to whatever they ought to expand |
| 674 | % to (spaces in files, or funny \verb*" " characters or whatever. |
| 675 | % |
| 676 | % \begin{macrocode} |
| 677 | \catcode13\active% |
| 678 | \catcode32\active% |
| 679 | % \end{macrocode} |
| 680 | % |
| 681 | % I'll use the |\if@tempswa| flag to tell me whether I ought to output the |
| 682 | % current line. This is a little messy, so I'll describe it later. I'll |
| 683 | % initialise it to false because this is the correct thing to do. |
| 684 | % |
| 685 | % \begin{macrocode} |
| 686 | \@tempswafalse% |
| 687 | % \end{macrocode} |
| 688 | % |
| 689 | % Most of the job is done by two submacros. I'll define them in terms of |
| 690 | % my current arguments (to save lots of token munging). The first just |
| 691 | % extracts the next line (which ends at the next newline character) and |
| 692 | % tries to match it. |
| 693 | % |
| 694 | % \begin{macrocode} |
| 695 | \lccode`\~13\lowercase{% |
| 696 | \def\sv@read@i##1~{\@match{##1}}% |
| 697 | }% |
| 698 | % \end{macrocode} |
| 699 | % |
| 700 | % The results of the match get passed here, along with the text of the |
| 701 | % line up to the matched text. |
| 702 | % |
| 703 | % \begin{macrocode} |
| 704 | \def\sv@read@ii##1{% |
| 705 | % \end{macrocode} |
| 706 | % |
| 707 | % The first job to do is to maybe strip off percent signs from the beginning, |
| 708 | % to keep \package{doc} happy. |
| 709 | % |
| 710 | % \begin{macrocode} |
| 711 | \sv@percent##1\relax\relax% |
| 712 | % \end{macrocode} |
| 713 | % |
| 714 | % Now I need to decide whether I ought to output this line. The method goes |
| 715 | % like this: if this is the first line (|\if@tempswa| is false) or the last |
| 716 | % (|\if@matched| is true), \emph{and} the text consists only of spaces, then |
| 717 | % I'll ignore it. |
| 718 | % |
| 719 | % The first thing to do is to notice the last line -- if |\if@matched| is |
| 720 | % true, then I'll make |\if@tempswa| false to make the first-line and |
| 721 | % last-line cases work the same way. |
| 722 | % |
| 723 | % \begin{macrocode} |
| 724 | \if@matched\@tempswafalse\fi% |
| 725 | % \end{macrocode} |
| 726 | % |
| 727 | % Now if this is the first or last line, I'll examine it for spaces. This |
| 728 | % is done in a separate macro. It will set |\if@tempswa| false if the |
| 729 | % text contains only spaces. |
| 730 | % |
| 731 | % \begin{macrocode} |
| 732 | \if@tempswa\else\@tempswatrue\expandafter\@isspaces\@tempa\relax\fi% |
| 733 | % \end{macrocode} |
| 734 | % |
| 735 | % Now, if |\if@tempswa| is still true, perform the \<process-line-proc> on |
| 736 | % the line of text. I'll provide a group, so that it doesn't upset me |
| 737 | % too much. |
| 738 | % |
| 739 | % \begin{macrocode} |
| 740 | \if@tempswa% |
| 741 | \begingroup% |
| 742 | \expandafter#2\expandafter{\@tempa}% |
| 743 | \endgroup% |
| 744 | \fi% |
| 745 | % \end{macrocode} |
| 746 | % |
| 747 | % The next line won't be the first one, so I'll set the flag true in |
| 748 | % readiness. |
| 749 | % |
| 750 | % \begin{macrocode} |
| 751 | \@tempswatrue% |
| 752 | % \end{macrocode} |
| 753 | % |
| 754 | % Now, if that wasn't the last line, go round again; otherwise end the group |
| 755 | % I started ages ago, and do the user's \<end-proc>. |
| 756 | % |
| 757 | % \begin{macrocode} |
| 758 | \if@matched\def\@tempa{\endgroup#3}\else\let\@tempa\sv@read@i\fi% |
| 759 | \@tempa% |
| 760 | }% |
| 761 | % \end{macrocode} |
| 762 | % |
| 763 | % Now to start the thing up. I'll read the first line. |
| 764 | % |
| 765 | % \begin{macrocode} |
| 766 | \sv@read@i% |
| 767 | } |
| 768 | % \end{macrocode} |
| 769 | % |
| 770 | % \end{macro} |
| 771 | % |
| 772 | % \begin{macro}{\sv@readenv} |
| 773 | % |
| 774 | % This macro works out an appropriate end-text for the current environment. |
| 775 | % If you say \syntax{"\\sv@readenv{"<macro-name>"}"}, it will expand do |
| 776 | % \begin{listinglist} \listingsize \synshorts |
| 777 | % <macro-name>"{\\"$_{12}$"end{"$_{12}$<current-env-name>"}"$_{12}$"}"^^A |
| 778 | % "{\\end{"<current-env-name>"}}" |
| 779 | % \end{listinglist} |
| 780 | % Easy, no? |
| 781 | % |
| 782 | % This is all done with mirrors. No, err\dots\ it's done with |
| 783 | % |\expandafter|. |
| 784 | % |
| 785 | % \begin{macrocode} |
| 786 | \begingroup |
| 787 | \lccode`\<=`\{ |
| 788 | \lccode`\>=`\} |
| 789 | \lccode`\|=`\\ |
| 790 | \lowercase{\endgroup |
| 791 | \def\sv@readenv#1{% |
| 792 | \expandafter\expandafter\expandafter% |
| 793 | #1\expandafter\sv@readenv@i\@currenvir\@@% |
| 794 | } |
| 795 | \def\sv@readenv@i#1\@@{{|end<#1>}{\end{#1}}} |
| 796 | } |
| 797 | % \end{macrocode} |
| 798 | % |
| 799 | % \end{macro} |
| 800 | % |
| 801 | % \begin{macro}{\sv@verbline} |
| 802 | % |
| 803 | % This macro typesets a line in a verbatim way, so you can construct a real |
| 804 | % verbatim environment from it. It's a bit tricky in the way that it catches |
| 805 | % the last line. Don't worry about this: it's easy really. Note the |
| 806 | % |\relax| after the |\par| -- this is because \package{doc} tries to do |
| 807 | % clever things with |\par| to strip `|%|' signs out. |
| 808 | % |
| 809 | % \begin{macrocode} |
| 810 | \def\sv@verbline#1{% |
| 811 | \setbox\z@\hbox{#1\sv@stripspc}% |
| 812 | \ifdim\wd\z@=\z@% |
| 813 | \if@matched\ifhmode\par\relax\fi\else\leavevmode\par\relax\fi% |
| 814 | \else% |
| 815 | \leavevmode\box\z@\par\relax% |
| 816 | \fi% |
| 817 | } |
| 818 | % \end{macrocode} |
| 819 | % |
| 820 | % \end{macro} |
| 821 | % |
| 822 | % \subsection{Listing environments} |
| 823 | % |
| 824 | % The \env{listing} environment is our equivalent of the standard |
| 825 | % \env{verbatim} environment. We do some slightly cleverer things, though, |
| 826 | % to make sure (for example) that even text which contains |\end{listing}| |
| 827 | % can be typeset. |
| 828 | % |
| 829 | % \begin{macro}{\listinglist} |
| 830 | % \begin{environment}{listinglist} |
| 831 | % |
| 832 | % This defines the layout for the \env{listing} environment. It starts a |
| 833 | % list with the appropriate shape. It's also made into an environment, so |
| 834 | % that the end-paragraph-environment bits work correctly. |
| 835 | % |
| 836 | % The |\listingindent| length parameter sets up the indentation of the |
| 837 | % listings. If there's a |\parindent| setting, I'll line listings up with |
| 838 | % that; otherwise I'll just choose something which looks right. |
| 839 | % |
| 840 | % \begin{macrocode} |
| 841 | \newdimen\listingindent |
| 842 | \AtBeginDocument{% |
| 843 | \ifdim\parindent=\z@\listingindent1em\else\listingindent\parindent\fi% |
| 844 | } |
| 845 | % \end{macrocode} |
| 846 | % |
| 847 | % Now to define a size hook for the environment. This is fairly simple |
| 848 | % stuff. |
| 849 | % |
| 850 | % \begin{macrocode} |
| 851 | \ifx\listingsize\@@undefined |
| 852 | \let\listingsize\small |
| 853 | \fi |
| 854 | % \end{macrocode} |
| 855 | % |
| 856 | % Now to define the environment itself. Suppress the indentation if we're |
| 857 | % first thing on a new list item, so that the listing lines up with |
| 858 | % everything else. |
| 859 | % |
| 860 | % \begin{macrocode} |
| 861 | \def\listinglist{% |
| 862 | \list{}{% |
| 863 | \if@inlabel% |
| 864 | \leftmargin\z@% |
| 865 | \else% |
| 866 | \leftmargin\listingindent% |
| 867 | \fi% |
| 868 | \rightmargin\z@% |
| 869 | \labelwidth\z@% |
| 870 | \labelsep\z@% |
| 871 | \itemindent\z@% |
| 872 | \listparindent\z@% |
| 873 | \let\makelabel\relax% |
| 874 | \parsep\z@skip% |
| 875 | }% |
| 876 | \parfillskip\@flushglue% |
| 877 | \item\relax% |
| 878 | } |
| 879 | \let\endlistinglist\endlist |
| 880 | % \end{macrocode} |
| 881 | % |
| 882 | % \end{environment} |
| 883 | % \end{macro} |
| 884 | % |
| 885 | % \begin{environment}{listing} |
| 886 | % |
| 887 | % The \env{listing} environment is the only real verbatim-like environment we |
| 888 | % create will all this kit, although it does the job very nicely. |
| 889 | % |
| 890 | % The environment indents its contents slightly, unlike \env{verbatim}, and |
| 891 | % uses a smaller typeface in an attempt to fit 77-column text on an A5~page. |
| 892 | % There is also a $*$-variant, which allows you to specify the terminating |
| 893 | % text. This enables you to include absolutely any text in the environment, |
| 894 | % including |\end{listing}|. |
| 895 | % |
| 896 | % First, we must define the |\listing| command. |
| 897 | % |
| 898 | % \begin{macrocode} |
| 899 | \def\listing{% |
| 900 | \listinglist% |
| 901 | \listingsize% |
| 902 | \sv@readenv\sv@listing% |
| 903 | } |
| 904 | % \end{macrocode} |
| 905 | % |
| 906 | % Now we define the |\@listing| command, which does most of the work. We |
| 907 | % base the \env{listing} environment on a \env{list}. |
| 908 | % |
| 909 | % \begin{macrocode} |
| 910 | \def\sv@listing#1#2{% |
| 911 | \sv@startlisting% |
| 912 | \sv@read{#1}\sv@verbline{\endlistinglist#2}% |
| 913 | } |
| 914 | % \end{macrocode} |
| 915 | % |
| 916 | % Now we define the starred version. The command name needs to include the |
| 917 | % `|*|' character, so we must use |\csname|. There's some hacking here to |
| 918 | % allow us to read the name using the appropriate catcodes for otherwise |
| 919 | % normal characters: \LaTeX\ activates some characters and makes them typeset |
| 920 | % themselves to suppress some ligaturing. |
| 921 | % |
| 922 | % \begin{macrocode} |
| 923 | \expandafter\def\csname listing*\endcsname{% |
| 924 | \listinglist% |
| 925 | \listingsize% |
| 926 | \begingroup% |
| 927 | \@noligs% |
| 928 | \def\@tempa##1{\endgroup\sv@listing{##1}{\end{listing*}}}% |
| 929 | \@tempa% |
| 930 | } |
| 931 | % \end{macrocode} |
| 932 | % |
| 933 | % \end{environment} |
| 934 | % |
| 935 | % \begin{environment}{ignore} |
| 936 | % |
| 937 | % The \env{ignore} environment entirely ignores its contents. Anything at |
| 938 | % all may be put into the environment: it is discarded utterly. |
| 939 | % |
| 940 | % We define some macros for defining ignoring environments, because this can |
| 941 | % be useful for version control, possibly. |
| 942 | % |
| 943 | % \begin{macrocode} |
| 944 | \def\sv@ignore#1#2{% |
| 945 | \@bsphack% |
| 946 | \let\do\@makeother\dospecials% |
| 947 | \sv@read{#1}\@gobble{\@esphack#2}% |
| 948 | } |
| 949 | \def\ignore{\sv@readenv\sv@ignore} |
| 950 | \def\ignoreenv#1{% |
| 951 | \expandafter\let\csname #1\endcsname\ignore% |
| 952 | } |
| 953 | \def\unignoreenv#1{% |
| 954 | \expandafter\def\csname #1\endcsname{\endgroup}% |
| 955 | \expandafter\def\csname end#1\endcsname% |
| 956 | {\begingroup\def\@currenvir{#1}}% |
| 957 | } |
| 958 | % \end{macrocode} |
| 959 | % |
| 960 | % \end{environment} |
| 961 | % |
| 962 | % \subsection{The \env{verbwrite} environment} |
| 963 | % |
| 964 | % The \env{verbwrite} environment allows text to be written to a file in a |
| 965 | % verbatim way. Note that tab characters don't work, because \TeX\ refuses |
| 966 | % to be nice. |
| 967 | % |
| 968 | % \begin{macro}{\sv@write} |
| 969 | % |
| 970 | % As seems to be traditional now, we first define a general hookable macro |
| 971 | % which allows a caller to specify the end-text and what to do afterwards. |
| 972 | % |
| 973 | % \begin{macrocode} |
| 974 | \newwrite\sv@writefile |
| 975 | \def\sv@write#1#2{% |
| 976 | \begingroup% |
| 977 | \@bsphack% |
| 978 | \let\do\@makeother\dospecials% |
| 979 | \sv@safespc% |
| 980 | \sv@read{#1}\sv@writeline{\sv@endwrite#2}% |
| 981 | } |
| 982 | \def\sv@writeline#1{% |
| 983 | \immediate\write\sv@writefile{#1}% |
| 984 | } |
| 985 | \def\sv@endwrite{% |
| 986 | \@esphack% |
| 987 | \endgroup% |
| 988 | } |
| 989 | % \end{macrocode} |
| 990 | % |
| 991 | % \end{macro} |
| 992 | % |
| 993 | % \begin{environment}{verbwrite} |
| 994 | % |
| 995 | % Now we can define the actual environment. We define a $*$-variant which |
| 996 | % allows the user to specify the end-text, just to make sure. |
| 997 | % |
| 998 | % \begin{macrocode} |
| 999 | \def\verbwrite#1{% |
| 1000 | \immediate\openout\sv@writefile#1\relax% |
| 1001 | \sv@readenv\sv@write% |
| 1002 | } |
| 1003 | \def\endverbwrite{\immediate\closeout\sv@writefile} |
| 1004 | \expandafter\def\csname verbwrite*\endcsname#1#2{% |
| 1005 | \immediate\openout\sv@writefile#2\relax% |
| 1006 | \sv@write{#1}{\immediate\closeout\sv@writefile\end{verbwrite*}}% |
| 1007 | } |
| 1008 | % \end{macrocode} |
| 1009 | % |
| 1010 | % \end{environment} |
| 1011 | % |
| 1012 | % \subsection{The \env{demo} environment} |
| 1013 | % |
| 1014 | % By way of tying all of this together, I present an environment for |
| 1015 | % displaying demonstrations of \LaTeX\ markup. We read the contents of the |
| 1016 | % environment, write it to a temporary file, and read it back twice, |
| 1017 | % typesetting it the first time and displaying it verbatim the second time. |
| 1018 | % |
| 1019 | % \begin{macro}{\sv@demoname} |
| 1020 | % |
| 1021 | % This macro expands to the filename to use for the temporary data. To |
| 1022 | % allow the package documentation to demonstrate the \env{demo} environment |
| 1023 | % itself, we need to keep a nesting count. This avoids too much hackery, |
| 1024 | % which unfortunately appears to plague all of my \TeX\ code. |
| 1025 | % |
| 1026 | % \begin{macrocode} |
| 1027 | \newcount\sv@nestcount |
| 1028 | \def\sv@demoname{demo\number\sv@nestcount.tmp} |
| 1029 | % \end{macrocode} |
| 1030 | % |
| 1031 | % \end{macro} |
| 1032 | % |
| 1033 | % \begin{macro}{\sv@demo} |
| 1034 | % |
| 1035 | % As for listing, we do all the business through a private macro. This is |
| 1036 | % good because it means we can leave the main macro readable. The argument |
| 1037 | % is the end-text to spot. |
| 1038 | % |
| 1039 | % \begin{macrocode} |
| 1040 | \def\sv@demo#1#2{% |
| 1041 | \@ifnextchar[{\sv@demo@i{#1}{#2}}{\sv@demo@i{#1}{#2}[n]}% |
| 1042 | } |
| 1043 | \def\sv@demo@i#1#2[#3]#4{% |
| 1044 | \advance\sv@nestcount by\@ne% |
| 1045 | \immediate\openout\sv@writefile\sv@demoname\relax% |
| 1046 | \sv@write{#1}{% |
| 1047 | \immediate\closeout\sv@writefile% |
| 1048 | \sv@dodemo{#2}{#3}{#4}% |
| 1049 | }% |
| 1050 | } |
| 1051 | % \end{macrocode} |
| 1052 | % |
| 1053 | % \end{macro} |
| 1054 | % |
| 1055 | % \begin{environment}{demo} |
| 1056 | % |
| 1057 | % This is the real environment. We provide \env{demo$*$} too, to allow the |
| 1058 | % user to choose the end-text. |
| 1059 | % |
| 1060 | % \begin{macrocode} |
| 1061 | \def\demo{\sv@readenv\sv@demo} |
| 1062 | \expandafter\def\csname demo*\endcsname#1{\sv@demo{#1}{\end{demo*}}} |
| 1063 | % \end{macrocode} |
| 1064 | % |
| 1065 | % \end{environment} |
| 1066 | % |
| 1067 | % \begin{macro}{\sv@dodemo} |
| 1068 | % |
| 1069 | % First, let's define some common bits of code in the stuff below. The |
| 1070 | % minipages used to typeset the material has some clever stuff to avoid |
| 1071 | % strange spacing in the output. |
| 1072 | % |
| 1073 | % \begin{macrocode} |
| 1074 | \def\sv@demosmp{% |
| 1075 | \begin{minipage}[t]{\@tempdima}% |
| 1076 | \vskip8\p@% |
| 1077 | \hrule\@height\z@% |
| 1078 | \raggedright% |
| 1079 | \vbox\bgroup% |
| 1080 | } |
| 1081 | \def\sv@demoemp{% |
| 1082 | \par\unpenalty\unskip% |
| 1083 | \egroup% |
| 1084 | \vskip8\p@% |
| 1085 | \hrule\@height\z@% |
| 1086 | \end{minipage}% |
| 1087 | } |
| 1088 | % \end{macrocode} |
| 1089 | % |
| 1090 | % This is the macro which actually typesets the demonstration. |
| 1091 | % |
| 1092 | % \begin{macrocode} |
| 1093 | \def\sv@dodemo#1#2#3{% |
| 1094 | % \end{macrocode} |
| 1095 | % |
| 1096 | % Now work out some values. We set |\hsize| to the line width leaving 2\,em |
| 1097 | % of space on either side. The size of the minipages is calculated depending |
| 1098 | % on the shape of the demonstration. This is all fairly simple. |
| 1099 | % |
| 1100 | % \begin{macrocode} |
| 1101 | \begingroup% |
| 1102 | \@tempdima\linewidth% |
| 1103 | \advance\@tempdima-2em% |
| 1104 | \hsize\@tempdima% |
| 1105 | \if#2w% |
| 1106 | \advance\@tempdima-2em% |
| 1107 | \else% |
| 1108 | \advance\@tempdima-3em% |
| 1109 | \divide\@tempdima2% |
| 1110 | \fi% |
| 1111 | % \end{macrocode} |
| 1112 | % |
| 1113 | % Now we open a big vertical box, and put in a header to mark off the |
| 1114 | % demonstration. |
| 1115 | % |
| 1116 | % \begin{macrocode} |
| 1117 | \par% |
| 1118 | \setbox\z@\hbox{\strut\enspace#3\enspace\strut}% |
| 1119 | \@tempdimb.5\dp\z@% |
| 1120 | \advance\@tempdimb-.5\ht\z@% |
| 1121 | \ht\z@\@tempdimb\dp\z@\@tempdimb% |
| 1122 | \noindent\hskip1em\vtop{% |
| 1123 | \hb@xt@\hsize{% |
| 1124 | \hrulefill% |
| 1125 | \raise\@tempdimb\box\z@% |
| 1126 | \hrulefill% |
| 1127 | }% |
| 1128 | \nointerlineskip% |
| 1129 | \hb@xt@\hsize{\vrule\@height5\p@\hfil\vrule\@height5\p@}% |
| 1130 | \nointerlineskip% |
| 1131 | % \end{macrocode} |
| 1132 | % |
| 1133 | % Now we insert the output text in the first minipage. I'll force `|%|' |
| 1134 | % to be a comment character, in case something like \package{doc} has had its |
| 1135 | % wicked way. |
| 1136 | % |
| 1137 | % \begin{macrocode} |
| 1138 | \vskip-\parskip% |
| 1139 | \noindent\hbox{}\hskip1em% |
| 1140 | \sv@demosmp% |
| 1141 | \catcode`\%14\relax% |
| 1142 | \input{\sv@demoname}% |
| 1143 | \sv@demoemp% |
| 1144 | % \end{macrocode} |
| 1145 | % |
| 1146 | % Insert some kind of separation between the two. In `wide' format, we start |
| 1147 | % a new line, and put a ruleoff between the two. In `narrow' format, we just |
| 1148 | % leave some space. |
| 1149 | % |
| 1150 | % \begin{macrocode} |
| 1151 | \if#2w% |
| 1152 | \vskip8\p@\hrule\vskip8\p@% |
| 1153 | \noindent\hbox{}% |
| 1154 | \fi% |
| 1155 | \hskip1em% |
| 1156 | % \end{macrocode} |
| 1157 | % |
| 1158 | % Now we put the verbatim copy of the text in the other minipage. |
| 1159 | % |
| 1160 | % \begin{macrocode} |
| 1161 | \sv@demosmp% |
| 1162 | \listingindent\z@% |
| 1163 | \verbinput\sv@demoname% |
| 1164 | \sv@demoemp% |
| 1165 | \par% |
| 1166 | \nointerlineskip% |
| 1167 | \hb@xt@\hsize{\vrule\@height5\p@\hfil\vrule\@height5\p@}% |
| 1168 | \hrule% |
| 1169 | }% |
| 1170 | \endgroup% |
| 1171 | \par% |
| 1172 | \vskip\baselineskip% |
| 1173 | #1% |
| 1174 | } |
| 1175 | % \end{macrocode} |
| 1176 | % |
| 1177 | % \end{macro} |
| 1178 | % |
| 1179 | % That's all there is. Have fun. |
| 1180 | % |
| 1181 | % \begin{macrocode} |
| 1182 | %</package> |
| 1183 | % \end{macrocode} |
| 1184 | % |
| 1185 | % \hfill Mark Wooding, \today |
| 1186 | % |
| 1187 | % \Finale |
| 1188 | % |
| 1189 | \endinput |