Commit | Line | Data |
---|---|---|
6055f9d4 | 1 | # Pod::Text -- Convert POD data to formatted ASCII text. |
2e20e14f | 2 | # $Id: Text.pm,v 2.3 1999/10/07 09:41:57 eagle Exp $ |
6055f9d4 GS |
3 | # |
4 | # Copyright 1999 by Russ Allbery <rra@stanford.edu> | |
5 | # | |
6 | # This program is free software; you can redistribute it and/or modify it | |
7 | # under the same terms as Perl itself. | |
8 | # | |
27f805f4 GS |
9 | # This module is intended to be a replacement for Pod::Text, and attempts to |
10 | # match its output except for some specific circumstances where other | |
11 | # decisions seemed to produce better output. It uses Pod::Parser and is | |
12 | # designed to be very easy to subclass. | |
6055f9d4 GS |
13 | |
14 | ############################################################################ | |
15 | # Modules and declarations | |
16 | ############################################################################ | |
69e00e79 | 17 | |
6055f9d4 | 18 | package Pod::Text; |
69e00e79 | 19 | |
6055f9d4 GS |
20 | require 5.004; |
21 | ||
27f805f4 | 22 | use Carp qw(carp croak); |
2e20e14f | 23 | use Exporter (); |
27f805f4 | 24 | use Pod::Select (); |
6055f9d4 GS |
25 | |
26 | use strict; | |
2e20e14f | 27 | use vars qw(@ISA @EXPORT %ESCAPES $VERSION); |
6055f9d4 | 28 | |
27f805f4 GS |
29 | # We inherit from Pod::Select instead of Pod::Parser so that we can be used |
30 | # by Pod::Usage. | |
2e20e14f | 31 | @ISA = qw(Pod::Select Exporter); |
6055f9d4 | 32 | |
2e20e14f GS |
33 | # We have to export pod2text for backward compatibility. |
34 | @EXPORT = qw(pod2text); | |
35 | ||
36 | ($VERSION = (split (' ', q$Revision: 2.3 $ ))[1]) =~ s/\.(\d)$/.0$1/; | |
6055f9d4 GS |
37 | |
38 | ||
39 | ############################################################################ | |
40 | # Table of supported E<> escapes | |
41 | ############################################################################ | |
42 | ||
9741dab0 | 43 | # This table is taken near verbatim from Pod::PlainText in Pod::Parser, |
27f805f4 GS |
44 | # which got it near verbatim from the original Pod::Text. It is therefore |
45 | # credited to Tom Christiansen, and I'm glad I didn't have to write it. :) | |
6055f9d4 GS |
46 | %ESCAPES = ( |
47 | 'amp' => '&', # ampersand | |
48 | 'lt' => '<', # left chevron, less-than | |
49 | 'gt' => '>', # right chevron, greater-than | |
50 | 'quot' => '"', # double quote | |
51 | ||
52 | "Aacute" => "\xC1", # capital A, acute accent | |
53 | "aacute" => "\xE1", # small a, acute accent | |
54 | "Acirc" => "\xC2", # capital A, circumflex accent | |
55 | "acirc" => "\xE2", # small a, circumflex accent | |
56 | "AElig" => "\xC6", # capital AE diphthong (ligature) | |
57 | "aelig" => "\xE6", # small ae diphthong (ligature) | |
58 | "Agrave" => "\xC0", # capital A, grave accent | |
59 | "agrave" => "\xE0", # small a, grave accent | |
60 | "Aring" => "\xC5", # capital A, ring | |
61 | "aring" => "\xE5", # small a, ring | |
62 | "Atilde" => "\xC3", # capital A, tilde | |
63 | "atilde" => "\xE3", # small a, tilde | |
64 | "Auml" => "\xC4", # capital A, dieresis or umlaut mark | |
65 | "auml" => "\xE4", # small a, dieresis or umlaut mark | |
66 | "Ccedil" => "\xC7", # capital C, cedilla | |
67 | "ccedil" => "\xE7", # small c, cedilla | |
68 | "Eacute" => "\xC9", # capital E, acute accent | |
69 | "eacute" => "\xE9", # small e, acute accent | |
70 | "Ecirc" => "\xCA", # capital E, circumflex accent | |
71 | "ecirc" => "\xEA", # small e, circumflex accent | |
72 | "Egrave" => "\xC8", # capital E, grave accent | |
73 | "egrave" => "\xE8", # small e, grave accent | |
74 | "ETH" => "\xD0", # capital Eth, Icelandic | |
75 | "eth" => "\xF0", # small eth, Icelandic | |
76 | "Euml" => "\xCB", # capital E, dieresis or umlaut mark | |
77 | "euml" => "\xEB", # small e, dieresis or umlaut mark | |
78 | "Iacute" => "\xCD", # capital I, acute accent | |
79 | "iacute" => "\xED", # small i, acute accent | |
80 | "Icirc" => "\xCE", # capital I, circumflex accent | |
81 | "icirc" => "\xEE", # small i, circumflex accent | |
82 | "Igrave" => "\xCD", # capital I, grave accent | |
83 | "igrave" => "\xED", # small i, grave accent | |
84 | "Iuml" => "\xCF", # capital I, dieresis or umlaut mark | |
85 | "iuml" => "\xEF", # small i, dieresis or umlaut mark | |
86 | "Ntilde" => "\xD1", # capital N, tilde | |
87 | "ntilde" => "\xF1", # small n, tilde | |
88 | "Oacute" => "\xD3", # capital O, acute accent | |
89 | "oacute" => "\xF3", # small o, acute accent | |
90 | "Ocirc" => "\xD4", # capital O, circumflex accent | |
91 | "ocirc" => "\xF4", # small o, circumflex accent | |
92 | "Ograve" => "\xD2", # capital O, grave accent | |
93 | "ograve" => "\xF2", # small o, grave accent | |
94 | "Oslash" => "\xD8", # capital O, slash | |
95 | "oslash" => "\xF8", # small o, slash | |
96 | "Otilde" => "\xD5", # capital O, tilde | |
97 | "otilde" => "\xF5", # small o, tilde | |
98 | "Ouml" => "\xD6", # capital O, dieresis or umlaut mark | |
99 | "ouml" => "\xF6", # small o, dieresis or umlaut mark | |
100 | "szlig" => "\xDF", # small sharp s, German (sz ligature) | |
101 | "THORN" => "\xDE", # capital THORN, Icelandic | |
102 | "thorn" => "\xFE", # small thorn, Icelandic | |
103 | "Uacute" => "\xDA", # capital U, acute accent | |
104 | "uacute" => "\xFA", # small u, acute accent | |
105 | "Ucirc" => "\xDB", # capital U, circumflex accent | |
106 | "ucirc" => "\xFB", # small u, circumflex accent | |
107 | "Ugrave" => "\xD9", # capital U, grave accent | |
108 | "ugrave" => "\xF9", # small u, grave accent | |
109 | "Uuml" => "\xDC", # capital U, dieresis or umlaut mark | |
110 | "uuml" => "\xFC", # small u, dieresis or umlaut mark | |
111 | "Yacute" => "\xDD", # capital Y, acute accent | |
112 | "yacute" => "\xFD", # small y, acute accent | |
113 | "yuml" => "\xFF", # small y, dieresis or umlaut mark | |
114 | ||
115 | "lchevron" => "\xAB", # left chevron (double less than) | |
116 | "rchevron" => "\xBB", # right chevron (double greater than) | |
117 | ); | |
69e00e79 | 118 | |
69e00e79 | 119 | |
6055f9d4 GS |
120 | ############################################################################ |
121 | # Initialization | |
122 | ############################################################################ | |
69e00e79 | 123 | |
6055f9d4 GS |
124 | # Initialize the object. Must be sure to call our parent initializer. |
125 | sub initialize { | |
126 | my $self = shift; | |
69e00e79 | 127 | |
6055f9d4 GS |
128 | $$self{alt} = 0 unless defined $$self{alt}; |
129 | $$self{indent} = 4 unless defined $$self{indent}; | |
130 | $$self{loose} = 0 unless defined $$self{loose}; | |
131 | $$self{sentence} = 0 unless defined $$self{sentence}; | |
132 | $$self{width} = 76 unless defined $$self{width}; | |
69e00e79 | 133 | |
6055f9d4 GS |
134 | $$self{INDENTS} = []; # Stack of indentations. |
135 | $$self{MARGIN} = $$self{indent}; # Current left margin in spaces. | |
69e00e79 | 136 | |
6055f9d4 GS |
137 | $self->SUPER::initialize; |
138 | } | |
69e00e79 | 139 | |
69e00e79 | 140 | |
6055f9d4 GS |
141 | ############################################################################ |
142 | # Core overrides | |
143 | ############################################################################ | |
144 | ||
145 | # Called for each command paragraph. Gets the command, the associated | |
146 | # paragraph, the line number, and a Pod::Paragraph object. Just dispatches | |
147 | # the command to a method named the same as the command. =cut is handled | |
148 | # internally by Pod::Parser. | |
149 | sub command { | |
150 | my $self = shift; | |
151 | my $command = shift; | |
152 | return if $command eq 'pod'; | |
153 | return if ($$self{EXCLUDE} && $command ne 'end'); | |
154 | $self->item ("\n") if defined $$self{ITEM}; | |
155 | $command = 'cmd_' . $command; | |
156 | $self->$command (@_); | |
157 | } | |
69e00e79 | 158 | |
6055f9d4 GS |
159 | # Called for a verbatim paragraph. Gets the paragraph, the line number, and |
160 | # a Pod::Paragraph object. Just output it verbatim, but with tabs converted | |
161 | # to spaces. | |
162 | sub verbatim { | |
163 | my $self = shift; | |
164 | return if $$self{EXCLUDE}; | |
165 | $self->item if defined $$self{ITEM}; | |
166 | local $_ = shift; | |
167 | return if /^\s*$/; | |
168 | s/^(\s*\S+)/(' ' x $$self{MARGIN}) . $1/gme; | |
169 | $self->output ($_); | |
170 | } | |
69e00e79 | 171 | |
6055f9d4 GS |
172 | # Called for a regular text block. Gets the paragraph, the line number, and |
173 | # a Pod::Paragraph object. Perform interpolation and output the results. | |
174 | sub textblock { | |
27f805f4 | 175 | my $self = shift; |
6055f9d4 | 176 | return if $$self{EXCLUDE}; |
27f805f4 GS |
177 | $self->output ($_[0]), return if $$self{VERBATIM}; |
178 | local $_ = shift; | |
179 | my $line = shift; | |
6055f9d4 GS |
180 | |
181 | # Perform a little magic to collapse multiple L<> references. This is | |
27f805f4 GS |
182 | # here mostly for backwards-compatibility. We'll just rewrite the whole |
183 | # thing into actual text at this part, bypassing the whole internal | |
184 | # sequence parsing thing. | |
6055f9d4 GS |
185 | s{ |
186 | ( | |
187 | L< # A link of the form L</something>. | |
188 | / | |
189 | ( | |
190 | [:\w]+ # The item has to be a simple word... | |
191 | (\(\))? # ...or simple function. | |
192 | ) | |
193 | > | |
194 | ( | |
195 | ,?\s+(and\s+)? # Allow lots of them, conjuncted. | |
196 | L< | |
197 | / | |
198 | ( | |
199 | [:\w]+ | |
200 | (\(\))? | |
201 | ) | |
202 | > | |
203 | )+ | |
204 | ) | |
205 | } { | |
206 | local $_ = $1; | |
207 | s%L</([^>]+)>%$1%g; | |
208 | my @items = split /(?:,?\s+(?:and\s+)?)/; | |
209 | my $string = "the "; | |
210 | my $i; | |
211 | for ($i = 0; $i < @items; $i++) { | |
212 | $string .= $items[$i]; | |
213 | $string .= ", " if @items > 2 && $i != $#items; | |
214 | $string .= " and " if ($i == $#items - 1); | |
215 | } | |
216 | $string .= " entries elsewhere in this document"; | |
217 | $string; | |
218 | }gex; | |
219 | ||
220 | # Now actually interpolate and output the paragraph. | |
221 | $_ = $self->interpolate ($_, $line); | |
222 | s/\s+$/\n/; | |
223 | if (defined $$self{ITEM}) { | |
224 | $self->item ($_ . "\n"); | |
225 | } else { | |
226 | $self->output ($self->reformat ($_ . "\n")); | |
227 | } | |
228 | } | |
69e00e79 | 229 | |
6055f9d4 GS |
230 | # Called for an interior sequence. Gets the command, argument, and a |
231 | # Pod::InteriorSequence object and is expected to return the resulting text. | |
232 | # Calls code, bold, italic, file, and link to handle those types of | |
233 | # sequences, and handles S<>, E<>, X<>, and Z<> directly. | |
234 | sub interior_sequence { | |
235 | my $self = shift; | |
236 | my $command = shift; | |
237 | local $_ = shift; | |
238 | return '' if ($command eq 'X' || $command eq 'Z'); | |
69e00e79 | 239 | |
6055f9d4 GS |
240 | # Expand escapes into the actual character now, carping if invalid. |
241 | if ($command eq 'E') { | |
2e20e14f GS |
242 | if (/^\d+$/) { |
243 | return chr; | |
244 | } else { | |
245 | return $ESCAPES{$_} if defined $ESCAPES{$_}; | |
246 | carp "Unknown escape: E<$_>"; | |
247 | return "E<$_>"; | |
248 | } | |
6055f9d4 | 249 | } |
69e00e79 | 250 | |
6055f9d4 | 251 | # For all the other sequences, empty content produces no output. |
27f805f4 | 252 | return if $_ eq ''; |
69e00e79 | 253 | |
6055f9d4 GS |
254 | # For S<>, compress all internal whitespace and then map spaces to \01. |
255 | # When we output the text, we'll map this back. | |
256 | if ($command eq 'S') { | |
257 | s/\s{2,}/ /g; | |
258 | tr/ /\01/; | |
259 | return $_; | |
260 | } | |
69e00e79 | 261 | |
6055f9d4 GS |
262 | # Anything else needs to get dispatched to another method. |
263 | if ($command eq 'B') { return $self->seq_b ($_) } | |
264 | elsif ($command eq 'C') { return $self->seq_c ($_) } | |
265 | elsif ($command eq 'F') { return $self->seq_f ($_) } | |
266 | elsif ($command eq 'I') { return $self->seq_i ($_) } | |
267 | elsif ($command eq 'L') { return $self->seq_l ($_) } | |
268 | else { carp "Unknown sequence $command<$_>" } | |
269 | } | |
f02a87df | 270 | |
6055f9d4 GS |
271 | # Called for each paragraph that's actually part of the POD. We take |
272 | # advantage of this opportunity to untabify the input. | |
273 | sub preprocess_paragraph { | |
274 | my $self = shift; | |
275 | local $_ = shift; | |
276 | 1 while s/^(.*?)(\t+)/$1 . ' ' x (length ($2) * 8 - length ($1) % 8)/me; | |
277 | $_; | |
278 | } | |
3ec07288 | 279 | |
69e00e79 | 280 | |
6055f9d4 GS |
281 | ############################################################################ |
282 | # Command paragraphs | |
283 | ############################################################################ | |
f2506fb2 | 284 | |
6055f9d4 | 285 | # All command paragraphs take the paragraph and the line number. |
69e00e79 | 286 | |
6055f9d4 GS |
287 | # First level heading. |
288 | sub cmd_head1 { | |
289 | my $self = shift; | |
290 | local $_ = shift; | |
291 | s/\s+$//; | |
27f805f4 | 292 | $_ = $self->interpolate ($_, shift); |
6055f9d4 GS |
293 | if ($$self{alt}) { |
294 | $self->output ("\n==== $_ ====\n\n"); | |
295 | } else { | |
296 | $_ .= "\n" if $$self{loose}; | |
297 | $self->output ($_ . "\n"); | |
298 | } | |
299 | } | |
69e00e79 | 300 | |
6055f9d4 GS |
301 | # Second level heading. |
302 | sub cmd_head2 { | |
303 | my $self = shift; | |
304 | local $_ = shift; | |
305 | s/\s+$//; | |
27f805f4 | 306 | $_ = $self->interpolate ($_, shift); |
6055f9d4 GS |
307 | if ($$self{alt}) { |
308 | $self->output ("\n== $_ ==\n\n"); | |
309 | } else { | |
310 | $self->output (' ' x ($$self{indent} / 2) . $_ . "\n\n"); | |
311 | } | |
312 | } | |
69e00e79 | 313 | |
6055f9d4 GS |
314 | # Start a list. |
315 | sub cmd_over { | |
316 | my $self = shift; | |
317 | local $_ = shift; | |
318 | unless (/^[-+]?\d+\s+$/) { $_ = $$self{indent} } | |
319 | push (@{ $$self{INDENTS} }, $$self{MARGIN}); | |
320 | $$self{MARGIN} += ($_ + 0); | |
321 | } | |
69e00e79 | 322 | |
6055f9d4 GS |
323 | # End a list. |
324 | sub cmd_back { | |
325 | my $self = shift; | |
326 | $$self{MARGIN} = pop @{ $$self{INDENTS} }; | |
327 | unless (defined $$self{MARGIN}) { | |
328 | carp "Unmatched =back"; | |
329 | $$self{MARGIN} = $$self{indent}; | |
330 | } | |
69e00e79 | 331 | } |
332 | ||
6055f9d4 GS |
333 | # An individual list item. |
334 | sub cmd_item { | |
335 | my $self = shift; | |
336 | if (defined $$self{ITEM}) { $self->item } | |
337 | local $_ = shift; | |
338 | s/\s+$//; | |
339 | $$self{ITEM} = $self->interpolate ($_); | |
340 | } | |
69e00e79 | 341 | |
27f805f4 GS |
342 | # Begin a block for a particular translator. Setting VERBATIM triggers |
343 | # special handling in textblock(). | |
6055f9d4 GS |
344 | sub cmd_begin { |
345 | my $self = shift; | |
346 | local $_ = shift; | |
347 | my ($kind) = /^(\S+)/ or return; | |
27f805f4 GS |
348 | if ($kind eq 'text') { |
349 | $$self{VERBATIM} = 1; | |
350 | } else { | |
351 | $$self{EXCLUDE} = 1; | |
352 | } | |
6055f9d4 | 353 | } |
f2506fb2 | 354 | |
6055f9d4 | 355 | # End a block for a particular translator. We assume that all =begin/=end |
27f805f4 | 356 | # pairs are properly closed. |
6055f9d4 GS |
357 | sub cmd_end { |
358 | my $self = shift; | |
27f805f4 GS |
359 | $$self{EXCLUDE} = 0; |
360 | $$self{VERBATIM} = 0; | |
6055f9d4 GS |
361 | } |
362 | ||
363 | # One paragraph for a particular translator. Ignore it unless it's intended | |
27f805f4 | 364 | # for text, in which case we treat it as a verbatim text block. |
6055f9d4 GS |
365 | sub cmd_for { |
366 | my $self = shift; | |
367 | local $_ = shift; | |
368 | my $line = shift; | |
27f805f4 GS |
369 | return unless s/^text\b[ \t]*\n?//; |
370 | $self->verbatim ($_, $line); | |
6055f9d4 | 371 | } |
f2506fb2 | 372 | |
69e00e79 | 373 | |
6055f9d4 GS |
374 | ############################################################################ |
375 | # Interior sequences | |
376 | ############################################################################ | |
69e00e79 | 377 | |
6055f9d4 GS |
378 | # The simple formatting ones. These are here mostly so that subclasses can |
379 | # override them and do more complicated things. | |
27f805f4 GS |
380 | sub seq_b { return $_[0]{alt} ? "``$_[1]''" : $_[1] } |
381 | sub seq_c { return $_[0]{alt} ? "``$_[1]''" : "`$_[1]'" } | |
382 | sub seq_f { return $_[0]{alt} ? "\"$_[1]\"" : $_[1] } | |
6055f9d4 | 383 | sub seq_i { return '*' . $_[1] . '*' } |
69e00e79 | 384 | |
6055f9d4 GS |
385 | # The complicated one. Handle links. Since this is plain text, we can't |
386 | # actually make any real links, so this is all to figure out what text we | |
387 | # print out. | |
388 | sub seq_l { | |
389 | my $self = shift; | |
390 | local $_ = shift; | |
69e00e79 | 391 | |
6055f9d4 GS |
392 | # Smash whitespace in case we were split across multiple lines. |
393 | s/\s+/ /g; | |
69e00e79 | 394 | |
6055f9d4 GS |
395 | # If we were given any explicit text, just output it. |
396 | if (/^([^|]+)\|/) { return $1 } | |
397 | ||
398 | # Okay, leading and trailing whitespace isn't important; get rid of it. | |
399 | s/^\s+//; | |
400 | s/\s+$//; | |
6055f9d4 GS |
401 | |
402 | # Default to using the whole content of the link entry as a section | |
403 | # name. Note that L<manpage/> forces a manpage interpretation, as does | |
404 | # something looking like L<manpage(section)>. The latter is an | |
405 | # enhancement over the original Pod::Text. | |
406 | my ($manpage, $section) = ('', $_); | |
407 | if (/^"\s*(.*?)\s*"$/) { | |
408 | $section = '"' . $1 . '"'; | |
409 | } elsif (m/^[-:.\w]+(?:\(\S+\))?$/) { | |
410 | ($manpage, $section) = ($_, ''); | |
411 | } elsif (m%/%) { | |
412 | ($manpage, $section) = split (/\s*\/\s*/, $_, 2); | |
8c634b6e KA |
413 | } |
414 | ||
6055f9d4 GS |
415 | # Now build the actual output text. |
416 | my $text = ''; | |
417 | if (!length $section) { | |
418 | $text = "the $manpage manpage" if length $manpage; | |
419 | } elsif ($section =~ /^[:\w]+(?:\(\))?/) { | |
420 | $text .= 'the ' . $section . ' entry'; | |
421 | $text .= (length $manpage) ? " in the $manpage manpage" | |
422 | : " elsewhere in this document"; | |
423 | } else { | |
424 | $section =~ s/^\"\s*//; | |
425 | $section =~ s/\s*\"$//; | |
426 | $text .= 'the section on "' . $section . '"'; | |
427 | $text .= " in the $manpage manpage" if length $manpage; | |
69e00e79 | 428 | } |
6055f9d4 | 429 | $text; |
69e00e79 | 430 | } |
431 | ||
6055f9d4 GS |
432 | |
433 | ############################################################################ | |
434 | # List handling | |
435 | ############################################################################ | |
436 | ||
437 | # This method is called whenever an =item command is complete (in other | |
438 | # words, we've seen its associated paragraph or know for certain that it | |
439 | # doesn't have one). It gets the paragraph associated with the item as an | |
440 | # argument. If that argument is empty, just output the item tag; if it | |
441 | # contains a newline, output the item tag followed by the newline. | |
442 | # Otherwise, see if there's enough room for us to output the item tag in the | |
443 | # margin of the text or if we have to put it on a separate line. | |
444 | sub item { | |
445 | my $self = shift; | |
446 | local $_ = shift; | |
447 | my $tag = $$self{ITEM}; | |
448 | unless (defined $tag) { | |
449 | carp "item called without tag"; | |
450 | return; | |
69e00e79 | 451 | } |
6055f9d4 GS |
452 | undef $$self{ITEM}; |
453 | my $indent = $$self{INDENTS}[-1]; | |
454 | unless (defined $indent) { $indent = $$self{indent} } | |
455 | my $space = ' ' x $indent; | |
456 | $space =~ s/^ /:/ if $$self{alt}; | |
457 | if (!$_ || /^\s+$/ || ($$self{MARGIN} - $indent < length ($tag) + 1)) { | |
27f805f4 GS |
458 | my $margin = $$self{MARGIN}; |
459 | $$self{MARGIN} = $indent; | |
460 | my $output = $self->reformat ($tag); | |
461 | $output =~ s/\n*$/\n/; | |
462 | $self->output ($output); | |
463 | $$self{MARGIN} = $margin; | |
6055f9d4 GS |
464 | $self->output ($self->reformat ($_)) if /\S/; |
465 | } else { | |
466 | $_ = $self->reformat ($_); | |
467 | s/^ /:/ if ($$self{alt} && $indent > 0); | |
468 | my $tagspace = ' ' x length $tag; | |
469 | s/^($space)$tagspace/$1$tag/ or warn "Bizarre space in item"; | |
470 | $self->output ($_); | |
69e00e79 | 471 | } |
472 | } | |
473 | ||
69e00e79 | 474 | |
6055f9d4 GS |
475 | ############################################################################ |
476 | # Output formatting | |
477 | ############################################################################ | |
69e00e79 | 478 | |
6055f9d4 GS |
479 | # Wrap a line, indenting by the current left margin. We can't use |
480 | # Text::Wrap because it plays games with tabs. We can't use formline, even | |
481 | # though we'd really like to, because it screws up non-printing characters. | |
482 | # So we have to do the wrapping ourselves. | |
483 | sub wrap { | |
484 | my $self = shift; | |
485 | local $_ = shift; | |
486 | my $output = ''; | |
487 | my $spaces = ' ' x $$self{MARGIN}; | |
488 | my $width = $$self{width} - $$self{MARGIN}; | |
489 | while (length > $width) { | |
490 | if (s/^([^\n]{0,$width})\s+// || s/^([^\n]{$width})//) { | |
491 | $output .= $spaces . $1 . "\n"; | |
492 | } else { | |
493 | last; | |
494 | } | |
69e00e79 | 495 | } |
6055f9d4 GS |
496 | $output .= $spaces . $_; |
497 | $output =~ s/\s+$/\n\n/; | |
498 | $output; | |
69e00e79 | 499 | } |
500 | ||
6055f9d4 GS |
501 | # Reformat a paragraph of text for the current margin. Takes the text to |
502 | # reformat and returns the formatted text. | |
503 | sub reformat { | |
504 | my $self = shift; | |
505 | local $_ = shift; | |
69e00e79 | 506 | |
6055f9d4 GS |
507 | # If we're trying to preserve two spaces after sentences, do some |
508 | # munging to support that. Otherwise, smash all repeated whitespace. | |
509 | if ($$self{sentence}) { | |
510 | s/ +$//mg; | |
511 | s/\.\n/. \n/g; | |
512 | s/\n/ /g; | |
513 | s/ +/ /g; | |
69e00e79 | 514 | } else { |
6055f9d4 | 515 | s/\s+/ /g; |
69e00e79 | 516 | } |
6055f9d4 | 517 | $self->wrap ($_); |
69e00e79 | 518 | } |
519 | ||
6055f9d4 GS |
520 | # Output text to the output device. |
521 | sub output { $_[1] =~ tr/\01/ /; print { $_[0]->output_handle } $_[1] } | |
69e00e79 | 522 | |
69e00e79 | 523 | |
6055f9d4 | 524 | ############################################################################ |
27f805f4 GS |
525 | # Backwards compatibility |
526 | ############################################################################ | |
527 | ||
528 | # The old Pod::Text module did everything in a pod2text() function. This | |
529 | # tries to provide the same interface for legacy applications. | |
530 | sub pod2text { | |
531 | my @args; | |
532 | ||
533 | # This is really ugly; I hate doing option parsing in the middle of a | |
534 | # module. But the old Pod::Text module supported passing flags to its | |
535 | # entry function, so handle -a and -<number>. | |
536 | while ($_[0] =~ /^-/) { | |
537 | my $flag = shift; | |
538 | if ($flag eq '-a') { push (@args, alt => 1) } | |
539 | elsif ($flag =~ /^-(\d+)$/) { push (@args, width => $1) } | |
540 | else { | |
541 | unshift (@_, $flag); | |
542 | last; | |
543 | } | |
544 | } | |
545 | ||
546 | # Now that we know what arguments we're using, create the parser. | |
547 | my $parser = Pod::Text->new (@args); | |
548 | ||
549 | # If two arguments were given, the second argument is going to be a file | |
550 | # handle. That means we want to call parse_from_filehandle(), which | |
551 | # means we need to turn the first argument into a file handle. Magic | |
552 | # open will handle the <&STDIN case automagically. | |
553 | if (defined $_[1]) { | |
554 | local *IN; | |
555 | unless (open (IN, $_[0])) { | |
556 | croak ("Can't open $_[0] for reading: $!\n"); | |
557 | return; | |
558 | } | |
559 | $_[0] = \*IN; | |
560 | return $parser->parse_from_filehandle (@_); | |
561 | } else { | |
562 | return $parser->parse_from_file (@_); | |
563 | } | |
564 | } | |
565 | ||
566 | ||
567 | ############################################################################ | |
6055f9d4 GS |
568 | # Module return value and documentation |
569 | ############################################################################ | |
69e00e79 | 570 | |
6055f9d4 GS |
571 | 1; |
572 | __END__ | |
69e00e79 | 573 | |
6055f9d4 | 574 | =head1 NAME |
69e00e79 | 575 | |
6055f9d4 | 576 | Pod::Text - Convert POD data to formatted ASCII text |
69e00e79 | 577 | |
6055f9d4 | 578 | =head1 SYNOPSIS |
69e00e79 | 579 | |
6055f9d4 GS |
580 | use Pod::Text; |
581 | my $parser = Pod::Text->new (sentence => 0, width => 78); | |
69e00e79 | 582 | |
6055f9d4 GS |
583 | # Read POD from STDIN and write to STDOUT. |
584 | $parser->parse_from_filehandle; | |
69e00e79 | 585 | |
6055f9d4 GS |
586 | # Read POD from file.pod and write to file.txt. |
587 | $parser->parse_from_file ('file.pod', 'file.txt'); | |
69e00e79 | 588 | |
6055f9d4 | 589 | =head1 DESCRIPTION |
5491a304 | 590 | |
27f805f4 GS |
591 | Pod::Text is a module that can convert documentation in the POD format (the |
592 | preferred language for documenting Perl) into formatted ASCII. It uses no | |
593 | special formatting controls or codes whatsoever, and its output is therefore | |
594 | suitable for nearly any device. | |
69e00e79 | 595 | |
27f805f4 GS |
596 | As a derived class from Pod::Parser, Pod::Text supports the same methods and |
597 | interfaces. See L<Pod::Parser> for all the details; briefly, one creates a | |
598 | new parser with C<Pod::Text-E<gt>new()> and then calls either | |
599 | parse_from_filehandle() or parse_from_file(). | |
6055f9d4 | 600 | |
27f805f4 | 601 | new() can take options, in the form of key/value pairs, that control the |
6055f9d4 GS |
602 | behavior of the parser. The currently recognized options are: |
603 | ||
604 | =over 4 | |
605 | ||
606 | =item alt | |
607 | ||
608 | If set to a true value, selects an alternate output format that, among other | |
609 | things, uses a different heading style and marks C<=item> entries with a | |
610 | colon in the left margin. Defaults to false. | |
611 | ||
612 | =item indent | |
613 | ||
614 | The number of spaces to indent regular text, and the default indentation for | |
615 | C<=over> blocks. Defaults to 4. | |
616 | ||
617 | =item loose | |
618 | ||
619 | If set to a true value, a blank line is printed after a C<=head1> heading. | |
620 | If set to false (the default), no blank line is printed after C<=head1>, | |
621 | although one is still printed after C<=head2>. This is the default because | |
622 | it's the expected formatting for manual pages; if you're formatting | |
623 | arbitrary text documents, setting this to true may result in more pleasing | |
624 | output. | |
625 | ||
626 | =item sentence | |
627 | ||
27f805f4 GS |
628 | If set to a true value, Pod::Text will assume that each sentence ends in two |
629 | spaces, and will try to preserve that spacing. If set to false, all | |
6055f9d4 GS |
630 | consecutive whitespace in non-verbatim paragraphs is compressed into a |
631 | single space. Defaults to true. | |
632 | ||
633 | =item width | |
634 | ||
635 | The column at which to wrap text on the right-hand side. Defaults to 76. | |
636 | ||
637 | =back | |
638 | ||
27f805f4 | 639 | The standard Pod::Parser method parse_from_filehandle() takes up to two |
6055f9d4 GS |
640 | arguments, the first being the file handle to read POD from and the second |
641 | being the file handle to write the formatted output to. The first defaults | |
642 | to STDIN if not given, and the second defaults to STDOUT. The method | |
27f805f4 GS |
643 | parse_from_file() is almost identical, except that its two arguments are the |
644 | input and output disk files instead. See L<Pod::Parser> for the specific | |
645 | details. | |
6055f9d4 GS |
646 | |
647 | =head1 DIAGNOSTICS | |
648 | ||
649 | =over 4 | |
650 | ||
27f805f4 GS |
651 | =item Bizarre space in item |
652 | ||
653 | (W) Something has gone wrong in internal C<=item> processing. This message | |
654 | indicates a bug in Pod::Text; you should never see it. | |
655 | ||
656 | =item Can't open %s for reading: %s | |
657 | ||
658 | (F) Pod::Text was invoked via the compatibility mode pod2text() interface | |
659 | and the input file it was given could not be opened. | |
660 | ||
6055f9d4 GS |
661 | =item Unknown escape: %s |
662 | ||
27f805f4 GS |
663 | (W) The POD source contained an C<EE<lt>E<gt>> escape that Pod::Text didn't |
664 | know about. | |
6055f9d4 GS |
665 | |
666 | =item Unknown sequence: %s | |
667 | ||
27f805f4 GS |
668 | (W) The POD source contained a non-standard internal sequence (something of |
669 | the form C<XE<lt>E<gt>>) that Pod::Text didn't know about. | |
6055f9d4 GS |
670 | |
671 | =item Unmatched =back | |
672 | ||
27f805f4 | 673 | (W) Pod::Text encountered a C<=back> command that didn't correspond to an |
6055f9d4 GS |
674 | C<=over> command. |
675 | ||
676 | =back | |
677 | ||
27f805f4 GS |
678 | =head1 RESTRICTIONS |
679 | ||
680 | Embedded Ctrl-As (octal 001) in the input will be mapped to spaces on | |
681 | output, due to an internal implementation detail. | |
682 | ||
6055f9d4 GS |
683 | =head1 NOTES |
684 | ||
27f805f4 GS |
685 | This is a replacement for an earlier Pod::Text module written by Tom |
686 | Christiansen. It has a revamped interface, since it now uses Pod::Parser, | |
687 | but an interface roughly compatible with the old Pod::Text::pod2text() | |
688 | function is still available. Please change to the new calling convention, | |
689 | though. | |
6055f9d4 GS |
690 | |
691 | The original Pod::Text contained code to do formatting via termcap | |
692 | sequences, although it wasn't turned on by default and it was problematic to | |
27f805f4 GS |
693 | get it to work at all. This rewrite doesn't even try to do that, but a |
694 | subclass of it does. Look for L<Pod::Text::Termcap|Pod::Text::Termcap>. | |
6055f9d4 GS |
695 | |
696 | =head1 SEE ALSO | |
697 | ||
27f805f4 GS |
698 | L<Pod::Parser|Pod::Parser>, L<Pod::Text::Termcap|Pod::Text::Termcap>, |
699 | pod2text(1) | |
6055f9d4 GS |
700 | |
701 | =head1 AUTHOR | |
702 | ||
703 | Russ Allbery E<lt>rra@stanford.eduE<gt>, based I<very> heavily on the | |
704 | original Pod::Text by Tom Christiansen E<lt>tchrist@mox.perl.comE<gt> and | |
705 | its conversion to Pod::Parser by Brad Appleton | |
706 | E<lt>bradapp@enteract.comE<gt>. | |
707 | ||
708 | =cut |