Commit | Line | Data |
6055f9d4 |
1 | # Pod::Text -- Convert POD data to formatted ASCII text. |
bf202ccd |
2 | # $Id: Text.pm,v 2.14 2001/11/15 08:03:18 eagle Exp $ |
6055f9d4 |
3 | # |
16ba52cf |
4 | # Copyright 1999, 2000, 2001 by Russ Allbery <rra@stanford.edu> |
6055f9d4 |
5 | # |
3c014959 |
6 | # This program is free software; you may redistribute it and/or modify it |
6055f9d4 |
7 | # under the same terms as Perl itself. |
8 | # |
3c014959 |
9 | # This module replaces the old Pod::Text that came with versions of Perl prior |
10 | # to 5.6.0, and attempts to match its output except for some specific |
11 | # circumstances where other decisions seemed to produce better output. It |
12 | # uses Pod::Parser and is designed to be very easy to subclass. |
13 | # |
14 | # Perl core hackers, please note that this module is also separately |
15 | # maintained outside of the Perl core as part of the podlators. Please send |
16 | # me any patches at the address above in addition to sending them to the |
17 | # standard Perl mailing lists. |
6055f9d4 |
18 | |
3c014959 |
19 | ############################################################################## |
6055f9d4 |
20 | # Modules and declarations |
3c014959 |
21 | ############################################################################## |
69e00e79 |
22 | |
6055f9d4 |
23 | package Pod::Text; |
69e00e79 |
24 | |
6055f9d4 |
25 | require 5.004; |
26 | |
27f805f4 |
27 | use Carp qw(carp croak); |
2e20e14f |
28 | use Exporter (); |
bf202ccd |
29 | use Pod::ParseLink qw(parselink); |
27f805f4 |
30 | use Pod::Select (); |
6055f9d4 |
31 | |
32 | use strict; |
2e20e14f |
33 | use vars qw(@ISA @EXPORT %ESCAPES $VERSION); |
6055f9d4 |
34 | |
3c014959 |
35 | # We inherit from Pod::Select instead of Pod::Parser so that we can be used by |
36 | # Pod::Usage. |
2e20e14f |
37 | @ISA = qw(Pod::Select Exporter); |
6055f9d4 |
38 | |
2e20e14f |
39 | # We have to export pod2text for backward compatibility. |
40 | @EXPORT = qw(pod2text); |
41 | |
3c014959 |
42 | # Don't use the CVS revision as the version, since this module is also in Perl |
43 | # core and too many things could munge CVS magic revision strings. This |
44 | # number should ideally be the same as the CVS revision in podlators, however. |
bf202ccd |
45 | $VERSION = 2.14; |
6055f9d4 |
46 | |
47 | |
3c014959 |
48 | ############################################################################## |
6055f9d4 |
49 | # Table of supported E<> escapes |
3c014959 |
50 | ############################################################################## |
6055f9d4 |
51 | |
3c014959 |
52 | # This table is taken near verbatim from Pod::PlainText in Pod::Parser, which |
53 | # got it near verbatim from the original Pod::Text. It is therefore credited |
54 | # to Tom Christiansen, and I'm glad I didn't have to write it. :) "iexcl" to |
55 | # "divide" added by Tim Jenness. |
6055f9d4 |
56 | %ESCAPES = ( |
57 | 'amp' => '&', # ampersand |
bf202ccd |
58 | 'apos' => "'", # apostrophe |
6055f9d4 |
59 | 'lt' => '<', # left chevron, less-than |
60 | 'gt' => '>', # right chevron, greater-than |
61 | 'quot' => '"', # double quote |
ab1f1d91 |
62 | 'sol' => '/', # solidus (forward slash) |
be3174d2 |
63 | 'verbar' => '|', # vertical bar |
5cdeb5a2 |
64 | |
6055f9d4 |
65 | "Aacute" => "\xC1", # capital A, acute accent |
66 | "aacute" => "\xE1", # small a, acute accent |
67 | "Acirc" => "\xC2", # capital A, circumflex accent |
68 | "acirc" => "\xE2", # small a, circumflex accent |
69 | "AElig" => "\xC6", # capital AE diphthong (ligature) |
70 | "aelig" => "\xE6", # small ae diphthong (ligature) |
71 | "Agrave" => "\xC0", # capital A, grave accent |
72 | "agrave" => "\xE0", # small a, grave accent |
73 | "Aring" => "\xC5", # capital A, ring |
74 | "aring" => "\xE5", # small a, ring |
75 | "Atilde" => "\xC3", # capital A, tilde |
76 | "atilde" => "\xE3", # small a, tilde |
77 | "Auml" => "\xC4", # capital A, dieresis or umlaut mark |
78 | "auml" => "\xE4", # small a, dieresis or umlaut mark |
79 | "Ccedil" => "\xC7", # capital C, cedilla |
80 | "ccedil" => "\xE7", # small c, cedilla |
81 | "Eacute" => "\xC9", # capital E, acute accent |
82 | "eacute" => "\xE9", # small e, acute accent |
83 | "Ecirc" => "\xCA", # capital E, circumflex accent |
84 | "ecirc" => "\xEA", # small e, circumflex accent |
85 | "Egrave" => "\xC8", # capital E, grave accent |
86 | "egrave" => "\xE8", # small e, grave accent |
87 | "ETH" => "\xD0", # capital Eth, Icelandic |
88 | "eth" => "\xF0", # small eth, Icelandic |
89 | "Euml" => "\xCB", # capital E, dieresis or umlaut mark |
90 | "euml" => "\xEB", # small e, dieresis or umlaut mark |
ee89c1da |
91 | "Iacute" => "\xCD", # capital I, acute accent |
92 | "iacute" => "\xED", # small i, acute accent |
6055f9d4 |
93 | "Icirc" => "\xCE", # capital I, circumflex accent |
94 | "icirc" => "\xEE", # small i, circumflex accent |
ee89c1da |
95 | "Igrave" => "\xCC", # capital I, grave accent |
96 | "igrave" => "\xEC", # small i, grave accent |
6055f9d4 |
97 | "Iuml" => "\xCF", # capital I, dieresis or umlaut mark |
98 | "iuml" => "\xEF", # small i, dieresis or umlaut mark |
99 | "Ntilde" => "\xD1", # capital N, tilde |
100 | "ntilde" => "\xF1", # small n, tilde |
101 | "Oacute" => "\xD3", # capital O, acute accent |
102 | "oacute" => "\xF3", # small o, acute accent |
103 | "Ocirc" => "\xD4", # capital O, circumflex accent |
104 | "ocirc" => "\xF4", # small o, circumflex accent |
105 | "Ograve" => "\xD2", # capital O, grave accent |
106 | "ograve" => "\xF2", # small o, grave accent |
107 | "Oslash" => "\xD8", # capital O, slash |
108 | "oslash" => "\xF8", # small o, slash |
109 | "Otilde" => "\xD5", # capital O, tilde |
110 | "otilde" => "\xF5", # small o, tilde |
111 | "Ouml" => "\xD6", # capital O, dieresis or umlaut mark |
112 | "ouml" => "\xF6", # small o, dieresis or umlaut mark |
113 | "szlig" => "\xDF", # small sharp s, German (sz ligature) |
114 | "THORN" => "\xDE", # capital THORN, Icelandic |
115 | "thorn" => "\xFE", # small thorn, Icelandic |
116 | "Uacute" => "\xDA", # capital U, acute accent |
117 | "uacute" => "\xFA", # small u, acute accent |
118 | "Ucirc" => "\xDB", # capital U, circumflex accent |
119 | "ucirc" => "\xFB", # small u, circumflex accent |
120 | "Ugrave" => "\xD9", # capital U, grave accent |
121 | "ugrave" => "\xF9", # small u, grave accent |
122 | "Uuml" => "\xDC", # capital U, dieresis or umlaut mark |
123 | "uuml" => "\xFC", # small u, dieresis or umlaut mark |
124 | "Yacute" => "\xDD", # capital Y, acute accent |
125 | "yacute" => "\xFD", # small y, acute accent |
126 | "yuml" => "\xFF", # small y, dieresis or umlaut mark |
5cdeb5a2 |
127 | |
a3e04946 |
128 | "laquo" => "\xAB", # left pointing double angle quotation mark |
129 | "lchevron" => "\xAB", # synonym (backwards compatibility) |
130 | "raquo" => "\xBB", # right pointing double angle quotation mark |
131 | "rchevron" => "\xBB", # synonym (backwards compatibility) |
132 | |
133 | "iexcl" => "\xA1", # inverted exclamation mark |
134 | "cent" => "\xA2", # cent sign |
135 | "pound" => "\xA3", # (UK) pound sign |
136 | "curren" => "\xA4", # currency sign |
137 | "yen" => "\xA5", # yen sign |
138 | "brvbar" => "\xA6", # broken vertical bar |
139 | "sect" => "\xA7", # section sign |
140 | "uml" => "\xA8", # diaresis |
141 | "copy" => "\xA9", # Copyright symbol |
142 | "ordf" => "\xAA", # feminine ordinal indicator |
143 | "not" => "\xAC", # not sign |
bf202ccd |
144 | "shy" => '', # soft (discretionary) hyphen |
a3e04946 |
145 | "reg" => "\xAE", # registered trademark |
146 | "macr" => "\xAF", # macron, overline |
147 | "deg" => "\xB0", # degree sign |
148 | "plusmn" => "\xB1", # plus-minus sign |
149 | "sup2" => "\xB2", # superscript 2 |
150 | "sup3" => "\xB3", # superscript 3 |
151 | "acute" => "\xB4", # acute accent |
152 | "micro" => "\xB5", # micro sign |
153 | "para" => "\xB6", # pilcrow sign = paragraph sign |
154 | "middot" => "\xB7", # middle dot = Georgian comma |
155 | "cedil" => "\xB8", # cedilla |
156 | "sup1" => "\xB9", # superscript 1 |
157 | "ordm" => "\xBA", # masculine ordinal indicator |
158 | "frac14" => "\xBC", # vulgar fraction one quarter |
159 | "frac12" => "\xBD", # vulgar fraction one half |
160 | "frac34" => "\xBE", # vulgar fraction three quarters |
161 | "iquest" => "\xBF", # inverted question mark |
162 | "times" => "\xD7", # multiplication sign |
163 | "divide" => "\xF7", # division sign |
bf202ccd |
164 | |
165 | "nbsp" => "\x01", # non-breaking space |
6055f9d4 |
166 | ); |
69e00e79 |
167 | |
69e00e79 |
168 | |
3c014959 |
169 | ############################################################################## |
6055f9d4 |
170 | # Initialization |
3c014959 |
171 | ############################################################################## |
69e00e79 |
172 | |
6055f9d4 |
173 | # Initialize the object. Must be sure to call our parent initializer. |
174 | sub initialize { |
175 | my $self = shift; |
69e00e79 |
176 | |
6055f9d4 |
177 | $$self{alt} = 0 unless defined $$self{alt}; |
178 | $$self{indent} = 4 unless defined $$self{indent}; |
179 | $$self{loose} = 0 unless defined $$self{loose}; |
180 | $$self{sentence} = 0 unless defined $$self{sentence}; |
181 | $$self{width} = 76 unless defined $$self{width}; |
69e00e79 |
182 | |
ab1f1d91 |
183 | # Figure out what quotes we'll be using for C<> text. |
50a3fd2a |
184 | $$self{quotes} ||= '"'; |
ab1f1d91 |
185 | if ($$self{quotes} eq 'none') { |
186 | $$self{LQUOTE} = $$self{RQUOTE} = ''; |
187 | } elsif (length ($$self{quotes}) == 1) { |
188 | $$self{LQUOTE} = $$self{RQUOTE} = $$self{quotes}; |
189 | } elsif ($$self{quotes} =~ /^(.)(.)$/ |
190 | || $$self{quotes} =~ /^(..)(..)$/) { |
191 | $$self{LQUOTE} = $1; |
192 | $$self{RQUOTE} = $2; |
193 | } else { |
194 | croak qq(Invalid quote specification "$$self{quotes}"); |
195 | } |
196 | |
6055f9d4 |
197 | $$self{INDENTS} = []; # Stack of indentations. |
198 | $$self{MARGIN} = $$self{indent}; # Current left margin in spaces. |
69e00e79 |
199 | |
6055f9d4 |
200 | $self->SUPER::initialize; |
59548eca |
201 | |
202 | # Tell Pod::Parser that we want the non-POD stuff too if code was set. |
203 | $self->parseopts ('-want_nonPODs' => 1) if $$self{code}; |
6055f9d4 |
204 | } |
69e00e79 |
205 | |
69e00e79 |
206 | |
3c014959 |
207 | ############################################################################## |
6055f9d4 |
208 | # Core overrides |
3c014959 |
209 | ############################################################################## |
6055f9d4 |
210 | |
211 | # Called for each command paragraph. Gets the command, the associated |
212 | # paragraph, the line number, and a Pod::Paragraph object. Just dispatches |
213 | # the command to a method named the same as the command. =cut is handled |
214 | # internally by Pod::Parser. |
215 | sub command { |
216 | my $self = shift; |
217 | my $command = shift; |
218 | return if $command eq 'pod'; |
219 | return if ($$self{EXCLUDE} && $command ne 'end'); |
220 | $self->item ("\n") if defined $$self{ITEM}; |
ab1f1d91 |
221 | if ($self->can ('cmd_' . $command)) { |
222 | $command = 'cmd_' . $command; |
223 | $self->$command (@_); |
224 | } else { |
225 | my ($text, $line, $paragraph) = @_; |
5cdeb5a2 |
226 | my $file; |
227 | ($file, $line) = $paragraph->file_line; |
ab1f1d91 |
228 | $text =~ s/\n+\z//; |
229 | $text = " $text" if ($text =~ /^\S/); |
230 | warn qq($file:$line: Unknown command paragraph "=$command$text"\n); |
231 | return; |
232 | } |
6055f9d4 |
233 | } |
69e00e79 |
234 | |
3c014959 |
235 | # Called for a verbatim paragraph. Gets the paragraph, the line number, and a |
236 | # Pod::Paragraph object. Just output it verbatim, but with tabs converted to |
237 | # spaces. |
6055f9d4 |
238 | sub verbatim { |
239 | my $self = shift; |
240 | return if $$self{EXCLUDE}; |
241 | $self->item if defined $$self{ITEM}; |
242 | local $_ = shift; |
243 | return if /^\s*$/; |
244 | s/^(\s*\S+)/(' ' x $$self{MARGIN}) . $1/gme; |
245 | $self->output ($_); |
246 | } |
69e00e79 |
247 | |
3c014959 |
248 | # Called for a regular text block. Gets the paragraph, the line number, and a |
249 | # Pod::Paragraph object. Perform interpolation and output the results. |
6055f9d4 |
250 | sub textblock { |
27f805f4 |
251 | my $self = shift; |
6055f9d4 |
252 | return if $$self{EXCLUDE}; |
27f805f4 |
253 | $self->output ($_[0]), return if $$self{VERBATIM}; |
254 | local $_ = shift; |
255 | my $line = shift; |
6055f9d4 |
256 | |
bf202ccd |
257 | # Interpolate and output the paragraph. |
6055f9d4 |
258 | $_ = $self->interpolate ($_, $line); |
259 | s/\s+$/\n/; |
260 | if (defined $$self{ITEM}) { |
261 | $self->item ($_ . "\n"); |
262 | } else { |
263 | $self->output ($self->reformat ($_ . "\n")); |
264 | } |
265 | } |
69e00e79 |
266 | |
6055f9d4 |
267 | # Called for an interior sequence. Gets the command, argument, and a |
268 | # Pod::InteriorSequence object and is expected to return the resulting text. |
3c014959 |
269 | # Calls code, bold, italic, file, and link to handle those types of sequences, |
270 | # and handles S<>, E<>, X<>, and Z<> directly. |
6055f9d4 |
271 | sub interior_sequence { |
bf202ccd |
272 | local $_; |
273 | my ($self, $command, $seq); |
274 | ($self, $command, $_, $seq) = @_; |
275 | |
276 | # We have to defer processing of the inside of an L<> formatting code. If |
277 | # this sequence is nested inside an L<> sequence, return the literal raw |
278 | # text of it. |
279 | my $parent = $seq->nested; |
280 | while (defined $parent) { |
281 | return $seq->raw_text if ($parent->cmd_name eq 'L'); |
282 | $parent = $parent->nested; |
283 | } |
284 | |
285 | # Index entries are ignored in plain text. |
6055f9d4 |
286 | return '' if ($command eq 'X' || $command eq 'Z'); |
69e00e79 |
287 | |
59548eca |
288 | # Expand escapes into the actual character now, warning if invalid. |
6055f9d4 |
289 | if ($command eq 'E') { |
2e20e14f |
290 | if (/^\d+$/) { |
291 | return chr; |
292 | } else { |
293 | return $ESCAPES{$_} if defined $ESCAPES{$_}; |
59548eca |
294 | my $seq = shift; |
295 | my ($file, $line) = $seq->file_line; |
296 | warn "$file:$line: Unknown escape: E<$_>\n"; |
2e20e14f |
297 | return "E<$_>"; |
298 | } |
6055f9d4 |
299 | } |
69e00e79 |
300 | |
6055f9d4 |
301 | # For all the other sequences, empty content produces no output. |
27f805f4 |
302 | return if $_ eq ''; |
69e00e79 |
303 | |
6055f9d4 |
304 | # For S<>, compress all internal whitespace and then map spaces to \01. |
305 | # When we output the text, we'll map this back. |
306 | if ($command eq 'S') { |
bf202ccd |
307 | s/\s+/ /g; |
6055f9d4 |
308 | tr/ /\01/; |
309 | return $_; |
310 | } |
69e00e79 |
311 | |
6055f9d4 |
312 | # Anything else needs to get dispatched to another method. |
313 | if ($command eq 'B') { return $self->seq_b ($_) } |
314 | elsif ($command eq 'C') { return $self->seq_c ($_) } |
315 | elsif ($command eq 'F') { return $self->seq_f ($_) } |
316 | elsif ($command eq 'I') { return $self->seq_i ($_) } |
bf202ccd |
317 | elsif ($command eq 'L') { return $self->seq_l ($_, $seq) } |
59548eca |
318 | else { |
319 | my $seq = shift; |
320 | my ($file, $line) = $seq->file_line; |
321 | warn "$file:$line: Unknown sequence $command<$_>\n"; |
322 | } |
6055f9d4 |
323 | } |
f02a87df |
324 | |
6055f9d4 |
325 | # Called for each paragraph that's actually part of the POD. We take |
59548eca |
326 | # advantage of this opportunity to untabify the input. Also, if given the |
327 | # code option, we may see paragraphs that aren't part of the POD and need to |
328 | # output them directly. |
6055f9d4 |
329 | sub preprocess_paragraph { |
330 | my $self = shift; |
331 | local $_ = shift; |
332 | 1 while s/^(.*?)(\t+)/$1 . ' ' x (length ($2) * 8 - length ($1) % 8)/me; |
59548eca |
333 | $self->output_code ($_) if $self->cutting; |
6055f9d4 |
334 | $_; |
335 | } |
3ec07288 |
336 | |
69e00e79 |
337 | |
3c014959 |
338 | ############################################################################## |
6055f9d4 |
339 | # Command paragraphs |
3c014959 |
340 | ############################################################################## |
f2506fb2 |
341 | |
6055f9d4 |
342 | # All command paragraphs take the paragraph and the line number. |
69e00e79 |
343 | |
6055f9d4 |
344 | # First level heading. |
345 | sub cmd_head1 { |
346 | my $self = shift; |
347 | local $_ = shift; |
348 | s/\s+$//; |
27f805f4 |
349 | $_ = $self->interpolate ($_, shift); |
6055f9d4 |
350 | if ($$self{alt}) { |
351 | $self->output ("\n==== $_ ====\n\n"); |
352 | } else { |
353 | $_ .= "\n" if $$self{loose}; |
354 | $self->output ($_ . "\n"); |
355 | } |
356 | } |
69e00e79 |
357 | |
6055f9d4 |
358 | # Second level heading. |
359 | sub cmd_head2 { |
360 | my $self = shift; |
361 | local $_ = shift; |
362 | s/\s+$//; |
27f805f4 |
363 | $_ = $self->interpolate ($_, shift); |
6055f9d4 |
364 | if ($$self{alt}) { |
365 | $self->output ("\n== $_ ==\n\n"); |
366 | } else { |
367 | $self->output (' ' x ($$self{indent} / 2) . $_ . "\n\n"); |
368 | } |
369 | } |
69e00e79 |
370 | |
50a3fd2a |
371 | # Third level heading. |
372 | sub cmd_head3 { |
373 | my $self = shift; |
374 | local $_ = shift; |
375 | s/\s+$//; |
376 | $_ = $self->interpolate ($_, shift); |
377 | if ($$self{alt}) { |
378 | $self->output ("\n= $_ =\n\n"); |
379 | } else { |
380 | $self->output (' ' x ($$self{indent} * 2 / 3 + 0.5) . $_ . "\n\n"); |
381 | } |
382 | } |
383 | |
384 | # Third level heading. |
385 | sub cmd_head4 { |
386 | my $self = shift; |
387 | local $_ = shift; |
388 | s/\s+$//; |
389 | $_ = $self->interpolate ($_, shift); |
390 | if ($$self{alt}) { |
391 | $self->output ("\n- $_ -\n\n"); |
392 | } else { |
393 | $self->output (' ' x ($$self{indent} * 3 / 4 + 0.5) . $_ . "\n\n"); |
394 | } |
395 | } |
396 | |
6055f9d4 |
397 | # Start a list. |
398 | sub cmd_over { |
399 | my $self = shift; |
400 | local $_ = shift; |
401 | unless (/^[-+]?\d+\s+$/) { $_ = $$self{indent} } |
402 | push (@{ $$self{INDENTS} }, $$self{MARGIN}); |
403 | $$self{MARGIN} += ($_ + 0); |
404 | } |
69e00e79 |
405 | |
6055f9d4 |
406 | # End a list. |
407 | sub cmd_back { |
59548eca |
408 | my ($self, $text, $line, $paragraph) = @_; |
6055f9d4 |
409 | $$self{MARGIN} = pop @{ $$self{INDENTS} }; |
410 | unless (defined $$self{MARGIN}) { |
59548eca |
411 | my $file; |
412 | ($file, $line) = $paragraph->file_line; |
413 | warn "$file:$line: Unmatched =back\n"; |
6055f9d4 |
414 | $$self{MARGIN} = $$self{indent}; |
415 | } |
69e00e79 |
416 | } |
417 | |
6055f9d4 |
418 | # An individual list item. |
419 | sub cmd_item { |
420 | my $self = shift; |
421 | if (defined $$self{ITEM}) { $self->item } |
422 | local $_ = shift; |
423 | s/\s+$//; |
424 | $$self{ITEM} = $self->interpolate ($_); |
425 | } |
69e00e79 |
426 | |
27f805f4 |
427 | # Begin a block for a particular translator. Setting VERBATIM triggers |
428 | # special handling in textblock(). |
6055f9d4 |
429 | sub cmd_begin { |
430 | my $self = shift; |
431 | local $_ = shift; |
432 | my ($kind) = /^(\S+)/ or return; |
27f805f4 |
433 | if ($kind eq 'text') { |
434 | $$self{VERBATIM} = 1; |
435 | } else { |
436 | $$self{EXCLUDE} = 1; |
437 | } |
6055f9d4 |
438 | } |
f2506fb2 |
439 | |
6055f9d4 |
440 | # End a block for a particular translator. We assume that all =begin/=end |
27f805f4 |
441 | # pairs are properly closed. |
6055f9d4 |
442 | sub cmd_end { |
443 | my $self = shift; |
27f805f4 |
444 | $$self{EXCLUDE} = 0; |
445 | $$self{VERBATIM} = 0; |
5cdeb5a2 |
446 | } |
6055f9d4 |
447 | |
448 | # One paragraph for a particular translator. Ignore it unless it's intended |
27f805f4 |
449 | # for text, in which case we treat it as a verbatim text block. |
6055f9d4 |
450 | sub cmd_for { |
451 | my $self = shift; |
452 | local $_ = shift; |
453 | my $line = shift; |
27f805f4 |
454 | return unless s/^text\b[ \t]*\n?//; |
455 | $self->verbatim ($_, $line); |
6055f9d4 |
456 | } |
f2506fb2 |
457 | |
69e00e79 |
458 | |
3c014959 |
459 | ############################################################################## |
6055f9d4 |
460 | # Interior sequences |
3c014959 |
461 | ############################################################################## |
69e00e79 |
462 | |
6055f9d4 |
463 | # The simple formatting ones. These are here mostly so that subclasses can |
464 | # override them and do more complicated things. |
27f805f4 |
465 | sub seq_b { return $_[0]{alt} ? "``$_[1]''" : $_[1] } |
27f805f4 |
466 | sub seq_f { return $_[0]{alt} ? "\"$_[1]\"" : $_[1] } |
6055f9d4 |
467 | sub seq_i { return '*' . $_[1] . '*' } |
3c014959 |
468 | |
469 | # Apply a whole bunch of messy heuristics to not quote things that don't |
470 | # benefit from being quoted. These originally come from Barrie Slaymaker and |
471 | # largely duplicate code in Pod::Man. |
ab1f1d91 |
472 | sub seq_c { |
3c014959 |
473 | my $self = shift; |
474 | local $_ = shift; |
475 | |
476 | # A regex that matches the portion of a variable reference that's the |
477 | # array or hash index, separated out just because we want to use it in |
478 | # several places in the following regex. |
479 | my $index = '(?: \[.*\] | \{.*\} )?'; |
480 | |
481 | # Check for things that we don't want to quote, and if we find any of |
482 | # them, return the string with just a font change and no quoting. |
483 | m{ |
484 | ^\s* |
485 | (?: |
486 | ( [\'\`\"] ) .* \1 # already quoted |
487 | | \` .* \' # `quoted' |
488 | | \$+ [\#^]? \S $index # special ($^Foo, $") |
489 | | [\$\@%&*]+ \#? [:\'\w]+ $index # plain var or func |
490 | | [\$\@%&*]* [:\'\w]+ (?: -> )? \(\s*[^\s,]\s*\) # 0/1-arg func call |
491 | | [+-]? [\d.]+ (?: [eE] [+-]? \d+ )? # a number |
492 | | 0x [a-fA-F\d]+ # a hex constant |
493 | ) |
494 | \s*\z |
495 | }xo && return $_; |
496 | |
497 | # If we didn't return, go ahead and quote the text. |
498 | return $$self{alt} ? "``$_''" : "$$self{LQUOTE}$_$$self{RQUOTE}"; |
ab1f1d91 |
499 | } |
69e00e79 |
500 | |
bf202ccd |
501 | # Handle links. Since this is plain text, we can't actually make any real |
502 | # links, so this is all to figure out what text we print out. Most of the |
503 | # work is done by Pod::ParseLink. |
6055f9d4 |
504 | sub seq_l { |
bf202ccd |
505 | my ($self, $link, $seq) = @_; |
506 | my ($text, $type) = (parselink ($link))[1,4]; |
507 | my ($file, $line) = $seq->file_line; |
508 | $text = $self->interpolate ($text, $line); |
509 | $text = '<' . $text . '>' if $type eq 'url'; |
510 | return $text || ''; |
69e00e79 |
511 | } |
512 | |
6055f9d4 |
513 | |
3c014959 |
514 | ############################################################################## |
6055f9d4 |
515 | # List handling |
3c014959 |
516 | ############################################################################## |
517 | |
518 | # This method is called whenever an =item command is complete (in other words, |
519 | # we've seen its associated paragraph or know for certain that it doesn't have |
520 | # one). It gets the paragraph associated with the item as an argument. If |
521 | # that argument is empty, just output the item tag; if it contains a newline, |
522 | # output the item tag followed by the newline. Otherwise, see if there's |
523 | # enough room for us to output the item tag in the margin of the text or if we |
524 | # have to put it on a separate line. |
6055f9d4 |
525 | sub item { |
526 | my $self = shift; |
527 | local $_ = shift; |
528 | my $tag = $$self{ITEM}; |
529 | unless (defined $tag) { |
59548eca |
530 | carp "Item called without tag"; |
6055f9d4 |
531 | return; |
69e00e79 |
532 | } |
6055f9d4 |
533 | undef $$self{ITEM}; |
534 | my $indent = $$self{INDENTS}[-1]; |
535 | unless (defined $indent) { $indent = $$self{indent} } |
536 | my $space = ' ' x $indent; |
537 | $space =~ s/^ /:/ if $$self{alt}; |
538 | if (!$_ || /^\s+$/ || ($$self{MARGIN} - $indent < length ($tag) + 1)) { |
27f805f4 |
539 | my $margin = $$self{MARGIN}; |
540 | $$self{MARGIN} = $indent; |
541 | my $output = $self->reformat ($tag); |
542 | $output =~ s/\n*$/\n/; |
543 | $self->output ($output); |
544 | $$self{MARGIN} = $margin; |
6055f9d4 |
545 | $self->output ($self->reformat ($_)) if /\S/; |
546 | } else { |
547 | $_ = $self->reformat ($_); |
548 | s/^ /:/ if ($$self{alt} && $indent > 0); |
549 | my $tagspace = ' ' x length $tag; |
550 | s/^($space)$tagspace/$1$tag/ or warn "Bizarre space in item"; |
551 | $self->output ($_); |
69e00e79 |
552 | } |
553 | } |
554 | |
69e00e79 |
555 | |
3c014959 |
556 | ############################################################################## |
6055f9d4 |
557 | # Output formatting |
3c014959 |
558 | ############################################################################## |
69e00e79 |
559 | |
3c014959 |
560 | # Wrap a line, indenting by the current left margin. We can't use Text::Wrap |
561 | # because it plays games with tabs. We can't use formline, even though we'd |
562 | # really like to, because it screws up non-printing characters. So we have to |
563 | # do the wrapping ourselves. |
6055f9d4 |
564 | sub wrap { |
565 | my $self = shift; |
566 | local $_ = shift; |
567 | my $output = ''; |
568 | my $spaces = ' ' x $$self{MARGIN}; |
569 | my $width = $$self{width} - $$self{MARGIN}; |
570 | while (length > $width) { |
571 | if (s/^([^\n]{0,$width})\s+// || s/^([^\n]{$width})//) { |
572 | $output .= $spaces . $1 . "\n"; |
573 | } else { |
574 | last; |
575 | } |
69e00e79 |
576 | } |
6055f9d4 |
577 | $output .= $spaces . $_; |
578 | $output =~ s/\s+$/\n\n/; |
579 | $output; |
69e00e79 |
580 | } |
581 | |
6055f9d4 |
582 | # Reformat a paragraph of text for the current margin. Takes the text to |
583 | # reformat and returns the formatted text. |
584 | sub reformat { |
585 | my $self = shift; |
586 | local $_ = shift; |
69e00e79 |
587 | |
3c014959 |
588 | # If we're trying to preserve two spaces after sentences, do some munging |
589 | # to support that. Otherwise, smash all repeated whitespace. |
6055f9d4 |
590 | if ($$self{sentence}) { |
591 | s/ +$//mg; |
592 | s/\.\n/. \n/g; |
593 | s/\n/ /g; |
594 | s/ +/ /g; |
69e00e79 |
595 | } else { |
6055f9d4 |
596 | s/\s+/ /g; |
69e00e79 |
597 | } |
6055f9d4 |
598 | $self->wrap ($_); |
69e00e79 |
599 | } |
600 | |
6055f9d4 |
601 | # Output text to the output device. |
602 | sub output { $_[1] =~ tr/\01/ /; print { $_[0]->output_handle } $_[1] } |
69e00e79 |
603 | |
59548eca |
604 | # Output a block of code (something that isn't part of the POD text). Called |
605 | # by preprocess_paragraph only if we were given the code option. Exists here |
606 | # only so that it can be overridden by subclasses. |
607 | sub output_code { $_[0]->output ($_[1]) } |
608 | |
69e00e79 |
609 | |
3c014959 |
610 | ############################################################################## |
27f805f4 |
611 | # Backwards compatibility |
3c014959 |
612 | ############################################################################## |
27f805f4 |
613 | |
614 | # The old Pod::Text module did everything in a pod2text() function. This |
615 | # tries to provide the same interface for legacy applications. |
616 | sub pod2text { |
617 | my @args; |
618 | |
619 | # This is really ugly; I hate doing option parsing in the middle of a |
620 | # module. But the old Pod::Text module supported passing flags to its |
621 | # entry function, so handle -a and -<number>. |
622 | while ($_[0] =~ /^-/) { |
623 | my $flag = shift; |
624 | if ($flag eq '-a') { push (@args, alt => 1) } |
625 | elsif ($flag =~ /^-(\d+)$/) { push (@args, width => $1) } |
626 | else { |
627 | unshift (@_, $flag); |
628 | last; |
629 | } |
630 | } |
631 | |
632 | # Now that we know what arguments we're using, create the parser. |
633 | my $parser = Pod::Text->new (@args); |
634 | |
635 | # If two arguments were given, the second argument is going to be a file |
3c014959 |
636 | # handle. That means we want to call parse_from_filehandle(), which means |
637 | # we need to turn the first argument into a file handle. Magic open will |
638 | # handle the <&STDIN case automagically. |
27f805f4 |
639 | if (defined $_[1]) { |
ab1f1d91 |
640 | my @fhs = @_; |
27f805f4 |
641 | local *IN; |
ab1f1d91 |
642 | unless (open (IN, $fhs[0])) { |
643 | croak ("Can't open $fhs[0] for reading: $!\n"); |
27f805f4 |
644 | return; |
645 | } |
ab1f1d91 |
646 | $fhs[0] = \*IN; |
647 | return $parser->parse_from_filehandle (@fhs); |
27f805f4 |
648 | } else { |
649 | return $parser->parse_from_file (@_); |
650 | } |
651 | } |
652 | |
653 | |
3c014959 |
654 | ############################################################################## |
6055f9d4 |
655 | # Module return value and documentation |
3c014959 |
656 | ############################################################################## |
69e00e79 |
657 | |
6055f9d4 |
658 | 1; |
659 | __END__ |
69e00e79 |
660 | |
6055f9d4 |
661 | =head1 NAME |
69e00e79 |
662 | |
6055f9d4 |
663 | Pod::Text - Convert POD data to formatted ASCII text |
69e00e79 |
664 | |
6055f9d4 |
665 | =head1 SYNOPSIS |
69e00e79 |
666 | |
6055f9d4 |
667 | use Pod::Text; |
668 | my $parser = Pod::Text->new (sentence => 0, width => 78); |
69e00e79 |
669 | |
6055f9d4 |
670 | # Read POD from STDIN and write to STDOUT. |
671 | $parser->parse_from_filehandle; |
69e00e79 |
672 | |
6055f9d4 |
673 | # Read POD from file.pod and write to file.txt. |
674 | $parser->parse_from_file ('file.pod', 'file.txt'); |
69e00e79 |
675 | |
6055f9d4 |
676 | =head1 DESCRIPTION |
5491a304 |
677 | |
27f805f4 |
678 | Pod::Text is a module that can convert documentation in the POD format (the |
679 | preferred language for documenting Perl) into formatted ASCII. It uses no |
680 | special formatting controls or codes whatsoever, and its output is therefore |
681 | suitable for nearly any device. |
69e00e79 |
682 | |
27f805f4 |
683 | As a derived class from Pod::Parser, Pod::Text supports the same methods and |
684 | interfaces. See L<Pod::Parser> for all the details; briefly, one creates a |
bf202ccd |
685 | new parser with C<< Pod::Text->new() >> and then calls either |
27f805f4 |
686 | parse_from_filehandle() or parse_from_file(). |
6055f9d4 |
687 | |
27f805f4 |
688 | new() can take options, in the form of key/value pairs, that control the |
6055f9d4 |
689 | behavior of the parser. The currently recognized options are: |
690 | |
691 | =over 4 |
692 | |
693 | =item alt |
694 | |
695 | If set to a true value, selects an alternate output format that, among other |
696 | things, uses a different heading style and marks C<=item> entries with a |
697 | colon in the left margin. Defaults to false. |
698 | |
59548eca |
699 | =item code |
700 | |
701 | If set to a true value, the non-POD parts of the input file will be included |
702 | in the output. Useful for viewing code documented with POD blocks with the |
703 | POD rendered and the code left intact. |
704 | |
6055f9d4 |
705 | =item indent |
706 | |
707 | The number of spaces to indent regular text, and the default indentation for |
708 | C<=over> blocks. Defaults to 4. |
709 | |
710 | =item loose |
711 | |
712 | If set to a true value, a blank line is printed after a C<=head1> heading. |
713 | If set to false (the default), no blank line is printed after C<=head1>, |
714 | although one is still printed after C<=head2>. This is the default because |
715 | it's the expected formatting for manual pages; if you're formatting |
716 | arbitrary text documents, setting this to true may result in more pleasing |
717 | output. |
718 | |
ab1f1d91 |
719 | =item quotes |
720 | |
721 | Sets the quote marks used to surround CE<lt>> text. If the value is a |
722 | single character, it is used as both the left and right quote; if it is two |
723 | characters, the first character is used as the left quote and the second as |
724 | the right quoted; and if it is four characters, the first two are used as |
725 | the left quote and the second two as the right quote. |
726 | |
727 | This may also be set to the special value C<none>, in which case no quote |
728 | marks are added around CE<lt>> text. |
729 | |
6055f9d4 |
730 | =item sentence |
731 | |
27f805f4 |
732 | If set to a true value, Pod::Text will assume that each sentence ends in two |
733 | spaces, and will try to preserve that spacing. If set to false, all |
6055f9d4 |
734 | consecutive whitespace in non-verbatim paragraphs is compressed into a |
735 | single space. Defaults to true. |
736 | |
737 | =item width |
738 | |
739 | The column at which to wrap text on the right-hand side. Defaults to 76. |
740 | |
741 | =back |
742 | |
27f805f4 |
743 | The standard Pod::Parser method parse_from_filehandle() takes up to two |
6055f9d4 |
744 | arguments, the first being the file handle to read POD from and the second |
745 | being the file handle to write the formatted output to. The first defaults |
746 | to STDIN if not given, and the second defaults to STDOUT. The method |
27f805f4 |
747 | parse_from_file() is almost identical, except that its two arguments are the |
748 | input and output disk files instead. See L<Pod::Parser> for the specific |
749 | details. |
6055f9d4 |
750 | |
751 | =head1 DIAGNOSTICS |
752 | |
753 | =over 4 |
754 | |
27f805f4 |
755 | =item Bizarre space in item |
756 | |
59548eca |
757 | =item Item called without tag |
758 | |
759 | (W) Something has gone wrong in internal C<=item> processing. These |
760 | messages indicate a bug in Pod::Text; you should never see them. |
27f805f4 |
761 | |
762 | =item Can't open %s for reading: %s |
763 | |
764 | (F) Pod::Text was invoked via the compatibility mode pod2text() interface |
765 | and the input file it was given could not be opened. |
766 | |
ab1f1d91 |
767 | =item Invalid quote specification "%s" |
768 | |
769 | (F) The quote specification given (the quotes option to the constructor) was |
770 | invalid. A quote specification must be one, two, or four characters long. |
771 | |
772 | =item %s:%d: Unknown command paragraph "%s". |
773 | |
774 | (W) The POD source contained a non-standard command paragraph (something of |
775 | the form C<=command args>) that Pod::Man didn't know about. It was ignored. |
776 | |
59548eca |
777 | =item %s:%d: Unknown escape: %s |
6055f9d4 |
778 | |
27f805f4 |
779 | (W) The POD source contained an C<EE<lt>E<gt>> escape that Pod::Text didn't |
780 | know about. |
6055f9d4 |
781 | |
59548eca |
782 | =item %s:%d: Unknown sequence: %s |
6055f9d4 |
783 | |
27f805f4 |
784 | (W) The POD source contained a non-standard internal sequence (something of |
785 | the form C<XE<lt>E<gt>>) that Pod::Text didn't know about. |
6055f9d4 |
786 | |
59548eca |
787 | =item %s:%d: Unmatched =back |
6055f9d4 |
788 | |
27f805f4 |
789 | (W) Pod::Text encountered a C<=back> command that didn't correspond to an |
6055f9d4 |
790 | C<=over> command. |
791 | |
792 | =back |
793 | |
27f805f4 |
794 | =head1 RESTRICTIONS |
795 | |
796 | Embedded Ctrl-As (octal 001) in the input will be mapped to spaces on |
797 | output, due to an internal implementation detail. |
798 | |
6055f9d4 |
799 | =head1 NOTES |
800 | |
27f805f4 |
801 | This is a replacement for an earlier Pod::Text module written by Tom |
802 | Christiansen. It has a revamped interface, since it now uses Pod::Parser, |
803 | but an interface roughly compatible with the old Pod::Text::pod2text() |
804 | function is still available. Please change to the new calling convention, |
805 | though. |
6055f9d4 |
806 | |
807 | The original Pod::Text contained code to do formatting via termcap |
808 | sequences, although it wasn't turned on by default and it was problematic to |
27f805f4 |
809 | get it to work at all. This rewrite doesn't even try to do that, but a |
bf202ccd |
810 | subclass of it does. Look for L<Pod::Text::Termcap>. |
6055f9d4 |
811 | |
812 | =head1 SEE ALSO |
813 | |
bf202ccd |
814 | L<Pod::Parser>, L<Pod::Text::Termcap>, L<pod2text(1)> |
6055f9d4 |
815 | |
816 | =head1 AUTHOR |
817 | |
bf202ccd |
818 | Russ Allbery <rra@stanford.edu>, based I<very> heavily on the original |
819 | Pod::Text by Tom Christiansen <tchrist@mox.perl.com> and its conversion to |
820 | Pod::Parser by Brad Appleton <bradapp@enteract.com>. |
6055f9d4 |
821 | |
3c014959 |
822 | =head1 COPYRIGHT AND LICENSE |
823 | |
824 | Copyright 1999, 2000, 2001 by Russ Allbery <rra@stanford.edu>. |
825 | |
826 | This program is free software; you may redistribute it and/or modify it |
827 | under the same terms as Perl itself. |
828 | |
6055f9d4 |
829 | =cut |