Commit | Line | Data |
6055f9d4 |
1 | # Pod::Text -- Convert POD data to formatted ASCII text. |
b616daaf |
2 | # $Id: Text.pm,v 2.15 2001/11/23 06:14:10 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. |
b616daaf |
45 | $VERSION = 2.15; |
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} && $command ne 'end'); |
ab1f1d91 |
220 | if ($self->can ('cmd_' . $command)) { |
221 | $command = 'cmd_' . $command; |
222 | $self->$command (@_); |
223 | } else { |
224 | my ($text, $line, $paragraph) = @_; |
5cdeb5a2 |
225 | my $file; |
226 | ($file, $line) = $paragraph->file_line; |
ab1f1d91 |
227 | $text =~ s/\n+\z//; |
228 | $text = " $text" if ($text =~ /^\S/); |
229 | warn qq($file:$line: Unknown command paragraph "=$command$text"\n); |
230 | return; |
231 | } |
6055f9d4 |
232 | } |
69e00e79 |
233 | |
3c014959 |
234 | # Called for a verbatim paragraph. Gets the paragraph, the line number, and a |
235 | # Pod::Paragraph object. Just output it verbatim, but with tabs converted to |
236 | # spaces. |
6055f9d4 |
237 | sub verbatim { |
238 | my $self = shift; |
239 | return if $$self{EXCLUDE}; |
240 | $self->item if defined $$self{ITEM}; |
241 | local $_ = shift; |
242 | return if /^\s*$/; |
243 | s/^(\s*\S+)/(' ' x $$self{MARGIN}) . $1/gme; |
244 | $self->output ($_); |
245 | } |
69e00e79 |
246 | |
3c014959 |
247 | # Called for a regular text block. Gets the paragraph, the line number, and a |
248 | # Pod::Paragraph object. Perform interpolation and output the results. |
6055f9d4 |
249 | sub textblock { |
27f805f4 |
250 | my $self = shift; |
6055f9d4 |
251 | return if $$self{EXCLUDE}; |
27f805f4 |
252 | $self->output ($_[0]), return if $$self{VERBATIM}; |
253 | local $_ = shift; |
254 | my $line = shift; |
6055f9d4 |
255 | |
bf202ccd |
256 | # Interpolate and output the paragraph. |
6055f9d4 |
257 | $_ = $self->interpolate ($_, $line); |
258 | s/\s+$/\n/; |
259 | if (defined $$self{ITEM}) { |
260 | $self->item ($_ . "\n"); |
261 | } else { |
262 | $self->output ($self->reformat ($_ . "\n")); |
263 | } |
264 | } |
69e00e79 |
265 | |
6055f9d4 |
266 | # Called for an interior sequence. Gets the command, argument, and a |
267 | # Pod::InteriorSequence object and is expected to return the resulting text. |
3c014959 |
268 | # Calls code, bold, italic, file, and link to handle those types of sequences, |
269 | # and handles S<>, E<>, X<>, and Z<> directly. |
6055f9d4 |
270 | sub interior_sequence { |
bf202ccd |
271 | local $_; |
272 | my ($self, $command, $seq); |
273 | ($self, $command, $_, $seq) = @_; |
274 | |
275 | # We have to defer processing of the inside of an L<> formatting code. If |
276 | # this sequence is nested inside an L<> sequence, return the literal raw |
277 | # text of it. |
278 | my $parent = $seq->nested; |
279 | while (defined $parent) { |
280 | return $seq->raw_text if ($parent->cmd_name eq 'L'); |
281 | $parent = $parent->nested; |
282 | } |
283 | |
284 | # Index entries are ignored in plain text. |
6055f9d4 |
285 | return '' if ($command eq 'X' || $command eq 'Z'); |
69e00e79 |
286 | |
59548eca |
287 | # Expand escapes into the actual character now, warning if invalid. |
6055f9d4 |
288 | if ($command eq 'E') { |
2e20e14f |
289 | if (/^\d+$/) { |
290 | return chr; |
291 | } else { |
292 | return $ESCAPES{$_} if defined $ESCAPES{$_}; |
59548eca |
293 | my $seq = shift; |
294 | my ($file, $line) = $seq->file_line; |
295 | warn "$file:$line: Unknown escape: E<$_>\n"; |
2e20e14f |
296 | return "E<$_>"; |
297 | } |
6055f9d4 |
298 | } |
69e00e79 |
299 | |
6055f9d4 |
300 | # For all the other sequences, empty content produces no output. |
27f805f4 |
301 | return if $_ eq ''; |
69e00e79 |
302 | |
6055f9d4 |
303 | # For S<>, compress all internal whitespace and then map spaces to \01. |
304 | # When we output the text, we'll map this back. |
305 | if ($command eq 'S') { |
bf202ccd |
306 | s/\s+/ /g; |
6055f9d4 |
307 | tr/ /\01/; |
308 | return $_; |
309 | } |
69e00e79 |
310 | |
6055f9d4 |
311 | # Anything else needs to get dispatched to another method. |
312 | if ($command eq 'B') { return $self->seq_b ($_) } |
313 | elsif ($command eq 'C') { return $self->seq_c ($_) } |
314 | elsif ($command eq 'F') { return $self->seq_f ($_) } |
315 | elsif ($command eq 'I') { return $self->seq_i ($_) } |
bf202ccd |
316 | elsif ($command eq 'L') { return $self->seq_l ($_, $seq) } |
59548eca |
317 | else { |
318 | my $seq = shift; |
319 | my ($file, $line) = $seq->file_line; |
320 | warn "$file:$line: Unknown sequence $command<$_>\n"; |
321 | } |
6055f9d4 |
322 | } |
f02a87df |
323 | |
6055f9d4 |
324 | # Called for each paragraph that's actually part of the POD. We take |
59548eca |
325 | # advantage of this opportunity to untabify the input. Also, if given the |
326 | # code option, we may see paragraphs that aren't part of the POD and need to |
327 | # output them directly. |
6055f9d4 |
328 | sub preprocess_paragraph { |
329 | my $self = shift; |
330 | local $_ = shift; |
331 | 1 while s/^(.*?)(\t+)/$1 . ' ' x (length ($2) * 8 - length ($1) % 8)/me; |
59548eca |
332 | $self->output_code ($_) if $self->cutting; |
6055f9d4 |
333 | $_; |
334 | } |
3ec07288 |
335 | |
69e00e79 |
336 | |
3c014959 |
337 | ############################################################################## |
6055f9d4 |
338 | # Command paragraphs |
3c014959 |
339 | ############################################################################## |
f2506fb2 |
340 | |
6055f9d4 |
341 | # All command paragraphs take the paragraph and the line number. |
69e00e79 |
342 | |
6055f9d4 |
343 | # First level heading. |
344 | sub cmd_head1 { |
b616daaf |
345 | my ($self, $text, $line) = @_; |
346 | $self->heading ($text, $line, 0, '===='); |
6055f9d4 |
347 | } |
69e00e79 |
348 | |
6055f9d4 |
349 | # Second level heading. |
350 | sub cmd_head2 { |
b616daaf |
351 | my ($self, $text, $line) = @_; |
352 | $self->heading ($text, $line, $$self{indent} / 2, '== '); |
6055f9d4 |
353 | } |
69e00e79 |
354 | |
50a3fd2a |
355 | # Third level heading. |
356 | sub cmd_head3 { |
b616daaf |
357 | my ($self, $text, $line) = @_; |
358 | $self->heading ($text, $line, $$self{indent} * 2 / 3 + 0.5, '= '); |
50a3fd2a |
359 | } |
360 | |
361 | # Third level heading. |
362 | sub cmd_head4 { |
b616daaf |
363 | my ($self, $text, $line) = @_; |
364 | $self->heading ($text, $line, $$self{indent} * 3 / 4 + 0.5, '- '); |
50a3fd2a |
365 | } |
366 | |
6055f9d4 |
367 | # Start a list. |
368 | sub cmd_over { |
369 | my $self = shift; |
370 | local $_ = shift; |
b616daaf |
371 | $self->item ("\n\n") if defined $$self{ITEM}; |
6055f9d4 |
372 | unless (/^[-+]?\d+\s+$/) { $_ = $$self{indent} } |
373 | push (@{ $$self{INDENTS} }, $$self{MARGIN}); |
374 | $$self{MARGIN} += ($_ + 0); |
375 | } |
69e00e79 |
376 | |
6055f9d4 |
377 | # End a list. |
378 | sub cmd_back { |
59548eca |
379 | my ($self, $text, $line, $paragraph) = @_; |
b616daaf |
380 | $self->item ("\n\n") if defined $$self{ITEM}; |
6055f9d4 |
381 | $$self{MARGIN} = pop @{ $$self{INDENTS} }; |
382 | unless (defined $$self{MARGIN}) { |
59548eca |
383 | my $file; |
384 | ($file, $line) = $paragraph->file_line; |
385 | warn "$file:$line: Unmatched =back\n"; |
6055f9d4 |
386 | $$self{MARGIN} = $$self{indent}; |
387 | } |
69e00e79 |
388 | } |
389 | |
6055f9d4 |
390 | # An individual list item. |
391 | sub cmd_item { |
392 | my $self = shift; |
393 | if (defined $$self{ITEM}) { $self->item } |
394 | local $_ = shift; |
395 | s/\s+$//; |
b616daaf |
396 | $$self{ITEM} = $_ ? $self->interpolate ($_) : '*'; |
6055f9d4 |
397 | } |
69e00e79 |
398 | |
27f805f4 |
399 | # Begin a block for a particular translator. Setting VERBATIM triggers |
400 | # special handling in textblock(). |
6055f9d4 |
401 | sub cmd_begin { |
402 | my $self = shift; |
403 | local $_ = shift; |
404 | my ($kind) = /^(\S+)/ or return; |
27f805f4 |
405 | if ($kind eq 'text') { |
406 | $$self{VERBATIM} = 1; |
407 | } else { |
408 | $$self{EXCLUDE} = 1; |
409 | } |
6055f9d4 |
410 | } |
f2506fb2 |
411 | |
6055f9d4 |
412 | # End a block for a particular translator. We assume that all =begin/=end |
27f805f4 |
413 | # pairs are properly closed. |
6055f9d4 |
414 | sub cmd_end { |
415 | my $self = shift; |
27f805f4 |
416 | $$self{EXCLUDE} = 0; |
417 | $$self{VERBATIM} = 0; |
5cdeb5a2 |
418 | } |
6055f9d4 |
419 | |
420 | # One paragraph for a particular translator. Ignore it unless it's intended |
27f805f4 |
421 | # for text, in which case we treat it as a verbatim text block. |
6055f9d4 |
422 | sub cmd_for { |
423 | my $self = shift; |
424 | local $_ = shift; |
425 | my $line = shift; |
27f805f4 |
426 | return unless s/^text\b[ \t]*\n?//; |
427 | $self->verbatim ($_, $line); |
6055f9d4 |
428 | } |
f2506fb2 |
429 | |
69e00e79 |
430 | |
3c014959 |
431 | ############################################################################## |
6055f9d4 |
432 | # Interior sequences |
3c014959 |
433 | ############################################################################## |
69e00e79 |
434 | |
6055f9d4 |
435 | # The simple formatting ones. These are here mostly so that subclasses can |
436 | # override them and do more complicated things. |
27f805f4 |
437 | sub seq_b { return $_[0]{alt} ? "``$_[1]''" : $_[1] } |
27f805f4 |
438 | sub seq_f { return $_[0]{alt} ? "\"$_[1]\"" : $_[1] } |
6055f9d4 |
439 | sub seq_i { return '*' . $_[1] . '*' } |
3c014959 |
440 | |
441 | # Apply a whole bunch of messy heuristics to not quote things that don't |
442 | # benefit from being quoted. These originally come from Barrie Slaymaker and |
443 | # largely duplicate code in Pod::Man. |
ab1f1d91 |
444 | sub seq_c { |
3c014959 |
445 | my $self = shift; |
446 | local $_ = shift; |
447 | |
448 | # A regex that matches the portion of a variable reference that's the |
449 | # array or hash index, separated out just because we want to use it in |
450 | # several places in the following regex. |
451 | my $index = '(?: \[.*\] | \{.*\} )?'; |
452 | |
453 | # Check for things that we don't want to quote, and if we find any of |
454 | # them, return the string with just a font change and no quoting. |
455 | m{ |
456 | ^\s* |
457 | (?: |
458 | ( [\'\`\"] ) .* \1 # already quoted |
459 | | \` .* \' # `quoted' |
460 | | \$+ [\#^]? \S $index # special ($^Foo, $") |
461 | | [\$\@%&*]+ \#? [:\'\w]+ $index # plain var or func |
462 | | [\$\@%&*]* [:\'\w]+ (?: -> )? \(\s*[^\s,]\s*\) # 0/1-arg func call |
463 | | [+-]? [\d.]+ (?: [eE] [+-]? \d+ )? # a number |
464 | | 0x [a-fA-F\d]+ # a hex constant |
465 | ) |
466 | \s*\z |
467 | }xo && return $_; |
468 | |
469 | # If we didn't return, go ahead and quote the text. |
470 | return $$self{alt} ? "``$_''" : "$$self{LQUOTE}$_$$self{RQUOTE}"; |
ab1f1d91 |
471 | } |
69e00e79 |
472 | |
bf202ccd |
473 | # Handle links. Since this is plain text, we can't actually make any real |
474 | # links, so this is all to figure out what text we print out. Most of the |
475 | # work is done by Pod::ParseLink. |
6055f9d4 |
476 | sub seq_l { |
bf202ccd |
477 | my ($self, $link, $seq) = @_; |
478 | my ($text, $type) = (parselink ($link))[1,4]; |
479 | my ($file, $line) = $seq->file_line; |
480 | $text = $self->interpolate ($text, $line); |
481 | $text = '<' . $text . '>' if $type eq 'url'; |
482 | return $text || ''; |
69e00e79 |
483 | } |
484 | |
6055f9d4 |
485 | |
3c014959 |
486 | ############################################################################## |
b616daaf |
487 | # Header handling |
488 | ############################################################################## |
489 | |
490 | # The common code for handling all headers. Takes the interpolated header |
491 | # text, the line number, the indentation, and the surrounding marker for the |
492 | # alt formatting method. |
493 | sub heading { |
494 | my ($self, $text, $line, $indent, $marker) = @_; |
495 | $self->item ("\n\n") if defined $$self{ITEM}; |
496 | $text =~ s/\s+$//; |
497 | $text = $self->interpolate ($text, $line); |
498 | if ($$self{alt}) { |
499 | my $closemark = reverse (split (//, $marker)); |
500 | $self->output ("\n" . "$marker $text $closemark" . "\n\n"); |
501 | } else { |
502 | $text .= "\n" if $$self{loose}; |
503 | $self->output (' ' x $indent . $text . "\n"); |
504 | } |
505 | } |
506 | |
507 | |
508 | ############################################################################## |
6055f9d4 |
509 | # List handling |
3c014959 |
510 | ############################################################################## |
511 | |
512 | # This method is called whenever an =item command is complete (in other words, |
513 | # we've seen its associated paragraph or know for certain that it doesn't have |
514 | # one). It gets the paragraph associated with the item as an argument. If |
515 | # that argument is empty, just output the item tag; if it contains a newline, |
516 | # output the item tag followed by the newline. Otherwise, see if there's |
517 | # enough room for us to output the item tag in the margin of the text or if we |
518 | # have to put it on a separate line. |
6055f9d4 |
519 | sub item { |
520 | my $self = shift; |
521 | local $_ = shift; |
522 | my $tag = $$self{ITEM}; |
523 | unless (defined $tag) { |
59548eca |
524 | carp "Item called without tag"; |
6055f9d4 |
525 | return; |
69e00e79 |
526 | } |
6055f9d4 |
527 | undef $$self{ITEM}; |
528 | my $indent = $$self{INDENTS}[-1]; |
529 | unless (defined $indent) { $indent = $$self{indent} } |
530 | my $space = ' ' x $indent; |
531 | $space =~ s/^ /:/ if $$self{alt}; |
532 | if (!$_ || /^\s+$/ || ($$self{MARGIN} - $indent < length ($tag) + 1)) { |
27f805f4 |
533 | my $margin = $$self{MARGIN}; |
534 | $$self{MARGIN} = $indent; |
535 | my $output = $self->reformat ($tag); |
536 | $output =~ s/\n*$/\n/; |
b616daaf |
537 | |
538 | # If the text is just whitespace, we have an empty item paragraph; |
539 | # this can result from =over/=item/=back without any intermixed |
540 | # paragraphs. Insert some whitespace to keep the =item from merging |
541 | # into the next paragraph. |
542 | $output .= "\n" if $_ && $_ =~ /^\s*$/; |
543 | |
27f805f4 |
544 | $self->output ($output); |
545 | $$self{MARGIN} = $margin; |
b616daaf |
546 | $self->output ($self->reformat ($_)) if $_ && /\S/; |
6055f9d4 |
547 | } else { |
548 | $_ = $self->reformat ($_); |
549 | s/^ /:/ if ($$self{alt} && $indent > 0); |
550 | my $tagspace = ' ' x length $tag; |
551 | s/^($space)$tagspace/$1$tag/ or warn "Bizarre space in item"; |
552 | $self->output ($_); |
69e00e79 |
553 | } |
554 | } |
555 | |
69e00e79 |
556 | |
3c014959 |
557 | ############################################################################## |
6055f9d4 |
558 | # Output formatting |
3c014959 |
559 | ############################################################################## |
69e00e79 |
560 | |
3c014959 |
561 | # Wrap a line, indenting by the current left margin. We can't use Text::Wrap |
562 | # because it plays games with tabs. We can't use formline, even though we'd |
563 | # really like to, because it screws up non-printing characters. So we have to |
564 | # do the wrapping ourselves. |
6055f9d4 |
565 | sub wrap { |
566 | my $self = shift; |
567 | local $_ = shift; |
568 | my $output = ''; |
569 | my $spaces = ' ' x $$self{MARGIN}; |
570 | my $width = $$self{width} - $$self{MARGIN}; |
571 | while (length > $width) { |
572 | if (s/^([^\n]{0,$width})\s+// || s/^([^\n]{$width})//) { |
573 | $output .= $spaces . $1 . "\n"; |
574 | } else { |
575 | last; |
576 | } |
69e00e79 |
577 | } |
6055f9d4 |
578 | $output .= $spaces . $_; |
579 | $output =~ s/\s+$/\n\n/; |
580 | $output; |
69e00e79 |
581 | } |
582 | |
6055f9d4 |
583 | # Reformat a paragraph of text for the current margin. Takes the text to |
584 | # reformat and returns the formatted text. |
585 | sub reformat { |
586 | my $self = shift; |
587 | local $_ = shift; |
69e00e79 |
588 | |
3c014959 |
589 | # If we're trying to preserve two spaces after sentences, do some munging |
590 | # to support that. Otherwise, smash all repeated whitespace. |
6055f9d4 |
591 | if ($$self{sentence}) { |
592 | s/ +$//mg; |
593 | s/\.\n/. \n/g; |
594 | s/\n/ /g; |
595 | s/ +/ /g; |
69e00e79 |
596 | } else { |
6055f9d4 |
597 | s/\s+/ /g; |
69e00e79 |
598 | } |
6055f9d4 |
599 | $self->wrap ($_); |
69e00e79 |
600 | } |
601 | |
6055f9d4 |
602 | # Output text to the output device. |
603 | sub output { $_[1] =~ tr/\01/ /; print { $_[0]->output_handle } $_[1] } |
69e00e79 |
604 | |
59548eca |
605 | # Output a block of code (something that isn't part of the POD text). Called |
606 | # by preprocess_paragraph only if we were given the code option. Exists here |
607 | # only so that it can be overridden by subclasses. |
608 | sub output_code { $_[0]->output ($_[1]) } |
609 | |
69e00e79 |
610 | |
3c014959 |
611 | ############################################################################## |
27f805f4 |
612 | # Backwards compatibility |
3c014959 |
613 | ############################################################################## |
27f805f4 |
614 | |
615 | # The old Pod::Text module did everything in a pod2text() function. This |
616 | # tries to provide the same interface for legacy applications. |
617 | sub pod2text { |
618 | my @args; |
619 | |
620 | # This is really ugly; I hate doing option parsing in the middle of a |
621 | # module. But the old Pod::Text module supported passing flags to its |
622 | # entry function, so handle -a and -<number>. |
623 | while ($_[0] =~ /^-/) { |
624 | my $flag = shift; |
625 | if ($flag eq '-a') { push (@args, alt => 1) } |
626 | elsif ($flag =~ /^-(\d+)$/) { push (@args, width => $1) } |
627 | else { |
628 | unshift (@_, $flag); |
629 | last; |
630 | } |
631 | } |
632 | |
633 | # Now that we know what arguments we're using, create the parser. |
634 | my $parser = Pod::Text->new (@args); |
635 | |
636 | # If two arguments were given, the second argument is going to be a file |
3c014959 |
637 | # handle. That means we want to call parse_from_filehandle(), which means |
638 | # we need to turn the first argument into a file handle. Magic open will |
639 | # handle the <&STDIN case automagically. |
27f805f4 |
640 | if (defined $_[1]) { |
ab1f1d91 |
641 | my @fhs = @_; |
27f805f4 |
642 | local *IN; |
ab1f1d91 |
643 | unless (open (IN, $fhs[0])) { |
644 | croak ("Can't open $fhs[0] for reading: $!\n"); |
27f805f4 |
645 | return; |
646 | } |
ab1f1d91 |
647 | $fhs[0] = \*IN; |
648 | return $parser->parse_from_filehandle (@fhs); |
27f805f4 |
649 | } else { |
650 | return $parser->parse_from_file (@_); |
651 | } |
652 | } |
653 | |
654 | |
3c014959 |
655 | ############################################################################## |
6055f9d4 |
656 | # Module return value and documentation |
3c014959 |
657 | ############################################################################## |
69e00e79 |
658 | |
6055f9d4 |
659 | 1; |
660 | __END__ |
69e00e79 |
661 | |
6055f9d4 |
662 | =head1 NAME |
69e00e79 |
663 | |
6055f9d4 |
664 | Pod::Text - Convert POD data to formatted ASCII text |
69e00e79 |
665 | |
6055f9d4 |
666 | =head1 SYNOPSIS |
69e00e79 |
667 | |
6055f9d4 |
668 | use Pod::Text; |
669 | my $parser = Pod::Text->new (sentence => 0, width => 78); |
69e00e79 |
670 | |
6055f9d4 |
671 | # Read POD from STDIN and write to STDOUT. |
672 | $parser->parse_from_filehandle; |
69e00e79 |
673 | |
6055f9d4 |
674 | # Read POD from file.pod and write to file.txt. |
675 | $parser->parse_from_file ('file.pod', 'file.txt'); |
69e00e79 |
676 | |
6055f9d4 |
677 | =head1 DESCRIPTION |
5491a304 |
678 | |
27f805f4 |
679 | Pod::Text is a module that can convert documentation in the POD format (the |
680 | preferred language for documenting Perl) into formatted ASCII. It uses no |
681 | special formatting controls or codes whatsoever, and its output is therefore |
682 | suitable for nearly any device. |
69e00e79 |
683 | |
27f805f4 |
684 | As a derived class from Pod::Parser, Pod::Text supports the same methods and |
685 | interfaces. See L<Pod::Parser> for all the details; briefly, one creates a |
bf202ccd |
686 | new parser with C<< Pod::Text->new() >> and then calls either |
27f805f4 |
687 | parse_from_filehandle() or parse_from_file(). |
6055f9d4 |
688 | |
27f805f4 |
689 | new() can take options, in the form of key/value pairs, that control the |
6055f9d4 |
690 | behavior of the parser. The currently recognized options are: |
691 | |
692 | =over 4 |
693 | |
694 | =item alt |
695 | |
696 | If set to a true value, selects an alternate output format that, among other |
697 | things, uses a different heading style and marks C<=item> entries with a |
698 | colon in the left margin. Defaults to false. |
699 | |
59548eca |
700 | =item code |
701 | |
702 | If set to a true value, the non-POD parts of the input file will be included |
703 | in the output. Useful for viewing code documented with POD blocks with the |
704 | POD rendered and the code left intact. |
705 | |
6055f9d4 |
706 | =item indent |
707 | |
708 | The number of spaces to indent regular text, and the default indentation for |
709 | C<=over> blocks. Defaults to 4. |
710 | |
711 | =item loose |
712 | |
713 | If set to a true value, a blank line is printed after a C<=head1> heading. |
714 | If set to false (the default), no blank line is printed after C<=head1>, |
715 | although one is still printed after C<=head2>. This is the default because |
716 | it's the expected formatting for manual pages; if you're formatting |
717 | arbitrary text documents, setting this to true may result in more pleasing |
718 | output. |
719 | |
ab1f1d91 |
720 | =item quotes |
721 | |
722 | Sets the quote marks used to surround CE<lt>> text. If the value is a |
723 | single character, it is used as both the left and right quote; if it is two |
724 | characters, the first character is used as the left quote and the second as |
725 | the right quoted; and if it is four characters, the first two are used as |
726 | the left quote and the second two as the right quote. |
727 | |
728 | This may also be set to the special value C<none>, in which case no quote |
729 | marks are added around CE<lt>> text. |
730 | |
6055f9d4 |
731 | =item sentence |
732 | |
27f805f4 |
733 | If set to a true value, Pod::Text will assume that each sentence ends in two |
734 | spaces, and will try to preserve that spacing. If set to false, all |
6055f9d4 |
735 | consecutive whitespace in non-verbatim paragraphs is compressed into a |
736 | single space. Defaults to true. |
737 | |
738 | =item width |
739 | |
740 | The column at which to wrap text on the right-hand side. Defaults to 76. |
741 | |
742 | =back |
743 | |
27f805f4 |
744 | The standard Pod::Parser method parse_from_filehandle() takes up to two |
6055f9d4 |
745 | arguments, the first being the file handle to read POD from and the second |
746 | being the file handle to write the formatted output to. The first defaults |
747 | to STDIN if not given, and the second defaults to STDOUT. The method |
27f805f4 |
748 | parse_from_file() is almost identical, except that its two arguments are the |
749 | input and output disk files instead. See L<Pod::Parser> for the specific |
750 | details. |
6055f9d4 |
751 | |
752 | =head1 DIAGNOSTICS |
753 | |
754 | =over 4 |
755 | |
27f805f4 |
756 | =item Bizarre space in item |
757 | |
59548eca |
758 | =item Item called without tag |
759 | |
760 | (W) Something has gone wrong in internal C<=item> processing. These |
761 | messages indicate a bug in Pod::Text; you should never see them. |
27f805f4 |
762 | |
763 | =item Can't open %s for reading: %s |
764 | |
765 | (F) Pod::Text was invoked via the compatibility mode pod2text() interface |
766 | and the input file it was given could not be opened. |
767 | |
ab1f1d91 |
768 | =item Invalid quote specification "%s" |
769 | |
770 | (F) The quote specification given (the quotes option to the constructor) was |
771 | invalid. A quote specification must be one, two, or four characters long. |
772 | |
773 | =item %s:%d: Unknown command paragraph "%s". |
774 | |
775 | (W) The POD source contained a non-standard command paragraph (something of |
776 | the form C<=command args>) that Pod::Man didn't know about. It was ignored. |
777 | |
59548eca |
778 | =item %s:%d: Unknown escape: %s |
6055f9d4 |
779 | |
27f805f4 |
780 | (W) The POD source contained an C<EE<lt>E<gt>> escape that Pod::Text didn't |
781 | know about. |
6055f9d4 |
782 | |
59548eca |
783 | =item %s:%d: Unknown sequence: %s |
6055f9d4 |
784 | |
27f805f4 |
785 | (W) The POD source contained a non-standard internal sequence (something of |
786 | the form C<XE<lt>E<gt>>) that Pod::Text didn't know about. |
6055f9d4 |
787 | |
59548eca |
788 | =item %s:%d: Unmatched =back |
6055f9d4 |
789 | |
27f805f4 |
790 | (W) Pod::Text encountered a C<=back> command that didn't correspond to an |
6055f9d4 |
791 | C<=over> command. |
792 | |
793 | =back |
794 | |
27f805f4 |
795 | =head1 RESTRICTIONS |
796 | |
797 | Embedded Ctrl-As (octal 001) in the input will be mapped to spaces on |
798 | output, due to an internal implementation detail. |
799 | |
6055f9d4 |
800 | =head1 NOTES |
801 | |
27f805f4 |
802 | This is a replacement for an earlier Pod::Text module written by Tom |
803 | Christiansen. It has a revamped interface, since it now uses Pod::Parser, |
804 | but an interface roughly compatible with the old Pod::Text::pod2text() |
805 | function is still available. Please change to the new calling convention, |
806 | though. |
6055f9d4 |
807 | |
808 | The original Pod::Text contained code to do formatting via termcap |
809 | sequences, although it wasn't turned on by default and it was problematic to |
27f805f4 |
810 | get it to work at all. This rewrite doesn't even try to do that, but a |
bf202ccd |
811 | subclass of it does. Look for L<Pod::Text::Termcap>. |
6055f9d4 |
812 | |
813 | =head1 SEE ALSO |
814 | |
bf202ccd |
815 | L<Pod::Parser>, L<Pod::Text::Termcap>, L<pod2text(1)> |
6055f9d4 |
816 | |
817 | =head1 AUTHOR |
818 | |
bf202ccd |
819 | Russ Allbery <rra@stanford.edu>, based I<very> heavily on the original |
820 | Pod::Text by Tom Christiansen <tchrist@mox.perl.com> and its conversion to |
821 | Pod::Parser by Brad Appleton <bradapp@enteract.com>. |
6055f9d4 |
822 | |
3c014959 |
823 | =head1 COPYRIGHT AND LICENSE |
824 | |
825 | Copyright 1999, 2000, 2001 by Russ Allbery <rra@stanford.edu>. |
826 | |
827 | This program is free software; you may redistribute it and/or modify it |
828 | under the same terms as Perl itself. |
829 | |
6055f9d4 |
830 | =cut |