1 | # Pod::Text -- Convert POD data to formatted ASCII text.
|
---|
2 | # $Id: Text.pm,v 2.21 2002/08/04 03:34:58 eagle Exp $
|
---|
3 | #
|
---|
4 | # Copyright 1999, 2000, 2001, 2002 by Russ Allbery <[email protected]>
|
---|
5 | #
|
---|
6 | # This program is free software; you may redistribute it and/or modify it
|
---|
7 | # under the same terms as Perl itself.
|
---|
8 | #
|
---|
9 | # This module converts POD to formatted text. It replaces the old Pod::Text
|
---|
10 | # module that came with versions of Perl prior to 5.6.0 and attempts to match
|
---|
11 | # its output except for some specific circumstances where other decisions
|
---|
12 | # seemed to produce better output. It uses Pod::Parser and is designed to be
|
---|
13 | # very easy to subclass.
|
---|
14 | #
|
---|
15 | # Perl core hackers, please note that this module is also separately
|
---|
16 | # maintained outside of the Perl core as part of the podlators. Please send
|
---|
17 | # me any patches at the address above in addition to sending them to the
|
---|
18 | # standard Perl mailing lists.
|
---|
19 |
|
---|
20 | ##############################################################################
|
---|
21 | # Modules and declarations
|
---|
22 | ##############################################################################
|
---|
23 |
|
---|
24 | package Pod::Text;
|
---|
25 |
|
---|
26 | require 5.004;
|
---|
27 |
|
---|
28 | use Carp qw(carp croak);
|
---|
29 | use Exporter ();
|
---|
30 | use Pod::ParseLink qw(parselink);
|
---|
31 | use Pod::Select ();
|
---|
32 |
|
---|
33 | use strict;
|
---|
34 | use vars qw(@ISA @EXPORT %ESCAPES $VERSION);
|
---|
35 |
|
---|
36 | # We inherit from Pod::Select instead of Pod::Parser so that we can be used by
|
---|
37 | # Pod::Usage.
|
---|
38 | @ISA = qw(Pod::Select Exporter);
|
---|
39 |
|
---|
40 | # We have to export pod2text for backward compatibility.
|
---|
41 | @EXPORT = qw(pod2text);
|
---|
42 |
|
---|
43 | # Don't use the CVS revision as the version, since this module is also in Perl
|
---|
44 | # core and too many things could munge CVS magic revision strings. This
|
---|
45 | # number should ideally be the same as the CVS revision in podlators, however.
|
---|
46 | $VERSION = 2.21;
|
---|
47 |
|
---|
48 |
|
---|
49 | ##############################################################################
|
---|
50 | # Table of supported E<> escapes
|
---|
51 | ##############################################################################
|
---|
52 |
|
---|
53 | # This table is taken near verbatim from Pod::PlainText in Pod::Parser, which
|
---|
54 | # got it near verbatim from the original Pod::Text. It is therefore credited
|
---|
55 | # to Tom Christiansen, and I'm glad I didn't have to write it. :) "iexcl" to
|
---|
56 | # "divide" added by Tim Jenness.
|
---|
57 | %ESCAPES = (
|
---|
58 | 'amp' => '&', # ampersand
|
---|
59 | 'apos' => "'", # apostrophe
|
---|
60 | 'lt' => '<', # left chevron, less-than
|
---|
61 | 'gt' => '>', # right chevron, greater-than
|
---|
62 | 'quot' => '"', # double quote
|
---|
63 | 'sol' => '/', # solidus (forward slash)
|
---|
64 | 'verbar' => '|', # vertical bar
|
---|
65 |
|
---|
66 | "Aacute" => "\xC1", # capital A, acute accent
|
---|
67 | "aacute" => "\xE1", # small a, acute accent
|
---|
68 | "Acirc" => "\xC2", # capital A, circumflex accent
|
---|
69 | "acirc" => "\xE2", # small a, circumflex accent
|
---|
70 | "AElig" => "\xC6", # capital AE diphthong (ligature)
|
---|
71 | "aelig" => "\xE6", # small ae diphthong (ligature)
|
---|
72 | "Agrave" => "\xC0", # capital A, grave accent
|
---|
73 | "agrave" => "\xE0", # small a, grave accent
|
---|
74 | "Aring" => "\xC5", # capital A, ring
|
---|
75 | "aring" => "\xE5", # small a, ring
|
---|
76 | "Atilde" => "\xC3", # capital A, tilde
|
---|
77 | "atilde" => "\xE3", # small a, tilde
|
---|
78 | "Auml" => "\xC4", # capital A, dieresis or umlaut mark
|
---|
79 | "auml" => "\xE4", # small a, dieresis or umlaut mark
|
---|
80 | "Ccedil" => "\xC7", # capital C, cedilla
|
---|
81 | "ccedil" => "\xE7", # small c, cedilla
|
---|
82 | "Eacute" => "\xC9", # capital E, acute accent
|
---|
83 | "eacute" => "\xE9", # small e, acute accent
|
---|
84 | "Ecirc" => "\xCA", # capital E, circumflex accent
|
---|
85 | "ecirc" => "\xEA", # small e, circumflex accent
|
---|
86 | "Egrave" => "\xC8", # capital E, grave accent
|
---|
87 | "egrave" => "\xE8", # small e, grave accent
|
---|
88 | "ETH" => "\xD0", # capital Eth, Icelandic
|
---|
89 | "eth" => "\xF0", # small eth, Icelandic
|
---|
90 | "Euml" => "\xCB", # capital E, dieresis or umlaut mark
|
---|
91 | "euml" => "\xEB", # small e, dieresis or umlaut mark
|
---|
92 | "Iacute" => "\xCD", # capital I, acute accent
|
---|
93 | "iacute" => "\xED", # small i, acute accent
|
---|
94 | "Icirc" => "\xCE", # capital I, circumflex accent
|
---|
95 | "icirc" => "\xEE", # small i, circumflex accent
|
---|
96 | "Igrave" => "\xCC", # capital I, grave accent
|
---|
97 | "igrave" => "\xEC", # small i, grave accent
|
---|
98 | "Iuml" => "\xCF", # capital I, dieresis or umlaut mark
|
---|
99 | "iuml" => "\xEF", # small i, dieresis or umlaut mark
|
---|
100 | "Ntilde" => "\xD1", # capital N, tilde
|
---|
101 | "ntilde" => "\xF1", # small n, tilde
|
---|
102 | "Oacute" => "\xD3", # capital O, acute accent
|
---|
103 | "oacute" => "\xF3", # small o, acute accent
|
---|
104 | "Ocirc" => "\xD4", # capital O, circumflex accent
|
---|
105 | "ocirc" => "\xF4", # small o, circumflex accent
|
---|
106 | "Ograve" => "\xD2", # capital O, grave accent
|
---|
107 | "ograve" => "\xF2", # small o, grave accent
|
---|
108 | "Oslash" => "\xD8", # capital O, slash
|
---|
109 | "oslash" => "\xF8", # small o, slash
|
---|
110 | "Otilde" => "\xD5", # capital O, tilde
|
---|
111 | "otilde" => "\xF5", # small o, tilde
|
---|
112 | "Ouml" => "\xD6", # capital O, dieresis or umlaut mark
|
---|
113 | "ouml" => "\xF6", # small o, dieresis or umlaut mark
|
---|
114 | "szlig" => "\xDF", # small sharp s, German (sz ligature)
|
---|
115 | "THORN" => "\xDE", # capital THORN, Icelandic
|
---|
116 | "thorn" => "\xFE", # small thorn, Icelandic
|
---|
117 | "Uacute" => "\xDA", # capital U, acute accent
|
---|
118 | "uacute" => "\xFA", # small u, acute accent
|
---|
119 | "Ucirc" => "\xDB", # capital U, circumflex accent
|
---|
120 | "ucirc" => "\xFB", # small u, circumflex accent
|
---|
121 | "Ugrave" => "\xD9", # capital U, grave accent
|
---|
122 | "ugrave" => "\xF9", # small u, grave accent
|
---|
123 | "Uuml" => "\xDC", # capital U, dieresis or umlaut mark
|
---|
124 | "uuml" => "\xFC", # small u, dieresis or umlaut mark
|
---|
125 | "Yacute" => "\xDD", # capital Y, acute accent
|
---|
126 | "yacute" => "\xFD", # small y, acute accent
|
---|
127 | "yuml" => "\xFF", # small y, dieresis or umlaut mark
|
---|
128 |
|
---|
129 | "laquo" => "\xAB", # left pointing double angle quotation mark
|
---|
130 | "lchevron" => "\xAB", # synonym (backwards compatibility)
|
---|
131 | "raquo" => "\xBB", # right pointing double angle quotation mark
|
---|
132 | "rchevron" => "\xBB", # synonym (backwards compatibility)
|
---|
133 |
|
---|
134 | "iexcl" => "\xA1", # inverted exclamation mark
|
---|
135 | "cent" => "\xA2", # cent sign
|
---|
136 | "pound" => "\xA3", # (UK) pound sign
|
---|
137 | "curren" => "\xA4", # currency sign
|
---|
138 | "yen" => "\xA5", # yen sign
|
---|
139 | "brvbar" => "\xA6", # broken vertical bar
|
---|
140 | "sect" => "\xA7", # section sign
|
---|
141 | "uml" => "\xA8", # diaresis
|
---|
142 | "copy" => "\xA9", # Copyright symbol
|
---|
143 | "ordf" => "\xAA", # feminine ordinal indicator
|
---|
144 | "not" => "\xAC", # not sign
|
---|
145 | "shy" => '', # soft (discretionary) hyphen
|
---|
146 | "reg" => "\xAE", # registered trademark
|
---|
147 | "macr" => "\xAF", # macron, overline
|
---|
148 | "deg" => "\xB0", # degree sign
|
---|
149 | "plusmn" => "\xB1", # plus-minus sign
|
---|
150 | "sup2" => "\xB2", # superscript 2
|
---|
151 | "sup3" => "\xB3", # superscript 3
|
---|
152 | "acute" => "\xB4", # acute accent
|
---|
153 | "micro" => "\xB5", # micro sign
|
---|
154 | "para" => "\xB6", # pilcrow sign = paragraph sign
|
---|
155 | "middot" => "\xB7", # middle dot = Georgian comma
|
---|
156 | "cedil" => "\xB8", # cedilla
|
---|
157 | "sup1" => "\xB9", # superscript 1
|
---|
158 | "ordm" => "\xBA", # masculine ordinal indicator
|
---|
159 | "frac14" => "\xBC", # vulgar fraction one quarter
|
---|
160 | "frac12" => "\xBD", # vulgar fraction one half
|
---|
161 | "frac34" => "\xBE", # vulgar fraction three quarters
|
---|
162 | "iquest" => "\xBF", # inverted question mark
|
---|
163 | "times" => "\xD7", # multiplication sign
|
---|
164 | "divide" => "\xF7", # division sign
|
---|
165 |
|
---|
166 | "nbsp" => "\x01", # non-breaking space
|
---|
167 | );
|
---|
168 |
|
---|
169 |
|
---|
170 | ##############################################################################
|
---|
171 | # Initialization
|
---|
172 | ##############################################################################
|
---|
173 |
|
---|
174 | # Initialize the object. Must be sure to call our parent initializer.
|
---|
175 | sub initialize {
|
---|
176 | my $self = shift;
|
---|
177 |
|
---|
178 | $$self{alt} = 0 unless defined $$self{alt};
|
---|
179 | $$self{indent} = 4 unless defined $$self{indent};
|
---|
180 | $$self{margin} = 0 unless defined $$self{margin};
|
---|
181 | $$self{loose} = 0 unless defined $$self{loose};
|
---|
182 | $$self{sentence} = 0 unless defined $$self{sentence};
|
---|
183 | $$self{width} = 76 unless defined $$self{width};
|
---|
184 |
|
---|
185 | # Figure out what quotes we'll be using for C<> text.
|
---|
186 | $$self{quotes} ||= '"';
|
---|
187 | if ($$self{quotes} eq 'none') {
|
---|
188 | $$self{LQUOTE} = $$self{RQUOTE} = '';
|
---|
189 | } elsif (length ($$self{quotes}) == 1) {
|
---|
190 | $$self{LQUOTE} = $$self{RQUOTE} = $$self{quotes};
|
---|
191 | } elsif ($$self{quotes} =~ /^(.)(.)$/
|
---|
192 | || $$self{quotes} =~ /^(..)(..)$/) {
|
---|
193 | $$self{LQUOTE} = $1;
|
---|
194 | $$self{RQUOTE} = $2;
|
---|
195 | } else {
|
---|
196 | croak qq(Invalid quote specification "$$self{quotes}");
|
---|
197 | }
|
---|
198 |
|
---|
199 | # Stack of indentations.
|
---|
200 | $$self{INDENTS} = [];
|
---|
201 |
|
---|
202 | # Current left margin.
|
---|
203 | $$self{MARGIN} = $$self{indent} + $$self{margin};
|
---|
204 |
|
---|
205 | $self->SUPER::initialize;
|
---|
206 |
|
---|
207 | # Tell Pod::Parser that we want the non-POD stuff too if code was set.
|
---|
208 | $self->parseopts ('-want_nonPODs' => 1) if $$self{code};
|
---|
209 | }
|
---|
210 |
|
---|
211 |
|
---|
212 | ##############################################################################
|
---|
213 | # Core overrides
|
---|
214 | ##############################################################################
|
---|
215 |
|
---|
216 | # Called for each command paragraph. Gets the command, the associated
|
---|
217 | # paragraph, the line number, and a Pod::Paragraph object. Just dispatches
|
---|
218 | # the command to a method named the same as the command. =cut is handled
|
---|
219 | # internally by Pod::Parser.
|
---|
220 | sub command {
|
---|
221 | my $self = shift;
|
---|
222 | my $command = shift;
|
---|
223 | return if $command eq 'pod';
|
---|
224 | return if ($$self{EXCLUDE} && $command ne 'end');
|
---|
225 | if ($self->can ('cmd_' . $command)) {
|
---|
226 | $command = 'cmd_' . $command;
|
---|
227 | $self->$command (@_);
|
---|
228 | } else {
|
---|
229 | my ($text, $line, $paragraph) = @_;
|
---|
230 | my $file;
|
---|
231 | ($file, $line) = $paragraph->file_line;
|
---|
232 | $text =~ s/\n+\z//;
|
---|
233 | $text = " $text" if ($text =~ /^\S/);
|
---|
234 | warn qq($file:$line: Unknown command paragraph: =$command$text\n);
|
---|
235 | return;
|
---|
236 | }
|
---|
237 | }
|
---|
238 |
|
---|
239 | # Called for a verbatim paragraph. Gets the paragraph, the line number, and a
|
---|
240 | # Pod::Paragraph object. Just output it verbatim, but with tabs converted to
|
---|
241 | # spaces.
|
---|
242 | sub verbatim {
|
---|
243 | my $self = shift;
|
---|
244 | return if $$self{EXCLUDE};
|
---|
245 | $self->item if defined $$self{ITEM};
|
---|
246 | local $_ = shift;
|
---|
247 | return if /^\s*$/;
|
---|
248 | s/^(\s*\S+)/(' ' x $$self{MARGIN}) . $1/gme;
|
---|
249 | $self->output ($_);
|
---|
250 | }
|
---|
251 |
|
---|
252 | # Called for a regular text block. Gets the paragraph, the line number, and a
|
---|
253 | # Pod::Paragraph object. Perform interpolation and output the results.
|
---|
254 | sub textblock {
|
---|
255 | my $self = shift;
|
---|
256 | return if $$self{EXCLUDE};
|
---|
257 | $self->output ($_[0]), return if $$self{VERBATIM};
|
---|
258 | local $_ = shift;
|
---|
259 | my $line = shift;
|
---|
260 |
|
---|
261 | # Interpolate and output the paragraph.
|
---|
262 | $_ = $self->interpolate ($_, $line);
|
---|
263 | s/\s+$/\n/;
|
---|
264 | if (defined $$self{ITEM}) {
|
---|
265 | $self->item ($_ . "\n");
|
---|
266 | } else {
|
---|
267 | $self->output ($self->reformat ($_ . "\n"));
|
---|
268 | }
|
---|
269 | }
|
---|
270 |
|
---|
271 | # Called for a formatting code. Gets the command, argument, and a
|
---|
272 | # Pod::InteriorSequence object and is expected to return the resulting text.
|
---|
273 | # Calls methods for code, bold, italic, file, and link to handle those types
|
---|
274 | # of codes, and handles S<>, E<>, X<>, and Z<> directly.
|
---|
275 | sub interior_sequence {
|
---|
276 | local $_;
|
---|
277 | my ($self, $command, $seq);
|
---|
278 | ($self, $command, $_, $seq) = @_;
|
---|
279 |
|
---|
280 | # We have to defer processing of the inside of an L<> formatting code. If
|
---|
281 | # this code is nested inside an L<> code, return the literal raw text of
|
---|
282 | # it.
|
---|
283 | my $parent = $seq->nested;
|
---|
284 | while (defined $parent) {
|
---|
285 | return $seq->raw_text if ($parent->cmd_name eq 'L');
|
---|
286 | $parent = $parent->nested;
|
---|
287 | }
|
---|
288 |
|
---|
289 | # Index entries are ignored in plain text.
|
---|
290 | return '' if ($command eq 'X' || $command eq 'Z');
|
---|
291 |
|
---|
292 | # Expand escapes into the actual character now, warning if invalid.
|
---|
293 | if ($command eq 'E') {
|
---|
294 | if (/^\d+$/) {
|
---|
295 | return chr;
|
---|
296 | } else {
|
---|
297 | return $ESCAPES{$_} if defined $ESCAPES{$_};
|
---|
298 | my ($file, $line) = $seq->file_line;
|
---|
299 | warn "$file:$line: Unknown escape: E<$_>\n";
|
---|
300 | return "E<$_>";
|
---|
301 | }
|
---|
302 | }
|
---|
303 |
|
---|
304 | # For all the other formatting codes, empty content produces no output.
|
---|
305 | return if $_ eq '';
|
---|
306 |
|
---|
307 | # For S<>, compress all internal whitespace and then map spaces to \01.
|
---|
308 | # When we output the text, we'll map this back.
|
---|
309 | if ($command eq 'S') {
|
---|
310 | s/\s+/ /g;
|
---|
311 | tr/ /\01/;
|
---|
312 | return $_;
|
---|
313 | }
|
---|
314 |
|
---|
315 | # Anything else needs to get dispatched to another method.
|
---|
316 | if ($command eq 'B') { return $self->seq_b ($_) }
|
---|
317 | elsif ($command eq 'C') { return $self->seq_c ($_) }
|
---|
318 | elsif ($command eq 'F') { return $self->seq_f ($_) }
|
---|
319 | elsif ($command eq 'I') { return $self->seq_i ($_) }
|
---|
320 | elsif ($command eq 'L') { return $self->seq_l ($_, $seq) }
|
---|
321 | else {
|
---|
322 | my ($file, $line) = $seq->file_line;
|
---|
323 | warn "$file:$line: Unknown formatting code: $command<$_>\n";
|
---|
324 | }
|
---|
325 | }
|
---|
326 |
|
---|
327 | # Called for each paragraph that's actually part of the POD. We take
|
---|
328 | # advantage of this opportunity to untabify the input. Also, if given the
|
---|
329 | # code option, we may see paragraphs that aren't part of the POD and need to
|
---|
330 | # output them directly.
|
---|
331 | sub preprocess_paragraph {
|
---|
332 | my $self = shift;
|
---|
333 | local $_ = shift;
|
---|
334 | 1 while s/^(.*?)(\t+)/$1 . ' ' x (length ($2) * 8 - length ($1) % 8)/me;
|
---|
335 | $self->output_code ($_) if $self->cutting;
|
---|
336 | $_;
|
---|
337 | }
|
---|
338 |
|
---|
339 |
|
---|
340 | ##############################################################################
|
---|
341 | # Command paragraphs
|
---|
342 | ##############################################################################
|
---|
343 |
|
---|
344 | # All command paragraphs take the paragraph and the line number.
|
---|
345 |
|
---|
346 | # First level heading.
|
---|
347 | sub cmd_head1 {
|
---|
348 | my ($self, $text, $line) = @_;
|
---|
349 | $self->heading ($text, $line, 0, '====');
|
---|
350 | }
|
---|
351 |
|
---|
352 | # Second level heading.
|
---|
353 | sub cmd_head2 {
|
---|
354 | my ($self, $text, $line) = @_;
|
---|
355 | $self->heading ($text, $line, $$self{indent} / 2, '== ');
|
---|
356 | }
|
---|
357 |
|
---|
358 | # Third level heading.
|
---|
359 | sub cmd_head3 {
|
---|
360 | my ($self, $text, $line) = @_;
|
---|
361 | $self->heading ($text, $line, $$self{indent} * 2 / 3 + 0.5, '= ');
|
---|
362 | }
|
---|
363 |
|
---|
364 | # Third level heading.
|
---|
365 | sub cmd_head4 {
|
---|
366 | my ($self, $text, $line) = @_;
|
---|
367 | $self->heading ($text, $line, $$self{indent} * 3 / 4 + 0.5, '- ');
|
---|
368 | }
|
---|
369 |
|
---|
370 | # Start a list.
|
---|
371 | sub cmd_over {
|
---|
372 | my $self = shift;
|
---|
373 | local $_ = shift;
|
---|
374 | $self->item ("\n\n") if defined $$self{ITEM};
|
---|
375 | unless (/^[-+]?\d+\s+$/) { $_ = $$self{indent} }
|
---|
376 | push (@{ $$self{INDENTS} }, $$self{MARGIN});
|
---|
377 | $$self{MARGIN} += ($_ + 0);
|
---|
378 | }
|
---|
379 |
|
---|
380 | # End a list.
|
---|
381 | sub cmd_back {
|
---|
382 | my ($self, $text, $line, $paragraph) = @_;
|
---|
383 | $self->item ("\n\n") if defined $$self{ITEM};
|
---|
384 | $$self{MARGIN} = pop @{ $$self{INDENTS} };
|
---|
385 | unless (defined $$self{MARGIN}) {
|
---|
386 | my $file;
|
---|
387 | ($file, $line) = $paragraph->file_line;
|
---|
388 | warn "$file:$line: Unmatched =back\n";
|
---|
389 | $$self{MARGIN} = $$self{indent};
|
---|
390 | }
|
---|
391 | }
|
---|
392 |
|
---|
393 | # An individual list item.
|
---|
394 | sub cmd_item {
|
---|
395 | my $self = shift;
|
---|
396 | if (defined $$self{ITEM}) { $self->item }
|
---|
397 | local $_ = shift;
|
---|
398 | s/\s+$//;
|
---|
399 | $$self{ITEM} = $_ ? $self->interpolate ($_) : '*';
|
---|
400 | }
|
---|
401 |
|
---|
402 | # Begin a block for a particular translator. Setting VERBATIM triggers
|
---|
403 | # special handling in textblock().
|
---|
404 | sub cmd_begin {
|
---|
405 | my $self = shift;
|
---|
406 | local $_ = shift;
|
---|
407 | my ($kind) = /^(\S+)/ or return;
|
---|
408 | if ($kind eq 'text') {
|
---|
409 | $$self{VERBATIM} = 1;
|
---|
410 | } else {
|
---|
411 | $$self{EXCLUDE} = 1;
|
---|
412 | }
|
---|
413 | }
|
---|
414 |
|
---|
415 | # End a block for a particular translator. We assume that all =begin/=end
|
---|
416 | # pairs are properly closed.
|
---|
417 | sub cmd_end {
|
---|
418 | my $self = shift;
|
---|
419 | $$self{EXCLUDE} = 0;
|
---|
420 | $$self{VERBATIM} = 0;
|
---|
421 | }
|
---|
422 |
|
---|
423 | # One paragraph for a particular translator. Ignore it unless it's intended
|
---|
424 | # for text, in which case we treat it as a verbatim text block.
|
---|
425 | sub cmd_for {
|
---|
426 | my $self = shift;
|
---|
427 | local $_ = shift;
|
---|
428 | my $line = shift;
|
---|
429 | return unless s/^text\b[ \t]*\n?//;
|
---|
430 | $self->verbatim ($_, $line);
|
---|
431 | }
|
---|
432 |
|
---|
433 |
|
---|
434 | ##############################################################################
|
---|
435 | # Formatting codes
|
---|
436 | ##############################################################################
|
---|
437 |
|
---|
438 | # The simple ones. These are here mostly so that subclasses can override them
|
---|
439 | # and do more complicated things.
|
---|
440 | sub seq_b { return $_[0]{alt} ? "``$_[1]''" : $_[1] }
|
---|
441 | sub seq_f { return $_[0]{alt} ? "\"$_[1]\"" : $_[1] }
|
---|
442 | sub seq_i { return '*' . $_[1] . '*' }
|
---|
443 |
|
---|
444 | # Apply a whole bunch of messy heuristics to not quote things that don't
|
---|
445 | # benefit from being quoted. These originally come from Barrie Slaymaker and
|
---|
446 | # largely duplicate code in Pod::Man.
|
---|
447 | sub seq_c {
|
---|
448 | my $self = shift;
|
---|
449 | local $_ = shift;
|
---|
450 |
|
---|
451 | # A regex that matches the portion of a variable reference that's the
|
---|
452 | # array or hash index, separated out just because we want to use it in
|
---|
453 | # several places in the following regex.
|
---|
454 | my $index = '(?: \[.*\] | \{.*\} )?';
|
---|
455 |
|
---|
456 | # Check for things that we don't want to quote, and if we find any of
|
---|
457 | # them, return the string with just a font change and no quoting.
|
---|
458 | m{
|
---|
459 | ^\s*
|
---|
460 | (?:
|
---|
461 | ( [\'\`\"] ) .* \1 # already quoted
|
---|
462 | | \` .* \' # `quoted'
|
---|
463 | | \$+ [\#^]? \S $index # special ($^Foo, $")
|
---|
464 | | [\$\@%&*]+ \#? [:\'\w]+ $index # plain var or func
|
---|
465 | | [\$\@%&*]* [:\'\w]+ (?: -> )? \(\s*[^\s,]\s*\) # 0/1-arg func call
|
---|
466 | | [+-]? ( \d[\d.]* | \.\d+ ) (?: [eE][+-]?\d+ )? # a number
|
---|
467 | | 0x [a-fA-F\d]+ # a hex constant
|
---|
468 | )
|
---|
469 | \s*\z
|
---|
470 | }xo && return $_;
|
---|
471 |
|
---|
472 | # If we didn't return, go ahead and quote the text.
|
---|
473 | return $$self{alt} ? "``$_''" : "$$self{LQUOTE}$_$$self{RQUOTE}";
|
---|
474 | }
|
---|
475 |
|
---|
476 | # Handle links. Since this is plain text, we can't actually make any real
|
---|
477 | # links, so this is all to figure out what text we print out. Most of the
|
---|
478 | # work is done by Pod::ParseLink.
|
---|
479 | sub seq_l {
|
---|
480 | my ($self, $link, $seq) = @_;
|
---|
481 | my ($text, $type) = (parselink ($link))[1,4];
|
---|
482 | my ($file, $line) = $seq->file_line;
|
---|
483 | $text = $self->interpolate ($text, $line);
|
---|
484 | $text = '<' . $text . '>' if $type eq 'url';
|
---|
485 | return $text || '';
|
---|
486 | }
|
---|
487 |
|
---|
488 |
|
---|
489 | ##############################################################################
|
---|
490 | # Header handling
|
---|
491 | ##############################################################################
|
---|
492 |
|
---|
493 | # The common code for handling all headers. Takes the interpolated header
|
---|
494 | # text, the line number, the indentation, and the surrounding marker for the
|
---|
495 | # alt formatting method.
|
---|
496 | sub heading {
|
---|
497 | my ($self, $text, $line, $indent, $marker) = @_;
|
---|
498 | $self->item ("\n\n") if defined $$self{ITEM};
|
---|
499 | $text =~ s/\s+$//;
|
---|
500 | $text = $self->interpolate ($text, $line);
|
---|
501 | if ($$self{alt}) {
|
---|
502 | my $closemark = reverse (split (//, $marker));
|
---|
503 | my $margin = ' ' x $$self{margin};
|
---|
504 | $self->output ("\n" . "$margin$marker $text $closemark" . "\n\n");
|
---|
505 | } else {
|
---|
506 | $text .= "\n" if $$self{loose};
|
---|
507 | my $margin = ' ' x ($$self{margin} + $indent);
|
---|
508 | $self->output ($margin . $text . "\n");
|
---|
509 | }
|
---|
510 | }
|
---|
511 |
|
---|
512 |
|
---|
513 | ##############################################################################
|
---|
514 | # List handling
|
---|
515 | ##############################################################################
|
---|
516 |
|
---|
517 | # This method is called whenever an =item command is complete (in other words,
|
---|
518 | # we've seen its associated paragraph or know for certain that it doesn't have
|
---|
519 | # one). It gets the paragraph associated with the item as an argument. If
|
---|
520 | # that argument is empty, just output the item tag; if it contains a newline,
|
---|
521 | # output the item tag followed by the newline. Otherwise, see if there's
|
---|
522 | # enough room for us to output the item tag in the margin of the text or if we
|
---|
523 | # have to put it on a separate line.
|
---|
524 | sub item {
|
---|
525 | my $self = shift;
|
---|
526 | local $_ = shift;
|
---|
527 | my $tag = $$self{ITEM};
|
---|
528 | unless (defined $tag) {
|
---|
529 | carp "Item called without tag";
|
---|
530 | return;
|
---|
531 | }
|
---|
532 | undef $$self{ITEM};
|
---|
533 | my $indent = $$self{INDENTS}[-1];
|
---|
534 | unless (defined $indent) { $indent = $$self{indent} }
|
---|
535 | my $margin = ' ' x $$self{margin};
|
---|
536 | if (!$_ || /^\s+$/ || ($$self{MARGIN} - $indent < length ($tag) + 1)) {
|
---|
537 | my $realindent = $$self{MARGIN};
|
---|
538 | $$self{MARGIN} = $indent;
|
---|
539 | my $output = $self->reformat ($tag);
|
---|
540 | $output =~ s/^$margin /$margin:/ if ($$self{alt} && $indent > 0);
|
---|
541 | $output =~ s/\n*$/\n/;
|
---|
542 |
|
---|
543 | # If the text is just whitespace, we have an empty item paragraph;
|
---|
544 | # this can result from =over/=item/=back without any intermixed
|
---|
545 | # paragraphs. Insert some whitespace to keep the =item from merging
|
---|
546 | # into the next paragraph.
|
---|
547 | $output .= "\n" if $_ && $_ =~ /^\s*$/;
|
---|
548 |
|
---|
549 | $self->output ($output);
|
---|
550 | $$self{MARGIN} = $realindent;
|
---|
551 | $self->output ($self->reformat ($_)) if $_ && /\S/;
|
---|
552 | } else {
|
---|
553 | my $space = ' ' x $indent;
|
---|
554 | $space =~ s/^$margin /$margin:/ if $$self{alt};
|
---|
555 | $_ = $self->reformat ($_);
|
---|
556 | s/^$margin /$margin:/ if ($$self{alt} && $indent > 0);
|
---|
557 | my $tagspace = ' ' x length $tag;
|
---|
558 | s/^($space)$tagspace/$1$tag/ or warn "Bizarre space in item";
|
---|
559 | $self->output ($_);
|
---|
560 | }
|
---|
561 | }
|
---|
562 |
|
---|
563 |
|
---|
564 | ##############################################################################
|
---|
565 | # Output formatting
|
---|
566 | ##############################################################################
|
---|
567 |
|
---|
568 | # Wrap a line, indenting by the current left margin. We can't use Text::Wrap
|
---|
569 | # because it plays games with tabs. We can't use formline, even though we'd
|
---|
570 | # really like to, because it screws up non-printing characters. So we have to
|
---|
571 | # do the wrapping ourselves.
|
---|
572 | sub wrap {
|
---|
573 | my $self = shift;
|
---|
574 | local $_ = shift;
|
---|
575 | my $output = '';
|
---|
576 | my $spaces = ' ' x $$self{MARGIN};
|
---|
577 | my $width = $$self{width} - $$self{MARGIN};
|
---|
578 | while (length > $width) {
|
---|
579 | if (s/^([^\n]{0,$width})\s+// || s/^([^\n]{$width})//) {
|
---|
580 | $output .= $spaces . $1 . "\n";
|
---|
581 | } else {
|
---|
582 | last;
|
---|
583 | }
|
---|
584 | }
|
---|
585 | $output .= $spaces . $_;
|
---|
586 | $output =~ s/\s+$/\n\n/;
|
---|
587 | $output;
|
---|
588 | }
|
---|
589 |
|
---|
590 | # Reformat a paragraph of text for the current margin. Takes the text to
|
---|
591 | # reformat and returns the formatted text.
|
---|
592 | sub reformat {
|
---|
593 | my $self = shift;
|
---|
594 | local $_ = shift;
|
---|
595 |
|
---|
596 | # If we're trying to preserve two spaces after sentences, do some munging
|
---|
597 | # to support that. Otherwise, smash all repeated whitespace.
|
---|
598 | if ($$self{sentence}) {
|
---|
599 | s/ +$//mg;
|
---|
600 | s/\.\n/. \n/g;
|
---|
601 | s/\n/ /g;
|
---|
602 | s/ +/ /g;
|
---|
603 | } else {
|
---|
604 | s/\s+/ /g;
|
---|
605 | }
|
---|
606 | $self->wrap ($_);
|
---|
607 | }
|
---|
608 |
|
---|
609 | # Output text to the output device.
|
---|
610 | sub output { $_[1] =~ tr/\01/ /; print { $_[0]->output_handle } $_[1] }
|
---|
611 |
|
---|
612 | # Output a block of code (something that isn't part of the POD text). Called
|
---|
613 | # by preprocess_paragraph only if we were given the code option. Exists here
|
---|
614 | # only so that it can be overridden by subclasses.
|
---|
615 | sub output_code { $_[0]->output ($_[1]) }
|
---|
616 |
|
---|
617 |
|
---|
618 | ##############################################################################
|
---|
619 | # Backwards compatibility
|
---|
620 | ##############################################################################
|
---|
621 |
|
---|
622 | # The old Pod::Text module did everything in a pod2text() function. This
|
---|
623 | # tries to provide the same interface for legacy applications.
|
---|
624 | sub pod2text {
|
---|
625 | my @args;
|
---|
626 |
|
---|
627 | # This is really ugly; I hate doing option parsing in the middle of a
|
---|
628 | # module. But the old Pod::Text module supported passing flags to its
|
---|
629 | # entry function, so handle -a and -<number>.
|
---|
630 | while ($_[0] =~ /^-/) {
|
---|
631 | my $flag = shift;
|
---|
632 | if ($flag eq '-a') { push (@args, alt => 1) }
|
---|
633 | elsif ($flag =~ /^-(\d+)$/) { push (@args, width => $1) }
|
---|
634 | else {
|
---|
635 | unshift (@_, $flag);
|
---|
636 | last;
|
---|
637 | }
|
---|
638 | }
|
---|
639 |
|
---|
640 | # Now that we know what arguments we're using, create the parser.
|
---|
641 | my $parser = Pod::Text->new (@args);
|
---|
642 |
|
---|
643 | # If two arguments were given, the second argument is going to be a file
|
---|
644 | # handle. That means we want to call parse_from_filehandle(), which means
|
---|
645 | # we need to turn the first argument into a file handle. Magic open will
|
---|
646 | # handle the <&STDIN case automagically.
|
---|
647 | if (defined $_[1]) {
|
---|
648 | my @fhs = @_;
|
---|
649 | local *IN;
|
---|
650 | unless (open (IN, $fhs[0])) {
|
---|
651 | croak ("Can't open $fhs[0] for reading: $!\n");
|
---|
652 | return;
|
---|
653 | }
|
---|
654 | $fhs[0] = \*IN;
|
---|
655 | return $parser->parse_from_filehandle (@fhs);
|
---|
656 | } else {
|
---|
657 | return $parser->parse_from_file (@_);
|
---|
658 | }
|
---|
659 | }
|
---|
660 |
|
---|
661 |
|
---|
662 | ##############################################################################
|
---|
663 | # Module return value and documentation
|
---|
664 | ##############################################################################
|
---|
665 |
|
---|
666 | 1;
|
---|
667 | __END__
|
---|
668 |
|
---|
669 | =head1 NAME
|
---|
670 |
|
---|
671 | Pod::Text - Convert POD data to formatted ASCII text
|
---|
672 |
|
---|
673 | =head1 SYNOPSIS
|
---|
674 |
|
---|
675 | use Pod::Text;
|
---|
676 | my $parser = Pod::Text->new (sentence => 0, width => 78);
|
---|
677 |
|
---|
678 | # Read POD from STDIN and write to STDOUT.
|
---|
679 | $parser->parse_from_filehandle;
|
---|
680 |
|
---|
681 | # Read POD from file.pod and write to file.txt.
|
---|
682 | $parser->parse_from_file ('file.pod', 'file.txt');
|
---|
683 |
|
---|
684 | =head1 DESCRIPTION
|
---|
685 |
|
---|
686 | Pod::Text is a module that can convert documentation in the POD format (the
|
---|
687 | preferred language for documenting Perl) into formatted ASCII. It uses no
|
---|
688 | special formatting controls or codes whatsoever, and its output is therefore
|
---|
689 | suitable for nearly any device.
|
---|
690 |
|
---|
691 | As a derived class from Pod::Parser, Pod::Text supports the same methods and
|
---|
692 | interfaces. See L<Pod::Parser> for all the details; briefly, one creates a
|
---|
693 | new parser with C<< Pod::Text->new() >> and then calls either
|
---|
694 | parse_from_filehandle() or parse_from_file().
|
---|
695 |
|
---|
696 | new() can take options, in the form of key/value pairs, that control the
|
---|
697 | behavior of the parser. The currently recognized options are:
|
---|
698 |
|
---|
699 | =over 4
|
---|
700 |
|
---|
701 | =item alt
|
---|
702 |
|
---|
703 | If set to a true value, selects an alternate output format that, among other
|
---|
704 | things, uses a different heading style and marks C<=item> entries with a
|
---|
705 | colon in the left margin. Defaults to false.
|
---|
706 |
|
---|
707 | =item code
|
---|
708 |
|
---|
709 | If set to a true value, the non-POD parts of the input file will be included
|
---|
710 | in the output. Useful for viewing code documented with POD blocks with the
|
---|
711 | POD rendered and the code left intact.
|
---|
712 |
|
---|
713 | =item indent
|
---|
714 |
|
---|
715 | The number of spaces to indent regular text, and the default indentation for
|
---|
716 | C<=over> blocks. Defaults to 4.
|
---|
717 |
|
---|
718 | =item loose
|
---|
719 |
|
---|
720 | If set to a true value, a blank line is printed after a C<=head1> heading.
|
---|
721 | If set to false (the default), no blank line is printed after C<=head1>,
|
---|
722 | although one is still printed after C<=head2>. This is the default because
|
---|
723 | it's the expected formatting for manual pages; if you're formatting
|
---|
724 | arbitrary text documents, setting this to true may result in more pleasing
|
---|
725 | output.
|
---|
726 |
|
---|
727 | =item margin
|
---|
728 |
|
---|
729 | The width of the left margin in spaces. Defaults to 0. This is the margin
|
---|
730 | for all text, including headings, not the amount by which regular text is
|
---|
731 | indented; for the latter, see the I<indent> option. To set the right
|
---|
732 | margin, see the I<width> option.
|
---|
733 |
|
---|
734 | =item quotes
|
---|
735 |
|
---|
736 | Sets the quote marks used to surround CE<lt>> text. If the value is a
|
---|
737 | single character, it is used as both the left and right quote; if it is two
|
---|
738 | characters, the first character is used as the left quote and the second as
|
---|
739 | the right quoted; and if it is four characters, the first two are used as
|
---|
740 | the left quote and the second two as the right quote.
|
---|
741 |
|
---|
742 | This may also be set to the special value C<none>, in which case no quote
|
---|
743 | marks are added around CE<lt>> text.
|
---|
744 |
|
---|
745 | =item sentence
|
---|
746 |
|
---|
747 | If set to a true value, Pod::Text will assume that each sentence ends in two
|
---|
748 | spaces, and will try to preserve that spacing. If set to false, all
|
---|
749 | consecutive whitespace in non-verbatim paragraphs is compressed into a
|
---|
750 | single space. Defaults to true.
|
---|
751 |
|
---|
752 | =item width
|
---|
753 |
|
---|
754 | The column at which to wrap text on the right-hand side. Defaults to 76.
|
---|
755 |
|
---|
756 | =back
|
---|
757 |
|
---|
758 | The standard Pod::Parser method parse_from_filehandle() takes up to two
|
---|
759 | arguments, the first being the file handle to read POD from and the second
|
---|
760 | being the file handle to write the formatted output to. The first defaults
|
---|
761 | to STDIN if not given, and the second defaults to STDOUT. The method
|
---|
762 | parse_from_file() is almost identical, except that its two arguments are the
|
---|
763 | input and output disk files instead. See L<Pod::Parser> for the specific
|
---|
764 | details.
|
---|
765 |
|
---|
766 | =head1 DIAGNOSTICS
|
---|
767 |
|
---|
768 | =over 4
|
---|
769 |
|
---|
770 | =item Bizarre space in item
|
---|
771 |
|
---|
772 | =item Item called without tag
|
---|
773 |
|
---|
774 | (W) Something has gone wrong in internal C<=item> processing. These
|
---|
775 | messages indicate a bug in Pod::Text; you should never see them.
|
---|
776 |
|
---|
777 | =item Can't open %s for reading: %s
|
---|
778 |
|
---|
779 | (F) Pod::Text was invoked via the compatibility mode pod2text() interface
|
---|
780 | and the input file it was given could not be opened.
|
---|
781 |
|
---|
782 | =item Invalid quote specification "%s"
|
---|
783 |
|
---|
784 | (F) The quote specification given (the quotes option to the constructor) was
|
---|
785 | invalid. A quote specification must be one, two, or four characters long.
|
---|
786 |
|
---|
787 | =item %s:%d: Unknown command paragraph: %s
|
---|
788 |
|
---|
789 | (W) The POD source contained a non-standard command paragraph (something of
|
---|
790 | the form C<=command args>) that Pod::Man didn't know about. It was ignored.
|
---|
791 |
|
---|
792 | =item %s:%d: Unknown escape: %s
|
---|
793 |
|
---|
794 | (W) The POD source contained an C<EE<lt>E<gt>> escape that Pod::Text didn't
|
---|
795 | know about.
|
---|
796 |
|
---|
797 | =item %s:%d: Unknown formatting code: %s
|
---|
798 |
|
---|
799 | (W) The POD source contained a non-standard formatting code (something of
|
---|
800 | the form C<XE<lt>E<gt>>) that Pod::Text didn't know about.
|
---|
801 |
|
---|
802 | =item %s:%d: Unmatched =back
|
---|
803 |
|
---|
804 | (W) Pod::Text encountered a C<=back> command that didn't correspond to an
|
---|
805 | C<=over> command.
|
---|
806 |
|
---|
807 | =back
|
---|
808 |
|
---|
809 | =head1 RESTRICTIONS
|
---|
810 |
|
---|
811 | Embedded Ctrl-As (octal 001) in the input will be mapped to spaces on
|
---|
812 | output, due to an internal implementation detail.
|
---|
813 |
|
---|
814 | =head1 NOTES
|
---|
815 |
|
---|
816 | This is a replacement for an earlier Pod::Text module written by Tom
|
---|
817 | Christiansen. It has a revamped interface, since it now uses Pod::Parser,
|
---|
818 | but an interface roughly compatible with the old Pod::Text::pod2text()
|
---|
819 | function is still available. Please change to the new calling convention,
|
---|
820 | though.
|
---|
821 |
|
---|
822 | The original Pod::Text contained code to do formatting via termcap
|
---|
823 | sequences, although it wasn't turned on by default and it was problematic to
|
---|
824 | get it to work at all. This rewrite doesn't even try to do that, but a
|
---|
825 | subclass of it does. Look for L<Pod::Text::Termcap>.
|
---|
826 |
|
---|
827 | =head1 SEE ALSO
|
---|
828 |
|
---|
829 | L<Pod::Parser>, L<Pod::Text::Termcap>, L<pod2text(1)>
|
---|
830 |
|
---|
831 | The current version of this module is always available from its web site at
|
---|
832 | L<http://www.eyrie.org/~eagle/software/podlators/>. It is also part of the
|
---|
833 | Perl core distribution as of 5.6.0.
|
---|
834 |
|
---|
835 | =head1 AUTHOR
|
---|
836 |
|
---|
837 | Russ Allbery <[email protected]>, based I<very> heavily on the original
|
---|
838 | Pod::Text by Tom Christiansen <[email protected]> and its conversion to
|
---|
839 | Pod::Parser by Brad Appleton <[email protected]>.
|
---|
840 |
|
---|
841 | =head1 COPYRIGHT AND LICENSE
|
---|
842 |
|
---|
843 | Copyright 1999, 2000, 2001, 2002 by Russ Allbery <[email protected]>.
|
---|
844 |
|
---|
845 | This program is free software; you may redistribute it and/or modify it
|
---|
846 | under the same terms as Perl itself.
|
---|
847 |
|
---|
848 | =cut
|
---|