Commit | Line | Data |
69e00e79 |
1 | package Pod::Text; |
2 | |
69e00e79 |
3 | =head1 NAME |
4 | |
5 | Pod::Text - convert POD data to formatted ASCII text |
6 | |
7 | =head1 SYNOPSIS |
8 | |
9 | use Pod::Text; |
10 | |
11 | pod2text("perlfunc.pod"); |
12 | |
13 | Also: |
14 | |
15 | pod2text < input.pod |
16 | |
17 | =head1 DESCRIPTION |
18 | |
19 | Pod::Text is a module that can convert documentation in the POD format (such |
20 | as can be found throughout the Perl distribution) into formatted ASCII. |
21 | Termcap is optionally supported for boldface/underline, and can enabled via |
22 | C<$Pod::Text::termcap=1>. If termcap has not been enabled, then backspaces |
23 | will be used to simulate bold and underlined text. |
24 | |
25 | A separate F<pod2text> program is included that is primarily a wrapper for |
26 | Pod::Text. |
27 | |
28 | The single function C<pod2text()> can take one or two arguments. The first |
1fef88e7 |
29 | should be the name of a file to read the pod from, or "E<lt>&STDIN" to read from |
69e00e79 |
30 | STDIN. A second argument, if provided, should be a filehandle glob where |
31 | output should be sent. |
32 | |
33 | =head1 AUTHOR |
34 | |
1fef88e7 |
35 | Tom Christiansen E<lt>F<tchrist@mox.perl.com>E<gt> |
69e00e79 |
36 | |
37 | =head1 TODO |
38 | |
39 | Cleanup work. The input and output locations need to be more flexible, |
40 | termcap shouldn't be a global variable, and the terminal speed needs to |
41 | be properly calculated. |
42 | |
43 | =cut |
44 | |
45 | use Term::Cap; |
46 | require Exporter; |
47 | @ISA = Exporter; |
48 | @EXPORT = qw(pod2text); |
49 | |
f02a87df |
50 | use vars qw($VERSION); |
51 | $VERSION = "1.0201"; |
52 | |
69e00e79 |
53 | $termcap=0; |
54 | |
55 | #$use_format=1; |
56 | |
57 | $UNDL = "\x1b[4m"; |
58 | $INV = "\x1b[7m"; |
59 | $BOLD = "\x1b[1m"; |
60 | $NORM = "\x1b[0m"; |
61 | |
62 | sub pod2text { |
63 | local($file,*OUTPUT) = @_; |
64 | *OUTPUT = *STDOUT if @_<2; |
65 | |
66 | if($termcap and !$setuptermcap) { |
67 | $setuptermcap=1; |
68 | |
69 | my($term) = Tgetent Term::Cap { TERM => undef, OSPEED => 9600 }; |
70 | $UNDL = $term->{'_us'}; |
71 | $INV = $term->{'_mr'}; |
72 | $BOLD = $term->{'_md'}; |
73 | $NORM = $term->{'_me'}; |
74 | } |
75 | |
76 | $SCREEN = ($_[0] =~ /^-(\d+)/ && (shift, $1)) |
69e00e79 |
77 | || $ENV{COLUMNS} |
36477c24 |
78 | || ($ENV{TERMCAP} =~ /co#(\d+)/)[0] |
69e00e79 |
79 | || (`stty -a 2>/dev/null` =~ /(\d+) columns/)[0] |
80 | || 72; |
81 | |
82 | $/ = ""; |
83 | |
84 | $FANCY = 0; |
85 | |
86 | $cutting = 1; |
87 | $DEF_INDENT = 4; |
88 | $indent = $DEF_INDENT; |
89 | $needspace = 0; |
8c634b6e |
90 | $begun = ""; |
69e00e79 |
91 | |
2e917f57 |
92 | open(IN, $file) || die "Couldn't open $file: $!"; |
69e00e79 |
93 | |
94 | POD_DIRECTIVE: while (<IN>) { |
95 | if ($cutting) { |
96 | next unless /^=/; |
97 | $cutting = 0; |
98 | } |
8c634b6e |
99 | if ($begun) { |
100 | if (/^=end\s+$begun/) { |
101 | $begun = ""; |
102 | } |
103 | elsif ($begun eq "text") { |
104 | print STDOUT $_; |
105 | } |
106 | next; |
107 | } |
69e00e79 |
108 | 1 while s{^(.*?)(\t+)(.*)$}{ |
109 | $1 |
110 | . (' ' x (length($2) * 8 - length($1) % 8)) |
111 | . $3 |
112 | }me; |
113 | # Translate verbatim paragraph |
114 | if (/^\s/) { |
115 | $needspace = 1; |
116 | output($_); |
117 | next; |
118 | } |
119 | |
f02a87df |
120 | if (/^=for\s+(\S+)\s*(.*)/s) { |
8c634b6e |
121 | if ($1 eq "text") { |
f02a87df |
122 | print STDOUT $2,""; |
8c634b6e |
123 | } else { |
124 | # ignore unknown for |
125 | } |
126 | next; |
127 | } |
f02a87df |
128 | elsif (/^=begin\s+(\S+)\s*(.*)/s) { |
8c634b6e |
129 | $begun = $1; |
130 | if ($1 eq "text") { |
f02a87df |
131 | print STDOUT $2.""; |
8c634b6e |
132 | } |
133 | next; |
134 | } |
135 | |
69e00e79 |
136 | sub prepare_for_output { |
137 | |
138 | s/\s*$/\n/; |
139 | &init_noremap; |
140 | |
141 | # need to hide E<> first; they're processed in clear_noremap |
142 | s/(E<[^<>]+>)/noremap($1)/ge; |
143 | $maxnest = 10; |
144 | while ($maxnest-- && /[A-Z]</) { |
145 | unless ($FANCY) { |
55497cff |
146 | s/C<(.*?)>/`$1'/sg; |
69e00e79 |
147 | } else { |
55497cff |
148 | s/C<(.*?)>/noremap("E<lchevron>${1}E<rchevron>")/sge; |
69e00e79 |
149 | } |
150 | # s/[IF]<(.*?)>/italic($1)/ge; |
55497cff |
151 | s/I<(.*?)>/*$1*/sg; |
69e00e79 |
152 | # s/[CB]<(.*?)>/bold($1)/ge; |
55497cff |
153 | s/X<.*?>//sg; |
69e00e79 |
154 | # LREF: a manpage(3f) |
155 | s:L<([a-zA-Z][^\s\/]+)(\([^\)]+\))?>:the $1$2 manpage:g; |
156 | # LREF: an =item on another manpage |
157 | s{ |
158 | L< |
159 | ([^/]+) |
160 | / |
161 | ( |
162 | [:\w]+ |
163 | (\(\))? |
164 | ) |
165 | > |
166 | } {the "$2" entry in the $1 manpage}gx; |
167 | |
168 | # LREF: an =item on this manpage |
169 | s{ |
170 | ((?: |
171 | L< |
172 | / |
173 | ( |
174 | [:\w]+ |
175 | (\(\))? |
176 | ) |
177 | > |
178 | (,?\s+(and\s+)?)? |
179 | )+) |
180 | } { internal_lrefs($1) }gex; |
181 | |
182 | # LREF: a =head2 (head1?), maybe on a manpage, maybe right here |
183 | # the "func" can disambiguate |
184 | s{ |
185 | L< |
186 | (?: |
187 | ([a-zA-Z]\S+?) / |
188 | )? |
189 | "?(.*?)"? |
190 | > |
191 | }{ |
192 | do { |
193 | $1 # if no $1, assume it means on this page. |
194 | ? "the section on \"$2\" in the $1 manpage" |
195 | : "the section on \"$2\"" |
196 | } |
55497cff |
197 | }sgex; |
69e00e79 |
198 | |
55497cff |
199 | s/[A-Z]<(.*?)>/$1/sg; |
69e00e79 |
200 | } |
201 | clear_noremap(1); |
202 | } |
203 | |
204 | &prepare_for_output; |
205 | |
206 | if (s/^=//) { |
207 | # $needspace = 0; # Assume this. |
208 | # s/\n/ /g; |
209 | ($Cmd, $_) = split(' ', $_, 2); |
210 | # clear_noremap(1); |
211 | if ($Cmd eq 'cut') { |
212 | $cutting = 1; |
213 | } |
214 | elsif ($Cmd eq 'head1') { |
215 | makespace(); |
216 | print OUTPUT; |
217 | # print OUTPUT uc($_); |
218 | } |
219 | elsif ($Cmd eq 'head2') { |
220 | makespace(); |
221 | # s/(\w+)/\u\L$1/g; |
222 | #print ' ' x $DEF_INDENT, $_; |
223 | # print "\xA7"; |
224 | s/(\w)/\xA7 $1/ if $FANCY; |
225 | print OUTPUT ' ' x ($DEF_INDENT/2), $_, "\n"; |
226 | } |
227 | elsif ($Cmd eq 'over') { |
228 | push(@indent,$indent); |
229 | $indent += ($_ + 0) || $DEF_INDENT; |
230 | } |
231 | elsif ($Cmd eq 'back') { |
232 | $indent = pop(@indent); |
233 | warn "Unmatched =back\n" unless defined $indent; |
234 | $needspace = 1; |
235 | } |
236 | elsif ($Cmd eq 'item') { |
237 | makespace(); |
238 | # s/\A(\s*)\*/$1\xb7/ if $FANCY; |
239 | # s/^(\s*\*\s+)/$1 /; |
240 | { |
241 | if (length() + 3 < $indent) { |
242 | my $paratag = $_; |
2e917f57 |
243 | $_ = <IN>; |
69e00e79 |
244 | if (/^=/) { # tricked! |
245 | local($indent) = $indent[$#index - 1] || $DEF_INDENT; |
246 | output($paratag); |
247 | redo POD_DIRECTIVE; |
248 | } |
249 | &prepare_for_output; |
250 | IP_output($paratag, $_); |
251 | } else { |
252 | local($indent) = $indent[$#index - 1] || $DEF_INDENT; |
253 | output($_); |
254 | } |
255 | } |
256 | } |
257 | else { |
258 | warn "Unrecognized directive: $Cmd\n"; |
259 | } |
260 | } |
261 | else { |
262 | # clear_noremap(1); |
263 | makespace(); |
264 | output($_, 1); |
265 | } |
266 | } |
267 | |
268 | close(IN); |
269 | |
270 | } |
271 | |
272 | ######################################################################### |
273 | |
274 | sub makespace { |
275 | if ($needspace) { |
276 | print OUTPUT "\n"; |
277 | $needspace = 0; |
278 | } |
279 | } |
280 | |
281 | sub bold { |
282 | my $line = shift; |
283 | return $line if $use_format; |
284 | if($termcap) { |
285 | $line = "$BOLD$line$NORM"; |
286 | } else { |
287 | $line =~ s/(.)/$1\b$1/g; |
288 | } |
289 | # $line = "$BOLD$line$NORM" if $ansify; |
290 | return $line; |
291 | } |
292 | |
293 | sub italic { |
294 | my $line = shift; |
295 | return $line if $use_format; |
296 | if($termcap) { |
297 | $line = "$UNDL$line$NORM"; |
298 | } else { |
299 | $line =~ s/(.)/$1\b_/g; |
300 | } |
301 | # $line = "$UNDL$line$NORM" if $ansify; |
302 | return $line; |
303 | } |
304 | |
305 | # Fill a paragraph including underlined and overstricken chars. |
306 | # It's not perfect for words longer than the margin, and it's probably |
307 | # slow, but it works. |
308 | sub fill { |
309 | local $_ = shift; |
310 | my $par = ""; |
311 | my $indent_space = " " x $indent; |
312 | my $marg = $SCREEN-$indent; |
313 | my $line = $indent_space; |
314 | my $line_length; |
315 | foreach (split) { |
316 | my $word_length = length; |
317 | $word_length -= 2 while /\010/g; # Subtract backspaces |
318 | |
319 | if ($line_length + $word_length > $marg) { |
320 | $par .= $line . "\n"; |
321 | $line= $indent_space . $_; |
322 | $line_length = $word_length; |
323 | } |
324 | else { |
325 | if ($line_length) { |
326 | $line_length++; |
327 | $line .= " "; |
328 | } |
329 | $line_length += $word_length; |
330 | $line .= $_; |
331 | } |
332 | } |
333 | $par .= "$line\n" if $line; |
334 | $par .= "\n"; |
335 | return $par; |
336 | } |
337 | |
338 | sub IP_output { |
339 | local($tag, $_) = @_; |
340 | local($tag_indent) = $indent[$#index - 1] || $DEF_INDENT; |
341 | $tag_cols = $SCREEN - $tag_indent; |
342 | $cols = $SCREEN - $indent; |
343 | $tag =~ s/\s*$//; |
344 | s/\s+/ /g; |
345 | s/^ //; |
346 | $str = "format OUTPUT = \n" |
347 | . (" " x ($tag_indent)) |
348 | . '@' . ('<' x ($indent - $tag_indent - 1)) |
349 | . "^" . ("<" x ($cols - 1)) . "\n" |
350 | . '$tag, $_' |
351 | . "\n~~" |
352 | . (" " x ($indent-2)) |
353 | . "^" . ("<" x ($cols - 5)) . "\n" |
354 | . '$_' . "\n\n.\n1"; |
355 | #warn $str; warn "tag is $tag, _ is $_"; |
356 | eval $str || die; |
357 | write OUTPUT; |
358 | } |
359 | |
360 | sub output { |
361 | local($_, $reformat) = @_; |
362 | if ($reformat) { |
363 | $cols = $SCREEN - $indent; |
364 | s/\s+/ /g; |
365 | s/^ //; |
366 | $str = "format OUTPUT = \n~~" |
367 | . (" " x ($indent-2)) |
368 | . "^" . ("<" x ($cols - 5)) . "\n" |
369 | . '$_' . "\n\n.\n1"; |
370 | eval $str || die; |
371 | write OUTPUT; |
372 | } else { |
373 | s/^/' ' x $indent/gem; |
374 | s/^\s+\n$/\n/gm; |
375 | print OUTPUT; |
376 | } |
377 | } |
378 | |
379 | sub noremap { |
380 | local($thing_to_hide) = shift; |
381 | $thing_to_hide =~ tr/\000-\177/\200-\377/; |
382 | return $thing_to_hide; |
383 | } |
384 | |
385 | sub init_noremap { |
386 | die "unmatched init" if $mapready++; |
26fb054b |
387 | #mask off high bit characters in input stream |
388 | s/([\200-\377])/"E<".ord($1).">"/ge; |
69e00e79 |
389 | } |
390 | |
391 | sub clear_noremap { |
392 | my $ready_to_print = $_[0]; |
393 | die "unmatched clear" unless $mapready--; |
394 | tr/\200-\377/\000-\177/; |
395 | # now for the E<>s, which have been hidden until now |
396 | # otherwise the interative \w<> processing would have |
397 | # been hosed by the E<gt> |
398 | s { |
26fb054b |
399 | E< |
400 | ( |
401 | ( \d+ ) |
402 | | ( [A-Za-z]+ ) |
403 | ) |
69e00e79 |
404 | > |
405 | } { |
406 | do { |
26fb054b |
407 | defined $2 |
408 | ? chr($2) |
409 | : |
410 | defined $HTML_Escapes{$3} |
411 | ? do { $HTML_Escapes{$3} } |
69e00e79 |
412 | : do { |
f02a87df |
413 | warn "Unknown escape: E<$1> in $_"; |
69e00e79 |
414 | "E<$1>"; |
415 | } |
416 | } |
417 | }egx if $ready_to_print; |
418 | } |
419 | |
420 | sub internal_lrefs { |
421 | local($_) = shift; |
422 | s{L</([^>]+)>}{$1}g; |
423 | my(@items) = split( /(?:,?\s+(?:and\s+)?)/ ); |
424 | my $retstr = "the "; |
425 | my $i; |
426 | for ($i = 0; $i <= $#items; $i++) { |
427 | $retstr .= "C<$items[$i]>"; |
428 | $retstr .= ", " if @items > 2 && $i != $#items; |
429 | $retstr .= " and " if $i+2 == @items; |
430 | } |
431 | |
432 | $retstr .= " entr" . ( @items > 1 ? "ies" : "y" ) |
433 | . " elsewhere in this document "; |
434 | |
435 | return $retstr; |
436 | |
437 | } |
438 | |
439 | BEGIN { |
440 | |
441 | %HTML_Escapes = ( |
442 | 'amp' => '&', # ampersand |
443 | 'lt' => '<', # left chevron, less-than |
444 | 'gt' => '>', # right chevron, greater-than |
445 | 'quot' => '"', # double quote |
446 | |
447 | "Aacute" => "\xC1", # capital A, acute accent |
448 | "aacute" => "\xE1", # small a, acute accent |
449 | "Acirc" => "\xC2", # capital A, circumflex accent |
450 | "acirc" => "\xE2", # small a, circumflex accent |
451 | "AElig" => "\xC6", # capital AE diphthong (ligature) |
452 | "aelig" => "\xE6", # small ae diphthong (ligature) |
453 | "Agrave" => "\xC0", # capital A, grave accent |
454 | "agrave" => "\xE0", # small a, grave accent |
455 | "Aring" => "\xC5", # capital A, ring |
456 | "aring" => "\xE5", # small a, ring |
457 | "Atilde" => "\xC3", # capital A, tilde |
458 | "atilde" => "\xE3", # small a, tilde |
459 | "Auml" => "\xC4", # capital A, dieresis or umlaut mark |
460 | "auml" => "\xE4", # small a, dieresis or umlaut mark |
461 | "Ccedil" => "\xC7", # capital C, cedilla |
462 | "ccedil" => "\xE7", # small c, cedilla |
463 | "Eacute" => "\xC9", # capital E, acute accent |
464 | "eacute" => "\xE9", # small e, acute accent |
465 | "Ecirc" => "\xCA", # capital E, circumflex accent |
466 | "ecirc" => "\xEA", # small e, circumflex accent |
467 | "Egrave" => "\xC8", # capital E, grave accent |
468 | "egrave" => "\xE8", # small e, grave accent |
469 | "ETH" => "\xD0", # capital Eth, Icelandic |
470 | "eth" => "\xF0", # small eth, Icelandic |
471 | "Euml" => "\xCB", # capital E, dieresis or umlaut mark |
472 | "euml" => "\xEB", # small e, dieresis or umlaut mark |
473 | "Iacute" => "\xCD", # capital I, acute accent |
474 | "iacute" => "\xED", # small i, acute accent |
475 | "Icirc" => "\xCE", # capital I, circumflex accent |
476 | "icirc" => "\xEE", # small i, circumflex accent |
477 | "Igrave" => "\xCD", # capital I, grave accent |
478 | "igrave" => "\xED", # small i, grave accent |
479 | "Iuml" => "\xCF", # capital I, dieresis or umlaut mark |
480 | "iuml" => "\xEF", # small i, dieresis or umlaut mark |
481 | "Ntilde" => "\xD1", # capital N, tilde |
482 | "ntilde" => "\xF1", # small n, tilde |
483 | "Oacute" => "\xD3", # capital O, acute accent |
484 | "oacute" => "\xF3", # small o, acute accent |
485 | "Ocirc" => "\xD4", # capital O, circumflex accent |
486 | "ocirc" => "\xF4", # small o, circumflex accent |
487 | "Ograve" => "\xD2", # capital O, grave accent |
488 | "ograve" => "\xF2", # small o, grave accent |
489 | "Oslash" => "\xD8", # capital O, slash |
490 | "oslash" => "\xF8", # small o, slash |
491 | "Otilde" => "\xD5", # capital O, tilde |
492 | "otilde" => "\xF5", # small o, tilde |
493 | "Ouml" => "\xD6", # capital O, dieresis or umlaut mark |
494 | "ouml" => "\xF6", # small o, dieresis or umlaut mark |
495 | "szlig" => "\xDF", # small sharp s, German (sz ligature) |
496 | "THORN" => "\xDE", # capital THORN, Icelandic |
497 | "thorn" => "\xFE", # small thorn, Icelandic |
498 | "Uacute" => "\xDA", # capital U, acute accent |
499 | "uacute" => "\xFA", # small u, acute accent |
500 | "Ucirc" => "\xDB", # capital U, circumflex accent |
501 | "ucirc" => "\xFB", # small u, circumflex accent |
502 | "Ugrave" => "\xD9", # capital U, grave accent |
503 | "ugrave" => "\xF9", # small u, grave accent |
504 | "Uuml" => "\xDC", # capital U, dieresis or umlaut mark |
505 | "uuml" => "\xFC", # small u, dieresis or umlaut mark |
506 | "Yacute" => "\xDD", # capital Y, acute accent |
507 | "yacute" => "\xFD", # small y, acute accent |
508 | "yuml" => "\xFF", # small y, dieresis or umlaut mark |
509 | |
510 | "lchevron" => "\xAB", # left chevron (double less than) |
511 | "rchevron" => "\xBB", # right chevron (double greater than) |
512 | ); |
513 | } |
514 | |
515 | 1; |