gcc/
[official-gcc.git] / contrib / texi2pod.pl
blob57d1d5bcd63202ec8d9358863f7fdf39249900d9
1 #! /usr/bin/perl -w
3 # Copyright (C) 1999, 2000, 2001, 2003 Free Software Foundation, Inc.
5 # This file is part of GCC.
7 # GCC is free software; you can redistribute it and/or modify
8 # it under the terms of the GNU General Public License as published by
9 # the Free Software Foundation; either version 2, or (at your option)
10 # any later version.
12 # GCC is distributed in the hope that it will be useful,
13 # but WITHOUT ANY WARRANTY; without even the implied warranty of
14 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
15 # GNU General Public License for more details.
17 # You should have received a copy of the GNU General Public License
18 # along with GCC; see the file COPYING. If not, write to
19 # the Free Software Foundation, 51 Franklin Street, Fifth Floor,
20 # Boston MA 02110-1301, USA.
22 # This does trivial (and I mean _trivial_) conversion of Texinfo
23 # markup to Perl POD format. It's intended to be used to extract
24 # something suitable for a manpage from a Texinfo document.
26 $output = 0;
27 $skipping = 0;
28 %sects = ();
29 $section = "";
30 @icstack = ();
31 @endwstack = ();
32 @skstack = ();
33 @instack = ();
34 $shift = "";
35 %defs = ();
36 $fnno = 1;
37 $inf = "";
38 $ibase = "";
39 @ipath = ();
41 while ($_ = shift) {
42 if (/^-D(.*)$/) {
43 if ($1 ne "") {
44 $flag = $1;
45 } else {
46 $flag = shift;
48 $value = "";
49 ($flag, $value) = ($flag =~ /^([^=]+)(?:=(.+))?/);
50 die "no flag specified for -D\n"
51 unless $flag ne "";
52 die "flags may only contain letters, digits, hyphens, dashes and underscores\n"
53 unless $flag =~ /^[a-zA-Z0-9_-]+$/;
54 $defs{$flag} = $value;
55 } elsif (/^-I(.*)$/) {
56 if ($1 ne "") {
57 $flag = $1;
58 } else {
59 $flag = shift;
61 push (@ipath, $flag);
62 } elsif (/^-/) {
63 usage();
64 } else {
65 $in = $_, next unless defined $in;
66 $out = $_, next unless defined $out;
67 usage();
71 if (defined $in) {
72 $inf = gensym();
73 open($inf, "<$in") or die "opening \"$in\": $!\n";
74 $ibase = $1 if $in =~ m|^(.+)/[^/]+$|;
75 } else {
76 $inf = \*STDIN;
79 if (defined $out) {
80 open(STDOUT, ">$out") or die "opening \"$out\": $!\n";
83 while(defined $inf) {
84 while(<$inf>) {
85 # Certain commands are discarded without further processing.
86 /^\@(?:
87 [a-z]+index # @*index: useful only in complete manual
88 |need # @need: useful only in printed manual
89 |(?:end\s+)?group # @group .. @end group: ditto
90 |page # @page: ditto
91 |node # @node: useful only in .info file
92 |(?:end\s+)?ifnottex # @ifnottex .. @end ifnottex: use contents
93 )\b/x and next;
95 chomp;
97 # Look for filename and title markers.
98 /^\@setfilename\s+([^.]+)/ and $fn = $1, next;
99 /^\@settitle\s+([^.]+)/ and $tl = postprocess($1), next;
101 # Identify a man title but keep only the one we are interested in.
102 /^\@c\s+man\s+title\s+([A-Za-z0-9-]+)\s+(.+)/ and do {
103 if (exists $defs{$1}) {
104 $fn = $1;
105 $tl = postprocess($2);
107 next;
110 # Look for blocks surrounded by @c man begin SECTION ... @c man end.
111 # This really oughta be @ifman ... @end ifman and the like, but such
112 # would require rev'ing all other Texinfo translators.
113 /^\@c\s+man\s+begin\s+([A-Z]+)\s+([A-Za-z0-9-]+)/ and do {
114 $output = 1 if exists $defs{$2};
115 $sect = $1;
116 next;
118 /^\@c\s+man\s+begin\s+([A-Z]+)/ and $sect = $1, $output = 1, next;
119 /^\@c\s+man\s+end/ and do {
120 $sects{$sect} = "" unless exists $sects{$sect};
121 $sects{$sect} .= postprocess($section);
122 $section = "";
123 $output = 0;
124 next;
127 # handle variables
128 /^\@set\s+([a-zA-Z0-9_-]+)\s*(.*)$/ and do {
129 $defs{$1} = $2;
130 next;
132 /^\@clear\s+([a-zA-Z0-9_-]+)/ and do {
133 delete $defs{$1};
134 next;
137 next unless $output;
139 # Discard comments. (Can't do it above, because then we'd never see
140 # @c man lines.)
141 /^\@c\b/ and next;
143 # End-block handler goes up here because it needs to operate even
144 # if we are skipping.
145 /^\@end\s+([a-z]+)/ and do {
146 # Ignore @end foo, where foo is not an operation which may
147 # cause us to skip, if we are presently skipping.
148 my $ended = $1;
149 next if $skipping && $ended !~ /^(?:ifset|ifclear|ignore|menu|iftex|copying)$/;
151 die "\@end $ended without \@$ended at line $.\n" unless defined $endw;
152 die "\@$endw ended by \@end $ended at line $.\n" unless $ended eq $endw;
154 $endw = pop @endwstack;
156 if ($ended =~ /^(?:ifset|ifclear|ignore|menu|iftex)$/) {
157 $skipping = pop @skstack;
158 next;
159 } elsif ($ended =~ /^(?:example|smallexample|display)$/) {
160 $shift = "";
161 $_ = ""; # need a paragraph break
162 } elsif ($ended =~ /^(?:itemize|enumerate|[fv]?table)$/) {
163 $_ = "\n=back\n";
164 $ic = pop @icstack;
165 } else {
166 die "unknown command \@end $ended at line $.\n";
170 # We must handle commands which can cause skipping even while we
171 # are skipping, otherwise we will not process nested conditionals
172 # correctly.
173 /^\@ifset\s+([a-zA-Z0-9_-]+)/ and do {
174 push @endwstack, $endw;
175 push @skstack, $skipping;
176 $endw = "ifset";
177 $skipping = 1 unless exists $defs{$1};
178 next;
181 /^\@ifclear\s+([a-zA-Z0-9_-]+)/ and do {
182 push @endwstack, $endw;
183 push @skstack, $skipping;
184 $endw = "ifclear";
185 $skipping = 1 if exists $defs{$1};
186 next;
189 /^\@(ignore|menu|iftex|copying)\b/ and do {
190 push @endwstack, $endw;
191 push @skstack, $skipping;
192 $endw = $1;
193 $skipping = 1;
194 next;
197 next if $skipping;
199 # Character entities. First the ones that can be replaced by raw text
200 # or discarded outright:
201 s/\@copyright\{\}/(c)/g;
202 s/\@dots\{\}/.../g;
203 s/\@enddots\{\}/..../g;
204 s/\@([.!? ])/$1/g;
205 s/\@[:-]//g;
206 s/\@bullet(?:\{\})?/*/g;
207 s/\@TeX\{\}/TeX/g;
208 s/\@pounds\{\}/\#/g;
209 s/\@minus(?:\{\})?/-/g;
210 s/\\,/,/g;
212 # Now the ones that have to be replaced by special escapes
213 # (which will be turned back into text by unmunge())
214 s/&/&amp;/g;
215 s/\@\{/&lbrace;/g;
216 s/\@\}/&rbrace;/g;
217 s/\@\@/&at;/g;
219 # Inside a verbatim block, handle @var specially.
220 if ($shift ne "") {
221 s/\@var\{([^\}]*)\}/<$1>/g;
224 # POD doesn't interpret E<> inside a verbatim block.
225 if ($shift eq "") {
226 s/</&lt;/g;
227 s/>/&gt;/g;
228 } else {
229 s/</&LT;/g;
230 s/>/&GT;/g;
233 # Single line command handlers.
235 /^\@include\s+(.+)$/ and do {
236 push @instack, $inf;
237 $inf = gensym();
238 $file = postprocess($1);
240 # Try cwd and $ibase, then explicit -I paths.
241 $done = 0;
242 foreach $path (".", $ibase, @ipath) {
243 open($inf, "<" . $path . "/" . $file) and ($done = 1, last);
245 die "cannot find $file" if !$done;
246 next;
249 /^\@(?:section|unnumbered|unnumberedsec|center)\s+(.+)$/
250 and $_ = "\n=head2 $1\n";
251 /^\@subsection\s+(.+)$/
252 and $_ = "\n=head3 $1\n";
254 # Block command handlers:
255 /^\@itemize(?:\s+(\@[a-z]+|\*|-))?/ and do {
256 push @endwstack, $endw;
257 push @icstack, $ic;
258 if (defined $1) {
259 $ic = $1;
260 } else {
261 $ic = '@bullet';
263 $_ = "\n=over 4\n";
264 $endw = "itemize";
267 /^\@enumerate(?:\s+([a-zA-Z0-9]+))?/ and do {
268 push @endwstack, $endw;
269 push @icstack, $ic;
270 if (defined $1) {
271 $ic = $1 . ".";
272 } else {
273 $ic = "1.";
275 $_ = "\n=over 4\n";
276 $endw = "enumerate";
279 /^\@([fv]?table)\s+(\@[a-z]+)/ and do {
280 push @endwstack, $endw;
281 push @icstack, $ic;
282 $endw = $1;
283 $ic = $2;
284 $ic =~ s/\@(?:samp|strong|key|gcctabopt|env)/B/;
285 $ic =~ s/\@(?:code|kbd)/C/;
286 $ic =~ s/\@(?:dfn|var|emph|cite|i)/I/;
287 $ic =~ s/\@(?:file)/F/;
288 $_ = "\n=over 4\n";
291 /^\@((?:small)?example|display)/ and do {
292 push @endwstack, $endw;
293 $endw = $1;
294 $shift = "\t";
295 $_ = ""; # need a paragraph break
298 /^\@itemx?\s*(.+)?$/ and do {
299 if (defined $1) {
300 # Entity escapes prevent munging by the <> processing below.
301 $_ = "\n=item $ic\&LT;$1\&GT;\n";
302 } else {
303 $_ = "\n=item $ic\n";
304 $ic =~ y/A-Ya-y/B-Zb-z/;
305 $ic =~ s/(\d+)/$1 + 1/eg;
309 $section .= $shift.$_."\n";
311 # End of current file.
312 close($inf);
313 $inf = pop @instack;
316 die "No filename or title\n" unless defined $fn && defined $tl;
318 $sects{NAME} = "$fn \- $tl\n";
319 $sects{FOOTNOTES} .= "=back\n" if exists $sects{FOOTNOTES};
321 for $sect (qw(NAME SYNOPSIS DESCRIPTION OPTIONS ENVIRONMENT FILES
322 BUGS NOTES FOOTNOTES SEEALSO AUTHOR COPYRIGHT)) {
323 if(exists $sects{$sect}) {
324 $head = $sect;
325 $head =~ s/SEEALSO/SEE ALSO/;
326 print "=head1 $head\n\n";
327 print scalar unmunge ($sects{$sect});
328 print "\n";
332 sub usage
334 die "usage: $0 [-D toggle...] [infile [outfile]]\n";
337 sub postprocess
339 local $_ = $_[0];
341 # @value{foo} is replaced by whatever 'foo' is defined as.
342 while (m/(\@value\{([a-zA-Z0-9_-]+)\})/g) {
343 if (! exists $defs{$2}) {
344 print STDERR "Option $2 not defined\n";
345 s/\Q$1\E//;
346 } else {
347 $value = $defs{$2};
348 s/\Q$1\E/$value/;
352 # Formatting commands.
353 # Temporary escape for @r.
354 s/\@r\{([^\}]*)\}/R<$1>/g;
355 s/\@(?:dfn|var|emph|cite|i)\{([^\}]*)\}/I<$1>/g;
356 s/\@(?:code|kbd)\{([^\}]*)\}/C<$1>/g;
357 s/\@(?:gccoptlist|samp|strong|key|option|env|command|b)\{([^\}]*)\}/B<$1>/g;
358 s/\@sc\{([^\}]*)\}/\U$1/g;
359 s/\@file\{([^\}]*)\}/F<$1>/g;
360 s/\@w\{([^\}]*)\}/S<$1>/g;
361 s/\@(?:dmn|math)\{([^\}]*)\}/$1/g;
363 # keep references of the form @ref{...}, print them bold
364 s/\@(?:ref)\{([^\}]*)\}/B<$1>/g;
366 # Change double single quotes to double quotes.
367 s/''/"/g;
368 s/``/"/g;
370 # Cross references are thrown away, as are @noindent and @refill.
371 # (@noindent is impossible in .pod, and @refill is unnecessary.)
372 # @* is also impossible in .pod; we discard it and any newline that
373 # follows it. Similarly, our macro @gol must be discarded.
375 s/\(?\@xref\{(?:[^\}]*)\}(?:[^.<]|(?:<[^<>]*>))*\.\)?//g;
376 s/\s+\(\@pxref\{(?:[^\}]*)\}\)//g;
377 s/;\s+\@pxref\{(?:[^\}]*)\}//g;
378 s/\@noindent\s*//g;
379 s/\@refill//g;
380 s/\@gol//g;
381 s/\@\*\s*\n?//g;
383 # @uref can take one, two, or three arguments, with different
384 # semantics each time. @url and @email are just like @uref with
385 # one argument, for our purposes.
386 s/\@(?:uref|url|email)\{([^\},]*)\}/&lt;B<$1>&gt;/g;
387 s/\@uref\{([^\},]*),([^\},]*)\}/$2 (C<$1>)/g;
388 s/\@uref\{([^\},]*),([^\},]*),([^\},]*)\}/$3/g;
390 # Un-escape <> at this point.
391 s/&LT;/</g;
392 s/&GT;/>/g;
394 # Now un-nest all B<>, I<>, R<>. Theoretically we could have
395 # indefinitely deep nesting; in practice, one level suffices.
396 1 while s/([BIR])<([^<>]*)([BIR])<([^<>]*)>/$1<$2>$3<$4>$1</g;
398 # Replace R<...> with bare ...; eliminate empty markup, B<>;
399 # shift white space at the ends of [BI]<...> expressions outside
400 # the expression.
401 s/R<([^<>]*)>/$1/g;
402 s/[BI]<>//g;
403 s/([BI])<(\s+)([^>]+)>/$2$1<$3>/g;
404 s/([BI])<([^>]+?)(\s+)>/$1<$2>$3/g;
406 # Extract footnotes. This has to be done after all other
407 # processing because otherwise the regexp will choke on formatting
408 # inside @footnote.
409 while (/\@footnote/g) {
410 s/\@footnote\{([^\}]+)\}/[$fnno]/;
411 add_footnote($1, $fnno);
412 $fnno++;
415 return $_;
418 sub unmunge
420 # Replace escaped symbols with their equivalents.
421 local $_ = $_[0];
423 s/&lt;/E<lt>/g;
424 s/&gt;/E<gt>/g;
425 s/&lbrace;/\{/g;
426 s/&rbrace;/\}/g;
427 s/&at;/\@/g;
428 s/&amp;/&/g;
429 return $_;
432 sub add_footnote
434 unless (exists $sects{FOOTNOTES}) {
435 $sects{FOOTNOTES} = "\n=over 4\n\n";
438 $sects{FOOTNOTES} .= "=item $fnno.\n\n"; $fnno++;
439 $sects{FOOTNOTES} .= $_[0];
440 $sects{FOOTNOTES} .= "\n\n";
443 # stolen from Symbol.pm
445 my $genseq = 0;
446 sub gensym
448 my $name = "GEN" . $genseq++;
449 my $ref = \*{$name};
450 delete $::{$name};
451 return $ref;