]> git.ipfire.org Git - thirdparty/gcc.git/blob - contrib/texi2pod.pl
contrib:
[thirdparty/gcc.git] / contrib / texi2pod.pl
1 #! /usr/bin/perl -w
2
3 # This does trivial (and I mean _trivial_) conversion of Texinfo
4 # markup to Perl POD format. It's intended to be used to extract
5 # something suitable for a manpage from a Texinfo document.
6
7 $output = 0;
8 $skipping = 0;
9 %sects = ();
10 $section = "";
11 @icstack = ();
12 @endwstack = ();
13 @skstack = ();
14 $shift = "";
15 %defs = ();
16 $fnno = 1;
17
18 while ($_ = shift) {
19 if (/^-D(.*)$/) {
20 if ($1 ne "") {
21 $flag = $1;
22 } else {
23 $flag = shift;
24 }
25 die "no flag specified for -D\n"
26 unless $flag ne "";
27 die "flags may only contain letters, digits, hyphens, and underscores\n"
28 unless $flag =~ /^[a-zA-Z0-9_-]+$/;
29 $defs{$flag} = "";
30 } elsif (/^-/) {
31 usage();
32 } else {
33 $in = $_, next unless defined $in;
34 $out = $_, next unless defined $out;
35 usage();
36 }
37 }
38
39 if (defined $in) {
40 open(STDIN, $in) or die "opening \"$in\": $!\n";
41 }
42 if (defined $out) {
43 open(STDOUT, ">$out") or die "opening \"$out\": $!\n";
44 }
45
46 while(<STDIN>)
47 {
48 # Certain commands are discarded without further processing.
49 /^\@(?:
50 [a-z]+index # @*index: useful only in complete manual
51 |need # @need: useful only in printed manual
52 |(?:end\s+)?group # @group .. @end group: ditto
53 |page # @page: ditto
54 |node # @node: useful only in .info file
55 )\b/x and next;
56
57 chomp;
58
59 # Look for filename and title markers.
60 /^\@setfilename\s+([^.]+)/ and $fn = $1, next;
61 /^\@settitle\s+([^.]+)/ and $tl = $1, next;
62
63 # Look for blocks surrounded by @c man begin SECTION ... @c man end.
64 # This really oughta be @ifman ... @end ifman and the like, but such
65 # would require rev'ing all other Texinfo translators.
66 /^\@c man begin ([A-Z]+)/ and $sect = $1, $output = 1, next;
67 /^\@c man end/ and do {
68 $sects{$sect} = "" unless exists $sects{$sect};
69 $sects{$sect} .= postprocess($section);
70 $section = "";
71 $output = 0;
72 next;
73 };
74 next unless $output;
75
76 # Discard comments. (Can't do it above, because then we'd never see
77 # @c man lines.)
78 /^\@c\b/ and next;
79
80 # End-block handler goes up here because it needs to operate even
81 # if we are skipping.
82 /^\@end\s+([a-z]+)/ and do {
83 # Ignore @end foo, where foo is not an operation which may
84 # cause us to skip, if we are presently skipping.
85 my $ended = $1;
86 next if $skipping && $ended !~ /^(?:ifset|ifclear|ignore|menu)$/;
87
88 die "\@end $ended without \@$ended at line $.\n" unless defined $endw;
89 die "\@$endw ended by \@end $ended at line $.\n" unless $ended eq $endw;
90
91 $endw = pop @endwstack;
92
93 if ($ended =~ /^(?:ifset|ifclear|ignore|menu)$/) {
94 $skipping = pop @skstack;
95 next;
96 } elsif ($ended =~ /^(?:example|smallexample)$/) {
97 $shift = "";
98 $_ = ""; # need a paragraph break
99 } elsif ($ended =~ /^(?:itemize|enumerate|table)$/) {
100 $_ = "\n=back\n";
101 $ic = pop @icstack;
102 } else {
103 die "unknown command \@end $ended at line $.\n";
104 }
105 };
106
107 # We must handle commands which can cause skipping even while we
108 # are skipping, otherwise we will not process nested conditionals
109 # correctly.
110 /^\@ifset\s+([a-zA-Z0-9_-]+)/ and do {
111 push @endwstack, $endw;
112 push @skstack, $skipping;
113 $endw = "ifset";
114 $skipping = 1 unless exists $defs{$1};
115 next;
116 };
117
118 /^\@ifclear\s+([a-zA-Z0-9_-]+)/ and do {
119 push @endwstack, $endw;
120 push @skstack, $skipping;
121 $endw = "ifclear";
122 $skipping = 1 if exists $defs{$1};
123 next;
124 };
125
126 /^\@(ignore|menu)\b/ and do {
127 push @endwstack, $endw;
128 push @skstack, $skipping;
129 $endw = $1;
130 $skipping = 1;
131 next;
132 };
133
134 next if $skipping;
135
136 # Character entities. First the ones that can be replaced by raw text
137 # or discarded outright:
138 s/\@copyright\{\}/(c)/g;
139 s/\@dots\{\}/.../g;
140 s/\@enddots\{\}/..../g;
141 s/\@([.!? ])/$1/g;
142 s/\@[:-]//g;
143 s/\@bullet(?:\{\})?/*/g;
144 s/\@TeX\{\}/TeX/g;
145 s/\@pounds\{\}/\#/g;
146 s/\@minus(?:\{\})?/-/g;
147
148 # Now the ones that have to be replaced by special escapes
149 # (which will be turned back into text by unmunge())
150 s/&/&amp;/g;
151 s/\@\{/&lbrace;/g;
152 s/\@\}/&rbrace;/g;
153 s/\@\@/&at;/g;
154 # POD doesn't interpret E<> inside a verbatim block.
155 if ($shift eq "") {
156 s/</&lt;/g;
157 s/>/&gt;/g;
158 } else {
159 s/</&LT;/g;
160 s/>/&GT;/g;
161 }
162
163 # Single line command handlers.
164 /^\@set\s+([a-zA-Z0-9_-]+)\s*(.*)$/ and $defs{$1} = $2, next;
165 /^\@clear\s+([a-zA-Z0-9_-]+)/ and delete $defs{$1}, next;
166
167 /^\@section\s+(.+)$/ and $_ = "\n=head2 $1\n";
168 /^\@subsection\s+(.+)$/ and $_ = "\n=head3 $1\n";
169
170 # Block command handlers:
171 /^\@itemize\s+(\@[a-z]+|\*|-)/ and do {
172 push @endwstack, $endw;
173 push @icstack, $ic;
174 $ic = $1;
175 $_ = "\n=over 4\n";
176 $endw = "itemize";
177 };
178
179 /^\@enumerate(?:\s+([A-Z0-9]+))?/ and do {
180 push @endwstack, $endw;
181 push @icstack, $ic;
182 if (defined $1) {
183 $ic = $1 . ".";
184 } else {
185 $ic = "1.";
186 }
187 $_ = "\n=over 4\n";
188 $endw = "enumerate";
189 };
190
191 /^\@table\s+(\@[a-z]+)/ and do {
192 push @endwstack, $endw;
193 push @icstack, $ic;
194 $ic = $1;
195 $ic =~ s/\@(?:samp|strong|key|gcctabopt|env)/B/;
196 $ic =~ s/\@(?:code|kbd)/C/;
197 $ic =~ s/\@(?:dfn|var|emph|cite|i)/I/;
198 $ic =~ s/\@(?:file)/F/;
199 $_ = "\n=over 4\n";
200 $endw = "table";
201 };
202
203 /^\@((?:small)?example)/ and do {
204 push @endwstack, $endw;
205 $endw = $1;
206 $shift = "\t";
207 $_ = ""; # need a paragraph break
208 };
209
210 /^\@itemx?\s*(.+)?$/ and do {
211 if (defined $1) {
212 # Entity escapes prevent munging by the <> processing below.
213 $_ = "\n=item $ic\&LT;$1\&GT;\n";
214 } else {
215 $_ = "\n=item $ic\n";
216 $ic =~ y/A-Ya-y1-8/B-Zb-z2-9/;
217 }
218 };
219
220 $section .= $shift.$_."\n";
221 }
222
223 die "No filename or title\n" unless defined $fn && defined $tl;
224
225 $sects{NAME} = "$fn \- $tl\n";
226 $sects{FOOTNOTES} .= "=back\n" if exists $sects{FOOTNOTES};
227
228 for $sect (qw(NAME SYNOPSIS DESCRIPTION OPTIONS ENVIRONMENT FILES
229 BUGS NOTES FOOTNOTES SEEALSO AUTHOR COPYRIGHT)) {
230 if(exists $sects{$sect}) {
231 $head = $sect;
232 $head =~ s/SEEALSO/SEE ALSO/;
233 print "=head1 $head\n\n";
234 print scalar unmunge ($sects{$sect});
235 print "\n";
236 }
237 }
238
239 sub usage
240 {
241 die "usage: $0 [-D toggle...] [infile [outfile]]\n";
242 }
243
244 sub postprocess
245 {
246 local $_ = $_[0];
247
248 # @value{foo} is replaced by whatever 'foo' is defined as.
249 s/\@value\{([a-zA-Z0-9_-]+)\}/$defs{$1}/g;
250
251 # Formatting commands.
252 # Temporary escape for @r.
253 s/\@r\{([^\}]*)\}/R<$1>/g;
254 s/\@(?:dfn|var|emph|cite|i)\{([^\}]*)\}/I<$1>/g;
255 s/\@(?:code|kbd)\{([^\}]*)\}/C<$1>/g;
256 s/\@(?:gccoptlist|samp|strong|key|option|env|command|b)\{([^\}]*)\}/B<$1>/g;
257 s/\@sc\{([^\}]*)\}/\U$1/g;
258 s/\@file\{([^\}]*)\}/F<$1>/g;
259 s/\@w\{([^\}]*)\}/S<$1>/g;
260 s/\@(?:dmn|math)\{([^\}]*)\}/$1/g;
261
262 # Handle @r inside bold.
263 1 while s/B<((?:[^<>]*|I<[^<>*]*>)*)R<([^>]*)>/B<$1>${2}B</g;
264
265 # Cross references are thrown away, as are @noindent and @refill.
266 # (@noindent is impossible in .pod, and @refill is unnecessary.)
267 # @* is also impossible in .pod; we discard it and any newline that
268 # follows it. Similarly, our macro @gol must be discarded.
269
270 s/\(?\@xref\{(?:[^\}]*)\}(?:[^.<]|(?:<[^<>]*>))*\.\)?//g;
271 s/\s+\(\@pxref\{(?:[^\}]*)\}\)//g;
272 s/;\s+\@pxref\{(?:[^\}]*)\}//g;
273 s/\@noindent\s*//g;
274 s/\@refill//g;
275 s/\@gol//g;
276 s/\@\*\s*\n?//g;
277
278 # @uref can take one, two, or three arguments, with different
279 # semantics each time. @url and @email are just like @uref with
280 # one argument, for our purposes.
281 s/\@(?:uref|url|email)\{([^\},]*)\}/&lt;B<$1>&gt;/g;
282 s/\@uref\{([^\},]*),([^\},]*)\}/$2 (C<$1>)/g;
283 s/\@uref\{([^\},]*),([^\},]*),([^\},]*)\}/$3/g;
284
285 # Turn B<blah I<blah> blah> into B<blah> I<blah> B<blah> to
286 # match Texinfo semantics of @emph inside @samp.
287 s/&LT;/</g;
288 s/&GT;/>/g;
289 1 while (s/B<([^<>]*)I<([^>]+)>/B<$1>I<$2>B</g);
290 1 while (s/I<([^<>]*)B<([^>]+)>/I<$1>B<$2>I</g);
291 s/[BI]<>//g;
292 s/([BI])<(\s+)([^>]+)>/$2$1<$3>/g;
293 s/([BI])<([^>]+?)(\s+)>/$1<$2>$3/g;
294
295 # Extract footnotes. This has to be done after all other
296 # processing because otherwise the regexp will choke on formatting
297 # inside @footnote.
298 while (/\@footnote/g) {
299 s/\@footnote\{([^\}]+)\}/[$fnno]/;
300 add_footnote($1, $fnno);
301 $fnno++;
302 }
303
304 return $_;
305 }
306
307 sub unmunge
308 {
309 # Replace escaped symbols with their equivalents.
310 local $_ = $_[0];
311
312 s/&lt;/E<lt>/g;
313 s/&gt;/E<gt>/g;
314 s/&lbrace;/\{/g;
315 s/&rbrace;/\}/g;
316 s/&at;/\@/g;
317 s/&amp;/&/g;
318 return $_;
319 }
320
321 sub add_footnote
322 {
323 unless (exists $sects{FOOTNOTES}) {
324 $sects{FOOTNOTES} = "\n=over 4\n\n";
325 }
326
327 $sects{FOOTNOTES} .= "=item $fnno.\n\n"; $fnno++;
328 $sects{FOOTNOTES} .= $_[0];
329 $sects{FOOTNOTES} .= "\n\n";
330 }
331