texi2pod.pl revision 1.1.1.5 
1#! /usr/bin/perl -w
2
3#   Copyright (C) 1999-2022 Free Software Foundation, Inc.
4
5# This file is part of GCC.
6
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.
11
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.
16
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.
21
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.
25
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$ibase = "";
39@ipath = ();
40
41while ($_ = shift) {
42    if (/^-D(.*)$/) {
43	if ($1 ne "") {
44	    $flag = $1;
45	} else {
46	    $flag = shift;
47	}
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;
60	}
61        push (@ipath, $flag);
62    } elsif (/^--no-split$/) {
63	# ignore option for makeinfo compatibility
64    } elsif (/^-/) {
65	usage();
66    } else {
67	$in = $_, next unless defined $in;
68	$out = $_, next unless defined $out;
69	usage();
70    }
71}
72
73if (defined $in) {
74    $inf = gensym();
75    open($inf, "<$in") or die "opening \"$in\": $!\n";
76    $ibase = $1 if $in =~ m|^(.+)/[^/]+$|;
77} else {
78    $inf = \*STDIN;
79}
80
81if (defined $out) {
82    open(STDOUT, ">$out") or die "opening \"$out\": $!\n";
83}
84
85while(defined $inf) {
86while(<$inf>) {
87    # Certain commands are discarded without further processing.
88    /^\@(?:
89	 [a-z]+index		# @*index: useful only in complete manual
90	 |need			# @need: useful only in printed manual
91	 |(?:end\s+)?group	# @group .. @end group: ditto
92	 |page			# @page: ditto
93	 |node			# @node: useful only in .info file
94	 |(?:end\s+)?ifnottex   # @ifnottex .. @end ifnottex: use contents
95	)\b/x and next;
96
97    chomp;
98
99    # Look for filename and title markers.
100    /^\@setfilename\s+([^.]+)/ and $fn = $1, next;
101    /^\@settitle\s+([^.]+)/ and $tl = postprocess($1), next;
102
103    # Identify a man title but keep only the one we are interested in.
104    /^\@c\s+man\s+title\s+([A-Za-z0-9-]+)\s+(.+)/ and do {
105	if (exists $defs{$1}) {
106	    $fn = $1;
107	    $tl = postprocess($2);
108	}
109	next;
110    };
111
112    # Look for blocks surrounded by @c man begin SECTION ... @c man end.
113    # This really oughta be @ifman ... @end ifman and the like, but such
114    # would require rev'ing all other Texinfo translators.
115    /^\@c\s+man\s+begin\s+([A-Z]+)\s+([A-Za-z0-9-]+)/ and do {
116	$output = 1 if exists $defs{$2};
117        $sect = $1;
118	next;
119    };
120    /^\@c\s+man\s+begin\s+([A-Z]+)/ and $sect = $1, $output = 1, next;
121    /^\@c\s+man\s+end/ and do {
122	$sects{$sect} = "" unless exists $sects{$sect};
123	$sects{$sect} .= postprocess($section);
124	$section = "";
125	$output = 0;
126	next;
127    };
128
129    # handle variables
130    /^\@set\s+([a-zA-Z0-9_-]+)\s*(.*)$/ and do {
131	$defs{$1} = $2;
132	next;
133    };
134    /^\@clear\s+([a-zA-Z0-9_-]+)/ and do {
135	delete $defs{$1};
136	next;
137    };
138
139    next unless $output;
140
141    # Discard comments.  (Can't do it above, because then we'd never see
142    # @c man lines.)
143    /^\@c\b/ and next;
144
145    # End-block handler goes up here because it needs to operate even
146    # if we are skipping.
147    /^\@end\s+([a-z]+)/ and do {
148	# Ignore @end foo, where foo is not an operation which may
149	# cause us to skip, if we are presently skipping.
150	my $ended = $1;
151	next if $skipping && $ended !~ /^(?:ifset|ifclear|ignore|menu|iftex|copying)$/;
152
153	die "\@end $ended without \@$ended at line $.\n" unless defined $endw;
154	die "\@$endw ended by \@end $ended at line $.\n" unless $ended eq $endw;
155
156	$endw = pop @endwstack;
157
158	if ($ended =~ /^(?:ifset|ifclear|ignore|menu|iftex)$/) {
159	    $skipping = pop @skstack;
160	    next;
161	} elsif ($ended =~ /^(?:example|smallexample|display)$/) {
162	    $shift = "";
163	    $_ = "";	# need a paragraph break
164	} elsif ($ended =~ /^(?:itemize|enumerate|[fv]?table)$/) {
165	    $_ = "\n=back\n";
166	    $ic = pop @icstack;
167	} elsif ($ended eq "multitable") {
168	    $_ = "\n=back\n";
169	} else {
170	    die "unknown command \@end $ended at line $.\n";
171	}
172    };
173
174    # We must handle commands which can cause skipping even while we
175    # are skipping, otherwise we will not process nested conditionals
176    # correctly.
177    /^\@ifset\s+([a-zA-Z0-9_-]+)/ and do {
178	push @endwstack, $endw;
179	push @skstack, $skipping;
180	$endw = "ifset";
181	$skipping = 1 unless exists $defs{$1};
182	next;
183    };
184
185    /^\@ifclear\s+([a-zA-Z0-9_-]+)/ and do {
186	push @endwstack, $endw;
187	push @skstack, $skipping;
188	$endw = "ifclear";
189	$skipping = 1 if exists $defs{$1};
190	next;
191    };
192
193    /^\@(ignore|menu|iftex|copying)\b/ and do {
194	push @endwstack, $endw;
195	push @skstack, $skipping;
196	$endw = $1;
197	$skipping = 1;
198	next;
199    };
200
201    next if $skipping;
202
203    # Character entities.  First the ones that can be replaced by raw text
204    # or discarded outright:
205    s/\@copyright\{\}/(c)/g;
206    s/\@dots\{\}/.../g;
207    s/\@enddots\{\}/..../g;
208    s/\@([.!? ])/$1/g;
209    s/\@[:-]//g;
210    s/\@bullet(?:\{\})?/*/g;
211    s/\@TeX\{\}/TeX/g;
212    s/\@pounds\{\}/\#/g;
213    s/\@minus(?:\{\})?/-/g;
214    s/\\,/,/g;
215
216    # Now the ones that have to be replaced by special escapes
217    # (which will be turned back into text by unmunge())
218    s/&/&amp;/g;
219    s/\@\{/&lbrace;/g;
220    s/\@\}/&rbrace;/g;
221    s/\@\@/&at;/g;
222
223    # Inside a verbatim block, handle @var specially.
224    if ($shift ne "") {
225	s/\@var\{([^\}]*)\}/<$1>/g;
226    }
227
228    # POD doesn't interpret E<> inside a verbatim block.
229    if ($shift eq "") {
230	s/</&lt;/g;
231	s/>/&gt;/g;
232    } else {
233	s/</&LT;/g;
234	s/>/&GT;/g;
235    }
236
237    # Single line command handlers.
238
239    /^\@include\s+(.+)$/ and do {
240	push @instack, $inf;
241	$inf = gensym();
242	$file = postprocess($1);
243
244	# Try cwd and $ibase, then explicit -I paths.
245	$done = 0;
246	foreach $path ("", $ibase, @ipath) {
247	    $mypath = $file;
248	    $mypath = $path . "/" . $mypath if ($path ne "");
249	    open($inf, "<" . $mypath) and ($done = 1, last);
250	}
251	die "cannot find $file" if !$done;
252	next;
253    };
254
255    /^\@(?:section|unnumbered|unnumberedsec|center)\s+(.+)$/
256	and $_ = "\n=head2 $1\n";
257    /^\@subsection\s+(.+)$/
258	and $_ = "\n=head3 $1\n";
259    /^\@subsubsection\s+(.+)$/
260	and $_ = "\n=head4 $1\n";
261
262    # Block command handlers:
263    /^\@itemize(?:\s+(\@[a-z]+|\*|-))?/ and do {
264	push @endwstack, $endw;
265	push @icstack, $ic;
266	if (defined $1) {
267	    $ic = $1;
268	} else {
269	    $ic = '*';
270	}
271	$_ = "\n=over 4\n";
272	$endw = "itemize";
273    };
274
275    /^\@enumerate(?:\s+([a-zA-Z0-9]+))?/ and do {
276	push @endwstack, $endw;
277	push @icstack, $ic;
278	if (defined $1) {
279	    $ic = $1 . ".";
280	} else {
281	    $ic = "1.";
282	}
283	$_ = "\n=over 4\n";
284	$endw = "enumerate";
285    };
286
287    /^\@multitable\s.*/ and do {
288	push @endwstack, $endw;
289	$endw = "multitable";
290	$_ = "\n=over 4\n";
291    };
292
293    /^\@([fv]?table)\s+(\@[a-z]+)/ and do {
294	push @endwstack, $endw;
295	push @icstack, $ic;
296	$endw = $1;
297	$ic = $2;
298	$ic =~ s/\@(?:samp|strong|key|gcctabopt|env)/B/;
299	$ic =~ s/\@(?:code|kbd)/C/;
300	$ic =~ s/\@(?:dfn|var|emph|cite|i)/I/;
301	$ic =~ s/\@(?:file)/F/;
302	$_ = "\n=over 4\n";
303    };
304
305    /^\@((?:small)?example|display)/ and do {
306	push @endwstack, $endw;
307	$endw = $1;
308	$shift = "\t";
309	$_ = "";	# need a paragraph break
310    };
311
312    /^\@item\s+(.*\S)\s*$/ and $endw eq "multitable" and do {
313	@columns = ();
314	for $column (split (/\s*\@tab\s*/, $1)) {
315	    # @strong{...} is used a @headitem work-alike
316	    $column =~ s/^\@strong\{(.*)\}$/$1/;
317	    push @columns, $column;
318	}
319	$_ = "\n=item ".join (" : ", @columns)."\n";
320    };
321
322    /^\@itemx?\s*(.+)?$/ and do {
323	if (defined $1) {
324	    # Entity escapes prevent munging by the <> processing below.
325	    $_ = "\n=item $ic\&LT;$1\&GT;\n";
326	} else {
327	    $_ = "\n=item $ic\n";
328	    $ic =~ y/A-Ya-y/B-Zb-z/;
329	    $ic =~ s/(\d+)/$1 + 1/eg;
330	}
331    };
332
333    $section .= $shift.$_."\n";
334}
335# End of current file.
336close($inf);
337$inf = pop @instack;
338}
339
340die "No filename or title\n" unless defined $fn && defined $tl;
341
342$sects{NAME} = "$fn \- $tl\n";
343$sects{FOOTNOTES} .= "=back\n" if exists $sects{FOOTNOTES};
344
345for $sect (qw(NAME SYNOPSIS TARGET DESCRIPTION OPTIONS ENVIRONMENT FILES
346	      BUGS NOTES FOOTNOTES SEEALSO AUTHOR COPYRIGHT)) {
347    if(exists $sects{$sect}) {
348	$head = $sect;
349	$head =~ s/SEEALSO/SEE ALSO/;
350	print "=head1 $head\n\n";
351	print scalar unmunge ($sects{$sect});
352	print "\n";
353    }
354}
355
356sub usage
357{
358    die "usage: $0 [-D toggle...] [infile [outfile]]\n";
359}
360
361sub postprocess
362{
363    local $_ = $_[0];
364
365    # @value{foo} is replaced by whatever 'foo' is defined as.
366    while (m/(\@value\{([a-zA-Z0-9_-]+)\})/g) {
367	if (! exists $defs{$2}) {
368	    print STDERR "Option $2 not defined\n";
369	    s/\Q$1\E//;
370	} else {
371	    $value = $defs{$2};
372	    s/\Q$1\E/$value/;
373	}
374    }
375
376    # Formatting commands.
377    # Temporary escape for @r.
378    s/\@r\{([^\}]*)\}/R<$1>/g;
379    s/\@(?:dfn|var|emph|cite|i)\{([^\}]*)\}/I<$1>/g;
380    s/\@(?:code|kbd)\{([^\}]*)\}/C<$1>/g;
381    s/\@(?:gccoptlist|samp|strong|key|option|env|command|b)\{([^\}]*)\}/B<$1>/g;
382    s/\@sc\{([^\}]*)\}/\U$1/g;
383    s/\@file\{([^\}]*)\}/F<$1>/g;
384    s/\@w\{([^\}]*)\}/S<$1>/g;
385    s/\@(?:dmn|math)\{([^\}]*)\}/$1/g;
386    s/\@t\{([^\}]*)\}/$1/g;
387
388    # keep references of the form @ref{...}, print them bold
389    s/\@(?:ref)\{([^\}]*)\}/B<$1>/g;
390
391    # Change double single quotes to double quotes.
392    s/''/"/g;
393    s/``/"/g;
394
395    # Cross references are thrown away, as are @noindent and @refill.
396    # (@noindent is impossible in .pod, and @refill is unnecessary.)
397    # @* is also impossible in .pod; we discard it and any newline that
398    # follows it.  Similarly, our macro @gol must be discarded.
399
400    s/\(?\@xref\{(?:[^\}]*)\}(?:[^.<]|(?:<[^<>]*>))*\.\)?//g;
401    s/\s+\(\@pxref\{(?:[^\}]*)\}\)//g;
402    s/;\s+\@pxref\{(?:[^\}]*)\}//g;
403    s/\@noindent\s*//g;
404    s/\@refill//g;
405    s/\@gol//g;
406    s/\@\*\s*\n?//g;
407
408    # Anchors are thrown away
409    s/\@anchor\{(?:[^\}]*)\}//g;
410
411    # @uref can take one, two, or three arguments, with different
412    # semantics each time.  @url and @email are just like @uref with
413    # one argument, for our purposes.
414    s/\@(?:uref|url|email)\{([^\},]*)\}/&lt;B<$1>&gt;/g;
415    s/\@uref\{([^\},]*),([^\},]*)\}/$2 (C<$1>)/g;
416    s/\@uref\{([^\},]*),([^\},]*),([^\},]*)\}/$3/g;
417
418    # Un-escape <> at this point.
419    s/&LT;/</g;
420    s/&GT;/>/g;
421
422    # Now un-nest all B<>, I<>, R<>.  Theoretically we could have
423    # indefinitely deep nesting; in practice, one level suffices.
424    1 while s/([BIR])<([^<>]*)([BIR])<([^<>]*)>/$1<$2>$3<$4>$1</g;
425
426    # Replace R<...> with bare ...; eliminate empty markup, B<>;
427    # shift white space at the ends of [BI]<...> expressions outside
428    # the expression.
429    s/R<([^<>]*)>/$1/g;
430    s/[BI]<>//g;
431    s/([BI])<(\s+)([^>]+)>/$2$1<$3>/g;
432    s/([BI])<([^>]+?)(\s+)>/$1<$2>$3/g;
433
434    # Extract footnotes.  This has to be done after all other
435    # processing because otherwise the regexp will choke on formatting
436    # inside @footnote.
437    while (/\@footnote/g) {
438	s/\@footnote\{([^\}]+)\}/[$fnno]/;
439	add_footnote($1, $fnno);
440	$fnno++;
441    }
442
443    return $_;
444}
445
446sub unmunge
447{
448    # Replace escaped symbols with their equivalents.
449    local $_ = $_[0];
450
451    s/&lt;/E<lt>/g;
452    s/&gt;/E<gt>/g;
453    s/&lbrace;/\{/g;
454    s/&rbrace;/\}/g;
455    s/&at;/\@/g;
456    s/&amp;/&/g;
457    return $_;
458}
459
460sub add_footnote
461{
462    unless (exists $sects{FOOTNOTES}) {
463	$sects{FOOTNOTES} = "\n=over 4\n\n";
464    }
465
466    $sects{FOOTNOTES} .= "=item $fnno.\n\n"; $fnno++;
467    $sects{FOOTNOTES} .= $_[0];
468    $sects{FOOTNOTES} .= "\n\n";
469}
470
471# stolen from Symbol.pm
472{
473    my $genseq = 0;
474    sub gensym
475    {
476	my $name = "GEN" . $genseq++;
477	my $ref = \*{$name};
478	delete $::{$name};
479	return $ref;
480    }
481}
482