1#! /usr/bin/perl -w
2
3#   Copyright (C) 1999, 2000, 2001, 2003 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 (/^-/) {
63	usage();
64    } else {
65	$in = $_, next unless defined $in;
66	$out = $_, next unless defined $out;
67	usage();
68    }
69}
70
71if (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;
77}
78
79if (defined $out) {
80    open(STDOUT, ">$out") or die "opening \"$out\": $!\n";
81}
82
83while(defined $inf) {
84while(<$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;
94
95    chomp;
96
97    # Look for filename and title markers.
98    /^\@setfilename\s+([^.]+)/ and $fn = $1, next;
99    /^\@settitle\s+([^.]+)/ and $tl = postprocess($1), next;
100
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);
106	}
107	next;
108    };
109
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;
117    };
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	$output = 0;
124	next;
125    };
126
127    # handle variables
128    /^\@set\s+([a-zA-Z0-9_-]+)\s*(.*)$/ and do {
129	$defs{$1} = $2;
130	next;
131    };
132    /^\@clear\s+([a-zA-Z0-9_-]+)/ and do {
133	delete $defs{$1};
134	next;
135    };
136
137    next unless $output;
138
139    # Discard comments.  (Can't do it above, because then we'd never see
140    # @c man lines.)
141    /^\@c\b/ and next;
142
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 && $ended !~ /^(?:ifset|ifclear|ignore|menu|iftex|copying)$/;
150
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;
153
154	$endw = pop @endwstack;
155
156	if ($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";
167	}
168    };
169
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;
179    };
180
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;
187    };
188
189    /^\@(ignore|menu|iftex|copying)\b/ and do {
190	push @endwstack, $endw;
191	push @skstack, $skipping;
192	$endw = $1;
193	$skipping = 1;
194	next;
195    };
196
197    next if $skipping;
198
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/\@([.!? ])/$1/g;
205    s/\@[:-]//g;
206    s/\@bullet(?:\{\})?/*/g;
207    s/\@TeX\{\}/TeX/g;
208    s/\@pounds\{\}/\#/g;
209    s/\@minus(?:\{\})?/-/g;
210    s/\\,/,/g;
211
212    # Now the ones that have to be replaced by special escapes
213    # (which will be turned back into text by unmunge())
214    s/&/&amp;/g;
215    s/\@\{/&lbrace;/g;
216    s/\@\}/&rbrace;/g;
217    s/\@\@/&at;/g;
218
219    # Inside a verbatim block, handle @var specially.
220    if ($shift ne "") {
221	s/\@var\{([^\}]*)\}/<$1>/g;
222    }
223
224    # POD doesn't interpret E<> inside a verbatim block.
225    if ($shift eq "") {
226	s/</&lt;/g;
227	s/>/&gt;/g;
228    } else {
229	s/</&LT;/g;
230	s/>/&GT;/g;
231    }
232
233    # Single line command handlers.
234
235    /^\@include\s+(.+)$/ and do {
236	push @instack, $inf;
237	$inf = gensym();
238	$file = postprocess($1);
239
240	# Try cwd and $ibase, then explicit -I paths.
241	$done = 0;
242	foreach $path ("", $ibase, @ipath) {
243	    $mypath = $file;
244	    $mypath = $path . "/" . $mypath if ($path ne "");
245	    open($inf, "<" . $mypath) and ($done = 1, last);
246	}
247	die "cannot find $file" if !$done;
248	next;
249    };
250
251    /^\@(?:section|unnumbered|unnumberedsec|center)\s+(.+)$/
252	and $_ = "\n=head2 $1\n";
253    /^\@subsection\s+(.+)$/
254	and $_ = "\n=head3 $1\n";
255
256    # Block command handlers:
257    /^\@itemize(?:\s+(\@[a-z]+|\*|-))?/ and do {
258	push @endwstack, $endw;
259	push @icstack, $ic;
260	if (defined $1) {
261	    $ic = $1;
262	} else {
263	    $ic = '@bullet';
264	}
265	$_ = "\n=over 4\n";
266	$endw = "itemize";
267    };
268
269    /^\@enumerate(?:\s+([a-zA-Z0-9]+))?/ and do {
270	push @endwstack, $endw;
271	push @icstack, $ic;
272	if (defined $1) {
273	    $ic = $1 . ".";
274	} else {
275	    $ic = "1.";
276	}
277	$_ = "\n=over 4\n";
278	$endw = "enumerate";
279    };
280
281    /^\@([fv]?table)\s+(\@[a-z]+)/ and do {
282	push @endwstack, $endw;
283	push @icstack, $ic;
284	$endw = $1;
285	$ic = $2;
286	$ic =~ s/\@(?:samp|strong|key|gcctabopt|env)/B/;
287	$ic =~ s/\@(?:code|kbd)/C/;
288	$ic =~ s/\@(?:dfn|var|emph|cite|i)/I/;
289	$ic =~ s/\@(?:file)/F/;
290	$_ = "\n=over 4\n";
291    };
292
293    /^\@((?:small)?example|display)/ and do {
294	push @endwstack, $endw;
295	$endw = $1;
296	$shift = "\t";
297	$_ = "";	# need a paragraph break
298    };
299
300    /^\@itemx?\s*(.+)?$/ and do {
301	if (defined $1) {
302	    # Entity escapes prevent munging by the <> processing below.
303	    $_ = "\n=item $ic\&LT;$1\&GT;\n";
304	} else {
305	    $_ = "\n=item $ic\n";
306	    $ic =~ y/A-Ya-y/B-Zb-z/;
307	    $ic =~ s/(\d+)/$1 + 1/eg;
308	}
309    };
310
311    $section .= $shift.$_."\n";
312}
313# End of current file.
314close($inf);
315$inf = pop @instack;
316}
317
318die "No filename or title\n" unless defined $fn && defined $tl;
319
320$sects{NAME} = "$fn \- $tl\n";
321$sects{FOOTNOTES} .= "=back\n" if exists $sects{FOOTNOTES};
322
323for $sect (qw(NAME SYNOPSIS DESCRIPTION OPTIONS ENVIRONMENT FILES
324	      BUGS NOTES FOOTNOTES SEEALSO AUTHOR COPYRIGHT)) {
325    if(exists $sects{$sect}) {
326	$head = $sect;
327	$head =~ s/SEEALSO/SEE ALSO/;
328	print "=head1 $head\n\n";
329	print scalar unmunge ($sects{$sect});
330	print "\n";
331    }
332}
333
334sub usage
335{
336    die "usage: $0 [-D toggle...] [infile [outfile]]\n";
337}
338
339sub postprocess
340{
341    local $_ = $_[0];
342
343    # @value{foo} is replaced by whatever 'foo' is defined as.
344    while (m/(\@value\{([a-zA-Z0-9_-]+)\})/g) {
345	if (! exists $defs{$2}) {
346	    print STDERR "Option $2 not defined\n";
347	    s/\Q$1\E//;
348	} else {
349	    $value = $defs{$2};
350	    s/\Q$1\E/$value/;
351	}
352    }
353
354    # Formatting commands.
355    # Temporary escape for @r.
356    s/\@r\{([^\}]*)\}/R<$1>/g;
357    s/\@(?:dfn|var|emph|cite|i)\{([^\}]*)\}/I<$1>/g;
358    s/\@(?:code|kbd)\{([^\}]*)\}/C<$1>/g;
359    s/\@(?:gccoptlist|samp|strong|key|option|env|command|b)\{([^\}]*)\}/B<$1>/g;
360    s/\@sc\{([^\}]*)\}/\U$1/g;
361    s/\@file\{([^\}]*)\}/F<$1>/g;
362    s/\@w\{([^\}]*)\}/S<$1>/g;
363    s/\@(?:dmn|math)\{([^\}]*)\}/$1/g;
364
365    # keep references of the form @ref{...}, print them bold
366    s/\@(?:ref)\{([^\}]*)\}/B<$1>/g;
367
368    # Change double single quotes to double quotes.
369    s/''/"/g;
370    s/``/"/g;
371
372    # Cross references are thrown away, as are @noindent and @refill.
373    # (@noindent is impossible in .pod, and @refill is unnecessary.)
374    # @* is also impossible in .pod; we discard it and any newline that
375    # follows it.  Similarly, our macro @gol must be discarded.
376
377    s/\(?\@xref\{(?:[^\}]*)\}(?:[^.<]|(?:<[^<>]*>))*\.\)?//g;
378    s/\s+\(\@pxref\{(?:[^\}]*)\}\)//g;
379    s/;\s+\@pxref\{(?:[^\}]*)\}//g;
380    s/\@noindent\s*//g;
381    s/\@refill//g;
382    s/\@gol//g;
383    s/\@\*\s*\n?//g;
384
385    # @uref can take one, two, or three arguments, with different
386    # semantics each time.  @url and @email are just like @uref with
387    # one argument, for our purposes.
388    s/\@(?:uref|url|email)\{([^\},]*)\}/&lt;B<$1>&gt;/g;
389    s/\@uref\{([^\},]*),([^\},]*)\}/$2 (C<$1>)/g;
390    s/\@uref\{([^\},]*),([^\},]*),([^\},]*)\}/$3/g;
391
392    # Un-escape <> at this point.
393    s/&LT;/</g;
394    s/&GT;/>/g;
395
396    # Now un-nest all B<>, I<>, R<>.  Theoretically we could have
397    # indefinitely deep nesting; in practice, one level suffices.
398    1 while s/([BIR])<([^<>]*)([BIR])<([^<>]*)>/$1<$2>$3<$4>$1</g;
399
400    # Replace R<...> with bare ...; eliminate empty markup, B<>;
401    # shift white space at the ends of [BI]<...> expressions outside
402    # the expression.
403    s/R<([^<>]*)>/$1/g;
404    s/[BI]<>//g;
405    s/([BI])<(\s+)([^>]+)>/$2$1<$3>/g;
406    s/([BI])<([^>]+?)(\s+)>/$1<$2>$3/g;
407
408    # Extract footnotes.  This has to be done after all other
409    # processing because otherwise the regexp will choke on formatting
410    # inside @footnote.
411    while (/\@footnote/g) {
412	s/\@footnote\{([^\}]+)\}/[$fnno]/;
413	add_footnote($1, $fnno);
414	$fnno++;
415    }
416
417    return $_;
418}
419
420sub unmunge
421{
422    # Replace escaped symbols with their equivalents.
423    local $_ = $_[0];
424
425    s/&lt;/E<lt>/g;
426    s/&gt;/E<gt>/g;
427    s/&lbrace;/\{/g;
428    s/&rbrace;/\}/g;
429    s/&at;/\@/g;
430    s/&amp;/&/g;
431    return $_;
432}
433
434sub add_footnote
435{
436    unless (exists $sects{FOOTNOTES}) {
437	$sects{FOOTNOTES} = "\n=over 4\n\n";
438    }
439
440    $sects{FOOTNOTES} .= "=item $fnno.\n\n"; $fnno++;
441    $sects{FOOTNOTES} .= $_[0];
442    $sects{FOOTNOTES} .= "\n\n";
443}
444
445# stolen from Symbol.pm
446{
447    my $genseq = 0;
448    sub gensym
449    {
450	my $name = "GEN" . $genseq++;
451	my $ref = \*{$name};
452	delete $::{$name};
453	return $ref;
454    }
455}
456