contrib/texi2pod.pl

   1 #! /usr/bin/perl -w
   2
   3 #   Copyright (C) 1999, 2000, 2001 Free Software Foundation, Inc.
   4
   5 # This file is part of GNU CC.
   6
   7 # GNU CC 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 # GNU CC 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 GNU CC; see the file COPYING.  If not, write to
  19 # the Free Software Foundation, 59 Temple Place - Suite 330,
  20 # Boston MA 02111-1307, 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 $shift = "";
  34 %defs = ();
  35 $fnno = 1;
  36
  37 while ($_ = shift) {
  38     if (/^-D(.*)$/) {
  39         if ($1 ne "") {
  40             $flag = $1;
  41         } else {
  42             $flag = shift;
  43         }
  44         die "no flag specified for -D\n"
  45             unless $flag ne "";
  46         die "flags may only contain letters, digits, hyphens, and underscores\n"
  47             unless $flag =~ /^[a-zA-Z0-9_-]+$/;
  48         $defs{$flag} = "";
  49     } elsif (/^-/) {
  50         usage();
  51     } else {
  52         $in = $_, next unless defined $in;
  53         $out = $_, next unless defined $out;
  54         usage();
  55     }
  56 }
  57
  58 if (defined $in) {
  59     open(STDIN, $in) or die "opening \"$in\": $!\n";
  60 }
  61 if (defined $out) {
  62     open(STDOUT, ">$out") or die "opening \"$out\": $!\n";
  63 }
  64
  65 while(<STDIN>)
  66 {
  67     # Certain commands are discarded without further processing.
  68     /^\@(?:
  69          [a-z]+index            # @*index: useful only in complete manual
  70          |need                  # @need: useful only in printed manual
  71          |(?:end\s+)?group      # @group .. @end group: ditto
  72          |page                  # @page: ditto
  73          |node                  # @node: useful only in .info file
  74         )\b/x and next;
  75
  76     chomp;
  77
  78     # Look for filename and title markers.
  79     /^\@setfilename\s+([^.]+)/ and $fn = $1, next;
  80     /^\@settitle\s+([^.]+)/ and $tl = $1, next;
  81
  82     # Look for blocks surrounded by @c man begin SECTION ... @c man end.
  83     # This really oughta be @ifman ... @end ifman and the like, but such
  84     # would require rev'ing all other Texinfo translators.
  85     /^\@c man begin ([A-Z]+)/ and $sect = $1, $output = 1, next;
  86     /^\@c man end/ and do {
  87         $sects{$sect} = "" unless exists $sects{$sect};
  88         $sects{$sect} .= postprocess($section);
  89         $section = "";
  90         $output = 0;
  91         next;
  92     };
  93     next unless $output;
  94
  95     # Discard comments.  (Can't do it above, because then we'd never see
  96     # @c man lines.)
  97     /^\@c\b/ and next;
  98
  99     # End-block handler goes up here because it needs to operate even
 100     # if we are skipping.
 101     /^\@end\s+([a-z]+)/ and do {
 102         # Ignore @end foo, where foo is not an operation which may
 103         # cause us to skip, if we are presently skipping.
 104         my $ended = $1;
 105         next if $skipping && $ended !~ /^(?:ifset|ifclear|ignore|menu)$/;
 106
 107         die "\@end $ended without \@$ended at line $.\n" unless defined $endw;
 108         die "\@$endw ended by \@end $ended at line $.\n" unless $ended eq $endw;
 109
 110         $endw = pop @endwstack;
 111
 112         if ($ended =~ /^(?:ifset|ifclear|ignore|menu)$/) {
 113             $skipping = pop @skstack;
 114             next;
 115         } elsif ($ended =~ /^(?:example|smallexample)$/) {
 116             $shift = "";
 117             $_ = "";    # need a paragraph break
 118         } elsif ($ended =~ /^(?:itemize|enumerate|table)$/) {
 119             $_ = "\n=back\n";
 120             $ic = pop @icstack;
 121         } else {
 122             die "unknown command \@end $ended at line $.\n";
 123         }
 124     };
 125
 126     # We must handle commands which can cause skipping even while we
 127     # are skipping, otherwise we will not process nested conditionals
 128     # correctly.
 129     /^\@ifset\s+([a-zA-Z0-9_-]+)/ and do {
 130         push @endwstack, $endw;
 131         push @skstack, $skipping;
 132         $endw = "ifset";
 133         $skipping = 1 unless exists $defs{$1};
 134         next;
 135     };
 136
 137     /^\@ifclear\s+([a-zA-Z0-9_-]+)/ and do {
 138         push @endwstack, $endw;
 139         push @skstack, $skipping;
 140         $endw = "ifclear";
 141         $skipping = 1 if exists $defs{$1};
 142         next;
 143     };
 144
 145     /^\@(ignore|menu)\b/ and do {
 146         push @endwstack, $endw;
 147         push @skstack, $skipping;
 148         $endw = $1;
 149         $skipping = 1;
 150         next;
 151     };
 152
 153     next if $skipping;
 154
 155     # Character entities.  First the ones that can be replaced by raw text
 156     # or discarded outright:
 157     s/\@copyright\{\}/(c)/g;
 158     s/\@dots\{\}/.../g;
 159     s/\@enddots\{\}/..../g;
 160     s/\@([.!? ])/$1/g;
 161     s/\@[:-]//g;
 162     s/\@bullet(?:\{\})?/*/g;
 163     s/\@TeX\{\}/TeX/g;
 164     s/\@pounds\{\}/\#/g;
 165     s/\@minus(?:\{\})?/-/g;
 166     s/\\,/,/g;
 167
 168     # Now the ones that have to be replaced by special escapes
 169     # (which will be turned back into text by unmunge())
 170     s/&/&amp;/g;
 171     s/\@\{/&lbrace;/g;
 172     s/\@\}/&rbrace;/g;
 173     s/\@\@/&at;/g;
 174     # POD doesn't interpret E<> inside a verbatim block.
 175     if ($shift eq "") {
 176         s/</&lt;/g;
 177         s/>/&gt;/g;
 178     } else {
 179         s/</&LT;/g;
 180         s/>/&GT;/g;
 181     }
 182
 183     # Single line command handlers.
 184     /^\@set\s+([a-zA-Z0-9_-]+)\s*(.*)$/ and $defs{$1} = $2, next;
 185     /^\@clear\s+([a-zA-Z0-9_-]+)/ and delete $defs{$1}, next;
 186
 187     /^\@section\s+(.+)$/ and $_ = "\n=head2 $1\n";
 188     /^\@subsection\s+(.+)$/ and $_ = "\n=head3 $1\n";
 189
 190     # Block command handlers:
 191     /^\@itemize\s+(\@[a-z]+|\*|-)/ and do {
 192         push @endwstack, $endw;
 193         push @icstack, $ic;
 194         $ic = $1;
 195         $_ = "\n=over 4\n";
 196         $endw = "itemize";
 197     };
 198
 199     /^\@enumerate(?:\s+([A-Z0-9]+))?/ and do {
 200         push @endwstack, $endw;
 201         push @icstack, $ic;
 202         if (defined $1) {
 203             $ic = $1 . ".";
 204         } else {
 205             $ic = "1.";
 206         }
 207         $_ = "\n=over 4\n";
 208         $endw = "enumerate";
 209     };
 210
 211     /^\@table\s+(\@[a-z]+)/ and do {
 212         push @endwstack, $endw;
 213         push @icstack, $ic;
 214         $ic = $1;
 215         $ic =~ s/\@(?:samp|strong|key|gcctabopt|env)/B/;
 216         $ic =~ s/\@(?:code|kbd)/C/;
 217         $ic =~ s/\@(?:dfn|var|emph|cite|i)/I/;
 218         $ic =~ s/\@(?:file)/F/;
 219         $_ = "\n=over 4\n";
 220         $endw = "table";
 221     };
 222
 223     /^\@((?:small)?example)/ and do {
 224         push @endwstack, $endw;
 225         $endw = $1;
 226         $shift = "\t";
 227         $_ = "";        # need a paragraph break
 228     };
 229
 230     /^\@itemx?\s*(.+)?$/ and do {
 231         if (defined $1) {
 232             # Entity escapes prevent munging by the <> processing below.
 233             $_ = "\n=item $ic\&LT;$1\&GT;\n";
 234         } else {
 235             $_ = "\n=item $ic\n";
 236             $ic =~ y/A-Ya-y1-8/B-Zb-z2-9/;
 237         }
 238     };
 239
 240     $section .= $shift.$_."\n";
 241 }
 242
 243 die "No filename or title\n" unless defined $fn && defined $tl;
 244
 245 $sects{NAME} = "$fn \- $tl\n";
 246 $sects{FOOTNOTES} .= "=back\n" if exists $sects{FOOTNOTES};
 247
 248 for $sect (qw(NAME SYNOPSIS DESCRIPTION OPTIONS ENVIRONMENT FILES
 249               BUGS NOTES FOOTNOTES SEEALSO AUTHOR COPYRIGHT)) {
 250     if(exists $sects{$sect}) {
 251         $head = $sect;
 252         $head =~ s/SEEALSO/SEE ALSO/;
 253         print "=head1 $head\n\n";
 254         print scalar unmunge ($sects{$sect});
 255         print "\n";
 256     }
 257 }
 258
 259 sub usage
 260 {
 261     die "usage: $0 [-D toggle...] [infile [outfile]]\n";
 262 }
 263
 264 sub postprocess
 265 {
 266     local $_ = $_[0];
 267
 268     # @value{foo} is replaced by whatever 'foo' is defined as.
 269     s/\@value\{([a-zA-Z0-9_-]+)\}/$defs{$1}/g;
 270
 271     # Formatting commands.
 272     # Temporary escape for @r.
 273     s/\@r\{([^\}]*)\}/R<$1>/g;
 274     s/\@(?:dfn|var|emph|cite|i)\{([^\}]*)\}/I<$1>/g;
 275     s/\@(?:code|kbd)\{([^\}]*)\}/C<$1>/g;
 276     s/\@(?:gccoptlist|samp|strong|key|option|env|command|b)\{([^\}]*)\}/B<$1>/g;
 277     s/\@sc\{([^\}]*)\}/\U$1/g;
 278     s/\@file\{([^\}]*)\}/F<$1>/g;
 279     s/\@w\{([^\}]*)\}/S<$1>/g;
 280     s/\@(?:dmn|math)\{([^\}]*)\}/$1/g;
 281
 282     # Cross references are thrown away, as are @noindent and @refill.
 283     # (@noindent is impossible in .pod, and @refill is unnecessary.)
 284     # @* is also impossible in .pod; we discard it and any newline that
 285     # follows it.  Similarly, our macro @gol must be discarded.
 286
 287     s/\(?\@xref\{(?:[^\}]*)\}(?:[^.<]|(?:<[^<>]*>))*\.\)?//g;
 288     s/\s+\(\@pxref\{(?:[^\}]*)\}\)//g;
 289     s/;\s+\@pxref\{(?:[^\}]*)\}//g;
 290     s/\@noindent\s*//g;
 291     s/\@refill//g;
 292     s/\@gol//g;
 293     s/\@\*\s*\n?//g;
 294
 295     # @uref can take one, two, or three arguments, with different
 296     # semantics each time.  @url and @email are just like @uref with
 297     # one argument, for our purposes.
 298     s/\@(?:uref|url|email)\{([^\},]*)\}/&lt;B<$1>&gt;/g;
 299     s/\@uref\{([^\},]*),([^\},]*)\}/$2 (C<$1>)/g;
 300     s/\@uref\{([^\},]*),([^\},]*),([^\},]*)\}/$3/g;
 301
 302     # Turn B<blah I<blah> blah> into B<blah> I<blah> B<blah> to
 303     # match Texinfo semantics of @emph inside @samp.  Also handle @r
 304     # inside bold.
 305     s/&LT;/</g;
 306     s/&GT;/>/g;
 307     1 while s/B<((?:[^<>]|I<[^<>]*>)*)R<([^>]*)>/B<$1>${2}B</g;
 308     1 while (s/B<([^<>]*)I<([^>]+)>/B<$1>I<$2>B</g);
 309     1 while (s/I<([^<>]*)B<([^>]+)>/I<$1>B<$2>I</g);
 310     s/[BI]<>//g;
 311     s/([BI])<(\s+)([^>]+)>/$2$1<$3>/g;
 312     s/([BI])<([^>]+?)(\s+)>/$1<$2>$3/g;
 313
 314     # Extract footnotes.  This has to be done after all other
 315     # processing because otherwise the regexp will choke on formatting
 316     # inside @footnote.
 317     while (/\@footnote/g) {
 318         s/\@footnote\{([^\}]+)\}/[$fnno]/;
 319         add_footnote($1, $fnno);
 320         $fnno++;
 321     }
 322
 323     return $_;
 324 }
 325
 326 sub unmunge
 327 {
 328     # Replace escaped symbols with their equivalents.
 329     local $_ = $_[0];
 330
 331     s/&lt;/E<lt>/g;
 332     s/&gt;/E<gt>/g;
 333     s/&lbrace;/\{/g;
 334     s/&rbrace;/\}/g;
 335     s/&at;/\@/g;
 336     s/&amp;/&/g;
 337     return $_;
 338 }
 339
 340 sub add_footnote
 341 {
 342     unless (exists $sects{FOOTNOTES}) {
 343         $sects{FOOTNOTES} = "\n=over 4\n\n";
 344     }
 345
 346     $sects{FOOTNOTES} .= "=item $fnno.\n\n"; $fnno++;
 347     $sects{FOOTNOTES} .= $_[0];
 348     $sects{FOOTNOTES} .= "\n\n";
 349 }
 350