contrib/texi2pod.pl

   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 3, 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 $ibase = "";
  39 @ipath = ();
  40
  41 while ($_ = 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
  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;
  77 }
  78
  79 if (defined $out) {
  80     open(STDOUT, ">$out") or die "opening \"$out\": $!\n";
  81 }
  82
  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;
  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             $ic = pop @icstack;
 165         } elsif ($ended eq "multitable") {
 166             $_ = "\n=back\n";
 167         } else {
 168             die "unknown command \@end $ended at line $.\n";
 169         }
 170     };
 171
 172     # We must handle commands which can cause skipping even while we
 173     # are skipping, otherwise we will not process nested conditionals
 174     # correctly.
 175     /^\@ifset\s+([a-zA-Z0-9_-]+)/ and do {
 176         push @endwstack, $endw;
 177         push @skstack, $skipping;
 178         $endw = "ifset";
 179         $skipping = 1 unless exists $defs{$1};
 180         next;
 181     };
 182
 183     /^\@ifclear\s+([a-zA-Z0-9_-]+)/ and do {
 184         push @endwstack, $endw;
 185         push @skstack, $skipping;
 186         $endw = "ifclear";
 187         $skipping = 1 if exists $defs{$1};
 188         next;
 189     };
 190
 191     /^\@(ignore|menu|iftex|copying)\b/ and do {
 192         push @endwstack, $endw;
 193         push @skstack, $skipping;
 194         $endw = $1;
 195         $skipping = 1;
 196         next;
 197     };
 198
 199     next if $skipping;
 200
 201     # Character entities.  First the ones that can be replaced by raw text
 202     # or discarded outright:
 203     s/\@copyright\{\}/(c)/g;
 204     s/\@dots\{\}/.../g;
 205     s/\@enddots\{\}/..../g;
 206     s/\@([.!? ])/$1/g;
 207     s/\@[:-]//g;
 208     s/\@bullet(?:\{\})?/*/g;
 209     s/\@TeX\{\}/TeX/g;
 210     s/\@pounds\{\}/\#/g;
 211     s/\@minus(?:\{\})?/-/g;
 212     s/\\,/,/g;
 213
 214     # Now the ones that have to be replaced by special escapes
 215     # (which will be turned back into text by unmunge())
 216     s/&/&amp;/g;
 217     s/\@\{/&lbrace;/g;
 218     s/\@\}/&rbrace;/g;
 219     s/\@\@/&at;/g;
 220
 221     # Inside a verbatim block, handle @var, @samp and @url specially.
 222     if ($shift ne "") {
 223         s/\@var\{([^\}]*)\}/<$1>/g;
 224         s/\@samp\{([^\}]*)\}/"$1"/g;
 225         s/\@url\{([^\}]*)\}/<$1>/g;
 226     }
 227
 228     # POD doesn't interpret E<> inside a verbatim block.
 229     if ($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|heading)\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         $ic =~ s/\@(?:asis)//;
 303         $_ = "\n=over 4\n";
 304     };
 305
 306     /^\@((?:small)?example|display)/ and do {
 307         push @endwstack, $endw;
 308         $endw = $1;
 309         $shift = "\t";
 310         $_ = "";        # need a paragraph break
 311     };
 312
 313     /^\@item\s+(.*\S)\s*$/ and $endw eq "multitable" and do {
 314         @columns = ();
 315         for $column (split (/\s*\@tab\s*/, $1)) {
 316             # @strong{...} is used a @headitem work-alike
 317             $column =~ s/^\@strong{(.*)}$/$1/;
 318             push @columns, $column;
 319         }
 320         $_ = "\n=item ".join (" : ", @columns)."\n";
 321     };
 322
 323     /^\@itemx?\s*(.+)?$/ and do {
 324         if (defined $1) {
 325             if ($ic) {
 326                 if ($endw eq "enumerate") {
 327                     $_ = "\n=item $ic $1\n";
 328                     $ic =~ s/(\d+)/$1 + 1/eg;
 329                 } else {
 330                     # Entity escapes prevent munging by the <>
 331                     # processing below.
 332                     $_ = "\n=item $ic\&LT;$1\&GT;\n";
 333                 }
 334             } else {
 335                 $_ = "\n=item $1\n";
 336             }
 337         } else {
 338             $_ = "\n=item $ic\n";
 339             $ic =~ y/A-Ya-y/B-Zb-z/;
 340             $ic =~ s/(\d+)/$1 + 1/eg;
 341         }
 342     };
 343
 344     $section .= $shift.$_."\n";
 345 }
 346 # End of current file.
 347 close($inf);
 348 $inf = pop @instack;
 349 }
 350
 351 die "No filename or title\n" unless defined $fn && defined $tl;
 352
 353 $sects{NAME} = "$fn \- $tl\n";
 354 $sects{FOOTNOTES} .= "=back\n" if exists $sects{FOOTNOTES};
 355
 356 for $sect (qw(NAME SYNOPSIS DESCRIPTION OPTIONS ENVIRONMENT FILES
 357               BUGS NOTES FOOTNOTES SEEALSO AUTHOR COPYRIGHT)) {
 358     if(exists $sects{$sect}) {
 359         $head = $sect;
 360         $head =~ s/SEEALSO/SEE ALSO/;
 361         print "=head1 $head\n\n";
 362         print scalar unmunge ($sects{$sect});
 363         print "\n";
 364     }
 365 }
 366
 367 sub usage
 368 {
 369     die "usage: $0 [-D toggle...] [infile [outfile]]\n";
 370 }
 371
 372 sub postprocess
 373 {
 374     local $_ = $_[0];
 375
 376     # @value{foo} is replaced by whatever 'foo' is defined as.
 377     while (m/(\@value\{([a-zA-Z0-9_-]+)\})/g) {
 378         if (! exists $defs{$2}) {
 379             print STDERR "Option $2 not defined\n";
 380             s/\Q$1\E//;
 381         } else {
 382             $value = $defs{$2};
 383             s/\Q$1\E/$value/;
 384         }
 385     }
 386
 387     # Formatting commands.
 388     # Temporary escape for @r.
 389     s/\@r\{([^\}]*)\}/R<$1>/g;
 390     s/\@(?:dfn|var|emph|cite|i)\{([^\}]*)\}/I<$1>/g;
 391     s/\@(?:code|kbd)\{([^\}]*)\}/C<$1>/g;
 392     s/\@(?:samp|strong|key|option|env|command|b)\{([^\}]*)\}/B<$1>/g;
 393     s/\@sc\{([^\}]*)\}/\U$1/g;
 394     s/\@file\{([^\}]*)\}/F<$1>/g;
 395     s/\@w\{([^\}]*)\}/S<$1>/g;
 396     s/\@(?:dmn|math)\{([^\}]*)\}/$1/g;
 397
 398     # keep references of the form @ref{...}, print them bold
 399     s/\@(?:ref)\{([^\}]*)\}/B<$1>/g;
 400
 401     # Change double single quotes to double quotes.
 402     s/''/"/g;
 403     s/``/"/g;
 404
 405     # Cross references are thrown away, as are @noindent and @refill.
 406     # (@noindent is impossible in .pod, and @refill is unnecessary.)
 407     # @* is also impossible in .pod; we discard it and any newline that
 408     # follows it.  Similarly, our macro @gol must be discarded.
 409
 410     s/\(?\@xref\{(?:[^\}]*)\}(?:[^.<]|(?:<[^<>]*>))*\.\)?//g;
 411     s/\s+\(\@pxref\{(?:[^\}]*)\}\)//g;
 412     s/;\s+\@pxref\{(?:[^\}]*)\}//g;
 413     s/\@noindent\s*//g;
 414     s/\@refill//g;
 415     s/\@gol//g;
 416     s/\@\*\s*\n?//g;
 417
 418     # Anchors are thrown away
 419     s/\@anchor\{(?:[^\}]*)\}//g;
 420
 421     # @uref can take one, two, or three arguments, with different
 422     # semantics each time.  @url and @email are just like @uref with
 423     # one argument, for our purposes.
 424     s/\@(?:uref|url|email)\{([^\},]*)\}/&lt;B<$1>&gt;/g;
 425     s/\@uref\{([^\},]*),([^\},]*)\}/$2 (C<$1>)/g;
 426     s/\@uref\{([^\},]*),([^\},]*),([^\},]*)\}/$3/g;
 427
 428     # Handle gccoptlist here, so it can contain the above formatting
 429     # commands.
 430     s/\@gccoptlist\{([^\}]*)\}/B<$1>/g;
 431
 432     # Un-escape <> at this point.
 433     s/&LT;/</g;
 434     s/&GT;/>/g;
 435
 436     # Now un-nest all B<>, I<>, R<>.  Theoretically we could have
 437     # indefinitely deep nesting; in practice, one level suffices.
 438     1 while s/([BIR])<([^<>]*)([BIR])<([^<>]*)>/$1<$2>$3<$4>$1</g;
 439
 440     # Replace R<...> with bare ...; eliminate empty markup, B<>;
 441     # shift white space at the ends of [BI]<...> expressions outside
 442     # the expression.
 443     s/R<([^<>]*)>/$1/g;
 444     s/[BI]<>//g;
 445     s/([BI])<(\s+)([^>]+)>/$2$1<$3>/g;
 446     s/([BI])<([^>]+?)(\s+)>/$1<$2>$3/g;
 447
 448     # Extract footnotes.  This has to be done after all other
 449     # processing because otherwise the regexp will choke on formatting
 450     # inside @footnote.
 451     while (/\@footnote/g) {
 452         s/\@footnote\{([^\}]+)\}/[$fnno]/;
 453         add_footnote($1, $fnno);
 454         $fnno++;
 455     }
 456
 457     return $_;
 458 }
 459
 460 sub unmunge
 461 {
 462     # Replace escaped symbols with their equivalents.
 463     local $_ = $_[0];
 464
 465     s/&lt;/E<lt>/g;
 466     s/&gt;/E<gt>/g;
 467     s/&lbrace;/\{/g;
 468     s/&rbrace;/\}/g;
 469     s/&at;/\@/g;
 470     s/&amp;/&/g;
 471     return $_;
 472 }
 473
 474 sub add_footnote
 475 {
 476     unless (exists $sects{FOOTNOTES}) {
 477         $sects{FOOTNOTES} = "\n=over 4\n\n";
 478     }
 479
 480     $sects{FOOTNOTES} .= "=item $fnno.\n\n"; $fnno++;
 481     $sects{FOOTNOTES} .= $_[0];
 482     $sects{FOOTNOTES} .= "\n\n";
 483 }
 484
 485 # stolen from Symbol.pm
 486 {
 487     my $genseq = 0;
 488     sub gensym
 489     {
 490         my $name = "GEN" . $genseq++;
 491         my $ref = \*{$name};
 492         delete $::{$name};
 493         return $ref;
 494     }
 495 }