4 use Getopt::Long; # package for handling command-line parameters
10 @EXPORT = qw(pod2html htmlify);
15 use locale; # make \w work right in non-ASCII lands
23 Pod::Html - module to convert pod files to HTML
32 Converts files from pod format (see L<perlpod>) to HTML format. It
33 can automatically generate indexes and cross-references, and it keeps
34 a cache of things it knows how to cross-reference.
38 Pod::Html takes the following arguments:
46 Displays the usage message.
52 Sets the directory in which the resulting HTML file is placed. This
53 is used to generate relative links to other files. Not passing this
54 causes all links to be absolute, since this is the value that tells
55 Pod::Html the root of the documentation tree.
61 Sets the base URL for the HTML files. When cross-references are made,
62 the HTML root is prepended to the URL.
68 Specify the pod file to convert. Input is taken from STDIN if no
75 Specify the HTML file to create. Output goes to STDOUT if no outfile
82 Specify the base directory for finding library pods.
86 --podpath=name:...:name
88 Specify which subdirectories of the podroot contain pod files whose
89 HTML converted forms can be linked-to in cross-references.
93 --libpods=name:...:name
95 List of page names (eg, "perlfunc") which contain linkable C<=item>s.
101 Use Netscape HTML directives when applicable.
107 Do not use Netscape HTML directives (default).
113 Generate an index at the top of the HTML file (default behaviour).
119 Do not generate an index at the top of the HTML file.
126 Recurse into subdirectories specified in podpath (default behaviour).
132 Do not recurse into subdirectories specified in podpath.
138 Specify the title of the resulting HTML file.
144 Specify the URL of a cascading style sheet.
150 Display progress messages.
156 Don't display I<mostly harmless> warning messages.
163 "--podpath=lib:ext:pod:vms",
164 "--podroot=/usr/src/perl",
165 "--htmlroot=/perl/nmanual",
166 "--libpods=perlfunc:perlguts:perlvar:perlrun:perlop",
169 "--outfile=/perl/nmanual/foo.html");
173 Uses $Config{pod2html} to setup default options.
177 Tom Christiansen, E<lt>tchrist@perl.comE<gt>.
185 This program is distributed under the Artistic License.
189 my $cache_ext = $^O eq 'VMS' ? ".tmp" : ".x~~";
190 my $dircache = "pod2htmd$cache_ext";
191 my $itemcache = "pod2htmi$cache_ext";
193 my @begin_stack = (); # begin/end stack
195 my @libpods = (); # files to search for links from C<> directives
196 my $htmlroot = "/"; # http-server base directory from which all
197 # relative paths in $podpath stem.
198 my $htmldir = ""; # The directory to which the html pages
199 # will (eventually) be written.
200 my $htmlfile = ""; # write to stdout by default
201 my $htmlfileurl = "" ; # The url that other files would use to
202 # refer to this file. This is only used
203 # to make relative urls that point to
205 my $podfile = ""; # read from stdin by default
206 my @podpath = (); # list of directories containing library pods.
207 my $podroot = "."; # filesystem base directory from which all
208 # relative paths in $podpath stem.
209 my $css = ''; # Cascading style sheet
210 my $recurse = 1; # recurse on subdirectories in $podpath.
211 my $quiet = 0; # not quiet by default
212 my $verbose = 0; # not verbose by default
213 my $doindex = 1; # non-zero if we should generate an index
214 my $listlevel = 0; # current list depth
215 my @listend = (); # the text to use to end the list.
216 my $after_lpar = 0; # set to true after a par in an =item
217 my $ignore = 1; # whether or not to format text. we don't
218 # format text until we hit our first pod
221 my %items_named = (); # for the multiples of the same item in perlfunc
223 my $netscape = 0; # whether or not to use netscape directives.
224 my $title; # title to give the pod(s)
225 my $header = 0; # produce block header/footer
226 my $top = 1; # true if we are at the top of the doc. used
227 # to prevent the first <HR> directive.
228 my $paragraph; # which paragraph we're processing (used
229 # for error messages)
230 my $ptQuote = 0; # status of double-quote conversion
231 my %pages = (); # associative array used to find the location
232 # of pages referenced by L<> links.
233 my %sections = (); # sections within this page
234 my %items = (); # associative array used to find the location
235 # of =item directives referenced by C<> links
236 my %local_items = (); # local items - avoid destruction of %items
237 my $Is83; # is dos with short filenames (8.3)
240 $dircache = "pod2htmd$cache_ext";
241 $itemcache = "pod2htmi$cache_ext";
243 @begin_stack = (); # begin/end stack
245 @libpods = (); # files to search for links from C<> directives
246 $htmlroot = "/"; # http-server base directory from which all
247 # relative paths in $podpath stem.
248 $htmlfile = ""; # write to stdout by default
249 $podfile = ""; # read from stdin by default
250 @podpath = (); # list of directories containing library pods.
251 $podroot = "."; # filesystem base directory from which all
252 # relative paths in $podpath stem.
253 $css = ''; # Cascading style sheet
254 $recurse = 1; # recurse on subdirectories in $podpath.
255 $quiet = 0; # not quiet by default
256 $verbose = 0; # not verbose by default
257 $doindex = 1; # non-zero if we should generate an index
258 $listlevel = 0; # current list depth
259 @listend = (); # the text to use to end the list.
260 $after_lpar = 0; # set to true after a par in an =item
261 $ignore = 1; # whether or not to format text. we don't
262 # format text until we hit our first pod
267 $netscape = 0; # whether or not to use netscape directives.
268 $header = 0; # produce block header/footer
269 $title = ''; # title to give the pod(s)
270 $top = 1; # true if we are at the top of the doc. used
271 # to prevent the first <HR> directive.
272 $paragraph = ''; # which paragraph we're processing (used
273 # for error messages)
274 %sections = (); # sections within this page
276 # These are not reinitialised here but are kept as a cache.
277 # See get_cache and related cache management code.
278 #%pages = (); # associative array used to find the location
279 # of pages referenced by L<> links.
280 #%items = (); # associative array used to find the location
281 # of =item directives referenced by C<> links
287 # clean_data: global clean-up of pod data
292 for( $i = 0; $i <= $#$dataref; $i++ ){
293 ${$dataref}[$i] =~ s/\s+\Z//;
295 # have a look for all-space lines
296 if( ${$dataref}[$i] =~ /^\s+$/m ){
297 my @chunks = split( /^\s+$/m, ${$dataref}[$i] );
298 splice( @$dataref, $i, 1, @chunks );
311 $Is83 = 0 if (defined (&Dos::UseLFN) && Dos::UseLFN());
313 # cache of %pages and %items from last time we ran pod2html
315 #undef $opt_help if defined $opt_help;
317 # parse the command-line parameters
318 parse_command_line();
320 # set some variables to their default values if necessary
322 unless (@ARGV && $ARGV[0]) {
323 $podfile = "-" unless $podfile; # stdin
324 open(POD, "<$podfile")
325 || die "$0: cannot open $podfile file for input: $!\n";
327 $podfile = $ARGV[0]; # XXX: might be more filenames
330 $htmlfile = "-" unless $htmlfile; # stdout
331 $htmlroot = "" if $htmlroot eq "/"; # so we don't get a //
332 $htmldir =~ s#/$## ; # so we don't get a //
334 && defined( $htmldir )
336 && substr( $htmlfile, 0, length( $htmldir ) ) eq $htmldir
339 # Set the 'base' url for this file, so that we can use it
340 # as the location from which to calculate relative links
341 # to other files. If this is '', then absolute links will
342 # be used throughout.
343 $htmlfileurl= "$htmldir/" . substr( $htmlfile, length( $htmldir ) + 1);
346 # read the pod a paragraph at a time
347 warn "Scanning for sections in input file(s)\n" if $verbose;
351 clean_data( \@poddata );
353 # scan the pod for =head[1-6] directives and build an index
354 my $index = scan_headings(\%sections, @poddata);
357 warn "No headings in $podfile\n" if $verbose;
360 # open the output file
361 open(HTML, ">$htmlfile")
362 || die "$0: cannot open $htmlfile file for output: $!\n";
364 # put a title in the HTML file if one wasn't specified
367 for (my $i = 0; $i < @poddata; $i++) {
368 if ($poddata[$i] =~ /^=head1\s*NAME\b/m) {
369 for my $para ( @poddata[$i, $i+1] ) {
371 if ($title) = $para =~ /(\S+\s+-+.*\S)/s;
378 if (!$title and $podfile =~ /\.pod$/) {
379 # probably a split pod so take first =head[12] as title
380 for (my $i = 0; $i < @poddata; $i++) {
381 last if ($title) = $poddata[$i] =~ /^=head[12]\s*(.*)/;
383 warn "adopted '$title' as title for $podfile\n"
384 if $verbose and $title;
387 $title =~ s/\s*\(.*\)//;
389 warn "$0: no title for $podfile" unless $quiet;
390 $podfile =~ /^(.*)(\.[^.\/]+)?$/;
391 $title = ($podfile eq "-" ? 'No Title' : $1);
392 warn "using $title" if $verbose;
394 my $csslink = $css ? qq(\n<LINK REL="stylesheet" HREF="$css" TYPE="text/css">) : '';
395 $csslink =~ s,\\,/,g;
396 $csslink =~ s,(/.):,$1|,;
398 my $block = $header ? <<END_OF_BLOCK : '';
399 <TABLE BORDER=0 CELLPADDING=0 CELLSPACING=0 WIDTH=100%>
400 <TR><TD CLASS=block VALIGN=MIDDLE WIDTH=100% BGCOLOR="#cccccc">
401 <FONT SIZE=+1><STRONG><P CLASS=block> $title</P></STRONG></FONT>
406 print HTML <<END_OF_HEAD;
409 <TITLE>$title</TITLE>$csslink
410 <LINK REV="made" HREF="mailto:$Config{perladmin}">
417 # load/reload/validate/cache %pages and %items
418 get_cache($dircache, $itemcache, \@podpath, $podroot, $recurse);
420 # scan the pod for =item directives
421 scan_items( \%local_items, "", @poddata);
423 # put an index at the top of the file. note, if $doindex is 0 we
424 # still generate an index, but surround it with an html comment.
425 # that way some other program can extract it if desired.
427 print HTML "<A NAME=\"__index__\"></A>\n";
428 print HTML "<!-- INDEX BEGIN -->\n";
429 print HTML "<!--\n" unless $doindex;
431 print HTML "-->\n" unless $doindex;
432 print HTML "<!-- INDEX END -->\n\n";
433 print HTML "<HR>\n" if $doindex and $index;
435 # now convert this file
436 my $after_item; # set to true after an =item
437 warn "Converting input file $podfile\n" if $verbose;
438 foreach my $i (0..$#poddata){
439 $ptQuote = 0; # status of quote conversion
443 if (/^(=.*)/s) { # is it a pod directive?
447 if (/^=begin\s+(\S+)\s*(.*)/si) {# =begin
448 process_begin($1, $2);
449 } elsif (/^=end\s+(\S+)\s*(.*)/si) {# =end
451 } elsif (/^=cut/) { # =cut
453 } elsif (/^=pod/) { # =pod
456 next if @begin_stack && $begin_stack[-1] ne 'html';
458 if (/^=(head[1-6])\s+(.*\S)/s) { # =head[1-6] heading
459 process_head( $1, $2, $doindex && $index );
460 } elsif (/^=item\s*(.*\S)?/sm) { # =item text
461 warn "$0: $podfile: =item without bullet, number or text"
462 . " in paragraph $paragraph.\n" if $1 eq '';
465 } elsif (/^=over\s*(.*)/) { # =over N
467 } elsif (/^=back/) { # =back
469 } elsif (/^=for\s+(\S+)\s+(.*)/si) {# =for
473 warn "$0: $podfile: unknown pod directive '$1' in "
474 . "paragraph $paragraph. ignoring.\n";
481 next if @begin_stack && $begin_stack[-1] ne 'html';
483 if( $text =~ /\A\s+/ ){
484 process_pre( \$text );
485 print HTML "<PRE>\n$text</PRE>\n";
488 process_text( \$text );
490 # experimental: check for a paragraph where all lines
491 # have some ...\t...\t...\n pattern
493 my @lines = split( "\n", $text );
496 foreach my $line ( @lines ){
497 if( $line =~ /\S/ && $line !~ /\t/ ){
503 $text =~ s/\t+/<TD>/g;
504 $text =~ s/^/<TR><TD>/gm;
505 $text = '<TABLE CELLSPACING=0 CELLPADDING=0>' .
510 ## end of experimental
513 print HTML "$text\n";
516 print HTML "<P>$text</P>\n";
523 # finish off any pending directives
527 print HTML "<P><A HREF=\"#__index__\"><SMALL>page index</SMALL></A></P>\n"
528 if $doindex and $index;
530 print HTML <<END_OF_TAIL;
537 # close the html file
540 warn "Finished\n" if $verbose;
543 ##############################################################################
545 my $usage; # see below
548 warn "$0: $podfile: @_\n" if @_;
552 $usage =<<END_OF_USAGE;
553 Usage: $0 --help --htmlroot=<name> --infile=<name> --outfile=<name>
554 --podpath=<name>:...:<name> --podroot=<name>
555 --libpods=<name>:...:<name> --recurse --verbose --index
556 --netscape --norecurse --noindex
558 --flush - flushes the item and directory caches.
559 --help - prints this message.
560 --htmlroot - http-server base directory from which all relative paths
561 in podpath stem (default is /).
562 --index - generate an index at the top of the resulting html
564 --infile - filename for the pod to convert (input taken from stdin
566 --libpods - colon-separated list of pages to search for =item pod
567 directives in as targets of C<> and implicit links (empty
568 by default). note, these are not filenames, but rather
569 page names like those that appear in L<> links.
570 --netscape - will use netscape html directives when applicable.
571 --nonetscape - will not use netscape directives (default).
572 --outfile - filename for the resulting html file (output sent to
574 --podpath - colon-separated list of directories containing library
575 pods. empty by default.
576 --podroot - filesystem base directory from which all relative paths
577 in podpath stem (default is .).
578 --noindex - don't generate an index at the top of the resulting html.
579 --norecurse - don't recurse on those subdirectories listed in podpath.
580 --recurse - recurse on those subdirectories listed in podpath
582 --title - title that will appear in resulting html file.
583 --header - produce block header/footer
584 --css - stylesheet URL
585 --verbose - self-explanatory
586 --quiet - supress some benign warning messages
590 sub parse_command_line {
591 my ($opt_flush,$opt_help,$opt_htmldir,$opt_htmlroot,$opt_index,$opt_infile,$opt_libpods,$opt_netscape,$opt_outfile,$opt_podpath,$opt_podroot,$opt_recurse,$opt_title,$opt_verbose,$opt_css,$opt_header,$opt_quiet);
592 unshift @ARGV, split ' ', $Config{pod2html} if $Config{pod2html};
593 my $result = GetOptions(
594 'flush' => \$opt_flush,
595 'help' => \$opt_help,
596 'htmldir=s' => \$opt_htmldir,
597 'htmlroot=s' => \$opt_htmlroot,
598 'index!' => \$opt_index,
599 'infile=s' => \$opt_infile,
600 'libpods=s' => \$opt_libpods,
601 'netscape!' => \$opt_netscape,
602 'outfile=s' => \$opt_outfile,
603 'podpath=s' => \$opt_podpath,
604 'podroot=s' => \$opt_podroot,
605 'recurse!' => \$opt_recurse,
606 'title=s' => \$opt_title,
607 'header' => \$opt_header,
608 'css=s' => \$opt_css,
609 'verbose' => \$opt_verbose,
610 'quiet' => \$opt_quiet,
612 usage("-", "invalid parameters") if not $result;
614 usage("-") if defined $opt_help; # see if the user asked for help
615 $opt_help = ""; # just to make -w shut-up.
617 $podfile = $opt_infile if defined $opt_infile;
618 $htmlfile = $opt_outfile if defined $opt_outfile;
619 $htmldir = $opt_htmldir if defined $opt_outfile;
621 @podpath = split(":", $opt_podpath) if defined $opt_podpath;
622 @libpods = split(":", $opt_libpods) if defined $opt_libpods;
624 warn "Flushing item and directory caches\n"
625 if $opt_verbose && defined $opt_flush;
626 unlink($dircache, $itemcache) if defined $opt_flush;
628 $htmlroot = $opt_htmlroot if defined $opt_htmlroot;
629 $podroot = $opt_podroot if defined $opt_podroot;
631 $doindex = $opt_index if defined $opt_index;
632 $recurse = $opt_recurse if defined $opt_recurse;
633 $title = $opt_title if defined $opt_title;
634 $header = defined $opt_header ? 1 : 0;
635 $css = $opt_css if defined $opt_css;
636 $verbose = defined $opt_verbose ? 1 : 0;
637 $quiet = defined $opt_quiet ? 1 : 0;
638 $netscape = $opt_netscape if defined $opt_netscape;
645 my($dircache, $itemcache, $podpath, $podroot, $recurse) = @_;
646 my @cache_key_args = @_;
648 # A first-level cache:
649 # Don't bother reading the cache files if they still apply
650 # and haven't changed since we last read them.
652 my $this_cache_key = cache_key(@cache_key_args);
654 return if $saved_cache_key and $this_cache_key eq $saved_cache_key;
656 # load the cache of %pages and %items if possible. $tests will be
657 # non-zero if successful.
659 if (-f $dircache && -f $itemcache) {
660 warn "scanning for item cache\n" if $verbose;
661 $tests = load_cache($dircache, $itemcache, $podpath, $podroot);
664 # if we didn't succeed in loading the cache then we must (re)build
667 warn "scanning directories in pod-path\n" if $verbose;
668 scan_podpath($podroot, $recurse, 0);
670 $saved_cache_key = cache_key(@cache_key_args);
674 my($dircache, $itemcache, $podpath, $podroot, $recurse) = @_;
675 return join('!', $dircache, $itemcache, $recurse,
676 @$podpath, $podroot, stat($dircache), stat($itemcache));
680 # load_cache - tries to find if the caches stored in $dircache and $itemcache
681 # are valid caches of %pages and %items. if they are valid then it loads
682 # them and returns a non-zero value.
685 my($dircache, $itemcache, $podpath, $podroot) = @_;
691 open(CACHE, "<$itemcache") ||
692 die "$0: error opening $itemcache for reading: $!\n";
695 # is it the same podpath?
698 $tests++ if (join(":", @$podpath) eq $_);
700 # is it the same podroot?
703 $tests++ if ($podroot eq $_);
705 # load the cache if its good
711 warn "loading item cache\n" if $verbose;
718 warn "scanning for directory cache\n" if $verbose;
719 open(CACHE, "<$dircache") ||
720 die "$0: error opening $dircache for reading: $!\n";
724 # is it the same podpath?
727 $tests++ if (join(":", @$podpath) eq $_);
729 # is it the same podroot?
732 $tests++ if ($podroot eq $_);
734 # load the cache if its good
740 warn "loading directory cache\n" if $verbose;
752 # scan_podpath - scans the directories specified in @podpath for directories,
753 # .pod files, and .pm files. it also scans the pod files specified in
754 # @libpods for =item directives.
757 my($podroot, $recurse, $append) = @_;
759 my($libpod, $dirname, $pod, @files, @poddata);
766 # scan each directory listed in @podpath
769 || die "$0: error changing to directory $podroot: $!\n";
770 foreach $dir (@podpath) {
771 scan_dir($dir, $recurse);
774 # scan the pods listed in @libpods for =item directives
775 foreach $libpod (@libpods) {
776 # if the page isn't defined then we won't know where to find it
778 next unless defined $pages{$libpod} && $pages{$libpod};
780 # if there is a directory then use the .pod and .pm files within it.
781 # NOTE: Only finds the first so-named directory in the tree.
782 # if ($pages{$libpod} =~ /([^:]*[^(\.pod|\.pm)]):/) {
783 if ($pages{$libpod} =~ /([^:]*(?<!\.pod)(?<!\.pm)):/) {
784 # find all the .pod and .pm files within the directory
786 opendir(DIR, $dirname) ||
787 die "$0: error opening directory $dirname: $!\n";
788 @files = grep(/(\.pod|\.pm)$/ && ! -d $_, readdir(DIR));
791 # scan each .pod and .pm file for =item directives
792 foreach $pod (@files) {
793 open(POD, "<$dirname/$pod") ||
794 die "$0: error opening $dirname/$pod for input: $!\n";
797 clean_data( \@poddata );
799 scan_items( \%items, "$dirname/$pod", @poddata);
802 # use the names of files as =item directives too.
803 ### Don't think this should be done this way - confuses issues.(WL)
804 ### foreach $pod (@files) {
805 ### $pod =~ /^(.*)(\.pod|\.pm)$/;
806 ### $items{$1} = "$dirname/$1.html" if $1;
808 } elsif ($pages{$libpod} =~ /([^:]*\.pod):/ ||
809 $pages{$libpod} =~ /([^:]*\.pm):/) {
810 # scan the .pod or .pm file for =item directives
812 open(POD, "<$pod") ||
813 die "$0: error opening $pod for input: $!\n";
816 clean_data( \@poddata );
818 scan_items( \%items, "$pod", @poddata);
820 warn "$0: shouldn't be here (line ".__LINE__."\n";
823 @poddata = (); # clean-up a bit
826 || die "$0: error changing to directory $pwd: $!\n";
828 # cache the item list for later use
829 warn "caching items for later use\n" if $verbose;
830 open(CACHE, ">$itemcache") ||
831 die "$0: error open $itemcache for writing: $!\n";
833 print CACHE join(":", @podpath) . "\n$podroot\n";
834 foreach my $key (keys %items) {
835 print CACHE "$key $items{$key}\n";
840 # cache the directory list for later use
841 warn "caching directories for later use\n" if $verbose;
842 open(CACHE, ">$dircache") ||
843 die "$0: error open $dircache for writing: $!\n";
845 print CACHE join(":", @podpath) . "\n$podroot\n";
846 foreach my $key (keys %pages) {
847 print CACHE "$key $pages{$key}\n";
854 # scan_dir - scans the directory specified in $dir for subdirectories, .pod
855 # files, and .pm files. notes those that it finds. this information will
856 # be used later in order to figure out where the pages specified in L<>
857 # links are on the filesystem.
860 my($dir, $recurse) = @_;
861 my($t, @subdirs, @pods, $pod, $dirname, @dirs);
867 opendir(DIR, $dir) ||
868 die "$0: error opening directory $dir: $!\n";
869 while (defined($_ = readdir(DIR))) {
870 if (-d "$dir/$_" && $_ ne "." && $_ ne "..") { # directory
871 $pages{$_} = "" unless defined $pages{$_};
872 $pages{$_} .= "$dir/$_:";
874 } elsif (/\.pod$/) { # .pod
876 $pages{$_} = "" unless defined $pages{$_};
877 $pages{$_} .= "$dir/$_.pod:";
878 push(@pods, "$dir/$_.pod");
879 } elsif (/\.pm$/) { # .pm
881 $pages{$_} = "" unless defined $pages{$_};
882 $pages{$_} .= "$dir/$_.pm:";
883 push(@pods, "$dir/$_.pm");
888 # recurse on the subdirectories if necessary
890 foreach my $subdir (@subdirs) {
891 scan_dir("$dir/$subdir", $recurse);
897 # scan_headings - scan a pod file for head[1-6] tags, note the tags, and
901 my($sections, @data) = @_;
902 my($tag, $which_head, $otitle, $listdepth, $index);
904 # here we need local $ignore = 0;
905 # unfortunately, we can't have it, because $ignore is lexical
911 # scan for =head directives, note their name, and build an index
912 # pointing to each of them.
913 foreach my $line (@data) {
914 if ($line =~ /^=(head)([1-6])\s+(.*)/) {
915 ($tag, $which_head, $otitle) = ($1,$2,$3);
917 my $title = depod( $otitle );
918 my $name = htmlify( $title );
919 $$sections{$name} = 1;
920 $title = process_text( \$otitle );
922 while ($which_head != $listdepth) {
923 if ($which_head > $listdepth) {
924 $index .= "\n" . ("\t" x $listdepth) . "<UL>\n";
926 } elsif ($which_head < $listdepth) {
928 $index .= "\n" . ("\t" x $listdepth) . "</UL>\n";
932 $index .= "\n" . ("\t" x $listdepth) . "<LI>" .
933 "<A HREF=\"#" . $name . "\">" .
934 $title . "</A></LI>";
938 # finish off the lists
939 while ($listdepth--) {
940 $index .= "\n" . ("\t" x $listdepth) . "</UL>\n";
943 # get rid of bogus lists
944 $index =~ s,\t*<UL>\s*</UL>\n,,g;
946 $ignore = 1; # restore old value;
952 # scan_items - scans the pod specified by $pod for =item directives. we
953 # will use this information later on in resolving C<> links.
956 my( $itemref, $pod, @poddata ) = @_;
961 $pod .= ".html" if $pod;
963 foreach $i (0..$#poddata) {
964 my $txt = depod( $poddata[$i] );
966 # figure out what kind of item it is.
967 # Build string for referencing this item.
968 if ( $txt =~ /\A=item\s+\*\s*(.*)\Z/s ) { # bullet
971 } elsif( $txt =~ /\A=item\s+(?>\d+\.?)\s*(.*)\Z/s ) { # numbered list
973 } elsif( $txt =~ /\A=item\s+(.*)\Z/s ) { # plain item
978 my $fid = fragment_id( $item );
979 $$itemref{$fid} = "$pod" if $fid;
984 # process_head - convert a pod head[1-6] tag and convert it to HTML format.
987 my($tag, $heading, $hasindex) = @_;
989 # figure out the level of the =head
990 $tag =~ /head([1-6])/;
994 warn "$0: $podfile: unterminated list at =head in paragraph $paragraph. ignoring.\n";
1001 if( $level == 1 && ! $top ){
1002 print HTML "<A HREF=\"#__index__\"><SMALL>page index</SMALL></A>\n"
1007 my $name = htmlify( depod( $heading ) );
1008 my $convert = process_text( \$heading );
1009 print HTML "<H$level><A NAME=\"$name\">$convert</A></H$level>\n";
1014 # emit_item_tag - print an =item's text
1015 # Note: The global $EmittedItem is used for inhibiting self-references.
1019 sub emit_item_tag($$$){
1020 my( $otext, $text, $compact ) = @_;
1021 my $item = fragment_id( $text );
1023 $EmittedItem = $item;
1024 ### print STDERR "emit_item_tag=$item ($text)\n";
1026 print HTML '<STRONG>';
1027 if ($items_named{$item}++) {
1028 print HTML process_text( \$otext );
1030 my $name = 'item_' . $item;
1031 print HTML qq{<A NAME="$name">}, process_text( \$otext ), '</A>';
1033 print HTML "</STRONG><BR>\n";
1034 undef( $EmittedItem );
1039 if( $items_seen[$listlevel]++ == 0 ){
1040 push( @listend, "</$tag>" );
1041 print HTML "<$tag>\n";
1043 print HTML $tag eq 'DL' ? '<DT>' : '<LI>';
1047 # process_item - convert a pod item tag and convert it to HTML format.
1052 # lots of documents start a list without doing an =over. this is
1053 # bad! but, the proper thing to do seems to be to just assume
1054 # they did do an =over. so warn them once and then continue.
1055 if( $listlevel == 0 ){
1056 warn "$0: $podfile: unexpected =item directive in paragraph $paragraph. ignoring.\n";
1060 # formatting: insert a paragraph if preceding item has >1 paragraph
1062 print HTML "<P></P>\n";
1066 # remove formatting instructions from the text
1067 my $text = depod( $otext );
1069 # all the list variants:
1070 if( $text =~ /\A\*/ ){ # bullet
1072 if ($text =~ /\A\*\s+(.+)\Z/s ) { # with additional text
1074 $otext =~ s/\A\*\s+//;
1075 emit_item_tag( $otext, $tag, 1 );
1078 } elsif( $text =~ /\A\d+/ ){ # numbered list
1080 if ($text =~ /\A(?>\d+\.?)\s*(.+)\Z/s ) { # with additional text
1082 $otext =~ s/\A\d+\.?\s*//;
1083 emit_item_tag( $otext, $tag, 1 );
1086 } else { # definition list
1088 if ($text =~ /\A(.+)\Z/s ){ # should have text
1089 emit_item_tag( $otext, $text, 1 );
1097 # process_over - process a pod over tag and start a corresponding HTML list.
1102 push( @items_seen, 0 );
1107 # process_back - process a pod back tag and convert it to HTML format.
1110 if( $listlevel == 0 ){
1111 warn "$0: $podfile: unexpected =back directive in paragraph $paragraph. ignoring.\n";
1115 # close off the list. note, I check to see if $listend[$listlevel] is
1116 # defined because an =item directive may have never appeared and thus
1117 # $listend[$listlevel] may have never been initialized.
1119 if( defined $listend[$listlevel] ){
1120 print HTML '<P></P>' if $after_lpar;
1121 print HTML $listend[$listlevel];
1127 # clean up item count
1132 # process_cut - process a pod cut tag, thus start ignoring pod directives.
1139 # process_pod - process a pod pod tag, thus stop ignoring pod directives
1140 # until we see a corresponding cut.
1143 # no need to set $ignore to 0 cause the main loop did it
1147 # process_for - process a =for pod tag. if it's for html, spit
1148 # it out verbatim, if illustration, center it, otherwise ignore it.
1151 my($whom, $text) = @_;
1152 if ( $whom =~ /^(pod2)?html$/i) {
1154 } elsif ($whom =~ /^illustration$/i) {
1155 1 while chomp $text;
1156 for my $ext (qw[.png .gif .jpeg .jpg .tga .pcl .bmp]) {
1157 $text .= $ext, last if -r "$text$ext";
1159 print HTML qq{<p align = "center"><img src = "$text" alt = "$text illustration"></p>};
1164 # process_begin - process a =begin pod tag. this pushes
1165 # whom we're beginning on the begin stack. if there's a
1166 # begin stack, we only print if it us.
1169 my($whom, $text) = @_;
1171 push (@begin_stack, $whom);
1172 if ( $whom =~ /^(pod2)?html$/) {
1173 print HTML $text if $text;
1178 # process_end - process a =end pod tag. pop the
1179 # begin stack. die if we're mismatched.
1182 my($whom, $text) = @_;
1184 if ($begin_stack[-1] ne $whom ) {
1185 die "Unmatched begin/end at chunk $paragraph\n"
1187 pop( @begin_stack );
1191 # process_pre - indented paragraph, made into <PRE></PRE>
1200 # insert spaces in place of tabs
1203 1 while $line =~ s/\t+/' ' x (length($&) * 8 - length($`) % 8)/e;
1207 # convert some special chars to HTML escapes
1208 $rest =~ s/&/&/g;
1209 $rest =~ s/</</g;
1210 $rest =~ s/>/>/g;
1211 $rest =~ s/"/"/g;
1213 # try and create links for all occurrences of perl.* within
1214 # the preformatted text.
1218 if ( defined $pages{$2} ){ # is a link
1219 qq($1<A HREF="$htmlroot/$pages{$2}">$2</A>);
1220 } elsif (defined $pages{dosify($2)}) { # is a link
1221 qq($1<A HREF="$htmlroot/$pages{dosify($2)}">$2</A>);
1227 (<A\ HREF="?) ([^>:]*:)? ([^>:]*) \.pod: ([^>:]*:)?
1230 if ( $htmlfileurl ne '' ){
1231 # Here, we take advantage of the knowledge
1232 # that $htmlfileurl ne '' implies $htmlroot eq ''.
1233 # Since $htmlroot eq '', we need to prepend $htmldir
1234 # on the fron of the link to get the absolute path
1235 # of the link's target. We check for a leading '/'
1236 # to avoid corrupting links that are #, file:, etc.
1238 $old_url = "$htmldir$old_url" if $old_url =~ m{^\/};
1239 $url = relativize_url( "$old_url.html", $htmlfileurl );
1246 # Look for embedded URLs and make them into links. We don't
1247 # relativize them since they are best left as the author intended.
1249 my $urls = '(' . join ('|', qw{
1262 my $gunk = '/#~:.?+=&%@!\-';
1264 my $any = "${ltrs}${gunk}${punc}";
1267 \b # start at word boundary
1269 $urls : # need resource and a colon
1270 (?!:) # Ignore File::, among others.
1271 [$any] +? # followed by on or more
1272 # of any valid character, but
1273 # be conservative and take only
1274 # what you need to....
1276 (?= # look-ahead non-consumptive assertion
1277 [$punc]* # either 0 or more puntuation
1278 [^$any] # followed by a non-url char
1280 $ # then end of the string
1282 }{<A HREF="$1">$1</A>}igox;
1284 # text should be as it is (verbatim)
1290 # pure text processing
1292 # pure_text/inIS_text: differ with respect to automatic C<> recognition.
1293 # we don't want this to happen within IS
1297 process_puretext( $text, \$ptQuote, 1 );
1302 process_puretext( $text, \$ptQuote, 0 );
1306 # process_puretext - process pure text (without pod-escapes) converting
1307 # double-quotes and handling implicit C<> links.
1309 sub process_puretext {
1310 my($text, $quote, $notinIS) = @_;
1312 ## Guessing at func() or [$@%&]*var references in plain text is destined
1313 ## to produce some strange looking ref's. uncomment to disable:
1316 my(@words, $lead, $trail);
1318 # convert double-quotes to single-quotes
1319 if( $$quote && $text =~ s/"/''/s ){
1322 while ($text =~ s/"([^"]*)"/``$1''/sg) {};
1323 $$quote = 1 if $text =~ s/"/``/s;
1325 # keep track of leading and trailing white-space
1326 $lead = ($text =~ s/\A(\s+)//s ? $1 : "");
1327 $trail = ($text =~ s/(\s+)\Z//s ? $1 : "");
1329 # split at space/non-space boundaries
1330 @words = split( /(?<=\s)(?=\S)|(?<=\S)(?=\s)/, $text );
1332 # process each word individually
1333 foreach my $word (@words) {
1335 next if $word =~ /^\s*$/;
1336 # see if we can infer a link
1337 if( $notinIS && $word =~ s/^(\w+)\((.*)\)\W*$/$1/ ) {
1338 # has parenthesis so should have been a C<> ref
1339 ## try for a pagename (perlXXX(1))?
1340 if( $2 =~ /^\d+$/ ){
1341 my $url = page_sect( $word, '' );
1343 $word = "<A HREF=\"$url\">the $word manpage</A>";
1347 $word = emit_C( $word );
1349 #### disabled. either all (including $\W, $\w+{.*} etc.) or nothing.
1350 ## } elsif( $notinIS && $word =~ /^[\$\@%&*]+\w+$/) {
1351 ## # perl variables, should be a C<> ref
1352 ## $word = emit_C( $word );
1354 } elsif ($word =~ m,^\w+://\w,) {
1356 # Don't relativize it: leave it as the author intended
1357 $word = qq(<A HREF="$word">$word</A>);
1358 } elsif ($word =~ /[\w.-]+\@[\w-]+\.\w/) {
1359 # looks like an e-mail address
1360 my ($w1, $w2, $w3) = ("", $word, "");
1361 ($w1, $w2, $w3) = ("(", $1, ")$2") if $word =~ /^\((.*?)\)(,?)/;
1362 ($w1, $w2, $w3) = ("<", $1, ">$2") if $word =~ /^<(.*?)>(,?)/;
1363 $word = qq($w1<A HREF="mailto:$w2">$w2</A>$w3);
1364 } elsif ($word !~ /[a-z]/ && $word =~ /[A-Z]/) { # all uppercase?
1365 $word = html_escape($word) if $word =~ /["&<>]/;
1366 $word = "\n<FONT SIZE=-1>$word</FONT>" if $netscape;
1368 $word = html_escape($word) if $word =~ /["&<>]/;
1372 # put everything back together
1373 return $lead . join( '', @words ) . $trail;
1378 # process_text - handles plaintext that appears in the input pod file.
1379 # there may be pod commands embedded within the text so those must be
1380 # converted to html commands.
1385 my $res = process_text1( 0, $tref );
1389 sub process_text1($$;$){
1390 my( $lev, $rstr, $func ) = @_;
1391 $lev++ unless defined $func;
1395 # B<text> - boldface
1396 $res = '<STRONG>' . process_text1( $lev, $rstr ) . '</STRONG>';
1398 } elsif( $func eq 'C' ){
1399 # C<code> - can be a ref or <CODE></CODE>
1400 # need to extract text
1401 my $par = go_ahead( $rstr, 'C' );
1403 ## clean-up of the link target
1404 my $text = depod( $par );
1406 ### my $x = $par =~ /[BI]</ ? 'yes' : 'no' ;
1407 ### print STDERR "-->call emit_C($par) lev=$lev, par with BI=$x\n";
1409 $res = emit_C( $text, $lev > 1 || ($par =~ /[BI]</) );
1411 } elsif( $func eq 'E' ){
1412 # E<x> - convert to character
1413 $$rstr =~ s/^(\w+)>//;
1416 } elsif( $func eq 'F' ){
1417 # F<filename> - italizice
1418 $res = '<EM>' . process_text1( $lev, $rstr ) . '</EM>';
1420 } elsif( $func eq 'I' ){
1421 # I<text> - italizice
1422 $res = '<EM>' . process_text1( $lev, $rstr ) . '</EM>';
1424 } elsif( $func eq 'L' ){
1426 ## L<text|cross-ref> => produce text, use cross-ref for linking
1427 ## L<cross-ref> => make text from cross-ref
1428 ## need to extract text
1429 my $par = go_ahead( $rstr, 'L' );
1431 # some L<>'s that shouldn't be:
1432 # a) full-blown URL's are emitted as-is
1433 if( $par =~ m{^\w+://}s ){
1434 return make_URL_href( $par );
1436 # b) C<...> is stripped and treated as C<>
1437 if( $par =~ /^C<(.*)>$/ ){
1438 my $text = depod( $1 );
1439 return emit_C( $text, $lev > 1 || ($par =~ /[BI]</) );
1442 # analyze the contents
1443 $par =~ s/\n/ /g; # undo word-wrapped tags
1446 if( $par =~ s{^([^|]+)\|}{} ){
1450 # make sure sections start with a /
1453 my( $page, $section, $ident );
1455 # check for link patterns
1456 if( $par =~ m{^([^/]+?)/(?!")(.*?)$} ){ # name/ident
1457 # we've got a name/ident (no quotes)
1458 ( $page, $ident ) = ( $1, $2 );
1459 ### print STDERR "--> L<$par> to page $page, ident $ident\n";
1461 } elsif( $par =~ m{^(.*?)/"?(.*?)"?$} ){ # [name]/"section"
1462 # even though this should be a "section", we go for ident first
1463 ( $page, $ident ) = ( $1, $2 );
1464 ### print STDERR "--> L<$par> to page $page, section $section\n";
1466 } elsif( $par =~ /\s/ ){ # this must be a section with missing quotes
1467 ( $page, $section ) = ( '', $par );
1468 ### print STDERR "--> L<$par> to void page, section $section\n";
1471 ( $page, $section ) = ( $par, '' );
1472 ### print STDERR "--> L<$par> to page $par, void section\n";
1475 # now, either $section or $ident is defined. the convoluted logic
1476 # below tries to resolve L<> according to what the user specified.
1477 # failing this, we try to find the next best thing...
1478 my( $url, $ltext, $fid );
1481 if( defined $ident ){
1482 ## try to resolve $ident as an item
1483 ( $url, $fid ) = coderef( $page, $ident );
1485 if( ! defined( $linktext ) ){
1487 $linktext .= " in " if $ident && $page;
1488 $linktext .= "the $page manpage" if $page;
1490 ### print STDERR "got coderef url=$url\n";
1493 ## no luck: go for a section (auto-quoting!)
1496 ## now go for a section
1497 my $htmlsection = htmlify( $section );
1498 $url = page_sect( $page, $htmlsection );
1500 if( ! defined( $linktext ) ){
1501 $linktext = $section;
1502 $linktext .= " in " if $section && $page;
1503 $linktext .= "the $page manpage" if $page;
1505 ### print STDERR "got page/section url=$url\n";
1508 ## no luck: go for an ident
1515 ( $url, $fid ) = coderef( $page, $ident );
1517 if( ! defined( $linktext ) ){
1519 $linktext .= " in " if $ident && $page;
1520 $linktext .= "the $page manpage" if $page;
1522 ### print STDERR "got section=>coderef url=$url\n";
1526 # warning; show some text.
1527 $linktext = $opar unless defined $linktext;
1528 warn "$0: $podfile: cannot resolve L<$opar> in paragraph $paragraph.";
1531 # now we have an URL or just plain code
1532 $$rstr = $linktext . '>' . $$rstr;
1533 if( defined( $url ) ){
1534 $res = "<A HREF=\"$url\">" . process_text1( $lev, $rstr ) . '</A>';
1536 $res = '<EM>' . process_text1( $lev, $rstr ) . '</EM>';
1539 } elsif( $func eq 'S' ){
1540 # S<text> - non-breaking spaces
1541 $res = process_text1( $lev, $rstr );
1542 $res =~ s/ / /g;
1544 } elsif( $func eq 'X' ){
1546 $$rstr =~ s/^[^>]*>//;
1548 } elsif( $func eq 'Z' ){
1550 warn "$0: $podfile: invalid X<> in paragraph $paragraph."
1551 unless $$rstr =~ s/^>//;
1554 while( $$rstr =~ s/\A(.*?)([BCEFILSXZ]<|>)//s ){
1555 # all others: either recurse into new function or
1556 # terminate at closing angle bracket
1558 $pt .= '>' if $2 eq '>' && $lev == 1;
1559 $res .= $lev == 1 ? pure_text( $pt ) : inIS_text( $pt );
1560 return $res if $2 eq '>' && $lev > 1;
1562 $res .= process_text1( $lev, $rstr, substr($2,0,1) );
1567 $res .= pure_text( $$rstr );
1569 warn "$0: $podfile: undelimited $func<> in paragraph $paragraph.";
1576 # go_ahead: extract text of an IS (can be nested)
1579 my( $rstr, $func ) = @_;
1582 while( $$rstr =~ s/\A(.*?)([BCEFILSXZ]<|>)//s ){
1585 return $res if --$level == 0;
1591 warn "$0: $podfile: undelimited $func<> in paragraph $paragraph.";
1596 # emit_C - output result of C<text>
1597 # $text is the depod-ed text
1600 my( $text, $nocode ) = @_;
1602 my( $url, $fid ) = coderef( undef(), $text );
1604 # need HTML-safe text
1605 my $linktext = html_escape( $text );
1607 if( defined( $url ) &&
1608 (!defined( $EmittedItem ) || $EmittedItem ne $fid ) ){
1609 $res = "<A HREF=\"$url\"><CODE>$linktext</CODE></A>";
1610 } elsif( 0 && $nocode ){
1613 $res = "<CODE>$linktext</CODE>";
1619 # html_escape: make text safe for HTML
1623 $rest =~ s/&/&/g;
1624 $rest =~ s/</</g;
1625 $rest =~ s/>/>/g;
1626 $rest =~ s/"/"/g;
1632 # dosify - convert filenames to 8.3
1636 return lc($str) if $^O eq 'VMS'; # VMS just needs casing
1639 $str =~ s/(\.\w+)/substr ($1,0,4)/ge;
1640 $str =~ s/(\w+)/substr ($1,0,8)/ge;
1646 # page_sect - make an URL from the text of a L<>
1649 my( $page, $section ) = @_;
1650 my( $linktext, $page83, $link); # work strings
1652 # check if we know that this is a section in this page
1653 if (!defined $pages{$page} && defined $sections{$page}) {
1656 ### print STDERR "reset page='', section=$section\n";
1659 $page83=dosify($page);
1660 $page=$page83 if (defined $pages{$page83});
1662 $link = "#" . htmlify( $section );
1663 } elsif ( $page =~ /::/ ) {
1665 # Search page cache for an entry keyed under the html page name,
1666 # then look to see what directory that page might be in. NOTE:
1667 # this will only find one page. A better solution might be to produce
1668 # an intermediate page that is an index to all such pages.
1669 my $page_name = $page ;
1670 $page_name =~ s,^.*/,, ;
1671 if ( defined( $pages{ $page_name } ) &&
1672 $pages{ $page_name } =~ /([^:]*$page)\.(?:pod|pm):/
1677 # NOTE: This branch assumes that all A::B pages are located in
1678 # $htmlroot/A/B.html . This is often incorrect, since they are
1679 # often in $htmlroot/lib/A/B.html or such like. Perhaps we could
1680 # analyze the contents of %pages and figure out where any
1681 # cousins of A::B are, then assume that. So, if A::B isn't found,
1682 # but A::C is found in lib/A/C.pm, then A::B is assumed to be in
1683 # lib/A/B.pm. This is also limited, but it's an improvement.
1684 # Maybe a hints file so that the links point to the correct places
1688 $link = "$htmlroot/$page.html";
1689 $link .= "#" . htmlify( $section ) if ($section);
1690 } elsif (!defined $pages{$page}) {
1693 $section = htmlify( $section ) if $section ne "";
1694 ### print STDERR "...section=$section\n";
1696 # if there is a directory by the name of the page, then assume that an
1697 # appropriate section will exist in the subdirectory
1698 # if ($section ne "" && $pages{$page} =~ /([^:]*[^(\.pod|\.pm)]):/) {
1699 if ($section ne "" && $pages{$page} =~ /([^:]*(?<!\.pod)(?<!\.pm)):/) {
1700 $link = "$htmlroot/$1/$section.html";
1701 ### print STDERR "...link=$link\n";
1703 # since there is no directory by the name of the page, the section will
1704 # have to exist within a .html of the same name. thus, make sure there
1705 # is a .pod or .pm that might become that .html
1707 $section = "#$section" if $section;
1708 ### print STDERR "...section=$section\n";
1710 # check if there is a .pod with the page name
1711 if ($pages{$page} =~ /([^:]*)\.pod:/) {
1712 $link = "$htmlroot/$1.html$section";
1713 } elsif ($pages{$page} =~ /([^:]*)\.pm:/) {
1714 $link = "$htmlroot/$1.html$section";
1722 # Here, we take advantage of the knowledge that $htmlfileurl ne ''
1723 # implies $htmlroot eq ''. This means that the link in question
1724 # needs a prefix of $htmldir if it begins with '/'. The test for
1725 # the initial '/' is done to avoid '#'-only links, and to allow
1726 # for other kinds of links, like file:, ftp:, etc.
1728 if ( $htmlfileurl ne '' ) {
1729 $link = "$htmldir$link" if $link =~ m{^/};
1730 $url = relativize_url( $link, $htmlfileurl );
1731 # print( " b: [$link,$htmlfileurl,$url]\n" );
1744 # relativize_url - convert an absolute URL to one relative to a base URL.
1745 # Assumes both end in a filename.
1747 sub relativize_url {
1748 my ($dest,$source) = @_ ;
1750 my ($dest_volume,$dest_directory,$dest_file) =
1751 File::Spec::Unix->splitpath( $dest ) ;
1752 $dest = File::Spec::Unix->catpath( $dest_volume, $dest_directory, '' ) ;
1754 my ($source_volume,$source_directory,$source_file) =
1755 File::Spec::Unix->splitpath( $source ) ;
1756 $source = File::Spec::Unix->catpath( $source_volume, $source_directory, '' ) ;
1759 if ( $dest ne '' ) {
1760 $rel_path = File::Spec::Unix->abs2rel( $dest, $source ) ;
1763 if ( $rel_path ne '' &&
1764 substr( $rel_path, -1 ) ne '/' &&
1765 substr( $dest_file, 0, 1 ) ne '#'
1767 $rel_path .= "/$dest_file" ;
1770 $rel_path .= "$dest_file" ;
1778 # coderef - make URL from the text of a C<>
1781 my( $page, $item ) = @_;
1784 my $fid = fragment_id( $item );
1786 if( defined( $page ) ){
1787 # we have been given a $page...
1790 # Do we take it? Item could be a section!
1791 my $base = $items{$fid};
1792 $base =~ s{[^/]*/}{};
1793 if( $base ne "$page.html" ){
1794 ### print STDERR "coderef( $page, $item ): items{$fid} = $items{$fid} = $base => discard page!\n";
1799 # no page - local items precede cached items
1800 if( exists $local_items{$fid} ){
1801 $page = $local_items{$fid};
1803 $page = $items{$fid};
1807 # if there was a pod file that we found earlier with an appropriate
1808 # =item directive, then create a link to that page.
1809 if( defined $page ){
1811 if( $pages{$page} =~ /([^:.]*)\.[^:]*:/){
1812 $page = $1 . '.html';
1814 my $link = "$htmlroot/$page#item_$fid";
1816 # Here, we take advantage of the knowledge that $htmlfileurl
1817 # ne '' implies $htmlroot eq ''.
1818 if ( $htmlfileurl ne '' ) {
1819 $link = "$htmldir$link" ;
1820 $url = relativize_url( $link, $htmlfileurl ) ;
1825 $url = "#item_" . $fid;
1828 confess "url has space: $url" if $url =~ /"[^"]*\s[^"]*"/;
1830 return( $url, $fid );
1836 # Adapted from Nick Ing-Simmons' PodToHtml package.
1838 my $source_file = shift ;
1839 my $destination_file = shift;
1841 my $source = URI::file->new_abs($source_file);
1842 my $uo = URI::file->new($destination_file,$source)->abs;
1843 return $uo->rel->as_string;
1848 # finish_list - finish off any pending HTML lists. this should be called
1849 # after the entire pod file has been read and converted.
1852 while ($listlevel > 0) {
1853 print HTML "</DL>\n";
1859 # htmlify - converts a pod section specification to a suitable section
1860 # specification for HTML. Note that we keep spaces and special characters
1861 # except ", ? (Netscape problem) and the hyphen (writer's problem...).
1865 $heading =~ s/(\s+)/ /g;
1866 $heading =~ s/\s+\Z//;
1867 $heading =~ s/\A\s+//;
1868 # The hyphen is a disgrace to the English language.
1869 $heading =~ s/[-"?]//g;
1870 $heading = lc( $heading );
1875 # depod - convert text by eliminating all interior sequences
1876 # Note: can be called with copy or modify semantics
1888 ${$_[0]} = depod1( \$string );
1896 my( $rstr, $func ) = @_;
1898 if( ! defined( $func ) ){
1899 # skip to next begin of an interior sequence
1900 while( $$rstr =~ s/\A(.*?)([BCEFILSXZ])<// ){
1901 # recurse into its text
1902 $res .= $1 . depod1( $rstr, $2 );
1905 } elsif( $func eq 'E' ){
1906 # E<x> - convert to character
1907 $$rstr =~ s/^(\w+)>//;
1909 } elsif( $func eq 'X' ){
1911 $$rstr =~ s/^[^>]*>//;
1912 } elsif( $func eq 'Z' ){
1916 # all others: either recurse into new function or
1917 # terminate at closing angle bracket
1918 while( $$rstr =~ s/\A(.*?)([BCEFILSXZ]<|>)// ){
1921 $res .= depod1( $rstr, substr($2,0,1) );
1923 ## If we're here and $2 ne '>': undelimited interior sequence.
1924 ## Ignored, as this is called without proper indication of where we are.
1925 ## Rely on process_text to produce diagnostics.
1931 # fragment_id - construct a fragment identifier from:
1933 # b) contents of C<...>
1938 $text =~ s/\s+\Z//s;
1940 # a method or function?
1941 return $1 if $text =~ /(\w+)\s*\(/;
1942 return $1 if $text =~ /->\s*(\w+)\s*\(?/;
1945 return $1 if $text =~ /^([$@%*]\S+)/;
1947 # some pattern matching operator?
1948 return $1 if $text =~ m|^(\w+/).*/\w*$|;
1950 # fancy stuff... like "do { }"
1951 return $1 if $text =~ m|^(\w+)\s*{.*}$|;
1953 # honour the perlfunc manpage: func [PAR[,[ ]PAR]...]
1954 # and some funnies with ... Module ...
1955 return $1 if $text =~ m{^([a-z\d]+)(\s+[A-Z\d,/& ]+)?$};
1956 return $1 if $text =~ m{^([a-z\d]+)\s+Module(\s+[A-Z\d,/& ]+)?$};
1959 $text =~ s/\s+/_/sg;
1961 defined( $hc[ord($1)] ) ? $hc[ord($1)]
1962 : ( $hc[ord($1)] = sprintf( "%%%02X", ord($1) ) ) }gxe;
1963 $text = substr( $text, 0, 50 );
1970 # make_URL_href - generate HTML href from URL
1971 # Special treatment for CGI queries.
1973 sub make_URL_href($){
1976 s{^(http:[-\w/#~:.+=&%@!]+)(\?.*)?$}{<A HREF="$1$2">$1</A>}i ){
1977 $url = "<A HREF=\"$url\">$url</A>";