5 use vars qw($VERSION @ISA @EXPORT @EXPORT_OK);
8 @EXPORT = qw(pod2html htmlify);
9 @EXPORT_OK = qw(anchorify);
18 use locale; # make \w work right in non-ASCII lands
22 Pod::Html - module to convert pod files to HTML
31 Converts files from pod format (see L<perlpod>) to HTML format. It
32 can automatically generate indexes and cross-references, and it keeps
33 a cache of things it knows how to cross-reference.
40 "--podpath=lib:ext:pod:vms",
41 "--podroot=/usr/src/perl",
42 "--htmlroot=/perl/nmanual",
43 "--libpods=perlfunc:perlguts:perlvar:perlrun:perlop",
46 "--outfile=/perl/nmanual/foo.html");
48 pod2html takes the following arguments:
54 --backlink="Back to Top"
56 Adds "Back to Top" links in front of every C<head1> heading (except for
57 the first). By default, no backlinks are generated.
63 Creates the item and directory caches in the given directory.
69 Specify the URL of a cascading style sheet. Also disables all HTML/CSS
70 C<style> attributes that are output by default (to avoid conflicts).
76 Flushes the item and directory caches.
83 Creates header and footer blocks containing the text of the C<NAME>
84 section. By default, no headers are generated.
90 Displays the usage message.
97 Include hidden directories in the search for POD's in podpath if recurse
99 The default is not to traverse any directory whose name begins with C<.>.
100 See L</"podpath"> and L</"recurse">.
102 [This option is for backward compatibility only.
103 It's hard to imagine that one would usefully create a module with a
104 name component beginning with C<.>.]
110 Sets the directory in which the resulting HTML file is placed. This
111 is used to generate relative links to other files. Not passing this
112 causes all links to be absolute, since this is the value that tells
113 Pod::Html the root of the documentation tree.
119 Sets the base URL for the HTML files. When cross-references are made,
120 the HTML root is prepended to the URL.
127 Generate an index at the top of the HTML file. This is the default
134 Specify the pod file to convert. Input is taken from STDIN if no
139 --libpods=name:...:name
141 List of page names (eg, "perlfunc") which contain linkable C<=item>s.
148 B<Deprecated>, has no effect. For backwards compatibility only.
154 Specify the HTML file to create. Output goes to STDOUT if no outfile
159 --podpath=name:...:name
161 Specify which subdirectories of the podroot contain pod files whose
162 HTML converted forms can be linked to in cross references.
168 Specify the base directory for finding library pods.
175 Don't display I<mostly harmless> warning messages. These messages
176 will be displayed by default. But this is not the same as C<verbose>
184 Recurse into subdirectories specified in podpath (default behaviour).
190 Specify the title of the resulting HTML file.
197 Display progress messages. By default, they won't be displayed.
205 Converts a pod section specification to a suitable section specification
206 for HTML. Note that we keep spaces and special characters except
207 C<", ?> (Netscape problem) and the hyphen (writer's problem...).
213 Similar to C<htmlify()>, but turns non-alphanumerics into underscores. Note
214 that C<anchorify()> is not exported by default.
218 Uses C<$Config{pod2html}> to setup default options.
222 Tom Christiansen, E<lt>tchrist@perl.comE<gt>.
230 This program is distributed under the Artistic License.
235 my($Dircache, $Itemcache);
238 my($Htmlroot, $Htmldir, $Htmlfile, $Htmlfileurl);
239 my($Podfile, @Podpath, $Podroot);
249 my($Listlevel, @Listend);
251 use vars qw($Ignore); # need to localize it later.
253 my(%Items_Named, @Items_Seen);
262 my %Pages = (); # associative array used to find the location
263 # of pages referenced by L<> links.
264 my %Items = (); # associative array used to find the location
265 # of =item directives referenced by C<> links
270 my $Curdir = File::Spec->curdir;
275 $Cachedir = "."; # The directory to which item and directory
276 # caches will be written.
278 $Dircache = "pod2htmd.tmp";
279 $Itemcache = "pod2htmi.tmp";
281 @Begin_Stack = (); # begin/end stack
283 @Libpods = (); # files to search for links from C<> directives
284 $Htmlroot = "/"; # http-server base directory from which all
285 # relative paths in $podpath stem.
286 $Htmldir = ""; # The directory to which the html pages
287 # will (eventually) be written.
288 $Htmlfile = ""; # write to stdout by default
289 $Htmlfileurl = "" ; # The url that other files would use to
290 # refer to this file. This is only used
291 # to make relative urls that point to
294 $Podfile = ""; # read from stdin by default
295 @Podpath = (); # list of directories containing library pods.
296 $Podroot = $Curdir; # filesystem base directory from which all
297 # relative paths in $podpath stem.
298 $Css = ''; # Cascading style sheet
299 $Recurse = 1; # recurse on subdirectories in $podpath.
300 $Quiet = 0; # not quiet by default
301 $Verbose = 0; # not verbose by default
302 $Doindex = 1; # non-zero if we should generate an index
303 $Backlink = ''; # text for "back to top" links
304 $Listlevel = 0; # current list depth
305 @Listend = (); # the text to use to end the list.
306 $After_Lpar = 0; # set to true after a par in an =item
307 $Ignore = 1; # whether or not to format text. we don't
308 # format text until we hit our first pod
311 @Items_Seen = (); # for multiples of the same item in perlfunc
313 $Header = 0; # produce block header/footer
314 $Title = ''; # title to give the pod(s)
315 $Top = 1; # true if we are at the top of the doc. used
316 # to prevent the first <hr /> directive.
317 $Paragraph = ''; # which paragraph we're processing (used
318 # for error messages)
319 %Sections = (); # sections within this page
322 $Is83 = $^O eq 'dos'; # Is it an 8.3 filesystem?
326 # clean_data: global clean-up of pod data
330 for my $i ( 0..$#{$dataref} ) {
331 ${$dataref}[$i] =~ s/\s+\Z//;
333 # have a look for all-space lines
334 if( ${$dataref}[$i] =~ /^\s+$/m and $dataref->[$i] !~ /^\s/ ){
335 my @chunks = split( /^\s+$/m, ${$dataref}[$i] );
336 splice( @$dataref, $i, 1, @chunks );
349 $Is83 = 0 if (defined (&Dos::UseLFN) && Dos::UseLFN());
351 # cache of %Pages and %Items from last time we ran pod2html
353 #undef $opt_help if defined $opt_help;
355 # parse the command-line parameters
356 parse_command_line();
358 # escape the backlink argument (same goes for title but is done later...)
359 $Backlink = html_escape($Backlink) if defined $Backlink;
361 # set some variables to their default values if necessary
363 unless (@ARGV && $ARGV[0]) {
364 $Podfile = "-" unless $Podfile; # stdin
365 open(POD, "<$Podfile")
366 || die "$0: cannot open $Podfile file for input: $!\n";
368 $Podfile = $ARGV[0]; # XXX: might be more filenames
371 $Htmlfile = "-" unless $Htmlfile; # stdout
372 $Htmlroot = "" if $Htmlroot eq "/"; # so we don't get a //
373 $Htmldir =~ s#/\z## ; # so we don't get a //
375 && defined( $Htmldir )
377 && substr( $Htmlfile, 0, length( $Htmldir ) ) eq $Htmldir
380 # Set the 'base' url for this file, so that we can use it
381 # as the location from which to calculate relative links
382 # to other files. If this is '', then absolute links will
383 # be used throughout.
384 $Htmlfileurl= "$Htmldir/" . substr( $Htmlfile, length( $Htmldir ) + 1);
387 # read the pod a paragraph at a time
388 warn "Scanning for sections in input file(s)\n" if $Verbose;
397 @poddata = map { s/\r\n/\n/g;
399 map { "$_\n\n" } split /\n\n/ :
402 @poddata = map { s/\r/\n/g;
404 map { "$_\n\n" } split /\n\n/ :
411 clean_data( \@poddata );
413 # scan the pod for =head[1-6] directives and build an index
414 my $index = scan_headings(\%Sections, @poddata);
417 warn "No headings in $Podfile\n" if $Verbose;
420 # open the output file
421 open(HTML, ">$Htmlfile")
422 || die "$0: cannot open $Htmlfile file for output: $!\n";
424 # put a title in the HTML file if one wasn't specified
427 for (my $i = 0; $i < @poddata; $i++) {
428 if ($poddata[$i] =~ /^=head1\s*NAME\b/m) {
429 for my $para ( @poddata[$i, $i+1] ) {
431 if ($Title) = $para =~ /(\S+\s+-+.*\S)/s;
438 if (!$Title and $Podfile =~ /\.pod\z/) {
439 # probably a split pod so take first =head[12] as title
440 for (my $i = 0; $i < @poddata; $i++) {
441 last if ($Title) = $poddata[$i] =~ /^=head[12]\s*(.*)/;
443 warn "adopted '$Title' as title for $Podfile\n"
444 if $Verbose and $Title;
447 $Title =~ s/\s*\(.*\)//;
449 warn "$0: no title for $Podfile.\n" unless $Quiet;
450 $Podfile =~ /^(.*)(\.[^.\/]+)?\z/s;
451 $Title = ($Podfile eq "-" ? 'No Title' : $1);
452 warn "using $Title" if $Verbose;
454 $Title = html_escape($Title);
457 my $bodystyle = ' style="background-color: white"';
458 my $tdstyle = ' style="background-color: #cccccc"';
461 $csslink = qq(\n<link rel="stylesheet" href="$Css" type="text/css" />);
462 $csslink =~ s,\\,/,g;
463 $csslink =~ s,(/.):,$1|,;
468 my $block = $Header ? <<END_OF_BLOCK : '';
469 <table border="0" width="100%" cellspacing="0" cellpadding="3">
470 <tr><td class="block"$tdstyle valign="middle">
471 <big><strong><span class="block"> $Title</span></strong></big>
476 print HTML <<END_OF_HEAD;
477 <?xml version="1.0" ?>
478 <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
479 <html xmlns="http://www.w3.org/1999/xhtml">
481 <title>$Title</title>$csslink
482 <meta http-equiv="content-type" content="text/html; charset=utf-8" />
483 <link rev="made" href="mailto:$Config{perladmin}" />
490 # load/reload/validate/cache %Pages and %Items
491 get_cache($Dircache, $Itemcache, \@Podpath, $Podroot, $Recurse);
493 # scan the pod for =item directives
494 scan_items( \%Local_Items, "", @poddata);
496 # put an index at the top of the file. note, if $Doindex is 0 we
497 # still generate an index, but surround it with an html comment.
498 # that way some other program can extract it if desired.
501 my $hr = ($Doindex and $index) ? qq(<hr name="index" />) : "";
505 $index = qq(<!--\n$index\n-->\n);
508 print HTML << "END_OF_INDEX";
512 <p><a name=\"__index__\"></a></p>
520 # now convert this file
521 my $after_item; # set to true after an =item
523 warn "Converting input file $Podfile\n" if $Verbose;
524 foreach my $i (0..$#poddata){
527 if (/^(=.*)/s) { # is it a pod directive?
532 if (/^=begin\s+(\S+)\s*(.*)/si) {# =begin
533 process_begin($1, $2);
534 } elsif (/^=end\s+(\S+)\s*(.*)/si) {# =end
536 } elsif (/^=cut/) { # =cut
538 } elsif (/^=pod/) { # =pod
541 next if @Begin_Stack && $Begin_Stack[-1] ne 'html';
543 if (/^=(head[1-6])\s+(.*\S)/s) { # =head[1-6] heading
544 process_head( $1, $2, $Doindex && $index );
545 } elsif (/^=item\s*(.*\S)?/sm) { # =item text
546 $need_dd = process_item( $1 );
548 } elsif (/^=over\s*(.*)/) { # =over N
550 } elsif (/^=back/) { # =back
551 process_back($need_dd);
552 } elsif (/^=for\s+(\S+)\s*(.*)/si) {# =for
556 warn "$0: $Podfile: unknown pod directive '$1' in "
557 . "paragraph $Paragraph. ignoring.\n" unless $Quiet;
564 next if @Begin_Stack && $Begin_Stack[-1] ne 'html';
565 print HTML and next if @Begin_Stack && $Begin_Stack[-1] eq 'html';
566 print HTML "<dd>\n" if $need_dd;
568 if( $text =~ /\A\s+/ ){
569 process_pre( \$text );
570 print HTML "<pre>\n$text</pre>\n";
573 process_text( \$text );
575 # experimental: check for a paragraph where all lines
576 # have some ...\t...\t...\n pattern
578 my @lines = split( "\n", $text );
581 foreach my $line ( @lines ){
582 if( $line =~ /\S/ && $line !~ /\t/ ){
588 $text =~ s/\t+/<td>/g;
589 $text =~ s/^/<tr><td>/gm;
590 $text = '<table cellspacing="0" cellpadding="0">' .
595 ## end of experimental
600 print HTML "<p>$text</p>\n";
602 print HTML "</dd>\n" if $need_dd;
607 # finish off any pending directives
611 print HTML "<p><a href=\"#__index__\"><small>$Backlink</small></a></p>\n"
612 if $Doindex and $index and $Backlink;
614 print HTML <<END_OF_TAIL;
621 # close the html file
624 warn "Finished\n" if $Verbose;
627 ##############################################################################
631 warn "$0: $podfile: @_\n" if @_;
633 Usage: $0 --help --htmlroot=<name> --infile=<name> --outfile=<name>
634 --podpath=<name>:...:<name> --podroot=<name>
635 --libpods=<name>:...:<name> --recurse --verbose --index
636 --netscape --norecurse --noindex --cachedir=<name>
638 --backlink - set text for "back to top" links (default: none).
639 --cachedir - directory for the item and directory cache files.
640 --css - stylesheet URL
641 --flush - flushes the item and directory caches.
642 --[no]header - produce block header/footer (default is no headers).
643 --help - prints this message.
644 --hiddendirs - search hidden directories in podpath
645 --htmldir - directory for resulting HTML files.
646 --htmlroot - http-server base directory from which all relative paths
647 in podpath stem (default is /).
648 --[no]index - generate an index at the top of the resulting html
650 --infile - filename for the pod to convert (input taken from stdin
652 --libpods - colon-separated list of pages to search for =item pod
653 directives in as targets of C<> and implicit links (empty
654 by default). note, these are not filenames, but rather
655 page names like those that appear in L<> links.
656 --outfile - filename for the resulting html file (output sent to
658 --podpath - colon-separated list of directories containing library
659 pods (empty by default).
660 --podroot - filesystem base directory from which all relative paths
661 in podpath stem (default is .).
662 --[no]quiet - suppress some benign warning messages (default is off).
663 --[no]recurse - recurse on those subdirectories listed in podpath
665 --title - title that will appear in resulting html file.
666 --[no]verbose - self-explanatory (off by default).
667 --[no]netscape - deprecated, has no effect. for backwards compatibility only.
673 sub parse_command_line {
674 my ($opt_backlink,$opt_cachedir,$opt_css,$opt_flush,$opt_header,$opt_help,
675 $opt_htmldir,$opt_htmlroot,$opt_index,$opt_infile,$opt_libpods,
676 $opt_netscape,$opt_outfile,$opt_podpath,$opt_podroot,$opt_quiet,
677 $opt_recurse,$opt_title,$opt_verbose,$opt_hiddendirs);
679 unshift @ARGV, split ' ', $Config{pod2html} if $Config{pod2html};
680 my $result = GetOptions(
681 'backlink=s' => \$opt_backlink,
682 'cachedir=s' => \$opt_cachedir,
683 'css=s' => \$opt_css,
684 'flush' => \$opt_flush,
685 'header!' => \$opt_header,
686 'help' => \$opt_help,
687 'hiddendirs!'=> \$opt_hiddendirs,
688 'htmldir=s' => \$opt_htmldir,
689 'htmlroot=s' => \$opt_htmlroot,
690 'index!' => \$opt_index,
691 'infile=s' => \$opt_infile,
692 'libpods=s' => \$opt_libpods,
693 'netscape!' => \$opt_netscape,
694 'outfile=s' => \$opt_outfile,
695 'podpath=s' => \$opt_podpath,
696 'podroot=s' => \$opt_podroot,
697 'quiet!' => \$opt_quiet,
698 'recurse!' => \$opt_recurse,
699 'title=s' => \$opt_title,
700 'verbose!' => \$opt_verbose,
702 usage("-", "invalid parameters") if not $result;
704 usage("-") if defined $opt_help; # see if the user asked for help
705 $opt_help = ""; # just to make -w shut-up.
707 @Podpath = split(":", $opt_podpath) if defined $opt_podpath;
708 @Libpods = split(":", $opt_libpods) if defined $opt_libpods;
710 $Backlink = $opt_backlink if defined $opt_backlink;
711 $Cachedir = $opt_cachedir if defined $opt_cachedir;
712 $Css = $opt_css if defined $opt_css;
713 $Header = $opt_header if defined $opt_header;
714 $Htmldir = $opt_htmldir if defined $opt_htmldir;
715 $Htmlroot = $opt_htmlroot if defined $opt_htmlroot;
716 $Doindex = $opt_index if defined $opt_index;
717 $Podfile = $opt_infile if defined $opt_infile;
718 $HiddenDirs = $opt_hiddendirs if defined $opt_hiddendirs;
719 $Htmlfile = $opt_outfile if defined $opt_outfile;
720 $Podroot = $opt_podroot if defined $opt_podroot;
721 $Quiet = $opt_quiet if defined $opt_quiet;
722 $Recurse = $opt_recurse if defined $opt_recurse;
723 $Title = $opt_title if defined $opt_title;
724 $Verbose = $opt_verbose if defined $opt_verbose;
726 warn "Flushing item and directory caches\n"
727 if $opt_verbose && defined $opt_flush;
728 $Dircache = "$Cachedir/pod2htmd.tmp";
729 $Itemcache = "$Cachedir/pod2htmi.tmp";
730 if (defined $opt_flush) {
731 1 while unlink($Dircache, $Itemcache);
739 my($dircache, $itemcache, $podpath, $podroot, $recurse) = @_;
740 my @cache_key_args = @_;
742 # A first-level cache:
743 # Don't bother reading the cache files if they still apply
744 # and haven't changed since we last read them.
746 my $this_cache_key = cache_key(@cache_key_args);
748 return if $Saved_Cache_Key and $this_cache_key eq $Saved_Cache_Key;
750 # load the cache of %Pages and %Items if possible. $tests will be
751 # non-zero if successful.
753 if (-f $dircache && -f $itemcache) {
754 warn "scanning for item cache\n" if $Verbose;
755 $tests = load_cache($dircache, $itemcache, $podpath, $podroot);
758 # if we didn't succeed in loading the cache then we must (re)build
761 warn "scanning directories in pod-path\n" if $Verbose;
762 scan_podpath($podroot, $recurse, 0);
764 $Saved_Cache_Key = cache_key(@cache_key_args);
768 my($dircache, $itemcache, $podpath, $podroot, $recurse) = @_;
769 return join('!', $dircache, $itemcache, $recurse,
770 @$podpath, $podroot, stat($dircache), stat($itemcache));
774 # load_cache - tries to find if the caches stored in $dircache and $itemcache
775 # are valid caches of %Pages and %Items. if they are valid then it loads
776 # them and returns a non-zero value.
779 my($dircache, $itemcache, $podpath, $podroot) = @_;
785 open(CACHE, "<$itemcache") ||
786 die "$0: error opening $itemcache for reading: $!\n";
789 # is it the same podpath?
792 $tests++ if (join(":", @$podpath) eq $_);
794 # is it the same podroot?
797 $tests++ if ($podroot eq $_);
799 # load the cache if its good
805 warn "loading item cache\n" if $Verbose;
812 warn "scanning for directory cache\n" if $Verbose;
813 open(CACHE, "<$dircache") ||
814 die "$0: error opening $dircache for reading: $!\n";
818 # is it the same podpath?
821 $tests++ if (join(":", @$podpath) eq $_);
823 # is it the same podroot?
826 $tests++ if ($podroot eq $_);
828 # load the cache if its good
834 warn "loading directory cache\n" if $Verbose;
846 # scan_podpath - scans the directories specified in @podpath for directories,
847 # .pod files, and .pm files. it also scans the pod files specified in
848 # @Libpods for =item directives.
851 my($podroot, $recurse, $append) = @_;
853 my($libpod, $dirname, $pod, @files, @poddata);
860 # scan each directory listed in @Podpath
863 || die "$0: error changing to directory $podroot: $!\n";
864 foreach $dir (@Podpath) {
865 scan_dir($dir, $recurse);
868 # scan the pods listed in @Libpods for =item directives
869 foreach $libpod (@Libpods) {
870 # if the page isn't defined then we won't know where to find it
872 next unless defined $Pages{$libpod} && $Pages{$libpod};
874 # if there is a directory then use the .pod and .pm files within it.
875 # NOTE: Only finds the first so-named directory in the tree.
876 # if ($Pages{$libpod} =~ /([^:]*[^(\.pod|\.pm)]):/) {
877 if ($Pages{$libpod} =~ /([^:]*(?<!\.pod)(?<!\.pm)):/) {
878 # find all the .pod and .pm files within the directory
880 opendir(DIR, $dirname) ||
881 die "$0: error opening directory $dirname: $!\n";
882 @files = grep(/(\.pod|\.pm)\z/ && ! -d $_, readdir(DIR));
885 # scan each .pod and .pm file for =item directives
886 foreach $pod (@files) {
887 open(POD, "<$dirname/$pod") ||
888 die "$0: error opening $dirname/$pod for input: $!\n";
891 clean_data( \@poddata );
893 scan_items( \%Items, "$dirname/$pod", @poddata);
896 # use the names of files as =item directives too.
897 ### Don't think this should be done this way - confuses issues.(WL)
898 ### foreach $pod (@files) {
899 ### $pod =~ /^(.*)(\.pod|\.pm)$/;
900 ### $Items{$1} = "$dirname/$1.html" if $1;
902 } elsif ($Pages{$libpod} =~ /([^:]*\.pod):/ ||
903 $Pages{$libpod} =~ /([^:]*\.pm):/) {
904 # scan the .pod or .pm file for =item directives
906 open(POD, "<$pod") ||
907 die "$0: error opening $pod for input: $!\n";
910 clean_data( \@poddata );
912 scan_items( \%Items, "$pod", @poddata);
914 warn "$0: shouldn't be here (line ".__LINE__."\n" unless $Quiet;
917 @poddata = (); # clean-up a bit
920 || die "$0: error changing to directory $pwd: $!\n";
922 # cache the item list for later use
923 warn "caching items for later use\n" if $Verbose;
924 open(CACHE, ">$Itemcache") ||
925 die "$0: error open $Itemcache for writing: $!\n";
927 print CACHE join(":", @Podpath) . "\n$podroot\n";
928 foreach my $key (keys %Items) {
929 print CACHE "$key $Items{$key}\n";
934 # cache the directory list for later use
935 warn "caching directories for later use\n" if $Verbose;
936 open(CACHE, ">$Dircache") ||
937 die "$0: error open $Dircache for writing: $!\n";
939 print CACHE join(":", @Podpath) . "\n$podroot\n";
940 foreach my $key (keys %Pages) {
941 print CACHE "$key $Pages{$key}\n";
948 # scan_dir - scans the directory specified in $dir for subdirectories, .pod
949 # files, and .pm files. notes those that it finds. this information will
950 # be used later in order to figure out where the pages specified in L<>
951 # links are on the filesystem.
954 my($dir, $recurse) = @_;
955 my($t, @subdirs, @pods, $pod, $dirname, @dirs);
961 opendir(DIR, $dir) ||
962 die "$0: error opening directory $dir: $!\n";
963 while (defined($_ = readdir(DIR))) {
964 if (-d "$dir/$_" && $_ ne "." && $_ ne ".."
965 && ($HiddenDirs || !/^\./)
967 $Pages{$_} = "" unless defined $Pages{$_};
968 $Pages{$_} .= "$dir/$_:";
970 } elsif (/\.pod\z/) { # .pod
972 $Pages{$_} = "" unless defined $Pages{$_};
973 $Pages{$_} .= "$dir/$_.pod:";
974 push(@pods, "$dir/$_.pod");
975 } elsif (/\.html\z/) { # .html
977 $Pages{$_} = "" unless defined $Pages{$_};
978 $Pages{$_} .= "$dir/$_.pod:";
979 } elsif (/\.pm\z/) { # .pm
981 $Pages{$_} = "" unless defined $Pages{$_};
982 $Pages{$_} .= "$dir/$_.pm:";
983 push(@pods, "$dir/$_.pm");
984 } elsif (-T "$dir/$_") { # script(?)
986 if (open(F, "$dir/$_")) {
988 while (defined($line = <F>)) {
989 if ($line =~ /^=(?:pod|head1)/) {
990 $Pages{$_} = "" unless defined $Pages{$_};
991 $Pages{$_} .= "$dir/$_.pod:";
1001 # recurse on the subdirectories if necessary
1003 foreach my $subdir (@subdirs) {
1004 scan_dir("$dir/$subdir", $recurse);
1010 # scan_headings - scan a pod file for head[1-6] tags, note the tags, and
1014 my($sections, @data) = @_;
1015 my($tag, $which_head, $otitle, $listdepth, $index);
1022 # scan for =head directives, note their name, and build an index
1023 # pointing to each of them.
1024 foreach my $line (@data) {
1025 if ($line =~ /^=(head)([1-6])\s+(.*)/) {
1026 ($tag, $which_head, $otitle) = ($1,$2,$3);
1028 my $title = depod( $otitle );
1029 my $name = anchorify( $title );
1030 $$sections{$name} = 1;
1031 $title = process_text( \$otitle );
1033 while ($which_head != $listdepth) {
1034 if ($which_head > $listdepth) {
1035 $index .= "\n" . ("\t" x $listdepth) . "<ul>\n";
1037 } elsif ($which_head < $listdepth) {
1039 $index .= "\n" . ("\t" x $listdepth) . "</ul>\n";
1043 $index .= "\n" . ("\t" x $listdepth) . "<li>" .
1044 "<a href=\"#" . $name . "\">" .
1045 $title . "</a></li>";
1049 # finish off the lists
1050 while ($listdepth--) {
1051 $index .= "\n" . ("\t" x $listdepth) . "</ul>\n";
1054 # get rid of bogus lists
1055 $index =~ s,\t*<ul>\s*</ul>\n,,g;
1061 # scan_items - scans the pod specified by $pod for =item directives. we
1062 # will use this information later on in resolving C<> links.
1065 my( $itemref, $pod, @poddata ) = @_;
1069 $pod =~ s/\.pod\z//;
1070 $pod .= ".html" if $pod;
1072 foreach $i (0..$#poddata) {
1073 my $txt = depod( $poddata[$i] );
1075 # figure out what kind of item it is.
1076 # Build string for referencing this item.
1077 if ( $txt =~ /\A=item\s+\*\s*(.*)\Z/s ) { # bullet
1080 } elsif( $txt =~ /\A=item\s+(?>\d+\.?)\s*(.*)\Z/s ) { # numbered list
1082 } elsif( $txt =~ /\A=item\s+(.*)\Z/s ) { # plain item
1087 my $fid = fragment_id( $item );
1088 $$itemref{$fid} = "$pod" if $fid;
1093 # process_head - convert a pod head[1-6] tag and convert it to HTML format.
1096 my($tag, $heading, $hasindex) = @_;
1098 # figure out the level of the =head
1099 $tag =~ /head([1-6])/;
1103 warn "$0: $Podfile: unterminated list at =head in paragraph $Paragraph. ignoring.\n" unless $Quiet;
1104 while( $Listlevel ){
1110 if( $level == 1 && ! $Top ){
1111 print HTML "<a href=\"#__index__\"><small>$Backlink</small></a>\n"
1112 if $hasindex and $Backlink;
1113 print HTML "</p>\n<hr />\n"
1115 print HTML "</p>\n";
1118 my $name = anchorify( depod( $heading ) );
1119 my $convert = process_text( \$heading );
1120 print HTML "<h$level><a name=\"$name\">$convert</a></h$level>\n";
1125 # emit_item_tag - print an =item's text
1126 # Note: The global $EmittedItem is used for inhibiting self-references.
1130 sub emit_item_tag($$$){
1131 my( $otext, $text, $compact ) = @_;
1132 my $item = fragment_id( $text , -generate);
1134 $EmittedItem = $item;
1135 ### print STDERR "emit_item_tag=$item ($text)\n";
1137 print HTML '<strong>';
1138 if ($Items_Named{$item}++) {
1139 print HTML process_text( \$otext );
1142 $name = anchorify($name);
1143 print HTML qq{<a name="$name" class="item">}, process_text( \$otext ), '</a>';
1145 print HTML "</strong>\n";
1146 undef( $EmittedItem );
1151 if( $Items_Seen[$Listlevel]++ == 0 ){
1152 push( @Listend, "</$tag>" );
1153 print HTML "<$tag>\n";
1155 my $emitted = $tag eq 'dl' ? 'dt' : 'li';
1156 print HTML "<$emitted>";
1161 # process_item - convert a pod item tag and convert it to HTML format.
1165 my $need_dd = 0; # set to 1 if we need a <dd></dd> after an item
1167 # lots of documents start a list without doing an =over. this is
1168 # bad! but, the proper thing to do seems to be to just assume
1169 # they did do an =over. so warn them once and then continue.
1170 if( $Listlevel == 0 ){
1171 warn "$0: $Podfile: unexpected =item directive in paragraph $Paragraph. ignoring.\n" unless $Quiet;
1175 # formatting: insert a paragraph if preceding item has >1 paragraph
1177 print HTML $need_dd ? "</dd>\n" : "</li>\n" if $After_Lpar;
1181 # remove formatting instructions from the text
1182 my $text = depod( $otext );
1184 my $emitted; # the tag actually emitted, used for closing
1186 # all the list variants:
1187 if( $text =~ /\A\*/ ){ # bullet
1188 $emitted = emit_li( 'ul' );
1189 if ($text =~ /\A\*\s+(.+)\Z/s ) { # with additional text
1191 $otext =~ s/\A\*\s+//;
1192 emit_item_tag( $otext, $tag, 1 );
1195 } elsif( $text =~ /\A\d+/ ){ # numbered list
1196 $emitted = emit_li( 'ol' );
1197 if ($text =~ /\A(?>\d+\.?)\s*(.+)\Z/s ) { # with additional text
1199 $otext =~ s/\A\d+\.?\s*//;
1200 emit_item_tag( $otext, $tag, 1 );
1203 } else { # definition list
1204 $emitted = emit_li( 'dl' );
1205 if ($text =~ /\A(.+)\Z/s ){ # should have text
1206 emit_item_tag( $otext, $text, 1 );
1215 # process_over - process a pod over tag and start a corresponding HTML list.
1220 push( @Items_Seen, 0 );
1225 # process_back - process a pod back tag and convert it to HTML format.
1228 my $need_dd = shift;
1229 if( $Listlevel == 0 ){
1230 warn "$0: $Podfile: unexpected =back directive in paragraph $Paragraph. ignoring.\n" unless $Quiet;
1234 # close off the list. note, I check to see if $Listend[$Listlevel] is
1235 # defined because an =item directive may have never appeared and thus
1236 # $Listend[$Listlevel] may have never been initialized.
1238 if( defined $Listend[$Listlevel] ){
1239 print HTML $need_dd ? "</dd>\n" : "</li>\n" if $After_Lpar;
1240 print HTML $Listend[$Listlevel];
1246 # clean up item count
1251 # process_cut - process a pod cut tag, thus start ignoring pod directives.
1258 # process_pod - process a pod tag, thus stop ignoring pod directives
1259 # until we see a corresponding cut.
1262 # no need to set $Ignore to 0 cause the main loop did it
1266 # process_for - process a =for pod tag. if it's for html, spit
1267 # it out verbatim, if illustration, center it, otherwise ignore it.
1270 my($whom, $text) = @_;
1271 if ( $whom =~ /^(pod2)?html$/i) {
1273 } elsif ($whom =~ /^illustration$/i) {
1274 1 while chomp $text;
1275 for my $ext (qw[.png .gif .jpeg .jpg .tga .pcl .bmp]) {
1276 $text .= $ext, last if -r "$text$ext";
1278 print HTML qq{<p align="center"><img src="$text" alt="$text illustration" /></p>};
1283 # process_begin - process a =begin pod tag. this pushes
1284 # whom we're beginning on the begin stack. if there's a
1285 # begin stack, we only print if it us.
1288 my($whom, $text) = @_;
1290 push (@Begin_Stack, $whom);
1291 if ( $whom =~ /^(pod2)?html$/) {
1292 print HTML $text if $text;
1297 # process_end - process a =end pod tag. pop the
1298 # begin stack. die if we're mismatched.
1301 my($whom, $text) = @_;
1303 if ($Begin_Stack[-1] ne $whom ) {
1304 die "Unmatched begin/end at chunk $Paragraph\n"
1306 pop( @Begin_Stack );
1310 # process_pre - indented paragraph, made into <pre></pre>
1319 # insert spaces in place of tabs
1322 1 while $line =~ s/(\t+)/' ' x ((length($1) * 8) - $-[0] % 8)/e;
1326 # convert some special chars to HTML escapes
1327 $rest = html_escape($rest);
1329 # try and create links for all occurrences of perl.* within
1330 # the preformatted text.
1334 if ( defined $Pages{$2} ){ # is a link
1335 qq($1<a href="$Htmlroot/$Pages{$2}">$2</a>);
1336 } elsif (defined $Pages{dosify($2)}) { # is a link
1337 qq($1<a href="$Htmlroot/$Pages{dosify($2)}">$2</a>);
1343 (<a\ href="?) ([^>:]*:)? ([^>:]*) \.pod: ([^>:]*:)?
1346 if ( $Htmlfileurl ne '' ){
1347 # Here, we take advantage of the knowledge
1348 # that $Htmlfileurl ne '' implies $Htmlroot eq ''.
1349 # Since $Htmlroot eq '', we need to prepend $Htmldir
1350 # on the fron of the link to get the absolute path
1351 # of the link's target. We check for a leading '/'
1352 # to avoid corrupting links that are #, file:, etc.
1354 $old_url = "$Htmldir$old_url" if $old_url =~ m{^\/};
1355 $url = relativize_url( "$old_url.html", $Htmlfileurl );
1362 # Look for embedded URLs and make them into links. We don't
1363 # relativize them since they are best left as the author intended.
1365 my $urls = '(' . join ('|', qw{
1378 my $gunk = '/#~:.?+=&%@!\-';
1379 my $punc = '.:!?\-;';
1380 my $any = "${ltrs}${gunk}${punc}";
1383 \b # start at word boundary
1385 $urls : # need resource and a colon
1386 (?!:) # Ignore File::, among others.
1387 [$any] +? # followed by one or more of any valid
1388 # character, but be conservative and
1389 # take only what you need to....
1392 " > # maybe pre-quoted '<a href="...">'
1394 [$punc]* # 0 or more punctuation
1396 [^$any] # by a non-url char
1398 $ # end of the string
1401 $ # then end of the string
1403 }{<a href="$1">$1</a>}igox;
1405 # text should be as it is (verbatim)
1411 # pure text processing
1413 # pure_text/inIS_text: differ with respect to automatic C<> recognition.
1414 # we don't want this to happen within IS
1418 process_puretext( $text, 1 );
1423 process_puretext( $text, 0 );
1427 # process_puretext - process pure text (without pod-escapes) converting
1428 # double-quotes and handling implicit C<> links.
1430 sub process_puretext {
1431 my($text, $notinIS) = @_;
1433 ## Guessing at func() or [\$\@%&]*var references in plain text is destined
1434 ## to produce some strange looking ref's. uncomment to disable:
1437 my(@words, $lead, $trail);
1439 # keep track of leading and trailing white-space
1440 $lead = ($text =~ s/\A(\s+)//s ? $1 : "");
1441 $trail = ($text =~ s/(\s+)\Z//s ? $1 : "");
1443 # split at space/non-space boundaries
1444 @words = split( /(?<=\s)(?=\S)|(?<=\S)(?=\s)/, $text );
1446 # process each word individually
1447 foreach my $word (@words) {
1449 next if $word =~ /^\s*$/;
1450 # see if we can infer a link
1451 if( $notinIS && $word =~ /^(\w+)\((.*)\)$/ ) {
1452 # has parenthesis so should have been a C<> ref
1453 ## try for a pagename (perlXXX(1))?
1454 my( $func, $args ) = ( $1, $2 );
1455 if( $args =~ /^\d+$/ ){
1456 my $url = page_sect( $word, '' );
1458 $word = "<a href=\"$url\">the $word manpage</a>";
1462 ## try function name for a link, append tt'ed argument list
1463 $word = emit_C( $func, '', "($args)");
1465 #### disabled. either all (including $\W, $\w+{.*} etc.) or nothing.
1466 ## } elsif( $notinIS && $word =~ /^[\$\@%&*]+\w+$/) {
1467 ## # perl variables, should be a C<> ref
1468 ## $word = emit_C( $word );
1470 } elsif ($word =~ m,^\w+://\w,) {
1472 # Don't relativize it: leave it as the author intended
1473 $word = qq(<a href="$word">$word</a>);
1474 } elsif ($word =~ /[\w.-]+\@[\w-]+\.\w/) {
1475 # looks like an e-mail address
1476 my ($w1, $w2, $w3) = ("", $word, "");
1477 ($w1, $w2, $w3) = ("(", $1, ")$2") if $word =~ /^\((.*?)\)(,?)/;
1478 ($w1, $w2, $w3) = ("<", $1, ">$2") if $word =~ /^<(.*?)>(,?)/;
1479 $word = qq($w1<a href="mailto:$w2">$w2</a>$w3);
1481 $word = html_escape($word) if $word =~ /["&<>]/;
1485 # put everything back together
1486 return $lead . join( '', @words ) . $trail;
1491 # process_text - handles plaintext that appears in the input pod file.
1492 # there may be pod commands embedded within the text so those must be
1493 # converted to html commands.
1496 sub process_text1($$;$$);
1497 sub pattern ($) { $_[0] ? '[^\S\n]+'.('>' x ($_[0] + 1)) : '>' }
1498 sub closing ($) { local($_) = shift; (defined && s/\s+$//) ? length : 0 }
1503 my $res = process_text1( 0, $tref );
1508 sub process_text_rfc_links {
1511 # For every "RFCnnnn" or "RFC nnn" link it to the authoritative
1512 # source. Do not use (i) option here. Require RFC to be written in
1513 # in capital letters.
1516 (?=^\S) # positive lookahead, make sure this is "text" paragraph
1517 (.*?) # Grab leading text
1518 (?<=[^<>]) # Make sure this is not an URL already
1519 (RFC\s*(\d{1,5}))(?=\s) # max 5 digits
1521 {$1<a href="http://www.ietf.org/rfc/rfc$3.txt">$2</a>}gx;
1526 sub process_text1($$;$$){
1527 my( $lev, $rstr, $func, $closing ) = @_;
1530 unless (defined $func) {
1536 # B<text> - boldface
1537 $res = '<strong>' . process_text1( $lev, $rstr ) . '</strong>';
1539 } elsif( $func eq 'C' ){
1540 # C<code> - can be a ref or <code></code>
1541 # need to extract text
1542 my $par = go_ahead( $rstr, 'C', $closing );
1544 ## clean-up of the link target
1545 my $text = depod( $par );
1547 ### my $x = $par =~ /[BI]</ ? 'yes' : 'no' ;
1548 ### print STDERR "-->call emit_C($par) lev=$lev, par with BI=$x\n";
1550 $res = emit_C( $text, $lev > 1 || ($par =~ /[BI]</) );
1552 } elsif( $func eq 'E' ){
1553 # E<x> - convert to character
1554 $$rstr =~ s/^([^>]*)>//;
1556 $escape =~ s/^(\d+|X[\dA-F]+)$/#$1/i;
1559 } elsif( $func eq 'F' ){
1560 # F<filename> - italicize
1561 $res = '<em class="file">' . process_text1( $lev, $rstr ) . '</em>';
1563 } elsif( $func eq 'I' ){
1564 # I<text> - italicize
1565 $res = '<em>' . process_text1( $lev, $rstr ) . '</em>';
1567 } elsif( $func eq 'L' ){
1569 ## L<text|cross-ref> => produce text, use cross-ref for linking
1570 ## L<cross-ref> => make text from cross-ref
1571 ## need to extract text
1572 my $par = go_ahead( $rstr, 'L', $closing );
1574 # some L<>'s that shouldn't be:
1575 # a) full-blown URL's are emitted as-is
1576 if( $par =~ m{^\w+://}s ){
1577 return make_URL_href( $par );
1579 # b) C<...> is stripped and treated as C<>
1580 if( $par =~ /^C<(.*)>$/ ){
1581 my $text = depod( $1 );
1582 return emit_C( $text, $lev > 1 || ($par =~ /[BI]</) );
1585 # analyze the contents
1586 $par =~ s/\n/ /g; # undo word-wrapped tags
1589 if( $par =~ s{^([^|]+)\|}{} ){
1593 # make sure sections start with a /
1596 my( $page, $section, $ident );
1598 # check for link patterns
1599 if( $par =~ m{^([^/]+?)/(?!")(.*?)$} ){ # name/ident
1600 # we've got a name/ident (no quotes)
1601 ( $page, $ident ) = ( $1, $2 );
1602 ### print STDERR "--> L<$par> to page $page, ident $ident\n";
1604 } elsif( $par =~ m{^(.*?)/"?(.*?)"?$} ){ # [name]/"section"
1605 # even though this should be a "section", we go for ident first
1606 ( $page, $ident ) = ( $1, $2 );
1607 ### print STDERR "--> L<$par> to page $page, section $section\n";
1609 } elsif( $par =~ /\s/ ){ # this must be a section with missing quotes
1610 ( $page, $section ) = ( '', $par );
1611 ### print STDERR "--> L<$par> to void page, section $section\n";
1614 ( $page, $section ) = ( $par, '' );
1615 ### print STDERR "--> L<$par> to page $par, void section\n";
1618 # now, either $section or $ident is defined. the convoluted logic
1619 # below tries to resolve L<> according to what the user specified.
1620 # failing this, we try to find the next best thing...
1621 my( $url, $ltext, $fid );
1624 if( defined $ident ){
1625 ## try to resolve $ident as an item
1626 ( $url, $fid ) = coderef( $page, $ident );
1628 if( ! defined( $linktext ) ){
1630 $linktext .= " in " if $ident && $page;
1631 $linktext .= "the $page manpage" if $page;
1633 ### print STDERR "got coderef url=$url\n";
1636 ## no luck: go for a section (auto-quoting!)
1639 ## now go for a section
1640 my $htmlsection = htmlify( $section );
1641 $url = page_sect( $page, $htmlsection );
1643 if( ! defined( $linktext ) ){
1644 $linktext = $section;
1645 $linktext .= " in " if $section && $page;
1646 $linktext .= "the $page manpage" if $page;
1648 ### print STDERR "got page/section url=$url\n";
1651 ## no luck: go for an ident
1658 ( $url, $fid ) = coderef( $page, $ident );
1660 if( ! defined( $linktext ) ){
1662 $linktext .= " in " if $ident && $page;
1663 $linktext .= "the $page manpage" if $page;
1665 ### print STDERR "got section=>coderef url=$url\n";
1669 # warning; show some text.
1670 $linktext = $opar unless defined $linktext;
1671 warn "$0: $Podfile: cannot resolve L<$opar> in paragraph $Paragraph.\n" unless $Quiet;
1674 # now we have a URL or just plain code
1675 $$rstr = $linktext . '>' . $$rstr;
1676 if( defined( $url ) ){
1677 $res = "<a href=\"$url\">" . process_text1( $lev, $rstr ) . '</a>';
1679 $res = '<em>' . process_text1( $lev, $rstr ) . '</em>';
1682 } elsif( $func eq 'S' ){
1683 # S<text> - non-breaking spaces
1684 $res = process_text1( $lev, $rstr );
1685 $res =~ s/ / /g;
1687 } elsif( $func eq 'X' ){
1689 $$rstr =~ s/^[^>]*>//;
1691 } elsif( $func eq 'Z' ){
1693 warn "$0: $Podfile: invalid X<> in paragraph $Paragraph.\n"
1694 unless $$rstr =~ s/^>// or $Quiet;
1697 my $term = pattern $closing;
1698 while( $$rstr =~ s/\A(.*?)(([BCEFILSXZ])<(<+[^\S\n]+)?|$term)//s ){
1699 # all others: either recurse into new function or
1700 # terminate at closing angle bracket(s)
1702 $pt .= $2 if !$3 && $lev == 1;
1703 $res .= $lev == 1 ? pure_text( $pt ) : inIS_text( $pt );
1704 return $res if !$3 && $lev > 1;
1706 $res .= process_text1( $lev, $rstr, $3, closing $4 );
1710 $res .= pure_text( $$rstr );
1712 warn "$0: $Podfile: undelimited $func<> in paragraph $Paragraph.\n" unless $Quiet;
1714 $res = process_text_rfc_links($res);
1720 # go_ahead: extract text of an IS (can be nested)
1723 my( $rstr, $func, $closing ) = @_;
1725 my @closing = ($closing);
1727 s/\A(.*?)(([BCEFILSXZ])<(<+[^\S\n]+)?|@{[pattern $closing[0]]})//s ){
1731 return $res unless @closing;
1733 unshift @closing, closing $4;
1737 warn "$0: $Podfile: undelimited $func<> in paragraph $Paragraph.\n" unless $Quiet;
1742 # emit_C - output result of C<text>
1743 # $text is the depod-ed text
1746 my( $text, $nocode, $args ) = @_;
1747 $args = '' unless defined $args;
1749 my( $url, $fid ) = coderef( undef(), $text );
1751 # need HTML-safe text
1752 my $linktext = html_escape( "$text$args" );
1754 if( defined( $url ) &&
1755 (!defined( $EmittedItem ) || $EmittedItem ne $fid ) ){
1756 $res = "<a href=\"$url\"><code>$linktext</code></a>";
1757 } elsif( 0 && $nocode ){
1760 $res = "<code>$linktext</code>";
1766 # html_escape: make text safe for HTML
1770 $rest =~ s/&/&/g;
1771 $rest =~ s/</</g;
1772 $rest =~ s/>/>/g;
1773 $rest =~ s/"/"/g;
1774 # ' is only in XHTML, not HTML4. Be conservative
1775 #$rest =~ s/'/'/g;
1781 # dosify - convert filenames to 8.3
1785 return lc($str) if $^O eq 'VMS'; # VMS just needs casing
1788 $str =~ s/(\.\w+)/substr ($1,0,4)/ge;
1789 $str =~ s/(\w+)/substr ($1,0,8)/ge;
1795 # page_sect - make a URL from the text of a L<>
1798 my( $page, $section ) = @_;
1799 my( $linktext, $page83, $link); # work strings
1801 # check if we know that this is a section in this page
1802 if (!defined $Pages{$page} && defined $Sections{$page}) {
1805 ### print STDERR "reset page='', section=$section\n";
1808 $page83=dosify($page);
1809 $page=$page83 if (defined $Pages{$page83});
1811 $link = "#" . anchorify( $section );
1812 } elsif ( $page =~ /::/ ) {
1814 # Search page cache for an entry keyed under the html page name,
1815 # then look to see what directory that page might be in. NOTE:
1816 # this will only find one page. A better solution might be to produce
1817 # an intermediate page that is an index to all such pages.
1818 my $page_name = $page ;
1819 $page_name =~ s,^.*/,,s ;
1820 if ( defined( $Pages{ $page_name } ) &&
1821 $Pages{ $page_name } =~ /([^:]*$page)\.(?:pod|pm):/
1826 # NOTE: This branch assumes that all A::B pages are located in
1827 # $Htmlroot/A/B.html . This is often incorrect, since they are
1828 # often in $Htmlroot/lib/A/B.html or such like. Perhaps we could
1829 # analyze the contents of %Pages and figure out where any
1830 # cousins of A::B are, then assume that. So, if A::B isn't found,
1831 # but A::C is found in lib/A/C.pm, then A::B is assumed to be in
1832 # lib/A/B.pm. This is also limited, but it's an improvement.
1833 # Maybe a hints file so that the links point to the correct places
1837 $link = "$Htmlroot/$page.html";
1838 $link .= "#" . anchorify( $section ) if ($section);
1839 } elsif (!defined $Pages{$page}) {
1842 $section = anchorify( $section ) if $section ne "";
1843 ### print STDERR "...section=$section\n";
1845 # if there is a directory by the name of the page, then assume that an
1846 # appropriate section will exist in the subdirectory
1847 # if ($section ne "" && $Pages{$page} =~ /([^:]*[^(\.pod|\.pm)]):/) {
1848 if ($section ne "" && $Pages{$page} =~ /([^:]*(?<!\.pod)(?<!\.pm)):/) {
1849 $link = "$Htmlroot/$1/$section.html";
1850 ### print STDERR "...link=$link\n";
1852 # since there is no directory by the name of the page, the section will
1853 # have to exist within a .html of the same name. thus, make sure there
1854 # is a .pod or .pm that might become that .html
1856 $section = "#$section" if $section;
1857 ### print STDERR "...section=$section\n";
1859 # check if there is a .pod with the page name.
1860 # for L<Foo>, Foo.(pod|pm) is preferred to A/Foo.(pod|pm)
1861 if ($Pages{$page} =~ /([^:]*)\.(?:pod|pm):/) {
1862 $link = "$Htmlroot/$1.html$section";
1870 # Here, we take advantage of the knowledge that $Htmlfileurl ne ''
1871 # implies $Htmlroot eq ''. This means that the link in question
1872 # needs a prefix of $Htmldir if it begins with '/'. The test for
1873 # the initial '/' is done to avoid '#'-only links, and to allow
1874 # for other kinds of links, like file:, ftp:, etc.
1876 if ( $Htmlfileurl ne '' ) {
1877 $link = "$Htmldir$link" if $link =~ m{^/}s;
1878 $url = relativize_url( $link, $Htmlfileurl );
1879 # print( " b: [$link,$Htmlfileurl,$url]\n" );
1892 # relativize_url - convert an absolute URL to one relative to a base URL.
1893 # Assumes both end in a filename.
1895 sub relativize_url {
1896 my ($dest,$source) = @_ ;
1898 my ($dest_volume,$dest_directory,$dest_file) =
1899 File::Spec::Unix->splitpath( $dest ) ;
1900 $dest = File::Spec::Unix->catpath( $dest_volume, $dest_directory, '' ) ;
1902 my ($source_volume,$source_directory,$source_file) =
1903 File::Spec::Unix->splitpath( $source ) ;
1904 $source = File::Spec::Unix->catpath( $source_volume, $source_directory, '' ) ;
1907 if ( $dest ne '' ) {
1908 $rel_path = File::Spec::Unix->abs2rel( $dest, $source ) ;
1911 if ( $rel_path ne '' &&
1912 substr( $rel_path, -1 ) ne '/' &&
1913 substr( $dest_file, 0, 1 ) ne '#'
1915 $rel_path .= "/$dest_file" ;
1918 $rel_path .= "$dest_file" ;
1926 # coderef - make URL from the text of a C<>
1929 my( $page, $item ) = @_;
1932 my $fid = fragment_id( $item );
1933 if( defined( $page ) && $page ne "" ){
1934 # we have been given a $page...
1937 # Do we take it? Item could be a section!
1938 my $base = $Items{$fid} || "";
1939 $base =~ s{[^/]*/}{};
1940 if( $base ne "$page.html" ){
1941 ### print STDERR "coderef( $page, $item ): items{$fid} = $Items{$fid} = $base => discard page!\n";
1946 # no page - local items precede cached items
1947 if( defined( $fid ) ){
1948 if( exists $Local_Items{$fid} ){
1949 $page = $Local_Items{$fid};
1951 $page = $Items{$fid};
1956 # if there was a pod file that we found earlier with an appropriate
1957 # =item directive, then create a link to that page.
1958 if( defined $page ){
1960 if( exists $Pages{$page} and $Pages{$page} =~ /([^:.]*)\.[^:]*:/){
1961 $page = $1 . '.html';
1963 my $link = "$Htmlroot/$page#item_" . anchorify($fid);
1965 # Here, we take advantage of the knowledge that $Htmlfileurl
1966 # ne '' implies $Htmlroot eq ''.
1967 if ( $Htmlfileurl ne '' ) {
1968 $link = "$Htmldir$link" ;
1969 $url = relativize_url( $link, $Htmlfileurl ) ;
1974 $url = "#item_" . anchorify($fid);
1977 confess "url has space: $url" if $url =~ /"[^"]*\s[^"]*"/;
1979 return( $url, $fid );
1985 # Adapted from Nick Ing-Simmons' PodToHtml package.
1987 my $source_file = shift ;
1988 my $destination_file = shift;
1990 my $source = URI::file->new_abs($source_file);
1991 my $uo = URI::file->new($destination_file,$source)->abs;
1992 return $uo->rel->as_string;
1997 # finish_list - finish off any pending HTML lists. this should be called
1998 # after the entire pod file has been read and converted.
2001 while ($Listlevel > 0) {
2002 print HTML "</dl>\n";
2008 # htmlify - converts a pod section specification to a suitable section
2009 # specification for HTML. Note that we keep spaces and special characters
2010 # except ", ? (Netscape problem) and the hyphen (writer's problem...).
2014 $heading =~ s/(\s+)/ /g;
2015 $heading =~ s/\s+\Z//;
2016 $heading =~ s/\A\s+//;
2017 # The hyphen is a disgrace to the English language.
2018 # $heading =~ s/[-"?]//g;
2019 $heading =~ s/["?]//g;
2020 $heading = lc( $heading );
2025 # similar to htmlify, but turns non-alphanumerics into underscores
2029 $anchor = htmlify($anchor);
2030 $anchor =~ s/\W/_/g;
2035 # depod - convert text by eliminating all interior sequences
2036 # Note: can be called with copy or modify semantics
2043 $E2c{amp} = '&'; # in Tk's pods
2051 ${$_[0]} = depod1( \$string );
2059 my( $rstr, $func, $closing ) = @_;
2061 return $res unless defined $$rstr;
2062 if( ! defined( $func ) ){
2063 # skip to next begin of an interior sequence
2064 while( $$rstr =~ s/\A(.*?)([BCEFILSXZ])<(<+[^\S\n]+)?//s ){
2065 # recurse into its text
2066 $res .= $1 . depod1( $rstr, $2, closing $3);
2069 } elsif( $func eq 'E' ){
2070 # E<x> - convert to character
2071 $$rstr =~ s/^([^>]*)>//;
2072 $res .= $E2c{$1} || "";
2073 } elsif( $func eq 'X' ){
2075 $$rstr =~ s/^[^>]*>//;
2076 } elsif( $func eq 'Z' ){
2080 # all others: either recurse into new function or
2081 # terminate at closing angle bracket
2082 my $term = pattern $closing;
2083 while( $$rstr =~ s/\A(.*?)(([BCEFILSXZ])<(<+[^\S\n]+)?|$term)//s ){
2086 $res .= depod1( $rstr, $3, closing $4 );
2088 ## If we're here and $2 ne '>': undelimited interior sequence.
2089 ## Ignored, as this is called without proper indication of where we are.
2090 ## Rely on process_text to produce diagnostics.
2096 my %seen; # static fragment record hash
2098 sub fragment_id_readable {
2100 my $generate = shift; # optional flag
2104 # just clean the punctuation and leave the words for the
2105 # fragment identifier.
2106 $text =~ s/([[:punct:]\s])+/$1/g;
2107 $text =~ s/[[:punct:]\s]+\Z//g;
2109 # "=item --version", remove leading punctuation.
2110 $text =~ s/^[-[:punct:]]//;
2114 # Nothing left after removing punctuation, so leave it as is
2115 # E.g. if option is named: "=item -#"
2121 if ( exists $seen{$text} ) {
2122 # This already exists, make it unique
2124 $text = $text . $seen{$text};
2126 $seen{$text} = 1; # first time seen this fragment
2134 sub fragment_id_obfusticated { # This was the old "_2d_2d__"
2136 my $generate = shift; # optional flag
2138 # text? Normalize by obfusticating the fragment id to make it unique
2139 $text =~ s/\s+/_/sg;
2142 defined( $HC[ord($1)] ) ? $HC[ord($1)]
2143 : ( $HC[ord($1)] = sprintf( "%%%02X", ord($1) ) ) }gxe;
2144 $text = substr( $text, 0, 50 );
2150 # fragment_id - construct a fragment identifier from:
2152 # b) contents of C<...>
2157 my $generate = shift; # optional flag
2159 $text =~ s/\s+\Z//s;
2161 # a method or function?
2162 return $1 if $text =~ /(\w+)\s*\(/;
2163 return $1 if $text =~ /->\s*(\w+)\s*\(?/;
2166 return $1 if $text =~ /^([\$\@%*]\S+)/;
2168 # some pattern matching operator?
2169 return $1 if $text =~ m|^(\w+/).*/\w*$|;
2171 # fancy stuff... like "do { }"
2172 return $1 if $text =~ m|^(\w+)\s*{.*}$|;
2174 # honour the perlfunc manpage: func [PAR[,[ ]PAR]...]
2175 # and some funnies with ... Module ...
2176 return $1 if $text =~ m{^([a-z\d_]+)(\s+[A-Z,/& ][A-Z\d,/& ]*)?$};
2177 return $1 if $text =~ m{^([a-z\d]+)\s+Module(\s+[A-Z\d,/& ]+)?$};
2179 fragment_id_readable($text, $generate);
2186 # make_URL_href - generate HTML href from URL
2187 # Special treatment for CGI queries.
2189 sub make_URL_href($){
2192 s{^(http:[-\w/#~:.+=&%@!]+)(\?.*)$}{<a href="$1$2">$1</a>}i ){
2193 $url = "<a href=\"$url\">$url</a>";