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.
37 Pod::Html takes the following arguments:
43 --backlink="Back to Top"
45 Adds "Back to Top" links in front of every C<head1> heading (except for
46 the first). By default, no backlinks are generated.
52 Creates the item and directory caches in the given directory.
58 Specify the URL of a cascading style sheet. Also disables all HTML/CSS
59 C<style> attributes that are output by default (to avoid conflicts).
65 Flushes the item and directory caches.
72 Creates header and footer blocks containing the text of the C<NAME>
73 section. By default, no headers are generated.
79 Displays the usage message.
86 Include hidden directories in the search for POD's in podpath if recurse
88 The default is not to traverse any directory whose name begins with C<.>.
89 See L</"podpath"> and L</"recurse">.
91 [This option is for backward compatibility only.
92 It's hard to imagine that one would usefully create a module with a
93 name component beginning with C<.>.]
99 Sets the directory in which the resulting HTML file is placed. This
100 is used to generate relative links to other files. Not passing this
101 causes all links to be absolute, since this is the value that tells
102 Pod::Html the root of the documentation tree.
108 Sets the base URL for the HTML files. When cross-references are made,
109 the HTML root is prepended to the URL.
116 Generate an index at the top of the HTML file. This is the default
123 Specify the pod file to convert. Input is taken from STDIN if no
128 --libpods=name:...:name
130 List of page names (eg, "perlfunc") which contain linkable C<=item>s.
137 B<Deprecated>, has no effect. For backwards compatibility only.
143 Specify the HTML file to create. Output goes to STDOUT if no outfile
148 --podpath=name:...:name
150 Specify which subdirectories of the podroot contain pod files whose
151 HTML converted forms can be linked to in cross references.
157 Specify the base directory for finding library pods.
164 Don't display I<mostly harmless> warning messages. These messages
165 will be displayed by default. But this is not the same as C<verbose>
173 Recurse into subdirectories specified in podpath (default behaviour).
179 Specify the title of the resulting HTML file.
186 Display progress messages. By default, they won't be displayed.
193 "--podpath=lib:ext:pod:vms",
194 "--podroot=/usr/src/perl",
195 "--htmlroot=/perl/nmanual",
196 "--libpods=perlfunc:perlguts:perlvar:perlrun:perlop",
199 "--outfile=/perl/nmanual/foo.html");
203 Uses C<$Config{pod2html}> to setup default options.
207 Tom Christiansen, E<lt>tchrist@perl.comE<gt>.
215 This program is distributed under the Artistic License.
221 my($Dircache, $Itemcache);
224 my($Htmlroot, $Htmldir, $Htmlfile, $Htmlfileurl);
225 my($Podfile, @Podpath, $Podroot);
235 my($Listlevel, @Listend);
237 use vars qw($Ignore); # need to localize it later.
239 my(%Items_Named, @Items_Seen);
248 my %Pages = (); # associative array used to find the location
249 # of pages referenced by L<> links.
250 my %Items = (); # associative array used to find the location
251 # of =item directives referenced by C<> links
257 my $Curdir = File::Spec->curdir;
262 $Cachedir = "."; # The directory to which item and directory
263 # caches will be written.
265 $Dircache = "pod2htmd.tmp";
266 $Itemcache = "pod2htmi.tmp";
268 @Begin_Stack = (); # begin/end stack
270 @Libpods = (); # files to search for links from C<> directives
271 $Htmlroot = "/"; # http-server base directory from which all
272 # relative paths in $podpath stem.
273 $Htmldir = ""; # The directory to which the html pages
274 # will (eventually) be written.
275 $Htmlfile = ""; # write to stdout by default
276 $Htmlfileurl = "" ; # The url that other files would use to
277 # refer to this file. This is only used
278 # to make relative urls that point to
281 $Podfile = ""; # read from stdin by default
282 @Podpath = (); # list of directories containing library pods.
283 $Podroot = $Curdir; # filesystem base directory from which all
284 # relative paths in $podpath stem.
285 $Css = ''; # Cascading style sheet
286 $Recurse = 1; # recurse on subdirectories in $podpath.
287 $Quiet = 0; # not quiet by default
288 $Verbose = 0; # not verbose by default
289 $Doindex = 1; # non-zero if we should generate an index
290 $Backlink = ''; # text for "back to top" links
291 $Listlevel = 0; # current list depth
292 @Listend = (); # the text to use to end the list.
293 $After_Lpar = 0; # set to true after a par in an =item
294 $Ignore = 1; # whether or not to format text. we don't
295 # format text until we hit our first pod
298 @Items_Seen = (); # for multiples of the same item in perlfunc
300 $Header = 0; # produce block header/footer
301 $Title = ''; # title to give the pod(s)
302 $Top = 1; # true if we are at the top of the doc. used
303 # to prevent the first <hr /> directive.
304 $Paragraph = ''; # which paragraph we're processing (used
305 # for error messages)
306 $PTQuote = 0; # status of double-quote conversion
307 %Sections = (); # sections within this page
310 $Is83 = $^O eq 'dos'; # Is it an 8.3 filesystem?
314 # clean_data: global clean-up of pod data
318 for my $i ( 0..$#{$dataref} ) {
319 ${$dataref}[$i] =~ s/\s+\Z//;
321 # have a look for all-space lines
322 if( ${$dataref}[$i] =~ /^\s+$/m and $dataref->[$i] !~ /^\s/ ){
323 my @chunks = split( /^\s+$/m, ${$dataref}[$i] );
324 splice( @$dataref, $i, 1, @chunks );
337 $Is83 = 0 if (defined (&Dos::UseLFN) && Dos::UseLFN());
339 # cache of %Pages and %Items from last time we ran pod2html
341 #undef $opt_help if defined $opt_help;
343 # parse the command-line parameters
344 parse_command_line();
346 # escape the backlink argument (same goes for title but is done later...)
347 $Backlink = html_escape($Backlink) if defined $Backlink;
349 # set some variables to their default values if necessary
351 unless (@ARGV && $ARGV[0]) {
352 $Podfile = "-" unless $Podfile; # stdin
353 open(POD, "<$Podfile")
354 || die "$0: cannot open $Podfile file for input: $!\n";
356 $Podfile = $ARGV[0]; # XXX: might be more filenames
359 $Htmlfile = "-" unless $Htmlfile; # stdout
360 $Htmlroot = "" if $Htmlroot eq "/"; # so we don't get a //
361 $Htmldir =~ s#/\z## ; # so we don't get a //
363 && defined( $Htmldir )
365 && substr( $Htmlfile, 0, length( $Htmldir ) ) eq $Htmldir
368 # Set the 'base' url for this file, so that we can use it
369 # as the location from which to calculate relative links
370 # to other files. If this is '', then absolute links will
371 # be used throughout.
372 $Htmlfileurl= "$Htmldir/" . substr( $Htmlfile, length( $Htmldir ) + 1);
375 # read the pod a paragraph at a time
376 warn "Scanning for sections in input file(s)\n" if $Verbose;
385 @poddata = map { s/\r\n/\n/g;
387 map { "$_\n\n" } split /\n\n/ :
390 @poddata = map { s/\r/\n/g;
392 map { "$_\n\n" } split /\n\n/ :
399 clean_data( \@poddata );
401 # scan the pod for =head[1-6] directives and build an index
402 my $index = scan_headings(\%Sections, @poddata);
405 warn "No headings in $Podfile\n" if $Verbose;
408 # open the output file
409 open(HTML, ">$Htmlfile")
410 || die "$0: cannot open $Htmlfile file for output: $!\n";
412 # put a title in the HTML file if one wasn't specified
415 for (my $i = 0; $i < @poddata; $i++) {
416 if ($poddata[$i] =~ /^=head1\s*NAME\b/m) {
417 for my $para ( @poddata[$i, $i+1] ) {
419 if ($Title) = $para =~ /(\S+\s+-+.*\S)/s;
426 if (!$Title and $Podfile =~ /\.pod\z/) {
427 # probably a split pod so take first =head[12] as title
428 for (my $i = 0; $i < @poddata; $i++) {
429 last if ($Title) = $poddata[$i] =~ /^=head[12]\s*(.*)/;
431 warn "adopted '$Title' as title for $Podfile\n"
432 if $Verbose and $Title;
435 $Title =~ s/\s*\(.*\)//;
437 warn "$0: no title for $Podfile.\n" unless $Quiet;
438 $Podfile =~ /^(.*)(\.[^.\/]+)?\z/s;
439 $Title = ($Podfile eq "-" ? 'No Title' : $1);
440 warn "using $Title" if $Verbose;
442 $Title = html_escape($Title);
445 my $bodystyle = ' style="background-color: white"';
446 my $tdstyle = ' style="background-color: #cccccc"';
449 $csslink = qq(\n<link rel="stylesheet" href="$Css" type="text/css" />);
450 $csslink =~ s,\\,/,g;
451 $csslink =~ s,(/.):,$1|,;
456 my $block = $Header ? <<END_OF_BLOCK : '';
457 <table border="0" width="100%" cellspacing="0" cellpadding="3">
458 <tr><td class="block"$tdstyle valign="middle">
459 <big><strong><span class="block"> $Title</span></strong></big>
464 print HTML <<END_OF_HEAD;
465 <?xml version="1.0" ?>
466 <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
467 <html xmlns="http://www.w3.org/1999/xhtml">
469 <title>$Title</title>$csslink
470 <meta http-equiv="content-type" content="text/html; charset=utf-8" />
471 <link rev="made" href="mailto:$Config{perladmin}" />
478 # load/reload/validate/cache %Pages and %Items
479 get_cache($Dircache, $Itemcache, \@Podpath, $Podroot, $Recurse);
481 # scan the pod for =item directives
482 scan_items( \%Local_Items, "", @poddata);
484 # put an index at the top of the file. note, if $Doindex is 0 we
485 # still generate an index, but surround it with an html comment.
486 # that way some other program can extract it if desired.
488 print HTML "<p><a name=\"__index__\"></a></p>\n";
489 print HTML "<!-- INDEX BEGIN -->\n";
490 print HTML "<!--\n" unless $Doindex;
492 print HTML "-->\n" unless $Doindex;
493 print HTML "<!-- INDEX END -->\n\n";
494 print HTML "<hr />\n" if $Doindex and $index;
496 # now convert this file
497 my $after_item; # set to true after an =item
499 warn "Converting input file $Podfile\n" if $Verbose;
500 foreach my $i (0..$#poddata){
501 $PTQuote = 0; # status of quote conversion
505 if (/^(=.*)/s) { # is it a pod directive?
510 if (/^=begin\s+(\S+)\s*(.*)/si) {# =begin
511 process_begin($1, $2);
512 } elsif (/^=end\s+(\S+)\s*(.*)/si) {# =end
514 } elsif (/^=cut/) { # =cut
516 } elsif (/^=pod/) { # =pod
519 next if @Begin_Stack && $Begin_Stack[-1] ne 'html';
521 if (/^=(head[1-6])\s+(.*\S)/s) { # =head[1-6] heading
522 process_head( $1, $2, $Doindex && $index );
523 } elsif (/^=item\s*(.*\S)?/sm) { # =item text
524 $need_dd = process_item( $1 );
526 } elsif (/^=over\s*(.*)/) { # =over N
528 } elsif (/^=back/) { # =back
529 process_back($need_dd);
530 } elsif (/^=for\s+(\S+)\s*(.*)/si) {# =for
534 warn "$0: $Podfile: unknown pod directive '$1' in "
535 . "paragraph $Paragraph. ignoring.\n" unless $Quiet;
542 next if @Begin_Stack && $Begin_Stack[-1] ne 'html';
543 print HTML and next if @Begin_Stack && $Begin_Stack[-1] eq 'html';
544 print HTML "<dd>\n" if $need_dd;
546 if( $text =~ /\A\s+/ ){
547 process_pre( \$text );
548 print HTML "<pre>\n$text</pre>\n";
551 process_text( \$text );
553 # experimental: check for a paragraph where all lines
554 # have some ...\t...\t...\n pattern
556 my @lines = split( "\n", $text );
559 foreach my $line ( @lines ){
560 if( $line =~ /\S/ && $line !~ /\t/ ){
566 $text =~ s/\t+/<td>/g;
567 $text =~ s/^/<tr><td>/gm;
568 $text = '<table cellspacing="0" cellpadding="0">' .
573 ## end of experimental
578 print HTML "<p>$text</p>\n";
580 print HTML "</dd>\n" if $need_dd;
585 # finish off any pending directives
589 print HTML "<p><a href=\"#__index__\"><small>$Backlink</small></a></p>\n"
590 if $Doindex and $index and $Backlink;
592 print HTML <<END_OF_TAIL;
599 # close the html file
602 warn "Finished\n" if $Verbose;
605 ##############################################################################
609 warn "$0: $podfile: @_\n" if @_;
611 Usage: $0 --help --htmlroot=<name> --infile=<name> --outfile=<name>
612 --podpath=<name>:...:<name> --podroot=<name>
613 --libpods=<name>:...:<name> --recurse --verbose --index
614 --netscape --norecurse --noindex --cachedir=<name>
616 --backlink - set text for "back to top" links (default: none).
617 --cachedir - directory for the item and directory cache files.
618 --css - stylesheet URL
619 --flush - flushes the item and directory caches.
620 --[no]header - produce block header/footer (default is no headers).
621 --help - prints this message.
622 --hiddendirs - search hidden directories in podpath
623 --htmldir - directory for resulting HTML files.
624 --htmlroot - http-server base directory from which all relative paths
625 in podpath stem (default is /).
626 --[no]index - generate an index at the top of the resulting html
628 --infile - filename for the pod to convert (input taken from stdin
630 --libpods - colon-separated list of pages to search for =item pod
631 directives in as targets of C<> and implicit links (empty
632 by default). note, these are not filenames, but rather
633 page names like those that appear in L<> links.
634 --outfile - filename for the resulting html file (output sent to
636 --podpath - colon-separated list of directories containing library
637 pods (empty by default).
638 --podroot - filesystem base directory from which all relative paths
639 in podpath stem (default is .).
640 --[no]quiet - suppress some benign warning messages (default is off).
641 --[no]recurse - recurse on those subdirectories listed in podpath
643 --title - title that will appear in resulting html file.
644 --[no]verbose - self-explanatory (off by default).
645 --[no]netscape - deprecated, has no effect. for backwards compatibility only.
651 sub parse_command_line {
652 my ($opt_backlink,$opt_cachedir,$opt_css,$opt_flush,$opt_header,$opt_help,
653 $opt_htmldir,$opt_htmlroot,$opt_index,$opt_infile,$opt_libpods,
654 $opt_netscape,$opt_outfile,$opt_podpath,$opt_podroot,$opt_quiet,
655 $opt_recurse,$opt_title,$opt_verbose,$opt_hiddendirs);
657 unshift @ARGV, split ' ', $Config{pod2html} if $Config{pod2html};
658 my $result = GetOptions(
659 'backlink=s' => \$opt_backlink,
660 'cachedir=s' => \$opt_cachedir,
661 'css=s' => \$opt_css,
662 'flush' => \$opt_flush,
663 'header!' => \$opt_header,
664 'help' => \$opt_help,
665 'hiddendirs!'=> \$opt_hiddendirs,
666 'htmldir=s' => \$opt_htmldir,
667 'htmlroot=s' => \$opt_htmlroot,
668 'index!' => \$opt_index,
669 'infile=s' => \$opt_infile,
670 'libpods=s' => \$opt_libpods,
671 'netscape!' => \$opt_netscape,
672 'outfile=s' => \$opt_outfile,
673 'podpath=s' => \$opt_podpath,
674 'podroot=s' => \$opt_podroot,
675 'quiet!' => \$opt_quiet,
676 'recurse!' => \$opt_recurse,
677 'title=s' => \$opt_title,
678 'verbose!' => \$opt_verbose,
680 usage("-", "invalid parameters") if not $result;
682 usage("-") if defined $opt_help; # see if the user asked for help
683 $opt_help = ""; # just to make -w shut-up.
685 @Podpath = split(":", $opt_podpath) if defined $opt_podpath;
686 @Libpods = split(":", $opt_libpods) if defined $opt_libpods;
688 $Backlink = $opt_backlink if defined $opt_backlink;
689 $Cachedir = $opt_cachedir if defined $opt_cachedir;
690 $Css = $opt_css if defined $opt_css;
691 $Header = $opt_header if defined $opt_header;
692 $Htmldir = $opt_htmldir if defined $opt_htmldir;
693 $Htmlroot = $opt_htmlroot if defined $opt_htmlroot;
694 $Doindex = $opt_index if defined $opt_index;
695 $Podfile = $opt_infile if defined $opt_infile;
696 $HiddenDirs = $opt_hiddendirs if defined $opt_hiddendirs;
697 $Htmlfile = $opt_outfile if defined $opt_outfile;
698 $Podroot = $opt_podroot if defined $opt_podroot;
699 $Quiet = $opt_quiet if defined $opt_quiet;
700 $Recurse = $opt_recurse if defined $opt_recurse;
701 $Title = $opt_title if defined $opt_title;
702 $Verbose = $opt_verbose if defined $opt_verbose;
704 warn "Flushing item and directory caches\n"
705 if $opt_verbose && defined $opt_flush;
706 $Dircache = "$Cachedir/pod2htmd.tmp";
707 $Itemcache = "$Cachedir/pod2htmi.tmp";
708 if (defined $opt_flush) {
709 1 while unlink($Dircache, $Itemcache);
717 my($dircache, $itemcache, $podpath, $podroot, $recurse) = @_;
718 my @cache_key_args = @_;
720 # A first-level cache:
721 # Don't bother reading the cache files if they still apply
722 # and haven't changed since we last read them.
724 my $this_cache_key = cache_key(@cache_key_args);
726 return if $Saved_Cache_Key and $this_cache_key eq $Saved_Cache_Key;
728 # load the cache of %Pages and %Items if possible. $tests will be
729 # non-zero if successful.
731 if (-f $dircache && -f $itemcache) {
732 warn "scanning for item cache\n" if $Verbose;
733 $tests = load_cache($dircache, $itemcache, $podpath, $podroot);
736 # if we didn't succeed in loading the cache then we must (re)build
739 warn "scanning directories in pod-path\n" if $Verbose;
740 scan_podpath($podroot, $recurse, 0);
742 $Saved_Cache_Key = cache_key(@cache_key_args);
746 my($dircache, $itemcache, $podpath, $podroot, $recurse) = @_;
747 return join('!', $dircache, $itemcache, $recurse,
748 @$podpath, $podroot, stat($dircache), stat($itemcache));
752 # load_cache - tries to find if the caches stored in $dircache and $itemcache
753 # are valid caches of %Pages and %Items. if they are valid then it loads
754 # them and returns a non-zero value.
757 my($dircache, $itemcache, $podpath, $podroot) = @_;
763 open(CACHE, "<$itemcache") ||
764 die "$0: error opening $itemcache for reading: $!\n";
767 # is it the same podpath?
770 $tests++ if (join(":", @$podpath) eq $_);
772 # is it the same podroot?
775 $tests++ if ($podroot eq $_);
777 # load the cache if its good
783 warn "loading item cache\n" if $Verbose;
790 warn "scanning for directory cache\n" if $Verbose;
791 open(CACHE, "<$dircache") ||
792 die "$0: error opening $dircache for reading: $!\n";
796 # is it the same podpath?
799 $tests++ if (join(":", @$podpath) eq $_);
801 # is it the same podroot?
804 $tests++ if ($podroot eq $_);
806 # load the cache if its good
812 warn "loading directory cache\n" if $Verbose;
824 # scan_podpath - scans the directories specified in @podpath for directories,
825 # .pod files, and .pm files. it also scans the pod files specified in
826 # @Libpods for =item directives.
829 my($podroot, $recurse, $append) = @_;
831 my($libpod, $dirname, $pod, @files, @poddata);
838 # scan each directory listed in @Podpath
841 || die "$0: error changing to directory $podroot: $!\n";
842 foreach $dir (@Podpath) {
843 scan_dir($dir, $recurse);
846 # scan the pods listed in @Libpods for =item directives
847 foreach $libpod (@Libpods) {
848 # if the page isn't defined then we won't know where to find it
850 next unless defined $Pages{$libpod} && $Pages{$libpod};
852 # if there is a directory then use the .pod and .pm files within it.
853 # NOTE: Only finds the first so-named directory in the tree.
854 # if ($Pages{$libpod} =~ /([^:]*[^(\.pod|\.pm)]):/) {
855 if ($Pages{$libpod} =~ /([^:]*(?<!\.pod)(?<!\.pm)):/) {
856 # find all the .pod and .pm files within the directory
858 opendir(DIR, $dirname) ||
859 die "$0: error opening directory $dirname: $!\n";
860 @files = grep(/(\.pod|\.pm)\z/ && ! -d $_, readdir(DIR));
863 # scan each .pod and .pm file for =item directives
864 foreach $pod (@files) {
865 open(POD, "<$dirname/$pod") ||
866 die "$0: error opening $dirname/$pod for input: $!\n";
869 clean_data( \@poddata );
871 scan_items( \%Items, "$dirname/$pod", @poddata);
874 # use the names of files as =item directives too.
875 ### Don't think this should be done this way - confuses issues.(WL)
876 ### foreach $pod (@files) {
877 ### $pod =~ /^(.*)(\.pod|\.pm)$/;
878 ### $Items{$1} = "$dirname/$1.html" if $1;
880 } elsif ($Pages{$libpod} =~ /([^:]*\.pod):/ ||
881 $Pages{$libpod} =~ /([^:]*\.pm):/) {
882 # scan the .pod or .pm file for =item directives
884 open(POD, "<$pod") ||
885 die "$0: error opening $pod for input: $!\n";
888 clean_data( \@poddata );
890 scan_items( \%Items, "$pod", @poddata);
892 warn "$0: shouldn't be here (line ".__LINE__."\n" unless $Quiet;
895 @poddata = (); # clean-up a bit
898 || die "$0: error changing to directory $pwd: $!\n";
900 # cache the item list for later use
901 warn "caching items for later use\n" if $Verbose;
902 open(CACHE, ">$Itemcache") ||
903 die "$0: error open $Itemcache for writing: $!\n";
905 print CACHE join(":", @Podpath) . "\n$podroot\n";
906 foreach my $key (keys %Items) {
907 print CACHE "$key $Items{$key}\n";
912 # cache the directory list for later use
913 warn "caching directories for later use\n" if $Verbose;
914 open(CACHE, ">$Dircache") ||
915 die "$0: error open $Dircache for writing: $!\n";
917 print CACHE join(":", @Podpath) . "\n$podroot\n";
918 foreach my $key (keys %Pages) {
919 print CACHE "$key $Pages{$key}\n";
926 # scan_dir - scans the directory specified in $dir for subdirectories, .pod
927 # files, and .pm files. notes those that it finds. this information will
928 # be used later in order to figure out where the pages specified in L<>
929 # links are on the filesystem.
932 my($dir, $recurse) = @_;
933 my($t, @subdirs, @pods, $pod, $dirname, @dirs);
939 opendir(DIR, $dir) ||
940 die "$0: error opening directory $dir: $!\n";
941 while (defined($_ = readdir(DIR))) {
942 if (-d "$dir/$_" && $_ ne "." && $_ ne ".."
943 && ($HiddenDirs || !/^\./)
945 $Pages{$_} = "" unless defined $Pages{$_};
946 $Pages{$_} .= "$dir/$_:";
948 } elsif (/\.pod\z/) { # .pod
950 $Pages{$_} = "" unless defined $Pages{$_};
951 $Pages{$_} .= "$dir/$_.pod:";
952 push(@pods, "$dir/$_.pod");
953 } elsif (/\.html\z/) { # .html
955 $Pages{$_} = "" unless defined $Pages{$_};
956 $Pages{$_} .= "$dir/$_.pod:";
957 } elsif (/\.pm\z/) { # .pm
959 $Pages{$_} = "" unless defined $Pages{$_};
960 $Pages{$_} .= "$dir/$_.pm:";
961 push(@pods, "$dir/$_.pm");
962 } elsif (-T "$dir/$_") { # script(?)
964 if (open(F, "$dir/$_")) {
966 while (defined($line = <F>)) {
967 if ($line =~ /^=(?:pod|head1)/) {
968 $Pages{$_} = "" unless defined $Pages{$_};
969 $Pages{$_} .= "$dir/$_.pod:";
979 # recurse on the subdirectories if necessary
981 foreach my $subdir (@subdirs) {
982 scan_dir("$dir/$subdir", $recurse);
988 # scan_headings - scan a pod file for head[1-6] tags, note the tags, and
992 my($sections, @data) = @_;
993 my($tag, $which_head, $otitle, $listdepth, $index);
1000 # scan for =head directives, note their name, and build an index
1001 # pointing to each of them.
1002 foreach my $line (@data) {
1003 if ($line =~ /^=(head)([1-6])\s+(.*)/) {
1004 ($tag, $which_head, $otitle) = ($1,$2,$3);
1006 my $title = depod( $otitle );
1007 my $name = anchorify( $title );
1008 $$sections{$name} = 1;
1009 $title = process_text( \$otitle );
1011 while ($which_head != $listdepth) {
1012 if ($which_head > $listdepth) {
1013 $index .= "\n" . ("\t" x $listdepth) . "<ul>\n";
1015 } elsif ($which_head < $listdepth) {
1017 $index .= "\n" . ("\t" x $listdepth) . "</ul>\n";
1021 $index .= "\n" . ("\t" x $listdepth) . "<li>" .
1022 "<a href=\"#" . $name . "\">" .
1023 $title . "</a></li>";
1027 # finish off the lists
1028 while ($listdepth--) {
1029 $index .= "\n" . ("\t" x $listdepth) . "</ul>\n";
1032 # get rid of bogus lists
1033 $index =~ s,\t*<ul>\s*</ul>\n,,g;
1039 # scan_items - scans the pod specified by $pod for =item directives. we
1040 # will use this information later on in resolving C<> links.
1043 my( $itemref, $pod, @poddata ) = @_;
1047 $pod =~ s/\.pod\z//;
1048 $pod .= ".html" if $pod;
1050 foreach $i (0..$#poddata) {
1051 my $txt = depod( $poddata[$i] );
1053 # figure out what kind of item it is.
1054 # Build string for referencing this item.
1055 if ( $txt =~ /\A=item\s+\*\s*(.*)\Z/s ) { # bullet
1058 } elsif( $txt =~ /\A=item\s+(?>\d+\.?)\s*(.*)\Z/s ) { # numbered list
1060 } elsif( $txt =~ /\A=item\s+(.*)\Z/s ) { # plain item
1065 my $fid = fragment_id( $item );
1066 $$itemref{$fid} = "$pod" if $fid;
1071 # process_head - convert a pod head[1-6] tag and convert it to HTML format.
1074 my($tag, $heading, $hasindex) = @_;
1076 # figure out the level of the =head
1077 $tag =~ /head([1-6])/;
1081 warn "$0: $Podfile: unterminated list at =head in paragraph $Paragraph. ignoring.\n" unless $Quiet;
1082 while( $Listlevel ){
1088 if( $level == 1 && ! $Top ){
1089 print HTML "<a href=\"#__index__\"><small>$Backlink</small></a>\n"
1090 if $hasindex and $Backlink;
1091 print HTML "</p>\n<hr />\n"
1093 print HTML "</p>\n";
1096 my $name = anchorify( depod( $heading ) );
1097 my $convert = process_text( \$heading );
1098 print HTML "<h$level><a name=\"$name\">$convert</a></h$level>\n";
1103 # emit_item_tag - print an =item's text
1104 # Note: The global $EmittedItem is used for inhibiting self-references.
1108 sub emit_item_tag($$$){
1109 my( $otext, $text, $compact ) = @_;
1110 my $item = fragment_id( $text );
1112 $EmittedItem = $item;
1113 ### print STDERR "emit_item_tag=$item ($text)\n";
1115 print HTML '<strong>';
1116 if ($Items_Named{$item}++) {
1117 print HTML process_text( \$otext );
1119 my $name = 'item_' . $item;
1120 $name = anchorify($name);
1121 print HTML qq{<a name="$name">}, process_text( \$otext ), '</a>';
1123 print HTML "</strong>\n";
1124 undef( $EmittedItem );
1129 if( $Items_Seen[$Listlevel]++ == 0 ){
1130 push( @Listend, "</$tag>" );
1131 print HTML "<$tag>\n";
1133 my $emitted = $tag eq 'dl' ? 'dt' : 'li';
1134 print HTML "<$emitted>";
1139 # process_item - convert a pod item tag and convert it to HTML format.
1143 my $need_dd = 0; # set to 1 if we need a <dd></dd> after an item
1145 # lots of documents start a list without doing an =over. this is
1146 # bad! but, the proper thing to do seems to be to just assume
1147 # they did do an =over. so warn them once and then continue.
1148 if( $Listlevel == 0 ){
1149 warn "$0: $Podfile: unexpected =item directive in paragraph $Paragraph. ignoring.\n" unless $Quiet;
1153 # formatting: insert a paragraph if preceding item has >1 paragraph
1155 print HTML $need_dd ? "</dd>\n" : "</li>\n" if $After_Lpar;
1159 # remove formatting instructions from the text
1160 my $text = depod( $otext );
1162 my $emitted; # the tag actually emitted, used for closing
1164 # all the list variants:
1165 if( $text =~ /\A\*/ ){ # bullet
1166 $emitted = emit_li( 'ul' );
1167 if ($text =~ /\A\*\s+(.+)\Z/s ) { # with additional text
1169 $otext =~ s/\A\*\s+//;
1170 emit_item_tag( $otext, $tag, 1 );
1173 } elsif( $text =~ /\A\d+/ ){ # numbered list
1174 $emitted = emit_li( 'ol' );
1175 if ($text =~ /\A(?>\d+\.?)\s*(.+)\Z/s ) { # with additional text
1177 $otext =~ s/\A\d+\.?\s*//;
1178 emit_item_tag( $otext, $tag, 1 );
1181 } else { # definition list
1182 $emitted = emit_li( 'dl' );
1183 if ($text =~ /\A(.+)\Z/s ){ # should have text
1184 emit_item_tag( $otext, $text, 1 );
1194 # process_over - process a pod over tag and start a corresponding HTML list.
1199 push( @Items_Seen, 0 );
1204 # process_back - process a pod back tag and convert it to HTML format.
1207 my $need_dd = shift;
1208 if( $Listlevel == 0 ){
1209 warn "$0: $Podfile: unexpected =back directive in paragraph $Paragraph. ignoring.\n" unless $Quiet;
1213 # close off the list. note, I check to see if $Listend[$Listlevel] is
1214 # defined because an =item directive may have never appeared and thus
1215 # $Listend[$Listlevel] may have never been initialized.
1217 if( defined $Listend[$Listlevel] ){
1218 print HTML $need_dd ? "</dd>\n" : "</li>\n" if $After_Lpar;
1219 print HTML $Listend[$Listlevel];
1225 # clean up item count
1230 # process_cut - process a pod cut tag, thus start ignoring pod directives.
1237 # process_pod - process a pod tag, thus stop ignoring pod directives
1238 # until we see a corresponding cut.
1241 # no need to set $Ignore to 0 cause the main loop did it
1245 # process_for - process a =for pod tag. if it's for html, spit
1246 # it out verbatim, if illustration, center it, otherwise ignore it.
1249 my($whom, $text) = @_;
1250 if ( $whom =~ /^(pod2)?html$/i) {
1252 } elsif ($whom =~ /^illustration$/i) {
1253 1 while chomp $text;
1254 for my $ext (qw[.png .gif .jpeg .jpg .tga .pcl .bmp]) {
1255 $text .= $ext, last if -r "$text$ext";
1257 print HTML qq{<p align="center"><img src="$text" alt="$text illustration" /></p>};
1262 # process_begin - process a =begin pod tag. this pushes
1263 # whom we're beginning on the begin stack. if there's a
1264 # begin stack, we only print if it us.
1267 my($whom, $text) = @_;
1269 push (@Begin_Stack, $whom);
1270 if ( $whom =~ /^(pod2)?html$/) {
1271 print HTML $text if $text;
1276 # process_end - process a =end pod tag. pop the
1277 # begin stack. die if we're mismatched.
1280 my($whom, $text) = @_;
1282 if ($Begin_Stack[-1] ne $whom ) {
1283 die "Unmatched begin/end at chunk $Paragraph\n"
1285 pop( @Begin_Stack );
1289 # process_pre - indented paragraph, made into <pre></pre>
1298 # insert spaces in place of tabs
1301 1 while $line =~ s/(\t+)/' ' x ((length($1) * 8) - $-[0] % 8)/e;
1305 # convert some special chars to HTML escapes
1306 $rest = html_escape($rest);
1308 # try and create links for all occurrences of perl.* within
1309 # the preformatted text.
1313 if ( defined $Pages{$2} ){ # is a link
1314 qq($1<a href="$Htmlroot/$Pages{$2}">$2</a>);
1315 } elsif (defined $Pages{dosify($2)}) { # is a link
1316 qq($1<a href="$Htmlroot/$Pages{dosify($2)}">$2</a>);
1322 (<a\ href="?) ([^>:]*:)? ([^>:]*) \.pod: ([^>:]*:)?
1325 if ( $Htmlfileurl ne '' ){
1326 # Here, we take advantage of the knowledge
1327 # that $Htmlfileurl ne '' implies $Htmlroot eq ''.
1328 # Since $Htmlroot eq '', we need to prepend $Htmldir
1329 # on the fron of the link to get the absolute path
1330 # of the link's target. We check for a leading '/'
1331 # to avoid corrupting links that are #, file:, etc.
1333 $old_url = "$Htmldir$old_url" if $old_url =~ m{^\/};
1334 $url = relativize_url( "$old_url.html", $Htmlfileurl );
1341 # Look for embedded URLs and make them into links. We don't
1342 # relativize them since they are best left as the author intended.
1344 my $urls = '(' . join ('|', qw{
1357 my $gunk = '/#~:.?+=&%@!\-';
1358 my $punc = '.:!?\-;';
1359 my $any = "${ltrs}${gunk}${punc}";
1362 \b # start at word boundary
1364 $urls : # need resource and a colon
1365 (?!:) # Ignore File::, among others.
1366 [$any] +? # followed by one or more of any valid
1367 # character, but be conservative and
1368 # take only what you need to....
1371 " > # maybe pre-quoted '<a href="...">'
1373 [$punc]* # 0 or more punctuation
1375 [^$any] # by a non-url char
1377 $ # end of the string
1380 $ # then end of the string
1382 }{<a href="$1">$1</a>}igox;
1384 # text should be as it is (verbatim)
1390 # pure text processing
1392 # pure_text/inIS_text: differ with respect to automatic C<> recognition.
1393 # we don't want this to happen within IS
1397 process_puretext( $text, \$PTQuote, 1 );
1402 process_puretext( $text, \$PTQuote, 0 );
1406 # process_puretext - process pure text (without pod-escapes) converting
1407 # double-quotes and handling implicit C<> links.
1409 sub process_puretext {
1410 my($text, $quote, $notinIS) = @_;
1412 ## Guessing at func() or [$@%&]*var references in plain text is destined
1413 ## to produce some strange looking ref's. uncomment to disable:
1416 my(@words, $lead, $trail);
1418 # convert double-quotes to single-quotes
1419 if( $$quote && $text =~ s/"/''/s ){
1422 while ($text =~ s/"([^"]*)"/``$1''/sg) {};
1423 $$quote = 1 if $text =~ s/"/``/s;
1425 # keep track of leading and trailing white-space
1426 $lead = ($text =~ s/\A(\s+)//s ? $1 : "");
1427 $trail = ($text =~ s/(\s+)\Z//s ? $1 : "");
1429 # split at space/non-space boundaries
1430 @words = split( /(?<=\s)(?=\S)|(?<=\S)(?=\s)/, $text );
1432 # process each word individually
1433 foreach my $word (@words) {
1435 next if $word =~ /^\s*$/;
1436 # see if we can infer a link
1437 if( $notinIS && $word =~ /^(\w+)\((.*)\)$/ ) {
1438 # has parenthesis so should have been a C<> ref
1439 ## try for a pagename (perlXXX(1))?
1440 my( $func, $args ) = ( $1, $2 );
1441 if( $args =~ /^\d+$/ ){
1442 my $url = page_sect( $word, '' );
1444 $word = "<a href=\"$url\">the $word manpage</a>";
1448 ## try function name for a link, append tt'ed argument list
1449 $word = emit_C( $func, '', "($args)");
1451 #### disabled. either all (including $\W, $\w+{.*} etc.) or nothing.
1452 ## } elsif( $notinIS && $word =~ /^[\$\@%&*]+\w+$/) {
1453 ## # perl variables, should be a C<> ref
1454 ## $word = emit_C( $word );
1456 } elsif ($word =~ m,^\w+://\w,) {
1458 # Don't relativize it: leave it as the author intended
1459 $word = qq(<a href="$word">$word</a>);
1460 } elsif ($word =~ /[\w.-]+\@[\w-]+\.\w/) {
1461 # looks like an e-mail address
1462 my ($w1, $w2, $w3) = ("", $word, "");
1463 ($w1, $w2, $w3) = ("(", $1, ")$2") if $word =~ /^\((.*?)\)(,?)/;
1464 ($w1, $w2, $w3) = ("<", $1, ">$2") if $word =~ /^<(.*?)>(,?)/;
1465 $word = qq($w1<a href="mailto:$w2">$w2</a>$w3);
1467 $word = html_escape($word) if $word =~ /["&<>]/;
1471 # put everything back together
1472 return $lead . join( '', @words ) . $trail;
1477 # process_text - handles plaintext that appears in the input pod file.
1478 # there may be pod commands embedded within the text so those must be
1479 # converted to html commands.
1482 sub process_text1($$;$$);
1483 sub pattern ($) { $_[0] ? '[^\S\n]+'.('>' x ($_[0] + 1)) : '>' }
1484 sub closing ($) { local($_) = shift; (defined && s/\s+$//) ? length : 0 }
1489 my $res = process_text1( 0, $tref );
1493 sub process_text1($$;$$){
1494 my( $lev, $rstr, $func, $closing ) = @_;
1497 unless (defined $func) {
1503 # B<text> - boldface
1504 $res = '<strong>' . process_text1( $lev, $rstr ) . '</strong>';
1506 } elsif( $func eq 'C' ){
1507 # C<code> - can be a ref or <code></code>
1508 # need to extract text
1509 my $par = go_ahead( $rstr, 'C', $closing );
1511 ## clean-up of the link target
1512 my $text = depod( $par );
1514 ### my $x = $par =~ /[BI]</ ? 'yes' : 'no' ;
1515 ### print STDERR "-->call emit_C($par) lev=$lev, par with BI=$x\n";
1517 $res = emit_C( $text, $lev > 1 || ($par =~ /[BI]</) );
1519 } elsif( $func eq 'E' ){
1520 # E<x> - convert to character
1521 $$rstr =~ s/^([^>]*)>//;
1523 $escape =~ s/^(\d+|X[\dA-F]+)$/#$1/i;
1526 } elsif( $func eq 'F' ){
1527 # F<filename> - italizice
1528 $res = '<em>' . process_text1( $lev, $rstr ) . '</em>';
1530 } elsif( $func eq 'I' ){
1531 # I<text> - italizice
1532 $res = '<em>' . process_text1( $lev, $rstr ) . '</em>';
1534 } elsif( $func eq 'L' ){
1536 ## L<text|cross-ref> => produce text, use cross-ref for linking
1537 ## L<cross-ref> => make text from cross-ref
1538 ## need to extract text
1539 my $par = go_ahead( $rstr, 'L', $closing );
1541 # some L<>'s that shouldn't be:
1542 # a) full-blown URL's are emitted as-is
1543 if( $par =~ m{^\w+://}s ){
1544 return make_URL_href( $par );
1546 # b) C<...> is stripped and treated as C<>
1547 if( $par =~ /^C<(.*)>$/ ){
1548 my $text = depod( $1 );
1549 return emit_C( $text, $lev > 1 || ($par =~ /[BI]</) );
1552 # analyze the contents
1553 $par =~ s/\n/ /g; # undo word-wrapped tags
1556 if( $par =~ s{^([^|]+)\|}{} ){
1560 # make sure sections start with a /
1563 my( $page, $section, $ident );
1565 # check for link patterns
1566 if( $par =~ m{^([^/]+?)/(?!")(.*?)$} ){ # name/ident
1567 # we've got a name/ident (no quotes)
1568 ( $page, $ident ) = ( $1, $2 );
1569 ### print STDERR "--> L<$par> to page $page, ident $ident\n";
1571 } elsif( $par =~ m{^(.*?)/"?(.*?)"?$} ){ # [name]/"section"
1572 # even though this should be a "section", we go for ident first
1573 ( $page, $ident ) = ( $1, $2 );
1574 ### print STDERR "--> L<$par> to page $page, section $section\n";
1576 } elsif( $par =~ /\s/ ){ # this must be a section with missing quotes
1577 ( $page, $section ) = ( '', $par );
1578 ### print STDERR "--> L<$par> to void page, section $section\n";
1581 ( $page, $section ) = ( $par, '' );
1582 ### print STDERR "--> L<$par> to page $par, void section\n";
1585 # now, either $section or $ident is defined. the convoluted logic
1586 # below tries to resolve L<> according to what the user specified.
1587 # failing this, we try to find the next best thing...
1588 my( $url, $ltext, $fid );
1591 if( defined $ident ){
1592 ## try to resolve $ident as an item
1593 ( $url, $fid ) = coderef( $page, $ident );
1595 if( ! defined( $linktext ) ){
1597 $linktext .= " in " if $ident && $page;
1598 $linktext .= "the $page manpage" if $page;
1600 ### print STDERR "got coderef url=$url\n";
1603 ## no luck: go for a section (auto-quoting!)
1606 ## now go for a section
1607 my $htmlsection = htmlify( $section );
1608 $url = page_sect( $page, $htmlsection );
1610 if( ! defined( $linktext ) ){
1611 $linktext = $section;
1612 $linktext .= " in " if $section && $page;
1613 $linktext .= "the $page manpage" if $page;
1615 ### print STDERR "got page/section url=$url\n";
1618 ## no luck: go for an ident
1625 ( $url, $fid ) = coderef( $page, $ident );
1627 if( ! defined( $linktext ) ){
1629 $linktext .= " in " if $ident && $page;
1630 $linktext .= "the $page manpage" if $page;
1632 ### print STDERR "got section=>coderef url=$url\n";
1636 # warning; show some text.
1637 $linktext = $opar unless defined $linktext;
1638 warn "$0: $Podfile: cannot resolve L<$opar> in paragraph $Paragraph.\n" unless $Quiet;
1641 # now we have a URL or just plain code
1642 $$rstr = $linktext . '>' . $$rstr;
1643 if( defined( $url ) ){
1644 $res = "<a href=\"$url\">" . process_text1( $lev, $rstr ) . '</a>';
1646 $res = '<em>' . process_text1( $lev, $rstr ) . '</em>';
1649 } elsif( $func eq 'S' ){
1650 # S<text> - non-breaking spaces
1651 $res = process_text1( $lev, $rstr );
1652 $res =~ s/ / /g;
1654 } elsif( $func eq 'X' ){
1656 $$rstr =~ s/^[^>]*>//;
1658 } elsif( $func eq 'Z' ){
1660 warn "$0: $Podfile: invalid X<> in paragraph $Paragraph.\n"
1661 unless $$rstr =~ s/^>// or $Quiet;
1664 my $term = pattern $closing;
1665 while( $$rstr =~ s/\A(.*?)(([BCEFILSXZ])<(<+[^\S\n]+)?|$term)//s ){
1666 # all others: either recurse into new function or
1667 # terminate at closing angle bracket(s)
1669 $pt .= $2 if !$3 && $lev == 1;
1670 $res .= $lev == 1 ? pure_text( $pt ) : inIS_text( $pt );
1671 return $res if !$3 && $lev > 1;
1673 $res .= process_text1( $lev, $rstr, $3, closing $4 );
1677 $res .= pure_text( $$rstr );
1679 warn "$0: $Podfile: undelimited $func<> in paragraph $Paragraph.\n" unless $Quiet;
1686 # go_ahead: extract text of an IS (can be nested)
1689 my( $rstr, $func, $closing ) = @_;
1691 my @closing = ($closing);
1693 s/\A(.*?)(([BCEFILSXZ])<(<+[^\S\n]+)?|@{[pattern $closing[0]]})//s ){
1697 return $res unless @closing;
1699 unshift @closing, closing $4;
1703 warn "$0: $Podfile: undelimited $func<> in paragraph $Paragraph.\n" unless $Quiet;
1708 # emit_C - output result of C<text>
1709 # $text is the depod-ed text
1712 my( $text, $nocode, $args ) = @_;
1713 $args = '' unless defined $args;
1715 my( $url, $fid ) = coderef( undef(), $text );
1717 # need HTML-safe text
1718 my $linktext = html_escape( "$text$args" );
1720 if( defined( $url ) &&
1721 (!defined( $EmittedItem ) || $EmittedItem ne $fid ) ){
1722 $res = "<a href=\"$url\"><code>$linktext</code></a>";
1723 } elsif( 0 && $nocode ){
1726 $res = "<code>$linktext</code>";
1732 # html_escape: make text safe for HTML
1736 $rest =~ s/&/&/g;
1737 $rest =~ s/</</g;
1738 $rest =~ s/>/>/g;
1739 $rest =~ s/"/"/g;
1740 # ' is only in XHTML, not HTML4. Be conservative
1741 #$rest =~ s/'/'/g;
1747 # dosify - convert filenames to 8.3
1751 return lc($str) if $^O eq 'VMS'; # VMS just needs casing
1754 $str =~ s/(\.\w+)/substr ($1,0,4)/ge;
1755 $str =~ s/(\w+)/substr ($1,0,8)/ge;
1761 # page_sect - make a URL from the text of a L<>
1764 my( $page, $section ) = @_;
1765 my( $linktext, $page83, $link); # work strings
1767 # check if we know that this is a section in this page
1768 if (!defined $Pages{$page} && defined $Sections{$page}) {
1771 ### print STDERR "reset page='', section=$section\n";
1774 $page83=dosify($page);
1775 $page=$page83 if (defined $Pages{$page83});
1777 $link = "#" . anchorify( $section );
1778 } elsif ( $page =~ /::/ ) {
1780 # Search page cache for an entry keyed under the html page name,
1781 # then look to see what directory that page might be in. NOTE:
1782 # this will only find one page. A better solution might be to produce
1783 # an intermediate page that is an index to all such pages.
1784 my $page_name = $page ;
1785 $page_name =~ s,^.*/,,s ;
1786 if ( defined( $Pages{ $page_name } ) &&
1787 $Pages{ $page_name } =~ /([^:]*$page)\.(?:pod|pm):/
1792 # NOTE: This branch assumes that all A::B pages are located in
1793 # $Htmlroot/A/B.html . This is often incorrect, since they are
1794 # often in $Htmlroot/lib/A/B.html or such like. Perhaps we could
1795 # analyze the contents of %Pages and figure out where any
1796 # cousins of A::B are, then assume that. So, if A::B isn't found,
1797 # but A::C is found in lib/A/C.pm, then A::B is assumed to be in
1798 # lib/A/B.pm. This is also limited, but it's an improvement.
1799 # Maybe a hints file so that the links point to the correct places
1803 $link = "$Htmlroot/$page.html";
1804 $link .= "#" . anchorify( $section ) if ($section);
1805 } elsif (!defined $Pages{$page}) {
1808 $section = anchorify( $section ) if $section ne "";
1809 ### print STDERR "...section=$section\n";
1811 # if there is a directory by the name of the page, then assume that an
1812 # appropriate section will exist in the subdirectory
1813 # if ($section ne "" && $Pages{$page} =~ /([^:]*[^(\.pod|\.pm)]):/) {
1814 if ($section ne "" && $Pages{$page} =~ /([^:]*(?<!\.pod)(?<!\.pm)):/) {
1815 $link = "$Htmlroot/$1/$section.html";
1816 ### print STDERR "...link=$link\n";
1818 # since there is no directory by the name of the page, the section will
1819 # have to exist within a .html of the same name. thus, make sure there
1820 # is a .pod or .pm that might become that .html
1822 $section = "#$section" if $section;
1823 ### print STDERR "...section=$section\n";
1825 # check if there is a .pod with the page name
1826 if ($Pages{$page} =~ /([^:]*)\.pod:/) {
1827 $link = "$Htmlroot/$1.html$section";
1828 } elsif ($Pages{$page} =~ /([^:]*)\.pm:/) {
1829 $link = "$Htmlroot/$1.html$section";
1837 # Here, we take advantage of the knowledge that $Htmlfileurl ne ''
1838 # implies $Htmlroot eq ''. This means that the link in question
1839 # needs a prefix of $Htmldir if it begins with '/'. The test for
1840 # the initial '/' is done to avoid '#'-only links, and to allow
1841 # for other kinds of links, like file:, ftp:, etc.
1843 if ( $Htmlfileurl ne '' ) {
1844 $link = "$Htmldir$link" if $link =~ m{^/}s;
1845 $url = relativize_url( $link, $Htmlfileurl );
1846 # print( " b: [$link,$Htmlfileurl,$url]\n" );
1859 # relativize_url - convert an absolute URL to one relative to a base URL.
1860 # Assumes both end in a filename.
1862 sub relativize_url {
1863 my ($dest,$source) = @_ ;
1865 my ($dest_volume,$dest_directory,$dest_file) =
1866 File::Spec::Unix->splitpath( $dest ) ;
1867 $dest = File::Spec::Unix->catpath( $dest_volume, $dest_directory, '' ) ;
1869 my ($source_volume,$source_directory,$source_file) =
1870 File::Spec::Unix->splitpath( $source ) ;
1871 $source = File::Spec::Unix->catpath( $source_volume, $source_directory, '' ) ;
1874 if ( $dest ne '' ) {
1875 $rel_path = File::Spec::Unix->abs2rel( $dest, $source ) ;
1878 if ( $rel_path ne '' &&
1879 substr( $rel_path, -1 ) ne '/' &&
1880 substr( $dest_file, 0, 1 ) ne '#'
1882 $rel_path .= "/$dest_file" ;
1885 $rel_path .= "$dest_file" ;
1893 # coderef - make URL from the text of a C<>
1896 my( $page, $item ) = @_;
1899 my $fid = fragment_id( $item );
1900 if( defined( $page ) && $page ne "" ){
1901 # we have been given a $page...
1904 # Do we take it? Item could be a section!
1905 my $base = $Items{$fid} || "";
1906 $base =~ s{[^/]*/}{};
1907 if( $base ne "$page.html" ){
1908 ### print STDERR "coderef( $page, $item ): items{$fid} = $Items{$fid} = $base => discard page!\n";
1913 # no page - local items precede cached items
1914 if( defined( $fid ) ){
1915 if( exists $Local_Items{$fid} ){
1916 $page = $Local_Items{$fid};
1918 $page = $Items{$fid};
1923 # if there was a pod file that we found earlier with an appropriate
1924 # =item directive, then create a link to that page.
1925 if( defined $page ){
1927 if( exists $Pages{$page} and $Pages{$page} =~ /([^:.]*)\.[^:]*:/){
1928 $page = $1 . '.html';
1930 my $link = "$Htmlroot/$page#item_" . anchorify($fid);
1932 # Here, we take advantage of the knowledge that $Htmlfileurl
1933 # ne '' implies $Htmlroot eq ''.
1934 if ( $Htmlfileurl ne '' ) {
1935 $link = "$Htmldir$link" ;
1936 $url = relativize_url( $link, $Htmlfileurl ) ;
1941 $url = "#item_" . anchorify($fid);
1944 confess "url has space: $url" if $url =~ /"[^"]*\s[^"]*"/;
1946 return( $url, $fid );
1952 # Adapted from Nick Ing-Simmons' PodToHtml package.
1954 my $source_file = shift ;
1955 my $destination_file = shift;
1957 my $source = URI::file->new_abs($source_file);
1958 my $uo = URI::file->new($destination_file,$source)->abs;
1959 return $uo->rel->as_string;
1964 # finish_list - finish off any pending HTML lists. this should be called
1965 # after the entire pod file has been read and converted.
1968 while ($Listlevel > 0) {
1969 print HTML "</dl>\n";
1975 # htmlify - converts a pod section specification to a suitable section
1976 # specification for HTML. Note that we keep spaces and special characters
1977 # except ", ? (Netscape problem) and the hyphen (writer's problem...).
1981 $heading =~ s/(\s+)/ /g;
1982 $heading =~ s/\s+\Z//;
1983 $heading =~ s/\A\s+//;
1984 # The hyphen is a disgrace to the English language.
1985 $heading =~ s/[-"?]//g;
1986 $heading = lc( $heading );
1991 # similar to htmlify, but turns non-alphanumerics into underscores
1995 $anchor = htmlify($anchor);
1996 $anchor =~ s/\W/_/g;
2001 # depod - convert text by eliminating all interior sequences
2002 # Note: can be called with copy or modify semantics
2009 $E2c{amp} = '&'; # in Tk's pods
2017 ${$_[0]} = depod1( \$string );
2025 my( $rstr, $func, $closing ) = @_;
2027 return $res unless defined $$rstr;
2028 if( ! defined( $func ) ){
2029 # skip to next begin of an interior sequence
2030 while( $$rstr =~ s/\A(.*?)([BCEFILSXZ])<(<+[^\S\n]+)?// ){
2031 # recurse into its text
2032 $res .= $1 . depod1( $rstr, $2, closing $3);
2035 } elsif( $func eq 'E' ){
2036 # E<x> - convert to character
2037 $$rstr =~ s/^([^>]*)>//;
2038 $res .= $E2c{$1} || "";
2039 } elsif( $func eq 'X' ){
2041 $$rstr =~ s/^[^>]*>//;
2042 } elsif( $func eq 'Z' ){
2046 # all others: either recurse into new function or
2047 # terminate at closing angle bracket
2048 my $term = pattern $closing;
2049 while( $$rstr =~ s/\A(.*?)(([BCEFILSXZ])<(<+[^\S\n]+)?|$term)// ){
2052 $res .= depod1( $rstr, $3, closing $4 );
2054 ## If we're here and $2 ne '>': undelimited interior sequence.
2055 ## Ignored, as this is called without proper indication of where we are.
2056 ## Rely on process_text to produce diagnostics.
2062 # fragment_id - construct a fragment identifier from:
2064 # b) contents of C<...>
2069 $text =~ s/\s+\Z//s;
2071 # a method or function?
2072 return $1 if $text =~ /(\w+)\s*\(/;
2073 return $1 if $text =~ /->\s*(\w+)\s*\(?/;
2076 return $1 if $text =~ /^([$@%*]\S+)/;
2078 # some pattern matching operator?
2079 return $1 if $text =~ m|^(\w+/).*/\w*$|;
2081 # fancy stuff... like "do { }"
2082 return $1 if $text =~ m|^(\w+)\s*{.*}$|;
2084 # honour the perlfunc manpage: func [PAR[,[ ]PAR]...]
2085 # and some funnies with ... Module ...
2086 return $1 if $text =~ m{^([a-z\d_]+)(\s+[A-Z\d,/& ]+)?$};
2087 return $1 if $text =~ m{^([a-z\d]+)\s+Module(\s+[A-Z\d,/& ]+)?$};
2090 $text =~ s/\s+/_/sg;
2092 defined( $HC[ord($1)] ) ? $HC[ord($1)]
2093 : ( $HC[ord($1)] = sprintf( "%%%02X", ord($1) ) ) }gxe;
2094 $text = substr( $text, 0, 50 );
2101 # make_URL_href - generate HTML href from URL
2102 # Special treatment for CGI queries.
2104 sub make_URL_href($){
2107 s{^(http:[-\w/#~:.+=&%@!]+)(\?.*)$}{<a href="$1$2">$1</a>}i ){
2108 $url = "<a href=\"$url\">$url</a>";