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, @Listtype);
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 @Listtype = (); # list types for open lists
306 $ListNewTerm = 0; # indicates new term in definition list; used
307 # to correctly open/close <dd> tags
308 $Ignore = 1; # whether or not to format text. we don't
309 # format text until we hit our first pod
312 @Items_Seen = (); # for multiples of the same item in perlfunc
314 $Header = 0; # produce block header/footer
315 $Title = ''; # title to give the pod(s)
316 $Top = 1; # true if we are at the top of the doc. used
317 # to prevent the first <hr /> directive.
318 $Paragraph = ''; # which paragraph we're processing (used
319 # for error messages)
320 %Sections = (); # sections within this page
323 $Is83 = $^O eq 'dos'; # Is it an 8.3 filesystem?
327 # clean_data: global clean-up of pod data
331 for my $i ( 0..$#{$dataref} ) {
332 ${$dataref}[$i] =~ s/\s+\Z//;
334 # have a look for all-space lines
335 if( ${$dataref}[$i] =~ /^\s+$/m and $dataref->[$i] !~ /^\s/ ){
336 my @chunks = split( /^\s+$/m, ${$dataref}[$i] );
337 splice( @$dataref, $i, 1, @chunks );
350 $Is83 = 0 if (defined (&Dos::UseLFN) && Dos::UseLFN());
352 # cache of %Pages and %Items from last time we ran pod2html
354 #undef $opt_help if defined $opt_help;
356 # parse the command-line parameters
357 parse_command_line();
359 # escape the backlink argument (same goes for title but is done later...)
360 $Backlink = html_escape($Backlink) if defined $Backlink;
362 # set some variables to their default values if necessary
364 unless (@ARGV && $ARGV[0]) {
365 $Podfile = "-" unless $Podfile; # stdin
366 open(POD, "<$Podfile")
367 || die "$0: cannot open $Podfile file for input: $!\n";
369 $Podfile = $ARGV[0]; # XXX: might be more filenames
372 $Htmlfile = "-" unless $Htmlfile; # stdout
373 $Htmlroot = "" if $Htmlroot eq "/"; # so we don't get a //
374 $Htmldir =~ s#/\z## ; # so we don't get a //
376 && defined( $Htmldir )
378 && substr( $Htmlfile, 0, length( $Htmldir ) ) eq $Htmldir
381 # Set the 'base' url for this file, so that we can use it
382 # as the location from which to calculate relative links
383 # to other files. If this is '', then absolute links will
384 # be used throughout.
385 $Htmlfileurl= "$Htmldir/" . substr( $Htmlfile, length( $Htmldir ) + 1);
388 # read the pod a paragraph at a time
389 warn "Scanning for sections in input file(s)\n" if $Verbose;
398 @poddata = map { s/\r\n/\n/g;
400 map { "$_\n\n" } split /\n\n/ :
403 @poddata = map { s/\r/\n/g;
405 map { "$_\n\n" } split /\n\n/ :
412 clean_data( \@poddata );
414 # scan the pod for =head[1-6] directives and build an index
415 my $index = scan_headings(\%Sections, @poddata);
418 warn "No headings in $Podfile\n" if $Verbose;
421 # open the output file
422 open(HTML, ">$Htmlfile")
423 || die "$0: cannot open $Htmlfile file for output: $!\n";
425 # put a title in the HTML file if one wasn't specified
428 for (my $i = 0; $i < @poddata; $i++) {
429 if ($poddata[$i] =~ /^=head1\s*NAME\b/m) {
430 for my $para ( @poddata[$i, $i+1] ) {
432 if ($Title) = $para =~ /(\S+\s+-+.*\S)/s;
439 if (!$Title and $Podfile =~ /\.pod\z/) {
440 # probably a split pod so take first =head[12] as title
441 for (my $i = 0; $i < @poddata; $i++) {
442 last if ($Title) = $poddata[$i] =~ /^=head[12]\s*(.*)/;
444 warn "adopted '$Title' as title for $Podfile\n"
445 if $Verbose and $Title;
448 $Title =~ s/\s*\(.*\)//;
450 warn "$0: no title for $Podfile.\n" unless $Quiet;
451 $Podfile =~ /^(.*)(\.[^.\/]+)?\z/s;
452 $Title = ($Podfile eq "-" ? 'No Title' : $1);
453 warn "using $Title" if $Verbose;
455 $Title = html_escape($Title);
458 my $bodystyle = ' style="background-color: white"';
459 my $tdstyle = ' style="background-color: #cccccc"';
462 $csslink = qq(\n<link rel="stylesheet" href="$Css" type="text/css" />);
463 $csslink =~ s,\\,/,g;
464 $csslink =~ s,(/.):,$1|,;
469 my $block = $Header ? <<END_OF_BLOCK : '';
470 <table border="0" width="100%" cellspacing="0" cellpadding="3">
471 <tr><td class="block"$tdstyle valign="middle">
472 <big><strong><span class="block"> $Title</span></strong></big>
477 print HTML <<END_OF_HEAD;
478 <?xml version="1.0" ?>
479 <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
480 <html xmlns="http://www.w3.org/1999/xhtml">
482 <title>$Title</title>$csslink
483 <meta http-equiv="content-type" content="text/html; charset=utf-8" />
484 <link rev="made" href="mailto:$Config{perladmin}" />
491 # load/reload/validate/cache %Pages and %Items
492 get_cache($Dircache, $Itemcache, \@Podpath, $Podroot, $Recurse);
494 # scan the pod for =item directives
495 scan_items( \%Local_Items, "", @poddata);
497 # put an index at the top of the file. note, if $Doindex is 0 we
498 # still generate an index, but surround it with an html comment.
499 # that way some other program can extract it if desired.
502 my $hr = ($Doindex and $index) ? qq(<hr name="index" />) : "";
506 $index = qq(<!--\n$index\n-->\n);
509 print HTML << "END_OF_INDEX";
513 <p><a name=\"__index__\"></a></p>
521 # now convert this file
522 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?
531 if (/^=begin\s+(\S+)\s*(.*)/si) {# =begin
532 process_begin($1, $2);
533 } elsif (/^=end\s+(\S+)\s*(.*)/si) {# =end
535 } elsif (/^=cut/) { # =cut
537 } elsif (/^=pod/) { # =pod
540 next if @Begin_Stack && $Begin_Stack[-1] ne 'html';
542 if (/^=(head[1-6])\s+(.*\S)/s) { # =head[1-6] heading
543 process_head( $1, $2, $Doindex && $index );
544 } elsif (/^=item\s*(.*\S)?/sm) { # =item text
547 } elsif (/^=over\s*(.*)/) { # =over N
549 } elsif (/^=back/) { # =back
551 } elsif (/^=for\s+(\S+)\s*(.*)/si) {# =for
555 warn "$0: $Podfile: unknown pod directive '$1' in "
556 . "paragraph $Paragraph. ignoring.\n" unless $Quiet;
563 next if @Begin_Stack && $Begin_Stack[-1] ne 'html';
564 print HTML and next if @Begin_Stack && $Begin_Stack[-1] eq 'html';
567 # Open tag for definition list as we have something to put in it
573 if( $text =~ /\A\s+/ ){
574 process_pre( \$text );
575 print HTML "<pre>\n$text</pre>\n";
578 process_text( \$text );
580 # experimental: check for a paragraph where all lines
581 # have some ...\t...\t...\n pattern
583 my @lines = split( "\n", $text );
586 foreach my $line ( @lines ){
587 if( $line =~ /\S/ && $line !~ /\t/ ){
593 $text =~ s/\t+/<td>/g;
594 $text =~ s/^/<tr><td>/gm;
595 $text = '<table cellspacing="0" cellpadding="0">' .
600 ## end of experimental
602 print HTML "<p>$text</p>\n";
608 # finish off any pending directives
612 print HTML "<p><a href=\"#__index__\"><small>$Backlink</small></a></p>\n"
613 if $Doindex and $index and $Backlink;
615 print HTML <<END_OF_TAIL;
622 # close the html file
625 warn "Finished\n" if $Verbose;
628 ##############################################################################
632 warn "$0: $podfile: @_\n" if @_;
634 Usage: $0 --help --htmlroot=<name> --infile=<name> --outfile=<name>
635 --podpath=<name>:...:<name> --podroot=<name>
636 --libpods=<name>:...:<name> --recurse --verbose --index
637 --netscape --norecurse --noindex --cachedir=<name>
639 --backlink - set text for "back to top" links (default: none).
640 --cachedir - directory for the item and directory cache files.
641 --css - stylesheet URL
642 --flush - flushes the item and directory caches.
643 --[no]header - produce block header/footer (default is no headers).
644 --help - prints this message.
645 --hiddendirs - search hidden directories in podpath
646 --htmldir - directory for resulting HTML files.
647 --htmlroot - http-server base directory from which all relative paths
648 in podpath stem (default is /).
649 --[no]index - generate an index at the top of the resulting html
651 --infile - filename for the pod to convert (input taken from stdin
653 --libpods - colon-separated list of pages to search for =item pod
654 directives in as targets of C<> and implicit links (empty
655 by default). note, these are not filenames, but rather
656 page names like those that appear in L<> links.
657 --outfile - filename for the resulting html file (output sent to
659 --podpath - colon-separated list of directories containing library
660 pods (empty by default).
661 --podroot - filesystem base directory from which all relative paths
662 in podpath stem (default is .).
663 --[no]quiet - suppress some benign warning messages (default is off).
664 --[no]recurse - recurse on those subdirectories listed in podpath
666 --title - title that will appear in resulting html file.
667 --[no]verbose - self-explanatory (off by default).
668 --[no]netscape - deprecated, has no effect. for backwards compatibility only.
674 sub parse_command_line {
675 my ($opt_backlink,$opt_cachedir,$opt_css,$opt_flush,$opt_header,$opt_help,
676 $opt_htmldir,$opt_htmlroot,$opt_index,$opt_infile,$opt_libpods,
677 $opt_netscape,$opt_outfile,$opt_podpath,$opt_podroot,$opt_quiet,
678 $opt_recurse,$opt_title,$opt_verbose,$opt_hiddendirs);
680 unshift @ARGV, split ' ', $Config{pod2html} if $Config{pod2html};
681 my $result = GetOptions(
682 'backlink=s' => \$opt_backlink,
683 'cachedir=s' => \$opt_cachedir,
684 'css=s' => \$opt_css,
685 'flush' => \$opt_flush,
686 'header!' => \$opt_header,
687 'help' => \$opt_help,
688 'hiddendirs!'=> \$opt_hiddendirs,
689 'htmldir=s' => \$opt_htmldir,
690 'htmlroot=s' => \$opt_htmlroot,
691 'index!' => \$opt_index,
692 'infile=s' => \$opt_infile,
693 'libpods=s' => \$opt_libpods,
694 'netscape!' => \$opt_netscape,
695 'outfile=s' => \$opt_outfile,
696 'podpath=s' => \$opt_podpath,
697 'podroot=s' => \$opt_podroot,
698 'quiet!' => \$opt_quiet,
699 'recurse!' => \$opt_recurse,
700 'title=s' => \$opt_title,
701 'verbose!' => \$opt_verbose,
703 usage("-", "invalid parameters") if not $result;
705 usage("-") if defined $opt_help; # see if the user asked for help
706 $opt_help = ""; # just to make -w shut-up.
708 @Podpath = split(":", $opt_podpath) if defined $opt_podpath;
709 @Libpods = split(":", $opt_libpods) if defined $opt_libpods;
711 $Backlink = $opt_backlink if defined $opt_backlink;
712 $Cachedir = $opt_cachedir if defined $opt_cachedir;
713 $Css = $opt_css if defined $opt_css;
714 $Header = $opt_header if defined $opt_header;
715 $Htmldir = $opt_htmldir if defined $opt_htmldir;
716 $Htmlroot = $opt_htmlroot if defined $opt_htmlroot;
717 $Doindex = $opt_index if defined $opt_index;
718 $Podfile = $opt_infile if defined $opt_infile;
719 $HiddenDirs = $opt_hiddendirs if defined $opt_hiddendirs;
720 $Htmlfile = $opt_outfile if defined $opt_outfile;
721 $Podroot = $opt_podroot if defined $opt_podroot;
722 $Quiet = $opt_quiet if defined $opt_quiet;
723 $Recurse = $opt_recurse if defined $opt_recurse;
724 $Title = $opt_title if defined $opt_title;
725 $Verbose = $opt_verbose if defined $opt_verbose;
727 warn "Flushing item and directory caches\n"
728 if $opt_verbose && defined $opt_flush;
729 $Dircache = "$Cachedir/pod2htmd.tmp";
730 $Itemcache = "$Cachedir/pod2htmi.tmp";
731 if (defined $opt_flush) {
732 1 while unlink($Dircache, $Itemcache);
740 my($dircache, $itemcache, $podpath, $podroot, $recurse) = @_;
741 my @cache_key_args = @_;
743 # A first-level cache:
744 # Don't bother reading the cache files if they still apply
745 # and haven't changed since we last read them.
747 my $this_cache_key = cache_key(@cache_key_args);
749 return if $Saved_Cache_Key and $this_cache_key eq $Saved_Cache_Key;
751 # load the cache of %Pages and %Items if possible. $tests will be
752 # non-zero if successful.
754 if (-f $dircache && -f $itemcache) {
755 warn "scanning for item cache\n" if $Verbose;
756 $tests = load_cache($dircache, $itemcache, $podpath, $podroot);
759 # if we didn't succeed in loading the cache then we must (re)build
762 warn "scanning directories in pod-path\n" if $Verbose;
763 scan_podpath($podroot, $recurse, 0);
765 $Saved_Cache_Key = cache_key(@cache_key_args);
769 my($dircache, $itemcache, $podpath, $podroot, $recurse) = @_;
770 return join('!', $dircache, $itemcache, $recurse,
771 @$podpath, $podroot, stat($dircache), stat($itemcache));
775 # load_cache - tries to find if the caches stored in $dircache and $itemcache
776 # are valid caches of %Pages and %Items. if they are valid then it loads
777 # them and returns a non-zero value.
780 my($dircache, $itemcache, $podpath, $podroot) = @_;
786 open(CACHE, "<$itemcache") ||
787 die "$0: error opening $itemcache for reading: $!\n";
790 # is it the same podpath?
793 $tests++ if (join(":", @$podpath) eq $_);
795 # is it the same podroot?
798 $tests++ if ($podroot eq $_);
800 # load the cache if its good
806 warn "loading item cache\n" if $Verbose;
813 warn "scanning for directory cache\n" if $Verbose;
814 open(CACHE, "<$dircache") ||
815 die "$0: error opening $dircache for reading: $!\n";
819 # is it the same podpath?
822 $tests++ if (join(":", @$podpath) eq $_);
824 # is it the same podroot?
827 $tests++ if ($podroot eq $_);
829 # load the cache if its good
835 warn "loading directory cache\n" if $Verbose;
847 # scan_podpath - scans the directories specified in @podpath for directories,
848 # .pod files, and .pm files. it also scans the pod files specified in
849 # @Libpods for =item directives.
852 my($podroot, $recurse, $append) = @_;
854 my($libpod, $dirname, $pod, @files, @poddata);
861 # scan each directory listed in @Podpath
864 || die "$0: error changing to directory $podroot: $!\n";
865 foreach $dir (@Podpath) {
866 scan_dir($dir, $recurse);
869 # scan the pods listed in @Libpods for =item directives
870 foreach $libpod (@Libpods) {
871 # if the page isn't defined then we won't know where to find it
873 next unless defined $Pages{$libpod} && $Pages{$libpod};
875 # if there is a directory then use the .pod and .pm files within it.
876 # NOTE: Only finds the first so-named directory in the tree.
877 # if ($Pages{$libpod} =~ /([^:]*[^(\.pod|\.pm)]):/) {
878 if ($Pages{$libpod} =~ /([^:]*(?<!\.pod)(?<!\.pm)):/) {
879 # find all the .pod and .pm files within the directory
881 opendir(DIR, $dirname) ||
882 die "$0: error opening directory $dirname: $!\n";
883 @files = grep(/(\.pod|\.pm)\z/ && ! -d $_, readdir(DIR));
886 # scan each .pod and .pm file for =item directives
887 foreach $pod (@files) {
888 open(POD, "<$dirname/$pod") ||
889 die "$0: error opening $dirname/$pod for input: $!\n";
892 clean_data( \@poddata );
894 scan_items( \%Items, "$dirname/$pod", @poddata);
897 # use the names of files as =item directives too.
898 ### Don't think this should be done this way - confuses issues.(WL)
899 ### foreach $pod (@files) {
900 ### $pod =~ /^(.*)(\.pod|\.pm)$/;
901 ### $Items{$1} = "$dirname/$1.html" if $1;
903 } elsif ($Pages{$libpod} =~ /([^:]*\.pod):/ ||
904 $Pages{$libpod} =~ /([^:]*\.pm):/) {
905 # scan the .pod or .pm file for =item directives
907 open(POD, "<$pod") ||
908 die "$0: error opening $pod for input: $!\n";
911 clean_data( \@poddata );
913 scan_items( \%Items, "$pod", @poddata);
915 warn "$0: shouldn't be here (line ".__LINE__."\n" unless $Quiet;
918 @poddata = (); # clean-up a bit
921 || die "$0: error changing to directory $pwd: $!\n";
923 # cache the item list for later use
924 warn "caching items for later use\n" if $Verbose;
925 open(CACHE, ">$Itemcache") ||
926 die "$0: error open $Itemcache for writing: $!\n";
928 print CACHE join(":", @Podpath) . "\n$podroot\n";
929 foreach my $key (keys %Items) {
930 print CACHE "$key $Items{$key}\n";
935 # cache the directory list for later use
936 warn "caching directories for later use\n" if $Verbose;
937 open(CACHE, ">$Dircache") ||
938 die "$0: error open $Dircache for writing: $!\n";
940 print CACHE join(":", @Podpath) . "\n$podroot\n";
941 foreach my $key (keys %Pages) {
942 print CACHE "$key $Pages{$key}\n";
949 # scan_dir - scans the directory specified in $dir for subdirectories, .pod
950 # files, and .pm files. notes those that it finds. this information will
951 # be used later in order to figure out where the pages specified in L<>
952 # links are on the filesystem.
955 my($dir, $recurse) = @_;
956 my($t, @subdirs, @pods, $pod, $dirname, @dirs);
962 opendir(DIR, $dir) ||
963 die "$0: error opening directory $dir: $!\n";
964 while (defined($_ = readdir(DIR))) {
965 if (-d "$dir/$_" && $_ ne "." && $_ ne ".."
966 && ($HiddenDirs || !/^\./)
968 $Pages{$_} = "" unless defined $Pages{$_};
969 $Pages{$_} .= "$dir/$_:";
971 } elsif (/\.pod\z/) { # .pod
973 $Pages{$_} = "" unless defined $Pages{$_};
974 $Pages{$_} .= "$dir/$_.pod:";
975 push(@pods, "$dir/$_.pod");
976 } elsif (/\.html\z/) { # .html
978 $Pages{$_} = "" unless defined $Pages{$_};
979 $Pages{$_} .= "$dir/$_.pod:";
980 } elsif (/\.pm\z/) { # .pm
982 $Pages{$_} = "" unless defined $Pages{$_};
983 $Pages{$_} .= "$dir/$_.pm:";
984 push(@pods, "$dir/$_.pm");
985 } elsif (-T "$dir/$_") { # script(?)
987 if (open(F, "$dir/$_")) {
989 while (defined($line = <F>)) {
990 if ($line =~ /^=(?:pod|head1)/) {
991 $Pages{$_} = "" unless defined $Pages{$_};
992 $Pages{$_} .= "$dir/$_.pod:";
1002 # recurse on the subdirectories if necessary
1004 foreach my $subdir (@subdirs) {
1005 scan_dir("$dir/$subdir", $recurse);
1011 # scan_headings - scan a pod file for head[1-6] tags, note the tags, and
1015 my($sections, @data) = @_;
1016 my($tag, $which_head, $otitle, $listdepth, $index);
1023 # scan for =head directives, note their name, and build an index
1024 # pointing to each of them.
1025 foreach my $line (@data) {
1026 if ($line =~ /^=(head)([1-6])\s+(.*)/) {
1027 ($tag, $which_head, $otitle) = ($1,$2,$3);
1029 my $title = depod( $otitle );
1030 my $name = anchorify( $title );
1031 $$sections{$name} = 1;
1032 $title = process_text( \$otitle );
1034 while ($which_head != $listdepth) {
1035 if ($which_head > $listdepth) {
1036 $index .= "\n" . ("\t" x $listdepth) . "<ul>\n";
1038 } elsif ($which_head < $listdepth) {
1040 $index .= "\n" . ("\t" x $listdepth) . "</ul>\n";
1044 $index .= "\n" . ("\t" x $listdepth) . "<li>" .
1045 "<a href=\"#" . $name . "\">" .
1046 $title . "</a></li>";
1050 # finish off the lists
1051 while ($listdepth--) {
1052 $index .= "\n" . ("\t" x $listdepth) . "</ul>\n";
1055 # get rid of bogus lists
1056 $index =~ s,\t*<ul>\s*</ul>\n,,g;
1062 # scan_items - scans the pod specified by $pod for =item directives. we
1063 # will use this information later on in resolving C<> links.
1066 my( $itemref, $pod, @poddata ) = @_;
1070 $pod =~ s/\.pod\z//;
1071 $pod .= ".html" if $pod;
1073 foreach $i (0..$#poddata) {
1074 my $txt = depod( $poddata[$i] );
1076 # figure out what kind of item it is.
1077 # Build string for referencing this item.
1078 if ( $txt =~ /\A=item\s+\*\s*(.*)\Z/s ) { # bulleted list
1081 } elsif( $txt =~ /\A=item\s+(?>\d+\.?)\s*(.*)\Z/s ) { # numbered list
1083 } elsif( $txt =~ /\A=item\s+(.*)\Z/s ) { # definition list
1088 my $fid = fragment_id( $item );
1089 $$itemref{$fid} = "$pod" if $fid;
1094 # process_head - convert a pod head[1-6] tag and convert it to HTML format.
1097 my($tag, $heading, $hasindex) = @_;
1099 # figure out the level of the =head
1100 $tag =~ /head([1-6])/;
1106 if( $level == 1 && ! $Top ){
1107 print HTML "<a href=\"#__index__\"><small>$Backlink</small></a>\n"
1108 if $hasindex and $Backlink;
1109 print HTML "</p>\n<hr />\n"
1111 print HTML "</p>\n";
1114 my $name = anchorify( depod( $heading ) );
1115 my $convert = process_text( \$heading );
1116 print HTML "<h$level><a name=\"$name\">$convert</a></h$level>\n";
1121 # emit_item_tag - print an =item's text
1122 # Note: The global $EmittedItem is used for inhibiting self-references.
1126 sub emit_item_tag($$$){
1127 my( $otext, $text, $compact ) = @_;
1128 my $item = fragment_id( depod($text) , -generate);
1129 Carp::confess("Undefined fragment '$text' (".depod($text).") from fragment_id() in emit_item_tag() in $Podfile")
1131 $EmittedItem = $item;
1132 ### print STDERR "emit_item_tag=$item ($text)\n";
1134 print HTML '<strong>';
1135 if ($Items_Named{$item}++) {
1136 print HTML process_text( \$otext );
1139 $name = anchorify($name);
1140 print HTML qq{<a name="$name" class="item">}, process_text( \$otext ), '</a>';
1142 print HTML "</strong>";
1143 undef( $EmittedItem );
1148 # Open tag for definition list as we have something to put in it
1149 if( ($tag ne 'dl') && ($ListNewTerm) ){
1150 print HTML "<dd>\n";
1154 if( $Items_Seen[$Listlevel]++ == 0 ){
1156 push( @Listtype, "$tag" );
1157 print HTML "<$tag>\n";
1159 # if this is not the first item, close the previous one
1160 if ( $tag eq 'dl' ){
1161 print HTML "</dd>\n" unless $ListNewTerm;
1163 print HTML "</li>\n";
1166 my $opentag = $tag eq 'dl' ? 'dt' : 'li';
1167 print HTML "<$opentag>";
1171 # process_item - convert a pod item tag and convert it to HTML format.
1176 # lots of documents start a list without doing an =over. this is
1177 # bad! but, the proper thing to do seems to be to just assume
1178 # they did do an =over. so warn them once and then continue.
1179 if( $Listlevel == 0 ){
1180 warn "$0: $Podfile: unexpected =item directive in paragraph $Paragraph. ignoring.\n" unless $Quiet;
1184 # remove formatting instructions from the text
1185 my $text = depod( $otext );
1187 # all the list variants:
1188 if( $text =~ /\A\*/ ){ # bullet
1189 new_listitem( 'ul' );
1190 if ($text =~ /\A\*\s+(\S.*)\Z/s ) { # with additional text
1192 $otext =~ s/\A\*\s+//;
1193 emit_item_tag( $otext, $tag, 1 );
1197 } elsif( $text =~ /\A\d+/ ){ # numbered list
1198 new_listitem( 'ol' );
1199 if ($text =~ /\A(?>\d+\.?)\s*(\S.*)\Z/s ) { # with additional text
1201 $otext =~ s/\A\d+\.?\s*//;
1202 emit_item_tag( $otext, $tag, 1 );
1206 } else { # definition list
1207 # new_listitem takes care of opening the <dt> tag
1208 new_listitem( 'dl' );
1209 if ($text =~ /\A(.+)\Z/s ){ # should have text
1210 emit_item_tag( $otext, $text, 1 );
1211 # write the definition term and close <dt> tag
1212 print HTML "</dt>\n";
1214 # trigger opening a <dd> tag for the actual definition; will not
1215 # happen if next paragraph is also a definition term (=item)
1222 # process_over - process a pod over tag and start a corresponding HTML list.
1227 push( @Items_Seen, 0 );
1231 # process_back - process a pod back tag and convert it to HTML format.
1234 if( $Listlevel == 0 ){
1235 warn "$0: $Podfile: unexpected =back directive in paragraph $Paragraph. ignoring.\n" unless $Quiet;
1239 # close off the list. note, I check to see if $Listtype[$Listlevel] is
1240 # defined because an =item directive may have never appeared and thus
1241 # $Listtype[$Listlevel] may have never been initialized.
1243 if( defined $Listtype[$Listlevel] ){
1244 if ( $Listtype[$Listlevel] eq 'dl' ){
1245 print HTML "</dd>\n" unless $ListNewTerm;
1247 print HTML "</li>\n";
1249 print HTML "</$Listtype[$Listlevel]>\n";
1254 # clean up item count
1259 # process_cut - process a pod cut tag, thus start ignoring pod directives.
1266 # process_pod - process a pod tag, thus stop ignoring pod directives
1267 # until we see a corresponding cut.
1270 # no need to set $Ignore to 0 cause the main loop did it
1274 # process_for - process a =for pod tag. if it's for html, spit
1275 # it out verbatim, if illustration, center it, otherwise ignore it.
1278 my($whom, $text) = @_;
1279 if ( $whom =~ /^(pod2)?html$/i) {
1281 } elsif ($whom =~ /^illustration$/i) {
1282 1 while chomp $text;
1283 for my $ext (qw[.png .gif .jpeg .jpg .tga .pcl .bmp]) {
1284 $text .= $ext, last if -r "$text$ext";
1286 print HTML qq{<p align="center"><img src="$text" alt="$text illustration" /></p>};
1291 # process_begin - process a =begin pod tag. this pushes
1292 # whom we're beginning on the begin stack. if there's a
1293 # begin stack, we only print if it us.
1296 my($whom, $text) = @_;
1298 push (@Begin_Stack, $whom);
1299 if ( $whom =~ /^(pod2)?html$/) {
1300 print HTML $text if $text;
1305 # process_end - process a =end pod tag. pop the
1306 # begin stack. die if we're mismatched.
1309 my($whom, $text) = @_;
1311 if (!defined $Begin_Stack[-1] or $Begin_Stack[-1] ne $whom ) {
1312 Carp::confess("Unmatched begin/end at chunk $Paragraph in pod $Podfile\n")
1314 pop( @Begin_Stack );
1318 # process_pre - indented paragraph, made into <pre></pre>
1327 # insert spaces in place of tabs
1330 1 while $line =~ s/(\t+)/' ' x ((length($1) * 8) - $-[0] % 8)/e;
1334 # convert some special chars to HTML escapes
1335 $rest = html_escape($rest);
1337 # try and create links for all occurrences of perl.* within
1338 # the preformatted text.
1342 if ( defined $Pages{$2} ){ # is a link
1343 qq($1<a href="$Htmlroot/$Pages{$2}">$2</a>);
1344 } elsif (defined $Pages{dosify($2)}) { # is a link
1345 qq($1<a href="$Htmlroot/$Pages{dosify($2)}">$2</a>);
1351 (<a\ href="?) ([^>:]*:)? ([^>:]*) \.pod: ([^>:]*:)?
1354 if ( $Htmlfileurl ne '' ){
1355 # Here, we take advantage of the knowledge
1356 # that $Htmlfileurl ne '' implies $Htmlroot eq ''.
1357 # Since $Htmlroot eq '', we need to prepend $Htmldir
1358 # on the fron of the link to get the absolute path
1359 # of the link's target. We check for a leading '/'
1360 # to avoid corrupting links that are #, file:, etc.
1362 $old_url = "$Htmldir$old_url" if $old_url =~ m{^\/};
1363 $url = relativize_url( "$old_url.html", $Htmlfileurl );
1370 # Look for embedded URLs and make them into links. We don't
1371 # relativize them since they are best left as the author intended.
1373 my $urls = '(' . join ('|', qw{
1386 my $gunk = '/#~:.?+=&%@!\-';
1387 my $punc = '.:!?\-;';
1388 my $any = "${ltrs}${gunk}${punc}";
1391 \b # start at word boundary
1393 $urls : # need resource and a colon
1394 (?!:) # Ignore File::, among others.
1395 [$any] +? # followed by one or more of any valid
1396 # character, but be conservative and
1397 # take only what you need to....
1400 " > # maybe pre-quoted '<a href="...">'
1402 [$punc]* # 0 or more punctuation
1404 [^$any] # by a non-url char
1406 $ # end of the string
1409 $ # then end of the string
1411 }{<a href="$1">$1</a>}igox;
1413 # text should be as it is (verbatim)
1419 # pure text processing
1421 # pure_text/inIS_text: differ with respect to automatic C<> recognition.
1422 # we don't want this to happen within IS
1426 process_puretext( $text, 1 );
1431 process_puretext( $text, 0 );
1435 # process_puretext - process pure text (without pod-escapes) converting
1436 # double-quotes and handling implicit C<> links.
1438 sub process_puretext {
1439 my($text, $notinIS) = @_;
1441 ## Guessing at func() or [\$\@%&]*var references in plain text is destined
1442 ## to produce some strange looking ref's. uncomment to disable:
1445 my(@words, $lead, $trail);
1447 # keep track of leading and trailing white-space
1448 $lead = ($text =~ s/\A(\s+)//s ? $1 : "");
1449 $trail = ($text =~ s/(\s+)\Z//s ? $1 : "");
1451 # split at space/non-space boundaries
1452 @words = split( /(?<=\s)(?=\S)|(?<=\S)(?=\s)/, $text );
1454 # process each word individually
1455 foreach my $word (@words) {
1457 next if $word =~ /^\s*$/;
1458 # see if we can infer a link or a function call
1460 # NOTE: This is a word based search, it won't automatically
1461 # mark "substr($var, 1, 2)" because the 1st word would be "substr($var"
1462 # User has to enclose those with proper C<>
1464 if( $notinIS && $word =~
1466 ^([a-z_]{2,}) # The function name
1468 ([0-9][a-z]* # Manual page(1) or page(1M)
1469 |[^)]*[\$\@\%][^)]+ # ($foo), (1, @foo), (%hash)
1473 ([.,;]?)$ # a possible punctuation follows
1476 # has parenthesis so should have been a C<> ref
1477 ## try for a pagename (perlXXX(1))?
1478 my( $func, $args, $rest ) = ( $1, $2, $3 || '' );
1479 if( $args =~ /^\d+$/ ){
1480 my $url = page_sect( $word, '' );
1482 $word = qq(<a href="$url" class="man">the $word manpage</a>$rest);
1486 ## try function name for a link, append tt'ed argument list
1487 $word = emit_C( $func, '', "($args)") . $rest;
1489 #### disabled. either all (including $\W, $\w+{.*} etc.) or nothing.
1490 ## } elsif( $notinIS && $word =~ /^[\$\@%&*]+\w+$/) {
1491 ## # perl variables, should be a C<> ref
1492 ## $word = emit_C( $word );
1494 } elsif ($word =~ m,^\w+://\w,) {
1496 # Don't relativize it: leave it as the author intended
1497 $word = qq(<a href="$word">$word</a>);
1498 } elsif ($word =~ /[\w.-]+\@[\w-]+\.\w/) {
1499 # looks like an e-mail address
1500 my ($w1, $w2, $w3) = ("", $word, "");
1501 ($w1, $w2, $w3) = ("(", $1, ")$2") if $word =~ /^\((.*?)\)(,?)/;
1502 ($w1, $w2, $w3) = ("<", $1, ">$2") if $word =~ /^<(.*?)>(,?)/;
1503 $word = qq($w1<a href="mailto:$w2">$w2</a>$w3);
1505 $word = html_escape($word) if $word =~ /["&<>]/;
1509 # put everything back together
1510 return $lead . join( '', @words ) . $trail;
1515 # process_text - handles plaintext that appears in the input pod file.
1516 # there may be pod commands embedded within the text so those must be
1517 # converted to html commands.
1520 sub process_text1($$;$$);
1521 sub pattern ($) { $_[0] ? '\s+'.('>' x ($_[0] + 1)) : '>' }
1522 sub closing ($) { local($_) = shift; (defined && s/\s+\z//) ? length : 0 }
1527 my $res = process_text1( 0, $tref );
1532 sub process_text_rfc_links {
1535 # For every "RFCnnnn" or "RFC nnn", link it to the authoritative
1536 # ource. Do not use the /i modifier here. Require "RFC" to be written in
1537 # in capital letters.
1540 (?<=[^<>[:alpha:]]) # Make sure this is not an URL already
1541 (RFC\s*([0-9]{1,5}))(?![0-9]) # max 5 digits
1543 {<a href="http://www.ietf.org/rfc/rfc$2.txt" class="rfc">$1</a>}gx;
1548 sub process_text1($$;$$){
1549 my( $lev, $rstr, $func, $closing ) = @_;
1552 unless (defined $func) {
1558 # B<text> - boldface
1559 $res = '<strong>' . process_text1( $lev, $rstr ) . '</strong>';
1561 } elsif( $func eq 'C' ){
1562 # C<code> - can be a ref or <code></code>
1563 # need to extract text
1564 my $par = go_ahead( $rstr, 'C', $closing );
1566 ## clean-up of the link target
1567 my $text = depod( $par );
1569 ### my $x = $par =~ /[BI]</ ? 'yes' : 'no' ;
1570 ### print STDERR "-->call emit_C($par) lev=$lev, par with BI=$x\n";
1572 $res = emit_C( $text, $lev > 1 || ($par =~ /[BI]</) );
1574 } elsif( $func eq 'E' ){
1575 # E<x> - convert to character
1576 $$rstr =~ s/^([^>]*)>//;
1578 $escape =~ s/^(\d+|X[\dA-F]+)$/#$1/i;
1581 } elsif( $func eq 'F' ){
1582 # F<filename> - italicize
1583 $res = '<em class="file">' . process_text1( $lev, $rstr ) . '</em>';
1585 } elsif( $func eq 'I' ){
1586 # I<text> - italicize
1587 $res = '<em>' . process_text1( $lev, $rstr ) . '</em>';
1589 } elsif( $func eq 'L' ){
1591 ## L<text|cross-ref> => produce text, use cross-ref for linking
1592 ## L<cross-ref> => make text from cross-ref
1593 ## need to extract text
1594 my $par = go_ahead( $rstr, 'L', $closing );
1596 # some L<>'s that shouldn't be:
1597 # a) full-blown URL's are emitted as-is
1598 if( $par =~ m{^\w+://}s ){
1599 return make_URL_href( $par );
1601 # b) C<...> is stripped and treated as C<>
1602 if( $par =~ /^C<(.*)>$/ ){
1603 my $text = depod( $1 );
1604 return emit_C( $text, $lev > 1 || ($par =~ /[BI]</) );
1607 # analyze the contents
1608 $par =~ s/\n/ /g; # undo word-wrapped tags
1611 if( $par =~ s{^([^|]+)\|}{} ){
1615 # make sure sections start with a /
1618 my( $page, $section, $ident );
1620 # check for link patterns
1621 if( $par =~ m{^([^/]+?)/(?!")(.*?)$} ){ # name/ident
1622 # we've got a name/ident (no quotes)
1624 ( $page, $ident ) = ( $1, $2 );
1626 ( $page, $section ) = ( $1, $2 );
1628 ### print STDERR "--> L<$par> to page $page, ident $ident\n";
1630 } elsif( $par =~ m{^(.*?)/"?(.*?)"?$} ){ # [name]/"section"
1631 # even though this should be a "section", we go for ident first
1632 ( $page, $ident ) = ( $1, $2 );
1633 ### print STDERR "--> L<$par> to page $page, section $section\n";
1635 } elsif( $par =~ /\s/ ){ # this must be a section with missing quotes
1636 ( $page, $section ) = ( '', $par );
1637 ### print STDERR "--> L<$par> to void page, section $section\n";
1640 ( $page, $section ) = ( $par, '' );
1641 ### print STDERR "--> L<$par> to page $par, void section\n";
1644 # now, either $section or $ident is defined. the convoluted logic
1645 # below tries to resolve L<> according to what the user specified.
1646 # failing this, we try to find the next best thing...
1647 my( $url, $ltext, $fid );
1650 if( defined $ident ){
1651 ## try to resolve $ident as an item
1652 ( $url, $fid ) = coderef( $page, $ident );
1654 if( ! defined( $linktext ) ){
1656 $linktext .= " in " if $ident && $page;
1657 $linktext .= "the $page manpage" if $page;
1659 ### print STDERR "got coderef url=$url\n";
1662 ## no luck: go for a section (auto-quoting!)
1665 ## now go for a section
1666 my $htmlsection = htmlify( $section );
1667 $url = page_sect( $page, $htmlsection );
1669 if( ! defined( $linktext ) ){
1670 $linktext = $section;
1671 $linktext .= " in " if $section && $page;
1672 $linktext .= "the $page manpage" if $page;
1674 ### print STDERR "got page/section url=$url\n";
1677 ## no luck: go for an ident
1684 ( $url, $fid ) = coderef( $page, $ident );
1686 if( ! defined( $linktext ) ){
1688 $linktext .= " in " if $ident && $page;
1689 $linktext .= "the $page manpage" if $page;
1691 ### print STDERR "got section=>coderef url=$url\n";
1695 # warning; show some text.
1696 $linktext = $opar unless defined $linktext;
1697 warn "$0: $Podfile: cannot resolve L<$opar> in paragraph $Paragraph.\n" unless $Quiet;
1700 # now we have a URL or just plain code
1701 $$rstr = $linktext . '>' . $$rstr;
1702 if( defined( $url ) ){
1703 $res = "<a href=\"$url\">" . process_text1( $lev, $rstr ) . '</a>';
1705 $res = '<em>' . process_text1( $lev, $rstr ) . '</em>';
1708 } elsif( $func eq 'S' ){
1709 # S<text> - non-breaking spaces
1710 $res = process_text1( $lev, $rstr );
1711 $res =~ s/ / /g;
1713 } elsif( $func eq 'X' ){
1715 warn "$0: $Podfile: invalid X<> in paragraph $Paragraph.\n"
1716 unless $$rstr =~ s/^[^>]*>// or $Quiet;
1717 } elsif( $func eq 'Z' ){
1719 warn "$0: $Podfile: invalid Z<> in paragraph $Paragraph.\n"
1720 unless $$rstr =~ s/^>// or $Quiet;
1723 my $term = pattern $closing;
1724 while( $$rstr =~ s/\A(.*?)(([BCEFILSXZ])<(<+[^\S\n]+)?|$term)//s ){
1725 # all others: either recurse into new function or
1726 # terminate at closing angle bracket(s)
1728 $pt .= $2 if !$3 && $lev == 1;
1729 $res .= $lev == 1 ? pure_text( $pt ) : inIS_text( $pt );
1730 return $res if !$3 && $lev > 1;
1732 $res .= process_text1( $lev, $rstr, $3, closing $4 );
1736 $res .= pure_text( $$rstr );
1737 } elsif( ! $Quiet ) {
1738 my $snippet = substr($$rstr,0,60);
1739 warn "$0: $Podfile: undelimited $func<> in paragraph $Paragraph: '$snippet'.\n"
1742 $res = process_text_rfc_links($res);
1748 # go_ahead: extract text of an IS (can be nested)
1751 my( $rstr, $func, $closing ) = @_;
1753 my @closing = ($closing);
1755 s/\A(.*?)(([BCEFILSXZ])<(<+\s+)?|@{[pattern $closing[0]]})//s ){
1759 return $res unless @closing;
1761 unshift @closing, closing $4;
1766 my $snippet = substr($$rstr,0,60);
1767 warn "$0: $Podfile: undelimited $func<> in paragraph $Paragraph (go_ahead): '$snippet'.\n"
1773 # emit_C - output result of C<text>
1774 # $text is the depod-ed text
1777 my( $text, $nocode, $args ) = @_;
1778 $args = '' unless defined $args;
1780 my( $url, $fid ) = coderef( undef(), $text );
1782 # need HTML-safe text
1783 my $linktext = html_escape( "$text$args" );
1785 if( defined( $url ) &&
1786 (!defined( $EmittedItem ) || $EmittedItem ne $fid ) ){
1787 $res = "<a href=\"$url\"><code>$linktext</code></a>";
1788 } elsif( 0 && $nocode ){
1791 $res = "<code>$linktext</code>";
1797 # html_escape: make text safe for HTML
1801 $rest =~ s/&/&/g;
1802 $rest =~ s/</</g;
1803 $rest =~ s/>/>/g;
1804 $rest =~ s/"/"/g;
1805 # ' is only in XHTML, not HTML4. Be conservative
1806 #$rest =~ s/'/'/g;
1812 # dosify - convert filenames to 8.3
1816 return lc($str) if $^O eq 'VMS'; # VMS just needs casing
1819 $str =~ s/(\.\w+)/substr ($1,0,4)/ge;
1820 $str =~ s/(\w+)/substr ($1,0,8)/ge;
1826 # page_sect - make a URL from the text of a L<>
1829 my( $page, $section ) = @_;
1830 my( $linktext, $page83, $link); # work strings
1832 # check if we know that this is a section in this page
1833 if (!defined $Pages{$page} && defined $Sections{$page}) {
1836 ### print STDERR "reset page='', section=$section\n";
1839 $page83=dosify($page);
1840 $page=$page83 if (defined $Pages{$page83});
1842 $link = "#" . anchorify( $section );
1843 } elsif ( $page =~ /::/ ) {
1845 # Search page cache for an entry keyed under the html page name,
1846 # then look to see what directory that page might be in. NOTE:
1847 # this will only find one page. A better solution might be to produce
1848 # an intermediate page that is an index to all such pages.
1849 my $page_name = $page ;
1850 $page_name =~ s,^.*/,,s ;
1851 if ( defined( $Pages{ $page_name } ) &&
1852 $Pages{ $page_name } =~ /([^:]*$page)\.(?:pod|pm):/
1857 # NOTE: This branch assumes that all A::B pages are located in
1858 # $Htmlroot/A/B.html . This is often incorrect, since they are
1859 # often in $Htmlroot/lib/A/B.html or such like. Perhaps we could
1860 # analyze the contents of %Pages and figure out where any
1861 # cousins of A::B are, then assume that. So, if A::B isn't found,
1862 # but A::C is found in lib/A/C.pm, then A::B is assumed to be in
1863 # lib/A/B.pm. This is also limited, but it's an improvement.
1864 # Maybe a hints file so that the links point to the correct places
1868 $link = "$Htmlroot/$page.html";
1869 $link .= "#" . anchorify( $section ) if ($section);
1870 } elsif (!defined $Pages{$page}) {
1873 $section = anchorify( $section ) if $section ne "";
1874 ### print STDERR "...section=$section\n";
1876 # if there is a directory by the name of the page, then assume that an
1877 # appropriate section will exist in the subdirectory
1878 # if ($section ne "" && $Pages{$page} =~ /([^:]*[^(\.pod|\.pm)]):/) {
1879 if ($section ne "" && $Pages{$page} =~ /([^:]*(?<!\.pod)(?<!\.pm)):/) {
1880 $link = "$Htmlroot/$1/$section.html";
1881 ### print STDERR "...link=$link\n";
1883 # since there is no directory by the name of the page, the section will
1884 # have to exist within a .html of the same name. thus, make sure there
1885 # is a .pod or .pm that might become that .html
1887 $section = "#$section" if $section;
1888 ### print STDERR "...section=$section\n";
1890 # check if there is a .pod with the page name.
1891 # for L<Foo>, Foo.(pod|pm) is preferred to A/Foo.(pod|pm)
1892 if ($Pages{$page} =~ /([^:]*)\.(?:pod|pm):/) {
1893 $link = "$Htmlroot/$1.html$section";
1901 # Here, we take advantage of the knowledge that $Htmlfileurl ne ''
1902 # implies $Htmlroot eq ''. This means that the link in question
1903 # needs a prefix of $Htmldir if it begins with '/'. The test for
1904 # the initial '/' is done to avoid '#'-only links, and to allow
1905 # for other kinds of links, like file:, ftp:, etc.
1907 if ( $Htmlfileurl ne '' ) {
1908 $link = "$Htmldir$link" if $link =~ m{^/}s;
1909 $url = relativize_url( $link, $Htmlfileurl );
1910 # print( " b: [$link,$Htmlfileurl,$url]\n" );
1923 # relativize_url - convert an absolute URL to one relative to a base URL.
1924 # Assumes both end in a filename.
1926 sub relativize_url {
1927 my ($dest,$source) = @_ ;
1929 my ($dest_volume,$dest_directory,$dest_file) =
1930 File::Spec::Unix->splitpath( $dest ) ;
1931 $dest = File::Spec::Unix->catpath( $dest_volume, $dest_directory, '' ) ;
1933 my ($source_volume,$source_directory,$source_file) =
1934 File::Spec::Unix->splitpath( $source ) ;
1935 $source = File::Spec::Unix->catpath( $source_volume, $source_directory, '' ) ;
1938 if ( $dest ne '' ) {
1939 $rel_path = File::Spec::Unix->abs2rel( $dest, $source ) ;
1942 if ( $rel_path ne '' &&
1943 substr( $rel_path, -1 ) ne '/' &&
1944 substr( $dest_file, 0, 1 ) ne '#'
1946 $rel_path .= "/$dest_file" ;
1949 $rel_path .= "$dest_file" ;
1957 # coderef - make URL from the text of a C<>
1960 my( $page, $item ) = @_;
1963 my $fid = fragment_id( $item );
1965 if( defined( $page ) && $page ne "" ){
1966 # we have been given a $page...
1969 Carp::confess("Undefined fragment '$item' from fragment_id() in coderef() in $Podfile")
1971 # Do we take it? Item could be a section!
1972 my $base = $Items{$fid} || "";
1973 $base =~ s{[^/]*/}{};
1974 if( $base ne "$page.html" ){
1975 ### print STDERR "coderef( $page, $item ): items{$fid} = $Items{$fid} = $base => discard page!\n";
1980 # no page - local items precede cached items
1981 if( defined( $fid ) ){
1982 if( exists $Local_Items{$fid} ){
1983 $page = $Local_Items{$fid};
1985 $page = $Items{$fid};
1990 # if there was a pod file that we found earlier with an appropriate
1991 # =item directive, then create a link to that page.
1992 if( defined $page ){
1994 if( exists $Pages{$page} and $Pages{$page} =~ /([^:.]*)\.[^:]*:/){
1995 $page = $1 . '.html';
1997 my $link = "$Htmlroot/$page#" . anchorify($fid);
1999 # Here, we take advantage of the knowledge that $Htmlfileurl
2000 # ne '' implies $Htmlroot eq ''.
2001 if ( $Htmlfileurl ne '' ) {
2002 $link = "$Htmldir$link" ;
2003 $url = relativize_url( $link, $Htmlfileurl ) ;
2008 $url = "#" . anchorify($fid);
2011 confess "url has space: $url" if $url =~ /"[^"]*\s[^"]*"/;
2013 return( $url, $fid );
2019 # Adapted from Nick Ing-Simmons' PodToHtml package.
2021 my $source_file = shift ;
2022 my $destination_file = shift;
2024 my $source = URI::file->new_abs($source_file);
2025 my $uo = URI::file->new($destination_file,$source)->abs;
2026 return $uo->rel->as_string;
2031 # finish_list - finish off any pending HTML lists. this should be called
2032 # after the entire pod file has been read and converted.
2036 warn "$0: $Podfile: unterminated list(s) at =head in paragraph $Paragraph. ignoring.\n" unless $Quiet;
2037 while( $Listlevel ){
2044 # htmlify - converts a pod section specification to a suitable section
2045 # specification for HTML. Note that we keep spaces and special characters
2046 # except ", ? (Netscape problem) and the hyphen (writer's problem...).
2050 $heading =~ s/(\s+)/ /g;
2051 $heading =~ s/\s+\Z//;
2052 $heading =~ s/\A\s+//;
2053 # The hyphen is a disgrace to the English language.
2054 # $heading =~ s/[-"?]//g;
2055 $heading =~ s/["?]//g;
2056 $heading = lc( $heading );
2061 # similar to htmlify, but turns non-alphanumerics into underscores
2065 $anchor = htmlify($anchor);
2066 $anchor =~ s/\W/_/g;
2071 # depod - convert text by eliminating all interior sequences
2072 # Note: can be called with copy or modify semantics
2079 $E2c{amp} = '&'; # in Tk's pods
2087 ${$_[0]} = depod1( \$string );
2095 my( $rstr, $func, $closing ) = @_;
2097 return $res unless defined $$rstr;
2098 if( ! defined( $func ) ){
2099 # skip to next begin of an interior sequence
2100 while( $$rstr =~ s/\A(.*?)([BCEFILSXZ])<(<+[^\S\n]+)?//s ){
2101 # recurse into its text
2102 $res .= $1 . depod1( $rstr, $2, closing $3);
2105 } elsif( $func eq 'E' ){
2106 # E<x> - convert to character
2107 $$rstr =~ s/^([^>]*)>//;
2108 $res .= $E2c{$1} || "";
2109 } elsif( $func eq 'X' ){
2111 $$rstr =~ s/^[^>]*>//;
2112 } elsif( $func eq 'Z' ){
2116 # all others: either recurse into new function or
2117 # terminate at closing angle bracket
2118 my $term = pattern $closing;
2119 while( $$rstr =~ s/\A(.*?)(([BCEFILSXZ])<(<+[^\S\n]+)?|$term)//s ){
2122 $res .= depod1( $rstr, $3, closing $4 );
2124 ## If we're here and $2 ne '>': undelimited interior sequence.
2125 ## Ignored, as this is called without proper indication of where we are.
2126 ## Rely on process_text to produce diagnostics.
2132 my %seen; # static fragment record hash
2134 sub fragment_id_readable {
2136 my $generate = shift; # optional flag
2140 # leave the words for the fragment identifier,
2141 # change everything else to underbars.
2142 $text =~ s/[^A-Za-z0-9_]+/_/g; # do not use \W to avoid locale dependency.
2143 $text =~ s/_{2,}/_/g;
2149 # Nothing left after removing punctuation, so leave it as is
2150 # E.g. if option is named: "=item -#"
2156 if ( exists $seen{$text} ) {
2157 # This already exists, make it unique
2159 $text = $text . $seen{$text};
2161 $seen{$text} = 1; # first time seen this fragment
2169 sub fragment_id_obfuscated { # This was the old "_2d_2d__"
2171 my $generate = shift; # optional flag
2173 # text? Normalize by obfuscating the fragment id to make it unique
2174 $text =~ s/\s+/_/sg;
2177 defined( $HC[ord($1)] ) ? $HC[ord($1)]
2178 : ( $HC[ord($1)] = sprintf( "%%%02X", ord($1) ) ) }gxe;
2179 $text = substr( $text, 0, 50 );
2185 # fragment_id - construct a fragment identifier from:
2187 # b) contents of C<...>
2192 my $generate = shift; # optional flag
2194 $text =~ s/\s+\Z//s;
2196 # a method or function?
2197 return $1 if $text =~ /(\w+)\s*\(/;
2198 return $1 if $text =~ /->\s*(\w+)\s*\(?/;
2201 return $1 if $text =~ /^([\$\@%*]\S+)/;
2203 # some pattern matching operator?
2204 return $1 if $text =~ m|^(\w+/).*/\w*$|;
2206 # fancy stuff... like "do { }"
2207 return $1 if $text =~ m|^(\w+)\s*{.*}$|;
2209 # honour the perlfunc manpage: func [PAR[,[ ]PAR]...]
2210 # and some funnies with ... Module ...
2211 return $1 if $text =~ m{^([a-z\d_]+)(\s+[A-Z,/& ][A-Z\d,/& ]*)?$};
2212 return $1 if $text =~ m{^([a-z\d]+)\s+Module(\s+[A-Z\d,/& ]+)?$};
2214 return fragment_id_readable($text, $generate);
2221 # make_URL_href - generate HTML href from URL
2222 # Special treatment for CGI queries.
2224 sub make_URL_href($){
2227 s{^(http:[-\w/#~:.+=&%@!]+)(\?.*)$}{<a href="$1$2">$1</a>}i ){
2228 $url = "<a href=\"$url\">$url</a>";