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.
236 my($Dircache, $Itemcache);
239 my($Htmlroot, $Htmldir, $Htmlfile, $Htmlfileurl);
240 my($Podfile, @Podpath, $Podroot);
250 my($Listlevel, @Listend);
252 use vars qw($Ignore); # need to localize it later.
254 my(%Items_Named, @Items_Seen);
263 my %Pages = (); # associative array used to find the location
264 # of pages referenced by L<> links.
265 my %Items = (); # associative array used to find the location
266 # of =item directives referenced by C<> links
271 my $Curdir = File::Spec->curdir;
276 $Cachedir = "."; # The directory to which item and directory
277 # caches will be written.
279 $Dircache = "pod2htmd.tmp";
280 $Itemcache = "pod2htmi.tmp";
282 @Begin_Stack = (); # begin/end stack
284 @Libpods = (); # files to search for links from C<> directives
285 $Htmlroot = "/"; # http-server base directory from which all
286 # relative paths in $podpath stem.
287 $Htmldir = ""; # The directory to which the html pages
288 # will (eventually) be written.
289 $Htmlfile = ""; # write to stdout by default
290 $Htmlfileurl = "" ; # The url that other files would use to
291 # refer to this file. This is only used
292 # to make relative urls that point to
295 $Podfile = ""; # read from stdin by default
296 @Podpath = (); # list of directories containing library pods.
297 $Podroot = $Curdir; # filesystem base directory from which all
298 # relative paths in $podpath stem.
299 $Css = ''; # Cascading style sheet
300 $Recurse = 1; # recurse on subdirectories in $podpath.
301 $Quiet = 0; # not quiet by default
302 $Verbose = 0; # not verbose by default
303 $Doindex = 1; # non-zero if we should generate an index
304 $Backlink = ''; # text for "back to top" links
305 $Listlevel = 0; # current list depth
306 @Listend = (); # the text to use to end the list.
307 $After_Lpar = 0; # set to true after a par in an =item
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
524 warn "Converting input file $Podfile\n" if $Verbose;
525 foreach my $i (0..$#poddata){
528 if (/^(=.*)/s) { # is it a pod directive?
533 if (/^=begin\s+(\S+)\s*(.*)/si) {# =begin
534 process_begin($1, $2);
535 } elsif (/^=end\s+(\S+)\s*(.*)/si) {# =end
537 } elsif (/^=cut/) { # =cut
539 } elsif (/^=pod/) { # =pod
542 next if @Begin_Stack && $Begin_Stack[-1] ne 'html';
544 if (/^=(head[1-6])\s+(.*\S)/s) { # =head[1-6] heading
545 process_head( $1, $2, $Doindex && $index );
546 } elsif (/^=item\s*(.*\S)?/sm) { # =item text
547 $need_dd = process_item( $1 );
549 } elsif (/^=over\s*(.*)/) { # =over N
551 } elsif (/^=back/) { # =back
552 process_back($need_dd);
553 } elsif (/^=for\s+(\S+)\s*(.*)/si) {# =for
557 warn "$0: $Podfile: unknown pod directive '$1' in "
558 . "paragraph $Paragraph. ignoring.\n" unless $Quiet;
565 next if @Begin_Stack && $Begin_Stack[-1] ne 'html';
566 print HTML and next if @Begin_Stack && $Begin_Stack[-1] eq 'html';
567 print HTML "<dd>\n" if $need_dd;
569 if( $text =~ /\A\s+/ ){
570 process_pre( \$text );
571 print HTML "<pre>\n$text</pre>\n";
574 process_text( \$text );
576 # experimental: check for a paragraph where all lines
577 # have some ...\t...\t...\n pattern
579 my @lines = split( "\n", $text );
582 foreach my $line ( @lines ){
583 if( $line =~ /\S/ && $line !~ /\t/ ){
589 $text =~ s/\t+/<td>/g;
590 $text =~ s/^/<tr><td>/gm;
591 $text = '<table cellspacing="0" cellpadding="0">' .
596 ## end of experimental
601 print HTML "<p>$text</p>\n";
603 print HTML "</dd>\n" if $need_dd;
608 # finish off any pending directives
612 print HTML "<p><a href=\"#__index__\"><small>$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 $$sections{$name} = 1;
1032 $title = process_text( \$otitle );
1034 while ($which_head != $listdepth) {
1035 if ($which_head > $listdepth) {
1036 $index .= "\n" . ("\t" x $listdepth) . "<ul>\n";
1038 } elsif ($which_head < $listdepth) {
1040 $index .= "\n" . ("\t" x $listdepth) . "</ul>\n";
1044 $index .= "\n" . ("\t" x $listdepth) . "<li>" .
1045 "<a href=\"#" . $name . "\">" .
1046 $title . "</a></li>";
1050 # finish off the lists
1051 while ($listdepth--) {
1052 $index .= "\n" . ("\t" x $listdepth) . "</ul>\n";
1055 # get rid of bogus lists
1056 $index =~ s,\t*<ul>\s*</ul>\n,,g;
1062 # scan_items - scans the pod specified by $pod for =item directives. we
1063 # will use this information later on in resolving C<> links.
1066 my( $itemref, $pod, @poddata ) = @_;
1070 $pod =~ s/\.pod\z//;
1071 $pod .= ".html" if $pod;
1073 foreach $i (0..$#poddata) {
1074 my $txt = depod( $poddata[$i] );
1076 # figure out what kind of item it is.
1077 # Build string for referencing this item.
1078 if ( $txt =~ /\A=item\s+\*\s*(.*)\Z/s ) { # bullet
1081 } elsif( $txt =~ /\A=item\s+(?>\d+\.?)\s*(.*)\Z/s ) { # numbered list
1083 } elsif( $txt =~ /\A=item\s+(.*)\Z/s ) { # plain item
1088 my $fid = fragment_id( $item );
1089 $$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])/;
1104 warn "$0: $Podfile: unterminated list at =head in paragraph $Paragraph. ignoring.\n" unless $Quiet;
1105 while( $Listlevel ){
1111 if( $level == 1 && ! $Top ){
1112 print HTML "<a href=\"#__index__\"><small>$Backlink</small></a>\n"
1113 if $hasindex and $Backlink;
1114 print HTML "</p>\n<hr />\n"
1116 print HTML "</p>\n";
1119 my $name = anchorify( depod( $heading ) );
1120 my $convert = process_text( \$heading );
1121 print HTML "<h$level><a name=\"$name\">$convert</a></h$level>\n";
1126 # emit_item_tag - print an =item's text
1127 # Note: The global $EmittedItem is used for inhibiting self-references.
1131 sub emit_item_tag($$$){
1132 my( $otext, $text, $compact ) = @_;
1133 my $item = fragment_id( $text );
1135 $EmittedItem = $item;
1136 ### print STDERR "emit_item_tag=$item ($text)\n";
1138 print HTML '<strong>';
1139 if ($Items_Named{$item}++) {
1140 print HTML process_text( \$otext );
1142 my $name = 'item_' . $item;
1143 $name = anchorify($name);
1144 print HTML qq{<a name="$name">}, process_text( \$otext ), '</a>';
1146 print HTML "</strong>\n";
1147 undef( $EmittedItem );
1152 if( $Items_Seen[$Listlevel]++ == 0 ){
1153 push( @Listend, "</$tag>" );
1154 print HTML "<$tag>\n";
1156 my $emitted = $tag eq 'dl' ? 'dt' : 'li';
1157 print HTML "<$emitted>";
1162 # process_item - convert a pod item tag and convert it to HTML format.
1166 my $need_dd = 0; # set to 1 if we need a <dd></dd> after an item
1168 # lots of documents start a list without doing an =over. this is
1169 # bad! but, the proper thing to do seems to be to just assume
1170 # they did do an =over. so warn them once and then continue.
1171 if( $Listlevel == 0 ){
1172 warn "$0: $Podfile: unexpected =item directive in paragraph $Paragraph. ignoring.\n" unless $Quiet;
1176 # formatting: insert a paragraph if preceding item has >1 paragraph
1178 print HTML $need_dd ? "</dd>\n" : "</li>\n" if $After_Lpar;
1182 # remove formatting instructions from the text
1183 my $text = depod( $otext );
1185 my $emitted; # the tag actually emitted, used for closing
1187 # all the list variants:
1188 if( $text =~ /\A\*/ ){ # bullet
1189 $emitted = emit_li( 'ul' );
1190 if ($text =~ /\A\*\s+(.+)\Z/s ) { # with additional text
1192 $otext =~ s/\A\*\s+//;
1193 emit_item_tag( $otext, $tag, 1 );
1196 } elsif( $text =~ /\A\d+/ ){ # numbered list
1197 $emitted = emit_li( 'ol' );
1198 if ($text =~ /\A(?>\d+\.?)\s*(.+)\Z/s ) { # with additional text
1200 $otext =~ s/\A\d+\.?\s*//;
1201 emit_item_tag( $otext, $tag, 1 );
1204 } else { # definition list
1205 $emitted = emit_li( 'dl' );
1206 if ($text =~ /\A(.+)\Z/s ){ # should have text
1207 emit_item_tag( $otext, $text, 1 );
1216 # process_over - process a pod over tag and start a corresponding HTML list.
1221 push( @Items_Seen, 0 );
1226 # process_back - process a pod back tag and convert it to HTML format.
1229 my $need_dd = shift;
1230 if( $Listlevel == 0 ){
1231 warn "$0: $Podfile: unexpected =back directive in paragraph $Paragraph. ignoring.\n" unless $Quiet;
1235 # close off the list. note, I check to see if $Listend[$Listlevel] is
1236 # defined because an =item directive may have never appeared and thus
1237 # $Listend[$Listlevel] may have never been initialized.
1239 if( defined $Listend[$Listlevel] ){
1240 print HTML $need_dd ? "</dd>\n" : "</li>\n" if $After_Lpar;
1241 print HTML $Listend[$Listlevel];
1247 # clean up item count
1252 # process_cut - process a pod cut tag, thus start ignoring pod directives.
1259 # process_pod - process a pod tag, thus stop ignoring pod directives
1260 # until we see a corresponding cut.
1263 # no need to set $Ignore to 0 cause the main loop did it
1267 # process_for - process a =for pod tag. if it's for html, spit
1268 # it out verbatim, if illustration, center it, otherwise ignore it.
1271 my($whom, $text) = @_;
1272 if ( $whom =~ /^(pod2)?html$/i) {
1274 } elsif ($whom =~ /^illustration$/i) {
1275 1 while chomp $text;
1276 for my $ext (qw[.png .gif .jpeg .jpg .tga .pcl .bmp]) {
1277 $text .= $ext, last if -r "$text$ext";
1279 print HTML qq{<p align="center"><img src="$text" alt="$text illustration" /></p>};
1284 # process_begin - process a =begin pod tag. this pushes
1285 # whom we're beginning on the begin stack. if there's a
1286 # begin stack, we only print if it us.
1289 my($whom, $text) = @_;
1291 push (@Begin_Stack, $whom);
1292 if ( $whom =~ /^(pod2)?html$/) {
1293 print HTML $text if $text;
1298 # process_end - process a =end pod tag. pop the
1299 # begin stack. die if we're mismatched.
1302 my($whom, $text) = @_;
1304 if ($Begin_Stack[-1] ne $whom ) {
1305 die "Unmatched begin/end at chunk $Paragraph\n"
1307 pop( @Begin_Stack );
1311 # process_pre - indented paragraph, made into <pre></pre>
1320 # insert spaces in place of tabs
1323 1 while $line =~ s/(\t+)/' ' x ((length($1) * 8) - $-[0] % 8)/e;
1327 # convert some special chars to HTML escapes
1328 $rest = html_escape($rest);
1330 # try and create links for all occurrences of perl.* within
1331 # the preformatted text.
1335 if ( defined $Pages{$2} ){ # is a link
1336 qq($1<a href="$Htmlroot/$Pages{$2}">$2</a>);
1337 } elsif (defined $Pages{dosify($2)}) { # is a link
1338 qq($1<a href="$Htmlroot/$Pages{dosify($2)}">$2</a>);
1344 (<a\ href="?) ([^>:]*:)? ([^>:]*) \.pod: ([^>:]*:)?
1347 if ( $Htmlfileurl ne '' ){
1348 # Here, we take advantage of the knowledge
1349 # that $Htmlfileurl ne '' implies $Htmlroot eq ''.
1350 # Since $Htmlroot eq '', we need to prepend $Htmldir
1351 # on the fron of the link to get the absolute path
1352 # of the link's target. We check for a leading '/'
1353 # to avoid corrupting links that are #, file:, etc.
1355 $old_url = "$Htmldir$old_url" if $old_url =~ m{^\/};
1356 $url = relativize_url( "$old_url.html", $Htmlfileurl );
1363 # Look for embedded URLs and make them into links. We don't
1364 # relativize them since they are best left as the author intended.
1366 my $urls = '(' . join ('|', qw{
1379 my $gunk = '/#~:.?+=&%@!\-';
1380 my $punc = '.:!?\-;';
1381 my $any = "${ltrs}${gunk}${punc}";
1384 \b # start at word boundary
1386 $urls : # need resource and a colon
1387 (?!:) # Ignore File::, among others.
1388 [$any] +? # followed by one or more of any valid
1389 # character, but be conservative and
1390 # take only what you need to....
1393 " > # maybe pre-quoted '<a href="...">'
1395 [$punc]* # 0 or more punctuation
1397 [^$any] # by a non-url char
1399 $ # end of the string
1402 $ # then end of the string
1404 }{<a href="$1">$1</a>}igox;
1406 # text should be as it is (verbatim)
1412 # pure text processing
1414 # pure_text/inIS_text: differ with respect to automatic C<> recognition.
1415 # we don't want this to happen within IS
1419 process_puretext( $text, 1 );
1424 process_puretext( $text, 0 );
1428 # process_puretext - process pure text (without pod-escapes) converting
1429 # double-quotes and handling implicit C<> links.
1431 sub process_puretext {
1432 my($text, $notinIS) = @_;
1434 ## Guessing at func() or [\$\@%&]*var references in plain text is destined
1435 ## to produce some strange looking ref's. uncomment to disable:
1438 my(@words, $lead, $trail);
1440 # keep track of leading and trailing white-space
1441 $lead = ($text =~ s/\A(\s+)//s ? $1 : "");
1442 $trail = ($text =~ s/(\s+)\Z//s ? $1 : "");
1444 # split at space/non-space boundaries
1445 @words = split( /(?<=\s)(?=\S)|(?<=\S)(?=\s)/, $text );
1447 # process each word individually
1448 foreach my $word (@words) {
1450 next if $word =~ /^\s*$/;
1451 # see if we can infer a link
1452 if( $notinIS && $word =~ /^(\w+)\((.*)\)$/ ) {
1453 # has parenthesis so should have been a C<> ref
1454 ## try for a pagename (perlXXX(1))?
1455 my( $func, $args ) = ( $1, $2 );
1456 if( $args =~ /^\d+$/ ){
1457 my $url = page_sect( $word, '' );
1459 $word = "<a href=\"$url\">the $word manpage</a>";
1463 ## try function name for a link, append tt'ed argument list
1464 $word = emit_C( $func, '', "($args)");
1466 #### disabled. either all (including $\W, $\w+{.*} etc.) or nothing.
1467 ## } elsif( $notinIS && $word =~ /^[\$\@%&*]+\w+$/) {
1468 ## # perl variables, should be a C<> ref
1469 ## $word = emit_C( $word );
1471 } elsif ($word =~ m,^\w+://\w,) {
1473 # Don't relativize it: leave it as the author intended
1474 $word = qq(<a href="$word">$word</a>);
1475 } elsif ($word =~ /[\w.-]+\@[\w-]+\.\w/) {
1476 # looks like an e-mail address
1477 my ($w1, $w2, $w3) = ("", $word, "");
1478 ($w1, $w2, $w3) = ("(", $1, ")$2") if $word =~ /^\((.*?)\)(,?)/;
1479 ($w1, $w2, $w3) = ("<", $1, ">$2") if $word =~ /^<(.*?)>(,?)/;
1480 $word = qq($w1<a href="mailto:$w2">$w2</a>$w3);
1482 $word = html_escape($word) if $word =~ /["&<>]/;
1486 # put everything back together
1487 return $lead . join( '', @words ) . $trail;
1492 # process_text - handles plaintext that appears in the input pod file.
1493 # there may be pod commands embedded within the text so those must be
1494 # converted to html commands.
1497 sub process_text1($$;$$);
1498 sub pattern ($) { $_[0] ? '[^\S\n]+'.('>' x ($_[0] + 1)) : '>' }
1499 sub closing ($) { local($_) = shift; (defined && s/\s+$//) ? length : 0 }
1504 my $res = process_text1( 0, $tref );
1509 sub process_text1($$;$$){
1510 my( $lev, $rstr, $func, $closing ) = @_;
1513 unless (defined $func) {
1519 # B<text> - boldface
1520 $res = '<strong>' . process_text1( $lev, $rstr ) . '</strong>';
1522 } elsif( $func eq 'C' ){
1523 # C<code> - can be a ref or <code></code>
1524 # need to extract text
1525 my $par = go_ahead( $rstr, 'C', $closing );
1527 ## clean-up of the link target
1528 my $text = depod( $par );
1530 ### my $x = $par =~ /[BI]</ ? 'yes' : 'no' ;
1531 ### print STDERR "-->call emit_C($par) lev=$lev, par with BI=$x\n";
1533 $res = emit_C( $text, $lev > 1 || ($par =~ /[BI]</) );
1535 } elsif( $func eq 'E' ){
1536 # E<x> - convert to character
1537 $$rstr =~ s/^([^>]*)>//;
1539 $escape =~ s/^(\d+|X[\dA-F]+)$/#$1/i;
1542 } elsif( $func eq 'F' ){
1543 # F<filename> - italicize
1544 $res = '<em class="file">' . process_text1( $lev, $rstr ) . '</em>';
1546 } elsif( $func eq 'I' ){
1547 # I<text> - italicize
1548 $res = '<em>' . process_text1( $lev, $rstr ) . '</em>';
1550 } elsif( $func eq 'L' ){
1552 ## L<text|cross-ref> => produce text, use cross-ref for linking
1553 ## L<cross-ref> => make text from cross-ref
1554 ## need to extract text
1555 my $par = go_ahead( $rstr, 'L', $closing );
1557 # some L<>'s that shouldn't be:
1558 # a) full-blown URL's are emitted as-is
1559 if( $par =~ m{^\w+://}s ){
1560 return make_URL_href( $par );
1562 # b) C<...> is stripped and treated as C<>
1563 if( $par =~ /^C<(.*)>$/ ){
1564 my $text = depod( $1 );
1565 return emit_C( $text, $lev > 1 || ($par =~ /[BI]</) );
1568 # analyze the contents
1569 $par =~ s/\n/ /g; # undo word-wrapped tags
1572 if( $par =~ s{^([^|]+)\|}{} ){
1576 # make sure sections start with a /
1579 my( $page, $section, $ident );
1581 # check for link patterns
1582 if( $par =~ m{^([^/]+?)/(?!")(.*?)$} ){ # name/ident
1583 # we've got a name/ident (no quotes)
1584 ( $page, $ident ) = ( $1, $2 );
1585 ### print STDERR "--> L<$par> to page $page, ident $ident\n";
1587 } elsif( $par =~ m{^(.*?)/"?(.*?)"?$} ){ # [name]/"section"
1588 # even though this should be a "section", we go for ident first
1589 ( $page, $ident ) = ( $1, $2 );
1590 ### print STDERR "--> L<$par> to page $page, section $section\n";
1592 } elsif( $par =~ /\s/ ){ # this must be a section with missing quotes
1593 ( $page, $section ) = ( '', $par );
1594 ### print STDERR "--> L<$par> to void page, section $section\n";
1597 ( $page, $section ) = ( $par, '' );
1598 ### print STDERR "--> L<$par> to page $par, void section\n";
1601 # now, either $section or $ident is defined. the convoluted logic
1602 # below tries to resolve L<> according to what the user specified.
1603 # failing this, we try to find the next best thing...
1604 my( $url, $ltext, $fid );
1607 if( defined $ident ){
1608 ## try to resolve $ident as an item
1609 ( $url, $fid ) = coderef( $page, $ident );
1611 if( ! defined( $linktext ) ){
1613 $linktext .= " in " if $ident && $page;
1614 $linktext .= "the $page manpage" if $page;
1616 ### print STDERR "got coderef url=$url\n";
1619 ## no luck: go for a section (auto-quoting!)
1622 ## now go for a section
1623 my $htmlsection = htmlify( $section );
1624 $url = page_sect( $page, $htmlsection );
1626 if( ! defined( $linktext ) ){
1627 $linktext = $section;
1628 $linktext .= " in " if $section && $page;
1629 $linktext .= "the $page manpage" if $page;
1631 ### print STDERR "got page/section url=$url\n";
1634 ## no luck: go for an ident
1641 ( $url, $fid ) = coderef( $page, $ident );
1643 if( ! defined( $linktext ) ){
1645 $linktext .= " in " if $ident && $page;
1646 $linktext .= "the $page manpage" if $page;
1648 ### print STDERR "got section=>coderef url=$url\n";
1652 # warning; show some text.
1653 $linktext = $opar unless defined $linktext;
1654 warn "$0: $Podfile: cannot resolve L<$opar> in paragraph $Paragraph.\n" unless $Quiet;
1657 # now we have a URL or just plain code
1658 $$rstr = $linktext . '>' . $$rstr;
1659 if( defined( $url ) ){
1660 $res = "<a href=\"$url\">" . process_text1( $lev, $rstr ) . '</a>';
1662 $res = '<em>' . process_text1( $lev, $rstr ) . '</em>';
1665 } elsif( $func eq 'S' ){
1666 # S<text> - non-breaking spaces
1667 $res = process_text1( $lev, $rstr );
1668 $res =~ s/ / /g;
1670 } elsif( $func eq 'X' ){
1672 $$rstr =~ s/^[^>]*>//;
1674 } elsif( $func eq 'Z' ){
1676 warn "$0: $Podfile: invalid X<> in paragraph $Paragraph.\n"
1677 unless $$rstr =~ s/^>// or $Quiet;
1680 my $term = pattern $closing;
1681 while( $$rstr =~ s/\A(.*?)(([BCEFILSXZ])<(<+[^\S\n]+)?|$term)//s ){
1682 # all others: either recurse into new function or
1683 # terminate at closing angle bracket(s)
1685 $pt .= $2 if !$3 && $lev == 1;
1686 $res .= $lev == 1 ? pure_text( $pt ) : inIS_text( $pt );
1687 return $res if !$3 && $lev > 1;
1689 $res .= process_text1( $lev, $rstr, $3, closing $4 );
1693 $res .= pure_text( $$rstr );
1695 warn "$0: $Podfile: undelimited $func<> in paragraph $Paragraph.\n" unless $Quiet;
1702 # go_ahead: extract text of an IS (can be nested)
1705 my( $rstr, $func, $closing ) = @_;
1707 my @closing = ($closing);
1709 s/\A(.*?)(([BCEFILSXZ])<(<+[^\S\n]+)?|@{[pattern $closing[0]]})//s ){
1713 return $res unless @closing;
1715 unshift @closing, closing $4;
1719 warn "$0: $Podfile: undelimited $func<> in paragraph $Paragraph.\n" unless $Quiet;
1724 # emit_C - output result of C<text>
1725 # $text is the depod-ed text
1728 my( $text, $nocode, $args ) = @_;
1729 $args = '' unless defined $args;
1731 my( $url, $fid ) = coderef( undef(), $text );
1733 # need HTML-safe text
1734 my $linktext = html_escape( "$text$args" );
1736 if( defined( $url ) &&
1737 (!defined( $EmittedItem ) || $EmittedItem ne $fid ) ){
1738 $res = "<a href=\"$url\"><code>$linktext</code></a>";
1739 } elsif( 0 && $nocode ){
1742 $res = "<code>$linktext</code>";
1748 # html_escape: make text safe for HTML
1752 $rest =~ s/&/&/g;
1753 $rest =~ s/</</g;
1754 $rest =~ s/>/>/g;
1755 $rest =~ s/"/"/g;
1756 # ' is only in XHTML, not HTML4. Be conservative
1757 #$rest =~ s/'/'/g;
1763 # dosify - convert filenames to 8.3
1767 return lc($str) if $^O eq 'VMS'; # VMS just needs casing
1770 $str =~ s/(\.\w+)/substr ($1,0,4)/ge;
1771 $str =~ s/(\w+)/substr ($1,0,8)/ge;
1777 # page_sect - make a URL from the text of a L<>
1780 my( $page, $section ) = @_;
1781 my( $linktext, $page83, $link); # work strings
1783 # check if we know that this is a section in this page
1784 if (!defined $Pages{$page} && defined $Sections{$page}) {
1787 ### print STDERR "reset page='', section=$section\n";
1790 $page83=dosify($page);
1791 $page=$page83 if (defined $Pages{$page83});
1793 $link = "#" . anchorify( $section );
1794 } elsif ( $page =~ /::/ ) {
1796 # Search page cache for an entry keyed under the html page name,
1797 # then look to see what directory that page might be in. NOTE:
1798 # this will only find one page. A better solution might be to produce
1799 # an intermediate page that is an index to all such pages.
1800 my $page_name = $page ;
1801 $page_name =~ s,^.*/,,s ;
1802 if ( defined( $Pages{ $page_name } ) &&
1803 $Pages{ $page_name } =~ /([^:]*$page)\.(?:pod|pm):/
1808 # NOTE: This branch assumes that all A::B pages are located in
1809 # $Htmlroot/A/B.html . This is often incorrect, since they are
1810 # often in $Htmlroot/lib/A/B.html or such like. Perhaps we could
1811 # analyze the contents of %Pages and figure out where any
1812 # cousins of A::B are, then assume that. So, if A::B isn't found,
1813 # but A::C is found in lib/A/C.pm, then A::B is assumed to be in
1814 # lib/A/B.pm. This is also limited, but it's an improvement.
1815 # Maybe a hints file so that the links point to the correct places
1819 $link = "$Htmlroot/$page.html";
1820 $link .= "#" . anchorify( $section ) if ($section);
1821 } elsif (!defined $Pages{$page}) {
1824 $section = anchorify( $section ) if $section ne "";
1825 ### print STDERR "...section=$section\n";
1827 # if there is a directory by the name of the page, then assume that an
1828 # appropriate section will exist in the subdirectory
1829 # if ($section ne "" && $Pages{$page} =~ /([^:]*[^(\.pod|\.pm)]):/) {
1830 if ($section ne "" && $Pages{$page} =~ /([^:]*(?<!\.pod)(?<!\.pm)):/) {
1831 $link = "$Htmlroot/$1/$section.html";
1832 ### print STDERR "...link=$link\n";
1834 # since there is no directory by the name of the page, the section will
1835 # have to exist within a .html of the same name. thus, make sure there
1836 # is a .pod or .pm that might become that .html
1838 $section = "#$section" if $section;
1839 ### print STDERR "...section=$section\n";
1841 # check if there is a .pod with the page name.
1842 # for L<Foo>, Foo.(pod|pm) is preferred to A/Foo.(pod|pm)
1843 if ($Pages{$page} =~ /([^:]*)\.(?:pod|pm):/) {
1844 $link = "$Htmlroot/$1.html$section";
1852 # Here, we take advantage of the knowledge that $Htmlfileurl ne ''
1853 # implies $Htmlroot eq ''. This means that the link in question
1854 # needs a prefix of $Htmldir if it begins with '/'. The test for
1855 # the initial '/' is done to avoid '#'-only links, and to allow
1856 # for other kinds of links, like file:, ftp:, etc.
1858 if ( $Htmlfileurl ne '' ) {
1859 $link = "$Htmldir$link" if $link =~ m{^/}s;
1860 $url = relativize_url( $link, $Htmlfileurl );
1861 # print( " b: [$link,$Htmlfileurl,$url]\n" );
1874 # relativize_url - convert an absolute URL to one relative to a base URL.
1875 # Assumes both end in a filename.
1877 sub relativize_url {
1878 my ($dest,$source) = @_ ;
1880 my ($dest_volume,$dest_directory,$dest_file) =
1881 File::Spec::Unix->splitpath( $dest ) ;
1882 $dest = File::Spec::Unix->catpath( $dest_volume, $dest_directory, '' ) ;
1884 my ($source_volume,$source_directory,$source_file) =
1885 File::Spec::Unix->splitpath( $source ) ;
1886 $source = File::Spec::Unix->catpath( $source_volume, $source_directory, '' ) ;
1889 if ( $dest ne '' ) {
1890 $rel_path = File::Spec::Unix->abs2rel( $dest, $source ) ;
1893 if ( $rel_path ne '' &&
1894 substr( $rel_path, -1 ) ne '/' &&
1895 substr( $dest_file, 0, 1 ) ne '#'
1897 $rel_path .= "/$dest_file" ;
1900 $rel_path .= "$dest_file" ;
1908 # coderef - make URL from the text of a C<>
1911 my( $page, $item ) = @_;
1914 my $fid = fragment_id( $item );
1915 if( defined( $page ) && $page ne "" ){
1916 # we have been given a $page...
1919 # Do we take it? Item could be a section!
1920 my $base = $Items{$fid} || "";
1921 $base =~ s{[^/]*/}{};
1922 if( $base ne "$page.html" ){
1923 ### print STDERR "coderef( $page, $item ): items{$fid} = $Items{$fid} = $base => discard page!\n";
1928 # no page - local items precede cached items
1929 if( defined( $fid ) ){
1930 if( exists $Local_Items{$fid} ){
1931 $page = $Local_Items{$fid};
1933 $page = $Items{$fid};
1938 # if there was a pod file that we found earlier with an appropriate
1939 # =item directive, then create a link to that page.
1940 if( defined $page ){
1942 if( exists $Pages{$page} and $Pages{$page} =~ /([^:.]*)\.[^:]*:/){
1943 $page = $1 . '.html';
1945 my $link = "$Htmlroot/$page#item_" . anchorify($fid);
1947 # Here, we take advantage of the knowledge that $Htmlfileurl
1948 # ne '' implies $Htmlroot eq ''.
1949 if ( $Htmlfileurl ne '' ) {
1950 $link = "$Htmldir$link" ;
1951 $url = relativize_url( $link, $Htmlfileurl ) ;
1956 $url = "#item_" . anchorify($fid);
1959 confess "url has space: $url" if $url =~ /"[^"]*\s[^"]*"/;
1961 return( $url, $fid );
1967 # Adapted from Nick Ing-Simmons' PodToHtml package.
1969 my $source_file = shift ;
1970 my $destination_file = shift;
1972 my $source = URI::file->new_abs($source_file);
1973 my $uo = URI::file->new($destination_file,$source)->abs;
1974 return $uo->rel->as_string;
1979 # finish_list - finish off any pending HTML lists. this should be called
1980 # after the entire pod file has been read and converted.
1983 while ($Listlevel > 0) {
1984 print HTML "</dl>\n";
1990 # htmlify - converts a pod section specification to a suitable section
1991 # specification for HTML. Note that we keep spaces and special characters
1992 # except ", ? (Netscape problem) and the hyphen (writer's problem...).
1996 $heading =~ s/(\s+)/ /g;
1997 $heading =~ s/\s+\Z//;
1998 $heading =~ s/\A\s+//;
1999 # The hyphen is a disgrace to the English language.
2000 $heading =~ s/[-"?]//g;
2001 $heading = lc( $heading );
2006 # similar to htmlify, but turns non-alphanumerics into underscores
2010 $anchor = htmlify($anchor);
2011 $anchor =~ s/\W/_/g;
2016 # depod - convert text by eliminating all interior sequences
2017 # Note: can be called with copy or modify semantics
2024 $E2c{amp} = '&'; # in Tk's pods
2032 ${$_[0]} = depod1( \$string );
2040 my( $rstr, $func, $closing ) = @_;
2042 return $res unless defined $$rstr;
2043 if( ! defined( $func ) ){
2044 # skip to next begin of an interior sequence
2045 while( $$rstr =~ s/\A(.*?)([BCEFILSXZ])<(<+[^\S\n]+)?//s ){
2046 # recurse into its text
2047 $res .= $1 . depod1( $rstr, $2, closing $3);
2050 } elsif( $func eq 'E' ){
2051 # E<x> - convert to character
2052 $$rstr =~ s/^([^>]*)>//;
2053 $res .= $E2c{$1} || "";
2054 } elsif( $func eq 'X' ){
2056 $$rstr =~ s/^[^>]*>//;
2057 } elsif( $func eq 'Z' ){
2061 # all others: either recurse into new function or
2062 # terminate at closing angle bracket
2063 my $term = pattern $closing;
2064 while( $$rstr =~ s/\A(.*?)(([BCEFILSXZ])<(<+[^\S\n]+)?|$term)//s ){
2067 $res .= depod1( $rstr, $3, closing $4 );
2069 ## If we're here and $2 ne '>': undelimited interior sequence.
2070 ## Ignored, as this is called without proper indication of where we are.
2071 ## Rely on process_text to produce diagnostics.
2077 # fragment_id - construct a fragment identifier from:
2079 # b) contents of C<...>
2084 $text =~ s/\s+\Z//s;
2086 # a method or function?
2087 return $1 if $text =~ /(\w+)\s*\(/;
2088 return $1 if $text =~ /->\s*(\w+)\s*\(?/;
2091 return $1 if $text =~ /^([\$\@%*]\S+)/;
2093 # some pattern matching operator?
2094 return $1 if $text =~ m|^(\w+/).*/\w*$|;
2096 # fancy stuff... like "do { }"
2097 return $1 if $text =~ m|^(\w+)\s*{.*}$|;
2099 # honour the perlfunc manpage: func [PAR[,[ ]PAR]...]
2100 # and some funnies with ... Module ...
2101 return $1 if $text =~ m{^([a-z\d_]+)(\s+[A-Z,/& ][A-Z\d,/& ]*)?$};
2102 return $1 if $text =~ m{^([a-z\d]+)\s+Module(\s+[A-Z\d,/& ]+)?$};
2105 $text =~ s/\s+/_/sg;
2107 defined( $HC[ord($1)] ) ? $HC[ord($1)]
2108 : ( $HC[ord($1)] = sprintf( "%%%02X", ord($1) ) ) }gxe;
2109 $text = substr( $text, 0, 50 );
2116 # make_URL_href - generate HTML href from URL
2117 # Special treatment for CGI queries.
2119 sub make_URL_href($){
2122 s{^(http:[-\w/#~:.+=&%@!]+)(\?.*)$}{<a href="$1$2">$1</a>}i ){
2123 $url = "<a href=\"$url\">$url</a>";