5 use vars qw($VERSION @ISA @EXPORT @EXPORT_OK);
8 @EXPORT = qw(pod2html htmlify);
9 @EXPORT_OK = qw(anchorify);
18 use locale; # make \w work right in non-ASCII lands
22 Pod::Html - module to convert pod files to HTML
31 Converts files from pod format (see L<perlpod>) to HTML format. It
32 can automatically generate indexes and cross-references, and it keeps
33 a cache of things it knows how to cross-reference.
37 Pod::Html takes the following arguments:
43 --backlink="Back to Top"
45 Adds "Back to Top" links in front of every C<head1> heading (except for
46 the first). By default, no backlinks are generated.
52 Creates the item and directory caches in the given directory.
58 Specify the URL of a cascading style sheet. Also disables all HTML/CSS
59 C<style> attributes that are output by default (to avoid conflicts).
65 Flushes the item and directory caches.
72 Creates header and footer blocks containing the text of the C<NAME>
73 section. By default, no headers are generated.
79 Displays the usage message.
86 Include hidden directories in the search for POD's in podpath if recurse
88 The default is not to traverse any directory whose name begins with C<.>.
89 See L</"podpath"> and L</"recurse">.
91 [This option is for backward compatibility only.
92 It's hard to imagine that one would usefully create a module with a
93 name component beginning with C<.>.]
99 Sets the directory in which the resulting HTML file is placed. This
100 is used to generate relative links to other files. Not passing this
101 causes all links to be absolute, since this is the value that tells
102 Pod::Html the root of the documentation tree.
108 Sets the base URL for the HTML files. When cross-references are made,
109 the HTML root is prepended to the URL.
116 Generate an index at the top of the HTML file. This is the default
123 Specify the pod file to convert. Input is taken from STDIN if no
128 --libpods=name:...:name
130 List of page names (eg, "perlfunc") which contain linkable C<=item>s.
137 B<Deprecated>, has no effect. For backwards compatibility only.
143 Specify the HTML file to create. Output goes to STDOUT if no outfile
148 --podpath=name:...:name
150 Specify which subdirectories of the podroot contain pod files whose
151 HTML converted forms can be linked to in cross references.
157 Specify the base directory for finding library pods.
164 Don't display I<mostly harmless> warning messages. These messages
165 will be displayed by default. But this is not the same as C<verbose>
173 Recurse into subdirectories specified in podpath (default behaviour).
179 Specify the title of the resulting HTML file.
186 Display progress messages. By default, they won't be displayed.
193 "--podpath=lib:ext:pod:vms",
194 "--podroot=/usr/src/perl",
195 "--htmlroot=/perl/nmanual",
196 "--libpods=perlfunc:perlguts:perlvar:perlrun:perlop",
199 "--outfile=/perl/nmanual/foo.html");
203 Uses C<$Config{pod2html}> to setup default options.
207 Tom Christiansen, E<lt>tchrist@perl.comE<gt>.
215 This program is distributed under the Artistic License.
221 my($Dircache, $Itemcache);
224 my($Htmlroot, $Htmldir, $Htmlfile, $Htmlfileurl);
225 my($Podfile, @Podpath, $Podroot);
235 my($Listlevel, @Listend);
237 use vars qw($Ignore); # need to localize it later.
239 my(%Items_Named, @Items_Seen);
248 my %Pages = (); # associative array used to find the location
249 # of pages referenced by L<> links.
250 my %Items = (); # associative array used to find the location
251 # of =item directives referenced by C<> links
257 my $Curdir = File::Spec->curdir;
262 $Cachedir = "."; # The directory to which item and directory
263 # caches will be written.
265 $Dircache = "pod2htmd.tmp";
266 $Itemcache = "pod2htmi.tmp";
268 @Begin_Stack = (); # begin/end stack
270 @Libpods = (); # files to search for links from C<> directives
271 $Htmlroot = "/"; # http-server base directory from which all
272 # relative paths in $podpath stem.
273 $Htmldir = ""; # The directory to which the html pages
274 # will (eventually) be written.
275 $Htmlfile = ""; # write to stdout by default
276 $Htmlfileurl = "" ; # The url that other files would use to
277 # refer to this file. This is only used
278 # to make relative urls that point to
281 $Podfile = ""; # read from stdin by default
282 @Podpath = (); # list of directories containing library pods.
283 $Podroot = $Curdir; # filesystem base directory from which all
284 # relative paths in $podpath stem.
285 $Css = ''; # Cascading style sheet
286 $Recurse = 1; # recurse on subdirectories in $podpath.
287 $Quiet = 0; # not quiet by default
288 $Verbose = 0; # not verbose by default
289 $Doindex = 1; # non-zero if we should generate an index
290 $Backlink = ''; # text for "back to top" links
291 $Listlevel = 0; # current list depth
292 @Listend = (); # the text to use to end the list.
293 $After_Lpar = 0; # set to true after a par in an =item
294 $Ignore = 1; # whether or not to format text. we don't
295 # format text until we hit our first pod
298 @Items_Seen = (); # for multiples of the same item in perlfunc
300 $Header = 0; # produce block header/footer
301 $Title = ''; # title to give the pod(s)
302 $Top = 1; # true if we are at the top of the doc. used
303 # to prevent the first <hr /> directive.
304 $Paragraph = ''; # which paragraph we're processing (used
305 # for error messages)
306 $PTQuote = 0; # status of double-quote conversion
307 %Sections = (); # sections within this page
310 $Is83 = $^O eq 'dos'; # Is it an 8.3 filesystem?
314 # clean_data: global clean-up of pod data
318 for my $i ( 0..$#{$dataref} ) {
319 ${$dataref}[$i] =~ s/\s+\Z//;
321 # have a look for all-space lines
322 if( ${$dataref}[$i] =~ /^\s+$/m and $dataref->[$i] !~ /^\s/ ){
323 my @chunks = split( /^\s+$/m, ${$dataref}[$i] );
324 splice( @$dataref, $i, 1, @chunks );
337 $Is83 = 0 if (defined (&Dos::UseLFN) && Dos::UseLFN());
339 # cache of %Pages and %Items from last time we ran pod2html
341 #undef $opt_help if defined $opt_help;
343 # parse the command-line parameters
344 parse_command_line();
346 # escape the backlink argument (same goes for title but is done later...)
347 $Backlink = html_escape($Backlink) if defined $Backlink;
349 # set some variables to their default values if necessary
351 unless (@ARGV && $ARGV[0]) {
352 $Podfile = "-" unless $Podfile; # stdin
353 open(POD, "<$Podfile")
354 || die "$0: cannot open $Podfile file for input: $!\n";
356 $Podfile = $ARGV[0]; # XXX: might be more filenames
359 $Htmlfile = "-" unless $Htmlfile; # stdout
360 $Htmlroot = "" if $Htmlroot eq "/"; # so we don't get a //
361 $Htmldir =~ s#/\z## ; # so we don't get a //
363 && defined( $Htmldir )
365 && substr( $Htmlfile, 0, length( $Htmldir ) ) eq $Htmldir
368 # Set the 'base' url for this file, so that we can use it
369 # as the location from which to calculate relative links
370 # to other files. If this is '', then absolute links will
371 # be used throughout.
372 $Htmlfileurl= "$Htmldir/" . substr( $Htmlfile, length( $Htmldir ) + 1);
375 # read the pod a paragraph at a time
376 warn "Scanning for sections in input file(s)\n" if $Verbose;
385 @poddata = map { s/\r\n/\n/g;
387 map { "$_\n\n" } split /\n\n/ :
390 @poddata = map { s/\r/\n/g;
392 map { "$_\n\n" } split /\n\n/ :
399 clean_data( \@poddata );
401 # scan the pod for =head[1-6] directives and build an index
402 my $index = scan_headings(\%Sections, @poddata);
405 warn "No headings in $Podfile\n" if $Verbose;
408 # open the output file
409 open(HTML, ">$Htmlfile")
410 || die "$0: cannot open $Htmlfile file for output: $!\n";
412 # put a title in the HTML file if one wasn't specified
415 for (my $i = 0; $i < @poddata; $i++) {
416 if ($poddata[$i] =~ /^=head1\s*NAME\b/m) {
417 for my $para ( @poddata[$i, $i+1] ) {
419 if ($Title) = $para =~ /(\S+\s+-+.*\S)/s;
426 if (!$Title and $Podfile =~ /\.pod\z/) {
427 # probably a split pod so take first =head[12] as title
428 for (my $i = 0; $i < @poddata; $i++) {
429 last if ($Title) = $poddata[$i] =~ /^=head[12]\s*(.*)/;
431 warn "adopted '$Title' as title for $Podfile\n"
432 if $Verbose and $Title;
435 $Title =~ s/\s*\(.*\)//;
437 warn "$0: no title for $Podfile.\n" unless $Quiet;
438 $Podfile =~ /^(.*)(\.[^.\/]+)?\z/s;
439 $Title = ($Podfile eq "-" ? 'No Title' : $1);
440 warn "using $Title" if $Verbose;
442 $Title = html_escape($Title);
445 my $bodystyle = ' style="background-color: white"';
446 my $tdstyle = ' style="background-color: #cccccc"';
449 $csslink = qq(\n<link rel="stylesheet" href="$Css" type="text/css" />);
450 $csslink =~ s,\\,/,g;
451 $csslink =~ s,(/.):,$1|,;
456 my $block = $Header ? <<END_OF_BLOCK : '';
457 <table border="0" width="100%" cellspacing="0" cellpadding="3">
458 <tr><td class="block"$tdstyle valign="middle">
459 <big><strong><span class="block"> $Title</span></strong></big>
464 print HTML <<END_OF_HEAD;
465 <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
466 <html xmlns="http://www.w3.org/1999/xhtml">
468 <title>$Title</title>$csslink
469 <link rev="made" href="mailto:$Config{perladmin}" />
476 # load/reload/validate/cache %Pages and %Items
477 get_cache($Dircache, $Itemcache, \@Podpath, $Podroot, $Recurse);
479 # scan the pod for =item directives
480 scan_items( \%Local_Items, "", @poddata);
482 # put an index at the top of the file. note, if $Doindex is 0 we
483 # still generate an index, but surround it with an html comment.
484 # that way some other program can extract it if desired.
486 print HTML "<p><a name=\"__index__\"></a></p>\n";
487 print HTML "<!-- INDEX BEGIN -->\n";
488 print HTML "<!--\n" unless $Doindex;
490 print HTML "-->\n" unless $Doindex;
491 print HTML "<!-- INDEX END -->\n\n";
492 print HTML "<hr />\n" if $Doindex and $index;
494 # now convert this file
495 my $after_item; # set to true after an =item
497 warn "Converting input file $Podfile\n" if $Verbose;
498 foreach my $i (0..$#poddata){
499 $PTQuote = 0; # status of quote conversion
503 if (/^(=.*)/s) { # is it a pod directive?
508 if (/^=begin\s+(\S+)\s*(.*)/si) {# =begin
509 process_begin($1, $2);
510 } elsif (/^=end\s+(\S+)\s*(.*)/si) {# =end
512 } elsif (/^=cut/) { # =cut
514 } elsif (/^=pod/) { # =pod
517 next if @Begin_Stack && $Begin_Stack[-1] ne 'html';
519 if (/^=(head[1-6])\s+(.*\S)/s) { # =head[1-6] heading
520 process_head( $1, $2, $Doindex && $index );
521 } elsif (/^=item\s*(.*\S)?/sm) { # =item text
522 $need_dd = process_item( $1 );
524 } elsif (/^=over\s*(.*)/) { # =over N
526 } elsif (/^=back/) { # =back
527 process_back($need_dd);
528 } elsif (/^=for\s+(\S+)\s*(.*)/si) {# =for
532 warn "$0: $Podfile: unknown pod directive '$1' in "
533 . "paragraph $Paragraph. ignoring.\n" unless $Quiet;
540 next if @Begin_Stack && $Begin_Stack[-1] ne 'html';
541 print HTML and next if @Begin_Stack && $Begin_Stack[-1] eq 'html';
542 print HTML "<dd>\n" if $need_dd;
544 if( $text =~ /\A\s+/ ){
545 process_pre( \$text );
546 print HTML "<pre>\n$text</pre>\n";
549 process_text( \$text );
551 # experimental: check for a paragraph where all lines
552 # have some ...\t...\t...\n pattern
554 my @lines = split( "\n", $text );
557 foreach my $line ( @lines ){
558 if( $line =~ /\S/ && $line !~ /\t/ ){
564 $text =~ s/\t+/<td>/g;
565 $text =~ s/^/<tr><td>/gm;
566 $text = '<table cellspacing="0" cellpadding="0">' .
571 ## end of experimental
576 print HTML "<p>$text</p>\n";
578 print HTML "</dd>\n" if $need_dd;
583 # finish off any pending directives
587 print HTML "<p><a href=\"#__index__\"><small>$Backlink</small></a></p>\n"
588 if $Doindex and $index and $Backlink;
590 print HTML <<END_OF_TAIL;
597 # close the html file
600 warn "Finished\n" if $Verbose;
603 ##############################################################################
607 warn "$0: $podfile: @_\n" if @_;
609 Usage: $0 --help --htmlroot=<name> --infile=<name> --outfile=<name>
610 --podpath=<name>:...:<name> --podroot=<name>
611 --libpods=<name>:...:<name> --recurse --verbose --index
612 --netscape --norecurse --noindex --cachedir=<name>
614 --backlink - set text for "back to top" links (default: none).
615 --cachedir - directory for the item and directory cache files.
616 --css - stylesheet URL
617 --flush - flushes the item and directory caches.
618 --[no]header - produce block header/footer (default is no headers).
619 --help - prints this message.
620 --hiddendirs - search hidden directories in podpath
621 --htmldir - directory for resulting HTML files.
622 --htmlroot - http-server base directory from which all relative paths
623 in podpath stem (default is /).
624 --[no]index - generate an index at the top of the resulting html
626 --infile - filename for the pod to convert (input taken from stdin
628 --libpods - colon-separated list of pages to search for =item pod
629 directives in as targets of C<> and implicit links (empty
630 by default). note, these are not filenames, but rather
631 page names like those that appear in L<> links.
632 --outfile - filename for the resulting html file (output sent to
634 --podpath - colon-separated list of directories containing library
635 pods (empty by default).
636 --podroot - filesystem base directory from which all relative paths
637 in podpath stem (default is .).
638 --[no]quiet - suppress some benign warning messages (default is off).
639 --[no]recurse - recurse on those subdirectories listed in podpath
641 --title - title that will appear in resulting html file.
642 --[no]verbose - self-explanatory (off by default).
643 --[no]netscape - deprecated, has no effect. for backwards compatibility only.
649 sub parse_command_line {
650 my ($opt_backlink,$opt_cachedir,$opt_css,$opt_flush,$opt_header,$opt_help,
651 $opt_htmldir,$opt_htmlroot,$opt_index,$opt_infile,$opt_libpods,
652 $opt_netscape,$opt_outfile,$opt_podpath,$opt_podroot,$opt_quiet,
653 $opt_recurse,$opt_title,$opt_verbose,$opt_hiddendirs);
655 unshift @ARGV, split ' ', $Config{pod2html} if $Config{pod2html};
656 my $result = GetOptions(
657 'backlink=s' => \$opt_backlink,
658 'cachedir=s' => \$opt_cachedir,
659 'css=s' => \$opt_css,
660 'flush' => \$opt_flush,
661 'header!' => \$opt_header,
662 'help' => \$opt_help,
663 'hiddendirs!'=> \$opt_hiddendirs,
664 'htmldir=s' => \$opt_htmldir,
665 'htmlroot=s' => \$opt_htmlroot,
666 'index!' => \$opt_index,
667 'infile=s' => \$opt_infile,
668 'libpods=s' => \$opt_libpods,
669 'netscape!' => \$opt_netscape,
670 'outfile=s' => \$opt_outfile,
671 'podpath=s' => \$opt_podpath,
672 'podroot=s' => \$opt_podroot,
673 'quiet!' => \$opt_quiet,
674 'recurse!' => \$opt_recurse,
675 'title=s' => \$opt_title,
676 'verbose!' => \$opt_verbose,
678 usage("-", "invalid parameters") if not $result;
680 usage("-") if defined $opt_help; # see if the user asked for help
681 $opt_help = ""; # just to make -w shut-up.
683 @Podpath = split(":", $opt_podpath) if defined $opt_podpath;
684 @Libpods = split(":", $opt_libpods) if defined $opt_libpods;
686 $Backlink = $opt_backlink if defined $opt_backlink;
687 $Cachedir = $opt_cachedir if defined $opt_cachedir;
688 $Css = $opt_css if defined $opt_css;
689 $Header = $opt_header if defined $opt_header;
690 $Htmldir = $opt_htmldir if defined $opt_htmldir;
691 $Htmlroot = $opt_htmlroot if defined $opt_htmlroot;
692 $Doindex = $opt_index if defined $opt_index;
693 $Podfile = $opt_infile if defined $opt_infile;
694 $HiddenDirs = $opt_hiddendirs if defined $opt_hiddendirs;
695 $Htmlfile = $opt_outfile if defined $opt_outfile;
696 $Podroot = $opt_podroot if defined $opt_podroot;
697 $Quiet = $opt_quiet if defined $opt_quiet;
698 $Recurse = $opt_recurse if defined $opt_recurse;
699 $Title = $opt_title if defined $opt_title;
700 $Verbose = $opt_verbose if defined $opt_verbose;
702 warn "Flushing item and directory caches\n"
703 if $opt_verbose && defined $opt_flush;
704 $Dircache = "$Cachedir/pod2htmd.tmp";
705 $Itemcache = "$Cachedir/pod2htmi.tmp";
706 if (defined $opt_flush) {
707 1 while unlink($Dircache, $Itemcache);
715 my($dircache, $itemcache, $podpath, $podroot, $recurse) = @_;
716 my @cache_key_args = @_;
718 # A first-level cache:
719 # Don't bother reading the cache files if they still apply
720 # and haven't changed since we last read them.
722 my $this_cache_key = cache_key(@cache_key_args);
724 return if $Saved_Cache_Key and $this_cache_key eq $Saved_Cache_Key;
726 # load the cache of %Pages and %Items if possible. $tests will be
727 # non-zero if successful.
729 if (-f $dircache && -f $itemcache) {
730 warn "scanning for item cache\n" if $Verbose;
731 $tests = load_cache($dircache, $itemcache, $podpath, $podroot);
734 # if we didn't succeed in loading the cache then we must (re)build
737 warn "scanning directories in pod-path\n" if $Verbose;
738 scan_podpath($podroot, $recurse, 0);
740 $Saved_Cache_Key = cache_key(@cache_key_args);
744 my($dircache, $itemcache, $podpath, $podroot, $recurse) = @_;
745 return join('!', $dircache, $itemcache, $recurse,
746 @$podpath, $podroot, stat($dircache), stat($itemcache));
750 # load_cache - tries to find if the caches stored in $dircache and $itemcache
751 # are valid caches of %Pages and %Items. if they are valid then it loads
752 # them and returns a non-zero value.
755 my($dircache, $itemcache, $podpath, $podroot) = @_;
761 open(CACHE, "<$itemcache") ||
762 die "$0: error opening $itemcache for reading: $!\n";
765 # is it the same podpath?
768 $tests++ if (join(":", @$podpath) eq $_);
770 # is it the same podroot?
773 $tests++ if ($podroot eq $_);
775 # load the cache if its good
781 warn "loading item cache\n" if $Verbose;
788 warn "scanning for directory cache\n" if $Verbose;
789 open(CACHE, "<$dircache") ||
790 die "$0: error opening $dircache for reading: $!\n";
794 # is it the same podpath?
797 $tests++ if (join(":", @$podpath) eq $_);
799 # is it the same podroot?
802 $tests++ if ($podroot eq $_);
804 # load the cache if its good
810 warn "loading directory cache\n" if $Verbose;
822 # scan_podpath - scans the directories specified in @podpath for directories,
823 # .pod files, and .pm files. it also scans the pod files specified in
824 # @Libpods for =item directives.
827 my($podroot, $recurse, $append) = @_;
829 my($libpod, $dirname, $pod, @files, @poddata);
836 # scan each directory listed in @Podpath
839 || die "$0: error changing to directory $podroot: $!\n";
840 foreach $dir (@Podpath) {
841 scan_dir($dir, $recurse);
844 # scan the pods listed in @Libpods for =item directives
845 foreach $libpod (@Libpods) {
846 # if the page isn't defined then we won't know where to find it
848 next unless defined $Pages{$libpod} && $Pages{$libpod};
850 # if there is a directory then use the .pod and .pm files within it.
851 # NOTE: Only finds the first so-named directory in the tree.
852 # if ($Pages{$libpod} =~ /([^:]*[^(\.pod|\.pm)]):/) {
853 if ($Pages{$libpod} =~ /([^:]*(?<!\.pod)(?<!\.pm)):/) {
854 # find all the .pod and .pm files within the directory
856 opendir(DIR, $dirname) ||
857 die "$0: error opening directory $dirname: $!\n";
858 @files = grep(/(\.pod|\.pm)\z/ && ! -d $_, readdir(DIR));
861 # scan each .pod and .pm file for =item directives
862 foreach $pod (@files) {
863 open(POD, "<$dirname/$pod") ||
864 die "$0: error opening $dirname/$pod for input: $!\n";
867 clean_data( \@poddata );
869 scan_items( \%Items, "$dirname/$pod", @poddata);
872 # use the names of files as =item directives too.
873 ### Don't think this should be done this way - confuses issues.(WL)
874 ### foreach $pod (@files) {
875 ### $pod =~ /^(.*)(\.pod|\.pm)$/;
876 ### $Items{$1} = "$dirname/$1.html" if $1;
878 } elsif ($Pages{$libpod} =~ /([^:]*\.pod):/ ||
879 $Pages{$libpod} =~ /([^:]*\.pm):/) {
880 # scan the .pod or .pm file for =item directives
882 open(POD, "<$pod") ||
883 die "$0: error opening $pod for input: $!\n";
886 clean_data( \@poddata );
888 scan_items( \%Items, "$pod", @poddata);
890 warn "$0: shouldn't be here (line ".__LINE__."\n" unless $Quiet;
893 @poddata = (); # clean-up a bit
896 || die "$0: error changing to directory $pwd: $!\n";
898 # cache the item list for later use
899 warn "caching items for later use\n" if $Verbose;
900 open(CACHE, ">$Itemcache") ||
901 die "$0: error open $Itemcache for writing: $!\n";
903 print CACHE join(":", @Podpath) . "\n$podroot\n";
904 foreach my $key (keys %Items) {
905 print CACHE "$key $Items{$key}\n";
910 # cache the directory list for later use
911 warn "caching directories for later use\n" if $Verbose;
912 open(CACHE, ">$Dircache") ||
913 die "$0: error open $Dircache for writing: $!\n";
915 print CACHE join(":", @Podpath) . "\n$podroot\n";
916 foreach my $key (keys %Pages) {
917 print CACHE "$key $Pages{$key}\n";
924 # scan_dir - scans the directory specified in $dir for subdirectories, .pod
925 # files, and .pm files. notes those that it finds. this information will
926 # be used later in order to figure out where the pages specified in L<>
927 # links are on the filesystem.
930 my($dir, $recurse) = @_;
931 my($t, @subdirs, @pods, $pod, $dirname, @dirs);
937 opendir(DIR, $dir) ||
938 die "$0: error opening directory $dir: $!\n";
939 while (defined($_ = readdir(DIR))) {
940 if (-d "$dir/$_" && $_ ne "." && $_ ne ".."
941 && ($HiddenDirs || !/^\./)
943 $Pages{$_} = "" unless defined $Pages{$_};
944 $Pages{$_} .= "$dir/$_:";
946 } elsif (/\.pod\z/) { # .pod
948 $Pages{$_} = "" unless defined $Pages{$_};
949 $Pages{$_} .= "$dir/$_.pod:";
950 push(@pods, "$dir/$_.pod");
951 } elsif (/\.html\z/) { # .html
953 $Pages{$_} = "" unless defined $Pages{$_};
954 $Pages{$_} .= "$dir/$_.pod:";
955 } elsif (/\.pm\z/) { # .pm
957 $Pages{$_} = "" unless defined $Pages{$_};
958 $Pages{$_} .= "$dir/$_.pm:";
959 push(@pods, "$dir/$_.pm");
960 } elsif (-T "$dir/$_") { # script(?)
962 if (open(F, "$dir/$_")) {
964 while (defined($line = <F>)) {
965 if ($line =~ /^=(?:pod|head1)/) {
966 $Pages{$_} = "" unless defined $Pages{$_};
967 $Pages{$_} .= "$dir/$_.pod:";
977 # recurse on the subdirectories if necessary
979 foreach my $subdir (@subdirs) {
980 scan_dir("$dir/$subdir", $recurse);
986 # scan_headings - scan a pod file for head[1-6] tags, note the tags, and
990 my($sections, @data) = @_;
991 my($tag, $which_head, $otitle, $listdepth, $index);
998 # scan for =head directives, note their name, and build an index
999 # pointing to each of them.
1000 foreach my $line (@data) {
1001 if ($line =~ /^=(head)([1-6])\s+(.*)/) {
1002 ($tag, $which_head, $otitle) = ($1,$2,$3);
1004 my $title = depod( $otitle );
1005 my $name = anchorify( $title );
1006 $$sections{$name} = 1;
1007 $title = process_text( \$otitle );
1009 while ($which_head != $listdepth) {
1010 if ($which_head > $listdepth) {
1011 $index .= "\n" . ("\t" x $listdepth) . "<ul>\n";
1013 } elsif ($which_head < $listdepth) {
1015 $index .= "\n" . ("\t" x $listdepth) . "</ul>\n";
1019 $index .= "\n" . ("\t" x $listdepth) . "<li>" .
1020 "<a href=\"#" . $name . "\">" .
1021 $title . "</a></li>";
1025 # finish off the lists
1026 while ($listdepth--) {
1027 $index .= "\n" . ("\t" x $listdepth) . "</ul>\n";
1030 # get rid of bogus lists
1031 $index =~ s,\t*<ul>\s*</ul>\n,,g;
1037 # scan_items - scans the pod specified by $pod for =item directives. we
1038 # will use this information later on in resolving C<> links.
1041 my( $itemref, $pod, @poddata ) = @_;
1045 $pod =~ s/\.pod\z//;
1046 $pod .= ".html" if $pod;
1048 foreach $i (0..$#poddata) {
1049 my $txt = depod( $poddata[$i] );
1051 # figure out what kind of item it is.
1052 # Build string for referencing this item.
1053 if ( $txt =~ /\A=item\s+\*\s*(.*)\Z/s ) { # bullet
1056 } elsif( $txt =~ /\A=item\s+(?>\d+\.?)\s*(.*)\Z/s ) { # numbered list
1058 } elsif( $txt =~ /\A=item\s+(.*)\Z/s ) { # plain item
1063 my $fid = fragment_id( $item );
1064 $$itemref{$fid} = "$pod" if $fid;
1069 # process_head - convert a pod head[1-6] tag and convert it to HTML format.
1072 my($tag, $heading, $hasindex) = @_;
1074 # figure out the level of the =head
1075 $tag =~ /head([1-6])/;
1079 warn "$0: $Podfile: unterminated list at =head in paragraph $Paragraph. ignoring.\n" unless $Quiet;
1080 while( $Listlevel ){
1086 if( $level == 1 && ! $Top ){
1087 print HTML "<a href=\"#__index__\"><small>$Backlink</small></a>\n"
1088 if $hasindex and $Backlink;
1089 print HTML "</p>\n<hr />\n"
1091 print HTML "</p>\n";
1094 my $name = anchorify( depod( $heading ) );
1095 my $convert = process_text( \$heading );
1096 print HTML "<h$level><a name=\"$name\">$convert</a></h$level>\n";
1101 # emit_item_tag - print an =item's text
1102 # Note: The global $EmittedItem is used for inhibiting self-references.
1106 sub emit_item_tag($$$){
1107 my( $otext, $text, $compact ) = @_;
1108 my $item = fragment_id( $text );
1110 $EmittedItem = $item;
1111 ### print STDERR "emit_item_tag=$item ($text)\n";
1113 print HTML '<strong>';
1114 if ($Items_Named{$item}++) {
1115 print HTML process_text( \$otext );
1117 my $name = 'item_' . $item;
1118 $name = anchorify($name);
1119 print HTML qq{<a name="$name">}, process_text( \$otext ), '</a>';
1121 print HTML "</strong>\n";
1122 undef( $EmittedItem );
1127 if( $Items_Seen[$Listlevel]++ == 0 ){
1128 push( @Listend, "</$tag>" );
1129 print HTML "<$tag>\n";
1131 my $emitted = $tag eq 'dl' ? 'dt' : 'li';
1132 print HTML "<$emitted>";
1137 # process_item - convert a pod item tag and convert it to HTML format.
1141 my $need_dd = 0; # set to 1 if we need a <dd></dd> after an item
1143 # lots of documents start a list without doing an =over. this is
1144 # bad! but, the proper thing to do seems to be to just assume
1145 # they did do an =over. so warn them once and then continue.
1146 if( $Listlevel == 0 ){
1147 warn "$0: $Podfile: unexpected =item directive in paragraph $Paragraph. ignoring.\n" unless $Quiet;
1151 # formatting: insert a paragraph if preceding item has >1 paragraph
1153 print HTML $need_dd ? "</dd>\n" : "</li>\n" if $After_Lpar;
1157 # remove formatting instructions from the text
1158 my $text = depod( $otext );
1160 my $emitted; # the tag actually emitted, used for closing
1162 # all the list variants:
1163 if( $text =~ /\A\*/ ){ # bullet
1164 $emitted = emit_li( 'ul' );
1165 if ($text =~ /\A\*\s+(.+)\Z/s ) { # with additional text
1167 $otext =~ s/\A\*\s+//;
1168 emit_item_tag( $otext, $tag, 1 );
1171 } elsif( $text =~ /\A\d+/ ){ # numbered list
1172 $emitted = emit_li( 'ol' );
1173 if ($text =~ /\A(?>\d+\.?)\s*(.+)\Z/s ) { # with additional text
1175 $otext =~ s/\A\d+\.?\s*//;
1176 emit_item_tag( $otext, $tag, 1 );
1179 } else { # definition list
1180 $emitted = emit_li( 'dl' );
1181 if ($text =~ /\A(.+)\Z/s ){ # should have text
1182 emit_item_tag( $otext, $text, 1 );
1191 # process_over - process a pod over tag and start a corresponding HTML list.
1196 push( @Items_Seen, 0 );
1201 # process_back - process a pod back tag and convert it to HTML format.
1204 my $need_dd = shift;
1205 if( $Listlevel == 0 ){
1206 warn "$0: $Podfile: unexpected =back directive in paragraph $Paragraph. ignoring.\n" unless $Quiet;
1210 # close off the list. note, I check to see if $Listend[$Listlevel] is
1211 # defined because an =item directive may have never appeared and thus
1212 # $Listend[$Listlevel] may have never been initialized.
1214 if( defined $Listend[$Listlevel] ){
1215 print HTML $need_dd ? "</dd>\n" : "</li>\n" if $After_Lpar;
1216 print HTML $Listend[$Listlevel];
1222 # clean up item count
1227 # process_cut - process a pod cut tag, thus start ignoring pod directives.
1234 # process_pod - process a pod tag, thus stop ignoring pod directives
1235 # until we see a corresponding cut.
1238 # no need to set $Ignore to 0 cause the main loop did it
1242 # process_for - process a =for pod tag. if it's for html, spit
1243 # it out verbatim, if illustration, center it, otherwise ignore it.
1246 my($whom, $text) = @_;
1247 if ( $whom =~ /^(pod2)?html$/i) {
1249 } elsif ($whom =~ /^illustration$/i) {
1250 1 while chomp $text;
1251 for my $ext (qw[.png .gif .jpeg .jpg .tga .pcl .bmp]) {
1252 $text .= $ext, last if -r "$text$ext";
1254 print HTML qq{<p align="center"><img src="$text" alt="$text illustration" /></p>};
1259 # process_begin - process a =begin pod tag. this pushes
1260 # whom we're beginning on the begin stack. if there's a
1261 # begin stack, we only print if it us.
1264 my($whom, $text) = @_;
1266 push (@Begin_Stack, $whom);
1267 if ( $whom =~ /^(pod2)?html$/) {
1268 print HTML $text if $text;
1273 # process_end - process a =end pod tag. pop the
1274 # begin stack. die if we're mismatched.
1277 my($whom, $text) = @_;
1279 if ($Begin_Stack[-1] ne $whom ) {
1280 die "Unmatched begin/end at chunk $Paragraph\n"
1282 pop( @Begin_Stack );
1286 # process_pre - indented paragraph, made into <pre></pre>
1295 # insert spaces in place of tabs
1298 1 while $line =~ s/(\t+)/' ' x ((length($1) * 8) - $-[0] % 8)/e;
1302 # convert some special chars to HTML escapes
1303 $rest = html_escape($rest);
1305 # try and create links for all occurrences of perl.* within
1306 # the preformatted text.
1310 if ( defined $Pages{$2} ){ # is a link
1311 qq($1<a href="$Htmlroot/$Pages{$2}">$2</a>);
1312 } elsif (defined $Pages{dosify($2)}) { # is a link
1313 qq($1<a href="$Htmlroot/$Pages{dosify($2)}">$2</a>);
1319 (<a\ href="?) ([^>:]*:)? ([^>:]*) \.pod: ([^>:]*:)?
1322 if ( $Htmlfileurl ne '' ){
1323 # Here, we take advantage of the knowledge
1324 # that $Htmlfileurl ne '' implies $Htmlroot eq ''.
1325 # Since $Htmlroot eq '', we need to prepend $Htmldir
1326 # on the fron of the link to get the absolute path
1327 # of the link's target. We check for a leading '/'
1328 # to avoid corrupting links that are #, file:, etc.
1330 $old_url = "$Htmldir$old_url" if $old_url =~ m{^\/};
1331 $url = relativize_url( "$old_url.html", $Htmlfileurl );
1338 # Look for embedded URLs and make them into links. We don't
1339 # relativize them since they are best left as the author intended.
1341 my $urls = '(' . join ('|', qw{
1354 my $gunk = '/#~:.?+=&%@!\-';
1355 my $punc = '.:!?\-;';
1356 my $any = "${ltrs}${gunk}${punc}";
1359 \b # start at word boundary
1361 $urls : # need resource and a colon
1362 (?!:) # Ignore File::, among others.
1363 [$any] +? # followed by one or more of any valid
1364 # character, but be conservative and
1365 # take only what you need to....
1368 " > # maybe pre-quoted '<a href="...">'
1370 [$punc]* # 0 or more punctuation
1372 [^$any] # by a non-url char
1374 $ # end of the string
1377 $ # then end of the string
1379 }{<a href="$1">$1</a>}igox;
1381 # text should be as it is (verbatim)
1387 # pure text processing
1389 # pure_text/inIS_text: differ with respect to automatic C<> recognition.
1390 # we don't want this to happen within IS
1394 process_puretext( $text, \$PTQuote, 1 );
1399 process_puretext( $text, \$PTQuote, 0 );
1403 # process_puretext - process pure text (without pod-escapes) converting
1404 # double-quotes and handling implicit C<> links.
1406 sub process_puretext {
1407 my($text, $quote, $notinIS) = @_;
1409 ## Guessing at func() or [$@%&]*var references in plain text is destined
1410 ## to produce some strange looking ref's. uncomment to disable:
1413 my(@words, $lead, $trail);
1415 # convert double-quotes to single-quotes
1416 if( $$quote && $text =~ s/"/''/s ){
1419 while ($text =~ s/"([^"]*)"/``$1''/sg) {};
1420 $$quote = 1 if $text =~ s/"/``/s;
1422 # keep track of leading and trailing white-space
1423 $lead = ($text =~ s/\A(\s+)//s ? $1 : "");
1424 $trail = ($text =~ s/(\s+)\Z//s ? $1 : "");
1426 # split at space/non-space boundaries
1427 @words = split( /(?<=\s)(?=\S)|(?<=\S)(?=\s)/, $text );
1429 # process each word individually
1430 foreach my $word (@words) {
1432 next if $word =~ /^\s*$/;
1433 # see if we can infer a link
1434 if( $notinIS && $word =~ /^(\w+)\((.*)\)$/ ) {
1435 # has parenthesis so should have been a C<> ref
1436 ## try for a pagename (perlXXX(1))?
1437 my( $func, $args ) = ( $1, $2 );
1438 if( $args =~ /^\d+$/ ){
1439 my $url = page_sect( $word, '' );
1441 $word = "<a href=\"$url\">the $word manpage</a>";
1445 ## try function name for a link, append tt'ed argument list
1446 $word = emit_C( $func, '', "($args)");
1448 #### disabled. either all (including $\W, $\w+{.*} etc.) or nothing.
1449 ## } elsif( $notinIS && $word =~ /^[\$\@%&*]+\w+$/) {
1450 ## # perl variables, should be a C<> ref
1451 ## $word = emit_C( $word );
1453 } elsif ($word =~ m,^\w+://\w,) {
1455 # Don't relativize it: leave it as the author intended
1456 $word = qq(<a href="$word">$word</a>);
1457 } elsif ($word =~ /[\w.-]+\@[\w-]+\.\w/) {
1458 # looks like an e-mail address
1459 my ($w1, $w2, $w3) = ("", $word, "");
1460 ($w1, $w2, $w3) = ("(", $1, ")$2") if $word =~ /^\((.*?)\)(,?)/;
1461 ($w1, $w2, $w3) = ("<", $1, ">$2") if $word =~ /^<(.*?)>(,?)/;
1462 $word = qq($w1<a href="mailto:$w2">$w2</a>$w3);
1464 $word = html_escape($word) if $word =~ /["&<>]/;
1468 # put everything back together
1469 return $lead . join( '', @words ) . $trail;
1474 # process_text - handles plaintext that appears in the input pod file.
1475 # there may be pod commands embedded within the text so those must be
1476 # converted to html commands.
1479 sub process_text1($$;$$);
1480 sub pattern ($) { $_[0] ? '[^\S\n]+'.('>' x ($_[0] + 1)) : '>' }
1481 sub closing ($) { local($_) = shift; (defined && s/\s+$//) ? length : 0 }
1486 my $res = process_text1( 0, $tref );
1490 sub process_text1($$;$$){
1491 my( $lev, $rstr, $func, $closing ) = @_;
1494 unless (defined $func) {
1500 # B<text> - boldface
1501 $res = '<strong>' . process_text1( $lev, $rstr ) . '</strong>';
1503 } elsif( $func eq 'C' ){
1504 # C<code> - can be a ref or <code></code>
1505 # need to extract text
1506 my $par = go_ahead( $rstr, 'C', $closing );
1508 ## clean-up of the link target
1509 my $text = depod( $par );
1511 ### my $x = $par =~ /[BI]</ ? 'yes' : 'no' ;
1512 ### print STDERR "-->call emit_C($par) lev=$lev, par with BI=$x\n";
1514 $res = emit_C( $text, $lev > 1 || ($par =~ /[BI]</) );
1516 } elsif( $func eq 'E' ){
1517 # E<x> - convert to character
1518 $$rstr =~ s/^([^>]*)>//;
1520 $escape =~ s/^(\d+|X[\dA-F]+)$/#$1/i;
1523 } elsif( $func eq 'F' ){
1524 # F<filename> - italizice
1525 $res = '<em>' . process_text1( $lev, $rstr ) . '</em>';
1527 } elsif( $func eq 'I' ){
1528 # I<text> - italizice
1529 $res = '<em>' . process_text1( $lev, $rstr ) . '</em>';
1531 } elsif( $func eq 'L' ){
1533 ## L<text|cross-ref> => produce text, use cross-ref for linking
1534 ## L<cross-ref> => make text from cross-ref
1535 ## need to extract text
1536 my $par = go_ahead( $rstr, 'L', $closing );
1538 # some L<>'s that shouldn't be:
1539 # a) full-blown URL's are emitted as-is
1540 if( $par =~ m{^\w+://}s ){
1541 return make_URL_href( $par );
1543 # b) C<...> is stripped and treated as C<>
1544 if( $par =~ /^C<(.*)>$/ ){
1545 my $text = depod( $1 );
1546 return emit_C( $text, $lev > 1 || ($par =~ /[BI]</) );
1549 # analyze the contents
1550 $par =~ s/\n/ /g; # undo word-wrapped tags
1553 if( $par =~ s{^([^|]+)\|}{} ){
1557 # make sure sections start with a /
1560 my( $page, $section, $ident );
1562 # check for link patterns
1563 if( $par =~ m{^([^/]+?)/(?!")(.*?)$} ){ # name/ident
1564 # we've got a name/ident (no quotes)
1565 ( $page, $ident ) = ( $1, $2 );
1566 ### print STDERR "--> L<$par> to page $page, ident $ident\n";
1568 } elsif( $par =~ m{^(.*?)/"?(.*?)"?$} ){ # [name]/"section"
1569 # even though this should be a "section", we go for ident first
1570 ( $page, $ident ) = ( $1, $2 );
1571 ### print STDERR "--> L<$par> to page $page, section $section\n";
1573 } elsif( $par =~ /\s/ ){ # this must be a section with missing quotes
1574 ( $page, $section ) = ( '', $par );
1575 ### print STDERR "--> L<$par> to void page, section $section\n";
1578 ( $page, $section ) = ( $par, '' );
1579 ### print STDERR "--> L<$par> to page $par, void section\n";
1582 # now, either $section or $ident is defined. the convoluted logic
1583 # below tries to resolve L<> according to what the user specified.
1584 # failing this, we try to find the next best thing...
1585 my( $url, $ltext, $fid );
1588 if( defined $ident ){
1589 ## try to resolve $ident as an item
1590 ( $url, $fid ) = coderef( $page, $ident );
1592 if( ! defined( $linktext ) ){
1594 $linktext .= " in " if $ident && $page;
1595 $linktext .= "the $page manpage" if $page;
1597 ### print STDERR "got coderef url=$url\n";
1600 ## no luck: go for a section (auto-quoting!)
1603 ## now go for a section
1604 my $htmlsection = htmlify( $section );
1605 $url = page_sect( $page, $htmlsection );
1607 if( ! defined( $linktext ) ){
1608 $linktext = $section;
1609 $linktext .= " in " if $section && $page;
1610 $linktext .= "the $page manpage" if $page;
1612 ### print STDERR "got page/section url=$url\n";
1615 ## no luck: go for an ident
1622 ( $url, $fid ) = coderef( $page, $ident );
1624 if( ! defined( $linktext ) ){
1626 $linktext .= " in " if $ident && $page;
1627 $linktext .= "the $page manpage" if $page;
1629 ### print STDERR "got section=>coderef url=$url\n";
1633 # warning; show some text.
1634 $linktext = $opar unless defined $linktext;
1635 warn "$0: $Podfile: cannot resolve L<$opar> in paragraph $Paragraph.\n" unless $Quiet;
1638 # now we have a URL or just plain code
1639 $$rstr = $linktext . '>' . $$rstr;
1640 if( defined( $url ) ){
1641 $res = "<a href=\"$url\">" . process_text1( $lev, $rstr ) . '</a>';
1643 $res = '<em>' . process_text1( $lev, $rstr ) . '</em>';
1646 } elsif( $func eq 'S' ){
1647 # S<text> - non-breaking spaces
1648 $res = process_text1( $lev, $rstr );
1649 $res =~ s/ / /g;
1651 } elsif( $func eq 'X' ){
1653 $$rstr =~ s/^[^>]*>//;
1655 } elsif( $func eq 'Z' ){
1657 warn "$0: $Podfile: invalid X<> in paragraph $Paragraph.\n"
1658 unless $$rstr =~ s/^>// or $Quiet;
1661 my $term = pattern $closing;
1662 while( $$rstr =~ s/\A(.*?)(([BCEFILSXZ])<(<+[^\S\n]+)?|$term)//s ){
1663 # all others: either recurse into new function or
1664 # terminate at closing angle bracket(s)
1666 $pt .= $2 if !$3 && $lev == 1;
1667 $res .= $lev == 1 ? pure_text( $pt ) : inIS_text( $pt );
1668 return $res if !$3 && $lev > 1;
1670 $res .= process_text1( $lev, $rstr, $3, closing $4 );
1674 $res .= pure_text( $$rstr );
1676 warn "$0: $Podfile: undelimited $func<> in paragraph $Paragraph.\n" unless $Quiet;
1683 # go_ahead: extract text of an IS (can be nested)
1686 my( $rstr, $func, $closing ) = @_;
1688 my @closing = ($closing);
1690 s/\A(.*?)(([BCEFILSXZ])<(<+[^\S\n]+)?|@{[pattern $closing[0]]})//s ){
1694 return $res unless @closing;
1696 unshift @closing, closing $4;
1700 warn "$0: $Podfile: undelimited $func<> in paragraph $Paragraph.\n" unless $Quiet;
1705 # emit_C - output result of C<text>
1706 # $text is the depod-ed text
1709 my( $text, $nocode, $args ) = @_;
1710 $args = '' unless defined $args;
1712 my( $url, $fid ) = coderef( undef(), $text );
1714 # need HTML-safe text
1715 my $linktext = html_escape( "$text$args" );
1717 if( defined( $url ) &&
1718 (!defined( $EmittedItem ) || $EmittedItem ne $fid ) ){
1719 $res = "<a href=\"$url\"><code>$linktext</code></a>";
1720 } elsif( 0 && $nocode ){
1723 $res = "<code>$linktext</code>";
1729 # html_escape: make text safe for HTML
1733 $rest =~ s/&/&/g;
1734 $rest =~ s/</</g;
1735 $rest =~ s/>/>/g;
1736 $rest =~ s/"/"/g;
1737 # ' is only in XHTML, not HTML4. Be conservative
1738 #$rest =~ s/'/'/g;
1744 # dosify - convert filenames to 8.3
1748 return lc($str) if $^O eq 'VMS'; # VMS just needs casing
1751 $str =~ s/(\.\w+)/substr ($1,0,4)/ge;
1752 $str =~ s/(\w+)/substr ($1,0,8)/ge;
1758 # page_sect - make a URL from the text of a L<>
1761 my( $page, $section ) = @_;
1762 my( $linktext, $page83, $link); # work strings
1764 # check if we know that this is a section in this page
1765 if (!defined $Pages{$page} && defined $Sections{$page}) {
1768 ### print STDERR "reset page='', section=$section\n";
1771 $page83=dosify($page);
1772 $page=$page83 if (defined $Pages{$page83});
1774 $link = "#" . anchorify( $section );
1775 } elsif ( $page =~ /::/ ) {
1777 # Search page cache for an entry keyed under the html page name,
1778 # then look to see what directory that page might be in. NOTE:
1779 # this will only find one page. A better solution might be to produce
1780 # an intermediate page that is an index to all such pages.
1781 my $page_name = $page ;
1782 $page_name =~ s,^.*/,,s ;
1783 if ( defined( $Pages{ $page_name } ) &&
1784 $Pages{ $page_name } =~ /([^:]*$page)\.(?:pod|pm):/
1789 # NOTE: This branch assumes that all A::B pages are located in
1790 # $Htmlroot/A/B.html . This is often incorrect, since they are
1791 # often in $Htmlroot/lib/A/B.html or such like. Perhaps we could
1792 # analyze the contents of %Pages and figure out where any
1793 # cousins of A::B are, then assume that. So, if A::B isn't found,
1794 # but A::C is found in lib/A/C.pm, then A::B is assumed to be in
1795 # lib/A/B.pm. This is also limited, but it's an improvement.
1796 # Maybe a hints file so that the links point to the correct places
1800 $link = "$Htmlroot/$page.html";
1801 $link .= "#" . anchorify( $section ) if ($section);
1802 } elsif (!defined $Pages{$page}) {
1805 $section = anchorify( $section ) if $section ne "";
1806 ### print STDERR "...section=$section\n";
1808 # if there is a directory by the name of the page, then assume that an
1809 # appropriate section will exist in the subdirectory
1810 # if ($section ne "" && $Pages{$page} =~ /([^:]*[^(\.pod|\.pm)]):/) {
1811 if ($section ne "" && $Pages{$page} =~ /([^:]*(?<!\.pod)(?<!\.pm)):/) {
1812 $link = "$Htmlroot/$1/$section.html";
1813 ### print STDERR "...link=$link\n";
1815 # since there is no directory by the name of the page, the section will
1816 # have to exist within a .html of the same name. thus, make sure there
1817 # is a .pod or .pm that might become that .html
1819 $section = "#$section" if $section;
1820 ### print STDERR "...section=$section\n";
1822 # check if there is a .pod with the page name
1823 if ($Pages{$page} =~ /([^:]*)\.pod:/) {
1824 $link = "$Htmlroot/$1.html$section";
1825 } elsif ($Pages{$page} =~ /([^:]*)\.pm:/) {
1826 $link = "$Htmlroot/$1.html$section";
1834 # Here, we take advantage of the knowledge that $Htmlfileurl ne ''
1835 # implies $Htmlroot eq ''. This means that the link in question
1836 # needs a prefix of $Htmldir if it begins with '/'. The test for
1837 # the initial '/' is done to avoid '#'-only links, and to allow
1838 # for other kinds of links, like file:, ftp:, etc.
1840 if ( $Htmlfileurl ne '' ) {
1841 $link = "$Htmldir$link" if $link =~ m{^/}s;
1842 $url = relativize_url( $link, $Htmlfileurl );
1843 # print( " b: [$link,$Htmlfileurl,$url]\n" );
1856 # relativize_url - convert an absolute URL to one relative to a base URL.
1857 # Assumes both end in a filename.
1859 sub relativize_url {
1860 my ($dest,$source) = @_ ;
1862 my ($dest_volume,$dest_directory,$dest_file) =
1863 File::Spec::Unix->splitpath( $dest ) ;
1864 $dest = File::Spec::Unix->catpath( $dest_volume, $dest_directory, '' ) ;
1866 my ($source_volume,$source_directory,$source_file) =
1867 File::Spec::Unix->splitpath( $source ) ;
1868 $source = File::Spec::Unix->catpath( $source_volume, $source_directory, '' ) ;
1871 if ( $dest ne '' ) {
1872 $rel_path = File::Spec::Unix->abs2rel( $dest, $source ) ;
1875 if ( $rel_path ne '' &&
1876 substr( $rel_path, -1 ) ne '/' &&
1877 substr( $dest_file, 0, 1 ) ne '#'
1879 $rel_path .= "/$dest_file" ;
1882 $rel_path .= "$dest_file" ;
1890 # coderef - make URL from the text of a C<>
1893 my( $page, $item ) = @_;
1896 my $fid = fragment_id( $item );
1897 if( defined( $page ) && $page ne "" ){
1898 # we have been given a $page...
1901 # Do we take it? Item could be a section!
1902 my $base = $Items{$fid} || "";
1903 $base =~ s{[^/]*/}{};
1904 if( $base ne "$page.html" ){
1905 ### print STDERR "coderef( $page, $item ): items{$fid} = $Items{$fid} = $base => discard page!\n";
1910 # no page - local items precede cached items
1911 if( defined( $fid ) ){
1912 if( exists $Local_Items{$fid} ){
1913 $page = $Local_Items{$fid};
1915 $page = $Items{$fid};
1920 # if there was a pod file that we found earlier with an appropriate
1921 # =item directive, then create a link to that page.
1922 if( defined $page ){
1924 if( exists $Pages{$page} and $Pages{$page} =~ /([^:.]*)\.[^:]*:/){
1925 $page = $1 . '.html';
1927 my $link = "$Htmlroot/$page#item_" . anchorify($fid);
1929 # Here, we take advantage of the knowledge that $Htmlfileurl
1930 # ne '' implies $Htmlroot eq ''.
1931 if ( $Htmlfileurl ne '' ) {
1932 $link = "$Htmldir$link" ;
1933 $url = relativize_url( $link, $Htmlfileurl ) ;
1938 $url = "#item_" . anchorify($fid);
1941 confess "url has space: $url" if $url =~ /"[^"]*\s[^"]*"/;
1943 return( $url, $fid );
1949 # Adapted from Nick Ing-Simmons' PodToHtml package.
1951 my $source_file = shift ;
1952 my $destination_file = shift;
1954 my $source = URI::file->new_abs($source_file);
1955 my $uo = URI::file->new($destination_file,$source)->abs;
1956 return $uo->rel->as_string;
1961 # finish_list - finish off any pending HTML lists. this should be called
1962 # after the entire pod file has been read and converted.
1965 while ($Listlevel > 0) {
1966 print HTML "</dl>\n";
1972 # htmlify - converts a pod section specification to a suitable section
1973 # specification for HTML. Note that we keep spaces and special characters
1974 # except ", ? (Netscape problem) and the hyphen (writer's problem...).
1978 $heading =~ s/(\s+)/ /g;
1979 $heading =~ s/\s+\Z//;
1980 $heading =~ s/\A\s+//;
1981 # The hyphen is a disgrace to the English language.
1982 $heading =~ s/[-"?]//g;
1983 $heading = lc( $heading );
1988 # similar to htmlify, but turns non-alphanumerics into underscores
1992 $anchor = htmlify($anchor);
1993 $anchor =~ s/\W/_/g;
1998 # depod - convert text by eliminating all interior sequences
1999 # Note: can be called with copy or modify semantics
2006 $E2c{amp} = '&'; # in Tk's pods
2014 ${$_[0]} = depod1( \$string );
2022 my( $rstr, $func, $closing ) = @_;
2024 return $res unless defined $$rstr;
2025 if( ! defined( $func ) ){
2026 # skip to next begin of an interior sequence
2027 while( $$rstr =~ s/\A(.*?)([BCEFILSXZ])<(<+[^\S\n]+)?// ){
2028 # recurse into its text
2029 $res .= $1 . depod1( $rstr, $2, closing $3);
2032 } elsif( $func eq 'E' ){
2033 # E<x> - convert to character
2034 $$rstr =~ s/^([^>]*)>//;
2035 $res .= $E2c{$1} || "";
2036 } elsif( $func eq 'X' ){
2038 $$rstr =~ s/^[^>]*>//;
2039 } elsif( $func eq 'Z' ){
2043 # all others: either recurse into new function or
2044 # terminate at closing angle bracket
2045 my $term = pattern $closing;
2046 while( $$rstr =~ s/\A(.*?)(([BCEFILSXZ])<(<+[^\S\n]+)?|$term)// ){
2049 $res .= depod1( $rstr, $3, closing $4 );
2051 ## If we're here and $2 ne '>': undelimited interior sequence.
2052 ## Ignored, as this is called without proper indication of where we are.
2053 ## Rely on process_text to produce diagnostics.
2059 # fragment_id - construct a fragment identifier from:
2061 # b) contents of C<...>
2066 $text =~ s/\s+\Z//s;
2068 # a method or function?
2069 return $1 if $text =~ /(\w+)\s*\(/;
2070 return $1 if $text =~ /->\s*(\w+)\s*\(?/;
2073 return $1 if $text =~ /^([$@%*]\S+)/;
2075 # some pattern matching operator?
2076 return $1 if $text =~ m|^(\w+/).*/\w*$|;
2078 # fancy stuff... like "do { }"
2079 return $1 if $text =~ m|^(\w+)\s*{.*}$|;
2081 # honour the perlfunc manpage: func [PAR[,[ ]PAR]...]
2082 # and some funnies with ... Module ...
2083 return $1 if $text =~ m{^([a-z\d_]+)(\s+[A-Z\d,/& ]+)?$};
2084 return $1 if $text =~ m{^([a-z\d]+)\s+Module(\s+[A-Z\d,/& ]+)?$};
2087 $text =~ s/\s+/_/sg;
2089 defined( $HC[ord($1)] ) ? $HC[ord($1)]
2090 : ( $HC[ord($1)] = sprintf( "%%%02X", ord($1) ) ) }gxe;
2091 $text = substr( $text, 0, 50 );
2098 # make_URL_href - generate HTML href from URL
2099 # Special treatment for CGI queries.
2101 sub make_URL_href($){
2104 s{^(http:[-\w/#~:.+=&%@!]+)(\?.*)$}{<a href="$1$2">$1</a>}i ){
2105 $url = "<a href=\"$url\">$url</a>";