5 use vars qw($VERSION @ISA @EXPORT);
8 @EXPORT = qw(pod2html htmlify);
17 use locale; # make \w work right in non-ASCII lands
21 Pod::Html - module to convert pod files to HTML
30 Converts files from pod format (see L<perlpod>) to HTML format. It
31 can automatically generate indexes and cross-references, and it keeps
32 a cache of things it knows how to cross-reference.
36 Pod::Html takes the following arguments:
42 --backlink="Back to Top"
44 Adds "Back to Top" links in front of every C<head1> heading (except for
45 the first). By default, no backlinks are generated.
51 Creates the item and directory caches in the given directory.
57 Specify the URL of a cascading style sheet. Also disables all HTML/CSS
58 C<style> attributes that are output by default (to avoid conflicts).
64 Flushes the item and directory caches.
71 Creates header and footer blocks containing the text of the C<NAME>
72 section. By default, no headers are generated.
78 Displays the usage message.
84 Sets the directory in which the resulting HTML file is placed. This
85 is used to generate relative links to other files. Not passing this
86 causes all links to be absolute, since this is the value that tells
87 Pod::Html the root of the documentation tree.
93 Sets the base URL for the HTML files. When cross-references are made,
94 the HTML root is prepended to the URL.
101 Generate an index at the top of the HTML file. This is the default
108 Specify the pod file to convert. Input is taken from STDIN if no
113 --libpods=name:...:name
115 List of page names (eg, "perlfunc") which contain linkable C<=item>s.
122 B<Deprecated>, has no effect. For backwards compatibility only.
128 Specify the HTML file to create. Output goes to STDOUT if no outfile
133 --podpath=name:...:name
135 Specify which subdirectories of the podroot contain pod files whose
136 HTML converted forms can be linked to in cross references.
142 Specify the base directory for finding library pods.
149 Don't display I<mostly harmless> warning messages. These messages
150 will be displayed by default. But this is not the same as C<verbose>
158 Recurse into subdirectories specified in podpath (default behaviour).
164 Specify the title of the resulting HTML file.
171 Display progress messages. By default, they won't be displayed.
178 "--podpath=lib:ext:pod:vms",
179 "--podroot=/usr/src/perl",
180 "--htmlroot=/perl/nmanual",
181 "--libpods=perlfunc:perlguts:perlvar:perlrun:perlop",
184 "--outfile=/perl/nmanual/foo.html");
188 Uses C<$Config{pod2html}> to setup default options.
192 Tom Christiansen, E<lt>tchrist@perl.comE<gt>.
200 This program is distributed under the Artistic License.
204 my $cachedir = "."; # The directory to which item and directory
205 # caches will be written.
206 my $cache_ext = $^O eq 'VMS' ? ".tmp" : ".x~~";
207 my $dircache = "pod2htmd$cache_ext";
208 my $itemcache = "pod2htmi$cache_ext";
210 my @begin_stack = (); # begin/end stack
212 my @libpods = (); # files to search for links from C<> directives
213 my $htmlroot = "/"; # http-server base directory from which all
214 # relative paths in $podpath stem.
215 my $htmldir = ""; # The directory to which the html pages
216 # will (eventually) be written.
217 my $htmlfile = ""; # write to stdout by default
218 my $htmlfileurl = "" ; # The url that other files would use to
219 # refer to this file. This is only used
220 # to make relative urls that point to
222 my $podfile = ""; # read from stdin by default
223 my @podpath = (); # list of directories containing library pods.
224 my $podroot = File::Spec->curdir; # filesystem base directory from which all
225 # relative paths in $podpath stem.
226 my $css = ''; # Cascading style sheet
227 my $recurse = 1; # recurse on subdirectories in $podpath.
228 my $quiet = 0; # not quiet by default
229 my $verbose = 0; # not verbose by default
230 my $doindex = 1; # non-zero if we should generate an index
231 my $backlink = ''; # text for "back to top" links
232 my $listlevel = 0; # current list depth
233 my @listend = (); # the text to use to end the list.
234 my $after_lpar = 0; # set to true after a par in an =item
235 my $ignore = 1; # whether or not to format text. we don't
236 # format text until we hit our first pod
239 my %items_named = (); # for the multiples of the same item in perlfunc
241 my $title; # title to give the pod(s)
242 my $header = 0; # produce block header/footer
243 my $top = 1; # true if we are at the top of the doc. used
244 # to prevent the first <hr /> directive.
245 my $paragraph; # which paragraph we're processing (used
246 # for error messages)
247 my $ptQuote = 0; # status of double-quote conversion
248 my %pages = (); # associative array used to find the location
249 # of pages referenced by L<> links.
250 my %sections = (); # sections within this page
251 my %items = (); # associative array used to find the location
252 # of =item directives referenced by C<> links
253 my %local_items = (); # local items - avoid destruction of %items
254 my $Is83; # is dos with short filenames (8.3)
257 $dircache = "pod2htmd$cache_ext";
258 $itemcache = "pod2htmi$cache_ext";
260 @begin_stack = (); # begin/end stack
262 @libpods = (); # files to search for links from C<> directives
263 $htmlroot = "/"; # http-server base directory from which all
264 # relative paths in $podpath stem.
265 $htmldir = ""; # The directory to which the html pages
266 # will (eventually) be written.
267 $htmlfile = ""; # write to stdout by default
268 $podfile = ""; # read from stdin by default
269 @podpath = (); # list of directories containing library pods.
270 $podroot = File::Spec->curdir; # filesystem base directory from which all
271 # relative paths in $podpath stem.
272 $css = ''; # Cascading style sheet
273 $recurse = 1; # recurse on subdirectories in $podpath.
274 $quiet = 0; # not quiet by default
275 $verbose = 0; # not verbose by default
276 $doindex = 1; # non-zero if we should generate an index
277 $backlink = ''; # text for "back to top" links
278 $listlevel = 0; # current list depth
279 @listend = (); # the text to use to end the list.
280 $after_lpar = 0; # set to true after a par in an =item
281 $ignore = 1; # whether or not to format text. we don't
282 # format text until we hit our first pod
287 $header = 0; # produce block header/footer
288 $title = ''; # title to give the pod(s)
289 $top = 1; # true if we are at the top of the doc. used
290 # to prevent the first <hr /> directive.
291 $paragraph = ''; # which paragraph we're processing (used
292 # for error messages)
293 %sections = (); # sections within this page
295 # These are not reinitialised here but are kept as a cache.
296 # See get_cache and related cache management code.
297 #%pages = (); # associative array used to find the location
298 # of pages referenced by L<> links.
299 #%items = (); # associative array used to find the location
300 # of =item directives referenced by C<> links
306 # clean_data: global clean-up of pod data
310 for my $i ( 0..$#{$dataref} ) {
311 ${$dataref}[$i] =~ s/\s+\Z//;
313 # have a look for all-space lines
314 if( ${$dataref}[$i] =~ /^\s+$/m and $dataref->[$i] !~ /^\s/ ){
315 my @chunks = split( /^\s+$/m, ${$dataref}[$i] );
316 splice( @$dataref, $i, 1, @chunks );
329 $Is83 = 0 if (defined (&Dos::UseLFN) && Dos::UseLFN());
331 # cache of %pages and %items from last time we ran pod2html
333 #undef $opt_help if defined $opt_help;
335 # parse the command-line parameters
336 parse_command_line();
338 # escape the backlink argument (same goes for title but is done later...)
339 $backlink = html_escape($backlink) if defined $backlink;
341 # set some variables to their default values if necessary
343 unless (@ARGV && $ARGV[0]) {
344 $podfile = "-" unless $podfile; # stdin
345 open(POD, "<$podfile")
346 || die "$0: cannot open $podfile file for input: $!\n";
348 $podfile = $ARGV[0]; # XXX: might be more filenames
351 $htmlfile = "-" unless $htmlfile; # stdout
352 $htmlroot = "" if $htmlroot eq "/"; # so we don't get a //
353 $htmldir =~ s#/\z## ; # so we don't get a //
355 && defined( $htmldir )
357 && substr( $htmlfile, 0, length( $htmldir ) ) eq $htmldir
360 # Set the 'base' url for this file, so that we can use it
361 # as the location from which to calculate relative links
362 # to other files. If this is '', then absolute links will
363 # be used throughout.
364 $htmlfileurl= "$htmldir/" . substr( $htmlfile, length( $htmldir ) + 1);
367 # read the pod a paragraph at a time
368 warn "Scanning for sections in input file(s)\n" if $verbose;
372 clean_data( \@poddata );
374 # scan the pod for =head[1-6] directives and build an index
375 my $index = scan_headings(\%sections, @poddata);
378 warn "No headings in $podfile\n" if $verbose;
381 # open the output file
382 open(HTML, ">$htmlfile")
383 || die "$0: cannot open $htmlfile file for output: $!\n";
385 # put a title in the HTML file if one wasn't specified
388 for (my $i = 0; $i < @poddata; $i++) {
389 if ($poddata[$i] =~ /^=head1\s*NAME\b/m) {
390 for my $para ( @poddata[$i, $i+1] ) {
392 if ($title) = $para =~ /(\S+\s+-+.*\S)/s;
399 if (!$title and $podfile =~ /\.pod\z/) {
400 # probably a split pod so take first =head[12] as title
401 for (my $i = 0; $i < @poddata; $i++) {
402 last if ($title) = $poddata[$i] =~ /^=head[12]\s*(.*)/;
404 warn "adopted '$title' as title for $podfile\n"
405 if $verbose and $title;
408 $title =~ s/\s*\(.*\)//;
410 warn "$0: no title for $podfile.\n" unless $quiet;
411 $podfile =~ /^(.*)(\.[^.\/]+)?\z/s;
412 $title = ($podfile eq "-" ? 'No Title' : $1);
413 warn "using $title" if $verbose;
415 $title = html_escape($title);
418 my $bodystyle = ' style="background-color: white"';
419 my $tdstyle = ' style="background-color: #cccccc"';
422 $csslink = qq(\n<link rel="stylesheet" href="$css" type="text/css" />);
423 $csslink =~ s,\\,/,g;
424 $csslink =~ s,(/.):,$1|,;
429 my $block = $header ? <<END_OF_BLOCK : '';
430 <table border="0" width="100%" cellspacing="0" cellpadding="3">
431 <tr><td class="block"$tdstyle valign="middle">
432 <big><strong><span class="block"> $title</span></strong></big>
437 print HTML <<END_OF_HEAD;
438 <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
439 <html xmlns="http://www.w3.org/1999/xhtml">
441 <title>$title</title>$csslink
442 <link rev="made" href="mailto:$Config{perladmin}" />
449 # load/reload/validate/cache %pages and %items
450 get_cache($dircache, $itemcache, \@podpath, $podroot, $recurse);
452 # scan the pod for =item directives
453 scan_items( \%local_items, "", @poddata);
455 # put an index at the top of the file. note, if $doindex is 0 we
456 # still generate an index, but surround it with an html comment.
457 # that way some other program can extract it if desired.
459 print HTML "<p><a name=\"__index__\"></a></p>\n";
460 print HTML "<!-- INDEX BEGIN -->\n";
461 print HTML "<!--\n" unless $doindex;
463 print HTML "-->\n" unless $doindex;
464 print HTML "<!-- INDEX END -->\n\n";
465 print HTML "<hr />\n" if $doindex and $index;
467 # now convert this file
468 my $after_item; # set to true after an =item
470 warn "Converting input file $podfile\n" if $verbose;
471 foreach my $i (0..$#poddata){
472 $ptQuote = 0; # status of quote conversion
476 if (/^(=.*)/s) { # is it a pod directive?
481 if (/^=begin\s+(\S+)\s*(.*)/si) {# =begin
482 process_begin($1, $2);
483 } elsif (/^=end\s+(\S+)\s*(.*)/si) {# =end
485 } elsif (/^=cut/) { # =cut
487 } elsif (/^=pod/) { # =pod
490 next if @begin_stack && $begin_stack[-1] ne 'html';
492 if (/^=(head[1-6])\s+(.*\S)/s) { # =head[1-6] heading
493 process_head( $1, $2, $doindex && $index );
494 } elsif (/^=item\s*(.*\S)?/sm) { # =item text
495 $need_dd = process_item( $1 );
497 } elsif (/^=over\s*(.*)/) { # =over N
499 } elsif (/^=back/) { # =back
501 } elsif (/^=for\s+(\S+)\s*(.*)/si) {# =for
505 warn "$0: $podfile: unknown pod directive '$1' in "
506 . "paragraph $paragraph. ignoring.\n";
513 next if @begin_stack && $begin_stack[-1] ne 'html';
514 print HTML and next if @begin_stack && $begin_stack[-1] eq 'html';
515 print HTML "<dd>\n" if $need_dd;
517 if( $text =~ /\A\s+/ ){
518 process_pre( \$text );
519 print HTML "<pre>\n$text</pre>\n";
522 process_text( \$text );
524 # experimental: check for a paragraph where all lines
525 # have some ...\t...\t...\n pattern
527 my @lines = split( "\n", $text );
530 foreach my $line ( @lines ){
531 if( $line =~ /\S/ && $line !~ /\t/ ){
537 $text =~ s/\t+/<td>/g;
538 $text =~ s/^/<tr><td>/gm;
539 $text = '<table cellspacing="0" cellpadding="0">' .
544 ## end of experimental
547 print HTML "$text\n";
550 print HTML "<p>$text</p>\n";
553 print HTML "</dd>\n" if $need_dd;
558 # finish off any pending directives
562 print HTML "<p><a href=\"#__index__\"><small>$backlink</small></a></p>\n"
563 if $doindex and $index and $backlink;
565 print HTML <<END_OF_TAIL;
572 # close the html file
575 warn "Finished\n" if $verbose;
578 ##############################################################################
580 my $usage; # see below
583 warn "$0: $podfile: @_\n" if @_;
587 $usage =<<END_OF_USAGE;
588 Usage: $0 --help --htmlroot=<name> --infile=<name> --outfile=<name>
589 --podpath=<name>:...:<name> --podroot=<name>
590 --libpods=<name>:...:<name> --recurse --verbose --index
591 --netscape --norecurse --noindex --cachedir=<name>
593 --backlink - set text for "back to top" links (default: none).
594 --cachedir - directory for the item and directory cache files.
595 --css - stylesheet URL
596 --flush - flushes the item and directory caches.
597 --[no]header - produce block header/footer (default is no headers).
598 --help - prints this message.
599 --htmldir - directory for resulting HTML files.
600 --htmlroot - http-server base directory from which all relative paths
601 in podpath stem (default is /).
602 --[no]index - generate an index at the top of the resulting html
604 --infile - filename for the pod to convert (input taken from stdin
606 --libpods - colon-separated list of pages to search for =item pod
607 directives in as targets of C<> and implicit links (empty
608 by default). note, these are not filenames, but rather
609 page names like those that appear in L<> links.
610 --outfile - filename for the resulting html file (output sent to
612 --podpath - colon-separated list of directories containing library
613 pods (empty by default).
614 --podroot - filesystem base directory from which all relative paths
615 in podpath stem (default is .).
616 --[no]quiet - supress some benign warning messages (default is off).
617 --[no]recurse - recurse on those subdirectories listed in podpath
619 --title - title that will appear in resulting html file.
620 --[no]verbose - self-explanatory (off by default).
621 --[no]netscape - deprecated, has no effect. for backwards compatibility only.
625 sub parse_command_line {
626 my ($opt_backlink,$opt_cachedir,$opt_css,$opt_flush,$opt_header,$opt_help,
627 $opt_htmldir,$opt_htmlroot,$opt_index,$opt_infile,$opt_libpods,
628 $opt_netscape,$opt_outfile,$opt_podpath,$opt_podroot,$opt_quiet,
629 $opt_recurse,$opt_title,$opt_verbose);
631 unshift @ARGV, split ' ', $Config{pod2html} if $Config{pod2html};
632 my $result = GetOptions(
633 'backlink=s' => \$opt_backlink,
634 'cachedir=s' => \$opt_cachedir,
635 'css=s' => \$opt_css,
636 'flush' => \$opt_flush,
637 'header!' => \$opt_header,
638 'help' => \$opt_help,
639 'htmldir=s' => \$opt_htmldir,
640 'htmlroot=s' => \$opt_htmlroot,
641 'index!' => \$opt_index,
642 'infile=s' => \$opt_infile,
643 'libpods=s' => \$opt_libpods,
644 'netscape!' => \$opt_netscape,
645 'outfile=s' => \$opt_outfile,
646 'podpath=s' => \$opt_podpath,
647 'podroot=s' => \$opt_podroot,
648 'quiet!' => \$opt_quiet,
649 'recurse!' => \$opt_recurse,
650 'title=s' => \$opt_title,
651 'verbose!' => \$opt_verbose,
653 usage("-", "invalid parameters") if not $result;
655 usage("-") if defined $opt_help; # see if the user asked for help
656 $opt_help = ""; # just to make -w shut-up.
658 @podpath = split(":", $opt_podpath) if defined $opt_podpath;
659 @libpods = split(":", $opt_libpods) if defined $opt_libpods;
661 $backlink = $opt_backlink if defined $opt_backlink;
662 $cachedir = $opt_cachedir if defined $opt_cachedir;
663 $css = $opt_css if defined $opt_css;
664 $header = $opt_header if defined $opt_header;
665 $htmldir = $opt_htmldir if defined $opt_htmldir;
666 $htmlroot = $opt_htmlroot if defined $opt_htmlroot;
667 $doindex = $opt_index if defined $opt_index;
668 $podfile = $opt_infile if defined $opt_infile;
669 $htmlfile = $opt_outfile if defined $opt_outfile;
670 $podroot = $opt_podroot if defined $opt_podroot;
671 $quiet = $opt_quiet if defined $opt_quiet;
672 $recurse = $opt_recurse if defined $opt_recurse;
673 $title = $opt_title if defined $opt_title;
674 $verbose = $opt_verbose if defined $opt_verbose;
676 warn "Flushing item and directory caches\n"
677 if $opt_verbose && defined $opt_flush;
678 $dircache = "$cachedir/pod2htmd$cache_ext";
679 $itemcache = "$cachedir/pod2htmi$cache_ext";
680 unlink($dircache, $itemcache) if defined $opt_flush;
687 my($dircache, $itemcache, $podpath, $podroot, $recurse) = @_;
688 my @cache_key_args = @_;
690 # A first-level cache:
691 # Don't bother reading the cache files if they still apply
692 # and haven't changed since we last read them.
694 my $this_cache_key = cache_key(@cache_key_args);
696 return if $saved_cache_key and $this_cache_key eq $saved_cache_key;
698 # load the cache of %pages and %items if possible. $tests will be
699 # non-zero if successful.
701 if (-f $dircache && -f $itemcache) {
702 warn "scanning for item cache\n" if $verbose;
703 $tests = load_cache($dircache, $itemcache, $podpath, $podroot);
706 # if we didn't succeed in loading the cache then we must (re)build
709 warn "scanning directories in pod-path\n" if $verbose;
710 scan_podpath($podroot, $recurse, 0);
712 $saved_cache_key = cache_key(@cache_key_args);
716 my($dircache, $itemcache, $podpath, $podroot, $recurse) = @_;
717 return join('!', $dircache, $itemcache, $recurse,
718 @$podpath, $podroot, stat($dircache), stat($itemcache));
722 # load_cache - tries to find if the caches stored in $dircache and $itemcache
723 # are valid caches of %pages and %items. if they are valid then it loads
724 # them and returns a non-zero value.
727 my($dircache, $itemcache, $podpath, $podroot) = @_;
733 open(CACHE, "<$itemcache") ||
734 die "$0: error opening $itemcache for reading: $!\n";
737 # is it the same podpath?
740 $tests++ if (join(":", @$podpath) eq $_);
742 # is it the same podroot?
745 $tests++ if ($podroot eq $_);
747 # load the cache if its good
753 warn "loading item cache\n" if $verbose;
760 warn "scanning for directory cache\n" if $verbose;
761 open(CACHE, "<$dircache") ||
762 die "$0: error opening $dircache for reading: $!\n";
766 # is it the same podpath?
769 $tests++ if (join(":", @$podpath) eq $_);
771 # is it the same podroot?
774 $tests++ if ($podroot eq $_);
776 # load the cache if its good
782 warn "loading directory cache\n" if $verbose;
794 # scan_podpath - scans the directories specified in @podpath for directories,
795 # .pod files, and .pm files. it also scans the pod files specified in
796 # @libpods for =item directives.
799 my($podroot, $recurse, $append) = @_;
801 my($libpod, $dirname, $pod, @files, @poddata);
808 # scan each directory listed in @podpath
811 || die "$0: error changing to directory $podroot: $!\n";
812 foreach $dir (@podpath) {
813 scan_dir($dir, $recurse);
816 # scan the pods listed in @libpods for =item directives
817 foreach $libpod (@libpods) {
818 # if the page isn't defined then we won't know where to find it
820 next unless defined $pages{$libpod} && $pages{$libpod};
822 # if there is a directory then use the .pod and .pm files within it.
823 # NOTE: Only finds the first so-named directory in the tree.
824 # if ($pages{$libpod} =~ /([^:]*[^(\.pod|\.pm)]):/) {
825 if ($pages{$libpod} =~ /([^:]*(?<!\.pod)(?<!\.pm)):/) {
826 # find all the .pod and .pm files within the directory
828 opendir(DIR, $dirname) ||
829 die "$0: error opening directory $dirname: $!\n";
830 @files = grep(/(\.pod|\.pm)\z/ && ! -d $_, readdir(DIR));
833 # scan each .pod and .pm file for =item directives
834 foreach $pod (@files) {
835 open(POD, "<$dirname/$pod") ||
836 die "$0: error opening $dirname/$pod for input: $!\n";
839 clean_data( \@poddata );
841 scan_items( \%items, "$dirname/$pod", @poddata);
844 # use the names of files as =item directives too.
845 ### Don't think this should be done this way - confuses issues.(WL)
846 ### foreach $pod (@files) {
847 ### $pod =~ /^(.*)(\.pod|\.pm)$/;
848 ### $items{$1} = "$dirname/$1.html" if $1;
850 } elsif ($pages{$libpod} =~ /([^:]*\.pod):/ ||
851 $pages{$libpod} =~ /([^:]*\.pm):/) {
852 # scan the .pod or .pm file for =item directives
854 open(POD, "<$pod") ||
855 die "$0: error opening $pod for input: $!\n";
858 clean_data( \@poddata );
860 scan_items( \%items, "$pod", @poddata);
862 warn "$0: shouldn't be here (line ".__LINE__."\n";
865 @poddata = (); # clean-up a bit
868 || die "$0: error changing to directory $pwd: $!\n";
870 # cache the item list for later use
871 warn "caching items for later use\n" if $verbose;
872 open(CACHE, ">$itemcache") ||
873 die "$0: error open $itemcache for writing: $!\n";
875 print CACHE join(":", @podpath) . "\n$podroot\n";
876 foreach my $key (keys %items) {
877 print CACHE "$key $items{$key}\n";
882 # cache the directory list for later use
883 warn "caching directories for later use\n" if $verbose;
884 open(CACHE, ">$dircache") ||
885 die "$0: error open $dircache for writing: $!\n";
887 print CACHE join(":", @podpath) . "\n$podroot\n";
888 foreach my $key (keys %pages) {
889 print CACHE "$key $pages{$key}\n";
896 # scan_dir - scans the directory specified in $dir for subdirectories, .pod
897 # files, and .pm files. notes those that it finds. this information will
898 # be used later in order to figure out where the pages specified in L<>
899 # links are on the filesystem.
902 my($dir, $recurse) = @_;
903 my($t, @subdirs, @pods, $pod, $dirname, @dirs);
909 opendir(DIR, $dir) ||
910 die "$0: error opening directory $dir: $!\n";
911 while (defined($_ = readdir(DIR))) {
912 if (-d "$dir/$_" && $_ ne "." && $_ ne "..") { # directory
913 $pages{$_} = "" unless defined $pages{$_};
914 $pages{$_} .= "$dir/$_:";
916 } elsif (/\.pod\z/) { # .pod
918 $pages{$_} = "" unless defined $pages{$_};
919 $pages{$_} .= "$dir/$_.pod:";
920 push(@pods, "$dir/$_.pod");
921 } elsif (/\.html\z/) { # .html
923 $pages{$_} = "" unless defined $pages{$_};
924 $pages{$_} .= "$dir/$_.pod:";
925 } elsif (/\.pm\z/) { # .pm
927 $pages{$_} = "" unless defined $pages{$_};
928 $pages{$_} .= "$dir/$_.pm:";
929 push(@pods, "$dir/$_.pm");
934 # recurse on the subdirectories if necessary
936 foreach my $subdir (@subdirs) {
937 scan_dir("$dir/$subdir", $recurse);
943 # scan_headings - scan a pod file for head[1-6] tags, note the tags, and
947 my($sections, @data) = @_;
948 my($tag, $which_head, $otitle, $listdepth, $index);
950 # here we need local $ignore = 0;
951 # unfortunately, we can't have it, because $ignore is lexical
957 # scan for =head directives, note their name, and build an index
958 # pointing to each of them.
959 foreach my $line (@data) {
960 if ($line =~ /^=(head)([1-6])\s+(.*)/) {
961 ($tag, $which_head, $otitle) = ($1,$2,$3);
963 my $title = depod( $otitle );
964 my $name = htmlify( $title );
965 $$sections{$name} = 1;
966 $name =~ s/\s/_/g; # htmlify keeps spaces but we don't want them here..
967 $title = process_text( \$otitle );
969 while ($which_head != $listdepth) {
970 if ($which_head > $listdepth) {
971 $index .= "\n" . ("\t" x $listdepth) . "<ul>\n";
973 } elsif ($which_head < $listdepth) {
975 $index .= "\n" . ("\t" x $listdepth) . "</ul>\n";
979 $index .= "\n" . ("\t" x $listdepth) . "<li>" .
980 "<a href=\"#" . $name . "\">" .
981 $title . "</a></li>";
985 # finish off the lists
986 while ($listdepth--) {
987 $index .= "\n" . ("\t" x $listdepth) . "</ul>\n";
990 # get rid of bogus lists
991 $index =~ s,\t*<ul>\s*</ul>\n,,g;
993 $ignore = 1; # restore old value;
999 # scan_items - scans the pod specified by $pod for =item directives. we
1000 # will use this information later on in resolving C<> links.
1003 my( $itemref, $pod, @poddata ) = @_;
1007 $pod =~ s/\.pod\z//;
1008 $pod .= ".html" if $pod;
1010 foreach $i (0..$#poddata) {
1011 my $txt = depod( $poddata[$i] );
1013 # figure out what kind of item it is.
1014 # Build string for referencing this item.
1015 if ( $txt =~ /\A=item\s+\*\s*(.*)\Z/s ) { # bullet
1018 } elsif( $txt =~ /\A=item\s+(?>\d+\.?)\s*(.*)\Z/s ) { # numbered list
1020 } elsif( $txt =~ /\A=item\s+(.*)\Z/s ) { # plain item
1025 my $fid = fragment_id( $item );
1026 $$itemref{$fid} = "$pod" if $fid;
1031 # process_head - convert a pod head[1-6] tag and convert it to HTML format.
1034 my($tag, $heading, $hasindex) = @_;
1036 # figure out the level of the =head
1037 $tag =~ /head([1-6])/;
1041 warn "$0: $podfile: unterminated list at =head in paragraph $paragraph. ignoring.\n";
1042 while( $listlevel ){
1048 if( $level == 1 && ! $top ){
1049 print HTML "<a href=\"#__index__\"><small>$backlink</small></a>\n"
1050 if $hasindex and $backlink;
1051 print HTML "</p>\n<hr />\n"
1053 print HTML "</p>\n";
1056 my $name = anchorify( depod( $heading ) );
1057 my $convert = process_text( \$heading );
1058 print HTML "<h$level><a name=\"$name\">$convert</a></h$level>\n";
1063 # emit_item_tag - print an =item's text
1064 # Note: The global $EmittedItem is used for inhibiting self-references.
1068 sub emit_item_tag($$$){
1069 my( $otext, $text, $compact ) = @_;
1070 my $item = fragment_id( $text );
1072 $EmittedItem = $item;
1073 ### print STDERR "emit_item_tag=$item ($text)\n";
1075 print HTML '<strong>';
1076 if ($items_named{$item}++) {
1077 print HTML process_text( \$otext );
1079 my $name = 'item_' . $item;
1080 $name = anchorify($name);
1081 print HTML qq{<a name="$name">}, process_text( \$otext ), '</a>';
1083 print HTML "</strong><br />\n";
1084 undef( $EmittedItem );
1089 if( $items_seen[$listlevel]++ == 0 ){
1090 push( @listend, "</$tag>" );
1091 print HTML "<$tag>\n";
1093 my $emitted = $tag eq 'dl' ? 'dt' : 'li';
1094 print HTML "<$emitted>";
1099 # process_item - convert a pod item tag and convert it to HTML format.
1103 my $need_dd = 0; # set to 1 if we need a <dd></dd> after an item
1105 # lots of documents start a list without doing an =over. this is
1106 # bad! but, the proper thing to do seems to be to just assume
1107 # they did do an =over. so warn them once and then continue.
1108 if( $listlevel == 0 ){
1109 warn "$0: $podfile: unexpected =item directive in paragraph $paragraph. ignoring.\n";
1113 # formatting: insert a paragraph if preceding item has >1 paragraph
1115 print HTML "<p></p>\n";
1119 # remove formatting instructions from the text
1120 my $text = depod( $otext );
1122 my $emitted; # the tag actually emitted, used for closing
1124 # all the list variants:
1125 if( $text =~ /\A\*/ ){ # bullet
1126 $emitted = emit_li( 'ul' );
1127 if ($text =~ /\A\*\s+(.+)\Z/s ) { # with additional text
1129 $otext =~ s/\A\*\s+//;
1130 emit_item_tag( $otext, $tag, 1 );
1133 } elsif( $text =~ /\A\d+/ ){ # numbered list
1134 $emitted = emit_li( 'ol' );
1135 if ($text =~ /\A(?>\d+\.?)\s*(.+)\Z/s ) { # with additional text
1137 $otext =~ s/\A\d+\.?\s*//;
1138 emit_item_tag( $otext, $tag, 1 );
1141 } else { # definition list
1142 $emitted = emit_li( 'dl' );
1143 if ($text =~ /\A(.+)\Z/s ){ # should have text
1144 emit_item_tag( $otext, $text, 1 );
1148 print HTML "</$emitted>" if $emitted;
1154 # process_over - process a pod over tag and start a corresponding HTML list.
1159 push( @items_seen, 0 );
1164 # process_back - process a pod back tag and convert it to HTML format.
1167 if( $listlevel == 0 ){
1168 warn "$0: $podfile: unexpected =back directive in paragraph $paragraph. ignoring.\n";
1172 # close off the list. note, I check to see if $listend[$listlevel] is
1173 # defined because an =item directive may have never appeared and thus
1174 # $listend[$listlevel] may have never been initialized.
1176 if( defined $listend[$listlevel] ){
1177 print HTML '<p></p>' if $after_lpar;
1178 print HTML $listend[$listlevel];
1184 # clean up item count
1189 # process_cut - process a pod cut tag, thus start ignoring pod directives.
1196 # process_pod - process a pod tag, thus stop ignoring pod directives
1197 # until we see a corresponding cut.
1200 # no need to set $ignore to 0 cause the main loop did it
1204 # process_for - process a =for pod tag. if it's for html, spit
1205 # it out verbatim, if illustration, center it, otherwise ignore it.
1208 my($whom, $text) = @_;
1209 if ( $whom =~ /^(pod2)?html$/i) {
1211 } elsif ($whom =~ /^illustration$/i) {
1212 1 while chomp $text;
1213 for my $ext (qw[.png .gif .jpeg .jpg .tga .pcl .bmp]) {
1214 $text .= $ext, last if -r "$text$ext";
1216 print HTML qq{<p align="center"><img src="$text" alt="$text illustration" /></p>};
1221 # process_begin - process a =begin pod tag. this pushes
1222 # whom we're beginning on the begin stack. if there's a
1223 # begin stack, we only print if it us.
1226 my($whom, $text) = @_;
1228 push (@begin_stack, $whom);
1229 if ( $whom =~ /^(pod2)?html$/) {
1230 print HTML $text if $text;
1235 # process_end - process a =end pod tag. pop the
1236 # begin stack. die if we're mismatched.
1239 my($whom, $text) = @_;
1241 if ($begin_stack[-1] ne $whom ) {
1242 die "Unmatched begin/end at chunk $paragraph\n"
1244 pop( @begin_stack );
1248 # process_pre - indented paragraph, made into <pre></pre>
1257 # insert spaces in place of tabs
1260 1 while $line =~ s/(\t+)/' ' x ((length($1) * 8) - $-[0] % 8)/e;
1264 # convert some special chars to HTML escapes
1265 $rest = html_escape($rest);
1267 # try and create links for all occurrences of perl.* within
1268 # the preformatted text.
1272 if ( defined $pages{$2} ){ # is a link
1273 qq($1<a href="$htmlroot/$pages{$2}">$2</a>);
1274 } elsif (defined $pages{dosify($2)}) { # is a link
1275 qq($1<a href="$htmlroot/$pages{dosify($2)}">$2</a>);
1281 (<a\ href="?) ([^>:]*:)? ([^>:]*) \.pod: ([^>:]*:)?
1284 if ( $htmlfileurl ne '' ){
1285 # Here, we take advantage of the knowledge
1286 # that $htmlfileurl ne '' implies $htmlroot eq ''.
1287 # Since $htmlroot eq '', we need to prepend $htmldir
1288 # on the fron of the link to get the absolute path
1289 # of the link's target. We check for a leading '/'
1290 # to avoid corrupting links that are #, file:, etc.
1292 $old_url = "$htmldir$old_url" if $old_url =~ m{^\/};
1293 $url = relativize_url( "$old_url.html", $htmlfileurl );
1300 # Look for embedded URLs and make them into links. We don't
1301 # relativize them since they are best left as the author intended.
1303 my $urls = '(' . join ('|', qw{
1316 my $gunk = '/#~:.?+=&%@!\-';
1317 my $punc = '.:!?\-;';
1318 my $any = "${ltrs}${gunk}${punc}";
1321 \b # start at word boundary
1323 $urls : # need resource and a colon
1324 (?!:) # Ignore File::, among others.
1325 [$any] +? # followed by one or more of any valid
1326 # character, but be conservative and
1327 # take only what you need to....
1329 (?= # look-ahead non-consumptive assertion
1330 [$punc]* # either 0 or more punctuation
1332 [^$any] # by a non-url char
1334 $ # end of the string
1337 $ # then end of the string
1339 }{<a href="$1">$1</a>}igox;
1341 # text should be as it is (verbatim)
1347 # pure text processing
1349 # pure_text/inIS_text: differ with respect to automatic C<> recognition.
1350 # we don't want this to happen within IS
1354 process_puretext( $text, \$ptQuote, 1 );
1359 process_puretext( $text, \$ptQuote, 0 );
1363 # process_puretext - process pure text (without pod-escapes) converting
1364 # double-quotes and handling implicit C<> links.
1366 sub process_puretext {
1367 my($text, $quote, $notinIS) = @_;
1369 ## Guessing at func() or [$@%&]*var references in plain text is destined
1370 ## to produce some strange looking ref's. uncomment to disable:
1373 my(@words, $lead, $trail);
1375 # convert double-quotes to single-quotes
1376 if( $$quote && $text =~ s/"/''/s ){
1379 while ($text =~ s/"([^"]*)"/``$1''/sg) {};
1380 $$quote = 1 if $text =~ s/"/``/s;
1382 # keep track of leading and trailing white-space
1383 $lead = ($text =~ s/\A(\s+)//s ? $1 : "");
1384 $trail = ($text =~ s/(\s+)\Z//s ? $1 : "");
1386 # split at space/non-space boundaries
1387 @words = split( /(?<=\s)(?=\S)|(?<=\S)(?=\s)/, $text );
1389 # process each word individually
1390 foreach my $word (@words) {
1392 next if $word =~ /^\s*$/;
1393 # see if we can infer a link
1394 if( $notinIS && $word =~ /^(\w+)\((.*)\)$/ ) {
1395 # has parenthesis so should have been a C<> ref
1396 ## try for a pagename (perlXXX(1))?
1397 my( $func, $args ) = ( $1, $2 );
1398 if( $args =~ /^\d+$/ ){
1399 my $url = page_sect( $word, '' );
1401 $word = "<a href=\"$url\">the $word manpage</a>";
1405 ## try function name for a link, append tt'ed argument list
1406 $word = emit_C( $func, '', "($args)");
1408 #### disabled. either all (including $\W, $\w+{.*} etc.) or nothing.
1409 ## } elsif( $notinIS && $word =~ /^[\$\@%&*]+\w+$/) {
1410 ## # perl variables, should be a C<> ref
1411 ## $word = emit_C( $word );
1413 } elsif ($word =~ m,^\w+://\w,) {
1415 # Don't relativize it: leave it as the author intended
1416 $word = qq(<a href="$word">$word</a>);
1417 } elsif ($word =~ /[\w.-]+\@[\w-]+\.\w/) {
1418 # looks like an e-mail address
1419 my ($w1, $w2, $w3) = ("", $word, "");
1420 ($w1, $w2, $w3) = ("(", $1, ")$2") if $word =~ /^\((.*?)\)(,?)/;
1421 ($w1, $w2, $w3) = ("<", $1, ">$2") if $word =~ /^<(.*?)>(,?)/;
1422 $word = qq($w1<a href="mailto:$w2">$w2</a>$w3);
1424 $word = html_escape($word) if $word =~ /["&<>]/;
1428 # put everything back together
1429 return $lead . join( '', @words ) . $trail;
1434 # process_text - handles plaintext that appears in the input pod file.
1435 # there may be pod commands embedded within the text so those must be
1436 # converted to html commands.
1439 sub process_text1($$;$$);
1440 sub pattern ($) { $_[0] ? '[^\S\n]+'.('>' x ($_[0] + 1)) : '>' }
1441 sub closing ($) { local($_) = shift; (defined && s/\s+$//) ? length : 0 }
1446 my $res = process_text1( 0, $tref );
1450 sub process_text1($$;$$){
1451 my( $lev, $rstr, $func, $closing ) = @_;
1454 unless (defined $func) {
1460 # B<text> - boldface
1461 $res = '<strong>' . process_text1( $lev, $rstr ) . '</strong>';
1463 } elsif( $func eq 'C' ){
1464 # C<code> - can be a ref or <code></code>
1465 # need to extract text
1466 my $par = go_ahead( $rstr, 'C', $closing );
1468 ## clean-up of the link target
1469 my $text = depod( $par );
1471 ### my $x = $par =~ /[BI]</ ? 'yes' : 'no' ;
1472 ### print STDERR "-->call emit_C($par) lev=$lev, par with BI=$x\n";
1474 $res = emit_C( $text, $lev > 1 || ($par =~ /[BI]</) );
1476 } elsif( $func eq 'E' ){
1477 # E<x> - convert to character
1478 $$rstr =~ s/^([^>]*)>//;
1480 $escape =~ s/^(\d+|X[\dA-F]+)$/#$1/i;
1483 } elsif( $func eq 'F' ){
1484 # F<filename> - italizice
1485 $res = '<em>' . process_text1( $lev, $rstr ) . '</em>';
1487 } elsif( $func eq 'I' ){
1488 # I<text> - italizice
1489 $res = '<em>' . process_text1( $lev, $rstr ) . '</em>';
1491 } elsif( $func eq 'L' ){
1493 ## L<text|cross-ref> => produce text, use cross-ref for linking
1494 ## L<cross-ref> => make text from cross-ref
1495 ## need to extract text
1496 my $par = go_ahead( $rstr, 'L', $closing );
1498 # some L<>'s that shouldn't be:
1499 # a) full-blown URL's are emitted as-is
1500 if( $par =~ m{^\w+://}s ){
1501 return make_URL_href( $par );
1503 # b) C<...> is stripped and treated as C<>
1504 if( $par =~ /^C<(.*)>$/ ){
1505 my $text = depod( $1 );
1506 return emit_C( $text, $lev > 1 || ($par =~ /[BI]</) );
1509 # analyze the contents
1510 $par =~ s/\n/ /g; # undo word-wrapped tags
1513 if( $par =~ s{^([^|]+)\|}{} ){
1517 # make sure sections start with a /
1520 my( $page, $section, $ident );
1522 # check for link patterns
1523 if( $par =~ m{^([^/]+?)/(?!")(.*?)$} ){ # name/ident
1524 # we've got a name/ident (no quotes)
1525 ( $page, $ident ) = ( $1, $2 );
1526 ### print STDERR "--> L<$par> to page $page, ident $ident\n";
1528 } elsif( $par =~ m{^(.*?)/"?(.*?)"?$} ){ # [name]/"section"
1529 # even though this should be a "section", we go for ident first
1530 ( $page, $ident ) = ( $1, $2 );
1531 ### print STDERR "--> L<$par> to page $page, section $section\n";
1533 } elsif( $par =~ /\s/ ){ # this must be a section with missing quotes
1534 ( $page, $section ) = ( '', $par );
1535 ### print STDERR "--> L<$par> to void page, section $section\n";
1538 ( $page, $section ) = ( $par, '' );
1539 ### print STDERR "--> L<$par> to page $par, void section\n";
1542 # now, either $section or $ident is defined. the convoluted logic
1543 # below tries to resolve L<> according to what the user specified.
1544 # failing this, we try to find the next best thing...
1545 my( $url, $ltext, $fid );
1548 if( defined $ident ){
1549 ## try to resolve $ident as an item
1550 ( $url, $fid ) = coderef( $page, $ident );
1552 if( ! defined( $linktext ) ){
1554 $linktext .= " in " if $ident && $page;
1555 $linktext .= "the $page manpage" if $page;
1557 ### print STDERR "got coderef url=$url\n";
1560 ## no luck: go for a section (auto-quoting!)
1563 ## now go for a section
1564 my $htmlsection = htmlify( $section );
1565 $url = page_sect( $page, $htmlsection );
1567 if( ! defined( $linktext ) ){
1568 $linktext = $section;
1569 $linktext .= " in " if $section && $page;
1570 $linktext .= "the $page manpage" if $page;
1572 ### print STDERR "got page/section url=$url\n";
1575 ## no luck: go for an ident
1582 ( $url, $fid ) = coderef( $page, $ident );
1584 if( ! defined( $linktext ) ){
1586 $linktext .= " in " if $ident && $page;
1587 $linktext .= "the $page manpage" if $page;
1589 ### print STDERR "got section=>coderef url=$url\n";
1593 # warning; show some text.
1594 $linktext = $opar unless defined $linktext;
1595 warn "$0: $podfile: cannot resolve L<$opar> in paragraph $paragraph.\n";
1598 # now we have a URL or just plain code
1599 $$rstr = $linktext . '>' . $$rstr;
1600 if( defined( $url ) ){
1601 $res = "<a href=\"$url\">" . process_text1( $lev, $rstr ) . '</a>';
1603 $res = '<em>' . process_text1( $lev, $rstr ) . '</em>';
1606 } elsif( $func eq 'S' ){
1607 # S<text> - non-breaking spaces
1608 $res = process_text1( $lev, $rstr );
1609 $res =~ s/ / /g;
1611 } elsif( $func eq 'X' ){
1613 $$rstr =~ s/^[^>]*>//;
1615 } elsif( $func eq 'Z' ){
1617 warn "$0: $podfile: invalid X<> in paragraph $paragraph.\n"
1618 unless $$rstr =~ s/^>//;
1621 my $term = pattern $closing;
1622 while( $$rstr =~ s/\A(.*?)(([BCEFILSXZ])<(<+[^\S\n]+)?|$term)//s ){
1623 # all others: either recurse into new function or
1624 # terminate at closing angle bracket(s)
1626 $pt .= $2 if !$3 && $lev == 1;
1627 $res .= $lev == 1 ? pure_text( $pt ) : inIS_text( $pt );
1628 return $res if !$3 && $lev > 1;
1630 $res .= process_text1( $lev, $rstr, $3, closing $4 );
1634 $res .= pure_text( $$rstr );
1636 warn "$0: $podfile: undelimited $func<> in paragraph $paragraph.\n";
1643 # go_ahead: extract text of an IS (can be nested)
1646 my( $rstr, $func, $closing ) = @_;
1648 my @closing = ($closing);
1650 s/\A(.*?)(([BCEFILSXZ])<(<+[^\S\n]+)?|@{[pattern $closing[0]]})//s ){
1654 return $res unless @closing;
1656 unshift @closing, closing $4;
1660 warn "$0: $podfile: undelimited $func<> in paragraph $paragraph.\n";
1665 # emit_C - output result of C<text>
1666 # $text is the depod-ed text
1669 my( $text, $nocode, $args ) = @_;
1670 $args = '' unless defined $args;
1672 my( $url, $fid ) = coderef( undef(), $text );
1674 # need HTML-safe text
1675 my $linktext = html_escape( "$text$args" );
1677 if( defined( $url ) &&
1678 (!defined( $EmittedItem ) || $EmittedItem ne $fid ) ){
1679 $res = "<a href=\"$url\"><code>$linktext</code></a>";
1680 } elsif( 0 && $nocode ){
1683 $res = "<code>$linktext</code>";
1689 # html_escape: make text safe for HTML
1693 $rest =~ s/&/&/g;
1694 $rest =~ s/</</g;
1695 $rest =~ s/>/>/g;
1696 $rest =~ s/"/"/g;
1697 $rest =~ s/'/'/g;
1703 # dosify - convert filenames to 8.3
1707 return lc($str) if $^O eq 'VMS'; # VMS just needs casing
1710 $str =~ s/(\.\w+)/substr ($1,0,4)/ge;
1711 $str =~ s/(\w+)/substr ($1,0,8)/ge;
1717 # page_sect - make a URL from the text of a L<>
1720 my( $page, $section ) = @_;
1721 my( $linktext, $page83, $link); # work strings
1723 # check if we know that this is a section in this page
1724 if (!defined $pages{$page} && defined $sections{$page}) {
1727 ### print STDERR "reset page='', section=$section\n";
1730 $page83=dosify($page);
1731 $page=$page83 if (defined $pages{$page83});
1733 $link = "#" . anchorify( $section );
1734 } elsif ( $page =~ /::/ ) {
1736 # Search page cache for an entry keyed under the html page name,
1737 # then look to see what directory that page might be in. NOTE:
1738 # this will only find one page. A better solution might be to produce
1739 # an intermediate page that is an index to all such pages.
1740 my $page_name = $page ;
1741 $page_name =~ s,^.*/,,s ;
1742 if ( defined( $pages{ $page_name } ) &&
1743 $pages{ $page_name } =~ /([^:]*$page)\.(?:pod|pm):/
1748 # NOTE: This branch assumes that all A::B pages are located in
1749 # $htmlroot/A/B.html . This is often incorrect, since they are
1750 # often in $htmlroot/lib/A/B.html or such like. Perhaps we could
1751 # analyze the contents of %pages and figure out where any
1752 # cousins of A::B are, then assume that. So, if A::B isn't found,
1753 # but A::C is found in lib/A/C.pm, then A::B is assumed to be in
1754 # lib/A/B.pm. This is also limited, but it's an improvement.
1755 # Maybe a hints file so that the links point to the correct places
1759 $link = "$htmlroot/$page.html";
1760 $link .= "#" . anchorify( $section ) if ($section);
1761 } elsif (!defined $pages{$page}) {
1764 $section = anchorify( $section ) if $section ne "";
1765 ### print STDERR "...section=$section\n";
1767 # if there is a directory by the name of the page, then assume that an
1768 # appropriate section will exist in the subdirectory
1769 # if ($section ne "" && $pages{$page} =~ /([^:]*[^(\.pod|\.pm)]):/) {
1770 if ($section ne "" && $pages{$page} =~ /([^:]*(?<!\.pod)(?<!\.pm)):/) {
1771 $link = "$htmlroot/$1/$section.html";
1772 ### print STDERR "...link=$link\n";
1774 # since there is no directory by the name of the page, the section will
1775 # have to exist within a .html of the same name. thus, make sure there
1776 # is a .pod or .pm that might become that .html
1778 $section = "#$section" if $section;
1779 ### print STDERR "...section=$section\n";
1781 # check if there is a .pod with the page name
1782 if ($pages{$page} =~ /([^:]*)\.pod:/) {
1783 $link = "$htmlroot/$1.html$section";
1784 } elsif ($pages{$page} =~ /([^:]*)\.pm:/) {
1785 $link = "$htmlroot/$1.html$section";
1793 # Here, we take advantage of the knowledge that $htmlfileurl ne ''
1794 # implies $htmlroot eq ''. This means that the link in question
1795 # needs a prefix of $htmldir if it begins with '/'. The test for
1796 # the initial '/' is done to avoid '#'-only links, and to allow
1797 # for other kinds of links, like file:, ftp:, etc.
1799 if ( $htmlfileurl ne '' ) {
1800 $link = "$htmldir$link" if $link =~ m{^/}s;
1801 $url = relativize_url( $link, $htmlfileurl );
1802 # print( " b: [$link,$htmlfileurl,$url]\n" );
1815 # relativize_url - convert an absolute URL to one relative to a base URL.
1816 # Assumes both end in a filename.
1818 sub relativize_url {
1819 my ($dest,$source) = @_ ;
1821 my ($dest_volume,$dest_directory,$dest_file) =
1822 File::Spec::Unix->splitpath( $dest ) ;
1823 $dest = File::Spec::Unix->catpath( $dest_volume, $dest_directory, '' ) ;
1825 my ($source_volume,$source_directory,$source_file) =
1826 File::Spec::Unix->splitpath( $source ) ;
1827 $source = File::Spec::Unix->catpath( $source_volume, $source_directory, '' ) ;
1830 if ( $dest ne '' ) {
1831 $rel_path = File::Spec::Unix->abs2rel( $dest, $source ) ;
1834 if ( $rel_path ne '' &&
1835 substr( $rel_path, -1 ) ne '/' &&
1836 substr( $dest_file, 0, 1 ) ne '#'
1838 $rel_path .= "/$dest_file" ;
1841 $rel_path .= "$dest_file" ;
1849 # coderef - make URL from the text of a C<>
1852 my( $page, $item ) = @_;
1855 my $fid = fragment_id( $item );
1856 if( defined( $page ) ){
1857 # we have been given a $page...
1860 # Do we take it? Item could be a section!
1861 my $base = $items{$fid} || "";
1862 $base =~ s{[^/]*/}{};
1863 if( $base ne "$page.html" ){
1864 ### print STDERR "coderef( $page, $item ): items{$fid} = $items{$fid} = $base => discard page!\n";
1869 # no page - local items precede cached items
1870 if( defined( $fid ) ){
1871 if( exists $local_items{$fid} ){
1872 $page = $local_items{$fid};
1874 $page = $items{$fid};
1879 # if there was a pod file that we found earlier with an appropriate
1880 # =item directive, then create a link to that page.
1881 if( defined $page ){
1883 if( exists $pages{$page} and $pages{$page} =~ /([^:.]*)\.[^:]*:/){
1884 $page = $1 . '.html';
1886 my $link = "$htmlroot/$page#item_$fid";
1888 # Here, we take advantage of the knowledge that $htmlfileurl
1889 # ne '' implies $htmlroot eq ''.
1890 if ( $htmlfileurl ne '' ) {
1891 $link = "$htmldir$link" ;
1892 $url = relativize_url( $link, $htmlfileurl ) ;
1897 $url = "#item_" . $fid;
1900 confess "url has space: $url" if $url =~ /"[^"]*\s[^"]*"/;
1902 return( $url, $fid );
1908 # Adapted from Nick Ing-Simmons' PodToHtml package.
1910 my $source_file = shift ;
1911 my $destination_file = shift;
1913 my $source = URI::file->new_abs($source_file);
1914 my $uo = URI::file->new($destination_file,$source)->abs;
1915 return $uo->rel->as_string;
1920 # finish_list - finish off any pending HTML lists. this should be called
1921 # after the entire pod file has been read and converted.
1924 while ($listlevel > 0) {
1925 print HTML "</dl>\n";
1931 # htmlify - converts a pod section specification to a suitable section
1932 # specification for HTML. Note that we keep spaces and special characters
1933 # except ", ? (Netscape problem) and the hyphen (writer's problem...).
1937 $heading =~ s/(\s+)/ /g;
1938 $heading =~ s/\s+\Z//;
1939 $heading =~ s/\A\s+//;
1940 # The hyphen is a disgrace to the English language.
1941 $heading =~ s/[-"?]//g;
1942 $heading = lc( $heading );
1947 # similar to htmlify, but turns spaces into underscores
1951 $anchor = htmlify($anchor);
1952 $anchor =~ s/\s/_/g; # fixup spaces left by htmlify
1957 # depod - convert text by eliminating all interior sequences
1958 # Note: can be called with copy or modify semantics
1965 $E2c{amp} = '&'; # in Tk's pods
1973 ${$_[0]} = depod1( \$string );
1981 my( $rstr, $func, $closing ) = @_;
1983 return $res unless defined $$rstr;
1984 if( ! defined( $func ) ){
1985 # skip to next begin of an interior sequence
1986 while( $$rstr =~ s/\A(.*?)([BCEFILSXZ])<(<+[^\S\n]+)?// ){
1987 # recurse into its text
1988 $res .= $1 . depod1( $rstr, $2, closing $3);
1991 } elsif( $func eq 'E' ){
1992 # E<x> - convert to character
1993 $$rstr =~ s/^([^>]*)>//;
1994 $res .= $E2c{$1} || "";
1995 } elsif( $func eq 'X' ){
1997 $$rstr =~ s/^[^>]*>//;
1998 } elsif( $func eq 'Z' ){
2002 # all others: either recurse into new function or
2003 # terminate at closing angle bracket
2004 my $term = pattern $closing;
2005 while( $$rstr =~ s/\A(.*?)(([BCEFILSXZ])<(<+[^\S\n]+)?|$term)// ){
2008 $res .= depod1( $rstr, $3, closing $4 );
2010 ## If we're here and $2 ne '>': undelimited interior sequence.
2011 ## Ignored, as this is called without proper indication of where we are.
2012 ## Rely on process_text to produce diagnostics.
2018 # fragment_id - construct a fragment identifier from:
2020 # b) contents of C<...>
2025 $text =~ s/\s+\Z//s;
2027 # a method or function?
2028 return $1 if $text =~ /(\w+)\s*\(/;
2029 return $1 if $text =~ /->\s*(\w+)\s*\(?/;
2032 return $1 if $text =~ /^([$@%*]\S+)/;
2034 # some pattern matching operator?
2035 return $1 if $text =~ m|^(\w+/).*/\w*$|;
2037 # fancy stuff... like "do { }"
2038 return $1 if $text =~ m|^(\w+)\s*{.*}$|;
2040 # honour the perlfunc manpage: func [PAR[,[ ]PAR]...]
2041 # and some funnies with ... Module ...
2042 return $1 if $text =~ m{^([a-z\d]+)(\s+[A-Z\d,/& ]+)?$};
2043 return $1 if $text =~ m{^([a-z\d]+)\s+Module(\s+[A-Z\d,/& ]+)?$};
2046 $text =~ s/\s+/_/sg;
2048 defined( $hc[ord($1)] ) ? $hc[ord($1)]
2049 : ( $hc[ord($1)] = sprintf( "%%%02X", ord($1) ) ) }gxe;
2050 $text = substr( $text, 0, 50 );
2057 # make_URL_href - generate HTML href from URL
2058 # Special treatment for CGI queries.
2060 sub make_URL_href($){
2063 s{^(http:[-\w/#~:.+=&%@!]+)(\?.*)$}{<a href="$1$2">$1</a>}i ){
2064 $url = "<a href=\"$url\">$url</a>";