3 # This file should really be a extracted from a .PL
5 use lib 'lib'; # use source library if present
7 use Config; # for config options in the makefile
8 use Getopt::Long; # for command-line parsing
16 installhtml - converts a collection of POD pages to HTML format.
20 installhtml [--help] [--podpath=<name>:...:<name>] [--podroot=<name>]
21 [--htmldir=<name>] [--htmlroot=<name>] [--norecurse] [--recurse]
22 [--splithead=<name>,...,<name>] [--splititem=<name>,...,<name>]
23 [--libpods=<name>,...,<name>] [--verbose]
27 I<installhtml> converts a collection of POD pages to a corresponding
28 collection of HTML pages. This is primarily used to convert the pod
29 pages found in the perl distribution.
39 =item B<--podroot> POD search path base directory
41 The base directory to search for all .pod and .pm files to be converted.
42 Default is current directory.
44 =item B<--podpath> POD search path
46 The list of directories to search for .pod and .pm files to be converted.
47 Default is `podroot/.'.
49 =item B<--recurse> recurse on subdirectories
51 Whether or not to convert all .pm and .pod files found in subdirectories
52 too. Default is to not recurse.
54 =item B<--htmldir> HTML destination directory
56 The base directory which all HTML files will be written to. This should
57 be a path relative to the filesystem, not the resulting URL.
59 =item B<--htmlroot> URL base directory
61 The base directory which all resulting HTML files will be visible at in
62 a URL. The default is `/'.
64 =item B<--splithead> POD files to split on =head directive
66 Comma-separated list of pod files to split by the =head directive. The
67 .pod suffix is optional. These files should have names specified
70 =item B<--splititem> POD files to split on =item directive
72 Comma-separated list of all pod files to split by the =item directive.
73 The .pod suffix is optional. I<installhtml> does not do the actual
74 split, rather it invokes I<splitpod> to do the dirty work. As with
75 --splithead, these files should have names specified relative to podroot.
77 =item B<--splitpod> Directory containing the splitpod program
79 The directory containing the splitpod program. The default is `podroot/pod'.
81 =item B<--libpods> library PODs for LE<lt>E<gt> links
83 Comma-separated list of "library" pod files. This is the same list that
84 will be passed to pod2html when any pod is converted.
86 =item B<--verbose> verbose output
94 The following command-line is an example of the one we use to convert
97 ./installhtml --podpath=lib:ext:pod:vms \
98 --podroot=/usr/src/perl \
99 --htmldir=/perl/nmanual \
100 --htmlroot=/perl/nmanual \
101 --splithead=pod/perlipc \
102 --splititem=pod/perlfunc \
103 --libpods=perlfunc,perlguts,perlvar,perlrun,perlop \
109 Chris Hall E<lt>hallc@cs.colorado.eduE<gt>
115 $usage =<<END_OF_USAGE;
116 Usage: $0 --help --podpath=<name>:...:<name> --podroot=<name>
117 --htmldir=<name> --htmlroot=<name> --norecurse --recurse
118 --splithead=<name>,...,<name> --splititem=<name>,...,<name>
119 --libpods=<name>,...,<name> --verbose
121 --help - this message
122 --podpath - colon-separated list of directories containing .pod and
123 .pm files to be converted (. by default).
124 --podroot - filesystem base directory from which all relative paths in
125 podpath stem (default is .).
126 --htmldir - directory to store resulting html files in relative
127 to the filesystem (\$podroot/html by default).
128 --htmlroot - http-server base directory from which all relative paths
129 in podpath stem (default is /).
130 --libpods - comma-separated list of files to search for =item pod
131 directives in as targets of C<> and implicit links (empty
133 --norecurse - don't recurse on those subdirectories listed in podpath.
135 --recurse - recurse on those subdirectories listed in podpath
136 --splithead - comma-separated list of .pod or .pm files to split. will
137 split each file into several smaller files at every occurrence
138 of a pod =head[1-6] directive.
139 --splititem - comma-separated list of .pod or .pm files to split using
141 --splitpod - directory where the program splitpod can be found
142 (\$podroot/pod by default).
143 --verbose - self-explanatory.
148 @podpath = ( "." ); # colon-separated list of directories containing .pod
149 # and .pm files to be converted.
150 $podroot = "."; # assume the pods we want are here
151 $htmldir = ""; # nothing for now...
152 $htmlroot = "/"; # default value
153 $recurse = 0; # default behavior
154 @splithead = (); # don't split any files by default
155 @splititem = (); # don't split any files by default
156 $splitpod = ""; # nothing for now.
158 $verbose = 0; # whether or not to print debugging info
160 $pod2html = "pod/pod2html";
162 usage("") unless @ARGV;
164 # parse the command-line
165 $result = GetOptions( qw(
178 usage("invalid parameters") unless $result;
179 parse_command_line();
182 # set these variables to appropriate values if the user didn't specify
184 $htmldir = "$htmlroot/html" unless $htmldir;
185 $splitpod = "$podroot/pod" unless $splitpod;
188 # make sure that the destination directory exists
189 (mkdir($htmldir, 0755) ||
190 die "$0: cannot make directory $htmldir: $!\n") if ! -d $htmldir;
193 # the following array will eventually contain files that are to be
194 # ignored in the conversion process. these are files that have been
195 # process by splititem or splithead and should not be converted as a
200 # split pods. its important to do this before convert ANY pods because
201 # it may effect some of the links
202 @splitdirs = (); # files in these directories won't get an index
203 split_on_head($podroot, $htmldir, \@splitdirs, \@ignore, @splithead);
204 split_on_item($podroot, \@splitdirs, \@ignore, @splititem);
207 # convert the pod pages found in @poddirs
208 #warn "converting files\n" if $verbose;
209 #warn "\@ignore\t= @ignore\n" if $verbose;
210 foreach $dir (@podpath) {
211 installdir($dir, $recurse, $podroot, \@splitdirs, \@ignore);
215 # now go through and create master indices for each pod we split
216 foreach $dir (@splititem) {
217 print "creating index $htmldir/$dir.html\n" if $verbose;
218 create_index("$htmldir/$dir.html", "$htmldir/$dir");
221 foreach $dir (@splithead) {
222 $dir .= ".pod" unless $dir =~ /(\.pod|\.pm)$/;
223 # let pod2html create the file
224 runpod2html($dir, 1);
226 # now go through and truncate after the index
227 $dir =~ /^(.*?)(\.pod|\.pm)?$/sm;
228 $file = "$htmldir/$1";
229 print "creating index $file.html\n" if $verbose;
231 # read in everything until what would have been the first =head
232 # directive, patching the index as we go.
233 open(H, "<$file.html") ||
234 die "$0: error opening $file.html for input: $!\n";
239 $_ =~ s{HREF="#(.*)">}{
240 my $url = "$file/$1.html" ;
241 $url = Pod::Html::relativize_url( $url, "$file.html" )
242 if ( ! defined $opt_htmlroot || $opt_htmlroot eq '' ) ;
249 # now rewrite the file
250 open(H, ">$file.html") ||
251 die "$0: error opening $file.html for output: $!\n";
256 ##############################################################################
260 warn "$0: @_\n" if @_;
265 sub parse_command_line {
266 usage() if defined $opt_help;
267 $opt_help = ""; # make -w shut up
269 # list of directories
270 @podpath = split(":", $opt_podpath) if defined $opt_podpath;
273 @splithead = split(",", $opt_splithead) if defined $opt_splithead;
274 @splititem = split(",", $opt_splititem) if defined $opt_splititem;
275 @libpods = split(",", $opt_libpods) if defined $opt_libpods;
277 $htmldir = $opt_htmldir if defined $opt_htmldir;
278 $htmlroot = $opt_htmlroot if defined $opt_htmlroot;
279 $podroot = $opt_podroot if defined $opt_podroot;
280 $splitpod = $opt_splitpod if defined $opt_splitpod;
282 $recurse = $opt_recurse if defined $opt_recurse;
283 $verbose = $opt_verbose if defined $opt_verbose;
288 my($cwd, $path) = @_;
289 return "$cwd/$path" unless $path =~ m:/:;
290 # add cwd if path is not already an absolute path
291 $path = "$cwd/$path" if (substr($path,0,1) ne '/');
297 my($html, $dir) = @_;
298 my(@files, @filedata, @index, $file);
302 # get the list of .html files in this directory
303 opendir(DIR, $dir) ||
304 die "$0: error opening directory $dir for reading: $!\n";
305 @files = sort(grep(/\.html?$/, readdir(DIR)));
308 open(HTML, ">$html") ||
309 die "$0: error opening $html for output: $!\n";
311 # for each .html file in the directory, extract the index
312 # embedded in the file and throw it into the big index.
313 print HTML "<DL COMPACT>\n";
314 foreach $file (@files) {
317 open(IN, "<$dir/$file") ||
318 die "$0: error opening $dir/$file for input: $!\n";
322 # pull out the NAME section
323 ($name) = grep(/NAME=/, @filedata);
324 ($lcp1,$lcp2) = ($name =~ m,/H1>\s(\S+)\s[\s-]*(.*?)\s*$,sm);
325 if (defined $lcp1 and $lcp1 eq '<P>') { # Uninteresting. Try again.
326 ($lcp1,$lcp2) = ($name =~ m,/H1>\s<P>\s(\S+)\s[\s-]*(.*?)\s*$,sm);
328 my $url= "$dir/$file" ;
329 if ( ! defined $opt_htmlroot || $opt_htmlroot eq '' ) {
330 $url = Pod::Html::relativize_url( "$dir/$file", $html ) ;
333 print HTML qq(<A HREF="$url">);
334 print HTML "<DT>$lcp1</A><DD>$lcp2\n" if defined $lcp1;
335 # print HTML qq(<A HREF="$url">$lcp1</A><BR>\n") if defined $lcp1;
339 @index = grep(/<!-- INDEX BEGIN -->.*<!-- INDEX END -->/s,
342 s/<!-- INDEX BEGIN -->(\s*<!--)(.*)(-->\s*)<!-- INDEX END -->/$lcp2/s;
345 print HTML "$_\n<P><HR><P>\n";
348 print HTML "</DL>\n";
355 my($podroot, $htmldir, $splitdirs, $ignore, @splithead) = @_;
356 my($pod, $dirname, $filename);
358 # split the files specified in @splithead on =head[1-6] pod directives
359 print "splitting files by head.\n" if $verbose && $#splithead >= 0;
360 foreach $pod (@splithead) {
361 # figure out the directory name and filename
362 $pod =~ s,^([^/]*)$,/$1,;
363 $pod =~ m,(.*?)/(.*?)(\.pod)?$,;
365 $filename = "$2.pod";
367 # since we are splitting this file it shouldn't be converted.
368 push(@$ignore, "$podroot/$dirname/$filename");
371 splitpod("$podroot/$dirname/$filename", "$podroot/$dirname", $htmldir,
378 my($podroot, $splitdirs, $ignore, @splititem) = @_;
379 my($pwd, $dirname, $filename);
381 print "splitting files by item.\n" if $verbose && $#splititem >= 0;
383 my $splitter = absolute_path($pwd, "$splitpod/splitpod");
384 foreach $pod (@splititem) {
385 # figure out the directory to split into
386 $pod =~ s,^([^/]*)$,/$1,;
387 $pod =~ m,(.*?)/(.*?)(\.pod)?$,;
389 $filename = "$2.pod";
391 # since we are splitting this file it shouldn't be converted.
392 push(@$ignore, "$podroot/$dirname.pod");
395 push(@$splitdirs, "$podroot/$dirname");
396 if (! -d "$podroot/$dirname") {
397 mkdir("$podroot/$dirname", 0755) ||
398 die "$0: error creating directory $podroot/$dirname: $!\n";
400 chdir("$podroot/$dirname") ||
401 die "$0: error changing to directory $podroot/$dirname: $!\n";
402 die "$splitter not found. Use '-splitpod dir' option.\n"
404 system("perl", $splitter, "../$filename") &&
405 warn "$0: error running '$splitter ../$filename'"
406 ." from $podroot/$dirname";
413 # splitpod - splits a .pod file into several smaller .pod files
414 # where a new file is started each time a =head[1-6] pod directive
415 # is encountered in the input file.
418 my($pod, $poddir, $htmldir, $splitdirs) = @_;
419 my(@poddata, @filedata, @heads);
420 my($file, $i, $j, $prevsec, $section, $nextsec);
422 print "splitting $pod\n" if $verbose;
424 # read the file in paragraphs
426 open(SPLITIN, "<$pod") ||
427 die "$0: error opening $pod for input: $!\n";
428 @filedata = <SPLITIN>;
430 die "$0: error closing $pod: $!\n";
432 # restore the file internally by =head[1-6] sections
434 for ($i = 0, $j = -1; $i <= $#filedata; $i++) {
435 $j++ if ($filedata[$i] =~ /^\s*=head[1-6]/);
437 $poddata[$j] = "" unless defined $poddata[$j];
438 $poddata[$j] .= "\n$filedata[$i]" if $j >= 0;
442 # create list of =head[1-6] sections so that we can rewrite
443 # L<> links as necessary.
445 foreach $i (0..$#poddata) {
446 $heads{htmlize($1)} = 1 if $poddata[$i] =~ /=head[1-6]\s+(.*)/;
449 # create a directory of a similar name and store all the
451 $pod =~ s,.*/(.*),$1,; # get the last part of the name
454 push(@$splitdirs, "$poddir/$dir");
455 mkdir("$poddir/$dir", 0755) ||
456 die "$0: could not create directory $poddir/$dir: $!\n"
457 unless -d "$poddir/$dir";
459 $poddata[0] =~ /^\s*=head[1-6]\s+(.*)/;
463 # for each section of the file create a separate pod file
464 for ($i = 0; $i <= $#poddata; $i++) {
465 # determine the "prev" and "next" links
468 if ($i < $#poddata) {
469 $poddata[$i+1] =~ /^\s*=head[1-6]\s+(.*)/;
475 # determine an appropriate filename (this must correspond with
476 # what pod2html will try and guess)
477 # $poddata[$i] =~ /^\s*=head[1-6]\s+(.*)/;
478 $file = "$dir/" . htmlize($section) . ".pod";
480 # create the new .pod file
481 print "\tcreating $poddir/$file\n" if $verbose;
482 open(SPLITOUT, ">$poddir/$file") ||
483 die "$0: error opening $poddir/$file for output: $!\n";
484 $poddata[$i] =~ s,L<([^<>]*)>,
485 defined $heads{htmlize($1)} ? "L<$dir/$1>" : "L<$1>"
487 print SPLITOUT $poddata[$i]."\n\n";
488 print SPLITOUT "=over 4\n\n";
489 print SPLITOUT "=item *\n\nBack to L<$dir/\"$prevsec\">\n\n" if $prevsec;
490 print SPLITOUT "=item *\n\nForward to L<$dir/\"$nextsec\">\n\n" if $nextsec;
491 print SPLITOUT "=item *\n\nUp to L<$dir>\n\n";
492 print SPLITOUT "=back\n\n";
494 die "$0: error closing $poddir/$file: $!\n";
500 # installdir - takes care of converting the .pod and .pm files in the
501 # current directory to .html files and then installing those.
504 my($dir, $recurse, $podroot, $splitdirs, $ignore) = @_;
505 my(@dirlist, @podlist, @pmlist, $doindex);
507 @dirlist = (); # directories to recurse on
508 @podlist = (); # .pod files to install
509 @pmlist = (); # .pm files to install
511 # should files in this directory get an index?
512 $doindex = (grep($_ eq "$podroot/$dir", @$splitdirs) ? 0 : 1);
514 opendir(DIR, "$podroot/$dir")
515 || die "$0: error opening directory $podroot/$dir: $!\n";
517 # find the directories to recurse on
518 @dirlist = map { if ($^O eq 'VMS') {/^(.*)\.dir$/i; "$dir/$1";} else {"$dir/$_";}}
519 grep(-d "$podroot/$dir/$_" && !/^\.{1,2}/, readdir(DIR)) if $recurse;
522 # find all the .pod files within the directory
523 @podlist = map { /^(.*)\.pod$/; "$dir/$1" }
524 grep(! -d "$podroot/$dir/$_" && /\.pod$/, readdir(DIR));
527 # find all the .pm files within the directory
528 @pmlist = map { /^(.*)\.pm$/; "$dir/$1" }
529 grep(! -d "$podroot/$dir/$_" && /\.pm$/, readdir(DIR));
533 # recurse on all subdirectories we kept track of
534 foreach $dir (@dirlist) {
535 installdir($dir, $recurse, $podroot, $splitdirs, $ignore);
538 # install all the pods we found
539 foreach $pod (@podlist) {
540 # check if we should ignore it.
541 next if grep($_ eq "$podroot/$pod.pod", @$ignore);
543 # check if a .pm files exists too
544 if (grep($_ eq "$pod.pm", @pmlist)) {
545 print "$0: Warning both `$podroot/$pod.pod' and "
546 . "`$podroot/$pod.pm' exist, using pod\n";
547 push(@ignore, "$pod.pm");
549 runpod2html("$pod.pod", $doindex);
552 # install all the .pm files we found
553 foreach $pm (@pmlist) {
554 # check if we should ignore it.
555 next if grep($_ eq "$pm.pm", @ignore);
557 runpod2html("$pm.pm", $doindex);
563 # runpod2html - invokes pod2html to convert a .pod or .pm file to a .html
567 my($pod, $doindex) = @_;
568 my($html, $i, $dir, @dirs);
571 $html =~ s/\.(pod|pm)$/.html/g;
573 # make sure the destination directories exist
574 @dirs = split("/", $html);
576 for ($i = 0; $i < $#dirs; $i++) {
577 if (! -d "$dir$dirs[$i]") {
578 mkdir("$dir$dirs[$i]", 0755) ||
579 die "$0: error creating directory $dir$dirs[$i]: $!\n";
581 $dir .= "$dirs[$i]/";
585 print "$podroot/$pod => $htmldir/$html\n" if $verbose;
586 #system("./pod2html",
588 #Pod::Html'pod2html($pod2html,
589 "--htmldir=$htmldir",
590 "--htmlroot=$htmlroot",
591 "--podpath=".join(":", @podpath),
592 "--podroot=$podroot", "--netscape",
593 ($doindex ? "--index" : "--noindex"),
594 "--" . ($recurse ? "" : "no") . "recurse",
595 ($#libpods >= 0) ? "--libpods=" . join(":", @libpods) : "",
596 "--infile=$podroot/$pod", "--outfile=$htmldir/$html");
597 die "$0: error running $pod2html: $!\n" if $?;
600 sub htmlize { htmlify(0, @_) }