3 use lib 'lib'; # use source library if present
5 use Config; # for config options in the makefile
6 use Getopt::Long; # for command-line parsing
14 installhtml - converts a collection of POD pages to HTML format.
18 installhtml [--help] [--podpath=<name>:...:<name>] [--podroot=<name>]
19 [--htmldir=<name>] [--htmlroot=<name>] [--norecurse] [--recurse]
20 [--splithead=<name>,...,<name>] [--splititem=<name>,...,<name>]
21 [--libpods=<name>,...,<name>] [--verbose]
25 I<installhtml> converts a collection of POD pages to a corresponding
26 collection of HTML pages. This is primarily used to convert the pod
27 pages found in the perl distribution.
37 =item B<--podroot> POD search path base directory
39 The base directory to search for all .pod and .pm files to be converted.
40 Default is current directory.
42 =item B<--podpath> POD search path
44 The list of directories to search for .pod and .pm files to be converted.
45 Default is `podroot/.'.
47 =item B<--recurse> recurse on subdirectories
49 Whether or not to convert all .pm and .pod files found in subdirectories
50 too. Default is to not recurse.
52 =item B<--htmldir> HTML destination directory
54 The base directory which all HTML files will be written to. This should
55 be a path relative to the filesystem, not the resulting URL.
57 =item B<--htmlroot> URL base directory
59 The base directory which all resulting HTML files will be visible at in
60 a URL. The default is `/'.
62 =item B<--splithead> POD files to split on =head directive
64 Colon-separated list of pod files to split by the =head directive. The
65 .pod suffix is optional. These files should have names specified
68 =item B<--splititem> POD files to split on =item directive
70 Colon-separated list of all pod files to split by the =item directive.
71 The .pod suffix is optional. I<installhtml> does not do the actual
72 split, rather it invokes I<splitpod> to do the dirty work. As with
73 --splithead, these files should have names specified relative to podroot.
75 =item B<--splitpod> Directory containing the splitpod program
77 The directory containing the splitpod program. The default is `podroot/pod'.
79 =item B<--libpods> library PODs for LE<lt>E<gt> links
81 Colon-separated list of "library" pod files. This is the same list that
82 will be passed to pod2html when any pod is converted.
84 =item B<--verbose> verbose output
92 The following command-line is an example of the one we use to convert
95 ./installhtml --podpath=lib:ext:pod:vms \
96 --podroot=/usr/src/perl \
97 --htmldir=/perl/nmanual \
98 --htmlroot=/perl/nmanual \
99 --splithead=pod/perlipc \
100 --splititem=pod/perlfunc \
101 --libpods=perlfunc:perlguts:perlvar:perlrun:perlop \
107 Chris Hall E<lt>hallc@cs.colorado.eduE<gt>
113 $usage =<<END_OF_USAGE;
114 Usage: $0 --help --podpath=<name>:...:<name> --podroot=<name>
115 --htmldir=<name> --htmlroot=<name> --norecurse --recurse
116 --splithead=<name>,...,<name> --splititem=<name>,...,<name>
117 --libpods=<name>,...,<name> --verbose
119 --help - this message
120 --podpath - colon-separated list of directories containing .pod and
121 .pm files to be converted (. by default).
122 --podroot - filesystem base directory from which all relative paths in
123 podpath stem (default is .).
124 --htmldir - directory to store resulting html files in relative
125 to the filesystem (\$podroot/html by default).
126 --htmlroot - http-server base directory from which all relative paths
127 in podpath stem (default is /).
128 --libpods - comma-separated list of files to search for =item pod
129 directives in as targets of C<> and implicit links (empty
131 --norecurse - don't recurse on those subdirectories listed in podpath.
133 --recurse - recurse on those subdirectories listed in podpath
134 --splithead - comma-separated list of .pod or .pm files to split. will
135 split each file into several smaller files at every occurrence
136 of a pod =head[1-6] directive.
137 --splititem - comma-separated list of .pod or .pm files to split using
139 --splitpod - directory where the program splitpod can be found
140 (\$podroot/pod by default).
141 --verbose - self-explanatory.
146 @podpath = ( "." ); # colon-separated list of directories containing .pod
147 # and .pm files to be converted.
148 $podroot = "."; # assume the pods we want are here
149 $htmldir = ""; # nothing for now...
150 $htmlroot = "/"; # default value
151 $recurse = 0; # default behavior
152 @splithead = (); # don't split any files by default
153 @splititem = (); # don't split any files by default
154 $splitpod = ""; # nothing for now.
156 $verbose = 0; # whether or not to print debugging info
158 $pod2html = "pod/pod2html";
160 usage("") unless @ARGV;
162 # parse the command-line
163 $result = GetOptions( qw(
176 usage("invalid parameters") unless $result;
177 parse_command_line();
180 # set these variables to appropriate values if the user didn't specify
182 $htmldir = "$htmlroot/html" unless $htmldir;
183 $splitpod = "$podroot/pod" unless $splitpod;
186 # make sure that the destination directory exists
187 (mkdir($htmldir, 0755) ||
188 die "$0: cannot make directory $htmldir: $!\n") if ! -d $htmldir;
191 # the following array will eventually contain files that are to be
192 # ignored in the conversion process. these are files that have been
193 # process by splititem or splithead and should not be converted as a
198 # split pods. its important to do this before convert ANY pods because
199 # it may effect some of the links
200 @splitdirs = (); # files in these directories won't get an index
201 split_on_head($podroot, $htmldir, \@splitdirs, \@ignore, @splithead);
202 split_on_item($podroot, \@splitdirs, \@ignore, @splititem);
205 # convert the pod pages found in @poddirs
206 #warn "converting files\n" if $verbose;
207 #warn "\@ignore\t= @ignore\n" if $verbose;
208 foreach $dir (@podpath) {
209 installdir($dir, $recurse, $podroot, \@splitdirs, \@ignore);
213 # now go through and create master indices for each pod we split
214 foreach $dir (@splititem) {
215 print "creating index $htmldir/$dir.html\n" if $verbose;
216 create_index("$htmldir/$dir.html", "$htmldir/$dir");
219 foreach $dir (@splithead) {
220 $dir .= ".pod" unless $dir =~ /(\.pod|\.pm)$/;
221 # let pod2html create the file
222 runpod2html($dir, 1);
224 # now go through and truncate after the index
225 $dir =~ /^(.*?)(\.pod|\.pm)?$/sm;
226 $file = "$htmldir/$1";
227 print "creating index $file.html\n" if $verbose;
229 # read in everything until what would have been the first =head
230 # directive, patching the index as we go.
231 open(H, "<$file.html") ||
232 die "$0: error opening $file.html for input: $!\n";
237 s,HREF="#(.*)">,HREF="$file/$1.html">,g;
242 # now rewrite the file
243 open(H, ">$file.html") ||
244 die "$0: error opening $file.html for output: $!\n";
249 ##############################################################################
253 warn "$0: @_\n" if @_;
258 sub parse_command_line {
259 usage() if defined $opt_help;
260 $opt_help = ""; # make -w shut up
262 # list of directories
263 @podpath = split(":", $opt_podpath) if defined $opt_podpath;
266 @splithead = split(",", $opt_splithead) if defined $opt_splithead;
267 @splititem = split(",", $opt_splititem) if defined $opt_splititem;
268 @libpods = split(",", $opt_libpods) if defined $opt_libpods;
270 $htmldir = $opt_htmldir if defined $opt_htmldir;
271 $htmlroot = $opt_htmlroot if defined $opt_htmlroot;
272 $podroot = $opt_podroot if defined $opt_podroot;
273 $splitpod = $opt_splitpod if defined $opt_splitpod;
275 $recurse = $opt_recurse if defined $opt_recurse;
276 $verbose = $opt_verbose if defined $opt_verbose;
281 my($cwd, $path) = @_;
282 return "$cwd/$path" unless $path =~ m:/:;
283 # add cwd if path is not already an absolute path
284 $path = "$cwd/$path" if (substr($path,0,1) ne '/');
290 my($html, $dir) = @_;
291 my(@files, @filedata, @index, $file);
293 # get the list of .html files in this directory
294 opendir(DIR, $dir) ||
295 die "$0: error opening directory $dir for reading: $!\n";
296 @files = sort(grep(/\.html$/, readdir(DIR)));
299 open(HTML, ">$html") ||
300 die "$0: error opening $html for output: $!\n";
302 # for each .html file in the directory, extract the index
303 # embedded in the file and throw it into the big index.
304 print HTML "<DL COMPACT>\n";
305 foreach $file (@files) {
308 open(IN, "<$dir/$file") ||
309 die "$0: error opening $dir/$file for input: $!\n";
313 # pull out the NAME section
314 ($name) = grep(/NAME=/, @filedata);
315 $name =~ m,/H1>\s(\S+)\s[\s-]*(.*?)\s*$,sm;
316 print HTML qq(<A HREF="$dir/$file">);
317 print HTML "<DT>$1</A><DD>$2\n" if defined $1;
318 # print HTML qq(<A HREF="$dir/$file">$1</A><BR>\n") if defined $1;
322 @index = grep(/<!-- INDEX BEGIN -->.*<!-- INDEX END -->/s,
325 s/<!-- INDEX BEGIN -->(\s*<!--)(.*)(-->\s*)<!-- INDEX END -->/$2/s;
328 print HTML "$_\n<P><HR><P>\n";
331 print HTML "</DL>\n";
338 my($podroot, $htmldir, $splitdirs, $ignore, @splithead) = @_;
339 my($pod, $dirname, $filename);
341 # split the files specified in @splithead on =head[1-6] pod directives
342 print "splitting files by head.\n" if $verbose && $#splithead >= 0;
343 foreach $pod (@splithead) {
344 # figure out the directory name and filename
345 $pod =~ s,^([^/]*)$,/$1,;
346 $pod =~ m,(.*?)/(.*?)(\.pod)?$,;
348 $filename = "$2.pod";
350 # since we are splitting this file it shouldn't be converted.
351 push(@$ignore, "$podroot/$dirname/$filename");
354 splitpod("$podroot/$dirname/$filename", "$podroot/$dirname", $htmldir,
361 my($podroot, $splitdirs, $ignore, @splititem) = @_;
362 my($pwd, $dirname, $filename);
364 print "splitting files by item.\n" if $verbose && $#splititem >= 0;
366 my $splitter = absolute_path($pwd, "$splitpod/splitpod");
367 foreach $pod (@splititem) {
368 # figure out the directory to split into
369 $pod =~ s,^([^/]*)$,/$1,;
370 $pod =~ m,(.*?)/(.*?)(\.pod)?$,;
372 $filename = "$2.pod";
374 # since we are splitting this file it shouldn't be converted.
375 push(@$ignore, "$podroot/$dirname.pod");
378 push(@$splitdirs, "$podroot/$dirname");
379 if (! -d "$podroot/$dirname") {
380 mkdir("$podroot/$dirname", 0755) ||
381 die "$0: error creating directory $podroot/$dirname: $!\n";
383 chdir("$podroot/$dirname") ||
384 die "$0: error changing to directory $podroot/$dirname: $!\n";
385 die "$splitter not found. Use '-splitpod dir' option.\n"
387 system("perl", $splitter, "../$filename") &&
388 warn "$0: error running '$splitter ../$filename'"
389 ." from $podroot/$dirname";
396 # splitpod - splits a .pod file into several smaller .pod files
397 # where a new file is started each time a =head[1-6] pod directive
398 # is encountered in the input file.
401 my($pod, $poddir, $htmldir, $splitdirs) = @_;
402 my(@poddata, @filedata, @heads);
403 my($file, $i, $j, $prevsec, $section, $nextsec);
405 print "splitting $pod\n" if $verbose;
407 # read the file in paragraphs
409 open(SPLITIN, "<$pod") ||
410 die "$0: error opening $pod for input: $!\n";
411 @filedata = <SPLITIN>;
413 die "$0: error closing $pod: $!\n";
415 # restore the file internally by =head[1-6] sections
417 for ($i = 0, $j = -1; $i <= $#filedata; $i++) {
418 $j++ if ($filedata[$i] =~ /^\s*=head[1-6]/);
420 $poddata[$j] = "" unless defined $poddata[$j];
421 $poddata[$j] .= "\n$filedata[$i]" if $j >= 0;
425 # create list of =head[1-6] sections so that we can rewrite
426 # L<> links as necessary.
428 foreach $i (0..$#poddata) {
429 $heads{htmlize($1)} = 1 if $poddata[$i] =~ /=head[1-6]\s+(.*)/;
432 # create a directory of a similar name and store all the
434 $pod =~ s,.*/(.*),$1,; # get the last part of the name
437 push(@$splitdirs, "$poddir/$dir");
438 mkdir("$poddir/$dir", 0755) ||
439 die "$0: could not create directory $poddir/$dir: $!\n"
440 unless -d "$poddir/$dir";
442 $poddata[0] =~ /^\s*=head[1-6]\s+(.*)/;
446 # for each section of the file create a separate pod file
447 for ($i = 0; $i <= $#poddata; $i++) {
448 # determine the "prev" and "next" links
451 if ($i < $#poddata) {
452 $poddata[$i+1] =~ /^\s*=head[1-6]\s+(.*)/;
458 # determine an appropriate filename (this must correspond with
459 # what pod2html will try and guess)
460 # $poddata[$i] =~ /^\s*=head[1-6]\s+(.*)/;
461 $file = "$dir/" . htmlize($section) . ".pod";
463 # create the new .pod file
464 print "\tcreating $poddir/$file\n" if $verbose;
465 open(SPLITOUT, ">$poddir/$file") ||
466 die "$0: error opening $poddir/$file for output: $!\n";
467 $poddata[$i] =~ s,L<([^<>]*)>,
468 defined $heads{htmlize($1)} ? "L<$dir/$1>" : "L<$1>"
470 print SPLITOUT $poddata[$i]."\n\n";
471 print SPLITOUT "=over 4\n\n";
472 print SPLITOUT "=item *\n\nBack to L<$dir/\"$prevsec\">\n\n" if $prevsec;
473 print SPLITOUT "=item *\n\nForward to L<$dir/\"$nextsec\">\n\n" if $nextsec;
474 print SPLITOUT "=item *\n\nUp to L<$dir>\n\n";
475 print SPLITOUT "=back\n\n";
477 die "$0: error closing $poddir/$file: $!\n";
483 # installdir - takes care of converting the .pod and .pm files in the
484 # current directory to .html files and then installing those.
487 my($dir, $recurse, $podroot, $splitdirs, $ignore) = @_;
488 my(@dirlist, @podlist, @pmlist, $doindex);
490 @dirlist = (); # directories to recurse on
491 @podlist = (); # .pod files to install
492 @pmlist = (); # .pm files to install
494 # should files in this directory get an index?
495 $doindex = (grep($_ eq "$podroot/$dir", @$splitdirs) ? 0 : 1);
497 opendir(DIR, "$podroot/$dir")
498 || die "$0: error opening directory $podroot/$dir: $!\n";
500 # find the directories to recurse on
501 @dirlist = map { "$dir/$_" }
502 grep(-d "$podroot/$dir/$_" && !/^\.{1,2}/, readdir(DIR)) if $recurse;
505 # find all the .pod files within the directory
506 @podlist = map { /^(.*)\.pod$/; "$dir/$1" }
507 grep(! -d "$podroot/$dir/$_" && /\.pod$/, readdir(DIR));
510 # find all the .pm files within the directory
511 @pmlist = map { /^(.*)\.pm$/; "$dir/$1" }
512 grep(! -d "$podroot/$dir/$_" && /\.pm$/, readdir(DIR));
516 # recurse on all subdirectories we kept track of
517 foreach $dir (@dirlist) {
518 installdir($dir, $recurse, $podroot, $splitdirs, $ignore);
521 # install all the pods we found
522 foreach $pod (@podlist) {
523 # check if we should ignore it.
524 next if grep($_ eq "$podroot/$pod.pod", @$ignore);
526 # check if a .pm files exists too
527 if (grep($_ eq "$pod.pm", @pmlist)) {
528 print "$0: Warning both `$podroot/$pod.pod' and "
529 . "`$podroot/$pod.pm' exist, using pod\n";
530 push(@ignore, "$pod.pm");
532 runpod2html("$pod.pod", $doindex);
535 # install all the .pm files we found
536 foreach $pm (@pmlist) {
537 # check if we should ignore it.
538 next if grep($_ eq "$pm.pm", @ignore);
540 runpod2html("$pm.pm", $doindex);
546 # runpod2html - invokes pod2html to convert a .pod or .pm file to a .html
550 my($pod, $doindex) = @_;
551 my($html, $i, $dir, @dirs);
554 $html =~ s/\.(pod|pm)$/.html/g;
556 # make sure the destination directories exist
557 @dirs = split("/", $html);
559 for ($i = 0; $i < $#dirs; $i++) {
560 if (! -d "$dir$dirs[$i]") {
561 mkdir("$dir$dirs[$i]", 0755) ||
562 die "$0: error creating directory $dir$dirs[$i]: $!\n";
564 $dir .= "$dirs[$i]/";
568 print "$podroot/$pod => $htmldir/$html\n" if $verbose;
569 #system("./pod2html",
571 #Pod::Html'pod2html($pod2html,
572 "--htmlroot=$htmlroot",
573 "--podpath=".join(":", @podpath),
574 "--podroot=$podroot", "--netscape",
575 ($doindex ? "--index" : "--noindex"),
576 "--" . ($recurse ? "" : "no") . "recurse",
577 ($#libpods >= 0) ? "--libpods=" . join(":", @libpods) : "",
578 "--infile=$podroot/$pod", "--outfile=$htmldir/$html");
579 die "$0: error running $pod2html: $!\n" if $?;
582 sub htmlize { htmlify(0, @_) }