This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
don't mess with the umask()
[perl5.git] / installhtml
CommitLineData
90248788
TB
1#!./perl -w
2
3# This file should really be a extracted from a .PL
54310121 4
3e3baf6d
TB
5use lib 'lib'; # use source library if present
6
54310121
PP
7use Config; # for config options in the makefile
8use Getopt::Long; # for command-line parsing
9use Cwd;
10use Pod::Html;
11
54310121
PP
12=head1 NAME
13
14installhtml - converts a collection of POD pages to HTML format.
15
16=head1 SYNOPSIS
17
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]
22
23=head1 DESCRIPTION
24
25I<installhtml> converts a collection of POD pages to a corresponding
26collection of HTML pages. This is primarily used to convert the pod
27pages found in the perl distribution.
28
29=head1 OPTIONS
30
31=over 4
32
33=item B<--help> help
34
35Displays the usage.
36
3e3baf6d
TB
37=item B<--podroot> POD search path base directory
38
39The base directory to search for all .pod and .pm files to be converted.
40Default is current directory.
41
54310121
PP
42=item B<--podpath> POD search path
43
e05c04ea 44The list of directories to search for .pod and .pm files to be converted.
54310121
PP
45Default is `podroot/.'.
46
3e3baf6d 47=item B<--recurse> recurse on subdirectories
54310121 48
3e3baf6d
TB
49Whether or not to convert all .pm and .pod files found in subdirectories
50too. Default is to not recurse.
54310121
PP
51
52=item B<--htmldir> HTML destination directory
53
54The base directory which all HTML files will be written to. This should
55be a path relative to the filesystem, not the resulting URL.
56
57=item B<--htmlroot> URL base directory
58
59The base directory which all resulting HTML files will be visible at in
60a URL. The default is `/'.
61
54310121
PP
62=item B<--splithead> POD files to split on =head directive
63
af19f77d 64Comma-separated list of pod files to split by the =head directive. The
3e3baf6d
TB
65.pod suffix is optional. These files should have names specified
66relative to podroot.
54310121
PP
67
68=item B<--splititem> POD files to split on =item directive
69
af19f77d 70Comma-separated list of all pod files to split by the =item directive.
3e3baf6d
TB
71The .pod suffix is optional. I<installhtml> does not do the actual
72split, rather it invokes I<splitpod> to do the dirty work. As with
73--splithead, these files should have names specified relative to podroot.
74
75=item B<--splitpod> Directory containing the splitpod program
76
77The directory containing the splitpod program. The default is `podroot/pod'.
54310121
PP
78
79=item B<--libpods> library PODs for LE<lt>E<gt> links
80
af19f77d 81Comma-separated list of "library" pod files. This is the same list that
54310121
PP
82will be passed to pod2html when any pod is converted.
83
84=item B<--verbose> verbose output
85
86Self-explanatory.
87
88=back
89
90=head1 EXAMPLE
91
92The following command-line is an example of the one we use to convert
93perl documentation:
94
95 ./installhtml --podpath=lib:ext:pod:vms \
84902520
TB
96 --podroot=/usr/src/perl \
97 --htmldir=/perl/nmanual \
98 --htmlroot=/perl/nmanual \
99 --splithead=pod/perlipc \
100 --splititem=pod/perlfunc \
af19f77d 101 --libpods=perlfunc,perlguts,perlvar,perlrun,perlop \
84902520
TB
102 --recurse \
103 --verbose
54310121
PP
104
105=head1 AUTHOR
106
107Chris Hall E<lt>hallc@cs.colorado.eduE<gt>
108
109=head1 TODO
110
111=cut
112
113$usage =<<END_OF_USAGE;
114Usage: $0 --help --podpath=<name>:...:<name> --podroot=<name>
115 --htmldir=<name> --htmlroot=<name> --norecurse --recurse
116 --splithead=<name>,...,<name> --splititem=<name>,...,<name>
117 --libpods=<name>,...,<name> --verbose
118
119 --help - this message
120 --podpath - colon-separated list of directories containing .pod and
121 .pm files to be converted (. by default).
122 --podroot - filesystem base directory from which all relative paths in
123 podpath stem (default is .).
124 --htmldir - directory to store resulting html files in relative
125 to the filesystem (\$podroot/html by default).
126 --htmlroot - http-server base directory from which all relative paths
127 in podpath stem (default is /).
128 --libpods - comma-separated list of files to search for =item pod
129 directives in as targets of C<> and implicit links (empty
130 by default).
131 --norecurse - don't recurse on those subdirectories listed in podpath.
132 (default behavior).
133 --recurse - recurse on those subdirectories listed in podpath
134 --splithead - comma-separated list of .pod or .pm files to split. will
e05c04ea 135 split each file into several smaller files at every occurrence
54310121
PP
136 of a pod =head[1-6] directive.
137 --splititem - comma-separated list of .pod or .pm files to split using
138 splitpod.
3e3baf6d
TB
139 --splitpod - directory where the program splitpod can be found
140 (\$podroot/pod by default).
54310121
PP
141 --verbose - self-explanatory.
142
143END_OF_USAGE
144
145@libpods = ();
146@podpath = ( "." ); # colon-separated list of directories containing .pod
147 # and .pm files to be converted.
148$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.
155
156$verbose = 0; # whether or not to print debugging info
157
158$pod2html = "pod/pod2html";
159
3e3baf6d 160usage("") unless @ARGV;
54310121
PP
161
162# parse the command-line
163$result = GetOptions( qw(
164 help
165 podpath=s
166 podroot=s
167 htmldir=s
168 htmlroot=s
169 libpods=s
170 recurse!
171 splithead=s
172 splititem=s
173 splitpod=s
174 verbose
175));
176usage("invalid parameters") unless $result;
177parse_command_line();
178
179
180# set these variables to appropriate values if the user didn't specify
181# values for them.
182$htmldir = "$htmlroot/html" unless $htmldir;
183$splitpod = "$podroot/pod" unless $splitpod;
184
185
186# make sure that the destination directory exists
187(mkdir($htmldir, 0755) ||
188 die "$0: cannot make directory $htmldir: $!\n") if ! -d $htmldir;
189
190
191# the following array will eventually contain files that are to be
192# ignored in the conversion process. these are files that have been
193# process by splititem or splithead and should not be converted as a
194# result.
195@ignore = ();
196
197
198# split pods. its important to do this before convert ANY pods because
199# it may effect some of the links
200@splitdirs = (); # files in these directories won't get an index
201split_on_head($podroot, $htmldir, \@splitdirs, \@ignore, @splithead);
3e3baf6d 202split_on_item($podroot, \@splitdirs, \@ignore, @splititem);
54310121
PP
203
204
205# convert the pod pages found in @poddirs
206#warn "converting files\n" if $verbose;
207#warn "\@ignore\t= @ignore\n" if $verbose;
208foreach $dir (@podpath) {
209 installdir($dir, $recurse, $podroot, \@splitdirs, \@ignore);
210}
211
212
213# now go through and create master indices for each pod we split
214foreach $dir (@splititem) {
215 print "creating index $htmldir/$dir.html\n" if $verbose;
216 create_index("$htmldir/$dir.html", "$htmldir/$dir");
217}
218
219foreach $dir (@splithead) {
220 $dir .= ".pod" unless $dir =~ /(\.pod|\.pm)$/;
221 # let pod2html create the file
222 runpod2html($dir, 1);
223
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 $verbose;
228
229 # read in everything until what would have been the first =head
230 # directive, patching the index as we go.
231 open(H, "<$file.html") ||
232 die "$0: error opening $file.html for input: $!\n";
233 $/ = "";
234 @data = ();
235 while (<H>) {
236 last if /NAME=/;
29f227c9
BS
237 $_ =~ s{HREF="#(.*)">}{
238 my $url = "$file/$1.html" ;
239 $url = Pod::Html::relativize_url( $url, "$file.html" )
240 if ( ! defined $opt_htmlroot || $opt_htmlroot eq '' ) ;
241 "HREF=\"$url\">" ;
242 }eg;
54310121
PP
243 push @data, $_;
244 }
245 close(H);
246
247 # now rewrite the file
248 open(H, ">$file.html") ||
249 die "$0: error opening $file.html for output: $!\n";
250 print H "@data\n";
251 close(H);
252}
253
254##############################################################################
255
256
257sub usage {
258 warn "$0: @_\n" if @_;
259 die $usage;
260}
261
262
263sub parse_command_line {
264 usage() if defined $opt_help;
265 $opt_help = ""; # make -w shut up
266
267 # list of directories
268 @podpath = split(":", $opt_podpath) if defined $opt_podpath;
269
270 # lists of files
271 @splithead = split(",", $opt_splithead) if defined $opt_splithead;
272 @splititem = split(",", $opt_splititem) if defined $opt_splititem;
273 @libpods = split(",", $opt_libpods) if defined $opt_libpods;
274
275 $htmldir = $opt_htmldir if defined $opt_htmldir;
276 $htmlroot = $opt_htmlroot if defined $opt_htmlroot;
277 $podroot = $opt_podroot if defined $opt_podroot;
278 $splitpod = $opt_splitpod if defined $opt_splitpod;
279
280 $recurse = $opt_recurse if defined $opt_recurse;
281 $verbose = $opt_verbose if defined $opt_verbose;
282}
283
284
3e3baf6d
TB
285sub absolute_path {
286 my($cwd, $path) = @_;
287 return "$cwd/$path" unless $path =~ m:/:;
288 # add cwd if path is not already an absolute path
289 $path = "$cwd/$path" if (substr($path,0,1) ne '/');
290 return $path;
291}
292
293
54310121
PP
294sub create_index {
295 my($html, $dir) = @_;
296 my(@files, @filedata, @index, $file);
af19f77d
LP
297 my($lcp1,$lcp2);
298
54310121
PP
299
300 # get the list of .html files in this directory
301 opendir(DIR, $dir) ||
302 die "$0: error opening directory $dir for reading: $!\n";
39e571d4 303 @files = sort(grep(/\.html?$/, readdir(DIR)));
54310121
PP
304 closedir(DIR);
305
306 open(HTML, ">$html") ||
307 die "$0: error opening $html for output: $!\n";
308
309 # for each .html file in the directory, extract the index
310 # embedded in the file and throw it into the big index.
311 print HTML "<DL COMPACT>\n";
312 foreach $file (@files) {
313 $/ = "";
314
315 open(IN, "<$dir/$file") ||
316 die "$0: error opening $dir/$file for input: $!\n";
317 @filedata = <IN>;
318 close(IN);
319
320 # pull out the NAME section
321 ($name) = grep(/NAME=/, @filedata);
af19f77d
LP
322 ($lcp1,$lcp2) = ($name =~ m,/H1>\s(\S+)\s[\s-]*(.*?)\s*$,sm);
323 if (defined $lcp1 and $lcp1 eq '<P>') { # Uninteresting. Try again.
324 ($lcp1,$lcp2) = ($name =~ m,/H1>\s<P>\s(\S+)\s[\s-]*(.*?)\s*$,sm);
325 }
29f227c9
BS
326 my $url= "$dir/$file" ;
327 if ( ! defined $opt_htmlroot || $opt_htmlroot eq '' ) {
328 $url = Pod::Html::relativize_url( "$dir/$file", $html ) ;
329 }
330
331 print HTML qq(<A HREF="$url">);
af19f77d 332 print HTML "<DT>$lcp1</A><DD>$lcp2\n" if defined $lcp1;
29f227c9 333# print HTML qq(<A HREF="$url">$lcp1</A><BR>\n") if defined $lcp1;
54310121
PP
334
335 next;
336
337 @index = grep(/<!-- INDEX BEGIN -->.*<!-- INDEX END -->/s,
338 @filedata);
339 for (@index) {
af19f77d 340 s/<!-- INDEX BEGIN -->(\s*<!--)(.*)(-->\s*)<!-- INDEX END -->/$lcp2/s;
54310121
PP
341 s,#,$dir/$file#,g;
342 # print HTML "$_\n";
343 print HTML "$_\n<P><HR><P>\n";
344 }
345 }
346 print HTML "</DL>\n";
347
348 close(HTML);
349}
350
351
352sub split_on_head {
353 my($podroot, $htmldir, $splitdirs, $ignore, @splithead) = @_;
354 my($pod, $dirname, $filename);
355
356 # split the files specified in @splithead on =head[1-6] pod directives
357 print "splitting files by head.\n" if $verbose && $#splithead >= 0;
358 foreach $pod (@splithead) {
359 # figure out the directory name and filename
360 $pod =~ s,^([^/]*)$,/$1,;
361 $pod =~ m,(.*?)/(.*?)(\.pod)?$,;
362 $dirname = $1;
363 $filename = "$2.pod";
364
365 # since we are splitting this file it shouldn't be converted.
366 push(@$ignore, "$podroot/$dirname/$filename");
367
368 # split the pod
369 splitpod("$podroot/$dirname/$filename", "$podroot/$dirname", $htmldir,
370 $splitdirs);
371 }
372}
373
374
375sub split_on_item {
376 my($podroot, $splitdirs, $ignore, @splititem) = @_;
377 my($pwd, $dirname, $filename);
378
379 print "splitting files by item.\n" if $verbose && $#splititem >= 0;
380 $pwd = getcwd();
3e3baf6d 381 my $splitter = absolute_path($pwd, "$splitpod/splitpod");
54310121
PP
382 foreach $pod (@splititem) {
383 # figure out the directory to split into
384 $pod =~ s,^([^/]*)$,/$1,;
385 $pod =~ m,(.*?)/(.*?)(\.pod)?$,;
386 $dirname = "$1/$2";
387 $filename = "$2.pod";
388
389 # since we are splitting this file it shouldn't be converted.
390 push(@$ignore, "$podroot/$dirname.pod");
391
392 # split the pod
393 push(@$splitdirs, "$podroot/$dirname");
394 if (! -d "$podroot/$dirname") {
395 mkdir("$podroot/$dirname", 0755) ||
396 die "$0: error creating directory $podroot/$dirname: $!\n";
397 }
398 chdir("$podroot/$dirname") ||
399 die "$0: error changing to directory $podroot/$dirname: $!\n";
3e3baf6d
TB
400 die "$splitter not found. Use '-splitpod dir' option.\n"
401 unless -f $splitter;
402 system("perl", $splitter, "../$filename") &&
403 warn "$0: error running '$splitter ../$filename'"
e05c04ea 404 ." from $podroot/$dirname";
54310121
PP
405 }
406 chdir($pwd);
407}
408
409
410#
411# splitpod - splits a .pod file into several smaller .pod files
412# where a new file is started each time a =head[1-6] pod directive
413# is encountered in the input file.
414#
415sub splitpod {
416 my($pod, $poddir, $htmldir, $splitdirs) = @_;
417 my(@poddata, @filedata, @heads);
418 my($file, $i, $j, $prevsec, $section, $nextsec);
419
420 print "splitting $pod\n" if $verbose;
421
422 # read the file in paragraphs
423 $/ = "";
424 open(SPLITIN, "<$pod") ||
425 die "$0: error opening $pod for input: $!\n";
426 @filedata = <SPLITIN>;
427 close(SPLITIN) ||
428 die "$0: error closing $pod: $!\n";
429
430 # restore the file internally by =head[1-6] sections
431 @poddata = ();
432 for ($i = 0, $j = -1; $i <= $#filedata; $i++) {
433 $j++ if ($filedata[$i] =~ /^\s*=head[1-6]/);
434 if ($j >= 0) {
435 $poddata[$j] = "" unless defined $poddata[$j];
436 $poddata[$j] .= "\n$filedata[$i]" if $j >= 0;
437 }
438 }
439
440 # create list of =head[1-6] sections so that we can rewrite
441 # L<> links as necessary.
442 %heads = ();
443 foreach $i (0..$#poddata) {
444 $heads{htmlize($1)} = 1 if $poddata[$i] =~ /=head[1-6]\s+(.*)/;
445 }
446
447 # create a directory of a similar name and store all the
448 # files in there
449 $pod =~ s,.*/(.*),$1,; # get the last part of the name
450 $dir = $pod;
451 $dir =~ s/\.pod//g;
452 push(@$splitdirs, "$poddir/$dir");
453 mkdir("$poddir/$dir", 0755) ||
454 die "$0: could not create directory $poddir/$dir: $!\n"
455 unless -d "$poddir/$dir";
456
457 $poddata[0] =~ /^\s*=head[1-6]\s+(.*)/;
458 $section = "";
459 $nextsec = $1;
460
461 # for each section of the file create a separate pod file
462 for ($i = 0; $i <= $#poddata; $i++) {
463 # determine the "prev" and "next" links
464 $prevsec = $section;
465 $section = $nextsec;
466 if ($i < $#poddata) {
467 $poddata[$i+1] =~ /^\s*=head[1-6]\s+(.*)/;
468 $nextsec = $1;
469 } else {
470 $nextsec = "";
471 }
472
473 # determine an appropriate filename (this must correspond with
474 # what pod2html will try and guess)
475 # $poddata[$i] =~ /^\s*=head[1-6]\s+(.*)/;
476 $file = "$dir/" . htmlize($section) . ".pod";
477
478 # create the new .pod file
479 print "\tcreating $poddir/$file\n" if $verbose;
480 open(SPLITOUT, ">$poddir/$file") ||
481 die "$0: error opening $poddir/$file for output: $!\n";
482 $poddata[$i] =~ s,L<([^<>]*)>,
483 defined $heads{htmlize($1)} ? "L<$dir/$1>" : "L<$1>"
484 ,ge;
485 print SPLITOUT $poddata[$i]."\n\n";
486 print SPLITOUT "=over 4\n\n";
487 print SPLITOUT "=item *\n\nBack to L<$dir/\"$prevsec\">\n\n" if $prevsec;
488 print SPLITOUT "=item *\n\nForward to L<$dir/\"$nextsec\">\n\n" if $nextsec;
489 print SPLITOUT "=item *\n\nUp to L<$dir>\n\n";
490 print SPLITOUT "=back\n\n";
491 close(SPLITOUT) ||
492 die "$0: error closing $poddir/$file: $!\n";
493 }
494}
495
496
497#
498# installdir - takes care of converting the .pod and .pm files in the
499# current directory to .html files and then installing those.
500#
501sub installdir {
502 my($dir, $recurse, $podroot, $splitdirs, $ignore) = @_;
503 my(@dirlist, @podlist, @pmlist, $doindex);
504
505 @dirlist = (); # directories to recurse on
506 @podlist = (); # .pod files to install
507 @pmlist = (); # .pm files to install
508
509 # should files in this directory get an index?
510 $doindex = (grep($_ eq "$podroot/$dir", @$splitdirs) ? 0 : 1);
511
512 opendir(DIR, "$podroot/$dir")
513 || die "$0: error opening directory $podroot/$dir: $!\n";
514
515 # find the directories to recurse on
fe4c6be1 516 @dirlist = map { if ($^O eq 'VMS') {/^(.*)\.dir$/i; "$dir/$1";} else {"$dir/$_";}}
54310121
PP
517 grep(-d "$podroot/$dir/$_" && !/^\.{1,2}/, readdir(DIR)) if $recurse;
518 rewinddir(DIR);
519
520 # find all the .pod files within the directory
521 @podlist = map { /^(.*)\.pod$/; "$dir/$1" }
522 grep(! -d "$podroot/$dir/$_" && /\.pod$/, readdir(DIR));
523 rewinddir(DIR);
524
525 # find all the .pm files within the directory
526 @pmlist = map { /^(.*)\.pm$/; "$dir/$1" }
527 grep(! -d "$podroot/$dir/$_" && /\.pm$/, readdir(DIR));
528
529 closedir(DIR);
530
531 # recurse on all subdirectories we kept track of
532 foreach $dir (@dirlist) {
533 installdir($dir, $recurse, $podroot, $splitdirs, $ignore);
534 }
535
536 # install all the pods we found
537 foreach $pod (@podlist) {
538 # check if we should ignore it.
539 next if grep($_ eq "$podroot/$pod.pod", @$ignore);
540
541 # check if a .pm files exists too
542 if (grep($_ eq "$pod.pm", @pmlist)) {
543 print "$0: Warning both `$podroot/$pod.pod' and "
544 . "`$podroot/$pod.pm' exist, using pod\n";
545 push(@ignore, "$pod.pm");
546 }
547 runpod2html("$pod.pod", $doindex);
548 }
549
550 # install all the .pm files we found
551 foreach $pm (@pmlist) {
552 # check if we should ignore it.
553 next if grep($_ eq "$pm.pm", @ignore);
554
555 runpod2html("$pm.pm", $doindex);
556 }
557}
558
559
560#
561# runpod2html - invokes pod2html to convert a .pod or .pm file to a .html
562# file.
563#
564sub runpod2html {
565 my($pod, $doindex) = @_;
566 my($html, $i, $dir, @dirs);
567
568 $html = $pod;
569 $html =~ s/\.(pod|pm)$/.html/g;
570
571 # make sure the destination directories exist
572 @dirs = split("/", $html);
573 $dir = "$htmldir/";
574 for ($i = 0; $i < $#dirs; $i++) {
575 if (! -d "$dir$dirs[$i]") {
576 mkdir("$dir$dirs[$i]", 0755) ||
577 die "$0: error creating directory $dir$dirs[$i]: $!\n";
578 }
579 $dir .= "$dirs[$i]/";
580 }
581
582 # invoke pod2html
583 print "$podroot/$pod => $htmldir/$html\n" if $verbose;
584#system("./pod2html",
585 Pod::Html'pod2html(
586 #Pod::Html'pod2html($pod2html,
5a039dd3 587 "--htmldir=$htmldir",
54310121
PP
588 "--htmlroot=$htmlroot",
589 "--podpath=".join(":", @podpath),
590 "--podroot=$podroot", "--netscape",
591 ($doindex ? "--index" : "--noindex"),
592 "--" . ($recurse ? "" : "no") . "recurse",
593 ($#libpods >= 0) ? "--libpods=" . join(":", @libpods) : "",
594 "--infile=$podroot/$pod", "--outfile=$htmldir/$html");
595 die "$0: error running $pod2html: $!\n" if $?;
596}
597
598sub htmlize { htmlify(0, @_) }