[perl5.git] / installhtml

#!./perl -w

# This file should really be a extracted from a .PL

use lib 'lib';		# use source library if present

use Config;		# for config options in the makefile
use Getopt::Long;	# for command-line parsing
use Cwd;
use Pod::Html;

umask 022;

=head1 NAME

installhtml - converts a collection of POD pages to HTML format.

=head1 SYNOPSIS

    installhtml  [--help] [--podpath=<name>:...:<name>] [--podroot=<name>]
         [--htmldir=<name>] [--htmlroot=<name>]  [--norecurse] [--recurse]
         [--splithead=<name>,...,<name>]   [--splititem=<name>,...,<name>]
	 [--libpods=<name>,...,<name>]  [--verbose]

=head1 DESCRIPTION

I<installhtml> converts a collection of POD pages to a corresponding
collection of HTML pages.  This is primarily used to convert the pod
pages found in the perl distribution.

=head1 OPTIONS

=over 4

=item B<--help> help

Displays the usage.

=item B<--podroot> POD search path base directory

The base directory to search for all .pod and .pm files to be converted.
Default is current directory.

=item B<--podpath> POD search path

The list of directories to search for .pod and .pm files to be converted.
Default is `podroot/.'.

=item B<--recurse> recurse on subdirectories

Whether or not to convert all .pm and .pod files found in subdirectories
too.  Default is to not recurse.

=item B<--htmldir> HTML destination directory

The base directory which all HTML files will be written to.  This should
be a path relative to the filesystem, not the resulting URL.

=item B<--htmlroot> URL base directory

The base directory which all resulting HTML files will be visible at in
a URL.  The default is `/'.

=item B<--splithead> POD files to split on =head directive

Colon-separated list of pod files to split by the =head directive.  The
.pod suffix is optional. These files should have names specified
relative to podroot.

=item B<--splititem> POD files to split on =item directive

Colon-separated list of all pod files to split by the =item directive.
The .pod suffix is optional.  I<installhtml> does not do the actual
split, rather it invokes I<splitpod> to do the dirty work.  As with
--splithead, these files should have names specified relative to podroot.

=item B<--splitpod> Directory containing the splitpod program

The directory containing the splitpod program. The default is `podroot/pod'.

=item B<--libpods> library PODs for LE<lt>E<gt> links

Colon-separated list of "library" pod files.  This is the same list that
will be passed to pod2html when any pod is converted.

=item B<--verbose> verbose output

Self-explanatory.

=back

=head1 EXAMPLE

The following command-line is an example of the one we use to convert
perl documentation:

    ./installhtml --podpath=lib:ext:pod:vms   \
			--podroot=/usr/src/perl     \
			--htmldir=/perl/nmanual     \
			--htmlroot=/perl/nmanual    \
			--splithead=pod/perlipc     \
			--splititem=pod/perlfunc    \
			--libpods=perlfunc:perlguts:perlvar:perlrun:perlop \
			--recurse \
			--verbose

=head1 AUTHOR

Chris Hall E<lt>hallc@cs.colorado.eduE<gt>

=head1 TODO

=cut

$usage =<<END_OF_USAGE;
Usage: $0 --help --podpath=<name>:...:<name> --podroot=<name>
         --htmldir=<name> --htmlroot=<name> --norecurse --recurse
         --splithead=<name>,...,<name> --splititem=<name>,...,<name>
	 --libpods=<name>,...,<name> --verbose

    --help      - this message
    --podpath   - colon-separated list of directories containing .pod and
                  .pm files to be converted (. by default).
    --podroot   - filesystem base directory from which all relative paths in
                  podpath stem (default is .).
    --htmldir   - directory to store resulting html files in relative
                  to the filesystem (\$podroot/html by default). 
    --htmlroot  - http-server base directory from which all relative paths
                  in podpath stem (default is /).
    --libpods   - comma-separated list of files to search for =item pod
                  directives in as targets of C<> and implicit links (empty
                  by default).
    --norecurse - don't recurse on those subdirectories listed in podpath.
                  (default behavior).
    --recurse   - recurse on those subdirectories listed in podpath
    --splithead - comma-separated list of .pod or .pm files to split.  will
                  split each file into several smaller files at every occurrence
                  of a pod =head[1-6] directive.
    --splititem - comma-separated list of .pod or .pm files to split using
                  splitpod.
    --splitpod  - directory where the program splitpod can be found
                  (\$podroot/pod by default).
    --verbose   - self-explanatory.

END_OF_USAGE

@libpods = ();
@podpath = ( "." );	# colon-separated list of directories containing .pod
			# and .pm files to be converted.
$podroot = ".";		# assume the pods we want are here
$htmldir = "";		# nothing for now...
$htmlroot = "/";	# default value
$recurse = 0;		# default behavior
@splithead = ();	# don't split any files by default
@splititem = ();	# don't split any files by default
$splitpod = "";		# nothing for now.

$verbose = 0; 	    	# whether or not to print debugging info

$pod2html = "pod/pod2html";

usage("") unless @ARGV;

# parse the command-line
$result = GetOptions( qw(
	help
	podpath=s
	podroot=s
	htmldir=s
	htmlroot=s
	libpods=s
	recurse!
	splithead=s
	splititem=s
	splitpod=s
	verbose
));
usage("invalid parameters") unless $result;
parse_command_line();


# set these variables to appropriate values if the user didn't specify
#  values for them.
$htmldir = "$htmlroot/html" unless $htmldir;
$splitpod = "$podroot/pod" unless $splitpod;


# make sure that the destination directory exists
(mkdir($htmldir, 0755) ||
	die "$0: cannot make directory $htmldir: $!\n") if ! -d $htmldir;


# the following array will eventually contain files that are to be
# ignored in the conversion process.  these are files that have been
# process by splititem or splithead and should not be converted as a
# result.
@ignore = ();


# split pods.  its important to do this before convert ANY pods because
#  it may effect some of the links
@splitdirs = ();    # files in these directories won't get an index
split_on_head($podroot, $htmldir, \@splitdirs, \@ignore, @splithead);
split_on_item($podroot,           \@splitdirs, \@ignore, @splititem);


# convert the pod pages found in @poddirs
#warn "converting files\n" if $verbose;
#warn "\@ignore\t= @ignore\n" if $verbose;
foreach $dir (@podpath) {
    installdir($dir, $recurse, $podroot, \@splitdirs, \@ignore);
}


# now go through and create master indices for each pod we split
foreach $dir (@splititem) {
    print "creating index $htmldir/$dir.html\n" if $verbose;
    create_index("$htmldir/$dir.html", "$htmldir/$dir");
}

foreach $dir (@splithead) {
    $dir .= ".pod" unless $dir =~ /(\.pod|\.pm)$/;
    # let pod2html create the file
    runpod2html($dir, 1);

    # now go through and truncate after the index
    $dir =~ /^(.*?)(\.pod|\.pm)?$/sm;
    $file = "$htmldir/$1";
    print "creating index $file.html\n" if $verbose;

    # read in everything until what would have been the first =head
    # directive, patching the index as we go.
    open(H, "<$file.html") ||
	die "$0: error opening $file.html for input: $!\n";
    $/ = "";
    @data = ();
    while (<H>) {
	last if /NAME=/;
	s,HREF="#(.*)">,HREF="$file/$1.html">,g;
	push @data, $_;
    } 
    close(H);

    # now rewrite the file 
    open(H, ">$file.html") ||
	die "$0: error opening $file.html for output: $!\n";
    print H "@data\n";
    close(H);
}

##############################################################################


sub usage {
    warn "$0: @_\n" if @_;
    die $usage;
}


sub parse_command_line {
    usage() if defined $opt_help;
    $opt_help = ""; 	    	    # make -w shut up

    # list of directories
    @podpath   = split(":", $opt_podpath) if defined $opt_podpath;

    # lists of files
    @splithead = split(",", $opt_splithead) if defined $opt_splithead;
    @splititem = split(",", $opt_splititem) if defined $opt_splititem;
    @libpods   = split(",", $opt_libpods) if defined $opt_libpods;

    $htmldir  = $opt_htmldir	    if defined $opt_htmldir;
    $htmlroot = $opt_htmlroot	    if defined $opt_htmlroot;
    $podroot  = $opt_podroot	    if defined $opt_podroot;
    $splitpod = $opt_splitpod	    if defined $opt_splitpod;

    $recurse  = $opt_recurse	    if defined $opt_recurse;
    $verbose  = $opt_verbose	    if defined $opt_verbose;
}


sub absolute_path {
    my($cwd, $path) = @_;
	return "$cwd/$path" unless $path =~ m:/:;
    # add cwd if path is not already an absolute path
    $path = "$cwd/$path" if (substr($path,0,1) ne '/');
    return $path;
}


sub create_index {
    my($html, $dir) = @_;
    my(@files, @filedata, @index, $file);

    # get the list of .html files in this directory
    opendir(DIR, $dir) ||
	die "$0: error opening directory $dir for reading: $!\n";
    @files = sort(grep(/\.html?$/, readdir(DIR)));
    closedir(DIR);

    open(HTML, ">$html") ||
	die "$0: error opening $html for output: $!\n";

    # for each .html file in the directory, extract the index
    #	embedded in the file and throw it into the big index.
    print HTML "<DL COMPACT>\n";
    foreach $file (@files) {
	$/ = "";

	open(IN, "<$dir/$file") ||
	    die "$0: error opening $dir/$file for input: $!\n";
	@filedata = <IN>;
	close(IN);

	# pull out the NAME section
	($name) = grep(/NAME=/, @filedata);
	$name =~ m,/H1>\s(\S+)\s[\s-]*(.*?)\s*$,sm;
	print HTML qq(<A HREF="$dir/$file">);
	print HTML "<DT>$1</A><DD>$2\n" if defined $1;
#	print HTML qq(<A HREF="$dir/$file">$1</A><BR>\n") if defined $1;

	next;

	@index = grep(/<!-- INDEX BEGIN -->.*<!-- INDEX END -->/s,
		    @filedata);
	for (@index) {
	    s/<!-- INDEX BEGIN -->(\s*<!--)(.*)(-->\s*)<!-- INDEX END -->/$2/s;
	    s,#,$dir/$file#,g;
	    # print HTML "$_\n";
	    print HTML "$_\n<P><HR><P>\n";
	}
    }
    print HTML "</DL>\n";

    close(HTML);
}


sub split_on_head {
    my($podroot, $htmldir, $splitdirs, $ignore, @splithead) = @_;
    my($pod, $dirname, $filename);

    # split the files specified in @splithead on =head[1-6] pod directives
    print "splitting files by head.\n" if $verbose && $#splithead >= 0;
    foreach $pod (@splithead) {
	# figure out the directory name and filename
	$pod      =~ s,^([^/]*)$,/$1,;
	$pod      =~ m,(.*?)/(.*?)(\.pod)?$,;
	$dirname  = $1;
	$filename = "$2.pod";

	# since we are splitting this file it shouldn't be converted.
	push(@$ignore, "$podroot/$dirname/$filename");

	# split the pod
	splitpod("$podroot/$dirname/$filename", "$podroot/$dirname", $htmldir,
	    $splitdirs);
    }
}


sub split_on_item {
    my($podroot, $splitdirs, $ignore, @splititem) = @_;
    my($pwd, $dirname, $filename);

    print "splitting files by item.\n" if $verbose && $#splititem >= 0;
    $pwd = getcwd();
	my $splitter = absolute_path($pwd, "$splitpod/splitpod");
    foreach $pod (@splititem) {
	# figure out the directory to split into
	$pod      =~ s,^([^/]*)$,/$1,;
	$pod      =~ m,(.*?)/(.*?)(\.pod)?$,;
	$dirname  = "$1/$2";
	$filename = "$2.pod";

	# since we are splitting this file it shouldn't be converted.
	push(@$ignore, "$podroot/$dirname.pod");

	# split the pod
	push(@$splitdirs, "$podroot/$dirname");
	if (! -d "$podroot/$dirname") {
	    mkdir("$podroot/$dirname", 0755) ||
		    die "$0: error creating directory $podroot/$dirname: $!\n";
	}
	chdir("$podroot/$dirname") ||
	    die "$0: error changing to directory $podroot/$dirname: $!\n";
	die "$splitter not found. Use '-splitpod dir' option.\n"
	    unless -f $splitter;
	system("perl", $splitter, "../$filename") &&
	    warn "$0: error running '$splitter ../$filename'"
		 ." from $podroot/$dirname";
    }
    chdir($pwd);
}


#
# splitpod - splits a .pod file into several smaller .pod files
#  where a new file is started each time a =head[1-6] pod directive
#  is encountered in the input file.
#
sub splitpod {
    my($pod, $poddir, $htmldir, $splitdirs) = @_;
    my(@poddata, @filedata, @heads);
    my($file, $i, $j, $prevsec, $section, $nextsec);

    print "splitting $pod\n" if $verbose;

    # read the file in paragraphs
    $/ = "";
    open(SPLITIN, "<$pod") ||
	die "$0: error opening $pod for input: $!\n";
    @filedata = <SPLITIN>;
    close(SPLITIN) ||
	die "$0: error closing $pod: $!\n";

    # restore the file internally by =head[1-6] sections
    @poddata = ();
    for ($i = 0, $j = -1; $i <= $#filedata; $i++) {
	$j++ if ($filedata[$i] =~ /^\s*=head[1-6]/);
	if ($j >= 0) { 
	    $poddata[$j]  = "" unless defined $poddata[$j];
	    $poddata[$j] .= "\n$filedata[$i]" if $j >= 0;
	}
    }

    # create list of =head[1-6] sections so that we can rewrite
    #  L<> links as necessary.
    %heads = ();
    foreach $i (0..$#poddata) {
	$heads{htmlize($1)} = 1 if $poddata[$i] =~ /=head[1-6]\s+(.*)/;
    }

    # create a directory of a similar name and store all the
    #  files in there
    $pod =~ s,.*/(.*),$1,;	# get the last part of the name
    $dir = $pod;
    $dir =~ s/\.pod//g;
    push(@$splitdirs, "$poddir/$dir");
    mkdir("$poddir/$dir", 0755) ||
	die "$0: could not create directory $poddir/$dir: $!\n"
	unless -d "$poddir/$dir";

    $poddata[0] =~ /^\s*=head[1-6]\s+(.*)/;
    $section    = "";
    $nextsec    = $1;

    # for each section of the file create a separate pod file
    for ($i = 0; $i <= $#poddata; $i++) {
	# determine the "prev" and "next" links
	$prevsec = $section;
	$section = $nextsec;
	if ($i < $#poddata) {
	    $poddata[$i+1] =~ /^\s*=head[1-6]\s+(.*)/;
	    $nextsec       = $1;
	} else {
	    $nextsec = "";
	}

	# determine an appropriate filename (this must correspond with
	#  what pod2html will try and guess)
	# $poddata[$i] =~ /^\s*=head[1-6]\s+(.*)/;
	$file = "$dir/" . htmlize($section) . ".pod";

	# create the new .pod file
	print "\tcreating $poddir/$file\n" if $verbose;
	open(SPLITOUT, ">$poddir/$file") ||
	    die "$0: error opening $poddir/$file for output: $!\n";
	$poddata[$i] =~ s,L<([^<>]*)>,
			    defined $heads{htmlize($1)} ? "L<$dir/$1>" : "L<$1>"
			 ,ge;
	print SPLITOUT $poddata[$i]."\n\n";
	print SPLITOUT "=over 4\n\n";
	print SPLITOUT "=item *\n\nBack to L<$dir/\"$prevsec\">\n\n" if $prevsec;
	print SPLITOUT "=item *\n\nForward to L<$dir/\"$nextsec\">\n\n" if $nextsec;
	print SPLITOUT "=item *\n\nUp to L<$dir>\n\n";
	print SPLITOUT "=back\n\n";
	close(SPLITOUT) ||
	    die "$0: error closing $poddir/$file: $!\n";
    }
}


#
# installdir - takes care of converting the .pod and .pm files in the
#  current directory to .html files and then installing those.
#
sub installdir {
    my($dir, $recurse, $podroot, $splitdirs, $ignore) = @_;
    my(@dirlist, @podlist, @pmlist, $doindex);

    @dirlist = ();	# directories to recurse on
    @podlist = ();	# .pod files to install
    @pmlist  = ();	# .pm files to install

    # should files in this directory get an index?
    $doindex = (grep($_ eq "$podroot/$dir", @$splitdirs) ? 0 : 1);

    opendir(DIR, "$podroot/$dir")
	|| die "$0: error opening directory $podroot/$dir: $!\n";

    # find the directories to recurse on
    @dirlist = map { "$dir/$_" }
	grep(-d "$podroot/$dir/$_" && !/^\.{1,2}/, readdir(DIR)) if $recurse;
    rewinddir(DIR);

    # find all the .pod files within the directory
    @podlist = map { /^(.*)\.pod$/; "$dir/$1" }
	grep(! -d "$podroot/$dir/$_" && /\.pod$/, readdir(DIR));
    rewinddir(DIR);

    # find all the .pm files within the directory
    @pmlist = map { /^(.*)\.pm$/; "$dir/$1" }
	grep(! -d "$podroot/$dir/$_" && /\.pm$/, readdir(DIR));

    closedir(DIR);

    # recurse on all subdirectories we kept track of
    foreach $dir (@dirlist) {
	installdir($dir, $recurse, $podroot, $splitdirs, $ignore);
    }

    # install all the pods we found
    foreach $pod (@podlist) {
	# check if we should ignore it.
	next if grep($_ eq "$podroot/$pod.pod", @$ignore);

	# check if a .pm files exists too
	if (grep($_ eq "$pod.pm", @pmlist)) {
	    print  "$0: Warning both `$podroot/$pod.pod' and "
		. "`$podroot/$pod.pm' exist, using pod\n";
	    push(@ignore, "$pod.pm");
	}
	runpod2html("$pod.pod", $doindex);
    }

    # install all the .pm files we found
    foreach $pm (@pmlist) {
	# check if we should ignore it.
	next if grep($_ eq "$pm.pm", @ignore);

	runpod2html("$pm.pm", $doindex);
    }
}


#
# runpod2html - invokes pod2html to convert a .pod or .pm file to a .html
#  file.
#
sub runpod2html {
    my($pod, $doindex) = @_;
    my($html, $i, $dir, @dirs);

    $html = $pod;
    $html =~ s/\.(pod|pm)$/.html/g;

    # make sure the destination directories exist
    @dirs = split("/", $html);
    $dir  = "$htmldir/";
    for ($i = 0; $i < $#dirs; $i++) {
	if (! -d "$dir$dirs[$i]") {
	    mkdir("$dir$dirs[$i]", 0755) ||
		die "$0: error creating directory $dir$dirs[$i]: $!\n";
	}
	$dir .= "$dirs[$i]/";
    }

    # invoke pod2html
    print "$podroot/$pod => $htmldir/$html\n" if $verbose;
#system("./pod2html",
        Pod::Html'pod2html(
        #Pod::Html'pod2html($pod2html,
	"--htmlroot=$htmlroot",
	"--podpath=".join(":", @podpath),
	"--podroot=$podroot", "--netscape",
	($doindex ? "--index" : "--noindex"),
	"--" . ($recurse ? "" : "no") . "recurse",
	($#libpods >= 0) ? "--libpods=" . join(":", @libpods) : "",
	"--infile=$podroot/$pod", "--outfile=$htmldir/$html");
    die "$0: error running $pod2html: $!\n" if $?;
}

sub htmlize { htmlify(0, @_) }
Commit	Line	Data
90248788 TB	1	#!./perl -w
	2
	3	# This file should really be a extracted from a .PL
54310121	4
3e3baf6d TB	5	use lib 'lib'; # use source library if present
3e3baf6d TB	6
54310121	7	use Config; # for config options in the makefile
	8	use Getopt::Long; # for command-line parsing
	9	use Cwd;
	10	use Pod::Html;
	11
	12	umask 022;
	13
	14	=head1 NAME
	15
	16	installhtml - converts a collection of POD pages to HTML format.
	17
	18	=head1 SYNOPSIS
	19
	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]
	24
	25	=head1 DESCRIPTION
	26
	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.
	30
	31	=head1 OPTIONS
	32
	33	=over 4
	34
	35	=item B<--help> help
	36
	37	Displays the usage.
	38
3e3baf6d TB	39	=item B<--podroot> POD search path base directory
	40
	41	The base directory to search for all .pod and .pm files to be converted.
	42	Default is current directory.
	43
54310121	44	=item B<--podpath> POD search path
54310121	45
e05c04ea	46	The list of directories to search for .pod and .pm files to be converted.
54310121	47	Default is `podroot/.'.
54310121	48
3e3baf6d	49	=item B<--recurse> recurse on subdirectories
54310121	50
3e3baf6d TB	51	Whether or not to convert all .pm and .pod files found in subdirectories
3e3baf6d TB	52	too. Default is to not recurse.
54310121	53
	54	=item B<--htmldir> HTML destination directory
	55
	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.
	58
	59	=item B<--htmlroot> URL base directory
	60
	61	The base directory which all resulting HTML files will be visible at in
	62	a URL. The default is `/'.
	63
54310121	64	=item B<--splithead> POD files to split on =head directive
54310121	65
3e3baf6d TB	66	Colon-separated list of pod files to split by the =head directive. The
	67	.pod suffix is optional. These files should have names specified
	68	relative to podroot.
54310121	69
	70	=item B<--splititem> POD files to split on =item directive
	71
e05c04ea	72	Colon-separated list of all pod files to split by the =item directive.
3e3baf6d TB	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.
	76
	77	=item B<--splitpod> Directory containing the splitpod program
	78
	79	The directory containing the splitpod program. The default is `podroot/pod'.
54310121	80
	81	=item B<--libpods> library PODs for LE<lt>E<gt> links
	82
e05c04ea	83	Colon-separated list of "library" pod files. This is the same list that
54310121	84	will be passed to pod2html when any pod is converted.
	85
	86	=item B<--verbose> verbose output
	87
	88	Self-explanatory.
	89
	90	=back
	91
	92	=head1 EXAMPLE
	93
	94	The following command-line is an example of the one we use to convert
	95	perl documentation:
	96
	97	./installhtml --podpath=lib:ext:pod:vms \
84902520 TB	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 \
	104	--recurse \
	105	--verbose
54310121	106
	107	=head1 AUTHOR
	108
	109	Chris Hall E<lt>hallc@cs.colorado.eduE<gt>
	110
	111	=head1 TODO
	112
	113	=cut
	114
	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
	120
	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
	132	by default).
	133	--norecurse - don't recurse on those subdirectories listed in podpath.
	134	(default behavior).
	135	--recurse - recurse on those subdirectories listed in podpath
	136	--splithead - comma-separated list of .pod or .pm files to split. will
e05c04ea	137	split each file into several smaller files at every occurrence
54310121	138	of a pod =head[1-6] directive.
	139	--splititem - comma-separated list of .pod or .pm files to split using
	140	splitpod.
3e3baf6d TB	141	--splitpod - directory where the program splitpod can be found
3e3baf6d TB	142	(\$podroot/pod by default).
54310121	143	--verbose - self-explanatory.
	144
	145	END_OF_USAGE
	146
	147	@libpods = ();
	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.
	157
	158	$verbose = 0; # whether or not to print debugging info
	159
	160	$pod2html = "pod/pod2html";
	161
3e3baf6d	162	usage("") unless @ARGV;
54310121	163
	164	# parse the command-line
	165	$result = GetOptions( qw(
	166	help
	167	podpath=s
	168	podroot=s
	169	htmldir=s
	170	htmlroot=s
	171	libpods=s
	172	recurse!
	173	splithead=s
	174	splititem=s
	175	splitpod=s
	176	verbose
	177	));
	178	usage("invalid parameters") unless $result;
	179	parse_command_line();
	180
	181
	182	# set these variables to appropriate values if the user didn't specify
	183	# values for them.
	184	$htmldir = "$htmlroot/html" unless $htmldir;
	185	$splitpod = "$podroot/pod" unless $splitpod;
	186
	187
	188	# make sure that the destination directory exists
	189	(mkdir($htmldir, 0755) \|\|
	190	die "$0: cannot make directory $htmldir: $!\n") if ! -d $htmldir;
	191
	192
	193	# the following array will eventually contain files that are to be
	194	# ignored in the conversion process. these are files that have been
	195	# process by splititem or splithead and should not be converted as a
	196	# result.
	197	@ignore = ();
	198
	199
	200	# split pods. its important to do this before convert ANY pods because
	201	# it may effect some of the links
	202	@splitdirs = (); # files in these directories won't get an index
	203	split_on_head($podroot, $htmldir, \@splitdirs, \@ignore, @splithead);
3e3baf6d	204	split_on_item($podroot, \@splitdirs, \@ignore, @splititem);
54310121	205
	206
	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);
	212	}
	213
	214
	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");
	219	}
	220
	221	foreach $dir (@splithead) {
	222	$dir .= ".pod" unless $dir =~ /(\.pod\|\.pm)$/;
	223	# let pod2html create the file
	224	runpod2html($dir, 1);
	225
	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 $verbose;
	230
	231	# read in everything until what would have been the first =head
	232	# directive, patching the index as we go.
	233	open(H, "<$file.html") \|\|
	234	die "$0: error opening $file.html for input: $!\n";
	235	$/ = "";
	236	@data = ();
	237	while (<H>) {
	238	last if /NAME=/;
	239	s,HREF="#(.*)">,HREF="$file/$1.html">,g;
	240	push @data, $_;
	241	}
	242	close(H);
	243
	244	# now rewrite the file
	245	open(H, ">$file.html") \|\|
	246	die "$0: error opening $file.html for output: $!\n";
	247	print H "@data\n";
	248	close(H);
	249	}
	250
	251	##############################################################################
	252
	253
	254	sub usage {
	255	warn "$0: @_\n" if @_;
	256	die $usage;
	257	}
	258
	259
	260	sub parse_command_line {
	261	usage() if defined $opt_help;
	262	$opt_help = ""; # make -w shut up
	263
	264	# list of directories
	265	@podpath = split(":", $opt_podpath) if defined $opt_podpath;
	266
	267	# lists of files
	268	@splithead = split(",", $opt_splithead) if defined $opt_splithead;
269	@splititem = split(",", $opt_splititem) if defined $opt_splititem;
270	@libpods = split(",", $opt_libpods) if defined $opt_libpods;
271
272	$htmldir = $opt_htmldir if defined $opt_htmldir;
273	$htmlroot = $opt_htmlroot if defined $opt_htmlroot;
274	$podroot = $opt_podroot if defined $opt_podroot;
275	$splitpod = $opt_splitpod if defined $opt_splitpod;
276
277	$recurse = $opt_recurse if defined $opt_recurse;
278	$verbose = $opt_verbose if defined $opt_verbose;
279	}
280
281
3e3baf6d TB	282	sub absolute_path {
	283	my($cwd, $path) = @_;
	284	return "$cwd/$path" unless $path =~ m:/:;
	285	# add cwd if path is not already an absolute path
	286	$path = "$cwd/$path" if (substr($path,0,1) ne '/');
	287	return $path;
	288	}
	289
	290
54310121	291	sub create_index {
	292	my($html, $dir) = @_;
	293	my(@files, @filedata, @index, $file);
	294
	295	# get the list of .html files in this directory
	296	opendir(DIR, $dir) \|\|
	297	die "$0: error opening directory $dir for reading: $!\n";
39e571d4	298	@files = sort(grep(/\.html?$/, readdir(DIR)));
54310121	299	closedir(DIR);
	300
	301	open(HTML, ">$html") \|\|
	302	die "$0: error opening $html for output: $!\n";
	303
	304	# for each .html file in the directory, extract the index
	305	# embedded in the file and throw it into the big index.
	306	print HTML "<DL COMPACT>\n";
	307	foreach $file (@files) {
	308	$/ = "";
	309
	310	open(IN, "<$dir/$file") \|\|
	311	die "$0: error opening $dir/$file for input: $!\n";
	312	@filedata = <IN>;
	313	close(IN);
	314
	315	# pull out the NAME section
	316	($name) = grep(/NAME=/, @filedata);
	317	$name =~ m,/H1>\s(\S+)\s[\s-](.?)\s*$,sm;
	318	print HTML qq(<A HREF="$dir/$file">);
	319	print HTML "<DT>$1</A><DD>$2\n" if defined $1;
	320	# print HTML qq(<A HREF="$dir/$file">$1</A><BR>\n") if defined $1;
	321
	322	next;
	323
	324	@index = grep(/<!-- INDEX BEGIN -->.*<!-- INDEX END -->/s,
	325	@filedata);
	326	for (@index) {
	327	s/<!-- INDEX BEGIN -->(\s<!--)(.)(-->\s*)<!-- INDEX END -->/$2/s;
	328	s,#,$dir/$file#,g;
	329	# print HTML "$_\n";
	330	print HTML "$_\n<P><HR><P>\n";
	331	}
	332	}
	333	print HTML "</DL>\n";
	334
	335	close(HTML);
	336	}
	337
	338
	339	sub split_on_head {
	340	my($podroot, $htmldir, $splitdirs, $ignore, @splithead) = @_;
	341	my($pod, $dirname, $filename);
	342
	343	# split the files specified in @splithead on =head[1-6] pod directives
	344	print "splitting files by head.\n" if $verbose && $#splithead >= 0;
	345	foreach $pod (@splithead) {
	346	# figure out the directory name and filename
	347	$pod =~ s,^([^/]*)$,/$1,;
	348	$pod =~ m,(.?)/(.?)(\.pod)?$,;
	349	$dirname = $1;
	350	$filename = "$2.pod";
	351
	352	# since we are splitting this file it shouldn't be converted.
	353	push(@$ignore, "$podroot/$dirname/$filename");
	354
	355	# split the pod
	356	splitpod("$podroot/$dirname/$filename", "$podroot/$dirname", $htmldir,
	357	$splitdirs);
	358	}
	359	}
	360
	361
	362	sub split_on_item {
363	my($podroot, $splitdirs, $ignore, @splititem) = @_;
364	my($pwd, $dirname, $filename);
365
366	print "splitting files by item.\n" if $verbose && $#splititem >= 0;
367	$pwd = getcwd();
3e3baf6d	368	my $splitter = absolute_path($pwd, "$splitpod/splitpod");
54310121	369	foreach $pod (@splititem) {
	370	# figure out the directory to split into
	371	$pod =~ s,^([^/]*)$,/$1,;
	372	$pod =~ m,(.?)/(.?)(\.pod)?$,;
	373	$dirname = "$1/$2";
	374	$filename = "$2.pod";
	375
	376	# since we are splitting this file it shouldn't be converted.
	377	push(@$ignore, "$podroot/$dirname.pod");
	378
	379	# split the pod
	380	push(@$splitdirs, "$podroot/$dirname");
	381	if (! -d "$podroot/$dirname") {
	382	mkdir("$podroot/$dirname", 0755) \|\|
	383	die "$0: error creating directory $podroot/$dirname: $!\n";
	384	}
	385	chdir("$podroot/$dirname") \|\|
	386	die "$0: error changing to directory $podroot/$dirname: $!\n";
3e3baf6d TB	387	die "$splitter not found. Use '-splitpod dir' option.\n"
	388	unless -f $splitter;
	389	system("perl", $splitter, "../$filename") &&
	390	warn "$0: error running '$splitter ../$filename'"
e05c04ea	391	." from $podroot/$dirname";
54310121	392	}
	393	chdir($pwd);
	394	}
	395
	396
	397	#
	398	# splitpod - splits a .pod file into several smaller .pod files
	399	# where a new file is started each time a =head[1-6] pod directive
	400	# is encountered in the input file.
	401	#
	402	sub splitpod {
	403	my($pod, $poddir, $htmldir, $splitdirs) = @_;
	404	my(@poddata, @filedata, @heads);
	405	my($file, $i, $j, $prevsec, $section, $nextsec);
	406
	407	print "splitting $pod\n" if $verbose;
	408
	409	# read the file in paragraphs
	410	$/ = "";
	411	open(SPLITIN, "<$pod") \|\|
	412	die "$0: error opening $pod for input: $!\n";
	413	@filedata = <SPLITIN>;
	414	close(SPLITIN) \|\|
	415	die "$0: error closing $pod: $!\n";
	416
	417	# restore the file internally by =head[1-6] sections
	418	@poddata = ();
	419	for ($i = 0, $j = -1; $i <= $#filedata; $i++) {
	420	$j++ if ($filedata[$i] =~ /^\s*=head[1-6]/);
	421	if ($j >= 0) {
	422	$poddata[$j] = "" unless defined $poddata[$j];
	423	$poddata[$j] .= "\n$filedata[$i]" if $j >= 0;
	424	}
	425	}
	426
	427	# create list of =head[1-6] sections so that we can rewrite
	428	# L<> links as necessary.
	429	%heads = ();
	430	foreach $i (0..$#poddata) {
	431	$heads{htmlize($1)} = 1 if $poddata[$i] =~ /=head[1-6]\s+(.*)/;
	432	}
	433
	434	# create a directory of a similar name and store all the
	435	# files in there
	436	$pod =~ s,./(.),$1,; # get the last part of the name
	437	$dir = $pod;
	438	$dir =~ s/\.pod//g;
	439	push(@$splitdirs, "$poddir/$dir");
	440	mkdir("$poddir/$dir", 0755) \|\|
	441	die "$0: could not create directory $poddir/$dir: $!\n"
	442	unless -d "$poddir/$dir";
	443
	444	$poddata[0] =~ /^\s=head[1-6]\s+(.)/;
	445	$section = "";
	446	$nextsec = $1;
	447
	448	# for each section of the file create a separate pod file
	449	for ($i = 0; $i <= $#poddata; $i++) {
	450	# determine the "prev" and "next" links
	451	$prevsec = $section;
	452	$section = $nextsec;
	453	if ($i < $#poddata) {
	454	$poddata[$i+1] =~ /^\s=head[1-6]\s+(.)/;
	455	$nextsec = $1;
456	} else {
457	$nextsec = "";
458	}
459
460	# determine an appropriate filename (this must correspond with
461	# what pod2html will try and guess)
462	# $poddata[$i] =~ /^\s=head[1-6]\s+(.)/;
463	$file = "$dir/" . htmlize($section) . ".pod";
464
465	# create the new .pod file
466	print "\tcreating $poddir/$file\n" if $verbose;
467	open(SPLITOUT, ">$poddir/$file") \|\|
468	die "$0: error opening $poddir/$file for output: $!\n";
469	$poddata[$i] =~ s,L<([^<>]*)>,
470	defined $heads{htmlize($1)} ? "L<$dir/$1>" : "L<$1>"
471	,ge;
472	print SPLITOUT $poddata[$i]."\n\n";
473	print SPLITOUT "=over 4\n\n";
474	print SPLITOUT "=item *\n\nBack to L<$dir/\"$prevsec\">\n\n" if $prevsec;
475	print SPLITOUT "=item *\n\nForward to L<$dir/\"$nextsec\">\n\n" if $nextsec;
476	print SPLITOUT "=item *\n\nUp to L<$dir>\n\n";
477	print SPLITOUT "=back\n\n";
478	close(SPLITOUT) \|\|
479	die "$0: error closing $poddir/$file: $!\n";
480	}
481	}
482
483
484	#
485	# installdir - takes care of converting the .pod and .pm files in the
486	# current directory to .html files and then installing those.
487	#
488	sub installdir {
489	my($dir, $recurse, $podroot, $splitdirs, $ignore) = @_;
490	my(@dirlist, @podlist, @pmlist, $doindex);
491
492	@dirlist = (); # directories to recurse on
493	@podlist = (); # .pod files to install
494	@pmlist = (); # .pm files to install
495
496	# should files in this directory get an index?
497	$doindex = (grep($_ eq "$podroot/$dir", @$splitdirs) ? 0 : 1);
498
499	opendir(DIR, "$podroot/$dir")
500	\|\| die "$0: error opening directory $podroot/$dir: $!\n";
501
502	# find the directories to recurse on
503	@dirlist = map { "$dir/$_" }
504	grep(-d "$podroot/$dir/$_" && !/^\.{1,2}/, readdir(DIR)) if $recurse;
505	rewinddir(DIR);
506
507	# find all the .pod files within the directory
508	@podlist = map { /^(.*)\.pod$/; "$dir/$1" }
509	grep(! -d "$podroot/$dir/$_" && /\.pod$/, readdir(DIR));
510	rewinddir(DIR);
511
512	# find all the .pm files within the directory
513	@pmlist = map { /^(.*)\.pm$/; "$dir/$1" }
514	grep(! -d "$podroot/$dir/$_" && /\.pm$/, readdir(DIR));
515
516	closedir(DIR);
517
518	# recurse on all subdirectories we kept track of
519	foreach $dir (@dirlist) {
520	installdir($dir, $recurse, $podroot, $splitdirs, $ignore);
521	}
522
523	# install all the pods we found
524	foreach $pod (@podlist) {
525	# check if we should ignore it.
526	next if grep($_ eq "$podroot/$pod.pod", @$ignore);
527
528	# check if a .pm files exists too
529	if (grep($_ eq "$pod.pm", @pmlist)) {
530	print "$0: Warning both `$podroot/$pod.pod' and "
531	. "`$podroot/$pod.pm' exist, using pod\n";
532	push(@ignore, "$pod.pm");
533	}
534	runpod2html("$pod.pod", $doindex);
535	}
536
537	# install all the .pm files we found
538	foreach $pm (@pmlist) {
539	# check if we should ignore it.
540	next if grep($_ eq "$pm.pm", @ignore);
541
542	runpod2html("$pm.pm", $doindex);
543	}
544	}
545
546
547	#
548	# runpod2html - invokes pod2html to convert a .pod or .pm file to a .html
549	# file.
550	#
551	sub runpod2html {
552	my($pod, $doindex) = @_;
553	my($html, $i, $dir, @dirs);
554
555	$html = $pod;
556	$html =~ s/\.(pod\|pm)$/.html/g;
557
558	# make sure the destination directories exist
559	@dirs = split("/", $html);
560	$dir = "$htmldir/";
561	for ($i = 0; $i < $#dirs; $i++) {
562	if (! -d "$dir$dirs[$i]") {
563	mkdir("$dir$dirs[$i]", 0755) \|\|
564	die "$0: error creating directory $dir$dirs[$i]: $!\n";
565	}
566	$dir .= "$dirs[$i]/";
567	}
568
569	# invoke pod2html
570	print "$podroot/$pod => $htmldir/$html\n" if $verbose;
571	#system("./pod2html",
572	Pod::Html'pod2html(
573	#Pod::Html'pod2html($pod2html,
574	"--htmlroot=$htmlroot",
575	"--podpath=".join(":", @podpath),
576	"--podroot=$podroot", "--netscape",
577	($doindex ? "--index" : "--noindex"),
578	"--" . ($recurse ? "" : "no") . "recurse",
579	($#libpods >= 0) ? "--libpods=" . join(":", @libpods) : "",
580	"--infile=$podroot/$pod", "--outfile=$htmldir/$html");
581	die "$0: error running $pod2html: $!\n" if $?;
582	}
583
584	sub htmlize { htmlify(0, @_) }