[ViewVC] Annotation of: cvs/cvsroot/staticperl/perl/installhtml

#!./perl -Ilib -w

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

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

=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

Comma-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

Comma-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

Comma-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

my $usage;

$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

my (@libpods, @podpath, $podroot, $htmldir, $htmlroot, $recurse, @splithead,
    @splititem, $splitpod, $verbose, $pod2html);

@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;

# Overcome shell's p1,..,p8 limitation.  
# See vms/descrip_mms.template -> descrip.mms for invokation.
if ( $^O eq 'VMS' ) { @ARGV = split(/\s+/,$ARGV[0]); }

use vars qw( %Options );

# parse the command-line
my $result = GetOptions( \%Options, 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.
my @ignore = ();
my @splitdirs;

# 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 my $dir (@podpath) {
    installdir($dir, $recurse, $podroot, \@splitdirs, \@ignore);
}


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

foreach my $dir (@splithead) {
    (my $pod = $dir) =~ s,^.*/,,;
    $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;
    my $file = "$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";
    $/ = "";
    my @data = ();
    while (<H>) {
        last if /name="name"/i;
        $_ =~ s{href="#(.*)">}{
            my $url = "$pod/$1.html" ;
            $url = Pod::Html::relativize_url( $url, "$file.html" )
            if ( ! defined $Options{htmlroot} || $Options{htmlroot} eq '' );
            "href=\"$url\">" ;
        }egi;
        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 $Options{help};
    $Options{help} = "";                    # make -w shut up

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

    # lists of files
    @splithead = split(",", $Options{splithead}) if defined $Options{splithead};
    @splititem = split(",", $Options{splititem}) if defined $Options{splititem};
    @libpods   = split(",", $Options{libpods}) if defined $Options{libpods};

    $htmldir  = $Options{htmldir}           if defined $Options{htmldir};
    $htmlroot = $Options{htmlroot}          if defined $Options{htmlroot};
    $podroot  = $Options{podroot}           if defined $Options{podroot};
    $splitpod = $Options{splitpod}          if defined $Options{splitpod};

    $recurse  = $Options{recurse}           if defined $Options{recurse};
    $verbose  = $Options{verbose}           if defined $Options{verbose};
}


sub create_index {
    my($html, $dir) = @_;
    (my $pod = $dir) =~ s,^.*/,,;
    my(@files, @filedata, @index, $file);
    my($lcp1,$lcp2);


    # 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
        my $name;
        ($name) = grep(/name="name"/i, @filedata);
        ($lcp1,$lcp2) = ($name =~ m,/H1>\s(\S+)\s[\s-]*(.*?)\s*$,smi);
        if (defined $lcp1 and $lcp1 =~ m,^<P>$,i) { # Uninteresting.  Try again.
            ($lcp1,$lcp2) = ($name =~ m,/H1>\s<P>\s*(\S+)\s[\s-]*(.*?)\s*$,smi);
        }
        my $url= "$pod/$file" ;
        if ( ! defined $Options{htmlroot} || $Options{htmlroot} eq '' ) {
            $url = Pod::Html::relativize_url( "$pod/$file", $html ) ;
        }

        if (defined $lcp1) {
            print HTML qq(<DT><A HREF="$url">);
            print HTML "$lcp1</A></DT><DD>$lcp2</DD>\n";
        }

        next;

        @index = grep(/<!-- INDEX BEGIN -->.*<!-- INDEX END -->/s,
                    @filedata);
        for (@index) {
            s/<!-- INDEX BEGIN -->(\s*<!--)(.*)(-->\s*)<!-- INDEX END -->/$lcp2/s;
            s,#,$dir/$file#,g;
            print HTML "$_\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 $splitter = File::Spec->rel2abs("$splitpod/splitpod", $pwd);
    my $perl = File::Spec->rel2abs($^X, $pwd);
    foreach my $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.
    my %heads = ();
    foreach $i (0..$#poddata) {
        $heads{anchorify($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
    my $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/" . anchorify($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{anchorify($1)} ? "L<$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 { if ($^O eq 'VMS') {/^(.*)\.dir$/i; "$dir/$1";} else {"$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 my $pod (@podlist) {
        # check if we should ignore it.
        next if grep($_ eq "$podroot/$pod.pod", @$ignore);

        # check if a .pm files exists too
        if (grep($_ eq $pod, @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 my $pm (@pmlist) {
        # check if we should ignore it.
        next if grep($_ 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 $verbose;
    Pod::Html::pod2html(
        "--htmldir=$htmldir",
        "--htmlroot=$htmlroot",
        "--podpath=".join(":", @podpath),
        "--podroot=$podroot", "--netscape",
        "--header",
        ($doindex ? "--index" : "--noindex"),
        "--" . ($recurse ? "" : "no") . "recurse",
        ($#libpods >= 0) ? "--libpods=" . join(":", @libpods) : "",
        "--infile=$podroot/$pod", "--outfile=$htmldir/$html");
    die "$0: error running $pod2html: $!\n" if $?;
}
Revision:	1.1
Committed:	Thu Jun 30 14:26:41 2005 UTC (21 years, 3 months ago) by root
Branch:	MAIN
CVS Tags:	PERL-5-8-7, HEAD
Branch point for:	PERL
Log Message:	* empty log message *
#	User	Rev	Content
1	root	1.1	#!./perl -Ilib -w
2
3			# This file should really be extracted from a .PL file
4
5			use strict;
6			use Config; # for config options in the makefile
7			use File::Spec;
8			use Getopt::Long; # for command-line parsing
9			use Cwd;
10			use Pod::Html 'anchorify';
11
12			=head1 NAME
13
14			installhtml - 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
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.
28
29			=head1 OPTIONS
30
31			=over 4
32
33			=item B<--help> help
34
35			Displays the usage.
36
37			=item B<--podroot> POD search path base directory
38
39			The base directory to search for all .pod and .pm files to be converted.
40			Default is current directory.
41
42			=item B<--podpath> POD search path
43
44			The list of directories to search for .pod and .pm files to be converted.
45			Default is `podroot/.'.
46
47			=item B<--recurse> recurse on subdirectories
48
49			Whether or not to convert all .pm and .pod files found in subdirectories
50			too. Default is to not recurse.
51
52			=item B<--htmldir> HTML destination directory
53
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.
56
57			=item B<--htmlroot> URL base directory
58
59			The base directory which all resulting HTML files will be visible at in
60			a URL. The default is `/'.
61
62			=item B<--splithead> POD files to split on =head directive
63
64			Comma-separated list of pod files to split by the =head directive. The
65			.pod suffix is optional. These files should have names specified
66			relative to podroot.
67
68			=item B<--splititem> POD files to split on =item directive
69
70			Comma-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.
74
75			=item B<--splitpod> Directory containing the splitpod program
76
77			The directory containing the splitpod program. The default is `podroot/pod'.
78
79			=item B<--libpods> library PODs for LE<lt>E<gt> links
80
81			Comma-separated list of "library" pod files. This is the same list that
82			will be passed to pod2html when any pod is converted.
83
84			=item B<--verbose> verbose output
85
86			Self-explanatory.
87
88			=back
89
90			=head1 EXAMPLE
91
92			The following command-line is an example of the one we use to convert
93			perl documentation:
94
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 \
102			--recurse \
103			--verbose
104
105			=head1 AUTHOR
106
107			Chris Hall E<lt>hallc@cs.colorado.eduE<gt>
108
109			=head1 TODO
110
111			=cut
112
113			my $usage;
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
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
140			splitpod.
141			--splitpod - directory where the program splitpod can be found
142			(\$podroot/pod by default).
143			--verbose - self-explanatory.
144
145			END_OF_USAGE
146
147			my (@libpods, @podpath, $podroot, $htmldir, $htmlroot, $recurse, @splithead,
148			@splititem, $splitpod, $verbose, $pod2html);
149
150			@libpods = ();
151			@podpath = ( "." ); # colon-separated list of directories containing .pod
152			# and .pm files to be converted.
153			$podroot = "."; # assume the pods we want are here
154			$htmldir = ""; # nothing for now...
155			$htmlroot = "/"; # default value
156			$recurse = 0; # default behavior
157			@splithead = (); # don't split any files by default
158			@splititem = (); # don't split any files by default
159			$splitpod = ""; # nothing for now.
160
161			$verbose = 0; # whether or not to print debugging info
162
163			$pod2html = "pod/pod2html";
164
165			usage("") unless @ARGV;
166
167			# Overcome shell's p1,..,p8 limitation.
168			# See vms/descrip_mms.template -> descrip.mms for invokation.
169			if ( $^O eq 'VMS' ) { @ARGV = split(/\s+/,$ARGV[0]); }
170
171			use vars qw( %Options );
172
173			# parse the command-line
174			my $result = GetOptions( \%Options, qw(
175			help
176			podpath=s
177			podroot=s
178			htmldir=s
179			htmlroot=s
180			libpods=s
181			recurse!
182			splithead=s
183			splititem=s
184			splitpod=s
185			verbose
186			));
187			usage("invalid parameters") unless $result;
188			parse_command_line();
189
190
191			# set these variables to appropriate values if the user didn't specify
192			# values for them.
193			$htmldir = "$htmlroot/html" unless $htmldir;
194			$splitpod = "$podroot/pod" unless $splitpod;
195
196
197			# make sure that the destination directory exists
198			(mkdir($htmldir, 0755) \|\|
199			die "$0: cannot make directory $htmldir: $!\n") if ! -d $htmldir;
200
201
202			# the following array will eventually contain files that are to be
203			# ignored in the conversion process. these are files that have been
204			# process by splititem or splithead and should not be converted as a
205			# result.
206			my @ignore = ();
207			my @splitdirs;
208
209			# split pods. its important to do this before convert ANY pods because
210			# it may effect some of the links
211			@splitdirs = (); # files in these directories won't get an index
212			split_on_head($podroot, $htmldir, \@splitdirs, \@ignore, @splithead);
213			split_on_item($podroot, \@splitdirs, \@ignore, @splititem);
214
215
216			# convert the pod pages found in @poddirs
217			#warn "converting files\n" if $verbose;
218			#warn "\@ignore\t= @ignore\n" if $verbose;
219			foreach my $dir (@podpath) {
220			installdir($dir, $recurse, $podroot, \@splitdirs, \@ignore);
221			}
222
223
224			# now go through and create master indices for each pod we split
225			foreach my $dir (@splititem) {
226			print "creating index $htmldir/$dir.html\n" if $verbose;
227			create_index("$htmldir/$dir.html", "$htmldir/$dir");
228			}
229
230			foreach my $dir (@splithead) {
231			(my $pod = $dir) =~ s,^.*/,,;
232			$dir .= ".pod" unless $dir =~ /(\.pod\|\.pm)$/;
233			# let pod2html create the file
234			runpod2html($dir, 1);
235
236			# now go through and truncate after the index
237			$dir =~ /^(.*?)(\.pod\|\.pm)?$/sm;
238			my $file = "$htmldir/$1";
239			print "creating index $file.html\n" if $verbose;
240
241			# read in everything until what would have been the first =head
242			# directive, patching the index as we go.
243			open(H, "<$file.html") \|\|
244			die "$0: error opening $file.html for input: $!\n";
245			$/ = "";
246			my @data = ();
247			while (<H>) {
248			last if /name="name"/i;
249			$_ =~ s{href="#(.*)">}{
250			my $url = "$pod/$1.html" ;
251			$url = Pod::Html::relativize_url( $url, "$file.html" )
252			if ( ! defined $Options{htmlroot} \|\| $Options{htmlroot} eq '' );
253			"href=\"$url\">" ;
254			}egi;
255			push @data, $_;
256			}
257			close(H);
258
259			# now rewrite the file
260			open(H, ">$file.html") \|\|
261			die "$0: error opening $file.html for output: $!\n";
262			print H "@data", "\n";
263			close(H);
264			}
265
266			##############################################################################
267
268
269			sub usage {
270			warn "$0: @_\n" if @_;
271			die $usage;
272			}
273
274
275			sub parse_command_line {
276			usage() if defined $Options{help};
277			$Options{help} = ""; # make -w shut up
278
279			# list of directories
280			@podpath = split(":", $Options{podpath}) if defined $Options{podpath};
281
282			# lists of files
283			@splithead = split(",", $Options{splithead}) if defined $Options{splithead};
284			@splititem = split(",", $Options{splititem}) if defined $Options{splititem};
285			@libpods = split(",", $Options{libpods}) if defined $Options{libpods};
286
287			$htmldir = $Options{htmldir} if defined $Options{htmldir};
288			$htmlroot = $Options{htmlroot} if defined $Options{htmlroot};
289			$podroot = $Options{podroot} if defined $Options{podroot};
290			$splitpod = $Options{splitpod} if defined $Options{splitpod};
291
292			$recurse = $Options{recurse} if defined $Options{recurse};
293			$verbose = $Options{verbose} if defined $Options{verbose};
294			}
295
296
297			sub create_index {
298			my($html, $dir) = @_;
299			(my $pod = $dir) =~ s,^.*/,,;
300			my(@files, @filedata, @index, $file);
301			my($lcp1,$lcp2);
302
303
304			# get the list of .html files in this directory
305			opendir(DIR, $dir) \|\|
306			die "$0: error opening directory $dir for reading: $!\n";
307			@files = sort(grep(/\.html?$/, readdir(DIR)));
308			closedir(DIR);
309
310			open(HTML, ">$html") \|\|
311			die "$0: error opening $html for output: $!\n";
312
313			# for each .html file in the directory, extract the index
314			# embedded in the file and throw it into the big index.
315			print HTML "<DL COMPACT>\n";
316			foreach $file (@files) {
317			$/ = "";
318
319			open(IN, "<$dir/$file") \|\|
320			die "$0: error opening $dir/$file for input: $!\n";
321			@filedata = <IN>;
322			close(IN);
323
324			# pull out the NAME section
325			my $name;
326			($name) = grep(/name="name"/i, @filedata);
327			($lcp1,$lcp2) = ($name =~ m,/H1>\s(\S+)\s[\s-](.?)\s*$,smi);
328			if (defined $lcp1 and $lcp1 =~ m,^<P>$,i) { # Uninteresting. Try again.
329			($lcp1,$lcp2) = ($name =~ m,/H1>\s<P>\s(\S+)\s[\s-](.?)\s$,smi);
330			}
331			my $url= "$pod/$file" ;
332			if ( ! defined $Options{htmlroot} \|\| $Options{htmlroot} eq '' ) {
333			$url = Pod::Html::relativize_url( "$pod/$file", $html ) ;
334			}
335
336			if (defined $lcp1) {
337			print HTML qq(<DT><A HREF="$url">);
338			print HTML "$lcp1</A></DT><DD>$lcp2</DD>\n";
339			}
340
341			next;
342
343			@index = grep(/<!-- INDEX BEGIN -->.*<!-- INDEX END -->/s,
344			@filedata);
345			for (@index) {
346			s/<!-- INDEX BEGIN -->(\s<!--)(.)(-->\s*)<!-- INDEX END -->/$lcp2/s;
347			s,#,$dir/$file#,g;
348			print HTML "$_\n<P><HR><P>\n";
349			}
350			}
351			print HTML "</DL>\n";
352
353			close(HTML);
354			}
355
356
357			sub split_on_head {
358			my($podroot, $htmldir, $splitdirs, $ignore, @splithead) = @_;
359			my($pod, $dirname, $filename);
360
361			# split the files specified in @splithead on =head[1-6] pod directives
362			print "splitting files by head.\n" if $verbose && $#splithead >= 0;
363			foreach $pod (@splithead) {
364			# figure out the directory name and filename
365			$pod =~ s,^([^/]*)$,/$1,;
366			$pod =~ m,(.)/(.?)(\.pod)?$,;
367			$dirname = $1;
368			$filename = "$2.pod";
369
370			# since we are splitting this file it shouldn't be converted.
371			push(@$ignore, "$podroot/$dirname/$filename");
372
373			# split the pod
374			splitpod("$podroot/$dirname/$filename", "$podroot/$dirname", $htmldir,
375			$splitdirs);
376			}
377			}
378
379
380			sub split_on_item {
381			my($podroot, $splitdirs, $ignore, @splititem) = @_;
382			my($pwd, $dirname, $filename);
383
384			print "splitting files by item.\n" if $verbose && $#splititem >= 0;
385			$pwd = getcwd();
386			my $splitter = File::Spec->rel2abs("$splitpod/splitpod", $pwd);
387			my $perl = File::Spec->rel2abs($^X, $pwd);
388			foreach my $pod (@splititem) {
389			# figure out the directory to split into
390			$pod =~ s,^([^/]*)$,/$1,;
391			$pod =~ m,(.)/(.?)(\.pod)?$,;
392			$dirname = "$1/$2";
393			$filename = "$2.pod";
394
395			# since we are splitting this file it shouldn't be converted.
396			push(@$ignore, "$podroot/$dirname.pod");
397
398			# split the pod
399			push(@$splitdirs, "$podroot/$dirname");
400			if (! -d "$podroot/$dirname") {
401			mkdir("$podroot/$dirname", 0755) \|\|
402			die "$0: error creating directory $podroot/$dirname: $!\n";
403			}
404			chdir("$podroot/$dirname") \|\|
405			die "$0: error changing to directory $podroot/$dirname: $!\n";
406			die "$splitter not found. Use '-splitpod dir' option.\n"
407			unless -f $splitter;
408			system($perl, $splitter, "../$filename") &&
409			warn "$0: error running '$splitter ../$filename'"
410			." from $podroot/$dirname";
411			}
412			chdir($pwd);
413			}
414
415
416			#
417			# splitpod - splits a .pod file into several smaller .pod files
418			# where a new file is started each time a =head[1-6] pod directive
419			# is encountered in the input file.
420			#
421			sub splitpod {
422			my($pod, $poddir, $htmldir, $splitdirs) = @_;
423			my(@poddata, @filedata, @heads);
424			my($file, $i, $j, $prevsec, $section, $nextsec);
425
426			print "splitting $pod\n" if $verbose;
427
428			# read the file in paragraphs
429			$/ = "";
430			open(SPLITIN, "<$pod") \|\|
431			die "$0: error opening $pod for input: $!\n";
432			@filedata = <SPLITIN>;
433			close(SPLITIN) \|\|
434			die "$0: error closing $pod: $!\n";
435
436			# restore the file internally by =head[1-6] sections
437			@poddata = ();
438			for ($i = 0, $j = -1; $i <= $#filedata; $i++) {
439			$j++ if ($filedata[$i] =~ /^\s*=head[1-6]/);
440			if ($j >= 0) {
441			$poddata[$j] = "" unless defined $poddata[$j];
442			$poddata[$j] .= "\n$filedata[$i]" if $j >= 0;
443			}
444			}
445
446			# create list of =head[1-6] sections so that we can rewrite
447			# L<> links as necessary.
448			my %heads = ();
449			foreach $i (0..$#poddata) {
450			$heads{anchorify($1)} = 1 if $poddata[$i] =~ /=head[1-6]\s+(.*)/;
451			}
452
453			# create a directory of a similar name and store all the
454			# files in there
455			$pod =~ s,./(.),$1,; # get the last part of the name
456			my $dir = $pod;
457			$dir =~ s/\.pod//g;
458			push(@$splitdirs, "$poddir/$dir");
459			mkdir("$poddir/$dir", 0755) \|\|
460			die "$0: could not create directory $poddir/$dir: $!\n"
461			unless -d "$poddir/$dir";
462
463			$poddata[0] =~ /^\s=head[1-6]\s+(.)/;
464			$section = "";
465			$nextsec = $1;
466
467			# for each section of the file create a separate pod file
468			for ($i = 0; $i <= $#poddata; $i++) {
469			# determine the "prev" and "next" links
470			$prevsec = $section;
471			$section = $nextsec;
472			if ($i < $#poddata) {
473			$poddata[$i+1] =~ /^\s=head[1-6]\s+(.)/;
474			$nextsec = $1;
475			} else {
476			$nextsec = "";
477			}
478
479			# determine an appropriate filename (this must correspond with
480			# what pod2html will try and guess)
481			# $poddata[$i] =~ /^\s=head[1-6]\s+(.)/;
482			$file = "$dir/" . anchorify($section) . ".pod";
483
484			# create the new .pod file
485			print "\tcreating $poddir/$file\n" if $verbose;
486			open(SPLITOUT, ">$poddir/$file") \|\|
487			die "$0: error opening $poddir/$file for output: $!\n";
488			$poddata[$i] =~ s,L<([^<>]*)>,
489			defined $heads{anchorify($1)} ? "L<$dir/$1>" : "L<$1>"
490			,ge;
491			print SPLITOUT $poddata[$i]."\n\n";
492			print SPLITOUT "=over 4\n\n";
493			print SPLITOUT "=item *\n\nBack to L<$dir/\"$prevsec\">\n\n" if $prevsec;
494			print SPLITOUT "=item *\n\nForward to L<$dir/\"$nextsec\">\n\n" if $nextsec;
495			print SPLITOUT "=item *\n\nUp to L<$dir>\n\n";
496			print SPLITOUT "=back\n\n";
497			close(SPLITOUT) \|\|
498			die "$0: error closing $poddir/$file: $!\n";
499			}
500			}
501
502
503			#
504			# installdir - takes care of converting the .pod and .pm files in the
505			# current directory to .html files and then installing those.
506			#
507			sub installdir {
508			my($dir, $recurse, $podroot, $splitdirs, $ignore) = @_;
509			my(@dirlist, @podlist, @pmlist, $doindex);
510
511			@dirlist = (); # directories to recurse on
512			@podlist = (); # .pod files to install
513			@pmlist = (); # .pm files to install
514
515			# should files in this directory get an index?
516			$doindex = (grep($_ eq "$podroot/$dir", @$splitdirs) ? 0 : 1);
517
518			opendir(DIR, "$podroot/$dir")
519			\|\| die "$0: error opening directory $podroot/$dir: $!\n";
520
521			# find the directories to recurse on
522			@dirlist = map { if ($^O eq 'VMS') {/^(.*)\.dir$/i; "$dir/$1";} else {"$dir/$_";}}
523			grep(-d "$podroot/$dir/$_" && !/^\.{1,2}/, readdir(DIR)) if $recurse;
524			rewinddir(DIR);
525
526			# find all the .pod files within the directory
527			@podlist = map { /^(.*)\.pod$/; "$dir/$1" }
528			grep(! -d "$podroot/$dir/$_" && /\.pod$/, readdir(DIR));
529			rewinddir(DIR);
530
531			# find all the .pm files within the directory
532			@pmlist = map { /^(.*)\.pm$/; "$dir/$1" }
533			grep(! -d "$podroot/$dir/$_" && /\.pm$/, readdir(DIR));
534
535			closedir(DIR);
536
537			# recurse on all subdirectories we kept track of
538			foreach $dir (@dirlist) {
539			installdir($dir, $recurse, $podroot, $splitdirs, $ignore);
540			}
541
542			# install all the pods we found
543			foreach my $pod (@podlist) {
544			# check if we should ignore it.
545			next if grep($_ eq "$podroot/$pod.pod", @$ignore);
546
547			# check if a .pm files exists too
548			if (grep($_ eq $pod, @pmlist)) {
549			print "$0: Warning both `$podroot/$pod.pod' and "
550			. "`$podroot/$pod.pm' exist, using pod\n";
551			push(@ignore, "$pod.pm");
552			}
553			runpod2html("$pod.pod", $doindex);
554			}
555
556			# install all the .pm files we found
557			foreach my $pm (@pmlist) {
558			# check if we should ignore it.
559			next if grep($_ eq "$pm.pm", @ignore);
560
561			runpod2html("$pm.pm", $doindex);
562			}
563			}
564
565
566			#
567			# runpod2html - invokes pod2html to convert a .pod or .pm file to a .html
568			# file.
569			#
570			sub runpod2html {
571			my($pod, $doindex) = @_;
572			my($html, $i, $dir, @dirs);
573
574			$html = $pod;
575			$html =~ s/\.(pod\|pm)$/.html/g;
576
577			# make sure the destination directories exist
578			@dirs = split("/", $html);
579			$dir = "$htmldir/";
580			for ($i = 0; $i < $#dirs; $i++) {
581			if (! -d "$dir$dirs[$i]") {
582			mkdir("$dir$dirs[$i]", 0755) \|\|
583			die "$0: error creating directory $dir$dirs[$i]: $!\n";
584			}
585			$dir .= "$dirs[$i]/";
586			}
587
588			# invoke pod2html
589			print "$podroot/$pod => $htmldir/$html\n" if $verbose;
590			Pod::Html::pod2html(
591			"--htmldir=$htmldir",
592			"--htmlroot=$htmlroot",
593			"--podpath=".join(":", @podpath),
594			"--podroot=$podroot", "--netscape",
595			"--header",
596			($doindex ? "--index" : "--noindex"),
597			"--" . ($recurse ? "" : "no") . "recurse",
598			($#libpods >= 0) ? "--libpods=" . join(":", @libpods) : "",
599			"--infile=$podroot/$pod", "--outfile=$htmldir/$html");
600			die "$0: error running $pod2html: $!\n" if $?;
601			}