[p5sagit/p5-mst-13.2.git] / 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 Getopt::Long;	# for command-line parsing
use Cwd;
use Pod::Html;

=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) {
    $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=/;
	$_ =~ s{HREF="#(.*)">}{
	    my $url = "$file/$1.html" ;
            $url = Pod::Html::relativize_url( $url, "$file.html" )
                if ( ! defined $Options{htmlroot} || $Options{htmlroot} eq '' );
            "HREF=\"$url\">" ;
        }eg;
	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 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, $file);
    my($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=/, @filedata);
	($lcp1,$lcp2) = ($name =~ m,/H1>\s(\S+)\s[\s-]*(.*?)\s*$,sm);
	if (defined $lcp1 and $lcp1 eq '<P>') { # Uninteresting.  Try again.
	   ($lcp1,$lcp2) = ($name =~ m,/H1>\s<P>\s(\S+)\s[\s-]*(.*?)\s*$,sm);
	}
	my $url= "$dir/$file" ;
	if ( ! defined $Options{htmlroot} || $Options{htmlroot} eq '' ) {
	    $url = Pod::Html::relativize_url( "$dir/$file", $html ) ;
	}

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

	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";
	    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 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{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
    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/" . 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 { 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.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 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 $verbose;
#system("./pod2html",
        Pod::Html'pod2html(
        #Pod::Html'pod2html($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 $?;
}

sub htmlize { htmlify(0, @_) }
Commit	Line	Data
3d8876fc	1	#!./perl -Ilib -w
90248788	2
53542a37	3	# This file should really be extracted from a .PL file
54310121	4
3d8876fc	5	use strict;
54310121	6	use Config; # for config options in the makefile
	7	use Getopt::Long; # for command-line parsing
	8	use Cwd;
	9	use Pod::Html;
	10
54310121	11	=head1 NAME
	12
	13	installhtml - converts a collection of POD pages to HTML format.
	14
	15	=head1 SYNOPSIS
	16
	17	installhtml [--help] [--podpath=<name>:...:<name>] [--podroot=<name>]
	18	[--htmldir=<name>] [--htmlroot=<name>] [--norecurse] [--recurse]
	19	[--splithead=<name>,...,<name>] [--splititem=<name>,...,<name>]
	20	[--libpods=<name>,...,<name>] [--verbose]
	21
	22	=head1 DESCRIPTION
	23
	24	I<installhtml> converts a collection of POD pages to a corresponding
	25	collection of HTML pages. This is primarily used to convert the pod
	26	pages found in the perl distribution.
	27
	28	=head1 OPTIONS
	29
	30	=over 4
	31
	32	=item B<--help> help
	33
	34	Displays the usage.
	35
3e3baf6d	36	=item B<--podroot> POD search path base directory
	37
	38	The base directory to search for all .pod and .pm files to be converted.
	39	Default is current directory.
	40
54310121	41	=item B<--podpath> POD search path
54310121	42
e05c04ea	43	The list of directories to search for .pod and .pm files to be converted.
54310121	44	Default is `podroot/.'.
54310121	45
3e3baf6d	46	=item B<--recurse> recurse on subdirectories
54310121	47
3e3baf6d	48	Whether or not to convert all .pm and .pod files found in subdirectories
3e3baf6d	49	too. Default is to not recurse.
54310121	50
	51	=item B<--htmldir> HTML destination directory
	52
	53	The base directory which all HTML files will be written to. This should
	54	be a path relative to the filesystem, not the resulting URL.
	55
	56	=item B<--htmlroot> URL base directory
	57
	58	The base directory which all resulting HTML files will be visible at in
	59	a URL. The default is `/'.
	60
54310121	61	=item B<--splithead> POD files to split on =head directive
54310121	62
af19f77d	63	Comma-separated list of pod files to split by the =head directive. The
3e3baf6d	64	.pod suffix is optional. These files should have names specified
3e3baf6d	65	relative to podroot.
54310121	66
	67	=item B<--splititem> POD files to split on =item directive
	68
af19f77d	69	Comma-separated list of all pod files to split by the =item directive.
3e3baf6d	70	The .pod suffix is optional. I<installhtml> does not do the actual
	71	split, rather it invokes I<splitpod> to do the dirty work. As with
	72	--splithead, these files should have names specified relative to podroot.
	73
	74	=item B<--splitpod> Directory containing the splitpod program
	75
	76	The directory containing the splitpod program. The default is `podroot/pod'.
54310121	77
	78	=item B<--libpods> library PODs for LE<lt>E<gt> links
	79
af19f77d	80	Comma-separated list of "library" pod files. This is the same list that
54310121	81	will be passed to pod2html when any pod is converted.
	82
	83	=item B<--verbose> verbose output
	84
	85	Self-explanatory.
	86
	87	=back
	88
	89	=head1 EXAMPLE
	90
	91	The following command-line is an example of the one we use to convert
	92	perl documentation:
	93
	94	./installhtml --podpath=lib:ext:pod:vms \
84902520	95	--podroot=/usr/src/perl \
	96	--htmldir=/perl/nmanual \
	97	--htmlroot=/perl/nmanual \
	98	--splithead=pod/perlipc \
	99	--splititem=pod/perlfunc \
af19f77d	100	--libpods=perlfunc,perlguts,perlvar,perlrun,perlop \
84902520	101	--recurse \
84902520	102	--verbose
54310121	103
	104	=head1 AUTHOR
	105
	106	Chris Hall E<lt>hallc@cs.colorado.eduE<gt>
	107
	108	=head1 TODO
	109
	110	=cut
	111
3d8876fc	112	my $usage;
3d8876fc	113
54310121	114	$usage =<<END_OF_USAGE;
	115	Usage: $0 --help --podpath=<name>:...:<name> --podroot=<name>
	116	--htmldir=<name> --htmlroot=<name> --norecurse --recurse
	117	--splithead=<name>,...,<name> --splititem=<name>,...,<name>
	118	--libpods=<name>,...,<name> --verbose
	119
	120	--help - this message
	121	--podpath - colon-separated list of directories containing .pod and
	122	.pm files to be converted (. by default).
	123	--podroot - filesystem base directory from which all relative paths in
	124	podpath stem (default is .).
	125	--htmldir - directory to store resulting html files in relative
	126	to the filesystem (\$podroot/html by default).
	127	--htmlroot - http-server base directory from which all relative paths
	128	in podpath stem (default is /).
	129	--libpods - comma-separated list of files to search for =item pod
	130	directives in as targets of C<> and implicit links (empty
	131	by default).
	132	--norecurse - don't recurse on those subdirectories listed in podpath.
	133	(default behavior).
	134	--recurse - recurse on those subdirectories listed in podpath
	135	--splithead - comma-separated list of .pod or .pm files to split. will
e05c04ea	136	split each file into several smaller files at every occurrence
54310121	137	of a pod =head[1-6] directive.
	138	--splititem - comma-separated list of .pod or .pm files to split using
	139	splitpod.
3e3baf6d	140	--splitpod - directory where the program splitpod can be found
3e3baf6d	141	(\$podroot/pod by default).
54310121	142	--verbose - self-explanatory.
	143
	144	END_OF_USAGE
	145
3d8876fc	146	my (@libpods, @podpath, $podroot, $htmldir, $htmlroot, $recurse, @splithead,
	147	@splititem, $splitpod, $verbose, $pod2html);
	148
54310121	149	@libpods = ();
	150	@podpath = ( "." ); # colon-separated list of directories containing .pod
	151	# and .pm files to be converted.
	152	$podroot = "."; # assume the pods we want are here
	153	$htmldir = ""; # nothing for now...
	154	$htmlroot = "/"; # default value
	155	$recurse = 0; # default behavior
	156	@splithead = (); # don't split any files by default
	157	@splititem = (); # don't split any files by default
	158	$splitpod = ""; # nothing for now.
	159
	160	$verbose = 0; # whether or not to print debugging info
	161
	162	$pod2html = "pod/pod2html";
	163
3e3baf6d	164	usage("") unless @ARGV;
54310121	165
c645ec3f	166	# Overcome shell's p1,..,p8 limitation.
	167	# See vms/descrip_mms.template -> descrip.mms for invokation.
	168	if ( $^O eq 'VMS' ) { @ARGV = split(/\s+/,$ARGV[0]); }
	169
65b03822	170	use vars qw( %Options );
3d8876fc	171
54310121	172	# parse the command-line
65b03822	173	my $result = GetOptions( \%Options, qw(
54310121	174	help
	175	podpath=s
	176	podroot=s
	177	htmldir=s
	178	htmlroot=s
	179	libpods=s
	180	recurse!
	181	splithead=s
	182	splititem=s
	183	splitpod=s
	184	verbose
	185	));
	186	usage("invalid parameters") unless $result;
	187	parse_command_line();
	188
	189
	190	# set these variables to appropriate values if the user didn't specify
	191	# values for them.
	192	$htmldir = "$htmlroot/html" unless $htmldir;
	193	$splitpod = "$podroot/pod" unless $splitpod;
	194
	195
	196	# make sure that the destination directory exists
	197	(mkdir($htmldir, 0755) \|\|
	198	die "$0: cannot make directory $htmldir: $!\n") if ! -d $htmldir;
	199
	200
	201	# the following array will eventually contain files that are to be
	202	# ignored in the conversion process. these are files that have been
	203	# process by splititem or splithead and should not be converted as a
	204	# result.
3d8876fc	205	my @ignore = ();
3d8876fc	206	my @splitdirs;
54310121	207
	208	# split pods. its important to do this before convert ANY pods because
	209	# it may effect some of the links
	210	@splitdirs = (); # files in these directories won't get an index
	211	split_on_head($podroot, $htmldir, \@splitdirs, \@ignore, @splithead);
3e3baf6d	212	split_on_item($podroot, \@splitdirs, \@ignore, @splititem);
54310121	213
	214
	215	# convert the pod pages found in @poddirs
	216	#warn "converting files\n" if $verbose;
	217	#warn "\@ignore\t= @ignore\n" if $verbose;
3d8876fc	218	foreach my $dir (@podpath) {
54310121	219	installdir($dir, $recurse, $podroot, \@splitdirs, \@ignore);
	220	}
	221
	222
	223	# now go through and create master indices for each pod we split
3d8876fc	224	foreach my $dir (@splititem) {
54310121	225	print "creating index $htmldir/$dir.html\n" if $verbose;
	226	create_index("$htmldir/$dir.html", "$htmldir/$dir");
	227	}
	228
3d8876fc	229	foreach my $dir (@splithead) {
54310121	230	$dir .= ".pod" unless $dir =~ /(\.pod\|\.pm)$/;
	231	# let pod2html create the file
	232	runpod2html($dir, 1);
	233
	234	# now go through and truncate after the index
	235	$dir =~ /^(.*?)(\.pod\|\.pm)?$/sm;
3d8876fc	236	my $file = "$htmldir/$1";
54310121	237	print "creating index $file.html\n" if $verbose;
	238
	239	# read in everything until what would have been the first =head
	240	# directive, patching the index as we go.
	241	open(H, "<$file.html") \|\|
	242	die "$0: error opening $file.html for input: $!\n";
	243	$/ = "";
3d8876fc	244	my @data = ();
54310121	245	while (<H>) {
54310121	246	last if /NAME=/;
29f227c9	247	$_ =~ s{HREF="#(.*)">}{
	248	my $url = "$file/$1.html" ;
	249	$url = Pod::Html::relativize_url( $url, "$file.html" )
65b03822	250	if ( ! defined $Options{htmlroot} \|\| $Options{htmlroot} eq '' );
29f227c9	251	"HREF=\"$url\">" ;
29f227c9	252	}eg;
54310121	253	push @data, $_;
	254	}
	255	close(H);
	256
	257	# now rewrite the file
	258	open(H, ">$file.html") \|\|
	259	die "$0: error opening $file.html for output: $!\n";
3d8876fc	260	print H "@data", "\n";
54310121	261	close(H);
	262	}
	263
	264	##############################################################################
	265
	266
	267	sub usage {
	268	warn "$0: @_\n" if @_;
	269	die $usage;
	270	}
	271
	272
	273	sub parse_command_line {
65b03822	274	usage() if defined $Options{help};
65b03822	275	$Options{help} = ""; # make -w shut up
54310121	276
54310121	277	# list of directories
65b03822	278	@podpath = split(":", $Options{podpath}) if defined $Options{podpath};
54310121	279
54310121	280	# lists of files
65b03822	281	@splithead = split(",", $Options{splithead}) if defined $Options{splithead};
	282	@splititem = split(",", $Options{splititem}) if defined $Options{splititem};
	283	@libpods = split(",", $Options{libpods}) if defined $Options{libpods};
54310121	284
65b03822	285	$htmldir = $Options{htmldir} if defined $Options{htmldir};
	286	$htmlroot = $Options{htmlroot} if defined $Options{htmlroot};
	287	$podroot = $Options{podroot} if defined $Options{podroot};
	288	$splitpod = $Options{splitpod} if defined $Options{splitpod};
54310121	289
65b03822	290	$recurse = $Options{recurse} if defined $Options{recurse};
65b03822	291	$verbose = $Options{verbose} if defined $Options{verbose};
54310121	292	}
	293
	294
3e3baf6d	295	sub absolute_path {
	296	my($cwd, $path) = @_;
	297	return "$cwd/$path" unless $path =~ m:/:;
	298	# add cwd if path is not already an absolute path
	299	$path = "$cwd/$path" if (substr($path,0,1) ne '/');
	300	return $path;
	301	}
	302
	303
54310121	304	sub create_index {
	305	my($html, $dir) = @_;
	306	my(@files, @filedata, @index, $file);
af19f77d	307	my($lcp1,$lcp2);
af19f77d	308
54310121	309
	310	# get the list of .html files in this directory
	311	opendir(DIR, $dir) \|\|
	312	die "$0: error opening directory $dir for reading: $!\n";
39e571d4	313	@files = sort(grep(/\.html?$/, readdir(DIR)));
54310121	314	closedir(DIR);
	315
	316	open(HTML, ">$html") \|\|
	317	die "$0: error opening $html for output: $!\n";
	318
	319	# for each .html file in the directory, extract the index
	320	# embedded in the file and throw it into the big index.
	321	print HTML "<DL COMPACT>\n";
	322	foreach $file (@files) {
	323	$/ = "";
	324
	325	open(IN, "<$dir/$file") \|\|
	326	die "$0: error opening $dir/$file for input: $!\n";
	327	@filedata = <IN>;
	328	close(IN);
	329
	330	# pull out the NAME section
3d8876fc	331	my $name;
54310121	332	($name) = grep(/NAME=/, @filedata);
af19f77d	333	($lcp1,$lcp2) = ($name =~ m,/H1>\s(\S+)\s[\s-](.?)\s*$,sm);
	334	if (defined $lcp1 and $lcp1 eq '<P>') { # Uninteresting. Try again.
	335	($lcp1,$lcp2) = ($name =~ m,/H1>\s<P>\s(\S+)\s[\s-](.?)\s*$,sm);
	336	}
29f227c9	337	my $url= "$dir/$file" ;
65b03822	338	if ( ! defined $Options{htmlroot} \|\| $Options{htmlroot} eq '' ) {
29f227c9	339	$url = Pod::Html::relativize_url( "$dir/$file", $html ) ;
	340	}
	341
	342	print HTML qq(<A HREF="$url">);
af19f77d	343	print HTML "<DT>$lcp1</A><DD>$lcp2\n" if defined $lcp1;
29f227c9	344	# print HTML qq(<A HREF="$url">$lcp1</A><BR>\n") if defined $lcp1;
54310121	345
	346	next;
	347
	348	@index = grep(/<!-- INDEX BEGIN -->.*<!-- INDEX END -->/s,
	349	@filedata);
	350	for (@index) {
af19f77d	351	s/<!-- INDEX BEGIN -->(\s<!--)(.)(-->\s*)<!-- INDEX END -->/$lcp2/s;
54310121	352	s,#,$dir/$file#,g;
	353	# print HTML "$_\n";
	354	print HTML "$_\n<P><HR><P>\n";
	355	}
	356	}
	357	print HTML "</DL>\n";
	358
	359	close(HTML);
	360	}
	361
	362
	363	sub split_on_head {
	364	my($podroot, $htmldir, $splitdirs, $ignore, @splithead) = @_;
	365	my($pod, $dirname, $filename);
	366
	367	# split the files specified in @splithead on =head[1-6] pod directives
	368	print "splitting files by head.\n" if $verbose && $#splithead >= 0;
	369	foreach $pod (@splithead) {
	370	# figure out the directory name and filename
	371	$pod =~ s,^([^/]*)$,/$1,;
	372	$pod =~ m,(.?)/(.?)(\.pod)?$,;
	373	$dirname = $1;
	374	$filename = "$2.pod";
	375
	376	# since we are splitting this file it shouldn't be converted.
	377	push(@$ignore, "$podroot/$dirname/$filename");
	378
	379	# split the pod
	380	splitpod("$podroot/$dirname/$filename", "$podroot/$dirname", $htmldir,
	381	$splitdirs);
	382	}
	383	}
	384
	385
	386	sub split_on_item {
	387	my($podroot, $splitdirs, $ignore, @splititem) = @_;
	388	my($pwd, $dirname, $filename);
	389
	390	print "splitting files by item.\n" if $verbose && $#splititem >= 0;
	391	$pwd = getcwd();
3e3baf6d	392	my $splitter = absolute_path($pwd, "$splitpod/splitpod");
3d8876fc	393	foreach my $pod (@splititem) {
54310121	394	# figure out the directory to split into
	395	$pod =~ s,^([^/]*)$,/$1,;
	396	$pod =~ m,(.?)/(.?)(\.pod)?$,;
	397	$dirname = "$1/$2";
	398	$filename = "$2.pod";
	399
	400	# since we are splitting this file it shouldn't be converted.
	401	push(@$ignore, "$podroot/$dirname.pod");
	402
	403	# split the pod
	404	push(@$splitdirs, "$podroot/$dirname");
	405	if (! -d "$podroot/$dirname") {
	406	mkdir("$podroot/$dirname", 0755) \|\|
	407	die "$0: error creating directory $podroot/$dirname: $!\n";
	408	}
	409	chdir("$podroot/$dirname") \|\|
	410	die "$0: error changing to directory $podroot/$dirname: $!\n";
3e3baf6d	411	die "$splitter not found. Use '-splitpod dir' option.\n"
	412	unless -f $splitter;
	413	system("perl", $splitter, "../$filename") &&
	414	warn "$0: error running '$splitter ../$filename'"
e05c04ea	415	." from $podroot/$dirname";
54310121	416	}
	417	chdir($pwd);
	418	}
	419
	420
	421	#
	422	# splitpod - splits a .pod file into several smaller .pod files
	423	# where a new file is started each time a =head[1-6] pod directive
	424	# is encountered in the input file.
	425	#
	426	sub splitpod {
	427	my($pod, $poddir, $htmldir, $splitdirs) = @_;
	428	my(@poddata, @filedata, @heads);
	429	my($file, $i, $j, $prevsec, $section, $nextsec);
	430
	431	print "splitting $pod\n" if $verbose;
	432
	433	# read the file in paragraphs
	434	$/ = "";
	435	open(SPLITIN, "<$pod") \|\|
	436	die "$0: error opening $pod for input: $!\n";
	437	@filedata = <SPLITIN>;
	438	close(SPLITIN) \|\|
	439	die "$0: error closing $pod: $!\n";
	440
	441	# restore the file internally by =head[1-6] sections
	442	@poddata = ();
	443	for ($i = 0, $j = -1; $i <= $#filedata; $i++) {
	444	$j++ if ($filedata[$i] =~ /^\s*=head[1-6]/);
	445	if ($j >= 0) {
	446	$poddata[$j] = "" unless defined $poddata[$j];
	447	$poddata[$j] .= "\n$filedata[$i]" if $j >= 0;
	448	}
	449	}
	450
	451	# create list of =head[1-6] sections so that we can rewrite
	452	# L<> links as necessary.
3d8876fc	453	my %heads = ();
54310121	454	foreach $i (0..$#poddata) {
	455	$heads{htmlize($1)} = 1 if $poddata[$i] =~ /=head[1-6]\s+(.*)/;
	456	}
	457
	458	# create a directory of a similar name and store all the
	459	# files in there
	460	$pod =~ s,./(.),$1,; # get the last part of the name
3d8876fc	461	my $dir = $pod;
54310121	462	$dir =~ s/\.pod//g;
	463	push(@$splitdirs, "$poddir/$dir");
	464	mkdir("$poddir/$dir", 0755) \|\|
	465	die "$0: could not create directory $poddir/$dir: $!\n"
	466	unless -d "$poddir/$dir";
	467
	468	$poddata[0] =~ /^\s=head[1-6]\s+(.)/;
	469	$section = "";
	470	$nextsec = $1;
	471
	472	# for each section of the file create a separate pod file
	473	for ($i = 0; $i <= $#poddata; $i++) {
	474	# determine the "prev" and "next" links
	475	$prevsec = $section;
	476	$section = $nextsec;
	477	if ($i < $#poddata) {
	478	$poddata[$i+1] =~ /^\s=head[1-6]\s+(.)/;
	479	$nextsec = $1;
	480	} else {
	481	$nextsec = "";
	482	}
	483
	484	# determine an appropriate filename (this must correspond with
	485	# what pod2html will try and guess)
	486	# $poddata[$i] =~ /^\s=head[1-6]\s+(.)/;
	487	$file = "$dir/" . htmlize($section) . ".pod";
	488
	489	# create the new .pod file
	490	print "\tcreating $poddir/$file\n" if $verbose;
	491	open(SPLITOUT, ">$poddir/$file") \|\|
	492	die "$0: error opening $poddir/$file for output: $!\n";
	493	$poddata[$i] =~ s,L<([^<>]*)>,
	494	defined $heads{htmlize($1)} ? "L<$dir/$1>" : "L<$1>"
	495	,ge;
	496	print SPLITOUT $poddata[$i]."\n\n";
	497	print SPLITOUT "=over 4\n\n";
	498	print SPLITOUT "=item *\n\nBack to L<$dir/\"$prevsec\">\n\n" if $prevsec;
	499	print SPLITOUT "=item *\n\nForward to L<$dir/\"$nextsec\">\n\n" if $nextsec;
	500	print SPLITOUT "=item *\n\nUp to L<$dir>\n\n";
	501	print SPLITOUT "=back\n\n";
	502	close(SPLITOUT) \|\|
	503	die "$0: error closing $poddir/$file: $!\n";
	504	}
	505	}
	506
	507
	508	#
	509	# installdir - takes care of converting the .pod and .pm files in the
	510	# current directory to .html files and then installing those.
	511	#
	512	sub installdir {
	513	my($dir, $recurse, $podroot, $splitdirs, $ignore) = @_;
	514	my(@dirlist, @podlist, @pmlist, $doindex);
	515
	516	@dirlist = (); # directories to recurse on
	517	@podlist = (); # .pod files to install
	518	@pmlist = (); # .pm files to install
	519
	520	# should files in this directory get an index?
	521	$doindex = (grep($_ eq "$podroot/$dir", @$splitdirs) ? 0 : 1);
	522
	523	opendir(DIR, "$podroot/$dir")
	524	\|\| die "$0: error opening directory $podroot/$dir: $!\n";
	525
526	# find the directories to recurse on
fe4c6be1	527	@dirlist = map { if ($^O eq 'VMS') {/^(.*)\.dir$/i; "$dir/$1";} else {"$dir/$_";}}
54310121	528	grep(-d "$podroot/$dir/$_" && !/^\.{1,2}/, readdir(DIR)) if $recurse;
	529	rewinddir(DIR);
	530
	531	# find all the .pod files within the directory
	532	@podlist = map { /^(.*)\.pod$/; "$dir/$1" }
	533	grep(! -d "$podroot/$dir/$_" && /\.pod$/, readdir(DIR));
	534	rewinddir(DIR);
	535
	536	# find all the .pm files within the directory
	537	@pmlist = map { /^(.*)\.pm$/; "$dir/$1" }
	538	grep(! -d "$podroot/$dir/$_" && /\.pm$/, readdir(DIR));
	539
	540	closedir(DIR);
	541
	542	# recurse on all subdirectories we kept track of
	543	foreach $dir (@dirlist) {
	544	installdir($dir, $recurse, $podroot, $splitdirs, $ignore);
	545	}
	546
	547	# install all the pods we found
3d8876fc	548	foreach my $pod (@podlist) {
54310121	549	# check if we should ignore it.
	550	next if grep($_ eq "$podroot/$pod.pod", @$ignore);
	551
	552	# check if a .pm files exists too
	553	if (grep($_ eq "$pod.pm", @pmlist)) {
	554	print "$0: Warning both `$podroot/$pod.pod' and "
	555	. "`$podroot/$pod.pm' exist, using pod\n";
	556	push(@ignore, "$pod.pm");
	557	}
	558	runpod2html("$pod.pod", $doindex);
	559	}
	560
	561	# install all the .pm files we found
3d8876fc	562	foreach my $pm (@pmlist) {
54310121	563	# check if we should ignore it.
	564	next if grep($_ eq "$pm.pm", @ignore);
	565
	566	runpod2html("$pm.pm", $doindex);
	567	}
	568	}
	569
	570
	571	#
	572	# runpod2html - invokes pod2html to convert a .pod or .pm file to a .html
	573	# file.
	574	#
	575	sub runpod2html {
	576	my($pod, $doindex) = @_;
	577	my($html, $i, $dir, @dirs);
	578
	579	$html = $pod;
	580	$html =~ s/\.(pod\|pm)$/.html/g;
	581
	582	# make sure the destination directories exist
	583	@dirs = split("/", $html);
	584	$dir = "$htmldir/";
	585	for ($i = 0; $i < $#dirs; $i++) {
	586	if (! -d "$dir$dirs[$i]") {
	587	mkdir("$dir$dirs[$i]", 0755) \|\|
	588	die "$0: error creating directory $dir$dirs[$i]: $!\n";
	589	}
	590	$dir .= "$dirs[$i]/";
	591	}
	592
	593	# invoke pod2html
	594	print "$podroot/$pod => $htmldir/$html\n" if $verbose;
	595	#system("./pod2html",
	596	Pod::Html'pod2html(
	597	#Pod::Html'pod2html($pod2html,
5a039dd3	598	"--htmldir=$htmldir",
54310121	599	"--htmlroot=$htmlroot",
	600	"--podpath=".join(":", @podpath),
	601	"--podroot=$podroot", "--netscape",
7c82de66	602	"--header",
54310121	603	($doindex ? "--index" : "--noindex"),
	604	"--" . ($recurse ? "" : "no") . "recurse",
	605	($#libpods >= 0) ? "--libpods=" . join(":", @libpods) : "",
	606	"--infile=$podroot/$pod", "--outfile=$htmldir/$html");
	607	die "$0: error running $pod2html: $!\n" if $?;
	608	}
	609
	610	sub htmlize { htmlify(0, @_) }