#!/usr/bin/perl -w
#
# Make a ranklist of tags from an XML file.
#
# Split before each start-tag/pi/comment/dcl
# Ditch all lines that aren't one of those
# Get rid of everything after the name
#
# ~2006-06-07: Written by Steven J. DeRose, sderoses@acm.org.
# 2007-08-15 sjd: Add -v, kill dcls and end-tags and comments.
# 2007-09-19 sjd: $version. Add -v, -alpha, -norank.
#

use strict;
use Getopt::Long;

my $version = "2010-09-12";

my $alpha   = 0;
my $norank  = 0;
my $quiet   = 0;
my $verbose = 0;


# Process options
#
Getopt::Long::Configure ("ignore_case");
my $result = GetOptions(
	"alpha!"            => \$alpha,
	"h|help|?"          => sub { system "perldoc counttags"; exit; },
	"q|quiet!"          => \$quiet,
	"norank"            => \$norank,
	"v|verbose+"        => \$verbose,
    "version"           => sub {
		die "Version of $version, by Steven J. DeRose.\n";
	}
	);

($result) || die "taglist: Bad options.\n";


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

if (!defined $ARGV[0] || !-f $ARGV[0]) {
	warn "Can't find file '$ARGV[0]'\n";
	showUsage();
	exit;
}

my $files = join(" ",@ARGV);

my $cmd = "cat $files"
	. " | sed 's/<!--[^>]*-->//g'"    # Remove (single-line) comments
	. " | sed 's/</~</g'"             # break lines before tags
	. " | tr '~' '\\012'"               
	. " | grep -v '<[?!/]'"           # delete end-tags, dcls and PIs
	. " | grep '^<'"                  # delete non-tag lines
	. " | sed 's/[ />].*\$//'"        # delete all but element type
	. " | sed 's/<//'"
	. " | grep -v '^[ \t]\$'"         # delete blank lines (shouldn't be any)
    . " | sort"
    . " | uniq";
if (!$norank) {
	$cmd .= " -c";
	if ($alpha) {
		$cmd .= " | sort -b -k 2,2";
	}
	else {
		$cmd .= " | sort -n";
	}
}

if ($verbose) {
	(my $pcmd = $cmd) =~ s/\|/\n    \|/g;
	warn "Running:\n$pcmd\n";
}
system "$cmd";
exit;


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

sub showUsage {
   print "

=head1 Usage

taglist [options] [filename(s)]

Makes a ranklist of element types from an XML file(s).


=head1 Options

=over

=item * B<-h>           Show this help and exit.

=item * B<-alpha>       Sort alphabetically by tag, not by frequency.

=item * B<-norank>      Don't show tag frequencies (implies -alpha).

=item * B<-v>           Show more detailed messages.

=item * B<-version>     Show version info and exit.

=back


=head1 Known bugs and limitations

Does not use an actual parser, so can be mislead by tags buried inside
multi-line comments, inside PIs, external entities that are not listed
on the command line, etc.


=head1 Related commands

    xmlstats: Calculates lots of statistics on an XML file, including the
    frequency list done here. 'taglist' is just quicker, and does not require
    that the file(s) be valid or even well-formed.


=head1 Ownership

This work by Steven J. DeRose is licensed under a Creative Commons 
Attribution-Share Alike 3.0 Unported License. For further information on
this license, see http://creativecommons.org/licenses/by-sa/3.0/.

The author's present email is sderose at acm.org.

For the most recent version, see http://www.derose.net/steve/utilities/.

=cut

";
}