# -----------------------------------------------------------------------------------------------
# Author: Dean Stringer
# Description:
#	This script is intended to be run from the command-line
#	and reads STDIN for a list of HTML filenames, each of which
#	it parses looking for MetaData and links.
# -----------------------------------------------------------------------------------------------
use strict;
use HTML::TokeParser;
my $totalFiles = 0;
my $showLinks = 0;
my $lineBreak = "-----------------------------------------------------------------------------------\n";

print "\n\n" . $lineBreak . "Note: See end of output for Summary\n" . $lineBreak;

while (my $sourceFile=<STDIN>) {

	chop $sourceFile;
	my $content;	# container for file contents
	if (-e $sourceFile) {
		local($/) = undef; 
		open(IN, '<'.$sourceFile) or dieNice ("Can't open include file [ $sourceFile ]\n"); 
			$content = <IN>; 
		close IN; 
	} else {
		dieNice ("File '$sourceFile' does not exist.");
	} 

	$totalFiles++;

	# Show filename
	print "\n\n" . $lineBreak . "File: $sourceFile\n" . $lineBreak;

	# Get the Title tag
	print "Title: " . parseTag($content, 'title') . "\n";

	# Get a META tag
	# my $description = parseMeta($content, 'keywords');
	# $pageBody .= "<br><strong>Description</strong> is <em>$description</em>";

	# Get ALL META Data values
	my %metaValues = parseAllMeta($content);
	my ($key, $value);
	print "Meta Data:\n";

	while (($key,$value) = each %metaValues) {
		# replace CRs with Newlines
		$value =~ s/\r/\r\n/g;
		# remove white-space
		$value =~ s/\s{2,}/\s/g;
        print "\t" . ucfirst($key) . ":" . $value . "\n";
	}

	if ($showLinks) {
		my %links = parseLinks($content);
		print "Links to:";
		while (($key,$value) = each %links) {
		        print "\n\t$value";
		    }
	}
}

print "\n\n\n" . $lineBreak . "Summary:\n\n\tFiles Read: " . 
    $totalFiles . "\n" . $lineBreak;
exit;


sub parseTag {
# -------------------------------------------------------------- 
#	Grab just the Title
# -------------------------------------------------------------- 
	my $webPage = shift;
	my $tag2Get = shift;
	my $parser = HTML::TokeParser->new(\$webPage);
	$parser->get_tag($tag2Get);		# Grab the next tag
	return $parser->get_trimmed_text;	# Trim off the white space
}


sub parseMeta {
# -------------------------------------------------------------- 
#	Grab a specific META tag
# -------------------------------------------------------------- 
	my $webPage = shift;
	my $tag2Get = shift;
	my $tagValue = '';
	my $parser = HTML::TokeParser->new(\$webPage);
	while (my $token=$parser->get_tag("meta")) {
	if ($token->[1]{name}=~/$tag2Get/i) {
		$tagValue = $token->[1]{content};
		}
	}
	return $tagValue;
}

sub parseAllMeta {
# -------------------------------------------------------------- 
#	Grab ALL META tags and return as a Hash
#	Much quicker than creating a new instance each
#	time (as in parseMeta) for each Meta element.
# -------------------------------------------------------------- 
	my $webPage = shift;
	my $tag2Get = shift;
	my %metaValues;
	my $parser = HTML::TokeParser->new(\$webPage);
	while (my $token=$parser->get_tag("meta")) {
		# Make sure the Meta item has a 'name' attribute, otherwise it
		# could be another HTTP header fiddle like HTTP-EQUIV or someit.
		if ($token->[1]{name} ne "") {
			$metaValues{$token->[1]{name}} = $token->[1]{content};
			}
	}
	return %metaValues;
}


sub parseLinks {
# -------------------------------------------------------------- 
#	Grab ALL META tags and return as a Hash
#	Much quicker than creating a new instance each
#	time (as in parseMeta) for each Meta element.
# -------------------------------------------------------------- 
	my $webPage = shift;
	my (%metaValues, $label, $URL);
	my $parser = HTML::TokeParser->new(\$webPage);
	while (my $token=$parser->get_tag("a")) {
		$label = $parser->get_trimmed_text;	# Trim off the white space
		$URL = $token->[1]{href};
		if ($URL ne "") {
			$metaValues{$label} = $URL;
			}
	}
	return %metaValues;
}

sub dieNice {
# -------------------------------------------------------------- 
#	Bomb out with a supplied error message and show syntax
# -------------------------------------------------------------- 
	my $errorMessage = shift;
	print "\nERROR !!!  $errorMessage\n";
	print "Syntax is :   metageta.pl <filename>\n";
	exit;
}