#!/usr/bin/perl
use Jcode;
#use Data::Dumper;

sub convert_jp {
	my ($text, $tocode, $fromcode) = @_;

	if($tocode ne '' and $tocode ne $fromcode){
		Jcode::convert(\$text, $tocode, $fromcode);
	}
	($text);
}

sub urldecode{
	my ($text) = @_;

	$text =~ s/\\x([0-9a-f][0-9a-f])/\%$1/gi;
	$text =~ s/%([0-9a-f][0-9a-f])/pack("C",hex($1))/egi;
	($text);
}

sub xml_to_array {
	my $xml = shift;
	my $tag_regex_ = q{<[^"'<>]*(?:"[^"]*"[^"'<>]*|'[^']*'[^"'<>]*)*(?:>|(?=<)|$(?!\n))};
	my $comment_regex = '<!(?:--[^-]*-(?:[^-]+-)*?-(?:[^>-]*(?:-[^>-]+)*?)??)*(?:>|$(?!\n)|--.*$)';
	my $cdata_regex   = q{<!\[CDATA\[.*?(:?\]\]>|$(?!\n))};
	my $tag_regex     = qq{$cdata_regex|$comment_regex|$tag_regex_};
	my $text_regex    = q{[^<]*};
	my @result;
	while ($xml =~ /($text_regex)($tag_regex)?/gso) {
		my ($text, $tag) = ($1, $2);
		last if ($text eq '' and $tag eq '');
		push (@result, grep {$_ ne ''} ($text, $tag));
	}
	return @result;
}

$charset2jcode{'ISO-2022-JP'}='jis';
$charset2jcode{'SHIFT_JIS'}='sjis';
$charset2jcode{'EUC-JP'}='euc';
$charset2jcode{'UTF-8'}='utf8';

$charset = "";
$data = join('', <>);
if($data =~ /<meta http-equiv="Content-Type" content="text\/html; charset=(.*)"/i){
	$charset = $1;
	$charset =~ tr/a-z/A-Z/;
}
$srccode = $charset2jcode{$charset};
if($srccode ne 'euc'){
	$eucdata = Jcode->new($data,$srccode)->euc;
}else{
	$eucdata = $data;
}
#print $eucdata;
@xml = &xml_to_array($eucdata);
#print Dumper([@xml]);
foreach(@xml){
	if(/^</){
		print convert_jp($_, $srccode, 'euc');
	}else{
		print convert_jp(urldecode($_), $srccode, '');
	}
}