use warnings;
use strict;

use Encode;
use HTML::Entities;
use XML::Twig;

my $objinfo_parser = XML::Twig->new (
    twig_handlers => { 'Product[@Id = "most3usb20on"]' => \&_test_handler }
);

$objinfo_parser->parse (decode_entities (join '', <DATA>) );
#$objinfo_parser->parseurl ('http://www.3btech.net/objinfo.xml');

sub _test_handler {
    my ($twig, $elt) = @_;
    
    my $str = substr ($elt->field ('Description'), -12);

    print "The string below passes a utf8 well-formedness test, why?\n" 
        if Encode::is_utf8 ($str, 1);
    
    print "$str\n\n";
    for (unpack ('U*', $str)) {
        printf "0x%X\n", $_;
    }
    
    print "\n";
    exit;
}

# The xml snippet below is taken directly from the above url
# (the url will take about 15s to download and parse)
# HTML encoded to preserve offending characters

__DATA__
<?xml version="1.0" encoding="UTF-8"?>
    <Product Id="most3usb20on">
      <Description>Moving Star 3.5" USB 2.0 One Button Backup Aluminum Hard Drive Enclosure &#xC2;&#x96; Black</Description>
    </Product>