#!/usr/bin/perl --
use strict;
use warnings;
use HTML::Encoding 'encoding_from_http_message';
use WWW::Mechanize;
use Encode;
use HTML::Tree;
my $file = shift or die "
Usage: xmlfixup.pl file:in.xml > out.xml
xmlfixup.pl http://example.com/foo.xml > out.utf8.xml
";
my $resp = WWW::Mechanize->new( autocheck => 1 )->get( $file );
my $enco = encoding_from_http_message( $resp );
my $utf8;
if( $enco ) {
$utf8 = decode( $enco => $resp->content );
} else {
$utf8 = $resp->content;
}
my $t = HTML::TreeBuilder->new(
qw(
ignore_unknown 0
no_space_compacting 1
ignore_ignorable_whitespace 0
implicit_tags 0
no_expand_entities 1
store_comments 1
store_pis 1
)
);
#~ $t->xml_mode( 1 );
$t->parse_content( $utf8 );
binmode STDOUT, ':utf8';
print $_->as_XML for $t->content_list;
__END__
