I have a problem using HTML::Parser. Maybe I'm missing something plain obviouse, but I don't see it. I have to say that it's the first time ever I use HTML::Parser.

What I thought I've done in the script shown below is:

1. invoked for every html file found the parser
2. dumped for every embed tag encountered all attributes

But what I really get is, for each embed tag in every file found, just one attribute.

Can someone please hit my head aginst the parts where I made a mistake?

#!/usr/bin/perl
use strict;
use warnings;
use HTML::Parser;
use File::Find;
use Data::Dumper;

my $hp= HTML::Parser->new(
    api_version => 3,
    start_h => [ \&start, 'tag,attr' ],
    # This should call, for every start tag the start subroutine
    # passign the tagname and a reference to the attribute hash
);

find( sub {
    return unless /\.html?$/;
    find_embed( $_ );
    # in find_embed I will invoke the parser
}, @ARGV) if scalar(@ARGV);

sub find_embed {
    my($filename)= @_;
    $hp->parse_file( $filename );
    # Now the parser is started
}    

sub start {
    my($tag, $attr)= @_;
    return unless $tag eq 'embed';
    # ignore every tag but embed
    print Dumper($attr);
    # Dump the attributes.
}
[download]