Untitled

use HTML::Parser::XML;

use Benchmark;

my $html = '<html>';

$html ~= '<div class="header">';
$html ~= '<div><a href="piko.html>piko</a></div>' x 10000;
$html ~= '<div><p>fake content</p>' x 10;
$html ~= '</div>' x 11;

$html ~= '<div class="content">';
$html ~= '<img src="nyo.png"/>' x 200;
$html ~= '<div class="content_footer"><p>fake content</p>' x 10;
$html ~= '</div>' x 10;
$html ~= '<div class="inside_content">';
$html ~= '<a href="nyo.html">nyo</a>' x 3;
$html ~= '</div> </div>';

$html ~= '<div><a href="piko.html>piko</a></div>' x 1000;
$html ~= '<div><p>fake content</p>' x 10;
$html ~= '</div>' x 10;

$html ~= '</html>';

say $html;

my $parser = HTML::Parser::XML.new;

$parser.parse($html);

my %result = timethese(10, {
        "nofilter" => sub {$parser.parse($html)},
        "partial filter" => sub {$parser.parse($html, :TAG<div>, :class<content>)},
        "specific filter" => sub {$parser.parse($html, :TAG<div>, :class<inside_content>)} } );

say %result.perl;
say map { $_ => %result{$_}[3] }, sort keys %result;