#!/usr/bin/perl
use strict;
use warnings;
use HTML::TreeBuilder;
use LWP::Simple;
use URI;
use Getopt::Std;
my %opts;
getopts('i', \%opts);
my ($tag, $href) = exists $opts{i} ? ('img', 'src') : ('a', 'href');
if (@ARGV < 1 or @ARGV > 2)
{
die "Invalid number of arguments";
}
my ($url, $regex) = @ARGV;
my $uri = URI->new($url);
my $tree;
$regex ||= '.';
if (-f $url)
{
$tree = HTML::TreeBuilder->new_from_file($url);
}
else
{
my $content = get($uri);
die unless defined $content;
$tree = HTML::TreeBuilder->new_from_content($content);
}
die unless defined $tree;
foreach my $link ($tree->look_down(_tag => $tag, $href => qr{$regex}))
{
my $link_url = URI->new_abs($link->attr($href), $uri);
print $link_url->as_string, "\n";
}
|