#!/usr/local/bin/perl

use HTML::TokeParser;
use LWP::Simple qw(!head);
use CGI;

my $path = '/full/path/to/database/dir';
my $site = 'http://www.domain.com'; #base URL that MUST be matched
$site = quotemeta($site);

my $IN = new CGI;

$IN->header("text/html");

open (IN, $path . "/links.db");
open (OUT, ">" . $path . "/recip.db");
while (<IN>)
{
next if /^#/;
next if /^\s*$/;

my ($id, $title, $url, @junk) = split /\|/;

my $file = get($url);
my $f = 0;

if ($file)
{
my $p = HTML::TokeParser->new(\$file);

while (my $link = $p->get_token("a"))
{
if ($link->[0]{href} =~ /^$site/i)
{
$f++;
last;
}
}

print OUT "$id|$url|$f\n";
print "$id: " . ($f ? "found" : "not found") . "\n";
}
else
{
print OUT "$id|$url|dead\n";
print "$id: dead\n";
}
}
close OUT;
close IN;