protected function check(SblamURI $link)
{
$domain = $link->getDomain();
$min = max(2, count(explode('.', $domain)) + 1);
$res = $this->blacklist->check($link->getHostname());
if ($res >= $min) {
return $res + 1 - $min;
}
return 0;
}
function addURI(array &$urls, SblamURI $link, $prefix = '')
{
if ($link->isTLD()) {
return;
}
if ($hostname = $link->getHostname()) {
$hostname = preg_replace(array('!^www\\.!', '!\\d\\d+!'), array('', 'D'), $hostname);
// normalise digits! (block bulk registrations)
$urls[$prefix . $hostname] = true;
if ($domain = $link->getDomain()) {
$urls[$prefix . $domain] = true;
}
if ($p = $link->getPath()) {
$p = preg_replace('!^(/[^#]{1,7}[^#/\\?]{0,5}).*$!', '\\1', $p);
// shorten path. its mainly for getting real tinyurl adresses, not every spammy subpage out there
if ($p !== '/') {
$urls[$prefix . $hostname . $p] = true;
}
}
}
if (preg_match('!\\b(?:site:|https?://)([a-zA-Z0-9.-]+)!', urldecode($link->getPath()), $m)) {
$this->addURI($urls, new SblamURI('http://' . $m[1]), $prefix);
}
}