Sub-domain parsing

00-ReadMe.md

Sub-domain parsing

Sometimes you have domains like a.really.long.url.thing.com and you need to get the base domain name. Yes, you could run:

$domains = explode('.', $host_name);
$target = array_slice($domains, -2, 1);
$root = $target[0];

But what if it is .co.uk? That is when you need this slightly badly written script.

It is worth noting that fr.co.uk is a valid URL despite fr, co and uk all being TLDs. So therefore, a.thing.fr.co.uk would be converted to thing.fr.co.uk instead of fr.co.uk. Obviously there are bugs.

01-ExampleUsage.php

<?php
// Fetch domains - might want to cache it
SubDomains::initTLds();


foreach(SubDomains::getSubDomains('some.thing.which.is.a.long.subdomain.bbc.co.uk') as $domain){
  echo sprintf('<a href="http://%s">%s</a>', $domain, $domain) . "\n";
  $contents = @file_get_contents('http://'.$domain);
  if($contents !== false){
    echo htmlspecialchars(substr($contents, 0, 255));
    break;
  }
  echo '<hr />';
}

02-SubDomains.php

<?php
class SubDomains {
  protected static $tlds = array();

	const TLD_URL = 'http://data.iana.org/TLD/tlds-alpha-by-domain.txt';
	const COMMENT_CHARACTER = '#';

	/**
	 * Fetches domains from data.iana site
	 */
	public static function initTLds() {
		$file = fopen(static::TLD_URL, 'r');
		if(!$file) throw new Exception('Unable to open TLD source');

		while(!feof($file)){
			$line = trim(fgets($file));

			if(substr($line, 0, 1) == static::COMMENT_CHARACTER) continue;

			$Tlds[] = strtolower($line);
		}
	}

	/**
	 * @param string $domain Domain of the URL
	 * @return array Each subdomain item in turn
	 */
	public static function getSubDomains($domain){
		$parts = explode('.', $domain);
		$current_domain = static::createDomain($parts);

		$domains = array($current_domain);
		for($i = count($parts); $i > 0; $i--){
      $part = array_pop($parts);
			$current_domain = $part . '.' . $current_domain;
			$domains[] = $current_domain;
		}

		return array_reverse($domains);
	}

	/**
	 * @param string $domain Domain of the URL
	 * @return string The root domain (www.bbc.co.uk maps to bbc.co.uk)
	 */
	public static function getRootDomain($domain){
		$parts = explode('.', $domain);
		return static::createDomain($domain);
	}

	/**
	 * @param string $subdomain The string to check
	 * @return string If the string is a TLD
	 */
	public static function isTld($subdomain){
		return in_array($subdomain, static::$tlds);
	}

	/**
	 * Finds the first instance of a non-tld url
	 * @param array $parts Parts of the domain
	 * @return string All elements of the domain upuntil that point
	 */
	protected static function createDomain(&$parts){
		$current_domain = array_pop($parts);
		$length = count($parts);

		for($i = 0; $i < $length; $i++){
			$subdomain = array_pop($parts);
			$current_domain = $subdomain . '.' . $current_domain;

			if(!static::isTld($subdomain)) break;
		}

		return $current_domain;
	}
}