Screenscraper

class ScreenScraper
{
	var $url;
	var $data;
	var $xpaths;
 
	// Download options
	var $cookie = 'cookie.txt';
	var $useragent = "Mozilla/5.0 (Windows; U; Windows NT 6.0; en-GB; rv:1.9.1.1) Gecko/20090715 Firefox/3.5.1 GTB5 (.NET CLR 4.0.20506)";
	var $cache_time = '300';
	var $cache_dir = 'cache/';
	var $cache_file;
 
	var $referer;
	var $postParams;
 
 
	function getData($url = null)
	{
		if (!empty($url))
		{
			$this->url = $url;
		}
 
		// Make sure the cookie file exists
		if (!file_exists($this->cookie))
		{
			file_put_contents($this->cookie, '');
		}
 
		if (!empty($this->url))
		{
			$this->cache_file = realpath($this->cache_dir). '/'. sha1($this->url. $this->cookie. $this->postParams. $this->useragent);
 
			// Cache the page...
			if (file_exists($this->cache_file) && (time() - @filemtime($this->cache_file) < $this->cache_time))
			{
				$file = file_get_contents($this->cache_file);
			} else {
				$ch = curl_init($this->url);
 
				$ch_opts = array(
								CURLOPT_URL => $this->url,
								CURLOPT_FOLLOWLOCATION => true,
								CURLOPT_RETURNTRANSFER => true,
								CURLOPT_MAXREDIRS => 3,
								CURLOPT_COOKIEFILE => $this->cookie,
								CURLOPT_COOKIEJAR => $this->cookie,
								CURLOPT_USERAGENT => $this->useragent,
							);
 
				if (!empty($this->postParams))
				{
					$ch_opts += array(CURLOPT_POSTFIELDS => $this->postParams);
				}
				if (!empty($this->referer))
				{
					$ch_opts += array(CURLOPT_REFERER => $this->referer);
				}
 
				curl_setopt_array($ch, $ch_opts);
 
				$file = curl_exec($ch);
				curl_close ($ch);
				file_put_contents($this->cache_file, $file);
			}
 
			$this->data = $file;
			$file = null;
 
			return !empty($this->data);
		} else {
			return false;
		}
	}
 
	function query($queries)
	{
		if (!empty($this->data))
		{
			$doc = new DomDocument();
			@$doc->loadHTML($this->data);
			$domXPath = new DomXPath($doc);
 
			foreach($queries as $label => $query)
			{
 
				$nodes = null;
 
				if (is_array($query))
				{
					$items = $domXPath->query($query['xpath']);
				}
				else
				{
					$items = $domXPath->query($query);
				}
 
				$count = 0;
 
				foreach($items as $item)
				{
					if (!empty($item))
					{
						if (is_array($query) && !isset($query['offset']) || $count >= $query['offset'])
						{
							$node = trim(preg_replace("%(?:\n\r|\n|\s{2,})%is", " ", str_replace('&nbsp;', ' ', $this->__xpathHTML($item))));
 
							if (is_array($query) && !empty($query['match']) && isset($query['match']))
							{
								preg_match_all($query['match'], $node, $matches);
 
								// Need a better array algorithm, this is messy...
								if (!empty($matches))
								{
									if (count($matches) >= 1)
									{
										if (count($matches[1]) <= 1)
										{
											$nodes[] = $matches[1][0];
										}
										else
										{
											$nodes[] = $matches[1];
										}
									}
									else
									{
										if (isset($matches[0][0]))
										{
											$nodes[] = $matches[0][0];
										}
										else
										{
											$nodes[] = $matches[0];
										}
 
									}
								}
 
							}
							else
							{
								$nodes[] = $node;
							}
						}
					}
 
					if (is_array($query) && isset($query['limit']) && ($count + 1) >= $query['limit'])
					{
						break;
					}
					else 
					{
						$count++;
					}
				}
 
				if (count($nodes) <= 1)
				{
					$data[$label] = $nodes[0];
				}
				else
				{
					$data[$label] = $nodes;
				}
			}
 
			return $data;
		} else {
			return false;
		}
	}
 
 
	function __xpathHTML($element)
	{
		$children = $element->childNodes;
		foreach ($children as $child)
		{
			$domdoc = new DOMDocument();
			$domdoc->appendChild($domdoc->importNode($child, true));
			$innerHTML.=trim($domdoc->saveHTML());
		}
		return $innerHTML;
	}
}