/** * Function to get the contents of the URL. It will get the contents using Gzip compression if possible in * order to save bandwidth. It uses the HTTP Client class from Simon Willison to do the dirty work. * * More information about the HTTP client class can be found on: http://scripts.incutio.com/httpclient/ * * If it fails to retrieve the data, it will raise a fatal error. * * By default, it will cache the downloaded data based on the etag and last-modified headers. The cache files * are stored in the temp directory of the Yellow Duck framework and have the extension "wch". You can delete * these automatically as they will be recreated on the fly if needed. * * For configuring the cache, there are two configuration variables you can redefine if needed: * YD_HTTP_CACHE_TIMEOUT: the lifetime of the cache in seconds (default: 3600). * YD_HTTP_CACHE_USEHEAD: if a HEAD HTTP request should be used to verify the cache validity (default: 1). * * @param $cache (optional) Indicate if the web content should be cached or not. By default, caching is * turned on. * @param $fail (optional) Whether to fail or not if the contents cannot be downloaded. Defaults to true. * * @returns Returns the contents of the URL. */ function getContents($cache = true, $fail = true) { // Check if caching is enabled $cacheFName = null; // Check the cache if ($cache == true) { // Include the filesystem library include_once YD_DIR_HOME_CLS . '/YDFileSystem.php'; // Check if we need to use the HTTP HEAD function if (YDConfig::get('YD_HTTP_CACHE_USEHEAD') == 1) { // Get the headers $headers = $this->getHeaders(); // Check if we have etag or last modified if (isset($headers['etag']) || isset($headers['last-modified'])) { $cacheFName = $this->getUrl(); if (isset($headers['etag'])) { $cacheFName .= $headers['etag']; } if (isset($headers['last-modified'])) { $cacheFName .= $headers['last-modified']; } if (isset($headers['content-length'])) { $cacheFName .= $headers['content-length']; } $cacheFName = YD_TMP_PRE . 'W_' . md5($cacheFName) . '.wch'; $cacheFName = YD_DIR_TEMP . '/' . $cacheFName; } } // If the cache filename is null, use the default one if ($cacheFName == null) { $cacheFName = YD_DIR_TEMP . '/' . YD_TMP_PRE . 'W_' . md5($this->getUrl()) . '.wch'; } // Use the cache file if any if (is_file($cacheFName)) { $file = new YDFSFile($cacheFName); $cacheValidTime = $file->getLastModified() + YDConfig::get('YD_HTTP_CACHE_TIMEOUT'); if (time() < $cacheValidTime) { return $file->getContents(); } } } // Create a new HTTP client $client = $this->_getHttpClient(); // Now send the request $result = @$client->doRequest(); // Check if there was a result if ($result == false) { if ($fail) { trigger_error('Failed to retrieve the data from the url "' . $this->getUrl() . '". ' . $client->getError(), YD_ERROR); } else { return false; } } else { $data = @$client->getContent(); } // Check if caching is enabled if ($cache == true) { // Save the cached data if ($cacheFName != null) { $dir = new YDFSDirectory(YD_DIR_TEMP); $dir->createFile($cacheFName, $data); } } // Return the data return $data; }