http = new p3k\HTTP(); if(Config::$cache && class_exists('Memcache')) { $this->mc = new Memcache(); $this->mc->addServer('127.0.0.1'); } } public static function debug($msg, $header='X-Parse-Debug') { syslog(LOG_INFO, $msg); if(array_key_exists('REMOTE_ADDR', $_SERVER)) header($header . ": " . $msg); } private function respond(Response $response, $code, $params, $headers=[]) { $response->setStatusCode($code); foreach($headers as $k=>$v) { $response->headers->set($k, $v); } $response->headers->set('Content-Type', 'application/json'); $opts = JSON_UNESCAPED_SLASHES; if($this->_pretty) $opts += JSON_PRETTY_PRINT; $response->setContent(json_encode($params, $opts)."\n"); return $response; } private static function toHtmlEntities($input) { return mb_convert_encoding($input, 'HTML-ENTITIES', mb_detect_encoding($input)); } public function parse(Request $request, Response $response) { if($request->get('timeout')) { // We might make 2 HTTP requests, so each request gets half the desired timeout $this->http->timeout = $request->get('timeout') / 2; } if($request->get('max_redirects')) { $this->http->max_redirects = (int)$request->get('max_redirects'); } if($request->get('pretty')) { $this->_pretty = true; } $url = $request->get('url'); $html = $request->get('html'); if(!$url && !$html) { return $this->respond($response, 400, [ 'error' => 'missing_url', 'error_description' => 'Provide a URL or HTML to fetch' ]); } if($html) { // If HTML is provided in the request, parse that, and use the URL provided as the base URL for mf2 resolving $result['body'] = $html; $result['url'] = $url; } else { // Attempt some basic URL validation $scheme = parse_url($url, PHP_URL_SCHEME); if(!in_array($scheme, ['http','https'])) { return $this->respond($response, 400, [ 'error' => 'invalid_url', 'error_description' => 'Only http and https URLs are supported' ]); } $host = parse_url($url, PHP_URL_HOST); if(!$host) { return $this->respond($response, 400, [ 'error' => 'invalid_url', 'error_description' => 'The URL provided was not valid' ]); } $url = \normalize_url($url); // Check if this is a Twitter URL and if they've provided API credentials, use the API if(preg_match('/https?:\/\/(?:mobile\.twitter\.com|twitter\.com|twtr\.io)\/(?:[a-z0-9_\/!#]+statuse?s?\/([0-9]+)|([a-zA-Z0-9_]+))/i', $url, $match)) { return $this->parseTwitterURL($request, $response, $url, $match); } if($host == 'github.com') { return $this->parseGitHubURL($request, $response, $url); } // Now fetch the URL and check for any curl errors // Don't cache the response if a token is used to fetch it if($this->mc && !$request->get('token')) { $cacheKey = 'xray-'.md5($url); if($cached=$this->mc->get($cacheKey)) { $result = json_decode($cached, true); self::debug('using HTML from cache', 'X-Cache-Debug'); } else { $result = $this->http->get($url, [self::useragent()]); $cacheData = json_encode($result); // App Engine limits the size of cached items, so don't cache ones larger than that if(strlen($cacheData) < 1000000) $this->mc->set($cacheKey, $cacheData, MEMCACHE_COMPRESSED, $this->_cacheTime); } } else { $headers = [self::useragent()]; if($request->get('token')) { $headers[] = 'Authorization: Bearer ' . $request->get('token'); } $result = $this->http->get($url, $headers); } if($result['error']) { return $this->respond($response, 200, [ 'error' => $result['error'], 'error_description' => $result['error_description'], 'url' => $result['url'], 'code' => $result['code'] ]); } if(trim($result['body']) == '') { if($result['code'] == 410) { // 410 Gone responses are valid and should not return an error return $this->respond($response, 200, [ 'data' => [ 'type' => 'unknown' ], 'url' => $result['url'], 'code' => $result['code'] ]); } return $this->respond($response, 200, [ 'error' => 'no_content', 'error_description' => 'We did not get a response body when fetching the URL', 'url' => $result['url'], 'code' => $result['code'] ]); } // Check for HTTP 401/403 if($result['code'] == 401) { return $this->respond($response, 200, [ 'error' => 'unauthorized', 'error_description' => 'The URL returned "HTTP 401 Unauthorized"', 'url' => $result['url'], 'code' => 401 ]); } if($result['code'] == 403) { return $this->respond($response, 200, [ 'error' => 'forbidden', 'error_description' => 'The URL returned "HTTP 403 Forbidden"', 'url' => $result['url'], 'code' => 403 ]); } } // Check for known services $host = parse_url($result['url'], PHP_URL_HOST); if(in_array($host, ['www.instagram.com','instagram.com'])) { list($data, $parsed) = Formats\Instagram::parse($result['body'], $result['url'], $this->http); if($request->get('include_original')) $data['original'] = $parsed; $data['url'] = $result['url']; $data['code'] = $result['code']; return $this->respond($response, 200, $data); } if($host == 'xkcd.com' && parse_url($url, PHP_URL_PATH) != '/') { $data = Formats\XKCD::parse($result['body'], $url); $data['url'] = $result['url']; $data['code'] = $result['code']; return $this->respond($response, 200, $data); } // attempt to parse the page as HTML $doc = new DOMDocument(); @$doc->loadHTML(self::toHtmlEntities($result['body'])); if(!$doc) { return $this->respond($response, 200, [ 'error' => 'invalid_content', 'error_description' => 'The document could not be parsed as HTML' ]); } $xpath = new DOMXPath($doc); // Check for meta http equiv and replace the status code if present foreach($xpath->query('//meta[translate(@http-equiv,\'STATUS\',\'status\')=\'status\']') as $el) { $equivStatus = ''.$el->getAttribute('content'); if($equivStatus && is_string($equivStatus)) { if(preg_match('/^(\d+)/', $equivStatus, $match)) { $result['code'] = (int)$match[1]; } } } // If a target parameter was provided, make sure a link to it exists on the page if($target=$request->get('target')) { $found = []; if($target) { self::xPathFindNodeWithAttribute($xpath, 'a', 'href', function($u) use($target, &$found){ if($u == $target) { $found[$u] = null; } }); self::xPathFindNodeWithAttribute($xpath, 'img', 'src', function($u) use($target, &$found){ if($u == $target) { $found[$u] = null; } }); self::xPathFindNodeWithAttribute($xpath, 'video', 'src', function($u) use($target, &$found){ if($u == $target) { $found[$u] = null; } }); self::xPathFindNodeWithAttribute($xpath, 'audio', 'src', function($u) use($target, &$found){ if($u == $target) { $found[$u] = null; } }); } if(!$found) { return $this->respond($response, 200, [ 'error' => 'no_link_found', 'error_description' => 'The source document does not have a link to the target URL', 'url' => $result['url'], 'code' => $result['code'], ]); } } // If the URL has a fragment ID, find the DOM starting at that node and parse it instead $html = $result['body']; $fragment = parse_url($url, PHP_URL_FRAGMENT); if($fragment) { $fragElement = self::xPathGetElementById($xpath, $fragment); if($fragElement) { $html = $doc->saveHTML($fragElement); $foundFragment = true; } else { $foundFragment = false; } } // Now start pulling in the data from the page. Start by looking for microformats2 $mf2 = mf2\Parse($html, $result['url']); if($mf2 && count($mf2['items']) > 0) { $data = Formats\Mf2::parse($mf2, $result['url'], $this->http); if($data) { if($fragment) { $data['info'] = [ 'found_fragment' => $foundFragment ]; } if($request->get('include_original')) $data['original'] = $html; $data['url'] = $result['url']; // this will be the effective URL after following redirects $data['code'] = $result['code']; return $this->respond($response, 200, $data); } } // TODO: look for other content like OEmbed or other known services later return $this->respond($response, 200, [ 'data' => [ 'type' => 'unknown', ], 'url' => $result['url'], 'code' => $result['code'] ]); } private static function xPathFindNodeWithAttribute($xpath, $node, $attr, $callback) { foreach($xpath->query('//'.$node.'[@'.$attr.']') as $el) { $v = $el->getAttribute($attr); $callback($v); } } private static function xPathGetElementById($xpath, $id) { $element = null; foreach($xpath->query("//*[@id='$id']") as $el) { $element = $el; } return $element; } private function parseTwitterURL(&$request, &$response, $url, $match) { $fields = ['twitter_api_key','twitter_api_secret','twitter_access_token','twitter_access_token_secret']; $creds = []; foreach($fields as $f) { if($v=$request->get($f)) $creds[$f] = $v; } $data = false; if(count($creds) == 4) { list($data, $parsed) = Formats\Twitter::parse($url, $match[1], $creds); } elseif(count($creds) > 0) { // If only some Twitter credentials were present, return an error return $this->respond($response, 400, [ 'error' => 'missing_parameters', 'error_description' => 'All 4 Twitter credentials must be included in the request' ]); } else { // Accept Tweet JSON and parse that if provided $json = $request->get('json'); if($json) { list($data, $parsed) = Formats\Twitter::parse($url, $match[1], null, $json); } // Skip parsing from the Twitter API if they didn't include credentials } if($data) { if($request->get('include_original')) $data['original'] = $parsed; $data['url'] = $url; $data['code'] = 200; return $this->respond($response, 200, $data); } else { return $this->respond($response, 200, [ 'data' => [ 'type' => 'unknown' ], 'url' => $url, 'code' => 0 ]); } } private function parseGitHubURL(&$request, &$response, $url) { $fields = ['github_access_token']; $creds = []; foreach($fields as $f) { if($v=$request->get($f)) $creds[$f] = $v; } $data = false; $json = $request->get('json'); if($json) { // Accept GitHub JSON and parse that if provided list($data, $json) = Formats\GitHub::parse($this->http, $url, null, $json); } else { // Otherwise fetch the post unauthenticated or with the provided access token list($data, $json) = Formats\GitHub::parse($this->http, $url, $creds); } if($data) { if($request->get('include_original')) $data['original'] = $json; $data['url'] = $url; $data['code'] = 200; return $this->respond($response, 200, $data); } else { return $this->respond($response, 200, [ 'data' => [ 'type' => 'unknown' ], 'url' => $url, 'code' => 0 ]); } } }