X-Git-Url: https://git.mxchange.org/?a=blobdiff_plain;f=classes%2FFile_redirection.php;h=f128b3e07c84179e92369de9b732800da1000d16;hb=a954fd65ba00328cd1a76e620113d2f639340aaf;hp=5054b4ff215adc142669903a98a6ed95ca6e8916;hpb=793a6a1155996a7bcdc068fc891e7063934bdb19;p=quix0rs-gnu-social.git diff --git a/classes/File_redirection.php b/classes/File_redirection.php index 5054b4ff21..f128b3e07c 100644 --- a/classes/File_redirection.php +++ b/classes/File_redirection.php @@ -1,7 +1,7 @@ . */ -if (!defined('LACONICA')) { exit(1); } +if (!defined('STATUSNET') && !defined('LACONICA')) { exit(1); } require_once INSTALLDIR.'/classes/Memcached_DataObject.php'; require_once INSTALLDIR.'/classes/File.php'; require_once INSTALLDIR.'/classes/File_oembed.php'; -define('USER_AGENT', 'Laconica user agent / file probe'); - +define('USER_AGENT', 'StatusNet user agent / file probe'); /** * Table Definition for file_redirection */ -class File_redirection extends Memcached_DataObject +class File_redirection extends Memcached_DataObject { ###START_AUTOCODE /* the code below is auto generated do not remove the above tag */ public $__table = 'file_redirection'; // table name - public $id; // int(11) not_null primary_key group_by - public $url; // varchar(255) unique_key - public $file_id; // int(11) group_by - public $redirections; // int(11) group_by - public $httpcode; // int(11) group_by + public $url; // varchar(255) primary_key not_null + public $file_id; // int(4) + public $redirections; // int(4) + public $httpcode; // int(4) + public $modified; // timestamp() not_null default_CURRENT_TIMESTAMP /* Static get */ - function staticGet($k,$v=NULL) { return DB_DataObject::staticGet('File_redirection',$k,$v); } + function staticGet($k,$v=NULL) { return Memcached_DataObject::staticGet('File_redirection',$k,$v); } /* the code above is auto generated do not remove the tag below */ ###END_AUTOCODE - - - function _commonCurl($url, $redirs) { - $curlh = curl_init(); - curl_setopt($curlh, CURLOPT_URL, $url); - curl_setopt($curlh, CURLOPT_AUTOREFERER, true); // # setup referer header when folowing redirects - curl_setopt($curlh, CURLOPT_CONNECTTIMEOUT, 10); // # seconds to wait - curl_setopt($curlh, CURLOPT_MAXREDIRS, $redirs); // # max number of http redirections to follow - curl_setopt($curlh, CURLOPT_USERAGENT, USER_AGENT); - curl_setopt($curlh, CURLOPT_FOLLOWLOCATION, true); // Follow redirects - curl_setopt($curlh, CURLOPT_RETURNTRANSFER, true); - curl_setopt($curlh, CURLOPT_FILETIME, true); - curl_setopt($curlh, CURLOPT_HEADER, true); // Include header in output - return $curlh; + static function _commonHttp($url, $redirs) { + $request = new HTTPClient($url); + $request->setConfig(array( + 'connect_timeout' => 10, // # seconds to wait + 'max_redirs' => $redirs, // # max number of http redirections to follow + 'follow_redirects' => true, // Follow redirects + 'store_body' => false, // We won't need body content here. + )); + return $request; } - function _redirectWhere_imp($short_url, $redirs = 10, $protected = false) { + /** + * Check if this URL is a redirect and return redir info. + * + * Most code should call File_redirection::where instead, to check if we + * already know that redirection and avoid extra hits to the web. + * + * The URL is hit and any redirects are followed, up to 10 levels or until + * a protected URL is reached. + * + * @param string $in_url + * @return mixed one of: + * string - target URL, if this is a direct link or can't be followed + * array - redirect info if this is an *unknown* redirect: + * associative array with the following elements: + * code: HTTP status code + * redirects: count of redirects followed + * url: URL string of final target + * type (optional): MIME type from Content-Type header + * size (optional): byte size from Content-Length header + * time (optional): timestamp from Last-Modified header + */ + public function lookupWhere($short_url, $redirs = 10, $protected = false) { if ($redirs < 0) return false; - // let's see if we know this... - $a = File::staticGet('url', $short_url); - if (empty($a->id)) { - $b = File_redirection::staticGet('url', $short_url); - if (empty($b->id)) { - // we'll have to figure it out - } else { - // this is a redirect to $b->file_id - $a = File::staticGet($b->file_id); - $url = $a->url; - } - } else { - // this is a direct link to $a->url - $url = $a->url; - } - if (isset($url)) { - return $url; + if(strpos($short_url,'://') === false){ + return $short_url; } - - - - $curlh = File_redirection::_commonCurl($short_url, $redirs); - // Don't include body in output - curl_setopt($curlh, CURLOPT_NOBODY, true); - curl_exec($curlh); - $info = curl_getinfo($curlh); - curl_close($curlh); - - if (405 == $info['http_code']) { - $curlh = File_redirection::_commonCurl($short_url, $redirs); - curl_exec($curlh); - $info = curl_getinfo($curlh); - curl_close($curlh); + try { + $request = self::_commonHttp($short_url, $redirs); + // Don't include body in output + $request->setMethod(HTTP_Request2::METHOD_HEAD); + $response = $request->send(); + + if (405 == $response->getStatus()) { + // Server doesn't support HEAD method? Can this really happen? + // We'll try again as a GET and ignore the response data. + $request = self::_commonHttp($short_url, $redirs); + $response = $request->send(); + } + } catch (Exception $e) { + // Invalid URL or failure to reach server + common_log(LOG_ERR, "Error while following redirects for $short_url: " . $e->getMessage()); + return $short_url; } - if (!empty($info['redirect_count']) && File::isProtected($info['url'])) { - return File_redirection::_redirectWhere_imp($short_url, $info['redirect_count'] - 1, true); + if ($response->getRedirectCount() && File::isProtected($response->getUrl())) { + // Bump back up the redirect chain until we find a non-protected URL + return self::lookupWhere($short_url, $response->getRedirectCount() - 1, true); } - $ret = array('code' => $info['http_code'] - , 'redirects' => $info['redirect_count'] - , 'url' => $info['url']); + $ret = array('code' => $response->getStatus() + , 'redirects' => $response->getRedirectCount() + , 'url' => $response->getUrl()); - if (!empty($info['content_type'])) $ret['type'] = $info['content_type']; + $type = $response->getHeader('Content-Type'); + if ($type) $ret['type'] = $type; if ($protected) $ret['protected'] = true; - if (!empty($info['download_content_length'])) $ret['size'] = $info['download_content_length']; - if (isset($info['filetime']) && ($info['filetime'] > 0)) $ret['time'] = $info['filetime']; + $size = $response->getHeader('Content-Length'); // @fixme bytes? + if ($size) $ret['size'] = $size; + $time = $response->getHeader('Last-Modified'); + if ($time) $ret['time'] = strtotime($time); return $ret; } - function where($in_url) { - $ret = File_redirection::_redirectWhere_imp($in_url); - return $ret; - } + /** + * Check if this URL is a redirect and return redir info. + * If a File record is present for this URL, it is not considered a redirect. + * If a File_redirection record is present for this URL, the recorded target is returned. + * + * If no File or File_redirect record is present, the URL is hit and any + * redirects are followed, up to 10 levels or until a protected URL is + * reached. + * + * @param string $in_url + * @return mixed one of: + * string - target URL, if this is a direct link or a known redirect + * array - redirect info if this is an *unknown* redirect: + * associative array with the following elements: + * code: HTTP status code + * redirects: count of redirects followed + * url: URL string of final target + * type (optional): MIME type from Content-Type header + * size (optional): byte size from Content-Length header + * time (optional): timestamp from Last-Modified header + */ + public function where($in_url) { + // let's see if we know this... + $a = File::staticGet('url', $in_url); - function makeShort($long_url) { - $long_url = File_redirection::_canonUrl($long_url); - // do we already know this long_url and have a short redirection for it? - $file = new File; - $file_redir = new File_redirection; - $file->url = $long_url; - $file->joinAdd($file_redir); - $file->selectAdd('length(file_redirection.url) as len'); - $file->limit(1); - $file->orderBy('len'); - $file->find(true); - if (!empty($file->url) && (strlen($file->url) < strlen($long_url))) { - return $file->url; + if (!empty($a)) { + // this is a direct link to $a->url + return $a->url; + } else { + $b = File_redirection::staticGet('url', $in_url); + if (!empty($b)) { + // this is a redirect to $b->file_id + $a = File::staticGet('id', $b->file_id); + return $a->url; + } } - // if yet unknown, we must find a short url according to user settings - $short_url = File_redirection::_userMakeShort($long_url, common_current_user()); - return $short_url; + $ret = File_redirection::lookupWhere($in_url); + return $ret; } - function _userMakeShort($long_url, $user) { - if (empty($user)) { - // common current user does not find a user when called from the XMPP daemon - // therefore we'll set one here fix, so that XMPP given URLs may be shortened - $user->urlshorteningservice = 'ur1.ca'; - } - $curlh = curl_init(); - curl_setopt($curlh, CURLOPT_CONNECTTIMEOUT, 20); // # seconds to wait - curl_setopt($curlh, CURLOPT_USERAGENT, 'Laconica'); - curl_setopt($curlh, CURLOPT_RETURNTRANSFER, true); - - switch($user->urlshorteningservice) { - case 'ur1.ca': - require_once INSTALLDIR.'/lib/Shorturl_api.php'; - $short_url_service = new LilUrl; - $short_url = $short_url_service->shorten($long_url); - break; - - case '2tu.us': - $short_url_service = new TightUrl; - require_once INSTALLDIR.'/lib/Shorturl_api.php'; - $short_url = $short_url_service->shorten($long_url); - break; + /** + * Shorten a URL with the current user's configured shortening + * options, if applicable. + * + * If it cannot be shortened or the "short" URL is longer than the + * original, the original is returned. + * + * If the referenced item has not been seen before, embedding data + * may be saved. + * + * @param string $long_url + * @return string + */ + function makeShort($long_url) { - case 'ptiturl.com': - require_once INSTALLDIR.'/lib/Shorturl_api.php'; - $short_url_service = new PtitUrl; - $short_url = $short_url_service->shorten($long_url); - break; + $canon = File_redirection::_canonUrl($long_url); - case 'bit.ly': - curl_setopt($curlh, CURLOPT_URL, 'http://bit.ly/api?method=shorten&long_url='.urlencode($long_url)); - $short_url = current(json_decode(curl_exec($curlh))->results)->hashUrl; - break; + $short_url = File_redirection::_userMakeShort($canon); - case 'is.gd': - curl_setopt($curlh, CURLOPT_URL, 'http://is.gd/api.php?longurl='.urlencode($long_url)); - $short_url = curl_exec($curlh); - break; - case 'snipr.com': - curl_setopt($curlh, CURLOPT_URL, 'http://snipr.com/site/snip?r=simple&link='.urlencode($long_url)); - $short_url = curl_exec($curlh); - break; - case 'metamark.net': - curl_setopt($curlh, CURLOPT_URL, 'http://metamark.net/api/rest/simple?long_url='.urlencode($long_url)); - $short_url = curl_exec($curlh); - break; - case 'tinyurl.com': - curl_setopt($curlh, CURLOPT_URL, 'http://tinyurl.com/api-create.php?url='.urlencode($long_url)); - $short_url = curl_exec($curlh); - break; - default: - $short_url = false; + // Did we get one? Is it shorter? + if (!empty($short_url) && mb_strlen($short_url) < mb_strlen($long_url)) { + return $short_url; + } else { + return $long_url; } + } - curl_close($curlh); - - if ($short_url) { + function _userMakeShort($long_url) { + $short_url = common_shorten_url($long_url); + if (!empty($short_url) && $short_url != $long_url) { $short_url = (string)$short_url; // store it $file = File::staticGet('url', $long_url); if (empty($file)) { + // Check if the target URL is itself a redirect... $redir_data = File_redirection::where($long_url); - $file = File::saveNew($redir_data, $long_url); - $file_id = $file->id; - if (!empty($redir_data['oembed']['json'])) { - File_oembed::saveNew($redir_data['oembed']['json'], $file_id); + if (is_array($redir_data)) { + // We haven't seen the target URL before. + // Save file and embedding data about it! + $file = File::saveNew($redir_data, $long_url); + $file_id = $file->id; + if (!empty($redir_data['oembed']['json'])) { + File_oembed::saveNew($redir_data['oembed']['json'], $file_id); + } + } else if (is_string($redir_data)) { + // The file is a known redirect target. + $file = File::staticGet('url', $redir_data); + $file_id = $file->id; } } else { $file_id = $file->id; @@ -222,7 +224,7 @@ class File_redirection extends Memcached_DataObject } return $short_url; } - return $long_url; + return null; } function _canonUrl($in_url, $default_scheme = 'http://') { @@ -252,7 +254,7 @@ class File_redirection extends Memcached_DataObject } } - if (('ftp' == $p['scheme']) || ('http' == $p['scheme']) || ('https' == $p['scheme'])) { + if (('ftp' == $p['scheme']) || ('ftps' == $p['scheme']) || ('http' == $p['scheme']) || ('https' == $p['scheme'])) { if (empty($p['host'])) return false; if (empty($p['path'])) { $out_url .= '/';