<?php
/*
- * Laconica - a distributed open-source microblogging tool
- * Copyright (C) 2008, 2009, Control Yourself, Inc.
+ * StatusNet - the distributed open-source microblogging tool
+ * Copyright (C) 2008, 2009, StatusNet, Inc.
*
* This program is free software: you can redistribute it and/or modify
* it under the terms of the GNU Affero General Public License as published by
* along with this program. If not, see <http://www.gnu.org/licenses/>.
*/
-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
/* 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;
+ if(strpos($short_url,'://') === false){
+ return $short_url;
+ }
+ 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 $short_url;
+ }
+
+ if ($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' => $response->getStatus()
+ , 'redirects' => $response->getRedirectCount()
+ , 'url' => $response->getUrl());
+
+ $type = $response->getHeader('Content-Type');
+ if ($type) $ret['type'] = $type;
+ if ($protected) $ret['protected'] = true;
+ $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;
+ }
+
+ /**
+ * 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', $short_url);
+ $a = File::staticGet('url', $in_url);
if (!empty($a)) {
// this is a direct link to $a->url
return $a->url;
} else {
- $b = File_redirection::staticGet('url', $short_url);
+ $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);
}
}
- $curlh = File_redirection::_commonCurl($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($curlh);
- }
-
- if (!empty($info['redirect_count']) && File::isProtected($info['url'])) {
- return File_redirection::_redirectWhere_imp($short_url, $info['redirect_count'] - 1, true);
- }
-
- $ret = array('code' => $info['http_code']
- , 'redirects' => $info['redirect_count']
- , 'url' => $info['url']);
-
- if (!empty($info['content_type'])) $ret['type'] = $info['content_type'];
- if ($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'];
- return $ret;
- }
-
- function where($in_url) {
- $ret = File_redirection::_redirectWhere_imp($in_url);
+ $ret = File_redirection::lookupWhere($in_url);
return $ret;
}
+ /**
+ * 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) {
$canon = File_redirection::_canonUrl($long_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', $redir_data);
+ if (empty($file)) {
+ // @fixme should we save a new one?
+ // this case was triggering sometimes for redirects
+ // with unresolvable targets; found while fixing
+ // "can't linkify" bugs with shortened links to
+ // SSL sites with cert issues.
+ return null;
+ }
+ $file_id = $file->id;
}
} else {
$file_id = $file->id;
}
}
- 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 .= '/';
$file_redir->insert();
}
}
-
projects / quix0rs-gnu-social.git / blobdiff