3 * StatusNet - the distributed open-source microblogging tool
4 * Copyright (C) 2008, 2009, StatusNet, Inc.
6 * This program is free software: you can redistribute it and/or modify
7 * it under the terms of the GNU Affero General Public License as published by
8 * the Free Software Foundation, either version 3 of the License, or
9 * (at your option) any later version.
11 * This program is distributed in the hope that it will be useful,
12 * but WITHOUT ANY WARRANTY; without even the implied warranty of
13 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
14 * GNU Affero General Public License for more details.
16 * You should have received a copy of the GNU Affero General Public License
17 * along with this program. If not, see <http://www.gnu.org/licenses/>.
20 if (!defined('STATUSNET') && !defined('LACONICA')) { exit(1); }
22 require_once INSTALLDIR.'/classes/Memcached_DataObject.php';
23 require_once INSTALLDIR.'/classes/File.php';
24 require_once INSTALLDIR.'/classes/File_oembed.php';
27 * Table Definition for file_redirection
30 class File_redirection extends Managed_DataObject
33 /* the code below is auto generated do not remove the above tag */
35 public $__table = 'file_redirection'; // table name
36 public $url; // varchar(255) primary_key not_null
37 public $file_id; // int(4)
38 public $redirections; // int(4)
39 public $httpcode; // int(4)
40 public $modified; // timestamp() not_null default_CURRENT_TIMESTAMP
42 /* the code above is auto generated do not remove the tag below */
45 public static function schemaDef()
49 'url' => array('type' => 'varchar', 'length' => 255, 'not null' => true, 'description' => 'short URL (or any other kind of redirect) for file (id)'),
50 'file_id' => array('type' => 'int', 'description' => 'short URL for what URL/file'),
51 'redirections' => array('type' => 'int', 'description' => 'redirect count'),
52 'httpcode' => array('type' => 'int', 'description' => 'HTTP status code (20x, 30x, etc.)'),
53 'modified' => array('type' => 'timestamp', 'not null' => true, 'description' => 'date this record was modified'),
55 'primary key' => array('url'),
56 'foreign keys' => array(
57 'file_redirection_file_id_fkey' => array('file' => array('file_id' => 'id')),
62 static function _commonHttp($url, $redirs) {
63 $request = new HTTPClient($url);
64 $request->setConfig(array(
65 'connect_timeout' => 10, // # seconds to wait
66 'max_redirs' => $redirs, // # max number of http redirections to follow
67 'follow_redirects' => true, // Follow redirects
68 'store_body' => false, // We won't need body content here.
74 * Check if this URL is a redirect and return redir info.
76 * Most code should call File_redirection::where instead, to check if we
77 * already know that redirection and avoid extra hits to the web.
79 * The URL is hit and any redirects are followed, up to 10 levels or until
80 * a protected URL is reached.
82 * @param string $in_url
83 * @return mixed one of:
84 * string - target URL, if this is a direct link or can't be followed
85 * array - redirect info if this is an *unknown* redirect:
86 * associative array with the following elements:
87 * code: HTTP status code
88 * redirects: count of redirects followed
89 * url: URL string of final target
90 * type (optional): MIME type from Content-Type header
91 * size (optional): byte size from Content-Length header
92 * time (optional): timestamp from Last-Modified header
94 public function lookupWhere($short_url, $redirs = 10, $protected = false) {
95 if ($redirs < 0) return false;
97 if(strpos($short_url,'://') === false){
101 $request = self::_commonHttp($short_url, $redirs);
102 // Don't include body in output
103 $request->setMethod(HTTP_Request2::METHOD_HEAD);
104 $response = $request->send();
106 if (405 == $response->getStatus() || 204 == $response->getStatus()) {
107 // HTTP 405 Unsupported Method
108 // Server doesn't support HEAD method? Can this really happen?
109 // We'll try again as a GET and ignore the response data.
111 // HTTP 204 No Content
112 // YFrog sends 204 responses back for our HEAD checks, which
113 // seems like it may be a logic error in their servers. If
114 // we get a 204 back, re-run it as a GET... if there's really
115 // no content it'll be cheap. :)
116 $request = self::_commonHttp($short_url, $redirs);
117 $response = $request->send();
119 } catch (Exception $e) {
120 // Invalid URL or failure to reach server
121 common_log(LOG_ERR, "Error while following redirects for $short_url: " . $e->getMessage());
125 if ($response->getRedirectCount() && File::isProtected($response->getUrl())) {
126 // Bump back up the redirect chain until we find a non-protected URL
127 return self::lookupWhere($short_url, $response->getRedirectCount() - 1, true);
130 $ret = array('code' => $response->getStatus()
131 , 'redirects' => $response->getRedirectCount()
132 , 'url' => $response->getUrl());
134 $type = $response->getHeader('Content-Type');
135 if ($type) $ret['type'] = $type;
136 if ($protected) $ret['protected'] = true;
137 $size = $response->getHeader('Content-Length'); // @fixme bytes?
138 if ($size) $ret['size'] = $size;
139 $time = $response->getHeader('Last-Modified');
140 if ($time) $ret['time'] = strtotime($time);
145 * Check if this URL is a redirect and return redir info.
146 * If a File record is present for this URL, it is not considered a redirect.
147 * If a File_redirection record is present for this URL, the recorded target is returned.
149 * If no File or File_redirect record is present, the URL is hit and any
150 * redirects are followed, up to 10 levels or until a protected URL is
153 * @param string $in_url
154 * @param boolean $discover true to attempt dereferencing the redirect if we don't know it already
155 * @return mixed one of:
156 * string - target URL, if this is a direct link or a known redirect
157 * array - redirect info if this is an *unknown* redirect:
158 * associative array with the following elements:
159 * code: HTTP status code
160 * redirects: count of redirects followed
161 * url: URL string of final target
162 * type (optional): MIME type from Content-Type header
163 * size (optional): byte size from Content-Length header
164 * time (optional): timestamp from Last-Modified header
166 public function where($in_url, $discover=true) {
167 // let's see if we know this...
168 $a = File::staticGet('url', $in_url);
171 // this is a direct link to $a->url
174 $b = File_redirection::staticGet('url', $in_url);
176 // this is a redirect to $b->file_id
177 $a = File::staticGet('id', $b->file_id);
183 $ret = File_redirection::lookupWhere($in_url);
186 // No manual dereferencing; leave the unknown URL as is.
192 * Shorten a URL with the current user's configured shortening
193 * options, if applicable.
195 * If it cannot be shortened or the "short" URL is longer than the
196 * original, the original is returned.
198 * If the referenced item has not been seen before, embedding data
201 * @param string $long_url
202 * @param User $user whose shortening options to use; defaults to the current web session user
205 function makeShort($long_url, $user=null)
207 $canon = File_redirection::_canonUrl($long_url);
209 $short_url = File_redirection::_userMakeShort($canon, $user);
211 // Did we get one? Is it shorter?
213 if (!empty($short_url)) {
221 * Shorten a URL with the current user's configured shortening
222 * options, if applicable.
224 * If it cannot be shortened or the "short" URL is longer than the
225 * original, the original is returned.
227 * If the referenced item has not been seen before, embedding data
230 * @param string $long_url
234 function forceShort($long_url, $user)
236 $canon = File_redirection::_canonUrl($long_url);
238 $short_url = File_redirection::_userMakeShort($canon, $user, true);
240 // Did we get one? Is it shorter?
241 if (!empty($short_url)) {
248 function _userMakeShort($long_url, User $user=null, $force = false) {
249 $short_url = common_shorten_url($long_url, $user, $force);
250 if (!empty($short_url) && $short_url != $long_url) {
251 $short_url = (string)$short_url;
253 $file = File::staticGet('url', $long_url);
255 // Check if the target URL is itself a redirect...
256 $redir_data = File_redirection::where($long_url);
257 if (is_array($redir_data)) {
258 // We haven't seen the target URL before.
259 // Save file and embedding data about it!
260 $file = File::saveNew($redir_data, $long_url);
261 $file_id = $file->id;
262 if (!empty($redir_data['oembed']['json'])) {
263 File_oembed::saveNew($redir_data['oembed']['json'], $file_id);
265 } else if (is_string($redir_data)) {
266 // The file is a known redirect target.
267 $file = File::staticGet('url', $redir_data);
269 // @fixme should we save a new one?
270 // this case was triggering sometimes for redirects
271 // with unresolvable targets; found while fixing
272 // "can't linkify" bugs with shortened links to
273 // SSL sites with cert issues.
276 $file_id = $file->id;
279 $file_id = $file->id;
281 $file_redir = File_redirection::staticGet('url', $short_url);
282 if (empty($file_redir)) {
283 $file_redir = new File_redirection;
284 $file_redir->url = $short_url;
285 $file_redir->file_id = $file_id;
286 $file_redir->insert();
294 * Basic attempt to canonicalize a URL, cleaning up some standard variants
295 * such as funny syntax or a missing path. Used internally when cleaning
296 * up URLs for storage and following redirect chains.
298 * Note that despite being on File_redirect, this function DOES NOT perform
299 * any dereferencing of redirects.
301 * @param string $in_url input URL
302 * @param string $default_scheme if given a bare link; defaults to 'http://'
305 function _canonUrl($in_url, $default_scheme = 'http://') {
306 if (empty($in_url)) return false;
308 $p = parse_url($out_url);
309 if (empty($p['host']) || empty($p['scheme'])) {
310 list($scheme) = explode(':', $in_url, 2);
311 switch (strtolower($scheme)) {
314 $out_url = str_replace('.-()', '', $out_url);
321 // don't touch anything
325 $out_url = $default_scheme . ltrim($out_url, '/');
326 $p = parse_url($out_url);
327 if (empty($p['scheme'])) return false;
332 if (('ftp' == $p['scheme']) || ('ftps' == $p['scheme']) || ('http' == $p['scheme']) || ('https' == $p['scheme'])) {
333 if (empty($p['host'])) return false;
334 if (empty($p['path'])) {
342 function saveNew($data, $file_id, $url) {
343 $file_redir = new File_redirection;
344 $file_redir->url = $url;
345 $file_redir->file_id = $file_id;
346 $file_redir->redirections = intval($data['redirects']);
347 $file_redir->httpcode = intval($data['code']);
348 $file_redir->insert();