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 $modified; // timestamp() not_null default_CURRENT_TIMESTAMP
43 function staticGet($k,$v=NULL) { return Memcached_DataObject::staticGet('File_redirection',$k,$v); }
45 /* the code above is auto generated do not remove the tag below */
48 public static function schemaDef()
52 'url' => array('type' => 'varchar', 'length' => 255, 'not null' => true, 'description' => 'short URL (or any other kind of redirect) for file (id)'),
53 'file_id' => array('type' => 'int', 'description' => 'short URL for what URL/file'),
54 'redirections' => array('type' => 'int', 'description' => 'redirect count'),
55 'httpcode' => array('type' => 'int', 'description' => 'HTTP status code (20x, 30x, etc.)'),
56 'modified' => array('type' => 'timestamp', 'not null' => true, 'description' => 'date this record was modified'),
58 'primary key' => array('url'),
59 'foreign keys' => array(
60 'file_redirection_file_id_fkey' => array('file' => array('file_id' => 'id')),
65 static function _commonHttp($url, $redirs) {
66 $request = new HTTPClient($url);
67 $request->setConfig(array(
68 'connect_timeout' => 10, // # seconds to wait
69 'max_redirs' => $redirs, // # max number of http redirections to follow
70 'follow_redirects' => true, // Follow redirects
71 'store_body' => false, // We won't need body content here.
77 * Check if this URL is a redirect and return redir info.
79 * Most code should call File_redirection::where instead, to check if we
80 * already know that redirection and avoid extra hits to the web.
82 * The URL is hit and any redirects are followed, up to 10 levels or until
83 * a protected URL is reached.
85 * @param string $in_url
86 * @return mixed one of:
87 * string - target URL, if this is a direct link or can't be followed
88 * array - redirect info if this is an *unknown* redirect:
89 * associative array with the following elements:
90 * code: HTTP status code
91 * redirects: count of redirects followed
92 * url: URL string of final target
93 * type (optional): MIME type from Content-Type header
94 * size (optional): byte size from Content-Length header
95 * time (optional): timestamp from Last-Modified header
97 public function lookupWhere($short_url, $redirs = 10, $protected = false) {
98 if ($redirs < 0) return false;
100 if(strpos($short_url,'://') === false){
104 $request = self::_commonHttp($short_url, $redirs);
105 // Don't include body in output
106 $request->setMethod(HTTP_Request2::METHOD_HEAD);
107 $response = $request->send();
109 if (405 == $response->getStatus() || 204 == $response->getStatus()) {
110 // HTTP 405 Unsupported Method
111 // Server doesn't support HEAD method? Can this really happen?
112 // We'll try again as a GET and ignore the response data.
114 // HTTP 204 No Content
115 // YFrog sends 204 responses back for our HEAD checks, which
116 // seems like it may be a logic error in their servers. If
117 // we get a 204 back, re-run it as a GET... if there's really
118 // no content it'll be cheap. :)
119 $request = self::_commonHttp($short_url, $redirs);
120 $response = $request->send();
122 } catch (Exception $e) {
123 // Invalid URL or failure to reach server
124 common_log(LOG_ERR, "Error while following redirects for $short_url: " . $e->getMessage());
128 if ($response->getRedirectCount() && File::isProtected($response->getUrl())) {
129 // Bump back up the redirect chain until we find a non-protected URL
130 return self::lookupWhere($short_url, $response->getRedirectCount() - 1, true);
133 $ret = array('code' => $response->getStatus()
134 , 'redirects' => $response->getRedirectCount()
135 , 'url' => $response->getUrl());
137 $type = $response->getHeader('Content-Type');
138 if ($type) $ret['type'] = $type;
139 if ($protected) $ret['protected'] = true;
140 $size = $response->getHeader('Content-Length'); // @fixme bytes?
141 if ($size) $ret['size'] = $size;
142 $time = $response->getHeader('Last-Modified');
143 if ($time) $ret['time'] = strtotime($time);
148 * Check if this URL is a redirect and return redir info.
149 * If a File record is present for this URL, it is not considered a redirect.
150 * If a File_redirection record is present for this URL, the recorded target is returned.
152 * If no File or File_redirect record is present, the URL is hit and any
153 * redirects are followed, up to 10 levels or until a protected URL is
156 * @param string $in_url
157 * @param boolean $discover true to attempt dereferencing the redirect if we don't know it already
158 * @return mixed one of:
159 * string - target URL, if this is a direct link or a known redirect
160 * array - redirect info if this is an *unknown* redirect:
161 * associative array with the following elements:
162 * code: HTTP status code
163 * redirects: count of redirects followed
164 * url: URL string of final target
165 * type (optional): MIME type from Content-Type header
166 * size (optional): byte size from Content-Length header
167 * time (optional): timestamp from Last-Modified header
169 public function where($in_url, $discover=true) {
170 // let's see if we know this...
171 $a = File::staticGet('url', $in_url);
174 // this is a direct link to $a->url
177 $b = File_redirection::staticGet('url', $in_url);
179 // this is a redirect to $b->file_id
180 $a = File::staticGet('id', $b->file_id);
186 $ret = File_redirection::lookupWhere($in_url);
189 // No manual dereferencing; leave the unknown URL as is.
195 * Shorten a URL with the current user's configured shortening
196 * options, if applicable.
198 * If it cannot be shortened or the "short" URL is longer than the
199 * original, the original is returned.
201 * If the referenced item has not been seen before, embedding data
204 * @param string $long_url
205 * @param User $user whose shortening options to use; defaults to the current web session user
208 function makeShort($long_url, $user=null)
210 $canon = File_redirection::_canonUrl($long_url);
212 $short_url = File_redirection::_userMakeShort($canon, $user);
214 // Did we get one? Is it shorter?
216 if (!empty($short_url)) {
224 * Shorten a URL with the current user's configured shortening
225 * options, if applicable.
227 * If it cannot be shortened or the "short" URL is longer than the
228 * original, the original is returned.
230 * If the referenced item has not been seen before, embedding data
233 * @param string $long_url
237 function forceShort($long_url, $user)
239 $canon = File_redirection::_canonUrl($long_url);
241 $short_url = File_redirection::_userMakeShort($canon, $user, true);
243 // Did we get one? Is it shorter?
244 if (!empty($short_url)) {
251 function _userMakeShort($long_url, User $user=null, $force = false) {
252 $short_url = common_shorten_url($long_url, $user, $force);
253 if (!empty($short_url) && $short_url != $long_url) {
254 $short_url = (string)$short_url;
256 $file = File::staticGet('url', $long_url);
258 // Check if the target URL is itself a redirect...
259 $redir_data = File_redirection::where($long_url);
260 if (is_array($redir_data)) {
261 // We haven't seen the target URL before.
262 // Save file and embedding data about it!
263 $file = File::saveNew($redir_data, $long_url);
264 $file_id = $file->id;
265 if (!empty($redir_data['oembed']['json'])) {
266 File_oembed::saveNew($redir_data['oembed']['json'], $file_id);
268 } else if (is_string($redir_data)) {
269 // The file is a known redirect target.
270 $file = File::staticGet('url', $redir_data);
272 // @fixme should we save a new one?
273 // this case was triggering sometimes for redirects
274 // with unresolvable targets; found while fixing
275 // "can't linkify" bugs with shortened links to
276 // SSL sites with cert issues.
279 $file_id = $file->id;
282 $file_id = $file->id;
284 $file_redir = File_redirection::staticGet('url', $short_url);
285 if (empty($file_redir)) {
286 $file_redir = new File_redirection;
287 $file_redir->url = $short_url;
288 $file_redir->file_id = $file_id;
289 $file_redir->insert();
297 * Basic attempt to canonicalize a URL, cleaning up some standard variants
298 * such as funny syntax or a missing path. Used internally when cleaning
299 * up URLs for storage and following redirect chains.
301 * Note that despite being on File_redirect, this function DOES NOT perform
302 * any dereferencing of redirects.
304 * @param string $in_url input URL
305 * @param string $default_scheme if given a bare link; defaults to 'http://'
308 function _canonUrl($in_url, $default_scheme = 'http://') {
309 if (empty($in_url)) return false;
311 $p = parse_url($out_url);
312 if (empty($p['host']) || empty($p['scheme'])) {
313 list($scheme) = explode(':', $in_url, 2);
317 $out_url = str_replace('.-()', '', $out_url);
324 // don't touch anything
328 $out_url = $default_scheme . ltrim($out_url, '/');
329 $p = parse_url($out_url);
330 if (empty($p['scheme'])) return false;
335 if (('ftp' == $p['scheme']) || ('ftps' == $p['scheme']) || ('http' == $p['scheme']) || ('https' == $p['scheme'])) {
336 if (empty($p['host'])) return false;
337 if (empty($p['path'])) {
345 function saveNew($data, $file_id, $url) {
346 $file_redir = new File_redirection;
347 $file_redir->url = $url;
348 $file_redir->file_id = $file_id;
349 $file_redir->redirections = intval($data['redirects']);
350 $file_redir->httpcode = intval($data['code']);
351 $file_redir->insert();