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('GNUSOCIAL')) { exit(1); }
23 * Table Definition for file_redirection
26 class File_redirection extends Managed_DataObject
29 /* the code below is auto generated do not remove the above tag */
31 public $__table = 'file_redirection'; // table name
32 public $urlhash; // varchar(64) primary_key not_null
34 public $file_id; // int(4)
35 public $redirections; // int(4)
36 public $httpcode; // int(4)
37 public $modified; // timestamp() not_null default_CURRENT_TIMESTAMP
39 /* the code above is auto generated do not remove the tag below */
42 public static function schemaDef()
46 'urlhash' => array('type' => 'varchar', 'length' => 64, 'not null' => true, 'description' => 'sha256 hash of the URL'),
47 'url' => array('type' => 'text', 'description' => 'short URL (or any other kind of redirect) for file (id)'),
48 'file_id' => array('type' => 'int', 'description' => 'short URL for what URL/file'),
49 'redirections' => array('type' => 'int', 'description' => 'redirect count'),
50 'httpcode' => array('type' => 'int', 'description' => 'HTTP status code (20x, 30x, etc.)'),
51 'modified' => array('type' => 'timestamp', 'not null' => true, 'description' => 'date this record was modified'),
53 'primary key' => array('urlhash'),
54 'foreign keys' => array(
55 'file_redirection_file_id_fkey' => array('file' => array('file_id' => 'id')),
60 static public function getByUrl($url)
62 return self::getByPK(array('urlhash' => File::hashurl($url)));
65 static function _commonHttp($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 $in_url
86 * @return mixed one of:
87 * string - target URL, if this is a direct link or can't be followed
88 * array - redirect info if this is an *unknown* redirect:
89 * associative array with the following elements:
90 * code: HTTP status code
91 * redirects: count of redirects followed
92 * url: URL string of final target
93 * type (optional): MIME type from Content-Type header
94 * size (optional): byte size from Content-Length header
95 * time (optional): timestamp from Last-Modified header
97 static function lookupWhere($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...
172 $a = File::getByUrl($in_url);
173 // this is a direct link to $a->url
175 } catch (NoResultException $e) {
177 $b = File_redirection::getByUrl($in_url);
178 // this is a redirect to $b->file_id
179 $a = File::getKV('id', $b->file_id);
181 } catch (NoResultException $e) {
182 // Oh well, let's keep going
187 $ret = File_redirection::lookupWhere($in_url);
190 // No manual dereferencing; leave the unknown URL as is.
196 * Shorten a URL with the current user's configured shortening
197 * options, if applicable.
199 * If it cannot be shortened or the "short" URL is longer than the
200 * original, the original is returned.
202 * If the referenced item has not been seen before, embedding data
205 * @param string $long_url
206 * @param User $user whose shortening options to use; defaults to the current web session user
209 function makeShort($long_url, $user=null)
211 $canon = File_redirection::_canonUrl($long_url);
213 $short_url = File_redirection::_userMakeShort($canon, $user);
215 // Did we get one? Is it shorter?
217 if (!empty($short_url)) {
225 * Shorten a URL with the current user's configured shortening
226 * options, if applicable.
228 * If it cannot be shortened or the "short" URL is longer than the
229 * original, the original is returned.
231 * If the referenced item has not been seen before, embedding data
234 * @param string $long_url
238 function forceShort($long_url, $user)
240 $canon = File_redirection::_canonUrl($long_url);
242 $short_url = File_redirection::_userMakeShort($canon, $user, true);
244 // Did we get one? Is it shorter?
245 if (!empty($short_url)) {
252 static function _userMakeShort($long_url, User $user=null, $force = false) {
253 $short_url = common_shorten_url($long_url, $user, $force);
254 if (!empty($short_url) && $short_url != $long_url) {
255 $short_url = (string)$short_url;
257 $file = File::getKV('url', $long_url);
258 if ($file instanceof File) {
259 $file_id = $file->getID();
261 // Check if the target URL is itself a redirect...
262 $redir_data = File_redirection::where($long_url);
263 if (is_array($redir_data)) {
264 // We haven't seen the target URL before.
265 // Save file and embedding data about it!
266 $file = File::saveNew($redir_data, $long_url);
267 $file_id = $file->getID();
268 } else if (is_string($redir_data)) {
269 // The file is a known redirect target.
270 $file = File::getKV('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->getID();
282 $file_redir = File_redirection::getKV('url', $short_url);
283 if (!$file_redir instanceof File_redirection) {
284 $file_redir = new File_redirection;
285 $file_redir->urlhash = File::hashurl($short_url);
286 $file_redir->url = $short_url;
287 $file_redir->file_id = $file_id;
288 $file_redir->insert();
296 * Basic attempt to canonicalize a URL, cleaning up some standard variants
297 * such as funny syntax or a missing path. Used internally when cleaning
298 * up URLs for storage and following redirect chains.
300 * Note that despite being on File_redirect, this function DOES NOT perform
301 * any dereferencing of redirects.
303 * @param string $in_url input URL
304 * @param string $default_scheme if given a bare link; defaults to 'http://'
307 static function _canonUrl($in_url, $default_scheme = 'http://') {
308 if (empty($in_url)) return false;
310 $p = parse_url($out_url);
311 if (empty($p['host']) || empty($p['scheme'])) {
312 list($scheme) = explode(':', $in_url, 2);
313 switch (strtolower($scheme)) {
316 $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($p['path'])) {
345 static function saveNew($data, $file_id, $url) {
346 $file_redir = new File_redirection;
347 $file_redir->urlhash = File::hashurl($short_url);
348 $file_redir->url = $url;
349 $file_redir->file_id = $file_id;
350 $file_redir->redirections = intval($data['redirects']);
351 $file_redir->httpcode = intval($data['code']);
352 $file_redir->insert();
355 static public function beforeSchemaUpdate()
357 $table = strtolower(get_called_class());
358 $schema = Schema::get();
359 $schemadef = $schema->getTableDef($table);
361 // 2015-02-19 We have to upgrade our table definitions to have the urlhash field populated
362 if (isset($schemadef['fields']['urlhash']) && in_array('urlhash', $schemadef['primary key'])) {
363 // We already have the urlhash field, so no need to migrate it.
366 echo "\nFound old $table table, upgrading it to contain 'urlhash' field...";
367 // We have to create a urlhash that is _not_ the primary key,
368 // transfer data and THEN run checkSchema
369 $schemadef['fields']['urlhash'] = array (
373 'description' => 'sha256 hash of the URL',
375 $schemadef['fields']['url'] = array (
377 'description' => 'short URL (or any other kind of redirect) for file (id)',
379 unset($schemadef['primary key']);
380 $schema->ensureTable($table, $schemadef);
383 $classname = ucfirst($table);
384 $tablefix = new $classname;
385 // urlhash is hash('sha256', $url) in the File table
386 echo "Updating urlhash fields in $table table...";
387 // Maybe very MySQL specific :(
388 $tablefix->query(sprintf('UPDATE %1$s SET %2$s=%3$s;',
389 $schema->quoteIdentifier($table),
391 // The line below is "result of sha256 on column `url`"
394 echo "Resuming core schema upgrade...";