MediaWiki REL1_31
PoolCounterRedis.php
Go to the documentation of this file.
1<?php
20use Psr\Log\LoggerInterface;
21
55 protected $ring;
57 protected $pool;
59 protected $logger;
61 protected $serversByLabel;
63 protected $keySha1;
65 protected $lockTTL;
66
68 protected $conn;
70 protected $slot;
72 protected $onRelease;
74 protected $session;
76 protected $slotTime;
77
78 const AWAKE_ONE = 1; // wake-up if when a slot can be taken from an existing process
79 const AWAKE_ALL = 2; // wake-up if an existing process finishes and wake up such others
80
82 protected static $active = null;
83
84 function __construct( $conf, $type, $key ) {
85 parent::__construct( $conf, $type, $key );
86
87 $this->serversByLabel = $conf['servers'];
88 $this->ring = new HashRing( array_fill_keys( array_keys( $conf['servers'] ), 100 ) );
89
90 $conf['redisConfig']['serializer'] = 'none'; // for use with Lua
91 $this->pool = RedisConnectionPool::singleton( $conf['redisConfig'] );
92 $this->logger = \MediaWiki\Logger\LoggerFactory::getInstance( 'redis' );
93
94 $this->keySha1 = sha1( $this->key );
95 $met = ini_get( 'max_execution_time' ); // usually 0 in CLI mode
96 $this->lockTTL = $met ? 2 * $met : 3600;
97
98 if ( self::$active === null ) {
99 self::$active = [];
100 register_shutdown_function( [ __CLASS__, 'releaseAll' ] );
101 }
102 }
103
107 protected function getConnection() {
108 if ( !isset( $this->conn ) ) {
109 $conn = false;
110 $servers = $this->ring->getLocations( $this->key, 3 );
111 ArrayUtils::consistentHashSort( $servers, $this->key );
112 foreach ( $servers as $server ) {
113 $conn = $this->pool->getConnection( $this->serversByLabel[$server], $this->logger );
114 if ( $conn ) {
115 break;
116 }
117 }
118 if ( !$conn ) {
119 return Status::newFatal( 'pool-servererror', implode( ', ', $servers ) );
120 }
121 $this->conn = $conn;
122 }
123 return Status::newGood( $this->conn );
124 }
125
126 function acquireForMe() {
127 $status = $this->precheckAcquire();
128 if ( !$status->isGood() ) {
129 return $status;
130 }
131
132 return $this->waitForSlotOrNotif( self::AWAKE_ONE );
133 }
134
135 function acquireForAnyone() {
136 $status = $this->precheckAcquire();
137 if ( !$status->isGood() ) {
138 return $status;
139 }
140
141 return $this->waitForSlotOrNotif( self::AWAKE_ALL );
142 }
143
144 function release() {
145 if ( $this->slot === null ) {
146 return Status::newGood( PoolCounter::NOT_LOCKED ); // not locked
147 }
148
149 $status = $this->getConnection();
150 if ( !$status->isOK() ) {
151 return $status;
152 }
153 $conn = $status->value;
154
155 // phpcs:disable Generic.Files.LineLength
156 static $script =
158<<<LUA
159 local kSlots,kSlotsNextRelease,kWakeup,kWaiting = unpack(KEYS)
160 local rMaxWorkers,rExpiry,rSlot,rSlotTime,rAwakeAll,rTime = unpack(ARGV)
161 -- Add the slots back to the list (if rSlot is "w" then it is not a slot).
162 -- Treat the list as expired if the "next release" time sorted-set is missing.
163 if rSlot ~= 'w' and redis.call('exists',kSlotsNextRelease) == 1 then
164 if 1*redis.call('zScore',kSlotsNextRelease,rSlot) ~= (rSlotTime + rExpiry) then
165 -- Slot lock expired and was released already
166 elseif redis.call('lLen',kSlots) >= 1*rMaxWorkers then
167 -- Slots somehow got out of sync; reset the list for sanity
168 redis.call('del',kSlots,kSlotsNextRelease)
169 elseif redis.call('lLen',kSlots) == (1*rMaxWorkers - 1) and redis.call('zCard',kWaiting) == 0 then
170 -- Slot list will be made full; clear it to save space (it re-inits as needed)
171 -- since nothing is waiting on being unblocked by a push to the list
172 redis.call('del',kSlots,kSlotsNextRelease)
173 else
174 -- Add slot back to pool and update the "next release" time
175 redis.call('rPush',kSlots,rSlot)
176 redis.call('zAdd',kSlotsNextRelease,rTime + 30,rSlot)
177 -- Always keep renewing the expiry on use
178 redis.call('expireAt',kSlots,math.ceil(rTime + rExpiry))
179 redis.call('expireAt',kSlotsNextRelease,math.ceil(rTime + rExpiry))
180 end
181 end
182 -- Update an ephemeral list to wake up other clients that can
183 -- reuse any cached work from this process. Only do this if no
184 -- slots are currently free (e.g. clients could be waiting).
185 if 1*rAwakeAll == 1 then
186 local count = redis.call('zCard',kWaiting)
187 for i = 1,count do
188 redis.call('rPush',kWakeup,'w')
189 end
190 redis.call('pexpire',kWakeup,1)
191 end
192 return 1
193LUA;
194 // phpcs:enable
195
196 try {
197 $conn->luaEval( $script,
198 [
199 $this->getSlotListKey(),
200 $this->getSlotRTimeSetKey(),
201 $this->getWakeupListKey(),
202 $this->getWaitSetKey(),
203 $this->workers,
204 $this->lockTTL,
205 $this->slot,
206 $this->slotTime, // used for CAS-style sanity check
207 ( $this->onRelease === self::AWAKE_ALL ) ? 1 : 0,
208 microtime( true )
209 ],
210 4 # number of first argument(s) that are keys
211 );
212 } catch ( RedisException $e ) {
213 return Status::newFatal( 'pool-error-unknown', $e->getMessage() );
214 }
215
216 $this->slot = null;
217 $this->slotTime = null;
218 $this->onRelease = null;
219 unset( self::$active[$this->session] );
220
221 $this->onRelease();
222
223 return Status::newGood( PoolCounter::RELEASED );
224 }
225
230 protected function waitForSlotOrNotif( $doWakeup ) {
231 if ( $this->slot !== null ) {
232 return Status::newGood( PoolCounter::LOCK_HELD ); // already acquired
233 }
234
235 $status = $this->getConnection();
236 if ( !$status->isOK() ) {
237 return $status;
238 }
239 $conn = $status->value;
240
241 $now = microtime( true );
242 try {
243 $slot = $this->initAndPopPoolSlotList( $conn, $now );
244 if ( ctype_digit( $slot ) ) {
245 // Pool slot acquired by this process
246 $slotTime = $now;
247 } elseif ( $slot === 'QUEUE_FULL' ) {
248 // Too many processes are waiting for pooled processes to finish
249 return Status::newGood( PoolCounter::QUEUE_FULL );
250 } elseif ( $slot === 'QUEUE_WAIT' ) {
251 // This process is now registered as waiting
252 $keys = ( $doWakeup == self::AWAKE_ALL )
253 // Wait for an open slot or wake-up signal (preferring the latter)
254 ? [ $this->getWakeupListKey(), $this->getSlotListKey() ]
255 // Just wait for an actual pool slot
256 : [ $this->getSlotListKey() ];
257
258 $res = $conn->blPop( $keys, $this->timeout );
259 if ( $res === [] ) {
260 $conn->zRem( $this->getWaitSetKey(), $this->session ); // no longer waiting
261 return Status::newGood( PoolCounter::TIMEOUT );
262 }
263
264 $slot = $res[1]; // pool slot or "w" for wake-up notifications
265 $slotTime = microtime( true ); // last microtime() was a few RTTs ago
266 // Unregister this process as waiting and bump slot "next release" time
268 } else {
269 return Status::newFatal( 'pool-error-unknown', "Server gave slot '$slot'." );
270 }
271 } catch ( RedisException $e ) {
272 return Status::newFatal( 'pool-error-unknown', $e->getMessage() );
273 }
274
275 if ( $slot !== 'w' ) {
276 $this->slot = $slot;
277 $this->slotTime = $slotTime;
278 $this->onRelease = $doWakeup;
279 self::$active[$this->session] = $this;
280 }
281
282 $this->onAcquire();
283
284 return Status::newGood( $slot === 'w' ? PoolCounter::DONE : PoolCounter::LOCKED );
285 }
286
292 protected function initAndPopPoolSlotList( RedisConnRef $conn, $now ) {
293 static $script =
295<<<LUA
296 local kSlots,kSlotsNextRelease,kSlotWaits = unpack(KEYS)
297 local rMaxWorkers,rMaxQueue,rTimeout,rExpiry,rSess,rTime = unpack(ARGV)
298 -- Initialize if the "next release" time sorted-set is empty. The slot key
299 -- itself is empty if all slots are busy or when nothing is initialized.
300 -- If the list is empty but the set is not, then it is the latter case.
301 -- For sanity, if the list exists but not the set, then reset everything.
302 if redis.call('exists',kSlotsNextRelease) == 0 then
303 redis.call('del',kSlots)
304 for i = 1,1*rMaxWorkers do
305 redis.call('rPush',kSlots,i)
306 redis.call('zAdd',kSlotsNextRelease,-1,i)
307 end
308 -- Otherwise do maintenance to clean up after network partitions
309 else
310 -- Find stale slot locks and add free them (avoid duplicates for sanity)
311 local staleLocks = redis.call('zRangeByScore',kSlotsNextRelease,0,rTime)
312 for k,slot in ipairs(staleLocks) do
313 redis.call('lRem',kSlots,0,slot)
314 redis.call('rPush',kSlots,slot)
315 redis.call('zAdd',kSlotsNextRelease,rTime + 30,slot)
316 end
317 -- Find stale wait slot entries and remove them
318 redis.call('zRemRangeByScore',kSlotWaits,0,rTime - 2*rTimeout)
319 end
320 local slot
321 -- Try to acquire a slot if possible now
322 if redis.call('lLen',kSlots) > 0 then
323 slot = redis.call('lPop',kSlots)
324 -- Update the slot "next release" time
325 redis.call('zAdd',kSlotsNextRelease,rTime + rExpiry,slot)
326 elseif redis.call('zCard',kSlotWaits) >= 1*rMaxQueue then
327 slot = 'QUEUE_FULL'
328 else
329 slot = 'QUEUE_WAIT'
330 -- Register this process as waiting
331 redis.call('zAdd',kSlotWaits,rTime,rSess)
332 redis.call('expireAt',kSlotWaits,math.ceil(rTime + 2*rTimeout))
333 end
334 -- Always keep renewing the expiry on use
335 redis.call('expireAt',kSlots,math.ceil(rTime + rExpiry))
336 redis.call('expireAt',kSlotsNextRelease,math.ceil(rTime + rExpiry))
337 return slot
338LUA;
339 return $conn->luaEval( $script,
340 [
341 $this->getSlotListKey(),
342 $this->getSlotRTimeSetKey(),
343 $this->getWaitSetKey(),
344 $this->workers,
345 $this->maxqueue,
346 $this->timeout,
347 $this->lockTTL,
348 $this->session,
349 $now
350 ],
351 3 # number of first argument(s) that are keys
352 );
353 }
354
361 protected function registerAcquisitionTime( RedisConnRef $conn, $slot, $now ) {
362 static $script =
364<<<LUA
365 local kSlots,kSlotsNextRelease,kSlotWaits = unpack(KEYS)
366 local rSlot,rExpiry,rSess,rTime = unpack(ARGV)
367 -- If rSlot is 'w' then the client was told to wake up but got no slot
368 if rSlot ~= 'w' then
369 -- Update the slot "next release" time
370 redis.call('zAdd',kSlotsNextRelease,rTime + rExpiry,rSlot)
371 -- Always keep renewing the expiry on use
372 redis.call('expireAt',kSlots,math.ceil(rTime + rExpiry))
373 redis.call('expireAt',kSlotsNextRelease,math.ceil(rTime + rExpiry))
374 end
375 -- Unregister this process as waiting
376 redis.call('zRem',kSlotWaits,rSess)
377 return 1
378LUA;
379 return $conn->luaEval( $script,
380 [
381 $this->getSlotListKey(),
382 $this->getSlotRTimeSetKey(),
383 $this->getWaitSetKey(),
384 $slot,
385 $this->lockTTL,
386 $this->session,
387 $now
388 ],
389 3 # number of first argument(s) that are keys
390 );
391 }
392
396 protected function getSlotListKey() {
397 return "poolcounter:l-slots-{$this->keySha1}-{$this->workers}";
398 }
399
403 protected function getSlotRTimeSetKey() {
404 return "poolcounter:z-renewtime-{$this->keySha1}-{$this->workers}";
405 }
406
410 protected function getWaitSetKey() {
411 return "poolcounter:z-wait-{$this->keySha1}-{$this->workers}";
412 }
413
417 protected function getWakeupListKey() {
418 return "poolcounter:l-wakeup-{$this->keySha1}-{$this->workers}";
419 }
420
424 public static function releaseAll() {
425 foreach ( self::$active as $poolCounter ) {
426 try {
427 if ( $poolCounter->slot !== null ) {
428 $poolCounter->release();
429 }
430 } catch ( Exception $e ) {
431 }
432 }
433 }
434}
and(b) You must cause any modified files to carry prominent notices stating that You changed the files
Apache License January AND DISTRIBUTION Definitions License shall mean the terms and conditions for and distribution as defined by Sections through of this document Licensor shall mean the copyright owner or entity authorized by the copyright owner that is granting the License Legal Entity shall mean the union of the acting entity and all other entities that control are controlled by or are under common control with that entity For the purposes of this definition control direct or to cause the direction or management of such whether by contract or including but not limited to software source documentation and configuration files Object form shall mean any form resulting from mechanical transformation or translation of a Source including but not limited to compiled object generated and conversions to other media types Work shall mean the work of whether in Source or Object made available under the as indicated by a copyright notice that is included in or attached to the work(an example is provided in the Appendix below). "Derivative Works" shall mean any work
Apache License January AND DISTRIBUTION Definitions License shall mean the terms and conditions for use
Apache License January AND DISTRIBUTION Definitions License shall mean the terms and conditions for and distribution as defined by Sections through of this document Licensor shall mean the copyright owner or entity authorized by the copyright owner that is granting the License Legal Entity shall mean the union of the acting entity and all other entities that control are controlled by or are under common control with that entity For the purposes of this definition control direct or to cause the direction or management of such whether by contract or including but not limited to software source documentation and configuration files Object form shall mean any form resulting from mechanical transformation or translation of a Source including but not limited to compiled object generated and conversions to other media types Work shall mean the work of whether in Source or Object made available under the as indicated by a copyright notice that is included in or attached to the whether in Source or Object that is based or other modifications as a an original work of authorship For the purposes of this Derivative Works shall not include works that remain separable or merely the Work and Derivative Works thereof Contribution shall mean any work of including the original version of the Work and any modifications or additions to that Work or Derivative Works that is intentionally submitted to Licensor for inclusion in the Work by the copyright owner or by an individual or Legal Entity authorized to submit on behalf of the copyright owner For the purposes of this submitted means any form of or written communication sent to the Licensor or its including but not limited to communication on electronic mailing source code control and issue tracking systems that are managed or on behalf the Licensor for the purpose of discussing and improving the but excluding communication that is conspicuously marked or otherwise designated in writing by the copyright owner as Not a Contribution Contributor shall mean Licensor and any individual or Legal Entity on behalf of whom a Contribution has been received by Licensor and subsequently incorporated within the Work Grant of Copyright License Subject to the terms and conditions of this each Contributor hereby grants to You a non no royalty free
Apache License January AND DISTRIBUTION Definitions License shall mean the terms and conditions for and distribution as defined by Sections through of this document Licensor shall mean the copyright owner or entity authorized by the copyright owner that is granting the License Legal Entity shall mean the union of the acting entity and all other entities that control are controlled by or are under common control with that entity For the purposes of this definition control direct or to cause the direction or management of such whether by contract or including but not limited to software source documentation and configuration files Object form shall mean any form resulting from mechanical transformation or translation of a Source including but not limited to compiled object generated and conversions to other media types Work shall mean the work of whether in Source or Object made available under the as indicated by a copyright notice that is included in or attached to the whether in Source or Object that is based or other modifications as a an original work of authorship For the purposes of this Derivative Works shall not include works that remain separable or merely the Work and Derivative Works thereof Contribution shall mean any work of including the original version of the Work and any modifications or additions to that Work or Derivative Works that is intentionally submitted to Licensor for inclusion in the Work by the copyright owner or by an individual or Legal Entity authorized to submit on behalf of the copyright owner For the purposes of this submitted means any form of or written communication sent to the Licensor or its including but not limited to communication on electronic mailing source code control and issue tracking systems that are managed by
Apache License January AND DISTRIBUTION Definitions License shall mean the terms and conditions for and distribution as defined by Sections through of this document Licensor shall mean the copyright owner or entity authorized by the copyright owner that is granting the License Legal Entity shall mean the union of the acting entity and all other entities that control are controlled by or are under common control with that entity For the purposes of this definition control direct or to cause the direction or management of such whether by contract or including but not limited to software source documentation and configuration files Object form shall mean any form resulting from mechanical transformation or translation of a Source including but not limited to compiled object generated and conversions to other media types Work shall mean the work of whether in Source or Object made available under the as indicated by a copyright notice that is included in or attached to the whether in Source or Object that is based or other modifications as a an original work of authorship For the purposes of this Derivative Works shall not include works that remain separable or merely the Work and Derivative Works thereof Contribution shall mean any work of including the original version of the Work and any modifications or additions to that Work or Derivative Works that is intentionally submitted to Licensor for inclusion in the Work by the copyright owner or by an individual or Legal Entity authorized to submit on behalf of the copyright owner For the purposes of this submitted means any form of or written communication sent to the Licensor or its including but not limited to communication on electronic mailing source code control and issue tracking systems that are managed or on behalf the Licensor for the purpose of discussing and improving the but excluding communication that is conspicuously marked or otherwise designated in writing by the copyright owner as Not a Contribution Contributor shall mean Licensor and any individual or Legal Entity on behalf of whom a Contribution has been received by Licensor and subsequently incorporated within the Work Grant of Copyright License Subject to the terms and conditions of this each Contributor hereby grants to You a non no royalty irrevocable copyright license to prepare Derivative Works publicly publicly and distribute the Work and such Derivative Works in Source or Object form Grant of Patent License Subject to the terms and conditions of this each Contributor hereby grants to You a non no royalty have made
Apache License January AND DISTRIBUTION Definitions License shall mean the terms and conditions for and distribution as defined by Sections through of this document Licensor shall mean the copyright owner or entity authorized by the copyright owner that is granting the License Legal Entity shall mean the union of the acting entity and all other entities that control are controlled by or are under common control with that entity For the purposes of this definition control direct or to cause the direction or management of such whether by contract or including but not limited to software source documentation and configuration files Object form shall mean any form resulting from mechanical transformation or translation of a Source including but not limited to compiled object generated and conversions to other media types Work shall mean the work of whether in Source or Object made available under the as indicated by a copyright notice that is included in or attached to the whether in Source or Object that is based or other modifications as a an original work of authorship For the purposes of this Derivative Works shall not include works that remain separable from
</td >< td > &</td >< td > t want your writing to be edited mercilessly and redistributed at will
Convenience class for weighted consistent hash rings.
Definition HashRing.php:28
Version of PoolCounter that uses Redis.
int $onRelease
AWAKE_* constant.
__construct( $conf, $type, $key)
LoggerInterface $logger
string $slot
Pool slot value.
int $slotTime
UNIX timestamp.
waitForSlotOrNotif( $doWakeup)
acquireForAnyone()
I want to do this task, but if anyone else does it instead, it's also fine for me.
initAndPopPoolSlotList(RedisConnRef $conn, $now)
array $serversByLabel
(server label => host) map
string $session
Unique string to identify this process.
static releaseAll()
Try to make sure that locks get released (even with exceptions and fatals)
release()
I have successfully finished my task.
static PoolCounterRedis[] $active
List of active PoolCounterRedis objects in this script.
registerAcquisitionTime(RedisConnRef $conn, $slot, $now)
acquireForMe()
I want to do this task and I need to do it myself.
int $lockTTL
TTL for locks to expire (work should finish in this time)
string $keySha1
SHA-1 of the key.
RedisConnectionPool $pool
When you have many workers (threads/servers) giving service, and a cached item expensive to produce e...
const QUEUE_FULL
onAcquire()
Update any lock tracking information when the lock is acquired.
string $key
All workers with the same key share the lock.
precheckAcquire()
Checks that the lock request is sane.
const NOT_LOCKED
const LOCK_HELD
onRelease()
Update any lock tracking information when the lock is released.
Helper class to handle automatically marking connectons as reusable (via RAII pattern)
luaEval( $script, array $params, $numKeys)
Helper class to manage Redis connections.
static singleton(array $options)
=Architecture==Two class hierarchies are used to provide the functionality associated with the different content models:*Content interface(and AbstractContent base class) define functionality that acts on the concrete content of a page, and *ContentHandler base class provides functionality specific to a content model, but not acting on concrete content. The most important function of ContentHandler is to act as a factory for the appropriate implementation of Content. These Content objects are to be used by MediaWiki everywhere, instead of passing page content around as text. All manipulation and analysis of page content must be done via the appropriate methods of the Content object. For each content model, a subclass of ContentHandler has to be registered with $wgContentHandlers. The ContentHandler object for a given content model can be obtained using ContentHandler::getForModelID($id). Also Title, WikiPage and Revision now have getContentHandler() methods for convenience. ContentHandler objects are singletons that provide functionality specific to the content type, but not directly acting on the content of some page. ContentHandler::makeEmptyContent() and ContentHandler::unserializeContent() can be used to create a Content object of the appropriate type. However, it is recommended to instead use WikiPage::getContent() resp. Revision::getContent() to get a page 's content as a Content object. These two methods should be the ONLY way in which page content is accessed. Another important function of ContentHandler objects is to define custom action handlers for a content model, see ContentHandler::getActionOverrides(). This is similar to what WikiPage::getActionOverrides() was already doing.==Serialization==With the ContentHandler facility, page content no longer has to be text based. Objects implementing the Content interface are used to represent and handle the content internally. For storage and data exchange, each content model supports at least one serialization format via ContentHandler::serializeContent($content). The list of supported formats for a given content model can be accessed using ContentHandler::getSupportedFormats(). Content serialization formats are identified using MIME type like strings. The following formats are built in:*text/x-wiki - wikitext *text/javascript - for js pages *text/css - for css pages *text/plain - for future use, e.g. with plain text messages. *text/html - for future use, e.g. with plain html messages. *application/vnd.php.serialized - for future use with the api and for extensions *application/json - for future use with the api, and for use by extensions *application/xml - for future use with the api, and for use by extensions In PHP, use the corresponding CONTENT_FORMAT_XXX constant. Note that when using the API to access page content, especially action=edit, action=parse and action=query &prop=revisions, the model and format of the content should always be handled explicitly. Without that information, interpretation of the provided content is not reliable. The same applies to XML dumps generated via maintenance/dumpBackup.php or Special:Export. Also note that the API will provide encapsulated, serialized content - so if the API was called with format=json, and contentformat is also json(or rather, application/json), the page content is represented as a string containing an escaped json structure. Extensions that use JSON to serialize some types of page content may provide specialized API modules that allow access to that content in a more natural form.==Compatibility==The ContentHandler facility is introduced in a way that should allow all existing code to keep functioning at least for pages that contain wikitext or other text based content. However, a number of functions and hooks have been deprecated in favor of new versions that are aware of the page 's content model, and will now generate warnings when used. Most importantly, the following functions have been deprecated:*Revisions::getText() is deprecated in favor Revisions::getContent() *WikiPage::getText() is deprecated in favor WikiPage::getContent() Also, the old Article::getContent()(which returns text) is superceded by Article::getContentObject(). However, both methods should be avoided since they do not provide clean access to the page 's actual content. For instance, they may return a system message for non-existing pages. Use WikiPage::getContent() instead. Code that relies on a textual representation of the page content should eventually be rewritten. However, ContentHandler::getContentText() provides a stop-gap that can be used to get text for a page. Its behavior is controlled by $wgContentHandlerTextFallback it
The ContentHandler facility adds support for arbitrary content types on wiki instead of relying on wikitext for everything It was introduced in MediaWiki Each kind of and so on Built in content types are
$res
Definition database.txt:21
We use the convention $dbr for read and $dbw for write to help you keep track of whether the database object is a the world will explode Or to be a subsequent write query which succeeded on the master may fail when replicated to the slave due to a unique key collision Replication on the slave will stop and it may take hours to repair the database and get it back online Setting read_only in my cnf on the slave will avoid this but given the dire we prefer to have as many checks as possible We provide a but the wrapper functions like please read the documentation for except in special pages derived from QueryPage It s a common pitfall for new developers to submit code containing SQL queries which examine huge numbers of rows Remember that COUNT * is(N), counting rows in atable is like counting beans in a bucket.------------------------------------------------------------------------ Replication------------------------------------------------------------------------The largest installation of MediaWiki, Wikimedia, uses a large set ofslave MySQL servers replicating writes made to a master MySQL server. Itis important to understand the issues associated with this setup if youwant to write code destined for Wikipedia.It 's often the case that the best algorithm to use for a given taskdepends on whether or not replication is in use. Due to our unabashedWikipedia-centrism, we often just use the replication-friendly version, but if you like, you can use wfGetLB() ->getServerCount() > 1 tocheck to see if replication is in use.===Lag===Lag primarily occurs when large write queries are sent to the master.Writes on the master are executed in parallel, but they are executed inserial when they are replicated to the slaves. The master writes thequery to the binlog when the transaction is committed. The slaves pollthe binlog and start executing the query as soon as it appears. They canservice reads while they are performing a write query, but will not readanything more from the binlog and thus will perform no more writes. Thismeans that if the write query runs for a long time, the slaves will lagbehind the master for the time it takes for the write query to complete.Lag can be exacerbated by high read load. MediaWiki 's load balancer willstop sending reads to a slave when it is lagged by more than 30 seconds.If the load ratios are set incorrectly, or if there is too much loadgenerally, this may lead to a slave permanently hovering around 30seconds lag.If all slaves are lagged by more than 30 seconds, MediaWiki will stopwriting to the database. All edits and other write operations will berefused, with an error returned to the user. This gives the slaves achance to catch up. Before we had this mechanism, the slaves wouldregularly lag by several minutes, making review of recent editsdifficult.In addition to this, MediaWiki attempts to ensure that the user seesevents occurring on the wiki in chronological order. A few seconds of lagcan be tolerated, as long as the user sees a consistent picture fromsubsequent requests. This is done by saving the master binlog positionin the session, and then at the start of each request, waiting for theslave to catch up to that position before doing any reads from it. Ifthis wait times out, reads are allowed anyway, but the request isconsidered to be in "lagged slave mode". Lagged slave mode can bechecked by calling wfGetLB() ->getLaggedSlaveMode(). The onlypractical consequence at present is a warning displayed in the pagefooter.===Lag avoidance===To avoid excessive lag, queries which write large numbers of rows shouldbe split up, generally to write one row at a time. Multi-row INSERT ...SELECT queries are the worst offenders should be avoided altogether.Instead do the select first and then the insert.===Working with lag===Despite our best efforts, it 's not practical to guarantee a low-lagenvironment. Lag will usually be less than one second, but mayoccasionally be up to 30 seconds. For scalability, it 's very importantto keep load on the master low, so simply sending all your queries tothe master is not the answer. So when you have a genuine need forup-to-date data, the following approach is advised:1) Do a quick query to the master for a sequence number or timestamp 2) Run the full query on the slave and check if it matches the data you gotfrom the master 3) If it doesn 't, run the full query on the masterTo avoid swamping the master every time the slaves lag, use of thisapproach should be kept to a minimum. In most cases you should just readfrom the slave and let the user deal with the delay.------------------------------------------------------------------------ Lock contention------------------------------------------------------------------------Due to the high write rate on Wikipedia(and some other wikis), MediaWiki developers need to be very careful to structure their writesto avoid long-lasting locks. By default, MediaWiki opens a transactionat the first query, and commits it before the output is sent. Locks willbe held from the time when the query is done until the commit. So youcan reduce lock time by doing as much processing as possible before youdo your write queries.Often this approach is not good enough, and it becomes necessary toenclose small groups of queries in their own transaction. Use thefollowing syntax:$dbw=wfGetDB(DB_MASTER
deferred txt A few of the database updates required by various functions here can be deferred until after the result page is displayed to the user For updating the view updating the linked to tables after a save
Definition deferred.txt:5
deferred txt A few of the database updates required by various functions here can be deferred until after the result page is displayed to the user For updating the view updating the linked to tables after a etc PHP does not yet have any way to tell the server to actually return and disconnect while still running these but it might have such a feature in the future We handle these by creating a deferred update object and putting those objects on a global list
Definition deferred.txt:11
deferred txt A few of the database updates required by various functions here can be deferred until after the result page is displayed to the user For updating the view updating the linked to tables after a etc PHP does not yet have any way to tell the server to actually return and disconnect while still running these but it might have such a feature in the future We handle these by creating a deferred update object and putting those objects on a global then executing the whole list after the page is displayed We don t do anything smart like collating updates to the same table or such because the list is almost always going to have just one item on if that
Definition deferred.txt:13
this class mediates it Skin Encapsulates a look and feel for the wiki All of the functions that render HTML and make choices about how to render it are here and are called from various other places when needed(most notably, OutputPage::addWikiText()). The StandardSkin object is a complete implementation
design txt This is a brief overview of the new design More thorough and up to date information is available on the documentation wiki at etc Handles the details of getting and saving to the user table of the and dealing with sessions and cookies OutputPage Encapsulates the entire HTML page that will be sent in response to any server request It is used by calling its functions to add in any and then calling but I prefer the flexibility This should also do the output encoding The system allocates a global one in $wgOut Title Represents the title of an and does all the work of translating among various forms such as plain database key
Definition design.txt:26
This document is intended to provide useful advice for parties seeking to redistribute MediaWiki to end users It s targeted particularly at maintainers for Linux since it s been observed that distribution packages of MediaWiki often break We ve consistently had to recommend that users seeking support use official tarballs instead of their distribution s and this often solves whatever problem the user is having It would be nice if this could such as
globals txt Globals are evil The original MediaWiki code relied on globals for processing context far too often MediaWiki development since then has been a story of slowly moving context out of global variables and into objects Storing processing context in object member variables allows those objects to be reused in a much more flexible way Consider the elegance of
database rows
Definition globals.txt:10
the array() calling protocol came about after MediaWiki 1.4rc1.
in this case you re responsible for computing and outputting the entire conflict i e
Definition hooks.txt:1421
An extension or a local will often add custom code to the function with or without a global variable For someone wanting email notification when an article is shown may add
Definition hooks.txt:56
null for the local wiki Added in
Definition hooks.txt:1591
this hook is for auditing only RecentChangesLinked and Watchlist RecentChangesLinked and Watchlist Do not use this to implement individual filters if they are compatible with the ChangesListFilter and ChangesListFilterGroup structure use sub classes of those in conjunction with the ChangesListSpecialPageStructuredFilters hook This hook can be used to implement filters that do not implement that or custom behavior that is not an individual filter e g Watchlist and Watchlist you will want to construct new ChangesListBooleanFilter or ChangesListStringOptionsFilter objects When constructing them
Definition hooks.txt:1018
Status::newGood()` to allow deletion, and then `return false` from the hook function. Ensure you consume the 'ChangeTagAfterDelete' hook to carry out custom deletion actions. $tag:name of the tag $user:user initiating the action & $status:Status object. See above. 'ChangeTagsListActive':Allows you to nominate which of the tags your extension uses are in active use. & $tags:list of all active tags. Append to this array. 'ChangeTagsAfterUpdateTags':Called after tags have been updated with the ChangeTags::updateTags function. Params:$addedTags:tags effectively added in the update $removedTags:tags effectively removed in the update $prevTags:tags that were present prior to the update $rc_id:recentchanges table id $rev_id:revision table id $log_id:logging table id $params:tag params $rc:RecentChange being tagged when the tagging accompanies the action or null $user:User who performed the tagging when the tagging is subsequent to the action or null 'ChangeTagsAllowedAdd':Called when checking if a user can add tags to a change. & $allowedTags:List of all the tags the user is allowed to add. Any tags the user wants to add( $addTags) that are not in this array will cause it to fail. You may add or remove tags to this array as required. $addTags:List of tags user intends to add. $user:User who is adding the tags. 'ChangeUserGroups':Called before user groups are changed. $performer:The User who will perform the change $user:The User whose groups will be changed & $add:The groups that will be added & $remove:The groups that will be removed 'Collation::factory':Called if $wgCategoryCollation is an unknown collation. $collationName:Name of the collation in question & $collationObject:Null. Replace with a subclass of the Collation class that implements the collation given in $collationName. 'ConfirmEmailComplete':Called after a user 's email has been confirmed successfully. $user:user(object) whose email is being confirmed 'ContentAlterParserOutput':Modify parser output for a given content object. Called by Content::getParserOutput after parsing has finished. Can be used for changes that depend on the result of the parsing but have to be done before LinksUpdate is called(such as adding tracking categories based on the rendered HTML). $content:The Content to render $title:Title of the page, as context $parserOutput:ParserOutput to manipulate 'ContentGetParserOutput':Customize parser output for a given content object, called by AbstractContent::getParserOutput. May be used to override the normal model-specific rendering of page content. $content:The Content to render $title:Title of the page, as context $revId:The revision ID, as context $options:ParserOptions for rendering. To avoid confusing the parser cache, the output can only depend on parameters provided to this hook function, not on global state. $generateHtml:boolean, indicating whether full HTML should be generated. If false, generation of HTML may be skipped, but other information should still be present in the ParserOutput object. & $output:ParserOutput, to manipulate or replace 'ContentHandlerDefaultModelFor':Called when the default content model is determined for a given title. May be used to assign a different model for that title. $title:the Title in question & $model:the model name. Use with CONTENT_MODEL_XXX constants. 'ContentHandlerForModelID':Called when a ContentHandler is requested for a given content model name, but no entry for that model exists in $wgContentHandlers. Note:if your extension implements additional models via this hook, please use GetContentModels hook to make them known to core. $modeName:the requested content model name & $handler:set this to a ContentHandler object, if desired. 'ContentModelCanBeUsedOn':Called to determine whether that content model can be used on a given page. This is especially useful to prevent some content models to be used in some special location. $contentModel:ID of the content model in question $title:the Title in question. & $ok:Output parameter, whether it is OK to use $contentModel on $title. Handler functions that modify $ok should generally return false to prevent further hooks from further modifying $ok. 'ContribsPager::getQueryInfo':Before the contributions query is about to run & $pager:Pager object for contributions & $queryInfo:The query for the contribs Pager 'ContribsPager::reallyDoQuery':Called before really executing the query for My Contributions & $data:an array of results of all contribs queries $pager:The ContribsPager object hooked into $offset:Index offset, inclusive $limit:Exact query limit $descending:Query direction, false for ascending, true for descending 'ContributionsLineEnding':Called before a contributions HTML line is finished $page:SpecialPage object for contributions & $ret:the HTML line $row:the DB row for this line & $classes:the classes to add to the surrounding< li > & $attribs:associative array of other HTML attributes for the< li > element. Currently only data attributes reserved to MediaWiki are allowed(see Sanitizer::isReservedDataAttribute). 'ContributionsToolLinks':Change tool links above Special:Contributions $id:User identifier $title:User page title & $tools:Array of tool links $specialPage:SpecialPage instance for context and services. Can be either SpecialContributions or DeletedContributionsPage. Extensions should type hint against a generic SpecialPage though. 'ConvertContent':Called by AbstractContent::convert when a conversion to another content model is requested. Handler functions that modify $result should generally return false to disable further attempts at conversion. $content:The Content object to be converted. $toModel:The ID of the content model to convert to. $lossy:boolean indicating whether lossy conversion is allowed. & $result:Output parameter, in case the handler function wants to provide a converted Content object. Note that $result->getContentModel() must return $toModel. 'CustomEditor':When invoking the page editor Return true to allow the normal editor to be used, or false if implementing a custom editor, e.g. for a special namespace, etc. $article:Article being edited $user:User performing the edit 'DatabaseOraclePostInit':Called after initialising an Oracle database $db:the DatabaseOracle object 'DeletedContribsPager::reallyDoQuery':Called before really executing the query for Special:DeletedContributions Similar to ContribsPager::reallyDoQuery & $data:an array of results of all contribs queries $pager:The DeletedContribsPager object hooked into $offset:Index offset, inclusive $limit:Exact query limit $descending:Query direction, false for ascending, true for descending 'DeletedContributionsLineEnding':Called before a DeletedContributions HTML line is finished. Similar to ContributionsLineEnding $page:SpecialPage object for DeletedContributions & $ret:the HTML line $row:the DB row for this line & $classes:the classes to add to the surrounding< li > & $attribs:associative array of other HTML attributes for the< li > element. Currently only data attributes reserved to MediaWiki are allowed(see Sanitizer::isReservedDataAttribute). 'DeleteUnknownPreferences':Called by the cleanupPreferences.php maintenance script to build a WHERE clause with which to delete preferences that are not known about. This hook is used by extensions that have dynamically-named preferences that should not be deleted in the usual cleanup process. For example, the Gadgets extension creates preferences prefixed with 'gadget-', and so anything with that prefix is excluded from the deletion. &where:An array that will be passed as the $cond parameter to IDatabase::select() to determine what will be deleted from the user_properties table. $db:The IDatabase object, useful for accessing $db->buildLike() etc. 'DifferenceEngineAfterLoadNewText':called in DifferenceEngine::loadNewText() after the new revision 's content has been loaded into the class member variable $differenceEngine->mNewContent but before returning true from this function. $differenceEngine:DifferenceEngine object 'DifferenceEngineLoadTextAfterNewContentIsLoaded':called in DifferenceEngine::loadText() after the new revision 's content has been loaded into the class member variable $differenceEngine->mNewContent but before checking if the variable 's value is null. This hook can be used to inject content into said class member variable. $differenceEngine:DifferenceEngine object 'DifferenceEngineMarkPatrolledLink':Allows extensions to change the "mark as patrolled" link which is shown both on the diff header as well as on the bottom of a page, usually wrapped in a span element which has class="patrollink". $differenceEngine:DifferenceEngine object & $markAsPatrolledLink:The "mark as patrolled" link HTML(string) $rcid:Recent change ID(rc_id) for this change(int) 'DifferenceEngineMarkPatrolledRCID':Allows extensions to possibly change the rcid parameter. For example the rcid might be set to zero due to the user being the same as the performer of the change but an extension might still want to show it under certain conditions. & $rcid:rc_id(int) of the change or 0 $differenceEngine:DifferenceEngine object $change:RecentChange object $user:User object representing the current user 'DifferenceEngineNewHeader':Allows extensions to change the $newHeader variable, which contains information about the new revision, such as the revision 's author, whether the revision was marked as a minor edit or not, etc. $differenceEngine:DifferenceEngine object & $newHeader:The string containing the various #mw-diff-otitle[1-5] divs, which include things like revision author info, revision comment, RevisionDelete link and more $formattedRevisionTools:Array containing revision tools, some of which may have been injected with the DiffRevisionTools hook $nextlink:String containing the link to the next revision(if any) $status
Definition hooks.txt:1255
We ve cleaned up the code here by removing clumps of infrequently used code and moving them off somewhere else It s much easier for someone working with this code to see what s _really_ going on
Definition hooks.txt:86
returning false will NOT prevent logging $e
Definition hooks.txt:2176
injection txt This is an overview of how MediaWiki makes use of dependency injection The design described here grew from the discussion of RFC T384 The term dependency this means that anything an object needs to operate should be injected from the the object itself should only know narrow no concrete implementation of the logic it relies on The requirement to inject everything typically results in an architecture that based on two main types of and essentially stateless service objects that use other service objects to operate on the value objects As of the beginning MediaWiki is only starting to use the DI approach Much of the code still relies on global state or direct resulting in a highly cyclical dependency which acts as the top level factory for services in MediaWiki which can be used to gain access to default instances of various services MediaWikiServices however also allows new services to be defined and default services to be redefined Services are defined or redefined by providing a callback the instantiator that will return a new instance of the service When it will create an instance of MediaWikiServices and populate it with the services defined in the files listed by thereby bootstrapping the DI framework Per $wgServiceWiringFiles lists includes ServiceWiring which defines all default service and specifies how they depend on each other("wiring"). When a new service is added to MediaWiki core
injection txt This is an overview of how MediaWiki makes use of dependency injection The design described here grew from the discussion of RFC T384 The term dependency this means that anything an object needs to operate should be injected from the the object itself should only know narrow no concrete implementation of the logic it relies on The requirement to inject everything typically results in an architecture that based on two main types of and essentially stateless service objects that use other service objects to operate on the value objects As of the beginning MediaWiki is only starting to use the DI approach Much of the code still relies on global state or direct resulting in a highly cyclical dependency which acts as the top level factory for services in MediaWiki which can be used to gain access to default instances of various services MediaWikiServices however also allows new services to be defined and default services to be redefined Services are defined or redefined by providing a callback the instantiator that will return a new instance of the service When it will create an instance of MediaWikiServices and populate it with the services defined in the files listed by thereby bootstrapping the DI framework Per $wgServiceWiringFiles lists includes ServiceWiring php
Definition injection.txt:37
MediaWiki has optional support for a high distributed memory object caching system For general information on but for a larger site with heavy like it should help lighten the load on the database servers by caching data and objects in Ubuntu and probably other Linux distributions If there s no package available for your you can compile it from epoll rt patch for Linux is current Memcached and libevent are under BSD style licenses The server should run on Linux and other Unix like systems you can run multiple servers on one machine or on multiple machines on a network
Definition memcached.txt:33
skin txt MediaWiki includes four core it has been set as the default in MediaWiki since the replacing Monobook it had been the default skin since then
Definition skin.txt:11