includes/jobqueue/jobs/RecentChangesUpdateJob.php

   1 <?php
   2 /**
   3  * This program is free software; you can redistribute it and/or modify
   4  * it under the terms of the GNU General Public License as published by
   5  * the Free Software Foundation; either version 2 of the License, or
   6  * (at your option) any later version.
   7  *
   8  * This program is distributed in the hope that it will be useful,
   9  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  10  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  11  * GNU General Public License for more details.
  12  *
  13  * You should have received a copy of the GNU General Public License along
  14  * with this program; if not, write to the Free Software Foundation, Inc.,
  15  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  16  * http://www.gnu.org/copyleft/gpl.html
  17  *
  18  * @file
  19  * @ingroup JobQueue
  20  */
  21 use MediaWiki\MediaWikiServices;
  22 use Wikimedia\Rdbms\DBReplicationWaitError;
  23
  24 /**
  25  * Job for pruning recent changes
  26  *
  27  * @ingroup JobQueue
  28  * @since 1.25
  29  */
  30 class RecentChangesUpdateJob extends Job {
  31         function __construct( Title $title, array $params ) {
  32                 parent::__construct( 'recentChangesUpdate', $title, $params );
  33
  34                 if ( !isset( $params['type'] ) ) {
  35                         throw new Exception( "Missing 'type' parameter." );
  36                 }
  37
  38                 $this->executionFlags |= self::JOB_NO_EXPLICIT_TRX_ROUND;
  39                 $this->removeDuplicates = true;
  40         }
  41
  42         /**
  43          * @return RecentChangesUpdateJob
  44          */
  45         final public static function newPurgeJob() {
  46                 return new self(
  47                         SpecialPage::getTitleFor( 'Recentchanges' ), [ 'type' => 'purge' ]
  48                 );
  49         }
  50
  51         /**
  52          * @return RecentChangesUpdateJob
  53          * @since 1.26
  54          */
  55         final public static function newCacheUpdateJob() {
  56                 return new self(
  57                         SpecialPage::getTitleFor( 'Recentchanges' ), [ 'type' => 'cacheUpdate' ]
  58                 );
  59         }
  60
  61         public function run() {
  62                 if ( $this->params['type'] === 'purge' ) {
  63                         $this->purgeExpiredRows();
  64                 } elseif ( $this->params['type'] === 'cacheUpdate' ) {
  65                         $this->updateActiveUsers();
  66                 } else {
  67                         throw new InvalidArgumentException(
  68                                 "Invalid 'type' parameter '{$this->params['type']}'." );
  69                 }
  70
  71                 return true;
  72         }
  73
  74         protected function purgeExpiredRows() {
  75                 global $wgRCMaxAge, $wgUpdateRowsPerQuery;
  76
  77                 $lockKey = wfWikiID() . ':recentchanges-prune';
  78
  79                 $dbw = wfGetDB( DB_MASTER );
  80                 if ( !$dbw->lock( $lockKey, __METHOD__, 0 ) ) {
  81                         // already in progress
  82                         return;
  83                 }
  84
  85                 $factory = MediaWikiServices::getInstance()->getDBLoadBalancerFactory();
  86                 $ticket = $factory->getEmptyTransactionTicket( __METHOD__ );
  87                 $cutoff = $dbw->timestamp( time() - $wgRCMaxAge );
  88                 $rcQuery = RecentChange::getQueryInfo();
  89                 do {
  90                         $rcIds = [];
  91                         $rows = [];
  92                         $res = $dbw->select(
  93                                 $rcQuery['tables'],
  94                                 $rcQuery['fields'],
  95                                 [ 'rc_timestamp < ' . $dbw->addQuotes( $cutoff ) ],
  96                                 __METHOD__,
  97                                 [ 'LIMIT' => $wgUpdateRowsPerQuery ],
  98                                 $rcQuery['joins']
  99                         );
 100                         foreach ( $res as $row ) {
 101                                 $rcIds[] = $row->rc_id;
 102                                 $rows[] = $row;
 103                         }
 104                         if ( $rcIds ) {
 105                                 $dbw->delete( 'recentchanges', [ 'rc_id' => $rcIds ], __METHOD__ );
 106                                 Hooks::run( 'RecentChangesPurgeRows', [ $rows ] );
 107                                 // There might be more, so try waiting for replica DBs
 108                                 try {
 109                                         $factory->commitAndWaitForReplication(
 110                                                 __METHOD__, $ticket, [ 'timeout' => 3 ]
 111                                         );
 112                                 } catch ( DBReplicationWaitError $e ) {
 113                                         // Another job will continue anyway
 114                                         break;
 115                                 }
 116                         }
 117                 } while ( $rcIds );
 118
 119                 $dbw->unlock( $lockKey, __METHOD__ );
 120         }
 121
 122         protected function updateActiveUsers() {
 123                 global $wgActiveUserDays;
 124
 125                 // Users that made edits at least this many days ago are "active"
 126                 $days = $wgActiveUserDays;
 127                 // Pull in the full window of active users in this update
 128                 $window = $wgActiveUserDays * 86400;
 129
 130                 $dbw = wfGetDB( DB_MASTER );
 131                 $factory = MediaWikiServices::getInstance()->getDBLoadBalancerFactory();
 132                 $ticket = $factory->getEmptyTransactionTicket( __METHOD__ );
 133
 134                 $lockKey = wfWikiID() . '-activeusers';
 135                 if ( !$dbw->lock( $lockKey, __METHOD__, 0 ) ) {
 136                         // Exclusive update (avoids duplicate entries)… it's usually fine to just
 137                         // drop out here, if the Job is already running.
 138                         return;
 139                 }
 140
 141                 // Long-running queries expected
 142                 $dbw->setSessionOptions( [ 'connTimeout' => 900 ] );
 143
 144                 $nowUnix = time();
 145                 // Get the last-updated timestamp for the cache
 146                 $cTime = $dbw->selectField( 'querycache_info',
 147                         'qci_timestamp',
 148                         [ 'qci_type' => 'activeusers' ]
 149                 );
 150                 $cTimeUnix = $cTime ? wfTimestamp( TS_UNIX, $cTime ) : 1;
 151
 152                 // Pick the date range to fetch from. This is normally from the last
 153                 // update to till the present time, but has a limited window for sanity.
 154                 // If the window is limited, multiple runs are need to fully populate it.
 155                 $sTimestamp = max( $cTimeUnix, $nowUnix - $days * 86400 );
 156                 $eTimestamp = min( $sTimestamp + $window, $nowUnix );
 157
 158                 // Get all the users active since the last update
 159                 $actorQuery = ActorMigration::newMigration()->getJoin( 'rc_user' );
 160                 $res = $dbw->select(
 161                         [ 'recentchanges' ] + $actorQuery['tables'],
 162                         [
 163                                 'rc_user_text' => $actorQuery['fields']['rc_user_text'],
 164                                 'lastedittime' => 'MAX(rc_timestamp)'
 165                         ],
 166                         [
 167                                 $actorQuery['fields']['rc_user'] . ' > 0', // actual accounts
 168                                 'rc_type != ' . $dbw->addQuotes( RC_EXTERNAL ), // no wikidata
 169                                 'rc_log_type IS NULL OR rc_log_type != ' . $dbw->addQuotes( 'newusers' ),
 170                                 'rc_timestamp >= ' . $dbw->addQuotes( $dbw->timestamp( $sTimestamp ) ),
 171                                 'rc_timestamp <= ' . $dbw->addQuotes( $dbw->timestamp( $eTimestamp ) )
 172                         ],
 173                         __METHOD__,
 174                         [
 175                                 'GROUP BY' => [ 'rc_user_text' ],
 176                                 'ORDER BY' => 'NULL' // avoid filesort
 177                         ],
 178                         $actorQuery['joins']
 179                 );
 180                 $names = [];
 181                 foreach ( $res as $row ) {
 182                         $names[$row->rc_user_text] = $row->lastedittime;
 183                 }
 184
 185                 // Find which of the recently active users are already accounted for
 186                 if ( count( $names ) ) {
 187                         $res = $dbw->select( 'querycachetwo',
 188                                 [ 'user_name' => 'qcc_title' ],
 189                                 [
 190                                         'qcc_type' => 'activeusers',
 191                                         'qcc_namespace' => NS_USER,
 192                                         'qcc_title' => array_keys( $names ),
 193                                         'qcc_value >= ' . $dbw->addQuotes( $nowUnix - $days * 86400 ), // TS_UNIX
 194                                  ],
 195                                 __METHOD__
 196                         );
 197                         // Note: In order for this to be actually consistent, we would need
 198                         // to update these rows with the new lastedittime.
 199                         foreach ( $res as $row ) {
 200                                 unset( $names[$row->user_name] );
 201                         }
 202                 }
 203
 204                 // Insert the users that need to be added to the list
 205                 if ( count( $names ) ) {
 206                         $newRows = [];
 207                         foreach ( $names as $name => $lastEditTime ) {
 208                                 $newRows[] = [
 209                                         'qcc_type' => 'activeusers',
 210                                         'qcc_namespace' => NS_USER,
 211                                         'qcc_title' => $name,
 212                                         'qcc_value' => wfTimestamp( TS_UNIX, $lastEditTime ),
 213                                         'qcc_namespacetwo' => 0, // unused
 214                                         'qcc_titletwo' => '' // unused
 215                                 ];
 216                         }
 217                         foreach ( array_chunk( $newRows, 500 ) as $rowBatch ) {
 218                                 $dbw->insert( 'querycachetwo', $rowBatch, __METHOD__ );
 219                                 $factory->commitAndWaitForReplication( __METHOD__, $ticket );
 220                         }
 221                 }
 222
 223                 // If a transaction was already started, it might have an old
 224                 // snapshot, so kludge the timestamp range back as needed.
 225                 $asOfTimestamp = min( $eTimestamp, (int)$dbw->trxTimestamp() );
 226
 227                 // Touch the data freshness timestamp
 228                 $dbw->replace( 'querycache_info',
 229                         [ 'qci_type' ],
 230                         [ 'qci_type' => 'activeusers',
 231                                 'qci_timestamp' => $dbw->timestamp( $asOfTimestamp ) ], // not always $now
 232                         __METHOD__
 233                 );
 234
 235                 $dbw->unlock( $lockKey, __METHOD__ );
 236
 237                 // Rotate out users that have not edited in too long (according to old data set)
 238                 $dbw->delete( 'querycachetwo',
 239                         [
 240                                 'qcc_type' => 'activeusers',
 241                                 'qcc_value < ' . $dbw->addQuotes( $nowUnix - $days * 86400 ) // TS_UNIX
 242                         ],
 243                         __METHOD__
 244                 );
 245         }
 246 }