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  * @author Aaron Schulz
  20  * @ingroup JobQueue
  21  */
  22 use MediaWiki\MediaWikiServices;
  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->removeDuplicates = true;
  39         }
  40
  41         /**
  42          * @return RecentChangesUpdateJob
  43          */
  44         final public static function newPurgeJob() {
  45                 return new self(
  46                         SpecialPage::getTitleFor( 'Recentchanges' ), [ 'type' => 'purge' ]
  47                 );
  48         }
  49
  50         /**
  51          * @return RecentChangesUpdateJob
  52          * @since 1.26
  53          */
  54         final public static function newCacheUpdateJob() {
  55                 return new self(
  56                         SpecialPage::getTitleFor( 'Recentchanges' ), [ 'type' => 'cacheUpdate' ]
  57                 );
  58         }
  59
  60         public function run() {
  61                 if ( $this->params['type'] === 'purge' ) {
  62                         $this->purgeExpiredRows();
  63                 } elseif ( $this->params['type'] === 'cacheUpdate' ) {
  64                         $this->updateActiveUsers();
  65                 } else {
  66                         throw new InvalidArgumentException(
  67                                 "Invalid 'type' parameter '{$this->params['type']}'." );
  68                 }
  69
  70                 return true;
  71         }
  72
  73         protected function purgeExpiredRows() {
  74                 global $wgRCMaxAge, $wgUpdateRowsPerQuery;
  75
  76                 $lockKey = wfWikiID() . ':recentchanges-prune';
  77
  78                 $dbw = wfGetDB( DB_MASTER );
  79                 if ( !$dbw->lockIsFree( $lockKey, __METHOD__ )
  80                         || !$dbw->lock( $lockKey, __METHOD__, 1 )
  81                 ) {
  82                         return; // already in progress
  83                 }
  84
  85                 $factory = MediaWikiServices::getInstance()->getDBLoadBalancerFactory();
  86                 $ticket = $factory->getEmptyTransactionTicket( __METHOD__ );
  87                 $cutoff = $dbw->timestamp( time() - $wgRCMaxAge );
  88                 do {
  89                         $rcIds = $dbw->selectFieldValues( 'recentchanges',
  90                                 'rc_id',
  91                                 [ 'rc_timestamp < ' . $dbw->addQuotes( $cutoff ) ],
  92                                 __METHOD__,
  93                                 [ 'LIMIT' => $wgUpdateRowsPerQuery ]
  94                         );
  95                         if ( $rcIds ) {
  96                                 $dbw->delete( 'recentchanges', [ 'rc_id' => $rcIds ], __METHOD__ );
  97                                 // There might be more, so try waiting for replica DBs
  98                                 try {
  99                                         $factory->commitAndWaitForReplication(
 100                                                 __METHOD__, $ticket, [ 'timeout' => 3 ]
 101                                         );
 102                                 } catch ( DBReplicationWaitError $e ) {
 103                                         // Another job will continue anyway
 104                                         break;
 105                                 }
 106                         }
 107                 } while ( $rcIds );
 108
 109                 $dbw->unlock( $lockKey, __METHOD__ );
 110         }
 111
 112         protected function updateActiveUsers() {
 113                 global $wgActiveUserDays;
 114
 115                 // Users that made edits at least this many days ago are "active"
 116                 $days = $wgActiveUserDays;
 117                 // Pull in the full window of active users in this update
 118                 $window = $wgActiveUserDays * 86400;
 119
 120                 $dbw = wfGetDB( DB_MASTER );
 121                 // JobRunner uses DBO_TRX, but doesn't call begin/commit itself;
 122                 // onTransactionIdle() will run immediately since there is no trx.
 123                 $dbw->onTransactionIdle(
 124                         function () use ( $dbw, $days, $window ) {
 125                                 $factory = MediaWikiServices::getInstance()->getDBLoadBalancerFactory();
 126                                 $ticket = $factory->getEmptyTransactionTicket( __METHOD__ );
 127                                 // Avoid disconnect/ping() cycle that makes locks fall off
 128                                 $dbw->setSessionOptions( [ 'connTimeout' => 900 ] );
 129
 130                                 $lockKey = wfWikiID() . '-activeusers';
 131                                 if ( !$dbw->lockIsFree( $lockKey, __METHOD__ ) || !$dbw->lock( $lockKey, __METHOD__, 1 ) ) {
 132                                         // Exclusive update (avoids duplicate entries)… it's usually fine to just drop out here,
 133                                         // if the Job is already running.
 134                                         return;
 135                                 }
 136
 137                                 $nowUnix = time();
 138                                 // Get the last-updated timestamp for the cache
 139                                 $cTime = $dbw->selectField( 'querycache_info',
 140                                         'qci_timestamp',
 141                                         [ 'qci_type' => 'activeusers' ]
 142                                 );
 143                                 $cTimeUnix = $cTime ? wfTimestamp( TS_UNIX, $cTime ) : 1;
 144
 145                                 // Pick the date range to fetch from. This is normally from the last
 146                                 // update to till the present time, but has a limited window for sanity.
 147                                 // If the window is limited, multiple runs are need to fully populate it.
 148                                 $sTimestamp = max( $cTimeUnix, $nowUnix - $days * 86400 );
 149                                 $eTimestamp = min( $sTimestamp + $window, $nowUnix );
 150
 151                                 // Get all the users active since the last update
 152                                 $res = $dbw->select(
 153                                         [ 'recentchanges' ],
 154                                         [ 'rc_user_text', 'lastedittime' => 'MAX(rc_timestamp)' ],
 155                                         [
 156                                                 'rc_user > 0', // actual accounts
 157                                                 'rc_type != ' . $dbw->addQuotes( RC_EXTERNAL ), // no wikidata
 158                                                 'rc_log_type IS NULL OR rc_log_type != ' . $dbw->addQuotes( 'newusers' ),
 159                                                 'rc_timestamp >= ' . $dbw->addQuotes( $dbw->timestamp( $sTimestamp ) ),
 160                                                 'rc_timestamp <= ' . $dbw->addQuotes( $dbw->timestamp( $eTimestamp ) )
 161                                         ],
 162                                         __METHOD__,
 163                                         [
 164                                                 'GROUP BY' => [ 'rc_user_text' ],
 165                                                 'ORDER BY' => 'NULL' // avoid filesort
 166                                         ]
 167                                 );
 168                                 $names = [];
 169                                 foreach ( $res as $row ) {
 170                                         $names[$row->rc_user_text] = $row->lastedittime;
 171                                 }
 172
 173                                 // Find which of the recently active users are already accounted for
 174                                 if ( count( $names ) ) {
 175                                         $res = $dbw->select( 'querycachetwo',
 176                                                 [ 'user_name' => 'qcc_title' ],
 177                                                 [
 178                                                         'qcc_type' => 'activeusers',
 179                                                         'qcc_namespace' => NS_USER,
 180                                                         'qcc_title' => array_keys( $names ),
 181                                                         'qcc_value >= ' . $dbw->addQuotes( $nowUnix - $days * 86400 ), // TS_UNIX
 182                                                  ],
 183                                                 __METHOD__
 184                                         );
 185                                         // Note: In order for this to be actually consistent, we would need
 186                                         // to update these rows with the new lastedittime.
 187                                         foreach ( $res as $row ) {
 188                                                 unset( $names[$row->user_name] );
 189                                         }
 190                                 }
 191
 192                                 // Insert the users that need to be added to the list
 193                                 if ( count( $names ) ) {
 194                                         $newRows = [];
 195                                         foreach ( $names as $name => $lastEditTime ) {
 196                                                 $newRows[] = [
 197                                                         'qcc_type' => 'activeusers',
 198                                                         'qcc_namespace' => NS_USER,
 199                                                         'qcc_title' => $name,
 200                                                         'qcc_value' => wfTimestamp( TS_UNIX, $lastEditTime ),
 201                                                         'qcc_namespacetwo' => 0, // unused
 202                                                         'qcc_titletwo' => '' // unused
 203                                                 ];
 204                                         }
 205                                         foreach ( array_chunk( $newRows, 500 ) as $rowBatch ) {
 206                                                 $dbw->insert( 'querycachetwo', $rowBatch, __METHOD__ );
 207                                                 $factory->commitAndWaitForReplication( __METHOD__, $ticket );
 208                                         }
 209                                 }
 210
 211                                 // If a transaction was already started, it might have an old
 212                                 // snapshot, so kludge the timestamp range back as needed.
 213                                 $asOfTimestamp = min( $eTimestamp, (int)$dbw->trxTimestamp() );
 214
 215                                 // Touch the data freshness timestamp
 216                                 $dbw->replace( 'querycache_info',
 217                                         [ 'qci_type' ],
 218                                         [ 'qci_type' => 'activeusers',
 219                                                 'qci_timestamp' => $dbw->timestamp( $asOfTimestamp ) ], // not always $now
 220                                         __METHOD__
 221                                 );
 222
 223                                 $dbw->unlock( $lockKey, __METHOD__ );
 224
 225                                 // Rotate out users that have not edited in too long (according to old data set)
 226                                 $dbw->delete( 'querycachetwo',
 227                                         [
 228                                                 'qcc_type' => 'activeusers',
 229                                                 'qcc_value < ' . $dbw->addQuotes( $nowUnix - $days * 86400 ) // TS_UNIX
 230                                         ],
 231                                         __METHOD__
 232                                 );
 233
 234                         },
 235                         __METHOD__
 236                 );
 237         }
 238 }