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
  23 /**
  24  * Job for pruning recent changes
  25  *
  26  * @ingroup JobQueue
  27  * @since 1.25
  28  */
  29 class RecentChangesUpdateJob extends Job {
  30         function __construct( $title, $params ) {
  31                 parent::__construct( 'recentChangesUpdate', $title, $params );
  32
  33                 if ( !isset( $params['type'] ) ) {
  34                         throw new Exception( "Missing 'type' parameter." );
  35                 }
  36
  37                 $this->removeDuplicates = true;
  38         }
  39
  40         /**
  41          * @return RecentChangesUpdateJob
  42          */
  43         final public static function newPurgeJob() {
  44                 return new self(
  45                         SpecialPage::getTitleFor( 'Recentchanges' ), array( 'type' => 'purge' )
  46                 );
  47         }
  48
  49         /**
  50          * @return RecentChangesUpdateJob
  51          * @since 1.26
  52          */
  53         final public static function newCacheUpdateJob() {
  54                 return new self(
  55                         SpecialPage::getTitleFor( 'Recentchanges' ), array( 'type' => 'cacheUpdate' )
  56                 );
  57         }
  58
  59         public function run() {
  60                 if ( $this->params['type'] === 'purge' ) {
  61                         $this->purgeExpiredRows();
  62                 } elseif ( $this->params['type'] === 'cacheUpdate' ) {
  63                         $this->updateActiveUsers();
  64                 } else {
  65                         throw new InvalidArgumentException(
  66                                 "Invalid 'type' parameter '{$this->params['type']}'." );
  67                 }
  68
  69                 return true;
  70         }
  71
  72         protected function purgeExpiredRows() {
  73                 global $wgRCMaxAge;
  74
  75                 $lockKey = wfWikiID() . ':recentchanges-prune';
  76
  77                 $dbw = wfGetDB( DB_MASTER );
  78                 if ( !$dbw->lock( $lockKey, __METHOD__, 1 ) ) {
  79                         return; // already in progress
  80                 }
  81                 $batchSize = 100; // Avoid slave lag
  82
  83                 $cutoff = $dbw->timestamp( time() - $wgRCMaxAge );
  84                 do {
  85                         $rcIds = $dbw->selectFieldValues( 'recentchanges',
  86                                 'rc_id',
  87                                 array( 'rc_timestamp < ' . $dbw->addQuotes( $cutoff ) ),
  88                                 __METHOD__,
  89                                 array( 'LIMIT' => $batchSize )
  90                         );
  91                         if ( $rcIds ) {
  92                                 $dbw->delete( 'recentchanges', array( 'rc_id' => $rcIds ), __METHOD__ );
  93                         }
  94                         // Commit in chunks to avoid slave lag
  95                         $dbw->commit( __METHOD__, 'flush' );
  96
  97                         if ( count( $rcIds ) === $batchSize ) {
  98                                 // There might be more, so try waiting for slaves
  99                                 if ( !wfWaitForSlaves( null, false, false, /* $timeout = */ 3 ) ) {
 100                                         // Another job will continue anyway
 101                                         break;
 102                                 }
 103                         }
 104                 } while ( $rcIds );
 105
 106                 $dbw->unlock( $lockKey, __METHOD__ );
 107         }
 108
 109         protected function updateActiveUsers() {
 110                 global $wgActiveUserDays;
 111
 112                 // Users that made edits at least this many days ago are "active"
 113                 $days = $wgActiveUserDays;
 114                 // Pull in the full window of active users in this update
 115                 $window = $wgActiveUserDays * 86400;
 116
 117                 $dbw = wfGetDB( DB_MASTER );
 118                 // JobRunner uses DBO_TRX, but doesn't call begin/commit itself;
 119                 // onTransactionIdle() will run immediately since there is no trx.
 120                 $dbw->onTransactionIdle( function() use ( $dbw, $days, $window ) {
 121                         // Avoid disconnect/ping() cycle that makes locks fall off
 122                         $dbw->setSessionOptions( array( 'connTimeout' => 900 ) );
 123
 124                         $lockKey = wfWikiID() . '-activeusers';
 125                         if ( !$dbw->lock( $lockKey, __METHOD__, 1 ) ) {
 126                                 return false; // exclusive update (avoids duplicate entries)
 127                         }
 128
 129                         $nowUnix = time();
 130                         // Get the last-updated timestamp for the cache
 131                         $cTime = $dbw->selectField( 'querycache_info',
 132                                 'qci_timestamp',
 133                                 array( 'qci_type' => 'activeusers' )
 134                         );
 135                         $cTimeUnix = $cTime ? wfTimestamp( TS_UNIX, $cTime ) : 1;
 136
 137                         // Pick the date range to fetch from. This is normally from the last
 138                         // update to till the present time, but has a limited window for sanity.
 139                         // If the window is limited, multiple runs are need to fully populate it.
 140                         $sTimestamp = max( $cTimeUnix, $nowUnix - $days * 86400 );
 141                         $eTimestamp = min( $sTimestamp + $window, $nowUnix );
 142
 143                         // Get all the users active since the last update
 144                         $res = $dbw->select(
 145                                 array( 'recentchanges' ),
 146                                 array( 'rc_user_text', 'lastedittime' => 'MAX(rc_timestamp)' ),
 147                                 array(
 148                                         'rc_user > 0', // actual accounts
 149                                         'rc_type != ' . $dbw->addQuotes( RC_EXTERNAL ), // no wikidata
 150                                         'rc_log_type IS NULL OR rc_log_type != ' . $dbw->addQuotes( 'newusers' ),
 151                                         'rc_timestamp >= ' . $dbw->addQuotes( $dbw->timestamp( $sTimestamp ) ),
 152                                         'rc_timestamp <= ' . $dbw->addQuotes( $dbw->timestamp( $eTimestamp ) )
 153                                 ),
 154                                 __METHOD__,
 155                                 array(
 156                                         'GROUP BY' => array( 'rc_user_text' ),
 157                                         'ORDER BY' => 'NULL' // avoid filesort
 158                                 )
 159                         );
 160                         $names = array();
 161                         foreach ( $res as $row ) {
 162                                 $names[$row->rc_user_text] = $row->lastedittime;
 163                         }
 164
 165                         // Rotate out users that have not edited in too long (according to old data set)
 166                         $dbw->delete( 'querycachetwo',
 167                                 array(
 168                                         'qcc_type' => 'activeusers',
 169                                         'qcc_value < ' . $dbw->addQuotes( $nowUnix - $days * 86400 ) // TS_UNIX
 170                                 ),
 171                                 __METHOD__
 172                         );
 173
 174                         // Find which of the recently active users are already accounted for
 175                         if ( count( $names ) ) {
 176                                 $res = $dbw->select( 'querycachetwo',
 177                                         array( 'user_name' => 'qcc_title' ),
 178                                         array(
 179                                                 'qcc_type' => 'activeusers',
 180                                                 'qcc_namespace' => NS_USER,
 181                                                 'qcc_title' => array_keys( $names ) ),
 182                                         __METHOD__
 183                                 );
 184                                 foreach ( $res as $row ) {
 185                                         unset( $names[$row->user_name] );
 186                                 }
 187                         }
 188
 189                         // Insert the users that need to be added to the list
 190                         if ( count( $names ) ) {
 191                                 $newRows = array();
 192                                 foreach ( $names as $name => $lastEditTime ) {
 193                                         $newRows[] = array(
 194                                                 'qcc_type' => 'activeusers',
 195                                                 'qcc_namespace' => NS_USER,
 196                                                 'qcc_title' => $name,
 197                                                 'qcc_value' => wfTimestamp( TS_UNIX, $lastEditTime ),
 198                                                 'qcc_namespacetwo' => 0, // unused
 199                                                 'qcc_titletwo' => '' // unused
 200                                         );
 201                                 }
 202                                 foreach ( array_chunk( $newRows, 500 ) as $rowBatch ) {
 203                                         $dbw->insert( 'querycachetwo', $rowBatch, __METHOD__ );
 204                                         wfWaitForSlaves();
 205                                 }
 206                         }
 207
 208                         // If a transaction was already started, it might have an old
 209                         // snapshot, so kludge the timestamp range back as needed.
 210                         $asOfTimestamp = min( $eTimestamp, (int)$dbw->trxTimestamp() );
 211
 212                         // Touch the data freshness timestamp
 213                         $dbw->replace( 'querycache_info',
 214                                 array( 'qci_type' ),
 215                                 array( 'qci_type' => 'activeusers',
 216                                         'qci_timestamp' => $dbw->timestamp( $asOfTimestamp ) ), // not always $now
 217                                 __METHOD__
 218                         );
 219
 220                         $dbw->unlock( $lockKey, __METHOD__ );
 221                 } );
 222         }
 223 }