maintenance/initEditCount.php

   1 <?php
   2 /**
   3  * Init the user_editcount database field based on the number of rows in the
   4  * revision table.
   5  *
   6  * This program is free software; you can redistribute it and/or modify
   7  * it under the terms of the GNU General Public License as published by
   8  * the Free Software Foundation; either version 2 of the License, or
   9  * (at your option) any later version.
  10  *
  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 General Public License for more details.
  15  *
  16  * You should have received a copy of the GNU General Public License along
  17  * with this program; if not, write to the Free Software Foundation, Inc.,
  18  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  19  * http://www.gnu.org/copyleft/gpl.html
  20  *
  21  * @file
  22  * @ingroup Maintenance
  23  */
  24
  25 require_once __DIR__ . '/Maintenance.php';
  26
  27 use MediaWiki\MediaWikiServices;
  28
  29 class InitEditCount extends Maintenance {
  30         public function __construct() {
  31                 parent::__construct();
  32                 $this->addOption( 'quick', 'Force the update to be done in a single query' );
  33                 $this->addOption( 'background', 'Force replication-friendly mode; may be inefficient but
  34                 avoids locking tables or lagging replica DBs with large updates;
  35                 calculates counts on a replica DB if possible.
  36
  37 Background mode will be automatically used if multiple servers are listed
  38 in the load balancer, usually indicating a replication environment.' );
  39                 $this->addDescription( 'Batch-recalculate user_editcount fields from the revision table' );
  40         }
  41
  42         public function execute() {
  43                 global $wgActorTableSchemaMigrationStage;
  44
  45                 $dbw = $this->getDB( DB_MASTER );
  46
  47                 // Autodetect mode...
  48                 if ( $this->hasOption( 'background' ) ) {
  49                         $backgroundMode = true;
  50                 } elseif ( $this->hasOption( 'quick' ) ) {
  51                         $backgroundMode = false;
  52                 } else {
  53                         $lb = MediaWikiServices::getInstance()->getDBLoadBalancer();
  54                         $backgroundMode = $lb->getServerCount() > 1;
  55                 }
  56
  57                 $actorQuery = ActorMigration::newMigration()->getJoin( 'rev_user' );
  58
  59                 $needSpecialQuery = ( $wgActorTableSchemaMigrationStage !== MIGRATION_OLD &&
  60                         $wgActorTableSchemaMigrationStage !== MIGRATION_NEW );
  61                 if ( $needSpecialQuery ) {
  62                         foreach ( $actorQuery['joins'] as &$j ) {
  63                                 $j[0] = 'JOIN'; // replace LEFT JOIN
  64                         }
  65                         unset( $j );
  66                 }
  67
  68                 if ( $backgroundMode ) {
  69                         $this->output( "Using replication-friendly background mode...\n" );
  70
  71                         $dbr = $this->getDB( DB_REPLICA );
  72                         $chunkSize = 100;
  73                         $lastUser = $dbr->selectField( 'user', 'MAX(user_id)', '', __METHOD__ );
  74
  75                         $start = microtime( true );
  76                         $migrated = 0;
  77                         for ( $min = 0; $min <= $lastUser; $min += $chunkSize ) {
  78                                 $max = $min + $chunkSize;
  79
  80                                 if ( $needSpecialQuery ) {
  81                                         // Use separate subqueries to collect counts with the old
  82                                         // and new schemas, to avoid having to do whole-table scans.
  83                                         $result = $dbr->select(
  84                                                 [
  85                                                         'user',
  86                                                         'rev1' => '('
  87                                                                 . $dbr->selectSQLText(
  88                                                                         [ 'revision', 'revision_actor_temp' ],
  89                                                                         [ 'rev_user', 'ct' => 'COUNT(*)' ],
  90                                                                         [
  91                                                                                 "rev_user > $min AND rev_user <= $max",
  92                                                                                 'revactor_rev' => null,
  93                                                                         ],
  94                                                                         __METHOD__,
  95                                                                         [ 'GROUP BY' => 'rev_user' ],
  96                                                                         [ 'revision_actor_temp' => [ 'LEFT JOIN', 'revactor_rev = rev_id' ] ]
  97                                                                 ) . ')',
  98                                                         'rev2' => '('
  99                                                                 . $dbr->selectSQLText(
 100                                                                         [ 'revision' ] + $actorQuery['tables'],
 101                                                                         [ 'actor_user', 'ct' => 'COUNT(*)' ],
 102                                                                         "actor_user > $min AND actor_user <= $max",
 103                                                                         __METHOD__,
 104                                                                         [ 'GROUP BY' => 'actor_user' ],
 105                                                                         $actorQuery['joins']
 106                                                                 ) . ')',
 107                                                 ],
 108                                                 [ 'user_id', 'user_editcount' => 'COALESCE(rev1.ct,0) + COALESCE(rev2.ct,0)' ],
 109                                                 "user_id > $min AND user_id <= $max",
 110                                                 __METHOD__,
 111                                                 [],
 112                                                 [
 113                                                         'rev1' => [ 'LEFT JOIN', 'user_id = rev_user' ],
 114                                                         'rev2' => [ 'LEFT JOIN', 'user_id = actor_user' ],
 115                                                 ]
 116                                         );
 117                                 } else {
 118                                         $revUser = $actorQuery['fields']['rev_user'];
 119                                         $result = $dbr->select(
 120                                                 [ 'user', 'rev' => [ 'revision' ] + $actorQuery['tables'] ],
 121                                                 [ 'user_id', 'user_editcount' => "COUNT($revUser)" ],
 122                                                 "user_id > $min AND user_id <= $max",
 123                                                 __METHOD__,
 124                                                 [ 'GROUP BY' => 'user_id' ],
 125                                                 [ 'rev' => [ 'LEFT JOIN', "user_id = $revUser" ] ] + $actorQuery['joins']
 126                                         );
 127                                 }
 128
 129                                 foreach ( $result as $row ) {
 130                                         $dbw->update( 'user',
 131                                                 [ 'user_editcount' => $row->user_editcount ],
 132                                                 [ 'user_id' => $row->user_id ],
 133                                                 __METHOD__ );
 134                                         ++$migrated;
 135                                 }
 136
 137                                 $delta = microtime( true ) - $start;
 138                                 $rate = ( $delta == 0.0 ) ? 0.0 : $migrated / $delta;
 139                                 $this->output( sprintf( "%s %d (%0.1f%%) done in %0.1f secs (%0.3f accounts/sec).\n",
 140                                         wfWikiID(),
 141                                         $migrated,
 142                                         min( $max, $lastUser ) / $lastUser * 100.0,
 143                                         $delta,
 144                                         $rate ) );
 145
 146                                 wfWaitForSlaves();
 147                         }
 148                 } else {
 149                         $this->output( "Using single-query mode...\n" );
 150
 151                         $user = $dbw->tableName( 'user' );
 152                         if ( $needSpecialQuery ) {
 153                                 $subquery1 = $dbw->selectSQLText(
 154                                         [ 'revision', 'revision_actor_temp' ],
 155                                         [ 'COUNT(*)' ],
 156                                         [
 157                                                 'user_id = rev_user',
 158                                                 'revactor_rev' => null,
 159                                         ],
 160                                         __METHOD__,
 161                                         [],
 162                                         [ 'revision_actor_temp' => [ 'LEFT JOIN', 'revactor_rev = rev_id' ] ]
 163                                 );
 164                                 $subquery2 = $dbw->selectSQLText(
 165                                         [ 'revision' ] + $actorQuery['tables'],
 166                                         [ 'COUNT(*)' ],
 167                                         'user_id = actor_user',
 168                                         __METHOD__,
 169                                         [],
 170                                         $actorQuery['joins']
 171                                 );
 172                                 $dbw->query(
 173                                         "UPDATE $user SET user_editcount=($subquery1) + ($subquery2)",
 174                                         __METHOD__
 175                                 );
 176                         } else {
 177                                 $subquery = $dbw->selectSQLText(
 178                                         [ 'revision' ] + $actorQuery['tables'],
 179                                         [ 'COUNT(*)' ],
 180                                         [ 'user_id = ' . $actorQuery['fields']['rev_user'] ],
 181                                         __METHOD__,
 182                                         [],
 183                                         $actorQuery['joins']
 184                                 );
 185                                 $dbw->query( "UPDATE $user SET user_editcount=($subquery)", __METHOD__ );
 186                         }
 187                 }
 188
 189                 $this->output( "Done!\n" );
 190         }
 191 }
 192
 193 $maintClass = InitEditCount::class;
 194 require_once RUN_MAINTENANCE_IF_MAIN;