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 class InitEditCount extends Maintenance {
  28         public function __construct() {
  29                 parent::__construct();
  30                 $this->addOption( 'quick', 'Force the update to be done in a single query' );
  31                 $this->addOption( 'background', 'Force replication-friendly mode; may be inefficient but
  32                 avoids locking tables or lagging replica DBs with large updates;
  33                 calculates counts on a replica DB if possible.
  34
  35 Background mode will be automatically used if multiple servers are listed
  36 in the load balancer, usually indicating a replication environment.' );
  37                 $this->addDescription( 'Batch-recalculate user_editcount fields from the revision table' );
  38         }
  39
  40         public function execute() {
  41                 global $wgActorTableSchemaMigrationStage;
  42
  43                 $dbw = $this->getDB( DB_MASTER );
  44
  45                 // Autodetect mode...
  46                 if ( $this->hasOption( 'background' ) ) {
  47                         $backgroundMode = true;
  48                 } elseif ( $this->hasOption( 'quick' ) ) {
  49                         $backgroundMode = false;
  50                 } else {
  51                         $backgroundMode = wfGetLB()->getServerCount() > 1;
  52                 }
  53
  54                 $actorQuery = ActorMigration::newMigration()->getJoin( 'rev_user' );
  55
  56                 $needSpecialQuery = ( $wgActorTableSchemaMigrationStage !== MIGRATION_OLD &&
  57                         $wgActorTableSchemaMigrationStage !== MIGRATION_NEW );
  58                 if ( $needSpecialQuery ) {
  59                         foreach ( $actorQuery['joins'] as &$j ) {
  60                                 $j[0] = 'JOIN'; // replace LEFT JOIN
  61                         }
  62                         unset( $j );
  63                 }
  64
  65                 if ( $backgroundMode ) {
  66                         $this->output( "Using replication-friendly background mode...\n" );
  67
  68                         $dbr = $this->getDB( DB_REPLICA );
  69                         $chunkSize = 100;
  70                         $lastUser = $dbr->selectField( 'user', 'MAX(user_id)', '', __METHOD__ );
  71
  72                         $start = microtime( true );
  73                         $migrated = 0;
  74                         for ( $min = 0; $min <= $lastUser; $min += $chunkSize ) {
  75                                 $max = $min + $chunkSize;
  76
  77                                 if ( $needSpecialQuery ) {
  78                                         // Use separate subqueries to collect counts with the old
  79                                         // and new schemas, to avoid having to do whole-table scans.
  80                                         $result = $dbr->select(
  81                                                 [
  82                                                         'user',
  83                                                         'rev1' => '('
  84                                                                 . $dbr->selectSQLText(
  85                                                                         [ 'revision', 'revision_actor_temp' ],
  86                                                                         [ 'rev_user', 'ct' => 'COUNT(*)' ],
  87                                                                         [
  88                                                                                 "rev_user > $min AND rev_user <= $max",
  89                                                                                 'revactor_rev' => null,
  90                                                                         ],
  91                                                                         __METHOD__,
  92                                                                         [ 'GROUP BY' => 'rev_user' ],
  93                                                                         [ 'revision_actor_temp' => [ 'LEFT JOIN', 'revactor_rev = rev_id' ] ]
  94                                                                 ) . ')',
  95                                                         'rev2' => '('
  96                                                                 . $dbr->selectSQLText(
  97                                                                         [ 'revision' ] + $actorQuery['tables'],
  98                                                                         [ 'actor_user', 'ct' => 'COUNT(*)' ],
  99                                                                         "actor_user > $min AND actor_user <= $max",
 100                                                                         __METHOD__,
 101                                                                         [ 'GROUP BY' => 'actor_user' ],
 102                                                                         $actorQuery['joins']
 103                                                                 ) . ')',
 104                                                 ],
 105                                                 [ 'user_id', 'user_editcount' => 'COALESCE(rev1.ct,0) + COALESCE(rev2.ct,0)' ],
 106                                                 "user_id > $min AND user_id <= $max",
 107                                                 __METHOD__,
 108                                                 [],
 109                                                 [
 110                                                         'rev1' => [ 'LEFT JOIN', 'user_id = rev_user' ],
 111                                                         'rev2' => [ 'LEFT JOIN', 'user_id = actor_user' ],
 112                                                 ]
 113                                         );
 114                                 } else {
 115                                         $revUser = $actorQuery['fields']['rev_user'];
 116                                         $result = $dbr->select(
 117                                                 [ 'user', 'rev' => [ 'revision' ] + $actorQuery['tables'] ],
 118                                                 [ 'user_id', 'user_editcount' => "COUNT($revUser)" ],
 119                                                 "user_id > $min AND user_id <= $max",
 120                                                 __METHOD__,
 121                                                 [ 'GROUP BY' => 'user_id' ],
 122                                                 [ 'rev' => [ 'LEFT JOIN', "user_id = $revUser" ] ] + $actorQuery['joins']
 123                                         );
 124                                 }
 125
 126                                 foreach ( $result as $row ) {
 127                                         $dbw->update( 'user',
 128                                                 [ 'user_editcount' => $row->user_editcount ],
 129                                                 [ 'user_id' => $row->user_id ],
 130                                                 __METHOD__ );
 131                                         ++$migrated;
 132                                 }
 133
 134                                 $delta = microtime( true ) - $start;
 135                                 $rate = ( $delta == 0.0 ) ? 0.0 : $migrated / $delta;
 136                                 $this->output( sprintf( "%s %d (%0.1f%%) done in %0.1f secs (%0.3f accounts/sec).\n",
 137                                         wfWikiID(),
 138                                         $migrated,
 139                                         min( $max, $lastUser ) / $lastUser * 100.0,
 140                                         $delta,
 141                                         $rate ) );
 142
 143                                 wfWaitForSlaves();
 144                         }
 145                 } else {
 146                         $this->output( "Using single-query mode...\n" );
 147
 148                         $user = $dbw->tableName( 'user' );
 149                         if ( $needSpecialQuery ) {
 150                                 $subquery1 = $dbw->selectSQLText(
 151                                         [ 'revision', 'revision_actor_temp' ],
 152                                         [ 'COUNT(*)' ],
 153                                         [
 154                                                 'user_id = rev_user',
 155                                                 'revactor_rev' => null,
 156                                         ],
 157                                         __METHOD__,
 158                                         [],
 159                                         [ 'revision_actor_temp' => [ 'LEFT JOIN', 'revactor_rev = rev_id' ] ]
 160                                 );
 161                                 $subquery2 = $dbw->selectSQLText(
 162                                         [ 'revision' ] + $actorQuery['tables'],
 163                                         [ 'COUNT(*)' ],
 164                                         'user_id = actor_user',
 165                                         __METHOD__,
 166                                         [],
 167                                         $actorQuery['joins']
 168                                 );
 169                                 $dbw->query(
 170                                         "UPDATE $user SET user_editcount=($subquery1) + ($subquery2)",
 171                                         __METHOD__
 172                                 );
 173                         } else {
 174                                 $subquery = $dbw->selectSQLText(
 175                                         [ 'revision' ] + $actorQuery['tables'],
 176                                         [ 'COUNT(*)' ],
 177                                         [ 'user_id = ' . $actorQuery['fields']['rev_user'] ],
 178                                         __METHOD__,
 179                                         [],
 180                                         $actorQuery['joins']
 181                                 );
 182                                 $dbw->query( "UPDATE $user SET user_editcount=($subquery)", __METHOD__ );
 183                         }
 184                 }
 185
 186                 $this->output( "Done!\n" );
 187         }
 188 }
 189
 190 $maintClass = InitEditCount::class;
 191 require_once RUN_MAINTENANCE_IF_MAIN;