maintenance/populateRevisionLength.php

   1 <?php
   2 /**
   3  * Populates the rev_len and ar_len fields when they are NULL.
   4  *
   5  * This program is free software; you can redistribute it and/or modify
   6  * it under the terms of the GNU General Public License as published by
   7  * the Free Software Foundation; either version 2 of the License, or
   8  * (at your option) any later version.
   9  *
  10  * This program is distributed in the hope that it will be useful,
  11  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  12  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  13  * GNU General Public License for more details.
  14  *
  15  * You should have received a copy of the GNU General Public License along
  16  * with this program; if not, write to the Free Software Foundation, Inc.,
  17  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  18  * http://www.gnu.org/copyleft/gpl.html
  19  *
  20  * @file
  21  * @ingroup Maintenance
  22  */
  23
  24 require_once __DIR__ . '/Maintenance.php';
  25
  26 /**
  27  * Maintenance script that populates the rev_len and ar_len fields when they are NULL.
  28  * This is the case for all revisions created before MW 1.10, as well as those affected
  29  * by T18748 (MW 1.10-1.13) and those affected by T135414 (MW 1.21-1.24).
  30  *
  31  * @ingroup Maintenance
  32  */
  33 class PopulateRevisionLength extends LoggedUpdateMaintenance {
  34         public function __construct() {
  35                 parent::__construct();
  36                 $this->addDescription( 'Populates the rev_len and ar_len fields' );
  37                 $this->setBatchSize( 200 );
  38         }
  39
  40         protected function getUpdateKey() {
  41                 return 'populate rev_len and ar_len';
  42         }
  43
  44         public function doDBUpdates() {
  45                 $dbw = $this->getDB( DB_MASTER );
  46                 if ( !$dbw->tableExists( 'revision' ) ) {
  47                         $this->fatalError( "revision table does not exist" );
  48                 } elseif ( !$dbw->tableExists( 'archive' ) ) {
  49                         $this->fatalError( "archive table does not exist" );
  50                 } elseif ( !$dbw->fieldExists( 'revision', 'rev_len', __METHOD__ ) ) {
  51                         $this->output( "rev_len column does not exist\n\n", true );
  52
  53                         return false;
  54                 }
  55
  56                 $this->output( "Populating rev_len column\n" );
  57                 $rev = $this->doLenUpdates( 'revision', 'rev_id', 'rev', Revision::getQueryInfo() );
  58
  59                 $this->output( "Populating ar_len column\n" );
  60                 $ar = $this->doLenUpdates( 'archive', 'ar_id', 'ar', Revision::getArchiveQueryInfo() );
  61
  62                 $this->output( "rev_len and ar_len population complete "
  63                         . "[$rev revision rows, $ar archive rows].\n" );
  64
  65                 return true;
  66         }
  67
  68         /**
  69          * @param string $table
  70          * @param string $idCol
  71          * @param string $prefix
  72          * @param array $queryInfo
  73          * @return int
  74          */
  75         protected function doLenUpdates( $table, $idCol, $prefix, $queryInfo ) {
  76                 $dbr = $this->getDB( DB_REPLICA );
  77                 $dbw = $this->getDB( DB_MASTER );
  78                 $batchSize = $this->getBatchSize();
  79                 $start = $dbw->selectField( $table, "MIN($idCol)", '', __METHOD__ );
  80                 $end = $dbw->selectField( $table, "MAX($idCol)", '', __METHOD__ );
  81                 if ( !$start || !$end ) {
  82                         $this->output( "...$table table seems to be empty.\n" );
  83
  84                         return 0;
  85                 }
  86
  87                 # Do remaining chunks
  88                 $blockStart = intval( $start );
  89                 $blockEnd = intval( $start ) + $batchSize - 1;
  90                 $count = 0;
  91
  92                 while ( $blockStart <= $end ) {
  93                         $this->output( "...doing $idCol from $blockStart to $blockEnd\n" );
  94                         $res = $dbr->select(
  95                                 $queryInfo['tables'],
  96                                 $queryInfo['fields'],
  97                                 [
  98                                         "$idCol >= $blockStart",
  99                                         "$idCol <= $blockEnd",
 100                                         "{$prefix}_len IS NULL"
 101                                 ],
 102                                 __METHOD__,
 103                                 [],
 104                                 $queryInfo['joins']
 105                         );
 106
 107                         if ( $res->numRows() > 0 ) {
 108                                 $this->beginTransaction( $dbw, __METHOD__ );
 109                                 # Go through and update rev_len from these rows.
 110                                 foreach ( $res as $row ) {
 111                                         if ( $this->upgradeRow( $row, $table, $idCol, $prefix ) ) {
 112                                                 $count++;
 113                                         }
 114                                 }
 115                                 $this->commitTransaction( $dbw, __METHOD__ );
 116                         }
 117
 118                         $blockStart += $batchSize;
 119                         $blockEnd += $batchSize;
 120                 }
 121
 122                 return $count;
 123         }
 124
 125         /**
 126          * @param stdClass $row
 127          * @param string $table
 128          * @param string $idCol
 129          * @param string $prefix
 130          * @return bool
 131          */
 132         protected function upgradeRow( $row, $table, $idCol, $prefix ) {
 133                 $dbw = $this->getDB( DB_MASTER );
 134
 135                 $rev = ( $table === 'archive' )
 136                         ? Revision::newFromArchiveRow( $row )
 137                         : new Revision( $row );
 138
 139                 $content = $rev->getContent();
 140                 if ( !$content ) {
 141                         # This should not happen, but sometimes does (T22757)
 142                         $id = $row->$idCol;
 143                         $this->output( "Content of $table $id unavailable!\n" );
 144
 145                         return false;
 146                 }
 147
 148                 # Update the row...
 149                 $dbw->update( $table,
 150                         [ "{$prefix}_len" => $content->getSize() ],
 151                         [ $idCol => $row->$idCol ],
 152                         __METHOD__
 153                 );
 154
 155                 return true;
 156         }
 157 }
 158
 159 $maintClass = PopulateRevisionLength::class;
 160 require_once RUN_MAINTENANCE_IF_MAIN;