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 use Wikimedia\Rdbms\IDatabase;
  25
  26 require_once __DIR__ . '/Maintenance.php';
  27
  28 /**
  29  * Maintenance script that populates the rev_len and ar_len fields when they are NULL.
  30  * This is the case for all revisions created before MW 1.10, as well as those affected
  31  * by T18748 (MW 1.10-1.13) and those affected by T135414 (MW 1.21-1.24).
  32  *
  33  * @ingroup Maintenance
  34  */
  35 class PopulateRevisionLength extends LoggedUpdateMaintenance {
  36         public function __construct() {
  37                 parent::__construct();
  38                 $this->addDescription( 'Populates the rev_len and ar_len fields' );
  39                 $this->setBatchSize( 200 );
  40         }
  41
  42         protected function getUpdateKey() {
  43                 return 'populate rev_len and ar_len';
  44         }
  45
  46         public function doDBUpdates() {
  47                 $dbw = $this->getDB( DB_MASTER );
  48                 if ( !$dbw->tableExists( 'revision' ) ) {
  49                         $this->fatalError( "revision table does not exist" );
  50                 } elseif ( !$dbw->tableExists( 'archive' ) ) {
  51                         $this->fatalError( "archive table does not exist" );
  52                 } elseif ( !$dbw->fieldExists( 'revision', 'rev_len', __METHOD__ ) ) {
  53                         $this->output( "rev_len column does not exist\n\n", true );
  54
  55                         return false;
  56                 }
  57
  58                 $this->output( "Populating rev_len column\n" );
  59                 $rev = $this->doLenUpdates( 'revision', 'rev_id', 'rev', Revision::getQueryInfo() );
  60
  61                 $this->output( "Populating ar_len column\n" );
  62                 $ar = $this->doLenUpdates( 'archive', 'ar_id', 'ar', Revision::getArchiveQueryInfo() );
  63
  64                 $this->output( "rev_len and ar_len population complete "
  65                         . "[$rev revision rows, $ar archive rows].\n" );
  66
  67                 return true;
  68         }
  69
  70         /**
  71          * @param string $table
  72          * @param string $idCol
  73          * @param string $prefix
  74          * @param array $queryInfo
  75          * @return int
  76          */
  77         protected function doLenUpdates( $table, $idCol, $prefix, $queryInfo ) {
  78                 $dbr = $this->getDB( DB_REPLICA );
  79                 $dbw = $this->getDB( DB_MASTER );
  80                 $batchSize = $this->getBatchSize();
  81                 $start = $dbw->selectField( $table, "MIN($idCol)", '', __METHOD__ );
  82                 $end = $dbw->selectField( $table, "MAX($idCol)", '', __METHOD__ );
  83                 if ( !$start || !$end ) {
  84                         $this->output( "...$table table seems to be empty.\n" );
  85
  86                         return 0;
  87                 }
  88
  89                 # Do remaining chunks
  90                 $blockStart = intval( $start );
  91                 $blockEnd = intval( $start ) + $batchSize - 1;
  92                 $count = 0;
  93
  94                 while ( $blockStart <= $end ) {
  95                         $this->output( "...doing $idCol from $blockStart to $blockEnd\n" );
  96                         $res = $dbr->select(
  97                                 $queryInfo['tables'],
  98                                 $queryInfo['fields'],
  99                                 [
 100                                         "$idCol >= $blockStart",
 101                                         "$idCol <= $blockEnd",
 102                                         $dbr->makeList( [
 103                                                 "{$prefix}_len IS NULL",
 104                                                 $dbr->makeList( [
 105                                                         "{$prefix}_len = 0",
 106                                                         "{$prefix}_sha1 != " . $dbr->addQuotes( 'phoiac9h4m842xq45sp7s6u21eteeq1' ), // sha1( "" )
 107                                                 ], IDatabase::LIST_AND )
 108                                         ], IDatabase::LIST_OR )
 109                                 ],
 110                                 __METHOD__,
 111                                 [],
 112                                 $queryInfo['joins']
 113                         );
 114
 115                         if ( $res->numRows() > 0 ) {
 116                                 $this->beginTransaction( $dbw, __METHOD__ );
 117                                 # Go through and update rev_len from these rows.
 118                                 foreach ( $res as $row ) {
 119                                         if ( $this->upgradeRow( $row, $table, $idCol, $prefix ) ) {
 120                                                 $count++;
 121                                         }
 122                                 }
 123                                 $this->commitTransaction( $dbw, __METHOD__ );
 124                         }
 125
 126                         $blockStart += $batchSize;
 127                         $blockEnd += $batchSize;
 128                 }
 129
 130                 return $count;
 131         }
 132
 133         /**
 134          * @param stdClass $row
 135          * @param string $table
 136          * @param string $idCol
 137          * @param string $prefix
 138          * @return bool
 139          */
 140         protected function upgradeRow( $row, $table, $idCol, $prefix ) {
 141                 $dbw = $this->getDB( DB_MASTER );
 142
 143                 $rev = ( $table === 'archive' )
 144                         ? Revision::newFromArchiveRow( $row )
 145                         : new Revision( $row );
 146
 147                 $content = $rev->getContent( Revision::RAW );
 148                 if ( !$content ) {
 149                         # This should not happen, but sometimes does (T22757)
 150                         $id = $row->$idCol;
 151                         $this->output( "Content of $table $id unavailable!\n" );
 152
 153                         return false;
 154                 }
 155
 156                 # Update the row...
 157                 $dbw->update( $table,
 158                         [ "{$prefix}_len" => $content->getSize() ],
 159                         [ $idCol => $row->$idCol ],
 160                         __METHOD__
 161                 );
 162
 163                 return true;
 164         }
 165 }
 166
 167 $maintClass = PopulateRevisionLength::class;
 168 require_once RUN_MAINTENANCE_IF_MAIN;