maintenance/populateRevisionSha1.php

   1 <?php
   2 /**
   3  * Fills the rev_sha1 and ar_sha1 columns of revision
   4  * and archive tables for revisions created before MW 1.19.
   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 /**
  28  * Maintenance script that fills the rev_sha1 and ar_sha1 columns of revision
  29  * and archive tables for revisions created before MW 1.19.
  30  *
  31  * @ingroup Maintenance
  32  */
  33 class PopulateRevisionSha1 extends LoggedUpdateMaintenance {
  34         public function __construct() {
  35                 parent::__construct();
  36                 $this->addDescription( 'Populates the rev_sha1 and ar_sha1 fields' );
  37                 $this->setBatchSize( 200 );
  38         }
  39
  40         protected function getUpdateKey() {
  41                 return 'populate rev_sha1';
  42         }
  43
  44         protected function doDBUpdates() {
  45                 $db = $this->getDB( DB_MASTER );
  46
  47                 if ( !$db->tableExists( 'revision' ) ) {
  48                         $this->error( "revision table does not exist", true );
  49                 } elseif ( !$db->tableExists( 'archive' ) ) {
  50                         $this->error( "archive table does not exist", true );
  51                 } elseif ( !$db->fieldExists( 'revision', 'rev_sha1', __METHOD__ ) ) {
  52                         $this->output( "rev_sha1 column does not exist\n\n", true );
  53
  54                         return false;
  55                 }
  56
  57                 $this->output( "Populating rev_sha1 column\n" );
  58                 $rc = $this->doSha1Updates( 'revision', 'rev_id', 'rev' );
  59
  60                 $this->output( "Populating ar_sha1 column\n" );
  61                 $ac = $this->doSha1Updates( 'archive', 'ar_rev_id', 'ar' );
  62                 $this->output( "Populating ar_sha1 column legacy rows\n" );
  63                 $ac += $this->doSha1LegacyUpdates();
  64
  65                 $this->output( "rev_sha1 and ar_sha1 population complete "
  66                         . "[$rc revision rows, $ac archive rows].\n" );
  67
  68                 return true;
  69         }
  70
  71         /**
  72          * @param string $table
  73          * @param string $idCol
  74          * @param string $prefix
  75          * @return int Rows changed
  76          */
  77         protected function doSha1Updates( $table, $idCol, $prefix ) {
  78                 $db = $this->getDB( DB_MASTER );
  79                 $start = $db->selectField( $table, "MIN($idCol)", false, __METHOD__ );
  80                 $end = $db->selectField( $table, "MAX($idCol)", false, __METHOD__ );
  81                 if ( !$start || !$end ) {
  82                         $this->output( "...$table table seems to be empty.\n" );
  83
  84                         return 0;
  85                 }
  86
  87                 $count = 0;
  88                 # Do remaining chunk
  89                 $end += $this->mBatchSize - 1;
  90                 $blockStart = $start;
  91                 $blockEnd = $start + $this->mBatchSize - 1;
  92                 while ( $blockEnd <= $end ) {
  93                         $this->output( "...doing $idCol from $blockStart to $blockEnd\n" );
  94                         $cond = "$idCol BETWEEN $blockStart AND $blockEnd
  95                                 AND $idCol IS NOT NULL AND {$prefix}_sha1 = ''";
  96                         $res = $db->select( $table, '*', $cond, __METHOD__ );
  97
  98                         $this->beginTransaction( $db, __METHOD__ );
  99                         foreach ( $res as $row ) {
 100                                 if ( $this->upgradeRow( $row, $table, $idCol, $prefix ) ) {
 101                                         $count++;
 102                                 }
 103                         }
 104                         $this->commitTransaction( $db, __METHOD__ );
 105
 106                         $blockStart += $this->mBatchSize;
 107                         $blockEnd += $this->mBatchSize;
 108                         wfWaitForSlaves();
 109                 }
 110
 111                 return $count;
 112         }
 113
 114         /**
 115          * @return int
 116          */
 117         protected function doSha1LegacyUpdates() {
 118                 $count = 0;
 119                 $db = $this->getDB( DB_MASTER );
 120                 $res = $db->select( 'archive', '*',
 121                         [ 'ar_rev_id IS NULL', 'ar_sha1' => '' ], __METHOD__ );
 122
 123                 $updateSize = 0;
 124                 $this->beginTransaction( $db, __METHOD__ );
 125                 foreach ( $res as $row ) {
 126                         if ( $this->upgradeLegacyArchiveRow( $row ) ) {
 127                                 ++$count;
 128                         }
 129                         if ( ++$updateSize >= 100 ) {
 130                                 $updateSize = 0;
 131                                 $this->commitTransaction( $db, __METHOD__ );
 132                                 $this->output( "Commited row with ar_timestamp={$row->ar_timestamp}\n" );
 133                                 wfWaitForSlaves();
 134                                 $this->beginTransaction( $db, __METHOD__ );
 135                         }
 136                 }
 137                 $this->commitTransaction( $db, __METHOD__ );
 138
 139                 return $count;
 140         }
 141
 142         /**
 143          * @param stdClass $row
 144          * @param string $table
 145          * @param string $idCol
 146          * @param string $prefix
 147          * @return bool
 148          */
 149         protected function upgradeRow( $row, $table, $idCol, $prefix ) {
 150                 $db = $this->getDB( DB_MASTER );
 151                 try {
 152                         $rev = ( $table === 'archive' )
 153                                 ? Revision::newFromArchiveRow( $row )
 154                                 : new Revision( $row );
 155                         $text = $rev->getSerializedData();
 156                 } catch ( Exception $e ) {
 157                         $this->output( "Data of revision with {$idCol}={$row->$idCol} unavailable!\n" );
 158
 159                         return false; // T24624?
 160                 }
 161                 if ( !is_string( $text ) ) {
 162                         # This should not happen, but sometimes does (T22757)
 163                         $this->output( "Data of revision with {$idCol}={$row->$idCol} unavailable!\n" );
 164
 165                         return false;
 166                 } else {
 167                         $db->update( $table,
 168                                 [ "{$prefix}_sha1" => Revision::base36Sha1( $text ) ],
 169                                 [ $idCol => $row->$idCol ],
 170                                 __METHOD__
 171                         );
 172
 173                         return true;
 174                 }
 175         }
 176
 177         /**
 178          * @param stdClass $row
 179          * @return bool
 180          */
 181         protected function upgradeLegacyArchiveRow( $row ) {
 182                 $db = $this->getDB( DB_MASTER );
 183                 try {
 184                         $rev = Revision::newFromArchiveRow( $row );
 185                 } catch ( Exception $e ) {
 186                         $this->output( "Text of revision with timestamp {$row->ar_timestamp} unavailable!\n" );
 187
 188                         return false; // T24624?
 189                 }
 190                 $text = $rev->getSerializedData();
 191                 if ( !is_string( $text ) ) {
 192                         # This should not happen, but sometimes does (T22757)
 193                         $this->output( "Data of revision with timestamp {$row->ar_timestamp} unavailable!\n" );
 194
 195                         return false;
 196                 } else {
 197                         # Archive table as no PK, but (NS,title,time) should be near unique.
 198                         # Any duplicates on those should also have duplicated text anyway.
 199                         $db->update( 'archive',
 200                                 [ 'ar_sha1' => Revision::base36Sha1( $text ) ],
 201                                 [
 202                                         'ar_namespace' => $row->ar_namespace,
 203                                         'ar_title' => $row->ar_title,
 204                                         'ar_timestamp' => $row->ar_timestamp,
 205                                         'ar_len' => $row->ar_len // extra sanity
 206                                 ],
 207                                 __METHOD__
 208                         );
 209
 210                         return true;
 211                 }
 212         }
 213 }
 214
 215 $maintClass = "PopulateRevisionSha1";
 216 require_once RUN_MAINTENANCE_IF_MAIN;