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  * @ingroup Maintenance
  22  */
  23
  24 require_once( dirname( __FILE__ ) . '/Maintenance.php' );
  25
  26 class PopulateRevisionSha1 extends LoggedUpdateMaintenance {
  27         public function __construct() {
  28                 parent::__construct();
  29                 $this->mDescription = "Populates the rev_sha1 and ar_sha1 fields";
  30                 $this->setBatchSize( 200 );
  31         }
  32
  33         protected function getUpdateKey() {
  34                 return 'populate rev_sha1';
  35         }
  36
  37         protected function doDBUpdates() {
  38                 $db = $this->getDB( DB_MASTER );
  39
  40                 if ( !$db->tableExists( 'revision' ) ) {
  41                         $this->error( "revision table does not exist", true );
  42                 } elseif ( !$db->tableExists( 'archive' ) ) {
  43                         $this->error( "archive table does not exist", true );
  44                 }
  45
  46                 $this->output( "Populating rev_sha1 column\n" );
  47                 $rc = $this->doSha1Updates( 'revision', 'rev_id', 'rev' );
  48
  49                 $this->output( "Populating ar_sha1 column\n" );
  50                 $ac = $this->doSha1Updates( 'archive', 'ar_rev_id', 'ar' );
  51                 $this->output( "Populating ar_sha1 column legacy rows\n" );
  52                 $ac += $this->doSha1LegacyUpdates();
  53
  54                 $this->output( "rev_sha1 and ar_sha1 population complete [$rc revision rows, $ac archive rows].\n" );
  55                 return true;
  56         }
  57
  58         /**
  59          * @param $table string
  60          * @param $idCol
  61          * @param $prefix string
  62          * @return Integer Rows changed
  63          */
  64         protected function doSha1Updates( $table, $idCol, $prefix ) {
  65                 $db = $this->getDB( DB_MASTER );
  66                 $start = $db->selectField( $table, "MIN($idCol)", false, __METHOD__ );
  67                 $end = $db->selectField( $table, "MAX($idCol)", false, __METHOD__ );
  68                 if ( !$start || !$end ) {
  69                         $this->output( "...$table table seems to be empty.\n" );
  70                         return 0;
  71                 }
  72
  73                 $count = 0;
  74                 # Do remaining chunk
  75                 $end += $this->mBatchSize - 1;
  76                 $blockStart = $start;
  77                 $blockEnd = $start + $this->mBatchSize - 1;
  78                 while ( $blockEnd <= $end ) {
  79                         $this->output( "...doing $idCol from $blockStart to $blockEnd\n" );
  80                         $cond = "$idCol BETWEEN $blockStart AND $blockEnd
  81                                 AND $idCol IS NOT NULL AND {$prefix}_sha1 = ''";
  82                         $res = $db->select( $table, '*', $cond, __METHOD__ );
  83
  84                         $db->begin( __METHOD__ );
  85                         foreach ( $res as $row ) {
  86                                 if ( $this->upgradeRow( $row, $table, $idCol, $prefix ) ) {
  87                                         $count++;
  88                                 }
  89                         }
  90                         $db->commit( __METHOD__ );
  91
  92                         $blockStart += $this->mBatchSize;
  93                         $blockEnd += $this->mBatchSize;
  94                         wfWaitForSlaves();
  95                 }
  96                 return $count;
  97         }
  98
  99         /**
 100          * @return int
 101          */
 102         protected function doSha1LegacyUpdates() {
 103                 $count = 0;
 104                 $db = $this->getDB( DB_MASTER );
 105                 $res = $db->select( 'archive', '*',
 106                         array( 'ar_rev_id IS NULL', 'ar_sha1' => '' ), __METHOD__ );
 107
 108                 $updateSize = 0;
 109                 $db->begin( __METHOD__ );
 110                 foreach ( $res as $row ) {
 111                         if ( $this->upgradeLegacyArchiveRow( $row ) ) {
 112                                 ++$count;
 113                         }
 114                         if ( ++$updateSize >= 100 ) {
 115                                 $updateSize = 0;
 116                                 $db->commit( __METHOD__ );
 117                                 $this->output( "Commited row with ar_timestamp={$row->ar_timestamp}\n" );
 118                                 wfWaitForSlaves();
 119                                 $db->begin( __METHOD__ );
 120                         }
 121                 }
 122                 $db->commit( __METHOD__ );
 123                 return $count;
 124         }
 125
 126         /**
 127          * @param $row
 128          * @param $table
 129          * @param $idCol
 130          * @param $prefix
 131          * @return bool
 132          */
 133         protected function upgradeRow( $row, $table, $idCol, $prefix ) {
 134                 $db = $this->getDB( DB_MASTER );
 135                 try {
 136                         $rev = ( $table === 'archive' )
 137                                 ? Revision::newFromArchiveRow( $row )
 138                                 : new Revision( $row );
 139                         $text = $rev->getSerializedData();
 140                 } catch ( MWException $e ) {
 141                         $this->output( "Data of revision with {$idCol}={$row->$idCol} unavailable!\n" );
 142                         return false; // bug 22624?
 143                 }
 144                 if ( !is_string( $text ) ) {
 145                         # This should not happen, but sometimes does (bug 20757)
 146                         $this->output( "Data of revision with {$idCol}={$row->$idCol} unavailable!\n" );
 147                         return false;
 148                 } else {
 149                         $db->update( $table,
 150                                 array( "{$prefix}_sha1" => Revision::base36Sha1( $text ) ),
 151                                 array( $idCol => $row->$idCol ),
 152                                 __METHOD__
 153                         );
 154                         return true;
 155                 }
 156         }
 157
 158         /**
 159          * @param $row
 160          * @return bool
 161          */
 162         protected function upgradeLegacyArchiveRow( $row ) {
 163                 $db = $this->getDB( DB_MASTER );
 164                 try {
 165                         $rev = Revision::newFromArchiveRow( $row );
 166                 } catch ( MWException $e ) {
 167                         $this->output( "Text of revision with timestamp {$row->ar_timestamp} unavailable!\n" );
 168                         return false; // bug 22624?
 169                 }
 170                 $text = $rev->getSerializedData();
 171                 if ( !is_string( $text ) ) {
 172                         # This should not happen, but sometimes does (bug 20757)
 173                         $this->output( "Data of revision with timestamp {$row->ar_timestamp} unavailable!\n" );
 174                         return false;
 175                 } else {
 176                         # Archive table as no PK, but (NS,title,time) should be near unique.
 177                         # Any duplicates on those should also have duplicated text anyway.
 178                         $db->update( 'archive',
 179                                 array( 'ar_sha1' => Revision::base36Sha1( $text ) ),
 180                                 array(
 181                                         'ar_namespace' => $row->ar_namespace,
 182                                         'ar_title'     => $row->ar_title,
 183                                         'ar_timestamp' => $row->ar_timestamp,
 184                                         'ar_len'       => $row->ar_len // extra sanity
 185                                 ),
 186                                 __METHOD__
 187                         );
 188                         return true;
 189                 }
 190         }
 191 }
 192
 193 $maintClass = "PopulateRevisionSha1";
 194 require_once( RUN_MAINTENANCE_IF_MAIN );