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->mDescription = "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                         return false;
  54                 }
  55
  56                 $this->output( "Populating rev_sha1 column\n" );
  57                 $rc = $this->doSha1Updates( 'revision', 'rev_id', 'rev' );
  58
  59                 $this->output( "Populating ar_sha1 column\n" );
  60                 $ac = $this->doSha1Updates( 'archive', 'ar_rev_id', 'ar' );
  61                 $this->output( "Populating ar_sha1 column legacy rows\n" );
  62                 $ac += $this->doSha1LegacyUpdates();
  63
  64                 $this->output( "rev_sha1 and ar_sha1 population complete "
  65                         . "[$rc revision rows, $ac archive rows].\n" );
  66                 return true;
  67         }
  68
  69         /**
  70          * @param string $table
  71          * @param string $idCol
  72          * @param string $prefix
  73          * @return int Rows changed
  74          */
  75         protected function doSha1Updates( $table, $idCol, $prefix ) {
  76                 $db = $this->getDB( DB_MASTER );
  77                 $start = $db->selectField( $table, "MIN($idCol)", false, __METHOD__ );
  78                 $end = $db->selectField( $table, "MAX($idCol)", false, __METHOD__ );
  79                 if ( !$start || !$end ) {
  80                         $this->output( "...$table table seems to be empty.\n" );
  81                         return 0;
  82                 }
  83
  84                 $count = 0;
  85                 # Do remaining chunk
  86                 $end += $this->mBatchSize - 1;
  87                 $blockStart = $start;
  88                 $blockEnd = $start + $this->mBatchSize - 1;
  89                 while ( $blockEnd <= $end ) {
  90                         $this->output( "...doing $idCol from $blockStart to $blockEnd\n" );
  91                         $cond = "$idCol BETWEEN $blockStart AND $blockEnd
  92                                 AND $idCol IS NOT NULL AND {$prefix}_sha1 = ''";
  93                         $res = $db->select( $table, '*', $cond, __METHOD__ );
  94
  95                         $db->begin( __METHOD__ );
  96                         foreach ( $res as $row ) {
  97                                 if ( $this->upgradeRow( $row, $table, $idCol, $prefix ) ) {
  98                                         $count++;
  99                                 }
 100                         }
 101                         $db->commit( __METHOD__ );
 102
 103                         $blockStart += $this->mBatchSize;
 104                         $blockEnd += $this->mBatchSize;
 105                         wfWaitForSlaves();
 106                 }
 107                 return $count;
 108         }
 109
 110         /**
 111          * @return int
 112          */
 113         protected function doSha1LegacyUpdates() {
 114                 $count = 0;
 115                 $db = $this->getDB( DB_MASTER );
 116                 $res = $db->select( 'archive', '*',
 117                         array( 'ar_rev_id IS NULL', 'ar_sha1' => '' ), __METHOD__ );
 118
 119                 $updateSize = 0;
 120                 $db->begin( __METHOD__ );
 121                 foreach ( $res as $row ) {
 122                         if ( $this->upgradeLegacyArchiveRow( $row ) ) {
 123                                 ++$count;
 124                         }
 125                         if ( ++$updateSize >= 100 ) {
 126                                 $updateSize = 0;
 127                                 $db->commit( __METHOD__ );
 128                                 $this->output( "Commited row with ar_timestamp={$row->ar_timestamp}\n" );
 129                                 wfWaitForSlaves();
 130                                 $db->begin( __METHOD__ );
 131                         }
 132                 }
 133                 $db->commit( __METHOD__ );
 134                 return $count;
 135         }
 136
 137         /**
 138          * @param stdClass $row
 139          * @param string $table
 140          * @param string $idCol
 141          * @param string $prefix
 142          * @return bool
 143          */
 144         protected function upgradeRow( $row, $table, $idCol, $prefix ) {
 145                 $db = $this->getDB( DB_MASTER );
 146                 try {
 147                         $rev = ( $table === 'archive' )
 148                                 ? Revision::newFromArchiveRow( $row )
 149                                 : new Revision( $row );
 150                         $text = $rev->getSerializedData();
 151                 } catch ( MWException $e ) {
 152                         $this->output( "Data of revision with {$idCol}={$row->$idCol} unavailable!\n" );
 153                         return false; // bug 22624?
 154                 }
 155                 if ( !is_string( $text ) ) {
 156                         # This should not happen, but sometimes does (bug 20757)
 157                         $this->output( "Data of revision with {$idCol}={$row->$idCol} unavailable!\n" );
 158                         return false;
 159                 } else {
 160                         $db->update( $table,
 161                                 array( "{$prefix}_sha1" => Revision::base36Sha1( $text ) ),
 162                                 array( $idCol => $row->$idCol ),
 163                                 __METHOD__
 164                         );
 165                         return true;
 166                 }
 167         }
 168
 169         /**
 170          * @param stdClass $row
 171          * @return bool
 172          */
 173         protected function upgradeLegacyArchiveRow( $row ) {
 174                 $db = $this->getDB( DB_MASTER );
 175                 try {
 176                         $rev = Revision::newFromArchiveRow( $row );
 177                 } catch ( MWException $e ) {
 178                         $this->output( "Text of revision with timestamp {$row->ar_timestamp} unavailable!\n" );
 179                         return false; // bug 22624?
 180                 }
 181                 $text = $rev->getSerializedData();
 182                 if ( !is_string( $text ) ) {
 183                         # This should not happen, but sometimes does (bug 20757)
 184                         $this->output( "Data of revision with timestamp {$row->ar_timestamp} unavailable!\n" );
 185                         return false;
 186                 } else {
 187                         # Archive table as no PK, but (NS,title,time) should be near unique.
 188                         # Any duplicates on those should also have duplicated text anyway.
 189                         $db->update( 'archive',
 190                                 array( 'ar_sha1' => Revision::base36Sha1( $text ) ),
 191                                 array(
 192                                         'ar_namespace' => $row->ar_namespace,
 193                                         'ar_title' => $row->ar_title,
 194                                         'ar_timestamp' => $row->ar_timestamp,
 195                                         'ar_len' => $row->ar_len // extra sanity
 196                                 ),
 197                                 __METHOD__
 198                         );
 199                         return true;
 200                 }
 201         }
 202 }
 203
 204 $maintClass = "PopulateRevisionSha1";
 205 require_once RUN_MAINTENANCE_IF_MAIN;