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                 }
  52
  53                 $this->output( "Populating rev_sha1 column\n" );
  54                 $rc = $this->doSha1Updates( 'revision', 'rev_id', 'rev' );
  55
  56                 $this->output( "Populating ar_sha1 column\n" );
  57                 $ac = $this->doSha1Updates( 'archive', 'ar_rev_id', 'ar' );
  58                 $this->output( "Populating ar_sha1 column legacy rows\n" );
  59                 $ac += $this->doSha1LegacyUpdates();
  60
  61                 $this->output( "rev_sha1 and ar_sha1 population complete [$rc revision rows, $ac archive rows].\n" );
  62                 return true;
  63         }
  64
  65         /**
  66          * @param $table string
  67          * @param $idCol
  68          * @param $prefix string
  69          * @return Integer Rows changed
  70          */
  71         protected function doSha1Updates( $table, $idCol, $prefix ) {
  72                 $db = $this->getDB( DB_MASTER );
  73                 $start = $db->selectField( $table, "MIN($idCol)", false, __METHOD__ );
  74                 $end = $db->selectField( $table, "MAX($idCol)", false, __METHOD__ );
  75                 if ( !$start || !$end ) {
  76                         $this->output( "...$table table seems to be empty.\n" );
  77                         return 0;
  78                 }
  79
  80                 $count = 0;
  81                 # Do remaining chunk
  82                 $end += $this->mBatchSize - 1;
  83                 $blockStart = $start;
  84                 $blockEnd = $start + $this->mBatchSize - 1;
  85                 while ( $blockEnd <= $end ) {
  86                         $this->output( "...doing $idCol from $blockStart to $blockEnd\n" );
  87                         $cond = "$idCol BETWEEN $blockStart AND $blockEnd
  88                                 AND $idCol IS NOT NULL AND {$prefix}_sha1 = ''";
  89                         $res = $db->select( $table, '*', $cond, __METHOD__ );
  90
  91                         $db->begin( __METHOD__ );
  92                         foreach ( $res as $row ) {
  93                                 if ( $this->upgradeRow( $row, $table, $idCol, $prefix ) ) {
  94                                         $count++;
  95                                 }
  96                         }
  97                         $db->commit( __METHOD__ );
  98
  99                         $blockStart += $this->mBatchSize;
 100                         $blockEnd += $this->mBatchSize;
 101                         wfWaitForSlaves();
 102                 }
 103                 return $count;
 104         }
 105
 106         /**
 107          * @return int
 108          */
 109         protected function doSha1LegacyUpdates() {
 110                 $count = 0;
 111                 $db = $this->getDB( DB_MASTER );
 112                 $res = $db->select( 'archive', '*',
 113                         array( 'ar_rev_id IS NULL', 'ar_sha1' => '' ), __METHOD__ );
 114
 115                 $updateSize = 0;
 116                 $db->begin( __METHOD__ );
 117                 foreach ( $res as $row ) {
 118                         if ( $this->upgradeLegacyArchiveRow( $row ) ) {
 119                                 ++$count;
 120                         }
 121                         if ( ++$updateSize >= 100 ) {
 122                                 $updateSize = 0;
 123                                 $db->commit( __METHOD__ );
 124                                 $this->output( "Commited row with ar_timestamp={$row->ar_timestamp}\n" );
 125                                 wfWaitForSlaves();
 126                                 $db->begin( __METHOD__ );
 127                         }
 128                 }
 129                 $db->commit( __METHOD__ );
 130                 return $count;
 131         }
 132
 133         /**
 134          * @param $row
 135          * @param $table
 136          * @param $idCol
 137          * @param $prefix
 138          * @return bool
 139          */
 140         protected function upgradeRow( $row, $table, $idCol, $prefix ) {
 141                 $db = $this->getDB( DB_MASTER );
 142                 try {
 143                         $rev = ( $table === 'archive' )
 144                                 ? Revision::newFromArchiveRow( $row )
 145                                 : new Revision( $row );
 146                         $text = $rev->getRawText();
 147                 } catch ( MWException $e ) {
 148                         $this->output( "Text of revision with {$idCol}={$row->$idCol} unavailable!\n" );
 149                         return false; // bug 22624?
 150                 }
 151                 if ( !is_string( $text ) ) {
 152                         # This should not happen, but sometimes does (bug 20757)
 153                         $this->output( "Text of revision with {$idCol}={$row->$idCol} unavailable!\n" );
 154                         return false;
 155                 } else {
 156                         $db->update( $table,
 157                                 array( "{$prefix}_sha1" => Revision::base36Sha1( $text ) ),
 158                                 array( $idCol => $row->$idCol ),
 159                                 __METHOD__
 160                         );
 161                         return true;
 162                 }
 163         }
 164
 165         /**
 166          * @param $row
 167          * @return bool
 168          */
 169         protected function upgradeLegacyArchiveRow( $row ) {
 170                 $db = $this->getDB( DB_MASTER );
 171                 try {
 172                         $rev = Revision::newFromArchiveRow( $row );
 173                 } catch ( MWException $e ) {
 174                         $this->output( "Text of revision with timestamp {$row->ar_timestamp} unavailable!\n" );
 175                         return false; // bug 22624?
 176                 }
 177                 $text = $rev->getRawText();
 178                 if ( !is_string( $text ) ) {
 179                         # This should not happen, but sometimes does (bug 20757)
 180                         $this->output( "Text of revision with timestamp {$row->ar_timestamp} unavailable!\n" );
 181                         return false;
 182                 } else {
 183                         # Archive table as no PK, but (NS,title,time) should be near unique.
 184                         # Any duplicates on those should also have duplicated text anyway.
 185                         $db->update( 'archive',
 186                                 array( 'ar_sha1' => Revision::base36Sha1( $text ) ),
 187                                 array(
 188                                         'ar_namespace' => $row->ar_namespace,
 189                                         'ar_title'     => $row->ar_title,
 190                                         'ar_timestamp' => $row->ar_timestamp,
 191                                         'ar_len'       => $row->ar_len // extra sanity
 192                                 ),
 193                                 __METHOD__
 194                         );
 195                         return true;
 196                 }
 197         }
 198 }
 199
 200 $maintClass = "PopulateRevisionSha1";
 201 require_once( RUN_MAINTENANCE_IF_MAIN );