maintenance/populateImageSha1.php

   1 <?php
   2 /**
   3  * Optional upgrade script to populate the img_sha1 field
   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  * @ingroup Maintenance
  21  */
  22
  23 require_once( dirname( __FILE__ ) . '/Maintenance.php' );
  24
  25 class PopulateImageSha1 extends LoggedUpdateMaintenance {
  26         public function __construct() {
  27                 parent::__construct();
  28                 $this->mDescription = "Populate the img_sha1 field";
  29                 $this->addOption( 'force', "Recalculate sha1 for rows that already have a value" );
  30                 $this->addOption( 'method', "Use 'pipe' to pipe to mysql command line,\n" .
  31                         "\t\tdefault uses Database class", false, true );
  32                 $this->addOption( 'file', 'Fix for a specific file, without File: namespace prefixed', false, true );
  33         }
  34
  35         protected function getUpdateKey() {
  36                 return 'populate img_sha1';
  37         }
  38
  39         protected function updateSkippedMessage() {
  40                 return 'img_sha1 column of image table already populated.';
  41         }
  42
  43         public function execute() {
  44                 if ( $this->getOption( 'file' ) ) {
  45                         $this->doDBUpdates(); // skip update log checks/saves
  46                 } else {
  47                         parent::execute();
  48                 }
  49         }
  50
  51         public function doDBUpdates() {
  52                 $method = $this->getOption( 'method', 'normal' );
  53                 $file = $this->getOption( 'file', '' );
  54                 $force = $this->getOption( 'force' );
  55                 $isRegen = ( $force || $file != '' ); // forced recalculation?
  56
  57                 $t = -microtime( true );
  58                 $dbw = wfGetDB( DB_MASTER );
  59                 if ( $file != '' ) {
  60                         $res = $dbw->select(
  61                                 'image',
  62                                 array( 'img_name' ),
  63                                 array( 'img_name' => $file ),
  64                                 __METHOD__
  65                         );
  66                         if ( !$res ) {
  67                                 $this->error( "No such file: $file", true );
  68                                 return false;
  69                         }
  70                         $this->output( "Populating img_sha1 field for specified files\n" );
  71                 } else {
  72                         if ( $force ) {
  73                                 $conds = array();
  74                                 $this->output( "Populating and recalculating img_sha1 field\n" );
  75                         } else {
  76                                 $conds = array( 'img_sha1' => '' );
  77                                 $this->output( "Populating img_sha1 field\n" );
  78                         }
  79                         $res = $dbw->select( 'image', array( 'img_name' ), $conds, __METHOD__ );
  80                 }
  81
  82                 $imageTable = $dbw->tableName( 'image' );
  83                 $oldImageTable = $dbw->tableName( 'oldimage' );
  84
  85                 if ( $method == 'pipe' ) {
  86                         // Opening a pipe allows the SHA-1 operation to be done in parallel
  87                         // with the database write operation, because the writes are queued
  88                         // in the pipe buffer. This can improve performance by up to a
  89                         // factor of 2.
  90                         global $wgDBuser, $wgDBserver, $wgDBpassword, $wgDBname;
  91                         $cmd = 'mysql -u' . wfEscapeShellArg( $wgDBuser ) .
  92                                 ' -h' . wfEscapeShellArg( $wgDBserver ) .
  93                                 ' -p' . wfEscapeShellArg( $wgDBpassword, $wgDBname );
  94                         $this->output( "Using pipe method\n" );
  95                         $pipe = popen( $cmd, 'w' );
  96                 }
  97
  98                 $numRows = $res->numRows();
  99                 $i = 0;
 100                 foreach ( $res as $row ) {
 101                         if ( $i % $this->mBatchSize == 0 ) {
 102                                 $this->output( sprintf(
 103                                         "Done %d of %d, %5.3f%%  \r", $i, $numRows, $i / $numRows * 100 ) );
 104                                 wfWaitForSlaves();
 105                         }
 106                         $file = wfLocalFile( $row->img_name );
 107                         if ( !$file ) {
 108                                 continue;
 109                         }
 110                         // Upgrade the current file version...
 111                         $sha1 = $file->getRepo()->getFileSha1( $file->getPath() );
 112                         if ( strval( $sha1 ) !== '' ) { // file on disk and hashed properly
 113                                 if ( $isRegen && $file->getSha1() !== $sha1 ) {
 114                                         // The population was probably done already. If the old SHA1
 115                                         // does not match, then both fix the SHA1 and the metadata.
 116                                         $file->upgradeRow();
 117                                 } else {
 118                                         $sql = "UPDATE $imageTable SET img_sha1=" . $dbw->addQuotes( $sha1 ) .
 119                                                 " WHERE img_name=" . $dbw->addQuotes( $file->getName() );
 120                                         if ( $method == 'pipe' ) {
 121                                                 fwrite( $pipe, "$sql;\n" );
 122                                         } else {
 123                                                 $dbw->query( $sql, __METHOD__ );
 124                                         }
 125                                 }
 126                         }
 127                         // Upgrade the old file versions...
 128                         foreach ( $file->getHistory() as $oldFile ) {
 129                                 $sha1 = $oldFile->getRepo()->getFileSha1( $oldFile->getPath() );
 130                                 if ( strval( $sha1 ) !== '' ) { // file on disk and hashed properly
 131                                         if ( $isRegen && $oldFile->getSha1() !== $sha1 ) {
 132                                                 // The population was probably done already. If the old SHA1
 133                                                 // does not match, then both fix the SHA1 and the metadata.
 134                                                 $oldFile->upgradeRow();
 135                                         } else {
 136                                                 $sql = "UPDATE $oldImageTable SET oi_sha1=" . $dbw->addQuotes( $sha1 ) .
 137                                                         " WHERE (oi_name=" . $dbw->addQuotes( $oldFile->getName() ) . " AND" .
 138                                                         " oi_archive_name=" . $dbw->addQuotes( $oldFile->getArchiveName() ) . ")";
 139                                                 if ( $method == 'pipe' ) {
 140                                                         fwrite( $pipe, "$sql;\n" );
 141                                                 } else {
 142                                                         $dbw->query( $sql, __METHOD__ );
 143                                                 }
 144                                         }
 145                                 }
 146                         }
 147                         $i++;
 148                 }
 149                 if ( $method == 'pipe' ) {
 150                         fflush( $pipe );
 151                         pclose( $pipe );
 152                 }
 153                 $t += microtime( true );
 154                 $this->output( sprintf( "\nDone %d files in %.1f seconds\n", $numRows, $t ) );
 155
 156                 return !$file; // we only updated *some* files, don't log
 157         }
 158 }
 159
 160 $maintClass = "PopulateImageSha1";
 161 require_once( RUN_MAINTENANCE_IF_MAIN );