maintenance/rebuildImages.php

   1 <?php
   2 /**
   3  * Script to update image metadata records
   4  *
   5  * Usage: php rebuildImages.php [--missing] [--dry-run]
   6  * Options:
   7  *   --missing  Crawl the uploads dir for images without records, and
   8  *              add them only.
   9  *
  10  * Copyright © 2005 Brion Vibber <brion@pobox.com>
  11  * http://www.mediawiki.org/
  12  *
  13  * This program is free software; you can redistribute it and/or modify
  14  * it under the terms of the GNU General Public License as published by
  15  * the Free Software Foundation; either version 2 of the License, or
  16  * (at your option) any later version.
  17  *
  18  * This program is distributed in the hope that it will be useful,
  19  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  20  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  21  * GNU General Public License for more details.
  22  *
  23  * You should have received a copy of the GNU General Public License along
  24  * with this program; if not, write to the Free Software Foundation, Inc.,
  25  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  26  * http://www.gnu.org/copyleft/gpl.html
  27  *
  28  * @file
  29  * @author Brion Vibber <brion at pobox.com>
  30  * @ingroup maintenance
  31  */
  32
  33 require_once( dirname( __FILE__ ) . '/Maintenance.php' );
  34
  35 class ImageBuilder extends Maintenance {
  36
  37         /**
  38          * @var DatabaseBase
  39          */
  40         protected $dbw;
  41
  42         function __construct() {
  43                 parent::__construct();
  44
  45                 global $wgUpdateCompatibleMetadata;
  46                 //make sure to update old, but compatible img_metadata fields.
  47                 $wgUpdateCompatibleMetadata = true;
  48
  49                 $this->mDescription = 'Script to update image metadata records';
  50
  51                 $this->addOption( 'missing', 'Check for files without associated database record' );
  52                 $this->addOption( 'dry-run', 'Only report, don\'t update the database' );
  53         }
  54
  55         public function execute() {
  56                 $this->dbw = wfGetDB( DB_MASTER );
  57                 $this->maxLag = 10; # if slaves are lagged more than 10 secs, wait
  58                 $this->dryrun = $this->hasOption( 'dry-run' );
  59                 if ( $this->dryrun ) {
  60                         $GLOBALS['wgReadOnly'] = 'Dry run mode, image upgrades are suppressed';
  61                 }
  62
  63                 if ( $this->hasOption( 'missing' ) ) {
  64                         $this->crawlMissing();
  65                 } else {
  66                         $this->build();
  67                 }
  68         }
  69
  70         function getRepo() {
  71                 if ( !isset( $this->repo ) ) {
  72                         $this->repo = RepoGroup::singleton()->getLocalRepo();
  73                 }
  74                 return $this->repo;
  75         }
  76
  77         function build() {
  78                 $this->buildImage();
  79                 $this->buildOldImage();
  80         }
  81
  82         function init( $count, $table ) {
  83                 $this->processed = 0;
  84                 $this->updated = 0;
  85                 $this->count = $count;
  86                 $this->startTime = wfTime();
  87                 $this->table = $table;
  88         }
  89
  90         function progress( $updated ) {
  91                 $this->updated += $updated;
  92                 $this->processed++;
  93                 if ( $this->processed % 100 != 0 ) {
  94                         return;
  95                 }
  96                 $portion = $this->processed / $this->count;
  97                 $updateRate = $this->updated / $this->processed;
  98
  99                 $now = wfTime();
 100                 $delta = $now - $this->startTime;
 101                 $estimatedTotalTime = $delta / $portion;
 102                 $eta = $this->startTime + $estimatedTotalTime;
 103                 $rate = $this->processed / $delta;
 104
 105                 $this->output( sprintf( "%s: %6.2f%% done on %s; ETA %s [%d/%d] %.2f/sec <%.2f%% updated>\n",
 106                         wfTimestamp( TS_DB, intval( $now ) ),
 107                         $portion * 100.0,
 108                         $this->table,
 109                         wfTimestamp( TS_DB, intval( $eta ) ),
 110                         $this->processed,
 111                         $this->count,
 112                         $rate,
 113                         $updateRate * 100.0 ) );
 114                 flush();
 115         }
 116
 117         function buildTable( $table, $key, $callback ) {
 118                 $count = $this->dbw->selectField( $table, 'count(*)', '', __METHOD__ );
 119                 $this->init( $count, $table );
 120                 $this->output( "Processing $table...\n" );
 121
 122                 $result = wfGetDB( DB_SLAVE )->select( $table, '*', array(), __METHOD__ );
 123
 124                 foreach ( $result as $row ) {
 125                         $update = call_user_func( $callback, $row, null );
 126                         if ( $update ) {
 127                                 $this->progress( 1 );
 128                         } else {
 129                                 $this->progress( 0 );
 130                         }
 131                 }
 132                 $this->output( "Finished $table... $this->updated of $this->processed rows updated\n" );
 133         }
 134
 135         function buildImage() {
 136                 $callback = array( $this, 'imageCallback' );
 137                 $this->buildTable( 'image', 'img_name', $callback );
 138         }
 139
 140         function imageCallback( $row, $copy ) {
 141                 // Create a File object from the row
 142                 // This will also upgrade it
 143                 $file = $this->getRepo()->newFileFromRow( $row );
 144                 return $file->getUpgraded();
 145         }
 146
 147         function buildOldImage() {
 148                 $this->buildTable( 'oldimage', 'oi_archive_name',
 149                         array( $this, 'oldimageCallback' ) );
 150         }
 151
 152         function oldimageCallback( $row, $copy ) {
 153                 // Create a File object from the row
 154                 // This will also upgrade it
 155                 if ( $row->oi_archive_name == '' ) {
 156                         $this->output( "Empty oi_archive_name for oi_name={$row->oi_name}\n" );
 157                         return false;
 158                 }
 159                 $file = $this->getRepo()->newFileFromRow( $row );
 160                 return $file->getUpgraded();
 161         }
 162
 163         function crawlMissing() {
 164                 $repo = RepoGroup::singleton()->getLocalRepo();
 165                 $repo->enumFilesInFS( array( $this, 'checkMissingImage' ) );
 166         }
 167
 168         function checkMissingImage( $fullpath ) {
 169                 $filename = wfBaseName( $fullpath );
 170                 if ( is_dir( $fullpath ) ) {
 171                         return;
 172                 }
 173                 if ( is_link( $fullpath ) ) {
 174                         $this->output( "skipping symlink at $fullpath\n" );
 175                         return;
 176                 }
 177                 $row = $this->dbw->selectRow( 'image',
 178                         array( 'img_name' ),
 179                         array( 'img_name' => $filename ),
 180                         __METHOD__ );
 181
 182                 if ( $row ) {
 183                         // already known, move on
 184                         return;
 185                 } else {
 186                         $this->addMissingImage( $filename, $fullpath );
 187                 }
 188         }
 189
 190         function addMissingImage( $filename, $fullpath ) {
 191                 $timestamp = $this->dbw->timestamp( filemtime( $fullpath ) );
 192
 193                 global $wgContLang;
 194                 $altname = $wgContLang->checkTitleEncoding( $filename );
 195                 if ( $altname != $filename ) {
 196                         if ( $this->dryrun ) {
 197                                 $filename = $altname;
 198                                 $this->output( "Estimating transcoding... $altname\n" );
 199                         } else {
 200                                 $filename = $this->renameFile( $filename );
 201                         }
 202                 }
 203
 204                 if ( $filename == '' ) {
 205                         $this->output( "Empty filename for $fullpath\n" );
 206                         return;
 207                 }
 208                 if ( !$this->dryrun ) {
 209                         $file = wfLocalFile( $filename );
 210                         if ( !$file->recordUpload( '', '(recovered file, missing upload log entry)', '', '', '',
 211                                 false, $timestamp ) )
 212                         {
 213                                 $this->output( "Error uploading file $fullpath\n" );
 214                                 return;
 215                         }
 216                 }
 217                 $this->output( $fullpath . "\n" );
 218         }
 219 }
 220
 221 $maintClass = 'ImageBuilder';
 222 require( RUN_MAINTENANCE_IF_MAIN );