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                 $this->mDescription = 'Script to update image metadata records';
  46
  47                 $this->addOption( 'missing', 'Check for files without associated database record' );
  48                 $this->addOption( 'dry-run', 'Only report, don\'t update the database' );
  49         }
  50
  51         public function execute() {
  52                 $this->dbw = wfGetDB( DB_MASTER );
  53                 $this->maxLag = 10; # if slaves are lagged more than 10 secs, wait
  54                 $this->dryrun = $this->hasOption( 'dry-run' );
  55                 if ( $this->dryrun ) {
  56                         $GLOBALS['wgReadOnly'] = 'Dry run mode, image upgrades are suppressed';
  57                 }
  58
  59                 if ( $this->hasOption( 'missing' ) ) {
  60                         $this->crawlMissing();
  61                 } else {
  62                         $this->build();
  63                 }
  64         }
  65
  66         function getRepo() {
  67                 if ( !isset( $this->repo ) ) {
  68                         $this->repo = RepoGroup::singleton()->getLocalRepo();
  69                 }
  70                 return $this->repo;
  71         }
  72
  73         function build() {
  74                 $this->buildImage();
  75                 $this->buildOldImage();
  76         }
  77
  78         function init( $count, $table ) {
  79                 $this->processed = 0;
  80                 $this->updated = 0;
  81                 $this->count = $count;
  82                 $this->startTime = wfTime();
  83                 $this->table = $table;
  84         }
  85
  86         function progress( $updated ) {
  87                 $this->updated += $updated;
  88                 $this->processed++;
  89                 if ( $this->processed % 100 != 0 ) {
  90                         return;
  91                 }
  92                 $portion = $this->processed / $this->count;
  93                 $updateRate = $this->updated / $this->processed;
  94
  95                 $now = wfTime();
  96                 $delta = $now - $this->startTime;
  97                 $estimatedTotalTime = $delta / $portion;
  98                 $eta = $this->startTime + $estimatedTotalTime;
  99                 $rate = $this->processed / $delta;
 100
 101                 $this->output( sprintf( "%s: %6.2f%% done on %s; ETA %s [%d/%d] %.2f/sec <%.2f%% updated>\n",
 102                         wfTimestamp( TS_DB, intval( $now ) ),
 103                         $portion * 100.0,
 104                         $this->table,
 105                         wfTimestamp( TS_DB, intval( $eta ) ),
 106                         $this->processed,
 107                         $this->count,
 108                         $rate,
 109                         $updateRate * 100.0 ) );
 110                 flush();
 111         }
 112
 113         function buildTable( $table, $key, $callback ) {
 114                 $count = $this->dbw->selectField( $table, 'count(*)', '', __METHOD__ );
 115                 $this->init( $count, $table );
 116                 $this->output( "Processing $table...\n" );
 117
 118                 $result = wfGetDB( DB_SLAVE )->select( $table, '*', array(), __METHOD__ );
 119
 120                 foreach ( $result as $row ) {
 121                         $update = call_user_func( $callback, $row, null );
 122                         if ( $update ) {
 123                                 $this->progress( 1 );
 124                         } else {
 125                                 $this->progress( 0 );
 126                         }
 127                 }
 128                 $this->output( "Finished $table... $this->updated of $this->processed rows updated\n" );
 129         }
 130
 131         function buildImage() {
 132                 $callback = array( $this, 'imageCallback' );
 133                 $this->buildTable( 'image', 'img_name', $callback );
 134         }
 135
 136         function imageCallback( $row, $copy ) {
 137                 // Create a File object from the row
 138                 // This will also upgrade it
 139                 $file = $this->getRepo()->newFileFromRow( $row );
 140                 return $file->getUpgraded();
 141         }
 142
 143         function buildOldImage() {
 144                 $this->buildTable( 'oldimage', 'oi_archive_name',
 145                         array( $this, 'oldimageCallback' ) );
 146         }
 147
 148         function oldimageCallback( $row, $copy ) {
 149                 // Create a File object from the row
 150                 // This will also upgrade it
 151                 if ( $row->oi_archive_name == '' ) {
 152                         $this->output( "Empty oi_archive_name for oi_name={$row->oi_name}\n" );
 153                         return false;
 154                 }
 155                 $file = $this->getRepo()->newFileFromRow( $row );
 156                 return $file->getUpgraded();
 157         }
 158
 159         function crawlMissing() {
 160                 $repo = RepoGroup::singleton()->getLocalRepo();
 161                 $repo->enumFilesInFS( array( $this, 'checkMissingImage' ) );
 162         }
 163
 164         function checkMissingImage( $fullpath ) {
 165                 $filename = wfBaseName( $fullpath );
 166                 if ( is_dir( $fullpath ) ) {
 167                         return;
 168                 }
 169                 if ( is_link( $fullpath ) ) {
 170                         $this->output( "skipping symlink at $fullpath\n" );
 171                         return;
 172                 }
 173                 $row = $this->dbw->selectRow( 'image',
 174                         array( 'img_name' ),
 175                         array( 'img_name' => $filename ),
 176                         __METHOD__ );
 177
 178                 if ( $row ) {
 179                         // already known, move on
 180                         return;
 181                 } else {
 182                         $this->addMissingImage( $filename, $fullpath );
 183                 }
 184         }
 185
 186         function addMissingImage( $filename, $fullpath ) {
 187                 $timestamp = $this->dbw->timestamp( filemtime( $fullpath ) );
 188
 189                 global $wgContLang;
 190                 $altname = $wgContLang->checkTitleEncoding( $filename );
 191                 if ( $altname != $filename ) {
 192                         if ( $this->dryrun ) {
 193                                 $filename = $altname;
 194                                 $this->output( "Estimating transcoding... $altname\n" );
 195                         } else {
 196                                 $filename = $this->renameFile( $filename );
 197                         }
 198                 }
 199
 200                 if ( $filename == '' ) {
 201                         $this->output( "Empty filename for $fullpath\n" );
 202                         return;
 203                 }
 204                 if ( !$this->dryrun ) {
 205                         $file = wfLocalFile( $filename );
 206                         if ( !$file->recordUpload( '', '(recovered file, missing upload log entry)', '', '', '',
 207                                 false, $timestamp ) )
 208                         {
 209                                 $this->output( "Error uploading file $fullpath\n" );
 210                                 return;
 211                         }
 212                 }
 213                 $this->output( $fullpath . "\n" );
 214         }
 215 }
 216
 217 $maintClass = 'ImageBuilder';
 218 require( RUN_MAINTENANCE_IF_MAIN );