maintenance/cleanupImages.php

   1 <?php
   2 /*
   3  * Script to clean up broken, unparseable upload filenames.
   4  *
   5  * Usage: php cleanupImages.php [--fix]
   6  * Options:
   7  *   --fix  Actually clean up titles; otherwise just checks for them
   8  *
   9  * Copyright (C) 2005-2006 Brion Vibber <brion@pobox.com>
  10  * http://www.mediawiki.org/
  11  *
  12  * This program is free software; you can redistribute it and/or modify
  13  * it under the terms of the GNU General Public License as published by
  14  * the Free Software Foundation; either version 2 of the License, or
  15  * (at your option) any later version.
  16  *
  17  * This program is distributed in the hope that it will be useful,
  18  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  19  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  20  * GNU General Public License for more details.
  21  *
  22  * You should have received a copy of the GNU General Public License along
  23  * with this program; if not, write to the Free Software Foundation, Inc.,
  24  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  25  * http://www.gnu.org/copyleft/gpl.html
  26  *
  27  * @author Brion Vibber <brion at pobox.com>
  28  * @package MediaWiki
  29  * @subpackage maintenance
  30  */
  31
  32 require_once( 'commandLine.inc' );
  33 require_once( 'cleanupTable.inc' );
  34
  35 class ImageCleanup extends TableCleanup {
  36         function __construct( $dryrun = false ) {
  37                 parent::__construct( 'image', $dryrun );
  38         }
  39
  40         function processPage( $row ) {
  41                 global $wgContLang;
  42
  43                 $source = $row->img_name;
  44                 if( $source == '' ) {
  45                         // Ye olde empty rows. Just kill them.
  46                         $this->killRow( $source );
  47                         return $this->progress( 1 );
  48                 }
  49
  50                 $cleaned = $source;
  51
  52                 // About half of old bad image names have percent-codes
  53                 $cleaned = rawurldecode( $cleaned );
  54
  55                 // Some are old latin-1
  56                 $cleaned = $wgContLang->checkTitleEncoding( $cleaned );
  57
  58                 // Many of remainder look like non-normalized unicode
  59                 $cleaned = UtfNormal::cleanUp( $cleaned );
  60
  61                 $title = Title::makeTitleSafe( NS_IMAGE, $cleaned );
  62
  63                 if( is_null( $title ) ) {
  64                         $this->log( "page $source ($cleaned) is illegal." );
  65                         $safe = $this->buildSafeTitle( $cleaned );
  66                         $this->pokeFile( $source, $safe );
  67                         return $this->progress( 1 );
  68                 }
  69
  70                 if( $title->getDbKey() !== $source ) {
  71                         $munged = $title->getDbKey();
  72                         $this->log( "page $source ($munged) doesn't match self." );
  73                         $this->pokeFile( $source, $munged );
  74                         return $this->progress( 1 );
  75                 }
  76
  77                 $this->progress( 0 );
  78         }
  79
  80         function killRow( $name ) {
  81                 if( $this->dryrun ) {
  82                         $this->log( "DRY RUN: would delete bogus row '$name'" );
  83                 } else {
  84                         $this->log( "deleting bogus row '$name'" );
  85                         $db = wfGetDB( DB_MASTER );
  86                         $db->delete( 'image',
  87                                 array( 'img_name' => $name ),
  88                                 __METHOD__ );
  89                 }
  90         }
  91
  92         function filePath( $name ) {
  93                 return wfImageDir( $name ) . "/$name";
  94         }
  95
  96         function pokeFile( $orig, $new ) {
  97                 $path = $this->filePath( $orig );
  98                 if( !file_exists( $path ) ) {
  99                         $this->log( "missing file: $path" );
 100                         return $this->killRow( $orig );
 101                 }
 102
 103                 $db = wfGetDB( DB_MASTER );
 104                 $version = 0;
 105                 $final = $new;
 106
 107                 while( $db->selectField( 'image', 'img_name',
 108                         array( 'img_name' => $final ), __METHOD__ ) ) {
 109                         $this->log( "Rename conflicts with '$final'..." );
 110                         $version++;
 111                         $final = $this->appendTitle( $new, "_$version" );
 112                 }
 113
 114                 $finalPath = $this->filePath( $final );
 115
 116                 if( $this->dryrun ) {
 117                         $this->log( "DRY RUN: would rename $path to $finalPath" );
 118                 } else {
 119                         $this->log( "renaming $path to $finalPath" );
 120                         $db->begin();
 121                         $db->update( 'image',
 122                                 array( 'img_name' => $final ),
 123                                 array( 'img_name' => $orig ),
 124                                 __METHOD__ );
 125                         $dir = dirname( $finalPath );
 126                         if( !file_exists( $dir ) ) {
 127                                 if( !mkdir( $dir, 0777, true ) ) {
 128                                         $this->log( "RENAME FAILED, COULD NOT CREATE $dir" );
 129                                         $db->rollback();
 130                                         return;
 131                                 }
 132                         }
 133                         if( rename( $path, $finalPath ) ) {
 134                                 $db->commit();
 135                         } else {
 136                                 $this->log( "RENAME FAILED" );
 137                                 $db->rollback();
 138                         }
 139                 }
 140         }
 141
 142         function appendTitle( $name, $suffix ) {
 143                 return preg_replace( '/^(.*)(\..*?)$/',
 144                         "\\1$suffix\\2", $name );
 145         }
 146
 147         function buildSafeTitle( $name ) {
 148                 global $wgLegalTitleChars;
 149                 $x = preg_replace_callback(
 150                         "/([^$wgLegalTitleChars])/",
 151                         array( $this, 'hexChar' ),
 152                         $name );
 153
 154                 $test = Title::makeTitleSafe( NS_IMAGE, $x );
 155                 if( is_null( $test ) || $test->getDbKey() !== $x ) {
 156                         $this->log( "Unable to generate safe title from '$name', got '$x'" );
 157                         return false;
 158                 }
 159
 160                 return $x;
 161         }
 162 }
 163
 164 $wgUser->setName( 'Conversion script' );
 165 $caps = new ImageCleanup( !isset( $options['fix'] ) );
 166 $caps->cleanup();
 167
 168 ?>