maintenance/cleanupTitles.php

   1 <?php
   2 /**
   3  * Clean up broken, unparseable titles.
   4  *
   5  * Copyright © 2005 Brion Vibber <brion@pobox.com>
   6  * https://www.mediawiki.org/
   7  *
   8  * This program is free software; you can redistribute it and/or modify
   9  * it under the terms of the GNU General Public License as published by
  10  * the Free Software Foundation; either version 2 of the License, or
  11  * (at your option) any later version.
  12  *
  13  * This program is distributed in the hope that it will be useful,
  14  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  15  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  16  * GNU General Public License for more details.
  17  *
  18  * You should have received a copy of the GNU General Public License along
  19  * with this program; if not, write to the Free Software Foundation, Inc.,
  20  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  21  * http://www.gnu.org/copyleft/gpl.html
  22  *
  23  * @file
  24  * @author Brion Vibber <brion at pobox.com>
  25  * @ingroup Maintenance
  26  */
  27
  28 use MediaWiki\MediaWikiServices;
  29
  30 require_once __DIR__ . '/cleanupTable.inc';
  31
  32 /**
  33  * Maintenance script to clean up broken, unparseable titles.
  34  *
  35  * @ingroup Maintenance
  36  */
  37 class TitleCleanup extends TableCleanup {
  38         public function __construct() {
  39                 parent::__construct();
  40                 $this->addDescription( 'Script to clean up broken, unparseable titles' );
  41         }
  42
  43         /**
  44          * @param object $row
  45          */
  46         protected function processRow( $row ) {
  47                 global $wgContLang;
  48                 $display = Title::makeName( $row->page_namespace, $row->page_title );
  49                 $verified = $wgContLang->normalize( $display );
  50                 $title = Title::newFromText( $verified );
  51
  52                 if ( !is_null( $title )
  53                         && $title->canExist()
  54                         && $title->getNamespace() == $row->page_namespace
  55                         && $title->getDBkey() === $row->page_title
  56                 ) {
  57                         $this->progress( 0 ); // all is fine
  58
  59                         return;
  60                 }
  61
  62                 if ( $row->page_namespace == NS_FILE && $this->fileExists( $row->page_title ) ) {
  63                         $this->output( "file $row->page_title needs cleanup, please run cleanupImages.php.\n" );
  64                         $this->progress( 0 );
  65                 } elseif ( is_null( $title ) ) {
  66                         $this->output( "page $row->page_id ($display) is illegal.\n" );
  67                         $this->moveIllegalPage( $row );
  68                         $this->progress( 1 );
  69                 } else {
  70                         $this->output( "page $row->page_id ($display) doesn't match self.\n" );
  71                         $this->moveInconsistentPage( $row, $title );
  72                         $this->progress( 1 );
  73                 }
  74         }
  75
  76         /**
  77          * @param string $name
  78          * @return bool
  79          */
  80         protected function fileExists( $name ) {
  81                 // XXX: Doesn't actually check for file existence, just presence of image record.
  82                 // This is reasonable, since cleanupImages.php only iterates over the image table.
  83                 $dbr = $this->getDB( DB_REPLICA );
  84                 $row = $dbr->selectRow( 'image', [ 'img_name' ], [ 'img_name' => $name ], __METHOD__ );
  85
  86                 return $row !== false;
  87         }
  88
  89         /**
  90          * @param object $row
  91          */
  92         protected function moveIllegalPage( $row ) {
  93                 $legal = 'A-Za-z0-9_/\\\\-';
  94                 $legalized = preg_replace_callback( "!([^$legal])!",
  95                         [ $this, 'hexChar' ],
  96                         $row->page_title );
  97                 if ( $legalized == '.' ) {
  98                         $legalized = '(dot)';
  99                 }
 100                 if ( $legalized == '_' ) {
 101                         $legalized = '(space)';
 102                 }
 103                 $legalized = 'Broken/' . $legalized;
 104
 105                 $title = Title::newFromText( $legalized );
 106                 if ( is_null( $title ) ) {
 107                         $clean = 'Broken/id:' . $row->page_id;
 108                         $this->output( "Couldn't legalize; form '$legalized' still invalid; using '$clean'\n" );
 109                         $title = Title::newFromText( $clean );
 110                 } elseif ( $title->exists() ) {
 111                         $clean = 'Broken/id:' . $row->page_id;
 112                         $this->output( "Legalized for '$legalized' exists; using '$clean'\n" );
 113                         $title = Title::newFromText( $clean );
 114                 }
 115
 116                 $dest = $title->getDBkey();
 117                 if ( $this->dryrun ) {
 118                         $this->output( "DRY RUN: would rename $row->page_id ($row->page_namespace," .
 119                                 "'$row->page_title') to ($row->page_namespace,'$dest')\n" );
 120                 } else {
 121                         $this->output( "renaming $row->page_id ($row->page_namespace," .
 122                                 "'$row->page_title') to ($row->page_namespace,'$dest')\n" );
 123                         $dbw = $this->getDB( DB_MASTER );
 124                         $dbw->update( 'page',
 125                                 [ 'page_title' => $dest ],
 126                                 [ 'page_id' => $row->page_id ],
 127                                 __METHOD__ );
 128                 }
 129         }
 130
 131         /**
 132          * @param object $row
 133          * @param Title $title
 134          */
 135         protected function moveInconsistentPage( $row, Title $title ) {
 136                 if ( $title->exists( Title::GAID_FOR_UPDATE )
 137                         || $title->getInterwiki()
 138                         || !$title->canExist()
 139                 ) {
 140                         if ( $title->getInterwiki() || !$title->canExist() ) {
 141                                 $prior = $title->getPrefixedDBkey();
 142                         } else {
 143                                 $prior = $title->getDBkey();
 144                         }
 145
 146                         # Old cleanupTitles could move articles there. See T25147.
 147                         $ns = $row->page_namespace;
 148                         if ( $ns < 0 ) {
 149                                 $ns = 0;
 150                         }
 151
 152                         # Namespace which no longer exists. Put the page in the main namespace
 153                         # since we don't have any idea of the old namespace name. See T70501.
 154                         if ( !MWNamespace::exists( $ns ) ) {
 155                                 $ns = 0;
 156                         }
 157
 158                         $clean = 'Broken/' . $prior;
 159                         $verified = Title::makeTitleSafe( $ns, $clean );
 160                         if ( !$verified || $verified->exists() ) {
 161                                 $blah = "Broken/id:" . $row->page_id;
 162                                 $this->output( "Couldn't legalize; form '$clean' exists; using '$blah'\n" );
 163                                 $verified = Title::makeTitleSafe( $ns, $blah );
 164                         }
 165                         $title = $verified;
 166                 }
 167                 if ( is_null( $title ) ) {
 168                         $this->error( "Something awry; empty title.", true );
 169                 }
 170                 $ns = $title->getNamespace();
 171                 $dest = $title->getDBkey();
 172
 173                 if ( $this->dryrun ) {
 174                         $this->output( "DRY RUN: would rename $row->page_id ($row->page_namespace," .
 175                                 "'$row->page_title') to ($ns,'$dest')\n" );
 176                 } else {
 177                         $this->output( "renaming $row->page_id ($row->page_namespace," .
 178                                 "'$row->page_title') to ($ns,'$dest')\n" );
 179                         $dbw = $this->getDB( DB_MASTER );
 180                         $dbw->update( 'page',
 181                                 [
 182                                         'page_namespace' => $ns,
 183                                         'page_title' => $dest
 184                                 ],
 185                                 [ 'page_id' => $row->page_id ],
 186                                 __METHOD__ );
 187                         MediaWikiServices::getInstance()->getLinkCache()->clear();
 188                 }
 189         }
 190 }
 191
 192 $maintClass = "TitleCleanup";
 193 require_once RUN_MAINTENANCE_IF_MAIN;