maintenance/cleanupTitles.php

   1 <?php
   2 /*
   3  * Script to clean up broken, unparseable titles.
   4  *
   5  * Usage: php cleanupTitles.php [--fix]
   6  * Options:
   7  *   --fix  Actually clean up titles; otherwise just checks for them
   8  *
   9  * Copyright (C) 2005 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 TitleCleanup extends TableCleanup {
  36         function __construct( $dryrun = false ) {
  37                 parent::__construct( 'page', $dryrun );
  38         }
  39
  40         function processPage( $row ) {
  41                 global $wgContLang;
  42
  43                 $current = Title::makeTitle( $row->page_namespace, $row->page_title );
  44                 $display = $current->getPrefixedText();
  45
  46                 $verified = UtfNormal::cleanUp( $display );
  47
  48                 $title = Title::newFromText( $verified );
  49
  50                 if( is_null( $title ) ) {
  51                         $this->log( "page $row->page_id ($display) is illegal." );
  52                         $this->moveIllegalPage( $row );
  53                         return $this->progress( 1 );
  54                 }
  55
  56                 if( !$title->equals( $current ) ) {
  57                         $this->log( "page $row->page_id ($display) doesn't match self." );
  58                         $this->moveInconsistentPage( $row, $title );
  59                         return $this->progress( 1 );
  60                 }
  61
  62                 $this->progress( 0 );
  63         }
  64
  65         function moveIllegalPage( $row ) {
  66                 $legal = 'A-Za-z0-9_/\\\\-';
  67                 $legalized = preg_replace_callback( "!([^$legal])!",
  68                         array( &$this, 'hexChar' ),
  69                         $row->page_title );
  70                 if( $legalized == '.' ) $legalized = '(dot)';
  71                 if( $legalized == '_' ) $legalized = '(space)';
  72                 $legalized = 'Broken/' . $legalized;
  73
  74                 $title = Title::newFromText( $legalized );
  75                 if( is_null( $title ) ) {
  76                         $clean = 'Broken/id:' . $row->page_id;
  77                         $this->log( "Couldn't legalize; form '$legalized' still invalid; using '$clean'" );
  78                         $title = Title::newFromText( $clean );
  79                 } elseif( $title->exists() ) {
  80                         $clean = 'Broken/id:' . $row->page_id;
  81                         $this->log( "Legalized for '$legalized' exists; using '$clean'" );
  82                         $title = Title::newFromText( $clean );
  83                 }
  84
  85                 $dest = $title->getDbKey();
  86                 if( $this->dryrun ) {
  87                         $this->log( "DRY RUN: would rename $row->page_id ($row->page_namespace,'$row->page_title') to ($row->page_namespace,'$dest')" );
  88                 } else {
  89                         $this->log( "renaming $row->page_id ($row->page_namespace,'$row->page_title') to ($row->page_namespace,'$dest')" );
  90                         $dbw =& wfGetDB( DB_MASTER );
  91                         $dbw->update( 'page',
  92                                 array( 'page_title' => $dest ),
  93                                 array( 'page_id' => $row->page_id ),
  94                                 'cleanupTitles::moveInconsistentPage' );
  95                 }
  96         }
  97
  98         function moveInconsistentPage( $row, $title ) {
  99                 if( $title->exists() || $title->getInterwiki() ) {
 100                         if( $title->getInterwiki() ) {
 101                                 $prior = $title->getPrefixedDbKey();
 102                         } else {
 103                                 $prior = $title->getDbKey();
 104                         }
 105                         $clean = 'Broken/' . $prior;
 106                         $verified = Title::makeTitleSafe( $row->page_namespace, $clean );
 107                         if( $verified->exists() ) {
 108                                 $blah = "Broken/id:" . $row->page_id;
 109                                 $this->log( "Couldn't legalize; form '$clean' exists; using '$blah'" );
 110                                 $verified = Title::makeTitleSafe( $row->page_namespace, $blah );
 111                         }
 112                         $title = $verified;
 113                 }
 114                 if( is_null( $title ) ) {
 115                         wfDie( "Something awry; empty title.\n" );
 116                 }
 117                 $ns = $title->getNamespace();
 118                 $dest = $title->getDbKey();
 119                 if( $this->dryrun ) {
 120                         $this->log( "DRY RUN: would rename $row->page_id ($row->page_namespace,'$row->page_title') to ($row->page_namespace,'$dest')" );
 121                 } else {
 122                         $this->log( "renaming $row->page_id ($row->page_namespace,'$row->page_title') to ($ns,'$dest')" );
 123                         $dbw =& wfGetDB( DB_MASTER );
 124                         $dbw->update( 'page',
 125                                 array(
 126                                         'page_namespace' => $ns,
 127                                         'page_title' => $dest
 128                                 ),
 129                                 array( 'page_id' => $row->page_id ),
 130                                 'cleanupTitles::moveInconsistentPage' );
 131                         $linkCache =& LinkCache::singleton();
 132                         $linkCache->clear();
 133                 }
 134         }
 135
 136         function hexChar( $matches ) {
 137                 return sprintf( "\\x%02x", ord( $matches[1] ) );
 138         }
 139 }
 140
 141 $wgUser->setName( 'Conversion script' );
 142 $caps = new TitleCleanup( !isset( $options['fix'] ) );
 143 $caps->cleanup();
 144
 145 ?>