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