maintenance/cleanupWatchlist.php

   1 <?php
   2 /*
   3  * Script to remove broken, unparseable titles in the Watchlist.
   4  *
   5  * Usage: php cleanupWatchlist.php [--fix]
   6  * Options:
   7  *   --fix  Actually remove entries; without will only report.
   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  * @addtogroup maintenance
  29  */
  30
  31 $options = array( 'fix' );
  32
  33 require_once( 'commandLine.inc' );
  34 require_once( 'FiveUpgrade.inc' );
  35
  36 class WatchlistCleanup extends FiveUpgrade {
  37         function WatchlistCleanup( $dryrun = false ) {
  38                 parent::FiveUpgrade();
  39
  40                 $this->maxLag = 10; # if slaves are lagged more than 10 secs, wait
  41                 $this->dryrun = $dryrun;
  42         }
  43
  44         function cleanup() {
  45                 $this->runTable( 'watchlist',
  46                         '',
  47                         array( &$this, 'processEntry' ) );
  48         }
  49
  50         function init( $count, $table ) {
  51                 $this->processed = 0;
  52                 $this->updated = 0;
  53                 $this->count = $count;
  54                 $this->startTime = wfTime();
  55                 $this->table = $table;
  56         }
  57
  58         function progress( $updated ) {
  59                 $this->updated += $updated;
  60                 $this->processed++;
  61                 if( $this->processed % 100 != 0 ) {
  62                         return;
  63                 }
  64                 $portion = $this->processed / $this->count;
  65                 $updateRate = $this->updated / $this->processed;
  66
  67                 $now = wfTime();
  68                 $delta = $now - $this->startTime;
  69                 $estimatedTotalTime = $delta / $portion;
  70                 $eta = $this->startTime + $estimatedTotalTime;
  71
  72                 printf( "%s %s: %6.2f%% done on %s; ETA %s [%d/%d] %.2f/sec <%.2f%% updated>\n",
  73                         wfWikiID(),
  74                         wfTimestamp( TS_DB, intval( $now ) ),
  75                         $portion * 100.0,
  76                         $this->table,
  77                         wfTimestamp( TS_DB, intval( $eta ) ),
  78                         $this->processed,
  79                         $this->count,
  80                         $this->processed / $delta,
  81                         $updateRate * 100.0 );
  82                 flush();
  83         }
  84
  85         function runTable( $table, $where, $callback ) {
  86                 $fname = 'WatchlistCleanup::runTable';
  87
  88                 $count = $this->dbw->selectField( $table, 'count(*)', '', $fname );
  89                 $this->init( $count, 'watchlist' );
  90                 $this->log( "Processing $table..." );
  91
  92                 $tableName = $this->dbr->tableName( $table );
  93                 $sql = "SELECT * FROM $tableName $where";
  94                 $result = $this->dbr->query( $sql, $fname );
  95
  96                 while( $row = $this->dbr->fetchObject( $result ) ) {
  97                         call_user_func( $callback, $row );
  98                 }
  99                 $this->log( "Finished $table... $this->updated of $this->processed rows updated" );
 100                 $this->dbr->freeResult( $result );
 101         }
 102
 103         function processEntry( $row ) {
 104                 $current = Title::makeTitle( $row->wl_namespace, $row->wl_title );
 105                 $display = $current->getPrefixedText();
 106
 107                 $verified = UtfNormal::cleanUp( $display );
 108
 109                 $title = Title::newFromText( $verified );
 110
 111                 if( $row->wl_user == 0 || is_null( $title ) || !$title->equals( $current ) ) {
 112                         $this->log( "invalid watch by {$row->wl_user} for ({$row->wl_namespace}, \"{$row->wl_title}\")" );
 113                         $this->removeWatch( $row );
 114                         return $this->progress( 1 );
 115                 }
 116
 117                 $this->progress( 0 );
 118         }
 119
 120         function removeWatch( $row ) {
 121                 if( !$this->dryrun) {
 122                         $dbw =& wfGetDB( DB_MASTER );
 123                         $dbw->delete( 'watchlist', array(
 124                                 'wl_user'      => $row->wl_user,
 125                                 'wl_namespace' => $row->wl_namespace,
 126                                 'wl_title'     => $row->wl_title ),
 127                         'WatchlistCleanup::removeWatch' );
 128                         $this->log( '- removed' );
 129                 }
 130         }
 131 }
 132
 133 $wgUser->setName( 'Conversion script' );
 134 $caps = new WatchlistCleanup( !isset( $options['fix'] ) );
 135 $caps->cleanup();
 136
 137 ?>