maintenance/cleanupSpam.php

   1 <?php
   2 /**
   3  * Cleanup all spam from a given hostname
   4  *
   5  * This program is free software; you can redistribute it and/or modify
   6  * it under the terms of the GNU General Public License as published by
   7  * the Free Software Foundation; either version 2 of the License, or
   8  * (at your option) any later version.
   9  *
  10  * This program is distributed in the hope that it will be useful,
  11  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  12  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  13  * GNU General Public License for more details.
  14  *
  15  * You should have received a copy of the GNU General Public License along
  16  * with this program; if not, write to the Free Software Foundation, Inc.,
  17  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  18  * http://www.gnu.org/copyleft/gpl.html
  19  *
  20  * @ingroup Maintenance
  21  */
  22
  23 require_once( "Maintenance.php" );
  24
  25 class CleanupSpam extends Maintenance {
  26         public function __construct() {
  27                 parent::__construct();
  28                 $this->mDescription = "Cleanup all spam from a given hostname";
  29                 $this->addOption( 'all', 'Check all wikis in $wgLocalDatabases' );
  30                 $this->addArgs( array( 'hostname' ) );
  31         }
  32
  33         public function execute() {
  34                 global $wgLocalDatabases;
  35
  36                 $username = wfMsg( 'spambot_username' );
  37                 $wgUser = User::newFromName( $username );
  38                 // Create the user if necessary
  39                 if ( !$wgUser->getId() ) {
  40                         $wgUser->addToDatabase();
  41                 }
  42                 $spec = $this->getArg();
  43                 $like = LinkFilter::makeLike( $spec );
  44                 if ( !$like ) {
  45                         $this->error( "Not a valid hostname specification: $spec", true );
  46                 }
  47
  48                 $dbr = wfGetDB( DB_SLAVE );
  49
  50                 if ( $this->hasOption('all') ) {
  51                         // Clean up spam on all wikis
  52                         $dbr = wfGetDB( DB_SLAVE );
  53                         $this->output( "Finding spam on " . count($wgLocalDatabases) . " wikis\n" );
  54                         $found = false;
  55                         foreach ( $wgLocalDatabases as $db ) {
  56                                 $count = $dbr->selectField( "`$db`.externallinks", 'COUNT(*)',
  57                                         array( 'el_index LIKE ' . $dbr->addQuotes( $like ) ), __METHOD__ );
  58                                 if ( $count ) {
  59                                         $found = true;
  60                                         passthru( "php cleanupSpam.php $db $spec | sed s/^/$db:  /" );
  61                                 }
  62                         }
  63                         if ( $found ) {
  64                                 $this->output( "All done\n" );
  65                         } else {
  66                                 $this->output( "None found\n" );
  67                         }
  68                 } else {
  69                         // Clean up spam on this wiki
  70                         $res = $dbr->select( 'externallinks', array( 'DISTINCT el_from' ),
  71                                 array( 'el_index LIKE ' . $dbr->addQuotes( $like ) ), __METHOD__ );
  72                         $count = $dbr->numRows( $res );
  73                         $this->output( "Found $count articles containing $spec\n" );
  74                         while ( $row = $dbr->fetchObject( $res ) ) {
  75                                 $this->cleanupArticle( $row->el_from, $spec );
  76                         }
  77                         if ( $count ) {
  78                                 $this->output( "Done\n" );
  79                         }
  80                 }
  81         }
  82
  83         private function cleanupArticle( $id, $domain ) {
  84                 $title = Title::newFromID( $id );
  85                 if ( !$title ) {
  86                         $this->error( "Internal error: no page for ID $id" );
  87                         return;
  88                 }
  89
  90                 $this->output( $title->getPrefixedDBkey() . " ..." );
  91                 $rev = Revision::newFromTitle( $title );
  92                 $revId = $rev->getId();
  93                 $currentRevId = $revId;
  94
  95                 while ( $rev && LinkFilter::matchEntry( $rev->getText() , $domain ) ) {
  96                         # Revision::getPrevious can't be used in this way before MW 1.6 (Revision.php 1.26)
  97                         #$rev = $rev->getPrevious();
  98                         $revId = $title->getPreviousRevisionID( $revId );
  99                         if ( $revId ) {
 100                                 $rev = Revision::newFromTitle( $title, $revId );
 101                         } else {
 102                                 $rev = false;
 103                         }
 104                 }
 105                 if ( $revId == $currentRevId ) {
 106                         // The regex didn't match the current article text
 107                         // This happens e.g. when a link comes from a template rather than the page itself
 108                         $this->output( "False match\n" );
 109                 } else {
 110                         $dbw = wfGetDB( DB_MASTER );
 111                         $dbw->immediateBegin();
 112                         if ( !$rev ) {
 113                                 // Didn't find a non-spammy revision, blank the page
 114                                 $this->output( "blanking\n" );
 115                                 $article = new Article( $title );
 116                                 $article->updateArticle( '', wfMsg( 'spam_blanking', $domain ),
 117                                         false, false );
 118
 119                         } else {
 120                                 // Revert to this revision
 121                                 $this->output( "reverting\n" );
 122                                 $article = new Article( $title );
 123                                 $article->updateArticle( $rev->getText(), wfMsg( 'spam_reverting', $domain ), false, false );
 124                         }
 125                         $dbw->immediateCommit();
 126                         wfDoUpdates();
 127                 }
 128         }
 129 }
 130
 131 $maintClass = "CleanupSpam";
 132 require_once( DO_MAINTENANCE );