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  * @file
  21  * @ingroup Maintenance
  22  */
  23
  24 require_once( dirname( __FILE__ ) . '/Maintenance.php' );
  25
  26 class CleanupSpam extends Maintenance {
  27
  28         public function __construct() {
  29                 parent::__construct();
  30                 $this->mDescription = "Cleanup all spam from a given hostname";
  31                 $this->addOption( 'all', 'Check all wikis in $wgLocalDatabases' );
  32                 $this->addOption( 'delete', 'Delete pages containing only spam instead of blanking them' );
  33                 $this->addArg( 'hostname', 'Hostname that was spamming, single * wildcard in the beginning allowed' );
  34         }
  35
  36         public function execute() {
  37                 global $wgLocalDatabases, $wgUser;
  38
  39                 $username = wfMsg( 'spambot_username' );
  40                 $wgUser = User::newFromName( $username );
  41                 if ( !$wgUser ) {
  42                         $this->error( "Invalid username", true );
  43                 }
  44                 // Create the user if necessary
  45                 if ( !$wgUser->getId() ) {
  46                         $wgUser->addToDatabase();
  47                 }
  48                 $spec = $this->getArg();
  49                 $like = LinkFilter::makeLikeArray( $spec );
  50                 if ( !$like ) {
  51                         $this->error( "Not a valid hostname specification: $spec", true );
  52                 }
  53
  54                 if ( $this->hasOption( 'all' ) ) {
  55                         // Clean up spam on all wikis
  56                         $this->output( "Finding spam on " . count( $wgLocalDatabases ) . " wikis\n" );
  57                         $found = false;
  58                         foreach ( $wgLocalDatabases as $wikiID ) {
  59                                 $dbr = wfGetDB( DB_SLAVE, array(), $wikiID );
  60
  61                                 $count = $dbr->selectField( 'externallinks', 'COUNT(*)',
  62                                         array( 'el_index' . $dbr->buildLike( $like ) ), __METHOD__ );
  63                                 if ( $count ) {
  64                                         $found = true;
  65                                         passthru( "php cleanupSpam.php --wiki='$wikiID' $spec | sed 's/^/$wikiID:  /'" );
  66                                 }
  67                         }
  68                         if ( $found ) {
  69                                 $this->output( "All done\n" );
  70                         } else {
  71                                 $this->output( "None found\n" );
  72                         }
  73                 } else {
  74                         // Clean up spam on this wiki
  75
  76                         $dbr = wfGetDB( DB_SLAVE );
  77                         $res = $dbr->select( 'externallinks', array( 'DISTINCT el_from' ),
  78                                 array( 'el_index' . $dbr->buildLike( $like ) ), __METHOD__ );
  79                         $count = $dbr->numRows( $res );
  80                         $this->output( "Found $count articles containing $spec\n" );
  81                         foreach ( $res as $row ) {
  82                                 $this->cleanupArticle( $row->el_from, $spec );
  83                         }
  84                         if ( $count ) {
  85                                 $this->output( "Done\n" );
  86                         }
  87                 }
  88         }
  89
  90         private function cleanupArticle( $id, $domain ) {
  91                 $title = Title::newFromID( $id );
  92                 if ( !$title ) {
  93                         $this->error( "Internal error: no page for ID $id" );
  94                         return;
  95                 }
  96
  97                 $this->output( $title->getPrefixedDBkey() . " ..." );
  98                 $rev = Revision::newFromTitle( $title );
  99                 $currentRevId = $rev->getId();
 100
 101                 //FIXME: LinkFilter needs to handle Content objects! Or rather, ContentHandler needs to provide the appropriate LinkFilter.
 102                 while ( $rev && ( $rev->isDeleted( Revision::DELETED_TEXT ) || LinkFilter::matchEntry( $rev->getText() , $domain ) ) ) {
 103                         $rev = $rev->getPrevious();
 104                 }
 105
 106                 if ( $rev && $rev->getId() == $currentRevId ) {
 107                         // The regex didn't match the current article text
 108                         // This happens e.g. when a link comes from a template rather than the page itself
 109                         $this->output( "False match\n" );
 110                 } else {
 111                         $dbw = wfGetDB( DB_MASTER );
 112                         $dbw->begin( __METHOD__ );
 113                         $page = WikiPage::factory( $title );
 114                         if ( $rev ) {
 115                                 // Revert to this revision
 116                                 $this->output( "reverting\n" );
 117                                 $page->doEdit( $rev->getText(), wfMsgForContent( 'spam_reverting', $domain ),
 118                                         EDIT_UPDATE, $rev->getId() );
 119                         } elseif ( $this->hasOption( 'delete' ) ) {
 120                                 // Didn't find a non-spammy revision, blank the page
 121                                 $this->output( "deleting\n" );
 122                                 $page->doDeleteArticle( wfMsgForContent( 'spam_deleting', $domain ) );
 123                         } else {
 124                                 // Didn't find a non-spammy revision, blank the page
 125                                 $this->output( "blanking\n" );
 126                                 $page->doEdit( '', wfMsgForContent( 'spam_blanking', $domain ) );
 127                         }
 128                         $dbw->commit( __METHOD__ );
 129                 }
 130         }
 131 }
 132
 133 $maintClass = "CleanupSpam";
 134 require_once( RUN_MAINTENANCE_IF_MAIN );