maintenance/cleanupDupes.inc

   1 <?php
   2 # Copyright (C) 2004 Brion Vibber <brion@pobox.com>
   3 # http://www.mediawiki.org/
   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 /**
  21  * If on the old non-unique indexes, check the cur table for duplicate
  22  * entries and remove them...
  23  *
  24  * @addtogroup Maintenance
  25  */
  26
  27 function fixDupes( $fixthem = false) {
  28         $dbw = wfGetDB( DB_MASTER );
  29         $cur = $dbw->tableName( 'cur' );
  30         $old = $dbw->tableName( 'old' );
  31         $dbw->query( "LOCK TABLES $cur WRITE, $old WRITE" );
  32         echo "Checking for duplicate cur table entries... (this may take a while on a large wiki)\n";
  33         $res = $dbw->query( <<<END
  34 SELECT cur_namespace,cur_title,count(*) as c,min(cur_id) as id
  35   FROM $cur
  36  GROUP BY cur_namespace,cur_title
  37 HAVING c > 1
  38 END
  39         );
  40         $n = $dbw->numRows( $res );
  41         echo "Found $n titles with duplicate entries.\n";
  42         if( $n > 0 ) {
  43                 if( $fixthem ) {
  44                         echo "Correcting...\n";
  45                 } else {
  46                         echo "Just a demo...\n";
  47                 }
  48                 while( $row = $dbw->fetchObject( $res ) ) {
  49                         $ns = intval( $row->cur_namespace );
  50                         $title = $dbw->addQuotes( $row->cur_title );
  51
  52                         # Get the first responding ID; that'll be the one we keep.
  53                         $id = $dbw->selectField( 'cur', 'cur_id', array(
  54                                 'cur_namespace' => $row->cur_namespace,
  55                                 'cur_title'     => $row->cur_title ) );
  56
  57                         echo "$ns:$row->cur_title (canonical ID $id)\n";
  58                         if( $id != $row->id ) {
  59                                 echo "  ** minimum ID $row->id; ";
  60                                 $timeMin = $dbw->selectField( 'cur', 'cur_timestamp', array(
  61                                         'cur_id' => $row->id ) );
  62                                 $timeFirst = $dbw->selectField( 'cur', 'cur_timestamp', array(
  63                                         'cur_id' => $id ) );
  64                                 if( $timeMin == $timeFirst ) {
  65                                         echo "timestamps match at $timeFirst; ok\n";
  66                                 } else {
  67                                         echo "timestamps don't match! min: $timeMin, first: $timeFirst; ";
  68                                         if( $timeMin > $timeFirst ) {
  69                                                 $id = $row->id;
  70                                                 echo "keeping minimum: $id\n";
  71                                         } else {
  72                                                 echo "keeping first: $id\n";
  73                                         }
  74                                 }
  75                         }
  76
  77                         if( $fixthem ) {
  78                                 $dbw->query( <<<END
  79 INSERT
  80   INTO $old
  81       (old_namespace, old_title,      old_text,
  82        old_comment,   old_user,       old_user_text,
  83        old_timestamp, old_minor_edit, old_flags,
  84        inverse_timestamp)
  85 SELECT cur_namespace, cur_title,      cur_text,
  86        cur_comment,   cur_user,       cur_user_text,
  87        cur_timestamp, cur_minor_edit, '',
  88        inverse_timestamp
  89   FROM $cur
  90  WHERE cur_namespace=$ns
  91    AND cur_title=$title
  92    AND cur_id != $id
  93 END
  94                                 );
  95                                 $dbw->query( <<<END
  96 DELETE
  97   FROM $cur
  98  WHERE cur_namespace=$ns
  99    AND cur_title=$title
 100    AND cur_id != $id
 101 END
 102                                 );
 103                         }
 104                 }
 105         }
 106         $dbw->query( 'UNLOCK TABLES' );
 107         if( $fixthem ) {
 108                 echo "Done.\n";
 109         } else {
 110                 echo "Run again with --fix option to delete the duplicates.\n";
 111         }
 112 }
 113
 114 function checkDupes( $fixthem = false, $indexonly = false ) {
 115         $dbw = wfGetDB( DB_MASTER );
 116         if( $dbw->indexExists( 'cur', 'name_title' ) &&
 117             $dbw->indexUnique( 'cur', 'name_title' ) ) {
 118                 echo wfWikiID().": cur table has the current unique index; no duplicate entries.\n";
 119         } elseif( $dbw->indexExists( 'cur', 'name_title_dup_prevention' ) ) {
 120                 echo wfWikiID().": cur table has a temporary name_title_dup_prevention unique index; no duplicate entries.\n";
 121         } else {
 122                 echo wfWikiID().": cur table has the old non-unique index and may have duplicate entries.\n";
 123                 if( !$indexonly ) {
 124                         fixDupes( $fixthem );
 125                 }
 126         }
 127 }
 128
 129 ?>