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