maintenance/rebuildtextindex.php

   1 <?php
   2 /**
   3  * Rebuild search index table from scratch.  This may take several
   4  * hours, depending on the database size and server configuration.
   5  *
   6  * Postgres is trigger-based and should never need rebuilding.
   7  *
   8  * This program is free software; you can redistribute it and/or modify
   9  * it under the terms of the GNU General Public License as published by
  10  * the Free Software Foundation; either version 2 of the License, or
  11  * (at your option) any later version.
  12  *
  13  * This program is distributed in the hope that it will be useful,
  14  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  15  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  16  * GNU General Public License for more details.
  17  *
  18  * You should have received a copy of the GNU General Public License along
  19  * with this program; if not, write to the Free Software Foundation, Inc.,
  20  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  21  * http://www.gnu.org/copyleft/gpl.html
  22  *
  23  * @ingroup Maintenance
  24  * @todo document
  25  */
  26
  27 require_once( dirname( __FILE__ ) . '/Maintenance.php' );
  28
  29 class RebuildTextIndex extends Maintenance {
  30         const RTI_CHUNK_SIZE = 500;
  31         private $db;
  32
  33         public function __construct() {
  34                 parent::__construct();
  35                 $this->mDescription = "Rebuild search index table from scratch";
  36         }
  37
  38         public function getDbType() {
  39                 return Maintenance::DB_ADMIN;
  40         }
  41
  42         public function execute() {
  43                 global $wgTitle, $wgDBtype;
  44
  45                 // Shouldn't be needed for Postgres
  46                 if ( $wgDBtype == 'postgres' ) {
  47                         $this->error( "This script is not needed when using Postgres.\n", true );
  48                 }
  49
  50                 $this->db = wfGetDB( DB_MASTER );
  51                 $wgTitle = Title::newFromText( "Rebuild text index script" );
  52
  53                 if ( $wgDBtype == 'mysql' ) {
  54                         $this->dropMysqlTextIndex();
  55                         $this->populateSearchIndex();
  56                         $this->createMysqlTextIndex();
  57                 } else {
  58                         $this->clearSearchIndex();
  59                         $this->populateSearchIndex();
  60                 }
  61
  62                 $this->output( "Done.\n" );
  63         }
  64
  65         /**
  66          * Populates the search index with content from all pages
  67          */
  68         protected function populateSearchIndex() {
  69                 $res = $this->db->select( 'page', 'MAX(page_id) AS count' );
  70                 $s = $this->db->fetchObject( $res );
  71                 $count = $s->count;
  72                 $this->output( "Rebuilding index fields for {$count} pages...\n" );
  73                 $n = 0;
  74
  75                 while ( $n < $count ) {
  76                         $this->output( $n . "\n" );
  77                         $end = $n + self::RTI_CHUNK_SIZE - 1;
  78
  79                         $res = $this->db->select( array( 'page', 'revision', 'text' ),
  80                                 array( 'page_id', 'page_namespace', 'page_title', 'old_flags', 'old_text' ),
  81                                 array( "page_id BETWEEN $n AND $end", 'page_latest = rev_id', 'rev_text_id = old_id' ),
  82                                 __METHOD__
  83                                 );
  84
  85                         foreach ( $res as $s ) {
  86                                 $revtext = Revision::getRevisionText( $s );
  87                                 $u = new SearchUpdate( $s->page_id, $s->page_title, $revtext );
  88                                 $u->doUpdate();
  89                         }
  90                         $this->db->freeResult( $res );
  91                         $n += self::RTI_CHUNK_SIZE;
  92                 }
  93         }
  94
  95         /**
  96          * (MySQL only) Drops fulltext index before populating the table.
  97          */
  98         private function dropMysqlTextIndex() {
  99                 $searchindex = $this->db->tableName( 'searchindex' );
 100                 if ( $this->db->indexExists( 'searchindex', 'si_title' ) ) {
 101                         $this->output( "Dropping index...\n" );
 102                         $sql = "ALTER TABLE $searchindex DROP INDEX si_title, DROP INDEX si_text";
 103                         $this->db->query( $sql, __METHOD__ );
 104                 }
 105         }
 106
 107         /**
 108          * (MySQL only) Adds back fulltext index after populating the table.
 109          */
 110         private function createMysqlTextIndex() {
 111                 $searchindex = $this->db->tableName( 'searchindex' );
 112                 $this->output( "\nRebuild the index...\n" );
 113                 $sql = "ALTER TABLE $searchindex ADD FULLTEXT si_title (si_title), " .
 114                   "ADD FULLTEXT si_text (si_text)";
 115                 $this->db->query( $sql, __METHOD__ );
 116         }
 117
 118         /**
 119          * Deletes everything from search index.
 120          */
 121         private function clearSearchIndex() {
 122                 $this->output( 'Clearing searchindex table...' );
 123                 $this->db->delete( 'searchindex', '*', __METHOD__ );
 124                 $this->output( "Done\n" );
 125         }
 126 }
 127
 128 $maintClass = "RebuildTextIndex";
 129 require_once( DO_MAINTENANCE );