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;
  44
  45                 // Shouldn't be needed for Postgres
  46                 $this->db = wfGetDB( DB_MASTER );
  47                 if ( $this->db->getType() == 'postgres' ) {
  48                         $this->error( "This script is not needed when using Postgres.\n", true );
  49                 }
  50
  51                 $this->db = wfGetDB( DB_MASTER );
  52                 if ( $this->db->getType() == 'sqlite' ) {
  53                         if ( !$this->db->getFulltextSearchModule() ) {
  54                                 $this->error( "Your version of SQLite module for PHP doesn't support full-text search (FTS3).\n" );
  55                         }
  56                         if ( !$this->db->checkForEnabledSearch() ) {
  57                                 $this->error( "Your database schema is not configured for full-text search support. Run update.php.\n" );
  58                         }
  59                 }
  60
  61                 $wgTitle = Title::newFromText( "Rebuild text index script" );
  62
  63                 if ( $this->db->getType() == 'mysql' ) {
  64                         $this->dropMysqlTextIndex();
  65                         $this->populateSearchIndex();
  66                         $this->createMysqlTextIndex();
  67                 } else {
  68                         $this->clearSearchIndex();
  69                         $this->populateSearchIndex();
  70                 }
  71
  72                 $this->output( "Done.\n" );
  73         }
  74
  75         /**
  76          * Populates the search index with content from all pages
  77          */
  78         protected function populateSearchIndex() {
  79                 $res = $this->db->select( 'page', 'MAX(page_id) AS count' );
  80                 $s = $this->db->fetchObject( $res );
  81                 $count = $s->count;
  82                 $this->output( "Rebuilding index fields for {$count} pages...\n" );
  83                 $n = 0;
  84
  85                 while ( $n < $count ) {
  86                         if ( $n ) {
  87                                 $this->output( $n . "\n" );
  88                         }
  89                         $end = $n + self::RTI_CHUNK_SIZE - 1;
  90
  91                         $res = $this->db->select( array( 'page', 'revision', 'text' ),
  92                                 array( 'page_id', 'page_namespace', 'page_title', 'old_flags', 'old_text' ),
  93                                 array( "page_id BETWEEN $n AND $end", 'page_latest = rev_id', 'rev_text_id = old_id' ),
  94                                 __METHOD__
  95                                 );
  96
  97                         foreach ( $res as $s ) {
  98                                 $revtext = Revision::getRevisionText( $s );
  99                                 $u = new SearchUpdate( $s->page_id, $s->page_title, $revtext );
 100                                 $u->doUpdate();
 101                         }
 102                         $n += self::RTI_CHUNK_SIZE;
 103                 }
 104         }
 105
 106         /**
 107          * (MySQL only) Drops fulltext index before populating the table.
 108          */
 109         private function dropMysqlTextIndex() {
 110                 $searchindex = $this->db->tableName( 'searchindex' );
 111                 if ( $this->db->indexExists( 'searchindex', 'si_title' ) ) {
 112                         $this->output( "Dropping index...\n" );
 113                         $sql = "ALTER TABLE $searchindex DROP INDEX si_title, DROP INDEX si_text";
 114                         $this->db->query( $sql, __METHOD__ );
 115                 }
 116         }
 117
 118         /**
 119          * (MySQL only) Adds back fulltext index after populating the table.
 120          */
 121         private function createMysqlTextIndex() {
 122                 $searchindex = $this->db->tableName( 'searchindex' );
 123                 $this->output( "\nRebuild the index...\n" );
 124                 $sql = "ALTER TABLE $searchindex ADD FULLTEXT si_title (si_title), " .
 125                   "ADD FULLTEXT si_text (si_text)";
 126                 $this->db->query( $sql, __METHOD__ );
 127         }
 128
 129         /**
 130          * Deletes everything from search index.
 131          */
 132         private function clearSearchIndex() {
 133                 $this->output( 'Clearing searchindex table...' );
 134                 $this->db->delete( 'searchindex', '*', __METHOD__ );
 135                 $this->output( "Done\n" );
 136         }
 137 }
 138
 139 $maintClass = "RebuildTextIndex";
 140 require_once( RUN_MAINTENANCE_IF_MAIN );