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