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  * @file
  24  * @ingroup Maintenance
  25  * @todo document
  26  */
  27
  28 require_once( __DIR__ . '/Maintenance.php' );
  29
  30 /**
  31  * Maintenance script that rebuilds search index table from scratch.
  32  *
  33  * @ingroup Maintenance
  34  */
  35 class RebuildTextIndex extends Maintenance {
  36         const RTI_CHUNK_SIZE = 500;
  37
  38         /**
  39          * @var DatabaseBase
  40          */
  41         private $db;
  42
  43         public function __construct() {
  44                 parent::__construct();
  45                 $this->mDescription = "Rebuild search index table from scratch";
  46         }
  47
  48         public function getDbType() {
  49                 return Maintenance::DB_ADMIN;
  50         }
  51
  52         public function execute() {
  53                 global $wgTitle;
  54
  55                 // Shouldn't be needed for Postgres
  56                 $this->db = wfGetDB( DB_MASTER );
  57                 if ( $this->db->getType() == 'postgres' ) {
  58                         $this->error( "This script is not needed when using Postgres.\n", true );
  59                 }
  60
  61                 $this->db = wfGetDB( DB_MASTER );
  62                 if ( $this->db->getType() == 'sqlite' ) {
  63                         if ( !DatabaseSqlite::getFulltextSearchModule() ) {
  64                                 $this->error( "Your version of SQLite module for PHP doesn't support full-text search (FTS3).\n", true );
  65                         }
  66                         if ( !$this->db->checkForEnabledSearch() ) {
  67                                 $this->error( "Your database schema is not configured for full-text search support. Run update.php.\n", true );
  68                         }
  69                 }
  70
  71                 $wgTitle = Title::newFromText( "Rebuild text index script" );
  72
  73                 if ( $this->db->getType() == 'mysql' ) {
  74                         $this->dropMysqlTextIndex();
  75                         $this->populateSearchIndex();
  76                         $this->createMysqlTextIndex();
  77                 } else {
  78                         $this->clearSearchIndex();
  79                         $this->populateSearchIndex();
  80                 }
  81
  82                 $this->output( "Done.\n" );
  83         }
  84
  85         /**
  86          * Populates the search index with content from all pages
  87          */
  88         protected function populateSearchIndex() {
  89                 $res = $this->db->select( 'page', 'MAX(page_id) AS count' );
  90                 $s = $this->db->fetchObject( $res );
  91                 $count = $s->count;
  92                 $this->output( "Rebuilding index fields for {$count} pages...\n" );
  93                 $n = 0;
  94
  95                 $fields = array_merge(
  96                         Revision::selectPageFields(),
  97                         Revision::selectFields(),
  98                         Revision::selectTextFields()
  99                 );
 100
 101                 while ( $n < $count ) {
 102                         if ( $n ) {
 103                                 $this->output( $n . "\n" );
 104                         }
 105                         $end = $n + self::RTI_CHUNK_SIZE - 1;
 106
 107                         $res = $this->db->select( array( 'page', 'revision', 'text' ), $fields,
 108                                 array( "page_id BETWEEN $n AND $end", 'page_latest = rev_id', 'rev_text_id = old_id' ),
 109                                 __METHOD__
 110                         );
 111
 112                         foreach ( $res as $s ) {
 113                                 try {
 114                                         $title = Title::makeTitle( $s->page_namespace, $s->page_title );
 115
 116                                         $rev = new Revision( $s );
 117                                         $content = $rev->getContent();
 118                                         $text = $content->getTextForSearchIndex();
 119
 120                                         $u = new SearchUpdate( $s->page_id, $title, $text );
 121                                         $u->doUpdate();
 122                                 } catch ( MWContentSerializationException $ex ) {
 123                                         $this->output( "Failed to deserialize content of revision {$s->rev_id} of page "
 124                                                 . "`" . $title->getPrefixedDBkey() . "`!\n" );
 125                                 }
 126                         }
 127                         $n += self::RTI_CHUNK_SIZE;
 128                 }
 129         }
 130
 131         /**
 132          * (MySQL only) Drops fulltext index before populating the table.
 133          */
 134         private function dropMysqlTextIndex() {
 135                 $searchindex = $this->db->tableName( 'searchindex' );
 136                 if ( $this->db->indexExists( 'searchindex', 'si_title', __METHOD__ ) ) {
 137                         $this->output( "Dropping index...\n" );
 138                         $sql = "ALTER TABLE $searchindex DROP INDEX si_title, DROP INDEX si_text";
 139                         $this->db->query( $sql, __METHOD__ );
 140                 }
 141         }
 142
 143         /**
 144          * (MySQL only) Adds back fulltext index after populating the table.
 145          */
 146         private function createMysqlTextIndex() {
 147                 $searchindex = $this->db->tableName( 'searchindex' );
 148                 $this->output( "\nRebuild the index...\n" );
 149                 $sql = "ALTER TABLE $searchindex ADD FULLTEXT si_title (si_title), " .
 150                         "ADD FULLTEXT si_text (si_text)";
 151                 $this->db->query( $sql, __METHOD__ );
 152         }
 153
 154         /**
 155          * Deletes everything from search index.
 156          */
 157         private function clearSearchIndex() {
 158                 $this->output( 'Clearing searchindex table...' );
 159                 $this->db->delete( 'searchindex', '*', __METHOD__ );
 160                 $this->output( "Done\n" );
 161         }
 162 }
 163
 164 $maintClass = "RebuildTextIndex";
 165 require_once( RUN_MAINTENANCE_IF_MAIN );