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