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