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