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