maintenance/updateSearchIndex.php

   1 <?php
   2 /**
   3  * Periodic off-peak updating of the search index.
   4  *
   5  * Usage: php updateSearchIndex.php [-s START] [-e END] [-p POSFILE] [-l LOCKTIME] [-q]
   6  * Where START is the starting timestamp
   7  * END is the ending timestamp
   8  * POSFILE is a file to load timestamps from and save them to, searchUpdate.WIKI_ID.pos by default
   9  * LOCKTIME is how long the searchindex and revision tables will be locked for
  10  * -q means quiet
  11  *
  12  * This program is free software; you can redistribute it and/or modify
  13  * it under the terms of the GNU General Public License as published by
  14  * the Free Software Foundation; either version 2 of the License, or
  15  * (at your option) any later version.
  16  *
  17  * This program is distributed in the hope that it will be useful,
  18  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  19  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  20  * GNU General Public License for more details.
  21  *
  22  * You should have received a copy of the GNU General Public License along
  23  * with this program; if not, write to the Free Software Foundation, Inc.,
  24  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  25  * http://www.gnu.org/copyleft/gpl.html
  26  *
  27  * @file
  28  * @ingroup Maintenance
  29  */
  30
  31 use MediaWiki\MediaWikiServices;
  32 use MediaWiki\Revision\SlotRecord;
  33
  34 require_once __DIR__ . '/Maintenance.php';
  35
  36 /**
  37  * Maintenance script for periodic off-peak updating of the search index.
  38  *
  39  * @ingroup Maintenance
  40  */
  41 class UpdateSearchIndex extends Maintenance {
  42
  43         public function __construct() {
  44                 parent::__construct();
  45                 $this->addDescription( 'Script for periodic off-peak updating of the search index' );
  46                 $this->addOption( 's', 'Starting timestamp', false, true );
  47                 $this->addOption( 'e', 'Ending timestamp', false, true );
  48                 $this->addOption(
  49                         'p',
  50                         'File for saving/loading timestamps, searchUpdate.WIKI_ID.pos by default',
  51                         false,
  52                         true
  53                 );
  54                 $this->addOption(
  55                         'l',
  56                         'Deprecated, has no effect (formerly lock time)',
  57                         false,
  58                         true
  59                 );
  60         }
  61
  62         public function getDbType() {
  63                 return Maintenance::DB_ADMIN;
  64         }
  65
  66         public function execute() {
  67                 $dbDomain = WikiMap::getCurrentWikiDbDomain()->getId();
  68                 $posFile = $this->getOption( 'p', 'searchUpdate.' . rawurlencode( $dbDomain ) . '.pos' );
  69                 $end = $this->getOption( 'e', wfTimestampNow() );
  70                 if ( $this->hasOption( 's' ) ) {
  71                         $start = $this->getOption( 's' );
  72                 } elseif ( is_readable( $posFile ) ) {
  73                         $start = file_get_contents( $posFile );
  74                 } else {
  75                         $start = wfTimestamp( TS_MW, time() - 86400 );
  76                 }
  77
  78                 $this->doUpdateSearchIndex( $start, $end );
  79                 $file = fopen( $posFile, 'w' );
  80                 if ( $file !== false ) {
  81                         fwrite( $file, $end );
  82                         fclose( $file );
  83                 } else {
  84                         $this->error( "*** Couldn't write to the $posFile!\n" );
  85                 }
  86         }
  87
  88         private function doUpdateSearchIndex( $start, $end ) {
  89                 global $wgDisableSearchUpdate;
  90
  91                 $wgDisableSearchUpdate = false;
  92
  93                 $dbw = $this->getDB( DB_PRIMARY );
  94
  95                 $this->output( "Updating searchindex between $start and $end\n" );
  96
  97                 # Select entries from recentchanges which are on top and between the specified times
  98                 $start = $dbw->timestamp( $start );
  99                 $end = $dbw->timestamp( $end );
 100
 101                 $res = $dbw->select(
 102                         [ 'recentchanges', 'page' ],
 103                         'rc_cur_id',
 104                         [
 105                                 'rc_type != ' . $dbw->addQuotes( RC_LOG ),
 106                                 'rc_timestamp BETWEEN ' . $dbw->addQuotes( $start ) . ' AND ' . $dbw->addQuotes( $end )
 107                         ],
 108                         __METHOD__,
 109                         [],
 110                         [
 111                                 'page' => [ 'JOIN', 'rc_cur_id=page_id AND rc_this_oldid=page_latest' ]
 112                         ]
 113                 );
 114
 115                 foreach ( $res as $row ) {
 116                         $this->updateSearchIndexForPage( (int)$row->rc_cur_id );
 117                 }
 118                 $this->output( "Done\n" );
 119         }
 120
 121         /**
 122          * Update the searchindex table for a given pageid
 123          * @param int $pageId The page ID to update.
 124          * @return null|string
 125          */
 126         private function updateSearchIndexForPage( int $pageId ) {
 127                 // Get current revision
 128                 $rev = MediaWikiServices::getInstance()
 129                         ->getRevisionLookup()
 130                         ->getRevisionByPageId( $pageId, 0, IDBAccessObject::READ_LATEST );
 131                 $title = null;
 132                 if ( $rev ) {
 133                         $titleObj = Title::newFromLinkTarget( $rev->getPageAsLinkTarget() );
 134                         $title = $titleObj->getPrefixedDBkey();
 135                         $this->output( "$title..." );
 136                         # Update searchindex
 137                         $u = new SearchUpdate( $pageId, $titleObj, $rev->getContent( SlotRecord::MAIN ) );
 138                         $u->doUpdate();
 139                         $this->output( "\n" );
 140                 }
 141
 142                 return $title;
 143         }
 144 }
 145
 146 $maintClass = UpdateSearchIndex::class;
 147 require_once RUN_MAINTENANCE_IF_MAIN;