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 require_once( __DIR__ . '/Maintenance.php' );
  32
  33 /**
  34  * Maintenance script for periodic off-peak updating of the search index.
  35  *
  36  * @ingroup Maintenance
  37  */
  38 class UpdateSearchIndex extends Maintenance {
  39
  40         public function __construct() {
  41                 parent::__construct();
  42                 $this->mDescription = "Script for periodic off-peak updating of the search index";
  43                 $this->addOption( 's', 'starting timestamp', false, true );
  44                 $this->addOption( 'e', 'Ending timestamp', false, true );
  45                 $this->addOption( 'p', 'File for saving/loading timestamps, searchUpdate.WIKI_ID.pos by default', false, true );
  46                 $this->addOption( 'l', 'How long the searchindex and revision tables will be locked for', false, true );
  47         }
  48
  49         public function getDbType() {
  50                 return Maintenance::DB_ADMIN;
  51         }
  52
  53         public function execute() {
  54                 $posFile = $this->getOption( 'p', 'searchUpdate.' . wfWikiId() . '.pos' );
  55                 $end = $this->getOption( 'e', wfTimestampNow() );
  56                 if ( $this->hasOption( 's' ) ) {
  57                         $start = $this->getOption( 's' );
  58                 } elseif ( is_readable( 'searchUpdate.pos' ) ) {
  59                         # B/c to the old position file name which was hardcoded
  60                         # We can safely delete the file when we're done though.
  61                         $start = file_get_contents( 'searchUpdate.pos' );
  62                         unlink( 'searchUpdate.pos' );
  63                 } elseif( is_readable( $posFile ) ) {
  64                         $start = file_get_contents( $posFile );
  65                 } else {
  66                         $start = wfTimestamp( TS_MW, time() - 86400 );
  67                 }
  68                 $lockTime = $this->getOption( 'l', 20 );
  69
  70                 $this->doUpdateSearchIndex( $start, $end, $lockTime );
  71                 if ( is_writable( dirname( realpath( $posFile ) ) ) ) {
  72                         $file = fopen( $posFile, 'w' );
  73                         if ( $file !== false ) {
  74                                 fwrite( $file, $end );
  75                                 fclose( $file );
  76                         } else {
  77                                 $this->error( "*** Couldn't write to the $posFile!\n" );
  78                         }
  79                 } else {
  80                         $this->error( "*** Couldn't write to the $posFile!\n" );
  81                 }
  82         }
  83
  84         private function doUpdateSearchIndex( $start, $end, $maxLockTime ) {
  85                 global $wgDisableSearchUpdate;
  86
  87                 $wgDisableSearchUpdate = false;
  88
  89                 $dbw = wfGetDB( DB_MASTER );
  90                 $recentchanges = $dbw->tableName( 'recentchanges' );
  91
  92                 $this->output( "Updating searchindex between $start and $end\n" );
  93
  94                 # Select entries from recentchanges which are on top and between the specified times
  95                 $start = $dbw->timestamp( $start );
  96                 $end = $dbw->timestamp( $end );
  97
  98                 $page = $dbw->tableName( 'page' );
  99                 $sql = "SELECT rc_cur_id FROM $recentchanges
 100                   JOIN $page ON rc_cur_id=page_id AND rc_this_oldid=page_latest
 101                   WHERE rc_type != " . RC_LOG . " AND rc_timestamp BETWEEN '$start' AND '$end'
 102                   ";
 103                 $res = $dbw->query( $sql, __METHOD__ );
 104
 105                 $this->updateSearchIndex( $maxLockTime, array( $this, 'searchIndexUpdateCallback' ), $dbw, $res );
 106
 107                 $this->output( "Done\n" );
 108         }
 109
 110         public function searchIndexUpdateCallback( $dbw, $row ) {
 111                 $this->updateSearchIndexForPage( $dbw, $row->rc_cur_id );
 112         }
 113 }
 114
 115 $maintClass = "UpdateSearchIndex";
 116 require_once( RUN_MAINTENANCE_IF_MAIN );