includes/jobqueue/jobs/RefreshLinksJob2.php

   1 <?php
   2 /**
   3  * Job to update links for a given title.
   4  *
   5  * This program is free software; you can redistribute it and/or modify
   6  * it under the terms of the GNU General Public License as published by
   7  * the Free Software Foundation; either version 2 of the License, or
   8  * (at your option) any later version.
   9  *
  10  * This program is distributed in the hope that it will be useful,
  11  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  12  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  13  * GNU General Public License for more details.
  14  *
  15  * You should have received a copy of the GNU General Public License along
  16  * with this program; if not, write to the Free Software Foundation, Inc.,
  17  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  18  * http://www.gnu.org/copyleft/gpl.html
  19  *
  20  * @file
  21  * @ingroup JobQueue
  22  */
  23
  24 /**
  25  * Background job to update links for titles in certain backlink range by page ID.
  26  * Newer version for high use templates. This is deprecated by RefreshLinksPartitionJob.
  27  *
  28  * @ingroup JobQueue
  29  * @deprecated since 1.23
  30  */
  31 class RefreshLinksJob2 extends Job {
  32         function __construct( $title, $params ) {
  33                 parent::__construct( 'refreshLinks2', $title, $params );
  34                 // Base jobs for large templates can easily be de-duplicated
  35                 $this->removeDuplicates = !isset( $params['start'] ) && !isset( $params['end'] );
  36         }
  37
  38         /**
  39          * Run a refreshLinks2 job
  40          * @return bool Success
  41          */
  42         function run() {
  43                 global $wgUpdateRowsPerJob;
  44
  45                 $linkCache = LinkCache::singleton();
  46                 $linkCache->clear();
  47
  48                 if ( is_null( $this->title ) ) {
  49                         $this->error = "refreshLinks2: Invalid title";
  50                         return false;
  51                 }
  52
  53                 // Back compat for pre-r94435 jobs
  54                 $table = isset( $this->params['table'] ) ? $this->params['table'] : 'templatelinks';
  55
  56                 // Avoid slave lag when fetching templates.
  57                 // When the outermost job is run, we know that the caller that enqueued it must have
  58                 // committed the relevant changes to the DB by now. At that point, record the master
  59                 // position and pass it along as the job recursively breaks into smaller range jobs.
  60                 // Hopefully, when leaf jobs are popped, the slaves will have reached that position.
  61                 if ( isset( $this->params['masterPos'] ) ) {
  62                         $masterPos = $this->params['masterPos'];
  63                 } elseif ( wfGetLB()->getServerCount() > 1 ) {
  64                         $masterPos = wfGetLB()->getMasterPos();
  65                 } else {
  66                         $masterPos = false;
  67                 }
  68
  69                 $tbc = $this->title->getBacklinkCache();
  70
  71                 $jobs = array(); // jobs to insert
  72                 if ( isset( $this->params['start'] ) && isset( $this->params['end'] ) ) {
  73                         # This is a partition job to trigger the insertion of leaf jobs...
  74                         $jobs = array_merge( $jobs, $this->getSingleTitleJobs( $table, $masterPos ) );
  75                 } else {
  76                         # This is a base job to trigger the insertion of partitioned jobs...
  77                         if ( $tbc->getNumLinks( $table, $wgUpdateRowsPerJob + 1 ) <= $wgUpdateRowsPerJob ) {
  78                                 # Just directly insert the single per-title jobs
  79                                 $jobs = array_merge( $jobs, $this->getSingleTitleJobs( $table, $masterPos ) );
  80                         } else {
  81                                 # Insert the partition jobs to make per-title jobs
  82                                 foreach ( $tbc->partition( $table, $wgUpdateRowsPerJob ) as $batch ) {
  83                                         list( $start, $end ) = $batch;
  84                                         $jobs[] = new RefreshLinksJob2( $this->title,
  85                                                 array(
  86                                                         'table' => $table,
  87                                                         'start' => $start,
  88                                                         'end' => $end,
  89                                                         'masterPos' => $masterPos,
  90                                                 ) + $this->getRootJobParams() // carry over information for de-duplication
  91                                         );
  92                                 }
  93                         }
  94                 }
  95
  96                 if ( count( $jobs ) ) {
  97                         JobQueueGroup::singleton()->push( $jobs );
  98                 }
  99
 100                 return true;
 101         }
 102
 103         /**
 104          * @param string $table
 105          * @param mixed $masterPos
 106          * @return array
 107          */
 108         protected function getSingleTitleJobs( $table, $masterPos ) {
 109                 # The "start"/"end" fields are not set for the base jobs
 110                 $start = isset( $this->params['start'] ) ? $this->params['start'] : false;
 111                 $end = isset( $this->params['end'] ) ? $this->params['end'] : false;
 112                 $titles = $this->title->getBacklinkCache()->getLinks( $table, $start, $end );
 113                 # Convert into single page refresh links jobs.
 114                 # This handles well when in sapi mode and is useful in any case for job
 115                 # de-duplication. If many pages use template A, and that template itself
 116                 # uses template B, then an edit to both will create many duplicate jobs.
 117                 # Roughly speaking, for each page, one of the "RefreshLinksJob" jobs will
 118                 # get run first, and when it does, it will remove the duplicates. Of course,
 119                 # one page could have its job popped when the other page's job is still
 120                 # buried within the logic of a refreshLinks2 job.
 121                 $jobs = array();
 122                 foreach ( $titles as $title ) {
 123                         $jobs[] = new RefreshLinksJob( $title,
 124                                 array( 'masterPos' => $masterPos ) + $this->getRootJobParams()
 125                         ); // carry over information for de-duplication
 126                 }
 127                 return $jobs;
 128         }
 129
 130         /**
 131          * @return array
 132          */
 133         public function getDeduplicationInfo() {
 134                 $info = parent::getDeduplicationInfo();
 135                 // Don't let highly unique "masterPos" values ruin duplicate detection
 136                 if ( is_array( $info['params'] ) ) {
 137                         unset( $info['params']['masterPos'] );
 138                 }
 139                 return $info;
 140         }
 141 }