includes/jobqueue/jobs/HTMLCacheUpdateJob.php

   1 <?php
   2 /**
   3  * HTML cache invalidation of all pages linking to 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  * @ingroup Cache
  23  */
  24
  25 /**
  26  * Job to purge the cache for all pages that link to or use another page or file
  27  *
  28  * This job comes in a few variants:
  29  *   - a) Recursive jobs to purge caches for backlink pages for a given title.
  30  *        These jobs have (recursive:true,table:<table>) set.
  31  *   - b) Jobs to purge caches for a set of titles (the job title is ignored).
  32  *            These jobs have (pages:(<page ID>:(<namespace>,<title>),...) set.
  33  *
  34  * @ingroup JobQueue
  35  */
  36 class HTMLCacheUpdateJob extends Job {
  37         function __construct( Title $title, array $params ) {
  38                 parent::__construct( 'htmlCacheUpdate', $title, $params );
  39                 // Base backlink purge jobs can be de-duplicated
  40                 $this->removeDuplicates = ( !isset( $params['range'] ) && !isset( $params['pages'] ) );
  41         }
  42
  43         function run() {
  44                 global $wgUpdateRowsPerJob, $wgUpdateRowsPerQuery;
  45
  46                 if ( isset( $this->params['table'] ) && !isset( $this->params['pages'] ) ) {
  47                         $this->params['recursive'] = true; // b/c; base job
  48                 }
  49
  50                 // Job to purge all (or a range of) backlink pages for a page
  51                 if ( !empty( $this->params['recursive'] ) ) {
  52                         // Convert this into no more than $wgUpdateRowsPerJob HTMLCacheUpdateJob per-title
  53                         // jobs and possibly a recursive HTMLCacheUpdateJob job for the rest of the backlinks
  54                         $jobs = BacklinkJobUtils::partitionBacklinkJob(
  55                                 $this,
  56                                 $wgUpdateRowsPerJob,
  57                                 $wgUpdateRowsPerQuery, // jobs-per-title
  58                                 // Carry over information for de-duplication
  59                                 array( 'params' => $this->getRootJobParams() )
  60                         );
  61                         JobQueueGroup::singleton()->push( $jobs );
  62                 // Job to purge pages for a set of titles
  63                 } elseif ( isset( $this->params['pages'] ) ) {
  64                         $this->invalidateTitles( $this->params['pages'] );
  65                 // Job to update a single title
  66                 } else {
  67                         $t = $this->title;
  68                         $this->invalidateTitles( array(
  69                                 $t->getArticleID() => array( $t->getNamespace(), $t->getDBkey() )
  70                         ) );
  71                 }
  72
  73                 return true;
  74         }
  75
  76         /**
  77          * @param array $pages Map of (page ID => (namespace, DB key)) entries
  78          */
  79         protected function invalidateTitles( array $pages ) {
  80                 global $wgUpdateRowsPerQuery, $wgUseFileCache, $wgUseSquid;
  81
  82                 // Get all page IDs in this query into an array
  83                 $pageIds = array_keys( $pages );
  84                 if ( !$pageIds ) {
  85                         return;
  86                 }
  87
  88                 $dbw = wfGetDB( DB_MASTER );
  89
  90                 // The page_touched field will need to be bumped for these pages.
  91                 // Only bump it to the present time if no "rootJobTimestamp" was known.
  92                 // If it is known, it can be used instead, which avoids invalidating output
  93                 // that was in fact generated *after* the relevant dependency change time
  94                 // (e.g. template edit). This is particularily useful since refreshLinks jobs
  95                 // save back parser output and usually run along side htmlCacheUpdate jobs;
  96                 // their saved output would be invalidated by using the current timestamp.
  97                 if ( isset( $this->params['rootJobTimestamp'] ) ) {
  98                         $touchTimestamp = $this->params['rootJobTimestamp'];
  99                 } else {
 100                         $touchTimestamp = wfTimestampNow();
 101                 }
 102
 103                 // Update page_touched (skipping pages already touched since the root job).
 104                 // Check $wgUpdateRowsPerQuery for sanity; batch jobs are sized by that already.
 105                 foreach ( array_chunk( $pageIds, $wgUpdateRowsPerQuery ) as $batch ) {
 106                         $dbw->update( 'page',
 107                                 array( 'page_touched' => $dbw->timestamp( $touchTimestamp ) ),
 108                                 array( 'page_id' => $batch,
 109                                         // don't invalidated pages that were already invalidated
 110                                         "page_touched < " . $dbw->addQuotes( $dbw->timestamp( $touchTimestamp ) )
 111                                 ),
 112                                 __METHOD__
 113                         );
 114                 }
 115                 // Get the list of affected pages (races only mean something else did the purge)
 116                 $titleArray = TitleArray::newFromResult( $dbw->select(
 117                         'page',
 118                         array( 'page_namespace', 'page_title' ),
 119                         array( 'page_id' => $pageIds, 'page_touched' => $dbw->timestamp( $touchTimestamp ) ),
 120                         __METHOD__
 121                 ) );
 122
 123                 // Update squid
 124                 if ( $wgUseSquid ) {
 125                         $u = SquidUpdate::newFromTitles( $titleArray );
 126                         $u->doUpdate();
 127                 }
 128
 129                 // Update file cache
 130                 if ( $wgUseFileCache ) {
 131                         foreach ( $titleArray as $title ) {
 132                                 HTMLFileCache::clearFileCache( $title );
 133                         }
 134                 }
 135         }
 136
 137         public function workItemCount() {
 138                 return isset( $this->params['pages'] ) ? count( $this->params['pages'] ) : 1;
 139         }
 140 }