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 Cache
  22  */
  23
  24 /**
  25  * Job to purge the cache for all pages that link to or use another page or file
  26  *
  27  * This job comes in a few variants:
  28  *   - a) Recursive jobs to purge caches for backlink pages for a given title.
  29  *        These jobs have have (recursive:true,table:<table>) set.
  30  *   - b) Jobs to purge caches for a set of titles (the job title is ignored).
  31  *            These jobs have have (pages:(<page ID>:(<namespace>,<title>),...) set.
  32  *
  33  * @ingroup JobQueue
  34  */
  35 class HTMLCacheUpdateJob extends Job {
  36         function __construct( $title, $params = '' ) {
  37                 parent::__construct( 'htmlCacheUpdate', $title, $params );
  38                 // Base backlink purge jobs can be de-duplicated
  39                 $this->removeDuplicates = ( !isset( $params['range'] ) && !isset( $params['pages'] ) );
  40         }
  41
  42         function run() {
  43                 global $wgUpdateRowsPerJob, $wgUpdateRowsPerQuery, $wgMaxBacklinksInvalidate;
  44
  45                 static $expected = array( 'recursive', 'pages' ); // new jobs have one of these
  46
  47                 $oldRangeJob = false;
  48                 if ( !array_intersect( array_keys( $this->params ), $expected ) ) {
  49                         // B/C for older job params formats that lack these fields:
  50                         // a) base jobs with just ("table") and b) range jobs with ("table","start","end")
  51                         if ( isset( $this->params['start'] ) && isset( $this->params['end'] ) ) {
  52                                 $oldRangeJob = true;
  53                         } else {
  54                                 $this->params['recursive'] = true; // base job
  55                         }
  56                 }
  57
  58                 // Job to purge all (or a range of) backlink pages for a page
  59                 if ( !empty( $this->params['recursive'] ) ) {
  60                         // @TODO: try to use delayed jobs if possible?
  61                         if ( !isset( $this->params['range'] ) && $wgMaxBacklinksInvalidate !== false ) {
  62                                 $numRows = $this->title->getBacklinkCache()->getNumLinks(
  63                                         $this->params['table'], $wgMaxBacklinksInvalidate );
  64                                 if ( $numRows > $wgMaxBacklinksInvalidate ) {
  65                                         return true;
  66                                 }
  67                         }
  68                         // Convert this into no more than $wgUpdateRowsPerJob HTMLCacheUpdateJob per-title
  69                         // jobs and possibly a recursive HTMLCacheUpdateJob job for the rest of the backlinks
  70                         $jobs = BacklinkJobUtils::partitionBacklinkJob(
  71                                 $this,
  72                                 $wgUpdateRowsPerJob,
  73                                 $wgUpdateRowsPerQuery, // jobs-per-title
  74                                 // Carry over information for de-duplication
  75                                 array( 'params' => $this->getRootJobParams() )
  76                         );
  77                         JobQueueGroup::singleton()->push( $jobs );
  78                 // Job to purge pages for for a set of titles
  79                 } elseif ( isset( $this->params['pages'] ) ) {
  80                         $this->invalidateTitles( $this->params['pages'] );
  81                 // B/C for job to purge a range of backlink pages for a given page
  82                 } elseif ( $oldRangeJob ) {
  83                         $titleArray = $this->title->getBacklinkCache()->getLinks(
  84                                 $this->params['table'], $this->params['start'], $this->params['end'] );
  85
  86                         $pages = array(); // same format BacklinkJobUtils uses
  87                         foreach ( $titleArray as $tl ) {
  88                                 $pages[$tl->getArticleId()] = array( $tl->getNamespace(), $tl->getDbKey() );
  89                         }
  90
  91                         $jobs = array();
  92                         foreach ( array_chunk( $pages, $wgUpdateRowsPerJob ) as $pageChunk ) {
  93                                 $jobs[] = new HTMLCacheUpdateJob( $this->title,
  94                                         array(
  95                                                 'table' => $this->params['table'],
  96                                                 'pages' => $pageChunk
  97                                         ) + $this->getRootJobParams() // carry over information for de-duplication
  98                                 );
  99                         }
 100                         JobQueueGroup::singleton()->push( $jobs );
 101                 }
 102
 103                 return true;
 104         }
 105
 106         /**
 107          * @param array $pages Map of (page ID => (namespace, DB key)) entries
 108          */
 109         protected function invalidateTitles( array $pages ) {
 110                 global $wgUpdateRowsPerQuery, $wgUseFileCache, $wgUseSquid;
 111
 112                 // Get all page IDs in this query into an array
 113                 $pageIds = array_keys( $pages );
 114                 if ( !$pageIds ) {
 115                         return;
 116                 }
 117
 118                 $dbw = wfGetDB( DB_MASTER );
 119
 120                 // The page_touched field will need to be bumped for these pages.
 121                 // Only bump it to the present time if no "rootJobTimestamp" was known.
 122                 // If it is known, it can be used instead, which avoids invalidating output
 123                 // that was in fact generated *after* the relevant dependency change time
 124                 // (e.g. template edit). This is particularily useful since refreshLinks jobs
 125                 // save back parser output and usually run along side htmlCacheUpdate jobs;
 126                 // their saved output would be invalidated by using the current timestamp.
 127                 if ( isset( $this->params['rootJobTimestamp'] ) ) {
 128                         $touchTimestamp = $this->params['rootJobTimestamp'];
 129                 } else {
 130                         $touchTimestamp = wfTimestampNow();
 131                 }
 132
 133                 // Update page_touched (skipping pages already touched since the root job).
 134                 // Check $wgUpdateRowsPerQuery for sanity; batch jobs are sized by that already.
 135                 foreach ( array_chunk( $pageIds, $wgUpdateRowsPerQuery ) as $batch ) {
 136                         $dbw->update( 'page',
 137                                 array( 'page_touched' => $dbw->timestamp( $touchTimestamp ) ),
 138                                 array( 'page_id' => $batch,
 139                                         // don't invalidated pages that were already invalidated
 140                                         "page_touched < " . $dbw->addQuotes( $dbw->timestamp( $touchTimestamp ) )
 141                                 ),
 142                                 __METHOD__
 143                         );
 144                 }
 145                 // Get the list of affected pages (races only mean something else did the purge)
 146                 $titleArray = TitleArray::newFromResult( $dbw->select(
 147                         'page',
 148                         array( 'page_namespace', 'page_title' ),
 149                         array( 'page_id' => $pageIds, 'page_touched' => $dbw->timestamp( $touchTimestamp ) ),
 150                         __METHOD__
 151                 ) );
 152
 153                 // Update squid
 154                 if ( $wgUseSquid ) {
 155                         $u = SquidUpdate::newFromTitles( $titleArray );
 156                         $u->doUpdate();
 157                 }
 158
 159                 // Update file cache
 160                 if ( $wgUseFileCache ) {
 161                         foreach ( $titleArray as $title ) {
 162                                 HTMLFileCache::clearFileCache( $title );
 163                         }
 164                 }
 165         }
 166
 167         public function workItemCount() {
 168                 return isset( $this->params['pages'] ) ? count( $this->params['pages'] ) : 1;
 169         }
 170 }