summaryrefslogtreecommitdiff
path: root/includes/jobqueue/jobs/HTMLCacheUpdateJob.php
blob: a9010c25d17a073337e82f15c4d2374994d3f763 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
<?php
/**
 * HTML cache invalidation of all pages linking to a given title.
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation; either version 2 of the License, or
 * (at your option) any later version.
 *
 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
 * GNU General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License along
 * with this program; if not, write to the Free Software Foundation, Inc.,
 * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
 * http://www.gnu.org/copyleft/gpl.html
 *
 * @file
 * @ingroup JobQueue
 * @ingroup Cache
 */

/**
 * Job to purge the cache for all pages that link to or use another page or file
 *
 * This job comes in a few variants:
 *   - a) Recursive jobs to purge caches for backlink pages for a given title.
 *        These jobs have (recursive:true,table:<table>) set.
 *   - b) Jobs to purge caches for a set of titles (the job title is ignored).
 *	      These jobs have (pages:(<page ID>:(<namespace>,<title>),...) set.
 *
 * @ingroup JobQueue
 */
class HTMLCacheUpdateJob extends Job {
	function __construct( Title $title, array $params ) {
		parent::__construct( 'htmlCacheUpdate', $title, $params );
		// Base backlink purge jobs can be de-duplicated
		$this->removeDuplicates = ( !isset( $params['range'] ) && !isset( $params['pages'] ) );
	}

	function run() {
		global $wgUpdateRowsPerJob, $wgUpdateRowsPerQuery;

		if ( isset( $this->params['table'] ) && !isset( $this->params['pages'] ) ) {
			$this->params['recursive'] = true; // b/c; base job
		}

		// Job to purge all (or a range of) backlink pages for a page
		if ( !empty( $this->params['recursive'] ) ) {
			// Convert this into no more than $wgUpdateRowsPerJob HTMLCacheUpdateJob per-title
			// jobs and possibly a recursive HTMLCacheUpdateJob job for the rest of the backlinks
			$jobs = BacklinkJobUtils::partitionBacklinkJob(
				$this,
				$wgUpdateRowsPerJob,
				$wgUpdateRowsPerQuery, // jobs-per-title
				// Carry over information for de-duplication
				array( 'params' => $this->getRootJobParams() )
			);
			JobQueueGroup::singleton()->push( $jobs );
		// Job to purge pages for a set of titles
		} elseif ( isset( $this->params['pages'] ) ) {
			$this->invalidateTitles( $this->params['pages'] );
		// Job to update a single title
		} else {
			$t = $this->title;
			$this->invalidateTitles( array(
				$t->getArticleID() => array( $t->getNamespace(), $t->getDBkey() )
			) );
		}

		return true;
	}

	/**
	 * @param array $pages Map of (page ID => (namespace, DB key)) entries
	 */
	protected function invalidateTitles( array $pages ) {
		global $wgUpdateRowsPerQuery, $wgUseFileCache, $wgUseSquid;

		// Get all page IDs in this query into an array
		$pageIds = array_keys( $pages );
		if ( !$pageIds ) {
			return;
		}

		$dbw = wfGetDB( DB_MASTER );

		// The page_touched field will need to be bumped for these pages.
		// Only bump it to the present time if no "rootJobTimestamp" was known.
		// If it is known, it can be used instead, which avoids invalidating output
		// that was in fact generated *after* the relevant dependency change time
		// (e.g. template edit). This is particularily useful since refreshLinks jobs
		// save back parser output and usually run along side htmlCacheUpdate jobs;
		// their saved output would be invalidated by using the current timestamp.
		if ( isset( $this->params['rootJobTimestamp'] ) ) {
			$touchTimestamp = $this->params['rootJobTimestamp'];
		} else {
			$touchTimestamp = wfTimestampNow();
		}

		// Update page_touched (skipping pages already touched since the root job).
		// Check $wgUpdateRowsPerQuery for sanity; batch jobs are sized by that already.
		foreach ( array_chunk( $pageIds, $wgUpdateRowsPerQuery ) as $batch ) {
			$dbw->update( 'page',
				array( 'page_touched' => $dbw->timestamp( $touchTimestamp ) ),
				array( 'page_id' => $batch,
					// don't invalidated pages that were already invalidated
					"page_touched < " . $dbw->addQuotes( $dbw->timestamp( $touchTimestamp ) )
				),
				__METHOD__
			);
		}
		// Get the list of affected pages (races only mean something else did the purge)
		$titleArray = TitleArray::newFromResult( $dbw->select(
			'page',
			array( 'page_namespace', 'page_title' ),
			array( 'page_id' => $pageIds, 'page_touched' => $dbw->timestamp( $touchTimestamp ) ),
			__METHOD__
		) );

		// Update squid
		if ( $wgUseSquid ) {
			$u = SquidUpdate::newFromTitles( $titleArray );
			$u->doUpdate();
		}

		// Update file cache
		if ( $wgUseFileCache ) {
			foreach ( $titleArray as $title ) {
				HTMLFileCache::clearFileCache( $title );
			}
		}
	}

	public function workItemCount() {
		return isset( $this->params['pages'] ) ? count( $this->params['pages'] ) : 1;
	}
}