Code Coverage |
||||||||||
Lines |
Functions and Methods |
Classes and Traits |
||||||||
| Total | |
81.67% |
49 / 60 |
|
40.00% |
2 / 5 |
CRAP | |
0.00% |
0 / 1 |
| HTMLCacheUpdater | |
83.05% |
49 / 59 |
|
40.00% |
2 / 5 |
30.55 | |
0.00% |
0 / 1 |
| __construct | |
100.00% |
5 / 5 |
|
100.00% |
1 / 1 |
1 | |||
| fieldHasFlag | |
100.00% |
1 / 1 |
|
100.00% |
1 / 1 |
1 | |||
| purgeUrls | |
72.73% |
8 / 11 |
|
0.00% |
0 / 1 |
6.73 | |||
| purgeTitleUrls | |
78.95% |
15 / 19 |
|
0.00% |
0 / 1 |
10.93 | |||
| getUrls | |
86.96% |
20 / 23 |
|
0.00% |
0 / 1 |
9.18 | |||
| 1 | <?php |
| 2 | /** |
| 3 | * This program is free software; you can redistribute it and/or modify |
| 4 | * it under the terms of the GNU General Public License as published by |
| 5 | * the Free Software Foundation; either version 2 of the License, or |
| 6 | * (at your option) any later version. |
| 7 | * |
| 8 | * This program is distributed in the hope that it will be useful, |
| 9 | * but WITHOUT ANY WARRANTY; without even the implied warranty of |
| 10 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
| 11 | * GNU General Public License for more details. |
| 12 | * |
| 13 | * You should have received a copy of the GNU General Public License along |
| 14 | * with this program; if not, write to the Free Software Foundation, Inc., |
| 15 | * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. |
| 16 | * http://www.gnu.org/copyleft/gpl.html |
| 17 | * |
| 18 | * @file |
| 19 | */ |
| 20 | |
| 21 | namespace MediaWiki\Cache; |
| 22 | |
| 23 | use MediaWiki\Deferred\CdnCacheUpdate; |
| 24 | use MediaWiki\Deferred\DeferredUpdates; |
| 25 | use MediaWiki\Deferred\HtmlFileCacheUpdate; |
| 26 | use MediaWiki\HookContainer\HookContainer; |
| 27 | use MediaWiki\HookContainer\HookRunner; |
| 28 | use MediaWiki\Page\PageIdentity; |
| 29 | use MediaWiki\Page\PageReference; |
| 30 | use MediaWiki\Title\TitleFactory; |
| 31 | use Traversable; |
| 32 | |
| 33 | /** |
| 34 | * Class to invalidate the CDN and HTMLFileCache entries associated with URLs/titles |
| 35 | * |
| 36 | * @ingroup Cache |
| 37 | * @since 1.35 |
| 38 | */ |
| 39 | class HTMLCacheUpdater { |
| 40 | /** @var int Seconds between initial and rebound purges; 0 if disabled */ |
| 41 | private $reboundDelay; |
| 42 | /** @var bool Whether filesystem-based HTML output caching is enabled */ |
| 43 | private $useFileCache; |
| 44 | /** @var int Max seconds for CDN to served cached objects without revalidation */ |
| 45 | private $cdnMaxAge; |
| 46 | |
| 47 | /** @var HookRunner */ |
| 48 | private $hookRunner; |
| 49 | |
| 50 | /** @var int Issue purge immediately and do not schedule a rebound purge */ |
| 51 | public const PURGE_NAIVE = 0; |
| 52 | /** |
| 53 | * @var int Defer purge via PRESEND deferred updates. The pending DeferrableUpdate instances |
| 54 | * will combined/de-duplicated into a single DeferrableUpdate instance; this lowers overhead |
| 55 | * and improves HTTP PURGE request pipelining. |
| 56 | */ |
| 57 | public const PURGE_PRESEND = 1; |
| 58 | /** |
| 59 | * @var int Upon purge, schedule a delayed CDN purge if rebound purges are enabled |
| 60 | * ($wgCdnReboundPurgeDelay). Rebound purges are schedule via the job queue. |
| 61 | */ |
| 62 | public const PURGE_REBOUND = 2; |
| 63 | |
| 64 | /** |
| 65 | * @var int Defer purge until no LBFactory transaction round is pending and then schedule |
| 66 | * a delayed rebound purge if rebound purges are enabled ($wgCdnReboundPurgeDelay). This is |
| 67 | * useful for CDN purges triggered by data changes in the current or last transaction round. |
| 68 | * Even if the origin server uses lagged replicas, the use of rebound purges corrects the |
| 69 | * cache in cases where lag is less than the rebound delay. If the lag is anywhere near the |
| 70 | * rebound delay, then auxiliary mechanisms should lower the cache TTL ($wgCdnMaxageLagged). |
| 71 | */ |
| 72 | public const PURGE_INTENT_TXROUND_REFLECTED = self::PURGE_PRESEND | self::PURGE_REBOUND; |
| 73 | |
| 74 | /** |
| 75 | * Reduce set of URLs to be purged to only those that may be affected by |
| 76 | * change propagation from LinksUpdate (e.g. after a used template was edited). |
| 77 | * |
| 78 | * Specifically, this means URLs only affected by direct revision edits, |
| 79 | * will not be purged. |
| 80 | * @var int |
| 81 | */ |
| 82 | public const PURGE_URLS_LINKSUPDATE_ONLY = 4; |
| 83 | |
| 84 | /** |
| 85 | * Do not bother purging cache items if the default max cache TTL implies that no objects |
| 86 | * can still be in cache from before the given timestamp. |
| 87 | * |
| 88 | * @var string |
| 89 | */ |
| 90 | public const UNLESS_CACHE_MTIME_AFTER = 'unless-timestamp-exceeds'; |
| 91 | |
| 92 | /** @var TitleFactory */ |
| 93 | private $titleFactory; |
| 94 | |
| 95 | /** |
| 96 | * @param HookContainer $hookContainer |
| 97 | * @param TitleFactory $titleFactory |
| 98 | * @param int $reboundDelay $wgCdnReboundPurgeDelay |
| 99 | * @param bool $useFileCache $wgUseFileCache |
| 100 | * @param int $cdnMaxAge $wgCdnMaxAge |
| 101 | * |
| 102 | * @internal For use with MediaWikiServices->getHtmlCacheUpdater() |
| 103 | */ |
| 104 | public function __construct( |
| 105 | HookContainer $hookContainer, |
| 106 | TitleFactory $titleFactory, |
| 107 | $reboundDelay, |
| 108 | $useFileCache, |
| 109 | $cdnMaxAge |
| 110 | ) { |
| 111 | $this->hookRunner = new HookRunner( $hookContainer ); |
| 112 | $this->titleFactory = $titleFactory; |
| 113 | $this->reboundDelay = $reboundDelay; |
| 114 | $this->useFileCache = $useFileCache; |
| 115 | $this->cdnMaxAge = $cdnMaxAge; |
| 116 | } |
| 117 | |
| 118 | /** |
| 119 | * @param int $flags Bit field |
| 120 | * @param int $flag Constant to check for |
| 121 | * @return bool If $flags contains $flag |
| 122 | */ |
| 123 | private function fieldHasFlag( $flags, $flag ) { |
| 124 | return ( ( $flags & $flag ) === $flag ); |
| 125 | } |
| 126 | |
| 127 | /** |
| 128 | * Purge the CDN for a URL or list of URLs |
| 129 | * |
| 130 | * @param string[]|string $urls URL or list of URLs |
| 131 | * @param int $flags Bit field of class PURGE_* constants |
| 132 | * [Default: HTMLCacheUpdater::PURGE_PRESEND] |
| 133 | * @param mixed[] $unless Optional map of (HTMLCacheUpdater::UNLESS_* constant => value) |
| 134 | */ |
| 135 | public function purgeUrls( $urls, $flags = self::PURGE_PRESEND, array $unless = [] ) { |
| 136 | $minFreshCacheMtime = $unless[self::UNLESS_CACHE_MTIME_AFTER] ?? null; |
| 137 | if ( $minFreshCacheMtime && time() > ( $minFreshCacheMtime + $this->cdnMaxAge ) ) { |
| 138 | return; |
| 139 | } |
| 140 | |
| 141 | $urls = is_string( $urls ) ? [ $urls ] : $urls; |
| 142 | |
| 143 | $reboundDelay = $this->fieldHasFlag( $flags, self::PURGE_REBOUND ) |
| 144 | ? $this->reboundDelay |
| 145 | : 0; // no second purge |
| 146 | |
| 147 | $update = new CdnCacheUpdate( $urls, [ 'reboundDelay' => $reboundDelay ] ); |
| 148 | if ( $this->fieldHasFlag( $flags, self::PURGE_PRESEND ) ) { |
| 149 | DeferredUpdates::addUpdate( $update, DeferredUpdates::PRESEND ); |
| 150 | } else { |
| 151 | $update->doUpdate(); |
| 152 | } |
| 153 | } |
| 154 | |
| 155 | /** |
| 156 | * Purge the CDN/HTMLFileCache for a title or the titles yielded by an iterator |
| 157 | * |
| 158 | * All cacheable canonical URLs associated with the titles will be purged from CDN. |
| 159 | * All cacheable actions associated with the titles will be purged from HTMLFileCache. |
| 160 | * |
| 161 | * @param Traversable|PageReference[]|PageReference $pages PageReference or iterator yielding |
| 162 | * PageReference instances |
| 163 | * @param int $flags Bit field of class PURGE_* constants |
| 164 | * [Default: HTMLCacheUpdater::PURGE_PRESEND] |
| 165 | * @param mixed[] $unless Optional map of (HTMLCacheUpdater::UNLESS_* constant => value) |
| 166 | */ |
| 167 | public function purgeTitleUrls( $pages, $flags = self::PURGE_PRESEND, array $unless = [] ) { |
| 168 | $pages = is_iterable( $pages ) ? $pages : [ $pages ]; |
| 169 | $pageIdentities = []; |
| 170 | |
| 171 | foreach ( $pages as $page ) { |
| 172 | // TODO: We really only need to cast to PageIdentity. We could use a LinkBatch for that. |
| 173 | $title = $this->titleFactory->newFromPageReference( $page ); |
| 174 | |
| 175 | if ( $title->canExist() ) { |
| 176 | $pageIdentities[] = $title; |
| 177 | } |
| 178 | } |
| 179 | |
| 180 | if ( !$pageIdentities ) { |
| 181 | return; |
| 182 | } |
| 183 | |
| 184 | if ( $this->useFileCache ) { |
| 185 | $update = HtmlFileCacheUpdate::newFromPages( $pageIdentities ); |
| 186 | if ( $this->fieldHasFlag( $flags, self::PURGE_PRESEND ) ) { |
| 187 | DeferredUpdates::addUpdate( $update, DeferredUpdates::PRESEND ); |
| 188 | } else { |
| 189 | $update->doUpdate(); |
| 190 | } |
| 191 | } |
| 192 | |
| 193 | $minFreshCacheMtime = $unless[self::UNLESS_CACHE_MTIME_AFTER] ?? null; |
| 194 | if ( !$minFreshCacheMtime || time() <= ( $minFreshCacheMtime + $this->cdnMaxAge ) ) { |
| 195 | $urls = []; |
| 196 | foreach ( $pageIdentities as $pi ) { |
| 197 | /** @var PageIdentity $pi */ |
| 198 | $urls = array_merge( $urls, $this->getUrls( $pi, $flags ) ); |
| 199 | } |
| 200 | $this->purgeUrls( $urls, $flags ); |
| 201 | } |
| 202 | } |
| 203 | |
| 204 | /** |
| 205 | * Get a list of URLs to purge from the CDN cache when this page changes. |
| 206 | * |
| 207 | * @param PageReference $page |
| 208 | * @param int $flags Bit field of `PURGE_URLS_*` class constants (optional). |
| 209 | * @return string[] URLs |
| 210 | */ |
| 211 | public function getUrls( PageReference $page, int $flags = 0 ): array { |
| 212 | $title = $this->titleFactory->newFromPageReference( $page ); |
| 213 | |
| 214 | if ( !$title->canExist() ) { |
| 215 | return []; |
| 216 | } |
| 217 | |
| 218 | // These urls are affected both by direct revisions as well, |
| 219 | // as re-rendering of the same content during a LinksUpdate. |
| 220 | $urls = [ |
| 221 | $title->getInternalURL() |
| 222 | ]; |
| 223 | // Language variant page views are currently not cached |
| 224 | // and thus not purged (T250511). |
| 225 | |
| 226 | // These urls are only affected by direct revisions, and do not require |
| 227 | // purging when a LinksUpdate merely rerenders the same content. |
| 228 | // This exists to avoid large amounts of redundant PURGE traffic (T250261). |
| 229 | if ( !$this->fieldHasFlag( $flags, self::PURGE_URLS_LINKSUPDATE_ONLY ) ) { |
| 230 | $urls[] = $title->getInternalURL( 'action=history' ); |
| 231 | |
| 232 | // Canonical action=raw URLs for user and site config pages (T58874, T261371). |
| 233 | if ( $title->isUserJsConfigPage() || $title->isSiteJsConfigPage() ) { |
| 234 | $urls[] = $title->getInternalURL( 'action=raw&ctype=text/javascript' ); |
| 235 | } elseif ( $title->isUserJsonConfigPage() || $title->isSiteJsonConfigPage() ) { |
| 236 | $urls[] = $title->getInternalURL( 'action=raw&ctype=application/json' ); |
| 237 | } elseif ( $title->isUserCssConfigPage() || $title->isSiteCssConfigPage() ) { |
| 238 | $urls[] = $title->getInternalURL( 'action=raw&ctype=text/css' ); |
| 239 | } |
| 240 | } |
| 241 | |
| 242 | // Extensions may add novel ways to access this content |
| 243 | $append = []; |
| 244 | $mode = $flags & self::PURGE_URLS_LINKSUPDATE_ONLY; |
| 245 | $this->hookRunner->onHtmlCacheUpdaterAppendUrls( $title, $mode, $append ); |
| 246 | $urls = array_merge( $urls, $append ); |
| 247 | |
| 248 | // Extensions may add novel ways to access the site overall |
| 249 | $append = []; |
| 250 | $this->hookRunner->onHtmlCacheUpdaterVaryUrls( $urls, $append ); |
| 251 | $urls = array_merge( $urls, $append ); |
| 252 | |
| 253 | // Legacy. TODO: Deprecate this |
| 254 | $this->hookRunner->onTitleSquidURLs( $title, $urls ); |
| 255 | |
| 256 | return $urls; |
| 257 | } |
| 258 | } |
| 259 | |
| 260 | /** @deprecated class alias since 1.42 */ |
| 261 | class_alias( HTMLCacheUpdater::class, 'HtmlCacheUpdater' ); |