Code Coverage |
||||||||||
Lines |
Functions and Methods |
Classes and Traits |
||||||||
| Total | |
82.00% |
82 / 100 |
|
57.14% |
8 / 14 |
CRAP | |
0.00% |
0 / 1 |
| LinkBatch | |
82.83% |
82 / 99 |
|
57.14% |
8 / 14 |
37.18 | |
0.00% |
0 / 1 |
| __construct | |
100.00% |
9 / 9 |
|
100.00% |
1 / 1 |
2 | |||
| setCaller | |
0.00% |
0 / 2 |
|
0.00% |
0 / 1 |
2 | |||
| addObj | |
64.29% |
9 / 14 |
|
0.00% |
0 / 1 |
4.73 | |||
| add | |
100.00% |
3 / 3 |
|
100.00% |
1 / 1 |
3 | |||
| setArray | |
0.00% |
0 / 1 |
|
0.00% |
0 / 1 |
2 | |||
| isEmpty | |
100.00% |
1 / 1 |
|
100.00% |
1 / 1 |
1 | |||
| getSize | |
100.00% |
1 / 1 |
|
100.00% |
1 / 1 |
1 | |||
| execute | |
100.00% |
1 / 1 |
|
100.00% |
1 / 1 |
1 | |||
| getPageIdentities | |
100.00% |
3 / 3 |
|
100.00% |
1 / 1 |
2 | |||
| executeInto | |
100.00% |
3 / 3 |
|
100.00% |
1 / 1 |
1 | |||
| addResultToCache | |
87.18% |
34 / 39 |
|
0.00% |
0 / 1 |
7.10 | |||
| doQuery | |
90.91% |
10 / 11 |
|
0.00% |
0 / 1 |
3.01 | |||
| doGenderQuery | |
100.00% |
4 / 4 |
|
100.00% |
1 / 1 |
3 | |||
| constructSet | |
57.14% |
4 / 7 |
|
0.00% |
0 / 1 |
2.31 | |||
| 1 | <?php |
| 2 | /** |
| 3 | * Batch query to determine page existence. |
| 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 | namespace MediaWiki\Cache; |
| 25 | |
| 26 | use InvalidArgumentException; |
| 27 | use Language; |
| 28 | use MediaWiki\Linker\LinksMigration; |
| 29 | use MediaWiki\Linker\LinkTarget; |
| 30 | use MediaWiki\Page\PageIdentityValue; |
| 31 | use MediaWiki\Page\PageReference; |
| 32 | use MediaWiki\Page\ProperPageIdentity; |
| 33 | use MediaWiki\Title\TitleFormatter; |
| 34 | use MediaWiki\Title\TitleValue; |
| 35 | use Psr\Log\LoggerInterface; |
| 36 | use RuntimeException; |
| 37 | use Wikimedia\Assert\Assert; |
| 38 | use Wikimedia\Rdbms\IConnectionProvider; |
| 39 | use Wikimedia\Rdbms\IResultWrapper; |
| 40 | use Wikimedia\Rdbms\Platform\ISQLPlatform; |
| 41 | |
| 42 | /** |
| 43 | * Class representing a list of titles |
| 44 | * The execute() method checks them all for existence and adds them to a LinkCache object |
| 45 | * |
| 46 | * @ingroup Cache |
| 47 | */ |
| 48 | class LinkBatch { |
| 49 | /** |
| 50 | * @var array<int,array<string,mixed>> 2-d array, first index namespace, second index dbkey, value arbitrary |
| 51 | */ |
| 52 | public $data = []; |
| 53 | |
| 54 | /** |
| 55 | * @var ProperPageIdentity[]|null page identity objects corresponding to the links in the batch |
| 56 | */ |
| 57 | private $pageIdentities = null; |
| 58 | |
| 59 | /** |
| 60 | * @var string|null For debugging which method is using this class. |
| 61 | */ |
| 62 | protected $caller; |
| 63 | |
| 64 | /** |
| 65 | * @var LinkCache |
| 66 | */ |
| 67 | private $linkCache; |
| 68 | |
| 69 | /** |
| 70 | * @var TitleFormatter |
| 71 | */ |
| 72 | private $titleFormatter; |
| 73 | |
| 74 | /** |
| 75 | * @var Language |
| 76 | */ |
| 77 | private $contentLanguage; |
| 78 | |
| 79 | /** |
| 80 | * @var GenderCache |
| 81 | */ |
| 82 | private $genderCache; |
| 83 | |
| 84 | /** |
| 85 | * @var IConnectionProvider |
| 86 | */ |
| 87 | private $dbProvider; |
| 88 | |
| 89 | /** @var LinksMigration */ |
| 90 | private $linksMigration; |
| 91 | |
| 92 | /** @var LoggerInterface */ |
| 93 | private $logger; |
| 94 | |
| 95 | /** |
| 96 | * @see \MediaWiki\Cache\LinkBatchFactory |
| 97 | * |
| 98 | * @internal |
| 99 | * @param iterable<LinkTarget>|iterable<PageReference> $arr Initial items to be added to the batch |
| 100 | * @param LinkCache $linkCache |
| 101 | * @param TitleFormatter $titleFormatter |
| 102 | * @param Language $contentLanguage |
| 103 | * @param GenderCache $genderCache |
| 104 | * @param IConnectionProvider $dbProvider |
| 105 | * @param LinksMigration $linksMigration |
| 106 | * @param LoggerInterface $logger |
| 107 | */ |
| 108 | public function __construct( |
| 109 | iterable $arr, |
| 110 | LinkCache $linkCache, |
| 111 | TitleFormatter $titleFormatter, |
| 112 | Language $contentLanguage, |
| 113 | GenderCache $genderCache, |
| 114 | IConnectionProvider $dbProvider, |
| 115 | LinksMigration $linksMigration, |
| 116 | LoggerInterface $logger |
| 117 | ) { |
| 118 | $this->linkCache = $linkCache; |
| 119 | $this->titleFormatter = $titleFormatter; |
| 120 | $this->contentLanguage = $contentLanguage; |
| 121 | $this->genderCache = $genderCache; |
| 122 | $this->dbProvider = $dbProvider; |
| 123 | $this->linksMigration = $linksMigration; |
| 124 | $this->logger = $logger; |
| 125 | |
| 126 | foreach ( $arr as $item ) { |
| 127 | $this->addObj( $item ); |
| 128 | } |
| 129 | } |
| 130 | |
| 131 | /** |
| 132 | * Use ->setCaller( __METHOD__ ) to indicate which code is using this |
| 133 | * class. Only used in debugging output. |
| 134 | * @since 1.17 |
| 135 | * |
| 136 | * @param string $caller |
| 137 | * @return self (since 1.32) |
| 138 | */ |
| 139 | public function setCaller( $caller ) { |
| 140 | $this->caller = $caller; |
| 141 | |
| 142 | return $this; |
| 143 | } |
| 144 | |
| 145 | /** |
| 146 | * @param LinkTarget|PageReference $link |
| 147 | */ |
| 148 | public function addObj( $link ) { |
| 149 | if ( !$link ) { |
| 150 | // Don't die if we got null, just skip. There is nothing to do anyway. |
| 151 | // For now, let's avoid things like T282180. We should be more strict in the future. |
| 152 | $this->logger->warning( |
| 153 | 'Skipping null link, probably due to a bad title.', |
| 154 | [ 'exception' => new RuntimeException() ] |
| 155 | ); |
| 156 | return; |
| 157 | } |
| 158 | if ( $link instanceof LinkTarget && $link->isExternal() ) { |
| 159 | $this->logger->warning( |
| 160 | 'Skipping interwiki link', |
| 161 | [ 'exception' => new RuntimeException() ] |
| 162 | ); |
| 163 | return; |
| 164 | } |
| 165 | |
| 166 | Assert::parameterType( [ LinkTarget::class, PageReference::class ], $link, '$link' ); |
| 167 | $this->add( $link->getNamespace(), $link->getDBkey() ); |
| 168 | } |
| 169 | |
| 170 | /** |
| 171 | * @param int $ns |
| 172 | * @param string $dbkey |
| 173 | */ |
| 174 | public function add( $ns, $dbkey ) { |
| 175 | if ( $ns < 0 || $dbkey === '' ) { |
| 176 | // T137083 |
| 177 | return; |
| 178 | } |
| 179 | $this->data[$ns][strtr( $dbkey, ' ', '_' )] = 1; |
| 180 | } |
| 181 | |
| 182 | /** |
| 183 | * Set the link list to a given 2-d array |
| 184 | * First key is the namespace, second is the DB key, value arbitrary |
| 185 | * |
| 186 | * @param array<int,array<string,mixed>> $array |
| 187 | */ |
| 188 | public function setArray( $array ) { |
| 189 | $this->data = $array; |
| 190 | } |
| 191 | |
| 192 | /** |
| 193 | * Returns true if no pages have been added, false otherwise. |
| 194 | * |
| 195 | * @return bool |
| 196 | */ |
| 197 | public function isEmpty() { |
| 198 | return $this->getSize() == 0; |
| 199 | } |
| 200 | |
| 201 | /** |
| 202 | * Returns the size of the batch. |
| 203 | * |
| 204 | * @return int |
| 205 | */ |
| 206 | public function getSize() { |
| 207 | return count( $this->data ); |
| 208 | } |
| 209 | |
| 210 | /** |
| 211 | * Do the query and add the results to the LinkCache object |
| 212 | * |
| 213 | * @return int[] Mapping PDBK to ID |
| 214 | */ |
| 215 | public function execute() { |
| 216 | return $this->executeInto( $this->linkCache ); |
| 217 | } |
| 218 | |
| 219 | /** |
| 220 | * Do the query, add the results to the LinkCache object, |
| 221 | * and return ProperPageIdentity instances corresponding to the pages in the batch. |
| 222 | * |
| 223 | * @since 1.37 |
| 224 | * @return ProperPageIdentity[] A list of ProperPageIdentities |
| 225 | */ |
| 226 | public function getPageIdentities(): array { |
| 227 | if ( $this->pageIdentities === null ) { |
| 228 | $this->execute(); |
| 229 | } |
| 230 | |
| 231 | return $this->pageIdentities; |
| 232 | } |
| 233 | |
| 234 | /** |
| 235 | * Do the query and add the results to a given LinkCache object |
| 236 | * Return an array mapping PDBK to ID |
| 237 | * |
| 238 | * @param LinkCache $cache |
| 239 | * @return int[] Remaining IDs |
| 240 | */ |
| 241 | protected function executeInto( $cache ) { |
| 242 | $res = $this->doQuery(); |
| 243 | $this->doGenderQuery(); |
| 244 | return $this->addResultToCache( $cache, $res ); |
| 245 | } |
| 246 | |
| 247 | /** |
| 248 | * Add a result wrapper containing IDs and titles to a LinkCache object. |
| 249 | * As normal, titles will go into the static Title cache field. |
| 250 | * This function *also* stores extra fields of the title used for link |
| 251 | * parsing to avoid extra DB queries. |
| 252 | * |
| 253 | * @param LinkCache $cache |
| 254 | * @param IResultWrapper $res |
| 255 | * @return int[] Array of remaining titles |
| 256 | */ |
| 257 | public function addResultToCache( $cache, $res ) { |
| 258 | if ( !$res ) { |
| 259 | return []; |
| 260 | } |
| 261 | |
| 262 | // For each returned entry, add it to the list of good links, and remove it from $remaining |
| 263 | |
| 264 | $this->pageIdentities ??= []; |
| 265 | |
| 266 | $ids = []; |
| 267 | $remaining = $this->data; |
| 268 | foreach ( $res as $row ) { |
| 269 | try { |
| 270 | $title = new TitleValue( (int)$row->page_namespace, $row->page_title ); |
| 271 | |
| 272 | $cache->addGoodLinkObjFromRow( $title, $row ); |
| 273 | $pdbk = $this->titleFormatter->getPrefixedDBkey( $title ); |
| 274 | $ids[$pdbk] = $row->page_id; |
| 275 | |
| 276 | $pageIdentity = new PageIdentityValue( |
| 277 | (int)$row->page_id, |
| 278 | (int)$row->page_namespace, |
| 279 | $row->page_title, |
| 280 | ProperPageIdentity::LOCAL |
| 281 | ); |
| 282 | |
| 283 | $key = CacheKeyHelper::getKeyForPage( $pageIdentity ); |
| 284 | $this->pageIdentities[$key] = $pageIdentity; |
| 285 | } catch ( InvalidArgumentException $ex ) { |
| 286 | $this->logger->warning( |
| 287 | 'Encountered invalid title', |
| 288 | [ 'title_namespace' => $row->page_namespace, 'title_dbkey' => $row->page_title ] |
| 289 | ); |
| 290 | } |
| 291 | |
| 292 | unset( $remaining[$row->page_namespace][$row->page_title] ); |
| 293 | } |
| 294 | |
| 295 | // The remaining links in $data are bad links, register them as such |
| 296 | foreach ( $remaining as $ns => $dbkeys ) { |
| 297 | foreach ( $dbkeys as $dbkey => $unused ) { |
| 298 | try { |
| 299 | $title = new TitleValue( (int)$ns, (string)$dbkey ); |
| 300 | |
| 301 | $cache->addBadLinkObj( $title ); |
| 302 | $pdbk = $this->titleFormatter->getPrefixedDBkey( $title ); |
| 303 | $ids[$pdbk] = 0; |
| 304 | |
| 305 | $pageIdentity = new PageIdentityValue( 0, (int)$ns, $dbkey, ProperPageIdentity::LOCAL ); |
| 306 | $key = CacheKeyHelper::getKeyForPage( $pageIdentity ); |
| 307 | $this->pageIdentities[$key] = $pageIdentity; |
| 308 | } catch ( InvalidArgumentException $ex ) { |
| 309 | $this->logger->warning( |
| 310 | 'Encountered invalid title', |
| 311 | [ 'title_namespace' => $ns, 'title_dbkey' => $dbkey ] |
| 312 | ); |
| 313 | } |
| 314 | } |
| 315 | } |
| 316 | |
| 317 | return $ids; |
| 318 | } |
| 319 | |
| 320 | /** |
| 321 | * Perform the existence test query, return a result wrapper with page_id fields |
| 322 | * @return IResultWrapper|false |
| 323 | */ |
| 324 | public function doQuery() { |
| 325 | if ( $this->isEmpty() ) { |
| 326 | return false; |
| 327 | } |
| 328 | |
| 329 | // This is similar to LinkHolderArray::replaceInternal |
| 330 | $dbr = $this->dbProvider->getReplicaDatabase(); |
| 331 | $queryBuilder = $dbr->newSelectQueryBuilder() |
| 332 | ->select( LinkCache::getSelectFields() ) |
| 333 | ->from( 'page' ) |
| 334 | ->where( $this->constructSet( 'page', $dbr ) ); |
| 335 | |
| 336 | $caller = __METHOD__; |
| 337 | if ( strval( $this->caller ) !== '' ) { |
| 338 | $caller .= " (for {$this->caller})"; |
| 339 | } |
| 340 | |
| 341 | return $queryBuilder->caller( $caller )->fetchResultSet(); |
| 342 | } |
| 343 | |
| 344 | /** |
| 345 | * Do (and cache) {{GENDER:...}} information for userpages in this LinkBatch |
| 346 | * |
| 347 | * @return bool Whether the query was successful |
| 348 | */ |
| 349 | public function doGenderQuery() { |
| 350 | if ( $this->isEmpty() || !$this->contentLanguage->needsGenderDistinction() ) { |
| 351 | return false; |
| 352 | } |
| 353 | |
| 354 | $this->genderCache->doLinkBatch( $this->data, $this->caller ); |
| 355 | |
| 356 | return true; |
| 357 | } |
| 358 | |
| 359 | /** |
| 360 | * Construct a WHERE clause which will match all the given titles. |
| 361 | * |
| 362 | * It is the caller's responsibility to only call this if the LinkBatch is |
| 363 | * not empty, because there is no safe way to represent a SQL conditional |
| 364 | * for the empty set. |
| 365 | * |
| 366 | * @param string $prefix The appropriate table's field name prefix ('page', 'pl', etc) |
| 367 | * @param ISQLPlatform $db DB object to use |
| 368 | * @return string String with SQL where clause fragment |
| 369 | */ |
| 370 | public function constructSet( $prefix, $db ) { |
| 371 | if ( isset( $this->linksMigration::$prefixToTableMapping[$prefix] ) ) { |
| 372 | [ $blNamespace, $blTitle ] = $this->linksMigration->getTitleFields( |
| 373 | $this->linksMigration::$prefixToTableMapping[$prefix] |
| 374 | ); |
| 375 | } else { |
| 376 | $blNamespace = "{$prefix}_namespace"; |
| 377 | $blTitle = "{$prefix}_title"; |
| 378 | } |
| 379 | return $db->makeWhereFrom2d( $this->data, $blNamespace, $blTitle ); |
| 380 | } |
| 381 | } |
| 382 | |
| 383 | /** @deprecated class alias since 1.42 */ |
| 384 | class_alias( LinkBatch::class, 'LinkBatch' ); |