Code Coverage |
||||||||||
Lines |
Functions and Methods |
Classes and Traits |
||||||||
| Total | |
0.00% |
0 / 57 |
|
0.00% |
0 / 8 |
CRAP | |
0.00% |
0 / 1 |
| OresMetadata | |
0.00% |
0 / 57 |
|
0.00% |
0 / 8 |
240 | |
0.00% |
0 / 1 |
| __construct | |
0.00% |
0 / 5 |
|
0.00% |
0 / 1 |
2 | |||
| newFromGlobalState | |
0.00% |
0 / 7 |
|
0.00% |
0 / 1 |
2 | |||
| getMetadata | |
0.00% |
0 / 1 |
|
0.00% |
0 / 1 |
2 | |||
| getArticleQualityClass | |
0.00% |
0 / 6 |
|
0.00% |
0 / 1 |
20 | |||
| getDraftQualityClass | |
0.00% |
0 / 2 |
|
0.00% |
0 / 1 |
2 | |||
| classToMessage | |
0.00% |
0 / 2 |
|
0.00% |
0 / 1 |
6 | |||
| fetchScores | |
0.00% |
0 / 17 |
|
0.00% |
0 / 1 |
20 | |||
| getORESScores | |
0.00% |
0 / 17 |
|
0.00% |
0 / 1 |
2 | |||
| 1 | <?php |
| 2 | |
| 3 | /** |
| 4 | * This program is free software: you can redistribute it and/or modify |
| 5 | * it under the terms of the GNU General Public License as published by |
| 6 | * the Free Software Foundation, either version 3 of the License, or |
| 7 | * (at your option) any later version. |
| 8 | * |
| 9 | * This program is distributed in the hope that it will be useful, |
| 10 | * but WITHOUT ANY WARRANTY; without even the implied warranty of |
| 11 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
| 12 | * GNU General Public License for more details. |
| 13 | * |
| 14 | * You should have received a copy of the GNU General Public License |
| 15 | * along with this program. If not, see <http://www.gnu.org/licenses/>. |
| 16 | */ |
| 17 | |
| 18 | namespace MediaWiki\Extension\PageTriage; |
| 19 | |
| 20 | use IContextSource; |
| 21 | use LogicException; |
| 22 | use ORES\Services\ORESServices; |
| 23 | use ORES\Storage\ModelLookup; |
| 24 | use ORES\Storage\ThresholdLookup; |
| 25 | use Wikimedia\Rdbms\IResultWrapper; |
| 26 | |
| 27 | /** |
| 28 | * Helper class to add metadata to articles in the list view of Special:NewPagesFeed. |
| 29 | * |
| 30 | * @package MediaWiki\Extension\PageTriage |
| 31 | */ |
| 32 | class OresMetadata { |
| 33 | |
| 34 | /** |
| 35 | * @var ThresholdLookup |
| 36 | */ |
| 37 | private $thresholdLookup; |
| 38 | |
| 39 | /** |
| 40 | * @var ModelLookup |
| 41 | */ |
| 42 | private $modelLookup; |
| 43 | |
| 44 | /** |
| 45 | * @var array |
| 46 | */ |
| 47 | private $oresModelClasses; |
| 48 | |
| 49 | /** |
| 50 | * @var IContextSource |
| 51 | */ |
| 52 | private $requestContext; |
| 53 | |
| 54 | /** |
| 55 | * @var array |
| 56 | */ |
| 57 | private $scores; |
| 58 | |
| 59 | /** |
| 60 | * @var string[] Map of ORES class names from thresholds lookup mapped to translatable strings. |
| 61 | */ |
| 62 | private const ORES_CLASS_TO_MSG_KEY = [ |
| 63 | 'Stub' => 'pagetriage-filter-stat-predicted-class-stub', |
| 64 | 'Start' => 'pagetriage-filter-stat-predicted-class-start', |
| 65 | 'C' => 'pagetriage-filter-stat-predicted-class-c', |
| 66 | 'B' => 'pagetriage-filter-stat-predicted-class-b', |
| 67 | 'GA' => 'pagetriage-filter-stat-predicted-class-good', |
| 68 | 'FA' => 'pagetriage-filter-stat-predicted-class-featured', |
| 69 | 'vandalism' => 'pagetriage-filter-stat-predicted-issues-vandalism', |
| 70 | 'attack' => 'pagetriage-filter-stat-predicted-issues-attack', |
| 71 | 'spam' => 'pagetriage-filter-stat-predicted-issues-spam', |
| 72 | 'OK' => false, |
| 73 | ]; |
| 74 | |
| 75 | /** |
| 76 | * OresMetadata constructor. |
| 77 | * @param ThresholdLookup $thresholdLookup |
| 78 | * @param ModelLookup $modelLookup |
| 79 | * @param array $oresModelClasses |
| 80 | * @param IContextSource $requestContext |
| 81 | * @param int[] $pageIds |
| 82 | */ |
| 83 | public function __construct( |
| 84 | ThresholdLookup $thresholdLookup, |
| 85 | ModelLookup $modelLookup, |
| 86 | array $oresModelClasses, |
| 87 | IContextSource $requestContext, |
| 88 | $pageIds |
| 89 | ) { |
| 90 | $this->thresholdLookup = $thresholdLookup; |
| 91 | $this->modelLookup = $modelLookup; |
| 92 | $this->oresModelClasses = $oresModelClasses; |
| 93 | $this->requestContext = $requestContext; |
| 94 | |
| 95 | // Pre-fetch the ORES scores for all the pages of interest |
| 96 | $this->scores = $this->fetchScores( $pageIds ); |
| 97 | } |
| 98 | |
| 99 | /** |
| 100 | * Create an instance of OresMetadata by getting dependencies from |
| 101 | * global variables and static ORESServices |
| 102 | * |
| 103 | * @param IContextSource $context |
| 104 | * @param int[] $pageIds |
| 105 | * @return OresMetadata |
| 106 | */ |
| 107 | public static function newFromGlobalState( IContextSource $context, $pageIds ) { |
| 108 | global $wgOresModelClasses; |
| 109 | return new self( |
| 110 | ORESServices::getThresholdLookup(), |
| 111 | ORESServices::getModelLookup(), |
| 112 | $wgOresModelClasses, |
| 113 | $context, |
| 114 | $pageIds |
| 115 | ); |
| 116 | } |
| 117 | |
| 118 | /** |
| 119 | * Get ORES metadata (articlequality, draftquality) from the database for an article. |
| 120 | * |
| 121 | * @param int $pageId |
| 122 | * @return array |
| 123 | * An array to merge in with other metadata for the article. |
| 124 | */ |
| 125 | public function getMetadata( $pageId ) { |
| 126 | return $this->scores[ $pageId ]; |
| 127 | } |
| 128 | |
| 129 | /** |
| 130 | * @param float $probability |
| 131 | * @return string Name of the class corresponding to the given probability |
| 132 | */ |
| 133 | private function getArticleQualityClass( $probability ) { |
| 134 | $thresholds = $this->thresholdLookup->getThresholds( 'articlequality' ); |
| 135 | foreach ( $thresholds as $className => $threshold ) { |
| 136 | if ( $probability >= $threshold[ 'min' ] && |
| 137 | $probability <= $threshold[ 'max' ] ) { |
| 138 | return $className; |
| 139 | } |
| 140 | } |
| 141 | |
| 142 | throw new LogicException( "Couldn't determine quality class for probability $probability" ); |
| 143 | } |
| 144 | |
| 145 | /** |
| 146 | * @param int $classId |
| 147 | * @return string Name of the class corresponding to the given class id |
| 148 | */ |
| 149 | private function getDraftQualityClass( $classId ) { |
| 150 | $modelClasses = array_flip( $this->oresModelClasses[ 'draftquality' ] ); |
| 151 | return $modelClasses[ $classId ]; |
| 152 | } |
| 153 | |
| 154 | /** |
| 155 | * @param string $className |
| 156 | * @return string Translated name of the class |
| 157 | */ |
| 158 | private function classToMessage( $className ) { |
| 159 | $key = self::ORES_CLASS_TO_MSG_KEY[ $className ]; |
| 160 | return $key ? $this->requestContext->msg( $key )->text() : ''; |
| 161 | } |
| 162 | |
| 163 | /** |
| 164 | * Fetch the 'articlequality' and 'draftquality' scores for the given page ids |
| 165 | * |
| 166 | * @param int[] $pageIds |
| 167 | * @return array |
| 168 | */ |
| 169 | private function fetchScores( $pageIds ) { |
| 170 | $pendingScore = $this->requestContext->msg( |
| 171 | 'pagetriage-filter-pending-ores-score' )->text(); |
| 172 | |
| 173 | $scores = []; |
| 174 | foreach ( $pageIds as $pageId ) { |
| 175 | $scores[ $pageId ] = [ |
| 176 | 'ores_articlequality' => $pendingScore, |
| 177 | 'ores_draftquality' => '', |
| 178 | ]; |
| 179 | } |
| 180 | |
| 181 | $result = $this->getORESScores( 'articlequality', $pageIds ); |
| 182 | foreach ( $result as $row ) { |
| 183 | $scores[$row->ptrp_page_id]['ores_articlequality'] = $this->classToMessage( |
| 184 | $this->getArticleQualityClass( $row->oresc_probability ) ); |
| 185 | } |
| 186 | |
| 187 | $result = $this->getORESScores( 'draftquality', $pageIds, [ 'oresc_is_predicted' => 1 ] ); |
| 188 | foreach ( $result as $row ) { |
| 189 | $scores[$row->ptrp_page_id]['ores_draftquality'] = $this->classToMessage( |
| 190 | $this->getDraftQualityClass( $row->oresc_class ) ); |
| 191 | } |
| 192 | |
| 193 | return $scores; |
| 194 | } |
| 195 | |
| 196 | /** |
| 197 | * Select ORES scores from the database. |
| 198 | * |
| 199 | * @param string $modelName |
| 200 | * @param int[] $pageIds |
| 201 | * @param array $extraConds |
| 202 | * @return IResultWrapper |
| 203 | */ |
| 204 | private function getORESScores( $modelName, $pageIds, $extraConds = [] ) { |
| 205 | $dbr = PageTriageUtil::getReplicaConnection(); |
| 206 | $result = $dbr->newSelectQueryBuilder() |
| 207 | ->select( [ |
| 208 | 'ptrp_page_id', |
| 209 | // used for articlequality |
| 210 | 'oresc_probability', |
| 211 | // used for draftquality |
| 212 | 'oresc_class', |
| 213 | ] ) |
| 214 | ->from( 'pagetriage_page' ) |
| 215 | ->join( 'page', 'page', 'ptrp_page_id=page_id' ) |
| 216 | ->leftJoin( 'ores_classification', 'ores_classification', 'page_latest=oresc_rev' ) |
| 217 | ->where( [ |
| 218 | 'oresc_model' => $this->modelLookup->getModelId( $modelName ), |
| 219 | 'ptrp_page_id' => $pageIds, |
| 220 | ] + $extraConds ) |
| 221 | ->caller( __METHOD__ ) |
| 222 | ->fetchResultSet(); |
| 223 | |
| 224 | return $result; |
| 225 | } |
| 226 | } |