Code Coverage |
||||||||||
Lines |
Functions and Methods |
Classes and Traits |
||||||||
| Total | |
76.25% |
61 / 80 |
|
53.85% |
7 / 13 |
CRAP | |
0.00% |
0 / 1 |
| Util | |
76.25% |
61 / 80 |
|
53.85% |
7 / 13 |
36.77 | |
0.00% |
0 / 1 |
| getWikiID | |
75.00% |
3 / 4 |
|
0.00% |
0 / 1 |
2.06 | |||
| getORESLanguageAgnosticModelName | |
100.00% |
1 / 1 |
|
100.00% |
1 / 1 |
1 | |||
| getORESMultiLingualModelName | |
100.00% |
1 / 1 |
|
100.00% |
1 / 1 |
1 | |||
| getRevertRiskModel | |
100.00% |
6 / 6 |
|
100.00% |
1 / 1 |
2 | |||
| getAutoModeratorUser | |
0.00% |
0 / 9 |
|
0.00% |
0 / 1 |
12 | |||
| getJsonUrl | |
70.59% |
12 / 17 |
|
0.00% |
0 / 1 |
4.41 | |||
| getRawUrl | |
77.78% |
7 / 9 |
|
0.00% |
0 / 1 |
4.18 | |||
| getRevertThreshold | |
92.86% |
13 / 14 |
|
0.00% |
0 / 1 |
3.00 | |||
| getMultiLingualThresholds | |
100.00% |
2 / 2 |
|
100.00% |
1 / 1 |
1 | |||
| isMultiLingualRevertRiskEnabled | |
100.00% |
2 / 2 |
|
100.00% |
1 / 1 |
1 | |||
| initializeLiftWingClient | |
91.67% |
11 / 12 |
|
0.00% |
0 / 1 |
3.01 | |||
| initializeApiClient | |
100.00% |
1 / 1 |
|
100.00% |
1 / 1 |
1 | |||
| getLanguageConfiguration | |
100.00% |
2 / 2 |
|
100.00% |
1 / 1 |
1 | |||
| 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 | * |
| 17 | * @file |
| 18 | */ |
| 19 | |
| 20 | namespace AutoModerator; |
| 21 | |
| 22 | use MediaWiki\Config\Config; |
| 23 | use MediaWiki\Context\RequestContext; |
| 24 | use MediaWiki\Http\HttpRequestFactory; |
| 25 | use MediaWiki\Json\FormatJson; |
| 26 | use MediaWiki\Linker\LinkTarget; |
| 27 | use MediaWiki\Logger\LoggerFactory; |
| 28 | use MediaWiki\Title\TitleFactory; |
| 29 | use MediaWiki\User\User; |
| 30 | use MediaWiki\User\UserGroupManager; |
| 31 | use MediaWiki\Utils\UrlUtils; |
| 32 | use MediaWiki\WikiMap\WikiMap; |
| 33 | use RuntimeException; |
| 34 | use StatusValue; |
| 35 | use UnexpectedValueException; |
| 36 | |
| 37 | class Util { |
| 38 | |
| 39 | /** |
| 40 | * @param Config $config |
| 41 | * |
| 42 | * @return string Wiki ID used by AutoModerator. |
| 43 | */ |
| 44 | public static function getWikiID( $config ): string { |
| 45 | $autoModeratorWikiId = $config->get( 'AutoModeratorWikiId' ); |
| 46 | if ( $autoModeratorWikiId ) { |
| 47 | return $autoModeratorWikiId; |
| 48 | } |
| 49 | return WikiMap::getCurrentWikiId(); |
| 50 | } |
| 51 | |
| 52 | /** |
| 53 | * |
| 54 | * @return string The name that ORES uses for the language-agnostic model |
| 55 | */ |
| 56 | public static function getORESLanguageAgnosticModelName() { |
| 57 | return 'revertrisklanguageagnostic'; |
| 58 | } |
| 59 | |
| 60 | /** |
| 61 | * |
| 62 | * @return string The name that ORES uses for the multilingual model |
| 63 | */ |
| 64 | public static function getORESMultiLingualModelName() { |
| 65 | // TODO: This hasn't been added to ORES; check if the name matches the model name when it is |
| 66 | return 'revertriskmultilingual'; |
| 67 | } |
| 68 | |
| 69 | /** |
| 70 | * Returns the revert risk model the revision will be scored from |
| 71 | * @param Config $config |
| 72 | * @return string |
| 73 | */ |
| 74 | public static function getRevertRiskModel( Config $config ) { |
| 75 | $languageAgnosticModel = self::getORESLanguageAgnosticModelName(); |
| 76 | $multiLingualModel = self::getORESMultiLingualModelName(); |
| 77 | |
| 78 | $isLanguageModelEnabledConfig = self::isMultiLingualRevertRiskEnabled( $config ); |
| 79 | |
| 80 | // Check if a multilingual configuration exists and is enabled |
| 81 | if ( $isLanguageModelEnabledConfig ) { |
| 82 | return $multiLingualModel; |
| 83 | } else { |
| 84 | return $languageAgnosticModel; |
| 85 | } |
| 86 | } |
| 87 | |
| 88 | /** |
| 89 | * Get a user to perform moderation actions. |
| 90 | * @param Config $config |
| 91 | * @param UserGroupManager $userGroupManager |
| 92 | * |
| 93 | * @return User |
| 94 | */ |
| 95 | public static function getAutoModeratorUser( $config, $userGroupManager ): User { |
| 96 | $username = $config->get( 'AutoModeratorUsername' ); |
| 97 | $autoModeratorUser = User::newSystemUser( $username, [ 'steal' => true ] ); |
| 98 | '@phan-var User $autoModeratorUser'; |
| 99 | if ( !$autoModeratorUser ) { |
| 100 | throw new UnexpectedValueException( |
| 101 | "{$username} is invalid. Please change it." |
| 102 | ); |
| 103 | } |
| 104 | // Assign the 'bot' group to the user, so that it looks like a bot |
| 105 | if ( !in_array( 'bot', $userGroupManager->getUserGroups( $autoModeratorUser ) ) ) { |
| 106 | $userGroupManager->addUserToGroup( $autoModeratorUser, 'bot' ); |
| 107 | } |
| 108 | return $autoModeratorUser; |
| 109 | } |
| 110 | |
| 111 | /** |
| 112 | * Fetch JSON data from a remote URL, parse it and return the results. |
| 113 | * @param HttpRequestFactory $requestFactory |
| 114 | * @param string $url |
| 115 | * @param bool $isSameFarm Is the URL on the same wiki farm we are making the request from? |
| 116 | * @return StatusValue A status object with the parsed JSON value, or any errors. |
| 117 | * (Warnings coming from the HTTP library will be logged and not included here.) |
| 118 | */ |
| 119 | public static function getJsonUrl( |
| 120 | HttpRequestFactory $requestFactory, $url, $isSameFarm = false |
| 121 | ): StatusValue { |
| 122 | $options = [ |
| 123 | 'method' => 'GET', |
| 124 | 'userAgent' => $requestFactory->getUserAgent() . ' AutoModerator', |
| 125 | ]; |
| 126 | if ( $isSameFarm ) { |
| 127 | $options['originalRequest'] = RequestContext::getMain()->getRequest(); |
| 128 | } |
| 129 | $request = $requestFactory->create( $url, $options, __METHOD__ ); |
| 130 | $status = $request->execute(); |
| 131 | if ( $status->isOK() ) { |
| 132 | $status->merge( FormatJson::parse( $request->getContent(), FormatJson::FORCE_ASSOC ), true ); |
| 133 | } |
| 134 | // Log warnings here. The caller is expected to handle errors so do not double-log them. |
| 135 | [ $errorStatus, $warningStatus ] = $status->splitByErrorType(); |
| 136 | if ( !$warningStatus->isGood() ) { |
| 137 | // @todo replace 'en' with correct language configuration |
| 138 | LoggerFactory::getInstance( 'AutoModerator' )->warning( |
| 139 | $warningStatus->getWikiText( false, false, 'en' ), |
| 140 | [ 'exception' => new RuntimeException ] |
| 141 | ); |
| 142 | } |
| 143 | return $errorStatus; |
| 144 | } |
| 145 | |
| 146 | /** |
| 147 | * Get the action=raw URL for a (probably remote) title. |
| 148 | * Normal title methods would return nice URLs, which are usually disallowed for action=raw. |
| 149 | * We assume both wikis use the same URL structure. |
| 150 | * @param LinkTarget $title |
| 151 | * @param TitleFactory $titleFactory |
| 152 | * @return string |
| 153 | */ |
| 154 | public static function getRawUrl( |
| 155 | LinkTarget $title, |
| 156 | TitleFactory $titleFactory, |
| 157 | UrlUtils $urlUtils |
| 158 | ): string { |
| 159 | // Use getFullURL to get the interwiki domain. |
| 160 | $url = $titleFactory->newFromLinkTarget( $title )->getFullURL(); |
| 161 | $parts = $urlUtils->parse( (string)$urlUtils->expand( $url, PROTO_CANONICAL ) ); |
| 162 | if ( !$parts ) { |
| 163 | throw new UnexpectedValueException( 'URL is expected to be valid' ); |
| 164 | } |
| 165 | $baseUrl = $parts['scheme'] . $parts['delimiter'] . $parts['host']; |
| 166 | if ( isset( $parts['port'] ) && $parts['port'] ) { |
| 167 | $baseUrl .= ':' . $parts['port']; |
| 168 | } |
| 169 | |
| 170 | $localPageTitle = $titleFactory->makeTitle( $title->getNamespace(), $title->getDBkey() ); |
| 171 | return $baseUrl . $localPageTitle->getLocalURL( [ 'action' => 'raw' ] ); |
| 172 | } |
| 173 | |
| 174 | /** |
| 175 | * @param Config $wikiConfig |
| 176 | * @param Config $config |
| 177 | * @param string $revertRiskModelName |
| 178 | * @return float An AutoModeratorRevertProbability threshold will be chosen depending on the model |
| 179 | */ |
| 180 | public static function getRevertThreshold( Config $wikiConfig, Config $config, |
| 181 | string $revertRiskModelName ): float { |
| 182 | $threshold = 0.990; |
| 183 | $cautionLevel = $wikiConfig->get( 'AutoModeratorCautionLevel' ); |
| 184 | if ( $revertRiskModelName === self::getORESLanguageAgnosticModelName() ) { |
| 185 | $languageAgnosticThresholds = [ |
| 186 | 'very-cautious' => 0.990, |
| 187 | 'cautious' => 0.985, |
| 188 | 'somewhat-cautious' => 0.980, |
| 189 | 'less-cautious' => 0.975 |
| 190 | ]; |
| 191 | return $languageAgnosticThresholds[ $cautionLevel ]; |
| 192 | } elseif ( $revertRiskModelName === self::getORESMultiLingualModelName() ) { |
| 193 | $multiLingualThresholds = self::getMultiLingualThresholds( $config ); |
| 194 | return $multiLingualThresholds[ $cautionLevel ]; |
| 195 | } else { |
| 196 | return $threshold; |
| 197 | } |
| 198 | } |
| 199 | |
| 200 | /** |
| 201 | * Returns the revert risk model the revision will be scored from |
| 202 | * @param Config $config |
| 203 | * @return array |
| 204 | */ |
| 205 | public static function getMultiLingualThresholds( Config $config ) { |
| 206 | $multiLingualModel = $config->get( 'AutoModeratorMultiLingualRevertRisk' ); |
| 207 | |
| 208 | return $multiLingualModel[ 'thresholds' ]; |
| 209 | } |
| 210 | |
| 211 | /** |
| 212 | * Checks if multilingual revert risk is enabled on the wiki |
| 213 | * See: https://meta.wikimedia.org/wiki/Machine_learning_models/Production/Multilingual_revert_risk#Motivation |
| 214 | * for more information |
| 215 | * @param Config $config |
| 216 | * @return bool |
| 217 | */ |
| 218 | public static function isMultiLingualRevertRiskEnabled( Config $config ) { |
| 219 | $multiLingualModel = $config->get( 'AutoModeratorMultiLingualRevertRisk' ); |
| 220 | |
| 221 | return $multiLingualModel[ 'enabled' ]; |
| 222 | } |
| 223 | |
| 224 | /** |
| 225 | * @param Config $config |
| 226 | * @return LiftWingClient |
| 227 | */ |
| 228 | public static function initializeLiftWingClient( Config $config ): LiftWingClient { |
| 229 | $isMultiLingualModelEnabled = self::isMultiLingualRevertRiskEnabled( $config ); |
| 230 | if ( $isMultiLingualModelEnabled ) { |
| 231 | $model = 'revertrisk-multilingual'; |
| 232 | } else { |
| 233 | $model = 'revertrisk-language-agnostic'; |
| 234 | } |
| 235 | |
| 236 | $lang = self::getLanguageConfiguration( $config ); |
| 237 | $hostHeader = $config->get( 'AutoModeratorLiftWingAddHostHeader' ) ? |
| 238 | $config->get( 'AutoModeratorLiftWingRevertRiskHostHeader' ) : null; |
| 239 | return new LiftWingClient( |
| 240 | $model, |
| 241 | $lang, |
| 242 | $config->get( 'AutoModeratorLiftWingBaseUrl' ), |
| 243 | $hostHeader ); |
| 244 | } |
| 245 | |
| 246 | /** |
| 247 | * @return ApiClient |
| 248 | */ |
| 249 | public static function initializeApiClient(): ApiClient { |
| 250 | return new ApiClient(); |
| 251 | } |
| 252 | |
| 253 | /** |
| 254 | * @param Config $config |
| 255 | * @return false|string |
| 256 | */ |
| 257 | public static function getLanguageConfiguration( Config $config ) { |
| 258 | $wikiId = self::getWikiID( $config ); |
| 259 | return substr( $wikiId, 0, strpos( $wikiId, "wiki" ) ); |
| 260 | } |
| 261 | } |