Code Coverage |
||||||||||
Lines |
Functions and Methods |
Classes and Traits |
||||||||
| Total | |
43.31% |
165 / 381 |
|
23.81% |
5 / 21 |
CRAP | |
0.00% |
0 / 1 |
| MediaInfoHooks | |
43.31% |
165 / 381 |
|
23.81% |
5 / 21 |
1186.60 | |
0.00% |
0 / 1 |
| __construct | |
100.00% |
1 / 1 |
|
100.00% |
1 / 1 |
1 | |||
| onParserOutputPostCacheTransform | |
100.00% |
5 / 5 |
|
100.00% |
1 / 1 |
1 | |||
| onRegistration | |
0.00% |
0 / 6 |
|
0.00% |
0 / 1 |
6 | |||
| isMediaInfoPage | |
100.00% |
3 / 3 |
|
100.00% |
1 / 1 |
3 | |||
| onBeforePageDisplay | |
75.31% |
61 / 81 |
|
0.00% |
0 / 1 |
26.02 | |||
| generateWbTermsLanguages | |
0.00% |
0 / 5 |
|
0.00% |
0 / 1 |
2 | |||
| generateWbMonolingualTextLanguages | |
0.00% |
0 / 5 |
|
0.00% |
0 / 1 |
2 | |||
| tabifyStructuredData | |
24.69% |
20 / 81 |
|
0.00% |
0 / 1 |
21.38 | |||
| extractStructuredDataHtml | |
64.71% |
11 / 17 |
|
0.00% |
0 / 1 |
2.18 | |||
| createEmptyStructuredData | |
100.00% |
17 / 17 |
|
100.00% |
1 / 1 |
1 | |||
| deleteMediaInfoData | |
0.00% |
0 / 5 |
|
0.00% |
0 / 1 |
2 | |||
| getProtectionMsg | |
100.00% |
17 / 17 |
|
100.00% |
1 / 1 |
6 | |||
| onCirrusSearchProfileService | |
93.75% |
30 / 32 |
|
0.00% |
0 / 1 |
6.01 | |||
| onCirrusSearchRegisterFullTextQueryClassifiers | |
0.00% |
0 / 4 |
|
0.00% |
0 / 1 |
6 | |||
| onGetPreferences | |
0.00% |
0 / 6 |
|
0.00% |
0 / 1 |
2 | |||
| onScribuntoExternalLibraries | |
0.00% |
0 / 9 |
|
0.00% |
0 / 1 |
20 | |||
| onRevisionUndeleted | |
0.00% |
0 / 58 |
|
0.00% |
0 / 1 |
90 | |||
| onArticleUndelete | |
0.00% |
0 / 4 |
|
0.00% |
0 / 1 |
12 | |||
| onSidebarBeforeOutput | |
0.00% |
0 / 14 |
|
0.00% |
0 / 1 |
12 | |||
| onCirrusSearchAddQueryFeatures | |
0.00% |
0 / 4 |
|
0.00% |
0 / 1 |
6 | |||
| onMultiContentSave | |
0.00% |
0 / 7 |
|
0.00% |
0 / 1 |
12 | |||
| 1 | <?php |
| 2 | |
| 3 | namespace Wikibase\MediaInfo; |
| 4 | |
| 5 | use CirrusSearch\Parser\ParsedQueryClassifiersRepository; |
| 6 | use CirrusSearch\Profile\SearchProfileService; |
| 7 | use MediaWiki\CommentStore\CommentStoreComment; |
| 8 | use MediaWiki\Config\ConfigException; |
| 9 | use MediaWiki\Context\RequestContext; |
| 10 | use MediaWiki\Extension\Scribunto\Hooks\ScribuntoExternalLibrariesHook; |
| 11 | use MediaWiki\Hook\ParserOutputPostCacheTransformHook; |
| 12 | use MediaWiki\Hook\SidebarBeforeOutputHook; |
| 13 | use MediaWiki\HookContainer\HookContainer; |
| 14 | use MediaWiki\Html\Html; |
| 15 | use MediaWiki\MediaWikiServices; |
| 16 | use MediaWiki\Output\Hook\BeforePageDisplayHook; |
| 17 | use MediaWiki\Output\OutputPage; |
| 18 | use MediaWiki\Page\Hook\ArticleUndeleteHook; |
| 19 | use MediaWiki\Page\Hook\RevisionUndeletedHook; |
| 20 | use MediaWiki\Parser\ParserOutput; |
| 21 | use MediaWiki\Preferences\Hook\GetPreferencesHook; |
| 22 | use MediaWiki\Registration\ExtensionDependencyError; |
| 23 | use MediaWiki\Registration\ExtensionRegistry; |
| 24 | use MediaWiki\Revision\RenderedRevision; |
| 25 | use MediaWiki\Revision\RevisionRecord; |
| 26 | use MediaWiki\Revision\SlotRecord; |
| 27 | use MediaWiki\Skin\Skin; |
| 28 | use MediaWiki\Status\Status; |
| 29 | use MediaWiki\Storage\BlobStore; |
| 30 | use MediaWiki\Storage\Hook\MultiContentSaveHook; |
| 31 | use MediaWiki\Title\Title; |
| 32 | use MediaWiki\User\User; |
| 33 | use MediaWiki\User\UserIdentity; |
| 34 | use OOUI\HtmlSnippet; |
| 35 | use OOUI\IndexLayout; |
| 36 | use OOUI\PanelLayout; |
| 37 | use OOUI\TabPanelLayout; |
| 38 | use Wikibase\Client\WikibaseClient; |
| 39 | use Wikibase\DataModel\Entity\NumericPropertyId; |
| 40 | use Wikibase\DataModel\Services\Lookup\PropertyDataTypeLookupException; |
| 41 | use Wikibase\DataModel\Statement\StatementGuid; |
| 42 | use Wikibase\Lib\LanguageFallbackChainFactory; |
| 43 | use Wikibase\MediaInfo\Content\MediaInfoContent; |
| 44 | use Wikibase\MediaInfo\DataAccess\Scribunto\WikibaseMediaInfoEntityLibrary; |
| 45 | use Wikibase\MediaInfo\DataAccess\Scribunto\WikibaseMediaInfoLibrary; |
| 46 | use Wikibase\MediaInfo\DataModel\MediaInfo; |
| 47 | use Wikibase\MediaInfo\Search\Feature\CustomMatchFeature; |
| 48 | use Wikibase\MediaInfo\Search\MediaSearchASTClassifier; |
| 49 | use Wikibase\MediaInfo\Search\MediaSearchQueryBuilder; |
| 50 | use Wikibase\MediaInfo\Services\MediaInfoServices; |
| 51 | use Wikibase\Repo\BabelUserLanguageLookup; |
| 52 | use Wikibase\Repo\Content\EntityInstanceHolder; |
| 53 | use Wikibase\Repo\MediaWikiLocalizedTextProvider; |
| 54 | use Wikibase\Repo\ParserOutput\DispatchingEntityViewFactory; |
| 55 | use Wikibase\Repo\WikibaseRepo; |
| 56 | |
| 57 | /** |
| 58 | * MediaWiki hook handlers for the Wikibase MediaInfo extension. |
| 59 | * |
| 60 | * @license GPL-2.0-or-later |
| 61 | * @author Bene* < benestar.wikimedia@gmail.com > |
| 62 | */ |
| 63 | class MediaInfoHooks implements |
| 64 | BeforePageDisplayHook, |
| 65 | ParserOutputPostCacheTransformHook, |
| 66 | GetPreferencesHook, |
| 67 | ScribuntoExternalLibrariesHook, |
| 68 | RevisionUndeletedHook, |
| 69 | ArticleUndeleteHook, |
| 70 | SidebarBeforeOutputHook, |
| 71 | MultiContentSaveHook |
| 72 | { |
| 73 | |
| 74 | public const MEDIAINFO_SLOT_HEADER_PLACEHOLDER = '<mediainfoslotheader />'; |
| 75 | |
| 76 | public function __construct( |
| 77 | private readonly HookContainer $hookContainer, |
| 78 | ) { |
| 79 | } |
| 80 | |
| 81 | /** |
| 82 | * The placeholder mw:slotheader is replaced by default with the name of the slot |
| 83 | * |
| 84 | * Replace it with a different placeholder so we can replace it with a message later |
| 85 | * on in onBeforePageDisplay() - can't replace it here because RequestContext (and therefore |
| 86 | * the language) is not available |
| 87 | * |
| 88 | * Won't be necessary when T205444 is done |
| 89 | * |
| 90 | * @see https://phabricator.wikimedia.org/T205444 |
| 91 | * @see onBeforePageDisplay() |
| 92 | * |
| 93 | * @param ParserOutput $parserOutput |
| 94 | * @param string &$text |
| 95 | * @param array &$options |
| 96 | */ |
| 97 | public function onParserOutputPostCacheTransform( |
| 98 | $parserOutput, |
| 99 | &$text, |
| 100 | &$options |
| 101 | ): void { |
| 102 | $text = str_replace( |
| 103 | '<mw:slotheader>mediainfo</mw:slotheader>', |
| 104 | self::MEDIAINFO_SLOT_HEADER_PLACEHOLDER, |
| 105 | $text |
| 106 | ); |
| 107 | } |
| 108 | |
| 109 | /** |
| 110 | * @throws ExtensionDependencyError |
| 111 | */ |
| 112 | public static function onRegistration() { |
| 113 | if ( !ExtensionRegistry::getInstance()->isLoaded( 'WikibaseRepository' ) ) { |
| 114 | // HACK: Declaring a dependency on Wikibase in extension.json |
| 115 | // requires more work. See T258822. |
| 116 | throw new ExtensionDependencyError( [ [ |
| 117 | 'msg' => 'WikibaseMediaInfo requires Wikibase to be installed.', |
| 118 | 'type' => 'missing-phpExtension', |
| 119 | 'missing' => 'Wikibase', |
| 120 | ] ] ); |
| 121 | } |
| 122 | } |
| 123 | |
| 124 | /** |
| 125 | * @param Title|null $title |
| 126 | * @return bool |
| 127 | */ |
| 128 | public static function isMediaInfoPage( ?Title $title = null ) { |
| 129 | // Check if the page exists and the page is a file |
| 130 | return $title !== null && |
| 131 | $title->exists() && |
| 132 | $title->inNamespace( NS_FILE ); |
| 133 | } |
| 134 | |
| 135 | /** |
| 136 | * Replace mediainfo-specific placeholders (if any), move structured data, add data and modules |
| 137 | * |
| 138 | * @param OutputPage $out |
| 139 | * @param Skin $skin |
| 140 | * @throws ConfigException |
| 141 | */ |
| 142 | public function onBeforePageDisplay( $out, $skin ): void { |
| 143 | $imgTitle = $out->getTitle(); |
| 144 | |
| 145 | // Hide any MediaInfo content and UI on a page, if the target page is a redirect. |
| 146 | if ( !$imgTitle || $imgTitle->isRedirect() ) { |
| 147 | self::deleteMediaInfoData( $out ); |
| 148 | return; |
| 149 | } |
| 150 | |
| 151 | $config = $out->getConfig(); |
| 152 | |
| 153 | $isMediaInfoPage = static::isMediaInfoPage( $imgTitle ) && |
| 154 | // … the page view is a read |
| 155 | $out->getActionName() === 'view'; |
| 156 | |
| 157 | $properties = $config->get( 'MediaInfoProperties' ); |
| 158 | $propertyTypes = []; |
| 159 | $propertyTitles = []; |
| 160 | foreach ( $properties as $name => $property ) { |
| 161 | try { |
| 162 | // some properties/statements may have custom titles, in addition to their property |
| 163 | // label, to help clarify what data is expected there |
| 164 | // possible messages include: |
| 165 | // wikibasemediainfo-statements-title-depicts |
| 166 | $message = $out->msg( 'wikibasemediainfo-statements-title-' . ( $name ?: '' ) ); |
| 167 | if ( $message->exists() ) { |
| 168 | $propertyTitles[$property] = $message->text(); |
| 169 | } |
| 170 | |
| 171 | // get data type for values associated with this property |
| 172 | $propertyTypes[$property] = WBMIHooksHelper::getPropertyType( new NumericPropertyId( $property ) ); |
| 173 | } catch ( PropertyDataTypeLookupException ) { |
| 174 | // ignore invalid properties... |
| 175 | } |
| 176 | } |
| 177 | |
| 178 | // Site-wide config |
| 179 | $jsConfigVars = [ |
| 180 | 'wbmiDefaultProperties' => array_values( $properties ), |
| 181 | 'wbmiPropertyTitles' => $propertyTitles, |
| 182 | 'wbmiPropertyTypes' => $propertyTypes, |
| 183 | 'wbmiRepoApiUrl' => wfScript( 'api' ), |
| 184 | 'wbmiHelpUrls' => $config->get( 'MediaInfoHelpUrls' ), |
| 185 | 'wbmiExternalEntitySearchBaseUri' => $config->get( 'MediaInfoExternalEntitySearchBaseUri' ), |
| 186 | 'wbmiSupportedDataTypes' => $config->get( 'MediaInfoSupportedDataTypes' ), |
| 187 | ]; |
| 188 | $modules = []; |
| 189 | $moduleStyles = []; |
| 190 | |
| 191 | if ( $isMediaInfoPage ) { |
| 192 | OutputPage::setupOOUI(); |
| 193 | $this->tabifyStructuredData( $out, WikibaseRepo::getEntityViewFactory() ); |
| 194 | $out->setPreventClickjacking( true ); |
| 195 | $entityId = MediaInfoServices::getMediaInfoIdLookup()->getEntityIdForTitle( $imgTitle ); |
| 196 | $user = $out->getUser(); |
| 197 | |
| 198 | $entityLookup = WikibaseRepo::getEntityLookup(); |
| 199 | $entityRevisionId = $entityLookup->hasEntity( $entityId ) ? $imgTitle->getLatestRevID() : null; |
| 200 | $entity = $entityLookup->getEntity( $entityId ); |
| 201 | $serializer = WikibaseRepo::getAllTypesEntitySerializer(); |
| 202 | $entityData = ( $entity ? $serializer->serialize( $entity ) : null ); |
| 203 | |
| 204 | $existingPropertyTypes = []; |
| 205 | if ( $entity instanceof MediaInfo ) { |
| 206 | foreach ( $entity->getStatements() as $statement ) { |
| 207 | $propertyId = $statement->getPropertyId(); |
| 208 | try { |
| 209 | $existingPropertyTypes[$propertyId->getSerialization()] = |
| 210 | WBMIHooksHelper::getPropertyType( $propertyId ); |
| 211 | } catch ( PropertyDataTypeLookupException ) { |
| 212 | // ignore when property can't be found - it likely no longer exists; |
| 213 | // either way, we can't find what datatype is has, so there's no |
| 214 | // useful data to be gathered here |
| 215 | } |
| 216 | foreach ( $statement->getQualifiers() as $qualifierSnak ) { |
| 217 | $qualifierPropertyId = $qualifierSnak->getPropertyId(); |
| 218 | try { |
| 219 | $existingPropertyTypes[$qualifierPropertyId->getSerialization()] = |
| 220 | WBMIHooksHelper::getPropertyType( $qualifierPropertyId ); |
| 221 | } catch ( PropertyDataTypeLookupException ) { |
| 222 | // ignore when property can't be found - it likely no longer exists; |
| 223 | // either way, we can't find what datatype is has, so there's no |
| 224 | // useful data to be gathered here |
| 225 | } |
| 226 | } |
| 227 | } |
| 228 | } |
| 229 | |
| 230 | $modules[] = 'wikibase.mediainfo.filePageDisplay'; |
| 231 | $moduleStyles[] = 'wikibase.mediainfo.filepage.styles'; |
| 232 | $moduleStyles[] = 'wikibase.mediainfo.statements.styles'; |
| 233 | $userLanguageLookup = new BabelUserLanguageLookup(); |
| 234 | $tempUserConfig = MediaWikiServices::getInstance()->getTempUserConfig(); |
| 235 | |
| 236 | $jsConfigVars = array_merge( $jsConfigVars, [ |
| 237 | 'wbUserSpecifiedLanguages' => $userLanguageLookup->getAllUserLanguages( $user ), |
| 238 | 'wbCurrentRevision' => $entityRevisionId, |
| 239 | 'wbEntityId' => $entityId->getSerialization(), |
| 240 | 'wbEntity' => $entityData, |
| 241 | 'wbmiMinCaptionLength' => 5, |
| 242 | 'wbmiMaxCaptionLength' => WBMIHooksHelper::getMaxCaptionLength(), |
| 243 | 'wbmiParsedMessageAnonEditWarning' => $out->msg( |
| 244 | 'anoneditwarning', |
| 245 | // Log-in link |
| 246 | '{{fullurl:Special:UserLogin|returnto={{FULLPAGENAMEE}}}}', |
| 247 | // Sign-up link |
| 248 | '{{fullurl:Special:UserLogin/signup|returnto={{FULLPAGENAMEE}}}}' |
| 249 | )->parseAsBlock(), |
| 250 | 'wbmiProtectionMsg' => $this->getProtectionMsg( $out ), |
| 251 | 'wbmiShowIPEditingWarning' => !$tempUserConfig->isEnabled(), |
| 252 | // extend/override wbmiPropertyTypes (which already contains a property type map |
| 253 | // for all default properties) with property types for existing statements |
| 254 | 'wbmiPropertyTypes' => $jsConfigVars['wbmiPropertyTypes'] + $existingPropertyTypes, |
| 255 | ] ); |
| 256 | |
| 257 | if ( ExtensionRegistry::getInstance()->isLoaded( 'WikibaseQualityConstraints' ) ) { |
| 258 | // Don't display constraints violations unless the user is logged in |
| 259 | // and has permission to edit the page and check constraints |
| 260 | if ( |
| 261 | $user->isNamed() && |
| 262 | $user->probablyCan( 'edit', $imgTitle ) && |
| 263 | $user->isAllowed( 'wbqc-check-constraints' ) |
| 264 | ) { |
| 265 | $modules[] = 'wikibase.quality.constraints.ui'; |
| 266 | $modules[] = 'wikibase.quality.constraints.icon'; |
| 267 | $jsConfigVars['wbmiDoConstraintCheck'] = true; |
| 268 | } |
| 269 | } |
| 270 | } |
| 271 | |
| 272 | $out->addJsConfigVars( $jsConfigVars ); |
| 273 | $out->addModuleStyles( $moduleStyles ); |
| 274 | $out->addModules( $modules ); |
| 275 | } |
| 276 | |
| 277 | /** |
| 278 | * Generate the list of languages that can be used in terms. |
| 279 | * This will be exposed as part of a ResourceLoader package module. |
| 280 | * |
| 281 | * @return string[] language codes as keys, autonyms as values |
| 282 | */ |
| 283 | public static function generateWbTermsLanguages() { |
| 284 | $services = MediaWikiServices::getInstance(); |
| 285 | $allLanguages = $services->getLanguageNameUtils()->getLanguageNames(); |
| 286 | $termsLanguages = WikibaseRepo::getTermsLanguages( $services ) |
| 287 | ->getLanguages(); |
| 288 | |
| 289 | // use <code> => <name> for known languages; and add |
| 290 | // <code> => <code> for all additional acceptable language |
| 291 | // (that are not known to mediawiki) |
| 292 | return $allLanguages + array_combine( $termsLanguages, $termsLanguages ); |
| 293 | } |
| 294 | |
| 295 | /** |
| 296 | * Generate the list of languages that can be used in monolingual text. |
| 297 | * This will be exposed as part of a ResourceLoader package module. |
| 298 | * |
| 299 | * @return string[] language codes as keys, autonyms as values |
| 300 | */ |
| 301 | public static function generateWbMonolingualTextLanguages() { |
| 302 | $services = MediaWikiServices::getInstance(); |
| 303 | $allLanguages = $services->getLanguageNameUtils()->getLanguageNames(); |
| 304 | $monolingualTextLanguages = WikibaseRepo::getMonolingualTextLanguages( $services ) |
| 305 | ->getLanguages(); |
| 306 | |
| 307 | // use <code> => <name> for known languages; and add |
| 308 | // <code> => <code> for all additional acceptable language |
| 309 | // (that are not known to mediawiki) |
| 310 | return $allLanguages + array_combine( $monolingualTextLanguages, $monolingualTextLanguages ); |
| 311 | } |
| 312 | |
| 313 | /** |
| 314 | * @param OutputPage $out |
| 315 | * @param DispatchingEntityViewFactory $entityViewFactory |
| 316 | */ |
| 317 | private function tabifyStructuredData( |
| 318 | OutputPage $out, |
| 319 | DispatchingEntityViewFactory $entityViewFactory |
| 320 | ) { |
| 321 | $html = $out->getHTML(); |
| 322 | $out->clearHTML(); |
| 323 | $textProvider = new MediaWikiLocalizedTextProvider( $out->getLanguage() ); |
| 324 | |
| 325 | // Remove the slot header, as it's made redundant by the tabs |
| 326 | $html = preg_replace( WBMIHooksHelper::getStructuredDataHeaderRegex(), '', $html ); |
| 327 | |
| 328 | // Snip out out the structured data sections ($captions, $statements) |
| 329 | $extractedHtml = $this->extractStructuredDataHtml( $html, $out, $entityViewFactory ); |
| 330 | if ( preg_match( |
| 331 | WBMIHooksHelper::getMediaInfoCaptionsRegex(), |
| 332 | $extractedHtml['structured'], |
| 333 | $matches |
| 334 | ) ) { |
| 335 | $captions = $matches[1]; |
| 336 | } |
| 337 | |
| 338 | if ( preg_match( |
| 339 | WBMIHooksHelper::getMediaInfoStatementsRegex(), |
| 340 | $extractedHtml['structured'], |
| 341 | $matches |
| 342 | ) ) { |
| 343 | $statements = $matches[1]; |
| 344 | } |
| 345 | |
| 346 | if ( empty( $captions ) || empty( $statements ) ) { |
| 347 | // Something has gone wrong - markup should have been created for empty/missing data. |
| 348 | // Return the html unmodified (this should not be reachable, it's here just in case) |
| 349 | $out->addHTML( $html ); |
| 350 | return; |
| 351 | } |
| 352 | |
| 353 | // Add a title to statements for no-js |
| 354 | $statements = Html::element( |
| 355 | 'h2', |
| 356 | [ 'class' => 'wbmi-structured-data-header' ], |
| 357 | $textProvider->get( 'wikibasemediainfo-filepage-structured-data-heading' ) |
| 358 | ) . $statements; |
| 359 | |
| 360 | // Tab 1 will be everything after (and including) <div id="mw-imagepage-content"> |
| 361 | // except for children of #mw-imagepage-content before .mw-parser-output (e.g. diffs) |
| 362 | $tab1ContentRegex = '/(<div\b[^>]*\bid=(\'|")mw-imagepage-content\\2[^>]*>)(.*)' . |
| 363 | '(<div\b[^>]*\bclass=(\'|")[^\'"]+mw-parser-output\\5[^>]*>.*$)/is'; |
| 364 | // Snip out the div, and replace with a placeholder |
| 365 | if ( |
| 366 | preg_match( |
| 367 | $tab1ContentRegex, |
| 368 | $extractedHtml['unstructured'], |
| 369 | $matches |
| 370 | ) |
| 371 | ) { |
| 372 | $tab1Html = $matches[1] . $matches[4]; |
| 373 | |
| 374 | // insert captions at the beginning of Tab1 |
| 375 | $tab1Html = $captions . $tab1Html; |
| 376 | |
| 377 | $html = preg_replace( |
| 378 | $tab1ContentRegex, |
| 379 | '$3<WBMI_TABS_PLACEHOLDER>', |
| 380 | $extractedHtml['unstructured'] |
| 381 | ); |
| 382 | // Add a title for no-js |
| 383 | $tab1Html = Html::element( |
| 384 | 'h2', |
| 385 | [ 'class' => 'wbmi-captions-header' ], |
| 386 | $textProvider->get( 'wikibasemediainfo-filepage-captions-title' ) |
| 387 | ) . $tab1Html; |
| 388 | } else { |
| 389 | // If the div isn't found, something has gone wrong - return unmodified html |
| 390 | // (this should not be reachable, it's here just in case) |
| 391 | $out->addHTML( $html ); |
| 392 | return; |
| 393 | } |
| 394 | |
| 395 | // Prepare tab panels |
| 396 | $tab1 = new TabPanelLayout( |
| 397 | 'wikiTextPlusCaptions', |
| 398 | [ |
| 399 | 'classes' => [ 'wbmi-tab' ], |
| 400 | 'label' => $textProvider->get( 'wikibasemediainfo-filepage-fileinfo-heading' ), |
| 401 | 'content' => new HtmlSnippet( $tab1Html ), |
| 402 | 'expanded' => false, |
| 403 | ] |
| 404 | ); |
| 405 | $tab2 = new TabPanelLayout( |
| 406 | 'statements', |
| 407 | [ |
| 408 | 'classes' => [ 'wbmi-tab' ], |
| 409 | 'label' => $textProvider->get( 'wikibasemediainfo-filepage-structured-data-heading' ), |
| 410 | 'content' => new HtmlSnippet( $statements ), |
| 411 | 'expanded' => false, |
| 412 | ] |
| 413 | ); |
| 414 | $tabs = new IndexLayout( [ |
| 415 | 'autoFocus' => false, |
| 416 | 'classes' => [ 'wbmi-tabs' ], |
| 417 | 'expanded' => false, |
| 418 | 'framed' => false, |
| 419 | ] ); |
| 420 | $tabs->addTabPanels( [ $tab1, $tab2 ] ); |
| 421 | // This shouldn't be needed, as this is the first tab, but it is (T340803) |
| 422 | $tabs->setTabPanel( 'wikiTextPlusCaptions' ); |
| 423 | $tabs->setInfusable( true ); |
| 424 | |
| 425 | $tabWrapper = new PanelLayout( [ |
| 426 | 'classes' => [ 'wbmi-tabs-container' ], |
| 427 | 'content' => $tabs, |
| 428 | 'expanded' => false, |
| 429 | 'framed' => false, |
| 430 | ] ); |
| 431 | |
| 432 | // Replace the placeholder with the tabs |
| 433 | $html = str_replace( '<WBMI_TABS_PLACEHOLDER>', $tabWrapper, $html ); |
| 434 | |
| 435 | $out->addHTML( $html ); |
| 436 | } |
| 437 | |
| 438 | /** |
| 439 | * Returns an array with 2 elements |
| 440 | * [ |
| 441 | * 'unstructured' => html output with structured data removed |
| 442 | * 'structured' => structured data as html ... if there is no structured data an empty |
| 443 | * mediainfoview is used to create the html |
| 444 | * ] |
| 445 | * |
| 446 | * @param string $html |
| 447 | * @param OutputPage $out |
| 448 | * @param DispatchingEntityViewFactory $entityViewFactory |
| 449 | * @return string[] |
| 450 | */ |
| 451 | private function extractStructuredDataHtml( |
| 452 | $html, |
| 453 | OutputPage $out, |
| 454 | DispatchingEntityViewFactory $entityViewFactory |
| 455 | ) { |
| 456 | if ( preg_match( |
| 457 | WBMIHooksHelper::getMediaInfoViewRegex(), |
| 458 | $html, |
| 459 | $matches |
| 460 | ) ) { |
| 461 | $structured = $matches[1]; |
| 462 | $unstructured = preg_replace( |
| 463 | WBMIHooksHelper::getMediaInfoViewRegex(), |
| 464 | '', |
| 465 | $html |
| 466 | ); |
| 467 | } else { |
| 468 | $unstructured = $html; |
| 469 | $structured = $this->createEmptyStructuredData( $out, $entityViewFactory ); |
| 470 | } |
| 471 | return [ |
| 472 | 'unstructured' => $unstructured, |
| 473 | 'structured' => $structured, |
| 474 | ]; |
| 475 | } |
| 476 | |
| 477 | private function createEmptyStructuredData( |
| 478 | OutputPage $out, |
| 479 | DispatchingEntityViewFactory $entityViewFactory |
| 480 | ): string { |
| 481 | $emptyMediaInfo = new MediaInfo(); |
| 482 | $fallbackChainFactory = new LanguageFallbackChainFactory(); |
| 483 | $view = $entityViewFactory->newEntityView( |
| 484 | $out->getLanguage(), |
| 485 | $fallbackChainFactory->newFromLanguage( $out->getLanguage() ), |
| 486 | $emptyMediaInfo |
| 487 | ); |
| 488 | |
| 489 | $structured = $view->getContent( |
| 490 | $emptyMediaInfo, |
| 491 | /* EntityRevision::UNSAVED_REVISION */ |
| 492 | 0 |
| 493 | )->getHtml(); |
| 494 | |
| 495 | // Strip out the surrounding <mediaInfoView> tag |
| 496 | $structured = preg_replace( |
| 497 | WBMIHooksHelper::getMediaInfoViewRegex(), |
| 498 | '$1', |
| 499 | $structured |
| 500 | ); |
| 501 | |
| 502 | return $structured; |
| 503 | } |
| 504 | |
| 505 | /** |
| 506 | * Delete all MediaInfo data from the output |
| 507 | * |
| 508 | * @param OutputPage $out |
| 509 | */ |
| 510 | private static function deleteMediaInfoData( $out ) { |
| 511 | $html = $out->getHTML(); |
| 512 | $out->clearHTML(); |
| 513 | $html = preg_replace( WBMIHooksHelper::getMediaInfoViewRegex(), '', $html ); |
| 514 | $html = preg_replace( WBMIHooksHelper::getStructuredDataHeaderRegex(), '', $html ); |
| 515 | $out->addHTML( $html ); |
| 516 | } |
| 517 | |
| 518 | /** |
| 519 | * If this file is protected, get the appropriate message for the user. |
| 520 | * |
| 521 | * Passing the message HTML to JS may not be ideal, but some messages are |
| 522 | * templates and template syntax isn't supported in JS. See |
| 523 | * https://www.mediawiki.org/wiki/Manual:Messages_API#Using_messages_in_JavaScript. |
| 524 | * |
| 525 | * @param OutputPage $out |
| 526 | * @return string|null |
| 527 | */ |
| 528 | private function getProtectionMsg( $out ) { |
| 529 | $imgTitle = $out->getTitle(); |
| 530 | $msg = null; |
| 531 | |
| 532 | $services = MediaWikiServices::getInstance(); |
| 533 | $restrictionStore = $services->getRestrictionStore(); |
| 534 | |
| 535 | // Full protection. |
| 536 | if ( $restrictionStore->isProtected( $imgTitle, 'edit' ) && |
| 537 | !$restrictionStore->isSemiProtected( $imgTitle, 'edit' ) |
| 538 | ) { |
| 539 | $msg = $out->msg( 'protectedpagetext', 'editprotected', 'edit' )->parseAsBlock(); |
| 540 | } |
| 541 | |
| 542 | // Semi-protection. |
| 543 | if ( $restrictionStore->isSemiProtected( $imgTitle, 'edit' ) ) { |
| 544 | $msg = $out->msg( 'protectedpagetext', 'editsemiprotected', 'edit' )->parseAsBlock(); |
| 545 | } |
| 546 | |
| 547 | // Cascading protection. |
| 548 | if ( $restrictionStore->isCascadeProtected( $imgTitle ) ) { |
| 549 | // Get the protected page(s) causing this file to be protected. |
| 550 | [ $cascadeSources ] = $restrictionStore->getCascadeProtectionSources( $imgTitle ); |
| 551 | $sources = ''; |
| 552 | $titleFormatter = $services->getTitleFormatter(); |
| 553 | foreach ( $cascadeSources as $pageIdentity ) { |
| 554 | $sources .= '* [[:' . $titleFormatter->getPrefixedText( $pageIdentity ) . "]]\n"; |
| 555 | } |
| 556 | |
| 557 | $msg = $out->msg( 'cascadeprotected', count( $cascadeSources ), $sources )->parseAsBlock(); |
| 558 | } |
| 559 | |
| 560 | return $msg; |
| 561 | } |
| 562 | |
| 563 | /** |
| 564 | * Register a ProfileContext for cirrus that will mean that queries in NS_FILE will use |
| 565 | * the MediaQueryBuilder class for searching |
| 566 | */ |
| 567 | public static function onCirrusSearchProfileService( SearchProfileService $service ) { |
| 568 | global $wgWBCSUseCirrus; |
| 569 | if ( !$wgWBCSUseCirrus ) { |
| 570 | // avoid leaking into CirrusSearch test suite, where $wgWBCSUseCirrus |
| 571 | // will be false |
| 572 | return; |
| 573 | } |
| 574 | |
| 575 | // Register the query builder profiles so that they are usable in interleaved A/B test |
| 576 | $service->registerFileRepository( SearchProfileService::FT_QUERY_BUILDER, |
| 577 | // this string is to prevent overwriting, not used for retrieval |
| 578 | 'mediainfo_base', |
| 579 | __DIR__ . '/Search/MediaSearchProfiles.php' ); |
| 580 | |
| 581 | $service->registerFileRepository( SearchProfileService::RESCORE, |
| 582 | // this string is to prevent overwriting, not used for retrieval |
| 583 | 'mediainfo_base', |
| 584 | __DIR__ . '/Search/MediaSearchRescoreProfiles.php' ); |
| 585 | |
| 586 | $service->registerFileRepository( SearchProfileService::RESCORE_FUNCTION_CHAINS, |
| 587 | // this string is to prevent overwriting, not used for retrieval |
| 588 | 'mediainfo_base', |
| 589 | __DIR__ . '/Search/MediaSearchRescoreFunctionChains.php' ); |
| 590 | |
| 591 | $searchProfileContextName = MediaSearchQueryBuilder::SEARCH_PROFILE_CONTEXT_NAME; |
| 592 | // array key in MediaSearchProfiles.php |
| 593 | $rescoreProfileName = 'classic_noboostlinks_max_boost_template'; |
| 594 | |
| 595 | // Need to register a rescore profile for the profile context |
| 596 | $service->registerDefaultProfile( SearchProfileService::RESCORE, |
| 597 | $searchProfileContextName, $rescoreProfileName ); |
| 598 | |
| 599 | $request = RequestContext::getMain()->getRequest(); |
| 600 | $mwServices = MediaWikiServices::getInstance(); |
| 601 | $config = $mwServices->getMainConfig(); |
| 602 | $profiles = array_keys( $config->get( 'MediaInfoMediaSearchProfiles' ) ?: [] ); |
| 603 | if ( $profiles ) { |
| 604 | // first profile is the default mediasearch profile |
| 605 | $fulltextProfileName = $profiles[0]; |
| 606 | |
| 607 | foreach ( $profiles as $profile ) { |
| 608 | if ( $request->getCheck( $profile ) ) { |
| 609 | // switch to non-default implementations (only) when explicitly requested |
| 610 | $fulltextProfileName = $profile; |
| 611 | } |
| 612 | } |
| 613 | |
| 614 | $service->registerDefaultProfile( SearchProfileService::FT_QUERY_BUILDER, |
| 615 | $searchProfileContextName, $fulltextProfileName ); |
| 616 | |
| 617 | $service->registerFTSearchQueryRoute( |
| 618 | $searchProfileContextName, |
| 619 | 1, |
| 620 | // only for NS_FILE searches |
| 621 | [ NS_FILE ], |
| 622 | // only when the search query is found to be something mediasearch |
| 623 | // is capable of dealing with (as determined by MediaSearchASTClassifier) |
| 624 | [ $fulltextProfileName ] |
| 625 | ); |
| 626 | } |
| 627 | } |
| 628 | |
| 629 | public static function onCirrusSearchRegisterFullTextQueryClassifiers( |
| 630 | ParsedQueryClassifiersRepository $repository |
| 631 | ) { |
| 632 | $mwServices = MediaWikiServices::getInstance(); |
| 633 | $config = $mwServices->getMainConfig(); |
| 634 | $profiles = array_keys( $config->get( 'MediaInfoMediaSearchProfiles' ) ?: [] ); |
| 635 | $repository->registerClassifier( new MediaSearchASTClassifier( $profiles ) ); |
| 636 | } |
| 637 | |
| 638 | /** |
| 639 | * Handler for the GetPreferences hook |
| 640 | * |
| 641 | * @param User $user |
| 642 | * @param array[] &$preferences |
| 643 | */ |
| 644 | public function onGetPreferences( $user, &$preferences ) { |
| 645 | $preferences['wbmi-cc0-confirmed'] = [ |
| 646 | 'type' => 'api' |
| 647 | ]; |
| 648 | |
| 649 | $preferences['wbmi-wikidata-link-notice-dismissed'] = [ |
| 650 | 'type' => 'api' |
| 651 | ]; |
| 652 | } |
| 653 | |
| 654 | /** |
| 655 | * External libraries for Scribunto |
| 656 | * |
| 657 | * @param string $engine |
| 658 | * @param string[] &$extraLibraries |
| 659 | */ |
| 660 | public function onScribuntoExternalLibraries( $engine, array &$extraLibraries ) { |
| 661 | if ( !ExtensionRegistry::getInstance()->isLoaded( 'WikibaseClient' ) ) { |
| 662 | return; |
| 663 | } |
| 664 | $allowDataTransclusion = WikibaseClient::getSettings()->getSetting( 'allowDataTransclusion' ); |
| 665 | if ( $engine === 'lua' && $allowDataTransclusion === true ) { |
| 666 | $extraLibraries['mw.wikibase.mediainfo'] = WikibaseMediaInfoLibrary::class; |
| 667 | $extraLibraries['mw.wikibase.mediainfo.entity'] = [ |
| 668 | 'class' => WikibaseMediaInfoEntityLibrary::class, |
| 669 | 'deferLoad' => true, |
| 670 | ]; |
| 671 | } |
| 672 | } |
| 673 | |
| 674 | /** |
| 675 | * @param RevisionRecord $revision |
| 676 | * @param ?int $oldPageID |
| 677 | */ |
| 678 | public function onRevisionUndeleted( $revision, $oldPageID ) { |
| 679 | $title = Title::newFromPageIdentity( $revision->getPage() ); |
| 680 | if ( !$title->inNamespace( NS_FILE ) ) { |
| 681 | // short-circuit if we're not even dealing with a file |
| 682 | return; |
| 683 | } |
| 684 | |
| 685 | if ( !$revision->hasSlot( 'mediainfo' ) ) { |
| 686 | // no mediainfo content found |
| 687 | return; |
| 688 | } |
| 689 | |
| 690 | $mwServices = MediaWikiServices::getInstance(); |
| 691 | $dbw = $mwServices->getDBLoadBalancerFactory()->getPrimaryDatabase(); |
| 692 | $blobStore = $mwServices->getBlobStoreFactory()->newSqlBlobStore(); |
| 693 | $statementGuidParser = WikibaseRepo::getStatementGuidParser( $mwServices ); |
| 694 | |
| 695 | // fetch existing entity data from old revision |
| 696 | $slot = $revision->getSlot( 'mediainfo', RevisionRecord::RAW ); |
| 697 | $existingContentId = $slot->getContentId(); |
| 698 | $existingContent = $slot->getContent(); |
| 699 | if ( !( $existingContent instanceof MediaInfoContent ) ) { |
| 700 | return; |
| 701 | } |
| 702 | $existingEntity = $existingContent->getEntity(); |
| 703 | $existingEntityId = $existingEntity->getId(); |
| 704 | |
| 705 | // generate actual correct entity id for this title |
| 706 | $entityIdLookup = MediaInfoServices::getMediaInfoIdLookup(); |
| 707 | $newEntityId = $entityIdLookup->getEntityIdForTitle( $title ); |
| 708 | if ( $existingEntityId === null || $newEntityId === null || $existingEntityId->equals( $newEntityId ) ) { |
| 709 | return; |
| 710 | } |
| 711 | |
| 712 | // create new content object with the same content, but this id |
| 713 | $newEntity = $existingEntity->copy(); |
| 714 | $newEntity->setId( $newEntityId ); |
| 715 | foreach ( $newEntity->getStatements()->toArray() as $statement ) { |
| 716 | // statement GUIDs also contain the M-id, so let's go fix those too |
| 717 | $existingStatementGuidString = $statement->getGuid(); |
| 718 | // cast GUID to non-null for Phan (we know it exists) |
| 719 | '@phan-var string $existingStatementGuidString'; |
| 720 | $existingStatementGuid = $statementGuidParser->parse( $existingStatementGuidString ); |
| 721 | if ( !$newEntityId->equals( $existingStatementGuid->getEntityId() ) ) { |
| 722 | $newStatementGuid = new StatementGuid( $newEntityId, $existingStatementGuid->getGuidPart() ); |
| 723 | $statement->setGuid( (string)$newStatementGuid ); |
| 724 | } |
| 725 | } |
| 726 | $newContent = new MediaInfoContent( |
| 727 | new MediaInfoWikibaseHookRunner( $this->hookContainer ), |
| 728 | new EntityInstanceHolder( $newEntity ) |
| 729 | ); |
| 730 | |
| 731 | // store updated content in blob store |
| 732 | $unsavedSlot = SlotRecord::newUnsaved( 'mediainfo', $newContent ); |
| 733 | $blobAddress = $blobStore->storeBlob( |
| 734 | $newContent->serialize( $newContent->getDefaultFormat() ), |
| 735 | [ |
| 736 | BlobStore::PAGE_HINT => $revision->getPageId(), |
| 737 | BlobStore::REVISION_HINT => $revision->getId(), |
| 738 | BlobStore::PARENT_HINT => $revision->getParentId(), |
| 739 | BlobStore::DESIGNATION_HINT => 'page-content', |
| 740 | BlobStore::ROLE_HINT => $unsavedSlot->getRole(), |
| 741 | BlobStore::SHA1_HINT => $unsavedSlot->getSha1(), |
| 742 | BlobStore::MODEL_HINT => $newContent->getModel(), |
| 743 | BlobStore::FORMAT_HINT => $newContent->getDefaultFormat(), |
| 744 | ] |
| 745 | ); |
| 746 | |
| 747 | // update content record to point to new, corrected, content blob |
| 748 | $dbw->newUpdateQueryBuilder() |
| 749 | ->update( 'content' ) |
| 750 | ->set( [ |
| 751 | 'content_size' => $unsavedSlot->getSize(), |
| 752 | 'content_sha1' => $unsavedSlot->getSha1(), |
| 753 | 'content_address' => $blobAddress, |
| 754 | ] ) |
| 755 | ->where( [ |
| 756 | 'content_id' => $existingContentId, |
| 757 | ] ) |
| 758 | ->caller( __METHOD__ ) |
| 759 | ->execute(); |
| 760 | } |
| 761 | |
| 762 | /** |
| 763 | * @param Title $title |
| 764 | * @param bool $create |
| 765 | * @param string $comment |
| 766 | * @param int $oldPageId |
| 767 | * @param array $restoredPages |
| 768 | */ |
| 769 | public function onArticleUndelete( $title, $create, $comment, $oldPageId, $restoredPages ) { |
| 770 | if ( !$title->inNamespace( NS_FILE ) || $oldPageId === $title->getArticleID() ) { |
| 771 | return; |
| 772 | } |
| 773 | |
| 774 | // above onArticleRevisionUndeleted hook has been fixing MediaInfo ids |
| 775 | // for every undeleted revision, but now that that process is done, we |
| 776 | // need to clear the parser caches that (may have) been created during |
| 777 | // the undelete process as they were based on incorrect entities |
| 778 | $page = MediaWikiServices::getInstance()->getWikiPageFactory()->newFromTitle( $title ); |
| 779 | $page->updateParserCache( [ 'causeAction' => 'mediainfo-id-splitting' ] ); |
| 780 | } |
| 781 | |
| 782 | /** |
| 783 | * Add Concept URI link to the toolbox section of the sidebar. |
| 784 | * |
| 785 | * @param Skin $skin |
| 786 | * @param string[] &$sidebar |
| 787 | * @return void |
| 788 | */ |
| 789 | public function onSidebarBeforeOutput( $skin, &$sidebar ): void { |
| 790 | $title = $skin->getTitle(); |
| 791 | if ( !static::isMediaInfoPage( $title ) ) { |
| 792 | return; |
| 793 | } |
| 794 | |
| 795 | $entityId = MediaInfoServices::getMediaInfoIdLookup()->getEntityIdForTitle( $title ); |
| 796 | if ( $entityId === null ) { |
| 797 | return; |
| 798 | } |
| 799 | |
| 800 | $baseConceptUri = WikibaseRepo::getLocalEntitySource() |
| 801 | ->getConceptBaseUri(); |
| 802 | |
| 803 | $sidebar['TOOLBOX']['wb-concept-uri'] = [ |
| 804 | 'id' => 't-wb-concept-uri', |
| 805 | 'text' => $skin->msg( 'wikibase-concept-uri' )->text(), |
| 806 | 'href' => $baseConceptUri . $entityId->getSerialization(), |
| 807 | 'title' => $skin->msg( 'wikibase-concept-uri-tooltip' )->text() |
| 808 | ]; |
| 809 | } |
| 810 | |
| 811 | /** |
| 812 | * Add extra cirrus search query features for wikibase |
| 813 | * |
| 814 | * @param \CirrusSearch\SearchConfig $config (not used, required by hook) |
| 815 | * @param array &$extraFeatures |
| 816 | */ |
| 817 | public static function onCirrusSearchAddQueryFeatures( $config, array &$extraFeatures ) { |
| 818 | $featureConfig = MediaWikiServices::getInstance()->getMainConfig() |
| 819 | ->get( 'MediaInfoCustomMatchFeature' ); |
| 820 | if ( $featureConfig ) { |
| 821 | $extraFeatures[] = new CustomMatchFeature( $featureConfig ); |
| 822 | } |
| 823 | } |
| 824 | |
| 825 | /** |
| 826 | * @param RenderedRevision $renderedRevision |
| 827 | * @param UserIdentity $author |
| 828 | * @param CommentStoreComment $summary |
| 829 | * @param int $flags |
| 830 | * @param Status $hookStatus |
| 831 | */ |
| 832 | public function onMultiContentSave( |
| 833 | $renderedRevision, |
| 834 | $author, |
| 835 | $summary, |
| 836 | $flags, |
| 837 | $hookStatus |
| 838 | ) { |
| 839 | if ( ( $flags & EDIT_AUTOSUMMARY ) !== 0 && $renderedRevision->getRevision()->hasSlot( 'mediainfo' ) ) { |
| 840 | // remove coordinates from edit summaries when deleting location statements |
| 841 | // @see https://phabricator.wikimedia.org/T298700 |
| 842 | $coordinate = '\d+°(\d+\'(\d+(\.\d+)?")?)?'; |
| 843 | $summary->text = preg_replace( |
| 844 | "/(\/\* wbremoveclaims-remove:.+? \*\/ .+?): {$coordinate}[NS], {$coordinate}[EW]/u", |
| 845 | '$1', |
| 846 | $summary->text |
| 847 | ); |
| 848 | } |
| 849 | } |
| 850 | } |