includes/content/ContentHandler.php

   1 <?php
   2
   3 /**
   4  * Base class for content handling.
   5  *
   6  * This program is free software; you can redistribute it and/or modify
   7  * it under the terms of the GNU General Public License as published by
   8  * the Free Software Foundation; either version 2 of the License, or
   9  * (at your option) any later version.
  10  *
  11  * This program is distributed in the hope that it will be useful,
  12  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  13  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  14  * GNU General Public License for more details.
  15  *
  16  * You should have received a copy of the GNU General Public License along
  17  * with this program; if not, write to the Free Software Foundation, Inc.,
  18  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  19  * http://www.gnu.org/copyleft/gpl.html
  20  *
  21  * @since 1.21
  22  *
  23  * @file
  24  * @ingroup Content
  25  *
  26  * @author Daniel Kinzler
  27  */
  28
  29 use Wikimedia\Assert\Assert;
  30 use MediaWiki\Logger\LoggerFactory;
  31 use MediaWiki\MediaWikiServices;
  32 use MediaWiki\Revision\SlotRenderingProvider;
  33 use MediaWiki\Search\ParserOutputSearchDataExtractor;
  34
  35 /**
  36  * A content handler knows how do deal with a specific type of content on a wiki
  37  * page. Content is stored in the database in a serialized form (using a
  38  * serialization format a.k.a. MIME type) and is unserialized into its native
  39  * PHP representation (the content model), which is wrapped in an instance of
  40  * the appropriate subclass of Content.
  41  *
  42  * ContentHandler instances are stateless singletons that serve, among other
  43  * things, as a factory for Content objects. Generally, there is one subclass
  44  * of ContentHandler and one subclass of Content for every type of content model.
  45  *
  46  * Some content types have a flat model, that is, their native representation
  47  * is the same as their serialized form. Examples would be JavaScript and CSS
  48  * code. As of now, this also applies to wikitext (MediaWiki's default content
  49  * type), but wikitext content may be represented by a DOM or AST structure in
  50  * the future.
  51  *
  52  * @ingroup Content
  53  */
  54 abstract class ContentHandler {
  55         /**
  56          * Convenience function for getting flat text from a Content object. This
  57          * should only be used in the context of backwards compatibility with code
  58          * that is not yet able to handle Content objects!
  59          *
  60          * If $content is null, this method returns the empty string.
  61          *
  62          * If $content is an instance of TextContent, this method returns the flat
  63          * text as returned by $content->getText().
  64          *
  65          * If $content is not a TextContent object, the behavior of this method
  66          * depends on the global $wgContentHandlerTextFallback:
  67          * - If $wgContentHandlerTextFallback is 'fail' and $content is not a
  68          *   TextContent object, an MWException is thrown.
  69          * - If $wgContentHandlerTextFallback is 'serialize' and $content is not a
  70          *   TextContent object, $content->serialize() is called to get a string
  71          *   form of the content.
  72          * - If $wgContentHandlerTextFallback is 'ignore' and $content is not a
  73          *   TextContent object, this method returns null.
  74          * - otherwise, the behavior is undefined.
  75          *
  76          * @since 1.21
  77          *
  78          * @param Content|null $content
  79          *
  80          * @throws MWException If the content is not an instance of TextContent and
  81          * wgContentHandlerTextFallback was set to 'fail'.
  82          * @return string|null Textual form of the content, if available.
  83          */
  84         public static function getContentText( Content $content = null ) {
  85                 global $wgContentHandlerTextFallback;
  86
  87                 if ( is_null( $content ) ) {
  88                         return '';
  89                 }
  90
  91                 if ( $content instanceof TextContent ) {
  92                         return $content->getText();
  93                 }
  94
  95                 wfDebugLog( 'ContentHandler', 'Accessing ' . $content->getModel() . ' content as text!' );
  96
  97                 if ( $wgContentHandlerTextFallback == 'fail' ) {
  98                         throw new MWException(
  99                                 "Attempt to get text from Content with model " .
 100                                 $content->getModel()
 101                         );
 102                 }
 103
 104                 if ( $wgContentHandlerTextFallback == 'serialize' ) {
 105                         return $content->serialize();
 106                 }
 107
 108                 return null;
 109         }
 110
 111         /**
 112          * Convenience function for creating a Content object from a given textual
 113          * representation.
 114          *
 115          * $text will be deserialized into a Content object of the model specified
 116          * by $modelId (or, if that is not given, $title->getContentModel()) using
 117          * the given format.
 118          *
 119          * @since 1.21
 120          *
 121          * @param string $text The textual representation, will be
 122          *    unserialized to create the Content object
 123          * @param Title|null $title The title of the page this text belongs to.
 124          *    Required if $modelId is not provided.
 125          * @param string|null $modelId The model to deserialize to. If not provided,
 126          *    $title->getContentModel() is used.
 127          * @param string|null $format The format to use for deserialization. If not
 128          *    given, the model's default format is used.
 129          *
 130          * @throws MWException If model ID or format is not supported or if the text can not be
 131          * unserialized using the format.
 132          * @return Content A Content object representing the text.
 133          */
 134         public static function makeContent( $text, Title $title = null,
 135                 $modelId = null, $format = null ) {
 136                 if ( is_null( $modelId ) ) {
 137                         if ( is_null( $title ) ) {
 138                                 throw new MWException( "Must provide a Title object or a content model ID." );
 139                         }
 140
 141                         $modelId = $title->getContentModel();
 142                 }
 143
 144                 $handler = self::getForModelID( $modelId );
 145
 146                 return $handler->unserializeContent( $text, $format );
 147         }
 148
 149         /**
 150          * Returns the name of the default content model to be used for the page
 151          * with the given title.
 152          *
 153          * Note: There should rarely be need to call this method directly.
 154          * To determine the actual content model for a given page, use
 155          * Title::getContentModel().
 156          *
 157          * Which model is to be used by default for the page is determined based
 158          * on several factors:
 159          * - The global setting $wgNamespaceContentModels specifies a content model
 160          *   per namespace.
 161          * - The hook ContentHandlerDefaultModelFor may be used to override the page's default
 162          *   model.
 163          * - Pages in NS_MEDIAWIKI and NS_USER default to the CSS or JavaScript
 164          *   model if they end in .js or .css, respectively.
 165          * - Pages in NS_MEDIAWIKI default to the wikitext model otherwise.
 166          * - The hook TitleIsCssOrJsPage may be used to force a page to use the CSS
 167          *   or JavaScript model. This is a compatibility feature. The ContentHandlerDefaultModelFor
 168          *   hook should be used instead if possible.
 169          * - The hook TitleIsWikitextPage may be used to force a page to use the
 170          *   wikitext model. This is a compatibility feature. The ContentHandlerDefaultModelFor
 171          *   hook should be used instead if possible.
 172          *
 173          * If none of the above applies, the wikitext model is used.
 174          *
 175          * Note: this is used by, and may thus not use, Title::getContentModel()
 176          *
 177          * @since 1.21
 178          * @deprecated since 1.33, use SlotRoleHandler::getDefaultModel() together with
 179          * SlotRoleRegistry::getRoleHandler().
 180          *
 181          * @param Title $title
 182          *
 183          * @return string Default model name for the page given by $title
 184          */
 185         public static function getDefaultModelFor( Title $title ) {
 186                 $slotRoleregistry = MediaWikiServices::getInstance()->getSlotRoleRegistry();
 187                 $mainSlotHandler = $slotRoleregistry->getRoleHandler( 'main' );
 188                 return $mainSlotHandler->getDefaultModel( $title );
 189         }
 190
 191         /**
 192          * Returns the appropriate ContentHandler singleton for the given title.
 193          *
 194          * @since 1.21
 195          *
 196          * @param Title $title
 197          *
 198          * @return ContentHandler
 199          */
 200         public static function getForTitle( Title $title ) {
 201                 $modelId = $title->getContentModel();
 202
 203                 return self::getForModelID( $modelId );
 204         }
 205
 206         /**
 207          * Returns the appropriate ContentHandler singleton for the given Content
 208          * object.
 209          *
 210          * @since 1.21
 211          *
 212          * @param Content $content
 213          *
 214          * @return ContentHandler
 215          */
 216         public static function getForContent( Content $content ) {
 217                 $modelId = $content->getModel();
 218
 219                 return self::getForModelID( $modelId );
 220         }
 221
 222         /**
 223          * @var array A Cache of ContentHandler instances by model id
 224          */
 225         protected static $handlers;
 226
 227         /**
 228          * Returns the ContentHandler singleton for the given model ID. Use the
 229          * CONTENT_MODEL_XXX constants to identify the desired content model.
 230          *
 231          * ContentHandler singletons are taken from the global $wgContentHandlers
 232          * array. Keys in that array are model names, the values are either
 233          * ContentHandler singleton objects, or strings specifying the appropriate
 234          * subclass of ContentHandler.
 235          *
 236          * If a class name is encountered when looking up the singleton for a given
 237          * model name, the class is instantiated and the class name is replaced by
 238          * the resulting singleton in $wgContentHandlers.
 239          *
 240          * If no ContentHandler is defined for the desired $modelId, the
 241          * ContentHandler may be provided by the ContentHandlerForModelID hook.
 242          * If no ContentHandler can be determined, an MWException is raised.
 243          *
 244          * @since 1.21
 245          *
 246          * @param string $modelId The ID of the content model for which to get a
 247          *    handler. Use CONTENT_MODEL_XXX constants.
 248          *
 249          * @throws MWException For internal errors and problems in the configuration.
 250          * @throws MWUnknownContentModelException If no handler is known for the model ID.
 251          * @return ContentHandler The ContentHandler singleton for handling the model given by the ID.
 252          */
 253         public static function getForModelID( $modelId ) {
 254                 global $wgContentHandlers;
 255
 256                 if ( isset( self::$handlers[$modelId] ) ) {
 257                         return self::$handlers[$modelId];
 258                 }
 259
 260                 if ( empty( $wgContentHandlers[$modelId] ) ) {
 261                         $handler = null;
 262
 263                         Hooks::run( 'ContentHandlerForModelID', [ $modelId, &$handler ] );
 264
 265                         if ( $handler === null ) {
 266                                 throw new MWUnknownContentModelException( $modelId );
 267                         }
 268
 269                         if ( !( $handler instanceof ContentHandler ) ) {
 270                                 throw new MWException( "ContentHandlerForModelID must supply a ContentHandler instance" );
 271                         }
 272                 } else {
 273                         $classOrCallback = $wgContentHandlers[$modelId];
 274
 275                         if ( is_callable( $classOrCallback ) ) {
 276                                 $handler = call_user_func( $classOrCallback, $modelId );
 277                         } else {
 278                                 $handler = new $classOrCallback( $modelId );
 279                         }
 280
 281                         if ( !( $handler instanceof ContentHandler ) ) {
 282                                 throw new MWException( "$classOrCallback from \$wgContentHandlers is not " .
 283                                         "compatible with ContentHandler" );
 284                         }
 285                 }
 286
 287                 wfDebugLog( 'ContentHandler', 'Created handler for ' . $modelId
 288                         . ': ' . get_class( $handler ) );
 289
 290                 self::$handlers[$modelId] = $handler;
 291
 292                 return self::$handlers[$modelId];
 293         }
 294
 295         /**
 296          * Clean up handlers cache.
 297          */
 298         public static function cleanupHandlersCache() {
 299                 self::$handlers = [];
 300         }
 301
 302         /**
 303          * Returns the localized name for a given content model.
 304          *
 305          * Model names are localized using system messages. Message keys
 306          * have the form content-model-$name, where $name is getContentModelName( $id ).
 307          *
 308          * @param string $name The content model ID, as given by a CONTENT_MODEL_XXX
 309          *    constant or returned by Revision::getContentModel().
 310          * @param Language|null $lang The language to parse the message in (since 1.26)
 311          *
 312          * @throws MWException If the model ID isn't known.
 313          * @return string The content model's localized name.
 314          */
 315         public static function getLocalizedName( $name, Language $lang = null ) {
 316                 // Messages: content-model-wikitext, content-model-text,
 317                 // content-model-javascript, content-model-css
 318                 $key = "content-model-$name";
 319
 320                 $msg = wfMessage( $key );
 321                 if ( $lang ) {
 322                         $msg->inLanguage( $lang );
 323                 }
 324
 325                 return $msg->exists() ? $msg->plain() : $name;
 326         }
 327
 328         public static function getContentModels() {
 329                 global $wgContentHandlers;
 330
 331                 $models = array_keys( $wgContentHandlers );
 332                 Hooks::run( 'GetContentModels', [ &$models ] );
 333                 return $models;
 334         }
 335
 336         public static function getAllContentFormats() {
 337                 global $wgContentHandlers;
 338
 339                 $formats = [];
 340
 341                 foreach ( $wgContentHandlers as $model => $class ) {
 342                         $handler = self::getForModelID( $model );
 343                         $formats = array_merge( $formats, $handler->getSupportedFormats() );
 344                 }
 345
 346                 $formats = array_unique( $formats );
 347
 348                 return $formats;
 349         }
 350
 351         // ------------------------------------------------------------------------
 352
 353         /**
 354          * @var string
 355          */
 356         protected $mModelID;
 357
 358         /**
 359          * @var string[]
 360          */
 361         protected $mSupportedFormats;
 362
 363         /**
 364          * Constructor, initializing the ContentHandler instance with its model ID
 365          * and a list of supported formats. Values for the parameters are typically
 366          * provided as literals by subclass's constructors.
 367          *
 368          * @param string $modelId (use CONTENT_MODEL_XXX constants).
 369          * @param string[] $formats List for supported serialization formats
 370          *    (typically as MIME types)
 371          */
 372         public function __construct( $modelId, $formats ) {
 373                 $this->mModelID = $modelId;
 374                 $this->mSupportedFormats = $formats;
 375         }
 376
 377         /**
 378          * Serializes a Content object of the type supported by this ContentHandler.
 379          *
 380          * @since 1.21
 381          *
 382          * @param Content $content The Content object to serialize
 383          * @param string|null $format The desired serialization format
 384          *
 385          * @return string Serialized form of the content
 386          */
 387         abstract public function serializeContent( Content $content, $format = null );
 388
 389         /**
 390          * Applies transformations on export (returns the blob unchanged per default).
 391          * Subclasses may override this to perform transformations such as conversion
 392          * of legacy formats or filtering of internal meta-data.
 393          *
 394          * @param string $blob The blob to be exported
 395          * @param string|null $format The blob's serialization format
 396          *
 397          * @return string
 398          */
 399         public function exportTransform( $blob, $format = null ) {
 400                 return $blob;
 401         }
 402
 403         /**
 404          * Unserializes a Content object of the type supported by this ContentHandler.
 405          *
 406          * @since 1.21
 407          *
 408          * @param string $blob Serialized form of the content
 409          * @param string|null $format The format used for serialization
 410          *
 411          * @return Content The Content object created by deserializing $blob
 412          */
 413         abstract public function unserializeContent( $blob, $format = null );
 414
 415         /**
 416          * Apply import transformation (per default, returns $blob unchanged).
 417          * This gives subclasses an opportunity to transform data blobs on import.
 418          *
 419          * @since 1.24
 420          *
 421          * @param string $blob
 422          * @param string|null $format
 423          *
 424          * @return string
 425          */
 426         public function importTransform( $blob, $format = null ) {
 427                 return $blob;
 428         }
 429
 430         /**
 431          * Creates an empty Content object of the type supported by this
 432          * ContentHandler.
 433          *
 434          * @since 1.21
 435          *
 436          * @return Content
 437          */
 438         abstract public function makeEmptyContent();
 439
 440         /**
 441          * Creates a new Content object that acts as a redirect to the given page,
 442          * or null if redirects are not supported by this content model.
 443          *
 444          * This default implementation always returns null. Subclasses supporting redirects
 445          * must override this method.
 446          *
 447          * Note that subclasses that override this method to return a Content object
 448          * should also override supportsRedirects() to return true.
 449          *
 450          * @since 1.21
 451          *
 452          * @param Title $destination The page to redirect to.
 453          * @param string $text Text to include in the redirect, if possible.
 454          *
 455          * @return Content Always null.
 456          */
 457         public function makeRedirectContent( Title $destination, $text = '' ) {
 458                 return null;
 459         }
 460
 461         /**
 462          * Returns the model id that identifies the content model this
 463          * ContentHandler can handle. Use with the CONTENT_MODEL_XXX constants.
 464          *
 465          * @since 1.21
 466          *
 467          * @return string The model ID
 468          */
 469         public function getModelID() {
 470                 return $this->mModelID;
 471         }
 472
 473         /**
 474          * @since 1.21
 475          *
 476          * @param string $model_id The model to check
 477          *
 478          * @throws MWException If the model ID is not the ID of the content model supported by this
 479          * ContentHandler.
 480          */
 481         protected function checkModelID( $model_id ) {
 482                 if ( $model_id !== $this->mModelID ) {
 483                         throw new MWException( "Bad content model: " .
 484                                 "expected {$this->mModelID} " .
 485                                 "but got $model_id." );
 486                 }
 487         }
 488
 489         /**
 490          * Returns a list of serialization formats supported by the
 491          * serializeContent() and unserializeContent() methods of this
 492          * ContentHandler.
 493          *
 494          * @since 1.21
 495          *
 496          * @return string[] List of serialization formats as MIME type like strings
 497          */
 498         public function getSupportedFormats() {
 499                 return $this->mSupportedFormats;
 500         }
 501
 502         /**
 503          * The format used for serialization/deserialization by default by this
 504          * ContentHandler.
 505          *
 506          * This default implementation will return the first element of the array
 507          * of formats that was passed to the constructor.
 508          *
 509          * @since 1.21
 510          *
 511          * @return string The name of the default serialization format as a MIME type
 512          */
 513         public function getDefaultFormat() {
 514                 return $this->mSupportedFormats[0];
 515         }
 516
 517         /**
 518          * Returns true if $format is a serialization format supported by this
 519          * ContentHandler, and false otherwise.
 520          *
 521          * Note that if $format is null, this method always returns true, because
 522          * null means "use the default format".
 523          *
 524          * @since 1.21
 525          *
 526          * @param string $format The serialization format to check
 527          *
 528          * @return bool
 529          */
 530         public function isSupportedFormat( $format ) {
 531                 if ( !$format ) {
 532                         return true; // this means "use the default"
 533                 }
 534
 535                 return in_array( $format, $this->mSupportedFormats );
 536         }
 537
 538         /**
 539          * Convenient for checking whether a format provided as a parameter is actually supported.
 540          *
 541          * @param string $format The serialization format to check
 542          *
 543          * @throws MWException If the format is not supported by this content handler.
 544          */
 545         protected function checkFormat( $format ) {
 546                 if ( !$this->isSupportedFormat( $format ) ) {
 547                         throw new MWException(
 548                                 "Format $format is not supported for content model "
 549                                 . $this->getModelID()
 550                         );
 551                 }
 552         }
 553
 554         /**
 555          * Returns overrides for action handlers.
 556          * Classes listed here will be used instead of the default one when
 557          * (and only when) $wgActions[$action] === true. This allows subclasses
 558          * to override the default action handlers.
 559          *
 560          * @since 1.21
 561          *
 562          * @return array An array mapping action names (typically "view", "edit", "history" etc.) to
 563          *  either the full qualified class name of an Action class, a callable taking ( Page $page,
 564          *  IContextSource $context = null ) as parameters and returning an Action object, or an actual
 565          *  Action object. An empty array in this default implementation.
 566          *
 567          * @see Action::factory
 568          */
 569         public function getActionOverrides() {
 570                 return [];
 571         }
 572
 573         /**
 574          * Factory for creating an appropriate DifferenceEngine for this content model.
 575          * Since 1.32, this is only used for page-level diffs; to diff two content objects,
 576          * use getSlotDiffRenderer.
 577          *
 578          * The DifferenceEngine subclass to use is selected in getDiffEngineClass(). The
 579          * GetDifferenceEngine hook will receive the DifferenceEngine object and can replace or
 580          * wrap it.
 581          * (Note that in older versions of MediaWiki the hook documentation instructed extensions
 582          * to return false from the hook; you should not rely on always being able to decorate
 583          * the DifferenceEngine instance from the hook. If the owner of the content type wants to
 584          * decorare the instance, overriding this method is a safer approach.)
 585          *
 586          * @todo This is page-level functionality so it should not belong to ContentHandler.
 587          *   Move it to a better place once one exists (e.g. PageTypeHandler).
 588          *
 589          * @since 1.21
 590          *
 591          * @param IContextSource $context Context to use, anything else will be ignored.
 592          * @param int $old Revision ID we want to show and diff with.
 593          * @param int|string $new Either a revision ID or one of the strings 'cur', 'prev' or 'next'.
 594          * @param int $rcid FIXME: Deprecated, no longer used. Defaults to 0.
 595          * @param bool $refreshCache If set, refreshes the diff cache. Defaults to false.
 596          * @param bool $unhide If set, allow viewing deleted revs. Defaults to false.
 597          *
 598          * @return DifferenceEngine
 599          */
 600         public function createDifferenceEngine( IContextSource $context, $old = 0, $new = 0,
 601                 $rcid = 0, // FIXME: Deprecated, no longer used
 602                 $refreshCache = false, $unhide = false
 603         ) {
 604                 $diffEngineClass = $this->getDiffEngineClass();
 605                 $differenceEngine = new $diffEngineClass( $context, $old, $new, $rcid, $refreshCache, $unhide );
 606                 Hooks::run( 'GetDifferenceEngine', [ $context, $old, $new, $refreshCache, $unhide,
 607                         &$differenceEngine ] );
 608                 return $differenceEngine;
 609         }
 610
 611         /**
 612          * Get an appropriate SlotDiffRenderer for this content model.
 613          * @since 1.32
 614          * @param IContextSource $context
 615          * @return SlotDiffRenderer
 616          */
 617         final public function getSlotDiffRenderer( IContextSource $context ) {
 618                 $slotDiffRenderer = $this->getSlotDiffRendererInternal( $context );
 619                 if ( get_class( $slotDiffRenderer ) === TextSlotDiffRenderer::class ) {
 620                         //  To keep B/C, when SlotDiffRenderer is not overridden for a given content type
 621                         // but DifferenceEngine is, use that instead.
 622                         $differenceEngine = $this->createDifferenceEngine( $context );
 623                         if ( get_class( $differenceEngine ) !== DifferenceEngine::class ) {
 624                                 // TODO turn this into a deprecation warning in a later release
 625                                 LoggerFactory::getInstance( 'diff' )->info(
 626                                         'Falling back to DifferenceEngineSlotDiffRenderer', [
 627                                                 'modelID' => $this->getModelID(),
 628                                                 'DifferenceEngine' => get_class( $differenceEngine ),
 629                                         ] );
 630                                 $slotDiffRenderer = new DifferenceEngineSlotDiffRenderer( $differenceEngine );
 631                         }
 632                 }
 633                 Hooks::run( 'GetSlotDiffRenderer', [ $this, &$slotDiffRenderer, $context ] );
 634                 return $slotDiffRenderer;
 635         }
 636
 637         /**
 638          * Return the SlotDiffRenderer appropriate for this content handler.
 639          * @param IContextSource $context
 640          * @return SlotDiffRenderer
 641          */
 642         protected function getSlotDiffRendererInternal( IContextSource $context ) {
 643                 $contentLanguage = MediaWikiServices::getInstance()->getContentLanguage();
 644                 $statsdDataFactory = MediaWikiServices::getInstance()->getStatsdDataFactory();
 645                 $slotDiffRenderer = new TextSlotDiffRenderer();
 646                 $slotDiffRenderer->setStatsdDataFactory( $statsdDataFactory );
 647                 // XXX using the page language would be better, but it's unclear how that should be injected
 648                 $slotDiffRenderer->setLanguage( $contentLanguage );
 649                 $slotDiffRenderer->setWikiDiff2MovedParagraphDetectionCutoff(
 650                         $context->getConfig()->get( 'WikiDiff2MovedParagraphDetectionCutoff' )
 651                 );
 652
 653                 $engine = DifferenceEngine::getEngine();
 654                 if ( $engine === false ) {
 655                         $slotDiffRenderer->setEngine( TextSlotDiffRenderer::ENGINE_PHP );
 656                 } elseif ( $engine === 'wikidiff2' ) {
 657                         $slotDiffRenderer->setEngine( TextSlotDiffRenderer::ENGINE_WIKIDIFF2 );
 658                 } else {
 659                         $slotDiffRenderer->setEngine( TextSlotDiffRenderer::ENGINE_EXTERNAL, $engine );
 660                 }
 661
 662                 return $slotDiffRenderer;
 663         }
 664
 665         /**
 666          * Get the language in which the content of the given page is written.
 667          *
 668          * This default implementation just returns the content language (except for pages
 669          * in the MediaWiki namespace)
 670          *
 671          * Note that the page's language is not cacheable, since it may in some
 672          * cases depend on user settings.
 673          *
 674          * Also note that the page language may or may not depend on the actual content of the page,
 675          * that is, this method may load the content in order to determine the language.
 676          *
 677          * @since 1.21
 678          *
 679          * @param Title $title The page to determine the language for.
 680          * @param Content|null $content The page's content, if you have it handy, to avoid reloading it.
 681          *
 682          * @return Language The page's language
 683          */
 684         public function getPageLanguage( Title $title, Content $content = null ) {
 685                 global $wgLang;
 686                 $pageLang = MediaWikiServices::getInstance()->getContentLanguage();
 687
 688                 if ( $title->inNamespace( NS_MEDIAWIKI ) ) {
 689                         // Parse mediawiki messages with correct target language
 690                         list( /* $unused */, $lang ) = MessageCache::singleton()->figureMessage( $title->getText() );
 691                         $pageLang = Language::factory( $lang );
 692                 }
 693
 694                 Hooks::run( 'PageContentLanguage', [ $title, &$pageLang, $wgLang ] );
 695
 696                 return wfGetLangObj( $pageLang );
 697         }
 698
 699         /**
 700          * Get the language in which the content of this page is written when
 701          * viewed by user. Defaults to $this->getPageLanguage(), but if the user
 702          * specified a preferred variant, the variant will be used.
 703          *
 704          * This default implementation just returns $this->getPageLanguage( $title, $content ) unless
 705          * the user specified a preferred variant.
 706          *
 707          * Note that the pages view language is not cacheable, since it depends on user settings.
 708          *
 709          * Also note that the page language may or may not depend on the actual content of the page,
 710          * that is, this method may load the content in order to determine the language.
 711          *
 712          * @since 1.21
 713          *
 714          * @param Title $title The page to determine the language for.
 715          * @param Content|null $content The page's content, if you have it handy, to avoid reloading it.
 716          *
 717          * @return Language The page's language for viewing
 718          */
 719         public function getPageViewLanguage( Title $title, Content $content = null ) {
 720                 $pageLang = $this->getPageLanguage( $title, $content );
 721
 722                 if ( $title->getNamespace() !== NS_MEDIAWIKI ) {
 723                         // If the user chooses a variant, the content is actually
 724                         // in a language whose code is the variant code.
 725                         $variant = $pageLang->getPreferredVariant();
 726                         if ( $pageLang->getCode() !== $variant ) {
 727                                 $pageLang = Language::factory( $variant );
 728                         }
 729                 }
 730
 731                 return $pageLang;
 732         }
 733
 734         /**
 735          * Determines whether the content type handled by this ContentHandler
 736          * can be used for the main slot of the given page.
 737          *
 738          * This default implementation always returns true.
 739          * Subclasses may override this to restrict the use of this content model to specific locations,
 740          * typically based on the namespace or some other aspect of the title, such as a special suffix
 741          * (e.g. ".svg" for SVG content).
 742          *
 743          * @note this calls the ContentHandlerCanBeUsedOn hook which may be used to override which
 744          * content model can be used where.
 745          *
 746          * @see SlotRoleHandler::isAllowedModel
 747          *
 748          * @param Title $title The page's title.
 749          *
 750          * @return bool True if content of this kind can be used on the given page, false otherwise.
 751          */
 752         public function canBeUsedOn( Title $title ) {
 753                 $ok = true;
 754
 755                 Hooks::run( 'ContentModelCanBeUsedOn', [ $this->getModelID(), $title, &$ok ] );
 756
 757                 return $ok;
 758         }
 759
 760         /**
 761          * Returns the name of the diff engine to use.
 762          *
 763          * @since 1.21
 764          *
 765          * @return string
 766          */
 767         protected function getDiffEngineClass() {
 768                 return DifferenceEngine::class;
 769         }
 770
 771         /**
 772          * Attempts to merge differences between three versions. Returns a new
 773          * Content object for a clean merge and false for failure or a conflict.
 774          *
 775          * This default implementation always returns false.
 776          *
 777          * @since 1.21
 778          *
 779          * @param Content $oldContent The page's previous content.
 780          * @param Content $myContent One of the page's conflicting contents.
 781          * @param Content $yourContent One of the page's conflicting contents.
 782          *
 783          * @return Content|bool Always false.
 784          */
 785         public function merge3( Content $oldContent, Content $myContent, Content $yourContent ) {
 786                 return false;
 787         }
 788
 789         /**
 790          * Return type of change if one exists for the given edit.
 791          *
 792          * @since 1.31
 793          *
 794          * @param Content|null $oldContent The previous text of the page.
 795          * @param Content|null $newContent The submitted text of the page.
 796          * @param int $flags Bit mask: a bit mask of flags submitted for the edit.
 797          *
 798          * @return string|null String key representing type of change, or null.
 799          */
 800         private function getChangeType(
 801                 Content $oldContent = null,
 802                 Content $newContent = null,
 803                 $flags = 0
 804         ) {
 805                 $oldTarget = $oldContent !== null ? $oldContent->getRedirectTarget() : null;
 806                 $newTarget = $newContent !== null ? $newContent->getRedirectTarget() : null;
 807
 808                 // We check for the type of change in the given edit, and return string key accordingly
 809
 810                 // Blanking of a page
 811                 if ( $oldContent && $oldContent->getSize() > 0 &&
 812                         $newContent && $newContent->getSize() === 0
 813                 ) {
 814                         return 'blank';
 815                 }
 816
 817                 // Redirects
 818                 if ( $newTarget ) {
 819                         if ( !$oldTarget ) {
 820                                 // New redirect page (by creating new page or by changing content page)
 821                                 return 'new-redirect';
 822                         } elseif ( !$newTarget->equals( $oldTarget ) ||
 823                                 $oldTarget->getFragment() !== $newTarget->getFragment()
 824                         ) {
 825                                 // Redirect target changed
 826                                 return 'changed-redirect-target';
 827                         }
 828                 } elseif ( $oldTarget ) {
 829                         // Changing an existing redirect into a non-redirect
 830                         return 'removed-redirect';
 831                 }
 832
 833                 // New page created
 834                 if ( $flags & EDIT_NEW && $newContent ) {
 835                         if ( $newContent->getSize() === 0 ) {
 836                                 // New blank page
 837                                 return 'newblank';
 838                         } else {
 839                                 return 'newpage';
 840                         }
 841                 }
 842
 843                 // Removing more than 90% of the page
 844                 if ( $oldContent && $newContent && $oldContent->getSize() > 10 * $newContent->getSize() ) {
 845                         return 'replace';
 846                 }
 847
 848                 // Content model changed
 849                 if ( $oldContent && $newContent && $oldContent->getModel() !== $newContent->getModel() ) {
 850                         return 'contentmodelchange';
 851                 }
 852
 853                 return null;
 854         }
 855
 856         /**
 857          * Return an applicable auto-summary if one exists for the given edit.
 858          *
 859          * @since 1.21
 860          *
 861          * @param Content|null $oldContent The previous text of the page.
 862          * @param Content|null $newContent The submitted text of the page.
 863          * @param int $flags Bit mask: a bit mask of flags submitted for the edit.
 864          *
 865          * @return string An appropriate auto-summary, or an empty string.
 866          */
 867         public function getAutosummary(
 868                 Content $oldContent = null,
 869                 Content $newContent = null,
 870                 $flags = 0
 871         ) {
 872                 $changeType = $this->getChangeType( $oldContent, $newContent, $flags );
 873
 874                 // There's no applicable auto-summary for our case, so our auto-summary is empty.
 875                 if ( !$changeType ) {
 876                         return '';
 877                 }
 878
 879                 // Decide what kind of auto-summary is needed.
 880                 switch ( $changeType ) {
 881                         case 'new-redirect':
 882                                 $newTarget = $newContent->getRedirectTarget();
 883                                 $truncatedtext = $newContent->getTextForSummary(
 884                                         250
 885                                         - strlen( wfMessage( 'autoredircomment' )->inContentLanguage()->text() )
 886                                         - strlen( $newTarget->getFullText() )
 887                                 );
 888
 889                                 return wfMessage( 'autoredircomment', $newTarget->getFullText() )
 890                                         ->plaintextParams( $truncatedtext )->inContentLanguage()->text();
 891                         case 'changed-redirect-target':
 892                                 $oldTarget = $oldContent->getRedirectTarget();
 893                                 $newTarget = $newContent->getRedirectTarget();
 894
 895                                 $truncatedtext = $newContent->getTextForSummary(
 896                                         250
 897                                         - strlen( wfMessage( 'autosumm-changed-redirect-target' )
 898                                                 ->inContentLanguage()->text() )
 899                                         - strlen( $oldTarget->getFullText() )
 900                                         - strlen( $newTarget->getFullText() )
 901                                 );
 902
 903                                 return wfMessage( 'autosumm-changed-redirect-target',
 904                                                 $oldTarget->getFullText(),
 905                                                 $newTarget->getFullText() )
 906                                         ->rawParams( $truncatedtext )->inContentLanguage()->text();
 907                         case 'removed-redirect':
 908                                 $oldTarget = $oldContent->getRedirectTarget();
 909                                 $truncatedtext = $newContent->getTextForSummary(
 910                                         250
 911                                         - strlen( wfMessage( 'autosumm-removed-redirect' )
 912                                                 ->inContentLanguage()->text() )
 913                                         - strlen( $oldTarget->getFullText() ) );
 914
 915                                 return wfMessage( 'autosumm-removed-redirect', $oldTarget->getFullText() )
 916                                         ->rawParams( $truncatedtext )->inContentLanguage()->text();
 917                         case 'newpage':
 918                                 // If they're making a new article, give its text, truncated, in the summary.
 919                                 $truncatedtext = $newContent->getTextForSummary(
 920                                         200 - strlen( wfMessage( 'autosumm-new' )->inContentLanguage()->text() ) );
 921
 922                                 return wfMessage( 'autosumm-new' )->rawParams( $truncatedtext )
 923                                         ->inContentLanguage()->text();
 924                         case 'blank':
 925                                 return wfMessage( 'autosumm-blank' )->inContentLanguage()->text();
 926                         case 'replace':
 927                                 $truncatedtext = $newContent->getTextForSummary(
 928                                         200 - strlen( wfMessage( 'autosumm-replace' )->inContentLanguage()->text() ) );
 929
 930                                 return wfMessage( 'autosumm-replace' )->rawParams( $truncatedtext )
 931                                         ->inContentLanguage()->text();
 932                         case 'newblank':
 933                                 return wfMessage( 'autosumm-newblank' )->inContentLanguage()->text();
 934                         default:
 935                                 return '';
 936                 }
 937         }
 938
 939         /**
 940          * Return an applicable tag if one exists for the given edit or return null.
 941          *
 942          * @since 1.31
 943          *
 944          * @param Content|null $oldContent The previous text of the page.
 945          * @param Content|null $newContent The submitted text of the page.
 946          * @param int $flags Bit mask: a bit mask of flags submitted for the edit.
 947          *
 948          * @return string|null An appropriate tag, or null.
 949          */
 950         public function getChangeTag(
 951                 Content $oldContent = null,
 952                 Content $newContent = null,
 953                 $flags = 0
 954         ) {
 955                 $changeType = $this->getChangeType( $oldContent, $newContent, $flags );
 956
 957                 // There's no applicable tag for this change.
 958                 if ( !$changeType ) {
 959                         return null;
 960                 }
 961
 962                 // Core tags use the same keys as ones returned from $this->getChangeType()
 963                 // but prefixed with pseudo namespace 'mw-', so we add the prefix before checking
 964                 // if this type of change should be tagged
 965                 $tag = 'mw-' . $changeType;
 966
 967                 // Not all change types are tagged, so we check against the list of defined tags.
 968                 if ( in_array( $tag, ChangeTags::getSoftwareTags() ) ) {
 969                         return $tag;
 970                 }
 971
 972                 return null;
 973         }
 974
 975         /**
 976          * Auto-generates a deletion reason
 977          *
 978          * @since 1.21
 979          *
 980          * @param Title $title The page's title
 981          * @param bool &$hasHistory Whether the page has a history
 982          *
 983          * @return mixed String containing deletion reason or empty string, or
 984          *    boolean false if no revision occurred
 985          *
 986          * @todo &$hasHistory is extremely ugly, it's here because
 987          * WikiPage::getAutoDeleteReason() and Article::generateReason()
 988          * have it / want it.
 989          */
 990         public function getAutoDeleteReason( Title $title, &$hasHistory ) {
 991                 $dbr = wfGetDB( DB_REPLICA );
 992
 993                 // Get the last revision
 994                 $rev = Revision::newFromTitle( $title );
 995
 996                 if ( is_null( $rev ) ) {
 997                         return false;
 998                 }
 999
1000                 // Get the article's contents
1001                 $content = $rev->getContent();
1002                 $blank = false;
1003
1004                 // If the page is blank, use the text from the previous revision,
1005                 // which can only be blank if there's a move/import/protect dummy
1006                 // revision involved
1007                 if ( !$content || $content->isEmpty() ) {
1008                         $prev = $rev->getPrevious();
1009
1010                         if ( $prev ) {
1011                                 $rev = $prev;
1012                                 $content = $rev->getContent();
1013                                 $blank = true;
1014                         }
1015                 }
1016
1017                 $this->checkModelID( $rev->getContentModel() );
1018
1019                 // Find out if there was only one contributor
1020                 // Only scan the last 20 revisions
1021                 $revQuery = Revision::getQueryInfo();
1022                 $res = $dbr->select(
1023                         $revQuery['tables'],
1024                         [ 'rev_user_text' => $revQuery['fields']['rev_user_text'] ],
1025                         [
1026                                 'rev_page' => $title->getArticleID(),
1027                                 $dbr->bitAnd( 'rev_deleted', Revision::DELETED_USER ) . ' = 0'
1028                         ],
1029                         __METHOD__,
1030                         [ 'LIMIT' => 20 ],
1031                         $revQuery['joins']
1032                 );
1033
1034                 if ( $res === false ) {
1035                         // This page has no revisions, which is very weird
1036                         return false;
1037                 }
1038
1039                 $hasHistory = ( $res->numRows() > 1 );
1040                 $row = $dbr->fetchObject( $res );
1041
1042                 if ( $row ) { // $row is false if the only contributor is hidden
1043                         $onlyAuthor = $row->rev_user_text;
1044                         // Try to find a second contributor
1045                         foreach ( $res as $row ) {
1046                                 if ( $row->rev_user_text != $onlyAuthor ) { // T24999
1047                                         $onlyAuthor = false;
1048                                         break;
1049                                 }
1050                         }
1051                 } else {
1052                         $onlyAuthor = false;
1053                 }
1054
1055                 // Generate the summary with a '$1' placeholder
1056                 if ( $blank ) {
1057                         // The current revision is blank and the one before is also
1058                         // blank. It's just not our lucky day
1059                         $reason = wfMessage( 'exbeforeblank', '$1' )->inContentLanguage()->text();
1060                 } else {
1061                         if ( $onlyAuthor ) {
1062                                 $reason = wfMessage(
1063                                         'excontentauthor',
1064                                         '$1',
1065                                         $onlyAuthor
1066                                 )->inContentLanguage()->text();
1067                         } else {
1068                                 $reason = wfMessage( 'excontent', '$1' )->inContentLanguage()->text();
1069                         }
1070                 }
1071
1072                 if ( $reason == '-' ) {
1073                         // Allow these UI messages to be blanked out cleanly
1074                         return '';
1075                 }
1076
1077                 // Max content length = max comment length - length of the comment (excl. $1)
1078                 $text = $content ? $content->getTextForSummary( 255 - ( strlen( $reason ) - 2 ) ) : '';
1079
1080                 // Now replace the '$1' placeholder
1081                 $reason = str_replace( '$1', $text, $reason );
1082
1083                 return $reason;
1084         }
1085
1086         /**
1087          * Get the Content object that needs to be saved in order to undo all revisions
1088          * between $undo and $undoafter. Revisions must belong to the same page,
1089          * must exist and must not be deleted.
1090          *
1091          * @since 1.21
1092          * @since 1.32 accepts Content objects for all parameters instead of Revision objects.
1093          *  Passing Revision objects is deprecated.
1094          *
1095          * @param Revision|Content $current The current text
1096          * @param Revision|Content $undo The content of the revision to undo
1097          * @param Revision|Content $undoafter Must be from an earlier revision than $undo
1098          * @param bool $undoIsLatest Set true if $undo is from the current revision (since 1.32)
1099          *
1100          * @return mixed Content on success, false on failure
1101          */
1102         public function getUndoContent( $current, $undo, $undoafter, $undoIsLatest = false ) {
1103                 Assert::parameterType( Revision::class . '|' . Content::class, $current, '$current' );
1104                 if ( $current instanceof Content ) {
1105                         Assert::parameter( $undo instanceof Content, '$undo',
1106                                 'Must be Content when $current is Content' );
1107                         Assert::parameter( $undoafter instanceof Content, '$undoafter',
1108                                 'Must be Content when $current is Content' );
1109                         $cur_content = $current;
1110                         $undo_content = $undo;
1111                         $undoafter_content = $undoafter;
1112                 } else {
1113                         Assert::parameter( $undo instanceof Revision, '$undo',
1114                                 'Must be Revision when $current is Revision' );
1115                         Assert::parameter( $undoafter instanceof Revision, '$undoafter',
1116                                 'Must be Revision when $current is Revision' );
1117
1118                         $cur_content = $current->getContent();
1119
1120                         if ( empty( $cur_content ) ) {
1121                                 return false; // no page
1122                         }
1123
1124                         $undo_content = $undo->getContent();
1125                         $undoafter_content = $undoafter->getContent();
1126
1127                         if ( !$undo_content || !$undoafter_content ) {
1128                                 return false; // no content to undo
1129                         }
1130
1131                         $undoIsLatest = $current->getId() === $undo->getId();
1132                 }
1133
1134                 try {
1135                         $this->checkModelID( $cur_content->getModel() );
1136                         $this->checkModelID( $undo_content->getModel() );
1137                         if ( !$undoIsLatest ) {
1138                                 // If we are undoing the most recent revision,
1139                                 // its ok to revert content model changes. However
1140                                 // if we are undoing a revision in the middle, then
1141                                 // doing that will be confusing.
1142                                 $this->checkModelID( $undoafter_content->getModel() );
1143                         }
1144                 } catch ( MWException $e ) {
1145                         // If the revisions have different content models
1146                         // just return false
1147                         return false;
1148                 }
1149
1150                 if ( $cur_content->equals( $undo_content ) ) {
1151                         // No use doing a merge if it's just a straight revert.
1152                         return $undoafter_content;
1153                 }
1154
1155                 $undone_content = $this->merge3( $undo_content, $undoafter_content, $cur_content );
1156
1157                 return $undone_content;
1158         }
1159
1160         /**
1161          * Get parser options suitable for rendering and caching the article
1162          *
1163          * @deprecated since 1.32, use WikiPage::makeParserOptions() or
1164          *  ParserOptions::newCanonical() instead.
1165          * @param IContextSource|User|string $context One of the following:
1166          *        - IContextSource: Use the User and the Language of the provided
1167          *                                            context
1168          *        - User: Use the provided User object and $wgLang for the language,
1169          *                                            so use an IContextSource object if possible.
1170          *        - 'canonical': Canonical options (anonymous user with default
1171          *                                            preferences and content language).
1172          *
1173          * @throws MWException
1174          * @return ParserOptions
1175          */
1176         public function makeParserOptions( $context ) {
1177                 wfDeprecated( __METHOD__, '1.32' );
1178                 return ParserOptions::newCanonical( $context );
1179         }
1180
1181         /**
1182          * Returns true for content models that support caching using the
1183          * ParserCache mechanism. See WikiPage::shouldCheckParserCache().
1184          *
1185          * @since 1.21
1186          *
1187          * @return bool Always false.
1188          */
1189         public function isParserCacheSupported() {
1190                 return false;
1191         }
1192
1193         /**
1194          * Returns true if this content model supports sections.
1195          * This default implementation returns false.
1196          *
1197          * Content models that return true here should also implement
1198          * Content::getSection, Content::replaceSection, etc. to handle sections..
1199          *
1200          * @return bool Always false.
1201          */
1202         public function supportsSections() {
1203                 return false;
1204         }
1205
1206         /**
1207          * Returns true if this content model supports categories.
1208          * The default implementation returns true.
1209          *
1210          * @return bool Always true.
1211          */
1212         public function supportsCategories() {
1213                 return true;
1214         }
1215
1216         /**
1217          * Returns true if this content model supports redirects.
1218          * This default implementation returns false.
1219          *
1220          * Content models that return true here should also implement
1221          * ContentHandler::makeRedirectContent to return a Content object.
1222          *
1223          * @return bool Always false.
1224          */
1225         public function supportsRedirects() {
1226                 return false;
1227         }
1228
1229         /**
1230          * Return true if this content model supports direct editing, such as via EditPage.
1231          *
1232          * @return bool Default is false, and true for TextContent and it's derivatives.
1233          */
1234         public function supportsDirectEditing() {
1235                 return false;
1236         }
1237
1238         /**
1239          * Whether or not this content model supports direct editing via ApiEditPage
1240          *
1241          * @return bool Default is false, and true for TextContent and derivatives.
1242          */
1243         public function supportsDirectApiEditing() {
1244                 return $this->supportsDirectEditing();
1245         }
1246
1247         /**
1248          * Get fields definition for search index
1249          *
1250          * @todo Expose title, redirect, namespace, text, source_text, text_bytes
1251          *       field mappings here. (see T142670 and T143409)
1252          *
1253          * @param SearchEngine $engine
1254          * @return SearchIndexField[] List of fields this content handler can provide.
1255          * @since 1.28
1256          */
1257         public function getFieldsForSearchIndex( SearchEngine $engine ) {
1258                 $fields['category'] = $engine->makeSearchFieldMapping(
1259                         'category',
1260                         SearchIndexField::INDEX_TYPE_TEXT
1261                 );
1262                 $fields['category']->setFlag( SearchIndexField::FLAG_CASEFOLD );
1263
1264                 $fields['external_link'] = $engine->makeSearchFieldMapping(
1265                         'external_link',
1266                         SearchIndexField::INDEX_TYPE_KEYWORD
1267                 );
1268
1269                 $fields['outgoing_link'] = $engine->makeSearchFieldMapping(
1270                         'outgoing_link',
1271                         SearchIndexField::INDEX_TYPE_KEYWORD
1272                 );
1273
1274                 $fields['template'] = $engine->makeSearchFieldMapping(
1275                         'template',
1276                         SearchIndexField::INDEX_TYPE_KEYWORD
1277                 );
1278                 $fields['template']->setFlag( SearchIndexField::FLAG_CASEFOLD );
1279
1280                 $fields['content_model'] = $engine->makeSearchFieldMapping(
1281                         'content_model',
1282                         SearchIndexField::INDEX_TYPE_KEYWORD
1283                 );
1284
1285                 return $fields;
1286         }
1287
1288         /**
1289          * Add new field definition to array.
1290          * @param SearchIndexField[] &$fields
1291          * @param SearchEngine $engine
1292          * @param string $name
1293          * @param int $type
1294          * @return SearchIndexField[] new field defs
1295          * @since 1.28
1296          */
1297         protected function addSearchField( &$fields, SearchEngine $engine, $name, $type ) {
1298                 $fields[$name] = $engine->makeSearchFieldMapping( $name, $type );
1299                 return $fields;
1300         }
1301
1302         /**
1303          * Return fields to be indexed by search engine
1304          * as representation of this document.
1305          * Overriding class should call parent function or take care of calling
1306          * the SearchDataForIndex hook.
1307          * @param WikiPage $page Page to index
1308          * @param ParserOutput $output
1309          * @param SearchEngine $engine Search engine for which we are indexing
1310          * @return array Map of name=>value for fields
1311          * @since 1.28
1312          */
1313         public function getDataForSearchIndex(
1314                 WikiPage $page,
1315                 ParserOutput $output,
1316                 SearchEngine $engine
1317         ) {
1318                 $fieldData = [];
1319                 $content = $page->getContent();
1320
1321                 if ( $content ) {
1322                         $searchDataExtractor = new ParserOutputSearchDataExtractor();
1323
1324                         $fieldData['category'] = $searchDataExtractor->getCategories( $output );
1325                         $fieldData['external_link'] = $searchDataExtractor->getExternalLinks( $output );
1326                         $fieldData['outgoing_link'] = $searchDataExtractor->getOutgoingLinks( $output );
1327                         $fieldData['template'] = $searchDataExtractor->getTemplates( $output );
1328
1329                         $text = $content->getTextForSearchIndex();
1330
1331                         $fieldData['text'] = $text;
1332                         $fieldData['source_text'] = $text;
1333                         $fieldData['text_bytes'] = $content->getSize();
1334                         $fieldData['content_model'] = $content->getModel();
1335                 }
1336
1337                 Hooks::run( 'SearchDataForIndex', [ &$fieldData, $this, $page, $output, $engine ] );
1338                 return $fieldData;
1339         }
1340
1341         /**
1342          * Produce page output suitable for indexing.
1343          *
1344          * Specific content handlers may override it if they need different content handling.
1345          *
1346          * @param WikiPage $page
1347          * @param ParserCache|null $cache
1348          * @return ParserOutput
1349          */
1350         public function getParserOutputForIndexing( WikiPage $page, ParserCache $cache = null ) {
1351                 // TODO: MCR: ContentHandler should be called per slot, not for the whole page.
1352                 // See T190066.
1353                 $parserOptions = $page->makeParserOptions( 'canonical' );
1354                 if ( $cache ) {
1355                         $parserOutput = $cache->get( $page, $parserOptions );
1356                 }
1357
1358                 if ( empty( $parserOutput ) ) {
1359                         $renderer = MediaWikiServices::getInstance()->getRevisionRenderer();
1360                         $parserOutput =
1361                                 $renderer->getRenderedRevision(
1362                                         $page->getRevision()->getRevisionRecord(),
1363                                         $parserOptions
1364                                 )->getRevisionParserOutput();
1365                         if ( $cache ) {
1366                                 $cache->save( $parserOutput, $page, $parserOptions );
1367                         }
1368                 }
1369                 return $parserOutput;
1370         }
1371
1372         /**
1373          * Returns a list of DeferrableUpdate objects for recording information about the
1374          * given Content in some secondary data store.
1375          *
1376          * Application logic should not call this method directly. Instead, it should call
1377          * DerivedPageDataUpdater::getSecondaryDataUpdates().
1378          *
1379          * @note Implementations must not return a LinksUpdate instance. Instead, a LinksUpdate
1380          * is created by the calling code in DerivedPageDataUpdater, on the combined ParserOutput
1381          * of all slots, not for each slot individually. This is in contrast to the old
1382          * getSecondaryDataUpdates method defined by AbstractContent, which returned a LinksUpdate.
1383          *
1384          * @note Implementations should not call $content->getParserOutput, they should call
1385          * $slotOutput->getSlotRendering( $role, false ) instead if they need to access a ParserOutput
1386          * of $content. This allows existing ParserOutput objects to be re-used, while avoiding
1387          * creating a ParserOutput when none is needed.
1388          *
1389          * @param Title $title The title of the page to supply the updates for
1390          * @param Content $content The content to generate data updates for.
1391          * @param string $role The role (slot) in which the content is being used. Which updates
1392          *        are performed should generally not depend on the role the content has, but the
1393          *        DeferrableUpdates themselves may need to know the role, to track to which slot the
1394          *        data refers, and to avoid overwriting data of the same kind from another slot.
1395          * @param SlotRenderingProvider $slotOutput A provider that can be used to gain access to
1396          *        a ParserOutput of $content by calling $slotOutput->getSlotParserOutput( $role, false ).
1397          * @return DeferrableUpdate[] A list of DeferrableUpdate objects for putting information
1398          *        about this content object somewhere. The default implementation returns an empty
1399          *        array.
1400          * @since 1.32
1401          */
1402         public function getSecondaryDataUpdates(
1403                 Title $title,
1404                 Content $content,
1405                 $role,
1406                 SlotRenderingProvider $slotOutput
1407         ) {
1408                 return [];
1409         }
1410
1411         /**
1412          * Returns a list of DeferrableUpdate objects for removing information about content
1413          * in some secondary data store. This is used when a page is deleted, and also when
1414          * a slot is removed from a page.
1415          *
1416          * Application logic should not call this method directly. Instead, it should call
1417          * WikiPage::getSecondaryDataUpdates().
1418          *
1419          * @note Implementations must not return a LinksDeletionUpdate instance. Instead, a
1420          * LinksDeletionUpdate is created by the calling code in WikiPage.
1421          * This is in contrast to the old getDeletionUpdates method defined by AbstractContent,
1422          * which returned a LinksUpdate.
1423          *
1424          * @note Implementations should not rely on the page's current content, but rather the current
1425          * state of the secondary data store.
1426          *
1427          * @param Title $title The title of the page to supply the updates for
1428          * @param string $role The role (slot) in which the content is being used. Which updates
1429          *        are performed should generally not depend on the role the content has, but the
1430          *        DeferrableUpdates themselves may need to know the role, to track to which slot the
1431          *        data refers, and to avoid overwriting data of the same kind from another slot.
1432          *
1433          * @return DeferrableUpdate[] A list of DeferrableUpdate objects for putting information
1434          *        about this content object somewhere. The default implementation returns an empty
1435          *        array.
1436          *
1437          * @since 1.32
1438          */
1439         public function getDeletionUpdates( Title $title, $role ) {
1440                 return [];
1441         }
1442
1443 }