includes/content/ContentHandler.php

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