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\Revision\SlotRenderingProvider;
  32 use MediaWiki\Search\ParserOutputSearchDataExtractor;
  33
  34 /**
  35  * A content handler knows how do deal with a specific type of content on a wiki
  36  * page. Content is stored in the database in a serialized form (using a
  37  * serialization format a.k.a. MIME type) and is unserialized into its native
  38  * PHP representation (the content model), which is wrapped in an instance of
  39  * the appropriate subclass of Content.
  40  *
  41  * ContentHandler instances are stateless singletons that serve, among other
  42  * things, as a factory for Content objects. Generally, there is one subclass
  43  * of ContentHandler and one subclass of Content for every type of content model.
  44  *
  45  * Some content types have a flat model, that is, their native representation
  46  * is the same as their serialized form. Examples would be JavaScript and CSS
  47  * code. As of now, this also applies to wikitext (MediaWiki's default content
  48  * type), but wikitext content may be represented by a DOM or AST structure in
  49  * the future.
  50  *
  51  * @ingroup Content
  52  */
  53 abstract class ContentHandler {
  54         /**
  55          * Convenience function for getting flat text from a Content object. This
  56          * should only be used in the context of backwards compatibility with code
  57          * that is not yet able to handle Content objects!
  58          *
  59          * If $content is null, this method returns the empty string.
  60          *
  61          * If $content is an instance of TextContent, this method returns the flat
  62          * text as returned by $content->getNativeData().
  63          *
  64          * If $content is not a TextContent object, the behavior of this method
  65          * depends on the global $wgContentHandlerTextFallback:
  66          * - If $wgContentHandlerTextFallback is 'fail' and $content is not a
  67          *   TextContent object, an MWException is thrown.
  68          * - If $wgContentHandlerTextFallback is 'serialize' and $content is not a
  69          *   TextContent object, $content->serialize() is called to get a string
  70          *   form of the content.
  71          * - If $wgContentHandlerTextFallback is 'ignore' and $content is not a
  72          *   TextContent object, this method returns null.
  73          * - otherwise, the behavior is undefined.
  74          *
  75          * @since 1.21
  76          *
  77          * @param Content|null $content
  78          *
  79          * @throws MWException If the content is not an instance of TextContent and
  80          * wgContentHandlerTextFallback was set to 'fail'.
  81          * @return string|null Textual form of the content, if available.
  82          */
  83         public static function getContentText( Content $content = null ) {
  84                 global $wgContentHandlerTextFallback;
  85
  86                 if ( is_null( $content ) ) {
  87                         return '';
  88                 }
  89
  90                 if ( $content instanceof TextContent ) {
  91                         return $content->getNativeData();
  92                 }
  93
  94                 wfDebugLog( 'ContentHandler', 'Accessing ' . $content->getModel() . ' content as text!' );
  95
  96                 if ( $wgContentHandlerTextFallback == 'fail' ) {
  97                         throw new MWException(
  98                                 "Attempt to get text from Content with model " .
  99                                 $content->getModel()
 100                         );
 101                 }
 102
 103                 if ( $wgContentHandlerTextFallback == 'serialize' ) {
 104                         return $content->serialize();
 105                 }
 106
 107                 return null;
 108         }
 109
 110         /**
 111          * Convenience function for creating a Content object from a given textual
 112          * representation.
 113          *
 114          * $text will be deserialized into a Content object of the model specified
 115          * by $modelId (or, if that is not given, $title->getContentModel()) using
 116          * the given format.
 117          *
 118          * @since 1.21
 119          *
 120          * @param string $text The textual representation, will be
 121          *    unserialized to create the Content object
 122          * @param Title|null $title The title of the page this text belongs to.
 123          *    Required if $modelId is not provided.
 124          * @param string|null $modelId The model to deserialize to. If not provided,
 125          *    $title->getContentModel() is used.
 126          * @param string|null $format The format to use for deserialization. If not
 127          *    given, the model's default format is used.
 128          *
 129          * @throws MWException If model ID or format is not supported or if the text can not be
 130          * unserialized using the format.
 131          * @return Content A Content object representing the text.
 132          */
 133         public static function makeContent( $text, Title $title = null,
 134                 $modelId = null, $format = null ) {
 135                 if ( is_null( $modelId ) ) {
 136                         if ( is_null( $title ) ) {
 137                                 throw new MWException( "Must provide a Title object or a content model ID." );
 138                         }
 139
 140                         $modelId = $title->getContentModel();
 141                 }
 142
 143                 $handler = self::getForModelID( $modelId );
 144
 145                 return $handler->unserializeContent( $text, $format );
 146         }
 147
 148         /**
 149          * Returns the name of the default content model to be used for the page
 150          * with the given title.
 151          *
 152          * Note: There should rarely be need to call this method directly.
 153          * To determine the actual content model for a given page, use
 154          * Title::getContentModel().
 155          *
 156          * Which model is to be used by default for the page is determined based
 157          * on several factors:
 158          * - The global setting $wgNamespaceContentModels specifies a content model
 159          *   per namespace.
 160          * - The hook ContentHandlerDefaultModelFor may be used to override the page's default
 161          *   model.
 162          * - Pages in NS_MEDIAWIKI and NS_USER default to the CSS or JavaScript
 163          *   model if they end in .js or .css, respectively.
 164          * - Pages in NS_MEDIAWIKI default to the wikitext model otherwise.
 165          * - The hook TitleIsCssOrJsPage may be used to force a page to use the CSS
 166          *   or JavaScript model. This is a compatibility feature. The ContentHandlerDefaultModelFor
 167          *   hook should be used instead if possible.
 168          * - The hook TitleIsWikitextPage may be used to force a page to use the
 169          *   wikitext model. This is a compatibility feature. The ContentHandlerDefaultModelFor
 170          *   hook should be used instead if possible.
 171          *
 172          * If none of the above applies, the wikitext model is used.
 173          *
 174          * Note: this is used by, and may thus not use, Title::getContentModel()
 175          *
 176          * @since 1.21
 177          *
 178          * @param Title $title
 179          *
 180          * @return string Default model name for the page given by $title
 181          */
 182         public static function getDefaultModelFor( Title $title ) {
 183                 // NOTE: this method must not rely on $title->getContentModel() directly or indirectly,
 184                 //       because it is used to initialize the mContentModel member.
 185
 186                 $ns = $title->getNamespace();
 187
 188                 $ext = false;
 189                 $m = null;
 190                 $model = MWNamespace::getNamespaceContentModel( $ns );
 191
 192                 // Hook can determine default model
 193                 if ( !Hooks::run( 'ContentHandlerDefaultModelFor', [ $title, &$model ] ) ) {
 194                         if ( !is_null( $model ) ) {
 195                                 return $model;
 196                         }
 197                 }
 198
 199                 // Could this page contain code based on the title?
 200                 $isCodePage = NS_MEDIAWIKI == $ns && preg_match( '!\.(css|js|json)$!u', $title->getText(), $m );
 201                 if ( $isCodePage ) {
 202                         $ext = $m[1];
 203                 }
 204
 205                 // Is this a user subpage containing code?
 206                 $isCodeSubpage = NS_USER == $ns
 207                         && !$isCodePage
 208                         && preg_match( "/\\/.*\\.(js|css|json)$/", $title->getText(), $m );
 209                 if ( $isCodeSubpage ) {
 210                         $ext = $m[1];
 211                 }
 212
 213                 // Is this wikitext, according to $wgNamespaceContentModels or the DefaultModelFor hook?
 214                 $isWikitext = is_null( $model ) || $model == CONTENT_MODEL_WIKITEXT;
 215                 $isWikitext = $isWikitext && !$isCodePage && !$isCodeSubpage;
 216
 217                 if ( !$isWikitext ) {
 218                         switch ( $ext ) {
 219                                 case 'js':
 220                                         return CONTENT_MODEL_JAVASCRIPT;
 221                                 case 'css':
 222                                         return CONTENT_MODEL_CSS;
 223                                 case 'json':
 224                                         return CONTENT_MODEL_JSON;
 225                                 default:
 226                                         return is_null( $model ) ? CONTENT_MODEL_TEXT : $model;
 227                         }
 228                 }
 229
 230                 // We established that it must be wikitext
 231
 232                 return CONTENT_MODEL_WIKITEXT;
 233         }
 234
 235         /**
 236          * Returns the appropriate ContentHandler singleton for the given title.
 237          *
 238          * @since 1.21
 239          *
 240          * @param Title $title
 241          *
 242          * @return ContentHandler
 243          */
 244         public static function getForTitle( Title $title ) {
 245                 $modelId = $title->getContentModel();
 246
 247                 return self::getForModelID( $modelId );
 248         }
 249
 250         /**
 251          * Returns the appropriate ContentHandler singleton for the given Content
 252          * object.
 253          *
 254          * @since 1.21
 255          *
 256          * @param Content $content
 257          *
 258          * @return ContentHandler
 259          */
 260         public static function getForContent( Content $content ) {
 261                 $modelId = $content->getModel();
 262
 263                 return self::getForModelID( $modelId );
 264         }
 265
 266         /**
 267          * @var array A Cache of ContentHandler instances by model id
 268          */
 269         protected static $handlers;
 270
 271         /**
 272          * Returns the ContentHandler singleton for the given model ID. Use the
 273          * CONTENT_MODEL_XXX constants to identify the desired content model.
 274          *
 275          * ContentHandler singletons are taken from the global $wgContentHandlers
 276          * array. Keys in that array are model names, the values are either
 277          * ContentHandler singleton objects, or strings specifying the appropriate
 278          * subclass of ContentHandler.
 279          *
 280          * If a class name is encountered when looking up the singleton for a given
 281          * model name, the class is instantiated and the class name is replaced by
 282          * the resulting singleton in $wgContentHandlers.
 283          *
 284          * If no ContentHandler is defined for the desired $modelId, the
 285          * ContentHandler may be provided by the ContentHandlerForModelID hook.
 286          * If no ContentHandler can be determined, an MWException is raised.
 287          *
 288          * @since 1.21
 289          *
 290          * @param string $modelId The ID of the content model for which to get a
 291          *    handler. Use CONTENT_MODEL_XXX constants.
 292          *
 293          * @throws MWException For internal errors and problems in the configuration.
 294          * @throws MWUnknownContentModelException If no handler is known for the model ID.
 295          * @return ContentHandler The ContentHandler singleton for handling the model given by the ID.
 296          */
 297         public static function getForModelID( $modelId ) {
 298                 global $wgContentHandlers;
 299
 300                 if ( isset( self::$handlers[$modelId] ) ) {
 301                         return self::$handlers[$modelId];
 302                 }
 303
 304                 if ( empty( $wgContentHandlers[$modelId] ) ) {
 305                         $handler = null;
 306
 307                         Hooks::run( 'ContentHandlerForModelID', [ $modelId, &$handler ] );
 308
 309                         if ( $handler === null ) {
 310                                 throw new MWUnknownContentModelException( $modelId );
 311                         }
 312
 313                         if ( !( $handler instanceof ContentHandler ) ) {
 314                                 throw new MWException( "ContentHandlerForModelID must supply a ContentHandler instance" );
 315                         }
 316                 } else {
 317                         $classOrCallback = $wgContentHandlers[$modelId];
 318
 319                         if ( is_callable( $classOrCallback ) ) {
 320                                 $handler = call_user_func( $classOrCallback, $modelId );
 321                         } else {
 322                                 $handler = new $classOrCallback( $modelId );
 323                         }
 324
 325                         if ( !( $handler instanceof ContentHandler ) ) {
 326                                 throw new MWException( "$classOrCallback from \$wgContentHandlers is not " .
 327                                         "compatible with ContentHandler" );
 328                         }
 329                 }
 330
 331                 wfDebugLog( 'ContentHandler', 'Created handler for ' . $modelId
 332                         . ': ' . get_class( $handler ) );
 333
 334                 self::$handlers[$modelId] = $handler;
 335
 336                 return self::$handlers[$modelId];
 337         }
 338
 339         /**
 340          * Clean up handlers cache.
 341          */
 342         public static function cleanupHandlersCache() {
 343                 self::$handlers = [];
 344         }
 345
 346         /**
 347          * Returns the localized name for a given content model.
 348          *
 349          * Model names are localized using system messages. Message keys
 350          * have the form content-model-$name, where $name is getContentModelName( $id ).
 351          *
 352          * @param string $name The content model ID, as given by a CONTENT_MODEL_XXX
 353          *    constant or returned by Revision::getContentModel().
 354          * @param Language|null $lang The language to parse the message in (since 1.26)
 355          *
 356          * @throws MWException If the model ID isn't known.
 357          * @return string The content model's localized name.
 358          */
 359         public static function getLocalizedName( $name, Language $lang = null ) {
 360                 // Messages: content-model-wikitext, content-model-text,
 361                 // content-model-javascript, content-model-css
 362                 $key = "content-model-$name";
 363
 364                 $msg = wfMessage( $key );
 365                 if ( $lang ) {
 366                         $msg->inLanguage( $lang );
 367                 }
 368
 369                 return $msg->exists() ? $msg->plain() : $name;
 370         }
 371
 372         public static function getContentModels() {
 373                 global $wgContentHandlers;
 374
 375                 $models = array_keys( $wgContentHandlers );
 376                 Hooks::run( 'GetContentModels', [ &$models ] );
 377                 return $models;
 378         }
 379
 380         public static function getAllContentFormats() {
 381                 global $wgContentHandlers;
 382
 383                 $formats = [];
 384
 385                 foreach ( $wgContentHandlers as $model => $class ) {
 386                         $handler = self::getForModelID( $model );
 387                         $formats = array_merge( $formats, $handler->getSupportedFormats() );
 388                 }
 389
 390                 $formats = array_unique( $formats );
 391
 392                 return $formats;
 393         }
 394
 395         // ------------------------------------------------------------------------
 396
 397         /**
 398          * @var string
 399          */
 400         protected $mModelID;
 401
 402         /**
 403          * @var string[]
 404          */
 405         protected $mSupportedFormats;
 406
 407         /**
 408          * Constructor, initializing the ContentHandler instance with its model ID
 409          * and a list of supported formats. Values for the parameters are typically
 410          * provided as literals by subclass's constructors.
 411          *
 412          * @param string $modelId (use CONTENT_MODEL_XXX constants).
 413          * @param string[] $formats List for supported serialization formats
 414          *    (typically as MIME types)
 415          */
 416         public function __construct( $modelId, $formats ) {
 417                 $this->mModelID = $modelId;
 418                 $this->mSupportedFormats = $formats;
 419         }
 420
 421         /**
 422          * Serializes a Content object of the type supported by this ContentHandler.
 423          *
 424          * @since 1.21
 425          *
 426          * @param Content $content The Content object to serialize
 427          * @param string|null $format The desired serialization format
 428          *
 429          * @return string Serialized form of the content
 430          */
 431         abstract public function serializeContent( Content $content, $format = null );
 432
 433         /**
 434          * Applies transformations on export (returns the blob unchanged per default).
 435          * Subclasses may override this to perform transformations such as conversion
 436          * of legacy formats or filtering of internal meta-data.
 437          *
 438          * @param string $blob The blob to be exported
 439          * @param string|null $format The blob's serialization format
 440          *
 441          * @return string
 442          */
 443         public function exportTransform( $blob, $format = null ) {
 444                 return $blob;
 445         }
 446
 447         /**
 448          * Unserializes a Content object of the type supported by this ContentHandler.
 449          *
 450          * @since 1.21
 451          *
 452          * @param string $blob Serialized form of the content
 453          * @param string|null $format The format used for serialization
 454          *
 455          * @return Content The Content object created by deserializing $blob
 456          */
 457         abstract public function unserializeContent( $blob, $format = null );
 458
 459         /**
 460          * Apply import transformation (per default, returns $blob unchanged).
 461          * This gives subclasses an opportunity to transform data blobs on import.
 462          *
 463          * @since 1.24
 464          *
 465          * @param string $blob
 466          * @param string|null $format
 467          *
 468          * @return string
 469          */
 470         public function importTransform( $blob, $format = null ) {
 471                 return $blob;
 472         }
 473
 474         /**
 475          * Creates an empty Content object of the type supported by this
 476          * ContentHandler.
 477          *
 478          * @since 1.21
 479          *
 480          * @return Content
 481          */
 482         abstract public function makeEmptyContent();
 483
 484         /**
 485          * Creates a new Content object that acts as a redirect to the given page,
 486          * or null if redirects are not supported by this content model.
 487          *
 488          * This default implementation always returns null. Subclasses supporting redirects
 489          * must override this method.
 490          *
 491          * Note that subclasses that override this method to return a Content object
 492          * should also override supportsRedirects() to return true.
 493          *
 494          * @since 1.21
 495          *
 496          * @param Title $destination The page to redirect to.
 497          * @param string $text Text to include in the redirect, if possible.
 498          *
 499          * @return Content Always null.
 500          */
 501         public function makeRedirectContent( Title $destination, $text = '' ) {
 502                 return null;
 503         }
 504
 505         /**
 506          * Returns the model id that identifies the content model this
 507          * ContentHandler can handle. Use with the CONTENT_MODEL_XXX constants.
 508          *
 509          * @since 1.21
 510          *
 511          * @return string The model ID
 512          */
 513         public function getModelID() {
 514                 return $this->mModelID;
 515         }
 516
 517         /**
 518          * @since 1.21
 519          *
 520          * @param string $model_id The model to check
 521          *
 522          * @throws MWException If the model ID is not the ID of the content model supported by this
 523          * ContentHandler.
 524          */
 525         protected function checkModelID( $model_id ) {
 526                 if ( $model_id !== $this->mModelID ) {
 527                         throw new MWException( "Bad content model: " .
 528                                 "expected {$this->mModelID} " .
 529                                 "but got $model_id." );
 530                 }
 531         }
 532
 533         /**
 534          * Returns a list of serialization formats supported by the
 535          * serializeContent() and unserializeContent() methods of this
 536          * ContentHandler.
 537          *
 538          * @since 1.21
 539          *
 540          * @return string[] List of serialization formats as MIME type like strings
 541          */
 542         public function getSupportedFormats() {
 543                 return $this->mSupportedFormats;
 544         }
 545
 546         /**
 547          * The format used for serialization/deserialization by default by this
 548          * ContentHandler.
 549          *
 550          * This default implementation will return the first element of the array
 551          * of formats that was passed to the constructor.
 552          *
 553          * @since 1.21
 554          *
 555          * @return string The name of the default serialization format as a MIME type
 556          */
 557         public function getDefaultFormat() {
 558                 return $this->mSupportedFormats[0];
 559         }
 560
 561         /**
 562          * Returns true if $format is a serialization format supported by this
 563          * ContentHandler, and false otherwise.
 564          *
 565          * Note that if $format is null, this method always returns true, because
 566          * null means "use the default format".
 567          *
 568          * @since 1.21
 569          *
 570          * @param string $format The serialization format to check
 571          *
 572          * @return bool
 573          */
 574         public function isSupportedFormat( $format ) {
 575                 if ( !$format ) {
 576                         return true; // this means "use the default"
 577                 }
 578
 579                 return in_array( $format, $this->mSupportedFormats );
 580         }
 581
 582         /**
 583          * Convenient for checking whether a format provided as a parameter is actually supported.
 584          *
 585          * @param string $format The serialization format to check
 586          *
 587          * @throws MWException If the format is not supported by this content handler.
 588          */
 589         protected function checkFormat( $format ) {
 590                 if ( !$this->isSupportedFormat( $format ) ) {
 591                         throw new MWException(
 592                                 "Format $format is not supported for content model "
 593                                 . $this->getModelID()
 594                         );
 595                 }
 596         }
 597
 598         /**
 599          * Returns overrides for action handlers.
 600          * Classes listed here will be used instead of the default one when
 601          * (and only when) $wgActions[$action] === true. This allows subclasses
 602          * to override the default action handlers.
 603          *
 604          * @since 1.21
 605          *
 606          * @return array An array mapping action names (typically "view", "edit", "history" etc.) to
 607          *  either the full qualified class name of an Action class, a callable taking ( Page $page,
 608          *  IContextSource $context = null ) as parameters and returning an Action object, or an actual
 609          *  Action object. An empty array in this default implementation.
 610          *
 611          * @see Action::factory
 612          */
 613         public function getActionOverrides() {
 614                 return [];
 615         }
 616
 617         /**
 618          * Factory for creating an appropriate DifferenceEngine for this content model.
 619          * Since 1.32, this is only used for page-level diffs; to diff two content objects,
 620          * use getSlotDiffRenderer.
 621          *
 622          * The DifferenceEngine subclass to use is selected in getDiffEngineClass(). The
 623          * GetDifferenceEngine hook will receive the DifferenceEngine object and can replace or
 624          * wrap it.
 625          * (Note that in older versions of MediaWiki the hook documentation instructed extensions
 626          * to return false from the hook; you should not rely on always being able to decorate
 627          * the DifferenceEngine instance from the hook. If the owner of the content type wants to
 628          * decorare the instance, overriding this method is a safer approach.)
 629          *
 630          * @todo This is page-level functionality so it should not belong to ContentHandler.
 631          *   Move it to a better place once one exists (e.g. PageTypeHandler).
 632          *
 633          * @since 1.21
 634          *
 635          * @param IContextSource $context Context to use, anything else will be ignored.
 636          * @param int $old Revision ID we want to show and diff with.
 637          * @param int|string $new Either a revision ID or one of the strings 'cur', 'prev' or 'next'.
 638          * @param int $rcid FIXME: Deprecated, no longer used. Defaults to 0.
 639          * @param bool $refreshCache If set, refreshes the diff cache. Defaults to false.
 640          * @param bool $unhide If set, allow viewing deleted revs. Defaults to false.
 641          *
 642          * @return DifferenceEngine
 643          */
 644         public function createDifferenceEngine( IContextSource $context, $old = 0, $new = 0,
 645                 $rcid = 0, // FIXME: Deprecated, no longer used
 646                 $refreshCache = false, $unhide = false
 647         ) {
 648                 $diffEngineClass = $this->getDiffEngineClass();
 649                 $differenceEngine = new $diffEngineClass( $context, $old, $new, $rcid, $refreshCache, $unhide );
 650                 Hooks::run( 'GetDifferenceEngine', [ $context, $old, $new, $refreshCache, $unhide,
 651                         &$differenceEngine ] );
 652                 return $differenceEngine;
 653         }
 654
 655         /**
 656          * Get an appropriate SlotDiffRenderer for this content model.
 657          * @since 1.32
 658          * @param IContextSource $context
 659          * @return SlotDiffRenderer
 660          */
 661         final public function getSlotDiffRenderer( IContextSource $context ) {
 662                 $slotDiffRenderer = $this->getSlotDiffRendererInternal( $context );
 663                 if ( get_class( $slotDiffRenderer ) === TextSlotDiffRenderer::class ) {
 664                         //  To keep B/C, when SlotDiffRenderer is not overridden for a given content type
 665                         // but DifferenceEngine is, use that instead.
 666                         $differenceEngine = $this->createDifferenceEngine( $context );
 667                         if ( get_class( $differenceEngine ) !== DifferenceEngine::class ) {
 668                                 // TODO turn this into a deprecation warning in a later release
 669                                 LoggerFactory::getInstance( 'diff' )->info(
 670                                         'Falling back to DifferenceEngineSlotDiffRenderer', [
 671                                                 'modelID' => $this->getModelID(),
 672                                                 'DifferenceEngine' => get_class( $differenceEngine ),
 673                                         ] );
 674                                 $slotDiffRenderer = new DifferenceEngineSlotDiffRenderer( $differenceEngine );
 675                         }
 676                 }
 677                 Hooks::run( 'GetSlotDiffRenderer', [ $this, &$slotDiffRenderer, $context ] );
 678                 return $slotDiffRenderer;
 679         }
 680
 681         /**
 682          * Return the SlotDiffRenderer appropriate for this content handler.
 683          * @param IContextSource $context
 684          * @return SlotDiffRenderer
 685          */
 686         protected function getSlotDiffRendererInternal( IContextSource $context ) {
 687                 $contentLanguage = MediaWikiServices::getInstance()->getContentLanguage();
 688                 $statsdDataFactory = MediaWikiServices::getInstance()->getStatsdDataFactory();
 689                 $slotDiffRenderer = new TextSlotDiffRenderer();
 690                 $slotDiffRenderer->setStatsdDataFactory( $statsdDataFactory );
 691                 // XXX using the page language would be better, but it's unclear how that should be injected
 692                 $slotDiffRenderer->setLanguage( $contentLanguage );
 693                 $slotDiffRenderer->setWikiDiff2MovedParagraphDetectionCutoff(
 694                         $context->getConfig()->get( 'WikiDiff2MovedParagraphDetectionCutoff' )
 695                 );
 696
 697                 $engine = DifferenceEngine::getEngine();
 698                 if ( $engine === false ) {
 699                         $slotDiffRenderer->setEngine( TextSlotDiffRenderer::ENGINE_PHP );
 700                 } elseif ( $engine === 'wikidiff2' ) {
 701                         $slotDiffRenderer->setEngine( TextSlotDiffRenderer::ENGINE_WIKIDIFF2 );
 702                 } else {
 703                         $slotDiffRenderer->setEngine( TextSlotDiffRenderer::ENGINE_EXTERNAL, $engine );
 704                 }
 705
 706                 return $slotDiffRenderer;
 707         }
 708
 709         /**
 710          * Get the language in which the content of the given page is written.
 711          *
 712          * This default implementation just returns the content language (except for pages
 713          * in the MediaWiki namespace)
 714          *
 715          * Note that the pages language is not cacheable, since it may in some
 716          * cases depend on user settings.
 717          *
 718          * Also note that the page language may or may not depend on the actual content of the page,
 719          * that is, this method may load the content in order to determine the language.
 720          *
 721          * @since 1.21
 722          *
 723          * @param Title $title The page to determine the language for.
 724          * @param Content|null $content The page's content, if you have it handy, to avoid reloading it.
 725          *
 726          * @return Language The page's language
 727          */
 728         public function getPageLanguage( Title $title, Content $content = null ) {
 729                 global $wgLang;
 730                 $pageLang = MediaWikiServices::getInstance()->getContentLanguage();
 731
 732                 if ( $title->getNamespace() == NS_MEDIAWIKI ) {
 733                         // Parse mediawiki messages with correct target language
 734                         list( /* $unused */, $lang ) = MessageCache::singleton()->figureMessage( $title->getText() );
 735                         $pageLang = Language::factory( $lang );
 736                 }
 737
 738                 Hooks::run( 'PageContentLanguage', [ $title, &$pageLang, $wgLang ] );
 739
 740                 return wfGetLangObj( $pageLang );
 741         }
 742
 743         /**
 744          * Get the language in which the content of this page is written when
 745          * viewed by user. Defaults to $this->getPageLanguage(), but if the user
 746          * specified a preferred variant, the variant will be used.
 747          *
 748          * This default implementation just returns $this->getPageLanguage( $title, $content ) unless
 749          * the user specified a preferred variant.
 750          *
 751          * Note that the pages view language is not cacheable, since it depends on user settings.
 752          *
 753          * Also note that the page language may or may not depend on the actual content of the page,
 754          * that is, this method may load the content in order to determine the language.
 755          *
 756          * @since 1.21
 757          *
 758          * @param Title $title The page to determine the language for.
 759          * @param Content|null $content The page's content, if you have it handy, to avoid reloading it.
 760          *
 761          * @return Language The page's language for viewing
 762          */
 763         public function getPageViewLanguage( Title $title, Content $content = null ) {
 764                 $pageLang = $this->getPageLanguage( $title, $content );
 765
 766                 if ( $title->getNamespace() !== NS_MEDIAWIKI ) {
 767                         // If the user chooses a variant, the content is actually
 768                         // in a language whose code is the variant code.
 769                         $variant = $pageLang->getPreferredVariant();
 770                         if ( $pageLang->getCode() !== $variant ) {
 771                                 $pageLang = Language::factory( $variant );
 772                         }
 773                 }
 774
 775                 return $pageLang;
 776         }
 777
 778         /**
 779          * Determines whether the content type handled by this ContentHandler
 780          * can be used on the given page.
 781          *
 782          * This default implementation always returns true.
 783          * Subclasses may override this to restrict the use of this content model to specific locations,
 784          * typically based on the namespace or some other aspect of the title, such as a special suffix
 785          * (e.g. ".svg" for SVG content).
 786          *
 787          * @note this calls the ContentHandlerCanBeUsedOn hook which may be used to override which
 788          * content model can be used where.
 789          *
 790          * @param Title $title The page's title.
 791          *
 792          * @return bool True if content of this kind can be used on the given page, false otherwise.
 793          */
 794         public function canBeUsedOn( Title $title ) {
 795                 $ok = true;
 796
 797                 Hooks::run( 'ContentModelCanBeUsedOn', [ $this->getModelID(), $title, &$ok ] );
 798
 799                 return $ok;
 800         }
 801
 802         /**
 803          * Returns the name of the diff engine to use.
 804          *
 805          * @since 1.21
 806          *
 807          * @return string
 808          */
 809         protected function getDiffEngineClass() {
 810                 return DifferenceEngine::class;
 811         }
 812
 813         /**
 814          * Attempts to merge differences between three versions. Returns a new
 815          * Content object for a clean merge and false for failure or a conflict.
 816          *
 817          * This default implementation always returns false.
 818          *
 819          * @since 1.21
 820          *
 821          * @param Content $oldContent The page's previous content.
 822          * @param Content $myContent One of the page's conflicting contents.
 823          * @param Content $yourContent One of the page's conflicting contents.
 824          *
 825          * @return Content|bool Always false.
 826          */
 827         public function merge3( Content $oldContent, Content $myContent, Content $yourContent ) {
 828                 return false;
 829         }
 830
 831         /**
 832          * Return type of change if one exists for the given edit.
 833          *
 834          * @since 1.31
 835          *
 836          * @param Content|null $oldContent The previous text of the page.
 837          * @param Content|null $newContent The submitted text of the page.
 838          * @param int $flags Bit mask: a bit mask of flags submitted for the edit.
 839          *
 840          * @return string|null String key representing type of change, or null.
 841          */
 842         private function getChangeType(
 843                 Content $oldContent = null,
 844                 Content $newContent = null,
 845                 $flags = 0
 846         ) {
 847                 $oldTarget = $oldContent !== null ? $oldContent->getRedirectTarget() : null;
 848                 $newTarget = $newContent !== null ? $newContent->getRedirectTarget() : null;
 849
 850                 // We check for the type of change in the given edit, and return string key accordingly
 851
 852                 // Blanking of a page
 853                 if ( $oldContent && $oldContent->getSize() > 0 &&
 854                         $newContent && $newContent->getSize() === 0
 855                 ) {
 856                         return 'blank';
 857                 }
 858
 859                 // Redirects
 860                 if ( $newTarget ) {
 861                         if ( !$oldTarget ) {
 862                                 // New redirect page (by creating new page or by changing content page)
 863                                 return 'new-redirect';
 864                         } elseif ( !$newTarget->equals( $oldTarget ) ||
 865                                 $oldTarget->getFragment() !== $newTarget->getFragment()
 866                         ) {
 867                                 // Redirect target changed
 868                                 return 'changed-redirect-target';
 869                         }
 870                 } elseif ( $oldTarget ) {
 871                         // Changing an existing redirect into a non-redirect
 872                         return 'removed-redirect';
 873                 }
 874
 875                 // New page created
 876                 if ( $flags & EDIT_NEW && $newContent ) {
 877                         if ( $newContent->getSize() === 0 ) {
 878                                 // New blank page
 879                                 return 'newblank';
 880                         } else {
 881                                 return 'newpage';
 882                         }
 883                 }
 884
 885                 // Removing more than 90% of the page
 886                 if ( $oldContent && $newContent && $oldContent->getSize() > 10 * $newContent->getSize() ) {
 887                         return 'replace';
 888                 }
 889
 890                 // Content model changed
 891                 if ( $oldContent && $newContent && $oldContent->getModel() !== $newContent->getModel() ) {
 892                         return 'contentmodelchange';
 893                 }
 894
 895                 return null;
 896         }
 897
 898         /**
 899          * Return an applicable auto-summary if one exists for the given edit.
 900          *
 901          * @since 1.21
 902          *
 903          * @param Content|null $oldContent The previous text of the page.
 904          * @param Content|null $newContent The submitted text of the page.
 905          * @param int $flags Bit mask: a bit mask of flags submitted for the edit.
 906          *
 907          * @return string An appropriate auto-summary, or an empty string.
 908          */
 909         public function getAutosummary(
 910                 Content $oldContent = null,
 911                 Content $newContent = null,
 912                 $flags = 0
 913         ) {
 914                 $changeType = $this->getChangeType( $oldContent, $newContent, $flags );
 915
 916                 // There's no applicable auto-summary for our case, so our auto-summary is empty.
 917                 if ( !$changeType ) {
 918                         return '';
 919                 }
 920
 921                 // Decide what kind of auto-summary is needed.
 922                 switch ( $changeType ) {
 923                         case 'new-redirect':
 924                                 $newTarget = $newContent->getRedirectTarget();
 925                                 $truncatedtext = $newContent->getTextForSummary(
 926                                         250
 927                                         - strlen( wfMessage( 'autoredircomment' )->inContentLanguage()->text() )
 928                                         - strlen( $newTarget->getFullText() )
 929                                 );
 930
 931                                 return wfMessage( 'autoredircomment', $newTarget->getFullText() )
 932                                         ->plaintextParams( $truncatedtext )->inContentLanguage()->text();
 933                         case 'changed-redirect-target':
 934                                 $oldTarget = $oldContent->getRedirectTarget();
 935                                 $newTarget = $newContent->getRedirectTarget();
 936
 937                                 $truncatedtext = $newContent->getTextForSummary(
 938                                         250
 939                                         - strlen( wfMessage( 'autosumm-changed-redirect-target' )
 940                                                 ->inContentLanguage()->text() )
 941                                         - strlen( $oldTarget->getFullText() )
 942                                         - strlen( $newTarget->getFullText() )
 943                                 );
 944
 945                                 return wfMessage( 'autosumm-changed-redirect-target',
 946                                                 $oldTarget->getFullText(),
 947                                                 $newTarget->getFullText() )
 948                                         ->rawParams( $truncatedtext )->inContentLanguage()->text();
 949                         case 'removed-redirect':
 950                                 $oldTarget = $oldContent->getRedirectTarget();
 951                                 $truncatedtext = $newContent->getTextForSummary(
 952                                         250
 953                                         - strlen( wfMessage( 'autosumm-removed-redirect' )
 954                                                 ->inContentLanguage()->text() )
 955                                         - strlen( $oldTarget->getFullText() ) );
 956
 957                                 return wfMessage( 'autosumm-removed-redirect', $oldTarget->getFullText() )
 958                                         ->rawParams( $truncatedtext )->inContentLanguage()->text();
 959                         case 'newpage':
 960                                 // If they're making a new article, give its text, truncated, in the summary.
 961                                 $truncatedtext = $newContent->getTextForSummary(
 962                                         200 - strlen( wfMessage( 'autosumm-new' )->inContentLanguage()->text() ) );
 963
 964                                 return wfMessage( 'autosumm-new' )->rawParams( $truncatedtext )
 965                                         ->inContentLanguage()->text();
 966                         case 'blank':
 967                                 return wfMessage( 'autosumm-blank' )->inContentLanguage()->text();
 968                         case 'replace':
 969                                 $truncatedtext = $newContent->getTextForSummary(
 970                                         200 - strlen( wfMessage( 'autosumm-replace' )->inContentLanguage()->text() ) );
 971
 972                                 return wfMessage( 'autosumm-replace' )->rawParams( $truncatedtext )
 973                                         ->inContentLanguage()->text();
 974                         case 'newblank':
 975                                 return wfMessage( 'autosumm-newblank' )->inContentLanguage()->text();
 976                         default:
 977                                 return '';
 978                 }
 979         }
 980
 981         /**
 982          * Return an applicable tag if one exists for the given edit or return null.
 983          *
 984          * @since 1.31
 985          *
 986          * @param Content|null $oldContent The previous text of the page.
 987          * @param Content|null $newContent The submitted text of the page.
 988          * @param int $flags Bit mask: a bit mask of flags submitted for the edit.
 989          *
 990          * @return string|null An appropriate tag, or null.
 991          */
 992         public function getChangeTag(
 993                 Content $oldContent = null,
 994                 Content $newContent = null,
 995                 $flags = 0
 996         ) {
 997                 $changeType = $this->getChangeType( $oldContent, $newContent, $flags );
 998
 999                 // There's no applicable tag for this change.
1000                 if ( !$changeType ) {
1001                         return null;
1002                 }
1003
1004                 // Core tags use the same keys as ones returned from $this->getChangeType()
1005                 // but prefixed with pseudo namespace 'mw-', so we add the prefix before checking
1006                 // if this type of change should be tagged
1007                 $tag = 'mw-' . $changeType;
1008
1009                 // Not all change types are tagged, so we check against the list of defined tags.
1010                 if ( in_array( $tag, ChangeTags::getSoftwareTags() ) ) {
1011                         return $tag;
1012                 }
1013
1014                 return null;
1015         }
1016
1017         /**
1018          * Auto-generates a deletion reason
1019          *
1020          * @since 1.21
1021          *
1022          * @param Title $title The page's title
1023          * @param bool &$hasHistory Whether the page has a history
1024          *
1025          * @return mixed String containing deletion reason or empty string, or
1026          *    boolean false if no revision occurred
1027          *
1028          * @todo &$hasHistory is extremely ugly, it's here because
1029          * WikiPage::getAutoDeleteReason() and Article::generateReason()
1030          * have it / want it.
1031          */
1032         public function getAutoDeleteReason( Title $title, &$hasHistory ) {
1033                 $dbr = wfGetDB( DB_REPLICA );
1034
1035                 // Get the last revision
1036                 $rev = Revision::newFromTitle( $title );
1037
1038                 if ( is_null( $rev ) ) {
1039                         return false;
1040                 }
1041
1042                 // Get the article's contents
1043                 $content = $rev->getContent();
1044                 $blank = false;
1045
1046                 // If the page is blank, use the text from the previous revision,
1047                 // which can only be blank if there's a move/import/protect dummy
1048                 // revision involved
1049                 if ( !$content || $content->isEmpty() ) {
1050                         $prev = $rev->getPrevious();
1051
1052                         if ( $prev ) {
1053                                 $rev = $prev;
1054                                 $content = $rev->getContent();
1055                                 $blank = true;
1056                         }
1057                 }
1058
1059                 $this->checkModelID( $rev->getContentModel() );
1060
1061                 // Find out if there was only one contributor
1062                 // Only scan the last 20 revisions
1063                 $revQuery = Revision::getQueryInfo();
1064                 $res = $dbr->select(
1065                         $revQuery['tables'],
1066                         [ 'rev_user_text' => $revQuery['fields']['rev_user_text'] ],
1067                         [
1068                                 'rev_page' => $title->getArticleID(),
1069                                 $dbr->bitAnd( 'rev_deleted', Revision::DELETED_USER ) . ' = 0'
1070                         ],
1071                         __METHOD__,
1072                         [ 'LIMIT' => 20 ],
1073                         $revQuery['joins']
1074                 );
1075
1076                 if ( $res === false ) {
1077                         // This page has no revisions, which is very weird
1078                         return false;
1079                 }
1080
1081                 $hasHistory = ( $res->numRows() > 1 );
1082                 $row = $dbr->fetchObject( $res );
1083
1084                 if ( $row ) { // $row is false if the only contributor is hidden
1085                         $onlyAuthor = $row->rev_user_text;
1086                         // Try to find a second contributor
1087                         foreach ( $res as $row ) {
1088                                 if ( $row->rev_user_text != $onlyAuthor ) { // T24999
1089                                         $onlyAuthor = false;
1090                                         break;
1091                                 }
1092                         }
1093                 } else {
1094                         $onlyAuthor = false;
1095                 }
1096
1097                 // Generate the summary with a '$1' placeholder
1098                 if ( $blank ) {
1099                         // The current revision is blank and the one before is also
1100                         // blank. It's just not our lucky day
1101                         $reason = wfMessage( 'exbeforeblank', '$1' )->inContentLanguage()->text();
1102                 } else {
1103                         if ( $onlyAuthor ) {
1104                                 $reason = wfMessage(
1105                                         'excontentauthor',
1106                                         '$1',
1107                                         $onlyAuthor
1108                                 )->inContentLanguage()->text();
1109                         } else {
1110                                 $reason = wfMessage( 'excontent', '$1' )->inContentLanguage()->text();
1111                         }
1112                 }
1113
1114                 if ( $reason == '-' ) {
1115                         // Allow these UI messages to be blanked out cleanly
1116                         return '';
1117                 }
1118
1119                 // Max content length = max comment length - length of the comment (excl. $1)
1120                 $text = $content ? $content->getTextForSummary( 255 - ( strlen( $reason ) - 2 ) ) : '';
1121
1122                 // Now replace the '$1' placeholder
1123                 $reason = str_replace( '$1', $text, $reason );
1124
1125                 return $reason;
1126         }
1127
1128         /**
1129          * Get the Content object that needs to be saved in order to undo all revisions
1130          * between $undo and $undoafter. Revisions must belong to the same page,
1131          * must exist and must not be deleted.
1132          *
1133          * @since 1.21
1134          * @since 1.32 accepts Content objects for all parameters instead of Revision objects.
1135          *  Passing Revision objects is deprecated.
1136          *
1137          * @param Revision|Content $current The current text
1138          * @param Revision|Content $undo The content of the revision to undo
1139          * @param Revision|Content $undoafter Must be from an earlier revision than $undo
1140          * @param bool $undoIsLatest Set true if $undo is from the current revision (since 1.32)
1141          *
1142          * @return mixed Content on success, false on failure
1143          */
1144         public function getUndoContent( $current, $undo, $undoafter, $undoIsLatest = false ) {
1145                 Assert::parameterType( Revision::class . '|' . Content::class, $current, '$current' );
1146                 if ( $current instanceof Content ) {
1147                         Assert::parameter( $undo instanceof Content, '$undo',
1148                                 'Must be Content when $current is Content' );
1149                         Assert::parameter( $undoafter instanceof Content, '$undoafter',
1150                                 'Must be Content when $current is Content' );
1151                         $cur_content = $current;
1152                         $undo_content = $undo;
1153                         $undoafter_content = $undoafter;
1154                 } else {
1155                         Assert::parameter( $undo instanceof Revision, '$undo',
1156                                 'Must be Revision when $current is Revision' );
1157                         Assert::parameter( $undoafter instanceof Revision, '$undoafter',
1158                                 'Must be Revision when $current is Revision' );
1159
1160                         $cur_content = $current->getContent();
1161
1162                         if ( empty( $cur_content ) ) {
1163                                 return false; // no page
1164                         }
1165
1166                         $undo_content = $undo->getContent();
1167                         $undoafter_content = $undoafter->getContent();
1168
1169                         if ( !$undo_content || !$undoafter_content ) {
1170                                 return false; // no content to undo
1171                         }
1172
1173                         $undoIsLatest = $current->getId() === $undo->getId();
1174                 }
1175
1176                 try {
1177                         $this->checkModelID( $cur_content->getModel() );
1178                         $this->checkModelID( $undo_content->getModel() );
1179                         if ( !$undoIsLatest ) {
1180                                 // If we are undoing the most recent revision,
1181                                 // its ok to revert content model changes. However
1182                                 // if we are undoing a revision in the middle, then
1183                                 // doing that will be confusing.
1184                                 $this->checkModelID( $undoafter_content->getModel() );
1185                         }
1186                 } catch ( MWException $e ) {
1187                         // If the revisions have different content models
1188                         // just return false
1189                         return false;
1190                 }
1191
1192                 if ( $cur_content->equals( $undo_content ) ) {
1193                         // No use doing a merge if it's just a straight revert.
1194                         return $undoafter_content;
1195                 }
1196
1197                 $undone_content = $this->merge3( $undo_content, $undoafter_content, $cur_content );
1198
1199                 return $undone_content;
1200         }
1201
1202         /**
1203          * Get parser options suitable for rendering and caching the article
1204          *
1205          * @deprecated since 1.32, use WikiPage::makeParserOptions() or
1206          *  ParserOptions::newCanonical() instead.
1207          * @param IContextSource|User|string $context One of the following:
1208          *        - IContextSource: Use the User and the Language of the provided
1209          *                                            context
1210          *        - User: Use the provided User object and $wgLang for the language,
1211          *                                            so use an IContextSource object if possible.
1212          *        - 'canonical': Canonical options (anonymous user with default
1213          *                                            preferences and content language).
1214          *
1215          * @throws MWException
1216          * @return ParserOptions
1217          */
1218         public function makeParserOptions( $context ) {
1219                 wfDeprecated( __METHOD__, '1.32' );
1220                 return ParserOptions::newCanonical( $context );
1221         }
1222
1223         /**
1224          * Returns true for content models that support caching using the
1225          * ParserCache mechanism. See WikiPage::shouldCheckParserCache().
1226          *
1227          * @since 1.21
1228          *
1229          * @return bool Always false.
1230          */
1231         public function isParserCacheSupported() {
1232                 return false;
1233         }
1234
1235         /**
1236          * Returns true if this content model supports sections.
1237          * This default implementation returns false.
1238          *
1239          * Content models that return true here should also implement
1240          * Content::getSection, Content::replaceSection, etc. to handle sections..
1241          *
1242          * @return bool Always false.
1243          */
1244         public function supportsSections() {
1245                 return false;
1246         }
1247
1248         /**
1249          * Returns true if this content model supports categories.
1250          * The default implementation returns true.
1251          *
1252          * @return bool Always true.
1253          */
1254         public function supportsCategories() {
1255                 return true;
1256         }
1257
1258         /**
1259          * Returns true if this content model supports redirects.
1260          * This default implementation returns false.
1261          *
1262          * Content models that return true here should also implement
1263          * ContentHandler::makeRedirectContent to return a Content object.
1264          *
1265          * @return bool Always false.
1266          */
1267         public function supportsRedirects() {
1268                 return false;
1269         }
1270
1271         /**
1272          * Return true if this content model supports direct editing, such as via EditPage.
1273          *
1274          * @return bool Default is false, and true for TextContent and it's derivatives.
1275          */
1276         public function supportsDirectEditing() {
1277                 return false;
1278         }
1279
1280         /**
1281          * Whether or not this content model supports direct editing via ApiEditPage
1282          *
1283          * @return bool Default is false, and true for TextContent and derivatives.
1284          */
1285         public function supportsDirectApiEditing() {
1286                 return $this->supportsDirectEditing();
1287         }
1288
1289         /**
1290          * Get fields definition for search index
1291          *
1292          * @todo Expose title, redirect, namespace, text, source_text, text_bytes
1293          *       field mappings here. (see T142670 and T143409)
1294          *
1295          * @param SearchEngine $engine
1296          * @return SearchIndexField[] List of fields this content handler can provide.
1297          * @since 1.28
1298          */
1299         public function getFieldsForSearchIndex( SearchEngine $engine ) {
1300                 $fields['category'] = $engine->makeSearchFieldMapping(
1301                         'category',
1302                         SearchIndexField::INDEX_TYPE_TEXT
1303                 );
1304                 $fields['category']->setFlag( SearchIndexField::FLAG_CASEFOLD );
1305
1306                 $fields['external_link'] = $engine->makeSearchFieldMapping(
1307                         'external_link',
1308                         SearchIndexField::INDEX_TYPE_KEYWORD
1309                 );
1310
1311                 $fields['outgoing_link'] = $engine->makeSearchFieldMapping(
1312                         'outgoing_link',
1313                         SearchIndexField::INDEX_TYPE_KEYWORD
1314                 );
1315
1316                 $fields['template'] = $engine->makeSearchFieldMapping(
1317                         'template',
1318                         SearchIndexField::INDEX_TYPE_KEYWORD
1319                 );
1320                 $fields['template']->setFlag( SearchIndexField::FLAG_CASEFOLD );
1321
1322                 $fields['content_model'] = $engine->makeSearchFieldMapping(
1323                         'content_model',
1324                         SearchIndexField::INDEX_TYPE_KEYWORD
1325                 );
1326
1327                 return $fields;
1328         }
1329
1330         /**
1331          * Add new field definition to array.
1332          * @param SearchIndexField[] &$fields
1333          * @param SearchEngine $engine
1334          * @param string $name
1335          * @param int $type
1336          * @return SearchIndexField[] new field defs
1337          * @since 1.28
1338          */
1339         protected function addSearchField( &$fields, SearchEngine $engine, $name, $type ) {
1340                 $fields[$name] = $engine->makeSearchFieldMapping( $name, $type );
1341                 return $fields;
1342         }
1343
1344         /**
1345          * Return fields to be indexed by search engine
1346          * as representation of this document.
1347          * Overriding class should call parent function or take care of calling
1348          * the SearchDataForIndex hook.
1349          * @param WikiPage $page Page to index
1350          * @param ParserOutput $output
1351          * @param SearchEngine $engine Search engine for which we are indexing
1352          * @return array Map of name=>value for fields
1353          * @since 1.28
1354          */
1355         public function getDataForSearchIndex(
1356                 WikiPage $page,
1357                 ParserOutput $output,
1358                 SearchEngine $engine
1359         ) {
1360                 $fieldData = [];
1361                 $content = $page->getContent();
1362
1363                 if ( $content ) {
1364                         $searchDataExtractor = new ParserOutputSearchDataExtractor();
1365
1366                         $fieldData['category'] = $searchDataExtractor->getCategories( $output );
1367                         $fieldData['external_link'] = $searchDataExtractor->getExternalLinks( $output );
1368                         $fieldData['outgoing_link'] = $searchDataExtractor->getOutgoingLinks( $output );
1369                         $fieldData['template'] = $searchDataExtractor->getTemplates( $output );
1370
1371                         $text = $content->getTextForSearchIndex();
1372
1373                         $fieldData['text'] = $text;
1374                         $fieldData['source_text'] = $text;
1375                         $fieldData['text_bytes'] = $content->getSize();
1376                         $fieldData['content_model'] = $content->getModel();
1377                 }
1378
1379                 Hooks::run( 'SearchDataForIndex', [ &$fieldData, $this, $page, $output, $engine ] );
1380                 return $fieldData;
1381         }
1382
1383         /**
1384          * Produce page output suitable for indexing.
1385          *
1386          * Specific content handlers may override it if they need different content handling.
1387          *
1388          * @param WikiPage $page
1389          * @param ParserCache|null $cache
1390          * @return ParserOutput
1391          */
1392         public function getParserOutputForIndexing( WikiPage $page, ParserCache $cache = null ) {
1393                 // TODO: MCR: ContentHandler should be called per slot, not for the whole page.
1394                 // See T190066.
1395                 $parserOptions = $page->makeParserOptions( 'canonical' );
1396                 if ( $cache ) {
1397                         $parserOutput = $cache->get( $page, $parserOptions );
1398                 }
1399
1400                 if ( empty( $parserOutput ) ) {
1401                         $renderer = MediaWikiServices::getInstance()->getRevisionRenderer();
1402                         $parserOutput =
1403                                 $renderer->getRenderedRevision(
1404                                         $page->getRevision()->getRevisionRecord(),
1405                                         $parserOptions
1406                                 )->getRevisionParserOutput();
1407                         if ( $cache ) {
1408                                 $cache->save( $parserOutput, $page, $parserOptions );
1409                         }
1410                 }
1411                 return $parserOutput;
1412         }
1413
1414         /**
1415          * Returns a list of DeferrableUpdate objects for recording information about the
1416          * given Content in some secondary data store.
1417          *
1418          * Application logic should not call this method directly. Instead, it should call
1419          * DerivedPageDataUpdater::getSecondaryDataUpdates().
1420          *
1421          * @note Implementations must not return a LinksUpdate instance. Instead, a LinksUpdate
1422          * is created by the calling code in DerivedPageDataUpdater, on the combined ParserOutput
1423          * of all slots, not for each slot individually. This is in contrast to the old
1424          * getSecondaryDataUpdates method defined by AbstractContent, which returned a LinksUpdate.
1425          *
1426          * @note Implementations should not call $content->getParserOutput, they should call
1427          * $slotOutput->getSlotRendering( $role, false ) instead if they need to access a ParserOutput
1428          * of $content. This allows existing ParserOutput objects to be re-used, while avoiding
1429          * creating a ParserOutput when none is needed.
1430          *
1431          * @param Title $title The title of the page to supply the updates for
1432          * @param Content $content The content to generate data updates for.
1433          * @param string $role The role (slot) in which the content is being used. Which updates
1434          *        are performed should generally not depend on the role the content has, but the
1435          *        DeferrableUpdates themselves may need to know the role, to track to which slot the
1436          *        data refers, and to avoid overwriting data of the same kind from another slot.
1437          * @param SlotRenderingProvider $slotOutput A provider that can be used to gain access to
1438          *        a ParserOutput of $content by calling $slotOutput->getSlotParserOutput( $role, false ).
1439          * @return DeferrableUpdate[] A list of DeferrableUpdate objects for putting information
1440          *        about this content object somewhere. The default implementation returns an empty
1441          *        array.
1442          * @since 1.32
1443          */
1444         public function getSecondaryDataUpdates(
1445                 Title $title,
1446                 Content $content,
1447                 $role,
1448                 SlotRenderingProvider $slotOutput
1449         ) {
1450                 return [];
1451         }
1452
1453         /**
1454          * Returns a list of DeferrableUpdate objects for removing information about content
1455          * in some secondary data store. This is used when a page is deleted, and also when
1456          * a slot is removed from a page.
1457          *
1458          * Application logic should not call this method directly. Instead, it should call
1459          * WikiPage::getSecondaryDataUpdates().
1460          *
1461          * @note Implementations must not return a LinksDeletionUpdate instance. Instead, a
1462          * LinksDeletionUpdate is created by the calling code in WikiPage.
1463          * This is in contrast to the old getDeletionUpdates method defined by AbstractContent,
1464          * which returned a LinksUpdate.
1465          *
1466          * @note Implementations should not rely on the page's current content, but rather the current
1467          * state of the secondary data store.
1468          *
1469          * @param Title $title The title of the page to supply the updates for
1470          * @param string $role The role (slot) in which the content is being used. Which updates
1471          *        are performed should generally not depend on the role the content has, but the
1472          *        DeferrableUpdates themselves may need to know the role, to track to which slot the
1473          *        data refers, and to avoid overwriting data of the same kind from another slot.
1474          *
1475          * @return DeferrableUpdate[] A list of DeferrableUpdate objects for putting information
1476          *        about this content object somewhere. The default implementation returns an empty
1477          *        array.
1478          *
1479          * @since 1.32
1480          */
1481         public function getDeletionUpdates( Title $title, $role ) {
1482                 return [];
1483         }
1484
1485 }