includes/content/ContentHandler.php

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