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 /**
  29  * Exception representing a failure to serialize or unserialize a content object.
  30  *
  31  * @ingroup Content
  32  */
  33 class MWContentSerializationException extends MWException {
  34 }
  35
  36 /**
  37  * A content handler knows how do deal with a specific type of content on a wiki
  38  * page. Content is stored in the database in a serialized form (using a
  39  * serialization format a.k.a. MIME type) and is unserialized into its native
  40  * PHP representation (the content model), which is wrapped in an instance of
  41  * the appropriate subclass of Content.
  42  *
  43  * ContentHandler instances are stateless singletons that serve, among other
  44  * things, as a factory for Content objects. Generally, there is one subclass
  45  * of ContentHandler and one subclass of Content for every type of content model.
  46  *
  47  * Some content types have a flat model, that is, their native representation
  48  * is the same as their serialized form. Examples would be JavaScript and CSS
  49  * code. As of now, this also applies to wikitext (MediaWiki's default content
  50  * type), but wikitext content may be represented by a DOM or AST structure in
  51  * the future.
  52  *
  53  * @ingroup Content
  54  */
  55 abstract class ContentHandler {
  56         /**
  57          * Switch for enabling deprecation warnings. Used by ContentHandler::deprecated()
  58          * and ContentHandler::runLegacyHooks().
  59          *
  60          * Once the ContentHandler code has settled in a bit, this should be set to true to
  61          * make extensions etc. show warnings when using deprecated functions and hooks.
  62          */
  63         protected static $enableDeprecationWarnings = false;
  64
  65         /**
  66          * Convenience function for getting flat text from a Content object. This
  67          * should only be used in the context of backwards compatibility with code
  68          * that is not yet able to handle Content objects!
  69          *
  70          * If $content is null, this method returns the empty string.
  71          *
  72          * If $content is an instance of TextContent, this method returns the flat
  73          * text as returned by $content->getNativeData().
  74          *
  75          * If $content is not a TextContent object, the behavior of this method
  76          * depends on the global $wgContentHandlerTextFallback:
  77          * - If $wgContentHandlerTextFallback is 'fail' and $content is not a
  78          *   TextContent object, an MWException is thrown.
  79          * - If $wgContentHandlerTextFallback is 'serialize' and $content is not a
  80          *   TextContent object, $content->serialize() is called to get a string
  81          *   form of the content.
  82          * - If $wgContentHandlerTextFallback is 'ignore' and $content is not a
  83          *   TextContent object, this method returns null.
  84          * - otherwise, the behavior is undefined.
  85          *
  86          * @since 1.21
  87          *
  88          * @param Content $content
  89          *
  90          * @throws MWException If the content is not an instance of TextContent and
  91          * wgContentHandlerTextFallback was set to 'fail'.
  92          * @return string|null Textual form of the content, if available.
  93          */
  94         public static function getContentText( Content $content = null ) {
  95                 global $wgContentHandlerTextFallback;
  96
  97                 if ( is_null( $content ) ) {
  98                         return '';
  99                 }
 100
 101                 if ( $content instanceof TextContent ) {
 102                         return $content->getNativeData();
 103                 }
 104
 105                 wfDebugLog( 'ContentHandler', 'Accessing ' . $content->getModel() . ' content as text!' );
 106
 107                 if ( $wgContentHandlerTextFallback == 'fail' ) {
 108                         throw new MWException(
 109                                 "Attempt to get text from Content with model " .
 110                                 $content->getModel()
 111                         );
 112                 }
 113
 114                 if ( $wgContentHandlerTextFallback == 'serialize' ) {
 115                         return $content->serialize();
 116                 }
 117
 118                 return null;
 119         }
 120
 121         /**
 122          * Convenience function for creating a Content object from a given textual
 123          * representation.
 124          *
 125          * $text will be deserialized into a Content object of the model specified
 126          * by $modelId (or, if that is not given, $title->getContentModel()) using
 127          * the given format.
 128          *
 129          * @since 1.21
 130          *
 131          * @param string $text The textual representation, will be
 132          *    unserialized to create the Content object
 133          * @param Title $title The title of the page this text belongs to.
 134          *    Required if $modelId is not provided.
 135          * @param string $modelId The model to deserialize to. If not provided,
 136          *    $title->getContentModel() is used.
 137          * @param string $format The format to use for deserialization. If not
 138          *    given, the model's default format is used.
 139          *
 140          * @throws MWException If model ID or format is not supported or if the text can not be
 141          * unserialized using the format.
 142          * @return Content A Content object representing the text.
 143          */
 144         public static function makeContent( $text, Title $title = null,
 145                 $modelId = null, $format = null ) {
 146                 if ( is_null( $modelId ) ) {
 147                         if ( is_null( $title ) ) {
 148                                 throw new MWException( "Must provide a Title object or a content model ID." );
 149                         }
 150
 151                         $modelId = $title->getContentModel();
 152                 }
 153
 154                 $handler = ContentHandler::getForModelID( $modelId );
 155
 156                 return $handler->unserializeContent( $text, $format );
 157         }
 158
 159         /**
 160          * Returns the name of the default content model to be used for the page
 161          * with the given title.
 162          *
 163          * Note: There should rarely be need to call this method directly.
 164          * To determine the actual content model for a given page, use
 165          * Title::getContentModel().
 166          *
 167          * Which model is to be used by default for the page is determined based
 168          * on several factors:
 169          * - The global setting $wgNamespaceContentModels specifies a content model
 170          *   per namespace.
 171          * - The hook ContentHandlerDefaultModelFor may be used to override the page's default
 172          *   model.
 173          * - Pages in NS_MEDIAWIKI and NS_USER default to the CSS or JavaScript
 174          *   model if they end in .js or .css, respectively.
 175          * - Pages in NS_MEDIAWIKI default to the wikitext model otherwise.
 176          * - The hook TitleIsCssOrJsPage may be used to force a page to use the CSS
 177          *   or JavaScript model. This is a compatibility feature. The ContentHandlerDefaultModelFor
 178          *   hook should be used instead if possible.
 179          * - The hook TitleIsWikitextPage may be used to force a page to use the
 180          *   wikitext model. This is a compatibility feature. The ContentHandlerDefaultModelFor
 181          *   hook should be used instead if possible.
 182          *
 183          * If none of the above applies, the wikitext model is used.
 184          *
 185          * Note: this is used by, and may thus not use, Title::getContentModel()
 186          *
 187          * @since 1.21
 188          *
 189          * @param Title $title
 190          *
 191          * @return string Default model name for the page given by $title
 192          */
 193         public static function getDefaultModelFor( Title $title ) {
 194                 // NOTE: this method must not rely on $title->getContentModel() directly or indirectly,
 195                 //       because it is used to initialize the mContentModel member.
 196
 197                 $ns = $title->getNamespace();
 198
 199                 $ext = false;
 200                 $m = null;
 201                 $model = MWNamespace::getNamespaceContentModel( $ns );
 202
 203                 // Hook can determine default model
 204                 if ( !Hooks::run( 'ContentHandlerDefaultModelFor', array( $title, &$model ) ) ) {
 205                         if ( !is_null( $model ) ) {
 206                                 return $model;
 207                         }
 208                 }
 209
 210                 // Could this page contain code based on the title?
 211                 $isCodePage = NS_MEDIAWIKI == $ns && preg_match( '!\.(css|js|json)$!u', $title->getText(), $m );
 212                 if ( $isCodePage ) {
 213                         $ext = $m[1];
 214                 }
 215
 216                 // Hook can force JS/CSS
 217                 Hooks::run( 'TitleIsCssOrJsPage', array( $title, &$isCodePage ), '1.25' );
 218
 219                 // Is this a user subpage containing code?
 220                 $isCodeSubpage = NS_USER == $ns
 221                         && !$isCodePage
 222                         && preg_match( "/\\/.*\\.(js|css|json)$/", $title->getText(), $m );
 223                 if ( $isCodeSubpage ) {
 224                         $ext = $m[1];
 225                 }
 226
 227                 // Is this wikitext, according to $wgNamespaceContentModels or the DefaultModelFor hook?
 228                 $isWikitext = is_null( $model ) || $model == CONTENT_MODEL_WIKITEXT;
 229                 $isWikitext = $isWikitext && !$isCodePage && !$isCodeSubpage;
 230
 231                 // Hook can override $isWikitext
 232                 Hooks::run( 'TitleIsWikitextPage', array( $title, &$isWikitext ), '1.25' );
 233
 234                 if ( !$isWikitext ) {
 235                         switch ( $ext ) {
 236                                 case 'js':
 237                                         return CONTENT_MODEL_JAVASCRIPT;
 238                                 case 'css':
 239                                         return CONTENT_MODEL_CSS;
 240                                 case 'json':
 241                                         return CONTENT_MODEL_JSON;
 242                                 default:
 243                                         return is_null( $model ) ? CONTENT_MODEL_TEXT : $model;
 244                         }
 245                 }
 246
 247                 // We established that it must be wikitext
 248
 249                 return CONTENT_MODEL_WIKITEXT;
 250         }
 251
 252         /**
 253          * Returns the appropriate ContentHandler singleton for the given title.
 254          *
 255          * @since 1.21
 256          *
 257          * @param Title $title
 258          *
 259          * @return ContentHandler
 260          */
 261         public static function getForTitle( Title $title ) {
 262                 $modelId = $title->getContentModel();
 263
 264                 return ContentHandler::getForModelID( $modelId );
 265         }
 266
 267         /**
 268          * Returns the appropriate ContentHandler singleton for the given Content
 269          * object.
 270          *
 271          * @since 1.21
 272          *
 273          * @param Content $content
 274          *
 275          * @return ContentHandler
 276          */
 277         public static function getForContent( Content $content ) {
 278                 $modelId = $content->getModel();
 279
 280                 return ContentHandler::getForModelID( $modelId );
 281         }
 282
 283         /**
 284          * @var array A Cache of ContentHandler instances by model id
 285          */
 286         protected static $handlers;
 287
 288         /**
 289          * Returns the ContentHandler singleton for the given model ID. Use the
 290          * CONTENT_MODEL_XXX constants to identify the desired content model.
 291          *
 292          * ContentHandler singletons are taken from the global $wgContentHandlers
 293          * array. Keys in that array are model names, the values are either
 294          * ContentHandler singleton objects, or strings specifying the appropriate
 295          * subclass of ContentHandler.
 296          *
 297          * If a class name is encountered when looking up the singleton for a given
 298          * model name, the class is instantiated and the class name is replaced by
 299          * the resulting singleton in $wgContentHandlers.
 300          *
 301          * If no ContentHandler is defined for the desired $modelId, the
 302          * ContentHandler may be provided by the ContentHandlerForModelID hook.
 303          * If no ContentHandler can be determined, an MWException is raised.
 304          *
 305          * @since 1.21
 306          *
 307          * @param string $modelId The ID of the content model for which to get a
 308          *    handler. Use CONTENT_MODEL_XXX constants.
 309          *
 310          * @throws MWException If no handler is known for the model ID.
 311          * @return ContentHandler The ContentHandler singleton for handling the model given by the ID.
 312          */
 313         public static function getForModelID( $modelId ) {
 314                 global $wgContentHandlers;
 315
 316                 if ( isset( ContentHandler::$handlers[$modelId] ) ) {
 317                         return ContentHandler::$handlers[$modelId];
 318                 }
 319
 320                 if ( empty( $wgContentHandlers[$modelId] ) ) {
 321                         $handler = null;
 322
 323                         Hooks::run( 'ContentHandlerForModelID', array( $modelId, &$handler ) );
 324
 325                         if ( $handler === null ) {
 326                                 throw new MWException( "No handler for model '$modelId' registered in \$wgContentHandlers" );
 327                         }
 328
 329                         if ( !( $handler instanceof ContentHandler ) ) {
 330                                 throw new MWException( "ContentHandlerForModelID must supply a ContentHandler instance" );
 331                         }
 332                 } else {
 333                         $class = $wgContentHandlers[$modelId];
 334                         $handler = new $class( $modelId );
 335
 336                         if ( !( $handler instanceof ContentHandler ) ) {
 337                                 throw new MWException( "$class from \$wgContentHandlers is not " .
 338                                         "compatible with ContentHandler" );
 339                         }
 340                 }
 341
 342                 wfDebugLog( 'ContentHandler', 'Created handler for ' . $modelId
 343                         . ': ' . get_class( $handler ) );
 344
 345                 ContentHandler::$handlers[$modelId] = $handler;
 346
 347                 return ContentHandler::$handlers[$modelId];
 348         }
 349
 350         /**
 351          * Returns the localized name for a given content model.
 352          *
 353          * Model names are localized using system messages. Message keys
 354          * have the form content-model-$name, where $name is getContentModelName( $id ).
 355          *
 356          * @param string $name The content model ID, as given by a CONTENT_MODEL_XXX
 357          *    constant or returned by Revision::getContentModel().
 358          *
 359          * @throws MWException If the model ID isn't known.
 360          * @return string The content model's localized name.
 361          */
 362         public static function getLocalizedName( $name ) {
 363                 // Messages: content-model-wikitext, content-model-text,
 364                 // content-model-javascript, content-model-css
 365                 $key = "content-model-$name";
 366
 367                 $msg = wfMessage( $key );
 368
 369                 return $msg->exists() ? $msg->plain() : $name;
 370         }
 371
 372         public static function getContentModels() {
 373                 global $wgContentHandlers;
 374
 375                 return array_keys( $wgContentHandlers );
 376         }
 377
 378         public static function getAllContentFormats() {
 379                 global $wgContentHandlers;
 380
 381                 $formats = array();
 382
 383                 foreach ( $wgContentHandlers as $model => $class ) {
 384                         $handler = ContentHandler::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                 $this->mModelName = preg_replace( '/(Content)?Handler$/', '', get_class( $this ) );
 419                 $this->mModelName = preg_replace( '/[_\\\\]/', '', $this->mModelName );
 420                 $this->mModelName = strtolower( $this->mModelName );
 421         }
 422
 423         /**
 424          * Serializes a Content object of the type supported by this ContentHandler.
 425          *
 426          * @since 1.21
 427          *
 428          * @param Content $content The Content object to serialize
 429          * @param string $format The desired serialization format
 430          *
 431          * @return string Serialized form of the content
 432          */
 433         abstract public function serializeContent( Content $content, $format = null );
 434
 435         /**
 436          * Applies transformations on export (returns the blob unchanged per default).
 437          * Subclasses may override this to perform transformations such as conversion
 438          * of legacy formats or filtering of internal meta-data.
 439          *
 440          * @param string $blob The blob to be exported
 441          * @param string|null $format The blob's serialization format
 442          *
 443          * @return string
 444          */
 445         public function exportTransform( $blob, $format = null ) {
 446                 return $blob;
 447         }
 448
 449         /**
 450          * Unserializes a Content object of the type supported by this ContentHandler.
 451          *
 452          * @since 1.21
 453          *
 454          * @param string $blob Serialized form of the content
 455          * @param string $format The format used for serialization
 456          *
 457          * @return Content The Content object created by deserializing $blob
 458          */
 459         abstract public function unserializeContent( $blob, $format = null );
 460
 461         /**
 462          * Apply import transformation (per default, returns $blob unchanged).
 463          * This gives subclasses an opportunity to transform data blobs on import.
 464          *
 465          * @since 1.24
 466          *
 467          * @param string $blob
 468          * @param string|null $format
 469          *
 470          * @return string
 471          */
 472         public function importTransform( $blob, $format = null ) {
 473                 return $blob;
 474         }
 475
 476         /**
 477          * Creates an empty Content object of the type supported by this
 478          * ContentHandler.
 479          *
 480          * @since 1.21
 481          *
 482          * @return Content
 483          */
 484         abstract public function makeEmptyContent();
 485
 486         /**
 487          * Creates a new Content object that acts as a redirect to the given page,
 488          * or null if redirects are not supported by this content model.
 489          *
 490          * This default implementation always returns null. Subclasses supporting redirects
 491          * must override this method.
 492          *
 493          * Note that subclasses that override this method to return a Content object
 494          * should also override supportsRedirects() to return true.
 495          *
 496          * @since 1.21
 497          *
 498          * @param Title $destination The page to redirect to.
 499          * @param string $text Text to include in the redirect, if possible.
 500          *
 501          * @return Content Always null.
 502          */
 503         public function makeRedirectContent( Title $destination, $text = '' ) {
 504                 return null;
 505         }
 506
 507         /**
 508          * Returns the model id that identifies the content model this
 509          * ContentHandler can handle. Use with the CONTENT_MODEL_XXX constants.
 510          *
 511          * @since 1.21
 512          *
 513          * @return string The model ID
 514          */
 515         public function getModelID() {
 516                 return $this->mModelID;
 517         }
 518
 519         /**
 520          * @since 1.21
 521          *
 522          * @param string $model_id The model to check
 523          *
 524          * @throws MWException If the model ID is not the ID of the content model supported by this
 525          * ContentHandler.
 526          */
 527         protected function checkModelID( $model_id ) {
 528                 if ( $model_id !== $this->mModelID ) {
 529                         throw new MWException( "Bad content model: " .
 530                                 "expected {$this->mModelID} " .
 531                                 "but got $model_id." );
 532                 }
 533         }
 534
 535         /**
 536          * Returns a list of serialization formats supported by the
 537          * serializeContent() and unserializeContent() methods of this
 538          * ContentHandler.
 539          *
 540          * @since 1.21
 541          *
 542          * @return string[] List of serialization formats as MIME type like strings
 543          */
 544         public function getSupportedFormats() {
 545                 return $this->mSupportedFormats;
 546         }
 547
 548         /**
 549          * The format used for serialization/deserialization by default by this
 550          * ContentHandler.
 551          *
 552          * This default implementation will return the first element of the array
 553          * of formats that was passed to the constructor.
 554          *
 555          * @since 1.21
 556          *
 557          * @return string The name of the default serialization format as a MIME type
 558          */
 559         public function getDefaultFormat() {
 560                 return $this->mSupportedFormats[0];
 561         }
 562
 563         /**
 564          * Returns true if $format is a serialization format supported by this
 565          * ContentHandler, and false otherwise.
 566          *
 567          * Note that if $format is null, this method always returns true, because
 568          * null means "use the default format".
 569          *
 570          * @since 1.21
 571          *
 572          * @param string $format The serialization format to check
 573          *
 574          * @return bool
 575          */
 576         public function isSupportedFormat( $format ) {
 577                 if ( !$format ) {
 578                         return true; // this means "use the default"
 579                 }
 580
 581                 return in_array( $format, $this->mSupportedFormats );
 582         }
 583
 584         /**
 585          * Convenient for checking whether a format provided as a parameter is actually supported.
 586          *
 587          * @param string $format The serialization format to check
 588          *
 589          * @throws MWException If the format is not supported by this content handler.
 590          */
 591         protected function checkFormat( $format ) {
 592                 if ( !$this->isSupportedFormat( $format ) ) {
 593                         throw new MWException(
 594                                 "Format $format is not supported for content model "
 595                                 . $this->getModelID()
 596                         );
 597                 }
 598         }
 599
 600         /**
 601          * Returns overrides for action handlers.
 602          * Classes listed here will be used instead of the default one when
 603          * (and only when) $wgActions[$action] === true. This allows subclasses
 604          * to override the default action handlers.
 605          *
 606          * @since 1.21
 607          *
 608          * @return array Always an empty array.
 609          */
 610         public function getActionOverrides() {
 611                 return array();
 612         }
 613
 614         /**
 615          * Factory for creating an appropriate DifferenceEngine for this content model.
 616          *
 617          * @since 1.21
 618          *
 619          * @param IContextSource $context Context to use, anything else will be ignored.
 620          * @param int $old Revision ID we want to show and diff with.
 621          * @param int|string $new Either a revision ID or one of the strings 'cur', 'prev' or 'next'.
 622          * @param int $rcid FIXME: Deprecated, no longer used. Defaults to 0.
 623          * @param bool $refreshCache If set, refreshes the diff cache. Defaults to false.
 624          * @param bool $unhide If set, allow viewing deleted revs. Defaults to false.
 625          *
 626          * @return DifferenceEngine
 627          */
 628         public function createDifferenceEngine( IContextSource $context, $old = 0, $new = 0,
 629                 $rcid = 0, //FIXME: Deprecated, no longer used
 630                 $refreshCache = false, $unhide = false ) {
 631
 632                 // hook: get difference engine
 633                 $differenceEngine = null;
 634                 if ( !Hooks::run( 'GetDifferenceEngine',
 635                         array( $context, $old, $new, $refreshCache, $unhide, &$differenceEngine )
 636                 ) ) {
 637                         return $differenceEngine;
 638                 }
 639                 $diffEngineClass = $this->getDiffEngineClass();
 640                 return new $diffEngineClass( $context, $old, $new, $rcid, $refreshCache, $unhide );
 641         }
 642
 643         /**
 644          * Get the language in which the content of the given page is written.
 645          *
 646          * This default implementation just returns $wgContLang (except for pages
 647          * in the MediaWiki namespace)
 648          *
 649          * Note that the pages language is not cacheable, since it may in some
 650          * cases depend on user settings.
 651          *
 652          * Also note that the page language may or may not depend on the actual content of the page,
 653          * that is, this method may load the content in order to determine the language.
 654          *
 655          * @since 1.21
 656          *
 657          * @param Title $title The page to determine the language for.
 658          * @param Content $content The page's content, if you have it handy, to avoid reloading it.
 659          *
 660          * @return Language The page's language
 661          */
 662         public function getPageLanguage( Title $title, Content $content = null ) {
 663                 global $wgContLang, $wgLang;
 664                 $pageLang = $wgContLang;
 665
 666                 if ( $title->getNamespace() == NS_MEDIAWIKI ) {
 667                         // Parse mediawiki messages with correct target language
 668                         list( /* $unused */, $lang ) = MessageCache::singleton()->figureMessage( $title->getText() );
 669                         $pageLang = wfGetLangObj( $lang );
 670                 }
 671
 672                 Hooks::run( 'PageContentLanguage', array( $title, &$pageLang, $wgLang ) );
 673
 674                 return wfGetLangObj( $pageLang );
 675         }
 676
 677         /**
 678          * Get the language in which the content of this page is written when
 679          * viewed by user. Defaults to $this->getPageLanguage(), but if the user
 680          * specified a preferred variant, the variant will be used.
 681          *
 682          * This default implementation just returns $this->getPageLanguage( $title, $content ) unless
 683          * the user specified a preferred variant.
 684          *
 685          * Note that the pages view language is not cacheable, since it depends on user settings.
 686          *
 687          * Also note that the page language may or may not depend on the actual content of the page,
 688          * that is, this method may load the content in order to determine the language.
 689          *
 690          * @since 1.21
 691          *
 692          * @param Title $title The page to determine the language for.
 693          * @param Content $content The page's content, if you have it handy, to avoid reloading it.
 694          *
 695          * @return Language The page's language for viewing
 696          */
 697         public function getPageViewLanguage( Title $title, Content $content = null ) {
 698                 $pageLang = $this->getPageLanguage( $title, $content );
 699
 700                 if ( $title->getNamespace() !== NS_MEDIAWIKI ) {
 701                         // If the user chooses a variant, the content is actually
 702                         // in a language whose code is the variant code.
 703                         $variant = $pageLang->getPreferredVariant();
 704                         if ( $pageLang->getCode() !== $variant ) {
 705                                 $pageLang = Language::factory( $variant );
 706                         }
 707                 }
 708
 709                 return $pageLang;
 710         }
 711
 712         /**
 713          * Determines whether the content type handled by this ContentHandler
 714          * can be used on the given page.
 715          *
 716          * This default implementation always returns true.
 717          * Subclasses may override this to restrict the use of this content model to specific locations,
 718          * typically based on the namespace or some other aspect of the title, such as a special suffix
 719          * (e.g. ".svg" for SVG content).
 720          *
 721          * @note this calls the ContentHandlerCanBeUsedOn hook which may be used to override which
 722          * content model can be used where.
 723          *
 724          * @param Title $title The page's title.
 725          *
 726          * @return bool True if content of this kind can be used on the given page, false otherwise.
 727          */
 728         public function canBeUsedOn( Title $title ) {
 729                 $ok = true;
 730
 731                 Hooks::run( 'ContentModelCanBeUsedOn', array( $this->getModelID(), $title, &$ok ) );
 732
 733                 return $ok;
 734         }
 735
 736         /**
 737          * Returns the name of the diff engine to use.
 738          *
 739          * @since 1.21
 740          *
 741          * @return string
 742          */
 743         protected function getDiffEngineClass() {
 744                 return 'DifferenceEngine';
 745         }
 746
 747         /**
 748          * Attempts to merge differences between three versions. Returns a new
 749          * Content object for a clean merge and false for failure or a conflict.
 750          *
 751          * This default implementation always returns false.
 752          *
 753          * @since 1.21
 754          *
 755          * @param Content $oldContent The page's previous content.
 756          * @param Content $myContent One of the page's conflicting contents.
 757          * @param Content $yourContent One of the page's conflicting contents.
 758          *
 759          * @return Content|bool Always false.
 760          */
 761         public function merge3( Content $oldContent, Content $myContent, Content $yourContent ) {
 762                 return false;
 763         }
 764
 765         /**
 766          * Return an applicable auto-summary if one exists for the given edit.
 767          *
 768          * @since 1.21
 769          *
 770          * @param Content $oldContent The previous text of the page.
 771          * @param Content $newContent The submitted text of the page.
 772          * @param int $flags Bit mask: a bit mask of flags submitted for the edit.
 773          *
 774          * @return string An appropriate auto-summary, or an empty string.
 775          */
 776         public function getAutosummary( Content $oldContent = null, Content $newContent = null,
 777                 $flags ) {
 778                 // Decide what kind of auto-summary is needed.
 779
 780                 // Redirect auto-summaries
 781
 782                 /**
 783                  * @var $ot Title
 784                  * @var $rt Title
 785                  */
 786
 787                 $ot = !is_null( $oldContent ) ? $oldContent->getRedirectTarget() : null;
 788                 $rt = !is_null( $newContent ) ? $newContent->getRedirectTarget() : null;
 789
 790                 if ( is_object( $rt ) ) {
 791                         if ( !is_object( $ot )
 792                                 || !$rt->equals( $ot )
 793                                 || $ot->getFragment() != $rt->getFragment()
 794                         ) {
 795                                 $truncatedtext = $newContent->getTextForSummary(
 796                                         250
 797                                         - strlen( wfMessage( 'autoredircomment' )->inContentLanguage()->text() )
 798                                         - strlen( $rt->getFullText() ) );
 799
 800                                 return wfMessage( 'autoredircomment', $rt->getFullText() )
 801                                         ->rawParams( $truncatedtext )->inContentLanguage()->text();
 802                         }
 803                 }
 804
 805                 // New page auto-summaries
 806                 if ( $flags & EDIT_NEW && $newContent->getSize() > 0 ) {
 807                         // If they're making a new article, give its text, truncated, in
 808                         // the summary.
 809
 810                         $truncatedtext = $newContent->getTextForSummary(
 811                                 200 - strlen( wfMessage( 'autosumm-new' )->inContentLanguage()->text() ) );
 812
 813                         return wfMessage( 'autosumm-new' )->rawParams( $truncatedtext )
 814                                 ->inContentLanguage()->text();
 815                 }
 816
 817                 // Blanking auto-summaries
 818                 if ( !empty( $oldContent ) && $oldContent->getSize() > 0 && $newContent->getSize() == 0 ) {
 819                         return wfMessage( 'autosumm-blank' )->inContentLanguage()->text();
 820                 } elseif ( !empty( $oldContent )
 821                         && $oldContent->getSize() > 10 * $newContent->getSize()
 822                         && $newContent->getSize() < 500
 823                 ) {
 824                         // Removing more than 90% of the article
 825
 826                         $truncatedtext = $newContent->getTextForSummary(
 827                                 200 - strlen( wfMessage( 'autosumm-replace' )->inContentLanguage()->text() ) );
 828
 829                         return wfMessage( 'autosumm-replace' )->rawParams( $truncatedtext )
 830                                 ->inContentLanguage()->text();
 831                 }
 832
 833                 // New blank article auto-summary
 834                 if ( $flags & EDIT_NEW && $newContent->isEmpty() ) {
 835                         return wfMessage( 'autosumm-newblank' )->inContentLanguage()->text();
 836                 }
 837
 838                 // If we reach this point, there's no applicable auto-summary for our
 839                 // case, so our auto-summary is empty.
 840                 return '';
 841         }
 842
 843         /**
 844          * Auto-generates a deletion reason
 845          *
 846          * @since 1.21
 847          *
 848          * @param Title $title The page's title
 849          * @param bool &$hasHistory Whether the page has a history
 850          *
 851          * @return mixed String containing deletion reason or empty string, or
 852          *    boolean false if no revision occurred
 853          *
 854          * @todo &$hasHistory is extremely ugly, it's here because
 855          * WikiPage::getAutoDeleteReason() and Article::generateReason()
 856          * have it / want it.
 857          */
 858         public function getAutoDeleteReason( Title $title, &$hasHistory ) {
 859                 $dbw = wfGetDB( DB_MASTER );
 860
 861                 // Get the last revision
 862                 $rev = Revision::newFromTitle( $title );
 863
 864                 if ( is_null( $rev ) ) {
 865                         return false;
 866                 }
 867
 868                 // Get the article's contents
 869                 $content = $rev->getContent();
 870                 $blank = false;
 871
 872                 // If the page is blank, use the text from the previous revision,
 873                 // which can only be blank if there's a move/import/protect dummy
 874                 // revision involved
 875                 if ( !$content || $content->isEmpty() ) {
 876                         $prev = $rev->getPrevious();
 877
 878                         if ( $prev ) {
 879                                 $rev = $prev;
 880                                 $content = $rev->getContent();
 881                                 $blank = true;
 882                         }
 883                 }
 884
 885                 $this->checkModelID( $rev->getContentModel() );
 886
 887                 // Find out if there was only one contributor
 888                 // Only scan the last 20 revisions
 889                 $res = $dbw->select( 'revision', 'rev_user_text',
 890                         array(
 891                                 'rev_page' => $title->getArticleID(),
 892                                 $dbw->bitAnd( 'rev_deleted', Revision::DELETED_USER ) . ' = 0'
 893                         ),
 894                         __METHOD__,
 895                         array( 'LIMIT' => 20 )
 896                 );
 897
 898                 if ( $res === false ) {
 899                         // This page has no revisions, which is very weird
 900                         return false;
 901                 }
 902
 903                 $hasHistory = ( $res->numRows() > 1 );
 904                 $row = $dbw->fetchObject( $res );
 905
 906                 if ( $row ) { // $row is false if the only contributor is hidden
 907                         $onlyAuthor = $row->rev_user_text;
 908                         // Try to find a second contributor
 909                         foreach ( $res as $row ) {
 910                                 if ( $row->rev_user_text != $onlyAuthor ) { // Bug 22999
 911                                         $onlyAuthor = false;
 912                                         break;
 913                                 }
 914                         }
 915                 } else {
 916                         $onlyAuthor = false;
 917                 }
 918
 919                 // Generate the summary with a '$1' placeholder
 920                 if ( $blank ) {
 921                         // The current revision is blank and the one before is also
 922                         // blank. It's just not our lucky day
 923                         $reason = wfMessage( 'exbeforeblank', '$1' )->inContentLanguage()->text();
 924                 } else {
 925                         if ( $onlyAuthor ) {
 926                                 $reason = wfMessage(
 927                                         'excontentauthor',
 928                                         '$1',
 929                                         $onlyAuthor
 930                                 )->inContentLanguage()->text();
 931                         } else {
 932                                 $reason = wfMessage( 'excontent', '$1' )->inContentLanguage()->text();
 933                         }
 934                 }
 935
 936                 if ( $reason == '-' ) {
 937                         // Allow these UI messages to be blanked out cleanly
 938                         return '';
 939                 }
 940
 941                 // Max content length = max comment length - length of the comment (excl. $1)
 942                 $text = $content ? $content->getTextForSummary( 255 - ( strlen( $reason ) - 2 ) ) : '';
 943
 944                 // Now replace the '$1' placeholder
 945                 $reason = str_replace( '$1', $text, $reason );
 946
 947                 return $reason;
 948         }
 949
 950         /**
 951          * Get the Content object that needs to be saved in order to undo all revisions
 952          * between $undo and $undoafter. Revisions must belong to the same page,
 953          * must exist and must not be deleted.
 954          *
 955          * @since 1.21
 956          *
 957          * @param Revision $current The current text
 958          * @param Revision $undo The revision to undo
 959          * @param Revision $undoafter Must be an earlier revision than $undo
 960          *
 961          * @return mixed String on success, false on failure
 962          */
 963         public function getUndoContent( Revision $current, Revision $undo, Revision $undoafter ) {
 964                 $cur_content = $current->getContent();
 965
 966                 if ( empty( $cur_content ) ) {
 967                         return false; // no page
 968                 }
 969
 970                 $undo_content = $undo->getContent();
 971                 $undoafter_content = $undoafter->getContent();
 972
 973                 if ( !$undo_content || !$undoafter_content ) {
 974                         return false; // no content to undo
 975                 }
 976
 977                 $this->checkModelID( $cur_content->getModel() );
 978                 $this->checkModelID( $undo_content->getModel() );
 979                 $this->checkModelID( $undoafter_content->getModel() );
 980
 981                 if ( $cur_content->equals( $undo_content ) ) {
 982                         // No use doing a merge if it's just a straight revert.
 983                         return $undoafter_content;
 984                 }
 985
 986                 $undone_content = $this->merge3( $undo_content, $undoafter_content, $cur_content );
 987
 988                 return $undone_content;
 989         }
 990
 991         /**
 992          * Get parser options suitable for rendering and caching the article
 993          *
 994          * @param IContextSource|User|string $context One of the following:
 995          *        - IContextSource: Use the User and the Language of the provided
 996          *                                            context
 997          *        - User: Use the provided User object and $wgLang for the language,
 998          *                                            so use an IContextSource object if possible.
 999          *        - 'canonical': Canonical options (anonymous user with default
1000          *                                            preferences and content language).
1001          *
1002          * @throws MWException
1003          * @return ParserOptions
1004          */
1005         public function makeParserOptions( $context ) {
1006                 global $wgContLang, $wgEnableParserLimitReporting;
1007
1008                 if ( $context instanceof IContextSource ) {
1009                         $options = ParserOptions::newFromContext( $context );
1010                 } elseif ( $context instanceof User ) { // settings per user (even anons)
1011                         $options = ParserOptions::newFromUser( $context );
1012                 } elseif ( $context === 'canonical' ) { // canonical settings
1013                         $options = ParserOptions::newFromUserAndLang( new User, $wgContLang );
1014                 } else {
1015                         throw new MWException( "Bad context for parser options: $context" );
1016                 }
1017
1018                 $options->enableLimitReport( $wgEnableParserLimitReporting ); // show inclusion/loop reports
1019                 $options->setTidy( true ); // fix bad HTML
1020
1021                 return $options;
1022         }
1023
1024         /**
1025          * Returns true for content models that support caching using the
1026          * ParserCache mechanism. See WikiPage::isParserCacheUsed().
1027          *
1028          * @since 1.21
1029          *
1030          * @return bool Always false.
1031          */
1032         public function isParserCacheSupported() {
1033                 return false;
1034         }
1035
1036         /**
1037          * Returns true if this content model supports sections.
1038          * This default implementation returns false.
1039          *
1040          * Content models that return true here should also implement
1041          * Content::getSection, Content::replaceSection, etc. to handle sections..
1042          *
1043          * @return bool Always false.
1044          */
1045         public function supportsSections() {
1046                 return false;
1047         }
1048
1049         /**
1050          * Returns true if this content model supports redirects.
1051          * This default implementation returns false.
1052          *
1053          * Content models that return true here should also implement
1054          * ContentHandler::makeRedirectContent to return a Content object.
1055          *
1056          * @return bool Always false.
1057          */
1058         public function supportsRedirects() {
1059                 return false;
1060         }
1061
1062         /**
1063          * Return true if this content model supports direct editing, such as via EditPage.
1064          *
1065          * @return bool Default is false, and true for TextContent and it's derivatives.
1066          */
1067         public function supportsDirectEditing() {
1068                 return false;
1069         }
1070
1071         /**
1072          * Whether or not this content model supports direct editing via ApiEditPage
1073          *
1074          * @return bool Default is false, and true for TextContent and derivatives.
1075          */
1076         public function supportsDirectApiEditing() {
1077                 return $this->supportsDirectEditing();
1078         }
1079
1080         /**
1081          * Logs a deprecation warning, visible if $wgDevelopmentWarnings, but only if
1082          * self::$enableDeprecationWarnings is set to true.
1083          *
1084          * @param string $func The name of the deprecated function
1085          * @param string $version The version since the method is deprecated. Usually 1.21
1086          *   for ContentHandler related stuff.
1087          * @param string|bool $component : Component to which the function belongs.
1088          *   If false, it is assumed the function is in MediaWiki core.
1089          *
1090          * @see ContentHandler::$enableDeprecationWarnings
1091          * @see wfDeprecated
1092          */
1093         public static function deprecated( $func, $version, $component = false ) {
1094                 if ( self::$enableDeprecationWarnings ) {
1095                         wfDeprecated( $func, $version, $component, 3 );
1096                 }
1097         }
1098
1099         /**
1100          * Call a legacy hook that uses text instead of Content objects.
1101          * Will log a warning when a matching hook function is registered.
1102          * If the textual representation of the content is changed by the
1103          * hook function, a new Content object is constructed from the new
1104          * text.
1105          *
1106          * @param string $event Event name
1107          * @param array $args Parameters passed to hook functions
1108          * @param bool $warn Whether to log a warning.
1109          *                    Default to self::$enableDeprecationWarnings.
1110          *                    May be set to false for testing.
1111          *
1112          * @return bool True if no handler aborted the hook
1113          *
1114          * @see ContentHandler::$enableDeprecationWarnings
1115          */
1116         public static function runLegacyHooks( $event, $args = array(),
1117                 $warn = null
1118         ) {
1119
1120                 if ( $warn === null ) {
1121                         $warn = self::$enableDeprecationWarnings;
1122                 }
1123
1124                 if ( !Hooks::isRegistered( $event ) ) {
1125                         return true; // nothing to do here
1126                 }
1127
1128                 if ( $warn ) {
1129                         // Log information about which handlers are registered for the legacy hook,
1130                         // so we can find and fix them.
1131
1132                         $handlers = Hooks::getHandlers( $event );
1133                         $handlerInfo = array();
1134
1135                         wfSuppressWarnings();
1136
1137                         foreach ( $handlers as $handler ) {
1138                                 if ( is_array( $handler ) ) {
1139                                         if ( is_object( $handler[0] ) ) {
1140                                                 $info = get_class( $handler[0] );
1141                                         } else {
1142                                                 $info = $handler[0];
1143                                         }
1144
1145                                         if ( isset( $handler[1] ) ) {
1146                                                 $info .= '::' . $handler[1];
1147                                         }
1148                                 } elseif ( is_object( $handler ) ) {
1149                                         $info = get_class( $handler[0] );
1150                                         $info .= '::on' . $event;
1151                                 } else {
1152                                         $info = $handler;
1153                                 }
1154
1155                                 $handlerInfo[] = $info;
1156                         }
1157
1158                         wfRestoreWarnings();
1159
1160                         wfWarn( "Using obsolete hook $event via ContentHandler::runLegacyHooks()! Handlers: " .
1161                                 implode( ', ', $handlerInfo ), 2 );
1162                 }
1163
1164                 // convert Content objects to text
1165                 $contentObjects = array();
1166                 $contentTexts = array();
1167
1168                 foreach ( $args as $k => $v ) {
1169                         if ( $v instanceof Content ) {
1170                                 /* @var Content $v */
1171
1172                                 $contentObjects[$k] = $v;
1173
1174                                 $v = $v->serialize();
1175                                 $contentTexts[$k] = $v;
1176                                 $args[$k] = $v;
1177                         }
1178                 }
1179
1180                 // call the hook functions
1181                 $ok = Hooks::run( $event, $args );
1182
1183                 // see if the hook changed the text
1184                 foreach ( $contentTexts as $k => $orig ) {
1185                         /* @var Content $content */
1186
1187                         $modified = $args[$k];
1188                         $content = $contentObjects[$k];
1189
1190                         if ( $modified !== $orig ) {
1191                                 // text was changed, create updated Content object
1192                                 $content = $content->getContentHandler()->unserializeContent( $modified );
1193                         }
1194
1195                         $args[$k] = $content;
1196                 }
1197
1198                 return $ok;
1199         }
1200 }