Code Coverage |
||||||||||
Lines |
Functions and Methods |
Classes and Traits |
||||||||
| Total | |
41.43% |
58 / 140 |
|
6.67% |
1 / 15 |
CRAP | |
0.00% |
0 / 1 |
| WikitextContentHandler | |
41.43% |
58 / 140 |
|
6.67% |
1 / 15 |
251.82 | |
0.00% |
0 / 1 |
| __construct | |
0.00% |
0 / 8 |
|
0.00% |
0 / 1 |
2 | |||
| getContentClass | |
0.00% |
0 / 1 |
|
0.00% |
0 / 1 |
2 | |||
| makeRedirectContent | |
93.75% |
15 / 16 |
|
0.00% |
0 / 1 |
5.01 | |||
| supportsRedirects | |
0.00% |
0 / 1 |
|
0.00% |
0 / 1 |
2 | |||
| supportsSections | |
0.00% |
0 / 1 |
|
0.00% |
0 / 1 |
2 | |||
| isParserCacheSupported | |
0.00% |
0 / 1 |
|
0.00% |
0 / 1 |
2 | |||
| supportsPreloadContent | |
0.00% |
0 / 1 |
|
0.00% |
0 / 1 |
2 | |||
| getFileHandler | |
0.00% |
0 / 10 |
|
0.00% |
0 / 1 |
2 | |||
| getFieldsForSearchIndex | |
0.00% |
0 / 12 |
|
0.00% |
0 / 1 |
2 | |||
| getDataForSearchIndex | |
100.00% |
14 / 14 |
|
100.00% |
1 / 1 |
2 | |||
| serializeContent | |
75.00% |
3 / 4 |
|
0.00% |
0 / 1 |
2.06 | |||
| preSaveTransform | |
0.00% |
0 / 14 |
|
0.00% |
0 / 1 |
6 | |||
| preloadTransform | |
0.00% |
0 / 9 |
|
0.00% |
0 / 1 |
2 | |||
| extractRedirectTargetAndText | |
0.00% |
0 / 14 |
|
0.00% |
0 / 1 |
42 | |||
| fillParserOutput | |
76.47% |
26 / 34 |
|
0.00% |
0 / 1 |
7.64 | |||
| 1 | <?php |
| 2 | /** |
| 3 | * Content handler for wiki text pages. |
| 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 | |
| 26 | use MediaWiki\Content\Renderer\ContentParseParams; |
| 27 | use MediaWiki\Content\Transform\PreloadTransformParams; |
| 28 | use MediaWiki\Content\Transform\PreSaveTransformParams; |
| 29 | use MediaWiki\Languages\LanguageNameUtils; |
| 30 | use MediaWiki\Linker\LinkRenderer; |
| 31 | use MediaWiki\Logger\LoggerFactory; |
| 32 | use MediaWiki\Parser\MagicWordFactory; |
| 33 | use MediaWiki\Parser\ParserOutput; |
| 34 | use MediaWiki\Parser\ParserOutputFlags; |
| 35 | use MediaWiki\Parser\Parsoid\ParsoidParserFactory; |
| 36 | use MediaWiki\Revision\RevisionRecord; |
| 37 | use MediaWiki\Title\Title; |
| 38 | use MediaWiki\Title\TitleFactory; |
| 39 | use Wikimedia\UUID\GlobalIdGenerator; |
| 40 | |
| 41 | /** |
| 42 | * Content handler for wiki text pages. |
| 43 | * |
| 44 | * @ingroup Content |
| 45 | */ |
| 46 | class WikitextContentHandler extends TextContentHandler { |
| 47 | |
| 48 | private TitleFactory $titleFactory; |
| 49 | private ParserFactory $parserFactory; |
| 50 | private GlobalIdGenerator $globalIdGenerator; |
| 51 | private LanguageNameUtils $languageNameUtils; |
| 52 | private LinkRenderer $linkRenderer; |
| 53 | private MagicWordFactory $magicWordFactory; |
| 54 | private ParsoidParserFactory $parsoidParserFactory; |
| 55 | |
| 56 | public function __construct( |
| 57 | string $modelId, |
| 58 | TitleFactory $titleFactory, |
| 59 | ParserFactory $parserFactory, |
| 60 | GlobalIdGenerator $globalIdGenerator, |
| 61 | LanguageNameUtils $languageNameUtils, |
| 62 | LinkRenderer $linkRenderer, |
| 63 | MagicWordFactory $magicWordFactory, |
| 64 | ParsoidParserFactory $parsoidParserFactory |
| 65 | ) { |
| 66 | // $modelId should always be CONTENT_MODEL_WIKITEXT |
| 67 | parent::__construct( $modelId, [ CONTENT_FORMAT_WIKITEXT ] ); |
| 68 | $this->titleFactory = $titleFactory; |
| 69 | $this->parserFactory = $parserFactory; |
| 70 | $this->globalIdGenerator = $globalIdGenerator; |
| 71 | $this->languageNameUtils = $languageNameUtils; |
| 72 | $this->linkRenderer = $linkRenderer; |
| 73 | $this->magicWordFactory = $magicWordFactory; |
| 74 | $this->parsoidParserFactory = $parsoidParserFactory; |
| 75 | } |
| 76 | |
| 77 | /** |
| 78 | * @return class-string<WikitextContent> |
| 79 | */ |
| 80 | protected function getContentClass() { |
| 81 | return WikitextContent::class; |
| 82 | } |
| 83 | |
| 84 | /** |
| 85 | * Returns a WikitextContent object representing a redirect to the given destination page. |
| 86 | * |
| 87 | * @param Title $destination The page to redirect to. |
| 88 | * @param string $text Text to include in the redirect, if possible. |
| 89 | * |
| 90 | * @return Content |
| 91 | * |
| 92 | * @see ContentHandler::makeRedirectContent |
| 93 | */ |
| 94 | public function makeRedirectContent( Title $destination, $text = '' ) { |
| 95 | $optionalColon = ''; |
| 96 | |
| 97 | if ( $destination->getNamespace() === NS_CATEGORY ) { |
| 98 | $optionalColon = ':'; |
| 99 | } else { |
| 100 | $iw = $destination->getInterwiki(); |
| 101 | if ( $iw && $this->languageNameUtils->getLanguageName( $iw, |
| 102 | LanguageNameUtils::AUTONYMS, |
| 103 | LanguageNameUtils::DEFINED |
| 104 | ) ) { |
| 105 | $optionalColon = ':'; |
| 106 | } |
| 107 | } |
| 108 | |
| 109 | $mwRedir = $this->magicWordFactory->get( 'redirect' ); |
| 110 | $redirectText = $mwRedir->getSynonym( 0 ) . |
| 111 | ' [[' . $optionalColon . $destination->getFullText() . ']]'; |
| 112 | |
| 113 | if ( $text != '' ) { |
| 114 | $redirectText .= "\n" . $text; |
| 115 | } |
| 116 | |
| 117 | $class = $this->getContentClass(); |
| 118 | return new $class( $redirectText ); |
| 119 | } |
| 120 | |
| 121 | /** |
| 122 | * Returns true because wikitext supports redirects. |
| 123 | * |
| 124 | * @return bool Always true. |
| 125 | * |
| 126 | * @see ContentHandler::supportsRedirects |
| 127 | */ |
| 128 | public function supportsRedirects() { |
| 129 | return true; |
| 130 | } |
| 131 | |
| 132 | /** |
| 133 | * Returns true because wikitext supports sections. |
| 134 | * |
| 135 | * @return bool Always true. |
| 136 | * |
| 137 | * @see ContentHandler::supportsSections |
| 138 | */ |
| 139 | public function supportsSections() { |
| 140 | return true; |
| 141 | } |
| 142 | |
| 143 | /** |
| 144 | * Returns true, because wikitext supports caching using the |
| 145 | * ParserCache mechanism. |
| 146 | * |
| 147 | * @since 1.21 |
| 148 | * |
| 149 | * @return bool Always true. |
| 150 | * |
| 151 | * @see ContentHandler::isParserCacheSupported |
| 152 | */ |
| 153 | public function isParserCacheSupported() { |
| 154 | return true; |
| 155 | } |
| 156 | |
| 157 | /** @inheritDoc */ |
| 158 | public function supportsPreloadContent(): bool { |
| 159 | return true; |
| 160 | } |
| 161 | |
| 162 | /** |
| 163 | * @return FileContentHandler |
| 164 | */ |
| 165 | protected function getFileHandler() { |
| 166 | return new FileContentHandler( |
| 167 | $this->getModelID(), |
| 168 | $this->titleFactory, |
| 169 | $this->parserFactory, |
| 170 | $this->globalIdGenerator, |
| 171 | $this->languageNameUtils, |
| 172 | $this->linkRenderer, |
| 173 | $this->magicWordFactory, |
| 174 | $this->parsoidParserFactory |
| 175 | ); |
| 176 | } |
| 177 | |
| 178 | public function getFieldsForSearchIndex( SearchEngine $engine ) { |
| 179 | $fields = parent::getFieldsForSearchIndex( $engine ); |
| 180 | |
| 181 | $fields['heading'] = |
| 182 | $engine->makeSearchFieldMapping( 'heading', SearchIndexField::INDEX_TYPE_TEXT ); |
| 183 | $fields['heading']->setFlag( SearchIndexField::FLAG_SCORING ); |
| 184 | |
| 185 | $fields['auxiliary_text'] = |
| 186 | $engine->makeSearchFieldMapping( 'auxiliary_text', SearchIndexField::INDEX_TYPE_TEXT ); |
| 187 | |
| 188 | $fields['opening_text'] = |
| 189 | $engine->makeSearchFieldMapping( 'opening_text', SearchIndexField::INDEX_TYPE_TEXT ); |
| 190 | $fields['opening_text']->setFlag( |
| 191 | SearchIndexField::FLAG_SCORING | SearchIndexField::FLAG_NO_HIGHLIGHT |
| 192 | ); |
| 193 | // Until we have the full first-class content handler for files, we invoke it explicitly here |
| 194 | return array_merge( $fields, $this->getFileHandler()->getFieldsForSearchIndex( $engine ) ); |
| 195 | } |
| 196 | |
| 197 | public function getDataForSearchIndex( |
| 198 | WikiPage $page, |
| 199 | ParserOutput $parserOutput, |
| 200 | SearchEngine $engine, |
| 201 | ?RevisionRecord $revision = null |
| 202 | ) { |
| 203 | $fields = parent::getDataForSearchIndex( $page, $parserOutput, $engine, $revision ); |
| 204 | |
| 205 | $structure = new WikiTextStructure( $parserOutput ); |
| 206 | $fields['heading'] = $structure->headings(); |
| 207 | // text fields |
| 208 | $fields['opening_text'] = $structure->getOpeningText(); |
| 209 | $fields['text'] = $structure->getMainText(); // overwrites one from ContentHandler |
| 210 | $fields['auxiliary_text'] = $structure->getAuxiliaryText(); |
| 211 | $fields['defaultsort'] = $structure->getDefaultSort(); |
| 212 | $fields['file_text'] = null; |
| 213 | |
| 214 | // Until we have the full first-class content handler for files, we invoke it explicitly here |
| 215 | if ( $page->getTitle()->getNamespace() === NS_FILE ) { |
| 216 | $fields = array_merge( |
| 217 | $fields, |
| 218 | $this->getFileHandler()->getDataForSearchIndex( $page, $parserOutput, $engine, $revision ) |
| 219 | ); |
| 220 | } |
| 221 | return $fields; |
| 222 | } |
| 223 | |
| 224 | /** |
| 225 | * Returns the content's text as-is. |
| 226 | * |
| 227 | * @param Content $content |
| 228 | * @param string|null $format The serialization format to check |
| 229 | * |
| 230 | * @return mixed |
| 231 | */ |
| 232 | public function serializeContent( Content $content, $format = null ) { |
| 233 | $this->checkFormat( $format ); |
| 234 | |
| 235 | // NOTE: MessageContent also uses CONTENT_MODEL_WIKITEXT, but it's not a TextContent! |
| 236 | // Perhaps MessageContent should use a separate ContentHandler instead. |
| 237 | if ( $content instanceof MessageContent ) { |
| 238 | return $content->getMessage()->plain(); |
| 239 | } |
| 240 | |
| 241 | return parent::serializeContent( $content, $format ); |
| 242 | } |
| 243 | |
| 244 | public function preSaveTransform( |
| 245 | Content $content, |
| 246 | PreSaveTransformParams $pstParams |
| 247 | ): Content { |
| 248 | '@phan-var WikitextContent $content'; |
| 249 | $text = $content->getText(); |
| 250 | |
| 251 | $parser = $this->parserFactory->getInstance(); |
| 252 | $pst = $parser->preSaveTransform( |
| 253 | $text, |
| 254 | $pstParams->getPage(), |
| 255 | $pstParams->getUser(), |
| 256 | $pstParams->getParserOptions() |
| 257 | ); |
| 258 | |
| 259 | if ( $text === $pst ) { |
| 260 | return $content; |
| 261 | } |
| 262 | |
| 263 | $contentClass = $this->getContentClass(); |
| 264 | $ret = new $contentClass( $pst ); |
| 265 | $ret->setPreSaveTransformFlags( $parser->getOutput()->getAllFlags() ); |
| 266 | return $ret; |
| 267 | } |
| 268 | |
| 269 | /** |
| 270 | * Returns a Content object with preload transformations applied (or this |
| 271 | * object if no transformations apply). |
| 272 | * |
| 273 | * @param Content $content |
| 274 | * @param PreloadTransformParams $pltParams |
| 275 | * |
| 276 | * @return Content |
| 277 | */ |
| 278 | public function preloadTransform( |
| 279 | Content $content, |
| 280 | PreloadTransformParams $pltParams |
| 281 | ): Content { |
| 282 | '@phan-var WikitextContent $content'; |
| 283 | $text = $content->getText(); |
| 284 | |
| 285 | $plt = $this->parserFactory->getInstance()->getPreloadText( |
| 286 | $text, |
| 287 | $pltParams->getPage(), |
| 288 | $pltParams->getParserOptions(), |
| 289 | $pltParams->getParams() |
| 290 | ); |
| 291 | |
| 292 | $contentClass = $this->getContentClass(); |
| 293 | return new $contentClass( $plt ); |
| 294 | } |
| 295 | |
| 296 | /** |
| 297 | * Extract the redirect target and the remaining text on the page. |
| 298 | * |
| 299 | * @since 1.41 (used to be a method on WikitextContent since 1.23) |
| 300 | * |
| 301 | * @return array List of two elements: LinkTarget|null and WikitextContent object. |
| 302 | */ |
| 303 | public function extractRedirectTargetAndText( WikitextContent $content ): array { |
| 304 | $redir = $this->magicWordFactory->get( 'redirect' ); |
| 305 | $text = ltrim( $content->getText() ); |
| 306 | |
| 307 | if ( !$redir->matchStartAndRemove( $text ) ) { |
| 308 | return [ null, $content ]; |
| 309 | } |
| 310 | |
| 311 | // Extract the first link and see if it's usable |
| 312 | // Ensure that it really does come directly after #REDIRECT |
| 313 | // Some older redirects included a colon, so don't freak about that! |
| 314 | $m = []; |
| 315 | if ( preg_match( '!^\s*:?\s*\[{2}(.*?)(?:\|.*?)?\]{2}\s*!', $text, $m ) ) { |
| 316 | // Strip preceding colon used to "escape" categories, etc. |
| 317 | // and URL-decode links |
| 318 | if ( strpos( $m[1], '%' ) !== false ) { |
| 319 | // Match behavior of inline link parsing here; |
| 320 | $m[1] = rawurldecode( ltrim( $m[1], ':' ) ); |
| 321 | } |
| 322 | |
| 323 | // TODO: Move isValidRedirectTarget() out Title, so we can use a TitleValue here. |
| 324 | $title = $this->titleFactory->newFromText( $m[1] ); |
| 325 | |
| 326 | // If the title is a redirect to bad special pages or is invalid, return null |
| 327 | if ( !$title instanceof Title || !$title->isValidRedirectTarget() ) { |
| 328 | return [ null, $content ]; |
| 329 | } |
| 330 | |
| 331 | $remainingContent = new WikitextContent( substr( $text, strlen( $m[0] ) ) ); |
| 332 | return [ $title, $remainingContent ]; |
| 333 | } |
| 334 | |
| 335 | return [ null, $content ]; |
| 336 | } |
| 337 | |
| 338 | /** |
| 339 | * Returns a ParserOutput object resulting from parsing the content's text |
| 340 | * using the global Parser service. |
| 341 | * |
| 342 | * @since 1.38 |
| 343 | * @param Content $content |
| 344 | * @param ContentParseParams $cpoParams |
| 345 | * @param ParserOutput &$parserOutput The output object to fill (reference). |
| 346 | */ |
| 347 | protected function fillParserOutput( |
| 348 | Content $content, |
| 349 | ContentParseParams $cpoParams, |
| 350 | ParserOutput &$parserOutput |
| 351 | ) { |
| 352 | '@phan-var WikitextContent $content'; |
| 353 | $title = $this->titleFactory->newFromPageReference( $cpoParams->getPage() ); |
| 354 | $parserOptions = $cpoParams->getParserOptions(); |
| 355 | $revId = $cpoParams->getRevId(); |
| 356 | |
| 357 | [ $redir, $contentWithoutRedirect ] = $this->extractRedirectTargetAndText( $content ); |
| 358 | if ( $parserOptions->getUseParsoid() ) { |
| 359 | $parser = $this->parsoidParserFactory->create(); |
| 360 | // Parsoid renders the #REDIRECT magic word as an invisible |
| 361 | // <link> tag and doesn't require it to be stripped. |
| 362 | // T349087: ...and in fact, RESTBase relies on getting |
| 363 | // redirect information from this <link> tag, so it needs |
| 364 | // to be present. |
| 365 | $text = $content->getText(); |
| 366 | } else { |
| 367 | // The legacy parser requires the #REDIRECT magic word to |
| 368 | // be stripped from the content before parsing. |
| 369 | $parser = $this->parserFactory->getInstance(); |
| 370 | $text = $contentWithoutRedirect->getText(); |
| 371 | } |
| 372 | |
| 373 | $time = -microtime( true ); |
| 374 | |
| 375 | $parserOutput = $parser |
| 376 | ->parse( $text, $title, $parserOptions, true, true, $revId ); |
| 377 | $time += microtime( true ); |
| 378 | |
| 379 | // Timing hack |
| 380 | if ( $time > 3 ) { |
| 381 | // TODO: Use Parser's logger (once it has one) |
| 382 | $channel = $parserOptions->getUseParsoid() ? 'slow-parsoid' : 'slow-parse'; |
| 383 | $logger = LoggerFactory::getInstance( $channel ); |
| 384 | $logger->info( 'Parsing {title} was slow, took {time} seconds', [ |
| 385 | 'time' => number_format( $time, 2 ), |
| 386 | 'title' => (string)$title, |
| 387 | 'trigger' => $parserOptions->getRenderReason(), |
| 388 | ] ); |
| 389 | } |
| 390 | |
| 391 | // T330667: Record the fact that we used the value of |
| 392 | // 'useParsoid' to influence this parse. Note that |
| 393 | // ::getUseParsoid() has a side-effect on $parserOutput here |
| 394 | // which didn't occur when we called ::getUseParsoid() earlier |
| 395 | // because $parserOutput didn't exist at that time. |
| 396 | $parserOptions->getUseParsoid(); |
| 397 | |
| 398 | // Add redirect indicator at the top |
| 399 | if ( $redir ) { |
| 400 | // Make sure to include the redirect link in pagelinks |
| 401 | $parserOutput->addLink( $redir ); |
| 402 | if ( $cpoParams->getGenerateHtml() ) { |
| 403 | $parserOutput->setRedirectHeader( |
| 404 | $this->linkRenderer->makeRedirectHeader( |
| 405 | $title->getPageLanguage(), $redir, false |
| 406 | ) |
| 407 | ); |
| 408 | $parserOutput->addModuleStyles( [ 'mediawiki.action.view.redirectPage' ] ); |
| 409 | } else { |
| 410 | $parserOutput->setRawText( null ); |
| 411 | } |
| 412 | } |
| 413 | |
| 414 | // Pass along user-signature flag |
| 415 | if ( in_array( 'user-signature', $content->getPreSaveTransformFlags() ) ) { |
| 416 | $parserOutput->setOutputFlag( ParserOutputFlags::USER_SIGNATURE ); |
| 417 | } |
| 418 | } |
| 419 | } |