Code Coverage |
||||||||||
Lines |
Functions and Methods |
Classes and Traits |
||||||||
| Total | |
25.89% |
29 / 112 |
|
0.00% |
0 / 6 |
CRAP | |
0.00% |
0 / 1 |
| ParsoidParser | |
25.89% |
29 / 112 |
|
0.00% |
0 / 6 |
165.92 | |
0.00% |
0 / 1 |
| __construct | |
0.00% |
0 / 5 |
|
0.00% |
0 / 1 |
2 | |||
| setParsoidRenderID | |
0.00% |
0 / 4 |
|
0.00% |
0 / 1 |
2 | |||
| genParserOutput | |
0.00% |
0 / 43 |
|
0.00% |
0 / 1 |
30 | |||
| parse | |
93.55% |
29 / 31 |
|
0.00% |
0 / 1 |
7.01 | |||
| parseFakeRevision | |
0.00% |
0 / 12 |
|
0.00% |
0 / 1 |
12 | |||
| makeLimitReport | |
0.00% |
0 / 17 |
|
0.00% |
0 / 1 |
6 | |||
| 1 | <?php |
| 2 | |
| 3 | namespace MediaWiki\Parser\Parsoid; |
| 4 | |
| 5 | use MediaWiki\Languages\LanguageConverterFactory; |
| 6 | use MediaWiki\MediaWikiServices; |
| 7 | use MediaWiki\Page\PageReference; |
| 8 | use MediaWiki\Parser\ParserOutput; |
| 9 | use MediaWiki\Parser\Parsoid\Config\PageConfigFactory; |
| 10 | use MediaWiki\Revision\MutableRevisionRecord; |
| 11 | use MediaWiki\Revision\RevisionRecord; |
| 12 | use MediaWiki\Revision\SlotRecord; |
| 13 | use MediaWiki\Title\Title; |
| 14 | use ParserFactory; |
| 15 | use ParserOptions; |
| 16 | use Wikimedia\Assert\Assert; |
| 17 | use Wikimedia\Parsoid\Config\PageConfig; |
| 18 | use Wikimedia\Parsoid\Parsoid; |
| 19 | use Wikimedia\UUID\GlobalIdGenerator; |
| 20 | use WikitextContent; |
| 21 | |
| 22 | /** |
| 23 | * Parser implementation which uses Parsoid. |
| 24 | * |
| 25 | * Currently incomplete; see T236809 for the long-term plan. |
| 26 | * |
| 27 | * @since 1.41 |
| 28 | * @unstable since 1.41; see T236809 for plan. |
| 29 | */ |
| 30 | class ParsoidParser /* eventually this will extend \Parser */ { |
| 31 | /** |
| 32 | * @unstable |
| 33 | * This should not be used widely right now since this may go away. |
| 34 | * This is being added to support DiscussionTools with Parsoid HTML |
| 35 | * and after initial exploration, this may be implemented differently. |
| 36 | */ |
| 37 | public const PARSOID_TITLE_KEY = "parsoid:title-dbkey"; |
| 38 | private Parsoid $parsoid; |
| 39 | private PageConfigFactory $pageConfigFactory; |
| 40 | private LanguageConverterFactory $languageConverterFactory; |
| 41 | private ParserFactory $legacyParserFactory; |
| 42 | private GlobalIdGenerator $globalIdGenerator; |
| 43 | |
| 44 | /** |
| 45 | * @param Parsoid $parsoid |
| 46 | * @param PageConfigFactory $pageConfigFactory |
| 47 | * @param LanguageConverterFactory $languageConverterFactory |
| 48 | * @param ParserFactory $legacyParserFactory |
| 49 | * @param GlobalIdGenerator $globalIdGenerator |
| 50 | */ |
| 51 | public function __construct( |
| 52 | Parsoid $parsoid, |
| 53 | PageConfigFactory $pageConfigFactory, |
| 54 | LanguageConverterFactory $languageConverterFactory, |
| 55 | ParserFactory $legacyParserFactory, |
| 56 | GlobalIdGenerator $globalIdGenerator |
| 57 | ) { |
| 58 | $this->parsoid = $parsoid; |
| 59 | $this->pageConfigFactory = $pageConfigFactory; |
| 60 | $this->languageConverterFactory = $languageConverterFactory; |
| 61 | $this->legacyParserFactory = $legacyParserFactory; |
| 62 | $this->globalIdGenerator = $globalIdGenerator; |
| 63 | } |
| 64 | |
| 65 | /** |
| 66 | * API users expect a ParsoidRenderID value set in the parser output's extension data. |
| 67 | * @param PageConfig $pageConfig |
| 68 | * @param ParserOutput $parserOutput |
| 69 | */ |
| 70 | private function setParsoidRenderID( PageConfig $pageConfig, ParserOutput $parserOutput ): void { |
| 71 | $parserOutput->setRenderId( $this->globalIdGenerator->newUUIDv1() ); |
| 72 | $parserOutput->setCacheRevisionId( $pageConfig->getRevisionId() ); |
| 73 | $parserOutput->setRevisionTimestamp( $pageConfig->getRevisionTimestamp() ); |
| 74 | $parserOutput->setCacheTime( wfTimestampNow() ); |
| 75 | } |
| 76 | |
| 77 | /** |
| 78 | * Internal helper to avoid code deuplication across two methods |
| 79 | * |
| 80 | * @param PageConfig $pageConfig |
| 81 | * @param ParserOptions $options |
| 82 | * @return ParserOutput |
| 83 | */ |
| 84 | private function genParserOutput( |
| 85 | PageConfig $pageConfig, ParserOptions $options |
| 86 | ): ParserOutput { |
| 87 | $parserOutput = new ParserOutput(); |
| 88 | |
| 89 | // Parsoid itself does not vary output by parser options right now. |
| 90 | // But, ensure that any option use by extensions, parser functions, |
| 91 | // recursive parses, or (in the unlikely future scenario) Parsoid itself |
| 92 | // are recorded as used. |
| 93 | $options->registerWatcher( [ $parserOutput, 'recordOption' ] ); |
| 94 | |
| 95 | // The enable/disable logic here matches that in Parser::internalParseHalfParsed(), |
| 96 | // although __NOCONTENTCONVERT__ is handled internal to Parsoid. |
| 97 | // |
| 98 | // T349137: It might be preferable to handle __NOCONTENTCONVERT__ here rather than |
| 99 | // by inspecting the DOM inside Parsoid. That will come in a separate patch. |
| 100 | $htmlVariantLanguage = null; |
| 101 | if ( !( $options->getDisableContentConversion() || $options->getInterfaceMessage() ) ) { |
| 102 | // NOTES (some of these are TODOs for read views integration) |
| 103 | // 1. This html variant conversion is a pre-cache transform. HtmlOutputRendererHelper |
| 104 | // has another variant conversion that is a post-cache transform based on the |
| 105 | // 'Accept-Language' header. If that header is set, there is really no reason to |
| 106 | // do this conversion here. So, eventually, we are likely to either not pass in |
| 107 | // the htmlVariantLanguage option below OR disable language conversion from the |
| 108 | // wt2html path in Parsoid and this and the Accept-Language variant conversion |
| 109 | // both would have to be handled as post-cache transforms. |
| 110 | // |
| 111 | // 2. Parser.php calls convert() which computes a preferred variant from the |
| 112 | // target language. But, we cannot do that unconditionally here because REST API |
| 113 | // requests specify the exact variant via the 'Content-Language' header. |
| 114 | // |
| 115 | // For Parsoid page views, either the callers will have to compute the |
| 116 | // preferred variant and set it in ParserOptions OR the REST API will have |
| 117 | // to set some other flag indicating that the preferred variant should not |
| 118 | // be computed. For now, I am adding a temporary hack, but this should be |
| 119 | // replaced with something more sensible (T267067). |
| 120 | // |
| 121 | // 3. Additionally, Parsoid's callers will have to set targetLanguage in ParserOptions |
| 122 | // to mimic the logic in Parser.php (missing right now). |
| 123 | $langCode = $pageConfig->getPageLanguageBcp47(); |
| 124 | if ( $options->getRenderReason() === 'page-view' ) { // TEMPORARY HACK |
| 125 | $langFactory = MediaWikiServices::getInstance()->getLanguageFactory(); |
| 126 | $lang = $langFactory->getLanguage( $langCode ); |
| 127 | $langConv = $this->languageConverterFactory->getLanguageConverter( $lang ); |
| 128 | $htmlVariantLanguage = $langFactory->getLanguage( $langConv->getPreferredVariant() ); |
| 129 | } else { |
| 130 | $htmlVariantLanguage = $langCode; |
| 131 | } |
| 132 | } |
| 133 | |
| 134 | $defaultOptions = [ |
| 135 | 'pageBundle' => true, |
| 136 | 'wrapSections' => true, |
| 137 | 'logLinterData' => true, |
| 138 | 'body_only' => false, |
| 139 | 'htmlVariantLanguage' => $htmlVariantLanguage, |
| 140 | 'offsetType' => 'byte', |
| 141 | 'outputContentVersion' => Parsoid::defaultHTMLVersion() |
| 142 | ]; |
| 143 | |
| 144 | $parserOutput->resetParseStartTime(); |
| 145 | |
| 146 | // This can throw ClientError or ResourceLimitExceededException. |
| 147 | // Callers are responsible for figuring out how to handle them. |
| 148 | $pageBundle = $this->parsoid->wikitext2html( |
| 149 | $pageConfig, |
| 150 | $defaultOptions, |
| 151 | $headers, |
| 152 | $parserOutput ); |
| 153 | |
| 154 | $parserOutput = PageBundleParserOutputConverter::parserOutputFromPageBundle( $pageBundle, $parserOutput ); |
| 155 | |
| 156 | // Record the page title in dbkey form so that post-cache transforms |
| 157 | // have access to the title. |
| 158 | $parserOutput->setExtensionData( |
| 159 | self::PARSOID_TITLE_KEY, |
| 160 | Title::newFromLinkTarget( $pageConfig->getLinkTarget() )->getPrefixedDBkey() |
| 161 | ); |
| 162 | |
| 163 | // Register a watcher again because the $parserOuptut arg |
| 164 | // and $parserOutput return value above are different objects! |
| 165 | $options->registerWatcher( [ $parserOutput, 'recordOption' ] ); |
| 166 | |
| 167 | $revId = $pageConfig->getRevisionId(); |
| 168 | if ( $revId !== null ) { |
| 169 | // T350538: This shouldn't be necessary so long as ContentRenderer |
| 170 | // is involved in the call chain somewhere, and should be turned |
| 171 | // into an assertion (and ::setParsoidRenderID() removed). |
| 172 | $this->setParsoidRenderID( $pageConfig, $parserOutput ); |
| 173 | } |
| 174 | |
| 175 | $parserOutput->setFromParserOptions( $options ); |
| 176 | |
| 177 | $parserOutput->recordTimeProfile(); |
| 178 | $this->makeLimitReport( $options, $parserOutput ); |
| 179 | |
| 180 | // Add Parsoid skinning module |
| 181 | $parserOutput->addModuleStyles( [ 'mediawiki.skinning.content.parsoid' ] ); |
| 182 | |
| 183 | // Record Parsoid version in extension data; this allows |
| 184 | // us to use the onRejectParserCacheValue hook to selectively |
| 185 | // expire "bad" generated content in the event of a rollback. |
| 186 | $parserOutput->setExtensionData( |
| 187 | 'core:parsoid-version', Parsoid::version() |
| 188 | ); |
| 189 | |
| 190 | return $parserOutput; |
| 191 | } |
| 192 | |
| 193 | /** |
| 194 | * Convert wikitext to HTML |
| 195 | * Do not call this function recursively. |
| 196 | * |
| 197 | * @param string $text Text we want to parse |
| 198 | * @param-taint $text escapes_htmlnoent |
| 199 | * @param PageReference $page |
| 200 | * @param ParserOptions $options |
| 201 | * @param bool $linestart |
| 202 | * @param bool $clearState |
| 203 | * @param int|null $revId ID of the revision being rendered. This is used to render |
| 204 | * REVISION* magic words. 0 means that any current revision will be used. Null means |
| 205 | * that {{REVISIONID}}/{{REVISIONUSER}} will be empty and {{REVISIONTIMESTAMP}} will |
| 206 | * use the current timestamp. |
| 207 | * @return ParserOutput |
| 208 | * @return-taint escaped |
| 209 | * @unstable since 1.41 |
| 210 | */ |
| 211 | public function parse( |
| 212 | string $text, PageReference $page, ParserOptions $options, |
| 213 | bool $linestart = true, bool $clearState = true, ?int $revId = null |
| 214 | ): ParserOutput { |
| 215 | Assert::invariant( $linestart, '$linestart=false is not yet supported' ); |
| 216 | Assert::invariant( $clearState, '$clearState=false is not yet supported' ); |
| 217 | $title = Title::newFromPageReference( $page ); |
| 218 | $lang = $options->getTargetLanguage(); |
| 219 | if ( $lang === null && $options->getInterfaceMessage() ) { |
| 220 | $lang = $options->getUserLangObj(); |
| 221 | } |
| 222 | $pageConfig = $revId === null ? null : $this->pageConfigFactory->create( |
| 223 | $title, |
| 224 | $options->getUserIdentity(), |
| 225 | $revId, |
| 226 | null, // unused |
| 227 | $lang // defaults to title page language if null |
| 228 | ); |
| 229 | if ( !( $pageConfig && $pageConfig->getPageMainContent() === $text ) ) { |
| 230 | // This is a bit awkward! But we really need to parse $text, which |
| 231 | // may or may not correspond to the $revId provided! |
| 232 | // T332928 suggests one solution: splitting the "have revid" |
| 233 | // callers from the "bare text, no associated revision" callers. |
| 234 | $revisionRecord = new MutableRevisionRecord( $title ); |
| 235 | if ( $revId !== null ) { |
| 236 | $revisionRecord->setId( $revId ); |
| 237 | } |
| 238 | $revisionRecord->setSlot( |
| 239 | SlotRecord::newUnsaved( |
| 240 | SlotRecord::MAIN, |
| 241 | new WikitextContent( $text ) |
| 242 | ) |
| 243 | ); |
| 244 | $pageConfig = $this->pageConfigFactory->create( |
| 245 | $title, |
| 246 | $options->getUserIdentity(), |
| 247 | $revisionRecord, |
| 248 | null, // unused |
| 249 | $lang // defaults to title page language if null |
| 250 | ); |
| 251 | } |
| 252 | |
| 253 | return $this->genParserOutput( $pageConfig, $options ); |
| 254 | } |
| 255 | |
| 256 | /** |
| 257 | * @internal |
| 258 | * |
| 259 | * Convert custom wikitext (stored in main slot of the $fakeRev arg) to HTML. |
| 260 | * Callers are expected NOT to stuff the result into ParserCache. |
| 261 | * |
| 262 | * @param RevisionRecord $fakeRev Revision to parse |
| 263 | * @param PageReference $page |
| 264 | * @param ParserOptions $options |
| 265 | * @return ParserOutput |
| 266 | * @unstable since 1.41 |
| 267 | */ |
| 268 | public function parseFakeRevision( |
| 269 | RevisionRecord $fakeRev, PageReference $page, ParserOptions $options |
| 270 | ): ParserOutput { |
| 271 | $title = Title::newFromPageReference( $page ); |
| 272 | $lang = $options->getTargetLanguage(); |
| 273 | if ( $lang === null && $options->getInterfaceMessage() ) { |
| 274 | $lang = $options->getUserLangObj(); |
| 275 | } |
| 276 | $pageConfig = $this->pageConfigFactory->create( |
| 277 | $title, |
| 278 | $options->getUserIdentity(), |
| 279 | $fakeRev, |
| 280 | null, // unused |
| 281 | $lang // defaults to title page language if null |
| 282 | ); |
| 283 | |
| 284 | return $this->genParserOutput( $pageConfig, $options ); |
| 285 | } |
| 286 | |
| 287 | /** |
| 288 | * Set the limit report data in the current ParserOutput. |
| 289 | * This is ported from Parser::makeLimitReport() and should eventually |
| 290 | * use the method from the superclass directly. |
| 291 | */ |
| 292 | protected function makeLimitReport( |
| 293 | ParserOptions $parserOptions, ParserOutput $parserOutput |
| 294 | ) { |
| 295 | $maxIncludeSize = $parserOptions->getMaxIncludeSize(); |
| 296 | |
| 297 | $cpuTime = $parserOutput->getTimeProfile( 'cpu' ); |
| 298 | if ( $cpuTime !== null ) { |
| 299 | $parserOutput->setLimitReportData( 'limitreport-cputime', |
| 300 | sprintf( "%.3f", $cpuTime ) |
| 301 | ); |
| 302 | } |
| 303 | |
| 304 | $wallTime = $parserOutput->getTimeProfile( 'wall' ); |
| 305 | $parserOutput->setLimitReportData( 'limitreport-walltime', |
| 306 | sprintf( "%.3f", $wallTime ) |
| 307 | ); |
| 308 | |
| 309 | $parserOutput->setLimitReportData( 'limitreport-timingprofile', [ 'not yet supported' ] ); |
| 310 | |
| 311 | // Add other cache related metadata |
| 312 | $parserOutput->setLimitReportData( 'cachereport-timestamp', |
| 313 | $parserOutput->getCacheTime() ); |
| 314 | $parserOutput->setLimitReportData( 'cachereport-ttl', |
| 315 | $parserOutput->getCacheExpiry() ); |
| 316 | $parserOutput->setLimitReportData( 'cachereport-transientcontent', |
| 317 | $parserOutput->hasReducedExpiry() ); |
| 318 | } |
| 319 | |
| 320 | } |