WikitextContentHandler.php

Go to the documentation of this file.

<?php
use MediaWiki\Content\Renderer\ContentParseParams;
use MediaWiki\Content\Transform\PreloadTransformParams;
use MediaWiki\Content\Transform\PreSaveTransformParams;
use MediaWiki\Languages\LanguageNameUtils;
use MediaWiki\Linker\LinkRenderer;
use MediaWiki\Parser\MagicWordFactory;
use MediaWiki\Parser\ParserOutputFlags;
use MediaWiki\Parser\Parsoid\ParsoidParserFactory;
use MediaWiki\Revision\RevisionRecord;
use MediaWiki\Title\Title;
use MediaWiki\Title\TitleFactory;
use Wikimedia\UUID\GlobalIdGenerator;
 
class WikitextContentHandler extends TextContentHandler {
 
    private $titleFactory;
 
    private $parserFactory;
 
    private $globalIdGenerator;
 
    private $languageNameUtils;
 
    private $linkRenderer;
 
    private $magicWordFactory;
 
    private $parsoidParserFactory;
 
    public function __construct(
        string $modelId,
        TitleFactory $titleFactory,
        ParserFactory $parserFactory,
        GlobalIdGenerator $globalIdGenerator,
        LanguageNameUtils $languageNameUtils,
        LinkRenderer $linkRenderer,
        MagicWordFactory $magicWordFactory,
        ParsoidParserFactory $parsoidParserFactory
    ) {
        // $modelId should always be CONTENT_MODEL_WIKITEXT
        parent::__construct( $modelId, [ CONTENT_FORMAT_WIKITEXT ] );
        $this->titleFactory = $titleFactory;
        $this->parserFactory = $parserFactory;
        $this->globalIdGenerator = $globalIdGenerator;
        $this->languageNameUtils = $languageNameUtils;
        $this->linkRenderer = $linkRenderer;
        $this->magicWordFactory = $magicWordFactory;
        $this->parsoidParserFactory = $parsoidParserFactory;
    }
 
    protected function getContentClass() {
        return WikitextContent::class;
    }
 
    public function makeRedirectContent( Title $destination, $text = '' ) {
        $optionalColon = '';
 
        if ( $destination->getNamespace() === NS_CATEGORY ) {
            $optionalColon = ':';
        } else {
            $iw = $destination->getInterwiki();
            if ( $iw && $this->languageNameUtils->getLanguageName( $iw,
                        LanguageNameUtils::AUTONYMS,
                        LanguageNameUtils::DEFINED )
            ) {
                $optionalColon = ':';
            }
        }
 
        $mwRedir = $this->magicWordFactory->get( 'redirect' );
        $redirectText = $mwRedir->getSynonym( 0 ) .
            ' [[' . $optionalColon . $destination->getFullText() . ']]';
 
        if ( $text != '' ) {
            $redirectText .= "\n" . $text;
        }
 
        $class = $this->getContentClass();
        return new $class( $redirectText );
    }
 
    public function supportsRedirects() {
        return true;
    }
 
    public function supportsSections() {
        return true;
    }
 
    public function isParserCacheSupported() {
        return true;
    }
 
    public function supportsPreloadContent(): bool {
        return true;
    }
 
    protected function getFileHandler() {
        return new FileContentHandler(
            $this->getModelID(),
            $this->titleFactory,
            $this->parserFactory,
            $this->globalIdGenerator,
            $this->languageNameUtils,
            $this->linkRenderer,
            $this->magicWordFactory,
            $this->parsoidParserFactory
        );
    }
 
    public function getFieldsForSearchIndex( SearchEngine $engine ) {
        $fields = parent::getFieldsForSearchIndex( $engine );
 
        $fields['heading'] =
            $engine->makeSearchFieldMapping( 'heading', SearchIndexField::INDEX_TYPE_TEXT );
        $fields['heading']->setFlag( SearchIndexField::FLAG_SCORING );
 
        $fields['auxiliary_text'] =
            $engine->makeSearchFieldMapping( 'auxiliary_text', SearchIndexField::INDEX_TYPE_TEXT );
 
        $fields['opening_text'] =
            $engine->makeSearchFieldMapping( 'opening_text', SearchIndexField::INDEX_TYPE_TEXT );
        $fields['opening_text']->setFlag(
            SearchIndexField::FLAG_SCORING | SearchIndexField::FLAG_NO_HIGHLIGHT
        );
        // Until we have full first-class content handler for files, we invoke it explicitly here
        $fields = array_merge( $fields, $this->getFileHandler()->getFieldsForSearchIndex( $engine ) );
 
        return $fields;
    }
 
    public function getDataForSearchIndex(
        WikiPage $page,
        ParserOutput $parserOutput,
        SearchEngine $engine,
        ?RevisionRecord $revision = null
    ) {
        $fields = parent::getDataForSearchIndex( $page, $parserOutput, $engine, $revision );
 
        $structure = new WikiTextStructure( $parserOutput );
        $fields['heading'] = $structure->headings();
        // text fields
        $fields['opening_text'] = $structure->getOpeningText();
        $fields['text'] = $structure->getMainText(); // overwrites one from ContentHandler
        $fields['auxiliary_text'] = $structure->getAuxiliaryText();
        $fields['defaultsort'] = $structure->getDefaultSort();
        $fields['file_text'] = null;
 
        // Until we have full first-class content handler for files, we invoke it explicitly here
        if ( $page->getTitle()->getNamespace() === NS_FILE ) {
            $fields = array_merge( $fields,
                    $this->getFileHandler()->getDataForSearchIndex( $page, $parserOutput, $engine, $revision ) );
        }
        return $fields;
    }
 
    public function serializeContent( Content $content, $format = null ) {
        $this->checkFormat( $format );
 
        // NOTE: MessageContent also uses CONTENT_MODEL_WIKITEXT, but it's not a TextContent!
        // Perhaps MessageContent should use a separate ContentHandler instead.
        if ( $content instanceof MessageContent ) {
            return $content->getMessage()->plain();
        }
 
        return parent::serializeContent( $content, $format );
    }
 
    public function preSaveTransform(
        Content $content,
        PreSaveTransformParams $pstParams
    ): Content {
        '@phan-var WikitextContent $content';
        $text = $content->getText();
 
        $parser = $this->parserFactory->getInstance();
        $pst = $parser->preSaveTransform(
            $text,
            $pstParams->getPage(),
            $pstParams->getUser(),
            $pstParams->getParserOptions()
        );
 
        if ( $text === $pst ) {
            return $content;
        }
 
        $contentClass = $this->getContentClass();
        $ret = new $contentClass( $pst );
        $ret->setPreSaveTransformFlags( $parser->getOutput()->getAllFlags() );
        return $ret;
    }
 
    public function preloadTransform(
        Content $content,
        PreloadTransformParams $pltParams
    ): Content {
        '@phan-var WikitextContent $content';
        $text = $content->getText();
 
        $plt = $this->parserFactory->getInstance()
            ->getPreloadText(
                $text,
                $pltParams->getPage(),
                $pltParams->getParserOptions(),
                $pltParams->getParams()
            );
 
        $contentClass = $this->getContentClass();
        return new $contentClass( $plt );
    }
 
    public function extractRedirectTargetAndText( WikitextContent $content ): array {
        $redir = $this->magicWordFactory->get( 'redirect' );
        $text = ltrim( $content->getText() );
 
        if ( !$redir->matchStartAndRemove( $text ) ) {
            return [ null, $content ];
        }
 
        // Extract the first link and see if it's usable
        // Ensure that it really does come directly after #REDIRECT
        // Some older redirects included a colon, so don't freak about that!
        $m = [];
        if ( preg_match( '!^\s*:?\s*\[{2}(.*?)(?:\|.*?)?\]{2}\s*!', $text, $m ) ) {
            // Strip preceding colon used to "escape" categories, etc.
            // and URL-decode links
            if ( strpos( $m[1], '%' ) !== false ) {
                // Match behavior of inline link parsing here;
                $m[1] = rawurldecode( ltrim( $m[1], ':' ) );
            }
 
            // TODO: Move isValidRedirectTarget() out Title, so we can use a TitleValue here.
            $title = $this->titleFactory->newFromText( $m[1] );
 
            // If the title is a redirect to bad special pages or is invalid, return null
            if ( !$title instanceof Title || !$title->isValidRedirectTarget() ) {
                return [ null, $content ];
            }
 
            $remainingContent = new WikitextContent( substr( $text, strlen( $m[0] ) ) );
            return [ $title, $remainingContent ];
        }
 
        return [ null, $content ];
    }
 
    protected function fillParserOutput(
        Content $content,
        ContentParseParams $cpoParams,
        ParserOutput &$parserOutput
    ) {
        '@phan-var WikitextContent $content';
        $title = $this->titleFactory->newFromPageReference( $cpoParams->getPage() );
        $parserOptions = $cpoParams->getParserOptions();
        $revId = $cpoParams->getRevId();
 
        [ $redir, $contentWithoutRedirect ] = $this->extractRedirectTargetAndText( $content );
        if ( $parserOptions->getUseParsoid() ) {
            $parser = $this->parsoidParserFactory->create();
            // Parsoid renders the #REDIRECT magic word as an invisible
            // <link> tag and doesn't require it to be stripped.
            // T349087: ...and in fact, RESTBase relies on getting
            // redirect information from this <link> tag, so it needs
            // to be present.
            $text = $content->getText();
        } else {
            // The legacy parser requires the #REDIRECT magic word to
            // be stripped from the content before parsing.
            $parser = $this->parserFactory->getInstance();
            $text = $contentWithoutRedirect->getText();
        }
 
        $parserOutput = $parser
            ->parse( $text, $title, $parserOptions, true, true, $revId );
 
        // T330667: Record the fact that we used the value of
        // 'useParsoid' to influence this parse.  Note that
        // ::getUseParsoid() has a side-effect on $parserOutput here
        // which didn't occur when we called ::getUseParsoid() earlier
        // because $parserOutput didn't exist at that time.
        $parserOptions->getUseParsoid();
 
        // Add redirect indicator at the top
        if ( $redir ) {
            // Make sure to include the redirect link in pagelinks
            $parserOutput->addLink( $redir );
            if ( $cpoParams->getGenerateHtml() ) {
                $parserOutput->setRedirectHeader(
                    $this->linkRenderer->makeRedirectHeader(
                        $title->getPageLanguage(), $redir, false
                    )
                );
                $parserOutput->addModuleStyles( [ 'mediawiki.action.view.redirectPage' ] );
            } else {
                $parserOutput->setText( null );
            }
        }
 
        // Pass along user-signature flag
        if ( in_array( 'user-signature', $content->getPreSaveTransformFlags() ) ) {
            $parserOutput->setOutputFlag( ParserOutputFlags::USER_SIGNATURE );
        }
    }
}