bertysentry commented on a change in pull request #49: URL: https://github.com/apache/maven-doxia/pull/49#discussion_r550357704
########## File path: doxia-modules/doxia-module-markdown/src/main/java/org/apache/maven/doxia/module/markdown/MarkdownParser.java ########## @@ -130,133 +184,98 @@ public void parse( Reader source, Sink sink ) * @return HTML content generated by flexmark-java * @throws IOException passed through */ - String toHtml( Reader source ) + CharSequence toHtml( Reader source ) throws IOException { + // Read the source String text = IOUtil.toString( source ); - MutableDataHolder flexmarkOptions = PegdownOptionsAdapter.flexmarkOptions( - Extensions.ALL & ~( Extensions.HARDWRAPS | Extensions.ANCHORLINKS ) ).toMutable(); - ArrayList<Extension> extensions = new ArrayList<>(); - for ( Extension extension : flexmarkOptions.get( com.vladsch.flexmark.parser.Parser.EXTENSIONS ) ) - { - extensions.add( extension ); - } - - extensions.add( FlexmarkDoxiaExtension.create() ); - flexmarkOptions.set( com.vladsch.flexmark.parser.Parser.EXTENSIONS, extensions ); - flexmarkOptions.set( HtmlRenderer.HTML_BLOCK_OPEN_TAG_EOL, false ); - flexmarkOptions.set( HtmlRenderer.HTML_BLOCK_CLOSE_TAG_EOL, false ); - flexmarkOptions.set( HtmlRenderer.MAX_TRAILING_BLANK_LINES, -1 ); - - com.vladsch.flexmark.parser.Parser parser = com.vladsch.flexmark.parser.Parser.builder( flexmarkOptions ) - .build(); - HtmlRenderer renderer = HtmlRenderer.builder( flexmarkOptions ) - .linkResolverFactory( new FlexmarkDoxiaLinkResolver.Factory() ) - .build(); - + // Now, build the HTML document StringBuilder html = new StringBuilder( 1000 ); html.append( "<html>" ); html.append( "<head>" ); - Pattern metadataPattern = Pattern.compile( MULTI_MARKDOWN_METADATA_SECTION, Pattern.MULTILINE ); - Matcher metadataMatcher = metadataPattern.matcher( text ); + + // First, we interpret the "metadata" section of the document and add the corresponding HTML headers + Matcher metadataMatcher = METADATA_SECTION_PATTERN.matcher( text ); boolean haveTitle = false; if ( metadataMatcher.find() ) { - metadataPattern = Pattern.compile( MULTI_MARKDOWN_METADATA_ENTRY, Pattern.MULTILINE ); - Matcher lineMatcher = metadataPattern.matcher( metadataMatcher.group( 1 ) ); - boolean first = true; - while ( lineMatcher.find() ) + Matcher entryMatcher = METADATA_ENTRY_PATTERN.matcher( metadataMatcher.group( 0 ) ); + while ( entryMatcher.find() ) { - String key = StringUtils.trimToEmpty( lineMatcher.group( 1 ) ); - if ( first ) - { - boolean found = false; - for ( String k : STANDARD_METADATA_KEYS ) - { - if ( k.equalsIgnoreCase( key ) ) - { - found = true; - break; - } - } - if ( !found ) - { - break; - } - first = false; - } - String value = StringUtils.trimToEmpty( lineMatcher.group( 2 ) ); + String key = entryMatcher.group( 1 ); + String value = entryMatcher.group( 2 ); if ( "title".equalsIgnoreCase( key ) ) { haveTitle = true; html.append( "<title>" ); - html.append( StringEscapeUtils.escapeXml( value ) ); + html.append( HtmlTools.escapeHTML( value, false ) ); html.append( "</title>" ); } - else if ( "author".equalsIgnoreCase( key ) ) - { - html.append( "<meta name=\'author\' content=\'" ); - html.append( StringEscapeUtils.escapeXml( value ) ); - html.append( "\' />" ); - } - else if ( "date".equalsIgnoreCase( key ) ) - { - html.append( "<meta name=\'date\' content=\'" ); - html.append( StringEscapeUtils.escapeXml( value ) ); - html.append( "\' />" ); - } else { - html.append( "<meta name=\'" ); - html.append( StringEscapeUtils.escapeXml( key ) ); - html.append( "\' content=\'" ); - html.append( StringEscapeUtils.escapeXml( value ) ); - html.append( "\' />" ); + html.append( "<meta name='" ); + html.append( HtmlTools.escapeHTML( key ) ); + html.append( "' content='" ); + html.append( HtmlTools.escapeHTML( value ) ); + html.append( "' />" ); } } - if ( !first ) - { - text = text.substring( metadataMatcher.end() ); - } + + // Trim the metadata from the source + text = text.substring( metadataMatcher.end( 0 ) ); + } - Node rootNode = parser.parse( text ); - String markdownHtml = renderer.render( rootNode ); + // Now is the time to parse the Markdown document + // (after we've trimmed out the metadatas, and before we check for its headings) + Node documentRoot = FLEXMARK_PARSER.parse( text ); - if ( !haveTitle && rootNode.hasChildren() ) + // Special trick: if there is no title specified as a metadata in the header, we will use the first + // heading as the document title + if ( !haveTitle && documentRoot.hasChildren() ) { - // use the first (non-comment) node only if it is a heading - Node firstNode = rootNode.getFirstChild(); - while ( firstNode != null && !( firstNode instanceof Heading ) ) + // Skip the comment nodes + Node firstNode = documentRoot.getFirstChild(); + while ( firstNode != null && firstNode instanceof HtmlCommentBlock ) { - if ( !( firstNode instanceof HtmlCommentBlock ) ) - { - break; - } firstNode = firstNode.getNext(); } - if ( firstNode instanceof Heading ) + // If this first non-comment node is a heading, we use it as the document title + if ( firstNode != null && firstNode instanceof Heading ) { html.append( "<title>" ); TextCollectingVisitor collectingVisitor = new TextCollectingVisitor(); String headingText = collectingVisitor.collectAndGetText( firstNode ); - html.append( StringEscapeUtils.escapeXml( headingText ) ); + html.append( HtmlTools.escapeHTML( headingText, false ) ); Review comment: Same answer as above. ---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org