Use JSoup to convert HTML to XHTML after parsing with Flexmark

doxia-modules/doxia-module-markdown/pom.xml

@@ -141,7 +141,12 @@ under the License.
       <artifactId>flexmark-ext-yaml-front-matter</artifactId>
       <version>${flexmarkVersion}</version>
     </dependency>
-
+    <dependency>
+      <!-- for converting html to xhtml -->
+      <groupId>org.jsoup</groupId>
+      <artifactId>jsoup</artifactId>
+      <version>1.21.2</version>
+    </dependency>
     <dependency>
       <groupId>org.slf4j</groupId>
       <artifactId>slf4j-api</artifactId>

doxia-modules/doxia-module-markdown/src/main/java/org/apache/maven/doxia/module/markdown/MarkdownParser.java

@@ -59,6 +59,8 @@
 import org.apache.maven.doxia.sink.Sink;
 import org.apache.maven.doxia.util.HtmlTools;
 import org.codehaus.plexus.util.xml.pull.XmlPullParser;
+import org.jsoup.Jsoup;
+import org.jsoup.nodes.Document;
 
 /**
  * <p>
@@ -172,14 +174,14 @@ public class MarkdownParser extends AbstractTextParser implements TextMarkup {
     public void parse(Reader source, Sink sink, String reference) throws ParseException {
         try {
             // Markdown to HTML (using flexmark-java library)
-            String html = toHtml(source);
+            String xhtml = toXhtml(source);
 
             // TODO: add locator for the markdown source (not the intermediate HTML format)
             // this requires writing a custom renderer not leveraging the XHTML parser
 
             // then HTML to Sink API
             parser.setEmitComments(isEmitComments());
-            parser.parse(html, getWrappedSink(sink), "Intermediate HTML from " + reference);
+            parser.parse(xhtml, getWrappedSink(sink), "Intermediate HTML from " + reference);
         } catch (IOException e) {
             throw new ParseException("Failed reading Markdown source document", e);
         }
@@ -272,7 +274,7 @@ private boolean writeHtmlMetadata(StringBuilder html, String key, List<String> v
      * @return HTML content generated by flexmark-java
      * @throws IOException passed through
      */
-    String toHtml(Reader source) throws IOException {
+    String toXhtml(Reader source) throws IOException {
         // Read the source
         StringBuilder markdownText = new StringBuilder(IOUtils.toString(source));
 
@@ -314,7 +316,13 @@ String toHtml(Reader source) throws IOException {
         html.append("</body>");
         html.append("</html>");
 
-        return html.toString();
+        return toXhtml(html.toString());
+    }
+
+    private String toXhtml(String html) {
+        final Document document = Jsoup.parse(html);
+        document.outputSettings().syntax(Document.OutputSettings.Syntax.xml).prettyPrint(false);
+        return document.html();
     }
 
     /**

doxia-modules/doxia-module-markdown/src/test/java/org/apache/maven/doxia/module/markdown/MarkdownParserTest.java

@@ -628,6 +628,7 @@ public void testHtmlContent() throws Exception {
                 "table",
                 "tableRows",
                 "text",
+                "unknown", // tbody start
                 "tableRow",
                 "tableHeaderCell",
                 "text",
@@ -640,6 +641,7 @@ public void testHtmlContent() throws Exception {
                 "tableCell_",
                 "tableRow_",
                 "text",
+                "unknown", // tbody end
                 "tableRows_",
                 "table_",
                 "text",
@@ -669,7 +671,7 @@ protected SinkEventTestingSink parseFileToEventTestingSink(String file) throws P
 
     protected String parseFileToHtml(String file) throws ParseException, IOException {
         try (Reader reader = getTestReader(file)) {
-            return parser.toHtml(reader).toString();
+            return parser.toXhtml(reader).toString();
         }
     }
 
@@ -860,6 +862,11 @@ public void testCommentsRemovedWithEmitCommentsFalse() throws ParseException, IO
         assertSinkDoesNotContain(eventList.iterator(), "comment", "comment_");
     }
 
+    @Test
+    public void testHtmlInMarkdown() throws ParseException, IOException {
+        parseFileToEventTestingSink("html");
+    }
+
     protected static void assertComment(ListIterator<SinkEventElement> it, String comment) {
         assertSinkEquals(it.next(), "comment", comment);
         // every comment ends with a line break in the emitted html which leads to an additional text event containing a

doxia-modules/doxia-module-markdown/src/test/resources/html.md

@@ -0,0 +1,4 @@
+
+# HTML in Markdown
+
+This is a <p> test.