in parseMarkdown, don't write empty class for code blocks, and don't encode their content

source/Ox/js/HTML.js

@@ -380,8 +380,8 @@
             .replace(
                 /\n```(.*)\n([^`]+)\n```/g,
                 function(match, a, b) {
-                    return '<pre><code class="'+ a + '">'
+                    return '<pre><code' + (a ? ' class="' + a + '"' : '') + '>'
-                        + Ox.encodeHTMLEntities(b) + '\n</code></pre>';
+                        + b + '\n</code></pre>';
                 }
             )
             .replace(