I'm creating documentation in XHTML5 on how to use... XHTML5. I want to show what the XML header should look like so I go to the source code and enter the following:
<figure><pre><code></code></pre></figure>
(I have to go into the source code to do this, because BlueGriffon apparently has no way to allow me to add a <pre> inside a <figure> using WYSIWYG interface. But that is another matter.)
I go back into the WYSIWYG interface and enter:
<?xml version="1.0" encoding="utf-8"?>
Unfortunately BlueGriffon does not property encode the < and > symbols, so the source code remains:
<figure>
<pre><code><?xml version="1.0" encoding="utf-8"?></code></pre>
</figure>
What it should show in the source is something like this:
<figure>
<pre><code><?xml version="1.0" encoding="utf-8"?></code></pre>
</figure>
Perhaps BlueGriffon thinks this is a processing instruction. But it doesn't matter; if I type the characters in the WYSIWYG interface, they should be properly encoded.
This document breaks compliance; the XML processor will issue a fatal error that an illegal processing instruction target has been encountered.
I consider this bug a blocker, as it produces XML documents that are not well-formed.