Inline elements getting parsed as text when doubly nested

[greenape]

import markdown
md = markdown.Markdown(extensions=["markdown.extensions.md_in_html"])
test = """
<div markdown="1">
<div>
<img src="" />
<code>Test</code>
<p>Test 2</p>
</div>
</div>
"""
md.convert(test)

Which gives:
'<div>\n<div>\n&lt;img src="" /&gt;\n&lt;code&gt;Test&lt;/code&gt;\n<p>Test 2</p>\n</div>\n</div>'

When I'd expect to get '<div>\n<div>\n<img src="" />\n<code>Test</code>\n<p>Test 2</p>\n</div>\n</div>'.

Happens with 3.3, 3.3.1.

[facelessuser]

Hmm, it seems the new HTML parser is escaping inline tags for some reason.

[facelessuser]

Seems like md_in_html subclasses from the new HTML parser, but overrides start tag handling. What it forgets to do is flag whether the content is "raw HTML" as that logic is in the overridden base element's start tag logic.

Doing a quick and dirty patch to add this flagging of "raw HTML" logic to md_in_html, we get the expected results:

<div markdown="1">
<div>
<img src=""/>
<code>Test</code>
<p>Test 2</p>
</div>
</div>

Now, I haven't done much testing on this, but I think I should be able to get this fix in with a couple of other HTML related parse issues.

[facelessuser]

The cause isn't quite right, but there is some inline handling logic missing. I'll spend some time narrowing it down.

[facelessuser]

This case has a solution in #1044.