Fix `ref` regex

[AntonOfTheWoods]

Currently the regex used for cleaning elements fails to take into account a somewhat common case where there is a / in an attribute:

This appears to be relatively common with <ref> , which has examples like:

<ref name="http://a.great.site/"/>

It appears that because there is a non-whitespace char between / and >, it fails. The current (warning, this will get refactored so don't just blindly search for the string!) regex is:

re.compile(r"<\s*%s\b[^>]*/\s*>" % tag, re.DOTALL | re.IGNORECASE) for tag in selfClosingTags

This needs tweaking