The current code in templating.py:StringHTMLProperty._hyper_repl_url
checks for a > *at the end* of the URL. If the url is e.g.
"<www.roundup.net>.":
 1) the string is first converted to "&lt;www.roundup.net&gt;.";
 2) the regex matches the whole string, including the trailing '.';
 3) _hyper_repl_url checks for a &gt; at the end of the string and
doesn't find it (because there's the dot);
 4) the whole string is converted to a link.

The attached patch checks for a &gt; in any point of the URL and convert
to a link only the part before it. It also refactors a bit the function
in order to apply the parentheses checks even when the &gt; is found and
improve it in case the ending ')' is missing.

The example is wrong, the problem is only in the 'path' part of the url,
i.e. "<http://roundup.net/>.".
The patch is still valid.

Looks good to me, thanks!

Thanks!