Merge lp://qastaging/~coalwater/lernid/fix-925702 into lp://qastaging/lernid

When i think about it i'm not sure if these changes broke the other checks or not

                if self._nick in text and sender != self._nick:
                    self._buffer.insert_with_tags_by_name(iend, text, 'highlight')
                elif self._on_faculty(sender):
                    self._buffer.insert_with_tags_by_name(iend, text, 'highlight')
                elif sender.lower() in self._classbotnames:
                    self._buffer.insert_with_tags_by_name(iend, text, 'italicize')

i think i should test them first

Yes, please. Testing those would be a good idea.

Can you tell me about the unicode conversion problem you found? Is it worth a bug report?

This is the old line of the url matching
for url in re.finditer(r"https?://\S+", unicode(text, 'utf-8')):

in the NativeClassroom.py line 181
http://bazaar.launchpad.net/~lernid-devs/lernid/trunk/view/head:/lernid/widgets/NativeChatroom.py#L181
self._append_to_buffer(_('IRC commands are not yet supported.'))
for some reason lernid sees this text as unicode and when trying to apply the unicode part [unicode(text,'utf-8')] it throws an error, you can easily simulate this error in python console by doing this
  >>> unicode(u'aaa','utf-8')
  Traceback (most recent call last):
    File "<stdin>", line 1, in <module>
  TypeError: decoding Unicode is not supported

when i added a test in the file before the unicode, [print type(text)] it always returned str but in this single case ( the irc command exception message ) it returned type unicode so i simple added the type check to avoid this.

Looks good to me. Thanks again.