Unicode support for TestCase.skip is broken because the caught SkipTest exception is passed through a str call.
except SkipTest as e:
    self._addSkip(result, str(e))
Could be fixed with by changing to unicode(e)

> Could be fixed with by changing to unicode(e)

Replacing str(e) with unicode(e) can introduce Unicode errors regressions. I'm not confident in doing such change late in the old 2.7 branch.

I suggest you to only use native strings (byte strings, "str" type) in Python 2, and slowly upgrade your application to Python 3.

I propose to close this issue as WONT FIX.

I concur with Victor. Unlikely this affects many people since skipping messages usually are ASCII strings.

As a workaround you can implement your own skip() method that encodes unicode to 8-bit string.

    def skip(self, reason):
        if isinstance(reason, unicode):
            reason = unicode.encode(reason, 'utf-8')
        TestCase.skip(self, reason)