UTF8 with MoinMoin and Python 2.7

[GoogleCodeExporter]

What steps will reproduce the problem?
1. Install MoinMoin
2. Configure a wiki(farm)
3. Go to any page with non-ASCII characters in name (e.g. a localized help 
page, HilfeFürAnfänger)

What is the expected output? What do you see instead?
Expected to see the page, instead gives page not found showing the name with 
wrong character set, giving a suggestion of the wanted page as a similar page, 
cannot go there either.

What version of the product are you using? On what operating system?
ISAPI_WSGI 0.4.2, Python 2.7, MoinMoin 1.9.7, Windows Server 2012, IIS8

Please provide any additional information below.
Seems to be fixable with this change:

    def _get_variable_py2k(self, key):
        # py2k - just use normal string objects.
        value = self.ecb.GetServerVariable(key)
        u_value = None
        try:
            u_value = self.ecb.GetServerVariable('UNICODE_' + key).encode("utf-8")
        except:
            pass            
        return u_value or value

Original issue reported on code.google.com by mickut on 1 Sep 2013 at 7:02