With a little bit more work, you could have something you could pass to codecs.register()¹, so that you could encode and decode in the standard way:
text = u'ႹЄLᒪo, ǃ'
enc = text.encode('utf-9')
dec = enc.decode('utf-9')
assert text == dec
¹:
https://docs.python.org/2.7/library/codecs.html#codecs.regis...