Getting Python to print in Cyrillic in a feature extractor

Question

I'm trying to train a program to learn to determine whether a newly given word in Russian is a noun or a verb.

def POS_features(word):
    return{'three_last_letters':word[-3:]}
print(POS_features(u'Богатир'))

Which returns {'three_last_letters': u'\u0442\u0438\u0440'}

Despite the

u'Богатир'

the last three letters print gobbledy-gook. How can I get Python to print in Cyrillic?

Answer 1

Your function returns a dict and that's what was printed. Containers frequently print their repr - that is, a python-like representation of their contents. If you process the dict yourself, you get the right value.

>>> def POS_features(word):
...     return{'three_last_letters':word[-3:]}
... 
>>> val = POS_features(u'Богатир')
\>>> for k,v in val.items():
...     print k, v
... 
three_last_letters тир

I pasted your printed result back into my shell and got a dict again. Its not guaranteed that a string representation of a object can be built back into an object, but it works for simple types.

>>> val = {'three_last_letters': u'\u0442\u0438\u0440'}
>>> val
{'three_last_letters': u'\u0442\u0438\u0440'}