word_count: test mixed case about python HOT 4 CLOSED

The reason for this is that a change was applied to the test to normalize case.

0314cbf
#207

It was justified because other language tracks expect normalization - specifically ruby and clojure in the pull request. I also noticed Haskell does this normalization.

from python.

The problem is not that normalization is tested. The problem is also not that the test is not opinionated about how to normalize case.
The problem is that the test is so minimalistic that a user wrote code that passed the test suite while the solution only works if the function gets the words in the right order.

My simple solution to the mixed case test is to expand the test string "go Go GO" to
"go Go GO Stop stop" and assume as result a dict with the values 3, 2. That would still not force the user to only normalize to lower case.

def test_mixed_case(self):
    self.assertEqual([2, 3], sorted(list(word_count('go Go GO Stop stop').values())))

from python.

I actually don't mind which direction the user normalizes to, and I think that it would be nice to leave that up to the person solving the problem. I like this suggested approach and wouldn't mind changing the other tracks to use it.

from python.

Okay, did it for the Python track.

from python.