Hamming distance definition differs in Python and Ruby

[typograph]

In Python the test suite seems either to be counting 'missing' characters, or to compare from the end of the string (strangely, this gives the same result in the provided cases).

    def test_hamming_different_length1(self):
        self.assertEqual(4, hamming('AAGCTAC','ACGTT'))
    def test_hamming_different_length2(self):
        self.assertEqual(5, hamming('AAGCTAC','ACGTTACGTC'))

In Ruby the test suite says to ignore the extra characters

    def test_ignores_extra_length_on_first_strand_when_longer
        assert_equal 1, Hamming.compute('AGAGACTTA', 'AAA')
    end
    def test_ignores_extra_length_on_other_strand_when_longer
        assert_equal 2, Hamming.compute('AGG', 'AAAACTGACCCACCCCAGG')
    end

And Wikipedia says the Hamming distance is not defined for strings of different length.

I am not saying the problems cannot be different, but for such a well-defined concept it would make sense to stick to one definition, especially when the READMEs provide so little information about what is expected from the implementation.

[kytrinyx]

I opened an issue in each of the language tracks that have either Hamming or the older Point Mutations exercise, that is also based on the Hamming distance. I'm going to close this issue and let each language track handle this separately. There's an open pull request in the Python track.

Thanks so much for bringing this up, it's definitely an improvement.

[kytrinyx]

FYI - very interesting discussion about this problem here: #1910