Description:

If the input array is empty consider it as: [0] (array with a zero).

But there are no empty arrays in the tests. That's too bad. Because in this case solutions with reduce would not work.

The reason of your problem is that you mix up, and seem to miss the difference, between numbers and digit characters. When you do odd = [1,3,5,7,9], then the list odd contains 5 numbers: 1, 3, 5, 7, and 9. When you do number = list(str(sum(arr))), the number list holds characters, for example: '1', '3', '5', '7', or '9'. That's why when you do number[-1] in odd you get no results: because for example '5' is not in [1, 3, 5, 7, 9].

5 is, but '5' is not. You need to find out what's the difference between '1' and 1, or between '7' and 7.

Thank you for the solution

Because it is not correct.

Without seeing your solution we do not know why it is not correct.

Why my solution doesn't work?

I disagree, checking a number's parity is 8 kyu, checking the parity of the sum of all elements in an array is 7.

too easy for 7 kyu

Thanks!

Approved

its 8kyu, not 7

I think it's okay to keep different types as inputs but the description is really off

Given a list of integers

in context of Haskell seems like a literal requirement for the arg to be [Integer] type

There should be Haskell-specific description clarifying that input is a list of Integral derivative

wouldn't it be fine if simply defining the type in the default code? it's quite odd for the solver to specify the type when they don't .. know what it should be

Yeah, that's nowhere remotely close to what the kata is supposed to test, which is a problem in itself.

It's solvable using typeclasses

But yeah it does put more of a challenge compared to other languages