'Its alr approved some time ago'

done: https://www.codewars.com/kumite/5d4a8296dbaa2400266a39c4?sel=5da142b9a82c1e0012169e0a

I fixed the merge issue within my fork, but the tests are timing out. Looks like you need to fork mine and try to optimize this some more.

Welp, thanks for the suggestion, though.

Bugger. I suspect, comparing the elm wiki page with the ruby one that the packages are currently limited to elm/core and elm-explorations/test. If so, that's pretty limiting. Without it, I suppose one could manually copypaste the whole Regex module in to the preloaded section, but that seems clumsy. Summary: I don't know.

I tried your solution, but the compiler threw an error saying they couldn't import the Regex module. How do we install it? Can we fiddle around with the elm.json?

I haven't really looked at the Regex library that much, but this solution seems a lot neater than mine haha. I do feel like the markers should continue to be a List Char, though.

This could also be solved arguably more neatly via a Regex e.g.

import Regex

stripComments : String -> List Char -> String
stripComments input markers =
    let
        markersString =
            String.concat <|
                List.map (\marker -> "\\" ++ String.fromChar marker) markers
        regex =
            Maybe.withDefault Regex.never <|
                Regex.fromString ("[ \t]*[" ++ markersString ++ "].*")
    in
    Regex.replace regex (\_ -> "") input

One other thought is that the markers input could be a List String rather than a List Char. I don't know what Elm best practice is here. I suppose there is some benefit to forcing the list to be single characters rather than allow a multi-character through.

Sorry, no random tests.

Now they just expect a constant...

To get this working I needed a shuffle which isn't natively provided by the Random library. I've pulled in the relevant functions from elm-community/random-extra and owanturist/elm-union-find packages. Bit annoying though.

That's a fair point. I'm still learning the Elm test framework so I'll have a think about how to go about this.

The fuzzed tests aren't really random when the answer always is the first number.

Elm translation for review.