Right, the end-handling of blocking the intersection is different by necessity (just imagine beyond the end is a cliff that cars fall off so the exit from the intersection is never blocked!). Anyway, the rules for those scenarios are covered by the spec:

• It is illegal to queue across an intersection. In other words, DO NOT enter an intersection unless you can see there is safe passage through it and out the other side!
• If there is a traffic light at the end of the road then its exit is never blocked

Which rules seem counter-intuitive?

Actually, I prefer for people to think they have a good solution after passing the sample tests only be to shocked that their solution is not complete at all. IMO it gives a good lesson to take more care to read the kata requirements carefully and not be too hasty in thinking you are finished without considering edge cases.

What are the missing tests?

Approved. Thanks!

Approved. Thanks!

Users are free to implement solutions any which way they like. Personally, I like to see lots of solution variations because occasionally someone will do something quite unexpected/terrific and you can learn something new.

It's just a formula that approximates reality. Don't try to read too much into it.

Approved

Half the fun of this kata was seeing the "prettyPrint" HTML result of the flap board in the test results. e.g. see the original Kata.

Is there no way to do similar in Haskell?

Fixed. Thanks for reporting.

Are you just asking about extra high cliffs? I thought the Kata desciption covers this case where it says:

Cliff characters | are repeated downwards to the lowest adacent ground level.

Cliffs may be higher than 2 units; in that case the same replacement rules as up/down 2 applies.

That first example was drawn wrong. Fixed. Thanks for reporting it.

Hm. Pretty similar to my histogram Kata series.

https://www.codewars.com/kata/histogram-v2/

Approved.