Please specify the language when raising issues. I'm assuming you're referring to the C translation.

Sorry, I didn't know that the discussion is lanugage agnostic.

No, the size of the allocated buffer is irrelevant as long as it's big enough. What's important is to set the value of nb_vowels appropriately (refer to the solution setup).

I see. I apologize.

Are you on macOS? I suspect that this is a distribution problem. I think macOS has a different echo then GNU/Linux.

echo is a built-in command, so it's the fault of the built-in bash being too old. This should work with newer bash.

Thank you for this solution! Btw, do you mind sharing what code formatter you use? This vertical alignment looks really neat.

Approved, thank you all for the effort!

I see, sorry for the misunderstanding :p

Though isn't that too big too? Or is that to ensure that there's enough cases to distinguish a good solution from a bad one? You only used up to 300 as list length in python.

Hello! Thank you for the translation!
Just one tiny thing, could you change the range (0,100_000) to (-100_000, 100_000)? This would be more coherent to the Python translation, which is the first imperitive translation.

Awesome, thank you! I have approved your fork.

Thanks for your translation :)

Could you complete the basic tests so that it's coherient with the description and other languages?
Also, do you think performance tests are needed? I decided to remove them in the end because there are some fun solutions that would time out.

I updated the draft with my new test cases :)

I have the same problem.
With the condition that "" is a valid tree, I decide to make it a minimal tree (so that this definition would work recursively.)
This has a problem, because that means the tree

a = Leaf :*: (Leaf :*: Leaf)
b = (Leaf :*: Leaf) :*: Leaf

are both represented by:

a' = (()) -- a tree with one subtree
b' = (())
a' == b' = True

If I choose "" to represent Leaf, information are lost: I can no longer tell whether a subtree is on the left or on the right if there's only one subtree.

What could be a better way to model trees?
Thank you.

I changed the random test so that it has a chance to use larger lists.

gen :: Gen [Integer]
gen = do
  len <- choose (0, 100000)
  vectorOf len (choose (-100000, 100000))

What do you think?
I do not know a "bad" algorithm which will time out, but my solution doesn't time out and I think these are relatively big lists (in length and in numbers contained).

Otherwise I think it's all good!

Update: I removed the perf tests, I wanted to keep this kata basic and fun :p

I can't seem to see the images in the description...

True! I'll think about how to do that practically. And thanks :)