guess you stopped me from asking this same question. time to read the description again..

Fixed

I have the same "No instance for (Applicative NIM)" issue.

It would be helpful if you could at least have the test suit available for trial runs.

Then people could help you debug.

Typo hopefully fixed. Tell me if it works now. It should do. Thanks for your post!

Doing Ocaml, I get an

Error: Unbound value range
Hint: Did you mean raise?

error when running the actual attempt that I don't get in the examples.

They are included in my "preloaded" section of my kata.

Where do the list handling functions come from?

Are they included in the criterion.h header in the test code?

After some experimentation, I've found the test code can't handle answers without a terminating newline.

echo -n and printf don't have one by default.

Moral of the story: don't post an issue too fast:-) Cheers!

Thanks, continuous composite numbers.

I am afraid that you read the description too fast an that you are mixing gap and steps.

Go test case expects Gap(6, 100, 110) to be nil, but the correct answer should be [101, 107].

Using C, I question why the return type is Pair**. I would expect it to be Pair* so I can allocate the array with Pair *dio = malloc(10 * sizeof(Pair)),
resize as needed, and return dio.

That is easier than needing a new malloc call for every pair.

First is that int in Ocaml is not automatically promoted to a bignum. Bignums are in the Num module.

It's up to you to open the module you need.

Why is this conversion not done in the testing code?

Until now the author of a kata can choose the output he wants.
Moreover I am not the Ocaml translator: post for @mcamuzat if you want more explanations.

The Ocaml starting code is wrong and misleading.

First is that int in Ocaml is not automatically promoted to a bignum. Bignums are in the Num module.

Second is that the user written function is required to output a string. Why is this conversion not done in the testing code?