@Blind4Basics That's alright I suppose; I just wanted to avoid copying the slice but it doesn't seem expensive in practice. The (sample) testing code creates and passes the slice at the function callsite so I assumed it wouldn't matter.

I've done a couple of combinatorics kata lately. They are very enjoyable - the math isn't terribly complex and you have the opportunity to count the whatever object in creative ways!

For people struggling with overflow issues, I recommend the wikipedia page on binomial coefficient.

Thanks for this one!

You should never mutate inputs, unless it's specified in the contract. Doing this is generating side effects (ie changing the state of the universe outside of the function), and you don't want to code like that. => Rust is right, here.

Please consider changing the type of the intervals parameter in the Rust version to &mut [(i32, i32)]. Most of other languages have mutable references types by default so they can modify the list in place whlist Rust requires both the caller and callee agree on reference mutability.

What's the point of a rust implementation returning a btreemap?

My solution is not correct.

No problem!

Looks clean! I'm in the process of learning Rust and I didn't even know about fold. Thank you!

Our solutions are pretty similar. I used fold instead of a for loop and a flag variable to detect double spaces.
Btw, the input is assumed to always be valid so MORSE_CODE[p] can't panic.