[Pre-pre-RFC] "splatting" for named arguments and function overloading

[CAD97]

@Nadrieril

You can only splat a generic that implements Tuple, which is only implemented for built-in tuples, so only the crate that defines the trait can define impls for it, unless I'm forgetting some piece of orphan rules magic.

I… somehow overlooked that restriction when writing that section of my post. So yes, the orphan rules for impl ForeiegnTrait for (…) are at least as restrictive as impl ForeiegnTrait<(…)> for ForeiegnType, which is what I suggested as an alternative.

[Nadrieril]

Oh, I didn't mean to say "I don't want Rust-to-Rust overloading", just "it's not a goal of mine to get that".

Where I'm coming from is: "tiny syntactic sugar gets us pretty nice FFI powers and also extends to named arguments which we wanted anyways". I love it when we get a lot of benefit for not much design complexity, something about this "fits well" in the language to my taste.

I don't mind that this gives a Rust-to-Rust overloading feature. I'm sure some ppl will find APIs where this is really beneficial to have and the rest won't bother with the weird trait dances required.