Add [T; N]::as_[mut_]slice

[LukasKalbertodt]

Part of me trying to populate arrays with a couple of basic useful methods, like slices already have. The ability to add methods to arrays were added in #75212. Tracking issue: #76118

This adds:

impl<T, const N: usize> [T; N] {
    pub fn as_slice(&self) -> &[T];
    pub fn as_mut_slice(&mut self) -> &mut [T];
}

These methods are like the ones on std::array::FixedSizeArray and in the crate arraytools.

[rust-highfive]

r? @Mark-Simulacrum

(rust_highfive has picked a reviewer for you, use r? to override)

[joshtriplett]

I've wanted these methods several times. Thanks!

[Mark-Simulacrum]

r=me on implementation, since you are a libs team member :)

It does feel a bit awkward to have these, since you get the same behavior just from coercion... but triggering coercion can be painful, and often means falling back on slicing, so this seems reasonable.

[scottmcm]

Given that Vec has these, it makes sense that arrays might as well too.

Hmm, it looks like this obviates FixedSizeArray -- could that just be removed entirely, now?

[LukasKalbertodt]

r=me on implementation, since you are a libs team member :)

Mhhhh... I'm not exactly sure what that means :D are you saying I can merge at my own convenience with r=Mark-Simulacrum?

Hmm, it looks like this obviates FixedSizeArray -- could that just be removed entirely, now?

Probably? Though I would not include its removal in this PR. We might also want to wait a cycle or so before removing it.

[Mark-Simulacrum]

I mean that we normally need libs sign off on unstable API but you can self grant that and the impl looks good, so yes, just r=me when ready.

[LukasKalbertodt]

From my side, this is ready. Let's merge it then, it's only unstable anyway.

@bors r=Mark-Simulacrum

[bors]

📌 Commit d7afe2a has been approved by Mark-Simulacrum

[bors]

⌛ Testing commit d7afe2a with merge 8247a30061a79a7923718edf5a68d8eac473ee00...

[Dylan-DPC-zz]

@bors retry yield