Tracking Issue for slice::split_at_unchecked() and split_at_mut_unchecked()

[sdroege]

This is a tracking issue for the slice::split_at_unchecked() and slice::split_at_mut_unchecked() functions.
The feature gate for the issue is #![feature(slice_split_at_unchecked)].

API

impl<T> [T] {
    pub const unsafe fn split_at_unchecked(&self, mid: usize) -> (&[T], &[T]);
    pub const unsafe fn split_at_mut_unchecked(&mut self, mid: usize) -> (&mut [T], &mut [T]);
}

Steps

• Implement (Get rid of bounds check in slice::chunks_exact() and related function… #75936)
• FCP (Tracking Issue for slice::split_at_unchecked() and split_at_mut_unchecked() #76014 (comment))
• Stabilization PR

Unresolved Questions

• Is this worth having if it's simple enough to implement manually around slice::get_unchecked() or slice::from_raw_parts()?

[sdroege]

This is now merged but not as public API.

I think this would be useful to have though, at least the mutable version as it's dealing with raw pointers and it's much harder to use split_at_mut_unchecked() wrong than to do a mistake when reimplementing it. I also checked some of my code and there are various places where this would be useful to use (and I have a couple of ad-hoc implementations of this).

There are probably also some places in std.

[WaffleLapkin]

Why the API is private? If there is no reason, I could make a PR making it public.

I've just found myself coping the methods from std, to use them...

[thomcc]

It's very strange that this API is both private and unstable with a tracking issue. Was this intentional?