Doc total order requirement of sort(_unstable)_by
I took the definition of what a total order is from the Ord trait
docs. I specifically put "elements of the slice" because if you
have a slice of f64s, but know none are NaN, then sorting by
partial ord is total in this case. I'm not sure if I should give
such an example in the docs or not.
r? @GuillaumeGomez
Add slice::rchunks(), rchunks_mut(), rchunks_exact() and rchunks_exact_mut()
These work exactly like the normal chunks iterators but start creating
chunks from the end of the slice.
----
The new iterators were motivated by a [comment](https://github.com/rust-lang/rust/issues/47115#issuecomment-424141121) by @DutchGhost.
~~~This currently includes the commits from https://github.com/rust-lang/rust/pull/54537 to not have to rename things twice or have merge conflicts. I'll force-push a new version of the branch ones those are in master.~~~
Also the stabilization tracking issue is just some number right now. I'll create the corresponding issue once this is reviewed and otherwise mergeable.
cc @DutchGhost
I took the definition of what a total order is from the Ord trait
docs. I specifically put "elements of the slice" because if you
have a slice of f64s, but know none are NaN, then sorting by
partial ord is total in this case. I'm not sure if I should give
such an example in the docs or not.
Use impl_header_lifetime_elision in libcore
The feature is approved for stabilization, so let's use it to remove about 300 `'a`s.
Tracking issue for the feature: https://github.com/rust-lang/rust/issues/15872
Simplify slice's first(_mut) and last(_mut) with get
This change makes these functions easier to read and interpret. I haven't detected any difference in performance locally.
r? @Mark-Simulacrum
This is a safe wrapper around ptr::copy, for regions within a single
slice. Previously, safe in-place copying was only available as a side
effect of Vec::drain.
stabilize slice_align_to
This is very hard to implement correctly, and leads to [serious bugs](https://github.com/llogiq/bytecount/pull/42) when done incorrectly. Moreover, this is needed to be able to run code that opportunistically exploits alignment on miri. So code using `align_to`/`align_to_mut` gets the benefit of a well-tested implementation *and* of being able to run in miri to test for (some kinds of) UB.
This PR also clarifies the guarantee wrt. the middle part being as long as possible. Should the docs say under which circumstances the middle part could be shorter? Currently, that can only happen when running in miri.
