Additional NonZero conversions

[clarfonthey]

View all comments

ACP: rust-lang/libs-team#145

Requires FCP due to insta-stable APIs added:

// can't be done generically; compiler can't infer that ZeroablePrimitive is sealed for coherence reasons
impl From<&NonZero<uN>> for &uN;
impl From<&NonZero<iN>> for &iN;
impl From<&NonZero<usize>> for &usize;
impl From<&NonZero<isize>> for &isize;
impl From<&NonZero<char>> for &char;

// excluded due to typechecking bug, but left commented out in code
// impl<T> From<&[NonZero<T>]> for &[T] where T: ZeroablePrimitive;

// rest are generic
impl<T, const N: usize> From<[NonZero<T>; N]> for [T; N] where T: ZeroablePrimitive;
impl<T> TryFrom<&T> for &NonZero<T> where T: ZeroablePrimitive;
impl<T> TryFrom<&[T]> for &[NonZero<T>] where T: ZeroablePrimitive;
impl<T, const N: usize> TryFrom<[T; N]> for [NonZero<T>; N] where T: ZeroablePrimitive;

Note that the Error for the TryFrom implementations is TryFromIntError to match the similar impls.

r? rust-lang/libs-api

[bors]

☔ The latest upstream changes (presumably #129750) made this pull request unmergeable. Please resolve the merge conflicts.

[Dylan-DPC]

@clarfonthey since this is waiting on a review, do you still want this to be as a draft? or is this ready for review?

[clarfonthey]

It is ready for review; see the OP.

I should just unmark it as draft and do what I said I was going to do, which is manually implement everything for all the NonZero types. I don't think it's going to be possible to fix the errors unless we actually come up with a way to track sealed traits for coherence, which I don't think is going to happen for a long time.

I'll unmark as draft once I make those changes.

[bors]

☔ The latest upstream changes (presumably #138177) made this pull request unmergeable. Please resolve the merge conflicts.

[Dylan-DPC]

@clarfonthey any updates on this? thanks

[clarfonthey]

@rustbot ready

One of the implementations was not possible due to inference issues, and one of them has to be done via a macro due to coherence issues, but the rest are working as intended.

[ChrisDenton]

Did you want to leave in the whole commented out block?

View changes since the review

[ChrisDenton]

Requires FCP due to insta-stable APIs added

@clarfonthey I've nominated this for that reason.

[clarfonthey]

You've found out how to force me to rebase my old PR…!

[clarfonthey]

@rustbot reroll (since BurntSushi isn't on the review rotation)

[rustbot]

This PR was rebased onto a different main commit. Here's a range-diff highlighting what actually changed.

Rebasing is a normal part of keeping PRs up to date, so no action is needed—this note is just to help reviewers.

[nia-e]

@rfcbot fcp merge libs

[rust-rfcbot]

@nia-e has proposed to merge this. The next step is review by the rest of the tagged team members:

• @Amanieu
• @BurntSushi
• @cuviper
• @dtolnay
• @joshtriplett
• @the8472
• @thomcc

No concerns currently listed.

Once a majority of reviewers approve (and at most 2 approvals are outstanding), this will enter its final comment period. If you spot a major issue that hasn't been raised at any point in this process, please speak up!

See this document for info about what commands tagged team members can give me.

[clarfonthey]

Also had temporarily committed some unrelated things with this, those are now fixed.

Also verified that the issue with the commented-out impl is real: we want <&[u8]>::from(&[]) to always work, but with an extra impl on From<&[NonZero<u8>]> it becomes ambiguous whether the empty slice is [NonZero<u8>] or [u8].