Stabilize some Result methods as const

[CDirkx]

Stabilize the following methods of Result as const:

• is_ok
• is_err
• as_ref

A test is also included, analogous to the test for const_option.

These methods are currently const under the unstable feature const_result (tracking issue: #67520).
I believe these methods to be eligible for stabilization because of the stabilization of #49146 (Allow if and match in constants) and the trivial implementations, see also: PR#75463 and PR#76135.

Note: these methods are the only methods currently under the const_result feature, thus this PR results in the removal of the feature.

Related: #76225

[rust-highfive]

r? @cramertj

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

[CDirkx]

r? @ecstatic-morse

[ecstatic-morse]

r? @KodrAus (Needs FCP)

[CDirkx]

Created an issue for similar stabilizations: #76225

[KodrAus]

For reference, there's also an equivalent push to constify some Option methods at #76135

@rfcbot fcp merge

[KodrAus]

@rfcbot fcp merge

[rfcbot]

Team member @KodrAus has proposed to merge this. The next step is review by the rest of the tagged team members:

• @Amanieu
• @KodrAus
• @dtolnay
• @sfackler
• @withoutboats

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.

[rfcbot]

🔔 This is now entering its final comment period, as per the review above. 🔔

[josephlr]

There are some other methods we should also make const (either in this PR or a followup). They all have trivial match implementations and doesn't use any trait bounds:

Result::ok()
Result::err()
Result::iter()
Result::and()
Result::or()
Result::unwrap_or()
Result::transpose()

Result::flatten() also has a trivial impl, but that's unstable anyway (see #70142 (comment))

EDIT: Nevermind, these all require const fn Drop
EDIT2: Actually, we can implement transpose without dropping anything
EDIT3: Nevermind, transpose requires const_precise_live_drops, which isn't stable.

[bors]

⌛ Testing commit 787b270 with merge a51855f645eb6b86d1c6eb5656ae3394c9949a05...

[bors]

💔 Test failed - checks-actions

[ecstatic-morse]

@CDirkx Test failures are legit. Clippy has a lint that suggests is_err() instead of if let Err(_), but it's disabled in const fn. With this PR, that exception is no longer necessary and the test output can be updated.

[CDirkx]

I updated the testcases, when this get's merged I'll update the testcases using Option as well and sync these changes to the Clippy repo.

[RalfJung]

How can it be enough to update the testcases, doesn't the clippy code emitting these lints also need to be updated?

[ebroto]

Clippy is checking if the suggested methods are const fn in const contexts, so it works without changing the logic because they are after this change.

But once the Option methods are stabilized as const fn the function can_suggest and calls to it should be removed as there would be no special cases left.

[CDirkx]

That's what I thought as well (#76135 (comment)), but the implementation of can_suggest was not specific to is_err etc., so at first I thought it could be used in the future if Clippy were to ever suggest other unstable const methods. However then it should probably be a more general pass and not a specific function that lives in matches.rs, so I'm indeed removing it.

I removed the can_suggest check for is_ok and is_err, but for now can_suggest is still used to not suggest the unstable const is_some and is_none from #76135, I will remove it entirely in that PR.

[RalfJung]

@bors r=dtolnay

[bors]

📌 Commit bf70e21 has been approved by dtolnay

[bors]

⌛ Testing commit bf70e21 with merge b873fa6...

[bors]

☀️ Test successful - checks-actions, checks-azure
Approved by: dtolnay
Pushing b873fa6 to master...