impl Fn/FnMut/FnOnce for Arc/Rc

[Centril]

This PR introduces the ability to have Rc/Arc<F>: Fn(A) -> B where F: Fn(A) -> B and Rc/Arc<Fn(A) -> B.

This was previously tried (and failed) in #34118 (comment) due to breakage with crater. But now that the new edition is approaching we might want to try this in edition 2018. We should probably do a new crater run to see how much the breakage has changed since the previous attempt.

cc @aturon

[rust-highfive]

r? @sfackler

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

[Centril]

r? @aturon

[kennytm]

Could you clarify how Rust 2018 is going to make this work? Both Rust 2015 and 2018 editions are going to share the same standard library.

[Centril]

@kennytm That's true. Honestly not sure at all here; but @aturon seemed to agree that we should try something, so I defer to them ;) And if we can't, we'll have to skip it.

[shepmaster]

Ping from triage, @aturon — we eagerly await your review!

[bors]

☔ The latest upstream changes (presumably #48333) made this pull request unmergeable. Please resolve the merge conflicts.

[aturon]

@rust-lang/infra Requesting crater run please!

[kennytm]

@bors try

[bors]

⌛ Trying commit 51b2ca5 with merge f131001...

[bors]

☀️ Test successful - status-travis
State: approved= try=True

[Mark-Simulacrum]

Crater run started. Expected ETA is in 5-6 days.

[bors]

☔ The latest upstream changes (presumably #50017) made this pull request unmergeable. Please resolve the merge conflicts.

[Mark-Simulacrum]

This probably cannot land as-is: the impls introduced appear to conflict with tokio-tls which is quite widely used: Crater results are at: http://cargobomb-reports.s3.amazonaws.com/pr-49224/index.html. 'Blacklisted' crates (spurious failures etc) can be found here. If you see any spurious failures not on the list, please make a PR against that file.

(interested observers: Crater is a tool for testing the impact of changes on the crates.io ecosystem. You can find out more at the repo if you're curious)

[emilyalbini]

Ping from triage @aturon! What should we do with this PR?

[alexcrichton]

Unfortunately that's a pretty high number of regressions, so I'm going to close this.

[Centril]

@alexcrichton Bummer on the regressions :(

I'll have to look through the regressions to see how specialization affects those, but do you think perhaps specialization could be our saving grace here?

[alexcrichton]

Unsure, I just know we can't land this with so many known regressions.

[Centril]

@alexcrichton Agreed on that :)