After partially consuming a Zip adapter and then wrapping it into
another Zip where the adapters use their TrustedRandomAccess specializations
leads to the outer adapter returning elements which should have already been
consumed.

As per T-lang consensus, this uses a branch to handle the dangling case.
The discussed optimization of only doing the branch in the T: ?Sized
case is left for a followup patch, as doing so is not trivial
(as it requires specialization for correctness, not just optimization).

Fix whitespace

Co-authored-by: Esteban Kuber <[email protected]>

There was one directory that had only a single test and there was also a
test in the top-level directory. This moves both of them to
`feature-gates/`.

Co-authored-by: Ralf Jung <[email protected]>

…s.rs

Co-authored-by: Esteban Kuber <[email protected]>

The use of Result allows for making use of a reconstructed original value on failed
projections.

AFAICT the test case never landed alongside the fix for the issue.

The first parse is to collect whether the code contains macros, has
`main`, and uses other crates. In that pass we ignore errors as those
will be reported when the test file is actually built.

For that we need to reset errors in the `Diagnostic` otherwise when
dropping it unhandled errors will be reported as compiler bugs.

Fixes rust-lang#80992

This has been fixed between 1.47.0-nightly (663d2f5 2020-08-22) and 1.47.0-nightly (5180f3d 2020-08-23).

Introduce {Ref, RefMut}::try_map for optional projections in RefCell

This fills a usability gap of `RefCell` I've personally encountered to perform optional projections, mostly into collections such as `RefCell<Vec<T>>` or `RefCell<HashMap<U, T>>`:

> This kind of API was briefly featured under Open questions in rust-lang#10514 back in 2013 (!)

```rust
let values = RefCell::new(vec![1, 2, 3, 4]);
let b = Ref::opt_map(values.borrow(), |vec| vec.get(2));
```

It primarily avoids this alternative approach to accomplish the same kind of projection which is both rather noisy and panicky:
```rust
let values = RefCell::new(vec![1, 2, 3, 4]);

let b = if values.get(2).is_some() {
    Some(Ref::map(values.borrow(), |vec| vec.get(2).unwrap()))
} else {
    None
};
```

### Open questions

The naming `opt_map` is preliminary. I'm not aware of prior art in std to lean on here, but this name should probably be improved if this functionality is desirable.

Since `opt_map` consumes the guard, and alternative syntax might be more appropriate which instead *tries* to perform the projection, allowing the original borrow to be recovered in case it fails:

```rust
pub fn try_map<U: ?Sized, F>(orig: Ref<'b, T>, f: F) -> Result<Ref<'b, U>, Self>
where
    F: FnOnce(&T) -> Option<&U>;
```

This would be more in line with the `try_map` method [provided by parking lot](https://docs.rs/lock_api/0/lock_api/struct.RwLockWriteGuard.html#method.try_map).

…atsakis

Remove giant badge in README

Is it meant to be this big? I haven't seen any other open source project with this sort of thing

Explain why borrows can't be held across yield point in async blocks

For rust-lang#78938.

…ss-composition, r=m-ou-se

TrustedRandomAaccess specialization composes incorrectly for nested iter::Zips

I found this while working on improvements for TRA.

After partially consuming a Zip adapter and then wrapping it into another Zip where the adapters use their `TrustedRandomAccess` specializations leads to the outer adapter returning elements which should have already been consumed.

If the optimizer gets tripped up by the addition this might affect performance for chained `zip()` iterators even when the inner one is not partially advanced but it would require more extensive fixes to `TrustedRandomAccess` to communicate those offsets earlier.

Included test fails on nightly, [playground link](https://play.rust-lang.org/?version=nightly&mode=debug&edition=2018&gist=24fa1edf8a104ff31f5a24830593b01f)

…-ou-se

Clarify what the effects of a 'logic error' are

This clarifies what a 'logic error' is (which is a term used to describe what happens if you put things in a hash table or btree and then use something like a refcell to break the internal ordering). This tries to be as vague as possible, as we don't really want to promise what happens, except "bad things, but not UB". This was discussed in rust-lang#80657

…RalfJung

Re-stabilize Weak::as_ptr and friends for unsized T

As per [T-lang consensus](https://hackmd.io/7r3_is6uTz-163fsOV8Vfg), this uses a branch to handle the dangling case. The discussed optimization of only doing the branch in the T: ?Sized case is left for a followup patch, as doing so is not trivial (as it requires specialization) and not _obviously_ better (as it requires using `wrapping_offset` rather than `offset` more).

<details><summary>Basically said optimization</summary>

Specialize on `T: Sized`:

```rust
fn as_ptr(&self) -> *const T {
    if [ T is Sized ] || !is_dangling(ptr) {
        (ptr as *mut T).set_ptr_value( (ptr as *mut u8).wrapping_offset(data_offset) )
    } else {
        ptr::null()
    }
}

fn from_raw(*const T) -> Self {
    if [ T is Sized ] || !ptr.is_null() {
        let ptr = (ptr as *mut RcBox).set_ptr_value( (ptr as *mut u8).wrapping_offset(-data_offset) );
        Weak { ptr }
    } else {
        Weak::new()
    }
}
```

(but with more `set_ptr_value` to avoid `Sized` restrictions and maintain metadata.)

Written in this fashion, this is not a correctness-critical specialization (i.e. so long as `[ T is Sized ]` is false for unsized `T`, it can be `rand()` for sized `T` without breaking correctness), but it's still touchy, so I'd rather do it in another PR with separate review.

---
</details>

This effectively reverts rust-lang#80422 and re-establishes rust-lang#74160. T-libs [previously signed off](rust-lang#74160 (comment)) on this stable API change in rust-lang#74160.

…crum

Make `x.py --color always` apply to logging too

Follow-up to rust-lang#78548, rust-lang#79004.

r? ```@Mark-Simulacrum```

…acrum

Add a regression test for rust-lang#76281

This has been fixed between 1.47.0-nightly (663d2f5 2020-08-22) and 1.47.0-nightly (5180f3d 2020-08-23). Maybe fixed by rust-lang#73526?

Created `wasm` dir, it currently has only one test but I'll move some wasm-related tests there as a follow-up.

Closes rust-lang#76281

…arkor

Do not suggest invalid code in pattern with loop

Fixes rust-lang#80913

…Simulacrum

Stabilize the poll_map feature

Stabilizes the `poll_map` feature as tracked by rust-lang#63514 (with a completed FCP).

…nization, r=Mark-Simulacrum

Put all feature gate tests under `feature-gates/`

There was one directory that had only a single test and there was also a
test in the top-level directory. This moves both of them to
`feature-gates/`.

…r=jyn514

Remove doctree::Import

Per the title. Part of cleaning up doctree

doctest: Reset errors before dropping the parse session

The first parse is to collect whether the code contains macros, has
`main`, and uses other crates. In that pass we ignore errors as those
will be reported when the test file is actually built.

For that we need to reset errors in the `Diagnostic` otherwise when
dropping it unhandled errors will be reported as compiler bugs.

Fixes rust-lang#80992

…ark-Simulacrum

Add a regression test for rust-lang#50041

AFAICT the test case never landed alongside the fix for the issue.

…=jyn514

codegen_cranelift: Fix redundant semicolon warn

…Simulacrum

Add sample code for Rc::new_cyclic

…imulacrum

Add test for rust-lang#34792

Closes rust-lang#34792