Rollup merge of #63934 - Aaron1011:fix/impl-trait-coherence, r=nikomatsakis
Fix coherence checking for impl trait in type aliases
**UPDATE**: This PR now treats all opaque types as remote. The original description appears below, but is no longer accurate.
Fixes #63677
[RFC 2071](https://github.com/rust-lang/rfcs/pull/2071) (impl-trait-existential-types) does not explicitly state how `type_alias_impl_trait` should interact with coherence. However, there's only one choice which makes sense - coherence should look at the underlying type (i.e. the *"defining"* type of the `impl Trait`) of the type alias, just like we do for non-`impl Trait` type aliases.
Specifically, `impl Trait` type aliases that resolve to a local type should be treated like a local type with respect to coherence (e.g. `impl Trait` type aliases which resolve to a foreign type should be treated as a foreign type, and those that resolve to a local type should be treated as a local type).
Since neither inherent impls nor direct trait impl (i.e. `impl MyType` or `impl MyTrait for MyType`) are allowed for type aliases, this usually does not come up. Before we ever attempt to do coherence checking, we will have errored out if an `impl Trait` type alias was used directly in an `impl` clause.
However, during trait selection, we sometimes need to prove bounds like `T: Sized` for some type `T`. If `T` is an impl trait type alias, this requires to know the coherence behavior for `impl Trait` type aliases when we perform coherence checking.
Note: Since determining the underlying type of an `impl Trait` type alias requires us to perform body type checking, this commit causes us to type check some bodies easier than we otherwise would have. However, since this is done through a query, this shouldn't cause any problems
For completeness, I've added an additional test of the coherence-related behavior of `impl Trait` type aliases.
Aaron Hill [Tue, 27 Aug 2019 01:32:14 +0000 (21:32 -0400)]
Fix coherence checking for impl trait in type aliases
Fixes #63677
RFC #2071 (impl-trait-existential-types) does not explicitly state how
impl trait type alises should interact with coherence. However, there's
only one choice which makes sense - coherence should look at the
underlying type (i.e. the 'defining' type of the impl trait) of the type
alias, just like we do for non-impl-trait type aliases.
Specifically, impl trait type alises which resolve to a local type
should be treated like a local type with respect to coherence (e.g.
impl trait type aliases which resolve to a forieign type should be
treated as a foreign type, and those that resolve to a local type should
be treated as a local type).
Since neither inherent impls nor direct trait impl (i.e. `impl MyType`
or `impl MyTrait for MyType`) are allowd for type aliases, this
usually does not come up. Before we ever attempt to do coherence
checking, we will have errored out if an impl trait type alias was used
directly in an 'impl' clause.
However, during trait selection, we sometimes need to prove bounds like
'T: Sized' for some type 'T'. If 'T' is an impl trait type alias, this
requires to know the coherence behavior for impl trait type aliases when
we perform coherence checking.
Note: Since determining the underlying type of an impl trait type alias
requires us to perform body type checking, this commit causes us to type
check some bodies easlier than we otherwise would have. However, since
this is done through a query, this shouldn't cause any problems
For completeness, I've added an additional test of the coherence-related
behavior of impl trait type aliases.
Auto merge of #64316 - alexcrichton:cleanup-shim, r=Mark-Simulacrum
Delete most of `src/bootstrap/bin/rustc.rs`
This commit is an attempt at deleting as much of the `rustc.rs` shim that we have in rustbuild as possible. This shim predates `RUSTFLAGS` and is as old as rustbuild itself. While useful for quick hacks, it subverts Cargo's knowledge of `rustc`, makes it more difficult to build crates out of rustbuild, and is generally a hazard/code smell due to its architecture.
Additionally since the inception of this script we've added a number of features to Cargo such as profile overrides and `RUSTFLAGS`. This commit attempts to use these features of Cargo as much as possible to delete almost all of `src/bootstrap/bin/rustc.rs`. It's hoped that all new configuration for the Rust compiler can be codified in rustbuild rather than in this shim, allowing Cargo to have more knowledge about what's going on and making it a bit easier to reproduce builds outside of Cargo itself.
This was primarily motivated by some recent work on std-aware Cargo, and is also generally a cleanup of the script itself. This internally resulted in a number of refactorings of rustbuild itself, and the commits should be readable one-at-a-time instead of having to digest them all at once.
Alex Crichton [Tue, 17 Sep 2019 18:23:09 +0000 (11:23 -0700)]
rustbuild: Pass `-Zsave-analysis` during tests
This is needed to ensure that the crates during a normal build are
shared with the crates during testing, otherwise they'll end up hasing
differently and we'll recompile crates like `core` during tests.
Alex Crichton [Mon, 9 Sep 2019 17:17:38 +0000 (10:17 -0700)]
Allow adding `RUSTFLAGS` after `Builder::cargo`
This commit changes the return type of `Builder::cargo` to return a
builder that allows dynamically adding more `RUSTFLAGS` values
after-the-fact. While not used yet, this will later be used to delete
more of `rustc.rs`
Alex Crichton [Thu, 15 Aug 2019 20:42:39 +0000 (13:42 -0700)]
bootstrap: Add a helper for managing RUSTFLAGS
Most of `bootstrap/bin/rustc.rs` doesn't need to exist with the advent
of `RUSTFLAGS` (yes this is super old) so this starts by refactoring a
bit to make it easier locally in the `Builder::cargo` method to append
to `RUSTFLAGS` that gets down to rustc.
Auto merge of #64272 - Mark-Simulacrum:parallel-handler, r=estebank
Refactor librustc_errors::Handler API
This should be reviewed by-commit.
The last commit moves all fields into an inner struct behind a single lock; this is done to prevent possible deadlocks in a multi-threaded compiler, as well as inconsistent state observation.
Auto merge of #64695 - Centril:rollup-t1xnl2c, r=Centril
Rollup of 7 pull requests
Successful merges:
- #64294 (Fix `Stdio::piped` example code and lint)
- #64670 (Cleanup syntax::ext::build)
- #64674 (Propagate `types.err` in locals further to avoid spurious knock-down errors)
- #64676 (Parse assoc type bounds in generic params and provide custom diagnostic)
- #64677 (remove outdated comment)
- #64679 (Infer consts more consistently)
- #64688 (Clarify the "since" tidy check)
Rollup merge of #64679 - skinny121:const-infer, r=varkor
Infer consts more consistently
Moved some duplicated logic in `TypeRelation` methods into `super_combined_consts`. Before some `TypeRelation`s like `Lub` wasn't using `replace_if_possible`, meaning some inference types were staying around longer than they should be.
Rollup merge of #64670 - Mark-Simulacrum:ext-build-simplify, r=petrochenkov
Cleanup syntax::ext::build
I suspect most of this code could be inlined but I only removed the bits where the inlining didn't really hurt readability (i.e., method call -> function call) or the completely unused code.
Rollup merge of #64294 - wchargin:wchargin-stdio-piped-docs, r=Dylan-DPC
Fix `Stdio::piped` example code and lint
Summary:
Invoking `rev` does not add a trailing newline when none is present in
the input (at least on my Debian). Nearby examples use `echo` rather
than `rev`, which probably explains the source of the discrepancy.
Also, a `mut` qualifier is unused.
Test Plan:
Copy the code block into <https://play.rust-lang.org> with a `fn main`
wrapper, and run it. Note that it compiles and runs cleanly; prior to
this commit, it would emit an `unused_mut` warning and then panic.
Auto merge of #64666 - Centril:rollup-tp98vlr, r=Centril
Rollup of 9 pull requests
Successful merges:
- #63907 (Add explanation to type mismatch involving type params and assoc types)
- #64615 (rustbuild: Turn down compression on exe installers)
- #64617 (rustbuild: Turn down compression on msi installers)
- #64618 (rustbuild: Improve output of `dist` step)
- #64619 (Fixes #63962. Hint about missing tuple parentheses in patterns)
- #64634 (Update to LLVM 9.0.0)
- #64635 (Allow using fn pointers in const fn with unleash miri)
- #64660 (unify errors for tuple/struct variants)
- #64664 (fully remove AstBuilder)
let x: i32 = foo(non_const_fn_ptr); // OK
let y: i32 = foo(const_fn_ptr); // OK
const X: i32 = foo(non_const_fn_ptr); // ERROR: `non_const_fn` is not `const fn`
const Y: i32 = foo(const_fn_ptr); // OK
```
Rollup merge of #64618 - alexcrichton:improve-dist-output, r=Mark-Simulacrum
rustbuild: Improve output of `dist` step
* Pass `/Q` to `iscc` on Windows to supress the thousands of lines of
output about compressing documentation.
* Print out what's happening before long steps
* Use `timeit` to print out timing information for long-running
installer assemblies.
* Try to scope output of `Dist ...` to not also encompass actual build steps
Rollup merge of #64617 - alexcrichton:smaller-msi, r=Mark-Simulacrum
rustbuild: Turn down compression on msi installers
This is the same as #64615 except applied to our MSI installers. The
same fix is applied effectively bringing these installers in line with
the gz tarball installers, which are about 3x faster to produce locally
and likely much faster to produce on CI.
Rollup merge of #64615 - alexcrichton:smaller-exe, r=Mark-Simulacrum
rustbuild: Turn down compression on exe installers
The Windows dist builders are the slowest builders right now, and the
distribution phase of them is enormously slow clocking in at around 20
minutes to build all the related installers. This commit starts to
optimize these by turning down the compression level in the `exe`
installers. These aren't super heavily used so there's no great need for
them to be so ultra-compressed, so let's dial back the compression
parameters to get closer to the rest of our xz archives. This brings the
installer in line with the gz tarball installer locally, and also brings
the compression settings on par with the rest of our xz installers.
Auto merge of #64658 - Centril:rollup-9s3raz6, r=Centril
Rollup of 9 pull requests
Successful merges:
- #64010 (Stabilize `param_attrs` in Rust 1.39.0)
- #64136 (Document From trait for LhsExpr in parser)
- #64342 (factor out pluralisation remains after #64280)
- #64347 (Add long error explanation for E0312)
- #64621 (Add Compatibility Notes to RELEASES.md for 1.38.0)
- #64632 (remove the extra comma after the match arm)
- #64640 (No home directory on vxWorks)
- #64641 (Exempt extern "Rust" from improper_ctypes)
- #64642 (Fix the span used to suggest avoiding for-loop moves)
Rollup merge of #64642 - cuviper:move-for-loop-snippet, r=varkor
Fix the span used to suggest avoiding for-loop moves
It was using the snippet from the "use" span, which often renders the
same, but with closures that snippet is on the start of the closure
where the value is captured. We should be using the snippet from the
span where it was moved into the `for` loop, which is `move_span`.
there are two case that doesn't not match the original macro pattern at [here](https://github.com/rust-lang/rust/blob/master/src/librustc_lint/unused.rs#L146) and [here](https://github.com/rust-lang/rust/blob/master/src/libsyntax/parse/diagnostics.rs#L539) as the provided param is already a bool or the check condition is not `x != 1`, so I change the macro accept a boolean expr instead of number to fit all the cases.
Rollup merge of #64136 - crgl:doc-from-parser-lhs, r=Centril
Document From trait for LhsExpr in parser
Add doc for From trait for converting P<Expr> and Option<ThinVec<Attribute>> to LhsExpr
As part of issue rust-lang#51430 (cc @skade).
Both of these should just be moving an address and setting a discriminant in an enum. The main thing I'm not sure about is whether it's worth documenting the branch in the From<Option<ThinVec<Attribute>>. As far as I can tell it doesn't seem like it is optimized away (although if the discriminant happened to work out you could just copy the pointer and the discriminant which might be cheaper, but that's not guaranteed). So it seems like if it's being called often, it's doubling the number of possible branch mispredictions on this Option, which could be a significant cost.
Let me know if there's anything that needs fixing and I'll get to it as soon as possible!
* Documentation comments like `/// Doc` on parameters.
* Code expansion of a user-defined `#[proc_macro_attribute]` macro used on parameters.
* Built-in attributes other than `cfg`, `cfg_attr`, `allow`, `warn`, `deny`, and `forbid`. Currently, only the lints `unused_variables` and `unused_mut` have effect and may be controlled on parameters.
## Motivation
The chief motivations for stabilizing `param_attrs` include:
* Finer conditional compilation with `#[cfg(..)]` and linting control of variables.
* Richer macro DSLs created by users.
* External tools and compiler internals can take advantage of the additional information that the parameters provide.
For more examples, see the [RFC][rfc motivation].
## Reference guide
In the grammar of function and function pointer, the grammar of variadic tails (`...`) and parameters are changed respectively from:
More generally, where there's a list of formal (value) parameters separated or terminated by `,` and delimited by `(` and `)`. Each parameter in that list may optionally be prefixed by `OuterAttr+`.
Note that in all cases, `OuterAttr*` applies to the whole parameter and not just the pattern. This distinction matters in pretty printing and in turn for macros.
## History
* On 2018-10-15, @Robbepop proposes [RFC 2565][rfc], "Attributes in formal function parameter position".
* On 2019-04-30, [RFC 2565][rfc] is merged and the tracking issue is made.
* On 2019-06-12, a partial implementation was completed. The implementation was done in [#60669][60669] by @c410-f3r and the PR was reviewed by @petrochenkov and @Centril.
* On 2019-07-29, [#61238][61238] was fixed in [#61856][61856]. The issue fixed was that lint attributes on function args had no effect. The PR was written by @c410-f3r and reviewed by @matthewjasper, @petrochenkov, and @oli-obk.
* On 2019-08-02, a bug [#63210][63210] was filed wherein the attributes on formal parameters would not be passed to macros. The issue was about forgetting to call the relevant method in `fn print_arg` in the pretty printer. In [#63212][63212], written by @Centril on 2019-08-02 and reviewed by @davidtwco, the issue aforementioned was fixed.
* This PR stabilizes `param_attrs`.
## Tests
* [On Rust 2018, attributes aren't permitted on function parameters without a pattern in trait definitions.](https://github.com/rust-lang/rust/blob/master/src/test/ui/rfc-2565-param-attrs/param-attrs-2018.rs)
* [All attributes that should be allowed. This includes `cfg`, `cfg_attr`, and lints check attributes.](https://github.com/rust-lang/rust/blob/master/src/test/ui/rfc-2565-param-attrs/param-attrs-allowed.rs)
* [Built-in attributes, which should be forbidden, e.g., `#[test]`, are.](https://github.com/rust-lang/rust/blob/master/src/test/ui/rfc-2565-param-attrs/param-attrs-builtin-attrs.rs)
* [`cfg` and `cfg_attr` are properly evaluated.](https://github.com/rust-lang/rust/blob/master/src/test/ui/rfc-2565-param-attrs/param-attrs-cfg.rs)
* [`unused_mut`](https://github.com/rust-lang/rust/blob/46f405ec4d7c6bf16fc2eaafe7541019f1da2996/src/test/ui/rfc-2565-param-attrs/param-attrs-cfg.rs) and [`unused_variables`](https://github.com/rust-lang/rust/blob/master/src/test/ui/lint/lint-unused-variables.rs) are correctly applied to parameter patterns.
* [Pretty printing takes formal parameter attributes into account.](https://github.com/rust-lang/rust/blob/master/src/test/ui/rfc-2565-param-attrs/param-attrs-pretty.rs)
## Possible future work
* Custom attributes inside function parameters aren't currently supported but it is something being worked on internally.
* Since documentation comments are syntactic sugar for `#[doc(...)]`, it is possible to allow literal `/// Foo` comments on function parameters.
`min_by` and `max_by` are somewhat trivial to implement, but not entirely because `min_by` returns the first value in case the two are equal (and `max_by` the second). `min` and `max` can be implemented in terms of `min_by` and `max_by`, but not as easily the other way around.
To give an example of why I think these functions could be useful: the `Iterator::{min_by, min_by_key, max_by, max_by_key}` methods all currently hard-code the behavior mentioned above which is an ever so small duplication of logic. If we delegate them to `cmp::{min_by, max_by}` methods instead, we get the correct behavior for free. (edit: this is now included in the PR)
I added `min_by_key` / `max_by_key` for consistency's sake but I wouldn't mind removing them. I don't have a particular use case in mind for them, and `min_by` / `max_by` seem to be more useful.