Namely, we thread down the `HirId` of the explicit type of the
argument. In the case of the special `self` variable with an implicit
type, we also thread down a description of its structure (`self`/`mut
self`/`&self`/`&mut self`).
As a drive-by: the ref_for_guards created by `fn declare_binding`
should not have been tagged as user_variables in the first
place. These secret internal locals are *pointers* to user variables,
but themselves are not such (IMO. For now at least.)
the originally generated code was highly suboptimal
this brings it close to the same code or even exactly the same as a
manual while-loop by eliminating a branch and the
double stepping of n-1 + 1 steps
The intermediate trait lets us circumvent the specialization
type inference bugs
Use String, not &str in some collection examples
Discussed in #46966
Overuse of borrowed values in data structures is a common mistake I see in Rust user forums. Users who copy&paste such examples end up fighting with the borrow checker as soon as they replace string literals with some real values.
This changes a couple of examples to use `String`, and it adds opportunity to demonstrate use of `Borrow`.
Add ability to apply custom derive to union types.
The Union item type has been included in the allowed types for a custom derive.
fyi @abonander
Closes#50223
`has_any_child_of` is hot. It allocates a `Vec` that almost always
doesn't exceed a length of 1.
This patch peels off the first iteration of the loop, avoiding the need
for the `Vec` creation in ~99% of cases.
The acronym is not descriptive unless one has seen it before.
* Rename the `oom` function to `handle_alloc_error`. It was **stabilized in 1.28**, so if we do this at all we need to land it this cycle.
* Rename `set_oom_hook` to `set_alloc_error_hook`
* Rename `take_oom_hook` to `take_alloc_error_hook`
Bikeshed: `alloc` v.s. `allocator`, `error` v.s. `failure`
Improve memoization and refactor NLL type check
I have a big branch that is refactoring NLL type check with the goal of introducing canonicalization-based memoization for all of the operations it does. This PR contains an initial prefix of that branch which, I believe, stands alone. It does introduce a few smaller optimizations of its own:
- Skip operations that are trivially a no-op
- Cache the results of the dropck-outlives computations done by liveness
- Skip resetting unifications if nothing changed
r? @pnkfelix
Add existential type definitions
Note: this does not allow creating named existential types, it just desugars `impl Trait` to a less (but still very) hacky version of actual `existential type` items.
r? @nikomatsakis
impl Hash for !
This was missing in some generic code I was writing and I figured that it would be worthwhile to add this. Blanket-requiring these traits to allow usage of errors in `HashSet`s and `HashMap`s isn't too unreasonable of a use case, and a prerequisite for allowing `!` as an error in such situations is this impl.
libstd: add an RAII utility for sys_common::mutex::Mutex
It is indeed debatable whether or not we should introduce more sophistication like this to the lowest layer of a system library. In fact, `Drop::drop()` cannot be `unsafe` (IIRC there was a discussion on introducing an unsafe variant of `Drop` whose entire scope must be within `unsafe`)
Removed two unused variables in os.rs
Issue #51419 suggested removing two unused variables in `libstd/sys/redox/os.rs`. This PR implements that change.
It compiles for me locally, but I haven't run any other tests.
Minor language change to CONTRIBUTING.md
In the `#building` section, the doc read that a default configuration "shall use around 3.5 GB of disk space." Changed "shall use" to "requires," as I don't think it's meant to sound so prescriptive.
Follow up to #51508, make parse_block public instead parse_block_expr
This is an follow up to #51508
I mistakenly made parse_block_expr public instead of parse_block.
This fixes this.
refactor: create multiple HIR items for imports
When lowering `use` statements into HIR, they get a `Def` of the thing they're pointing at. This is great for things that need to know what was just pulled into scope. However, this is a bit misleading, because a `use` statement can pull things from multiple namespaces if their names collide. This is a problem for rustdoc, because if there are a module and a function with the same name (for example) then it will only document the module import, because that's that the lowered `use` statement points to.
The current version of this PR does the following:
* Whenever the resolver comes across a `use` statement, it loads the definitions into a new `import_map` instead of the existing `def_map`. This keeps the resolutions per-namespace so that all the target definitions are available.
* When lowering `use` statements, it looks up the resolutions in the `import_map` and creates multiple `Item`s if there is more than one resolution.
* To ensure the `NodeId`s are properly tracked in the lowered module, they need to be created in the AST, and pulled out as needed if multiple resolutions are available.
Fixes https://github.com/rust-lang/rust/issues/34843
Add Ref/RefMut map_split method
As proposed [here](https://internals.rust-lang.org/t/make-refcell-support-slice-splitting/7707).
TLDR: Add a `map_split` method that allows multiple `RefMut`s to exist simultaneously so long as they refer to non-overlapping regions of the original `RefCell`. This is useful for things like the slice `split_at_mut` method.
Add lint for intra link resolution failure
This PR is almost done, just remains this note:
```
note: requested on the command line with `-W intra-link-resolution-failure`
```
I have no idea why my lint is considered as being passed through command line and wasn't able to find where it was set. If anyone has an idea, it'd be very helpful!
cc @QuietMisdreavus
Prevent Windows filesystem races in bootstrap tests
Fixes#51595.
This also makes bootstrap tests run near last in `./x.py test` invocations
since they are unlikely to fail.
r? @petrochenkov
