On mismatched delimiters, only point at empty blocks that are in the same line
We point at empty blocks when we have mismatched braces to detect cases where editors auto insert `}` after writing `{`. Gate this to only the case where the entire span is in the same line so we never point at explicitly empty blocks.
resolve, inconsistent binding mode: tweak wording
Now that we can have e.g. `let Ok(x) | Err(x) = res;`, it's no longer appropriate to refer to "the same *match arm*", so let's tweak the wording.
r? @estebank
test(pattern): add tests for combinations of pattern features
Reference issue #67311
Tests combinations of the following pattern features:
- bindings_after_at
- or_patterns
- slice_patterns
- box_patterns
r? @Centril
Parse & reject postfix operators after casts
This adds an explicit error messages for when parsing `x as Type[0]` or similar expressions. Our add an extra parse case for parsing any postfix operator (dot, indexing, method calls, await) that triggers directly after parsing `as` expressions.
My friend and I worked on this together, but they're still deciding on a github username and thus I'm submitting this for both of us.
It will immediately error out, but will also provide the rest of the parser with a useful parse tree to deal with.
There's one decision we made in how this produces the parse tree. In the situation `&x as T[0]`, one could imagine this parsing as either `&((x as T)[0])` or `((&x) as T)[0]`. We chose the latter for ease of implementation, and as it seemed the most intuitive.
Feedback welcome! This is our first change to the parser section, and it might be completely horrible.
Fixes#35813.
When encountering an Item in a pat context, point at the item def
```
error[E0308]: mismatched types
--> $DIR/const-in-struct-pat.rs:8:17
|
LL | struct foo;
| ----------- `foo` defined here
...
LL | let Thing { foo } = t;
| ^^^ expected struct `std::string::String`, found struct `foo`
|
= note: `foo` is interpreted as a unit struct, not a new binding
help: you can bind the struct field to a different name
|
LL | let Thing { foo: other_foo } = t;
| ^^^^^^^^^^^^^^
```
```
error[E0308]: mismatched types
--> $DIR/const.rs:14:9
|
LL | const FOO: Foo = Foo{bar: 5};
| ----------------------------- constant defined here
...
LL | FOO => {},
| ^^^
| |
| expected `&Foo`, found struct `Foo`
| `FOO` is interpreted as a constant, not a new binding
| help: use different name to introduce a new binding: `other_foo`
```
Fix#55631, fix#48062, cc #42876.
`delay_span_bug` when codegen cannot select obligation
Fix#69602, introduced in #60126 by letting the compiler continue past
type checking after encountering errors.
Invoke OptimizerLastEPCallbacks in PreLinkThinLTO
The default ThinLTO pre-link pipeline does not include optimizer last
extension points. Thus, when using the new LLVM pass manager & ThinLTO
& sanitizers on any opt-level different from zero, the sanitizer
function passes would be omitted from the pipeline.
Add optimizer last extensions points manually to the pipeline, but guard
registration with stage check in the case this behaviour changes in the
future.
encode `;` stmt without expr as `StmtKind::Empty`
Instead of encoding `;` statements without a an expression as a tuple in AST, encode it as `ast::StmtKind::Empty`.
r? @petrochenkov
Improve linking of crates with circular dependencies
Previously, the code responsible for handling the cycles between crates
introduces through weak lang items, would keep a set of missing language
items:
* extending it with items missing from the current crate,
* removing items provided by the current crate,
* grouping the crates when the set changed from non-empty back to empty.
This could produce incorrect results, if a lang item was missing from a
crate that comes after the crate that provides it (in the loop iteration
order). In that case the grouping would not take place.
The changes here address this specific failure scenario by keeping track
of two separate sets of crates. Those that are required to link successfully,
and those that are available for linking.
Verified using test case from #69368.
Remove `usable_size` APIs
This removes the usable size APIs:
- remove `usable_size` (obv)
- change return type of allocating methods to include the allocated size
- remove `_excess` API
r? @Amanieu
closesrust-lang/wg-allocators#17
Additionally verify that the current implementation of LLVM version
check (which uses lexicographic ordering) is good enough to exclude
versions before LLVM 9, where the new LLVM pass manager is unsupported.
Adjust Miri value visitor, and doc-comment layout components
I realized that I still didn't have quite the right intuition for how our `LayoutDetails` work, so I had to adjust the Miri value visitor to the things I understood better now. I also added some doc-comments to `LayoutDetails` as a hopefully canonical place to note such things.
The main visitor change is that we *first* look at all the fields (according to `FieldPlacement`), and *then* check the variants and handle `Multiple` appropriately. I did not quite realize how orthogonal "fields" and "variants" are.
I also moved the check for the scalar ABI to *after* checking all the fields; this leads to better (more type-driven) error messages.
And it looks like we can finally remove that magic hack for `ty::Generator`. :D
r? @oli-obk for the Miri/visitor changes and @eddyb for the layout docs
The Miri PR is at: https://github.com/rust-lang/miri/pull/1178
* Add a new method `has_infer_types_or_consts` that's used instead most
of the time, since there's generally no reason to only consider types.
* Remove use of `has_closure_types`, because closures are no longer
implicitly linked to the `InferCtxt`.
