Record proc macro harness order for use during metadata deserialization
Fixes#68690
When we generate the proc macro harness, we now explicitly recorder the
order in which we generate entries. We then use this ordering data to
deserialize the correct proc-macro-data from the crate metadata.
Fixes#68690
When we generate the proc macro harness, we now explicitly recorder the
order in which we generate entries. We then use this ordering data to
deserialize the correct proc-macro-data from the crate metadata.
add regression test for issue #68794
This is a minimal regression test for the issue #68794: "TEXTREL in
i686", which was fixed with e86019c4a0.
The test links a minimal rust static library into a shared library, and
checks that the linker didn't have to add the TEXTREL flag.
Rollup of 6 pull requests
Successful merges:
- #64069 (Added From<Vec<NonZeroU8>> for CString)
- #66721 (implement LowerExp and UpperExp for integers)
- #69106 (Fix std::fs::copy on WASI target)
- #69154 (Avoid calling `fn_sig` on closures)
- #69166 (Check `has_typeck_tables` before calling `typeck_tables_of`)
- #69180 (Suggest a comma if a struct initializer field fails to parse)
Failed merges:
r? @ghost
Suggest a comma if a struct initializer field fails to parse
Currently, we emit a "try adding a comma" suggestion if a comma is
missing in a struct definition. However, we emit no such suggestion if a
comma is missing in a struct initializer.
This commit adds a "try adding a comma" suggestion when we don't find a
comma during the parsing of a struct initializer field.
The change to `src/test/ui/parser/removed-syntax-with-1.stderr` isn't
great, but I don't see a good way of avoiding it.
Fix std::fs::copy on WASI target
Previously `std::fs::copy` on wasm32-wasi would reuse code from the `sys_common` module and would successfully copy contents of the file just to fail right before closing it.
This was happening because `sys_common::copy` tries to copy permissions of the file, but permissions are not a thing in WASI (at least yet) and `set_permissions` is implemented as an unconditional runtime error.
This change instead adds a custom working implementation of `std::fs::copy` (like Rust already has on some other targets) that doesn't try to call `set_permissions` and is essentially a thin wrapper around `std::io::copy`.
Fixes#68560.
implement LowerExp and UpperExp for integers
Addresses https://github.com/rust-lang/rust/issues/39479
This implementation is heavily based on the preexisting `macro_rules! impl_Display` in the same file. I don't like the liberal use of unsafe in that macro and would like to modify it so `unsafe` is only present where necessary. What is Rust's policy on doing such modifications?
Also, I couldn't figure out where to put tests, can I have some help with that?
Added From<Vec<NonZeroU8>> for CString
Added a `From<Vec<NonZeroU8>>` `impl` for `CString`
# Rationale
- `CString::from_vec_unchecked` is a subtle function, that makes `unsafe` code harder to audit when the generated `Vec`'s creation is non-trivial. This `impl` allows to write safer `unsafe` code thanks to the very explicit semantics of the `Vec<NonZeroU8>` type.
- One such situation is when trying to `.read()` a `CString`, see issue #59229.
- this lead to a PR: #59314, that was closed for being too specific / narrow (it only targetted being able to `.read()` a `CString`, when this pattern could have been generalized).
- the issue suggested another route, based on `From<Vec<NonZeroU8>>`, which is indeed a less general and more concise code pattern.
- quoting @shnatsel:
- > For me the main thing about making this safe is simplifying auditing - people have spent like an hour looking at just this one unsafe block in libflate because it's not clear what exactly is unchecked, so you have to look it up when auditing anyway. This has distracted us from much more serious memory safety issues the library had.
Having this trivial impl in stdlib would turn this into safe code with compiler more or less guaranteeing that it's fine, and save anyone auditing the code a whole lot of time.
Currently, we emit a "try adding a comma" suggestion if a comma is
missing in a struct definition. However, we emit no such suggestion if a
comma is missing in a struct initializer.
This commit adds a "try adding a comma" suggestion when we don't find a
comma during the parsing of a struct initializer field.
The change to `src/test/ui/parser/removed-syntax-with-1.stderr` isn't
great, but I don't see a good way of avoiding it.
Infer regions for opaque types in borrowck
This is a step towards the goal of typeck not doing region inference.
The commits up to `Arena allocate the result of mir_borrowck` are various bug fixes and prerequisites.
The remaining commits move opaque type inference to borrow checking.
r? @nikomatsakis
* Use better span for member constraint errors
* Avoid a bad suggestion
* Don't report member constraint errors if we have other universal
region errors.
Follow-up to #68848
This PR contains some late changes to #68848 that somehow didn't get included when that PR was merged in a roll-up.
r? @petrochenkov
Fix extra subslice lowering
We are currently ICEing on e.g.
```rust
fn main() {
let [.., b @ ..] = [1, 2];
b;
}
```
This happens because `b @ ..` registers a binding such that `b;` is OK, but then we forget to lower that binding in `rustc_ast_lowering`.
Fixes#69103.
r? @davidtwco
typeck: clarify def_bm adjustments & add tests for or-patterns
Clarify the adjustment algorithm for the expected type / default binding-modes when type checking patterns with more documentation and tweaks that make the algorithm more independent of the pattern forms.
Also resolve the FIXME noted for or-patterns by deciding that the current implementation is correct, noting the rationale and adding tests for the current implementation.
cc https://github.com/rust-lang/rust/issues/54883
r? @oli-obk @varkor
Use a `ParamEnvAnd<Predicate>` for caching in `ObligationForest`
Previously, we used a plain `Predicate` to cache results (e.g. successes
and failures) in ObligationForest. However, fulfillment depends on the
precise `ParamEnv` used, so this is unsound in general.
This commit changes the impl of `ForestObligation` for
`PendingPredicateObligation` to use `ParamEnvAnd<Predicate>` instead of
`Predicate` for the associated type. The associated type and method are
renamed from 'predicate' to 'cache_key' to reflect the fact that type is
no longer just a predicate.
