This PR fixes an issues where rustc would ignore subsequent
`#[diagnostic::on_unimplemented]` attributes. The [corresponding
RFC](https://rust-lang.github.io/rfcs/3368-diagnostic-attribute-namespace.html)
specifies that the first matching instance of each option is used.
Invalid attributes are linted and otherwise ignored.
Rollup of 5 pull requests
Successful merges:
- #116219 (Relate alias ty with variance)
- #116315 (Do not check for impossible predicates in const-prop lint.)
- #116436 (Structurally normalize for closure)
- #116597 (Prevent showing methods from blanket impls of not available foreign traits to show up in the search results)
- #116627 (small cleanup)
r? `@ghost`
`@rustbot` modify labels: rollup
When encountering a for loop that is rejected by the borrow checker
because it is being advanced within its body, provide a structured
suggestion for `while let Some(pat) = iter.next()`.
Prevent showing methods from blanket impls of not available foreign traits to show up in the search results
Fixes https://github.com/rust-lang/rust/issues/115480.
In the case that the blanket impl trait is not available in the current crate, we prevent adding its methods in the search index.
Now how I found how to fix the issue: the `equivalent` method is not generated in the documentation pages but was still added to the search index. To render impls, we iterate over `cache.impls` so I took a look at how this was generated. Inside `formats/cache.rs`, we have `CacheBuilder::populate` where we push impls into `impls` but with this condition:
```rust
if cx.cache.traits.contains_key(&trait_did) {
```
I re-used this condition in `CacheBuilder::fold_item` to prevent this method from being added in `cache.search_index` or `cache.orphan_impl_items`.
PS: If you want to double-check if the added test works, just comment the code I added in `cache.rs` and it should fail.
r? ``@notriddle``
Do not check for impossible predicates in const-prop lint.
The enclosing query already checks for them, and replaces the body with a single `unreachable` if they are indeed impossible.
Relate alias ty with variance
In the new solver, turns out that the subst-relate branch of the alias-relate predicate was relating args invariantly even for opaques, which have variance 💀.
This change is a bit more invasive, but I'd rather not special-case it [here](aeaa5c30e5/compiler/rustc_trait_selection/src/solve/alias_relate.rs (L171-L190)) and then have it break elsewhere. I'm doing a perf run to see if the extra call to `def_kind` is that expensive, if it is, I'll reconsider.
r? ``@lcnr``
Compute NLL loan scopes using the polonius model
For a *location-insensitive* analysis (that is, without expressiveness improvements for users yet), this PR implements loans going out of scope using reachability and liveness, rather than checking if the issuing region's values contain a given CFG point. This is equivalent to NLL scopes and computes the same data.
r? `@matthewjasper`
A couple of notes:
- there are some assumptions about SCC representatives, placeholders, free regions, and member constraints that I believe hold, and they're documented in the code
- this passes all the UI tests with `-Zpolonius=next` -- the perf is [not terrible](https://github.com/rust-lang/rust/pull/112432#issuecomment-1749685862) and there are a bunch of ways to improve it in the future.
- there's a fixme left, hopefully Matthew you know a clean way to get the information it mentions.
Implement `-Clink-self-contained=-linker` opt out
This implements the `-Clink-self-contained` opt out necessary to switch to lld by changing rustc's defaults instead of cargo's.
Components that are enabled and disabled on the CLI are recorded, for the purpose of being merged with the ones which the target spec will declare (I'll open another PR for that tomorrow, for easier review).
For MCP510, we now check whether using the self-contained linker is disabled on the CLI. Right now it would only be sensible to with `-Zgcc-ld=lld` (and I'll add some checks that we don't both enable and disable a component on the CLI in a future PR), but the goal is to simplify adding the check of the target's enabled components here in the follow-up PRs.
r? `@petrochenkov`
Fix overflow checking in range patterns
When a range pattern contains an overflowing literal, if we're not careful we might not notice the overflow and use the wrapped value. This makes for confusing error messages because linting against overflowing literals is only done in a later pass. So when a range is invalid we check for overflows to provide a better error.
This check didn't use to handle negative types; this PR fixes that. First commit adds tests, second cleans up without changing behavior, third does the fix.
EDIT: while I was at it, I fixed a small annoyance about the span of the overflow lint on negated literals.
Fixes https://github.com/rust-lang/rust/issues/94239
Implement `slice::split_once` and `slice::rsplit_once`
Feature gate is `slice_split_once` and tracking issue is #112811. These are equivalents to the existing `str::split_once` and `str::rsplit_once` methods.
Update cargo
5 commits in 794d0a82547f3081044c0aca7b6083733ce51344..6fa6fdc7606cfa664f9bee2fb33ee2ed904f4e88
2023-10-03 23:19:33 +0000 to 2023-10-10 23:06:08 +0000
- test(build): generalize test assertion for non-rustup env (rust-lang/cargo#12804)
- chore: Sort dependency tables (rust-lang/cargo#12803)
- fix(install): Suggest an alternative version on MSRV failure (rust-lang/cargo#12798)
- rustdoc: remove the word "Version" from test cases (rust-lang/cargo#12800)
- Add unsupported lowercase `-z` flag suggestion for `-Z` flag (rust-lang/cargo#12788)
r? ghost
Document `diagnostic_namespace` feature
This adds it to the rust unstable book.
FWIW: I couldn't find a way to serve the book locally (please send help), so I can't check that this renders correctly.
cc `@weiznich`
Add explicit-endian String::from_utf16 variants
This adds the following APIs under `feature(str_from_utf16_endian)`:
```rust
impl String {
pub fn from_utf16le(v: &[u8]) -> Result<String, FromUtf16Error>;
pub fn from_utf16le_lossy(v: &[u8]) -> String;
pub fn from_utf16be(v: &[u8]) -> Result<String, FromUtf16Error>;
pub fn from_utf16be_lossy(v: &[u8]) -> String;
}
```
These are versions of `String::from_utf16` that explicitly take [UTF-16LE and UTF-16BE](https://unicode.org/faq/utf_bom.html#gen7). Notably, we can do better than just the obvious `decode_utf16(v.array_chunks::<2>().copied().map(u16::from_le_bytes)).collect()` in that:
- We handle the case where the byte slice is not an even number of bytes, and
- In the case that the UTF-16 is native endian and the slice is aligned, we can forward to `String::from_utf16`.
If the Unicode Consortium actively defines how to handle character replacement when decoding a UTF-16 bytestream with a trailing odd byte, I was unable to find reference. However, the behavior implemented here is fairly self-evidently correct: replace the single errant byte with the replacement character.
Also consider call and yield as MIR SSA.
The SSA analysis on MIR only considered `Assign` statements as defining a SSA local.
This PR adds assignments as part of a `Call` or `Yield` terminator in that category.
This mainly allows to perform CopyProp on a call return place.
The only subtlety is in the dominance property: the assignment is only complete at the beginning of the target block.
Rollup of 7 pull requests
Successful merges:
- #109422 (rustdoc-search: add impl disambiguator to duplicate assoc items)
- #116250 (On type error of closure call argument, point at earlier calls that affected inference)
- #116444 (add test for const-eval error in dead code during monomorphization)
- #116503 (Update docs for mips target tier demotion.)
- #116559 (Mark `new_in` as `const` for BTree collections)
- #116560 (In smir use `FxIndexMap` to store indexed ids)
- #116574 (Update books)
r? `@ghost`
`@rustbot` modify labels: rollup
