Commit Graph

28060 Commits

Author SHA1 Message Date
bors
ea82cc47b9 Auto merge of #16692 - Veykril:methres, r=Veykril
internal: Remove dead branches in `method_resolution::is_valid_candidate`
2024-02-27 09:03:21 +00:00
Lukas Wirth
3a1b4c29b3 internal: Remove dead branches in method_resolutiopn::is_valid_candidate 2024-02-27 10:00:45 +01:00
bors
d4d9d0c85a Auto merge of #16691 - Veykril:completion-analysis-panic, r=Veykril
fix: Fix completions panicking with certain macro setups

Unable to figure out a test case for this but managed to run into it in r-a reproducably.

Fixes https://github.com/rust-lang/rust-analyzer/issues/16266 Fixes https://github.com/rust-lang/rust-analyzer/issues/13255
2024-02-27 08:39:34 +00:00
Lukas Wirth
cc7fe32ba3 fix: Fix completions panicking with certain macro setups 2024-02-27 09:35:57 +01:00
Tavo Annus
a2bf15eede Filter out false positive errors 2024-02-26 20:17:09 +02:00
Tavo Annus
be6f8e2648 Add make_tuple tactic 2024-02-26 20:17:09 +02:00
Tavo Annus
8bd30e9b3f Improve generics handling in term search 2024-02-26 20:17:09 +02:00
bors
5fead606bc Auto merge of #16555 - davidbarsky:david/speedup-completions-by-exploiting-orphan-rules, r=Veykril
performance: Speed up Method Completions By Taking Advantage of Orphan Rules

(Continues https://github.com/rust-lang/rust-analyzer/pull/16498)

This PR speeds up method completions by doing two things without regressing `analysis-stats`[^1]:
- Filter candidate traits prior to calling `iterate_path_candidates` by relying on orphan rules (see below for a slightly more in-depth explanation). When generating completions [on `slog::Logger`](5e9e59c312/common/src/ledger.rs (L78)) in `oxidecomputer/omicron` as a test, this PR halved my completion times—it's now 454ms cold and 281ms warm. Before this PR, it was 808ms cold and 579ms warm.
- Inline some of the method candidate checks into `is_valid_method_candidate` and remove some unnecessary visibility checks. This was suggested by `@Veykril` in [this comment](https://github.com/rust-lang/rust-analyzer/pull/16498#issuecomment-1929864427).

We filter candidate traits by taking advantage of orphan rules. For additional details, I'll rely on `@WaffleLapkin's` explanation  [from Zulip](https://rust-lang.zulipchat.com/#narrow/stream/185405-t-compiler.2Frust-analyzer/topic/Trait.20Checking/near/420942417):

> A type `A` can only implements traits which
> 1. Have a blanket implementation (`impl<T> Trait for T {}`)
> 2. Have implementation for `A` (`impl Trait for A {}`)
>
> Blanket implementation can only exist in `Trait`'s crate. Implementation for `A` can only exist in `A`'s or `Trait`'s crate.

Big thanks to Waffle for its keen observation!

---

I think some additional improvements are possible:
- `for_trait_and_self_ty` seemingly does not distinguish between `&T`, `&mut T`, or `T`, resulting in seemingly irrelevant traits like `tokio::io::AsyncWrite` being being included for, e.g., `&slog::Logger`. I don't know they're being considered due to the [autoref/autoderef behavior](a02a219773/crates/hir-ty/src/method_resolution.rs (L945-L962)), but I wonder if it'd make sense to filter by mutability earlier and not consider trait implementations that require `&mut T` when we only have a `&T`.
- The method completions [spend a _lot_ of time in unification](https://rust-lang.zulipchat.com/#narrow/stream/185405-t-compiler.2Frust-analyzer/topic/Trait.20Checking/near/421072356), and while there might be low-hanging fruit there, it might make more sense to wait for the new trait solver in `rustc`. I dunno.

[^1]: The filtering occurs outside of typechecking, after all.
2024-02-26 18:05:52 +00:00
David Barsky
c246a93046 completions: speed up completions by filtering non-applicable traits 2024-02-26 12:49:11 -05:00
bors
1d3558bfe1 Auto merge of #16684 - Veykril:synthetic-fmt-args, r=Veykril
fix: Don't panic on synthetic syntax in inference diagnostics

Temporary fix for https://github.com/rust-lang/rust-analyzer/issues/16682

We ought to rethink how we attach diagnostics to things, as IDs don't work for `format_args` like that!
2024-02-26 16:47:57 +00:00
Lukas Wirth
3856648b52 fix: Don't panic on synthetic syntax in inference diagnostics 2024-02-26 17:46:03 +01:00
dfireBird
8fa903a447
add test for checking struct generated through macro 2024-02-26 21:32:59 +05:30
dfireBird
2ea70662f0
make assist not applicable if there is no missing field 2024-02-26 21:21:56 +05:30
Shoyu Vanilla
a4021f6ed5 Use lang-item instead of db lookup for FnTrait kind 2024-02-27 00:38:53 +09:00
Shoyu Vanilla
46cdce1b53 Update expectation tests 2024-02-27 00:18:40 +09:00
Shoyu Vanilla
e36a31bd95 Port closure kind deduction logic from rustc 2024-02-27 00:18:38 +09:00
Shoyu Vanilla
763714145a Add a new failing test 2024-02-27 00:15:25 +09:00
bors
96505787b4 Auto merge of #16678 - roife:fix-issue-16660, r=lnicola
fix: panic when inlining callsites inside macros' parameters

Close #16660, #12429, #10695.

When `inline_into_callers` encounters callsites in macros parameters, it can lead to panics. Since there is no perfect way to handle macros, this PR directly filters out these cases.
2024-02-26 14:44:22 +00:00
roife
61b576c5ab fix: fmt 2024-02-26 22:36:47 +08:00
bors
f7f63ccd8d Auto merge of #16680 - lnicola:salsa-doc-test, r=Veykril
internal: Remove salsa compile_fail tests

I can't figure out how to reproduce them, but hopefully fixes these:

![image](https://github.com/rust-lang/rust-analyzer/assets/308347/462dbfc3-5414-42fd-b81d-a3d699fa9a8d)
2024-02-26 14:08:38 +00:00
bors
40bf8bf757 Auto merge of #16679 - Veykril:block-mod, r=Veykril
fix: Fix modules in blocks not resolving in ide layer

Fixes https://github.com/rust-lang/rust-analyzer/issues/16511
2024-02-26 13:56:01 +00:00
Lukas Wirth
91554e0ae7 fix: Fix modules in blocks not resolving in ide layer 2024-02-26 14:54:47 +01:00
bors
2251d06200 Auto merge of #16673 - szepeviktor:patch-1, r=lnicola
Narrow down typos ignores

- remove ignored files
- check dotfiles
- reorder config sections: files, regexp, words, IDs
- remove non-existent ignores
2024-02-26 13:41:48 +00:00
Viktor Szépe
18801171b2 Narrow down typos ignores 2024-02-26 13:40:01 +00:00
Laurențiu Nicola
1ef7a9971d Remove salsa compile_fail tests 2024-02-26 15:26:55 +02:00
roife
38a50cf1a4 test: callsites inside inline_into_callers 2024-02-26 20:24:44 +08:00
roife
36298c622e fix:do not handle callsites in macros' parameters 2024-02-26 20:23:36 +08:00
bors
8929853df7 Auto merge of #16647 - Young-Flash:fix_replace_filter_map_next_with_find_map, r=Veykril
fix: replace_filter_map_next_with_find_map shouldn't work for dyn trait

close https://github.com/rust-lang/rust-analyzer/issues/16596
2024-02-26 11:31:44 +00:00
Young-Flash
7a58a23f16 internal: format code 2024-02-26 19:10:50 +08:00
Young-Flash
fd0cddf655 internal: move strip_references into dyn trait check 2024-02-26 18:57:46 +08:00
bors
e1983d2b90 Auto merge of #16669 - ShoyuVanilla:merge-borrow-kind-unique, r=Veykril
Merge `BorrowKind::Unique` into `BorrowKind::Mut`

Resolves #15170
2024-02-26 10:57:42 +00:00
bors
6903a5c7d6 Auto merge of #16665 - MabezDev:prio-rustup-when-searching, r=Veykril
Prioritise rustup sysroots over system ones

`get_path_for_executable` will now first check `$CARGO_HOME` before falling back to searching `$PATH`.

`rustup` is the recommended way to manage rust toolchains, therefore should be picked before the system toolchain.

Closes #16661
2024-02-26 10:45:11 +00:00
bors
4585b461e5 Auto merge of #16670 - rikhuijzer:rh/inlay-hints-nvim, r=Veykril
Document nvim 0.10 `inlay_hint`

It took me a few hours to figure out how to enable inlay type hints on Nvim 0.10 with `lspconfig`. `rustaceanvim` has already dropped support for them (https://github.com/mrcjkb/rustaceanvim/discussions/46#discussioncomment-7518669) and most other plugins, like `rust-tools`, are depreciated in favor of `rustaceanvim`.

This PR documents how to enable the inlay hints on Neovim 0.10 and later. I've tested this and changing the `on_attach` function is the only step that is required.
2024-02-26 10:32:30 +00:00
bors
309e087426 Auto merge of #16627 - Young-Flash:issue_template, r=Veykril
update bug report issue template
2024-02-26 10:18:15 +00:00
Rik Huijzer
4060377c07 Add note 2024-02-25 18:05:52 +01:00
Rik Huijzer
ea837d181e Document nvim 0.10 inlay_hint 2024-02-25 17:59:19 +01:00
Shoyu Vanilla
a336bdc142 Merge BorrowKind::Unique into BorrowKind::Mut 2024-02-26 01:57:03 +09:00
bors
5346002d07 Auto merge of #16668 - lnicola:sync-from-rust, r=lnicola
internal: sync from downstream
2024-02-25 16:29:22 +00:00
Laurențiu Nicola
fb31610ae8 Merge branch 'master' into sync-from-rust 2024-02-25 18:27:30 +02:00
Laurențiu Nicola
f206d8b902 Avoid using cfg(FALSE) 2024-02-25 09:58:11 +02:00
Laurențiu Nicola
2edd74be7e Add missing imports 2024-02-25 09:56:19 +02:00
Laurențiu Nicola
9e4ecc60a5 Merge commit '4a8d0f7f565b6df45da5522dd7366a4df3460cd7' into sync-from-ra 2024-02-25 09:45:26 +02:00
Scott Mabin
9f1d4aa4b9 prioritise rustup sysroots over system ones
`get_path_for_executable` will now first check `$CARGO_HOME` before falling back to searching `$PATH`.

rustup is the recommended way to manage rust toolchains, therefore should be picked before the
system toolchain.
2024-02-24 21:19:10 +00:00
bors
4a8d0f7f56 Auto merge of #16654 - graemer957:feature/short_version_flag, r=lnicola
Add short flag -V for consistency with other rust tooling

Minor change to add the `-V` short flag to `rust-analyzer` to bring it in-line with other rust tooling such as:

> rustc -V
```bash
rustc 1.76.0 (07dca489a 2024-02-04)
```

> rustup -V
```bash
rustup 1.26.0 (5af9b9484 2023-04-05)
info: This is the version for the rustup toolchain manager, not the rustc compiler.
info: The currently active `rustc` version is `rustc 1.76.0 (07dca489a 2024-02-04)`
```

> cargo -V
```bash
cargo 1.76.0 (c84b36747 2024-01-18)
```
2024-02-24 09:51:32 +00:00
Graeme Read
30429f8ece
feat: Add short flag -V for consistency with other rust tooling 2024-02-24 06:45:00 +00:00
dfireBird
6f4354f6ad
add assist for filling fields by replacing ellipsis in record syntax 2024-02-24 11:59:48 +05:30
bors
03b3cb6be9 Auto merge of #16652 - davidbarsky:david/deadlock-fix-for-16643, r=Veykril
internal: fix deadlock introduced by #16643

This fixes a deadlock introduced by #16643 ([backtrace](https://gist.github.com/davidbarsky/00f17598f5496a9c41aff31fec1c42d6)). `maybe_changed_after` calls back into other queries, so the cloning here is unfortunately inevitable.

(Zulip conversation: https://rust-lang.zulipchat.com/#narrow/stream/185405-t-compiler.2Frust-analyzer/topic/Fixing.20proc-macro.20dirtying.20with.20unindexed.20projects)
2024-02-23 20:39:13 +00:00
David Barsky
64779737db internal: fix deadlock introduced by #16643 2024-02-23 15:34:23 -05:00
bors
12144045fe Auto merge of #16650 - Veykril:metrics-rustc, r=Veykril
Fix: Fix metrics CI failing
2024-02-23 19:23:45 +00:00
Lukas Wirth
0c3a524acb Fix: Fix metrics CI failing 2024-02-23 20:22:01 +01:00