10995: internal: switch from trait-based TokenSource to simple struct of arrays r=matklad a=matklad
cc #10765
The idea here is to try to simplify the interface as best as we can. The original trait-based approach is a bit over-engineered and hard to debug. Here, we replace callback with just data. The next PR in series will replace the output `TreeSink` trait with data as well.
The biggest drawback here is that we now require to materialize all parser's input up-front. This is a bad fit for macro by example: when you parse `$e:expr`, you might consume only part of the input. However, today's trait-based solution doesn't really help -- we were already materializing the whole thing! So, let's keep it simple!
Co-authored-by: Aleksey Kladov <aleksey.kladov@gmail.com>
10988: Fix expected type calculation in struct literal if followed by comma r=flodiebold a=flodiebold
Co-authored-by: Florian Diebold <flodiebold@gmail.com>
10987: fix: respect inner attributes for Structs and Enums r=lnicola a=rainy-me
fix: #10980 (the allow/deny issue is not fully resolved though.)
Co-authored-by: rainy-me <github@yue.coffee>
10965: minor: Remove dead assists-related function r=Veykril a=lnicola
Unless I'm missing something, `assists_with_fixes` seems to be always used and ends up in the same code path.
Co-authored-by: Laurențiu Nicola <lnicola@dend.ro>
10956: minor: Bump deps r=Veykril a=lnicola
bors r+
10986: fix: Fix lint completions not working for unclosed attributes r=Veykril a=Veykril
Fixes#10682
Uses keywords and nested `TokenTree`s as a heuristic to figure out when to stop parsing in case the attribute is unclosed which should work pretty well as attributes are usually followed by either of those.
bors r+
Co-authored-by: Laurențiu Nicola <lnicola@dend.ro>
Co-authored-by: Lukas Wirth <lukastw97@gmail.com>
10877: feat: make hightlighting linear r=matklad a=matklad
In https://youtu.be/qvIZZf5dmTE, we've noticed that AstIdMap does a
linear lookup when going from SyntaxNode to Id. This leads to
accidentally quadratic overall performance. Replace linear lookup with a
O(1) hashmap lookup.
Future work: don't duplicate `SyntaxNodePtr` in `AstIdMap` and switch to
"call site dependency injection" style storage (eg, store a
`HashSet<ErasedFileAstId>`).
See the explanation of the work here on YouTube :-)
As you can see from then benchmark results, this doesn't actually make analysis stats fastre. I am a bit mystified as to why this is happening to be honest.
Baseline
```
Database loaded: 598.40ms, 304minstr, 118mb (metadata 390.57ms, 21minstr, 841kb; build 111.31ms, 8764kinstr, -214kb)
crates: 39, mods: 824, decls: 18647, fns: 13910
Item Collection: 9.70s, 75ginstr, 377mb
exprs: 382426, ??ty: 387 (0%), ?ty: 285 (0%), !ty: 145
Inference: 43.16s, 342ginstr, 641mb
Total: 52.86s, 417ginstr, 1018mb
```
This PR:
```
Database loaded: 626.34ms, 304minstr, 118mb (metadata 416.26ms, 21minstr, 841kb; build 113.67ms, 8750kinstr, -209kb)
crates: 39, mods: 824, decls: 18647, fns: 13910
Item Collection: 10.16s, 75ginstr, 389mb
exprs: 382426, ??ty: 387 (0%), ?ty: 285 (0%), !ty: 145
Inference: 44.51s, 342ginstr, 644mb
Total: 54.67s, 417ginstr, 1034mb
```
I think we probably should merge the first commit here, but not the second.
Co-authored-by: Aleksey Kladov <aleksey.kladov@gmail.com>
10976: fix: Show case-insensitive exact matches instead of fuzzy flyimport for short paths r=Veykril a=Veykril
And raise the fuzzy trigger from 2 to 3 chars. This way we keep the ability of flyimporting short names like `Rc`.
Fixes https://github.com/rust-analyzer/rust-analyzer/issues/6917
bors r+
Co-authored-by: Lukas Wirth <lukastw97@gmail.com>
10978: internal: Only prime direct dependencies of the workspace crates r=jonas-schievink a=lnicola
This reduces the number of primed crates on `self` from 177 to 105.
```text
# baseline
86521 me 20 0 2939M 562M 14320 S 0.0 0.9 0:07.10 37 rust-analyzer
# pr
99947 me 20 0 2935M 549M 14424 S 0.0 0.9 0:07.10 37 rust-analyzer
```
The RAM usage seems to vary a lot, so I'm not sure the improvement shown above is real.
We also drop the topological sorting, although it might still make sense to do the dependencies before the workspace crates. But since it's only used to make the progress reporting a bit more fine, it probably doesn't matter too much.
Co-authored-by: Laurențiu Nicola <lnicola@dend.ro>
