Exclude non-identifier aliases from completion filtering text
When building `CompletionItem`s, this excludes aliases that aren't valid identifiers from the "lookup" text used to filter completions in the LSP client. Including them results in weird completion filtering behavior e.g. `Partial>` matching a completion for the `PartialOrd` trait because it has a doc alias of ">".
Closes#14692
Add ExternCrateDecl to HIR
Adding these doesn't really require much design effort as they represent a single import, unlike use trees which are one item that represent 0 or more imports.
We only resolve to this definition when actually resolving on the name or alias of an `extern crate name as alias` item, not usages yet as that requires far more changes that won't lead anywhere without giving it more thought. Nevertheless the changes slightly improve IDE things, an example being hover on the decl showing the merged doc comments for example.
cc https://github.com/rust-lang/rust-analyzer/issues/14079
Filter out short-lived LLVM diagnostics before they reach the rustc handler
During profiling I saw remark passes being unconditionally enabled: for example `Machine Optimization Remark Emitter`.
The diagnostic remarks enabled by default are [from missed optimizations and opt analyses](https://github.com/rust-lang/rust/pull/113339#discussion_r1259480303). They are created by LLVM, passed to the diagnostic handler on the C++ side, emitted to rust, where they are unpacked, C++ strings are converted to rust, etc.
Then they are discarded in the vast majority of the time (i.e. unless some kind of `-Cremark` has enabled some of these passes' output to be printed).
These unneeded allocations are very short-lived, basically only lasting between the LLVM pass emitting them and the rust handler where they are discarded. So it doesn't hugely impact max-rss, and is only a slight reduction in instruction count (cachegrind reports a reduction between 0.3% and 0.5%) _on linux_. It's possible that targets without `jemalloc` or with a worse allocator, may optimize these less.
It is however significant in the aggregate, looking at the total number of allocated bytes:
- it's the biggest source of allocations according to dhat, on the benchmarks I've tried e.g. `syn` or `cargo`
- allocations on `syn` are reduced by 440MB, 17% (from 2440722647 bytes total, to 2030461328 bytes)
- allocations on `cargo` are reduced by 6.6GB, 19% (from 35371886402 bytes total, to 28723987743 bytes)
Some of these diagnostics objects [are allocated in LLVM](https://github.com/rust-lang/rust/pull/113339#discussion_r1252387484) *before* they're emitted to our diagnostic handler, where they'll be filtered out. So we could remove those in the future, but that will require changing a few LLVM call-sites upstream, so I left a FIXME.
Added remove unused imports assist
This resolves the most important part of #5131. I needed to make a couple of cosmetic changes to the search infrastructure to do this.
A few open questions:
* Should imports that don't resolve to anything be considered unused? I figured probably not, but it would be a trivial change to make if we want it.
* Is there a cleaner way to make the edits to the use list?
* Is there a cleaner way to get the list of uses that intersect the current selection?
* Is the performance acceptable? When testing this on itself, it takes a good couple seconds to perform the assist.
* Is there a way to hide the rustc diagnostics that overlap with this functionality?
internal: Defer structured snippet rendering to allow escaping snippet bits
Since we know exactly where snippets are, we can transparently escape snippet bits to the exact text edits that need it, and not have to do it for anything other text edits.
Also will eventually fix#11006 once all assists are migrated. This comes as a side-effect of text edits that don't have snippets get marked as having no insert formatting at all.
Don't provide `add_missing_match_arms` assist when upmapping match arm list failed
Fixes#15310
We shouldn't provide the assist when we fail to find the original match arm list.
Note that this PR will temporarily make the assist not applicable when attribute macro operates on the match expression in question, just like the case in #15310, for most of the current stable toolchain users. This is because the sysroot-abi proc-macro-srv on the current stable [discards] spans for `Group` delimiters in some code paths, which the popular `proc-macro2` crate almost always calls, and it makes the identity of match arm list's brackets lost, leading to the upmapping failure. This has been fixed by #14960, which will land in the next stable, 1.71.
[discards]: 8ede3aae28/src/tools/rust-analyzer/crates/proc-macro-srv/src/abis/abi_sysroot/ra_server.rs (L231)
bugfix : skip doc(hidden) default members
fixes #14957 . I have two questions :
1. I am definitely looking for a more idiomatic way for the things I added in `crates/ide-assists/src/utils.rs`. See `FIXME` in that file.
2. Would it be actually better to change `DefaultMethods` to something like
```rust
enum DefaultMethods {
Only( IgnoreHidden ( bool ) ) ,
None
}
```
instead of adding a boolean to every function that calls `crates/ide-assists/src/utils.rs::filter_assoc_items`
fix: Expand eager macros to delimited comma separated expression list
Prior to this, we were just parsing it as an expression which works fine for `()` and `[]` calls as those are tuple and array expressions respectively, but if tails for `{}` calls which with my recent changes reported errors for such eager macro invocations.
Fixup path fragments upon MBE transcription
Fixes#14367
There are roughly two types of paths: paths in expression context, where a separator `::` between an identifier and its following generic argument list is mandatory, and paths in type context, where `::` can be omitted.
Unlike rustc, we need to transform the parsed fragments back into tokens during transcription. When the matched path fragment is a type-context path and is transcribed as an expression-context path, verbatim transcription would cause a syntax error.
This PR fixes up path fragments by inserting `::` to make sure they are syntactically correct in all contexts. Note that this works because expression-context paths are a strict superset of type-context paths.
Remove redundant example of `BTreeSet::iter`
The usage and that `Values returned by the iterator are returned in ascending order` are already demonstrated by the other example and the description, so I removed the useless one.
