This appears to have been introduced ages ago in
be742a5877
but has since been removed.
As it stands, it is problematic if multiple instances of the
rust-analyzer LSP are launched during the same VS Code session because
VS Code complains about multiple LSP servers trying to register the
same command.
Most LSP servers workaround this by parameterizing the command by the
process id. For example, this is where `rls` does this:
ff0b9057c8/rls/src/server/mod.rs (L413-L421)
Though `apply_code_action` does not seems to be used, so it seems better
to delete it than to parameterize it.
This wasn't a right decision in the first place, the feature flag was
broken in the last rustfmt release, and syntax highlighting of imports
is more important anyway
My workflow in Visual Studio Code + Rust Analyzer has become:
1. Make a change to Rust source code using all the analysis magic
2. Save the file to trigger `cargo watch`. I have format on save enabled
for all file types so this also runs `rustfmt`
3. Fix any diagnostics that `cargo watch` finds
Unfortunately if the Rust source has any syntax errors the act of saving
will pop up a scary "command has failed" message and will switch to the
"Output" tab to show the `rustfmt` error and exit code.
I did a quick survey of what other Language Servers do in this case.
Both the JSON and TypeScript servers will swallow the error and return
success. This is consistent with how I remember my workflow in those
languages. The syntax error will show up as a diagnostic so it should
be clear why the file isn't formatting.
I checked the `rustfmt` source code and while it does distinguish "parse
errors" from "operational errors" internally they both result in exit
status of 1. However, more catastrophic errors (missing `rustfmt`,
SIGSEGV, etc) will return 127+ error codes which we can distinguish from
a normal failure.
This changes our handler to log an info message and feign success if
`rustfmt` exits with status 1.
Another option I considered was only swallowing the error if the
formatting request came from format-on-save. However, the Language
Server Protocol doesn't seem to distinguish those cases.
The issue was windows specific -- cancellation caused collection of
bracktraces at some point, and that was slow on windows.
The proper fix here is to make sure that we don't collect bracktraces
unnecessary (which we currently do due to failure), but, as a
temporary fix, let's just not force their collection in the first
place!
Very simple approach: For each identifier, set the hash of the range
where it's defined as its 'id' and use it in the VSCode extension to
generate unique colors.
Thus, the generated colors are per-file. They are also quite fragile,
and I'm not entirely sure why. Looks like we need to make sure the
same ranges aren't overwritten by a later request?
`ProjectWorkspace::to_roots` now returns a new `ProjectRoot` which contains
information regarding whether or not the given path is part of the current
workspace or an external dependency. This information can then be used in
`ra_batch` and `ra_lsp_server` to implement more advanced filtering. This allows
us to filter some unnecessary folders from external dependencies such as tests,
examples and benches.
If the client doesn't specify this explicitly, that very likely means it doesn't
know about it and so we shouldn't send decorations. In particular, the recent
change to this default caused decorations to be sent to emacs, resulting in a
lot of warning spam.
This allows users to control whether or not they want to see the "workspace
loaded" notification.
This is done on the server side using InitializationOptions which are provided
by the client. By default show_workspace_loaded is true, meaning the
notification is sent.
This now allows us to send a notification that can be shown in the UI when the
workspace has been loaded.
Additionally this removes the need for internal_mode flag.
901: Add basic support for showing fn signature when hovering r=matklad a=vipentti
This adds basic support for displaying function signature when hovering over a usage of a function.
Additionally refactored `hover` to return `HoverResult` to ease with testing and in general to be more robust.
Co-authored-by: Ville Penttinen <villem.penttinen@gmail.com>
vscode would report "A request has failed" when it got "Content modified"
message and this would cause a pop-up to appear. This works around the issue by
returning an "empty" response that vscode can ignore.
742: Extern crate r=matklad a=flodiebold
This implements `extern crate` declarations by lowering them to (absolute) imports, and adds support for absolute paths. It also extracts the extern prelude from the per-module item map, and handles the special case of extern crates in the crate root adding to the extern prelude.
This means we finally resolve `Arc`, so it fixes#523😄
Co-authored-by: Florian Diebold <flodiebold@gmail.com>
702: Go to Implementation r=matklad a=kjeremy
First half of #620
Co-authored-by: Jeremy Kolb <kjeremy@gmail.com>
Co-authored-by: kjeremy <kjeremy@gmail.com>
574: refactor completions to use TextEdit instead of InsertText r=matklad a=gfreezy
1. migrate from `insertText` to `TextEdit` from `CompleteItem`
2. use `insta` to test completions
Co-authored-by: gfreezy <gfreezy@gmail.com>
517: gracefully handle cycles in crate graph r=matklad a=matklad
518: Add an explanatory message when we use the Query fallback r=matklad a=DJMcNab
Fixes https://github.com/rust-analyzer/rust-analyzer/issues/269.
There is no good way to explain it for go_to_def, so I've just fallen back on to_vec.
Co-authored-by: Aleksey Kladov <aleksey.kladov@gmail.com>
Co-authored-by: DJMcNab <36049421+djmcnab@users.noreply.github.com>
505: Inherent methods r=matklad a=flodiebold
This adds resolution, type checking and completion for inherent methods.
The main open question here is the caching, I think. I'm not sure whether we should be caching method resolutions in a more fine grained way (currently we just build a hash map of types -> impl blocks, and iterate through all potential impl blocks when looking for a method).
Co-authored-by: Florian Diebold <flodiebold@gmail.com>
500: Code lens support for running tests r=matklad a=kjeremy
Supports running individual and mod tests.
I feel like this kind of abuses the `Runnables` infrastructure but it works. Maybe later on down the line we should introduce a struct that is really just a tuple of binary, arguments, and environment and pass that back to the client instead. `run_single.ts` is just a paired down version of `runnables.ts` and there is duplication because I think run_single will probably change independent of runnables.
Co-authored-by: Jeremy A. Kolb <jkolb@ara.com>
Co-authored-by: Jeremy Kolb <kjeremy@gmail.com>
If we index gazillion libraries simultaneously, we fill the threadpool
and so the main loop fails to turn, although there isn't really any
significant blocking inside the loop itself.
