- prevent `pat` from matching before `path` in metavariable types
- reduce the precedence of math operators so that assignment operators match correctly
- remove comment scope from ignored params
- underscores will automatically receive variable/param scope
- add raw ID syntax to modules, functions, and variables
6251: Semantic Highlight: Add Callable modifier for variables r=matklad a=GrayJack
This PR added the `HighlightModifier::Callable` variant and assigned it to variables and parameters that are fn pointers, closures and implements FnOnce trait.
This allows to colorize these variables/parameters when used in call expression.
6310: Rewrite algo::diff to support insertion and deletion r=matklad a=Veykril
This in turn also makes `algo::diff` generate finer diffs(maybe even minimal diffs?) as insertions and deletions aren't always represented as as replacements of parent nodes now.
Required for #6287 to go on.
Co-authored-by: GrayJack <gr41.j4ck@gmail.com>
Co-authored-by: Lukas Wirth <lukastw97@gmail.com>
6274: Check cargoExtraArgs for undefined before using r=matklad a=feileacan
Fixes#6273
Co-authored-by: feileacan <54381366+feileacan@users.noreply.github.com>
6137: add a new TextMate grammar r=matklad a=dustypomerleau
Thanks to everyone working hard on Rust Analyzer - my impression is that it's quickly becoming the community default.
I think it would be helpful to have a more robust TextMate grammar to fall back on, for those who wish to disable semantic highlighting for any reason. It should allow theming of punctuation, and provide scopes for all tokens on the page. This can be done at zero cost to those who enable semantic highlighting, as the TextMate scopes will be invisible to those users.
I can see a couple ways of accomplishing this:
1. Ship a new grammar by merging this PR.
1. Ship no TextMate grammar at all (like the [Rust](https://marketplace.visualstudio.com/items?itemName=rust-lang.rust) extension), and allow users to install a separate extension that provides the grammar of their choice (I have released this one as [Rust Syntax](https://marketplace.visualstudio.com/items?itemName=dustypomerleau.rust-syntax)). If no grammar were installed, they would simply fall back to the default grammar provided by their editor. In the case of VS Code, the default grammar already matches what is currently being shipped, so users who choose not to override it would see no difference.
I have tried to choose sensible default scopes, in the hopes that a wider variety of themes would work out of the box with Rust, even if those themes do not yet supply scopes for semantic highlighting. There is definitely some interest in using this grammar with Rust Analyzer, as this was the very first issue after the syntax extension was shipped: https://github.com/dustypomerleau/rust-syntax/issues/1.
I considered simply using an alternative grammar alongside Rust Analyzer, but this doesn't seem possible. When RA starts, any existing grammar/extension is overridden, and I haven't been able to find a workaround.
Co-authored-by: Dusty Pomerleau <dustypomerleau@users.noreply.github.com>
Currently a method only has defaultness if it is a provided trait
method, but this will change when specialisation is available and may
need to become a concept known to hir.
I opted to go for a 'fewest changes' approach given specialisation is
still under development.
* Chalk very recently (like an hour ago) merged a fix that prevents rust analyzer from panicking. This allows it to be usable again for code that hits those situations. See #6134, #6145, Probably #6120
The previous version would have interpreted an empty token as
an abort of the dialog and would have not properly cleared the token.
This is now fixed by checking for `undefined` for a an abort and
by setting the token to `undefined` in order to clear it.
This change allows to use a authorization token provided by Github in
order to fetch metadata for a RA release. Using an authorization token
prevents to get rate-limited in environments where lots of RA users use
a shared client IP (e.g. behind a company NAT).
The auth token is stored in `ExtensionContext.globalState`.
As far as I could observe through testing with a local WSL2 environment
that state is synced between an extension installed locally and a remote
version.
The change provides no explicit command to query for an auth token.
However in case a download fails it will provide a retry option as well
as an option to enter the auth token. This should be more discoverable
for most users.
Closes#3688
5985: Make MergeBehaviour configurable r=jonas-schievink a=Veykril
This should make the newly implemented `MergeBehaviour` for import insertion configurable as roughly outlined in https://github.com/rust-analyzer/rust-analyzer/pull/5935#issuecomment-685834257. For the config name and the like I just picked what came to mind so that might be up for bikeshedding.
Co-authored-by: Lukas Wirth <lukastw97@gmail.com>
5910: Fix some typos r=matklad a=SomeoneToIgnore
5912: Remove fixme from inlay_hints.ts r=matklad a=Veetaha
I have reevaluated the fixme and it doesn't seem necessary to pass an array of files
to the inlay hints request.
This will (a) make the request more compilcated (b), make us wait for
inlay hints for `all` active editors resolve at once before rendering and (c)
doesn't seem required because 99% of the time there is a single active editor
in the IDE
Co-authored-by: Kirill Bulatov <mail4score@gmail.com>
Co-authored-by: Veetaha <veetaha2@gmail.com>
I have reevaluated the fixme and it doesn't seem necessary to pass an array of files
to the inlay hints request.
This will (a) make the request more compilcated (b), make us wait for
inlay hints for `all` active editors resolve at once before rendering and (c)
doesn't seem required because 99% of the time there is a single active editor
in the IDE
5682: Add an option to disable diagnostics r=matklad a=popzxc
As far as I know, currently it's not possible to disable a selected type of diagnostics provided by `rust-analyzer`.
This causes an inconvenient situation with a false-positive warnings: you either have to disable all the diagnostics, or you have to ignore these warnings.
There are some open issues related to this problem, e.g.: https://github.com/rust-analyzer/rust-analyzer/issues/5412, https://github.com/rust-analyzer/rust-analyzer/issues/5502
This PR attempts to make it possible to selectively disable some diagnostics on per-project basis.
Co-authored-by: Igor Aleksanov <popzxc@yandex.ru>
5776: Fix eslint errors on .eslintrc.js and rollup.config.js r=matklad a=fuafa
Eslint complains if these two files does not include in the `tsconfig.json`.
```
Parsing error: "parserOptions.project" has been set for @typescript-eslint/parser.
The file does not match your project config: .eslintrc.js.
The file must be included in at least one of the projects provided.eslint
```
5780: Fixup whitespace when adding missing impl items r=matklad a=jDomantas
Generate properly formatted whitespace when adding impl items - with an empty line between items and removing extra whitespace that often appears at the end.
This is my first time working on rust analyzer so I'm not very familiar with its internal APIs. If there's a better way to do such syntax tree editing I'd be glad to hear it.
Co-authored-by: xiaofa <xiaofalzx@gmail.com>
Co-authored-by: jDomantas <djadenkus@gmail.com>
No we return ContentModified during the workspace loading. This signifies the language
client to retry the operation (i.e. the client will
continue polling the server while it returns ContentModified).
I believe that there might be cases of overly big projects where the backoff
logic we have setup in `sendRequestWithRetry` (which we use for inlay hints)
might bail too early (currently the largest retry standby time is 10 seconds).
However, I've tried on one of my project with 500+ dependencies and it is still enough.
5513: Try figure out correct workspace in vscode multi root workspace r=vsrs a=urbandove
the code to replace the root with the `${workspaceRoot}` arg breaks in multi root workspaces as it needs a qualifier `${workspaceRoot:workspaceName}`
This PR attempts to figure out the root workspace - and if it cant find it falls back to the first workspace
Co-authored-by: Urban Dove <urbandove80@gmail.com>
5480: Fix snippetTextEdits applying to other files r=matklad a=TimoFreiberg
Fixes#4551
`vscode.window.visibleTextEditors` only contains editors whose contents are being displayed at the moment, so the previous logic only worked if the other file for which a snippetTextEdit is being received was visible in a separate split.
I feel that this is a hacky approach, so feel free to reject it for something nicer :)
Co-authored-by: Timo Freiberg <timo.freiberg@gmail.com>
vscode.window.visibleTextEditors only contains editors whose contents
are being displayed at the moment, so the previous logic only worked if
the other file for which a snippetTextEdit is being received was visible
in a separate split.
Co-authored-by: bjorn3 <bjorn3@users.noreply.github.com>
Override miniz_oxide to build it with optimizations
Building this crate with optimizations decreases the gzipping
part of `cargo xtask dist` from `30-40s` down to `3s`,
the overhead for `rustc` to apply optimizations is miserable on this background
5202: Runnable env r=matklad a=vsrs
This PR adds on option to specify (in the settings.json) environment variables passed to the runnable.
The simplest way for all runnables in a bunch:
```jsonc
"rust-analyzer.runnableEnv": {
"RUN_SLOW_TESTS": "1"
}
```
Or it is possible to specify vars more granularly:
```jsonc
"rust-analyzer.runnableEnv": [
{
// "mask": null, // null mask means that this rule will be applied for all runnables
env: {
"APP_ID": "1",
"APP_DATA": "asdf"
}
},
{
"mask": "test_name",
"env": {
"APP_ID": "2", // overwrites only APP_ID
}
}
]
```
You can use any valid RegExp as a mask. Also note that a full runnable name is something like *run bin_or_example_name*, *test some::mod::test_name* or *test-mod some::mod*, so it is possible to distinguish binaries, single tests, and test modules with this masks: `"^run"`, `"^test "` (the trailing space matters!), and `"^test-mod"` respectively.
Fixes#4450
I suppose this info should be somewhere in the docs, but unsure where is the best place.
Co-authored-by: vsrs <vit@conrlab.com>
5089: Disable auto-complete on comments r=matklad a=BGluth
Resolves#4907 by disabling any auto-completion on comments.
As flodiebold [pointed out](https://github.com/rust-analyzer/rust-analyzer/issues/4907#issuecomment-648439979), in the future we may want to support some form of auto-completion within doc comments, but for now it was suggested to just disable auto-completion on them entirely.
The implementation involves adding a new field `is_comment` to `CompletionContext` and checking if the immediate token we auto-completed on is a comment. I couldn't see a case where we need to check any of the ancestors, but let me know if this is not sufficient. I also wasn't sure if it was necessary to add a new field to this struct, but I decided it's probably the best option if we want to potentially do auto-completion on doc comments in the future.
Finally, the three tests I added should I think ideally not filter results by `CompletionKind::Keyword`, but if I want to get unfiltered results, I need access to a non-public function [get_all_completion_items](9a4d02faf9/crates/ra_ide/src/completion/test_utils.rs (L32-L39)) which I don't know if I should make public just for this.
5161: SSR: Add initial support for placeholder constraints r=matklad a=davidlattimore
5184: Always install required nightly extension if current one is not nightly r=matklad a=Veetaha
This is weird, but having switched back to stable by uninstalling the extension appears that vscode doesn't destroy the `PersistentState` and thus changing to `nightly` channel doesn't work because the last check for nightly extension was less than 1 hour ago. The simple solution is to skip this check if we know that the current extension version is not nightly.
5185: Force showing extension activation error pop-up notification r=matklad a=Veetaha
Fixes https://github.com/rust-analyzer/rust-analyzer/issues/5091
5186: fix: correct pd/ppd/tfn/tmod completion doc r=matklad a=fannheyward
a33eefa3b2/crates/ra_ide/src/completion/complete_snippet.rs (L23-L24)
Co-authored-by: BGluth <gluthb@gmail.com>
Co-authored-by: David Lattimore <dml@google.com>
Co-authored-by: Veetaha <veetaha2@gmail.com>
Co-authored-by: Heyward Fann <fannheyward@gmail.com>
5101: Add expect -- a light-weight alternative to insta r=matklad a=matklad
This PR implements a small snapshot-testing library. Snapshot updating is done by setting an env var, or by using editor feature (which runs a test with env-var set).
Here's workflow for updating a failing test:
Here's workflow for adding a new test:
Note that colorized diffs are not implemented in this PR, but should be easy to add (we already use them in test_utils).
Main differences from insta (which is essential for rust-analyzer development, thanks @mitsuhiko!):
* self-updating tests, no need for a separate tool
* fewer features (only inline snapshots, no redactions)
* fewer deps (no yaml, no persistence)
* tighter integration with editor
* first-class snapshot object, which can be used to write test functions (as opposed to testing macros)
* trivial to tweak for rust-analyzer needs, by virtue of being a workspace member.
I think eventually we should converge to a single snapshot testing library, but I am not sure that `expect` is exactly right, so I suggest rolling with both insta and expect for some time (if folks agree that expect might be better in the first place!).
# Editor Integration Implementation
The thing I am most excited about is the ability to update a specific snapshot from the editor. I want this to be available to other snapshot-testing libraries (cc @mitsuhiko, @aaronabramov), so I want to document how this works.
The ideal UI here would be a code action (💡). Unfortunately, it seems like it is impossible to implement without some kind of persistence (if you save test failures into some kind of a database, like insta does, than you can read the database from the editor plugin). Note that it is possible to highlight error by outputing error message in rustc's format. Unfortunately, one can't use the same trick to implement a quick fix.
For this reason, expect makes use of another rust-analyzer feature -- ability to run a single test at the cursor position. This does need some expect-specific code in rust-analyzer unfortunately. Specifically, if rust-analyzer notices that the cursor is on `expect!` macro, it adds a special flag to runnable's JSON. However, given #5017 it is possible to approximate this well-enough without rust-analyzer integration. Specifically, an extension can register a special runner which checks (using regexes) if rust-anlyzer runnable covers text with specific macro invocation and do special magic in that case.
closes#3835
Co-authored-by: Aleksey Kladov <aleksey.kladov@gmail.com>
5119: Show notification while SSR is in progress r=matklad a=davidlattimore
Ideally we would (a) show progress and (b) allow cancellation, but at least now there's some indication to the user that something is happening.
Co-authored-by: David Lattimore <dml@google.com>
5025: Don't mess with messy temp dir and just download into extension dir r=matklad a=Veetaha
Temp dirs are messy. Dealing with them requires handling quite a bunch of
edge cases. As proposed by lnicola this seems better to just put the temp files
in the extension dir and not care much about suddenly leaving garbage.
Instead we get shorter and less platform-caveat-y code.
We will also assume users don't try to issue a download in different vscode windows simultaneously.
Fixes#5019
Co-authored-by: Veetaha <veetaha2@gmail.com>
Temp dirs are messy. Dealing with them requires handling quite a bunch of
edge cases. As proposed by lnicola this seems better to just put the temp files
in the extension dir and not care much about suddenly leaving garbage.
Instead we get shorter and less platform-caveat-y code.
We will also assume users don't try to issue a download in different vscode windows simultaneously
4992: Never disable error logging on the frontend r=matklad a=Veetaha
4993: Make bootstrap error message more informative and better-fitting r=matklad a=Veetaha
Now this better fits standard vscode extension activation failure message and suggests enabling verbose logs.
4994: Decouple http file stream logic from temp dir logic r=matklad a=Veetaha
Followup for #4989
4997: Update manual.adoc r=matklad a=gwutz
GNOME Builder (Nightly) supports now rust-analyzer
4998: Disrecommend trace.server: "verbose" for regular users r=matklad a=Veetaha
This option has never been useful for me, I wonder if anyone finds regular users can use this for sending logs
Co-authored-by: Veetaha <veetaha2@gmail.com>
Co-authored-by: Günther Wagner <info@gunibert.de>
4580: Fix invoking cargo without consulting CARGO env var or standard installation paths r=matklad a=Veetaha
Followup for #4329
The pr essentially fixes [this bug](https://youtu.be/EzQ7YIIo1rY?t=2189)
cc @lefticus
Co-authored-by: veetaha <veetaha2@gmail.com>
This is useful when an extension (e.g. Nix Environment Selector) or launch
configuration sets one or more environment variables.
When `env` is not explicitly specified in the options passed to
`child_process.spawn()` or `vscode.ShellExecution()`, then `process.env` gets
applied automatically. But when an explicit `env` is set, it should inherit from
`process.env` rather than replace it completely.
As per matklad, we now pass the responsibility for finding the binary to the frontend.
Also, added caching for finding the binary path to reduce
the amount of filesystem interactions.
4625: Partially fix displaying inlay hints in Github PR diff views r=matklad a=Veetaha
See the comment in https://github.com/rust-analyzer/rust-analyzer/issues/4608#issuecomment-63424257
It partially fixes the left side of diff view (the one where old code is displayed), but the diff editor with new code changes still has `file` scheme and will proceed displaying inlay hints...
4629: Fix the `should_panic` snippet r=matklad a=eminence
Closes#4628
Co-authored-by: veetaha <veetaha2@gmail.com>
Co-authored-by: Andrew Chin <achin@eminence32.net>
4602: Add boolean literal semantic token type to package.json r=matklad a=lnicola
Closes#4583.
CC @GrayJack
4603: Add self keyword semantic token type r=matklad a=lnicola
Not sure if this is warranted a new token type or just a modifier.
---
CC #4583, @GrayJack
Co-authored-by: Laurențiu Nicola <lnicola@dend.ro>
4555: VSCode: added patchelf after download for NixOS support r=matklad a=cab404
This adds Nix support, and fixes#4542
4575: Use Chalk's built-in representations for fn items and pointers r=matklad a=flodiebold
The `TypeName::FnDef` was just added; the function pointer variant has existed for a while, I just forgot about it because it's special (because fn pointers can be higher-ranked over lifetimes).
We *could* also make `FnPtr` a separate `Ty` variant instead of a `TypeCtor` variant, which would make the conversion code a bit less special-casey, but it doesn't seem worth doing right now.
Co-authored-by: Vladimir Serov <me@cab404.ru>
Co-authored-by: Cabia Rangris <me@cab404.ru>
Co-authored-by: Florian Diebold <florian.diebold@freiheit.com>
4552: Transition OnEnter to WorkspaceSnippetEdit r=matklad a=matklad
This also changes our handiling of snippet edits on the client side.
`editor.insertSnippet` unfortunately forces indentation, which we
really don't want to have to deal with. So, let's just implement our
manual hacky way of dealing with a simple subset of snippets we
actually use in rust-analyzer
bors r+
🤖
Co-authored-by: Aleksey Kladov <aleksey.kladov@gmail.com>
This also changes our handiling of snippet edits on the client side.
`editor.insertSnippet` unfortunately forces indentation, which we
really don't want to have to deal with. So, let's just implement our
manual hacky way of dealing with a simple subset of snippets we
actually use in rust-analyzer
4499: CodeLens configuration options r=vsrs a=vsrs
This PR
- adds an option to granularly enable\disable all CodeLens, just like the TypeScript extension.
- fixes a minor bug for doctests. It makes no sense to show `Debug` lens for them as cargo `Can't skip running doc tests with --no-run`.
Co-authored-by: vsrs <vit@conrlab.com>
4448: Generate configuration for launch.json r=vsrs a=vsrs
This PR adds two new commands: `"rust-analyzer.debug"` and `"rust-analyzer.newDebugConfig"`. The former is a supplement to the existing `"rust-analyzer.run"` command and works the same way: asks for a runnable and starts new debug session. The latter allows adding a new configuration to **launch.json** (or to update an existing one).
If the new option `"rust-analyzer.debug.useLaunchJson"` is set to true then `"rust-analyzer.debug"` and Debug Lens will first look for existing debug configuration in **launch.json**. That is, it has become possible to specify startup arguments, env variables, etc.
`"rust-analyzer.debug.useLaunchJson"` is false by default, but it might be worth making true the default value. Personally I prefer true, but I'm not sure if it is good for all value.
----
I think that this PR also solves https://github.com/rust-analyzer/rust-analyzer/issues/3441.
Both methods to update launch.json mentioned in the issue do not work:
1. Menu. It is only possible to add a launch.json configuration template via a debug adapter. And anyway it's only a template and it is impossible to specify arguments from an extension.
2. DebugConfigurationProvider. The exact opposite situation: it is possible to specify all debug session settings, but it is impossible to export these settings to launch.json.
Separate `"rust-analyzer.newDebugConfig"` command looks better for me.
----
Fixes#4450Fixes#3441
Co-authored-by: vsrs <vit@conrlab.com>
Co-authored-by: vsrs <62505555+vsrs@users.noreply.github.com>
4083: Smol documentation for ast nodes r=matklad a=Veetaha
There is a tremendous amount of TODOs to clarify the topics I am not certain about.
Please @matklad, @edwin0cheng review carefully, I even left some mentions of your names in todos to put your attention where you most probably can give comments.
In order to simplify the review, I separated the codegen (i.e. changes in `ast/generated/nodes.rs`) from `ast_src` changes (they in fact just duplicate one another) into two commits.
Also, I had to hack a little bit to let the docs be generated as doc comments and not as doc attributes because it's easier to read them this way and IIRC we don't support hints for `#[doc = ""]` attributes for now...
Closes#3682
Co-authored-by: veetaha <veetaha2@gmail.com>
4329: Look for `cargo`, `rustc`, and `rustup` in standard installation path r=matklad a=cdisselkoen
Discussed in #3118. This is approximately a 90% fix for the issue described there.
This PR creates a new crate `ra_env` with a function `get_path_for_executable()`; see docs there. `get_path_for_executable()` improves and generalizes the function `cargo_binary()` which was previously duplicated in the `ra_project_model` and `ra_flycheck` crates. (Both of those crates now depend on the new `ra_env` crate.) The new function checks (e.g.) `$CARGO` and `$PATH`, but also falls back on `~/.cargo/bin` manually before erroring out. This should allow most users to not have to worry about setting the `$CARGO` or `$PATH` variables for VSCode, which can be difficult e.g. on macOS as discussed in #3118.
I've attempted to replace all calls to `cargo`, `rustc`, and `rustup` in rust-analyzer with appropriate invocations of `get_path_for_executable()`; I don't think I've missed any in Rust code, but there is at least one invocation in TypeScript code which I haven't fixed. (I'm not sure whether it's affected by the same problem or not.) a4778ddb7a/editors/code/src/cargo.ts (L79)
I'm sure this PR could be improved a bunch, so I'm happy to take feedback/suggestions on how to solve this problem better, or just bikeshedding variable/function/crate names etc.
cc @Veetaha
Fixes#3118.
Co-authored-by: Craig Disselkoen <craigdissel@gmail.com>
Co-authored-by: veetaha <veetaha2@gmail.com>
4166: Defining a default target to support cross-compilation targets r=matklad a=FuriouZz
Related to #4163
Co-authored-by: Christophe MASSOLIN <christophe.massolin@gmail.com>
The `workspaceLoaded` notification setting was originally designed to
control the display of a popup message that said:
"workspace loaded, {} rust packages"
This popup was removed and replaced by a much sleeker message in the
VSCode status bar that provides a real-time status while loading:
rust-analyzer: {}/{} packages
This was done as part of #3587
The new status-bar indicator is unobtrusive and shouldn't need to be
disabled. So this setting is removed.
4222: Introduce C/C++ for Visual Studio Code extension as an alternative debug engine for Debug Code lens. r=matklad a=vsrs
At the moment Debug Code Lens can use only one debug engine: lldb via [CodeLLDB](https://marketplace.visualstudio.com/items?itemName=vadimcn.vscode-lldb) extension.
This PR adds support of the debug engine from the [MS C++ tools](https://marketplace.visualstudio.com/items?itemName=ms-vscode.cpptools) extension, as well as the configuration option. If both extensions are installed, `CodeLLDB` will be used by default.
Another new option `rust-analyzer.debug.sourceFileMap` allows, for example, to step into Rust std library during debugging. Works only with `MS C++ tools`.
On Windows:
```json
"rust-analyzer.debug.sourceFileMap": {
"/rustc/4fb7144ed159f94491249e86d5bbd033b5d60550": "${env:USERPROFILE}/.rustup/toolchains/stable-x86_64-pc-windows-msvc/lib/rustlib/src/rust"
}
```
On Linux:
```json
"rust-analyzer.debug.sourceFileMap": {
"/rustc/4fb7144ed159f94491249e86d5bbd033b5d60550": "~/.rustup/toolchains/stable-x86_64-unknown-linux-gnu/lib/rustlib/src/rust"
}
```
Co-authored-by: vsrs <vit@conrlab.com>
3998: Make add_function generate functions in other modules via qualified path r=matklad a=TimoFreiberg
Additional feature for #3639
- [x] Add tests for paths with more segments
- [x] Make generating the function in another file work
- [x] Add `pub` or `pub(crate)` to the generated function if it's generated in a different module
- [x] Make the assist jump to the edited file
- [x] Enable file support in the `check_assist` helper
4006: Syntax highlighting for format strings r=matklad a=ltentrup
I have an implementation for syntax highlighting for format string modifiers `{}`.
The first commit refactors the changes in #3826 into a separate struct.
The second commit implements the highlighting: first we check in a macro call whether the macro is a format macro from `std`. In this case, we remember the format string node. If we encounter this node during syntax highlighting, we check for the format modifiers `{}` using regular expressions.
There are a few places which I am not quite sure:
- Is the way I extract the macro names correct?
- Is the `HighlightTag::Attribute` suitable for highlighting the `{}`?
Let me know what you think, any feedback is welcome!
Co-authored-by: Timo Freiberg <timo.freiberg@gmail.com>
Co-authored-by: Leander Tentrup <leander.tentrup@gmail.com>
Co-authored-by: Leander Tentrup <ltentrup@users.noreply.github.com>
Move the task provider anonymous class into a real class, as this seems
to be how Microsoft do this in their documentation.
resolveTask is now implemented, which is used by VSCode to determine how
to execute tasks that the user has defined in tasks.json.
This is a quick way to implement unresolved reference diagnostics.
For example, adding to VS Code config
"editor.tokenColorCustomizationsExperimental": {
"unresolvedReference": "#FF0000"
},
will highlight all unresolved refs in red.
This is covered under vscode's "editor.semanticHighlighting.enabled"
setting plus the user has to have a theme that has opted into highlighting.
Bumps required vscode stable to 1.44
In textmate, keyword.control is used for all kinds of things; in fact,
the default scope mapping for keyword is keyword.control!
So let's add a less ambiguous controlFlow modifier
See Microsoft/vscode#94367
3844: vscode: restore removed default values r=matklad a=Veetaha
After refactoring the config we forgot to set defaults for
some properties like workspaceLoaded, callInfo.full, etc.
This commit restored them to being turned on by defult,
as well added defaults for other props to be more explicit
on their defualt value.
cc @matklad
Co-authored-by: veetaha <veetaha2@gmail.com>
After refactoring the config we forgot to set defaults for
some properties like workspaceLoaded, callInfo.full, etc.
This commit restored them to being turned on by defult,
as well added defaults for other props to be more explicit
on their defualt value.
3817: vscode: highlight syntax tree ro editor r=matklad a=Veetaha
Small textmate grammar declaration to make rust-analyzer syntax tree more easily inspectable:
Btw, if we change the file extension of our `ra_syntax/test_data/**` files to `.rast` they should be highlighted in vscode too.
The colors of the tokens are actually going to be color-theme dependent, or you can customize them via:
```jsonc
{
"editor.tokenColorCustomizations": {
"textMateRules": [ { "scope": "name", "settings": { /* */ } } ]
}
}
```
Related: #3682
Co-authored-by: veetaha <veetaha2@gmail.com>
3780: Simplify r=matklad a=Veetaha
I absolutely love tha fact that removing `.clone()` simplifies the code comparing to other languages where it's actually the contrary (ahem ~~`std::move()`~~)
3787: vscode: add syntax tree inspection hovers and highlights r=matklad a=Veetaha
I implemented the reverse mapping (when you hover in the rust editor), but it seems overcomplicated, so I removed it
Related #3682
Co-authored-by: veetaha <veetaha2@gmail.com>
Co-authored-by: Veetaha <veetaha2@gmail.com>
3695: vscode: simplify and refactor config r=matklad a=Veetaha
Removed unnecessary interfaces, changed `cfg` to be a getter to ensure the fresh values any time possible.
Migrated from explicit casts to implicit.
Co-authored-by: veetaha <veetaha2@gmail.com>
3696: vscode: more type safety r=matklad a=Veetaha
3698: Consider references when applying postfix completions r=matklad a=SomeoneToIgnore
Sometimes my RA debugging workflow breaks because `.dbg` is applied to the variable that is used later in the code.
It's safer to consider the refences to avoid this for completions that may trigger the move.
3703: Don't try to enable proposed API's on stable r=matklad a=matklad
bors r+
🤖
Co-authored-by: veetaha <veetaha2@gmail.com>
Co-authored-by: Kirill Bulatov <mail4score@gmail.com>
Co-authored-by: Aleksey Kladov <aleksey.kladov@gmail.com>
This cancel is unnecessary since we cancel the previous inlay hints requests in `fetchHints()` method itself. This is not a hard error, we just called cancel() 2 times.
3629: Alternative aproach to plugin auto update r=matklad a=matklad
This is very much WIP (as in, I haven't run this once), but I like the result so far.
cc @Veetaha
The primary focus here on simplification:
* local simplification of data structures and control-flow: using union of strings instead of an enum, using unwrapped GitHub API responses
* global simplification of control flow: all logic is now in `main.ts`, implemented as linear functions without abstractions. This is stateful side-effective code, so arguments from [Carmack](http://number-none.com/blow/john_carmack_on_inlined_code.html) very much apply. We need all user interractions, all mutations, and all network requests to happen in a single file.
* as a side-effect of condensing everything to functions, we can get rid of various enums. The enums were basically a reified control flow:
```
enum E { A, B }
fn foo() -> E {
if cond { E::A } else { E::B }
}
fn bar(e: E) {
match e {
E::A => do_a(),
E::B => do_b(),
}
}
==>>
fn all() {
if cond { do_a() } else { do_b() }
}
```
* simplification of model: we don't need to reinstall on settings update, we can just ask the user to reload, we don't need to handle nightly=>stable fallback, we can ask the user to reinstall extension, (todo) we don't need to parse out the date from the version, we can use build id for nightly and for stable we can write the info directly into package.json.
Co-authored-by: Aleksey Kladov <aleksey.kladov@gmail.com>
Everything now happens in main.ts, in the bootstrap family of
functions. The current flow is:
* check everything only on extension installation.
* if the user is on nightly channel, try to download the nightly
extension and reload.
* when we install nightly extension, we persist its release id, so
that we can check if the current release is different.
* if server binary was not downloaded by the current version of the
extension, redownload it (we persist the version of ext that
downloaded the server).
3534: Feature: vscode impl nightlies download and installation r=Veetaha a=Veetaha
I need to test things more, but the core shape of the code is quite well-formed.
The main problem is that we save the release date only for nightlies and there are no means to get the release date of the stable extension (i.e. for this we would need to consult the github releases via a network request, or we would need to somehow save this info into package.json or any other file accessible from the extension code during the deployment step, but this will be very hard I guess).
So there is an invariant that the users can install nightly only from our extension and they can't do it manually, because when installing the nightly `.vsix` we actually save its release date to `globalState`
Closes: #3402
TODO:
- [x] More manual tests and documentation
cc @matklad @lnicola
Co-authored-by: Veetaha <gerzoh1@gmail.com>
Co-authored-by: Veetaha <veetaha2@gmail.com>
3561: feat: add debug code lens r=matklad a=hdevalke
Refs #3539
3577: Protect against infinite macro expansion in def collector r=edwin0cheng a=flodiebold
Something I noticed while trying to make macro expansion more resilient against errors.
There was a test for this, but it wasn't actually working because the first recursive expansion failed. (The comma...)
Even with this limit, that test (when fixed) still takes some time to pass because of the exponential growth of the expansions, so I disabled it and added a different one without growth.
CC @edwin0cheng
Co-authored-by: Hannes De Valkeneer <hannes@de-valkeneer.be>
Co-authored-by: hdevalke <2261239+hdevalke@users.noreply.github.com>
Co-authored-by: Florian Diebold <florian.diebold@freiheit.com>
3543: Parameter inlay hint separate from variable type inlay? #2876 r=matklad a=slyngbaek
Add setting to allow enabling either type inlay hints or parameter
inlay hints or both. Group the the max inlay hint length option
into the object.
- Add a new type for the inlayHint options.
- Add tests to ensure the inlays don't happen on the server side
Co-authored-by: Steffen Lyngbaek <steffenlyngbaek@gmail.com>
- Instead of a single object type, use several individual nested types
to allow toggling from the settings GUI
- Remove unused struct definitions
- Install and test that the toggles work
3549: Implement env! macro r=matklad a=edwin0cheng
This PR implements `env!` macro by adding following things:
1. Added `additional_outdirs` settings in vscode. (naming to be bikeshed)
2. Added `ExternSourceId` which is a wrapping for SourceRootId but only used in extern sources. It is because `OUT_DIR` is not belonged to any crate and we have to access it behind an `AstDatabase`.
3. This PR does not implement the `OUT_DIR` parsing from `cargo check`. I don't have general design about this, @kiljacken could we reuse some cargo watch code for that ?
~~Block on [#3536]~~
PS: After this PR , we (kind of) completed the `include!(concat!(env!('OUT_DIR'), "foo.rs")` macro call combo. [Exodia Obliterate!](https://www.youtube.com/watch?v=RfqNH3FoGi0)
Co-authored-by: Edwin Cheng <edwin0cheng@gmail.com>
- Updated naming of config
- Define struct in ra_ide and use remote derive in rust-analyzer/config
- Make inlayConfig type more flexible to support more future types
- Remove constructor only used in tests
Add setting to allow enabling either type inlay hints or parameter
inlay hints or both. Group the the max inlay hint length option
into the object.
- Add a new type for the inlayHint options.
- Add tests to ensure the inlays don't happen on the server side
3514: vscode: askBeforeDownload option r=matklad a=Veetaha
This is a small step towards #3402, also implements my proposal stated in #3403
Also renamed `BinarySource` to `ArtifactSource` in anticipation of nightlies installation that requires downloading not a binary itself but `.vsix` package, thus generalized to `artifact` term.
@matklad @lnicola
Co-authored-by: Veetaha <gerzoh1@gmail.com>
The new name seems much simpler and it doesn't limit
this config value only to downloading the server binary.
Thus we wouldn't need to create another config
properties to handle other downloads whatsoever.
Anyway, I believe (heuristically) that most of the users
would want to set "askBeforeDownload": false once
and never bother clicking on the notification again
(because otherwise there is no big point in installing rust-analyzer if it cannot install the server)
Also renamed BinarySource to ArtifactSource in anticipation of
nightlies installation that requires downloading
not a binary itself but .vsix package, thus generalized
to `artifact` term
3388: Remove inlay hint in diff views r=matklad a=vbfox
If the left side of a diff view that contain the old version of the file apply inlays they are misplaced and produce a weird display:
After the change:
The detection is done by blacklisting the url schemes used by git and subversion scm extensions, whitelisting `file` is also possible but neither is perfect as VSCode now support both pluggable scm extensions and pluggable remote filesystems. But I suspect that the list of scm extensions is more easily manageable.
**Note**: I can rebase on #3378 if needed as it touches the same lines of code
Co-authored-by: Julien Roncaglia <julien@roncaglia.fr>
If the left side of a diff view that contain the old
version of the file apply inlays they are misplaced.
The detection is done by blacklisting the url schemes used
by git and subversion scm extensions.
3361: Reset fontStyle for inlay hints r=matklad a=matklad
Otherwise, we get bold hints on `true` and `false`
bors r+
🤖
Co-authored-by: Aleksey Kladov <aleksey.kladov@gmail.com>
3308: vscode: fix vscode-vim keybindings conflict r=matklad a=Veetaha
Closes#3013 I hope
vscode-vim extension overrides the `type` command so that it prevents
some keypresses to reach the text document editor.
It conflicts with our `onEnter` keybinding
that is used to support automatic doc comments extending and
indentation insertion.
The VSCode-native way to implement this would be
to use regular expressions, but as per matklad it is
considered not acceptable for the language server.
Thus we implement it via a `Enter` keybinding that
invokes our `onEnter` command which does it via the language-server.
At the end of the day we may only apply
ad hoc workarounds for conflicting extensions.
But vscode has another bug for that. You
either cannot use parantheses in `when` condition
of a keybinding or it just malfunctions.
See an issue about that here: https://github.com/microsoft/vscode/issues/91473
To get the ultimate context, follow this [zulip thread](https://rust-lang.zulipchat.com/#narrow/stream/185405-t-compiler.2Fwg-rls-2.2E0/topic/Enhanced.20typing)
Co-authored-by: Veetaha <gerzoh1@gmail.com>
vscode-vim extension overrides the `type` command so that it prevents
some keypresses to reach the text document editor.
It conflicts with our `onEnter` keybinding
that is used to support automatic doc comments extending and
indentation insertion.
The VSCode-native way to implement this would be
to use regular expressions, but as per matklad it is
considered not acceptable for the language server.
Thus we implement it via a `Enter` keybinding that
invokes our `onEnter` command which sends
a request to rust-analyzer process and applies
the appropriate source change recieved from it.
At the end of the day we may only apply
ad hoc workarounds for conflicting extensions.
But vscode has another bug for that. You
either cannot use parantheses in `when` condition
of a keybinding or it just malfunctions.
See an issue about that here: https://github.com/microsoft/vscode/issues/91473
To get the ultimate context, follow this zulip thread: https://rust-lang.zulipchat.com/#narrow/stream/185405-t-compiler.2Fwg-rls-2.2E0/topic/Enhanced.20typing
3099: Init implementation of structural search replace r=matklad a=mikhail-m1
next steps:
* ignore space and other minor difference
* add support to ra_cli
* call rust parser to check pattern
* documentation
original issue #2267
Co-authored-by: Mikhail Modin <mikhailm1@gmail.com>
3162: Feature: vscode always downloads only the matching ra_lsp_server version r=matklad a=Veetaha
I tried to separate logically connected changes into separate commits, so enjoy!
Now TypeScript extension saves installed binary version in global state and always checks that the installed binary version equals the version of the TypeScript extension itself (to prevent version drifts).
Also, changed `fetchLatestArtifactReleaseInfo()` to `fetchArtifactReleaseInfo()` that takes an optional release tag (when not specified fetches the latest release). The version without a release tag will be useful in the future when adding auto-checking for updates.
I decided not to do `Download latest language server` command (I have stated the rationale for this in #3073) and let the extension itself decide which version of the binary it wants. This way the users will be able to get the latest `ra_lsp_server` binary after the approaching 2020-02-17 release, without having to manually delete the outdated one from `~/.config/Code/User/globalStorage/matklad.rust-analyzer`!
Closes#3073
Co-authored-by: Veetaha <gerzoh1@gmail.com>
3131: vscode: simplified config and to removed one source of truth of default values r=matklad a=Veetaha
Though not intended initially, the implementation of config design is alike [dart's one](https://github.com/Dart-Code/Dart-Code/blob/master/src/extension/config.ts) as pointed by @matklad in PM.
Co-authored-by: Veetaha <gerzoh1@gmail.com>
3137: Do not register all proposed features r=matklad a=kjeremy
Instead only opt-in to CallHierarchy since it has a vscode API but LSP support
is still proposed.
Discovered while working on SemanticTokens which does not have a vscode API
and is still in the proposed state. Somehow enabling it would crash the
language server.
See https://github.com/microsoft/vscode-languageserver-node/issues/572
Co-authored-by: kjeremy <kjeremy@gmail.com>
Instead only opt-in to CallHierarchy since it has a vscode API but LSP support
is still proposed.
Discovered while working on SemanticTokens which does not have a vscode API
and is still in the proposed state. Somehow enabling it would crash the
language server.
See https://github.com/microsoft/vscode-languageserver-node/issues/572
3083: Update some crates r=matklad a=kjeremy
3101: vscode: filter out arm linux from using prebuilt binaries r=matklad a=Veetaha
Closes#3076
Co-authored-by: kjeremy <kjeremy@gmail.com>
Co-authored-by: Veetaha <gerzoh1@gmail.com>
2979: vscode: now we are actually using tslib r=matklad a=Veetaha
We had an incorrect setup where `tslib` was in `devDependencies`.
FYI:
tslib is a runtime dependency, it contains functions that are used by transpiled JavaScript in order not to inline them in each file.
For example:
```ts
// foo.ts (source code)
import * as foo from "foo";
// ---------------------------
// foo.js (compiled output)
"use strict";
var __importStar = (this && this.__importStar) || function (mod) {
if (mod && mod.__esModule) return mod;
var result = {};
if (mod != null) for (var k in mod) if (Object.hasOwnProperty.call(mod, k)) result[k] = mod[k];
result["default"] = mod;
return result;
};
Object.defineProperty(exports, "__esModule", { value: true });
const foo = __importStar(require("foo"));
```
As you see, `tsc` generated that `__importStar` helper function in compiled output. And it generates it per each file if you don't enable `"importHelpers": true`. Now with `importHelpers` enabled we get the following picture:
```ts
// foo.ts (source code)
import * as foo from "foo";
// ---------------------------
// foo.js (compiled output)
"use strict";
Object.defineProperty(exports, "__esModule", { value: true });
const tslib_1 = require("tslib");
const foo = tslib_1.__importStar(require("foo"));
```
It saves some bundle size, but I am not entirely sure wheter we want that. Discussions are welcome!
Co-authored-by: Veetaha <gerzoh1@gmail.com>
2964: Improve responsiveness of the cargo check status label r=matklad a=lnicola
This is still not ideal because the label displays the crate that was just checked, not the one that's currently being checked. But it should give the impression of being faster.
Co-authored-by: Laurențiu Nicola <lnicola@dend.ro>
2061: Theme loading and "editor.tokenColorCustomizations" support. r=matklad a=seivan
Fixes: [Issue#1294](https://github.com/rust-analyzer/rust-analyzer/issues/1294#issuecomment-497450325)
TODO:
- [x] Load themes
- [x] Load existing `ralsp`-prefixed overrides from `"workbench.colorCustomizations"`.
- [x] Load overrides from `"editor.tokenColorCustomizations.textMateRules"`.
- [x] Use RA tags to load `vscode.DecorationRenderOptions` (colors) from theme & overrides.
- [x] Map RA tags to common TextMate scopes before loading colors.
- [x] Add default scope mappings in extension.
- [x] Cache mappings between settings updates.
- [x] Add scope mapping configuration manifest in `package.json`
- [x] Load configurable scope mappings from settings.
- [x] Load JSON Scheme for text mate scope rules in settings.
- [x] Update [Readme](https://github.com/seivan/rust-analyzer/blob/feature/themes/docs/user/README.md#settings).
Borrowed the theme loading (`scopes.ts`) from `Tree Sitter` with some modifications to reading `"editor.tokenColorCustomizations"` for merging with loaded themes and had to remove the async portions to be able to load it from settings updates.
~Just a PoC and an idea I toyed around with a lot of room for improvement.~
For starters, certain keywords aren't part of the standard TextMate grammar, so it still reads colors from the `ralsp` prefixed values in `"workbench.colorCustomizations"`.
But I think there's more value making the extension work with existing themes by maping some of the decoration tags to existing key or keys.
<img width="453" alt="Screenshot 2019-11-09 at 17 43 18" src="https://user-images.githubusercontent.com/55424/68531968-71b4e380-0318-11ea-924e-cdbb8d5eae06.png">
<img width="780" alt="Screenshot 2019-11-09 at 17 41 45" src="https://user-images.githubusercontent.com/55424/68531950-4b8f4380-0318-11ea-8f85-24a84efaf23b.png">
<img width="468" alt="Screenshot 2019-11-09 at 17 40 29" src="https://user-images.githubusercontent.com/55424/68531952-51852480-0318-11ea-800a-6ae9215f5368.png">
These will merge with the default ones coming with the extension, so you don't have to implement all of them and works well with overrides defined in settings.
```jsonc
"editor.tokenColorCustomizations": {
"textMateRules": [
{
"scope": "keyword",
"settings": {
"fontStyle": "bold",
}
},
]
},
```
Edit: The idea is to work with 90% of the themes out there by working within existing scopes available that are generally styled. It's not to say I want to erase the custom Rust scopes - those should still remain and eventually worked into a custom grammar bundle for Rust specific themes that target those, I just want to make it work with generic themes offered on the market place for now.
A custom grammar bundle and themes for Rust specific scopes is out of... scope for this PR.
We'll make another round to tackle those issues.
Current fallbacks implemented
```typescript
[
'comment',
[
'comment',
'comment.block',
'comment.line',
'comment.block.documentation'
]
],
['string', ['string']],
['keyword', ['keyword']],
['keyword.control', ['keyword.control', 'keyword', 'keyword.other']],
[
'keyword.unsafe',
['storage.modifier', 'keyword.other', 'keyword.control', 'keyword']
],
['function', ['entity.name.function']],
['parameter', ['variable.parameter']],
['constant', ['constant', 'variable']],
['type', ['entity.name.type']],
['builtin', ['variable.language', 'support.type', 'support.type']],
['text', ['string', 'string.quoted', 'string.regexp']],
['attribute', ['keyword']],
['literal', ['string', 'string.quoted', 'string.regexp']],
['macro', ['support.other']],
['variable', ['variable']],
['variable.mut', ['variable', 'storage.modifier']],
[
'field',
[
'variable.object.property',
'meta.field.declaration',
'meta.definition.property',
'variable.other'
]
],
['module', ['entity.name.section', 'entity.other']]
```
Co-authored-by: Seivan Heidari <seivan.heidari@icloud.com>
2583: Use prettier settings in ts-lint r=matklad a=edwin0cheng
This PR add `tslint-plugin-prettier` extension in ts-lint, which "runs prettier rules as tslint rules." and remove `quotemark` from ts-lint and let prettier to handle it.
And also fix#2515
Co-authored-by: Edwin Cheng <edwin0cheng@gmail.com>
2568: Add option to disable all-targets. r=matklad a=pftbest
Can be useful in embedded.
Co-authored-by: Vadzim Dambrouski <vadzim.dambrouski@promwad.com>
2571: Fixed a typo in settings r=matklad a=omerbenamram
@lnicola found a typo in the description for one of the settings introduced in #2559.
Co-authored-by: Omer Ben-Amram <omerbenamram@gmail.com>
2565: Fixed syntax highlighting not refreshing on windows. r=matklad a=omerbenamram
I was encoutering the same probelm described in #1690.
It seems that events initiated by the frontend with `rust-analyzer/decorationsRequest` would go through.
So whenever a user switches tabs, highlighting will update.
However, when decorations are initiated by a notification with `rust-analyzer/publishDecorations`, it would fail on this check here 6cbd8a4a4b/editors/code/src/notifications/publish_decorations.ts (L15) (`targetEditor` will always be `undefined`).
This is because it's trying to match the uri `rust-analyzer` sends (which uses an uppercase drive letter) to the uri provided at `editor.document.uri.toString()`, which is both escaped (uses `%3a` for `:`), and uses a lowercase letter drive.
Aparrently this was an issue for some other extensions aswell - https://github.com/Microsoft/vscode/issues/68325.
But this is the defined behavior - c110d84460/src/vs/vscode.d.ts (L1304)
This fix is only relevant for windows.
I've opted for a server-side fix, since rust will always return uppercase letters for drives, there seems to be no other easy solution than manipulating the Url string before sending it to the frontend.
Closes#1690.
Co-authored-by: Omer Ben-Amram <omerbenamram@gmail.com>
2508: Code: don't check for ra_lsp_server on Windows r=matklad a=lnicola
Workaround for https://github.com/rust-analyzer/rust-analyzer/pull/2503#issuecomment-562980020.
~~(not yet tested on Windows)~~
We can't run `ra_lsp_server --version` right now because the server doesn't seem to handle arguments (so it hangs).
Co-authored-by: Laurențiu Nicola <lnicola@dend.ro>
This commit implements a general truncation framework for HirFormatter
that keeps track of how much has been output so far. This information
can then be used to perform truncation inside the language server,
instead of relying on the client.
Initial support is implemented for truncating types hints using the
maxInlayHintLength server config option. The existing solution in the
VSCode extension has been removed in favor of letting the server
truncate type hints.
1984: Bump rollup and vsce r=matklad a=kjeremy
I got sick of the vsce warning on install and noticed that rollup was also out of date.
Co-authored-by: kjeremy <kjeremy@gmail.com>
The old `vscode` package is outdated and it is recommened to switch to
these two new packages. This also solves a problem of a missing `.d.ts`
for `vscode` in Nixos.
1652: Improve type hints behavior r=matklad a=SomeoneToIgnore
This PR fixed the following type hints issues:
* Restructures the `InlayKind` enum contents based on the discussion here: https://github.com/rust-analyzer/rust-analyzer/pull/1606#issuecomment-515968055
* Races described in #1639
* Caches the latest decorations received for each file to show them the next time the file is opened (instead of a new server request)
Co-authored-by: Kirill Bulatov <mail4score@gmail.com>
1458: Run VS Code tests on CI r=matklad a=etaoins
This is actually much faster than I expected; it takes about 13 seconds to download VS Code and run the unit tests. This means the VS Code tests are still significantly faster than the Rust ones.
If this ends up being unreliable we can always remove it later or move it to a separate optional job.
We also need to ignore the `.vscode-test` directory when running `prettier` or it will get upset about some temporary JSON files VS Code creates.
cc @killercup
Co-authored-by: Ryan Cumming <etaoins@gmail.com>
1459: Include primary span label in VS Code diagnostics r=matklad a=etaoins
In most cases the primary label span repeats information found elsewhere in the diagnostic. For example, with E0061:
```json
{
"message": "this function takes 2 parameters but 3 parameters were supplied",
"spans": [{"label": "expected 2 parameters"}]
}
```
However, with some mismatched type errors (E0308) the expected type only appears in the primary span's label, e.g.:
```json
{
"message": "mismatched types",
"spans": [{"label": "expected usize, found u32"}]
}
```
I initially added the primary span label to the message unconditionally. However, for most error types the child diagnostics repeat the primary span label with more detail. `rustc` also renders the duplicate text but because the span label and child diagnostics appear in visually distinct places it's not as confusing.
This takes a heuristic approach where it will only add the primary span label if there are no child message lines. For most error types the child messages repeat the primary span label with more detail.
Co-authored-by: Ryan Cumming <etaoins@gmail.com>
This adds `unreachable_code` to the list of diagnostic codes we map to
`Unnecessary` in Visual Studio Code. This is consistent with what the
TypeScript language server does.
In most cases the primary label span repeats information found elsewhere
in the diagnostic. For example, with E0061:
```
{
"message": "this function takes 2 parameters but 3 parameters were supplied",
"spans": [{"label": "expected 2 parameters"}]
}
```
However, with some mismatched type errors (E0308) the expected type only
appears in the primary span's label, e.g.:
```
{
"message": "mismatched types",
"spans": [{"label": "expected usize, found u32"}]
}
```
I initially added the primary span label to the message unconditionally.
However, for most error types the child diagnostics repeat the primary
span label with more detail. `rustc` also renders the duplicate text but
because the span label and child diagnostics appear in visually distinct
places it's not as confusing.
This takes a heuristic approach where it will only add the primary span
label if there are no child message lines.
This is actually much faster than I expected; it takes about 13 seconds
to download VS Code and run the unit tests. This means the VS Code tests
are still significantly faster than the Rust ones.
If this ends up being unreliable we can always remove it later or move
it to a separate optional job.
We also need to ignore the `.vscode-test` directory when running
`prettier` or it will get upset about some temporary JSON files VS Code
creates.
1454: Fix `cargo watch` code action filtering r=etaoins a=etaoins
There are two issues with the implementation of `provideCodeActions` introduced in #1439:
1. We're returning the code action based on the file its diagnostic is in; not the file the suggested fix is in. I'm not sure how often fixes are suggested cross-file but it's something we should handle.
2. We're not filtering code actions based on the passed range. The means if there is any suggestion in a file we'll show an action for every line of the file. I naively thought that VS Code would filter for us but that was wrong.
Unfortunately the VS Code `CodeAction` object is very complex - it can handle edits across multiple files, run commands, etc. This makes it complex to check them for equality or see if any of their edits intersects with a specified range.
To make it easier to work with suggestions this introduces a `SuggestedFix` model object and a `SuggestFixCollection` code action provider. This is a layer between the raw Rust JSON and VS Code's `CodeAction`s. I was reluctant to introduce another layer of abstraction here but my attempt to work directly with VS Code's model objects was worse.
Co-authored-by: Ryan Cumming <etaoins@gmail.com>
`tslint` doesn't catch this because TypeScript has had this check
builtin since 2.9. However, it's disabled by default so right now
nothing is checking for unused variables.
There are two issues with the implementation of `provideCodeActions`
introduced in #1439:
1. We're returning the code action based on the file its diagnostic is
in; not the file the suggested fix is in. I'm not sure how often
fixes are suggested cross-file but it's something we should handle.
2. We're not filtering code actions based on the passed range. The means
if there is any suggestion in a file we'll show an action for every
line of the file. I naively thought that VS Code would filter for us
but that was wrong.
Unfortunately the VS Code `CodeAction` object is very complex - it can
handle edits across multiple files, run commands, etc. This makes it
complex to check them for equality or see if any of their edits
intersects with a specified range.
To make it easier to work with suggestions this introduces a
`SuggestedFix` model object and a `SuggestFixCollection` code action
provider. This is a layer between the raw Rust JSON and VS Code's
`CodeAction`s. I was reluctant to introduce another layer of abstraction
here but my attempt to work directly with VS Code's model objects was
worse.
Currently all of our VS Code diagnostics are given the source of
`rustc`. However, if you have something like `cargo-watch.command` set
to `clippy` it will also watch for Clippy lints. The `rustc` source is a
bit misleading in that case.
Fortunately, Rust's tool lints (RFC 2103) line up perfectly with VS
Code's concept of `source`. This checks for lints scoped to a given tool
and then splits them in to a `source` and tool-specific `code`.
As promised in #1439 this is an initial attempt at unit testing the
VSCode extension. There are two separate parts to this: getting the test
framework working and unit testing the code in #1439.
The test framework nearly intact from the VSCode extension generator.
The main thing missing was `test/index.ts` which acts as an entry point
for Mocha. This was simply copied back in. I also needed to open the
test VSCode instance inside a workspace as our file URI generation
depends on a workspace being open.
There are two ways to run the test framework:
1. Opening the extension's source in VSCode, pressing F5 and selecting
the "Extensions Test" debug target.
2. Closing all copies of VSCode and running `npm test`. This is started
from the command line but actually opens a temporary VSCode window to
host the tests.
This doesn't attempt to wire this up to CI. That requires running a
headless X11 server which is a bit daunting. I'll assess the difficulty
of that in a follow-up branch. This PR is at least helpful for local
development without having to induce errors on a Rust project.
For the actual tests this uses snapshots of `rustc` output from a real
Rust project captured from the command line. Except for extracting the
`message` object and reformatting they're copied verbatim into fixture
JSON files.
Only four different types of diagnostics are tested but they represent
the main combinations of code actions and related information possible.
They can be considered the happy path tests; as we encounter
corner-cases we can introduce new tests fixtures.
The first cut was a bit rough with the blanket `unused_*` rule. This
trigger for things like `unused_mut` where the code is used but it's
suboptimal. It's misleading to grey out the code in those cases.
Instead, use an explicit list of things known to be dead code.
Currently we depend on the ASCII rendering string that `rustc` provides
to populate Visual Studio Code's diagnostic. This has a number of
shortcomings:
1. It's not a very good use of space in the error list
2. We can't jump to secondary spans (e.g. where a called function is
defined)
3. We can't use Code Actions aka Quick Fix
This moves all of the low-level parsing and mapping to a
`rust_diagnostics.ts`. This uses some heuristics to map Rust diagnostics
to VsCode:
1. As before, the Rust diagnostic message and primary span is used for
the root diagnostic. However, we now just use the message instead of
the rendered version.
2. Every secondary span is converted to "related information". This
shows as child in the error list and can be jumped to.
3. Every child diagnostic is categorised in to three buckets:
1. If they have no span they're treated as another line of the root
messages
2. If they have replacement text they're treated as a Code Action
3. If they have a span but no replacement text they're treated as
related information (same as secondary spans).
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?
