New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Update fork to all the newest deps and code #162
Commits on Mar 27, 2024
-
-
bugfix: Make sure proper runtime classpath is used when debugging
Previously, we would use the default config, which would not always be correct. Now, we use runtime config with a fallback to normal platform config.
-
-
-
-
-
-
improvement: Change information about test framework to warn
Previously, this would be set to error, which would mean Metals would show it to the users more prominently. Now, we change it to warn, which means it's still present, but since this might be actually be related to non compiled code, it's prominence is reduced,
-
-
-
-
-
-
-
refactor: Remove forked parts of scala js envs
As far as I can see this should work properly and it hits us even more to have duplicated part of code that is not maintained.
-
-
-
-
bugfix: Print copying error together with stacktrace
It seems we don't get that properly forwarded to Bloop otherwise.
-
-
-
-
-
-
-
-
-
-
Revert "Update upload/download artifact to v4"
This reverts commit 41ce771.
-
-
-
-
-
-
refactor: don't send task start/end for no-op compilations
This might be a bit opinionated and I'm not 100% sure on the reason this was done in the first place, but this changes the current behavior around reporting start and end compilations for no-ops. This still retains the compile report if it was a no-op and also still checks the diagnostics and reports them, but skips the task start/end. Originally this was added in a commit without any context... but there was a comment that said: > // When no-op, we keep reporting the start and the end of compilation for consistency However, I'm not really sure that consistency matters here. In reality this ends up creating a bunch of noise on the client side, especially when these tasks turn into LSP progress notifications that aren't useful for the user to see. You can see more context about this change in the issue reported [here](scalameta/metals#6099) and also the discussion found [here](build-server-protocol/build-server-protocol#654). It also seems that in some projects like scala-cli these are even [ignored](https://github.com/VirtusLab/scala-cli/blob/6b7a10007e4eefde717079255e0df38c027f788b/modules/build/src/main/scala/scala/build/ConsoleBloopBuildClient.scala#L109).
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Revert "refactor: Remove Bloop4j as it seems not to be used at all"
This reverts commit 2a4ee42.