Metals: Metals will not connect to sbt if compilation fails

[tgodzik]

Describe the bug

If we try to connect to the sbt BSP server on a project that has a compile error, it will not succeed

To Reproduce

Steps to reproduce the behavior:

1. Create a new project
2. Introduce an error
3. Try to switch to sbt bsp

Expected behavior

sbt just reports the error.

Installation:

• Operating system: macOS/Windows/Linux
• Editor: Visual Studio Code/Atom/Vim/Sublime/Emacs
• Metals version: v0.9.8

Additional context

This was raised on gitter

Search terms

sbt BSP compile error

[ckipp01]

Thanks for reporting this @tgodzik! Although I'm unable to reproduce it. This is what I did to test it.

• Create a new sbt project (has sbt 1.4.6)
• Imported with Bloop, went to Main.scala and added a val hi: String = 3.
• Executed a metals.generate-bsp-config which created the .bsp/sbt.json and then attempted to connect.

It did connect correctly, and in the logs I see this:

2021.01.04 16:50:17 INFO  Attempting to connect to the build server...
2021.01.04 16:50:17 INFO  tracing is enabled: /Users/ckipp/Library/Caches/org.scalameta.metals/bsp.trace.json
2021.01.04 16:50:23 INFO  time: connected to build server in 6.03s
2021.01.04 16:50:23 INFO  Connected to Build server v1.4.6
2021.01.04 16:50:23 INFO  Processing
2021.01.04 16:50:23 INFO  Processing
2021.01.04 16:50:23 INFO  Processing
2021.01.04 16:50:23 INFO  Processing
2021.01.04 16:50:23 INFO  Processing
2021.01.04 16:50:23 INFO  Processing
2021.01.04 16:50:23 INFO  Processing
2021.01.04 16:50:24 INFO  time: imported build in 1.43s
2021.01.04 16:50:24 INFO  Processing
2021.01.04 16:50:24 INFO  time: indexed workspace in 0.6s
2021.01.04 16:50:25 WARN  no build target for: /Users/ckipp/Documents/scala-workspace/tester/build.sbt
2021.01.04 16:50:24 INFO  Processing
2021.01.04 16:50:24 INFO  compiling root
2021.01.04 16:50:24 INFO  compiling 2 Scala sources to /Users/ckipp/Documents/scala-workspace/tester/target/scala-2.13/classes ...
2021.01.04 16:50:25 INFO  Non-compiled module 'compiler-bridge_2.13' for Scala 2.13.4. Compiling...
2021.01.04 16:50:32 INFO    Compilation completed in 6.949s.
2021.01.04 16:50:32 ERROR /Users/ckipp/Documents/scala-workspace/tester/src/main/scala/example/Main.scala:13:20: type mismatch;
 found   : Int(3)
 required: String
2021.01.04 16:50:32 ERROR   val hi: String = 3
2021.01.04 16:50:32 ERROR                    ^
2021.01.04 16:50:32 ERROR one error found
2021.01.04 16:50:32 INFO  time: compiled root in 7.69s
2021.01.04 16:50:32 ERROR (Compile / compileIncremental) Compilation failed

I'm assuming this isn't what you're seeing?

[tgodzik]

It seems I can no longer reproduce it :thinking: I thought I've seen it a number of times, but maybe new sbt version fixed it? Or I was doing something totally wrong. Either way I will close it and we can reopen if I ever encounter it again.

[asflierl]

Hi, I was the person who reported this on Gitter. I can (still) reproduce this.

here's what I do:

1. git clone https://github.com/asflierl/sglicko2.git (it is a very small & simple project)
2. cd sglicko2
3. git checkout db0b84787dcbf1bcceb37db3abf44ba2129c78ec
4. sbtn compile (this will report a compile error)
5. code .
6. in VSCode issue "Metals: switch build server" command and switch to SBT
7. BOOM :)

I got SBT 1.4.6 installed, Metals updated to 0.9.8 and using GraalVM CE 20.3.0 for Java 11.

Anything else I can do to help diagnose this? @tgodzik @ckipp01

[ckipp01]

Hey @asflierl I just pulled down your repo and indeed, it does fail for me as well!

2021.01.05 16:37:03 ERROR     }x
2021.01.05 16:37:03 ERROR      ^
2021.01.05 16:37:03 ERROR one error found
2021.01.05 16:37:03 ERROR (Compile / compileIncremental) Compilation failed
2021.01.05 16:37:03 INFO  Processing
2021.01.05 16:37:03 INFO  Disconnecting from sbt session...
2021.01.05 16:37:03 INFO  Shut down connection with build server.
2021.01.05 16:37:03 ERROR Failed to connect with build server, no functionality will work.
org.eclipse.lsp4j.jsonrpc.ResponseErrorException:
        at org.eclipse.lsp4j.jsonrpc.RemoteEndpoint.handleResponse(RemoteEndpoint.java:209)
        at org.eclipse.lsp4j.jsonrpc.RemoteEndpoint.consume(RemoteEndpoint.java:193)
        at org.eclipse.lsp4j.jsonrpc.TracingMessageConsumer.consume(TracingMessageConsumer.java:114)
        at org.eclipse.lsp4j.jsonrpc.json.StreamMessageProducer.handleMessage(StreamMessageProducer.java:194)
        at org.eclipse.lsp4j.jsonrpc.json.StreamMessageProducer.listen(StreamMessageProducer.java:94)
        at org.eclipse.lsp4j.jsonrpc.json.ConcurrentMessageProcessor.run(ConcurrentMessageProcessor.java:113)
        at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
        at java.util.concurrent.FutureTask.run(FutureTask.java:266)
        at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
        at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
        at java.lang.Thread.run(Thread.java:748)

🤔 I'm not fully sure why in this instance it behaves differently than with a minimal reproduction. I'll have to dive into this a bit more to figure out why it's disconnecting in this scenario and not the other. Thanks for following up!