Guillaume Martres

> Can the scalac genbcode ever inline something that is not marked as @inline? Yes with call-site `@inline`, as mentioned in https://github.com/sbt/zinc/issues/537#issuecomment-386005581 and I also still stand by https://github.com/sbt/zinc/issues/537#issuecomment-386006868, this...

I suggest making a PR to scala/scala that drops the feature and put a warning in its place.

Can you provide a testcase?

@adpi2 If you miss working on sbt ^ :).

> but Zinc is not at fault here since the error happens in the first compilation of the modified sources Zinc is assuming that compiling the modified sources by themselves...

Sometimes it's useful to print the logs when the test doesn't fail for debugging purpose too.

> User adds a name of a scripted test to a setting in the build that enables debugging for that scripted test. That requires reloading the build which is always...

> Currently, test discoverability is done via reflection Actually, it's done using the incremental compilation Analysis output, at least for junit: see https://github.com/sbt/junit-interface/blob/master/src/main/java/com/novocode/junit/JUnitFingerprint.java and grep for fingerprint in https://github.com/sbt/sbt, I...

> This is arguably not such a big problem because java dependencies are always there and never change. This is true for now, but in the future the Scala compiler...

@retronym What's the proper way to go from a PlainNioFile to the corresponding file in the real file system if one exists? Is calling .container good enough ?