sbt 1.10.1 (was 1.9.9)

[SethTisue]

not for merge until 1.10.0 final; but let's see if CI likes it

[SethTisue]

@eed3si9n the test failures here worry me somewhat, but I might not have time this week to look into it

[eed3si9n]

Acknowledged. Thanks for starting this, @SethTisue!

[eed3si9n]

Opened sbt/zinc#1351

[eed3si9n]

FYI @bishabosha - new Zinc 1.10.x breaks scripted test added by #18880.
https://github.com/scala/scala3/actions/runs/8690208123/job/23829593105#step:9:342

[info] [error] java.lang.RuntimeException: Can't find source file for a.A.Inner_sel
[info] [error] 	at scala.sys.package$.error(package.scala:30)
[info] [error] 	at sbt.internal.inc.Analysis$.$anonfun$sources$1(Analysis.scala:105)
[info] [error] 	at scala.Option.getOrElse(Option.scala:189)
[info] [error] 	at sbt.internal.inc.Analysis$.sourceFileForClass$1(Analysis.scala:105)
[info] [error] 	at sbt.internal.inc.Analysis$.isJavaClass$1(Analysis.scala:107)
[info] [error] 	at sbt.internal.inc.Analysis$.$anonfun$sources$2(Analysis.scala:108)
[info] [error] 	at sbt.internal.inc.Analysis$.$anonfun$sources$2$adapted(Analysis.scala:108)
[info] [error] 	at scala.collection.TraversableLike.$anonfun$partition$1(TraversableLike.scala:450)
[info] [error] 	at scala.collection.immutable.HashMap$HashMapKeys.$anonfun$foreach$1(HashMap.scala:150)
[info] [error] 	at scala.collection.immutable.HashMap$HashMap1.foreachEntry(HashMap.scala:401)
[info] [error] 	at scala.collection.immutable.HashMap$HashTrieMap.foreachEntry(HashMap.scala:735)
[info] [error] 	at scala.collection.immutable.HashMap$HashTrieMap.foreachEntry(HashMap.scala:735)
[info] [error] 	at scala.collection.immutable.HashMap$HashMapKeys.foreach(HashMap.scala:150)
[info] [error] 	at scala.collection.TraversableLike.partition(TraversableLike.scala:450)
[info] [error] 	at scala.collection.TraversableLike.partition$(TraversableLike.scala:448)
[info] [error] 	at scala.collection.AbstractTraversable.partition(Traversable.scala:108)
[info] [error] 	at sbt.internal.inc.Analysis$.sources(Analysis.scala:108)
[info] [error] 	at sbt.internal.inc.IncrementalCommon$CycleState$IncrementalCallbackImpl.mergeAndInvalidate(IncrementalCommon.scala:193)
[info] [error] 	at sbt.internal.inc.IncrementalCommon$CycleState$IncrementalCallbackImpl.completeCycle(IncrementalCommon.scala:209)
[info] [error] 	at sbt.internal.inc.AnalysisCallback.getCycleResultOnce(Incremental.scala:940)
[info] [error] 	at sbt.internal.inc.Incremental$$anon$2.run(Incremental.scala:455)
[info] [error] 	at sbt.internal.inc.IncrementalCommon$CycleState.next(IncrementalCommon.scala:117)
[info] [error] 	at sbt.internal.inc.IncrementalCommon$$anon$1.next(IncrementalCommon.scala:56)
[info] [error] 	at sbt.internal.inc.IncrementalCommon$$anon$1.next(IncrementalCommon.scala:52)
[info] [error] 	at sbt.internal.inc.IncrementalCommon.cycle(IncrementalCommon.scala:265)
[info] [error] 	at sbt.internal.inc.Incremental$.$anonfun$incrementalCompile$8(Incremental.scala:409)
[info] [error] 	at sbt.internal.inc.Incremental$.withClassfileManager(Incremental.scala:496)
[info] [error] 	at sbt.internal.inc.Incremental$.incrementalCompile(Incremental.scala:396)
[info] [error] 	at sbt.internal.inc.Incremental$.apply(Incremental.scala:170)
[info] [error] 	at sbt.internal.inc.IncrementalCompilerImpl.compileInternal(IncrementalCompilerImpl.scala:534)
[info] [error] 	at sbt.internal.inc.IncrementalCompilerImpl.$anonfun$compileIncrementally$1(IncrementalCompilerImpl.scala:488)
[info] [error] 	at sbt.internal.inc.IncrementalCompilerImpl.handleCompilationError(IncrementalCompilerImpl.scala:332)
[info] [error] 	at sbt.internal.inc.IncrementalCompilerImpl.compileIncrementally(IncrementalCompilerImpl.scala:425)
[info] [error] 	at sbt.internal.inc.IncrementalCompilerImpl.compile(IncrementalCompilerImpl.scala:137)

I can' tell if this is uncovering some changes Zinc made in 1.10.0 (https://eed3si9n.com/sbt-1.10.0-beta), or uncovering some issue with the compiler bridge.

[bishabosha]

@eed3si9n on the current main branch with sbt 1.9.9, I printed out the relations after a/compile in that task and it seems some classes are missing (such as a.A.Inner_sel):

class names: Relation [
  ${BASE}/a/src/main/scala/a/A.java -> a.A
  ${BASE}/a/src/main/scala/a/A.java -> a.A$.Inner
  ${BASE}/a/src/main/scala/a/A.java -> a.A$.Inner_sel
  ${BASE}/a/src/main/scala/a/AImport.java -> a.AImport
  ${BASE}/a/src/main/scala/a/AImport.java -> a.AImport.Inner
]

here are the set of apis.allInternalClasses:

[
  a.A,
  a.AImport.Inner,
  a.A$.Inner_sel,
  a.A.Inner_sel,
  a.AImport,
  a.A$.Inner,
  a.A.Inner,
  a.AImport$.Inner
]

[eed3si9n]

@bishabosha Thanks. To clarify, you mean it's there in sbt 1.9.9, but those entries that should be coming from Java sources are missing in sbt 1.10.0-RC2, right?

[bishabosha]

@eed3si9n both those outputs are with sbt 1.9.9. From what I understand (please verify), the class names relation comes from generatedNonLocalClass callback, and allInternalClasses are just whatever was recorded by ExtractAPI phase - so should these contain the same classes?

[eed3si9n]

https://github.com/scala/scala3/blob/05114dd6bcd007a6becc83011fe9916d35ba4716/sbt-test/pipelining/Yjava-tasty-fromjavaobject/a/src/main/scala/a/A.java

If Java compiler works like Scala compiler, I guess javac only emits a.A$.Inner_sel.class, but it generates some forwarder so humans can write a.A.Inner_sel?

[SethTisue]

The scripted tests are now running afoul of SIP-51 😆 I'll update it soon-ish

[mbovel]

Out of curiosity, why not also updating the sbt version in tests and community projects as you did in 19a6990?

[SethTisue]

oh, I suppose I'd have gotten to that before undraft-ing it

anyway, I've just updated it

if CI still fails, I won't have time to look at it until October, so if somebody's in more of a hurry than that, please feel free to take it over