[WIP] Warn unused unused annotation #10241
Conversation
|
ha, and the code says |
som-snytt
force-pushed
the
issue/11631-lint-unused
branch
from
fc4d104 to
240ada0
Compare
som-snytt
force-pushed
the
issue/11631-lint-unused
branch
from
240ada0 to
fc934ce
Compare
| class C { | ||
| def n: Int = 42 | ||
| def f(@unused i: Int) = n + 1 | ||
| def g(@unused i: Int) = i + 1 // unused unused |
There was a problem hiding this comment.
unused unused doesn't seem to be reported?
There was a problem hiding this comment.
I wasn't able to workaround the spurious report for ChromeTrace.EventType, so I put in a version trip switch.
That is also why I fixed the ordering of ScalaVersion so that 2.13.11-development < 2.13.11.
There was a problem hiding this comment.
Actually, the problem of the unused diagnostic not seeing trees of constant values made me want to address the few well-known limitations with the unused checks for 2.13.11, depending perhaps on when it is scheduled.
| if (sym != null && sym.pos.isDefined) | ||
| sym.getAnnotation(UnusedClass).foreach { annot => | ||
| if (!runReporting.suppressionExists(annot.pos)) | ||
| runReporting.addSuppression(Suppression(annot.pos, List(MessageFilter.Category(WarningCategory.Unused)), sym.pos.start, sym.pos.end, synthetic = true)) |
There was a problem hiding this comment.
I'm not sure how reliable it is to use sym.pos here...
➜ sandbox qs -Wunused
Welcome to Scala 2.13.11-20221222-071937-84bf5ca (OpenJDK 64-Bit Server VM, Java 17.0.1).
Type in expressions for evaluation. Or try :help.
scala> object t { @annotation.unused private var x = 2 }
java.lang.UnsupportedOperationException: Position.point on NoPosition
at scala.reflect.internal.util.Position.fail(Position.scala:24)
See rangeFinder in Typers
There was a problem hiding this comment.
I had thrown in sym.pos.isDefined for what I thought was good measure.
| def GEN_== (other: Tree, kind: ClassSymbol) = fn(target, getMember(kind, nme.EQ), other) | ||
| @unused("avoid warning for multiple parameters") |
There was a problem hiding this comment.
I think this was due to an early version of multiarg infix that warned for this name, but it doesn't seem to warn now.
som-snytt
changed the title
Fixes scala/bug#11631