Re: [scala-reflect] arguments get replaced with null in MethodMirror.apply when there are both implicit and non-implicit parameters

Your fix fixes it. It's the AnyVal that sends it down that code path.

I bet they would appreciate a Jira issue.



On Mon, Jan 19, 2015 at 7:20 AM, John Sullivan <[email protected]
> wrote:

> Good morning everyone and happy MLK day!
>
> I seem to have found a bug in scala-reflect - when invoking apply on a
> MethodMirror, I end up getting a null where I have supplied a non-null
> value.
>
> Here is some sample code. You can see I get a null where I should get a
> B(7):
>
> object Test extends App {
>   object a {
>     case class B(i: Int)
>     implicit class C(val i: Int) extends AnyVal
>     case class D(b: B, c: C)
>   }
>
>   import a._
>   import scala.reflect.runtime.currentMirror
>   import scala.reflect.runtime.universe._
>
>   val tpe = typeOf[D]
>   val classSymbol: ClassSymbol = tpe.typeSymbol.asClass
>   val moduleSymbol: ModuleSymbol = classSymbol.companion.asModule
>   val moduleMirror: ModuleMirror =
> currentMirror.reflectModule(moduleSymbol)
>   val moduleInstance: Any = moduleMirror.instance
>   val instanceMirror = currentMirror.reflect(moduleInstance)
>   val typeSignature = instanceMirror.symbol.typeSignature
>   val applyMethod = typeSignature.member(TermName("apply")).asMethod
>   val applyMirror = instanceMirror.reflectMethod(applyMethod)
>   val applyResult = applyMirror(B(7), new C(8))
>   println(applyResult)
>   // output is D(null,Test$a$C@8)
>   // the null seems to be due to a fall-through at line 433 of
> JavaMirrors.scala
> }
>
> This seems to happen whenever I have a MethodMirror with both implicit and
> non-implicit parameters. I am not familiar with this code at all, but to me
> it seems like an "else args1(i) = arg" is missing just below line 433 of
> JavaMirrors.scala.
>
> This test was run against Scala 2.11.5, but I got the same results against
> 2.11.2, 2.11.3, and 2.11.4. I tried against 2.10 as well, but the API has
> changed a lot and I would have had to refactor to get it to compile there.
>
> If you like I can submit a bug report.
>
> Thanks! -John
>
> --
> You received this message because you are subscribed to the Google Groups
> "scala-language" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to [email protected].
> For more options, visit https://groups.google.com/d/optout.
>

-- 
You received this message because you are subscribed to the Google Groups "scala-language" group.
To unsubscribe from this group and stop receiving emails from it, send an email to [email protected].
For more options, visit https://groups.google.com/d/optout.