Re: [Jastadd] [bug report] absence of AbstractWildcardType.createQualifiedAccess()

From: Jesper Öqvist <jesper.oqvist_at_cs.lth.se>
Date: Thu, 03 May 2012 12:35:44 +0200

On 05/03/2012 02:30 AM, Na, Hyunik wrote:
>
> I thought about the reason of this difference.
>
> I guess that the cause might be the static fields of the classes of
> JastAddJ.
>
> Each invocation of JavaCompiler.compile() in your script might not be
> run in a completely ?fresh? state of the compiler,
>
> because some classes have static fields.
>
> So, each invocation of JavaCompiler.compile() might be affected by
> earlier invocations,
>
> and the compiler might run into a state which is unreachable from a
> normal execution.
>
> ( This is just a ?guess?. I didn?t verify it. )
>

I did consider this possibility before I posted my build script and
searched for static (non-final) fields in JastAddJ. I didn't find any.
Maybe I just missed them, or maybe there is something else at play here.

It seems very likely that the state of JastAddJ is altered between
invocations.

By the way, I had a typo in the build script I posted here. There was a
$ missing in the -bootclasspath in front of OPENJDK6. When I restored
that, suddenly JastAddJ did not parse the source files but rather just
used the compiled class files from Open JDK. My results were then
similar to what you describe with the very short build time and very
many files not being compiled at all. If I remove the $OPENJDK6/... part
then the build time is at least 5 minutes. I do get a lot of NPEs but I
think this is because of missing class files from Open JDK (which is
still odd - missing class files should not cause NPEs).

/Jesper
Received on Thu May 03 2012 - 12:34:43 CEST