Re: [ajdt-dev] strange problem using AJDT with Eclipse 3.0 m4

[Martin Lippert <lippert@xxxxxxx>]

Hi Ron,

> I found that using incremental compilation in Eclipse 3.0 m3 broke badly for 
> me (and submitted a bug report). 

I have the same problems running AJDT 1.1.4 without incremental 
compilation. The same wrong output is produced.

> I believe the problem is that the AspectJ compiler reuses some of the JDT 
> compiler classes and it has to use internal interfaces; so a major shift like 
> from 2.1 to 3.0 requires work on the compiler core. There was a need to work 
> on this in the move from 2.0 to 2.1.

Ah, I see. Is there anybody working on this at the moment? I would love 
to use m4 with AJDT for my OOPSLA demo. Any chance?

Best regards,
Martin



> On Thu, 16 Oct 2003 21:11:03 +0200, Martin Lippert wrote
>
> > Hi,
> >
> > I have a very strange problem using AJDT 1.1.4 with Eclipse 3.0 m4. 
> > I compiled an aspect that contains something like:
> >
> > thisJoinPoint.getSignature();
> >
> > The compiler output for this (using the -noweave option) is wrong. 
> > The bytecode for the compiled aspect contains a call to the class 
> > org.aspectj.lang.JoinPoint.getSignature (instead of 
> > JoinPoint$StaticPart.getSignature which would be right).
> >
> > Now the strange thing: If I compile the same aspect using AJDT 1.1.4 
> > within Eclipse 2.1.1 the bytecode of the compiled aspect is correct.
> >
> > Any idea on that??????
> >
> > Best regards,
> > Martin
>
>
>
>
> Ron Bodkin
> Chief Technology Officer
> New Aspects
> m: (415) 509-2895
> _______________________________________________
> ajdt-dev mailing list
> ajdt-dev@xxxxxxxxxxxxxxx
> http://dev.eclipse.org/mailman/listinfo/ajdt-dev

Attachment: smime.p7s
Description: S/MIME Cryptographic Signature