Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [aspectj-users]"can't determine precedence between two or more pieces of advice that apply to the same join point"

Eric
You may read more on a thread on this topic that I had started in
January, based on the difference with AspectWerkz semantics where we
have implemented what you suggest instead of the current AspectJ
semantics.
http://dev.eclipse.org/mhonarc/lists/aspectj-users/msg03412.html

One interesting topic: try to ask yourself what should happen when you
have an around advice defined somewhere in your aspect for the same
join point, and wonder if that 'd be good if ones could not surround a
before advice and the shadow with an around advice f.e.

Alex


On 5/7/05, Eric Bodden <eric@xxxxxxxxx> wrote:
> 
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> Wes Isberg wrote:
> > The behavior is well-defined, but the programmer can define a
> > precedence cycle. See circularity discussion in programming guide
> > semantic appendix section on advice precedence:
> >
> > http://www.eclipse.org/aspectj/doc/released/progguide/semantics-advi
> > ce.html#d0e6220
> Thanks for pointing that out, Wes.
> 
> > (A good rule is to always put after advice last.)
> Well, given that, would it not be more intuitive to execute *all*
> before advice matching one joinpoint simply before the joinpoint in
> the order in which they appear, followed by the shadow, followed by
> all after advice that match this joinpoint?
> 
> Eric
> 
> - --
> Eric Bodden
> Chair I2 for Programming Languages and Program Analysis
> RWTH Aachen University
> 
> -----BEGIN PGP SIGNATURE-----
> Version: PGP 8.0.3
> 
> iQA/AwUBQnygEswiFCm7RlWCEQJSPACgyQ22erpCGUVrnEZqNR11tGXzGcgAn3jN
> hUa/folTblvCnOYH6Hk+TFdZ
> =hCwN
> -----END PGP SIGNATURE-----
> 
> _______________________________________________
> aspectj-users mailing list
> aspectj-users@xxxxxxxxxxx
> https://dev.eclipse.org/mailman/listinfo/aspectj-users
>


Back to the top