RE: [abc-dev] Valid combinations of primitive pointcuts

 
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Sascha Kuzins wrote:
> Dear Eric,
>
> It is not entirely clear what you mean by "invalid". The example
> pointcut you present never matches, but that should be no reason to
> consider it invalid (what about the pointcut "if(false)", for
> example?).

Yes, you are making valid points here and the more I thought about
the problem, the more I began to agree with you. I had a look at the
document about the IR and it seems that conjucts of pointcuts
employing different StmtShadowMatches *may* be considered invalid. I
think that covers most cases. (e.g. a "field get" and a "call" can
never happen at the same time) Also I think multiple
BodyShadowMatches (again nonnegated in a conjunct) make no sense,
since e.g. you cannot be in an interface initilization and in a
method body at the same time.

In the case of similar matches (e.g. two "method call" matches as
before, I agree that this should indeed be no error since they *can*
actually be used in valid combinations.

Also this all should maybe be optional, but I think it would be a
nice thing to have for early error detection.

Cheers,
Eric

- --
Eric Bodden
Chair I2 for Programming Languages and Program Analysis
RWTH Aachen University

-----BEGIN PGP SIGNATURE-----
Version: PGP 8.0.3

iQA/AwUBQlp5dMwiFCm7RlWCEQJQXgCeNiVnDUR/dB/H8Ak5zlOtck6JVLwAn061
+BCWoZS5NAFkAa2dxAucH8Ku
=zWHr
-----END PGP SIGNATURE-----
Received on Mon Apr 11 14:20:09 2005