Re: [abc] oopsla outline

On Tue, Mar 07, 2006 at 10:21:16PM +0000, Pavel Avgustinov wrote:
> Ondrej Lhotak wrote:
>
> >Is the reason that most of our tracematches cannot be implemented in PQL
> >that there is no way to implement anything like filtering? i.e. there is
> >no way to say "an event, with this particular binding, does not occur"?
> >
> >If so, then this prevents implementation of the reweaving tracematch
> >in PQL.
> >
> >
> No, that is not the reason. Indeed, you can explicitly forbid events
> (with bindings) by prefixing them with ~, so ~event(binding1) could be
> used to get the effect you want.
>
> The reasons a lot of our examples don't work in PQL are actually quite
> varied. Nulltracker fails due to PQL's lack of constants, hashcode --
> due to its lack of support for primitive bindings. LOR can't talk about
> threads.

Hmm...

This means I'm still looking for a convincing reason why reweaving
cannot be implemented in PQL.

Is there a way to implement Julian's poincut:
    pointcut field(String f) :
        let(f, thisJoinPoint.getSignature().toString().intern());
i.e. I want to match a read/write of an arbitrary field, and get some
string to identify which field it was. I don't see any way to do this
in the PQL paper. Am I missing something?

Ondrej

>
> - P
>
Received on Tue Mar 07 22:43:43 2006