Perfect. Darn, I need to brush up qi4j again. I'm going to Brisbane for 3 weeks next week, I'll have plenty of time during cold winter night.
Regards, Edward Yakop On Tue, Jul 6, 2010 at 23:09, Rickard Öberg <[email protected]> wrote: > On 2010-07-07 13.34, Edward Yakop wrote: > >> In general, to keep things aligned with DCI, I think I'd prefer >> "withRoles" instead of "withAbilities" as previously discussed. What >> do you think? >> >> It feels more natural in terms of UL. >> Do you think that we needs to stop at withMixins() ? >> How do we address "For some roles, we would need x Concerns and x >> Side-effects to be declared?" >> > > I think this will be orthogonal to the already existing functionality. We > already have ability to declare additional concerns, side-effects and mixin > implementations. All we need is to be able to add more interfaces (=roles) > that we want the composite to implement. > > So the main difference is that when we resolve the composite, instead of > just implementing the composite type, we also add the additionally declared > roles, and then the usual algorithms for selecting mixin, concerns and > side-effects apply. > > > /Rickard > > _______________________________________________ > qi4j-dev mailing list > [email protected] > http://lists.ops4j.org/mailman/listinfo/qi4j-dev >
_______________________________________________ qi4j-dev mailing list [email protected] http://lists.ops4j.org/mailman/listinfo/qi4j-dev
