Re: Lightweight peer implementation

["Scott Gilbertson" <scottg at mantatest dot com>]

> > Bryce wrote:
> >> Yup, I agree. Lightweights should be implemented entirely in the
> >> "generic" part of the code, not in the peer code.
> > Thanks, guys.  I've been trying to get someone to express an opinion
> > one way
> > or the other on this issue for some time.  I look forward to getting
> > the
> > lightweight components issue off my "pending" list.
> >
> > I agree with all of what Bryce said:
> >  - lightweights should be generic
> >  - there should be no need to check for a null peer
>
> Well, I actually said there might be reason to check for null peer. In
> the JRE implementation, creation of the peers doesn't happen until the
> component is added to a container and/or its made visible. Just
> guessing - but this might be because some windowing systems don't allow
> the native widget to be created unless it already has a parent?

An alternative would be to have a single static instance of a "really
lightweight" peer, which returns nulls for everything, and make that the
Component's peer until the real one is set.  You'd get a slight performance
boost, by eliminating a comparison statement in the common case where the
peer has already been set up (at least on architectures where a virtual
method call is faster than a conditional branch).