[Kde-perl] My response to comments

[Ashley Winters]

My request for comments was quite informative. I've come to believe
that Adam and Marcus were right on the mark as to how to implement Qt#
in the first place, before I brought up Smoke. QtC is exactly what they
need, but they needed it up to date with full functionality, including
virtual functions. As it is now, it's unusable. So is Smoke. However,
QtC is far closer to what they need than Smoke.

Now I see how QtC is the backbone of the language binding efforts...
Qt#, QtJava, basically any strongly typed language. If I want to put
forth effort toward making a "universal" binding package, it needs to
be for C.

Since Qt# is already abandoning QtC, perhaps I can interfere... err...
contribute to their design of the "new" QtC so Smoke can use it. Mind
if I call the new QtC "Puff"? It was Adam's name for the usable part of
Smoke.

So, instead of hacking into the gcc name mangling, you can just define
whatever C mangling system you want for Puff. As for virtual functions,
you *still* need to override them all, but you just provide
function-pointer-structs in C for those delegate pointers like QtC
should have done. Fun.

Before I go all technical and stuff, let me stop there. I had only
intended to say that I changed my mind about where the "shared" efforts
should go instead of Smoke. I would also recommend that a QtC wrapper
should be built over Puff after it's developed for compatibility.

That's enough for today.
Ashley Winters

__________________________________________________
Do you Yahoo!?
Yahoo! Mail Plus - Powerful. Affordable. Sign up now.
http://mailplus.yahoo.com