DINPUT: Generic FF effect implementation (FF #4)

[Alexandre Julliard]

Daniel Remenak <dtremenak at gmail.com> writes:

> Submitted #5 a couple days ago, along with a short explanation of how
> it all worked together.  Any comments?  If you'd like me to change the
> architecture I'm willing to do so, if you can give me some hints about
> what direction to go with it.  If you want some other changes made,
> please drop some clues about what they might be.  If you want to see
> more, I have more that can be cleaned up for submission.  If you need
> more clarification on some issue I can give it.  I just loathe sitting
> around wondering if these are good enough or not when I could be
> working more on them if they need revision or the next batch if they
> don't ;)

It looks a bit overdesigned to me. With this patch you add even more
infrastructure, generic interfaces, etc. and it still doesn't do
anything. I'd suggest going the other way around: implement one
working version of the code without worrying about being generic,
without trying to abstract interfaces or anything, just a simple
source file implementing the COM interface directly. Then if there's
really a need for other implementations to share a lot of code we can
look at abstracting parts of it.

-- 
Alexandre Julliard
julliard at winehq.org