[1/2] d3dx9: Implement the different dlls [try 2]

[Alexandre Julliard]

tony.wasserka at freenet.de writes:

> Okay, I now have spent some time on analysing the differences of the dlls
> (at least regarding to what functions they provide). So just a few questions
> from my side to make sure the next version of the patch will be what you expect:
> 1. Really just to ensure that one: You DO want me to forward even the stubbed
> functions to the "d3dx9_36 stubs", don't you? If there's more to the forwards,
> then tell me please.

All functions that can be forwarded must be forwarded, you don't want to
have to go back and fix 12 spec files every time one is implemented. If
some functions are not compatible then of course they shouldn't be
forwarded. You need to verify this for each function.

> 2. Regarding the copyright thing, I'm really sorry for ignoring David's copyright, I'll
> change that back. However, you didn't mean the git changelog with
> "use his name in the changelog", did you?
> Well, if you did, It'd be possible to mention both, as I'll have rewritten all
> the specfiles. (It's not that I wouldn't respect David for his work, but the
> time I spent on this shouldn't be left out either)

If you have really rewritten them, then of course they are copyright by
you. If all you did is s/d3dx8/d3dx9_36/ (as seems likely since you have
the exact same bugs) then David should be credited for them.

> 3. When it's ready, should I post my patch to a pastebin and ask at #winehackers
> first if it's ok like that?

Yes, and then you should submit first a standalone d3dx9_36, and once
this is applied, a single other dll like d3dx9_24, and not worry about
the other ones until you can get that one in.

-- 
Alexandre Julliard
julliard at winehq.org