[Bug 421] Implement a DIB engine

http://bugs.winehq.org/show_bug.cgi?id=421

--- Comment #101 from [e-mail]  2009-01-24 17:13:47 ---
(In reply to comment #100)
> Hi Max!  Thanks for posting that.  
> You've preserved Jesse and Huw's copyright notices, which is great,
> but it'd be nice if you also credited Jesse and Huw whenever you post this.
> Can you give a rough percentage of who wrote what
> (e.g. is it 20% Jesse, 50% Huw, and 30% you)?

Eh, the percentage.... quite difficult.
I started with Jesse's engine, then I added the complete Huw engine, changing
about all off Jesse's one besides parts on blitting (which were missing in
Huw's, but patched by me) and the big amount of stubs (quite comfortable to
have for testing, BTW), then I changed code again, moving driver loading from
SelectBitmap to CreateDC, CreateCompatibleDC and DeleteDC, added registry and
environment entries for enabling driver, corrected some stuffs which made it
crash if driver is not present/disabled, added the missing ExtTextOut() stuffs,
displaced color table grabbing hack to RealizeDefaultPalette() which is called
earlier.... Who knows ? :-)

If you look to code lines, I'd say 33 33 33, if looking on implemented stuffs
I'd say 25 30 45.... If looking on correctness, it's really a mystery :-)

BTW, I think the most important would be having it working :-)
I'm somehow in touch with Jesse, which (AFAIK) is building a GIT repository
with the new engine, even if I think the best would be to have it in main tree,
if possible.
Now that it is disabled by default it should do no harm at all to existing
applications (GDI with disabled dib engine behaves exactly as before), so it
should make no regressions at all.

I'd like to know Alexandre's opinion about it.

Ciao

Max