Andrew Riedi wrote:
On Sun, Nov 9, 2008 at 6:25 AM, Krzysztof Drewniak krzysdrewniak@gmail.com wrote:
There is a .ANI cursor patch in hacks.git. Would it be possible to integrate it into wine? I know of many applications (especcially games) that would benefit. Thanks Krzys D.
Yes and no. Those patches were originally written by Henri Verbeet way back when (Thanks a bunch Henri, BTW). Since then I (and some other people) have merged what I was able to. As an example, 32-bit cursors from Wine beta. IIRC, the problem with the patches had to do with the wineserver code. The wineserver code needs to be altered to be more efficient and more correctly written. Unfortunately, I am not experienced enough to correct the wineserver code.
The trick is that to implement .ani cursors (without causing regressions in 16-bit programs or having ugly code) we need the cursors to be stored in the wineserver (which they are not currently). We also need the cursors to be in the wineserver for other reasons - I found this out when I wrote some cursor tests a while back. Anyhow, what I *can* do is strip out and clean up (so as to not have dead code) the code that loads the .ani, and then have it only load the first frame. This would basically just convert the .ani into a regular .cur cursor. There would be no fancy animation anymore, but should work for most purposes.
The question is then whether AJ would accept such a patch (as long as it was written cleanly). Of course, the patch would be written in such a way as to avoid making it harder to extend to full .ani support. (And be mostly Henri's code.)
I like that idea. I think that would help a lot of people with games that use .ani. Please post here when that's done