On Tue Feb 25 20:39:47 2025 +0000, Nikolay Sivov wrote:
> This isn't about percentage or how common it is. It's not clear this is
> the right direction to me, for example groups could contain path
> geometries with different filling modes, or other groups.
So, what is the behavior you expect? It is clear the behavior is correct when the fill mode is WINDING, or when the rendering does not depend on the fill rule - that also includes overlapping geometries with correct path direction, e.g. paths generated from font outlines.
You should provide a way forward, not block it because it is not perfect. It definitely *does* work properly in the majority of cases, the rendering becomes more correct. If you want changes, please be specific.
Wine is full of partially incomplete and incorrect implementations, according to your line of thoughts all this should be deleted.
--
https://gitlab.winehq.org/wine/wine/-/merge_requests/6492#note_95959
On Tue Feb 25 20:02:24 2025 +0000, Elizabeth Figura wrote:
> > No, I think it is desirable. I'm going to move some more tests out, to
> build with either msxml2.idl or msxml6.idl, and then we'll see what's left.
> What's the progress on this? Is this something that could be delegated
> to someone else invested in upstreaming wine-staging patches?
I sent some more changes in !7427. But sure, we might need to duplicate some things between dlls too.
--
https://gitlab.winehq.org/wine/wine/-/merge_requests/1060#note_95957