On Sun May 11 15:06:09 2025 +0000, Connor McAdams wrote:
changed this line in [version 4 of the diff](/wine/wine/-/merge_requests/7711/diffs?diff_id=177017&start_sha=3ee9cf5fc6a568625d0fa74f58976d89fdbb09f2#8fffeae09861a99974e55842964630933974cdc8_2432_2443)
So, IIRC, I originally did have it the way you're describing here and then I later changed it once I remembered we always initialize missing channels to the maximum value (0xff) when we convert to the color key format. So I think either way would work, having a range of `0xff->0xff` or `0x00->0xff`. But, having it as `0x00->0xff` would make it clearer that we're checking against a channel that isn't present in the source format.
I might be missing something though, I can't think of a case where this being `0xff->0xff` might fail.