On Thu, 19 Jun 2025 10:10:46 GMT, Sergey Bylokhov <s...@openjdk.org> wrote:
> It would be good if we could eliminate the reference to WindowsButtonUI. But we want the regression to reproduce the original problem. The problem was specific to `WindowsButtonUI`, in `WindowsGraphicsUtils` according to comments in the bug. It may be impossible to reproduce the problem without referencing `WindowsButtonUI`. If it's possible, I'm all for it. > It might also be useful to check the behavior across all L&Fs. The bug was in Windows L&F, I don't think it's applicable to other L&F. On Windows, the disabled text was painted with “shadow” with offset of (1, 1) which could be painted over another part of the button. Other L&Fs are free to paint the button differently. ------------- PR Review Comment: https://git.openjdk.org/jdk/pull/25883#discussion_r2156668056
