Alexandre Julliard julliard@winehq.org wrote:
I think that I've already described the technical details about both the source of the regression that Jacek is trying to fix, and the fact that the tests that could confirm that the fix is correct simply don't exist at all.
I haven't seen Jacek claim that the patch fixes the regression. I think we all agree that more changes are needed, with corresponding tests of course.
As far as I can tell, your objection is not to this specific patch, but to the fact that we shouldn't commit anything until we have a full solution? If so, I certainly don't have a problem with waiting.
My objection is that this patch pretends to be a partial fix for the regression while there is no any tests for the regression source at all, or the tests that actually demonstrate how all of this is supposed to work with correct implementation.
At the same time we see numerous claims that my tests are not good enough and the fix I've created for this regression is not correct, while there nothing has been shown as a viable alternative to them, just empty claims and hints (stay tuned, I'm working on something!) and accusations that I'm not understanding the proposed fix.