On 12/19/18 15:16, Alexandre Julliard wrote:
Gabriel Ivăncescu gabrielopcode@gmail.com writes:
Cause of bug discovered by Jason Edmeades.
Some applications partially overlap the two buffers. Handle such rare corner cases without affecting performance in the general case.
Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=38558 Signed-off-by: Gabriel Ivăncescu gabrielopcode@gmail.com
There are no risks of regression here, as the only case where it is slower now is when previously gave a completely wrong result.
I assume performance is important for this function, since it had the switch statement in the first place...
Yes, that's why it would be better to do it the same way as the wcstombs case.
Note that unless you can show that it affects more apps than the one from the bug report, it will still have to wait until after code freeze.
Oh ok, I'll revise and send it after the code freeze then.