https://bugs.kde.org/show_bug.cgi?id=354724
--- Comment #46 from Thomas Lübking <thomas.luebk...@gmail.com> --- Well, "shrug, I don't care then" - Qt seems to be dropping desktop support anyway, so it probably doesn't really matter. For principal reasins I'd however object " proven to be not broken by being around for over 10 years": It wasn't "broken" because one major client class (KDE) simply replaced that "unbroken" code. Nor did I mean to support directly "changing behavior" - I simply suggested to raise awareness of the problem by intercepting "this looks unimplemented" conditions to allow getting rid of the "unbroken" code in Qt6 w/o pain. Closing windows certainly is wrong (and apparently wasn't done before Qt4, seems more like a closeEvent was faked previously) because it alters the session even on a canceled session exit, regardless on whether it breaks restoring them (which one would kludge away in the session manager) Ok, bottom line is SM is FUBAR on Qt5 and there's no intention to ever fix that, merely re-allow client code to bypass the broken Qt behavior. I'll tag future bugs respectively. -- You are receiving this mail because: You are watching all bug changes.
