https://bugs.kde.org/show_bug.cgi?id=382604
--- Comment #18 from Colin Griffith <tyna...@gmail.com> --- It looks like the code was ported to C++ from a bash script. The relevant code is here: https://invent.kde.org/plasma/plasma-workspace/-/blob/master/startkde/startplasma.cpp This was done in this commit: https://invent.kde.org/plasma/plasma-workspace/-/commit/5df9f0bcc9259e595d97d0e55399753325f7a010#31c42a873c0b74c54dc71f1f4a2ca9a6f7643718 And the code was just copied mostly without changing what it actually did, which is why the behavior didn't only start happening in 2019. For that, I had to dig REALLY far back... And eventually found this in the old SVN repository: https://websvn.kde.org/?view=revision&revision=396433 And in particular, here: https://websvn.kde.org/trunk/KDE/kdebase/startkde?r1=396411&r2=396433&pathrev=528175 Way back on March 10th, 2005. It looks like the reason the environment variable was set back then was for kded and ksmserver. Is this still the case today? It seems like code that old working around problems in other packages (albeit other KDE packages) might not even be relevant anymore, and so safe to pull out. Additionally, I don't have KDE's source code installed on my drive, so I can't actually try changing this, compiling, and testing... Hence just talking on this bug report instead. It could be that this isn't at all related to the bug we're experiencing, and I just went on a wild goose chase for nothing. Still, it would explain the current behavior for the most part. -- You are receiving this mail because: You are watching all bug changes.
