amantia added a comment.
I have changed the code to not rely on screen removal/addition signals inside the FolderView plugin, but use the existing code from shellcorona. This also ensures that we get the signals for the right id, and works correctly as long as the screen layout detection in shellcorona detection works properly. What I tested: - removal of a secondary screen from the right: the removal for id == 2 is emited - moving the secondary screen so it slightly overlaps with the primary: no removal signal is emited - moving the secondary screen so it completely overlaps, but it is aligned to the right of the primary screen (my secondary screen has smaller resolution): removal for id == 2 is emited - moving the secondary screen so it completely overlaps, but it is left/top aligned: removal for id == 2 is emited - moving the secondary screen so it is a little bit left of the primary, but mostly overlaps: no removal signal is emited. This is actually a problem as visually I see only the secondary screen on both monitors, but there is no removal signal for the primary screen (id == 0), thus plasma and folderview thinks the primary screen is still available. I consider this a bug in shellcorona, although probably it is a corner-case. For all cases I tested if the add signal arrives correctly when the secondary screen is moved back to the right of the primary one, and it does. Do you think this is acceptable? I have no idea about how wayland works, but if plasma shell does the right thing, the folderview should also do the right thing as well. REPOSITORY R119 Plasma Desktop REVISION DETAIL https://phabricator.kde.org/D8493 To: amantia, #plasma, ervin, mlaurent, dvratil, hein, aacid, davidedmundson Cc: davidedmundson, broulik, mart, plasma-devel, ZrenBot, progwolff, lesliezhai, ali-mohamed, jensreuterberg, abetts, sebas, apol
