[systemtray] Keep config entries after modifying
ClosedPublic

Authored by anthonyfieroni on Nov 15 2018, 8:00 AM.

Details

Summary

It still has one problem, after configuration is changed in Entries MouseArea index is invalid, i can't understand why, @broulik ?

BUG: 390890

Test Plan
  1. Right button on systemtray
  2. Check / uncheck extra imtem
  3. Click on Entries
  4. Show / hidden some of them
  5. Return to General

again 2
again 4

All good, but after close configuration dialog clicking on item in systray is an invalid index.

Diff Detail

Repository
R120 Plasma Workspace
Lint
Automatic diff as part of commit; lint not applicable.
Unit
Automatic diff as part of commit; unit tests not applicable.
anthonyfieroni created this revision.Nov 15 2018, 8:00 AM
Restricted Application added a project: Plasma. · View Herald TranscriptNov 15 2018, 8:00 AM
Restricted Application added a subscriber: plasma-devel. · View Herald Transcript
anthonyfieroni requested review of this revision.Nov 15 2018, 8:00 AM

Weird. If configuration window stays open, clicks in systray work correct, when it's dismissed clicks are incorrect.

I think that's a Qt bug, i did an investigation in another systray bug report but can't find it right now

Weird. If configuration window stays open, clicks in systray work correct, when it's dismissed clicks are incorrect.

That's a separate bug: https://bugs.kde.org/show_bug.cgi?id=393630

ngraham accepted this revision.Nov 18 2018, 3:54 AM

This does fix 390890 for me though! Everything else still seems to work.

This revision is now accepted and ready to land.Nov 18 2018, 3:54 AM

I don't understand.

The only usage of plasmoidItemComponent is in " Containment.onAppletAdded:", so how is this going to make any difference?

But for some reason, on second Entries show warnings says: cannot call createObject on null object. Looks like it is treated as local object or something.

trmdi added a subscriber: trmdi.Nov 28 2018, 1:35 PM

To commit it @mart @broulik ?

I think that is not the root cause. You should fix the root cause instead of just a part of it.
See my comment on https://bugs.kde.org/show_bug.cgi?id=393630#c25

Yep, it's Qt bug.

If it's a Qt bug, our options are to fix it upstream, or work around it here with a comment in the code indicating the workaround and the Qt bug report. Can you add that?

davidedmundson accepted this revision.Jan 14 2019, 5:30 PM

The bug has been reported before!

https://bugreports.qt.io/browse/QTBUG-71238 / https://bugreports.qt.io/browse/QTBUG-72004

Ignore the title and context, read the comment from Kai about the component without a parent getting cleaned up.

That moves this from working round an "unknown", to working round a "known" issue.
Seems a tiny difference, but means a lot to whether I accept it.

@anthonyfieroni can you add a comment in the code that indicates the Qt issues this is working around?

We've got three days until 5.15 tagging. Let's try to get this in!

This revision was automatically updated to reflect the committed changes.