Under certain circumstances, changing the client geometry in a script (like one made for tiling window management) may lead to corrupted visuals (the former visuals are filled in to the new size) and unresponsive windows. This happens mostly when changing geometry in the handler for client.geometry{Shape,}Changed, but can also happen when changing the geometry too early after the client is just added or when violating the clients min/maxSize. Clients particularly prone to this are vlc (when violating minSize, which is what the linked script does for ease of reproduction), firefox (when restoring a session) and emacs (pretty much always). Reproducible: Always Steps to Reproduce: 1. Run the attached script 2. Open vlc 3. Resize vlc Actual Results: The old client image is filled in to the new client size and kwin shows _tons_ of >Creating window pixmap failed: SOMETHING in the console. Expected Results: The client either sticks to its old size and the script is notified (by throwing an error it can react to) or (my preferred solution) the client is properly resized to the geometry the script specified. As far as I remember, this has already been present in 4.9 and is also present up to and including 5.1.1. I would also be okay with an API change, i.e. providing a "client.setGeometry(rectangle)" method that provides a success return value instead of directly accessing the property if that helps.
Created attachment 89700 [details] test script
Just to be sure: happens with oxygen decoration, not aurorae?
Yes, I just tested again to be sure. It also happens with client.noBorder set to true (that's what I use for the most part), though I'm not sure how much of the decoration code that disables.
if it's about violating minSize: we should probably not allow to do that from the script.
Just to be clear: It can also be triggered in non-minSize related cases. It's just that that case is the easiest and most reliable to reproduce. Another requires resizing firefox after it tries to resize to the value from the previous session, which it only does when it shows about:sessionrestore. There may also be some races involved - as far as I can see in the firefox case creating the pixmap fails because the geometry was changed between kwin requesting and receiving a pixmap (i.e. the important test seems to be "windowGeometry->width != toplevel()->width()" or the same for height in scene.cpp). This may require some ordering guarantees for the scripting API - something I already mentioned regarding FFM (in a bug I can't find right now).
Yes, that certainly needs better improvement. I wanted just to point out the obvious: resizing below minSize doesn't make much sense.
*** Bug 365821 has been marked as a duplicate of this bug. ***
This issue report is quite old. Can you please confirm, that it still persists with KDE 5.23?
Dear Bug Submitter, This bug has been in NEEDSINFO status with no change for at least 15 days. Please provide the requested information as soon as possible and set the bug status as REPORTED. Due to regular bug tracker maintenance, if the bug is still in NEEDSINFO status with no change in 30 days the bug will be closed as RESOLVED > WORKSFORME due to lack of needed information. For more information about our bug triaging procedures please read the wiki located here: https://community.kde.org/Guidelines_and_HOWTOs/Bug_triaging If you have already provided the requested information, please mark the bug as REPORTED so that the KDE team knows that the bug is ready to be confirmed. Thank you for helping us make KDE software even better for everyone!
This bug has been in NEEDSINFO status with no change for at least 30 days. The bug is now closed as RESOLVED > WORKSFORME due to lack of needed information. For more information about our bug triaging procedures please read the wiki located here: https://community.kde.org/Guidelines_and_HOWTOs/Bug_triaging Thank you for helping us make KDE software even better for everyone!