marco / mate-session-restore misplaces windows

[pF-arQon]

Expected behaviour

Windows appear in the right place after logout/login

Actual behaviour

Windows appear in the wrong place

Steps to reproduce the behaviour

Log out with session restore enabled and a terminal window in the bottom right corner. Log back in. Look at the terminal window.

MATE general version

1.26

Package version

1.26.0-1

Linux Distribution

Ubuntu 22.04

Link to bugreport of your Distribution (requirement)

https://bugs.launchpad.net/ubuntu/+source/marco/+bug/1845822

ugh - I fixed this bug last year, but apparently ran out of time to file all the duplicate bug reports. Patch is in the original thread.

[lukefromdc]

Did a fix get committed here, or was that just a patch in Ubuntu?

[pF-arQon]

The patch is against the MATE source, as documented in great detail in the linked thread, as well as the parent hash in the patch itself. I'm not even sure I understand the question TBH, but - and I mean this with no sarcasm - did you actually read the thread?

There's a limit to how much time I have available to investigate bugs, fix them, file reports in triplicate, provide patches and PRs around the net, and chase people to actually apply them; and unfortunately I've reached it. If Ubuntu MATE no longer has any maintainers, I can't help with that. I wish I could. If bugs weren't managed properly during handovers and site changes, I get it, but I can't retroactively fix that. If the request is that I burn hours more of my time to save someone else literally seconds of non-effort, then consider me appropriately insulted but I don't have the time for that either. I've done the work to file (repeatedly), investigate, and fix a bug that is now over 4 years old, and that's where my involvement is going to have to end for now. No offense, but even if I had more free time I've already more than pulled my weight on this.

If nobody knows how to apply a git patch outside of a web interface then, well, I don't believe that :P, but it'll still have to wait until life cuts me enough slack to babysit things. Hopefully that'll just be a couple of months rather than a couple of years again, but either way it's not going to be any time soon. (and it'll miss the Ubuntu LTS anyway at this point, so there's no real urgency to it).

If you (or others) have questions about the patch itself that aren't already answered, ask away and I'll respond as best I can when time permits. gl.

[lukefromdc]

While I work on MATE, I am NOT one of the maintainers of Ubuntu-MATE ( I run a modified version of Debian) and do not know who is currently bottomlining Ubuntu-MATE. Also note that our team is small (we need more people) with zero paid staff and having to harvest patches from around the Internet adds a lot of work and significant time. A patch against any distro version may not apply clean if the distro added any patches, and may have to be done by hand.

I usually do NOT read far into reports on Launchpad etc unless I suspect an issue is distro-specific. Someone else here might, but I do not. Since marco is not one of the packages I normally focus on (normally using compiz on x11 and now wayfire on wayland) I don't know my way around it as well as others here might. Session saving is also not part of my normal workflow and as a result, it is an additional session shutdown and restart to test it here.

I am working on a live, production system with zero VM experience too. To test this, I would need to kill a session, including my delberately stateless firefox setup, open another session and start everything over. Then repeat at least twice to test session saving. I thus usually leave this part of the job to others who can test this on VM's or otherwise not interfere with all their other work.

Generally, the procedure for upstreaming a distro-level patch is to open a PR here. I suspect that's true of almost every package from the kernel on down.