I also can't reproduce the issue anymore. Great work, thank you!
I agree. Sadly, also no new discoverys on my side regarding bug #1319.
Full ack. However, as you said, this is quite some effort. I think it is worth to silence the GTK assertion until then to not confuse users. We could add a comment saying that this is only a workaround?
Yeah, some pointer gets corrupted. Error messages I also saw indicating this: "free(): invalid pointer" and "Bus error" (points into the void). Bug #1438 is probably a duplicate of this, I'd think.
I found this yesterday. Similar problem to patch #292 which I just submitted. The UI is not even there when the PSID is loaded. Also, there is currently no interface to pass the needed filename to the UI. I guess this can be added but I wonder if this is the way to go.
Better patch, avoiding the sprintf...
GTK3: VSID: fix GTK3 assertions
For the record, my Debain (Bullseye) also doesn't have libxcb-util0 anymore but only -util1. Maybe time to update the Debian build machine? :)