I don't know where the problem is, and I don't know where to report it upstream (found no contact anywhere and issues on its repo are disabled), but anki is dysfunctional for me.
I already recently had reinstalled the whole system and the problem didn't change.
$ anki
qt: QStandardPaths: XDG_RUNTIME_DIR not set, defaulting to '/tmp/runtime-user'
qt: QStandardPaths: XDG_RUNTIME_DIR not set, defaulting to '/tmp/runtime-user'
webview proxy use disabled
xuname: Void 4.19.14_1 x86_64-musl AuthenticAMD uptodate rF
Package: anki-2.1.8_1
Do not know if it will help, I do not use anki, but with other programs using Qt (specifically with qutebrowser) deleting the qtshadercache dir in ~/.cache would fix the problem.
Thank you for your suggestion, but I don't even have that directory.
I am unable to reproduce! With and without XDG_RUNTIME_DIR set...
That's strange and unfortunate. It must have something to do with musl, then, maybe?
That's strange and unfortunate. It must have something to do with musl, then, maybe?
Yes, qt5-webview is currently broken on musl.
While getting anki to run in a glibc-chroot, I found that it won't start without dbus installed. It tries to access /var/lib/dbus/machine-id.
You may be able to get by by just copying the file over...
That's not the problem though. Just noticing it seems to be a dependency, but it's not in the template. It's unrelated to the original problem.
It will also refuse to be run as root. Just taking notes.
(It does work in that glibc-chroot, by the way)
I am facing the similar issue. Has anyone found any fix yet?
I am not able to reproduce this on musl.
I couldn't even run it this time. Some missing qt library. I guess this template is incomplete. I'll look again once FF unbreaks (they share a library).
Thanks for the update, @PaperMountainStudio!
In the meantime, you can use flatpak.
Hmm, I tried this in a VM with only xorg, st and dwm. And still I can't replicate it.
Hey, it seems to work now! Awesome!
@PaperMountainStudio, thanks for reminding me.