Closed dvdesolve closed 4 years ago
Hi @dvdesolve ,
I don't know exactly the reason for the first error, looks like there is a problem with the possibility to change the font added in the new release.
To compile from source: actually I still have to try Python ver. 3.8 to run Artemis. I'm not sure all the dependencies are satisfied. Apart from that, note that the PyQt version you are using is not the one specified in the requirements.txt file. In particular note that in your PyQt version the 'exclusive' parameter is in fact not available: https://github.com/metabrainz/picard/pull/1424
I suggest to build a Python 3.7.0 environment and install the dependencies as per the requirements.txt file and try again.
I can't figure out why it complains about canberra-gtk-module
- I already have libcanberra
installed:
[viktor@desolve-nettop ~]$ pacman -Ql libcanberra | grep canberra-gtk-module
libcanberra /etc/X11/xinit/xinitrc.d/40-libcanberra-gtk-module.sh
libcanberra /usr/lib/gnome-settings-daemon-3.0/gtk-modules/canberra-gtk-module.desktop
libcanberra /usr/lib/gtk-2.0/modules/libcanberra-gtk-module.so
libcanberra /usr/lib/gtk-3.0/modules/libcanberra-gtk-module.so
Regarding GdkPixbuf error: seems like path /usr/lib/x86_64-linux-gnu/gdk-pixbuf-2.0/2.10.0/loaders.cache
is hardcoded in binary somehow so it tries to look in non-existent place.
Don't know why there are errors about Fontconfig but how can I figure out the cause of the segfault?
Thank you for Python 3.7 suggestion. I'll give it a try. According to your link it seems like exclusive
parameter is not needed anymore. Will new release have support for Qt 5.14?
The cause of the segfault seems to be in the setupUi function, which is a PyQt function. We tested the binary versions on Linux Mint and Fedora, so I think it is possible that the compiled version could not function properly in all linux distros.
I would try to just run the script python artemis.py
first (given you are in a proper python environment with all the requirements satisfied). If that works, I'd try to compile from the .spec file.
About the Qt 5.14 support: chances are that the exclusive
parameter may be removed also in the currently supported version. I will try to see if everything works the same and eventually remove it.
Ok, thank you PS: I always thought that binary version depends almost on nothing and self-consistent as much as possible. So why PyQt interferes in that case? Also I remember that older version (3.0.1) worked just fine
Not sure if it helps, but I modified the code to remove the 'exclusive' parameter
Artemis/src/themesmanager.py line 143
old: ag = QActionGroup(self._owner, exclusive=True)
new: ag = QActionGroup(self._owner)
and so far it seems to be working fine. (Fairly new Artemis user, so I'm not sure if I'm missing something).
Trying to run it different ways. I'm on the latest kernel, Python version is 3.8.1 and Qt is 5.14.1. All requirements are installed.
Binary from aresvalley.com
When running from terminal I can see splash screen for a moment and then nothing happens. In terminal output I see the following messages:
Should note that there is no
/usr/lib/x86_64-linux-gnu
directory exists at all. Also version 3.0.1 binary worked fine.Building binary from source
I've downloaded the same 3.2.0 release source code from GitHub and tried to build from
Artemis.spec
file:When trying to launch
dist/Artemis
binary I have the following error message:I can workaround it by changing
hiddenimports=[],
line to thehiddenimports=['packaging.requirements'],
inArtemis.spec
. However I can't still run it after rebuild because of the following: