@algent mutagen also uses pyflakes for its code quality check, but it has no functional dependency on pyflakes. mutagen builds successfully and its python3 part of unit tests pass. python-pycurl also only uses pyflakes for unit test. I am able to build python-pycurl successfully. Its unit tests failed. However, this failure happens regardless of whether I use local repository or not, so I don't think it's related to the upgrade of pyflakes.

@algent autopep8 is also updated to have compatible versions with pycodestyle. mutagen uses pycodestyle and pyflakes only for code quality check, but no functional dependency. mutagen builds successfully with new versions of pycodestyle and the python3 part of unit tests passed. python2 part of unit test fails because the drop of python2 support in pycodestyle.

Tested against pylint. Updated test plan accordingly.

Switched source to pypi since the github source does not contain a setup.py file.

@serebit Some suggestions to this patch:

Now that LLVM 10.0.1 is out, hopefully we can see it being upgraded in solus repo very soon.

@g66925 As your link mentioned, you can try switching to strongswan. You can do that with the following commands:

@g66925 Just so you know, the patches have been landed. You should be able to find them in the unstable repo now.

@g66925 They have not been landed. It would be better if you test them and verify that they work as expected before landing these patches. If you feel comfortable, I can build these packages and send the resulting eopkg file for you to test. It is also understandable if you prefer to build it yourself. In the latter case, you can download the aforementioned patches and build them locally. This is a very good introduction to setting up solbuild: https://www.youtube.com/watch?v=rlPnHjUBpJ8

Add a child revision to its stack

@g66925 D9261 and D9262 are relevant patches aiming to fix this issue. Can you try them out?

I have to say I have the same issue. I also confirmed that using mbedtls as the crypto-library caused the issue.

Closing as all patches are merged.

Fix indentation of package.yml

Understood.

Add ldfalgs

Add MAINTAINERS.md

Kindly poking @JoshStrobl for any feedback on this patch.

I guess I can try updating this but my employer has since changed to use openVPN so I won't be able to test things as I did before.

Move the homepage entry to right after source in package.yml.

Add MAINTAINERS.md

Add MAINTAINERS.md

Undo editor formatting.

@JoshStrobl The symlink /etc/xprofile is now removed. lightdm will source /usr/share/defaults/etc/xprofile, as well as /etc/xprofile and $HOME/.xprofile if they exist. Once you patch gdm and sddm in the same way, the three display managers should behave in a consistent manner.

@JoshStrobl I made lightdm source the vendor-provided xprofile and removed the symlink of /etc/xprofile to /usr/share/defaults/etc/xprofile in D8989 for a stateless setup.

Make lightdm source /usr/share/defaults/etc/xprofile for a stateless setup.

Do not symlink to /usr/share/defaults/etc/xprofile.

Revert last change.

Remove symlink to /usr/share/defaults/etc/xprofile.

@JoshStrobl I will remove the symlink line and update the other patch accordingly (line 34). So I guess we should not land this until that is done.

@ermo Yeah I totally understand that. There is no pressure from me on you and the Core team. I believe this will eventually get resolved.

@ermo I made some modifications to D8989 based on your suggestions. I do think this is a better way to organize things.

Add system-wide xprofile and user-specific xprofile.

@ermo That sounds a good idea. One question is which package /usr/share/defaults/etc/xprofile goes to? Still baselayout or some other package?

@ermo Yeah understood. If there is anything else I can do to improve the current package, just let know.

@ermo I went ahead and created D8989 for adding default xprofile. Let me know if you have any suggestions.

Forgot to bump release.

@ermo I also cooked up a .xprofile and tested it on all four DEs. My question is, how should I proceed to create a patch for that? Should I create a patch for baselayout? My concern is that I have no way to test whether the new baselayout package works as intended.

@ermo I have moved the part of loading key maps out of lightdm-wrapper so I think this D8984 is ready for review again.