I'm very interested in upgrading Python to 3.7, so I'm willing offer any help I can. Probably "Python 3.7" hackfest is worth considering to happen :)

Oh, I see. In fact, this is completely true - library path changes, so all packages must be rebuilt.

In general, Python 3.7 introduces no breaking changes compared to Python 3.6. So, any Python code which works on Python 3.6 expected to work on 3.7 as well. So, IMHO, upgrading Python to 3.7 is pretty much safe operation. I also don't see the need of rebuilding _all_ the packages of python-* series - they should continue to work.

I'm going to listen to your advice, thanks.

I think this is clear in some way. If I would ask you about enabling kexec in shipped kernel, I would create another task, in appropriate tracker (or with appropriate tags). Starting from now I will be more precise in describing things.

Sorry about that.

I don't understand the complexity of having this package included in mainline repo... It's package.yml extremely straightforward. If you don't want enable kexec in shipped kernel - well, don't do that. If someone else decided to enable it - it would be helpful to have kexec-tools in repo rather than building it by hand.

I'm not talking about having kexec enabled (in shipped kernel, I mean)

Is is not so hard to build kernel with kexec enabled. And if someone do so - it is better to provide some batteries included for him. It would be pleasant.