Cristian Le

> If any critical issue will be found with last supporting 3.7 python version always on top of the version tag is possible to add branch and add fix issue...

> For example in Fedora only ~1/3 of all python modules are executing included test suites. It's a bit complicated because: tests mightbe run from `tox`, unit-test (files directly), outside...

> > It's a bit complicated because: tests mightbe run from tox, unit-test (files directly), outside with tmt > > Trust me .. it is not 😋 I was referring...

It's mostly up to how your developers tend to work. E.g. sometimes it's useful to have the build directory be python-version dependent, sometimes you want a predictable path so that...

Does it work when run locally outside of `SKBUILD`? A dockerfile or github wotkflow would be helpful to navigate the build log

Yes, this is bound to happen because the cmake file-api does not follow the latest cmake version, nor is it version aware. I've tried to look into how to do...

When you said you tested with plain CMake and it worked fine, can you be more explicit on the steps there? Did you do an install and tested from those...

> I'm not sure what you were referring to by transient link failures. I mean issues where libA compiles with dependency of libfoo but when you are building libB that...

@martinling did you check my comment in https://github.com/greatscottgadgets/hackrf/pull/1514#discussion_r2214442243. The `...` is the recommended setup to keep backwards and forwards compatibility specifications when you have a CI testing multiple CMake versions....

> Setting that kind of compatibility range would make CMake's behaviour differ more between versions, more so than we're really set up to test for effectively. For the most parts...