This avoids having to hardcode profile targets in multiple places
and also fixes the confusing entry points into scripts both by
arguments and environment variables.
It also makes the setup script a lot more debuggable and scalable.
OS detection is a lot more robust, where it will try to use pip to
set up the distro python package, and if pip is missing try to install
it.
If OS detection fails now, a user can use --os on contrib/setup for
specifying it.
The end year is legally and functionally redundant, and more importantly causes
cherry-pick conflicts when trying to maintain old branches. Use git for history.
Sometimes it is desirable to create a build environment
outside of docker.
Move dependencies parser to a standalone python script
and call it from generate_docker.py
Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Fedora doesn't distribute docker anymore, instead
it uses podman for the containers.
It is possible to alias podman to docker, but
it's less hassle if it will work just out of the box.
The fix here is simple the podman is a fallback if
docker is not found.
Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
- Moved version discovery routine to PKGBUILD
- Set PKGEXT to .pkg.tar to avoid the package being compressed
- Added --needed to pacman arguments when installing the dependencies to
avoid reinstalling packages
Signed-off-by: Filipe Laíns <lains@archlinux.org>
There are a lot of hacks here;
* Pulling newer libappstream-glib from Fedora
* Pulling a systemd backport
* Manually installing pillow and pygobject
* PKCS7 is turned off (gnutls is too old)
Building:
The Dockerfile really is just an intermediary file. While building it the
container can be built too all in one shot. No need to git ignore the
intermediary file any more as a result.
Running:
Dockerfiles support default entry points which means one docker command
can be used for starting all of them.
