Hi team,
I want to clarify and document our branching and release strategy
to set the correct expectations for different distributions and future
389-ds-base maintainers.
Recently, I assumed that we have this branching strategy:
- The development happens in the *main.* We cherry-pick the needed fixes to
other branches as required.
- When we release a new *X.Y *version (bump version, create tag, create
release), we create a separate branch *389-ds-base-X.Y.*
- The main at this point becomes future *389-ds-base-X.Y+1* (without
bumping the version there, for now), and we continue the development there
while cherry-picking to other branches + 389-ds-base-X.Y.
It ignited a discussion... So please, if you disagree with the above,
describe the release flow you think we have/should have.
We must make it very transparent for other distributions besides
RHEL/Fedora.
Also, an important question will be:
- Do SUSE/Debian/Archlinux/etc. need the *389-ds-base-X.Y *branch created
right away after the *X.Y* official GitHub release? Or can it be postponed?
Please, maintainers, let us know.
Regards,
Simon