© 2004
Canonical Ltd.
•
Terms of use
•
Data privacy
•
Contact Launchpad Support
•
Blog
•
Careers
•
System status
•
1b1ed1a
(Get the code!)
Hello everyone,
From a release point of view this kind of problem is now under control.
Indeed we recently introduced new checks [1] to ensure to not falls in a similar scenario.
These checks are now merged.
Concerning the existing tags, those who are problematic, they have been generated before we adding these new checks. This is a side effect of the trailing project model. Indeed those projects are trailing projects and they have been forgotten from a release point of view during the previous trailing deadline of the two previous series (ussuri and victoria). We created new check/process to handle similar cases and ensure to never forget to branch trailing projects in during next series.
To get rid of the current issue I agree we could generate new tags based gitreview changes discussed in the previous comment.
This will lead us more or less to a patch similar to this one https:/
Our check [1] is based on pbr so we can't fail again on this point.
Concerning semver this is another discussion... Lot of projects release major versions for each cycle without releasing unbackward compatible changes...
This is an historical story, I'm not even have all the pieces of the puzzle who leaded us to this choice.
I don't think we want to mix the both topics and I'm not even sure that we can move away from this tradition (starting a new series with major version) (in a normal context without tag issue like here).
I seen many time people complain about this point in the past, I think that here are pro and con.
Maybe this could be a topics to discuss about during the next series. For now I think that we are too much closer from the final deadline to revolution this kind of tradition.
Thoughts?
[1] https:/
[2] https:/