Community
Participate
Working Groups
I know from experience, one of the most frequent "user errors" made in updating aggregator models for the yearly Simultaneous Release is that people add or remove the "contacts" element in their b3aggrcon file, but forget to (or, do not know to) update with the model editor, so that the change is made in both the main file (simrel.b3aggr) as well as their projects b3aggrcon file. Is it really required to define the model this way? Or is there some use-case I'm not aware of? Seems it would be easier if indeed users could just update their b3aggrcon file, instead of having to touch the main file, just to change a contact person.
The general idea was that the set of contacts was managed globally for the aggregation and that each contribution then could select from that global set. I agree that this is a bad fit for the release train use case where the contributions reside in separate files. It should be easy enough to change so that the contact is contained in the contribution rather than referenced and add a compatibility layer. Once we do, we must ensure that all b3aggrcon files are changed simultaneously since the change will affect their content as well as the master.
[Bookkeeping change only. Moving bugs to the new "home" of aggregator, CBI.]
If this issue is still relevant, please move it to https://github.com/eclipse-cbi/p2repo-aggregator/issues.
