Join GitHub today
GitHub is home to over 40 million developers working together to host and review code, manage projects, and build software together.Sign up
Dataset add `sources_policy` is confusing and should be split into two options #451
I'm talking about this one:
There are really two axis here
So this all quite confusing, but worst of all we bubble up that confusion all the way to the end-user, have a look at
@uchchwhash not sure I would do it at this level, wholesale lineage rejection is a useful "app-level" feature we should have as part of
This is index API level and at this point having an option like "just ignore missing lineage datasets and proceed like nothing happened" is not ideal.
But in general I agree that we need to have a serious discussion about lineage management options for summary datasets, as it gets quite unwieldy in that case. Also cross datacube data movement is made practically impossible because of the way we deal with lineage.
Users should not be calling