Fixes: https://pagure.io/koji/issue/2780
:thumbsup:
I don't think that readTaggedBuilds needs a with_owners option. This is a different owner field -- package owner vs build owner. It should be enough to just pass with_owners=False to readPackageList here.
readTaggedBuilds
with_owners
Cheers for QueryProcessor conversions, but I don't like splitting up fields/aliases declarations. Putting them in separate lists makes it hard to ensure alignment and invites future bugs. Best to keep them together in a list of pairs or a dict and split them before the query, as we do in other places. I'm a fan of using zip (as in readTaggedRPMS), but the approach in _split_fields works too.
The sanity check on the userID param should be if userID is None. It is not the same case if a client passes userID=0 or userID="".
if userID is None
userID=0
userID=""
rebased onto c16e382054ffa15964cc6590d1065f535797e7d0
Thanks for the updates!
The listTagged call wrapper is still passing with_owners to readTaggedBuilds even though the option has been removed, and several client calls to listTagged are still passing this option.
listTagged
1 new commit added
cleanup removed options
backward compatibility for CLI
Ouch, cleaned. I've also modified CLI to work with older hub.
rebased onto 0be3b6c9d36421a16ac01edd074d336316c71b5e
Metadata Update from @tkopecek: - Pull-request tagged with: testing-ready
rebased onto 216935617595f09fbe954149833765ab35f3a21f
11 new commits added
remove with_owners from readTaggedBuilds
updated API docs
use listTagged with_owner=False
x
readTaggedBuilds with_owners option
update API docs
fix tests for new option
use with_owners=False
hub: add with_owners option to readPackageList
Commit 0414ff7d fixes this pull-request
Pull-Request has been merged by tkopecek
Metadata Update from @mfilip: - Pull-request tagged with: testing-done
Fixes: https://pagure.io/koji/issue/2780