I wonder if we are treating a symptom instead of a rootcause. Do I understand right that the test fails because integers and strings happen to be inside the same collection, which is being sorted? If that's the case, and if that's a change in behavior from what happens on Python 2, do we understand why the change is happening?
[ Full content available at: https://github.com/apache/beam/pull/6602 ] This message was relayed via gitbox.apache.org for devnull@infra.apache.org
