Naming nit: it's not obvious what the "right" answers are by name, you have to go back and forth to the test stream above. If instead the suppose-to-match values were, say 'a' and 'aa', 'b' and 'bb' ... then it'd be a lot more obvious what's supposed to match (especially if the test failed for some runner).
[ Full content available at: https://github.com/apache/beam/pull/6304 ] This message was relayed via gitbox.apache.org for [email protected]
