Hi, Thanks for the reply.
I had already almost completely lost hope in using Flink SQL. You have confirmed that. But, like I said, I don't know how to reduce the large amount of boilerplate I foresee this requiring with the DataStream API. Can you help me with that? You mention "parameterizable aggregation functions", can you show an example? I don't know how to do this without reinventing AVG and COUNT over and over again. Thanks -- Sent from: http://apache-flink-user-mailing-list-archive.2336050.n4.nabble.com/
