Github user jose-torres commented on the issue:
https://github.com/apache/spark/pull/21617
LGTM, but note that the rows being counted here are the rows persisted into
the state store, which aren't necessarily the input rows. So the side-channel
described in the JIRA would be orthogonal to this.--- --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
