Tim Armstrong has posted comments on this change. Change subject: IMPALA-3567: Part 1: groundwork to make Join build sides DataSinks ......................................................................
Patch Set 1: (1 comment) http://gerrit.cloudera.org:8080/#/c/3842/1/be/src/exec/nested-loop-join-builder.h File be/src/exec/nested-loop-join-builder.h: Line 35: /// tuple memory, the non-copying mode is used and row batches are simply accumulated in > what about this question? I was confused since I thought you were talking about the different memory ownership cases. There are only two places that drive it so it's true that there are only two patterns of use. It's not clear to me that we should enshrine the behaviour of those two callers in the interface. E.g. I don't know what meaningful name I'd give this flag. I think the logical distinction is between the DataSink interface (that doesn't expose AddBuildBatch()) and the NestedLoopJoinBuilder interface, where you can call AddBuildBatch() if you know that the input node won't return batches with the need_to_return flag set and Send() otherwise. Send() also calls AddBuildBatch() internally. -- To view, visit http://gerrit.cloudera.org:8080/3842 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-MessageType: comment Gerrit-Change-Id: I9d7608181eeacfe706a09c1e153d0a3e1ee9b475 Gerrit-PatchSet: 1 Gerrit-Project: Impala-ASF Gerrit-Branch: master Gerrit-Owner: Tim Armstrong <[email protected]> Gerrit-Reviewer: Alex Behm <[email protected]> Gerrit-Reviewer: Marcel Kornacker <[email protected]> Gerrit-Reviewer: Tim Armstrong <[email protected]> Gerrit-HasComments: Yes
