Matthew Jacobs has posted comments on this change. Change subject: IMPALA-3742: partitions DMLs for Kudu tables ......................................................................
Patch Set 7: (5 comments) http://gerrit.cloudera.org:8080/#/c/6037/7/be/src/runtime/data-stream-partitioner.cc File be/src/runtime/data-stream-partitioner.cc: PS7, Line 51: void* value = ctx->GetValue(row); : PrimitiveType type = ctx->root()->type().type; : WriteKuduRowValue(row_.get(), col, type, value); probably a codegen opportunity here for later, can you leave a TODO? http://gerrit.cloudera.org:8080/#/c/6037/7/be/src/runtime/data-stream-partitioner.h File be/src/runtime/data-stream-partitioner.h: PS7, Line 45: switch (partition_.type) { : case TPartitionType::HASH_PARTITIONED: I'm sorry if I wasn't clear when we spoke last, but I think it is still better to keep the classes separate and do this switching in the data-stream-sender. It's kind of unfortunate to have to have shared members in this class among the different branches. PS7, Line 63: UNKOWN typo PS7, Line 87: next_partition_ rename to indicate used for R.R. only, e.g. next_unknown_partition_ or something http://gerrit.cloudera.org:8080/#/c/6037/6/testdata/workloads/functional-planner/queries/PlannerTest/kudu-update.test File testdata/workloads/functional-planner/queries/PlannerTest/kudu-update.test: PS6, Line 64: [KUDU(a.id)] > What should the behavior be? There isn't an alias to use in the above queri I'm not sure, it's just interesting that it varies. Doesn't seem like a problem but it'd be good for the FE reviewer to confirm this is OK. -- To view, visit http://gerrit.cloudera.org:8080/6037 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-MessageType: comment Gerrit-Change-Id: Ic10b3295159354888efcde3df76b0edb24161515 Gerrit-PatchSet: 7 Gerrit-Project: Impala-ASF Gerrit-Branch: master Gerrit-Owner: Thomas Tauber-Marshall <[email protected]> Gerrit-Reviewer: Henry Robinson <[email protected]> Gerrit-Reviewer: Marcel Kornacker <[email protected]> Gerrit-Reviewer: Matthew Jacobs <[email protected]> Gerrit-Reviewer: Thomas Tauber-Marshall <[email protected]> Gerrit-HasComments: Yes
