Michael Smith has posted comments on this change. ( http://gerrit.cloudera.org:8080/22409 )
Change subject: IMPALA-13042: Calcite Planner; Enable partition pruning ...................................................................... Patch Set 5: (4 comments) http://gerrit.cloudera.org:8080/#/c/22409/4//COMMIT_MSG Commit Message: http://gerrit.cloudera.org:8080/#/c/22409/4//COMMIT_MSG@9 PS4, Line 9: Enables partition pruning in the HdfsScan RelNode. > Yes, and there are tests in the Impala test framework that check for partit Ack http://gerrit.cloudera.org:8080/#/c/22409/4/fe/src/main/java/org/apache/impala/planner/HdfsPartitionFilter.java File fe/src/main/java/org/apache/impala/planner/HdfsPartitionFilter.java: http://gerrit.cloudera.org:8080/#/c/22409/4/fe/src/main/java/org/apache/impala/planner/HdfsPartitionFilter.java@123 PS4, Line 123: Expr literalPredicate = predicate_.substitute(sMap, analyzer, true); > I tested this months ago, so I can't quite remember which query caused the The 3rd argument comes from IMPALA-1340. It's interesting that the comment on https://github.com/apache/impala/commit/b6204dff59c013ec3f04f2867538f56c66c6e61e suggests there are other places where it may be necessary, but they didn't identify any besides union. So seems reasonable. http://gerrit.cloudera.org:8080/#/c/22409/4/java/calcite-planner/src/main/java/org/apache/impala/calcite/rel/util/PrunedPartitionHelper.java File java/calcite-planner/src/main/java/org/apache/impala/calcite/rel/util/PrunedPartitionHelper.java: http://gerrit.cloudera.org:8080/#/c/22409/4/java/calcite-planner/src/main/java/org/apache/impala/calcite/rel/util/PrunedPartitionHelper.java@60 PS4, Line 60: // IMPALA-13849: tblref is null. Tablesampling is disabled. > Filed IMPALA-13849 Ack http://gerrit.cloudera.org:8080/#/c/22409/5/java/calcite-planner/src/main/java/org/apache/impala/calcite/rel/util/PrunedPartitionHelper.java File java/calcite-planner/src/main/java/org/apache/impala/calcite/rel/util/PrunedPartitionHelper.java: http://gerrit.cloudera.org:8080/#/c/22409/5/java/calcite-planner/src/main/java/org/apache/impala/calcite/rel/util/PrunedPartitionHelper.java@102 PS5, Line 102: private static class ConjunctsHelper { This class seems kind of arbitrary. Why doesn't separateConjuncts populate partitionedConjuncts_ and nonPartitionedConjuncts_ directly in the parent class? -- To view, visit http://gerrit.cloudera.org:8080/22409 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: Impala-ASF Gerrit-Branch: master Gerrit-MessageType: comment Gerrit-Change-Id: I8c698b857555baeae347835b4a6b39d035f12405 Gerrit-Change-Number: 22409 Gerrit-PatchSet: 5 Gerrit-Owner: Steve Carlin <scar...@cloudera.com> Gerrit-Reviewer: Aman Sinha <amsi...@cloudera.com> Gerrit-Reviewer: Fang-Yu Rao <fangyu....@cloudera.com> Gerrit-Reviewer: Impala Public Jenkins <impala-public-jenk...@cloudera.com> Gerrit-Reviewer: Joe McDonnell <joemcdonn...@cloudera.com> Gerrit-Reviewer: Michael Smith <michael.sm...@cloudera.com> Gerrit-Reviewer: Riza Suminto <riza.sumi...@cloudera.com> Gerrit-Reviewer: Steve Carlin <scar...@cloudera.com> Gerrit-Comment-Date: Mon, 10 Mar 2025 23:37:19 +0000 Gerrit-HasComments: Yes
