Aman Sinha has posted comments on this change. ( http://gerrit.cloudera.org:8080/16346 )
Change subject: IMPALA-10064: Support constant propagation for eligible range predicates ...................................................................... Patch Set 9: (2 comments) http://gerrit.cloudera.org:8080/#/c/16346/9/testdata/workloads/functional-planner/queries/PlannerTest/constant-propagation.test File testdata/workloads/functional-planner/queries/PlannerTest/constant-propagation.test: http://gerrit.cloudera.org:8080/#/c/16346/9/testdata/workloads/functional-planner/queries/PlannerTest/constant-propagation.test@419 PS9, Line 419: predicates: timestamp_col <= TIMESTAMP '2010-12-01 00:00:00', timestamp_col >= TIMESTAMP '2009-12-01 00:00:00' > Don't we still need to keep the date_col = cast(timestamp_col as date) pred Good point. All the use cases I have seen so far were ones where date_col was derived from the timestamp column. Yeah, for your example, we need to keep the cast predicate if the constant is a range predicate. I think the code change isn't much but I need to think about how to create a test data set for this. http://gerrit.cloudera.org:8080/#/c/16346/9/testdata/workloads/functional-query/queries/QueryTest/range-constant-propagation.test File testdata/workloads/functional-query/queries/QueryTest/range-constant-propagation.test: http://gerrit.cloudera.org:8080/#/c/16346/9/testdata/workloads/functional-query/queries/QueryTest/range-constant-propagation.test@4 PS9, Line 4: functional_parquet > We generally don't include database names in the test files, since the infr I would be ok with running with other data sets but I had some struggles in loading the alltypes_date_partition table and had offline discussion with Shant. For Text format loading, the following error occurred since it went through HIve load process rather than Impala: The load-functional-planner-core-hive-generated-text-none-none.sql.log had the following error: "Caused by: org.apache.hadoop.hive.ql.parse.SemanticException: Dynamic partition strict mode requires at least one static partition column. To turn this off set hive.exec.dynamic.partition.mode=nonstrict" Setting the partition.mode to nonstrict got past that but ran into a default limit of the # dynamic partitions: "The maximum number of dynamic partitions is controlled by hive.exec.max.dynamic.partitions and hive.exec.max.dynamic.partitions.pernode. Maximum was set to 100 partitions per node, number of dynamic partitions on this node: 101" I could bump this up too .. but the Tez job does take much longer to execute..so I wasn't sure if it is worthwhile. I could move this to TestQueriesParquetTables unless you have other suggestions. -- To view, visit http://gerrit.cloudera.org:8080/16346 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: Impala-ASF Gerrit-Branch: master Gerrit-MessageType: comment Gerrit-Change-Id: I811a1f8d605c27c7704d7fc759a91510c6db3c2b Gerrit-Change-Number: 16346 Gerrit-PatchSet: 9 Gerrit-Owner: Aman Sinha <[email protected]> Gerrit-Reviewer: Aman Sinha <[email protected]> Gerrit-Reviewer: Impala Public Jenkins <[email protected]> Gerrit-Reviewer: Qifan Chen <[email protected]> Gerrit-Reviewer: Shant Hovsepian <[email protected]> Gerrit-Reviewer: Tim Armstrong <[email protected]> Gerrit-Comment-Date: Tue, 01 Sep 2020 02:20:28 +0000 Gerrit-HasComments: Yes
