Details
-
Sub-task
-
Status: Resolved
-
Major
-
Resolution: Won't Fix
-
3.1.0
-
None
-
None
Description
spark.sql("create table t1(id string) partitioned by (part string) stored as parquet") spark.sql("insert into t1 values('1', '2019-01-01')") spark.sql("insert into t1 values('2', '2019-01-02')") spark.sql("select * from t1 where part = date '2019-01-01' ").show
We can pushdown {{cast(part as date) = date '2019-01-01' }} to Hive metastore.