Yash Datta created SPARK-4762:
---------------------------------
Summary: Add support for tuples in where in clause query
Key: SPARK-4762
URL: https://issues.apache.org/jira/browse/SPARK-4762
Project: Spark
Issue Type: Improvement
Components: SQL
Affects Versions: 1.1.0
Reporter: Yash Datta
Fix For: 1.3.0
Currently, in the where in clause the filter is applied only on a single
column. We can enhance it to accept filter on multiple columns.
So current support is for queries like :
Select * from table where c1 in (value1,value2,...value n);
Need to add support for queries like :
Select * from table where (c1,c2,... cn) in ((value1,value2...value n),
(value1' , value2' ... ,value n').... )
Also, we can add optimized version of where in clause of tuples , where we
create a hashset of the filter tuples for matching rows.
This also requires a change in the hive parser since currently there is no
support for multiple columns in IN clause.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
