Predicate pushdown should work with single insert in the form of FROM TABLE
INSERT ... SELECT ... WHERE. ..pred
---------------------------------------------------------------------------------------------------------------
Key: HIVE-1034
URL: https://issues.apache.org/jira/browse/HIVE-1034
Project: Hadoop Hive
Issue Type: Improvement
Reporter: Ning Zhang
One sample query is
FROM a JOIN b ON a.eid = b.eid AND a.ds>='2009-10-01' AND a.ds<='2009-12-01'
AND b.ds>='2009-10-01' AND b.ds<='2009-12-01'
INSERT OVERWRITE TABLE t SELECT a.user, b.user, a.eid, a.time
WHERE (a.time < 1259625600) and a.user <> b.user;
Here a.time<1259625600 should be pushed down before the join since there is
only 1 insert. This optimization should NOT be done if there are multiple
inserts.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
