blob: 2783047698e78bd97002ebb5054df6c37b031e70 [file] [log] [blame]
set hive.new.job.grouping.set.cardinality=2;
CREATE TABLE T1(a STRING, b STRING, c STRING) ROW FORMAT DELIMITED FIELDS TERMINATED BY ' ' STORED AS TEXTFILE;
-- Since 4 grouping sets would be generated for the query below, an additional MR job should be created
-- This is not allowed with distincts.
SELECT a, b, count(distinct c) from T1 group by a, b with cube;