hive-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "huo.zhu" <>
Subject Column Name in Nested Select of Hive Join Operator
Date Tue, 15 Jul 2014 10:32:04 GMT


I recently find a problem about column name in Hive join operator.  The HQL
is like this


select * from  -- possible column names are  key,  value


    select * from  -- possible column names are   tb_1.key,  tb_1.value,
tb_2.key,   tb_2.value


        select key ,value

        from test_tb_1

    ) tb_1





        select key ,value

        from test_tb_2

) tb_2


    on tb_1.key = tb_2.key


) tb_1_2


Why the outer select has only two columns?

 I check the explain output , and find these two columns are from tb_2,  is
this an Optimization in Hive Or a Bug?



View raw message