[jira] [Created] (HIVE-12874) dynamic partition insert project wrong column

["bin wang (JIRA)" <ji...@apache.org>]

bin wang created HIVE-12874:
-------------------------------

             Summary: dynamic partition insert project wrong column
                 Key: HIVE-12874
                 URL: https://issues.apache.org/jira/browse/HIVE-12874
             Project: Hive
          Issue Type: Bug
          Components: SQL
    Affects Versions: 0.14.1
         Environment: hive 1.1.0-cdh5.4.8
            Reporter: bin wang
            Assignee: Alan Gates


We have two table as below:
create table  test (
id bigint comment ' id',
)
PARTITIONED BY(etl_dt string)
STORED AS ORC;

create table  test1 (
id bigint
start_time int,
)
PARTITIONED BY(etl_dt string)
STORED AS ORC;

we use sql like below to import rows from test1 to test:

insert overwrite table test PARTITION(etl_dt)
select id
,from_unixtime(start_time,'yyyy-MM-dd') as  etl_dt                                                                                                               
from test1
where test1.etl_dt='2016-01-12';

but it behave wrong, it use test1.etl_dt as the test's partition value, not the 'etl_dt' in select.
We think it's a bug, anyone to fix it? 



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)