[jira] [Updated] (TAJO-1556) "insert into select" with reordered column list does not work.

["Jihoon Son (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/TAJO-1556?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Jihoon Son updated TAJO-1556:
-----------------------------
    Attachment: TAJO-1556_add_test.patch

Since we scheduled this issue for 0.10.1 but [~yongjin.choi] looks busy, I've added a new test for query result verification.
Here is the patch. Please review it.

> "insert into select" with reordered column list does not work.
> --------------------------------------------------------------
>
>                 Key: TAJO-1556
>                 URL: https://issues.apache.org/jira/browse/TAJO-1556
>             Project: Tajo
>          Issue Type: Bug
>          Components: compilation
>            Reporter: Yongjin Choi
>            Assignee: Yongjin Choi
>            Priority: Minor
>             Fix For: 0.11.0, 0.10.1
>
>         Attachments: TAJO-1556.patch, TAJO-1556_add_test.patch
>
>
> Using "insert into table1 (colums, ...) select " statement, users can fill part of table and tajo already had this feature.
> If columns are given in the same order as table definition, it seems to work well.
> Otherwise, the results are reversed.
> This bug can be reproduced easily as shown below (table1 is same as in http://tajo.apache.org/docs/0.8.0/getting_started/first_query.html).
> {code:sql}
> create table t1 (id int, name text, score float, type text);
> insert into t1 (type, name) select type, name from table1;
> default> select * from t1;
> id,  name,  score,  type
> -------------------------------
> ,  a,  ,  abc
> ,  b,  ,  def
> {code}
> But, the expected result is as follows.
> {code:sql}
> default> select * from t1;
> id,  name,  score,  type
> -------------------------------
> ,  abc,  ,  a
> ,  def,  ,  b
> {code}



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