[ 
https://issues.apache.org/jira/browse/PHOENIX-3348?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Julian Jaffe updated PHOENIX-3348:
----------------------------------
    Summary: SELECT DISTINCT with a non-rowkey array and another column returns 
incorrect results  (was: SELECT DISTINCT with a non-rowkey array and an id 
returns incorrect results)

> SELECT DISTINCT with a non-rowkey array and another column returns incorrect 
> results
> ------------------------------------------------------------------------------------
>
>                 Key: PHOENIX-3348
>                 URL: https://issues.apache.org/jira/browse/PHOENIX-3348
>             Project: Phoenix
>          Issue Type: Bug
>    Affects Versions: 4.6.0
>            Reporter: Julian Jaffe
>
> Attempting to select distinct combinations from an array and another column 
> returns incorrect results if the columns are not part of the primary key.
> {code:sql}
> CREATE TABLE IF NOT EXISTS TEST.DESERTEST (COLONE BIGINT NOT NULL PRIMARY 
> KEY, COLTWO BIGINT, COLTHREE BIGINT[]);
> No rows affected (1.368 seconds)
> UPSERT INTO TEST.DESERTEST VALUES (10, 7, ARRAY[1,2]);
> 1 row affected (0.161 seconds)
> SELECT DISTINCT COLTWO, COLTHREE FROM TEST.DESERTEST WHERE COLONE = 10;
> +------------------------------------------+------------------------------------------+
> |                  COLTWO                  |                 COLTHREE         
>         |
> +------------------------------------------+------------------------------------------+
> | -9223372036854677504                     | []                               
>         |
> +------------------------------------------+------------------------------------------+
> 1 row selected (0.367 seconds)
> SELECT COLTWO, COLTHREE FROM TEST.DESERTEST WHERE COLONE = 10;
> +------------------------------------------+------------------------------------------+
> |                  COLTWO                  |                 COLTHREE         
>         |
> +------------------------------------------+------------------------------------------+
> | 7                                        | [1, 2]                           
>         |
> +------------------------------------------+------------------------------------------+
> 1 row selected (0.336 seconds)
> SELECT DISTINCT COLONE, COLTHREE FROM TEST.DESERTEST WHERE COLONE = 10;
> +------------------------------------------+------------------------------------------+
> |                  COLONE                  |                 COLTHREE         
>         |
> +------------------------------------------+------------------------------------------+
> | 10                                       | [1, 2]                           
>         |
> +------------------------------------------+------------------------------------------+
> 1 row selected (0.32 seconds)
> {code}
> We've confirmed this bug in 4.6, and confirmed that it does not occur in 4.8, 
> although I haven't been able to locate a corresponding JIRA. However, it will 
> occur if using the 4.6 client to talk to 4.8, so it appears to be at least 
> partially client-side.



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

Reply via email to