drill-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "amit hadke (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (DRILL-3991) Support schema changes in hash join operator
Date Wed, 28 Oct 2015 18:58:28 GMT

     [ https://issues.apache.org/jira/browse/DRILL-3991?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

amit hadke updated DRILL-3991:
------------------------------
    Description: 
Hash join should be able to support schema changes during execution.
It should resolve edge cases when join columns are missing.

Example:

|Table A | Table B|
| k1&nbsp;&nbsp;&nbsp; v1 | k2 &nbsp;&nbsp;&nbsp; v2|
| 1 &nbsp;&nbsp;&nbsp;  "a" | "2" &nbsp;&nbsp;&nbsp; "b"|
| 2 &nbsp;&nbsp;&nbsp; "b" | 1  &nbsp;&nbsp;&nbsp;  "a"|
| 2.0 &nbsp;&nbsp;&nbsp;"b" | 2.0  &nbsp;&nbsp;&nbsp;"b"|
| 3 &nbsp;&nbsp;&nbsp;"c" | |
   

A INNER JOIN B on A.k1=B.k2
|k1 |  v1  |     k2|    v2|
| 1 | "a" | 1 | "a" | 
| 2  | "b" | 2.0 | "b" |
| 2.0 | "b" | 2.0 | "b" |

Where in output
    
    k1 is of union type (INTEGER, DOUBLE)
    k2 is of union type (INTEGER, DOUBLE, VARCHAR)

  was:
Hash join should be able to support schema changes during execution.
It should resolve edge cases when join columns are missing.

Example:

|Table A | Table B|
| k1&nbsp;&nbsp;&nbsp; v1 | k2 &nbsp;&nbsp;&nbsp; v2|
| 1 &nbsp;&nbsp;&nbsp;  "a" | "2" &nbsp;&nbsp;&nbsp; "b"|
| 2 &nbsp;&nbsp;&nbsp; "b" | 1  &nbsp;&nbsp;&nbsp;  "a"|
| 2.0 &nbsp;&nbsp;&nbsp;"b" | 2.0  &nbsp;&nbsp;&nbsp;"b"|
| 3 &nbsp;&nbsp;&nbsp;"c" | |
   

A INNER JOIN B on A.k1=B.k2
|k1 |  v1  |     k2|    v2|
| 1 | "a" | 1 | "a" | 
| 2  | "b" | 2.0 | "b" |
| 2.0 | "b" | 2.0 | "b" |

Where in output
    
    k1 is a union type (INTEGER, DOUBLE)
    k2 is a union type (INTEGER, DOUBLE, VARCHAR)


> Support schema changes in hash join operator
> --------------------------------------------
>
>                 Key: DRILL-3991
>                 URL: https://issues.apache.org/jira/browse/DRILL-3991
>             Project: Apache Drill
>          Issue Type: Improvement
>            Reporter: amit hadke
>
> Hash join should be able to support schema changes during execution.
> It should resolve edge cases when join columns are missing.
> Example:
> |Table A | Table B|
> | k1&nbsp;&nbsp;&nbsp; v1 | k2 &nbsp;&nbsp;&nbsp; v2|
> | 1 &nbsp;&nbsp;&nbsp;  "a" | "2" &nbsp;&nbsp;&nbsp; "b"|
> | 2 &nbsp;&nbsp;&nbsp; "b" | 1  &nbsp;&nbsp;&nbsp;  "a"|
> | 2.0 &nbsp;&nbsp;&nbsp;"b" | 2.0  &nbsp;&nbsp;&nbsp;"b"|
> | 3 &nbsp;&nbsp;&nbsp;"c" | |
>    
> A INNER JOIN B on A.k1=B.k2
> |k1 |  v1  |     k2|    v2|
> | 1 | "a" | 1 | "a" | 
> | 2  | "b" | 2.0 | "b" |
> | 2.0 | "b" | 2.0 | "b" |
> Where in output
>     
>     k1 is of union type (INTEGER, DOUBLE)
>     k2 is of union type (INTEGER, DOUBLE, VARCHAR)



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

Mime
View raw message