[jira] [Commented] (TAJO-904) ORDER BY with a null column miss some data.

["ASF GitHub Bot (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/TAJO-904?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14060322#comment-14060322 ] 

ASF GitHub Bot commented on TAJO-904:
-------------------------------------

GitHub user babokim opened a pull request:

    https://github.com/apache/tajo/pull/70

    TAJO-904: ORDER BY with a null column miss some data.

    Currently Tajo ignores null value when calculates range partition range.  I added a MaxValueNull flag variable in ColumnStats. If ColumnStats's max value is null, ColumnStats's max value is a non-null value and  MaxValueNull flag is true. So Repartitioner calculates range cardinality with non-null value. Finally Repartitioner set null value to last range.

You can merge this pull request into a Git repository by running:

    $ git pull https://github.com/babokim/tajo TAJO-904

Alternatively you can review and apply these changes as the patch at:

    https://github.com/apache/tajo/pull/70.patch

To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:

    This closes #70
    
----
commit 3acb265add415a9f4722935f9a69d6cd97b6f987
Author: 김형준 <ba...@babokim-mbp.server.gruter.com>
Date:   2014-07-14T03:31:21Z

    TAJO-904: ORDER BY with a null column miss some data.

commit 89a35fd24bc7a8f9ab3ec7db0edfa5dad8e328a6
Author: 김형준 <ba...@babokim-mbp.server.gruter.com>
Date:   2014-07-14T03:31:59Z

    Merge branch 'master' of https://git-wip-us.apache.org/repos/asf/tajo

----


> ORDER BY with a null column miss some data.
> -------------------------------------------
>
>                 Key: TAJO-904
>                 URL: https://issues.apache.org/jira/browse/TAJO-904
>             Project: Tajo
>          Issue Type: Bug
>            Reporter: Hyoungjun Kim
>            Assignee: Hyoungjun Kim
>            Priority: Trivial
>
> ORDER BY with a null column miss some data. I found RowStoreEncoder has a bug dealing with NULL values. 
> "continue" statement should be add.
> {code}
>       Column col;
>       for (int i = 0; i < schema.size(); i++) {
>         if (tuple.isNull(i)) {
>           nullFlags.set(i);
>           continue;
>         }
>         col = schema.getColumn(i);
> {code}



--
This message was sent by Atlassian JIRA
(v6.2#6252)