[
https://issues.apache.org/jira/browse/DRILL-3476?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16498289#comment-16498289
]
ASF GitHub Bot commented on DRILL-3476:
---------------------------------------
ilooner commented on issue #83: DRILL-3476: Merge paths in FieldSelection
URL: https://github.com/apache/drill/pull/83#issuecomment-393950759
This was already merged.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
> Filter on nested element gives wrong results
> --------------------------------------------
>
> Key: DRILL-3476
> URL: https://issues.apache.org/jira/browse/DRILL-3476
> Project: Apache Drill
> Issue Type: Bug
> Reporter: Steven Phillips
> Assignee: Chun Chang
> Priority: Critical
> Fix For: 1.2.0
>
>
> Take this query for example:
> {code}
> 0: jdbc:drill:drillbit=localhost> select * from t;
> +----------------+
> | a |
> +----------------+
> | {"b":1,"c":1} |
> +----------------+
> {code}
> if I instead run:
> {code}
> 0: jdbc:drill:drillbit=localhost> select a from t where t.a.b = 1;
> +----------------+
> | a |
> +----------------+
> | {"b":1} |
> +----------------+
> {code}
> Only a.b was returned, but the select specified a. In this case, it should
> have returned all of the elements of a, not just the one specified in the
> filter.
> This is because the logic in FieldSelection does not correctly handle the
> case where a selected column is a child of another selected column. In such a
> case, the record reader should ignore the child column, and just return the
> full selected parent column.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
