Mailing-List: contact issues-help@drill.incubator.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@drill.incubator.apache.org
Date: Fri, 7 Nov 2014 03:20:33 +0000 (UTC)
From: "Venkata krishnan Sowrirajan (JIRA)" <jira@apache.org>
To: issues@drill.incubator.apache.org
Message-ID: <JIRA.12753477.1415301393000.441968.1415330433641@Atlassian.JIRA>
In-Reply-To: <JIRA.12753477.1415301393000@Atlassian.JIRA>
References: <JIRA.12753477.1415301393000@Atlassian.JIRA>
 <JIRA.12753477.1415301393686@arcas>
Subject: [jira] [Commented] (DRILL-1646) Drill cannot handle empty
 array/list types in JSON documents
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


    [ https://issues.apache.org/jira/browse/DRILL-1646?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14201517#comment-14201517 ] 

Venkata krishnan Sowrirajan commented on DRILL-1646:
----------------------------------------------------

If the second record has actual data, then its not failing. Its giving the second record as result. I am seeing the same behavior as you said, the empty column it didn't treat as an existing column.

Data 
{
"a":[],
"b":[1,2,3]
}

0: jdbc:drill:> select * from `json/sample.json`;
+------------+
|     b      |
+------------+
| ["1","2","3"] |
+------------+
1 row selected (0.529 seconds)


> Drill cannot handle empty array/list types in JSON documents
> ------------------------------------------------------------
>
>                 Key: DRILL-1646
>                 URL: https://issues.apache.org/jira/browse/DRILL-1646
>             Project: Apache Drill
>          Issue Type: Improvement
>            Reporter: Venkata krishnan Sowrirajan
>
> To reproduce
> JSON sample data:
> {
> "a":[]
> }
> Problem: Empty array
> Exception:
> java.lang.IllegalStateException: Incoming batch has an empty schema. This is not allowed


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