Mailing-List: contact dev-help@hive.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@hive.apache.org
Received-SPF: pass (nike.apache.org: domain of chinnarao@huawei.com designates
 119.145.14.66 as permitted sender)
Date: Mon, 01 Aug 2011 19:59:11 +0530
From: Chinna <chinnarao@huawei.com>
Subject: Query on retrieving data using smallint datatype
To: dev@hive.apache.org
Cc: subramanian@huawei.com, venugopalarao.kotha@huawei.com
Reply-to: chinnarao@huawei.com
Message-id: <7817A4783B024271818AB9677F1BF654@china.huawei.com>
Organization: HTIPL
MIME-version: 1.0
Content-type: text/plain; charset=us-ascii
Content-transfer-encoding: 7BIT
Thread-index: AcxQV2HCQ3F7bzlrSqCzYBQgEbyW2g==

Hi,

I have a query. I have execute the following statements

  1. create table test(a smallint,b tinyint,c int,d bouble) row format
delimeted fields terminated by'@';
  2. load data local inpath '/home/data/input.txt' into table test;
  3. select a,b from test;

But one of my Input data record is:
40000@300@10000@4000.00

This data record appeared due to a special scenario/bug in the application.
So the values exceeded the given data type.

The Output returned by Hive is :   null, null

Means, when the data size exceeds the data type range, Hive is returning
NULL values. From code I can see this is explicitly implemented to do this
way. 

Can any one explain why we choose to return NULL instead of this can we
throw an exception or rotate the value

Thanks&Regards,
Chinna Rao Lalam