Mailing-List: contact user-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Reply-To: user@hadoop.apache.org
Received-SPF: pass (athena.apache.org: domain of john.meagher@gmail.com
 designates 209.85.217.173 as permitted sender)
MIME-Version: 1.0
In-Reply-To: 
 <E5CD9C0EEBC5954A95554ACE79F770BB88661E@IE1AEX3007.global.ds.honeywell.com>
References: 
 <E5CD9C0EEBC5954A95554ACE79F770BB88661E@IE1AEX3007.global.ds.honeywell.com>
Date: Tue, 1 Oct 2013 08:53:53 -0400
Message-ID: 
 <CAKb7Ddh=BQ+VFzwcqK-=hK_oAsgwECmv2QuzHW8sQ9BikJ7-pA@mail.gmail.com>
Subject: Re: Question on BytesWritable
From: John Meagher <john.meagher@gmail.com>
To: user@hadoop.apache.org
Content-Type: text/plain; charset=windows-1252
Content-Transfer-Encoding: quoted-printable

https://issues.apache.org/jira/browse/HADOOP-6298

On Tue, Oct 1, 2013 at 12:39 AM, Chandra Mohan, Ananda Vel Murugan
<Ananda.Murugan@honeywell.com> wrote:
> Hi,
>
>
>
> I am using Hadoop 1.0.2. I have written a map reduce job. I have a
> requirement to process the whole file without splitting. So I have writte=
n a
> new input format to process the file as a whole by overriding the
> isSplittable() method. I have also created a new Record reader
> implementation to read the whole file. I followed the sample in Chapter 7=
 of
> =93Hadoop- The Definitive Guide=94 book. In my map reduce job, my mapper =
emits
> BytesWritable as value. I want to get the bytes and read some specific
> information from the bytes. I use ByteArrayInputStream and do further
> processing. But strangely the following code shows different numbers.
> Because of this I am getting errors.
>
>
>
> //value -> BytesWritable
>
> System.out.println(=93Bytes length =94 + value.getLength()); // Bytes len=
gth
> 1931650
>
> byte[] bytes =3D value.getBytes();
>
> System.out.println("Bytes array length"+bytes.length); //Bytes array leng=
th
> 2897340
>
>
>
> My file size is 1931650 bytes. I don=92t know why byte array is bigger th=
an
> the original file.
>
>
>
> Any idea what is going wrong. Please help. Thanks in advance.
>
>
>
> Regards,
>
> Anand.C