Mailing-List: contact hdfs-user-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Reply-To: hdfs-user@hadoop.apache.org
Received-SPF: pass (nike.apache.org: domain of harsh@cloudera.com designates
 74.125.83.48 as permitted sender)
MIME-Version: 1.0
In-Reply-To: <1311849652.2056.YahooMailClassic@web29216.mail.ird.yahoo.com>
References: <1311338086.96751.YahooMailClassic@web29201.mail.ird.yahoo.com>
 <1311849652.2056.YahooMailClassic@web29216.mail.ird.yahoo.com>
From: Harsh J <harsh@cloudera.com>
Date: Thu, 28 Jul 2011 19:48:23 +0530
Message-ID: 
 <CAOcnVr29eNJF-=V4rYgN-m8Nm-1Gd-WxrS5sZJxVXpxRhb=Tig@mail.gmail.com>
Subject: Re: Obtain the filename that is procesed by Map class when
 CombineFileInputFormat is used
To: hdfs-user@hadoop.apache.org
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

Florin,

I believe you answered yourself accidentally?

On Thu, Jul 28, 2011 at 4:10 PM, Florin P <florinpico@yahoo.com> wrote:
> --- On Fri, 7/22/11, Florin P <florinpico@yahoo.com> wrote:
>
> From: Florin P <florinpico@yahoo.com>
> Subject: Obtain the filename that is procesed by Map class when CombineFi=
leInputFormat is used
> To: hdfs-user@hadoop.apache.org
> Date: Friday, July 22, 2011, 8:34 AM
>
> Hello!
> =A0 I would like to ask you, how can you obtain the filenames that is pro=
cessed by Map class when CombineFileInputFormat is used?
> =A0=A0 As far as I know when using CombineFileInputFormat, multiple files=
 will be processed by the same mapper. In my case, I would like to know how=
 to obtain these file names.

Depending on your you have implemented your per-FileSplit record
readers in the CFIP, you can set "map.input.file" in the Configuration
instance in each of its initialization. This is somewhat self managed
here since several record readers may be initialized. Let me know if
you would like to see a simple example along as well.

--
Harsh J