I would like to use MultipleTextOutputFormat, which is only available with the old Hadoop API (mapred).
The mapred version of TableMapReduceUtil does not seem to support the use of a Scan object.

Is there any way around this?

Can you MTOF to the mapreduce package?  See how in hbase, we have a
hadoopbackport sub-package under mapreduce into which some classes
from the mapred were ported.

St.Ack
On Mon, Jun 27, 2011 at 1:26 PM, Chan, Tim <[EMAIL PROTECTED]> wrote:
> I would like to use MultipleTextOutputFormat, which is only available with the old Hadoop API (mapred).
> The mapred version of TableMapReduceUtil does not seem to support the use of a Scan object.
>
> Is there any way around this?
>

I do not understand what MTOF stands for.

I'm not interested in using the new mapreduce API, I would like to use the old mapred API, but with the ability to pass in Scan to the TableMapReduceUtil.

________________________________________
From: [EMAIL PROTECTED] [[EMAIL PROTECTED]] On Behalf Of Stack [[EMAIL PROTECTED]]
Sent: Monday, June 27, 2011 1:39 PM
To: [EMAIL PROTECTED]
Subject: Re: How can I use Scan with mapred version of TableMapReduceUtil?

Can you MTOF to the mapreduce package?  See how in hbase, we have a
hadoopbackport sub-package under mapreduce into which some classes
from the mapred were ported.

St.Ack
On Mon, Jun 27, 2011 at 1:26 PM, Chan, Tim <[EMAIL PROTECTED]> wrote:
> I would like to use MultipleTextOutputFormat, which is only available with the old Hadoop API (mapred).
> The mapred version of TableMapReduceUtil does not seem to support the use of a Scan object.
>
> Is there any way around this?
>

I think Stack was asking if you could port MultipleTextOutputFormat to
the new API. And the pointers he gave was how to wrap an old API
implementation with the new API.

-Joey

On Mon, Jun 27, 2011 at 1:44 PM, Chan, Tim <[EMAIL PROTECTED]> wrote:
> I do not understand what MTOF stands for.
>
> I'm not interested in using the new mapreduce API, I would like to use the old mapred API, but with the ability to pass in Scan to the TableMapReduceUtil.
>
> ________________________________________
> From: [EMAIL PROTECTED] [[EMAIL PROTECTED]] On Behalf Of Stack [[EMAIL PROTECTED]]
> Sent: Monday, June 27, 2011 1:39 PM
> To: [EMAIL PROTECTED]
> Subject: Re: How can I use Scan with mapred version of TableMapReduceUtil?
>
> Can you MTOF to the mapreduce package?  See how in hbase, we have a
> hadoopbackport sub-package under mapreduce into which some classes
> from the mapred were ported.
>
> St.Ack
>
>
> On Mon, Jun 27, 2011 at 1:26 PM, Chan, Tim <[EMAIL PROTECTED]> wrote:
>> I would like to use MultipleTextOutputFormat, which is only available with the old Hadoop API (mapred).
>> The mapred version of TableMapReduceUtil does not seem to support the use of a Scan object.
>>
>> Is there any way around this?
>>
>
>

--
Joseph Echeverria
Cloudera, Inc.
443.305.9434

Thanks Joey.

That makes sense now. Heh. I just figured out MTOF is MultipleTextOutputFormat.

________________________________________
From: Joey Echeverria [[EMAIL PROTECTED]]
Sent: Monday, June 27, 2011 2:11 PM
To: [EMAIL PROTECTED]
Subject: Re: How can I use Scan with mapred version of TableMapReduceUtil?

I think Stack was asking if you could port MultipleTextOutputFormat to
the new API. And the pointers he gave was how to wrap an old API
implementation with the new API.

-Joey

On Mon, Jun 27, 2011 at 1:44 PM, Chan, Tim <[EMAIL PROTECTED]> wrote:
> I do not understand what MTOF stands for.
>
> I'm not interested in using the new mapreduce API, I would like to use the old mapred API, but with the ability to pass in Scan to the TableMapReduceUtil.
>
> ________________________________________
> From: [EMAIL PROTECTED] [[EMAIL PROTECTED]] On Behalf Of Stack [[EMAIL PROTECTED]]
> Sent: Monday, June 27, 2011 1:39 PM
> To: [EMAIL PROTECTED]
> Subject: Re: How can I use Scan with mapred version of TableMapReduceUtil?
>
> Can you MTOF to the mapreduce package?  See how in hbase, we have a
> hadoopbackport sub-package under mapreduce into which some classes
> from the mapred were ported.
>
> St.Ack
>
>
> On Mon, Jun 27, 2011 at 1:26 PM, Chan, Tim <[EMAIL PROTECTED]> wrote:
>> I would like to use MultipleTextOutputFormat, which is only available with the old Hadoop API (mapred).
>> The mapred version of TableMapReduceUtil does not seem to support the use of a Scan object.
>>
>> Is there any way around this?
>>
>
>

--
Joseph Echeverria
Cloudera, Inc.
443.305.9434

Pardon my curtness (Thanks Joey for doing the interpreting).
St.Ack

On Mon, Jun 27, 2011 at 2:27 PM, Chan, Tim <[EMAIL PROTECTED]> wrote:
> Thanks Joey.
>
> That makes sense now. Heh. I just figured out MTOF is MultipleTextOutputFormat.
>
> ________________________________________
> From: Joey Echeverria [[EMAIL PROTECTED]]
> Sent: Monday, June 27, 2011 2:11 PM
> To: [EMAIL PROTECTED]
> Subject: Re: How can I use Scan with mapred version of TableMapReduceUtil?
>
> I think Stack was asking if you could port MultipleTextOutputFormat to
> the new API. And the pointers he gave was how to wrap an old API
> implementation with the new API.
>
> -Joey
>
> On Mon, Jun 27, 2011 at 1:44 PM, Chan, Tim <[EMAIL PROTECTED]> wrote:
>> I do not understand what MTOF stands for.
>>
>> I'm not interested in using the new mapreduce API, I would like to use the old mapred API, but with the ability to pass in Scan to the TableMapReduceUtil.
>>
>> ________________________________________
>> From: [EMAIL PROTECTED] [[EMAIL PROTECTED]] On Behalf Of Stack [[EMAIL PROTECTED]]
>> Sent: Monday, June 27, 2011 1:39 PM
>> To: [EMAIL PROTECTED]
>> Subject: Re: How can I use Scan with mapred version of TableMapReduceUtil?
>>
>> Can you MTOF to the mapreduce package?  See how in hbase, we have a
>> hadoopbackport sub-package under mapreduce into which some classes
>> from the mapred were ported.
>>
>> St.Ack
>>
>>
>> On Mon, Jun 27, 2011 at 1:26 PM, Chan, Tim <[EMAIL PROTECTED]> wrote:
>>> I would like to use MultipleTextOutputFormat, which is only available with the old Hadoop API (mapred).
>>> The mapred version of TableMapReduceUtil does not seem to support the use of a Scan object.
>>>
>>> Is there any way around this?
>>>
>>
>>
>
>
>
> --
> Joseph Echeverria
> Cloudera, Inc.
> 443.305.9434
>
>