Home | About | Sematext search-lucene.com search-hadoop.com
 Search Hadoop and all its subprojects:

Switch to Threaded View
MapReduce >> mail # user >> InputFormat in mapred vs. mapreduce.


Copy link to this message
-
Re: InputFormat in mapred vs. mapreduce.
Harsh, thank you for your response.  That's what I guessed.

In 0.20, Interface InputFormat under mapred package was deprecated.  In 0.21, it is no longer deprecated.  Why is that?

Thanks,
Jane

--- On Tue, 12/7/10, Harsh J <[EMAIL PROTECTED]> wrote:

> From: Harsh J <[EMAIL PROTECTED]>
> Subject: Re: InputFormat in mapred vs. mapreduce.
> To: [EMAIL PROTECTED]
> Date: Tuesday, December 7, 2010, 5:51 AM
> mapred.* is the old, "stable" API,
> while the mapreduce.* stuff is
> newer. With 0.21, I'd suggest using mapreduce.*
>
> mapred.* stuff isn't compatible directly with mapreduce.*.
> If you are
> using the New API all over (driver, mapper, etc.), you
> should use the
> mapreduce.* only, right?
>
> On Tue, Dec 7, 2010 at 3:05 AM, Jane Chen <[EMAIL PROTECTED]>
> wrote:
> > In Hadoop 0.21, I found InputFormat as an Interface in
> package mapred, and as an abstract class in package
> mapreduce.  The APIs are slightly different.  Which one
> should I choose to extend from or implement?  How are the
> two packages intended to be used differently?
> >
> > Thanks,
> > Jane
> >
> >
> >
> >
>
>
>
> --
> Harsh J
> www.harshj.com
>