You are viewing a plain text version of this content. The canonical link for it is here.
Posted to common-user@hadoop.apache.org by steven zhuang <zh...@gmail.com> on 2010/01/28 14:14:45 UTC

map(K1 key, V1 value, OutputCollector output, Reporter reporter) deprecated in 0.20.2?

hello, all,
                As a newbie, I have been used to the (k1,v1,k2,v2) format
parameter list for map and reduce methods in mapper and reducer(as is
written in many books), but after several failures, I found in 0.20+, if we
extends from base class org.apache.hadoop.mapreduce.Mapper, the map should
be something like this:

         void map(KEYIN key, VALUEIN value, Context context) throws
IOException, InterruptedException
               A little confusing to me.
               My question is why the old fashion map interface deprecated?
thanks!


-- 
       best wishes.
                            steven

Re: map(K1 key, V1 value, OutputCollector output, Reporter reporter) deprecated in 0.20.2?

Posted by steven zhuang <zh...@gmail.com>.
thanks you guys  :D


On Sat, Jan 30, 2010 at 2:16 AM, Jim Twensky <ji...@gmail.com> wrote:

> Steven,
>
> I recently had the same issues, and I found this blog post very
> helpful on migrating from 0.19 to 0.20.2
>
> http://sonerbalkir.blogspot.com/2010/01/new-hadoop-api-020x.html
>
> You can download the sample code at the end, which contains a Hadoop
> word count program written using the new API.
>
> Hope this helps.
>
> -Jim
>
> On Thu, Jan 28, 2010 at 9:43 AM, Edward Capriolo <ed...@gmail.com>
> wrote:
> > On Thu, Jan 28, 2010 at 8:14 AM, steven zhuang <zh...@gmail.com>
> wrote:
> >> hello, all,
> >>                As a newbie, I have been used to the (k1,v1,k2,v2) format
> >> parameter list for map and reduce methods in mapper and reducer(as is
> >> written in many books), but after several failures, I found in 0.20+, if
> we
> >> extends from base class org.apache.hadoop.mapreduce.Mapper, the map
> should
> >> be something like this:
> >>
> >>         void map(KEYIN key, VALUEIN value, Context context) throws
> >> IOException, InterruptedException
> >>               A little confusing to me.
> >>               My question is why the old fashion map interface
> deprecated?
> >> thanks!
> >>
> >>
> >> --
> >>       best wishes.
> >>                            steven
> >>
> >
> > Steven,
> >
> > The old map/reduce api is still available. org.apache.hadoop.mapred
> >
> >>               My question is why the old fashion map interface
> deprecated?
> >
> > Cause hadoop is like a freight train, either hop on or get out of the way
> ! :)
> > Just kidding,
> >
> > Great presentation about how to update code:
> > http://www.slideshare.net/sh1mmer/upgrading-to-the-new-map-reduce-api
> >
> > Some information on the 'why'
> >
> http://www.cloudera.com/blog/2009/05/07/what%E2%80%99s-new-in-hadoop-core-020/
> >
>



-- 
       best wishes.
                            steven

Re: map(K1 key, V1 value, OutputCollector output, Reporter reporter) deprecated in 0.20.2?

Posted by Jim Twensky <ji...@gmail.com>.
Steven,

I recently had the same issues, and I found this blog post very
helpful on migrating from 0.19 to 0.20.2

http://sonerbalkir.blogspot.com/2010/01/new-hadoop-api-020x.html

You can download the sample code at the end, which contains a Hadoop
word count program written using the new API.

Hope this helps.

-Jim

On Thu, Jan 28, 2010 at 9:43 AM, Edward Capriolo <ed...@gmail.com> wrote:
> On Thu, Jan 28, 2010 at 8:14 AM, steven zhuang <zh...@gmail.com> wrote:
>> hello, all,
>>                As a newbie, I have been used to the (k1,v1,k2,v2) format
>> parameter list for map and reduce methods in mapper and reducer(as is
>> written in many books), but after several failures, I found in 0.20+, if we
>> extends from base class org.apache.hadoop.mapreduce.Mapper, the map should
>> be something like this:
>>
>>         void map(KEYIN key, VALUEIN value, Context context) throws
>> IOException, InterruptedException
>>               A little confusing to me.
>>               My question is why the old fashion map interface deprecated?
>> thanks!
>>
>>
>> --
>>       best wishes.
>>                            steven
>>
>
> Steven,
>
> The old map/reduce api is still available. org.apache.hadoop.mapred
>
>>               My question is why the old fashion map interface deprecated?
>
> Cause hadoop is like a freight train, either hop on or get out of the way ! :)
> Just kidding,
>
> Great presentation about how to update code:
> http://www.slideshare.net/sh1mmer/upgrading-to-the-new-map-reduce-api
>
> Some information on the 'why'
> http://www.cloudera.com/blog/2009/05/07/what%E2%80%99s-new-in-hadoop-core-020/
>

Re: map(K1 key, V1 value, OutputCollector output, Reporter reporter) deprecated in 0.20.2?

Posted by Edward Capriolo <ed...@gmail.com>.
On Thu, Jan 28, 2010 at 8:14 AM, steven zhuang <zh...@gmail.com> wrote:
> hello, all,
>                As a newbie, I have been used to the (k1,v1,k2,v2) format
> parameter list for map and reduce methods in mapper and reducer(as is
> written in many books), but after several failures, I found in 0.20+, if we
> extends from base class org.apache.hadoop.mapreduce.Mapper, the map should
> be something like this:
>
>         void map(KEYIN key, VALUEIN value, Context context) throws
> IOException, InterruptedException
>               A little confusing to me.
>               My question is why the old fashion map interface deprecated?
> thanks!
>
>
> --
>       best wishes.
>                            steven
>

Steven,

The old map/reduce api is still available. org.apache.hadoop.mapred

>               My question is why the old fashion map interface deprecated?

Cause hadoop is like a freight train, either hop on or get out of the way ! :)
Just kidding,

Great presentation about how to update code:
http://www.slideshare.net/sh1mmer/upgrading-to-the-new-map-reduce-api

Some information on the 'why'
http://www.cloudera.com/blog/2009/05/07/what%E2%80%99s-new-in-hadoop-core-020/