hadoop-0.20.2 FileOutputFormat

[ANKITBHATNAGAR <ab...@vantage.com>]

Hi,

I am upgrading a mapreduce program from 0.18.3 to hadoop-0.20.2

in 0.18.3 in FileOutputFormat had a method - 
getRecordWriter(FileSystem ignored, JobConf job, String name, Progressable
progress) 

In 0.20.2 in FileOutputFormat declaration has changed
getRecordWriter(TaskAttemptContext job) 

I am extending this class-
Does anybody know how to get the "name" ?

name here was showing the partition name like "part-00000" and so on 

I could generate it using some weird technique but wanted to know is there a
better technique?

Thanks
Ankit

-- 
View this message in context: http://old.nabble.com/hadoop-0.20.2-FileOutputFormat-tp28830534p28830534.html
Sent from the Hadoop core-user mailing list archive at Nabble.com.

Re: hadoop-0.20.2 FileOutputFormat

[Ted Yu <yu...@gmail.com>]

Interestingly,
http://hadoop.apache.org/common/docs/r0.20.2/api/org/apache/hadoop/mapred/TaskAttemptContext.html
shows that all methods of TaskAttemptContext are deprecated.

On Wed, Jun 9, 2010 at 6:37 AM, ANKITBHATNAGAR <ab...@vantage.com>wrote:

>
> Hi,
>
> I am upgrading a mapreduce program from 0.18.3 to hadoop-0.20.2
>
> in 0.18.3 in FileOutputFormat had a method -
> getRecordWriter(FileSystem ignored, JobConf job, String name, Progressable
> progress)
>
> In 0.20.2 in FileOutputFormat declaration has changed
> getRecordWriter(TaskAttemptContext job)
>
> I am extending this class-
> Does anybody know how to get the "name" ?
>
> name here was showing the partition name like "part-00000" and so on
>
> I could generate it using some weird technique but wanted to know is there
> a
> better technique?
>
> Thanks
> Ankit
>
> --
> View this message in context:
> http://old.nabble.com/hadoop-0.20.2-FileOutputFormat-tp28830534p28830534.html
> Sent from the Hadoop core-user mailing list archive at Nabble.com.
>
>