[
https://issues.apache.org/jira/browse/MAPREDUCE-4616?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13459025#comment-13459025
]
Harsh J commented on MAPREDUCE-4616:
------------------------------------
Thanks Tony.
A few comments we may address before this gets committed (mostly nits and
typos):
- "Use in conjuction" -> "Useful when used in conjunction" perhaps?
- "Use your own code in <code>generateFileName()</code>". Your code sample
references a generateFileName method but doesn't show an implementation.
Perhaps add in a sample implementation that returns "part", "foo" or whatever?
- "in your Hadoop job task-level setup." -> Simpler to say "in your Job
configuration."?
> Improvement to MultipleOutputs javadocs
> ---------------------------------------
>
> Key: MAPREDUCE-4616
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-4616
> Project: Hadoop Map/Reduce
> Issue Type: Improvement
> Components: documentation
> Affects Versions: 1.0.3
> Reporter: Tony Burton
> Priority: Minor
> Attachments: MAPREDUCE-4616.patch
>
>
> In the new API, and using MultipleOutputs it is possible to segment output
> into directories by using MultipleOutputs.write(KEYOUT key, VALUEOUT value,
> String baseOutputPath) in the Reducer to determine the output directory, and
> by using LazyOutputFormat at the job-level config to suppress normal output
> [eg use LazyOutputFormat.setOutputFormatClass(job, TextOutputFormat.class);
> instead of job.setOutputFormatClass(TextOutputFormat.class);]
> This recreates the functionality previously provided in the old API by using
> MultipleTextOutputFormat (etc)
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
