[
https://issues.apache.org/jira/browse/MAPREDUCE-1634?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12852096#action_12852096
]
Doug Cutting commented on MAPREDUCE-1634:
-----------------------------------------
> Wouldn't this create a circular dependency?
The added Avro dependency on Hadoop is only at compile/test time. It compiles
against Hadoop 0.20.2, as retrieved from Maven. Avro's POM does not require
Hadoop. The code should be compatible with Hadoop 0.18+ and perhaps even
earlier, although I have not tested that yet.
> What are the advantages to adding this to Avro instead of MapReduce?
Primarily that it can be used with older Hadoop releases.
The Avro version of the patch is at AVRO-493.
> add input and output formats for Avro value wrappers
> ----------------------------------------------------
>
> Key: MAPREDUCE-1634
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-1634
> Project: Hadoop Map/Reduce
> Issue Type: New Feature
> Components: client
> Reporter: Doug Cutting
> Fix For: 0.22.0
>
> Attachments: MAPREDUCE-1634.patch
>
>
> HADOOP-6660 proposes adding an AvroValue wrapper for Avro data. We should
> add InputFormat and OutputFormat implementations for Avro data files that use
> this.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
