[
https://issues.apache.org/jira/browse/HADOOP-7264?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13087419#comment-13087419
]
Luke Lu commented on HADOOP-7264:
---------------------------------
I just noticed that v4 (since v2) keeps avro source (*.av{sc,pr}) in java
source directory, which requires significant explicit configuration of
avro-maven-plugin, unlike in MRv2, all avro source is in avro's source
directory, which I (and Sidd) think, is more convenient and conforming to maven
conventions, which means default configuration would work with the plugin. But
I can see some benefit of keep foreign source in java source as well. What
other people think about this?
> Bump avro version to at least 1.4.1
> -----------------------------------
>
> Key: HADOOP-7264
> URL: https://issues.apache.org/jira/browse/HADOOP-7264
> Project: Hadoop Common
> Issue Type: Improvement
> Components: io
> Affects Versions: 0.21.0
> Reporter: Luke Lu
> Assignee: Luke Lu
> Labels: avro
> Fix For: 0.23.0
>
> Attachments: HADOOP-7264v2.patch, HADOOP-7264v3.patch,
> HADOOP-7264v4.patch, hadoop-7264-mr2-v1.patch, hadoop-7264-v1.patch
>
>
> Needed by mapreduce 2.0 avro support. Maybe we could jump to Avro 1.5. There
> is incompatible API changes from 1.3x to 1.4x (Utf8 to CharSequence in user
> facing APIs) not sure about 1.5x though.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
