[
https://issues.apache.org/jira/browse/HIVE-20678?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16639847#comment-16639847
]
Yongzhi Chen commented on HIVE-20678:
-------------------------------------
Do we still need implement the method?
public org.apache.hadoop.mapred.RecordWriter<ImmutableBytesWritable, Object>
getRecordWriter(
And do we still need implement mapred.RecordWriter in the MyRecordWriter?
static private class MyRecordWriter
implements org.apache.hadoop.mapred.RecordWriter<ImmutableBytesWritable,
Object>,
org.apache.hadoop.hive.ql.exec.FileSinkOperator.RecordWriter {
> HiveHBaseTableOutputFormat should implement HiveOutputFormat to ensure
> compatibility
> ------------------------------------------------------------------------------------
>
> Key: HIVE-20678
> URL: https://issues.apache.org/jira/browse/HIVE-20678
> Project: Hive
> Issue Type: Bug
> Components: HBase Handler, Hive, Spark
> Affects Versions: 4.0.0
> Reporter: Alice Fan
> Assignee: Alice Fan
> Priority: Major
> Attachments: HIVE-20678.1.patch, HIVE-20678.2.patch
>
>
> This issue was caused by same reason described in
> https://issues.apache.org/jira/browse/HIVE-11166. As the original Jira seems
> to be abandoned, so create this one to resolve the issue.
> HiveHBaseTableOutputFormat does not implemented HiveOutputFormat, it will
> break the compatibility in spark's API when it is using hive.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
