[
https://issues.apache.org/jira/browse/HIVE-10461?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Ferdinand Xu updated HIVE-10461:
--------------------------------
Attachment: HIVE-10461.wip.patch
Hi [~spena], for record updater, we can implement it like orc did. This patch
is not completed and only used for sharing thoughts. For Merger, I think we
need to discuss how to define a data structure like <ReaderKey, ReaderPair> in
ORC. And this part is not included in my work-in-progress patch. Thank you.
> Implement Record Updater and Raw Merger for Parquet as well
> -----------------------------------------------------------
>
> Key: HIVE-10461
> URL: https://issues.apache.org/jira/browse/HIVE-10461
> Project: Hive
> Issue Type: Sub-task
> Reporter: Ferdinand Xu
> Assignee: Ferdinand Xu
> Attachments: HIVE-10461.wip.patch
>
>
> The Record updater will create the data with acid information. And for the
> raw record merger it can provide the user-view data. In this jira, we should
> implement these two classes and make the basic acid w/r case work. For the
> upper layer like FileSinkOperator, CompactorMR and TxnManager, we can file
> new jiras to fix them.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
