[
https://issues.apache.org/jira/browse/HIVE-2354?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13087938#comment-13087938
]
Syed S. Albiz commented on HIVE-2354:
-------------------------------------
Moved a bunch of the logic to update index timestamps into the AlterIndex
DDLTask rather than the DDLSemanticAnalyzer, so that the timestamp is
guaranteed to be updated *after* the base table is touched. This change is
reflected in index_auto_update.q, as now we see that the check for index
staleness in the SELECT passes and we use the index in index_auto_update.q as
desired.
> Support automatic rebuilding of indexes when they go stale
> ----------------------------------------------------------
>
> Key: HIVE-2354
> URL: https://issues.apache.org/jira/browse/HIVE-2354
> Project: Hive
> Issue Type: Improvement
> Reporter: Syed S. Albiz
> Assignee: Syed S. Albiz
> Attachments: HIVE-2354.1.patch, HIVE-2354.2.patch, HIVE-2354.3.patch,
> HIVE-2354.4.patch, HIVE-2354.5.patch, HIVE-2354.6.patch
>
>
> Support a mode where indexes will be automatically rebuilt when the
> table/partition they are based on is modified. So if index foo is built on
> table bar, and bar has it's contents overwritten, we should support a mode
> where index foo will automatically rebuild itself.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
