[
https://issues.apache.org/jira/browse/HIVE-15944?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15960256#comment-15960256
]
Ashutosh Chauhan commented on HIVE-15944:
-----------------------------------------
Likely column order is changed in ColumnPruner. Better approach here is to
generate columns in correct order in Column Pruner while visiting Limit
Operator.
> The order of cols is error in ColumnPrunerReduceSinkProc because of sort
> operator
> ---------------------------------------------------------------------------------
>
> Key: HIVE-15944
> URL: https://issues.apache.org/jira/browse/HIVE-15944
> Project: Hive
> Issue Type: Bug
> Components: Query Processor
> Affects Versions: 1.1.0, 2.2.0
> Reporter: wan kun
> Assignee: wan kun
> Fix For: 2.2.0
>
> Attachments: HIVE-15944.1.patch,
> HIVE-15944.2-branch-1.1.1.path.erroroutput, HIVE-15944.3.patch,
> HIVE-15944.4-branch-1.1.1.patch, HIVE-15944.4.patch, HIVE-15944.5.patch,
> HIVE-15944.6.patch, HIVE-15944.7.patch, HIVE-15944.8.patch,
> HIVE-15944-branch-1.1.patch, HIVE-15944.patch, STAGE_DEPENDENCIES
>
> Original Estimate: 168h
> Remaining Estimate: 168h
>
> If one sql have two job.
> job 1:The order of cols is updated in ColumnPrunerReduceSinkProc because of
> sort operator.
> job 2 will read error in map operation because the cols order is old.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
