[HIVE-2597] Repeated key in GROUP BY is erroneously displayed when using DISTINCT

• Type: Bug
• Status: Closed
• Priority: Major
• Resolution: Fixed
• Fix Version/s: 0.14.0
• Assignee: Navis Ryu
• Reporter: Alex Rovner
• Created: 18/Nov/11 20:06
• Updated: 13/Nov/14 19:43
• Resolved: 30/Jun/14 15:22

Description

The following query was simplified for illustration purposes.

This works correctly:

select client_tid, "" as myvalue1, "" as myvalue2 from clients cluster by client_tid

The intent here is to produce two empty columns in between data.

The following query does not work:

select distinct client_tid, "" as myvalue1, "" as myvalue2 from clients cluster by client_tid

FAILED: Error in semantic analysis: Line 1:44 Repeated key in GROUP BY ""

The key is not repeated since the aliases were given. Seems like Hive is ignoring the aliases when the "distinct" keyword is specified.

Issue Links

• links to: review board