Some Eclipse Foundation services are deprecated, or will be soon. Please ensure you've read this important communication.

Bug 224320

Summary: The dataset count aggregation return 0
Product: z_Archived Reporter: Tianli Zhang <tzhang>
Component: BIRTAssignee: Xiaoxiao Wu <xwu>
Status: RESOLVED FIXED QA Contact:
Severity: normal    
Priority: P3 CC: bluesoldier, lzhu, wenfeng.fwd, whe, xxue
Version: 2.3.0Keywords: plan
Target Milestone: 3.7.0 RC3   
Hardware: PC   
OS: Windows XP   
Whiteboard:

Description Tianli Zhang CLA 2008-03-27 05:03:21 EDT
Description:
The dataset count aggregation return 0, so I guess they do aggregate at the same time right? Do we support to aggregate on another computed column which has aggregation?

Build number: 
2.3.0.v20080327-0800

Steps to reproduce:
1, New a dataset from customer
2, New a computed column COUNT on STATE
3, New another computed column COUNT on the computed column in step2

Actual result:
In step3, the second computed column return 0;

Error log:
Comment 1 Lin Zhu CLA 2008-03-27 07:05:46 EDT
This is about the nested aggregation support in data set level. Need to discuss whether to support this feature.
Comment 2 Xiaoxiao Wu CLA 2011-05-20 02:23:07 EDT
Since this kind of aggregation is not supported, we made the enhancement for the data set computed columns:
we will prevent the creation of those computed columns that refer to other computed column(s) involving aggregation.
As for the UI layer, an error message will be displayed to indicate the invalid behavior. 

Now mark this bug as FIXED.