Aggregation problem in graphical HANA calculation view

[stefan_voltz1]

Hello,

I am puzzled by the aggregation behavior in in the Join node of my graphical calculation view. I hope someone can give me a hint on how to get the result I expect.

The scenario is the following:

• There are two projections I need to join. I am using an Inner Join.
  

  

  
  
  In the Join node there is one Calculated Column RECORD_COUNT set constantly to "1".
  
• The projection on the left (Druckbeleg_Kopf) contains the following data:
  There are three document numbers (OPBEL).
  

  

• The projection on the right (Druckbeleg_Abrechnungsbelege) contains the following data:
  Each of the three document numbers (OPBEL above, PRINTDOC in this projection) contains four records with different BILLDOC values.
  

  

• The RAW DATA of the Join node looks as expected. There are 12 records with all available combinations of OPBEL/PRINTDOC and BILLDOC.
  

  

  
  
• So far so good.
  In the Analysis tab, we can see the 12 records as well.
  

  

  
  
• When removing field OPBEL, the aggregation of RECORD_COUNT looks as expected on field BILLDOC:
  

  

• Now when removing field BILLDOC and dragging field OPBEL back into the labels axis, RECORD_COUNT is not aggregated as I expect.
  I would expect the sum of RECORD_COUNT to be 4 for each entry in OPBEL. Because as we can see in the RAW DATA of the Join result, there are four records each.
  

  

Can someone explain why the aggregation behaves as it does?

And more important: Can anyone give me a hint to how I do get the result I expect? How do I get the aggregation of RECORD_COUNT based on the joined data set?

I appreciate any help.

Thanks a lot in advance 🙂

[philippe_dobrowlanski · Accepted solution]

hello,

try to change the keep flag proprety to true on your field BILLDOC at the aggregration node.