cancel
Showing results for 
Search instead for 
Did you mean: 

Group By Error

Former Member
0 Kudos

I created a view with the following mappings in in the output schema:

Column 1 :     "VENDOR_BANKING".BANKN_Bank_Account_Number

Column2 :     substr("VENDOR_BANKING".BVTYP_Partner_Bank_Type, 1, 3)

Column 3 :     Count("VENDOR_BANKING".BANKN_Bank_Account_Number)

To enable the aggregation of C3, I obviously want to GROUP BY the first 2 columns.

However, in the Group By tab, there are only 2 options:

> Propose Columns

> Remove

"Propose" adds in all non aggregation columns, but ignores my custom mapping.

So where above, I have selected a SUBSTR of field BVTYP,  the Propose action actually adds in :

"VENDOR_BANKING".BVTYP_Partner_Bank_Type

.
i.e. the Entire column, with no substring.

For the life of me, I cannot see an option to manually edit this mapping !?

Obviously this will produce incorrect output .... can anyone advise if I'm missing something here ?
or is a bug ?

Many Thanks

Accepted Solutions (1)

Accepted Solutions (1)

former_member187605
Active Contributor
0 Kudos

Not a bug, just missing fuctionality.

Aggregate views in IS group by columns only, no expressions allowed.
Try your luck at Idea Place: Enterprise Home.

My workaround? I know it defeats the purpose of the tool but I define the view at database level.

Former Member
0 Kudos

Thanks for confirming that Dirk, at least it's good to know I'm not missing any understanding at my end.

Bug vs Missing functionality ... hmm, we'll agree to differ on that one

Yes, likely I'll end up going to the DB for a work around as well.
It's a shame because this is the 3rd workaround in as many days, and you're right ... it is absolutely defeating the purpose (& Cost) of the tool.

Let's hope SAP pull their finger out and get some fixes sorted.

0 Kudos

Hi Simon,

If you are using IS 4.2, you can achieve the desired result by defining the view in 2 stages:

1) Create IS view (IS_VIEW_1) to select:

Column 1: "VENDOR_BANKING".BANKN_Bank_Account_Number

Column 2: substr("VENDOR_BANKING".BVTYP_Partner_Bank_Type, 1, 3)

2) Create IS view based on view created above:

Column 1: <IS_VIEW_1>.<Column 1>

Column 2: <IS_VIEW_1>.<Column 2>

Column 3: count(<IS_VIEW_1>.<Column 1>)

GROUP BY <IS_VIEW_1>.<Column 1>, <IS_VIEW_1>.<Column 2>

Regards

Nigel

Answers (0)