Managing the first application of the MERGE statement in an ETL logic - SQL Server 2012

Question

Hi,
in order to feed a fact table of a dwh from a staging table I'm using the MERGE statement in order to control insert and update operations. It is possible that the staging table has duplicate rows respect to the fields controlled in the merge condition:
when I run the first time the MERGE statement unwanted rows could be inserted in the fact table.
Does the MERGE statement allow to manage this case or do I need to filter data from the staging table before to write them into the fact table?
Many thanks

Answer 1

Hi Anuj,
thank for your reply. Unfortunately the DISTINCT in the using clause isn't enough for my case: I need to read a subset of fields of a staging table and for this subset I could have duplicate rows, but also I need to the change date of the rows inserted in
the staging table and I cannot include this date in the distincted columns.
However the answer is that I need to handle the staging data before to write them into the fact table.
Thanks
Yes
One more thing to note is when running MERGE statement subsequently in case there are multiple records which matches the update criteria then default behavior of MERGE is to throw an error indicating that it tried to update same record twice.
So in any case you need to make sure you eliminate duplicates before applying MERGE over it at first.
Please Mark This As Answer if it solved your issue
Please Vote This As Helpful if it helps to solve your issue
Visakh
My Wiki User Page
My MSDN Page
My Personal Blog
My Facebook Page