Reputation: 2762
In a Data Flow I get the data with a Script Component that queries a webservice. The output is a table with payments from customers.
| PaymentId | CustomerId | Amount | PaymentDate |
| 2 | 1 | 10 | 01/01/2009 |
| 5 | 1 | 15 | 08/07/2013 |
| 1 | 2 | 20 | 05/08/2007 |
| 3 | 2 | 18 | 07/03/2011 |
| 6 | 2 | 32 | 02/09/2013 |
| 4 | 3 | 70 | 04/06/2012 |
| 7 | 3 | 61 | 02/03/2013 |
Among other things, I have to filter out following rows per customer:
And for each I need the PaymentId, CustomerId, Amount and PaymentDate for later processing.
I know how to use the Aggregate Transforamtion, group by CustomerId and use the appropriate Min/Max functions to get the requested VALUES.
But what is the best approach to get the complete ROWS?
Thanks in advance!
Upvotes: 0
Views: 908
Reputation: 1317
If you're data source is a SQL database, do what you can with SQL, before you get to SSIS. This will generally give you the best performance.
Otherwise, suggest dumping data from the web service into a RAW
file. Raw files are quite useful for staging data to do assorted transformations. more on RAW files
Upvotes: 1