DSXchange

Hi,

I have a very headache problem. I have a sequential file which contains more than 3 million records, every records has around 20 fields. The file contains product information about customers.

Here is the samples

Customer_ID, GAS, Eelectricty,Heating_Protection_Plan, Cooling_Protection_Plan, ....
1000,1, 0, 1, 0,....

The value 1 indicates customer has the product, 0 means not. The requrement is to genearte a producting_holding string (by concacatenating the product that customer owns), , but it is also required the sequence of each product within the string should be decided by a product table.

The sample of product is listed as below

Product sequence
Heating_Protection_Plan 1
Electricity 2
Cooling_Protection_Plan 3
GAS 4
......
For above example, the result would be 'Heating_Protection_Plan GAS", besides, the product table changes every months. I could not figure out any easy way to do in Datastage, any suggestions?

Thanks,

Carol

If you use your product sequence table in a multirow lookup, but don't specify a key, you should get all the rows from the table returned in the lookup in the order you want. Then you can string together the product name if there is a 1 on input. Then output that string when the customer id changes.

How about stage variable
variablesum, variable key

where variablesum becomes variablesum:filed:variablekey when field value is 1 or remains variablesum
and the variablekey becomes variablekey+1 when field value is 1 and remains variablekey when field value is 0..........

You have option to distribute the record and collect them again.

Hai Sainath,

Can you provide some more inputs on how to distribute and recollect to achieve the result...please provide the logic to implement the same..

Thanks
Sudharsanan A N

Sudharsanan,

There is no straightforward command or stage that can do what I tried to convey. It is a set of tasks on its own involving few steps and having more information of the requirement.

That is why I left it to be decoded by the reader.