Unduplicating a file

[DSkkk]

There is a file consisting of Name, Phone Number. My job is to verify if the records have duplicate phone numbers with the same name. If say a person A appears twice in the file,
First time with Phone No 1 and second time with Phone No. 2 then only the record with Phone no. 1 has to survive where as the record with phone no 2 should go into the residual (or Reject file).
How Can I achieve this? Can anyone help me out on what basis do I need to proceed i.e., my Un-dup option and the extract file option.

Thanks.
Kalyan

[kcbland]

Read the file and use the Aggregator stage. Use the FIRST derivation for the phone number, group by the name. You'll only get the first phone number as output for each customer name.

[ashok]

There is a file consisting of Name, Phone Number. My job is to verify if the records have duplicate phone numbers with the same name. If say a person A appears twice in the file,
First time with Phone No 1 and second time with Phone No. 2 then only the record with Phone no. 1 has to survive where as the record with phone no 2 should go into the residual (or Reject file).
How Can I achieve this? Can anyone help me out on what basis do I need to proceed i.e., my Un-dup option and the extract file option.

Thanks.
Kalyan

Hi Kalyan,
use match stage in which use un-dup option, when you create the pass use customer name as blocking strategy, then compare the customer and phone number, refer stage guide how to build match stage,
Then use survive stage and give the necessary rules for survivorship of records,

Bye,
Ashok

[PilotBaha]

Better than anything else: Read the matching strategies document that comes with the software.. There is a very deep explanation on how to construct match stages, including UNDUP