Find out duplicate records using aggregator

How to find out duplicate records using aggregator. Pls explain with example. What is the need for placing joiner after aggregator. I just tried like the following:

Count the records and group by all the columns. if the count=1 then it will be a valid record. what to do if the count >2.means file contains duplicate records.
Why we need to use joiner after aggregator?

buvanalogu
Profile Answers by buvanalogu Questions by buvanalogu
Dec 13th, 2007
12
16686

Questions by buvanalogu

Informatica

Answer

Showing Answers 1 - 12 of 12 Answers

sekhar.ganta
Profile Answers by sekhar.ganta Questions by sekhar.ganta

Dec 17th, 2007

When we are removing duplicate rows using aggregator we have to check all the ports by group by ports.
Without using summation, we can eliminate duplicates by using aggregator.

arun.menon
Profile Answers by arun.menon

Oct 24th, 2008

With respect to your query. Try getting the session log details regarding the session run. In case if you are not able to get sufficient informations, running the workflow in Verbose data mode. That will give you an exact account where things went wrong.

Cheers
Arun

priyank_divya
Profile Answers by priyank_divya

Aug 7th, 2012

Hi,

Its similar to the SQL query,

SELECT * FROM ,,....
FROM TABLE_NAME
GROUP BY ,,.....
HAVING COUNT(*)=1

Similarly in Informatica Aggregator transformation, select group by for all the columns and add one output port,OUT_CNT_RCRDS=count(*)

In the next transformation, use a Router transformation and put a condition,
G1_OUT_CNT_RCRDS=1
G2_OUT_CNT_RCRDS>1

G1_OUT_CNT_RCRDS --> TGT_NO_DUPLICATES
G2_OUT_CNT_RCRDS --> TGT_DUPLICATES

Hope this helps.
Thanks.

VISHNU VARDHAN

Nov 17th, 2015
Use HAVING Condition in SELECT clause, You can write code like below to test, if it is greater than one then it is duplicate otherwise Unique.
Code
SIMPLE EXAMPLE I AM WRITTEN HERE


SELECT EMPNO,COUNT(*) FROM EMP


GROUP BY EMPNO


HAVING COUNT(*)>1;