Answer / reddevilzzzz

@ Shalu
Your answer is almost correct. The question says, you have
to use Aggregator transformation.
Select all the rows from SQ.
Pass them to aggregator transformation. Group By on all
ports.
Create a Output port in Aggregator(lets call it TOTAL) and
give expression as COUNT(Col1).
Create a Router transformation, with 2 groups. In one group
(lets call it UNIQUE), put condition as TOTAL = 1.
In another group (lets call it DUPLICATES), put condition
as TOTAL>=2.
Pass the output from UNIQUE group to table where we want
unique rows.
Pass the output from DUPLICATE group to table where we want
duplicate rows.

P.S - tried and tested :):)

Answer / shalu

giving one example, lets say my table temp is having
following -

col1 col2
1 2
1 2
1 2
3 4
3 4
5 6

In the SQL Qualifier, override the query as
select col1,col2,count(1) total from temp group by col1,col2

which shows the output as
col1 col2 total
1 2 3
3 4 2
5 6 1

Now, use one router transformation where one condition is
where total >=1
and second condition where total>1

So first condition will return you all the unique records
1 2
3 4
5 6


and second condition will return you duplicate records
1 2
3 4

Answer / sankar

AS PER Reddevilzzzz ANS ALMOST OK

BUT NO NEED TO SELECT GROUP BY IN AGGREGATOR T/R BCOZ IF U
SELECT GROUP BY THERE IS NO DUPLICATE SO WITH OUT
DUPLICATES HOW WE PASS THE DUPLICATES IN T2.

Answer / hitesh

but in this we ll get distinct values in duplicate table.
how can i get all values in duplicate table like:
col1 col2
1 2
1 2
1 2
3 4
3 4
5 6
and i want
unique table:
5 6
and
duplicate table :
1 2
1 2
1 2
3 4
3 4

i know this is of no use, but can we do this??
pls rply

Answer / nikita jain

for this query we can use aggregator row wise calculation and handle then via router
col1 col2
1 2
1 2
1 2
3 4
3 4
5 6

O/P
Table with unique records:
1 2
3 4
5 6

Table with rest of the records
1 2
1 2
3 4

After SQ take a sorter transformation sort on col1 asc then an expression transformation
col1
col2
v_count iif(col1=prev_col1 and col2=prev_col2, vcount+1,1)
o_count v_count
prev_col1 col1
prev_col2 col2

Take a Router transformation , make 2 groups
Group1 : 0_count=1
Group2 : Default (it will come automatically)

Connect first group with unique target table
and second with other table

Answer / lokendra

wt ever above saying is correct.

Answer / srinu

I have a idea after sql transformation go thruogh 2 Agg
Trans,2 Router Trans
Agg1-gorup by col count=1 to router trans
Agg2-group by col count<>1 to router trans

I am not confident check itonce let me know,,

Thanks
Srinu

What is staging area?

11 Answers  

What are connected or unconnected transformations?

0 Answers   Informatica,

What is a shortcut and copy in Informatica and how two are different with each other?

0 Answers  

What are pre and post-session shell commands?

0 Answers  

Can we create two repositories on same database instance in oracle?

0 Answers  

Stage is which type of load is it a full load or incremental load Please let me know in detail with best example

3 Answers   Infosys,

How to transform normalized data to denormalized form in informatica? Is there any logic or any transformations to achieve this?

3 Answers  

What are the various test procedures used to check whether the data is loaded in the backend, performance of the mapping, and quality of the data loaded in informatica?

0 Answers  

HOw can we load the normalised data ( Vertical data) to (Horizontal data)with out using decode in the expression transformation and the aggregator transformation. But, what exactly I want is .. If we have million of records,it is not possible using decode and search for that Particular Id to pivot into Horizontal data... Can any one of guys please help me in this regard.....

3 Answers   TCS,

HOE DO U IMPLIMENT SCHEDULING IN INFORMATICA?

7 Answers   Patni, Wipro,

What is the term PIPELINE in informatica ?

7 Answers   Deloitte,

8.what is mapplet ?give example?

2 Answers