Answer / shridhar kasat

For Unique and Duplicate , we can go for same flow :

Source-->SrcQualifier-->Aggregator(group
by this column and find count also)--> then use joiner (use
the column from Source qualifier and join the data from
Aggregator as source qualifier.col=aggregator.col) -->
Router (insert into t1 for count>1 and for count=1 insert
into t2). t1 is for duplicate records and it will cintain
1,1,3,3 and t2 will have 2,4.

Records in Source Q :
col
1,1,2,3,3,4

The records after agregator will look like this :
col count
1 2
2 1
3 2
4 1

Records after Joiner : source q.col=aggregator.col
col count
1 2
1 2
2 1
3 2
3 2
4 1

These records will flow in router and we define two groups;
one for count>1 (for duplicates record) and second for
count=1 ( for unique record)

Hope this answer suffice your requirements.

Regards
Shri

Answer / suman chattopadhyay

Shirish,

Part 1 of your answer is not correct. If you use distinct
option in sorter you will get all the distinct values even
for the duplicate. i.e if the values are 1,1,2,3,3,4, you
will get 1,2,3,4 in target.

Part 2 of your answer will do all the trick for duplicate
and unique ones. If you use group by and count the number
of occurences you will get 1 as the count for unique values
and more than one as count for duplicate values. Pass those
records with count 1 to 1 target and the other to the 2nd
target.

Hope this helps.

Suman

Answer / sanju_yu

source> dynamic lookup>router,2 conditions in router
1. condition if column_lkp port is null then insert into
target1(unqie)
2. condtion if COLumn_lkp port is not null then insert into
target2(duplicates

Plz let me know if am wrong

Sanjay

Answer / shirish

1.For unique: Source-->SrcQualifier--> sorter (use distinct
option)--> Target
2. For duplicate : Source-->SrcQualifier-->Aggregator(group
by this column and find count also)-->filter(use count
greater than 1 )--> left outer join with source qualifier --
> target 2.
Hope i am clear.

Answer / dinesh

can u give prepare solution

Answer / rahul

hi sanjay,

will this logic work when we are loading for the 1st time

What is the difference between procedure and stored procedur?

2 Answers   IBM,

if i hv 6 table as source table but can i make 12 as a dim table and 6 as fact table belongs to that src table.

0 Answers   iGate,

I am hvaing SOURCE as first line: 1000,null,null,null second line as:null,2000,null,null 3rd line as :null,null,3000,null and final line as: null,null,null,4000 ............................Now i want the OUTPUT as 1000,2000,3000,4000 For more clarification i want to elimate nulls and want in a single line. Please help me out

2 Answers   IBM,

I have a flat file, in which i have two fields, emp_id, emp_name. The data is like this, emp_id,emp_name 101,soha 101,ali 101,khan 102,siva 102,shanker 102,reddy. how to merge the names so that my output is like this Emp_id Emp_name 101 Soha ali kahn 102 siva shenkar reddy please provide solution

5 Answers  

Which is costliest transformation? costly means occupying more memory?

3 Answers  

-Which expression we can not use in Maplets?, -Can we join(relate) two dimensions in a schema? -Why and where we use 'sorted input' option?

1 Answers   TCS,

Write the prerequisite tasks to achieve the session partition?

0 Answers  

can any one explain me about junk dimension

4 Answers  

What is different between the data warehouse and data mart?

0 Answers  

My source table look like ID Name Location Salary 1 ran NULL NULL 1 NULL DELHI NULL 1 NULL NULL 1000 I want my output to look like ID Name Location Salary 1 ran Delhi 1000 What should be my approach to design a informatica mapping. Assuming my source is flat file not a Relation table so i cant use SQL.

2 Answers   BirlaSoft,

COL1,COL2 ABC,1 XYZ,2 HERE IN COL2 VALUES 1,2 NOT STSANDARD(IE MEANS NOT FIXED VALUES LIKE OTHER SOME VALUES LIKE 10,20) O/P IS COL1,COL2 ABC,2 XYZ,1

0 Answers  

why do u use shortcuts in informatica.?

4 Answers   iFlex,