Plz help me on below tech tip

[nagarjuna.dudam]

Hi All,
can anybody help me on this below.

1)I have a source below.
a
a
a
b
b
c
c

I need output like below.
a1
a2
a3
b1
b2
c1
c2



2)
src:
a
b
a
b
c
d

I need the result should be as non duplicate values like below.
c
d

Thanks,
Nag

[Ravi.K]

What is your source? Whether it is Database (Oracle / DB2) or Flat file.

If it is Database then you can handle at Database itself by using the below queries.

1. select col||row_number() over(partition by col order by col) from yourtable;

2. select col from yourtable group by col having count(col) <=1;

The above queries are work for Oracle if your database is DB2 then simulate accordingly.

[ray.wurlod]

(1) Sort and partition data using Sort stage generating key change column then use stage variables to increment/reset counter depending on whether a change has occurred.

(2) Fork join. Make a copy of your data (Copy stage) and run it through an Aggregator to count the rows for each group. Downstream of that join the two streams then filter based on count = 1.

[XRAY]

2)
a) One of the way is to use column generator ( assign 1 to a new column ) - Aggregator ( Sum ) - Filter

b) Another way is

Parallel Aggregator ( Count Row ) -> Sequential Aggregator ( Sum ) with Sort Merge -> Filter

[vishal_rastogi]

for the second thing you can use the aggregator and then use where link count = 1 in the transformer constraint

[Vidyut]

for 1st--
StageVar1 = (If DSLink2.Col1 = StageVar Then StageVar1 +1 Else 1)
StageVar = (DSLink2.Col1 )

Output
DSLink2.Col1:StageVar1

for 2nd: vishal_rastogi has already answered