Answer / kamlesh mishra

The PowerCenter Server skips the specified number of rows
before reading the file. Use this to skip header
rows in the file.
And for last rows implement following logic.

1) Take sorter transformation reverse the rows on the basis of some key field like empno.
2)take expr transformation add nextval from sequence generator
3) Take filter transformation add condition nextval !=1 .
4)(optional)reverse the rows again using sorter transformation.
5) From filter to tgt connect the respective ports.

Src---->SQ--->SORTER--->EXPR---->FILTER---->SORTER---->TGT

Answer / allen

Skipping the firt row in a flat file is easy:
The PowerCenter Server skips the specified number of rows
before reading the file. Use this to skip title or header
rows in the file.

Skipping the last row is not as easy. In my case, the last
row was a record count. I was able to check ports that
shouldn't be null, spaces or zeros and reject the row. The
LENGTH function can also be used to check the length of
data. Delimited or not has to be considered in any checks.

Answer / manoj subramanian

I have done it using sequence generator,rank and filter transformation.
Filtering first record:
Generated sequence no using sequence transformation and given
as input to rank transformation.In properties selected
top and no of ranks as maximum rank allowed in informatica
then followed by filter transformation with filter condition
as rankindex<>1.
Filtering last Record:
Followed by another set of rank and filter wit rank property as bottom.

Answer / sagarreddy

in source qualifier your using this query like this u get it
easily
select rowid from tablename
minus
(select min(rowid) from tablename
union
select max(rowid) from tablename )
if u write this query will work it plz go through it

Answer / maneesh

To eliminate first row, simply use a counter in Expression and then Filter.

To eliminate the last row, make two pipelines with the same source
In first pipeline, use AGGREGATOR without any Group By ports In second pipeline, join it with the above with JOINER. Use and Outer Join and then other than that Joined row .. load all the rows.

get me output as if input is like 1 x o/p-->1 x,y,z 1 y 2 a,b 1 z 3 c 2 a 2 b 3 c

0 Answers   Infosys,

without matching columns in two tables. how can you join

5 Answers   BirlaSoft,

What is the fact table?

0 Answers  

Can we have a Mapping without a Source Qualifier?

14 Answers   Deloitte,

what is materialized view?

2 Answers  

i did MBA in 2008. i got job as a Software Engineer(Informatica) in 2008 February in our college campus interviews through consultancy. my problem is when iam going interview HR people ask " YOu are MBA graduate how u get software(informatica) job". iam saying i got job in campus interviews. i have knowledge in informatica in dataware housing. is this answer correct or not. plese give me guidence

1 Answers   IBM,

Hi, In Router transformation I created two groups . One is Passthrough=> True Second one is CorrectId’s => Invest>50000 Here I have one doubt. Can’t I treat default group as Passthrough group (fist group) . Is there any difference between default group and Passthrough group in this scenario? Let me know if you want more information about this scenario. Advance thanks.

3 Answers   IBM,

what is the logic will you implement to load data into a fact table from n dimension tables?

4 Answers   TCS,

Diffrence between 7.1,8.1 and 8.6

3 Answers   HCL, Wipro,

Two Default User groups created in the repository are ____ and ______

2 Answers  

Whats there in global repository

0 Answers  

What is sorter transformation?

0 Answers