DSXchange

Hi,

I have data like the following:

Col1 ColA ColB ColC ColA1 ColB1 ColC1
--------------------------------------------------------------

123 abc def ghi jkl mno pqr
456 stu vwx yza bcd efg hij



My requirement is that I need the data in the following format:


Col1 ColA ColB ColC
------------------------------
123 abc def ghi
123 jkl mno pqr
456 stu vwx yza
456 bcd efg hij


Could anyone please let me know which stage i need to use and a bit of insight into the logic..

Thanks.

Transformer stage with two outputs followed by a Funnel stage. The key goes to both output links. The first three columns go to the first output link, the other three columns go to the second output link.

ray.wurlod wrote:Transformer stage with two outputs followed by a Funnel stage. The key goes to both output links. The first three columns go to the first output link, the other three columns go to the second output ...

Hi Ray,

That's a cool way of doing it. Instead, you think I can use a Copy stage and Funnel stage to funnel the data into single output..? Is there any particular need to use the Transformer here?

Thanks.

Hi,
No Need Transformer stage ,you can use copy stage ,have two output links and rename the 2nd link column names to first link column names before funneling.

Excellent.

(My server job background came through for a bit, there.)

ravireddy25 wrote:Hi,
No Need Transformer stage ,you can use copy stage ,have two output links and rename the 2nd link column names to first link column names before funneling.

Cool.. That's exactly.. what i had in mind. I mean, the Funnelling wont work if I dint. Thanks..

ray.wurlod wrote:Excellent.

(My server job background came through for a bit, there.) ...

It's the same with me. It's just the necessity that I have to decrease the complexity of the job. I am from server jobs background too. Thanks Ray.

Hi,

Can we do all these operations using a pivot stage??

samratisking wrote:Hi,

Can we do all these operations using a pivot stage??

Yes, it's possible

Only if you combine the triplets of columns into single columns (then back again into separate columns after pivoting if needed).