Field that returns 1/0 based on all similar values in the same column

PatM
PatM
in Magic ETL

Hello,

 

This should be pretty simple but I can't find the solution. I tried fiddling around in ETL and beastmode, as well as googling, but was unsuccessful.

 

Here is the problem:

 

I am trying to create a column that returns 1 when any unique Job has a PO#. For example, in the second row there is no PO, however the 1st row is also Job 1 and has a PO. So I want row 2 to return 1.

 

Job

PO #

Goal

1

123

1

1

 

1

2

124

1

2

125

1

3

 

0

3

 

0

3

 

0

 

Thank you in advance for your help.

Comments

  • ST_-Superman-_
    ST_-Superman-_ Coach

    If you are creating a table card, and you have it sorted by `Job` then you should be able to get something like this to accomplish what you are looking for:

    COUNT(DISTINCT `PO #`) OVER (PARTITION BY `Job`)

    However, if you sort your table by something else, then this beastmode will break


    “There is a superhero in all of us, we just need the courage to put on the cape.” -Superman
  • PatM
    PatM

    Hi there, I tried your method and unfortunately I did not manage to make it work.

     

    However, I copied the idea by doing a rank & window in the dataflow, with a function count of PO's and partitioned by Job's. It is a bit of patchwork as it was mandatory that I had a "frame range" (which I would rather not have) but other than that, it has worked.

     

    Thank you

  • ST_-Superman-_
    ST_-Superman-_ Coach

    You could keep it unbounded in a redshift dataflow


    “There is a superhero in all of us, we just need the courage to put on the cape.” -Superman

Categories