Want to get involved? We're always looking for ideas and content for Weekly Challenges.
SUBMIT YOUR IDEAStruggled with this one, I think I knew I needed some kind of cartesian join, which in my head meant Append Fields, but I then struggled to filter out all the cases that weren't needed. Turns out a simple L-R join did the job much better because you can join by UserID. Then it was a relatively simple job to filter out inverted pairs and left=right cases.
Looks different than the output file as I put the non-null values in the Merchant1 column but the results are the same. My solution:
Here is my solution to this rather old challenge.
Btw I think the given solution is wrong. There should be only 65, not 77 records, in the output. The solution contains 12 records where the first merchant is null.
In my solution, I adopted a statistic method of finding the number of combinations.
Dawn