This challenge is about finding second highest salary for each department...

I didn't even test this....

I didn't even test this....

I learned this recently when working with a different data source. I expect this version not to time out. Instead of using a correlated subquery, I made the magic happen in the join.

I learned this recently when working with a different data source. I expect this version not to time out. Instead of using a correlated subquery, I made the magic happen in the join.

I used a with statement to get a signed amount (neg for payments) and create a rownumber for the sort. then in main, I got the 4 main columns and used a sum(signed amount) where custid = custid and rownum(outside)<=rownum.

I used a with statement to get a signed amount (neg for payments) and create a rownumber for the sort. then in main, I got the 4 main columns and used a sum(signed amount) where custid = custid and rownum(outside)<=rownum.