I have an unusual request. I'm trying to remove duplicates, but as long as they're in ANY of given columns. I have LCD serial numbers, and sometimes they are present in different rows in different orders. (Don't ask, Windows randomly switches LCD order)
so, for example
PC# LCD_SN1 LCD_SN2 LCD_SN3
i001 123 624 791
i001 624 791 123
In both cases, both entries have the same connected LCDs. But I want to remove all other entries, as long as it has all of the connected LCDs and the PC# matches. So, the pseudocode would be-
remove duplicate rows that match PC# AND all of LCD_SN#s have the same values, in any order. So, if multiple entries with same PC# have lessSNs (as in missing values), remove those as well. Since they're incomplete.