Extract unique value from column using DAX

[Jorsch]

I've come across what I thought should be a simple problem, but I can't quite figure it out. I have a table that's the result of an expression that could have a column like this in certain instances:

Row Key Index

AH4000	1
AH4000	2
AP9999	3

<tbody>
</tbody>

What I want is to keep only rows with a unique row key so that I can get the correct single index value (the red value). Using distinct/values wouldn't help because it would still show the 1st or 2nd row in the table, where as I want those rows that contain a row key that exists in the table more than once to be fully excluded. I can't rely on position of the index because it varies, so I don't think min/max would work. I have this calculation as part of a complicated calculated column so there are memory concerns as well. Is there an easy answer to this that I'm missing?

Thanks,
Josh

 

[macfuller]

If I understand your question, you want your filter to exclude all rows where the row key appears more than once and to capture the index value. What about a COUNTROWS?

[Permitted Index] =VAR MyIndex = MyTable[Index]
RETURN
    IF (
        CALCULATE (
            COUNTROWS ( MyTable[Row Key] ),
            FILTER ( ALL ( MyTable ), MyTable[Row Key] = EARLIER ( MyTable[Row Key] ) )
        ) > 1,
        BLANK (),
        MyIndex
    )

Shame on me, I can't remember if the countrows returns the original row too (I think so). If not, you would have to set the test to > 0.