topic Using DAX variables in the filter function in DAX Commands and Tips

Using DAX variables in the filter function

Anonymous — Tue, 09 Nov 2021 20:29:25 GMT

Hello,

I'm trying to create a measure and at a given moment I use the following expression:

Top =
VAR Nth = 2
VAR Category = ALLSELECTED(Table[Category])
VAR Sales =
ADDCOLUMNS(
VALUES(Table[Category]),,"Rank",RANKX(Category,[Total Sales]))

VAR FilterNThCategory = FILTER(Sales , [Rank] = Nth)
VAR NthLineName = MINX(FilterNThLine,Table[Category])
VAR Subtop = CALCULATE([Total Sales],Table[Category] = NthLineName)

return Subtop

in theory the return of the variable "NthLineName" is a string, but when I use it inside the FILTER function, I don't get the expected result. However, when I replace "NthLineName" with the string "Cars", it works perfectly.

Does the FILTER function not support the use of variables or am I doing something wrong? is there any way to get around it?

thank you very much in advance

Re: Using DAX variables in the filter function

AlexisOlson — Tue, 09 Nov 2021 21:25:12 GMT

Does it help if you replace VALUES ( Table[Category] ) with Category?

If not, then can you provide some example data and desired results to test against?

Re: Using DAX variables in the filter function

v-cazheng-msft — Fri, 12 Nov 2021 02:55:08 GMT

Hi Anonymous

Please make a little bit change to your measure.

Top =

VAR Nth = 2

VAR Category =

ALLSELECTED ( 'Table'[Category] )

VAR Sales_ =

ADDCOLUMNS(

VALUES ('Table'[Category]),

"Rank", RANKX ( Category, [Total Sales] )

)

VAR FilterNThCategory =

FILTER(Sales_,[Rank]=Nth)

VAR NthLineName =

MINX(FilterNThCategory,'Table'[Category])

VAR Subtop =

CALCULATE ( [Total Sales], 'Table'[Category] =NthLineName)

RETURN

Subtop

Result should look like this:

Attached the pbix file as reference.

Best Regards,

Community Support Team _ Caiyun

If this post helps, then please consider Accept it as the solution to help the other members find it more quickly. If you still have problems on it, please feel free to let us know. Thanks a lot!