Filtered measures using SQL filter expressions

Beto
Participant II

The problem starts when looker suggests that you can use sql as a filter expression inside a measure, but I have been looking around and I can’t find anything.

I need to filter a measure based on a condition. example:

measure: dinamyc_tons {

type:sum

sql: ${Tons}

filters: [month: (month calculated by another dimension)]

}

It is easy to add month: “December” or month: “12” but instead or hardcoding the filter and the measure, i need it to be dinamyc and change based on the value provided by another dimension.

if you stop typing after filters:[] , looker throws an error that says:

Expecting '[', 'string', 'numeric', 'yes', 'no', 'asc', 'desc', 'sql', ';;', 'expression', 'html', 'js', 'identifier', '-', '*', got ']'

from there i understand i can then use sql to perform the filtering action but i can’t find how, can you guys help me ???

0 2 3,766
2 REPLIES 2

tonya_mina
Participant I

Hi Beto, 

try to add another new dimension of type YesNo, which will check if month is equal calculated value

and then add that new dimension as a filter for your measure.

Beto
Participant II

Thanks Tonya, i tried to implement something like that but it didnt work.

I need to grab the user input for the date filter on the dashboard and then based on that date i perform certain calculations, one of those calculations is to set a value in another dimension which is the one i really want to filter by.

Because in my DB table i have all possible combinations (catalogue) when i compare to a yes/no it will always return yes and then perform the wrong operation.

At least thats what i think its happening, because i checked and it says yes all the time and then returns a measure with wrong values.

Top Labels in this Space
Top Solution Authors