Forum Discussion

rpinxt's avatar
rpinxt
Icon for Solution Sage rankSolution Sage
3 years ago
Solved

Why do I get Circular Dependancy

Making calculated columns my first one is this : IF('AVN Rework Merge'[Flow] = "Reworked" && [Amt USD] > -50,[Quantity],0)   It works fine and gives me the quantity that I expected.   Now I want...
Need Help
  • m_alireza's avatar
    m_alireza
    3 years ago

    rpinxt Instead of referencing the measure [FinAmtUSD], try referencing the column directly instead.

    so:

    Reworked USD < 50 = IF('AVN Rework Merge'[Flow] = "Reworked" && 'AVN Rework Merge'[FinAmtUSD] > -50,'AVN Rework Merge'[FinAmtUSD],0)
  • SivaMani's avatar
    SivaMani
    3 years ago

    rpinxt I agree with m_alireza. Also, any specific reason why you went with Calculated Columns instead measures?
    I rewrote the logic as a measure for you. Check if this meets your requirement

    Reworked Qty <50__ = 
VAR Result = CALCULATE(
     [Quantity],
     FILTER('AVN Rework Merge', 'AVN Rework Merge'[Flow] = "Reworked" &&  [Amt USD] > -50)
     )
RETURN
COALESCE(Result,0)
SivaMani's avatar
SivaMani
Icon for Resident Rockstar rankResident Rockstar
3 years ago

rpinxt I agree with m_alireza. Also, any specific reason why you went with Calculated Columns instead measures?
I rewrote the logic as a measure for you. Check if this meets your requirement

Reworked Qty <50__ = 
VAR Result = CALCULATE(
     [Quantity],
     FILTER('AVN Rework Merge', 'AVN Rework Merge'[Flow] = "Reworked" &&  [Amt USD] > -50)
     )
RETURN
COALESCE(Result,0)
rpinxt's avatar
rpinxt
Icon for Solution Sage rankSolution Sage
3 years ago

SivaMani m_alireza I want to thank you both because both solutions seem to work! 😁

 

Well I chose for calculated column because I merged two sources and thought to get the correct amount on every line a calculated column would be better.

Also I tried a measure before but could not succeed 😑

 

Must say I do not fully understand your measure.

What does COALESCE do?

 

And what would you guys think is best in ways of performance?

Would the measure be better (lighter) on the data then the calculated columns?