Learn from the best! Meet the four finalists headed to the FINALS of the Power BI Dataviz World Championships! Register now
If I have a "master" measure that will return a different calculation based on a disconnected slicer, will the variables be calculated only if needed or will they all be calculated before the RETURN section is evaluated?
For example in this measure, should I remove the variables and put the associated DAX directly in the SWITCH statement to ensure I am not calculating all possible output measures before the RETURN section determines which cacluation I actually need, or is PBI smart enough to know which variables are needed by the SWITCH and only calculate what is needed?
Time_Calculation_Total_Labor_Dollars =
VAR filter_week =
MAX ( dimSalesDates[Week_End_Date] )
VAR prior_year =
CALCULATE (
[Total_Labor_Dollars],
FILTER (
ALLEXCEPT (
dimSalesDates,
dimSalesDates[Sales_Week_Number],
dimSalesDates[Period_Number]
),
dimSalesDates[Year]
= MAX ( dimSalesDates[Year] ) - 1
)
)
VAR prior_year_growth = [Total_Labor_Dollars] - prior_year
VAR prior_year_growth_percent =
FORMAT ( DIVIDE ( prior_year_growth, prior_year ), "Percent" )
VAR prior_year_to_date =
CALCULATE (
[Total_Labor_Dollars],
ALL ( dimSalesDates ),
dimSalesDates[Week_End_Date] <= filter_week - 364
)
VAR year_to_date =
CALCULATE (
[Total_Labor_Dollars],
ALLEXCEPT ( dimSalesDates, dimSalesDates[Year] ),
dimSalesDates[Week_End_Date] <= filter_week
)
VAR ytd_growth = year_to_date - prior_year_to_date
VAR ytd_growth_percent =
FORMAT ( DIVIDE ( ytd_growth, prior_year_to_date ), "Percent" )
RETURN
IF (
HASONEVALUE ( filterTimeCalculations[time_calculation] ),
SWITCH (
VALUES ( filterTimeCalculations[time_calculation] ),
"Current Year", [Total_Labor_Dollars],
"Prior Year", prior_year,
"Prior Year Growth", prior_year_growth,
"Prior Year Growth Percent", prior_year_growth_percent,
"Prior Year YTD", prior_year_to_date,
"Current YTD", year_to_date,
"YTD Growth", ytd_growth,
"YTD Growth %", ytd_growth_percent
),
[Total_Labor_Dollars]
)
Solved! Go to Solution.
Hello @Anonymous
I believe each Variable will calc before it makes it to the switch but what you can do is turn each variable into a measure then just read your filterTimeCalculations[time_calculation] and do the switch. This also makes it easier to add additional measures to your list and the measures you do write can be used in other parts of the report.
prior_year =
CALCULATE (
[Total_Labor_Dollars],
FILTER (
ALLEXCEPT (
dimSalesDates,
dimSalesDates[Sales_Week_Number],
dimSalesDates[Period_Number]
),
dimSalesDates[Year]
= MAX ( dimSalesDates[Year] ) - 1
)
)prior_year_growth = [Total_Labor_Dollars] - [prior_year]
prior_year_growth_percent =
FORMAT ( DIVIDE ( [prior_year_growth], [prior_year] ), "Percent" )prior_year_to_date =
VAR filter_week =
MAX ( dimSalesDates[Week_End_Date] )
RETURN
CALCULATE (
[Total_Labor_Dollars],
ALL ( dimSalesDates ),
dimSalesDates[Week_End_Date] <= filter_week - 364)year_to_date =
VAR filter_week =
MAX ( dimSalesDates[Week_End_Date] )
RETURN
CALCULATE (
[Total_Labor_Dollars],
ALLEXCEPT ( dimSalesDates, dimSalesDates[Year] ),
dimSalesDates[Week_End_Date] <= filter_week
)ytd_growth = [year_to_date] - [prior_year_to_date]
ytd_growth_percent =
FORMAT ( DIVIDE ( [ytd_growth], [prior_year_to_date] ), "Percent" )Time_Calculation_Total_Labor_Dollars_2 =
VAR TimeCalcSelection =
SELECTEDVALUE( filterTimeCalculations[time_calculation], "Current Year")
RETURN
SWITCH (
TRUE(),
TimeCalcSelection = "Current Year", [Total_Labor_Dollars],
TimeCalcSelection = "Prior Year", [prior_year],
TimeCalcSelection = "Prior Year Growth", [prior_year_growth],
TimeCalcSelection = "Prior Year Growth Percent", [prior_year_growth_percent],
TimeCalcSelection = "Prior Year YTD", [prior_year_to_date],
TimeCalcSelection = "Current YTD", [year_to_date],
TimeCalcSelection = "YTD Growth", [ytd_growth],
TimeCalcSelection = "YTD Growth %", [ytd_growth_percent],
[Total_Labor_Dollars]
)
Hello @Anonymous
I believe each Variable will calc before it makes it to the switch but what you can do is turn each variable into a measure then just read your filterTimeCalculations[time_calculation] and do the switch. This also makes it easier to add additional measures to your list and the measures you do write can be used in other parts of the report.
prior_year =
CALCULATE (
[Total_Labor_Dollars],
FILTER (
ALLEXCEPT (
dimSalesDates,
dimSalesDates[Sales_Week_Number],
dimSalesDates[Period_Number]
),
dimSalesDates[Year]
= MAX ( dimSalesDates[Year] ) - 1
)
)prior_year_growth = [Total_Labor_Dollars] - [prior_year]
prior_year_growth_percent =
FORMAT ( DIVIDE ( [prior_year_growth], [prior_year] ), "Percent" )prior_year_to_date =
VAR filter_week =
MAX ( dimSalesDates[Week_End_Date] )
RETURN
CALCULATE (
[Total_Labor_Dollars],
ALL ( dimSalesDates ),
dimSalesDates[Week_End_Date] <= filter_week - 364)year_to_date =
VAR filter_week =
MAX ( dimSalesDates[Week_End_Date] )
RETURN
CALCULATE (
[Total_Labor_Dollars],
ALLEXCEPT ( dimSalesDates, dimSalesDates[Year] ),
dimSalesDates[Week_End_Date] <= filter_week
)ytd_growth = [year_to_date] - [prior_year_to_date]
ytd_growth_percent =
FORMAT ( DIVIDE ( [ytd_growth], [prior_year_to_date] ), "Percent" )Time_Calculation_Total_Labor_Dollars_2 =
VAR TimeCalcSelection =
SELECTEDVALUE( filterTimeCalculations[time_calculation], "Current Year")
RETURN
SWITCH (
TRUE(),
TimeCalcSelection = "Current Year", [Total_Labor_Dollars],
TimeCalcSelection = "Prior Year", [prior_year],
TimeCalcSelection = "Prior Year Growth", [prior_year_growth],
TimeCalcSelection = "Prior Year Growth Percent", [prior_year_growth_percent],
TimeCalcSelection = "Prior Year YTD", [prior_year_to_date],
TimeCalcSelection = "Current YTD", [year_to_date],
TimeCalcSelection = "YTD Growth", [ytd_growth],
TimeCalcSelection = "YTD Growth %", [ytd_growth_percent],
[Total_Labor_Dollars]
)
Turns out that is not true: in the Definitive Guide to DAX there is a section on variables and it says DAX uses "lazy execution" meaning that variables are only evaluated when they are called.
If you have recently started exploring Fabric, we'd love to hear how it's going. Your feedback can help with product improvements.
A new Power BI DataViz World Championship is coming this June! Don't miss out on submitting your entry.
Share feedback directly with Fabric product managers, participate in targeted research studies and influence the Fabric roadmap.
| User | Count |
|---|---|
| 57 | |
| 38 | |
| 33 | |
| 19 | |
| 16 |
| User | Count |
|---|---|
| 68 | |
| 66 | |
| 41 | |
| 34 | |
| 24 |