PowerBI中DAX度量值过滤、行计数及列返回问题咨询

Hey there! Since you're transitioning from Python to DAX in Power BI, I totally get figuring out how to translate filtering logic over—let's break down your two questions clearly.

1. Modify Measure.Controllable to Count Filtered Rows #

Your original measure already uses CALCULATE with filters and a relationship; we just need to add two more conditions to exclude rows where column_two is blank or equals "Acceptance_mwap".

The most efficient way (since DAX optimizes native CALCULATE filters better than nested FILTER) is to add the new conditions directly as additional filter arguments:

Measure.Controllable = 
CALCULATE(
    COUNTROWS(table_adj_spec_uno),
    table_adj_spec_uno[column_uno] = "variable 1",
    // Exclude blank values and "Acceptance_mwap"
    NOT ISBLANK(table_adj_spec_uno[column_two]),
    table_adj_spec_uno[column_two] <> "Acceptance_mwap",
    USERELATIONSHIP('table_adj_spec_uno'[IncidentDate], 'Table.Date'[DateOnly])
)

If you prefer a more explicit filter structure (great for readability when logic gets complex), you can wrap the table in a FILTER function instead:

Measure.Controllable = 
CALCULATE(
    COUNTROWS(table_adj_spec_uno),
    FILTER(
        table_adj_spec_uno,
        table_adj_spec_uno[column_uno] = "variable 1" &&
        NOT ISBLANK(table_adj_spec_uno[column_two]) &&
        table_adj_spec_uno[column_two] <> "Acceptance_mwap"
    ),
    USERELATIONSHIP('table_adj_spec_uno'[IncidentDate], 'Table.Date'[DateOnly])
)

Both versions work, but the first one will perform better on large datasets.

2. Create a Measure to Return Filtered column_two Values #

DAX measures return aggregated results, so we need to adjust based on what you want to display:

If you expect a single value per context (e.g., filtered to one row) #

Use SELECTEDVALUE to get the single matching value (returns blank if there are multiple values, or you can set a default message):

Measure.FilteredColumnTwo = 
CALCULATE(
    SELECTEDVALUE(table_adj_spec_uno[column_two], "Multiple values found"), // Optional default text
    table_adj_spec_uno[column_uno] = "variable 1",
    NOT ISBLANK(table_adj_spec_uno[column_two]),
    table_adj_spec_uno[column_two] <> "Acceptance_mwap",
    USERELATIONSHIP('table_adj_spec_uno'[IncidentDate], 'Table.Date'[DateOnly])
)

If you want to list all matching values (concatenated into a string) #

Use CONCATENATEX to iterate over filtered rows and join values with a separator (like commas or new lines):

Measure.FilteredColumnTwo_List = 
CALCULATE(
    CONCATENATEX(
        table_adj_spec_uno,
        table_adj_spec_uno[column_two],
        ", " // Adjust separator as needed (e.g., "; ", CHAR(10) for new lines)
    ),
    table_adj_spec_uno[column_uno] = "variable 1",
    NOT ISBLANK(table_adj_spec_uno[column_two]),
    table_adj_spec_uno[column_two] <> "Acceptance_mwap",
    USERELATIONSHIP('table_adj_spec_uno'[IncidentDate], 'Table.Date'[DateOnly])
)

This works great in card visuals to show all qualifying column_two values at once.

内容的提问来源于stack exchange，提问作者LeCoda