问题
I am trying to use Power BI with my SQL DB. We have a three layer structure
- A Main table, with information about the year and unique key for each firm
- A bridge table which an information about the type of the firm
- Many end Tables with information about sales and stuff
On some occasions these end tables are jointly used by different firm types and some only by one firm type.
When I load the tables into Power BI all the relationships are correctly imported and I was expecting no issue to create a summary table like Sum(Var) by Year, even though I am getting the Var from an end table and the Year from the main table. However they show the same sum for all years.
I am having a hard time creating a MWE and I do not have access to a remote server.
This picture shows the structure, which is similar to our data base, I did not add the type column. I now created bi directional realtion with the end Tables
Basically I just created a MWE with this data split into 4 tables. I simplified the End tables to just be 1 and 0 and added a third firm
When taking the sums of X and Y, I am getting the same value and I would have expected 2 in 2018 and 1 in 2019 for X and 1 in 2018 and 2 in 2019 for Y
I also tried to merge the Year into the bridge table when loading the data, but I still get the wrong results.
I find many post about wrong totals etc. and needing to use different measures, but I do not have any complicated calculations just the sum of the raw data per year.
Anything else I can do to clarify my question? Should i create a new one? I could add the raw data as a csv/xlsx file(s)
thx for the answer which I can accept later given that it solves my MWE, but somehow it does not solve my original problem. As explained in the comment here is my data model:
I am using the Import function from an MS SQL Data Base and your DAX Code
dax1 = sumx(unternehmen; RELATED(at_sonstiges[abschluss_erwartet]))
回答1:
I'm assuming this is what you're after:
I've defined these measures as
SumX = SUMX ( Main_Table, RELATED ( End_Table_X[X] ) )
SumY = SUMX ( Main_Table, RELATED ( End_Table_Y[Y] ) )
This solution does not require a year column on the Bridge_Table or bidirectional filters with the end tables.
Instead of trying to propagate the year, it iterates over all of the rows of the Main_Table (within the local filter context) and essentially does a lookup of X or Y for each row.
来源：https://stackoverflow.com/questions/65197622/power-bi-report-with-bridge-table