How to join tables which does not have unique value columns or have nulls in it

[Anonymous]

i have a table with raw data and i need to  calculate some aggregations by grouping some columns and i need to join this aggregated table bak to source table to get some details. but the common columns in both tables have nulls in it and does not have a column with unique values in it. I am not able to join those becoz it says, one  of the table must have unique column values and no null values in it. How do i join in this case?

[BobBI · Accepted solution]

Hi Reddy ,

You can combine 2 or more coulmns to create Composite Key .

OR 

you can copy the source table , strip out all columns except the one or tow  you think can be use as primary key ,

U se remove duplicate  ( in query editor) and  remove duplicate from the column , Now you will have only unique value in thsi New table then use this new table  in the middle and join source table and aggregated table to this this new table

source table   << New Table>>>aggregated table.

If it doesn't help , share  some data to simulate the solution.

thanks,

SS

[vishnurk · Accepted solution]

I believe the power bi july update have new feature introduced for many to many relationship

https://www.youtube.com/watch?v=ZrANsDNnZug

[PattemManohar]

@Anonymous Sample data and expected output will be helpful to provide any suggestion or solution. 

[BobBI · Accepted solution]

Hi Reddy ,

You can combine 2 or more coulmns to create Composite Key .

OR 

you can copy the source table , strip out all columns except the one or tow  you think can be use as primary key ,

U se remove duplicate  ( in query editor) and  remove duplicate from the column , Now you will have only unique value in thsi New table then use this new table  in the middle and join source table and aggregated table to this this new table

source table   << New Table>>>aggregated table.

If it doesn't help , share  some data to simulate the solution.

thanks,

SS