Power BI is turning 10! Tune in for a special live episode on July 24 with behind-the-scenes stories, product evolution highlights, and a sneak peek at what’s in store for the future.
Save the dateEnhance your career with this limited time 50% discount on Fabric and Power BI exams. Ends August 31st. Request your voucher.
Hi,
I know there are many questions like this, but I don't have enough DAX to customize them to solve my problem.
In short,
I have this daily summary table
and I need to transform it like this
Until now I am just able to create a table which is the cartesian product of [ID] and my calendar [Date] with CROSSJOIN.
But I don't know how to associate the existing Val 1,2 and 3 to the proper record and populate the missing values with zeroes.
Thanks for any help.
maclura
Solved! Go to Solution.
Once you have the cross join, one option to populate the columns is with a LOOKUPVALUE.
For example:
Val1 =
LOOKUPVALUE (
Summary[Val1],
Summary[ID], Transform[ID],
Summary[Date], Transform[Date],
0
)
Once you have the cross join, one option to populate the columns is with a LOOKUPVALUE.
For example:
Val1 =
LOOKUPVALUE (
Summary[Val1],
Summary[ID], Transform[ID],
Summary[Date], Transform[Date],
0
)
Dear @AlexisOlson ,
LOOKUPVALUE just works great! It's the solution to my problem.
It also allows me to avoid to SUMMARIZE duplicate entries in the source table. This is great!
Thank you.
maclura
@maclura , Do a left join in power query( Date table should be first or on left) , create a date table in power query
https://www.sqlbi.com/articles/from-sql-to-dax-joining-tables/
https://www.mssqltips.com/sqlservertip/6756/power-bi-calendar-table/
DAX Joines. or related columns
https://www.sqlbi.com/articles/from-sql-to-dax-joining-tables/
Thank you @amitchandak
Even if this is not the solution to my problem, thanks to your suggested links, I refreshed a bit my knowledge on join types. Very interesting the article on DAX join with related or not related tables!
For my situation the CROSSJOIN (cartesian product of 2 tables) is the only solution to avoid losing "date" records.
Thank you for your input, you deserve a Kudo!
maclura
User | Count |
---|---|
25 | |
11 | |
8 | |
6 | |
6 |
User | Count |
---|---|
30 | |
13 | |
11 | |
9 | |
6 |