The ultimate Fabric, Power BI, SQL, and AI community-led learning event. Save €200 with code FABCOMM.
Get registeredCompete to become Power BI Data Viz World Champion! First round ends August 18th. Get started.
So if have data that looks like this, A can happen in week 1, 2 or 3. And did a sum of A,B,C,D in power query to add the data to a donut chart. I REALLY dont want to unpivot this data, since i know its not ideal for working on BI.
The tooltip here is fake just for show.
What i need is filtering data on a table like the image shows (i can recognize this on a measure by "IF ( ISFILTERED( ) )", and have a tooltip that tells me, i had 2 "A" results, but what week each of them corresponds. The measure should have something like "IF value selected is A, and week1 / A = 1, then "week1: 1" else "week1: 0"...." and so on.
At the moment im failing to grab the "hover" value on the donut chart, for column charts it was simple, just using "ISFILTERED" did the job, here i cannot reference what value is "hovered" to create a measure. Any ideas?
HI @DanS446,
Perhaps you can create a table with all of the table field names and use its feild on legend field.
Then you can write a measure formula to lookup raw table field value and summary corresponding value field values based on current category values. (they will also work for the tooltip hover effects)
Regards,
Xiaoxin Sheng
Greetings. Creating a separate legends table and applying it to the graph just repeats the totals in every row. Using SELECTEDVALUE works to see if you are hovering over a section of a donut chart when you have legends, but since i have a bunch of values from different columns thrown together, im trying to find if the values selected are from that column.
HI @DanS446,
If you want to accurately map these records, I'd like to suggest duplicate the fact table and apply unpivot column on this new table. Then you can use this new table as source to write formulas without cross multiple fields.
Regards,
Xiaoxin Sheng