Hello,
I wish to filter data on date, to show data within a date range (which currently works) OR with no date (ISBLANK() returns True). As I'm using multiple data sources, I created a 'Dates' table as follows :
Dates =
DISTINCT(
UNION(
SUMMARIZE('Table 1', 'Table 1'[Date comptable]),
SUMMARIZE('Table 2', 'Table 1'[Date comptable]),
SUMMARIZE('Table 3', 'Table 1'[Date comptable]),
...
)
)
with the following relations :
Dates[Date comptable] (*) -> (*) 'Table 1'[Date comptable]
Dates[Date comptable] (*) -> (*) 'Table 2'[Date comptable]
Dates[Date comptable] (*) -> (*) 'Table 3'[Date comptable]
...
As excepted, the Dates[Date comptable] column contains one blank row, and all the dates from the dataset.
I'm using the following filter on the report :
And it shows no data /
However, when I use another table, for example 'Table 1', PBI sucessfuly filters the empty dates and counts the elements :
I tried to modify the relations from -> to <->, and it doesn't work. I also wanted to create one-to-many relations, as I use 'Distinct' on my Dates table, but the editor won't let me chose this option (it says the cardinality isn't valid for this relation).
Am I doing something wrong ?
Thanks in advance for helping me.
Solved! Go to Solution.
Do not derive your Dates table from the values in your fact tables. You can use the min and max dates for the boundaries, but the Dates table itself must have a contiguous (ie no gaps) Date column.
Do not derive your Dates table from the values in your fact tables. You can use the min and max dates for the boundaries, but the Dates table itself must have a contiguous (ie no gaps) Date column.
User | Count |
---|---|
106 | |
88 | |
69 | |
52 | |
49 |
User | Count |
---|---|
148 | |
94 | |
79 | |
71 | |
70 |