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.
Hi Team
Is it possible for me to remove rows from a union table created in Dax based on the value of a column? Need to remove so table has unique values so I can use the Related function. Currently many to many.
From 'UNIONTABLE-BKM-NDC' I think I want to remove all rows where the value in column 'APPR DATE CHECK' is "NOT OLDEST DATE'. I think this should leave me with only unique values....
Solved! Go to Solution.
Hi @pmadam ,
According to your statement, I think "UNIONTABLE-BKM-NDC" is an UNION table by dax. [_APPR DATE CHECK] should be a calculated column to return flag based on some conditions.
You can create a virtual table and filter it.
UNIONTABLE-BKM-NDC =
VAR _BASIC_UNION =
UNION ( 'Table1', 'Table2' )
VAR _ADD_APPR_DATE_CHECK =
ADDCOLUMNS (
_BASIC_UNION,
"_APPR DATE CHECK",
IF (
[AddedDate]
= MINX (
FILTER ( _BASIC_UNION, [OrderNumber] = EARLIER ( [OrderNumber] ) ),
[AddedDate]
),
"OLDEST DATE",
"NOT OLDEST DATE"
)
)
VAR _FILTER =
FILTER ( _ADD_APPR_DATE_CHECK, [_APPR DATE CHECK] <> "NOT OLDEST DATE" )
RETURN
_FILTER
Best Regards,
Rico Zhou
If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.
Hi @pmadam ,
According to your statement, I think "UNIONTABLE-BKM-NDC" is an UNION table by dax. [_APPR DATE CHECK] should be a calculated column to return flag based on some conditions.
You can create a virtual table and filter it.
UNIONTABLE-BKM-NDC =
VAR _BASIC_UNION =
UNION ( 'Table1', 'Table2' )
VAR _ADD_APPR_DATE_CHECK =
ADDCOLUMNS (
_BASIC_UNION,
"_APPR DATE CHECK",
IF (
[AddedDate]
= MINX (
FILTER ( _BASIC_UNION, [OrderNumber] = EARLIER ( [OrderNumber] ) ),
[AddedDate]
),
"OLDEST DATE",
"NOT OLDEST DATE"
)
)
VAR _FILTER =
FILTER ( _ADD_APPR_DATE_CHECK, [_APPR DATE CHECK] <> "NOT OLDEST DATE" )
RETURN
_FILTER
Best Regards,
Rico Zhou
If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.
@pmadam
This should work for a calculated column. What results are you getting?
User | Count |
---|---|
26 | |
10 | |
8 | |
6 | |
6 |
User | Count |
---|---|
31 | |
11 | |
10 | |
10 | |
9 |