Flag duplicates with DAX

[uif19085]

Hello, i'm trying to find a way to flag first apprence of a duplicates across muliple columns using DAX, my table looks like this:
ID           DATE             FLAG
1       01/01/2011           1
1       01/01/2011           0
2      02/02/2011            1
2      02/02/2011            0
2      02/02/2011            0

I tried with this calculated column but didn't work: 

Column =

IF(CALCULATE(COUNTROWS('Weeks'),ALLEXCEPT('Weeks','Weeks'[id],'Weeks'[date]))>1,1,0)

[Ashish_Mathur]

Hi,

It may be easier to do this in the Query Editor.  Would you be interested in that solution?

[v-yueyunzh-msft]

Hi , @uif19085 

According to your description, you want to "Flag duplicates with DAX".

First, to distinguish this difference, we need a column to distinguish the order of the data, and we need to add an index column to the Power Query.

Then we can add a calculated column use this dax code:

Column = var _t =FILTER('Table','Table'[ID]=EARLIER('Table'[ID]) && 'Table'[DATE]=EARLIER('Table'[DATE]) && 'Table'[Index]<= EARLIER('Table'[Index]))
return
IF(COUNTROWS(_t)=1,1,0)

Then we can meet your need:

Thank you for your time and sharing, and thank you for your support and understanding of PowerBI! 

Best Regards,

Aniya Zhang

If this post helps, then please consider Accept it as the solution to help the other members find it more quickly

[smpa01 · Accepted solution]

@uif19085  can you please test out the following meaure , you probably need a rowNum/Index Column

dup =
VAR __val =
    CALCULATE (
        [current],
        OFFSET (
            -1,
            DISTINCT ( ALL ( 'Table' ) ),
            ORDERBY ( 'Table'[Index] ),
            KEEP,
            PARTITIONBY ( 'Table'[id] )
        )
    )
VAR currentId =
    MAX ( 'Table'[id] )
VAR currentIndex =
    MAX ( 'Table'[Index] )
VAR dateCount =
    CALCULATE (
        DISTINCTCOUNT ( 'Table'[date] ),
        FILTER (
            ALL ( 'Table' ),
            'Table'[id] = currentId
                && 'Table'[Index] <= currentIndex
        )
    )
VAR final =
    SWITCH (
        TRUE (),
        __val == BLANK (), 1,
        [current] - 1 = __val
            && dateCount = 1, 0,
        1
    )
RETURN
    final

or like this

Measure = 
VAR currDate =
    MAX ( 'Table'[date] )
VAR currDate2 = currDate + 0
VAR currentId =
    MAX ( 'Table'[id] )
VAR currentIndex =
    MAX ( 'Table'[Index] )
VAR minIndexbyID =
    CALCULATE (
        [currentIndex],
        INDEX (
            1,
            DISTINCT ( ALL ( 'Table' ) ),
            ORDERBY ( 'Table'[Index] ),
            KEEP,
            PARTITIONBY ( 'Table'[id] )
        )
    )
VAR filterTbl1 =
    FILTER (
        ALL ( 'Table' ),
        'Table'[id] = currentId
            && 'Table'[Index] >= minIndexbyID
            && currentIndex >= 'Table'[Index]
    )
VAR filterTbl2 =
    FILTER (
        ALL ( 'Table' ),
        'Table'[id] = currentId
            && 'Table'[Index] >= minIndexbyID
    )
VAR leftTbl1 =
    SELECTCOLUMNS ( filterTbl1, "dt", 'Table'[date] + 0 )
VAR leftTbl2 =
    SELECTCOLUMNS ( filterTbl2, "dt", 'Table'[date] + 0 )
VAR rightTbl =
    SELECTCOLUMNS ( { currDate }, "dt", [Value] + 0 )
VAR dates1 =
    NATURALINNERJOIN ( leftTbl1, rightTbl )
VAR dates2 =
    NATURALINNERJOIN ( leftTbl2, rightTbl )
VAR val1 =
    COUNTROWS ( dates1 )
VAR val2 =
    COUNTROWS ( dates2 )
VAR ternary =
    SWITCH (
        TRUE (),
        val1 = 1
            && val2 = 1, val1 & "-unique entry",
        val1 = 1
            && val2 <> 1, val1 & "-has duplicate entry",
        "0" & "-has duplicate entry"
    )
RETURN
    ternary

[FreemanZ]

hi @uif19085 

do you have other columns?

without the flag column, row1 and row2 are identical.