Forum Discussion

sun-sboyanapall's avatar
sun-sboyanapall
Advocate I
4 years ago
Solved

convert SQL(case) to Dax

Hello guys,    I have this query in sql which I am trying to make a measure in powerbi but having issues,   SQL query:  SELECT COUNT (DISTINCT(ID)) -  COUNT( DISTINCT (case when MONTH(ISNULL(Co...
Need Help
  • smpa01's avatar
    smpa01
    4 years ago

    sun-sboyanapall  can you try this measure

     

    Measure =
CALCULATE (
    DISTINCTCOUNT ( tbl[ID] ),
    VAR _base =
        FILTER (
            tbl,
            tbl[TransactionDate] <> BLANK ()
                && tbl[ConvertedDate] <> BLANK ()
                && tbl[Date_of_Last_Status_Change__c] <> BLANK ()
        )
    VAR _left =
        FILTER ( tbl, tbl[ID] <> BLANK () )
    VAR _join =
        EXCEPT (
            _left,
            FILTER (
                _base,
                MONTH ( COALESCE ( tbl[ConvertedDate], tbl[Date_of_Last_Status_Change__c] ) )
                    <> MONTH ( tbl[TransactionDate] )
                    && YEAR ( COALESCE ( tbl[ConvertedDate], tbl[Date_of_Last_Status_Change__c] ) )
                        <> YEAR ( tbl[TransactionDate] )
            )
        )
    RETURN
        _join
)

     

sun-sboyanapall's avatar
sun-sboyanapall
Advocate I
4 years ago

Thank you, another way I did was created a calculated column, is consumed that is taking care of my case statement.

 

IsConsumed = 
var yearTranDate = YEAR(Tbl[TransactionDate])
var monthTranDate = MONTH(Tbl[TransactionDate])
var yearConvDate = YEAR(Tbl[ConvertedDate]))
var monthConvDate = MONTH(Tbl[ConvertedDate]))
var yeardateLastSC = YEAR(Tbl[Date_of_Last_Status_Change__c]))
var monthdateLastSC = MONTH(Tbl[Date_of_Last_Status_Change__c]))

RETURN  IF(ISBLANK(vLeadConsumedWithSales[ID]), FALSE(), IF(ISBLANK(yearConvDate), yearDateLastSC <> yearTranDate && monthDateLastSC <> monthTranDate, yearConvDate <> yearTranDate && monthConvDate <> monthTranDate))
smpa01's avatar
smpa01
Community Champion
4 years ago

sun-sboyanapall  well done.

 

One rule of thumb to always remember with DAX; if you can create a measure, don't create a column.

  • sun-sboyanapall's avatar
    sun-sboyanapall
    Advocate I
    4 years ago

    can you explain your measure too please, in its current form it is hard to understand. smpa01 

    • smpa01's avatar
      smpa01
      Community Champion
      4 years ago

      sun-sboyanapall 

       

      Measure = 
    VAR _base =
        FILTER (    //#1- returns a table that has no tbl[ID]or tbl[ConvertedDate]or tbl[Date_of_Last_Status_Change__c] as BLANK
            tbl,
            tbl[TransactionDate] <> BLANK ()
                && tbl[ConvertedDate] <> BLANK ()
                && tbl[Date_of_Last_Status_Change__c] <> BLANK ()
        )
    VAR _left =
        FILTER ( tbl, tbl[ID] <> BLANK () ) //#2- returns a table that has no tbl[ID] as BLANK
    VAR _join = 
        EXCEPT ( // #4 - this returns a left anti joined tables between 2 and 3, i.e. return everything from 2 that are not in 3
            _left,
            FILTER ( // #3- previously created base table is further filteres with SQL IS NULL conditions
                _base,
                MONTH ( COALESCE ( tbl[ConvertedDate], tbl[Date_of_Last_Status_Change__c] ) )
                    <> MONTH ( tbl[TransactionDate] )
                    && YEAR ( COALESCE ( tbl[ConvertedDate], tbl[Date_of_Last_Status_Change__c] ) )
                        <> YEAR ( tbl[TransactionDate] )
            )
        )
RETURN CALCULATE ( //5. DISTINCTCOUNT is performed on the above derived table
    DISTINCTCOUNT ( tbl[ID] ),_join)
      • sun-sboyanapall's avatar
        sun-sboyanapall
        Advocate I
        4 years ago

        That's clear, Thank you Sir