Reg:Handling distinct Values

[#1]

Hi All,

     I am having following data structure

   TYPE  MAKE  MODEL  KW

   A1A3  MAK1   MDU1    1

   A1A3  MAK1   MDU1    2

   A1A3  MAK1   MDU1    1

   B1B2  MAK2   MDU2    2

   B1B2  MAK2   MDU2    1

   B1B2  MAK2   MDU2    3

Now I want to show the sum of KW values irrespetive of Make/Model..Means I need to show distinct values of KW irrespetive of MAKE and Model

Actual O/P : SUM(KW)=10

Expected O/P:             =6

Note:

   Though it is having diffrent MAKES & MODELS need to get only distict KW values,in the straight table.

-Jai

[#2]

You mean sum should be based on Type like 2+1+3 =6 for B1B2. If not i didnt get ur question

[#3]

Hi vikram,

               It should sum based on KW values not by make/model .So that as a distinct value 1 and 2 will come from

A1A3 and 3 from B1B2. It should neglect remaining occurance for the same.

Thanks,

Jai

[#4]

Try the following expression:

    Sum({<MAKE=, MODEL=>} Distinct KW)

THANKS,

Narasimha

[SunilChauhan]

use below code

sum( distinct KW)

hope thiss helps

[jagannalla]

If you want to do it in script. Use this code.

T3:

LOAD * INLINE [

       TYPE,  MAKE,  MODEL,  KW

       A1A3,  MAK1,   MDU1,    1

       A1A3,  MAK1,   MDU1,    2

       A1A3,  MAK1,   MDU1,    1

       B1B2,  MAK2,   MDU2,    2

       B1B2,  MAK2,   MDU2,    1

       B1B2,  MAK2,   MDU2,    3

];

R_T3:

LOAD KW, KW as Kvalue,Sum(KW) as Kvalue1

Resident T3

Group By KW;

-In ST take dim as TYPE and Exp as Sum(kValue)

[#7]

Hi,

You can use TOTAL inside of sum function:

sum(TOTAL KW)

Thanks,

Niranjan M.