How to use group by in the load script?

[Anonymous]

Table 1 in load script

Period	Customer	Rep
201301	C1	Rep1
201301	C2	Rep1
201301	C2	Rep2
201301	C3	Rep2
201302	C1	Rep1
201302	C2	Rep2
201302	C3	Rep2

Table 2 in load script

Period	Customer	Feedback
201301	C1	good
201301	C2	better
201301	C3	best
201302	C1	better
201302	C2	better
201302	C3	best

I want to get table 3 in load script as following, but how to write the script?

Table 3 in load script

Period	Customer	Feedback	Overlap_flag
201301	C1	good	0
201301	C2	better	1
201301	C3	best	0
201302	C1	better	0
201302	C2	better	0
201302	C3	best	0

[kumarnatarajan]

Hi,

Find the attached file

[#3]

try this

in script--

table1:

load if(Previous(Customer)<>Customer,0,1) as Overlap_flag,* Inline [

Period,    Customer,    Rep

201301,    C1,    Rep1

201301,    C2,    Rep1

201301,    C2,    Rep2

201301,    C3,    Rep2

201302,    C1,    Rep1

201302,    C2,    Rep2

201302,    C3,    Rep2

];

join

table2:

load * Inline [

Period,    Customer,    Feedback

201301,    C1,    good

201301,    C2,    better

201301,    C3,    best

201302,    C1,    better

201302,    C2,    better

201302,    C3,    best

];

table3:

load

Period,   

Customer,

only(Feedback) as Feedback

Resident table1

Group by Period,Customer;

then take pivote chart--

Dimension1--- Period

dimension2--- Customer

expression1--  Feedback

expression2--   Aggr(max(Overlap_flag),Customer,Period)        //Overlap_flag    or simly use  max(Overlap_flag)

then output like this

Period	Customer	Feedback	Overlap_flag
201301	C1	good	0
201301	C2	better	1
201301	C3	best	0
201302	C1	better	0
201302	C2	better	0
201302	C3	best	0

[#4]

hi

if your problem resolved then marked it as either correct or helpful according to your question so that it might be helpful for other developer

Regards

vishwaranjan