My problem is that i have different sums when my graph use a dynamic dimension.
I try to use Firstsortedvalue() but i don't succeed (but perhaps i don't how to use it correctly).
My expectation is to have in the result of the set analyse, one and only one row per ticket, whatever the dimension(s) used.
so in my example, if i want to display the ticket in input for the month of september, i want to count 1 for my ticket 1025639 and i want to use only the row 10365 which is the first time the ticket came into the grp Grp2.
And if i display my input with a second dimension like the status, i want to count 1 for my ticket 1025639 and status Open (and i don't want to count 1 for the status Resolved).
You will find enclosed my app.
the same expression is used in all the graphs (i want to count tickets in input and output for the month of September).
The result is not the same if i add a dimension, the sum is higher because i count severals time a ticket (1 for each value of the dimension).
What i want to do, is to select in the Set Analysis only one row per ticket (so when i will add new dimension, the sum will be still the same).
Pb-Set Analysis.qvw 9.8 MB
Choose the group 'FR IT SIAM N2 BACK OFFICE', then you will see that in the global graph, i have :
1981 Tickets in Input
1666 Tickets in output
==> thoses numbers are correct.
But when you look at the graph 'Details' and choose the dimension 'Phase' per exemple, you will find :
2123 Tickets in Input
1900 tickets in output.
==> those numbers are not correct because it count several times the same tickets but with a different phase.
What i want to do is to select directly in the Set Analysis the last rows for each ticket (regarding the field 'date'), if i can do that my detail graph will be correct.
Thanks for your time and your help.
Note : yes input and output formula are different, but input formula is the same between global graph, detail graph, and detail table.