QlikView App Dev

[apthansh]

I have data as below to calculate the Total Amount.

ID	Amount ID	Amount	Start Date	End Date
800	1	10855.35	3/1/2018 0:00	10/31/2018 23:59
800	2	11107.8	11/1/2018 0:00	10/31/2019 23:59
800	1	11410.74	11/1/2019 0:00	10/31/2020 23:59

In this FY is from July 1st - June 30th.

Below should be the calculation.

FY for 2019 is July 2018 - June 2019.Based on the start date and End date there are 4 Months in 1st row and 8 months in 2nd row that falls under the 2019 FY.so Expense would be (10,855 * 4) + (11,107*8) = 132,283.

FY for 2020 is July 2019 - June 2020.Based on the start date and End date there are 4 Months in 2nd row and 8 months in 3rd row that falls under the 2020 FY.so Expense would be (11,107*4) + (11,410*8) = 135,717.

I want to show Expense by Year in bar chart.

Thank you much

[apthansh]

@Sunny @sunny_123 @sunny1 @sunnyc129 @swuehl @jagan 

You guys are experts...any thoughts on this please ? I appreciate your help.

[Vegar]

You could use intervalmatch() to solve this.

See attached qvw

[apthansh]

Question - Why do you have Min and Max date hardcoded.

LET _mindate = '3/1/2018';
LET _maxdate = '10/31/2020';

[Vegar]

Because it was quick and easy to do for my demo. I recommend you to dynamically calculate your month values from your source data.

[apthansh]

Also - I will be showing data for- 2016 to 2020.

So - I have set min/max date as below.

LET _mindate = '7/01/2016';
LET _maxdate = '6/30/2020';

and added where clause to the source file as below.

where [expense_start] >='7/01/2016' and expense_end <='6/31/2020';

But this did not help.

[apthansh]

Hi - I have attached the sample file that I am testing.Can you please check ID - 749926 ?Some months are double counting.

[apthansh]

Can you please check the attached sample that I am working on ? For some reason Months are duplicating and Expense as well.Thank you so mcuh.

[Vegar]

I see. There is a misstake done in the join. The join should be on the calendar not the transaction. When joining an non distinct table in to transactions your tranactions will duplicate.  

You will get an synthetic key, but dont worry about it. It is ok to leave an syntethic key when using intervalmatch. If you do not want it then you can remove it after the join by creating a concatenated field of expence start and end. 

LET _mindate = '7/01/2016';
LET _maxdate = '6/30/2020';

Calendar:
LOAD
  YearName(Month,0,7) as Year,
  Month,
;
LOAD 
  MonthName('$(_mindate)',iterno()-1) as Month
autogenerate 1
While 
  MonthName('$(_mindate)',iterno()-1)  < '$(_maxdate)'
;
Trans:
LOAD 
  RowNo() as TransID,
  record_id, 
  expense_amount, 
  date(expense_start) AS expense_start, 
  date(expense_end) AS expense_end
FROM
  expense_schedule.xlsx
  (ooxml, embedded labels, table is expense_schedule)
where 
  date(expense_start,'M/DD/YYYY') >='7/01/2016' and 
  date(expense_end,'M/DD/YYYY') <='6/30/2020';

/*IntervalMatch your Start and End against the Month-value (All three are duals with a numeric value)*/
Left join (Calendar)
IntervalMatch(Month)
LOAD 
  expense_start, 
  expense_end
Resident
  Trans
;

[alex00321]

You are awesome!! I updated the code a little bit to remove the synthetic keys, but I could still not figure out how to summarize the number, seems it's not a total, it's showing each month's value: