I have a table with the following values:

The input of the function is a range of date, for example, as follows:

date_start = '2019-01-01'
date_end = '2021-11-07'

Expected output:

For all users, for each date in this range of date, I want to return the sum of the amounts, from the current date to all previous dates like this:

My efforts

def get_sum_amount(self, date_start=None, date_end=None):
    date_detail = {}
    # Minimum date
    if date_start:
        date_detail['date__gt'] = date_start

    # Maximum date
    if date_end:
        date_detail['date__lt'] = date_end

    detail = Financial.objects.filter(Q(**date_detail)) \
        .values('user') \
        .annotate(sum_amount=Coalesce(Sum(F('amount')), Value(0)))

but no result.

UPDATED

The output of the function for the above example is:

CodePudding user response：

You can use Window functions with a partition onuser to run a cumulative sum per user, ordered by the date:

detail = Financial.objects.filter(**date_detail).annotate(
    running_amount=Window(
        expression=Sum('amount'),
        partition_by=[F('user')],
        order_by=F('date').asc(),
    )
)