147

I have a model:

class ItemPrice(models.Model):
     price = models.DecimalField(max_digits=8, decimal_places=2)
     # ...

I tried this to calculate the sum of price in this queryset:

items = ItemPrice.objects.all().annotate(Sum('price'))

What's wrong in this query? or is there any other way to calculate the Sum of price column?

I know this can be done by using for loop on queryset but i need an elegant solution.

Thanks!

Super Kai - Kazuya Ito
  • 22,221
  • 10
  • 124
  • 129
Ahsan
  • 11,516
  • 12
  • 52
  • 79
  • Does this answer your question? [Django SUM Query?](https://stackoverflow.com/questions/6481279/django-sum-query) – Flimm Aug 06 '20 at 13:42

7 Answers7

306

You're probably looking for aggregate

from django.db.models import Sum

ItemPrice.objects.aggregate(Sum('price'))
# returns {'price__sum': 1000} for example
Flimm
  • 136,138
  • 45
  • 251
  • 267
MattH
  • 37,273
  • 11
  • 82
  • 84
64

Use .aggregate(Sum('column'))['column__sum'] reefer my example below

sum = Sale.objects.filter(type='Flour').aggregate(Sum('column'))['column__sum']
ugali soft
  • 2,719
  • 28
  • 25
52

Annotate adds a field to results:

>> Order.objects.annotate(total_price=Sum('price'))
<QuerySet [<Order: L-555>, <Order: L-222>]>

>> orders.first().total_price
Decimal('340.00')

Aggregate returns a dict with asked result:

>> Order.objects.aggregate(total_price=Sum('price'))
{'total_price': Decimal('1260.00')}
Ibrohim Ermatov
  • 2,169
  • 19
  • 13
11

Using cProfile profiler, I find that in my development environment, it is more efficient (faster) to sum the values of a list than to aggregate using Sum(). eg:

sum_a = sum([item.column for item in queryset]) # Definitely takes more memory.
sum_b = queryset.aggregate(Sum('column')).get('column__sum') # Takes about 20% more time.

I tested this in different contexts and it seems like using aggregate takes always longer to produce the same result. Although I suspect there might be advantages memory-wise to use it instead of summing a list.

UncleSaam
  • 346
  • 2
  • 8
  • 6
    Alternatively, use a generator expression instead of a list: `sum_a = sum(item.column for item in queryset)`. The only difference is the removed `[]`s. This saves the memory space for calculating the entire list before `sum()` iterates over it. – Code-Apprentice Oct 08 '20 at 17:28
  • getting an error : 'decimal.Decimal' object is not iterable – Roni X Dec 13 '20 at 19:26
7

Previous answers are pretty well, also, you may get that total with a line of vanilla code...

items = ItemPrice.objects.all()
total_price = sum(items.values_list('price', flat=True))
Jcc.Sanabria
  • 629
  • 1
  • 12
  • 22
4

You could also get the sum this way:

def total_sale(self):
    total = Sale.objects.aggregate(TOTAL = Sum('amount'))['TOTAL']
    return total

Replace the 'amount' with the column name from your model you want to calculate the sum of and replace 'Sale' with your model name.

thordarson
  • 5,943
  • 2
  • 17
  • 36
2

You need to use aggregate() and Sum() to calculate the sum of price column as shown below. *The query with all() is equivalent to the query without all() as shown below:

from django.db.models import Sum

print(ItemPrice.objects.all().aggregate(Sum('price')))
print(ItemPrice.objects.aggregate(Sum('price')))

Then, these dictionaries below are outputted on console:

{'price__sum': Decimal('150.00')}
{'price__sum': Decimal('150.00')}

And, you can change the default key price__sum to priceSum for price column as shown below:

from django.db.models import Sum
                                        # ↓ Here
print(ItemPrice.objects.all().aggregate(priceSum=Sum('price')))
print(ItemPrice.objects.aggregate(priceSum=Sum('price')))
                                  # ↑ Here

Then, the default key is changed as shown below:

{'priceSum': Decimal('150.00')}
{'priceSum': Decimal('150.00')}
Super Kai - Kazuya Ito
  • 22,221
  • 10
  • 124
  • 129