How can you include annotated results in a serialized QuerySet?
data = serializer.serialize(Books.objects.filter(publisher__id=id).annotate(num_books=Count('related_books')), use_natural_keys=True)
However the key/value pare {'num_books': number} is not include into the json result.
I've been searching for similar questions on the internet, but i didn't found a solution that worked for me.
Here is a similar case: http://python.6.x6.nabble.com/How-can-you-include-annotated-results-in-a-serialized-QuerySet-td67238.html
Thanks!
I did some research and found that serializer.serialize
can only serialize queryset, and annotation just adds an attribute with each object of the queryset, so when you try to serialize a query, annotated fields aren't shown. This is my way of implementation:
from django.core.serializers.json import DjangoJSONEncoder
books = Books.objects.filter(publisher__id=id).annotate(num_books=Count('related_books')).values()
json_data = json.dumps(list(books), cls=DjangoJSONEncoder)
To get count from specific columns, you must declare them via values
method
>>>> Books.objects.filter(publisher__id=id).values('<group by field>').annotate(num_books=Count('related_books'))
[{'num_books': 1, '<group by field>': X}]
Based on the link, this has been solved by pull request (https://github.com/django/django/pull/1176) some months ago.
You need to add num_books
as a property:
class Publisher():
....
@property
def num_books(self):
return some_way_to_count('related_books')
and then call it like so:
data = serializer.serialize(Books.objects.filter(publisher__id=id)), use_natural_keys=True, extra=['num_books'])
I'm not too sure about the exact syntax, since I don't work much with serializers.