Added paginate query set as qs method for limit offset pagination

[saichander17]

Description

Have a method in the pagination classes which returns the queryset instead of a list so that users can use the pagination classes even better.

Current Behaviour:

We currently have paginate_queryset method which paginates and returns the exact data as a list.

Required Behaviour:

Add a new method paginate_queryset_as_qs (name can be changed) which does everything as paginate_queryset but returns a queryset instead of a list.

This is a rough PR which I'm raising without really putting any efforts just to get inputs from everyone to understand if the proposal is even valid or not. If we think it's useful and can be added to the library, I can try to go in more detail and raise a PR which does changes for all the pagination classes.

[auvipy]

I'm not sure it is the right approach for the problem you are describing. and the tests are also failing for this change proposed!

[saichander17]

Hi @auvipy ,
Sorry for the delayed response. Was away for while.
I want to first understand if you all think it's a valid use case or not to have paginate_queryset_as_qs in the first place. If it's agreed it could be useful, then I can focus on changing the code and getting the test cases to pass.

UseCase I have currently:
I am creating a view which returns data of tableA with some aggregated counts for each entry of tableA aggregated using multiple joins.

Current code I have:

@action(detail=False, methods=["GET"], url_name="abc", url_path="abc")
def abc(self, request):
    ordering = request.query_params.get("ordering", "-created_ts")
    input_queryset = self.filter_class(
        request.query_params, queryset=self.get_queryset().order_by(f"{ordering}"), request=request
    ).qs
    qs = aggregate_counts(input_queryset=qs)
    page = self.paginate_queryset(qs)
    serializer = WaferBookResponseSerializer(page, many=True)

    response = self.get_paginated_response(serializer.data)
    return response

def aggregate_counts(qs):
    qs.annotate(
        count1=Count('pqrs', filter=Q(pqrs__is_automated=True, pqrs__mnops__tuv__isnull=False))
    )
    for el in qs:
        print(el.count1)
    return qs

Problem in the above code is that I want to provide already paginated queryset to aggregate_counts method because otherwise it is computed over all the records in qs which is not needed.