Django - working example of graphene aiodataloader

Question

I've been looking into using a DataLoader in Django to optimise Graphene resolvers. I have been having trouble finding a working example of this as it requires multi threading. Based on what I can tell, there is no longer a way to specify executors in the latest versions of GraphQL in python. Does anyone have a working setup as an example?

I was following the example below, and everytime I use async for my resolvers, I get:

RuntimeError: There is no current event loop in thread 'ThreadPoolExecutor-1_0'.

Which I think makes sense because GraphQL is using a sync executor by default (I stand corrected). I got this example from: https://docs.graphene-python.org/en/latest/execution/dataloader/

Source code:

dataloaders.py

from dataloader import DataLoader

class UserDataLoader(DataLoader):
    async def batch_load_fn(self, keys):
        pass

class PostDataLoader(DataLoader):
    async def batch_load_fn(self, keys):
        pass

graphql.py

import graphene
from graphene_django.types import DjangoObjectType
from .models import User, Post
from .dataloaders import UserDataLoader, PostDataLoader

class UserType(DjangoObjectType):
    class Meta:
        model = User

class PostType(DjangoObjectType):
    class Meta:
        model = Post

class Query(graphene.ObjectType):
    user = graphene.Field(UserType, id=graphene.Int())
    post = graphene.Field(PostType, id=graphene.Int())

    async def resolve_user(self, info, id):
        user_data_loader = UserDataLoader(info.context)
        return await user_data_loader.load(id)

    async def resolve_post(self, info, id):
        post_data_loader = PostDataLoader(info.context)
        return await post_data_loader.load(id)

schema = graphene.Schema(query=Query)

view.py

from graphene import BatchingExecutor # <- Does not exist in the library anymore
from graphene_django.views import GraphQLView

class NewDataLoaderGraphQLView(GraphQLView):
    executor = BatchingExecutor()

    def get_context(self, request):
        # Create new data loader instances for each request
        context = super().get_context(request)
        context.user_data_loader = UserDataLoader(context)
        context.post_data_loader = PostDataLoader(context)
        return context

urls.py

from django.urls import path
from .views import NewDataLoaderGraphQLView

urlpatterns = [
    path("graphql/", NewDataLoaderGraphQLView.as_view(graphiql=True)),
]

I'm running this using

./manage.py runserver 0.0.0.0:80

Note: This is not production code because I'm still testing this.

---

APPROACH ABANDONED

At this time, there are a lot of inconsistencies in concurrency support for Django + GraphQL. At least I didn't have any luck finding anything reliable. The closest I came was the following library which I am trialing. https://github.com/jkimbo/graphql-sync-dataloaders