Differentiate explicit null vs. not provided fields

[weilu]

• What is the current behavior?

Currently explicitly supplying null value in a mutation query and leaving out a field & its value resulted in the same {'field': None} value in the mutate function, instead of what was described by @jkimbo in #103

• If the current behavior is a bug, please provide the steps to reproduce and if possible a minimal demo of the problem via
  a github repo, https://repl.it or similar (you can use this template as a starting point: https://repl.it/@jkimbo/Graphene-Django-Example).

Say this is my schema:

type UserType {
  id: ID!
  name: String!
  phone: String!
}

This is the UserMutation defined in schema.py:

class MediatorMutation(graphene.Mutation):
    class Arguments:
        id = graphene.ID(required=True)
        name = graphene.String()
        phone = graphene.String()

    user = graphene.Field(UserType)

    def mutate(self, info, **args):
        print(args)
        return UserMutation(user=user)

This is my mutation query:

mutation {
  updateUser(id: 1, phone: null){
    user {
      id
      name
      phone
    }
  }
}

• What is the expected behavior?

I expect the print line to be {'id': '1', 'phone': None}, but it's currently printing {'id': '1', 'name': None, 'phone': None}

• What is the motivation / use case for changing the behavior?

This is important for updating when we want to set a nullable field to null/None. See relevant GraphQL spec change: graphql/graphql-spec#83

• Please tell us about your environment:

  • Version: graphene-django==3.0.0b7
  • Platform: macos

• Other information (e.g. detailed explanation, stacktraces, related issues, suggestions how to fix, links for us to have context, eg. stackoverflow)

I just want to hear maintainer & community's thoughts on this and if it's just a user issue on my end. Happy to send a patch/PR once we confirm this is indeed a bug.

[zbyte64]

This would probably be a PR with graphene so I took a look and there is an open PR for it: graphql-python/graphene#1300

[weilu]

Makes sense. Thanks for looking into this. I’m closing this issue since there’s already one upstream