Including `id` field in query aborts "only" optimization

In my API I have a type which is based on Django’s model using DjangoObjectType and relay.Node interface. Using the latter automatically adds the id (graphene.ID) field in the type.

class Product(DjangoObjectType):
    name = graphene.String()

    class Meta:
        interfaces = [relay.Node]
        model = models.Product

Now, when I query the API (with optimizer enabled) and include the id field in the query, the “only optimization” is aborted as a result of this check - my automatically added id field has no name. As a result, this function is called which sets the only_list to None and disables any future “only optimizations” I’d have in my case.

@tfoxy Could you validate the issue? Maybe I broke something on my side, but maybe it is a bug in the library.

Also, I don’t fully understand why do you abort the optimization and set None to the only_list. I experimented with the code and assigned [] instead of None in that function and it solved the issue.

[yardensachs]

@maarcingebala @tfoxy I submitted PR for this suggestion #15

[tfoxy]

Maybe there can be an option to enable or disable that feature.