0148401 Sep 20, 2016
@syrusakbary @mwilliamson
195 lines (132 sloc) 4.82 KB

v1.0 Upgrade Guide

Big changes from v0.10.x to 1.0. While on the surface a lot of this just looks like shuffling around API, the entire codebase has been rewritten to handle some really great use cases and improved performance.

Backwards Compatibility and Deprecation Warnings

This has been a community project from the start, we need your help making the upgrade as smooth as possible for everybody! We have done our best to provide backwards compatibility with deprecated APIs.


  • with_context is no longer needed. Resolvers now always take the context argument. Before:

    def resolve_xxx(self, args, info):
        # ...

    With 1.0:

    def resolve_xxx(self, args, context, info):
        # ...
  • ObjectType and Interface no longer accept the abstract option in the Meta. Inheriting fields should be now achieved using AbstractType inheritance.


    class MyBaseQuery(graphene.ObjectType):
        my_field = String()
        class Meta:
            abstract = True
    class Query(MyBaseQuery):

    With 1.0:

    class MyBaseQuery(graphene.AbstractType):
        my_field = String()
    class Query(MyBaseQuery, graphene.ObjectType):
  • The type_name option in the Meta in types is now name

  • Type references no longer work with strings, but with functions.


    class Query(graphene.ObjectType):
        user = graphene.Field('User')
        users = graphene.List('User')

    With 1.0:

    class Query(graphene.ObjectType):
        user = graphene.Field(lambda: User)
        users = graphene.List(lambda: User)


Schemas in graphene 1.0 are Immutable, that means that once you create a graphene.Schema any change in their attributes will not have any effect. The name argument is removed from the Schema.

The arguments executor and middlewares are also removed from the Schema definition. You can still use them, but by calling explicitly in the execute method in graphql.

# Old way
schema = graphene.Schema(name='My Schema')
schema.query = Query
schema.mutation = Mutation

# New way
schema = graphene.Schema(


For implementing an Interface in a ObjectType, you have to it onto Meta.interfaces.


from graphene import Interface, ObjectType, String

class Character(Interface):
    name = String()

class Human(Character): # Old way, Human will still be an Interface

class Droid(ObjectType): # New way, you have to specify the ObjectType
    class Meta:
        interfaces = (Character, )


Mutation fields have changed the way of usage, before if you have the mutation MyMutation you only have to reference with graphene.Field(MyMutation) now it's simply MyMutation.Field()


from graphene import ObjectType, Mutation, String

class ReverseString(Mutation):
    class Input:
        input = String(required=True)

    reversed = String()

    def mutate(self, args, context, info):
        reversed = args.get('input')[::-1]
        return ReverseString(reversed=reversed)

class Query(ObjectType):
    reverse_string = graphene.Field(ReverseString) # Old way, will not include the mutation arguments by default
    reverse_string = ReverseString.Field()


Apart of implementing as showed in the previous section, for use the node field you have to specify the node Type.


from graphene import ObjectType, relay

class Query(ObjectType):
    node = relay.NodeField() # Old way, NodeField no longer exists. Use Node.Field
    node = relay.Node.Field() # New way

Also, if wanted to create an ObjectType that implements Node, you have to do it explicity.


The Django integration with Graphene now have an independent package: graphene-django. For installing, you have to replace the old graphene[django] with graphene-django.

  • As the package is now independent, you have to import now from graphene_django.
  • DjangoNode no longer exists, please use relay.Node instead:

    from graphene.relay import Node
    from graphene_django import DjangoObjectType
    class Droid(DjangoObjectType):
        class Meta:
            interfaces = (Node, )


The SQLAlchemy integration with Graphene now have an independent package: graphene-sqlalchemy. For installing, you have to replace the old graphene[sqlalchemy] with graphene-sqlalchemy.

  • As the package is now independent, you have to import now from graphene_sqlalchemy.
  • SQLAlchemyNode no longer exists, please use relay.Node instead:

    from graphene.relay import Node
    from graphene_sqlalchemy import SQLAlchemyObjectType
    class Droid(SQLAlchemyObjectType):
        class Meta:
            interfaces = (Node, )