# Define models

class Author(Model):
    name = CharField()
    rating = IntegerField()

class Book(Model):
    name = CharField()
    year = IntegerField()
    author = ForeignKeyField(Author)

# Create nodes

class BookNode(PeeweeObjectType):
    class Meta:
        model = Book
        manager = db_manager

class AuthorNode(PeeweeObjectType):
    class Meta:
        model = Author
        manager = db_manager

# Create connections

class BookConnection(PeeweeConnection):
    class Meta:
        node = BookNode

# Aggregate queries

class Query(ObjectType):
    books = PeeweeConnectionField(BookConnection)

# Create schema

schema = Schema(query=Query, auto_camelcase=False)

# Execute graphql query

result = schema.execute('''
    query {
        books (filters: {author__name__ilike: "%Lovecraft%"}) {
            total
            edges {
                node {
                    id
                    name
                    author {
                        id
                        name
                    }
                }
            }
        }
    }''',
    return_promise=True,
    executor=AsyncioExecutor()
)

# Await result if required (failed queries are usually returning result
#                           synchronously with non-empty `result.errors`
#                           while successful ones requires awaiting
#                           of peewee/DB level queries of course)

if not isinstance(result, ExecutionResult):
    result = await result

# Enjoy the result :)

print(result.data)
#
# ===>
#
# {'books': {
#     'total': 2,
#     'edges': [
#         {'node': {
#             'id': 5,
#             'name': 'Dagon',
#             'author': {
#                 'id': 1,
#                 'name': 'Howard Lovecraft'
#             }
#         }},
#         {'node': {
#             'id': 6,
#             'name': 'At the Mountains of Madness',
#             'author': {
#                 'id': 1,
#                 'name': 'H.P. Lovecraft'
#             }
#         }}
#     ]
# }}