Use pydantic model as a scalar with `strawberry.auto`

[benzolium]

Add way of using pydantic model scalars without manually defining fields - just by using strawberry.auto

Feature Request Type

• [x] Alteration (enhancement/optimization) of existing feature(s)

Description

I have a pydantic model, lets say:

class SomeNameModel(pydantic.BaseModel):
    name: str
    number: int

And I have a scalar:

@strawberry.scalar(
    name='SomeName',
    serialize=marshal_some_name,
    parse_value=unmarshal_some_name,
)
class SomeName(SomeNameModel):
    """SomeName scalar."""

I have a model which has uses that scalar:

class User(pydantic.BaseModel):
    some_name: SomeNameModel

When I'm trying to create a type from it:

@strawberry.experimental.pydantic.type(
    model=User,
)
class UserType:
    some_name: strawberry.auto

I'm getting an error:

Cannot find a Strawberry Type for <class 'SomeNameModel'> did you forget to register it?

If I change User to:

class User(pydantic.BaseModel):
    some_name: SomeName

I have a pydantic error:

RuntimeError: error checking inheritance of <strawberry.custom_scalar.ScalarWrapper object at 0x10adc2ce0> (type: ScalarWrapper)

However, this works (manual annotation of type):

@strawberry.experimental.pydantic.type(
    model=User,
)
class UserType:
    some_name: SomeName

Upvote & Fund

• We're using Polar.sh so you can upvote and help fund this issue.
• We receive the funding once the issue is completed & confirmed by you.
• Thank you in advance for helping prioritize & fund our backlog.

[filwaline]

I am also encountering the same issue because strawberry.auto is not working, and setting the flag all_field=True in strawberry.experimental.pydantic.type has also failed.

Even when scalar_overrides are set, strawberry.auto still fails to resolve the type.

def create_schema() -> strawberry.Schema:
    return strawberry.Schema(
        query=Query,
        mutation=Mutation,
        scalar_overrides={SomeNameModel: SomeName},
    )