Pyright type errors when using SQLModel models as attributes in other pydantic classes or dataclasses

[li-dennis]

First Check

• I added a very descriptive title here.
• I used the GitHub search to find a similar question and didn't find it.
• I searched the SQLModel documentation, with the integrated search.
• I already searched in Google "How to X in SQLModel" and didn't find any information.
• I already read and followed all the tutorial in the docs and didn't find an answer.
• I already checked if it is not related to SQLModel but to Pydantic.
• I already checked if it is not related to SQLModel but to SQLAlchemy.

Commit to Help

• I commit to help with one of those options 👆

Example Code

import uuid

import sqlalchemy as sa
from pydantic import BaseModel
from sqlmodel import Field, SQLModel, Session, create_engine

engine = create_engine("postgres://postgres:postgres@localhost:5432/database", echo=True)  

class Foo(SQLModel, table=True):
    __tablename__ = "foo"  # type: ignore

    id: uuid.UUID = Field(
        default=None,
        sa_column=sa.Column(
            sa.UUID(as_uuid=True),
            primary_key=True,
            nullable=False,
            server_default=sa.text("gen_random_uuid()"),
        ),
    )

    data: str = Field(sa_column=sa.Column(sa.String, nullable=False))


class Bar(BaseModel):
    foo: Foo
  

def main():
  bar = Bar(foo=Foo(data="hello"))

  session = Session(engine)
  session.add(bar.foo) # bar.foo unknown type

Description

Type of "foo" is unknown Pylance[reportUnknownMemberType]

Moreover, Bar resolves to...

class Bar(
    *,
    foo: Any
)

instead of foo: Foo

However, if I change Foo back to a pydantic BaseModel, it works as expected without typing issues.

Full stack trace:

(reportUnknownVariableType)
  /path/to/example.py:36:5 - error: Type of "foo" is unknown (reportUnknownVariableType)
  /path/to/example.py:36:11 - error: Type of "foo" is unknown (reportUnknownMemberType)
  /path/to/example.py:38:17 - error: Type of "foo" is unknown (reportUnknownMemberType)
  /path/to/example.py:38:17 - error: Argument type is unknown
    Argument corresponds to parameter "instance" in function "add" (reportUnknownArgumentType)

(Of note: Field also throws a type error

  /path/to/example.py:5:22 - error: Type of "Field" is partially unknown
    Type of "Field" is "Overload[(default: Any = Undefined, *, default_factory: (() -> Any) | None = None, alias: str | None = None, title: str | None = None, description: str | None = None, exclude: AbstractSet[int | str] | Mapping[int | str, Any] | Any = None, include: AbstractSet[int | str] | Mapping[int | str, Any] | Any = None, const: bool | None = None, gt: float | None = None, ge: float | None = None, lt: float | None = None, le: float | None = None, multiple_of: float | None = None, max_digits: int | None = None, decimal_places: int | None = None, min_items: int | None = None, max_items: int | None = None, unique_items: bool | None = None, min_length: int | None = None, max_length: int | None = None, allow_mutation: bool = True, regex: str | None = None, discriminator: str | None = None, repr: bool = True, primary_key: bool | Unknown = Undefined, foreign_key: Any = Undefined, unique: bool | Unknown = Undefined, nullable: bool | Unknown = Undefined, index: bool | Unknown = Undefined, sa_type: Type[Any] | Unknown = Undefined, sa_column_args: Sequence[Any] | Unknown = Undefined, sa_column_kwargs: Mapping[str, Any] | Unknown = Undefined, schema_extra: Dict[str, Any] | None = None) -> Any, (default: Any = Undefined, *, default_factory: (() -> Any) | None = None, alias: str | None = None, title: str | None = None, description: str | None = None, exclude: AbstractSet[int | str] | Mapping[int | str, Any] | Any = None, include: AbstractSet[int | str] | Mapping[int | str, Any] | Any = None, const: bool | None = None, gt: float | None = None, ge: float | None = None, lt: float | None = None, le: float | None = None, multiple_of: float | None = None, max_digits: int | None = None, decimal_places: int | None = None, min_items: int | None = None, max_items: int | None = None, unique_items: bool | None = None, min_length: int | None = None, max_length: int | None = None, allow_mutation: bool = True, regex: str | None = None, discriminator: str | None = None, repr: bool = True, sa_column: Column[Unknown] | Unknown = Undefined, schema_extra: Dict[str, Any] | None = None) -> Any]" 
```)

### Operating System

macOS

### Operating System Details

Mac OS Sonoma 14.2.1 

### SQLModel Version

 0.0.16

### Python Version

3.12.2

### Additional Context

N/A

[sirEven]

Hi, I'm seeing "Type of "Field" is partially unknown" when importing Field from sqlmodel.
Have you found a solution?

[pradeeptammalioneflow]

Hi,
I am seeing the same error. To avoid adding type ignore every time I import "Field" from sqlmodel, I am dong this as workaround.

from sqlmodel import Field as _Field  # type: ignore[reportUnknownVariableType]


@functools.wraps(_Field)  # type: ignore[reportUnknownVariableType]
def Field(*args: Any, **kwargs: Any):  # pylint: disable=invalid-name
    return _Field(*args, **kwargs)

[m345054]

Any fix for Alembic auto-generated columns?

sa.Column("name", sqlmodel.sql.sqltypes.AutoString(length=255), nullable=False),

Argument type is partially unknown
Argument corresponds to parameter "columns" in function "create_table"
Argument type is "Column[Unknown]"

[YuriiMotov]

Works fine in current version (0.0.24)