marshmallow_dataclass.dataclass

marshmallow_dataclass.dataclass(_cls: Type[_U], *, repr: bool = True, eq: bool = True, order: bool = False, unsafe_hash: bool = False, frozen: bool = False, base_schema: Type[Schema] | None = None, cls_frame: FrameType | None = None) → Type[_U]

marshmallow_dataclass.dataclass(*, repr: bool = True, eq: bool = True, order: bool = False, unsafe_hash: bool = False, frozen: bool = False, base_schema: Type[Schema] | None = None, cls_frame: FrameType | None = None) → Callable[[Type[_U]], Type[_U]]

This decorator does the same as dataclasses.dataclass, but also applies add_schema(). It adds a .Schema attribute to the class object

Parameters:

• base_schema – marshmallow schema used as a base class when deriving dataclass schema

• cls_frame – frame of cls definition, used to obtain locals with other classes definitions. If None is passed the caller frame will be treated as cls_frame

>>> @dataclass
... class Artist:
...    name: str
>>> Artist.Schema
<class 'marshmallow.schema.Artist'>

>>> from typing import ClassVar
>>> from marshmallow import Schema
>>> @dataclass(order=True) # preserve field order
... class Point:
...   x:float
...   y:float
...   Schema: ClassVar[Type[Schema]] = Schema # For the type checker
...
>>> Point.Schema().load({'x':0, 'y':0}) # This line can be statically type checked
Point(x=0.0, y=0.0)