Support Field in dataclass + 'metadata' kwarg of dataclasses.field (pydantic#2384)

* Support `Field` in `dataclass` + `'metadata'` kwarg of `dataclasses.field`

Please enter the commit message for your changes. Lines starting

* add `__has_field_info_default__` for minimal effect on perf

* lower complexity of `_process_class`

Author: PrettyWood

changes/2384-PrettyWood.md

@@ -0,0 +1 @@
+Improve field declaration for _pydantic_ `dataclass` by allowing the usage of _pydantic_ `Field` or `'metadata'` kwarg of `dataclasses.field`

docs/examples/dataclasses_default_schema.py

@@ -1,5 +1,7 @@
 import dataclasses
-from typing import List
+from typing import List, Optional
+
+from pydantic import Field
 from pydantic.dataclasses import dataclass
 
 
@@ -8,6 +10,11 @@ class User:
     id: int
     name: str = 'John Doe'
     friends: List[int] = dataclasses.field(default_factory=lambda: [0])
+    age: Optional[int] = dataclasses.field(
+        default=None,
+        metadata=dict(title='The age of the user', description='do not lie!')
+    )
+    height: Optional[int] = Field(None, title='The height in cm', ge=50, le=300)
 
 
 user = User(id='42')

pydantic/dataclasses.py

@@ -3,14 +3,14 @@
 from .class_validators import gather_all_validators
 from .error_wrappers import ValidationError
 from .errors import DataclassTypeError
-from .fields import Required
+from .fields import Field, FieldInfo, Required, Undefined
 from .main import create_model, validate_model
 from .typing import resolve_annotations
 from .utils import ClassAttribute
 
 if TYPE_CHECKING:
     from .main import BaseConfig, BaseModel  # noqa: F401
-    from .typing import CallableGenerator
+    from .typing import CallableGenerator, NoArgAnyCallable
 
     DataclassT = TypeVar('DataclassT', bound='Dataclass')
 
@@ -19,6 +19,7 @@ class Dataclass:
         __initialised__: bool
         __post_init_original__: Optional[Callable[..., None]]
         __processed__: Optional[ClassAttribute]
+        __has_field_info_default__: bool  # whether or not a `pydantic.Field` is used as default value
 
         def __init__(self, *args: Any, **kwargs: Any) -> None:
             pass
@@ -80,6 +81,30 @@ def is_builtin_dataclass(_cls: Type[Any]) -> bool:
     return not hasattr(_cls, '__processed__') and dataclasses.is_dataclass(_cls)
 
 
+def _generate_pydantic_post_init(
+    post_init_original: Optional[Callable[..., None]], post_init_post_parse: Optional[Callable[..., None]]
+) -> Callable[..., None]:
+    def _pydantic_post_init(self: 'Dataclass', *initvars: Any) -> None:
+        if post_init_original is not None:
+            post_init_original(self, *initvars)
+
+        if getattr(self, '__has_field_info_default__', False):
+            # We need to remove `FieldInfo` values since they are not valid as input
+            # It's ok to do that because they are obviously the default values!
+            input_data = {k: v for k, v in self.__dict__.items() if not isinstance(v, FieldInfo)}
+        else:
+            input_data = self.__dict__
+        d, _, validation_error = validate_model(self.__pydantic_model__, input_data, cls=self.__class__)
+        if validation_error:
+            raise validation_error
+        object.__setattr__(self, '__dict__', d)
+        object.__setattr__(self, '__initialised__', True)
+        if post_init_post_parse is not None:
+            post_init_post_parse(self, *initvars)
+
+    return _pydantic_post_init
+
+
 def _process_class(
     _cls: Type[Any],
     init: bool,
@@ -100,16 +125,7 @@ def _process_class(
 
     post_init_post_parse = getattr(_cls, '__post_init_post_parse__', None)
 
-    def _pydantic_post_init(self: 'Dataclass', *initvars: Any) -> None:
-        if post_init_original is not None:
-            post_init_original(self, *initvars)
-        d, _, validation_error = validate_model(self.__pydantic_model__, self.__dict__, cls=self.__class__)
-        if validation_error:
-            raise validation_error
-        object.__setattr__(self, '__dict__', d)
-        object.__setattr__(self, '__initialised__', True)
-        if post_init_post_parse is not None:
-            post_init_post_parse(self, *initvars)
+    _pydantic_post_init = _generate_pydantic_post_init(post_init_original, post_init_post_parse)
 
     # If the class is already a dataclass, __post_init__ will not be called automatically
     # so no validation will be added.
@@ -144,22 +160,31 @@ def _pydantic_post_init(self: 'Dataclass', *initvars: Any) -> None:
     )
     cls.__processed__ = ClassAttribute('__processed__', True)
 
-    fields: Dict[str, Any] = {}
+    field_definitions: Dict[str, Any] = {}
     for field in dataclasses.fields(cls):
+        default: Any = Undefined
+        default_factory: Optional['NoArgAnyCallable'] = None
+        field_info: FieldInfo
 
-        if field.default != dataclasses.MISSING:
-            field_value = field.default
+        if field.default is not dataclasses.MISSING:
+            default = field.default
         # mypy issue 7020 and 708
-        elif field.default_factory != dataclasses.MISSING:  # type: ignore
-            field_value = field.default_factory()  # type: ignore
+        elif field.default_factory is not dataclasses.MISSING:  # type: ignore
+            default_factory = field.default_factory  # type: ignore
+        else:
+            default = Required
+
+        if isinstance(default, FieldInfo):
+            field_info = default
+            cls.__has_field_info_default__ = True
         else:
-            field_value = Required
+            field_info = Field(default=default, default_factory=default_factory, **field.metadata)
 
-        fields[field.name] = (field.type, field_value)
+        field_definitions[field.name] = (field.type, field_info)
 
     validators = gather_all_validators(cls)
     cls.__pydantic_model__ = create_model(
-        cls.__name__, __config__=config, __module__=_cls.__module__, __validators__=validators, **fields
+        cls.__name__, __config__=config, __module__=_cls.__module__, __validators__=validators, **field_definitions
     )
 
     cls.__initialised__ = False

tests/test_dataclasses.py

@@ -429,7 +429,7 @@ class User:
     assert fields['id'].default is None
 
     assert fields['aliases'].required is False
-    assert fields['aliases'].default == {'John': 'Joey'}
+    assert fields['aliases'].default_factory() == {'John': 'Joey'}
 
 
 def test_default_factory_singleton_field():
@@ -456,6 +456,10 @@ class User:
         name: str = 'John Doe'
         aliases: Dict[str, str] = dataclasses.field(default_factory=lambda: {'John': 'Joey'})
         signup_ts: datetime = None
+        age: Optional[int] = dataclasses.field(
+            default=None, metadata=dict(title='The age of the user', description='do not lie!')
+        )
+        height: Optional[int] = pydantic.Field(None, title='The height in cm', ge=50, le=300)
 
     user = User(id=123)
     assert user.__pydantic_model__.schema() == {
@@ -466,11 +470,21 @@ class User:
             'name': {'title': 'Name', 'default': 'John Doe', 'type': 'string'},
             'aliases': {
                 'title': 'Aliases',
-                'default': {'John': 'Joey'},
                 'type': 'object',
                 'additionalProperties': {'type': 'string'},
             },
             'signup_ts': {'title': 'Signup Ts', 'type': 'string', 'format': 'date-time'},
+            'age': {
+                'title': 'The age of the user',
+                'description': 'do not lie!',
+                'type': 'integer',
+            },
+            'height': {
+                'title': 'The height in cm',
+                'minimum': 50,
+                'maximum': 300,
+                'type': 'integer',
+            },
         },
         'required': ['id'],
     }