Python Marshmallow Field 可以是两种不同的类型

Question

我想指定一个棉花糖模式。对于我的一个字段，我希望对其进行验证，但它可以是字符串或字符串列表。我尝试过原始字段类型，但这允许一切通过。有没有办法只验证我想要的两种类型？

就像是，

value = fields.Str() or fields.List()

Answer 1

我今天遇到了同样的问题，我想出了这个解决方案：

class ValueField(fields.Field):
    def _deserialize(self, value, attr, data, **kwargs):
        if isinstance(value, str) or isinstance(value, list):
            return value
        else:
            raise ValidationError('Field should be str or list')


class Foo(Schema):
    value = ValueField()
    other_field = fields.Integer()

您可以创建自定义字段并重载该_deserialize方法，以便它验证代码是否isinstance符合所需类型。我希望它对你有用。

foo.load({'value': 'asdf', 'other_field': 1})
>>> {'other_field': 1, 'value': 'asdf'}
foo.load({'value': ['asdf'], 'other_field': 1})
>>> {'other_field': 1, 'value': ['asdf']}
foo.load({'value': 1, 'other_field': 1})
Traceback (most recent call last):
  File "<input>", line 1, in <module>
  File "/Users/webinterpret/Envs/gl-gs-onboarding-api/lib/python3.7/site-packages/marshmallow/schema.py", line 723, in load
    data, many=many, partial=partial, unknown=unknown, postprocess=True
  File "/Users/webinterpret/Envs/gl-gs-onboarding-api/lib/python3.7/site-packages/marshmallow/schema.py", line 904, in _do_load
    raise exc
marshmallow.exceptions.ValidationError: {'value': ['Field should be str or list']}

Answer 2

映射的解决方案，类似于上面的：

from typing import List, Mapping, Any
from marshmallow import Schema, fields
from marshmallow.exceptions import ValidationError


class UnionField(fields.Field):
    """Field that deserializes multi-type input data to app-level objects."""

    def __init__(self, val_types: List[fields.Field]):
        self.valid_types = val_types
        super().__init__()

    def _deserialize(
        self, value: Any, attr: str = None, data: Mapping[str, Any] = None, **kwargs
    ):
        """
        _deserialize defines a custom Marshmallow Schema Field that takes in mutli-type input data to
        app-level objects.
        
        Parameters
        ----------
        value : {Any}
            The value to be deserialized.
        
        Keyword Parameters
        ----------
        attr : {str} [Optional]
            The attribute/key in data to be deserialized. (default: {None})
        data : {Optional[Mapping[str, Any]]}
            The raw input data passed to the Schema.load. (default: {None})
        
        Raises
        ----------
        ValidationError : Exception
            Raised when the validation fails on a field or schema.
        """
        errors = []
        # iterate through the types being passed into UnionField via val_types
        for field in self.valid_types:
            try:
                # inherit deserialize method from Fields class
                return field.deserialize(value, attr, data, **kwargs)
            # if error, add error message to error list
            except ValidationError as error:
                errors.append(error.messages)
                raise ValidationError(errors)

使用：

class SampleSchema(Schema):
    ex_attr = fields.Dict(keys=fields.Str(), values=UnionField([fields.Str(), fields.Number()]))

图片来源：安娜·K