madtypes 0.0.6

Python typing that raise TypeError at runtime

madtypes

• 💢 Python typing that raise TypeError at runtime
• 📖 Render to dict or json
• 🌐 Generate Json-Schema
• 💪 Type hints cheat sheet

from madtypes import Schema

class Item(Schema)
    name: str

Item() # raise TypeError, name is missing
Item(name=2) # raise TypeError, 2 is not an str
Item(name="foo") # ok

repr(Item(name="foo")) == {"name": "foo"}
json.dumps(Item(name="foo")) => '{"name": "foo"}'

from typing import Optional
class ItemWithOptional(Schema):
    name: Optional[str]

ItemWithOptional() # ok

• json-schema

from madtypes import json_schema, Schema
from typing import Optional

def test_simple_json_schema():
    class Item(Schema):
        name: Optional[str]

    class Basket(Schema):
        items: list[Item]

    assert json_schema(Basket) == {
        "type": "object",
        "properties": {
            "items": {
                "type": "array",
                "items": {
                    "type": "object",
                    "properties": {"name": {"type": "string"}},
                },
            }
        },
        "required": ["items"]
    }


def test_set_json_schema():
    class Foo(Schema):
        my_set: set[int]

    schema = json_schema(Foo)
    print(json.dumps(schema, indent=4))
    assert schema == {
        "type": "object",
        "properties": {
            "my_set": {
                "type": "array",
                "items": {"type": "integer"},
                "uniqueItems": True,
            }
        },
        "required": ["my_set"],
    }
    with pytest.raises(TypeError):
        Foo(my_set=[1, 2, 3])
    Foo(my_set={1, 2, 3})


def test_enum():
    class SomeEnum(Enum):
        FOO = "Foo"
        BAR = "Bar"
        BAZ = "Baz"

    class Item(Schema):
        key: SomeEnum

    schema = json_schema(Item)
    print(schema)
    assert schema == {
        "type": "object",
        "properties": {
            "key": {"type": "string", "enum": ["Foo", "Bar", "Baz"]}
        },
        "required": ["key"],
    }
    Item(key=SomeEnum.FOO)
    with pytest.raises(TypeError):
        Item(key="Foo")

• 🔥 Annotation attributes

It is possible to use the Annotation metaclass customize a type.

class SomeStringAttribute(str, metaclass=Annotation):
   pass

SomeDescriptedAttribute(2) # raise type error

It is possible to use this to further describe a field.

class SomeDescriptedAttribute(str, metaclass=Annotation):
    annotation = str
    description = "Some description"

using json_schema on SomeDescription will include the description attribute

class DescriptedString(str, metaclass=Annotation):
    description = "Some description"
    annotation = str

class DescriptedItem(Schema):
    descripted: DescriptedString

assert json_schema(DescriptedItem) == {
    "type": "object",
    "properties": {
        "descripted": {
            "type": "string",
            "description": "Some description",
        },
    },
    "required": ["descripted"],
}

• Regular expression

Regex can be defined on an Annotated type using the pattern attribute.

:warning: be careful to respect the json-schema specifications when using json_schema At the moment it is not checked nor tested, and will probably render an invalid json-schema without warning nor error

def test_pattern_definition_allows_normal_usage():
    class PhoneNumber(str, metaclass=Annotation):
        annotation = str
        pattern = r"\d{3}-\d{3}-\d{4}"

    PhoneNumber("000-000-0000")


def test_pattern_raise_type_error():
    class PhoneNumber(str, metaclass=Annotation):
        annotation = str
        pattern = r"\d{3}-\d{3}-\d{4}"

    with pytest.raises(TypeError):
        PhoneNumber("oops")


def test_pattern_is_rendered_in_json_schema():
    class PhoneNumber(str, metaclass=Annotation):
        annotation = str
        pattern = r"^\d{3}-\d{3}-\d{4}$"
        description = "A phone number in the format XXX-XXX-XXXX"

    class Contact(Schema):
        phone: PhoneNumber

    schema = json_schema(Contact)
    print(json.dumps(schema, indent=4))
    assert schema == {
        "type": "object",
        "properties": {
            "phone": {
                "pattern": "^\\d{3}-\\d{3}-\\d{4}$",
                "description": "A phone number in the format XXX-XXX-XXXX",
                "type": "string",
            }
        },
        "required": ["phone"],
    }

• Object validation

It is possible to define a is_valid method on a Schema object, which is during instantiation to allow restrictions based on multiple fields.

def test_object_validation():
    class Item(Schema):
        title: Optional[str]
        content: Optional[str]

        def is_valid(self, **kwargs):
            """title is mandatory if content is absent"""
            return (
                False
                if not kwargs.get("content", None)
                and not kwargs.get("title", None)
                else True
            )

    Item(
        title="foo"
    )  # we should be able to create with only one of title or content
    Item(content="foo")
    with pytest.raises(TypeError):
        Item()

Multiple inheritance

Sometimes technical contraints should not be rendered publicly, and you still want to use the existing class definitions.

For instance one of those occurances is multiple objects that have different realities in the code, but have the same buisness organisation.

Instead of having multiple keys pointing to each object, we would prefer to have a unique item with fields from both classes.

In that occurance we can use multiple inheritance to define a class by combining existing definitions.

def test_multiple_inheritance_json_schema():
    class Foo(Schema):
        foo: str

    class Bar(Schema):
        bar: str

    class FooBar(Foo, Bar):
        pass

    assert len(FooBar.get_fields()) == 2
    schema = json_schema(FooBar)
    print(schema)
    assert schema == {
        "type": "object",
        "properties": {"foo": {"type": "string"}, "bar": {"type": "string"}},
        "required": ["foo", "bar"],
    }

• Immutables

from madtypes import Immutable # Immutable inherits from Schema

class Foo(Immutable):
    name: str
    age: Optional[int]

e = Foo(name="foo")

e.name = "bar" # raise TypeError


b = Foo(**e) # intianciate a new copy
b = Foo(age=2, **e) # create a copy with changes

Installation

pip3 install madtypes

• Context

madtypes is a Python3.9+ library that provides enhanced data type checking capabilities. It offers features beyond the scope of PEP 589 and is built toward an industrial use-case that require reliability.

• The library introduces a Schema class that allows you to define classes with strict type enforcement. By inheriting from Schema, you can specify the expected data structure and enforce type correctness at runtime. If an incorrect type is assigned to an attribute, madtypes raises a TypeError.

• Schema class and it's attributes inherit from dict. Attributes are considered values of the dictionnary.

• It renders natively to JSON, facilitating data serialization and interchange.

• The library also includes a json_schema() function that generates JSON-Schema representations based on class definitions.