Support for Serializing/Deserializing Built-in IP Address Types #1856
Description
Description
Will there be support for serializing/deserializing Python's built-in IP addresse types? Currently, they are treated as unknown types.
I am aware of serialize_unknown parameter and it can remedy this particular case, but there can be cases where raising an error for other types is appropriate. It just feels reasonable that Python’s built-in IP address types should be treated as known types and automatically serialized/deserialized to/from strings (as they already have a standard __str__ representation).
my_data: dict[str, Any] = {
"ipv4": IPv4Address("1.1.1.1"),
"my_decimal": Decimal(123),
"more_data": [{"kissa": "koira"}],
"nested_ip": {"ip": IPv4Address("1.2.2.3")},
}
to_json(my_data)Raises:
Traceback (most recent call last):
File "/xx/xx.py", line xx, in <module>
to_json(my_data)
^^^^^^^^^^^^^^^^
pydantic_core._pydantic_core.PydanticSerializationError: Unable to serialize unknown type: <class 'ipaddress.IPv4Address'>I am aware of serialize_unknown and it remedies this particular case, but there are cases when raising an error is appropriate. I just think that it would be reasonable to consider Python's built-in IP addresses as known types that can be serialized/deserialized.
My solution for now
my_data: dict[str, Any] = {
"ipv4": IPv4Address("1.1.1.1"),
"my_decimal": Decimal(123),
"more_data": [{"kissa": "koira"}],
"nested_ip": {"ip": IPv4Address("1.2.2.3")},
}
type AnyDict = dict[str, Any | AnyDict | IPvAnyAddress]
adapter: TypeAdapter[AnyDict] = TypeAdapter(AnyDict)
adapter.dump_python(my_data, mode="json")
>>> {'ipv4': '1.1.1.1', 'my_decimal': '123', 'more_data': [{'kissa': 'koira'}], 'nested_ip': {'ip': '1.2.2.3'}}Version info
pydantic version: 2.12.3
pydantic-core version: 2.41.4
pydantic-core build: profile=release pgo=false
python version: 3.12.11 (main, Oct 6 2025, 09:15:07) [GCC 13.3.0]
platform: Linux-6.14.11-061411-generic-x86_64-with-glibc2.39
related packages: typing_extensions-4.15.0Related: pydantic/pydantic#8942
Related: pydantic/pydantic#10559