[Python] Python Pydantic `TypeAdapter` validate with default for missing attribute

Suppose I have:

class Thing(pydantic.BaseModel):
one: int
two: str
three: bool

adapter = pydantic.TypeAdapter(dict[str, Thing])

raw_data = {"this": {"one": 1, "two": "zwei"}, "that": {"one": 111, "two": "dos"}}


You can see that the raw_data, which I want to convert (validate) using adapter, contains two items that are missing the attribute three. I want to be able to still validate this, but give it a default value True for the missing attribute three. However, I don't want to do this in the definition of class Thing, because in my use-case this default behaviour is not an intrinsic part of the class Thing. I don't want users of this class to assume there's always a default for the attribute three; I only want to provide a default value this time only, for this particular raw_data only, because I'm expecting this particular raw_data to be missing three for some reason. (In particular, this question looks similar but does not solve my problem.) So, what I want is something like:

things_i_want = adapter.validate_python(raw_data, default={"three": True})


Or:

or_this = adapter.validate_python(raw_data, default=pydantic.Default(three=True))


Is this possible at all in Pydantic?

Continue reading...