When assigning an incorrect attribute to a Pydantic model field, no validation error occurs.

from pydantic import BaseModelclass pyUser(BaseModel):username: strclass Config:validate_all = Truevalidate_assignment = Trueperson = pyUser(username=1234)
person.username
>>>1234
try_again = pyUser()pydantic.error_wrappers.ValidationError:
[ErrorWrapper(exc=MissingError(), loc=('username',))]
<class '__main__.pyUser'>

How can I get pydantic to validate on assignment?

It is expected behaviour according to the documentation:

str

strings are accepted as-is, int, float and Decimal are coerced using str(v)

You can use the StrictStr, StrictInt, StrictFloat, and StrictBool types to prevent coercion from compatible types.

from pydantic import BaseModel, StrictStrclass pyUser(BaseModel):username: StrictStrclass Config:validate_all = Truevalidate_assignment = Trueperson = pyUser(username=1234)  # ValidationError `str type expected`
print(person.username)