FastAPI does not handle ValidationError when updating

[ecosys-technology]

When using the PUT endpoint to update the Student data, if gpa is higher than 4.0, FastAPI returns a 500 Internal Server Error.

A quick fix is to add the validation directly to the UpdateStudentModel's gpa like this :

class UpdateStudentModel(BaseModel):
    name: Optional[str]
    email: Optional[EmailStr]
    course: Optional[str]
    gpa: Optional[float] = Path(..., le=4.0)

    class Config:
        arbitrary_types_allowed = True
        json_encoders = {ObjectId: str}
        schema_extra = {
            "example": {
                "name": "Jane Doe",
                "email": "jdoe@example.com",
                "course": "Experiments, Science, and Fashion in Nanophotonics",
                "gpa": "3.0",
            }
        }

But this solution is a little bit redundant. Is there a better solution ?

This is relevant : https://fastapi.tiangolo.com/tutorial/handling-errors/#requestvalidationerror-vs-validationerror

[sanjeebtiwary]

One alternative solution is to define a custom validation function for the GPA field using the validator decorator provided by Pydantic.
`from pydantic import validator

class UpdateStudentModel(BaseModel):
name: Optional[str]
email: Optional[EmailStr]
course: Optional[str]
gpa: Optional[float]

@validator('gpa')
def validate_gpa(cls, value):
    if value and value > 4.0:
        raise ValueError('gpa must be less than or equal to 4.0')
    return value

class Config:
    arbitrary_types_allowed = True
    json_encoders = {ObjectId: str}
    schema_extra = {
        "example": {
            "name": "Jane Doe",
            "email": "jdoe@example.com",
            "course": "Experiments, Science, and Fashion in Nanophotonics",
            "gpa": "3.0",
        }
    }

`
In this implementation, the validator decorator is used to define a validation function for the gpa field. The function checks if the gpa value is greater than 4.0 and raises a ValueError with an appropriate message if it is. Otherwise, it returns the gpa value unchanged.

Using this approach, the validation logic is separated from the field definition, making the code more modular and easier to maintain.

[judy2k]

I need to spend some time testing how errors are raised and returned, and then work out the correct way to serialize them for the API client.