Advance Usage
Advance¶
This is guideline for how to use the library in more advanced ways.
Customization of the OpenAPI schema¶
All you need to do, is to return type HTTPExceptionSchema correctly on the method get_json_schema of your exception class which you inherited from BaseHTTPException.
Imagine you are writing an exception that is returning a user balance and a time to retry again a failed bank transaction, you can define the schema like this:
from richapi import BaseHTTPException
from richapi.exc_parser import HTTPExceptionSchema
from pydantic import create_model
class YourException(BaseHTTPException):
user_balance: int
retry_again_in: int
status_code: int = 400
@classmethod
def get_json_schema(cls) -> HTTPExceptionSchema:
schema = create_model(name, user_balance=(int, ...), retry_again_in=(int, ...))
return HTTPExceptionSchema(
schema_name="YourException",
schema_desc="This is a custom exception",
status_code=cls.status_code,
response_schema_json=schema.model_json_schema(),
)
If your intention is to have a custom response schema, then recommended way is to use custom exception feature.
Finding exceptions in other modules¶
By default, the library only looks at the FastAPI app module to search for exceptions. This is to avoid parsing countless dependency modules that are not related to the FastAPI app. But sometimes you may be using some other libraries that raises HTTPExceptions. For example using a third library for JWT authentication.
In that case, you can simply pass the module name to the parse_exceptions function.