How to handle JSON payload validation in a custom API endpoint?

[bc230426875hal-cloud]

I am creating a simple endpoint to process incoming JSON payloads, but I want to make sure missing or malformed keys return a clear 400 Bad Request instead of throwing an unhandled exception.

What is the cleanest way to validate optional and required fields in a dictionary before processing the data? Should I rely on try/except KeyError blocks or a structured schema library? Any small code snippet would be appreciated!

[Hassan0703]

Using a structured schema library is the cleanest, most scalable way to validate JSON data. While try/except KeyError blocks work for tiny payloads, they quickly become messy, hard to maintain, and difficult to format into clear error messages for the client.Why Schema Libraries WinCentralized Logic: Validation rules sit in one clean structure.Automatic Error Handling: They catch missing fields, wrong data types, and malformed structures simultaneously.Better Client Feedback: They return a precise list of what went wrong, which is ideal for a 400 Bad Request response.Implementation OptionsHere are the two best approaches using standard Python tools.Option 1: Pydantic (Recommended for modern Python/FastAPI)Pydantic uses Python type hints. It is the industry standard for fast, clean data validation.pythonfrom typing import Optional
from pydantic import BaseModel, ValidationError

1. Define your structure

class UserPayload(BaseModel):
username: str # Required
email: str # Required
age: Optional[int] = None # Optional (Defaults to None)

2. Validate incoming data

def process_request(json_data):
try:
# Pydantic validates keys and data types automatically
user = UserPayload(**json_data)
return {"status": "success", "data": user.dict()}, 200

except ValidationError as e:
    # e.errors() returns a clean list of exactly what is missing or wrong
    return {"error": "Bad Request", "details": e.errors()}, 400

Use code with caution.Option 2: Marshmallow (Great for Flask/Django)If you are working with legacy systems or prefer explicit schema objects, Marshmallow is a highly flexible alternative.pythonfrom marshmallow import Schema, fields, ValidationError

1. Define your structure

class UserSchema(Schema):
username = fields.Str(required=True)
email = fields.Str(required=True)
age = fields.Int(required=False) # Optional by default

2. Validate incoming data

def process_request(json_data):
try:
# Validate and deserialize
result = UserSchema().load(json_data)
return {"status": "success", "data": result}, 200

except ValidationError as e:
    # e.messages contains structured key-error pairs
    return {"error": "Bad Request", "details": e.messages}, 400