How should exceptions be handled in production Python?

Python uses exceptions for both unexpected failures and expected conditions. The try statement supports except for handling, else for code that runs only when no exception occurred, and finally for unconditional cleanup.

class PaymentError(Exception):
    pass

def charge(amount):
    if amount <= 0:
        raise ValueError("amount must be positive")
    try:
        return gateway.charge(amount)
    except TimeoutError as exc:
        raise PaymentError("gateway timed out") from exc

try:
    charge(-1)
except ValueError as exc:
    print("invalid input:", exc)
except PaymentError as exc:
    print("payment failed:", exc)

Practices that hold up

• Never use a bare except: it also catches KeyboardInterrupt and SystemExit
• Use EAFP: try the operation and handle the specific failure rather than pre-checking everything
• Log once at the boundary where you handle the error, not at every layer
• Define a small exception hierarchy for the domain and translate third-party errors at the boundary

Swallowing an exception with pass hides bugs and makes incidents hard to diagnose. If you catch and continue, record why and what state was affected. Raising a new exception without from loses the original traceback chain, which is exactly the information you need in production.