Retry Logic with Python Decorator | Dataford Interview Questions

Write a Python script using a decorator that retries a function up to 3 times if it raises an exception.

Implement retry(failure_count, arguments). The input describes a deterministic function call: failure_count specifies how many initial attempts raise RuntimeError, and arguments contains JSON-compatible args and kwargs. The decorated function returns args[0] * kwargs.get("multiplier", 2) after the failures. Return { "result": value, "calls": count } on success. If all four total attempts fail, return { "raised": "RuntimeError from attempt N", "calls": count }, where the retry policy allows one initial attempt plus three retries.

Examples: retry(0, {"args": [5], "kwargs": {}}) returns { "result": 10, "calls": 1 }. retry(4, {"args": [7], "kwargs": {}}) returns { "raised": "RuntimeError from attempt 4", "calls": 4 }.

Constraints

0 <= failure_count <= 4
arguments contains exactly the keys args and kwargs
args contains one integer
kwargs contains zero or one multiplier key
The retry policy makes at most 4 total calls

Function Signature

def retry(failure_count, arguments):

Your question is Retry Logic with Python Decorator. Start with the requirements in the Question tab.

Run and submit as often as you like. When you're ready, talk me through your approach or go straight to the code.

Problem

Write a Python script using a decorator that retries a function up to 3 times if it raises an exception.

Examples: retry(0, {"args": [5], "kwargs": {}}) returns { "result": 10, "calls": 1 }. retry(4, {"args": [7], "kwargs": {}}) returns { "raised": "RuntimeError from attempt 4", "calls": 4 }.

Constraints

0 <= failure_count <= 4
arguments contains exactly the keys args and kwargs
args contains one integer
kwargs contains zero or one multiplier key
The retry policy makes at most 4 total calls

Function Signature

def retry(failure_count, arguments):

Your question is Retry Logic with Python Decorator. Start with the requirements in the Question tab.

Run and submit as often as you like. When you're ready, talk me through your approach or go straight to the code.