Tenacity Client -

@retry(retry=retry_if_result(lambda x: x is None)) def get_user(): ... Control delays between retries:

retryer = Retrying(stop=stop_after_attempt(3)) with retryer: unreliable_operation() Built-in logging of retry events (supports custom logger). 11. Integration with asyncio , Tornado , gevent Event-loop aware waiting. Example: Resilient HTTP Client from tenacity import ( retry, stop_after_attempt, wait_exponential, retry_if_exception_type, before_sleep_log ) import requests import logging logging.basicConfig(level=logging.INFO) tenacity client

@retry(retry=retry_if_exception_type(requests.RequestException)) def fetch_data(): ... Retry based on return value or exception message. Integration with asyncio , Tornado , gevent Event-loop

from tenacity import retry @retry def unstable_http_call(): # Automatically retries on any exception ... Retry only on specific exceptions (or retry except some). pybreaker ) for advanced resilience patterns.

Tenacity is a powerful, Apache 2.0-licensed library that simplifies adding retry behavior to Python functions, methods, or any callable. It’s a modern alternative to retrying and provides fine-grained control over retry policies, especially for HTTP clients, database connectors, and API wrappers. Core Features 1. Decorator-Based Retries ( @retry ) Apply retry logic with minimal boilerplate.

@retry(...) def op(): ... op.retry.statistics # e.g., 'attempt_number': 3, 'delay': 2.1 Native support for async def functions. 9. Retrying Arbitrary Blocks (not just functions) Use Retrying context manager:

from tenacity import retry @retry def never_succeed(): raise Exception("Always fails") Tenacity is the go-to retry library for Python clients. Its expressive API, extensive wait strategies, and async support make it suitable for everything from simple scripts to production microservices. It replaces the now-deprecated retrying library and offers more control than a manual while loop. ✅ Recommendation: Use Tenacity for any client that communicates over unreliable networks or resources. Pair with circuit breakers (e.g., pybreaker ) for advanced resilience patterns.