Rate-Limiting API Calls¶

Protect external APIs from excessive traffic using the retry module with custom backoff strategies.

The Problem¶

External APIs have rate limits. When you exceed them, requests fail. You need sensible retry logic that respects rate limits without hammering the API.

The Solution¶

Combine forge's @retry decorator with the circuit breaker to handle rate limits gracefully.

Basic Rate-Limiting¶

from forge.retry import retry

@retry(attempts=5, backoff="exponential")
async def call_api(endpoint: str):
    async with httpx.AsyncClient() as client:
        response = await client.get(f"https://api.example.com{endpoint}")
        response.raise_for_status()
        return response.json()

Rate-Limit Aware with Circuit Breaker¶

from forge.retry import retry, CircuitBreaker

breaker = CircuitBreaker(
    failure_threshold=10,    # Open circuit after 10 failures
    recovery_time=30.0,      # Wait 30 seconds before retrying
)

@retry(
    attempts=3,
    backoff="exponential",
    circuit_breaker=breaker,
)
async def call_api(endpoint: str):
    ...

Manual Retry with Backoff¶

from forge.retry import attempt

async def fetch_with_retry(url: str):
    async with attempt(max_attempts=5) as ctx:
        response = await http_client.get(url)
        if response.status_code == 429:  # Rate limited
            ctx.retry()  # Will wait with backoff
        response.raise_for_status()
        return response.json()