🏎️ pyopenf1

A production-grade, asynchronous Python wrapper for the OpenF1 API.

✨ Features

• Fully Async — Built on httpx.AsyncClient with first-class async/await support.
• Sync Wrapper — OpenF1Client for users who don't need async.
• All 18 Endpoints — Complete coverage: telemetry, laps, drivers, sessions, weather, pit stops, stints, overtakes, race control, team radio, championship standings, results, starting grid, and more.
• Pydantic V2 Models — Every response is validated and returned as a strict, typed model.
• Auto-Retry — Exponential backoff on 429/5xx via tenacity.
• Rate Limiting — Built-in client-side throttle (3 req/s free tier).
• Response Caching — Optional in-memory TTL cache.
• CLI Tool — Query the API from your terminal.
• DataFrame Support — Optional pandas integration.
• Production-Ready — Custom exceptions, structured logging, connection pooling.

📦 Installation

pip install pyopenf1

With pandas support:

pip install pyopenf1[pandas]

🚀 Quickstart

Async (recommended)

import asyncio
from pyopenf1 import AsyncOpenF1Client

async def main() -> None:
    async with AsyncOpenF1Client() as client:
        # Fetch telemetry
        car_data = await client.telemetry.get_car_data(driver_number=1, session_key=9159)
        for entry in car_data[:5]:
            print(f"Speed: {entry.speed} km/h | Gear: {entry.n_gear}")

        # Fetch drivers
        drivers = await client.drivers.get_drivers(session_key=9158)
        for d in drivers[:3]:
            print(f"#{d.driver_number} {d.full_name} - {d.team_name}")

asyncio.run(main())

Sync

from pyopenf1 import OpenF1Client

with OpenF1Client() as client:
    drivers = client.drivers.get_drivers(session_key=9158)
    for d in drivers:
        print(f"{d.name_acronym} - {d.team_name}")

CLI

pyopenf1 car-data --driver 1 --session 9159 --format table
pyopenf1 drivers --session 9158 --format json
pyopenf1 weather --meeting 1208 --format csv --output weather.csv

DataFrame

from pyopenf1.ext.pandas import to_dataframe

data = await client.telemetry.get_car_data(driver_number=1)
df = to_dataframe(data)
print(df.describe())

⚙️ Configuration

async with AsyncOpenF1Client(
    cache_ttl=300.0,        # Cache responses for 5 minutes
    max_retries=5,          # Retry up to 5 times
    max_per_second=6.0,     # Sponsor-tier rate limit
    max_per_minute=60.0,
) as client:
    ...

📡 Available Endpoints

🏗️ Architecture

pyopenf1/
├── __init__.py              # Public API surface
├── client.py                # AsyncOpenF1Client facade
├── sync_client.py           # OpenF1Client (sync wrapper)
├── cli.py                   # Click CLI
├── exceptions.py            # Custom exception hierarchy
├── core/
│   ├── http_client.py       # httpx wrapper + retry + logging
│   ├── rate_limiter.py      # Token-bucket rate limiter
│   └── cache.py             # In-memory TTL cache
├── models/                  # Pydantic V2 data models
│   ├── telemetry.py         # CarData, Location
│   ├── session.py           # Session, Meeting
│   ├── driver.py            # Driver
│   ├── timing.py            # Lap, Interval, Position
│   ├── race.py              # RaceControl, Pit, Stint, Overtake
│   ├── championship.py      # ChampionshipDriver, ChampionshipTeam
│   ├── results.py           # SessionResult, StartingGrid
│   ├── weather.py           # Weather
│   └── team_radio.py        # TeamRadio
├── endpoints/               # API endpoint classes
│   ├── telemetry_api.py
│   ├── session_api.py
│   ├── driver_api.py
│   ├── timing_api.py
│   ├── race_api.py
│   ├── championship_api.py
│   ├── results_api.py
│   ├── weather_api.py
│   └── team_radio_api.py
└── ext/
    └── pandas.py            # Optional DataFrame integration

🧪 Development

# Install dev dependencies
poetry install

# Lint & format
poetry run ruff check .
poetry run ruff format .

# Type check
poetry run mypy pyopenf1/

# Test
poetry run pytest -v

# Build docs
poetry run mkdocs serve

📄 License

Distributed under the MIT License. See LICENSE for details.