chess-guru 0.1.1

Async Chess.com public API client

chess-guru

Async Python client for the Chess.com Public API.

Built on aiohttp with automatic retries for rate limits and server errors. Doesn't mess with your logging config. Supports time-based filtering and concurrent month fetching.

Install

pip install chess-guru

Requires Python 3.9+.

Quickstart

import asyncio
import aiohttp
from chess_guru import ChesscomAPI

async def main():
    async with aiohttp.ClientSession() as session:
        api = ChesscomAPI(session, user_agent="my-app/1.0")
        
        profile = await api.get_player("erik")
        print(profile.get("username"), profile.get("country"))
        
        data = await api.get_games("erik", max_concurrency=8)
        print("months fetched:", len(data["months"]))
        
        first_month = next(iter(data["months"].values()))
        print("games in first month:", len(first_month.get("games", [])))

asyncio.run(main())

You can also pass a custom user agent via headers:

api = ChesscomAPI(session, headers={"User-Agent": "my-app/1.0"})

Return structure

get_games() returns a dict with these keys:

{
  "username": "erik",
  "archives": ["https://api.chess.com/pub/player/erik/games/2024/06", ...],
  "months": {
    "https://api.chess.com/pub/player/erik/games/2024/06": {
      "games": [{...}, {...}, ...]
      # full month payload from chess.com
    }
  },
  "errors": {
    "https://...": "ClientError(...)"
    # month URLs that failed to fetch
  },
  "from_ts": "2024-01-01T00:00:00+00:00",  # or None
  "to_ts": "2024-06-30T23:59:59+00:00"     # or None
}

If a monthly archive fails to fetch, it shows up in errors but doesn't kill the whole request.

Time filtering

Pass from_ts and/or to_ts as datetime objects to filter by game end time. Both should be UTC (naive datetimes are assumed to be UTC).

from datetime import datetime, timezone

from_ts = datetime(2024, 1, 1, tzinfo=timezone.utc)
to_ts = datetime(2024, 6, 30, 23, 59, 59, tzinfo=timezone.utc)

data = await api.get_games("erik", from_ts=from_ts, to_ts=to_ts)

The filter works in two passes: first it skips months outside your time range, then it filters individual games by end_time within the months that remain.

Concurrency control

max_concurrency limits how many month requests run in parallel:

data = await api.get_games("erik", max_concurrency=5)

Default is 10. If you're hitting rate limits, lower it. If you want faster fetching and the API can handle it, raise it.

Logging

The library uses Python's standard logging but doesn't configure anything by default. If you want to see retry warnings or debug output:

import logging
logging.basicConfig(level=logging.WARNING)

Set level to INFO or DEBUG for more detail.

API methods

ChesscomAPI(
    session,
    base_url="https://api.chess.com/pub",
    timeout_sec=20,
    user_agent="chess-guru/0.1.0",
    headers=None
)

api.get_player(username)
api.get_archives(username)
api.get_games(username, max_concurrency=10, from_ts=None, to_ts=None)

License

PolyForm Noncommercial 1.0.0 — free for noncommercial use.
Commercial licensing: [e-mail]