xfinance 0.3.0

The definitive open-source yfinance alternative — multi-source, failover-capable, fully normalized

findata

A multi-source, failover-capable Python financial data library.

findata delivers yfinance's beloved simplicity with 10× the reliability through automatic failover across multiple data sources, intelligent three-tier caching, and a community plugin architecture.

Why findata?

Problem with yfinance	findata solution
Breaks 2–4×/year when Yahoo changes endpoints	Circuit-breaker failover to SEC, ECB, Binance
No built-in caching; hammers APIs on every call	Three-tier cache (memory → diskcache → Parquet)
No rate limiting; users get IP-blocked	Per-source adaptive rate limiters
Silent failures (empty DataFrames)	Rich exceptions with source provenance
Single unofficial API	5+ sources, plugin architecture for more

Quick start

pip install findata

import findata

# One line — works immediately, no API keys needed
df = findata.prices("AAPL", period="1mo")

# Multi-source with validation
client = findata.Client(sources=["yahoo", "sec", "ecb"])
df = client.prices("AAPL", start="2024-01-01", validate=True)

# Async batch fetching
import asyncio

async def main():
    async with findata.AsyncClient() as client:
        data = await client.prices(["AAPL", "GOOGL", "MSFT"], period="1y")

asyncio.run(main())

# Fundamentals from SEC EDGAR (free, no key needed)
info = findata.info("AAPL")

# Forex from ECB (free, no key, no rate limits)
fx = findata.forex("USD/EUR", period="1y")

# Crypto from Binance public API (free, no key)
btc = findata.prices("BTC/USDT", period="30d")

Optional extras

pip install findata[cache]    # diskcache for persistent local caching
pip install findata[duckdb]   # DuckDB + Parquet for historical storage
pip install findata[polars]   # Polars DataFrame output
pip install findata[all]      # Everything

Data sources (v0.1)

Source	Data	Auth	Rate limit
Yahoo Finance	Prices, options, info	None (unofficial)	~950 req/session
SEC EDGAR	Fundamentals, filings	None	10 req/s
ECB / Frankfurter	Forex	None	No limit
Binance	Crypto OHLCV	None	1200 req/min
CoinGecko	Crypto info	None (free tier)	30 req/min

See DATA_SOURCES.md for full terms-of-service details and risk tiers.

Community plugins

Add new sources as pip-installable packages:

# In your plugin's pyproject.toml
[project.entry-points."findata.sources"]
tradingview = "findata_tradingview:TradingViewSource"

A new source adapter can be implemented in under 100 lines by implementing the DataSource protocol.

Legal notice

findata fetches data on behalf of users as a client library. Users are responsible for complying with each data provider's terms of service. Government data (SEC EDGAR) is in the public domain. See DATA_SOURCES.md for source-specific guidance.

License

Apache 2.0