Database¶

FinanceLib provides database storage for market data and news articles.

SQLite (Built-in)¶

from financelib.database.sqlite import SQLite
from financelib.database.models import NewsModel, StockDataModel

# Context manager ensures proper cleanup
with SQLite("market_data") as db:
    # Insert news
    news = NewsModel(
        title="BIST Up 3%",
        content="Turkish stocks surged...",
        source="Bloomberg",
        published_at="2026-01-15",
    )
    db.insert_article(news)

    # Insert stock data
    stock = StockDataModel(
        symbol="THYAO.IS",
        price=172.45,
        open_price=170.00,
        high=173.50,
        low=169.80,
        volume=24532100,
        change=2.45,
        change_percent=1.44,
    )
    db.insert_stock_data(stock)

    # Query
    results = db.fetchall("SELECT * FROM stock_data WHERE symbol = ?", ("THYAO.IS",))

PostgreSQL¶

For production deployments, use the PostgreSQL backend.

pip install financelib[db]

from financelib.database.postgres import PostgreSQL
from financelib.database.models import StockDataModel

# Connect via connection string or DATABASE_URL env var
with PostgreSQL("postgresql://user:pass@localhost:5432/financelib") as db:
    stock = StockDataModel(
        symbol="THYAO.IS",
        price=172.45,
        volume=24532100,
    )
    db.insert_stock_data(stock)

    # Store trade history
    db.insert_trade(
        symbol="BTC/USDT",
        side="buy",
        price=65000.0,
        amount=0.1,
        market="spot",
    )

    # Store indicator snapshots
    db.insert_indicator_snapshot(
        symbol="THYAO.IS",
        rsi=45.2,
        macd=0.0023,
        bb_upper=175.0,
        bb_lower=168.0,
    )

    results = db.fetchall("SELECT * FROM stock_data")

Tip

Set DATABASE_URL in your environment and you can create a connection with just PostgreSQL().

Additional PostgreSQL Tables¶

The PostgreSQL backend includes extra tables not in SQLite:

trade_history — Records all trades (symbol, side, price, amount, market, leverage, pnl)
indicator_snapshots — Stores all 11 indicator values at a point in time
created_at timestamps on all tables

Data Models¶

NewsModel¶

Field	Type	Description
`title`	str	Article title
`content`	str	Article body
`published_at`	str	Publication date
`author`	str	Author name
`category`	str	Article category
`source`	str	Source name
`article_url`	str	Article URL
`article_thumbnail_url`	str	Thumbnail URL

StockDataModel¶

Field	Type	Description
`symbol`	str	Stock ticker
`price`	float	Current price
`open_price`	float	Opening price
`high`	float	Day high
`low`	float	Day low
`volume`	int	Trading volume
`change`	float	Price change
`change_percent`	float	Change %
`currency`	str	Currency code
`timestamp`	str	Data timestamp