pykabutan

A clean, beginner-friendly Python library for scraping kabutan.jp (Japanese stock information site).

Installation

pip install pykabutan

Quick Start

import pykabutan as pk

# Get stock information
ticker = pk.Ticker("7203")  # Toyota
print(ticker.profile.name)      # トヨタ自動車
print(ticker.profile.market)    # 東証P
print(ticker.profile.per)       # 10.5

# Get price history (yfinance-style), indexed by date
df = ticker.history(period="30d")
print(df)

# Get financial statements (dict of DataFrames, values in 百万円)
financials = ticker.financials()
print(financials["annual"])

# Search by industry
results = pk.search_by_industry("電気機器")
for t in results[:5]:
    print(t.code, t.profile.name)

# Search by theme (Japanese terms only)
results = pk.search_by_theme("人工知能")

Features

• Simple, yfinance-style API
• Lazy loading for performance
• Works out of the box with sensible defaults
• No Selenium dependency (lightweight)

What's New in 0.2.0

• Real financials(): parses the finance page into a dict of DataFrames (annual, interim, quarterly, cashflow, profitability, financial_position, records), values in 百万円 (millions of yen).
• Calendar-accurate history() periods: period="30d" / "6mo" / "1y" now mean actual calendar time, not a row count.
• Connection reuse + retries: HTTP requests share a pooled requests session with automatic retries on transient errors (429/500/502/503).
• ScrapingError: raised (with a .what attribute) when kabutan.jp's page structure changes and a page can't be parsed.

Development

# Install dependencies
uv sync

# Run tests
uv run pytest

# Run integration tests (real HTTP)
uv run pytest -m integration

# Format code
uv run ruff format .

# Lint
uv run ruff check .

License

MIT