Unit and integration tests for the Kalshi API client library.

# Run all tests
pytest tests/

# Run with verbose output
pytest tests/ -v

# Run specific test file
pytest tests/test_exchange.py

# Run specific test class
pytest tests/test_exchange.py::TestExchangeStatus

# Run with coverage
pytest tests/ --cov=pykalshi

• mock_response: Factory for creating mock HTTP responses with JSON data and status codes
• client: Pre-configured KalshiClient with mocked auth and HTTP session (no real API calls)

• Auth header generation with RSA-PSS signatures
• HTTP response handling (success, errors)
• Custom exceptions: AuthenticationError, ResourceNotFoundError, InsufficientFundsError
• Error message and code extraction

• All Pydantic models validate correctly
• Enum parsing (Action, Side, OrderStatus, MarketStatus)
• Optional fields handled
• Extra fields ignored (forward compatibility)
• Type coercion and validation errors

• get_positions() with filters (ticker, event_ticker, count_filter)
• get_fills() with filters (ticker, order_id, min_ts, max_ts)
• get_order() by ID
• 404 handling for missing orders

• portfolio.get_balance() method
• portfolio.place_order() with Market objects
• market.get_orderbook() returns typed response

• URL selection (demo vs production)
• Handler registration (direct and decorator)
• Subscription management
• Message dispatch and parsing for all channel types:
  • ticker, trade, fill
  • orderbook_snapshot, orderbook_delta
• Error handling (malformed JSON, handler exceptions)
• Auth header generation

• client.exchange.get_status() - operational status
• client.exchange.is_trading() - quick boolean check
• client.exchange.get_schedule() - trading hours
• client.exchange.get_user_data_timestamp() - data validation timestamp
• Cached property behavior

• client.api_keys.list() - list all keys
• client.api_keys.create() - create with public key
• client.api_keys.generate() - generate key pair
• client.api_keys.delete() - delete key
• client.api_keys.get_limits() - rate limit info
• 404 handling for non-existent keys

• client.get_series() - fetch single series
• client.get_all_series() - list with category filter
• series.get_markets() - markets in a series
• client.get_trades() - public trade history with filters
• client.get_candlesticks_batch() - batch candlestick retrieval
• Pagination with fetch_all=True

• client.get_market() - fetch single market
• client.get_markets() - list with filters (series, event, status)
• market.get_candlesticks() - historical OHLC data
• Series ticker resolution from event
• client.get_event() - fetch single event
• client.get_events() - list with filters
• Market equality and hashing
• Attribute delegation to underlying model

def test_example(client, mock_response):
    client._session.request.return_value = mock_response({
        "data": {"key": "value"}
    })

    result = client.get("/endpoint")

    assert result["data"]["key"] == "value"

def test_not_found(client, mock_response):
    client._session.request.return_value = mock_response(
        {"message": "Not found"}, status_code=404
    )

    with pytest.raises(ResourceNotFoundError):
        client.get("/missing")

def test_pagination(client, mock_response):
    client._session.request.side_effect = [
        mock_response({"items": [1], "cursor": "page2"}),
        mock_response({"items": [2], "cursor": ""}),
    ]

    result = client.get_items(fetch_all=True)

    assert len(result) == 2
    assert client._session.request.call_count == 2

def test_filters(client, mock_response):
    client._session.request.return_value = mock_response({"items": []})

    client.get_items(status="active", limit=50)

    call_url = client._session.request.call_args.args[1]
    assert "status=active" in call_url
    assert "limit=50" in call_url

test_integration.py contains tests that hit the real Kalshi API. These are skipped by default unless valid credentials are configured:

# Set environment variables
export KALSHI_API_KEY_ID="your_key"
export KALSHI_PRIVATE_KEY_PATH="/path/to/key.pem"

# Run integration tests
pytest tests/test_integration.py -v

Integration tests verify:

• Balance retrieval
• Market listing and fetching
• Orderbook data
• Position queries
• Fill history
• Order queries