Alpaca-Py SDK

Warnings

• breaking The `get_corporate_actions` method's response structure changed significantly in v0.42.0 to include ID/CUSIPs fields, potentially altering the order of response/dataframe columns. Relying on column order is no longer safe.
  Fix: Always access data by field/column names (e.g., `response.field_name` or `df['column_name']`) instead of positional indexing.
• deprecated The `get_corporate_announcements` method was deprecated in v0.42.1.
  Fix: Use the `get_corporate_actions` method instead for retrieving corporate action data.
• breaking Migration from the older `alpaca-trade-api` SDK to `alpaca-py` involves significant architectural changes, including a new object-oriented design and different import paths.
  Fix: Review the official `alpaca-py` documentation for updated client initialization, request object patterns, and module import paths. Many examples from the old SDK are incompatible.
• gotcha Alpaca-py utilizes an object-oriented design where most API calls require specific 'request objects' (e.g., `MarketOrderRequest`, `StockBarsRequest`) instead of direct keyword arguments. Failing to provide the correct request object will result in errors.
  Fix: Consult the `alpaca-py` documentation for the correct request object class corresponding to each API method (e.g., `TradingClient.submit_order` requires a subclass of `OrderRequest`).
• gotcha API keys (ALPACA_API_KEY_ID and ALPACA_SECRET_KEY) are crucial for authentication. Hardcoding them is insecure and against best practices.
  Fix: Store API keys securely as environment variables (recommended) or use a secrets management system. The SDK will automatically pick them up if named correctly, or they can be passed explicitly during client initialization.
• gotcha Alpaca-py employs multiple client classes for different API services and asset types (e.g., `TradingClient`, `StockHistoricalDataClient`, `CryptoHistoricalDataClient`). You must instantiate the correct client for your desired operation.
  Fix: Identify whether you need to perform trading operations, retrieve historical stock data, or historical crypto data, and instantiate the corresponding client (e.g., `TradingClient`, `StockHistoricalDataClient`, `CryptoHistoricalDataClient`).

Install

• pip install alpaca-py Install latest version

Imports

• TradingClient

  from alpaca.trading.client import TradingClient

• StockHistoricalDataClient

  from alpaca.data.historical import StockHistoricalDataClient

• MarketOrderRequest

  from alpaca.trading.requests import MarketOrderRequest

• OrderSide

  from alpaca.trading.enums import OrderSide

• TimeFrame

  from alpaca.data.timeframe import TimeFrame

• StockBarsRequest

  from alpaca.data.requests import StockBarsRequest

• alpaca_trade_api

  Not applicable in alpaca-py

  This is the old, deprecated SDK. alpaca-py uses a different, more modular import structure.

Quickstart

This quickstart demonstrates how to initialize the `TradingClient` using API keys from environment variables, fetch account details, and submit a basic market order for a stock. It uses a paper trading account by default for risk-free testing.

import os
from alpaca.trading.client import TradingClient
from alpaca.trading.requests import MarketOrderRequest
from alpaca.trading.enums import OrderSide, TimeInForce

# Ensure API keys are set as environment variables for security
API_KEY = os.environ.get('ALPACA_API_KEY_ID', '')
API_SECRET = os.environ.get('ALPACA_SECRET_KEY', '')

if not API_KEY or not API_SECRET:
    print("Error: ALPACA_API_KEY_ID and ALPACA_SECRET_KEY environment variables must be set.")
    exit()

# Initialize TradingClient (paper=True for paper trading account)
trading_client = TradingClient(api_key=API_KEY, secret_key=API_SECRET, paper=True)

try:
    # Get account information
    account = trading_client.get_account()
    print(f"Account ID: {account.id}")
    print(f"Status: {account.status}")
    print(f"Equity: {account.equity}")

    # Place a market order (example: buy 1 share of AAPL)
    market_order_data = MarketOrderRequest(
        symbol='AAPL',
        qty=1,
        side=OrderSide.BUY,
        time_in_force=TimeInForce.DAY
    )
    market_order = trading_client.submit_order(market_order_data)
    print(f"\nSubmitted Market Order:")
    print(f"  Order ID: {market_order.id}")
    print(f"  Symbol: {market_order.symbol}")
    print(f"  Side: {market_order.side}")
    print(f"  Quantity: {market_order.qty}")

except Exception as e:
    print(f"An error occurred: {e}")