Metadata-Version: 2.1
Name: btgsolutions-dataservices-python-client
Version: 0.4
Summary: Python package containing several classes and data for extracting and manipulating market and trading data.
Home-page: https://github.com/BTG-Pactual-Solutions/btgsolutions-dataservices-python-client
Author: BTG Solutions Data Services powered by BTG Pactual Solutions
License: UNKNOWN
Description: # BTG Solutions - Data Service
        
        It's a Python library to get Brazilian Financial Market Data.
        
        ## Installation
        
        ```bash
        pip3 install btgsolutions-dataservices-python-client
        ```
        
        ## Example - WebSocket Books
        ```python
        import btgsolutions_dataservices as btg
        ws = btg.WebSocketClient(api_key='YOUR_API_KEY', ws_type='books', instruments=['PETR4', 'VALE3'])
        ws.run(on_message=lambda message: print(message))
        ```
        ## Example - WebSocket Trades Delayed
        ```python
        import btgsolutions_dataservices as btg
        ws = btg.WebSocketClient(api_key='YOUR_API_KEY', ws_type='trades', target='delayed', instruments=['PETR4', 'VALE3'])
        ws.run(on_message=lambda message: print(message))
        ```
        
        ## Example - IntradayCandles
        ```python
        import btgsolutions_dataservices as btg
        int_candles = btg.IntradayCandles(api_key='YOUR_API_KEY')
        int_candles.get_intraday_candles(market_type='stocks', tickers=['PETR4', 'VALE3'], candle_period='1m', mode='relative', raw_data=True)
        ```
        ## Example - Plot HistoricalCandles
        ```python
        import btgsolutions_dataservices as btg
        hist_candles = btg.HistoricalCandles(api_key='YOUR_API_KEY')
        hist_candles.get_historical_candles(ticker='PETR4', lookback='5D', mode='absolute').plot(x='date', y='close_price', kind='scatter')
        ```
        ## Documentation
        The official documentation is hosted at https://python-client-docs.dataservices.btgpactualsolutions.com/
Platform: UNKNOWN
Requires-Python: >=3.7
Description-Content-Type: text/markdown
