Skip to content

Latest commit

 

History

History
299 lines (211 loc) · 8.13 KB

basic_usage.rst

File metadata and controls

299 lines (211 loc) · 8.13 KB

Basic Usage

There are three levels in the dxfeed package. The lowest is the C API library, the highest is Python wrapper classes. Cython level in the middle aims to connect these two. Here we are going to look into Python level.

Python level, in its turn, mainly consists of three class types:

  1. Endpoint
  2. Subscription
  3. EventHandler

The Endpoint is responsible for connection management and creating dependent classes, for example Subscription. One Endpoint may have several different Subscriptions, but each Subscription is related to one Endpoint.

Subscription class sets the type of subscription (stream or timed), the type of events (e.g. Trade, Candle), etc.

After you specified the data you want to receive, you have to specify how to process upcoming events. This is where the EventHandler class and its children come into play. Every time an event arrives Cython event listener will call self.update(event) method. You have to inherit from the EventHandler class and redefine the update method. Or you may use DefaultHandler which stores upcoming data in deque of the length 100k.

Import package

import dxfeed as dx
from datetime import datetime  # for timed subscription

Configure and create connection with Endpoint class

Create instance of Endpoint class which will connect provided address.

endpoint = dx.Endpoint('demo.dxfeed.com:7300')

Endpoint instance contains information about the connection, e.g. connection address or status

print(f'Connected address: {endpoint.address}')
print(f'Connection status: {endpoint.connection_status}')
Connected address: demo.dxfeed.com:7300
Connection status: Connected and authorized

Configure and create subscription

You should specify event type. For timed subscription (conflated stream) you should also provide time to start subscription from.

trade_sub = endpoint.create_subscription('Trade')

Set event handler - class that process incoming events. Here we use default one

trade_handler = dx.DefaultHandler()
trade_sub.set_event_handler(trade_handler);

Add tikers you want to recieve events for

trade_sub = trade_sub.add_symbols(['C', 'IBM'])

For timed subscription you should provide either datetime object or string. String might be incomlete, in this case you will get warning with how your provided date parsed automatically. For Candle event type along with base symbol, you should specify an aggregation period. You can also set price type. More details: https://kb.dxfeed.com/display/DS/REST+API#RESTAPI-Candlesymbols.

tns_sub = endpoint.create_subscription('TimeAndSale', date_time=datetime.now()) \
                  .add_symbols(['AMZN'])
candle_sub = endpoint.create_subscription('Candle', date_time='2020-04-16 13:05')
candle_sub = candle_sub.add_symbols(['AAPL{=d}', 'MSFT{=d}'])
c:\job\python-api\dxfeed\wrappers\class_utils.py:38: UserWarning: Datetime argument does not exactly match %Y-%m-%d %H:%M:%S.%f format, date was parsed automatically as 2020-04-16 13:05:00.000000
  warn(warn_message, UserWarning)

Note Two previous subscriptions attached DefaultHandler implicitly. To retrieve instances just call get_event_handler() method.

tns_handler = tns_sub.get_event_handler()
candle_handler = candle_sub.get_event_handler()

Subscription instance properties

print(f'TimeAndSale subscription event type: {tns_sub.event_type}')
print(f'Candle subscription event type: {candle_sub.event_type}')
print(f'Candle subscription symbols: {candle_sub.symbols}')
TimeAndSale subscription event type: TimeAndSale
Candle subscription event type: Candle
Candle subscription symbols: ['AAPL{=d}', 'MSFT{=d}']

Access data from DefaultHandler instance

You can get colums, list or dataframe. You are also allowed to write handler that stores no data.

print(f'Trade columns: {trade_handler.columns}')
print(f'Candle columns: {candle_handler.columns}')
Trade columns: ['Symbol', 'Price', 'ExchangeCode', 'Size', 'Tick', 'Change', 'DayVolume', 'Time', 'IsETH']
Candle columns: ['Symbol', 'Index', 'Time', 'Sequence', 'Count', 'Open', 'High', 'Low', 'Close', 'Volume', 'VWap', 'BidVolume', 'AskVolume', 'OpenInterest', 'ImpVolatility']

candle_handler.get_list()[-5:]
[['MSFT{=d}', 6816463568083353600, 1587081600000, 0, 189986.0, 179.5, 180.0, 175.87, 178.6, 52765625.0, 177.90622, 24188832.0, 22094602.0, 0, 0.4384],
 ['MSFT{=d}', 6816294775868620800, 1587042300000, 0, 189986.0, 179.5, 180.0, 175.87, 178.6, 52765625.0, 177.90622, 24188832.0, 22094602.0, 0, 0.4384],
 ['AAPL{=d}', 6839841934068940800, 1592524800000, 0, 827.0, 354.05, 355.55, 353.35, 354.72, 188804.0, 354.45941, 78039.0, 110765.0, 0, 0.3691],
 ['AAPL{=d}', 6839841934068940800, 1592524800000, 0, 831.0, 354.05, 355.55, 353.35, 354.9, 189555.0, 354.4611, 78039.0, 111516.0, 0, 0.3691],
 ['AAPL{=d}', 6839841934068940800, 1592524800000, 0, 832.0, 354.05, 355.55, 353.35, 354.72, 190055.0, 354.46178, 78539.0, 111516.0, 0, 0.3691]]

candle_handler.get_dataframe().head(3)
.dataframe tbody tr th:only-of-type { vertical-align: middle; } .dataframe tbody tr th { vertical-align: top; } .dataframe thead th { text-align: right; }
Symbol Index Time Sequence Count Open High Low Close Volume VWap BidVolume AskVolume OpenInterest ImpVolatility
0 AAPL{=d} 6839841934068940800 2020-06-19 0 827.0 354.05 355.55 353.35 354.72 188804.0 354.45941 78039.0 110765.0 0 0.3691
1 AAPL{=d} 6839470848894566400 2020-06-18 0 96172.0 351.41 353.45 349.22 351.73 24205096.0 351.56873 8565421.0 10394906.0 0 0.3673
2 AAPL{=d} 6839099763720192000 2020-06-17 0 110438.0 355.15 355.40 351.09 351.59 28601626.0 353.70998 10686232.0 12141490.0 0 0.3713

Close subscription

trade_sub.close_subscription()
tns_sub.close_subscription()
candle_sub.close_subscription()

Close connection

endpoint.close_connection()
print(f'Connection status: {endpoint.connection_status}')
Connection status: Not connected