Subscribe to Public channels

Initialise client

By default all incoming messages are being logged at INFO level without any modifications. So you would want to set logging accordingly in case you need to visually inspect results.

import time
import logging
from pprint import pprint

from bcx.client import BlockchainWebsocketClient

logging.basicConfig(level=logging.INFO)

client = BlockchainWebsocketClient()

Subscribe for monitoring

Subscription can happen on a fly, i.e. you don’t need to restart client (or re-run script) to be able to subscribe to new channels. This quite handy if you work in interactive environment, e.g. Jupyter Lab (remember that all logged information will be stored within a notebook itself, and there can be a lot of messages depending on channel activity)

client.subscribe_to_heartbeat()

client.subscribe_to_prices("BTC-USD", granularity=60)

client.subscribe_to_prices("ETH-USD", granularity=60)

client.subscribe_to_ticker("BTC-USD")

client.subscribe_to_trades("BTC-USD")

time.sleep(2)
pprint(client.connected_channels)

Out:

[HeartbeatChannel(is_subscribed=True),
 PricesChannel(symbol=BTC-USD, granularity=60, is_subscribed=True),
 PricesChannel(symbol=ETH-USD, granularity=60, is_subscribed=True),
 TickerChannel(symbol=BTC-USD, is_subscribed=True),
 TradesChannel(symbol=BTC-USD, is_subscribed=True)]

Some channels send a snapshot of when you connect to them for the very first time.

time.sleep(2)
client.subscribe_to_symbols()

client.subscribe_to_orderbook_l2("BTC-USD")

client.subscribe_to_orderbook_l3("BTC-USD")

time.sleep(2)
pprint(client.connected_channels)

Out:

[HeartbeatChannel(is_subscribed=True),
 OrderbookL2Channel(symbol=BTC-USD, is_subscribed=True),
 OrderbookL3Channel(symbol=BTC-USD, is_subscribed=True),
 PricesChannel(symbol=BTC-USD, granularity=60, is_subscribed=True),
 PricesChannel(symbol=ETH-USD, granularity=60, is_subscribed=True),
 SymbolsChannel(is_subscribed=True),
 TickerChannel(symbol=BTC-USD, is_subscribed=True),
 TradesChannel(symbol=BTC-USD, is_subscribed=True)]

This is a simple use case showing that the client not only reads incoming messages and logs them, but also can parse them and store the result in memory in order to be used later on, i.e. by external API.

time.sleep(7)
print(f"Last Heart Beat: {client.get_last_heartbeat()}")

Out:

Last Heart Beat: 2020-05-17 22:56:29.039731