BitEx

BitEx is a collection of API Clients for Crypto Currency Exchanges.

It comes with two parts - bitex.api represents the base level API interfaces, on top of which the second part - bitex.interfaces - builds upon. bitex.api classes can be used without making use of the interface classes.

Donations welcome! BTC @ 3D4yuyf84eQUauyZLoQKyouPuThoxMMRZa

State

RESTAPI : Completed

WSSAPI : BETA

Interfaces : WIP

Supported Exchanges

Additional clients will be added to (or removed from) this list, according to their liquidity and market volume.

[^1]: This table considers standardized methods only, when describing the state. See section Standardized Methods for more

bitex.api.REST

Classes found in bitex.api.REST provide wrapper classes and methods for Python's requests module, including handling of each exchange's specific authentication procedure.

An example:

from bitex.api.REST import KrakenREST

k = KrakenREST()
k.load_key('kraken.key')  # loads key and secret from given file;

# Query a public endpoint
k.query('GET','public/Depth', params={'pair': 'XXBTZUSD'})

# Query a private (authenticated) endpoint
q = {'pair': 'XXBTZUSD', 'type': 'sell', 'ordertype': 'limit', 'price': 1000.0,
     'volume': 0.01, 'validate': True}
k.query('POST','private/AddOrder', authenticate=True, params=q)

Example .key file:

>>>dummy.key
my_api_key
my_fancy_api_secret

If the api requires further details, for example a userid or account number (for example for bitstamp), you should check the class method's doc string, although usually this information needs to go after the api key and secret, on a separate line each.

>>>dummy2.key
my_api_key
my_fancy_api_secret
Userid
accountname

bitex.api.WSS

bitex.api.WSS offers Queue()-based Websocket interface for a select few exchanges. The classes found within are very basic, and subject to further development. Private endpoints and trading are only sporadically implemented.

Their prime objective is to provide a raw, realtime interface to all of an exchange's Websocket endpoint.

Usage

from.bitex.api.WSS import GeminiWSS
import time

wss = GeminiWSS()
wss.start()
time.sleep(5)
wss.stop()

while not wss.data_q.empty():
    print(wss.data_q.get())
    

You can of course also access data_q while the WebSocket is still running (i.e. before calling stop()).

bitex.interfaces

Built on top of bitex.api's api classes are the slightly more sophisticated exchange interfaces in bitex.interfaces. These have been written to unify the diverse REST APIs of the implemented exchanges, by providing the same methods and method parameters across all of them (see next section, Standardized Methods, for more information).

For example, querying tickers looks the same on all exchanges, as well as placing an order, using bitex.interface:

from bitex import Kraken, Bitstamp, Gemini
k = Kraken(key_file='krkn.key')
b = Bitstamp(key_file='btst.key')
g = Gemini(key_file='gmni.key')

k.ticker('XBTUSD')
b.ticker('btceur')
g.ticker('BTC-USD')

k.ask(pair, price, size)
b.ask(pair, price, size)
g.ask(pair, price, size)

Standardized Methods

As explained in the previous section, standardized methods refer to the methods of each interface which have been deemed as part of the set of minimal methods and functions required to trade at an exchange via its API. They feature the following characteristics:

• Each method has an identical method header across all interfaces
• Its output is identical across all interfaces
• Each method returns a bitex.api.response.APIResponse object; these behave like requests.Request objects, with the addition of a new attribute, formatted, which stores a standardized representation of the data queried.

bitex.formatters

This module provide formatters for the standardized methods, formatting their json output into a uniform layout. They are a work in progress feature.

Be mindful that, in order to provide a unified output format, some fields have been dropped in the formatted output! If you rely on one of these dropped fields, be sure to use the APIResponse's json attribute instead, and parse the json yourself:

from bitex import Kraken
k = Kraken()
response = k.ticker()
print(response.formatted)  # show formatted data
print(response.json())  # Returns all json data

The following is a table of all formatters currently implemented - any method not marked as Done will not do any formatting.

Installation

Manually, using the supplied setup.py file:

python3 setup.py install

or via pip

pip install BitEx