This library provides tools that help develop code that access the 3commas api very fast and easy. The library is built on top of the py3cw library. Main features are prebuilt functions for the api, models for easier access of the returned data, automatic attribute parsing of the returned data, built in error parsing.
WIP: this library is still in construction. Some endpoints implementations are missing. Feel free to create a PR in github if you desire a particular endpoint implementation https://github.com/badass-blockchain/python-three-commas
pip install three-commas
The package is built mirroring the names of the api paths. For example:
from three_commas import api
api.ver1.bots # bots endpoint
api.ver1.accounts # account endpoint
api.ver1.deals # deals endpoint
api.ver1.users # users endpoint
api.v2.smart_trades # v2 smart_trades endpoint
The endpoints are mirrored from the paths too.
# GET /ver1/bots/pairs_black_list
error, black_list_pairs = api.ver1.bots.get_pairs_black_list()
# GET /v2/smart_trades/{id}
error, smart_trades_list = api.v2.smart_trades.get_by_id(id=<your_smart_trade_id>)
You can get all bots with:
error, bots_list = api.ver1.bots.get()
Or a single bot with:
error, bot = api.ver1.bots.get_show_by_id(bot_id=<your_bot_id>)
Or a smart trade
error, smart_trade = api.v2.smart_trades.get_by_id(id=9993000)
The endpoints return a dict object with added functionality. You can use the object like a normal dictionary (exactly how you receive from py3cw), or use the added functions. For example if you want to get the bot max_active_deals you can do both:
error, bot = api.ver1.bots.get_show_by_id(bot_id=9999999)
if not error:
max_active_deals = bot['max_active_deals']
max_active_deals = bot.max_active_deals
You can easily connect to the websockets You can use annotations.
import three_commas
from three_commas.model import DealEntity, SmartTradeV2Entity
@three_commas.streams.smart_trades
def handle_smart_trades(smart_trade: SmartTradeV2Entity):
# Do here something with the smart trade
# Every new smart trade is passed to this function
print(smart_trade)
@three_commas.streams.deals
def handle_deals(deal: DealEntity):
# do your awesome stuff with the deal
print(deal) # {'id': 1311811868, 'type': 'Deal', 'bot_id': 6313165, 'max_safety_orders': 6, 'deal_has_error': False ....
print(deal.account_id) # 99648312
print(deal.created_at) # string object '2022-02-18T05:26:06.803Z'
print(deal.parsed(True).created_at) # datetime.datetime object
In order to use the websocket streams you need to set the api key and secret in your environment. [Later in the document you can find how to set up the environment variables](#Set the api key and secret)
Or you can pass the keys to the decorator:
@three_commas.streams.deals(api_key='<your_api_key>', secret='<your_secret>')
def handle_deals(deal):
...
For debugging you can turn on debug level
import logging
logging.getLogger('three_commas.streams').setLevel(level=logging.DEBUG)
You will see a lot of websocket messages including pings:
> DEBUG:three_commas.streams.streams: {"type": "welcome"}
> DEBUG:three_commas.streams.streams: {"type": "ping", "message": 1645286932}
> DEBUG:three_commas.streams.streams: {"type": "ping", "message": 1645286935}
> DEBUG:three_commas.streams.streams: {"type": "ping", "message": 1645286938}
You can also set up the level to info for less verbose loging logging.getLogger('three_commas.streams').setLevel(level=logging.INFO)
One of the features of this library is the automatic parsing of the returned data. Some numeric data fetched from the api is returned as string. For example in the bot object:
...
"base_order_volume": "0.003",
"safety_order_volume": "0.003",
"safety_order_step_percentage": "1.0",
...
Now you do not need to bother checking the type of the field and parsing it into the desired type. This library auto parses these fields:
error, bot = api.ver1.bots.get_show(9999999)
# base_order_volume is a float
base_order_volume = bot.get_base_order_volume()
Parsing (except datetime fields) is done by default. If you do not want the field to be parsed, and you want the original string to be returned use parsed=False
error, bot = api.ver1.bots.get_show(9999999)
# base_order_volume is a str
base_order_volume = bot.parsed(False).base_order_volume
Some fields like "created_at" are timestamps. You can parse these fields to a python datetime object. Timestamp fields are NOT parsed by default, only on demand:
error, account = api.ver1.accounts.get_by_id(8888888)
# the original string returned by the api
created_at_str = account.created_at
# parsed into a datetime.datetime object
created_at_datetime = account.parsed(True).created_at
In order to use the api you need to set the api key and secret. This could be done globally or per request. To set it globally you need to set the environment variables THREE_COMMAS_API_KEY and THREE_COMMAS_API_SECRET. To do it per request just pass them into the function:
error, account = api.ver1.accounts.get_by_id(8888888, api_key='my_key', api_secret='my_secret')
Request keys have priority. If both global and request keys are set, then the request keys will be used.
You can set the forced mode globally or also per request. To set it globally set the environment variable THREE_COMMAS_FORCED_MODE to either paper or real. To use the forced mode per request pass it as an argument:
error, paper_deals = api.ver1.deals.get(forced_mode='paper')
error, real_deals = api.ver1.deals.get(forced_mode='real')
Some enum fields have functionality.
error, accounts_list = api_v1.accounts.get_accounts()
if not error:
for account in accounts_list:
if account.get_market_code().is_binance():
# do stuff with the binance account.
else:
# deal with error here
You can check what enums are available in the three_commas.model.generated_enums package
Always make sure your api keys are stored securely and are not pushed into any public repositories
You can set key and secret as an environment variables. The preferred more secure way is to use a .env file:
create a file called ".env" with the following content:
THREE_COMMAS_API_KEY=<your_api_key>
THREE_COMMAS_API_SECRET=<your_api_secret>
Use a .env loader as python-dotenv. Install it on your machine with pip: "pip install python-dotenv". If your project is a git project make sure you gitignore the .env file. Then in your code
from dotenv import load_dotenv
load_dotenv()
Now your variables are loaded. Enjoy using the library.
You can also set the api key and secret directly in the code. This method is less secure and not recommended.
import os
os.environ["THREE_COMMAS_API_KEY"] = <your_api_key>
os.environ["THREE_COMMAS_API_SECRET"] = <your_api_secret>
smart_trade = {
'account_id': 99999999,
'pair': 'USDT_FUN',
'position': {
'type': 'buy',
'order_type': 'market',
'units': {
'value': "30526.0",
},
"total": {
"value": "600.61907506"
}
},
'take_profit': {
'enabled': False,
},
'stop_loss': {
'enabled': False,
}
}
error, smart_trade_response = api.v2.smart_trades.post(smart_trade)
error, smart_trade = api.v2.smart_trades.get_by_id(13819196)
if not error:
# Do your awesome stuff with the smart trade
print(smart_trade.profit)