fastlob 0.0.21

Fast & minimalist limit-order-book (LOB) implementation in pure Python.

fastlob | Python Limit-Order-Book

Fast & minimalist limit-order-book (LOB) implementation in Python, with almost no dependencies.

Package currently in development, bugs are expected.

---

I know that it does not make sense to call "fast" a single-threaded order-book implementation written in an interpreted language such as Python. And, in fact, this project is not fast at all yet.

This is just the very first version of the project, the idea was to first have a working and clean pure-Python version. The next step is to rewrite the core order processing parts in concurrent C/C++. This will be done during summer 2025 on a separate branch and then merged to the main branch.

---

Functionalities:

• Place limit orders.
• Execute market orders.
• Orders can be good-till-cancel (GTC), fill-or-kill (FOK) or good-till-date (GTD).
• Cancel pending or partially filled orders.
• Query order status (pending, filled, partially filled, canceled...).
• Set custom tick size for price and quantities.
• Extract spread, midprice, volume, etc.
• Simulate on historical data.

The goal is to build an efficient and easy to use package, with a clean and comprehensible API.

We aim to keep it minimalist and simple, while keeping reasonable performances (for a pure Python implementation). We intend the final project to contain no more than ~1000 lines of code.

We implement three types of orders: FOK, GTC and GTD. Every order is initially defined as limit, but will be executed as a market order if its price matches the best (bid or ask) limit price in the book.

In the case of GTD orders, the book only supports whole seconds for the order expiry (order can not be set to expire in 3.8 seconds, in this case it will be rounded to 4, nearest integer).

We do not implement multi-threaded order processing yet, check out the corresponding issue for more infos.

Installation

The package is available on PyPI, you can simply install it using

pip install fastlob

Otherwise, one can install it from source

git clone [e-mail]:mrochk/fastlob.git
cd fastlob
pip install -r requirements.txt
pip install .

Testing

To run the tests and check that everything is okay, run make test or python3 -m unittest discover test.

Usage

This book runs at a fixed decimal precision through the Python decimal package. The decimal precision (also called tick size) can be set via the FASTLOB_DECIMAL_PRECISION_PRICE and FASTLOB_DECIMAL_PRECISION_QTY environment variables, if not set it defaults to 2.

# examples/basics.py

import time, logging

from fastlob import Orderbook, OrderParams, OrderSide, OrderType

logging.basicConfig(level=logging.INFO) # maximum logging

lob = Orderbook(name='MYLOB', start=True) # init and start lob

# every order must be created this way 
params = OrderParams(
    side=OrderSide.BID, # is it a buy or sell order
    price=123.32, quantity=3.42, # by default runs at 2 digits decimal precision
    otype=OrderType.GTD, # good-till-date order
    expiry=time.time() + 120 # order will expire in two minutes
    # since order is GTD, expiry must be set to some future timestamp
)

# -> at this point an exception will be raised if invalid attributes are provided

result = lob(params) # let the book process the order
assert result.success() # result object can be used to see various infos about the order execution

# order uuid is used to query our order after it's been placed
status, quantity_left = lob.get_order_status(result.orderid())
print(f'Current order status: {status.name}, quantity left: {quantity_left}.\n')

lob.render() # pretty-print the book

lob.stop() # stop the background processes

In the notebook simulate you will find an example of simulating the arrival of many orders using different distributions.

In load_snapshot_updates you'll find an example of running the lob on historical data.

For more information and examples check out https://fastlob.com.

Contributing

As mentioned earlier, this package is still in early development, and contributions are more than welcome.

Please do not hesitate to contact me or directly submit a pull request if you'd like to contribute, there are also various issues open on Github.