Ape-based SDK for working with deployments of Uniswap protocol
- python3 version 3.10 or greater, python3-dev
You can install the latest release via pip:
pip install uniswap_sdkYou can clone the repository and use setuptools for the most up-to-date version:
git clone https://github.com/SilverBackLtd/uniswap-sdk.git
cd uniswap-sdk
python3 setup.py installThe SDK can be used for any scripting task very easily:
>>> from ape_tokens import tokens
>>> from uniswap_sdk import Uniswap
>>> uni = Uniswap(use_v3=False) # Can skip versions and only index certain tokens
>>> list(uni.index(tokens=tokens)) # Takes time, but makes planning faster (recommended for scripting)
>>> uni.price("UNI", "USDC") # Get liquidity-weighted prices of entire index in real-time
Decimal("4.75")
>>> usdc = tokens["USDC"]
>>> tx = uni.swap(
... "UNI",
... usdc, # Can use any ContractInstance type
... amount_in="12 UNI", # Uses Ape's conversion system
... slippage=0.3,
... deadline=timedelta(minutes=2),
... sender=trader,
... )This SDK installs a special CLI command uni.
You can use this command to do common tasks with the SDK such as finding prices or performing swaps.
Try uni --help after installing the SDK to learn more about what the CLI can do.
The SDK has special support for use within Silverback bots, which takes advantage of real-time processing to drastically reduce the overhead of certain search and solver functions of the SDK:
from ape_tokens import tokens
from silverback import SilverbackBot
from uniswap_sdk import Uniswap
bot = SilverbackBot()
uni = Uniswap()
uni.install(bot) # This replaces having to do `uni.index()`
# NOTE: The bot will now process all swaps in the background to keep it's indexes up-to-date!
@bot.cron("* * * * *")
async def weth_price(t):
# So now when you use top-level functions, it takes advantage of cached data in the index
return uni.price("WETH", "USDC") # This executes faster w/ Silverback!The SDK comes with a default Solver that should be performant enough for most situations.
However, it is likely that you will want to design a custom solver function or class in order
to obtain better results when performing actions like uni.swap which leverage the solver.
You can override the default solver by providing a function or object which matches the following interface:
from uniswap_sdk import Order
Route = tuple[PairType, ...] # 1 (or more) `PairType`s (e.g. `UniswapV2Pair`, etc.)
Solution = dict[Route, Decimal] # mapping of Route -> amount to swap via Route
SolverType = Callable[[Order, Iterable[Route]], Solution]
# Given `amount` of `token` and `*routes`, find `solution`This can be a class, allowing more flexibility in how you design your solver:
class Solver:
def __call__(self, order: Order, routes: Iterable[Route]) -> Solution:
# This function must match `SolverType` to work
my_solver = Solver(...)
uni = Uniswap(use_solver=my_solver)
uni.solve(...) # Will now use `my_solver` to find solutions (also `uni.swap`)This project is in development and should be considered a beta. Things might not be in their final state and breaking changes may occur. Comments, questions, criticisms and pull requests are welcomed.
Support for various Uniswap-related protocols:
- V1
- V2
- V3
- V4
- Permit2
- UniversalRouter
This project is licensed under the Apache 2.0.