An Asyncio based, high performance, REST client libary for interacting with the polygon REST api.

The two main python rest-client libs for polygon.io(alpaca-trade-api, polygonio) do not provide an effective means to gather more than 50,000 trade bars at once. This library aims to address that by providing an easy and performant solution to getting results from timespans where the resultset exceeds 50,000 trade bars.

This library makes use of some high performance packages written in C/Rust (uvloop, orjson) so it may require sudo apt install python3-dev on Ubuntu or similar on other OS's. It is currently only compatible with Python 3.8.x but aims to be compatible with 3.9 once it's dependendencies support 3.9 and all future CPython versions moving forward.

pip/poetry w/ venv

#!/bin/env bash
python3.8 -m venv .venv && source .venv/bin/activate

poetry add upoly
# or
pip install upoly

Reccomend to create a copy of ./env.sample as ./env. Make sure .env is listed in .gitignore.

# ./.env
POLYGON_KEY_ID=REPACEWITHPOLYGONORALPACAKEYHERE

Many alternatives to .env exist. One such alternative is exporting like so:

#!/bin/env bash
export POLYGON_KEY_ID=REPACEWITHPOLYGONORALPACAKEYHERE

or adding to your shell startup script, either .zshrc or .bashrc to have it be globally available to all projects.

#/home/youruseraccount/.bashrc
...
export POLYGON_KEY_ID=REPACEWITHPOLYGONORALPACAKEYHERE
...

# ./yourscript.py
import pytz
from dotenv import load_dotenv
import pandas as pd

# load Polygon key from .env file
load_dotenv()
# alternatively run from cli with:
# POLYGON_KEY_ID=@#*$sdfasd python yourscript.py

# Not recommend but can be set with os.environ["POLYGON_KEY_ID"] as well

from upoly import async_polygon_aggs


NY = pytz.timezone("America/New_York")

# Must be a NY, pandas Timestamp
start = pd.Timestamp("2015-01-01", tz=NY)
end = pd.Timestamp("2020-01-01", tz=NY)

df = async_polygon_aggs("AAPL", start, end)

• unit tests
• regression tests
• integration tests
• /trade endpoint functionality for tick data