many_requests

Dead easy interface for executing many HTTP requests asynchronously. It has been tested in the wild with over 10 million requests. Automatically handles errors and executes retries.

Built on-top of Trio and asks. Interface heavily inspired by Requests and joblib.

Also provides helper functions for executing embarrassingly parallel async coroutines.

To install:

pip install many-requests

Example Usage

Execute 10 GET requests for example.org:

from many_requests import ManyRequests
responses = ManyRequests(n_workers=5, n_connections=5)(
                method='GET',
                url=['https://example.org' for i in range(10)])

Query HackNews API for 10 items and parse JSON output:

responses = ManyRequests(n_workers=5, n_connections=5, json=True)(
                method='GET',
                url=[f'https://hacker-news.firebaseio.com/v0/item/{i}.json?print=pretty' for i in range(10)])

To use basic authentication with all requests:

from asks import BasicAuth
username = 'user'
password = 'pw'
responses = ManyRequests(n_workers=5, n_connections=5)(
                method='GET',
                url=['https://example.org' for i in range(10)],
                auth=BasicAuth((username, password)))

To execute embarrassingly parallel async coroutines, for example 10 trio.sleep calls:

from many_requests import EasyAsync, delayed
import trio
outputs = EasyAsync(n_workers = 4)(delayed(trio.sleep)(i) for i in range(10))