Despite multiple efforts, I couldn't figure out how to evaluate generators in parallel. Ideas are welcome.

parallel_generators.py

def get_repo_stars(org, repo):
    url = 'https://api.github.com/repos/{org}/{repo}'.format(
        org=org, repo=repo)
    print("GET ", url)
    resp = requests.get(url)
    return resp.json()['stargazers_count']

params = [
    ('requests', 'requests'),
    ('requests', 'httpbin'),
    ('django', 'django'),
    ('Lukasa', 'hyper'),
]

generator = (get_repo_stars(org, repo) for org, repo in params)

def parallel(generator):
    def _next():
        try:
            val = next(generator)
            print(val)
        except StopIteration:
            pass
    threads = [threading.Thread(target=_next) for _ in range(2)]
    [t.start() for t in threads]
    [t.join() for t in threads]
    

parallel(generator)

from concurrent import futures

import requests


def get_repo_stars(org, repo):
    url = 'https://api.github.com/repos/{org}/{repo}'.format(
        org=org, repo=repo)
    resp = requests.get(url)
    return (repo, resp.json()['stargazers_count'])


params = [
    ('requests', 'requests'),
    ('requests', 'httpbin'),
    ('django', 'django'),
    ('Lukasa', 'hyper'),
]

with futures.ThreadPoolExecutor() as ex:
    results = ex.map(lambda p: get_repo_stars(*p), params)
    for repo, stars in results:
        print(f'{repo} has {stars} stars')

Hey, thanks! Yes, I know about map of concurrent.futures and multiprocessing.Pool. I was just trying to see if it was possible to make generators parallel.

Generators are always synchronous, but you can make the request async (and therefore do it concurrently).