Skip to main content

Manage coroutine execution speed

Project description

Execute coroutines with limitations. Take a look at following examples.

Uniform and steady execution

Let’s imagine we have some service. We want to get something like 10k results from that server. While we don’t want to cost trouble target server we can use Pulemet with rps parameter. It will run just rps requests per second. And we won’t damage server. Also, we may need to use pool_size parameter if server doesn’t answer fast enough. That parameter will prevent creating new connections above the limit if current still working.

import asyncio

   from pulemet.pulemet import Pulemet


   async def http_request(t: float = 0):
       """Let's say we go somewhere by http here."""
       await asyncio.sleep(t)
       return 1


   async def sum_results(coros):
       s = 0
       for elem in asyncio.as_completed(coros):
           s += await elem

       return s


   pulemet = Pulemet(rps=1, pool_size=20)
   coroutines = pulemet.process([http_request() for _ in range(10)])

   result = asyncio.get_event_loop().run_until_complete(sum_results(coroutines))

Functions and retries

You can run some async function with list of arguments and catch certain exceptions and even try call it again(few times). All of these in following example.

import asyncio

from pulemet.pulemet import Pulemet


async def func(ind):
    await asyncio.sleep(0.001)
    if ind % 2 == 0:
        raise ValueError
    return ind


def main():
    pulemet = Pulemet(rps=10)

    coros_pulemet = pulemet.process_funcs(
        coro_func=func,
        coros_kwargs=({'ind': i} for i in range(0, 20)),
        exceptions=(ValueError,),
        exceptions_max=5,
    )
    coroutines = asyncio.gather(*coros_pulemet, return_exceptions=True)

    asyncio.get_event_loop().run_until_complete(coroutines)


if __name__ == '__main__':
main()

Progress Bar Integration

That example explain how you can see execution progress this tqdm.

import asyncio

from tqdm.auto import tqdm

from pulemet.pulemet import Pulemet


async def target(t: float = 0):
    await asyncio.sleep(t)
    return 1


async def sum_results(coros):
    s = 0
    for elem in asyncio.as_completed(coros):
        s += await elem

    return s


pulemet = Pulemet(rps=1, pbar=tqdm)
coroutines = pulemet.process([target() for _ in range(10)])

result = asyncio.get_event_loop().run_until_complete(sum_results(coroutines))

You will see something like that.

Total: 0it [00:00, ?it/s]
Per second: 0it [00:00, ?it/s]

Total:   0%|          | 0/10 [00:00<?, ?it/s]
Total:  20%|██        | 2/10 [00:01<00:04,  1.99it/s]
Total:  30%|███       | 3/10 [00:02<00:04,  1.40it/s]
Total:  40%|████      | 4/10 [00:03<00:04,  1.22it/s]
Total:  50%|█████     | 5/10 [00:04<00:04,  1.13it/s]
Total:  60%|██████    | 6/10 [00:05<00:03,  1.08it/s]
Total:  70%|███████   | 7/10 [00:06<00:02,  1.05it/s]
Total:  80%|████████  | 8/10 [00:07<00:01,  1.04it/s]
Total:  90%|█████████ | 9/10 [00:08<00:00,  1.02it/s]
Total: 100%|██████████| 10/10 [00:09<00:00,  1.02it/s]

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

pulemet-1.0.1a1.tar.gz (4.8 kB view hashes)

Uploaded Source

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page