Skip to main content

Understanding AsyncIO by Examples

·584 words·3 mins
Posts async http python
Table of Contents
AsyncIO - This article is part of a series.
Part 2: This Article
Learn AsyncIO by example, including aiothttp for concurrent network connections.

In learning AsyncIO by code, we explain how to use async/await in asyncio.

Here, we will learn AsyncIO by 3 examples, from basic to more practical use (aiohttp).

Example_1
#

In this example, task1 and task2 start almost the same time. And you will see task1 completes after 2 sec, and task 2 completes after 5 sec.

# Example 1 - Demo asyncio with 2 tasks.
import asyncio

async def fetch_data():
    print('# [t1] start fetching....')
    await asyncio.sleep(2)
    print('# [t1] done fetching.')
    return {'data': 1}

async def print_numbers():
    for i in range(10):
        print(f'# [t2] {i}')
        await asyncio.sleep(0.5)

async def main():
    task1 = asyncio.create_task(fetch_data())
    task2 = asyncio.create_task(print_numbers())

    value = await task1
    print(f'# [t1] {value}')
    await task2

asyncio.run(main())

# Output
# [t1] start fetching....
# [t2] 0
# [t2] 1
# [t2] 2
# [t2] 3
# [t1] done fetching.
# [t1] {'data': 1}
# [t2] 4
# [t2] 5
# [t2] 6
# [t2] 7
# [t2] 8
# [t2] 9

Example_2
#

In this example, it demos the completion of 3 tasks. Each task will calculate and find the highest prime number. And we start the first task by 100000 and follow by 10000 and 1000.

# Example 2 - Demo asyncio with 3 tasks.
import time
import asyncio

def is_prime(x):
    return not any(x//i == x/i for i in range(x-1, 1, -1))

async def highest_prime_below(x):
    print(f'Highest prime below {x}')
    for y in range(x-1, 0, -1):
        if is_prime(y):
            print(f'--> Highest prime below {x} is {y}.')
            return y
        await asyncio.sleep(0.01)
    return None

async def main():
    p = [100000, 10000, 1000]
    tasks = []
    t0 = time.time()

    for x in p:
        task = asyncio.create_task(highest_prime_below(x))
        tasks.append(task)

    await asyncio.gather(*tasks)

    t1 = time.time()
    print(f'Took {1000*(t1-t0):.2f} ms')

loop = asyncio.get_event_loop()
loop.run_until_complete(main())
#loop.close()

# Output
# Highest prime below 100000
# Highest prime below 10000
# Highest prime below 1000
# --> Highest prime below 1000 is 997.
# --> Highest prime below 100000 is 99991.
# --> Highest prime below 10000 is 9973.
# Took 311.18 ms

Example_3 (aiohttp)
#

In this example, it demos the completion of 5 network connections (tasks). It will trace the elapsed time taken for individual URL. Some tasks complete faster and some are slower.

# Example 3 - Demo asyncio with 5 tasks [network connections] .
import aiohttp
import asyncio
from datetime import datetime

async def fetch_with(client, url):
    async with client.get(url) as resp:
        await resp.text()
        return f'# [{datetime.now():%X}] - {url} [{resp.status}]'

async def fetch_all(client, urls):
    tasks = []
    for url in urls:
        task = asyncio.create_task(fetch_with(client,url))
        print(f'# [{datetime.now():%X}] : Added [{url}] into tasklist.')
        tasks.append(task)

    print(f'')
    results = await asyncio.gather(*tasks)
    return results

async def on_request_start(session, trace_config_ctx, params):
    trace_config_ctx.start = asyncio.get_event_loop().time()

async def on_request_end(session, trace_config_ctx, params):
    elapsed = asyncio.get_event_loop().time() - trace_config_ctx.start
    print(f'# [{elapsed:.5f}s] : {params.url}')

async def main(urls):
    trace_config = aiohttp.TraceConfig()
    trace_config.on_request_start.append(on_request_start)
    trace_config.on_request_end.append(on_request_end)
    async with aiohttp.ClientSession(trace_configs=[trace_config]) as client:
        data = await fetch_all(client, urls)
        return data

if __name__ == "__main__":
    urls = [
        'https://securityheaders.com/',
        'https://hstspreload.org/',
        'https://www.isc2.org/',
        'https://www.sans.org/',
        'https://scotthelme.co.uk/'
    ]

    results = asyncio.run(main(urls))
    print(f'')
    for r in results:
        print(r)

# Output
# [21:38:50] : Added [https://securityheaders.com/] into tasklist.
# [21:38:50] : Added [https://hstspreload.org/] into tasklist.
# [21:38:50] : Added [https://www.isc2.org/] into tasklist.
# [21:38:50] : Added [https://www.sans.org/] into tasklist.
# [21:38:50] : Added [https://scotthelme.co.uk/] into tasklist.

# [0.12875s] : https://scotthelme.co.uk/
# [0.93167s] : https://hstspreload.org/
# [1.08564s] : https://www.isc2.org/
# [1.39669s] : https://securityheaders.com/
# [2.20264s] : https://www.sans.org/apac/

# [21:38:51] - https://securityheaders.com/ [200]
# [21:38:50] - https://hstspreload.org/ [200]
# [21:38:51] - https://www.isc2.org/ [200]
# [21:38:53] - https://www.sans.org/ [200]
# [21:38:50] - https://scotthelme.co.uk/ [200]
AsyncIO - This article is part of a series.
Part 2: This Article

Related

Insecurity in HTTP Headers
·2195 words·11 mins
Posts Essential async cli http python
Based on essential security, here is how to protect users by securing HTTP headers for a website.
HTTP Versioning
·740 words·4 mins
Posts async cli http python tools
HTTP/1 vs HTTP/1.1 vs HTTP/2 vs HTTP/3
Understanding AsyncIO by Code
·691 words·4 mins
Posts 101 code async python
Let’s learn AsyncIO by code.