Skip to content

A Python Wrapper for httpx that combines the httpx.AsyncClient with the httpx.Client to allow both async and sync requests

License

Notifications You must be signed in to change notification settings

GrowthEngineAI/aiohttpx

Repository files navigation

aiohttpx

A Python Wrapper for httpx that combines the httpx.AsyncClient with the httpx.Client to allow both async and sync requests

Latest Version: PyPI version


Installation

# Install from pypi
pip install --upgrade aiohttpx

# Install from Github
pip install --upgrade git+https://github.com/trisongz/aiohttpx

Usage

aiohttpx is a wrapper around httpx that provides a unified async + sync interface for making HTTP requests. This is useful for making HTTP requests in both async and sync codebases.

Additionally, it includes a ProxyClient that can be used for scraping / high volume requests that would otherwise be blocked by the target server by using a rotating proxy pool through AWS API Gateway.

import asyncio

import aiohttpx

async def test_requests():
    # Notice it utilizes async context manager but can use sync methods.

    async with aiohttpx.Client() as client:
        # Make an Async GET request
        response = await client.async_get("https://httpbin.org/get")
        print(response.json())

        # Make a Sync GET request
        response = client.get("https://httpbin.org/get")
        print(response.json())
    
    # The same applies with the sync context manager
    with aiohttpx.Client() as client:
        # Make an Async GET request
        response = await client.async_get("https://httpbin.org/get")
        print(response.json())

        # Make a Sync GET request
        response = client.get("https://httpbin.org/get")
        print(response.json())


async def test_proxies():
    # Here we will test the ProxyClient
    # some magic/notes:
    
    # there is a wrapper for BeautifulSoup that is enabled for GET 
    # requests. This can be triggered by passing `soup_enabled=True` 
    # to the request method.
    
    # the ProxyClient will automatically terminate the api gateways upon 
    # exit from the context manager in both sync and async.

    # however if no context manager is used, then the ProxyClient will 
    # need to be manually terminated by calling 
    # `client.shutdown()` | `await client.async_shutdown()`

    # You can choose to perserve the api gateways by passing 
    # `reuse_gateways=True` to the ProxyClient constructor. 
    # This is useful if you want to reuse the same api gateways 
    # for multiple requests.

    # You can also increase the number of gateways per region to 
    # increase the number of concurrent requests. This can be done by 
    # passing `gateways_per_region=10` to the ProxyClient constructor.

    # by default the ProxyClient will use the `us-east-1` region. 
    # You can change this by passing `regions=["us-west-2"]` or 
    # `regions="us"` for all us regions to the ProxyClient constructor.


    base_url = "https://www.google.com"

    async with aiohttpx.ProxyClient(base_url = base_url) as client:
        # Make an Async GET request
        response = await client.async_get(
            "/search", 
            params = {"q": "httpx"},
            soup_enabled = True
        )
        print(response.soup)
        print(response.soup.title.text)

        # Make a Sync GET request
        response = client.get(
            "/search", 
            params = {"q": "httpx"},
            soup_enabled = True
        )
        print(response.soup)
        print(response.soup.title.text)
    
    # Upon exiting the context manager, the api gateways will be terminated.


async def run_tests():
    await asyncio.gather(
        test_requests(),
        test_proxies()
    )

asyncio.run(run_tests())

About

A Python Wrapper for httpx that combines the httpx.AsyncClient with the httpx.Client to allow both async and sync requests

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published