[QUESTION] : Move to a single http request library

[baseplate-admin]

Hi there,

It seems that you guys are using both requests and aiohttp to perform HTTP Request. Can we switch to httpx ? It has requests compatibility and a similar interface like aiohttp

We can probably remove this file :

Automatic-Udemy-Course-Enroller-GET-PAID-UDEMY-COURSES-for-FREE/udemy_enroller/http.py

Lines 8 to 24 in b9325b6

	async def get(url, headers=None):
	"""
	Send REST get request to the url passed in
	:param url: The Url to get call get request on
	:param headers: The headers to pass with the get request
	:return: data if any exists
	"""
	if headers is None:
	headers = {}
	try:
	async with aiohttp.ClientSession() as session:
	async with session.get(url, headers=headers) as response:
	text = await response.read()
	return text
	except Exception as e:
	logger.error(f"Error in get request: {e}")

[fakeid30]

@cullzie what do you think?

[baseplate-admin]

This is not a good take on this issue as cloudscraper internally uses requests and requests-toolbelt and so including httpx will introduce yet another HTTP Client library ( which we are trying to avoid )

[baseplate-admin]

I have another proposal to move everything to requests.

@cullzie what do you think ?

[cullzie]

Hey @baseplate-admin,
The cloudscraper dependency is the only reason we have both requests and aiohttp.
I would be happy to have a look at your proposal on this

[baseplate-admin]

It's been a couple of months.

So here's my take on this issue,

It's tiresome to wait for aiohttp to bump their dependencies every major python upgrade.

So, here are my proposals :

1. My first proposal is to move everything to httpx. It has a synchronous ( and on demand asynchronous ) api that would suffice our needs pretty well.

2. My second proposal is that, we dont use aiohttp and use requests for everything. The requests ecosystem is amazing and faster-cpython project willl only increase the performance of requests