Can a python async http transfer by easily added to an existing code base? - python-asyncio

Question

I'm writing a Fusion 360 python add-in, which is an event-driven way to extent their product (their code calls my functions that hooked in to their events).
Inside my code, I would like to send a single HTTP GET (or POST) request to a remote server without making the user wait (e.g. if they're offline, I want no delay - it just needs to fail silently).
There are many dozens of async examples around, but all of them appear to require that you're running a "normal" program, and that every part of the program is async to start with (i.e. I can't find any examples of a regular program, with an async bit added).
I'm new to python, and the async Doc is drowning me :-(
That said - I do kinda know what I'm doing in other languages, and I understand how processes work (not so much threads though).
I did manage to partly "solve" my own question with this:
subprocess.Popen([get_exec(),os.path.join(prog_folder,"send_data.py"),str(VERSION)])
and a second script - except that pops open an ugly black "DOS" box which hangs around until the transfer completes and looks highly unprofessional. All attempts at avoiding the black box failed (I do not get the luxury of specifying my user's environment, and there is no "windows UI build" python version shipped that works.)
So basically - two questions
a) is it even possible for an event-driven python function to even "spawn" a thread at all? Perhaps imagine it this way: you've written a python module, and any caller can call a function in your module, which returns immediately, but your function then continues to do work for another minute in parallel - but crucially - the caller does not need to do anything special.
b) assuming it more-or-less is possible - can anyone give me a hint or a pointer to an example or something might might give me a clue where to start?
Python 3.7.6+ is my minimum environment.
My main problem (pardon the pun) is that all examples I can find do this:
loop.run_until_complete(asyncio.wait(print_http_headers(url)))
or this:
asyncio.run(main())
both of which block. Even the asyncio doc's "hello world" example is non-async as well (if only they had printed "world" first (after a 1s delay) and then printed "hello" second with no delay - that would have solved everything!!!)
All other suggestions gratefully received (there's bound to be an "outside the box" alternative I've not realized yet I expect - so long as the box isn't black and in-your-face that is :-)

Answer 1

Thanks! #user4815162342 - that totally did the trick!!
import _thread, time, socket
def YoBlably(stuff):
s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
s.connect(("example.com", 80))
sent=s.send(b'GET /my_path/check_update.asp?u=u1.20200505&v=1.20200503&a=_aft&p=my_prog HTTP/1.1\x0d\x0aHost: example.com\x0d\x0aAccept-Encoding: identity\x0d\x0aUser-Agent: Python\x0d\x0aConnection: close\x0d\x0a\x0d\x0a')
if sent==0:
print("s Problem")
chunk = s.recv(1024000)
if chunk == b'':
print("r Problem")
print('got {}.'.format(chunk))
s.close()
print("Starting in 1s...");
time.sleep(1)
_thread.start_new_thread(YoBlably, ('foo',))
print("Started...");
for i in range(0,6):
time.sleep(1)
print('{}...'.format(i))
print("The end...");
outputs:-
$ python pythreadsock.py
Starting in 1s...
Started...
got b'HTTP/1.1 200 OK\r\nDate: Wed, 20 May 2020 01:47:28 GMT\r\nServer: Apache/2.0.52\r\nExpires: Sun, 17 May 2020 23:58:28 GMT\r\nPragma: no-cache\r\nCache-Control: no-cache\r\nContent-Length: 35\r\nConnection: close\r\nContent-Type: application/json; charset=UTF-8\r\n\r\n{"current_version_xyz":1.20200502}\n'.
0...
1...
2...
3...
4...
5...
The end...