Greetings!

I'm able to authenticate with code that goes something like this:

request.request("POST", "/services/oauth2/token", urllib.parse.urlencode(body), headers)
response = request.getresponse()
print("status and reason: ", response.status, response.reason)
if(response.status == 200):
    response_string = str(response.read())
    formatted_for_json_string = response_string[-(len(response_string)-2):-1]
    json_response = json.loads(formatted_for_json_string)
    for value in json_response:
        if(value == "access_token"):
            access_token = json_response[value]
        if(value == "instance_url"):
            _endpoint = json_response[value]
        if(value == "token_type"):
            token_type = json_response[value]

That part works and sets up some of the required data for the Bulk Load Job creation.
 

headers = {
    "Authorization" : token_type + " " + access_token,
    "Content-Type": "application/json",
    "Accept": "*/*",
    "Cache-Control": "no-cache",
    "Host": host,
    "Accept-Encoding": "gzip, deflate, br",
    "Connection": "keep-alive"
}

body = {
    "object":"Account",
    "contentType":"CSV",
    "operation":"update"
}

request = client.HTTPSConnection(host, timeout=10)
request.request("POST", "/", urllib.parse.urlencode(body), headers)
response = request.getresponse()

The reponse is 200, but the gzip decompression of the response is:
b''

This is an indication that the response was empty. This makes sense considering a Bulk Data Load job was not created as a result of the transaction.

This all works in Postman and I've been attempting to mirror as much of the Postman behavior as possible except for adding the Postman-specific header bits and cookie info. The Authorization part has "Bearer key" in it, though the behavior is no different if I don't have a key.

If I use the key obtained in Postman, that also doesn't change the behavior.

Curious what I'm missing.