i've never heard of a depth argument. is this new? does it need to be documented?

This could probably be handled better. It's to track how many attempts have been made

Maybe just use a loop...?

@harlowja can you elaborate on what that might look like? 🤔

Something like:

reply = self.api_requester.do(self.token,
                              connect_method, timeout=timeout,
                              post_data=kwargs)
retry_attempt = 0
while reply.status_code != 200:
    if reply.status_code == 429:
        ratelimit_timeout = int(reply.headers.get('retry-after', 120))
        time.sleep(ratelimit_timeout)
    retry_attempt += 1
    if retry_attempt > 10:
        raise SlackConnectionError("RTM connection attempt was rate limited 10 times.")
    else:
        reply = self.api_requester.do(self.token,
                              connect_method, timeout=timeout,
                              post_data=kwargs)

You can also use http://tenacity.readthedocs.org/ and it will do alot of this for u.

I updated it to work more like the reconnect_count login in the block above the retry logic

tenacity looks like it'd be a very good addition to the work I have planned for 2.x

Ok, I am co-author of that library, so feel free to ask questions.

oh i see, its used to measure the number of recursions. if this is strictly for private use, then i guess it doesn't need to be documented.