How to check errors returned by `CreateChatCompletion` was a network or rate limiting issues ?

[guoqiangqi]

I am retring my request with the exponential backoff when encountering network or rate limiting issues. Currently, it seems that go-openai does not provide specific network errors for requests by client.CreateChatCompletion and other, openai python client do a good job by providing errors like openai.error.RateLimitError, openai.error.APIError, openai.error.Timeout, openai.error.APIConnectionError, openai.error.ServiceUnavailableError, how can i check errors returned by CreateChatCompletion was a network or rate limiting issues ?

[dannysievers]

This can be solved by using errors.As() as seen in the example README. Check out PR comments #293 for implementation details.

e := &openai.APIError{}
if errors.As(err, &e) {
  switch e.HTTPStatusCode {
    case 401:
      // invalid auth or key (do not retry)
    case 429:
      // rate limiting or engine overload (wait and retry) 
    case 500:
      // openai server error (retry)
    default:
      // unhandled
  }
}

[guoqiangqi]

@dannysievers it works, thanks for your code!