HTTP Requests response status code handlings #42
Conversation
| class RLRequestFailed(RateLimiterError): | ||
| def __init__(self, request, *args, **kwargs): | ||
| super().__init__(request, *args, **kwargs) | ||
|
|
||
| class RLRequestClientError(RLRequestFailed): | ||
| def __init__(self, request, *args, **kwargs): | ||
| super().__init__(request, *args, **kwargs) |
There was a problem hiding this comment.
Maybe RLClientError, RLClientError4xx would be a better name?
(same with Server)
| class RLRequestClientError(RLRequestFailed): | ||
| def __init__(self, request, *args, **kwargs): | ||
| super().__init__(request, *args, **kwargs) |
There was a problem hiding this comment.
Maybe RLClientError, RLClientError4xx would be a better name?
(same with Server)
There was a problem hiding this comment.
also the name RLRequestServerError wrongly give the impression of inheriting from RLRequestFailed
| class RLRequestFailed(RateLimiterError): | ||
| def __init__(self, request, *args, **kwargs): | ||
| super().__init__(request, *args, **kwargs) |
There was a problem hiding this comment.
If I'm not mistaken the whole point of this exception is to allow future subclassing (because today only RLRequestClientError inherits from it)
However it made the code a bit more obscure (especially because you except the parent class not this one)
So in the interest of readability maybe we could get rid of this one and instead except RLClientError later?
…, server error mean longer time to wait, client error means request failed
…ass the exception to calling process
ctmbl
force-pushed
the
request-response-code-handlings
branch
from
b284a6d to
4b72f13
Compare
ctmbl
force-pushed
the
request-response-code-handlings
branch
from
4b72f13 to
9121289
Compare
The goal of this PR is to close #41 & close #28.
This aims to make the bot handle more HTTP responde status codes, 40X errors that mean there is a problem with the request and 50X errors that means there is a problem on server side.