Limiting OAS3.1 to RFC7231 status codes?

[ioggstream]

I expect

To reference all registered status codes

Instead

The spec says

The available status codes are defined by RFC7231

Note

It is not clear to whether the current formulation limits the available status codes to the one in 7231

[philsturgeon]

What else would it mean? You can use any status codes defined in RFC 7231. Using custom status codes is undefined and therefore undefined behavior is to be expected.

Was there a particular status code you were looking to use thats not in 7231 or are you trying to use custom codes?

For extra context, the JSON Schema validates the following:

patternProperties:
          '[1-5](?:\d{2}|XX)':

So you could go using custom status codes if you wanted to, but if its outside of those rules some validators will barf errors at you.

[ioggstream]

429 status code is rfc6585, but not in 7231 ;)

I'm just saying that there are http-extensions beyond 7231 and they are standards too.

[hkosova]

OAS refers not only to RFC 7231 but also to the IANA registry:

https://github.com/OAI/OpenAPI-Specification/blob/master/versions/3.1.0.md#http-status-codes

HTTP Status Codes
The available status codes are defined by RFC7231 and registered status codes are listed in the IANA Status Code Registry.

RFC 7231 itself points to the IANA registry for the complete list of codes:

Note that this list is not exhaustive -- it does not include extension status codes defined in other specifications. The complete list of status codes is maintained by IANA.

Maybe the intended meaning here is that the definition of "HTTP status code" is taken from RFC 7231 (i.e. it's a 3-digit integer in the 100-599 range), and for specific registered status codes the reader should refer to the IANA registry?

[philsturgeon]

@hkosova sounds good!

[ioggstream]

@hkosova exactly, the text just need a slight rewording. Feel free to PR with your text... or I'll do it tomorrow if you don't have time.

[MikeRalphson]

@ioggstream @hkosova woudl one of you like to take a go at that proposed PR?

[ioggstream]

The httpwg is releasing the new http spec. I will create a PR next week but we could even wait for the new spec to be released so that we have the latest references.

[MikeRalphson]

#2630 was merged. Thanks all.

[ioggstream]

@MikeRalphson sorry for the delay. The PR is ok. Thanks!