Re: [core] Open erratas to core documents

[Esko Dijk <[e-mail]>]

Hi Francesca,

Agree with your suggestions.

The following 2 errata about the fragment identifier may also require some more text (e.g. for a next document update?):

https://www.rfc-editor.org/errata/eid4946 verified
https://www.rfc-editor.org/errata/eid4947 verified

While I do agree with the intention of allowing the fragment identifier in the CoAP URI, just adding it only in the URI definition without further text may be a bit confusing to the reader.
Especially since the CoAP URI parsing algorithm says “fail this algorithm” whenever a URI has a <fragment> component. At first sight, this looks strange: add the fragment and suddenly my URI cannot be parsed anymore? Then why allow a fragment in the first place?

It may take a bit of explanation to point to RFC 3986 section 3.5 and what that means for CoAP.

Section 5.10.1 currently says:

   Note: Fragments ([RFC3986], Section 3.5<https://datatracker.ietf.org/doc/html/rfc3986#section-3-5>) are not part of the request

   URI and thus will not be transmitted in a CoAP request.

Adding the “fragment” back to the CoAP URI will thus be confusing as the above says there is no fragment. Then we have a “CoAP URI” which is different from the “request URI”, and we also have a string “|url|” for which it is unclear whether that equals the CoAP URI or the request URI.
Maybe explaining this clearly in the erratum text could be sufficient for now, but it’s not yet as clear as we would like the CoAP specification to be w.r.t. fragment handling.

Regards
Esko

From: core <[e-mail]> On Behalf Of Francesca Palombini
Sent: Monday, November 7, 2022 16:02
To: [e-mail]
Cc: [e-mail]
Subject: [core] Open erratas to core documents

Hi,

I have been looking at the list of erratas still in reported state for the CoRE working group: https://www.rfc-editor.org/errata_search.php?rec_status=2&wg_acronym=core&presentation=table and I’d like the working group opinion on how to resolve them. You can see my suggestion below. Please let me know if you agree (or not).

Thanks,
Francesca (as ART AD)

https://www.rfc-editor.org/errata/eid4895 Hold for doc update
https://www.rfc-editor.org/errata/eid4946 verified
https://www.rfc-editor.org/errata/eid4947 verified
https://www.rfc-editor.org/errata/eid4948 verified (to double check - also I am not sure “(...)” is allowed in the errata, might need to edit it)
https://www.rfc-editor.org/errata/eid4949 verified
https://www.rfc-editor.org/errata/eid4954 Hold for doc update? (or verified)
https://www.rfc-editor.org/errata/eid5078 verified
https://www.rfc-editor.org/errata/eid5254 rejected (changes the RFC original intention)
https://www.rfc-editor.org/errata/eid5284 verified
https://www.rfc-editor.org/errata/eid5429 rejected (changes the RFC original intention)