As a co-author for uri-path-abbrev, and given christian time constraints, I'll see if I can process the issues into a PR.
", "time": "2026-02-11T15:07:36.000Z"}, {"author": "Marco Tiloca", "text": "Thanks, Michael
", "time": "2026-02-11T15:07:52.000Z"}, {"author": "Michael Richardson", "text": "I didn't know I needed an XPath tutorial today. but really I do....
", "time": "2026-02-11T15:13:10.000Z"}, {"author": "Michael Richardson", "text": "Andy is not here. We'll have to do this again with him :-(
", "time": "2026-02-11T15:22:01.000Z"}, {"author": "Marco Tiloca", "text": "On where building the tentative text for the erratum, there's indeed the GitHub repo at https://github.com/core-wg/yang-cbor --- Perhaps just start a new file there?
", "time": "2026-02-11T15:36:45.000Z"}, {"author": "Esko Dijk", "text": "@Marco good idea!
", "time": "2026-02-11T15:39:46.000Z"}, {"author": "Michael Richardson", "text": "the choice node doesn't represent anything concrete.
", "time": "2026-02-11T15:41:21.000Z"}, {"author": "Michael Richardson", "text": "the fact that you ought not transmit both 1739 and 1740 is a semantic thing. It wouldn't be syntactically wrong.
", "time": "2026-02-11T15:42:48.000Z"}, {"author": "Carsten Bormann", "text": "Expressio unius est exclusio alterius
", "time": "2026-02-11T15:46:58.000Z"}, {"author": "Michael Richardson", "text": "The problem with the full XPath stuff is that it's probably missing from a partial (constrained) implementation. Quality of implementation I guess.
", "time": "2026-02-11T15:52:03.000Z"}, {"author": "Esko Dijk", "text": "YANG data structures
", "time": "2026-02-11T15:52:10.000Z"}, {"author": "Vojt\u011bch Vil\u00edmek", "text": "We also miss XPath constrained alternative that would be simpler than the string representation and more powerful than the raw SID-based 'instance-identifier'.
", "time": "2026-02-11T15:54:23.000Z"}, {"author": "Vojt\u011bch Vil\u00edmek", "text": "give me a sec
", "time": "2026-02-11T15:58:29.000Z"}, {"author": "Vojt\u011bch Vil\u00edmek", "text": "ok
", "time": "2026-02-11T15:58:40.000Z"}, {"author": "Vojt\u011bch Vil\u00edmek", "text": "+1
", "time": "2026-02-11T16:00:47.000Z"}, {"author": "Michael Richardson", "text": "Vojt\u011bch Vil\u00edmek said:
\n\n\nWe also miss XPath constrained alternative that would be simpler than the string representation and more powerful than the raw SID-based 'instance-identifier'.
\n
I would love to solve this problem; please articulate some concrete use cases, as I come to YANG in a round-about way.
", "time": "2026-02-11T16:01:02.000Z"}, {"author": "Carsten Bormann", "text": "+1 michael
", "time": "2026-02-11T16:01:18.000Z"}, {"author": "Esko Dijk", "text": "ok to start the adoption call!
", "time": "2026-02-11T16:06:52.000Z"}, {"author": "Maria Mat\u011bjka", "text": "don't we need then explicitly signal whether the other side supports metadata?
", "time": "2026-02-11T16:08:10.000Z"}, {"author": "Vojt\u011bch Vil\u00edmek", "text": "I think that this is closely coupled with the YANG-CBOR
", "time": "2026-02-11T16:08:18.000Z"}, {"author": "Vojt\u011bch Vil\u00edmek", "text": "I think the clients should only use things from modules. So if the metadata is in the module, the client may use it.
", "time": "2026-02-11T16:09:35.000Z"}, {"author": "Vojt\u011bch Vil\u00edmek", "text": "double tree is too heavy
", "time": "2026-02-11T16:09:50.000Z"}, {"author": "Vojt\u011bch Vil\u00edmek", "text": "I would like to see a process that could restrict a use of metadata for e.g. a subtree, This is maybe better to discuss on NETMOD
", "time": "2026-02-11T16:11:58.000Z"}, {"author": "Maria Mat\u011bjka", "text": "+1 to the restriction requirement
", "time": "2026-02-11T16:13:14.000Z"}, {"author": "Maria Mat\u011bjka", "text": "also: imposing the metadata on top of an imported module looks like a major implementation problem if one wants to delegate parsing of such data trees to another module \u2026 but that's more of a concern about RFC 7952 itself
", "time": "2026-02-11T16:14:53.000Z"}, {"author": "Vojt\u011bch Vil\u00edmek", "text": "I am also not sure if the RFC 7952 forbids to add metadata to metadata
", "time": "2026-02-11T16:15:50.000Z"}, {"author": "Marco Tiloca", "text": "Yes, proposed text on piggybacked vs. separate response is in the issue description
", "time": "2026-02-11T16:20:47.000Z"}, {"author": "Marco Tiloca", "text": "(together with side points)
", "time": "2026-02-11T16:20:54.000Z"}, {"author": "Michael Richardson", "text": "Not having implemented exactly... if I know the code (it's in the RFC), but I don't do it (turned off, no code, configured off, compiled out...), then 4.05. If it's some new code that I never heard of, then 5.01.
", "time": "2026-02-11T16:22:52.000Z"}, {"author": "Michael Richardson", "text": "gotta go, Thank you Carsten.
", "time": "2026-02-11T16:23:59.000Z"}, {"author": "Michael Richardson", "text": "(not claiming I'm right!)
", "time": "2026-02-11T16:25:27.000Z"}, {"author": "Michael Richardson", "text": "If I thought about supporting PATCH or OBSERVE, and decided not to, (or if the code was compiled out due to code space), then I would know about it.
", "time": "2026-02-11T16:26:38.000Z"}, {"author": "Michael Richardson", "text": "I did subissues for an AD review.
\nMeh. Not an advantage.
ok
", "time": "2026-02-11T16:29:39.000Z"}]