> Does the manufacturer-private satisfy your need here?
Yes, this looks fine. The issue of "hidden communication channel" that was
raised for this field, would equally be there for standardized extensions that
import from other YANG modules. E.g. I can always find a bstr field from some
arbitrary other YANG module, use it as extension, and store my hidden data in
there.
> Carsten has made a suggestion during the meeting for the problem of strings
> (JSON) vs integers/SIDs (CBOR). I haven't not yet understood the suggestion
> as to what to do here. I hope to close the loop soon though.
The idea was to use a single type that's already defined in RFC 9254 to be
serializable to both text and to SID value.
E.g. like below:
leaf-list extensions {
type instance-identifier {
}
description
"A list of extension modules that are used in this Voucher
file.";
}
But this may be the wrong one. There also "yang-identifier", "identifier" and a
few more candidates perhaps.
The one I picked doesn't seem to have the desired (short) string serialization?
Otherwise, maybe a YANG "union" can be used between type "sid" (RFC 9254) and
type "module-name" (whatever type defines the module name, I couldn't find
this.)
Esko
-----Original Message-----
From: Michael Richardson <[email protected]>
Sent: woensdag 19 maart 2025 11:21
To: Esko Dijk <[email protected]>
Cc: [email protected]
Subject: Re: Details of proposed voucher "extensions" in Github PR
Esko Dijk <[email protected]> wrote:
> Reading the 8366bis Pull Request for voucher extendibility,
> https://github.com/anima-wg/voucher/pull/79 , a few questions came up.
Does the manufacturer-private satisfy your need here?
Carsten has made a suggestion during the meeting for the problem of strings
(JSON) vs integers/SIDs (CBOR). I haven't not yet understood the suggestion
as to what to do here. I hope to close the loop soon though.
--
Michael Richardson <[email protected]>, Sandelman Software Works
-= IPv6 IoT consulting =- *I*LIKE*TRAINS*
_______________________________________________
Anima mailing list -- [email protected]
To unsubscribe send an email to [email protected]
