Re: [core] [netconf] YANG encoding in CBOR

Hello,

I also support value outside and string inside unions. Michel gave a rather
good rationale why this is a good option and for me this seems as the best
option available right now.

Best regards,
Ivaylo

On Thu, Mar 28, 2019 at 3:32 PM Juergen Schoenwaelder <
j.schoenwaelder@jacobs-university.de> wrote:

> I agree with your points. It seems this is the best we can do right
> now and hopefully we can do better in the future.
>
> /js
>
> On Thu, Mar 28, 2019 at 02:06:17PM +0000, Michel Veillette wrote:
> > +1 for value outside, string inside union
> >
> > My rational:
> >
> > - SID can have up to 8 bytes and will provide a marginal compression.
> > - Generating SID for all 'enumeration' and 'bits' in case they are used
> in 'union' is lot of overhead to revolve a corner case.
> > - YANG also support the 'choice' statement (
> https://tools.ietf.org/html/rfc7950#section-4.2.7) which is similar to
> >    the concept of type tagging within union. This construct can be used
> by YANG modules specifically created for
> >   embedded applications to take advantage of 'enumeration' encoded as
> value and 'bits' encoded as bit.
> >
> > Michel
> >
> >
> > -----Original Message-----
> > From: Carsten Bormann <cabo@tzi.org>
> > Sent: Thursday, March 28, 2019 9:47 AM
> > To: Juergen Schoenwaelder <j.schoenwaelder@jacobs-university.de>
> > Cc: Michel Veillette <Michel.Veillette@trilliant.com>; Ladislav Lhotka <
> lhotka@nic.cz>; netconf@ietf.org; core@ietf.org
> > Subject: Re: [netconf] YANG encoding in CBOR
> >
> > So for enum/bits we are back to value outside, string inside union?
> > Can we use SIDs instead of the strings for the latter?
> > This can’t be much more complicated than assigning a SID to each string
> that occurs in an enum/bits in a union.
> >
> > Grüße, Carsten
> >
>
> --
> Juergen Schoenwaelder           Jacobs University Bremen gGmbH
> Phone: +49 421 200 3587         Campus Ring 1 | 28759 Bremen | Germany
> Fax:   +49 421 200 3103         <https://www.jacobs-university.de/>
>
> _______________________________________________
> core mailing list
> core@ietf.org
> https://www.ietf.org/mailman/listinfo/core
>