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

[Carsten Bormann <[e-mail]>]

On Mar 25, 2019, at 07:23, Carsten Bormann <[e-mail]> wrote:
> 
> We can maybe cheat:  Use strings for enums inside unions for now, and values only outside.

Or, if it takes too long to do the identifier scheme:
We could define SIDs for each module for the enum and bits strings used in that module.
That would mean we’d have multiple SIDs for the same strings all over the YANG universe, but it is something we don’t have to wait for and can do locally.

So a decoder knowing the module at all would also know what string to imagine (it never has to be reified!) for that SID.  An encoder has a slightly harder time, as it has to use a SID from a module the decoder will know; I’d imagine that won’t be too hard to glean from the context though.

Grüße, Carsten