Re: [Netconf] [yang-doctors] YANG Doctor question: empty mandatory choice?

[Martin Bjorklund <[e-mail]>]

Kent Watsen <[e-mail]> wrote:
> Hi Juergen,
> 
> 
> > Why is the transport leaf needed if the choice cases identify the transport? 
> 
> My understanding is that this leaf is needed (exclusively?) to support the
> enforcement that all the transports for a given subscription use the same
> transport.
> 
> I've also been asking about this, partly from the "difficultly to enforce"
> reason Eric mentions (though see the "when" statement in Eric's 7/20 email
> where he resolves this, I thought), but also from a "are we making this
> more difficult than needed?" perspective.
> 
> 
> > Why do all receivers of a subscription have to use the same transport?
> 
> This was something that Martin and Eric worked out before we did the first
> Last Call.  Eric doesn't seem to know the particular reason, other than 
> Martin seems to think it’s easier.

No; I personally also prefer a design where each receiver has its own
transport + encoding.  The original model had a common "encoding" for
all receivers, and then a receiver-specific transport - I think this
is even worse, and suggested to have transport + encoding defined
together preferrably receiver-specifc or else common for all
receivers.

If the WG now believes that the transport + encoding should be done per
receiver, this should be fairly easy to change.


> Thinking out loud, I can sort-of, but not really, understand the motivation
> for wanting to enforce the same encoding (xml, json, etc.) across receivers
> and, since some transports may require certain encodings (i.e. netconf-->xml,
> coap-->cbor, etc.), it carries that the transports should be the same too?
> 
> What I don't like about this argument is:
> 
> 1) It's not user-friendly.  If a server really does have to send similar 
>    subscriptions to a set receivers using different transports and/or 
>    encodings, we've introduced what looks like an artificial need to 
>    duplicate the subscriptions for each unique transport+encoding 
>    combination.
> 
> 2) I don't understand the implementation issue.  My view is that slightly
>    clever code that caches encodings used could do this in one loop.  
>    Here's some pseudo-pythonic code:
> 
>    send-notification-to-receivers(receivers, native-notification):
>      xml-notification=None
>      json-notification=None
>      for receiver in receivers:
>        if receiver.encoding is XML:
>          if xml-notification is None:
>            xml-notification = to-xml(native-notification)
>          receiver.send(xml-notification)
>        if receiver.encoding is JSON:
>          if json-notification is None:
>            json-notification = to-json(native-notification)
>          receiver.send(json-notification)
> 
>    
> Kent // as a contributor


/martin