Re: [netconf] Wrong example in RFC 8040 Section 3.1

[Kent Watsen <[e-mail]>]

Hi Henning,


>> No, the examples are for the Root Resource Discovery, i.e. what is the
>> path to {+restconf} ; not for the Root Resource, i.e. accessing the actual
>> {+restconf} path.
>> 
>> Note that the examples show different values (/restconf, and
>> /top/restconf respectively) for {+restconf}, the RESTCONF Root
>> Resource.
> 
> Okay... was a little bit subtil but I think you are right that I were wrong.
> 
> I was a bit confused that there is no normative example of querying
> the root resource, only one in the appendix.

Per is correct and, yes, that section could’ve been better formatted 


>>> Is it correct to report data/operations always with an empty
>>> dictionary for this root query, even if there are data/operations
>>> available?
>> 
>> Yes, I would say it is correct even though it is not explicitly stated
>> in RFC 8040. The API resource is an overview of supported
>> functionality by the RESTCONF server.
> ....
>> Note that {+restconf}/data and {+restconf}/yang-library-version are mandatory
>> while {+restconf}/operations is optional. If the RESTCONF Root Resource would
>> also emit all data it would be very difficult to get an overvew of what the
>> RESTCONF server supports.
> 
> Good, that one more "done" point in my implementation...
> 
> currently I am slowly working myself through RFC8040 to extract the
> examples as test cases for my code.

I tend to agree again with Per here, though my argument rests on that the example would otherwise be invalid since, “ietf-yang-library” is required and the example doesn’t pass the “content=confg” query parameter...


Kent // co-author