[Teas] Re: [OPS-DIR]Re: Opsdir last call review of draft-ietf-teas-actn-vn-yang-26

[Mahesh Jethanandani <[e-mail]>]

Hi Dhruv,

> On Jun 10, 2024, at 12:58 PM, Dhruv Dhody <[e-mail]> wrote:
> 
> Hi Mahesh,
> 
> On Sat, Jun 8, 2024 at 9:47 PM Mahesh Jethanandani <[e-mail] <mailto:[e-mail]>> wrote:
> Hi Dhruv,
> 
>> On Jun 7, 2024, at 2:57 AM, Dhruv Dhody <[e-mail] <mailto:[e-mail]>> wrote:
>> 
>>  
>> >From this sentence, it seems that LTP refers to the LTP in the native TE
>> topology, because in YANG, PE is defined as the PE node in the native TE
>> Topology. While in the YANG model and the Examples in B.1. VN JSON, it looks
>> like "ltp" is the LTP within the abstract node, correct? If so, could the YANG
>> path be modified to a relative path of the abstract node?
>> 
>> 
>> Dhruv: Since the 'te-node-id' is not a key, it is not possible to change to relative path like "path "/nw:networks/nw:network/nw:node[tet:te-node-id=current()/../abstract-node]/nt:termination-point/tet:te-tp-id";"
>> Instead I will update the description. 
> 
> I am not able to understand this comment. I understand that you cannot provide a path like you mention above, but at the same time, Wubo does make a good point. Which LTP is being referred to? The one in the underlying layer or in the abstract layer?
> 
> 
> Dhruv: To handle this comment from Bo, I updated the description of this leaf to explicitly say abstract-node. 
> 
>            leaf ltp {
>              type leafref {
>                path "/nw:networks/nw:network/nw:node/"
>                   + "nt:termination-point/tet:te-tp-id";
>              }
>              description
>                "A reference to Link Termination Point (LTP) in the
>                 abstract-node.";
>              reference
>                "RFC 8795: YANG Data Model for Traffic Engineering (TE)
>                 Topologies";
>            }
> 
> Is there anything else that I should be doing? 

It is entirely possible that I have not understood the model, so do educate me if necessary.

By providing an absolute reference, and using the XPath that you have, I get this to be a reference to the underlying layer, not the abstract layer. But maybe the XPaths in the underlying layer are the same as in the abstract layer, in which case I am wrong. Regardless, by using a relative XPath, you will stay in the layer that you are currently in without any confusion. Won’t you?

> 
> Thanks! 
> Dhruv
>  
> 
> Mahesh Jethanandani
> [e-mail] <mailto:[e-mail]>

Mahesh Jethanandani
[e-mail]