Re: [netmod] yang-data-ext issues

[Kent Watsen <[e-mail]>]

Another and somewhat radical idea is to think of 'yang-data' as defining a data node, like a 'container', but not a config or opstate node.  Yes, this is different from rc:yang-data, which defines a transparent node, like 'choice', but maybe it's okay if we can get the substitution groups part working.  For instance:

From draft-ietf-anima-voucher:
    OLD:
       rc:yang-data voucher-artifact {
          container voucher {…}
        }
    NEW:
       yd:yang-data voucher {
          …
        }

From draft-ietf-netconf-zerotouch:
    OLD
         rc:yang-data "zerotouch-information" {
           choice information-type {
             container redirect-information {…}
             container onboarding-information {…}
           }
         }
    NEW
         yd:yang-data "redirect-information" {
             yd:substitution-group zerotouch-information;
             …
         }
         yd:yang-data "onboarding-information" {
             yd:substitution-group zerotouch-information;
             …
         }

And for the error-info use-case:

    yd:yang-data "error-info" {
        choice error-type {
            leaf my-error1 {…}
            leaf my-error2 {…}
        }
   }


Thoughts?

Kent