Re: [YANG] insert and key attributes

Phil Shafer wrote:
> Andy Bierman writes:
>> The NETCONF WG should agree on this sort of new NETCONF protocol feature,
>> the same as any other.
> 
> The feature relies on the data modeling rules of YANG.  It will be
> a standard part of YANG-based models, but not a standard part of
> NETCONF.  Other data modeling languages are free to define their
> own ordering schemes, which need not be compatible with what we
> have in YANG.  The dependency would mean that spliting it into
> another document is just adding world.  Leave it in the YANG draft.
> 

I disagree that this is a good design approach for a standard protocol.
There is nothing YANG-specific about user-ordered lists or the
concept of inserting this entry before/after that one.  If the
WG agrees to add this sort of standard operation to the protocol,
then DMLs like YANG can take advantage of it (ala error-app-tag).

IMO, the NETCONF protocol should define a standard set of features
that does not change depending on the DML used to define NETCONF content.

> Thanks,
>  Phil
> 
> 

Andy

_______________________________________________
YANG mailing list
YANG@ietf.org
https://www1.ietf.org/mailman/listinfo/yang