Re: [Softwires] draft-ietf-softwire-yang: YANG compilation issues

Hi Benoit and Per,

Thanks for the info, we will correct this in the next version.

Best,
Linhui

2016-08-31 18:59 GMT+08:00 Benoit Claise <bclaise@cisco.com>:

> draft-ietf-softwire-yang authors,
>
> Per and I have been investigating the YANG compilation error for this
> document.
> See http://www.claise.be/IETFYANGPageCompilation.html:
>
> ietf-softwire@2016-06-04.yang:215: error: a 'default' value cannot be
> given when 'mandatory' is "true"
>
> The error refers to:
>
>   container port-set {
>       description
>         "For Lightweight 4over6, the default value
>         of offset should be 0, to configure one contiguous
>         port range.";
>       uses port-set {
>         refine psid-offset {
>           default "0";                   <=========
>         }
>       }
>
>
>
>   grouping port-set {
>     description
>       "Use the PSID algorithm to represent a range of transport layer
>       ports.";
>     leaf psid-offset {
>       type uint8 {
>         range 0..16;
>       }
>       mandatory true;               <==== If I remove this line, it works fine.
>       description
>         "The number of offset bits. In Lightweight 4over6, the default
>         value is 0 for assigning one contiguous port range. In MAP-E/T,
>         the default value is 6, which excludes system ports by default
>         and assigns distributed port ranges. If the this parameter is
>         larger than 0, the value of offset MUST be greater than 0.";
>     }
>
>
> You can refine-add/modify a default, but the result must be valid YANG,
> and 6020 7.6.4 (and draft-ietf-netmod-rfc6020bis-14) says
>
>    The "default" statement MUST NOT be present on nodes where
>    "mandatory" is true.
>
> I.e. if you want to refine-add a default to a mandatory leaf, you must
> also refine-remove the mandatory-ness:
>
>        refine psid-offset {
>          default "0";
> 	 mandatory false;
>        }
>
> (Having a default on a mandatory leaf doesn't make sense, since it can
> never take effect - you are required to configure a value for the leaf.)
>
> Regards, Per and Benoit
>
>
> _______________________________________________
> Softwires mailing list
> Softwires@ietf.org
> https://www.ietf.org/mailman/listinfo/softwires
>
>