Re: [netmod] yang-json document

[Andy Bierman <[e-mail]>]

On Thu, Feb 5, 2015 at 2:09 PM, Juergen Schoenwaelder
<[e-mail]> wrote:
> On Thu, Feb 05, 2015 at 01:57:16PM -0800, Andy Bierman wrote:
>>
>> That YANG text was written without any use-cases in mind.
>> If a YANG module is written that has a use-case for unconstrained XML,
>> then anyxml is ready to use.
>>
>> I have never seen anyxml actually used that way.
>> In every case, there is a schema to apply, known to
>> the implementation somehow.  In every case there are
>> only YANG nodes expected. No PIs. No mixed context.
>>
>> So I do not believe this is a real problem.
>> I think YANG 1.1 could spend a lot of effort defining
>> schema rules for unconstrained data, but this might not
>> help address the real uses of anyxml.
>>
>
> The argument "my code has a more specific interpretation and I believe
> everybody else does the same" only works if everybody confirms it. I
> have not seen that confirmation. Hence, to fix this, we have Y34-05
> for YANG 1.1.
>
> This thread is about the current JSON mapping document, which is based
> on YANG 1.0. Again, the argument "my code has a more specific
> interpretation and I believe everybody else does the same" only works
> if everybody confirms it. As long as we do not have a confirmation
> that everybody happens to do what your code does, I think we have to
> assume that anyxml means what RFC 6020 says, any xml.
>

Maybe you misunderstood.
anyxml is being used as a placeholder for a "real schema"
that YANG is unable to express.  "my code knows how
to pick the real schema for this anyxml blob" is part of
the data model or protocol definition.

I don't think Y34-05 addresses this issue.

Even the "blog" use-case expects valid HTML to be specified
in the anyxml, not free-form, any XML.


> /js
>

Andy

> --
> Juergen Schoenwaelder           Jacobs University Bremen gGmbH
> Phone: [phone]         Campus Ring 1 | 28759 Bremen | Germany
> Fax:   [phone]         <http://www.jacobs-university.de/>