Re: analysis of YANG vs. RELAX NG

[Andy Bierman <[e-mail]>]

Juergen Schoenwaelder wrote:
> On Mon, Dec 03, 2007 at 02:56:02PM +0100, Ladislav Lhotka wrote:
>   
>> Tim Bray p????e v Ne 02. 12. 2007 v 12:42 -0800:
>>     
>>> I've been staying off this thread because I don't understand what a
>>> NETCONF is or what you'd do with one.  But Julian's point is massively
>>> important.  If you're going to design your own modeling language, you
>>> need to spend a *lot* of time thinking about the extensibility model.
>>> It's really easy to get wrong.  -Tim
>>>       
>> Indeed. An extensibility problem in YANG is IMO the leaf statement. It
>> is encoded as an XML element but the fact that it is declared as leaf
>> effectively makes it into an XML attribute: It is impossible to extend
>> it (e.g., add a qualifying subelement) without changing the leaf into
>> container. In a RELAX NG schema, if properly designed, such an extension
>> wouldn't have to touch the parent schema.
>>     
>
> Leafs carry data and we strongly dislike mixed contents in NETCONF. So
> this is actually a YANG feature and not a bug.
>  
>   

I strongly agree with Juergen.
YANG is based (in part) on 18+ years experience
with SNMP and SMI.

It s absolutely forbidden in NM to redefine the syntax and semantics
of a managed object in this way.


Andy


>> Another problem might be the default statement in leaves (or it is
>> proper to say leafs here?:-) or typedefs. What if a standard data model
>> defines an item, say a hello timer, with such a default, but an
>> implementation wants to use another value for the default? Should it
>> come up with another schema?
>>     
>
> No. There can only be one standard default. If you are thinking about
> reusable groupings, then defaults can actually be changed wherever you
> apply a grouping as far as I know.
>
> /js
>
>