> On 22 May 2015, at 16:11, Andy Bierman <[email protected]> wrote:
> 
> On Fri, May 22, 2015 at 1:37 AM, Martin Bjorklund <[email protected]> wrote:
>> Ladislav Lhotka <[email protected]> wrote:
>>> Martin Bjorklund <[email protected]> writes:
>>> 
>>>> Andy,
>>>> 
>>>> I don't think the implementation burden on the server is that heavy.
>>>> A YANG 1.0 compliant server today supports:
>>>> 
>>>>   leaf a in module A of type foo from foo@2001-01-01
>>>>   leaf b in module B of type foo from foo@2002-02-02
>>> 
>>> In fact, the same can be done in two submodules of the same module, or
>>> main module and submodule, which leads exactly to the situation below -
>>> and that's allowed even in YANG 1.0.
>> 
>> Yes, you're right!  Which makes the current MUST NOT more of a CLR.
>> 
> 
> Completely wrong because each submodule is its own context and you
> cannot combine
> multiple revisions of any module in any 1 submodule.  All XPath and other
> statements in the submodule will only be using 1 revision of any of the
> submodules or modules it can access.

This looks like your wishful thinking - the CLR in section 7.1.5 is totally 
unqualified, and my interpretation is that it’s meant for the text of a single 
module or submodule because an import in one submodule doesn’t apply to other 
submodules nor the main module. Moreover, I can easily combine *groupings and 
typedefs* from multiple revisions. For everything else, including XPath, the 
revisions are irrelevant.

Lada

> 
> 
>> 
>> /martin
> 
> Andy

--
Ladislav Lhotka, CZ.NIC Labs
PGP Key ID: E74E8C0C




_______________________________________________
netmod mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/netmod

Reply via email to