Re: [YANG] 7.8.3 unique statement

Andy Bierman <ietf@andybierman.com> wrote:
> Martin Bjorklund wrote:
> >> It is also OK to apply the default once.
> > 
> > So you would allow one instance with the default, and the rest has to
> > have values set?
> > 
> 
> no -- the value specified by the default can be used once.
> If the leaf is not provided, and its value has not been used
> yet, then it is OK to apply the default and still be unique.

I think that this translates to treating a leaf that has its default
the same way as a non-existing leaf (in the unique test).

> >> <tangent>
> >> One augments concern: what does it mean to declare conformance to a module?
> >> If module FOO-MIB includes X and Y, but the vendor also has module
> >> MY-FOO-MIB that adds a mandatory Z, does that vendor really conform to FOO-MIB?
> >> A manager that only knows about FOO-MIB and not MY-FOO-MIB will not work,
> >> so IMO the agent MUST NOT advertise that it supports FOO-MIB.
> >> </tangent>
> > 
> > But that won't work.  If I impplement FOO-MIB, how do I know that
> > there is some other module somewhere that augments it?  I think we
> > have to restrict the augment to not augment with any mandatory nodes.
> > 
> 
> If the manager sees FOO-MIB in the schema-discovery or capabilities,
> then it will assume this MIB can be used.  No other modules or
> capabilities can override the contract that is implied by the
> definitions in FOO-MIB.  Period.  So MY-FOO-MIB MUST NOT
> add anything that breaks the FOO-MIB contract, if the vendor
> wants to claim conformance to FOO-MIB.

Exactly.

/martin

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