Hi Martin,
Please see inline [RW] ...
On 09/12/2015 15:13, Martin Bjorklund wrote:
Ladislav Lhotka <[email protected]> wrote:
On 09 Dec 2015, at 15:28, Martin Bjorklund <[email protected]> wrote:
Ladislav Lhotka <[email protected]> wrote:
or when augmenting a choice.
Sure, but this already works in YANG 1.0, since it technically doesn't
add any mandatory nodes.
module A:
choice foo {
...
}
module B:
augment "A:/foo" {
container top {
leaf bar { type empty; }
leaf baz { mandatory true; type empty; }
}
}
IMO this is also perfectly safe yet not allowed by your wording.
It is allowed even in YANG 1.
Note that there is a precise definition of "mandatory node", and note
that the augment really is:
[RW]
Agreed, but I find the distinction to be a bit subtle. I suspect that
quite a lot of readers are likely to think that a mandatory node is any
node with an associated "mandatory true" condition, and not make the
mental leap to the above example being OK.
Hence I would suggest that we call out this case as being explicitly
allowed either in 6020bis or perhaps more appropriately in 6087bis.
Thanks,
Rob
_______________________________________________
netmod mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/netmod
