Hi,

we have to specify a subset of schema node types that are permitted as
candidates for mount points. The proposals were seen so far:

1. container and list
2. container, list, case, anydata
3. anydata only

I prefer #2 because it is closest to the restrictions on augment's
target node. I think that anydata should be allowed in any case, but
only the other hand I don't support #3 because

- anydata (as defined in 6020bis) makes the schema weaker: permitted
  child nodes aren't limited to those specified in mounted subschemas;

- according to the existing rules, old clients have to be prepared to
  face unknown data pretty much anywhere: for example, if a client only
  supports module A, and the server advertises A + B where B augments A,
  then the client may see data from B.

Lada

-- 
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