Chris Lauwers

We don't have support in TOSCA for "disabling" a feature (none of our refinement rules allow it) so I would be surprised if anybody would assume that "null" could be...

We could also borrow from YANG, and use the keyword `must`

@pmjordan I agree. Technically it should be `must_satisfy`. But since that is rather long, I would vote for the shorter `must`

I added the following test case that illustrates the problem: https://github.com/oasis-open/tosca-community-contributions/tree/master/tests/tosca_2_0/profile_versions

I completely agree. TOSCA should support (semantic) versioning of profiles and service templates.

I'd like to revisit this issue during one of our upcoming TOSCA Language ad-hoc meetings. In particular, I'd like us to decide exactly what entities need to be (uniquely) named...

Thanks Philippe. I have corrected and fully validated versions of all (most?) of these examples. Perhaps we can have a brief discussion about the following: * How/where to organize these...

In the Ubicity implementation, the get_attribute function automatically returns a list of values when multiple requirements exist of the same name.

This renders correctly on github.com: https://github.com/lauwers/tosca-specs/blob/published/tosca_2_0/TOSCA-v2.0-csd06.md#641-common-keynames-in-type-definitions Unlike other tables in the document, this table does not have a blank line immediately preceding it. I added a blank line in the...