I agree that ideally it could be inherited from the parent node type if not
specified. That would indeed be my opinion.

Small quibble: the fact that it's called "Standard" is arbitrary. The
interface name does *not* have to conform to the interface type, it's just
what they decided for the normative Root node type. The name does *not*
contain type information. For example, in your own node type you can
override the interface named "Standard" to be of a different interface
type, which does not have "Standard" in its name. (ARIA will insist that
your overriding type inherits from the Standard node type -- see our other
discussion about type inheritance.)

On Tue, Nov 28, 2017 at 12:32 PM, DeWayne Filppi <[email protected]>
wrote:

> Seems rather ugly.  Do you have an opinion?  If I specify that I'm using
> the Standard interface, what point is there in the "type" field?
>
> On Tue, Nov 28, 2017 at 10:27 AM, Tal Liron <[email protected]> wrote:
>
> > I mentioned this to you in the previous thread: the "type" field is
> > required for interface definitions according to TOSCA syntax. So, even if
> > it's the same as what you are inheriting, you must specify it.
> >
> > On Tue, Nov 28, 2017 at 12:04 PM, DeWayne Filppi <[email protected]>
> > wrote:
> >
> > > Now that the 'subclassing' problem has been resolved,  overriding
> > interface
> > > methods is breaking.  Simple example:
> > >
> > > tosca_definitions_version: tosca_simple_yaml_1_0
> > >
> > > imports:
> > >
> > >   - aria-1.0
> > >
> > > node_types:
> > >
> > >   T1:
> > >     derived_from: tosca.nodes.Root
> > >     interfaces:
> > >       Standard:
> > >         create:
> > >           implementation:
> > >             primary: i1.sh
> > >         delete:
> > >           implementation:
> > >             primary: i1.sh
> > >
> > > The error, using Aria in the ARIA-1 branch:
> > >
> > > Validation issues:
> > >   2: required field "type" in
> > > "aria_extension_tosca.simple_v1_0.definitions.InterfaceDefinition"
> does
> > > not
> > > have a value
> > >
> >
>

Reply via email to