On Thu, Apr 24, 2008 at 8:49 AM, Simon Laws <[EMAIL PROTECTED]>
wrote:
>
>
> On Thu, Apr 24, 2008 at 8:08 AM, ant elder <[EMAIL PROTECTED]> wrote:
>
> > I'm going to close TUSCANY-2145 as working as designed in a few days
> > unless
> > someone says otherwise.
> >
> > I still get the feeling that there is actually no way for
> > o.a.t.s.host.imbedded.impl.DefaultSCADomain to find any new services at
> > invocation time so it could just throw the exception as suggested in the
> > JIRA, but it seems harder to prove that than to just go with the flow
> > and
> > keep the current behaviour.
> >
> > ...ant
> >
> > On Sun, Apr 13, 2008 at 3:35 PM, Yang Lei <[EMAIL PROTECTED]> wrote:
> >
> > > I think we should keep the following piece. As the comment
> > > said,"create a remote service ref" , so it can delegate to the binding
> > > implementation to handle remote case.
> > >
> > > // Lookup the component in the domain
> > > Component component =
> > componentManager.getComponent(componentName);
> > > if (component == null) {
> > > // The component is not local in the partition, try to
> > > create a remote service ref
> > > return createServiceReference(businessInterface, name);
> > > }
> > >
> > > I agree if we already find the component ( component !=null case), we
> > > should check if the service is defined on the component and throw the
> > > service not found exception if it is not defined. I think the current
> > > implementation already does it in the ComponentContextImpl...
> > >
> > > Yang.
> > >
> > > On Thu, Apr 10, 2008 at 4:17 AM, Wang Feng <[EMAIL PROTECTED]> wrote:
> > > > +1 throw an exception.
> > > >
> > > > The scenario like this class.getMethod(methodName).
> > > > If a matching method is not found,it will throw
> > NoSuchMethodException.
> > > >
> > > > Thanks,
> > > > Wang Feng
> > > >
> > > >
> > > > On 2008-04-10,ant elder <[EMAIL PROTECTED]> wrote:
> > > >
> > > > >TUSCANY-2145 asks about SCADomain.getService () returning a proxy
> > even
> > > when
> > > > >the service doesn't exist, but looking back through the SVN history
> > it
> > > looks
> > > > >like this is intentional. Before I close the JIRA does anyone have
> > any
> > > > >comments on if this is/isn't the correct behaviour?
> > > > >
> > > > > ...ant
> > > > >
> > > >
> > > >
> > > >
> > ---------------------------------------------------------------------
> > > > To unsubscribe, e-mail: [EMAIL PROTECTED]
> > > > For additional commands, e-mail: [EMAIL PROTECTED]
> > > >
> > > >
> > >
> > > ---------------------------------------------------------------------
> > > To unsubscribe, e-mail: [EMAIL PROTECTED]
> > > For additional commands, e-mail: [EMAIL PROTECTED]
> > >
> > >
> >
>
> Hi Ant
>
> I agree with the sentiment that given the way that we have the code in
> Tuscany organized at the moment there is no way for DefaultSCADomain to find
> any new service at invocation time. However in the past we have have had
> Nodes that have allowed this and I'm currently changing the builder code [1]
> to allow for this more generally in the future. So I agree we should just
> close this JIRA with the expectation that, in the future, returning a
> reference that is not, yet, resolved is a useful feature.
>
> Simon
>
oops, forgot the reference which talks about some experiments to make the
binding of references more flexible. I'm not done just yet but I hope to
check some code in shortly that we can take a look at and discuss to see if
we think it's a sensible thing to do i.e. move from a brainstorm into a
concrete feature.
[1] http://www.mail-archive.com/tuscany-dev@ws.apache.org/msg30805.html
