On 4/19/07, ant elder <[EMAIL PROTECTED]> wrote:
That sounds good. I actually started doing that a while ago on the old trunk code and had a "dynamic" operation and a "dynamic" interface, but "any" seems like a better name. I got quite far but then I hit some issue that blocked progress, I can't remember what now though, i'll have to go dig up the old code again... ...ant On 4/19/07, Raymond Feng <[EMAIL PROTECTED]> wrote: > > Hi, Ant. > > The scripting support would be a good driving force for #3. A list of > requirements will be really helpful. > > Just for your information, I'm thinking of a special "any" operation > bounded > invocation chain on the wire to deal with the dynamic invocations. This > way, > we might be able to benefit both the static and dynamic invocation styles. > > Thanks, > Raymond > > ----- Original Message ----- > From: "ant elder" <[EMAIL PROTECTED]> > To: <[email protected]> > Sent: Thursday, April 19, 2007 7:21 AM > Subject: Re: Improve/simplify the runtime further > > > > On 4/18/07, Raymond Feng <[EMAIL PROTECTED]> wrote: > >> > >> Hi, > >> > >> I'm thinking of improving and simplifying the runtime further in the > >> following directions: > >> > >> 1. Have the Tuscany runtime to maintain the SCA domain composite and > >> refactor the DeployerImpl into the Domain service. > >> > >> 2. Simplify the relationship between runtime metadata > >> (o.a.t.spi.Component/Service/Reference) for the invocations and model > >> (o.a.t.assembly.*). > >> > >> 3. Improve the Wire/InvocationChain to support dynamic style invocation > >> (in > >> cases that we don't know or don't care about the interface/operation > >> until > >> runtime) in addtion to the operation-based invocation chain. > >> > >> 4. Implement the CallableReference/ServiceReference/CallbackReference. > >> > >> The list is not meant to be complete. Please feel free to add your > ideas. > > > > > > All those sound good to me, I'm particularly interested in 3 for the > > script > > implementation so I'd like to help with that one to make sure it does > what > > I > > need. Will have to wait a day or two while I get the script > implementation > > more fully functional. > > > > ...ant > > > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: [EMAIL PROTECTED] > >
Re. the implementation of the SCA Domain. I'm starting to think about how we get multiple nodes up in a runtime node again now that the module structure is starting to settle down. So I'd like to help out with with makeing sure the SCA domain is able to operate in the multi node scenarion. Regards Simon
