Niclas Hedhman wrote:
On Thu, Jul 23, 2009 at 1:29 AM, Gregg Wonderly<[email protected]> wrote:
The issue which I think is never fully considered is that lookup is based on
the Java type system, including complex types, and not based on string
matching. If it was just string matching, we'd have RE support now. But
there is no defined RE syntax for "derived from" or "implements X" etc. The
Java type system provides that.
Yes, that is the view of "Jini Zealot" ;-) and their argument is that
Jini Lookup can't do algorithms and arithmetic. And as long as either
side don't want to look for the converged solution, this status quo
will remain, just like when I first brought this up 2-3 years ago. And
Kriens & Co at OSGi will have the same 'we are so much better'
attitude as well...
As just another perspective point, I consider this the "everything useful must
be part of the platform" confusion. Everything useful needs to be possible with
a platform in the space the platform is "used". But everything that everyone
finds useful just can't be part of the platform.
When I rant about programming languages reinventing the wheel and only syntax
being used to do something different that can be expressed in other languages, I
think of the same narrow minded view. Anything that is not provided "natively"
in a language or platform, either can't be done, or requires that you right some
code to do it. We are still at such low coding levels with such powerful
languages, that "can't be done" just doesn't come out as the point of contention.
And as I said in my other long winded response, in this case, someone can write
a service that provides such a lookup mechanism and encapsulate such logic
there. It would make most sense for such a service to use the ServiceRegistrar
Entry data to perform the searches.
My changes to Reggie (http://reef.dev.java.net) for "delayed unmarshalling"
would make it possible to then send the "unmarshalled" service data to the
caller as the results so that the codebase would remain uncontaminated etc.
I will conceded that a multi-layer service lookup is not really possible without
having access to the unmarshalled service data to be forwarded to the calling party.
Gregg Wonderly
