Re: Java and Objective-C

FROM : Jean-Daniel Dupas
DATE : Sun Jun 08 13:05:00 2008

Le 8 juin 08 à 12:43, Michael Ash a écrit :

> On Sun, Jun 8, 2008 at 1:48 AM, Peter Duniho <<email_removed>> wrote:

>> So, when you write "true proxying of method invocations", what does 
>> that
>> mean, exactly?

>
> Distributed Objects is probably the best example in terms of
> real-world use of a technique which is difficult in stricter
> languages. Consider:
>
> Foo *obj = [connection rootProxy];
> [obj doSomethingWithArgument:arg];
>
> This transparently proxies the method invocation off to another
> machine. I seem to recall from the last time that this is possible to
> implement in C# but not Java, but I could be wrong.
>
> And as I pointed out in the last discussion, a key point is not only
> whether this sort of thing is possible but whether it is easy. Message
> capturing and forwarding is fairly trivial in ObjC, which makes it
> practical for the end programmer to write his own code which takes
> advantage of it, rather than using system APIs which use it. It's very
> nice to be able to compose two objects together with forwarding using
> only a couple of lines of code, or to capture an invocation for later
> use with a similar amount of code.
>
> Mike


Of couse, it took me only a couple of line to implements my own 
interthread messaging proxy.

I don't think this is worth restarting this long and boring discussion 
about what are the advantage of Obj-C runtime.