[1327] in java-interest
Re: "perform:" and Java
daemon@ATHENA.MIT.EDU (Charles L. Perkins)
Wed Aug 30 08:56:06 1995
Date: Wed, 30 Aug 95 05:40:42 -0400
From: "Charles L. Perkins" <clp@home.HarvardSq.com>
To: java-interest@java.sun.com
> From: David Griffiths <dave@prim.demon.co.uk>
> Date: Tue, 29 Aug 95 20:05:02 +0100
> Subject: Re: "perform:" and Java
> /* Call Java method from C. The java method may have any number of arguments
> * and may return a single 4 byte value which may be an integer, a pointer to
> * an object handle, whatever.
> long
> execute_java_dynamic_method(ExecEnv *ee, HObject *obj,
> char *method_name, char *signature, ...)
> So couldn't you use this to add a native perform method to Object? It
> appears as though the Tcl stuff makes use of it.
Yep, I noticed this, too. It doesn't help, since using a native method will be
(a) less secure (some people won't allow it), (b) non-standard (it's not Java).
[This also means, by the way, that using TCL has these same two drawbacks.]
I started this thread a while back to generate some feedback from the Sun/Java
guys on their designs choices, but I was also hoping to stimulate some interest
in the mailing list.
Is it really the case that *none* of you miss "perform:" enough to speak up?
Let's get a show of virtual hands....
Charles L. Perkins
Virtual Rendezvous
-
Note to Sun employees: this is an EXTERNAL mailing list!
Info: send 'help' to java-interest-request@java.sun.com
