[825] in java-interest
Undeliverable Message
daemon@ATHENA.MIT.EDU (MAILER-DAEMON@bangate.compaq.com)
Fri Jul 21 11:55:21 1995
Date: Fri, 21 Jul 95 9:12:45 CDT
To: <java-interest@java.sun.com>
From: <MAILER-DAEMON@bangate.compaq.com>
To: <java-interest-digest@java.sun.com>
Cc:
Subject: java-interest-digest V1 #93
Message not delivered to recipients below. Press F1 for help with VNM
error codes.
VNM3043: Erik Sawyer@TS SysCall@CS Hou
VNM3043 -- MAILBOX IS FULL
The message cannot be delivered because the
recipient's mailbox contains the maximum number of
messages, as set by the system administrator. The
recipient must delete some messages before any
other messages can be delivered.
The maximum message limit for a user's mailbox is
10,000. The default message limit is 1000 messages.
Administrators can set message limits using the
Mailbox Settings function available in the
Manage User menu (MUSER).
When a user's mailbox reaches the limit, the
user must delete some of the messages before
the mailbox can accept any more incoming messages.
---------------------- Original Message Follows ----------------------java-interest-digest Friday, 21 July 1995 Volume 01 : Number 093
In this issue:
Re: Native Tk
native OpenGL
re: throw specs in functions
problems with windowing
Execution Environment in "interpreter"
----------------------------------------------------------------------
From: Markus Buchhorn <markus@octavia.anu.edu.au>
Date: Fri, 21 Jul 1995 08:22:54 +1000
Subject: Re: Native Tk
[Tk written in C ?]
>It is, although you need parts of Tcl as well to make Tk work. It's not (yet?)
>fully seperable from Tcl.
There is a library called pTk which is standalone, and very useful for hooking
into other languages. At the moment it is very nicely hooked into Perl 5.
Try getting in touch with the Perl Tk author: Nick.Ing-Simmons@tiuk.ti.com
as I'm not sure what's involved here. Nick has mentioned other languages
to be worked on later...
Cheers,
Markus
Markus Buchhorn, Parallel Computing Research Facility
email = markus@octavia.anu.edu.au snail = CISR, I Block, OAA, ANU
Australian National University, Canberra, 0200 , Australia.
[International = +61 6, Australia = 06] [Phone = 2492930, Fax = 2490747]
------------------------------
From: dswain@condor.citi.doc.ca (Duncan Swain)
Date: Thu, 20 Jul 1995 19:06:06 -0400
Subject: native OpenGL
I have been trying unsuccessfully to call native methods written in C which in turn make OpenGL library calls. As an example, consider the following code which is built (along with it's stub) into a dynamic library called libogltest.so: (this example works fine when the GL code is replaced with a simple printf() call)
#include "StubPreamble.h"
#include "javaString.h"
#include "project_OpenGLTest.h"
#include <GL/gl.h>
#include "aux.h"
void project_OpenGLTest_openWindow(struct Hproject_OpenGLTest *this)
{
auxInitDisplayMode (AUX_SINGLE | AUX_RGB);
auxInitPosition (0, 0, 500, 500);
auxInitWindow ("OGL test");
glClearColor (0.0, 0.0, 0.0, 0.0);
glClear(GL_COLOR_BUFFER_BIT);
sleep (10);
}
ld.so.1: /usr/ncis/hotjava/bin/sun4/java: fatal: relocation error: symbol not found: auxInitDisplayMode: referenced in /usr/ncis/MyApp/lib/libogltest.so
My java class looks like this: (libGL.so, libaux.so, and libogltest.so are in my LD_LIBRARY _PATH)
class OpenGLTest {
static {
Linker.loadLibrary("GL"); <<<<<???
Linker.loadLibrary("aux"); <<<<<???
Linker.loadLibrary("ogltest");
}
public native void openWindow();
}
How am I supposed to dynamically link to the GL libs?
------------------------------
From: pelegri@calterra.Eng.Sun.COM (Eduardo Pelegri-Llopart)
Date: Thu, 20 Jul 1995 21:00:06 -0700
Subject: re: throw specs in functions
I'd also like to have throw specs in function specs. In addition to
allowing additional type-checking, an implementation might be able to
take advantage of the extra information to generate better code. That
is certainly the case for C++ exceptions (with the latest semantics)
and a PC-range based implementation of exceptions.
- eduardo
------------------------------
From: furm@npac.syr.edu (Wojtek Furmanski)
Date: Fri, 21 Jul 1995 00:21:18 -0400
Subject: problems with windowing
I found the following two problems with hotjava windowing where I would
appreciate some help/hints:
1) My applet tries to create an editor in a new Frame that contains a MenuBar
and a Window, which hosts a TextArea. One of MenuItems, called 'File',
starts a FileDialog which reads a file and tries to display it using
textArea.setText. This doesn't seem to work as the TextArea ignores
my attempts to (re)set anything after the Frame is mapped. I tried
various linear combinations of map/unMap/dispose applied to
Frame/Window/TextArea and the only way I found so far to get a new file
displayed is to dispose and recreate the whole frame on each File->Open
request - which is hardly an acceptable solution as the frame disappears
then each time for a while.
2) My other applet tries to customize MenuBar in the HotJava frame by
adding some applet specific items in the start() method and disposing
them in the stop() method. These new items work fine but the whole
construct is rather unstable and getting worse when moving from alpha2
to alpha3. In alpha2, hotjava dies when I try to Reload this applet.
In alpha3, hotjava also dies when I try to move to another applet.
.hotjava/weblog reports segmentation fault which has likely something
to do with disposing my custom menu items. I use the same items here
as in example 1) where they are also disposed in stop() (together with
the whole frame) and don't create any segmentation problems there.
Thanks,
Wojtek Furmanski
NPAC, Syracuse University
------------------------------
From: nataraj@cat.syr.edu
Date: Fri, 21 Jul 95 01:24:57 -0400
Subject: Execution Environment in "interpreter"
Hi,
I need someone to clarify some doubts re: the way java code is
interpreted by the "interpreter" and the neccesary environement to be
set up.
background:
What we are trying to do, is to implement "remote objects" by using a
spl keyword (this is more like a compiler project than a major work in
java). On seeing the 'remote' keyword, those objects should be
created on a remote machine and the methods invoked on those objects
(later in the code) got to be executed in the remote machine.
eg.
a = remotenew A(); //create 'a' ona remote machine R
a.foo(); // foo(0 is invoked on the object 'a' on remote machine
Clarifications:
Our understanding from the code is (as far as we can browse and 'look-thru' the
code) :
On a new object creation there are 2 scenerio:
1. The object is the first of its class
In this case, a new ExecEnv ( ee = EE() ) is created;
2. There is/are other object(s) of the same type already created
In this case, the new object is added to the pool (and a ref to
that object is stored)
Now for our "remote" execution to work ..the MAJOR factors as far as we
understand are
1. Exec Env to be either created (if our above assumptions are right)
or the existing one should be passed on (which will be tough 8-( )
2. Constant Pool (cb) for access to objects for "method invocation" etc
3. "pc" should be passed on and result accordingly managed into the
"pc" of the primary host.
Is our idea about "ee" right? If so (or if not) any hints/pointers to
"How to simulate a 'ee' on the 'remote host'" ??
So any information about this "ee" as to the
a. interpreter
b. machine (address space)
c. context (already created objects/methods invoked)
is appreciated,
thanks,
- -Nataraj
------------------------------
End of java-interest-digest V1 #93
**********************************
-
Note to Sun employees: this is an EXTERNAL mailing list!
Info: send 'help' to java-interest-request@java.sun.com
