Re: GNU Classpath Linuxtag meeting minutes

[Sascha Brawer]

Mark Wielaard <[EMAIL PROTECTED]> wrote on Mon, 14 Jul 2003 00:46:08 +0200:

> http://www.dandelis.ch/development/fonts/

I've updated the page to reflect the current state. I did not bother
updating the code snapshot, since I'd like to clean up things and bring
my implementation a bit further. If you want the code right now, please
write me.
 
>  Tim Tyler his public domain, pure java font rendering program was
>  mentioned 

I've written to Tim, and he said he would not mind us using his code.
I'll make sure that the legal issues get resolved; Mark receives a copy
of everything.


>- Aicas is working on some RMI fixups.

In the context of RMI, maybe not everyone knows that there also exists a
variant which does RMI over IIOP. It uses the CORBA protocol for RMI, but
without using CORBA IDL. Actually, CORBA had to be extended to cope for
this, but it now is part of the standard. Of course, I don't know whether
this is relevant for Aicas -- in any case, [1] has more information. But
implementing RMI-IIOP seems like a big task, unless parts of some free
Java ORB can be used. IMHO, the main advantage is that IIOP is a better
protocol, and that it is possible to replace either client or server by
something not written in Java. There exist CORBA bindings for relatively
exotic languages, including Lisp, Ada, Objective C, or Smalltalk. There
also is an LGPLed project for IIOP on .NET [2].


>- We should have a standard NotYetImplementedException. Just returning
>  something random (like null) from stubs is really not
>  acceptable. Kaffe for example has one.

What about java.lang.UnsupportedOperationException?


>- We really have a complete framework now to run significatly large
>  applications. Getting semi-free application written in the java
>  language (but which currently uses a proprietary runtime
>  environment) running with a free VM is a great way to test and/or
>  expand our classes and keeps us focused on the needs of free software
>  developers.

Georg Greve said at LinuxTag that he would not mind including a text
about Classpath in one of the next issues of Brave GNU World.  I've heard
that [3] was perceived by some people as a statement that Free Software
developers should stay away from Java, although the text doesn't say
this. IMHO, we'd best send him a draft text for the column -- any
volunteers? I think this would be important.

[1] http://java.sun.com/products/rmi-iiop/
[2] http://www.gnu.org/brave-gnu-world/issue-49.en.html
[3] http://iiop-net.sourceforge.net/

Best,

-- Sascha

Sascha Brawer, [EMAIL PROTECTED], http://www.dandelis.ch/people/brawer/ 




___
Classpath mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/classpath

Re: GNU Classpath Linuxtag meeting minutes

[Tom Tromey]

> "Sascha" == Sascha Brawer <[EMAIL PROTECTED]> writes:

Sascha> But implementing RMI-IIOP seems like a big task, unless parts
Sascha> of some free Java ORB can be used.

I haven't looked at RMI-IIOP, but I do know that there are at least
two free Java ORBs out there.  One is OpenORB.  I forgot the name of
the other.

Occasionally I've tinkered with putting OpenORB into rhug, but the
build system is complex and it is a fair amount of work.  For just
building the .jar files, though, I think we should be there already.

The other ORB, whatever it is, once built with gcj.

Tom


___
Classpath mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/classpath

Re: GNU Classpath Linuxtag meeting minutes

[Per Bothner]

Mark Wielaard wrote:

- We should have a standard NotYetImplementedException. Just returning
  something random (like null) from stubs is really not
  acceptable. Kaffe for example has one.
In the gcj list I suggested UnsupportedOperationException, with a 
refinement:

throw new UnsupportedOperationException (REASON);
where REASON is a string literal that includes "not implemented"
and all on a single line (to make it easy to grep for).
See http://gcc.gnu.org/ml/java/2003-03/msg00016.html
--
--Per Bothner
[EMAIL PROTECTED]   http://per.bothner.com/


___
Classpath mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/classpath

Re: GNU Classpath Linuxtag meeting minutes

[Dalibor Topic]

Tom Tromey wrote:
"Sascha" == Sascha Brawer <[EMAIL PROTECTED]> writes:


Sascha> But implementing RMI-IIOP seems like a big task, unless parts
Sascha> of some free Java ORB can be used.
I haven't looked at RMI-IIOP, but I do know that there are at least
two free Java ORBs out there.  One is OpenORB.  I forgot the name of
the other.
Occasionally I've tinkered with putting OpenORB into rhug, but the
build system is complex and it is a fair amount of work.  For just
building the .jar files, though, I think we should be there already.
The other ORB, whatever it is, once built with gcj.
the other ORB I know is Jacorb. See http://www.jacorb.org/ for more 
information. One day I'll try to integreate one of them into kaffe.

cheers,
dalibor topic


___
Classpath mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/classpath

Re: GNU Classpath Linuxtag meeting minutes

[Andy Walter]

On Monday 14 July 2003 18:08, Per Bothner wrote:
> Mark Wielaard wrote:
> > - [NotYetImplementedException]
>
> In the gcj list I suggested UnsupportedOperationException, with a
> refinement:
>
> throw new UnsupportedOperationException (REASON);
> where REASON is a string literal that includes "not implemented"
> and all on a single line (to make it easy to grep for).

What about subclassing UnsupportedOperationException like this:

--- cut ---

package gnu.org.classpath;

public class NYIException extends UnsupportedOperationException {
  static String nyi =
"This method has not yet been implemented in GNU Classpath. Please " +
"contact [EMAIL PROTECTED] if you want to use it.";

  public NYIException() {
super(nyi);
  }

  public NYIException(String s) {
super(s + "\n" + nyi);
  }
}

--- cut ---

Thus, we make sure the user always receives a helpful description and it is 
easy for VM implementations to replace the contact information by whatever 
else. I prefer grepping for class names to grepping for strings because there 
are no undetected typos in class names.


Cheers,

Andy.

-- 
aicas GmbH   /"\  ASCII Ribbon Campaign
Haid-und-Neu-Straße 18 * 76131 Karlsruhe \ /  No HTML or RTF in mail
http://www.aicas.com  X   No MS-Word in mail
Tel: +49-721-663 968-24; Fax: +49-721-663 968-94 / \  Respect Open Standards



___
Classpath mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/classpath

Re: GNU Classpath Linuxtag meeting minutes

[Dalibor Topic]

Andy Walter wrote:
On Monday 14 July 2003 18:08, Per Bothner wrote:

Mark Wielaard wrote:

- [NotYetImplementedException]
In the gcj list I suggested UnsupportedOperationException, with a
refinement:
throw new UnsupportedOperationException (REASON);
where REASON is a string literal that includes "not implemented"
and all on a single line (to make it easy to grep for).


What about subclassing UnsupportedOperationException like this:

Thus, we make sure the user always receives a helpful description and it is 
easy for VM implementations to replace the contact information by whatever 
else. I prefer grepping for class names to grepping for strings because there 
are no undetected typos in class names.
I agree with Andy, classes are nicer, since the compiler checks that the 
names are actually correct. But since we also talked about attracting 
new contributors on our meeting, I'd like to propose a sightly modified 
message string:

 static String nyi =
"This method has not yet been implemented in GNU Classpath.\n\n"
+ " The GNU Classpath project would appreciate receiving"
+ " an implementation for it.\n"
+ " If you intend to contribute, please take a look at"
+ " http://www.gnu.org/software/classpath/docs/hacking.html and"
+ " get in touch with other developers on [EMAIL PROTECTED]";
I hope that we could get more contributors fixing small missing bits 
that way. I guess people who dearly want to use a missing feature 
shouldn't be that hard to convince to implement it ;)

cheers,
dalibor topic


___
Classpath mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/classpath

Re: GNU Classpath Linuxtag meeting minutes

[Andy Walter]

On Tuesday 15 July 2003 15:36, Dalibor Topic wrote:
> >>>- [NotYetImplementedException]
>   static String nyi =
>   "This method has not yet been implemented in GNU Classpath.\n\n"
>   + " The GNU Classpath project would appreciate receiving"
>   + " an implementation for it.\n"
>   + " If you intend to contribute, please take a look at"
>   + " http://www.gnu.org/software/classpath/docs/hacking.html and"
>   + " get in touch with other developers on [EMAIL PROTECTED]";

Thanks, Dalibor. This message string seems more suitable for GNU Classpath 
than my proposal.


Cheers,

Andy.

-- 
aicas GmbH   /"\  ASCII Ribbon Campaign
Haid-und-Neu-Straße 18 * 76131 Karlsruhe \ /  No HTML or RTF in mail
http://www.aicas.com  X   No MS-Word in mail
Tel: +49-721-663 968-24; Fax: +49-721-663 968-94 / \  Respect Open Standards



___
Classpath mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/classpath

Re: GNU Classpath Linuxtag meeting minutes

[Andy Walter]

On Monday 14 July 2003 14:39, Sascha Brawer wrote:
> Mark Wielaard <[EMAIL PROTECTED]> wrote on Mon, 14 Jul 2003 00:46:08 +0200:
> >- Aicas is working on some RMI fixups.
>
> In the context of RMI, maybe not everyone knows that there also exists a
> variant which does RMI over IIOP. It uses the CORBA protocol for RMI, but
> without using CORBA IDL.
> [...] http://java.sun.com/products/rmi-iiop/

Thanks for the link, Sascha. For the time being, we'd like to get standard RMI 
running with GNU Classpath. RMI IIOP might be an interesting approach after 
that.


Cheers,

Andy.

-- 
aicas GmbH   /"\  ASCII Ribbon Campaign
Haid-und-Neu-Straße 18 * 76131 Karlsruhe \ /  No HTML or RTF in mail
http://www.aicas.com  X   No MS-Word in mail
Tel: +49-721-663 968-24; Fax: +49-721-663 968-94 / \  Respect Open Standards



___
Classpath mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/classpath