From MAILER-DAEMON Sun Aug 01 12:03:28 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BrIo4-00018g-IY
	for mharc-classpath@gnu.org; Sun, 01 Aug 2004 12:03:28 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BrIo2-00017l-1b
	for classpath@gnu.org; Sun, 01 Aug 2004 12:03:26 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BrIo0-00017C-Tb
	for classpath@gnu.org; Sun, 01 Aug 2004 12:03:25 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BrIo0-000177-Rr
	for classpath@gnu.org; Sun, 01 Aug 2004 12:03:24 -0400
Received: from [139.165.32.78] (helo=serv09.segi.ulg.ac.be)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BrIkN-0008Nd-UC
	for classpath@gnu.org; Sun, 01 Aug 2004 11:59:40 -0400
Received: (qmail 1896 invoked by uid 504); 1 Aug 2004 17:59:38 +0200
Received: from arnaud.vandyck@ulg.ac.be by serv09.segi.ulg.ac.be by uid 501
	with qmail-scanner-1.16 (spamassassin: 2.55.  Clear:. 
	Processed in 0.61909 secs); 01 Aug 2004 15:59:38 -0000
Received: from unknown (HELO oz.fapse.ulg.ac.be) ([139.165.201.14])
	(envelope-sender <arnaud.vandyck@ulg.ac.be>)
	by serv09.segi.ulg.ac.be (qmail-ldap-1.03) with SMTP
	for <kaffe@kaffe.org>; 1 Aug 2004 17:59:37 +0200
Received: from arnaud by oz.fapse.ulg.ac.be with local (Exim 4.34)
	id 1BrIkK-000161-IP; Sun, 01 Aug 2004 17:59:36 +0200
To: Kaffe Mailing List <kaffe@kaffe.org>
From: Arnaud Vandyck <avdyk@debian.org>
X-Face: yuSM.z0$PasG_!+)P;
	ugu5P+@#JEocHIpArGcQZ^hcGos8:DBJ-tfTQYWyf`$2r0vfaoo7F| h.;
	Agl'@x8v]?{#ZLQDqSB:L^6RXGfF_fD+G9$c:)p<yycF[Da]*=*
Mail-Followup-To: Kaffe Mailing List <kaffe@kaffe.org>, ClassPath List
	<classpath@gnu.org>, Alan Tam <Tam@SiuLung.com>
Date: Sun, 01 Aug 2004 17:59:36 +0200
Message-ID: <87d62argrb.fsf@oz.fapse.ulg.ac.be>
User-Agent: Gnus/5.1006 (Gnus v5.10.6) Emacs/21.3 (gnu/linux)
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="=-=-="
Sender: Arnaud Vandyck <arnaud.vandyck@ulg.ac.be>
Cc: Alan Tam <Tam@SiuLung.com>, ClassPath List <classpath@gnu.org>
Subject: Building log4j with free compiler (was: [Alan Tam] Re: IMPORTANT
 [RFS] liblog4j1.2-java build from sources)
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Sun, 01 Aug 2004 16:03:26 -0000

--=-=-=

System.out.println("Hi!");

I've tried to build log4j with free compilers but it fails. Alan Tam
investigate a lot more and here is the result.

Thanks to him,

Regards,


--=-=-=
Content-Type: message/rfc822
Content-Disposition: inline

Old-Return-Path: <Tam@SiuLung.com>
Gnus-Warning: This is a duplicate of message <410A2B7A.9060608@SiuLung.com>
Message-ID: <410A2B7A.9060608@SiuLung.com>
Date: Fri, 30 Jul 2004 19:05:30 +0800
From: Alan Tam <Tam@SiuLung.com>
User-Agent: Mozilla Thunderbird 0.7+ (Windows/20040722)
To: Arnaud Vandyck <avdyk@debian.org>
Cc: debian-java@lists.debian.org, Benoit Joly <benoit@debian.org>,
	221236@bugs.debian.org
Subject: Re: IMPORTANT [RFS] liblog4j1.2-java build from sources
References: <87d62gmhn9.fsf@oz.fapse.ulg.ac.be>
In-Reply-To: <87d62gmhn9.fsf@oz.fapse.ulg.ac.be>
Resent-Message-ID: <_rLRkD.A.vVG.QuiCBB@murphy>
Resent-From: debian-java@lists.debian.org
Resent-Sender: debian-java-request@lists.debian.org
Resent-Date: Fri, 30 Jul 2004 06:05:53 -0500 (CDT)
Lines: 88
MIME-Version: 1.0

Hi,

On 2004-07-28 20:38, Arnaud Vandyck wrote:

>[...]
>
>So I built log4j with sablevm, ant1.6 and jikes... but it fails. I don't
>know why but I've been unable to 'exclude' classes to build. So at the
>moment, log4j is built using jikes-classpath (oups, I forgot 'classpath'
>dependency) and fastjar... and all classes are not built!
>
>This is the line to find all the sources to be built:
>  files = $(shell find src/java -name \*.java \
>            | grep -v chainsaw | grep -v lf5 \
>            | grep -v jms | grep -v jmx | grep -v JMS)
>
>And this is the lines two install the package:
>  deb http://people.debian.org/~avdyk/debian ./
>  deb-src http://people.debian.org/~avdyk/debian ./
>  
>
I have just tested building log4j1.2-1.2.8 with sablevm, ant1.6 and 
jikes. Only 2 build problems occurred:

1. lf5/viewer failed to build because of the following missing methods:
   - javax.swing.JTable.getRowHeight
   - javax.swing.JTable.sizeColumnsToFit
   - javax.swing.JTable.setRowHeight
   - javax.swing.JTextArea.setWrapStyleWord
   - javax.swing.JTree.setCellEditor
   - javax.swing.JTree.setShowsRootHandles
   - javax.swing.JTree.expandPath
   - javax.swing.JTree.collapsePath
   - javax.swing.JTree.getPathForLocation
   - javax.swing.JTree.getRowBounds
   - javax.swing.JTree.getRowCount
   - javax.swing.JTree.isExpanded
in
   - 
src/java/org/apache/log4j/lf5/viewer/categoryexplorer/CategoryExplorerTree.java
   - 
src/java/org/apache/log4j/lf5/viewer/categoryexplorer/CategoryImmediateEditor.java
   - 
src/java/org/apache/log4j/lf5/viewer/categoryexplorer/CategoryNodeEditor.java
   - 
src/java/org/apache/log4j/lf5/viewer/configure/ConfigurationManager.java
   - src/java/org/apache/log4j/lf5/viewer/LF5SwingUtils.java
   - src/java/org/apache/log4j/lf5/viewer/LogBrokerMonitor.java
   - src/java/org/apache/log4j/lf5/viewer/LogTable.java
This should be solved if lf5/viewer (or the whole lf5, which was 
introduced in version 1.2.5) is removed from the package, maybe moved to 
a second package.

2. rmic fails even after set to use the one from kaffe. But this affects 
the RMI example only, which is safe to ignore.

>[...]
>
>So here are the possible way (if there are more, thanks to comment):
>
>. log4j stays in free but without some packages (chainsaw, lf5, jmx and
>  jms)
>
>. log4j goes back to contrib and will be built with non-free JVM's
>
>. we make two packages available.
>
>Thanks for your time and comments,
>  
>
Given "This version of LogFactor5 is not completely integrated with 
log4j [1], I suppose we can first remove the offending packages to fix 
the rc bug and after the problem it solved, it is split into a second 
package.

-- 
Regards,
Alan

[1] http://logging.apache.org/log4j/docs/api/org/apache/log4j/lf5/package-summary.html


-- 
To UNSUBSCRIBE, email to debian-java-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org




--=-=-=



-- 
  .''`. 
 : :' :rnaud
 `. `'  
   `-    

--=-=-=--



From MAILER-DAEMON Sun Aug 01 19:00:34 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BrPJi-0003qp-Ec
	for mharc-classpath@gnu.org; Sun, 01 Aug 2004 19:00:34 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BrPJg-0003qk-GM
	for classpath@gnu.org; Sun, 01 Aug 2004 19:00:32 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BrPJf-0003qX-It
	for classpath@gnu.org; Sun, 01 Aug 2004 19:00:32 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BrPJe-0003qU-Se
	for classpath@gnu.org; Sun, 01 Aug 2004 19:00:31 -0400
Received: from [82.161.94.186] (helo=gnu.wildebeest.org)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BrPFt-0005jV-UK
	for classpath@gnu.org; Sun, 01 Aug 2004 18:56:38 -0400
Received: from elsschot.wildebeest.org ([192.168.1.26])
	by gnu.wildebeest.org with esmtp (Exim 3.35 #1 (Debian))
	id 1BrPFO-0006yc-00
	for <classpath@gnu.org>; Mon, 02 Aug 2004 00:56:06 +0200
From: Mark Wielaard <mark@klomp.org>
To: classpath@gnu.org
Content-Type: multipart/signed; micalg=pgp-sha1;
	protocol="application/pgp-signature";
	boundary="=-yueFvLxvHcvAi0fy0WY9"
Message-Id: <1091400952.11966.189.camel@localhost>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.4.6 
Date: Mon, 02 Aug 2004 00:55:53 +0200
Subject: GNU Classpath GUI projects (AWT/Swing)
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Sun, 01 Aug 2004 23:00:32 -0000


--=-yueFvLxvHcvAi0fy0WY9
Content-Type: text/plain
Content-Transfer-Encoding: quoted-printable

Hi all,

After all the GUI (AWT/Swing) work going on in CVS it is probably a good
idea to give a little overview of the process.

Thomas Fitzsimmons gave a very inspiring talk on "GCJ and the Desktop"
at the Desktop Developers' Conference describing our AWT, Swing and
java-gnome support.
Slides and screenshots at http://people.redhat.com/fitzsim/ddc2004/
A report by Christopher Blizzard can be found at
http://www.0xdeadbeef.com/html/2004/07/#200407201126

MERGING. We decided that it was very important to share these advances
to our AWT and Swing code faster with the other projects so this last
week we made sure that the libgcj gui branch and GNU Classpath CVS have
been (almost) completely merged. Kaffe has merged the AWT and Swing code
again from GNU Classpath into Kaffe CVS and jamvm (1.1.4) and kissme
(CVS) can be made to work out of the box with the new GNU Classpath CVS
gui code.

Starting immediately we are going to start merging between the libgcj
gui branch and GNU Classpath CVS directly and more frequently. Any
future work in [gnu]/java[x]/(awt/swing)/ and native/jni/gtk-peer
directories should go directly to the libgcj gui branch or GNU Classpath
CVS, *not* gcc trunk, since this will ease integration and reduce the
risk of a merge clobbering changes.

We used to do merges like:
  kaffe/[other-projects] <=3D> classpath <=3D> libgcj <=3D> gui-branch
Since most changes for java/awt, javax/swing and native/jni/gtk-peers
take place on the gui branch we will now do it like:
   kaffe/[other-projects] <=3D> classpath <=3D> gui-branch <=3D> libgcj
at least for those (sub)dirs.

(Note that doesn't mean you cannot develop gui stuff directly on GNU
Classpath CVS, but if you do either check it also in on the gui-branch,
or ask someone to do that for you.)

I used to be a bit skeptical about splitting off the gui work on a
separate branch. But these days more and more people seem to have no
problem just working with the branch directly. Thomas his jhbuild setup
even makes it fun and easy to do:
http://people.redhat.com/fitzsim/gcj-and-jhbuild.html
And I noticed that being able to just use gdb with native gcj/C/JNI code
is invaluable for debugging at least the gtk-peers.

Tom Tromey setup a comparison script to show the things not completely
merged yet between GNU Classpath and libgcj. For the non-gui parts
(libgcj head) see: http://gcc.gnu.org/java/libgcj-classpath-compare.html
For the GUI parts (against libgcj gui-branch) see:
http://gcc.gnu.org/java/gui-compare/libgcj-classpath-compare.html

Jim Pick has offered us some shared resources for the GNU Classpath, gcj
and kaffe projects so we will setup a similar comparison script against
kaffe soon. (And we will then also be able to setup a
developer.classpath.org environment to make it easier to do testing,
autobuilding, scripting and generating dynamic webpages.)

Of course we hope to get both gcj and kaffe working out of the box with
GNU Classpath (CVS). This is a slow process though. Hopefully for gcj
the BC-ABI work will make this easier. And Guilhem has recently switched
kaffe to use the GNU Classpath VMObject and VMThread platform interfaces
to be able to share more low level code.

EXAMPLES. To make it easier to show off all the nice new GUI work we now
have a new examples directory.  The idea is that we will distribute a
small collection of example code to show what can be done with the GNU
Classpath library. The first examples included are the old TestAWT and
Test code merged into a AWTDemo and the activityboard and testswing code
merged into a SwingDemo.

This is how it currently looks (actually it looks a bit better now
against GNU Classpath CVS from today, the screenshots are already old!):
http://www.klomp.org/mark/classpath/awtdemo-2004-jul-25.png
http://www.klomp.org/mark/classpath/swingdemo-2004-jul-25.png

When you do a make && make install a precompiled examples.zip and all
the source code will be installed in ${prefix}/share/classpath/examples
to play with. See the examples/README for more information.

This is not just for GUI code though. It should be easy to add new
examples (IODemo, NetDemo, BeanDemo, CollectionsDemo, RegexDemo,
RMIDemo, NIODemo, TextDemo, SQLDemo, MathDemo, GeomDemo, ...) by just
adding a new directory under gnu/classpath/examples/ in the examples dir
and to add a Demo.java (and possibily some helper classes). Also when
gcj 3.5 is out we want to include examples of how to create native
binaries directly.

BUGS. Red Hat had an internal bug database with AWT and Swing related
issues for libgcj. These have all been exported to the public bugzilla
on gcc.gnu.org (both old and resolved bugs and new still open bugs). I
made sure that all AWT and SWING related issues in the savannah tracker
are now also mirrored there. The gcc.gnu.org bugzilla maintainer setup a
special AWT and SWING component for us to keep track of both bugs and
new features people want to work on. Please coordinate and list missing
issues at:
http://gcc.gnu.org/bugzilla/buglist.cgi?component=3DAWT
http://gcc.gnu.org/bugzilla/buglist.cgi?component=3DSWING
For everything else please keep using the savannah bug tracker:
https://savannah.gnu.org/bugs/?group=3Dclasspath
(Bugzilla support for Savannah will hopefully come in the future.)

NOT MERGED YET. The main gui things not merged yet between libgcj and
GNU Classpath are the build infrastructure to get the Cairo Graphics2D
backend working (all code is already in the tree). And libgcj has a xlib
AWT peers backend which has not yet been merged into GNU Classpath (it
seemed wiser to concentrate on the GTK+ AWT peers and get them 100%
perfect first). As always, volunteers to help with this are welcome!

Happy hacking!

Mark

--=-yueFvLxvHcvAi0fy0WY9
Content-Type: application/pgp-signature; name=signature.asc
Content-Description: This is a digitally signed message part

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)

iD8DBQBBDXT4xVhZCJWr9QwRAss2AJ9AeXbcxoAnhQvsYqz0jQHNHjtreQCgq1Ux
cOmWMQxokxNZWoZFYkjKOuM=
=3wgO
-----END PGP SIGNATURE-----

--=-yueFvLxvHcvAi0fy0WY9--




From MAILER-DAEMON Sun Aug 01 22:57:23 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BrT0t-000090-KS
	for mharc-classpath@gnu.org; Sun, 01 Aug 2004 22:57:23 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BrT0r-00008t-NJ
	for classpath@gnu.org; Sun, 01 Aug 2004 22:57:21 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BrT0p-00008a-Sp
	for classpath@gnu.org; Sun, 01 Aug 2004 22:57:21 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BrT0p-00008Q-QF
	for classpath@gnu.org; Sun, 01 Aug 2004 22:57:19 -0400
Received: from [199.232.41.8] (helo=mx20.gnu.org)
	by monty-python.gnu.org with esmtp (TLSv1:DES-CBC3-SHA:168)
	(Exim 4.34) id 1BrSxI-0006NE-Od
	for classpath@gnu.org; Sun, 01 Aug 2004 22:53:40 -0400
Received: from [24.25.9.102] (helo=ms-smtp-03-eri0.southeast.rr.com)
	by mx20.gnu.org with esmtp (Exim 4.34) id 1BrSmj-0008Uk-0F
	for classpath@gnu.org; Sun, 01 Aug 2004 22:42:45 -0400
Received: from [192.168.0.3] (rdu74-160-248.nc.rr.com [24.74.160.248])
	by ms-smtp-03-eri0.southeast.rr.com (8.12.10/8.12.7) with ESMTP id
	i722gdiA008821; Sun, 1 Aug 2004 22:42:41 -0400 (EDT)
From: "C. Brian Jones" <cbj@gnu.org>
To: Arnaud Vandyck <avdyk@debian.org>
In-Reply-To: <87d62argrb.fsf@oz.fapse.ulg.ac.be>
References: <87d62argrb.fsf@oz.fapse.ulg.ac.be>
Content-Type: text/plain
Message-Id: <1091414558.4640.123.camel@lyta.haphazard.org>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.4.5 (1.4.5-7) 
Date: Sun, 01 Aug 2004 22:42:39 -0400
Content-Transfer-Encoding: 7bit
X-Virus-Scanned: Symantec AntiVirus Scan Engine
Cc: Alan Tam <Tam@SiuLung.com>, ClassPath List <classpath@gnu.org>,
	Kaffe Mailing List <kaffe@kaffe.org>
Subject: Re: Building log4j with free compiler (was: [Alan Tam] Re:
	IMPORTANT [RFS] liblog4j1.2-java build from sources)
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Mon, 02 Aug 2004 02:57:21 -0000

On Sun, 2004-08-01 at 11:59, Arnaud Vandyck wrote:
> System.out.println("Hi!");
> 
> I've tried to build log4j with free compilers but it fails. Alan Tam
> investigate a lot more and here is the result.

If it helps, I believe log4j 1.3.x will separate out the log browser
stuff since it's useless bloatware if all you're interested in is
logging with log4j.  But there is the java.util.logging interface that
you can use, though it lacks a few handlers for log output already
available in log4j appenders.

Brian
-- 
Brian Jones <cbj@gnu.org>




From MAILER-DAEMON Mon Aug 02 03:21:04 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BrX83-0007xY-Nm
	for mharc-classpath@gnu.org; Mon, 02 Aug 2004 03:21:03 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BrWog-0003qT-VS
	for classpath@gnu.org; Mon, 02 Aug 2004 03:01:03 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BrWoe-0003kd-7P
	for classpath@gnu.org; Mon, 02 Aug 2004 03:01:01 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BrWod-0003ja-R0
	for classpath@gnu.org; Mon, 02 Aug 2004 03:00:59 -0400
Received: from [218.190.66.159] (helo=SiuLung.com)
	by monty-python.gnu.org with esmtp (Exim 4.34)
	id 1BrWkT-0000RL-LB; Mon, 02 Aug 2004 02:56:42 -0400
Received: from 192.168.3.52 (authenticated user Tam@siulung.com) by siulung.com
	(MDaemon.PRO.v6.8.5.R) with ESMTP id 33-md50000000007.tmp;
	Mon, 02 Aug 2004 14:56:11 +0800
Message-ID: <410DE5A3.4020905@SiuLung.com>
Date: Mon, 02 Aug 2004 14:56:35 +0800
From: Alan Tam <Tam@SiuLung.com>
User-Agent: Mozilla Thunderbird 0.7+ (Windows/20040730)
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: "C. Brian Jones" <cbj@gnu.org>
References: <87d62argrb.fsf@oz.fapse.ulg.ac.be>
	<1091414558.4640.123.camel@lyta.haphazard.org>
In-Reply-To: <1091414558.4640.123.camel@lyta.haphazard.org>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
X-Authenticated-Sender: Tam@siulung.com
X-Spam-Processed: siulung.com, Mon, 02 Aug 2004 14:56:11 +0800
	(not processed: message from valid local sender)
X-Return-Path: Tam@SiuLung.com
X-Mailman-Approved-At: Mon, 02 Aug 2004 03:21:01 -0400
Cc: Arnaud Vandyck <avdyk@debian.org>, ClassPath List <classpath@gnu.org>,
	Kaffe Mailing List <kaffe@kaffe.org>
Subject: Re: [kaffe] Re: Building log4j with free compiler (was: [Alan Tam]
 Re:	IMPORTANT [RFS] liblog4j1.2-java build from sources)
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Mon, 02 Aug 2004 07:01:03 -0000

Hi,

On 2004-08-02 10:42, C. Brian Jones wrote:

>If it helps, I believe log4j 1.3.x will separate out the log browser
>stuff since it's useless bloatware if all you're interested in is
>logging with log4j.  But there is the java.util.logging interface that
>you can use, though it lacks a few handlers for log output already
>available in log4j appenders.
>  
>
I think Debian has no time to wait for 1.3, which was planned to go 
alpha 11 months ago. Maybe we can take the lead to remove the 
"bloatware" part from the1.2.8 package first. No matter there is 
java.util.logging or not, packages still tend to use commons-logging, 
and log4j is the de facto choice of backend. This is the significance of 
making sure log4j will be in sarge.

In the mean time, classpath (or kaffe, sablevm, ...) can go ahead to 
implement the missing Swing features, at least stub them so that we can 
compile.

-- 
Regards,
Alan




From MAILER-DAEMON Mon Aug 02 10:26:20 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1Brdlc-00067L-2Y
	for mharc-classpath@gnu.org; Mon, 02 Aug 2004 10:26:20 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1Brdlb-00066Q-0b
	for classpath@gnu.org; Mon, 02 Aug 2004 10:26:19 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1Brdla-00065h-0U
	for classpath@gnu.org; Mon, 02 Aug 2004 10:26:18 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BrdlZ-00065e-TQ
	for classpath@gnu.org; Mon, 02 Aug 2004 10:26:17 -0400
Received: from [62.253.162.42] (helo=mta02-svc.ntlworld.com)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1Brdhg-0006tg-P6
	for classpath@gnu.org; Mon, 02 Aug 2004 10:22:17 -0400
Received: from [192.168.1.100] ([82.6.126.159]) by mta02-svc.ntlworld.com
	(InterMail vM.4.01.03.37 201-229-121-137-20020806) with ESMTP
	id <20040802142044.JDVH12282.mta02-svc.ntlworld.com@[192.168.1.100]>
	for <classpath@gnu.org>; Mon, 2 Aug 2004 15:20:44 +0100
From: David Gilbert <david.gilbert@object-refinery.com>
To: ClassPath List <classpath@gnu.org>
Content-Type: multipart/mixed; boundary="=-1rEHhAFyfm+MOtxPgaTo"
Message-Id: <1091456621.6948.37.camel@linux42.overstone>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.4.6 
Date: Mon, 02 Aug 2004 15:23:43 +0100
Subject: Line2D intersection
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Mon, 02 Aug 2004 14:26:19 -0000


--=-1rEHhAFyfm+MOtxPgaTo
Content-Type: text/plain
Content-Transfer-Encoding: 7bit

Hello All,

I've recently started exploring Classpath to see where I can help out. 
In general, I am interested in Java2D, since the project I work on
(JFreeChart) depends on this quite a lot.  My FSF copyright assignment
is already completed.

To begin with, I've written some test cases for the Line2D class, and
sent these (just now) to the Mauve patches mailing list.  Another
developer (Sven de Marothy) previously posted a test case that
highlights a problem with the Line2D.linesIntersect() method.  I didn't
find any additional problems.  

I modified the Line2D class to correct the linesIntersect() method - a
patch is attached.  The fix is based on an algorithm described in
"Computational Geometry in C" (2nd edition) by Joseph O'Rourke.  I have
no background in computational geometry, by the way, just a mild
interest - so my code may be rubbish (but it passes the test cases,
which the existing code doesn't).

Re-reading Sven's earlier post, it sounds as though he has a fix for the
problem also, which I'll try to find out more about (I couldn't find it
posted anywhere).  Of course, if I had read his post properly the first
time, I would have found out about his fix before writing one myself...

Regards,

Dave Gilbert

--=-1rEHhAFyfm+MOtxPgaTo
Content-Disposition: attachment; filename=Line2D-patch.txt
Content-Type: text/x-patch; name=Line2D-patch.txt; charset=utf-8
Content-Transfer-Encoding: 7bit

Index: java/awt/geom/Line2D.java
===================================================================
RCS file: /cvsroot/classpath/classpath/java/awt/geom/Line2D.java,v
retrieving revision 1.6
diff -u -r1.6 Line2D.java
--- java/awt/geom/Line2D.java	18 Jul 2003 19:35:02 -0000	1.6
+++ java/awt/geom/Line2D.java	2 Aug 2004 12:33:11 -0000
@@ -235,11 +235,57 @@
   }
 
   /**
-   * Test if the line segment (x1,y1)-&gt;(x2,y2) intersects the line segment
+   * Computes twice the (signed) area of the triangle defined by the three
+   * points.  This method is used for intersection testing.
+   * 
+   * @param x1  the x-coordinate of the first point.
+   * @param y1  the y-coordinate of the first point.
+   * @param x2  the x-coordinate of the second point.
+   * @param y2  the y-coordinate of the second point.
+   * @param x3  the x-coordinate of the third point.
+   * @param y3  the y-coordinate of the third point.
+   * 
+   * @return Twice the area.
+   */
+  private static double area2(double x1, double y1,
+                             double x2, double y2,
+                             double x3, double y3) 
+  {
+    return (x2 - x1) * (y3 - y1) - (x3 - x1) * (y2 - y1);    
+  }
+
+  /**
+   * Returns <code>true</code> if (x3, y3) lies between (x1, y1) and (x2, y2),
+   * and false otherwise,  This test assumes that the three points are 
+   * collinear, and is used for intersection testing.
+   * 
+   * @param x1  the x-coordinate of the first point.
+   * @param y1  the y-coordinate of the first point.
+   * @param x2  the x-coordinate of the second point.
+   * @param y2  the y-coordinate of the second point.
+   * @param x3  the x-coordinate of the third point.
+   * @param y3  the y-coordinate of the third point.
+   * 
+   * @return A boolean.
+   */
+  private static boolean between(double x1, double y1, 
+                                double x2, double y2, 
+                                double x3, double y3) 
+  {
+    if (x1 != x2) {
+      return (x1 <= x3 && x3 <= x2) || (x1 >= x3 && x3 >= x2);   
+    }
+    else {
+      return (y1 <= y3 && y3 <= y2) || (y1 >= y3 && y3 >= y2);   
+    }
+  }
+
+  /**
+   * Test if the line segment (x1,y1)-&gt;(x2,y2) intersects the line segment 
    * (x3,y3)-&gt;(x4,y4).
    *
    * @param x1 the first x coordinate of the first segment
-   * @param y1 the first y coordinate of the first segment
+   * @param y1 the first y coordinate of the first segment 
    * @param x2 the second x coordinate of the first segment
    * @param y2 the second y coordinate of the first segment
    * @param x3 the first x coordinate of the second segment
@@ -249,16 +295,64 @@
    * @return true if the segments intersect
    */
   public static boolean linesIntersect(double x1, double y1,
-                                       double x2, double y2,
-                                       double x3, double y3,
-                                       double x4, double y4)
-  {
-    double beta = (((y1 - y3) * (x4 - x3) + (x1 - x3) * (y4 - y3))
-                   / ((y2 - y1) * (x4 - x3) + (x2 - x1) * (y4 - y3)));
-    if (beta < 0.0 || beta > 1.0)
-      return false;
-    double alpha = (x1 + beta * (x2 - x1) - x3) / (x4 - x3);
-    return alpha >= 0.0 && alpha <= 1.0;
+                                      double x2, double y2,
+                                      double x3, double y3,
+                                      double x4, double y4)
+  {
+    double a1, a2, a3, a4;
+  
+    // deal with special cases
+    if ((a1 = area2(x1, y1, x2, y2, x3, y3)) == 0.0) 
+    {
+      // check if p3 is between p1 and p2 OR
+      // p4 is collinear also AND either between p1 and p2 OR at opposite ends
+      if (between(x1, y1, x2, y2, x3, y3)) 
+      {
+        return true;
+      }
+      else 
+      {
+        if (area2(x1, y1, x2, y2, x4, y4) == 0.0) 
+        {
+          return between(x3, y3, x4, y4, x1, y1) 
+                 || between (x3, y3, x4, y4, x2, y2);
+        }
+        else {
+          return false;
+        }
+      }
+    }
+    else if ((a2 = area2(x1, y1, x2, y2, x4, y4)) == 0.0) 
+    {
+      // check if p4 is between p1 and p2 (we already know p3 is not
+      // collinear)
+      return between(x1, y1, x2, y2, x4, y4);
+    }
+  
+    if ((a3 = area2(x3, y3, x4, y4, x1, y1)) == 0.0) {
+      // check if p1 is between p3 and p4 OR
+      // p2 is collinear also AND either between p1 and p2 OR at opposite ends
+      if (between(x3, y3, x4, y4, x1, y1)) {
+        return true;
+      }
+      else {
+        if (area2(x3, y3, x4, y4, x2, y2) == 0.0) {
+          return between(x1, y1, x2, y2, x3, y3) 
+                 || between (x1, y1, x2, y2, x4, y4);
+        }
+        else {
+          return false;
+        }
+      }
+    }
+    else if ((a4 = area2(x3, y3, x4, y4, x2, y2)) == 0.0) {
+      // check if p2 is between p3 and p4 (we already know p1 is not
+      // collinear)
+      return between(x3, y3, x4, y4, x2, y2);
+    }
+    else {  // test for regular intersection
+      return ((a1 > 0.0) ^ (a2 > 0.0)) && ((a3 > 0.0) ^ (a4 > 0.0));
+    } 
   }
 
   /**

--=-1rEHhAFyfm+MOtxPgaTo--




From MAILER-DAEMON Tue Aug 03 04:39:28 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BrupT-00017z-UQ
	for mharc-classpath@gnu.org; Tue, 03 Aug 2004 04:39:28 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BrupQ-00016E-4O
	for classpath@gnu.org; Tue, 03 Aug 2004 04:39:24 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BrupL-00015d-5u
	for classpath@gnu.org; Tue, 03 Aug 2004 04:39:19 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BrupK-00015a-Uv
	for classpath@gnu.org; Tue, 03 Aug 2004 04:39:19 -0400
Received: from [213.228.0.62] (helo=postfix4-1.free.fr)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BrulW-0003sb-CG
	for classpath@gnu.org; Tue, 03 Aug 2004 04:35:22 -0400
Received: from kaffe.org (lns-vlq-17f-81-56-169-129.adsl.proxad.net
	[81.56.169.129])
	by postfix4-1.free.fr (Postfix) with ESMTP id B69A616CB45
	for <classpath@gnu.org>; Tue,  3 Aug 2004 10:35:20 +0200 (CEST)
Message-ID: <410F4E47.4040907@kaffe.org>
Date: Tue, 03 Aug 2004 10:35:19 +0200
From: Guilhem Lavaux <guilhem@kaffe.org>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.6) Gecko/20040115
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: classpath@gnu.org
Content-Type: multipart/mixed; boundary="------------030402020805070002080907"
Subject: [Fwd: [kaffe] (no subject)] (jni.h)
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Tue, 03 Aug 2004 08:39:24 -0000

This is a multi-part message in MIME format.
--------------030402020805070002080907
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit

Hi,

As it is a Classpath's file, I'm forwarding this mail here. I've added a 
small amount of lines for win32 at the top but it is not related.

Cheers,

Guilhem.

--------------030402020805070002080907
Content-Type: message/rfc822;
 name="[kaffe] (no subject)"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline;
 filename="[kaffe] (no subject)"

Return-path: <kaffe-bounces@kaffe.org>
Envelope-to: guilhem@kaffe.org
Received: from mail by pogo.kaffe.org with spam-scanned (Exim 3.35 #1 (Debian))
	id 1BrofH-0006eC-00; Mon, 02 Aug 2004 19:04:46 -0700
Received: from localhost ([127.0.0.1] helo=pogo.kaffe.org)
	by pogo.kaffe.org with esmtp (Exim 3.35 #1 (Debian))
	id 1BrYC7-0005oV-00; Mon, 02 Aug 2004 01:29:19 -0700
Received: from mail by pogo.kaffe.org with spam-scanned (Exim 3.35 #1 (Debian))
	id 1BrYBS-0005en-00
	for <kaffe@kaffe.org>; Mon, 02 Aug 2004 01:28:52 -0700
Received: from serv09.segi.ulg.ac.be ([139.165.32.78])
	by pogo.kaffe.org with esmtp (Exim 3.35 #1 (Debian))
	id 1BrYBR-0005bq-00
	for <kaffe@kaffe.org>; Mon, 02 Aug 2004 01:28:37 -0700
Received: (qmail 5877 invoked by uid 504); 2 Aug 2004 10:22:32 +0200
Received: from arnaud.vandyck@ulg.ac.be by serv09.segi.ulg.ac.be by uid 501
	with qmail-scanner-1.16 (spamassassin: 2.55.  Clear:. 
	Processed in 0.398945 secs); 02 Aug 2004 08:22:32 -0000
Received: from unknown (HELO oz.fapse.ulg.ac.be) ([139.165.201.14])
	(envelope-sender <arnaud.vandyck@ulg.ac.be>)
	by serv09.segi.ulg.ac.be (qmail-ldap-1.03) with SMTP
	for <kaffe@kaffe.org>; 2 Aug 2004 10:22:31 +0200
Received: from arnaud by oz.fapse.ulg.ac.be with local (Exim 4.34)
	id 1BrY5W-00014j-MH
	for kaffe@kaffe.org; Mon, 02 Aug 2004 10:22:30 +0200
To: Kaffe Mailing List <kaffe@kaffe.org>
Message-Id: <E1BrY5W-00014j-MH@oz.fapse.ulg.ac.be>
From: Arnaud Vandyck <arnaud.vandyck@ulg.ac.be>
Date: Mon, 02 Aug 2004 10:22:30 +0200
Subject: [kaffe] (no subject)
X-BeenThere: kaffe@kaffe.org
X-Mailman-Version: 2.1.4
Precedence: list
List-Id: Kaffe mailing list <kaffe.kaffe.org>
List-Unsubscribe: <http://kaffe.org/cgi-bin/mailman/listinfo/kaffe>,
	<mailto:kaffe-request@kaffe.org?subject=unsubscribe>
List-Archive: <http://pogo.kaffe.org/pipermail/kaffe>
List-Post: <mailto:kaffe@kaffe.org>
List-Help: <mailto:kaffe-request@kaffe.org?subject=help>
List-Subscribe: <http://kaffe.org/cgi-bin/mailman/listinfo/kaffe>,
	<mailto:kaffe-request@kaffe.org?subject=subscribe>
Sender: kaffe-bounces@kaffe.org
Errors-To: kaffe-bounces@kaffe.org
X-Spam-Checker-Version: SpamAssassin 2.63 (2004-01-11) on pogo.kaffe.org
X-Spam-Status: No, hits=-4.7 required=5.0 tests=AWL,BAYES_00,RCVD_IN_SORBS 
	autolearn=no version=2.63
X-Spam-Level: 

Cc; 262760@bugs.debian.org
Subject: [Jan Schulz] Bug#262760: kaffe-dev: include/jni.h misses #include
 <stdio.h> to compile eclipse
From: Arnaud Vandyck <avdyk@debian.org>
Mail-Followup-To: Kaffe Mailing List <kaffe@kaffe.org>
Date: Mon, 02 Aug 2004 10:22:30 +0200
Message-ID: <87acxec5kp.fsf@oz.fapse.ulg.ac.be>
User-Agent: Gnus/5.1006 (Gnus v5.10.6) Emacs/21.3 (gnu/linux)
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="=-=-="

--=-=-=

For your information, Jan Schulz has just submitted this bug to the
debian package of kaffe from cvs head July 21 2004.

Regards,


--=-=-=
Content-Type: message/rfc822
Content-Disposition: inline

Subject: Bug#262760: kaffe-dev: include/jni.h misses #include <stdio.h> to
    compile eclipse
Reply-To: Jan Schulz <debian@katzien.de>, 262760@bugs.debian.org
Resent-From: Jan Schulz <debian@katzien.de>
Original-Sender: Jan Schulz <jasc@gmx.net>
Resent-To: debian-bugs-dist@lists.debian.org
Resent-CC: ean@novare.net (Ean R. Schuessler)
Resent-Date: Sun, 01 Aug 2004 21:03:04 UTC
Resent-Message-Id: <handler.262760.B.10913933653542@bugs.debian.org>
From: Jan Schulz <debian@katzien.de>
To: Debian Bug Tracking System <submit@bugs.debian.org>
Date: Sun, 01 Aug 2004 22:49:52 +0200
Message-Id: <E1BrNHE-0006Kn-00@snoopy.katzien.de>
Resent-Sender: Debian BTS <debbugs@bugs.debian.org>
Lines: 41
MIME-Version: 1.0

Package: kaffe-dev
Version: 2:1.1.4.PRECVS-20040721
Severity: normal
Tags: patch

The /usr/lib/kaffe/include/jni.h misses a #include <stdio.h> to compile 
the jni files of eclipse 3.0. With this added, it compiles just fine.

Pseudo diff:
------
#ifndef __CLASSPATH_JNI_H__
#define __CLASSPATH_JNI_H__

#include <stdarg.h>
+#include <stdio.h>
/* Linkage and calling conventions. */
#if defined (_WIN32) || defined (__WIN32__) || defined (WIN32)

#define JNIIMPORT        __declspec(dllimport)
#define JNIEXPORT        __declspec(dllexport)
------

Nice freetings, Jan

-- System Information:
Debian Release: 3.1
  APT prefers experimental
  APT policy: (990, 'experimental'), (500, 'unstable')
Architecture: i386 (i686)
Kernel: Linux 2.6.6-1-k7
Locale: LANG=de_DE@euro, LC_CTYPE=de_DE@euro

Versions of packages kaffe-dev depends on:
ii  kaffe            2:1.1.4.PRECVS-20040721 A JVM to run Java bytecode
ii  kaffe-jthreads [ 2:1.1.4.PRECVS-20040721 A green threads compile of the Kaf

-- no debconf information





--=-=-=



-- 
Arnaud Vandyck

<fortune>
Bleeding Ponytail:
	An elderly, sold-out baby boomer who pines for hippie or
presellout days.
		-- Douglas Coupland, "Generation X: Tales for an Accelerated
		   Culture"
</fortune>

--=-=-=--

_______________________________________________
kaffe mailing list
kaffe@kaffe.org
http://kaffe.org/cgi-bin/mailman/listinfo/kaffe


--------------030402020805070002080907--



From MAILER-DAEMON Tue Aug 03 07:32:32 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BrxWy-0003dR-4W
	for mharc-classpath@gnu.org; Tue, 03 Aug 2004 07:32:32 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BrxWw-0003dL-Gz
	for classpath@gnu.org; Tue, 03 Aug 2004 07:32:30 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BrxWu-0003d9-RD
	for classpath@gnu.org; Tue, 03 Aug 2004 07:32:30 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BrxWu-0003d6-Os
	for classpath@gnu.org; Tue, 03 Aug 2004 07:32:28 -0400
Received: from [212.6.122.12] (helo=mail1.ewetel.de)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BrxT6-0007dy-KS
	for classpath@gnu.org; Tue, 03 Aug 2004 07:28:33 -0400
Received: from majestix.konqueror.de (dynadsl-080-228-93-080.ewetel.net
	[80.228.93.80])
	by mail1.ewetel.de (8.12.1/8.12.9) with ESMTP id i73BSUdf014306;
	Tue, 3 Aug 2004 13:28:30 +0200 (MEST)
Received: from miraculix.konqueror.de ([192.168.1.1])
	by majestix.konqueror.de with esmtp (Exim 3.35 #1 (Debian))
	id 1BrxbO-0002r6-00; Tue, 03 Aug 2004 13:37:06 +0200
From: Michael Koch <konqueror@gmx.de>
To: classpath@gnu.org, pingo@aicas.com
Date: Tue, 3 Aug 2004 13:29:01 +0200
User-Agent: KMail/1.6.2
MIME-Version: 1.0
Content-Disposition: inline
Content-Type: text/plain;
  charset="iso-8859-15"
Content-Transfer-Encoding: 7bit
Message-Id: <200408031329.01616.konqueror@gmx.de>
X-CheckCompat: OK
Cc: 
Subject: Removing the TARGET_* layer or not ?
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Tue, 03 Aug 2004 11:32:30 -0000

Hi list, hi Ingo,



I just started porting/testing GNU classpath to solaris (2.6). Its not 
really different to other unices (we currently support linux and 
*BSD) but some things are. When I wanted to do the needed changes I 
stumbled over the TARGET_* layer which adds (in my eyes) some 
unneeded complexity. It makes it more hard to read the code and all 
the stuff seem to be implemented either in target/generic or 
target/Linux with no real rule what have to go where. It's just 
difficult to understand. In fact I try to understand it since a long 
time and always fail because of its "indirectness". Personally I 
think some AUTOCONF checks would be more appropriate and would make 
the code much more readable and bugfixable. There are some known bugs 
in it but noone attacked them because none understands the code it 
seems.

Ingo: Now my question are you really using the TARGET_* system or is 
it only rotting around in GNU classpath ? I really wonder if some 
AUTOCONF macros would be more helpful for you ?

If it is consensus that noone understands the cod and its good to 
rewrite it I can start it. Then I will do it slowly part for part and 
try to introduce as less as possible bugs. Unfortunately there will 
be bugs introduced. We are all humans. So this will need some testing 
on some archs, Linux, *.BSD, AIX, Solaris ... etc.

What is your opinion on the TARGET_* system ?

It would be nice to get a statement from the AICAS people too as they 
introduced it.


Michael



From MAILER-DAEMON Tue Aug 03 09:43:15 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BrzZT-0006x4-D8
	for mharc-classpath@gnu.org; Tue, 03 Aug 2004 09:43:15 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BrzZR-0006vH-HF
	for classpath@gnu.org; Tue, 03 Aug 2004 09:43:13 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BrzZQ-0006ud-C9
	for classpath@gnu.org; Tue, 03 Aug 2004 09:43:13 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BrzZQ-0006uT-7l
	for classpath@gnu.org; Tue, 03 Aug 2004 09:43:12 -0400
Received: from [212.227.126.184] (helo=moutng.kundenserver.de)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BrzVe-0008La-Rj
	for classpath@gnu.org; Tue, 03 Aug 2004 09:39:19 -0400
Received: from [212.227.126.209] (helo=mrelayng.kundenserver.de)
	by moutng.kundenserver.de with esmtp (Exim 3.35 #1)
	id 1BrzVP-0007bA-00; Tue, 03 Aug 2004 15:39:03 +0200
Received: from [212.126.219.82] (helo=aicas.com)
	by mrelayng.kundenserver.de with asmtp (TLSv1:RC4-MD5:128)
	(Exim 3.35 #1) id 1BrzVO-0003Zs-00; Tue, 03 Aug 2004 15:39:02 +0200
Message-ID: <410F9574.8040408@aicas.com>
Date: Tue, 03 Aug 2004 15:39:00 +0200
From: =?ISO-8859-15?Q?Ingo_Pr=F6tel?= <proetel@aicas.com>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.6) Gecko/20040113
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: Michael Koch <konqueror@gmx.de>
References: <200408031329.01616.konqueror@gmx.de>
In-Reply-To: <200408031329.01616.konqueror@gmx.de>
X-Enigmail-Version: 0.83.6.0
X-Enigmail-Supports: pgp-inline, pgp-mime
Content-Type: text/plain; charset=ISO-8859-15; format=flowed
Content-Transfer-Encoding: 8bit
X-Provags-ID: kundenserver.de abuse@kundenserver.de
	auth:ebd3e0372cc3ac236a2e2611be1a8528
Cc: classpath@gnu.org
Subject: Re: Removing the TARGET_* layer or not ?
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Tue, 03 Aug 2004 13:43:13 -0000

Hi Michael,

Michael Koch wrote:
> Hi list, hi Ingo,
> 
> 
> 
> I just started porting/testing GNU classpath to solaris (2.6). Its not 
> really different to other unices (we currently support linux and 
> *BSD) but some things are. When I wanted to do the needed changes I 
> stumbled over the TARGET_* layer which adds (in my eyes) some 
> unneeded complexity. It makes it more hard to read the code and all 
> the stuff seem to be implemented either in target/generic or 
> target/Linux with no real rule what have to go where. It's just 
> difficult to understand. In fact I try to understand it since a long 
> time and always fail because of its "indirectness". Personally I 
> think some AUTOCONF checks would be more appropriate and would make 
> the code much more readable and bugfixable. There are some known bugs 
> in it but noone attacked them because none understands the code it 
> seems.
> 
> Ingo: Now my question are you really using the TARGET_* system or is 
> it only rotting around in GNU classpath ? I really wonder if some 
> AUTOCONF macros would be more helpful for you ?

Yes, the TARGET-layer is in active use here.

The TARGET-layer has two advantages:
1. It makes JNI-code more readable,
2. It makes porting simple.

It makes JNI-code more readable because whenever a native function is called there is only one macro call. The other possibility would be that one has 
ifdefs for every system (possibly for every system+architecture) that one wants to support with this code. This would make the JNI-code difficult to 
read.

It makes porting easier in that one can reuse generic macros and only needs to override specific macros that differ for a given new system. And this 
all can happen without modifing the actual JNI-code. So the main step in porting is creating a new subdirectory in tartget name it Solaris. Copy over 
all the files in the Linux target. Add this to the include path. Then just work on the error that the compiler will find and where necessary create 
Solaris specific macros.

If there is interest we could provide our Solaris, Darwin, and MinGW TARGET-layer. This would probably help to understand how this layer is helpful.
In any case we can help if anyone needs help understanding the code.

> 
> If it is consensus that noone understands the cod and its good to 
> rewrite it I can start it. Then I will do it slowly part for part and 
> try to introduce as less as possible bugs. Unfortunately there will 
> be bugs introduced. We are all humans. So this will need some testing 
> on some archs, Linux, *.BSD, AIX, Solaris ... etc.
> 
> What is your opinion on the TARGET_* system ?
> 
> It would be nice to get a statement from the AICAS people too as they 
> introduced it.
> 
> 
> Michael
> 
> 
> _______________________________________________
> Classpath mailing list
> Classpath@gnu.org
> http://lists.gnu.org/mailman/listinfo/classpath
> 

Ingo

-- 
Ingo Prötel                                          proetel@aicas.com
aicas GmbH                                        http://www.aicas.com
Haid-und-Neu-Str. 18                        phone   +49 721 663 968-32
76131 Karlsruhe                             fax     +49 721 663 968-93
Germany



From MAILER-DAEMON Tue Aug 03 10:07:34 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1Brzwz-0008Sc-RM
	for mharc-classpath@gnu.org; Tue, 03 Aug 2004 10:07:33 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1Brzwx-0008RR-Py
	for classpath@gnu.org; Tue, 03 Aug 2004 10:07:31 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1Brzwv-0008QJ-Ns
	for classpath@gnu.org; Tue, 03 Aug 2004 10:07:30 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1Brzwu-0008PA-HP
	for classpath@gnu.org; Tue, 03 Aug 2004 10:07:28 -0400
Received: from [212.6.122.32] (helo=mail5.ewetel.de)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BrzsP-0005GF-RA
	for classpath@gnu.org; Tue, 03 Aug 2004 10:02:50 -0400
Received: from majestix.konqueror.de (dynadsl-080-228-93-080.ewetel.net
	[80.228.93.80])
	by mail5.ewetel.de (8.12.1/8.12.9) with ESMTP id i73E128D026101;
	Tue, 3 Aug 2004 16:01:03 +0200 (MEST)
Received: from miraculix.konqueror.de ([192.168.1.1])
	by majestix.konqueror.de with esmtp (Exim 3.35 #1 (Debian))
	id 1Brzz1-0003E1-00; Tue, 03 Aug 2004 16:09:39 +0200
From: Michael Koch <konqueror@gmx.de>
To: Ingo =?iso-8859-15?q?Pr=F6tel?= <proetel@aicas.com>
Date: Tue, 3 Aug 2004 16:01:34 +0200
User-Agent: KMail/1.6.2
References: <200408031329.01616.konqueror@gmx.de> <410F9574.8040408@aicas.com>
In-Reply-To: <410F9574.8040408@aicas.com>
MIME-Version: 1.0
Content-Disposition: inline
Content-Type: text/plain;
  charset="iso-8859-15"
Content-Transfer-Encoding: quoted-printable
Message-Id: <200408031601.34628.konqueror@gmx.de>
X-CheckCompat: OK
Cc: classpath@gnu.org
Subject: Re: Removing the TARGET_* layer or not ?
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Tue, 03 Aug 2004 14:07:33 -0000

Am Dienstag, 3. August 2004 15:39 schrieb Ingo Pr=F6tel:
> Hi Michael,
>
> Michael Koch wrote:
> > Hi list, hi Ingo,
> >
> >
> >
> > I just started porting/testing GNU classpath to solaris (2.6).
> > Its not really different to other unices (we currently support
> > linux and *BSD) but some things are. When I wanted to do the
> > needed changes I stumbled over the TARGET_* layer which adds (in
> > my eyes) some unneeded complexity. It makes it more hard to read
> > the code and all the stuff seem to be implemented either in
> > target/generic or target/Linux with no real rule what have to go
> > where. It's just difficult to understand. In fact I try to
> > understand it since a long time and always fail because of its
> > "indirectness". Personally I think some AUTOCONF checks would be
> > more appropriate and would make the code much more readable and
> > bugfixable. There are some known bugs in it but noone attacked
> > them because none understands the code it seems.
> >
> > Ingo: Now my question are you really using the TARGET_* system or
> > is it only rotting around in GNU classpath ? I really wonder if
> > some AUTOCONF macros would be more helpful for you ?
>
> Yes, the TARGET-layer is in active use here.
>
> The TARGET-layer has two advantages:
> 1. It makes JNI-code more readable,
> 2. It makes porting simple.
>
> It makes JNI-code more readable because whenever a native function
> is called there is only one macro call. The other possibility would
> be that one has ifdefs for every system (possibly for every
> system+architecture) that one wants to support with this code. This
> would make the JNI-code difficult to read.

The MACRO names are really long and they dont really make the code=20
more readable for me. Perhaps for the compiler ... I dont know.

> It makes porting easier in that one can reuse generic macros and
> only needs to override specific macros that differ for a given new
> system. And this all can happen without modifing the actual
> JNI-code. So the main step in porting is creating a new
> subdirectory in tartget name it Solaris. Copy over all the files in
> the Linux target. Add this to the include path. Then just work on
> the error that the compiler will find and where necessary create
> Solaris specific macros.

I tried to port to Solaris and you system just made it hard for me=20
(and I know mjw had problems with it too when he dbugged some=20
filesystem stuff).

> If there is interest we could provide our Solaris, Darwin, and
> MinGW TARGET-layer. This would probably help to understand how this
> layer is helpful. In any case we can help if anyone needs help
> understanding the code.

I don't get the benefit of the whole system. BSD's are already=20
suppored by the "Linux" layer. I guess Darwin (as its a BSD too) is=20
supported too and if not its easy to do so. I tried to support=20
solaris in the "Linux" layer and I needed to add around 10 lines for=20
it. So why add a lot oc code duplication to support an additional OS=20
when its enough to add 10 lines that are much more easy to=20
understand, adds no bloat and is more easy to maintain ? I haven't=20
tried mingw yet but I guess its similar easy.

Sorry, but I'm currently upset because this layer made it hard for me=20
to do something normally really simple. Autoconf was created to make=20
it easy to write portable apps. We use it already in GNU classpath.=20
Why not use it to write portable apps in it ?


Michael



From MAILER-DAEMON Tue Aug 03 10:24:40 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1Bs0DY-0005R6-Ke
	for mharc-classpath@gnu.org; Tue, 03 Aug 2004 10:24:40 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1Bs0DX-0005R1-K4
	for classpath@gnu.org; Tue, 03 Aug 2004 10:24:39 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1Bs0DV-0005Qp-R2
	for classpath@gnu.org; Tue, 03 Aug 2004 10:24:39 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1Bs0DV-0005Ql-JP
	for classpath@gnu.org; Tue, 03 Aug 2004 10:24:37 -0400
Received: from [213.9.1.84] (helo=box84.elkhouse.de)
	by monty-python.gnu.org with esmtp (TLSv1:RC4-SHA:128) (Exim 4.34)
	id 1Bs09i-000844-Sf
	for classpath@gnu.org; Tue, 03 Aug 2004 10:20:43 -0400
Received: from 1cust48.vr1.dtm1.alter.net ([149.229.96.48] helo=[192.168.1.3])
	by box84.elkhouse.de with asmtp (Exim 4.32)
	id 1Bs0BT-0008Sm-7k; Tue, 03 Aug 2004 16:22:31 +0200
From: Roman Kennke <roman@ontographics.com>
To: Michael Koch <konqueror@gmx.de>
In-Reply-To: <200408031329.01616.konqueror@gmx.de>
References: <200408031329.01616.konqueror@gmx.de>
Content-Type: text/plain
Message-Id: <1091550034.719.13.camel@moonlight>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.4.6 
Date: Tue, 03 Aug 2004 18:20:34 +0200
Content-Transfer-Encoding: 7bit
Cc: pingo@aicas.com, classpath@gnu.org
Subject: Re: Removing the TARGET_* layer or not ?
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Tue, 03 Aug 2004 14:24:39 -0000

Hi there,

I agree, the TARGET layer makes things difficult to understand. And I
really don't like multi-line macros in C code. This is bad style IMHO.
What makes it worse is, that it is not clear at first sight, where 
these macros are defined. There are better/easier-to-understand ways to
make code portable, I think.

This target layer makes maintainence of the code more difficult, at
least to those (like me), who are not familiar to it. I once tried to
find a bug in the network code, but couldn't figure out the indirections
in the code. Sorry.

Just my 2c.

/Roman

Am Di, den 03.08.2004 schrieb Michael Koch um 13:29:
> Hi list, hi Ingo,
> 
> 
> 
> I just started porting/testing GNU classpath to solaris (2.6). Its not 
> really different to other unices (we currently support linux and 
> *BSD) but some things are. When I wanted to do the needed changes I 
> stumbled over the TARGET_* layer which adds (in my eyes) some 
> unneeded complexity. It makes it more hard to read the code and all 
> the stuff seem to be implemented either in target/generic or 
> target/Linux with no real rule what have to go where. It's just 
> difficult to understand. In fact I try to understand it since a long 
> time and always fail because of its "indirectness". Personally I 
> think some AUTOCONF checks would be more appropriate and would make 
> the code much more readable and bugfixable. There are some known bugs 
> in it but noone attacked them because none understands the code it 
> seems.
> 
> Ingo: Now my question are you really using the TARGET_* system or is 
> it only rotting around in GNU classpath ? I really wonder if some 
> AUTOCONF macros would be more helpful for you ?
> 
> If it is consensus that noone understands the cod and its good to 
> rewrite it I can start it. Then I will do it slowly part for part and 
> try to introduce as less as possible bugs. Unfortunately there will 
> be bugs introduced. We are all humans. So this will need some testing 
> on some archs, Linux, *.BSD, AIX, Solaris ... etc.
> 
> What is your opinion on the TARGET_* system ?
> 
> It would be nice to get a statement from the AICAS people too as they 
> introduced it.
> 
> 
> Michael
> 
> 
> _______________________________________________
> Classpath mailing list
> Classpath@gnu.org
> http://lists.gnu.org/mailman/listinfo/classpath




From MAILER-DAEMON Tue Aug 03 10:42:41 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1Bs0Uz-00031l-Qh
	for mharc-classpath@gnu.org; Tue, 03 Aug 2004 10:42:41 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1Bs0Uw-00031a-A7
	for classpath@gnu.org; Tue, 03 Aug 2004 10:42:38 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1Bs0Uv-00031L-Kn
	for classpath@gnu.org; Tue, 03 Aug 2004 10:42:37 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1Bs0Uv-00031I-I2
	for classpath@gnu.org; Tue, 03 Aug 2004 10:42:37 -0400
Received: from [212.6.122.32] (helo=mail5.ewetel.de)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1Bs0RO-0001wp-1q
	for classpath@gnu.org; Tue, 03 Aug 2004 10:38:58 -0400
Received: from majestix.konqueror.de (dynadsl-080-228-93-080.ewetel.net
	[80.228.93.80])
	by mail5.ewetel.de (8.12.1/8.12.9) with ESMTP id i73Ecu8D020798;
	Tue, 3 Aug 2004 16:38:56 +0200 (MEST)
Received: from miraculix.konqueror.de ([192.168.1.1])
	by majestix.konqueror.de with esmtp (Exim 3.35 #1 (Debian))
	id 1Bs0Zh-0003FZ-00; Tue, 03 Aug 2004 16:47:33 +0200
From: Michael Koch <konqueror@gmx.de>
To: classpath@gnu.org
Date: Tue, 3 Aug 2004 16:39:28 +0200
User-Agent: KMail/1.6.2
References: <200408031329.01616.konqueror@gmx.de> <410F9574.8040408@aicas.com>
In-Reply-To: <410F9574.8040408@aicas.com>
MIME-Version: 1.0
Content-Disposition: inline
Content-Type: text/plain;
  charset="iso-8859-15"
Content-Transfer-Encoding: quoted-printable
Message-Id: <200408031639.28188.konqueror@gmx.de>
X-CheckCompat: OK
Cc: 
Subject: Re: Removing the TARGET_* layer or not ?
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Tue, 03 Aug 2004 14:42:38 -0000

Am Dienstag, 3. August 2004 15:39 schrieb Ingo Pr=F6tel:
> Hi Michael,
>
> Michael Koch wrote:
> > Hi list, hi Ingo,
> >
> >
> >
> > I just started porting/testing GNU classpath to solaris (2.6).
> > Its not really different to other unices (we currently support
> > linux and *BSD) but some things are. When I wanted to do the
> > needed changes I stumbled over the TARGET_* layer which adds (in
> > my eyes) some unneeded complexity. It makes it more hard to read
> > the code and all the stuff seem to be implemented either in
> > target/generic or target/Linux with no real rule what have to go
> > where. It's just difficult to understand. In fact I try to
> > understand it since a long time and always fail because of its
> > "indirectness". Personally I think some AUTOCONF checks would be
> > more appropriate and would make the code much more readable and
> > bugfixable. There are some known bugs in it but noone attacked
> > them because none understands the code it seems.
> >
> > Ingo: Now my question are you really using the TARGET_* system or
> > is it only rotting around in GNU classpath ? I really wonder if
> > some AUTOCONF macros would be more helpful for you ?
>
> Yes, the TARGET-layer is in active use here.
>
> The TARGET-layer has two advantages:
> 1. It makes JNI-code more readable,
> 2. It makes porting simple.
>
> It makes JNI-code more readable because whenever a native function
> is called there is only one macro call. The other possibility would
> be that one has ifdefs for every system (possibly for every
> system+architecture) that one wants to support with this code. This
> would make the JNI-code difficult to read.
>
> It makes porting easier in that one can reuse generic macros and
> only needs to override specific macros that differ for a given new
> system. And this all can happen without modifing the actual
> JNI-code. So the main step in porting is creating a new
> subdirectory in tartget name it Solaris. Copy over all the files in
> the Linux target. Add this to the include path. Then just work on
> the error that the compiler will find and where necessary create
> Solaris specific macros.
>
> If there is interest we could provide our Solaris, Darwin, and
> MinGW TARGET-layer. This would probably help to understand how this
> layer is helpful. In any case we can help if anyone needs help
> understanding the code.

Another thing unmentioned yet is that if you fix a bug in one=20
TARGET-layer you need to check all other too to be sure the bug is=20
totally fixed. If all is in one base you need to fix a bug only once.=20
With more care but only once.


Michael



From MAILER-DAEMON Tue Aug 03 15:18:44 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1Bs4o7-0008GP-U8
	for mharc-classpath@gnu.org; Tue, 03 Aug 2004 15:18:44 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1Bs4o6-0008Fs-SI
	for classpath@gnu.org; Tue, 03 Aug 2004 15:18:42 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1Bs4o3-0008Eo-JA
	for classpath@gnu.org; Tue, 03 Aug 2004 15:18:42 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1Bs4o3-0008Ee-Es
	for classpath@gnu.org; Tue, 03 Aug 2004 15:18:39 -0400
Received: from [130.242.128.36] (helo=mx.physto.se)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1Bs4kH-0005ti-Hb
	for classpath@gnu.org; Tue, 03 Aug 2004 15:14:45 -0400
Received: from localhost (localhost [127.0.0.1])
	by mx.physto.se (Postfix) with ESMTP id 91C372EE69
	for <classpath@gnu.org>; Tue,  3 Aug 2004 21:14:43 +0200 (CEST)
Received: from mx.physto.se ([127.0.0.1])
	by localhost (syslx09.physto.se [127.0.0.1]) (amavisd-new, port 10024)
	with ESMTP id 23511-07 for <classpath@gnu.org>;
	Tue,  3 Aug 2004 21:14:40 +0200 (CEST)
Received: from qcplx01.physto.se (qcplx01.physto.se [130.242.128.121])
	by mx.physto.se (Postfix) with ESMTP id A185F2EE35
	for <classpath@gnu.org>; Tue,  3 Aug 2004 21:14:40 +0200 (CEST)
From: Sven de Marothy <sven@physto.se>
To: classpath@gnu.org
Content-Type: text/plain
Message-Id: <1091560480.7431.51.camel@qcplx01.physto.se>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.4.6 
Date: Tue, 03 Aug 2004 21:14:40 +0200
Content-Transfer-Encoding: 7bit
X-Virus-Scanned: by amavisd-new at physto.se
Subject: Re: Line2D intersection
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Tue, 03 Aug 2004 19:18:43 -0000

Hi,

>I modified the Line2D class to correct the linesIntersect() method - a
>patch is attached.  The fix is based on an algorithm described in
>"Computational Geometry in C" (2nd edition) by Joseph O'Rourke.  

Great! I've tried it out, and it seems to work just fine. And it appears
to be relatively robust too. (But not invincible)

For instance, Sun's implementation can't handle lines which are colinear
within numerical error, for instance try:
(492.4307556152344,315.69091796875) to
(213.55988088439742,243.2761970031803)
and (201.19495192826076,240.0653816715973) to
(195.03655523160162,238.4662237143015)

(There are open-source CAD/GIS libraries like JTS
http://www.vividsolutions.com/jts/ which provide numerically robust
algorithms for those who really need 14 significant digits for every
calculation and don't mind speed.)

Actually most of my work on Area() at the moment is trying out the
robustness. At the moment it gets into some trouble with shapes that
have > 1000 or so self-intersections.

>Re-reading Sven's earlier post, it sounds as though he has a fix for
>the problem also, which I'll try to find out more about (I couldn't
>find it posted anywhere).

Yes, I've written intersection-point code for Area, so I figured I'd
just adapt that to return true or false instead of a point or null.
But I think your code seems better (haven't compared them against
eachother though).

Just one suggestion: I'm (hopefully) going to be done with Area before
the end of the month, and the remaining Arc2D stuff (and whatever else
is left in geom.*) should be pretty trivial to do. So instead of making
a bunch of small commits, I'd like to roll this all into one big geom.*
patch, since any updates to this will need to be sent over to the
gcj-gui-branch guys, and it'd be less work for them.

Does that sound OK?

Oh, and keep up the good work on writing tests! It's very useful!

(If only someone with Mauve CVS would commit them for you..
doesn't seem to be much action on that mailing-list)

/Sven




From MAILER-DAEMON Wed Aug 04 02:45:18 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BsFWX-0008MB-Uy
	for mharc-classpath@gnu.org; Wed, 04 Aug 2004 02:45:18 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BsFWW-0008K3-8i
	for classpath@gnu.org; Wed, 04 Aug 2004 02:45:16 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BsFWV-0008JY-JM
	for classpath@gnu.org; Wed, 04 Aug 2004 02:45:15 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BsFWV-0008JV-Fl
	for classpath@gnu.org; Wed, 04 Aug 2004 02:45:15 -0400
Received: from [213.9.1.84] (helo=box84.elkhouse.de)
	by monty-python.gnu.org with esmtp (TLSv1:RC4-SHA:128) (Exim 4.34)
	id 1BsFSG-0006oV-T2
	for classpath@gnu.org; Wed, 04 Aug 2004 02:40:53 -0400
Received: from 1cust91.vr2.fft4.alter.net ([149.229.88.91] helo=[192.168.1.3])
	by box84.elkhouse.de with asmtp (Exim 4.32)
	id 1BsFU1-00072v-R6; Wed, 04 Aug 2004 08:42:42 +0200
From: Roman Kennke <roman@ontographics.com>
To: Ingo =?ISO-8859-1?Q?Pr=F6tel?= <proetel@aicas.com>
In-Reply-To: <410F9574.8040408@aicas.com>
References: <200408031329.01616.konqueror@gmx.de> <410F9574.8040408@aicas.com>
Content-Type: text/plain; charset=iso-8859-15
Message-Id: <1091608847.726.10.camel@moonlight>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.4.6 
Date: Wed, 04 Aug 2004 10:40:47 +0200
Content-Transfer-Encoding: quoted-printable
Cc: classpath@gnu.org, Michael Koch <konqueror@gmx.de>
Subject: Re: Removing the TARGET_* layer or not ?
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Wed, 04 Aug 2004 06:45:16 -0000

Am Di, den 03.08.2004 schrieb Ingo Pr=F6tel um 15:39:
> Hi Michael,
>=20
> Michael Koch wrote:
> > Hi list, hi Ingo,
> >=20
> >=20
> >=20
> > I just started porting/testing GNU classpath to solaris (2.6). Its not=20
> > really different to other unices (we currently support linux and=20
> > *BSD) but some things are. When I wanted to do the needed changes I=20
> > stumbled over the TARGET_* layer which adds (in my eyes) some=20
> > unneeded complexity. It makes it more hard to read the code and all=20
> > the stuff seem to be implemented either in target/generic or=20
> > target/Linux with no real rule what have to go where. It's just=20
> > difficult to understand. In fact I try to understand it since a long=20
> > time and always fail because of its "indirectness". Personally I=20
> > think some AUTOCONF checks would be more appropriate and would make=20
> > the code much more readable and bugfixable. There are some known bugs=20
> > in it but noone attacked them because none understands the code it=20
> > seems.
> >=20
> > Ingo: Now my question are you really using the TARGET_* system or is=20
> > it only rotting around in GNU classpath ? I really wonder if some=20
> > AUTOCONF macros would be more helpful for you ?
>=20
> Yes, the TARGET-layer is in active use here.
>=20
> The TARGET-layer has two advantages:
> 1. It makes JNI-code more readable,
> 2. It makes porting simple.
>=20
> It makes JNI-code more readable because whenever a native function is cal=
led there is only one macro call. The other possibility would be that one h=
as=20
> ifdefs for every system (possibly for every system+architecture) that one=
 wants to support with this code. This would make the JNI-code difficult to=
=20
> read.
>=20
> It makes porting easier in that one can reuse generic macros and only nee=
ds to override specific macros that differ for a given new system. And this=
=20
> all can happen without modifing the actual JNI-code. So the main step in =
porting is creating a new subdirectory in tartget name it Solaris. Copy ove=
r=20
> all the files in the Linux target. Add this to the include path. Then jus=
t work on the error that the compiler will find and where necessary create=20
> Solaris specific macros.
>=20
> If there is interest we could provide our Solaris, Darwin, and MinGW TARG=
ET-layer. This would probably help to understand how this layer is helpful.
> In any case we can help if anyone needs help understanding the code.

I have thought about how I would design this stuff in order to write
portable code:
1. first of all, try to stick to POSIX standard stuff ;)
2. Of course there are situations where this won't do. In this case I
would split out the non-portable stuff on a case-by-case basis in the
following scheme: Suppose you have the following expressions in a
function (pseudocode):

func() {
  exp_a;
  exp_b;
  exp_c;
}

of which exp_b is not portable. Then make a function call out of it:

func() {
  exp_a;
  do_b();
  exp_c;
}

and define the function in a file in a file in an OS-specific subdir.

the autotools should check for OSs/arches and include the correct
subdirs here. These functions must be kept minimal, in order to avoid
the effect that Michael mentioned, where you have to fix a bug multiple
times.
Note, that with this system we would avoid the use of macros altogether
(You see, I don't like them ;) I better go with a well-thought-out
design). Sure, there is the 'overhead' of 1 additional function call.
But:
1. IMO this it is still better to write readable code (hey, this is free
software and the code is supposed to be read).
2. a compiler should (when in optimization mode) try to inline such
small function calls.

As I said: I strongly vote for the removal of the TARGET_* stuff and go
with a good,readable,debuggable,etc design instead.

Best regards,
Roman





From MAILER-DAEMON Wed Aug 04 04:10:17 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BsGqn-0000K2-GQ
	for mharc-classpath@gnu.org; Wed, 04 Aug 2004 04:10:17 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BsGqm-0000Jw-40
	for classpath@gnu.org; Wed, 04 Aug 2004 04:10:16 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BsGqk-0000JV-7S
	for classpath@gnu.org; Wed, 04 Aug 2004 04:10:15 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BsGqh-0000JM-K3
	for classpath@gnu.org; Wed, 04 Aug 2004 04:10:14 -0400
Received: from [212.227.126.173] (helo=moutng.kundenserver.de)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BsGmq-0007Xs-Hb
	for classpath@gnu.org; Wed, 04 Aug 2004 04:06:12 -0400
Received: from [212.227.126.206] (helo=mrelayng.kundenserver.de)
	by moutng.kundenserver.de with esmtp (Exim 3.35 #1)
	id 1BsGmp-0005au-00; Wed, 04 Aug 2004 10:06:11 +0200
Received: from [212.126.219.82] (helo=aicas.com)
	by mrelayng.kundenserver.de with asmtp (TLSv1:RC4-MD5:128)
	(Exim 3.35 #1) id 1BsGmp-0008HC-00; Wed, 04 Aug 2004 10:06:11 +0200
Message-ID: <411098F1.50509@aicas.com>
Date: Wed, 04 Aug 2004 10:06:09 +0200
From: =?ISO-8859-15?Q?Ingo_Pr=F6tel?= <proetel@aicas.com>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.6) Gecko/20040113
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: Roman Kennke <roman@ontographics.com>
References: <200408031329.01616.konqueror@gmx.de>	 <410F9574.8040408@aicas.com>
	<1091608847.726.10.camel@moonlight>
In-Reply-To: <1091608847.726.10.camel@moonlight>
X-Enigmail-Version: 0.83.6.0
X-Enigmail-Supports: pgp-inline, pgp-mime
Content-Type: text/plain; charset=iso-8859-15; format=flowed
Content-Transfer-Encoding: 8bit
X-Provags-ID: kundenserver.de abuse@kundenserver.de
	auth:ebd3e0372cc3ac236a2e2611be1a8528
Cc: classpath@gnu.org, Michael Koch <konqueror@gmx.de>
Subject: Re: Removing the TARGET_* layer or not ?
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Wed, 04 Aug 2004 08:10:16 -0000

Hi,

Roman Kennke wrote:
> Am Di, den 03.08.2004 schrieb Ingo Prötel um 15:39:
> 
>>Hi Michael,
>>
>>Michael Koch wrote:
>>
>>>Hi list, hi Ingo,
>>>
>>>
>>>
>>>I just started porting/testing GNU classpath to solaris (2.6). Its not 
>>>really different to other unices (we currently support linux and 
>>>*BSD) but some things are. When I wanted to do the needed changes I 
>>>stumbled over the TARGET_* layer which adds (in my eyes) some 
>>>unneeded complexity. It makes it more hard to read the code and all 
>>>the stuff seem to be implemented either in target/generic or 
>>>target/Linux with no real rule what have to go where. It's just 
>>>difficult to understand. In fact I try to understand it since a long 
>>>time and always fail because of its "indirectness". Personally I 
>>>think some AUTOCONF checks would be more appropriate and would make 
>>>the code much more readable and bugfixable. There are some known bugs 
>>>in it but noone attacked them because none understands the code it 
>>>seems.
>>>
>>>Ingo: Now my question are you really using the TARGET_* system or is 
>>>it only rotting around in GNU classpath ? I really wonder if some 
>>>AUTOCONF macros would be more helpful for you ?
>>
>>Yes, the TARGET-layer is in active use here.
>>
>>The TARGET-layer has two advantages:
>>1. It makes JNI-code more readable,
>>2. It makes porting simple.
>>
>>It makes JNI-code more readable because whenever a native function is called there is only one macro call. The other possibility would be that one has 
>>ifdefs for every system (possibly for every system+architecture) that one wants to support with this code. This would make the JNI-code difficult to 
>>read.
>>
>>It makes porting easier in that one can reuse generic macros and only needs to override specific macros that differ for a given new system. And this 
>>all can happen without modifing the actual JNI-code. So the main step in porting is creating a new subdirectory in tartget name it Solaris. Copy over 
>>all the files in the Linux target. Add this to the include path. Then just work on the error that the compiler will find and where necessary create 
>>Solaris specific macros.
>>
>>If there is interest we could provide our Solaris, Darwin, and MinGW TARGET-layer. This would probably help to understand how this layer is helpful.
>>In any case we can help if anyone needs help understanding the code.
> 
> 
> I have thought about how I would design this stuff in order to write
> portable code:
> 1. first of all, try to stick to POSIX standard stuff ;)
> 2. Of course there are situations where this won't do. In this case I
> would split out the non-portable stuff on a case-by-case basis in the
> following scheme: Suppose you have the following expressions in a
> function (pseudocode):
> 
> func() {
>   exp_a;
>   exp_b;
>   exp_c;
> }
> 
> of which exp_b is not portable. Then make a function call out of it:
> 
> func() {
>   exp_a;
>   do_b();
     TARGET_NATIVE_B();  // this is what it currently looks like
>   exp_c;
> }
> 
> and define the function in a file in a file in an OS-specific subdir.
This is actually what the whole idea is about. Except that macros are used instead of functions.
The problem with functions is that many embedded compilers cannot inline function calls. This would be ok if the overhead would really be only one 
function call but if you got through the JNI-code and replace every TARGET-macro with a function call you will see that there are more function calls 
involved. Maybe we could get some mixed mode where we leave the macros in place and the macros for Linux just point to functions. This would make the 
Linux code more debuggable.
> 
> the autotools should check for OSs/arches and include the correct
> subdirs here. These functions must be kept minimal, in order to avoid
> the effect that Michael mentioned, where you have to fix a bug multiple
> times.
If one fixes a bug in the generic part it will be automatically picked up in every system that uses the generic stuff. If a system does not use the 
generic stuff most likely there will be no system that will depend on this bug fix. Especially if one fixes a bug somewhere this might as a side 
effect break another system that depends on the bogous behavior. There is no way of testing it if one doesn't have a target setup to test it on. So 
sometimes this sort of isolation is good.
> Note, that with this system we would avoid the use of macros altogether
> (You see, I don't like them ;) I better go with a well-thought-out
> design). Sure, there is the 'overhead' of 1 additional function call.
That is basically on function call for every TARGET-macro.
> But:
> 1. IMO this it is still better to write readable code (hey, this is free
> software and the code is supposed to be read).
> 2. a compiler should (when in optimization mode) try to inline such
> small function calls.
I guess that might work with gcc on  Linux / Solaris but as a general rule?
> 
> As I said: I strongly vote for the removal of the TARGET_* stuff and go
> with a good,readable,debuggable,etc design instead.
> 
> Best regards,
> Roman
> 
> 
> 

Ingo

-- 
Ingo Prötel                                          proetel@aicas.com
aicas GmbH                                        http://www.aicas.com
Haid-und-Neu-Str. 18                        phone   +49 721 663 968-32
76131 Karlsruhe                             fax     +49 721 663 968-93
Germany



From MAILER-DAEMON Wed Aug 04 04:19:52 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BsH04-0002Ic-OB
	for mharc-classpath@gnu.org; Wed, 04 Aug 2004 04:19:52 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BsH02-0002IX-2V
	for classpath@gnu.org; Wed, 04 Aug 2004 04:19:50 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BsH01-0002IL-6W
	for classpath@gnu.org; Wed, 04 Aug 2004 04:19:49 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BsH00-0002II-P7
	for classpath@gnu.org; Wed, 04 Aug 2004 04:19:49 -0400
Received: from [212.227.126.173] (helo=moutng.kundenserver.de)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BsGw6-0000Aa-P7
	for classpath@gnu.org; Wed, 04 Aug 2004 04:15:47 -0400
Received: from [212.227.126.179] (helo=mrelayng.kundenserver.de)
	by moutng.kundenserver.de with esmtp (Exim 3.35 #1)
	id 1BsGw5-0007ei-00; Wed, 04 Aug 2004 10:15:45 +0200
Received: from [212.126.219.82] (helo=aicas.com)
	by mrelayng.kundenserver.de with asmtp (TLSv1:RC4-MD5:128)
	(Exim 3.35 #1) id 1BsGw4-0005JZ-00; Wed, 04 Aug 2004 10:15:45 +0200
Message-ID: <41109B2E.9080701@aicas.com>
Date: Wed, 04 Aug 2004 10:15:42 +0200
From: =?ISO-8859-15?Q?Ingo_Pr=F6tel?= <proetel@aicas.com>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.6) Gecko/20040113
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: Michael Koch <konqueror@gmx.de>
References: <200408031329.01616.konqueror@gmx.de> <410F9574.8040408@aicas.com>
	<200408031601.34628.konqueror@gmx.de>
In-Reply-To: <200408031601.34628.konqueror@gmx.de>
X-Enigmail-Version: 0.83.6.0
X-Enigmail-Supports: pgp-inline, pgp-mime
Content-Type: multipart/mixed; boundary="------------050205050809040505080501"
X-Provags-ID: kundenserver.de abuse@kundenserver.de
	auth:ebd3e0372cc3ac236a2e2611be1a8528
Cc: classpath@gnu.org
Subject: Re: Removing the TARGET_* layer or not ?
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Wed, 04 Aug 2004 08:19:50 -0000

This is a multi-part message in MIME format.
--------------050205050809040505080501
Content-Type: text/plain; charset=ISO-8859-15; format=flowed
Content-Transfer-Encoding: 8bit

[I'm sending this again since I did not get it back via the list]

Hi Michael,

Michael Koch wrote:
> Am Dienstag, 3. August 2004 15:39 schrieb Ingo Prötel:
> 
>>Hi Michael,
>>
>>Michael Koch wrote:
>>
>>>Hi list, hi Ingo,
>>>
>>>
>>>
>>>I just started porting/testing GNU classpath to solaris (2.6).
>>>Its not really different to other unices (we currently support
>>>linux and *BSD) but some things are. When I wanted to do the
>>>needed changes I stumbled over the TARGET_* layer which adds (in
>>>my eyes) some unneeded complexity. It makes it more hard to read
>>>the code and all the stuff seem to be implemented either in
>>>target/generic or target/Linux with no real rule what have to go
>>>where. It's just difficult to understand. In fact I try to
>>>understand it since a long time and always fail because of its
>>>"indirectness". Personally I think some AUTOCONF checks would be
>>>more appropriate and would make the code much more readable and
>>>bugfixable. There are some known bugs in it but noone attacked
>>>them because none understands the code it seems.
>>>
>>>Ingo: Now my question are you really using the TARGET_* system or
>>>is it only rotting around in GNU classpath ? I really wonder if
>>>some AUTOCONF macros would be more helpful for you ?
>>
>>Yes, the TARGET-layer is in active use here.
>>
>>The TARGET-layer has two advantages:
>>1. It makes JNI-code more readable,
>>2. It makes porting simple.
>>
>>It makes JNI-code more readable because whenever a native function
>>is called there is only one macro call. The other possibility would
>>be that one has ifdefs for every system (possibly for every
>>system+architecture) that one wants to support with this code. This
>>would make the JNI-code difficult to read.
> 
> 
> The MACRO names are really long and they dont really make the code 
> more readable for me. Perhaps for the compiler ... I dont know.
> 
> 
>>It makes porting easier in that one can reuse generic macros and
>>only needs to override specific macros that differ for a given new
>>system. And this all can happen without modifing the actual
>>JNI-code. So the main step in porting is creating a new
>>subdirectory in tartget name it Solaris. Copy over all the files in
>>the Linux target. Add this to the include path. Then just work on
>>the error that the compiler will find and where necessary create
>>Solaris specific macros.
> 
> 
> I tried to port to Solaris and you system just made it hard for me 
> (and I know mjw had problems with it too when he dbugged some 
> filesystem stuff).
> 
> 
>>If there is interest we could provide our Solaris, Darwin, and
>>MinGW TARGET-layer. This would probably help to understand how this
>>layer is helpful. In any case we can help if anyone needs help
>>understanding the code.
> 
> 
> I don't get the benefit of the whole system. BSD's are already 
> suppored by the "Linux" layer. I guess Darwin (as its a BSD too) is 
> supported too and if not its easy to do so. I tried to support 
> solaris in the "Linux" layer and I needed to add around 10 lines for 
> it. So why add a lot oc code duplication to support an additional OS 
> when its enough to add 10 lines that are much more easy to 
> understand, adds no bloat and is more easy to maintain ? I haven't 
> tried mingw yet but I guess its similar easy.
> 
> Sorry, but I'm currently upset because this layer made it hard for me 
> to do something normally really simple. Autoconf was created to make 
> it easy to write portable apps. We use it already in GNU classpath. 
> Why not use it to write portable apps in it ?
> 
> 
> Michael
> 
You are quite right. Solaris is still pretty close to Linux. So the only difference is in the target_native_network.h file.
The difference to Darwin is even less. But we have been able to port to those platforms without having to modify the actual JNI-code.
If the TRAGET-layer would be removed all changes would need to be done in the JNI-file. We try to support embedded targets which use at least seven
different operating systems, which often have no native support for 64-bit values, or have weired endianess. If we would need to add all these
differences as as ifdefs in the JNI-code it would make this code difficult to read.
Debugging the macros is actually a hard thing. But once a macro works correctly it normally does not need to be changed any more. Much more often the
code around the macro must be changed and that is easier if it is not cluttered with ifdefs.

I have attached the target_native_network.h files for Solaris and Darwin. I hope that helps you a bit.

Ingo


-- 
Ingo Prötel                                          proetel@aicas.com
aicas GmbH                                        http://www.aicas.com
Haid-und-Neu-Str. 18                        phone   +49 721 663 968-32
76131 Karlsruhe                             fax     +49 721 663 968-93
Germany


--------------050205050809040505080501
Content-Type: text/plain;
 name="Darwin-target_native_network.h"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline;
 filename="Darwin-target_native_network.h"

/* ???.h - ???
   Copyright (C) 1998 Free Software Foundation, Inc.

This file is part of GNU Classpath.

GNU Classpath is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 2, or (at your option)
any later version.
 
GNU Classpath is distributed in the hope that it will be useful, but
WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
General Public License for more details.

You should have received a copy of the GNU General Public License
along with GNU Classpath; see the file COPYING.  If not, write to the
Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
02111-1307 USA.

Linking this library statically or dynamically with other modules is
making a combined work based on this library.  Thus, the terms and
conditions of the GNU General Public License cover the whole
combination.

As a special exception, the copyright holders of this library give you
permission to link this library with independent modules to produce an
executable, regardless of the license terms of these independent
modules, and to copy and distribute the resulting executable under
terms of your choice, provided that you also meet, for each linked
independent module, the terms and conditions of the license of that
module.  An independent module is a module which is not derived from
or based on this library.  If you modify this library, you may extend
this exception to your version of the library, but you are not
obligated to do so.  If you do not wish to do so, delete this
exception statement from your version. */

/*
Description: Darwin/Mac OS-X target defintions of network functions
Systems    : Darwin/Mac OS-X

*/

#ifndef __TARGET_NATIVE_NETWORK__
#define __TARGET_NATIVE_NETWORK__

/****************************** Includes *******************************/
/* do not move; needed here because of some macro definitions */
#include <config.h>

#include <stdlib.h>

/****************** Conditional compilation switches *******************/

/***************************** Constants *******************************/

/***************************** Datatypes *******************************/

/***************************** Variables *******************************/

/****************************** Macros *********************************/

#define TARGET_NATIVE_NETWORK_SOCKET_SEND(socketDescriptor,buffer,length,bytesSent) \
  do { \
    bytesSent=send(socketDescriptor,buffer,length,0); \
  } while (0)

#define TARGET_NATIVE_NETWORK_SOCKET_SEND_WITH_ADDRESS_PORT(socketDescriptor,buffer,length,address,port,bytesSent) \
  do { \
    struct sockaddr_in __socketAddress; \
    \
    memset(&__socketAddress,0,sizeof(__socketAddress)); \
    __socketAddress.sin_family      = AF_INET; \
    __socketAddress.sin_addr.s_addr = htonl(address); \
    __socketAddress.sin_port        = htons((short)port); \
    bytesSent=sendto(socketDescriptor,buffer,length,0,(struct sockaddr*)&__socketAddress,sizeof(__socketAddress)); \
  } while (0)

/***************************** Functions *******************************/

#ifdef __cplusplus
extern "C" {
#endif

#ifdef __cplusplus
}
#endif

/* include rest of definitions from generic file (do not move it to 
   another position!) */
#include "target_generic_network.h"

#endif /* __TARGET_NATIVE_NETWORK__ */

/* end of file */



--------------050205050809040505080501
Content-Type: text/plain;
 name="SunOS-target_native_network.h"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline;
 filename="SunOS-target_native_network.h"

/* ???.h - ???
   Copyright (C) 1998 Free Software Foundation, Inc.

This file is part of GNU Classpath.

GNU Classpath is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 2, or (at your option)
any later version.
 
GNU Classpath is distributed in the hope that it will be useful, but
WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
General Public License for more details.

You should have received a copy of the GNU General Public License
along with GNU Classpath; see the file COPYING.  If not, write to the
Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
02111-1307 USA.

Linking this library statically or dynamically with other modules is
making a combined work based on this library.  Thus, the terms and
conditions of the GNU General Public License cover the whole
combination.

As a special exception, the copyright holders of this library give you
permission to link this library with independent modules to produce an
executable, regardless of the license terms of these independent
modules, and to copy and distribute the resulting executable under
terms of your choice, provided that you also meet, for each linked
independent module, the terms and conditions of the license of that
module.  An independent module is a module which is not derived from
or based on this library.  If you modify this library, you may extend
this exception to your version of the library, but you are not
obligated to do so.  If you do not wish to do so, delete this
exception statement from your version. */

/*
Description: SunOS/Solaris target defintions of network functions
Systems    : all
*/

#ifndef __TARGET_NATIVE_NETWORK__
#define __TARGET_NATIVE_NETWORK__

/****************************** Includes *******************************/
/* do not move; needed here because of some macro definitions */
#include <config.h>

#include <stdlib.h>
#include <stropts.h>
#include <sys/conf.h>
#include <netinet/in_systm.h>
#include <sys/ioctl.h>

/****************** Conditional compilation switches *******************/

/***************************** Constants *******************************/

/***************************** Datatypes *******************************/

/***************************** Variables *******************************/

/****************************** Macros *********************************/

/***********************************************************************\
* Name       : TARGET_NATIVE_NETWORK_GET_HOSTNAME_BY_NAME
* Purpose    : get hostname by name
* Input      : name           - hostname
*              maxAddressSize - max. size of address array
* Output     : addresses    - host addresses (array, in host byte
*                             order!)
*              addressCount - number of entries in address array
*              result       - TARGET_NATIVE_OK if no error occurred,
*                             TARGET_NATIVE_ERROR otherwise
* Return     : -
* Side-effect: unknown
* Notes      : -
\***********************************************************************/

#ifndef TARGET_NATIVE_NETWORK_GET_HOSTNAME_BY_NAME
  #include <netdb.h>
  #define TARGET_NATIVE_NETWORK_GET_HOSTNAME_BY_NAME(name,addresses,maxAddressSize,addressCount,result) \
    do { \
      struct hostent __hostEntry; \
      char           __buffer[5000]; \
      int            __h_errno; \
      \
      addressCount = 0; \
      \
      if (gethostbyname_r(name, &__hostEntry, __buffer, sizeof(__buffer), &__h_errno) != NULL) \
        { \
          while ((addressCount < maxAddressSize) && (__hostEntry.h_addr_list[addressCount] != NULL)) \
            { \
              addresses[addressCount] = ntohl(*(int*)(__hostEntry.h_addr_list[addressCount])); \
              addressCount++; \
            } \
          result = TARGET_NATIVE_OK; \
        } \
      else \
        { \
          result = TARGET_NATIVE_ERROR; \
        } \
    } while (0)
#endif

/***********************************************************************\
* Name       : TARGET_NATIVE_NETWORK_SOCKET_RECEIVE_AVAILABLE
* Purpose    : get number of available bytes for receive
* Input      : socketDescriptor - socket descriptor
* Output     : bytesAvailable - available bytes for receive
*            : result         - TARGET_NATIVE_OK if no error occurred,
*                               TARGET_NATIVE_ERROR otherwise
* Return     : -
* Side-effect: unknown
* Notes      : -
\***********************************************************************/

#define TARGET_NATIVE_NETWORK_SOCKET_RECEIVE_AVAILABLE(socketDescriptor,bytesAvailable,result) \
  do { \
    int __value; \
    \
    bytesAvailable=0; \
    \
    result = (ioctl(socketDescriptor, I_NREAD, &__value) == 0)?TARGET_NATIVE_OK:TARGET_NATIVE_ERROR; \
    if (result == TARGET_NATIVE_OK) \
      { \
        bytesAvailable = __value; \
      } \
  } while (0)

#if 1
#include <sys/types.h>
#include <sys/socket.h>
#include <arpa/inet.h>
#include <netinet/in.h>
#include <netdb.h>
#define TARGET_NATIVE_NETWORK_SOCKET_CONNECT(socketDescriptor,address,port,result) \
  do { \
    struct sockaddr_in __socketAddress; \
    int nnn; \
    \
fprintf(stderr, "address %08x %d\n", address, (int)port); \
    memset(&__socketAddress,0,sizeof(__socketAddress)); \
    __socketAddress.sin_family      = AF_INET; \
    __socketAddress.sin_addr.s_addr = htonl(address); \
    __socketAddress.sin_port        = htons(((short)port)); \
    \
    nnn=connect(socketDescriptor,(struct sockaddr*)&__socketAddress,sizeof(__socketAddress)); \
    result=(nnn==0)?TARGET_NATIVE_OK:TARGET_NATIVE_ERROR; \
fprintf(stderr,"%s,%d: connect: %d %d %d\n",__FILE__,__LINE__,socketDescriptor,nnn,errno); \
  } while (0)
#endif /* 0 */

/***********************************************************************\
* Name       : TARGET_NATIVE_NETWORK_SOCKET_SEND
* Purpose    : send data to socket
* Input      : socketDescriptor - socket descriptor
*            : buffer  - data to send
*              length  - length of data to send
* Output     : bytesSent - number of bytes sent, -1 otherwise
* Side-effect: unknown
* Notes      : -
\***********************************************************************/

#ifndef TARGET_NATIVE_NETWORK_SOCKET_SEND
  #include <sys/types.h>
  #include <sys/socket.h>
  #include <netinet/in.h>
  #define TARGET_NATIVE_NETWORK_SOCKET_SEND(socketDescriptor, buffer, length, bytesSent) \
    do { \
      bytesSent = send(socketDescriptor, \
                       buffer, \
                       length, \
                       0 /* MSG_NOSIGNAL not supported by Solaris */ \
                      ); \
    } while (0)
#endif

/***********************************************************************\
* Name       : TARGET_NATIVE_NETWORK_SOCKET_SEND_WITH_ADDRESS_PORT
* Purpose    : send data to socket
* Input      : socketDescriptor - socket descriptor
*            : buffer  - data to send
*              length  - length of data to send
*              Address - to address (NOT in network byte order!)
*              Port    - to port (NOT in network byte order!)
* Output     : bytesSent - number of bytes sent, -1 otherwise
* Side-effect: unknown
* Notes      : -
\***********************************************************************/

#ifndef TARGET_NATIVE_NETWORK_SOCKET_SEND_WITH_ADDRESS_PORT
  #include <sys/types.h>
  #include <sys/socket.h>
  #include <netinet/in.h>
  #define TARGET_NATIVE_NETWORK_SOCKET_SEND_WITH_ADDRESS_PORT(socketDescriptor, buffer, length, address, port, bytesSent) \
    do { \
      struct sockaddr_in __socketAddress; \
      \
      memset(&__socketAddress,0,sizeof(__socketAddress)); \
      __socketAddress.sin_family      = AF_INET; \
      __socketAddress.sin_addr.s_addr = htonl(address); \
      __socketAddress.sin_port        = htons((short)port); \
      bytesSent = sendto(socketDescriptor, \
                         buffer, \
                         length, \
			 0 /* MSG_NOSIGNAL not supported by solaris */, \
			 (struct sockaddr*)&__socketAddress, \
			 sizeof(__socketAddress) \
                        ); \
    } while (0)
#endif

/***************************** Functions *******************************/

#ifdef __cplusplus
extern "C" {
#endif

#ifdef __cplusplus
}
#endif

/* include rest of definitions from generic file (do not move it to 
   another position!) */
#include "target_generic_network.h"

#endif /* __TARGET_NATIVE_NETWORK__ */

/* end of file */


--------------050205050809040505080501--



From MAILER-DAEMON Wed Aug 04 04:31:53 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BsHBh-0005Pj-AW
	for mharc-classpath@gnu.org; Wed, 04 Aug 2004 04:31:53 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BsHBe-0005N2-JT
	for classpath@gnu.org; Wed, 04 Aug 2004 04:31:50 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BsHBc-0005Lj-I0
	for classpath@gnu.org; Wed, 04 Aug 2004 04:31:50 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BsHBc-0005Lg-5f
	for classpath@gnu.org; Wed, 04 Aug 2004 04:31:48 -0400
Received: from [62.253.162.94] (helo=mta10-svc.ntlworld.com)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BsH7p-0001aT-SO
	for classpath@gnu.org; Wed, 04 Aug 2004 04:27:54 -0400
Received: from cuddles.cambridge.redhat.com ([81.96.64.123])
	by mta10-svc.ntlworld.com
	(InterMail vM.4.01.03.37 201-229-121-137-20020806) with ESMTP id
	<20040804082637.FSLB10838.mta10-svc.ntlworld.com@cuddles.cambridge.redhat.com>;
	Wed, 4 Aug 2004 09:26:37 +0100
Received: from redhat.com (localhost.localdomain [127.0.0.1])
	by cuddles.cambridge.redhat.com (8.12.8/8.12.8) with ESMTP id
	i748RQ2v026649; Wed, 4 Aug 2004 09:27:26 +0100
Received: (from aph@localhost)
	by redhat.com (8.12.8/8.12.8/Submit) id i748RPWf026645;
	Wed, 4 Aug 2004 09:27:25 +0100
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
Message-ID: <16656.40429.72617.324082@cuddles.cambridge.redhat.com>
Date: Wed, 4 Aug 2004 09:27:25 +0100
From: Andrew Haley <aph@redhat.com>
To: =?ISO-8859-15?Q?Ingo_Pr=F6tel?= <proetel@aicas.com>
In-Reply-To: <411098F1.50509@aicas.com>
References: <200408031329.01616.konqueror@gmx.de> <410F9574.8040408@aicas.com>
	<1091608847.726.10.camel@moonlight> <411098F1.50509@aicas.com>
X-Mailer: VM 7.14 under Emacs 21.3.50.1
Cc: classpath@gnu.org, Michael Koch <konqueror@gmx.de>
Subject: Re: Removing the TARGET_* layer or not ?
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Wed, 04 Aug 2004 08:31:50 -0000

Ingo Prötel writes:

 > Roman Kennke wrote:
 > > I have thought about how I would design this stuff in order to write
 > > portable code:
 > > 1. first of all, try to stick to POSIX standard stuff ;)
 > > 2. Of course there are situations where this won't do. In this case I
 > > would split out the non-portable stuff on a case-by-case basis in the
 > > following scheme: Suppose you have the following expressions in a
 > > function (pseudocode):
 > > 
 > > func() {
 > >   exp_a;
 > >   exp_b;
 > >   exp_c;
 > > }
 > > 
 > > of which exp_b is not portable. Then make a function call out of it:
 > > 
 > > func() {
 > >   exp_a;
 > >   do_b();
 >      TARGET_NATIVE_B();  // this is what it currently looks like
 > >   exp_c;
 > > }
 > > 
 > > and define the function in a file in a file in an OS-specific subdir.

 > This is actually what the whole idea is about. Except that macros
 > are used instead of functions.  The problem with functions is that
 > many embedded compilers cannot inline function calls. This would be
 > ok if the overhead would really be only one function call but if
 > you got through the JNI-code and replace every TARGET-macro with a
 > function call you will see that there are more function calls
 > involved. Maybe we could get some mixed mode where we leave the
 > macros in place and the macros for Linux just point to
 > functions. This would make the Linux code more debuggable.

I guess as long as we're stuck with C90 we can't use inline.
Am I right in thinking that GNU Classpath targets C90?

IMO, we shouldn't really be designing Classpath around the
infelicities of unfree compilers.

> > Note, that with this system we would avoid the use of macros altogether
 > > (You see, I don't like them ;) I better go with a well-thought-out
 > > design). Sure, there is the 'overhead' of 1 additional function call.

Macros are bad for a bunch of reasons, esp. debuggability and type
safety.

 > That is basically on function call for every TARGET-macro.
 > > But:
 > > 1. IMO this it is still better to write readable code (hey, this is free
 > > software and the code is supposed to be read).
 > > 2. a compiler should (when in optimization mode) try to inline such
 > > small function calls.

 > I guess that might work with gcc on Linux / Solaris but as a
 > general rule?

Well, that's the question: do we make Classpath harder to debug / read
in order to optimize the performance of some unfree C compilers?

Andrew.



From MAILER-DAEMON Wed Aug 04 05:45:23 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BsIKp-0004dK-LF
	for mharc-classpath@gnu.org; Wed, 04 Aug 2004 05:45:23 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BsIKn-0004bh-FO
	for classpath@gnu.org; Wed, 04 Aug 2004 05:45:21 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BsIKm-0004bV-OD
	for classpath@gnu.org; Wed, 04 Aug 2004 05:45:21 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BsIKm-0004bL-Lc
	for classpath@gnu.org; Wed, 04 Aug 2004 05:45:20 -0400
Received: from [212.6.122.16] (helo=mail2.ewetel.de)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BsIGy-0001lW-Oo
	for classpath@gnu.org; Wed, 04 Aug 2004 05:41:25 -0400
Received: from majestix.konqueror.de (dynadsl-080-228-94-135.ewetel.net
	[80.228.94.135])
	by mail2.ewetel.de (8.12.1/8.12.9) with ESMTP id i749eJO0022185;
	Wed, 4 Aug 2004 11:40:20 +0200 (MEST)
Received: from miraculix.konqueror.de ([192.168.1.1])
	by majestix.konqueror.de with esmtp (Exim 3.35 #1 (Debian))
	id 1BsIOK-0002ZI-00; Wed, 04 Aug 2004 11:49:00 +0200
From: Michael Koch <konqueror@gmx.de>
To: Andrew Haley <aph@redhat.com>
Date: Wed, 4 Aug 2004 11:40:52 +0200
User-Agent: KMail/1.6.2
References: <200408031329.01616.konqueror@gmx.de> <411098F1.50509@aicas.com>
	<16656.40429.72617.324082@cuddles.cambridge.redhat.com>
In-Reply-To: <16656.40429.72617.324082@cuddles.cambridge.redhat.com>
MIME-Version: 1.0
Content-Disposition: inline
Content-Type: text/plain;
  charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Message-Id: <200408041140.52568.konqueror@gmx.de>
X-CheckCompat: OK
Cc: classpath@gnu.org
Subject: Re: Removing the TARGET_* layer or not ?
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Wed, 04 Aug 2004 09:45:22 -0000

Am Mittwoch, 4. August 2004 10:27 schrieb Andrew Haley:
> Ingo Pr=F6tel writes:
>  > Roman Kennke wrote:
>  > > I have thought about how I would design this stuff in order to
>  > > write portable code:
>  > > 1. first of all, try to stick to POSIX standard stuff ;)
>  > > 2. Of course there are situations where this won't do. In this
>  > > case I would split out the non-portable stuff on a
>  > > case-by-case basis in the following scheme: Suppose you have
>  > > the following expressions in a function (pseudocode):
>  > >
>  > > func() {
>  > >   exp_a;
>  > >   exp_b;
>  > >   exp_c;
>  > > }
>  > >
>  > > of which exp_b is not portable. Then make a function call out
>  > > of it:
>  > >
>  > > func() {
>  > >   exp_a;
>  > >   do_b();
>  >
>  >      TARGET_NATIVE_B();  // this is what it currently looks like
>  >
>  > >   exp_c;
>  > > }
>  > >
>  > > and define the function in a file in a file in an OS-specific
>  > > subdir.
>  >
>  > This is actually what the whole idea is about. Except that
>  > macros are used instead of functions.  The problem with
>  > functions is that many embedded compilers cannot inline function
>  > calls. This would be ok if the overhead would really be only one
>  > function call but if you got through the JNI-code and replace
>  > every TARGET-macro with a function call you will see that there
>  > are more function calls involved. Maybe we could get some mixed
>  > mode where we leave the macros in place and the macros for Linux
>  > just point to
>  > functions. This would make the Linux code more debuggable.
>
> I guess as long as we're stuck with C90 we can't use inline.
> Am I right in thinking that GNU Classpath targets C90?
>
> IMO, we shouldn't really be designing Classpath around the
> infelicities of unfree compilers.
>
> > > Note, that with this system we would avoid the use of macros
> > > altogether
> > >
>  > > (You see, I don't like them ;) I better go with a
>  > > well-thought-out design). Sure, there is the 'overhead' of 1
>  > > additional function call.
>
> Macros are bad for a bunch of reasons, esp. debuggability and type
> safety.
>
>  > That is basically on function call for every TARGET-macro.
>  >
>  > > But:
>  > > 1. IMO this it is still better to write readable code (hey,
>  > > this is free software and the code is supposed to be read).
>  > > 2. a compiler should (when in optimization mode) try to inline
>  > > such small function calls.
>  >
>  > I guess that might work with gcc on Linux / Solaris but as a
>  > general rule?
>
> Well, that's the question: do we make Classpath harder to debug /
> read in order to optimize the performance of some unfree C
> compilers?

It's not bad to support other compilers too but this should should not=20
be used to make the code unreadable and hard to understand. If its=20
hard to achieve better drop support for some compilers.


Michael



From MAILER-DAEMON Wed Aug 04 05:56:51 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BsIVu-0007IB-Ta
	for mharc-classpath@gnu.org; Wed, 04 Aug 2004 05:56:50 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BsIVt-0007Hv-8n
	for classpath@gnu.org; Wed, 04 Aug 2004 05:56:49 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BsIVr-0007HW-VW
	for classpath@gnu.org; Wed, 04 Aug 2004 05:56:48 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BsIVr-0007HF-UH
	for classpath@gnu.org; Wed, 04 Aug 2004 05:56:47 -0400
Received: from [212.6.122.32] (helo=mail5.ewetel.de)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BsIRq-0003FH-Sz
	for classpath@gnu.org; Wed, 04 Aug 2004 05:52:39 -0400
Received: from majestix.konqueror.de (dynadsl-080-228-94-135.ewetel.net
	[80.228.94.135])
	by mail5.ewetel.de (8.12.1/8.12.9) with ESMTP id i749oL8D000539;
	Wed, 4 Aug 2004 11:50:22 +0200 (MEST)
Received: from miraculix.konqueror.de ([192.168.1.1])
	by majestix.konqueror.de with esmtp (Exim 3.35 #1 (Debian))
	id 1BsIY2-0002Zj-00; Wed, 04 Aug 2004 11:59:02 +0200
From: Michael Koch <konqueror@gmx.de>
To: classpath@gnu.org
Date: Wed, 4 Aug 2004 11:50:54 +0200
User-Agent: KMail/1.6.2
References: <200408031329.01616.konqueror@gmx.de>
	<1091608847.726.10.camel@moonlight> <411098F1.50509@aicas.com>
In-Reply-To: <411098F1.50509@aicas.com>
MIME-Version: 1.0
Content-Disposition: inline
Content-Type: text/plain;
  charset="iso-8859-15"
Content-Transfer-Encoding: quoted-printable
Message-Id: <200408041150.54758.konqueror@gmx.de>
X-CheckCompat: OK
Cc: 
Subject: Re: Removing the TARGET_* layer or not ?
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Wed, 04 Aug 2004 09:56:49 -0000

Am Mittwoch, 4. August 2004 10:06 schrieb Ingo Pr=F6tel:
> Hi,
>
> Roman Kennke wrote:
> > Am Di, den 03.08.2004 schrieb Ingo Pr=F6tel um 15:39:
> >>Hi Michael,
> >>
> >>Michael Koch wrote:
> >>>Hi list, hi Ingo,
> >>>
> >>>
> >>>
> >>>I just started porting/testing GNU classpath to solaris (2.6).
> >>> Its not really different to other unices (we currently support
> >>> linux and *BSD) but some things are. When I wanted to do the
> >>> needed changes I stumbled over the TARGET_* layer which adds
> >>> (in my eyes) some unneeded complexity. It makes it more hard to
> >>> read the code and all the stuff seem to be implemented either
> >>> in target/generic or target/Linux with no real rule what have
> >>> to go where. It's just difficult to understand. In fact I try
> >>> to understand it since a long time and always fail because of
> >>> its "indirectness". Personally I think some AUTOCONF checks
> >>> would be more appropriate and would make the code much more
> >>> readable and bugfixable. There are some known bugs in it but
> >>> noone attacked them because none understands the code it seems.
> >>>
> >>>Ingo: Now my question are you really using the TARGET_* system
> >>> or is it only rotting around in GNU classpath ? I really wonder
> >>> if some AUTOCONF macros would be more helpful for you ?
> >>
> >>Yes, the TARGET-layer is in active use here.
> >>
> >>The TARGET-layer has two advantages:
> >>1. It makes JNI-code more readable,
> >>2. It makes porting simple.
> >>
> >>It makes JNI-code more readable because whenever a native
> >> function is called there is only one macro call. The other
> >> possibility would be that one has ifdefs for every system
> >> (possibly for every system+architecture) that one wants to
> >> support with this code. This would make the JNI-code difficult
> >> to read.
> >>
> >>It makes porting easier in that one can reuse generic macros and
> >> only needs to override specific macros that differ for a given
> >> new system. And this all can happen without modifing the actual
> >> JNI-code. So the main step in porting is creating a new
> >> subdirectory in tartget name it Solaris. Copy over all the files
> >> in the Linux target. Add this to the include path. Then just
> >> work on the error that the compiler will find and where
> >> necessary create Solaris specific macros.
> >>
> >>If there is interest we could provide our Solaris, Darwin, and
> >> MinGW TARGET-layer. This would probably help to understand how
> >> this layer is helpful. In any case we can help if anyone needs
> >> help understanding the code.
> >
> > I have thought about how I would design this stuff in order to
> > write portable code:
> > 1. first of all, try to stick to POSIX standard stuff ;)
> > 2. Of course there are situations where this won't do. In this
> > case I would split out the non-portable stuff on a case-by-case
> > basis in the following scheme: Suppose you have the following
> > expressions in a function (pseudocode):
> >
> > func() {
> >   exp_a;
> >   exp_b;
> >   exp_c;
> > }
> >
> > of which exp_b is not portable. Then make a function call out of
> > it:
> >
> > func() {
> >   exp_a;
> >   do_b();
>
>      TARGET_NATIVE_B();  // this is what it currently looks like
>
> >   exp_c;
> > }
> >
> > and define the function in a file in a file in an OS-specific
> > subdir.
>
> This is actually what the whole idea is about. Except that macros
> are used instead of functions. The problem with functions is that
> many embedded compilers cannot inline function calls. This would be
> ok if the overhead would really be only one function call but if
> you got through the JNI-code and replace every TARGET-macro with a
> function call you will see that there are more function calls
> involved. Maybe we could get some mixed mode where we leave the
> macros in place and the macros for Linux just point to functions.
> This would make the Linux code more debuggable.
>
> > the autotools should check for OSs/arches and include the correct
> > subdirs here. These functions must be kept minimal, in order to
> > avoid the effect that Michael mentioned, where you have to fix a
> > bug multiple times.
>
> If one fixes a bug in the generic part it will be automatically
> picked up in every system that uses the generic stuff. If a system
> does not use the generic stuff most likely there will be no system
> that will depend on this bug fix. Especially if one fixes a bug
> somewhere this might as a side effect break another system that
> depends on the bogous behavior. There is no way of testing it if
> one doesn't have a target setup to test it on. So sometimes this
> sort of isolation is good.
>
> > Note, that with this system we would avoid the use of macros
> > altogether (You see, I don't like them ;) I better go with a
> > well-thought-out design). Sure, there is the 'overhead' of 1
> > additional function call.
>
> That is basically on function call for every TARGET-macro.

I'm not a friend of this one-OS-one-dir approach. I would more test=20
for features with autoconf and use the features if present. This=20
makes it more easy to port to another arch. When the feature is=20
available there, use it. If not, add some code to make it work.

The TARGET api is really not intutive and the very long MACRO names=20
makes it even harder to understand. Sometimes the names just badly=20
choosed. The "Linux" layer is full of bugs. And I suppose your other=20
ports are too because they are copies of the "Linux" layer. Some bugs=20
are fixed in the classpath version of "Linux". I wonder how many of=20
them were ported to you other ports. When an autoconf approach you=20
wouldnt even think about this. I know that with the autoconf approach=20
the patches need a closed review and more testing before they go in=20
but IMO this is much better then duplicatin bugs over several trees.

I was pretty surprised when ypu wrote yesterday, that you have more=20
"ports" then just "Linux". Why don't you submitted them ? I think it=20
would be a very good idea to support as much OSes upstream as=20
possible, even Plan9 and QNX and ...


Michael



From MAILER-DAEMON Wed Aug 04 06:49:52 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BsJLE-0001tG-1j
	for mharc-classpath@gnu.org; Wed, 04 Aug 2004 06:49:52 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BsJLC-0001tB-TP
	for classpath@gnu.org; Wed, 04 Aug 2004 06:49:50 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BsJLB-0001sy-9m
	for classpath@gnu.org; Wed, 04 Aug 2004 06:49:50 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BsJLB-0001sv-5u
	for classpath@gnu.org; Wed, 04 Aug 2004 06:49:49 -0400
Received: from [62.253.162.41] (helo=mta01-svc.ntlworld.com)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BsJHP-0000b0-UF
	for classpath@gnu.org; Wed, 04 Aug 2004 06:45:56 -0400
Received: from cuddles.cambridge.redhat.com ([81.96.64.123])
	by mta01-svc.ntlworld.com
	(InterMail vM.4.01.03.37 201-229-121-137-20020806) with ESMTP id
	<20040804104520.PUVJ7107.mta01-svc.ntlworld.com@cuddles.cambridge.redhat.com>;
	Wed, 4 Aug 2004 11:45:20 +0100
Received: from redhat.com (localhost.localdomain [127.0.0.1])
	by cuddles.cambridge.redhat.com (8.12.8/8.12.8) with ESMTP id
	i74Ajq2v027142; Wed, 4 Aug 2004 11:45:52 +0100
Received: (from aph@localhost)
	by redhat.com (8.12.8/8.12.8/Submit) id i74AjoCO027138;
	Wed, 4 Aug 2004 11:45:50 +0100
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Message-ID: <16656.48734.711075.788373@cuddles.cambridge.redhat.com>
Date: Wed, 4 Aug 2004 11:45:50 +0100
From: Andrew Haley <aph@redhat.com>
To: Michael Koch <konqueror@gmx.de>
In-Reply-To: <200408041140.52568.konqueror@gmx.de>
References: <200408031329.01616.konqueror@gmx.de> <411098F1.50509@aicas.com>
	<16656.40429.72617.324082@cuddles.cambridge.redhat.com>
	<200408041140.52568.konqueror@gmx.de>
X-Mailer: VM 7.14 under Emacs 21.3.50.1
Cc: classpath@gnu.org
Subject: Re: Removing the TARGET_* layer or not ?
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Wed, 04 Aug 2004 10:49:51 -0000

Michael Koch writes:
 > Am Mittwoch, 4. August 2004 10:27 schrieb Andrew Haley:
 > >
 > > Well, that's the question: do we make Classpath harder to debug /
 > > read in order to optimize the performance of some unfree C
 > > compilers?
 > 
 > It's not bad to support other compilers too but this should should not 
 > be used to make the code unreadable and hard to understand. If its 
 > hard to achieve better drop support for some compilers.

We weren't talking about dropping any support: merely a performance
decrease for some old unfree compilers.

Andrew.



From MAILER-DAEMON Wed Aug 04 06:56:24 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BsJRY-0003J7-6y
	for mharc-classpath@gnu.org; Wed, 04 Aug 2004 06:56:24 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BsJRW-0003J2-O2
	for classpath@gnu.org; Wed, 04 Aug 2004 06:56:22 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BsJRV-0003Iq-3k
	for classpath@gnu.org; Wed, 04 Aug 2004 06:56:22 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BsJRU-0003In-Vp
	for classpath@gnu.org; Wed, 04 Aug 2004 06:56:21 -0400
Received: from [212.6.122.28] (helo=mail4.ewetel.de)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BsJNW-0001OF-LA
	for classpath@gnu.org; Wed, 04 Aug 2004 06:52:14 -0400
Received: from majestix.konqueror.de (dynadsl-080-228-94-135.ewetel.net
	[80.228.94.135])
	by mail4.ewetel.de (8.12.1/8.12.9) with ESMTP id i74ApfMv020110;
	Wed, 4 Aug 2004 12:51:42 +0200 (MEST)
Received: from miraculix.konqueror.de ([192.168.1.1])
	by majestix.konqueror.de with esmtp (Exim 3.35 #1 (Debian))
	id 1BsJVO-0002hy-00; Wed, 04 Aug 2004 13:00:22 +0200
From: Michael Koch <konqueror@gmx.de>
To: Andrew Haley <aph@redhat.com>
Date: Wed, 4 Aug 2004 12:52:14 +0200
User-Agent: KMail/1.6.2
References: <200408031329.01616.konqueror@gmx.de>
	<200408041140.52568.konqueror@gmx.de>
	<16656.48734.711075.788373@cuddles.cambridge.redhat.com>
In-Reply-To: <16656.48734.711075.788373@cuddles.cambridge.redhat.com>
MIME-Version: 1.0
Content-Disposition: inline
Content-Type: text/plain;
  charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
Message-Id: <200408041252.14749.konqueror@gmx.de>
X-CheckCompat: OK
Cc: classpath@gnu.org
Subject: Re: Removing the TARGET_* layer or not ?
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Wed, 04 Aug 2004 10:56:22 -0000

Am Mittwoch, 4. August 2004 12:45 schrieb Andrew Haley:
> Michael Koch writes:
>  > Am Mittwoch, 4. August 2004 10:27 schrieb Andrew Haley:
>  > > Well, that's the question: do we make Classpath harder to
>  > > debug / read in order to optimize the performance of some
>  > > unfree C compilers?
>  >
>  > It's not bad to support other compilers too but this should
>  > should not be used to make the code unreadable and hard to
>  > understand. If its hard to achieve better drop support for some
>  > compilers.
>
> We weren't talking about dropping any support: merely a performance
> decrease for some old unfree compilers.

As I understood Ingo this means dropping for him as the overhead is 
too much for him.


Michael



From MAILER-DAEMON Wed Aug 04 07:03:23 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BsJYJ-0004l8-PA
	for mharc-classpath@gnu.org; Wed, 04 Aug 2004 07:03:23 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BsJYI-0004k3-NG
	for classpath@gnu.org; Wed, 04 Aug 2004 07:03:22 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BsJYH-0004je-S0
	for classpath@gnu.org; Wed, 04 Aug 2004 07:03:22 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BsJYH-0004jY-My
	for classpath@gnu.org; Wed, 04 Aug 2004 07:03:21 -0400
Received: from [62.253.162.41] (helo=mta01-svc.ntlworld.com)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BsJUj-0002Aq-IV
	for classpath@gnu.org; Wed, 04 Aug 2004 06:59:43 -0400
Received: from cuddles.cambridge.redhat.com ([81.96.64.123])
	by mta01-svc.ntlworld.com
	(InterMail vM.4.01.03.37 201-229-121-137-20020806) with ESMTP id
	<20040804105905.QTMU7107.mta01-svc.ntlworld.com@cuddles.cambridge.redhat.com>;
	Wed, 4 Aug 2004 11:59:05 +0100
Received: from redhat.com (localhost.localdomain [127.0.0.1])
	by cuddles.cambridge.redhat.com (8.12.8/8.12.8) with ESMTP id
	i74Axb2v027216; Wed, 4 Aug 2004 11:59:38 +0100
Received: (from aph@localhost)
	by redhat.com (8.12.8/8.12.8/Submit) id i74AxaaI027212;
	Wed, 4 Aug 2004 11:59:36 +0100
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Message-ID: <16656.49560.233267.917140@cuddles.cambridge.redhat.com>
Date: Wed, 4 Aug 2004 11:59:36 +0100
From: Andrew Haley <aph@redhat.com>
To: Michael Koch <konqueror@gmx.de>
In-Reply-To: <200408041252.14749.konqueror@gmx.de>
References: <200408031329.01616.konqueror@gmx.de>
	<200408041140.52568.konqueror@gmx.de>
	<16656.48734.711075.788373@cuddles.cambridge.redhat.com>
	<200408041252.14749.konqueror@gmx.de>
X-Mailer: VM 7.14 under Emacs 21.3.50.1
Cc: classpath@gnu.org
Subject: Re: Removing the TARGET_* layer or not ?
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Wed, 04 Aug 2004 11:03:23 -0000

Michael Koch writes:
 > Am Mittwoch, 4. August 2004 12:45 schrieb Andrew Haley:
 > > Michael Koch writes:
 > >  > Am Mittwoch, 4. August 2004 10:27 schrieb Andrew Haley:
 > >  > > Well, that's the question: do we make Classpath harder to
 > >  > > debug / read in order to optimize the performance of some
 > >  > > unfree C compilers?
 > >  >
 > >  > It's not bad to support other compilers too but this should
 > >  > should not be used to make the code unreadable and hard to
 > >  > understand. If its hard to achieve better drop support for some
 > >  > compilers.
 > >
 > > We weren't talking about dropping any support: merely a performance
 > > decrease for some old unfree compilers.
 > 
 > As I understood Ingo this means dropping for him as the overhead is 
 > too much for him.

This is such a weird conversation!  Surely the bug is in the C
compiler that still doesn't inline.  gcc got inlining in 1988, I think
with version 1.20.

Andrew.



From MAILER-DAEMON Wed Aug 04 08:29:33 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BsKtg-0003GH-QP
	for mharc-classpath@gnu.org; Wed, 04 Aug 2004 08:29:33 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BsKte-0003FI-0s
	for classpath@gnu.org; Wed, 04 Aug 2004 08:29:30 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BsKtZ-0003DF-N2
	for classpath@gnu.org; Wed, 04 Aug 2004 08:29:29 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BsKtZ-0003D0-El
	for classpath@gnu.org; Wed, 04 Aug 2004 08:29:25 -0400
Received: from [199.232.41.8] (helo=mx20.gnu.org)
	by monty-python.gnu.org with esmtp (TLSv1:DES-CBC3-SHA:168)
	(Exim 4.34) id 1BsKpp-00089F-At
	for classpath@gnu.org; Wed, 04 Aug 2004 08:25:33 -0400
Received: from [212.227.126.185] (helo=moutng.kundenserver.de)
	by mx20.gnu.org with esmtp (Exim 4.34) id 1Bs0c3-0007GC-Ut
	for classpath@gnu.org; Tue, 03 Aug 2004 10:50:00 -0400
Received: from [212.227.126.162] (helo=mrelayng.kundenserver.de)
	by moutng.kundenserver.de with esmtp (Exim 3.35 #1)
	id 1Bs0bz-0001KG-00; Tue, 03 Aug 2004 16:49:55 +0200
Received: from [212.126.219.82] (helo=aicas.com)
	by mrelayng.kundenserver.de with asmtp (TLSv1:RC4-MD5:128)
	(Exim 3.35 #1) id 1Bs0bz-0002GF-00; Tue, 03 Aug 2004 16:49:55 +0200
Message-ID: <410FA611.7020407@aicas.com>
Date: Tue, 03 Aug 2004 16:49:53 +0200
From: =?ISO-8859-15?Q?Ingo_Pr=F6tel?= <proetel@aicas.com>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.6) Gecko/20040113
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: Michael Koch <konqueror@gmx.de>
References: <200408031329.01616.konqueror@gmx.de> <410F9574.8040408@aicas.com>
	<200408031601.34628.konqueror@gmx.de>
In-Reply-To: <200408031601.34628.konqueror@gmx.de>
X-Enigmail-Version: 0.83.6.0
X-Enigmail-Supports: pgp-inline, pgp-mime
Content-Type: multipart/mixed; boundary="------------030502010108090303000507"
X-Provags-ID: kundenserver.de abuse@kundenserver.de
	auth:ebd3e0372cc3ac236a2e2611be1a8528
Cc: classpath@gnu.org
Subject: Re: Removing the TARGET_* layer or not ?
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Wed, 04 Aug 2004 12:29:30 -0000

This is a multi-part message in MIME format.
--------------030502010108090303000507
Content-Type: text/plain; charset=ISO-8859-15; format=flowed
Content-Transfer-Encoding: 8bit

Hi Michael,

Michael Koch wrote:
> Am Dienstag, 3. August 2004 15:39 schrieb Ingo Prötel:
> 
>>Hi Michael,
>>
>>Michael Koch wrote:
>>
>>>Hi list, hi Ingo,
>>>
>>>
>>>
>>>I just started porting/testing GNU classpath to solaris (2.6).
>>>Its not really different to other unices (we currently support
>>>linux and *BSD) but some things are. When I wanted to do the
>>>needed changes I stumbled over the TARGET_* layer which adds (in
>>>my eyes) some unneeded complexity. It makes it more hard to read
>>>the code and all the stuff seem to be implemented either in
>>>target/generic or target/Linux with no real rule what have to go
>>>where. It's just difficult to understand. In fact I try to
>>>understand it since a long time and always fail because of its
>>>"indirectness". Personally I think some AUTOCONF checks would be
>>>more appropriate and would make the code much more readable and
>>>bugfixable. There are some known bugs in it but noone attacked
>>>them because none understands the code it seems.
>>>
>>>Ingo: Now my question are you really using the TARGET_* system or
>>>is it only rotting around in GNU classpath ? I really wonder if
>>>some AUTOCONF macros would be more helpful for you ?
>>
>>Yes, the TARGET-layer is in active use here.
>>
>>The TARGET-layer has two advantages:
>>1. It makes JNI-code more readable,
>>2. It makes porting simple.
>>
>>It makes JNI-code more readable because whenever a native function
>>is called there is only one macro call. The other possibility would
>>be that one has ifdefs for every system (possibly for every
>>system+architecture) that one wants to support with this code. This
>>would make the JNI-code difficult to read.
> 
> 
> The MACRO names are really long and they dont really make the code 
> more readable for me. Perhaps for the compiler ... I dont know.
> 
> 
>>It makes porting easier in that one can reuse generic macros and
>>only needs to override specific macros that differ for a given new
>>system. And this all can happen without modifing the actual
>>JNI-code. So the main step in porting is creating a new
>>subdirectory in tartget name it Solaris. Copy over all the files in
>>the Linux target. Add this to the include path. Then just work on
>>the error that the compiler will find and where necessary create
>>Solaris specific macros.
> 
> 
> I tried to port to Solaris and you system just made it hard for me 
> (and I know mjw had problems with it too when he dbugged some 
> filesystem stuff).
> 
> 
>>If there is interest we could provide our Solaris, Darwin, and
>>MinGW TARGET-layer. This would probably help to understand how this
>>layer is helpful. In any case we can help if anyone needs help
>>understanding the code.
> 
> 
> I don't get the benefit of the whole system. BSD's are already 
> suppored by the "Linux" layer. I guess Darwin (as its a BSD too) is 
> supported too and if not its easy to do so. I tried to support 
> solaris in the "Linux" layer and I needed to add around 10 lines for 
> it. So why add a lot oc code duplication to support an additional OS 
> when its enough to add 10 lines that are much more easy to 
> understand, adds no bloat and is more easy to maintain ? I haven't 
> tried mingw yet but I guess its similar easy.
> 
> Sorry, but I'm currently upset because this layer made it hard for me 
> to do something normally really simple. Autoconf was created to make 
> it easy to write portable apps. We use it already in GNU classpath. 
> Why not use it to write portable apps in it ?
> 
> 
> Michael
> 
You are quite right. Solaris is still pretty close to Linux. So the only difference is in the target_native_network.h file.
The difference to Darwin is even less. But we have been able to port to those platforms without having to modify the actual JNI-code.
If the TRAGET-layer would be removed all changes would need to be done in the JNI-file. We try to support embedded targets which use at least seven 
different operating systems, which often have no native support for 64-bit values, or have weired endianess. If we would need to add all these 
differences as as ifdefs in the JNI-code it would make this code difficult to read.
Debugging the macros is actually a hard thing. But once a macro works correctly it normally does not need to be changed any more. Much more often the 
code around the macro must be changed and that is easier if it is not cluttered with ifdefs.

I have attached the target_native_network.h files for Solaris and Darwin. I hope that helps you a bit.

Ingo


-- 
Ingo Prötel                                          proetel@aicas.com
aicas GmbH                                        http://www.aicas.com
Haid-und-Neu-Str. 18                        phone   +49 721 663 968-32
76131 Karlsruhe                             fax     +49 721 663 968-93
Germany

--------------030502010108090303000507
Content-Type: text/plain;
 name="Darwin-target_native_network.h"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline;
 filename="Darwin-target_native_network.h"

/* ???.h - ???
   Copyright (C) 1998 Free Software Foundation, Inc.

This file is part of GNU Classpath.

GNU Classpath is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 2, or (at your option)
any later version.
 
GNU Classpath is distributed in the hope that it will be useful, but
WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
General Public License for more details.

You should have received a copy of the GNU General Public License
along with GNU Classpath; see the file COPYING.  If not, write to the
Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
02111-1307 USA.

Linking this library statically or dynamically with other modules is
making a combined work based on this library.  Thus, the terms and
conditions of the GNU General Public License cover the whole
combination.

As a special exception, the copyright holders of this library give you
permission to link this library with independent modules to produce an
executable, regardless of the license terms of these independent
modules, and to copy and distribute the resulting executable under
terms of your choice, provided that you also meet, for each linked
independent module, the terms and conditions of the license of that
module.  An independent module is a module which is not derived from
or based on this library.  If you modify this library, you may extend
this exception to your version of the library, but you are not
obligated to do so.  If you do not wish to do so, delete this
exception statement from your version. */

/*
Description: Darwin/Mac OS-X target defintions of network functions
Systems    : Darwin/Mac OS-X

*/

#ifndef __TARGET_NATIVE_NETWORK__
#define __TARGET_NATIVE_NETWORK__

/****************************** Includes *******************************/
/* do not move; needed here because of some macro definitions */
#include <config.h>

#include <stdlib.h>

/****************** Conditional compilation switches *******************/

/***************************** Constants *******************************/

/***************************** Datatypes *******************************/

/***************************** Variables *******************************/

/****************************** Macros *********************************/

#define TARGET_NATIVE_NETWORK_SOCKET_SEND(socketDescriptor,buffer,length,bytesSent) \
  do { \
    bytesSent=send(socketDescriptor,buffer,length,0); \
  } while (0)

#define TARGET_NATIVE_NETWORK_SOCKET_SEND_WITH_ADDRESS_PORT(socketDescriptor,buffer,length,address,port,bytesSent) \
  do { \
    struct sockaddr_in __socketAddress; \
    \
    memset(&__socketAddress,0,sizeof(__socketAddress)); \
    __socketAddress.sin_family      = AF_INET; \
    __socketAddress.sin_addr.s_addr = htonl(address); \
    __socketAddress.sin_port        = htons((short)port); \
    bytesSent=sendto(socketDescriptor,buffer,length,0,(struct sockaddr*)&__socketAddress,sizeof(__socketAddress)); \
  } while (0)

/***************************** Functions *******************************/

#ifdef __cplusplus
extern "C" {
#endif

#ifdef __cplusplus
}
#endif

/* include rest of definitions from generic file (do not move it to 
   another position!) */
#include "target_generic_network.h"

#endif /* __TARGET_NATIVE_NETWORK__ */

/* end of file */


--------------030502010108090303000507
Content-Type: text/plain;
 name="SunOS-target_native_network.h"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline;
 filename="SunOS-target_native_network.h"

/* ???.h - ???
   Copyright (C) 1998 Free Software Foundation, Inc.

This file is part of GNU Classpath.

GNU Classpath is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 2, or (at your option)
any later version.
 
GNU Classpath is distributed in the hope that it will be useful, but
WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
General Public License for more details.

You should have received a copy of the GNU General Public License
along with GNU Classpath; see the file COPYING.  If not, write to the
Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
02111-1307 USA.

Linking this library statically or dynamically with other modules is
making a combined work based on this library.  Thus, the terms and
conditions of the GNU General Public License cover the whole
combination.

As a special exception, the copyright holders of this library give you
permission to link this library with independent modules to produce an
executable, regardless of the license terms of these independent
modules, and to copy and distribute the resulting executable under
terms of your choice, provided that you also meet, for each linked
independent module, the terms and conditions of the license of that
module.  An independent module is a module which is not derived from
or based on this library.  If you modify this library, you may extend
this exception to your version of the library, but you are not
obligated to do so.  If you do not wish to do so, delete this
exception statement from your version. */

/*
Description: SunOS/Solaris target defintions of network functions
Systems    : all
*/

#ifndef __TARGET_NATIVE_NETWORK__
#define __TARGET_NATIVE_NETWORK__

/****************************** Includes *******************************/
/* do not move; needed here because of some macro definitions */
#include <config.h>

#include <stdlib.h>
#include <stropts.h>
#include <sys/conf.h>
#include <netinet/in_systm.h>
#include <sys/ioctl.h>

/****************** Conditional compilation switches *******************/

/***************************** Constants *******************************/

/***************************** Datatypes *******************************/

/***************************** Variables *******************************/

/****************************** Macros *********************************/

/***********************************************************************\
* Name       : TARGET_NATIVE_NETWORK_GET_HOSTNAME_BY_NAME
* Purpose    : get hostname by name
* Input      : name           - hostname
*              maxAddressSize - max. size of address array
* Output     : addresses    - host addresses (array, in host byte
*                             order!)
*              addressCount - number of entries in address array
*              result       - TARGET_NATIVE_OK if no error occurred,
*                             TARGET_NATIVE_ERROR otherwise
* Return     : -
* Side-effect: unknown
* Notes      : -
\***********************************************************************/

#ifndef TARGET_NATIVE_NETWORK_GET_HOSTNAME_BY_NAME
  #include <netdb.h>
  #define TARGET_NATIVE_NETWORK_GET_HOSTNAME_BY_NAME(name,addresses,maxAddressSize,addressCount,result) \
    do { \
      struct hostent __hostEntry; \
      char           __buffer[5000]; \
      int            __h_errno; \
      \
      addressCount = 0; \
      \
      if (gethostbyname_r(name, &__hostEntry, __buffer, sizeof(__buffer), &__h_errno) != NULL) \
        { \
          while ((addressCount < maxAddressSize) && (__hostEntry.h_addr_list[addressCount] != NULL)) \
            { \
              addresses[addressCount] = ntohl(*(int*)(__hostEntry.h_addr_list[addressCount])); \
              addressCount++; \
            } \
          result = TARGET_NATIVE_OK; \
        } \
      else \
        { \
          result = TARGET_NATIVE_ERROR; \
        } \
    } while (0)
#endif

/***********************************************************************\
* Name       : TARGET_NATIVE_NETWORK_SOCKET_RECEIVE_AVAILABLE
* Purpose    : get number of available bytes for receive
* Input      : socketDescriptor - socket descriptor
* Output     : bytesAvailable - available bytes for receive
*            : result         - TARGET_NATIVE_OK if no error occurred,
*                               TARGET_NATIVE_ERROR otherwise
* Return     : -
* Side-effect: unknown
* Notes      : -
\***********************************************************************/

#define TARGET_NATIVE_NETWORK_SOCKET_RECEIVE_AVAILABLE(socketDescriptor,bytesAvailable,result) \
  do { \
    int __value; \
    \
    bytesAvailable=0; \
    \
    result = (ioctl(socketDescriptor, I_NREAD, &__value) == 0)?TARGET_NATIVE_OK:TARGET_NATIVE_ERROR; \
    if (result == TARGET_NATIVE_OK) \
      { \
        bytesAvailable = __value; \
      } \
  } while (0)

#if 1
#include <sys/types.h>
#include <sys/socket.h>
#include <arpa/inet.h>
#include <netinet/in.h>
#include <netdb.h>
#define TARGET_NATIVE_NETWORK_SOCKET_CONNECT(socketDescriptor,address,port,result) \
  do { \
    struct sockaddr_in __socketAddress; \
    int nnn; \
    \
fprintf(stderr, "address %08x %d\n", address, (int)port); \
    memset(&__socketAddress,0,sizeof(__socketAddress)); \
    __socketAddress.sin_family      = AF_INET; \
    __socketAddress.sin_addr.s_addr = htonl(address); \
    __socketAddress.sin_port        = htons(((short)port)); \
    \
    nnn=connect(socketDescriptor,(struct sockaddr*)&__socketAddress,sizeof(__socketAddress)); \
    result=(nnn==0)?TARGET_NATIVE_OK:TARGET_NATIVE_ERROR; \
fprintf(stderr,"%s,%d: connect: %d %d %d\n",__FILE__,__LINE__,socketDescriptor,nnn,errno); \
  } while (0)
#endif /* 0 */

/***********************************************************************\
* Name       : TARGET_NATIVE_NETWORK_SOCKET_SEND
* Purpose    : send data to socket
* Input      : socketDescriptor - socket descriptor
*            : buffer  - data to send
*              length  - length of data to send
* Output     : bytesSent - number of bytes sent, -1 otherwise
* Side-effect: unknown
* Notes      : -
\***********************************************************************/

#ifndef TARGET_NATIVE_NETWORK_SOCKET_SEND
  #include <sys/types.h>
  #include <sys/socket.h>
  #include <netinet/in.h>
  #define TARGET_NATIVE_NETWORK_SOCKET_SEND(socketDescriptor, buffer, length, bytesSent) \
    do { \
      bytesSent = send(socketDescriptor, \
                       buffer, \
                       length, \
                       0 /* MSG_NOSIGNAL not supported by Solaris */ \
                      ); \
    } while (0)
#endif

/***********************************************************************\
* Name       : TARGET_NATIVE_NETWORK_SOCKET_SEND_WITH_ADDRESS_PORT
* Purpose    : send data to socket
* Input      : socketDescriptor - socket descriptor
*            : buffer  - data to send
*              length  - length of data to send
*              Address - to address (NOT in network byte order!)
*              Port    - to port (NOT in network byte order!)
* Output     : bytesSent - number of bytes sent, -1 otherwise
* Side-effect: unknown
* Notes      : -
\***********************************************************************/

#ifndef TARGET_NATIVE_NETWORK_SOCKET_SEND_WITH_ADDRESS_PORT
  #include <sys/types.h>
  #include <sys/socket.h>
  #include <netinet/in.h>
  #define TARGET_NATIVE_NETWORK_SOCKET_SEND_WITH_ADDRESS_PORT(socketDescriptor, buffer, length, address, port, bytesSent) \
    do { \
      struct sockaddr_in __socketAddress; \
      \
      memset(&__socketAddress,0,sizeof(__socketAddress)); \
      __socketAddress.sin_family      = AF_INET; \
      __socketAddress.sin_addr.s_addr = htonl(address); \
      __socketAddress.sin_port        = htons((short)port); \
      bytesSent = sendto(socketDescriptor, \
                         buffer, \
                         length, \
			 0 /* MSG_NOSIGNAL not supported by solaris */, \
			 (struct sockaddr*)&__socketAddress, \
			 sizeof(__socketAddress) \
                        ); \
    } while (0)
#endif

/***************************** Functions *******************************/

#ifdef __cplusplus
extern "C" {
#endif

#ifdef __cplusplus
}
#endif

/* include rest of definitions from generic file (do not move it to 
   another position!) */
#include "target_generic_network.h"

#endif /* __TARGET_NATIVE_NETWORK__ */

/* end of file */

--------------030502010108090303000507--



From MAILER-DAEMON Wed Aug 04 12:19:43 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BsOUR-0007iE-8K
	for mharc-classpath@gnu.org; Wed, 04 Aug 2004 12:19:43 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BsOUP-0007hT-TK
	for classpath@gnu.org; Wed, 04 Aug 2004 12:19:41 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BsOUM-0007gM-Af
	for classpath@gnu.org; Wed, 04 Aug 2004 12:19:41 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BsOUM-0007g5-6Y
	for classpath@gnu.org; Wed, 04 Aug 2004 12:19:38 -0400
Received: from [212.227.126.171] (helo=moutng.kundenserver.de)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BsOQR-0007Dg-Js
	for classpath@gnu.org; Wed, 04 Aug 2004 12:15:35 -0400
Received: from [212.227.126.209] (helo=mrelayng.kundenserver.de)
	by moutng.kundenserver.de with esmtp (Exim 3.35 #1)
	id 1BsOQQ-0002R9-00; Wed, 04 Aug 2004 18:15:34 +0200
Received: from [212.126.219.82] (helo=aicas.com)
	by mrelayng.kundenserver.de with asmtp (TLSv1:RC4-MD5:128)
	(Exim 3.35 #1) id 1BsOQQ-0004o0-00; Wed, 04 Aug 2004 18:15:34 +0200
Message-ID: <41110BA5.6030702@aicas.com>
Date: Wed, 04 Aug 2004 18:15:33 +0200
From: =?ISO-8859-15?Q?Ingo_Pr=F6tel?= <proetel@aicas.com>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.6) Gecko/20040113
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: Michael Koch <konqueror@gmx.de>
References: <200408031329.01616.konqueror@gmx.de>
	<1091608847.726.10.camel@moonlight> <411098F1.50509@aicas.com>
	<200408041150.54758.konqueror@gmx.de>
In-Reply-To: <200408041150.54758.konqueror@gmx.de>
X-Enigmail-Version: 0.83.6.0
X-Enigmail-Supports: pgp-inline, pgp-mime
Content-Type: text/plain; charset=iso-8859-15; format=flowed
Content-Transfer-Encoding: 8bit
X-Provags-ID: kundenserver.de abuse@kundenserver.de
	auth:ebd3e0372cc3ac236a2e2611be1a8528
Cc: classpath@gnu.org
Subject: Re: Removing the TARGET_* layer or not ?
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Wed, 04 Aug 2004 16:19:42 -0000



Michael Koch wrote:
> Am Mittwoch, 4. August 2004 10:06 schrieb Ingo Prötel:
> 
>>Hi,
>>
>>Roman Kennke wrote:
>>
>>>Am Di, den 03.08.2004 schrieb Ingo Prötel um 15:39:
>>>
>>>>Hi Michael,
>>>>
>>>>Michael Koch wrote:
>>>>
>>>>>Hi list, hi Ingo,
>>>>>
>>>>>
>>>>>
>>>>>I just started porting/testing GNU classpath to solaris (2.6).
>>>>>Its not really different to other unices (we currently support
>>>>>linux and *BSD) but some things are. When I wanted to do the
>>>>>needed changes I stumbled over the TARGET_* layer which adds
>>>>>(in my eyes) some unneeded complexity. It makes it more hard to
>>>>>read the code and all the stuff seem to be implemented either
>>>>>in target/generic or target/Linux with no real rule what have
>>>>>to go where. It's just difficult to understand. In fact I try
>>>>>to understand it since a long time and always fail because of
>>>>>its "indirectness". Personally I think some AUTOCONF checks
>>>>>would be more appropriate and would make the code much more
>>>>>readable and bugfixable. There are some known bugs in it but
>>>>>noone attacked them because none understands the code it seems.
>>>>>
>>>>>Ingo: Now my question are you really using the TARGET_* system
>>>>>or is it only rotting around in GNU classpath ? I really wonder
>>>>>if some AUTOCONF macros would be more helpful for you ?
>>>>
>>>>Yes, the TARGET-layer is in active use here.
>>>>
>>>>The TARGET-layer has two advantages:
>>>>1. It makes JNI-code more readable,
>>>>2. It makes porting simple.
>>>>
>>>>It makes JNI-code more readable because whenever a native
>>>>function is called there is only one macro call. The other
>>>>possibility would be that one has ifdefs for every system
>>>>(possibly for every system+architecture) that one wants to
>>>>support with this code. This would make the JNI-code difficult
>>>>to read.
>>>>
>>>>It makes porting easier in that one can reuse generic macros and
>>>>only needs to override specific macros that differ for a given
>>>>new system. And this all can happen without modifing the actual
>>>>JNI-code. So the main step in porting is creating a new
>>>>subdirectory in tartget name it Solaris. Copy over all the files
>>>>in the Linux target. Add this to the include path. Then just
>>>>work on the error that the compiler will find and where
>>>>necessary create Solaris specific macros.
>>>>
>>>>If there is interest we could provide our Solaris, Darwin, and
>>>>MinGW TARGET-layer. This would probably help to understand how
>>>>this layer is helpful. In any case we can help if anyone needs
>>>>help understanding the code.
>>>
>>>I have thought about how I would design this stuff in order to
>>>write portable code:
>>>1. first of all, try to stick to POSIX standard stuff ;)
>>>2. Of course there are situations where this won't do. In this
>>>case I would split out the non-portable stuff on a case-by-case
>>>basis in the following scheme: Suppose you have the following
>>>expressions in a function (pseudocode):
>>>
>>>func() {
>>>  exp_a;
>>>  exp_b;
>>>  exp_c;
>>>}
>>>
>>>of which exp_b is not portable. Then make a function call out of
>>>it:
>>>
>>>func() {
>>>  exp_a;
>>>  do_b();
>>
>>     TARGET_NATIVE_B();  // this is what it currently looks like
>>
>>
>>>  exp_c;
>>>}
>>>
>>>and define the function in a file in a file in an OS-specific
>>>subdir.
>>
>>This is actually what the whole idea is about. Except that macros
>>are used instead of functions. The problem with functions is that
>>many embedded compilers cannot inline function calls. This would be
>>ok if the overhead would really be only one function call but if
>>you got through the JNI-code and replace every TARGET-macro with a
>>function call you will see that there are more function calls
>>involved. Maybe we could get some mixed mode where we leave the
>>macros in place and the macros for Linux just point to functions.
>>This would make the Linux code more debuggable.
>>
>>
>>>the autotools should check for OSs/arches and include the correct
>>>subdirs here. These functions must be kept minimal, in order to
>>>avoid the effect that Michael mentioned, where you have to fix a
>>>bug multiple times.
>>
>>If one fixes a bug in the generic part it will be automatically
>>picked up in every system that uses the generic stuff. If a system
>>does not use the generic stuff most likely there will be no system
>>that will depend on this bug fix. Especially if one fixes a bug
>>somewhere this might as a side effect break another system that
>>depends on the bogous behavior. There is no way of testing it if
>>one doesn't have a target setup to test it on. So sometimes this
>>sort of isolation is good.
>>
>>
>>>Note, that with this system we would avoid the use of macros
>>>altogether (You see, I don't like them ;) I better go with a
>>>well-thought-out design). Sure, there is the 'overhead' of 1
>>>additional function call.
>>
>>That is basically on function call for every TARGET-macro.
> 
> 
> I'm not a friend of this one-OS-one-dir approach. I would more test 
> for features with autoconf and use the features if present. This 
> makes it more easy to port to another arch. When the feature is 
> available there, use it. If not, add some code to make it work.
> 
The advantage of having one-OS-one-dir is that one knows which files are involved. Otherwise one needs the latest autoconf run to determine all the 
variables to figure out what parts of the code actually runs. Or do I have it wrong?

> The TARGET api is really not intutive and the very long MACRO names 
> makes it even harder to understand. Sometimes the names just badly 
> choosed. The "Linux" layer is full of bugs. And I suppose your other 
> ports are too because they are copies of the "Linux" layer. Some bugs 
> are fixed in the classpath version of "Linux". I wonder how many of 
> them were ported to you other ports. When an autoconf approach you 
> wouldnt even think about this. I know that with the autoconf approach 
> the patches need a closed review and more testing before they go in 
> but IMO this is much better then duplicatin bugs over several trees.
> 
I actually didn't consider long names a problem since there is word completion every where. If the names are badly chosen the can be changed. They 
actually follow a convention:(Quote form the readme.txt)
   The naming pattern for native macros is like follows:

     TARGET_NATIVE_<module name>_<macro name>

   where <module name> is a name of a module, e. g. FILE; <macro name> is
   the name of the specific native macro, e. g. OPEN_READ.

I guess the TARGET_NATIVE_ part could be shortend.

> I was pretty surprised when ypu wrote yesterday, that you have more 
> "ports" then just "Linux". Why don't you submitted them ? I think it 
> would be a very good idea to support as much OSes upstream as 
> possible, even Plan9 and QNX and ...
>
I will see what I can put in. Though I don't think we have Plan9 (At least I don't know it). But we have an RTEMS port.

> 
> Michael
> 


Ingo
-- 
Ingo Prötel                                          proetel@aicas.com
aicas GmbH                                        http://www.aicas.com
Haid-und-Neu-Str. 18                        phone   +49 721 663 968-32
76131 Karlsruhe                             fax     +49 721 663 968-93
Germany



From MAILER-DAEMON Wed Aug 04 12:32:40 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BsOgx-0003It-Sq
	for mharc-classpath@gnu.org; Wed, 04 Aug 2004 12:32:39 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BsOgw-0003Ih-6T
	for classpath@gnu.org; Wed, 04 Aug 2004 12:32:38 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BsOgv-0003IR-Ij
	for classpath@gnu.org; Wed, 04 Aug 2004 12:32:37 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BsOgv-0003IN-ES
	for classpath@gnu.org; Wed, 04 Aug 2004 12:32:37 -0400
Received: from [212.6.122.28] (helo=mail4.ewetel.de)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BsOdL-0000Wu-3X
	for classpath@gnu.org; Wed, 04 Aug 2004 12:28:55 -0400
Received: from majestix.konqueror.de (dynadsl-080-228-94-135.ewetel.net
	[80.228.94.135])
	by mail4.ewetel.de (8.12.1/8.12.9) with ESMTP id i74GSMMv013049;
	Wed, 4 Aug 2004 18:28:23 +0200 (MEST)
Received: from miraculix.konqueror.de ([192.168.1.1])
	by majestix.konqueror.de with esmtp (Exim 3.35 #1 (Debian))
	id 1BsOlE-0003QT-00; Wed, 04 Aug 2004 18:37:04 +0200
From: Michael Koch <konqueror@gmx.de>
To: Ingo =?iso-8859-15?q?Pr=F6tel?= <proetel@aicas.com>
Date: Wed, 4 Aug 2004 18:28:55 +0200
User-Agent: KMail/1.6.2
References: <200408031329.01616.konqueror@gmx.de>
	<200408041150.54758.konqueror@gmx.de> <41110BA5.6030702@aicas.com>
In-Reply-To: <41110BA5.6030702@aicas.com>
MIME-Version: 1.0
Content-Disposition: inline
Content-Type: text/plain;
  charset="iso-8859-15"
Content-Transfer-Encoding: quoted-printable
Message-Id: <200408041828.55700.konqueror@gmx.de>
X-CheckCompat: OK
Cc: classpath@gnu.org
Subject: Re: Removing the TARGET_* layer or not ?
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Wed, 04 Aug 2004 16:32:38 -0000

Am Mittwoch, 4. August 2004 18:15 schrieb Ingo Pr=F6tel:

> > I'm not a friend of this one-OS-one-dir approach. I would more
> > test for features with autoconf and use the features if present.
> > This makes it more easy to port to another arch. When the feature
> > is available there, use it. If not, add some code to make it
> > work.
>
> The advantage of having one-OS-one-dir is that one knows which
> files are involved. Otherwise one needs the latest autoconf run to
> determine all the variables to figure out what parts of the code
> actually runs. Or do I have it wrong?

The advantages of one-dir-fits-all was described in other mails from=20
me already.

No, you are don't wrong. But the #ifdefs are normally really clear=20
from the code to understand. More see below.

> > The TARGET api is really not intutive and the very long MACRO
> > names makes it even harder to understand. Sometimes the names
> > just badly choosed. The "Linux" layer is full of bugs. And I
> > suppose your other ports are too because they are copies of the
> > "Linux" layer. Some bugs are fixed in the classpath version of
> > "Linux". I wonder how many of them were ported to you other
> > ports. When an autoconf approach you wouldnt even think about
> > this. I know that with the autoconf approach the patches need a
> > closed review and more testing before they go in but IMO this is
> > much better then duplicatin bugs over several trees.
>
> I actually didn't consider long names a problem since there is word
> completion every where. If the names are badly chosen the can be
> changed. They actually follow a convention:(Quote form the
> readme.txt)
>    The naming pattern for native macros is like follows:
>
>      TARGET_NATIVE_<module name>_<macro name>
>
>    where <module name> is a name of a module, e. g. FILE; <macro
> name> is the name of the specific native macro, e. g. OPEN_READ.

Whoever invented this convention must have been drunk or so. I dont=20
like like the ultra-long names because they are hard to type (that's=20
no reason as you said yourself). I don't like them because they make=20
the code more unreasable. When I open javanet.c in an editor I get=20
blind because I only see "TARGET_NATIVE" all over the place.

> I guess the TARGET_NATIVE_ part could be shortend.

Or better removed. This prefix is really not needed.

> > I was pretty surprised when ypu wrote yesterday, that you have
> > more "ports" then just "Linux". Why don't you submitted them ? I
> > think it would be a very good idea to support as much OSes
> > upstream as possible, even Plan9 and QNX and ...
>
> I will see what I can put in. Though I don't think we have Plan9
> (At least I don't know it). But we have an RTEMS port.

That would be great.


Thanks,

Michael



From MAILER-DAEMON Wed Aug 04 13:00:09 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BsP7Z-0002Dp-7p
	for mharc-classpath@gnu.org; Wed, 04 Aug 2004 13:00:09 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BsP7W-0002BZ-J5
	for classpath@gnu.org; Wed, 04 Aug 2004 13:00:06 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BsP7T-0002Au-Sz
	for classpath@gnu.org; Wed, 04 Aug 2004 13:00:05 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BsP7T-0002Ar-NS
	for classpath@gnu.org; Wed, 04 Aug 2004 13:00:03 -0400
Received: from [207.174.178.92] (helo=email2.peakpeak.com)
	by monty-python.gnu.org with esmtp (TLSv1:DES-CBC3-SHA:168)
	(Exim 4.34) id 1BsP3j-0004Jy-7d
	for classpath@gnu.org; Wed, 04 Aug 2004 12:56:11 -0400
Received: (qmail 15541 invoked by uid 513); 4 Aug 2004 17:02:23 -0000
Received: from tromey@redhat.com by email2 by uid 504 with qmail-scanner-1.20 
	(clamscan: 0.65.  Clear:RC:1(204.144.239.10):. 
	Processed in 0.027182 secs); 04 Aug 2004 17:02:23 -0000
Received: from tf0010.peakpeak.com (HELO fleche.redhat.com) ([204.144.239.10])
	(envelope-sender <tromey@redhat.com>)
	by email2.peakpeak.com (qmail-ldap-1.03) with SMTP
	for <proetel@aicas.com>; 4 Aug 2004 17:02:23 -0000
Received: by fleche.redhat.com (Postfix, from userid 1000)
	id 9F5DC4F805A; Wed,  4 Aug 2004 10:40:07 -0600 (MDT)
To: Roman Kennke <roman@ontographics.com>
References: <200408031329.01616.konqueror@gmx.de> <410F9574.8040408@aicas.com>
	<1091608847.726.10.camel@moonlight>
From: Tom Tromey <tromey@redhat.com>
X-Attribution: Tom
Date: 04 Aug 2004 10:40:07 -0600
In-Reply-To: <1091608847.726.10.camel@moonlight>
Message-ID: <87d626amc8.fsf@fleche.redhat.com>
Lines: 20
User-Agent: Gnus/5.09 (Gnus v5.9.0) Emacs/21.3.50
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Cc: classpath@gnu.org, Michael Koch <konqueror@gmx.de>
Subject: Re: Removing the TARGET_* layer or not ?
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
Reply-To: tromey@redhat.com
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Wed, 04 Aug 2004 17:00:07 -0000

>>>>> "Roman" == Roman Kennke <roman@ontographics.com> writes:

Roman> the autotools should check for OSs/arches and include the correct
Roman> subdirs here. These functions must be kept minimal, in order to avoid
Roman> the effect that Michael mentioned, where you have to fix a bug multiple
Roman> times.

I just wanted to interject a note about auto* here -- if you do use
it, it is generally better not to look at OS/arch and instead just
write feature tests.  This gets the most benefit and there is a long
history showing this has less maintenance burden and better
portability.

I don't have an opinion on whether to keep TARGET_*, as libgcj has its
own native layer and is not likely to switch.  Just FYI, libgcj's
approach is a hybrid -- for Unixy systems we use autoconf feature
tests, but we treat Windows as a separate "flavor" with its own
implementation of native things.

Tom



From MAILER-DAEMON Wed Aug 04 13:05:25 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BsPCf-0003uM-PH
	for mharc-classpath@gnu.org; Wed, 04 Aug 2004 13:05:25 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BsPCd-0003rf-N5
	for classpath@gnu.org; Wed, 04 Aug 2004 13:05:23 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BsPCb-0003pw-Qn
	for classpath@gnu.org; Wed, 04 Aug 2004 13:05:23 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BsPCb-0003pg-Mr
	for classpath@gnu.org; Wed, 04 Aug 2004 13:05:21 -0400
Received: from [62.253.162.43] (helo=mta03-svc.ntlworld.com)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BsP8t-00051R-RC
	for classpath@gnu.org; Wed, 04 Aug 2004 13:01:32 -0400
Received: from cuddles.cambridge.redhat.com ([81.96.64.123])
	by mta03-svc.ntlworld.com
	(InterMail vM.4.01.03.37 201-229-121-137-20020806) with ESMTP id
	<20040804170009.URLD291.mta03-svc.ntlworld.com@cuddles.cambridge.redhat.com>;
	Wed, 4 Aug 2004 18:00:09 +0100
Received: from redhat.com (localhost.localdomain [127.0.0.1])
	by cuddles.cambridge.redhat.com (8.12.8/8.12.8) with ESMTP id
	i74H1S2v031992; Wed, 4 Aug 2004 18:01:28 +0100
Received: (from aph@localhost)
	by redhat.com (8.12.8/8.12.8/Submit) id i74H1RSa031983;
	Wed, 4 Aug 2004 18:01:27 +0100
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Message-ID: <16657.5735.622933.194807@cuddles.cambridge.redhat.com>
Date: Wed, 4 Aug 2004 18:01:27 +0100
From: Andrew Haley <aph@redhat.com>
To: java-patches@gcc.gnu.org, classpath@gnu.org
X-Mailer: VM 7.14 under Emacs 21.3.50.1
Cc: 
Subject: Patch: BasicPermission constructor
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Wed, 04 Aug 2004 17:05:23 -0000

2004-08-04  Andrew Haley  <aph@redhat.com>

	* java/security/BasicPermission.java: Don't check wildcards.

Index: BasicPermission.java
===================================================================
RCS file: /cvs/gcc/gcc/libjava/java/security/BasicPermission.java,v
retrieving revision 1.6
diff -u -r1.6 BasicPermission.java
--- BasicPermission.java	14 Jun 2003 05:45:12 -0000	1.6
+++ BasicPermission.java	4 Aug 2004 16:58:01 -0000
@@ -80,9 +80,8 @@
   private static final long serialVersionUID = 6279438298436773498L;
 
   /**
-   * Create a new instance with the specified permission name. If the name
-   * is empty, or contains an illegal wildcard character, an exception is
-   * thrown.
+   * Create a new instance with the specified permission name. If the
+   * name is empty an exception is thrown.
    *
    * @param name the name of this permission
    * @throws NullPointerException if name is null
@@ -91,12 +90,11 @@
   public BasicPermission(String name)
   {
     super(name);
-    if (name.indexOf("*") != -1)
-      {
-        if ((! name.endsWith(".*") && ! name.equals("*"))
-            || name.indexOf("*") != name.lastIndexOf("*"))
-          throw new IllegalArgumentException("Bad wildcard: " + name);
-      }
+
+    // This routine used to check for illegal wildcards, but no such
+    // requirement exists in the specification and Sun's runtime
+    // doesn't appear to do it.
+
     if ("".equals(name))
       throw new IllegalArgumentException("Empty name");
   }



From MAILER-DAEMON Wed Aug 04 14:23:16 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BsQQ0-0004nv-MD
	for mharc-classpath@gnu.org; Wed, 04 Aug 2004 14:23:16 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BsQPy-0004lq-Lw
	for classpath@gnu.org; Wed, 04 Aug 2004 14:23:14 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BsQPw-0004ki-HE
	for classpath@gnu.org; Wed, 04 Aug 2004 14:23:14 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BsQPv-0004jw-WF
	for classpath@gnu.org; Wed, 04 Aug 2004 14:23:12 -0400
Received: from [209.100.227.1] (helo=e4500.ripco.com)
	by monty-python.gnu.org with esmtp (TLSv1:DES-CBC3-SHA:168)
	(Exim 4.34) id 1BsQLZ-0006qA-JK
	for classpath@gnu.org; Wed, 04 Aug 2004 14:18:41 -0400
Received: from shell2.ripco.com (root@shell2.ripco.com [209.100.225.144])
	by e4500.ripco.com (8.12.10/8.12.9) with ESMTP id i74IIabN006397
	for <classpath@gnu.org>; Wed, 4 Aug 2004 13:18:36 -0500 (CDT)
Received: from shell2.ripco.com (arenn@localhost [127.0.0.1])
	by shell2.ripco.com (8.12.9/8.12.9) with ESMTP id i74IIZhJ019021
	for <classpath@gnu.org>; Wed, 4 Aug 2004 13:18:35 -0500 (CDT)
Received: (from arenn@localhost)
	by shell2.ripco.com (8.12.9/8.12.9/Submit) id i74IIZow019020
	for classpath@gnu.org; Wed, 4 Aug 2004 13:18:35 -0500 (CDT)
Date: Wed, 4 Aug 2004 13:18:35 -0500
From: "Aaron M. Renn" <arenn@urbanophile.com>
To: classpath@gnu.org
Message-ID: <20040804131835.A18964@urbanophile.com>
Mail-Followup-To: "Aaron M. Renn" <arenn@urbanophile.com>,
	classpath@gnu.org
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5i
Organization: GNU's Not Unix!
Subject: OT: Amsterdam
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Wed, 04 Aug 2004 18:23:15 -0000

Hello, all.  I'm traveling to Amsterdam on business for the next few
weeks and wondered if any Classpath hackers were in the vicinity.  I
really have not gotten a chance to meet most of the folks here, and
would love to do so if possible, particularly in a place where I'm
not likely to travel frequently.  Happy hacking, everyone!

-- 
Aaron M. Renn (arenn@urbanophile.com) http://www.urbanophile.com/arenn/




From MAILER-DAEMON Wed Aug 04 19:14:46 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BsUy5-0000qG-Tb
	for mharc-classpath@gnu.org; Wed, 04 Aug 2004 19:14:46 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BsUy3-0000pw-63
	for classpath@gnu.org; Wed, 04 Aug 2004 19:14:43 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BsUxz-0000pN-EL
	for classpath@gnu.org; Wed, 04 Aug 2004 19:14:42 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BsUxx-0000pC-W6
	for classpath@gnu.org; Wed, 04 Aug 2004 19:14:38 -0400
Received: from [62.253.162.43] (helo=mta03-svc.ntlworld.com)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BsUuP-00005U-1d
	for classpath@gnu.org; Wed, 04 Aug 2004 19:10:57 -0400
Received: from [192.168.1.100] ([82.6.126.159]) by mta03-svc.ntlworld.com
	(InterMail vM.4.01.03.37 201-229-121-137-20020806) with ESMTP
	id <20040804230932.SEQH291.mta03-svc.ntlworld.com@[192.168.1.100]>;
	Thu, 5 Aug 2004 00:09:32 +0100
From: David Gilbert <david.gilbert@object-refinery.com>
To: Sven de Marothy <sven@physto.se>
In-Reply-To: <1091560480.7431.51.camel@qcplx01.physto.se>
References: <1091560480.7431.51.camel@qcplx01.physto.se>
Content-Type: text/plain
Message-Id: <1091661142.6920.14.camel@linux42.overstone>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.4.6 
Date: Thu, 05 Aug 2004 00:12:24 +0100
Content-Transfer-Encoding: 7bit
Cc: classpath@gnu.org
Subject: Re: Line2D intersection
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Wed, 04 Aug 2004 23:14:43 -0000

On Tue, 2004-08-03 at 20:14, Sven de Marothy wrote:
> Just one suggestion: I'm (hopefully) going to be done with Area before
> the end of the month, and the remaining Arc2D stuff (and whatever else
> is left in geom.*) should be pretty trivial to do. So instead of making
> a bunch of small commits, I'd like to roll this all into one big geom.*
> patch, since any updates to this will need to be sent over to the
> gcj-gui-branch guys, and it'd be less work for them.
> 
> Does that sound OK?

Sounds reasonable to me.  I will write some tests for the other classes
in geom.* to run your changes against.

> Oh, and keep up the good work on writing tests! It's very useful!
> 
> (If only someone with Mauve CVS would commit them for you..
> doesn't seem to be much action on that mailing-list)

Maybe I'll request CVS write access (for Mauve), although I'd like to
avoid that until I know my way around a little better.

Regards,

Dave.




From MAILER-DAEMON Thu Aug 05 00:51:23 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BsaDq-000324-Uk
	for mharc-classpath@gnu.org; Thu, 05 Aug 2004 00:51:23 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BsaDp-00031z-Nb
	for classpath@gnu.org; Thu, 05 Aug 2004 00:51:21 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BsaDo-00031n-MY
	for classpath@gnu.org; Thu, 05 Aug 2004 00:51:21 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BsaDo-00031k-KX
	for classpath@gnu.org; Thu, 05 Aug 2004 00:51:20 -0400
Received: from [24.25.9.100] (helo=ms-smtp-01-eri0.southeast.rr.com)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BsaA3-0004wr-9L
	for classpath@gnu.org; Thu, 05 Aug 2004 00:47:27 -0400
Received: from [192.168.0.3] (rdu74-160-248.nc.rr.com [24.74.160.248])
	by ms-smtp-01-eri0.southeast.rr.com (8.12.10/8.12.7) with ESMTP id
	i754lNPf022382; Thu, 5 Aug 2004 00:47:23 -0400 (EDT)
From: "C. Brian Jones" <cbj@gnu.org>
To: "Aaron M. Renn" <arenn@urbanophile.com>
In-Reply-To: <20040804131835.A18964@urbanophile.com>
References: <20040804131835.A18964@urbanophile.com>
Content-Type: text/plain
Message-Id: <1091681242.4640.132.camel@lyta.haphazard.org>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.4.5 (1.4.5-7) 
Date: Thu, 05 Aug 2004 00:47:23 -0400
Content-Transfer-Encoding: 7bit
X-Virus-Scanned: Symantec AntiVirus Scan Engine
Cc: classpath@gnu.org
Subject: Re: OT: Amsterdam
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Thu, 05 Aug 2004 04:51:21 -0000

On Wed, 2004-08-04 at 14:18, Aaron M. Renn wrote:
> Hello, all.  I'm traveling to Amsterdam on business for the next few
> weeks and wondered if any Classpath hackers were in the vicinity.  I
> really have not gotten a chance to meet most of the folks here, and
> would love to do so if possible, particularly in a place where I'm
> not likely to travel frequently.  Happy hacking, everyone!

Aaron, you're alive!  ;)  I'm traveling to France the last two weeks of
Sept. for personal vacation, but I don't know of anyone using Classpath
there.  I will be in Strasbourg for a day or two, Paris, Avignon,
Mont-Saint-Michel, is that anywhere near anyone?  :)

-- 
Brian Jones <cbj@gnu.org>




From MAILER-DAEMON Thu Aug 05 06:01:22 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1Bsf3q-0007gU-Dq
	for mharc-classpath@gnu.org; Thu, 05 Aug 2004 06:01:22 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1Bsf3m-0007gA-Tv
	for classpath@gnu.org; Thu, 05 Aug 2004 06:01:19 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1Bsf3k-0007fu-VX
	for classpath@gnu.org; Thu, 05 Aug 2004 06:01:18 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1Bsf3k-0007fq-8U
	for classpath@gnu.org; Thu, 05 Aug 2004 06:01:16 -0400
Received: from [139.19.1.2] (helo=francois.mpi-sb.mpg.de)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BsezQ-0006ba-Do
	for classpath@gnu.org; Thu, 05 Aug 2004 05:56:48 -0400
Received: from amavis by francois.mpi-sb.mpg.de with scanned-ok (Exim 3.35 #1
	(Debian)) id 1BsezP-00089a-00
	for <classpath@gnu.org>; Thu, 05 Aug 2004 11:56:47 +0200
Received: from maildist ([139.19.90.238] helo=mpi-sb.mpg.de)
	by francois.mpi-sb.mpg.de with esmtp (Exim 3.35 #1 (Debian))
	id 1BsezM-00089Q-00; Thu, 05 Aug 2004 11:56:44 +0200
Received: from kaffe.org (topic@mpiat2313 [139.19.20.102])
	by mpi-sb.mpg.de (8.12.9+Sun/8.12.9) with ESMTP id i759uidx001579;
	Thu, 5 Aug 2004 11:56:44 +0200 (MEST)
Message-ID: <4112045C.1010107@kaffe.org>
Date: Thu, 05 Aug 2004 11:56:44 +0200
From: Dalibor Topic <robilad@kaffe.org>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US;
	rv:1.6) Gecko/20040413 Debian/1.6-5
X-Accept-Language: en
MIME-Version: 1.0
To: Guilhem Lavaux <guilhem@kaffe.org>
References: <410F4E47.4040907@kaffe.org>
In-Reply-To: <410F4E47.4040907@kaffe.org>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
X-Virus-Scanned: by AMaViS perl-11
Cc: classpath@gnu.org
Subject: Re: [Fwd: [kaffe] (no subject)] (jni.h)
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Thu, 05 Aug 2004 10:01:19 -0000

Guilhem Lavaux wrote:
> Hi,
> 
> As it is a Classpath's file, I'm forwarding this mail here. I've added a 
> small amount of lines for win32 at the top but it is not related.

Sounds OK to me. It's bad programming style to depend on includes in 
other included files, but if the single line change is required to build 
eclipse's native parts, so be it.

cheers,
dalibor topic



From MAILER-DAEMON Thu Aug 05 16:25:59 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BsooI-0007mF-U6
	for mharc-classpath@gnu.org; Thu, 05 Aug 2004 16:25:59 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BsooH-0007lm-7H
	for classpath@gnu.org; Thu, 05 Aug 2004 16:25:57 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BsooG-0007lT-FG
	for classpath@gnu.org; Thu, 05 Aug 2004 16:25:56 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BsooG-0007lQ-Cw
	for classpath@gnu.org; Thu, 05 Aug 2004 16:25:56 -0400
Received: from [207.189.223.52] (helo=email3.peakpeak.com)
	by monty-python.gnu.org with esmtp (TLSv1:DES-CBC3-SHA:168)
	(Exim 4.34) id 1Bsokc-0003D5-4B
	for classpath@gnu.org; Thu, 05 Aug 2004 16:22:10 -0400
Received: (qmail 2850 invoked by uid 513); 5 Aug 2004 20:31:28 -0000
Received: from tromey@redhat.com by email3 by uid 504 with qmail-scanner-1.20 
	(clamscan: 0.65.  Clear:RC:1(204.144.239.10):. 
	Processed in 0.028879 secs); 05 Aug 2004 20:31:28 -0000
Received: from tf0010.peakpeak.com (HELO fleche.redhat.com) ([204.144.239.10])
	(envelope-sender <tromey@redhat.com>)
	by email3.peakpeak.com (qmail-ldap-1.03) with SMTP
	for <classpath@gnu.org>; 5 Aug 2004 20:31:28 -0000
Received: by fleche.redhat.com (Postfix, from userid 1000)
	id D48F74F8059; Thu,  5 Aug 2004 14:06:01 -0600 (MDT)
To: GNU Classpath Project <classpath@gnu.org>
From: Tom Tromey <tromey@redhat.com>
X-Attribution: Tom
Date: 05 Aug 2004 14:06:01 -0600
Message-ID: <87zn5973km.fsf@fleche.redhat.com>
Lines: 35
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Subject: Branch for generics
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
Reply-To: tromey@redhat.com
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Thu, 05 Aug 2004 20:25:57 -0000

I'm going to make a branch for generics work, including all the other
parts of 1.5; for the most part these rely on generics and can't be
done independently anyway.  I plan to name it "generics-branch".

* Commit rules.
  Only commit changes related to generics or other 1.5 work; other
  changes should go on the trunk (so no sneaking random changes in
  via the branch :-).
  Post your patches as usual, but put `[generics]' in the Subject.
  There is no need for approval for these patches.

* Lifetime.
  I propose we merge the branch to mainline when two criteria are met:

  1. A majority of Classpath-using VM implementors agree that they
  have access to a compiler suitable for compiling the branch.

  2. A majority of Classpath-using VMs can actually use this version
  of Classpath, though perhaps not perfectly.  (In particular here
  I'm thinking that it is probably ok if the new reflection code does
  not work -- the VM will still be able to bootstrap, it just won't
  implement all of 1.5 faithfully.)

* What's there.
  So far I've written Iterable and converted a good chunk of the
  existing collections classes.  None of this has been compiled at
  all yet.  Feel free to help out, there is plenty to do :-)


I'm personally planning to use this code to help debug the generics
(and other 1.5 stuff) in gcjx; hopefully it can serve the same purpose
for the generics branch of jikes.  (Probably a quicker route to a free
1.5-ish compiler will be via Eclipse ...)

Tom



From MAILER-DAEMON Fri Aug 06 01:12:35 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1Bsx1v-0008Dj-Dq
	for mharc-classpath@gnu.org; Fri, 06 Aug 2004 01:12:35 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1Bsx1u-0008DN-9e
	for classpath@gnu.org; Fri, 06 Aug 2004 01:12:34 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1Bsx1t-0008D7-PU
	for classpath@gnu.org; Fri, 06 Aug 2004 01:12:34 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33)
	id 1Bsx1t-0008Cq-0A; Fri, 06 Aug 2004 01:12:33 -0400
Received: from [167.206.5.85] (helo=mta10.srv.hcvlny.cv.net)
	by monty-python.gnu.org with esmtp (Exim 4.34)
	id 1Bswxq-0004xm-ME; Fri, 06 Aug 2004 01:08:22 -0400
Received: from arcadia.augart.com
	(ool-182ca563.dyn.optonline.net [24.44.165.99]) by
	mta10.srv.hcvlny.cv.net
	(iPlanet Messaging Server 5.2 HotFix 1.25 (built Mar  3 2004))
	with ESMTP id <0I2000EKDE9JII@mta10.srv.hcvlny.cv.net>; Fri,
	06 Aug 2004 01:08:07 -0400 (EDT)
Received: from watson.ibm.com (dyn-101.augart.com [10.0.0.101])
	by arcadia.augart.com (Postfix on SuSE Linux 8.0 (i386))
	with ESMTP	id 7193CAB8E; Fri, 06 Aug 2004 01:08:21 -0400 (EDT)
Date: Fri, 06 Aug 2004 01:08:20 -0400
From: Steven Augart <augart@watson.ibm.com>
In-reply-to: <16657.5735.622933.194807@cuddles.cambridge.redhat.com>
To: Andrew Haley <aph@redhat.com>
Message-id: <41131244.4040609@watson.ibm.com>
MIME-version: 1.0
Content-type: text/plain; charset=ISO-8859-1; format=flowed
Content-transfer-encoding: 7BIT
X-Accept-Language: en-us, en
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.4)
	Gecko/20030624
References: <16657.5735.622933.194807@cuddles.cambridge.redhat.com>
Cc: classpath@gnu.org, classpath-patches@gnu.org, java-patches@gcc.gnu.org
Subject: Re: Patch: BasicPermission constructor
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Fri, 06 Aug 2004 05:12:34 -0000

I approve

Andrew Haley wrote:

>2004-08-04  Andrew Haley  <aph@redhat.com>
>
>	* java/security/BasicPermission.java: Don't check wildcards.
>  
>
[....]

>+    // This routine used to check for illegal wildcards, but no such
>+    // requirement exists in the specification and Sun's runtime
>+    // doesn't appear to do it.
>
This looks fine.  I reviwed the API spec on 
java.security.BasicPermission and I agree with your conclusion.  I 
approve committing it to classpath.

I'm CC'ing this reply to classpath-patches because that's the list where 
most of this stuff is getting processed, as you probably know already.  
In this case, it's enough of a functionality change that I understand 
why you posted it to classpath.  Makes sense.

--Steve Augart






From MAILER-DAEMON Fri Aug 06 01:17:33 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1Bsx6j-0003ZE-SK
	for mharc-classpath@gnu.org; Fri, 06 Aug 2004 01:17:33 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1Bsx6i-0003Yz-9P
	for classpath@gnu.org; Fri, 06 Aug 2004 01:17:32 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1Bsx6h-0003Yn-Oo
	for classpath@gnu.org; Fri, 06 Aug 2004 01:17:32 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1Bsx6h-0003Yk-Mj
	for classpath@gnu.org; Fri, 06 Aug 2004 01:17:31 -0400
Received: from [167.206.5.85] (helo=mta10.srv.hcvlny.cv.net)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1Bsx2k-0005a6-9x
	for classpath@gnu.org; Fri, 06 Aug 2004 01:13:26 -0400
Received: from arcadia.augart.com
	(ool-182ca563.dyn.optonline.net [24.44.165.99]) by
	mta10.srv.hcvlny.cv.net
	(iPlanet Messaging Server 5.2 HotFix 1.25 (built Mar  3 2004))
	with ESMTP id <0I2000GMTEHZK1@mta10.srv.hcvlny.cv.net> for
	classpath@gnu.org; Fri, 06 Aug 2004 01:13:11 -0400 (EDT)
Received: from watson.ibm.com (dyn-101.augart.com [10.0.0.101])
	by arcadia.augart.com (Postfix on SuSE Linux 8.0 (i386))
	with ESMTP	id A2A1CAB8E; Fri, 06 Aug 2004 01:13:25 -0400 (EDT)
Date: Fri, 06 Aug 2004 01:13:25 -0400
From: Steven Augart <augart@watson.ibm.com>
In-reply-to: <87zn5973km.fsf@fleche.redhat.com>
To: tromey@redhat.com
Message-id: <41131375.5090702@watson.ibm.com>
MIME-version: 1.0
Content-type: text/plain; charset=ISO-8859-1; format=flowed
Content-transfer-encoding: 7BIT
X-Accept-Language: en-us, en
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.4)
	Gecko/20030624
References: <87zn5973km.fsf@fleche.redhat.com>
Cc: GNU Classpath Project <classpath@gnu.org>
Subject: Re: Branch for generics
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Fri, 06 Aug 2004 05:17:32 -0000

Tom Tromey wrote:

>I'm going to make a branch for generics work, including all the other
>parts of 1.5; for the most part these rely on generics and can't be
>done independently anyway.  I plan to name it "generics-branch".
>
This makse sense. 


>
>* Commit rules. [....]
>
I agree with the commit rule.  They seem clean.  I suppose it makes 
sense not to wait for approval, since only a few people will be working 
on that branch for now.  Your suggestion of putting [generics] in the 
subject line is wise.

>* Lifetime.
>  I propose we merge the branch to mainline when two criteria are met:
>
>  1. A majority of Classpath-using VM implementors agree that they
>  have access to a compiler suitable for compiling the branch.
>
>  2. A majority of Classpath-using VMs can actually use this version
>  of Classpath, though perhaps not perfectly.  (In particular here
>  I'm thinking that it is probably ok if the new reflection code does
>  not work -- the VM will still be able to bootstrap, it just won't
>  implement all of 1.5 faithfully.)
>
This is a good idea too. 

I'm personally not going to be using it for a while; other fish to fry....

Incidentally, I should mention here that I'll be on a much-needed 
vacation Saturday, August 7th, through Sunday, August 15th. 

--Steve Augart






From MAILER-DAEMON Fri Aug 06 01:28:35 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BsxHO-0005kk-Um
	for mharc-classpath@gnu.org; Fri, 06 Aug 2004 01:28:35 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BsxHM-0005kO-N6
	for classpath@gnu.org; Fri, 06 Aug 2004 01:28:32 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BsxHM-0005kC-4R
	for classpath@gnu.org; Fri, 06 Aug 2004 01:28:32 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BsxHM-0005k9-1G
	for classpath@gnu.org; Fri, 06 Aug 2004 01:28:32 -0400
Received: from [167.206.5.70] (helo=mta4.srv.hcvlny.cv.net)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BsxDN-0006kD-Mp
	for classpath@gnu.org; Fri, 06 Aug 2004 01:24:25 -0400
Received: from arcadia.augart.com
	(ool-182ca563.dyn.optonline.net [24.44.165.99]) by
	mta4.srv.hcvlny.cv.net
	(iPlanet Messaging Server 5.2 HotFix 1.25 (built Mar  3 2004))
	with ESMTP id <0I2000D34F0P84@mta4.srv.hcvlny.cv.net> for
	classpath@gnu.org; Fri, 06 Aug 2004 01:24:26 -0400 (EDT)
Received: from watson.ibm.com (dyn-101.augart.com [10.0.0.101])
	by arcadia.augart.com (Postfix on SuSE Linux 8.0 (i386))
	with ESMTP	id 153C4AB8E; Fri, 06 Aug 2004 01:24:25 -0400 (EDT)
Date: Fri, 06 Aug 2004 01:24:24 -0400
From: Steven Augart <augart@watson.ibm.com>
In-reply-to: <1091043617.28981.41.camel@localhost>
To: Mark Wielaard <mark@klomp.org>
Message-id: <41131608.6050606@watson.ibm.com>
MIME-version: 1.0
Content-type: text/plain; charset=ISO-8859-1; format=flowed
Content-transfer-encoding: 7BIT
X-Accept-Language: en-us, en
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.4)
	Gecko/20030624
References: <87zn5kb08b.fsf@fleche.redhat.com>
	<200407282104.30462.konqueror@gmx.de>
	<1091042022.28981.17.camel@localhost>
	<200407282132.07137.konqueror@gmx.de>
	<1091043617.28981.41.camel@localhost>
Cc: tromey@redhat.com, classpath@gnu.org, Michael Koch <konqueror@gmx.de>
Subject: Re: building without gtk
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Fri, 06 Aug 2004 05:28:32 -0000

Mark Wielaard wrote:

>Hi,
>
>On Wed, 2004-07-28 at 21:32, Michael Koch wrote:
>  
>
>>I think we should make it enabled by 
>>default if gtk+ gets found, and disable otherwise. We will still be 
>>able to disable it. That's not hard to do.
>>    
>>
>
>No, we should not silently give the user a crippled AWT. 
>
This makes sense, in general. 

However, it would be really really nice if the version checks happened 
first.  It would be even nicer if several version checks were performed, 
so that the user gets all of the complaints at once.  When I was having 
trouble with this, it was annoying having a bunch of checks succeed only 
then to find out I had the wrong version of package Z, then to rerun, 
wait, and find that I had the wrong version of package Y, etc.

I finally started building on AIX with --disable-gtk-peer, but it wasted 
a fair amount of time.  For some reason that I do not understand, 
Classpath's configure script runs really slowly on a new-ish 64-bit PPC 
AIX machine -- several times slower than it does on the laptop I'm 
typing this on (Pentium III/1.1 GHz, Linux).

--Steven Augart





From MAILER-DAEMON Fri Aug 06 01:51:07 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BsxdC-0000h9-OO
	for mharc-classpath@gnu.org; Fri, 06 Aug 2004 01:51:06 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BsxdA-0000h0-AS
	for classpath@gnu.org; Fri, 06 Aug 2004 01:51:04 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1Bsxd9-0000go-OR
	for classpath@gnu.org; Fri, 06 Aug 2004 01:51:04 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1Bsxd9-0000gl-M6
	for classpath@gnu.org; Fri, 06 Aug 2004 01:51:03 -0400
Received: from [167.206.5.75] (helo=mta8.srv.hcvlny.cv.net)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BsxZH-0000xZ-W2
	for classpath@gnu.org; Fri, 06 Aug 2004 01:47:04 -0400
Received: from arcadia.augart.com
	(ool-182ca563.dyn.optonline.net [24.44.165.99]) by
	mta8.srv.hcvlny.cv.net
	(iPlanet Messaging Server 5.2 HotFix 1.25 (built Mar  3 2004))
	with ESMTP id <0I2000BFIG2628@mta8.srv.hcvlny.cv.net> for
	classpath@gnu.org; Fri, 06 Aug 2004 01:46:54 -0400 (EDT)
Received: from watson.ibm.com (dyn-101.augart.com [10.0.0.101])
	by arcadia.augart.com (Postfix on SuSE Linux 8.0 (i386))
	with ESMTP	id 4AC72AB8E; Fri, 06 Aug 2004 01:47:03 -0400 (EDT)
Date: Fri, 06 Aug 2004 01:47:02 -0400
From: Steven Augart <augart@watson.ibm.com>
In-reply-to: <200408031329.01616.konqueror@gmx.de>
To: Michael Koch <konqueror@gmx.de>
Message-id: <41131B56.5080707@watson.ibm.com>
MIME-version: 1.0
Content-type: text/plain; charset=ISO-8859-15; format=flowed
Content-transfer-encoding: 7BIT
X-Accept-Language: en-us, en
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.4)
	Gecko/20030624
References: <200408031329.01616.konqueror@gmx.de>
Cc: pingo@aicas.com, classpath@gnu.org
Subject: Re: Removing the TARGET_* layer or not ?
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Fri, 06 Aug 2004 05:51:04 -0000

Michael Koch wrote:

>[...] When I wanted to do the needed changes I 
>stumbled over the TARGET_* layer which adds (in my eyes) some 
>unneeded complexity. It makes it more hard to read the code and all 
>the stuff seem to be implemented either in target/generic or 
>target/Linux with no real rule what have to go where. It's just 
>difficult to understand. In fact I try to understand it since a long 
>time and always fail because of its "indirectness". Personally I 
>think some AUTOCONF checks would be more appropriate and would make 
>the code much more readable and bugfixable. There are some known bugs 
>in it but noone attacked them because none understands the code it 
>seems.
>
You expressed my own feelings very well, Michael.   I found the TARGET_ 
layer very unintuitive.  Perhaps if I understood the problems of these 
embedded systems that use unfree compilers, I might feel differently.

Part of the problem is that Classpath is currently targeted to C '89/C 
'90, not to C '99.  So we don't get features like 64-bit integers.  
Perhaps if we were to go over to the newer standard (now five years 
old!), we could dispense with stuff like:
  #define TARGET_NATIVE_MATH_INT_UINT64_TO_UINT32(v) ((jint)(v))
which appears in generic/target_generic_math_int.h.

(I would also like to point out that the above definition appears wrong 
to me -- it casts the (unsigned 64 bit) value v to a signed integer.  I 
have my copy of the ANSI C '89 standard back at the office, so I can't 
quote the standard section on it.)

A great many of the operations in generic/target_generic_math_int.h 
don't seem to be used by Classpath.  I wonder whether AICAS uses this 
stuff on a variety of systems, not just GNU Classpath.

If we are going to be using this stuff, we need a clear justification 
for it.  I would suggest that we create subdirectories for a couple of 
these embedded platforms that Classpath is running on, and populate 
them, so that we can see what these extreme cases actually look like.  
As I understand matters, if AICAS is distributing software that uses 
Classpath on such platforms, then they also must be distributing the 
source code for the portability layer that Classpath requires in order 
to work on those platforms.  If so, then we should have that all in the 
source tree, so that we can see why this is needed.  Does this make 
sense?  (It's almost two in the morning here in New York, so it may not :))

--Steve Augart







From MAILER-DAEMON Fri Aug 06 09:28:28 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1Bt4lo-0005Ku-Dx
	for mharc-classpath@gnu.org; Fri, 06 Aug 2004 09:28:28 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1Bt4lk-0005JU-LF
	for classpath@gnu.org; Fri, 06 Aug 2004 09:28:25 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1Bt4lj-0005Ie-Qo
	for classpath@gnu.org; Fri, 06 Aug 2004 09:28:24 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1Bt4lj-0005I4-MW
	for classpath@gnu.org; Fri, 06 Aug 2004 09:28:23 -0400
Received: from [82.161.94.186] (helo=gnu.wildebeest.org)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1Bt4ho-0001G2-83
	for classpath@gnu.org; Fri, 06 Aug 2004 09:24:20 -0400
Received: from elsschot.wildebeest.org ([192.168.1.26])
	by gnu.wildebeest.org with esmtp (Exim 3.35 #1 (Debian))
	id 1Bt4hG-00048f-00
	for <classpath@gnu.org>; Fri, 06 Aug 2004 15:23:46 +0200
From: Mark Wielaard <mark@klomp.org>
To: classpath@gnu.org
In-Reply-To: <41131B56.5080707@watson.ibm.com>
References: <200408031329.01616.konqueror@gmx.de>
	<41131B56.5080707@watson.ibm.com>
Content-Type: multipart/signed; micalg=pgp-sha1;
	protocol="application/pgp-signature";
	boundary="=-J0OG3FIinl/vt30FDyS0"
Message-Id: <1091798611.7864.164.camel@localhost>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.4.6 
Date: Fri, 06 Aug 2004 15:23:31 +0200
Subject: Re: Removing the TARGET_* layer or not ?
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Fri, 06 Aug 2004 13:28:26 -0000


--=-J0OG3FIinl/vt30FDyS0
Content-Type: text/plain
Content-Transfer-Encoding: quoted-printable

Hi,

 BTW for people who have no idea what this is actually about please read
 about the TARGET layer in native/target/readme.txt (GNU Classpath CVS).
 Autoconfiscating is explained in "The Goat Book":
 http://sources.redhat.com/autobook/autobook/autobook_237.html

On Fri, 2004-08-06 at 07:47, Steven Augart wrote:
> If we are going to be using this stuff, we need a clear justification=20
> for it.  I would suggest that we create subdirectories for a couple of=20
> these embedded platforms that Classpath is running on, and populate=20
> them, so that we can see what these extreme cases actually look like. =20
> As I understand matters, if AICAS is distributing software that uses=20
> Classpath on such platforms, then they also must be distributing the=20
> source code for the portability layer that Classpath requires in order=20
> to work on those platforms.  If so, then we should have that all in the=20
> source tree, so that we can see why this is needed.  Does this make=20
> sense?  (It's almost two in the morning here in New York, so it may not :=
))

I think this makes sense and explains why people see the TARGET layer as
just an extra (obfuscated) layer of indirection. We don't have any
evidence that it actually works since we don't use it ourselves. And
people are looking at libgcj and kaffe and notice they do have many
ports to different platforms, are mostly similar to GNU Classpath, and
do use autoconf to select the implementation they need for each
platform. So clearly people want to mimic that. And it is what the GNU C
coding standards recommend: http://www.gnu.org/prep/standards_27.html

Tom Tromey said that for libgcj they use a hybrid approach. For Posix
like systems they use just one source file that uses autoconf checks to
see what/how things work. For completely different systems they just
compile/link against a completely different implementation source file.

I think this is the way forward if we want to both keep the TARGET layer
for some systems and autoconf for those systems were it works. The trick
would be to come up with an good design/interface to make this work. My
suggestion would be to better split the JNI functions from the actual
implementation. In each native package directory we already have a
java[pacakge].c file that defines common pacakge functions
(native/jni/java-io/javaio.c, native/jni/java-nio/java_nio.c,
 native/jni/java-net/javanet.c, etc).
I propose we put all real functionality in these files and make the
java_[package]_[class].c JNI source files do only the JNI wrapping of
these real implementation methods. So all TARGET layer or
autoconfiscated stuff would be moved inside the common package functions
C source files.

I haven't actually tried doing this, so it might not work, but it might
be a way to support both autoconfiscation and a target layer for those
platforms that aren't autoconfiscatable.

One additional benefit might be that if the common package functions C
source files don't contain any JNI constructs that we might be able to
more easily share native implementations between JNI and CNI based
systems like libgcj. (This is just speculation though, I haven't
investigated if we can actually do this.)

Cheers,

Mark

--=-J0OG3FIinl/vt30FDyS0
Content-Type: application/pgp-signature; name=signature.asc
Content-Description: This is a digitally signed message part

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)

iD8DBQBBE4ZTxVhZCJWr9QwRAj6VAJ0cK1RyEohz1sUTkB6majp1Juvx6gCfXWN/
0FOkHAZAkPl/3KE9Pq4eu3Y=
=85V4
-----END PGP SIGNATURE-----

--=-J0OG3FIinl/vt30FDyS0--




From MAILER-DAEMON Fri Aug 06 12:08:30 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1Bt7Gg-0002kb-Ju
	for mharc-classpath@gnu.org; Fri, 06 Aug 2004 12:08:30 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1Bt7Ge-0002iy-B7
	for classpath@gnu.org; Fri, 06 Aug 2004 12:08:28 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1Bt7GZ-0002gG-Rg
	for classpath@gnu.org; Fri, 06 Aug 2004 12:08:27 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1Bt7GZ-0002fq-Jt
	for classpath@gnu.org; Fri, 06 Aug 2004 12:08:23 -0400
Received: from [139.19.1.2] (helo=francois.mpi-sb.mpg.de)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1Bt7Cl-0007N4-66
	for classpath@gnu.org; Fri, 06 Aug 2004 12:04:27 -0400
Received: from amavis by francois.mpi-sb.mpg.de with scanned-ok (Exim 3.35 #1
	(Debian)) id 1Bt7Cj-0002UO-00
	for <classpath@gnu.org>; Fri, 06 Aug 2004 18:04:25 +0200
Received: from maildist ([139.19.90.238] helo=mpi-sb.mpg.de)
	by francois.mpi-sb.mpg.de with esmtp (Exim 3.35 #1 (Debian))
	id 1Bt7Ch-0002UG-00; Fri, 06 Aug 2004 18:04:23 +0200
Received: from kaffe.org (topic@mpiat2313 [139.19.20.102])
	by mpi-sb.mpg.de (8.12.9+Sun/8.12.9) with ESMTP id i76G4Mdx005031;
	Fri, 6 Aug 2004 18:04:22 +0200 (MEST)
Message-ID: <4113AC06.6070802@kaffe.org>
Date: Fri, 06 Aug 2004 18:04:22 +0200
From: Dalibor Topic <robilad@kaffe.org>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US;
	rv:1.6) Gecko/20040413 Debian/1.6-5
X-Accept-Language: en
MIME-Version: 1.0
To: "C. Brian Jones" <cbj@gnu.org>
References: <20040804131835.A18964@urbanophile.com>
	<1091681242.4640.132.camel@lyta.haphazard.org>
In-Reply-To: <1091681242.4640.132.camel@lyta.haphazard.org>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
X-Virus-Scanned: by AMaViS perl-11
Cc: classpath@gnu.org, "Aaron M. Renn" <arenn@urbanophile.com>
Subject: Re: OT: Amsterdam
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Fri, 06 Aug 2004 16:08:29 -0000

C. Brian Jones wrote:
> On Wed, 2004-08-04 at 14:18, Aaron M. Renn wrote:
> 
>>Hello, all.  I'm traveling to Amsterdam on business for the next few
>>weeks and wondered if any Classpath hackers were in the vicinity.  I
>>really have not gotten a chance to meet most of the folks here, and
>>would love to do so if possible, particularly in a place where I'm
>>not likely to travel frequently.  Happy hacking, everyone!
> 
> 
> Aaron, you're alive!  ;)  I'm traveling to France the last two weeks of
> Sept. for personal vacation, but I don't know of anyone using Classpath
> there.  I will be in Strasbourg for a day or two, Paris, Avignon,
> Mont-Saint-Michel, is that anywhere near anyone?  :)
> 

Hi Aaron, hello Brian,

I'm living in Saarbruecken[1][2], Germany, a small border town between 
Germany and France, tucked in some 100 km away from Luxembourg or 
Strasbourg. Feel free to drop by if you've got a day or two to spare :)

cheers,
dalibor topic

[1] http://de.wikipedia.org/wiki/Saarbr%C3%BCcken
[2] 
http://world-map.perbang.dk/?map.x=248&map.y=119&zoom=2500&msize=1&c11=1&c12=1&c13=1&border=1&river=1&lake=1&grid=1&cords=1&gcircle=1&cwc=1&timezone=1&olat=2956.28&aid=62&olon=391.96&oyoffset=-73907&oxoffset=9799&ozoom=2500



From MAILER-DAEMON Fri Aug 06 12:20:54 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1Bt7Sg-00063F-LO
	for mharc-classpath@gnu.org; Fri, 06 Aug 2004 12:20:54 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1Bt7Sf-00062R-N1
	for classpath@gnu.org; Fri, 06 Aug 2004 12:20:53 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1Bt7Sd-00060Q-Ku
	for classpath@gnu.org; Fri, 06 Aug 2004 12:20:53 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1Bt7Sd-00060G-HS
	for classpath@gnu.org; Fri, 06 Aug 2004 12:20:51 -0400
Received: from [212.227.126.191] (helo=moutng.kundenserver.de)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1Bt7Oz-0000mq-Kp
	for classpath@gnu.org; Fri, 06 Aug 2004 12:17:05 -0400
Received: from [212.227.126.205] (helo=mrelayng.kundenserver.de)
	by moutng.kundenserver.de with esmtp (Exim 3.35 #1)
	id 1Bt7Oy-0007Pi-00
	for classpath@gnu.org; Fri, 06 Aug 2004 18:17:04 +0200
Received: from [212.126.219.82] (helo=wau)
	by mrelayng.kundenserver.de with asmtp (TLSv1:RC4-MD5:128)
	(Exim 3.35 #1) id 1Bt7Oy-0007pX-00; Fri, 06 Aug 2004 18:17:04 +0200
Content-Type: text/plain;
  charset="iso-8859-15"
From: Andy Walter <anwalt_pop@aicas.com>
To: classpath@gnu.org
Date: Fri, 6 Aug 2004 18:17:03 +0200
X-Mailer: KMail [version 1.4]
References: <200408031329.01616.konqueror@gmx.de>
	<41131B56.5080707@watson.ibm.com>
	<1091798611.7864.164.camel@localhost>
In-Reply-To: <1091798611.7864.164.camel@localhost>
MIME-Version: 1.0
Content-Transfer-Encoding: quoted-printable
Message-Id: <200408061817.03221.anwalt_pop@aicas.com>
X-Provags-ID: kundenserver.de abuse@kundenserver.de
	auth:b72082335eb2dc13c82d55d68fe167f3
Cc: Fridtjof Siebert <siebert@aicas.com>, James Jeffers Hunt <jjh@aicas.com>
Subject: Re: Removing the TARGET_* layer or not ?
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Fri, 06 Aug 2004 16:20:54 -0000

Hi,

I'm sorry for the trouble which was caused by the target layer. Actually,=
 I'm=20
quite surprised about the massive critique now. When we asked to check it=
 in=20
more than one year ago, there was very little feedback and the feedback w=
e=20
got was mainly quite positive.

We are currently preparing the native layers of

- RTEMS
- Solaris
- embOS

for check in into the Classpath repository next week. Especially the embO=
S=20
layer for file I/O illustrates some bigger differences of embedded operat=
ed=20
systems. Since embOS is unfree, I expect the RTEMS port will be more=20
interesting to most people in this list. (I have to admit that the=20
differences are not as big there).

Some of the native interfaces we are using here are even more complicated=
 than=20
the embOS one, because they are for special operating systems which are o=
nly=20
used internally in a customer's company, but not publicly available as a=20
product. Those systems would probably be of little interest here (because=
=20
they are not available at all) and we would not even be allowed to publis=
h=20
the implementations for them. We might be able to provide more platforms =
in=20
the future.

It's okay by us to modify or change things in the native layer. Removing =
it=20
completely means kicking aicas out of the project. After one and a half y=
ears=20
of working on the project, this would be really sad for us. I would be gl=
ad=20
if we can come to a solution which is okay for everybody.


I wish you a nice weekend,

cheers,

=09Andy.

--=20
aicas GmbH                                     =20
Haid-und-Neu-Stra=DFe 18 * 76131 Karlsruhe       =20
http://www.aicas.com                           =20
Tel: +49-721-663 968-24; Fax: +49-721-663 968-94




From MAILER-DAEMON Fri Aug 06 23:31:57 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BtHw5-0005dg-AK
	for mharc-classpath@gnu.org; Fri, 06 Aug 2004 23:31:57 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BtHw3-0005dZ-RG
	for classpath@gnu.org; Fri, 06 Aug 2004 23:31:55 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BtHw0-0005dN-Vh
	for classpath@gnu.org; Fri, 06 Aug 2004 23:31:55 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BtHw0-0005dK-RV
	for classpath@gnu.org; Fri, 06 Aug 2004 23:31:52 -0400
Received: from [129.34.20.6] (helo=igw2.watson.ibm.com)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BtHs0-00084P-Uh
	for classpath@gnu.org; Fri, 06 Aug 2004 23:27:45 -0400
Received: from sp1n293en1.watson.ibm.com (sp1n293en1.watson.ibm.com
	[129.34.20.41])
	by igw2.watson.ibm.com (8.11.7-20030924/8.11.4) with ESMTP id
	i773Q5c99094; Fri, 6 Aug 2004 23:26:05 -0400
Received: from [9.65.196.75] (sig-9-65-196-75.mts.ibm.com [9.65.196.75])
	by sp1n293en1.watson.ibm.com (8.11.7-20030924/8.11.7/8.11.7-01-14-2004)
	with ESMTP id i773RfD38678; Fri, 6 Aug 2004 23:27:41 -0400
Message-ID: <41144C2B.6000402@watson.ibm.com>
Date: Fri, 06 Aug 2004 23:27:39 -0400
From: Steven Augart <augart@watson.ibm.com>
Organization: IBM Watson Research Ctr., Jikes RVM Project
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7) Gecko/20040616
X-Accept-Language: ast, gl, en-us, en, it, eo, pt, es-ec, es-us, es, fr, de, el
MIME-Version: 1.0
To: Andy Walter <anwalt_pop@aicas.com>
References: <200408031329.01616.konqueror@gmx.de>	<41131B56.5080707@watson.ibm.com>	<1091798611.7864.164.camel@localhost>
	<200408061817.03221.anwalt_pop@aicas.com>
In-Reply-To: <200408061817.03221.anwalt_pop@aicas.com>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Cc: classpath@gnu.org, Fridtjof Siebert <siebert@aicas.com>,
	James Jeffers Hunt <jjh@aicas.com>
Subject: Classpath License, Phillosophy, and AICAS (was: Re: Removing the
 TARGET_* layer or not ?)
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Sat, 07 Aug 2004 03:31:56 -0000

Andy Walter wrote:
> Some of the native interfaces we are using here are even more complicated than 
> the embOS one, because they are for special operating systems which are only 
> used internally in a customer's company, but not publicly available as a 
> product. Those systems would probably be of little interest here (because 
> they are not available at all) and we would not even be allowed to publish 
> the implementations for them. We might be able to provide more platforms in 
> the future.

I assume you're absolutely certain that those programs on those 
operating systems are not being further redistributed by the companies 
in question, that they're only being used internally.  And I assume 
those customers are fully aware that any distribution they do of the 
resulting software (including as part of an embedded system) must 
include source code or an offer of source code, pursuant to GPL 
clauses 3a and 3b.

Are there any of the implementations that don't have a 64-bit type 
that you can share?

> It's okay by us to modify or change things in the native layer. Removing it 
> completely means kicking aicas out of the project. After one and a half years 
> of working on the project, this would be really sad for us. I would be glad 
> if we can come to a solution which is okay for everybody.
This is a difficult issue here.  I am writing up a technical (software 
engineering perspective) note on the topic, which I will send 
separately to avoid mixing licensing issues with technical ones.

I'll be on a week's vacation starting Saturday (New York time); my 
responses may be delayed.


-- 
Steven Augart

Jikes RVM, a free, open source, Virtual Machine:
http://oss.software.ibm.com/jikesrvm



From MAILER-DAEMON Sat Aug 07 00:33:55 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BtIu2-0001Rz-SX
	for mharc-classpath@gnu.org; Sat, 07 Aug 2004 00:33:55 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BtIu2-0001Ru-66
	for classpath@gnu.org; Sat, 07 Aug 2004 00:33:54 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BtIu1-0001Rh-Fd
	for classpath@gnu.org; Sat, 07 Aug 2004 00:33:53 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BtIu1-0001Re-By
	for classpath@gnu.org; Sat, 07 Aug 2004 00:33:53 -0400
Received: from [129.34.20.6] (helo=igw2.watson.ibm.com)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BtIqM-0006uc-26
	for classpath@gnu.org; Sat, 07 Aug 2004 00:30:06 -0400
Received: from sp1n293en1.watson.ibm.com (sp1n293en1.watson.ibm.com
	[129.34.20.41])
	by igw2.watson.ibm.com (8.11.7-20030924/8.11.4) with ESMTP id
	i774SQc34056; Sat, 7 Aug 2004 00:28:26 -0400
Received: from [9.65.196.75] (sig-9-65-196-75.mts.ibm.com [9.65.196.75])
	by sp1n293en1.watson.ibm.com (8.11.7-20030924/8.11.7/8.11.7-01-14-2004)
	with ESMTP id i774U2D35752; Sat, 7 Aug 2004 00:30:02 -0400
Message-ID: <41145AC7.3060602@watson.ibm.com>
Date: Sat, 07 Aug 2004 00:29:59 -0400
From: Steven Augart <augart@watson.ibm.com>
Organization: IBM Watson Research Ctr., Jikes RVM Project
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7) Gecko/20040616
X-Accept-Language: ast, gl, en-us, en, it, eo, pt, es-ec, es-us, es, fr, de, el
MIME-Version: 1.0
To: Andy Walter <anwalt_pop@aicas.com>
References: <200408031329.01616.konqueror@gmx.de>	<41131B56.5080707@watson.ibm.com>	<1091798611.7864.164.camel@localhost>
	<200408061817.03221.anwalt_pop@aicas.com>
In-Reply-To: <200408061817.03221.anwalt_pop@aicas.com>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Cc: classpath@gnu.org, Fridtjof Siebert <siebert@aicas.com>,
	James Jeffers Hunt <jjh@aicas.com>
Subject: Re: Removing the TARGET_* layer or not ?
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Sat, 07 Aug 2004 04:33:54 -0000

Andy Walter wrote:
> Hi,
> 
> I'm sorry for the trouble which was caused by the target layer. Actually, I'm 
> quite surprised about the massive critique now. When we asked to check it in 
> more than one year ago, there was very little feedback and the feedback we 
> got was mainly quite positive.
Andy,

Thank you for the kind note.  I think some of the difference may be 
that several of us (I'm thinking of Michael Koch in particular) now 
have been doing considerable work in C on JNI.

> We are currently preparing the native layers of
> 
> - RTEMS
> - Solaris
> - embOS

> for check in into the Classpath repository next week. Especially the embOS 
> layer for file I/O illustrates some bigger differences of embedded operated 
> systems. Since embOS is unfree, I expect the RTEMS port will be more 
> interesting to most people in this list. (I have to admit that the 
> differences are not as big there).


I appreciate you preparing the native layers for inclusion into the 
classpath repository.  Is any one of those platforms a platform which 
has a compiler that lacks a 64-bit type?

General Questions:

1) Is any classpath user other than AICAS on systems where there are 
only C '89 compilers available?

2) Anyone other than AICAS on systems where there is no 64-bit type 
available?

3) Anyone other than AICAS on systems where there is no support for 
"inline"?  And, if you are, is avoiding the overhead of two or three 
function calls when performing a C call file handling functions very 
very important to you?

4) Anyone, other than AICAS, on a system that doesn't support POSIX?

If only AICAS has some of those four constraints (I suspect this is 
so), then I think the Classpath project (that's us, folks!) will have 
to look at this from a software project management point of view, and 
make a direction decision about whether it is appropriate for the 
project's native code to meet those constraints, given the negative 
consequences that follow:

The consequence of constraint (1) is that we are restricting ourselves 
to C '89.  (For example, I define a variable in gthread-jni.c that I 
do not use until about a hundred lines later.  This is complicated.) 
The consequence of constraint (2) is that all operations on a "jlong", 
except for passing them as parameters (this includes comparison for 
equality and assignment to anything other than another jlong) have to 
be wrapped with macros whose names take up half of a screen line.  The 
consequence of the two parts of constraint (3) is that we apparently 
have to define use macros for operations that on every other 
multi-platform system I've worked on are handled via a uniform 
function interface (such as Mozilla's "nspr" portable runtime).   The 
consequence of (4) is that we have to worry about multi-platform 
support for what would otherwise be simple calls to open(2), seek(2), 
and fstat(2).

The megative consequences of all of these put together is to hide bugs 
that would otherwise be obvious to experienced C programmers, and to 
discourage some programmers from working on the native code.  This is 
  the case because the JNI code in Classpath is not really written in 
C.  It's written in a stack of macros on top of macros on top of 
macros, which eventually turn into C.

I am looking forward to learning more about the portability needs of 
the many projects that use Classpath.

-- 
Steven Augart

Jikes RVM, a free, open source, Virtual Machine:
http://oss.software.ibm.com/jikesrvm



From MAILER-DAEMON Sat Aug 07 12:25:01 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BtU0C-0002ri-PI
	for mharc-classpath@gnu.org; Sat, 07 Aug 2004 12:25:00 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BtU0B-0002rL-0I
	for classpath@gnu.org; Sat, 07 Aug 2004 12:24:59 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BtU08-0002r0-7k
	for classpath@gnu.org; Sat, 07 Aug 2004 12:24:58 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BtU08-0002qw-4w
	for classpath@gnu.org; Sat, 07 Aug 2004 12:24:56 -0400
Received: from [80.91.224.249] (helo=main.gmane.org)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BtTw6-0006Qt-PO
	for classpath@gnu.org; Sat, 07 Aug 2004 12:20:47 -0400
Received: from root by main.gmane.org with local (Exim 3.35 #1 (Debian))
	id 1BtTw5-0005fy-00
	for <classpath@gnu.org>; Sat, 07 Aug 2004 18:20:45 +0200
Received: from pppfree174-164-tn.aknet.it ([213.21.174.164])
	by main.gmane.org with esmtp (Gmexim 0.1 (Debian))
	id 1AlnuQ-0007hv-00
	for <classpath@gnu.org>; Sat, 07 Aug 2004 18:20:45 +0200
Received: from motse by pppfree174-164-tn.aknet.it with local (Gmexim 0.1
	(Debian)) id 1AlnuQ-0007hv-00
	for <classpath@gnu.org>; Sat, 07 Aug 2004 18:20:45 +0200
X-Injected-Via-Gmane: http://gmane.org/
To: classpath@gnu.org
From: "news.gmane.org" <motse@complang.tuwien.ac.at>
Date: Sat, 07 Aug 2004 18:15:16 +0200
Lines: 27
Message-ID: <cf2v27$q7r$1@sea.gmane.org>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
X-Complaints-To: usenet@sea.gmane.org
X-Gmane-NNTP-Posting-Host: pppfree174-164-tn.aknet.it
User-Agent: Mozilla Thunderbird 0.7.1 (X11/20040731)
X-Accept-Language: en-us, en
X-Enigmail-Version: 0.84.2.0
X-Enigmail-Supports: pgp-inline, pgp-mime
Sender: news <news@sea.gmane.org>
Subject: Missing GdkClasspathFontPeer files in classpath 0.10
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Sat, 07 Aug 2004 16:24:59 -0000

Hi all !


I would like to know why the files

native/jni/gtk-peer/gnu_java_awt_peer_gtk_GdkClasspathFontPeer.c
native/jni/gtk-peer/gnu_java_awt_peer_gtk_GdkClasspathFontPeerMetrics.c

have not been included in the classpath 0.10 release ?

Due to this I get the following errors:

LOG: warning: native function initStaticState: ()V not found in class 
gnu/java/awt/peer/gtk/GdkClasspathFontPeer
LOG: warning: native function initState: ()V not found in class 
gnu/java/awt/peer/gtk/GdkClasspathFontPeer
LOG: warning: native function dispose: ()V not found in class 
gnu/java/awt/peer/gtk/GdkClasspathFontPeer
LOG: warning: native function setFont: (Ljava/lang/String;II)V not found 
in class gnu/java/awt/peer/gtk/GdkClasspathFontPeer


Is it possible to get this running without those files or do I have to 
use the cvs version ?


Martin




From MAILER-DAEMON Sat Aug 07 18:29:00 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BtZgR-0000jP-VP
	for mharc-classpath@gnu.org; Sat, 07 Aug 2004 18:29:00 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BtZgR-0000jK-7i
	for classpath@gnu.org; Sat, 07 Aug 2004 18:28:59 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BtZgQ-0000j8-KW
	for classpath@gnu.org; Sat, 07 Aug 2004 18:28:58 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BtZgQ-0000j5-Gx
	for classpath@gnu.org; Sat, 07 Aug 2004 18:28:58 -0400
Received: from [207.189.223.52] (helo=email3.peakpeak.com)
	by monty-python.gnu.org with esmtp (TLSv1:DES-CBC3-SHA:168)
	(Exim 4.34) id 1BtZcN-0002wl-JE
	for classpath@gnu.org; Sat, 07 Aug 2004 18:24:47 -0400
Received: (qmail 21401 invoked by uid 513); 7 Aug 2004 22:34:48 -0000
Received: from tromey@redhat.com by email3 by uid 504 with qmail-scanner-1.20 
	(clamscan: 0.65.  Clear:RC:1(204.144.239.10):. 
	Processed in 0.021919 secs); 07 Aug 2004 22:34:48 -0000
Received: from tf0010.peakpeak.com (HELO fleche.redhat.com) ([204.144.239.10])
	(envelope-sender <tromey@redhat.com>)
	by email3.peakpeak.com (qmail-ldap-1.03) with SMTP
	for <motse@complang.tuwien.ac.at>; 7 Aug 2004 22:34:48 -0000
Received: by fleche.redhat.com (Postfix, from userid 1000)
	id 4CB844F82A5; Sat,  7 Aug 2004 16:08:32 -0600 (MDT)
To: "news.gmane.org" <motse@complang.tuwien.ac.at>
References: <cf2v27$q7r$1@sea.gmane.org>
From: Tom Tromey <tromey@redhat.com>
X-Attribution: Tom
Date: 07 Aug 2004 16:08:31 -0600
In-Reply-To: <cf2v27$q7r$1@sea.gmane.org>
Message-ID: <87pt6238kg.fsf@fleche.redhat.com>
Lines: 19
User-Agent: Gnus/5.09 (Gnus v5.9.0) Emacs/21.3.50
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Cc: classpath@gnu.org
Subject: Re: Missing GdkClasspathFontPeer files in classpath 0.10
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
Reply-To: tromey@redhat.com
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Sat, 07 Aug 2004 22:28:59 -0000

>>>>> "Martin" == news gmane org <motse@complang.tuwien.ac.at> writes:

Martin> I would like to know why the files
Martin> native/jni/gtk-peer/gnu_java_awt_peer_gtk_GdkClasspathFontPeer.c
Martin> native/jni/gtk-peer/gnu_java_awt_peer_gtk_GdkClasspathFontPeerMetrics.c
Martin> have not been included in the classpath 0.10 release ?

Looks like a Makefile bug to me.
In libgcj these files are conditionally built when Cairo is used.
In Classpath they don't appear in the Makefile at all.  Classpath
also doesn't seem to have configury to select whether Cairo is used.

Martin> Is it possible to get this running without those files or do I
Martin> have to use the cvs version ?

I'm not sure using cvs will fix the problem, since as far as I can
tell these files are never compiled.

Tom



From MAILER-DAEMON Sat Aug 07 19:11:03 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BtaL9-00057g-2I
	for mharc-classpath@gnu.org; Sat, 07 Aug 2004 19:11:03 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BtaL6-00056T-7S
	for classpath@gnu.org; Sat, 07 Aug 2004 19:11:00 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BtaL4-00055x-KC
	for classpath@gnu.org; Sat, 07 Aug 2004 19:10:58 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BtaL3-00055N-Pz
	for classpath@gnu.org; Sat, 07 Aug 2004 19:10:57 -0400
Received: from [24.24.2.58] (helo=ms-smtp-04.nyroc.rr.com)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BtaGU-0007uP-4s
	for classpath@gnu.org; Sat, 07 Aug 2004 19:06:14 -0400
Received: from [192.168.0.43] (alb-24-195-113-26.nycap.rr.com [24.195.113.26])
	by ms-smtp-04.nyroc.rr.com (8.12.10/8.12.10) with ESMTP id
	i77N5nt2000943
	for <classpath@gnu.org>; Sat, 7 Aug 2004 19:05:50 -0400 (EDT)
From: James Damour <jdamour@nycap.rr.com>
To: classpath@gnu.org
Content-Type: multipart/signed; micalg=pgp-sha1;
	protocol="application/pgp-signature";
	boundary="=-o6DVPR6DUejZNf1kaeRR"
Organization: SourceForge.net
Message-Id: <1091920958.12611.451.camel@hovel.lepus.org>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.4.6 
Date: Sat, 07 Aug 2004 19:22:38 -0400
X-Virus-Scanned: Symantec AntiVirus Scan Engine
Subject: Re: Branch for generics
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
Reply-To: "James Damour \(Suvarov454\)" <suvarov454@users.sourceforge.net>
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Sat, 07 Aug 2004 23:11:01 -0000


--=-o6DVPR6DUejZNf1kaeRR
Content-Type: text/plain
Content-Transfer-Encoding: quoted-printable

Tom Tromey wrote:
>
>  I propose we merge the branch to mainline when two criteria are met:
>
>  1. A majority of Classpath-using VM implementors agree that they
>  have access to a compiler suitable for compiling the branch.

Maybe you should pick an actual number instead of saying "majority";
otherwise people can claim that we aren't including **every**
Classpath-using VM in our count. =20

Alternatively, you could just say "When the following Classpath-using VM
implementors agree that they have access to a compiler suitable for
compiling the branch: Kaffe, SableVM, GCJ, JikesRVM, JamVM, and
Kissme".  Modify the list to be whomever you want, but IMHO if these
guys can't use it then even if everyone else can, it isn't ready.

--=20
James Damour (Suvarov454) <suvarov454@users.sourceforge.net>

--=-o6DVPR6DUejZNf1kaeRR
Content-Type: application/pgp-signature; name=signature.asc
Content-Description: This is a digitally signed message part

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)

iD8DBQBBFWQ+JtJj6pH+EFQRAv+4AJ4+oxa82O+P+ZRliZIRSErEuWCsVgCfa9N1
nfN/wYBWYb5MwCRJkfp/r/E=
=4EYy
-----END PGP SIGNATURE-----

--=-o6DVPR6DUejZNf1kaeRR--




From MAILER-DAEMON Sun Aug 08 17:08:27 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1Btuu3-0006zu-Da
	for mharc-classpath@gnu.org; Sun, 08 Aug 2004 17:08:27 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1Btuu2-0006zj-AR
	for classpath@gnu.org; Sun, 08 Aug 2004 17:08:26 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1Btuu0-0006zK-GT
	for classpath@gnu.org; Sun, 08 Aug 2004 17:08:25 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1Btuu0-0006zH-Dx
	for classpath@gnu.org; Sun, 08 Aug 2004 17:08:24 -0400
Received: from [82.161.94.186] (helo=gnu.wildebeest.org)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1Btupb-0001PP-AR
	for classpath@gnu.org; Sun, 08 Aug 2004 17:03:51 -0400
Received: from elsschot.wildebeest.org ([192.168.1.26])
	by gnu.wildebeest.org with esmtp (Exim 3.35 #1 (Debian))
	id 1Btup3-00032s-00; Sun, 08 Aug 2004 23:03:17 +0200
From: Mark Wielaard <mark@klomp.org>
To: classpath@gnu.org
In-Reply-To: <1091560480.7431.51.camel@qcplx01.physto.se>
References: <1091560480.7431.51.camel@qcplx01.physto.se>
Content-Type: multipart/signed; micalg=pgp-sha1;
	protocol="application/pgp-signature";
	boundary="=-FSNdDNK+M1QVetHiwSB1"
Message-Id: <1091998982.7864.640.camel@localhost>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.4.6 
Date: Sun, 08 Aug 2004 23:03:02 +0200
Cc: tromey@redhat.com
Subject: Re: Line2D intersection
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Sun, 08 Aug 2004 21:08:26 -0000


--=-FSNdDNK+M1QVetHiwSB1
Content-Type: text/plain
Content-Transfer-Encoding: quoted-printable

Hi,

On Tue, 2004-08-03 at 21:14, Sven de Marothy wrote:
> Just one suggestion: I'm (hopefully) going to be done with Area before
> the end of the month, and the remaining Arc2D stuff (and whatever else
> is left in geom.*) should be pretty trivial to do. So instead of making
> a bunch of small commits, I'd like to roll this all into one big geom.*
> patch, since any updates to this will need to be sent over to the
> gcj-gui-branch guys, and it'd be less work for them.
>=20
> Does that sound OK?

We are luckily in a bit better position these days.
http://gcc.gnu.org/java/gui-compare/libgcj-classpath-compare.html
And I try to keep the trees in sync at the end of each week (10 of the
diffs shown today at that URL have been merged). That seems doable. So
if you have smaller chunks it should also be OK now.

> Oh, and keep up the good work on writing tests! It's very useful!
>=20
> (If only someone with Mauve CVS would commit them for you..
> doesn't seem to be much action on that mailing-list)

Yeah, sorry about missing those patches. But I have committed them now.
Tom Tromey can provide people with commit access to mauve to make
contributing easier.

BTW Tom, the reason I missed the patches was because they were send to
the mauve-patches mailinglist which is not used very much. Maybe you
could just forward everything send to mauve-patches to the mauve-discuss
mailinglist? mauve-discuss has only a few postings each month and
mauve-patches only a few each year.

Cheers,

Mark

--=-FSNdDNK+M1QVetHiwSB1
Content-Type: application/pgp-signature; name=signature.asc
Content-Description: This is a digitally signed message part

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)

iD8DBQBBFpUFxVhZCJWr9QwRAuG5AJsGyLKfv+Ixfp1PPR0KFMuzha8xNgCdEeL1
o84m8C93UU6WFB4dicysUqA=
=gv2w
-----END PGP SIGNATURE-----

--=-FSNdDNK+M1QVetHiwSB1--




From MAILER-DAEMON Sun Aug 08 17:10:05 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1Btuva-0008Hn-6e
	for mharc-classpath@gnu.org; Sun, 08 Aug 2004 17:10:02 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BtuvX-0008Gb-BL
	for classpath@gnu.org; Sun, 08 Aug 2004 17:09:59 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BtuvV-0008Fm-DY
	for classpath@gnu.org; Sun, 08 Aug 2004 17:09:57 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BtuvV-0008FA-BJ
	for classpath@gnu.org; Sun, 08 Aug 2004 17:09:57 -0400
Received: from [82.161.94.186] (helo=gnu.wildebeest.org)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BturK-0001bG-SE
	for classpath@gnu.org; Sun, 08 Aug 2004 17:05:39 -0400
Received: from elsschot.wildebeest.org ([192.168.1.26])
	by gnu.wildebeest.org with esmtp (Exim 3.35 #1 (Debian))
	id 1Btuqm-00033H-00; Sun, 08 Aug 2004 23:05:04 +0200
From: Mark Wielaard <mark@klomp.org>
To: "Aaron M. Renn" <arenn@urbanophile.com>
In-Reply-To: <20040804131835.A18964@urbanophile.com>
References: <20040804131835.A18964@urbanophile.com>
Content-Type: multipart/signed; micalg=pgp-sha1;
	protocol="application/pgp-signature";
	boundary="=-I/zEWDifUJkOAxxqq8Kd"
Message-Id: <1091999088.7864.643.camel@localhost>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.4.6 
Date: Sun, 08 Aug 2004 23:04:48 +0200
Cc: classpath@gnu.org
Subject: Re: OT: Amsterdam
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Sun, 08 Aug 2004 21:10:00 -0000


--=-I/zEWDifUJkOAxxqq8Kd
Content-Type: text/plain
Content-Transfer-Encoding: quoted-printable

Hi,

On Wed, 2004-08-04 at 20:18, Aaron M. Renn wrote:
> Hello, all.  I'm traveling to Amsterdam on business for the next few
> weeks and wondered if any Classpath hackers were in the vicinity.  I
> really have not gotten a chance to meet most of the folks here, and
> would love to do so if possible, particularly in a place where I'm
> not likely to travel frequently.  Happy hacking, everyone!

I live in Amsterdam and would love to meet up!

Cheers,

Mark

--=-I/zEWDifUJkOAxxqq8Kd
Content-Type: application/pgp-signature; name=signature.asc
Content-Description: This is a digitally signed message part

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)

iD8DBQBBFpVwxVhZCJWr9QwRAnTYAJsEwKuh17vYUqBhFiU8LxkaBojxRACfW1yV
QgJDuL2h6wyz/09Hd/32+pQ=
=Qvh2
-----END PGP SIGNATURE-----

--=-I/zEWDifUJkOAxxqq8Kd--




From MAILER-DAEMON Sun Aug 08 17:19:41 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1Btv4v-000344-ET
	for mharc-classpath@gnu.org; Sun, 08 Aug 2004 17:19:41 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1Btv4t-00033z-Nd
	for classpath@gnu.org; Sun, 08 Aug 2004 17:19:39 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1Btv4s-00033n-7d
	for classpath@gnu.org; Sun, 08 Aug 2004 17:19:39 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1Btv4s-00033k-4M
	for classpath@gnu.org; Sun, 08 Aug 2004 17:19:38 -0400
Received: from [82.161.94.186] (helo=gnu.wildebeest.org)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1Btv0i-0002rW-J3
	for classpath@gnu.org; Sun, 08 Aug 2004 17:15:20 -0400
Received: from elsschot.wildebeest.org ([192.168.1.26])
	by gnu.wildebeest.org with esmtp (Exim 3.35 #1 (Debian))
	id 1Btv0A-00033c-00; Sun, 08 Aug 2004 23:14:46 +0200
From: Mark Wielaard <mark@klomp.org>
To: Dalibor Topic <robilad@kaffe.org>
In-Reply-To: <4112045C.1010107@kaffe.org>
References: <410F4E47.4040907@kaffe.org>  <4112045C.1010107@kaffe.org>
Content-Type: multipart/signed; micalg=pgp-sha1;
	protocol="application/pgp-signature";
	boundary="=-+qrtESvmSTAN90mbxcTf"
Message-Id: <1091999670.7864.650.camel@localhost>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.4.6 
Date: Sun, 08 Aug 2004 23:14:30 +0200
Cc: classpath@gnu.org
Subject: Re: [Fwd: [kaffe] (no subject)] (jni.h)
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Sun, 08 Aug 2004 21:19:39 -0000


--=-+qrtESvmSTAN90mbxcTf
Content-Type: text/plain
Content-Transfer-Encoding: quoted-printable

Hi,

On Thu, 2004-08-05 at 11:56, Dalibor Topic wrote:
> > As it is a Classpath's file, I'm forwarding this mail here. I've added =
a=20
> > small amount of lines for win32 at the top but it is not related.
>=20
> Sounds OK to me. It's bad programming style to depend on includes in=20
> other included files, but if the single line change is required to build=20
> eclipse's native parts, so be it.

I think it is best to report this as a bug to the Eclipse hackers.
Even the simplest JNI example that I could find
http://java.sun.com/docs/books/jni/html/start.html#26865
says you should include stdio.h separately.

Cheers,

Mark

--=-+qrtESvmSTAN90mbxcTf
Content-Type: application/pgp-signature; name=signature.asc
Content-Description: This is a digitally signed message part

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)

iD8DBQBBFpe1xVhZCJWr9QwRAjmaAJ4r8O5111du+XhcliW92bce1G7NfACcDS+/
jY7oEJSsfyeOaUiDRHXyU1Y=
=vFj3
-----END PGP SIGNATURE-----

--=-+qrtESvmSTAN90mbxcTf--




From MAILER-DAEMON Sun Aug 08 17:28:09 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BtvD7-00058w-53
	for mharc-classpath@gnu.org; Sun, 08 Aug 2004 17:28:09 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BtvD5-00058g-GN
	for classpath@gnu.org; Sun, 08 Aug 2004 17:28:07 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BtvD4-00058U-V8
	for classpath@gnu.org; Sun, 08 Aug 2004 17:28:07 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BtvD4-00058R-QA
	for classpath@gnu.org; Sun, 08 Aug 2004 17:28:06 -0400
Received: from [82.161.94.186] (helo=gnu.wildebeest.org)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1Btv8v-0003tf-VG
	for classpath@gnu.org; Sun, 08 Aug 2004 17:23:50 -0400
Received: from elsschot.wildebeest.org ([192.168.1.26])
	by gnu.wildebeest.org with esmtp (Exim 3.35 #1 (Debian))
	id 1Btv86-000343-00; Sun, 08 Aug 2004 23:22:58 +0200
From: Mark Wielaard <mark@klomp.org>
To: Steven Augart <augart@watson.ibm.com>
In-Reply-To: <41131608.6050606@watson.ibm.com>
References: <87zn5kb08b.fsf@fleche.redhat.com>
	<200407282104.30462.konqueror@gmx.de>
	<1091042022.28981.17.camel@localhost>
	<200407282132.07137.konqueror@gmx.de>
	<1091043617.28981.41.camel@localhost>
	<41131608.6050606@watson.ibm.com>
Content-Type: multipart/signed; micalg=pgp-sha1;
	protocol="application/pgp-signature";
	boundary="=-2RXqt+z9i9KmVtC59c1X"
Message-Id: <1092000162.7864.657.camel@localhost>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.4.6 
Date: Sun, 08 Aug 2004 23:22:42 +0200
Cc: tromey@redhat.com, classpath@gnu.org, Michael Koch <konqueror@gmx.de>
Subject: Re: building without gtk
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Sun, 08 Aug 2004 21:28:07 -0000


--=-2RXqt+z9i9KmVtC59c1X
Content-Type: text/plain
Content-Transfer-Encoding: quoted-printable

Hi,

On Fri, 2004-08-06 at 07:24, Steven Augart wrote:
> Mark Wielaard wrote:
> >No, we should not silently give the user a crippled AWT.=20
> >
> This makes sense, in general.=20
>=20
> However, it would be really really nice if the version checks happened=20
> first.  It would be even nicer if several version checks were performed,=20
> so that the user gets all of the complaints at once.  When I was having=20
> trouble with this, it was annoying having a bunch of checks succeed only=20
> then to find out I had the wrong version of package Z, then to rerun,=20
> wait, and find that I had the wrong version of package Y, etc.

They should all be listed in the INSTALL file and/or the hackers guide.
The idea is that when you get a configure error telling you are missing
a package you read the INSTALL file and make sure you have all
dependencies before trying to reconfigure again. If some things aren't
listed there could you add them?

I believe IBM distributes all the dependencies for GNU Classpath on AIX:
http://www-1.ibm.com/servers/aix/products/aixos/linux/download.html

> I finally started building on AIX with --disable-gtk-peer, but it wasted=20
> a fair amount of time.  For some reason that I do not understand,=20
> Classpath's configure script runs really slowly on a new-ish 64-bit PPC=20
> AIX machine -- several times slower than it does on the laptop I'm=20
> typing this on (Pentium III/1.1 GHz, Linux).

I would recommend upgrading to a better, faster and free OS :)

Cheers,

Mark

--=-2RXqt+z9i9KmVtC59c1X
Content-Type: application/pgp-signature; name=signature.asc
Content-Description: This is a digitally signed message part

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)

iD8DBQBBFpmixVhZCJWr9QwRAuVrAJ0RAY/yptT6BOnWfx7EHa/RAYP8VgCgrZhw
r8S/AEx7HSyd+JLVTQVBT5Y=
=jXAA
-----END PGP SIGNATURE-----

--=-2RXqt+z9i9KmVtC59c1X--




From MAILER-DAEMON Sun Aug 08 17:49:14 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BtvXW-0005GD-7z
	for mharc-classpath@gnu.org; Sun, 08 Aug 2004 17:49:14 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BtvXU-0005G7-PN
	for classpath@gnu.org; Sun, 08 Aug 2004 17:49:12 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BtvXU-0005Fv-8A
	for classpath@gnu.org; Sun, 08 Aug 2004 17:49:12 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BtvXU-0005Fs-5J
	for classpath@gnu.org; Sun, 08 Aug 2004 17:49:12 -0400
Received: from [82.161.94.186] (helo=gnu.wildebeest.org)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BtvTj-0006Re-Jf
	for classpath@gnu.org; Sun, 08 Aug 2004 17:45:20 -0400
Received: from elsschot.wildebeest.org ([192.168.1.26])
	by gnu.wildebeest.org with esmtp (Exim 3.35 #1 (Debian))
	id 1BtvT0-00035f-00; Sun, 08 Aug 2004 23:44:34 +0200
From: Mark Wielaard <mark@klomp.org>
To: Andrew Haley <aph@redhat.com>
In-Reply-To: <16656.40429.72617.324082@cuddles.cambridge.redhat.com>
References: <200408031329.01616.konqueror@gmx.de>
	<410F9574.8040408@aicas.com> <1091608847.726.10.camel@moonlight>
	<411098F1.50509@aicas.com>
	<16656.40429.72617.324082@cuddles.cambridge.redhat.com>
Content-Type: multipart/signed; micalg=pgp-sha1;
	protocol="application/pgp-signature";
	boundary="=-7Da7PbnIh9G1XqSBWETk"
Message-Id: <1092001458.7864.670.camel@localhost>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.4.6 
Date: Sun, 08 Aug 2004 23:44:18 +0200
Cc: classpath@gnu.org, Michael Koch <konqueror@gmx.de>
Subject: Re: Removing the TARGET_* layer or not ?
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Sun, 08 Aug 2004 21:49:13 -0000


--=-7Da7PbnIh9G1XqSBWETk
Content-Type: text/plain
Content-Transfer-Encoding: quoted-printable

Hi,

On Wed, 2004-08-04 at 10:27, Andrew Haley wrote:
> I guess as long as we're stuck with C90 we can't use inline.
> Am I right in thinking that GNU Classpath targets C90?

Yes. More accurately would probably be that we try to support platforms
that still depend on gcc-2.95. This is not that uncommon. And it is what
the GNU Coding Standards recommend:
http://www.gnu.org/prep/standards_10.html

We are currently compiling our C code with:
    dnl We want ISO C90 pedantic ansi, but with longlong (jlong) support
    dnl and modern POSIX and BSD C library functions/prototypes.
    AM_CFLAGS=3D'-ansi -pedantic -Wmissing-declarations -Wmissing-prototype=
s -Wstrict-prototypes -Wall -Wno-long-long -D_BSD_SOURCE '

I didn't know that would prevent inlining in general.
But even if it does then we are probably better of with a little slower
code that is easier to debug. GNU Classpath is still very much in rapid
prototyping mode and we should be careful about premature optimization.

> IMO, we shouldn't really be designing Classpath around the
> infelicities of unfree compilers.
>=20
> > > Note, that with this system we would avoid the use of macros altogeth=
er
>  > > (You see, I don't like them ;) I better go with a well-thought-out
>  > > design). Sure, there is the 'overhead' of 1 additional function call=
.
>=20
> Macros are bad for a bunch of reasons, esp. debuggability and type
> safety.

I think this is the thing that most people found at this last year. That
the use of C macros makes debugging very difficult. There are other
inconveniences that come from the use the a extra layer of indirection,
but I believe we can work around most of them as long as we get rid of C
#defines in so many places or make it possible to more easily debug code
that uses them.

Cheers,

Mark

--=-7Da7PbnIh9G1XqSBWETk
Content-Type: application/pgp-signature; name=signature.asc
Content-Description: This is a digitally signed message part

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)

iD8DBQBBFp6yxVhZCJWr9QwRAp2lAJ4zkDDeDqa7TztmqUPDC9tMfkM54gCeIU2N
7b+eoipqcuaLdCpQ+33YPW4=
=wAtj
-----END PGP SIGNATURE-----

--=-7Da7PbnIh9G1XqSBWETk--




From MAILER-DAEMON Sun Aug 08 22:56:42 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1Bu0L4-0008Oh-NI
	for mharc-classpath@gnu.org; Sun, 08 Aug 2004 22:56:42 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1Bu0L2-0008LD-Qq
	for classpath@gnu.org; Sun, 08 Aug 2004 22:56:40 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1Bu0L1-0008Jl-7I
	for classpath@gnu.org; Sun, 08 Aug 2004 22:56:40 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1Bu0L1-0008Ji-3k
	for classpath@gnu.org; Sun, 08 Aug 2004 22:56:39 -0400
Received: from [132.208.250.6] (helo=anis.telecom.uqam.ca)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1Bu0H2-00083Y-Dg
	for classpath@gnu.org; Sun, 08 Aug 2004 22:52:32 -0400
Received: from anis4.telecom.uqam.ca (anis4.telecom.uqam.ca [132.208.250.236])
	by sortant.uqam.ca (8.12.10/8.12.1) with SMTP id i792VB4K020884;
	Sun, 8 Aug 2004 22:33:04 -0400 (EDT)
Received: from antivirus.uqam.ca ([132.208.250.6])
	by anis4.telecom.uqam.ca (SAVSMTP 3.1.1.32) with SMTP id
	M2004080822350706462 ; Sun, 08 Aug 2004 22:35:07 -0400
Received: from 244.30.internet.uqam.ca (244.30.internet.uqam.ca
	[132.208.30.244])
	by intrant.uqam.ca (8.12.10/8.12.2/uqam-filtres) with ESMTP id
	i792HR0p014761; Sun, 8 Aug 2004 22:17:28 -0400 (EDT)
X-Spam-Filter: Filtre-Uqam re:  abuse@uqam.ca
From: "Grzegorz B. Prokopski" <gadek@debian.org>
To: Michael Koch <konqueror@gmx.de>
In-Reply-To: <200408031329.01616.konqueror@gmx.de>
References: <200408031329.01616.konqueror@gmx.de>
Content-Type: text/plain; charset=ISO-8859-2
Organization: Debian GNU/Linux http://www.debian.org
Message-Id: <1092017840.4138.73.camel@gadek.homelinux.org>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.4.6 
Date: Sun, 08 Aug 2004 22:17:21 -0400
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by anis.telecom.uqam.ca id
	i792VB4K020884
Cc: GNU Classpath <classpath@gnu.org>
Subject: Re: Removing the TARGET_* layer or not ?
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Mon, 09 Aug 2004 02:56:41 -0000

W li=B6cie z wto, 03-08-2004, godz. 07:29, Michael Koch pisze:=20
> Ingo: Now my question are you really using the TARGET_* system or is=20
> it only rotting around in GNU classpath ? I really wonder if some=20
> AUTOCONF macros would be more helpful for you ?

I personally have never had a positive experience with these overcomplex
redirections, multiline C macros (yuck!), 100-characters long uppercase
names...

> If it is consensus that noone understands the cod and its good to=20
> rewrite it I can start it. Then I will do it slowly part for part and=20
> try to introduce as less as possible bugs. Unfortunately there will=20
> be bugs introduced. We are all humans. So this will need some testing=20
> on some archs, Linux, *.BSD, AIX, Solaris ... etc.

Yes, that would be really great.  At least it would be easier to
understand the code and remove bugs.  Can we simply have the
implementations directly in i.e. java_lang_...*.c files?
I have already seen Mark proposing another layer of inderection inside
these files, but I am not sure it's really needed.  When it's really
needed we can have i.e java_lang*_WIN.c files that are used instead
of the default ones, or have separate directory for these files.

> What is your opinion on the TARGET_* system ?

I really wish it could just go away.

> It would be nice to get a statement from the AICAS people too as they=20
> introduced it.

They surely had their own, good reasons.  But apparenlty we actually
only have one working implementation of these native parts in GNU CP
anyway and it's been so since the layer was introduced...

I think GCJ's approach is reasonable, if I understood it well.

The basic implementation is for POSIX systems + autoconf for detecting
system's features.  And for the systems that are really too different
from POSIX there might be separate files that replace whole parts of
native layers.  I guess MS Windows port would be (the only?) one such.

Cheers,

				Grzegorz B. Prokopski

--=20
Grzegorz B. Prokopski      <gadek@debian.org>
Debian GNU/Linux           http://www.debian.org
SableVM - LGPL'ed Java VM  http://www.sablevm.org
Why SableVM ?!?            http://devel.sablevm.org/wiki/Features




From MAILER-DAEMON Mon Aug 09 05:32:22 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1Bu6Vy-0001T5-2s
	for mharc-classpath@gnu.org; Mon, 09 Aug 2004 05:32:22 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1Bu6Vw-0001Sx-7B
	for classpath@gnu.org; Mon, 09 Aug 2004 05:32:20 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1Bu6Vu-0001SL-RP
	for classpath@gnu.org; Mon, 09 Aug 2004 05:32:19 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1Bu6Vu-0001SB-OK
	for classpath@gnu.org; Mon, 09 Aug 2004 05:32:18 -0400
Received: from [62.253.162.49] (helo=mta09-svc.ntlworld.com)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1Bu6S2-0005QW-Us
	for classpath@gnu.org; Mon, 09 Aug 2004 05:28:19 -0400
Received: from cuddles.cambridge.redhat.com ([81.96.64.123])
	by mta09-svc.ntlworld.com
	(InterMail vM.4.01.03.37 201-229-121-137-20020806) with ESMTP id
	<20040809092900.KKXW15901.mta09-svc.ntlworld.com@cuddles.cambridge.redhat.com>;
	Mon, 9 Aug 2004 10:29:00 +0100
Received: from redhat.com (localhost.localdomain [127.0.0.1])
	by cuddles.cambridge.redhat.com (8.12.8/8.12.8) with ESMTP id
	i799S42v025355; Mon, 9 Aug 2004 10:28:04 +0100
Received: (from aph@localhost)
	by redhat.com (8.12.8/8.12.8/Submit) id i799S2ql025351;
	Mon, 9 Aug 2004 10:28:02 +0100
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Message-ID: <16663.17314.645718.977374@cuddles.cambridge.redhat.com>
Date: Mon, 9 Aug 2004 10:28:02 +0100
From: Andrew Haley <aph@redhat.com>
To: Mark Wielaard <mark@klomp.org>
In-Reply-To: <1092001458.7864.670.camel@localhost>
References: <200408031329.01616.konqueror@gmx.de> <410F9574.8040408@aicas.com>
	<1091608847.726.10.camel@moonlight> <411098F1.50509@aicas.com>
	<16656.40429.72617.324082@cuddles.cambridge.redhat.com>
	<1092001458.7864.670.camel@localhost>
X-Mailer: VM 7.14 under Emacs 21.3.50.1
Cc: classpath@gnu.org, Michael Koch <konqueror@gmx.de>
Subject: Re: Removing the TARGET_* layer or not ?
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Mon, 09 Aug 2004 09:32:20 -0000

Mark Wielaard writes:
 > 
 > On Wed, 2004-08-04 at 10:27, Andrew Haley wrote:
 > > I guess as long as we're stuck with C90 we can't use inline.
 > > Am I right in thinking that GNU Classpath targets C90?
 > 
 > Yes. More accurately would probably be that we try to support platforms
 > that still depend on gcc-2.95. This is not that uncommon.

Really?  Not uncommon that platforms still depend on 2.95?  Are these
platforms that have been orphande due to lack of maintainers, or some
other reason?

 > And it is what the GNU Coding Standards recommend:
 > http://www.gnu.org/prep/standards_10.html
 > 
 > We are currently compiling our C code with:
 >     dnl We want ISO C90 pedantic ansi, but with longlong (jlong) support
 >     dnl and modern POSIX and BSD C library functions/prototypes.
 >     AM_CFLAGS='-ansi -pedantic -Wmissing-declarations -Wmissing-prototypes -Wstrict-prototypes -Wall -Wno-long-long -D_BSD_SOURCE '
 > 
 > I didn't know that would prevent inlining in general.
 > But even if it does then we are probably better of with a little slower
 > code that is easier to debug. GNU Classpath is still very much in rapid
 > prototyping mode and we should be careful about premature optimization.

Right.  All we have to do to support C90 is `#define inline'
appropriately.  

 > I think this is the thing that most people found at this last year. That
 > the use of C macros makes debugging very difficult. There are other
 > inconveniences that come from the use the a extra layer of indirection,
 > but I believe we can work around most of them as long as we get rid of C
 > #defines in so many places or make it possible to more easily debug code
 > that uses them.

Good.

Andrew.



From MAILER-DAEMON Mon Aug 09 06:03:36 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1Bu70B-0006ez-Am
	for mharc-classpath@gnu.org; Mon, 09 Aug 2004 06:03:35 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1Bu708-0006e7-43
	for classpath@gnu.org; Mon, 09 Aug 2004 06:03:32 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1Bu705-0006d1-Nv
	for classpath@gnu.org; Mon, 09 Aug 2004 06:03:31 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1Bu704-0006cy-LJ
	for classpath@gnu.org; Mon, 09 Aug 2004 06:03:29 -0400
Received: from [139.19.1.2] (helo=francois.mpi-sb.mpg.de)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1Bu6w5-0003WM-Ji
	for classpath@gnu.org; Mon, 09 Aug 2004 05:59:21 -0400
Received: from amavis by francois.mpi-sb.mpg.de with scanned-ok (Exim 3.35 #1
	(Debian)) id 1Bu6w4-000223-00
	for <classpath@gnu.org>; Mon, 09 Aug 2004 11:59:20 +0200
Received: from maildist ([139.19.90.238] helo=mpi-sb.mpg.de)
	by francois.mpi-sb.mpg.de with esmtp (Exim 3.35 #1 (Debian))
	id 1Bu6vu-000209-00; Mon, 09 Aug 2004 11:59:10 +0200
Received: from kaffe.org (topic@mpiat2313 [139.19.20.102])
	by mpi-sb.mpg.de (8.12.9+Sun/8.12.9) with ESMTP id i799x7wg023989;
	Mon, 9 Aug 2004 11:59:08 +0200 (MEST)
Message-ID: <41174AEB.80806@kaffe.org>
Date: Mon, 09 Aug 2004 11:59:07 +0200
From: Dalibor Topic <robilad@kaffe.org>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US;
	rv:1.6) Gecko/20040413 Debian/1.6-5
X-Accept-Language: en
MIME-Version: 1.0
To: Mark Wielaard <mark@klomp.org>
References: <410F4E47.4040907@kaffe.org> <4112045C.1010107@kaffe.org>
	<1091999670.7864.650.camel@localhost>
In-Reply-To: <1091999670.7864.650.camel@localhost>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
X-Virus-Scanned: by AMaViS perl-11
Cc: classpath@gnu.org, Jan Schulz <jan@katzien.de>
Subject: Re: [Fwd: [kaffe] (no subject)] (jni.h)
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Mon, 09 Aug 2004 10:03:32 -0000

Mark Wielaard wrote:
> Hi,
> 
> On Thu, 2004-08-05 at 11:56, Dalibor Topic wrote:
> 
>>>As it is a Classpath's file, I'm forwarding this mail here. I've added a 
>>>small amount of lines for win32 at the top but it is not related.
>>
>>Sounds OK to me. It's bad programming style to depend on includes in 
>>other included files, but if the single line change is required to build 
>>eclipse's native parts, so be it.
> 
> 
> I think it is best to report this as a bug to the Eclipse hackers.
> Even the simplest JNI example that I could find
> http://java.sun.com/docs/books/jni/html/start.html#26865
> says you should include stdio.h separately.

Okay, that makes sense. Thanks for checking, Mark, I'll forward this 
discussion to Jan Schulz, Debian Eclipse packager, and ask him to file a 
bug against eclipse upstream.

cheers,
dalibor topic



From MAILER-DAEMON Mon Aug 09 06:25:37 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1Bu7LV-0001W0-Jb
	for mharc-classpath@gnu.org; Mon, 09 Aug 2004 06:25:37 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1Bu7LS-0001Vg-DT
	for classpath@gnu.org; Mon, 09 Aug 2004 06:25:34 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1Bu7LR-0001VU-Ov
	for classpath@gnu.org; Mon, 09 Aug 2004 06:25:34 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1Bu7LR-0001VC-Lm
	for classpath@gnu.org; Mon, 09 Aug 2004 06:25:33 -0400
Received: from [139.19.1.2] (helo=francois.mpi-sb.mpg.de)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1Bu7HO-0000Og-EP
	for classpath@gnu.org; Mon, 09 Aug 2004 06:21:22 -0400
Received: from amavis by francois.mpi-sb.mpg.de with scanned-ok (Exim 3.35 #1
	(Debian)) id 1Bu7HN-0004t9-00
	for <classpath@gnu.org>; Mon, 09 Aug 2004 12:21:21 +0200
Received: from maildist ([139.19.90.238] helo=mpi-sb.mpg.de)
	by francois.mpi-sb.mpg.de with esmtp (Exim 3.35 #1 (Debian))
	id 1Bu7HJ-0004rz-00; Mon, 09 Aug 2004 12:21:17 +0200
Received: from kaffe.org (topic@mpiat2313 [139.19.20.102])
	by mpi-sb.mpg.de (8.12.9+Sun/8.12.9) with ESMTP id i79ALGwg006889;
	Mon, 9 Aug 2004 12:21:16 +0200 (MEST)
Message-ID: <4117501C.2040904@kaffe.org>
Date: Mon, 09 Aug 2004 12:21:16 +0200
From: Dalibor Topic <robilad@kaffe.org>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US;
	rv:1.6) Gecko/20040413 Debian/1.6-5
X-Accept-Language: en
MIME-Version: 1.0
To: Andrew Haley <aph@redhat.com>
References: <200408031329.01616.konqueror@gmx.de>
	<410F9574.8040408@aicas.com>	<1091608847.726.10.camel@moonlight>
	<411098F1.50509@aicas.com>	<16656.40429.72617.324082@cuddles.cambridge.redhat.com>	<1092001458.7864.670.camel@localhost>
	<16663.17314.645718.977374@cuddles.cambridge.redhat.com>
In-Reply-To: <16663.17314.645718.977374@cuddles.cambridge.redhat.com>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
X-Virus-Scanned: by AMaViS perl-11
Cc: Mark Wielaard <mark@klomp.org>, classpath@gnu.org,
	Michael Koch <konqueror@gmx.de>
Subject: Re: Removing the TARGET_* layer or not ?
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Mon, 09 Aug 2004 10:25:34 -0000

Andrew Haley wrote:
> Mark Wielaard writes:
>  > 
>  > On Wed, 2004-08-04 at 10:27, Andrew Haley wrote:
>  > > I guess as long as we're stuck with C90 we can't use inline.
>  > > Am I right in thinking that GNU Classpath targets C90?
>  > 
>  > Yes. More accurately would probably be that we try to support platforms
>  > that still depend on gcc-2.95. This is not that uncommon.
> 
> Really?  Not uncommon that platforms still depend on 2.95?  Are these
> platforms that have been orphande due to lack of maintainers, or some
> other reason?

Current stable *BSD releases use gcc2. NetBSD is switching to 3.3.1 for 
2.0. FreeBSD 5.3 may end up with gcc 3.3.3. OpenBSD 3.5 uses 2.95.3 for 
some arches, 3.3.2 on others.

I assume that gcc2 will be history in a year or two ;)

cheers,
dalibor topic



From MAILER-DAEMON Mon Aug 09 12:38:54 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BuDAk-0000ad-6w
	for mharc-classpath@gnu.org; Mon, 09 Aug 2004 12:38:54 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BuDAi-0000Yu-9y
	for classpath@gnu.org; Mon, 09 Aug 2004 12:38:52 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BuDAg-0000Wa-BE
	for classpath@gnu.org; Mon, 09 Aug 2004 12:38:51 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BuDAg-0000WG-8h
	for classpath@gnu.org; Mon, 09 Aug 2004 12:38:50 -0400
Received: from [207.174.178.92] (helo=email2.peakpeak.com)
	by monty-python.gnu.org with esmtp (TLSv1:DES-CBC3-SHA:168)
	(Exim 4.34) id 1BuD6b-0000ks-BJ
	for classpath@gnu.org; Mon, 09 Aug 2004 12:34:37 -0400
Received: (qmail 22826 invoked by uid 513); 9 Aug 2004 16:30:48 -0000
Received: from tromey@redhat.com by email2 by uid 504 with qmail-scanner-1.20 
	(clamscan: 0.65.  Clear:RC:1(204.144.239.10):. 
	Processed in 0.463576 secs); 09 Aug 2004 16:30:48 -0000
Received: from tf0010.peakpeak.com (HELO fleche.redhat.com) ([204.144.239.10])
	(envelope-sender <tromey@redhat.com>)
	by email2.peakpeak.com (qmail-ldap-1.03) with SMTP
	for <classpath@gnu.org>; 9 Aug 2004 16:30:47 -0000
Received: by fleche.redhat.com (Postfix, from userid 1000)
	id 6340D4F8059; Mon,  9 Aug 2004 10:07:49 -0600 (MDT)
To: "James Damour (Suvarov454)" <suvarov454@users.sourceforge.net>
References: <1091920958.12611.451.camel@hovel.lepus.org>
From: Tom Tromey <tromey@redhat.com>
X-Attribution: Tom
Date: 09 Aug 2004 10:07:49 -0600
In-Reply-To: <1091920958.12611.451.camel@hovel.lepus.org>
Message-ID: <87ekmg1ei2.fsf@fleche.redhat.com>
Lines: 15
User-Agent: Gnus/5.09 (Gnus v5.9.0) Emacs/21.3.50
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Cc: classpath@gnu.org
Subject: Re: Branch for generics
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
Reply-To: tromey@redhat.com
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Mon, 09 Aug 2004 16:38:52 -0000

>>>>> "James" == James Damour <jdamour@nycap.rr.com> writes:

>> 1. A majority of Classpath-using VM implementors agree that they
>> have access to a compiler suitable for compiling the branch.

James> Maybe you should pick an actual number instead of saying "majority";
James> otherwise people can claim that we aren't including **every**
James> Classpath-using VM in our count.  

Yeah, suggesting a majority might have actually been too
specific...  maybe it is better to just leave it at "once the branch
is ready, we can negotiate".  Basically I don't want folks to worry
that this stuff will suddenly show up on the trunk without warning.

Tom



From MAILER-DAEMON Mon Aug 09 18:30:50 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BuIfK-0001f6-BQ
	for mharc-classpath@gnu.org; Mon, 09 Aug 2004 18:30:50 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BuIfH-0001eP-OO
	for classpath@gnu.org; Mon, 09 Aug 2004 18:30:48 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BuIfF-0001dn-S0
	for classpath@gnu.org; Mon, 09 Aug 2004 18:30:47 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BuIfF-0001dc-Oz
	for classpath@gnu.org; Mon, 09 Aug 2004 18:30:45 -0400
Received: from [82.161.94.186] (helo=gnu.wildebeest.org)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BuIbR-00030d-2y
	for classpath@gnu.org; Mon, 09 Aug 2004 18:26:49 -0400
Received: from elsschot.wildebeest.org ([192.168.1.26])
	by gnu.wildebeest.org with esmtp (Exim 3.35 #1 (Debian))
	id 1BuIan-0006L7-00; Tue, 10 Aug 2004 00:26:09 +0200
From: Mark Wielaard <mark@klomp.org>
To: "news.gmane.org" <motse@complang.tuwien.ac.at>
In-Reply-To: <cf2v27$q7r$1@sea.gmane.org>
References: <cf2v27$q7r$1@sea.gmane.org>
Content-Type: multipart/signed; micalg=pgp-sha1;
	protocol="application/pgp-signature";
	boundary="=-W7gCrMB8gYiH4Wo7RJcJ"
Message-Id: <1092090353.29997.76.camel@localhost>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.4.6 
Date: Tue, 10 Aug 2004 00:25:53 +0200
Cc: classpath@gnu.org, graydon@redhat.com
Subject: Re: Missing GdkClasspathFontPeer files in classpath 0.10
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Mon, 09 Aug 2004 22:30:48 -0000


--=-W7gCrMB8gYiH4Wo7RJcJ
Content-Type: multipart/mixed; boundary="=-fFTP9/BBqAzhcQ62LpeK"


--=-fFTP9/BBqAzhcQ62LpeK
Content-Type: text/plain
Content-Transfer-Encoding: quoted-printable

Hi,

On Sat, 2004-08-07 at 18:15, news.gmane.org wrote:
> native/jni/gtk-peer/gnu_java_awt_peer_gtk_GdkClasspathFontPeer.c
> native/jni/gtk-peer/gnu_java_awt_peer_gtk_GdkClasspathFontPeerMetrics.c
>=20
> have not been included in the classpath 0.10 release ?

They are part of the cairo/pango setup (for Graphics2D written by
Graydon Hoare) which were never enabled in GNU Classpath.

> Due to this I get the following errors:
>=20
> LOG: warning: native function initStaticState: ()V not found in class=20
> gnu/java/awt/peer/gtk/GdkClasspathFontPeer
> LOG: warning: native function initState: ()V not found in class=20
> gnu/java/awt/peer/gtk/GdkClasspathFontPeer
> LOG: warning: native function dispose: ()V not found in class=20
> gnu/java/awt/peer/gtk/GdkClasspathFontPeer
> LOG: warning: native function setFont: (Ljava/lang/String;II)V not found=20
> in class gnu/java/awt/peer/gtk/GdkClasspathFontPeer

Interesting. Which tool gives this output?
We could use something that warns us about missing native methods.

> Is it possible to get this running without those files or do I have to=20
> use the cvs version ?

No, CVS still has the same problem.
You can try the attached patch though (and get those two C files from
CVS). It adds all build infrastructure. But does not detect whether the
correct cairo libs have been installed and have not been tested at all.
You will have to configure with --enable-regen-headers.

Graydon, how ready is this code and how would people test it out?

Cheers,

Mark



--=-fFTP9/BBqAzhcQ62LpeK
Content-Disposition: inline; filename=build.patch
Content-Transfer-Encoding: base64
Content-Type: text/x-patch; name=build.patch; charset=ISO-8859-1
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--=-fFTP9/BBqAzhcQ62LpeK--

--=-W7gCrMB8gYiH4Wo7RJcJ
Content-Type: application/pgp-signature; name=signature.asc
Content-Description: This is a digitally signed message part

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)

iD8DBQBBF/nxxVhZCJWr9QwRAu1qAJ9c9+EaXchZ72uVbT4j5k6yHhqPkQCfeqxB
UshuumBsVvFEJxZj7A2jyvg=
=Mfba
-----END PGP SIGNATURE-----

--=-W7gCrMB8gYiH4Wo7RJcJ--




From MAILER-DAEMON Mon Aug 09 19:16:17 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BuJNJ-0005lm-7U
	for mharc-classpath@gnu.org; Mon, 09 Aug 2004 19:16:17 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BuJNH-0005lN-DH
	for classpath@gnu.org; Mon, 09 Aug 2004 19:16:15 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BuJNG-0005lB-P5
	for classpath@gnu.org; Mon, 09 Aug 2004 19:16:15 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BuJNG-0005l8-MM
	for classpath@gnu.org; Mon, 09 Aug 2004 19:16:14 -0400
Received: from [66.187.233.31] (helo=mx1.redhat.com)
	by monty-python.gnu.org with esmtp (TLSv1:DES-CBC3-SHA:168)
	(Exim 4.34) id 1BuJJM-0002Fl-Fk
	for classpath@gnu.org; Mon, 09 Aug 2004 19:12:12 -0400
Received: from int-mx1.corp.redhat.com (int-mx1.corp.redhat.com
	[172.16.52.254])
	by mx1.redhat.com (8.12.10/8.12.10) with ESMTP id i79NBte1032062;
	Mon, 9 Aug 2004 19:11:55 -0400
Received: from dub.venge.net (sebastian-int.corp.redhat.com [172.16.52.221])
	by int-mx1.corp.redhat.com (8.11.6/8.11.6) with ESMTP id i79NBsa26751; 
	Mon, 9 Aug 2004 19:11:54 -0400
Date: Mon, 09 Aug 2004 19:11:55 -0400
To: "Mark Wielaard" <mark@klomp.org>,
	"news.gmane.org" <motse@complang.tuwien.ac.at>
References: <cf2v27$q7r$1@sea.gmane.org> <1092090353.29997.76.camel@localhost>
From: "graydon hoare" <graydon@redhat.com>
Organization: red hat
Content-Type: text/plain; format=flowed; delsp=yes; charset=utf-8
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Message-ID: <opschmp5y7tgmzw2@dub.venge.net>
In-Reply-To: <1092090353.29997.76.camel@localhost>
User-Agent: Opera M2/7.53 (Linux, build 737)
Cc: classpath@gnu.org
Subject: Re: Missing GdkClasspathFontPeer files in classpath 0.10
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Mon, 09 Aug 2004 23:16:15 -0000

On Tue, 10 Aug 2004 00:25:53 +0200, Mark Wielaard <mark@klomp.org> wrote:

> You can try the attached patch though (and get those two C files from
> CVS). It adds all build infrastructure. But does not detect whether the
> correct cairo libs have been installed and have not been tested at all.
> You will have to configure with --enable-regen-headers.
>
> Graydon, how ready is this code and how would people test it out?

it works, more or less. it's slower than gdk, but then, it's doing a lot
more (and nobody's profiled or optimized it). also some of the font stuff
is more fragile, and some of the fancier stuff (custom compositing operators)
just fail outright.

to test it out, you have to get it built (and get the configury finding the
appropriate cflags, etc. as in libgcj), and then there's a system property you
set to switch between GdkGraphics and GdkGraphics2D (which uses cairo):

  -Dgnu.java.awt.peer.gtk.Graphics=Graphics     <- uses GDK (the default)
  -Dgnu.java.awt.peer.gtk.Graphics=Graphics2D   <- uses cairo

-graydon



From MAILER-DAEMON Tue Aug 10 07:34:55 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BuUu7-0000z0-Fz
	for mharc-classpath@gnu.org; Tue, 10 Aug 2004 07:34:55 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BuUu5-0000yZ-L4
	for classpath@gnu.org; Tue, 10 Aug 2004 07:34:53 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BuUu4-0000y1-1D
	for classpath@gnu.org; Tue, 10 Aug 2004 07:34:53 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BuUu3-0000xr-SU
	for classpath@gnu.org; Tue, 10 Aug 2004 07:34:51 -0400
Received: from [212.227.126.191] (helo=moutng.kundenserver.de)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BuUpb-0003eo-EX
	for classpath@gnu.org; Tue, 10 Aug 2004 07:30:16 -0400
Received: from [212.227.126.208] (helo=mrelayng.kundenserver.de)
	by moutng.kundenserver.de with esmtp (Exim 3.35 #1)
	id 1BuUpX-0005e6-00
	for classpath@gnu.org; Tue, 10 Aug 2004 13:30:11 +0200
Received: from [212.126.219.82] (helo=aicas.com)
	by mrelayng.kundenserver.de with asmtp (TLSv1:RC4-MD5:128)
	(Exim 3.35 #1) id 1BuUpW-0002j0-00; Tue, 10 Aug 2004 13:30:10 +0200
Message-ID: <4118B1C1.7000408@aicas.com>
Date: Tue, 10 Aug 2004 13:30:09 +0200
From: "Dr. Torsten Rupp" <rupp@aicas.com>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.4) Gecko/20030624
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: classpath@gnu.org
X-Enigmail-Version: 0.76.5.0
X-Enigmail-Supports: pgp-inline, pgp-mime
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
X-Provags-ID: kundenserver.de abuse@kundenserver.de
	auth:f9c39743fddc2557f7fa7c1a513f7ea5
Subject: Target native layer
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Tue, 10 Aug 2004 11:34:54 -0000

Dear Classpath-developers,

I read many emails in the list about the TARGET_*-layer. I'm not very 
happy about the discussion, because we discussed this already around 1 
year ago and at that time it seemed everybody was happy to get the
abstraction layer TARGET_*. And of course I'm not happy about it, 
because there is a discussion to remove it completely without imho 
understanding the idea behind the layer. After exchanging some private 
emails I like to post some of my thoughts in public.

1. Advantages and disadvantages

Advantages of TARGET_*:
- efficient code (independent of compiler)
- easy to port (just override the macros which are some
   special case for the target compared to the generic
   implementation)
- macros are usually small (only 3-5 lines)
- functions can also used (e. g. for complex native code)
- there is no "dead-code"
- no need for extensive ifdef-elif-else-endif constructs
- target dependent implementation is located in a single file
   (header-file with macros)

Disadvantages of TARGET_*
- debugging is more complicated
- not type-save

Advantages of target-layer functions like do_*()
- debugging is easier
- type-safe like other C-code

Disadvantages of target-layer functions do_*()
- less efficient (additional function-call)
- "dead-code" if some function in a object-file is not used
   (problem with linker; see comments below)
- code cluttered with ifdef-elif-else-endif constructs
- running autoconf for embedded systems is difficult, thus
   many "hard-wired" predefines are needed
- target dependent implementation are located in several
   files (configure, header-file with predefines, c-file)

2. Naming and other "cosmetic" things

The naming convention of the TARGET_* is like the following:

TARGET_NATIVE_<module>_<function>

<module> stands for some group of functions, e. g. file-functions.
<function> stands for the name of a function, usually the name of the
corresponding OS-function from Linux extended by some suffix, e. g.
OPEN_READ which stands for open(...,RD_READONLY). There is (or at
least: there should be) no exception, thus some names could become
a little bit long. But there are no "arbitrary" abbreviations
which are difficult to understand.

For the math-macros the naming is also like this, e. g.

TARGET_NATIVE_MATH_FLOAT_DOUBLE_ISNAN

module: MATH_FLOAT (floating point macros)
function: DOUBLE_ISNAN (check if a double is NaN)

The prefix TARGET_NATIVE_ is always used to avoid naming conflicts with
existing names in the OS-includes. E. g. using OPEN_READ() only would be
dangerous, because OPEN_READ could be some already existing constant,
macro or function in the specific target OS.

OF COURSE.... this naming is _not_ fixed and I do _not_ assume it is the
best solution at all (but it is some solution; before that we had many
conflicts with different OSs, because of the so "convenient" short 
names). If needed this can be changed without to much confusion and pain 
(it would be some pain of course for aicas, but this would be acceptable).

Cosmetics are:
  - length of names for macros (usually I never typeing a macro name,
    instead I use cut+copy; emacs-users can use auto-completion,
    Eclipse-users also have some help be the IDE). By the way: the
    longest name is currently 55 characters long.
  - prefix TARGET_NATIVE: some other would also be good
  - length of lines

3. complexity of macros, debugging

It is true that #defines are difficult to debug. They are also difficult
to write, but of course it depends always on the specific macro. Usually
the macros are of the form like this

#define TARGET_NATIVE_<name>(...) \
   INCLUDE
   do { \
     FUNCTION
     RESULT
   } while (0)

e. g.:

#define 
TARGET_NATIVE_FILE_OPEN(filename,filedescriptor,flags,permissions,result) \
   #include <sys/types.h>
   #include <sys/stat.h>
   #include <fcntl.h>
   do { \
     filedescriptor=open(filename, \
                         flags, \
                         permissions \
                         ); \
     result=(filedescriptor>=0)?TARGET_NATIVE_OK:TARGET_NATIVE_ERROR; \
   } while (0)

The standard implementation (generic) contain 138 of these macros. Only 
9 are more complex, because of different possible implementations 
(selected by autoconf) or transformation of values. Thus in most of the 
cases the macros are only "wrappers" for some OS-specific call including 
adaption of parameters (e. g. types or units, result value).

Imho the complexity of the macros is usually not very high (if it 
becomes complex also a function can be implemented; then a macro is only 
an "alias"). They are multi-lined to make them more readable. The 
"include"-statements are needed in the generic implementation and the
"do...while" is a construct for safe usage of the macro.

Debugging of macros is difficult - if they are complex. If a macro is 
only a wrapper then only a OS-specific function is called with some 
additional calculations, e. g. evaluation of the return value. It is a 
good idea to keep the macros as simple as possible. And it is possible, 
because the TARGET_*-layer does not add additional functionally, it only 
"maps" some functionality.

3. autoconf, POSIX - porting

autoconf is a nice tool which is also used at aicas heavily (half of my
time I'm doing "autoconf"). But autoconf is also limited in its usage. 
For Unix-like systems autoconf is a good solution (for this autoconf is
written, I assume), but for non-Unix-like systems autoconf could be 
become a problem. We discussed at aicas around 1 year ago if we should 
use autoconf only, but we detected this is not really possible. 
Especially for embedded systems and "strange" systems (e. g. MinGW or 
embOS) it is a big challenge to "trim" autoconf in such a way that the 
right configuration is selected. I detected some complicacy of autoconf 
and the specific OS which makes it very difficult to use autoconf only. 
I will give you a few examples:

- for embeeded system it is not always feasible to check if a function
exists by compiling and linking a small example-program, because 
sometimes linkage is done only partially on the host. Final linkage is 
done on the target when loading the program or when creating the system 
image with the included application. Thus AC_CHECK_FUNC is not feasible. 
The same problem occurs for other checks, e. g. constants or datatypes.

- some systems have very strange and even wrong header-files. E. g. for
Windows/MinGW the headers sys/stat.h, io.h, windows.h, winbase.h are
needed for checking chsize() (truncate) or mkdir(). For embOS even some
header-files can not be included, because they are wrong (but cannot be
changed/fixed by aicas). These things make autoconf very complicate to
use, because if there are a lot of possible functions which can be used 
to implement some feature, it is not clear which function is detected by
autoconf for some specific system. There even could be very bad
side-effects if more than one function is available (e. g. f1() and 
f2()) and at some time f2() is used instead f1() (with different 
behaviors or limitations), if there is some change for another target 
system (e. g. you add some changes for RTEMS, but this will also have 
effects on e. g. embOS. You will not detect this problem until testing 
again all targets for any change in autoconf). It is a little bit 
"undeterministic" which features are detected and if they are usable.

- some features are not detectable by autoconf, e. g. the ordering of
parameters for functions like inb() and outb() (we had that problem) or
additional parameters (which usually only produce a warning which is
discarded), e. g. gethostbyname_r() under Solaris.

There are much more difficult things which occur with autoconf. To 
replace some target layer (e. g. TARGET_*) by autoconf only will imho 
make implementations for non-Unix-like systems very difficult and will 
only shift the so called "complex" C-macro-implementation into "complex"
autoconf-macros-implementations (imho M4 is not much better then a
C-preprocessor and difficult to debug).

4. Multiple code - some statistics:

In the current implementation we use at aicas we have the following
systems. The numbers below count the number of macros at all (functions
and constants) which are different from the standard (generic)
implementation:

generic macros: 220

Linux: 0
Solaris: 11
RTEMS: 3
MinGW: 46
embOS: 10 (only partially implemented)

There are 0 (Linux) upto 46 (MinGW) special cases of macros. Some have 
to be implemented because of different OS functions, some are 
implemented for efficiency (e. g. some OSes offer a POSIX-thread 
interface, but the native thread-interface is usually more efficient. It 
can be used with the macros-technique without any overhead). Thus for 
some targets 0..20% of the macros have to be reimplemented to cover 
special causes. For "Unix"-like systems this is only usually less then 5%.

Some additional comments:

Efficient code: wrapper-functions are nice, but in some cases an 
overkill, e. g. when calling simple function like sin(). In general C 
compilers do not optimize this (imho that is some reason why "inline" 
was introduced). If "inline" can be used, macros are almost not needed 
anymore.

autoconf: autoconf is a good idea and aicas is also heavily using it, 
but there are some limitations especially for embedded systems: because
autoconf can not run test-programs on embedded target-systems, some test
can not be done with autoconf (see above). I had to replace many
autoconf-test-functions by special versions which can be used for 
embedded systems. And still there are many things which are difficult to 
handle, e. g. which include files have to be included for some 
native-function. Some target systems make it really hard to use autoconf 
in the right way.

dead-code: the standard GNU linker does not remove functions which are 
not used (dead-code). Thus if at least one function is needed from a 
object file, all other function from that object file are linked to the
application, too. There is only one automatic way to remove dead-code 
(-ffunction-sections), but this have other disadvantages. Even the 
man-page do not recommend it. Thus to remove dead-code of a function 
some #ifdef-endif around a function is needed. On the other side: A 
non-used macro will not produce any dead code.

Some personal view:

By the way: I like "long" names and I hate uncommon abbreviations e. g.
like "fnctn" instead "function". I also like some prefix which indicate 
the location of, e. g. "file_open()" instead "open()". I usually have no 
problems with long names if the naming is consistent and useful. I also 
have no problems with lines longer then 80 characters, because my editor 
does not have some "optimal" line length.



These are my thoughts to this topic. I hope all developers who are 
interrested in some target native layer will reconsider the current 
discussion. And I hope we will find a solution which can satisfy everybody.

Sincerely,

Torsten




From MAILER-DAEMON Tue Aug 10 17:12:09 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1Buduj-00014X-Mv
	for mharc-classpath@gnu.org; Tue, 10 Aug 2004 17:12:09 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1Buduh-00011s-Rk
	for classpath@gnu.org; Tue, 10 Aug 2004 17:12:07 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1Budug-00010R-Rw
	for classpath@gnu.org; Tue, 10 Aug 2004 17:12:07 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1Budug-00010D-Nr
	for classpath@gnu.org; Tue, 10 Aug 2004 17:12:06 -0400
Received: from [82.161.94.186] (helo=gnu.wildebeest.org)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BudqW-0000ek-99
	for classpath@gnu.org; Tue, 10 Aug 2004 17:07:48 -0400
Received: from elsschot.wildebeest.org ([192.168.1.26])
	by gnu.wildebeest.org with esmtp (Exim 3.35 #1 (Debian))
	id 1Budpy-000102-00
	for <classpath@gnu.org>; Tue, 10 Aug 2004 23:07:14 +0200
From: Mark Wielaard <mark@klomp.org>
To: classpath@gnu.org
In-Reply-To: <4118B1C1.7000408@aicas.com>
References: <4118B1C1.7000408@aicas.com>
Content-Type: multipart/signed; micalg=pgp-sha1;
	protocol="application/pgp-signature";
	boundary="=-kzBYXHjFwfTGYCX2Fubi"
Message-Id: <1092172018.6388.158.camel@localhost>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.4.6 
Date: Tue, 10 Aug 2004 23:06:59 +0200
Subject: Re: Target native layer
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Tue, 10 Aug 2004 21:12:08 -0000


--=-kzBYXHjFwfTGYCX2Fubi
Content-Type: text/plain
Content-Transfer-Encoding: quoted-printable

Hi,

On Tue, 2004-08-10 at 13:30, Dr. Torsten Rupp wrote:
> I read many emails in the list about the TARGET_*-layer. I'm not very=20
> happy about the discussion, because we discussed this already around 1=20
> year ago and at that time it seemed everybody was happy to get the
> abstraction layer TARGET_*.

That is true, but since then we haven't (as GNU Classpath group) used
the abstraction layer at all to make our code more portable.

It was introduced by you in March 2003 and refined in July 2003. But
since then there hasn't been any real use/development of it. Michael
Koch made a couple of fixes late last year and during this year so that
it works on BSD systems and I made some some fixes this year to make
available() work on non-file streams (like system.in) and to remove
non-ansi C comments when the GNU Classpath code was converted to really
use ISO C 90 conventions only.

Both Michael and I take the libgcj code (that already works and is
ported to a few different systems) as inspiration for our changes. But
that code is written to use autoconf style checks. This is kind of hard
for us since the multi-line macro style used in the generic layer is
hard to read/write.

Since nobody ever commented on this and since except for all the empty
files in the target/Linux directory nobody ever made a real
implementation of the abstraction layer I always thought that we never
actually used it. That is why we have the discussion now. We have an
abstraction layer without a real implementation layer.

> 3. complexity of macros, debugging
>=20
> It is true that #defines are difficult to debug. They are also difficult
> to write, but of course it depends always on the specific macro.

Yes, this is my main complaint personally. If we can get at least rid of
the multi-line defines then I think it would be much easier to debug and
write native code.

> #define=20
> TARGET_NATIVE_FILE_OPEN(filename,filedescriptor,flags,permissions,result)=
 \
>    #include <sys/types.h>
>    #include <sys/stat.h>
>    #include <fcntl.h>
>    do { \
>      filedescriptor=3Dopen(filename, \
>                          flags, \
>                          permissions \
>                          ); \
>      result=3D(filedescriptor>=3D0)?TARGET_NATIVE_OK:TARGET_NATIVE_ERROR;=
 \
>    } while (0)

It is interesting that you quote this macro since this is actually one
which Michael Koch fixed so it now reads:

#ifndef TARGET_NATIVE_FILE_OPEN
  #include <sys/types.h>
  #include <sys/stat.h>
  #include <fcntl.h>
  #define TARGET_NATIVE_FILE_OPEN(filename,filedescriptor,flags,permissions=
,result) \
    do { \
      filedescriptor=3Dopen(filename, \
                          flags, \
                          permissions \
                          ); \
      if (filedescriptor >=3D 0) \
        fcntl (filedescriptor,F_SETFD,FD_CLOEXEC); \
      result=3D(filedescriptor>=3D0)?TARGET_NATIVE_OK:TARGET_NATIVE_ERROR; =
\
   } while (0)
#endif

Do you actually use the GNU Classpath version of the target/generic
implementation or do you use another implementation?

> Imho the complexity of the macros is usually not very high (if it=20
> becomes complex also a function can be implemented; then a macro is only=20
> an "alias"). They are multi-lined to make them more readable. The=20
> "include"-statements are needed in the generic implementation and the
> "do...while" is a construct for safe usage of the macro.
>
> Debugging of macros is difficult - if they are complex. If a macro is=20
> only a wrapper then only a OS-specific function is called with some=20
> additional calculations, e. g. evaluation of the return value. It is a=20
> good idea to keep the macros as simple as possible. And it is possible,=20
> because the TARGET_*-layer does not add additional functionally, it only=20
> "maps" some functionality.

One thing that worries me about a lot of these macro definitions is that
it seems they can trigger unexpected multiple evaluation of the
arguments. A couple of the macros their arguments multiple times this
might trip people who don't expect a function to be called multiple
times or who use such a macro with arguments such as (i++).
I guess you must be very careful when you use them.

> - for embeeded system it is not always feasible to check if a function
> exists by compiling and linking a small example-program, because=20
> sometimes linkage is done only partially on the host. Final linkage is=20
> done on the target when loading the program or when creating the system=20
> image with the included application. Thus AC_CHECK_FUNC is not feasible.=20
> The same problem occurs for other checks, e. g. constants or datatypes.

Kaffe seems to be able to handle that. You can give defaults for cases
where you need to cross-compile for example.

> - some systems have very strange and even wrong header-files. E. g. for
> Windows/MinGW the headers sys/stat.h, io.h, windows.h, winbase.h are
> needed for checking chsize() (truncate) or mkdir(). For embOS even some
> header-files can not be included, because they are wrong (but cannot be
> changed/fixed by aicas). These things make autoconf very complicate to
> use, because if there are a lot of possible functions which can be used=20
> to implement some feature, it is not clear which function is detected by
> autoconf for some specific system.

That is handled in libgcj by having a hybrid system. autoconf for normal
Posix like systems and an os-layer for systems that are completely
different.

> There even could be very bad
> side-effects if more than one function is available (e. g. f1() and=20
> f2()) and at some time f2() is used instead f1() (with different=20
> behaviors or limitations), if there is some change for another target=20
> system (e. g. you add some changes for RTEMS, but this will also have=20
> effects on e. g. embOS. You will not detect this problem until testing=20
> again all targets for any change in autoconf). It is a little bit=20
> "undeterministic" which features are detected and if they are usable.

But on the other hand that is autoconf its strength. You put in tests
for features of different systems. If a platform has those features you
get your port for free.

> - some features are not detectable by autoconf, e. g. the ordering of
> parameters for functions like inb() and outb() (we had that problem) or
> additional parameters (which usually only produce a warning which is
> discarded), e. g. gethostbyname_r() under Solaris.

Have you looked at The GNU Autoconf Macro Archive?
http://www.gnu.org/software/ac-archive/
It has a macro specifically for this gethostbyname_r() parameter
detection for example. And there is also the ac-archive on sf.
http://ac-archive.sourceforge.net/

> There are much more difficult things which occur with autoconf. To=20
> replace some target layer (e. g. TARGET_*) by autoconf only will imho=20
> make implementations for non-Unix-like systems very difficult and will=20
> only shift the so called "complex" C-macro-implementation into "complex"
> autoconf-macros-implementations (imho M4 is not much better then a
> C-preprocessor and difficult to debug).

But using autoconf has the advantage that "everybody" else is using it.
At least the two most important projects to GNU Classpath gcj and kaffe
are build around it. So solving something with autoconf helps everybody.
Having our own target porting layer will make it more difficult to work
together with others that do use autoconf like systems. And autoconf is
very well documented out own layer isn't really well documented and
actually has no examples how to use it properly.

> 4. Multiple code - some statistics:
>=20
> In the current implementation we use at aicas we have the following
> systems. The numbers below count the number of macros at all (functions
> and constants) which are different from the standard (generic)
> implementation:
>=20
> generic macros: 220
>=20
> Linux: 0
> Solaris: 11
> RTEMS: 3
> MinGW: 46
> embOS: 10 (only partially implemented)

I think this sums up pretty well why most people don't really see the
point of the target layer. We only have the "empty implementation" for
GNU/Linux systems.=20

It would be interesting to compare the MinGW implementation with the way
libgcj does it (mostly also through a os-layer).

Also you say "the current implementation we use at aicas". To really
consider the strength of the target-layer we must make it our
implementation (for GNU Classpath as group). Otherwise it is only this
abstraction that we don't really use.

> Efficient code: wrapper-functions are nice, but in some cases an=20
> overkill, e. g. when calling simple function like sin(). In general C=20
> compilers do not optimize this (imho that is some reason why "inline"=20
> was introduced). If "inline" can be used, macros are almost not needed=20
> anymore.

autoconf comes with AC_C_INLINE which handles this case.

> dead-code: the standard GNU linker does not remove functions which are=20
> not used (dead-code). Thus if at least one function is needed from a=20
> object file, all other function from that object file are linked to the
> application, too. There is only one automatic way to remove dead-code=20
> (-ffunction-sections), but this have other disadvantages. Even the=20
> man-page do not recommend it. Thus to remove dead-code of a function=20
> some #ifdef-endif around a function is needed. On the other side: A=20
> non-used macro will not produce any dead code.

If this matters wouldn't you just put these functions that might not be
used into their own source file so you have multiple object files?

> These are my thoughts to this topic. I hope all developers who are=20
> interrested in some target native layer will reconsider the current=20
> discussion. And I hope we will find a solution which can satisfy everybod=
y.

I am very interested. And although I clearly favor some autoconf like
solution (since I know how to copycat that from libgcj, kaffe or other
GNU projects) I also don't provide any code at the moment.

I think that has been the real problem in this discussion. Neither the
supporters of the TARGET_NATIVE layer nor the supporters of a autoconf
approach have actually shown the code to really do the ports. Having
real code to argue about would make a lot of the arguments a lot easier
to settle.

Cheers,

Mark

--=-kzBYXHjFwfTGYCX2Fubi
Content-Type: application/pgp-signature; name=signature.asc
Content-Description: This is a digitally signed message part

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)

iD8DBQBBGTjyxVhZCJWr9QwRAj6jAKCZo7qhZboaZTtuzJ5X+sIe1kaB1QCfVZOI
cHerGWYlg2CKI7y/fraD918=
=uxbZ
-----END PGP SIGNATURE-----

--=-kzBYXHjFwfTGYCX2Fubi--




From MAILER-DAEMON Tue Aug 10 18:59:08 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BufaG-00072O-4T
	for mharc-classpath@gnu.org; Tue, 10 Aug 2004 18:59:08 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BufaE-0006zs-Fi
	for classpath@gnu.org; Tue, 10 Aug 2004 18:59:06 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BufaC-0006y6-TI
	for classpath@gnu.org; Tue, 10 Aug 2004 18:59:06 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BufaC-0006xw-RU
	for classpath@gnu.org; Tue, 10 Aug 2004 18:59:04 -0400
Received: from [213.9.1.84] (helo=box84.elkhouse.de)
	by monty-python.gnu.org with esmtp (TLSv1:RC4-SHA:128) (Exim 4.34)
	id 1BufWI-0006Qj-JE
	for classpath@gnu.org; Tue, 10 Aug 2004 18:55:02 -0400
Received: from 1cust181.vr2.fft4.alter.net ([149.229.88.181]
	helo=[192.168.1.3]) by box84.elkhouse.de with asmtp (Exim 4.34)
	id 1BufXS-0003d1-M4; Wed, 11 Aug 2004 00:56:14 +0200
From: Roman Kennke <roman@ontographics.com>
To: Mark Wielaard <mark@klomp.org>
In-Reply-To: <1092172018.6388.158.camel@localhost>
References: <4118B1C1.7000408@aicas.com> <1092172018.6388.158.camel@localhost>
Content-Type: text/plain
Message-Id: <1092178428.72752.29.camel@moonlight>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.4.6 
Date: Wed, 11 Aug 2004 00:53:48 +0200
Content-Transfer-Encoding: 7bit
Cc: classpath@gnu.org
Subject: Re: Target native layer
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Tue, 10 Aug 2004 22:59:06 -0000

Hi all,

may I once more comment on that.

> Also you say "the current implementation we use at aicas". To really
> consider the strength of the target-layer we must make it our
> implementation (for GNU Classpath as group). Otherwise it is only this
> abstraction that we don't really use.

This is a real argument, IMO. I can see some ways to solve the problem:

1. We get rid of the TARGET_* layer (because we don't use it), and the
folks at AICAS could keep it on their own (much like some specialities
in kaffe, gcj and sablevm are handled). Of course, this would make
synchronizing their sources with the sources of GNU CP more difficult. 

2. AICAS could contribute all their TARGET_* layer using code to GNU CP.
This way this layer would kinde make sense to GNU CP hackers. Maybe we
all can help to get rid of it nevertheless, but keep CP portable on
those rare systems.

3. I dunno about the details, but we could go with a hybrid system: Most
of the portable stuff goes into the generic layer and is
autoconfiscated, like it was discussed to great lengths now. In really
extreme cases the TARGET_* layer could be used. BUT: We should think
about if we need all of the indirections in this layer and how we could
make it more readable and debuggable. This means:
- get rid of (multiline) macros where feasible,
- shorten macro names,
- group files together more intelligently and
- reduce (better: completely avoid) redundant code (bugs
multiplication!!).

If this can be fullfilled somehow, then I could go with the TARGET
layer.

just my 2 cents again,

/Roman





From MAILER-DAEMON Tue Aug 10 19:43:11 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BugGt-0007cs-RC
	for mharc-classpath@gnu.org; Tue, 10 Aug 2004 19:43:11 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BugGs-0007cn-8A
	for classpath@gnu.org; Tue, 10 Aug 2004 19:43:10 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BugGq-0007cb-LP
	for classpath@gnu.org; Tue, 10 Aug 2004 19:43:09 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BugGq-0007cY-Id
	for classpath@gnu.org; Tue, 10 Aug 2004 19:43:08 -0400
Received: from [66.33.197.214] (helo=ludo.dreamhost.com)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BugCz-0003jp-0E
	for classpath@gnu.org; Tue, 10 Aug 2004 19:39:09 -0400
Received: from dexter (c-24-6-46-122.client.comcast.net [24.6.46.122])
	by ludo.dreamhost.com (Postfix) with ESMTP id 8148143B71
	for <classpath@gnu.org>; Tue, 10 Aug 2004 16:39:08 -0700 (PDT)
Received: by dexter (Postfix, from userid 1000)
	id 871BF39; Tue, 10 Aug 2004 16:39:03 -0700 (PDT)
To: classpath@gnu.org
References: <200408031329.01616.konqueror@gmx.de>
	<41131B56.5080707@watson.ibm.com>
	<1091798611.7864.164.camel@localhost>
	<200408061817.03221.anwalt_pop@aicas.com>
	<41144C2B.6000402@watson.ibm.com>
From: Casey Marshall <csm@gnu.org>
Date: Tue, 10 Aug 2004 16:39:03 -0700
In-Reply-To: <41144C2B.6000402@watson.ibm.com> (Steven Augart's message of
	"Fri, 06 Aug 2004 23:27:39 -0400")
Message-ID: <87k6w67eco.fsf@gnu.org>
User-Agent: Gnus/5.1002 (Gnus v5.10.2) Emacs/21.2 (gnu/linux)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Subject: Re: Classpath License, Phillosophy, and AICAS
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Tue, 10 Aug 2004 23:43:10 -0000

Ok, so if I understand this thread, the question is whether or not to
drop a fairly clumsy, ugly, and hard-to-debug OS-dependent layer from
Classpath that is really only used by a non-free virtual machine.

Why is this a difficult decision? What exactly does your proprietary
VM do for us?

-- 
Casey Marshall || csm@gnu.org



From MAILER-DAEMON Wed Aug 11 00:55:14 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1Bul8s-0006oz-Mf
	for mharc-classpath@gnu.org; Wed, 11 Aug 2004 00:55:14 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1Bul8r-0006o8-WC
	for classpath@gnu.org; Wed, 11 Aug 2004 00:55:14 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1Bul8q-0006lR-6K
	for classpath@gnu.org; Wed, 11 Aug 2004 00:55:13 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1Bul8q-0006lK-3l
	for classpath@gnu.org; Wed, 11 Aug 2004 00:55:12 -0400
Received: from [212.6.122.28] (helo=mail4.ewetel.de)
	by monty-python.gnu.org with esmtp (Exim 4.34)
	id 1Bul4i-00018o-Vu; Wed, 11 Aug 2004 00:50:57 -0400
Received: from majestix.konqueror.de (dynadsl-080-228-94-126.ewetel.net
	[80.228.94.126])
	by mail4.ewetel.de (8.12.1/8.12.9) with ESMTP id i7B4odMv025127;
	Wed, 11 Aug 2004 06:50:40 +0200 (MEST)
Received: from miraculix.konqueror.de ([192.168.1.1])
	by majestix.konqueror.de with esmtp (Exim 3.36 #1 (Debian))
	id 1BulDL-0000NS-00; Wed, 11 Aug 2004 06:59:51 +0200
From: Michael Koch <konqueror@gmx.de>
To: classpath@gnu.org
Date: Wed, 11 Aug 2004 06:51:21 +0200
User-Agent: KMail/1.6.2
References: <200408031329.01616.konqueror@gmx.de>
	<41144C2B.6000402@watson.ibm.com> <87k6w67eco.fsf@gnu.org>
In-Reply-To: <87k6w67eco.fsf@gnu.org>
MIME-Version: 1.0
Content-Disposition: inline
Content-Type: text/plain;
  charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
Message-Id: <200408110651.22519.konqueror@gmx.de>
X-CheckCompat: OK
Cc: Casey Marshall <csm@gnu.org>
Subject: Re: Classpath License, Phillosophy, and AICAS
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Wed, 11 Aug 2004 04:55:14 -0000

Am Mittwoch, 11. August 2004 01:39 schrieb Casey Marshall:
> Ok, so if I understand this thread, the question is whether or not
> to drop a fairly clumsy, ugly, and hard-to-debug OS-dependent layer
> from Classpath that is really only used by a non-free virtual
> machine.
>
> Why is this a difficult decision? What exactly does your
> proprietary VM do for us?

Thats not true. Its currently used by jamvm, kissme and I think 
sablevm, probably more.


Michael



From MAILER-DAEMON Wed Aug 11 03:53:17 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BunvA-0004CO-TW
	for mharc-classpath@gnu.org; Wed, 11 Aug 2004 03:53:17 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1Bunv9-00048v-0s
	for classpath@gnu.org; Wed, 11 Aug 2004 03:53:15 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1Bunv7-00046C-8R
	for classpath@gnu.org; Wed, 11 Aug 2004 03:53:14 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1Bunv7-000464-5e
	for classpath@gnu.org; Wed, 11 Aug 2004 03:53:13 -0400
Received: from [217.72.192.224] (helo=smtp06.web.de)
	by monty-python.gnu.org with esmtp (TLSv1:DES-CBC3-SHA:168)
	(Exim 4.34) id 1Bunqn-00021e-Re
	for classpath@gnu.org; Wed, 11 Aug 2004 03:48:46 -0400
Received: from pd9e6173b.dip.t-dialin.net ([217.230.23.59] helo=web.de)
	by smtp06.web.de with asmtp (WEB.DE 4.101 #44) id 1Bunq9-0004zK-00
	for classpath@gnu.org; Wed, 11 Aug 2004 09:48:05 +0200
Message-ID: <4119CF35.3000702@web.de>
Date: Wed, 11 Aug 2004 09:48:05 +0200
From: Matthias Pfisterer <Matthias.Pfisterer@web.de>
User-Agent: Mozilla/5.0 (X11; U; Linux i586; en-US;
	rv:1.0.0) Gecko/20020623 Debian/1.0.0-0.woody.1
X-Accept-Language: en-us, en, de
MIME-Version: 1.0
To: GNU Classpath Project <classpath@gnu.org>
References: <87zn5973km.fsf@fleche.redhat.com>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Sender: Matthias.Pfisterer@web.de
X-Sender: Matthias.Pfisterer@web.de
Subject: Re: Branch for generics
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Wed, 11 Aug 2004 07:53:15 -0000

Hi,

I' happy to see movement in this area!

In spring, I started implementing generics. However, I'm now doing my 
master's thesis, so I'm busy and stalled development. I'll send you my 
development state in a private mail. Use at your will.

And a remark to the conditions: "a majority of VMs" could perhaps really 
mean "a majority of compilers". As far as I remember, the changes in the 
class format used for generics are backward-compatible. Indeed, I had no 
problems to run code using generics and compiled by the javac of the JDK 
1.5 with the VM of the JDK 1.4.

Matthias


Tom Tromey wrote:
> I'm going to make a branch for generics work, including all the other
> parts of 1.5; for the most part these rely on generics and can't be
> done independently anyway.  I plan to name it "generics-branch".
> 
> * Commit rules.
>   Only commit changes related to generics or other 1.5 work; other
>   changes should go on the trunk (so no sneaking random changes in
>   via the branch :-).
>   Post your patches as usual, but put `[generics]' in the Subject.
>   There is no need for approval for these patches.
> 
> * Lifetime.
>   I propose we merge the branch to mainline when two criteria are met:
> 
>   1. A majority of Classpath-using VM implementors agree that they
>   have access to a compiler suitable for compiling the branch.
> 
>   2. A majority of Classpath-using VMs can actually use this version
>   of Classpath, though perhaps not perfectly.  (In particular here
>   I'm thinking that it is probably ok if the new reflection code does
>   not work -- the VM will still be able to bootstrap, it just won't
>   implement all of 1.5 faithfully.)
> 
> * What's there.
>   So far I've written Iterable and converted a good chunk of the
>   existing collections classes.  None of this has been compiled at
>   all yet.  Feel free to help out, there is plenty to do :-)
> 
> 
> I'm personally planning to use this code to help debug the generics
> (and other 1.5 stuff) in gcjx; hopefully it can serve the same purpose
> for the generics branch of jikes.  (Probably a quicker route to a free
> 1.5-ish compiler will be via Eclipse ...)
> 
> Tom
> 
> 
> _______________________________________________
> Classpath mailing list
> Classpath@gnu.org
> http://lists.gnu.org/mailman/listinfo/classpath


-- 
Matthias Pfisterer	Matthias.Pfisterer@web.de
Reuchlinstrasse 28	phone ++49-711-62 87 12
D-70176 Stuttgart	(in Deutschland 0711-62 87 12)
GERMANY

Work like you don't need the money.
Love like you've never been hurt.
Dance like nobody is watching.

Java Sound Resources (examples, FAQ, applications):
http://www.jsresources.org/

Tritonus, the open source implementation of the Java Sound API:
http://www.tritonus.org/
--------------------------------------------------------------




From MAILER-DAEMON Wed Aug 11 06:43:02 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BuqZS-0001AS-6F
	for mharc-classpath@gnu.org; Wed, 11 Aug 2004 06:43:02 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BuqZQ-0001AA-8b
	for classpath@gnu.org; Wed, 11 Aug 2004 06:43:00 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BuqZO-00018X-CS
	for classpath@gnu.org; Wed, 11 Aug 2004 06:43:00 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BuqZO-00017H-7w
	for classpath@gnu.org; Wed, 11 Aug 2004 06:42:58 -0400
Received: from [212.227.126.171] (helo=moutng.kundenserver.de)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BuqUl-0001jL-MA
	for classpath@gnu.org; Wed, 11 Aug 2004 06:38:12 -0400
Received: from [212.227.126.207] (helo=mrelayng.kundenserver.de)
	by moutng.kundenserver.de with esmtp (Exim 3.35 #1)
	id 1BuqUF-0006tU-00; Wed, 11 Aug 2004 12:37:39 +0200
Received: from [212.126.219.82] (helo=aicas.com)
	by mrelayng.kundenserver.de with asmtp (TLSv1:RC4-MD5:128)
	(Exim 3.35 #1) id 1BuqUF-0007I9-00; Wed, 11 Aug 2004 12:37:39 +0200
Message-ID: <4119F6F3.5070306@aicas.com>
Date: Wed, 11 Aug 2004 12:37:39 +0200
From: "Dr. Torsten Rupp" <rupp@aicas.com>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.4) Gecko/20030624
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: Roman Kennke <roman@ontographics.com>,  classpath@gnu.org
References: <4118B1C1.7000408@aicas.com> <1092172018.6388.158.camel@localhost>
	<1092178428.72752.29.camel@moonlight>
In-Reply-To: <1092178428.72752.29.camel@moonlight>
X-Enigmail-Version: 0.76.5.0
X-Enigmail-Supports: pgp-inline, pgp-mime
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
X-Provags-ID: kundenserver.de abuse@kundenserver.de
	auth:f9c39743fddc2557f7fa7c1a513f7ea5
Cc: 
Subject: Re: Target native layer
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Wed, 11 Aug 2004 10:43:00 -0000

Dear Roman,

> 1. We get rid of the TARGET_* layer (because we don't use it), and the
> folks at AICAS could keep it on their own (much like some specialities
> in kaffe, gcj and sablevm are handled). Of course, this would make
> synchronizing their sources with the sources of GNU CP more difficult. 

I assume if TARGET_* is removed, aicas have to do their own development 
of the native code. This would be possible, but of course this would not 
be very useful. For aicas it would be difficult, because we are almost 
"kicked-out" of the project (at least for the native code).

> 2. AICAS could contribute all their TARGET_* layer using code to GNU CP.
> This way this layer would kinde make sense to GNU CP hackers. Maybe we
> all can help to get rid of it nevertheless, but keep CP portable on
> those rare systems.

You are right, we are late with contributing more layers. Because of 
limited manpower we still did not contributed some more layers, e.g . 
Solaris. We like to do this.

> 3. I dunno about the details, but we could go with a hybrid system: Most
> of the portable stuff goes into the generic layer and is
> autoconfiscated, like it was discussed to great lengths now.

The generic target is already doing this (at least partially). It is 
designed to use autoconf results to select the right OS-function or the 
right constants. Thus most of the code is located there. And that is 
also the reason why Linux seems to be "empty". It is not empty, it 
includes the generic cases of the target layer, thus there is not 
special implementation in Linux (please have a look at the final 
"include"-statement at the end of the files). If needed we can rename 
"generic" into "Linux" and discard the current "Linux". I do not want to 
do this, because this would look like the layer is limited to 
"Linux-like" systems.

> In really
> extreme cases the TARGET_* layer could be used.

The extrem cases are the problem. I spend a lot of time to handle such 
cases and it is difficult to do this with autoconf only. If you only use 
Unix-like systems, probably you will not run into this problem, thus 
autoconf only is nice. But imho soon you like to use other systems, too, 
e. g. Windows or embedded systems it becomes a challange. I can imagine 
it is difficult for someone to understand my viewpoint, if that person 
is not developing software for embedded systems.

> - get rid of (multiline) macros where feasible,

Of course it would be possible to remove the "linefeeds", but that would 
not help (imho it would make it even more difficult to read).

I have the following suggestion:

Because the multiline-macros are to difficult to handle, we can replace 
that macros e. g. by direct function-calls. Thus a macro would become 
some "alias" only. We also can change (shorten) the names. For this we 
have to change the generic implementation. The advantages would be:

- debugging is possible
- type-safety
- easier to read

and for aicas

- we can keep most of the code we already have written
- we can continue to participate in Classpath

For aicas this would be some compromise, because we can still use the 
target-layer. What would be still important is, that in the native _no_ 
  OS-specific feature is used directly, like constants or datatypes or 
OS-functions calls. With such a implementation we can survive.

What do you think?

> - shorten macro names,

Please make some suggestion. If TARGET_NATIVE is to long, what about

TARGET or
NATIVE or
NC or
...?

Please take care that there is some risk for naming-collisions if the
name becomes to short (thus "OS" would probably be a bad choice).

> - group files together more intelligently and

I do not understand what is wrong with the current grouping. The 
grouping is done by function-types, e. g. file, network, memory and so 
on. How to make it better/more clear?

> - reduce (better: completely avoid) redundant code (bugs
> multiplication!!).

Ok. There are some parts of the layer which could be improved. If we 
decide to keep the target layer in the code, then we should do these 
improvements.

Sincerely,

Torsten




From MAILER-DAEMON Wed Aug 11 07:14:45 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1Bur48-0001RQ-TG
	for mharc-classpath@gnu.org; Wed, 11 Aug 2004 07:14:44 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1Bur47-0001R2-5W
	for classpath@gnu.org; Wed, 11 Aug 2004 07:14:43 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1Bur45-0001Qa-0G
	for classpath@gnu.org; Wed, 11 Aug 2004 07:14:42 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1Bur44-0001QJ-SJ
	for classpath@gnu.org; Wed, 11 Aug 2004 07:14:40 -0400
Received: from [199.232.41.8] (helo=mx20.gnu.org)
	by monty-python.gnu.org with esmtp (TLSv1:DES-CBC3-SHA:168)
	(Exim 4.34) id 1Buqzz-0006st-PG
	for classpath@gnu.org; Wed, 11 Aug 2004 07:10:28 -0400
Received: from [212.6.122.12] (helo=mail1.ewetel.de)
	by mx20.gnu.org with esmtp (Exim 4.34) id 1BuqqR-00013j-6K
	for classpath@gnu.org; Wed, 11 Aug 2004 07:00:36 -0400
Received: from majestix.konqueror.de (dynadsl-080-228-94-126.ewetel.net
	[80.228.94.126])
	by mail1.ewetel.de (8.12.1/8.12.9) with ESMTP id i7BAxwdf021536;
	Wed, 11 Aug 2004 12:59:58 +0200 (MEST)
Received: from miraculix.konqueror.de ([192.168.1.1])
	by majestix.konqueror.de with esmtp (Exim 3.36 #1 (Debian))
	id 1Buqyl-00035z-00; Wed, 11 Aug 2004 13:09:11 +0200
From: Michael Koch <konqueror@gmx.de>
To: classpath@gnu.org
Date: Wed, 11 Aug 2004 13:00:44 +0200
User-Agent: KMail/1.6.2
References: <4118B1C1.7000408@aicas.com> <1092178428.72752.29.camel@moonlight>
	<4119F6F3.5070306@aicas.com>
In-Reply-To: <4119F6F3.5070306@aicas.com>
MIME-Version: 1.0
Content-Disposition: inline
Content-Type: text/plain;
  charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
Message-Id: <200408111300.44745.konqueror@gmx.de>
X-CheckCompat: OK
Cc: "Dr. Torsten Rupp" <rupp@aicas.com>
Subject: Re: Target native layer
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Wed, 11 Aug 2004 11:14:43 -0000

Am Mittwoch, 11. August 2004 12:37 schrieb Dr. Torsten Rupp:
> Dear Roman,
>
> > 1. We get rid of the TARGET_* layer (because we don't use it),
> > and the folks at AICAS could keep it on their own (much like some
> > specialities in kaffe, gcj and sablevm are handled). Of course,
> > this would make synchronizing their sources with the sources of
> > GNU CP more difficult.
>
> I assume if TARGET_* is removed, aicas have to do their own
> development of the native code. This would be possible, but of
> course this would not be very useful. For aicas it would be
> difficult, because we are almost "kicked-out" of the project (at
> least for the native code).

I don't wanna kick somebody out of the project. I just want to improve 
our Free Software project. I hope you dont got me wrong from the 
start.

> > 2. AICAS could contribute all their TARGET_* layer using code to
> > GNU CP. This way this layer would kinde make sense to GNU CP
> > hackers. Maybe we all can help to get rid of it nevertheless, but
> > keep CP portable on those rare systems.
>
> You are right, we are late with contributing more layers. Because
> of limited manpower we still did not contributed some more layers,
> e.g . Solaris. We like to do this.
>
> > 3. I dunno about the details, but we could go with a hybrid
> > system: Most of the portable stuff goes into the generic layer
> > and is autoconfiscated, like it was discussed to great lengths
> > now.
>
> The generic target is already doing this (at least partially). It
> is designed to use autoconf results to select the right OS-function
> or the right constants. Thus most of the code is located there. And
> that is also the reason why Linux seems to be "empty". It is not
> empty, it includes the generic cases of the target layer, thus
> there is not special implementation in Linux (please have a look at
> the final "include"-statement at the end of the files). If needed
> we can rename "generic" into "Linux" and discard the current
> "Linux". I do not want to do this, because this would look like the
> layer is limited to "Linux-like" systems.

Wouldnt it be better to call it "posix" (note: no captial letter) as 
it currently supports Linux, *BSD, AIX ... We can and should add 
Solaris too. With autoconf its really simple. Then AICAS can remove 
their own Solaris port totally and we have a clean non-redundant 
codebase for posix-like OSes.

> > In really
> > extreme cases the TARGET_* layer could be used.
>
> The extrem cases are the problem. I spend a lot of time to handle
> such cases and it is difficult to do this with autoconf only. If
> you only use Unix-like systems, probably you will not run into this
> problem, thus autoconf only is nice. But imho soon you like to use
> other systems, too, e. g. Windows or embedded systems it becomes a
> challange. I can imagine it is difficult for someone to understand
> my viewpoint, if that person is not developing software for
> embedded systems.
>
> > - get rid of (multiline) macros where feasible,
>
> Of course it would be possible to remove the "linefeeds", but that
> would not help (imho it would make it even more difficult to read).
>
> I have the following suggestion:
>
> Because the multiline-macros are to difficult to handle, we can
> replace that macros e. g. by direct function-calls. Thus a macro
> would become some "alias" only. We also can change (shorten) the
> names. For this we have to change the generic implementation. The
> advantages would be:
>
> - debugging is possible
> - type-safety
> - easier to read
>
> and for aicas
>
> - we can keep most of the code we already have written
> - we can continue to participate in Classpath
>
> For aicas this would be some compromise, because we can still use
> the target-layer. What would be still important is, that in the
> native _no_ OS-specific feature is used directly, like constants or
> datatypes or OS-functions calls. With such a implementation we can
> survive.

I like this idea. In fact I had the same in mind. 

> What do you think?
>
> > - shorten macro names,
>
> Please make some suggestion. If TARGET_NATIVE is to long, what
> about
>
> TARGET or
> NATIVE or
> NC or
> ...?

Perhaps a "CP" prefix (for ClassPath).

> Please take care that there is some risk for naming-collisions if
> the name becomes to short (thus "OS" would probably be a bad
> choice).
>
> > - group files together more intelligently and
>
> I do not understand what is wrong with the current grouping. The
> grouping is done by function-types, e. g. file, network, memory and
> so on. How to make it better/more clear?

There are several ways to deal with the files. Either do it the AICAS 
way and group one OS in one dir or do it the libgcj way and put all 
into one dir, havean OS suffix/prefix in the filename and choose the 
files are configure time.

> > - reduce (better: completely avoid) redundant code (bugs
> > multiplication!!).
>
> Ok. There are some parts of the layer which could be improved. If
> we decide to keep the target layer in the code, then we should do
> these improvements.

Yeah, we should really improve this. This needs much discussions and 
some time but will (hopefully) improve classpath' code.

I'm really glad that we now on the way to tackle the issues with the 
TARGE_NATIVE layer.


Michael



From MAILER-DAEMON Wed Aug 11 13:33:36 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1Buwym-0000u3-Do
	for mharc-classpath@gnu.org; Wed, 11 Aug 2004 13:33:36 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1Buwyk-0000tp-EW
	for classpath@gnu.org; Wed, 11 Aug 2004 13:33:34 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1Buwyi-0000tW-Ks
	for classpath@gnu.org; Wed, 11 Aug 2004 13:33:34 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1Buwyi-0000tT-Hw
	for classpath@gnu.org; Wed, 11 Aug 2004 13:33:32 -0400
Received: from [66.33.193.59] (helo=sara.dreamhost.com)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1Buwup-00032Q-PK
	for classpath@gnu.org; Wed, 11 Aug 2004 13:29:31 -0400
Received: from dexter (c-24-6-46-122.client.comcast.net [24.6.46.122])
	by sara.dreamhost.com (Postfix) with ESMTP
	id D4D969F70F; Wed, 11 Aug 2004 10:29:30 -0700 (PDT)
Received: by dexter (Postfix, from userid 1000)
	id 84ADB39; Wed, 11 Aug 2004 10:29:23 -0700 (PDT)
To: Michael Koch <konqueror@gmx.de>
References: <200408031329.01616.konqueror@gmx.de>
	<41144C2B.6000402@watson.ibm.com> <87k6w67eco.fsf@gnu.org>
	<200408110651.22519.konqueror@gmx.de>
From: Casey Marshall <csm@gnu.org>
Date: Wed, 11 Aug 2004 10:29:23 -0700
In-Reply-To: <200408110651.22519.konqueror@gmx.de> (Michael Koch's message
	of "Wed, 11 Aug 2004 06:51:21 +0200")
Message-ID: <87ekmd1t3g.fsf@gnu.org>
User-Agent: Gnus/5.1002 (Gnus v5.10.2) Emacs/21.2 (gnu/linux)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Cc: classpath@gnu.org
Subject: Re: Classpath License, Phillosophy, and AICAS
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Wed, 11 Aug 2004 17:33:34 -0000

>>>>> "Michael" == Michael Koch <konqueror@gmx.de> writes:

Michael> Am Mittwoch, 11. August 2004 01:39 schrieb Casey Marshall:
>> Ok, so if I understand this thread, the question is whether or not
>> to drop a fairly clumsy, ugly, and hard-to-debug OS-dependent layer
>> from Classpath that is really only used by a non-free virtual
>> machine.
>> 
>> Why is this a difficult decision? What exactly does your
>> proprietary VM do for us?

Michael> Thats not true. Its currently used by jamvm, kissme and I think 
Michael> sablevm, probably more.

Ok, I didn't know that. From the thread it didn't seem like it.

I assume you mean it's used by them because it is in Classpath, not
because they needed it. But you seem to be trying to make it sound
more useful than it is; why?

-- 
Casey Marshall || csm@gnu.org



From MAILER-DAEMON Wed Aug 11 13:42:03 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1Bux6x-0002qn-8Z
	for mharc-classpath@gnu.org; Wed, 11 Aug 2004 13:42:03 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1Bux6u-0002qX-6m
	for classpath@gnu.org; Wed, 11 Aug 2004 13:42:01 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1Bux6t-0002qH-EW
	for classpath@gnu.org; Wed, 11 Aug 2004 13:41:59 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1Bux6t-0002q7-Av
	for classpath@gnu.org; Wed, 11 Aug 2004 13:41:59 -0400
Received: from [212.6.122.12] (helo=mail1.ewetel.de)
	by monty-python.gnu.org with esmtp (Exim 4.34)
	id 1Bux1J-0003yd-Qo; Wed, 11 Aug 2004 13:36:14 -0400
Received: from majestix.konqueror.de (dynadsl-080-228-94-126.ewetel.net
	[80.228.94.126])
	by mail1.ewetel.de (8.12.1/8.12.9) with ESMTP id i7BHaBdf019439;
	Wed, 11 Aug 2004 19:36:12 +0200 (MEST)
Received: from miraculix.konqueror.de ([192.168.1.1])
	by majestix.konqueror.de with esmtp (Exim 3.36 #1 (Debian))
	id 1BuxAE-0003td-00; Wed, 11 Aug 2004 19:45:26 +0200
From: Michael Koch <konqueror@gmx.de>
To: classpath@gnu.org
Date: Wed, 11 Aug 2004 19:36:58 +0200
User-Agent: KMail/1.6.2
References: <200408031329.01616.konqueror@gmx.de>
	<200408110651.22519.konqueror@gmx.de> <87ekmd1t3g.fsf@gnu.org>
In-Reply-To: <87ekmd1t3g.fsf@gnu.org>
MIME-Version: 1.0
Content-Disposition: inline
Content-Type: text/plain;
  charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
Message-Id: <200408111936.58199.konqueror@gmx.de>
X-CheckCompat: OK
Cc: Casey Marshall <csm@gnu.org>
Subject: Re: Classpath License, Phillosophy, and AICAS
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Wed, 11 Aug 2004 17:42:01 -0000

Am Mittwoch, 11. August 2004 19:29 schrieb Casey Marshall:
> >>>>> "Michael" == Michael Koch <konqueror@gmx.de> writes:
>
> Michael> Am Mittwoch, 11. August 2004 01:39 schrieb Casey Marshall:
> >> Ok, so if I understand this thread, the question is whether or
> >> not to drop a fairly clumsy, ugly, and hard-to-debug
> >> OS-dependent layer from Classpath that is really only used by a
> >> non-free virtual machine.
> >>
> >> Why is this a difficult decision? What exactly does your
> >> proprietary VM do for us?
>
> Michael> Thats not true. Its currently used by jamvm, kissme and I
> think Michael> sablevm, probably more.
>
> Ok, I didn't know that. From the thread it didn't seem like it.
>
> I assume you mean it's used by them because it is in Classpath, not
> because they needed it. But you seem to be trying to make it sound
> more useful than it is; why?

Well its used by them because it is currently (?) the native 
implementation for some stuff in classpath. I dont want to make it 
more/less useful then it is. I just wanna have maintainable native 
code.


Michael



From MAILER-DAEMON Wed Aug 11 18:22:40 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1Bv1UV-00071n-Vn
	for mharc-classpath@gnu.org; Wed, 11 Aug 2004 18:22:40 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1Bv1UU-00071M-Ly
	for classpath@gnu.org; Wed, 11 Aug 2004 18:22:38 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1Bv1US-00070C-Hz
	for classpath@gnu.org; Wed, 11 Aug 2004 18:22:38 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1Bv1US-000708-ES
	for classpath@gnu.org; Wed, 11 Aug 2004 18:22:36 -0400
Received: from [203.12.22.20] (helo=power.connexus.net.au)
	by monty-python.gnu.org with esmtp (TLSv1:DES-CBC3-SHA:168)
	(Exim 4.34) id 1Bv1QT-00073W-AE
	for classpath@gnu.org; Wed, 11 Aug 2004 18:18:29 -0400
Received: from dholmeslaptop (dholmes.dltech.com.au [203.222.72.2])
	by power.connexus.net.au (8.12.4/8.11.6) with SMTP id i7BMGHJD033166
	for <classpath@gnu.org>; Thu, 12 Aug 2004 08:16:18 +1000 (EST)
	(envelope-from dholmes@dltech.com.au)
XAntiVirus: This e-mail has been scanned for viruses via the Connexus Internet
	Service
From: "David Holmes" <dholmes@dltech.com.au>
To: "classpath" <classpath@gnu.org>
Date: Thu, 12 Aug 2004 08:16:42 +1000
Message-ID: <NFBBKALFDCPFIDBNKAPCIEAIEIAA.dholmes@dltech.com.au>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: 7bit
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook IMO, Build 9.0.6604 (9.0.2911.0)
In-Reply-To: <4119CF35.3000702@web.de>
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1441
Importance: Normal
Subject: RE: Branch for generics
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Wed, 11 Aug 2004 22:22:39 -0000

Matthias writes:
> And a remark to the conditions: "a majority of VMs" could perhaps really
> mean "a majority of compilers". As far as I remember, the changes in the
> class format used for generics are backward-compatible. Indeed, I had no
> problems to run code using generics and compiled by the javac of the JDK
> 1.5 with the VM of the JDK 1.4.

The may be just the luck of a shared code base. The class file format
version number is increased once again for 1.5 (to 49.0) and that means pre
1.5 VMs need not support it. The official definition from Sun is that:

"The Java virtual machine implementation of Sun's JDK release 1.0.2 supports
class file format versions 45.0 through 45.3 inclusive. Sun's JDK releases
1.1.X can support class file formats of versions in the range 45.0 through
45.65535 inclusive. For  k >= 2 implementations of version 1.k of the Java 2
platform can support class file formats of versions in the range 45.0
through 44+k.0 inclusive."

So Sun's JDK 1.4 VM is not required to support 1.5 class files. Of course it
is up to the individual non-Sun VM's whether they will be able to load such
class files and simply ignore the extra information that is now present.
Remember that more than generics has changed the class file format.

David Holmes




From MAILER-DAEMON Wed Aug 11 19:48:24 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1Bv2pU-00013r-8U
	for mharc-classpath@gnu.org; Wed, 11 Aug 2004 19:48:24 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1Bv2pS-00011l-Mu
	for classpath@gnu.org; Wed, 11 Aug 2004 19:48:22 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1Bv2pS-000110-3E
	for classpath@gnu.org; Wed, 11 Aug 2004 19:48:22 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1Bv2pR-00010Z-TM
	for classpath@gnu.org; Wed, 11 Aug 2004 19:48:21 -0400
Received: from [207.189.223.52] (helo=email3.peakpeak.com)
	by monty-python.gnu.org with esmtp (TLSv1:DES-CBC3-SHA:168)
	(Exim 4.34) id 1Bv2lM-00028i-O1
	for classpath@gnu.org; Wed, 11 Aug 2004 19:44:09 -0400
Received: (qmail 31602 invoked by uid 513); 11 Aug 2004 23:54:44 -0000
Received: from tromey@redhat.com by email3 by uid 504 with qmail-scanner-1.20 
	(clamscan: 0.65.  Clear:RC:1(204.144.239.10):. 
	Processed in 0.152695 secs); 11 Aug 2004 23:54:44 -0000
Received: from tf0010.peakpeak.com (HELO fleche.redhat.com) ([204.144.239.10])
	(envelope-sender <tromey@redhat.com>)
	by email3.peakpeak.com (qmail-ldap-1.03) with SMTP
	for <classpath@gnu.org>; 11 Aug 2004 23:54:44 -0000
Received: by fleche.redhat.com (Postfix, from userid 1000)
	id CAD374F8A1C; Wed, 11 Aug 2004 17:26:49 -0600 (MDT)
To: Matthias Pfisterer <Matthias.Pfisterer@web.de>
References: <87zn5973km.fsf@fleche.redhat.com> <4119CF35.3000702@web.de>
From: Tom Tromey <tromey@redhat.com>
X-Attribution: Tom
Date: 11 Aug 2004 17:26:49 -0600
In-Reply-To: <4119CF35.3000702@web.de>
Message-ID: <874qn9s1c6.fsf@fleche.redhat.com>
Lines: 11
User-Agent: Gnus/5.09 (Gnus v5.9.0) Emacs/21.3.50
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Cc: GNU Classpath Project <classpath@gnu.org>
Subject: Re: Branch for generics
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
Reply-To: tromey@redhat.com
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Wed, 11 Aug 2004 23:48:22 -0000

>>>>> "Matthias" == Matthias Pfisterer <Matthias.Pfisterer@web.de> writes:

Matthias> In spring, I started implementing generics. However, I'm now
Matthias> doing my master's thesis, so I'm busy and stalled
Matthias> development. I'll send you my development state in a private
Matthias> mail. Use at your will.

Just for everybody's information, I'm (slowly) going through this,
comparing it against the current generics branch.

Tom



From MAILER-DAEMON Wed Aug 11 19:48:26 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1Bv2pW-00017O-N2
	for mharc-classpath@gnu.org; Wed, 11 Aug 2004 19:48:26 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1Bv2pT-00012m-Hw
	for classpath@gnu.org; Wed, 11 Aug 2004 19:48:23 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1Bv2pR-00010l-Rr
	for classpath@gnu.org; Wed, 11 Aug 2004 19:48:23 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1Bv2pR-00010Z-OM
	for classpath@gnu.org; Wed, 11 Aug 2004 19:48:21 -0400
Received: from [207.189.223.52] (helo=email3.peakpeak.com)
	by monty-python.gnu.org with esmtp (TLSv1:DES-CBC3-SHA:168)
	(Exim 4.34) id 1Bv2lI-00028O-HA
	for classpath@gnu.org; Wed, 11 Aug 2004 19:44:04 -0400
Received: (qmail 30514 invoked by uid 513); 11 Aug 2004 23:54:08 -0000
Received: from tromey@redhat.com by email3 by uid 504 with qmail-scanner-1.20 
	(clamscan: 0.65.  Clear:RC:1(204.144.239.10):. 
	Processed in 1.927523 secs); 11 Aug 2004 23:54:08 -0000
Received: from tf0010.peakpeak.com (HELO fleche.redhat.com) ([204.144.239.10])
	(envelope-sender <tromey@redhat.com>)
	by email3.peakpeak.com (qmail-ldap-1.03) with SMTP
	for <dholmes@dltech.com.au>; 11 Aug 2004 23:54:04 -0000
Received: by fleche.redhat.com (Postfix, from userid 1000)
	id A193E4F8A1C; Wed, 11 Aug 2004 17:26:09 -0600 (MDT)
To: "David Holmes" <dholmes@dltech.com.au>
References: <NFBBKALFDCPFIDBNKAPCIEAIEIAA.dholmes@dltech.com.au>
From: Tom Tromey <tromey@redhat.com>
X-Attribution: Tom
Date: 11 Aug 2004 17:26:09 -0600
In-Reply-To: <NFBBKALFDCPFIDBNKAPCIEAIEIAA.dholmes@dltech.com.au>
Message-ID: <878ycls1da.fsf@fleche.redhat.com>
Lines: 25
User-Agent: Gnus/5.09 (Gnus v5.9.0) Emacs/21.3.50
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Cc: classpath <classpath@gnu.org>
Subject: Re: Branch for generics
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
Reply-To: tromey@redhat.com
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Wed, 11 Aug 2004 23:48:23 -0000

>>>>> "David" == David Holmes <dholmes@dltech.com.au> writes:

David> So Sun's JDK 1.4 VM is not required to support 1.5 class
David> files. Of course it is up to the individual non-Sun VM's
David> whether they will be able to load such class files and simply
David> ignore the extra information that is now present.  Remember
David> that more than generics has changed the class file format.

Yeah.  As I understand it, Sun decided that their 1.5 compiler would
not be able to handle "-source 1.5 -target 1.4", since that would mean
"subsetting the language" (since, e.g., enum and foreach depend on
classes not in 1.4).  And in particular, I think 1.5 extends the ldc
opcode to allow constant classes (a very nice addition if you ask me)
-- so VMs would need at least one change.

Ultimately, of course, it is us here on this list implementing
Classpath.  I'm sure when the time comes we can competently evaluate
the merge according to our collective needs.  Probably by that time
we'll have a good idea of the minimal changes required to actually run
the resulting Classpath... I do suspect that full handling of all the
new 1.5 features (there are class file extensions for generic
signatures, enums, varargs, and attributes, at least) won't be
required.

Tom



From MAILER-DAEMON Thu Aug 12 06:21:19 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BvChz-0000ad-2q
	for mharc-classpath@gnu.org; Thu, 12 Aug 2004 06:21:19 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BvChw-0000aA-Kz
	for classpath@gnu.org; Thu, 12 Aug 2004 06:21:16 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BvChv-0000Zr-Rp
	for classpath@gnu.org; Thu, 12 Aug 2004 06:21:16 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BvChv-0000Zo-MS
	for classpath@gnu.org; Thu, 12 Aug 2004 06:21:15 -0400
Received: from [212.227.126.177] (helo=moutng.kundenserver.de)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BvCdS-0001cF-Fk
	for classpath@gnu.org; Thu, 12 Aug 2004 06:16:38 -0400
Received: from [212.227.126.205] (helo=mrelayng.kundenserver.de)
	by moutng.kundenserver.de with esmtp (Exim 3.35 #1)
	id 1BvCdM-0001B8-00
	for classpath@gnu.org; Thu, 12 Aug 2004 12:16:32 +0200
Received: from [212.126.219.82] (helo=aicas.com)
	by mrelayng.kundenserver.de with asmtp (TLSv1:RC4-MD5:128)
	(Exim 3.35 #1) id 1BvCdM-0005dL-00; Thu, 12 Aug 2004 12:16:32 +0200
Message-ID: <411B4380.2010006@aicas.com>
Date: Thu, 12 Aug 2004 12:16:32 +0200
From: "Dr. Torsten Rupp" <rupp@aicas.com>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.4) Gecko/20030624
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: classpath@gnu.org
References: <4118B1C1.7000408@aicas.com>
	<1092178428.72752.29.camel@moonlight>	<4119F6F3.5070306@aicas.com>
	<200408111300.44745.konqueror@gmx.de>
In-Reply-To: <200408111300.44745.konqueror@gmx.de>
X-Enigmail-Version: 0.76.5.0
X-Enigmail-Supports: pgp-inline, pgp-mime
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
X-Provags-ID: kundenserver.de abuse@kundenserver.de
	auth:f9c39743fddc2557f7fa7c1a513f7ea5
Subject: Re: Target native layer
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Thu, 12 Aug 2004 10:21:17 -0000

Dear Michael,

 > Wouldnt it be better to call it "posix" (note: no captial letter) as 
it currently supports Linux, *BSD, AIX ... We can and should add Solaris 
too. With autoconf its really simple. Then AICAS can remove their own 
Solaris port totally and we have a clean non-redundant codebase for 
posix-like OSes.


It is also not redundant code now, because the special implementation 
also have to be used in a single-directory code-basis (currently there 
are separated files which are a "design"-feature; the existence of the 
files could be called some redundancy).

 >> I have the following suggestion:
 >>
 >> Because the multiline-macros are to difficult to handle, we can
 >> replace that macros e. g. by direct function-calls. Thus a macro
 >> would become some "alias" only. We also can change (shorten) the
 >> names. For this we have to change the generic implementation. The

...

 > I like this idea. In fact I had the same in mind.


I'm very happy to hear this.

 >>> - shorten macro names,

...

 > Perhaps a "CP" prefix (for ClassPath).


I would prefer some prefix which indicates the "class" of the feature 
instead, e. g. "NC" (native code) or "TN" (target native), not some 
relation to the project name (everthing in Classpath is "Classpath", 
thus should probably have the prefix "CP"). But if "CP" would make you 
and other members happy, it is ok for me, too.

 > There are several ways to deal with the files. Either do it the AICAS 
way and group one OS in one dir or do it the libgcj way and put all into 
one dir, havean OS suffix/prefix in the filename and choose the files 
are configure time.


I would like to keep the files in separated directories to keep it more 
clear what files belong to which OS. Usually files are stored in a 
hierarchical order by using sub-directories. For the "hierachical" 
include-tree in the target-layer this was also a good idea to avoid 
confusion in names of files with the same function-group (e. g. 
"network"), but for different systems (different implementations).

 > I'm really glad that we now on the way to tackle the issues with the 
TARGE_NATIVE layer.


I'm sorry for this hard discussion. But I'm happy we are one a way to 
keep going on together. If everybody becomes satisfied with the 
following new idea for a target layer, I will prepare the changes.

New target native layer:

- transform code inside macros into functions (this will make debugging 
possible)
- the current macros become an "alias" to the functions (this is 
important to be able to replace a function by something special if needed)
- combine posix-like systems - Linux, BSD, AIX, Solaris - into a single 
set of files in one directory (usage of autoconf for posix-like systems)
- add some more layers from aicas

Sincerely,

Torsten




From MAILER-DAEMON Thu Aug 12 06:46:15 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BvD67-0006e8-5A
	for mharc-classpath@gnu.org; Thu, 12 Aug 2004 06:46:15 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BvD65-0006dj-HN
	for classpath@gnu.org; Thu, 12 Aug 2004 06:46:13 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BvD64-0006dX-G7
	for classpath@gnu.org; Thu, 12 Aug 2004 06:46:13 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BvD64-0006dU-6y
	for classpath@gnu.org; Thu, 12 Aug 2004 06:46:12 -0400
Received: from [212.6.122.10] (helo=mail0.ewetel.de)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BvD1v-0004rr-Pq
	for classpath@gnu.org; Thu, 12 Aug 2004 06:41:56 -0400
Received: from majestix.konqueror.de (dynadsl-080-228-94-132.ewetel.net
	[80.228.94.132])
	by mail0.ewetel.de (8.12.1/8.12.9) with ESMTP id i7CAfmt4009516;
	Thu, 12 Aug 2004 12:41:49 +0200 (MEST)
Received: from miraculix.konqueror.de ([192.168.1.1])
	by majestix.konqueror.de with esmtp (Exim 3.36 #1 (Debian))
	id 1BvDAo-000136-00; Thu, 12 Aug 2004 12:51:06 +0200
From: Michael Koch <konqueror@gmx.de>
To: classpath@gnu.org
Date: Thu, 12 Aug 2004 12:42:36 +0200
User-Agent: KMail/1.6.2
References: <4118B1C1.7000408@aicas.com> <200408111300.44745.konqueror@gmx.de>
	<411B4380.2010006@aicas.com>
In-Reply-To: <411B4380.2010006@aicas.com>
MIME-Version: 1.0
Content-Disposition: inline
Content-Type: text/plain;
  charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
Message-Id: <200408121242.36132.konqueror@gmx.de>
X-CheckCompat: OK
Cc: "Dr. Torsten Rupp" <rupp@aicas.com>
Subject: Re: Target native layer
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Thu, 12 Aug 2004 10:46:13 -0000

Am Donnerstag, 12. August 2004 12:16 schrieb Dr. Torsten Rupp:
> Dear Michael,
>
>  > Wouldnt it be better to call it "posix" (note: no captial
>  > letter) as
>
> it currently supports Linux, *BSD, AIX ... We can and should add
> Solaris too. With autoconf its really simple. Then AICAS can remove
> their own Solaris port totally and we have a clean non-redundant
> codebase for posix-like OSes.
>
>
> It is also not redundant code now, because the special
> implementation also have to be used in a single-directory
> code-basis (currently there are separated files which are a
> "design"-feature; the existence of the files could be called some
> redundancy).

But a autoconf implementation would be much more understandable by 
Free Software developers. Its stuff they act with all day long. Thre 
currently TRAGET_NATIVE solution is ... well ... uncommon and hard to 
understand. What you call a "design" feature will be called "hard to 
understand" because its distributed over the whole (not really true) 
directory tree. Persoanlly I think this is a mis-feature.

>  > Perhaps a "CP" prefix (for ClassPath).
>
> I would prefer some prefix which indicates the "class" of the
> feature instead, e. g. "NC" (native code) or "TN" (target native),
> not some relation to the project name (everthing in Classpath is
> "Classpath", thus should probably have the prefix "CP"). But if
> "CP" would make you and other members happy, it is ok for me, too.

NC (native code): We know its native code, its written in C.

TN (target native): We know its for a special target. Its in a special 
directory when using your solution.

CP (ClassPath): We know its classpath BUT this makes it easy to 
dinstiguish code defined in classpath or outside of classpath, e.g. 
system headers.

>  > There are several ways to deal with the files. Either do it the
>  > AICAS
>
> way and group one OS in one dir or do it the libgcj way and put all
> into one dir, havean OS suffix/prefix in the filename and choose
> the files are configure time.
>
>
> I would like to keep the files in separated directories to keep it
> more clear what files belong to which OS. Usually files are stored
> in a hierarchical order by using sub-directories. For the
> "hierachical" include-tree in the target-layer this was also a good
> idea to avoid confusion in names of files with the same
> function-group (e. g. "network"), but for different systems
> (different implementations).

But we should put ALL possible OSes into ONE posix layer. Solaris 
needs no special layer. Its easy to support in a posix layer. That is 
the way Free Software developers handle it since a long time. No need 
to have it Solaris in an extra directory.

>  > I'm really glad that we now on the way to tackle the issues with
>  > the
>
> TARGE_NATIVE layer.
>
>
> I'm sorry for this hard discussion. But I'm happy we are one a way
> to keep going on together. If everybody becomes satisfied with the
> following new idea for a target layer, I will prepare the changes.
>
> New target native layer:
>
> - transform code inside macros into functions (this will make
> debugging possible)
> - the current macros become an "alias" to the functions (this is
> important to be able to replace a function by something special if
> needed) - combine posix-like systems - Linux, BSD, AIX, Solaris -
> into a single set of files in one directory (usage of autoconf for
> posix-like systems) - add some more layers from aicas

I still have one question: what layers do you want to add and when ? I 
still saw no other layer then the mis-named "Linux" one. I ask this 
because layers for embedded OSes that we can never get in touch with 
or somehow nonsense to all outside of AICAS. We could look at the 
code but nothing else. This doesnt really help either.


Michael



From MAILER-DAEMON Thu Aug 12 08:14:55 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BvETv-0000Co-GD
	for mharc-classpath@gnu.org; Thu, 12 Aug 2004 08:14:55 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BvETt-0000Bv-15
	for classpath@gnu.org; Thu, 12 Aug 2004 08:14:53 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BvETp-00009i-Hk
	for classpath@gnu.org; Thu, 12 Aug 2004 08:14:52 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BvETp-00009Y-5Z
	for classpath@gnu.org; Thu, 12 Aug 2004 08:14:49 -0400
Received: from [212.227.126.173] (helo=moutng.kundenserver.de)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BvEPS-00038U-GI
	for classpath@gnu.org; Thu, 12 Aug 2004 08:10:18 -0400
Received: from [212.227.126.179] (helo=mrelayng.kundenserver.de)
	by moutng.kundenserver.de with esmtp (Exim 3.35 #1)
	id 1BvEPS-0001zO-00; Thu, 12 Aug 2004 14:10:18 +0200
Received: from [212.126.219.82] (helo=aicas.com)
	by mrelayng.kundenserver.de with asmtp (TLSv1:RC4-MD5:128)
	(Exim 3.35 #1) id 1BvEPR-0006C3-00; Thu, 12 Aug 2004 14:10:17 +0200
Message-ID: <411B5E29.3030601@aicas.com>
Date: Thu, 12 Aug 2004 14:10:17 +0200
From: "Dr. Torsten Rupp" <rupp@aicas.com>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.4) Gecko/20030624
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: classpath@gnu.org
References: <4118B1C1.7000408@aicas.com> <200408111300.44745.konqueror@gmx.de>
	<411B4380.2010006@aicas.com> <200408121242.36132.konqueror@gmx.de>
In-Reply-To: <200408121242.36132.konqueror@gmx.de>
X-Enigmail-Version: 0.76.5.0
X-Enigmail-Supports: pgp-inline, pgp-mime
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
X-Provags-ID: kundenserver.de abuse@kundenserver.de
	auth:f9c39743fddc2557f7fa7c1a513f7ea5
Cc: Michael Koch <konqueror@gmx.de>
Subject: Re: Target native layer
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Thu, 12 Aug 2004 12:14:53 -0000

Dear Michael,

> NC (native code): We know its native code, its written in C.
> 
> TN (target native): We know its for a special target. Its in a special 
> directory when using your solution.
> 
> CP (ClassPath): We know its classpath BUT this makes it easy to 
> dinstiguish code defined in classpath or outside of classpath, e.g. 
> system headers.

I would like to do one more step: we know it is native code, but I would 
also like to know _which_ part it belongs to in the native code. I see 
two big parts: target-independent code and target-dependent code (one is 
above the target native layer, the other is inside the target native 
laxer). It would be nice to distinguish these two parts by a suitable 
prefix. Of course "CP" would do this (I can make an "alias" in my brain 
to map all "CP" to "this is target dependent code"), but I think some 
prefix which is indicates this more clearly would be better.

BUT: this detail is not really important. "CP" is ok for me, if it is ok 
for other people, too.

> But we should put ALL possible OSes into ONE posix layer. Solaris 
> needs no special layer. Its easy to support in a posix layer. That is 
> the way Free Software developers handle it since a long time. No need 
> to have it Solaris in an extra directory.

Yes, I following you in this point. That was it, I wanted to say with:

- combine posix-like systems - Linux, BSD, AIX, Solaris - into a single 
set of files in one directory (usage of autoconf for posix-like systems)

> I still have one question: what layers do you want to add and when ? I 
> still saw no other layer then the mis-named "Linux" one. I ask this 
> because layers for embedded OSes that we can never get in touch with 
> or somehow nonsense to all outside of AICAS. We could look at the 
> code but nothing else. This doesnt really help either.

I first wanted to avoid to offer here things in advance (otherwise I 
fear somebody will become upset if I can not do it in time, because of 
limitated manpower). Currently on the list is: Solaris, RTEMS, embOS and 
MinGW (Windows). Please do not ask me exactly _when_ I will supply this, 
but I promise I will do my best to add these as soon as possible.

By the way: I think "Linux" is not mis-named, because it is "Linux". It 
is only a "coincidence" that is 100% identical to the generic case. The 
background is: we are developing basically with Linux.

Sincerly,

Torsten




From MAILER-DAEMON Thu Aug 12 08:25:37 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BvEeG-000353-JY
	for mharc-classpath@gnu.org; Thu, 12 Aug 2004 08:25:36 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BvEeC-00033z-1Y
	for classpath@gnu.org; Thu, 12 Aug 2004 08:25:32 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BvEe9-00032H-4K
	for classpath@gnu.org; Thu, 12 Aug 2004 08:25:31 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BvEe8-00031g-W9
	for classpath@gnu.org; Thu, 12 Aug 2004 08:25:29 -0400
Received: from [212.6.122.12] (helo=mail1.ewetel.de)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BvEa9-0004rF-Rq
	for classpath@gnu.org; Thu, 12 Aug 2004 08:21:22 -0400
Received: from majestix.konqueror.de (dynadsl-080-228-94-132.ewetel.net
	[80.228.94.132])
	by mail1.ewetel.de (8.12.1/8.12.9) with ESMTP id i7CCLKdf025898;
	Thu, 12 Aug 2004 14:21:20 +0200 (MEST)
Received: from miraculix.konqueror.de ([192.168.1.1])
	by majestix.konqueror.de with esmtp (Exim 3.36 #1 (Debian))
	id 1BvEj8-0001JX-00; Thu, 12 Aug 2004 14:30:38 +0200
From: Michael Koch <konqueror@gmx.de>
To: "Dr. Torsten Rupp" <rupp@aicas.com>
Date: Thu, 12 Aug 2004 14:22:07 +0200
User-Agent: KMail/1.6.2
References: <4118B1C1.7000408@aicas.com> <200408121242.36132.konqueror@gmx.de>
	<411B5E29.3030601@aicas.com>
In-Reply-To: <411B5E29.3030601@aicas.com>
MIME-Version: 1.0
Content-Disposition: inline
Content-Type: text/plain;
  charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
Message-Id: <200408121422.07415.konqueror@gmx.de>
X-CheckCompat: OK
Cc: classpath@gnu.org
Subject: Re: Target native layer
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Thu, 12 Aug 2004 12:25:32 -0000

Am Donnerstag, 12. August 2004 14:10 schrieb Dr. Torsten Rupp:

> I first wanted to avoid to offer here things in advance (otherwise
> I fear somebody will become upset if I can not do it in time,
> because of limitated manpower). Currently on the list is: Solaris,
> RTEMS, embOS and MinGW (Windows). Please do not ask me exactly
> _when_ I will supply this, but I promise I will do my best to add
> these as soon as possible.

That sounds great. Can you just post the code without added to CVS ? 
This way we can at least look at it. We should discuss all before 
commiting to CVS anyway.

> By the way: I think "Linux" is not mis-named, because it is
> "Linux". It is only a "coincidence" that is 100% identical to the
> generic case. The background is: we are developing basically with
> Linux.

Its a difference if code is target for Linux only or if its developed 
on Linux only. You use only Linux for it. Roman use it on FreeBSD.
I use this code on NetBSD too. I have access to Solaris machines and 
they want to use this code too. So in fact is not Linux anymore, its 
Posix.


Michael

> Sincerly,
>
> Torsten



From MAILER-DAEMON Thu Aug 12 13:41:56 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BvJaN-0007w6-PR
	for mharc-classpath@gnu.org; Thu, 12 Aug 2004 13:41:55 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BvJaK-0007ph-H4
	for classpath@gnu.org; Thu, 12 Aug 2004 13:41:52 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BvJaJ-0007na-7I
	for classpath@gnu.org; Thu, 12 Aug 2004 13:41:51 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BvJaI-0007n3-W9
	for classpath@gnu.org; Thu, 12 Aug 2004 13:41:51 -0400
Received: from [207.174.178.92] (helo=email2.peakpeak.com)
	by monty-python.gnu.org with esmtp (TLSv1:DES-CBC3-SHA:168)
	(Exim 4.34) id 1BvJWC-0004gt-VG
	for classpath@gnu.org; Thu, 12 Aug 2004 13:37:37 -0400
Received: (qmail 17326 invoked by uid 513); 12 Aug 2004 17:42:06 -0000
Received: from tromey@redhat.com by email2 by uid 504 with qmail-scanner-1.20 
	(clamscan: 0.65.  Clear:RC:1(204.144.239.10):. 
	Processed in 0.209624 secs); 12 Aug 2004 17:42:06 -0000
Received: from tf0010.peakpeak.com (HELO fleche.redhat.com) ([204.144.239.10])
	(envelope-sender <tromey@redhat.com>)
	by email2.peakpeak.com (qmail-ldap-1.03) with SMTP
	for <classpath@gnu.org>; 12 Aug 2004 17:42:05 -0000
Received: by fleche.redhat.com (Postfix, from userid 1000)
	id 1FB784F8A1C; Thu, 12 Aug 2004 11:18:41 -0600 (MDT)
To: GNU Classpath Project <classpath@gnu.org>
From: Tom Tromey <tromey@redhat.com>
X-Attribution: Tom
Date: 12 Aug 2004 11:18:40 -0600
Message-ID: <87pt5wp95b.fsf@fleche.redhat.com>
Lines: 11
User-Agent: Gnus/5.09 (Gnus v5.9.0) Emacs/21.3.50
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Subject: VMClassLoader
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
Reply-To: tromey@redhat.com
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Thu, 12 Aug 2004 17:41:54 -0000

I notice that VMClassLoader still has this:

   * <strong>For backward compatibility, this just ignores the protection
   * domain; that is the wrong behavior, and you should directly implement
   * this method natively if you can.</strong>

I suggest we pick a flag day to switch this.  In particular I think we
ought to simply remove the old form of defineClass() and only have the
form that accepts a ProtectionDomain.

Tom



From MAILER-DAEMON Fri Aug 13 08:03:58 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1Bval1-0000XV-F1
	for mharc-classpath@gnu.org; Fri, 13 Aug 2004 08:02:04 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1Bvaky-0000Vz-JD
	for classpath@gnu.org; Fri, 13 Aug 2004 08:02:00 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1Bvakt-0000U0-Ps
	for classpath@gnu.org; Fri, 13 Aug 2004 08:01:57 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1Bvakt-0000QV-Ge
	for classpath@gnu.org; Fri, 13 Aug 2004 08:01:55 -0400
Received: from [82.161.94.186] (helo=gnu.wildebeest.org)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BvagT-0000up-Sg
	for classpath@gnu.org; Fri, 13 Aug 2004 07:57:22 -0400
Received: from slauerhoff.wildebeest.org ([192.168.1.27] ident=mark)
	by gnu.wildebeest.org with esmtp (Exim 3.35 #1 (Debian))
	id 1BvagR-0000Uq-00; Fri, 13 Aug 2004 13:57:19 +0200
From: Mark Wielaard <mark@klomp.org>
To: Guilhem Lavaux <guilhem@kaffe.org>
In-Reply-To: <41090A12.2020807@kaffe.org>
References: <41090A12.2020807@kaffe.org>
Content-Type: multipart/signed; micalg=pgp-sha1;
	protocol="application/pgp-signature";
	boundary="=-XF+4HyC1RU9GS1rnoO77"
Message-Id: <1092398238.1592.21.camel@localhost>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.4.6 
Date: Fri, 13 Aug 2004 13:57:18 +0200
Cc: classpath@gnu.org
Subject: Re: About locale data ... (again)
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Fri, 13 Aug 2004 12:02:02 -0000


--=-XF+4HyC1RU9GS1rnoO77
Content-Type: text/plain
Content-Transfer-Encoding: quoted-printable

Hi,

On Thu, 2004-07-29 at 16:30, Guilhem Lavaux wrote:
> As we always need a good locale database for GNU Classpath, I was=20
> thinking of a way to get it. The ICU locale converter is only partially=20
> satisfying as it does not have any collation rules generator (at least).=20
> I've seen that www.openi18n.org[1][2] has begun a new standard file=20
> format to interchange locale data. It is purely written in XML form and=20
> looks well documented. On IBM[3], we can find the locale database=20
> waiting to be converted.

This looks like a very promising way to get good locale data for GNU
Classpath. Is there a statement from IBM[3] that the locale database is
free for use by projects like ours (GPL compatible)? If not, then we
need to have a contact address at IBM to explicitly ask for permission.
(I can do that if you want.)

Cheers,

Mark

--=-XF+4HyC1RU9GS1rnoO77
Content-Type: application/pgp-signature; name=signature.asc
Content-Description: This is a digitally signed message part

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)

iD8DBQBBHKyexVhZCJWr9QwRAsw2AKCknbfV2pTAx/hRygO8yTc3DVspsACeMm0n
NMLV3D8cDutZHvO2Wgvx/bg=
=LHZH
-----END PGP SIGNATURE-----

--=-XF+4HyC1RU9GS1rnoO77--




From MAILER-DAEMON Fri Aug 13 08:30:42 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BvbCj-0000Pi-Pt
	for mharc-classpath@gnu.org; Fri, 13 Aug 2004 08:30:41 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BvbCi-0000PQ-QB
	for classpath@gnu.org; Fri, 13 Aug 2004 08:30:41 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BvbCi-0000PE-1H
	for classpath@gnu.org; Fri, 13 Aug 2004 08:30:40 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BvbCh-0000P4-Ug
	for classpath@gnu.org; Fri, 13 Aug 2004 08:30:39 -0400
Received: from [213.228.0.169] (helo=postfix3-2.free.fr)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1Bvb8T-0006A4-0Y
	for classpath@gnu.org; Fri, 13 Aug 2004 08:26:17 -0400
Received: from kaffe.org (lns-vlq-17f-81-56-169-129.adsl.proxad.net
	[81.56.169.129]) by postfix3-2.free.fr (Postfix) with ESMTP
	id BE6DBC322; Fri, 13 Aug 2004 16:21:17 +0200 (CEST)
Message-ID: <411CB369.9090500@kaffe.org>
Date: Fri, 13 Aug 2004 14:26:17 +0200
From: Guilhem Lavaux <guilhem@kaffe.org>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.6) Gecko/20040115
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: Mark Wielaard <mark@klomp.org>
References: <41090A12.2020807@kaffe.org> <1092398238.1592.21.camel@localhost>
In-Reply-To: <1092398238.1592.21.camel@localhost>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Cc: classpath@gnu.org
Subject: Re: About locale data ... (again)
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Fri, 13 Aug 2004 12:30:41 -0000

Mark Wielaard wrote:

>Hi,
>
>On Thu, 2004-07-29 at 16:30, Guilhem Lavaux wrote:
>  
>
>>As we always need a good locale database for GNU Classpath, I was 
>>thinking of a way to get it. The ICU locale converter is only partially 
>>satisfying as it does not have any collation rules generator (at least). 
>>I've seen that www.openi18n.org[1][2] has begun a new standard file 
>>format to interchange locale data. It is purely written in XML form and 
>>looks well documented. On IBM[3], we can find the locale database 
>>waiting to be converted.
>>    
>>
>
>This looks like a very promising way to get good locale data for GNU
>Classpath. Is there a statement from IBM[3] that the locale database is
>free for use by projects like ours (GPL compatible)? If not, then we
>need to have a contact address at IBM to explicitly ask for permission.
>(I can do that if you want.)
>
>Cheers,
>
>Mark
>  
>

Hi Mark,

The Copyright & Permission Notice for LDML and the locale data 
repository is contained in the following file:

http://oss.software.ibm.com/cvs/icu/~checkout~/locale/CLDR_status.html

It's the same IBM Licence "do what you want with it" from what I can see.

Cheers,

Guilhem.



From MAILER-DAEMON Fri Aug 13 08:38:44 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BvbKW-0002tJ-Ps
	for mharc-classpath@gnu.org; Fri, 13 Aug 2004 08:38:44 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BvbKV-0002sd-JI
	for classpath@gnu.org; Fri, 13 Aug 2004 08:38:43 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BvbKU-0002s4-Sv
	for classpath@gnu.org; Fri, 13 Aug 2004 08:38:43 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BvbKU-0002rs-NX
	for classpath@gnu.org; Fri, 13 Aug 2004 08:38:42 -0400
Received: from [213.228.0.169] (helo=postfix3-2.free.fr)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BvbG4-0007hh-5M
	for classpath@gnu.org; Fri, 13 Aug 2004 08:34:08 -0400
Received: from kaffe.org (lns-vlq-17f-81-56-169-129.adsl.proxad.net
	[81.56.169.129]) by postfix3-2.free.fr (Postfix) with ESMTP
	id 2BD33C113; Fri, 13 Aug 2004 16:29:10 +0200 (CEST)
Message-ID: <411CB542.50100@kaffe.org>
Date: Fri, 13 Aug 2004 14:34:10 +0200
From: Guilhem Lavaux <guilhem@kaffe.org>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.6) Gecko/20040115
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: Mark Wielaard <mark@klomp.org>
References: <41090A12.2020807@kaffe.org> <1092398238.1592.21.camel@localhost>
In-Reply-To: <1092398238.1592.21.camel@localhost>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Cc: classpath@gnu.org
Subject: Re: About locale data ... (again)
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Fri, 13 Aug 2004 12:38:44 -0000

Mark Wielaard wrote:

>Hi,
>
>On Thu, 2004-07-29 at 16:30, Guilhem Lavaux wrote:
>  
>
>>As we always need a good locale database for GNU Classpath, I was 
>>thinking of a way to get it. The ICU locale converter is only partially 
>>satisfying as it does not have any collation rules generator (at least). 
>>I've seen that www.openi18n.org[1][2] has begun a new standard file 
>>format to interchange locale data. It is purely written in XML form and 
>>looks well documented. On IBM[3], we can find the locale database 
>>waiting to be converted.
>>    
>>
>
>This looks like a very promising way to get good locale data for GNU
>Classpath. Is there a statement from IBM[3] that the locale database is
>free for use by projects like ours (GPL compatible)? If not, then we
>need to have a contact address at IBM to explicitly ask for permission.
>(I can do that if you want.)
>
>Cheers,
>
>Mark
>  
>
Here are the details, for the list...

Cheers,

Guilhem.


COPYRIGHT AND PERMISSION NOTICE
 
 Copyright (c) 2002-2003 International Business Machines Corporation and 
others
 All rights reserved.
 
 Permission is hereby granted, free of charge, to any person obtaining a 
copy of the Common Locale data (the "Data"), to deal in the Data without 
restriction, including without limitation the rights to use, copy, 
modify, merge, publish, distribute, and/or sell copies of the Data, and 
to permit persons to whom the Data is furnished to do so, provided both 
the above copyright notice(s) and this permission notice appear in 
supporting documentation.


THE DATA IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT OF THIRD PARTY 
RIGHTS. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR HOLDERS INCLUDED IN 
THIS NOTICE BE LIABLE FOR ANY CLAIM, OR ANY SPECIAL INDIRECT OR 
CONSEQUENTIAL DAMAGES, OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF 
USE, DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR 
OTHER TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR 
PERFORMANCE OF THIS DATA.


Except as contained in this notice, the name of a copyright holder shall 
not be used in advertising or otherwise to promote the sale, use or 
other dealings in this Data without prior written authorization of the 
copyright holder.



From MAILER-DAEMON Fri Aug 13 12:00:32 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BveTn-0004C5-PC
	for mharc-classpath@gnu.org; Fri, 13 Aug 2004 12:00:31 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BveTi-0004BM-EO
	for classpath@gnu.org; Fri, 13 Aug 2004 12:00:26 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BveTh-0004Aw-0g
	for classpath@gnu.org; Fri, 13 Aug 2004 12:00:25 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33)
	id 1BveTg-0004Ar-Qc; Fri, 13 Aug 2004 12:00:24 -0400
Received: from [82.161.94.186] (helo=gnu.wildebeest.org)
	by monty-python.gnu.org with esmtp (Exim 4.34)
	id 1BvePg-0006NR-2K; Fri, 13 Aug 2004 11:56:16 -0400
Received: from slauerhoff.wildebeest.org ([192.168.1.27] ident=mark)
	by gnu.wildebeest.org with esmtp (Exim 3.35 #1 (Debian))
	id 1BvePf-00014u-00; Fri, 13 Aug 2004 17:56:15 +0200
From: Mark Wielaard <mark@klomp.org>
To: Casey Marshall <csm@gnu.org>
In-Reply-To: <87wu0acpkx.fsf@gnu.org>
References: <87wu0acpkx.fsf@gnu.org>
Content-Type: multipart/signed; micalg=pgp-sha1;
	protocol="application/pgp-signature";
	boundary="=-fTBlzxB9RbYQcXYij3ng"
Message-Id: <1092412573.1592.84.camel@localhost>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.4.6 
Date: Fri, 13 Aug 2004 17:56:14 +0200
Cc: classpath@gnu.org, classpath-patches@gnu.org, java@gcc.gnu.org
Subject: Re: [cp-patches] Updated: crypto merge patch
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Fri, 13 Aug 2004 16:00:27 -0000


--=-fTBlzxB9RbYQcXYij3ng
Content-Type: text/plain
Content-Transfer-Encoding: quoted-printable

Hi,

CCed the main classpath and gcj mailinglists to make everybody aware of
this.

On Sun, 2004-08-08 at 04:48, Casey Marshall wrote:
> I've updated the patch to import the APIs from GNU Crypto and Jessie,
> which is still here:
>=20
>    <http://syzygy.metastatic.org/crypto.patch.gz>
>    <http://syzygy.metastatic.org/crypto.ChangeLog>
>=20
> This cleans up a few comments, and adds the javax.security.sasl
> package.
>=20
> I don't know when this will go in, nor am I sure that this is
> completely compatible with the changes to the build scripts.

Thanks! Sorry for the long delay. Yes, it can fianlly go in (but see
below). I just wanted to clear everything with FSF legal first. So we
know what the impact is on projects that are based on GNU Classpath. The
situation with respect to U.S Export Regulations when we distribute
crypto code is as follows:

Free Software implementing crypto can in principle be exported from the
U.S. when a certain BIS/ENC notification process is followed. The FSF
has the required notfications on savannah and ftp.gnu.org to cover all
GNU projects (this means GCC is automatically covered). This
notification also covers all mirrors of GNU Classpath (or any project
hosted on savannah/ftp.gnu.org).

Note that it does not automatically cover new projects based on GNU
Classpath that incorporate the new encryption source code when it is
(re)exported from the U.S (like Kaffe). If you want to do that then=20
please see the Bureau of Industry and Security (www.bxa.doc.gov) for
more information about current U.S. regulations. In particular you will
have to contact the BIS and the ENC Encryption Request Coordinator:
http://www.bxa.doc.gov/Encryption/PubAvailEncSourceCodeNofify.html

With that out of the way I want to thanks Casey and the other GNU Crypto
hackers for making this possible in the first place.

About the patch and the ChangeLog. They look fine to me. I only had two
issues.

First javax/net/ssl/HandshakeCompletedEvent.java  and
javax/net/ssl/SSLSession.java use javax.security.cert.X509Certificate. I
don't know if we even want to provide the javax.security.cert package
since it is obviously not needed by anyone (or is it?)

Second you need to add the new Makefiles in the resources subdirectories
to the configure.ac file (see below).

With that fixed make distcheck works correctly and the source compiles
with both jikes and gcj. So please check this in.

And maybe we want to add a little NEWS blurb like:

* New javax.crypto, javax.crypto.interfaces, javax.crypto.spec, javax.net,
  javax.net.ssl, javax.security.auth, javax.security.auth.callback,
  javax.security.auth.login, javax.security.auth.x500, javax.security.sasl
  and org.ietf.jgss are now officially part of GNU Classpath.
  Extra crypto algorithms can be obtained from the GNU Crypto project,
  a full TLS implementation is provided by the Jessie project.
  http://www.gnu.org/software/gnu-crypto/
  http://www.nongnu.org/jessie/

Thanks,

Mark


Patch for configure.ac:

--- configure.ac        1 Aug 2004 21:31:11 -0000       1.43
+++ configure.ac        13 Aug 2004 15:40:18 -0000
@@ -324,6 +324,9 @@
 resource/java/Makefile
 resource/java/util/Makefile
 resource/java/security/Makefile
+resource/org/Makefile
+resource/org/ietf/Makefile
+resource/org/ietf/jgss/Makefile
 lib/Makefile
 lib/gen-classlist.sh
 examples/Makefile])



--=-fTBlzxB9RbYQcXYij3ng
Content-Type: application/pgp-signature; name=signature.asc
Content-Description: This is a digitally signed message part

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)

iD8DBQBBHOSdxVhZCJWr9QwRAoU0AJ40sKXCZuD/bGtmntBH2n/OebeJrwCdGgZo
Y6Yg1AIacP/2AV4ooIrJq5E=
=2FkZ
-----END PGP SIGNATURE-----

--=-fTBlzxB9RbYQcXYij3ng--




From MAILER-DAEMON Fri Aug 13 13:58:56 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BvgKO-0001Wp-I9
	for mharc-classpath@gnu.org; Fri, 13 Aug 2004 13:58:56 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BvgKM-0001VU-M5
	for classpath@gnu.org; Fri, 13 Aug 2004 13:58:54 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BvgKM-0001VI-0R
	for classpath@gnu.org; Fri, 13 Aug 2004 13:58:54 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BvgKL-0001VF-UM
	for classpath@gnu.org; Fri, 13 Aug 2004 13:58:53 -0400
Received: from [209.152.205.226] (helo=NR-EXCHANGE.netreach.core)
	by monty-python.gnu.org with esmtp (Exim 4.34)
	id 1BvgGA-0007d6-7o; Fri, 13 Aug 2004 13:54:34 -0400
Received: from [209.152.205.123] ([209.152.205.123]) by
	NR-EXCHANGE.netreach.core with Microsoft SMTPSVC(6.0.3790.0); 
	Fri, 13 Aug 2004 13:57:39 -0400
Message-ID: <411D0007.6050109@netreach.com>
Date: Fri, 13 Aug 2004 13:53:11 -0400
From: Stuart Ballard <sballard@netreach.com>
User-Agent: Mozilla Thunderbird 0.6 (X11/20040605)
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: Mark Wielaard <mark@klomp.org>
References: <87wu0acpkx.fsf@gnu.org> <1092412573.1592.84.camel@localhost>
In-Reply-To: <1092412573.1592.84.camel@localhost>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
X-OriginalArrivalTime: 13 Aug 2004 17:57:39.0407 (UTC)
	FILETIME=[05E645F0:01C4815F]
Cc: Casey Marshall <csm@gnu.org>, classpath@gnu.org
Subject: Re: [cp-patches] Updated: crypto merge patch
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Fri, 13 Aug 2004 17:58:54 -0000

Mark Wielaard wrote:
>>I've updated the patch to import the APIs from GNU Crypto and Jessie,
>>which is still here:

I just noticed how great Classpath is doing these days in Japitools 
coverage.

Versus 1.2, the only oranges and reds (below 80% coverage of any 
package) are in Swing and CORBA (the latter being totally missing, of 
course).

Versus 1.3, only Sound is added to that list. Oh, and java.awt becomes a 
very light orange (79%).

Versus 1.4, Imageio and Security get added to that list (not sure 
whether the current run includes this new security work, though). Oh, 
and javax.print.attribute.standard which is a very light orange (78%).


The big holes, therefore, are only:
- Swing (obviously heavily in progress)
- CORBA (no work is being done here AFAIK)
- Sound (Kaffe scores almost 100% here, some merging needs doing?)
- Imageio (not sure of the state of this)
- Security (which is the subject of the message I'm replying to).

Getting to the point of only four major holes in an API as large as Java 
1.4 is incredible, unbelievable progress. Obviously API coverage doesn't 
tell the whole story, but I still think congratulations are in order :)

Stuart.

PS I'm still interested if anyone wants to step up to maintain or 
co-maintain Japitools. Preferably someone who, unlike me, understands 
the bytecode file format, because we'll shortly be faced with making 
Japitools handle generics, and I have NO clue how to approach that problem.


-- 
Stuart Ballard, Senior Web Developer
NetReach, Inc.
(215) 283-2300, ext. 126
http://www.netreach.com/



From MAILER-DAEMON Fri Aug 13 16:04:56 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BviIJ-0001TB-Uk
	for mharc-classpath@gnu.org; Fri, 13 Aug 2004 16:04:56 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BviIH-0001SO-SS
	for classpath@gnu.org; Fri, 13 Aug 2004 16:04:53 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BviIG-0001RT-18
	for classpath@gnu.org; Fri, 13 Aug 2004 16:04:53 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BviIF-0001RQ-UC
	for classpath@gnu.org; Fri, 13 Aug 2004 16:04:51 -0400
Received: from [66.33.197.214] (helo=ludo.dreamhost.com)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BviE2-0000xX-8a
	for classpath@gnu.org; Fri, 13 Aug 2004 16:00:30 -0400
Received: from dexter (c-24-6-46-122.client.comcast.net [24.6.46.122])
	by ludo.dreamhost.com (Postfix) with ESMTP
	id B54E543A39; Fri, 13 Aug 2004 13:00:29 -0700 (PDT)
Received: by dexter (Postfix, from userid 1000)
	id A5C9139; Fri, 13 Aug 2004 13:00:26 -0700 (PDT)
To: Stuart Ballard <sballard@netreach.com>
References: <87wu0acpkx.fsf@gnu.org> <1092412573.1592.84.camel@localhost>
	<411D0007.6050109@netreach.com>
From: Casey Marshall <csm@gnu.org>
Date: Fri, 13 Aug 2004 13:00:26 -0700
In-Reply-To: <411D0007.6050109@netreach.com> (Stuart Ballard's message of
	"Fri, 13 Aug 2004 13:53:11 -0400")
Message-ID: <87smaqbyg5.fsf@gnu.org>
User-Agent: Gnus/5.1002 (Gnus v5.10.2) Emacs/21.2 (gnu/linux)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Cc: Mark Wielaard <mark@klomp.org>, classpath@gnu.org
Subject: Re: [cp-patches] Updated: crypto merge patch
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Fri, 13 Aug 2004 20:04:54 -0000

>>>>> "Stuart" == Stuart Ballard <sballard@netreach.com> writes:

Stuart> The big holes, therefore, are only:
Stuart> - Swing (obviously heavily in progress)
Stuart> - CORBA (no work is being done here AFAIK)

I think there are free implementations of this being done by others. I
don't know about licensing etc.

Stuart> - Sound (Kaffe scores almost 100% here, some merging needs
Stuart> doing?)

Kaffe gets javax.sound from Tritonus, which is LGPL, so we can't
include it in Classpath. We could ask them for permission to do so
under Classpath's license, if they are still around (the news page has
its latest item in 2003).

<http://www.tritonus.org/>

-- 
Casey Marshall || csm@gnu.org



From MAILER-DAEMON Fri Aug 13 17:48:55 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1Bvjux-00031T-0H
	for mharc-classpath@gnu.org; Fri, 13 Aug 2004 17:48:55 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1Bvjuv-00031O-Sr
	for classpath@gnu.org; Fri, 13 Aug 2004 17:48:53 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1Bvjuu-00031C-BD
	for classpath@gnu.org; Fri, 13 Aug 2004 17:48:53 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1Bvjuu-000319-7P
	for classpath@gnu.org; Fri, 13 Aug 2004 17:48:52 -0400
Received: from [82.161.94.186] (helo=gnu.wildebeest.org)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1Bvjqb-0007Ts-8x
	for classpath@gnu.org; Fri, 13 Aug 2004 17:44:25 -0400
Received: from slauerhoff.wildebeest.org ([192.168.1.27] ident=mark)
	by gnu.wildebeest.org with esmtp (Exim 3.35 #1 (Debian))
	id 1BvjqO-0001sT-00; Fri, 13 Aug 2004 23:44:12 +0200
From: Mark Wielaard <mark@klomp.org>
To: Guilhem Lavaux <guilhem@kaffe.org>
In-Reply-To: <411CB369.9090500@kaffe.org>
References: <41090A12.2020807@kaffe.org>
	<1092398238.1592.21.camel@localhost>  <411CB369.9090500@kaffe.org>
Content-Type: multipart/signed; micalg=pgp-sha1;
	protocol="application/pgp-signature";
	boundary="=-hqejsE+exjc2g+5jjdnc"
Message-Id: <1092433451.1582.126.camel@localhost>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.4.6 
Date: Fri, 13 Aug 2004 23:44:11 +0200
Cc: classpath@gnu.org
Subject: Re: About locale data ... (again)
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Fri, 13 Aug 2004 21:48:54 -0000


--=-hqejsE+exjc2g+5jjdnc
Content-Type: text/plain
Content-Transfer-Encoding: quoted-printable

Hi,

On Fri, 2004-08-13 at 14:26, Guilhem Lavaux wrote:
> Mark Wielaard wrote:
>
> >This looks like a very promising way to get good locale data for GNU
> >Classpath. Is there a statement from IBM[3] that the locale database is
> >free for use by projects like ours (GPL compatible)? If not, then we
> >need to have a contact address at IBM to explicitly ask for permission.
> >(I can do that if you want.)
>
> The Copyright & Permission Notice for LDML and the locale data=20
> repository is contained in the following file:
>=20
> http://oss.software.ibm.com/cvs/icu/~checkout~/locale/CLDR_status.html
>=20
> It's the same IBM Licence "do what you want with it" from what I can see.

Yes, thanks it will be no problem to use this data for our own locale
databases.

Thanks,

Mark

--=-hqejsE+exjc2g+5jjdnc
Content-Type: application/pgp-signature; name=signature.asc
Content-Description: This is a digitally signed message part

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)

iD8DBQBBHTYrxVhZCJWr9QwRAvHBAJ0emO3IwEkgEfgWZhYpoFi12iJAhgCgkAp4
RrgZX3bhVbZm7Od2z9etbn4=
=J+2I
-----END PGP SIGNATURE-----

--=-hqejsE+exjc2g+5jjdnc--




From MAILER-DAEMON Fri Aug 13 18:35:37 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1Bvke9-0001Wn-Cn
	for mharc-classpath@gnu.org; Fri, 13 Aug 2004 18:35:37 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1Bvke7-0001WW-GI
	for classpath@gnu.org; Fri, 13 Aug 2004 18:35:35 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1Bvke5-0001WC-JN
	for classpath@gnu.org; Fri, 13 Aug 2004 18:35:35 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1Bvke5-0001W2-Gw
	for classpath@gnu.org; Fri, 13 Aug 2004 18:35:33 -0400
Received: from [207.189.223.52] (helo=email3.peakpeak.com)
	by monty-python.gnu.org with esmtp (TLSv1:DES-CBC3-SHA:168)
	(Exim 4.34) id 1Bvka2-0005cG-TB
	for classpath@gnu.org; Fri, 13 Aug 2004 18:31:23 -0400
Received: (qmail 30669 invoked by uid 513); 13 Aug 2004 22:43:17 -0000
Received: from tromey@redhat.com by email3 by uid 504 with qmail-scanner-1.20 
	(clamscan: 0.65.  Clear:RC:1(204.144.239.10):. 
	Processed in 0.129619 secs); 13 Aug 2004 22:43:17 -0000
Received: from tf0010.peakpeak.com (HELO fleche.redhat.com) ([204.144.239.10])
	(envelope-sender <tromey@redhat.com>)
	by email3.peakpeak.com (qmail-ldap-1.03) with SMTP
	for <classpath@gnu.org>; 13 Aug 2004 22:43:17 -0000
Received: by fleche.redhat.com (Postfix, from userid 1000)
	id 6527E4F89EC; Fri, 13 Aug 2004 16:14:32 -0600 (MDT)
To: Stuart Ballard <sballard@netreach.com>
References: <87wu0acpkx.fsf@gnu.org> <1092412573.1592.84.camel@localhost>
	<411D0007.6050109@netreach.com>
From: Tom Tromey <tromey@redhat.com>
X-Attribution: Tom
Date: 13 Aug 2004 16:14:31 -0600
In-Reply-To: <411D0007.6050109@netreach.com>
Message-ID: <87657mn0s8.fsf@fleche.redhat.com>
Lines: 38
User-Agent: Gnus/5.09 (Gnus v5.9.0) Emacs/21.3.50
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Cc: Casey Marshall <csm@gnu.org>, Mark Wielaard <mark@klomp.org>,
	classpath@gnu.org
Subject: Re: [cp-patches] Updated: crypto merge patch
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
Reply-To: tromey@redhat.com
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Fri, 13 Aug 2004 22:35:35 -0000

>>>>> "Stuart" == Stuart Ballard <sballard@netreach.com> writes:

Stuart> I just noticed how great Classpath is doing these days in
Stuart> Japitools coverage.

Yeah, it is awesome.  I wish I'd kept all those old comparisons in
cvs, so I could make a chart showing the growth.

Stuart> - CORBA (no work is being done here AFAIK)

There is JacORB and OpenORB.  They have some license trouble, but
nothing really fatal.  We couldn't put them in classpath though.

I've considered running my nightly comparison including JacORB, but
haven't set that up yet.  I don't think it is fully building with free
tools... Anthony has looked into this a bit.

Stuart> - Imageio (not sure of the state of this)

I thought Michael Koch had some of this... ?

Stuart> PS I'm still interested if anyone wants to step up to maintain or
Stuart> co-maintain Japitools. Preferably someone who, unlike me, understands
Stuart> the bytecode file format, because we'll shortly be faced with making
Stuart> Japitools handle generics, and I have NO clue how to approach that
Stuart> problem.

There is a new flag for enum (and I thought another one, but I
forget), new signatures for generics, and also additions for
attributes.  These would all need to be included in the comparison, I
think.

We definitely need someone to work on this actively... it will help
the generics effort along.  Also perhaps we could fix the one or two
bugs in japi (I don't know what they are exactly, but I think some of
the differences on that page are spurious...)

Tom



From MAILER-DAEMON Fri Aug 13 20:37:58 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BvmYY-0003Xy-5H
	for mharc-classpath@gnu.org; Fri, 13 Aug 2004 20:37:58 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BvmYU-0003Xf-Kk
	for classpath@gnu.org; Fri, 13 Aug 2004 20:37:56 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BvmYS-0003XT-U2
	for classpath@gnu.org; Fri, 13 Aug 2004 20:37:54 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BvmYS-0003XQ-Ou
	for classpath@gnu.org; Fri, 13 Aug 2004 20:37:52 -0400
Received: from [217.12.12.140] (helo=smtp803.mail.ukl.yahoo.com)
	by monty-python.gnu.org with smtp (Exim 4.34) id 1BvmUA-0004cV-P9
	for classpath@gnu.org; Fri, 13 Aug 2004 20:33:27 -0400
Received: from unknown (HELO rivendell) (classpath@gnu.org@81.152.170.101 with
	poptime)
	by smtp803.mail.ukl.yahoo.com with SMTP; 14 Aug 2004 00:33:25 -0000
Received: from localhost ([127.0.0.1]) by rivendell with esmtp (Exim 4.34)
	id 1BvmU9-0003TU-AP
	for classpath@gnu.org; Sat, 14 Aug 2004 01:33:25 +0100
From: Andrew John Hughes <gnu_andrew@member.fsf.org>
To: GNU Classpath <classpath@gnu.org>
In-Reply-To: <1092412573.1592.84.camel@localhost>
References: <87wu0acpkx.fsf@gnu.org>  <1092412573.1592.84.camel@localhost>
Content-Type: multipart/signed; micalg=pgp-sha1;
	protocol="application/pgp-signature";
	boundary="=-MKuw4r4vJ5YIRklWZ8Oj"
Message-Id: <1092443605.2477.3.camel@rivendell>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.4.6 
Date: Sat, 14 Aug 2004 01:33:25 +0100
Subject: Re: [cp-patches] Updated: crypto merge patch
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
Reply-To: gnu_andrew@member.fsf.org
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Sat, 14 Aug 2004 00:37:56 -0000


--=-MKuw4r4vJ5YIRklWZ8Oj
Content-Type: text/plain
Content-Transfer-Encoding: quoted-printable

On Fri, 2004-08-13 at 16:56, Mark Wielaard wrote:
> Hi,
>=20
> CCed the main classpath and gcj mailinglists to make everybody aware of
> this.
>=20
> On Sun, 2004-08-08 at 04:48, Casey Marshall wrote:
> > I've updated the patch to import the APIs from GNU Crypto and Jessie,
> > which is still here:
> >=20
> >    <http://syzygy.metastatic.org/crypto.patch.gz>
> >    <http://syzygy.metastatic.org/crypto.ChangeLog>
> >=20
> > This cleans up a few comments, and adds the javax.security.sasl
> > package.
> >=20
> > I don't know when this will go in, nor am I sure that this is
> > completely compatible with the changes to the build scripts.
>=20
> Thanks! Sorry for the long delay. Yes, it can fianlly go in (but see
> below). I just wanted to clear everything with FSF legal first. So we
> know what the impact is on projects that are based on GNU Classpath. The
> situation with respect to U.S Export Regulations when we distribute
> crypto code is as follows:
>=20
> Free Software implementing crypto can in principle be exported from the
> U.S. when a certain BIS/ENC notification process is followed. The FSF
> has the required notfications on savannah and ftp.gnu.org to cover all
> GNU projects (this means GCC is automatically covered). This
> notification also covers all mirrors of GNU Classpath (or any project
> hosted on savannah/ftp.gnu.org).
>=20
> Note that it does not automatically cover new projects based on GNU
> Classpath that incorporate the new encryption source code when it is
> (re)exported from the U.S (like Kaffe). If you want to do that then=20
> please see the Bureau of Industry and Security (www.bxa.doc.gov) for
> more information about current U.S. regulations. In particular you will
> have to contact the BIS and the ENC Encryption Request Coordinator:
> http://www.bxa.doc.gov/Encryption/PubAvailEncSourceCodeNofify.html
>=20
> With that out of the way I want to thanks Casey and the other GNU Crypto
> hackers for making this possible in the first place.
>=20
> About the patch and the ChangeLog. They look fine to me. I only had two
> issues.
>=20
> First javax/net/ssl/HandshakeCompletedEvent.java  and
> javax/net/ssl/SSLSession.java use javax.security.cert.X509Certificate. I
> don't know if we even want to provide the javax.security.cert package
> since it is obviously not needed by anyone (or is it?)
>=20
> Second you need to add the new Makefiles in the resources subdirectories
> to the configure.ac file (see below).
>=20
> With that fixed make distcheck works correctly and the source compiles
> with both jikes and gcj. So please check this in.
>=20
> And maybe we want to add a little NEWS blurb like:
>=20
> * New javax.crypto, javax.crypto.interfaces, javax.crypto.spec, javax.net=
,
>   javax.net.ssl, javax.security.auth, javax.security.auth.callback,
>   javax.security.auth.login, javax.security.auth.x500, javax.security.sas=
l
>   and org.ietf.jgss are now officially part of GNU Classpath.
>   Extra crypto algorithms can be obtained from the GNU Crypto project,
>   a full TLS implementation is provided by the Jessie project.
>   http://www.gnu.org/software/gnu-crypto/
>   http://www.nongnu.org/jessie/
>=20
> Thanks,
>=20
> Mark
>=20
>=20
> Patch for configure.ac:
>=20
> --- configure.ac        1 Aug 2004 21:31:11 -0000       1.43
> +++ configure.ac        13 Aug 2004 15:40:18 -0000
> @@ -324,6 +324,9 @@
>  resource/java/Makefile
>  resource/java/util/Makefile
>  resource/java/security/Makefile
> +resource/org/Makefile
> +resource/org/ietf/Makefile
> +resource/org/ietf/jgss/Makefile
>  lib/Makefile
>  lib/gen-classlist.sh
>  examples/Makefile])
>=20
>=20
>=20
> ______________________________________________________________________
> _______________________________________________
> Classpath mailing list
> Classpath@gnu.org
> http://lists.gnu.org/mailman/listinfo/classpath

Classpath still includes deprecated methods, so why not a deprecated
package?  Some older code may need it (possibly something that is no
longer maintained), and Sun appear to be continuing support for it in
the current 1.5 beta.
--=20
Andrew :-)
=20
Please avoid sending me Word or PowerPoint attachments.
See http://www.fsf.org/philosophy/no-word-attachments.html
=20
Value your freedom, or you will lose it, teaches history.
`Don't bother us with politics' respond those who don't want to learn.
=20


--=-MKuw4r4vJ5YIRklWZ8Oj
Content-Type: application/pgp-signature; name=signature.asc
Content-Description: This is a digitally signed message part

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)

iD8DBQBBHV3Uu+fW6pI0dq4RAlrFAJ9NEcHDJUDmEwOmsCx3b1ExvcRD2QCglMkv
9VsNm0+JauFN0iM7mL8T/uI=
=CElM
-----END PGP SIGNATURE-----

--=-MKuw4r4vJ5YIRklWZ8Oj--




From MAILER-DAEMON Fri Aug 13 20:57:58 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1Bvmru-0006eL-H4
	for mharc-classpath@gnu.org; Fri, 13 Aug 2004 20:57:58 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1Bvmrs-0006df-VB
	for classpath@gnu.org; Fri, 13 Aug 2004 20:57:57 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1Bvmrs-0006dP-Bx
	for classpath@gnu.org; Fri, 13 Aug 2004 20:57:56 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33)
	id 1Bvmrq-0006cy-G2; Fri, 13 Aug 2004 20:57:54 -0400
Received: from [66.33.193.59] (helo=sara.dreamhost.com)
	by monty-python.gnu.org with esmtp (Exim 4.34)
	id 1Bvmnp-00074E-R9; Fri, 13 Aug 2004 20:53:46 -0400
Received: from dexter (c-24-6-46-122.client.comcast.net [24.6.46.122])
	by sara.dreamhost.com (Postfix) with ESMTP
	id 1AACD9F148; Fri, 13 Aug 2004 17:53:45 -0700 (PDT)
Received: by dexter (Postfix, from userid 1000)
	id 76B4539; Fri, 13 Aug 2004 17:53:44 -0700 (PDT)
To: Mark Wielaard <mark@klomp.org>
References: <87wu0acpkx.fsf@gnu.org> <1092412573.1592.84.camel@localhost>
From: Casey Marshall <csm@gnu.org>
Date: Fri, 13 Aug 2004 17:53:44 -0700
In-Reply-To: <1092412573.1592.84.camel@localhost> (Mark Wielaard's message
	of "Fri, 13 Aug 2004 17:56:14 +0200")
Message-ID: <87k6w2wndz.fsf@gnu.org>
User-Agent: Gnus/5.1002 (Gnus v5.10.2) Emacs/21.2 (gnu/linux)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Cc: classpath@gnu.org, classpath-patches@gnu.org, java@gcc.gnu.org
Subject: Re: [cp-patches] Updated: crypto merge patch
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Sat, 14 Aug 2004 00:57:57 -0000

>>>>> "Mark" == Mark Wielaard <mark@klomp.org> writes:

Mark> First javax/net/ssl/HandshakeCompletedEvent.java and
Mark> javax/net/ssl/SSLSession.java use
Mark> javax.security.cert.X509Certificate. I don't know if we even
Mark> want to provide the javax.security.cert package since it is
Mark> obviously not needed by anyone (or is it?)

I would not be surprised if there is code that uses that package, and
won't be (or can't be) updated. I think including it (since it's
already done and complete) is the safest thing to do.

Also, none of these classes are actually marked as deprecated. Sun
just includes a warning that people should use java.security.cert
instead. It's more green in japitools, too ;)

Mark> Second you need to add the new Makefiles in the resources
Mark> subdirectories to the configure.ac file (see below).

Ok, thanks for the heads up.

Mark> With that fixed make distcheck works correctly and the source
Mark> compiles with both jikes and gcj. So please check this in.

Ok. This will probably happen this weekend.

-- 
Casey Marshall || csm@gnu.org



From MAILER-DAEMON Sat Aug 14 07:46:24 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BvwzQ-0002y1-Ef
	for mharc-classpath@gnu.org; Sat, 14 Aug 2004 07:46:24 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BvwzN-0002wd-RO
	for classpath@gnu.org; Sat, 14 Aug 2004 07:46:21 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BvwzL-0002vA-FM
	for classpath@gnu.org; Sat, 14 Aug 2004 07:46:20 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BvwzK-0002tf-Eo
	for classpath@gnu.org; Sat, 14 Aug 2004 07:46:18 -0400
Received: from [82.161.94.186] (helo=gnu.wildebeest.org)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1Bvwuy-0000lv-VC
	for classpath@gnu.org; Sat, 14 Aug 2004 07:41:49 -0400
Received: from slauerhoff.wildebeest.org ([192.168.1.27] ident=mark)
	by gnu.wildebeest.org with esmtp (Exim 3.35 #1 (Debian))
	id 1Bvwuv-0003u1-00
	for <classpath@gnu.org>; Sat, 14 Aug 2004 13:41:45 +0200
From: Mark Wielaard <mark@klomp.org>
To: classpath@gnu.org
In-Reply-To: <87smaqbyg5.fsf@gnu.org>
References: <87wu0acpkx.fsf@gnu.org> <1092412573.1592.84.camel@localhost>
	<411D0007.6050109@netreach.com>  <87smaqbyg5.fsf@gnu.org>
Content-Type: multipart/signed; micalg=pgp-sha1;
	protocol="application/pgp-signature";
	boundary="=-5/95Ae99ua2s5GIK2GQF"
Message-Id: <1092483703.1586.282.camel@localhost>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.4.6 
Date: Sat, 14 Aug 2004 13:41:44 +0200
Subject: Re: [cp-patches] Updated: crypto merge patch
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Sat, 14 Aug 2004 11:46:22 -0000


--=-5/95Ae99ua2s5GIK2GQF
Content-Type: text/plain
Content-Transfer-Encoding: quoted-printable

Hi,

On Fri, 2004-08-13 at 22:00, Casey Marshall wrote:
> Stuart> - Sound (Kaffe scores almost 100% here, some merging needs
> Stuart> doing?)
>=20
> Kaffe gets javax.sound from Tritonus, which is LGPL, so we can't
> include it in Classpath.

We can include it into GNU Classpath. We have an external directory
which contains convenience libraries not directly part of the GNU
Classpath core libraries that are possibly under other (but
GPL-compatible) licenses.
The reason I don't want to do this with Tritonus is because we don't
have a very good track record of following upstream sources.  We
currently only have GNU JAXP as external library shipped with GNU
Classpath, and we are horribly out of date (libgcj is even more out of
date), so I am actually thinking of dropping it and point people to the
official GNU JAXP upstream sources.

It looks like people that create whole development environments out of
GNU Classpath, like kaffe is doing, are in a much better position to do
this kind of integration. We are mostly very busy with hacking on our
own sources. And it gives really independent package implementations a
chance of following their own release cycle.

Cheers,

Mark

--=-5/95Ae99ua2s5GIK2GQF
Content-Type: application/pgp-signature; name=signature.asc
Content-Description: This is a digitally signed message part

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)

iD8DBQBBHfp3xVhZCJWr9QwRAmN5AKCrs4PyFcBBXZVQevQas3REA2UkZgCfX+HH
mu9ow3SRRNR0CH0G+hqsGQc=
=sqr4
-----END PGP SIGNATURE-----

--=-5/95Ae99ua2s5GIK2GQF--




From MAILER-DAEMON Sat Aug 14 08:09:05 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BvxLL-0002Ea-Vd
	for mharc-classpath@gnu.org; Sat, 14 Aug 2004 08:09:04 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BvxLH-0002Bq-73
	for classpath@gnu.org; Sat, 14 Aug 2004 08:08:59 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BvxLE-000299-Kc
	for classpath@gnu.org; Sat, 14 Aug 2004 08:08:56 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33)
	id 1BvxLC-00028d-Rh; Sat, 14 Aug 2004 08:08:54 -0400
Received: from [82.161.94.186] (helo=gnu.wildebeest.org)
	by monty-python.gnu.org with esmtp (Exim 4.34)
	id 1BvxFn-0006sY-12; Sat, 14 Aug 2004 08:03:19 -0400
Received: from slauerhoff.wildebeest.org ([192.168.1.27] ident=mark)
	by gnu.wildebeest.org with esmtp (Exim 3.35 #1 (Debian))
	id 1BvxFj-0003zo-00; Sat, 14 Aug 2004 14:03:15 +0200
From: Mark Wielaard <mark@klomp.org>
To: Casey Marshall <csm@gnu.org>
In-Reply-To: <87k6w2wndz.fsf@gnu.org>
References: <87wu0acpkx.fsf@gnu.org> <1092412573.1592.84.camel@localhost>
	<87k6w2wndz.fsf@gnu.org>
Content-Type: multipart/signed; micalg=pgp-sha1;
	protocol="application/pgp-signature";
	boundary="=-ySdjX3U//CD0jVonBpyR"
Message-Id: <1092484994.1592.304.camel@localhost>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.4.6 
Date: Sat, 14 Aug 2004 14:03:14 +0200
Cc: classpath@gnu.org, classpath-patches@gnu.org, java@gcc.gnu.org
Subject: Re: [cp-patches] Updated: crypto merge patch
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Sat, 14 Aug 2004 12:09:01 -0000


--=-ySdjX3U//CD0jVonBpyR
Content-Type: text/plain
Content-Transfer-Encoding: quoted-printable

Hi,

On Sat, 2004-08-14 at 02:53, Casey Marshall wrote:
> >>>>> "Mark" =3D=3D Mark Wielaard <mark@klomp.org> writes:
>=20
> Mark> First javax/net/ssl/HandshakeCompletedEvent.java and
> Mark> javax/net/ssl/SSLSession.java use
> Mark> javax.security.cert.X509Certificate. I don't know if we even
> Mark> want to provide the javax.security.cert package since it is
> Mark> obviously not needed by anyone (or is it?)
>=20
> I would not be surprised if there is code that uses that package, and
> won't be (or can't be) updated. I think including it (since it's
> already done and complete) is the safest thing to do.
>=20
> Also, none of these classes are actually marked as deprecated. Sun
> just includes a warning that people should use java.security.cert
> instead. It's more green in japitools, too ;)

O, more green in japitool! Yes, lets include it then.
Yes, I see the point, and it looks like even though this is a bit of
code duplication the implementation is so minimal that it will need not
much (any) maintenance at all.
But I still think our own code (HandshakeCompletedEvent and SSLSession)
should use java.security.cert directly and not javax.security.cert.

Cheers,

Mark

--=-ySdjX3U//CD0jVonBpyR
Content-Type: application/pgp-signature; name=signature.asc
Content-Description: This is a digitally signed message part

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)

iD8DBQBBHf+CxVhZCJWr9QwRAsaoAJsHSEcm7AnV26TTOqlY5KucY0dh4QCgkYuW
s+DiYAn7RgHCh6F112/fJ6c=
=xrc/
-----END PGP SIGNATURE-----

--=-ySdjX3U//CD0jVonBpyR--




From MAILER-DAEMON Sat Aug 14 10:23:11 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BvzR8-0005hY-S0
	for mharc-classpath@gnu.org; Sat, 14 Aug 2004 10:23:10 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BvzR7-0005hT-Mu
	for classpath@gnu.org; Sat, 14 Aug 2004 10:23:09 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BvzR6-0005hE-25
	for classpath@gnu.org; Sat, 14 Aug 2004 10:23:09 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BvzR5-0005hA-UA
	for classpath@gnu.org; Sat, 14 Aug 2004 10:23:08 -0400
Received: from [139.19.1.2] (helo=francois.mpi-sb.mpg.de)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BvzMn-0006sc-6z
	for classpath@gnu.org; Sat, 14 Aug 2004 10:18:41 -0400
Received: from amavis by francois.mpi-sb.mpg.de with scanned-ok (Exim 3.35 #1
	(Debian)) id 1BvzMm-0000zs-00
	for <classpath@gnu.org>; Sat, 14 Aug 2004 16:18:40 +0200
Received: from maildist ([139.19.90.238] helo=mpi-sb.mpg.de)
	by francois.mpi-sb.mpg.de with esmtp (Exim 3.35 #1 (Debian))
	id 1BvzMj-0000zk-00; Sat, 14 Aug 2004 16:18:37 +0200
Received: from kaffe.org (topic@mpiat2313 [139.19.20.102])
	by mpi-sb.mpg.de (8.12.9+Sun/8.12.9) with ESMTP id i7EEIawg013001;
	Sat, 14 Aug 2004 16:18:36 +0200 (MEST)
Message-ID: <411E1F3B.9070604@kaffe.org>
Date: Sat, 14 Aug 2004 16:18:35 +0200
From: Dalibor Topic <robilad@kaffe.org>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US;
	rv:1.6) Gecko/20040413 Debian/1.6-5
X-Accept-Language: en
MIME-Version: 1.0
To: Casey Marshall <csm@gnu.org>
References: <87wu0acpkx.fsf@gnu.org>
	<1092412573.1592.84.camel@localhost>	<411D0007.6050109@netreach.com>
	<87smaqbyg5.fsf@gnu.org>
In-Reply-To: <87smaqbyg5.fsf@gnu.org>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
X-Virus-Scanned: by AMaViS perl-11
Cc: Mark Wielaard <mark@klomp.org>, classpath@gnu.org,
	Stuart Ballard <sballard@netreach.com>
Subject: Re: [cp-patches] Updated: crypto merge patch
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Sat, 14 Aug 2004 14:23:09 -0000

Casey Marshall wrote:
>>>>>>"Stuart" == Stuart Ballard <sballard@netreach.com> writes:
> 
> 
> Stuart> The big holes, therefore, are only:
> Stuart> - Swing (obviously heavily in progress)
> Stuart> - CORBA (no work is being done here AFAIK)
> 
> I think there are free implementations of this being done by others. I
> don't know about licensing etc.
> 
> Stuart> - Sound (Kaffe scores almost 100% here, some merging needs
> Stuart> doing?)
> 
> Kaffe gets javax.sound from Tritonus, which is LGPL, so we can't
> include it in Classpath. We could ask them for permission to do so
> under Classpath's license, if they are still around (the news page has
> its latest item in 2003).

Yep, they are still around ;) Michael Pfisterer is on this list, and has 
updated tritonus to 1.5 including the generics bits, afaik. He was one 
of the persons that volunteered to help with the generics part of the 
class libraries.

cheers,
dalibor topic



From MAILER-DAEMON Sat Aug 14 10:28:27 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BvzWF-0006yn-FW
	for mharc-classpath@gnu.org; Sat, 14 Aug 2004 10:28:27 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BvzWC-0006ye-Nx
	for classpath@gnu.org; Sat, 14 Aug 2004 10:28:24 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BvzWC-0006yL-6b
	for classpath@gnu.org; Sat, 14 Aug 2004 10:28:24 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BvzWC-0006yI-21
	for classpath@gnu.org; Sat, 14 Aug 2004 10:28:24 -0400
Received: from [199.232.41.8] (helo=mx20.gnu.org)
	by monty-python.gnu.org with esmtp (TLSv1:DES-CBC3-SHA:168)
	(Exim 4.34) id 1BvzRY-0007de-LK
	for classpath@gnu.org; Sat, 14 Aug 2004 10:23:36 -0400
Received: from [139.19.1.2] (helo=francois.mpi-sb.mpg.de)
	by mx20.gnu.org with esmtp (Exim 4.34) id 1BvzR6-0007Ma-9k
	for classpath@gnu.org; Sat, 14 Aug 2004 10:23:08 -0400
Received: from amavis by francois.mpi-sb.mpg.de with scanned-ok (Exim 3.35 #1
	(Debian)) id 1BvzQM-0001Ei-00
	for <classpath@gnu.org>; Sat, 14 Aug 2004 16:22:22 +0200
Received: from maildist ([139.19.90.238] helo=mpi-sb.mpg.de)
	by francois.mpi-sb.mpg.de with esmtp (Exim 3.35 #1 (Debian))
	id 1BvzQF-0001DX-00; Sat, 14 Aug 2004 16:22:15 +0200
Received: from kaffe.org (topic@mpiat2313 [139.19.20.102])
	by mpi-sb.mpg.de (8.12.9+Sun/8.12.9) with ESMTP id i7EEMFwg014800;
	Sat, 14 Aug 2004 16:22:15 +0200 (MEST)
Message-ID: <411E2017.7080708@kaffe.org>
Date: Sat, 14 Aug 2004 16:22:15 +0200
From: Dalibor Topic <robilad@kaffe.org>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US;
	rv:1.6) Gecko/20040413 Debian/1.6-5
X-Accept-Language: en
MIME-Version: 1.0
To: tromey@redhat.com
References: <87wu0acpkx.fsf@gnu.org>
	<1092412573.1592.84.camel@localhost>	<411D0007.6050109@netreach.com>
	<87657mn0s8.fsf@fleche.redhat.com>
In-Reply-To: <87657mn0s8.fsf@fleche.redhat.com>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
X-Virus-Scanned: by AMaViS perl-11
Cc: Casey Marshall <csm@gnu.org>, Mark Wielaard <mark@klomp.org>,
	classpath@gnu.org, Stuart Ballard <sballard@netreach.com>
Subject: Re: [cp-patches] Updated: crypto merge patch
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Sat, 14 Aug 2004 14:28:25 -0000

Tom Tromey wrote:
>>>>>>"Stuart" == Stuart Ballard <sballard@netreach.com> writes:

> Stuart> - CORBA (no work is being done here AFAIK)
> 
> There is JacORB and OpenORB.  They have some license trouble, but
> nothing really fatal.  We couldn't put them in classpath though.

I have not managed to build JacORB with Kaffe. there is a class missing 
in their org.omg.* packages, unfortunately. Asking on the JacORB mailing 
list about it produced no result, so far. But I guess we'll need to have 
our own org.omg implementation anyway, due to the original 
implementation's license being 'don't touch', i.e. not GPL-compatible.

OpenORB is not an option for kaffe because it is licensed under ASL, so 
I haven't tried building it.

cheers,
dalibor topic



From MAILER-DAEMON Sat Aug 14 15:35:44 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1Bw4Jc-0006s0-6p
	for mharc-classpath@gnu.org; Sat, 14 Aug 2004 15:35:44 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1Bw4Ja-0006rC-5y
	for classpath@gnu.org; Sat, 14 Aug 2004 15:35:42 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1Bw4JZ-0006qw-Kv
	for classpath@gnu.org; Sat, 14 Aug 2004 15:35:41 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33)
	id 1Bw4JX-0006qf-KZ; Sat, 14 Aug 2004 15:35:39 -0400
Received: from [66.33.193.59] (helo=sara.dreamhost.com)
	by monty-python.gnu.org with esmtp (Exim 4.34)
	id 1Bw4Ex-00046q-Ck; Sat, 14 Aug 2004 15:30:55 -0400
Received: from dexter (c-24-6-46-122.client.comcast.net [24.6.46.122])
	by sara.dreamhost.com (Postfix) with ESMTP
	id 7A5599F418; Sat, 14 Aug 2004 12:30:54 -0700 (PDT)
Received: by dexter (Postfix, from userid 1000)
	id A996E39; Sat, 14 Aug 2004 12:30:47 -0700 (PDT)
To: Mark Wielaard <mark@klomp.org>
References: <87wu0acpkx.fsf@gnu.org> <1092412573.1592.84.camel@localhost>
From: Casey Marshall <csm@gnu.org>
Date: Sat, 14 Aug 2004 12:30:47 -0700
In-Reply-To: <1092412573.1592.84.camel@localhost> (Mark Wielaard's message
	of "Fri, 13 Aug 2004 17:56:14 +0200")
Message-ID: <87oeldikk8.fsf@gnu.org>
User-Agent: Gnus/5.1002 (Gnus v5.10.2) Emacs/21.2 (gnu/linux)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Cc: classpath@gnu.org, classpath-patches@gnu.org, java@gcc.gnu.org
Subject: Re: [cp-patches] Updated: crypto merge patch
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Sat, 14 Aug 2004 19:35:42 -0000

>>>>> "Mark" == Mark Wielaard <mark@klomp.org> writes:

Mark> * New javax.crypto, javax.crypto.interfaces, javax.crypto.spec, javax.net,
Mark>   javax.net.ssl, javax.security.auth, javax.security.auth.callback,
Mark>   javax.security.auth.login, javax.security.auth.x500, javax.security.sasl
Mark>   and org.ietf.jgss are now officially part of GNU Classpath.
Mark>   Extra crypto algorithms can be obtained from the GNU Crypto project,
Mark>   a full TLS implementation is provided by the Jessie project.
Mark>   http://www.gnu.org/software/gnu-crypto/
Mark>   http://www.nongnu.org/jessie/

Ok. These classes are in.

Happy hacking.

-- 
Casey Marshall || csm@gnu.org



From MAILER-DAEMON Sun Aug 15 06:12:27 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BwI02-0003ql-Om
	for mharc-classpath@gnu.org; Sun, 15 Aug 2004 06:12:26 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BwI00-0003qf-L9
	for classpath@gnu.org; Sun, 15 Aug 2004 06:12:24 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BwHzy-0003qT-V1
	for classpath@gnu.org; Sun, 15 Aug 2004 06:12:24 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BwHzs-0003qP-UN
	for classpath@gnu.org; Sun, 15 Aug 2004 06:12:22 -0400
Received: from [82.161.94.186] (helo=gnu.wildebeest.org)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BwHvj-000134-3c
	for classpath@gnu.org; Sun, 15 Aug 2004 06:07:59 -0400
Received: from slauerhoff.wildebeest.org ([192.168.1.27] ident=mark)
	by gnu.wildebeest.org with esmtp (Exim 3.35 #1 (Debian))
	id 1BwHvb-0007Za-00; Sun, 15 Aug 2004 12:07:51 +0200
From: Mark Wielaard <mark@klomp.org>
To: tromey@redhat.com
In-Reply-To: <87pt5wp95b.fsf@fleche.redhat.com>
References: <87pt5wp95b.fsf@fleche.redhat.com>
Content-Type: multipart/signed; micalg=pgp-sha1;
	protocol="application/pgp-signature";
	boundary="=-KR02DJWVG9oDItP2kOR8"
Message-Id: <1092564468.1591.450.camel@localhost>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.4.6 
Date: Sun, 15 Aug 2004 12:07:48 +0200
Cc: GNU Classpath Project <classpath@gnu.org>
Subject: Re: VMClassLoader
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Sun, 15 Aug 2004 10:12:24 -0000


--=-KR02DJWVG9oDItP2kOR8
Content-Type: text/plain
Content-Transfer-Encoding: quoted-printable

Hi,

On Thu, 2004-08-12 at 19:18, Tom Tromey wrote:
> I notice that VMClassLoader still has this:
>=20
>    * <strong>For backward compatibility, this just ignores the protection
>    * domain; that is the wrong behavior, and you should directly implemen=
t
>    * this method natively if you can.</strong>
>=20
> I suggest we pick a flag day to switch this.  In particular I think we
> ought to simply remove the old form of defineClass() and only have the
> form that accepts a ProtectionDomain.

I agree. And a since the old form of defineClass() isn't even used from
the rest of GNU Claspath I think we can remove it immediatly. Runtimes
will probably have provided their own VMClassLoader implementation
already.

If nobody objects I want to commit the attached patch next week.

2004-08-14  Mark Wielaard  <mark@klomp.org>

        * vm/reference/java/lang/VMClassLoader.java (defineClass): Removed
        version that didn't take a ProtectionDomain.

Cheers,

Mark

--=-KR02DJWVG9oDItP2kOR8
Content-Type: application/pgp-signature; name=signature.asc
Content-Description: This is a digitally signed message part

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)

iD8DBQBBHzX0xVhZCJWr9QwRApCmAJ40Aa/Jqm29DZG6SlYgy26AyJOjjQCdFHYv
jPCNB32CnfXhQN07qCiunRY=
=Ti82
-----END PGP SIGNATURE-----

--=-KR02DJWVG9oDItP2kOR8--




From MAILER-DAEMON Sun Aug 15 06:25:26 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BwICb-000633-5I
	for mharc-classpath@gnu.org; Sun, 15 Aug 2004 06:25:25 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BwICX-00062r-HQ
	for classpath@gnu.org; Sun, 15 Aug 2004 06:25:21 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BwICW-000623-7R
	for classpath@gnu.org; Sun, 15 Aug 2004 06:25:20 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BwICW-00061z-5e
	for classpath@gnu.org; Sun, 15 Aug 2004 06:25:20 -0400
Received: from [82.161.94.186] (helo=gnu.wildebeest.org)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BwI8I-0002aU-NS
	for classpath@gnu.org; Sun, 15 Aug 2004 06:20:59 -0400
Received: from slauerhoff.wildebeest.org ([192.168.1.27] ident=mark)
	by gnu.wildebeest.org with esmtp (Exim 3.35 #1 (Debian))
	id 1BwI86-0007av-00; Sun, 15 Aug 2004 12:20:46 +0200
From: Mark Wielaard <mark@klomp.org>
To: tromey@redhat.com
In-Reply-To: <1092564468.1591.450.camel@localhost>
References: <87pt5wp95b.fsf@fleche.redhat.com>
	<1092564468.1591.450.camel@localhost>
Content-Type: multipart/signed; micalg=pgp-sha1;
	protocol="application/pgp-signature";
	boundary="=-NdSqZn+J0NOjfBqBE3uu"
Message-Id: <1092565243.1591.459.camel@localhost>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.4.6 
Date: Sun, 15 Aug 2004 12:20:43 +0200
Cc: GNU Classpath Project <classpath@gnu.org>
Subject: Re: VMClassLoader
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Sun, 15 Aug 2004 10:25:22 -0000


--=-NdSqZn+J0NOjfBqBE3uu
Content-Type: multipart/mixed; boundary="=-HPapk/vwo8D0fbjp/vYV"


--=-HPapk/vwo8D0fbjp/vYV
Content-Type: text/plain
Content-Transfer-Encoding: quoted-printable

Hi,

On Sun, 2004-08-15 at 12:07, Mark Wielaard wrote:
> If nobody objects I want to commit the attached patch next week.
>=20
> 2004-08-14  Mark Wielaard  <mark@klomp.org>
>=20
>         * vm/reference/java/lang/VMClassLoader.java (defineClass): Remove=
d
>         version that didn't take a ProtectionDomain.

That would be the attached patch.

Cheers,

Mark

--=-HPapk/vwo8D0fbjp/vYV
Content-Disposition: inline; filename=vmclassloader.patch
Content-Transfer-Encoding: base64
Content-Type: text/x-patch; name=vmclassloader.patch; charset=ANSI_X3.4-1968
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--=-HPapk/vwo8D0fbjp/vYV--

--=-NdSqZn+J0NOjfBqBE3uu
Content-Type: application/pgp-signature; name=signature.asc
Content-Description: This is a digitally signed message part

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)

iD8DBQBBHzj7xVhZCJWr9QwRAhK4AJ4+a3yKQM8fsBGEqW0yny3wSc3nPQCeMslg
eDQ/DyGg/7ipXkoBc9oTXNs=
=dRI2
-----END PGP SIGNATURE-----

--=-NdSqZn+J0NOjfBqBE3uu--




From MAILER-DAEMON Sun Aug 15 08:32:53 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BwKBx-0005oF-QH
	for mharc-classpath@gnu.org; Sun, 15 Aug 2004 08:32:53 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BwKBv-0005ni-LO
	for classpath@gnu.org; Sun, 15 Aug 2004 08:32:52 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BwKBt-0005n1-U8
	for classpath@gnu.org; Sun, 15 Aug 2004 08:32:50 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BwKBt-0005mr-RZ
	for classpath@gnu.org; Sun, 15 Aug 2004 08:32:49 -0400
Received: from [64.233.170.206] (helo=mproxy.gmail.com)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BwK79-0000p5-Mg
	for classpath@gnu.org; Sun, 15 Aug 2004 08:27:55 -0400
Received: by mproxy.gmail.com with SMTP id 73so84391rnl
	for <classpath@gnu.org>; Sun, 15 Aug 2004 05:27:54 -0700 (PDT)
Received: by 10.38.72.37 with SMTP id u37mr58824rna;
	Sun, 15 Aug 2004 05:27:54 -0700 (PDT)
Message-ID: <d58509a804081505275d57f873@mail.gmail.com>
Date: Sun, 15 Aug 2004 08:27:54 -0400
From: Robert Lougher <rob.lougher@gmail.com>
To: Mark Wielaard <mark@klomp.org>
In-Reply-To: <1092564468.1591.450.camel@localhost>
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
References: <87pt5wp95b.fsf@fleche.redhat.com>
	<1092564468.1591.450.camel@localhost>
Cc: tromey@redhat.com, GNU Classpath Project <classpath@gnu.org>
Subject: Re: VMClassLoader
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
Reply-To: Robert Lougher <rob.lougher@gmail.com>
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Sun, 15 Aug 2004 12:32:52 -0000

On Sun, 15 Aug 2004 12:07:48 +0200, Mark Wielaard <mark@klomp.org> wrote:
> Hi,
> 
> On Thu, 2004-08-12 at 19:18, Tom Tromey wrote:
> > I notice that VMClassLoader still has this:
> >
> >    * <strong>For backward compatibility, this just ignores the protection
> >    * domain; that is the wrong behavior, and you should directly implement
> >    * this method natively if you can.</strong>
> >
> > I suggest we pick a flag day to switch this.  In particular I think we
> > ought to simply remove the old form of defineClass() and only have the
> > form that accepts a ProtectionDomain.
> 
> I agree. And a since the old form of defineClass() isn't even used from
> the rest of GNU Claspath I think we can remove it immediatly. Runtimes
> will probably have provided their own VMClassLoader implementation
> already.
> 

Hi All,

FYI, JamVM implements the version with the ProtectionDomain natively,
as suggested in the comments.  I also implement getPrimitiveClass
natively with a type char, rather than relying on the Java method that
converts it to a string (e.g. D->double).  This method is also marked
as destined to be removed.

Would it be a good idea to tidy up both at the same time?  Of course,
the class would still need modifications by the runtime because
loadClass/getResource, etc. are stubs which must be implemented.

Rob.

P.S.  I'm going by the Classpath-0.10 release snapshot, and not CVS.

> If nobody objects I want to commit the attached patch next week.
> 
> 2004-08-14  Mark Wielaard  <mark@klomp.org>
> 
>         * vm/reference/java/lang/VMClassLoader.java (defineClass): Removed
>         version that didn't take a ProtectionDomain.
> 
> Cheers,
> 
> Mark



From MAILER-DAEMON Sun Aug 15 08:40:57 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BwKJl-00079K-Rr
	for mharc-classpath@gnu.org; Sun, 15 Aug 2004 08:40:57 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BwKJk-00079A-Cf
	for classpath@gnu.org; Sun, 15 Aug 2004 08:40:56 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BwKJj-00078q-Jb
	for classpath@gnu.org; Sun, 15 Aug 2004 08:40:55 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BwKJj-00078g-F7
	for classpath@gnu.org; Sun, 15 Aug 2004 08:40:55 -0400
Received: from [217.12.12.199] (helo=smtp809.mail.ukl.yahoo.com)
	by monty-python.gnu.org with smtp (Exim 4.34) id 1BwKEp-00026I-D9
	for classpath@gnu.org; Sun, 15 Aug 2004 08:35:52 -0400
Received: from unknown (HELO rivendell) (classpath@gnu.org@81.152.170.101 with
	poptime)
	by smtp809.mail.ukl.yahoo.com with SMTP; 15 Aug 2004 12:35:50 -0000
Received: from localhost ([127.0.0.1]) by rivendell with esmtp (Exim 4.34)
	id 1BwKEo-0006Tn-0k; Sun, 15 Aug 2004 13:35:50 +0100
From: Andrew John Hughes <gnu_andrew@member.fsf.org>
To: classpath-patches@gnu.org
In-Reply-To: <1092521725.1591.437.camel@localhost>
References: <1092521725.1591.437.camel@localhost>
Content-Type: multipart/signed; micalg=pgp-sha1;
	protocol="application/pgp-signature";
	boundary="=-ZQtb6DNhmZgF6i5BPXPn"
Message-Id: <1092573349.2352.1.camel@rivendell>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.4.6 
Date: Sun, 15 Aug 2004 13:35:49 +0100
Cc: GNU Classpath <classpath@gnu.org>
Subject: Re: [cp-patches] Add JSpinner and friends added
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
Reply-To: gnu_andrew@member.fsf.org
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Sun, 15 Aug 2004 12:40:56 -0000


--=-ZQtb6DNhmZgF6i5BPXPn
Content-Type: text/plain
Content-Transfer-Encoding: quoted-printable

On Sat, 2004-08-14 at 23:15, Mark Wielaard wrote:
> Hi,
>=20
> This finally adds JSpinner and friends that Ka-Hing Cheung implemented.
>=20
> 2004-08-14  Ka-Hing Cheung  <kahing@javabsp.org>
>=20
>        * javax/swing/AbstractSpinnerModel.java,
>        javax/swing/JSpinner.java,
>        javax/swing/SpinnerNumberModel.java,
>        javax/swing/plaf/basic/BasicSpinnerUI.java:
>        New files.
>        * javax/swing/plaf/basic/BasicLookAndFeel.java
>        (initClassDefaults): Added defaults for BasicSpinnerUI.
>=20
> And to make sure people can play with it I added a little example to our
> swing example Demo class:
>=20
> 2004-08-14  Mark Wielaard  <mark@klomp.org>
>=20
>         * examples/gnu/classpath/examples/swing/Demo.java (mkSpinner):
>         New method.
>         (mkButtonBar): Add Spinner.
>=20
> Committed both.
>=20
> Sorry, Ka-Hing, I still haven't looked at the Timer patch.
>=20
> Anybody wants to implement SpinnerDateModel and SpinnerListModel?
>=20
> Cheers,
>=20
> Mark

If no-one else is already working on them, I'd like to take these two
on.

Cheers,
--=20
Andrew :-)
=20
Please avoid sending me Word or PowerPoint attachments.
See http://www.fsf.org/philosophy/no-word-attachments.html
=20
Value your freedom, or you will lose it, teaches history.
`Don't bother us with politics' respond those who don't want to learn.
=20


--=-ZQtb6DNhmZgF6i5BPXPn
Content-Type: application/pgp-signature; name=signature.asc
Content-Description: This is a digitally signed message part

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)

iD8DBQBBH1ilu+fW6pI0dq4RAgCLAJ49OZqwa00Sd6jyDLWopXn3ufDoKQCeJVmV
L0R6aqqKsGR9MgLLpFn5cS4=
=dCdj
-----END PGP SIGNATURE-----

--=-ZQtb6DNhmZgF6i5BPXPn--




From MAILER-DAEMON Sun Aug 15 21:20:10 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BwWAU-0006Pb-K4
	for mharc-classpath@gnu.org; Sun, 15 Aug 2004 21:20:10 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BwWAS-0006Nv-8m
	for classpath@gnu.org; Sun, 15 Aug 2004 21:20:08 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BwWAR-0006NR-8h
	for classpath@gnu.org; Sun, 15 Aug 2004 21:20:07 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33)
	id 1BwWAN-0006N6-OV; Sun, 15 Aug 2004 21:20:04 -0400
Received: from [66.33.211.27] (helo=pippin.dreamhost.com)
	by monty-python.gnu.org with esmtp (Exim 4.34)
	id 1BwW5u-0004IG-Nf; Sun, 15 Aug 2004 21:15:26 -0400
Received: from struggle.javabsp (adsl-69-105-158-127.dsl.pltn13.pacbell.net
	[69.105.158.127]) by pippin.dreamhost.com (Postfix) with ESMTP
	id 58CE716D203; Sun, 15 Aug 2004 18:15:25 -0700 (PDT)
From: Ka-Hing Cheung <kahing@javabsp.org>
To: gnu_andrew@member.fsf.org
In-Reply-To: <1092573349.2352.1.camel@rivendell>
References: <1092521725.1591.437.camel@localhost>
	<1092573349.2352.1.camel@rivendell>
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
Message-Id: <1092618897.3835.4.camel@struggle.javabsp>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.4.6 
Date: Sun, 15 Aug 2004 18:15:24 -0700
Cc: GNU Classpath <classpath@gnu.org>, classpath-patches@gnu.org
Subject: Re: [cp-patches] Add JSpinner and friends added
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Mon, 16 Aug 2004 01:20:08 -0000

On Sun, 2004-08-15 at 05:35, Andrew John Hughes wrote:
> > Sorry, Ka-Hing, I still haven't looked at the Timer patch.

No problem. I think there maybe a potential problem with my timer code,
so I will send you an update once I take a look at it again.

> > 
> > Anybody wants to implement SpinnerDateModel and SpinnerListModel?

> If no-one else is already working on them, I'd like to take these two
> on.
> 
> Cheers,

I didn't write any code for it since without JFormattedTextField I
couldn't write the cooresponding editors anyways. Although the editors
are not strictly necessary for the models...

-khc




From MAILER-DAEMON Mon Aug 16 04:53:45 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BwdFQ-0007Zz-AG
	for mharc-classpath@gnu.org; Mon, 16 Aug 2004 04:53:44 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BwdFM-0007Xf-GV
	for classpath@gnu.org; Mon, 16 Aug 2004 04:53:40 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BwdFI-0007Wy-6Y
	for classpath@gnu.org; Mon, 16 Aug 2004 04:53:40 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BwdFH-0007Wv-TJ
	for classpath@gnu.org; Mon, 16 Aug 2004 04:53:36 -0400
Received: from [212.227.126.185] (helo=moutng.kundenserver.de)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BwdAc-00050j-4o
	for classpath@gnu.org; Mon, 16 Aug 2004 04:48:48 -0400
Received: from [212.227.126.207] (helo=mrelayng.kundenserver.de)
	by moutng.kundenserver.de with esmtp (Exim 3.35 #1)
	id 1BwdAa-00013v-00
	for classpath@gnu.org; Mon, 16 Aug 2004 10:48:44 +0200
Received: from [212.126.219.82] (helo=aicas.com)
	by mrelayng.kundenserver.de with asmtp (TLSv1:RC4-MD5:128)
	(Exim 3.35 #1) id 1BwdAZ-0004Dr-00; Mon, 16 Aug 2004 10:48:43 +0200
Message-ID: <412074EA.7090409@aicas.com>
Date: Mon, 16 Aug 2004 10:48:42 +0200
From: "Dr. Torsten Rupp" <rupp@aicas.com>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.4) Gecko/20030624
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: classpath <classpath@gnu.org>
X-Enigmail-Version: 0.76.5.0
X-Enigmail-Supports: pgp-inline, pgp-mime
Content-Type: multipart/mixed; boundary="------------040505090707060700070708"
X-Provags-ID: kundenserver.de abuse@kundenserver.de
	auth:f9c39743fddc2557f7fa7c1a513f7ea5
Subject: Target native layer patches: SunOS, RTEMS, embOS, MinGW
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Mon, 16 Aug 2004 08:53:41 -0000

This is a multi-part message in MIME format.
--------------040505090707060700070708
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit

Dear Classpath-members,

as discussed please find below our target native layers for the 
following OSs:

- SunOS
- RTEMS
- embOS
- MinGW

Please take into account that these patches are only for people who like 
to see what aicas is currently using and what was/is the idea of the
target native layer. It will _not_ be integrated like shown in the 
patches (please see discussion about this topic in the other messages). 
I'm currently preparing a "posix"-compliant layer which will replace the 
Linux and Solaris layers (as suggested by Michael).

Sincerely,

Torsten

--------------040505090707060700070708
Content-Type: text/plain;
 name="target-layer-MinGW.patch"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline;
 filename="target-layer-MinGW.patch"

diff -ruN --exclude='.*' --exclude='Entries*' --exclude=Repository --exclude=Root MinGW/target_native.h MinGW-new/target_native.h
--- MinGW/target_native.h	Thu Jan  1 01:00:00 1970
+++ MinGW-new/target_native.h	Thu Apr  1 12:33:05 2004
@@ -0,0 +1,78 @@
+/* ???.h - ???
+   Copyright (C) 1998 Free Software Foundation, Inc.
+
+This file is part of GNU Classpath.
+
+GNU Classpath is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2, or (at your option)
+any later version.
+ 
+GNU Classpath is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Classpath; see the file COPYING.  If not, write to the
+Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
+02111-1307 USA.
+
+Linking this library statically or dynamically with other modules is
+making a combined work based on this library.  Thus, the terms and
+conditions of the GNU General Public License cover the whole
+combination.
+
+As a special exception, the copyright holders of this library give you
+permission to link this library with independent modules to produce an
+executable, regardless of the license terms of these independent
+modules, and to copy and distribute the resulting executable under
+terms of your choice, provided that you also meet, for each linked
+independent module, the terms and conditions of the license of that
+module.  An independent module is a module which is not derived from
+or based on this library.  If you modify this library, you may extend
+this exception to your version of the library, but you are not
+obligated to do so.  If you do not wish to do so, delete this
+exception statement from your version. */
+
+/*
+Description: MinGW target global defintions
+Systems    : all
+*/
+
+#ifndef __TARGET_NATIVE__
+#define __TARGET_NATIVE__
+
+/****************************** Includes *******************************/
+/* do not move; needed here because of some macro definitions */
+#include <config.h>
+
+#include <stdlib.h>
+
+/****************** Conditional compilation switches *******************/
+
+/***************************** Constants *******************************/
+
+/***************************** Datatypes *******************************/
+
+/***************************** Variables *******************************/
+
+/****************************** Macros *********************************/
+
+/***************************** Functions *******************************/
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#ifdef __cplusplus
+}
+#endif
+
+/* include rest of definitions from generic file (do not move it to 
+   another position!) */
+#include "target_generic.h"
+
+#endif /* __TARGET_NATIVE__ */
+
+/* end of file */
diff -ruN --exclude='.*' --exclude='Entries*' --exclude=Repository --exclude=Root MinGW/target_native_file.h MinGW-new/target_native_file.h
--- MinGW/target_native_file.h	Thu Jan  1 01:00:00 1970
+++ MinGW-new/target_native_file.h	Thu Apr  1 12:33:00 2004
@@ -0,0 +1,101 @@
+/* ???.h - ???
+   Copyright (C) 1998 Free Software Foundation, Inc.
+
+This file is part of GNU Classpath.
+
+GNU Classpath is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2, or (at your option)
+any later version.
+ 
+GNU Classpath is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Classpath; see the file COPYING.  If not, write to the
+Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
+02111-1307 USA.
+
+Linking this library statically or dynamically with other modules is
+making a combined work based on this library.  Thus, the terms and
+conditions of the GNU General Public License cover the whole
+combination.
+
+As a special exception, the copyright holders of this library give you
+permission to link this library with independent modules to produce an
+executable, regardless of the license terms of these independent
+modules, and to copy and distribute the resulting executable under
+terms of your choice, provided that you also meet, for each linked
+independent module, the terms and conditions of the license of that
+module.  An independent module is a module which is not derived from
+or based on this library.  If you modify this library, you may extend
+this exception to your version of the library, but you are not
+obligated to do so.  If you do not wish to do so, delete this
+exception statement from your version. */
+
+/*
+Description: MinGW target defintions of file functions
+Systems    : all
+*/
+
+#ifndef __TARGET_NATIVE_FILE__
+#define __TARGET_NATIVE_FILE__
+
+/****************************** Includes *******************************/
+/* do not move; needed here because of some macro definitions */
+#include <config.h>
+
+#include <stdlib.h>
+#include <sys/stat.h>
+#include <io.h>
+#include <windows.h>
+#include <winbase.h>
+
+#include "target_native.h"
+
+/****************** Conditional compilation switches *******************/
+
+/***************************** Constants *******************************/
+#define TARGET_NATIVE_FILE_FILEFLAG_SYNC  0
+#define TARGET_NATIVE_FILE_FILEFLAG_DSYNC 0
+
+#define TARGET_NATIVE_FILE_FILEPERMISSION_NORMAL (S_IRUSR | S_IWUSR)
+
+#define TARGET_NATIVE_FILE_FILEPERMISSION_PRIVATE (S_IRUSR | S_IWUSR)
+
+#define TARGET_NATIVE_FILE_FILEPERMISSION_READONLY (~(S_IWRITE))
+
+/***************************** Datatypes *******************************/
+
+/***************************** Variables *******************************/
+
+/****************************** Macros *********************************/
+#define TARGET_NATIVE_FILE_TRUNCATE(filedescriptor,offset,result) \
+  do { \
+    result=(chsize(filedescriptor,offset)==0)?TARGET_NATIVE_OK:TARGET_NATIVE_ERROR; \
+  } while (0)
+
+#define TARGET_NATIVE_FILE_MAKE_DIR(name,result) \
+  do { \
+    result=((mkdir(name)==0)?TARGET_NATIVE_OK:TARGET_NATIVE_ERROR); \
+  } while (0)
+
+/***************************** Functions *******************************/
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#ifdef __cplusplus
+}
+#endif
+
+/* include rest of definitions from generic file (do not move it to 
+   another position!) */
+#include "target_generic_file.h"
+
+#endif /* __TARGET_NATIVE_FILE__ */
+
+/* end of file */
diff -ruN --exclude='.*' --exclude='Entries*' --exclude=Repository --exclude=Root MinGW/target_native_io.h MinGW-new/target_native_io.h
--- MinGW/target_native_io.h	Thu Jan  1 01:00:00 1970
+++ MinGW-new/target_native_io.h	Thu Jun 24 12:53:39 2004
@@ -0,0 +1,85 @@
+/* generic_math_int64.h - Native methods for 64bit math operations
+   Copyright (C) 1998 Free Software Foundation, Inc.
+
+This file is part of GNU Classpath.
+
+GNU Classpath is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2, or (at your option)
+any later version.
+ 
+GNU Classpath is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Classpath; see the file COPYING.  If not, write to the
+Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
+02111-1307 USA.
+
+Linking this library statically or dynamically with other modules is
+making a combined work based on this library.  Thus, the terms and
+conditions of the GNU General Public License cover the whole
+combination.
+
+As a special exception, the copyright holders of this library give you
+permission to link this library with independent modules to produce an
+executable, regardless of the license terms of these independent
+modules, and to copy and distribute the resulting executable under
+terms of your choice, provided that you also meet, for each linked
+independent module, the terms and conditions of the license of that
+module.  An independent module is a module which is not derived from
+or based on this library.  If you modify this library, you may extend
+this exception to your version of the library, but you are not
+obligated to do so.  If you do not wish to do so, delete this
+exception statement from your version. */
+
+/*
+Description: MinGW target defintions of I/O functions
+Systems    : all
+*/
+
+#ifndef __TARGET_NATIVE_IO__
+#define __TARGET_NATIVE_IO__
+
+/* check if target_native_io.h included */
+#ifndef __TARGET_NATIVE_IO__
+  #error Do NOT INCLUDE generic target files! Include the corresponding native target files instead!
+#endif
+
+/****************************** Includes *******************************/
+/* do not move; needed here because of some macro definitions */
+#include "config.h"
+
+#include <stdlib.h>
+
+#include "target_native.h"
+
+/****************** Conditional compilation switches *******************/
+
+/***************************** Constants *******************************/
+
+/***************************** Datatypes *******************************/
+
+/***************************** Variables *******************************/
+
+/****************************** Macros *********************************/
+
+/***************************** Functions *******************************/
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#ifdef __cplusplus
+}
+#endif
+
+/* include rest of definitions from generic file (do not move it to 
+   another position!) */
+#include "target_generic_io.h"
+
+#endif /* __TARGET_NATIVE_IO__ */
+
+/* end of file */
diff -ruN --exclude='.*' --exclude='Entries*' --exclude=Repository --exclude=Root MinGW/target_native_math_float.h MinGW-new/target_native_math_float.h
--- MinGW/target_native_math_float.h	Thu Jan  1 01:00:00 1970
+++ MinGW-new/target_native_math_float.h	Thu Apr  1 12:33:07 2004
@@ -0,0 +1,81 @@
+/* ???.h - ???
+   Copyright (C) 1998 Free Software Foundation, Inc.
+
+This file is part of GNU Classpath.
+
+GNU Classpath is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2, or (at your option)
+any later version.
+ 
+GNU Classpath is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Classpath; see the file COPYING.  If not, write to the
+Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
+02111-1307 USA.
+
+Linking this library statically or dynamically with other modules is
+making a combined work based on this library.  Thus, the terms and
+conditions of the GNU General Public License cover the whole
+combination.
+
+As a special exception, the copyright holders of this library give you
+permission to link this library with independent modules to produce an
+executable, regardless of the license terms of these independent
+modules, and to copy and distribute the resulting executable under
+terms of your choice, provided that you also meet, for each linked
+independent module, the terms and conditions of the license of that
+module.  An independent module is a module which is not derived from
+or based on this library.  If you modify this library, you may extend
+this exception to your version of the library, but you are not
+obligated to do so.  If you do not wish to do so, delete this
+exception statement from your version. */
+
+/*
+Description: MinGW target defintions of float/double constants/
+             macros/functions
+Systems    : all
+*/
+
+#ifndef __TARGET_NATIVE_MATH_FLOAT__
+#define __TARGET_NATIVE_MATH_FLOAT__
+
+/****************************** Includes *******************************/
+/* do not move; needed here because of some macro definitions */
+#include <config.h>
+
+#include <stdlib.h>
+
+#include "target_native.h"
+
+/****************** Conditional compilation switches *******************/
+
+/***************************** Constants *******************************/
+
+/***************************** Datatypes *******************************/
+
+/***************************** Variables *******************************/
+
+/****************************** Macros *********************************/
+
+/***************************** Functions *******************************/
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#ifdef __cplusplus
+}
+#endif
+
+/* include rest of definitions from generic file (do not move it to 
+   another position!) */
+#include "target_generic_math_float.h"
+
+#endif /* __TARGET_NATIVE_MATH_FLOAT__ */
+
+/* end of file */
diff -ruN --exclude='.*' --exclude='Entries*' --exclude=Repository --exclude=Root MinGW/target_native_math_int.h MinGW-new/target_native_math_int.h
--- MinGW/target_native_math_int.h	Thu Jan  1 01:00:00 1970
+++ MinGW-new/target_native_math_int.h	Thu Apr  1 12:32:51 2004
@@ -0,0 +1,81 @@
+/* ???.h - ???
+   Copyright (C) 1998 Free Software Foundation, Inc.
+
+This file is part of GNU Classpath.
+
+GNU Classpath is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2, or (at your option)
+any later version.
+ 
+GNU Classpath is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Classpath; see the file COPYING.  If not, write to the
+Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
+02111-1307 USA.
+
+Linking this library statically or dynamically with other modules is
+making a combined work based on this library.  Thus, the terms and
+conditions of the GNU General Public License cover the whole
+combination.
+
+As a special exception, the copyright holders of this library give you
+permission to link this library with independent modules to produce an
+executable, regardless of the license terms of these independent
+modules, and to copy and distribute the resulting executable under
+terms of your choice, provided that you also meet, for each linked
+independent module, the terms and conditions of the license of that
+module.  An independent module is a module which is not derived from
+or based on this library.  If you modify this library, you may extend
+this exception to your version of the library, but you are not
+obligated to do so.  If you do not wish to do so, delete this
+exception statement from your version. */
+
+/*
+Description: MinGW target defintions of int/int64 constants/
+             macros/functions
+Systems    : all
+*/
+
+#ifndef __TARGET_NATIVE_MATH_INT__
+#define __TARGET_NATIVE_MATH_INT__
+
+/****************************** Includes *******************************/
+/* do not move; needed here because of some macro definitions */
+#include <config.h>
+
+#include <stdlib.h>
+
+#include "target_native.h"
+
+/****************** Conditional compilation switches *******************/
+
+/***************************** Constants *******************************/
+
+/***************************** Datatypes *******************************/
+
+/***************************** Variables *******************************/
+
+/****************************** Macros *********************************/
+
+/***************************** Functions *******************************/
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#ifdef __cplusplus
+}
+#endif
+
+/* include rest of definitions from generic file (do not move it to 
+   another position!) */
+#include "target_generic_math_int.h"
+
+#endif /* __TARGET_NATIVE_MATH_INT__ */
+
+/* end of file */
diff -ruN --exclude='.*' --exclude='Entries*' --exclude=Repository --exclude=Root MinGW/target_native_memory.h MinGW-new/target_native_memory.h
--- MinGW/target_native_memory.h	Thu Jan  1 01:00:00 1970
+++ MinGW-new/target_native_memory.h	Thu Apr  1 12:27:13 2004
@@ -0,0 +1,78 @@
+/* ???.h - ???
+   Copyright (C) 1998 Free Software Foundation, Inc.
+
+This file is part of GNU Classpath.
+
+GNU Classpath is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2, or (at your option)
+any later version.
+ 
+GNU Classpath is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Classpath; see the file COPYING.  If not, write to the
+Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
+02111-1307 USA.
+
+Linking this library statically or dynamically with other modules is
+making a combined work based on this library.  Thus, the terms and
+conditions of the GNU General Public License cover the whole
+combination.
+
+As a special exception, the copyright holders of this library give you
+permission to link this library with independent modules to produce an
+executable, regardless of the license terms of these independent
+modules, and to copy and distribute the resulting executable under
+terms of your choice, provided that you also meet, for each linked
+independent module, the terms and conditions of the license of that
+module.  An independent module is a module which is not derived from
+or based on this library.  If you modify this library, you may extend
+this exception to your version of the library, but you are not
+obligated to do so.  If you do not wish to do so, delete this
+exception statement from your version. */
+
+/*
+Description: MinGW target defintions of memory functions
+Systems    : all
+*/
+
+#ifndef __TARGET_NATIVE_MEMORY__
+#define __TARGET_NATIVE_MEMORY__
+
+/****************************** Includes *******************************/
+/* do not move; needed here because of some macro definitions */
+#include <config.h>
+
+#include <stdlib.h>
+
+/****************** Conditional compilation switches *******************/
+
+/***************************** Constants *******************************/
+
+/***************************** Datatypes *******************************/
+
+/***************************** Variables *******************************/
+
+/****************************** Macros *********************************/
+
+/***************************** Functions *******************************/
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#ifdef __cplusplus
+}
+#endif
+
+/* include rest of definitions from generic file (do not move it to 
+   another position!) */
+#include "target_generic_memory.h"
+
+#endif /* __TARGET_NATIVE_MEMORY__ */
+
+/* end of file */
diff -ruN --exclude='.*' --exclude='Entries*' --exclude=Repository --exclude=Root MinGW/target_native_misc.h MinGW-new/target_native_misc.h
--- MinGW/target_native_misc.h	Thu Jan  1 01:00:00 1970
+++ MinGW-new/target_native_misc.h	Thu Apr  1 12:32:55 2004
@@ -0,0 +1,80 @@
+/* ???.h - ???
+   Copyright (C) 1998 Free Software Foundation, Inc.
+
+This file is part of GNU Classpath.
+
+GNU Classpath is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2, or (at your option)
+any later version.
+ 
+GNU Classpath is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Classpath; see the file COPYING.  If not, write to the
+Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
+02111-1307 USA.
+
+Linking this library statically or dynamically with other modules is
+making a combined work based on this library.  Thus, the terms and
+conditions of the GNU General Public License cover the whole
+combination.
+
+As a special exception, the copyright holders of this library give you
+permission to link this library with independent modules to produce an
+executable, regardless of the license terms of these independent
+modules, and to copy and distribute the resulting executable under
+terms of your choice, provided that you also meet, for each linked
+independent module, the terms and conditions of the license of that
+module.  An independent module is a module which is not derived from
+or based on this library.  If you modify this library, you may extend
+this exception to your version of the library, but you are not
+obligated to do so.  If you do not wish to do so, delete this
+exception statement from your version. */
+
+/*
+Description: MinGW target defintions of miscellaneous functions
+Systems    : all
+*/
+
+#ifndef __TARGET_NATIVE_MISC__
+#define __TARGET_NATIVE_MISC__
+
+/****************************** Includes *******************************/
+/* do not move; needed here because of some macro definitions */
+#include <config.h>
+
+#include <stdlib.h>
+
+#include "target_native.h"
+
+/****************** Conditional compilation switches *******************/
+
+/***************************** Constants *******************************/
+
+/***************************** Datatypes *******************************/
+
+/***************************** Variables *******************************/
+
+/****************************** Macros *********************************/
+
+/***************************** Functions *******************************/
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#ifdef __cplusplus
+}
+#endif
+
+/* include rest of definitions from generic file (do not move it to 
+   another position!) */
+#include "target_generic_misc.h"
+
+#endif /* __TARGET_NATIVE_MISC__ */
+
+/* end of file */
diff -ruN --exclude='.*' --exclude='Entries*' --exclude=Repository --exclude=Root MinGW/target_native_network.h MinGW-new/target_native_network.h
--- MinGW/target_native_network.h	Thu Jan  1 01:00:00 1970
+++ MinGW-new/target_native_network.h	Tue Apr 27 14:02:52 2004
@@ -0,0 +1,503 @@
+/* ???.h - ???
+   Copyright (C) 1998 Free Software Foundation, Inc.
+
+This file is part of GNU Classpath.
+
+GNU Classpath is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2, or (at your option)
+any later version.
+ 
+GNU Classpath is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Classpath; see the file COPYING.  If not, write to the
+Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
+02111-1307 USA.
+
+Linking this library statically or dynamically with other modules is
+making a combined work based on this library.  Thus, the terms and
+conditions of the GNU General Public License cover the whole
+combination.
+
+As a special exception, the copyright holders of this library give you
+permission to link this library with independent modules to produce an
+executable, regardless of the license terms of these independent
+modules, and to copy and distribute the resulting executable under
+terms of your choice, provided that you also meet, for each linked
+independent module, the terms and conditions of the license of that
+module.  An independent module is a module which is not derived from
+or based on this library.  If you modify this library, you may extend
+this exception to your version of the library, but you are not
+obligated to do so.  If you do not wish to do so, delete this
+exception statement from your version. */
+
+/*
+Description: MinGW target defintions of network functions
+Systems    : all
+*/
+
+#ifndef __TARGET_NATIVE_NETWORK__
+#define __TARGET_NATIVE_NETWORK__
+
+/****************************** Includes *******************************/
+/* do not move; needed here because of some macro definitions */
+#include <config.h>
+
+#include <stdlib.h>
+#include <windows.h>
+#include <winbase.h>
+#include <winsock.h>
+
+#include "target_native.h"
+
+/****************** Conditional compilation switches *******************/
+
+/***************************** Constants *******************************/
+
+/***************************** Datatypes *******************************/
+
+/***************************** Variables *******************************/
+
+/****************************** Macros *********************************/
+
+/* Hint: target generic macros can not be used, because of completely
+   different header files and some different datatypes
+*/
+
+#define TARGET_NATIVE_NETWORK_GET_HOSTNAME_BY_ADDRESS(address,name,maxNameLen,result) \
+  do { \
+    int            __networkAddress; \
+    struct hostent *__hostEntry; \
+    \
+    __networkAddress=htonl(address); \
+    __hostEntry = gethostbyaddr((char*)&__networkAddress,sizeof(__networkAddress),AF_INET); \
+    if (__hostEntry!=NULL) \
+    { \
+      strncpy(name,__hostEntry->h_name,maxNameLen-1); \
+      name[maxNameLen]='\0'; \
+      result=TARGET_NATIVE_OK; \
+    } \
+    else \
+    { \
+      result=TARGET_NATIVE_ERROR; \
+    } \
+  } while (0)
+#define TARGET_NATIVE_NETWORK_GET_HOSTNAME_BY_NAME(name,addresses,maxAddressSize,addressCount,result) \
+  do { \
+    struct hostent *__hostEntry; \
+    \
+    addressCount = 0; \
+    __hostEntry = gethostbyname(name); \
+    if (__hostEntry != NULL) \
+    { \
+      while ((addressCount<maxAddressSize) && (__hostEntry->h_addr_list[addressCount]!=NULL)) \
+      { \
+        addresses[addressCount]=ntohl(*(int*)(__hostEntry->h_addr_list[addressCount])); \
+        addressCount++; \
+      } \
+      result=TARGET_NATIVE_OK; \
+    } \
+    else \
+    { \
+        result=TARGET_NATIVE_ERROR; \
+    } \
+  } while (0)
+
+#define TARGET_NATIVE_NETWORK_SOCKET_OPEN_STREAM(socketDescriptor,result) \
+  do { \
+    SOCKET __socket; \
+    \
+    socketDescriptor = 0; \
+    \
+    __socket = socket(AF_INET,SOCK_STREAM,0); \
+    if (__socket != INVALID_SOCKET) \
+    { \
+      socketDescriptor = (int)__socket; \
+      result = TARGET_NATIVE_OK; \
+    } \
+    else \
+    { \
+      result = TARGET_NATIVE_ERROR; \
+    } \
+  } while (0)
+#define TARGET_NATIVE_NETWORK_SOCKET_OPEN_DATAGRAM(socketDescriptor,result) \
+  do { \
+    SOCKET __socket; \
+    \
+    socketDescriptor = 0; \
+    \
+    __socket = socket(AF_INET,SOCK_DGRAM,0); \
+    if (__socket != INVALID_SOCKET) \
+    { \
+      socketDescriptor = (int)__socket; \
+      result = TARGET_NATIVE_OK; \
+    } \
+    else \
+    { \
+      result = TARGET_NATIVE_ERROR; \
+    } \
+  } while (0)
+#define TARGET_NATIVE_NETWORK_SOCKET_CLOSE(socketDescriptor,result) \
+  do { \
+    result=(closesocket(socketDescriptor)==0)?TARGET_NATIVE_OK:TARGET_NATIVE_ERROR; \
+  } while (0)
+
+#define TARGET_NATIVE_NETWORK_SOCKET_CONNECT(socketDescriptor,address,port,result) \
+  do { \
+    struct sockaddr_in __socketAddress; \
+    \
+    memset(&__socketAddress,0,sizeof(__socketAddress)); \
+    __socketAddress.sin_family      = AF_INET; \
+    __socketAddress.sin_addr.s_addr = htonl(address); \
+    __socketAddress.sin_port        = htons(((short)port)); \
+    \
+    result=(connect(socketDescriptor,(const struct sockaddr FAR*)&__socketAddress,sizeof(__socketAddress))==0)?TARGET_NATIVE_OK:TARGET_NATIVE_ERROR; \
+  } while (0)
+#define TARGET_NATIVE_NETWORK_SOCKET_BIND(socketDescriptor,address,port,result) \
+  do { \
+    struct sockaddr_in __socketAddress; \
+    \
+    memset(&__socketAddress,0,sizeof(__socketAddress)); \
+    __socketAddress.sin_family      = AF_INET; \
+    __socketAddress.sin_addr.s_addr = htonl(address); \
+    __socketAddress.sin_port        = htons(((short)port)); \
+    \
+    result=(bind(socketDescriptor,(const struct sockaddr FAR*)&__socketAddress,sizeof(__socketAddress))==0)?TARGET_NATIVE_OK:TARGET_NATIVE_ERROR; \
+  } while (0)
+#define TARGET_NATIVE_NETWORK_SOCKET_LISTEN(socketDescriptor,maxQueueLength,result) \
+  do { \
+    result=(listen(socketDescriptor,maxQueueLength)==0)?TARGET_NATIVE_OK:TARGET_NATIVE_ERROR; \
+  } while (0)
+#define TARGET_NATIVE_NETWORK_SOCKET_ACCEPT(socketDescriptor,newSocketDescriptor,result) \
+  do { \
+    struct sockaddr_in __socketAddress; \
+    int                __socketAddressLength; \
+    \
+    memset(&__socketAddress,0,sizeof(__socketAddress)); \
+    __socketAddressLength=sizeof(__socketAddress); \
+    newSocketDescriptor=accept(socketDescriptor,(struct sockaddr FAR*)&__socketAddress,&__socketAddressLength); \
+    result=(newSocketDescriptor!=INVALID_SOCKET)?TARGET_NATIVE_OK:TARGET_NATIVE_ERROR; \
+  } while (0)
+
+#define TARGET_NATIVE_NETWORK_SOCKET_GET_LOCAL_INFO(socketDescriptor,localAddress,localPort,result) \
+  do { \
+    struct sockaddr_in __socketAddress; \
+    int                __socketAddressLength; \
+    \
+    localAddress=0; \
+    localPort   =0; \
+    \
+    __socketAddressLength=sizeof(__socketAddress); \
+    result=(getsockname(socketDescriptor,(struct sockaddr FAR*)&__socketAddress,&__socketAddressLength)==0)?TARGET_NATIVE_OK:TARGET_NATIVE_ERROR; \
+    if (result==TARGET_NATIVE_OK) \
+    { \
+      assert(__socketAddressLength>=sizeof(__socketAddress)); \
+      localAddress=ntohl(__socketAddress.sin_addr.s_addr); \
+      localPort   =ntohs(__socketAddress.sin_port); \
+    } \
+  } while (0)
+#define TARGET_NATIVE_NETWORK_SOCKET_GET_REMOTE_INFO(socketDescriptor,remoteAddress,remotePort,result) \
+  do { \
+    struct sockaddr_in __socketAddress; \
+    int                __socketAddressLength; \
+    \
+    remoteAddress=0; \
+    remotePort   =0; \
+    \
+    __socketAddressLength=sizeof(__socketAddress); \
+    result=(getpeername(socketDescriptor,(struct sockaddr*)&__socketAddress,&__socketAddressLength)==0)?TARGET_NATIVE_OK:TARGET_NATIVE_ERROR; \
+    if (result==TARGET_NATIVE_OK) \
+    { \
+      assert(__socketAddressLength>=sizeof(__socketAddress)); \
+      remoteAddress=ntohl(__socketAddress.sin_addr.s_addr); \
+      remotePort   =ntohs(__socketAddress.sin_port); \
+    } \
+  } while (0)
+
+#define TARGET_NATIVE_NETWORK_SOCKET_RECEIVE_AVAILABLE(socketDescriptor,bytesAvailable,result) \
+  do { \
+    u_long __value; \
+    \
+    bytesAvailable=0; \
+    \
+    result=(ioctlsocket(socketDescriptor,FIONREAD,&__value)==0)?TARGET_NATIVE_OK:TARGET_NATIVE_ERROR; \
+    if (result==TARGET_NATIVE_OK) \
+    { \
+      bytesAvailable=__value; \
+    } \
+  } while (0)
+#define TARGET_NATIVE_NETWORK_SOCKET_RECEIVE(socketDescriptor,buffer,maxLength,bytesReceived) \
+  do { \
+    bytesReceived=recv(socketDescriptor,buffer,maxLength,0); \
+  } while (0)
+#define TARGET_NATIVE_NETWORK_SOCKET_RECEIVE_WITH_ADDRESS_PORT(socketDescriptor,buffer,maxLength,address,port,bytesReceived) \
+  do { \
+    struct sockaddr_in __socketAddress; \
+    int                __socketAddressLength; \
+    \
+    port=0; \
+    \
+    memset(&__socketAddress,0,sizeof(__socketAddress)); \
+    __socketAddressLength=sizeof(__socketAddress); \
+    bytesReceived=recvfrom(socketDescriptor,buffer,maxLength,0,(struct sockaddr FAR*)&__socketAddress,&__socketAddressLength); \
+    if (__socketAddressLength==sizeof(__socketAddress)) \
+    { \
+      address=ntohl(__socketAddress.sin_addr.s_addr); \
+      port   =ntohs(__socketAddress.sin_port); \
+    } \
+  } while (0)
+
+#define TARGET_NATIVE_NETWORK_SOCKET_SEND(socketDescriptor,buffer,length,bytesSent) \
+  do { \
+    bytesSent=send(socketDescriptor,buffer,length,0); \
+  } while (0)
+#define TARGET_NATIVE_NETWORK_SOCKET_SEND_WITH_ADDRESS_PORT(socketDescriptor,buffer,length,address,port,bytesSent) \
+  do { \
+    struct sockaddr_in __socketAddress; \
+    \
+    memset(&__socketAddress,0,sizeof(__socketAddress)); \
+    __socketAddress.sin_family      = AF_INET; \
+    __socketAddress.sin_addr.s_addr = htonl(address); \
+    __socketAddress.sin_port        = htons((short)port); \
+    bytesSent=sendto(socketDescriptor,buffer,length,0,&__socketAddress,sizeof(__socketAddress)); \
+  } while (0)
+
+#define TARGET_NATIVE_NETWORK_SOCKET_SET_OPTION_TCP_NODELAY(socketDescriptor,flag,result) \
+  do { \
+    BOOL __value; \
+    \
+    __value=flag; \
+    result=(setsockopt(socketDescriptor,IPPROTO_TCP,TCP_NODELAY,(const char FAR*)&__value,sizeof(__value))==0)?TARGET_NATIVE_OK:TARGET_NATIVE_ERROR; \
+  } while (0)
+#define TARGET_NATIVE_NETWORK_SOCKET_SET_OPTION_SO_LINGER(socketDescriptor,flag,value,result) \
+  do { \
+    struct linger __linger; \
+    \
+    memset(&__linger,0,sizeof(__linger)); \
+    if (flag) \
+    { \
+      __linger.l_onoff=0; \
+    } \
+    else \
+    { \
+      __linger.l_linger=value; \
+      __linger.l_onoff =1; \
+    } \
+    result=(setsockopt(socketDescriptor,SOL_SOCKET,SO_LINGER,(const char FAR*)&__linger,sizeof(__linger))==0)?TARGET_NATIVE_OK:TARGET_NATIVE_ERROR; \
+  } while (0)
+#define TARGET_NATIVE_NETWORK_SOCKET_SET_OPTION_SO_TIMEOUT(socketDescriptor,flag,result) \
+  do { \
+    result = TARGET_NATIVE_OK; \
+  } while (0)
+#define TARGET_NATIVE_NETWORK_SOCKET_SET_OPTION_SO_SNDBUF(socketDescriptor,size,result) \
+  do { \
+    int __value; \
+    \
+    __value=size; \
+    result=(setsockopt(socketDescriptor,SOL_SOCKET,SO_SNDBUF,(const char FAR*)&__value,sizeof(__value))==0)?TARGET_NATIVE_OK:TARGET_NATIVE_ERROR; \
+  } while (0)
+#define TARGET_NATIVE_NETWORK_SOCKET_SET_OPTION_SO_RCDBUF(socketDescriptor,size,result) \
+  do { \
+    int __value; \
+    \
+    __value=size; \
+    result=(setsockopt(socketDescriptor,SOL_SOCKET,SO_RCVBUF,(const char FAR*)&__value,sizeof(__value))==0)?TARGET_NATIVE_OK:TARGET_NATIVE_ERROR; \
+  } while (0)
+#define TARGET_NATIVE_NETWORK_SOCKET_SET_OPTION_IP_TTL(socketDescriptor,value,result) \
+  do { \
+    result = TARGET_NATIVE_OK; \
+  } while (0)
+#define TARGET_NATIVE_NETWORK_SOCKET_SET_OPTION_IP_MULTICAST_IF(socketDescriptor,address,result) \
+  do { \
+    result = TARGET_NATIVE_OK; \
+  } while (0)
+#define TARGET_NATIVE_NETWORK_SOCKET_SET_OPTION_REUSE_ADDRESS(socketDescriptor,flag,result) \
+  do { \
+      BOOL __value; \
+      \
+      __value=flag; \
+      result=(setsockopt(socketDescriptor,SOL_SOCKET,SO_REUSEADDR,(const char FAR*)&__value,sizeof(__value))==0)?TARGET_NATIVE_OK:TARGET_NATIVE_ERROR; \
+  } while (0)
+#define TARGET_NATIVE_NETWORK_SOCKET_SET_OPTION_ADD_MEMBERSHIP(socketDescriptor,address,result) \
+  do { \
+    result = TARGET_NATIVE_ERROR; \
+  } while (0)
+#define TARGET_NATIVE_NETWORK_SOCKET_SET_OPTION_DROP_MEMBERSHIP(socketDescriptor,address,result) \
+  do { \
+    result = TARGET_NATIVE_ERROR; \
+  } while (0)
+#define TARGET_NATIVE_NETWORK_SOCKET_SET_OPTION_KEEP_ALIVE(socketDescriptor,flag,result) \
+  do { \
+    BOOL __value; \
+    \
+    __value=flag; \
+    result=(setsockopt(socketDescriptor,SOL_SOCKET,SO_KEEPALIVE,(const char FAR*)&__value,sizeof(__value))==0)?TARGET_NATIVE_OK:TARGET_NATIVE_ERROR; \
+  } while (0)
+#define TARGET_NATIVE_NETWORK_SOCKET_SET_OPTION_BROADCAST(socketDescriptor,flag,result) \
+  do { \
+    BOOL __value; \
+    \
+    __value=flag; \
+    result=(setsockopt(socketDescriptor,SOL_SOCKET,SO_BROADCAST,(const char FAR*)&__value,sizeof(__value))==0)?TARGET_NATIVE_OK:TARGET_NATIVE_ERROR; \
+  } while (0)
+
+#define TARGET_NATIVE_NETWORK_SOCKET_GET_OPTION_TCP_NODELAY(socketDescriptor,flag,result) \
+  do { \
+    BOOL __value; \
+    int  __len; \
+    \
+    flag=0; \
+    \
+    __len=sizeof(__value); \
+    result=(getsockopt(socketDescriptor,IPPROTO_TCP,TCP_NODELAY,(char FAR*)&__value,&__len)==0)?TARGET_NATIVE_OK:TARGET_NATIVE_ERROR; \
+    if (result==TARGET_NATIVE_OK) \
+    { \
+      assert(__len>=sizeof(__value)); \
+      flag=__value; \
+    } \
+  } while (0)
+#define TARGET_NATIVE_NETWORK_SOCKET_GET_OPTION_SO_LINGER(socketDescriptor,flag,value,result) \
+  do { \
+    struct linger __linger; \
+    int           __len; \
+    \
+    flag =0; \
+    value=0; \
+    \
+    __len=sizeof(__linger); \
+    result=(getsockopt(socketDescriptor,SOL_SOCKET,SO_LINGER,(char FAR*)&__linger,&__len)==0)?TARGET_NATIVE_OK:TARGET_NATIVE_ERROR; \
+    if (result==TARGET_NATIVE_OK) \
+    { \
+      assert(__len>=sizeof(__linger)); \
+      flag =__linger.l_onoff; \
+      value=__linger.l_linger; \
+    } \
+  } while (0)
+#define TARGET_NATIVE_NETWORK_SOCKET_GET_OPTION_SO_TIMEOUT(socketDescriptor,flag,result) \
+  do { \
+    flag = 0; \
+    result = TARGET_NATIVE_OK; \
+  } while (0)
+#define TARGET_NATIVE_NETWORK_SOCKET_GET_OPTION_SO_SNDBUF(socketDescriptor,size,result) \
+  do { \
+    int __value; \
+    int __len; \
+    \
+    size=0; \
+    \
+    __len=sizeof(__value); \
+    result=(getsockopt(socketDescriptor,SOL_SOCKET,SO_SNDBUF,(char FAR*)&__value,&__len)==0)?TARGET_NATIVE_OK:TARGET_NATIVE_ERROR; \
+    if (result==TARGET_NATIVE_OK) \
+    { \
+      assert(__len>=sizeof(__value)); \
+      size=__value; \
+    } \
+  } while (0)
+#define TARGET_NATIVE_NETWORK_SOCKET_GET_OPTION_SO_RCDBUF(socketDescriptor,size,result) \
+  do { \
+    int __value; \
+    int __len; \
+    \
+    size=0; \
+    \
+    __len=sizeof(__value); \
+    result=(getsockopt(socketDescriptor,SOL_SOCKET,SO_RCVBUF,(char FAR*)&__value,&__len)==0)?TARGET_NATIVE_OK:TARGET_NATIVE_ERROR; \
+    if (result==TARGET_NATIVE_OK) \
+    { \
+      assert(__len>=sizeof(__value)); \
+      size=__value; \
+    } \
+  } while (0)
+#define TARGET_NATIVE_NETWORK_SOCKET_GET_OPTION_IP_TTL(socketDescriptor,flag,result) \
+  do { \
+    flag = 0; \
+    result = TARGET_NATIVE_OK; \
+  } while (0)
+#define TARGET_NATIVE_NETWORK_SOCKET_GET_OPTION_IP_MULTICAST_IF(socketDescriptor,address,result) \
+  do { \
+    address = 0; \
+    result = TARGET_NATIVE_OK; \
+  } while (0)
+#define TARGET_NATIVE_NETWORK_SOCKET_GET_OPTION_BIND_ADDRESS(socketDescriptor,address,result) \
+  do { \
+    struct sockaddr_in __socketAddress; \
+    int                __socketAddressLength; \
+    \
+    address=0;\
+    \
+    memset(&__socketAddress,0,sizeof(__socketAddress)); \
+    __socketAddressLength=sizeof(__socketAddress); \
+    result=(getsockname(socketDescriptor,(struct sockaddr*)&__socketAddress,&__socketAddressLength)==0)?TARGET_NATIVE_OK:TARGET_NATIVE_ERROR; \
+    if (result==TARGET_NATIVE_OK) \
+    { \
+      assert(__socketAddressLength>=sizeof(__socketAddress)); \
+      address=ntohl(__socketAddress.sin_addr.s_addr); \
+    } \
+  } while (0)
+#define TARGET_NATIVE_NETWORK_SOCKET_GET_OPTION_REUSE_ADDRESS(socketDescriptor,flag,result) \
+  do { \
+    BOOL __value; \
+    int  __len; \
+    \
+    flag=0; \
+    \
+    __len=sizeof(__value); \
+    result=(getsockopt(socketDescriptor,SOL_SOCKET,SO_REUSEADDR,(char FAR*)&__value,&__len)==0)?TARGET_NATIVE_OK:TARGET_NATIVE_ERROR; \
+    if (result==TARGET_NATIVE_OK) \
+    { \
+      assert(__len>=sizeof(__value)); \
+      flag=__value; \
+    } \
+  } while (0)
+#define TARGET_NATIVE_NETWORK_SOCKET_GET_OPTION_KEEP_ALIVE(socketDescriptor,flag,result) \
+  do { \
+    BOOL __value; \
+    int  __len; \
+    \
+    flag=0; \
+    \
+    __len=sizeof(__value); \
+    result=(getsockopt(socketDescriptor,SOL_SOCKET,SO_KEEPALIVE,(char FAR*)&__value,&__len)==0)?TARGET_NATIVE_OK:TARGET_NATIVE_ERROR; \
+    if (result==TARGET_NATIVE_OK) \
+    { \
+      assert(__len>=sizeof(__value)); \
+      flag=__value; \
+    } \
+  } while (0)
+#define TARGET_NATIVE_NETWORK_SOCKET_GET_OPTION_BROADCAST(socketDescriptor,flag,result) \
+  do { \
+    BOOL __value; \
+    int  __len; \
+    \
+    flag=0; \
+    \
+    __len=sizeof(__value); \
+    result=(getsockopt(socketDescriptor,SOL_SOCKET,SO_BROADCAST,(char FAR*)&__value,&__len)==0)?TARGET_NATIVE_OK:TARGET_NATIVE_ERROR; \
+    if (result==TARGET_NATIVE_OK) \
+    { \
+      assert(__len>=sizeof(__value)); \
+      flag=__value; \
+    } \
+  } while (0)
+
+/***************************** Functions *******************************/
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#ifdef __cplusplus
+}
+#endif
+
+/* include rest of definitions from generic file (do not move it to 
+   another position!) */
+#include "target_generic_network.h"
+
+#endif /* __TARGET_NATIVE_NETWORK__ */
+
+/* end of file */

--------------040505090707060700070708
Content-Type: text/plain;
 name="target-layer-RTEMS.patch"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline;
 filename="target-layer-RTEMS.patch"

diff -ruN --exclude='.*' --exclude='Entries*' --exclude=Repository --exclude=Root RTEMS/Makefile.am RTEMS-new/Makefile.am
--- RTEMS/Makefile.am	Thu Jan  1 01:00:00 1970
+++ RTEMS-new/Makefile.am	Fri Aug  6 15:51:03 2004
@@ -0,0 +1,11 @@
+## Input file for automake to generate the Makefile.in used by configure
+
+EXTRA_DIST = \
+target_native.h \
+target_native_file.h \
+target_native_io.h \
+target_native_math_float.h \
+target_native_math_int.h
+target_native_memory.h
+target_native_misc.h \
+target_native_network.h \
diff -ruN --exclude='.*' --exclude='Entries*' --exclude=Repository --exclude=Root RTEMS/target_native.h RTEMS-new/target_native.h
--- RTEMS/target_native.h	Thu Jan  1 01:00:00 1970
+++ RTEMS-new/target_native.h	Fri Aug  6 15:51:23 2004
@@ -0,0 +1,79 @@
+/* ???.h - ???
+   Copyright (C) 1998 Free Software Foundation, Inc.
+
+This file is part of GNU Classpath.
+
+GNU Classpath is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2, or (at your option)
+any later version.
+ 
+GNU Classpath is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Classpath; see the file COPYING.  If not, write to the
+Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
+02111-1307 USA.
+
+Linking this library statically or dynamically with other modules is
+making a combined work based on this library.  Thus, the terms and
+conditions of the GNU General Public License cover the whole
+combination.
+
+As a special exception, the copyright holders of this library give you
+permission to link this library with independent modules to produce an
+executable, regardless of the license terms of these independent
+modules, and to copy and distribute the resulting executable under
+terms of your choice, provided that you also meet, for each linked
+independent module, the terms and conditions of the license of that
+module.  An independent module is a module which is not derived from
+or based on this library.  If you modify this library, you may extend
+this exception to your version of the library, but you are not
+obligated to do so.  If you do not wish to do so, delete this
+exception statement from your version. */
+
+/*
+Description: RTEMS target global defintions
+Systems    : RTEMS
+*/
+
+#ifndef __TARGET_NATIVE__
+#define __TARGET_NATIVE__
+
+/****************************** Includes *******************************/
+/* do not move; needed here because of some macro definitions */
+#include <config.h>
+
+#include <stdlib.h>
+
+/****************** Conditional compilation switches *******************/
+
+/***************************** Constants *******************************/
+
+/***************************** Datatypes *******************************/
+
+/***************************** Variables *******************************/
+
+/****************************** Macros *********************************/
+
+/***************************** Functions *******************************/
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#ifdef __cplusplus
+}
+#endif
+
+/* include rest of definitions from generic file (do not move it to 
+   another position!) */
+#include "target_generic.h"
+
+#endif /* __TARGET_NATIVE__ */
+
+/* end of file */
+
diff -ruN --exclude='.*' --exclude='Entries*' --exclude=Repository --exclude=Root RTEMS/target_native_file.h RTEMS-new/target_native_file.h
--- RTEMS/target_native_file.h	Thu Jan  1 01:00:00 1970
+++ RTEMS-new/target_native_file.h	Fri Aug  6 16:04:40 2004
@@ -0,0 +1,79 @@
+/* ???.h - ???
+   Copyright (C) 1998 Free Software Foundation, Inc.
+
+This file is part of GNU Classpath.
+
+GNU Classpath is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2, or (at your option)
+any later version.
+ 
+GNU Classpath is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Classpath; see the file COPYING.  If not, write to the
+Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
+02111-1307 USA.
+
+Linking this library statically or dynamically with other modules is
+making a combined work based on this library.  Thus, the terms and
+conditions of the GNU General Public License cover the whole
+combination.
+
+As a special exception, the copyright holders of this library give you
+permission to link this library with independent modules to produce an
+executable, regardless of the license terms of these independent
+modules, and to copy and distribute the resulting executable under
+terms of your choice, provided that you also meet, for each linked
+independent module, the terms and conditions of the license of that
+module.  An independent module is a module which is not derived from
+or based on this library.  If you modify this library, you may extend
+this exception to your version of the library, but you are not
+obligated to do so.  If you do not wish to do so, delete this
+exception statement from your version. */
+
+/*
+Description: RTEMS target defintions of file functions
+Systems    : RTEMS
+*/
+
+#ifndef __TARGET_NATIVE_FILE__
+#define __TARGET_NATIVE_FILE__
+
+/****************************** Includes *******************************/
+/* do not move; needed here because of some macro definitions */
+#include <config.h>
+
+#include <stdlib.h>
+
+/****************** Conditional compilation switches *******************/
+
+/***************************** Constants *******************************/
+
+/***************************** Datatypes *******************************/
+
+/***************************** Variables *******************************/
+
+/****************************** Macros *********************************/
+
+/***************************** Functions *******************************/
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#ifdef __cplusplus
+}
+#endif
+
+/* include rest of definitions from generic file (do not move it to 
+   another position!) */
+#include "target_generic_file.h"
+
+#endif /* __TARGET_NATIVE_FILE__ */
+
+/* end of file */
+
diff -ruN --exclude='.*' --exclude='Entries*' --exclude=Repository --exclude=Root RTEMS/target_native_io.h RTEMS-new/target_native_io.h
--- RTEMS/target_native_io.h	Thu Jan  1 01:00:00 1970
+++ RTEMS-new/target_native_io.h	Fri Aug  6 15:51:23 2004
@@ -0,0 +1,78 @@
+/* ???.h - ???
+   Copyright (C) 1998 Free Software Foundation, Inc.
+
+This file is part of GNU Classpath.
+
+GNU Classpath is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2, or (at your option)
+any later version.
+ 
+GNU Classpath is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Classpath; see the file COPYING.  If not, write to the
+Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
+02111-1307 USA.
+
+Linking this library statically or dynamically with other modules is
+making a combined work based on this library.  Thus, the terms and
+conditions of the GNU General Public License cover the whole
+combination.
+
+As a special exception, the copyright holders of this library give you
+permission to link this library with independent modules to produce an
+executable, regardless of the license terms of these independent
+modules, and to copy and distribute the resulting executable under
+terms of your choice, provided that you also meet, for each linked
+independent module, the terms and conditions of the license of that
+module.  An independent module is a module which is not derived from
+or based on this library.  If you modify this library, you may extend
+this exception to your version of the library, but you are not
+obligated to do so.  If you do not wish to do so, delete this
+exception statement from your version. */
+
+/*
+Description: RTEMS target defintions of miscellaneous functions
+Systems    : RTEMS
+*/
+
+#ifndef __TARGET_NATIVE_IO__
+#define __TARGET_NATIVE_IO__
+
+/****************************** Includes *******************************/
+/* do not move; needed here because of some macro definitions */
+#include <config.h>
+
+#include <stdlib.h>
+
+/****************** Conditional compilation switches *******************/
+
+/***************************** Constants *******************************/
+
+/***************************** Datatypes *******************************/
+
+/***************************** Variables *******************************/
+
+/****************************** Macros *********************************/
+
+/***************************** Functions *******************************/
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#ifdef __cplusplus
+}
+#endif
+
+/* include rest of definitions from generic file (do not move it to 
+   another position!) */
+#include "target_generic_io.h"
+
+#endif /* __TARGET_NATIVE_MISC__ */
+
+/* end of file */
diff -ruN --exclude='.*' --exclude='Entries*' --exclude=Repository --exclude=Root RTEMS/target_native_math_float.h RTEMS-new/target_native_math_float.h
--- RTEMS/target_native_math_float.h	Thu Jan  1 01:00:00 1970
+++ RTEMS-new/target_native_math_float.h	Fri Aug  6 15:51:23 2004
@@ -0,0 +1,79 @@
+/* ???.h - ???
+   Copyright (C) 1998 Free Software Foundation, Inc.
+
+This file is part of GNU Classpath.
+
+GNU Classpath is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2, or (at your option)
+any later version.
+ 
+GNU Classpath is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Classpath; see the file COPYING.  If not, write to the
+Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
+02111-1307 USA.
+
+Linking this library statically or dynamically with other modules is
+making a combined work based on this library.  Thus, the terms and
+conditions of the GNU General Public License cover the whole
+combination.
+
+As a special exception, the copyright holders of this library give you
+permission to link this library with independent modules to produce an
+executable, regardless of the license terms of these independent
+modules, and to copy and distribute the resulting executable under
+terms of your choice, provided that you also meet, for each linked
+independent module, the terms and conditions of the license of that
+module.  An independent module is a module which is not derived from
+or based on this library.  If you modify this library, you may extend
+this exception to your version of the library, but you are not
+obligated to do so.  If you do not wish to do so, delete this
+exception statement from your version. */
+
+/*
+Description: RTEMS target defintions of floating point functions
+Systems    : RTEMS
+*/
+
+#ifndef __TARGET_NATIVE_MATH_FLOAT__
+#define __TARGET_NATIVE_MATH_FLOAT__
+
+/****************************** Includes *******************************/
+/* do not move; needed here because of some macro definitions */
+#include <config.h>
+
+#include <stdlib.h>
+
+/****************** Conditional compilation switches *******************/
+
+/***************************** Constants *******************************/
+
+/***************************** Datatypes *******************************/
+
+/***************************** Variables *******************************/
+
+/****************************** Macros *********************************/
+
+/***************************** Functions *******************************/
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#ifdef __cplusplus
+}
+#endif
+
+/* include rest of definitions from generic file (do not move it to 
+   another position!) */
+#include "target_generic_math_float.h"
+
+#endif /* __TARGET_NATIVE_MATH_FLOAT__ */
+
+/* end of file */
+
diff -ruN --exclude='.*' --exclude='Entries*' --exclude=Repository --exclude=Root RTEMS/target_native_math_int.h RTEMS-new/target_native_math_int.h
--- RTEMS/target_native_math_int.h	Thu Jan  1 01:00:00 1970
+++ RTEMS-new/target_native_math_int.h	Fri Aug  6 15:51:23 2004
@@ -0,0 +1,80 @@
+/* ???.h - ???
+   Copyright (C) 1998 Free Software Foundation, Inc.
+
+This file is part of GNU Classpath.
+
+GNU Classpath is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2, or (at your option)
+any later version.
+ 
+GNU Classpath is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Classpath; see the file COPYING.  If not, write to the
+Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
+02111-1307 USA.
+
+Linking this library statically or dynamically with other modules is
+making a combined work based on this library.  Thus, the terms and
+conditions of the GNU General Public License cover the whole
+combination.
+
+As a special exception, the copyright holders of this library give you
+permission to link this library with independent modules to produce an
+executable, regardless of the license terms of these independent
+modules, and to copy and distribute the resulting executable under
+terms of your choice, provided that you also meet, for each linked
+independent module, the terms and conditions of the license of that
+module.  An independent module is a module which is not derived from
+or based on this library.  If you modify this library, you may extend
+this exception to your version of the library, but you are not
+obligated to do so.  If you do not wish to do so, delete this
+exception statement from your version. */
+
+/*
+Description: RTEMS target defintions of int/int64 constants/
+             macros/functions
+Systems    : RTEMS
+*/
+
+#ifndef __TARGET_NATIVE_MATH_INT__
+#define __TARGET_NATIVE_MATH_INT__
+
+/****************************** Includes *******************************/
+/* do not move; needed here because of some macro definitions */
+#include <config.h>
+
+#include <stdlib.h>
+
+/****************** Conditional compilation switches *******************/
+
+/***************************** Constants *******************************/
+
+/***************************** Datatypes *******************************/
+
+/***************************** Variables *******************************/
+
+/****************************** Macros *********************************/
+
+/***************************** Functions *******************************/
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#ifdef __cplusplus
+}
+#endif
+
+/* include rest of definitions from generic file (do not move it to 
+   another position!) */
+#include "target_generic_math_int.h"
+
+#endif /* __TARGET_NATIVE_MATH_INT__ */
+
+/* end of file */
+
diff -ruN --exclude='.*' --exclude='Entries*' --exclude=Repository --exclude=Root RTEMS/target_native_memory.h RTEMS-new/target_native_memory.h
--- RTEMS/target_native_memory.h	Thu Jan  1 01:00:00 1970
+++ RTEMS-new/target_native_memory.h	Fri Aug  6 15:51:23 2004
@@ -0,0 +1,78 @@
+/* ???.h - ???
+   Copyright (C) 1998 Free Software Foundation, Inc.
+
+This file is part of GNU Classpath.
+
+GNU Classpath is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2, or (at your option)
+any later version.
+ 
+GNU Classpath is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Classpath; see the file COPYING.  If not, write to the
+Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
+02111-1307 USA.
+
+Linking this library statically or dynamically with other modules is
+making a combined work based on this library.  Thus, the terms and
+conditions of the GNU General Public License cover the whole
+combination.
+
+As a special exception, the copyright holders of this library give you
+permission to link this library with independent modules to produce an
+executable, regardless of the license terms of these independent
+modules, and to copy and distribute the resulting executable under
+terms of your choice, provided that you also meet, for each linked
+independent module, the terms and conditions of the license of that
+module.  An independent module is a module which is not derived from
+or based on this library.  If you modify this library, you may extend
+this exception to your version of the library, but you are not
+obligated to do so.  If you do not wish to do so, delete this
+exception statement from your version. */
+
+/*
+Description: RTEMS target defintions of memory functions
+Systems    : all
+*/
+
+#ifndef __TARGET_NATIVE_MEMORY__
+#define __TARGET_NATIVE_MEMORY__
+
+/****************************** Includes *******************************/
+/* do not move; needed here because of some macro definitions */
+#include <config.h>
+
+#include <stdlib.h>
+
+/****************** Conditional compilation switches *******************/
+
+/***************************** Constants *******************************/
+
+/***************************** Datatypes *******************************/
+
+/***************************** Variables *******************************/
+
+/****************************** Macros *********************************/
+
+/***************************** Functions *******************************/
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#ifdef __cplusplus
+}
+#endif
+
+/* include rest of definitions from generic file (do not move it to 
+   another position!) */
+#include "target_generic_memory.h"
+
+#endif /* __TARGET_NATIVE_MEMORY__ */
+
+/* end of file */
diff -ruN --exclude='.*' --exclude='Entries*' --exclude=Repository --exclude=Root RTEMS/target_native_misc.h RTEMS-new/target_native_misc.h
--- RTEMS/target_native_misc.h	Thu Jan  1 01:00:00 1970
+++ RTEMS-new/target_native_misc.h	Fri Aug  6 16:04:05 2004
@@ -0,0 +1,87 @@
+/* ???.h - ???
+   Copyright (C) 1998 Free Software Foundation, Inc.
+
+This file is part of GNU Classpath.
+
+GNU Classpath is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2, or (at your option)
+any later version.
+ 
+GNU Classpath is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Classpath; see the file COPYING.  If not, write to the
+Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
+02111-1307 USA.
+
+Linking this library statically or dynamically with other modules is
+making a combined work based on this library.  Thus, the terms and
+conditions of the GNU General Public License cover the whole
+combination.
+
+As a special exception, the copyright holders of this library give you
+permission to link this library with independent modules to produce an
+executable, regardless of the license terms of these independent
+modules, and to copy and distribute the resulting executable under
+terms of your choice, provided that you also meet, for each linked
+independent module, the terms and conditions of the license of that
+module.  An independent module is a module which is not derived from
+or based on this library.  If you modify this library, you may extend
+this exception to your version of the library, but you are not
+obligated to do so.  If you do not wish to do so, delete this
+exception statement from your version. */
+
+/*
+Description: RTEMS target defintions of miscellaneous functions
+Systems    : RTEMS
+*/
+
+#ifndef __TARGET_NATIVE_MISC__
+#define __TARGET_NATIVE_MISC__
+
+/****************************** Includes *******************************/
+/* do not move; needed here because of some macro definitions */
+#include <config.h>
+
+#include <stdlib.h>
+
+/****************** Conditional compilation switches *******************/
+
+/***************************** Constants *******************************/
+
+/***************************** Datatypes *******************************/
+
+/***************************** Variables *******************************/
+
+/****************************** Macros *********************************/
+
+// NYI
+#define TARGET_NATIVE_MISC_GET_TIMEZONE_STRING(string,maxStringLength,result) \
+  do { \
+    result = TARGET_NATIVE_ERROR; \
+  } while (0)
+
+/***************************** Functions *******************************/
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#ifdef __cplusplus
+}
+#endif
+
+
+
+/* include rest of definitions from generic file (do not move it to 
+   another position!) */
+#include "target_generic_misc.h"
+
+#endif /* __TARGET_NATIVE_MISC__ */
+
+/* end of file */
+
diff -ruN --exclude='.*' --exclude='Entries*' --exclude=Repository --exclude=Root RTEMS/target_native_network.h RTEMS-new/target_native_network.h
--- RTEMS/target_native_network.h	Thu Jan  1 01:00:00 1970
+++ RTEMS-new/target_native_network.h	Fri Aug  6 16:02:53 2004
@@ -0,0 +1,97 @@
+/* ???.h - ???
+   Copyright (C) 1998 Free Software Foundation, Inc.
+
+This file is part of GNU Classpath.
+
+GNU Classpath is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2, or (at your option)
+any later version.
+ 
+GNU Classpath is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Classpath; see the file COPYING.  If not, write to the
+Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
+02111-1307 USA.
+
+Linking this library statically or dynamically with other modules is
+making a combined work based on this library.  Thus, the terms and
+conditions of the GNU General Public License cover the whole
+combination.
+
+As a special exception, the copyright holders of this library give you
+permission to link this library with independent modules to produce an
+executable, regardless of the license terms of these independent
+modules, and to copy and distribute the resulting executable under
+terms of your choice, provided that you also meet, for each linked
+independent module, the terms and conditions of the license of that
+module.  An independent module is a module which is not derived from
+or based on this library.  If you modify this library, you may extend
+this exception to your version of the library, but you are not
+obligated to do so.  If you do not wish to do so, delete this
+exception statement from your version. */
+
+/*
+Description: RTEMS target defintions of network functions
+Systems    : RTEMS
+
+*/
+
+#ifndef __TARGET_NATIVE_NETWORK__
+#define __TARGET_NATIVE_NETWORK__
+
+/****************************** Includes *******************************/
+/* do not move; needed here because of some macro definitions */
+#include <config.h>
+
+#include <stdlib.h>
+#include <sys/socket.h>
+
+/****************** Conditional compilation switches *******************/
+
+/***************************** Constants *******************************/
+
+/***************************** Datatypes *******************************/
+
+/***************************** Variables *******************************/
+
+/****************************** Macros *********************************/
+
+#define TARGET_NATIVE_NETWORK_SOCKET_SEND(socketDescriptor,buffer,length,bytesSent) \
+  do { \
+    bytesSent = send(socketDescriptor,buffer,length,0); \
+  } while (0)
+
+#define TARGET_NATIVE_NETWORK_SOCKET_SEND_WITH_ADDRESS_PORT(socketDescriptor,buffer,length,address,port,bytesSent) \
+  do { \
+    struct sockaddr_in __socketAddress; \
+    \
+    memset(&__socketAddress,0,sizeof(__socketAddress)); \
+    __socketAddress.sin_family      = AF_INET; \
+    __socketAddress.sin_addr.s_addr = htonl(address); \
+    __socketAddress.sin_port        = htons((short)port); \
+    bytesSent = sendto(socketDescriptor,buffer,length,0,(struct sockaddr*)&__socketAddress,sizeof(__socketAddress)); \
+  } while (0)
+
+/***************************** Functions *******************************/
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#ifdef __cplusplus
+}
+#endif
+
+/* include rest of definitions from generic file (do not move it to 
+   another position!) */
+#include "target_generic_network.h"
+
+#endif /* __TARGET_NATIVE_NETWORK__ */
+
+/* end of file */
+

--------------040505090707060700070708
Content-Type: text/plain;
 name="target-layer-SunOS.patch"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline;
 filename="target-layer-SunOS.patch"

diff -ruN --exclude='.*' --exclude='Entries*' --exclude=Repository --exclude=Root SunOS/Makefile.am SunOS-new/Makefile.am
--- SunOS/Makefile.am	Thu Jan  1 01:00:00 1970
+++ SunOS-new/Makefile.am	Fri Aug  6 15:49:40 2004
@@ -0,0 +1,11 @@
+## Input file for automake to generate the Makefile.in used by configure
+
+EXTRA_DIST = \
+target_native.h \
+target_native_file.h \
+target_native_io.h \
+target_native_math_float.h \
+target_native_math_int.h
+target_native_memory.h
+target_native_misc.h \
+target_native_network.h \
diff -ruN --exclude='.*' --exclude='Entries*' --exclude=Repository --exclude=Root SunOS/target_native.h SunOS-new/target_native.h
--- SunOS/target_native.h	Thu Jan  1 01:00:00 1970
+++ SunOS-new/target_native.h	Tue Aug  3 14:13:23 2004
@@ -0,0 +1,78 @@
+/* ???.h - ???
+   Copyright (C) 1998 Free Software Foundation, Inc.
+
+This file is part of GNU Classpath.
+
+GNU Classpath is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2, or (at your option)
+any later version.
+ 
+GNU Classpath is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Classpath; see the file COPYING.  If not, write to the
+Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
+02111-1307 USA.
+
+Linking this library statically or dynamically with other modules is
+making a combined work based on this library.  Thus, the terms and
+conditions of the GNU General Public License cover the whole
+combination.
+
+As a special exception, the copyright holders of this library give you
+permission to link this library with independent modules to produce an
+executable, regardless of the license terms of these independent
+modules, and to copy and distribute the resulting executable under
+terms of your choice, provided that you also meet, for each linked
+independent module, the terms and conditions of the license of that
+module.  An independent module is a module which is not derived from
+or based on this library.  If you modify this library, you may extend
+this exception to your version of the library, but you are not
+obligated to do so.  If you do not wish to do so, delete this
+exception statement from your version. */
+
+/*
+Description: SunOS/Solaris target global defintions
+Systems    : all
+*/
+
+#ifndef __TARGET_NATIVE__
+#define __TARGET_NATIVE__
+
+/****************************** Includes *******************************/
+/* do not move; needed here because of some macro definitions */
+#include <config.h>
+
+#include <stdlib.h>
+
+/****************** Conditional compilation switches *******************/
+
+/***************************** Constants *******************************/
+
+/***************************** Datatypes *******************************/
+
+/***************************** Variables *******************************/
+
+/****************************** Macros *********************************/
+
+/***************************** Functions *******************************/
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#ifdef __cplusplus
+}
+#endif
+
+/* include rest of definitions from generic file (do not move it to 
+   another position!) */
+#include "target_generic.h"
+
+#endif /* __TARGET_NATIVE__ */
+
+/* end of file */
diff -ruN --exclude='.*' --exclude='Entries*' --exclude=Repository --exclude=Root SunOS/target_native_file.h SunOS-new/target_native_file.h
--- SunOS/target_native_file.h	Thu Jan  1 01:00:00 1970
+++ SunOS-new/target_native_file.h	Tue Aug  3 14:13:23 2004
@@ -0,0 +1,79 @@
+/* ???.h - ???
+   Copyright (C) 1998 Free Software Foundation, Inc.
+
+This file is part of GNU Classpath.
+
+GNU Classpath is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2, or (at your option)
+any later version.
+ 
+GNU Classpath is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Classpath; see the file COPYING.  If not, write to the
+Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
+02111-1307 USA.
+
+Linking this library statically or dynamically with other modules is
+making a combined work based on this library.  Thus, the terms and
+conditions of the GNU General Public License cover the whole
+combination.
+
+As a special exception, the copyright holders of this library give you
+permission to link this library with independent modules to produce an
+executable, regardless of the license terms of these independent
+modules, and to copy and distribute the resulting executable under
+terms of your choice, provided that you also meet, for each linked
+independent module, the terms and conditions of the license of that
+module.  An independent module is a module which is not derived from
+or based on this library.  If you modify this library, you may extend
+this exception to your version of the library, but you are not
+obligated to do so.  If you do not wish to do so, delete this
+exception statement from your version. */
+
+/*
+Description: Linux target defintions of file functions
+Systems    : all
+*/
+
+#ifndef __TARGET_NATIVE_FILE__
+#define __TARGET_NATIVE_FILE__
+
+/****************************** Includes *******************************/
+/* do not move; needed here because of some macro definitions */
+#include <config.h>
+
+#include <stdlib.h>
+
+/****************** Conditional compilation switches *******************/
+
+/***************************** Constants *******************************/
+
+/***************************** Datatypes *******************************/
+
+/***************************** Variables *******************************/
+
+/****************************** Macros *********************************/
+
+/***************************** Functions *******************************/
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#ifdef __cplusplus
+}
+#endif
+
+/* include rest of definitions from generic file (do not move it to 
+   another position!) */
+#include "target_generic_file.h"
+
+#endif /* __TARGET_NATIVE_FILE__ */
+
+/* end of file */
+
diff -ruN --exclude='.*' --exclude='Entries*' --exclude=Repository --exclude=Root SunOS/target_native_io.h SunOS-new/target_native_io.h
--- SunOS/target_native_io.h	Thu Jan  1 01:00:00 1970
+++ SunOS-new/target_native_io.h	Tue Aug  3 14:13:23 2004
@@ -0,0 +1,80 @@
+/* generic_math_int64.h - Native methods for 64bit math operations
+   Copyright (C) 1998 Free Software Foundation, Inc.
+
+This file is part of GNU Classpath.
+
+GNU Classpath is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2, or (at your option)
+any later version.
+ 
+GNU Classpath is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Classpath; see the file COPYING.  If not, write to the
+Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
+02111-1307 USA.
+
+Linking this library statically or dynamically with other modules is
+making a combined work based on this library.  Thus, the terms and
+conditions of the GNU General Public License cover the whole
+combination.
+
+As a special exception, the copyright holders of this library give you
+permission to link this library with independent modules to produce an
+executable, regardless of the license terms of these independent
+modules, and to copy and distribute the resulting executable under
+terms of your choice, provided that you also meet, for each linked
+independent module, the terms and conditions of the license of that
+module.  An independent module is a module which is not derived from
+or based on this library.  If you modify this library, you may extend
+this exception to your version of the library, but you are not
+obligated to do so.  If you do not wish to do so, delete this
+exception statement from your version. */
+
+/*
+Description: SunOS/Solaris target defintions of I/O functions
+Systems    : all
+*/
+
+#ifndef __TARGET_NATIVE_IO__
+#define __TARGET_NATIVE_IO__
+
+/****************************** Includes *******************************/
+/* do not move; needed here because of some macro definitions */
+#include "config.h"
+
+#include <stdlib.h>
+
+#include "target_native.h"
+
+/****************** Conditional compilation switches *******************/
+
+/***************************** Constants *******************************/
+
+/***************************** Datatypes *******************************/
+
+/***************************** Variables *******************************/
+
+/****************************** Macros *********************************/
+
+/***************************** Functions *******************************/
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#ifdef __cplusplus
+}
+#endif
+
+/* include rest of definitions from generic file (do not move it to 
+   another position!) */
+#include "target_generic_io.h"
+
+#endif /* __TARGET_NATIVE_IO__ */
+
+/* end of file */
diff -ruN --exclude='.*' --exclude='Entries*' --exclude=Repository --exclude=Root SunOS/target_native_math_float.h SunOS-new/target_native_math_float.h
--- SunOS/target_native_math_float.h	Thu Jan  1 01:00:00 1970
+++ SunOS-new/target_native_math_float.h	Fri Aug  6 15:59:36 2004
@@ -0,0 +1,90 @@
+/* ???.h - ???
+   Copyright (C) 1998 Free Software Foundation, Inc.
+
+This file is part of GNU Classpath.
+
+GNU Classpath is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2, or (at your option)
+any later version.
+ 
+GNU Classpath is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Classpath; see the file COPYING.  If not, write to the
+Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
+02111-1307 USA.
+
+Linking this library statically or dynamically with other modules is
+making a combined work based on this library.  Thus, the terms and
+conditions of the GNU General Public License cover the whole
+combination.
+
+As a special exception, the copyright holders of this library give you
+permission to link this library with independent modules to produce an
+executable, regardless of the license terms of these independent
+modules, and to copy and distribute the resulting executable under
+terms of your choice, provided that you also meet, for each linked
+independent module, the terms and conditions of the license of that
+module.  An independent module is a module which is not derived from
+or based on this library.  If you modify this library, you may extend
+this exception to your version of the library, but you are not
+obligated to do so.  If you do not wish to do so, delete this
+exception statement from your version. */
+
+/*
+Description: Linux target defintions of float/double constants/
+             macros/functions
+Systems    : all
+*/
+
+#ifndef __TARGET_NATIVE_MATH_FLOAT__
+#define __TARGET_NATIVE_MATH_FLOAT__
+
+/****************************** Includes *******************************/
+/* do not move; needed here because of some macro definitions */
+#include <config.h>
+
+#include <stdlib.h>
+
+/****************** Conditional compilation switches *******************/
+
+/***************************** Constants *******************************/
+
+/***************************** Datatypes *******************************/
+
+/***************************** Variables *******************************/
+
+/****************************** Macros *********************************/
+
+// NYI ??? correct?
+#define TARGET_NATIVE_MATH_FLOAT_FLOAT_ISINF(f) (((f)==1.0/0.0) || ((f)==-1.0/0.0))
+#define TARGET_NATIVE_MATH_FLOAT_FLOAT_POS_ISINF(f) ((f)==1.0/0.0)
+#define TARGET_NATIVE_MATH_FLOAT_FLOAT_NEG_ISINF(f) ((f)==-1.0/0.0)
+
+// NYI ??? correct?
+#define TARGET_NATIVE_MATH_FLOAT_DOUBLE_ISINF(d) (((d)==1.0/0.0) || ((d)==-1.0/0.0))
+#define TARGET_NATIVE_MATH_FLOAT_DOUBLE_POS_ISINF(d) ((d)==1.0/0.0)
+#define TARGET_NATIVE_MATH_FLOAT_DOUBLE_NEG_ISINF(d) ((d)==-1.0/0.0)
+
+/***************************** Functions *******************************/
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#ifdef __cplusplus
+}
+#endif
+
+/* include rest of definitions from generic file (do not move it to 
+   another position!) */
+#include "target_generic_math_float.h"
+
+#endif /* __TARGET_NATIVE_MATH_FLOAT__ */
+
+/* end of file */
+
diff -ruN --exclude='.*' --exclude='Entries*' --exclude=Repository --exclude=Root SunOS/target_native_math_int.h SunOS-new/target_native_math_int.h
--- SunOS/target_native_math_int.h	Thu Jan  1 01:00:00 1970
+++ SunOS-new/target_native_math_int.h	Tue Aug  3 14:13:23 2004
@@ -0,0 +1,80 @@
+/* ???.h - ???
+   Copyright (C) 1998 Free Software Foundation, Inc.
+
+This file is part of GNU Classpath.
+
+GNU Classpath is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2, or (at your option)
+any later version.
+ 
+GNU Classpath is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Classpath; see the file COPYING.  If not, write to the
+Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
+02111-1307 USA.
+
+Linking this library statically or dynamically with other modules is
+making a combined work based on this library.  Thus, the terms and
+conditions of the GNU General Public License cover the whole
+combination.
+
+As a special exception, the copyright holders of this library give you
+permission to link this library with independent modules to produce an
+executable, regardless of the license terms of these independent
+modules, and to copy and distribute the resulting executable under
+terms of your choice, provided that you also meet, for each linked
+independent module, the terms and conditions of the license of that
+module.  An independent module is a module which is not derived from
+or based on this library.  If you modify this library, you may extend
+this exception to your version of the library, but you are not
+obligated to do so.  If you do not wish to do so, delete this
+exception statement from your version. */
+
+/*
+Description: Linux target defintions of int/int64 constants/
+             macros/functions
+Systems    : all
+*/
+
+#ifndef __TARGET_NATIVE_MATH_INT__
+#define __TARGET_NATIVE_MATH_INT__
+
+/****************************** Includes *******************************/
+/* do not move; needed here because of some macro definitions */
+#include <config.h>
+
+#include <stdlib.h>
+
+/****************** Conditional compilation switches *******************/
+
+/***************************** Constants *******************************/
+
+/***************************** Datatypes *******************************/
+
+/***************************** Variables *******************************/
+
+/****************************** Macros *********************************/
+
+/***************************** Functions *******************************/
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#ifdef __cplusplus
+}
+#endif
+
+/* include rest of definitions from generic file (do not move it to 
+   another position!) */
+#include "target_generic_math_int.h"
+
+#endif /* __TARGET_NATIVE_MATH_INT__ */
+
+/* end of file */
+
diff -ruN --exclude='.*' --exclude='Entries*' --exclude=Repository --exclude=Root SunOS/target_native_memory.h SunOS-new/target_native_memory.h
--- SunOS/target_native_memory.h	Thu Jan  1 01:00:00 1970
+++ SunOS-new/target_native_memory.h	Tue Aug  3 14:13:23 2004
@@ -0,0 +1,78 @@
+/* ???.h - ???
+   Copyright (C) 1998 Free Software Foundation, Inc.
+
+This file is part of GNU Classpath.
+
+GNU Classpath is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2, or (at your option)
+any later version.
+ 
+GNU Classpath is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Classpath; see the file COPYING.  If not, write to the
+Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
+02111-1307 USA.
+
+Linking this library statically or dynamically with other modules is
+making a combined work based on this library.  Thus, the terms and
+conditions of the GNU General Public License cover the whole
+combination.
+
+As a special exception, the copyright holders of this library give you
+permission to link this library with independent modules to produce an
+executable, regardless of the license terms of these independent
+modules, and to copy and distribute the resulting executable under
+terms of your choice, provided that you also meet, for each linked
+independent module, the terms and conditions of the license of that
+module.  An independent module is a module which is not derived from
+or based on this library.  If you modify this library, you may extend
+this exception to your version of the library, but you are not
+obligated to do so.  If you do not wish to do so, delete this
+exception statement from your version. */
+
+/*
+Description: SunOS/Solaris target defintions of memory functions
+Systems    : all
+*/
+
+#ifndef __TARGET_NATIVE_MEMORY__
+#define __TARGET_NATIVE_MEMORY__
+
+/****************************** Includes *******************************/
+/* do not move; needed here because of some macro definitions */
+#include <config.h>
+
+#include <stdlib.h>
+
+/****************** Conditional compilation switches *******************/
+
+/***************************** Constants *******************************/
+
+/***************************** Datatypes *******************************/
+
+/***************************** Variables *******************************/
+
+/****************************** Macros *********************************/
+
+/***************************** Functions *******************************/
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#ifdef __cplusplus
+}
+#endif
+
+/* include rest of definitions from generic file (do not move it to 
+   another position!) */
+#include "target_generic_memory.h"
+
+#endif /* __TARGET_NATIVE_MEMORY__ */
+
+/* end of file */
diff -ruN --exclude='.*' --exclude='Entries*' --exclude=Repository --exclude=Root SunOS/target_native_misc.h SunOS-new/target_native_misc.h
--- SunOS/target_native_misc.h	Thu Jan  1 01:00:00 1970
+++ SunOS-new/target_native_misc.h	Tue Aug  3 14:13:24 2004
@@ -0,0 +1,79 @@
+/* ???.h - ???
+   Copyright (C) 1998 Free Software Foundation, Inc.
+
+This file is part of GNU Classpath.
+
+GNU Classpath is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2, or (at your option)
+any later version.
+ 
+GNU Classpath is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Classpath; see the file COPYING.  If not, write to the
+Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
+02111-1307 USA.
+
+Linking this library statically or dynamically with other modules is
+making a combined work based on this library.  Thus, the terms and
+conditions of the GNU General Public License cover the whole
+combination.
+
+As a special exception, the copyright holders of this library give you
+permission to link this library with independent modules to produce an
+executable, regardless of the license terms of these independent
+modules, and to copy and distribute the resulting executable under
+terms of your choice, provided that you also meet, for each linked
+independent module, the terms and conditions of the license of that
+module.  An independent module is a module which is not derived from
+or based on this library.  If you modify this library, you may extend
+this exception to your version of the library, but you are not
+obligated to do so.  If you do not wish to do so, delete this
+exception statement from your version. */
+
+/*
+Description: Linux target defintions of miscellaneous functions
+Systems    : all
+*/
+
+#ifndef __TARGET_NATIVE_MISC__
+#define __TARGET_NATIVE_MISC__
+
+/****************************** Includes *******************************/
+/* do not move; needed here because of some macro definitions */
+#include <config.h>
+
+#include <stdlib.h>
+
+/****************** Conditional compilation switches *******************/
+
+/***************************** Constants *******************************/
+
+/***************************** Datatypes *******************************/
+
+/***************************** Variables *******************************/
+
+/****************************** Macros *********************************/
+
+/***************************** Functions *******************************/
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#ifdef __cplusplus
+}
+#endif
+
+/* include rest of definitions from generic file (do not move it to 
+   another position!) */
+#include "target_generic_misc.h"
+
+#endif /* __TARGET_NATIVE_MISC__ */
+
+/* end of file */
+
diff -ruN --exclude='.*' --exclude='Entries*' --exclude=Repository --exclude=Root SunOS/target_native_network.h SunOS-new/target_native_network.h
--- SunOS/target_native_network.h	Thu Jan  1 01:00:00 1970
+++ SunOS-new/target_native_network.h	Fri Aug  6 15:58:57 2004
@@ -0,0 +1,227 @@
+/* ???.h - ???
+   Copyright (C) 1998 Free Software Foundation, Inc.
+
+This file is part of GNU Classpath.
+
+GNU Classpath is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2, or (at your option)
+any later version.
+ 
+GNU Classpath is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Classpath; see the file COPYING.  If not, write to the
+Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
+02111-1307 USA.
+
+Linking this library statically or dynamically with other modules is
+making a combined work based on this library.  Thus, the terms and
+conditions of the GNU General Public License cover the whole
+combination.
+
+As a special exception, the copyright holders of this library give you
+permission to link this library with independent modules to produce an
+executable, regardless of the license terms of these independent
+modules, and to copy and distribute the resulting executable under
+terms of your choice, provided that you also meet, for each linked
+independent module, the terms and conditions of the license of that
+module.  An independent module is a module which is not derived from
+or based on this library.  If you modify this library, you may extend
+this exception to your version of the library, but you are not
+obligated to do so.  If you do not wish to do so, delete this
+exception statement from your version. */
+
+/*
+Description: SunOS/Solaris target defintions of network functions
+Systems    : all
+*/
+
+#ifndef __TARGET_NATIVE_NETWORK__
+#define __TARGET_NATIVE_NETWORK__
+
+/****************************** Includes *******************************/
+/* do not move; needed here because of some macro definitions */
+#include <config.h>
+
+#include <stdlib.h>
+#include <netdb.h>
+#include <stropts.h>
+#include <sys/conf.h>
+#include <netinet/in_systm.h>
+#include <sys/ioctl.h>
+
+/****************** Conditional compilation switches *******************/
+
+/***************************** Constants *******************************/
+
+/***************************** Datatypes *******************************/
+
+/***************************** Variables *******************************/
+
+/****************************** Macros *********************************/
+
+/***********************************************************************\
+* Name       : TARGET_NATIVE_NETWORK_GET_HOSTNAME_BY_NAME
+* Purpose    : get hostname by name
+* Input      : name           - hostname
+*              maxAddressSize - max. size of address array
+* Output     : addresses    - host addresses (array, in host byte
+*                             order!)
+*              addressCount - number of entries in address array
+*              result       - TARGET_NATIVE_OK if no error occurred,
+*                             TARGET_NATIVE_ERROR otherwise
+* Return     : -
+* Side-effect: unknown
+* Notes      : -
+\***********************************************************************/
+
+#define TARGET_NATIVE_NETWORK_GET_HOSTNAME_BY_NAME(name,addresses,maxAddressSize,addressCount,result) \
+  do { \
+    struct hostent __hostEntry; \
+    char           __buffer[5000]; \
+    int            __h_errno; \
+    \
+    addressCount = 0; \
+    \
+    if (gethostbyname_r(name, &__hostEntry, __buffer, sizeof(__buffer), &__h_errno) != NULL) \
+      { \
+        while ((addressCount < maxAddressSize) && (__hostEntry.h_addr_list[addressCount] != NULL)) \
+          { \
+            addresses[addressCount] = ntohl(*(int*)(__hostEntry.h_addr_list[addressCount])); \
+            addressCount++; \
+          } \
+        result = TARGET_NATIVE_OK; \
+      } \
+    else \
+      { \
+        result = TARGET_NATIVE_ERROR; \
+      } \
+  } while (0)
+
+/***********************************************************************\
+* Name       : TARGET_NATIVE_NETWORK_SOCKET_RECEIVE_AVAILABLE
+* Purpose    : get number of available bytes for receive
+* Input      : socketDescriptor - socket descriptor
+* Output     : bytesAvailable - available bytes for receive
+*            : result         - TARGET_NATIVE_OK if no error occurred,
+*                               TARGET_NATIVE_ERROR otherwise
+* Return     : -
+* Side-effect: unknown
+* Notes      : -
+\***********************************************************************/
+
+#define TARGET_NATIVE_NETWORK_SOCKET_RECEIVE_AVAILABLE(socketDescriptor,bytesAvailable,result) \
+  do { \
+    int __value; \
+    \
+    bytesAvailable=0; \
+    result = (ioctl(socketDescriptor, I_NREAD, &__value) == 0)?TARGET_NATIVE_OK:TARGET_NATIVE_ERROR; \
+    if (result == TARGET_NATIVE_OK) \
+    { \
+      bytesAvailable = __value; \
+    } \
+  } while (0)
+
+/***********************************************************************\
+* Name       : TARGET_NATIVE_NETWORK_SOCKET_SEND
+* Purpose    : send data to socket
+* Input      : socketDescriptor - socket descriptor
+*            : buffer  - data to send
+*              length  - length of data to send
+* Output     : bytesSent - number of bytes sent, -1 otherwise
+* Side-effect: unknown
+* Notes      : -
+\***********************************************************************/
+
+#define TARGET_NATIVE_NETWORK_SOCKET_SEND(socketDescriptor, buffer, length, bytesSent) \
+  do { \
+    bytesSent = send(socketDescriptor, \
+                     buffer, \
+                     length, \
+                     0 /* MSG_NOSIGNAL not supported by Solaris */ \
+                    ); \
+  } while (0)
+
+/***********************************************************************\
+* Name       : TARGET_NATIVE_NETWORK_SOCKET_SEND_WITH_ADDRESS_PORT
+* Purpose    : send data to socket
+* Input      : socketDescriptor - socket descriptor
+*            : buffer  - data to send
+*              length  - length of data to send
+*              Address - to address (NOT in network byte order!)
+*              Port    - to port (NOT in network byte order!)
+* Output     : bytesSent - number of bytes sent, -1 otherwise
+* Side-effect: unknown
+* Notes      : -
+\***********************************************************************/
+
+#define TARGET_NATIVE_NETWORK_SOCKET_SEND_WITH_ADDRESS_PORT(socketDescriptor, buffer, length, address, port, bytesSent) \
+  do { \
+    struct sockaddr_in __socketAddress; \
+    \
+    memset(&__socketAddress,0,sizeof(__socketAddress)); \
+    __socketAddress.sin_family      = AF_INET; \
+    __socketAddress.sin_addr.s_addr = htonl(address); \
+    __socketAddress.sin_port        = htons((short)port); \
+    bytesSent = sendto(socketDescriptor, \
+                       buffer, \
+                       length, \
+		       0 /* MSG_NOSIGNAL not supported by solaris */, \
+		       (struct sockaddr*)&__socketAddress, \
+		       sizeof(__socketAddress) \
+                      ); \
+  } while (0)
+
+/***********************************************************************\
+* Name       : TARGET_NATIVE_NETWORK_SOCKET_SET_OPTION_SO_TIMEOUT
+* Purpose    : set socket option SO_TIMEOUT
+* Input      : socketDescriptor - socket descriptor
+*              flag             - 1 or 0
+* Output     : result - TARGET_NATIVE_OK
+* Return     : -
+* Side-effect: unknown
+* Notes      : SO_TIMEOUT not available under Solaris
+\***********************************************************************/
+
+#define TARGET_NATIVE_NETWORK_SOCKET_SET_OPTION_SO_TIMEOUT(socketDescriptor,flag,result) \
+  do { \
+    result = TARGET_NATIVE_OK; \
+  } while (0)
+
+/***********************************************************************\
+* Name       : TARGET_NATIVE_NETWORK_SOCKET_GET_OPTION_SO_TIMEOUT
+* Purpose    : get socket option SO_TIMEOUT
+* Input      : socketDescriptor - socket descriptor
+* Output     : flag   - 1 or 0
+*              result - TARGET_NATIVE_OK
+* Return     : -
+* Side-effect: unknown
+* Notes      : SO_TIMEOUT not available under Solaris
+\***********************************************************************/
+
+#define TARGET_NATIVE_NETWORK_SOCKET_GET_OPTION_SO_TIMEOUT(socketDescriptor,flag,result) \
+  do { \
+    result = TARGET_NATIVE_OK; \
+  } while (0)
+
+/***************************** Functions *******************************/
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#ifdef __cplusplus
+}
+#endif
+
+/* include rest of definitions from generic file (do not move it to 
+   another position!) */
+#include "target_generic_network.h"
+
+#endif /* __TARGET_NATIVE_NETWORK__ */
+
+/* end of file */

--------------040505090707060700070708
Content-Type: text/plain;
 name="target-layer-embOS.patch"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline;
 filename="target-layer-embOS.patch"

diff -ruN --exclude='.*' --exclude='Entries*' --exclude=Repository --exclude=Root embOS/Makefile.am embOS-new/Makefile.am
--- embOS/Makefile.am	Thu Jan  1 01:00:00 1970
+++ embOS-new/Makefile.am	Fri Aug  6 16:35:21 2004
@@ -0,0 +1,11 @@
+## Input file for automake to generate the Makefile.in used by configure
+
+EXTRA_DIST = \
+target_native.h \
+target_native_file.h \
+target_native_io.h \
+target_native_math_float.h \
+target_native_math_int.h
+target_native_memory.h
+target_native_misc.h \
+target_native_network.h \
diff -ruN --exclude='.*' --exclude='Entries*' --exclude=Repository --exclude=Root embOS/target_native.h embOS-new/target_native.h
--- embOS/target_native.h	Thu Jan  1 01:00:00 1970
+++ embOS-new/target_native.h	Fri Aug  6 16:31:36 2004
@@ -0,0 +1,78 @@
+/* ???.h - ???
+   Copyright (C) 1998 Free Software Foundation, Inc.
+
+This file is part of GNU Classpath.
+
+GNU Classpath is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2, or (at your option)
+any later version.
+ 
+GNU Classpath is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Classpath; see the file COPYING.  If not, write to the
+Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
+02111-1307 USA.
+
+Linking this library statically or dynamically with other modules is
+making a combined work based on this library.  Thus, the terms and
+conditions of the GNU General Public License cover the whole
+combination.
+
+As a special exception, the copyright holders of this library give you
+permission to link this library with independent modules to produce an
+executable, regardless of the license terms of these independent
+modules, and to copy and distribute the resulting executable under
+terms of your choice, provided that you also meet, for each linked
+independent module, the terms and conditions of the license of that
+module.  An independent module is a module which is not derived from
+or based on this library.  If you modify this library, you may extend
+this exception to your version of the library, but you are not
+obligated to do so.  If you do not wish to do so, delete this
+exception statement from your version. */
+
+/*
+Description: embOS target global defintions
+Systems    : all
+*/
+
+#ifndef __TARGET_NATIVE__
+#define __TARGET_NATIVE__
+
+/****************************** Includes *******************************/
+/* do not move; needed here because of some macro definitions */
+#include <config.h>
+
+#include <stdlib.h>
+
+/****************** Conditional compilation switches *******************/
+
+/***************************** Constants *******************************/
+
+/***************************** Datatypes *******************************/
+
+/***************************** Variables *******************************/
+
+/****************************** Macros *********************************/
+
+/***************************** Functions *******************************/
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#ifdef __cplusplus
+}
+#endif
+
+/* include rest of definitions from generic file (do not move it to 
+   another position!) */
+#include "target_generic.h"
+
+#endif /* __TARGET_NATIVE__ */
+
+/* end of file */
diff -ruN --exclude='.*' --exclude='Entries*' --exclude=Repository --exclude=Root embOS/target_native_file.h embOS-new/target_native_file.h
--- embOS/target_native_file.h	Thu Jan  1 01:00:00 1970
+++ embOS-new/target_native_file.h	Fri Aug  6 16:31:36 2004
@@ -0,0 +1,123 @@
+/* ???.h - ???
+   Copyright (C) 1998 Free Software Foundation, Inc.
+
+This file is part of GNU Classpath.
+
+GNU Classpath is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2, or (at your option)
+any later version.
+ 
+GNU Classpath is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Classpath; see the file COPYING.  If not, write to the
+Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
+02111-1307 USA.
+
+Linking this library statically or dynamically with other modules is
+making a combined work based on this library.  Thus, the terms and
+conditions of the GNU General Public License cover the whole
+combination.
+
+As a special exception, the copyright holders of this library give you
+permission to link this library with independent modules to produce an
+executable, regardless of the license terms of these independent
+modules, and to copy and distribute the resulting executable under
+terms of your choice, provided that you also meet, for each linked
+independent module, the terms and conditions of the license of that
+module.  An independent module is a module which is not derived from
+or based on this library.  If you modify this library, you may extend
+this exception to your version of the library, but you are not
+obligated to do so.  If you do not wish to do so, delete this
+exception statement from your version. */
+
+/*
+Description: embOS target defintions of file functions
+Systems    : all
+*/
+
+#ifndef __TARGET_NATIVE_FILE__
+#define __TARGET_NATIVE_FILE__
+
+/****************************** Includes *******************************/
+/* do not move; needed here because of some macro definitions */
+#include <config.h>
+
+#include <stdlib.h>
+#include "filsystypedef.h"
+/* can not be used, because of collisions with stdio.h: #include "functionConverter.h" */
+
+/****************** Conditional compilation switches *******************/
+
+/***************************** Constants *******************************/
+
+/***************************** Datatypes *******************************/
+
+/***************************** Variables *******************************/
+
+/****************************** Macros *********************************/
+
+#define TARGET_NATIVE_FILE_OPEN_READ(filename,filedescriptor,result) \
+  do  { \
+    FFFILE *__fileHandle; \
+    \
+    __fileHandle = IOF_fopen(filename,"r"); \
+    if (__fileHandle != NULL) \
+    { \
+      filedescriptor = (int)(__fileHandle); \
+      result = TARGET_NATIVE_OK; \
+    } \
+    else \
+    { \
+      result = TARGET_NATIVE_ERROR; \
+    } \
+ } while (0)
+
+#define TARGET_NATIVE_FILE_CLOSE(filedescriptor,result) \
+  do  { \
+    FFFILE *__fileHandle; \
+    \
+    __fileHandle = (FFFILE*)filedescriptor; \
+    result = (IOF_fclose(__fileHandle)==0)?TARGET_NATIVE_OK:TARGET_NATIVE_ERROR; \
+ } while (0)
+
+#define TARGET_NATIVE_FILE_SIZE(filedescriptor,length,result) \
+  do { \
+    FFFILE *__fileHandle; \
+    \
+    __fileHandle = (FFFILE*)filedescriptor; \
+    length = IOF_flengthOpened(__fileHandle); \
+    result = (length >= 0)?TARGET_NATIVE_OK:TARGET_NATIVE_ERROR; \
+  } while (0)
+
+#define TARGET_NATIVE_FILE_READ(filedescriptor,buffer,length,bytesRead,result) \
+  do { \
+    FFFILE *__fileHandle; \
+    \
+    __fileHandle = (FFFILE*)filedescriptor; \
+    bytesRead = IOF_fread(__fileHandle,buffer,length); \
+    result = (bytesRead!=-1)?TARGET_NATIVE_OK:TARGET_NATIVE_ERROR; \
+  } while (0)
+
+/***************************** Functions *******************************/
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#ifdef __cplusplus
+}
+#endif
+
+/* include rest of definitions from generic file (do not move it to 
+   another position!) */
+#include "target_generic_file.h"
+
+#endif /* __TARGET_NATIVE_FILE__ */
+
+/* end of file */
+
diff -ruN --exclude='.*' --exclude='Entries*' --exclude=Repository --exclude=Root embOS/target_native_io.c embOS-new/target_native_io.c
--- embOS/target_native_io.c	Thu Jan  1 01:00:00 1970
+++ embOS-new/target_native_io.c	Fri Aug  6 16:33:13 2004
@@ -0,0 +1,122 @@
+/* generic_math_int64.h - Native methods for 64bit math operations
+   Copyright (C) 1998 Free Software Foundation, Inc.
+
+This file is part of GNU Classpath.
+
+GNU Classpath is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2, or (at your option)
+any later version.
+ 
+GNU Classpath is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Classpath; see the file COPYING.  If not, write to the
+Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
+02111-1307 USA.
+
+Linking this library statically or dynamically with other modules is
+making a combined work based on this library.  Thus, the terms and
+conditions of the GNU General Public License cover the whole
+combination.
+
+As a special exception, the copyright holders of this library give you
+permission to link this library with independent modules to produce an
+executable, regardless of the license terms of these independent
+modules, and to copy and distribute the resulting executable under
+terms of your choice, provided that you also meet, for each linked
+independent module, the terms and conditions of the license of that
+module.  An independent module is a module which is not derived from
+or based on this library.  If you modify this library, you may extend
+this exception to your version of the library, but you are not
+obligated to do so.  If you do not wish to do so, delete this
+exception statement from your version. */
+
+/*
+Description: embOS target defintions of I/O-functions
+Systems    : all
+*/
+
+/****************************** Includes *******************************/
+/* do not move; needed here because of some macro definitions */
+#include "config.h"
+
+#include <stdlib.h>
+#include <stdio.h>
+#include <stdarg.h>
+#include <assert.h>
+
+#include "target_native.h"
+
+#include "target_native_io.h"
+
+/****************** Conditional compilation switches *******************/
+
+/***************************** Constants *******************************/
+
+/***************************** Datatypes *******************************/
+
+/***************************** Variables *******************************/
+
+/****************************** Macros *********************************/
+
+/***************************** Functions *******************************/
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+int JamaicaNativeembOS_print(const char *buffer, int length)
+{
+  static char tmpBuffer[512];
+  int         z1,z2;
+
+  assert(buffer!=NULL);
+
+  /* convert \n -> \r\n */
+  z1=0;
+  z2=0;
+  while ((z1<length) && (z2<sizeof(tmpBuffer)-2))
+  {
+    if (buffer[z1]=='\n')
+    {
+      tmpBuffer[z2]='\r';z2++;
+    }
+    tmpBuffer[z2]=buffer[z1];;
+    z1++;
+    z2++;
+  }
+  tmpBuffer[z2]='\0';
+
+  /* output */
+  tx_putstr(2,tmpBuffer);
+
+  return z2;
+}
+
+int JamaicaNativeembOS_printf(const char *format, ...)
+{
+  va_list arguments;
+  int     n;
+
+  assert(format!=NULL);
+
+  /* format string */
+  va_start(arguments,format);
+  n = vsnprintf(buffer,sizeof(buffer),format,arguments);
+  va_end(arguments);
+
+  /* output */
+  n = JamaicaNativeembOS_print(buffer,n);
+
+  return(n);
+}
+
+#ifdef __cplusplus
+}
+#endif
+
+/* end of file */
diff -ruN --exclude='.*' --exclude='Entries*' --exclude=Repository --exclude=Root embOS/target_native_io.h embOS-new/target_native_io.h
--- embOS/target_native_io.h	Thu Jan  1 01:00:00 1970
+++ embOS-new/target_native_io.h	Fri Aug  6 16:31:36 2004
@@ -0,0 +1,117 @@
+/* ???.h - ???
+   Copyright (C) 1998 Free Software Foundation, Inc.
+
+This file is part of GNU Classpath.
+
+GNU Classpath is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2, or (at your option)
+any later version.
+ 
+GNU Classpath is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Classpath; see the file COPYING.  If not, write to the
+Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
+02111-1307 USA.
+
+Linking this library statically or dynamically with other modules is
+making a combined work based on this library.  Thus, the terms and
+conditions of the GNU General Public License cover the whole
+combination.
+
+As a special exception, the copyright holders of this library give you
+permission to link this library with independent modules to produce an
+executable, regardless of the license terms of these independent
+modules, and to copy and distribute the resulting executable under
+terms of your choice, provided that you also meet, for each linked
+independent module, the terms and conditions of the license of that
+module.  An independent module is a module which is not derived from
+or based on this library.  If you modify this library, you may extend
+this exception to your version of the library, but you are not
+obligated to do so.  If you do not wish to do so, delete this
+exception statement from your version. */
+
+/*
+Description: embOS target defintions of miscellaneous functions
+Systems    : all
+*/
+
+#ifndef __TARGET_NATIVE_IO__
+#define __TARGET_NATIVE_IO__
+
+/****************************** Includes *******************************/
+/* do not move; needed here because of some macro definitions */
+#include <config.h>
+
+#include <stdlib.h>
+#include <stdarg.h>
+
+#include "platform/comm/async_nativeserialport.h"
+
+/****************** Conditional compilation switches *******************/
+
+/***************************** Constants *******************************/
+
+/***************************** Datatypes *******************************/
+
+/***************************** Variables *******************************/
+
+/****************************** Macros *********************************/
+
+#define TARGET_NATIVE_IO_PRINT \
+ JamaicaNativeembOS_printf
+#define TARGET_NATIVE_IO_PRINT_ERROR \
+  JamaicaNativeembOS_printf
+
+#define TARGET_NATIVE_IO_READ_STDIN(buffer,length,bytesRead,result) \
+  do { \
+    bytesRead = 0; \
+    result = TARGET_NATIVE_ERROR; \
+  } while(0)
+
+// NYI ??? buffer length?
+#define TARGET_NATIVE_IO_WRITE_STDOUT(buffer,length,bytesWritten,result) \
+  do { \
+    bytesWritten = JamaicaNativeembOS_print(buffer,length); \
+    result = TARGET_NATIVE_OK; \
+  } while (0)
+#define TARGET_NATIVE_IO_WRITE_STDERR(buffer,length,bytesWritten,result) \
+  do { \
+    bytesWritten = JamaicaNativeembOS_print(buffer,length); \
+    result = TARGET_NATIVE_OK; \
+  } while (0)
+
+/***************************** Functions *******************************/
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+/** print string (convert LF -> CRLF)
+ * @param buffer string
+ * @param length length
+ * @return number of characters printed
+ */
+int JamaicaNativeembOS_print(const char *buffer, int length);
+
+/** print formated string to stdout
+ * @param format format string
+ * @return number of characters printed
+ */
+int JamaicaNativeembOS_printf(const char *format, ...);
+
+#ifdef __cplusplus
+}
+#endif
+
+/* include rest of definitions from generic file (do not move it to 
+   another position!) */
+#include "target_generic_io.h"
+
+#endif /* __TARGET_NATIVE_MISC__ */
+
+/* end of file */
diff -ruN --exclude='.*' --exclude='Entries*' --exclude=Repository --exclude=Root embOS/target_native_math_float.h embOS-new/target_native_math_float.h
--- embOS/target_native_math_float.h	Thu Jan  1 01:00:00 1970
+++ embOS-new/target_native_math_float.h	Fri Aug  6 16:31:36 2004
@@ -0,0 +1,80 @@
+/* ???.h - ???
+   Copyright (C) 1998 Free Software Foundation, Inc.
+
+This file is part of GNU Classpath.
+
+GNU Classpath is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2, or (at your option)
+any later version.
+ 
+GNU Classpath is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Classpath; see the file COPYING.  If not, write to the
+Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
+02111-1307 USA.
+
+Linking this library statically or dynamically with other modules is
+making a combined work based on this library.  Thus, the terms and
+conditions of the GNU General Public License cover the whole
+combination.
+
+As a special exception, the copyright holders of this library give you
+permission to link this library with independent modules to produce an
+executable, regardless of the license terms of these independent
+modules, and to copy and distribute the resulting executable under
+terms of your choice, provided that you also meet, for each linked
+independent module, the terms and conditions of the license of that
+module.  An independent module is a module which is not derived from
+or based on this library.  If you modify this library, you may extend
+this exception to your version of the library, but you are not
+obligated to do so.  If you do not wish to do so, delete this
+exception statement from your version. */
+
+/*
+Description: embOS target defintions of float/double constants/
+             macros/functions
+Systems    : all
+*/
+
+#ifndef __TARGET_NATIVE_MATH_FLOAT__
+#define __TARGET_NATIVE_MATH_FLOAT__
+
+/****************************** Includes *******************************/
+/* do not move; needed here because of some macro definitions */
+#include <config.h>
+
+#include <stdlib.h>
+
+/****************** Conditional compilation switches *******************/
+
+/***************************** Constants *******************************/
+
+/***************************** Datatypes *******************************/
+
+/***************************** Variables *******************************/
+
+/****************************** Macros *********************************/
+
+/***************************** Functions *******************************/
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#ifdef __cplusplus
+}
+#endif
+
+/* include rest of definitions from generic file (do not move it to 
+   another position!) */
+#include "target_generic_math_float.h"
+
+#endif /* __TARGET_NATIVE_MATH_FLOAT__ */
+
+/* end of file */
+
diff -ruN --exclude='.*' --exclude='Entries*' --exclude=Repository --exclude=Root embOS/target_native_math_int.h embOS-new/target_native_math_int.h
--- embOS/target_native_math_int.h	Thu Jan  1 01:00:00 1970
+++ embOS-new/target_native_math_int.h	Fri Aug  6 16:31:36 2004
@@ -0,0 +1,80 @@
+/* ???.h - ???
+   Copyright (C) 1998 Free Software Foundation, Inc.
+
+This file is part of GNU Classpath.
+
+GNU Classpath is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2, or (at your option)
+any later version.
+ 
+GNU Classpath is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Classpath; see the file COPYING.  If not, write to the
+Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
+02111-1307 USA.
+
+Linking this library statically or dynamically with other modules is
+making a combined work based on this library.  Thus, the terms and
+conditions of the GNU General Public License cover the whole
+combination.
+
+As a special exception, the copyright holders of this library give you
+permission to link this library with independent modules to produce an
+executable, regardless of the license terms of these independent
+modules, and to copy and distribute the resulting executable under
+terms of your choice, provided that you also meet, for each linked
+independent module, the terms and conditions of the license of that
+module.  An independent module is a module which is not derived from
+or based on this library.  If you modify this library, you may extend
+this exception to your version of the library, but you are not
+obligated to do so.  If you do not wish to do so, delete this
+exception statement from your version. */
+
+/*
+Description: embOS target defintions of int/int64 constants/
+             macros/functions
+Systems    : all
+*/
+
+#ifndef __TARGET_NATIVE_MATH_INT__
+#define __TARGET_NATIVE_MATH_INT__
+
+/****************************** Includes *******************************/
+/* do not move; needed here because of some macro definitions */
+#include <config.h>
+
+#include <stdlib.h>
+
+/****************** Conditional compilation switches *******************/
+
+/***************************** Constants *******************************/
+
+/***************************** Datatypes *******************************/
+
+/***************************** Variables *******************************/
+
+/****************************** Macros *********************************/
+
+/***************************** Functions *******************************/
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#ifdef __cplusplus
+}
+#endif
+
+/* include rest of definitions from generic file (do not move it to 
+   another position!) */
+#include "target_generic_math_int.h"
+
+#endif /* __TARGET_NATIVE_MATH_INT__ */
+
+/* end of file */
+
diff -ruN --exclude='.*' --exclude='Entries*' --exclude=Repository --exclude=Root embOS/target_native_memory.h embOS-new/target_native_memory.h
--- embOS/target_native_memory.h	Thu Jan  1 01:00:00 1970
+++ embOS-new/target_native_memory.h	Fri Aug  6 16:31:36 2004
@@ -0,0 +1,78 @@
+/* ???.h - ???
+   Copyright (C) 1998 Free Software Foundation, Inc.
+
+This file is part of GNU Classpath.
+
+GNU Classpath is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2, or (at your option)
+any later version.
+ 
+GNU Classpath is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Classpath; see the file COPYING.  If not, write to the
+Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
+02111-1307 USA.
+
+Linking this library statically or dynamically with other modules is
+making a combined work based on this library.  Thus, the terms and
+conditions of the GNU General Public License cover the whole
+combination.
+
+As a special exception, the copyright holders of this library give you
+permission to link this library with independent modules to produce an
+executable, regardless of the license terms of these independent
+modules, and to copy and distribute the resulting executable under
+terms of your choice, provided that you also meet, for each linked
+independent module, the terms and conditions of the license of that
+module.  An independent module is a module which is not derived from
+or based on this library.  If you modify this library, you may extend
+this exception to your version of the library, but you are not
+obligated to do so.  If you do not wish to do so, delete this
+exception statement from your version. */
+
+/*
+Description: embOS target defintions of memory functions
+Systems    : all
+*/
+
+#ifndef __TARGET_NATIVE_MEMORY__
+#define __TARGET_NATIVE_MEMORY__
+
+/****************************** Includes *******************************/
+/* do not move; needed here because of some macro definitions */
+#include <config.h>
+
+#include <stdlib.h>
+
+/****************** Conditional compilation switches *******************/
+
+/***************************** Constants *******************************/
+
+/***************************** Datatypes *******************************/
+
+/***************************** Variables *******************************/
+
+/****************************** Macros *********************************/
+
+/***************************** Functions *******************************/
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#ifdef __cplusplus
+}
+#endif
+
+/* include rest of definitions from generic file (do not move it to 
+   another position!) */
+#include "target_generic_memory.h"
+
+#endif /* __TARGET_NATIVE_MEMORY__ */
+
+/* end of file */
diff -ruN --exclude='.*' --exclude='Entries*' --exclude=Repository --exclude=Root embOS/target_native_misc.h embOS-new/target_native_misc.h
--- embOS/target_native_misc.h	Thu Jan  1 01:00:00 1970
+++ embOS-new/target_native_misc.h	Fri Aug  6 16:31:36 2004
@@ -0,0 +1,84 @@
+/* ???.h - ???
+   Copyright (C) 1998 Free Software Foundation, Inc.
+
+This file is part of GNU Classpath.
+
+GNU Classpath is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2, or (at your option)
+any later version.
+ 
+GNU Classpath is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Classpath; see the file COPYING.  If not, write to the
+Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
+02111-1307 USA.
+
+Linking this library statically or dynamically with other modules is
+making a combined work based on this library.  Thus, the terms and
+conditions of the GNU General Public License cover the whole
+combination.
+
+As a special exception, the copyright holders of this library give you
+permission to link this library with independent modules to produce an
+executable, regardless of the license terms of these independent
+modules, and to copy and distribute the resulting executable under
+terms of your choice, provided that you also meet, for each linked
+independent module, the terms and conditions of the license of that
+module.  An independent module is a module which is not derived from
+or based on this library.  If you modify this library, you may extend
+this exception to your version of the library, but you are not
+obligated to do so.  If you do not wish to do so, delete this
+exception statement from your version. */
+
+/*
+Description: embOS target defintions of miscellaneous functions
+Systems    : all
+*/
+
+#ifndef __TARGET_NATIVE_MISC__
+#define __TARGET_NATIVE_MISC__
+
+/****************************** Includes *******************************/
+/* do not move; needed here because of some macro definitions */
+#include <config.h>
+
+#include <stdlib.h>
+
+/****************** Conditional compilation switches *******************/
+
+/***************************** Constants *******************************/
+
+/***************************** Datatypes *******************************/
+
+/***************************** Variables *******************************/
+
+/****************************** Macros *********************************/
+
+#define TARGET_NATIVE_MISC_GET_TIMEZONE_STRING(string,maxStringLength,result) \
+  do { \
+    string[0]='\0'; \
+  } while (0)
+
+/***************************** Functions *******************************/
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#ifdef __cplusplus
+}
+#endif
+
+/* include rest of definitions from generic file (do not move it to 
+   another position!) */
+#include "target_generic_misc.h"
+
+#endif /* __TARGET_NATIVE_MISC__ */
+
+/* end of file */
+
diff -ruN --exclude='.*' --exclude='Entries*' --exclude=Repository --exclude=Root embOS/target_native_network.h embOS-new/target_native_network.h
--- embOS/target_native_network.h	Thu Jan  1 01:00:00 1970
+++ embOS-new/target_native_network.h	Fri Aug  6 16:31:36 2004
@@ -0,0 +1,80 @@
+/* ???.h - ???
+   Copyright (C) 1998 Free Software Foundation, Inc.
+
+This file is part of GNU Classpath.
+
+GNU Classpath is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2, or (at your option)
+any later version.
+ 
+GNU Classpath is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Classpath; see the file COPYING.  If not, write to the
+Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
+02111-1307 USA.
+
+Linking this library statically or dynamically with other modules is
+making a combined work based on this library.  Thus, the terms and
+conditions of the GNU General Public License cover the whole
+combination.
+
+As a special exception, the copyright holders of this library give you
+permission to link this library with independent modules to produce an
+executable, regardless of the license terms of these independent
+modules, and to copy and distribute the resulting executable under
+terms of your choice, provided that you also meet, for each linked
+independent module, the terms and conditions of the license of that
+module.  An independent module is a module which is not derived from
+or based on this library.  If you modify this library, you may extend
+this exception to your version of the library, but you are not
+obligated to do so.  If you do not wish to do so, delete this
+exception statement from your version. */
+
+/*
+Description: embOS target defintions of network functions
+Systems    : all
+*/
+
+#ifndef __TARGET_NATIVE_NETWORK__
+#define __TARGET_NATIVE_NETWORK__
+
+/****************************** Includes *******************************/
+/* do not move; needed here because of some macro definitions */
+#include <config.h>
+
+#include <stdlib.h>
+
+#include "target_native.h"
+
+/****************** Conditional compilation switches *******************/
+
+/***************************** Constants *******************************/
+
+/***************************** Datatypes *******************************/
+
+/***************************** Variables *******************************/
+
+/****************************** Macros *********************************/
+
+/***************************** Functions *******************************/
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#ifdef __cplusplus
+}
+#endif
+
+/* include rest of definitions from generic file (do not move it to 
+   another position!) */
+#include "target_generic_network.h"
+
+#endif /* __TARGET_NATIVE_NETWORK__ */
+
+/* end of file */

--------------040505090707060700070708--




From MAILER-DAEMON Mon Aug 16 11:07:26 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1Bwj54-0005Ty-8W
	for mharc-classpath@gnu.org; Mon, 16 Aug 2004 11:07:26 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1Bwj51-0005PZ-4w
	for classpath@gnu.org; Mon, 16 Aug 2004 11:07:23 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1Bwj4y-0005N8-FW
	for classpath@gnu.org; Mon, 16 Aug 2004 11:07:22 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1Bwj4y-0005Md-Bd
	for classpath@gnu.org; Mon, 16 Aug 2004 11:07:20 -0400
Received: from [209.152.205.226] (helo=NR-EXCHANGE.netreach.core)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1Bwj0n-00008t-DZ
	for classpath@gnu.org; Mon, 16 Aug 2004 11:03:01 -0400
Received: from [209.152.205.123] ([209.152.205.123]) by
	NR-EXCHANGE.netreach.core with Microsoft SMTPSVC(6.0.3790.0); 
	Mon, 16 Aug 2004 11:05:34 -0400
Message-ID: <4120CC9F.4050604@netreach.com>
Date: Mon, 16 Aug 2004 11:02:55 -0400
From: Stuart Ballard <sballard@netreach.com>
User-Agent: Mozilla Thunderbird 0.6 (X11/20040605)
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: Mark Wielaard <mark@klomp.org>
References: <87wu0acpkx.fsf@gnu.org>
	<1092412573.1592.84.camel@localhost>	<411D0007.6050109@netreach.com>
	<87smaqbyg5.fsf@gnu.org> <1092483703.1586.282.camel@localhost>
In-Reply-To: <1092483703.1586.282.camel@localhost>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
X-OriginalArrivalTime: 16 Aug 2004 15:05:34.0487 (UTC)
	FILETIME=[7B01C270:01C483A2]
Cc: classpath@gnu.org
Subject: Re: [cp-patches] Updated: crypto merge patch
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Mon, 16 Aug 2004 15:07:23 -0000

Mark Wielaard wrote:
> The reason I don't want to do this with Tritonus is because we don't
> have a very good track record of following upstream sources.  We
> currently only have GNU JAXP as external library shipped with GNU
> Classpath, and we are horribly out of date (libgcj is even more out of
> date), so I am actually thinking of dropping it and point people to the
> official GNU JAXP upstream sources.

If JAXP, Tritonus etc can be used out of the box with Classpath, would 
it be possible to make the Classpath build process simply grab the 
latest released version from their respective CVS repositories, or 
something?

It seems a shame to just have to say "Classpath out of the box is 
incomplete, go get these other libraries too before you have a complete 
Java environment".

We could make them "dependencies" of classpath, but then you end up with 
some weird inverted dependency relationships where the other packages 
require classpath to build, but classpath is marked as requiring the 
other packages.

Stuart.

-- 
Stuart Ballard, Senior Web Developer
NetReach, Inc.
(215) 283-2300, ext. 126
http://www.netreach.com/



From MAILER-DAEMON Wed Aug 18 20:42:32 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1Bxb0i-0008GO-FQ
	for mharc-classpath@gnu.org; Wed, 18 Aug 2004 20:42:32 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1Bxb0g-0008EM-08
	for classpath@gnu.org; Wed, 18 Aug 2004 20:42:30 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1Bxb0e-0008CI-Dc
	for classpath@gnu.org; Wed, 18 Aug 2004 20:42:29 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1Bxb0e-0008CF-BA
	for classpath@gnu.org; Wed, 18 Aug 2004 20:42:28 -0400
Received: from [217.12.12.141] (helo=smtp804.mail.ukl.yahoo.com)
	by monty-python.gnu.org with smtp (Exim 4.34) id 1Bxavx-0003yg-Ru
	for classpath@gnu.org; Wed, 18 Aug 2004 20:37:38 -0400
Received: from unknown (HELO rivendell) (classpath@gnu.org@81.152.170.101 with
	poptime)
	by smtp804.mail.ukl.yahoo.com with SMTP; 19 Aug 2004 00:37:36 -0000
Received: from gandalf by rivendell with local (Exim 4.34) id 1Bxavw-0006Oh-Gn
	for classpath@gnu.org; Thu, 19 Aug 2004 01:37:36 +0100
Date: Thu, 19 Aug 2004 01:37:36 +0100
To: classpath@gnu.org
Message-ID: <20040819003736.GA24589@middle-earth.co.uk>
Mime-Version: 1.0
Content-Type: multipart/signed; micalg=pgp-sha1;
	protocol="application/pgp-signature"; boundary="uZ3hkaAS1mZxFaxD"
Content-Disposition: inline
User-Agent: Mutt/1.5.6+20040722i
From: Andrew John Hughes <gnu_andrew@member.fsf.org>
Subject: Mauve patches for SpinnerListModel
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Thu, 19 Aug 2004 00:42:30 -0000


--uZ3hkaAS1mZxFaxD
Content-Type: multipart/mixed; boundary="45Z9DzgjV8m4Oswq"
Content-Disposition: inline


--45Z9DzgjV8m4Oswq
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

Attached is a series of Mauve tests for SpinnerListModel
to go in the directory gnu/testlet/javax/swing/SpinnerListModel.
What do I do about adding these?

Also, does anyone have any idea as to why Mauve only seems to test java.io =
and
java.net when I run make check?  Using ant, I can get it to compile all the=
 tests
but it segfaults in java.lang.reflect.Array.newInstance.

Cheers,
--=20
Andrew :-)

Please avoid sending me Word or PowerPoint attachments.
See http://www.fsf.org/philosophy/no-word-attachments.html

Value your freedom, or you will lose it, teaches history.=20
`Don't bother us with politics' respond those who don't want to learn.=20


--45Z9DzgjV8m4Oswq
Content-Type: text/plain; charset=us-ascii
Content-Disposition: attachment; filename="spinnerlistmodel-tests.diff"
Content-Transfer-Encoding: quoted-printable

--- /dev/null	2004-08-16 21:12:55.000000000 +0100
+++ ArrayModel.java	2004-08-19 01:34:49.000000000 +0100
@@ -0,0 +1,62 @@
+// Tags: JDK1.4
+
+// Copyright (C) 2004 Andrew John Hughes <gnu_andrew@member.fsf.org>
+
+// This file is part of Mauve.
+
+// Mauve is free software; you can redistribute it and/or modify
+// it under the terms of the GNU General Public License as published by
+// the Free Software Foundation; either version 2, or (at your option)
+// any later version.
+
+// Mauve is distributed in the hope that it will be useful,
+// but WITHOUT ANY WARRANTY; without even the implied warranty of
+// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+// GNU General Public License for more details.
+
+// You should have received a copy of the GNU General Public License
+// along with Mauve; see the file COPYING.  If not, write to
+// the Free Software Foundation, 59 Temple Place - Suite 330,
+// Boston, MA 02111-1307, USA.
+
+package gnu.testlet.javax.swing.SpinnerListModel;
+
+import gnu.testlet.Testlet;
+import gnu.testlet.TestHarness;
+import javax.swing.SpinnerListModel;
+
+public class ArrayModel implements Testlet
+{
+
+  public void test(TestHarness harness)
+  {
+    SpinnerListModel model;
+    Object[] array;
+   =20
+    /* Create array */
+    array =3D new Object[]{"GNU", "Classpath"};
+    /* Create model */
+    model =3D new SpinnerListModel(array);
+    /* Check retrieval */
+    harness.check(model.getList() !=3D null, "Array model creation check");
+    harness.check(model.getValue(), "GNU", "Array model current value chec=
k");
+    harness.check(model.getNextValue(), "Classpath", "Array model next val=
ue check");
+    harness.check(model.getValue(), "GNU", "Array model no change of curre=
nt value after next check");
+    harness.check(model.getPreviousValue(), null, "Array model previous va=
lue check");
+    /* Value change check */
+    array[0] =3D "GNU's Not UNIX";
+    harness.check(model.getValue(), "GNU's Not UNIX", "Array model backing=
 list change check");
+    /* Value setting check */
+    model.setValue("Classpath");
+    harness.check(model.getValue(), "Classpath", "Array model successful s=
et check");
+    try
+      {
+        model.setValue("Sun");
+        harness.fail("Exception not thrown for non-existant value with arr=
ay model.");
+      }
+    catch (IllegalArgumentException exception)
+      {
+        /* Success */
+      }
+  }
+}
--- /dev/null	2004-08-16 21:12:55.000000000 +0100
+++ Constructors.java	2004-08-19 01:34:49.000000000 +0100
@@ -0,0 +1,73 @@
+// Tags: JDK1.4
+
+// Copyright (C) 2004 Andrew John Hughes <gnu_andrew@member.fsf.org>
+
+// This file is part of Mauve.
+
+// Mauve is free software; you can redistribute it and/or modify
+// it under the terms of the GNU General Public License as published by
+// the Free Software Foundation; either version 2, or (at your option)
+// any later version.
+
+// Mauve is distributed in the hope that it will be useful,
+// but WITHOUT ANY WARRANTY; without even the implied warranty of
+// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+// GNU General Public License for more details.
+
+// You should have received a copy of the GNU General Public License
+// along with Mauve; see the file COPYING.  If not, write to
+// the Free Software Foundation, 59 Temple Place - Suite 330,
+// Boston, MA 02111-1307, USA.
+
+package gnu.testlet.javax.swing.SpinnerListModel;
+
+import gnu.testlet.Testlet;
+import gnu.testlet.TestHarness;
+import java.util.ArrayList;
+import javax.swing.SpinnerListModel;
+
+public class Constructors implements Testlet
+{
+
+  public void test(TestHarness harness)
+  {
+    SpinnerListModel model;
+
+    /* Invalid values */
+    try
+      {
+        model =3D new SpinnerListModel((Object[]) null);
+        harness.fail("Null array supplied to constructor failed to throw a=
n exception.");
+      }
+    catch (IllegalArgumentException exception)
+      {
+        /* Success */
+      }
+    try
+      {
+        model =3D new SpinnerListModel(new ArrayList());
+        harness.fail("Empty list supplied to constructor failed to throw a=
n exception.");
+      }
+    catch (IllegalArgumentException exception)
+      {
+        /* Success */
+      }
+    try
+      {
+        model =3D new SpinnerListModel(new Object[]{});
+        harness.fail("Empty array supplied to constructor failed to throw =
an exception.");
+      }
+    catch (IllegalArgumentException exception)
+      {
+        /* Success */
+      }
+
+    /* Test the default model */
+    model =3D new SpinnerListModel();
+    harness.check(model.getList() !=3D null, "Default list construction ch=
eck.");
+    harness.check(model.getValue(), "empty", "Default list current value c=
heck.");
+    harness.check(model.getNextValue(), null, "Default list next value che=
ck.");
+    harness.check(model.getPreviousValue(), null, "Default list previous v=
alue check.");
+
+  }
+}
--- /dev/null	2004-08-16 21:12:55.000000000 +0100
+++ ListModel.java	2004-08-19 01:34:49.000000000 +0100
@@ -0,0 +1,66 @@
+// Tags: JDK1.4
+
+// Copyright (C) 2004 Andrew John Hughes <gnu_andrew@member.fsf.org>
+
+// This file is part of Mauve.
+
+// Mauve is free software; you can redistribute it and/or modify
+// it under the terms of the GNU General Public License as published by
+// the Free Software Foundation; either version 2, or (at your option)
+// any later version.
+
+// Mauve is distributed in the hope that it will be useful,
+// but WITHOUT ANY WARRANTY; without even the implied warranty of
+// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+// GNU General Public License for more details.
+
+// You should have received a copy of the GNU General Public License
+// along with Mauve; see the file COPYING.  If not, write to
+// the Free Software Foundation, 59 Temple Place - Suite 330,
+// Boston, MA 02111-1307, USA.
+
+package gnu.testlet.javax.swing.SpinnerListModel;
+
+import gnu.testlet.Testlet;
+import gnu.testlet.TestHarness;
+import java.util.ArrayList;
+import java.util.List;
+import javax.swing.SpinnerListModel;
+
+public class ListModel implements Testlet
+{
+
+  public void test(TestHarness harness)
+  {
+    SpinnerListModel model;
+    List list;
+   =20
+    /* Create list */
+    list =3D new ArrayList();
+    list.add("GNU");
+    list.add("Classpath");
+    /* Create model */
+    model =3D new SpinnerListModel(list);
+    /* Check retrieval */
+    harness.check(model.getList() !=3D null, "List model creation check");
+    harness.check(model.getValue(), "GNU", "List model current value check=
");
+    harness.check(model.getNextValue(), "Classpath", "List model next valu=
e check");
+    harness.check(model.getValue(), "GNU", "List model no change of curren=
t value after next check");
+    harness.check(model.getPreviousValue(), null, "List model previous val=
ue check");
+    /* Value change check */
+    list.set(0, "GNU's Not UNIX");
+    harness.check(model.getValue(), "GNU's Not UNIX", "List model backing =
list change check");
+    /* Value setting check */
+    model.setValue("Classpath");
+    harness.check(model.getValue(), "Classpath", "List model successful se=
t check");
+    try
+      {
+        model.setValue("Sun");
+        harness.fail("Exception not thrown for non-existant value with lis=
t model.");
+      }
+    catch (IllegalArgumentException exception)
+      {
+        /* Success */
+      }
+  }
+}
--- /dev/null	2004-08-16 21:12:55.000000000 +0100
+++ Ordering.java	2004-08-19 01:34:49.000000000 +0100
@@ -0,0 +1,58 @@
+// Tags: JDK1.4
+
+// Copyright (C) 2004 Andrew John Hughes <gnu_andrew@member.fsf.org>
+
+// This file is part of Mauve.
+
+// Mauve is free software; you can redistribute it and/or modify
+// it under the terms of the GNU General Public License as published by
+// the Free Software Foundation; either version 2, or (at your option)
+// any later version.
+
+// Mauve is distributed in the hope that it will be useful,
+// but WITHOUT ANY WARRANTY; without even the implied warranty of
+// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+// GNU General Public License for more details.
+
+// You should have received a copy of the GNU General Public License
+// along with Mauve; see the file COPYING.  If not, write to
+// the Free Software Foundation, 59 Temple Place - Suite 330,
+// Boston, MA 02111-1307, USA.
+
+package gnu.testlet.javax.swing.SpinnerListModel;
+
+import gnu.testlet.Testlet;
+import gnu.testlet.TestHarness;
+import java.util.ArrayList;
+import java.util.List;
+import javax.swing.SpinnerListModel;
+
+public class Ordering implements Testlet
+{
+
+  public void test(TestHarness harness)
+  {
+    SpinnerListModel model;
+    List list;
+   =20
+    /* Create list */
+    list =3D new ArrayList();
+    list.add("a");
+    list.add("z");
+    list.add("a");
+    list.add("b");
+    /* Create model */
+    model =3D new SpinnerListModel(list);
+    /* Check retrieval */
+    harness.check(model.getList() !=3D null, "Array model ordering creatio=
n check");
+    harness.check(model.getValue(), "a", "Array model ordering current val=
ue check");
+    harness.check(model.getNextValue(), "z", "Array model ordering next va=
lue check");
+    harness.check(model.getValue(), "a", "Array model ordering no change o=
f current value after next check");
+    harness.check(model.getPreviousValue(), null, "Array model ordering pr=
evious value check");
+    /* Value ordering of setting check */
+    model.setValue("a");
+    harness.check(model.getValue(), "a", "Array model ordering successful =
set check");
+    harness.check(model.getPreviousValue(), null, "Array model ordering po=
st-set previous value check");
+    harness.check(model.getNextValue(), "z", "Array model ordering post-se=
t next value check");
+  }
+}
--- /dev/null	2004-08-16 21:12:55.000000000 +0100
+++ SetList.java	2004-08-19 01:34:49.000000000 +0100
@@ -0,0 +1,67 @@
+// Tags: JDK1.4
+
+// Copyright (C) 2004 Andrew John Hughes <gnu_andrew@member.fsf.org>
+
+// This file is part of Mauve.
+
+// Mauve is free software; you can redistribute it and/or modify
+// it under the terms of the GNU General Public License as published by
+// the Free Software Foundation; either version 2, or (at your option)
+// any later version.
+
+// Mauve is distributed in the hope that it will be useful,
+// but WITHOUT ANY WARRANTY; without even the implied warranty of
+// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+// GNU General Public License for more details.
+
+// You should have received a copy of the GNU General Public License
+// along with Mauve; see the file COPYING.  If not, write to
+// the Free Software Foundation, 59 Temple Place - Suite 330,
+// Boston, MA 02111-1307, USA.
+
+package gnu.testlet.javax.swing.SpinnerListModel;
+
+import gnu.testlet.Testlet;
+import gnu.testlet.TestHarness;
+import java.util.ArrayList;
+import java.util.List;
+import javax.swing.SpinnerListModel;
+
+public class SetList implements Testlet
+{
+
+public void test(TestHarness harness)
+{
+    SpinnerListModel model;
+    List list;
+
+    /* Create default model */
+    model =3D new SpinnerListModel();
+    /* Invalid values */
+    try
+      {
+        model.setList((ArrayList) null);
+        harness.fail("Null list supplied to setList failed to throw an exc=
eption.");
+      }
+    catch (IllegalArgumentException exception)
+      {
+        /* Success */
+      }
+    try
+      {
+        model.setList(new ArrayList());
+        harness.fail("Empty list supplied to setList failed to throw an ex=
ception.");
+      }
+    catch (IllegalArgumentException exception)
+      {
+        /* Success */
+      }
+
+    /* Test a successful change */
+    list =3D new ArrayList();
+    list.add("GNU");
+    model.setList(list);
+    harness.check(model.getList(), list, "Model allowed successful change =
of list.");
+  }
+
+}
--- /dev/null	2004-08-16 21:12:55.000000000 +0100
+++ SpinnerListModelTest.java	2004-08-19 01:34:49.000000000 +0100
@@ -0,0 +1,98 @@
+import java.util.ArrayList;
+import java.util.List;
+import javax.swing.SpinnerListModel;
+
+public class SpinnerListModelTest
+{
+    public static void main(String[] args)
+    {
+	SpinnerListModel model;
+	List list;
+	Object[] array;
+
+	/* Quick constructor tests */
+	try
+	    {
+		model =3D new SpinnerListModel((Object[]) null);
+	    }
+	catch (IllegalArgumentException exception)
+	    {
+		System.out.println("Null correctly threw exception.");
+	    }
+	try
+	    {
+		model =3D new SpinnerListModel(new ArrayList());
+	    }
+	catch (IllegalArgumentException exception)
+	    {
+		System.out.println("Zero elements correctly threw exception.");
+	    }
+	try
+	    {
+		model =3D new SpinnerListModel(new Object[]{});
+	    }
+	catch (IllegalArgumentException exception)
+	    {
+		System.out.println("Zero elements correctly threw exception.");
+	    }
+
+	/* Test the default model */
+	model =3D new SpinnerListModel();
+	System.out.println("Model created with list " + model.getList() + ".");
+	System.out.println("Current value should be 'empty', is " + model.getValu=
e() + ".");
+	System.out.println("Next value should be null, is " + model.getNextValue(=
) + ".");
+	System.out.println("Previous value should be null, is " + model.getPrevio=
usValue() + ".");
+ =09
+	/* Test with a list */
+	list =3D new ArrayList();
+	list.add("GNU");
+	list.add("Classpath");
+	model =3D new SpinnerListModel(list);
+	System.out.println("Model created with list " + model.getList() + ".");
+	System.out.println("Current value should be 'GNU', is " + model.getValue(=
) + ".");
+	System.out.println("Next value should be 'Classpath', is " + model.getNex=
tValue() + ".");
+	System.out.println("Current value should be 'GNU', is " + model.getValue(=
) + ".");=09
+	System.out.println("Previous value should be null, is " + model.getPrevio=
usValue() + ".");=09
+	/* Change a value */
+	list.set(0, "GNU's Not UNIX");
+	System.out.println("Current value should be 'GNU's Not UNIX', is " + mode=
l.getValue() + ".");	=09
+	/* Move to 'Classpath' */
+	model.setValue("Classpath");
+	System.out.println("Current value should be 'Classpath', is " + model.get=
Value() + ",");
+	/* Check exception throwing */
+	try
+	    {
+		model.setValue("Sun");
+		System.out.println("Exception not thrown for non-existant value.");
+	    }
+	catch (IllegalArgumentException exception)
+	    {
+		System.out.println("Value 'Sun' correctly threw exception.");
+	    }
+	/* Test with a array */
+	array =3D new Object[]{"GNU", "Classpath"};
+	model =3D new SpinnerListModel(array);
+	System.out.println("Model created with list " + model.getList() + ".");
+	System.out.println("Current value should be 'GNU', is " + model.getValue(=
) + ".");
+	System.out.println("Next value should be 'Classpath', is " + model.getNex=
tValue() + ".");
+	System.out.println("Current value should be 'GNU', is " + model.getValue(=
) + ".");=09
+	System.out.println("Previous value should be null, is " + model.getPrevio=
usValue() + ".");=09
+	/* Change a value */
+	array[0] =3D "GNU's Not UNIX";
+	System.out.println("Current value should be 'GNU's Not UNIX', is " + mode=
l.getValue() + ".");	=09
+	/* Move to 'Classpath' */
+	model.setValue("Classpath");
+	System.out.println("Current value should be 'Classpath', is " + model.get=
Value() + ",");
+	/* Check exception throwing */
+	try
+	    {
+		model.setValue("Sun");
+		System.out.println("Exception not thrown for non-existant value.");
+	    }
+	catch (IllegalArgumentException exception)
+	    {
+		System.out.println("Value 'Sun' correctly threw exception.");
+	    }
+
+    }
+}

--45Z9DzgjV8m4Oswq--

--uZ3hkaAS1mZxFaxD
Content-Type: application/pgp-signature; name="signature.asc"
Content-Description: Digital signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)

iD8DBQFBI/ZPu+fW6pI0dq4RAvmRAJ9IXrqkZ52NSvx1wPTflw8+C0Dx6ACgi+UX
0c50fQP0LnjFM+lWBa+b3nk=
=NDLL
-----END PGP SIGNATURE-----

--uZ3hkaAS1mZxFaxD--



From MAILER-DAEMON Thu Aug 19 07:43:01 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1BxlJt-0002KV-Jm
	for mharc-classpath@gnu.org; Thu, 19 Aug 2004 07:43:01 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1BxlJr-0002KP-W4
	for classpath@gnu.org; Thu, 19 Aug 2004 07:43:00 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1BxlJq-0002KD-Da
	for classpath@gnu.org; Thu, 19 Aug 2004 07:42:59 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1BxlJq-0002KA-93
	for classpath@gnu.org; Thu, 19 Aug 2004 07:42:58 -0400
Received: from [82.161.94.186] (helo=gnu.wildebeest.org)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1BxlFY-0003ZZ-In
	for classpath@gnu.org; Thu, 19 Aug 2004 07:38:32 -0400
Received: from slauerhoff.wildebeest.org ([192.168.1.27] ident=mark)
	by gnu.wildebeest.org with esmtp (Exim 3.35 #1 (Debian))
	id 1BxlFT-0005O7-00; Thu, 19 Aug 2004 13:38:27 +0200
From: Mark Wielaard <mark@klomp.org>
To: Stuart Ballard <sballard@netreach.com>
In-Reply-To: <4120CC9F.4050604@netreach.com>
References: <87wu0acpkx.fsf@gnu.org> <1092412573.1592.84.camel@localhost>
	<411D0007.6050109@netreach.com>  <87smaqbyg5.fsf@gnu.org>
	<1092483703.1586.282.camel@localhost> <4120CC9F.4050604@netreach.com>
Content-Type: multipart/signed; micalg=pgp-sha1;
	protocol="application/pgp-signature";
	boundary="=-0XbQ0YUgNIZ4fs/2+zSW"
Message-Id: <1092915508.1605.83.camel@localhost>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.4.6 
Date: Thu, 19 Aug 2004 13:38:28 +0200
Cc: classpath@gnu.org
Subject: Re: [cp-patches] Updated: crypto merge patch
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Thu, 19 Aug 2004 11:43:00 -0000


--=-0XbQ0YUgNIZ4fs/2+zSW
Content-Type: text/plain
Content-Transfer-Encoding: quoted-printable

Hi,

On Mon, 2004-08-16 at 17:02, Stuart Ballard wrote:
> Mark Wielaard wrote:
> > The reason I don't want to do this with Tritonus is because we don't
> > have a very good track record of following upstream sources.  We
> > currently only have GNU JAXP as external library shipped with GNU
> > Classpath, and we are horribly out of date (libgcj is even more out of
> > date), so I am actually thinking of dropping it and point people to the
> > official GNU JAXP upstream sources.
>=20
> If JAXP, Tritonus etc can be used out of the box with Classpath, would=20
> it be possible to make the Classpath build process simply grab the=20
> latest released version from their respective CVS repositories, or=20
> something?

We should really make a general extension classloader that works out of
the box with GNU Classpath based compilers and runtimes so you can just
drop new the latest releases of those projects in that directory and
have them used automatically.

Any volunteers?
gcj actually has support for that now. But it is very gcj specific:
http://gcc.gnu.org/ml/java-patches/2004-q2/msg00700.html

Cheers,

Mark

--=-0XbQ0YUgNIZ4fs/2+zSW
Content-Type: application/pgp-signature; name=signature.asc
Content-Description: This is a digitally signed message part

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)

iD8DBQBBJJE0xVhZCJWr9QwRAgf1AKCbeF1huHtfssrXPqwPsHYQH9VUXwCglYNv
z+LaoIpxiDq53p7QzBWJALI=
=B4v6
-----END PGP SIGNATURE-----

--=-0XbQ0YUgNIZ4fs/2+zSW--




From MAILER-DAEMON Thu Aug 26 18:09:14 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1C0SQk-0005c9-MU
	for mharc-classpath@gnu.org; Thu, 26 Aug 2004 18:09:14 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1C0SQj-0005c3-Iq
	for classpath@gnu.org; Thu, 26 Aug 2004 18:09:13 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1C0SQi-0005br-TM
	for classpath@gnu.org; Thu, 26 Aug 2004 18:09:13 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1C0SQi-0005bo-PZ
	for classpath@gnu.org; Thu, 26 Aug 2004 18:09:12 -0400
Received: from [129.34.20.6] (helo=igw2.watson.ibm.com)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1C0SM3-0001Ug-MZ
	for classpath@gnu.org; Thu, 26 Aug 2004 18:04:23 -0400
Received: from sp1n293en1.watson.ibm.com (sp1n293en1.watson.ibm.com
	[129.34.20.41])
	by igw2.watson.ibm.com (8.11.7-20030924/8.11.4) with ESMTP id
	i7QM28M33158 for <classpath@gnu.org>; Thu, 26 Aug 2004 18:02:08 -0400
Received: from [9.2.32.161] (localhost [127.0.0.1])
	by sp1n293en1.watson.ibm.com (8.11.7-20030924/8.11.7/8.11.7-01-14-2004)
	with ESMTP id i7QM4LQ25834; Thu, 26 Aug 2004 18:04:21 -0400
Message-ID: <412E5E62.4030900@watson.ibm.com>
Date: Thu, 26 Aug 2004 18:04:18 -0400
From: Steven Augart <augart@watson.ibm.com>
Organization: IBM Watson Research Ctr., Jikes RVM Project
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7) Gecko/20040616
X-Accept-Language: ast, gl, en-us, en, it, eo, pt, es-ec, es-us, es, fr, de, el
MIME-Version: 1.0
To: classpath@gnu.org
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Subject: Eclipse 3.0 and java.nio
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Thu, 26 Aug 2004 22:09:13 -0000

This letter is to note that Eclipse 3.0, with Classpath, still 
requires that we disable file locking, since Classpath 0.10's java.nio 
implementation does not support it.   There was a brief thread on this 
in November, 2003: 
http://lists.gnu.org/archive/html/classpath/2003-11/msg00119.html  ; 
however, the name of the property has changed.  You now need to set 
the system property osgi.locking=none to turn off the java.nio file 
locking.

--Steve Augart

-- 
Steven Augart

Jikes RVM, a free, open source, Virtual Machine:
http://oss.software.ibm.com/jikesrvm



From MAILER-DAEMON Tue Aug 31 17:55:15 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1C2Gax-0003qr-NJ
	for mharc-classpath@gnu.org; Tue, 31 Aug 2004 17:55:15 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1C2Gav-0003qV-BG
	for classpath@gnu.org; Tue, 31 Aug 2004 17:55:13 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1C2Gau-0003q9-7R
	for classpath@gnu.org; Tue, 31 Aug 2004 17:55:12 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1C2Gau-0003pz-1F
	for classpath@gnu.org; Tue, 31 Aug 2004 17:55:12 -0400
Received: from [82.161.94.186] (helo=gnu.wildebeest.org)
	by monty-python.gnu.org with esmtp (Exim 4.34)
	id 1C2GVp-0004WQ-Ri; Tue, 31 Aug 2004 17:49:58 -0400
Received: from slauerhoff.wildebeest.org ([192.168.1.27] ident=mark)
	by gnu.wildebeest.org with esmtp (Exim 3.35 #1 (Debian))
	id 1C2GVi-0001vf-00; Tue, 31 Aug 2004 23:49:50 +0200
From: Mark Wielaard <mark@klomp.org>
To: classpath@gnu.org
Content-Type: multipart/signed; micalg=pgp-sha1;
	protocol="application/pgp-signature";
	boundary="=-dyNVDEuQkqP0NgTi30ZT"
Message-Id: <1093988996.1144.68.camel@localhost>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.4.6 
Date: Tue, 31 Aug 2004 23:49:56 +0200
Cc: avdyk@gnu.org
Subject: End of the month - where is the new snapshot?
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Tue, 31 Aug 2004 21:55:13 -0000


--=-dyNVDEuQkqP0NgTi30ZT
Content-Type: text/plain
Content-Transfer-Encoding: quoted-printable

Hi all (avdyk and dog CCed for the last point),

It is already that time. End of the (second) month. So we should get a
new snapshot release out. Please try to get everything that you want to
get in for this new 0.11 release snapshot in on Friday September 3.
Then we do the real release on Monday September 6.

This is the NEWS entry I have up to now.
I have probably forgotten something. Please yell and scream if your
favorate new feature is missing!

New in release 0.11 (not released yet)

* javax.swing.Spring and SpringLayout support.
* Added pluggable look and feel support for BasicTextFieldUI and
  BasicToolBarSeparatorUI.
* java.swing.text support for (Default and Layered) Highlighter, FieldView,
  PlainView, TabExpander and TabableView added.
* Start of JTable and JTree implementation.
* Internal Swing frames work.
* JMenu and JPopupMenu work.
* New gtk+ AWT FileDialog peer now based on gtk+2.4 or higher.
* java.awt.image LookupTables and kernel support.
* Improved java.awt.image.BufferedImage support.
* AWT 1.0 event model support.
* GNU Classpath now comes with some example programs (see examples/README).
* New javax.crypto, javax.crypto.interfaces, javax.crypto.spec, javax.net,
  javax.net.ssl, javax.security.auth, javax.security.auth.callback,
  javax.security.auth.login, javax.security.auth.x500, javax.security.sasl
  and org.ietf.jgss packages are now officially part of GNU Classpath.
  Extra crypto algorithms can be obtained from the GNU Crypto project,
  a full TLS implementation is provided by the Jessie project.
  http://www.gnu.org/software/gnu-crypto/
  http://www.nongnu.org/jessie/
* Frame.setIconImage() support.
* AWT GDKGraphics scaling.
* New configure flag --enable-gtk-cairo to build Graphics2D implementation
  build on cairo and pangoft2.  Enabled at runtime by defining the system
  property gnu.java.awt.peer.gtk.Graphics=3DGraphics2D.
* javax.swing.JSpinner implemented.
* Extensive documentation update for java.util collection classes.
* java.awt.geom completed. Area, Arc2D, Ellipse2D and Line2D implemented.

Runtime interface Changes:

* java.lang.Compiler now uses the new java.lang.VMCompiler; there is
  a reference implementation that most VMs can use.
* java.lang.VMSystem has a new getenv(String) method and a reference C/JNI
  implementation that should work on most Posix like systems.
* java.util.TimeZone has been split into a platform independent class and
  a platform dependent class VMTimeZone.  GNU Classpath comes with a generi=
c
  way to get at the default time zone for Posix/GNU-like platforms.
* [VM]AccessController improvements.  In particular it handles
  `doPrivileged' calls better, and allows for recursive `doPrivileged'
  calls in the same Thread. (see vm/reference/java/security/)

There is still a long list of things that I want to see in. We will have
to see how much time I actually have for doing it all. I don't believe
anything (except probably the gjdoc thing) is a real show stopper.

- resync gui-branch (JColor Chooser, JComboBox, javax.swing.text, ...)
- New examples (Drop Animation Applet, JColor Chooser, JComboBox, Area)
- URLConnection.getContent()
- Get gjdoc working again (see rsdio patches)
- Fix jikes 1.21 warnings (lots...)
- khc Timer patch
- dog JNDI file provider
- dog inet HTTP handler
- GNU JAXP sync?

About that last point. What do people think about dropping GNU JAXP as
external library and let runtime, compiler, tool integrators depend on
the GNU JAXP upstream release? Kaffe already does that and I think GNU
JAXP is independent enough to point people at it as external extension
library.

It is not something I really do want to do or don't want to do, but I
notice that for that last two releases we didn't actually sync with GNU
JAXP so it seems we don't really have time for that. In that case it
would be a good idea to just drop it. If on the other hand people really
want to see this shipped together with GNU Classpath (as external
library) then I will put in some effort to finally upgrade it.

Cheers,

Mark.

--=-dyNVDEuQkqP0NgTi30ZT
Content-Type: application/pgp-signature; name=signature.asc
Content-Description: This is a digitally signed message part

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)

iD8DBQBBNPKExVhZCJWr9QwRArp6AJ9CYTBA7VioDEZxMjG2kgjqMkgLAQCfXcFu
h6hse5JqH3V+K90TqZV+jWM=
=J2my
-----END PGP SIGNATURE-----

--=-dyNVDEuQkqP0NgTi30ZT--




From MAILER-DAEMON Tue Aug 31 18:50:29 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1C2HSP-0004Bn-6S
	for mharc-classpath@gnu.org; Tue, 31 Aug 2004 18:50:29 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1C2HSM-0004Ai-JB
	for classpath@gnu.org; Tue, 31 Aug 2004 18:50:26 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1C2HSK-00049T-DM
	for classpath@gnu.org; Tue, 31 Aug 2004 18:50:25 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1C2HSK-000497-9P
	for classpath@gnu.org; Tue, 31 Aug 2004 18:50:24 -0400
Received: from [209.225.28.220] (helo=mxsf20.cluster1.charter.net)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1C2HNE-0006Pj-1J
	for classpath@gnu.org; Tue, 31 Aug 2004 18:45:08 -0400
Received: from mxip02.cluster1.charter.net (mxip02a.cluster1.charter.net
	[209.225.28.132])
	by mxsf20.cluster1.charter.net (8.12.11/8.12.11) with ESMTP id
	i7VMj38J030923
	for <classpath@gnu.org>; Tue, 31 Aug 2004 18:45:03 -0400
Received: from cable-24-196-25-11.mtv.al.charter.com (HELO
	InterJet.dellroad.org) (24.196.25.11)
	by mxip02.cluster1.charter.net with ESMTP; 31 Aug 2004 18:45:03 -0400
X-Ironport-AV: i="3.84,120,1091419200"; 
	d="scan'208"; a="245119629:sNHT12047832"
Received: from arch20m.dellroad.org ([10.104.180.64])
	by InterJet.dellroad.org (8.9.1a/8.9.1) with ESMTP id RAA77515;
	Tue, 31 Aug 2004 17:41:08 -0500 (CDT)
Received: from arch20m.dellroad.org (localhost [127.0.0.1])
	by arch20m.dellroad.org (8.12.9p2/8.12.9) with ESMTP id i7VMf4mW002572; 
	Tue, 31 Aug 2004 17:41:04 -0500 (CDT)
	(envelope-from archie@arch20m.dellroad.org)
Received: (from archie@localhost)
	by arch20m.dellroad.org (8.12.9p2/8.12.9/Submit) id i7VMf4U8002571;
	Tue, 31 Aug 2004 17:41:04 -0500 (CDT) (envelope-from archie)
From: Archie Cobbs <archie@dellroad.org>
Message-Id: <200408312241.i7VMf4U8002571@arch20m.dellroad.org>
In-Reply-To: <1093988996.1144.68.camel@localhost>
To: Mark Wielaard <mark@klomp.org>
Date: Tue, 31 Aug 2004 17:41:04 -0500 (CDT)
X-Mailer: ELM [version 2.4ME+ PL99b (25)]
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Type: text/plain; charset=US-ASCII
Cc: classpath@gnu.org
Subject: Re: End of the month - where is the new snapshot?
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Tue, 31 Aug 2004 22:50:26 -0000

Mark Wielaard wrote:
> - GNU JAXP sync?
> 
> About that last point. What do people think about dropping GNU JAXP as
> external library and let runtime, compiler, tool integrators depend on
> the GNU JAXP upstream release? Kaffe already does that and I think GNU
> JAXP is independent enough to point people at it as external extension
> library.

I agree.. Classpath should avoid trying to duplicate work that
can easily be obtained (freely, of course :-) elsewhere..

-Archie

__________________________________________________________________________
Archie Cobbs      *        CTO, Awarix        *      http://www.awarix.com



From MAILER-DAEMON Tue Aug 31 19:09:22 2004
Received: from mailman by lists.gnu.org with archive (Exim 4.33)
	id 1C2Hkg-00015E-Gr
	for mharc-classpath@gnu.org; Tue, 31 Aug 2004 19:09:22 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.33)
	id 1C2Hkd-00014I-Px
	for classpath@gnu.org; Tue, 31 Aug 2004 19:09:19 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.33)
	id 1C2Hkb-00012w-D5
	for classpath@gnu.org; Tue, 31 Aug 2004 19:09:19 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.33) id 1C2Hka-00012n-V6
	for classpath@gnu.org; Tue, 31 Aug 2004 19:09:17 -0400
Received: from [66.33.197.214] (helo=ludo.dreamhost.com)
	by monty-python.gnu.org with esmtp (Exim 4.34) id 1C2HfN-0001JH-7q
	for classpath@gnu.org; Tue, 31 Aug 2004 19:03:53 -0400
Received: from dexter (c-24-6-46-122.client.comcast.net [24.6.46.122])
	by ludo.dreamhost.com (Postfix) with ESMTP
	id A2A7643CF0; Tue, 31 Aug 2004 16:03:52 -0700 (PDT)
Received: by dexter (Postfix, from userid 1000)
	id 6DA5A39; Tue, 31 Aug 2004 16:02:52 -0700 (PDT)
To: Archie Cobbs <archie@dellroad.org>
References: <200408312241.i7VMf4U8002571@arch20m.dellroad.org>
From: Casey Marshall <csm@gnu.org>
Date: Tue, 31 Aug 2004 16:02:52 -0700
In-Reply-To: <200408312241.i7VMf4U8002571@arch20m.dellroad.org> (Archie
	Cobbs's message of "Tue, 31 Aug 2004 17:41:04 -0500 (CDT)")
Message-ID: <87acwaud0j.fsf@gnu.org>
User-Agent: Gnus/5.1002 (Gnus v5.10.2) Emacs/21.2 (gnu/linux)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Cc: Mark Wielaard <mark@klomp.org>, classpath@gnu.org
Subject: Re: End of the month - where is the new snapshot?
X-BeenThere: classpath@gnu.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussion of the implementation of GNU Classpath libraries
	<classpath.gnu.org>
List-Unsubscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/classpath>
List-Post: <mailto:classpath@gnu.org>
List-Help: <mailto:classpath-request@gnu.org?subject=help>
List-Subscribe: <http://lists.gnu.org/mailman/listinfo/classpath>,
	<mailto:classpath-request@gnu.org?subject=subscribe>
X-List-Received-Date: Tue, 31 Aug 2004 23:09:20 -0000

>>>>> "Archie" == Archie Cobbs <archie@dellroad.org> writes:

Archie> Mark Wielaard wrote:
>> - GNU JAXP sync?
>> 
>> About that last point. What do people think about dropping GNU JAXP
>> as external library and let runtime, compiler, tool integrators
>> depend on the GNU JAXP upstream release? Kaffe already does that
>> and I think GNU JAXP is independent enough to point people at it as
>> external extension library.

Archie> I agree.. Classpath should avoid trying to duplicate work that
Archie> can easily be obtained (freely, of course :-) elsewhere..

Can't we instead just import javax.xml, org.w3c.dom, and org.xml.sax
into classpath, and let JAXP work on the concrete implementations of
that abstract API?

We just did a similar thing with GNU Crypto and Jessie.

-- 
Casey Marshall || csm@gnu.org



