Attached is a patch that fixes the warnings in the NIO files I
changed, and fixes the problem on Darwin (adding the 'volatile'
keyword to the MappedByteBufferImpl instance we create and return
appears to fix this). A bug in GCC 4.0 on that platform, perhaps?
I can't test this properly becau
On Jul 26, 2005, at 12:42 AM, Mark Wielaard wrote:
Hi,
On Sat, 2005-07-23 at 20:37 -0700, Casey Marshall wrote:
Committed. I changed a few minor things, such as using Mark's
suggestion to use 'madvise' to load the file into core.
I can't get this to work with Jamvm on Darwin; I get a bizarre
Ok, I've committed this patch updated with a very minor doc fix
thanks, Aaron
--- /dev/null 2005-06-09 16:29:11.371620296 -0400
+++ gnu/classpath/jdwp/processor/ArrayReferenceCommandSet.java 2005-07-26 20:32:15.0 -0400
@@ -0,0 +1,178 @@
+/* ArrayReferenceCommandSet.java -- class to impleme
> "Roman" == Roman Kennke <[EMAIL PROTECTED]> writes:
Roman> This fix 'outsources' the functionality of the getProperty methods
Roman> into a private method. This avoids infinite recursion if one of the
Roman> getProperty methods is overridden in subclasses.
While this is a totally sensible w
On Wed, 2005-07-20 at 11:14 -0600, Tom Tromey wrote:
> > "Aaron" == Aaron Luchko <[EMAIL PROTECTED]> writes:
>
> Aaron> 2005-07-12 Aaron Luchko <[EMAIL PROTECTED]>
> Aaron> *
> gnu/classpath/jdwp/processor/ThreadGroupReferenceCommandSet.java:
> Aaron>Implemented the ThreadGroup
I'm checking this in.
This fixes PR classpath/22989. We don't need all these overrides that
do nothing, but it was simpler to fix the two bogus calls.
Tom
Index: ChangeLog
from Tom Tromey <[EMAIL PROTECTED]>
PR classpath/22989:
* java/net/Inet4Address.java (isMCNodeLocal): Use
I'm checking this in.
Another bug found by FindBugs. The code in question is obviously
useless; the sense of the test was wrong.
Tom
Index: ChangeLog
from Tom Tromey <[EMAIL PROTECTED]>
PR classpath/22992:
* javax/security/sasl/Sasl.java (getSaslClientFactories): Fix
s
I'm checking this in.
Another fixlet suggested by FindBugs.
Tom
Index: ChangeLog
from Tom Tromey <[EMAIL PROTECTED]>
* gnu/java/security/provider/PKIXCertPathValidatorImpl.java
(engineValidate): Fix sense of test. PR classpath/22991.
Index: gnu/java/security/provider/PKIXCert
Thomas Fitzsimmons wrote:
This patch adds two new state tables to GdkGraphics.c. We were storing
Graphics objects in the GtkGenericPeer state table. Though technically
incorrect, this didn't cause problems on JamVM nor on libgcj but it did
cause problems on VMs with stricter JNI checking. This
On Tue, 2005-07-26 at 18:31 +0200, Mark Wielaard wrote:
> Hi,
>
> On Tue, 2005-07-26 at 12:16 -0400, Thomas Fitzsimmons wrote:
> > On Tue, 2005-07-26 at 07:48 +0200, Mark Wielaard wrote:
> >
> > > I don't think you want to create a (non-daemon) Timer each and every
> > > time here. That means tha
Hi,
This patch adds two new state tables to GdkGraphics.c. We were storing
Graphics objects in the GtkGenericPeer state table. Though technically
incorrect, this didn't cause problems on JamVM nor on libgcj but it did
cause problems on VMs with stricter JNI checking. This patch eliminates
the a
First one simply implements CTRL-Space action (toggle). Second one
replaces calls to KeyEvent.isControlDown() and isShiftDown() with
comparisons between KeyEvent.getModifiers() and InputEvent.SHIFT_MASK
and CTRL_MASK. This way we ignore invalid modifiers for events without
explicitly writing in w
> "Chris" == Chris Burdess <[EMAIL PROTECTED]> writes:
Chris> If the libgcj way is better we should
Chris> use that instead since as I understand it it would save a layer of
Chris> indirection, but perhaps it needs VM-specific support to work?
Yeah, it needs special vm treatment to work.
My i
> "Tom" == Tom Tromey <[EMAIL PROTECTED]> writes:
Tom> This fixes a buglet found by FindBugs and reported as classpath/22988.
Tom> Index: javax/swing/text/AbstractWriter.java
Tom> Index: javax/swing/text/ElementIterator.java
I didn't mean to include these in the patch and I didn't check them
I'm checking this in.
This fixes PR classpath/22987. We were comparing an OID to a String.
The simplest fix seems to be to compare the OID's String
representation instead.
Tom
Index: ChangeLog
from Tom Tromey <[EMAIL PROTECTED]>
* java/security/cert/X509CertSelector.java (match): Conv
I'm checking this in.
This fixes a typo in OID.
Tom
Index: ChangeLog
from Tom Tromey <[EMAIL PROTECTED]>
* gnu/java/security/OID.java: Typo fix.
Index: gnu/java/security/OID.java
===
RCS file: /cvsroot/classpath/classpath
Hi,
On Tue, 2005-07-26 at 12:16 -0400, Thomas Fitzsimmons wrote:
> On Tue, 2005-07-26 at 07:48 +0200, Mark Wielaard wrote:
>
> > I don't think you want to create a (non-daemon) Timer each and every
> > time here. That means that on each repaint() a new Thread is created
> > which is never destroy
I'm checking this in.
This fixes a buglet found by FindBugs and reported as classpath/22988.
Tom
Index: ChangeLog
from Tom Tromey <[EMAIL PROTECTED]>
* javax/security/auth/x500/X500Principal.java (getName): Don't
compare oid to String. PR classpath/22988.
Index: javax/securi
On Tue, 2005-07-26 at 07:48 +0200, Mark Wielaard wrote:
> I don't think you want to create a (non-daemon) Timer each and every
> time here. That means that on each repaint() a new Thread is created
> which is never destroyed and which will prevent the application to ever
> stop since the Timers wi
Hi added and fixed some API comments for Swing.
2005-07-26 Roman Kennke <[EMAIL PROTECTED]>
* javax/swing/AbstractAction.java
Fixed / Added API doc comments for the class description.
* javax/swing/AbstractButton.java: Likewise.
* javax/swing/AbstractCellEditor.
>> (Whoops, accidentally commited a change to Component. Did that revert
>> correctly?)
Mark> No. You committed the attached.
I thought I'd mention a nice cvs trick for reverting a patch. (I've
had to do this more times than I care to admit...)
If you accidentally check in a commit to some file
> "Archit" == Archit Shah <[EMAIL PROTECTED]> writes:
Archit> Attached is a patch to the cp-tools rmic that centralizes calls to
Archit> System.exit and makes handling of exceptions and error conditions more
Archit> uniform.
Please write a ChangeLog entry and check it in.
Thanks.
Tom
_
> "Andreas" == Andreas Tobler <[EMAIL PROTECTED]> writes:
Andreas> a little patchlet for the README in the examples directory.
Andreas> Ok?
Yes, thanks.
Tom
___
Classpath-patches mailing list
Classpath-patches@gnu.org
http://lists.gnu.org/mailman
> "Andreas" == Andreas Tobler <[EMAIL PROTECTED]> writes:
Andreas> 2005-07-25 Andreas Tobler <[EMAIL PROTECTED]>
Andreas> * examples/gnu/classpath/examples/swing/Demo.java: Exit when pressing
Andreas> 'Close' on the panel.
Andreas> + System.exit(1);
exit(0) would be more usual her
Hi David,
On Tue, 2005-07-26 at 15:22 +, David Gilbert wrote:
> I wasn't sure from your description exactly which test fails.
Only number 10 fails. Where you set the TimeZone of the Calendar to a
new one with the same rules.
> I can't run the tests against Classpath myself right now because
2005-07-26 Roman Kennke <[EMAIL PROTECTED]>
* javax/swing/plaf/basic/BasicButtonUI.java
(paint): Only call paintFocus if the button is actually focused.
(paintFocus): The default implementation in the BasicLookAndFeel
does nothing. I removed the code here.
/Roma
I added the border that David implemented to the MetalLookAndFeel.
2005-07-26 Roman Kennke <[EMAIL PROTECTED]>
* javax/swing/plaf/metal/MetalLookAndFeel.java
(initComponentDefaults): Added new borders for MenuBar and
MenuItem.
/Roman
Index: javax/swing/plaf/metal/MetalLookAn
Mark Wielaard wrote:
>Hi,
>
>This is a reimplementation of DateFormat.equals() following the JCL book
>description. While testing this I found that David already created Mauve
>tests for this which closely follow this description already.
>
>2005-07-26 Mark Wielaard <[EMAIL PROTECTED]>
>
>
Hi,
I was seeing a loop with an application using getPreferredSize() because
BasicToolBarUI was calling the layoutmanager directly. It should
actually just return null (as the default implementation of UIComponent
does) to get the layoutmanager to calculate the preferred size. On irc
Roman and I f
This fix 'outsources' the functionality of the getProperty methods into
a private method. This avoids infinite recursion if one of the
getProperty methods is overridden in subclasses.
2005-07-26 Roman Kennke <[EMAIL PROTECTED]>
* java/util/Properties.java
(getPropertyInternal
As suggested by Mark, I committed the following formatting fix.
2005-07-26 Roman Kennke <[EMAIL PROTECTED]>
* gnu/java/awt/AWTUtilities.java:
Fixed formatting of this file.
/Roman
Index: gnu/java/awt/AWTUtilities.java
===
I committed this patch:
2005-07-26 David Gilbert <[EMAIL PROTECTED]>
* javax/swing/plaf/metal/MetalBorders.java:
(MenuBarBorder): new inner class.
Regards,
Dave Gilbert
Index: javax/swing/plaf/metal/MetalBorders.java
===
Hi,
This is a reimplementation of DateFormat.equals() following the JCL book
description. While testing this I found that David already created Mauve
tests for this which closely follow this description already.
2005-07-26 Mark Wielaard <[EMAIL PROTECTED]>
* java/text/DateFormat.java (
I committed this patch:
2005-07-26 David Gilbert <[EMAIL PROTECTED]>
* javax/swing/plaf/metal/MetalBorders.java:
(MenuItemBorder): new inner class.
Regards,
Dave Gilbert
Index: javax/swing/plaf/metal/MetalBorders.java
==
Hi,
A close and select call (possibly used by available) can be interrupted.
In that case we have to retry the system call.
2005-07-26 Mark Wielaard <[EMAIL PROTECTED]>
* native/jni/java-nio/gnu_java_nio_channels_FileChannelImpl.c
(Java_gnu_java_nio_channels_FileChannelImpl_init): Mark
Andrew Haley wrote:
> All of this is fairly ill-specified.
As usual :-( I just re-read the relevant part of the 1.5 API spec and
JLS (which it refers to) and I think you can actually interpret it to
prohibit loading array classes via ClassLoader.loadClass (indirectly
based on the definition of "b
Hi,
This moves the sanity checking of lock() and tryLock() into a shared
private method checkLock() so that all argument and channel property
checking is done in the same way.
2005-07-26 Mark Wielaard <[EMAIL PROTECTED]>
* gnu/java/nio/channels/FileChannelImpl.java (lockCheck): New meth
Jeroen Frijters writes:
> Andrew Haley wrote:
> > Tom Tromey writes:
> > > > "Jeroen" == Jeroen Frijters <[EMAIL PROTECTED]> writes:
> > >
> > > Jeroen> * java/rmi/server/RMIClassLoader.java
> > > Jeroen> (loadClass(String, String)): Use
> > Class.forName() instead
Hi,
I had this patch on my disk for a long time. It helps with debugging
file related errors/exceptions a lot.
2005-07-26 Mark Wielaard <[EMAIL PROTECTED]>
* gnu/java/nio/channels/FileChannelImpl.java (description):
New final field.
(FileChannelImpl): Set description.
(init): L
Andrew Haley wrote:
> Tom Tromey writes:
> > > "Jeroen" == Jeroen Frijters <[EMAIL PROTECTED]> writes:
> >
> > Jeroen>* java/rmi/server/RMIClassLoader.java
> > Jeroen>(loadClass(String, String)): Use
> Class.forName() instead of
> > Jeroen>directly calling ClassLoader.loadCla
Hi Christian,
On Tue, 2005-07-26 at 11:29 +0200, Christian Thalinger wrote:
> On Tue, 2005-07-26 at 11:12 +0200, Mark Wielaard wrote:
> > That is what the autoconf manual recommends.
> > Does that work for you?
>
> Yes.
Thanks. Committed as follows:
2005-07-26 Christian Thalinger <[EMAIL PROT
On Tue, 2005-07-26 at 11:12 +0200, Mark Wielaard wrote:
> Hi,
>
> On Tue, 2005-07-26 at 10:20 +0200, Christian Thalinger wrote:
> > There is an include missing (for timeval struct).
> >
> > #ifndef TARGET_NATIVE_NETWORK_SOCKET_SET_OPTION_SO_TIMEOUT
> > + #include
> >#include
> >#incl
Hi,
I moved to native methods in Proxy to VMProxy and moved the
configuration flags that control whether the native methods are used
from gnu.classpath.Configuration to VMProxy as well.
Regards,
Jeroen
2005-07-26 Jeroen Frijters <[EMAIL PROTECTED]>
* NEWS: Added comment about new VMPr
Hi,
On Tue, 2005-07-26 at 10:20 +0200, Christian Thalinger wrote:
> There is an include missing (for timeval struct).
>
> #ifndef TARGET_NATIVE_NETWORK_SOCKET_SET_OPTION_SO_TIMEOUT
> + #include
>#include
>#include
Should we enclose this as done in TimeZone.c?
#if TIME_WITH_SYS_TIM
Tom Tromey writes:
> > "Jeroen" == Jeroen Frijters <[EMAIL PROTECTED]> writes:
>
> Jeroen> * java/rmi/server/RMIClassLoader.java
> Jeroen> (loadClass(String, String)): Use Class.forName() instead of
> Jeroen> directly calling ClassLoader.loadClass(), to handle array
> Jero
Hi!
There is an include missing (for timeval struct).
TWISTI
2005-7-26 Christian Thalinger <[EMAIL PROTECTED]>
* native/target/generic/target_generic_network.h
(TARGET_NATIVE_NETWORK_SOCKET_SET_OPTION_SO_TIMEOUT): Added
missing sys/time.h include.
Index: native/targe
I committed this patch:
2005-07-26 David Gilbert <[EMAIL PROTECTED]>
* javax/swing/text/DateFormatter.java: API doc fixes,
* javax/swing/text/InternationalFormatter.java: likewise,
* javax/swing/text/JTextComponent.java: likewise.
Regards,
Dave Gilbert
Index: javax/sw
Hi Jeroen,
On Mon, 2005-07-25 at 11:43 +0200, Jeroen Frijters wrote:
> I committed the attached patch, that allows a VM implementer to hook
> Classloader.findLoadedClass.
When adding new VM hooks could you please add a little note to the NEWS
file. That makes it much easier for VM implementers to
Hi,
On Sat, 2005-07-23 at 20:37 -0700, Casey Marshall wrote:
> Committed. I changed a few minor things, such as using Mark's
> suggestion to use 'madvise' to load the file into core.
>
> I can't get this to work with Jamvm on Darwin; I get a bizarre
> segfault when trying to create the new Ma
Mark Wielaard wrote:
> On Mon, 2005-07-25 at 11:43 +0200, Jeroen Frijters wrote:
> > I committed the attached patch, that allows a VM implementer to hook
> > Classloader.findLoadedClass.
>
> When adding new VM hooks could you please add a little note
> to the NEWS file. That makes it much easier
Hi,
On Sat, 2005-07-23 at 18:00 +0200, Sven de Marothy wrote:
> This agrees with the JDK behaviour better. And my peers expect it too.
>
> (Whoops, accidentally commited a change to Component. Did that revert
> correctly?)
No. You committed the attached.
The first hunk seems to have been changed
Hi,
This makes make clean in examples work even when the compile somehow
failed and the classes directory was still present.
2005-07-26 Mark Wielaard <[EMAIL PROTECTED]>
* examples/Makefile.am (clean-local): Use rm -rf.
Committed,
Mark
--- examples/Makefile.am25 May 2005 22:
Mark Wielaard wrote:
> On Mon, 2005-07-25 at 18:17 +0200, Jeroen Frijters wrote:
> > Is this an important scenario? I thought you long haired
> types didn't
> > like binary distributions ;-)
>
> He he. Sure we don't. But if you follow the harmony effort of the
> cleanly shaved apache types you wi
53 matches
Mail list logo
