This has been lying around for a long time here. It implements the basic
inheritedChannel() infrastructure. The default implementation can't do
much more than return null (which is OK, and actually what the JDK does
for me). If a VM can do something more useful here, it would have to
provide its ow
Hi,
> This avoids a possible NPE when the byteorder system property is not
> set.
Some people argued (on IRC IIRC) that we should rather let this fail if
the property is not setup correcly. I think I agree and revert that like
attached.
2007-04-03 Roman Kennke <[EMAIL PROTECTED]>
* ja
This fixes BasicTreeUI.getPathBounds() to also consider the tree's
insets, plus it adds a bunch of null checks for safety (all of this
stuff is public API and can thus be overridden to return null, so...).
2007-04-03 Roman Kennke <[EMAIL PROTECTED]>
* javax/swing/plaf/basic/BasicTreeUI.
This improves the BasicLabelUI:
- It caches the Insets instance.
- It makes the font metrics fetching more fail-safe.
- It removes the graphics color restoring, this is already performed by
Swing's painting mechanism.
2007-04-03 Roman Kennke <[EMAIL PROTECTED]>
* javax/swing/plaf/basic/
Ok, I've got to flush some older patches. These are a small bunch of
minor fixes for the Metal L&F.
2007-04-03 Roman Kennke <[EMAIL PROTECTED]>
* javax/swing/plaf/metal/MetalButtonUI.java
(paintButtonPressed): Fill the whole button not only visibleRect.
* javax/swing/pla
There were some .cvsignores missing. Here they are.
2007-04-03 Roman Kennke <[EMAIL PROTECTED]>
* native/target/.cvsignore,
* native/target/Linux/.cvsignore,
* native/target/generic/.cvsignore: Some new ignores.
/Roman
--
http://kennke.org/blog/
Index: native/target/.
Hi Thomas,
> For now this is not enabled by default. I propose to thouroughly test
> > this and make it default when things turn out to work.
>
> This may make sense for portableNativeSync -- have you talked to the JikesRVM
> developers about this change? I don't think making it the default is
I've run through a number of tests and have not noticed any
regressions in performance or functionality, but please let me know if
there are any problems.
That didn't take long. This patch clears up a null pointer exception
introduced by the last patch.
One last one, a tiny fix for people
Keith Seitz wrote:
Kyle Galloway wrote:
ChangeLog
2007-04-03 Kyle Galloway <[EMAIL PROTECTED]>
* gnu/classpath/jdwp/value/ArrayValue.java: New file.
Questions/comments/concerns?
Looks good. Please commit.
Committed, thanks.
Kyle
Kyle Galloway wrote:
ChangeLog
2007-04-03 Kyle Galloway <[EMAIL PROTECTED]>
* gnu/classpath/jdwp/value/ArrayValue.java: New file.
Questions/comments/concerns?
Looks good. Please commit.
Keith
Hi Roman,
Roman Kennke wrote:
This patch makes the GDK lock reentrant. Basically it replaces the
portableNativeSync functionality by a simple new implementation. As of
GTK 2.4 (according to the docs), there is a function
gdk_threads_set_lock_functions() which allows to define a pair of
functions
Sometimes, it helps to attach a patch :-).
Kyle Galloway wrote:
JDWP needs to distinguish between and Object and an Array to properly
get and set values, so this patch adds the ArrayValue type, which is
used for this purpose. It is very similar to ObjectValue, but with a
different tag value.
JDWP needs to distinguish between and Object and an Array to properly
get and set values, so this patch adds the ArrayValue type, which is
used for this purpose. It is very similar to ObjectValue, but with a
different tag value.
ChangeLog
2007-04-03 Kyle Galloway <[EMAIL PROTECTED]>
* g
This patch makes the GDK lock reentrant. Basically it replaces the
portableNativeSync functionality by a simple new implementation. As of
GTK 2.4 (according to the docs), there is a function
gdk_threads_set_lock_functions() which allows to define a pair of
functions for locking and unlocking. I imp
14 matches
Mail list logo
