Combo button on right side is not displayed correctly. I tried more themes - it is the same problem for all themes I tried eg Human, Clearlooks. It happens only on Ubuntu 7.10 but not 7.04 so it might be problem with GTK libs. It happens with JDK 6u3 and JDK 7 build 22. On console I get (6u3 for JDK 7b22 just numbers are different) :
(<unknown>:8407): Gtk-WARNING **: Attempting to add a widget with type GtkButton to a GtkComboBoxEntry (need an instance of GtkEntry or of a subclass)

(<unknown>:8407): Gtk-CRITICAL **: gtk_widget_realize: assertion `GTK_WIDGET_ANCHORED (widget) || GTK_IS_INVISIBLE (widget)' failed

(<unknown>:8407): Gtk-CRITICAL **: gtk_paint_box: assertion `style->depth == gdk_drawable_get_depth (window)' failed

(<unknown>:8407): Gtk-CRITICAL **: gtk_paint_box: assertion `style->depth == gdk_drawable_get_depth (window)' failed
Posted Date : 2007-11-01 13:03:59.0

N/A

Rather then adding an arrow_button to combo_box_entry as a container directly we should
the parent of the arrow_button to the combo_box_entry.
Things work good and look good in this case.
Posted Date : 2008-04-07 11:25:25.0

I have the same problem on Ubuntu 7.10. In all GTK+ themes. I've tried all recent Sun JVM's, also the Java SE 6 Update N build 05. The same problem everywhere.

Also JDK 6u2. 

And in JDK 6u3 as well.

Also on debian sid

Didn't works correct on ubuntu gutsy

Also on OpenSUSE 10.3 JDK 6u3

also on JDK 6u3  + ubuntu 7.10

I also have the problem on 7.10 and I have tried all recent Sun JVM's. :-(

I also have the problem on 7.10 and I have tried all recent Sun JVM's. :-(

Count me as another unhappy Java/Ruby developer who is running Ubuntu 7.10 and has tried all recent Sun JVM's with no success. :-(

Another Ubuntu 7.10, sun java 1.6.0_03, with netbeans 6.0 final produces the following on install:
 (<unknown>:26158): Gtk-WARNING **: Attempting to add a widget with type GtkButton to a GtkComboBoxEntry (need an instance of GtkEntry or of a subclass)

Also on Fedora 8 JDK 6u3

same here using Ubuntu 7.10

java version "1.6.0_03"
Java(TM) SE Runtime Environment (build 1.6.0_03-b05)
Java HotSpot(TM) Client VM (build 1.6.0_03-b05, mixed mode, sharing)

Same problem here Ubuntu 7.10/Netbeans6/Java 1.6.0_03-b05

The same problem with JDK 6 Update 4. Will this bug ever get fixed???

... and not working on Fedora 8 j1.6 update 5

+1
openSUSE 10.3 
java-1_6_0-sun-1.6.0.u4-3.1.i586.rpm

Great developers, who can't fixed this bug in the last six month!

There's a new duplicate of this bug, it's ID #6657509 (JComboBox display errors in GTK L&F on Linux). Unfortunately, there's still nof bugfix :-(

This problem occur here also 

Yeah, I get the same error with the HSQLDB Swing Client. 

org.hsqldb.util.DatabaseManagerSwing

Still not fixed.
Ubuntu 7.10
Java SE 6 update 10 (build 1.6.0_10-ea-b12)

Also on Zenwalk 5.0 with gtk 2.12.2 and jdk 6 u2

It's so sad for Java at the Linux desktop that this bug can't get fixed for such a long time. This bug is the most ugly one. It appears on all recent Linux distributions and also on all recent OpenSolaris builds :-(

There are so many more GTK L&F problems left, e.g. the FileChooser-dialog looks like from the 1980's (very ugly), JTable components have very ugly table headers and so on. Nothing happens :-(

Same here, all versions of Ubuntu, as well as all versions of OpenSolaris I played with so far. If only Java provided a pleasant LaF for CDE on Solaris - so far it seems Sun is discouraging the use of their own operating systems by preventing their dev tools to seamlessly integrate with the desktop systems given... whyever.

Same here
Ubuntu 7.10
Java SE 6 update 10 (build 1.6.0_10-beta-b13)

The link below demonstrates the failure and a patch to fix it.  The
patch eliminates theme engines' ability to special-case the arrow button/text field combination.

http://www.mail-archive.com/swing-dev@openjdk.java.net/msg00090.html

Unbelievable. This patch is available such a long time and it was not included??? Too sad...

4 Month later and this isn't fixed.
Not in jdk6 "build 1.6.0_05-b13"
Not in jdk7 "build 1.7.0-ea-b24"
Patch from  kozmoz02 is not accepted by SUN.
Is Linux/Solaris platform no longer supported by SUN?
Or is providing more features in the win platform more important for SUN as fixing bugs in other platforms.  

What's happaning with this bug ??? When you gonna fixed, I have ubntu 7.10 too...

Same problem here on Ubuntu 7.10. Its ridiculous that a bug like this could exists so long, poor form Sun!

sorry, i'm using ubuntu 8.04, not 7.10 ^
java-6-sun-1.6.0.05

Getting the same with Ubuntu 7.10, Java 1.5.0_03 and 1.6.0_04 and GTK 2.12.0, 100% reproducible on NetBeans 6.0.1 and 6.1 rc1.

Quoting Kirill Kirichenko  Apr 11, 2008:
http://mail.openjdk.java.net/pipermail/swing-dev/2008-April/000148.html

"I already have an approved fix for the issue.
Wait up for little while. I just have to find out how to put it into the
repository. I have some troubles with the new version control system."

Unfortunately not available in build 1.6.0_10-beta-b22. Hopefully in the next build - I'm really looking forward to this fix. 

There is still discussion going on to determine if this fix is going to be included in JDK6. Please mail kirill.kirichenko at sun.com to ask for inclusion in the JDK6 update release...

Discussion reference:
http://mail.openjdk.java.net/pipermail/jdk6-dev/2008-April/000043.html

It's fixed in Ubuntu 8.04 Hardy Heron OpenJDK6 though.

It does seem to work in Hardy Heron, but OpenJDK gives me totally messed up fonts (kinda blueish, slurrish) and seems to default back to the Swing L&F no matter what I do. Anyone else experiencing these problems?

It's happening in the last ubuntu 8

Why is it only fixed in Java 7(b27) and not in the latest Java6 release (1.6.0_06  or 1.6.0_10 b23)???

So sad, the fix is not available in the latest Java6 release 1.6.0_10 b24  :-(
Do we need to wait for final Java7 release for this bugfix which occurs on all Linux distribution? Java success on the Linux Desktop must fail with such L&F problems (and strange file choosers, ...).

also on last hardy
down to jdk 5 fix it

also on last hardy
down to jdk 5 fix it

This is ridiculous. I do not consider this bug closed by any means, and I do not see any available fix - openJDK is completely immature and cant be seriously used. The question is why this ostensible fix has not been incorporated into current JDK release? Is it really so difficult to do that? I am really  fed up with that stupid feature, it spoils any GTK/swing GUI just greatly. 

"This is ridiculous. I do not consider this bug closed by any means, and I do not see any available fix - openJDK is completely immature and cant be seriously used. The question is why this ostensible fix has not been incorporated into current JDK release? Is it really so difficult to do that? I am really  fed up with that stupid feature, it spoils any GTK/swing GUI just greatly. "
This is exactly how any java desktop user/developer feels!! Fix it in Java6!!!! 

The fix is being ported to openjdk6 (so it will get into the major linux distributions.

Dmitri
Java2D Team

Damn it.
I thought I was doing something wrong and spend 2 days trying to fix it :(
I hope the fix get ported really quick to Debian Testing.

This bug must be urgently fixed. If Sun wants to promote Solaris and open S.O.s, this fix is crucial to desktop applications in these operational systems (how Linux).

Only fix it in the JDK 7.0 is an absurd. And OpenJDK is still immature.

I've tested it on the latest OpenJDK6. OK, it's fixed there, but the OpenJDK6 is yet not usable for Swing applications. Font handling and layout management is completely unacceptable.

PLEASE, PLEASE, PLEASE fix it in Sun's JDK 6 !!!

It's importand to fix it in jdk 6 not in some future release jdk 7
please fix it 

I cannot vote for it, does it mean that it will be fixed only in jdk 7!?... lol, I'm surprised that sun release netbeans for linux users, totally disrespect.
Is it so hard to fix it in jdk 6!???
FIX IT! wrrrr

I saw that in list  of TOP 25 recent Bugs there are lot of those which has less than 76 votes, so why there isn't this bug!

Still not fixed in the latest  1.6.0_10-rc-b26. It's so sad ....

Deliver the fix in java 6u10 !!!

Please fix this.

The bug has also not been fixed in 1.6.0_10-rc-b27 :-(
Why is it so hard to down-merge the in JDK 7 available bugfix?

So, dear Sun folks, if I can NOT trust in Java as an effective MULTI-platform technology that provides a serious competitive alternative to Windows:

 1. Please let me know: How do I transmit confidence to my customers regarding the Java software they are delivered, when these applications look so 'cute and proffesional' like this combo in gtk+?

 2. Please let me know: How do I explain my customers that Java is a solid and trustable technology when they have to put up with the 'cute and proffesional' l&f  details like this one for months until JDK7 is delivered?

 3. Please let me know: How can I continue advocating for Xnix systems + Java tech. as a quality and cost effective technological environment that really worths adopting? How can I continue believing this is yet one of my competitive advantages?

 4. Please let me know: According to the above, what's the point of having Java as a business strategy? Wouldn't it be better to deliver native software? If Java is thought of as a Windows-centric it is better to focus in Windows or to use C++ (for instance) for Xnix, but not Java in either case.

Java support for non-windows platform lacks a lot of care if Sun still wants Java to keep the reason of its existence. Under my businessman standpoint I have already started to reconsider the strategy of my company regarding Java.

I hope Sun guys reach this, but I guess this would not be the first time they hear something like this.

Kind Regards.

This problem still occur in my openSUSE11, java version "1.6.0_06"
Java(TM) SE Runtime Environment (build 1.6.0_06-b02)

Please fix this problem.

thanks.

Still not fixed in 1.6.0_10-rc-b28. It will probably never be included in Java 6 :-( Terrible ...

Product Version: NetBeans IDE 6.0.1 (Build 080320)
Java: 1.6.0_06; Java HotSpot(TM) Client VM 10.0-b22
System: Linux version 2.6.24-19-generic running on i386; UTF-8; en_US (nb)
Hardy

[though it looks as though I should get 1.6.0_10 :)

Just to add, a reinstall from Synaptic is stuck at update 6 in Ubuntu Hardy; so there's even more work to do to *deliver* a bugfix

[now 1.6.0_10-rc-b28:]
Gtk-WARNING **: Attempting to add a widget with type GtkButton to a GtkComboBoxEntry (need an instance of GtkEntry or of a subclass)

Gtk-CRITICAL **: gtk_widget_realize: assertion `GTK_WIDGET_ANCHORED (widget) || GTK_IS_INVISIBLE (widget)' failed

Gtk-CRITICAL **: gtk_paint_box: assertion `style->depth == gdk_drawable_get_depth (window)' failed

etc :)

The problem occurs when I want to open the Update Center GlassFish V2:

(<unknown>:24221): Gtk-WARNING **: Attempting to add a widget with type GtkButton to a GtkComboBoxEntry (need an instance of GtkEntry or of a subclass)

(<unknown>:24221): Gtk-CRITICAL **: gtk_widget_realize: assertion `GTK_WIDGET_ANCHORED (widget) || GTK_IS_INVISIBLE (widget)' failed

Oh yeah, 7(b31) fixes it all right, now it just crashes the whole program.  At least the program works with 6(b28) while it still has the Gtk-WARNINGs

Exception in thread "AWT-EventQueue-0" java.lang.NullPointerException
	at javax.swing.JComponent.setBorder(JComponent.java:1783)
	at javax.swing.DefaultListCellRenderer.getListCellRendererComponent(DefaultListCellRenderer.java:159)
	at javax.swing.plaf.basic.BasicComboBoxUI.getDefaultSize(BasicComboBoxUI.java:1252)
	at javax.swing.plaf.basic.BasicComboBoxUI.getDisplaySize(BasicComboBoxUI.java:1323)
	at javax.swing.plaf.basic.BasicComboBoxUI.getMinimumSize(BasicComboBoxUI.java:892)
	at javax.swing.JComponent.getMinimumSize(JComponent.java:1729)
	at javax.swing.BoxLayout.checkRequests(BoxLayout.java:482)
	at javax.swing.BoxLayout.minimumLayoutSize(BoxLayout.java:325)
	at java.awt.Container.minimumSize(Container.java:1713)
	at java.awt.Container.getMinimumSize(Container.java:1698)
	at javax.swing.JComponent.getMinimumSize(JComponent.java:1731)
	at javax.swing.BoxLayout.checkRequests(BoxLayout.java:482)
	at javax.swing.BoxLayout.layoutContainer(BoxLayout.java:424)
	at java.awt.Container.layout(Container.java:1477)
	at java.awt.Container.doLayout(Container.java:1466)
	at java.awt.Container.validateTree(Container.java:1564)
	at java.awt.Container.validateTree(Container.java:1571)
	at java.awt.Container.validateTree(Container.java:1571)
	at java.awt.Container.validateTree(Container.java:1571)
	at java.awt.Container.validateTree(Container.java:1571)
	at java.awt.Container.validate(Container.java:1536)
	at java.awt.Window.pack(Window.java:718)

The disrespect with us - Linux programmers - is evident: there is a low priority with Java to Linux. I have systems in Windows and Linux (Ubuntu and Fedora). My desktop systems in Ubuntu and Fedora are with this big problem, and my customers are very dissatisfied.

I don't  want use the metal L&F in my systems, I don't want wait months to get the JDK 7, and I don't want use the OpenJDK (now). I have problems with my Linux customers because this bug.

I am extremely, extremely disappointed with Sun and Java to Linux. And starting to analize the Phyton language. ¬¬

The problem reported by 'this_is_my_screen_name' is unrelated and is due to the LookAndFeel being initialized AFTER the JList is created.

Dear Members,
By trial and error, I have noticed that Decorated look and feel is unaffected fy the bug. So, for now we can use the fork:
if(LINUX)
JFrame.setDefaultLookAndFeelDecorated(true);
else
UIManager.setLookAndFeel(UIManager.getSystemLookAndFeelClassName());
(pseudocode)
Worked for the time
Vlad

What should be the benefit of JFrame.setDefaultLookAndFeelDecorated(true) ???

It does not help on the corrupted combo box problem on  GTK L&F !

I will move on , I guess python or tcl-tk are a better solution .
Maybe one day it will be fixed. I hope .
rg.  

I steel have the problem:

using:
> java -version
java version "1.6.0_10-rc2"
Java(TM) SE Runtime Environment (build 1.6.0_10-rc2-b32)
Java HotSpot(TM) Server VM (build 11.0-b15, mixed mode)

(openSUSE 11)

(<unknown>:8680): Gtk-WARNING **: Attempting to add a widget with type GtkButton to a GtkComboBoxEntry (need an instance of GtkEntry or of a subclass)

(<unknown>:8680): Gtk-CRITICAL **: gtk_widget_realize: assertion `GTK_WIDGET_ANCHORED (widget) || GTK_IS_INVISIBLE (widget)' failed

(<unknown>:8680): Gtk-CRITICAL **: gtk_paint_box: assertion `style->depth == gdk_drawable_get_depth (window)' failed

(<unknown>:8680): Gtk-CRITICAL **: gtk_paint_box: assertion `style->depth == gdk_drawable_get_depth (window)' failed

(<unknown>:8680): Gtk-CRITICAL **: gtk_paint_box: assertion `style->depth == gdk_drawable_get_depth (window)' failed

(<unknown>:8680): Gtk-CRITICAL **: gtk_paint_box: assertion `style->depth == gdk_drawable_get_depth (window)' failed

(<unknown>:8680): Gtk-CRITICAL **: gtk_paint_box: assertion `style->depth == gdk_drawable_get_depth (window)' failed

...

It's so sad and almost impossible. This bugfix is still missing in the new Java6 Update 10 release. Why is Sun so ignorant to Linux users?

I have the same problem in Ubuntu 8.04.  It even shows up in the Java plugin control panel.

Ubuntu 8.04, Java build 1.6.0_07-b06:  Another Linux developer here facing this bug. In any serious "platform independent" software environment bugs as visible as this one should just not take over a year to fix...

The bug still persists and affects many of our users.. Would be just great to get some feedback here why this problem has not been fixed in Java 6! Thanks.

Still the same problem, using Fedora Core 9, JRE 1.6.0_10. 

The bug celebrates it's first birthday today!!! 

java version "1.6.0_10", Fedora 8 and the bug is still there!

Fedora Core 9 , jdk "1.6.0_10, jxplorer ldap browser

[telman@localhost jxplorer]$ uname -a
Linux localhost.localdomain 2.6.26.6-79.fc9.i686 #1 SMP Fri Oct 17 14:52:14 EDT 2008 i686 athlon i386 GNU/Linux

[telman@localhost jxplorer]$ java -version
java version "1.6.0_10"
Java(TM) SE Runtime Environment (build 1.6.0_10-b33)
Java HotSpot(TM) Client VM (build 11.0-b15, mixed mode, sharing)

[telman@localhost jxplorer]$ sh jxplorer.sh -version
starting JXplorer...

Use "jxstart.sh console" if you want logging to the console
[telman@localhost jxplorer]$ sh jxplorer.sh  console
starting JXplorer...

main start
TIME: Fri Nov 14 14:14:26 EET 2008  (398)

Fri Nov 14 14:14:26 EET 2008: running JXplorer version JXplorer (build JXv3.1 2005-04-07 16:18 EST)
Fri Nov 14 14:14:26 EET 2008: running java from: /usr/java/jdk1.6.0_10/jre
Fri Nov 14 14:14:26 EET 2008: running java version 1.6.0_10
localDir = /home/telman/jxplorer/
Fri Nov 14 14:14:26 EET 2008: set configuration file to /home/telman/jxplorer/dxconfig.txt
Fri Nov 14 14:14:27 EET 2008: no search filter config file found at: /home/telman/jxplorer/search_filters.txt
can't find: /home/telman/jxplorer/htmldocs/start_en.html

(<unknown>:18297): Gtk-WARNING **: Attempting to add a widget with type GtkButton to a GtkComboBoxEntry (need an instance of GtkEntry or of a subclass)

(<unknown>:18297): Gtk-CRITICAL **: gtk_widget_realize: assertion `GTK_WIDGET_ANCHORED (widget) || GTK_IS_INVISIBLE (widget)' failed

(<unknown>:18297): Gtk-CRITICAL **: gtk_paint_box: assertion `style->depth == gdk_drawable_get_depth (window)' failed

(<unknown>:18297): Gtk-CRITICAL **: gtk_paint_box: assertion `style->depth == gdk_drawable_get_depth (window)' failed

(<unknown>:18297): Gtk-CRITICAL **: gtk_paint_box: assertion `style->depth == gdk_drawable_get_depth (window)' failed

(<unknown>:18297): Gtk-CRITICAL **: gtk_paint_box: assertion `style->depth == gdk_drawable_get_depth (window)' failed
main end
TIME: Fri Nov 14 14:14:28 EET 2008  (863)

I have alse the bug on Ubuntu 8.10 with sun-jdk 1.6.0_10 but I don't have the problem on openjdk 1.60_0-b12, nor openjdk 1.7.0-ea-b39.

This bug is in 1.6.0_11

Tested FileChooserDemo.jar in Ubuntu.

Unbelievable! This bug still occurs in JDK 1.6.0_11. 
Linux and Solaris are equally affected. It's really ridiculous...

Unbelievable - this really important bug has finally been fixed in Sun's Java SE 6 u12 Prerelease (build 1.6.0_12-ea-b02). Right in time for christmas :-)
Probably all these angry comments here had helped.

Thanks a lot Sun. Hopefully all this other nasty Linux @ Desktop problems can get fixed too!

A simple workaround is to run with Metal LnF:

-Dswing.systemlaf=javax.swing.plaf.metal.MetalLookAndFeel
-Dswing.defaultlaf=javax.swing.plaf.metal.MetalLookAndFeel

Quote: A simple workaround is to run with Metal LnF:

Yeah right... I'd rather go code Visual Basic instead.

I have the same problem.

java version "1.6.0_07"

Ubuntu 8.04

(<unknown>:6922): Gtk-CRITICAL **: gtk_paint_box: assertion `style->depth == gdk_drawable_get_depth (window)' failed

Didn't works correct on ubuntu hardy 1.6.0_07-b06

Again:

Ubuntu 8.10 intrepid,

java version "1.6.0_14-ea"
Java(TM) SE Runtime Environment (build 1.6.0_14-ea-b02)
Java HotSpot(TM) Client VM (build 14.0-b11, mixed mode, sharing),

netbeans-6.7_m2-linux.sh

Unbelievable! This bug still occurs.

Vladimir Kokovic,

vladimir.kokovic@a-asoft.com

I just updated to JDK6U12 from JDK6U11 and it fixed this bug (Linux 8.10 Intrepid).

Must say, about time.

(<unknown>:10880): Gtk-WARNING **: Attempting to add a widget with type GtkButton to a GtkComboBoxEntry (need an instance of GtkEntry or of a subclass)

(<unknown>:10880): Gtk-CRITICAL **: gtk_widget_realize: assertion `GTK_WIDGET_ANCHORED (widget) || GTK_IS_INVISIBLE (widget)' failed

(<unknown>:10880): Gtk-CRITICAL **: gtk_paint_box: assertion `style->depth == gdk_drawable_get_depth (window)' failed

(<unknown>:10880): Gtk-CRITICAL **: gtk_paint_box: assertion `style->depth == gdk_drawable_get_depth (window)' failed

On Mandriva too !!!!!