[cp-patches] FYI: BoxView fixes

This fixes some problems with BoxView, makes the Intel testsuite happy and removes one debug output statement that I left in there. 2006-08-10 Roman Kennke <[EMAIL PROTECTED]> * javax/swing/text/BoxView.java (calculateMajorAxisRequirements): Sum up the preferred and ma

[cp-patches] FYI: Bidi support for AbstractDocument

This implements full bidi support for AbstractDocument. This means that there is now an additional element structure that reflects the direction of bidi runs in the document. 'Full' in the sense that the Intel testcases for AbstractDocument and subclasses all pass now. And since the testsuite i

[cp-patches] FYI: GapContent optimization/fix

This optimizes GapContent.getChars() to only copy the buffer when really really necessary. It also makes this method respect the Segment.isPartialReturn() property for further optimization. This makes a couple of tests in Intel's testsuite happy and should be good for performance. 2006-08-10

[cp-patches] Patch: GtkDragSourceContextPeer fixlet

Small fix 2006-08-10 Lillian Angel <[EMAIL PROTECTED]> * gnu/java/awt/dnd/peer/gtk/GtkDragSourceContextPeer.java (getComponentPeer): Added check to prevent NPE. Index: gnu/java/awt/dnd/peer/gtk/GtkDragSourceContextPeer.java ==

[cp-patches] FYI: AccessControlContext speedup

Hi all, This commit avoids calling AccessController.getContext() twice when creating an AccessControlContext with a security manager enabled. I've been using BC-compiled Tomcat on gcj to benchmark the security stuff, and this commit improves gcj's performance from 2200 to 2400 requests per second.