[cp-patches] Patch: IdentityHashMap - using null for empty-marker and introducing null-value-marker

2006-04-01 Thread Martin Egholm Nielsen
Hi there, For those having followed the discussion on the gcj-list, here is a patch that replaces the use of a dedicated java.lang.Object as marker for empty entries in the map. Instead null is used to indicate these empty entries, and a new java.lang.Object is used to represent any null-valu

[cp-patches] FYI: Generating jarsigner script

2006-04-01 Thread Raif S. Naffah
hello there, the attached patch --already committed-- generates a jarsigner shell script which attempts to find a java executable before invoking it with the tool's main class. the ChangeLog entry follows: 2006-04-02 Raif S. Naffah <[EMAIL PROTECTED]> * configure.ac: Added tools/jarsi

[cp-patches] FYI: jarsigner is now GPL

2006-04-01 Thread Raif S. Naffah
hello there, following Mark's suggestion, i changed the licence of the jarsigner sources to be GPL. here is the patch --already committed-- and the ChangeLog entry: 2006-04-02 Raif S. Naffah <[EMAIL PROTECTED]> * tools/gnu/classpath/tools/jarsigner/HashUtils.java: Use GPL. *

[cp-patches] FYI: more WrappedPlainView fixes

2006-04-01 Thread Robert Schuster
Hi, this fixes another bunch of problems with WrappedPlainView. There were all kinds of oddities which I have described in PR26676[0] and these are now fixed. Included is a fix to Utilities.getTabbedTextOffset() which makes this method a bit more eager to break text. I changed a comparison from >

[cp-patches] Patch: FYI: PR 25924

2006-04-01 Thread Tom Tromey
I'm checking this in. This is the patch from PR 25924. It seems reasonable to me. Tom 2006-04-01 Bernhard Rosenkraenzer <[EMAIL PROTECTED]> PR classpath/25924: * java/awt/image/DirectColorModel.java (extractAndNormalizeSample): Handle case where alpha==0. Index: java

[cp-patches] FYI: Segment.toString() robustness fix

2006-04-01 Thread Robert Schuster
Hi, this patch makes Segment a bit more robust. Before the patch you could not create a new Segment and immediately call toString() on it as it would have thrown a NullPointerException (segment.array is null). No a check is done in toString and an empty string is returned (like in the RI). Change

[cp-patches] FYI: use SwingUtilities.computeUnion in BasicTextUI

2006-04-01 Thread Robert Schuster
Hi, this patch replaces Rectangle.union with the more efficient SwingUtilities.computeUnion in BasicTextUI. ChangeLog: 2006-04-02 Robert Schuster <[EMAIL PROTECTED]> * javax/swing/plaf/basic/BasicTextUI.java: (damageRange): Use SwingUtilities.computeUnion to avoid unnec

Re: [cp-patches] FYI: fix for up/down movement in JTextArea /w lineWrap=true - 2nd attempt

2006-04-01 Thread Robert Schuster
Please ignore that mail the 2nd changelog fix was not needed. Robert Schuster wrote: >>(WrappedLine.viewToModel): Changed '<' to '<=' in if-expression, > >^^^ > All this viewToModelToView stuff makes me dizzy. This time it should have been > modelToVie

[cp-patches] Patch: FYI: serialVersionUID fix

2006-04-01 Thread Tom Tromey
I'm checking this in. This fixes the serialVersionUID for Certificate. Japi pointed out that it was wrong. Tom 2006-04-01 Tom Tromey <[EMAIL PROTECTED]> * java/security/cert/Certificate.java (serialVersionUID): Fixed. Index: java/security/cert/Certificate.java ===

Re: [cp-patches] Patch: FYI: speed up split-for-gcj

2006-04-01 Thread Tom Tromey
> "Mark" == Mark Wielaard <[EMAIL PROTECTED]> writes: Mark> Unfortunately it seems to trigger some gcj bug with the svn trunk Mark> development version on builder (see below). I checked against 4.0.3 and Mark> 4.1 and they both seem fine. I am not 100% sure it is caused by this Mark> patch (st

Re: [cp-patches] FYI: fix for up/down movement in JTextArea /w lineWrap=true - 2nd attempt

2006-04-01 Thread Robert Schuster
> * javax/swing/text/WrappedPlainView.java: > (WrappedLine.viewToModel): Changed '<' to '<=' in if-expression, ^ ^^ Great ... a typo in the ChangeLog entry. I fixed this in CVS. The corrected one is: 2006-04-01 Robert Schuster

Re: [cp-patches] FYI: fix for up/down movement in JTextArea /w lineWrap=true - 2nd attempt

2006-04-01 Thread Robert Schuster
> (WrappedLine.viewToModel): Changed '<' to '<=' in if-expression, ^^^ All this viewToModelToView stuff makes me dizzy. This time it should have been modelToView. Corrected ChangeLog is: 2006-04-01 Robert Schuster <[EMAIL PROTECTED]> * javax/swin

Re: [cp-patches] Patch: FYI: speed up split-for-gcj

2006-04-01 Thread Mark Wielaard
Hi Tom, On Fri, 2006-03-31 at 15:30 -0700, Tom Tromey wrote: > This hugely speeds up the split-for-gcj script. On my machine it goes > from more than two minutes to about 6 seconds. Very cool. It definitely helps me a lot on a make distcheck in classpath. Unfortunately it seems to trigger some

[cp-patches] FYI: copyright headers updated

2006-04-01 Thread Robert Schuster
Hi, this patches updates some copyright headers in files I worked on recently. The ChangeLog: 2006-04-01 Robert Schuster <[EMAIL PROTECTED]> * javax/swing/text/CompositeView.java: Fixed copyright header. * javax/swing/text/BoxView.java: Fixed copyright header. * javax/s

Re: [cp-patches] RFC: gnu.regexp: a fix that makes \G usable

2006-04-01 Thread Ito Kazumitsu
From: Ito Kazumitsu <[EMAIL PROTECTED]> Date: Fri, 31 Mar 2006 02:05:56 +0900 (JST) > ChangeLog: > 2006-03-30 Ito Kazumitsu <[EMAIL PROTECTED]> New class gnu.regexp.CharIndexedCharSequence can be used as a superclass of other classes. ChangeLog: 2006-04-01 Ito Kazumitsu <[EMAIL PROTECTED]>

[cp-patches] FYI: fix for up/down movement in JTextArea /w lineWrap=true - 2nd attempt

2006-04-01 Thread Robert Schuster
Hi, this patch fixes the broken up/down movement in JTextArea /w lineWrap=true[0] (which use a WrappedPlainView) for real!. The faulty behavior was spread along many classes in javax.swing.text but now its fixed. The ChangeLog: 2006-04-01 Robert Schuster <[EMAIL PROTECTED]> * javax/sw

[cp-patches] FYI: fix for up/down movement in JTextArea /w lineWrap=true

2006-04-01 Thread Robert Schuster
Hi, this patch fixes the broken up/down movement in JTextArea /w lineWrap=true[0] (which use a WrappedPlainView). The faulty behavior was actually caused by some tiny bugs in the implementation of BoxView.getViewAtPoint and WrappedPlainView.WrappedLine.viewToModel(). The ChangeLog: 2006-04-01 Ro

[cp-patches] FYI: jikes compile fix

2006-04-01 Thread Mark Wielaard
Hi, Seems there is a bug in jikes (1.22) when compiling jar.Attributes. This fixes it by fully qualifying the implements. 2006-04-01 Mark Wielaard <[EMAIL PROTECTED]> Fixes PR26973 * java/util/jar/Attributes.java: Fully qualify java.util.Map. Committed, Mark diff -u -r1.15 Att