[cp-patches] Patch: FYI: getopt refactoring

2006-09-21 Thread Tom Tromey
I'm checking this in. This is a getopt refactoring I've wanted to do for a while now. It: * Adds support for "joined" short options. These are one-character options whose argument is part of the same string, like "-Ifoo". I updated javah's -I option to be joined and I removed some special

Re: [cp-patches] FYI: Epoll and DatagramSocket bugs

2006-09-21 Thread Casey Marshall
David Daney wrote: > Casey Marshall wrote: >> -channel.send(buf, new InetSocketAddress(remote, port)); >> +while (true) >> + { >> +try >> + { >> +channel.send(buf, new InetSocketAddress(remote, port)); >> + } >> +

Re: [cp-patches] FYI: Epoll and DatagramSocket bugs

2006-09-21 Thread David Daney
Casey Marshall wrote: Index: gnu/java/net/PlainDatagramSocketImpl.java === RCS file: /cvsroot/classpath/classpath/gnu/java/net/PlainDatagramSocketImpl.java,v retrieving revision 1.11 diff -u -r1.11 PlainDatagramSocketImpl.java ---

[cp-patches] FYI: Epoll and DatagramSocket bugs

2006-09-21 Thread Casey Marshall
Paul Jenner helpfully pointed out some problems when running Azureus with the epoll selector, and the modified DatagramSocket implementation. 2006-09-21 Casey Marshall <[EMAIL PROTECTED]> * gnu/java/net/PlainDatagramSocketImpl.java (send): ignore `InterruptedIOException;' try ag

[cp-patches] FYI: More Glyph Vector

2006-09-21 Thread Francis Kung
Hey, This patch (committed) continues fixing FreetypeGlyphVector. The previous implementation treated glyph transformations as relative to the origin, when instead they should be relative to the glyph's individual position (and thus independent of the position). This also makes it much easier to

Re: [cp-patches] FYI: Colour Model fixes

2006-09-21 Thread Francis Kung
Ooops, just realised I removed a line of useful code while cleaning up debug messages. This patch fixes it. Francis 2006-09-21 Francis Kung <[EMAIL PROTECTED]> * java/awt/image/ColorModel.java (coerceData): Set alpha premultiplied flag. On Thu, 2006-09-21 at 17:49 -0400, Francis Kun

[cp-patches] FYI: Colour Model fixes

2006-09-21 Thread Francis Kung
Hi, This fixes the coerceData method in java.awt.image.ColorModel and its relevant subclasses. Cheers, Francis 2006-09-21 Francis Kung <[EMAIL PROTECTED]> * java/awt/image/ColorModel.java (coerceData): Return proper colour model. * java/awt/image/ComponentColorModel.java (coe

Re: [cp-patches] RFC: GConfNativePeer fix

2006-09-21 Thread Thomas Fitzsimmons
Hi, Tania Bento wrote: Hey, This patch fixes the GConf-Critical Bug of the Yahoo! Finance Stock Screener application (Bugzilla Bug #203674). [...] dir = JCL_jstring_to_cstring (env, node); - if (dir == NULL) + if (dir != NULL) This seems wrong; this will prevent valid directories fro

[cp-patches] RFC: GConfNativePeer fix

2006-09-21 Thread Tania Bento
Hey, This patch fixes the GConf-Critical Bug of the Yahoo! Finance Stock Screener application (Bugzilla Bug #203674). Could someone kindly comment on/approve this patch? It's a minor change, but it's a native file. So, I didn't really feel comfortable just committing it. Thanks, Tania 2006-

[cp-patches] FYI: BasicInternalFrameTitlePane.java - flip icon when maximising

2006-09-21 Thread David Gilbert
This patch (committed) flips the icon on the maximise/minimise button for internal frames: 2006-09-21 David Gilbert <[EMAIL PROTECTED]> * javax/swing/plaf/basic/BasicInternalFrameTitlePane.java (MaximizeAction.actionPerformed): Change icon on maxButton. Regards, Dave Index:

[cp-patches] FYI: JScrollPane related fixes

2006-09-21 Thread Roman Kennke
My last patch obviously made JScrollPane unusable. This was caused by a small typo. While debugging this I found a couple more buglets, which I don't want to hold back. 2006-09-21 Roman Kennke <[EMAIL PROTECTED]> * javax/swing/plaf/basic/BasicScrollPaneUI.java (syncScrollPaneWit

Re: [cp-patches] FYI: AWT font handling fixes

2006-09-21 Thread Roman Kennke
Hi Thomas, > > Some more stuff that I had trouble with Espial/Espresso. This should fix > > the font handling in AWT. > > [...] > > >/** > > * Called to ensure that the layout for this component is valid. This is > > * usually called on containers. > > * > > * @see #invalidat