Jim Graham wrote:
I'm curious if you need this behavior or not.
No, I don't specifically need this behavior. I am simply working my way
through JCK failures on OpenJDK/icedtea.
So, I guess it is reasonable to fix (though I'm still curious if anyone
but the JCK tests might rely on the behavior), but please fix it in a
way that is compatible with the SystemColor rgb mutation so that we
don't regress that bug at the same time...
Ultimately, the decision to fix/not fix is up to you guys. I am simply
following the specifications and the JCK.
I think the attached patch addresses the regression concern, but I have
not tested it (I am not even sure how to test it).
Thank you for your extensive reply,
Keith
--- openjdk/jdk/src/share/classes/java/awt/Color.java.orig 2008-04-18 12:31:17.000000000 -0700
+++ openjdk/jdk/src/share/classes/java/awt/Color.java 2008-04-18 12:31:42.000000000 -0700
@@ -251,6 +251,12 @@
*/
private ColorSpace cs = null;
+ /**
+ * The <code>PaintContext</code> for this solid color.
+ * @see #createContext
+ */
+ private transient ColorPaintContext context;
+
/*
* JDK 1.1 serialVersionUID
*/
@@ -1205,7 +1211,10 @@
Rectangle2D r2d,
AffineTransform xform,
RenderingHints hints) {
- return new ColorPaintContext(getRGB(), cm);
+ if (context == null || !context.getColorModel().equals(cm)
+ || context.getRGB() != getRGB())
+ context = new ColorPaintContext(getRGB(), cm);
+ return context;
}
/**
