[Commit] feBlend & Initial enable-background

[Thomas E Deweese <th...@kodak.com>]

1) Added feBlend (we now pass the BE for blend).  I had to add
   Messages to Messages.properties, and Constants to SVGConstants
   (I'm surprised by this since all the constants I've used in the past
   have already been there, does this perhaps mean there the blend values
   are missing from somewhere in the spec?).

   This is implemented by simply adding additional rules to CompositeRule.
   The modes are all implemented in SVGComposite as well.

2) Added Preliminary support for enable-background.  It's currently
   only supported on 'g' element, and it currently doesn't support
   bounds.  I believe both of these are simply bridge issues (I have
   the code for bounds in the BackgroundRable but right now I don't
   excersise it).  I'm not certain of the best way to handle this in
   the bridge.  Also given the work going on there I wanted to tread
   as little as possible.

3) Fixed a bug in EnableBackgroundFactory (didn't handle new with no bounds).
   People many want to review my fix here.

4) All references to coerceData & copyData now use the ones in
   GraphicsUtil not GaussianBlurOp.  I've removed the methods from
   GaussianBlurOp.

CVS: ----------------------------------------------------------------------
CVS: Enter Log.  Lines beginning with `CVS:' are removed automatically
CVS:
CVS: Committing in .
CVS:
CVS: Modified Files:
CVS:    xml-batik/resources/org/apache/batik/refimpl/bridge/resources/Messages.\
properties
CVS:    xml-batik/sources/org/apache/batik/css/svg/EnableBackgroundFactory.java\
 
CVS:    xml-batik/sources/org/apache/batik/gvt/CompositeGraphicsNode.java
CVS:    xml-batik/sources/org/apache/batik/gvt/filter/CompositeRule.java
CVS:    xml-batik/sources/org/apache/batik/gvt/filter/GraphicsNodeRable.java
CVS:    xml-batik/sources/org/apache/batik/refimpl/bridge/CSSUtilities.java
CVS:    xml-batik/sources/org/apache/batik/refimpl/bridge/SVGBridgeContext.java\
 
CVS:    xml-batik/sources/org/apache/batik/refimpl/bridge/SVGGElementBridge.jav\
a
CVS:    xml-batik/sources/org/apache/batik/refimpl/gvt/ConcreteCompositeGraphic\
sNode.java
CVS:    xml-batik/sources/org/apache/batik/refimpl/gvt/filter/ConcreteGraphicsN\
odeRable.java
CVS:    xml-batik/sources/org/apache/batik/refimpl/gvt/filter/GaussianBlurOp.ja\
va
CVS:    xml-batik/sources/org/apache/batik/refimpl/gvt/filter/MorphologyOp.java\
 
CVS:    xml-batik/sources/org/apache/batik/refimpl/gvt/filter/SVGComposite.java\
 
CVS:    xml-batik/sources/org/apache/batik/util/SVGConstants.java
CVS:    xml-batik/sources/org/apache/batik/util/awt/image/GraphicsUtil.java
CVS: Added Files:
CVS:    xml-batik/sources/org/apache/batik/refimpl/bridge/SVGFeBlendElementBrid\
ge.java
CVS:    xml-batik/sources/org/apache/batik/refimpl/gvt/filter/ConcreteBackgroun\
dRable.java
CVS: ----------------------------------------------------------------------

Re: [Commit] feBlend & Initial enable-background

[Stephane Hillion <St...@sophia.inria.fr>]

Thomas E Deweese wrote:

> 2) Added Preliminary support for enable-background.  It's currently
>    only supported on 'g' element, and it currently doesn't support
>    bounds.  I believe both of these are simply bridge issues (I have
>    the code for bounds in the BackgroundRable but right now I don't
>    excersise it).  I'm not certain of the best way to handle this in
>    the bridge.  Also given the work going on there I wanted to tread
>    as little as possible.

The enable-background property can be an identifier ('accumulate') or a list ('new'
or 'new x y w h').
See http://lists.w3.org/Archives/Public/www-svg/2000Oct/0074.html for confirmation.

In fact the changes you made to the CSS engine
(batik/css/svg/EnableBackgroundFactory.java) to return an identifier when 'new' is
parsed was not the right thing to do :-(
I will go back to the previous version and fix the bridge to make it works
correctly.

You and everybody here, please ask me before changing anything in batik/dom,
batik/css or if you have a question about how to manage a CSS value.
--
    Stephane.