Swing Hacks

Turn the Spotlight on Swing #64

Chapter 8, Rendering

|

321

HACK

paint( ) method of the component you want to see through the Java3D

scene is less efficient than taking a screen capture with the help of

java.awt.

Robot

; however, it is a lot easier because it works even when the

Canvas3D

has already been added to the window. This allows, for instance, changing

the texture when the window is resized and the gradient changes due to the

new dimensions.

Once the content panel has been fully drawn in

image, you must clip it to

retrieve the exact part covered by the

Canvas3D. This is done with another

BufferedImage

called

subImage

. This new picture has the same dimensions as

the 3D scene. The second step is to draw

image on subImage (without forget-

ting to change the origin of the drawing when you call

drawImage( )). When

the two last parameters of this method are both

0, the image is drawn on the

target surface with its top-left corner at the target’s top-left corner. With the

coordinates –

c3d.getX( ) and –c3d.getY( ), the pixel of image drawn at the

top-left corner of

subImage is the pixel where the Canvas3D is located on

screen. This ensures the code paints the exact part of the content panel that

lies behind the 3D scene.

Then, a

Background object is created with subImage as a texture. To achieve

this, you need to create an

ImageComponent2D from the BufferedImage—and

don’t forget to make the pixel formats compatible! ...