Few Image class (and related) questions

[Luc The Perverse]

First - this class is not done. Right now it moves the unit right through
walls just to simulate some kind of animation. This is my first animation
of any kind that wasn't copied straight from a book. (Still pretty close.)

1 - I was trying to get the size of the image, but the page $Page states
that Image.getWidth( . . .) asks for an ImageObserver object. But I cannot
figure out what an ImageObserver is. I am concerned that I am missing the
bigger picture here (no pun intended.) Is an image not considered simply a
collection of pixels? Can someone explain to me not only how to get the
image, but why I need an ImageObserver object?

2 - I have a sneaking suspicion that the way I am loading the images is the
wrong way to do it. Other code snippets I have found use a Toolkit object
inexplicably. Is this Toolkit related to running the app as an applet - or
does it give you the "flexibility" to load from a JAR file if that is where
you files are located. I am using a MediaTracker object after I resize, but
not after I load, because AFAIK the ImageIcon class constructor is a
blocking call. In this case I don't know enough about what I'm supposed to
be doing to know what to ask.Can some advise me?

Here is are the two classes.
Discrete2DMaze and Positioner are just two classes which define a maze and
maze geometry on the screen. They are quite simple - any use of them
should be pretty obvious from context. I did not paste them in because they
are poorly structured, in need of revisions, and not directly related to my
question. My maze generation algorithm came from an old C function I wrote.

class JavaMazePanel extends javax.swing.JPanel{
Discrete2DMaze MyMaze; //abstract parent defines
Positioner DrawingAssistant;
public void paintComponent(Graphics comp) {
//to do: Make color dynamic, include "defaults" in constructor with
seperate constructor to allowing passing
if(MyMaze == null || MyMaze.X <1 || MyMaze.Y < 1) //something is wrong
return;
Graphics2D comp2D = (Graphics2D)comp;
int he = getHeight();
int wi = getWidth();
comp2D.setColor(Color.black);
comp2D.fillRect(0, 0, wi, he);
comp2D.setColor(Color.white);

DrawingAssistant=new Positioner(MyMaze.X, MyMaze.Y, wi, he, 0 ,0);
comp2D.setColor(Color.white);
int i, j;
for(i=0;i<MyMaze.X;i++){
if(MyMaze.isHorizWall(i, 0))
comp2D.drawLine(DrawingAssistant.X(i, MyMaze.Y), DrawingAssistant.Y(i,
MyMaze.Y), DrawingAssistant.X(i+1, MyMaze.Y), DrawingAssistant.Y(i+1,
MyMaze.Y));
for(j=0;j<MyMaze.Y;j++){
if(MyMaze.isVertWall(i, j))
comp2D.drawLine(DrawingAssistant.X(i, j), DrawingAssistant.Y(i, j),
DrawingAssistant.X(i, j+1), DrawingAssistant.Y(i, j+1));
if(MyMaze.isHorizWall(i, j))
comp2D.drawLine(DrawingAssistant.X(i, j), DrawingAssistant.Y(i, j),
DrawingAssistant.X(i+1, j), DrawingAssistant.Y(i+1, j));
}
}
for(j=0;j<MyMaze.Y;j++)
if(MyMaze.isVertWall(0, j))
comp2D.drawLine(DrawingAssistant.X(MyMaze.X, j), DrawingAssistant.Y(i,
j), DrawingAssistant.X(MyMaze.X, j+1), DrawingAssistant.Y(MyMaze.X,j+1));
}
}

class Walkable2DMaze extends JavaMazePanel implements Runnable{
Thread runner;
Image ScaledSprite;
Image UnscaledWalkerSprite;
Image[] UnscaledWalkerSprites;
float posX; //Maze position of walker
float posY;
//Positioner DrawingAssistant maintained and controlled by parent class
Walkable2DMaze(){
super();
UnscaledWalkerSprites = new Image[4];
for(int i=1;i<=4;i++)
UnscaledWalkerSprites[i-1] = new
ImageIcon("./img/Drone"+i+".GIF").getImage();
UnscaledWalkerSprite=UnscaledWalkerSprites[0];
if (runner == null) {
runner = new Thread(this);
runner.start();
}
}
protected void finalize() throws Throwable{
runner = null;
super.finalize();
}
public void paintComponent(Graphics comp) {
Graphics2D comp2D = (Graphics2D)comp;
super.paintComponent(comp); //draw maze background
if (UnscaledWalkerSprite != null){ //If a picture has been
specified
if(ScaledSprite == null/* ||
ScaledSprite.getWidth()!=DrawingAssistant.CellWidth*/)
ScaledSprite =
UnscaledWalkerSprite.getScaledInstance(DrawingAssistant.CellWidth ,
DrawingAssistant.CellHeight , 0);
MediaTracker track = new MediaTracker(this);
track.addImage(ScaledSprite, 0);
try{
track.waitForAll();
//to do: Make a "getCenterPoint" function in Positioner class
int PixelLoc_X = DrawingAssistant.X(posX + 0.5f, posY + 0.5f) -
DrawingAssistant.CellWidth/2;
int PixelLoc_Y = DrawingAssistant.Y(posX + 0.5f, posY + 0.5f) -
DrawingAssistant.CellHeight/2;
comp2D.drawImage(ScaledSprite, PixelLoc_X, PixelLoc_Y, this); }
catch(InterruptedException e){} //didn't work

}
}
public void run() {
int g=0;
while(runner!=null){
walk();
pause(1500);
g++;
if(g>3)
g=0;
UnscaledWalkerSprite = UnscaledWalkerSprites[g];
ScaledSprite = null;
}
}
public void walk() {
for (float i = 0; i < 10; i += .01) {
posX = i;
repaint();
pause(11);
}
}
public void pause(int time) {
try {
Thread.sleep(time);
} catch (InterruptedException e) { }
}


}

 

[Knute Johnson]

First - this class is not done. Right now it moves the unit right through
walls just to simulate some kind of animation. This is my first animation
of any kind that wasn't copied straight from a book. (Still pretty close.)

1 - I was trying to get the size of the image, but the page $Page states
that Image.getWidth( . . .) asks for an ImageObserver object. But I cannot
figure out what an ImageObserver is. I am concerned that I am missing the
bigger picture here (no pun intended.) Is an image not considered simply a
collection of pixels? Can someone explain to me not only how to get the
image, but why I need an ImageObserver object?

2 - I have a sneaking suspicion that the way I am loading the images is the
wrong way to do it. Other code snippets I have found use a Toolkit object
inexplicably. Is this Toolkit related to running the app as an applet - or
does it give you the "flexibility" to load from a JAR file if that is where
you files are located. I am using a MediaTracker object after I resize, but
not after I load, because AFAIK the ImageIcon class constructor is a
blocking call. In this case I don't know enough about what I'm supposed to
be doing to know what to ask.Can some advise me?

Here is are the two classes.
Discrete2DMaze and Positioner are just two classes which define a maze and
maze geometry on the screen. They are quite simple - any use of them
should be pretty obvious from context. I did not paste them in because they
are poorly structured, in need of revisions, and not directly related to my
question. My maze generation algorithm came from an old C function I wrote.

LTP:

Well there are some things we need to know. Is this an Applet that
needs to run on an old JRE or an application than can use modern (1.4 or
later) JRE? If you are writing an application, lose the MediaTracker,
use BufferedImages and load them with the ImageIO class. And if you are
using Swing, the EDT runs at a very high priority and is a very good way
to get good timing for your video updates. See the javax.swing.Timer class.

The ImageObserver is "An asynchronous update interface for receiving
notifications about Image information as the Image is constructed." So
it won't have all the information about your image until it is loaded
completely. There are probably a hundred classes that implement the
ImageObserver interface. So if you make method calls that take an
ImageObserver (eg. Image.getWidth() and Graphics.drawImage()) you put in
a reference to your ImageObserver, usually 'this'. Read the description
of Graphics.drawImage() for the note about drawing what is available.

Here is a simple example of using the ImageObserver interface to load an
image asynchronously. When you run it, note how many times paint() is
called before the image is completely drawn.

import java.awt.*;

public class test2 extends Canvas {
Image image;

public test2() {
image = Toolkit.getDefaultToolkit().createImage("saturn.jpg");
setSize(640,480);
}

public void paint(Graphics g) {
System.out.println("paint");
g.drawImage(image,0,0,this);
}

public static void main(String[] args) {
Frame f = new Frame();
f.add(new test2());
f.pack();
f.setVisible(true);
}
}

 

[Roedy Green]

1 - I was trying to get the size of the image, but the page $Page states
that Image.getWidth( . . .) asks for an ImageObserver object. But I cannot
figure out what an ImageObserver is.

for the big picture see http://mindprod.com/jgloss/image.html
and http://mindprod.com/jgloss/paint.html

 

[Luc The Perverse]

Oh dear. $Page was supposed to be a link to Sun's Image class page.

for the big picture see http://mindprod.com/jgloss/image.html
and http://mindprod.com/jgloss/paint.html

I have bookmarked these. I will now proceed to read them throughly.

Thanks Roedy!