[ Previous Section | Next Section | Chapter Index | Main Index ]

An Example
Where is main()?

Section 6.2

Graphics and Painting

Everything you see on a computer screen has to be drawn there, even the text. The Java API includes a range of classes and methods that are devoted to drawing. In this section, I'll look at some of the most basic of these. Some of this material was already covered in preliminary form in Section 3.9.

The physical structure of a GUI is built of components. The term component refers to a visual element in a GUI, including buttons, menus, text-input boxes, scroll bars, check boxes, and so on. In Java, GUI components are represented by objects belonging to subclasses of the class java.awt.Component. Most components in the Swing GUI toolkit -- although not top-level components like JFrame -- belong to subclasses of the class javax.swing.JComponent, which is itself a subclass of java.awt.Component. Every component is responsible for drawing itself. If you want to use a standard component, you only have to add it to your program. You don't have to worry about painting it on the screen. That will happen automatically, since it already knows how to draw itself.

Sometimes, however, you do want to draw on a component. You will have to do this whenever you want to display something that is not included among the standard, pre-defined component classes. When you want to do this, you have to define your own component class and provide a method in that class for drawing the component. I will always use a subclass of JPanel when I need a drawing surface of this kind, as I did for the HelloWorldDisplay class in the example HelloWorldGUI2.java in the previous section. A JPanel, like any JComponent, draws its content in the method

public void paintComponent(Graphics g)

To create a drawing surface, you should define a subclass of JPanel and provide a custom paintComponent() method. Create an object belonging to this class and use it in your program. When the time comes for your component to be drawn on the screen, the system will call its paintComponent() to do the drawing. That is, the code that you put into the paintComponent() method will be executed whenever the panel needs to be drawn on the screen; by writing this method, you determine the picture that will be displayed in the panel. Note that you are not likely to call a paintComponent() method any more than you are likely to call a main() routine. The system calls the method. You write the method to say what will happen when the system calls it.

Note that the paintComponent() method has a parameter of type Graphics. The Graphics object will be provided by the system when it calls your method. You need this object to do the actual drawing. To do any drawing at all in Java, you need a graphics context. A graphics context is an object belonging to the class java.awt.Graphics. Instance methods are provided in this class for drawing shapes, text, and images. Any given Graphics object can draw to only one location. In this chapter, that location will always be a GUI component belonging to some subclass of JPanel. The Graphics class is an abstract class, which means that it is impossible to create a graphics context directly, with a constructor. There are actually two ways to get a graphics context for drawing on a component: First of all, of course, when the paintComponent() method of a component is called by the system, the parameter to that method is a graphics context for drawing on the component. Second, every component has an instance method called getGraphics(). This method is a function that returns a graphics context that can be used for drawing on the component outside its paintComponent() method. The official line is that you should not do this, and I will almost always avoid it. But I have found it convenient to use getGraphics() in a few examples. (Note that if g is a graphics context created with getGraphics(), it is good form to call g.dispose() when finished using it. This releases any operating system resources that might be held by g.)

The paintComponent() method in the JPanel class simply fills the panel with the panel's background color. When defining a subclass of JPanel for use as a drawing surface, you will usually want to fill the panel with the background color before drawing other content onto the panel (although it is not necessary to do this if the drawing commands in the method cover the background of the component completely). This is traditionally done with a call to super.paintComponent(g), so most paintComponent() methods that you write will have the form:

public void paintComponent(g) {
   . . . // Draw the content of the component.

In general, a component should do all drawing operations in its paintComponent() method. What happens if, in the middle of some other method, you realize that the content of the component needs to be changed? You should not call paintComponent() directly to make the change. Instead, you have to inform the system that the component needs to be redrawn, and let the system do its job by calling paintComponent(). You do this by calling the component's repaint() method. The method

public void repaint();

is defined in the Component class, and so can be used with any component. You should call repaint() to inform the system that the component needs to be redrawn. It is important to understand that the repaint() method returns immediately, without doing any painting itself. The system will call the component's paintComponent() method later, as soon as it gets a chance to do so, after processing other pending events if there are any. It is even possible that many calls to repaint() will all be handled by one call to paintComponent(), if the calls to repaint() occur in a very short timespan.

Note that the system can also call paintComponent() for other reasons. It is called when the component first appears on the screen. It will also be called if the size of the component changes, which can happen when the user resizes the window that contains the component. This means that paintComponent() should be capable of redrawing the content of the component on demand. As you will see, however, some of our early examples will not be able to do this correctly.

This means that, to work properly, the paintComponent() method must be smart enough to correctly redraw the component at any time. To make this possible, a program should store data in its instance variables about the state of the component. These variables should contain all the information necessary to redraw the component completely. The paintComponent() method should use the data in these variables to decide what to draw. When the program wants to change the content of the component, it should not simply draw the new content. It should change the values of the relevant variables and call repaint(). When the system calls paintComponent(), that method will use the new values of the variables and will draw the component with the desired modifications. This might seem a roundabout way of doing things. Why not just draw the modifications directly? There are at least two reasons. First of all, it really does turn out to be easier to get things right if all drawing is done in one method. Second, even if you could directly draw the modifications, you would still have to save enough information about the modifications to enable paintComponent() to redraw the component correctly on demand.

You will see how all this works in practice as we work through examples in the rest of this chapter. For now, we will spend the rest of this section looking at how to get some actual drawing done.

6.2.1  Coordinates

The screen of a computer is a grid of little squares called pixels. The color of each pixel can be set individually, and drawing on the screen just means setting the colors of individual pixels.

pixel coordinate system

A graphics context draws in a rectangle made up of pixels. A position in the rectangle is specified by a pair of integer coordinates, (x,y). The upper left corner has coordinates (0,0). The x coordinate increases from left to right, and the y coordinate increases from top to bottom. The illustration shows a 20-pixel by 12-pixel component (with very large pixels). A small line, rectangle, and oval are shown as they would be drawn by coloring individual pixels. (Note that, properly speaking, the coordinates don't belong to the pixels but to the grid lines between them.)

For any component, you can find out the size of the rectangle that it occupies by calling the instance methods getWidth() and getHeight(), which return the number of pixels in the horizontal and vertical directions, respectively. In general, it's not a good idea to assume that you know the size of a component, since the size is often set by a layout manager and can even change if the component is in a window and that window is resized by the user. This means that it's good form to check the size of a component before doing any drawing on that component. For example, you can use a paintComponent() method that looks like:

public void paintComponent(Graphics g) {
   int width =  getWidth();   // Find out the width of this component.
   int height = getHeight();  // Find out its height.
   . . .   // Draw the content of the component.

Of course, your drawing commands will have to take the size into account. That is, they will have to use (x,y) coordinates that are calculated based on the actual height and width of the component. (However, if you are sure that you know the size, using constants for the width and height can make the drawing easier.)

6.2.2  Colors

You will probably want to use some color when you draw. Java is designed to work with the RGB color system. An RGB color is specified by three numbers that give the level of red, green, and blue, respectively, in the color. A color in Java is an object of the class, java.awt.Color. You can construct a new color by specifying its red, blue, and green components. For example,

Color myColor = new Color(r,g,b);

There are two constructors that you can call in this way. In the one that I almost always use, r, g, and b are integers in the range 0 to 255. In the other, they are numbers of type float in the range 0.0F to 1.0F. (Recall that a literal of type float is written with an "F" to distinguish it from a double number.) Often, you can avoid constructing new colors altogether, since the Color class defines several named constants representing common colors: Color.WHITE, Color.BLACK, Color.RED, Color.GREEN, Color.BLUE, Color.CYAN, Color.MAGENTA, Color.YELLOW, Color.PINK, Color.ORANGE, Color.LIGHT_GRAY, Color.GRAY, and Color.DARK_GRAY. (There are older, alternative names for these constants that use lower case rather than upper case constants, such as Color.red instead of Color.RED, but the upper case versions are preferred because they follow the convention that constant names should be upper case.)

An alternative to RGB is the HSB color system. In the HSB system, a color is specified by three numbers called the hue, the saturation, and the brightness. The hue is the basic color, ranging from red through orange through all the other colors of the rainbow. The brightness is pretty much what it sounds like. A fully saturated color is a pure color tone. Decreasing the saturation is like mixing white or gray paint into the pure color. In Java, the hue, saturation and brightness are always specified by values of type float in the range from 0.0F to 1.0F. The Color class has a static member function named getHSBColor for creating HSB colors. To create the color with HSB values given by h, s, and b, you can say:

Color myColor = Color.getHSBColor(h,s,b);

For example, to make a color with a random hue that is as bright and as saturated as possible, you could use:

Color randomColor = Color.getHSBColor( (float)Math.random(), 1.0F, 1.0F );

The type cast is necessary because the value returned by Math.random() is of type double, and Color.getHSBColor() requires values of type float. (By the way, you might ask why RGB colors are created using a constructor while HSB colors are created using a static member function. The problem is that we would need two different constructors, both of them with three parameters of type float. Unfortunately, this is impossible. You can have two constructors only if the number of parameters or the parameter types differ.)

The RGB system and the HSB system are just different ways of describing the same set of colors. It is possible to translate between one system and the other. The best way to understand the color systems is to experiment with them. (The sample program SimpleColorChooser.java lets you do that. You won't understand the source code at this time, but you can run it to play with color selection or to find the RGB or HSB values for the color that want.)

One of the properties of a Graphics object is the current drawing color, which is used for all drawing of shapes and text. If g is a graphics context, you can change the current drawing color for g using the method g.setColor(c), where c is a Color. For example, if you want to draw in green, you would just say g.setColor(Color.GREEN) before doing the drawing. The graphics context continues to use the color until you explicitly change it with another setColor() command. If you want to know what the current drawing color is, you can call the function g.getColor(), which returns an object of type Color. This can be useful if you want to change to another drawing color temporarily and then restore the previous drawing color.

Every component has an associated foreground color and background color. Generally, the component is filled with the background color before anything else is drawn (although some components are "transparent," meaning that the background color is ignored). When a new graphics context is created for a component, the current drawing color is set to the foreground color. Note that the foreground color and background color are properties of the component, not of a graphics context.

The foreground and background colors of a component can be set by calling instance methods component.setForeground(color) and component.setBackground(color), which are defined in the Component class and therefore are available for use with any component. This can be useful even for standard components, if you want them to use colors that are different from the defaults.

6.2.3  Fonts

A font represents a particular size and style of text. The same character will appear different in different fonts. In Java, a font is characterized by a font name, a style, and a size. The available font names are system dependent, but you can always use the following four strings as font names: "Serif", "SansSerif", "Monospaced", and "Dialog". (A "serif" is a little decoration on a character, such as a short horizontal line at the bottom of the letter i. "SansSerif" means "without serifs." "Monospaced" means that all the characters in the font have the same width. The "Dialog" font is the one that is typically used in dialog boxes.)

The style of a font is specified using named constants that are defined in the Font class. You can specify the style as one of the four values:

The size of a font is an integer. Size typically ranges from about 9 to 36, although larger sizes can also be used. The size of a font is usually about equal to the height of the largest characters in the font, in pixels, but this is not an exact rule. The size of the default font is 12.

Java uses the class named java.awt.Font for representing fonts. You can construct a new font by specifying its font name, style, and size in a constructor:

Font plainFont = new Font("Serif", Font.PLAIN, 12);
Font bigBoldFont = new Font("SansSerif", Font.BOLD, 24);

Every graphics context has a current font, which is used for drawing text. You can change the current font with the setFont() method. For example, if g is a graphics context and bigBoldFont is a font, then the command g.setFont(bigBoldFont) will set the current font of g to bigBoldFont. The new font will be used for any text that is drawn after the setFont() command is given. You can find out the current font of g by calling the method g.getFont(), which returns an object of type Font.

Every component also has an associated font. It can be set with the instance method component.setFont(font), which is defined in the Component class. When a graphics context is created for drawing on a component, the graphic context's current font is set equal to the font of the component.

6.2.4  Shapes

The Graphics class includes a large number of instance methods for drawing various shapes, such as lines, rectangles, and ovals. The shapes are specified using the (x,y) coordinate system described above. They are drawn in the current drawing color of the graphics context. The current drawing color is set to the foreground color of the component when the graphics context is created, but it can be changed at any time using the setColor() method.

Some drawing methods were already listed in Subsection 3.9.1. Here, I describe those methods in more detail and add a few more. With all these commands, any drawing that is done outside the boundaries of the component is ignored. Note that all these methods are in the Graphics class, so they all must be called through an object of type Graphics. It is shown here as g, but of course the name of the graphics context is up to the programmer.

6.2.5  Graphics2D

All drawing in Java is done through an object of type Graphics. The Graphics class provides basic commands for such things as drawing shapes and text and for selecting a drawing color. These commands are adequate in many cases, but they fall far short of what's needed in a serious computer graphics program. Java has another class, Graphics2D, that provides a larger set of drawing operations. Graphics2D is a sub-class of Graphics, so all the methods from the Graphics class are also available in a Graphics2D.

The paintComponent() method of a JComponent gives you a graphics context of type Graphics that you can use for drawing on the component. In fact, the graphics context actually belongs to the sub-class Graphics2D, and can be type-cast to gain access to the advanced Graphics2D drawing methods:

public void paintComponent(Graphics g) {
   Graphics2D g2;
   g2 = (Graphics2D)g;
    . // Draw on the component using g2.

I mention Graphics2D here for completeness. I will cover some important aspects of Graphics2D in Section 13.2, but a full treatment is more than we will have time for in this book. However, there are two simple applications that I would like to start using now, without explaining how they work. If g2 is a variable of type Graphics2D, as in the paintComponent() method above, then the intimidating-looking command

g2.setRenderingHint( RenderingHints.KEY_ANTIALIASING, 
                                RenderingHints.VALUE_ANTIALIAS_ON );

turns on antialiasing in the graphics context. Aliasing is a jagged appearance that can be seen when shapes are drawn using pixels. Antialiasing tries to reduce the jaggedness. It can make diagonal lines and the outlines of ovals look much nicer. It can also improve the appearance of text. Another useful command is

g2.setStroke( new BasicStroke(lineWidth) );

where lineWidth is an integer or a float. This command can be used to draw thicker lines. Lines drawn after the command will be lineWidth pixels wide. This also affects the thickness of the outlined shapes drawn by methods such as g.drawRect and g.drawOval().

6.2.6  An Example

Let's use some of the material covered in this section to write a subclass of JPanel for use as a drawing surface. All the drawing will be done in the paintComponent() method of the panel class. The panel will draw multiple copies of a message on a black background. Each copy of the message is in a random color. Five different fonts are used, with different sizes and styles. The message can be specified in the constructor; if the default constructor is used, the message is the string "Java!". The panel works OK no matter what its size.

There is one problem with the way this class works. When the panel's paintComponent() method is called, it chooses random colors, fonts, and locations for the messages. The information about which colors, fonts, and locations are used is not stored anywhere. The next time paintComponent() is called, it will make different random choices and will draw a different picture. If you resize a window containing the panel, the picture will be continually redrawn as the size of the window is changed! To avoid that, you would store enough information about the picture in instance variables to enable the paintComponent() method to draw the same picture each time it is called.

The source code for the panel class is shown below. I use an instance variable called message to hold the message that the panel will display. There are five instance variables of type Font that represent different sizes and styles of text. These variables are initialized in the constructor and are used in the paintComponent() method.

The paintComponent() method for the panel simply draws 25 copies of the message. For each copy, it chooses one of the five fonts at random, and it uses g.setFont() to select that font for drawing. It creates a random HSB color and uses g.setColor() to select that color for drawing. It then chooses random (x,y) coordinates for the location of the message. The x coordinate gives the horizontal position of the left end of the string. The formula used for the x coordinate is "-50 + (int)(Math.random() * (width+40))". This gives a random integer in the range from -50 to width-10. This makes it possible for the string to extend beyond the left edge or the right edge of the panel. Similarly, the formula for y allows the string to extend beyond the top and bottom.

Here is the complete source code for the RandomStringsPanel:

import java.awt.*;
import javax.swing.JPanel;

 * This panel displays 25 copies of a message.  The color and 
 * position of each message is selected at random.  The font
 * of each message is randomly chosen from among five possible
 * fonts.  The messages are displayed on a black background.
 * Note:  The style of drawing used here is poor, because every
 * time the paintComponent() method is called, new random values are
 * used.  This means that a different picture will be drawn each time.  
public class RandomStringsPanel extends JPanel {

    private String message;  // The message to be displayed.  This can be set in
                             // the constructor.  If no value is provided in the
                             // constructor, then the string "Java!" is used.

    private Font font1, font2, font3, font4, font5;  // The five fonts.

     * Default constructor creates a panel that displays the message "Java!".
    public RandomStringsPanel() {
        this(null);  // Call the other constructor, with parameter null.

     * Constructor creates a panel to display 25 copies of a specified message.
     * @param messageString The message to be displayed.  If this is null,
     * then the default message "Java!" is displayed.
    public RandomStringsPanel(String messageString) {

        message = messageString;
        if (message == null)
            message = "Java!";

        font1 = new Font("Serif", Font.BOLD, 14);
        font2 = new Font("SansSerif", Font.BOLD + Font.ITALIC, 24);
        font3 = new Font("Monospaced", Font.PLAIN, 30);
        font4 = new Font("Dialog", Font.PLAIN, 36);
        font5 = new Font("Serif", Font.ITALIC, 48);



     * The paintComponent method is responsible for drawing the content of the panel.
     * It draws 25 copies of the message string, using a random color, font, and
     * position for each string.
    public void paintComponent(Graphics g) {

        super.paintComponent(g);  // Call the paintComponent method from the 
                                  // superclass, JPanel.  This simply fills the 
                                  // entire panel with the background color, black.
        Graphics2D g2 = (Graphics2D)g;  // (To make the text smoother.)
        g2.setRenderingHint( RenderingHints.KEY_ANTIALIASING, 
                                RenderingHints.VALUE_ANTIALIAS_ON );

        int width = getWidth();
        int height = getHeight();

        for (int i = 0; i < 25; i++) {

            // Draw one string.  First, set the font to be one of the five
            // available fonts, at random.  

            int fontNum = (int)(5*Math.random()) + 1;
            switch (fontNum) {
            case 1:
            case 2:
            case 3:
            case 4:
            case 5:
            } // end switch

            // Set the color to a bright, saturated color, with random hue.

            float hue = (float)Math.random();
            g.setColor( Color.getHSBColor(hue, 1.0F, 1.0F) );

            // Select the position of the string, at random.

            int x,y;
            x = -50 + (int)(Math.random()*(width+40));
            y = (int)(Math.random()*(height+20));

            // Draw the message.


        } // end for

    } // end paintComponent()

}  // end class RandomStringsPanel

6.2.7  Where is main()?

The source code for the RandomStringsPanel class can be found in the example file RandomStringsPanel.java. You can compile that file, but you won't be able to run the compiled class. The problem is that the class doesn't have a main() routine. Only a class that has a main() routine can be run as a program.

Another problem is that a JPanel is not something that can stand on its own. It has to be placed into a container such as another panel or a window. In general, to make a complete program, we need a main() routine that will create a window of type JFrame. It can then create a panel and place the panel in the window. Here is a class with a main() routine that does this:

import javax.swing.JFrame;

public class RandomStrings {
    public static void main(String[] args) {
        JFrame window = new JFrame("Java!");
        RandomStringsPanel content = new RandomStringsPanel();


This class is defined by the file RandomStrings.java. You can compile and run the program, as long as the RandomStringsPanel class is also available.

The main routine is not logically a part of the panel class. It is just one way of using a panel. However, it's possible to include main() as part of the panel class, even if it doesn't logically belong there. This makes it possible to run the panel class as a program, and it has the advantage of keeping everything in one file. For an example, you can look at RandomStringsPanelWithMain.java, which is identical to the original class except for the addition of a main() routine. Although it might not be great style, I will usually take a similar approach in future examples.

I am not going to discuss the details of using JFrame here, but you can look ahead and find them in Subsection 6.7.3. You won't completely understand my main() routines until you read that section.

[ Previous Section | Next Section | Chapter Index | Main Index ]