Colors and Fonts in Java Graphics

Suppose you want a Swing window to show a red outline of a square and a solid blue circle. You might expect Java to offer one built-in command that means put this shape on the screen. It does not. Instead, drawing is carried out through an object that knows how to paint and that remembers how it is currently configured. That object is called a graphics context.

Once you see the graphics context as a bundle of drawing tools plus a set of current settings, the rest of this topic becomes a matter of learning two things: which tools exist, and how the settings behave over time. The second question is the one that trips up most readers of drawing code, so this article spends most of its time on it.

Before reading any definitions, trace a short sequence of calls and keep track of one question: what color is the context holding right now? Imagine the following fragment, where g is a graphics context.

Read the diagram from left to right. The color is set once and then persists. The rectangle outline is red because RED was the current setting when drawRect ran. The setColor call in the middle changes the setting, and only operations after it use BLUE. Notice that the earlier outline is not repainted. The change affects what happens next, not what has already been drawn.

The subroutines are the things the context can do. They include drawing basic shapes such as lines, rectangles, and ovals, and they also include drawing text, because characters on the screen have to be drawn by the computer just as any other shape is. The data are the settings currently in force, such as the selected color and the selected font. A font determines the style and size of characters.

The third piece is easy to overlook: the drawing surface, which is the place where the drawing actually lands. Different graphics context objects can draw to different drawing surfaces. In this course the drawing surface is the content area of a window, which does not include the border or the title bar. When you call a drawing subroutine, you ask the context to put pixels onto that area, using whatever color and font it currently holds.

In a program the context is represented by a variable whose type is Graphics, just as a string variable has the type String. The variable is often named g, but the programmer may choose any name. Every numerical parameter passed to its subroutines is of type int, so coordinates and sizes are whole numbers.

The font works the same way as the color. The font is a setting stored in the context, so text drawn after the font is selected is affected by it. Coordinates on the drawing surface are measured in pixels from the top-left of the content area, so the point (10, 10) lies 10 pixels from the left edge and 10 pixels down from the top.

Swing offers two ways to put a shape on the screen. Filling a shape sets the color of every pixel inside it. Drawing a shape sets the color of only those pixels that lie along its border. A helpful picture is dragging a pen around the edge of the shape. The pen's properties, such as its size or whether its line is solid or dashed, belong to the graphics context and not to the shape itself.

The core subroutines form a practical starter set. The rectangle and oval subroutines come in pairs: each has a draw version and a fill version, and both take the same four arguments. The only difference is whether the pixels inside the shape are colored. Lines are different, since they can only be drawn. For rectangles and ovals, the four parameters give the top-left corner (x, y) and the size (w, h). For lines, the parameters are the two endpoints.

Before revealing the answer, try this on your own. Given the fragment below, which shapes are drawn in red and which in blue? Remember that the color is a setting stored in the context and that calls take effect in the order they are written.