How to Use Key Bindings
Trail: Creating a GUI With JFC/Swing
Lesson: Using Other Swing Features

How to Use Key Bindings

The JComponent class supports key bindings as a way of responding to individual keys typed by a user. Here are some examples of when key bindings are appropriate:

You often don't need to use key bindings directly. They're used behind the scenes by mnemonics (supported by all buttons and by tabbed panes as well as by JLabel) and accelerators (supported by menu items). You can find coverage of mnemonics and accelerators in the section Enabling Keyboard Operation.

An alternative to key bindings is using key listeners. Key listeners have their place as a low-level interface to keyboard input, but for responding to individual keys key bindings are more appropriate and tend to result in more easily maintained code. Key listeners are also difficult if the key binding is to be active when the component doesn't have focus. Some of the advantages of key bindings are they're somewhat self documenting, take the containment hierarchy into account, encourage reusable chunks of code (Action objects), and allow actions to be easily removed, customized, or shared. Also, they make it easy to change the key to which an action is bound. Another advantage of Actions is that they have an enabled state which provides an easy way to disable the action without having to track which component it is attached to.

The rest of this section gives you the details you need to use key bindings:

How Key Bindings Work

The key binding support provided by JComponent relies on the InputMap and ActionMap classes. An input map binds key strokes to action names, and an action map specifies the action corresponding to each action name. Technically, you don't need to use action names in the maps; you can use any object as the "key" into the maps. By convention, however, you use a string that names an action.

Each InputMap/ActionMap has a parent that typically comes from the UI. Any time the look and feel is changed, the parent is reset. In this way, any bindings specified by the developer are never lost on look and feel changes.

Each JComponent has one action map and three input maps. The input maps correspond to the following focus situations:

The component has the keyboard focus. The WHEN_FOCUSED input map is typically used when the component has no children. For example, buttons bind the Space key using the WHEN_FOCUSED map.
These bindings are only effective when the component has focus.
The component contains (or is) the component that has the focus. This input map is commonly used for a composite component — a component whose implementation depends on child components. For example, JTables make all their bindings using WHEN_ANCESTOR_OF_FOCUSED_COMPONENT so that if the user is editing, the up-arrow key (for example) still changes the selected cell.
The component's window either has the focus or contains the component that has the focus. This input map is commonly used for mnemonics or accelerators, which need to be active regardless of where focus is in the window.

When the user types a key, the JComponent key event processing code searches through one or more input maps to find a valid binding for the key. When it finds a binding, it looks up the corresponding action in the action map. If the action is enabled, the binding is valid and the action is executed. If it's disabled, the search for a valid binding continues.

If more than one binding exists for the key, only the first valid one found is used. Input maps are checked in this order:

  1. The focused component's WHEN_FOCUSED input map.
  2. The focused component's WHEN_ANCESTOR_OF_FOCUSED_COMPONENT input map.
  3. The WHEN_ANCESTOR_OF_FOCUSED_COMPONENT input maps of the focused component's parent, and then its parent's parent, and so on, continuing up the containment hierarchy. Note: Input maps for disabled components are skipped.
  4. The WHEN_IN_FOCUSED_WINDOW input maps of all the enabled components in the focused window are searched. Because the order of searching the components is unpredictable, avoid duplicate WHEN_IN_FOCUSED_WINDOW bindings!

Let's consider what happens in two typical key binding cases: a button reacting to the Space key, and a frame with a default button reacting to the Enter key.

In the first case, assume the user presses the Space key while a JButton has the keyboard focus. First, the button's key listeners are notified of the event. Assuming none of the key listeners consumes the event (by invoking the consume method on the KeyEvent) the button's WHEN_FOCUSED input map is consulted. A binding is found because JButton uses that input map to bind Space to an action name. The action name is looked up in the button's action map, and the actionPerformed method of the action is invoked. The KeyEvent is consumed, and processing stops.

In the second case, assume the Enter key is pressed while the focus is anywhere inside a frame that has a default button (set using the JRootPane setDefaultButton method). Whatever the focused component is, its key listeners are first notified. Assuming none of them consumes the key event the focused component's WHEN_FOCUSED input map is consulted. If it has no binding for the key or the Action bound to the key is disabled, the focused component's WHEN_ANCESTOR_OF_FOCUSED_COMPONENT input map is consulted and then (if no binding is found or the Action bound to the key is disabled) the WHEN_ANCESTOR_OF_FOCUSED_COMPONENT input maps of each of the component's ancestors in the containment hierarchy. Eventually, the root pane's WHEN_ANCESTOR_OF_FOCUSED_COMPONENT input map is searched. Since that input map has a valid binding for Enter, the action is executed, causing the default button to be clicked.

How to Make and Remove Key Bindings

Here is an example of specifying that a component should react to the F2 key:

//where anAction is a javax.swing.Action

As the preceding code shows, to get a component's action map you use the getActionMap method (inherited from JComponent). To get an input map, you can use the getInputMap(int) method, where the integer is one of the JComponent.WHEN_*FOCUSED* constants shown in the preceding list. Or, in the usual case where the constant is JComponent.WHEN_FOCUSED, you can just use getInputMap with no arguments.

To add an entry to one of the maps, use the put method. You specify a key using a KeyStroke object, which you can get using the KeyStroke.getKeyStroke(String) method. You can find examples of creating an Action (to put in an action map) in How to Use Actions.

Here's a slightly more complex example that specifies that a component should react to the Space key as if the user clicked the mouse.

component.getInputMap().put(KeyStroke.getKeyStroke("released SPACE"),
//where pressedAction and releasedAction are javax.swing.Action objects

To make a component ignore a key that it normally responds to, you can use the special action name "none". For example, the following code makes a component ignore the F2 key.



The preceding code doesn't prevent the relevant WHEN_ANCESTOR_OF_FOCUSED_COMPONENT and WHEN_IN_FOCUSED_WINDOW input maps from being searched for an F2 key binding. To prevent this search, you must use a valid action instead of "none". For example:

Action doNothing = new AbstractAction() {
    public void actionPerformed(ActionEvent e) {
        //do nothing

The Key Binding API

The following tables list the commonly used API for key bindings. Also see the API table Creating and Using an Action, in the section How to Use Actions.

Getting and Using InputMaps
Method Purpose
InputMap getInputMap()
InputMap getInputMap(int)
(in JComponent)
Get one of the input maps for the component. The arguments can be one of these JComponent constants: WHEN_FOCUSED, WHEN_IN_FOCUSED_WINDOW, or WHEN_ANCESTOR_OF_FOCUSED_COMPONENT. The no-argument method gets the WHEN_FOCUSED input map.
void put(KeyStroke, Object)
(in InputMap)
Set the action name associated with the specified key stroke. If the second argument is null, this method removes the binding for the key stroke. To make the key stroke be ignored, use "none" as the second argument.
static KeyStroke getKeyStroke(String)
(in KeyStroke)
Get the object specifying a particular user keyboard activity. Typical arguments are "alt shift X", "INSERT", and "typed a". See the KeyStroke API documentation for full details and for other forms of the getKeyStroke method.
Getting and Using ActionMaps
Method Purpose
ActionMap getActionMap()
(in JComponent)
Get the object that maps names into actions for the component.
void put(Object, Action)
(in ActionMap)
Set the action associated with the specified name. If the second argument is null, this method removes the binding for the name.

Examples that Use Key Bindings

The following table lists examples that use key bindings:

Example Where Described Notes
TableFTFEditDemo How to Use Tables The IntegerEditor class registers a key binding on a formatted text field to validate the input when the user presses the Enter key.
TextComponentDemo Text Component Features Key bindings are registered on a text pane to navigate through the text when the user presses the Control-B, Control-F, Control-P, and Control-N keys.

Previous page: How to Use the Focus Subsystem
Next page: How to Use Modality in Dialogs