I have a project that takes time to load everything so I create a splash screen that tells the user through a progressbar how much time it will take to fully load and show the UI, but I'm facing a problem.
When I create my splash, this shows up correctly but then I create and initialize the Principal frame and everything freeze until this has fully load.
So, I try to load my Principal frame in a thread using SwingWorker (and it works) but after unknown NullPointerExceptions and reading a lot I found that this is a terrible idea because I am not creating my UI in the EDT, so here I am stuck.
I know that I must do Swing Calls in the Event Dispatch Thread (EDT) and non-swing heavy work in SwingWorkers but initialize the Swing Components of my Principal Frame are a heavy work too so, what should I do?
I have read some question here, specially this, and I think I get it but I have doubts. Taking that example:
SwingUtilities.invokeAndWait(new Runnable() {
public void run() {
new SplashScreen();
}
});
// Code to start system (nothing that touches the GUI)
SwingUtilities.invokeAndWait(new Runnable() {
public void run() {
new MainFrame();
}
});
//.. etc
And reading this site that says:
The Swing framework manages component drawing, updates, and event handlers on the EDT.
Is creating a new component a Swing Call? If it is, What should I do if new MainFrame() will take some time because the project has a lot of components to initialize?
How do I tell the Splash something like "Program loaded 50%"?
What does a Swing Call means and how can I do a correct use of invokeLater and SwingWorker? Maybe the solution is too obvious or have already an answer, but I can't see it and I apologize if this is the case.
Thanks!
You're on a right track. But don't use invokeAndWait (if you have to only) - use invokeLater:
invokeAndWait
Causes doRun.run() to be executed synchronously on the AWT event dispatching thread.
invokeLater
Causes doRun.run() to be executed asynchronously on the AWT event dispatching thread.
Consider that block wrapped doLater is run on EDT thread and code wrapped in doOutside is invoked in another thread (and that's why you don't block the UI):
EDIT:
As pointed out in the comments I add the explanations for the concepts I'll use.
doLater {
// here goes the code
}
is a concept for:
SwingUtilities.invokeLater(new Runnable() {
public void run() {
// here goes the code
}
});
And
doOutside {
// here goes the code
}
is a concept for:
new Thread(new Runnable() {
#Override
public void run() {
// here goes the code
}
}).start();
doLater {
final MainFrame m = new MainFrame();
doOutside {
// handle heavy operation
final int result = 1;
doLater {
m.setResult(result);
}
}
}
Conclusion: everything that touches Swing in some way must be run on EDT.
If you want to update percentages:
doLater {
final MainFrame m = new MainFrame();
doOutside {
// handle progress
for(int i = 0; i < someSize; ++i) {
final int progress = i;
doLater {
m.getProgressBar().setProgress(progress);
}
}
}
}
I hope you understand the concept now. The SwingWorker just do exectly something as doOutside === doInBackground & doLater === done/progress
Btw. The code above is a real code: lookup Griffon framework in Groovy.
Related
Java is not my mother tongue and I've been fighting with this problem for a little while.
Basically, I am finding a behavioural difference between calling method switchApplets() directly from init(), and calling it from within a new thread spawned by init().
The consequence of calling it from inside the new thread is that the new applet whitescreens -- until/unless the user resizes or minimizes their browser. If called at the end of init(), the new UI renders immediately without any input from the user. But that's not an option because it doesn't wait for the thread to finish its prep work.
Trimmed-down code:
public class PreLoader extends Applet implements AppletStub {
static JProgressBar pBar = null;
static JLabel message;
public void switchApplets() {
try {
Class main_class = Class.forName("MainClass");
Applet main_applet = (Applet)main_class.newInstance();
removeAll();
setSize(0,0);
setLayout(new GridLayout(1,0));
add(main_applet);
main_applet.init();
main_applet.start();
main_applet.setStub(this);
}
catch (Exception e) {
}
}
public void init() {
pBar = new JProgressBar(0, 100);
pBar.setValue(0);
pBar.setStringPainted(true);
message = new JLabel("Beginning work!");
add(message);
add(pBar);
FlowLayout flow = new FlowLayout();
setLayout(flow);
Thread t = new Thread ( new Runnable () {
public void run ()
{
longRunningFunction1();
longRunningFunction2();
message.setText("Work complete! Stand by..");
switchApplets(); //does NOT work as intended from here
return;
}
} );
t.start();
//switchApplets(); //works as intended if called HERE
}
public void longRunningFunction1() {
//perform some tasks, advance progress bar
}
public void longRunningFunction2() {
//perform some tasks, advance progress bar
}
public void start() {
return;
}
public void appletResize(int width, int height) {
return;
}
}
I tried making init() wait for the thread to finish so that I could call switchApplets() from there, but that only blocked the EDT and prevented the UI from updating. Also tried playing with SwingUtilities' invokeLater/invokeAndWait, but even though switchApplets() gets run on the EDT, it seems that it MUST be called directly from init() (or at least the thread init is running on) to have the desired effect.
Why does calling switchApplets() from within a new thread result in a slightly different (and unwanted) UI behaviour?
The consequence of calling it from inside the new thread is that the new applet whitescreens -- until/unless the user resizes or minimizes their browser.
It's likely a deadlock caused by trying to do UI code on the wrong thread.
I tried making init() wait for the thread to finish so that I could call switchApplets() from there, but that only blocked the EDT and prevented the UI from updating.
You're on the right track. You need to call switchApplets() only from the EDT, and only after the work is done on the other thread.
Are you sure you tried using invokeLater() or invokeAndWait() from within the spawned thread after the long running functions were done? It's been a long while since I did applets but I'm not aware of any applet-specific reason why it wouldn't work, and it would work in any other case. I.e.,
public void run()
{
longRunningFunction1();
longRunningFunction2();
SwingUtilities.invokeLater(new Runnable() {
public void run() {
message.setText("Work complete! Stand by..");
switchApplets();
}
});
}
However, the most proper way to do this is with a SwingWorker rather than a manually created thread. SwingWorker (which is not nearly as well-known as it should be) is designed exactly for the goal of performing background tasks on a separate thread while still being able to update the GUI with progress updates and the results. E.g.,
new SwingWorker<Void,Void>() {
#Override
protected Void doInBackground() { // is called on a background thread
longRunningFunction1();
longRunningFunction2();
return null;
}
#Override
protected void done() { // is called on the Swing thread
message.setText("Work complete! Stand by..");
switchApplets();
}
}.execute();
The Void stuff is because SwingWorker is also capable of returning results and sending intermediate progress updates, but this example doesn't use those features.
You indicated that your long running functions are also updating a progress bar. That's another thing that should happen only on the Swing thread. In practice you can often get away without it, but it's dodgy. Your progress updates can use one of the SwingUtilities.invoke methods, or the mechanisms of SwingWorker; either should work. (SwingWorker itself provides two different ways to do it: Call addPropertyChangeListener (Swing thread) and setProgress (background thread), or call publish (background thread) and override process (Swing thread).)
Also, a small suggestion: if it's inconvenient to deal with a checked exception (or impossible to usefully do so), rather than catching and ignoring it, you should at least catch & rethrow it as an unchecked exception:
catch (Exception e) {
throw new RuntimeException(e);
}
That way, the stacktrace and error message of any exception will not be lost.
I'm making a chess program for a project. I'm trying to add a move history box to the side of the board. The move history works fine, and the data is properly sent to the text area, but the text inside the JTextArea disappears while the AI is thinking about his move.
public void aiMove(){
if (!playing){ return; }
paintImmediately(0,0,totalX,totalY);
ai = eve.getMove(chess,wtm,aiOut); //text disappears here
chess.makeMove(ai);
wtm = !wtm;
humanMove = true;
writeMove(ai); //updates move history, text reappears here
playing = stillPlaying();
repaint();
}
private void writeMove(Move move){
char c = "abcdefgh".charAt(7-move.fromY);
char h ="abcdefgh".charAt(7-move.toY);
String s = Character.toString(c)+(move.fromX+1)+" - "+Character.toString(h)+(move.toX+1)+" ";
if (!wtm){
String q = chess.getFullMove()+". "+s+" ";
moves.setText(moves.getText()+q);
}
else {
moves.setText(moves.getText()+s+"\n");
}
}
Here's a print screen of what's happening.
http://s13.postimage.org/mh7hltfk7/JText_Area_disappear.png
SOLVED
Thanks to all replies. I changed aiMove() so it creates a thread. Here is what I did.
Attempt #3... swing is still so foreign to me. I didn't want to change writeMove to getMove or I would have to rewrite the human's turn slightly. Since the project is essentially done, I am trying to avoid as much work as possible :)
The GUI is entirely optional anyways, I was just doing it for fun, and to try and learn a bit of swing.
public void aiMove(){
if (!playing){ return; }
if (!aiThread.isAlive()){
aiThread = new Thread(){
public void run(){
ai = eve.getMove(chess,wtm,aiOut);
chess.makeMove(ai);
wtm = !wtm;
humanMove = true;
SwingUtilities.invokeLater(new Runnable(){
public void run(){
writeMove(ai);
}
});
repaint();
playing = stillPlaying();
}
};
aiThread.start();
}
}
It also fixed a problem I had before, in that if I were to hold down the 'a' key (force ai move), it would queue up many forced ai moves. Now that doesn't happen.
The problem is your AI thinking is CPU intensive/time consuming, thus it is considered a long running task. You should not do long running tasks on GUI Event Dispatch Thread as this will cause the UI to seem frozen and thus only show updates after the task has finished.
Fortunately there are 2 different approaches you could use:
Use a Swing Worker which as the tutorial states:
The SwingWorker subclass can define a method, done, which is
automatically invoked on the event dispatch thread when the background
task is finished.
SwingWorker implements java.util.concurrent.Future.
This interface allows the background task to provide a return value to
the other thread. Other methods in this interface allow cancellation
of the background task and discovering whether the background task has
finished or been cancelled.
The background task can provide
intermediate results by invoking SwingWorker.publish, causing
SwingWorker.process to be invoked from the event dispatch thread.
The background task can define bound properties. Changes to these
properties trigger events, causing event-handling methods to be
invoked on the event dispatch thread.
Alternatively create separate Thread for AI thinking and wrap setText call in SwingUtilities.invokeLater(...);
Thread t=new Thread(new Runnable() {
#Override
public void run() {
}
});
t.start();
UPDATE
After reading MadProgrammers comment (+1 to it) please remember to create/manipulate your GUI/Swing components on EDT via the SwingUtilities.invokeLater(..) block. You can read more on it here.
UPDATE 2:
That edit is defeating the point, the only call on EDT in SwingUtilitites block should be the setText or atleast only code that manipulates a Swing component i.e
public void aiMove(){
if (!playing){ return; }
if (!aiThread.isAlive()){ //originally initialized by constructor
aiThread = new Thread(){
public void run(){
ai = eve.getMove(chess,wtm,aiOut);
chess.makeMove(ai);
wtm = !wtm;
humanMove = true;
SwingUtilities.invokeLater(new Runnable(){
public void run(){
writeMove(ai);
}
});
repaint();
playing = stillPlaying();
}
};
aiThread.start();
}
}
I have been doing some research on this but I am still VERY confused to say the least.
Can anyone give me a concrete example of when to use Task and when to use Platform.runLater(Runnable);? What exactly is the difference? Is there a golden rule to when to use any of these?
Also correct me if I'm wrong but aren't these two "Objects" a way of creating another thread inside the main thread in a GUI (used for updating the GUI)?
Use Platform.runLater(...) for quick and simple operations and Task for complex and big operations .
Use case for Platform.runLater(...)
Use case for Task: Task Example in Ensemble App
Example: Why Can't we use Platform.runLater(...) for long calculations (Taken from below reference).
Problem: Background thread which just counts from 0 to 1 million and update progress bar in UI.
Code using Platform.runLater(...):
final ProgressBar bar = new ProgressBar();
new Thread(new Runnable() {
#Override public void run() {
for (int i = 1; i <= 1000000; i++) {
final int counter = i;
Platform.runLater(new Runnable() {
#Override public void run() {
bar.setProgress(counter / 1000000.0);
}
});
}
}).start();
This is a hideous hunk of code, a crime against nature (and
programming in general). First, you’ll lose brain cells just looking
at this double nesting of Runnables. Second, it is going to swamp the
event queue with little Runnables — a million of them in fact.
Clearly, we needed some API to make it easier to write background
workers which then communicate back with the UI.
Code using Task :
Task task = new Task<Void>() {
#Override public Void call() {
static final int max = 1000000;
for (int i = 1; i <= max; i++) {
updateProgress(i, max);
}
return null;
}
};
ProgressBar bar = new ProgressBar();
bar.progressProperty().bind(task.progressProperty());
new Thread(task).start();
it suffers from none of the flaws exhibited in the previous code
Reference :
Worker Threading in JavaFX 2.0
Platform.runLater: If you need to update a GUI component from a non-GUI thread, you can use that to put your update in a queue and it will be handled by the GUI thread as soon as possible.
Task implements the Worker interface which is used when you need to run a long task outside the GUI thread (to avoid freezing your application) but still need to interact with the GUI at some stage.
If you are familiar with Swing, the former is equivalent to SwingUtilities.invokeLater and the latter to the concept of SwingWorker.
The javadoc of Task gives many examples which should clarify how they can be used. You can also refer to the tutorial on concurrency.
It can now be changed to lambda version
#Override
public void actionPerformed(ActionEvent e) {
Platform.runLater(() -> {
try {
//an event with a button maybe
System.out.println("button is clicked");
} catch (IOException | COSVisitorException ex) {
Exceptions.printStackTrace(ex);
}
});
}
One reason to use an explicite Platform.runLater() could be that you bound a property in the ui to a service (result) property. So if you update the bound service property, you have to do this via runLater():
In UI thread also known as the JavaFX Application thread:
...
listView.itemsProperty().bind(myListService.resultProperty());
...
in Service implementation (background worker):
...
Platform.runLater(() -> result.add("Element " + finalI));
...
I have the following
public static void main(String[] args) {
boolean running = true;
boolean foo= false;
while(running)
{
doSomeTask(); // might set foo true
if(foo) {
//This call waits/blocks until gui is done working.
fireUpSwingGui(); //does work...
foo=false;
godModeReleaseGUIandALLResourcesOnlyWantMainThreadLeft();
}
}
}
Hope that godModeReleaseGUIandALLResourcesOnlyWantMainThreadLeft() says it all.
Bear in mind that we might fire up the gui again at a later stage when foo becomes true again somewhere inside doSomeTask().
Take a look at AWT Threading Issues which explains the criteria for an AWT application to exit. The part you want to focus on is the following:
Therefore, a stand-alone AWT application that wishes to exit cleanly
without calling System.exit must:
Make sure that all AWT or Swing components are made undisplayable when the application finishes. This can be done by calling Window.dispose on all top-level Windows. See Frame.getFrames.
Make sure that no method of AWT event listeners registered by the application with any AWT or Swing component can run into an infinite loop or hang indefinitely. For example, an AWT listener method triggered by some AWT event can post a new AWT event of the same type to the EventQueue. The argument is that methods of AWT event listeners are typically executed on helper threads.
A quick sample app to demonstrate...
import java.awt.Frame;
import javax.swing.JOptionPane;
import javax.swing.SwingUtilities;
public class CloseAWT
{
private static boolean running = true;
private static int response = -1;
public static void main(String[] args)
{
boolean showSwing = true;
boolean checkFrames = true;
while (running)
{
if (showSwing)
{
SwingUtilities.invokeLater(new Runnable()
{
public void run()
{
response = JOptionPane.showConfirmDialog(null, "Hello World?");
}
});
showSwing = false;
}
else
{
if (response >= 0 && checkFrames)
{
SwingUtilities.invokeLater(new Runnable()
{
public void run()
{
// topFrame.dispose();
Frame[] frames = Frame.getFrames();
System.out.printf("frames.length=%d\n", frames.length);
}
});
checkFrames = false;
}
}
}
}
}
To confirm the behavior was as expected, I ran this in JProfiler. After clicking 'yes' to dismiss the confirmation dialog, the 'AWT-EventQueue-0' thread was marked as dead. The only threads alive after this were the 'main' and the thread which listens for Ctrl-Break.
I highly recommend using something like JProfiler, YourKit, JProbe or one of the free profilers to make sure you've properly released all the components and removed all the listeners.
One final thought... You might want to consider spawning your GUI as a separate process and using some sort of IPC to pass information between your daemon process and GUI. Although this incurs the additional overhead of an extra process and the IPC, it would give you a greater assurance that your GUI is completely cleaned up when it is no longer needed.
Assuming you're opening JFrame isntances, why don't you just store them in a collection and inside godModeReleaseGUIandALLResourcesOnlyWantMainThreadLeft() you iterate over them ans call setVisible(false);
I'm not sure whether you actually can stop the AWT event queue that drives the gui.
I have some code which takes a few minutes to process, it has to connect to the web for each string in a long array, each string is a url. I want to make it so that everytime it connects, it should refresh the jtextarea so that the user is not staring into a blank page that looks frozen for 20 min. or however long it takes. here is an example of something i tried and didnt work:
try {
ArrayList<String> myLinks = LinkParser.getmyLinksArray(jTextArea1.getText());
for (String s : myLinks) {
jTextArea2.append(LinkChecker.checkFileStatus(s) + "\n");
}
} catch (IOException ex) {
JOptionPane.showMessageDialog(jTextArea1, "Parsing Error", "Parsing Error", JOptionPane.ERROR_MESSAGE);
Logger.getLogger(MYView.class.getName()).log(Level.SEVERE, null, ex);
}
The problem is that you need to perform the computation asynchronously. You should create a background thread that performs the computation, and then use SwingUtilities.invokeLater to update the JTextArea.
final ArrayList<String> myLinks = //...
(new Thread()
{
public void run(){
for (String s : myLinks) {
try{
final String result = LinkChecker.checkFileStatus(s) + "\n";
SwingUtilities.invokeLater(new Runnable(){
public void run(){
jtextArea2.append(result);
}
});
}catch(IOException error){
// handle error
}
}
}
}).start();
Edit
It has been pointed out that JTextArea's append function actually is thread safe (unlike most Swing functions). Therefore, for this particular, case it is not necessary to update it via invokeLater. However, you should still do you processing in a background thread so as to allow the GUI to update, so the code is:
final ArrayList<String> myLinks = //...
(new Thread()
{
public void run(){
for (String s : myLinks) {
try{
jtextArea2.append(LinkChecker.checkFileStatus(s) + "\n");
}catch(IOException error){
// handle error
}
}
}
}).start();
However, for pretty much any other operation that modifies a Swing object, you will need to use invokeLater (to ensure the modification occurs in the GUI thread), since almost all the Swing functions aren't thread safe.
You need to investigate threading and its relationship to GUI updates in Swing. Anything that affects or makes use of GUI components in Swing must done on a special thread called the Event Dispatch Thread (EDT).
If your code snippet, if it's freezing the GUI, I imagine that it is being run in the EDT. Performing a long-running action on the EDT will make the GUI unresponsive, because no further updates can be done while your long-running process is using the thread.
There is a helper class called SwingWorker that allows you to offload long-running computations to a background thread, and then make updates to the GUI thread when it is complete. The SwingWorker looks after the context switches between the GUI thread and the background thread. You can also display progress bars to let the user know the state of the long-running process, so they know your application hasn't hung.
swing/awt is a single threaded library, so once a component is shown, just changing it's appearance won't work correctly. You need to change the component on the GUI Thread, not from your thread. To do this wrap any code that updates a component with SwingUtilities.invokeLater... as in
SwingUtilities.invokeLater(new Runnable()
{
public void run()
{
jTextArea2.append(LinkChecker.checkFileStatus(s) + "\n");
}
});
also you want to limit what you do on the gui thread to avoid the gui from becoming sluggish, so if checkFileStatus is time consuming, execute it outside the run method and store the result in a final local variable, and just access the variable in the run() code.