We Keep Coding

Java program with multiple threads not working

So, I'm having a problem with a Gui i'm designing for a java app that renames all the files in a given directory to junk (Just for fun). This is the main block of code behind it all:
import java.io.File;
import java.io.IOException;
import java.io.PrintWriter;
import java.util.ArrayList;
import java.util.HashMap;
import java.util.Scanner;
import org.json.simple.JSONObject;
import org.json.simple.parser.JSONParser;
import org.json.simple.parser.ParseException;
/**
* Class for renaming files to garbage names.
* All methods are static, hence private constructor.
* #author The Shadow Hacker
*/
public class RenameFiles {
private static int renamedFiles = 0;
private static int renamedFolders = 0;
public static char theChar = '#';
public static ArrayList<File> fileWhitelist = new ArrayList<>();
public static HashMap<File, File> revert = new HashMap<>();
public static int getRenamedFiles() {
return renamedFiles;
}
public static int getRenamedFolders() {
return renamedFolders;
}
/**
* All methods are static, hence private constructor.
*/
private RenameFiles() {
// Private constructor, nothing to do.
}
/**
* #param file The file to rename.
* #param renameTo The current value of the name to rename it to.
* #return A new value for renameTo.
*/
private static String renameFile(File file, String renameTo) {
for (File whitelistedFile : fileWhitelist) {
if (whitelistedFile.getAbsolutePath().equals(file.getAbsolutePath())) {
return renameTo;
}
}
if (new File(file.getParentFile().getAbsolutePath() + "/" + renameTo).exists()) {
renameTo += theChar;
renameFile(file, renameTo);
} else {
revert.put(new File(file.getParent() + "/" + renameTo), file);
file.renameTo(new File(file.getParent() + "/" + renameTo));
if (new File(file.getParent() + "/" + renameTo).isDirectory()) {
renamedFolders++;
} else {
renamedFiles++;
}
}
return renameTo;
}
/**
* TODO Add exception handling.
* #param dir The root directory.
* #throws NullPointerException if it can't open the dir
*/
public static void renameAllFiles(File dir) {
String hashtags = Character.toString(theChar);
for (File file : dir.listFiles()) {
if (file.isDirectory()) {
renameAllFiles(file);
hashtags = renameFile(file, hashtags);
} else {
hashtags = renameFile(file, hashtags);
}
}
}
public static void renameAllFiles(String dir) {
renameAllFiles(new File(dir));
}
/**
* This uses the revert HashMap to change the files back to their orignal names,
* if the user decides he didn't want to change the names of the files later.
* #param dir The directory in which to search.
*/
public static void revert(File dir) {
for (File file : dir.listFiles()) {
if (file.isDirectory()) {
revert(file);
}
revert.forEach((name, renameTo) -> {
if (file.getName().equals(name.getName())) {
file.renameTo(renameTo);
}
});
}
}
public static void revert(String dir) {
revert(new File(dir));
}
/**
* Saves the revert configs to a JSON file; can't use obj.writeJSONString(out)
* because a File's toString() method just calls getName(), and we want full
* paths.
* #param whereToSave The file to save the config to.
* #throws IOException
*/
#SuppressWarnings("unchecked")
public static void saveRevertConfigs(String whereToSave) throws IOException {
PrintWriter out = new PrintWriter(whereToSave);
JSONObject obj = new JSONObject();
revert.forEach((k, v) -> {
obj.put(k.getAbsolutePath(), v.getAbsolutePath());
});
out.write(obj.toJSONString());
out.close();
}
/**
* Warning - clears revert.
* Can't use obj.putAll(revert) because that puts the strings
* into revert, and we want Files.
* TODO Add exception handling.
* #param whereToLoad The path to the file to load.
* #throws ParseException If the file can't be read.
*/
#SuppressWarnings("unchecked")
public static void loadRevertConfigs(String whereToLoad) throws ParseException {
revert.clear();
((JSONObject) new JSONParser().parse(whereToLoad)).forEach((k, v) -> {
revert.put(new File((String) k), new File((String) v));
});
}
/**
* This static block is here because the program uses forEach
* loops, and we don't want the methods that call them to
* return errors.
*/
static {
if (!(System.getProperty("java.version").startsWith("1.8") || System.getProperty("java.version").startsWith("1.9"))) {
System.err.println("Must use java version 1.8 or above.");
System.exit(1);
}
}
/**
* Even though I made a gui for this, it still has a complete command-line interface
* because Reasons.
* #param argv[0] The folder to rename files in; defaults to the current directory.
* #throws IOException
*/
public static void main(String[] argv) throws IOException {
Scanner scanner = new Scanner(System.in);
String accept;
if (argv.length == 0) {
System.out.print("Are you sure you want to proceed? This could potentially damage your system! (y/n) : ");
accept = scanner.nextLine();
scanner.close();
if (!(accept.equalsIgnoreCase("y") || accept.equalsIgnoreCase("yes"))) {
System.exit(1);
}
renameAllFiles(System.getProperty("user.dir"));
} else if (argv.length == 1 && new File(argv[0]).exists()) {
System.out.print("Are you sure you want to proceed? This could potentially damage your system! (y/n) : ");
accept = scanner.nextLine();
scanner.close();
if (!(accept.equalsIgnoreCase("y") || accept.equalsIgnoreCase("yes"))) {
System.exit(1);
}
renameAllFiles(argv[0]);
} else {
System.out.println("Usage: renameAllFiles [\033[3mpath\033[0m]");
scanner.close();
System.exit(1);
}
System.out.println("Renamed " + (renamedFiles != 0 ? renamedFiles : "no") + " file" + (renamedFiles == 1 ? "" : "s")
+ " and " + (renamedFolders != 0 ? renamedFolders : "no") + " folder" + (renamedFolders == 1 ? "." : "s."));
}
}
As you can see, all of it's methods are static. Now here is my (Only partially completed) event handler class:
import java.io.File;
/**
* Seperate class for the gui event handlers.
* Mostly just calls methods from RenameFiles.
* Like RenameFiles, all methods are static.
* #author The Shadow Hacker
*/
public class EventHandlers {
private static Thread t;
/**
* The reason this is in a new thread is so we can check
* if it is done or not (For the 'cancel' option).
* #param dir The root directory used by RenameFiles.renameAllFiles.
*/
public static void start(File dir) {
t = new Thread(() -> {
RenameFiles.renameAllFiles(dir);
});
t.start();
}
/**
* #param dir The root directory used by RenameFiles.revert(dir).
* #throws InterruptedException
*/
public static void cancel(File dir) throws InterruptedException {
new Thread(() -> {
while (t.isAlive()) {
// Nothing to do; simply waiting for t to end.
}
RenameFiles.revert(dir);
}).start();
}
public static void main(String[] args) throws InterruptedException {
start(new File("rename"));
cancel(new File("rename"));
}
}
The problem I'm having is that when I run revert from the RenameFiles class it works fine, but while running it from the multithreaded (We don't want the handlers to have to wait for the method to finish before reacting to another button press) EventHandlers class, revert dosn't work. Does this have something to do with RenameFiles being a class with all static methods, or something else? Please help!
Edit: #Douglas, when I run:
import java.io.File;
import java.util.concurrent.CountDownLatch;
import java.util.concurrent.ExecutorService;
import java.util.concurrent.Executors;
/**
* Seperate class for the gui event handlers.
* Mostly just calls methods from RenameFiles.
* Like RenameFiles, all methods are static.
* #author The Shadow Hacker
*/
public class EventHandlers {
private static ExecutorService service = Executors.newSingleThreadExecutor();
private static volatile CountDownLatch latch;
/**
* The reason this is in a new thread is so we can check
* if it is done or not (For the 'cancel' option).
* #param dir The root directory used by RenameFiles.renameAllFiles.
*/
public static void start(File dir) {
latch = new CountDownLatch(1);
service.submit(() -> {
RenameFiles.renameAllFiles(dir);
latch.countDown();
});
}
/**
* #param dir The root directory used by RenameFiles.revert(dir).
* #throws InterruptedException
*/
public static void cancel(File dir) throws InterruptedException {
service.submit(() -> {
try {
latch.await();
} catch (InterruptedException e) {
e.printStackTrace();
}
RenameFiles.revert(dir);
});
}
The program just runs forever, without terminating.

You have two major problems here.
First, you are sharing variables between threads. Default variable handling in Java has no guarantee that two threads will agree on what value any given variable has. You can fix this one by giving each variable the volatile modifier (note: this can decrease performance, which is why it's not default).
Second, you have no mechanism in place to guarantee anything about thread execution order. As written, it is entirely possible for EventHandlers.main to run cancel to completion before the renameAllFiles call even starts. It is also possible for the renaming to start, get paused by the thread scheduler, cancel run from beginning to end, and then renaming finish, or any of a bunch of other combinations. You attempted to do something about this with the t.isAlive() check, but your redundant creation of yet another Thread in main means there's no guarantee t is even initialized before the main thread gets there. It would be an unlikely but valid by the spec possibility for you to get a NullPointerException from that line.
This second problem is a much harder one to fix in general, and is the primary reason working with threads is infamously difficult. Fortunately this particular problem is a fairly simple case. Instead of looping forever on the isAlive() check, create a CountDownLatch when you start the thread, count it down when the thread finishes, and simply await() it in cancel. This will incidentally also solve the first problem at the same time without any need for volatile, because in addition to its scheduling coordination a CountDownLatch guarantees that any thread that awaited on it will see the results of everything done in any thread that counted it down.
So, long story short, steps to fix this:
Remove the new Thread in main and just call start directly. start creates a Thread itself, there's no need to nest that inside another Thread.
Replace the Thread t with a CountDownLatch.
In start, initialize the CountDownLatch with a count of 1.
In start, after initializing the CountDownLatch, get an ExecutorService by calling Executors.newSingleThreadExecutor(), and then submit the renameAllFiles call to it. Do this instead of using a Thread directly. Among other things, the specification guarantees that anything done before that will be visible as expected in the new thread, and I don't see any such guarantee in the documentation of Thread.start(). It's also got a lot more convenience and utility methods.
Inside what you submit to the ExecutorService, after the renaming, call countDown() on the latch.
After the submit, call shutdown() on the ExecutorService. This will prevent you from reusing the same one, but stops it from waiting indefinitely for reuse that will never happen.
In cancel, replace the while loop with a call to await() on the latch. In addition to the memory consistency guarantee, this will improve performance by letting the system thread scheduler handle the wait instead of spending CPU time on looping.
Additional changes will be needed if you want to account for multiple rename operations in the same run of the program.

Signaling between threads in Delphi from Java Code

package org.helioviewer.viewmodel.view.jp2view.concurrency;
/**
* Very simple way of signaling between threads. Has no sense of ownership and
* thus any thread can signal or wait for a signal. In general it is not a
* problem if many different threads call the signal method, but only one thread
* should be calling the waitForSignal method, since there is no way to tell
* which thread will be woken up.
*
* #author caplins
*
*/
public class BooleanSignal {
/** Signal flag */
private volatile boolean isSignaled;
/**
* Default constructor. Assigns the initial value of the isSignaled flag.
*
* #param _intitialVal
*/
public BooleanSignal(boolean _intitialVal) {
isSignaled = _intitialVal;
}
/**
* Used to wait for a signal. Waits until the flag is set, then it resets
* the flag and returns. The waiting thread can be interrupted and that
* exception is thrown immediately.
*
* #throws InterruptedException
*/
public synchronized void waitForSignal() throws InterruptedException {
while (!isSignaled)
this.wait();
isSignaled = false;
}
public synchronized void waitForSignal(long timeout) throws InterruptedException {
while (!isSignaled) {
this.wait(timeout);
isSignaled = true;
}
isSignaled = false;
}
/**
* Sets the isSignaled flag and wakes up one waiting thread. Doesn't bother
* to notifyAll since the first thread woken up resets the flag anyway.
*/
public synchronized void signal() {
isSignaled = true;
this.notify();
}
/**
* Returns the signal state.
*
* #return Current signal state
*/
public synchronized boolean isSignaled() /* throws InterruptedException */{
/*
* if(Thread.interrupted()) throw new InterruptedException();
*/
return isSignaled;
}
};
I would like to signaling between threads using Delphi. I have a sample from Java. How can I convert this code to Delphi or how can I signaling between threads.
I couldn't call thread in another thread directly. So I don't know how to do this in Delphi?

Generic array creation for java.lang.Thread

Intellij IDEA is complaining about Generic array creation
public abstract class BaseImageLoader<CacheItem>
{
private ImageLoaderThread[] workerThreads;
public BaseImageLoader(Context context)
{
...
workerThreads = new ImageLoaderThread[DEFAULT_CACHE_THREAD_POOL_SIZE];//Generic array creation error
...
}
}
ImageLoaderThread is in fact a subclass of java.lang.Thread, its not generic
I dont get what im doing wrong
this works fine:
Thread[] threads = new Thread[DEFAULT_CACHE_THREAD_POOL_SIZE];
ImageLoaderThread class
private class ImageLoaderThread extends Thread
{
/**
* The queue of requests to service.
*/
private final BlockingQueue<ImageData> mQueue;
/**
* Used for telling us to die.
*/
private volatile boolean mQuit = false;
/**
* Creates a new cache thread. You must call {#link #start()}
* in order to begin processing.
*
* #param queue Queue of incoming requests for triage
*/
public ImageLoaderThread(BlockingQueue<ImageData> queue)
{
mQueue = queue;
}
/**
* Forces this thread to quit immediately. If any requests are still in
* the queue, they are not guaranteed to be processed.
*/
public void quit()
{
mQuit = true;
interrupt();
}
#Override
public void run()
{
android.os.Process.setThreadPriority(Process.THREAD_PRIORITY_BACKGROUND);
ImageData data;
while (true)
{
try
{
// Take a request from the queue.
data = mQueue.take();
}
catch (InterruptedException e)
{
// We may have been interrupted because it was time to quit.
if (mQuit)
{
return;
}
continue;
}
...
//other unrelated stuff
}
}
}

How to detect when main thread freeze GUI in java?

I want to detect when some time consumption operations in main thread cause gui freeze.
My target is to set and unset wait cursor automatically.
thanks

I think you're putting the cart before the horse: Your main thread shouldn't do any time-consuming operations in the first place - they should always be externalized in separate threads, so that your GUI can stay responsive (and e.g. show status on the operations, or provide the possibility to abort them).

I think this could be helpful: http://www.javaspecialists.eu/archive/Issue075.html and http://www.javaworld.com/javaworld/javatips/jw-javatip87.html.

You can have a thread which polls the GUI thread's stack trace to determine whether it is idle or busy. If busy too often, you can log what it is doing (the stack trace) to a log. Initially it might be interesting to record every non-idle stack trace and work out which ones are not worth logging.

This EDT lockup detection code will do the job by adding watch dogs.
EventQueueWithWD.java:
import java.awt.*;
import java.awt.event.*;
import java.util.*;
/**
* Alternative events dispatching queue. The benefit over the
* default Event Dispatch queue is that you can add as many
* watchdog timers as you need and they will trigger arbitrary
* actions when processing of single event will take longer than
* one timer period.
* <p/>
* Timers can be of two types:
* <ul>
* <li><b>Repetitive</b> - action can be triggered multiple times
* for the same "lengthy" event dispatching.
* </li>
* <li><b>Non-repetitive</b> - action can be triggered only once
* per event dispatching.</li>
* </ul>
* <p/>
* The queue records time of the event dispatching start. This
* time is used by the timers to check if dispatching takes
* longer than their periods. If so the timers trigger associated
* actions.
* <p/>
* In order to use this queue application should call
* <code>install()</code> method. This method will create,
* initialize and register the alternative queue as appropriate.
* It also will return the instance of the queue for further
* interactions. Here's an example of how it can be done:
* <p/>
* <pre>
* <p/>
* EventQueueWithWD queue = EventQueueWithWD.install();
* Action edtOverloadReport = ...;
* <p/>
* // install single-shot wg to report EDT overload after
* // 10-seconds timeout
* queue.addWatchdog(10000, edtOverloadReport, false);
* <p/>
* </pre>
*/
public class EventQueueWithWD extends EventQueue {
// Main timer
private final java.util.Timer timer = new java.util.Timer(true);
// Group of informational fields for describing the event
private final Object eventChangeLock = new Object();
private volatile long eventDispatchingStart = -1;
private volatile AWTEvent event = null;
/**
* Hidden utility constructor.
*/
private EventQueueWithWD() { }
/**
* Install alternative queue.
*
* #return instance of queue installed.
*/
public static EventQueueWithWD install() {
EventQueue eventQueue =
Toolkit.getDefaultToolkit().getSystemEventQueue();
EventQueueWithWD newEventQueue = new EventQueueWithWD();
eventQueue.push(newEventQueue);
return newEventQueue;
}
/**
* Record the event and continue with usual dispatching.
*
* #param anEvent event to dispatch.
*/
protected void dispatchEvent(AWTEvent anEvent) {
setEventDispatchingStart(anEvent, System.currentTimeMillis());
super.dispatchEvent(anEvent);
setEventDispatchingStart(null, -1);
}
/**
* Register event and dispatching start time.
*
* #param anEvent event.
* #param timestamp dispatching start time.
*/
private void setEventDispatchingStart(AWTEvent anEvent,
long timestamp) {
synchronized (eventChangeLock) {
event = anEvent;
eventDispatchingStart = timestamp;
}
}
/**
* Add watchdog timer. Timer will trigger <code>listener</code>
* if the queue dispatching event longer than specified
* <code>maxProcessingTime</code>. If the timer is
* <code>repetitive</code> then it will trigger additional
* events if the processing 2x, 3x and further longer than
* <code>maxProcessingTime</code>.
*
* #param maxProcessingTime maximum processing time.
* #param listener listener for events. The listener
* will receive <code>AWTEvent</code>
* as source of event.
* #param repetitive TRUE to trigger consequent events
* for 2x, 3x and further periods.
*/
public void addWatchdog(long maxProcessingTime,
ActionListener listener,
boolean repetitive) {
Watchdog checker = new Watchdog(maxProcessingTime, listener,
repetitive);
timer.schedule(checker, maxProcessingTime,
maxProcessingTime);
}
/**
* Checks if the processing of the event is longer than the
* specified <code>maxProcessingTime</code>. If so then
* listener is notified.
*/
private class Watchdog extends TimerTask {
// Settings
private final long maxProcessingTime;
private final ActionListener listener;
private final boolean repetitive;
// Event reported as "lengthy" for the last time. Used to
// prevent repetitive behaviour in non-repeatitive timers.
private AWTEvent lastReportedEvent = null;
/**
* Creates timer.
*
* #param maxProcessingTime maximum event processing time
* before listener is notified.
* #param listener listener to notify.
* #param repetitive TRUE to allow consequent
* notifications for the same event
*/
private Watchdog(long maxProcessingTime,
ActionListener listener,
boolean repetitive) {
if (listener == null)
throw new IllegalArgumentException(
"Listener cannot be null.");
if (maxProcessingTime < 0)
throw new IllegalArgumentException(
"Max locking period should be greater than zero");
this.maxProcessingTime = maxProcessingTime;
this.listener = listener;
this.repetitive = repetitive;
}
public void run() {
long time;
AWTEvent currentEvent;
// Get current event requisites
synchronized (eventChangeLock) {
time = eventDispatchingStart;
currentEvent = event;
}
long currentTime = System.currentTimeMillis();
// Check if event is being processed longer than allowed
if (time != -1 && (currentTime - time > maxProcessingTime) &&
(repetitive || currentEvent != lastReportedEvent)) {
listener.actionPerformed(
new ActionEvent(currentEvent, -1, null));
lastReportedEvent = currentEvent;
}
}
}
}
SampleEQUsage.java:
import javax.swing.*;
import java.awt.event.ActionEvent;
import java.util.Date;
/**
* Sample usage of <code>EventQueueWithWD</code> class.
*/
public class SampleEQUsage extends JFrame
{
public SampleEQUsage()
{
super("Sample EQ Usage");
setDefaultCloseOperation(EXIT_ON_CLOSE);
getContentPane().add(new JButton(new AbstractAction("Go")
{
public void actionPerformed(ActionEvent e)
{
System.out.println();
System.out.println(new Date());
try
{
// Sleep for 10 seconds
Thread.sleep(10000);
} catch (InterruptedException e1)
{
}
}
}));
setSize(100, 100);
}
public static void main(String[] args)
{
initQueue();
SampleEQUsage sequ = new SampleEQUsage();
sequ.setVisible(true);
}
// Install and init the alternative queue
private static void initQueue()
{
EventQueueWithWD queue = EventQueueWithWD.install();
// Install 3-seconds single-shot watchdog timer
queue.addWatchdog(3000, new AbstractAction()
{
public void actionPerformed(ActionEvent e)
{
System.out.println(new Date() + " 3 seconds - single-shot");
}
}, false);
// Install 3-seconds multi-shot watchdog timer
queue.addWatchdog(3000, new AbstractAction()
{
public void actionPerformed(ActionEvent e)
{
System.out.println(new Date() + " 3 seconds - multi-shot");
}
}, true);
// Install 11-seconds multi-shot watchdog timer
queue.addWatchdog(11000, new AbstractAction()
{
public void actionPerformed(ActionEvent e)
{
System.out.println(new Date() + " 11 seconds - multi-shot");
}
}, true);
}
}

Resettable CountdownLatch

I need something which is directly equivalent to CountDownLatch, but is resettable (remaining thread-safe!). I can't use classic synchronisation constructs as they simply don't work in this situation (complex locking issues). At the moment, I'm creating many CountDownLatch objects, each replacing the previous one. I believe this is doing in the young generation in the GC (due to the sheer number of objects). You can see the code which uses the latches below (it's part of the java.net mock for a ns-3 network simulator interface).
Some ideas might be to try CyclicBarrier (JDK5+) or Phaser (JDK7)
I can test code and get back to anyone that finds a solution to this problem, since I'm the only one who can insert it into the running system to see what happens :)
/**
*
*/
package kokunet;
import java.io.IOException;
import java.nio.channels.ClosedSelectorException;
import java.util.HashMap;
import java.util.Map;
import java.util.concurrent.CountDownLatch;
import java.util.concurrent.TimeUnit;
import kokuks.IConnectionSocket;
import kokuks.KKSAddress;
import kokuks.KKSSocket;
import kokuks.KKSSocketListener;
/**
* KSelector
* #version 1.0
* #author Chris Dennett
*/
public class KSelector extends SelectorImpl {
// True if this Selector has been closed
private volatile boolean closed = false;
// Lock for close and cleanup
final class CloseLock {}
private final Object closeLock = new CloseLock();
private volatile boolean selecting = false;
private volatile boolean wakeup = false;
class SocketListener implements KKSSocketListener {
protected volatile CountDownLatch latch = null;
/**
*
*/
public SocketListener() {
newLatch();
}
protected synchronized CountDownLatch newLatch() {
return this.latch = new CountDownLatch(1);
}
protected synchronized void refreshReady(KKSSocket socket) {
if (!selecting) return;
synchronized (socketToChannel) {
SelChImpl ch = socketToChannel.get(socket);
if (ch == null) {
System.out.println("ks sendCB: channel not found for socket: " + socket);
return;
}
synchronized (channelToKey) {
SelectionKeyImpl sk = channelToKey.get(ch);
if (sk != null) {
if (handleSelect(sk)) {
latch.countDown();
}
}
}
}
}
#Override
public void connectionSucceeded(KKSSocket socket) {
refreshReady(socket);
}
#Override
public void connectionFailed(KKSSocket socket) {
refreshReady(socket);
}
#Override
public void dataSent(KKSSocket socket, long bytesSent) {
refreshReady(socket);
}
#Override
public void sendCB(KKSSocket socket, long bytesAvailable) {
refreshReady(socket);
}
#Override
public void onRecv(KKSSocket socket) {
refreshReady(socket);
}
#Override
public void newConnectionCreated(KKSSocket socket, KKSSocket newSocket, KKSAddress remoteaddress) {
refreshReady(socket);
}
#Override
public void normalClose(KKSSocket socket) {
wakeup();
}
#Override
public void errorClose(KKSSocket socket) {
wakeup();
}
}
protected final Map<KKSSocket, SelChImpl> socketToChannel = new HashMap<KKSSocket, SelChImpl>();
protected final Map<SelChImpl, SelectionKeyImpl> channelToKey = new HashMap<SelChImpl, SelectionKeyImpl>();
protected final SocketListener currListener = new SocketListener();
protected Thread selectingThread = null;
SelChImpl getChannelForSocket(KKSSocket s) {
synchronized (socketToChannel) {
return socketToChannel.get(s);
}
}
SelectionKeyImpl getSelKeyForChannel(KKSSocket s) {
synchronized (channelToKey) {
return channelToKey.get(s);
}
}
protected boolean markRead(SelectionKeyImpl impl) {
synchronized (impl) {
if (!impl.isValid()) return false;
impl.nioReadyOps(impl.readyOps() | SelectionKeyImpl.OP_READ);
return selectedKeys.add(impl);
}
}
protected boolean markWrite(SelectionKeyImpl impl) {
synchronized (impl) {
if (!impl.isValid()) return false;
impl.nioReadyOps(impl.readyOps() | SelectionKeyImpl.OP_WRITE);
return selectedKeys.add(impl);
}
}
protected boolean markAccept(SelectionKeyImpl impl) {
synchronized (impl) {
if (!impl.isValid()) return false;
impl.nioReadyOps(impl.readyOps() | SelectionKeyImpl.OP_ACCEPT);
return selectedKeys.add(impl);
}
}
protected boolean markConnect(SelectionKeyImpl impl) {
synchronized (impl) {
if (!impl.isValid()) return false;
impl.nioReadyOps(impl.readyOps() | SelectionKeyImpl.OP_CONNECT);
return selectedKeys.add(impl);
}
}
/**
* #param provider
*/
protected KSelector(SelectorProvider provider) {
super(provider);
}
/* (non-Javadoc)
* #see kokunet.SelectorImpl#implClose()
*/
#Override
protected void implClose() throws IOException {
provider().getApp().printMessage("implClose: closed: " + closed);
synchronized (closeLock) {
if (closed) return;
closed = true;
for (SelectionKey sk : keys) {
provider().getApp().printMessage("dereg1");
deregister((AbstractSelectionKey)sk);
provider().getApp().printMessage("dereg2");
SelectableChannel selch = sk.channel();
if (!selch.isOpen() && !selch.isRegistered())
((SelChImpl)selch).kill();
}
implCloseInterrupt();
}
}
protected void implCloseInterrupt() {
wakeup();
}
private boolean handleSelect(SelectionKey k) {
synchronized (k) {
boolean notify = false;
if (!k.isValid()) {
k.cancel();
((SelectionKeyImpl)k).channel.socket().removeListener(currListener);
return false;
}
SelectionKeyImpl ski = (SelectionKeyImpl)k;
if ((ski.interestOps() & SelectionKeyImpl.OP_READ) != 0) {
if (ski.channel.socket().getRxAvailable() > 0) {
notify |= markRead(ski);
}
}
if ((ski.interestOps() & SelectionKeyImpl.OP_WRITE) != 0) {
if (ski.channel.socket().getTxAvailable() > 0) {
notify |= markWrite(ski);
}
}
if ((ski.interestOps() & SelectionKeyImpl.OP_CONNECT) != 0) {
if (!ski.channel.socket().isConnectionless()) {
IConnectionSocket cs = (IConnectionSocket)ski.channel.socket();
if (!ski.channel.socket().isAccepting() && !cs.isConnecting() && !cs.isConnected()) {
notify |= markConnect(ski);
}
}
}
if ((ski.interestOps() & SelectionKeyImpl.OP_ACCEPT) != 0) {
//provider().getApp().printMessage("accept check: ski: " + ski + ", connectionless: " + ski.channel.socket().isConnectionless() + ", listening: " + ski.channel.socket().isListening() + ", hasPendingConn: " + (ski.channel.socket().isConnectionless() ? "nope!" : ((IConnectionSocket)ski.channel.socket()).hasPendingConnections()));
if (!ski.channel.socket().isConnectionless() && ski.channel.socket().isListening()) {
IConnectionSocket cs = (IConnectionSocket)ski.channel.socket();
if (cs.hasPendingConnections()) {
notify |= markAccept(ski);
}
}
}
return notify;
}
}
private boolean handleSelect() {
boolean notify = false;
// get initial status
for (SelectionKey k : keys) {
notify |= handleSelect(k);
}
return notify;
}
/* (non-Javadoc)
* #see kokunet.SelectorImpl#doSelect(long)
*/
#Override
protected int doSelect(long timeout) throws IOException {
processDeregisterQueue();
long timestartedms = System.currentTimeMillis();
synchronized (selectedKeys) {
synchronized (currListener) {
wakeup = false;
selectingThread = Thread.currentThread();
selecting = true;
}
try {
handleSelect();
if (!selectedKeys.isEmpty() || timeout == 0) {
return selectedKeys.size();
}
//TODO: useless op if we have keys available
for (SelectionKey key : keys) {
((SelectionKeyImpl)key).channel.socket().addListener(currListener);
}
try {
while (!wakeup && isOpen() && selectedKeys.isEmpty()) {
CountDownLatch latch = null;
synchronized (currListener) {
if (wakeup || !isOpen() || !selectedKeys.isEmpty()) {
break;
}
latch = currListener.newLatch();
}
try {
if (timeout > 0) {
long currtimems = System.currentTimeMillis();
long remainingMS = (timestartedms + timeout) - currtimems;
if (remainingMS > 0) {
latch.await(remainingMS, TimeUnit.MILLISECONDS);
} else {
break;
}
} else {
latch.await();
}
} catch (InterruptedException e) {
}
}
return selectedKeys.size();
} finally {
for (SelectionKey key : keys) {
((SelectionKeyImpl)key).channel.socket().removeListener(currListener);
}
}
} finally {
synchronized (currListener) {
selecting = false;
selectingThread = null;
wakeup = false;
}
}
}
}
/* (non-Javadoc)
* #see kokunet.SelectorImpl#implRegister(kokunet.SelectionKeyImpl)
*/
#Override
protected void implRegister(SelectionKeyImpl ski) {
synchronized (closeLock) {
if (closed) throw new ClosedSelectorException();
synchronized (channelToKey) {
synchronized (socketToChannel) {
keys.add(ski);
socketToChannel.put(ski.channel.socket(), ski.channel);
channelToKey.put(ski.channel, ski);
}
}
}
}
/* (non-Javadoc)
* #see kokunet.SelectorImpl#implDereg(kokunet.SelectionKeyImpl)
*/
#Override
protected void implDereg(SelectionKeyImpl ski) throws IOException {
synchronized (channelToKey) {
synchronized (socketToChannel) {
keys.remove(ski);
socketToChannel.remove(ski.channel.socket());
channelToKey.remove(ski.channel);
SelectableChannel selch = ski.channel();
if (!selch.isOpen() && !selch.isRegistered())
((SelChImpl)selch).kill();
}
}
}
/* (non-Javadoc)
* #see kokunet.SelectorImpl#wakeup()
*/
#Override
public Selector wakeup() {
synchronized (currListener) {
if (selecting) {
wakeup = true;
selecting = false;
selectingThread.interrupt();
selectingThread = null;
}
}
return this;
}
}
Cheers,
Chris

I copied CountDownLatch and implemented a reset() method that resets the internal Sync class to its initial state (starting count) :) Appears to work fine. No more unnecessary object creation \o/ It was not possible to subclass because sync was private. Boo.
import java.util.concurrent.CyclicBarrier;
import java.util.concurrent.TimeUnit;
import java.util.concurrent.locks.AbstractQueuedSynchronizer;
/**
* A synchronization aid that allows one or more threads to wait until
* a set of operations being performed in other threads completes.
*
* <p>A {#code CountDownLatch} is initialized with a given <em>count</em>.
* The {#link #await await} methods block until the current count reaches
* zero due to invocations of the {#link #countDown} method, after which
* all waiting threads are released and any subsequent invocations of
* {#link #await await} return immediately. This is a one-shot phenomenon
* -- the count cannot be reset. If you need a version that resets the
* count, consider using a {#link CyclicBarrier}.
*
* <p>A {#code CountDownLatch} is a versatile synchronization tool
* and can be used for a number of purposes. A
* {#code CountDownLatch} initialized with a count of one serves as a
* simple on/off latch, or gate: all threads invoking {#link #await await}
* wait at the gate until it is opened by a thread invoking {#link
* #countDown}. A {#code CountDownLatch} initialized to <em>N</em>
* can be used to make one thread wait until <em>N</em> threads have
* completed some action, or some action has been completed N times.
*
* <p>A useful property of a {#code CountDownLatch} is that it
* doesn't require that threads calling {#code countDown} wait for
* the count to reach zero before proceeding, it simply prevents any
* thread from proceeding past an {#link #await await} until all
* threads could pass.
*
* <p><b>Sample usage:</b> Here is a pair of classes in which a group
* of worker threads use two countdown latches:
* <ul>
* <li>The first is a start signal that prevents any worker from proceeding
* until the driver is ready for them to proceed;
* <li>The second is a completion signal that allows the driver to wait
* until all workers have completed.
* </ul>
*
* <pre>
* class Driver { // ...
* void main() throws InterruptedException {
* CountDownLatch startSignal = new CountDownLatch(1);
* CountDownLatch doneSignal = new CountDownLatch(N);
*
* for (int i = 0; i < N; ++i) // create and start threads
* new Thread(new Worker(startSignal, doneSignal)).start();
*
* doSomethingElse(); // don't let run yet
* startSignal.countDown(); // let all threads proceed
* doSomethingElse();
* doneSignal.await(); // wait for all to finish
* }
* }
*
* class Worker implements Runnable {
* private final CountDownLatch startSignal;
* private final CountDownLatch doneSignal;
* Worker(CountDownLatch startSignal, CountDownLatch doneSignal) {
* this.startSignal = startSignal;
* this.doneSignal = doneSignal;
* }
* public void run() {
* try {
* startSignal.await();
* doWork();
* doneSignal.countDown();
* } catch (InterruptedException ex) {} // return;
* }
*
* void doWork() { ... }
* }
*
* </pre>
*
* <p>Another typical usage would be to divide a problem into N parts,
* describe each part with a Runnable that executes that portion and
* counts down on the latch, and queue all the Runnables to an
* Executor. When all sub-parts are complete, the coordinating thread
* will be able to pass through await. (When threads must repeatedly
* count down in this way, instead use a {#link CyclicBarrier}.)
*
* <pre>
* class Driver2 { // ...
* void main() throws InterruptedException {
* CountDownLatch doneSignal = new CountDownLatch(N);
* Executor e = ...
*
* for (int i = 0; i < N; ++i) // create and start threads
* e.execute(new WorkerRunnable(doneSignal, i));
*
* doneSignal.await(); // wait for all to finish
* }
* }
*
* class WorkerRunnable implements Runnable {
* private final CountDownLatch doneSignal;
* private final int i;
* WorkerRunnable(CountDownLatch doneSignal, int i) {
* this.doneSignal = doneSignal;
* this.i = i;
* }
* public void run() {
* try {
* doWork(i);
* doneSignal.countDown();
* } catch (InterruptedException ex) {} // return;
* }
*
* void doWork() { ... }
* }
*
* </pre>
*
* <p>Memory consistency effects: Actions in a thread prior to calling
* {#code countDown()}
* <i>happen-before</i>
* actions following a successful return from a corresponding
* {#code await()} in another thread.
*
* #since 1.5
* #author Doug Lea
*/
public class ResettableCountDownLatch {
/**
* Synchronization control For CountDownLatch.
* Uses AQS state to represent count.
*/
private static final class Sync extends AbstractQueuedSynchronizer {
private static final long serialVersionUID = 4982264981922014374L;
public final int startCount;
Sync(int count) {
this.startCount = count;
setState(startCount);
}
int getCount() {
return getState();
}
public int tryAcquireShared(int acquires) {
return getState() == 0? 1 : -1;
}
public boolean tryReleaseShared(int releases) {
// Decrement count; signal when transition to zero
for (;;) {
int c = getState();
if (c == 0)
return false;
int nextc = c-1;
if (compareAndSetState(c, nextc))
return nextc == 0;
}
}
public void reset() {
setState(startCount);
}
}
private final Sync sync;
/**
* Constructs a {#code CountDownLatch} initialized with the given count.
*
* #param count the number of times {#link #countDown} must be invoked
* before threads can pass through {#link #await}
* #throws IllegalArgumentException if {#code count} is negative
*/
public ResettableCountDownLatch(int count) {
if (count < 0) throw new IllegalArgumentException("count < 0");
this.sync = new Sync(count);
}
/**
* Causes the current thread to wait until the latch has counted down to
* zero, unless the thread is {#linkplain Thread#interrupt interrupted}.
*
* <p>If the current count is zero then this method returns immediately.
*
* <p>If the current count is greater than zero then the current
* thread becomes disabled for thread scheduling purposes and lies
* dormant until one of two things happen:
* <ul>
* <li>The count reaches zero due to invocations of the
* {#link #countDown} method; or
* <li>Some other thread {#linkplain Thread#interrupt interrupts}
* the current thread.
* </ul>
*
* <p>If the current thread:
* <ul>
* <li>has its interrupted status set on entry to this method; or
* <li>is {#linkplain Thread#interrupt interrupted} while waiting,
* </ul>
* then {#link InterruptedException} is thrown and the current thread's
* interrupted status is cleared.
*
* #throws InterruptedException if the current thread is interrupted
* while waiting
*/
public void await() throws InterruptedException {
sync.acquireSharedInterruptibly(1);
}
public void reset() {
sync.reset();
}
/**
* Causes the current thread to wait until the latch has counted down to
* zero, unless the thread is {#linkplain Thread#interrupt interrupted},
* or the specified waiting time elapses.
*
* <p>If the current count is zero then this method returns immediately
* with the value {#code true}.
*
* <p>If the current count is greater than zero then the current
* thread becomes disabled for thread scheduling purposes and lies
* dormant until one of three things happen:
* <ul>
* <li>The count reaches zero due to invocations of the
* {#link #countDown} method; or
* <li>Some other thread {#linkplain Thread#interrupt interrupts}
* the current thread; or
* <li>The specified waiting time elapses.
* </ul>
*
* <p>If the count reaches zero then the method returns with the
* value {#code true}.
*
* <p>If the current thread:
* <ul>
* <li>has its interrupted status set on entry to this method; or
* <li>is {#linkplain Thread#interrupt interrupted} while waiting,
* </ul>
* then {#link InterruptedException} is thrown and the current thread's
* interrupted status is cleared.
*
* <p>If the specified waiting time elapses then the value {#code false}
* is returned. If the time is less than or equal to zero, the method
* will not wait at all.
*
* #param timeout the maximum time to wait
* #param unit the time unit of the {#code timeout} argument
* #return {#code true} if the count reached zero and {#code false}
* if the waiting time elapsed before the count reached zero
* #throws InterruptedException if the current thread is interrupted
* while waiting
*/
public boolean await(long timeout, TimeUnit unit)
throws InterruptedException {
return sync.tryAcquireSharedNanos(1, unit.toNanos(timeout));
}
/**
* Decrements the count of the latch, releasing all waiting threads if
* the count reaches zero.
*
* <p>If the current count is greater than zero then it is decremented.
* If the new count is zero then all waiting threads are re-enabled for
* thread scheduling purposes.
*
* <p>If the current count equals zero then nothing happens.
*/
public void countDown() {
sync.releaseShared(1);
}
/**
* Returns the current count.
*
* <p>This method is typically used for debugging and testing purposes.
*
* #return the current count
*/
public long getCount() {
return sync.getCount();
}
/**
* Returns a string identifying this latch, as well as its state.
* The state, in brackets, includes the String {#code "Count ="}
* followed by the current count.
*
* #return a string identifying this latch, as well as its state
*/
public String toString() {
return super.toString() + "[Count = " + sync.getCount() + "]";
}
}

Phaser has more options, we can implement resettable countdownLatch using that.
Please read below basic concepts from the following sites
https://examples.javacodegeeks.com/core-java/util/concurrent/phaser/java-util-concurrent-phaser-example/
http://netjs.blogspot.in/2016/01/phaser-in-java-concurrency.html
import java.util.concurrent.Phaser;
/**
* Resettable countdownLatch using phaser
*/
public class PhaserExample {
public static void main(String[] args) throws InterruptedException {
Phaser phaser = new Phaser(3); // you can use constructor hint or
// register() or mixture of both
// register self... so parties are incremented to 4 (3+1) now
phaser.register();
//register is one time call for all the phases.
//means no need to register for every phase
int phasecount = phaser.getPhase();
System.out.println("Phasecount is " + phasecount);
new PhaserExample().testPhaser(phaser, 2000);
new PhaserExample().testPhaser(phaser, 4000);
new PhaserExample().testPhaser(phaser, 6000);
// similar to await() in countDownLatch/CyclicBarrier
// parties are decremented to 3 (4+1) now
phaser.arriveAndAwaitAdvance();
// once all the thread arrived at same level, barrier opens
System.out.println("Barrier has broken.");
phasecount = phaser.getPhase();
System.out.println("Phasecount is " + phasecount);
//second phase
new PhaserExample().testPhaser(phaser, 2000);
new PhaserExample().testPhaser(phaser, 4000);
new PhaserExample().testPhaser(phaser, 6000);
phaser.arriveAndAwaitAdvance();
// once all the thread arrived at same level, barrier opens
System.out.println("Barrier has broken.");
phasecount = phaser.getPhase();
System.out.println("Phasecount is " + phasecount);
}
private void testPhaser(final Phaser phaser, final int sleepTime) {
// phaser.register(); //Already constructor hint is given so not
// required
new Thread() {
#Override
public void run() {
try {
Thread.sleep(sleepTime);
System.out.println(Thread.currentThread().getName() + " arrived");
// phaser.arrive(); //similar to CountDownLatch#countDown()
phaser.arriveAndAwaitAdvance();// thread will wait till Barrier opens
// arriveAndAwaitAdvance is similar to CyclicBarrier#await()
}
catch (InterruptedException e) {
e.printStackTrace();
}
System.out.println(Thread.currentThread().getName() + " after passing barrier");
}
}.start();
}
}

Based on #Fidel -s answer, I made a drop-in replacement for ResettableCountDownLatch. The changes I made
mLatch is private volatile
mInitialCount is private final
the return type of the simple await() has changed to void.
Otherwise, the original code is cool too. So, this is the full, enhanced code:
public class ResettableCountDownLatch {
private final int initialCount;
private volatile CountDownLatch latch;
public ResettableCountDownLatch(int count) {
initialCount = count;
latch = new CountDownLatch(count);
}
public void reset() {
latch = new CountDownLatch(initialCount);
}
public void countDown() {
latch.countDown();
}
public void await() throws InterruptedException {
latch.await();
}
public boolean await(long timeout, TimeUnit unit) throws InterruptedException {
return latch.await(timeout, unit);
}
}
Update
Based on #Systemplanet-s comment, here is a safer version of reset():
// An atomic reference is required because reset() is not that atomic anymore, not even with `volatile`.
private final AtomicReference<CountDownLatch> latchHolder = new AtomicReference<>();
public void reset() {
// obtaining a local reference for modifying the required latch
final CountDownLatch oldLatch = latchHolder.getAndSet(null);
if (oldLatch != null) {
// checking the count each time to prevent unnecessary countdowns due to parallel countdowns
while (0L < oldLatch.getCount()) {
oldLatch.countDown();
}
}
}
Basically, it's a choice between simplicity and safety. I.e. if you are willing to move the responsibility to the client of your code, then it's enough to set the reference null in reset().
On the other hand, if you want to make it easy for the users of this code, then you need to use a little more tricks.

I'm not sure if this is fatally flawed but I recently had the same problem and solved it by simply instantiating a new CountDownLatch object each time I wanted to reset. Something like this:
Waiter:
bla();
latch.await();
//now the latch has counted down to 0
blabla();
CountDowner
foo();
latch.countDown();
//now the latch has counted down to 0
latch = new CountDownLatch(1);
Waiter.receiveReferenceToNewLatch(latch);
bar();
Obviously this is a heavy abstraction but thus far it has worked for me and doesn't require you to tinker with any class definitions.

Use Phaser.
if only one thread should to do work. U can join AtomicBoolean and Phaser
AtomicBoolean someConditionInProgress = new AtomicBoolean("false"); Phaser onConditionalPhaser = new Phaser(1);
(some function) if (!someConditionInProgress.compareAndSet(false, true)) {
try {
//do something
} finally {
someConditionInProgress.set(false);
//release barier
onConditionalPhaser.arrive();
}
} else {
onConditionalPhaser.awaitAdvance(onConditionalPhaser.getPhase());
}

Looks like you want to turn asynchronous I/O to synchronous. The whole idea of using asynchronous I/O is to avoid threads, but CountDownLatch requres using threads. This is an obvious contradiction in your question. So, you can:
keep using threads and employ synchronous I/O instead of Selectors and the suff. This will be much more simple and reliable
keep using asynchronous I/0 and give up CountDownLatch. Then you need an asynchronous library - look at RxJava, Akka, or df4j.
continue to develop your project for fun. Then you can try to use java.util.Semaphore instead of CountDownLatch, or program your own synchronization class using synchronized/wait/notify.

public class ResettableLatch {
private static final class Sync extends AbstractQueuedSynchronizer {
Sync(int count) {
setState(count);
}
int getCount() {
return getState();
}
protected int tryAcquireShared(int acquires) {
return getState() == 0 ? 1 : -1;
}
public void reset(int count) {
setState(count);
}
protected boolean tryReleaseShared(int releases) {
for (;;) {
int c = getState();
if (c == 0)
return false;
int nextc = c - 1;
if (compareAndSetState(c, nextc))
return nextc == 0;
}
}
}
private final Sync sync;
public ResettableLatch(int count) {
if (count < 0)
throw new IllegalArgumentException("count < 0");
this.sync = new Sync(count);
}
public void await() throws InterruptedException {
sync.acquireSharedInterruptibly(1);
}
public boolean await(long timeout, TimeUnit unit) throws InterruptedException {
return sync.tryAcquireSharedNanos(1, unit.toNanos(timeout));
}
public void countDown() {
sync.releaseShared(1);
}
public long getCount() {
return sync.getCount();
}
public void reset(int count) {
sync.reset(count);
}
}
This worked for me.

From what I was able to understand from the OP explanation and source code, the resettable CountDownLatch is not quite adequate concept for the problem he was going to solve. The documentation of the CountDownLatch itself mentions the OP's use case as simple gate initialized with a count of one:
CountDownLatch initialized with a count of one serves as a simple
on/off latch, or gate: all threads invoking await wait at the gate
until it is opened by a thread invoking countDown.
, but CountDownLatch implementation does not go any further in this direction.
So, myself having a problem similar to that of OP's I decided to introduce a SimpleGate class with the following properties:
Number of permits is one, which means it can be either in On or Off state;
There is a dedicated thread, called Gate Keeper that is only allowed to shut off or open up the Gate;
The right of gate keeping is transferable;
the opening up the Gate immediately allows the threads, that tried to come through the Gate, to do it (this very logical feature has been overlooked in the other answers);
as the thread contention is expected to be high, fairness is supported as an option, this allows to decrease an effect of thread barging.
public class SimpleGate {
private static class Sync extends AbstractQueuedSynchronizer {
// State
private static final int SHUT = 1;
private static final int OPEN = 0;
private boolean fair;
public void setFair(boolean fair) {
this.fair = fair;
}
public void shutOff() {
super.setState(SHUT);
}
#Override
protected int tryAcquireShared(int arg) {
if (fair && super.hasQueuedPredecessors())
return -1;
return super.getState() == OPEN ? 1 : -1;
}
#Override
protected boolean tryReleaseShared(int arg) {
super.setState(OPEN);
return true;
}
}
private Sync sync = new Sync();
private volatile Thread gateKeeper = Thread.currentThread();
public SimpleGate(){
this(true);
}
public SimpleGate(boolean shutOff){
this(shutOff, false);
}
public SimpleGate(boolean shutOff, boolean fair){
if (shutOff)
sync.shutOff();
sync.setFair(fair);
}
public void comeThrough(){
if (Thread.currentThread() == gateKeeper)
throw new IllegalStateException("Gate Keeper thread is not supposed to come through the gate");
sync.acquireShared(0);
}
public void shutOff(){
if (Thread.currentThread() != gateKeeper)
throw new IllegalStateException("Only a Gate Keeper thread is allowed to shut off");
sync.shutOff();
}
public void openUp(){
if (Thread.currentThread() != gateKeeper)
throw new IllegalStateException("Only a Gate Keeper thread is allowed to open up");
sync.releaseShared(0);
}
public void transferOwnership(Thread newGateKeeper){
this.gateKeeper = newGateKeeper;
}
// an addition of waiting interruptibly and waiting for specified amount of time,
//if they are needed, is trivial
}

Another drop-in replacement
import java.util.concurrent.CountDownLatch;
import java.util.concurrent.TimeUnit;
public class ResettableCountDownLatch {
int mInitialCount;
CountDownLatch mLatch;
public ResettableCountDownLatch(int count) {
mInitialCount = count;
mLatch = new CountDownLatch(count);
}
public void reset() {
mLatch = new CountDownLatch(mInitialCount);
}
public void countDown() {
mLatch.countDown();
}
public boolean await() throws InterruptedException {
boolean result = mLatch.await();
return result;
}
public boolean await(long timeout, TimeUnit unit) throws InterruptedException {
boolean result = mLatch.await(timeout, unit);
return result;
}
}