Important: Please read the Qt Code of Conduct -

QMutexLocker - SignalWaiter - how does the example work?

  • Hi all,

    in the docs for the class QMutexLocker there is an outline of an example for a SignalWaiter given.
    It is only an outline. Can somebody give me the details how it will work?

    I have to solve the following problem:

    I have a method like this:
    int SomeClass::myMethod()
    // here I have access to a socket. I must start some network-communication
    // over the socket to get the result for the method. E.g.

       mySocket->sendCommand( "someCommandId" );
       // The method should now wait till the result is there (or some error occurs).
       // So I want to wait for the readyRead-Signal now.
       // Can the SignalWaiter-class be used for it? How?


    I know that the Qt socket class already has some waitFor...-methods. But using these methods on Windows is not encouraged (...using this method will randomly fail on Windows...). What can I say? I'm on Windows.... So I'm looking for an alternative.

    Here is the example from the Qt-docs of QMutexLocker:

    class SignalWaiter
    QMutexLocker locker;

    SignalWaiter(QMutex *mutex)
    : locker(mutex)

    void waitForSignal()
        while (!signalled)


    Thanks in advance,
    Best regards


  • Lifetime Qt Champion


    Not a direct answer but I would have rather used a QEventLoop for that or the old QxtSignalWaiter class which does the job well.

    Out of curiosity, why do you need such a blocking behavior ?

  • Hi Samuel,

    thanks for the suggestions. I will take a look at the QxtSignalWaiter. I think it will do exactly what I want. Of course it is a "little overhead" using a whole new library for solving my little problem.

    What do you mean by "using the QEventLoop"? So far I have solved my problem with code like this:

    while ( ! _isReadyRead ) {
         // the boolean-variable _isReadyRead will be set inside of the slot which is called by the readyRead()-Signal

    But very often code like this is described as an "ugly hack".

    So I'm just looking for a better and elegant solution....

    ** Why do you need such a blocking behavior ?
    Little bit complicated to explain, a try: I have a main-program which I can't modify. The code which I can implement and modify is some plugin-code for the main-program. The main-program expects immediate results. When it calls the plugin-method myMethod which is implemented inside of the plugin a result has to be returned. I can't switch the main-program to asynchronous signal-slot-communication.

    int SomeClass::myMethod()
    // do some calculation here
    // and return the result of the calculation
    return myResult;


    Best regards

  • Lifetime Qt Champion

    IIRC, it's self-contained so You can extract it from the library.

    What I have seen in use:

    QEventLoop loop;
    connect(mySocket, &QTcpSocket::readyRead, &loop, &QEventLoop::exit);

    Which work a bit like a modal dialog.

    Ok, I see your use case, a classic.

Log in to reply