CanExecute 仅部分起作用

Question

我正在开发需要使用按钮命令处理较长任务的 WPF .NET 5 应用程序。在任务完成之前，该按钮应该被禁用。我正在使用来自 Microsoft.Toolkit.Mvvm:
的 RelayCommand BtnCmd = new RelayCommand(DoSomething, CanDoSomething);
DoSomething 方法所做的第一件事是使 CanDoSomething 的 return 值为假。这可以防止再次执行 DoSomething，但它在 Button 上不可见。
在研究时，我发现 https://github.com/CommunityToolkit/MVVM-Samples/issues/41:

确实是这种情况

"Sergio0694 commented on Mar 28, 2021": "That is correct, by default RelayCommand will not automatically update its visual state on WPF...".

他推荐的解决方案是使用：https://gist.github.com/Sergio0694/130bc344e552e501563546454bd4e62a 和

<button xmlns:input="using:Microsoft.Toolkit.Mvvm.Wpf.Input"    
        Command="{Binding Command}"
        input:RelayCommandExtensions.IsCommandUpdateEnabled="True"/>

我的 DoSomething 方法如下所示：

private async void DoSomething()
{
    PropertyCheckedByCanDoSomething = false;
    await Task.Delay(1000);
    PropertyCheckedByCanDoSomething = true;
}

它会给出想要的视觉效果，但仅限于：PropertyCheckedByCanDoSomething = false; 使用 PropertyCheckedByCanDoSomething = true; 效果仅在单击应用程序或执行 window 切换后可见。
我该如何解决这个问题？
非常感谢您的支持。

Answer 1

在不知道你所有的绑定等是如何进行的情况下，我可能会通过子类化你的中继命令来处理它，比如

using System;
using System.Windows.Input;
namespace MyTestApp
{
    public class MyRelayCommand : ICommand
    {
        private readonly Action<object> _execute;
        private readonly Func<object, bool> _canExecute;

        public MyRelayCommand(Action<object> execute) : this(execute, CanAlwaysExecute)
        { }

        public MyRelayCommand(Action<object> execute, Func<object, bool> canExecute)
        {
            // Lamda expression to execute each respectively
            _execute = execute;
            _canExecute = canExecute;
        }

        public bool CanExecute(object cmdParm)
        { return _canExecute(cmdParm); }

        public static bool CanAlwaysExecute(object cmdParm)
        {   return true;    }

        public void Execute(object cmdParm)
        {
            if (!_doingWithCallback)
                _execute(cmdParm);
            else
                Execute2(cmdParm);
        }

        // The CanExecuteChanged event handler is required from the ICommand interface
        public event EventHandler CanExecuteChanged;
        public void RaiseCanExecuteChanged()
        {
            if (CanExecuteChanged != null)
                CanExecuteChanged(this, new EventArgs());
        }

        private bool _isMyTaskRunning = false;
        public bool IsMyTaskRunning
        { get { return _isMyTaskRunning; } }

        private bool _doingWithCallback;
        private readonly Action<object, Action> _executeWithCallback;

        public MyRelayCommand(Action<object, Action> executeWithCallback) : this( executeWithCallback, CanAlwaysExecute)
        { }

        public MyRelayCommand(Action<object, Action> executeWithCallback, Func<object, bool> canExecute)
        {
            // new flag, so when the default "Execute" method is called, it can then redirect to
            // calling the Execute2() method that checks to prevent the double-click and then
            // calls your function with the additional parameter of the action method to call upon completion.
            _doingWithCallback = true;
            _executeWithCallback = executeWithCallback;
            _canExecute = canExecute;
        }

        public void Execute2(object cmdParm)
        {
            // prevent double action if running vs not
            if (_isMyTaskRunning)
                return;

            // flag it to prevent double action
            _isMyTaskRunning = true;

            // trigger raising the CanExecute changed which will update the user interface
            RaiseCanExecuteChanged();

            // turn off when done, but if being done from a "task()" process, 
            // you probably want to have a return function be called when the 
            // TASK is finished to re-enable the button... maybe like

            // NOW, call your execute function that accepts TWO parameters.
            // the first is whatever parameter MAY come from the button click itself.
            // the SECOND parameter will be to accept MY ACTION HERE to reset the flag when finished
            System.Threading.Tasks.Task.Run(() => _executeWithCallback(cmdParm, ButtonTaskIsComplete));
        }

        public void ButtonTaskIsComplete()
        {
            _isMyTaskRunning = false;
            System.Windows.Application.Current.Dispatcher.Invoke(() => { RaiseCanExecuteChanged(); });
        }
    }
}

可能不是最合适的，但可能会为您提供一个可能的包装器解决方案。

这里是一个示例实现，可以在您现有的表单区域中调用它。

private MyRelayCommand _myFormButton;
public MyRelayCommand MyFormButton
{ get { return _myFormButton ?? ( _myFormButton = new MyRelayCommand( YourFormMethod )); } }

public void YourFormMethod(object cmdParm, System.Action doThisWhenFinished)
{
    MessageBox.Show("Something from after the click event of the button");

    // NOW, the callback function so the button re-enables itself once finished.
    doThisWhenFinished();
}

Answer 2

我遇到了与您提到的 RelayCommandExtension 类似的问题。

有一个 Github issue regarding this. The following solution 被张贴在那里：

在 RelayCommandExtensions.cs 中添加以下行：

notifiers.Add(obj, notifier);

到 OnIsCommandUpdateEnabledChanged(...) 方法，使其看起来像这样：

// Ensure a notifier is enabled so that we can monitor for changes to the command
// property in the future. This is also needed in case this attached property is
// set before the binding engine has actually resolved the target command instance.
if (!notifiers.TryGetValue(obj, out _))
{
    PropertyChangeNotifier notifier = new(obj, nameof(ICommandSource.Command));

    notifier.PropertyChanged += (_, _) => ToggleIsCommandUpdateEnabled(obj);
    
    notifiers.Add(obj, notifier);
}

这对我有用。

CanExecute 仅部分起作用

CanExecute works only partial

c#

wpf

mvvm

relaycommand

mvvm-toolkit