我一直在网上寻找.NET中MVC设置的示例.我发现了许多例子,但它们在某些方面似乎都有所不同.我有一本关于设计模式的书,描述了MVC起源于Smalltalk,因此我读了几个人讨论它在该语言中的实现.以下是我用自己收集的内容编写的一个示例项目是一个正确的实现,但我对一些细节感到困惑.
我遇到的一个问题是对象构造的正确顺序.这是我的Program.cs中的impl
Model mdl = new Model(); Controller ctrl = new Controller(mdl); Application.Run(new Form1(ctrl,mdl));
风景:
我马上就遇到了一些我不确定的问题.首先,如果视图应该只从模型中读取数据以进行更新,但包含对它的引用,那么是什么阻止我从控制器对视图中的模型进行调用?程序员是否应该忽略它们暴露给模型成员函数的事实?我的另一个想法,也许是通知视图模型更新的事件,将为视图发送某种状态对象以更新自身.
public interface IView { double TopSpeed { get; } double ZeroTo60 { get; } int VehicleID { get; } string VehicleName { get; } } /// <summary> /// Assume the form has the following controls /// A button with a click event OnSaveClicked /// A comboBox with a selected index changed event OnSelectedIndexChanged /// A textBox that displays the vehicles top speed named mTextTopSpeed /// A textBox that displays the vehicles zero to 60 time named mTextZeroTo60 /// </summary> public partial class Form1 : Form,IView { private IController mController; private IModel mModel; public Form1(IController controller,IModel model) { InitializeComponent(); mController = controller; mController.SetListener(this); mModel = model; mModel.ModelChanged += new ModelUpdated(mModel_ModelChanged); } void mModel_ModelChanged(object sender,EventArgs e) { mTextTopSpeed.Text = mModel.TopSpeed.ToString(); mTextZeroTo60.Text = mModel.ZeroTo60.ToString(); } public double TopSpeed { get { return Double.Parse(mTextTopSpeed.Text); } } public double ZeroTo60 { get { return Double.Parse(mTextZeroTo60.Text); } } public int VehicleID { get { return (int)mComboVehicles.SelectedValue; } } public string VehicleName { get { return mComboVehicles.SelectedText; } } #region Form Events private void OnFormLoad(object sender,EventArgs e) { mComboVehicles.ValueMember = "Key"; mComboVehicles.DisplayMember = "Value"; mComboVehicles.DataSource = new BindingSource(mModel.VehicleList,null); } private void OnSelectedIndexChanged(object sender,EventArgs e) { mController.OnSelectedVehicleChanged(); } private void OnSaveClicked(object sender,EventArgs e) { mController.OnUpdateVehicle(); } #endregion }
控制者:
我实现控制器的方式唯一真正的问题是,对我来说似乎有点奇怪,可以构建控制器而不必明确分配视图.我可以完全忽略视图但这意味着我会将参数传递给控制器的函数以更新模型,这似乎完全忽略了这一点.
public interface IController { void OnUpdateVehicle(); void OnSelectedVehicleChanged(); void SetListener(IView view); } class Controller : IController { private IModel mModel; private IView mView = null; public Controller(IModel model) { mModel = model; } public void OnUpdateVehicle() { if(mView == null) return; mModel.UpdateVehicle(mView.VehicleID,mView.TopSpeed,mView.ZeroTo60); } public void SetListener(IView view) { mView = view; } public void OnSelectedVehicleChanged() { if (mView == null) return; mModel.SelectVehicle(mView.VehicleID); } }
该模型:
在我的表单中,我有一个组合框,它是我的伪数据库中给出的车辆列表.我觉得我的表单实际上应该实现多个视图/模型.特定于列出具有相应控制器/模型的可能车辆的视图,以及用于使用其自己的控制器/模型显示关于所选车辆的信息的视图.
public delegate void ModelUpdated(object sender,EventArgs e); public interface IModel { event ModelUpdated ModelChanged; void UpdateVehicle(int id,double topSpeed,double zeroTo60); void SelectVehicle(int id); double TopSpeed { get; } double ZeroTo60 { get; } IDictionary<int,string> VehicleList { get; } } // class for the sake of a pseudo database object class Vehicle { public int ID { get; set; } public string Name { get; set; } public double TopSpeed { get; set; } public double ZeroTo60 { get; set; } public Vehicle(int id,string name,double zeroTo60) { ID = id; Name = name; TopSpeed = topSpeed; ZeroTo60 = zeroTo60; } } class Model : IModel { private List<Vehicle> mVehicles = new List<Vehicle>() { new Vehicle(1,"Civic",120.0,5.0),new Vehicle(2,"Batmobile",9000.0,1.0),new Vehicle(3,"Tricycle",5.0,0.0) }; private Vehicle mCurrentVehicle; public Model() { mCurrentVehicle = mVehicles[0]; } public event ModelUpdated ModelChanged; public void OnModelChanged() { if (ModelChanged != null) { ModelChanged(this,new EventArgs()); } } public double TopSpeed { get { return mCurrentVehicle.TopSpeed; } } public double ZeroTo60 { get { return mCurrentVehicle.ZeroTo60; } } public IDictionary<int,string> VehicleList { get { Dictionary<int,string> vDict = new Dictionary<int,string>(); foreach (Vehicle v in mVehicles) { vDict.Add(v.ID,v.Name); } return vDict as IDictionary<int,string>; } } #region Pseudo Database Calls public void SelectVehicle(int id) { foreach (Vehicle v in mVehicles) { if (v.ID == id) { mCurrentVehicle = v; OnModelChanged(); // send notification to registered views break; } } } public void UpdateVehicle(int id,double zeroTo60) { foreach (Vehicle v in mVehicles) { if (v.ID == id) { mCurrentVehicle.TopSpeed = topSpeed; mCurrentVehicle.ZeroTo60 = zeroTo60; OnModelChanged(); // send notification to registered views break; } } } #endregion }
在这个tl; dr的结论中,我想我正在寻找的是关于我在这里做什么代表一个真正的MVC实现的一些指导,也许是某些人对上述问题有所了解.任何建议将不胜感激.
我们这取决于你想做什么.您目前有一个监督控制器的实现.如果您希望从视图中删除模型(以及任何数据绑定),则可以实现Passive View模式.有关更多差异,请参阅
this文章.
Passive View and Supervising Controller http://i.msdn.microsoft.com/dynimg/IC281772.png
而马丁福勒是国王(GUI Architectures).