Question

I wanted to use a “presenter first” approach (the presenter attaches itself to it’s view and model in the Constructor using dependency injection via the constructor call).

My form is a single MainForm containing two user controls which each have their own presenters, but share the model, so I create all the presenters in Main, passing in the relevant user control to the presenter by having a property which exposes these controls from FormMain, and passing in a single instance of the model to all presenters.

static void Main()
{
Application.EnableVisualStyles();
Application.SetCompatibleTextRenderingDefault(false);

IDocumotiveCaptureView view = new DocumotiveCaptureView(); //this is the MainForm
IDocumentModel model = new DocumentModel(); //single model
IDocumotiveCapturePresenter Presenter = new DocumotiveCapturePresenter(view, model); //MainForm's presenter
IControlsPresenter ControlsPresenter = new ControlsPresenter(view.ControlsView, model); //ControlsPresenter - attached to the user control via view.ControlsView
IDocumentPresenter DocumentPresenter = new DocumentPresenter(view.DocumentView, model);//DocumentPresenter - attached to the user control via view.DocumentView)

Application.Run((Form)view);                                                         
}

Can anyone see anything inherently bad or wrong in this approach? I realise this MAY be subjective, but I’m very new to MVP and would like some feedback.

Answer 1

If it works for you, then of course it’s ok. MVP isn’t a religion. It’s not even considered a single pattern anymore. The two most popular interpretations seems now to be Passive View and Supervising controller.

The Main method in your case is your composition root (search for any answers by Mark Seeman) and it’s an excellent place to do what you do. But it’s not the only place to do it.

For more MVP in WinForms, please see Model-View-Presenter in WinForms