Question

Hi I have a problem when I am working on my Windows Phone silverlight C# application.

So I want to do something like this

1. Press a button on page one. The button click handler calls a async method MakeRequest in other class to retrieve data. The async method will fire a event DataReadyEvent and has the result wrapped as DataEventArgs.The handler of this event will be in Page2. So after add a handler to this event, I navigate to Page2 from current page.
2. I want to retrieve data by a event handler in Page 2 code behind and update that on UI. But the event handler is static (so that I can add it by using Page2.handler_method_name in page1 code without creating a new instance of the page.). Since the handler method is static, I cannot use Dispatcher.Invoke and get back to the UI thread and update UI.

So in this case, anyone has any idea to it? I just want to call a async method in page1, and update result to UI in page2. Thank you

Answer 1

Here is an idea: don’t make it static. Don’t try to create problems for yourself by breaking simple OOP rules like encapsulation, etc and by finding some crazy workarounds around the framework you work with.

When you are in such a situation you should stop, look back and think because it is an indication that you do something completely wrong. Don’t try to push it even further by finding hacks and workarounds. Rather you should refactor and reuse the correct paradigm.

For example, if you want to display the result on Page2, then there IS a Page2 ALREADY. So there IS an instance of it. Why do you want to use static handler then?
Probably because you don’t have a reference to this page. That’s fine, normally you shouldn’t.

But when you finish your computation you can publish an event saying “hey, here is the task done”. At that point you shouldn’t care who is interested in this result, that’s not the worker’s concern.
Which means that the logic of the computation itself should probably be moved out from Page1. Really, pages concern is dome presentation logic, nothing more.
Page1 should make a request that some computation needs to be done. And here will be an external component (perhaps something in your ViewModel) to actually make it happen.

So when the result is ready to be consumed, you can simply push it into a ViewModel (update some observable properties or collections, etc), so if there is any UI (or many of them, or other components) interested in this data it will be automatically notified and the data will be displayed.

But please don’t try to hack around, it will lead you to bigger pain in the future.