I have a Windows Phone 8.1 MVVM Light project and I am struggling to keep it Blendable.
As I see it I have a few options. I can load different view models depending on whether ViewModelBase.IsInDesignModeStatic is true in the ViewModelLocator constructor, or I can test ViewModelBase.IsInDesignModeStatic in the view model constructor and load data appropriately.
If ViewModelBase.IsInDesignModeStatic is true I need to load data from file. Here's my code:
public async Task<ThingsSampleDataSource> GetSampleDataAsync()
{
if (_DeserializedThingsSampleDataSource == null)
{
var dataUri = new Uri(_SampleDataJSONFile);
var file = await StorageFile.GetFileFromApplicationUriAsync(dataUri);
var jsonText = await FileIO.ReadTextAsync(file);
_DeserializedThingsSampleDataSource = JsonConvert.DeserializeObject<ThingsSampleDataSource>(jsonText);
}
return _DeserializedThingsSampleDataSource;
}
When I call that method I need to mark the call await and thus the calling method async. But constructors cannot be marked async.
Or I can provide a ContinueWith continuation instead of awaiting return of the asynchronous code. But Blend loads the page before the ContinueWith is complete.
Given that sample data is loaded in the view model or the locator service constructors and it has to load data from a file, an asynchronous activity, how do I do this in MVVM Light so that the sample data is available in Blend?
(N.B. other answers I have found, for example this one, do not use MVVM Light.)
Load your data on the page load event, using a Command, so you can take advantage of the await/async stuff. I don't know how this works with blend as I don't use it much.
View:
<i:Interaction.Triggers>
<i:EventTrigger EventName="Loaded">
<i:InvokeCommandAction Command="{Binding PageLoadedCommand}"/>
</i:EventTrigger>
</i:Interaction.Triggers>
ViewModel:
public RelayCommand PageLoadedCommand { get; private set; }
public MyConstructor(IService serviceInjected)
{
PageLoadedCommand = new RelayCommand(async()=>await OnPageLoaded());
....
}
private async Task OnPageLoaded()
{
if(ViewModelBase.IsInDesignModeStatic)
{
var data = await GetSampleDataAsync();
//Do something..
}
}
Related
My aim is to access bindable property across the the App. But My current framework ViewModel Instance create multiple time
My Requirement : I have the cart count in the bottomTray(CheckuoutViewModel) i want to increase the cart count any where in the app page, but in this cart count not update when back click, its only working on forward navigation, the reason behind CheckoutViewModel instance create each and every time. so that i'm try to instant creation at earlier.
Here I'm list out sample ViewModel and calling method
Login ViewModel
Checkuout ViewModel(This view model common for all page)
BaseNavigationViewModel(Its BaseViewModel)
As of now i'm calling when BindinContext each and every time like,
new LoginViewMode(navigation)
new CheckoutViewModel(navigation)
what will do to create all ViewModel instance when app start time like ViewModel Locator?
Im tried
public static ViewModelLocator Locator
{
get { return locator ?? (locator = new ViewModelLocator()); }
}
And ViewModel Locator
public ViewModelLocator()
{
navigation = App.Current.MainPage.Navigation;
}
internal CustomTabBarViewModel CustomTabBarVM
{
get
{
return customTabBarVM ?? (customTabBarVM = new CustomTabBarViewModel(navigation));
}
}
And CustomTabBar.xaml.cs
public CustomTabBar()
{
viewModel = App.Locator.CustomTabBarVM;
InitializeComponent();
BindingContext = viewModel;
}
and Expectation
App.Locator.CustomTabBarVM.BadgeCartCount = OrderObject.Instance.ORDER_OBJECT.Items.Count;
This approach is working fine but it's create some navigation issues
A singleton instance is a common feature of virtually all MVVM frameworks (Prism, FreshMVVM etc). If you aren't using a framework (if you aren't, I would STRONGLY advise you consider using one), below is a solution.
To obtain a single instance of a ViewModel you can use the App class to host the object and access it whenever you need.
Create a public static property of your ViewModel:
public static MyViewModel MyViewModelInstance { get; }
Create an instance in the constructor of the app
public App()
{
InitializeComponent();
MyViewModelInstance = new MyViewModel();
var myPage = new MyPage()
{
BindingContext = MyViewModelInstance
};
var navPage = new NavigationPage(myPage);
MainPage = navPage;
}
Whenever you create a new page, access the shared instance
// This method is just an example of how you might create a new page and wire up the view model
async void GoNextClicked(System.Object sender, System.EventArgs e)
{
var myPage = new MyPage()
{
BindingContext = App.MyViewModelInstance
};
await this.Navigation.PushAsync(myPage);
}
This approach comes with a few caveats, you are creating instances when the app loads not when they are needed (Eagerly loading). So a performance optimisation would be to use Lazy<T> to handle the creation of these objects. However this is logic that has already been written for you in MVVM frameworks, they are there to help you and you should be using them.
Lazy Load
You can save memory and performance at startup by lazy loading the viewmodel, here is this example rewritten to support this pattern:
public static MyViewModel MyViewModelInstance
{
get => _myViewModelInstanceFactory.Value;
}
private static Lazy<MyViewModel> _myViewModelInstanceFactory = new Lazy<MyViewModel>(() => new MyViewModel(), true);
public App()
{
InitializeComponent();
var myPage = new MyPage()
{
BindingContext = MyViewModelInstance
};
var navPage = new NavigationPage(myPage);
MainPage = navPage;
}
Now this object won't be created until it is accessed by your code, and once it has been accessed once it has already been created and will go on to live in memory for the rest of your apps lifecycle.
Axemasta has good answer about re-use of a shared view model instance.
I'll give an alternative approach to the underlying need given in one comment: how to have a static property (so the value is common), and Bind to it when Binding to an instance.
Use this approach if you do want a different CheckoutViewModel for each new page. For example, if there are other properties that should be set up differently, depending on the page.
public class CheckoutViewModel : : INotifyPropertyChanged // or your MVVM library's base class for ViewModels.
{
public static int SharedCount { get; set; }
public void IncrementCount()
{
Count = Count + 1;
}
public int Count {
get => SharedCount;
set {
// Exact code might be simpler if using an MVVM library.
if (SharedCount != value)
{
SharedCount = value;
OnPropertyChanged("Count");
}
}
}
}
}
LIMITATION: This assumes that only the current instance of CheckoutViewModel is visible; if you need to "notify" OTHER Views (update other CheckoutViewModel instances), then you'll need a "publish/subscribe" solution. In Xamarin Forms, one such solution is MessagingCenter.
I have two ViewModels, MainWindowShellViewModel(shellVm) and MainWindowContentViewModel(contentVm). The shellVm publishes an event and the contentVm subscribes to it.
The shell VM looks something like the following. I have omitted many details.
// ctor
public MainWindowShellViewModel(IEventAggregator eventAggregator)
{
_EventAggregator = eventAggregator ?? throw new ArgumentNullException(nameof(IEventAggregator) + " service injected is null!!!");
_AppStartingClosingEventToken = _EventAggregator.GetEvent<AppStartingClosingEvent>();
}
private void MainWindowShellLoaded()
{
var payload = new AppStartingClosingEventData();
payload.Data = "MainWindowStarting";
_AppStartingClosingEventToken.Publish(payload);
}
The AppStartingClosingEvent is a no brainer type as follows.
public class AppStartingClosingEvent : PubSubEvent<AppStartingClosingEventData>
{ }
public class AppStartingClosingEventData
{
public string Data { get; set; }
}
And finally, the contentVm looks as follows.
public MainWindowContentViewModel(IEventAggregator eventAggregator)
{
_AppClosingEventToken.Subscribe(AppStartingClosing);
}
private void AppStartingClosing(AppStartingClosingEventData appStartingClosingEventData)
{
if (appStartingClosingEventData.Data == "MainWindowStarting")
LoadState(appStartingClosingEventData);
if (appStartingClosingEventData.Data == "MainWindowClosing")
SaveState(appStartingClosingEventData);
}
I want to test the that the method AppStartingClosing inside of contentVm is called with proper data. I am using Moq
I am running out of ideas. Please suggest. Tried the following but so far no success.
How do I test Prism event aggregator subscriptions, on the UIThread?
Using Moq to verify a Prism event subscription fails
Unit testing with Moq, Prism 6, and Event Aggregation
Moq Event Aggregator Is it possible
// Verifying a delegate was called with Moq
EDIT
Here is what I have tried.
// Arrange
var mockingKernel = new MoqMockingKernel();
var eventAggregatorMock = mockingKernel.GetMock<IEventAggregator>();
var eventBeingListenedTo = new AppStartingClosingEvent();
eventAggregatorMock.Setup(e => e.GetEvent<AppStartingClosingEvent>()).Returns(eventBeingListenedTo);
var vm = mockingKernel.Get<MainWindowContentViewModel>();
var evData = new AppStartingClosingEventData();
evData.Data = "MainWindowStarting";
// Act
eventBeingListenedTo.Publish(evData);
Now, what should I do? I am not even clear if I have approached correctly.
Now what should I do?
After eventBeingListenedTo.Publish(evData); look whether whatever effect SaveState should have is actually happening.
I am not even clear if I have approached correctly.
You do not want to test whether one method in a class is called by another method of that class.
So instead of trying to do
subjectUnderTest.DoStuff();
MagicallyVerifyThatThisGotCalled( () => subjectUnderTest.SomeEffect() );
you should do
var subjectUnderTest = new SubjectUnderTest( serviceMock.Object );
subjectUnderTest.DoStuff();
serviceMock.Verify( x => x.SomeEffectOnTheService(), Times.Once );
Assert.That( subjectUnderTest.SomePropertyThatsChanged, Is.EqualTo( newValue ) );
Whatever SubjectUnderTest does internally to achieve the desired effect, is not in the scope of the test. It's private to SubjectUnderTest, you don't care how it is done as long as it is done at all. When testing, look at the externally visible state of your subject under test, and what it does to its dependencies.
I'm new to ASP.NET and I'm refactoring some functionalities in my MVC-structured ASP.NET application into area's. This has already lead to controller-methods not able to find their views anymore, which results in the following page:
To test if all controllers can find their views, I'd like to write some automated unit tests for this.
I have came up with the following:
[TestMethod]
public void AboutTest()
{
var controller = new HomeController();
var result = controller.About() as ViewResult;
Assert.IsNotNull(result);
}
which tests the About-method in the following code:
public class HomeController : Controller
{
public ActionResult About()
{
return View();
}
public ActionResult Contact()
{
return View("~/Views/SomeFolder/Contact.cshtml");
}
}
But even when the HomeControllers About-method can not find a view, this assert succeeds, so this does not work for me.
I have found a solution online to use use ViewEngine.FindView() here. I don't think I can use this, since in some controllers the views are referenced by a hardcoded string (see the contact method in the example controller above) instead of just returning the default view (simularly named as its method). The ViewEngine.FindView(controller.ControllerContext, "about", "about"); will then fail, but the controller-method would not.
Another solution states to use Assert.IsEqual() and check if the result.ViewName is equal to a hardcoded string (for example: "About"). Since I do not set or know the title of the views I'm expecting to get returned, this would not be a solution either.
(How) would I be able to test my application for this?
You shouldn't check for null, it will return a ViewResult even when it doesn't render.
To test whether it actually renders use AssertViewRendered from mvccontrib.
[TestMethod]
public void AboutTest()
{
var controller = new HomeController();
var result = controller.About().AssertViewRendered();
}
You can even check for a specific view like so:
result.AssertViewRendered().ForView(MVC.Your.Views.AboutView);
Or supply data like so:
controller.page().AssertViewRendered().ForView("page").WithViewData<SomeModel>();
For an interactive tutorial with lots of pictures I can recommend: http://toreaurstad.blogspot.nl/2011/09/adventures-with-mvccontrib-testhelper.html
Edit:
You might also check out Selenium to test your entire app (incl. rendering of 200 routes).
I am asking this because after long time searching I haven't found a good answer on this yet...
Here is what I want:
Example: I have a domain model "JobPosting" which a user should be able to change state to published, if it is still a draft. Before publishing I must not only validate the model properties I must also validate many different requirements regarding the user account, it's registered company etc. All this validation logic is put into a service layer. So far so good...
This is how my service layer looks like:
public IValidationResult ValidatePublish(JobPosting jobPosting){
...
}
public void Publish(JobPosting jobPosting){
jobPosting.State = JobPostingState.Published;
...
}
Any my controller:
public ActionResult Publish(PublishViewModel model){
...
var validationResult = _jobService.ValidatePublish(jobPosting);
if(validationResult.Success){
_jobService.Publish(jobPosting);
...
}
...
}
And here now my questions:
I want to be able to call the ValidatePublish from the controller to show validation errors in the view. However I must never be able to publish a job when validation fails.
So to have my code more robust I added a second validation check in my Publish method in service layer:
public void Publish(JobPosting jobPosting){
if(ValidatePublish(jobPosting).Success){
jobPosting.State = JobPostingState.Published;
...
}
}
but I have not such a good feeling with this approach because now I am calling the validation twice when validation is OK during each controller publish request.
What do you think. Is the second call to much? Is there a better approach?
I am asking because my whole application looks like that and if I would ever forget a validation call in controller I might end up with an not allowed domain model state in database. That's why I added the second validation check in each service method.
Thanks in advance for your thoughts on this!!!
One quick solution might be to have the Publisher class require the JobPosting and IValidationResult objects as arguments.
public void Publish(JobPosting jobPosting, IValidationResult validation)
{
if (validation.IsValid)
{
jobPosting.State = JobPostingState.Published;
// other work here...
}
}
Your Controller can then call the Validator, receive an IValidationResult and pass that back to the presentation layer if needed. Otherwise pass on to Publisher
public ActionResult Publish(PublishViewModel model)
{
var validationResult = _jobService.ValidatePublish(jobPosting);
if(validationResult.Success) _jobService.Publish(jobPosting, validationResult);
else return View("error", validationResult);
}
Edit:
A cleaner solution may be to have the Publisher class return a PublishAttempt result.
public class PublishAttempt : IValidationResult
{
public enum AttemptOutcome {get; set;}
}
public ActionResult Publish(PublishViewModel model)
{
var attempt = _jobService.Publish(jobPosting);
if (attempt.Success) return View("success");
else return View("error", attempt.ValidationResults);
}
The following just came into my mind... what do you think:
I change my service method to:
public IValidationResult Publish(JobPosting jobPosting, bool validateOnly = false){
var validationResult = ValidatePublish(jobPosting);
if(validateOnly) return validationResult;
jobPosting.State = JobPostingState.Published;
...
return validationResult;
}
And then in controller I always call only the Publish method and not the extra ValidatePublish anymore:
public ActionResult Publish(PublishViewModel model)
{
var validationResult = _jobService.Publish(jobPosting);
if(!validationResult.Success) return View("error", validationResult);
}
And when I need only simple validation I do
var validationResult = _jobService.Publish(jobPosting, true);
Is this okey to do it like that?
Or is it not good looking if a normal service call returns IValidationResult?
I want to know what the advantage are using the relay command to call functions that refresh the screen. In my application I have the following relay command setup.
private RelayCommand _refreshSitesCommand;
public RelayCommand RefreshSitesCommand
{
get { return _refreshSitesCommand ?? (_refreshSitesCommand = new RelayCommand(RefreshSites)); }
}
private RelayCommand _refreshProvidersCommand;
public RelayCommand RefreshProvidersCommand
{
get { return _refreshProvidersCommand ?? (_refreshProvidersCommand = new RelayCommand(RefreshProviders)); }
}
private async void RefreshSites()
{
var sitesStats = await _dataService.GetSiteStats();
if (sitesStats != null)
{
SiteStats.Clear();
foreach (var site in sitesStats)
{
SiteStats.Add(new SiteStatsViewModel(site));
}
SelectedSite = SiteStats[0];
}
}
private async void RefreshProviders()
{
var providers = await _dataService.GetProviders();
if (providers != null)
{
Providers.Clear();
foreach (var provider in providers)
{
Providers.Add(new ProviderViewModel(provider));
}
SelectedProvider = Providers[0];
}
}
Then in my code I have the following calls to execute it.
RefreshProvidersCommand.Execute(null);
RefreshSitesCommand.Execute(null);
So why is that better than just calling the RefreshSites and RefreshProviders functions. Then I would not need the code for the RelayCommand objects. Other than exposing the functionality of the 2 private functions, what benefit does using the RelayCommand object have over just making the functions public and calling them.
MVVM is in part about avoiding code-behind in the View class.
If, for example, you want an action to be taken in response to a button click, then you can either assign a Click event handler or assign the Command property to command methods. (Commands have certain advantages over Click event handlers, but that was not the question.)
There is no other good option for handling the Click event other than defining a method in the View class. You cannot directly assign the Click event to a handler method in a different class than the View and you can bind only to properties, not methods.
However, you can assign the Command property to a binding to an object that implements the ICommand interface, e.g. a RelayCommand, and that binding can be to a property of your ViewModel object. This avoids having to define Click event handlers in the view's code behind file and at the same time gives your ViewModel the ability to easily enable/disable commands without needing to know anything about the View's specific implementation.
One can argue about the merits of religiously avoiding code-behind, but that was not the question asked.
Because you can bind to a Command in your view. You can't bind to methods in your views (well you can but binding to Commands is much cleaner)
RelayCommand also implements a CanExecute method which, when binding your RelayCommand to a button, is used to automatically toggle the button's IsEnabled property based on the action you specified for the CanExecute method.