Should an authentication class be static or created from an instance ?
The problem I have is that I cannot access the authentication without injecting an instance of it in any controllers and that a bit problematic.
public class HomeController : Controller
{
//=> I WANT TO AVOID THIS! <=
//protected IAuthentification Authentification { get; private set; }
//public HomeController(IAuthentification authentification)
//{
// Authentification = authentification;
//}
public ActionResult Index()
{
//Authentification.CreerUtilisateur(new Utilisateur { Nom = "Test", NomUsage = "ttt", MotDePasse = "ddsss" });
return View();
}
}
that a bit problematic.
What is the problem? What is 'problematic' ? Constructor injection is worth the effort. If not, are you looking for Property Setter Injection or Setter Method Injection ?
Beware that even if you think they are prettier, the clearest way to manage all of this is simply by using Constructor Injection.
Or is there something I'm missing - you didn't detail your actual problem in the question.
Related
Am learning about data sync from API to WPF app. Got a demo from https://github.com/Azure/azure-mobile-apps/tree/main/samples. But I got into a problem that all the data inside the tables are collected on the call but I need to select specific data using Id. Tried a query etc all came to nothing. Please guide me
Thank you
PatientsController.cs
[Route("tables/Patients")]
public class PatientsController : TableController<Patients>
{
public PatientsController(AppDbContext context)
: base(new EntityTableRepository<Patients>(context))
{
}
}
AppDbContext.cs
public class AppDbContext : DbContext
{
public AppDbContext(DbContextOptions<AppDbContext> options) : base(options)
{
}
public DbSet<Patients> Patients => Set<Patients>();
}
Try to use the following code:
public class AppDbContext : DbContext
{
public AppDbContext(DbContextOptions<AppDbContext> options) : base(options)
{
}
public DbSet<Patients> Patients {get;set;}
}
controller:
[Route("tables/Patients")]
public class PatientsController : TableController<Patients>
{
private readonly AppDbContext _context;
public PatientsController(AppDbContext context)
: base(new EntityTableRepository<Patients>(context))
{
_context=context;
}
public async Task<IActionResult> Index(int id){
var Patients=_context.Patients.FindAsync(id);
return View(Patients);
}
}
If you just need to get a record by Id, you use the URL https://mysite/tables/myTable/id - no search required - it will go directly to the entity you want.
If you need to limit what a user can see, you will need to implement an access control provider (which is an implementation of IAccessControlProvider). This provides three methods to limit what a user can see and do model updates for ensuring the write operations store the right thing. You can read more here: https://learn.microsoft.com/en-us/azure/developer/mobile-apps/azure-mobile-apps/howto/server/dotnet-core#configure-access-permissions
If it's something else, you'll need to be a little more specific. I hang out in the issues and discussions of the azure/azure-mobile-apps repo, so feel free to file a question there.
I am programming an ASP.NET Core 5.0 Web API.
I have a CurrentUser class like this. I want to access the header info anywhere in the project by using CurrentUser.Id. How can I do that? (or where can I initialize httpContext variable?)
public class CurrentUser
{
private static HttpContext _context;
private static UserDto _myUserObj;
public static void Initialize(AuthorizationFilterContext context)
{
_context = context.HttpContext;
_myUserObj = context.HttpContext.Request.Headers["User"] as UserDto;
}
public static int Id()
{
return _myUserObj.Id
}
}
I found some solutions. But I think its not best practice solutions. (There must be another and better way..)
MySolutions is that:
I am building a custom Attribute and i am adding it every API Controller like this:
[MyCustomAttribute]
public class UserController : ControllerBase{
//blablabla
}
After this : I am setting my CurrentUser.Initilaze() method in this Attribute like this:
[AttributeUsage(AttributeTargets.Class | AttributeTargets.Method)]
public class MyCustomAttribute: Attribute, IAuthorizationFilter
{
public void OnAuthorization(AuthorizationFilterContext context)
{
//fortest
var userObj = context.HttpContext.Request.Headers["User"];//working
var userObj2 = JsonSerializer.Deserialize<UserDto>(userObj, new JsonSerializerOptions()
{
PropertyNameCaseInsensitive = true
});//working
CurrentUser.Initialize(context);//working
var testResult = CurrentUser.Test();//working
}
}
So I can use CurrentUser static class everywhere in N-Layerd project and I am accessing UserId or another variables..
But i am saying again, I think its not best solution. Because I have to add [MyCustomAttribute] to all Controllers. There must be a better solition.
I've got 2 different viewmodel which inherits from a base one. there're some services that are resolved currently in the base controller via
ServiceLocator.Default.ResolveType<xxx>
I know its not the best approach since it breaks the DI. I've also read that its an anti-pattern to have more than one constructor. How should I threat that scenario? Since most of the code is the same for both the viewmodels
Thanks
Why don't you inject these services in a constructors? Like this:
public class BaseVM
{
protected readonly IService1 _service1;
public BaseVM(IService1 service1)
{
_service1 = service1
}
}
public class InheritedVM1: BaseVM
{
public InheritedVM1(IService1 service1, ...other args...): base(service1
{
//...
}
}
public class InheritedVM2: BaseVM
{
public InheritedVM2(IService1 service1, ...other args...): base(service1
{
//...
}
}
IoC will inject your services through constructors of InheritedVM, and after calling base with the injected services you can use them using _service1 field.
This question might look naive, but I couldn't understand this code in the ViewModelLocator.cs file:
static ViewModelLocator()
{
ServiceLocator.SetLocatorProvider(() => SimpleIoc.Default);
if (ViewModelBase.IsInDesignModeStatic)
{
SimpleIoc.Default.Register<IDataService, Design.DesignDataService>();
}
else
{
SimpleIoc.Default.Register<IDataService, DataService>();
}
SimpleIoc.Default.Register<MainViewModel>();
}
I see that we use a DataService to get data (from WCF service for example) and assigning it to the MainViewModel. But what if I'm registering more than one ViewModel? like this:
static ViewModelLocator()
{
....
SimpleIoc.Default.Register<MainViewModel>();
SimpleIoc.Default.Register<Page2ViewModel>();
}
and let's say I have another DataService (DataService2 for example), but this one I'll use with the Page2ViewModel. how can I do that?
Also, if someone can help me (or even give me a link to read) about the above code. I have no clue what it means.
You are not assigning any IDataService to the MainViewModel here. You are registering a type mapping, so your container will be aware that it should return a DataService whenever an IDataService required.
This is related to dependency injection http://en.wikipedia.org/wiki/Dependency_injection
The DI container auto-wires the dependencies, so when you need a specific type, you can call
ServiceLocator.Current.GetInstance<IDataService>()
or
ServiceLocator.Current.GetInstance<MainViewModel>()
etc. If it can build it (so you registered your types), it will resolve the full dependency graph for you.
For example, if your MainViewModel has a constructor dependency on IDataService, and you are not in design mode, a DataService will be injected to the MainViewModel constructor. Don't be afraid from the buzzword injected, it is just a call to the MainViewModel constructor with the appropriate parameters :).
So, MainViewModel will not interference with Page2ViewModel here.
I made a simple sample for you to demonstrate what happens (I used Unity, http://unity.codeplex.com/ , but the syntax is almost the same):
class Program
{
static void Main(string[] args)
{
var container = new UnityContainer();
container.RegisterType<IService, Service1>();
container.RegisterType<IService, Service2>("MySpecificService");
container.RegisterType<IRepository, Repository>();
ServiceLocator.SetLocatorProvider(() => new UnityServiceLocator(container));
var viewModel = ServiceLocator.Current.GetInstance<MainViewModel>();
viewModel.Foo();
}
}
interface IService
{
}
interface IRepository
{
}
class Service1 : IService
{
public Service1(IRepository repository)
{
Console.WriteLine("Service1 created");
}
}
class Service2 : IService
{
public Service2()
{
Console.WriteLine("Service2 created");
}
}
class Repository : IRepository
{
public Repository()
{
Console.WriteLine("Repository created");
}
}
class MainViewModel
{
public MainViewModel(IService service)
{
Console.WriteLine("MainViewModel created");
}
public void Foo()
{
var specificService = ServiceLocator.Current.GetInstance<IService>("MySpecificService");
}
}
the output is:
Repository created
Service1 created
MainViewModel created
Service2 created
Because you need a MainViewModel (maybe in SimpleIoC you need to register MainViewModel too, in Unity, it can resolve concrete classes without mapping), the container trying to create one, but it realizes that MainViewModel needs an IService, and it finds the default one from the mapping, which is Service1, but it realizes that Service1 needs an IRepository, and it finds the default one, so it can pass a Repository to the Service1 constructor, then the Service1 instance to the MainViewModel constructor. All the dependencies resolved.
The Foo call is an example how you can register more than one type to the same interface. Dependency injection is a much more bigger topic, but auto-wiring is an important part of it.
I'm not sure the best way to get this accomplished. Here's my view:
public partial class MyPage : Page
{
[Import]
public MyVM ViewModel
{
get { return DataContext as MyVM ; }
set { DataContext = value; }
}
public String EventName { get; set; }
public MyPage()
{
InitializeComponent();
CompositionInitializer.SatisfyImports(this);
}
// Executes when the user navigates to this page.
protected override void OnNavigatedTo(NavigationEventArgs e)
{ }
}
And my VM:
[Export]
public class MyVM : ViewModelBase
{
public MyVM ()
{
}
}
This works great. However, I need to get data from either the viewmodel that has my string, or the URL. Either way, I'm not sure the best way to get the string to MyVW using MEF.
I thought ok I'll use Messaging from MVVMLight, but the MyVM class isn't instantiated yet to receive the broadcast from the other ViewModel. So then I thought well, I'll try this:
[Export]
public class MyVM : ViewModelBase
{
public MyVM ([Import("hello")]string hello)
{
}
}
and then put this in the view:
[Export("hello")]
public String MyHello { get; set; }
but that gave me an error. Cannot call SatisfyImports on a object of type 'Form A' because it is marked with one or more ExportAttributes.
So what's the best way to accomplish this?
To share data between views I usually inject a SharedData object into my ViewModels.
[Import(RequiredCreationPolicy = CreationPolicy.Shared)]
public ISharedData SharedData { get; set; }
I'm also using the Caliburn Micro framework so I'm not passing data around via the URL querystring. By convention CM will parse out URL parameters and inject them into properties on your VM but I'm not sure if this functionality only applies to Windows Phone development.
from here
Examine the Page’s QueryString. Look
for properties on the VM that match
the QueryString parameters and inject
them, performing the necessary type
coercion.
When you say you want to possibly pass data from the view to the vm, that should happen through databinding.