I'm thinking about how to create a main servlet that the rest of the servlets extends from this. I have some properties that I would like to be accesible for the whole servlets in my app and I want to init them. I think it could be something like this:
public abstract class MainServlet extends HttpServlet {
protected String errorURL = null;
protected String adminMenuURL = null;
protected String AdminLoginServlet = null;
protected String ValidateServlet = null;
// here, more properties and methods...
#Override
public void init() {
errorURL = context.getInitParameter("errorURL");
adminMenuURL = context.getInitParameter("adminMenuURL");
AdminLoginServlet = context.getInitParameter("AdminLoginServlet");
ValidateServlet = context.getInitParameter("ValidatePicsServlet");
// here, some more inits...
}
}
When I create a new Servlet like the following...
public class AdminLoginServlet extends MainServlet {
}
If I forward to AdminLoginServlet, would the parameters (errorURL, adminMenuURL, etc.) be assigned again?
If i override the init method in AdminLoginServlet (and others servlets that extends from MainServlet)... these properties won't be never assigned, isn't it?
How would you do what I pretend to do?
Thanks.
the members adminMenuUrl and so on are assigned when the servlet is first created by the engine, because this is the moment when init() is invoked. Every user request is handled by the same servlet instance, in different threads. When you forward a request to a servlet, the method init() won't be invoked.
you're right, because there can be no instance of MainServlet
I'm not sure how your web app should look like
You should rename the fields AdminLoginServlet and ValidateServlet, because it's bad practice to start fieldnames with upper case letters.
Related
I'm having trouble injecting the dependency I pass into the constructor of my Asegurador class.
When I want to instantiate, _instance = new Asegurador(); I don't have the parameter required by the constructor (IGeneralRepository), how can I solve this problem?
Note that my Asegurador class is a singleton.
private Asegurador(IGeneralRepository generalRepository)
{
_token = GetTokenAsync().Result;
_repository = generalRepository;
}
public static Asegurador Instance
{
get
{
if (_instance == null)
{
_local = System.Environment.GetEnvironmentVariable("SEGUROS_LOCAL") ?? "local";
_instance = new Asegurador();
}
return _instance;
}
}
When using a DI container you can (and should) let it take care of handling the Lifetime of a dependency.
.Net core's dependency injection lets you define 3 different lifetimes for your services (Docs):
Transient: a transient service is recreated each time it is injected
Scoped: a scoped service is created once for each request
Singleton: a singleton is created once in the whole application lifetime.
The best approach to achieve what you are trying to do is the following:
Amend your Asegurador class so that it has a public constructor and get rid of the static Instance property
public class Asegurador {
public Asegurador(IGeneralRepository generalRepository)
{
_token = GetTokenAsync().Result; //I know too few about it but I would try to pass it as a dependency as well
_repository = generalRepository;
}
}
instead of calling Asegurador.Instance inject the dependency in the client class
public class IUseTheAsegurador {
private Asegurador _asegurador;
public IUseTheAsegurador(Asegurador asegurador)
{
_asegurador = asegurador;
}
}
Register all in the DI in your Startup.cs:
public void ConfigureServices(IServiceCollection services)
{
...
services.AddSingleton<Asegurador>();
services.AddScoped<IUseAsegurador>(); //This can be Singleton or Transient as well, depending on your needs
...
}
I (a lot of people actually :D) prefer this approach because it leaves the responsability of guaranteeing a single instance to the DI and also because lets you write unit tests for the client class (IUseTheAsegurador in the example) in an easier way.
My theme has some sort of breadcrumb. The controller is always the category. To avoid repeat myself, I want to set it in the constructor of the controller for all actions like this:
class MyController:Controller{
public MyController() {
ViewBag.BreadcrumbCategory = "MyCategory";
}
}
When I access ViewBag.BreadcrumbCategory in the layout-view, its null. In a Action it works:
class MyController:Controller{
public IActionResult DoSomething() {
ViewBag.BreadcrumbCategory = "MyCategory";
}
}
I'm wondering that setting a ViewBag property is not possible in a constructor? It would be annoying and no good practice to have a function called on every action which do this work. In another question using the constructor was an accepted answear, but as I said this doesn't work, at least for ASP.NET Core.
There is an GitHub issue about it and it's stated that this is by design. The answer you linked is about ASP.NET MVC3, the old legacy ASP.NET stack.
ASP.NET Core is written from scratch and uses different concepts, designed for both portability (multiple platforms) as well as for performance and modern practices like built-in support for Dependency Injection.
The last one makes it impossible to set ViewBag in the constructor, because certain properties of the Constructor base class must be injected via Property Injection as you may have noticed that you don't have to pass these dependencies in your derived controllers.
This means, when the Controller's constructor is called, the properties for HttpContext, ControllerContext etc. are not set. They are only set after the constructor is called and there is a valid instance/reference to this object.
And as pointed in the GitHub issues, it won't be fixed because this is by design.
As you can see here, ViewBag has a dependency on ViewData and ViewData is populated after the controller is initialized. If you call ViewBag.Something = "something", then you it will create a new instance of the DynamicViewData class, which will be replaced by the one after the constructor gets initialized.
As #SLaks pointed out, you can use an action filter which you configure per controller.
The following example assumes that you always derive your controllers from Controller base class.
public class BreadCrumbAttribute : IActionFilter
{
private readonly string _name;
public BreadCrumbAttribute(string name)
{
_name = name;
}
public void OnActionExecuting(ActionExecutingContext context)
{
base.OnActionExecuting(context);
var controller = context.Controller as Controller;
if (controller != null)
{
controller.ViewBag.BreadcrumbCategory = _name;
}
}
}
Now you should be able to decorate your controller with it.
[BreadCrumb("MyCategory")]
class MyController:Controller
{
}
I have the same issue and solve it overriding the OnActionExecuted method of the controller:
public override void OnActionExecuted(ActionExecutedContext context)
{
base.OnActionExecuted(context);
ViewBag.Module = "Production";
}
Here is a better way to do this for .NET Core 3.x, use the ResultFilterAttribute:
Create your own custom filter attribute that inherits from ResultFilterAttribute as shown below:
public class PopulateViewBagAttribute : ResultFilterAttribute
{
public PopulateViewBagAttribute()
{
}
public override void OnResultExecuting(ResultExecutingContext context)
{
// context.HttpContext.Response.Headers.Add(_name, new string[] { _value });
(context.Controller as MyController).SetViewBagItems();
base.OnResultExecuting(context);
}
}
You'll need to implement the method SetViewBagItems to populate your ViewBag
public void SetViewBagItems()
{
ViewBag.Orders = Orders;
}
Then Decorate your Controller class with the new attribute:
[PopulateViewBag]
public class ShippingManifestController : Controller
That's all there is to it! If you are populating ViewBags all over the place from your constructor, then you may consider creating a controller base class with the abstract method SetViewBagItems. Then you only need one ResultFilterAttribute class to do all the work.
In order to retrieve a list in a Spring MVC application I would like to write something like:
public String myMethod(#RequestParam("foo") List<FooUi> foos)
But the only solution I've found so far is the following :
public String myMethod(FooListWrapperUi fooListWrapperUi)
I don't like this solution because I have to write a wrapper each time I need to retrieve a list. In this example, the wrapper is the following :
#Data
#JsonIgnoreProperties(ignoreUnknown = true)
public class FooListWrapperUi
{
private ArrayList<FooUi> fooList;
}
So my question is, is it possible to use something like the first solution or is it impossible and I need to write a wrapper?
Thanks.
You can accommodate your use case by creating your own HandlerMethodArgumentResolver:
public class FooUiResolver implements HandlerMethodArgumentResolver {
#Override
public boolean supportsParameter(MethodParameter methodParameter) {
return (methodParameter.getParameterType().equals(FooUi.class) ||
(methodParameter instanceof Collection<?> && ((ParameterizedType) methodParameter.getParameterType().getGenericSuperclass()).getActualTypeArguments()[0] == FooUi.class));
}
#Override
public Object resolveArgument(MethodParameter methodParameter,
ModelAndViewContainer modelAndViewContainer, NativeWebRequest nativeWebRequest,
WebDataBinderFactory webDataBinderFactory) throws Exception {
// Create instances of FooUi by accessing requests parameters in nativeWebRequest.getParameterMap()
}
}
The actual implementation will depend on how you would create one or more FooUi instances from the request parameters or body. You then need to register FooUiResolver in your servlet config:
#Override
public void addArgumentResolvers(List<HandlerMethodArgumentResolver> argumentResolvers){
argumentResolvers.add(new FooUiResolver());
super.addArgumentResolvers(argumentResolvers);
}
Once registered, you can use FooUi in your controller method arguments without RequestParam or any other annotation:
#RequestMapping(value = "/foo")
public String myMethod(List<FooUi> foos){}
I've got an abstract class shown below which gets inherited by all the other controllers. Is it possible to test this method at all? Btw, I'm trying to use MOQ but no luck. If you could help me will be much appreciated:
public abstract class ApplicationController : Controller
{
protected override void Initialize(System.Web.Routing.RequestContext requestContext)
{
base.Initialize(requestContext);
//do some stuff here
}
}
If you take a look at the source code of base Initialize method you will find out that what it does is that it sets up ControllerContext and url stuff. Now, download MvcContrib TestHelper and check out TestControllerBuilder . The builder sets up everything you need in order to have controller context and other stuff which you depend upon.
Ok, we are not over yet - you wanted to test your own override of Initialize right?
TestControllerBuilder doesnt call your Initialize because it does initialization in different way. I suggest you to factor out your custom Initialize() logic out into different method. Then create fake (stub) subclass with public method that calls this factored out protected Initialize. Are you with me?
something like:
public abstract class ApplicationController : Controller
{
protected override void Initialize(System.Web.Routing.RequestContext requestContext)
{
base.Initialize(requestContext);
MyInitialzie()
}
protected void MyInitialize()
{
ControllerContext.XXX // do whatewer you want here. Context is already setted up
}
}
class FakeController: ApplicationController
{
public void CallMyInitialize()
{
MyInitialize();
}
}
Later in test class:
[Test]
public void MyInitializeTest()
{
TestControllerBuilder builder = new TestControllerBuilder();
FakeController controller = new FakeController();
builder.InitializeController(controller);
controller.CallMyInitialize();
//TODO: verification of MyInitialize assumptions
}
Is that clear?
I have a java class which has a static member created using Facade (Singleton).
Class A implements InterfaceA {
private static DataStore db = DataStoreFacade.getInstance("BDB"); //singleton instance
public void save(final String key, final String val) {
db.save(key,val);
}
};
Here Class A is used as a member variable for webservice (stateless bean).
I can't test this code using EasyMock because there is no way to override the DataStore instance.
There are two options.
Have a constructor taking the instance of DataStore which will set to db member variable. The problem is I don't want webservice class to know which datastore instance has been created.
Provide an additional protected Set Method to override the db object. This is what I have used where I create a Easy Mock object of DataStore and override the member variable. Is it the correct design.
What are the other possibilities?
You're right that is bad for testability. Use dependency injection and don't go for static variable:
public class A implements InterfaceA {
private DataStore db;
public A(DataStore db) {
this.db = db;
}
...
}
to inject or build either use dependency-injection framework (e.g. spring) or build the object somewhere in bootstrap factory code yourself.
production code:
new A(DataStoreFacade.getInstance("...");
test-code:
public void test_xxx(){
DataStore db = EasyMock.createMock(DataStore.class);
//... do some expectations and replay(db)
InterfaceA a=new A(db);
//...
}
Well, the original code is already testable. Here is a unit test for it, using JMockit:
#Test
public void testSave(final DataStore mockDb)
{
final String key = "aKey";
final String value = "aValue";
new A().save(aKey, aValue);
new Verifications()
{{
mockDb.save(key, value);
}};
}
If needed, the DataStoreFacade class could be mocked too.
Why not make the db member protected, and in your test project inherit it and override that member:
project
{
Class A
{
protected static db = ...
public void Save(...) { ... }
}
}
test_project
{
Class B : A
{
protected override static db = ... (create test db)
}
Class testB
{
public A a;
public void Setup()
{
this.a = new B();
}
public void TearDown()
{
// delete a
}
public void TestSaveKey()
{
// test a
}
}
}
It's still hidden from consumers of the code/library, the test object isn't cluttering the production code, and the behaviour will be tested as if it were the production version.
Beware though, that having a static member for your db object could cause troubles for your tests if it's not cleaned up properly after each test.*
I know that you probably already know this, but I'm saying it for completeness.
Use the Supersede Instance pattern...
http://goodcoffeegoodcode.blogspot.com/2010/01/supercede-instance-pattern.html