I just recently read about "Mocking objects" for unit testing and currently I'm having a difficulties implementing this approach in my application. Please let me explain my problem.
I have a User model class, which is dependent on 2 data sources (database and facebook web service). The controller class simply use this User model as an interface to access data and it doesn't care about where the data came from.
Currently I never done any unit test to this User model because it is dependent on an external web service. But just a while ago, I read about object mocking and now I know that it is a common approach to unit test a class that depends on external resources (like in my case).
Now I want to create a unit test for the User model, but then I encountered a design issue:
In order for the User model to use a mocked Facebook SDK, I have to inject this mocked Facebook SDK to the User object (probably using a setter). Therefore I can't construct the Facebook SDK inside the User object. I have to construct it outside the User object, and inject the SDK into the User object.
The real client of my User model is the application's controller. Therefore I have to construct the Facebook SDK inside the controller and inject it to the user object. Well, this is a problem because I want my controller to be as clean as possible. I want my controller to be ignorant about the application's data source.
I'm not good at explaining something systematically, so you'll probably sleeping before reading this last paragraph. But anyway, I want to ask if anyone here ever encountered the same problem as mine? How do you solve this problem?
Regards,
Andree
P.S: I'm using Zend framework, PHP 5.3.
One way to solve this problem is to create two constructors in User: the default one instantiates the real-life data sources, the other one gets them as parameters. This way your controller can use the default constructor, while your tests use the parameterized one to pass in mock data sources.
Since you haven't specified your language, I show you an example in Java, hopefully this helps get the idea:
class User {
private DataBase database;
private WebService webService;
// default constructor
public User() {
database = new OracleDataBase();
webService = new FacebookWebService();
}
// constructor for unit testing
public User(DataBase database, WebService webService) {
this.database = database;
this.webService = webService;
}
}
This isn't really a question about mocking, but about dependencies--and trying to unit test has forced the issue. It sounds like currently you create your User object within your Controller.
If the User and Controller have the same lifetime (they're created at the same time), then you can pass the User into the Controller's constructor, which is where you can make the substitution.
If there is a User object per call, then perhaps the User object should be passed in by the environment, or returned from some Context object.
If you make the Facebook SDK Object(s) publicly accessible, your User object can still create it when it sets up but you can replace it with a mock before you actually do anything on the User object.
[Test]
public void Test()
{
User u = new User(); // let's say that the object on User gets created in the ctor
u.FacebookObj = new DynamicMock(typeof(FacebookSDK)).MockInstance;
Assert.That(u.Method(), Does.Stuff, "u.Method didn't do stuff");
}
You don't say what language you are using, but I use Ruby and Mocha for mocking objects, and it's quite easy.
See http://mocha.rubyforge.org/.
here the fourth example shows that any call to Product.name from the unit under test is intercepted and the value 'stubbed_name' is returned.
I guess there are similar mechanisms in Java, etc.
Since you will be Unit Testing your test class will play the role of the Controller, so that one should not be touched. So that can remain clean.
You are well underway of reinventing Dependency Injection. So it may be worthwhile to look at Spring or Guice to help you with the plumbing. (If you are in Java land).
Your test can indeed do the injection of the Mocked Facebook SDK and your Database service using setter or constructor arguments.
your tests will now do what the controllers (and maybe views) will do and verify the proper routines are called in the SDK's and that the resources are properly obtained and disposed of.
Related
Im trying to get familiar with using MOQ and mocking in general. So I want to test an api controller which uses an assembly which serves as a repository for getting/updating data etc.
eg a structure like this.
HomeController
Index
Repository.GetSomeData (returns JSON object)
This repository class has an interface, and that is what is injected via the .net core startup class. The method in this case, GetSomeData does a number of steps via calls to the Db, as well as reading a file from the file system, parsing it and moving it off to another folder.
Question: How can a "mocked" Repository work without doing the things that the "real" object does? All the examples I see are simple addition, returning strings etc.
When you mock something like your repository, you going to stub out methods on the repository to return some canned result. Calls to those methods on the repository mock, then, bypass the real methods and instead just do what you've stubbed.
Essentially, you need to first identity what methods will be utilized. Then, you should determined appropriate responses those methods should return based on the particular scenario you're attempting to unit test. Then, you create the mock and add stubs for those methods with those responses.
The whole point of mocking is to remove variables, so you're intentionally trying to get to the "happy path": the set of internal responses that put the action in the state you need it to be in for specific test you're conducting.
I am connecting to a REST service. It's QuickBlox, to be specific, but it should not matter other than that the REST API is defined like a SQL query (but notably minus the joins) and I can do CRUD on it.
QuickBlox provides iOS SDK, which gives me an interface to construct the REST query with a callback.
In my program I want objects like User, UserResourceTable, Card, for example. (It's not exactly like this as User resides in a special QuickBlox module, but for the sake of the question it's OK to ignore this.)
I have constructed something I think is similar to the ActiveRecord pattern (I hope). So in my User class I have CRUD methods, and Card class also. And these CRUD have callbacks (or rather, Objective-C blocks which are closures).
Originally I have "read" as class method (like a Java static method), since before calling read there is nothing about the object that's read so I thought I would provide a static factory method on the aforementioned classes.
UserResourceTable is not directly expose to whoever uses the User class. It's created when User is created and read when User is read.
Now this all went OK until I started to think about unit-testing. I end up deciding to do dependency injection (DI), then I want UserResourceTable to be injected into User. But it now seems difficult to do DI while read is a class method/static method and I now think I want read to be an instance method. (Especially there is no static variable in Objective-C and even if I can use static variable in the file to kind of use it I wouldn't know when to inject it. It just smell bad to me.)
Now read would be something like this (pseudo code):
User user = User(); // user just so we can call read()
User userFetched = user.read();
A few questions:
Is it a good idea to have read as an instance method?
Is ActiveRecord suitable for my situation? Perhaps a better model?
Right now the User, UserResourceTable, and Card all directly uses QuickBlox sdk (or rather, via a thin wrapper). And I intend to inject UserResourceTable into User. I also like to inject Quickblox's interface (or the thin wrapper) into User, UserResourceTable, and Card. I would have some factories to do this. Does that sound any good?
Quick terminology question that's somewhat related to my main question: What is the correct term for a model class and the term for a instance of that class?
I am learning Test Driven Development, and want to make sure I am learning it the right so I can form good habits.
My current project has a SalesmanController, which is pretty much a basic resource controller. Here's my current issue (I can get it working, but I want to make sure its done as "right" as possible)
I have a 'Salesman' model.
The 'Salesman' is mapped as having many 'Sales' using my ORM.
The 'Sales' model is mapped as belongsTo 'Salesman' using my ORM.
I have created a ORMSalesmanRepository which implements the SalesmanRepositoryInterface.
My controller has SalesmanRepositoryInterface passed to it upon construction(constructor dependency injection).
My controller calls the find method on the SalesmanRepositoryInterface implementation it has been given to find the correct salesman.
My view needs information on the salesman and information on all 'Sales' records that belong to him.
The current implementation of SalesmanRepositoryInterface returns an instance of a ORMRecord, and then passes that to the view, which retrieves the sales from the ORMRecord.
My gut tells me this implementation is wrong. The orm record implements Array Access, so it still behaves like an array as far as the view knows.
However, when trying to go back and implement my unit tests I am running into issues mocking my dependencies. (I now know with TDD I'm supposed to make my unit tests and then develop the actual implementation, didn't figure this out till recently).
Salesman = MockedSalesman;
SalesRecords = MockedSalesman->Sales;
Is it poor programming to expect my Salesman to return a ORMObject for the controller to use (for chaining relationships maybe?) or is a controller becoming to 'fat' if I'm allowing it to call more than just basic get methods (using arrayaccess []) on the ORMObject? Should the controller just assume that whatever it gets back is an array (or at least acts like one?)
Also, should it ever come up where something one of my mocked classes returns needs to be mocked again?
Thanks in advance everybody.
What is the correct term for a model class and the term for a instance of that class?
Depends on what you mean with "model classes"? Technically model is a layer, that contains several groups of classes. Most notable ones would be: mappers, services an domain objects. Domain objects as whole are implementation of accumulated knowledge about business requirements, insight from specialists and project goals. This knowledge is referred to as "domain model".
Basically, there is no such thing as "model class". There are classes that are part of model.
What is the controller allowed to assume about what it recieves from a service?
Nothing, because controller should not receive anything from model layer. The responsibility of controller is to alter the state of model layer (and in rare cases - state of current view).
Controller is NOT RESPONSIBLE for:
gather data from model layer,
initializing views
passing data from model layer to views
dealing with authorization checks
The current implementation of SalesmanRepositoryInterface returns an instance of a ORMRecord, and then passes that to the view, which retrieves the sales from the ORMRecord.
It sounds like you are implementing active record pattern. It has very limited use-case, where is is appropriate to use AR - when object mostly consists of getters and setters tht are directly stored in a single table. For anything beyond that active record becomes an anti-pattern because it violates SRP and you loose the ability to test your domain logic without database.
Also, are repository should be returning an instance of domain object and makes sure that you are not retrieving data repeatedly. Repositories are not factories for active record instances.
Is it poor programming to expect my Salesman to return a ORMObject for the controller to use (for chaining relationships maybe?) or is a controller becoming to 'fat' if I'm allowing it to call more than just basic get methods (using arrayaccess []) on the ORMObject?
Yes, it's bad code. Your application logic (one that would usually be contained in services) is leaking in the presentation layer.
At this stage I wouldn't stress too much about your implementation - if you try and write proper tests, you'll quickly find out what works and what doesn't.
The trick is to think hard about what each component is trying to achieve. What is your controller method supposed to do? Most likely it is intended to create a ViewModel of some kind, and then choose which View to render. So there's a few tests right there:
When I call my controller method with given arguments (ShowSalesmanDetail(5))
It should pick the correct View to render ('ShowSalesmanDetail')
It should construct the ViewModel that I expect (A Salesman object with some Sales)
In theory, at this point you don't care how the controller constructs the model, only that it does. In practice though you do need to care, because the controller has dependencies (the big one being the database), which you need to cater for. You've chosen to abstract this with a Repository class that talks to an ORM, but that shouldn't impact the purpose of your tests (though it will definitely alter how you implement those tests).
Ideally the Salesman object in your example would be a regular class, with no dependencies of its own. This way, your repository can construct a Salesman object by populating it from the database/ORM, and your unit tests can also use Salesman objects that you've populated yourself with test data. You shouldn't need to mock your models or data classes.
My personal preference is that you don't take your 'data entities' (what you get back from your ORM) and put them into Views. I would construct a ViewModel class that is tied to one View, and then map from your data entities to your ViewModel. In your example, you might have a Salesman class which represents the data in the database, then a SalesmanModel which represents the information you are displaying on the page (which is usually a subset of what's in the db).
So you might end up with a unit test looking something like this:
public void CallingShowSalesmanShouldReturnAValidModel()
{
ISalesmanRepository repository = A.Fake<ISalesmanRepository>();
SalesmanController controller = new SalesmanController(repository);
const int salesmanId = 5;
Salesman salesman = new Salesman
{
Id = salesmanId,
Name = "Joe Bloggs",
Address = "123 Sesame Street",
Sales = new[]
{
new Sale { OrderId = 123, SaleDate = DateTime.Now.AddMonths(-1) }
}
};
A.CallTo(() => repository.Find(salesmanId)).Returns(salesman);
ViewResult result = controller.ShowSalesman(salesmanId) as ViewResult;
SalesmanModel model = result.Model as SalesmanModel;
Assert.AreEqual(salesman.Id, model.Id);
Assert.AreEqual(salesman.Name, model.Name);
SaleModel saleModel = model.Sales.First();
Assert.AreEqual(salesman.Sales.First().OrderId, saleModel.OrderId);
}
This test is by no means ideal but hopefully gives you an idea of the structure. For reference, the A.Fake<> and A.CallTo() stuff is from FakeItEasy, you could replace that with your mocking framework of choice.
If you were doing proper TDD, you wouldn't have started with the Repository - you'd have written your Controller tests, probably got them passing, and then realised that having all this ORM/DB code in the controller is a bad thing, and refactored it out. The same approach should be taken for the Repository itself and so on down the dependency chain, until you run out of things that you can mock (the ORM layer most likely).
I'm trying to write unit tests for parts of my Node app. I'm using Mongoose for my ORM.
I've searched a bunch for how to do testing with Mongoose and Node but not come with anything. The solutions/frameworks all seem to be full-stack or make no mention of mocking stuff.
Is there a way I can mock my Mongoose DB so I can return static data in my tests? I'd rather not have to set up a test DB and fill it with data for every unit test.
Has anyone else encountered this?
I too went looking for answers, and ended up here. This is what I did:
I started off using mockery to mock out the module that my models were in. An then creating my own mock module with each model hanging off it as a property. These properties wrapped the real models (so that child properties exist for the code under test). And then I override the methods I want to manipulate for the test like save. This had the advantage of mockery being able to undo the mocking.
but...
I don't really care enough about undoing the mocking to write wrapper properties for every model. So now I just require my module and override the functions I want to manipulate. I will probably run tests in separate processes if it becomes an issue.
In the arrange part of my tests:
// mock out database saves
var db = require("../../schema");
db.Model1.prototype.save = function(callback) {
console.log("in the mock");
callback();
};
db.Model2.prototype.save = function(callback) {
console.log("in the mock");
callback("mock staged an error for testing purposes");
};
I solved this by structuring my code a little. I'm keeping all my mongoose-related stuff in separate classes with APIs like "save", "find", "delete" and no other class does direct access to the database. Then I simply mock those in tests that rely on data.
I did something similar with the actual objects that are returned. For every model I have in mongoose, I have a corresponding class that wraps it and provides access-methods to fields. Those are also easily mocked.
Also worth mentioning:
mockgoose - In-memory DB that mocks Mongoose, for testing purposes.
monckoose - Similar, but takes a different approach (Implements a fake driver). Monckoose seems to be unpublished as of March 2015.
I am writing a spock unit test that tests a controller method.
The controller action under test instantiates a new domain instance object and
calls validate on it before it saves.
Is there anyway of mocking the call to domainInstance.validate() so I
can make it return whatever I want? Or do I have to hide this
instanciation and saving behind a service method to achieve this?
I do this this way, because within the context of a unit test for a
controller, the constraints of a domain object should not be involved.
I test those elsewhere (in the MyDomainClassTests, obviously). If I
wanted to take those into into account my test would be an integration
test.
If you didn't place the validate on the domain instance itself, but rather in a service,
you could let your controller take a Service in its constructor (or rather an interface of a service). Let that service handle the validation.
Now for your unittest of that controller, you would pass in a Mock of that interface(service) to the controller and configure the mock to return whatever you want.
For .net i can recommend Moq (http://code.google.com/p/moq/)
After a while I have come to the conclusion that what I wanted is rather tricky. If you're in a scenario where you don't have to use mockDomain(), you could add a groovy metaclass method and it's implementation (return true or false, whichever you want)
If you do need mockDomain() because you need to mock pre-existing instances you are out of options, at least for now because mockDomain() and fiddling with metaclass methods that mockDomain actually provides will not mix.