Why Stub objects in JUnit? - Purpose & Use Cases

The Big Idea

What if you could test your code without waiting or risking real-world consequences?

The Scenario

Imagine testing a car's navigation system but you have to drive the whole route yourself every time to check if the system reacts correctly.

Or testing a payment system that requires real money transactions for every test.

The Problem

Manually performing these tests is slow, tiring, and risky.

It's easy to make mistakes or miss cases because you can't repeat exactly the same conditions every time.

Also, relying on real external systems can cause delays and unpredictable failures.

The Solution

Stub objects act like simple, fake versions of real parts your code talks to.

They give back fixed answers instantly, so you can test your code quickly and safely without depending on real systems.

Before vs After

✗ Before

RealPaymentService payment = new RealPaymentService();
payment.process(amount); // actually sends money

✓ After

StubPaymentService payment = new StubPaymentService();
payment.process(amount); // returns fixed success response

What It Enables

Stub objects let you test your code in isolation, making tests fast, reliable, and easy to repeat.

Real Life Example

When testing a shopping app, you use a stub for the payment gateway so you don't spend real money but still check if your app handles payment success or failure correctly.

Key Takeaways

Manual testing with real dependencies is slow and error-prone.

Stub objects simulate parts of the system with fixed, simple behavior.

This makes automated tests faster, safer, and more reliable.

Practice

1. What is the main purpose of a stub object in JUnit testing?

easy

A. To replace a real object with fixed responses for controlled testing

B. To generate random data during test execution

C. To speed up the compilation of test classes

D. To automatically fix bugs found during testing

Solution

1. Step 1: Understand the role of stub objects

   Stub objects are used to replace real parts of a system with simple implementations that return fixed data.

2. Step 2: Identify the purpose in testing

   This allows tests to run in isolation and control the test environment by providing predictable responses.

3. Final Answer:

   To replace a real object with fixed responses for controlled testing -> Option A

4. Quick Check:

   Stub objects = fixed responses for isolation [OK]

Hint: Stub = fixed answers to isolate tests quickly [OK]

Common Mistakes:

• Confusing stubs with mocks or spies
• Thinking stubs generate random data
• Believing stubs fix bugs automatically

2. Which of the following is the correct way to create a simple stub in JUnit using an anonymous class?

easy

A. MyService service = new MyService(); service.getValue() = 5;

B. MyService service = new MyService.getValue(5);

C. MyService service = new MyService() { public int getValue() { return 5; } };

D. MyService service = new MyService() -> 5;

Solution

1. Step 1: Recognize Java anonymous class syntax

   Creating a stub often uses an anonymous class overriding methods to return fixed values.

2. Step 2: Check each option's syntax

   MyService service = new MyService() { public int getValue() { return 5; } }; correctly overrides getValue() to return 5. Options A, C, and D have invalid Java syntax.

3. Final Answer:

   MyService service = new MyService() { public int getValue() { return 5; } }; -> Option C

4. Quick Check:

   Anonymous class override = MyService service = new MyService() { public int getValue() { return 5; } }; [OK]

Hint: Anonymous class overrides method with fixed return [OK]

Common Mistakes:

• Trying to assign a value directly to a method call
• Using incorrect method call syntax
• Confusing lambda syntax with anonymous classes

3. Given the stub below, what will be the output of testService.getValue()?

class TestService {
  int getValue() { return 0; }
}

TestService testService = new TestService() {
  @Override
  int getValue() { return 10; }
};
System.out.println(testService.getValue());

medium

A. Runtime exception

B. 0

C. Compilation error

D. 10

Solution

1. Step 1: Understand method overriding in anonymous class

   The anonymous class overrides getValue() to return 10 instead of 0.

2. Step 2: Determine output of the print statement

   Calling testService.getValue() invokes the overridden method, so it returns 10.

3. Final Answer:

   10 -> Option D

4. Quick Check:

   Overridden method returns 10 [OK]

Hint: Overridden method in stub returns new value [OK]

Common Mistakes:

• Assuming original method runs instead of overridden
• Thinking anonymous class causes compile error
• Expecting runtime exceptions without errors

4. Identify the error in the following stub creation code snippet:

MyService service = new MyService() {
  public int getValue() {
    return "five";
  }
};

medium

A. Cannot override getValue method

B. Method getValue returns wrong data type

C. Missing semicolon after class definition

D. Anonymous class syntax is incorrect

Solution

1. Step 1: Check method return type consistency

   The method getValue() is declared to return int, but returns a String "five".

2. Step 2: Verify syntax correctness

   The anonymous class syntax and semicolon are correct; overriding is allowed.

3. Final Answer:

   Method getValue returns wrong data type -> Option B

4. Quick Check:

   Return type mismatch causes error [OK]

Hint: Return type must match method signature [OK]

Common Mistakes:

• Ignoring return type mismatch
• Assuming syntax error instead of type error
• Thinking overriding is disallowed

5. You want to test a component that depends on a slow external service. Which stub implementation approach is best to speed up tests while ensuring predictable results?

hard

A. Create a stub class that returns fixed, pre-defined data instantly

B. Use the real external service but limit the number of calls

C. Mock the external service to record calls without returning data

D. Skip testing the component until the external service is faster

Solution

1. Step 1: Identify the problem with the real service

   The external service is slow, which slows down tests and makes results unpredictable.

2. Step 2: Choose a stub approach for speed and predictability

   Creating a stub that returns fixed data instantly allows fast, reliable tests without depending on the real service.

3. Step 3: Evaluate other options

   Using the real service or skipping tests delays feedback; mocking without return data doesn't help test results.

4. Final Answer:

   Create a stub class that returns fixed, pre-defined data instantly -> Option A

5. Quick Check:

   Stub fixed data = fast, predictable tests [OK]

Hint: Stub fixed data for fast, reliable tests [OK]

Common Mistakes:

• Using real slow service in tests
• Confusing mocks with stubs
• Skipping tests instead of isolating dependencies