Contract testing basics in Testing Fundamentals - Build an Automation Script

Verify API contract between consumer and provider

Preconditions (3)

Step 1: Run the consumer test to generate the contract file

Step 2: Publish the contract file to the contract broker or share with provider

Step 3: Run the provider verification test using the contract file

Step 4: Check that the provider response matches the contract expectations

✅ Expected Result: Provider verification test passes confirming the API contract is fulfilled

Automation Requirements - Pact JVM (Java) or Pact JS (JavaScript)

Assertions Needed:

Verify that provider response status code matches contract

Verify that provider response body matches contract schema and values

Verify that provider headers match contract

Best Practices:

Use explicit contract files shared between consumer and provider

Automate contract verification as part of provider CI pipeline

Use clear and stable request/response examples in contract

Fail tests if provider does not meet contract exactly

Automated Solution

Testing Fundamentals

import au.com.dius.pact.provider.junit5.PactVerificationContext;
import au.com.dius.pact.provider.junit5.PactVerificationInvocationContextProvider;
import au.com.dius.pact.provider.junit5.Provider;
import au.com.dius.pact.provider.junit5.State;
import org.junit.jupiter.api.BeforeEach;
import org.junit.jupiter.api.TestTemplate;
import org.junit.jupiter.api.extension.ExtendWith;

@Provider("UserService")
public class UserServiceProviderTest {

  @BeforeEach
  void before(PactVerificationContext context) {
    context.setTarget(new HttpTestTarget("localhost", 8080));
  }

  @TestTemplate
  @ExtendWith(PactVerificationInvocationContextProvider.class)
  void pactVerificationTestTemplate(PactVerificationContext context) {
    context.verifyInteraction();
  }

  @State("User exists")
  public void userExists() {
    // Setup provider state: ensure user data exists in database
  }

  @State("User does not exist")
  public void userDoesNotExist() {
    // Setup provider state: ensure user data does not exist
  }

}

This code uses the Pact JVM framework to verify the provider API against the contract.

The @Provider annotation names the provider service.

The @BeforeEach method sets the target URL where the provider API runs.

The @TestTemplate method runs all interactions defined in the contract file.

The @State methods set up provider data states required by the contract scenarios.

This structure ensures the provider API responses match the consumer expectations defined in the contract.

Common Mistakes - 3 Pitfalls

Not setting up provider states before verification

Running provider verification without the correct contract file

Ignoring header verification in contract tests

Bonus Challenge

Now add data-driven contract tests for multiple user IDs with different expected responses

Show Hint

Practice

1. What is the main purpose of contract testing in software development?

easy

A. To ensure two systems agree on the format and data of their communication

B. To test the user interface for visual bugs

C. To check the performance speed of an application

D. To verify database schema changes

Solution

1. Step 1: Understand contract testing definition

   Contract testing focuses on the agreement between two systems on how they communicate, including request and response formats.

2. Step 2: Compare options with contract testing purpose

   Only To ensure two systems agree on the format and data of their communication describes this agreement on communication format and data, which is the core of contract testing.

3. Final Answer:

   To ensure two systems agree on the format and data of their communication -> Option A

4. Quick Check:

   Contract testing = communication agreement [OK]

Hint: Contract testing checks communication agreements between systems [OK]

Common Mistakes:

• Confusing contract testing with UI testing
• Thinking contract testing measures performance
• Assuming contract testing checks database structure

2. Which of the following is the correct way to describe a contract in contract testing?

easy

A. A set of UI elements and their styles

B. A list of database tables and columns

C. A defined request and expected response between services

D. A performance benchmark for API calls

Solution

1. Step 1: Identify what a contract defines

   A contract in contract testing defines the request sent and the expected response received between two services.

2. Step 2: Match options to contract definition

   A defined request and expected response between services correctly describes this request-response definition, while others describe unrelated concepts.

3. Final Answer:

   A defined request and expected response between services -> Option C

4. Quick Check:

   Contract = request + expected response [OK]

Hint: Contracts define requests and expected responses [OK]

Common Mistakes:

• Mixing contracts with UI or database details
• Thinking contracts are about performance
• Confusing contracts with test cases

3. Given a contract test that expects a JSON response with a field status equal to "success", what will happen if the actual response is {"status": "fail"}?

medium

A. The test will pass because the response is JSON

B. The test will fail because the status value does not match

C. The test will pass because the field status exists

D. The test will fail due to a syntax error in the response

Solution

1. Step 1: Understand the contract expectation

   The contract expects the status field to have the value "success" exactly.

2. Step 2: Compare actual response to expected

   The actual response has "status": "fail", which does not match the expected value, causing the test to fail.

3. Final Answer:

   The test will fail because the status value does not match -> Option B

4. Quick Check:

   Expected value mismatch = test fail [OK]

Hint: Exact value mismatch causes contract test failure [OK]

Common Mistakes:

• Assuming presence of field is enough to pass
• Ignoring value differences in contract tests
• Confusing syntax errors with value mismatches

4. You wrote a contract test but it always passes even when the provider sends incorrect data. What is the most likely mistake?

medium

A. The contract test has syntax errors and does not run

B. The contract test is missing the request URL

C. The contract test is running on the wrong environment

D. The contract test is not validating the response content properly

Solution

1. Step 1: Analyze why test passes despite incorrect data

   If the test passes even with wrong data, it likely means the test is not checking the response content correctly.

2. Step 2: Evaluate other options

   Missing URL or wrong environment would cause errors or failures, and syntax errors prevent running, so these are less likely.

3. Final Answer:

   The contract test is not validating the response content properly -> Option D

4. Quick Check:

   Incorrect validation = false pass [OK]

Hint: Check if test actually validates response content [OK]

Common Mistakes:

• Ignoring validation logic in tests
• Assuming environment issues cause false passes
• Confusing syntax errors with test logic errors

5. You have two microservices: Service A (consumer) and Service B (provider). Service A expects Service B to return a JSON with a userId as a number and active as a boolean. Which contract test scenario best ensures compatibility?

hard

A. Test that Service B returns {"userId": 123, "active": true} and reject if types differ

B. Test that Service B returns any JSON object with keys userId and active regardless of types

C. Test that Service B returns a string containing userId and active anywhere

D. Test that Service B returns a JSON with userId as string and active as number

Solution

1. Step 1: Identify expected data types in contract

   Service A expects userId as a number and active as a boolean, so the contract must enforce these types.

2. Step 2: Evaluate options for strictness and correctness

   Test that Service B returns {"userId": 123, "active": true} and reject if types differ tests exact types and values, ensuring compatibility. Options A, B, and C ignore or mismatch types, risking integration errors.

3. Final Answer:

   Test that Service B returns {"userId": 123, "active": true} and reject if types differ -> Option A

4. Quick Check:

   Strict type check = compatibility ensured [OK]

Hint: Contract tests must check exact data types and values [OK]

Common Mistakes:

• Ignoring data types in contract tests
• Accepting any JSON structure without validation
• Confusing string and number types in contracts