Integration testing in Rest API - Step-by-Step Execution

Concept Flow - Integration testing

Write individual modules

↓

Develop integration tests

↓

Run integration tests

↓

Check if modules work together

↓

Deploy

Integration testing checks if different parts of a program work well together by running tests that combine modules.

Execution Sample

Rest API

POST /login {"user":"alice","pass":"123"}
GET /profile (with token)
Assert profile data correct

This test sends a login request, then uses the returned token to get the profile, checking if the data matches expected.

Execution Table

Step	Action	Request Sent	Response Received	Check	Result
1	Send login request	POST /login {"user":"alice","pass":"123"}	{"token":"abc123"}	Token received?	Pass
2	Send profile request with token	GET /profile Authorization: Bearer abc123	{"name":"Alice","email":"[e-mail]"}	Profile data correct?	Pass
3	Integration test result	-	-	All steps passed?	Pass

💡 All integration steps passed, modules work together correctly.

Variable Tracker

Variable	Start	After Step 1	After Step 2	Final
token	None	"abc123"	"abc123"	"abc123"
profile_data	None	None	{"name":"Alice","email":"[e-mail]"}	{"name":"Alice","email":"[e-mail]"}

Key Moments - 2 Insights

Why do we need to use the token from the login response in the profile request?

Because the profile endpoint requires authentication, the token proves the user is logged in. See execution_table step 2 where the token is sent.

What happens if the profile data does not match expected?

The integration test fails at step 2, indicating modules do not work together properly and need fixing before deployment.

Visual Quiz - 3 Questions

Look at the execution table, what is the token value after step 1?

A "xyz789"

B "abc123"

C None

D "token"

💡 Hint

Check the 'Response Received' column in step 1 of the execution_table.

At which step does the test check if profile data is correct?

A Step 2

B Step 1

C Step 3

D No step checks profile data

💡 Hint

Look at the 'Check' column in the execution_table.

If the login response did not return a token, what would happen in the execution table?

A Step 3 would say Pass

B Step 2 would pass anyway

C Step 1 result would be Fail

D No change in the table

💡 Hint

Refer to the 'Result' column in step 1 of the execution_table.

Concept Snapshot

Integration testing combines multiple modules to test them working together.
Write tests that send real requests between modules.
Check responses and data flow between parts.
Fail tests if modules don't cooperate correctly.
Fix issues before deploying the whole system.

Full Transcript

Integration testing is the process of checking if different parts of a program work well together. First, individual modules are written. Then, integration tests are developed that send requests combining these modules. For example, a login request is sent, and the returned token is used to request a user profile. The test checks if the profile data matches what is expected. If all steps pass, the modules work together correctly and the system can be deployed. If any step fails, the problem must be fixed and tests rerun. This ensures the whole system functions as intended.

Practice

1. What is the main goal of integration testing in REST API development?

easy

A. To measure the speed of the server hardware

B. To test the user interface design

C. To check if different parts of the application work together correctly

D. To write documentation for the API

Solution

1. Step 1: Understand integration testing purpose

   Integration testing focuses on verifying that multiple components or services interact correctly.

2. Step 2: Compare options with integration testing goal

   Only To check if different parts of the application work together correctly describes checking the cooperation of different parts, which matches integration testing.

3. Final Answer:

   To check if different parts of the application work together correctly -> Option C

4. Quick Check:

   Integration testing = check parts work together [OK]

Hint: Integration tests check connections between parts, not UI or speed [OK]

Common Mistakes:

• Confusing integration testing with UI testing
• Thinking integration tests measure performance
• Believing integration tests are for documentation

2. Which of the following is the correct way to send a GET request in a Python integration test using the requests library?

easy

A. response = requests.get('http://api.example.com/data')

B. response = requests.post('http://api.example.com/data')

C. response = requests.fetch('http://api.example.com/data')

D. response = requests.send('http://api.example.com/data')

Solution

1. Step 1: Recall correct method for GET request in requests library

   The method to send a GET request is requests.get().

2. Step 2: Check options for correct syntax

   Only response = requests.get('http://api.example.com/data') uses requests.get() correctly with a URL string.

3. Final Answer:

   response = requests.get('http://api.example.com/data') -> Option A

4. Quick Check:

   GET request method = requests.get() [OK]

Hint: Use requests.get() for GET requests, not post, fetch, or send [OK]

Common Mistakes:

• Using requests.post() instead of get() for GET requests
• Using non-existent methods like fetch() or send()
• Missing quotes around the URL string

3. Consider this Python integration test snippet using requests:

response = requests.post('http://api.example.com/login', json={'user':'admin', 'pass':'1234'})
print(response.status_code)

What will be the printed output if the login is successful and the server returns HTTP 200?

medium

A. 200

B. 403

C. 500

D. 404

Solution

1. Step 1: Understand HTTP status codes

   HTTP 200 means success, 404 means not found, 500 means server error, 403 means forbidden.

2. Step 2: Analyze the code output

   The code prints response.status_code. If login is successful, server returns 200, so output is 200.

3. Final Answer:

   200 -> Option A

4. Quick Check:

   Success status code = 200 [OK]

Hint: Successful HTTP response status is 200 [OK]

Common Mistakes:

• Confusing 404 (not found) with success
• Assuming 500 means success
• Mixing 403 (forbidden) with success

4. You wrote this integration test code:

response = requests.get('http://api.example.com/items')
assert response.status_code == 200
assert response.json()['items']

But the test fails with a KeyError on 'items'. What is the likely cause?

medium

A. The URL is incorrect and returns 404

B. The requests library is not imported

C. The status code is not 200

D. The response JSON does not contain the key 'items'

Solution

1. Step 1: Understand KeyError cause

   A KeyError means the key 'items' is missing in the JSON response.

2. Step 2: Analyze test failure reason

   Since status code is asserted as 200, the URL is likely correct and requests imported. The error is missing 'items' key in JSON.

3. Final Answer:

   The response JSON does not contain the key 'items' -> Option D

4. Quick Check:

   KeyError means missing key in JSON [OK]

Hint: KeyError means JSON key missing, not URL or import error [OK]

Common Mistakes:

• Assuming URL is wrong without checking status code
• Ignoring that status code 200 means request succeeded
• Forgetting to import requests library

5. You want to write an integration test that verifies your REST API endpoint /users returns only active users. The API returns JSON like {"users": [{"id":1,"active":true}, {"id":2,"active":false}]}. Which test code snippet correctly checks this condition?

hard

A. response = requests.get('http://api.example.com/users')
users = response.json()['users']
assert any(user['active'] for user in users)

B. response = requests.get('http://api.example.com/users')
users = response.json()['users']
assert all(user['active'] for user in users)

C. response = requests.get('http://api.example.com/users')
users = response.json()['users']
assert user['active'] == True

D. response = requests.get('http://api.example.com/users')
assert response.status_code == 404

Solution

1. Step 1: Understand requirement to check all users are active

   The test must confirm every user in the list has active set to true.

2. Step 2: Analyze code snippets for correctness

   response = requests.get('http://api.example.com/users') users = response.json()['users'] assert all(user['active'] for user in users) uses all() to check all users are active, which matches requirement. response = requests.get('http://api.example.com/users') users = response.json()['users'] assert any(user['active'] for user in users) checks if any user is active (not enough). response = requests.get('http://api.example.com/users') users = response.json()['users'] assert user['active'] == True references undefined variable user outside loop. response = requests.get('http://api.example.com/users') assert response.status_code == 404 checks for 404 status, which is wrong.

3. Final Answer:

   response = requests.get('http://api.example.com/users') users = response.json()['users'] assert all(user['active'] for user in users) -> Option B

4. Quick Check:

   Use all() to verify all users active [OK]

Hint: Use all() to confirm every user is active [OK]

Common Mistakes:

• Using any() instead of all() to check all users
• Referencing undefined variables in assertions
• Checking wrong HTTP status code