Class 5 - Continuous Integration & TDD

Question 1

Stinky Fish Exercise

Answer 1

An exercise where team members share their individual concerns & worries about the future of the project as a way to start a conversation and begin to confront or overcome concerns

Question 2

Continuous Integration (CI)

Answer 2

Developers regularly save, commit, pull, and push their code. An automated build occurs after which automated tests are run to assert code correctness before integration.

Question 3

What is accomplished by saving, committing, pulling, and pushing often?

Answer 3

You are less likely to lose work or have to spend time merging two developer’s code

Question 4

What is accomplished by running an automated build after every push?

Answer 4

You can quickly catch bugs that cause your code to not compile

Question 5

What is accomplished by running automated tests after every push?

Answer 5

Automated tests give developers a better idea of the state of the codebase. They tell developers what functionalities are not working properly and what functionalities need to be worked on.

Question 6

Continuous Delivery

Answer 6

Deploy app to staging environment after every push to ensure the app can be reliably released at any time. It is more comprehensive than continuous integration and it provides a more accurate picture of the current state of the codebase

Question 7

Continuous Deployment

Answer 7

Same procedure as continuous delivery (push, build, run tests, deploy) with the caveat that the codebase is only deployed to production if all tests pass. Otherwise, a rollback occurs.

Question 8

What is a potential benefit of continuous deployment

Answer 8

The development team is forced to write tests everyone REALLY believes in

Question 9

Quality Assurance (QA)

Answer 9

Essentially, making sure your code works before something terrible happens

Question 10

What are the different branches of QA?

Answer 10

• Fault Avoidance (verification, reviews, configuration management)
• Fault Detection (testing, debugging)
• Fault Tolerance
• Usability Testing (scenario testing, prototype testing, product testing)

Question 11

True or False: Testing is a proof of correctness

Answer 11

False. Testing can never completely identify all defects within a piece of software

Question 12

Testing

Answer 12

The process of validating and verifying that a software program meets business and technical requirements. Testing is used to determine if software works as expected.

Question 13

What are the five dimensions of testing?

Answer 13

• Testers
• Coverage
• Potential Problems
• Activities
• Evaluation

Question 14

Testers (Dimension of Testing)

Answer 14

Who does the testing. This includes developers and independent testers.

Question 15

Coverage (Dimension of Testing)

Answer 15

What gets tested

Question 16

Potential Problems (Dimension of Testing)

Answer 16

Why you’re testing

Question 17

Activities (Dimension of Testing)

Answer 17

How you test

Question 18

Evaluation (Dimension of Testing)

Answer 18

How to tell whether the test passed or failed

Question 19

What is the disadvantage of having developers test code?

Answer 19

Developers are too close to the code and understand it too well. Furthermore, they are often driven by delivery schedules and are busy fixing bugs

Question 20

What is the disadvantage of having independent testers test code?

Answer 20

It will take them some additional time to learn the system (since they did not write the codebase)

Question 21

What is needed to develop an effective test?

Answer 21

• Detailed understanding of the system
• Knowledge of testing techniques
• Skill to apply testing techniques effectively and efficiently

Question 22

Why does testing require creativity?

Answer 22

Programmers have a tendency to stick to the data set that makes the program work (it is intuitive to them since they wrote the program). However, this strategy often does not cover all possible actions a user might take.

Question 23

What are the different types of tests?

Answer 23

• UI tests
• End-to-end tests (system tests)
• Integration tests
• Unit tests

Question 24

UI Tests

Answer 24

Simulates a user interacting with the application. Runs in either a real environment or an isolated full development environment

Question 25

End-to-end tests (system tests)

Answer 25

Tests the whole program at once. Runs in either a real environment or an isolated full development environment

Question 26

Integration tests

Answer 26

Tests involving multiple classes and modules working together. Runs in either a real environment or an isolated full development environment

Question 27

Unit tests

Answer 27

Tests a single class/method/function or even a single branch or conditional (if statement). Runs in the development environment.

Question 28

Runtime Assertions

Answer 28

A statement that is expected to always be true at the point in the code it appears. This allows programmers to catch problems early before they become bigger problems in production

Question 29

What is the downside of using runtime assertions

Answer 29

They can make code hard to read if not used carefully

Question 30

Order the different types of tests from most tests to lest tests in the codebase

Answer 30

Highest to lowest: Unit tests, integration tests, end-to-end tests, UI tests

Question 31

In what order do the different types of tests run?

Answer 31

First to last: Unit tests, integration tests, end-to-end tests, UI tests

Question 32

Order the different types of tests by the amount of code tested by 1 test

Answer 32

Highest to lowest: UI tests, end-to-end tests, integration tests, unit tests

Question 33

Test-Driven Development (TDD)

Answer 33

1. Write test for functionality that does not exist
2. Test fails (since no implementation has been provided)
3. Write code to make the test pass

Question 34

What are the 3 rules of TDD?

Answer 34

1. Do not write any production code until you have first written a test that fails due to the lack of that code
2. Do not write more of a test than is sufficient to fail. Failing to compile counts as a failure
3. Do not write more production code than is sufficient to pass the currently failing test