Question 1

What is unit testing?

Accepted Answer

Unit testing: testing individual functions/components in isolation. Purpose: verify each unit works correctly before integration. Characteristics: fast execution, isolated (mock dependencies), automated, repeatable. Benefits: catch bugs early, safe refactoring, documentation of behavior, confidence in changes. Write tests for: logic functions, utility functions, components, edge cases. Use frameworks: Jest (JS), pytest (Python), JUnit (Java). Unit tests = foundation of testing pyramid.

Question 2

How do I write good unit tests?

Accepted Answer

Good unit tests: Test behavior, not implementation (what it does, not how). One assertion per test (clear pass/fail). Descriptive names (describe what is tested). Arrange-Act-Assert pattern (setup, execute, verify). Test edge cases (boundary, errors, empty). Keep tests fast (quick feedback). Independent tests (no shared state). Mock external dependencies (isolate unit). Avoid: testing private methods, over-mocking, flaky tests, giant test files. Refactor tests when refactoring code.

Question 3

What is test coverage?

Accepted Answer

Test coverage: percentage of code executed by tests. Types: line coverage (lines run), branch coverage (if/else branches tested), function coverage (functions called), statement coverage. Target: 80%+ for critical code, less for non-critical. Tools: Jest coverage, pytest-cov, JaCoCo. Note: coverage != quality - tests must be meaningful, not just hitting lines. High coverage with poor tests = false confidence. Focus coverage on critical paths, business logic, edge cases. Track coverage in CI, fail below threshold.

Question 4

What is the testing pyramid?

Accepted Answer

Testing pyramid: Unit tests (base, most tests) - fast, isolated, many. Integration tests (middle) - test component interactions, moderate speed. End-to-End tests (top, few) - test full flows, slow, expensive. Ratio: ~70% unit, ~20% integration, ~10% E2E. Reason: unit tests cheap/fast, catch most bugs. E2E tests slow/flaky, test complete system. Anti-pattern: inverted pyramid (many E2E, few unit) = slow CI, fragile tests. Balance: comprehensive coverage with reasonable runtime.

Question 5

What is mocking in unit testing?

Accepted Answer

Mocking: replacing real dependencies with fake implementations. Purpose: isolate unit, control behavior, avoid external calls. Mock types: Mock (replace entire object), Stub (provide canned responses), Spy (wrap real object, record calls), Fake (simplified working implementation). Mock what: external APIs, database calls, file system, network, time-dependent code. Don't mock: simple utilities, internal logic, domain objects. Over-mocking = tests don't test real behavior. Tools: Jest mocks, unittest.mock (Python), Mockito (Java).

Unit Testing Guide

Testing Concepts

Testing Frameworks

Best Practices

Anti-Patterns to Avoid

Testing Checklist