Testing & Error Handling

Unit tests check small pieces of code, integration tests check how components and data work together, and end-to-end tests simulate browser behavior. Robinson recommends tests that resemble real user actions and prioritizing high-value flows, such as login or payment, over pursuing full unit-test coverage without considering the effort.

Robinson describes Jest as the test runner and React Testing Library as a way to query rendered DOM in a user-oriented way. He demonstrates setting up a shared render helper with Chakra providers, then testing a bar card's text, image, and deal-count badge. He advises using snapshots cautiously, since large snapshots can be hard to review and easy to accept without inspection.

The sign-in example checks a critical path by entering credentials, submitting the form, and verifying the result. The tests mock the authentication hook and router, then check successful sign-in and navigation, an API error displayed to the user, and validation messages when fields are blank. Robinson recommends limiting mocks while testing user-visible behavior.

Robinson recommends exception monitoring such as Sentry to catch errors in production and lower environments, noting that Next.js apps need to account for both server-side and client-side errors. He shows how a custom `_error` page can present different experiences for errors such as 404s and 500s, and suggests checking Next.js's Sentry plugin and examples for setup.