A React Native-based, simple groceries shop.

• Expostack: A really good app bootstrapper, which allows you to create advanced boilerplates with a few keyboard clicks.
• TanStack Query: Quite powerful fetching and state management library. A good alternative to Redux when you do not need too much complexity.
• Tamagui UI: Modern, a bit "on hype" UI KIT. Fast and multiplatform.
• Expo Router: File-based navigation. A good alternative to classic React Navigation.
• Jest: Reliable and worldwide beloved testing framework.
• Prettier: Almost de facto standard, code formatting tool.
• ESLint: Also almost an industry standard for linting and code style.

• To start app in the dev mode, run the folllowing comands:

npm install
npm run start 

• npm run test - run unit tests for the Api and Data Layer logic
• npm run lint - code style check
• npm run lint:formatting - check code formatting

To simulate API calls, I have chosen a strategy commonly used in full-scale API services. There are two types of entities:

• Data Layer - responsible for direct communication with the data source.
• API - holds the business logic.

This approach enables you to achieve the following benefits:

• Decoupling - You can replace parts of the system, making it capable of working with different data sources very easily. This technique especially shines when used in conjunction with dependency injection/inversion of control (as seen in frameworks like Nest.js or Angular, for example).

• Maintainability/Readability - A system comprised of small, well-structured pieces is much easier to read, develop, and maintain.

• Testability - A system composed of small pieces is much easier to test, allowing for independent testing of its components.

There are two types of tests in this application:

• Unit Tests - These tests check separated pieces of API logic independently. API tests verify business controllers. Data tests evaluate the Data Layer by itself. The implementation of these tests is somewhat artificial because in real-life scenarios, Data Layer tests would be integration tests due to the external location of the data source. Similarly, API tests because the way they work after they were provided with the fake Data Layer API. However, these tests showcase two important aspects. First, they demonstrate how easily separate code modules can be tested. Second, they illustrate how you can simulate API functionality working with various data providers just by consuming appropriately prepared fake data layers, which can simulate all kinds of real life remote systems.

• Integration - These tests evaluate the API subsistem as a whole. They validate the current business requirements of API simulation. So, these tests are the ones asked to provide in the task requirements, while the first ones were created solely for showcasing purposes.