Homework 2: Type Validation

In this assignment, you will be validating inventory data for an fictional online store that sells JavaScript merchandise called BazaarJS. The only items in your starter code are multiple store inventory catalogs in the form of JSON Files. Imagine you were given these JSON files by a co-worker to validate, or you queried this data from an API or a database. Your task is to build a TypeScript project that can perform this validation.

You will need to following dependencies to develop this application:

• typescript
• @types/node
• zod
• eslint & typescript-eslint
• prettier
• jest & ts-jest

This may not be an exhaustive list of the dependencies you need for this project. You are free to install any other packages that may help for this project, so long as the core of the project still uses TypeScript and Zod.

Your project must have a script called build that invokes tsc to compile your TypeScript files into an output directory. You may use any TypeScript configuration that works for you so long as the build command works to compile your code. While there's no strict file naming conventions for this project, we recommend you use some default names, such as src for your source code/input directory and dist for your output directory where your .ts files will compile into .js files.

Part 1: Installation and Setup

You should first setup your project and install your dependencies. Make sure to keep your project root in one folder so that you can easily submit all the files in your project at the end.

Fill out your package.json with proper metadata, including the project name, description, author, the name of the main script (the entry point for this application), and a license name. For the license,

• Choose some open-source license (we recommend ISC)
• Create a LICENSE.txt file next to your package.json
• Add the placeholder text from the license, then fill in placeholders like year and owner

At this point, your project is setup, and ready to publish to npm if you so choose! (through we won't require you to publish it for this homework)

Part 2: Schema

Your first task is to create a Zod schema for the data within the provided JSON file. This data consists of products within the online store that may have some common properties, and some differing properties based on their category. Start by taking a look at the JSON file to familiarize yourself with the shape of the data you will be validating.

Each product must contain the following common properties:

• id — a positive integer
• name — a non-empty string
• price — a number greater than 0
• category — a string, among the literals "apparel", "accessories", or "stickers"
• inStock — a boolean
• tags — an array containing at least one string
• seller — an object containing a valid seller schema (see below)

A seller must contain the following properties:

• id — a positive integer
• name — a non-empty string
• rating — a number between 0 and 5

The category property determines which additional properties are required for a product. Your schema must represent these different product types as a discriminated union using category as the discriminator.

Finally, create an array schema that represents the entire contents of the JSON file. The root value of the file should be an array containing only valid products. This schema should represent your discriminated union of the three types of products.

Part 3: Parsing and Validation

Write a function that takes in a file name (the JSON data) and outputs various console logs based on the contents of the file alongside a validation result object. Within this function, parse the input JSON using fs.readFileSync() (similar to HW1). Then, use Zod to validate that JSON data against your defined product array schema (not the inferred TypeScript type) with safeParse().

Validation will either succeed or fail.

If validation succeeds:

• Print out a success message.
• Print out the total number of products.
• List each product id, name, category, and at least one of the additional properties of the category in a single formatted string.
• Return the result object

If validation fails:

• Print out a failure message.
• List out the products that had validation errors, and their associated error message.
• Return the result object

Part 4: Unit Testing

In preparation to test, export the validation function you just wrote so that your test files can use it as a module.

You will set up and write some unit tests for your project. We will be using Jest for this purpose, so look up how to install and use the syntax of Jest in order to prepare your test suite. We'd recommend using ts-jest alongside this, but feel free to find your own solution if you come across one. Additionally, set up a script in package.json to be able to quickly run the test suite with either npm run test or npm test.

For your unit tests, you must write:

• 1 test containing a valid catalog
• 1 test containing an invalid catalog
• 2 tests containing custom catalog(s) you wrote yourself (not provided by the starter)

Since you are writing your own JSON catalogs, be sure to include these files with your submission!

Refer to the rubric at the bottom of the page on how these tests will be graded. These tests can take any form, so long as they represent an attempt to make use of Jest's features to check that validation is performed correctly.

Part 5: Style Checking

Finally, you should polish up your project and standardize the style within the project. With eslint, add 4 extra style rules beyond any recommended default rules for JavaScript and TypeScript. With prettier, add 2 extra rules beyond the defaults. This process will give you a taste of setting up style checking as part of the production pipeline of projects. Don't just copy the rules from Homework 1 (since some of them require extra dependencies), try branching out a bit and exploring how different rules work!

README

Answer the provided reflection questions within the starter code README file. In this reflection, you will also indicate whether or not you used AI, and also document your usage of AI as well. Please don't forget this step, as it is important feedback for the homework and the content of the course!

Submission

Submit your code through Gradescope as a .zip file that contains your project. Make sure your project includes all files you worked on during this homework and your README.md file, all config files for TS, eslint, and prettier, your test script, and the every JSON file used in the tests. You should NOT include the node_modules folder in the .zip file (feel free to delete it before submission), as it is quite heavy and we will reinstall the dependencies for grading anyways. Make sure the submitted file structure within your submission is exactly or similar to the file structure you used to run and develop the project. Points will be taken off for malformed project structures in the final submission!

Since you are preparing eslint in your own, we will subject your code to your defined linting rules instead of the standard rules we would apply, so make sure you pass your own set of style rules!

Rubric