About the project

customer-management-backend

To display a customer list based on search set to customer's First Name or Last Name and filters set to Company Name. I have used MERN tech stack creating the backend end in node, express and mongoose. The dependencies installed and work tasks are tracked in GITHUB issues url - https://github.com/RajiDurai/customer-management-backend/issues. I started this project by implementing the back end first and tested the api using POSTMAN inputting the endpoints and fetched test data from local mongo db set up.

Breakdown of how I started the project:

1. Create a repository in GHE
2. Clone repository

git clone <repository_url>

3. I like to start with backend so creating a backend project using Node.js. Front project using react.js is added later.

cd backend
npm init -y

4. Commit and push

git add .
git commit -m "Initial commit"
git push origin main

5. Backend implemenation(API endpoints)
   • create routes for data retrieval, handle search and filter requests.
   • Implement database queries to fetch customers based on search string on names and company. Note: I have handled filtering and search in the backend query itself.
6. Front end implementation
   • set up UI componenets - search text field for names, drop down for company name and list for data retrieval
   • Make requests to teh back end with passing search string
   • List display of results in UI
7. Testing
   • Write tests for backend routes, front end components

The implementation:

1. Setting Up the Project:

• Make sure you have Node.js and npm installed.(https://medium.com/@hayasnc/how-to-install-nodejs-and-npm-on-mac-using-homebrew-b33780287d8f)

• Create a new Node project: cd customer-management-backend npm init -y

• Install any necessary dependencies (e.g., express for server coding, cors - cross origin request and mongoose - Object database modelling): npm i express cors mongoose nodemon dotenv

2. Create routes(API endpoints implentation):

• Create a /routes folder and create customer.js for implementing customer search and filter
• Harcode imput values in json and test the end points using postman

3. Create database (database implemenation and add mock data):

• Install mongo db before proceeding to next step(source: https://www.sqliz.com/posts/install-mongodb-on-macos/)
• To start mongo service, use brew services start mongodb/brew/mongodb-community
• To access mongo shell, use mongosh. (Note. command 'mongo' doesn't work after version 6)
• Create a /models folder and create customer.js for creating customer schema.

4. Test the backend

Database schema:

 [
   { id: 1, firstName: 'Raji', lastName: 'Durai', company: 'NewRelic' },
   { id: 2, firstName: 'Rachel', lastName: 'Smith', company: 'Google' },
   { id: 2, firstName: 'Dhara', lastName: 'Thamil', company: 'Google' }
 ];

• Search functionality:

  • Input : GET request to endpoints - http://localhost:5000/api/customers?search=Raji. This request is currently hardcoded in url (to be implemented in front end)
  • Output : returns json data of customers list with matching first name or last name.
  • Edge cases : non-English characters

• Filter functionality:

  • Input : GET request to endpoints - http://localhost:5000/api/customers?filter_by_company_name=Google. This request is currently hardcoded in url (to be implemented in front end)
  • Output : returns json data of customers list with matching company name.
  • Edge cases : non-English characters

Here are the steps involved to run the project locally:

Clone the project from the git URL -

In the project directory, you can run:

npm start

Runs the app in the development mode.
Open http://localhost:5000 to view it in your browser or use POSTMAN.

npm test

Launches the test runner in the interactive watch mode.
See the section about running tests for more information.

Your app is ready to be deployed!

Future goals

• write tests covering all possible end points and different inputs mismatching reg ex and non-english characters unit tests.
• write routing for pagination and sort
• Install newrelic monitoring to know about the performance
• Seed the database from /faker and test robustically