E-Commerce Trust Score Chrome Extension - Backend

This is the backend server for the E-Commerce Trust Score Chrome Extension, which provides trust scores, feedback, and reviews for e-commerce websites. The server is built using Node.js, Express, and MongoDB, and integrates with OpenAI for AI-generated feedback.

---

Features

• Website Verification: Check if a website is verified and retrieve its trust score.
• Feedback Management: Add user feedback and calculate ratings using Bayesian and time-decay algorithms.
• AI-Generated Feedback: Use OpenAI to analyze review messages and categorize feedback.
• Error Handling: Comprehensive error handling for validation, database, and runtime errors.

---

ER Diagram & Database Design

• Miro

Project Structure

.env
.env.example
.gitignore
.prettierrc.json
eslint.config.mjs
package.json
README.md
src/
  app.ts
  server.ts
  app/
    configaration/
      index.ts
    errors/
      AppError.ts
      handleCastError.ts
      handleDuplicateError.ts
      handleValidationError.ts
      handleZodError.ts
    interface/
      error.ts
      index.d.ts
      website.ts
    middlewares/
      globalErrorHandler.ts
      notFound.ts
      validateRequests.ts
    modules/
      verifiedWebsite/
        verifiedwebsite.controller.ts
        verifiedwebsite.route.ts
        verifiedwebsite.services.ts
        verifiedwebsite.model.ts
        verifiedwebsite.interface.ts
        verifiedWebsite.consts.ts
    routes/
      index.ts
    utils/
      sendResponse.ts
      extractValues.ts
      catchAsync.ts
      calculateRating.ts

---

Installation

1. Clone the repository:

   git clone https://github.com/wolfiee42/E-CommerceTrustScoreChromeExtensionServer.git
   cd E-CommerceTrustScoreChromeExtensionServer

2. Install dependencies:

   npm install

3. Create a .env file in the root directory and configure the following variables:

   NODE_ENVIRONMENT=development
   PORT=8080
   DATABASE_URL=<your_mongodb_connection_string>
   OPENAI_API_KEY=<your_openai_api_key>

4. Build the project:

   npm run build

5. Start the server:

   • Development mode:

     npm run dev

   • Production mode:

     npm start

---

API Endpoints

1. Check Verified Website

• URL: /api/v1/website/check
• Method: GET
• Query Parameters:
  • name (string): The website URL to check.
• Response:

  {
    "success": true,
    "message": "The website has been successfully retrieved.",
    "data": {
      "website": { "verified": true, "rating": 4.5 },
      "recentReviews": [...],
      "mostCommonFeedback": [...]
    }
  }

2. Add Feedback

• URL: /api/v1/website/review
• Method: POST
• Body:

  {
    "reviewMessage": "The website is beautiful, but the quality of their clothes is not good.",
    "siteId": "677bdcdad3ad95823e4eda0b",
    "feedback": {
      "ProductQuality": [{ "name": "Durability", "baseRating": 4 }],
      "CustomerServices": [{ "name": "Responsiveness", "baseRating": 5 }],
      "PlatformExperience": [{ "name": "Ease of Use", "baseRating": 5 }]
    }
  }

• Response:

  {
    "success": true,
    "message": "The website feedback has been successfully retrieved.",
    "data": { ... }
  }

3. AI-Generated Feedback

• URL: /api/v1/website/ai-generated-feedback/:id
• Method: GET
• Response:

  {
    "success": true,
    "message": "The website feedback has been successfully retrieved.",
    "data": {
      "ProductQuality": [...],
      "CustomerServices": [...],
      "PlatformExperience": [...]
    }
  }

---

Technologies Used

• Node.js: Backend runtime environment.
• Express: Web framework for building APIs.
• MongoDB: NoSQL database for storing website, feedback, and review data.
• OpenAI: AI integration for analyzing review messages.
• TypeScript: Strongly typed JavaScript for better code quality.
• Zod: Schema validation for request payloads.
• Day.js: Lightweight library for date manipulation.

---

Development Scripts

• Start Development Server:

  npm run dev

• Build Project:

  npm run build

• Run Linter:

  npm run lint

• Fix Linter Issues:

  npm run lint:fix

• Format Code:

  npm run format

---

Error Handling

The server includes a global error handler to manage various types of errors:

• Validation Errors: Handled using Zod and Mongoose validation.
• Duplicate Errors: Managed for unique constraints in MongoDB.
• Unhandled Rejections: Gracefully shuts down the server.
• Uncaught Exceptions: Ensures the server exits safely.

---

Environment Variables

Variable	Description
NODE_ENVIRONMENT	Application environment (e.g., development, production).
PORT	Port number for the server.
DATABASE_URL	MongoDB connection string.
OPENAI_API_KEY	API key for OpenAI integration.

---

License

This project is licensed under the MIT License.

---

Author

Developed by Saif.