File Parser QA Application

This repository contains a powerful application that allows users to upload files or provide Google Drive links and extract answers strictly based on the content of those files using OpenAI API.

Features

• User Authentication:

  • Users can register and log in using their email.

• File Upload & Parsing:

  • Supports major file formats, including:
    • .pdf, .txt, .docx, .xlsx, .csv, .json, .html, .xml, .jpg, .jpeg, .png.
  • Users can upload multiple files simultaneously.
  • Users can switch between currently uploaded files and previously uploaded files.

• Efficient Parsing:

  • Extracts accurate answers from the file content.
  • Provides the context of the answer, specifying where in the file the answer is located.

• Question History:

  • Maintains a history of user questions and answers for future reference.

• Seamless Integration:

  • Allows users to provide Google Drive links for file access.

Project Setup

• Instructions for running the application locally

  cd server
  npm install
  npm run dev
  
  cd ../frontend
  npm install
  npm start
  
  

Challenges Faced During Development

-API Rate Limits:

-Using OpenAI API heavily can result in rate limits being exceeded. -To mitigate this, multiple API keys can be used in a rotation strategy. -Performance on Free Cloud Servers:

-The deployed application may experience slower performance due to limited resources. -Upgrading to a paid cloud service with higher CPU capacity can enhance performance.

Environment Variables

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

MONGO_URI=your_mongo_database_uri
OPENAI_API_KEY=your_openai_api_key
JWT_SECRET=your_jwt_secret_key