Mini-Drive Lite

A Django 5 application that lets authenticated users upload, download, and semantically search files. Built for Windows-compatibility with SQLite and without Docker.

Features

• Upload and download files (stored on-disk using Django's FileField)
• Automatic processing of uploaded files:
  • PDF thumbnails (first page rendered to PNG)
  • Text extraction (from PDFs and text files)
  • OpenAI embedding generation for semantic search
• Semantic search using OpenAI embeddings and cosine similarity
• Expiring share links for file sharing (valid for 12 hours)
• REST API endpoints for file upload and search
• Bootstrap 5 UI with light/dark theme support
• HTMX for progressive enhancement (upload progress, etc.)

Setup

Prerequisites

• Python 3.12
• Windows 10/11 (also works on Linux/Mac)
• OpenAI API key

Installation

# Clone the repository
git clone https://github.com/your-username/mini-drive-lite.git
cd mini-drive-lite

# Create a virtual environment
python -m venv .venv
.venv\Scripts\activate  # Windows
# source .venv/bin/activate  # Linux/Mac

# Install dependencies
pip install -r requirements.txt

# Set up environment variables
cp .env.example .env
# Edit .env to add your OpenAI API key

# Run migrations
python manage.py migrate

# Create admin user
python manage.py createsuperuser

# Start the development server
python manage.py runserver

Environment Variables

Variable	Example	Description
SECRET_KEY	django-insecure-...	Django secret key
DEBUG	1	Development mode (1=True, 0=False)
ALLOWED_HOSTS	localhost,127.0.0.1	Comma-separated list of allowed hosts
OPENAI_API_KEY	sk-***	Your OpenAI API key
MEDIA_ROOT	media	Directory for file storage
EMBED_MODEL	text-embedding-3-small	OpenAI embedding model
MAX_STORAGE_MB	5000	Per-user storage quota in MB

Usage

Web Interface

1. Login: Visit http://localhost:8000/login/ and log in with your credentials
2. Upload: Upload files via the Upload page
3. View: View your files on the Dashboard
4. Search: Use the semantic search to find files by content
5. Share: Create temporary share links that expire after 12 hours

API Endpoints

• POST /api/upload/ - Upload a file (multipart/form-data)
• GET /api/search/?q=query - Search files by content
• GET /api/files/ - List all your files

Background Tasks

File processing runs in the foreground (Celery is configured in eager mode, so no separate worker is needed).

Notes

• PDF thumbnails are generated for the first page only
• Embeddings are stored in a JSON field (SQLite-compatible)
• Files are stored in the MEDIA_ROOT directory
• Storage quota is enforced per user

Deployment on Render.com

Prerequisites

• A Render.com account
• Your project pushed to a Git repository (GitHub, GitLab, or Bitbucket)

Steps to Deploy

1. Create a New Web Service

   • Log in to your Render.com dashboard
   • Click "New +" and select "Web Service"
   • Connect your Git repository
   • Choose the repository containing your Mini-Drive project

2. Configure the Web Service

   • Name: Choose a name for your service
   • Environment: Python
   • Build Command: ./build.sh
   • Start Command: gunicorn minidrive.wsgi:application
   • Plan: Choose a plan that fits your needs (Free tier available for testing)

3. Environment Variables Add the following environment variables in Render's dashboard:

   SECRET_KEY=your-secret-key
   DEBUG=0
   ALLOWED_HOSTS=your-render-app-url.onrender.com
   OPENAI_API_KEY=your-openai-api-key
   MEDIA_ROOT=media
   EMBED_MODEL=text-embedding-3-small
   MAX_STORAGE_MB=5000
   DJANGO_SUPERUSER_USERNAME=testuser (change accordingly)
   DJANGO_SUPERUSER_EMAIL=testuser@example.com (change accordingly)
   DJANGO_SUPERUSER_PASSWORD=testuser123 (change accordingly)
   

4. Database Configuration

   • The project uses SQLite by default, which is suitable for small deployments
   • For production, consider using PostgreSQL:
     • Add dj-database-url to requirements.txt
     • Update settings.py to use PostgreSQL when DATABASE_URL is present

5. Static and Media Files

   • Static files are automatically collected during build
   • For media files, consider using a service like AWS S3:
     • Add django-storages to requirements.txt
     • Configure S3 settings in settings.py

6. Deploy

   • Click "Create Web Service"
   • Render will automatically build and deploy your application
   • The first deploy might take a few minutes

Post-Deployment

1. Create Admin User

   • Access your deployed application
   • Run the following command in Render's shell:

     python manage.py createsuperuser

2. Monitor Your Application

   • Use Render's dashboard to monitor logs and performance
   • Set up alerts for any issues

Important Notes

• The free tier of Render.com will spin down after 15 minutes of inactivity
• Consider upgrading to a paid plan for production use
• Regularly backup your database
• Monitor your OpenAI API usage and costs
• Keep your dependencies updated for security