Dreamstime Automation Bot - Enhanced Edition

Automated image submission workflow for Dreamstime with advanced AI-powered features.

Features

Core Automation

• Automatic login to Dreamstime
• Navigate to upload section
• Process uploaded images
• Title sanitization (remove colons, limit to 115 chars)
• Auto-submit images

Advanced Features

• Template-based description enhancement - Two template sets with 40 descriptive phrases each
• AI image auto-categorization - Automatically categorize AI-generated images
• Model release management - Add model releases automatically
• Exclusive content marking - Mark images as exclusive
• Batch processing - Process up to 10,000 images per run
• Pause intervals - Pause every 20 images for 2 minutes by default
• Duplicate detection - Skip or stop on duplicate image IDs
• Configurable delays - Fast (5-10s) or Slow (10-15s) modes
• Stop functionality - Gracefully stop automation anytime
• Real-time progress tracking - See exactly what's happening
• Gemini AI Analysis - Auto-generate descriptive titles and descriptions
• Health monitoring - Automatic browser recovery on failure

Requirements

• Python 3.8+
• Playwright
• Flask
• python-dotenv
• Google Gemini API key (optional, for AI features)

Installation

1. Clone the repository

   git clone <repository-url>
   cd Dreamstime-AutoUploader

2. Install dependencies

   pip install -r requirements.txt

3. Install Playwright browsers

   playwright install chromium

4. Configure credentials

   Create .env file with your settings:

   DREAMSTIME_USERNAME=your_username
   DREAMSTIME_PASSWORD=your_password
   FLASK_SECRET_KEY=your-secret-key-here
   FLASK_DEBUG=True
   GEMINI_API_KEY=your-gemini-api-key
   

Usage

Start the Web Interface

python app.py

Then open your browser to: http://localhost:5000

Configure Options

The web interface provides comprehensive options:

1. Description Enhancement

• Template: Choose from professional or creative phrase sets
• Manual Description: Add custom text to all descriptions

2. Image Properties

• AI Generated: Auto-categorize as AI (category 172, subcategory 212)
• Model Release: Add first available model release
• Exclusive: Mark as exclusive content

3. Processing Configuration

• Speed: Fast (5-10s delays) or Slow (10-15s delays)
• Number of Images: How many to process (default 10,000)
• Pause Settings: Pause every 20 images for 2 minutes by default
• Duplicate Action: Skip or stop on duplicate image IDs

Example Configurations

Simple Batch Processing

Template: None
Number of Images: 10
Speed: Fast
Duplicate Action: Skip

AI Images with Professional Descriptions

Template: Template 1 (Professional)
AI Generated: Yes
Number of Images: 100
Speed: Fast
Pause After: 20 images
Pause Duration: 120 seconds

Exclusive Content with Model Releases

Template: Template 2 (Creative)
Manual Description: "High quality commercial content"
Model Release: Yes
Exclusive: Yes
Number of Images: 500
Pause After: 20
Duplicate Action: Stop

API Integration

The bot exposes RESTful API endpoints for integration with n8n, webhooks, or other automation tools:

Endpoints

POST /api/start      - Start automation with optional parameters
POST /api/stop       - Stop running automation
GET  /api/status     - Get current automation status
GET  /health         - Health check for the bot service

Example API Call

curl -X POST https://n8n.shravanpandala.me/dreamstime/api/start \
  -H "Content-Type: application/json" \
  -d '{"repeatCount": 100}'

Health Check

The bot includes automatic health monitoring:

• Systemd timer checks browser connection every 2 minutes
• Auto-restarts browser if remote debugging port is unavailable
• Maintains persistent browser session to prevent captcha challenges
• VNC interface available at: https://vnc.shravanpandala.me

Architecture

Backend (Python + Playwright)

• automation.py: Core automation logic with Playwright

  • AutomationState: State management class
  • DreamstimeBot: Main automation orchestrator
  • Gemini AI integration for image analysis

• app.py: Flask web server

  • REST API endpoints for control
  • Real-time status updates
  • Options handling and validation

• config.py: Configuration management

  • Credentials and API key loading
  • Default option values
  • URL configurations

• utils.py: Utility functions

  • Template management (80+ phrases)
  • Delay calculation with randomization
  • Title sanitization
  • Stop-safe wait operations

• gemini_analyzer.py: AI analysis module

  • Google Gemini API integration
  • Image-to-text analysis
  • Title and description generation

Frontend (HTML + CSS + JavaScript)

• templates/index.html: Web interface

  • Comprehensive options panel
  • Real-time progress display
  • Start/Stop controls

• static/style.css: Modern, professional styling

  • Dark mode design
  • Smooth animations

Services

• chromium-dreamstime.service: Systemd service for persistent browser
• dreamstime-bot.service: Systemd service for Flask app
• chromium-healthcheck.service: Health monitoring and recovery
• chromium-healthcheck.timer: Runs health checks every 2 minutes

Progress Tracking

Real-time updates show:

• Current step and action
• Success/error status
• Progress percentage
• Processed vs. successful count
• Timestamps for all actions

UI Features

• Modern dark theme with gradient design
• Real-time status with color-coded indicators
• Scrollable progress log with icons
• Responsive design for all screen sizes
• Smooth animations and transitions

Debugging

Check console logs for detailed execution information:

# View bot service logs
sudo journalctl -u dreamstime-bot -f

# View browser service logs
sudo journalctl -u chromium-dreamstime -f

# Check health check status
sudo systemctl status chromium-healthcheck.timer

# Verify browser connection
curl http://localhost:9222/json/version

Important Notes

1. Persistent Browser Session: The browser maintains a persistent session to avoid repeated captcha challenges
2. Health Monitoring: Automatic health checks (every 2 minutes) restart the browser if needed
3. Initial Login: First login may require manual captcha solving via VNC viewer
4. Credentials Security: Never commit your .env file to version control
5. Rate Limiting: Default delays (5-15 seconds) help avoid triggering anti-bot measures
6. Duplicate Detection: The bot tracks image IDs to prevent reprocessing
7. Max Images Per Run: Default 10,000 images per API call (customizable via payload)
8. AI Analysis: Gemini API is optional; titles fall back to descriptions if unavailable

Changelog

Version 2.1 (Current)

• Added Gemini AI image analysis for descriptive titles/descriptions
• Improved fallback title logic to use descriptions when AI unavailable
• Increased default batch size to 10,000 images per run
• Added health check system with automatic browser recovery
• Fixed browser session persistence to prevent captcha loops
• Reduced delay ranges (5-10s fast, 10-15s slow) for faster processing
• Professional README with reduced emoji usage

Version 2.0

• Added template system with 80 descriptive phrases
• AI image auto-categorization
• Model release management
• Exclusive content marking
• Configurable batch processing
• Pause interval support
• Duplicate image detection
• Configurable processing speeds
• Stop functionality
• Enhanced progress tracking

Version 1.0

• Basic automation workflow
• Login and navigation
• Image processing
• Basic progress tracking

Support

For issues or questions:

1. Check browser status via VNC: https://vnc.shravanpandala.me
2. Review bot logs: sudo journalctl -u dreamstime-bot -f
3. Check browser service: sudo systemctl status chromium-dreamstime
4. Verify health check: systemctl status chromium-healthcheck.timer
5. Test connection: curl http://localhost:9222/json/version
6. Ensure .env file is properly configured

Acknowledgments

• Built with Playwright for reliable browser automation
• Flask for the web interface
• Google Gemini AI for image analysis
• Systemd for service management and health monitoring