Image Swipe App

A web application for sorting and organizing images using swipe gestures, similar to dating apps.

Features

• Swipe Interface: Swipe images in four directions (left, right, up, down) to categorize them
• Full-size Image View: Click on any image to view it in full resolution with metadata
• History Page: View all your past selections with rich filtering and sort options
• NSFW Filtering: Toggle to include/exclude NSFW images on both the main swiper and history pages
• NSFW Blur: Optional blur for NSFW thumbnails on the history page with a toolbar toggle
• Orientation & Action Filters: Filter results by orientation (portrait/landscape/square) and by action taken
• Database Storage: All selections are saved in a SQLite database
• Reset Functionality: Option to clear all selections and start fresh

File Structure

• server.py: Entry point that starts the HTTP server
• handler.py: HTTP request handler implementing all API endpoints
• database.py: SQLite helpers (initialisation, queries, sync)
• config.py: Centralised constants (paths, port)
• index.html: Main page with the swipe interface
• history.html: Page to view and manage past selections
• js/: Front-end JavaScript (main.js, utils.js, etc.)
• styles.css: CSS styling for the application
• update_nsfw_flags.py: Utility script to (re)calculate NSFW flags for existing images after changing keyword lists

How to Use

1. Run the server: python server.py
2. Open a web browser and navigate to http://localhost:8000
3. Use the on-screen buttons or swipe gestures to categorize images:
   • Left: Discard
   • Right: Keep
   • Up: Favorite
   • Down: Review Later
4. Click on an image to view it in full resolution
5. Use the "View History" link to see all your selections. The toolbar lets you:
   • Filter by action/orientation/NSFW status
   • Toggle blur of NSFW thumbnails
   • Sort by creation date, swipe date or resolution
6. Use the "Reset Database" button in the history page to clear all selections

NSFW Detection & Maintenance

NSFW status is inferred from prompt keywords defined in config.py (NSFW_KEYWORDS).

After changing the keyword list or adding new images you can refresh the database flags:

python update_nsfw_flags.py

This will reset and recalculate all nsfw flags so that the new filters work correctly.

Requirements

• Python 3.x
• Standard Python libraries (http.server, sqlite3, etc.)
• Web browser with JavaScript enabled

Docker

If you have Docker installed you can build and run the application in an isolated container:

# Build image (only once)
docker build -t swiper-app .

# Run the container on port 8888
docker run --rm -p 8888:8888 swiper-app

Then open http://localhost:8888 in your browser.

The Dockerfile uses a minimal Python 3.11 image, installs the Python dependencies from requirements.txt, exposes port 8888 and starts the app with python server.py.

Docker Compose

A docker-compose.yml file is included so you can run the container with your image folders mounted automatically.

# Build image (first time or when code changes)
docker compose up --build

This will:

• Publish port 8888 on your host → container.
• Mount the portrait and landscape image folders (update paths in docker-compose.yml as needed) read-only into /media/Portrait and /media/Landscape inside the container.
• Set IMAGE_DIRS=/media/Portrait:/media/Landscape so the server finds the images.

Adjust ports, folders or environment variables by editing docker-compose.yml. Remember to rebuild (docker compose up --build) after changes.