A Flask-based web application for searching and downloading character cards from the Character Archive. Features a modern Tailwind CSS interface with full-text search, character previews, and card downloads.

• Full-text search across all character sources (Chub, JanitorAI, RisuAI, Character Tavern, Booru, Webring, Generic)
• Character card previews with lazy-loaded images
• Detailed character modal showing tags, description, and first message
• Card downloads in PNG format with embedded character data
• Source filtering to search specific platforms
• Responsive design with dark theme and purple gradient accents
• Pagination with customizable results per page

Flask API with the following endpoints:

Single-page application using:

• Tailwind CSS (via CDN) for styling
• Vanilla JavaScript for interactivity
• Responsive grid layout (2-6 columns based on screen size)
• Modal animations and hover effects
• Custom text cleaning for character card formatting

• Docker Engine (20.10+)
• Docker Compose (v2.0+)
• PostgreSQL database with Character Archive data
• Image archive at ../archive/hashed-data/

Create a .env file or set environment variables:

DB_HOST=postgres              # PostgreSQL host
DB_PORT=5432                  # PostgreSQL port
DB_NAME=char_archive          # Database name
DB_USER=char_archive          # Database user
DB_PASSWORD=your_password     # Database password
ARCHIVE_PATH=/archive         # Path to archive images

Add this service to your docker-compose.yml:

services:
  frontend:
    build:
      context: ./small_front
      dockerfile: Dockerfile
    environment:
      DB_HOST: postgres
      DB_PORT: "5432"
      DB_NAME: char_archive
      DB_USER: char_archive
      DB_PASSWORD: your_password_here
      ARCHIVE_PATH: /archive
    volumes:
      - ./archive:/archive:ro
    ports:
      - "8080:5000"
    depends_on:
      - postgres
    networks:
      - char_archive_network

Start the services:

docker compose up -d frontend

Open your browser and navigate to:

• Local: http://localhost:8080
• Tailscale: http://100.108.69.91:8080 (if configured)

1. Clone the repository:

git clone https://github.com/sproutingnerd/char-archive-small_frontend.git
cd char-archive-small_frontend/small_front

2. Install dependencies:

pip install -r requirements.txt

3. Set environment variables:

export DB_HOST=localhost
export DB_PORT=5432
export DB_NAME=char_archive
export DB_USER=char_archive
export DB_PASSWORD=your_password
export ARCHIVE_PATH=/path/to/archive

4. Run the Flask app:

python app.py

The app will be available at http://localhost:5000

docker compose build frontend && docker compose up -d frontend

docker compose logs -f frontend

curl "http://localhost:8080/api/search?q=test&source=chub&page=1&per_page=24"

Parameters:

• q (required): Search query
• source (optional): Filter by source (chub, generic, booru, webring, char_tavern, risuai, nyaime, all)
• page (optional): Page number (default: 1)
• per_page (optional): Results per page (default: 24)

Response:

{
  "results": [
    {
      "id": "123456",
      "author": "username",
      "name": "Character Name",
      "image_hash": "abc123...",
      "added": "2024-12-01T00:00:00+00:00",
      "source": "chub"
    }
  ],
  "total": 1000,
  "page": 1,
  "per_page": 24,
  "pages": 42
}

curl "http://localhost:8080/api/character/chub/123456"

curl "http://localhost:8080/api/card/chub/123456" -o character.png

curl "http://localhost:8080/api/stats"

The frontend queries these tables per platform:

Each table follows the pattern:

• {platform}_character - Scraped metadata with JSON data
• {platform}_character_def - Card definitions with:
  • definition (JSONB): Character data
  • raw (bytea): zlib-compressed original card
  • image_hash: MD5 hash for image lookup
  • metadata (JSON): Safety ratings and token info

Images are stored in archive/hashed-data/ using MD5 hash sharding:

Path format: /{h[0]}/{h[1]}/{h[2]}/{h[3:]}

Example:

• Hash: 54db4830ceab552d4824dd5b016f4b06
• Path: /5/4/d/b4830ceab552d4824dd5b016f4b06

• Check database connection in logs
• Verify PostgreSQL is running and healthy
• Test database directly: docker compose exec postgres psql -U char_archive -d char_archive

• Verify archive volume is mounted correctly
• Check image hash exists in archive/hashed-data/
• Verify file permissions on archive folder

• Check logs: docker compose logs frontend
• Ensure PostgreSQL is healthy before frontend starts
• Verify all environment variables are set correctly

• Ensure DB_HOST matches PostgreSQL service name in docker-compose
• Check database credentials
• Verify PostgreSQL port is accessible

char-archive-small_frontend/
├── README.md                      # This file
├── docker-compose.yml             # Production Docker Compose config
├── docker-compose.import.yml      # Docker Compose with DB import
├── init-db.sh                     # Database initialization script
├── docs/                          # Documentation
│   ├── MIGRATION.md               # Server migration guide
│   ├── setup-guide.md             # Complete Docker setup
│   ├── frontend-guide.md          # API and architecture details
│   ├── DATABASE_STRUCTURE.md      # Full database schema
│   └── FILE_STRUCTURE.md          # Image storage layout
└── small_front/                   # Frontend application
    ├── app.py                     # Flask API backend
    ├── templates/
    │   └── index.html             # Tailwind CSS frontend
    ├── requirements.txt           # Python dependencies
    └── Dockerfile                 # Container configuration

• Flask 3.1.0 - Web framework
• psycopg2-binary 2.9.10 - PostgreSQL adapter
• Pillow 11.0.0 - Image processing
• gunicorn 23.0.0 - Production WSGI server

See requirements.txt for complete list.

• Full Setup Guide - Complete Docker setup instructions
• Frontend Architecture - Detailed API and schema documentation
• Database Structure - Complete database schema reference
• File Structure - Image storage and file organization
• Migration Guide - Server migration instructions

Part of the Character Archive preservation project. Original server and scraper source code:

• char-archive-server
• char-archive-scraper