Ontology Viewer

A web application for viewing and exploring OWL ontologies with discourse relation examples. Built with FastAPI backend and Angular frontend.

Features

• Corpus Discovery: Automatically loads OWL files following the <corpus>-<lang>.owl naming convention
• Advanced Filtering: Filter examples by language, connective, relation type, symmetry, and full-text search
• Pagination: Browse through examples with configurable page sizes
• Responsive Design: Mobile-friendly interface
• Detailed View: Modal popup with complete example information
• RESTful API: Comprehensive JSON API with filtering capabilities

Project Structure

ontology-viewer/
├── backend/                          # Python FastAPI backend
│   ├── src/
│   │   ├── main.py                   # FastAPI application
│   │   ├── models.py                 # Pydantic models
│   │   ├── parser.py                 # OWL file parsing
│   │   └── config.py                 # Configuration
│   ├── tests/
│   │   └── test_parser.py
│   ├── requirements.txt
│   └── Dockerfile
├── frontend/                         # Angular frontend
│   ├── src/
│   │   ├── app/
│   │   │   ├── components/
│   │   │   │   ├── corpus-list/
│   │   │   │   ├── example-detail/
│   │   │   │   ├── examples-list/
│   │   │   │   └── filter-bar/
│   │   │   ├── services/
│   │   │   │   └── api.service.ts
│   │   │   ├── models/
│   │   │   │   └── example.model.ts
│   │   │   ├── app.component.ts
│   │   │   └── app.module.ts
│   │   ├── index.html
│   │   ├── main.ts
│   │   └── styles.css
│   ├── angular.json
│   ├── tsconfig.json
│   ├── tsconfig.app.json
│   ├── tsconfig.spec.json
│   ├── package.json
│   ├── package-lock.json
│   ├── postcss.config.cjs
│   ├── tailwind.config.cjs
│   ├── proxy.conf.json
│   ├── nginx.conf
│   ├── .dockerignore
│   └── Dockerfile
├── nginx/                            # Nginx reverse proxy config
│   ├── conf.d/
│   │   └── security.conf
│   ├── html/
│   │   ├── 429.html
│   │   └── 50x.html
│   └── nginx.conf
├── data/                             # OWL ontology files
│   ├── discourse.ISO.UDisc.owl
│   ├── ISO-24617-8-en.owl
│   ├── PDC-pl.owl
│   └── PDC-Guidelines-pl.owl
├── docker-compose.yml
├── .env.example
├── .gitignore
└── README.md

Quick Start

Prerequisites

• Docker and Docker Compose

Deployment

1. Clone the repository and configure environment:

   cp .env.example .env

2. Edit .env — set your domain and allowed origins:

   DOMAIN_NAME=example.com
   ALLOWED_ORIGINS=https://example.com

3. Prepare SSL certificates:

   mkdir -p nginx/ssl
   cp /path/to/fullchain.pem nginx/ssl/fullchain.pem
   cp /path/to/privkey.pem nginx/ssl/privkey.pem
   openssl dhparam -out nginx/ssl/dhparam.pem 2048

4. Start the stack:

   docker compose up -d

   The application will be available at https://<your-domain>.

5. View logs:

   docker compose logs -f

6. Stop the stack:

   docker compose down

Adding New Corpora

1. Place OWL files in the data/ directory
2. Follow naming convention: <corpus-name>-<language-code>.owl
3. Ensure files import the main ontology discourse.ISO.UDisc.owl
4. Restart the stack: docker compose restart backend

API Endpoints

Health Check

• GET /api/health — Returns server status

Corpora

• GET /api/corpora — List available corpora with counts

Examples

• GET /api/examples — Get paginated examples with filtering
  • Query parameters: corpus, language, connective, relation, symmetric, q (search), page, page_size
• GET /api/examples/{id} — Get specific example by ID

Ontology

• GET /api/ontology/classes — Get ontology class hierarchy
• GET /api/ontology/relations — Get relation classes with symmetry info

Data Format

Example Object

{
  "id": "http://purl.org/olia/discourse/discourse.ISO.owl#Example_Purpose",
  "text": "Idź do babci, żeby dała ci 20 złotych",
  "language": "pl",
  "connective": "żeby",
  "relation": "Purpose",
  "symmetric": false,
  "corpus": "PDC",
  "ontology_types": [
    "http://purl.org/olia/discourse/discourse.ISO.owl#Purpose"
  ],
  "provenance": "PDC-pl.owl"
}

Corpus Object

{
  "name": "PDC",
  "language": "pl",
  "file": "data/PDC-pl.owl",
  "count": 8
}

Acknowledgements

This research was funded in whole by the National Science Centre, Poland, grant 2023/50/A/HS2/00559 ("Universal Discourse: a multilingual model of discourse relations").