Auctus v2 Documentation Portal
Welcome to the internal engineering documentation for Auctus v2—our specialized dataset discovery engine and open-data harvesting pipeline.
This platform is hosted and managed under our organization credentials to streamline catalog syncing, vector-accelerated dataset searches, and automated metadata profiling.
🧭 Document Navigation
Use the top navigation bar or the sidebar menus to explore the system layout:
- Architecture Overview: A deep dive into our multi-tier infrastructure, data persistence layers (OpenSearch & MinIO), and background task worker lifecycles.
- Installation & Local Setup: Step-by-step developer environment assembly, database initializations (
storage.initialize_os), testing seed rules, and live data ingestion runs. - Search API Reference: Automated code and docstring specifications for our vector and keyword search routing engine.
- Portals API Reference: Documentation covering dynamic multi-provider registry tracking and analytics aggregation.
🛠️ Quick Developer Stack Reference
| Service Component | Technology | Default Local Port | Purpose |
|---|---|---|---|
| Frontend Web UI | React / TailwindCSS | 5173 |
Client dashboard & map interfaces |
| API Gateway | FastAPI (Python) | 8000 |
Gateway & dynamic payload streaming |
| Search Engine | OpenSearch | 9200 |
Core catalog index (Keyword & k-NN vector scoring) |
| Search Dashboard | OpenSearch Dashboards | 5601 |
Cluster administration & index visualization |
| Task Coordinator | Redis / ARQ | 6379 |
Asynchronous harvesting pipeline orchestration |
| Object Data Lake | MinIO Storage | 9000 / 9001 |
High-volume structured JSON profile matrix storage |