CLEAN CODE PRACTICE: 0. pip install ruff mypy
- Installiere das Tool: pip install pre-commit
- Installiere die Hooks in deinem Repo: pre-commit install (nicht sicher ob man es nur 1mal pro repo macht)
pre-commit autoupdate -> um ein automatischen update der revs durchzuführen
Wir nutzen Ruff und mypy zum automatisierten sichern von Clean code practices
Auth directory is used to story the Authorization for the Neo4j database -> Offload to .env
ESCO_DB contians the ESCO dataset in RDF format (json-id or ttl or rdf): https://esco.ec.europa.eu/en/use-esco/download
Neo4J Aura access: console.neo4j.io
root/ ├── data/ # Raw and processed datasets (PDFs, CSVs) ├── docs/ # Project documentation and API specs ├── src/ │ ├── main.py # Entry point for the application │ ├── database/ │ │ ├── init.py │ │ └── db_adapter.py # Database connection and CRUD operations │ ├── ingestion/ │ │ ├── init.py │ │ ├── cv_ingestor.py # CV-specific parsing and loading │ │ └── rfq_ingestor.py# RFQ-specific parsing and loading │ ├── services/ # Business logic and external API wrappers │ │ ├── init.py │ │ ├── llm_service.py # NEW: Ollama request handling & prompting │ │ └── embedding.py # Embedding creator / vectorization logic │ ├── retrieval/ # Logic for querying vector stores │ │ ├── init.py │ │ └── search.py │ ├── schemas/ # Pydantic models for data validation │ │ ├── init.py │ │ ├── cv_schema.py # Schema file 1 │ │ └── rfq_schema.py # Schema file 2 │ └── graph/ # LangGraph / Workflow logic │ ├── init.py │ ├── nodes.py # Individual graph step functions │ └── graph_init.py # graph_initiation.py (workflow definition) ├── .env # Environment variables (Ollama URL, DB creds) ├── pyproject.toml # Dependencies and build system └── README.md