How epfl-enac/ccfatigueplatform works
Project type and purpose
• This is a full-stack web application (“CCFatigue Platform”) for managing, harmonizing, storing, and analyzing composite-materials testing datasets.
• The backend (folder “backend/”) is a Python FastAPI service with database migrations (Alembic), data-preprocessing utilities, and core analysis logic.
• The frontend (folder “frontend/”) is a Vue.js single-page application, using an OpenAPI-generated client to call the backend.
• Data (folder “Data/”) holds raw/preprocessed CSV/JSON experimental files and conventions.
File structure
Top-level directories:
• Data/ ‑- raw & preprocessed datasets, sample files, metadata.
• backend/ ‑- Python service:
– api.py, routers/, services/, models/, config, Alembic migrations under backend/alembic/
– analysis modules under ccfatigue/analysis/
– preprocessing scripts under backend/preprocessing/
– tests/ for pytest.
• frontend/ ‑- Vue.js app:
– public/ (static assets), src/ (components/, views/, store/, router/, backend/ generated OpenAPI client)
– build config (babel, vue.config.js), Dockerfile.
• docker-compose.yml, traefik.yaml, deploy/ for container orchestration & reverse-proxy.
Insights from README
• The README outlines “make dev-backend”, “make dev-frontend”, preprocessing, API regeneration, and links to a high-level flowchart (flowchart_CCFATIGUE.png).
• Key tech: PostgreSQL (configured via env secrets), Docker, FastAPI, Vue.js, OpenAPI.
Essential elements
• Do not omit the preprocessing pipeline (a separate step/service).
• Analysis engine and external data sources (ASTM CSVs).
• Reverse-proxy (Traefik), particularly in a production context.