BI Platform (Single Folder, FastAPI + Jinja + PostgreSQL)

This revision consolidates backend and frontend into one deployable app folder (bi_app/) with FastAPI APIs + Jinja frontend templates.

Folder structure

.
├── bi_app/
│   ├── app/
│   │   ├── core/                # settings + security helpers
│   │   ├── db/                  # engine/session/base
│   │   ├── routers/             # auth, data, analysis, pages
│   │   ├── services/            # file extraction, pattern engine, AI layer
│   │   ├── static/              # CSS/JS assets for Jinja frontend
│   │   ├── templates/           # Jinja UI pages
│   │   ├── deps.py
│   │   ├── models.py
│   │   ├── schemas.py
│   │   └── main.py
│   ├── tests/
│   ├── Dockerfile
│   └── requirements.txt
├── docker-compose.yml
└── README.md

Why this change

• Use Jinja as requested for the frontend layer.
• Keep frontend + backend in a single folder for simpler deployment and ownership.
• Keep API contracts required by the original BI workflow.

Run with Docker

docker-compose up --build

• App: http://localhost:8000
• Health: http://localhost:8000/health
• API docs: http://localhost:8000/docs

Core API endpoints

Auth

• POST /api/v1/auth/signup
• POST /api/v1/auth/login
• GET /api/v1/me

Data

• POST /api/v1/upload
• GET /api/v1/uploads
• GET /api/v1/company/{id}
• GET /api/v1/summary/{id}

Analysis

• POST /api/v1/analyze
• GET /api/v1/relationship/{id}

Business flow

1. Sign up / Login from Jinja page (/)
2. Upload Company A and Target Company (/app)
3. Parsing + extraction + metadata storage
4. Pattern signature check
   • Known: rule normalization
   • Unknown/conflict: AI normalization + persist new pattern/rule
5. Save normalized companies + summaries
6. Run relationship analysis
7. View output in dashboard panel / history endpoint

Notes

• AI service is pluggable; mock mode is used without provider key.
• Upload supports DOC/DOCX/PDF/Excel/JSON/TXT + copy-paste text.
• Max upload size is 25MB.