# Music Analyser

A personal analytics dashboard for your music listening habits, powered by Python, FastAPI, and Google Gemini AI.

## Project Structure

- `backend/`: FastAPI backend for data ingestion and API.
  - `app/ingest.py`: Background worker that polls Spotify.
  - `app/services/`: Logic for Spotify and Gemini APIs.
  - `app/models.py`: Database schema (Tracks, PlayHistory).
- `frontend/`: (Coming Soon) React/Vite frontend.

## Getting Started

### Prerequisites

- Docker & Docker Compose (optional, for containerization)
- Python 3.11+ (for local dev)
- A Spotify Developer App (Client ID & Secret)
- A Google Gemini API Key

### 1. Setup Environment Variables

Create a `.env` file in the `backend/` directory:

```bash
SPOTIFY_CLIENT_ID="your_client_id"
SPOTIFY_CLIENT_SECRET="your_client_secret"
SPOTIFY_REFRESH_TOKEN="your_refresh_token"
GEMINI_API_KEY="your_gemini_key"
```

To get the `SPOTIFY_REFRESH_TOKEN`, run the helper script:

```bash
python backend/scripts/get_refresh_token.py
```

### 2. Run Locally

Install dependencies:

```bash
cd backend
pip install -r requirements.txt
```

Run the server:

```bash
uvicorn app.main:app --reload
```

The API will be available at `http://localhost:8000`.

### 3. Run Ingestion (Manually)

You can trigger the ingestion process via the API:

```bash
curl -X POST http://localhost:8000/trigger-ingest
```

Or run the ingestion logic directly via python shell (see `app/ingest.py`).

### 4. Docker Build

To build the image locally:

```bash
docker build -t music-analyser-backend ./backend
```