app.py

from fastapi import FastAPI, HTTPException, Query
from weather_pipeline import WeatherPipeline
from model import TrafficRiskModel
import os
from dotenv import load_dotenv

# Load environment variables
load_dotenv()

app = FastAPI(title="Weather-Traffic Risk API")

# Setup
WEATHER_API_KEY = os.getenv("WEATHER_API_KEY", "your_api_key_here")
weather_pipeline = WeatherPipeline(WEATHER_API_KEY)
risk_model = TrafficRiskModel()

@app.get("/")
async def root():
    return {
        "message": "Welcome to the Weather-Traffic Risk Prediction API",
        "endpoints": {
            "/weather/{city}": "Get weather data and traffic risk prediction for a city"
        }
    }

@app.get("/weather/{city}")
async def get_weather_risk(
    city: str, 
    lat: float = Query(None, description="Optional latitude for high precision"),
    lon: float = Query(None, description="Optional longitude for high precision")
):
    # 1. Fetch and process weather data (Lat/Lon takes priority for precision)
    data = weather_pipeline.get_weather(city=city, lat=lat, lon=lon)
    
    if "error" in data:
        raise HTTPException(status_code=400, detail=data["error"])
    
    # 2. Predict Traffic Risk using ML model
    risk_level, confidence = risk_model.predict(
        data["temp"], 
        data["rain"], 
        data["condition"]
    )
    
    # 3. Combine results in the FINAL God-Level format
    result = {
        "city": city,
        "timestamp": data.get("timestamp"),
        "source": data.get("source"),
        "coordinates": {"lat": lat, "lon": lon} if lat and lon else "Default City Center",
        "weather": {
            "actual_temp": data["temp"],
            "feels_like": data["feels_like"],
            "rainfall_mm": data["rain"],
            "condition": data["condition"],
            "condition_raw": data["condition_raw"]
        },
        "derived_weather_metrics": data["derived_metrics"],
        "ml_prediction": {
            "traffic_risk_level": risk_level,
            "confidence": round(float(confidence), 2),
            "status_color": get_risk_color(risk_level)
        }
    }
    
    return result

def get_risk_color(risk_level):
    mapping = {
        "Low": "🟢 Green",
        "Medium": "🟡 Yellow",
        "High": "🔴 Red"
    }
    return mapping.get(risk_level, "⚪ Unknown")

if __name__ == "__main__":
    import uvicorn
    # Train model on startup if needed
    if not risk_model.load():
        print("Training model...")
        risk_model.train()
    
    uvicorn.run(app, host="0.0.0.0", port=8000)