boc/life-agents/real_data_pipeline_v2.py

#!/usr/bin/env python3
"""
LIFE Real Data Pipeline v2
Hämtar RIKTIG data från flera öppna källor
"""
import requests
import sqlite3
import json
from datetime import datetime

DB_PATH = "/home/bernt/.openclaw/workspace/rivp-pilot-1/rivp.db"

def get_db():
    conn = sqlite3.connect(DB_PATH)
    conn.row_factory = sqlite3.Row
    return conn

def fetch_smhi_temperature():
    """Hämta RIKTIG temperaturdata från SMHI"""
    station_id = "97530"  # Uppsala Flygplats
    url = f"https://opendata-download-metobs.smhi.se/api/version/latest/parameter/1/station/{station_id}/period/latest-hour/data.json"
    
    try:
        response = requests.get(url, timeout=30)
        if response.status_code == 200:
            data = response.json()
            values = data.get("value", [])
            if values:
                latest = values[-1]
                return {
                    "station": data["station"]["name"],
                    "temperature": float(latest["value"]),
                    "timestamp": latest["date"],
                    "unit": "celsius"
                }
    except Exception as e:
        print(f"SMHI temp error: {e}")
    return None

def fetch_smhi_precipitation():
    """Hämta RIKTIG nederbördsdata från SMHI"""
    station_id = "97530"
    url = f"https://opendata-download-metobs.smhi.se/api/version/latest/parameter/7/station/{station_id}/period/latest-day/data.json"
    
    try:
        response = requests.get(url, timeout=30)
        if response.status_code == 200:
            data = response.json()
            values = data.get("value", [])
            if values:
                total = sum(float(v["value"]) for v in values if float(v["value"]) >= 0)
                return {
                    "station": data["station"]["name"],
                    "precipitation_mm": total,
                    "measurements": len(values)
                }
    except Exception as e:
        print(f"SMHI precip error: {e}")
    return None

def fetch_smhi_wind():
    """Hämta RIKTIG vinddata från SMHI"""
    station_id = "97530"
    url = f"https://opendata-download-metobs.smhi.se/api/version/latest/parameter/4/station/{station_id}/period/latest-hour/data.json"
    
    try:
        response = requests.get(url, timeout=30)
        if response.status_code == 200:
            data = response.json()
            values = data.get("value", [])
            if values:
                latest = values[-1]
                return {
                    "station": data["station"]["name"],
                    "wind_speed": float(latest["value"]),
                    "unit": "m/s"
                }
    except Exception as e:
        print(f"SMHI wind error: {e}")
    return None

def fetch_osm_roads():
    """Hämta vägdata från OpenStreetMap"""
    query = """
    [out:json];
    area["name"="Uppsala"]->.searchArea;
    way["highway"~"motorway|trunk|primary|secondary"](area.searchArea);
    out body;
    """
    
    try:
        response = requests.post(
            "https://overpass-api.de/api/interpreter",
            data={"data": query},
            timeout=30
        )
        if response.status_code == 200:
            data = response.json()
            ways = [e for e in data.get("elements", []) if e.get("type") == "way"]
            return ways
    except Exception as e:
        print(f"OSM error: {e}")
    return []

def fetch_wikidata_sweden():
    """Hämta data om Sverige från Wikidata"""
    query = """
    SELECT ?item ?itemLabel WHERE {
      ?item wdt:P31 wd:Q34442.
      ?item wdt:P17 wd:Q34.
      SERVICE wikibase:label { bd:serviceParam wikibase:language "sv,en". }
    }
    LIMIT 10
    """
    
    try:
        response = requests.get(
            "https://query.wikidata.org/sparql",
            params={"query": query, "format": "json"},
            headers={"Accept": "application/sparql-results+json"},
            timeout=30
        )
        if response.status_code == 200:
            return response.json()
    except Exception as e:
        print(f"Wikidata error: {e}")
    return None

def save_weather_observations(temp_data, precip_data, wind_data):
    """Spara väderobservationer i databasen"""
    conn = get_db()
    c = conn.cursor()
    
    # Hitta vägar i Uppsala
    c.execute("SELECT id FROM roads WHERE county = 'Uppsala' LIMIT 10")
    roads = c.fetchall()
    
    count = 0
    for road in roads:
        # Temperatur
        if temp_data:
            severity = "high" if temp_data['temperature'] < 0 else "medium" if temp_data['temperature'] > 25 else "low"
            c.execute('''
                INSERT INTO observations 
                (road_id, observation_type, latitude, longitude, confidence, severity, detected_date, source)
                VALUES (?, ?, ?, ?, ?, ?, ?, ?)
            ''', (
                road[0],
                f"temperature_{temp_data['temperature']}c",
                59.85, 17.65,
                0.95,
                severity,
                datetime.now().strftime('%Y-%m-%d'),
                'smhi_real'
            ))
            count += 1
        
        # Nederbörd
        if precip_data and precip_data['precipitation_mm'] > 0:
            severity = "high" if precip_data['precipitation_mm'] > 10 else "medium" if precip_data['precipitation_mm'] > 2 else "low"
            c.execute('''
                INSERT INTO observations 
                (road_id, observation_type, latitude, longitude, confidence, severity, detected_date, source)
                VALUES (?, ?, ?, ?, ?, ?, ?, ?)
            ''', (
                road[0],
                f"precipitation_{precip_data['precipitation_mm']}mm",
                59.85, 17.65,
                0.9,
                severity,
                datetime.now().strftime('%Y-%m-%d'),
                'smhi_real'
            ))
            count += 1
        
        # Vind
        if wind_data:
            severity = "high" if wind_data['wind_speed'] > 15 else "medium" if wind_data['wind_speed'] > 10 else "low"
            c.execute('''
                INSERT INTO observations 
                (road_id, observation_type, latitude, longitude, confidence, severity, detected_date, source)
                VALUES (?, ?, ?, ?, ?, ?, ?, ?)
            ''', (
                road[0],
                f"wind_{wind_data['wind_speed']}ms",
                59.85, 17.65,
                0.9,
                severity,
                datetime.now().strftime('%Y-%m-%d'),
                'smhi_real'
            ))
            count += 1
    
    conn.commit()
    conn.close()
    return count

def run_real_pipeline():
    """Kör pipeline med RIKTIG data"""
    print(f"[{datetime.now().isoformat()}] Fetching REAL data...")
    
    # 1. Hämta riktig väderdata
    temp = fetch_smhi_temperature()
    precip = fetch_smhi_precipitation()
    wind = fetch_smhi_wind()
    
    if temp:
        print(f"  SMHI Temp: {temp['temperature']}°C at {temp['station']}")
    if precip:
        print(f"  SMHI Precip: {precip['precipitation_mm']}mm")
    if wind:
        print(f"  SMHI Wind: {wind['wind_speed']} m/s")
    
    # 2. Spara i databasen
    saved = save_weather_observations(temp, precip, wind)
    print(f"  Saved {saved} observations")
    
    # 3. Hämta vägar från OSM
    roads = fetch_osm_roads()
    if roads:
        print(f"  OSM Roads: {len(roads)} found")
    
    print(f"[{datetime.now().isoformat()}] Real data pipeline complete")

if __name__ == "__main__":
    print("="*60)
    print("LIFE REAL DATA PIPELINE v2")
    print("="*60)
    run_real_pipeline()
    print("="*60)