"""Standalone OSM civic feature fetcher — dumps to data/civic_intel/osm_features/. No database needed. Reads city_bboxes.json and queries Overpass API for civic infrastructure: streetlights, traffic signals, bus stops, speed bumps, CCTV, zebra crossings, toll booths. Usage: python scripts/data/fetch_osm_civic_features.py python scripts/data/fetch_osm_civic_features.py --cities mumbai,chennai,delhi python scripts/data/fetch_osm_civic_features.py --all """ from __future__ import annotations import argparse import csv import json import os import sys import time from pathlib import Path try: import httpx except ImportError: print('[ERROR] httpx is required. Run: pip install httpx') sys.exit(1) PROJECT_ROOT = Path(__file__).resolve().parents[2] DATA_DIR = PROJECT_ROOT / 'data' / 'civic_intel' BBOXES_FILE = DATA_DIR / 'city_bboxes.json' OUTPUT_DIR = DATA_DIR / 'osm_features' OVERPASS_URL = os.getenv('OVERPASS_URL', 'https://overpass-api.de/api/interpreter') # OSM feature queries — maps feature_type to Overpass tag filters FEATURE_QUERIES = { 'streetlight': 'node["highway"="street_lamp"]', 'traffic_signal': 'node["highway"="traffic_signals"]', 'bus_stop': 'node["highway"="bus_stop"]', 'speed_bump': 'node["traffic_calming"="bump"]', 'cctv': 'node["man_made"="surveillance"]["surveillance:type"="camera"]', 'zebra_crossing': 'node["highway"="crossing"]["crossing"="zebra"]', 'toll_booth': 'node["barrier"="toll_booth"]', 'police_station': 'node["amenity"="police"]', 'fire_station': 'node["amenity"="fire_station"]', 'hospital': 'node["amenity"="hospital"]', 'fuel_station': 'node["amenity"="fuel"]', 'parking': 'node["amenity"="parking"]', } def build_overpass_query(bbox: list[float], feature_filter: str) -> str: """Build Overpass QL query for a bounding box.""" s, w, n, e = bbox return f'[out:json][timeout:60];({feature_filter}({s},{w},{n},{e}););out center;' def fetch_features_for_city( client: httpx.Client, city: str, bbox: list[float], feature_types: list[str] | None = None, ) -> dict[str, list[dict]]: """Fetch all feature types for a city.""" results: dict[str, list[dict]] = {} types_to_fetch = feature_types or list(FEATURE_QUERIES.keys()) for ftype in types_to_fetch: if ftype not in FEATURE_QUERIES: print(f' ⚠ Unknown feature type: {ftype}') continue query = build_overpass_query(bbox, FEATURE_QUERIES[ftype]) try: resp = client.get( OVERPASS_URL, params={'data': query}, headers={'User-Agent': 'SafeVixAI-CivicIntel/1.0'}, timeout=90, ) resp.raise_for_status() data = resp.json() elements = data.get('elements', []) features = [] for el in elements: lat = el.get('lat') or el.get('center', {}).get('lat') lon = el.get('lon') or el.get('center', {}).get('lon') if lat and lon: features.append({ 'osm_id': el.get('id'), 'lat': round(lat, 6), 'lon': round(lon, 6), 'feature_type': ftype, 'city': city, 'tags': json.dumps(el.get('tags', {})), }) results[ftype] = features print(f' ✓ {city}/{ftype}: {len(features)} features') # Rate limit: 1 request per second (Overpass courtesy) time.sleep(1.2) except httpx.TimeoutException: print(f' ✗ {city}/{ftype}: TIMEOUT (bbox may be too large)') results[ftype] = [] except Exception as exc: print(f' ✗ {city}/{ftype}: {exc}') results[ftype] = [] return results def save_features(city: str, features: dict[str, list[dict]]) -> dict[str, int]: """Save features to CSV files, one per feature type.""" OUTPUT_DIR.mkdir(parents=True, exist_ok=True) counts = {} for ftype, items in features.items(): counts[ftype] = len(items) if not items: continue # Resolve and validate the target file path to prevent path traversal resolved_output_dir = os.path.realpath(str(OUTPUT_DIR)) outfile = os.path.realpath(str(OUTPUT_DIR / f'{city}_{ftype}.csv')) if not outfile.startswith(resolved_output_dir + os.sep): raise ValueError(f"Path traversal detected: {outfile} is outside of {resolved_output_dir}") fieldnames = ['osm_id', 'lat', 'lon', 'feature_type', 'city', 'tags'] with open(outfile, 'w', newline='', encoding='utf-8') as f: writer = csv.DictWriter(f, fieldnames=fieldnames) writer.writeheader() writer.writerows(items) return counts def main(): parser = argparse.ArgumentParser(description='Fetch OSM civic features for Indian cities') parser.add_argument('--cities', type=str, help='Comma-separated city names (e.g., mumbai,chennai)') parser.add_argument('--all', action='store_true', help='Fetch for all cities in city_bboxes.json') parser.add_argument('--features', type=str, help='Comma-separated feature types to fetch') args = parser.parse_args() # Load city bounding boxes if not BBOXES_FILE.exists(): print(f'[ERROR] City bboxes file not found: {BBOXES_FILE}') sys.exit(1) with open(BBOXES_FILE, 'r', encoding='utf-8') as f: raw_bboxes = json.load(f) # city_bboxes.json has nested structure: {"metros": {"mumbai": {"bbox": [...], ...}}} metro_data = raw_bboxes.get('metros', raw_bboxes) all_bboxes = {} for city_name, city_info in metro_data.items(): if isinstance(city_info, dict) and 'bbox' in city_info: all_bboxes[city_name] = city_info['bbox'] elif isinstance(city_info, list): all_bboxes[city_name] = city_info # Determine which cities to process if args.cities: city_names = [c.strip().lower() for c in args.cities.split(',')] elif args.all: city_names = list(all_bboxes.keys()) else: # Default: top 10 metro cities top_metros = ['mumbai', 'delhi', 'chennai', 'kolkata', 'bengaluru', 'hyderabad', 'ahmedabad', 'pune', 'jaipur', 'lucknow'] city_names = [c for c in top_metros if c in all_bboxes] feature_types = [f.strip() for f in args.features.split(',')] if args.features else None print(f'\n╔══════════════════════════════════════════╗') print(f'║ SafeVixAI OSM Civic Feature Fetcher ║') print(f'╚══════════════════════════════════════════╝') print(f' Cities: {len(city_names)}') print(f' Features: {", ".join(feature_types) if feature_types else "all"}') print(f' Output: {OUTPUT_DIR}') print(f' Overpass: {OVERPASS_URL}') print() summary: dict[str, dict[str, int]] = {} with httpx.Client() as client: for i, city in enumerate(city_names, 1): if city not in all_bboxes: print(f'[{i}/{len(city_names)}] ⚠ {city}: not in city_bboxes.json — skipping') continue bbox = all_bboxes[city] print(f'[{i}/{len(city_names)}] Fetching {city}...') features = fetch_features_for_city(client, city, bbox, feature_types) counts = save_features(city, features) summary[city] = counts # Extra pause between cities if i < len(city_names): time.sleep(2) # Save summary OUTPUT_DIR.mkdir(parents=True, exist_ok=True) summary_file = OUTPUT_DIR / 'features_summary.json' with open(summary_file, 'w', encoding='utf-8') as f: json.dump(summary, f, indent=2) # Print summary print(f'\n{"═" * 55}') print(f' SUMMARY') print(f'{"═" * 55}') total_all = 0 for city, counts in summary.items(): total_city = sum(counts.values()) total_all += total_city top_features = sorted(counts.items(), key=lambda x: x[1], reverse=True)[:3] top_str = ', '.join(f'{k}={v}' for k, v in top_features if v > 0) print(f' {city:20s} {total_city:>6,} features [{top_str}]') print(f'{"─" * 55}') print(f' {"TOTAL":20s} {total_all:>6,} features') print(f' Output: {OUTPUT_DIR}') print() if __name__ == '__main__': main()