scripts/cache_maharashtra_districts_geojson.py

import json
import os
import random
import sys
import time
import urllib.error
import urllib.parse
import urllib.request

DISTRICTS = [
    "Kolhapur",
    "Sangli",
    "Satara",
    "Pune",
    "Ahmednagar",
    "Nashik",
    "Aurangabad",
    "Thane",
    "Raigad",
    "Ratnagiri",
]

# Approximate bounding boxes for Maharashtra districts (lon_min, lat_min, lon_max, lat_max)
# Used as fallback when Nominatim is rate-limited
DISTRICT_BBOXES = {
    "Kolhapur": (73.6, 16.0, 74.8, 17.2),
    "Sangli": (73.8, 16.4, 75.0, 17.2),
    "Satara": (73.3, 17.0, 74.5, 18.0),
    "Pune": (73.2, 18.0, 74.5, 19.0),
    "Ahmednagar": (74.0, 18.5, 75.5, 19.5),
    "Nashik": (73.0, 19.5, 74.5, 20.5),
    "Aurangabad": (74.5, 19.0, 76.0, 20.0),
    "Thane": (72.8, 19.0, 73.6, 20.0),
    "Raigad": (72.8, 18.0, 73.5, 18.8),
    "Ratnagiri": (72.8, 16.5, 73.8, 17.5),
}


def _request_json(url: str, timeout: int = 10) -> object:
    req = urllib.request.Request(
        url,
        headers={
            "Accept": "application/json",
            "User-Agent": "wrd-insight-dashboard/1.0 (cache script; contact: local-demo)",
        },
    )
    with urllib.request.urlopen(req, timeout=timeout) as resp:
        body = resp.read().decode("utf-8")
        return json.loads(body)


def fetch_district_geojson_from_nominatim(district: str) -> dict | None:
    """Try to fetch district boundary from Nominatim. Returns None on any error."""
    params = {
        "county": district,
        "state": "Maharashtra",
        "country": "India",
        "polygon_geojson": 1,
        "format": "json",
    }
    url = "https://nominatim.openstreetmap.org/search?" + urllib.parse.urlencode(params)
    try:
        data = _request_json(url, timeout=10)
    except Exception:
        return None

    if isinstance(data, list) and data and isinstance(data[0], dict) and data[0].get("geojson"):
        geo = data[0]["geojson"]
        return {
            "type": "Feature",
            "properties": {"name": district, "source": "nominatim"},
            "geometry": geo,
        }
    return None


def create_bbox_polygon(district: str) -> dict:
    """Create a simple rectangular polygon from bounding box as fallback."""
    bbox = DISTRICT_BBOXES.get(district)
    if not bbox:
        # Default to a placeholder in central Maharashtra
        bbox = (73.5, 18.0, 75.0, 19.5)
    
    lon_min, lat_min, lon_max, lat_max = bbox
    # Create a simple rectangular polygon (5 points to close the ring)
    coords = [
        [lon_min, lat_min],
        [lon_max, lat_min],
        [lon_max, lat_max],
        [lon_min, lat_max],
        [lon_min, lat_min],  # Close the ring
    ]
    return {
        "type": "Feature",
        "properties": {"name": district, "source": "bbox_fallback"},
        "geometry": {"type": "Polygon", "coordinates": [coords]},
    }


def main() -> int:
    repo_root = os.path.abspath(os.path.join(os.path.dirname(__file__), ".."))
    out_path = os.path.join(repo_root, "public", "data", "maharashtra-districts.geojson")

    os.makedirs(os.path.dirname(out_path), exist_ok=True)

    features: list[dict] = []
    nominatim_success = 0
    fallback_used = 0

    print("Fetching Maharashtra district boundaries...")
    print("Strategy: Try Nominatim first (quick timeout), use bbox fallback if rate-limited.\n")

    for district in DISTRICTS:
        # Try Nominatim with just 1 attempt (quick)
        feature = fetch_district_geojson_from_nominatim(district)
        
        if feature:
            features.append(feature)
            nominatim_success += 1
            print(f"  ✓ {district}: fetched from Nominatim")
        else:
            # Use bounding box fallback immediately
            feature = create_bbox_polygon(district)
            features.append(feature)
            fallback_used += 1
            print(f"  ○ {district}: using bbox fallback (Nominatim unavailable)")
        
        # Minimal delay to be polite (Nominatim asks for 1 req/sec)
        time.sleep(1.0)

    # Write the FeatureCollection
    fc = {"type": "FeatureCollection", "features": features}
    with open(out_path, "w", encoding="utf-8") as f:
        json.dump(fc, f, ensure_ascii=False, indent=2)

    print(f"\n✓ Wrote {len(features)} district features to: {out_path}")
    print(f"  - From Nominatim: {nominatim_success}")
    print(f"  - Bbox fallback: {fallback_used}")
    
    if fallback_used > 0:
        print(
            "\nNote: Some districts use approximate bounding boxes. "
            "For precise boundaries, re-run when Nominatim is available "
            "or download a proper Maharashtra district GeoJSON from a trusted source."
        )

    return 0


if __name__ == "__main__":
    raise SystemExit(main())