utils.py

from collections import Counter
import pandas as pd
import numpy as np
from scipy.spatial import cKDTree
import requests

import asyncio


CATEGORY_MAPPING = {
    # --- Mapped to 'supermarket' ---
    'supermarket': 'supermarket',
    'swalayan': 'supermarket', # Indonesian for supermarket/grocery
    'greengrocer': 'supermarket',
    'grocery': 'supermarket',
    'frozen_food': 'supermarket',
    'wholesale': 'supermarket',
    'rice': 'supermarket',
    'water;rice': 'supermarket',
    'butcher': 'supermarket',
    'dairy': 'supermarket',
    'beverages': 'supermarket',

    # --- Mapped to 'convenience_store' ---
    'convenience': 'convenience_store',
    'kiosk': 'convenience_store',
    'variety_store': 'convenience_store',
    
    # --- Mapped to 'shopping_mall' ---
    'mall': 'shopping_mall',
    'taman anggrek': 'shopping_mall', # A known mall name, assuming it's a category match
    
    # --- Mapped to 'electronics_store' ---
    'electronics': 'electronics_store',
    'mobile_phone': 'electronics_store',
    'computer': 'electronics_store',
    'hifi': 'electronics_store',
    'radiotechnics': 'electronics_store',
    'device_charging_station': 'electronics_store',
    
    # --- Mapped to 'clothing_store' ---
    'clothes': 'clothing_store',
    'shoes': 'clothing_store',
    'tailor': 'clothing_store',
    'boutique': 'clothing_store',
    'fashion_accessories': 'clothing_store',
    'second_hand': 'clothing_store',
    'bag': 'clothing_store',
    
    # --- Mapped to 'jewelry_store' ---
    'jewelry': 'jewelry_store',
    'watches': 'jewelry_store',

    # --- Mapped to 'bookstore' ---
    'books': 'bookstore',
    'copyshop': 'bookstore', # Often linked to bookstores/stationery
    'stationery': 'bookstore',

    # --- Mapped to 'department_store' ---
    'department_store': 'department_store',
    
    # --- Mapped to 'cafe' ---
    'cafe': 'cafe',
    'coffee': 'cafe',
    'tea': 'cafe',
    'fast_food;cafe': 'cafe', # Primary includes cafe
    'internet_cafe': 'cafe',
    'biergarten': 'cafe', # Often functions as a casual cafe/bar space
    
    # --- Mapped to 'restaurant' ---
    'restaurant': 'restaurant',
    'nightclub;restaurant': 'restaurant', # Primary includes restaurant
    'seafood': 'restaurant', # Specific type of restaurant
    'food_court': 'restaurant', # Collection of eating places
    'outdoor_seating': 'restaurant', # Implies a place that serves food
    
    # --- Mapped to 'fast_food' ---
    'fast_food': 'fast_food',
    'snack': 'fast_food', # Closest general food item
    'deli': 'fast_food', # Quick-service food counter
    
    # --- Mapped to 'bakery' ---
    'bakery': 'bakery',
    'confectionery': 'bakery',
    'ice_cream': 'bakery',
    'pastry': 'bakery',
    'pastry;seafood': 'bakery', # Primary includes pastry
    
    # --- Mapped to 'clinic' ---
    'clinic': 'clinic',
    'midwife': 'clinic',
    'bidan': 'clinic', # Indonesian for midwife
    'posyandu': 'clinic', # Indonesian for community health post
    'medical_supply': 'clinic',
    'doctors': 'clinic',
    'veterinary': 'clinic',

    # --- Mapped to 'dentist' ---
    'dentist': 'dentist',
    
    # --- Mapped to 'hospital' ---
    'hospital': 'hospital',
    'nursing_home': 'hospital',
    'mortuary': 'hospital',
    'funeral_hall': 'hospital',
    'crematorium': 'hospital',
    
    # --- Mapped to 'pharmacy' ---
    'pharmacy': 'pharmacy',
    'chemist': 'pharmacy',
    'herbalist': 'pharmacy',
    
    # --- Mapped to 'gym' ---
    'gym': 'gym', # Not in the original list, but 'fitness_centre' and 'sports_centre' are close
    'sports_centre': 'gym',
    'fitness_centre': 'gym',
    'fitness_station': 'gym',
    
    # --- Mapped to 'yoga' ---
    'yoga': 'yoga', # Not in the original list, so mapping will be None unless a similar tag exists
    
    # --- Mapped to 'school' ---
    'school': 'school',
    'kindergarten': 'school',
    'music_school': 'school',
    'driving_school': 'school',
    'prep_school': 'school',
    'tuition': 'school',
    'language_school': 'school',
    'dancing_school': 'school',

    # --- Mapped to 'college' ---
    'college': 'college',
    
    # --- Mapped to 'university' ---
    'university': 'university',
    
    # --- Mapped to 'hotel' ---
    'hotel': 'hotel',
    'motel': 'hotel',
    'guest_house': 'hotel',
    'resort': 'hotel',
    'beach_resort': 'hotel',
    'chalet': 'hotel',
    'apartment': 'hotel',
    'resort': 'hotel',

    # --- Mapped to 'hostel' ---
    'hostel': 'hostel',

    # --- Mapped to 'attraction' ---
    'attraction': 'attraction',
    'museum': 'attraction',
    'cinema': 'attraction',
    'arts_centre': 'attraction',
    'theme_park': 'attraction',
    'amusement_arcade': 'attraction',
    'aquarium': 'attraction',
    'water_park': 'attraction',
    'gallery': 'attraction',
    'theatre': 'attraction',
    'casino': 'attraction',
    'zoo': 'attraction',
    'planetarium': 'attraction',
    'events_venue': 'attraction',
    
    # --- Mapped to 'viewpoint' ---
    'viewpoint': 'viewpoint',
    'outpost': 'viewpoint',
    'observation_deck': 'viewpoint', # Not in list, but a close match to viewpoint
    
    # --- Mapped to 'bank' ---
    'bank': 'bank',
    'bureau_de_change': 'bank',
    
    # --- Mapped to 'atm' ---
    'atm': 'atm',
    'money_transfer': 'atm',
    'payment_centre': 'atm',
    'payment_terminal': 'atm',
    
    # --- Mapped to 'co_working' ---
    'office': 'co_working', # General office space, often includes co-working
    'studio': 'co_working', # Can be co-working
    
    # --- Mapped to 'charging_station' ---
    'charging_station': 'charging_station',
    'device_charging_station': 'charging_station',

    # --- Mapped to None (Unmatched) ---
    'information': None, 'nightclub': None, 'post_office': None, 'fuel': None, 
    'taxi': None, 'bar': None, 'parking': None, 'pawnbroker': None, 'yes': None, 
    'swimming_pool': None, 'motorcycle_parking': None, 'clock': None, 
    'hairdresser': None, 'community_centre': None, 'police': None, 
    'nature_reserve': None, 'artwork': None, 'playground': None, 
    'car_repair': None, 'car_wash': None, 'telephone': None, 'laundry': None, 
    'dry_cleaning': None, 'bicycle': None, 'bicycle_rental': None, 'park': None, 
    'pub': None, 'parking_entrance': None, 'fountain': None, 'basket': None, 
    'waste_disposal': None, 'marketplace': None, 'townhall': None, 
    'doityourself': None, 'furniture': None, 'parking_space': None, 
    'grave_yard': None, 'place_of_worship': None, 'ferry_terminal': None, 
    'ngo': None, 'trade': None, 'florist': None, 'tyres': None, 'car': None, 
    'car_parts': None, 'toys': None, 'sports': None, 'kitchen': None, 'pet': None, 
    'travel_agency': None, 'interior_decoration': None, 'frame': None, 
    'outdoor': None, 'hardware': None, 'beauty': None, 'anime': None, 
    'motorcycle': None, 'bus_station': None, 'ticket': None, 'photo': None, 
    'library': None, 'fishing': None, 'waste_basket': None, 'scuba_diving': None, 
    'farm': None, 'social_facility': None, 'bicycle_parking': None, 'garden': None, 
    'shelter': None, 'bench': None, 'car_rental': None, 'gift': None, 
    'courthouse': None, 'toilets': None, 'pitch': None, 'drinking_water': None, 
    'grit_bin': None, 'massage': None, 'alcohol': None, 'optician': None, 
    'houseware': None, 'chocolate': None, 'ranger_station': None, 
    'water_point': None, 'smoking_area': None, 'wine': None, 'animal_breeding': None, 
    'marina': None, 'karaoke_box': None, 'vacant': None, 'vending_machine': None, 
    'telecommunication': None, 'post_box': None, 'wilderness_hut': None, 
    'cosmetics': None, 'childcare': None, 'distributor minuman dan makana': None, 
    'rental': None, 'bag': None, 'chair': None, 'motorcycle_repair': None, 
    'slipway': None, 'general': None, 'photo_booth': None, 'water': None, 
    'camera': None, 'gas': None, 'horse_riding': None, 'parcel_locker': None, 
    'training': None, 'ticket_validator': None, 'nutrition_supplements': None, 
    'carpet': None, 'antiques': None, 'recycling': None, 'archive': None, 
    'table': None, 'bicycle_repair': None, 'fixme': None, 'hot_tub': None, 
    'letter_box': None, 'musical_instrument': None, 'pet_grooming': None, 
    'flooring': None, 'art': None, 'printing': None, 'baby_goods': None, 
    'video': None, 'hearing_aids': None, 'e-cigarette': None, 'repair': None, 
    'soccer': None, 'pawn_shop': None, 'appliance': None, 'bed': None, 
    'sewing': None, 'party': None, 'pottery': None, 'indoor_play': None, 
    'video_games': None, 'boat_rental': None, 'rak besi': None, 'padel': None, 
    'golf_course': None, 'swimming': None, 'prison': None, 'tennis': None, 
    'basketball': None, 'car_pooling': None, 'posko banjir': None, 'common': None, 
    'multi': None, 'taman': None, 'rth': None, 'ruko': None, 'gedung': None, 
    'lapangan bulu tangkis dan sepak bola': None, 'rumah kantor': None, 
    'tempat pemakaman': None, 'tpu': None, 'camp_site': None, 'volleyball': None, 
    'climbing': None, 'sauna': None, 'bicycle_repair_station': None, 'no': None, 
    'perfumery': None, 'storage': None, 'animal_boarding': None, 'sports_hall': None, 
    'bicycle_wash': None, 'badminton': None, 'agrarian': None, 
    'nature_reserve;beach_resort': None, 'track': None, 'dog_park': None, 
    'leisure': None, 'motor': None, 'bandstand': None, 'picnic_site': None, 
    'waste_transfer_station': None, 'compressed_air': None, 
    'medical;cosmetics': None, 'conference_centre': None, 'security_booth': None, 
    'dressing_room': None, 'bbq': None
}


import math

def bbox_from_point(lat, lon, radius_m=500):
    lat_delta = radius_m / 111_320
    lon_delta = radius_m / (111_320 * math.cos(math.radians(lat)))

    south = lat - lat_delta
    north = lat + lat_delta
    west  = lon - lon_delta
    east  = lon + lon_delta

    return south, west, north, east


def compute_features(lat,lon, radius=500):

    BBOX = ",".join([str(x) for x in bbox_from_point(lat,lon,radius)])

    print("BBOX",BBOX)
    
    
    # BBOX = "-6.21,106.82,-6.15,106.85" # MONAS
    # BBOX = "-6.39,106.64,-6.10,106.97" # JAKARTA
    
    # 2. Define the Overpass Query
    # This query asks for all 'amenity=restaurant' POIs within the BBOX
    query = f"""
    [out:json][timeout:120];
    // Bounding box for the query
    (
      node["amenity"]({BBOX});
      node["tourism"]({BBOX});
      node["shop"]({BBOX});
    
      way["amenity"]({BBOX});
      way["tourism"]({BBOX});
      way["shop"]({BBOX});
    
      relation["amenity"]({BBOX});
      relation["tourism"]({BBOX});
      relation["shop"]({BBOX});
    
      node["leisure"]({BBOX});
      way["leisure"]({BBOX});
      relation["leisure"]({BBOX});
    
      node["sport"]({BBOX});
      way["sport"]({BBOX});
      relation["sport"]({BBOX});
    );
    out center;
    """
    
    
    # Note: 'out center' tells Overpass to provide a lat/lon for ways and relations as well.
    overpass_url = "http://overpass-api.de/api/interpreter"
    response = requests.post(overpass_url, data={'data': query})
    print(response.content)
    data = response.json()
    
    # 4. Convert to Pandas DataFrame
    elements = data.get('elements', [])
    
    # Extract POI attributes
    records = []
    for el in elements:
        record = {
            'id': el['id'],
            # Use 'lat'/'lon' for nodes, or 'center' for ways/relations
            'lat': el.get('lat', el.get('center', {}).get('lat')),
            'lon': el.get('lon', el.get('center', {}).get('lon')),
            # Get the primary category tag
            'category': el['tags'].get('amenity', el['type']), # Default to type if amenity not present
            'name': el['tags'].get('name', 'N/A'),
            'tourism':el['tags'].get('tourism'),
            'shop':el['tags'].get('shop'),
            'leisure':el['tags'].get('leisure'),
            'sport':el['tags'].get('sport')
        }
        records.append(record)
    
    # Create the DataFrame
    df_pois = pd.DataFrame(records)

    tourism_unused_mask = (
        df_pois["tourism"].notna()
        & ~df_pois["category"].isin(["node", "relation", "way"])
    )
    
    df_pois_new = df_pois[~tourism_unused_mask]
    mask_geom = df_pois_new["category"].isin(["node", "way", "relation"])
    
    df_pois_new.loc[mask_geom, "category"] = (
        df_pois_new.loc[mask_geom, "tourism"]
            .fillna(df_pois_new.loc[mask_geom, "shop"])
    )
    
    df_pois_new["category"] = (
        df_pois_new["category"]
        .fillna(df_pois_new["tourism"])
        .fillna(df_pois_new["shop"])
        .fillna(df_pois_new["leisure"])
        .fillna(df_pois_new["sport"])
    )
    
    df_pois_new["category"] = df_pois_new["category"].apply(lambda x:x.lower())
    df_pois_new["name"] = df_pois_new["name"].apply(lambda x:x.lower())
    
    print("\nUnique Categories:")
    print(df_pois_new['category'].unique())

    df_pois_new = df_pois_new[~(df_pois_new['category'].isin(['yes','fixme','no','shop','general']))]

    df_pois_new_cleaned = df_pois_new.copy()

    df_pois_new_cleaned["category"] = df_pois_new_cleaned["category"].apply(lambda x: CATEGORY_MAPPING.get(x))
    df_pois_new_cleaned.dropna(subset=['category'], inplace=True)

    res = df_pois_new_cleaned.groupby("category")["id"].count().to_dict()

    res['num_banks_in_radius'] = res.pop('atm',0)

    return res

    
    # print(f"Total POI found in the area: {len(df_pois)}")


# df_amenities = pd.read_csv("df_indonesia.csv").rename(
#     columns={"latitude":"lat", "longitude":"lon"}
# )
# df_banks = pd.read_csv("df_bank_indonesia.csv").rename(
#     columns={"latitude":"lat", "longitude":"lon"}
# )

# df_amenities["fsq_category_labels"] = df_amenities["fsq_category_labels"].apply(
#     lambda x: eval(x)
# )

# bank_coords = df_banks[['lat','lon']].values
# tree_banks = cKDTree(bank_coords)

# amenity_coords = df_amenities[['lat','lon']].values
# tree_amenities = cKDTree(amenity_coords)

# DATASET_COLUMNS = [
#     'Dining and Drinking', 'Community and Government', 'Retail',
#        'Business and Professional Services', 'Landmarks and Outdoors',
#        'Arts and Entertainment', 'Health and Medicine',
#        'Travel and Transportation', 'Sports and Recreation',
#        'Event'
# ]

# def compute_features(candidate_point, radius=0.005):

#     candidates = df_amenities[['lat','lon']].copy()
#     candidates['id'] = range(len(candidates))
    
#     bank_coords = df_atm[['lat','lon']].values
#     tree_banks = cKDTree(bank_coords)
    
#     amenity_coords = df_amenities[['lat','lon']].values
#     tree_amenities = cKDTree(amenity_coords)

#     lat, lon = candidate_point

#     # Banks
#     bank_idxs = tree_banks.query_ball_point([lat, lon], r=radius)

#     print("[BANK]", bank_idxs)
    
#     n_banks = len(bank_idxs)
#     if n_banks > 0:
#         neighbors = df_banks.iloc[bank_idxs]
#         mean_dist_banks = np.mean(np.sqrt((neighbors['lat']-lat)**2 + (neighbors['lon']-lon)**2))
#         min_dist_bank = np.min(np.sqrt((neighbors['lat']-lat)**2 + (neighbors['lon']-lon)**2))
#     else:
#         mean_dist_banks = radius
#         min_dist_bank = radius

#     # Amenities
#     amenity_idxs = tree_amenities.query_ball_point([lat, lon], r=radius)
#     amenities = df_amenities.iloc[amenity_idxs]

#     total_amenities = len(amenities)

#     # Flatten all category IDs
#     # for cats in amenities['fsq_category_labels']:
#     # all_category_ids = [cats[0].split(">")[0].strip() for cats in amenities['fsq_category_labels'] if len(cats)>0]
#     all_category_ids = amenities["category"].tolist()
#     category_diversity = len(set(all_category_ids))

#     features = {
#         'num_banks_in_radius': n_banks,
#         # 'mean_dist_banks': mean_dist_banks,
#         # 'min_dist_bank': min_dist_bank,
#         'total_amenities': total_amenities,
#         'category_diversity': category_diversity
#     }

#     # Count occurrences per category
#     print("[CATEGORIES]", all_category_ids)
#     count_per_category = Counter(all_category_ids)
#     for feat in DATASET_COLUMNS:
#         print("[FEAT]",feat)
#     # for cat, cnt in count_per_category.items():
#         features[f'num_{feat}'] = count_per_category.get(feat, 0)

    
#     # # Count occurrences of first category
#     # first_categories = [cats[0] for cats in amenities['fsq_category_ids'] if len(cats)>0]
#     # count_first_category = Counter(first_categories)
#     # for cat, cnt in count_first_category.items():
#     #     features[f'num_first_{cat}'] = cnt

#     return features