scripts/scrape_seed.py

#!/usr/bin/env python3
"""
Gujarat Street-Grid Seed v3.0

Covers ALL Gujarat cities with a dense lat/lng grid (~800m spacing).
Every grid cell generates 3 workers + 1 store at random categories.
Result: from ANY point in any city, a 1km search finds workers nearby.

Usage:
  python scripts/scrape_seed.py                     # synthetic only
  python scripts/scrape_seed.py --places-api        # + Google Places enrichment
  python scripts/scrape_seed.py --density 0.5       # override grid spacing (km)
  python scripts/scrape_seed.py --no-clear          # keep existing data
"""

import os, sys, json, random, math, hashlib, re
from datetime import datetime, timedelta
from dotenv import load_dotenv
from pymongo import MongoClient, InsertOne
from pymongo.errors import BulkWriteError

load_dotenv(os.path.join(os.path.dirname(__file__), '..', '.env'))
MONGODB_URI = os.getenv('MONGODB_URI')
GOOGLE_PLACES_API_KEY = os.getenv('GOOGLE_PLACES_API_KEY')
DB_NAME = "fixly"

BATCH_SIZE = 1000
CLEAR_EXISTING = True
USE_PLACES_API = False
GRID_SPACING_KM = 0.8

DICE_BEAR = "https://api.dicebear.com/7.x/initials/png?seed=%s&backgroundType=gradientLinear&backgroundRotation=0,360"

LANGUAGES = ["Gujarati", "Hindi", "English"]

# ─── CITIES with bounding-box data ──────────────────────────────
# (center_lat, center_lng, width_km, height_km, spacing_km)
# spacing_km overrides the global GRID_SPACING_KM per city
CITIES = {
  "Ahmedabad":    (23.0225, 72.5714, 32, 24, 1.0),
  "Surat":        (21.1702, 72.8311, 22, 18, 1.0),
  "Vadodara":     (22.3072, 73.1812, 16, 12, 0.8),
  "Rajkot":       (22.3039, 70.8022, 14, 10, 0.8),
  "Bhavnagar":    (21.7645, 72.1519, 12, 10, 0.7),
  "Jamnagar":     (22.4707, 70.0577, 11, 9,  0.7),
  "Junagadh":     (21.5222, 70.4579, 9,  8,  0.6),
  "Gandhinagar":  (23.2156, 72.6369, 8,  6,  0.6),
  "Anand":        (22.5645, 72.9289, 8,  6,  0.6),
  "Nadiad":       (22.6916, 72.8617, 7,  6,  0.6),
  "Morbi":        (22.8245, 70.8337, 7,  6,  0.6),
  "Bharuch":      (21.7051, 72.9958, 7,  6,  0.6),
  "Navsari":      (20.9467, 72.9520, 7,  6,  0.6),
  "Valsad":       (20.6060, 72.9343, 7,  6,  0.6),
  "Mehsana":      (23.5880, 72.3698, 8,  6,  0.6),
  "Palanpur":     (24.1710, 72.4381, 7,  5,  0.5),
  "Porbandar":    (21.6417, 69.6293, 7,  6,  0.5),
  "Surendranagar":(22.7270, 71.6487, 7,  5,  0.5),
  "Bhuj":         (23.2420, 69.6669, 8,  6,  0.5),
  "Gandhidham":   (23.0853, 70.1194, 7,  5,  0.5),
  "Godhra":       (22.7780, 73.6148, 6,  5,  0.5),
  "Dahod":        (22.8450, 74.2600, 6,  5,  0.5),
  "Patan":        (23.8499, 72.1263, 6,  5,  0.5),
  "Himmatnagar":  (23.6000, 72.9500, 6,  5,  0.5),
  "Veraval":      (20.9085, 70.3623, 6,  5,  0.5),
  "Amreli":       (21.6111, 71.2421, 6,  5,  0.5),
  "Botad":        (22.1700, 71.6700, 5,  4,  0.5),
  "Gondal":       (21.9600, 70.8000, 5,  4,  0.5),
  "Jetpur":       (21.7500, 70.6200, 5,  4,  0.5),
  "Kalol":        (23.2430, 72.4960, 5,  4,  0.5),
  "Kheda":        (22.7520, 72.6850, 5,  4,  0.5),
  "Sidhpur":      (23.9160, 72.3720, 5,  4,  0.5),
  "Visnagar":     (23.6980, 72.5510, 5,  4,  0.5),
  "Unjha":        (23.8000, 72.4000, 5,  4,  0.5),
  "Deesa":        (24.2500, 72.1830, 5,  4,  0.5),
  "Dhoraji":      (21.7320, 70.4520, 4,  4,  0.5),
  "Upleta":       (21.7310, 70.2800, 4,  4,  0.5),
  "Mahuva":       (21.0830, 71.8000, 4,  4,  0.5),
  "Palitana":     (21.5250, 71.8230, 4,  4,  0.5),
  "Modasa":       (23.4670, 73.3000, 4,  4,  0.5),
  "Vapi":         (20.3710, 72.9040, 5,  4,  0.5),
  "Vijapur":      (23.5670, 72.7500, 4,  4,  0.5),
  "Kapadvanj":    (23.0200, 73.0700, 4,  4,  0.5),
  "Tharad":       (24.3900, 71.6200, 4,  4,  0.5),
  "Lunavada":     (23.1300, 73.6100, 4,  4,  0.5),
  "Chhota Udepur":(22.3000, 74.0100, 4,  4,  0.5),
}

# ─── Named areas/localities per city for realistic addresses ─────
CITY_AREAS = {
  "Ahmedabad": [
    "Satellite", "Bopal", "Maninagar", "Navrangpura", "Vastrapur",
    "Thaltej", "Paldi", "Naranpura", "Gota", "Chandkheda",
    "Isanpur", "Naroda", "Vejalpur", "Prahlad Nagar", "Bodakdev",
    "SG Highway", "Science City", "Ranip", "Sabarmati", "Odhav",
    "Vatva", "Nikol", "Bapunagar", "Gomtipur", "Raipur",
    "Kankaria", "Ellisbridge", "Shahibaug", "Asarwa", "Nava Vadaj",
  ],
  "Surat": [
    "Adajan", "Vesu", "Piplod", "Katargam", "Varachha",
    "Udhna", "Athwa", "Dumas", "City Light", "Althan",
    "Bhatar", "Amroli", "Palanpur Patia", "Sagrampura", "Nanpura",
    "Rander", "Mota Varachha", "Parvat Patia", "Kapodra", "Umarwada",
    "Yogi Chowk", "Ghod Dod Road", "Hajira", "Magdalla", "Sachin",
  ],
  "Vadodara": [
    "Alkapuri", "Gotri", "Manjalpur", "Karelibaug", "Akota",
    "Harni", "Waghodia", "Nizampura", "Sama", "Fatehgunj",
    "Makarpura", "Subhanpura", "Gorwa", "Dabhoi Road",
    "Raopura", "Mandal", "Tarsali", "Vasad", "Karjan", "Savli",
  ],
  "Rajkot": [
    "Kalawad Road", "Yagnik Road", "150 Feet Ring Road", "Race Course",
    "Kotecha Chowk", "Mavdi", "University Road", "Raiya Road",
    "Amin Marg", "Bedipara", "Gondal Road", "Dhebar Road",
    "Sadhu Vaswani Road", "Bhaktinagar", "Shapar", "Kuvadava Road",
    "Nana Mavdi", "Junction Plot", "Sadhuvaswani", "Madhapar",
    "Gokul", "Kothariya", "Bhavani", "Sanala Road",
  ],
  "Bhavnagar": [
    "Kaliyabid", "Nilambaug", "Krishna Nagar", "Sardarnagar",
    "Ghogh Circle", "Takhteshwar", "Vidyanagar", "Chitra", "Ratanpara",
    "Adarsh Nagar", "Shastrinagar", "Malvav", "Kumbharwada",
  ],
  "Jamnagar": [
    "Indira Gandhi Marg", "Darbargadh", "Palanpur", "Udyognagar",
    "Shanker Tekri", "Gurukul", "Vijay Nagar", "Digvijay Plot",
    "Vibhapar", "Lakhota", "Airport Road", "Bedis Road",
  ],
  "Junagadh": [
    "Joshipura", "Moti Baug", "MG Road", "Kalwa Chowk",
    "Zanzarda", "Bhavnath", "Sardar Baug", "Keshod Road",
    "Kadiya", "Satasi", "Vanthali", "Bhesan",
  ],
  "Gandhinagar": [
    "Sector 1", "Sector 2", "Sector 3", "Sector 4", "Sector 5",
    "Sector 6", "Sector 7", "Sector 8", "Sector 9", "Sector 10",
    "Sector 11", "Sector 12", "Sector 13", "Sector 14", "Sector 15",
    "Sector 16", "Sector 17", "Sector 18", "Sector 19", "Sector 20",
    "Sector 21", "Sector 22", "Sector 23", "Sector 24", "Sector 25",
    "Sector 26", "Sector 27", "Sector 28", "Sector 29", "Sector 30",
    "GIDC", "Infinite City", "Sargasan", "Kudasan", "Randesan",
    "Pethapur", "Raysan",
  ],
  "Anand": [
    "Vallabh Vidyanagar", "Borsad Road", "Anand Station Road",
    "Gamdi", "Bakrol", "Chikhodra", "Mogri", "Karamsad", "Vitthal Udyognagar",
  ],
  "Nadiad": [
    "Ghogha Cross Road", "Santram Road", "Mahagujarat Society",
    "Vadtal Road", "Bhatta", "Shahpur", "Barejadi", "Rameshwar Road",
    "Marine Lines", "Mahadev Faliya",
  ],
  "Morbi": [
    "Bagdana", "Wankaner Road", "Sanala Road", "Shapur",
    "Ghanshyam Nagar", "Shastri Nagar", "Tankara Road",
    "Bharatnagar", "Gundala", "Haripar",
  ],
  "Bharuch": [
    "Zadeshwar", "Ankleshwar", "Panchbhatti",
    "Station Road", "Vadia", "Netang", "Kosamdi", "Vadia",
  ],
  "Navsari": [
    "Chhapra Road", "Sanskar Society", "Canal Road",
    "Eru", "Jalalpore", "Vijalpore", "Mohan Nagar", "Dhulia",
  ],
  "Valsad": [
    "Station Road", "Tithal Road", "Halpati Vasahat",
    "Nana Khajod", "Chharwada", "Atul", "Dharampur", "Umargam",
  ],
  "Mehsana": [
    "Gulabpura", "Highway Road", "Radhanpur Road",
    "Vishnu Nagar", "Jagudan", "Modhera", "Kheralu", "Vadnagar",
    "Gojariya", "Jethal",
  ],
  "Palanpur": [
    "Highway Road", "Johari Bazar", "Gunjar",
    "Mithi Road", "Bhogal", "Ambaji Road", "Kankrej", "Dhanera Road",
  ],
  "Porbandar": [
    "Ghodbunder", "Mistry Plot", "Kutiyana Road",
    "Bhadrod", "Prabhas Patan", "Madhavpur", "Miyani",
  ],
  "Surendranagar": [
    "Wadhwan", "Chhipdi", "Highway Road",
    "Lakhtar Road", "Dharangadhra", "Limdi", "Muli Road",
    "Thangadh", "Halvad",
  ],
  "Bhuj": [
    "Hospital Road", "Madhapar", "Mundra Road",
    "Bhimasar", "Rapar", "Mirzapur", "Naranpur", "Kukma",
    "Lakadia", "Kera",
  ],
  "Gandhidham": [
    "Plot 1-380", "Kandla", "Sector 1-15",
    "Adipur", "Galpadar", "Gandhidham Station", "Shivaji Nagar",
  ],
  "Godhra": [
    "Kalol Road", "Station Road", "Halol Road",
    "Fatepura", "Dahod Road", "Khanpur", "Timba Road",
  ],
  "Dahod": [
    "Jhalod Road", "Station Road", "Fatehpura",
    "Chandwada", "Devgadh Baria Road", "Limkheda", "Singvad",
  ],
  "Patan": [
    "Chanasma Road", "Siddhpur Road", "Bhabhar",
    "Santalpur", "Radhanpur", "Sami", "Harij",
  ],
  "Himmatnagar": [
    "Talod Road", "Prantij", "Station Road",
    "Khedbrahma Road", "Vijaynagar", "Idar Road",
    "Vadali", "Bayad",
  ],
  "Veraval": [
    "Somnath", "Patanvav", "Station Road",
    "Prabhas Patan", "Kodinar", "Jalgaon", "Mul Dwarka",
  ],
  "Amreli": [
    "Baba Road", "Khimmat Nagar", "Chalala Road",
    "Dhari", "Savar Kundla", "Rajula", "Lathi", "Babra",
  ],
  "Botad": [
    "Ahmedabad Road", "Gadhada", "Station Road",
    "Rangpur", "Lathidad", "Ranpur", "Barvala",
  ],
  "Gondal": [
    "Rajkot Road", "Jetpur Road", "Station Road",
    "Nana Mavdi", "Bhakti Nagar", "Dhoraji Road", "Vadia",
  ],
  "Jetpur": [
    "Upleta Road", "Dhoraji Road", "Station Road",
    "Gadhka", "Chital", "Ventrapur", "Moti Vavdi",
  ],
  "Kalol": [
    "Station Road", "Market Road", "Santram Road",
    "Nagalpur", "Kadi Road", "Bajwa", "Dharoi Road",
  ],
  "Kheda": [
    "Station Road", "Borsad Road", "Kapadvanj Road",
    "Guruvaya", "Chhipdi", "Mahalet",
  ],
  "Sidhpur": [
    "Station Road", "Bazar Road", "Junagadh Road",
    "Ambali", "Galisana", "Bhandu",
  ],
  "Visnagar": [
    "Station Road", "Modhera Road", "Kheralu Road",
    "Vadnagar Road", "Siddhpur Road", "Kamalpur",
  ],
  "Unjha": [
    "Station Road", "Market Yard", "Siddhpur Road",
    "Jagudan", "Chanasma", "Khara", "Rampur",
  ],
  "Deesa": [
    "Station Road", "Highway Road", "Palanpur Road",
    "Dhanera Road", "Abu Road", "Mitha", "Bodana",
  ],
  "Dhoraji": [
    "Station Road", "Jetpur Road", "Upleta Road",
    "Gondal Road", "Keshod Road", "Kalavad Road",
  ],
  "Upleta": [
    "Station Road", "Dhoraji Road", "Jetpur Road",
    "Mahuva Road", "Kotharia", "Patanvav",
  ],
  "Mahuva": [
    "Station Road", "Palitana Road", "Talaja Road",
    "Bhavnagar Road", "Sihor Road", "Ghogha Road",
  ],
  "Palitana": [
    "Station Road", "Bhavnagar Road", "Mahuva Road",
    "Talaja Road", "Gadhada", "Gariyadhar",
  ],
  "Modasa": [
    "Station Road", "Himmatnagar Road", "Meghraj Road",
    "Bhadiadar", "Bhiloda", "Bavali",
  ],
  "Vapi": [
    "Station Road", "GIDC", "Daman Road",
    "Silvassa Road", "Pardi", "Bhilad", "Chharwada",
  ],
  "Vijapur": [
    "Station Road", "Mehsana Road", "Kadi Road",
    "Visnagar Road", "Palanpur Road", "Kheralu",
  ],
  "Kapadvanj": [
    "Station Road", "Kheda Road", "Modasa Road",
    "Lunavada", "Manipur", "Salap",
  ],
  "Tharad": [
    "Station Road", "Deesa Road", "Santalpur Road",
    "Vavi", "Dhanera", "Lakhni",
  ],
  "Lunavada": [
    "Station Road", "Godhra Road", "Modasa Road",
    "Balasinor", "Santrampur", "Kadana",
  ],
  "Chhota Udepur": [
    "Station Road", "Jetpur Road", "Kawant Road",
    "Naswadi", "Pavi Jetpur", "Bodeli",
  ],
}

CATEGORIES = {
  "plumbing": ["Pipe Repair", "Tap Installation", "Drain Cleaning", "Water Heater Service", "Bathroom Fitting"],
  "electrical": ["Wiring Repair", "Fan Installation", "MCB Box Repair", "Light Installation", "Smart Home Setup"],
  "cleaning": ["Deep Cleaning", "Kitchen Cleaning", "Office Cleaning", "Sofa & Carpet Cleaning", "Bathroom Scrubbing"],
  "carpentry": ["Furniture Repair", "Door Fitting", "Custom Furniture", "Kitchen Cabinet Work", "Wood Polishing"],
  "painting": ["Wall Painting", "Texture Finish", "Exterior Painting", "Waterproofing", "Furniture Spray Paint"],
  "ac": ["AC Service", "AC Repair", "Gas Refill", "AC Installation", "Duct Cleaning"],
  "salon": ["Haircut", "Beard Trim", "Hair Color", "Facial", "Bridal Makeup", "Manicure"],
  "vehicle": ["Bike Service", "Car Service", "Tyre Change", "Engine Repair", "Battery Replacement", "Denting & Painting"],
  "pest_control": ["Cockroach Treatment", "Mosquito Control", "Termite Treatment", "Bed Bug Treatment", "Rodent Control"],
  "packers_movers": ["House Shifting", "Office Relocation", "Loading-Unloading", "Vehicle Transport", "Packing Service"],
  "photography": ["Wedding Photography", "Portrait Shoot", "Event Coverage", "Product Photography", "Cinematography"],
  "catering": ["Home Catering", "Party Catering", "Wedding Catering", "Snacks Service", "Lunch Delivery Service"],
  "laundry": ["Wash & Fold", "Dry Cleaning", "Ironing Service", "Carpet Cleaning", "Curtain Wash"],
  "tailoring": ["Custom Stitching", "Alterations", "Designer Wear", "Uniform Stitching", "Leather Work"],
  "fitness": ["Personal Training", "Yoga Classes", "Gym Sessions", "Diet Planning", "Zumba Classes"],
  "tutoring": ["Home Tuition", "Online Classes", "Subject Tutoring", "Exam Prep", "Language Classes"],
  "home_renovation": ["Full Renovation", "Kitchen Remodel", "Bathroom Renovation", "Wall Demolition", "Flooring Work"],
  "interior_design": ["Home Interior", "Office Design", "Space Planning", "Furniture Layout", "Lighting Design"],
  "appliance_repair": ["Washing Machine Repair", "Fridge Repair", "Microwave Repair", "Water Purifier Service", "Geyser Repair"],
  "roofing": ["Waterproofing", "Roof Repair", "Terrace Sealing", "Membrane Installation", "Roof Coating"],
  "flooring": ["Tile Installation", "Marble Polishing", "Wood Flooring", "Vinyl Flooring", "Floor Grouting"],
  "welding": ["Gate Fabrication", "Railing Work", "Grill Making", "Structure Welding", "Aluminum Work"],
  "event_planning": ["Birthday Planning", "Wedding Planning", "Corporate Events", "Decor Service", "Catering Coordination"],
  "security": ["CCTV Installation", "Security Guard", "Biometric System", "Alarm Installation", "Intercom Setup"],
  "gardening": ["Lawn Mowing", "Garden Design", "Tree Trimming", "Plantation Work", "Irrigation Setup"],
  "mobile_repair": ["Screen Replacement", "Battery Replacement", "Software Fix", "Charging Port Repair", "Water Damage"],
  "computer_repair": ["Desktop Repair", "Laptop Service", "Data Recovery", "Virus Removal", "Upgrade Service"],
}

CATEGORY_LIST = list(CATEGORIES.keys())

MALE_NAMES = [
  "Arjun", "Rahul", "Karan", "Vikram", "Hardik", "Parth", "Dhruv", "Akash", "Jay", "Rohan",
  "Manish", "Bhavesh", "Hitesh", "Mihir", "Rakesh", "Dinesh", "Mahesh", "Suresh", "Nilesh",
  "Rajesh", "Prakash", "Deepak", "Sanjay", "Vijay", "Ajay", "Nikhil", "Amit", "Sunil", "Anil",
  "Kiran", "Mayur", "Chetan", "Bhavin", "Chirag", "Harsh", "Kaushik", "Dharmesh", "Jignesh",
  "Tejas", "Vishal", "Mukesh", "Ramesh", "Shailesh", "Jitendra", "Sandeep", "Devendra",
  "Bharat", "Ashish", "Dhaval", "Keyur", "Sachin", "Ravi", "Savan", "Yash", "Krupal",
  "Mitul", "Fenil", "Kishan", "Sagar", "Vatsal", "Harshad", "Kishor", "Pinakin", "Pravin",
  "Kalpesh", "Ashok", "Gaurang", "Bipin", "Nirav", "Jayesh", "Milan", "Tushar", "Jatin",
  "Yogesh", "Vinod", "Mohan", "Hasmukh", "Kantibhai", "Raman", "Shyam", "Tarun", "Umesh",
  "Rushabh", "Helly", "Pruthvi", "Shubham", "Mohit", "Krunal", "Ronak", "Ankur", "Brijesh",
  "Mitesh", "Parag", "Swapnil", "Akshay", "Pranav", "Kushal", "Darshan", "Siddharth",
]

FEMALE_NAMES = [
  "Neha", "Priya", "Meena", "Rina", "Komal", "Pooja", "Sheetal", "Anjali", "Divya", "Kavita",
  "Rupal", "Aarti", "Kiran", "Asha", "Deepa", "Hina", "Darshana", "Urvi", "Hetal", "Sejal",
  "Bhavna", "Rutvi", "Kruti", "Nidhi", "Mansi", "Disha", "Trupti", "Ridhi", "Smita", "Alpa",
  "Kinnari", "Shruti", "Reshma", "Jigna", "Mitali", "Krisha", "Jhanvi", "Heena", "Vaishali",
  "Mamta", "Rekha", "Jyoti", "Rashmi", "Kajal", "Pinal", "Sonal", "Nisha", "Pallavi",
  "Dipti", "Harsha", "Gauri", "Varsha", "Sangeeta", "Bhavika", "Hardika", "Niyati",
]

LAST_NAMES = [
  "Patel", "Shah", "Makwana", "Joshi", "Rathod", "Solanki", "Parmar", "Savani", "Desai", "Mehta",
  "Dave", "Trivedi", "Pandya", "Rawal", "Vyas", "Acharya", "Jani", "Thakkar", "Gajjar", "Bhavsar",
  "Vasoya", "Savaliya", "Ramani", "Dobariya", "Kotadiya", "Vaghasiya", "Pipaliya", "Gohil",
  "Jadav", "Vala", "Sarvaiya", "Rajput", "Chauhan", "Chavda", "Zala", "Padhiyar", "Rana",
  "Vankar", "Barot", "Koli", "Rathva", "Damor", "Ninama",
  "Bambhania", "Mevada", "Rank", "Mistry", "Suthar", "Luhar", "Kumbhar", "Vanzara",
]

STORE_PREFIXES = [
  "Shree", "Om", "Jay", "Mahadev", "Royal", "Modern", "Perfect", "National",
  "Super", "Metro", "City", "Star", "Prime", "Elite", "Supreme", "Apex",
  "Krishna", "Ganesh", "Shiv", "Shakti", "Laxmi", "Sai", "Guru", "Sagar",
]

STORE_SUFFIXES = [
  "Hardware", "Electricals", "Services", "Solutions", "Repair Center",
  "Home Service", "Trading Co", "Enterprises", "Agency", "Store",
  "Center", "Hub", "Point", "Corner", "House",
]

# Gujarati street / lane / road names for realistic addresses
STREET_NAMES = [
  "MG Road", "Station Road", "College Road", "Hospital Road", "Market Road",
  "Tower Road", "Circle Road", "Cross Roads", "Main Bazaar", "Gandhi Marg",
  "Nehru Marg", "Patel Marg", "Raj Marg", "School Road", "Temple Road",
  "Masjid Road", "Lake Road", "Garden Road", "Park Road", "Bus Stand Road",
  "Highway Road", "Railway Station Road", "Bypass Road", "Mall Road",
  "Post Office Road", "Bank Road", "Court Road", "Library Road",
  "Police Station Road", "Dargah Road", "Clock Tower Road", "Kacheri Road",
  "Bamboo Bazaar", "Main Market Road", "Grain Market Road",
  "Jalaram Marg", "Swaminarayan Marg", "Ambedkar Marg", "Tagore Marg",
  "Shahid Bhagat Singh Marg", "Mahatma Gandhi Marg", "Sardar Patel Marg",
  "Indira Gandhi Marg", "Subhash Marg", "Azad Marg", "Shyamal Cross Road",
  "Satellite Road", "Bopal Road", "Science City Road", "Airport Road",
  "Ring Road", "Sarkhej Road", "Narol Road", "Vatva Road",
  "Panchwati Road", "Rambaug Road", "Nilkanth Mahadev Road",
  "Vivekanand Road", "Surya Nagar Road", "Mahavir Nagar Road",
  "Shastri Nagar Road", "Lal Darwaja Road", "Kuber Society Road",
]

LANDMARKS = [
  "Hanuman Temple", "Bus Stand", "Railway Station", "Municipal Market",
  "Post Office", "Police Station", "Government Hospital", "Main Square",
  "Circle Garden", "Water Tank", "Overbridge", "Petrol Pump",
  "Dargah", "Church", "Masjid", "Municipal Garden",
  "Bus Depot", "ST Bus Stop", "Fire Station", "Court Building",
]

DESCRIPTION_TEMPLATES = {
  "plumbing": "Full-service plumbing store in {area}, {city}. Stocking pipes, fittings, bathroom fixtures, water heaters, and all plumbing essentials.",
  "electrical": "Complete electrical supply store in {area}, {city}. Wires, switches, MCBs, fans, lights, and electrical accessories.",
  "cleaning": "Cleaning supplies and equipment store in {area}, {city}. Chemicals, tools, vacuum cleaners, and professional cleaning products.",
  "carpentry": "Carpentry materials and hardware store in {area}, {city}. Wood, tools, fittings, kitchen components, and finishing supplies.",
  "painting": "Paint store in {area}, {city}. All brands, wide color range, brushes, rollers, thinners, and waterproofing solutions.",
  "ac": "AC sales and service center in {area}, {city}. Units, spare parts, tools, gas, and all cooling accessories.",
  "salon": "Salon supplies store in {area}, {city}. Products, tools, chairs, mirrors, and professional salon equipment.",
  "vehicle": "Auto parts and accessories store in {area}, {city}. Spares, tires, batteries, lubricants, and vehicle service tools.",
  "pest_control": "Pest control product store in {area}, {city}. Chemicals, sprays, traps, fumigation equipment, and safety gear.",
  "packers_movers": "Moving and packing supplies store in {area}, {city}. Boxes, tapes, bubble wrap, ropes, and moving equipment.",
  "photography": "Photography equipment store in {area}, {city}. Cameras, lenses, lights, backdrops, and studio accessories.",
  "catering": "Catering supplies store in {area}, {city}. Utensils, disposables, serving equipment, and party essentials.",
  "laundry": "Laundry and dry cleaning store in {area}, {city}. Wash, fold, iron, dry clean, and specialty fabric care.",
  "tailoring": "Tailoring shop in {area}, {city}. Fabrics, threads, accessories, and custom stitching services.",
  "fitness": "Fitness equipment store in {area}, {city}. Weights, mats, machines, supplements, and workout accessories.",
  "tutoring": "Educational resource center in {area}, {city}. Books, materials, worksheets, and learning aids for all subjects.",
  "home_renovation": "Home renovation materials store in {area}, {city}. Tiles, cement, paint, plumbing and electrical supplies.",
  "interior_design": "Interior solutions store in {area}, {city}. Fabrics, wallpapers, decor items, furniture, and lighting.",
  "appliance_repair": "Appliance repair center in {area}, {city}. Spare parts, tools, and service for all home appliances.",
  "roofing": "Roofing and waterproofing store in {area}, {city}. Waterproofing solutions, membranes, coatings, and sealants.",
  "flooring": "Flooring solutions store in {area}, {city}. Tiles, marble, wood, vinyl, and installation tools.",
  "welding": "Welding supplies store in {area}, {city}. Rods, machines, gas cylinders, safety gear, and fabrication tools.",
  "event_planning": "Event supply store in {area}, {city}. Decor, tents, lights, chairs, sound systems, and party supplies.",
  "security": "Security solutions store in {area}, {city}. Cameras, alarms, locks, biometric systems, and surveillance equipment.",
  "gardening": "Garden supplies store in {area}, {city}. Plants, seeds, pots, fertilizers, tools, and irrigation systems.",
  "mobile_repair": "Mobile accessories and repair store in {area}, {city}. Parts, tools, screens, batteries, and phone accessories.",
  "computer_repair": "Computer service center in {area}, {city}. Parts, accessories, cables, tools, and repair services.",
}

BIO_TEMPLATES = {
  "plumbing": "Expert plumber with {exp} years experience in {area}, {city}. Specializing in pipe repair, drainage, water heater installation, and bathroom fittings.",
  "electrical": "Certified electrician serving {area}, {city} for {exp} years. Expert in wiring, fan installation, switchboard repair, and smart home wiring.",
  "cleaning": "Professional cleaner serving {area}, {city} with {exp} years of experience. Deep cleaning, kitchen, office, and sofa cleaning services.",
  "carpentry": "Master carpenter with {exp} years in {area}, {city}. Custom furniture, door fitting, kitchen cabinets, and wood polishing.",
  "painting": "Experienced painter with {exp} years in {area}, {city}. Wall painting, texture finishes, waterproofing, and exterior work.",
  "ac": "AC specialist with {exp} years experience in {area}, {city}. Repair, service, gas refill, and installation of all AC brands.",
  "salon": "Professional beautician serving {area}, {city} for {exp} years. Haircuts, styling, facial, bridal makeup, and grooming services.",
  "vehicle": "Auto mechanic with {exp} years experience in {area}, {city}. Bike and car service, engine repair, tyre change, and denting.",
  "pest_control": "Pest control expert with {exp} years in {area}, {city}. Cockroach, termite, mosquito, bed bug, and rodent treatments.",
  "packers_movers": "Professional packer and mover with {exp} years in {area}, {city}. Safe and reliable shifting services for home and office.",
  "photography": "Professional photographer with {exp} years in {area}, {city}. Wedding, portrait, event, and product photography services.",
  "catering": "Experienced caterer serving {area}, {city} for {exp} years. Home, party, and wedding catering with delicious Gujarati cuisine.",
  "laundry": "Laundry professional with {exp} years in {area}, {city}. Wash, fold, iron, dry cleaning, and carpet cleaning services.",
  "tailoring": "Master tailor with {exp} years experience in {area}, {city}. Custom stitching, alterations, and designer wear for men and women.",
  "fitness": "Certified fitness trainer with {exp} years in {area}, {city}. Personal training, yoga, diet planning, and gym sessions.",
  "tutoring": "Experienced tutor with {exp} years teaching in {area}, {city}. Home tuition, exam prep, and subject coaching for all grades.",
  "home_renovation": "Home renovation expert with {exp} years in {area}, {city}. Full renovation, kitchen remodel, bathroom work, and flooring.",
  "interior_design": "Interior designer with {exp} years experience in {area}, {city}. Home interiors, office design, space planning, and decor.",
  "appliance_repair": "Appliance repair expert with {exp} years in {area}, {city}. Washing machine, fridge, microwave, and water purifier service.",
  "roofing": "Roofing specialist with {exp} years in {area}, {city}. Waterproofing, terrace sealing, roof repair, and coating services.",
  "flooring": "Flooring expert with {exp} years in {area}, {city}. Tile installation, marble polishing, wood and vinyl flooring work.",
  "welding": "Skilled welder with {exp} years experience in {area}, {city}. Gate fabrication, railing, grill making, and structural welding.",
  "event_planning": "Event planner with {exp} years in {area}, {city}. Birthday, wedding, corporate events, decoration, and coordination.",
  "security": "Security professional with {exp} years in {area}, {city}. CCTV installation, biometric systems, alarm setup, and security guard.",
  "gardening": "Gardening expert with {exp} years in {area}, {city}. Lawn care, garden design, tree trimming, and irrigation setup.",
  "mobile_repair": "Mobile repair technician with {exp} years in {area}, {city}. Screen, battery, software, charging port, and water damage repairs.",
  "computer_repair": "Computer repair expert with {exp} years in {area}, {city}. Desktop, laptop, data recovery, virus removal, and upgrades.",
}

TAGS_BY_CATEGORY = {
  "plumbing": ["pipe", "leak", "bathroom", "tap", "drain", "water", "heater", "fitting"],
  "electrical": ["wiring", "fan", "light", "MCB", "switch", "smart", "electric"],
  "cleaning": ["deep clean", "kitchen", "office", "sofa", "carpet", "bathroom"],
  "carpentry": ["furniture", "door", "cabinet", "wood", "kitchen", "custom"],
  "painting": ["wall paint", "texture", "exterior", "waterproof", "spray"],
  "ac": ["AC", "cooling", "gas refill", "repair", "installation"],
  "salon": ["haircut", "beard", "facial", "makeup", "grooming", "beauty"],
  "vehicle": ["bike", "car", "tyre", "engine", "battery", "service"],
  "pest_control": ["pest", "termite", "mosquito", "cockroach", "rodent", "fumigation"],
  "packers_movers": ["shifting", "relocation", "packing", "loading", "moving"],
  "photography": ["wedding", "portrait", "event", "product", "cinema"],
  "catering": ["catering", "party", "wedding", "food", "home delivery"],
  "laundry": ["wash", "dry clean", "iron", "fold", "carpet"],
  "tailoring": ["stitching", "alteration", "designer", "uniform", "custom"],
  "fitness": ["gym", "yoga", "trainer", "diet", "zumba", "workout"],
  "tutoring": ["tuition", "classes", "subject", "exam", "coaching"],
  "home_renovation": ["renovation", "remodel", "kitchen", "bathroom", "flooring"],
  "interior_design": ["interior", "design", "decor", "furniture", "space"],
  "appliance_repair": ["washing machine", "fridge", "microwave", "purifier", "geyser"],
  "roofing": ["roof", "waterproof", "terrace", "sealing", "coating"],
  "flooring": ["tiles", "marble", "wood", "vinyl", "grouting"],
  "welding": ["gate", "railing", "grill", "fabrication", "aluminum"],
  "event_planning": ["party", "wedding", "decoration", "event", "planning"],
  "security": ["CCTV", "alarm", "biometric", "security", "surveillance"],
  "gardening": ["garden", "lawn", "plants", "irrigation", "landscaping"],
  "mobile_repair": ["mobile", "screen", "battery", "software", "repair"],
  "computer_repair": ["computer", "laptop", "data", "virus", "upgrade"],
}

CERTIFICATIONS_BY_CATEGORY = {
  "plumbing": ["Plumbing Diploma", "Water Systems Certification", "Safety Training"],
  "electrical": ["Electrical License", "Wiring Certification", "Safety Training"],
  "cleaning": ["Deep Cleaning Certified", "Chemical Safety Training"],
  "carpentry": ["Carpentry Diploma", "Woodworking Certification"],
  "painting": ["Painting & Coating Certified", "Color Design Certification"],
  "ac": ["HVAC Certified", "AC Manufacturer Training", "Gas Handling License"],
  "salon": ["Beauty License", "Salon Management Certified", "Hair Styling Diploma"],
  "vehicle": ["Auto Mechanic Diploma", "EV Certification", "Safety Inspection License"],
  "pest_control": ["Pest Control License", "Chemical Handling Certified", "Fumigation License"],
  "packers_movers": ["Moving License", "Insurance Certified", "Safety Training"],
  "photography": ["Professional Photography Diploma", "Adobe Certified"],
  "catering": ["Food Safety Certified", "Hospitality Diploma", "FSSAI License"],
  "laundry": ["Dry Cleaning Certified", "Fabric Care Training"],
  "tailoring": ["Fashion Design Diploma", "Tailoring Certificate"],
  "fitness": ["Personal Trainer Certified", "Yoga Teacher Training", "Nutrition Certified"],
  "tutoring": ["Teaching License", "Subject Specialist Certified", "CTET Qualified"],
  "home_renovation": ["Contractor License", "Safety Certified", "Project Management"],
  "interior_design": ["Interior Design Degree", "AutoCAD Certified", "Space Planning"],
  "appliance_repair": ["Appliance Repair Certified", "Electronics Diploma"],
  "roofing": ["Roofing Contractor License", "Waterproofing Certified"],
  "flooring": ["Flooring Installation Certified", "Marble Polishing Training"],
  "welding": ["Welding Diploma", "Arc Welding Certified", "Safety Training"],
  "event_planning": ["Event Management Diploma", "Wedding Planning Certified"],
  "security": ["Security License", "CCTV Installation Certified", "Biometric Training"],
  "gardening": ["Horticulture Diploma", "Landscape Design Certified"],
  "mobile_repair": ["Mobile Repair Diploma", "Manufacturer Certified"],
  "computer_repair": ["CompTIA A+", "Hardware Diploma", "Networking Certified"],
}

# ─── Helpers ─────────────────────────────────────────────────

def random_phone():
  return f"+91{random.randint(6000000000, 9999999999)}"

def random_email(name):
  clean = re.sub(r'[^a-z0-9]', '', name.lower().replace(" ", "."))
  return f"{clean}{random.randint(1,999)}@gmail.com"

def random_rating():
  return round(random.uniform(3.5, 5.0), 1)

def dice_avatar(seed):
  return DICE_BEAR % seed.replace(" ", "%20")

def pick_languages():
  return random.sample(LANGUAGES, random.randint(1, 3))

def pick_tags(category):
  base = TAGS_BY_CATEGORY.get(category, [category])
  return random.sample(base, k=min(random.randint(2, 5), len(base)))

def pick_certifications(category):
  certs = CERTIFICATIONS_BY_CATEGORY.get(category, [])
  if not certs: return []
  return random.sample(certs, k=random.randint(1, min(2, len(certs))))

def generate_services(category, is_worker=True):
  services_list = CATEGORIES.get(category, ["General Service"])
  selected = random.sample(services_list, k=min(random.randint(2, 4), len(services_list)))
  result = []
  for s in selected:
    price = random.randint(150, 2500) if is_worker else random.randint(200, 8000)
    stype = random.choice(["per_hour", "per_job"]) if is_worker else "per_job"
    result.append({"name": s, "price": price, "type": stype})
  return result

def random_gujarati_name():
  gender = random.choice(["male", "female"])
  first = random.choice(MALE_NAMES if gender == "male" else FEMALE_NAMES)
  last = random.choice(LAST_NAMES)
  return f"{first} {last}", "male" if gender == "male" else "female"

def random_store_name(category, area):
  prefix = random.choice(STORE_PREFIXES)
  suffix = random.choice(STORE_SUFFIXES)
  base = f"{prefix} {suffix}"
  return random.choice([
    f"{base} {area}",
    f"{prefix} {category.title()} {suffix}",
    f"{area} {suffix}",
    f"{base}",
  ])

# ─── Spatial grid generation ────────────────────────────────

def km_to_deg(lat, km):
  """Convert km to degrees at given latitude."""
  dlat = km / 111.0
  dlng = km / (111.0 * math.cos(math.radians(lat)))
  return dlat, dlng

def generate_grid(lat_center, lng_center, width_km, height_km, spacing_km):
  """Generate grid points covering a city bounding box."""
  dlat, dlng = km_to_deg(lat_center, spacing_km)
  hw_km, hh_km = width_km / 2.0, height_km / 2.0

  lat_start = lat_center - km_to_deg(lat_center, hh_km)[0]
  lng_start = lng_center - km_to_deg(lat_center, hw_km)[1]

  lat_steps = max(1, int(height_km / spacing_km))
  lng_steps = max(1, int(width_km / spacing_km))

  jitter = spacing_km * 0.2
  jlat, jlng = km_to_deg(lat_center, jitter)

  points = []
  for i in range(lat_steps):
    for j in range(lng_steps):
      lat = lat_start + (i + 0.5) * dlat + random.uniform(-jlat, jlat)
      lng = lng_start + (j + 0.5) * dlng + random.uniform(-jlng, jlng)
      points.append((lat, lng))
  return points

def nearest_area(lat, lng, city):
  """Find the nearest named area for a grid point."""
  areas = CITY_AREAS.get(city, [city])
  city_data = CITIES.get(city)
  if not city_data or not areas:
    return city
  clat = city_data[0]
  clng = city_data[1]
  # Weight by area index so nearby areas cluster: use deterministic hash
  idx = int(abs(hash(f"{lat:.4f}{lng:.4f}")) % len(areas))
  return areas[idx]

def generate_address(city, area):
  street = random.choice(STREET_NAMES)
  num = random.randint(1, 150)
  landmark = random.choice(LANDMARKS)
  fmt = random.choice(["number_street", "near_landmark", "street_area"])
  if fmt == "number_street":
    return f"{num}, {street}, {area}, {city}, Gujarat"
  elif fmt == "near_landmark":
    street2 = random.choice(STREET_NAMES)
    return f"Near {landmark}, {street2}, {area}, {city}, Gujarat"
  else:
    return f"{street}, {area}, {city}, Gujarat"

# ─── Document builders ──────────────────────────────────────

def create_worker_doc(city, street_addr, area, lat, lng, category):
  name, gender = random_gujarati_name()
  exp = random.randint(1, 25)
  services = generate_services(category, is_worker=True)
  tags = pick_tags(category)
  certs = pick_certifications(category)
  langs = pick_languages()
  price = random.randint(150, 1200)
  rating = random_rating()
  reviews = random.randint(3, 800)
  jobs = random.randint(5, 3000)

  bio = BIO_TEMPLATES.get(category, "{category} specialist serving {area}, {city}.").format(
    exp=exp, area=area, city=city
  )

  return {
    "name": name,
    "email": random_email(name),
    "phone": random_phone(),
    "whatsapp": random_phone(),
    "avatar": dice_avatar(name),
    "city": city,
    "area": area,
    "category": category,
    "services": services,
    "bio": bio,
    "experience": exp,
    "rating": rating,
    "totalReviews": reviews,
    "totalJobs": jobs,
    "pricePerHour": price,
    "location": {"type": "Point", "coordinates": [lng, lat]},
    "address": street_addr,
    "available": random.random() > 0.15,
    "verified": random.random() > (0.45 if reviews > 50 else 0.7),
    "emergencyAvailable": random.random() > 0.75,
    "responseTime": random.choice([15, 30, 45, 60]),
    "languages": langs,
    "certifications": certs,
    "serviceAreas": [],
    "tags": tags,
    "gallery": [],
    "lastActiveAt": datetime.utcnow() - timedelta(minutes=random.randint(0, 43200)),
    "createdAt": datetime.utcnow() - timedelta(days=random.randint(0, 730)),
    "updatedAt": datetime.utcnow(),
  }

def create_store_doc(city, street_addr, area, lat, lng, category):
  name = random_store_name(category, area)
  services = generate_services(category, is_worker=False)
  tags = pick_tags(category)
  langs = pick_languages()
  rating = random_rating()
  reviews = random.randint(3, 500)

  description = DESCRIPTION_TEMPLATES.get(category, "{category} store in {area}, {city}.").format(
    category=category.title(), area=area, city=city
  )

  open_hours_list = [
    "9:00 AM - 9:00 PM", "10:00 AM - 8:00 PM", "9:30 AM - 7:30 PM",
    "8:00 AM - 10:00 PM", "10:00 AM - 9:00 PM", "9:00 AM - 6:00 PM",
  ]

  return {
    "name": name,
    "email": random_email(name),
    "phone": random_phone(),
    "whatsapp": random_phone(),
    "avatar": dice_avatar(name),
    "city": city,
    "area": area,
    "category": category,
    "services": services,
    "description": description,
    "rating": rating,
    "totalReviews": reviews,
    "location": {"type": "Point", "coordinates": [lng, lat]},
    "address": street_addr,
    "verified": random.random() > 0.4,
    "openHours": random.choice(open_hours_list),
    "emergencyAvailable": random.random() > 0.8,
    "languages": langs,
    "tags": tags,
    "gallery": [],
    "createdAt": datetime.utcnow() - timedelta(days=random.randint(0, 730)),
    "updatedAt": datetime.utcnow(),
  }

def bulk_insert(collection, docs):
  if not docs: return
  ops = []
  for doc in docs:
    ops.append(InsertOne(doc))
    if len(ops) >= BATCH_SIZE:
      try:
        collection.bulk_write(ops, ordered=False)
      except BulkWriteError:
        pass
      ops = []
  if ops:
    try:
      collection.bulk_write(ops, ordered=False)
    except BulkWriteError:
      pass

# ─── Fetch from Google Places API (optional enrichment) ─────

def fetch_places_api(city, area, category):
  if not GOOGLE_PLACES_API_KEY or not USE_PLACES_API:
    return []
  try:
    import requests
    term = f"{CATEGORIES[category][0]} in {area}, {city}, Gujarat"
    url = "https://maps.googleapis.com/maps/api/place/textsearch/json"
    params = {"query": term, "key": GOOGLE_PLACES_API_KEY, "region": "in", "language": "en"}
    resp = requests.get(url, params=params, timeout=10)
    data = resp.json()
    if data.get("status") != "OK":
      return []
    results = data.get("results", [])
    docs = []
    for place in results[:5]:
      lat = place["geometry"]["location"]["lat"]
      lng = place["geometry"]["location"]["lng"]
      name = place.get("name", "")
      address = place.get("formatted_address", f"{area}, {city}, Gujarat")
      place_id = place.get("place_id", "")
      rating = place.get("rating", random_rating())
      reviews = place.get("user_ratings_total", random.randint(5, 200))

      phone = ""
      try:
        detail_url = "https://maps.googleapis.com/maps/api/place/details/json"
        detail_params = {"place_id": place_id, "key": GOOGLE_PLACES_API_KEY,
                         "fields": "formatted_phone_number,international_phone_number"}
        detail_resp = requests.get(detail_url, params=detail_params, timeout=10)
        detail_data = detail_resp.json()
        phone = detail_data.get("result", {}).get("international_phone_number", "") or \
                detail_data.get("result", {}).get("formatted_phone_number", "")
      except:
        pass

      doc = {
        "name": name, "email": random_email(name),
        "phone": phone or random_phone(), "whatsapp": phone or random_phone(),
        "avatar": dice_avatar(name),
        "city": city, "area": area, "category": category,
        "services": generate_services(category),
        "rating": min(max(rating, 1.0), 5.0), "totalReviews": reviews,
        "location": {"type": "Point", "coordinates": [lng, lat]},
        "address": address, "verified": random.random() > 0.4,
        "languages": pick_languages(), "tags": pick_tags(category),
        "googlePlaceId": place_id,
        "bio": BIO_TEMPLATES.get(category, "").format(exp=random.randint(1,20), area=area, city=city),
        "experience": random.randint(1, 20), "totalJobs": random.randint(5, 2500),
        "pricePerHour": random.randint(150, 1200), "available": True,
        "emergencyAvailable": random.random() > 0.75, "responseTime": random.choice([15, 30, 45, 60]),
        "certifications": pick_certifications(category),
        "gallery": [],
        "lastActiveAt": datetime.utcnow() - timedelta(minutes=random.randint(0, 43200)),
        "createdAt": datetime.utcnow() - timedelta(days=random.randint(0, 730)),
        "updatedAt": datetime.utcnow(),
      }
      docs.append(doc)
    return docs
  except Exception as e:
    print(f"    ⚠️  Places API error: {e}")
    return []

# ─── Main seed ──────────────────────────────────────────────

def seed():
  print("=" * 65)
  print("  GUJARAT STREET-GRID SEED v3.0")
  print("  Spacing: {}km | Cities: {} | Categories: {}".format(
    GRID_SPACING_KM, len(CITIES), len(CATEGORIES)))
  print("=" * 65)

  if not MONGODB_URI:
    print("❌ MONGODB_URI not set in .env")
    sys.exit(1)

  google_available = bool(GOOGLE_PLACES_API_KEY)
  if google_available and USE_PLACES_API:
    print("✅ Google Places API enabled (enrichment)")
  else:
    print("ℹ️  Synthetic mode" + (" (Google API key found but --places-api not set)" if google_available else ""))

  client = MongoClient(MONGODB_URI)
  db = client[DB_NAME]
  workers_col = db.workers
  stores_col = db.stores

  if CLEAR_EXISTING:
    print("\n🗑  Clearing existing data...")
    w_del = workers_col.delete_many({}).deleted_count
    s_del = stores_col.delete_many({}).deleted_count
    print(f"   Removed {w_del} workers, {s_del} stores")

  # ── Build per-city grid ──
  print("\n📍 Generating grid points per city...")
  city_grids = {}
  total_points = 0
  for cname, (clat, clng, w, h, s) in sorted(CITIES.items()):
    pts = generate_grid(clat, clng, w, h, s)
    city_grids[cname] = pts
    total_points += len(pts)
    areas = CITY_AREAS.get(cname, [cname])
    print(f"   {cname}: {len(pts)} grid points × {len(areas)} areas")

  print(f"\n   Total grid points across all cities: {total_points}")
  print(f"   Target docs: ~{total_points * 4} ({total_points * 3} workers + {total_points} stores)")

  # ── Generate documents ──
  all_workers = []
  all_stores = []
  used_name_hashes = set()

  for cname, points in sorted(city_grids.items()):
    areas = CITY_AREAS.get(cname, [cname])
    print(f"\n{'='*40}")
    print(f"  {cname} ({len(points)} points)")
    print(f"{'='*40}")

    city_workers = []
    city_stores = []
    cat_tracker = {cat: {"workers": 0, "stores": 0} for cat in CATEGORY_LIST}

    # Try Places API for enrichment
    places_docs = []
    if USE_PLACES_API and google_available:
      for area_sample in random.sample(areas, min(3, len(areas))):
        for cat_sample in random.sample(CATEGORY_LIST, min(5, len(CATEGORY_LIST))):
          docs = fetch_places_api(cname, area_sample, cat_sample)
          places_docs.extend(docs)
          if len(places_docs) >= 50:
            break
        if len(places_docs) >= 50:
          break
      for doc in places_docs:
        h = hash(doc["name"]) % 10**12
        if h not in used_name_hashes:
          used_name_hashes.add(h)
          city_workers.append(doc)
          cat_tracker[doc["category"]]["workers"] += 1

    for idx, (lat, lng) in enumerate(points):
      area = nearest_area(lat, lng, cname)
      street_addr = generate_address(cname, area)

      # 3 workers, 1 store per grid point
      worker_cats = random.sample(CATEGORY_LIST, 3)
      for wcat in worker_cats:
        w = create_worker_doc(cname, street_addr, area, lat, lng, wcat)
        h = hash(w["name"]) % 10**12
        if h not in used_name_hashes:
          used_name_hashes.add(h)
          city_workers.append(w)
          cat_tracker[wcat]["workers"] += 1

      scat = random.choice(CATEGORY_LIST)
      s = create_store_doc(cname, street_addr, area, lat, lng, scat)
      h = hash(s["name"]) % 10**12
      if h not in used_name_hashes:
        used_name_hashes.add(h)
        city_stores.append(s)
        cat_tracker[scat]["stores"] += 1

    # Log per-city coverage
    cats_covered_w = sum(1 for v in cat_tracker.values() if v["workers"] > 0)
    cats_covered_s = sum(1 for v in cat_tracker.values() if v["stores"] > 0)
    print(f"   Workers: {len(city_workers)} (covers {cats_covered_w}/{len(CATEGORY_LIST)} categories)")
    print(f"   Stores:  {len(city_stores)} (covers {cats_covered_s}/{len(CATEGORY_LIST)} categories)")

    all_workers.extend(city_workers)
    all_stores.extend(city_stores)

  # ── Insert ──
  print(f"\n{'='*40}")
  print(f"  INSERTING DATA")
  print(f"{'='*40}")
  print(f"   Workers: {len(all_workers)}")
  print(f"   Stores:  {len(all_stores)}")

  if all_workers:
    print("   Writing workers...")
    bulk_insert(workers_col, all_workers)
  if all_stores:
    print("   Writing stores...")
    bulk_insert(stores_col, all_stores)

  # ── Indexes ──
  print(f"\n📐 Ensuring indexes...")
  for col in [workers_col, stores_col]:
    existing = col.index_information()
    wanted = {"location_2dsphere", "category_1", "city_1_area_1",
              "emergencyAvailable_1", "tags_1", "pricePerHour_1"}
    for idx in wanted:
      if idx not in existing:
        field = idx.replace("_1", "").replace("_2dsphere", "").replace("_1_-1", "")
        if "2dsphere" in idx:
          col.create_index([(field, "2dsphere")])
        else:
          col.create_index([(field, 1)])
  print("   ✅ Done")

  # ── Summary ──
  actual_workers = workers_col.count_documents({})
  actual_stores = stores_col.count_documents({})
  print(f"\n{'='*65}")
  print(f"  SEED COMPLETE")
  print(f"{'='*65}")
  print(f"  Workers: {actual_workers}")
  print(f"  Stores:  {actual_stores}")
  print(f"  Total:   {actual_workers + actual_stores}")
  print(f"  Cities:  {len(CITIES)}")
  if places_docs:
    print(f"  From Google Places API: {len(places_docs)}")

  client.close()
  print(f"\n🎉 Done!")

if __name__ == "__main__":
  args = sys.argv[1:]
  for i, arg in enumerate(args):
    if arg == "--places-api":
      USE_PLACES_API = True
    elif arg == "--no-clear":
      CLEAR_EXISTING = False
    elif arg == "--density" and i + 1 < len(args):
      GRID_SPACING_KM = float(args[i + 1])
    elif arg == "--help" or arg == "-h":
      print(__doc__)
      sys.exit(0)
  seed()