"""
Scraper Zillow: usa el search box como humano + extrae JSON de la página.
"""
import re, time, random, json
from playwright.sync_api import sync_playwright

CHROME_PATH  = r"C:\Program Files\Google\Chrome\Application\chrome.exe"
TEMP_PROFILE = r"C:\Temp\chrome_casa_hunter"

def hd(a=1.0, b=2.5):
    time.sleep(random.uniform(a, b))

def scroll(page, steps=4):
    for _ in range(steps):
        page.mouse.wheel(0, random.randint(250, 550))
        time.sleep(random.uniform(0.4, 0.9))

def parse_listings(html, min_p=40000, max_p=230000):
    results = []
    m = re.search(r'<script[^>]*id="__NEXT_DATA__"[^>]*>(.*?)</script>', html, re.DOTALL)
    if m:
        try:
            data = json.loads(m.group(1))
            list_results = (data.get("props",{}).get("pageProps",{})
                           .get("searchPageState",{}).get("cat1",{})
                           .get("searchResults",{}).get("listResults",[]))
            for p in list_results:
                price = p.get("unformattedPrice", 0)
                if min_p <= price <= max_p:
                    city = p.get("addressCity", "")
                    state = p.get("addressState", "")
                    results.append({
                        "source": "zillow",
                        "address": p.get("address","?"),
                        "price": price,
                        "beds": p.get("beds", 0),
                        "baths": p.get("baths", 0),
                        "sqft": p.get("area", 0),
                        "city": city,
                        "state": state,
                        "zip": p.get("addressZipcode",""),
                        "status": p.get("statusType",""),
                        "url": "https://www.zillow.com" + p.get("detailUrl",""),
                        "img": p.get("imgSrc",""),
                        "type": p.get("hdpData",{}).get("homeInfo",{}).get("homeType",""),
                    })
        except Exception as e:
            print(f"  Parse error: {e}")
    return results

def search_city(page, city_query, max_price=230000):
    """Busca una ciudad en Zillow usando el search box."""
    print(f"\n--- Buscando: {city_query} ---")
    try:
        # Ir a zillow.com
        page.goto("https://www.zillow.com", wait_until="load", timeout=30000)
        hd(1.5, 2.5)

        # Encontrar el search box y escribir la ciudad
        search_box = page.query_selector("input[id*='search'], input[placeholder*='address'], input[placeholder*='city']")
        if not search_box:
            # Probar selectores alternativos
            search_box = page.query_selector("#search-box-input, [data-testid='search-input'], input[name='searchQueryState']")

        if search_box:
            search_box.click()
            hd(0.3, 0.6)
            page.keyboard.down("Control")
            page.keyboard.press("a")
            page.keyboard.up("Control")
            hd(0.2, 0.4)
            page.keyboard.press("Delete")
            hd(0.3, 0.5)
            # Escribir como humano, caracter por caracter
            for char in city_query:
                page.keyboard.type(char)
                time.sleep(random.uniform(0.07, 0.18))
            hd(1.0, 1.8)
            page.keyboard.press("Enter")
            page.wait_for_load_state("load", timeout=30000)
            hd(2, 3)
            scroll(page, 4)
            hd(1, 2)
        else:
            # Si no encuentra search box, usar URL directamente
            city_slug = city_query.lower().replace(" ", "-").replace(",", "")
            url = f"https://www.zillow.com/homes/for_sale/{city_slug}/?searchQueryState=%7B%22filterState%22%3A%7B%22price%22%3A%7B%22max%22%3A{max_price}%2C%22min%22%3A40000%7D%7D%7D"
            page.goto(url, wait_until="load", timeout=45000)
            hd(2, 3)
            scroll(page, 4)

        title = page.title()
        html  = page.content()
        print(f"  Título: {title[:60]}")

        listings = parse_listings(html)
        print(f"  Encontrados: {len(listings)} en rango $40K-$230K")
        for l in listings[:3]:
            print(f"    ${l['price']:,} | {l.get('beds','?')}bd | {l['address'][:50]}, {l['city']}")
        return listings

    except Exception as e:
        print(f"  ERROR: {e}")
        return []

# Ciudades objetivo
CITIES = [
    "Vero Beach, FL",
    "Melbourne, FL",
    "Jacksonville, FL",
    "Stuart, FL",
    "Daytona Beach, FL",
    "St. Augustine, FL",
    "Palm Coast, FL",
    "New Smyrna Beach, FL",
]

all_results = []

with sync_playwright() as p:
    ctx = p.chromium.launch_persistent_context(
        user_data_dir=TEMP_PROFILE,
        executable_path=CHROME_PATH,
        headless=False,
        args=[
            "--profile-directory=Default",
            "--disable-blink-features=AutomationControlled",
            "--start-maximized",
            "--no-first-run",
            "--no-default-browser-check",
        ],
        viewport={"width": 1366, "height": 768},
    )
    page = ctx.new_page()

    for city in CITIES:
        listings = search_city(page, city)
        all_results.extend(listings)
        hd(2, 4)  # pausa entre ciudades

    ctx.close()

# Deduplicar por dirección
seen = set()
unique = []
for r in all_results:
    key = r["address"].lower().strip()
    if key not in seen:
        seen.add(key)
        unique.append(r)

print(f"\n=== TOTAL: {len(unique)} listings únicos en $40K-$230K ===")
with open("zillow_final.json", "w", encoding="utf-8") as f:
    json.dump(unique, f, indent=2, ensure_ascii=False)
print("Guardado en zillow_final.json")