88 lines
3.3 KiB
Python
88 lines
3.3 KiB
Python
import requests, json, re
|
|
from bs4 import BeautifulSoup
|
|
|
|
headers = {
|
|
"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 Chrome/124 Safari/537.36",
|
|
"Accept": "text/html,application/xhtml+xml,*/*;q=0.8",
|
|
"Accept-Language": "en-US,en;q=0.9",
|
|
"Accept-Encoding": "gzip, deflate, br",
|
|
"Connection": "keep-alive",
|
|
}
|
|
|
|
# 1. Test Brevard County Property Appraiser (government API - no anti-bot)
|
|
print("=== Brevard County Property API ===")
|
|
try:
|
|
r = requests.get(
|
|
"https://www.bcpao.us/api/search?activeOnly=true&saleAmountMax=230000"
|
|
"&saleAmountMin=50000&town=MELBOURNE&count=10&page=1",
|
|
headers=headers, timeout=15
|
|
)
|
|
print(f"Status: {r.status_code}, size: {len(r.text)}")
|
|
print(r.text[:500])
|
|
except Exception as e:
|
|
print(f"ERROR: {e}")
|
|
|
|
# 2. Test Indian River County (Vero Beach) Property Appraiser
|
|
print("\n=== Indian River County Property API ===")
|
|
try:
|
|
r = requests.get(
|
|
"https://www.ircpa.net/propertysearch.aspx",
|
|
headers=headers, timeout=15
|
|
)
|
|
print(f"Status: {r.status_code}, size: {len(r.text)}")
|
|
except Exception as e:
|
|
print(f"ERROR: {e}")
|
|
|
|
# 3. Test Point2Homes (no JS required usually)
|
|
print("\n=== Point2Homes ===")
|
|
try:
|
|
r = requests.get(
|
|
"https://www.point2homes.com/US/Real-Estate-Listings/FL/Indian-River-County.html"
|
|
"?PriceMin=50000&PriceMax=230000",
|
|
headers=headers, timeout=15
|
|
)
|
|
print(f"Status: {r.status_code}, size: {len(r.text)}")
|
|
soup = BeautifulSoup(r.text, "html.parser")
|
|
listings = soup.select(".item-listing, .listing, [class*='listing-card']")
|
|
print(f"Listings found: {len(listings)}")
|
|
for l in listings[:2]:
|
|
print(f" {l.get_text()[:100].strip()}")
|
|
except Exception as e:
|
|
print(f"ERROR: {e}")
|
|
|
|
# 4. Test Homes.com
|
|
print("\n=== Homes.com ===")
|
|
try:
|
|
r = requests.get(
|
|
"https://www.homes.com/homes-for-sale/fl/vero-beach/?min_price=50000&max_price=230000",
|
|
headers=headers, timeout=15
|
|
)
|
|
print(f"Status: {r.status_code}, size: {len(r.text)}")
|
|
soup = BeautifulSoup(r.text, "html.parser")
|
|
# Try to find embedded JSON data
|
|
scripts = soup.find_all("script", type="application/json")
|
|
print(f" JSON scripts found: {len(scripts)}")
|
|
scripts2 = [s for s in soup.find_all("script") if s.string and "listPrice" in (s.string or "")]
|
|
print(f" Scripts with listPrice: {len(scripts2)}")
|
|
except Exception as e:
|
|
print(f"ERROR: {e}")
|
|
|
|
# 5. Test HUD proper endpoint
|
|
print("\n=== HUD Homes (correct endpoint) ===")
|
|
try:
|
|
# Try different HUD endpoints
|
|
urls = [
|
|
"https://www.hudhomestore.gov/HudHomes/SearchProperties.aspx?state=FL&county=Indian+River&maxprice=230000",
|
|
"https://www.hudhomestore.gov/HudHomes/SearchProperties.aspx?state=FL&county=Brevard&maxprice=230000",
|
|
]
|
|
for url in urls:
|
|
r = requests.get(url, headers=headers, timeout=15)
|
|
soup = BeautifulSoup(r.text, "html.parser")
|
|
listings = soup.select("tr.datarow, .propRow, [class*='prop-row']")
|
|
prices = re.findall(r'\$[\d,]+', r.text)
|
|
print(f" {url.split('county=')[1].split('&')[0]}: status={r.status_code}, listings={len(listings)}, prices found={len(prices)}")
|
|
if prices:
|
|
print(f" First prices: {prices[:5]}")
|
|
except Exception as e:
|
|
print(f"ERROR: {e}")
|