import requests
from bs4 import BeautifulSoup

def scrape_zameen(location: str, property_type: str):
    url = f"https://www.zameen.com/search/?city={location}&purpose=sale&type={property_type}"
    response = requests.get(url)
    soup = BeautifulSoup(response.content, 'html.parser')

    properties = []
    for listing in soup.find_all('div', class_='property-card'):
        price = listing.find('span', class_='price').text
        size = listing.find('span', class_='size').text
        location = listing.find('span', class_='location').text
        properties.append({'price': price, 'size': size, 'location': location})

    return properties

import pandas as pd
from sklearn.preprocessing import StandardScaler

# Load data
df = pd.read_csv('properties.csv')

# Features for model
features = {
    'price': df['price'],
    'size_sqft': df['size'],
    'bedrooms': df['bedrooms'],
    'bathrooms': df['bathrooms'],
    'age_years': 2026 - df['construction_year'],
    'distance_to_main_road_km': df['distance_main_road'],
    'proximity_to_dha': df['phase_dha'],  # 0-10 (0=far, 10=DHA Phase 1)
    'amenities_count': df['parking'] + df['garden'] + df['gym'],
}

# Normalize
scaler = StandardScaler()
features_scaled = scaler.fit_transform(features)

from sklearn.ensemble import RandomForestRegressor

X = features_scaled
y = df['price']

model = RandomForestRegressor(n_estimators=100, max_depth=15)
model.fit(X, y)

# Accuracy
from sklearn.metrics import mean_absolute_percentage_error
predictions = model.predict(X)
accuracy = 100 - mean_absolute_percentage_error(y, predictions)
print(f"Model Accuracy: {accuracy:.1f}%")  # Expected: 82-87%

from anthropic import Anthropic

def claude_valuation(property_details: dict):
    comparable_properties = fetch_comparables(
        location=property_details['location'],
        size=property_details['size'],
        bedrooms=property_details['bedrooms']
    )

    prompt = f"""
    I'm valuing a property in {property_details['location']}:
    - Size: {property_details['size']} sq ft
    - Bedrooms: {property_details['bedrooms']}
    - Bathrooms: {property_details['bathrooms']}
    - Age: {property_details['age']} years
    - Amenities: {property_details['amenities']}

    Here are 5 comparable properties recently sold:
    {comparable_properties}

    Based on comparable sales, market trends, and location factors, estimate the fair market value in PKR.
    Explain your valuation reasoning.
    """

    response = client.messages.create(
        model="claude-opus-4-6",
        max_tokens=500,
        messages=[{"role": "user", "content": prompt}]
    )

    return response.content[0].text