pythonScrapper

from playwright.sync_api import sync_playwright
import openai
import json

# Extract elements using Playwright
def extract_elements(url):
    with sync_playwright() as p:
        browser = p.chromium.launch(headless=True)
        page = browser.new_page()
        page.goto(url)
        
        elements = page.locator('*').all()  # Extract all elements
        element_data = []
        for element in elements:
            tag_name = element.evaluate('el => el.tagName')
            attributes = element.evaluate('el => Object.fromEntries([...el.attributes].map(attr => [attr.name, attr.value]))')
            element_data.append({'tag_name': tag_name, 'attributes': attributes})
        browser.close()
        return element_data

# Use GPT to classify elements
def classify_elements(elements):
    openai.api_key = "your_openai_api_key"
    for element in elements:
        prompt = f"Given the following HTML attributes: {element['attributes']}, classify the type of element (e.g., button, input, link) and suggest an XPath."
        response = openai.Completion.create(engine="text-davinci-003", prompt=prompt, max_tokens=100)
        element['classification'] = response.choices[0].text.strip()
    return elements

# Generate POM
def generate_pom(elements, class_name="GeneratedPage"):
    pom_template = f"class {class_name}:\n    def __init__(self, driver):\n        self.driver = driver\n\n"
    for element in elements:
        if 'classification' in element:
            pom_template += f"    def {element['classification'].replace(' ', '_')}:\n"
            pom_template += f"        return self.driver.find_element(By.XPATH, \"{element['xpath']}\")\n\n"
    return pom_template

# Run
url = "https://example.com"
elements = extract_elements(url)
classified_elements = classify_elements(elements)
pom_code = generate_pom(classified_elements)
print(pom_code)