# mcp_server.py
import requests
import json
from bs4 import BeautifulSoup
from fastmcp import FastMCP
from concurrent.futures import ThreadPoolExecutor

mcp = FastMCP("BatchWebContentExtraction")

def scrape_single_url(url: str) -> dict:
    """Helper function to scrape and clean a single URL."""
    try:
        headers = {"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64)"}
        response = requests.get(url, headers=headers, timeout=10)
        response.raise_for_status()
        
        soup = BeautifulSoup(response.text, 'html.parser')
        for script in soup(["script", "style", "nav", "footer", "header", "aside"]):
            script.decompose()
            
        raw_text = soup.get_text(separator=' ')
        lines = (line.strip() for line in raw_text.splitlines())
        chunks = (phrase.strip() for line in lines for phrase in line.split("  "))
        clean_text = '\n'.join(chunk for chunk in chunks if chunk)
        
        return {"url": url, "status": "success", "content": clean_text[:5000]}
        
    except Exception as e:
        return {"url": url, "status": "failed", "content": str(e)}

@mcp.tool()
def batch_web_content_extraction(urls: list[str]) -> str:
    """
    Fetches text content from a list of multiple URLs simultaneously in parallel.
    Returns a unified JSON string of results.
    """
    if not urls:
        return json.dumps({"error": "No URLs provided"})
        
    results = []
    with ThreadPoolExecutor(max_workers=min(len(urls), 5)) as executor:
        futures_map = {executor.submit(scrape_single_url, url): url for url in urls}
        for future in futures_map:
            results.append(future.result())
            
    return json.dumps(results)

if __name__ == "__main__":
    # Use standard streamable-http protocol on port 8000
    mcp.run(transport="streamable-http", host="0.0.0.0", port=8000)