# this is  asmolagent too to fetch html content from a url
from smolagents import tool
import requests
from markdownify import markdownify as md
import time
from bs4 import BeautifulSoup

@tool
def fetch_webpage(url: str, convert_to_markdown: bool = True) -> str:
    """
    Fetches the HTML content of a given URL. 
    if markdown conversion is enabled, it will remove script and style and return the text content as markdown else return raw unfiltered HTML
    Args:
        url (str): The URL to fetch.
        convert_to_markdown (bool): If True, convert the HTML content to Markdown format. else return the raw HTML.
    Returns:
        str: The HTML content of the URL.
    """    
    response = requests.get(url)
    if (convert_to_markdown):
        soup = BeautifulSoup(response.text, "html.parser")
        # remove script and style tags
        for script in soup(["script", "style"]):
            script.extract()

        # for wikipedia only keep the main content
        if "wikipedia.org" in url:
            main_content = soup.find("main",{"id":"content"})
            if main_content:
                content = md(str(main_content),strip=['script', 'style']).strip()
            else:
                content = md(response.text,strip=['script', 'style']).strip()
    else:
        content = response.text
    
    try:
        # save content to a file in test folder before returning
        # compute filepath with correct extension based on convert_to_markdown and add a timestamp for unicity
        file_extension = ".md" if convert_to_markdown else ".html"
        unicity_suffix = str(int(time.time()))
        file_name = f"test/fetched_content_{unicity_suffix}{file_extension}"
        with open(file_name, "w", encoding="utf-8") as f:
            f.write(content)
    except Exception as e:
        print(f"Error saving content to file: {e}")
        
    return content

@tool
# this tool allow web search on a local SearXNG instance
def search_web(query: str, num_results: int = 5) -> list:
    """
    Perform a web search using local SearXNG instance.
    Args:
        query (str): The search query.
        num_results (int): The number of results to return.
    Returns:
        list: A list of search results sorted by score with {url, title, content, score} for each result.
    """
    # local metaserach engine searxng, run on localhost:8888
    searxng_url = "http://localhost:8888/search"
    params = {"q": query, "format": 'json'}
    response = requests.get(searxng_url, params=params)
    if response.status_code == 200:
        ret = response.json()
        # keep only the response'results' array
        results = ret.get("results", [])
        # keep only the first num_results   
        results = results[:num_results]
        # for each result keep only the url, title and content ans score
        results = [
            {
                "url": result.get("url"),
                "title": result.get("title"),
                "content": result.get("content"),
                "score": result.get("score"),
            }
            for result in results
        ]
        
        return results
        
    else:
        print(f"Error: {response.status_code}")
        return []

if __name__ == "__main__":
   
    try:
        # Test the function
        query = "What is the capital of France?"
        results = search_web(query,3)
        print(results)
    except Exception as e:
        print(f"An error occurred: {e}")

    try:
        # Test the function
        video_id = "L1vXCYZAYYM"  # Replace with your YouTube video ID
        video_url = "https://www.youtube.com/watch?v=" + video_id
        url = "https://en.wikipedia.org/wiki/Malko_Competition"
        # page_content = fetch_webpage(video_url)
        page_content = fetch_webpage(url, convert_to_markdown=True)
        print(page_content.encode("utf-8"))
    except Exception as e:
        print(f"An error occurred: {e}")