g1mp/plugins/url_title_sniffer.py

# -*- coding: utf-8 -*-
"""
IRC3 Bot Plugin: URL Title Fetcher

A plugin for IRC3 bots that monitors chat messages for URLs, fetches their webpage titles, and displays them
with formatted styling in the chat. Provides visual enhancement to URL sharing in IRC channels.

Features:
    - Asynchronous URL processing using aiohttp for efficient network operations
    - Robust HTML parsing with lxml for accurate title extraction
    - Configurable message styling with color and formatting options
    - Built-in exclusion of YouTube URLs to avoid conflicts with dedicated YouTube plugins
    - Error handling for network and parsing operations
    - Proper resource cleanup through session management

Dependencies:
    - aiohttp: For asynchronous HTTP requests
    - irc3: Core IRC bot functionality
    - ircstyle: IRC text formatting utilities
    - lxml: HTML parsing capabilities

Author: Zodiac (simplified by Claude)
Date: 2025-02-14
"""

import re
import aiohttp
import ircstyle
from lxml import html
import irc3
from irc3 import event
from plugins.services.permissions import check_ignore


@irc3.plugin
class URLTitlePlugin:
    """Plugin for fetching and displaying webpage titles from URLs shared in IRC messages.
    
    Monitors IRC messages for URLs, retrieves their webpage titles, and posts formatted responses
    back to the channel. Supports styled text output with configurable formatting options.

    Attributes:
        bot (irc3.IrcBot): Reference to the main IRC bot instance
        session (aiohttp.ClientSession): Persistent HTTP session for making web requests
        url_pattern (re.Pattern): Compiled regex for URL detection in messages
    """

    def __init__(self, bot):
        """Initialize plugin with bot instance and set up HTTP session.
        
        Args:
            bot (irc3.IrcBot): The IRC bot instance this plugin will be attached to
        """
        self.bot = bot
        # Create persistent HTTP session for better performance
        self.session = aiohttp.ClientSession(loop=self.bot.loop)
        # Regex pattern matches both http(s):// URLs and www. domains
        self.url_pattern = re.compile(r"https?://[^\s<>\"']+|www\.[^\s<>\"']+")

    @event(irc3.rfc.PRIVMSG)
    @check_ignore
    async def on_privmsg(self, mask, event, target, data):
        """Handle incoming private messages by processing any URLs they contain.
        
        Args:
            mask (str): IRC user mask of message sender
            event (str): IRC event type
            target (str): Channel or user the message was sent to
            data (str): Content of the message
        """
        # Extract all URLs from the message
        urls = self.url_pattern.findall(data)
        
        for url in urls:
            # Skip YouTube URLs as they're typically handled by dedicated plugins
            if "youtube.com" in url.lower() or "youtu.be" in url.lower():
                continue
                
            try:
                title = await self.fetch_title(url)
                if title:
                    formatted_message = self.format_message(title, url)
                    await self.bot.privmsg(target, formatted_message)
            except Exception as e:
                # Log errors but continue processing other URLs
                self.bot.log.error(f"Error processing URL {url}: {e}")

    def format_message(self, title, url):
        """Create a styled IRC message containing the webpage title and source URL.
        
        Args:
            title (str): The webpage title to display
            url (str): The source URL
            
        Returns:
            str: Formatted IRC message with styling applied
        """
        # Define styled components for the message
        prefix = ircstyle.style("►", fg="cyan", bold=True, reset=True)
        title_label = ircstyle.style("Title", fg="blue", bold=True, reset=True)
        title_text = ircstyle.style(title, fg="green", italics=True, underline=True, reset=True)
        separator = ircstyle.style("❘", fg="grey", bold=True, reset=True)
        url_label = ircstyle.style("Source", fg="blue", bold=True, underline=True, reset=True)
        url_text = ircstyle.style(url, fg="cyan", italics=True, reset=True)
        suffix = ircstyle.style("◄", fg="cyan", bold=True, reset=True)

        return f"{prefix} {title_label}: {title_text} {separator} {url_label}: {url_text} {suffix}"

    async def fetch_title(self, url):
        """Retrieve the title of a webpage using asynchronous HTTP requests.
        
        Args:
            url (str): The URL to fetch the title from
            
        Returns:
            str: The webpage title or "No title found" if title cannot be extracted
            
        Raises:
            Any exceptions from aiohttp or lxml processing
        """
        # Use modern browser User-Agent to avoid being blocked
        headers = {
            "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 "
                         "(KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36"
        }

        async with self.session.get(url, headers=headers, timeout=10) as response:
            response.raise_for_status()
            content = await response.text()
            tree = html.fromstring(content)
            title = tree.findtext(".//title")
            return title.strip() if title else "No title found"

    async def close(self):
        """Clean up resources by closing the HTTP session."""
        await self.session.close()

    def __del__(self):
        """Ensure proper cleanup when the plugin is destroyed."""
        self.bot.create_task(self.close())