# -*- coding: utf-8 -*-
import re
import irc3
import requests
import io
from lxml import etree

from . import Plugin

class URLInfo(Plugin):
    BLACKLIST = [
        "^https?:\/\/www\.youtube\.com",
        "^https?:\/\/youtu\.be",
        "^https?:\/\/w0bm\.com",
        "^https?:\/\/f0ck\.me"
    ]

    SIZE_LIMIT = 2 * 1024 ** 2

    @irc3.event(r'(?i)^:\S+ PRIVMSG (?P<target>\S+) :.*(?P<url>https?:\/\/\S+\.\S+).*')
    def url_parser(self, target: str, url: str):
        for regex in self.BLACKLIST:
            if re.match(regex, url):
                return
        bytes_io = io.BytesIO()
        try:
            with requests.get(url, timeout=10, stream=True) as r:
                r.raise_for_status()
                size = 0
                for chunk in r.iter_content(chunk_size=1024 ** 2):
                    size += len(chunk)
                    if size >= self.SIZE_LIMIT:
                        return
                    bytes_io.write(chunk)
        except (requests.exceptions.ConnectionError, requests.exceptions.HTTPError) as e:
            return

        mime_type = r.headers.get("content-type")
        if mime_type is not None:
            if mime_type.split(";")[0] != "text/html":
                return
        bytes_io.seek(0)
        tree = etree.parse(bytes_io, etree.HTMLParser()).getroot()
        title = tree.xpath("/html/head/title")
        if len(title) > 0:
            self.bot.privmsg(target, '\x02[URLInfo]\x02 ' + title[0].text.strip())