info = {
	'author': 'Jon Bergli Heier',
	'title': 'URL Titles',
	'description': 'Fetches the title tags off of URLs.',
}

cfg_section = 'module/url_titles'

import re, urllib2, htmlentitydefs, gzip, cStringIO, time
from PIL import ImageFile

class Module:
	re_http = re.compile(r'(https?://[^\ #]+)')
	re_title = re.compile(r'<title[^>]*?>(.*?)</title>', re.S | re.I)
	re_meta = re.compile(r'<meta\s+http-equiv="content-type"\s+content="[^"]+charset=([^"]+)"', re.I)

	def __init__(self, bot):
		self.irc = bot
		if bot:
			self.irc.register_callback(self)

	def get_titles(self, s):
		def parse_url(url):
			s = url[7:].split('/', 1)[0:]
			host = s[0]
			path = '/' if len(s) == 1 else '/' + s[1]
			return host, path

		def unescape(text):
			def fixup(m):
				text = m.group(0)
				if text[:2] == "&#":
					# character reference
					try:
						if text[:3] == "&#x":
							return unichr(int(text[3:-1], 16)).encode('utf-8')
						else:
							return unichr(int(text[2:-1])).encode('utf-8')
					except ValueError:
						pass
				else:
					# named entity
					try:
						text = unichr(htmlentitydefs.name2codepoint[text[1:-1]]).encode('utf-8')
					except KeyError:
						pass
				return text # leave as is
			return re.sub("&#?\w+;", fixup, text)

		def format_text(s):
			s = s.replace('\n', ' ').replace('\r', ' ').replace('\t', ' ')
			while '  ' in s:
				s = s.replace('  ', ' ')
			s = unescape(s)
			return s

		m = self.re_http.findall(s)
		titles = []
		for url in m:
			# ignore spotify URLs
			if 'open.spotify.com' in url:
				continue
			
			# Run AniDB URLs through AniDB module.
			m = re.match('http://anidb.net/(perl-bin/animedb.pl\?show=anime&aid=|a)(\d+)', url)
			if m and 'anidb' in self.irc.modules:
				aid = int(m.groups()[1])
				titles.append(self.irc.modules['anidb'].get_info(aid))
				continue
			
			t = time.time()
			try:
				u = urllib2.urlopen(url, timeout = config.getfloat(cfg_section, 'timeout'))
			except:
				return
			if not 'content-type' in u.headers:
				u.close()
				continue
			if u.headers['content-type'].startswith('text/html'):
				#s = u.read()
				if 'content-encoding' in u.headers and u.headers['content-encoding'] == 'gzip':
					s = cStringIO.StringIO(u.read())
					s.seek(0)
					s = gzip.GzipFile(fileobj = s).read()
					m = self.re_title.search(s)
					meta_enc = self.re_meta.search(s)
				else:
					s = ''
					m = None
					meta_enc = None
					buf = u.read(1024)
					while buf and time.time() - t < 5.0:
						s += buf
						m = self.re_title.search(s)
						meta_enc = self.re_meta.search(s)
						if m and (meta_enc or '</head>' in s):
							break
						buf = u.read(1024)
				ct = u.headers['content-type']
				enc = ct.lower().split('charset=')
				if len(enc) == 2:
					enc = enc[1]
				elif meta_enc:
					enc = meta_enc.groups()[0]
				else:
					enc = None

				if m:
					s = m.groups()[0]
					if enc:
						s = s.decode(enc, 'replace').encode('utf8', 'replace')
					titles.append(s)
			elif u.headers['content-type'] in ('image/gif', 'image/png', 'image/jpeg'):
				def pretty_size(size):
					suffixes = (('B', 2**10), ('KiB', 2**20), ('MiB', 2**30), ('GiB', 2**40), ('TiB', 2**50))
					for suf, lim in suffixes:
						if size > lim:
							continue
						else:
							return '%s %s' % (str(round(size/float(lim/2**10), 2)), suf)
				p = ImageFile.Parser()
				size = 0
				while time.time() - t < 5.0:
					s = u.read(1024)
					size += len(s)
					if not s:
						break
					p.feed(s)
				try:
					im = None
					im = p.close()
					titles.append('%s image: %dx%d (%s)' % ((im.format,) + tuple(im.size) + (pretty_size(size),)))
				except:
					pass
				finally:
					del im
			u.close()
		if len(titles) == 0:
			return
		elif len(titles) == 1:
			s = format_text(titles[0])
		else:
			s = ''
			for i in range(len(titles)):
				s += '\002[%d]\002 %s ' % (i+1, format_text(titles[i]))
		return s.strip()

	def privmsg(self, nick, channel, msg):
		titles = self.get_titles(msg)
		if titles:
			self.irc.msg(channel if not channel == self.irc.nickname else nick.split('!')[0], titles)

if __name__ == '__main__':
	import sys, ConfigParser, os
	config = ConfigParser.ConfigParser()
	config.read([os.path.expanduser('~/.fot')])
	m = Module(None)
	print m.get_titles(' '.join(sys.argv[1:]))