modules/urls.py

   1 # Erebus IRC bot - Author: Erebus Team
   2 # URL Checker
   3 # This file is released into the public domain; see http://unlicense.org/
   4
   5 # module info
   6 modinfo = {
   7         'author': 'Erebus Team',
   8         'license': 'public domain',
   9         'compatible': [1,2],
  10         'depends': [],
  11         'softdeps': [],
  12 }
  13
  14 # http://embed.ly/tools/generator
  15
  16 # preamble
  17 import modlib
  18 lib = modlib.modlib(__name__)
  19 modstart = lib.modstart
  20 modstop = lib.modstop
  21
  22 # module code
  23 import re, urllib2, urlparse, json, HTMLParser
  24 from BeautifulSoup import BeautifulSoup
  25
  26 html_parser = HTMLParser.HTMLParser()
  27
  28 hostmask_regex = re.compile(r'^(.*)!(.*)@(.*)$')
  29 url_regex = re.compile(r'((?:https?://|spotify:)[^\s]+)')
  30 spotify_regex = (
  31         re.compile(r'spotify:(?P<type>\w+):(?P<track_id>\w{22})'),
  32         re.compile(r'https?://open.spotify.com/(?P<type>\w+)/(?P<track_id>\w{22})')
  33 )
  34 youtube_regex = (
  35         re.compile(r'https?://(?:www\.)?youtube\.com/watch\?[a-zA-Z0-9=&_\-]+'),
  36 )
  37 twitch_regex = (
  38         re.compile(r'https?:\/\/(?:www\.)?twitch.tv\/([A-Za-z0-9]*)'),
  39 )
  40
  41 def parser_hostmask(hostmask):
  42         if isinstance(hostmask, dict):
  43                 return hostmask
  44
  45         nick = None
  46         user = None
  47         host = None
  48
  49         if hostmask is not None:
  50                 match = hostmask_regex.match(hostmask)
  51
  52                 if not match:
  53                         nick = hostmask
  54                 else:
  55                         nick = match.group(1)
  56                         user = match.group(2)
  57                         host = match.group(3)
  58
  59         return {
  60                 'nick': nick,
  61                 'user': user,
  62                 'host': host
  63         }
  64
  65 class SmartRedirectHandler(urllib2.HTTPRedirectHandler):
  66         def http_error_301(self, req, fp, code, msg, headers):
  67                 result = urllib2.HTTPRedirectHandler.http_error_301(
  68                                 self, req, fp, code, msg, headers)
  69                 result.status = code
  70                 return result
  71
  72         def http_error_302(self, req, fp, code, msg, headers):
  73                 result = urllib2.HTTPRedirectHandler.http_error_302(
  74                                 self, req, fp, code, msg, headers)
  75                 result.status = code
  76                 return result
  77
  78 @lib.hooknum("PRIVMSG")
  79 def privmsg_hook(bot, textline):
  80         user = parser_hostmask(textline[1:textline.find(' ')])
  81         chan = textline.split()[2]
  82
  83         try:
  84                 line = textline.split(None, 3)[3][1:]
  85         except IndexError:
  86                 line = ''
  87
  88         for match in url_regex.findall(line):
  89                 if match:
  90                         if 'open.spotify.com' in match or 'spotify:' in match:
  91                                 for r in spotify_regex:
  92                                         for sptype, track in r.findall(match):
  93                                                 bot.msg(chan, gotspotify(sptype, track))
  94
  95                         elif 'youtube.com' in match or 'youtu.be' in match:
  96                                 for r in youtube_regex:
  97                                         for url in r.findall(match):
  98                                                 bot.msg(chan, gotyoutube(url))
  99
 100                         elif 'twitch.tv' in match:
 101                                 for r in twitch_regex:
 102                                         for uri in r.findall(match):
 103                                                 bot.msg(chan, gottwitch(uri))
 104
 105                         else:
 106                                 bot.msg(chan, goturl(match))
 107
 108 def unescape(line):
 109         return html_parser.unescape(line)
 110
 111 def gotspotify(type, track):
 112         url = 'http://ws.spotify.com/lookup/1/?uri=spotify:%s:%s' % (type, track)
 113         xml = urllib2.urlopen(url).read()
 114         soup = BeautifulSoup(xml, convertEntities=BeautifulSoup.HTML_ENTITIES)
 115         lookup_type = soup.contents[2].name
 116
 117         if lookup_type == 'track':
 118                 name = soup.find('name').string
 119                 album_name = soup.find('album').find('name').string
 120                 artist_name = soup.find('artist').find('name').string
 121                 popularity = soup.find('popularity')
 122                 if popularity:
 123                         popularity = float(popularity.string)*100
 124                 length = float(soup.find('length').string)
 125                 minutes = int(length)/60
 126                 seconds =  int(length)%60
 127
 128                 return unescape('Track: %s - %s / %s %s:%.2d %2d%%' % (artist_name, name, album_name, minutes, seconds, popularity))
 129
 130         elif lookup_type == 'album':
 131                 album_name = soup.find('album').find('name').string
 132                 artist_name = soup.find('artist').find('name').string
 133                 released = soup.find('released').string
 134                 return unescape('Album: %s - %s - %s' % (artist_name, album_name, released))
 135
 136         else:
 137                 return 'Unsupported type.'
 138
 139 def gotyoutube(url):
 140         url_data = urlparse.urlparse(url)
 141         query = urlparse.parse_qs(url_data.query)
 142         video = query["v"][0]
 143         api_url = 'http://gdata.youtube.com/feeds/api/videos/%s?alt=json&v=2' % video
 144         try:
 145                 respdata = urllib2.urlopen(api_url).read()
 146                 video_info = json.loads(respdata)
 147
 148                 title = video_info['entry']['title']["$t"]
 149                 author = video_info['entry']['author'][0]['name']['$t']
 150
 151                 return unescape("Youtube: %s (%s)" % (title, author))
 152         except:
 153                 pass
 154
 155 def gottwitch(uri):
 156                 url = 'http://api.justin.tv/api/stream/list.json?channel=%s' % uri.split('/')[0]
 157                 respdata = urllib2.urlopen(url).read()
 158                 twitch = json.loads(respdata)
 159                 try:
 160                         return unescape('Twitch: %s (%s playing %s)' % (twitch[0]['channel']['status'], twitch[0]['channel']['login'], twitch[0]['channel']['meta_game']))
 161                 except:
 162                         return 'Twitch: Channel offline.'
 163
 164 def goturl(url):
 165         request = urllib2.Request(url)
 166         opener = urllib2.build_opener(SmartRedirectHandler())
 167         try:
 168                 soup = BeautifulSoup(opener.open(request, timeout=2))
 169                 return unescape('Title: %s' % (soup.title.string))
 170         except:
 171                 return 'Invalid URL/Timeout'