modules/urls.py

   1 # Erebus IRC bot - Author: Conny Sjoblom
   2 # vim: fileencoding=utf-8
   3 # URL Checker
   4 # This file is released into the public domain; see http://unlicense.org/
   5
   6 # module info
   7 modinfo = {
   8         'author': 'Erebus Team',
   9         'license': 'public domain',
  10         'compatible': [0],
  11         'depends': [],
  12         'softdeps': [],
  13 }
  14
  15 # http://embed.ly/tools/generator
  16
  17 # preamble
  18 import modlib
  19 lib = modlib.modlib(__name__)
  20 modstart = lib.modstart
  21 modstop = lib.modstop
  22
  23 # module code
  24 import sys
  25 if sys.version_info.major < 3:
  26         stringbase = basestring
  27         import urllib2
  28         import urlparse
  29         import HTMLParser
  30         html = HTMLParser.HTMLParser()
  31         from BeautifulSoup import BeautifulSoup
  32 else:
  33         stringbase = str
  34         import urllib.request as urllib2
  35         import urllib.parse as urlparse
  36         import html
  37         from bs4 import BeautifulSoup
  38
  39 import re, json, datetime
  40
  41 hostmask_regex = re.compile(r'^(.*)!(.*)@(.*)$')
  42
  43 def parser_hostmask(hostmask):
  44         if isinstance(hostmask, dict):
  45                 return hostmask
  46
  47         nick = None
  48         user = None
  49         host = None
  50
  51         if hostmask is not None:
  52                 match = hostmask_regex.match(hostmask)
  53
  54                 if not match:
  55                         nick = hostmask
  56                 else:
  57                         nick = match.group(1)
  58                         user = match.group(2)
  59                         host = match.group(3)
  60
  61         return {
  62                 'nick': nick,
  63                 'user': user,
  64                 'host': host
  65         }
  66
  67 class SmartRedirectHandler(urllib2.HTTPRedirectHandler):
  68         def http_error_301(self, req, fp, code, msg, headers):
  69                 result = urllib2.HTTPRedirectHandler.http_error_301(
  70                                 self, req, fp, code, msg, headers)
  71                 result.status = code
  72                 return result
  73
  74         def http_error_302(self, req, fp, code, msg, headers):
  75                 result = urllib2.HTTPRedirectHandler.http_error_302(
  76                                 self, req, fp, code, msg, headers)
  77                 result.status = code
  78                 return result
  79
  80 def process_line(line):
  81         responses = []
  82         num_found = 0
  83         limit = lib.parent.cfg.getint('urls', 'limit', 2)
  84         for action, group, prefix in regexes:
  85                 for regex in group:
  86                         for match in regex.findall(line):
  87                                 if match:
  88                                         num_found += 1
  89                                         if num_found > limit:
  90                                                 return responses
  91                                         if isinstance(match, stringbase):
  92                                                 resp = action(match)
  93                                         else:
  94                                                 resp = action(*match)
  95                                         if resp is not None:
  96                                                 responses.append("%s: %s" % (prefix, resp))
  97         return responses
  98
  99 @lib.hooknum("PRIVMSG")
 100 def privmsg_hook(bot, textline):
 101         user = parser_hostmask(textline[1:textline.find(' ')])
 102         chan = textline.split()[2]
 103
 104         try:
 105                 line = textline.split(None, 3)[3][1:]
 106         except IndexError:
 107                 line = ''
 108
 109         responses = process_line(line)
 110         if len(responses) > 0:
 111                 if lib.parent.cfg.getboolean('urls', 'multiline'):
 112                         for r in responses:
 113                                 bot.msg(chan, r, True)
 114                 else:
 115                         bot.msg(chan, ' | '.join(responses), True)
 116
 117 def unescape(line):
 118         return re.sub('\s+', ' ', html.unescape(line))
 119
 120 def gotspotify(type, track):
 121         url = 'http://ws.spotify.com/lookup/1/?uri=spotify:%s:%s' % (type, track)
 122         xml = urllib2.urlopen(url).read()
 123         soup = BeautifulSoup(xml, convertEntities=BeautifulSoup.HTML_ENTITIES)
 124         lookup_type = soup.contents[2].name
 125
 126         if lookup_type == 'track':
 127                 name = soup.find('name').string
 128                 album_name = soup.find('album').find('name').string
 129                 artist_name = soup.find('artist').find('name').string
 130                 popularity = soup.find('popularity')
 131                 if popularity:
 132                         popularity = float(popularity.string)*100
 133                 length = float(soup.find('length').string)
 134                 minutes = int(length)/60
 135                 seconds = int(length)%60
 136
 137                 return unescape('Track: %s - %s / %s %s:%.2d %2d%%' % (artist_name, name, album_name, minutes, seconds, popularity))
 138
 139         elif lookup_type == 'album':
 140                 album_name = soup.find('album').find('name').string
 141                 artist_name = soup.find('artist').find('name').string
 142                 released = soup.find('released').string
 143                 return unescape('Album: %s - %s - %s' % (artist_name, album_name, released))
 144
 145         else:
 146                 return 'Unsupported type.'
 147
 148 def _yt_duration(s):
 149         mo = re.match(r'P(\d+D)?T(\d+H)?(\d+M)?(\d+S)?', s)
 150         pcs = [x for x in mo.groups() if x]
 151         return ''.join(pcs).lower()
 152 def _yt_date(s, f):
 153         mo = re.match(r'(\d{4})-(\d{2})-(\d{2})T(\d{2}):(\d{2}):(\d{2})\.(\d+)Z', s)
 154         return datetime.datetime(*(int(x) for x in mo.groups())).strftime(f)
 155 def _yt_round(n):
 156         n = float(n)
 157         if n >= 10**12:
 158                 return '%.1ft' % (n/10**12)
 159         elif n >= 10**9:
 160                 return '%.1fb' % (n/10**9)
 161         elif n >= 10**6:
 162                 return '%.1fm' % (n/10**6)
 163         elif n >= 10**3:
 164                 return '%.1fk' % (n/10**3)
 165         else:
 166                 return int(n)
 167
 168 def gotyoutube(url):
 169         url_data = urlparse.urlparse(url)
 170         query = urlparse.parse_qs(url_data.query)
 171         video = query["v"][0]
 172         api_url = 'https://www.googleapis.com/youtube/v3/videos?part=snippet,contentDetails,statistics&id=%s&key=%s' % (video, lib.parent.cfg.get('urls', 'api_key'))
 173         try:
 174                 respdata = urllib2.urlopen(api_url).read()
 175                 v = json.loads(respdata)
 176                 v = v['items'][0]
 177
 178                 return unescape(lib.parent.cfg.get('urls', 'yt_format', "\002%(author)s\002: \037%(title)s\037 [%(duration)s, uploaded %(uploaded)s, %(views)s v/%(likes)s l/%(dislikes)s d]") % {
 179                         'title': v['snippet']['title'],
 180                         'author': v['snippet']['channelTitle'],
 181                         'duration': _yt_duration(v['contentDetails']['duration']),
 182                         'uploaded': _yt_date(v['snippet']['publishedAt'], lib.parent.cfg.get('urls', 'yt_date_format', '%b %d %Y')),
 183                         'views': _yt_round(v['statistics']['viewCount']),
 184                         'likes': _yt_round(v['statistics']['likeCount']),
 185                         'dislikes': _yt_round(v['statistics']['dislikeCount']),
 186                 })
 187         except urllib2.HTTPError as e:
 188                 if e.getcode() == 403:
 189                         return 'API limit exceeded'
 190                 else:
 191                         return str(e)
 192         except IndexError:
 193                 return 'no results'
 194         except Exception as e:
 195                 return str(e)
 196
 197 def gottwitch(uri):
 198         url = 'https://api.twitch.tv/helix/streams?user_login=%s' % uri.split('/')[0]
 199         opener = urllib2.build_opener()
 200         opener.addheaders = [('Client-ID', lib.parent.cfg.get('urls', 'twitch_api_key'))]
 201         respdata = opener.open(url).read()
 202         twitch = json.loads(respdata)['data']
 203         try:
 204                 # TODO: add current game.
 205                 return unescape('\037%s\037 is %s (%s)' % (twitch[0]['user_name'], twitch[0]['type'], twitch[0]['title']))
 206         except:
 207                 return 'Channel offline.'
 208
 209 def goturl(url):
 210         for _, group, _ in other_regexes:
 211                 for regex in group:
 212                         if regex.match(url):
 213                                 return None
 214         request = urllib2.Request(url)
 215         opener = urllib2.build_opener(SmartRedirectHandler())
 216         try:
 217                 soup = BeautifulSoup(opener.open(request, timeout=0.5))
 218                 if soup.title:
 219                         return unescape('%s' % (soup.title.string))
 220                 else:
 221                         return None
 222         except urllib2.HTTPError as e:
 223                 return 'Error: %s %s' % (e.code, e.reason)
 224         except urllib2.URLError as e:
 225                 return 'Error: %s' % (e.reason)
 226         except Exception as e:
 227                 return 'Error: %r' % (e.args)
 228
 229 url_regex = (
 230         re.compile(r'https?://[^/\s]+\.[^/\s]+(?:/\S+)?'),
 231 )
 232 spotify_regex = (
 233         re.compile(r'spotify:(?P<type>\w+):(?P<track_id>\w{22})'),
 234         re.compile(r'https?://open\.spotify\.com/(?P<type>\w+)/(?P<track_id>\w+)')
 235 )
 236 youtube_regex = (
 237         re.compile(r'https?://(?:www\.)?youtube\.com/watch\?[a-zA-Z0-9=&_\-]+'),
 238 )
 239 twitch_regex = (
 240         re.compile(r'https?:\/\/(?:www\.)?twitch.tv\/([A-Za-z0-9]*)'),
 241 )
 242 other_regexes = (
 243         (gotspotify, spotify_regex, 'Spotify'),
 244         (gotyoutube, youtube_regex, 'YouTube'),
 245         (gottwitch, twitch_regex, 'Twitch'),
 246 )
 247 regexes = other_regexes + (
 248         (goturl, url_regex, 'Title'),
 249 )