youtube_dl/extractor/beeg.py

   1 from __future__ import unicode_literals
   2
   3 from .common import InfoExtractor
   4 from ..compat import (
   5     compat_chr,
   6     compat_ord,
   7     compat_urllib_parse_unquote,
   8 )
   9 from ..utils import (
  10     int_or_none,
  11     parse_iso8601,
  12 )
  13
  14
  15 class BeegIE(InfoExtractor):
  16     _VALID_URL = r'https?://(?:www\.)?beeg\.com/(?P<id>\d+)'
  17     _TEST = {
  18         'url': 'http://beeg.com/5416503',
  19         'md5': '46c384def73b33dbc581262e5ee67cef',
  20         'info_dict': {
  21             'id': '5416503',
  22             'ext': 'mp4',
  23             'title': 'Sultry Striptease',
  24             'description': 'md5:d22219c09da287c14bed3d6c37ce4bc2',
  25             'timestamp': 1391813355,
  26             'upload_date': '20140207',
  27             'duration': 383,
  28             'tags': list,
  29             'age_limit': 18,
  30         }
  31     }
  32
  33     def _real_extract(self, url):
  34         video_id = self._match_id(url)
  35
  36         webpage = self._download_webpage(url, video_id)
  37
  38         cpl_url = self._search_regex(
  39             r'<script[^>]+src=(["\'])(?P<url>(?:https?:)?//static\.beeg\.com/cpl/\d+\.js.*?)\1',
  40             webpage, 'cpl', default=None, group='url')
  41
  42         beeg_version, beeg_salt = [None] * 2
  43
  44         if cpl_url:
  45             cpl = self._download_webpage(
  46                 self._proto_relative_url(cpl_url), video_id,
  47                 'Downloading cpl JS', fatal=False)
  48             if cpl:
  49                 beeg_version = self._search_regex(
  50                     r'beeg_version\s*=\s*(\d+)', cpl,
  51                     'beeg version', default=None) or self._search_regex(
  52                     r'/(\d+)\.js', cpl_url, 'beeg version', default=None)
  53                 beeg_salt = self._search_regex(
  54                     r'beeg_salt\s*=\s*(["\'])(?P<beeg_salt>.+?)\1', cpl, 'beeg beeg_salt',
  55                     default=None, group='beeg_salt')
  56
  57         beeg_version = beeg_version or '1750'
  58         beeg_salt = beeg_salt or 'MIDtGaw96f0N1kMMAM1DE46EC9pmFr'
  59
  60         video = self._download_json(
  61             'http://api.beeg.com/api/v6/%s/video/%s' % (beeg_version, video_id),
  62             video_id)
  63
  64         def split(o, e):
  65             def cut(s, x):
  66                 n.append(s[:x])
  67                 return s[x:]
  68             n = []
  69             r = len(o) % e
  70             if r > 0:
  71                 o = cut(o, r)
  72             while len(o) > e:
  73                 o = cut(o, e)
  74             n.append(o)
  75             return n
  76
  77         def decrypt_key(key):
  78             # Reverse engineered from http://static.beeg.com/cpl/1738.js
  79             a = beeg_salt
  80             e = compat_urllib_parse_unquote(key)
  81             o = ''.join([
  82                 compat_chr(compat_ord(e[n]) - compat_ord(a[n % len(a)]) % 21)
  83                 for n in range(len(e))])
  84             return ''.join(split(o, 3)[::-1])
  85
  86         def decrypt_url(encrypted_url):
  87             encrypted_url = self._proto_relative_url(
  88                 encrypted_url.replace('{DATA_MARKERS}', ''), 'https:')
  89             key = self._search_regex(
  90                 r'/key=(.*?)%2Cend=', encrypted_url, 'key', default=None)
  91             if not key:
  92                 return encrypted_url
  93             return encrypted_url.replace(key, decrypt_key(key))
  94
  95         formats = []
  96         for format_id, video_url in video.items():
  97             if not video_url:
  98                 continue
  99             height = self._search_regex(
 100                 r'^(\d+)[pP]$', format_id, 'height', default=None)
 101             if not height:
 102                 continue
 103             formats.append({
 104                 'url': decrypt_url(video_url),
 105                 'format_id': format_id,
 106                 'height': int(height),
 107             })
 108         self._sort_formats(formats)
 109
 110         title = video['title']
 111         video_id = video.get('id') or video_id
 112         display_id = video.get('code')
 113         description = video.get('desc')
 114
 115         timestamp = parse_iso8601(video.get('date'), ' ')
 116         duration = int_or_none(video.get('duration'))
 117
 118         tags = [tag.strip() for tag in video['tags'].split(',')] if video.get('tags') else None
 119
 120         return {
 121             'id': video_id,
 122             'display_id': display_id,
 123             'title': title,
 124             'description': description,
 125             'timestamp': timestamp,
 126             'duration': duration,
 127             'tags': tags,
 128             'formats': formats,
 129             'age_limit': self._rta_search(webpage),
 130         }