[beeg] Fix extraction (closes #14403)
[ytdl] / 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     urljoin,
13 )
14
15
16 class BeegIE(InfoExtractor):
17     _VALID_URL = r'https?://(?:www\.)?beeg\.com/(?P<id>\d+)'
18     _TEST = {
19         'url': 'http://beeg.com/5416503',
20         'md5': 'a1a1b1a8bc70a89e49ccfd113aed0820',
21         'info_dict': {
22             'id': '5416503',
23             'ext': 'mp4',
24             'title': 'Sultry Striptease',
25             'description': 'md5:d22219c09da287c14bed3d6c37ce4bc2',
26             'timestamp': 1391813355,
27             'upload_date': '20140207',
28             'duration': 383,
29             'tags': list,
30             'age_limit': 18,
31         }
32     }
33
34     def _real_extract(self, url):
35         video_id = self._match_id(url)
36
37         webpage = self._download_webpage(url, video_id)
38
39         cpl_url = self._search_regex(
40             r'<script[^>]+src=(["\'])(?P<url>(?:/static|(?:https?:)?//static\.beeg\.com)/cpl/\d+\.js.*?)\1',
41             webpage, 'cpl', default=None, group='url')
42
43         cpl_url = urljoin(url, cpl_url)
44
45         beeg_version, beeg_salt = [None] * 2
46
47         if cpl_url:
48             cpl = self._download_webpage(
49                 self._proto_relative_url(cpl_url), video_id,
50                 'Downloading cpl JS', fatal=False)
51             if cpl:
52                 beeg_version = int_or_none(self._search_regex(
53                     r'beeg_version\s*=\s*([^\b]+)', cpl,
54                     'beeg version', default=None)) or self._search_regex(
55                     r'/(\d+)\.js', cpl_url, 'beeg version', default=None)
56                 beeg_salt = self._search_regex(
57                     r'beeg_salt\s*=\s*(["\'])(?P<beeg_salt>.+?)\1', cpl, 'beeg salt',
58                     default=None, group='beeg_salt')
59
60         beeg_version = beeg_version or '2185'
61         beeg_salt = beeg_salt or 'pmweAkq8lAYKdfWcFCUj0yoVgoPlinamH5UE1CB3H'
62
63         for api_path in ('', 'api.'):
64             video = self._download_json(
65                 'https://%sbeeg.com/api/v6/%s/video/%s'
66                 % (api_path, beeg_version, video_id), video_id,
67                 fatal=api_path == 'api.')
68             if video:
69                 break
70
71         def split(o, e):
72             def cut(s, x):
73                 n.append(s[:x])
74                 return s[x:]
75             n = []
76             r = len(o) % e
77             if r > 0:
78                 o = cut(o, r)
79             while len(o) > e:
80                 o = cut(o, e)
81             n.append(o)
82             return n
83
84         def decrypt_key(key):
85             # Reverse engineered from http://static.beeg.com/cpl/1738.js
86             a = beeg_salt
87             e = compat_urllib_parse_unquote(key)
88             o = ''.join([
89                 compat_chr(compat_ord(e[n]) - compat_ord(a[n % len(a)]) % 21)
90                 for n in range(len(e))])
91             return ''.join(split(o, 3)[::-1])
92
93         def decrypt_url(encrypted_url):
94             encrypted_url = self._proto_relative_url(
95                 encrypted_url.replace('{DATA_MARKERS}', ''), 'https:')
96             key = self._search_regex(
97                 r'/key=(.*?)%2Cend=', encrypted_url, 'key', default=None)
98             if not key:
99                 return encrypted_url
100             return encrypted_url.replace(key, decrypt_key(key))
101
102         formats = []
103         for format_id, video_url in video.items():
104             if not video_url:
105                 continue
106             height = self._search_regex(
107                 r'^(\d+)[pP]$', format_id, 'height', default=None)
108             if not height:
109                 continue
110             formats.append({
111                 'url': decrypt_url(video_url),
112                 'format_id': format_id,
113                 'height': int(height),
114             })
115         self._sort_formats(formats)
116
117         title = video['title']
118         video_id = video.get('id') or video_id
119         display_id = video.get('code')
120         description = video.get('desc')
121
122         timestamp = parse_iso8601(video.get('date'), ' ')
123         duration = int_or_none(video.get('duration'))
124
125         tags = [tag.strip() for tag in video['tags'].split(',')] if video.get('tags') else None
126
127         return {
128             'id': video_id,
129             'display_id': display_id,
130             'title': title,
131             'description': description,
132             'timestamp': timestamp,
133             'duration': duration,
134             'tags': tags,
135             'formats': formats,
136             'age_limit': self._rta_search(webpage),
137         }