[noovo] Fix extraction (closes #14214)
[youtube-dl] / youtube_dl / extractor / openload.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 import re
5
6 from .common import InfoExtractor
7 from ..compat import compat_chr
8 from ..utils import (
9     determine_ext,
10     ExtractorError,
11 )
12
13
14 class OpenloadIE(InfoExtractor):
15     _VALID_URL = r'https?://(?:openload\.(?:co|io)|oload\.tv)/(?:f|embed)/(?P<id>[a-zA-Z0-9-_]+)'
16
17     _TESTS = [{
18         'url': 'https://openload.co/f/kUEfGclsU9o',
19         'md5': 'bf1c059b004ebc7a256f89408e65c36e',
20         'info_dict': {
21             'id': 'kUEfGclsU9o',
22             'ext': 'mp4',
23             'title': 'skyrim_no-audio_1080.mp4',
24             'thumbnail': r're:^https?://.*\.jpg$',
25         },
26     }, {
27         'url': 'https://openload.co/embed/rjC09fkPLYs',
28         'info_dict': {
29             'id': 'rjC09fkPLYs',
30             'ext': 'mp4',
31             'title': 'movie.mp4',
32             'thumbnail': r're:^https?://.*\.jpg$',
33             'subtitles': {
34                 'en': [{
35                     'ext': 'vtt',
36                 }],
37             },
38         },
39         'params': {
40             'skip_download': True,  # test subtitles only
41         },
42     }, {
43         'url': 'https://openload.co/embed/kUEfGclsU9o/skyrim_no-audio_1080.mp4',
44         'only_matching': True,
45     }, {
46         'url': 'https://openload.io/f/ZAn6oz-VZGE/',
47         'only_matching': True,
48     }, {
49         'url': 'https://openload.co/f/_-ztPaZtMhM/',
50         'only_matching': True,
51     }, {
52         # unavailable via https://openload.co/f/Sxz5sADo82g/, different layout
53         # for title and ext
54         'url': 'https://openload.co/embed/Sxz5sADo82g/',
55         'only_matching': True,
56     }, {
57         'url': 'https://oload.tv/embed/KnG-kKZdcfY/',
58         'only_matching': True,
59     }]
60
61     @staticmethod
62     def _extract_urls(webpage):
63         return re.findall(
64             r'<iframe[^>]+src=["\']((?:https?://)?(?:openload\.(?:co|io)|oload\.tv)/embed/[a-zA-Z0-9-_]+)',
65             webpage)
66
67     def _real_extract(self, url):
68         video_id = self._match_id(url)
69         webpage = self._download_webpage('https://openload.co/embed/%s/' % video_id, video_id)
70
71         if 'File not found' in webpage or 'deleted by the owner' in webpage:
72             raise ExtractorError('File not found', expected=True)
73
74         ol_id = self._search_regex(
75             '<span[^>]+id="[^"]+"[^>]*>([0-9A-Za-z]+)</span>',
76             webpage, 'openload ID')
77
78         decoded = ''
79         a = ol_id[0:24]
80         b = []
81         for i in range(0, len(a), 8):
82             b.append(int(a[i:i + 8] or '0', 16))
83         ol_id = ol_id[24:]
84         j = 0
85         k = 0
86         while j < len(ol_id):
87             c = 128
88             d = 0
89             e = 0
90             f = 0
91             _more = True
92             while _more:
93                 if j + 1 >= len(ol_id):
94                     c = 143
95                 f = int(ol_id[j:j + 2] or '0', 16)
96                 j += 2
97                 d += (f & 127) << e
98                 e += 7
99                 _more = f >= c
100             g = d ^ b[k % 3]
101             for i in range(4):
102                 char_dec = (g >> 8 * i) & (c + 127)
103                 char = compat_chr(char_dec)
104                 if char != '#':
105                     decoded += char
106             k += 1
107
108         video_url = 'https://openload.co/stream/%s?mime=true'
109         video_url = video_url % decoded
110
111         title = self._og_search_title(webpage, default=None) or self._search_regex(
112             r'<span[^>]+class=["\']title["\'][^>]*>([^<]+)', webpage,
113             'title', default=None) or self._html_search_meta(
114             'description', webpage, 'title', fatal=True)
115
116         entries = self._parse_html5_media_entries(url, webpage, video_id)
117         subtitles = entries[0]['subtitles'] if entries else None
118
119         info_dict = {
120             'id': video_id,
121             'title': title,
122             'thumbnail': self._og_search_thumbnail(webpage, default=None),
123             'url': video_url,
124             # Seems all videos have extensions in their titles
125             'ext': determine_ext(title, 'mp4'),
126             'subtitles': subtitles,
127         }
128         return info_dict