]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/yandexdisk.py
debian/control: Update list of extractors in long description.
[youtubedl] / youtube_dl / extractor / yandexdisk.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 from .common import InfoExtractor
5 from ..compat import compat_str
6 from ..utils import (
7 determine_ext,
8 float_or_none,
9 int_or_none,
10 try_get,
11 urlencode_postdata,
12 )
13
14
15 class YandexDiskIE(InfoExtractor):
16 _VALID_URL = r'https?://yadi\.sk/[di]/(?P<id>[^/?#&]+)'
17
18 _TESTS = [{
19 'url': 'https://yadi.sk/i/VdOeDou8eZs6Y',
20 'md5': '33955d7ae052f15853dc41f35f17581c',
21 'info_dict': {
22 'id': 'VdOeDou8eZs6Y',
23 'ext': 'mp4',
24 'title': '4.mp4',
25 'duration': 168.6,
26 'uploader': 'y.botova',
27 'uploader_id': '300043621',
28 'view_count': int,
29 },
30 }, {
31 'url': 'https://yadi.sk/d/h3WAXvDS3Li3Ce',
32 'only_matching': True,
33 }]
34
35 def _real_extract(self, url):
36 video_id = self._match_id(url)
37
38 status = self._download_webpage(
39 'https://disk.yandex.com/auth/status', video_id, query={
40 'urlOrigin': url,
41 'source': 'public',
42 'md5': 'false',
43 })
44
45 sk = self._search_regex(
46 r'(["\'])sk(?:External)?\1\s*:\s*(["\'])(?P<value>(?:(?!\2).)+)\2',
47 status, 'sk', group='value')
48
49 webpage = self._download_webpage(url, video_id)
50
51 models = self._parse_json(
52 self._search_regex(
53 r'<script[^>]+id=["\']models-client[^>]+>\s*(\[.+?\])\s*</script',
54 webpage, 'video JSON'),
55 video_id)
56
57 data = next(
58 model['data'] for model in models
59 if model.get('model') == 'resource')
60
61 video_hash = data['id']
62 title = data['name']
63
64 models = self._download_json(
65 'https://disk.yandex.com/models/', video_id,
66 data=urlencode_postdata({
67 '_model.0': 'videoInfo',
68 'id.0': video_hash,
69 '_model.1': 'do-get-resource-url',
70 'id.1': video_hash,
71 'version': '13.6',
72 'sk': sk,
73 }), query={'_m': 'videoInfo'})['models']
74
75 videos = try_get(models, lambda x: x[0]['data']['videos'], list) or []
76 source_url = try_get(
77 models, lambda x: x[1]['data']['file'], compat_str)
78
79 formats = []
80 if source_url:
81 formats.append({
82 'url': source_url,
83 'format_id': 'source',
84 'ext': determine_ext(title, 'mp4'),
85 'quality': 1,
86 })
87 for video in videos:
88 format_url = video.get('url')
89 if not format_url:
90 continue
91 if determine_ext(format_url) == 'm3u8':
92 formats.extend(self._extract_m3u8_formats(
93 format_url, video_id, 'mp4', entry_protocol='m3u8_native',
94 m3u8_id='hls', fatal=False))
95 else:
96 formats.append({
97 'url': format_url,
98 })
99 self._sort_formats(formats)
100
101 duration = float_or_none(try_get(
102 models, lambda x: x[0]['data']['duration']), 1000)
103 uploader = try_get(
104 data, lambda x: x['user']['display_name'], compat_str)
105 uploader_id = try_get(
106 data, lambda x: x['user']['uid'], compat_str)
107 view_count = int_or_none(try_get(
108 data, lambda x: x['meta']['views_counter']))
109
110 return {
111 'id': video_id,
112 'title': title,
113 'duration': duration,
114 'uploader': uploader,
115 'uploader_id': uploader_id,
116 'view_count': view_count,
117 'formats': formats,
118 }