]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/hotstar.py
Fix extraction from youtube.
[youtubedl] / youtube_dl / extractor / hotstar.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 import hashlib
5 import hmac
6 import time
7
8 from .common import InfoExtractor
9 from ..compat import compat_HTTPError
10 from ..utils import (
11 determine_ext,
12 ExtractorError,
13 int_or_none,
14 try_get,
15 )
16
17
18 class HotStarBaseIE(InfoExtractor):
19 _AKAMAI_ENCRYPTION_KEY = b'\x05\xfc\x1a\x01\xca\xc9\x4b\xc4\x12\xfc\x53\x12\x07\x75\xf9\xee'
20
21 def _call_api(self, path, video_id, query_name='contentId'):
22 st = int(time.time())
23 exp = st + 6000
24 auth = 'st=%d~exp=%d~acl=/*' % (st, exp)
25 auth += '~hmac=' + hmac.new(self._AKAMAI_ENCRYPTION_KEY, auth.encode(), hashlib.sha256).hexdigest()
26 response = self._download_json(
27 'https://api.hotstar.com/' + path,
28 video_id, headers={
29 'hotstarauth': auth,
30 'x-country-code': 'IN',
31 'x-platform-code': 'JIO',
32 }, query={
33 query_name: video_id,
34 'tas': 10000,
35 })
36 if response['statusCode'] != 'OK':
37 raise ExtractorError(
38 response['body']['message'], expected=True)
39 return response['body']['results']
40
41
42 class HotStarIE(HotStarBaseIE):
43 IE_NAME = 'hotstar'
44 _VALID_URL = r'https?://(?:www\.)?hotstar\.com/(?:.+?[/-])?(?P<id>\d{10})'
45 _TESTS = [{
46 # contentData
47 'url': 'https://www.hotstar.com/can-you-not-spread-rumours/1000076273',
48 'info_dict': {
49 'id': '1000076273',
50 'ext': 'mp4',
51 'title': 'Can You Not Spread Rumours?',
52 'description': 'md5:c957d8868e9bc793ccb813691cc4c434',
53 'timestamp': 1447248600,
54 'upload_date': '20151111',
55 'duration': 381,
56 },
57 'params': {
58 # m3u8 download
59 'skip_download': True,
60 }
61 }, {
62 # contentDetail
63 'url': 'https://www.hotstar.com/movies/radha-gopalam/1000057157',
64 'only_matching': True,
65 }, {
66 'url': 'http://www.hotstar.com/sports/cricket/rajitha-sizzles-on-debut-with-329/2001477583',
67 'only_matching': True,
68 }, {
69 'url': 'http://www.hotstar.com/1000000515',
70 'only_matching': True,
71 }]
72 _GEO_BYPASS = False
73
74 def _real_extract(self, url):
75 video_id = self._match_id(url)
76
77 webpage = self._download_webpage(url, video_id)
78 app_state = self._parse_json(self._search_regex(
79 r'<script>window\.APP_STATE\s*=\s*({.+?})</script>',
80 webpage, 'app state'), video_id)
81 video_data = {}
82 getters = list(
83 lambda x, k=k: x['initialState']['content%s' % k]['content']
84 for k in ('Data', 'Detail')
85 )
86 for v in app_state.values():
87 content = try_get(v, getters, dict)
88 if content and content.get('contentId') == video_id:
89 video_data = content
90 break
91
92 title = video_data['title']
93
94 if video_data.get('drmProtected'):
95 raise ExtractorError('This video is DRM protected.', expected=True)
96
97 formats = []
98 format_data = self._call_api('h/v1/play', video_id)['item']
99 format_url = format_data['playbackUrl']
100 ext = determine_ext(format_url)
101 if ext == 'm3u8':
102 try:
103 formats.extend(self._extract_m3u8_formats(
104 format_url, video_id, 'mp4', m3u8_id='hls'))
105 except ExtractorError as e:
106 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
107 self.raise_geo_restricted(countries=['IN'])
108 raise
109 elif ext == 'f4m':
110 # produce broken files
111 pass
112 else:
113 formats.append({
114 'url': format_url,
115 'width': int_or_none(format_data.get('width')),
116 'height': int_or_none(format_data.get('height')),
117 })
118 self._sort_formats(formats)
119
120 return {
121 'id': video_id,
122 'title': title,
123 'description': video_data.get('description'),
124 'duration': int_or_none(video_data.get('duration')),
125 'timestamp': int_or_none(video_data.get('broadcastDate') or video_data.get('startDate')),
126 'formats': formats,
127 'channel': video_data.get('channelName'),
128 'channel_id': video_data.get('channelId'),
129 'series': video_data.get('showName'),
130 'season': video_data.get('seasonName'),
131 'season_number': int_or_none(video_data.get('seasonNo')),
132 'season_id': video_data.get('seasonId'),
133 'episode': title,
134 'episode_number': int_or_none(video_data.get('episodeNo')),
135 }
136
137
138 class HotStarPlaylistIE(HotStarBaseIE):
139 IE_NAME = 'hotstar:playlist'
140 _VALID_URL = r'https?://(?:www\.)?hotstar\.com/tv/[^/]+/s-\w+/list/[^/]+/t-(?P<id>\w+)'
141 _TESTS = [{
142 'url': 'https://www.hotstar.com/tv/savdhaan-india/s-26/list/popular-clips/t-3_2_26',
143 'info_dict': {
144 'id': '3_2_26',
145 },
146 'playlist_mincount': 20,
147 }, {
148 'url': 'https://www.hotstar.com/tv/savdhaan-india/s-26/list/extras/t-2480',
149 'only_matching': True,
150 }]
151
152 def _real_extract(self, url):
153 playlist_id = self._match_id(url)
154
155 collection = self._call_api('o/v1/tray/find', playlist_id, 'uqId')
156
157 entries = [
158 self.url_result(
159 'https://www.hotstar.com/%s' % video['contentId'],
160 ie=HotStarIE.ie_key(), video_id=video['contentId'])
161 for video in collection['assets']['items']
162 if video.get('contentId')]
163
164 return self.playlist_result(entries, playlist_id)