youtube_dl/extractor/hotstar.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import hashlib
   5 import hmac
   6 import json
   7 import re
   8 import time
   9 import uuid
  10
  11 from .common import InfoExtractor
  12 from ..compat import (
  13     compat_HTTPError,
  14     compat_str,
  15 )
  16 from ..utils import (
  17     determine_ext,
  18     ExtractorError,
  19     int_or_none,
  20     str_or_none,
  21     try_get,
  22     url_or_none,
  23 )
  24
  25
  26 class HotStarBaseIE(InfoExtractor):
  27     _AKAMAI_ENCRYPTION_KEY = b'\x05\xfc\x1a\x01\xca\xc9\x4b\xc4\x12\xfc\x53\x12\x07\x75\xf9\xee'
  28
  29     def _call_api_impl(self, path, video_id, headers, query, data=None):
  30         st = int(time.time())
  31         exp = st + 6000
  32         auth = 'st=%d~exp=%d~acl=/*' % (st, exp)
  33         auth += '~hmac=' + hmac.new(self._AKAMAI_ENCRYPTION_KEY, auth.encode(), hashlib.sha256).hexdigest()
  34         h = {'hotstarauth': auth}
  35         h.update(headers)
  36         return self._download_json(
  37             'https://api.hotstar.com/' + path,
  38             video_id, headers=h, query=query, data=data)
  39
  40     def _call_api(self, path, video_id, query_name='contentId'):
  41         response = self._call_api_impl(path, video_id, {
  42             'x-country-code': 'IN',
  43             'x-platform-code': 'JIO',
  44         }, {
  45             query_name: video_id,
  46             'tas': 10000,
  47         })
  48         if response['statusCode'] != 'OK':
  49             raise ExtractorError(
  50                 response['body']['message'], expected=True)
  51         return response['body']['results']
  52
  53     def _call_api_v2(self, path, video_id, headers, query=None, data=None):
  54         h = {'X-Request-Id': compat_str(uuid.uuid4())}
  55         h.update(headers)
  56         try:
  57             return self._call_api_impl(
  58                 path, video_id, h, query, data)
  59         except ExtractorError as e:
  60             if isinstance(e.cause, compat_HTTPError):
  61                 if e.cause.code == 402:
  62                     self.raise_login_required()
  63                 message = self._parse_json(e.cause.read().decode(), video_id)['message']
  64                 if message in ('Content not available in region', 'Country is not supported'):
  65                     raise self.raise_geo_restricted(message)
  66                 raise ExtractorError(message)
  67             raise e
  68
  69
  70 class HotStarIE(HotStarBaseIE):
  71     IE_NAME = 'hotstar'
  72     _VALID_URL = r'https?://(?:www\.)?hotstar\.com/(?:.+[/-])?(?P<id>\d{10})'
  73     _TESTS = [{
  74         # contentData
  75         'url': 'https://www.hotstar.com/can-you-not-spread-rumours/1000076273',
  76         'info_dict': {
  77             'id': '1000076273',
  78             'ext': 'mp4',
  79             'title': 'Can You Not Spread Rumours?',
  80             'description': 'md5:c957d8868e9bc793ccb813691cc4c434',
  81             'timestamp': 1447248600,
  82             'upload_date': '20151111',
  83             'duration': 381,
  84         },
  85         'params': {
  86             # m3u8 download
  87             'skip_download': True,
  88         }
  89     }, {
  90         # contentDetail
  91         'url': 'https://www.hotstar.com/movies/radha-gopalam/1000057157',
  92         'only_matching': True,
  93     }, {
  94         'url': 'http://www.hotstar.com/sports/cricket/rajitha-sizzles-on-debut-with-329/2001477583',
  95         'only_matching': True,
  96     }, {
  97         'url': 'http://www.hotstar.com/1000000515',
  98         'only_matching': True,
  99     }, {
 100         # only available via api v2
 101         'url': 'https://www.hotstar.com/tv/ek-bhram-sarvagun-sampanna/s-2116/janhvi-targets-suman/1000234847',
 102         'only_matching': True,
 103     }, {
 104         'url': 'https://www.hotstar.com/in/tv/start-music/1260005217/cooks-vs-comalis/1100039717',
 105         'only_matching': True,
 106     }]
 107     _GEO_BYPASS = False
 108     _DEVICE_ID = None
 109     _USER_TOKEN = None
 110
 111     def _real_extract(self, url):
 112         video_id = self._match_id(url)
 113
 114         webpage = self._download_webpage(url, video_id)
 115         app_state = self._parse_json(self._search_regex(
 116             r'<script>window\.APP_STATE\s*=\s*({.+?})</script>',
 117             webpage, 'app state'), video_id)
 118         video_data = {}
 119         getters = list(
 120             lambda x, k=k: x['initialState']['content%s' % k]['content']
 121             for k in ('Data', 'Detail')
 122         )
 123         for v in app_state.values():
 124             content = try_get(v, getters, dict)
 125             if content and content.get('contentId') == video_id:
 126                 video_data = content
 127                 break
 128
 129         title = video_data['title']
 130
 131         if video_data.get('drmProtected'):
 132             raise ExtractorError('This video is DRM protected.', expected=True)
 133
 134         headers = {'Referer': url}
 135         formats = []
 136         geo_restricted = False
 137
 138         if not self._USER_TOKEN:
 139             self._DEVICE_ID = compat_str(uuid.uuid4())
 140             self._USER_TOKEN = self._call_api_v2('um/v3/users', video_id, {
 141                 'X-HS-Platform': 'PCTV',
 142                 'Content-Type': 'application/json',
 143             }, data=json.dumps({
 144                 'device_ids': [{
 145                     'id': self._DEVICE_ID,
 146                     'type': 'device_id',
 147                 }],
 148             }).encode())['user_identity']
 149
 150         playback_sets = self._call_api_v2(
 151             'play/v2/playback/content/' + video_id, video_id, {
 152                 'X-HS-Platform': 'web',
 153                 'X-HS-AppVersion': '6.99.1',
 154                 'X-HS-UserToken': self._USER_TOKEN,
 155             }, query={
 156                 'device-id': self._DEVICE_ID,
 157                 'desired-config': 'encryption:plain',
 158                 'os-name': 'Windows',
 159                 'os-version': '10',
 160             })['data']['playBackSets']
 161         for playback_set in playback_sets:
 162             if not isinstance(playback_set, dict):
 163                 continue
 164             format_url = url_or_none(playback_set.get('playbackUrl'))
 165             if not format_url:
 166                 continue
 167             format_url = re.sub(
 168                 r'(?<=//staragvod)(\d)', r'web\1', format_url)
 169             tags = str_or_none(playback_set.get('tagsCombination')) or ''
 170             if tags and 'encryption:plain' not in tags:
 171                 continue
 172             ext = determine_ext(format_url)
 173             try:
 174                 if 'package:hls' in tags or ext == 'm3u8':
 175                     formats.extend(self._extract_m3u8_formats(
 176                         format_url, video_id, 'mp4',
 177                         entry_protocol='m3u8_native',
 178                         m3u8_id='hls', headers=headers))
 179                 elif 'package:dash' in tags or ext == 'mpd':
 180                     formats.extend(self._extract_mpd_formats(
 181                         format_url, video_id, mpd_id='dash', headers=headers))
 182                 elif ext == 'f4m':
 183                     # produce broken files
 184                     pass
 185                 else:
 186                     formats.append({
 187                         'url': format_url,
 188                         'width': int_or_none(playback_set.get('width')),
 189                         'height': int_or_none(playback_set.get('height')),
 190                     })
 191             except ExtractorError as e:
 192                 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
 193                     geo_restricted = True
 194                 continue
 195         if not formats and geo_restricted:
 196             self.raise_geo_restricted(countries=['IN'])
 197         self._sort_formats(formats)
 198
 199         for f in formats:
 200             f.setdefault('http_headers', {}).update(headers)
 201
 202         image = try_get(video_data, lambda x: x['image']['h'], compat_str)
 203
 204         return {
 205             'id': video_id,
 206             'title': title,
 207             'thumbnail': 'https://img1.hotstarext.com/image/upload/' + image if image else None,
 208             'description': video_data.get('description'),
 209             'duration': int_or_none(video_data.get('duration')),
 210             'timestamp': int_or_none(video_data.get('broadcastDate') or video_data.get('startDate')),
 211             'formats': formats,
 212             'channel': video_data.get('channelName'),
 213             'channel_id': str_or_none(video_data.get('channelId')),
 214             'series': video_data.get('showName'),
 215             'season': video_data.get('seasonName'),
 216             'season_number': int_or_none(video_data.get('seasonNo')),
 217             'season_id': str_or_none(video_data.get('seasonId')),
 218             'episode': title,
 219             'episode_number': int_or_none(video_data.get('episodeNo')),
 220         }
 221
 222
 223 class HotStarPlaylistIE(HotStarBaseIE):
 224     IE_NAME = 'hotstar:playlist'
 225     _VALID_URL = r'https?://(?:www\.)?hotstar\.com/(?:[a-z]{2}/)?tv/[^/]+/s-\w+/list/[^/]+/t-(?P<id>\w+)'
 226     _TESTS = [{
 227         'url': 'https://www.hotstar.com/tv/savdhaan-india/s-26/list/popular-clips/t-3_2_26',
 228         'info_dict': {
 229             'id': '3_2_26',
 230         },
 231         'playlist_mincount': 20,
 232     }, {
 233         'url': 'https://www.hotstar.com/tv/savdhaan-india/s-26/list/extras/t-2480',
 234         'only_matching': True,
 235     }, {
 236         'url': 'https://www.hotstar.com/us/tv/masterchef-india/s-830/list/episodes/t-1_2_830',
 237         'only_matching': True,
 238     }]
 239
 240     def _real_extract(self, url):
 241         playlist_id = self._match_id(url)
 242
 243         collection = self._call_api('o/v1/tray/find', playlist_id, 'uqId')
 244
 245         entries = [
 246             self.url_result(
 247                 'https://www.hotstar.com/%s' % video['contentId'],
 248                 ie=HotStarIE.ie_key(), video_id=video['contentId'])
 249             for video in collection['assets']['items']
 250             if video.get('contentId')]
 251
 252         return self.playlist_result(entries, playlist_id)