yt_dlp/extractor/hotstar.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import hashlib
   5 import hmac
   6 import re
   7 import time
   8 import uuid
   9 import json
  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, query, st=None, cookies=None):
  30         st = int_or_none(st) or 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
  35         if cookies and cookies.get('userUP'):
  36             token = cookies.get('userUP').value
  37         else:
  38             token = self._download_json(
  39                 'https://api.hotstar.com/um/v3/users',
  40                 video_id, note='Downloading token',
  41                 data=json.dumps({"device_ids": [{"id": compat_str(uuid.uuid4()), "type": "device_id"}]}).encode('utf-8'),
  42                 headers={
  43                     'hotstarauth': auth,
  44                     'x-hs-platform': 'PCTV',  # or 'web'
  45                     'Content-Type': 'application/json',
  46                 })['user_identity']
  47
  48         response = self._download_json(
  49             'https://api.hotstar.com/' + path, video_id, headers={
  50                 'hotstarauth': auth,
  51                 'x-hs-appversion': '6.72.2',
  52                 'x-hs-platform': 'web',
  53                 'x-hs-usertoken': token,
  54             }, query=query)
  55
  56         if response['message'] != "Playback URL's fetched successfully":
  57             raise ExtractorError(
  58                 response['message'], expected=True)
  59         return response['data']
  60
  61     def _call_api(self, path, video_id, query_name='contentId'):
  62         return self._download_json('https://api.hotstar.com/' + path, video_id=video_id, query={
  63             query_name: video_id,
  64             'tas': 10000,
  65         }, headers={
  66             'x-country-code': 'IN',
  67             'x-platform-code': 'PCTV',
  68         })
  69
  70     def _call_api_v2(self, path, video_id, st=None, cookies=None):
  71         return self._call_api_impl(
  72             '%s/content/%s' % (path, video_id), video_id, st=st, cookies=cookies, query={
  73                 'desired-config': 'audio_channel:stereo|dynamic_range:sdr|encryption:plain|ladder:tv|package:dash|resolution:hd|subs-tag:HotstarVIP|video_codec:vp9',
  74                 'device-id': cookies.get('device_id').value if cookies.get('device_id') else compat_str(uuid.uuid4()),
  75                 'os-name': 'Windows',
  76                 'os-version': '10',
  77             })
  78
  79
  80 class HotStarIE(HotStarBaseIE):
  81     IE_NAME = 'hotstar'
  82     _VALID_URL = r'''(?x)
  83                         (?:
  84                             hotstar\:|
  85                             https?://(?:www\.)?hotstar\.com(?:/in)?/(?!in/)
  86                         )
  87                         (?:
  88                             (?P<type>movies|sports|episode|(?P<tv>tv))
  89                             (?:
  90                                 \:|
  91                                 /[^/?#]+/
  92                                 (?(tv)
  93                                     (?:[^/?#]+/){2}|
  94                                     (?:[^/?#]+/)*
  95                                 )
  96                             )|
  97                             [^/?#]+/
  98                         )?
  99                         (?P<id>\d{10})
 100                    '''
 101     _TESTS = [{
 102         'url': 'https://www.hotstar.com/can-you-not-spread-rumours/1000076273',
 103         'info_dict': {
 104             'id': '1000076273',
 105             'ext': 'mp4',
 106             'title': 'Can You Not Spread Rumours?',
 107             'description': 'md5:c957d8868e9bc793ccb813691cc4c434',
 108             'timestamp': 1447248600,
 109             'upload_date': '20151111',
 110             'duration': 381,
 111         },
 112     }, {
 113         'url': 'hotstar:1000076273',
 114         'only_matching': True,
 115     }, {
 116         'url': 'https://www.hotstar.com/movies/radha-gopalam/1000057157',
 117         'info_dict': {
 118             'id': '1000057157',
 119             'ext': 'mp4',
 120             'title': 'Radha Gopalam',
 121             'description': 'md5:be3bc342cc120bbc95b3b0960e2b0d22',
 122             'timestamp': 1140805800,
 123             'upload_date': '20060224',
 124             'duration': 9182,
 125         },
 126     }, {
 127         'url': 'hotstar:movies:1000057157',
 128         'only_matching': True,
 129     }, {
 130         'url': 'https://www.hotstar.com/in/sports/cricket/follow-the-blues-2021/recap-eng-fight-back-on-day-2/1260066104',
 131         'only_matching': True,
 132     }, {
 133         'url': 'https://www.hotstar.com/in/sports/football/most-costly-pl-transfers-ft-grealish/1260065956',
 134         'only_matching': True,
 135     }, {
 136         # contentData
 137         'url': 'hotstar:sports:1260065956',
 138         'only_matching': True,
 139     }, {
 140         # contentData
 141         'url': 'hotstar:sports:1260066104',
 142         'only_matching': True,
 143     }, {
 144         'url': 'https://www.hotstar.com/tv/ek-bhram-sarvagun-sampanna/s-2116/janhvi-targets-suman/1000234847',
 145         'info_dict': {
 146             'id': '1000234847',
 147             'ext': 'mp4',
 148             'title': 'Janhvi Targets Suman',
 149             'description': 'md5:78a85509348910bd1ca31be898c5796b',
 150             'timestamp': 1556670600,
 151             'upload_date': '20190501',
 152             'duration': 1219,
 153             'channel': 'StarPlus',
 154             'channel_id': 3,
 155             'series': 'Ek Bhram - Sarvagun Sampanna',
 156             'season': 'Chapter 1',
 157             'season_number': 1,
 158             'season_id': 6771,
 159             'episode': 'Janhvi Targets Suman',
 160             'episode_number': 8,
 161         },
 162     }, {
 163         'url': 'hotstar:episode:1000234847',
 164         'only_matching': True,
 165     }]
 166     _GEO_BYPASS = False
 167     _TYPE = {
 168         'movies': 'movie',
 169         'sports': 'match',
 170         'episode': 'episode',
 171         'tv': 'episode',
 172         None: 'content',
 173     }
 174
 175     def _real_extract(self, url):
 176         mobj = self._match_valid_url(url)
 177         video_id = mobj.group('id')
 178         video_type = mobj.group('type')
 179         cookies = self._get_cookies(url)
 180         video_type = self._TYPE.get(video_type, video_type)
 181         video_data = self._call_api(f'o/v1/{video_type}/detail', video_id)['body']['results']['item']
 182         title = video_data['title']
 183
 184         if not self.get_param('allow_unplayable_formats') and video_data.get('drmProtected'):
 185             self.report_drm(video_id)
 186
 187         headers = {'Referer': 'https://www.hotstar.com/in'}
 188         formats = []
 189         subs = {}
 190         geo_restricted = False
 191         _, urlh = self._download_webpage_handle('https://www.hotstar.com/in', video_id)
 192         # Required to fix https://github.com/yt-dlp/yt-dlp/issues/396
 193         st = urlh.headers.get('x-origin-date')
 194         # change to v2 in the future
 195         playback_sets = self._call_api_v2('play/v1/playback', video_id, st=st, cookies=cookies)['playBackSets']
 196         for playback_set in playback_sets:
 197             if not isinstance(playback_set, dict):
 198                 continue
 199             format_url = url_or_none(playback_set.get('playbackUrl'))
 200             if not format_url:
 201                 continue
 202             format_url = re.sub(
 203                 r'(?<=//staragvod)(\d)', r'web\1', format_url)
 204             tags = str_or_none(playback_set.get('tagsCombination')) or ''
 205             if tags and 'encryption:plain' not in tags:
 206                 continue
 207             ext = determine_ext(format_url)
 208             try:
 209                 if 'package:hls' in tags or ext == 'm3u8':
 210                     hls_formats, hls_subs = self._extract_m3u8_formats_and_subtitles(
 211                         format_url, video_id, 'mp4',
 212                         entry_protocol='m3u8_native',
 213                         m3u8_id='hls', headers=headers)
 214                     formats.extend(hls_formats)
 215                     subs = self._merge_subtitles(subs, hls_subs)
 216                 elif 'package:dash' in tags or ext == 'mpd':
 217                     dash_formats, dash_subs = self._extract_mpd_formats_and_subtitles(
 218                         format_url, video_id, mpd_id='dash', headers=headers)
 219                     formats.extend(dash_formats)
 220                     subs = self._merge_subtitles(subs, dash_subs)
 221                 elif ext == 'f4m':
 222                     # produce broken files
 223                     pass
 224                 else:
 225                     formats.append({
 226                         'url': format_url,
 227                         'width': int_or_none(playback_set.get('width')),
 228                         'height': int_or_none(playback_set.get('height')),
 229                     })
 230             except ExtractorError as e:
 231                 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
 232                     geo_restricted = True
 233                 continue
 234         if not formats and geo_restricted:
 235             self.raise_geo_restricted(countries=['IN'], metadata_available=True)
 236         self._sort_formats(formats)
 237
 238         for f in formats:
 239             f.setdefault('http_headers', {}).update(headers)
 240
 241         return {
 242             'id': video_id,
 243             'title': title,
 244             'description': video_data.get('description'),
 245             'duration': int_or_none(video_data.get('duration')),
 246             'timestamp': int_or_none(video_data.get('broadcastDate') or video_data.get('startDate')),
 247             'formats': formats,
 248             'subtitles': subs,
 249             'channel': video_data.get('channelName'),
 250             'channel_id': video_data.get('channelId'),
 251             'series': video_data.get('showName'),
 252             'season': video_data.get('seasonName'),
 253             'season_number': int_or_none(video_data.get('seasonNo')),
 254             'season_id': video_data.get('seasonId'),
 255             'episode': title,
 256             'episode_number': int_or_none(video_data.get('episodeNo')),
 257         }
 258
 259
 260 class HotStarPlaylistIE(HotStarBaseIE):
 261     IE_NAME = 'hotstar:playlist'
 262     _VALID_URL = r'https?://(?:www\.)?hotstar\.com/tv/[^/]+/s-\w+/list/[^/]+/t-(?P<id>\w+)'
 263     _TESTS = [{
 264         'url': 'https://www.hotstar.com/tv/savdhaan-india/s-26/list/popular-clips/t-3_2_26',
 265         'info_dict': {
 266             'id': '3_2_26',
 267         },
 268         'playlist_mincount': 20,
 269     }, {
 270         'url': 'https://www.hotstar.com/tv/savdhaan-india/s-26/list/extras/t-2480',
 271         'only_matching': True,
 272     }]
 273
 274     def _real_extract(self, url):
 275         playlist_id = self._match_id(url)
 276
 277         collection = self._call_api('o/v1/tray/find', playlist_id, 'uqId')['body']['results']
 278         entries = [
 279             self.url_result(
 280                 'https://www.hotstar.com/%s' % video['contentId'],
 281                 ie=HotStarIE.ie_key(), video_id=video['contentId'])
 282             for video in collection['assets']['items']
 283             if video.get('contentId')]
 284
 285         return self.playlist_result(entries, playlist_id)
 286
 287
 288 class HotStarSeriesIE(HotStarBaseIE):
 289     IE_NAME = 'hotstar:series'
 290     _VALID_URL = r'(?:https?://)(?:www\.)?hotstar\.com(?:/in)?/tv/[^/]+/(?P<id>\d+)'
 291     _TESTS = [{
 292         'url': 'https://www.hotstar.com/in/tv/radhakrishn/1260000646',
 293         'info_dict': {
 294             'id': '1260000646',
 295         },
 296         'playlist_mincount': 690,
 297     }, {
 298         'url': 'https://www.hotstar.com/tv/dancee-/1260050431',
 299         'info_dict': {
 300             'id': '1260050431',
 301         },
 302         'playlist_mincount': 43,
 303     }, {
 304         'url': 'https://www.hotstar.com/in/tv/mahabharat/435/',
 305         'info_dict': {
 306             'id': '435',
 307         },
 308         'playlist_mincount': 269,
 309     }]
 310
 311     def _real_extract(self, url):
 312         series_id = self._match_id(url)
 313         headers = {
 314             'x-country-code': 'IN',
 315             'x-platform-code': 'PCTV',
 316         }
 317         detail_json = self._download_json('https://api.hotstar.com/o/v1/show/detail?contentId=' + series_id,
 318                                           video_id=series_id, headers=headers)
 319         id = compat_str(try_get(detail_json, lambda x: x['body']['results']['item']['id'], int))
 320         item_json = self._download_json('https://api.hotstar.com/o/v1/tray/g/1/items?etid=0&tao=0&tas=10000&eid=' + id,
 321                                         video_id=series_id, headers=headers)
 322         entries = [
 323             self.url_result(
 324                 'hotstar:episode:%d' % video['contentId'],
 325                 ie=HotStarIE.ie_key(), video_id=video['contentId'])
 326             for video in item_json['body']['results']['items']
 327             if video.get('contentId')]
 328
 329         return self.playlist_result(entries, series_id)