yt_dlp/extractor/hotstar.py

   1 import hashlib
   2 import hmac
   3 import json
   4 import re
   5 import time
   6 import uuid
   7
   8 from .common import InfoExtractor
   9 from ..compat import compat_HTTPError, compat_str
  10 from ..utils import (
  11     ExtractorError,
  12     determine_ext,
  13     int_or_none,
  14     join_nonempty,
  15     str_or_none,
  16     traverse_obj,
  17     url_or_none,
  18 )
  19
  20
  21 class HotStarBaseIE(InfoExtractor):
  22     _BASE_URL = 'https://www.hotstar.com'
  23     _API_URL = 'https://api.hotstar.com'
  24     _AKAMAI_ENCRYPTION_KEY = b'\x05\xfc\x1a\x01\xca\xc9\x4b\xc4\x12\xfc\x53\x12\x07\x75\xf9\xee'
  25
  26     def _call_api_v1(self, path, *args, **kwargs):
  27         return self._download_json(
  28             f'{self._API_URL}/o/v1/{path}', *args, **kwargs,
  29             headers={'x-country-code': 'IN', 'x-platform-code': 'PCTV'})
  30
  31     def _call_api_impl(self, path, video_id, query, st=None, cookies=None):
  32         st = int_or_none(st) or int(time.time())
  33         exp = st + 6000
  34         auth = 'st=%d~exp=%d~acl=/*' % (st, exp)
  35         auth += '~hmac=' + hmac.new(self._AKAMAI_ENCRYPTION_KEY, auth.encode(), hashlib.sha256).hexdigest()
  36
  37         if cookies and cookies.get('userUP'):
  38             token = cookies.get('userUP').value
  39         else:
  40             token = self._download_json(
  41                 f'{self._API_URL}/um/v3/users',
  42                 video_id, note='Downloading token',
  43                 data=json.dumps({"device_ids": [{"id": compat_str(uuid.uuid4()), "type": "device_id"}]}).encode('utf-8'),
  44                 headers={
  45                     'hotstarauth': auth,
  46                     'x-hs-platform': 'PCTV',  # or 'web'
  47                     'Content-Type': 'application/json',
  48                 })['user_identity']
  49
  50         response = self._download_json(
  51             f'{self._API_URL}/{path}', video_id, query=query,
  52             headers={
  53                 'hotstarauth': auth,
  54                 'x-hs-appversion': '6.72.2',
  55                 'x-hs-platform': 'web',
  56                 'x-hs-usertoken': token,
  57             })
  58
  59         if response['message'] != "Playback URL's fetched successfully":
  60             raise ExtractorError(
  61                 response['message'], expected=True)
  62         return response['data']
  63
  64     def _call_api_v2(self, path, video_id, st=None, cookies=None):
  65         return self._call_api_impl(
  66             f'{path}/content/{video_id}', video_id, st=st, cookies=cookies, query={
  67                 'desired-config': 'audio_channel:stereo|container:fmp4|dynamic_range:hdr|encryption:plain|ladder:tv|package:dash|resolution:fhd|subs-tag:HotstarVIP|video_codec:h265',
  68                 'device-id': cookies.get('device_id').value if cookies.get('device_id') else compat_str(uuid.uuid4()),
  69                 'os-name': 'Windows',
  70                 'os-version': '10',
  71             })
  72
  73     def _playlist_entries(self, path, item_id, root=None, **kwargs):
  74         results = self._call_api_v1(path, item_id, **kwargs)['body']['results']
  75         for video in traverse_obj(results, (('assets', None), 'items', ...)):
  76             if video.get('contentId'):
  77                 yield self.url_result(
  78                     HotStarIE._video_url(video['contentId'], root=root), HotStarIE, video['contentId'])
  79
  80
  81 class HotStarIE(HotStarBaseIE):
  82     IE_NAME = 'hotstar'
  83     _VALID_URL = r'''(?x)
  84         https?://(?:www\.)?hotstar\.com(?:/in)?/(?!in/)
  85         (?:
  86             (?P<type>movies|sports|episode|(?P<tv>tv))/
  87             (?(tv)(?:[^/?#]+/){2}|[^?#]*)
  88         )?
  89         [^/?#]+/
  90         (?P<id>\d{10})
  91     '''
  92
  93     _TESTS = [{
  94         'url': 'https://www.hotstar.com/can-you-not-spread-rumours/1000076273',
  95         'info_dict': {
  96             'id': '1000076273',
  97             'ext': 'mp4',
  98             'title': 'Can You Not Spread Rumours?',
  99             'description': 'md5:c957d8868e9bc793ccb813691cc4c434',
 100             'timestamp': 1447248600,
 101             'upload_date': '20151111',
 102             'duration': 381,
 103             'episode': 'Can You Not Spread Rumours?',
 104         },
 105         'params': {'skip_download': 'm3u8'},
 106     }, {
 107         'url': 'https://www.hotstar.com/tv/ek-bhram-sarvagun-sampanna/s-2116/janhvi-targets-suman/1000234847',
 108         'info_dict': {
 109             'id': '1000234847',
 110             'ext': 'mp4',
 111             'title': 'Janhvi Targets Suman',
 112             'description': 'md5:78a85509348910bd1ca31be898c5796b',
 113             'timestamp': 1556670600,
 114             'upload_date': '20190501',
 115             'duration': 1219,
 116             'channel': 'StarPlus',
 117             'channel_id': 3,
 118             'series': 'Ek Bhram - Sarvagun Sampanna',
 119             'season': 'Chapter 1',
 120             'season_number': 1,
 121             'season_id': 6771,
 122             'episode': 'Janhvi Targets Suman',
 123             'episode_number': 8,
 124         }
 125     }, {
 126         'url': 'https://www.hotstar.com/movies/radha-gopalam/1000057157',
 127         'only_matching': True,
 128     }, {
 129         'url': 'https://www.hotstar.com/in/sports/cricket/follow-the-blues-2021/recap-eng-fight-back-on-day-2/1260066104',
 130         'only_matching': True,
 131     }, {
 132         'url': 'https://www.hotstar.com/in/sports/football/most-costly-pl-transfers-ft-grealish/1260065956',
 133         'only_matching': True,
 134     }]
 135     _GEO_BYPASS = False
 136
 137     _TYPE = {
 138         'movies': 'movie',
 139         'sports': 'match',
 140         'episode': 'episode',
 141         'tv': 'episode',
 142         None: 'content',
 143     }
 144
 145     _IGNORE_MAP = {
 146         'res': 'resolution',
 147         'vcodec': 'video_codec',
 148         'dr': 'dynamic_range',
 149     }
 150
 151     _TAG_FIELDS = {
 152         'language': 'language',
 153         'acodec': 'audio_codec',
 154         'vcodec': 'video_codec',
 155     }
 156
 157     @classmethod
 158     def _video_url(cls, video_id, video_type=None, *, slug='ignore_me', root=None):
 159         assert None in (video_type, root)
 160         if not root:
 161             root = join_nonempty(cls._BASE_URL, video_type, delim='/')
 162         return f'{root}/{slug}/{video_id}'
 163
 164     def _real_extract(self, url):
 165         video_id, video_type = self._match_valid_url(url).group('id', 'type')
 166         video_type = self._TYPE.get(video_type, video_type)
 167         cookies = self._get_cookies(url)  # Cookies before any request
 168
 169         video_data = self._call_api_v1(f'{video_type}/detail', video_id,
 170                                        query={'tas': 10000, 'contentId': video_id})['body']['results']['item']
 171         if not self.get_param('allow_unplayable_formats') and video_data.get('drmProtected'):
 172             self.report_drm(video_id)
 173
 174         # See https://github.com/yt-dlp/yt-dlp/issues/396
 175         st = self._download_webpage_handle(f'{self._BASE_URL}/in', video_id)[1].headers.get('x-origin-date')
 176
 177         geo_restricted = False
 178         formats, subs = [], {}
 179         headers = {'Referer': f'{self._BASE_URL}/in'}
 180
 181         # change to v2 in the future
 182         playback_sets = self._call_api_v2('play/v1/playback', video_id, st=st, cookies=cookies)['playBackSets']
 183         for playback_set in playback_sets:
 184             if not isinstance(playback_set, dict):
 185                 continue
 186             tags = str_or_none(playback_set.get('tagsCombination')) or ''
 187             if any(f'{prefix}:{ignore}' in tags
 188                    for key, prefix in self._IGNORE_MAP.items()
 189                    for ignore in self._configuration_arg(key)):
 190                 continue
 191             tag_dict = dict((t.split(':', 1) + [None])[:2] for t in tags.split(';'))
 192
 193             format_url = url_or_none(playback_set.get('playbackUrl'))
 194             if not format_url:
 195                 continue
 196             format_url = re.sub(r'(?<=//staragvod)(\d)', r'web\1', format_url)
 197             ext = determine_ext(format_url)
 198
 199             current_formats, current_subs = [], {}
 200             try:
 201                 if 'package:hls' in tags or ext == 'm3u8':
 202                     current_formats, current_subs = self._extract_m3u8_formats_and_subtitles(
 203                         format_url, video_id, ext='mp4', headers=headers)
 204                 elif 'package:dash' in tags or ext == 'mpd':
 205                     current_formats, current_subs = self._extract_mpd_formats_and_subtitles(
 206                         format_url, video_id, headers=headers)
 207                 elif ext == 'f4m':
 208                     pass  # XXX: produce broken files
 209                 else:
 210                     current_formats = [{
 211                         'url': format_url,
 212                         'width': int_or_none(playback_set.get('width')),
 213                         'height': int_or_none(playback_set.get('height')),
 214                     }]
 215             except ExtractorError as e:
 216                 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
 217                     geo_restricted = True
 218                 continue
 219
 220             if tag_dict.get('encryption') not in ('plain', None):
 221                 for f in current_formats:
 222                     f['has_drm'] = True
 223             for f in current_formats:
 224                 for k, v in self._TAG_FIELDS.items():
 225                     if not f.get(k):
 226                         f[k] = tag_dict.get(v)
 227                 if f.get('vcodec') != 'none' and not f.get('dynamic_range'):
 228                     f['dynamic_range'] = tag_dict.get('dynamic_range')
 229                 if f.get('acodec') != 'none' and not f.get('audio_channels'):
 230                     f['audio_channels'] = {
 231                         'stereo': 2,
 232                         'dolby51': 6,
 233                     }.get(tag_dict.get('audio_channel'))
 234                 f['format_note'] = join_nonempty(
 235                     tag_dict.get('ladder'),
 236                     tag_dict.get('audio_channel') if f.get('acodec') != 'none' else None,
 237                     f.get('format_note'),
 238                     delim=', ')
 239
 240             formats.extend(current_formats)
 241             subs = self._merge_subtitles(subs, current_subs)
 242
 243         if not formats and geo_restricted:
 244             self.raise_geo_restricted(countries=['IN'], metadata_available=True)
 245         self._remove_duplicate_formats(formats)
 246         for f in formats:
 247             f.setdefault('http_headers', {}).update(headers)
 248
 249         return {
 250             'id': video_id,
 251             'title': video_data.get('title'),
 252             'description': video_data.get('description'),
 253             'duration': int_or_none(video_data.get('duration')),
 254             'timestamp': int_or_none(traverse_obj(video_data, 'broadcastDate', 'startDate')),
 255             'formats': formats,
 256             'subtitles': subs,
 257             'channel': video_data.get('channelName'),
 258             'channel_id': video_data.get('channelId'),
 259             'series': video_data.get('showName'),
 260             'season': video_data.get('seasonName'),
 261             'season_number': int_or_none(video_data.get('seasonNo')),
 262             'season_id': video_data.get('seasonId'),
 263             'episode': video_data.get('title'),
 264             'episode_number': int_or_none(video_data.get('episodeNo')),
 265         }
 266
 267
 268 class HotStarPrefixIE(InfoExtractor):
 269     """ The "hotstar:" prefix is no longer in use, but this is kept for backward compatibility """
 270     IE_DESC = False
 271     _VALID_URL = r'hotstar:(?:(?P<type>\w+):)?(?P<id>\d+)$'
 272     _TESTS = [{
 273         'url': 'hotstar:1000076273',
 274         'only_matching': True,
 275     }, {
 276         'url': 'hotstar:movies:1260009879',
 277         'info_dict': {
 278             'id': '1260009879',
 279             'ext': 'mp4',
 280             'title': 'Nuvvu Naaku Nachav',
 281             'description': 'md5:d43701b1314e6f8233ce33523c043b7d',
 282             'timestamp': 1567525674,
 283             'upload_date': '20190903',
 284             'duration': 10787,
 285             'episode': 'Nuvvu Naaku Nachav',
 286         },
 287     }, {
 288         'url': 'hotstar:episode:1000234847',
 289         'only_matching': True,
 290     }, {
 291         # contentData
 292         'url': 'hotstar:sports:1260065956',
 293         'only_matching': True,
 294     }, {
 295         # contentData
 296         'url': 'hotstar:sports:1260066104',
 297         'only_matching': True,
 298     }]
 299
 300     def _real_extract(self, url):
 301         video_id, video_type = self._match_valid_url(url).group('id', 'type')
 302         return self.url_result(HotStarIE._video_url(video_id, video_type), HotStarIE, video_id)
 303
 304
 305 class HotStarPlaylistIE(HotStarBaseIE):
 306     IE_NAME = 'hotstar:playlist'
 307     _VALID_URL = r'https?://(?:www\.)?hotstar\.com(?:/in)?/tv(?:/[^/]+){2}/list/[^/]+/t-(?P<id>\w+)'
 308     _TESTS = [{
 309         'url': 'https://www.hotstar.com/tv/savdhaan-india/s-26/list/popular-clips/t-3_2_26',
 310         'info_dict': {
 311             'id': '3_2_26',
 312         },
 313         'playlist_mincount': 20,
 314     }, {
 315         'url': 'https://www.hotstar.com/tv/savdhaan-india/s-26/list/extras/t-2480',
 316         'only_matching': True,
 317     }, {
 318         'url': 'https://www.hotstar.com/in/tv/karthika-deepam/15457/list/popular-clips/t-3_2_1272',
 319         'only_matching': True,
 320     }]
 321
 322     def _real_extract(self, url):
 323         id_ = self._match_id(url)
 324         return self.playlist_result(
 325             self._playlist_entries('tray/find', id_, query={'tas': 10000, 'uqId': id_}), id_)
 326
 327
 328 class HotStarSeasonIE(HotStarBaseIE):
 329     IE_NAME = 'hotstar:season'
 330     _VALID_URL = r'(?P<url>https?://(?:www\.)?hotstar\.com(?:/in)?/tv/[^/]+/\w+)/seasons/[^/]+/ss-(?P<id>\w+)'
 331     _TESTS = [{
 332         'url': 'https://www.hotstar.com/tv/radhakrishn/1260000646/seasons/season-2/ss-8028',
 333         'info_dict': {
 334             'id': '8028',
 335         },
 336         'playlist_mincount': 35,
 337     }, {
 338         'url': 'https://www.hotstar.com/in/tv/ishqbaaz/9567/seasons/season-2/ss-4357',
 339         'info_dict': {
 340             'id': '4357',
 341         },
 342         'playlist_mincount': 30,
 343     }, {
 344         'url': 'https://www.hotstar.com/in/tv/bigg-boss/14714/seasons/season-4/ss-8208/',
 345         'info_dict': {
 346             'id': '8208',
 347         },
 348         'playlist_mincount': 19,
 349     }]
 350
 351     def _real_extract(self, url):
 352         url, season_id = self._match_valid_url(url).groups()
 353         return self.playlist_result(self._playlist_entries(
 354             'season/asset', season_id, url, query={'tao': 0, 'tas': 0, 'size': 10000, 'id': season_id}), season_id)
 355
 356
 357 class HotStarSeriesIE(HotStarBaseIE):
 358     IE_NAME = 'hotstar:series'
 359     _VALID_URL = r'(?P<url>https?://(?:www\.)?hotstar\.com(?:/in)?/tv/[^/]+/(?P<id>\d+))/?(?:[#?]|$)'
 360     _TESTS = [{
 361         'url': 'https://www.hotstar.com/in/tv/radhakrishn/1260000646',
 362         'info_dict': {
 363             'id': '1260000646',
 364         },
 365         'playlist_mincount': 690,
 366     }, {
 367         'url': 'https://www.hotstar.com/tv/dancee-/1260050431',
 368         'info_dict': {
 369             'id': '1260050431',
 370         },
 371         'playlist_mincount': 43,
 372     }, {
 373         'url': 'https://www.hotstar.com/in/tv/mahabharat/435/',
 374         'info_dict': {
 375             'id': '435',
 376         },
 377         'playlist_mincount': 267,
 378     }]
 379
 380     def _real_extract(self, url):
 381         url, series_id = self._match_valid_url(url).groups()
 382         id_ = self._call_api_v1(
 383             'show/detail', series_id, query={'contentId': series_id})['body']['results']['item']['id']
 384
 385         return self.playlist_result(self._playlist_entries(
 386             'tray/g/1/items', series_id, url, query={'tao': 0, 'tas': 10000, 'etid': 0, 'eid': id_}), series_id)