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_str
  10 from ..networking.exceptions import HTTPError
  11 from ..utils import (
  12     ExtractorError,
  13     determine_ext,
  14     int_or_none,
  15     join_nonempty,
  16     str_or_none,
  17     traverse_obj,
  18     url_or_none,
  19 )
  20
  21
  22 class HotStarBaseIE(InfoExtractor):
  23     _BASE_URL = 'https://www.hotstar.com'
  24     _API_URL = 'https://api.hotstar.com'
  25     _AKAMAI_ENCRYPTION_KEY = b'\x05\xfc\x1a\x01\xca\xc9\x4b\xc4\x12\xfc\x53\x12\x07\x75\xf9\xee'
  26
  27     def _call_api_v1(self, path, *args, **kwargs):
  28         return self._download_json(
  29             f'{self._API_URL}/o/v1/{path}', *args, **kwargs,
  30             headers={'x-country-code': 'IN', 'x-platform-code': 'PCTV'})
  31
  32     def _call_api_impl(self, path, video_id, query, st=None, cookies=None):
  33         st = int_or_none(st) or int(time.time())
  34         exp = st + 6000
  35         auth = 'st=%d~exp=%d~acl=/*' % (st, exp)
  36         auth += '~hmac=' + hmac.new(self._AKAMAI_ENCRYPTION_KEY, auth.encode(), hashlib.sha256).hexdigest()
  37
  38         if cookies and cookies.get('userUP'):
  39             token = cookies.get('userUP').value
  40         else:
  41             token = self._download_json(
  42                 f'{self._API_URL}/um/v3/users',
  43                 video_id, note='Downloading token',
  44                 data=json.dumps({"device_ids": [{"id": compat_str(uuid.uuid4()), "type": "device_id"}]}).encode('utf-8'),
  45                 headers={
  46                     'hotstarauth': auth,
  47                     'x-hs-platform': 'PCTV',  # or 'web'
  48                     'Content-Type': 'application/json',
  49                 })['user_identity']
  50
  51         response = self._download_json(
  52             f'{self._API_URL}/{path}', video_id, query=query,
  53             headers={
  54                 'hotstarauth': auth,
  55                 'x-hs-appversion': '6.72.2',
  56                 'x-hs-platform': 'web',
  57                 'x-hs-usertoken': token,
  58             })
  59
  60         if response['message'] != "Playback URL's fetched successfully":
  61             raise ExtractorError(
  62                 response['message'], expected=True)
  63         return response['data']
  64
  65     def _call_api_v2(self, path, video_id, st=None, cookies=None):
  66         return self._call_api_impl(
  67             f'{path}/content/{video_id}', video_id, st=st, cookies=cookies, query={
  68                 'desired-config': 'audio_channel:stereo|container:fmp4|dynamic_range:hdr|encryption:plain|ladder:tv|package:dash|resolution:fhd|subs-tag:HotstarVIP|video_codec:h265',
  69                 'device-id': cookies.get('device_id').value if cookies.get('device_id') else compat_str(uuid.uuid4()),
  70                 'os-name': 'Windows',
  71                 'os-version': '10',
  72             })
  73
  74     def _playlist_entries(self, path, item_id, root=None, **kwargs):
  75         results = self._call_api_v1(path, item_id, **kwargs)['body']['results']
  76         for video in traverse_obj(results, (('assets', None), 'items', ...)):
  77             if video.get('contentId'):
  78                 yield self.url_result(
  79                     HotStarIE._video_url(video['contentId'], root=root), HotStarIE, video['contentId'])
  80
  81
  82 class HotStarIE(HotStarBaseIE):
  83     IE_NAME = 'hotstar'
  84     _VALID_URL = r'''(?x)
  85         https?://(?:www\.)?hotstar\.com(?:/in)?/(?!in/)
  86         (?:
  87             (?P<type>movies|sports|clips|episode|(?P<tv>tv|shows))/
  88             (?(tv)(?:[^/?#]+/){2}|[^?#]*)
  89         )?
  90         [^/?#]+/
  91         (?P<id>\d{10})
  92     '''
  93
  94     _TESTS = [{
  95         'url': 'https://www.hotstar.com/can-you-not-spread-rumours/1000076273',
  96         'info_dict': {
  97             'id': '1000076273',
  98             'ext': 'mp4',
  99             'title': 'Can You Not Spread Rumours?',
 100             'description': 'md5:c957d8868e9bc793ccb813691cc4c434',
 101             'timestamp': 1447248600,
 102             'upload_date': '20151111',
 103             'duration': 381,
 104             'episode': 'Can You Not Spread Rumours?',
 105         },
 106         'params': {'skip_download': 'm3u8'},
 107     }, {
 108         'url': 'https://www.hotstar.com/tv/ek-bhram-sarvagun-sampanna/s-2116/janhvi-targets-suman/1000234847',
 109         'info_dict': {
 110             'id': '1000234847',
 111             'ext': 'mp4',
 112             'title': 'Janhvi Targets Suman',
 113             'description': 'md5:78a85509348910bd1ca31be898c5796b',
 114             'timestamp': 1556670600,
 115             'upload_date': '20190501',
 116             'duration': 1219,
 117             'channel': 'StarPlus',
 118             'channel_id': '3',
 119             'series': 'Ek Bhram - Sarvagun Sampanna',
 120             'season': 'Chapter 1',
 121             'season_number': 1,
 122             'season_id': '6771',
 123             'episode': 'Janhvi Targets Suman',
 124             'episode_number': 8,
 125         }
 126     }, {
 127         'url': 'https://www.hotstar.com/in/shows/anupama/1260022017/anupama-anuj-share-a-moment/1000282843',
 128         'info_dict': {
 129             'id': '1000282843',
 130             'ext': 'mp4',
 131             'title': 'Anupama, Anuj Share a Moment',
 132             'season': 'Chapter 1',
 133             'description': 'md5:8d74ed2248423b8b06d5c8add4d7a0c0',
 134             'timestamp': 1678149000,
 135             'channel': 'StarPlus',
 136             'series': 'Anupama',
 137             'season_number': 1,
 138             'season_id': '7399',
 139             'upload_date': '20230307',
 140             'episode': 'Anupama, Anuj Share a Moment',
 141             'episode_number': 853,
 142             'duration': 1272,
 143             'channel_id': '3',
 144         },
 145         'skip': 'HTTP Error 504: Gateway Time-out',  # XXX: Investigate 504 errors on some episodes
 146     }, {
 147         'url': 'https://www.hotstar.com/in/shows/kana-kaanum-kaalangal/1260097087/back-to-school/1260097320',
 148         'info_dict': {
 149             'id': '1260097320',
 150             'ext': 'mp4',
 151             'title': 'Back To School',
 152             'season': 'Chapter 1',
 153             'description': 'md5:b0d6a4c8a650681491e7405496fc7e13',
 154             'timestamp': 1650564000,
 155             'channel': 'Hotstar Specials',
 156             'series': 'Kana Kaanum Kaalangal',
 157             'season_number': 1,
 158             'season_id': '9441',
 159             'upload_date': '20220421',
 160             'episode': 'Back To School',
 161             'episode_number': 1,
 162             'duration': 1810,
 163             'channel_id': '54',
 164         },
 165     }, {
 166         'url': 'https://www.hotstar.com/in/clips/e3-sairat-kahani-pyaar-ki/1000262286',
 167         'info_dict': {
 168             'id': '1000262286',
 169             'ext': 'mp4',
 170             'title': 'E3 - SaiRat, Kahani Pyaar Ki',
 171             'description': 'md5:e3b4b3203bc0c5396fe7d0e4948a6385',
 172             'episode': 'E3 - SaiRat, Kahani Pyaar Ki',
 173             'upload_date': '20210606',
 174             'timestamp': 1622943900,
 175             'duration': 5395,
 176         },
 177     }, {
 178         'url': 'https://www.hotstar.com/in/movies/premam/1000091195',
 179         'info_dict': {
 180             'id': '1000091195',
 181             'ext': 'mp4',
 182             'title': 'Premam',
 183             'release_year': 2015,
 184             'description': 'md5:d833c654e4187b5e34757eafb5b72d7f',
 185             'timestamp': 1462149000,
 186             'upload_date': '20160502',
 187             'episode': 'Premam',
 188             'duration': 8994,
 189         },
 190     }, {
 191         'url': 'https://www.hotstar.com/movies/radha-gopalam/1000057157',
 192         'only_matching': True,
 193     }, {
 194         'url': 'https://www.hotstar.com/in/sports/cricket/follow-the-blues-2021/recap-eng-fight-back-on-day-2/1260066104',
 195         'only_matching': True,
 196     }, {
 197         'url': 'https://www.hotstar.com/in/sports/football/most-costly-pl-transfers-ft-grealish/1260065956',
 198         'only_matching': True,
 199     }]
 200     _GEO_BYPASS = False
 201
 202     _TYPE = {
 203         'movies': 'movie',
 204         'sports': 'match',
 205         'episode': 'episode',
 206         'tv': 'episode',
 207         'shows': 'episode',
 208         'clips': 'content',
 209         None: 'content',
 210     }
 211
 212     _IGNORE_MAP = {
 213         'res': 'resolution',
 214         'vcodec': 'video_codec',
 215         'dr': 'dynamic_range',
 216     }
 217
 218     _TAG_FIELDS = {
 219         'language': 'language',
 220         'acodec': 'audio_codec',
 221         'vcodec': 'video_codec',
 222     }
 223
 224     @classmethod
 225     def _video_url(cls, video_id, video_type=None, *, slug='ignore_me', root=None):
 226         assert None in (video_type, root)
 227         if not root:
 228             root = join_nonempty(cls._BASE_URL, video_type, delim='/')
 229         return f'{root}/{slug}/{video_id}'
 230
 231     def _real_extract(self, url):
 232         video_id, video_type = self._match_valid_url(url).group('id', 'type')
 233         video_type = self._TYPE.get(video_type, video_type)
 234         cookies = self._get_cookies(url)  # Cookies before any request
 235
 236         video_data = traverse_obj(
 237             self._call_api_v1(
 238                 f'{video_type}/detail', video_id, fatal=False, query={'tas': 10000, 'contentId': video_id}),
 239             ('body', 'results', 'item', {dict})) or {}
 240         if not self.get_param('allow_unplayable_formats') and video_data.get('drmProtected'):
 241             self.report_drm(video_id)
 242
 243         # See https://github.com/yt-dlp/yt-dlp/issues/396
 244         st = self._download_webpage_handle(f'{self._BASE_URL}/in', video_id)[1].headers.get('x-origin-date')
 245
 246         geo_restricted = False
 247         formats, subs = [], {}
 248         headers = {'Referer': f'{self._BASE_URL}/in'}
 249
 250         # change to v2 in the future
 251         playback_sets = self._call_api_v2('play/v1/playback', video_id, st=st, cookies=cookies)['playBackSets']
 252         for playback_set in playback_sets:
 253             if not isinstance(playback_set, dict):
 254                 continue
 255             tags = str_or_none(playback_set.get('tagsCombination')) or ''
 256             if any(f'{prefix}:{ignore}' in tags
 257                    for key, prefix in self._IGNORE_MAP.items()
 258                    for ignore in self._configuration_arg(key)):
 259                 continue
 260             tag_dict = dict((t.split(':', 1) + [None])[:2] for t in tags.split(';'))
 261
 262             format_url = url_or_none(playback_set.get('playbackUrl'))
 263             if not format_url:
 264                 continue
 265             format_url = re.sub(r'(?<=//staragvod)(\d)', r'web\1', format_url)
 266             ext = determine_ext(format_url)
 267
 268             current_formats, current_subs = [], {}
 269             try:
 270                 if 'package:hls' in tags or ext == 'm3u8':
 271                     current_formats, current_subs = self._extract_m3u8_formats_and_subtitles(
 272                         format_url, video_id, ext='mp4', headers=headers)
 273                 elif 'package:dash' in tags or ext == 'mpd':
 274                     current_formats, current_subs = self._extract_mpd_formats_and_subtitles(
 275                         format_url, video_id, headers=headers)
 276                 elif ext == 'f4m':
 277                     pass  # XXX: produce broken files
 278                 else:
 279                     current_formats = [{
 280                         'url': format_url,
 281                         'width': int_or_none(playback_set.get('width')),
 282                         'height': int_or_none(playback_set.get('height')),
 283                     }]
 284             except ExtractorError as e:
 285                 if isinstance(e.cause, HTTPError) and e.cause.status == 403:
 286                     geo_restricted = True
 287                 continue
 288
 289             if tag_dict.get('encryption') not in ('plain', None):
 290                 for f in current_formats:
 291                     f['has_drm'] = True
 292             for f in current_formats:
 293                 for k, v in self._TAG_FIELDS.items():
 294                     if not f.get(k):
 295                         f[k] = tag_dict.get(v)
 296                 if f.get('vcodec') != 'none' and not f.get('dynamic_range'):
 297                     f['dynamic_range'] = tag_dict.get('dynamic_range')
 298                 if f.get('acodec') != 'none' and not f.get('audio_channels'):
 299                     f['audio_channels'] = {
 300                         'stereo': 2,
 301                         'dolby51': 6,
 302                     }.get(tag_dict.get('audio_channel'))
 303                 f['format_note'] = join_nonempty(
 304                     tag_dict.get('ladder'),
 305                     tag_dict.get('audio_channel') if f.get('acodec') != 'none' else None,
 306                     f.get('format_note'),
 307                     delim=', ')
 308
 309             formats.extend(current_formats)
 310             subs = self._merge_subtitles(subs, current_subs)
 311
 312         if not formats and geo_restricted:
 313             self.raise_geo_restricted(countries=['IN'], metadata_available=True)
 314         self._remove_duplicate_formats(formats)
 315         for f in formats:
 316             f.setdefault('http_headers', {}).update(headers)
 317
 318         return {
 319             'id': video_id,
 320             'title': video_data.get('title'),
 321             'description': video_data.get('description'),
 322             'duration': int_or_none(video_data.get('duration')),
 323             'timestamp': int_or_none(traverse_obj(video_data, 'broadcastDate', 'startDate')),
 324             'release_year': int_or_none(video_data.get('year')),
 325             'formats': formats,
 326             'subtitles': subs,
 327             'channel': video_data.get('channelName'),
 328             'channel_id': str_or_none(video_data.get('channelId')),
 329             'series': video_data.get('showName'),
 330             'season': video_data.get('seasonName'),
 331             'season_number': int_or_none(video_data.get('seasonNo')),
 332             'season_id': str_or_none(video_data.get('seasonId')),
 333             'episode': video_data.get('title'),
 334             'episode_number': int_or_none(video_data.get('episodeNo')),
 335         }
 336
 337
 338 class HotStarPrefixIE(InfoExtractor):
 339     """ The "hotstar:" prefix is no longer in use, but this is kept for backward compatibility """
 340     IE_DESC = False
 341     _VALID_URL = r'hotstar:(?:(?P<type>\w+):)?(?P<id>\d+)$'
 342     _TESTS = [{
 343         'url': 'hotstar:1000076273',
 344         'only_matching': True,
 345     }, {
 346         'url': 'hotstar:movies:1260009879',
 347         'info_dict': {
 348             'id': '1260009879',
 349             'ext': 'mp4',
 350             'title': 'Nuvvu Naaku Nachav',
 351             'description': 'md5:d43701b1314e6f8233ce33523c043b7d',
 352             'timestamp': 1567525674,
 353             'upload_date': '20190903',
 354             'duration': 10787,
 355             'episode': 'Nuvvu Naaku Nachav',
 356         },
 357     }, {
 358         'url': 'hotstar:episode:1000234847',
 359         'only_matching': True,
 360     }, {
 361         # contentData
 362         'url': 'hotstar:sports:1260065956',
 363         'only_matching': True,
 364     }, {
 365         # contentData
 366         'url': 'hotstar:sports:1260066104',
 367         'only_matching': True,
 368     }]
 369
 370     def _real_extract(self, url):
 371         video_id, video_type = self._match_valid_url(url).group('id', 'type')
 372         return self.url_result(HotStarIE._video_url(video_id, video_type), HotStarIE, video_id)
 373
 374
 375 class HotStarPlaylistIE(HotStarBaseIE):
 376     IE_NAME = 'hotstar:playlist'
 377     _VALID_URL = r'https?://(?:www\.)?hotstar\.com(?:/in)?/(?:tv|shows)(?:/[^/]+){2}/list/[^/]+/t-(?P<id>\w+)'
 378     _TESTS = [{
 379         'url': 'https://www.hotstar.com/tv/savdhaan-india/s-26/list/popular-clips/t-3_2_26',
 380         'info_dict': {
 381             'id': '3_2_26',
 382         },
 383         'playlist_mincount': 20,
 384     }, {
 385         'url': 'https://www.hotstar.com/shows/savdhaan-india/s-26/list/popular-clips/t-3_2_26',
 386         'only_matching': True,
 387     }, {
 388         'url': 'https://www.hotstar.com/tv/savdhaan-india/s-26/list/extras/t-2480',
 389         'only_matching': True,
 390     }, {
 391         'url': 'https://www.hotstar.com/in/tv/karthika-deepam/15457/list/popular-clips/t-3_2_1272',
 392         'only_matching': True,
 393     }]
 394
 395     def _real_extract(self, url):
 396         id_ = self._match_id(url)
 397         return self.playlist_result(
 398             self._playlist_entries('tray/find', id_, query={'tas': 10000, 'uqId': id_}), id_)
 399
 400
 401 class HotStarSeasonIE(HotStarBaseIE):
 402     IE_NAME = 'hotstar:season'
 403     _VALID_URL = r'(?P<url>https?://(?:www\.)?hotstar\.com(?:/in)?/(?:tv|shows)/[^/]+/\w+)/seasons/[^/]+/ss-(?P<id>\w+)'
 404     _TESTS = [{
 405         'url': 'https://www.hotstar.com/tv/radhakrishn/1260000646/seasons/season-2/ss-8028',
 406         'info_dict': {
 407             'id': '8028',
 408         },
 409         'playlist_mincount': 35,
 410     }, {
 411         'url': 'https://www.hotstar.com/in/tv/ishqbaaz/9567/seasons/season-2/ss-4357',
 412         'info_dict': {
 413             'id': '4357',
 414         },
 415         'playlist_mincount': 30,
 416     }, {
 417         'url': 'https://www.hotstar.com/in/tv/bigg-boss/14714/seasons/season-4/ss-8208/',
 418         'info_dict': {
 419             'id': '8208',
 420         },
 421         'playlist_mincount': 19,
 422     }, {
 423         'url': 'https://www.hotstar.com/in/shows/bigg-boss/14714/seasons/season-4/ss-8208/',
 424         'only_matching': True,
 425     }]
 426
 427     def _real_extract(self, url):
 428         url, season_id = self._match_valid_url(url).groups()
 429         return self.playlist_result(self._playlist_entries(
 430             'season/asset', season_id, url, query={'tao': 0, 'tas': 0, 'size': 10000, 'id': season_id}), season_id)
 431
 432
 433 class HotStarSeriesIE(HotStarBaseIE):
 434     IE_NAME = 'hotstar:series'
 435     _VALID_URL = r'(?P<url>https?://(?:www\.)?hotstar\.com(?:/in)?/(?:tv|shows)/[^/]+/(?P<id>\d+))/?(?:[#?]|$)'
 436     _TESTS = [{
 437         'url': 'https://www.hotstar.com/in/tv/radhakrishn/1260000646',
 438         'info_dict': {
 439             'id': '1260000646',
 440         },
 441         'playlist_mincount': 690,
 442     }, {
 443         'url': 'https://www.hotstar.com/tv/dancee-/1260050431',
 444         'info_dict': {
 445             'id': '1260050431',
 446         },
 447         'playlist_mincount': 43,
 448     }, {
 449         'url': 'https://www.hotstar.com/in/tv/mahabharat/435/',
 450         'info_dict': {
 451             'id': '435',
 452         },
 453         'playlist_mincount': 267,
 454     }, {
 455         'url': 'https://www.hotstar.com/in/shows/anupama/1260022017/',
 456         'info_dict': {
 457             'id': '1260022017',
 458         },
 459         'playlist_mincount': 940,
 460     }]
 461
 462     def _real_extract(self, url):
 463         url, series_id = self._match_valid_url(url).groups()
 464         id_ = self._call_api_v1(
 465             'show/detail', series_id, query={'contentId': series_id})['body']['results']['item']['id']
 466
 467         return self.playlist_result(self._playlist_entries(
 468             'tray/g/1/items', series_id, url, query={'tao': 0, 'tas': 10000, 'etid': 0, 'eid': id_}), series_id)