Raphaël G. Git Repositories - youtubedl/blob - 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 import uuid
   8
   9 from .common import InfoExtractor
  10 from ..compat import (
  11     compat_HTTPError,
  12     compat_str,
  13 )
  14 from ..utils import (
  15     determine_ext,
  16     ExtractorError,
  17     int_or_none,
  18     str_or_none,
  19     try_get,
  20     url_or_none,
  21 )
  22
  23
  24 class HotStarBaseIE(InfoExtractor):
  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_impl(self, path, video_id, query):
  28         st = int(time.time())
  29         exp = st + 6000
  30         auth = 'st=%d~exp=%d~acl=/*' % (st, exp)
  31         auth += '~hmac=' + hmac.new(self._AKAMAI_ENCRYPTION_KEY, auth.encode(), hashlib.sha256).hexdigest()
  32         response = self._download_json(
  33             'https://api.hotstar.com/' + path, video_id, headers={
  34                 'hotstarauth': auth,
  35                 'x-country-code': 'IN',
  36                 'x-platform-code': 'JIO',
  37             }, query=query)
  38         if response['statusCode'] != 'OK':
  39             raise ExtractorError(
  40                 response['body']['message'], expected=True)
  41         return response['body']['results']
  42
  43     def _call_api(self, path, video_id, query_name='contentId'):
  44         return self._call_api_impl(path, video_id, {
  45             query_name: video_id,
  46             'tas': 10000,
  47         })
  48
  49     def _call_api_v2(self, path, video_id):
  50         return self._call_api_impl(
  51             '%s/in/contents/%s' % (path, video_id), video_id, {
  52                 'desiredConfig': 'encryption:plain;ladder:phone,tv;package:hls,dash',
  53                 'client': 'mweb',
  54                 'clientVersion': '6.18.0',
  55                 'deviceId': compat_str(uuid.uuid4()),
  56                 'osName': 'Windows',
  57                 'osVersion': '10',
  58             })
  59
  60
  61 class HotStarIE(HotStarBaseIE):
  62     IE_NAME = 'hotstar'
  63     _VALID_URL = r'https?://(?:www\.)?hotstar\.com/(?:.+?[/-])?(?P<id>\d{10})'
  64     _TESTS = [{
  65         # contentData
  66         'url': 'https://www.hotstar.com/can-you-not-spread-rumours/1000076273',
  67         'info_dict': {
  68             'id': '1000076273',
  69             'ext': 'mp4',
  70             'title': 'Can You Not Spread Rumours?',
  71             'description': 'md5:c957d8868e9bc793ccb813691cc4c434',
  72             'timestamp': 1447248600,
  73             'upload_date': '20151111',
  74             'duration': 381,
  75         },
  76         'params': {
  77             # m3u8 download
  78             'skip_download': True,
  79         }
  80     }, {
  81         # contentDetail
  82         'url': 'https://www.hotstar.com/movies/radha-gopalam/1000057157',
  83         'only_matching': True,
  84     }, {
  85         'url': 'http://www.hotstar.com/sports/cricket/rajitha-sizzles-on-debut-with-329/2001477583',
  86         'only_matching': True,
  87     }, {
  88         'url': 'http://www.hotstar.com/1000000515',
  89         'only_matching': True,
  90     }, {
  91         # only available via api v2
  92         'url': 'https://www.hotstar.com/tv/ek-bhram-sarvagun-sampanna/s-2116/janhvi-targets-suman/1000234847',
  93         'only_matching': True,
  94     }]
  95     _GEO_BYPASS = False
  96
  97     def _real_extract(self, url):
  98         video_id = self._match_id(url)
  99
 100         webpage = self._download_webpage(url, video_id)
 101         app_state = self._parse_json(self._search_regex(
 102             r'<script>window\.APP_STATE\s*=\s*({.+?})</script>',
 103             webpage, 'app state'), video_id)
 104         video_data = {}
 105         getters = list(
 106             lambda x, k=k: x['initialState']['content%s' % k]['content']
 107             for k in ('Data', 'Detail')
 108         )
 109         for v in app_state.values():
 110             content = try_get(v, getters, dict)
 111             if content and content.get('contentId') == video_id:
 112                 video_data = content
 113                 break
 114
 115         title = video_data['title']
 116
 117         if video_data.get('drmProtected'):
 118             raise ExtractorError('This video is DRM protected.', expected=True)
 119
 120         formats = []
 121         geo_restricted = False
 122         playback_sets = self._call_api_v2('h/v2/play', video_id)['playBackSets']
 123         for playback_set in playback_sets:
 124             if not isinstance(playback_set, dict):
 125                 continue
 126             format_url = url_or_none(playback_set.get('playbackUrl'))
 127             if not format_url:
 128                 continue
 129             tags = str_or_none(playback_set.get('tagsCombination')) or ''
 130             if tags and 'encryption:plain' not in tags:
 131                 continue
 132             ext = determine_ext(format_url)
 133             try:
 134                 if 'package:hls' in tags or ext == 'm3u8':
 135                     formats.extend(self._extract_m3u8_formats(
 136                         format_url, video_id, 'mp4', m3u8_id='hls'))
 137                 elif 'package:dash' in tags or ext == 'mpd':
 138                     formats.extend(self._extract_mpd_formats(
 139                         format_url, video_id, mpd_id='dash'))
 140                 elif ext == 'f4m':
 141                     # produce broken files
 142                     pass
 143                 else:
 144                     formats.append({
 145                         'url': format_url,
 146                         'width': int_or_none(playback_set.get('width')),
 147                         'height': int_or_none(playback_set.get('height')),
 148                     })
 149             except ExtractorError as e:
 150                 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
 151                     geo_restricted = True
 152                 continue
 153         if not formats and geo_restricted:
 154             self.raise_geo_restricted(countries=['IN'])
 155         self._sort_formats(formats)
 156
 157         return {
 158             'id': video_id,
 159             'title': title,
 160             'description': video_data.get('description'),
 161             'duration': int_or_none(video_data.get('duration')),
 162             'timestamp': int_or_none(video_data.get('broadcastDate') or video_data.get('startDate')),
 163             'formats': formats,
 164             'channel': video_data.get('channelName'),
 165             'channel_id': video_data.get('channelId'),
 166             'series': video_data.get('showName'),
 167             'season': video_data.get('seasonName'),
 168             'season_number': int_or_none(video_data.get('seasonNo')),
 169             'season_id': video_data.get('seasonId'),
 170             'episode': title,
 171             'episode_number': int_or_none(video_data.get('episodeNo')),
 172         }
 173
 174
 175 class HotStarPlaylistIE(HotStarBaseIE):
 176     IE_NAME = 'hotstar:playlist'
 177     _VALID_URL = r'https?://(?:www\.)?hotstar\.com/tv/[^/]+/s-\w+/list/[^/]+/t-(?P<id>\w+)'
 178     _TESTS = [{
 179         'url': 'https://www.hotstar.com/tv/savdhaan-india/s-26/list/popular-clips/t-3_2_26',
 180         'info_dict': {
 181             'id': '3_2_26',
 182         },
 183         'playlist_mincount': 20,
 184     }, {
 185         'url': 'https://www.hotstar.com/tv/savdhaan-india/s-26/list/extras/t-2480',
 186         'only_matching': True,
 187     }]
 188
 189     def _real_extract(self, url):
 190         playlist_id = self._match_id(url)
 191
 192         collection = self._call_api('o/v1/tray/find', playlist_id, 'uqId')
 193
 194         entries = [
 195             self.url_result(
 196                 'https://www.hotstar.com/%s' % video['contentId'],
 197                 ie=HotStarIE.ie_key(), video_id=video['contentId'])
 198             for video in collection['assets']['items']
 199             if video.get('contentId')]
 200
 201         return self.playlist_result(entries, playlist_id)