youtube_dlc/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):
  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         token = self._download_json(
  35             'https://api.hotstar.com/in/aadhar/v2/web/in/user/guest-signup',
  36             video_id, note='Downloading token',
  37             data=json.dumps({"idType": "device", "id": compat_str(uuid.uuid4())}).encode('utf-8'),
  38             headers={
  39                 'hotstarauth': auth,
  40                 'Content-Type': 'application/json',
  41             })['description']['userIdentity']
  42         response = self._download_json(
  43             'https://api.hotstar.com/' + path, video_id, headers={
  44                 'hotstarauth': auth,
  45                 'x-hs-appversion': '6.72.2',
  46                 'x-hs-platform': 'web',
  47                 'x-hs-usertoken': token,
  48             }, query=query)
  49         if response['message'] != "Playback URL's fetched successfully":
  50             raise ExtractorError(
  51                 response['message'], expected=True)
  52         return response['data']
  53
  54     def _call_api(self, path, video_id, query_name='contentId'):
  55         return self._call_api_impl(path, video_id, {
  56             query_name: video_id,
  57             'tas': 10000,
  58         })
  59
  60     def _call_api_v2(self, path, video_id):
  61         return self._call_api_impl(
  62             '%s/content/%s' % (path, video_id), video_id, {
  63                 'desired-config': 'encryption:plain;ladder:phone,tv;package:hls,dash',
  64                 'device-id': compat_str(uuid.uuid4()),
  65                 'os-name': 'Windows',
  66                 'os-version': '10',
  67             })
  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     _GEO_BYPASS = False
 105
 106     def _real_extract(self, url):
 107         video_id = self._match_id(url)
 108
 109         webpage = self._download_webpage(url, video_id)
 110         app_state = self._parse_json(self._search_regex(
 111             r'<script>window\.APP_STATE\s*=\s*({.+?})</script>',
 112             webpage, 'app state'), video_id)
 113         video_data = {}
 114         getters = list(
 115             lambda x, k=k: x['initialState']['content%s' % k]['content']
 116             for k in ('Data', 'Detail')
 117         )
 118         for v in app_state.values():
 119             content = try_get(v, getters, dict)
 120             if content and content.get('contentId') == video_id:
 121                 video_data = content
 122                 break
 123
 124         title = video_data['title']
 125
 126         if video_data.get('drmProtected'):
 127             raise ExtractorError('This video is DRM protected.', expected=True)
 128
 129         headers = {'Referer': url}
 130         formats = []
 131         geo_restricted = False
 132         playback_sets = self._call_api_v2('play/v1/playback', video_id)['playBackSets']
 133         for playback_set in playback_sets:
 134             if not isinstance(playback_set, dict):
 135                 continue
 136             format_url = url_or_none(playback_set.get('playbackUrl'))
 137             if not format_url:
 138                 continue
 139             format_url = re.sub(
 140                 r'(?<=//staragvod)(\d)', r'web\1', format_url)
 141             tags = str_or_none(playback_set.get('tagsCombination')) or ''
 142             if tags and 'encryption:plain' not in tags:
 143                 continue
 144             ext = determine_ext(format_url)
 145             try:
 146                 if 'package:hls' in tags or ext == 'm3u8':
 147                     formats.extend(self._extract_m3u8_formats(
 148                         format_url, video_id, 'mp4',
 149                         entry_protocol='m3u8_native',
 150                         m3u8_id='hls', headers=headers))
 151                 elif 'package:dash' in tags or ext == 'mpd':
 152                     formats.extend(self._extract_mpd_formats(
 153                         format_url, video_id, mpd_id='dash', headers=headers))
 154                 elif ext == 'f4m':
 155                     # produce broken files
 156                     pass
 157                 else:
 158                     formats.append({
 159                         'url': format_url,
 160                         'width': int_or_none(playback_set.get('width')),
 161                         'height': int_or_none(playback_set.get('height')),
 162                     })
 163             except ExtractorError as e:
 164                 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
 165                     geo_restricted = True
 166                 continue
 167         if not formats and geo_restricted:
 168             self.raise_geo_restricted(countries=['IN'])
 169         self._sort_formats(formats)
 170
 171         for f in formats:
 172             f.setdefault('http_headers', {}).update(headers)
 173
 174         return {
 175             'id': video_id,
 176             'title': title,
 177             'description': video_data.get('description'),
 178             'duration': int_or_none(video_data.get('duration')),
 179             'timestamp': int_or_none(video_data.get('broadcastDate') or video_data.get('startDate')),
 180             'formats': formats,
 181             'channel': video_data.get('channelName'),
 182             'channel_id': video_data.get('channelId'),
 183             'series': video_data.get('showName'),
 184             'season': video_data.get('seasonName'),
 185             'season_number': int_or_none(video_data.get('seasonNo')),
 186             'season_id': video_data.get('seasonId'),
 187             'episode': title,
 188             'episode_number': int_or_none(video_data.get('episodeNo')),
 189         }
 190
 191
 192 class HotStarPlaylistIE(HotStarBaseIE):
 193     IE_NAME = 'hotstar:playlist'
 194     _VALID_URL = r'https?://(?:www\.)?hotstar\.com/tv/[^/]+/s-\w+/list/[^/]+/t-(?P<id>\w+)'
 195     _TESTS = [{
 196         'url': 'https://www.hotstar.com/tv/savdhaan-india/s-26/list/popular-clips/t-3_2_26',
 197         'info_dict': {
 198             'id': '3_2_26',
 199         },
 200         'playlist_mincount': 20,
 201     }, {
 202         'url': 'https://www.hotstar.com/tv/savdhaan-india/s-26/list/extras/t-2480',
 203         'only_matching': True,
 204     }]
 205
 206     def _real_extract(self, url):
 207         playlist_id = self._match_id(url)
 208
 209         collection = self._call_api('o/v1/tray/find', playlist_id, 'uqId')
 210
 211         entries = [
 212             self.url_result(
 213                 'https://www.hotstar.com/%s' % video['contentId'],
 214                 ie=HotStarIE.ie_key(), video_id=video['contentId'])
 215             for video in collection['assets']['items']
 216             if video.get('contentId')]
 217
 218         return self.playlist_result(entries, playlist_id)