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
   8 from .common import InfoExtractor
   9 from ..compat import compat_HTTPError
  10 from ..utils import (
  11     determine_ext,
  12     ExtractorError,
  13     int_or_none,
  14 )
  15
  16
  17 class HotStarBaseIE(InfoExtractor):
  18     _AKAMAI_ENCRYPTION_KEY = b'\x05\xfc\x1a\x01\xca\xc9\x4b\xc4\x12\xfc\x53\x12\x07\x75\xf9\xee'
  19
  20     def _call_api(self, path, video_id, query_name='contentId'):
  21         st = int(time.time())
  22         exp = st + 6000
  23         auth = 'st=%d~exp=%d~acl=/*' % (st, exp)
  24         auth += '~hmac=' + hmac.new(self._AKAMAI_ENCRYPTION_KEY, auth.encode(), hashlib.sha256).hexdigest()
  25         response = self._download_json(
  26             'https://api.hotstar.com/' + path,
  27             video_id, headers={
  28                 'hotstarauth': auth,
  29                 'x-country-code': 'IN',
  30                 'x-platform-code': 'JIO',
  31             }, query={
  32                 query_name: video_id,
  33                 'tas': 10000,
  34             })
  35         if response['statusCode'] != 'OK':
  36             raise ExtractorError(
  37                 response['body']['message'], expected=True)
  38         return response['body']['results']
  39
  40
  41 class HotStarIE(HotStarBaseIE):
  42     IE_NAME = 'hotstar'
  43     _VALID_URL = r'https?://(?:www\.)?hotstar\.com/(?:.+?[/-])?(?P<id>\d{10})'
  44     _TESTS = [{
  45         'url': 'https://www.hotstar.com/can-you-not-spread-rumours/1000076273',
  46         'info_dict': {
  47             'id': '1000076273',
  48             'ext': 'mp4',
  49             'title': 'Can You Not Spread Rumours?',
  50             'description': 'md5:c957d8868e9bc793ccb813691cc4c434',
  51             'timestamp': 1447248600,
  52             'upload_date': '20151111',
  53             'duration': 381,
  54         },
  55         'params': {
  56             # m3u8 download
  57             'skip_download': True,
  58         }
  59     }, {
  60         'url': 'http://www.hotstar.com/sports/cricket/rajitha-sizzles-on-debut-with-329/2001477583',
  61         'only_matching': True,
  62     }, {
  63         'url': 'http://www.hotstar.com/1000000515',
  64         'only_matching': True,
  65     }]
  66     _GEO_BYPASS = False
  67
  68     def _real_extract(self, url):
  69         video_id = self._match_id(url)
  70
  71         webpage = self._download_webpage(url, video_id)
  72         app_state = self._parse_json(self._search_regex(
  73             r'<script>window\.APP_STATE\s*=\s*({.+?})</script>',
  74             webpage, 'app state'), video_id)
  75         video_data = list(app_state.values())[0]['initialState']['contentData']['content']
  76
  77         title = video_data['title']
  78
  79         if video_data.get('drmProtected'):
  80             raise ExtractorError('This video is DRM protected.', expected=True)
  81
  82         formats = []
  83         format_data = self._call_api('h/v1/play', video_id)['item']
  84         format_url = format_data['playbackUrl']
  85         ext = determine_ext(format_url)
  86         if ext == 'm3u8':
  87             try:
  88                 formats.extend(self._extract_m3u8_formats(
  89                     format_url, video_id, 'mp4', m3u8_id='hls'))
  90             except ExtractorError as e:
  91                 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
  92                     self.raise_geo_restricted(countries=['IN'])
  93                 raise
  94         elif ext == 'f4m':
  95             # produce broken files
  96             pass
  97         else:
  98             formats.append({
  99                 'url': format_url,
 100                 'width': int_or_none(format_data.get('width')),
 101                 'height': int_or_none(format_data.get('height')),
 102             })
 103         self._sort_formats(formats)
 104
 105         return {
 106             'id': video_id,
 107             'title': title,
 108             'description': video_data.get('description'),
 109             'duration': int_or_none(video_data.get('duration')),
 110             'timestamp': int_or_none(video_data.get('broadcastDate') or video_data.get('startDate')),
 111             'formats': formats,
 112             'channel': video_data.get('channelName'),
 113             'channel_id': video_data.get('channelId'),
 114             'series': video_data.get('showName'),
 115             'season': video_data.get('seasonName'),
 116             'season_number': int_or_none(video_data.get('seasonNo')),
 117             'season_id': video_data.get('seasonId'),
 118             'episode': title,
 119             'episode_number': int_or_none(video_data.get('episodeNo')),
 120         }
 121
 122
 123 class HotStarPlaylistIE(HotStarBaseIE):
 124     IE_NAME = 'hotstar:playlist'
 125     _VALID_URL = r'https?://(?:www\.)?hotstar\.com/tv/[^/]+/s-\w+/list/[^/]+/t-(?P<id>\w+)'
 126     _TESTS = [{
 127         'url': 'https://www.hotstar.com/tv/savdhaan-india/s-26/list/popular-clips/t-3_2_26',
 128         'info_dict': {
 129             'id': '3_2_26',
 130         },
 131         'playlist_mincount': 20,
 132     }, {
 133         'url': 'https://www.hotstar.com/tv/savdhaan-india/s-26/list/extras/t-2480',
 134         'only_matching': True,
 135     }]
 136
 137     def _real_extract(self, url):
 138         playlist_id = self._match_id(url)
 139
 140         collection = self._call_api('o/v1/tray/find', playlist_id, 'uqId')
 141
 142         entries = [
 143             self.url_result(
 144                 'https://www.hotstar.com/%s' % video['contentId'],
 145                 ie=HotStarIE.ie_key(), video_id=video['contentId'])
 146             for video in collection['assets']['items']
 147             if video.get('contentId')]
 148
 149         return self.playlist_result(entries, playlist_id)