_ Git - youtube-dl/blob - youtube_dl/extractor/hotstar.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     ExtractorError,
   7     determine_ext,
   8     int_or_none,
   9 )
  10
  11
  12 class HotStarIE(InfoExtractor):
  13     _VALID_URL = r'https?://(?:www\.)?hotstar\.com/(?:.+?[/-])?(?P<id>\d{10})'
  14     _GEO_COUNTRIES = ['IN']
  15     _TESTS = [{
  16         'url': 'http://www.hotstar.com/on-air-with-aib--english-1000076273',
  17         'info_dict': {
  18             'id': '1000076273',
  19             'ext': 'mp4',
  20             'title': 'On Air With AIB - English',
  21             'description': 'md5:c957d8868e9bc793ccb813691cc4c434',
  22             'timestamp': 1447227000,
  23             'upload_date': '20151111',
  24             'duration': 381,
  25         },
  26         'params': {
  27             # m3u8 download
  28             'skip_download': True,
  29         }
  30     }, {
  31         'url': 'http://www.hotstar.com/sports/cricket/rajitha-sizzles-on-debut-with-329/2001477583',
  32         'only_matching': True,
  33     }, {
  34         'url': 'http://www.hotstar.com/1000000515',
  35         'only_matching': True,
  36     }]
  37
  38     def _download_json(self, url_or_request, video_id, note='Downloading JSON metadata', fatal=True, query=None):
  39         json_data = super(HotStarIE, self)._download_json(
  40             url_or_request, video_id, note, fatal=fatal, query=query)
  41         if json_data['resultCode'] != 'OK':
  42             if fatal:
  43                 raise ExtractorError(json_data['errorDescription'])
  44             return None
  45         return json_data['resultObj']
  46
  47     def _real_extract(self, url):
  48         video_id = self._match_id(url)
  49         video_data = self._download_json(
  50             'http://account.hotstar.com/AVS/besc', video_id, query={
  51                 'action': 'GetAggregatedContentDetails',
  52                 'channel': 'PCTV',
  53                 'contentId': video_id,
  54             })['contentInfo'][0]
  55         title = video_data['episodeTitle']
  56
  57         if video_data.get('encrypted') == 'Y':
  58             raise ExtractorError('This video is DRM protected.', expected=True)
  59
  60         formats = []
  61         for f in ('JIO',):
  62             format_data = self._download_json(
  63                 'http://getcdn.hotstar.com/AVS/besc',
  64                 video_id, 'Downloading %s JSON metadata' % f,
  65                 fatal=False, query={
  66                     'action': 'GetCDN',
  67                     'asJson': 'Y',
  68                     'channel': f,
  69                     'id': video_id,
  70                     'type': 'VOD',
  71                 })
  72             if format_data:
  73                 format_url = format_data.get('src')
  74                 if not format_url:
  75                     continue
  76                 ext = determine_ext(format_url)
  77                 if ext == 'm3u8':
  78                     formats.extend(self._extract_m3u8_formats(
  79                         format_url, video_id, 'mp4',
  80                         m3u8_id='hls', fatal=False))
  81                 elif ext == 'f4m':
  82                     # produce broken files
  83                     continue
  84                 else:
  85                     formats.append({
  86                         'url': format_url,
  87                         'width': int_or_none(format_data.get('width')),
  88                         'height': int_or_none(format_data.get('height')),
  89                     })
  90         self._sort_formats(formats)
  91
  92         return {
  93             'id': video_id,
  94             'title': title,
  95             'description': video_data.get('description'),
  96             'duration': int_or_none(video_data.get('duration')),
  97             'timestamp': int_or_none(video_data.get('broadcastDate')),
  98             'formats': formats,
  99             'episode': title,
 100             'episode_number': int_or_none(video_data.get('episodeNumber')),
 101             'series': video_data.get('contentTitle'),
 102         }