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