youtube_dl/extractor/cbs.py

   1 from __future__ import unicode_literals
   2
   3 from .theplatform import ThePlatformFeedIE
   4 from ..utils import (
   5     int_or_none,
   6     find_xpath_attr,
   7     xpath_element,
   8     xpath_text,
   9     update_url_query,
  10 )
  11
  12
  13 class CBSBaseIE(ThePlatformFeedIE):
  14     def _parse_smil_subtitles(self, smil, namespace=None, subtitles_lang='en'):
  15         closed_caption_e = find_xpath_attr(smil, self._xpath_ns('.//param', namespace), 'name', 'ClosedCaptionURL')
  16         return {
  17             'en': [{
  18                 'ext': 'ttml',
  19                 'url': closed_caption_e.attrib['value'],
  20             }]
  21         } if closed_caption_e is not None and closed_caption_e.attrib.get('value') else []
  22
  23
  24 class CBSIE(CBSBaseIE):
  25     _VALID_URL = r'(?:cbs:|https?://(?:www\.)?(?:cbs\.com/shows/[^/]+/video|colbertlateshow\.com/(?:video|podcasts))/)(?P<id>[\w-]+)'
  26
  27     _TESTS = [{
  28         'url': 'http://www.cbs.com/shows/garth-brooks/video/_u7W953k6la293J7EPTd9oHkSPs6Xn6_/connect-chat-feat-garth-brooks/',
  29         'info_dict': {
  30             'id': '_u7W953k6la293J7EPTd9oHkSPs6Xn6_',
  31             'ext': 'mp4',
  32             'title': 'Connect Chat feat. Garth Brooks',
  33             'description': 'Connect with country music singer Garth Brooks, as he chats with fans on Wednesday November 27, 2013. Be sure to tune in to Garth Brooks: Live from Las Vegas, Friday November 29, at 9/8c on CBS!',
  34             'duration': 1495,
  35             'timestamp': 1385585425,
  36             'upload_date': '20131127',
  37             'uploader': 'CBSI-NEW',
  38         },
  39         'params': {
  40             # m3u8 download
  41             'skip_download': True,
  42         },
  43         '_skip': 'Blocked outside the US',
  44     }, {
  45         'url': 'http://colbertlateshow.com/video/8GmB0oY0McANFvp2aEffk9jZZZ2YyXxy/the-colbeard/',
  46         'only_matching': True,
  47     }, {
  48         'url': 'http://www.colbertlateshow.com/podcasts/dYSwjqPs_X1tvbV_P2FcPWRa_qT6akTC/in-the-bad-room-with-stephen/',
  49         'only_matching': True,
  50     }]
  51
  52     def _extract_video_info(self, content_id):
  53         items_data = self._download_xml(
  54             'http://can.cbs.com/thunder/player/videoPlayerService.php',
  55             content_id, query={'partner': 'cbs', 'contentId': content_id})
  56         video_data = xpath_element(items_data, './/item')
  57         title = xpath_text(video_data, 'videoTitle', 'title', True)
  58         tp_path = 'dJ5BDC/media/guid/2198311517/%s' % content_id
  59         tp_release_url = 'http://link.theplatform.com/s/' + tp_path
  60
  61         asset_types = []
  62         subtitles = {}
  63         formats = []
  64         for item in items_data.findall('.//item'):
  65             asset_type = xpath_text(item, 'assetType')
  66             if not asset_type or asset_type in asset_types:
  67                 continue
  68             asset_types.append(asset_type)
  69             query = {
  70                 'mbr': 'true',
  71                 'assetTypes': asset_type,
  72             }
  73             if asset_type.startswith('HLS') or asset_type in ('OnceURL', 'StreamPack'):
  74                 query['formats'] = 'MPEG4,M3U'
  75             elif asset_type in ('RTMP', 'WIFI', '3G'):
  76                 query['formats'] = 'MPEG4,FLV'
  77             tp_formats, tp_subtitles = self._extract_theplatform_smil(
  78                 update_url_query(tp_release_url, query), content_id,
  79                 'Downloading %s SMIL data' % asset_type)
  80             formats.extend(tp_formats)
  81             subtitles = self._merge_subtitles(subtitles, tp_subtitles)
  82         self._sort_formats(formats)
  83
  84         info = self._extract_theplatform_metadata(tp_path, content_id)
  85         info.update({
  86             'id': content_id,
  87             'title': title,
  88             'series': xpath_text(video_data, 'seriesTitle'),
  89             'season_number': int_or_none(xpath_text(video_data, 'seasonNumber')),
  90             'episode_number': int_or_none(xpath_text(video_data, 'episodeNumber')),
  91             'duration': int_or_none(xpath_text(video_data, 'videoLength'), 1000),
  92             'thumbnail': xpath_text(video_data, 'previewImageURL'),
  93             'formats': formats,
  94             'subtitles': subtitles,
  95         })
  96         return info
  97
  98     def _real_extract(self, url):
  99         content_id = self._match_id(url)
 100         return self._extract_video_info(content_id)