yt_dlp/extractor/cbs.py

   1 from __future__ import unicode_literals
   2
   3 from .theplatform import ThePlatformFeedIE
   4 from ..utils import (
   5     ExtractorError,
   6     int_or_none,
   7     find_xpath_attr,
   8     xpath_element,
   9     xpath_text,
  10     update_url_query,
  11     url_or_none,
  12 )
  13
  14
  15 class CBSBaseIE(ThePlatformFeedIE):
  16     def _parse_smil_subtitles(self, smil, namespace=None, subtitles_lang='en'):
  17         subtitles = {}
  18         for k, ext in [('sMPTE-TTCCURL', 'tt'), ('ClosedCaptionURL', 'ttml'), ('webVTTCaptionURL', 'vtt')]:
  19             cc_e = find_xpath_attr(smil, self._xpath_ns('.//param', namespace), 'name', k)
  20             if cc_e is not None:
  21                 cc_url = cc_e.get('value')
  22                 if cc_url:
  23                     subtitles.setdefault(subtitles_lang, []).append({
  24                         'ext': ext,
  25                         'url': cc_url,
  26                     })
  27         return subtitles
  28
  29     def _extract_common_video_info(self, content_id, asset_types, mpx_acc, extra_info):
  30         tp_path = 'dJ5BDC/media/guid/%d/%s' % (mpx_acc, content_id)
  31         tp_release_url = f'https://link.theplatform.com/s/{tp_path}'
  32         info = self._extract_theplatform_metadata(tp_path, content_id)
  33
  34         formats, subtitles = [], {}
  35         last_e = None
  36         for asset_type, query in asset_types.items():
  37             try:
  38                 tp_formats, tp_subtitles = self._extract_theplatform_smil(
  39                     update_url_query(tp_release_url, query), content_id,
  40                     'Downloading %s SMIL data' % asset_type)
  41             except ExtractorError as e:
  42                 last_e = e
  43                 if asset_type != 'fallback':
  44                     continue
  45                 query['formats'] = ''  # blank query to check if expired
  46                 try:
  47                     tp_formats, tp_subtitles = self._extract_theplatform_smil(
  48                         update_url_query(tp_release_url, query), content_id,
  49                         'Downloading %s SMIL data, trying again with another format' % asset_type)
  50                 except ExtractorError as e:
  51                     last_e = e
  52                     continue
  53             formats.extend(tp_formats)
  54             subtitles = self._merge_subtitles(subtitles, tp_subtitles)
  55         if last_e and not formats:
  56             self.raise_no_formats(last_e, True, content_id)
  57         self._sort_formats(formats)
  58
  59         extra_info.update({
  60             'id': content_id,
  61             'formats': formats,
  62             'subtitles': subtitles,
  63         })
  64         info.update({k: v for k, v in extra_info.items() if v is not None})
  65         return info
  66
  67     def _extract_video_info(self, *args, **kwargs):
  68         # Extract assets + metadata and call _extract_common_video_info
  69         raise NotImplementedError('This method must be implemented by subclasses')
  70
  71     def _real_extract(self, url):
  72         return self._extract_video_info(self._match_id(url))
  73
  74
  75 class CBSIE(CBSBaseIE):
  76     _VALID_URL = r'''(?x)
  77         (?:
  78             cbs:|
  79             https?://(?:www\.)?(?:
  80                 cbs\.com/(?:shows/[^/]+/video|movies/[^/]+)/|
  81                 colbertlateshow\.com/(?:video|podcasts)/)
  82         )(?P<id>[\w-]+)'''
  83
  84     # All tests are blocked outside US
  85     _TESTS = [{
  86         'url': 'https://www.cbs.com/shows/garth-brooks/video/_u7W953k6la293J7EPTd9oHkSPs6Xn6_/connect-chat-feat-garth-brooks/',
  87         'info_dict': {
  88             'id': '_u7W953k6la293J7EPTd9oHkSPs6Xn6_',
  89             'ext': 'mp4',
  90             'title': 'Connect Chat feat. Garth Brooks',
  91             '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!',
  92             'duration': 1495,
  93             'timestamp': 1385585425,
  94             'upload_date': '20131127',
  95             'uploader': 'CBSI-NEW',
  96         },
  97         'params': {
  98             # m3u8 download
  99             'skip_download': True,
 100         },
 101     }, {
 102         'url': 'https://www.cbs.com/shows/the-late-show-with-stephen-colbert/video/60icOhMb9NcjbcWnF_gub9XXHdeBcNk2/the-late-show-6-23-21-christine-baranski-joy-oladokun-',
 103         'info_dict': {
 104             'id': '60icOhMb9NcjbcWnF_gub9XXHdeBcNk2',
 105             'title': 'The Late Show - 6/23/21 (Christine Baranski, Joy Oladokun)',
 106             'timestamp': 1624507140,
 107             'description': 'md5:e01af24e95c74d55e8775aef86117b95',
 108             'uploader': 'CBSI-NEW',
 109             'upload_date': '20210624',
 110         },
 111         'params': {
 112             'ignore_no_formats_error': True,
 113             'skip_download': True,
 114         },
 115         'expected_warnings': [
 116             'This content expired on', 'No video formats found', 'Requested format is not available'],
 117     }, {
 118         'url': 'http://colbertlateshow.com/video/8GmB0oY0McANFvp2aEffk9jZZZ2YyXxy/the-colbeard/',
 119         'only_matching': True,
 120     }, {
 121         'url': 'http://www.colbertlateshow.com/podcasts/dYSwjqPs_X1tvbV_P2FcPWRa_qT6akTC/in-the-bad-room-with-stephen/',
 122         'only_matching': True,
 123     }]
 124
 125     def _extract_video_info(self, content_id, site='cbs', mpx_acc=2198311517):
 126         items_data = self._download_xml(
 127             'https://can.cbs.com/thunder/player/videoPlayerService.php',
 128             content_id, query={'partner': site, 'contentId': content_id})
 129         video_data = xpath_element(items_data, './/item')
 130         title = xpath_text(video_data, 'videoTitle', 'title') or xpath_text(video_data, 'videotitle', 'title')
 131
 132         asset_types = {}
 133         has_drm = False
 134         for item in items_data.findall('.//item'):
 135             asset_type = xpath_text(item, 'assetType')
 136             query = {
 137                 'mbr': 'true',
 138                 'assetTypes': asset_type,
 139             }
 140             if not asset_type:
 141                 # fallback for content_ids that videoPlayerService doesn't return anything for
 142                 asset_type = 'fallback'
 143                 query['formats'] = 'M3U+none,MPEG4,M3U+appleHlsEncryption,MP3'
 144                 del query['assetTypes']
 145             if asset_type in asset_types:
 146                 continue
 147             elif any(excluded in asset_type for excluded in ('HLS_FPS', 'DASH_CENC', 'OnceURL')):
 148                 if 'DASH_CENC' in asset_type:
 149                     has_drm = True
 150                 continue
 151             if asset_type.startswith('HLS') or 'StreamPack' in asset_type:
 152                 query['formats'] = 'MPEG4,M3U'
 153             elif asset_type in ('RTMP', 'WIFI', '3G'):
 154                 query['formats'] = 'MPEG4,FLV'
 155             asset_types[asset_type] = query
 156
 157         if not asset_types and has_drm:
 158             self.report_drm(content_id)
 159
 160         return self._extract_common_video_info(content_id, asset_types, mpx_acc, extra_info={
 161             'title': title,
 162             'series': xpath_text(video_data, 'seriesTitle'),
 163             'season_number': int_or_none(xpath_text(video_data, 'seasonNumber')),
 164             'episode_number': int_or_none(xpath_text(video_data, 'episodeNumber')),
 165             'duration': int_or_none(xpath_text(video_data, 'videoLength'), 1000),
 166             'thumbnail': url_or_none(xpath_text(video_data, 'previewImageURL')),
 167         })