yt_dlp/extractor/cbs.py

   1 from .theplatform import ThePlatformFeedIE
   2 from ..utils import (
   3     ExtractorError,
   4     int_or_none,
   5     find_xpath_attr,
   6     xpath_element,
   7     xpath_text,
   8     update_url_query,
   9     url_or_none,
  10 )
  11
  12
  13 class CBSBaseIE(ThePlatformFeedIE):
  14     def _parse_smil_subtitles(self, smil, namespace=None, subtitles_lang='en'):
  15         subtitles = {}
  16         for k, ext in [('sMPTE-TTCCURL', 'tt'), ('ClosedCaptionURL', 'ttml'), ('webVTTCaptionURL', 'vtt')]:
  17             cc_e = find_xpath_attr(smil, self._xpath_ns('.//param', namespace), 'name', k)
  18             if cc_e is not None:
  19                 cc_url = cc_e.get('value')
  20                 if cc_url:
  21                     subtitles.setdefault(subtitles_lang, []).append({
  22                         'ext': ext,
  23                         'url': cc_url,
  24                     })
  25         return subtitles
  26
  27     def _extract_common_video_info(self, content_id, asset_types, mpx_acc, extra_info):
  28         tp_path = 'dJ5BDC/media/guid/%d/%s' % (mpx_acc, content_id)
  29         tp_release_url = f'https://link.theplatform.com/s/{tp_path}'
  30         info = self._extract_theplatform_metadata(tp_path, content_id)
  31
  32         formats, subtitles = [], {}
  33         last_e = None
  34         for asset_type, query in asset_types.items():
  35             try:
  36                 tp_formats, tp_subtitles = self._extract_theplatform_smil(
  37                     update_url_query(tp_release_url, query), content_id,
  38                     'Downloading %s SMIL data' % asset_type)
  39             except ExtractorError as e:
  40                 last_e = e
  41                 if asset_type != 'fallback':
  42                     continue
  43                 query['formats'] = ''  # blank query to check if expired
  44                 try:
  45                     tp_formats, tp_subtitles = self._extract_theplatform_smil(
  46                         update_url_query(tp_release_url, query), content_id,
  47                         'Downloading %s SMIL data, trying again with another format' % asset_type)
  48                 except ExtractorError as e:
  49                     last_e = e
  50                     continue
  51             formats.extend(tp_formats)
  52             subtitles = self._merge_subtitles(subtitles, tp_subtitles)
  53         if last_e and not formats:
  54             self.raise_no_formats(last_e, True, content_id)
  55         self._sort_formats(formats)
  56
  57         extra_info.update({
  58             'id': content_id,
  59             'formats': formats,
  60             'subtitles': subtitles,
  61         })
  62         info.update({k: v for k, v in extra_info.items() if v is not None})
  63         return info
  64
  65     def _extract_video_info(self, *args, **kwargs):
  66         # Extract assets + metadata and call _extract_common_video_info
  67         raise NotImplementedError('This method must be implemented by subclasses')
  68
  69     def _real_extract(self, url):
  70         return self._extract_video_info(self._match_id(url))
  71
  72
  73 class CBSIE(CBSBaseIE):
  74     _VALID_URL = r'''(?x)
  75         (?:
  76             cbs:|
  77             https?://(?:www\.)?(?:
  78                 cbs\.com/(?:shows|movies)/(?:video|[^/]+/video|[^/]+)/|
  79                 colbertlateshow\.com/(?:video|podcasts)/)
  80         )(?P<id>[\w-]+)'''
  81
  82     # All tests are blocked outside US
  83     _TESTS = [{
  84         'url': 'https://www.cbs.com/shows/video/xrUyNLtl9wd8D_RWWAg9NU2F_V6QpB3R/',
  85         'info_dict': {
  86             'id': 'xrUyNLtl9wd8D_RWWAg9NU2F_V6QpB3R',
  87             'ext': 'mp4',
  88             'title': 'Tough As Nails - Dreams Never Die',
  89             'description': 'md5:a3535a62531cdd52b0364248a2c1ae33',
  90             'duration': 2588,
  91             'timestamp': 1639015200,
  92             'upload_date': '20211209',
  93             'uploader': 'CBSI-NEW',
  94         },
  95         'params': {
  96             # m3u8 download
  97             'skip_download': True,
  98         },
  99     }, {
 100         'url': 'https://www.cbs.com/shows/video/sZH1MGgomIosZgxGJ1l263MFq16oMtW1/',
 101         'info_dict': {
 102             'id': 'sZH1MGgomIosZgxGJ1l263MFq16oMtW1',
 103             'title': 'The Late Show - 3/16/22 (Michael Buble, Rose Matafeo)',
 104             'timestamp': 1647488100,
 105             'description': 'md5:d0e6ec23c544b7fa8e39a8e6844d2439',
 106             'uploader': 'CBSI-NEW',
 107             'upload_date': '20220317',
 108         },
 109         'params': {
 110             'ignore_no_formats_error': True,
 111             'skip_download': True,
 112         },
 113         'expected_warnings': [
 114             'This content expired on', 'No video formats found', 'Requested format is not available'],
 115     }, {
 116         'url': 'http://colbertlateshow.com/video/8GmB0oY0McANFvp2aEffk9jZZZ2YyXxy/the-colbeard/',
 117         'only_matching': True,
 118     }, {
 119         'url': 'http://www.colbertlateshow.com/podcasts/dYSwjqPs_X1tvbV_P2FcPWRa_qT6akTC/in-the-bad-room-with-stephen/',
 120         'only_matching': True,
 121     }]
 122
 123     def _extract_video_info(self, content_id, site='cbs', mpx_acc=2198311517):
 124         items_data = self._download_xml(
 125             'https://can.cbs.com/thunder/player/videoPlayerService.php',
 126             content_id, query={'partner': site, 'contentId': content_id})
 127         video_data = xpath_element(items_data, './/item')
 128         title = xpath_text(video_data, 'videoTitle', 'title') or xpath_text(video_data, 'videotitle', 'title')
 129
 130         asset_types = {}
 131         has_drm = False
 132         for item in items_data.findall('.//item'):
 133             asset_type = xpath_text(item, 'assetType')
 134             query = {
 135                 'mbr': 'true',
 136                 'assetTypes': asset_type,
 137             }
 138             if not asset_type:
 139                 # fallback for content_ids that videoPlayerService doesn't return anything for
 140                 asset_type = 'fallback'
 141                 query['formats'] = 'M3U+none,MPEG4,M3U+appleHlsEncryption,MP3'
 142                 del query['assetTypes']
 143             if asset_type in asset_types:
 144                 continue
 145             elif any(excluded in asset_type for excluded in ('HLS_FPS', 'DASH_CENC', 'OnceURL')):
 146                 if 'DASH_CENC' in asset_type:
 147                     has_drm = True
 148                 continue
 149             if asset_type.startswith('HLS') or 'StreamPack' in asset_type:
 150                 query['formats'] = 'MPEG4,M3U'
 151             elif asset_type in ('RTMP', 'WIFI', '3G'):
 152                 query['formats'] = 'MPEG4,FLV'
 153             asset_types[asset_type] = query
 154
 155         if not asset_types and has_drm:
 156             self.report_drm(content_id)
 157
 158         return self._extract_common_video_info(content_id, asset_types, mpx_acc, extra_info={
 159             'title': title,
 160             'series': xpath_text(video_data, 'seriesTitle'),
 161             'season_number': int_or_none(xpath_text(video_data, 'seasonNumber')),
 162             'episode_number': int_or_none(xpath_text(video_data, 'episodeNumber')),
 163             'duration': int_or_none(xpath_text(video_data, 'videoLength'), 1000),
 164             'thumbnail': url_or_none(xpath_text(video_data, 'previewImageURL')),
 165         })