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 )
  12
  13
  14 class CBSBaseIE(ThePlatformFeedIE):
  15     def _parse_smil_subtitles(self, smil, namespace=None, subtitles_lang='en'):
  16         subtitles = {}
  17         for k, ext in [('sMPTE-TTCCURL', 'tt'), ('ClosedCaptionURL', 'ttml'), ('webVTTCaptionURL', 'vtt')]:
  18             cc_e = find_xpath_attr(smil, self._xpath_ns('.//param', namespace), 'name', k)
  19             if cc_e is not None:
  20                 cc_url = cc_e.get('value')
  21                 if cc_url:
  22                     subtitles.setdefault(subtitles_lang, []).append({
  23                         'ext': ext,
  24                         'url': cc_url,
  25                     })
  26         return subtitles
  27
  28
  29 class CBSIE(CBSBaseIE):
  30     _VALID_URL = r'''(?x)
  31         (?:
  32             cbs:|
  33             https?://(?:www\.)?(?:
  34                 (?:cbs|paramountplus)\.com/(?:shows/[^/]+/video|movies/[^/]+)/|
  35                 colbertlateshow\.com/(?:video|podcasts)/)
  36         )(?P<id>[\w-]+)'''
  37
  38     _TESTS = [{
  39         'url': 'https://www.cbs.com/shows/garth-brooks/video/_u7W953k6la293J7EPTd9oHkSPs6Xn6_/connect-chat-feat-garth-brooks/',
  40         'info_dict': {
  41             'id': '_u7W953k6la293J7EPTd9oHkSPs6Xn6_',
  42             'ext': 'mp4',
  43             'title': 'Connect Chat feat. Garth Brooks',
  44             '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!',
  45             'duration': 1495,
  46             'timestamp': 1385585425,
  47             'upload_date': '20131127',
  48             'uploader': 'CBSI-NEW',
  49         },
  50         'params': {
  51             # m3u8 download
  52             'skip_download': True,
  53         },
  54         '_skip': 'Blocked outside the US',
  55     }, {
  56         'url': 'http://colbertlateshow.com/video/8GmB0oY0McANFvp2aEffk9jZZZ2YyXxy/the-colbeard/',
  57         'only_matching': True,
  58     }, {
  59         'url': 'http://www.colbertlateshow.com/podcasts/dYSwjqPs_X1tvbV_P2FcPWRa_qT6akTC/in-the-bad-room-with-stephen/',
  60         'only_matching': True,
  61     }, {
  62         'url': 'https://www.paramountplus.com/shows/all-rise/video/QmR1WhNkh1a_IrdHZrbcRklm176X_rVc/all-rise-space/',
  63         'only_matching': True,
  64     }, {
  65         'url': 'https://www.paramountplus.com/movies/million-dollar-american-princesses-meghan-and-harry/C0LpgNwXYeB8txxycdWdR9TjxpJOsdCq',
  66         'only_matching': True,
  67     }]
  68
  69     def _extract_video_info(self, content_id, site='cbs', mpx_acc=2198311517):
  70         items_data = self._download_xml(
  71             'https://can.cbs.com/thunder/player/videoPlayerService.php',
  72             content_id, query={'partner': site, 'contentId': content_id})
  73         video_data = xpath_element(items_data, './/item')
  74         title = xpath_text(video_data, 'videoTitle', 'title') or xpath_text(video_data, 'videotitle', 'title')
  75         tp_path = 'dJ5BDC/media/guid/%d/%s' % (mpx_acc, content_id)
  76         tp_release_url = 'https://link.theplatform.com/s/' + tp_path
  77
  78         asset_types = []
  79         subtitles = {}
  80         formats = []
  81         last_e = None
  82         for item in items_data.findall('.//item'):
  83             asset_type = xpath_text(item, 'assetType')
  84             if not asset_type or asset_type in asset_types or 'HLS_FPS' in asset_type or 'DASH_CENC' in asset_type:
  85                 continue
  86             asset_types.append(asset_type)
  87             query = {
  88                 'mbr': 'true',
  89                 'assetTypes': asset_type,
  90             }
  91             if asset_type.startswith('HLS') or asset_type in ('OnceURL', 'StreamPack'):
  92                 query['formats'] = 'MPEG4,M3U'
  93             elif asset_type in ('RTMP', 'WIFI', '3G'):
  94                 query['formats'] = 'MPEG4,FLV'
  95             try:
  96                 tp_formats, tp_subtitles = self._extract_theplatform_smil(
  97                     update_url_query(tp_release_url, query), content_id,
  98                     'Downloading %s SMIL data' % asset_type)
  99             except ExtractorError as e:
 100                 last_e = e
 101                 continue
 102             formats.extend(tp_formats)
 103             subtitles = self._merge_subtitles(subtitles, tp_subtitles)
 104         if last_e and not formats:
 105             raise last_e
 106         self._sort_formats(formats)
 107
 108         info = self._extract_theplatform_metadata(tp_path, content_id)
 109         info.update({
 110             'id': content_id,
 111             'title': title,
 112             'series': xpath_text(video_data, 'seriesTitle'),
 113             'season_number': int_or_none(xpath_text(video_data, 'seasonNumber')),
 114             'episode_number': int_or_none(xpath_text(video_data, 'episodeNumber')),
 115             'duration': int_or_none(xpath_text(video_data, 'videoLength'), 1000),
 116             'thumbnail': xpath_text(video_data, 'previewImageURL'),
 117             'formats': formats,
 118             'subtitles': subtitles,
 119         })
 120         return info
 121
 122     def _real_extract(self, url):
 123         content_id = self._match_id(url)
 124         return self._extract_video_info(content_id)