yt_dlp/extractor/msn.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     ExtractorError,
   6     determine_ext,
   7     int_or_none,
   8     unescapeHTML,
   9 )
  10
  11
  12 class MSNIE(InfoExtractor):
  13     _WORKING = False
  14     _VALID_URL = r'https?://(?:(?:www|preview)\.)?msn\.com/(?:[^/]+/)+(?P<display_id>[^/]+)/[a-z]{2}-(?P<id>[\da-zA-Z]+)'
  15     _TESTS = [{
  16         'url': 'https://www.msn.com/en-in/money/video/7-ways-to-get-rid-of-chest-congestion/vi-BBPxU6d',
  17         'md5': '087548191d273c5c55d05028f8d2cbcd',
  18         'info_dict': {
  19             'id': 'BBPxU6d',
  20             'display_id': '7-ways-to-get-rid-of-chest-congestion',
  21             'ext': 'mp4',
  22             'title': 'Seven ways to get rid of chest congestion',
  23             'description': '7 Ways to Get Rid of Chest Congestion',
  24             'duration': 88,
  25             'uploader': 'Health',
  26             'uploader_id': 'BBPrMqa',
  27         },
  28     }, {
  29         # Article, multiple Dailymotion Embeds
  30         'url': 'https://www.msn.com/en-in/money/sports/hottest-football-wags-greatest-footballers-turned-managers-and-more/ar-BBpc7Nl',
  31         'info_dict': {
  32             'id': 'BBpc7Nl',
  33         },
  34         'playlist_mincount': 4,
  35     }, {
  36         'url': 'http://www.msn.com/en-ae/news/offbeat/meet-the-nine-year-old-self-made-millionaire/ar-BBt6ZKf',
  37         'only_matching': True,
  38     }, {
  39         'url': 'http://www.msn.com/en-ae/video/watch/obama-a-lot-of-people-will-be-disappointed/vi-AAhxUMH',
  40         'only_matching': True,
  41     }, {
  42         # geo restricted
  43         'url': 'http://www.msn.com/en-ae/foodanddrink/joinourtable/the-first-fart-makes-you-laugh-the-last-fart-makes-you-cry/vp-AAhzIBU',
  44         'only_matching': True,
  45     }, {
  46         'url': 'http://www.msn.com/en-ae/entertainment/bollywood/watch-how-salman-khan-reacted-when-asked-if-he-would-apologize-for-his-‘raped-woman’-comment/vi-AAhvzW6',
  47         'only_matching': True,
  48     }, {
  49         # Vidible(AOL) Embed
  50         'url': 'https://www.msn.com/en-us/money/other/jupiter-is-about-to-come-so-close-you-can-see-its-moons-with-binoculars/vi-AACqsHR',
  51         'only_matching': True,
  52     }, {
  53         # Dailymotion Embed
  54         'url': 'https://www.msn.com/es-ve/entretenimiento/watch/winston-salem-paire-refait-des-siennes-en-perdant-sa-raquette-au-service/vp-AAG704L',
  55         'only_matching': True,
  56     }, {
  57         # YouTube Embed
  58         'url': 'https://www.msn.com/en-in/money/news/meet-vikram-%E2%80%94-chandrayaan-2s-lander/vi-AAGUr0v',
  59         'only_matching': True,
  60     }, {
  61         # NBCSports Embed
  62         'url': 'https://www.msn.com/en-us/money/football_nfl/week-13-preview-redskins-vs-panthers/vi-BBXsCDb',
  63         'only_matching': True,
  64     }]
  65
  66     def _real_extract(self, url):
  67         display_id, page_id = self._match_valid_url(url).groups()
  68
  69         webpage = self._download_webpage(url, display_id)
  70
  71         entries = []
  72         for _, metadata in re.findall(r'data-metadata\s*=\s*(["\'])(?P<data>.+?)\1', webpage):
  73             video = self._parse_json(unescapeHTML(metadata), display_id)
  74
  75             provider_id = video.get('providerId')
  76             player_name = video.get('playerName')
  77             if player_name and provider_id:
  78                 entry = None
  79                 if player_name == 'AOL':
  80                     if provider_id.startswith('http'):
  81                         provider_id = self._search_regex(
  82                             r'https?://delivery\.vidible\.tv/video/redirect/([0-9a-f]{24})',
  83                             provider_id, 'vidible id')
  84                     entry = self.url_result(
  85                         'aol-video:' + provider_id, 'Aol', provider_id)
  86                 elif player_name == 'Dailymotion':
  87                     entry = self.url_result(
  88                         'https://www.dailymotion.com/video/' + provider_id,
  89                         'Dailymotion', provider_id)
  90                 elif player_name == 'YouTube':
  91                     entry = self.url_result(
  92                         provider_id, 'Youtube', provider_id)
  93                 elif player_name == 'NBCSports':
  94                     entry = self.url_result(
  95                         'http://vplayer.nbcsports.com/p/BxmELC/nbcsports_embed/select/media/' + provider_id,
  96                         'NBCSportsVPlayer', provider_id)
  97                 if entry:
  98                     entries.append(entry)
  99                     continue
 100
 101             video_id = video['uuid']
 102             title = video['title']
 103
 104             formats = []
 105             for file_ in video.get('videoFiles', []):
 106                 format_url = file_.get('url')
 107                 if not format_url:
 108                     continue
 109                 if 'format=m3u8-aapl' in format_url:
 110                     # m3u8_native should not be used here until
 111                     # https://github.com/ytdl-org/youtube-dl/issues/9913 is fixed
 112                     formats.extend(self._extract_m3u8_formats(
 113                         format_url, display_id, 'mp4',
 114                         m3u8_id='hls', fatal=False))
 115                 elif 'format=mpd-time-csf' in format_url:
 116                     formats.extend(self._extract_mpd_formats(
 117                         format_url, display_id, 'dash', fatal=False))
 118                 elif '.ism' in format_url:
 119                     if format_url.endswith('.ism'):
 120                         format_url += '/manifest'
 121                     formats.extend(self._extract_ism_formats(
 122                         format_url, display_id, 'mss', fatal=False))
 123                 else:
 124                     format_id = file_.get('formatCode')
 125                     formats.append({
 126                         'url': format_url,
 127                         'ext': 'mp4',
 128                         'format_id': format_id,
 129                         'width': int_or_none(file_.get('width')),
 130                         'height': int_or_none(file_.get('height')),
 131                         'vbr': int_or_none(self._search_regex(r'_(\d+)\.mp4', format_url, 'vbr', default=None)),
 132                         'quality': 1 if format_id == '1001' else None,
 133                     })
 134
 135             subtitles = {}
 136             for file_ in video.get('files', []):
 137                 format_url = file_.get('url')
 138                 format_code = file_.get('formatCode')
 139                 if not format_url or not format_code:
 140                     continue
 141                 if str(format_code) == '3100':
 142                     subtitles.setdefault(file_.get('culture', 'en'), []).append({
 143                         'ext': determine_ext(format_url, 'ttml'),
 144                         'url': format_url,
 145                     })
 146
 147             entries.append({
 148                 'id': video_id,
 149                 'display_id': display_id,
 150                 'title': title,
 151                 'description': video.get('description'),
 152                 'thumbnail': video.get('headlineImage', {}).get('url'),
 153                 'duration': int_or_none(video.get('durationSecs')),
 154                 'uploader': video.get('sourceFriendly'),
 155                 'uploader_id': video.get('providerId'),
 156                 'creator': video.get('creator'),
 157                 'subtitles': subtitles,
 158                 'formats': formats,
 159             })
 160
 161         if not entries:
 162             error = unescapeHTML(self._search_regex(
 163                 r'data-error=(["\'])(?P<error>.+?)\1',
 164                 webpage, 'error', group='error'))
 165             raise ExtractorError(f'{self.IE_NAME} said: {error}', expected=True)
 166
 167         return self.playlist_result(entries, page_id)