Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/msn.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import compat_str
   8 from ..utils import (
   9     determine_ext,
  10     ExtractorError,
  11     int_or_none,
  12     unescapeHTML,
  13 )
  14
  15
  16 class MSNIE(InfoExtractor):
  17     _VALID_URL = r'https?://(?:(?:www|preview)\.)?msn\.com/(?:[^/]+/)+(?P<display_id>[^/]+)/[a-z]{2}-(?P<id>[\da-zA-Z]+)'
  18     _TESTS = [{
  19         'url': 'https://www.msn.com/en-in/money/video/7-ways-to-get-rid-of-chest-congestion/vi-BBPxU6d',
  20         'md5': '087548191d273c5c55d05028f8d2cbcd',
  21         'info_dict': {
  22             'id': 'BBPxU6d',
  23             'display_id': '7-ways-to-get-rid-of-chest-congestion',
  24             'ext': 'mp4',
  25             'title': 'Seven ways to get rid of chest congestion',
  26             'description': '7 Ways to Get Rid of Chest Congestion',
  27             'duration': 88,
  28             'uploader': 'Health',
  29             'uploader_id': 'BBPrMqa',
  30         },
  31     }, {
  32         # Article, multiple Dailymotion Embeds
  33         'url': 'https://www.msn.com/en-in/money/sports/hottest-football-wags-greatest-footballers-turned-managers-and-more/ar-BBpc7Nl',
  34         'info_dict': {
  35             'id': 'BBpc7Nl',
  36         },
  37         'playlist_mincount': 4,
  38     }, {
  39         'url': 'http://www.msn.com/en-ae/news/offbeat/meet-the-nine-year-old-self-made-millionaire/ar-BBt6ZKf',
  40         'only_matching': True,
  41     }, {
  42         'url': 'http://www.msn.com/en-ae/video/watch/obama-a-lot-of-people-will-be-disappointed/vi-AAhxUMH',
  43         'only_matching': True,
  44     }, {
  45         # geo restricted
  46         'url': 'http://www.msn.com/en-ae/foodanddrink/joinourtable/the-first-fart-makes-you-laugh-the-last-fart-makes-you-cry/vp-AAhzIBU',
  47         'only_matching': True,
  48     }, {
  49         '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',
  50         'only_matching': True,
  51     }, {
  52         # Vidible(AOL) Embed
  53         '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',
  54         'only_matching': True,
  55     }, {
  56         # Dailymotion Embed
  57         'url': 'https://www.msn.com/es-ve/entretenimiento/watch/winston-salem-paire-refait-des-siennes-en-perdant-sa-raquette-au-service/vp-AAG704L',
  58         'only_matching': True,
  59     }, {
  60         # YouTube Embed
  61         'url': 'https://www.msn.com/en-in/money/news/meet-vikram-%E2%80%94-chandrayaan-2s-lander/vi-AAGUr0v',
  62         'only_matching': True,
  63     }, {
  64         # NBCSports Embed
  65         'url': 'https://www.msn.com/en-us/money/football_nfl/week-13-preview-redskins-vs-panthers/vi-BBXsCDb',
  66         'only_matching': True,
  67     }]
  68
  69     def _real_extract(self, url):
  70         display_id, page_id = re.match(self._VALID_URL, url).groups()
  71
  72         webpage = self._download_webpage(url, display_id)
  73
  74         entries = []
  75         for _, metadata in re.findall(r'data-metadata\s*=\s*(["\'])(?P<data>.+?)\1', webpage):
  76             video = self._parse_json(unescapeHTML(metadata), display_id)
  77
  78             provider_id = video.get('providerId')
  79             player_name = video.get('playerName')
  80             if player_name and provider_id:
  81                 entry = None
  82                 if player_name == 'AOL':
  83                     if provider_id.startswith('http'):
  84                         provider_id = self._search_regex(
  85                             r'https?://delivery\.vidible\.tv/video/redirect/([0-9a-f]{24})',
  86                             provider_id, 'vidible id')
  87                     entry = self.url_result(
  88                         'aol-video:' + provider_id, 'Aol', provider_id)
  89                 elif player_name == 'Dailymotion':
  90                     entry = self.url_result(
  91                         'https://www.dailymotion.com/video/' + provider_id,
  92                         'Dailymotion', provider_id)
  93                 elif player_name == 'YouTube':
  94                     entry = self.url_result(
  95                         provider_id, 'Youtube', provider_id)
  96                 elif player_name == 'NBCSports':
  97                     entry = self.url_result(
  98                         'http://vplayer.nbcsports.com/p/BxmELC/nbcsports_embed/select/media/' + provider_id,
  99                         'NBCSportsVPlayer', provider_id)
 100                 if entry:
 101                     entries.append(entry)
 102                     continue
 103
 104             video_id = video['uuid']
 105             title = video['title']
 106
 107             formats = []
 108             for file_ in video.get('videoFiles', []):
 109                 format_url = file_.get('url')
 110                 if not format_url:
 111                     continue
 112                 if 'format=m3u8-aapl' in format_url:
 113                     # m3u8_native should not be used here until
 114                     # https://github.com/ytdl-org/youtube-dl/issues/9913 is fixed
 115                     formats.extend(self._extract_m3u8_formats(
 116                         format_url, display_id, 'mp4',
 117                         m3u8_id='hls', fatal=False))
 118                 elif 'format=mpd-time-csf' in format_url:
 119                     formats.extend(self._extract_mpd_formats(
 120                         format_url, display_id, 'dash', fatal=False))
 121                 elif '.ism' in format_url:
 122                     if format_url.endswith('.ism'):
 123                         format_url += '/manifest'
 124                     formats.extend(self._extract_ism_formats(
 125                         format_url, display_id, 'mss', fatal=False))
 126                 else:
 127                     format_id = file_.get('formatCode')
 128                     formats.append({
 129                         'url': format_url,
 130                         'ext': 'mp4',
 131                         'format_id': format_id,
 132                         'width': int_or_none(file_.get('width')),
 133                         'height': int_or_none(file_.get('height')),
 134                         'vbr': int_or_none(self._search_regex(r'_(\d+)\.mp4', format_url, 'vbr', default=None)),
 135                         'preference': 1 if format_id == '1001' else None,
 136                     })
 137             self._sort_formats(formats)
 138
 139             subtitles = {}
 140             for file_ in video.get('files', []):
 141                 format_url = file_.get('url')
 142                 format_code = file_.get('formatCode')
 143                 if not format_url or not format_code:
 144                     continue
 145                 if compat_str(format_code) == '3100':
 146                     subtitles.setdefault(file_.get('culture', 'en'), []).append({
 147                         'ext': determine_ext(format_url, 'ttml'),
 148                         'url': format_url,
 149                     })
 150
 151             entries.append({
 152                 'id': video_id,
 153                 'display_id': display_id,
 154                 'title': title,
 155                 'description': video.get('description'),
 156                 'thumbnail': video.get('headlineImage', {}).get('url'),
 157                 'duration': int_or_none(video.get('durationSecs')),
 158                 'uploader': video.get('sourceFriendly'),
 159                 'uploader_id': video.get('providerId'),
 160                 'creator': video.get('creator'),
 161                 'subtitles': subtitles,
 162                 'formats': formats,
 163             })
 164
 165         if not entries:
 166             error = unescapeHTML(self._search_regex(
 167                 r'data-error=(["\'])(?P<error>.+?)\1',
 168                 webpage, 'error', group='error'))
 169             raise ExtractorError('%s said: %s' % (self.IE_NAME, error), expected=True)
 170
 171         return self.playlist_result(entries, page_id)