_ Git - youtube-dl/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\.)?msn\.com/(?:[^/]+/)+(?P<display_id>[^/]+)/[a-z]{2}-(?P<id>[\da-zA-Z]+)'
  18     _TESTS = [{
  19         'url': 'http://www.msn.com/en-ae/foodanddrink/joinourtable/criminal-minds-shemar-moore-shares-a-touching-goodbye-message/vp-BBqQYNE',
  20         'md5': '8442f66c116cbab1ff7098f986983458',
  21         'info_dict': {
  22             'id': 'BBqQYNE',
  23             'display_id': 'criminal-minds-shemar-moore-shares-a-touching-goodbye-message',
  24             'ext': 'mp4',
  25             'title': 'Criminal Minds - Shemar Moore Shares A Touching Goodbye Message',
  26             'description': 'md5:e8e89b897b222eb33a6b5067a8f1bc25',
  27             'duration': 104,
  28             'uploader': 'CBS Entertainment',
  29             'uploader_id': 'IT0X5aoJ6bJgYerJXSDCgFmYPB1__54v',
  30         },
  31     }, {
  32         'url': 'http://www.msn.com/en-ae/news/offbeat/meet-the-nine-year-old-self-made-millionaire/ar-BBt6ZKf',
  33         'only_matching': True,
  34     }, {
  35         'url': 'http://www.msn.com/en-ae/video/watch/obama-a-lot-of-people-will-be-disappointed/vi-AAhxUMH',
  36         'only_matching': True,
  37     }, {
  38         # geo restricted
  39         'url': 'http://www.msn.com/en-ae/foodanddrink/joinourtable/the-first-fart-makes-you-laugh-the-last-fart-makes-you-cry/vp-AAhzIBU',
  40         'only_matching': True,
  41     }, {
  42         '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',
  43         'only_matching': True,
  44     }, {
  45         # Vidible(AOL) Embed
  46         'url': 'https://www.msn.com/en-us/video/animals/yellowstone-park-staffers-catch-deer-engaged-in-behavior-they-cant-explain/vi-AAGfdg1',
  47         'only_matching': True,
  48     }, {
  49         # Dailymotion Embed
  50         'url': 'https://www.msn.com/es-ve/entretenimiento/watch/winston-salem-paire-refait-des-siennes-en-perdant-sa-raquette-au-service/vp-AAG704L',
  51         'only_matching': True,
  52     }]
  53
  54     def _real_extract(self, url):
  55         mobj = re.match(self._VALID_URL, url)
  56         video_id, display_id = mobj.group('id', 'display_id')
  57
  58         webpage = self._download_webpage(url, display_id)
  59
  60         video = self._parse_json(
  61             self._search_regex(
  62                 r'data-metadata\s*=\s*(["\'])(?P<data>.+?)\1',
  63                 webpage, 'video data', default='{}', group='data'),
  64             display_id, transform_source=unescapeHTML)
  65
  66         if not video:
  67             error = unescapeHTML(self._search_regex(
  68                 r'data-error=(["\'])(?P<error>.+?)\1',
  69                 webpage, 'error', group='error'))
  70             raise ExtractorError('%s said: %s' % (self.IE_NAME, error), expected=True)
  71
  72         player_name = video.get('playerName')
  73         if player_name:
  74             provider_id = video.get('providerId')
  75             if provider_id:
  76                 if player_name == 'AOL':
  77                     return self.url_result(
  78                         'aol-video:' + provider_id, 'Aol', provider_id)
  79                 elif player_name == 'Dailymotion':
  80                     return self.url_result(
  81                         'https://www.dailymotion.com/video/' + provider_id,
  82                         'Dailymotion', provider_id)
  83
  84         title = video['title']
  85
  86         formats = []
  87         for file_ in video.get('videoFiles', []):
  88             format_url = file_.get('url')
  89             if not format_url:
  90                 continue
  91             if 'm3u8' in format_url:
  92                 # m3u8_native should not be used here until
  93                 # https://github.com/ytdl-org/youtube-dl/issues/9913 is fixed
  94                 m3u8_formats = self._extract_m3u8_formats(
  95                     format_url, display_id, 'mp4',
  96                     m3u8_id='hls', fatal=False)
  97                 formats.extend(m3u8_formats)
  98             elif determine_ext(format_url) == 'ism':
  99                 formats.extend(self._extract_ism_formats(
 100                     format_url + '/Manifest', display_id, 'mss', fatal=False))
 101             else:
 102                 formats.append({
 103                     'url': format_url,
 104                     'ext': 'mp4',
 105                     'format_id': 'http',
 106                     'width': int_or_none(file_.get('width')),
 107                     'height': int_or_none(file_.get('height')),
 108                 })
 109         self._sort_formats(formats)
 110
 111         subtitles = {}
 112         for file_ in video.get('files', []):
 113             format_url = file_.get('url')
 114             format_code = file_.get('formatCode')
 115             if not format_url or not format_code:
 116                 continue
 117             if compat_str(format_code) == '3100':
 118                 subtitles.setdefault(file_.get('culture', 'en'), []).append({
 119                     'ext': determine_ext(format_url, 'ttml'),
 120                     'url': format_url,
 121                 })
 122
 123         return {
 124             'id': video_id,
 125             'display_id': display_id,
 126             'title': title,
 127             'description': video.get('description'),
 128             'thumbnail': video.get('headlineImage', {}).get('url'),
 129             'duration': int_or_none(video.get('durationSecs')),
 130             'uploader': video.get('sourceFriendly'),
 131             'uploader_id': video.get('providerId'),
 132             'creator': video.get('creator'),
 133             'subtitles': subtitles,
 134             'formats': formats,
 135         }