_ Git - youtube-dl/blob - youtube_dl/extractor/nbc.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..compat import (
   7     compat_str,
   8     compat_HTTPError,
   9 )
  10 from ..utils import (
  11     ExtractorError,
  12     find_xpath_attr,
  13 )
  14
  15
  16 class NBCIE(InfoExtractor):
  17     _VALID_URL = r'https?://www\.nbc\.com/(?:[^/]+/)+(?P<id>n?\d+)'
  18
  19     _TESTS = [
  20         {
  21             'url': 'http://www.nbc.com/the-tonight-show/segments/112966',
  22             # md5 checksum is not stable
  23             'info_dict': {
  24                 'id': 'c9xnCo0YPOPH',
  25                 'ext': 'flv',
  26                 'title': 'Jimmy Fallon Surprises Fans at Ben & Jerry\'s',
  27                 'description': 'Jimmy gives out free scoops of his new "Tonight Dough" ice cream flavor by surprising customers at the Ben & Jerry\'s scoop shop.',
  28             },
  29         },
  30         {
  31             'url': 'http://www.nbc.com/the-tonight-show/episodes/176',
  32             'info_dict': {
  33                 'id': 'XwU9KZkp98TH',
  34                 'ext': 'flv',
  35                 'title': 'Ricky Gervais, Steven Van Zandt, ILoveMakonnen',
  36                 'description': 'A brand new episode of The Tonight Show welcomes Ricky Gervais, Steven Van Zandt and ILoveMakonnen.',
  37             },
  38             'skip': 'Only works from US',
  39         },
  40     ]
  41
  42     def _real_extract(self, url):
  43         video_id = self._match_id(url)
  44         webpage = self._download_webpage(url, video_id)
  45         theplatform_url = self._search_regex(
  46             '(?:class="video-player video-player-full" data-mpx-url|class="player" src)="(.*?)"',
  47             webpage, 'theplatform url').replace('_no_endcard', '')
  48         if theplatform_url.startswith('//'):
  49             theplatform_url = 'http:' + theplatform_url
  50         return self.url_result(theplatform_url)
  51
  52
  53 class NBCSportsVPlayerIE(InfoExtractor):
  54     _VALID_URL = r'https?://vplayer\.nbcsports\.com/(?:[^/]+/)+(?P<id>[0-9a-zA-Z_]+)'
  55
  56     _TESTS = [{
  57         'url': 'https://vplayer.nbcsports.com/p/BxmELC/nbcsports_share/select/9CsDKds0kvHI',
  58         'info_dict': {
  59             'id': '9CsDKds0kvHI',
  60             'ext': 'flv',
  61             'description': 'md5:df390f70a9ba7c95ff1daace988f0d8d',
  62             'title': 'Tyler Kalinoski hits buzzer-beater to lift Davidson',
  63         }
  64     }, {
  65         'note': 'This video is already expired. It\'s for testing _VALID_URL',
  66         'url': 'http://vplayer.nbcsports.com/p/BxmELC/nbc_embedshare/select/_hqLjQ95yx8Z',
  67         'only_matching': True,
  68     }]
  69
  70     @staticmethod
  71     def _extract_url(webpage):
  72         iframe_m = re.search(
  73             r'<iframe[^>]+src="(?P<url>https?://vplayer\.nbcsports\.com/[^"]+)"', webpage)
  74         if iframe_m:
  75             return iframe_m.group('url')
  76
  77     def _real_extract(self, url):
  78         video_id = self._match_id(url)
  79         webpage = self._download_webpage(url, video_id)
  80         theplatform_url = self._og_search_video_url(webpage)
  81         return self.url_result(theplatform_url, 'ThePlatform')
  82
  83
  84 class NBCSportsIE(InfoExtractor):
  85     # Does not include https becuase its certificate is invalid
  86     _VALID_URL = r'http://www\.nbcsports\.com//?(?:[^/]+/)+(?P<id>[0-9a-z-]+)'
  87
  88     _TEST = {
  89         'url': 'http://www.nbcsports.com//college-basketball/ncaab/tom-izzo-michigan-st-has-so-much-respect-duke',
  90         'info_dict': {
  91             'id': 'PHJSaFWbrTY9',
  92             'ext': 'flv',
  93             'title': 'Tom Izzo, Michigan St. has \'so much respect\' for Duke',
  94             'description': 'md5:ecb459c9d59e0766ac9c7d5d0eda8113',
  95         }
  96     }
  97
  98     def _real_extract(self, url):
  99         video_id = self._match_id(url)
 100         webpage = self._download_webpage(url, video_id)
 101         return self.url_result(
 102             NBCSportsVPlayerIE._extract_url(webpage), 'NBCSportsVPlayer')
 103
 104
 105 class NBCNewsIE(InfoExtractor):
 106     _VALID_URL = r'''(?x)https?://(?:www\.)?nbcnews\.com/
 107         (?:video/.+?/(?P<id>\d+)|
 108         (?:feature|nightly-news)/[^/]+/(?P<title>.+))
 109         '''
 110
 111     _TESTS = [
 112         {
 113             'url': 'http://www.nbcnews.com/video/nbc-news/52753292',
 114             'md5': '47abaac93c6eaf9ad37ee6c4463a5179',
 115             'info_dict': {
 116                 'id': '52753292',
 117                 'ext': 'flv',
 118                 'title': 'Crew emerges after four-month Mars food study',
 119                 'description': 'md5:24e632ffac72b35f8b67a12d1b6ddfc1',
 120             },
 121         },
 122         {
 123             'url': 'http://www.nbcnews.com/feature/edward-snowden-interview/how-twitter-reacted-snowden-interview-n117236',
 124             'md5': 'b2421750c9f260783721d898f4c42063',
 125             'info_dict': {
 126                 'id': 'I1wpAI_zmhsQ',
 127                 'ext': 'mp4',
 128                 'title': 'How Twitter Reacted To The Snowden Interview',
 129                 'description': 'md5:65a0bd5d76fe114f3c2727aa3a81fe64',
 130             },
 131             'add_ie': ['ThePlatform'],
 132         },
 133         {
 134             'url': 'http://www.nbcnews.com/feature/dateline-full-episodes/full-episode-family-business-n285156',
 135             'md5': 'fdbf39ab73a72df5896b6234ff98518a',
 136             'info_dict': {
 137                 'id': 'Wjf9EDR3A_60',
 138                 'ext': 'mp4',
 139                 'title': 'FULL EPISODE: Family Business',
 140                 'description': 'md5:757988edbaae9d7be1d585eb5d55cc04',
 141             },
 142         },
 143         {
 144             'url': 'http://www.nbcnews.com/nightly-news/video/nightly-news-with-brian-williams-full-broadcast-february-4-394064451844',
 145             'md5': 'b5dda8cddd8650baa0dcb616dd2cf60d',
 146             'info_dict': {
 147                 'id': 'sekXqyTVnmN3',
 148                 'ext': 'mp4',
 149                 'title': 'Nightly News with Brian Williams Full Broadcast (February 4)',
 150                 'description': 'md5:1c10c1eccbe84a26e5debb4381e2d3c5',
 151             },
 152         },
 153     ]
 154
 155     def _real_extract(self, url):
 156         mobj = re.match(self._VALID_URL, url)
 157         video_id = mobj.group('id')
 158         if video_id is not None:
 159             all_info = self._download_xml('http://www.nbcnews.com/id/%s/displaymode/1219' % video_id, video_id)
 160             info = all_info.find('video')
 161
 162             return {
 163                 'id': video_id,
 164                 'title': info.find('headline').text,
 165                 'ext': 'flv',
 166                 'url': find_xpath_attr(info, 'media', 'type', 'flashVideo').text,
 167                 'description': compat_str(info.find('caption').text),
 168                 'thumbnail': find_xpath_attr(info, 'media', 'type', 'thumbnail').text,
 169             }
 170         else:
 171             # "feature" and "nightly-news" pages use theplatform.com
 172             title = mobj.group('title')
 173             webpage = self._download_webpage(url, title)
 174             bootstrap_json = self._search_regex(
 175                 r'var\s+(?:bootstrapJson|playlistData)\s*=\s*({.+});?\s*$',
 176                 webpage, 'bootstrap json', flags=re.MULTILINE)
 177             bootstrap = self._parse_json(bootstrap_json, video_id)
 178             info = bootstrap['results'][0]['video']
 179             mpxid = info['mpxId']
 180
 181             base_urls = [
 182                 info['fallbackPlaylistUrl'],
 183                 info['associatedPlaylistUrl'],
 184             ]
 185
 186             for base_url in base_urls:
 187                 if not base_url:
 188                     continue
 189                 playlist_url = base_url + '?form=MPXNBCNewsAPI'
 190
 191                 try:
 192                     all_videos = self._download_json(playlist_url, title)
 193                 except ExtractorError as ee:
 194                     if isinstance(ee.cause, compat_HTTPError):
 195                         continue
 196                     raise
 197
 198                 if not all_videos or 'videos' not in all_videos:
 199                     continue
 200
 201                 try:
 202                     info = next(v for v in all_videos['videos'] if v['mpxId'] == mpxid)
 203                     break
 204                 except StopIteration:
 205                     continue
 206
 207             if info is None:
 208                 raise ExtractorError('Could not find video in playlists')
 209
 210             return {
 211                 '_type': 'url',
 212                 # We get the best quality video
 213                 'url': info['videoAssets'][-1]['publicUrl'],
 214                 'ie_key': 'ThePlatform',
 215             }