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             'url': 'http://www.nbc.com/saturday-night-live/video/star-wars-teaser/2832821',
  42             'info_dict': {
  43                 'id': '8iUuyzWDdYUZ',
  44                 'ext': 'flv',
  45                 'title': 'Star Wars Teaser',
  46                 'description': 'md5:0b40f9cbde5b671a7ff62fceccc4f442',
  47             },
  48             'skip': 'Only works from US',
  49         }
  50     ]
  51
  52     def _real_extract(self, url):
  53         video_id = self._match_id(url)
  54         webpage = self._download_webpage(url, video_id)
  55         theplatform_url = self._search_regex(
  56             [
  57                 r'(?:class="video-player video-player-full" data-mpx-url|class="player" src)="(.*?)"',
  58                 r'"embedURL"\s*:\s*"([^"]+)"'
  59             ],
  60             webpage, 'theplatform url').replace('_no_endcard', '')
  61         if theplatform_url.startswith('//'):
  62             theplatform_url = 'http:' + theplatform_url
  63         return self.url_result(theplatform_url)
  64
  65
  66 class NBCSportsVPlayerIE(InfoExtractor):
  67     _VALID_URL = r'https?://vplayer\.nbcsports\.com/(?:[^/]+/)+(?P<id>[0-9a-zA-Z_]+)'
  68
  69     _TESTS = [{
  70         'url': 'https://vplayer.nbcsports.com/p/BxmELC/nbcsports_share/select/9CsDKds0kvHI',
  71         'info_dict': {
  72             'id': '9CsDKds0kvHI',
  73             'ext': 'flv',
  74             'description': 'md5:df390f70a9ba7c95ff1daace988f0d8d',
  75             'title': 'Tyler Kalinoski hits buzzer-beater to lift Davidson',
  76         }
  77     }, {
  78         'url': 'http://vplayer.nbcsports.com/p/BxmELC/nbc_embedshare/select/_hqLjQ95yx8Z',
  79         'only_matching': True,
  80     }]
  81
  82     @staticmethod
  83     def _extract_url(webpage):
  84         iframe_m = re.search(
  85             r'<iframe[^>]+src="(?P<url>https?://vplayer\.nbcsports\.com/[^"]+)"', webpage)
  86         if iframe_m:
  87             return iframe_m.group('url')
  88
  89     def _real_extract(self, url):
  90         video_id = self._match_id(url)
  91         webpage = self._download_webpage(url, video_id)
  92         theplatform_url = self._og_search_video_url(webpage)
  93         return self.url_result(theplatform_url, 'ThePlatform')
  94
  95
  96 class NBCSportsIE(InfoExtractor):
  97     # Does not include https becuase its certificate is invalid
  98     _VALID_URL = r'http://www\.nbcsports\.com//?(?:[^/]+/)+(?P<id>[0-9a-z-]+)'
  99
 100     _TEST = {
 101         'url': 'http://www.nbcsports.com//college-basketball/ncaab/tom-izzo-michigan-st-has-so-much-respect-duke',
 102         'info_dict': {
 103             'id': 'PHJSaFWbrTY9',
 104             'ext': 'flv',
 105             'title': 'Tom Izzo, Michigan St. has \'so much respect\' for Duke',
 106             'description': 'md5:ecb459c9d59e0766ac9c7d5d0eda8113',
 107         }
 108     }
 109
 110     def _real_extract(self, url):
 111         video_id = self._match_id(url)
 112         webpage = self._download_webpage(url, video_id)
 113         return self.url_result(
 114             NBCSportsVPlayerIE._extract_url(webpage), 'NBCSportsVPlayer')
 115
 116
 117 class NBCNewsIE(InfoExtractor):
 118     _VALID_URL = r'''(?x)https?://(?:www\.)?nbcnews\.com/
 119         (?:video/.+?/(?P<id>\d+)|
 120         (?:feature|nightly-news)/[^/]+/(?P<title>.+))
 121         '''
 122
 123     _TESTS = [
 124         {
 125             'url': 'http://www.nbcnews.com/video/nbc-news/52753292',
 126             'md5': '47abaac93c6eaf9ad37ee6c4463a5179',
 127             'info_dict': {
 128                 'id': '52753292',
 129                 'ext': 'flv',
 130                 'title': 'Crew emerges after four-month Mars food study',
 131                 'description': 'md5:24e632ffac72b35f8b67a12d1b6ddfc1',
 132             },
 133         },
 134         {
 135             'url': 'http://www.nbcnews.com/feature/edward-snowden-interview/how-twitter-reacted-snowden-interview-n117236',
 136             'md5': 'b2421750c9f260783721d898f4c42063',
 137             'info_dict': {
 138                 'id': 'I1wpAI_zmhsQ',
 139                 'ext': 'mp4',
 140                 'title': 'How Twitter Reacted To The Snowden Interview',
 141                 'description': 'md5:65a0bd5d76fe114f3c2727aa3a81fe64',
 142             },
 143             'add_ie': ['ThePlatform'],
 144         },
 145         {
 146             'url': 'http://www.nbcnews.com/feature/dateline-full-episodes/full-episode-family-business-n285156',
 147             'md5': 'fdbf39ab73a72df5896b6234ff98518a',
 148             'info_dict': {
 149                 'id': 'Wjf9EDR3A_60',
 150                 'ext': 'mp4',
 151                 'title': 'FULL EPISODE: Family Business',
 152                 'description': 'md5:757988edbaae9d7be1d585eb5d55cc04',
 153             },
 154         },
 155         {
 156             'url': 'http://www.nbcnews.com/nightly-news/video/nightly-news-with-brian-williams-full-broadcast-february-4-394064451844',
 157             'md5': 'b5dda8cddd8650baa0dcb616dd2cf60d',
 158             'info_dict': {
 159                 'id': 'sekXqyTVnmN3',
 160                 'ext': 'mp4',
 161                 'title': 'Nightly News with Brian Williams Full Broadcast (February 4)',
 162                 'description': 'md5:1c10c1eccbe84a26e5debb4381e2d3c5',
 163             },
 164         },
 165     ]
 166
 167     def _real_extract(self, url):
 168         mobj = re.match(self._VALID_URL, url)
 169         video_id = mobj.group('id')
 170         if video_id is not None:
 171             all_info = self._download_xml('http://www.nbcnews.com/id/%s/displaymode/1219' % video_id, video_id)
 172             info = all_info.find('video')
 173
 174             return {
 175                 'id': video_id,
 176                 'title': info.find('headline').text,
 177                 'ext': 'flv',
 178                 'url': find_xpath_attr(info, 'media', 'type', 'flashVideo').text,
 179                 'description': compat_str(info.find('caption').text),
 180                 'thumbnail': find_xpath_attr(info, 'media', 'type', 'thumbnail').text,
 181             }
 182         else:
 183             # "feature" and "nightly-news" pages use theplatform.com
 184             title = mobj.group('title')
 185             webpage = self._download_webpage(url, title)
 186             bootstrap_json = self._search_regex(
 187                 r'var\s+(?:bootstrapJson|playlistData)\s*=\s*({.+});?\s*$',
 188                 webpage, 'bootstrap json', flags=re.MULTILINE)
 189             bootstrap = self._parse_json(bootstrap_json, video_id)
 190             info = bootstrap['results'][0]['video']
 191             mpxid = info['mpxId']
 192
 193             base_urls = [
 194                 info['fallbackPlaylistUrl'],
 195                 info['associatedPlaylistUrl'],
 196             ]
 197
 198             for base_url in base_urls:
 199                 if not base_url:
 200                     continue
 201                 playlist_url = base_url + '?form=MPXNBCNewsAPI'
 202
 203                 try:
 204                     all_videos = self._download_json(playlist_url, title)
 205                 except ExtractorError as ee:
 206                     if isinstance(ee.cause, compat_HTTPError):
 207                         continue
 208                     raise
 209
 210                 if not all_videos or 'videos' not in all_videos:
 211                     continue
 212
 213                 try:
 214                     info = next(v for v in all_videos['videos'] if v['mpxId'] == mpxid)
 215                     break
 216                 except StopIteration:
 217                     continue
 218
 219             if info is None:
 220                 raise ExtractorError('Could not find video in playlists')
 221
 222             return {
 223                 '_type': 'url',
 224                 # We get the best quality video
 225                 'url': info['videoAssets'][-1]['publicUrl'],
 226                 'ie_key': 'ThePlatform',
 227             }