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