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 compat_HTTPError
   7 from ..utils import (
   8     ExtractorError,
   9     find_xpath_attr,
  10     lowercase_escape,
  11     smuggle_url,
  12     unescapeHTML,
  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             # This video has expired but with an escaped embedURL
  52             'url': 'http://www.nbc.com/parenthood/episode-guide/season-5/just-like-at-home/515',
  53             'skip': 'Expired'
  54         }
  55     ]
  56
  57     def _real_extract(self, url):
  58         video_id = self._match_id(url)
  59         webpage = self._download_webpage(url, video_id)
  60         theplatform_url = unescapeHTML(lowercase_escape(self._html_search_regex(
  61             [
  62                 r'(?:class="video-player video-player-full" data-mpx-url|class="player" src)="(.*?)"',
  63                 r'<iframe[^>]+src="((?:https?:)?//player\.theplatform\.com/[^"]+)"',
  64                 r'"embedURL"\s*:\s*"([^"]+)"'
  65             ],
  66             webpage, 'theplatform url').replace('_no_endcard', '').replace('\\/', '/')))
  67         if theplatform_url.startswith('//'):
  68             theplatform_url = 'http:' + theplatform_url
  69         return self.url_result(smuggle_url(theplatform_url, {'source_url': url}))
  70
  71
  72 class NBCSportsVPlayerIE(InfoExtractor):
  73     _VALID_URL = r'https?://vplayer\.nbcsports\.com/(?:[^/]+/)+(?P<id>[0-9a-zA-Z_]+)'
  74
  75     _TESTS = [{
  76         'url': 'https://vplayer.nbcsports.com/p/BxmELC/nbcsports_share/select/9CsDKds0kvHI',
  77         'info_dict': {
  78             'id': '9CsDKds0kvHI',
  79             'ext': 'flv',
  80             'description': 'md5:df390f70a9ba7c95ff1daace988f0d8d',
  81             'title': 'Tyler Kalinoski hits buzzer-beater to lift Davidson',
  82         }
  83     }, {
  84         'url': 'http://vplayer.nbcsports.com/p/BxmELC/nbc_embedshare/select/_hqLjQ95yx8Z',
  85         'only_matching': True,
  86     }]
  87
  88     @staticmethod
  89     def _extract_url(webpage):
  90         iframe_m = re.search(
  91             r'<iframe[^>]+src="(?P<url>https?://vplayer\.nbcsports\.com/[^"]+)"', webpage)
  92         if iframe_m:
  93             return iframe_m.group('url')
  94
  95     def _real_extract(self, url):
  96         video_id = self._match_id(url)
  97         webpage = self._download_webpage(url, video_id)
  98         theplatform_url = self._og_search_video_url(webpage)
  99         return self.url_result(theplatform_url, 'ThePlatform')
 100
 101
 102 class NBCSportsIE(InfoExtractor):
 103     # Does not include https becuase its certificate is invalid
 104     _VALID_URL = r'http://www\.nbcsports\.com//?(?:[^/]+/)+(?P<id>[0-9a-z-]+)'
 105
 106     _TEST = {
 107         'url': 'http://www.nbcsports.com//college-basketball/ncaab/tom-izzo-michigan-st-has-so-much-respect-duke',
 108         'info_dict': {
 109             'id': 'PHJSaFWbrTY9',
 110             'ext': 'flv',
 111             'title': 'Tom Izzo, Michigan St. has \'so much respect\' for Duke',
 112             'description': 'md5:ecb459c9d59e0766ac9c7d5d0eda8113',
 113         }
 114     }
 115
 116     def _real_extract(self, url):
 117         video_id = self._match_id(url)
 118         webpage = self._download_webpage(url, video_id)
 119         return self.url_result(
 120             NBCSportsVPlayerIE._extract_url(webpage), 'NBCSportsVPlayer')
 121
 122
 123 class NBCNewsIE(InfoExtractor):
 124     _VALID_URL = r'''(?x)https?://(?:www\.)?nbcnews\.com/
 125         (?:video/.+?/(?P<id>\d+)|
 126         (?:watch|feature|nightly-news)/[^/]+/(?P<title>.+))
 127         '''
 128
 129     _TESTS = [
 130         {
 131             'url': 'http://www.nbcnews.com/video/nbc-news/52753292',
 132             'md5': '47abaac93c6eaf9ad37ee6c4463a5179',
 133             'info_dict': {
 134                 'id': '52753292',
 135                 'ext': 'flv',
 136                 'title': 'Crew emerges after four-month Mars food study',
 137                 'description': 'md5:24e632ffac72b35f8b67a12d1b6ddfc1',
 138             },
 139         },
 140         {
 141             'url': 'http://www.nbcnews.com/feature/edward-snowden-interview/how-twitter-reacted-snowden-interview-n117236',
 142             'md5': 'b2421750c9f260783721d898f4c42063',
 143             'info_dict': {
 144                 'id': 'I1wpAI_zmhsQ',
 145                 'ext': 'mp4',
 146                 'title': 'How Twitter Reacted To The Snowden Interview',
 147                 'description': 'md5:65a0bd5d76fe114f3c2727aa3a81fe64',
 148             },
 149             'add_ie': ['ThePlatform'],
 150         },
 151         {
 152             'url': 'http://www.nbcnews.com/feature/dateline-full-episodes/full-episode-family-business-n285156',
 153             'md5': 'fdbf39ab73a72df5896b6234ff98518a',
 154             'info_dict': {
 155                 'id': 'Wjf9EDR3A_60',
 156                 'ext': 'mp4',
 157                 'title': 'FULL EPISODE: Family Business',
 158                 'description': 'md5:757988edbaae9d7be1d585eb5d55cc04',
 159             },
 160         },
 161         {
 162             'url': 'http://www.nbcnews.com/nightly-news/video/nightly-news-with-brian-williams-full-broadcast-february-4-394064451844',
 163             'md5': 'b5dda8cddd8650baa0dcb616dd2cf60d',
 164             'info_dict': {
 165                 'id': 'sekXqyTVnmN3',
 166                 'ext': 'mp4',
 167                 'title': 'Nightly News with Brian Williams Full Broadcast (February 4)',
 168                 'description': 'md5:1c10c1eccbe84a26e5debb4381e2d3c5',
 169             },
 170         },
 171         {
 172             'url': 'http://www.nbcnews.com/watch/dateline/full-episode--deadly-betrayal-386250819952',
 173             'only_matching': True,
 174         },
 175     ]
 176
 177     def _real_extract(self, url):
 178         mobj = re.match(self._VALID_URL, url)
 179         video_id = mobj.group('id')
 180         if video_id is not None:
 181             all_info = self._download_xml('http://www.nbcnews.com/id/%s/displaymode/1219' % video_id, video_id)
 182             info = all_info.find('video')
 183
 184             return {
 185                 'id': video_id,
 186                 'title': info.find('headline').text,
 187                 'ext': 'flv',
 188                 'url': find_xpath_attr(info, 'media', 'type', 'flashVideo').text,
 189                 'description': info.find('caption').text,
 190                 'thumbnail': find_xpath_attr(info, 'media', 'type', 'thumbnail').text,
 191             }
 192         else:
 193             # "feature" and "nightly-news" pages use theplatform.com
 194             title = mobj.group('title')
 195             webpage = self._download_webpage(url, title)
 196             bootstrap_json = self._search_regex(
 197                 r'var\s+(?:bootstrapJson|playlistData)\s*=\s*({.+});?\s*$',
 198                 webpage, 'bootstrap json', flags=re.MULTILINE)
 199             bootstrap = self._parse_json(bootstrap_json, video_id)
 200             info = bootstrap['results'][0]['video']
 201             mpxid = info['mpxId']
 202
 203             base_urls = [
 204                 info['fallbackPlaylistUrl'],
 205                 info['associatedPlaylistUrl'],
 206             ]
 207
 208             for base_url in base_urls:
 209                 if not base_url:
 210                     continue
 211                 playlist_url = base_url + '?form=MPXNBCNewsAPI'
 212
 213                 try:
 214                     all_videos = self._download_json(playlist_url, title)
 215                 except ExtractorError as ee:
 216                     if isinstance(ee.cause, compat_HTTPError):
 217                         continue
 218                     raise
 219
 220                 if not all_videos or 'videos' not in all_videos:
 221                     continue
 222
 223                 try:
 224                     info = next(v for v in all_videos['videos'] if v['mpxId'] == mpxid)
 225                     break
 226                 except StopIteration:
 227                     continue
 228
 229             if info is None:
 230                 raise ExtractorError('Could not find video in playlists')
 231
 232             return {
 233                 '_type': 'url',
 234                 # We get the best quality video
 235                 'url': info['videoAssets'][-1]['publicUrl'],
 236                 'ie_key': 'ThePlatform',
 237             }
 238
 239
 240 class MSNBCIE(InfoExtractor):
 241     # https URLs redirect to corresponding http ones
 242     _VALID_URL = r'http://www\.msnbc\.com/[^/]+/watch/(?P<id>[^/]+)'
 243     _TEST = {
 244         'url': 'http://www.msnbc.com/all-in-with-chris-hayes/watch/the-chaotic-gop-immigration-vote-314487875924',
 245         'md5': '6d236bf4f3dddc226633ce6e2c3f814d',
 246         'info_dict': {
 247             'id': 'n_hayes_Aimm_140801_272214',
 248             'ext': 'mp4',
 249             'title': 'The chaotic GOP immigration vote',
 250             'description': 'The Republican House votes on a border bill that has no chance of getting through the Senate or signed by the President and is drawing criticism from all sides.',
 251             'thumbnail': 're:^https?://.*\.jpg$',
 252             'timestamp': 1406937606,
 253             'upload_date': '20140802',
 254             'categories': ['MSNBC/Topics/Franchise/Best of last night', 'MSNBC/Topics/General/Congress'],
 255         },
 256     }
 257
 258     def _real_extract(self, url):
 259         video_id = self._match_id(url)
 260         webpage = self._download_webpage(url, video_id)
 261         embed_url = self._html_search_meta('embedURL', webpage)
 262         return self.url_result(embed_url)