_ 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     lowercase_escape,
  14     smuggle_url,
  15     unescapeHTML,
  16 )
  17
  18
  19 class NBCIE(InfoExtractor):
  20     _VALID_URL = r'https?://www\.nbc\.com/(?:[^/]+/)+(?P<id>n?\d+)'
  21
  22     _TESTS = [
  23         {
  24             'url': 'http://www.nbc.com/the-tonight-show/segments/112966',
  25             # md5 checksum is not stable
  26             'info_dict': {
  27                 'id': 'c9xnCo0YPOPH',
  28                 'ext': 'flv',
  29                 'title': 'Jimmy Fallon Surprises Fans at Ben & Jerry\'s',
  30                 'description': 'Jimmy gives out free scoops of his new "Tonight Dough" ice cream flavor by surprising customers at the Ben & Jerry\'s scoop shop.',
  31             },
  32         },
  33         {
  34             'url': 'http://www.nbc.com/the-tonight-show/episodes/176',
  35             'info_dict': {
  36                 'id': 'XwU9KZkp98TH',
  37                 'ext': 'flv',
  38                 'title': 'Ricky Gervais, Steven Van Zandt, ILoveMakonnen',
  39                 'description': 'A brand new episode of The Tonight Show welcomes Ricky Gervais, Steven Van Zandt and ILoveMakonnen.',
  40             },
  41             'skip': 'Only works from US',
  42         },
  43         {
  44             'url': 'http://www.nbc.com/saturday-night-live/video/star-wars-teaser/2832821',
  45             'info_dict': {
  46                 'id': '8iUuyzWDdYUZ',
  47                 'ext': 'flv',
  48                 'title': 'Star Wars Teaser',
  49                 'description': 'md5:0b40f9cbde5b671a7ff62fceccc4f442',
  50             },
  51             'skip': 'Only works from US',
  52         },
  53         {
  54             # This video has expired but with an escaped embedURL
  55             'url': 'http://www.nbc.com/parenthood/episode-guide/season-5/just-like-at-home/515',
  56             'skip': 'Expired'
  57         }
  58     ]
  59
  60     def _real_extract(self, url):
  61         video_id = self._match_id(url)
  62         webpage = self._download_webpage(url, video_id)
  63         theplatform_url = unescapeHTML(lowercase_escape(self._html_search_regex(
  64             [
  65                 r'(?:class="video-player video-player-full" data-mpx-url|class="player" src)="(.*?)"',
  66                 r'"embedURL"\s*:\s*"([^"]+)"'
  67             ],
  68             webpage, 'theplatform url').replace('_no_endcard', '').replace('\\/', '/')))
  69         if theplatform_url.startswith('//'):
  70             theplatform_url = 'http:' + theplatform_url
  71         return self.url_result(smuggle_url(theplatform_url, {'source_url': url}))
  72
  73
  74 class NBCSportsVPlayerIE(InfoExtractor):
  75     _VALID_URL = r'https?://vplayer\.nbcsports\.com/(?:[^/]+/)+(?P<id>[0-9a-zA-Z_]+)'
  76
  77     _TESTS = [{
  78         'url': 'https://vplayer.nbcsports.com/p/BxmELC/nbcsports_share/select/9CsDKds0kvHI',
  79         'info_dict': {
  80             'id': '9CsDKds0kvHI',
  81             'ext': 'flv',
  82             'description': 'md5:df390f70a9ba7c95ff1daace988f0d8d',
  83             'title': 'Tyler Kalinoski hits buzzer-beater to lift Davidson',
  84         }
  85     }, {
  86         'url': 'http://vplayer.nbcsports.com/p/BxmELC/nbc_embedshare/select/_hqLjQ95yx8Z',
  87         'only_matching': True,
  88     }]
  89
  90     @staticmethod
  91     def _extract_url(webpage):
  92         iframe_m = re.search(
  93             r'<iframe[^>]+src="(?P<url>https?://vplayer\.nbcsports\.com/[^"]+)"', webpage)
  94         if iframe_m:
  95             return iframe_m.group('url')
  96
  97     def _real_extract(self, url):
  98         video_id = self._match_id(url)
  99         webpage = self._download_webpage(url, video_id)
 100         theplatform_url = self._og_search_video_url(webpage)
 101         return self.url_result(theplatform_url, 'ThePlatform')
 102
 103
 104 class NBCSportsIE(InfoExtractor):
 105     # Does not include https becuase its certificate is invalid
 106     _VALID_URL = r'http://www\.nbcsports\.com//?(?:[^/]+/)+(?P<id>[0-9a-z-]+)'
 107
 108     _TEST = {
 109         'url': 'http://www.nbcsports.com//college-basketball/ncaab/tom-izzo-michigan-st-has-so-much-respect-duke',
 110         'info_dict': {
 111             'id': 'PHJSaFWbrTY9',
 112             'ext': 'flv',
 113             'title': 'Tom Izzo, Michigan St. has \'so much respect\' for Duke',
 114             'description': 'md5:ecb459c9d59e0766ac9c7d5d0eda8113',
 115         }
 116     }
 117
 118     def _real_extract(self, url):
 119         video_id = self._match_id(url)
 120         webpage = self._download_webpage(url, video_id)
 121         return self.url_result(
 122             NBCSportsVPlayerIE._extract_url(webpage), 'NBCSportsVPlayer')
 123
 124
 125 class NBCNewsIE(InfoExtractor):
 126     _VALID_URL = r'''(?x)https?://(?:www\.)?nbcnews\.com/
 127         (?:video/.+?/(?P<id>\d+)|
 128         (?:watch|feature|nightly-news)/[^/]+/(?P<title>.+))
 129         '''
 130
 131     _TESTS = [
 132         {
 133             'url': 'http://www.nbcnews.com/video/nbc-news/52753292',
 134             'md5': '47abaac93c6eaf9ad37ee6c4463a5179',
 135             'info_dict': {
 136                 'id': '52753292',
 137                 'ext': 'flv',
 138                 'title': 'Crew emerges after four-month Mars food study',
 139                 'description': 'md5:24e632ffac72b35f8b67a12d1b6ddfc1',
 140             },
 141         },
 142         {
 143             'url': 'http://www.nbcnews.com/feature/edward-snowden-interview/how-twitter-reacted-snowden-interview-n117236',
 144             'md5': 'b2421750c9f260783721d898f4c42063',
 145             'info_dict': {
 146                 'id': 'I1wpAI_zmhsQ',
 147                 'ext': 'mp4',
 148                 'title': 'How Twitter Reacted To The Snowden Interview',
 149                 'description': 'md5:65a0bd5d76fe114f3c2727aa3a81fe64',
 150             },
 151             'add_ie': ['ThePlatform'],
 152         },
 153         {
 154             'url': 'http://www.nbcnews.com/feature/dateline-full-episodes/full-episode-family-business-n285156',
 155             'md5': 'fdbf39ab73a72df5896b6234ff98518a',
 156             'info_dict': {
 157                 'id': 'Wjf9EDR3A_60',
 158                 'ext': 'mp4',
 159                 'title': 'FULL EPISODE: Family Business',
 160                 'description': 'md5:757988edbaae9d7be1d585eb5d55cc04',
 161             },
 162         },
 163         {
 164             'url': 'http://www.nbcnews.com/nightly-news/video/nightly-news-with-brian-williams-full-broadcast-february-4-394064451844',
 165             'md5': 'b5dda8cddd8650baa0dcb616dd2cf60d',
 166             'info_dict': {
 167                 'id': 'sekXqyTVnmN3',
 168                 'ext': 'mp4',
 169                 'title': 'Nightly News with Brian Williams Full Broadcast (February 4)',
 170                 'description': 'md5:1c10c1eccbe84a26e5debb4381e2d3c5',
 171             },
 172         },
 173         {
 174             'url': 'http://www.nbcnews.com/watch/dateline/full-episode--deadly-betrayal-386250819952',
 175             'only_matching': True,
 176         },
 177     ]
 178
 179     def _real_extract(self, url):
 180         mobj = re.match(self._VALID_URL, url)
 181         video_id = mobj.group('id')
 182         if video_id is not None:
 183             all_info = self._download_xml('http://www.nbcnews.com/id/%s/displaymode/1219' % video_id, video_id)
 184             info = all_info.find('video')
 185
 186             return {
 187                 'id': video_id,
 188                 'title': info.find('headline').text,
 189                 'ext': 'flv',
 190                 'url': find_xpath_attr(info, 'media', 'type', 'flashVideo').text,
 191                 'description': compat_str(info.find('caption').text),
 192                 'thumbnail': find_xpath_attr(info, 'media', 'type', 'thumbnail').text,
 193             }
 194         else:
 195             # "feature" and "nightly-news" pages use theplatform.com
 196             title = mobj.group('title')
 197             webpage = self._download_webpage(url, title)
 198             bootstrap_json = self._search_regex(
 199                 r'var\s+(?:bootstrapJson|playlistData)\s*=\s*({.+});?\s*$',
 200                 webpage, 'bootstrap json', flags=re.MULTILINE)
 201             bootstrap = self._parse_json(bootstrap_json, video_id)
 202             info = bootstrap['results'][0]['video']
 203             mpxid = info['mpxId']
 204
 205             base_urls = [
 206                 info['fallbackPlaylistUrl'],
 207                 info['associatedPlaylistUrl'],
 208             ]
 209
 210             for base_url in base_urls:
 211                 if not base_url:
 212                     continue
 213                 playlist_url = base_url + '?form=MPXNBCNewsAPI'
 214
 215                 try:
 216                     all_videos = self._download_json(playlist_url, title)
 217                 except ExtractorError as ee:
 218                     if isinstance(ee.cause, compat_HTTPError):
 219                         continue
 220                     raise
 221
 222                 if not all_videos or 'videos' not in all_videos:
 223                     continue
 224
 225                 try:
 226                     info = next(v for v in all_videos['videos'] if v['mpxId'] == mpxid)
 227                     break
 228                 except StopIteration:
 229                     continue
 230
 231             if info is None:
 232                 raise ExtractorError('Could not find video in playlists')
 233
 234             return {
 235                 '_type': 'url',
 236                 # We get the best quality video
 237                 'url': info['videoAssets'][-1]['publicUrl'],
 238                 'ie_key': 'ThePlatform',
 239             }
 240
 241
 242 class MSNBCIE(InfoExtractor):
 243     # https URLs redirect to corresponding http ones
 244     _VALID_URL = r'http://www\.msnbc\.com/[^/]+/watch/(?P<id>[^/]+)'
 245     _TEST = {
 246         'url': 'http://www.msnbc.com/all-in-with-chris-hayes/watch/the-chaotic-gop-immigration-vote-314487875924',
 247         'md5': '6d236bf4f3dddc226633ce6e2c3f814d',
 248         'info_dict': {
 249             'id': 'n_hayes_Aimm_140801_272214',
 250             'ext': 'mp4',
 251             'title': 'The chaotic GOP immigration vote',
 252             '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.',
 253             'thumbnail': 're:^https?://.*\.jpg$',
 254             'timestamp': 1406937606,
 255             'upload_date': '20140802',
 256             'categories': ['MSNBC/Topics/Franchise/Best of last night', 'MSNBC/Topics/General/Congress'],
 257         },
 258     }
 259
 260     def _real_extract(self, url):
 261         video_id = self._match_id(url)
 262         webpage = self._download_webpage(url, video_id)
 263         embed_url = self._html_search_meta('embedURL', webpage)
 264         return self.url_result(embed_url)