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             'info_dict': {
  23                 'id': '112966',
  24                 'ext': 'mp4',
  25                 'title': 'Jimmy Fallon Surprises Fans at Ben & Jerry\'s',
  26                 'description': 'Jimmy gives out free scoops of his new "Tonight Dough" ice cream flavor by surprising customers at the Ben & Jerry\'s scoop shop.',
  27             },
  28             'params': {
  29                 # m3u8 download
  30                 'skip_download': True,
  31             },
  32         },
  33         {
  34             'url': 'http://www.nbc.com/the-tonight-show/episodes/176',
  35             'info_dict': {
  36                 'id': '176',
  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': '404 Not Found',
  42         },
  43         {
  44             'url': 'http://www.nbc.com/saturday-night-live/video/star-wars-teaser/2832821',
  45             'info_dict': {
  46                 'id': '2832821',
  47                 'ext': 'mp4',
  48                 'title': 'Star Wars Teaser',
  49                 'description': 'md5:0b40f9cbde5b671a7ff62fceccc4f442',
  50             },
  51             'params': {
  52                 # m3u8 download
  53                 'skip_download': True,
  54             },
  55             'skip': 'Only works from US',
  56         },
  57         {
  58             # This video has expired but with an escaped embedURL
  59             'url': 'http://www.nbc.com/parenthood/episode-guide/season-5/just-like-at-home/515',
  60             'only_matching': True,
  61         }
  62     ]
  63
  64     def _real_extract(self, url):
  65         video_id = self._match_id(url)
  66         webpage = self._download_webpage(url, video_id)
  67         theplatform_url = unescapeHTML(lowercase_escape(self._html_search_regex(
  68             [
  69                 r'(?:class="video-player video-player-full" data-mpx-url|class="player" src)="(.*?)"',
  70                 r'<iframe[^>]+src="((?:https?:)?//player\.theplatform\.com/[^"]+)"',
  71                 r'"embedURL"\s*:\s*"([^"]+)"'
  72             ],
  73             webpage, 'theplatform url').replace('_no_endcard', '').replace('\\/', '/')))
  74         if theplatform_url.startswith('//'):
  75             theplatform_url = 'http:' + theplatform_url
  76         return {
  77             '_type': 'url_transparent',
  78             'url': smuggle_url(theplatform_url, {'source_url': url}),
  79             'id': video_id,
  80         }
  81
  82
  83 class NBCSportsVPlayerIE(InfoExtractor):
  84     _VALID_URL = r'https?://vplayer\.nbcsports\.com/(?:[^/]+/)+(?P<id>[0-9a-zA-Z_]+)'
  85
  86     _TESTS = [{
  87         'url': 'https://vplayer.nbcsports.com/p/BxmELC/nbcsports_share/select/9CsDKds0kvHI',
  88         'info_dict': {
  89             'id': '9CsDKds0kvHI',
  90             'ext': 'flv',
  91             'description': 'md5:df390f70a9ba7c95ff1daace988f0d8d',
  92             'title': 'Tyler Kalinoski hits buzzer-beater to lift Davidson',
  93         }
  94     }, {
  95         'url': 'http://vplayer.nbcsports.com/p/BxmELC/nbc_embedshare/select/_hqLjQ95yx8Z',
  96         'only_matching': True,
  97     }]
  98
  99     @staticmethod
 100     def _extract_url(webpage):
 101         iframe_m = re.search(
 102             r'<iframe[^>]+src="(?P<url>https?://vplayer\.nbcsports\.com/[^"]+)"', webpage)
 103         if iframe_m:
 104             return iframe_m.group('url')
 105
 106     def _real_extract(self, url):
 107         video_id = self._match_id(url)
 108         webpage = self._download_webpage(url, video_id)
 109         theplatform_url = self._og_search_video_url(webpage)
 110         return self.url_result(theplatform_url, 'ThePlatform')
 111
 112
 113 class NBCSportsIE(InfoExtractor):
 114     # Does not include https because its certificate is invalid
 115     _VALID_URL = r'http://www\.nbcsports\.com//?(?:[^/]+/)+(?P<id>[0-9a-z-]+)'
 116
 117     _TEST = {
 118         'url': 'http://www.nbcsports.com//college-basketball/ncaab/tom-izzo-michigan-st-has-so-much-respect-duke',
 119         'info_dict': {
 120             'id': 'PHJSaFWbrTY9',
 121             'ext': 'flv',
 122             'title': 'Tom Izzo, Michigan St. has \'so much respect\' for Duke',
 123             'description': 'md5:ecb459c9d59e0766ac9c7d5d0eda8113',
 124         }
 125     }
 126
 127     def _real_extract(self, url):
 128         video_id = self._match_id(url)
 129         webpage = self._download_webpage(url, video_id)
 130         return self.url_result(
 131             NBCSportsVPlayerIE._extract_url(webpage), 'NBCSportsVPlayer')
 132
 133
 134 class NBCNewsIE(InfoExtractor):
 135     _VALID_URL = r'''(?x)https?://(?:www\.)?nbcnews\.com/
 136         (?:video/.+?/(?P<id>\d+)|
 137         (?:watch|feature|nightly-news)/[^/]+/(?P<title>.+))
 138         '''
 139
 140     _TESTS = [
 141         {
 142             'url': 'http://www.nbcnews.com/video/nbc-news/52753292',
 143             'md5': '47abaac93c6eaf9ad37ee6c4463a5179',
 144             'info_dict': {
 145                 'id': '52753292',
 146                 'ext': 'flv',
 147                 'title': 'Crew emerges after four-month Mars food study',
 148                 'description': 'md5:24e632ffac72b35f8b67a12d1b6ddfc1',
 149             },
 150         },
 151         {
 152             'url': 'http://www.nbcnews.com/feature/edward-snowden-interview/how-twitter-reacted-snowden-interview-n117236',
 153             'md5': 'b2421750c9f260783721d898f4c42063',
 154             'info_dict': {
 155                 'id': 'I1wpAI_zmhsQ',
 156                 'ext': 'mp4',
 157                 'title': 'How Twitter Reacted To The Snowden Interview',
 158                 'description': 'md5:65a0bd5d76fe114f3c2727aa3a81fe64',
 159             },
 160             'add_ie': ['ThePlatform'],
 161         },
 162         {
 163             'url': 'http://www.nbcnews.com/feature/dateline-full-episodes/full-episode-family-business-n285156',
 164             'md5': 'fdbf39ab73a72df5896b6234ff98518a',
 165             'info_dict': {
 166                 'id': 'Wjf9EDR3A_60',
 167                 'ext': 'mp4',
 168                 'title': 'FULL EPISODE: Family Business',
 169                 'description': 'md5:757988edbaae9d7be1d585eb5d55cc04',
 170             },
 171         },
 172         {
 173             'url': 'http://www.nbcnews.com/nightly-news/video/nightly-news-with-brian-williams-full-broadcast-february-4-394064451844',
 174             'md5': 'b5dda8cddd8650baa0dcb616dd2cf60d',
 175             'info_dict': {
 176                 'id': 'sekXqyTVnmN3',
 177                 'ext': 'mp4',
 178                 'title': 'Nightly News with Brian Williams Full Broadcast (February 4)',
 179                 'description': 'md5:1c10c1eccbe84a26e5debb4381e2d3c5',
 180             },
 181         },
 182         {
 183             'url': 'http://www.nbcnews.com/watch/dateline/full-episode--deadly-betrayal-386250819952',
 184             'only_matching': True,
 185         },
 186     ]
 187
 188     def _real_extract(self, url):
 189         mobj = re.match(self._VALID_URL, url)
 190         video_id = mobj.group('id')
 191         if video_id is not None:
 192             all_info = self._download_xml('http://www.nbcnews.com/id/%s/displaymode/1219' % video_id, video_id)
 193             info = all_info.find('video')
 194
 195             return {
 196                 'id': video_id,
 197                 'title': info.find('headline').text,
 198                 'ext': 'flv',
 199                 'url': find_xpath_attr(info, 'media', 'type', 'flashVideo').text,
 200                 'description': info.find('caption').text,
 201                 'thumbnail': find_xpath_attr(info, 'media', 'type', 'thumbnail').text,
 202             }
 203         else:
 204             # "feature" and "nightly-news" pages use theplatform.com
 205             title = mobj.group('title')
 206             webpage = self._download_webpage(url, title)
 207             bootstrap_json = self._search_regex(
 208                 r'var\s+(?:bootstrapJson|playlistData)\s*=\s*({.+});?\s*$',
 209                 webpage, 'bootstrap json', flags=re.MULTILINE)
 210             bootstrap = self._parse_json(bootstrap_json, video_id)
 211             info = bootstrap['results'][0]['video']
 212             mpxid = info['mpxId']
 213
 214             base_urls = [
 215                 info['fallbackPlaylistUrl'],
 216                 info['associatedPlaylistUrl'],
 217             ]
 218
 219             for base_url in base_urls:
 220                 if not base_url:
 221                     continue
 222                 playlist_url = base_url + '?form=MPXNBCNewsAPI'
 223
 224                 try:
 225                     all_videos = self._download_json(playlist_url, title)
 226                 except ExtractorError as ee:
 227                     if isinstance(ee.cause, compat_HTTPError):
 228                         continue
 229                     raise
 230
 231                 if not all_videos or 'videos' not in all_videos:
 232                     continue
 233
 234                 try:
 235                     info = next(v for v in all_videos['videos'] if v['mpxId'] == mpxid)
 236                     break
 237                 except StopIteration:
 238                     continue
 239
 240             if info is None:
 241                 raise ExtractorError('Could not find video in playlists')
 242
 243             return {
 244                 '_type': 'url',
 245                 # We get the best quality video
 246                 'url': info['videoAssets'][-1]['publicUrl'],
 247                 'ie_key': 'ThePlatform',
 248             }
 249
 250
 251 class MSNBCIE(InfoExtractor):
 252     # https URLs redirect to corresponding http ones
 253     _VALID_URL = r'http://www\.msnbc\.com/[^/]+/watch/(?P<id>[^/]+)'
 254     _TEST = {
 255         'url': 'http://www.msnbc.com/all-in-with-chris-hayes/watch/the-chaotic-gop-immigration-vote-314487875924',
 256         'md5': '6d236bf4f3dddc226633ce6e2c3f814d',
 257         'info_dict': {
 258             'id': 'n_hayes_Aimm_140801_272214',
 259             'ext': 'mp4',
 260             'title': 'The chaotic GOP immigration vote',
 261             '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.',
 262             'thumbnail': 're:^https?://.*\.jpg$',
 263             'timestamp': 1406937606,
 264             'upload_date': '20140802',
 265             'categories': ['MSNBC/Topics/Franchise/Best of last night', 'MSNBC/Topics/General/Congress'],
 266         },
 267     }
 268
 269     def _real_extract(self, url):
 270         video_id = self._match_id(url)
 271         webpage = self._download_webpage(url, video_id)
 272         embed_url = self._html_search_meta('embedURL', webpage)
 273         return self.url_result(embed_url)