_ 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 .theplatform import ThePlatformIE
   7 from ..utils import (
   8     find_xpath_attr,
   9     lowercase_escape,
  10     smuggle_url,
  11     unescapeHTML,
  12     update_url_query,
  13     int_or_none,
  14     HEADRequest,
  15     parse_iso8601,
  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             'info_dict': {
  26                 'id': '112966',
  27                 'ext': 'mp4',
  28                 'title': 'Jimmy Fallon Surprises Fans at Ben & Jerry\'s',
  29                 'description': 'Jimmy gives out free scoops of his new "Tonight Dough" ice cream flavor by surprising customers at the Ben & Jerry\'s scoop shop.',
  30                 'timestamp': 1424246400,
  31                 'upload_date': '20150218',
  32                 'uploader': 'NBCU-COM',
  33             },
  34             'params': {
  35                 # m3u8 download
  36                 'skip_download': True,
  37             },
  38         },
  39         {
  40             'url': 'http://www.nbc.com/the-tonight-show/episodes/176',
  41             'info_dict': {
  42                 'id': '176',
  43                 'ext': 'flv',
  44                 'title': 'Ricky Gervais, Steven Van Zandt, ILoveMakonnen',
  45                 'description': 'A brand new episode of The Tonight Show welcomes Ricky Gervais, Steven Van Zandt and ILoveMakonnen.',
  46             },
  47             'skip': '404 Not Found',
  48         },
  49         {
  50             'url': 'http://www.nbc.com/saturday-night-live/video/star-wars-teaser/2832821',
  51             'info_dict': {
  52                 'id': '2832821',
  53                 'ext': 'mp4',
  54                 'title': 'Star Wars Teaser',
  55                 'description': 'md5:0b40f9cbde5b671a7ff62fceccc4f442',
  56                 'timestamp': 1417852800,
  57                 'upload_date': '20141206',
  58                 'uploader': 'NBCU-COM',
  59             },
  60             'params': {
  61                 # m3u8 download
  62                 'skip_download': True,
  63             },
  64             'skip': 'Only works from US',
  65         },
  66         {
  67             # This video has expired but with an escaped embedURL
  68             'url': 'http://www.nbc.com/parenthood/episode-guide/season-5/just-like-at-home/515',
  69             'only_matching': True,
  70         }
  71     ]
  72
  73     def _real_extract(self, url):
  74         video_id = self._match_id(url)
  75         webpage = self._download_webpage(url, video_id)
  76         theplatform_url = unescapeHTML(lowercase_escape(self._html_search_regex(
  77             [
  78                 r'(?:class="video-player video-player-full" data-mpx-url|class="player" src)="(.*?)"',
  79                 r'<iframe[^>]+src="((?:https?:)?//player\.theplatform\.com/[^"]+)"',
  80                 r'"embedURL"\s*:\s*"([^"]+)"'
  81             ],
  82             webpage, 'theplatform url').replace('_no_endcard', '').replace('\\/', '/')))
  83         if theplatform_url.startswith('//'):
  84             theplatform_url = 'http:' + theplatform_url
  85         return {
  86             '_type': 'url_transparent',
  87             'ie_key': 'ThePlatform',
  88             'url': smuggle_url(theplatform_url, {'source_url': url}),
  89             'id': video_id,
  90         }
  91
  92
  93 class NBCSportsVPlayerIE(InfoExtractor):
  94     _VALID_URL = r'https?://vplayer\.nbcsports\.com/(?:[^/]+/)+(?P<id>[0-9a-zA-Z_]+)'
  95
  96     _TESTS = [{
  97         'url': 'https://vplayer.nbcsports.com/p/BxmELC/nbcsports_share/select/9CsDKds0kvHI',
  98         'info_dict': {
  99             'id': '9CsDKds0kvHI',
 100             'ext': 'flv',
 101             'description': 'md5:df390f70a9ba7c95ff1daace988f0d8d',
 102             'title': 'Tyler Kalinoski hits buzzer-beater to lift Davidson',
 103             'timestamp': 1426270238,
 104             'upload_date': '20150313',
 105             'uploader': 'NBCU-SPORTS',
 106         }
 107     }, {
 108         'url': 'http://vplayer.nbcsports.com/p/BxmELC/nbc_embedshare/select/_hqLjQ95yx8Z',
 109         'only_matching': True,
 110     }]
 111
 112     @staticmethod
 113     def _extract_url(webpage):
 114         iframe_m = re.search(
 115             r'<iframe[^>]+src="(?P<url>https?://vplayer\.nbcsports\.com/[^"]+)"', webpage)
 116         if iframe_m:
 117             return iframe_m.group('url')
 118
 119     def _real_extract(self, url):
 120         video_id = self._match_id(url)
 121         webpage = self._download_webpage(url, video_id)
 122         theplatform_url = self._og_search_video_url(webpage)
 123         return self.url_result(theplatform_url, 'ThePlatform')
 124
 125
 126 class NBCSportsIE(InfoExtractor):
 127     # Does not include https because its certificate is invalid
 128     _VALID_URL = r'https?://www\.nbcsports\.com//?(?:[^/]+/)+(?P<id>[0-9a-z-]+)'
 129
 130     _TEST = {
 131         'url': 'http://www.nbcsports.com//college-basketball/ncaab/tom-izzo-michigan-st-has-so-much-respect-duke',
 132         'info_dict': {
 133             'id': 'PHJSaFWbrTY9',
 134             'ext': 'flv',
 135             'title': 'Tom Izzo, Michigan St. has \'so much respect\' for Duke',
 136             'description': 'md5:ecb459c9d59e0766ac9c7d5d0eda8113',
 137             'uploader': 'NBCU-SPORTS',
 138             'upload_date': '20150330',
 139             'timestamp': 1427726529,
 140         }
 141     }
 142
 143     def _real_extract(self, url):
 144         video_id = self._match_id(url)
 145         webpage = self._download_webpage(url, video_id)
 146         return self.url_result(
 147             NBCSportsVPlayerIE._extract_url(webpage), 'NBCSportsVPlayer')
 148
 149
 150 class CSNNEIE(InfoExtractor):
 151     _VALID_URL = r'https?://www\.csnne\.com/video/(?P<id>[0-9a-z-]+)'
 152
 153     _TEST = {
 154         'url': 'http://www.csnne.com/video/snc-evening-update-wright-named-red-sox-no-5-starter',
 155         'info_dict': {
 156             'id': 'yvBLLUgQ8WU0',
 157             'ext': 'mp4',
 158             'title': 'SNC evening update: Wright named Red Sox\' No. 5 starter.',
 159             'description': 'md5:1753cfee40d9352b19b4c9b3e589b9e3',
 160             'timestamp': 1459369979,
 161             'upload_date': '20160330',
 162             'uploader': 'NBCU-SPORTS',
 163         }
 164     }
 165
 166     def _real_extract(self, url):
 167         display_id = self._match_id(url)
 168         webpage = self._download_webpage(url, display_id)
 169         return {
 170             '_type': 'url_transparent',
 171             'ie_key': 'ThePlatform',
 172             'url': self._html_search_meta('twitter:player:stream', webpage),
 173             'display_id': display_id,
 174         }
 175
 176
 177 class NBCNewsIE(ThePlatformIE):
 178     _VALID_URL = r'''(?x)https?://(?:www\.)?(?:nbcnews|today)\.com/
 179         (?:video/.+?/(?P<id>\d+)|
 180         ([^/]+/)*(?P<display_id>[^/?]+))
 181         '''
 182
 183     _TESTS = [
 184         {
 185             'url': 'http://www.nbcnews.com/video/nbc-news/52753292',
 186             'md5': '47abaac93c6eaf9ad37ee6c4463a5179',
 187             'info_dict': {
 188                 'id': '52753292',
 189                 'ext': 'flv',
 190                 'title': 'Crew emerges after four-month Mars food study',
 191                 'description': 'md5:24e632ffac72b35f8b67a12d1b6ddfc1',
 192             },
 193         },
 194         {
 195             'url': 'http://www.nbcnews.com/watch/nbcnews-com/how-twitter-reacted-to-the-snowden-interview-269389891880',
 196             'md5': 'af1adfa51312291a017720403826bb64',
 197             'info_dict': {
 198                 'id': '269389891880',
 199                 'ext': 'mp4',
 200                 'title': 'How Twitter Reacted To The Snowden Interview',
 201                 'description': 'md5:65a0bd5d76fe114f3c2727aa3a81fe64',
 202             },
 203         },
 204         {
 205             'url': 'http://www.nbcnews.com/feature/dateline-full-episodes/full-episode-family-business-n285156',
 206             'md5': 'fdbf39ab73a72df5896b6234ff98518a',
 207             'info_dict': {
 208                 'id': 'Wjf9EDR3A_60',
 209                 'ext': 'mp4',
 210                 'title': 'FULL EPISODE: Family Business',
 211                 'description': 'md5:757988edbaae9d7be1d585eb5d55cc04',
 212             },
 213             'skip': 'This page is unavailable.',
 214         },
 215         {
 216             'url': 'http://www.nbcnews.com/nightly-news/video/nightly-news-with-brian-williams-full-broadcast-february-4-394064451844',
 217             'md5': '73135a2e0ef819107bbb55a5a9b2a802',
 218             'info_dict': {
 219                 'id': '394064451844',
 220                 'ext': 'mp4',
 221                 'title': 'Nightly News with Brian Williams Full Broadcast (February 4)',
 222                 'description': 'md5:1c10c1eccbe84a26e5debb4381e2d3c5',
 223             },
 224         },
 225         {
 226             'url': 'http://www.nbcnews.com/business/autos/volkswagen-11-million-vehicles-could-have-suspect-software-emissions-scandal-n431456',
 227             'md5': 'a49e173825e5fcd15c13fc297fced39d',
 228             'info_dict': {
 229                 'id': '529953347624',
 230                 'ext': 'mp4',
 231                 'title': 'Volkswagen U.S. Chief: We \'Totally Screwed Up\'',
 232                 'description': 'md5:d22d1281a24f22ea0880741bb4dd6301',
 233             },
 234             'expected_warnings': ['http-6000 is not available']
 235         },
 236         {
 237             'url': 'http://www.today.com/video/see-the-aurora-borealis-from-space-in-stunning-new-nasa-video-669831235788',
 238             'md5': '118d7ca3f0bea6534f119c68ef539f71',
 239             'info_dict': {
 240                 'id': '669831235788',
 241                 'ext': 'mp4',
 242                 'title': 'See the aurora borealis from space in stunning new NASA video',
 243                 'description': 'md5:74752b7358afb99939c5f8bb2d1d04b1',
 244                 'upload_date': '20160420',
 245                 'timestamp': 1461152093,
 246             },
 247         },
 248         {
 249             'url': 'http://www.nbcnews.com/watch/dateline/full-episode--deadly-betrayal-386250819952',
 250             'only_matching': True,
 251         },
 252     ]
 253
 254     def _real_extract(self, url):
 255         mobj = re.match(self._VALID_URL, url)
 256         video_id = mobj.group('id')
 257         if video_id is not None:
 258             all_info = self._download_xml('http://www.nbcnews.com/id/%s/displaymode/1219' % video_id, video_id)
 259             info = all_info.find('video')
 260
 261             return {
 262                 'id': video_id,
 263                 'title': info.find('headline').text,
 264                 'ext': 'flv',
 265                 'url': find_xpath_attr(info, 'media', 'type', 'flashVideo').text,
 266                 'description': info.find('caption').text,
 267                 'thumbnail': find_xpath_attr(info, 'media', 'type', 'thumbnail').text,
 268             }
 269         else:
 270             # "feature" and "nightly-news" pages use theplatform.com
 271             display_id = mobj.group('display_id')
 272             webpage = self._download_webpage(url, display_id)
 273             info = None
 274             bootstrap_json = self._search_regex(
 275                 r'(?m)var\s+(?:bootstrapJson|playlistData)\s*=\s*({.+});?\s*$',
 276                 webpage, 'bootstrap json', default=None)
 277             if bootstrap_json:
 278                 bootstrap = self._parse_json(bootstrap_json, display_id)
 279                 info = bootstrap['results'][0]['video']
 280             else:
 281                 player_instance_json = self._search_regex(
 282                     r'videoObj\s*:\s*({.+})', webpage, 'player instance', default=None)
 283                 if not player_instance_json:
 284                     player_instance_json = self._html_search_regex(
 285                         r'data-video="([^"]+)"', webpage, 'video json')
 286                 info = self._parse_json(player_instance_json, display_id)
 287             video_id = info['mpxId']
 288             title = info['title']
 289
 290             subtitles = {}
 291             caption_links = info.get('captionLinks')
 292             if caption_links:
 293                 for (sub_key, sub_ext) in (('smpte-tt', 'ttml'), ('web-vtt', 'vtt'), ('srt', 'srt')):
 294                     sub_url = caption_links.get(sub_key)
 295                     if sub_url:
 296                         subtitles.setdefault('en', []).append({
 297                             'url': sub_url,
 298                             'ext': sub_ext,
 299                         })
 300
 301             formats = []
 302             for video_asset in info['videoAssets']:
 303                 video_url = video_asset.get('publicUrl')
 304                 if not video_url:
 305                     continue
 306                 container = video_asset.get('format')
 307                 asset_type = video_asset.get('assetType') or ''
 308                 if container == 'ISM' or asset_type == 'FireTV-Once':
 309                     continue
 310                 elif asset_type == 'OnceURL':
 311                     tp_formats, tp_subtitles = self._extract_theplatform_smil(
 312                         video_url, video_id)
 313                     formats.extend(tp_formats)
 314                     subtitles = self._merge_subtitles(subtitles, tp_subtitles)
 315                 else:
 316                     tbr = int_or_none(video_asset.get('bitRate') or video_asset.get('bitrate'), 1000)
 317                     format_id = 'http%s' % ('-%d' % tbr if tbr else '')
 318                     video_url = update_url_query(
 319                         video_url, {'format': 'redirect'})
 320                     # resolve the url so that we can check availability and detect the correct extension
 321                     head = self._request_webpage(
 322                         HEADRequest(video_url), video_id,
 323                         'Checking %s url' % format_id,
 324                         '%s is not available' % format_id,
 325                         fatal=False)
 326                     if head:
 327                         video_url = head.geturl()
 328                         formats.append({
 329                             'format_id': format_id,
 330                             'url': video_url,
 331                             'width': int_or_none(video_asset.get('width')),
 332                             'height': int_or_none(video_asset.get('height')),
 333                             'tbr': tbr,
 334                             'container': video_asset.get('format'),
 335                         })
 336             self._sort_formats(formats)
 337
 338             return {
 339                 'id': video_id,
 340                 'title': title,
 341                 'description': info.get('description'),
 342                 'thumbnail': info.get('thumbnail'),
 343                 'duration': int_or_none(info.get('duration')),
 344                 'timestamp': parse_iso8601(info.get('pubDate') or info.get('pub_date')),
 345                 'formats': formats,
 346                 'subtitles': subtitles,
 347             }
 348
 349
 350 class MSNBCIE(InfoExtractor):
 351     # https URLs redirect to corresponding http ones
 352     _VALID_URL = r'https?://www\.msnbc\.com/[^/]+/watch/(?P<id>[^/]+)'
 353     _TEST = {
 354         'url': 'http://www.msnbc.com/all-in-with-chris-hayes/watch/the-chaotic-gop-immigration-vote-314487875924',
 355         'md5': '6d236bf4f3dddc226633ce6e2c3f814d',
 356         'info_dict': {
 357             'id': 'n_hayes_Aimm_140801_272214',
 358             'ext': 'mp4',
 359             'title': 'The chaotic GOP immigration vote',
 360             '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.',
 361             'thumbnail': 're:^https?://.*\.jpg$',
 362             'timestamp': 1406937606,
 363             'upload_date': '20140802',
 364             'uploader': 'NBCU-NEWS',
 365             'categories': ['MSNBC/Topics/Franchise/Best of last night', 'MSNBC/Topics/General/Congress'],
 366         },
 367     }
 368
 369     def _real_extract(self, url):
 370         video_id = self._match_id(url)
 371         webpage = self._download_webpage(url, video_id)
 372         embed_url = self._html_search_meta('embedURL', webpage)
 373         return self.url_result(embed_url)