youtube_dlc/extractor/nbc.py

   1 from __future__ import unicode_literals
   2
   3 import base64
   4 import json
   5 import re
   6
   7 from .common import InfoExtractor
   8 from .theplatform import ThePlatformIE
   9 from .adobepass import AdobePassIE
  10 from ..compat import compat_urllib_parse_unquote
  11 from ..utils import (
  12     int_or_none,
  13     parse_duration,
  14     smuggle_url,
  15     try_get,
  16     unified_timestamp,
  17     update_url_query,
  18 )
  19
  20
  21 class NBCIE(AdobePassIE):
  22     _VALID_URL = r'https?(?P<permalink>://(?:www\.)?nbc\.com/(?:classic-tv/)?[^/]+/video/[^/]+/(?P<id>n?\d+))'
  23
  24     _TESTS = [
  25         {
  26             'url': 'http://www.nbc.com/the-tonight-show/video/jimmy-fallon-surprises-fans-at-ben-jerrys/2848237',
  27             'info_dict': {
  28                 'id': '2848237',
  29                 'ext': 'mp4',
  30                 'title': 'Jimmy Fallon Surprises Fans at Ben & Jerry\'s',
  31                 'description': 'Jimmy gives out free scoops of his new "Tonight Dough" ice cream flavor by surprising customers at the Ben & Jerry\'s scoop shop.',
  32                 'timestamp': 1424246400,
  33                 'upload_date': '20150218',
  34                 'uploader': 'NBCU-COM',
  35             },
  36             'params': {
  37                 # m3u8 download
  38                 'skip_download': True,
  39             },
  40         },
  41         {
  42             'url': 'http://www.nbc.com/saturday-night-live/video/star-wars-teaser/2832821',
  43             'info_dict': {
  44                 'id': '2832821',
  45                 'ext': 'mp4',
  46                 'title': 'Star Wars Teaser',
  47                 'description': 'md5:0b40f9cbde5b671a7ff62fceccc4f442',
  48                 'timestamp': 1417852800,
  49                 'upload_date': '20141206',
  50                 'uploader': 'NBCU-COM',
  51             },
  52             'params': {
  53                 # m3u8 download
  54                 'skip_download': True,
  55             },
  56             'skip': 'Only works from US',
  57         },
  58         {
  59             # HLS streams requires the 'hdnea3' cookie
  60             'url': 'http://www.nbc.com/Kings/video/goliath/n1806',
  61             'info_dict': {
  62                 'id': '101528f5a9e8127b107e98c5e6ce4638',
  63                 'ext': 'mp4',
  64                 'title': 'Goliath',
  65                 'description': 'When an unknown soldier saves the life of the King\'s son in battle, he\'s thrust into the limelight and politics of the kingdom.',
  66                 'timestamp': 1237100400,
  67                 'upload_date': '20090315',
  68                 'uploader': 'NBCU-COM',
  69             },
  70             'params': {
  71                 'skip_download': True,
  72             },
  73             'skip': 'Only works from US',
  74         },
  75         {
  76             'url': 'https://www.nbc.com/classic-tv/charles-in-charge/video/charles-in-charge-pilot/n3310',
  77             'only_matching': True,
  78         },
  79         {
  80             # Percent escaped url
  81             'url': 'https://www.nbc.com/up-all-night/video/day-after-valentine%27s-day/n2189',
  82             'only_matching': True,
  83         }
  84     ]
  85
  86     def _real_extract(self, url):
  87         permalink, video_id = re.match(self._VALID_URL, url).groups()
  88         permalink = 'http' + compat_urllib_parse_unquote(permalink)
  89         video_data = self._download_json(
  90             'https://friendship.nbc.co/v2/graphql', video_id, query={
  91                 'query': '''query bonanzaPage(
  92   $app: NBCUBrands! = nbc
  93   $name: String!
  94   $oneApp: Boolean
  95   $platform: SupportedPlatforms! = web
  96   $type: EntityPageType! = VIDEO
  97   $userId: String!
  98 ) {
  99   bonanzaPage(
 100     app: $app
 101     name: $name
 102     oneApp: $oneApp
 103     platform: $platform
 104     type: $type
 105     userId: $userId
 106   ) {
 107     metadata {
 108       ... on VideoPageData {
 109         description
 110         episodeNumber
 111         keywords
 112         locked
 113         mpxAccountId
 114         mpxGuid
 115         rating
 116         resourceId
 117         seasonNumber
 118         secondaryTitle
 119         seriesShortTitle
 120       }
 121     }
 122   }
 123 }''',
 124                 'variables': json.dumps({
 125                     'name': permalink,
 126                     'oneApp': True,
 127                     'userId': '0',
 128                 }),
 129             })['data']['bonanzaPage']['metadata']
 130         query = {
 131             'mbr': 'true',
 132             'manifest': 'm3u',
 133         }
 134         video_id = video_data['mpxGuid']
 135         title = video_data['secondaryTitle']
 136         if video_data.get('locked'):
 137             resource = self._get_mvpd_resource(
 138                 video_data.get('resourceId') or 'nbcentertainment',
 139                 title, video_id, video_data.get('rating'))
 140             query['auth'] = self._extract_mvpd_auth(
 141                 url, video_id, 'nbcentertainment', resource)
 142         theplatform_url = smuggle_url(update_url_query(
 143             'http://link.theplatform.com/s/NnzsPC/media/guid/%s/%s' % (video_data.get('mpxAccountId') or '2410887629', video_id),
 144             query), {'force_smil_url': True})
 145         return {
 146             '_type': 'url_transparent',
 147             'id': video_id,
 148             'title': title,
 149             'url': theplatform_url,
 150             'description': video_data.get('description'),
 151             'tags': video_data.get('keywords'),
 152             'season_number': int_or_none(video_data.get('seasonNumber')),
 153             'episode_number': int_or_none(video_data.get('episodeNumber')),
 154             'episode': title,
 155             'series': video_data.get('seriesShortTitle'),
 156             'ie_key': 'ThePlatform',
 157         }
 158
 159
 160 class NBCSportsVPlayerIE(InfoExtractor):
 161     _VALID_URL = r'https?://vplayer\.nbcsports\.com/(?:[^/]+/)+(?P<id>[0-9a-zA-Z_]+)'
 162
 163     _TESTS = [{
 164         'url': 'https://vplayer.nbcsports.com/p/BxmELC/nbcsports_embed/select/9CsDKds0kvHI',
 165         'info_dict': {
 166             'id': '9CsDKds0kvHI',
 167             'ext': 'mp4',
 168             'description': 'md5:df390f70a9ba7c95ff1daace988f0d8d',
 169             'title': 'Tyler Kalinoski hits buzzer-beater to lift Davidson',
 170             'timestamp': 1426270238,
 171             'upload_date': '20150313',
 172             'uploader': 'NBCU-SPORTS',
 173         }
 174     }, {
 175         'url': 'https://vplayer.nbcsports.com/p/BxmELC/nbcsports_embed/select/media/_hqLjQ95yx8Z',
 176         'only_matching': True,
 177     }]
 178
 179     @staticmethod
 180     def _extract_url(webpage):
 181         iframe_m = re.search(
 182             r'<iframe[^>]+src="(?P<url>https?://vplayer\.nbcsports\.com/[^"]+)"', webpage)
 183         if iframe_m:
 184             return iframe_m.group('url')
 185
 186     def _real_extract(self, url):
 187         video_id = self._match_id(url)
 188         webpage = self._download_webpage(url, video_id)
 189         theplatform_url = self._og_search_video_url(webpage).replace(
 190             'vplayer.nbcsports.com', 'player.theplatform.com')
 191         return self.url_result(theplatform_url, 'ThePlatform')
 192
 193
 194 class NBCSportsIE(InfoExtractor):
 195     # Does not include https because its certificate is invalid
 196     _VALID_URL = r'https?://(?:www\.)?nbcsports\.com//?(?:[^/]+/)+(?P<id>[0-9a-z-]+)'
 197
 198     _TEST = {
 199         'url': 'http://www.nbcsports.com//college-basketball/ncaab/tom-izzo-michigan-st-has-so-much-respect-duke',
 200         'info_dict': {
 201             'id': 'PHJSaFWbrTY9',
 202             'ext': 'flv',
 203             'title': 'Tom Izzo, Michigan St. has \'so much respect\' for Duke',
 204             'description': 'md5:ecb459c9d59e0766ac9c7d5d0eda8113',
 205             'uploader': 'NBCU-SPORTS',
 206             'upload_date': '20150330',
 207             'timestamp': 1427726529,
 208         }
 209     }
 210
 211     def _real_extract(self, url):
 212         video_id = self._match_id(url)
 213         webpage = self._download_webpage(url, video_id)
 214         return self.url_result(
 215             NBCSportsVPlayerIE._extract_url(webpage), 'NBCSportsVPlayer')
 216
 217
 218 class NBCSportsStreamIE(AdobePassIE):
 219     _VALID_URL = r'https?://stream\.nbcsports\.com/.+?\bpid=(?P<id>\d+)'
 220     _TEST = {
 221         'url': 'http://stream.nbcsports.com/nbcsn/generic?pid=206559',
 222         'info_dict': {
 223             'id': '206559',
 224             'ext': 'mp4',
 225             'title': 'Amgen Tour of California Women\'s Recap',
 226             'description': 'md5:66520066b3b5281ada7698d0ea2aa894',
 227         },
 228         'params': {
 229             # m3u8 download
 230             'skip_download': True,
 231         },
 232         'skip': 'Requires Adobe Pass Authentication',
 233     }
 234
 235     def _real_extract(self, url):
 236         video_id = self._match_id(url)
 237         live_source = self._download_json(
 238             'http://stream.nbcsports.com/data/live_sources_%s.json' % video_id,
 239             video_id)
 240         video_source = live_source['videoSources'][0]
 241         title = video_source['title']
 242         source_url = None
 243         for k in ('source', 'msl4source', 'iossource', 'hlsv4'):
 244             sk = k + 'Url'
 245             source_url = video_source.get(sk) or video_source.get(sk + 'Alt')
 246             if source_url:
 247                 break
 248         else:
 249             source_url = video_source['ottStreamUrl']
 250         is_live = video_source.get('type') == 'live' or video_source.get('status') == 'Live'
 251         resource = self._get_mvpd_resource('nbcsports', title, video_id, '')
 252         token = self._extract_mvpd_auth(url, video_id, 'nbcsports', resource)
 253         tokenized_url = self._download_json(
 254             'https://token.playmakerservices.com/cdn',
 255             video_id, data=json.dumps({
 256                 'requestorId': 'nbcsports',
 257                 'pid': video_id,
 258                 'application': 'NBCSports',
 259                 'version': 'v1',
 260                 'platform': 'desktop',
 261                 'cdn': 'akamai',
 262                 'url': video_source['sourceUrl'],
 263                 'token': base64.b64encode(token.encode()).decode(),
 264                 'resourceId': base64.b64encode(resource.encode()).decode(),
 265             }).encode())['tokenizedUrl']
 266         formats = self._extract_m3u8_formats(tokenized_url, video_id, 'mp4')
 267         self._sort_formats(formats)
 268         return {
 269             'id': video_id,
 270             'title': self._live_title(title) if is_live else title,
 271             'description': live_source.get('description'),
 272             'formats': formats,
 273             'is_live': is_live,
 274         }
 275
 276
 277 class CSNNEIE(InfoExtractor):
 278     _VALID_URL = r'https?://(?:www\.)?csnne\.com/video/(?P<id>[0-9a-z-]+)'
 279
 280     _TEST = {
 281         'url': 'http://www.csnne.com/video/snc-evening-update-wright-named-red-sox-no-5-starter',
 282         'info_dict': {
 283             'id': 'yvBLLUgQ8WU0',
 284             'ext': 'mp4',
 285             'title': 'SNC evening update: Wright named Red Sox\' No. 5 starter.',
 286             'description': 'md5:1753cfee40d9352b19b4c9b3e589b9e3',
 287             'timestamp': 1459369979,
 288             'upload_date': '20160330',
 289             'uploader': 'NBCU-SPORTS',
 290         }
 291     }
 292
 293     def _real_extract(self, url):
 294         display_id = self._match_id(url)
 295         webpage = self._download_webpage(url, display_id)
 296         return {
 297             '_type': 'url_transparent',
 298             'ie_key': 'ThePlatform',
 299             'url': self._html_search_meta('twitter:player:stream', webpage),
 300             'display_id': display_id,
 301         }
 302
 303
 304 class NBCNewsIE(ThePlatformIE):
 305     _VALID_URL = r'(?x)https?://(?:www\.)?(?:nbcnews|today|msnbc)\.com/([^/]+/)*(?:.*-)?(?P<id>[^/?]+)'
 306
 307     _TESTS = [
 308         {
 309             'url': 'http://www.nbcnews.com/watch/nbcnews-com/how-twitter-reacted-to-the-snowden-interview-269389891880',
 310             'md5': 'cf4bc9e6ce0130f00f545d80ecedd4bf',
 311             'info_dict': {
 312                 'id': '269389891880',
 313                 'ext': 'mp4',
 314                 'title': 'How Twitter Reacted To The Snowden Interview',
 315                 'description': 'md5:65a0bd5d76fe114f3c2727aa3a81fe64',
 316                 'timestamp': 1401363060,
 317                 'upload_date': '20140529',
 318             },
 319         },
 320         {
 321             'url': 'http://www.nbcnews.com/feature/dateline-full-episodes/full-episode-family-business-n285156',
 322             'md5': 'fdbf39ab73a72df5896b6234ff98518a',
 323             'info_dict': {
 324                 'id': '529953347624',
 325                 'ext': 'mp4',
 326                 'title': 'FULL EPISODE: Family Business',
 327                 'description': 'md5:757988edbaae9d7be1d585eb5d55cc04',
 328             },
 329             'skip': 'This page is unavailable.',
 330         },
 331         {
 332             'url': 'http://www.nbcnews.com/nightly-news/video/nightly-news-with-brian-williams-full-broadcast-february-4-394064451844',
 333             'md5': '8eb831eca25bfa7d25ddd83e85946548',
 334             'info_dict': {
 335                 'id': '394064451844',
 336                 'ext': 'mp4',
 337                 'title': 'Nightly News with Brian Williams Full Broadcast (February 4)',
 338                 'description': 'md5:1c10c1eccbe84a26e5debb4381e2d3c5',
 339                 'timestamp': 1423104900,
 340                 'upload_date': '20150205',
 341             },
 342         },
 343         {
 344             'url': 'http://www.nbcnews.com/business/autos/volkswagen-11-million-vehicles-could-have-suspect-software-emissions-scandal-n431456',
 345             'md5': '4a8c4cec9e1ded51060bdda36ff0a5c0',
 346             'info_dict': {
 347                 'id': 'n431456',
 348                 'ext': 'mp4',
 349                 'title': "Volkswagen U.S. Chief:  We 'Totally Screwed Up'",
 350                 'description': 'md5:d22d1281a24f22ea0880741bb4dd6301',
 351                 'upload_date': '20150922',
 352                 'timestamp': 1442917800,
 353             },
 354         },
 355         {
 356             'url': 'http://www.today.com/video/see-the-aurora-borealis-from-space-in-stunning-new-nasa-video-669831235788',
 357             'md5': '118d7ca3f0bea6534f119c68ef539f71',
 358             'info_dict': {
 359                 'id': '669831235788',
 360                 'ext': 'mp4',
 361                 'title': 'See the aurora borealis from space in stunning new NASA video',
 362                 'description': 'md5:74752b7358afb99939c5f8bb2d1d04b1',
 363                 'upload_date': '20160420',
 364                 'timestamp': 1461152093,
 365             },
 366         },
 367         {
 368             'url': 'http://www.msnbc.com/all-in-with-chris-hayes/watch/the-chaotic-gop-immigration-vote-314487875924',
 369             'md5': '6d236bf4f3dddc226633ce6e2c3f814d',
 370             'info_dict': {
 371                 'id': '314487875924',
 372                 'ext': 'mp4',
 373                 'title': 'The chaotic GOP immigration vote',
 374                 '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.',
 375                 'thumbnail': r're:^https?://.*\.jpg$',
 376                 'timestamp': 1406937606,
 377                 'upload_date': '20140802',
 378             },
 379         },
 380         {
 381             'url': 'http://www.nbcnews.com/watch/dateline/full-episode--deadly-betrayal-386250819952',
 382             'only_matching': True,
 383         },
 384         {
 385             # From http://www.vulture.com/2016/06/letterman-couldnt-care-less-about-late-night.html
 386             'url': 'http://www.nbcnews.com/widget/video-embed/701714499682',
 387             'only_matching': True,
 388         },
 389     ]
 390
 391     def _real_extract(self, url):
 392         video_id = self._match_id(url)
 393         webpage = self._download_webpage(url, video_id)
 394
 395         data = self._parse_json(self._search_regex(
 396             r'<script[^>]+id="__NEXT_DATA__"[^>]*>({.+?})</script>',
 397             webpage, 'bootstrap json'), video_id)['props']['initialState']
 398         video_data = try_get(data, lambda x: x['video']['current'], dict)
 399         if not video_data:
 400             video_data = data['article']['content'][0]['primaryMedia']['video']
 401         title = video_data['headline']['primary']
 402
 403         formats = []
 404         for va in video_data.get('videoAssets', []):
 405             public_url = va.get('publicUrl')
 406             if not public_url:
 407                 continue
 408             if '://link.theplatform.com/' in public_url:
 409                 public_url = update_url_query(public_url, {'format': 'redirect'})
 410             format_id = va.get('format')
 411             if format_id == 'M3U':
 412                 formats.extend(self._extract_m3u8_formats(
 413                     public_url, video_id, 'mp4', 'm3u8_native',
 414                     m3u8_id=format_id, fatal=False))
 415                 continue
 416             tbr = int_or_none(va.get('bitrate'), 1000)
 417             if tbr:
 418                 format_id += '-%d' % tbr
 419             formats.append({
 420                 'format_id': format_id,
 421                 'url': public_url,
 422                 'width': int_or_none(va.get('width')),
 423                 'height': int_or_none(va.get('height')),
 424                 'tbr': tbr,
 425                 'ext': 'mp4',
 426             })
 427         self._sort_formats(formats)
 428
 429         subtitles = {}
 430         closed_captioning = video_data.get('closedCaptioning')
 431         if closed_captioning:
 432             for cc_url in closed_captioning.values():
 433                 if not cc_url:
 434                     continue
 435                 subtitles.setdefault('en', []).append({
 436                     'url': cc_url,
 437                 })
 438
 439         return {
 440             'id': video_id,
 441             'title': title,
 442             'description': try_get(video_data, lambda x: x['description']['primary']),
 443             'thumbnail': try_get(video_data, lambda x: x['primaryImage']['url']['primary']),
 444             'duration': parse_duration(video_data.get('duration')),
 445             'timestamp': unified_timestamp(video_data.get('datePublished')),
 446             'formats': formats,
 447             'subtitles': subtitles,
 448         }
 449
 450
 451 class NBCOlympicsIE(InfoExtractor):
 452     IE_NAME = 'nbcolympics'
 453     _VALID_URL = r'https?://www\.nbcolympics\.com/video/(?P<id>[a-z-]+)'
 454
 455     _TEST = {
 456         # Geo-restricted to US
 457         'url': 'http://www.nbcolympics.com/video/justin-roses-son-leo-was-tears-after-his-dad-won-gold',
 458         'md5': '54fecf846d05429fbaa18af557ee523a',
 459         'info_dict': {
 460             'id': 'WjTBzDXx5AUq',
 461             'display_id': 'justin-roses-son-leo-was-tears-after-his-dad-won-gold',
 462             'ext': 'mp4',
 463             'title': 'Rose\'s son Leo was in tears after his dad won gold',
 464             'description': 'Olympic gold medalist Justin Rose gets emotional talking to the impact his win in men\'s golf has already had on his children.',
 465             'timestamp': 1471274964,
 466             'upload_date': '20160815',
 467             'uploader': 'NBCU-SPORTS',
 468         },
 469     }
 470
 471     def _real_extract(self, url):
 472         display_id = self._match_id(url)
 473
 474         webpage = self._download_webpage(url, display_id)
 475
 476         drupal_settings = self._parse_json(self._search_regex(
 477             r'jQuery\.extend\(Drupal\.settings\s*,\s*({.+?})\);',
 478             webpage, 'drupal settings'), display_id)
 479
 480         iframe_url = drupal_settings['vod']['iframe_url']
 481         theplatform_url = iframe_url.replace(
 482             'vplayer.nbcolympics.com', 'player.theplatform.com')
 483
 484         return {
 485             '_type': 'url_transparent',
 486             'url': theplatform_url,
 487             'ie_key': ThePlatformIE.ie_key(),
 488             'display_id': display_id,
 489         }
 490
 491
 492 class NBCOlympicsStreamIE(AdobePassIE):
 493     IE_NAME = 'nbcolympics:stream'
 494     _VALID_URL = r'https?://stream\.nbcolympics\.com/(?P<id>[0-9a-z-]+)'
 495     _TEST = {
 496         'url': 'http://stream.nbcolympics.com/2018-winter-olympics-nbcsn-evening-feb-8',
 497         'info_dict': {
 498             'id': '203493',
 499             'ext': 'mp4',
 500             'title': 're:Curling, Alpine, Luge [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
 501         },
 502         'params': {
 503             # m3u8 download
 504             'skip_download': True,
 505         },
 506     }
 507     _DATA_URL_TEMPLATE = 'http://stream.nbcolympics.com/data/%s_%s.json'
 508
 509     def _real_extract(self, url):
 510         display_id = self._match_id(url)
 511         webpage = self._download_webpage(url, display_id)
 512         pid = self._search_regex(r'pid\s*=\s*(\d+);', webpage, 'pid')
 513         resource = self._search_regex(
 514             r"resource\s*=\s*'(.+)';", webpage,
 515             'resource').replace("' + pid + '", pid)
 516         event_config = self._download_json(
 517             self._DATA_URL_TEMPLATE % ('event_config', pid),
 518             pid)['eventConfig']
 519         title = self._live_title(event_config['eventTitle'])
 520         source_url = self._download_json(
 521             self._DATA_URL_TEMPLATE % ('live_sources', pid),
 522             pid)['videoSources'][0]['sourceUrl']
 523         media_token = self._extract_mvpd_auth(
 524             url, pid, event_config.get('requestorId', 'NBCOlympics'), resource)
 525         formats = self._extract_m3u8_formats(self._download_webpage(
 526             'http://sp.auth.adobe.com/tvs/v1/sign', pid, query={
 527                 'cdn': 'akamai',
 528                 'mediaToken': base64.b64encode(media_token.encode()),
 529                 'resource': base64.b64encode(resource.encode()),
 530                 'url': source_url,
 531             }), pid, 'mp4')
 532         self._sort_formats(formats)
 533
 534         return {
 535             'id': pid,
 536             'display_id': display_id,
 537             'title': title,
 538             'formats': formats,
 539             'is_live': True,
 540         }