yt_dlp/extractor/nbc.py

   1 import base64
   2 import json
   3 import re
   4
   5 from .common import InfoExtractor
   6 from .theplatform import ThePlatformIE
   7 from .adobepass import AdobePassIE
   8 from ..compat import compat_urllib_parse_unquote
   9 from ..utils import (
  10     int_or_none,
  11     parse_age_limit,
  12     parse_duration,
  13     RegexNotFoundError,
  14     smuggle_url,
  15     try_get,
  16     unified_timestamp,
  17     update_url_query,
  18 )
  19
  20
  21 class NBCIE(ThePlatformIE):
  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 = self._match_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         tp_path = 'NnzsPC/media/guid/%s/%s' % (video_data.get('mpxAccountId') or '2410887629', video_id)
 136         tpm = self._download_theplatform_metadata(tp_path, video_id)
 137         title = tpm.get('title') or video_data.get('secondaryTitle')
 138         if video_data.get('locked'):
 139             resource = self._get_mvpd_resource(
 140                 video_data.get('resourceId') or 'nbcentertainment',
 141                 title, video_id, video_data.get('rating'))
 142             query['auth'] = self._extract_mvpd_auth(
 143                 url, video_id, 'nbcentertainment', resource)
 144         theplatform_url = smuggle_url(update_url_query(
 145             'http://link.theplatform.com/s/NnzsPC/media/guid/%s/%s' % (video_data.get('mpxAccountId') or '2410887629', video_id),
 146             query), {'force_smil_url': True})
 147
 148         # Empty string or 0 can be valid values for these. So the check must be `is None`
 149         description = video_data.get('description')
 150         if description is None:
 151             description = tpm.get('description')
 152         episode_number = int_or_none(video_data.get('episodeNumber'))
 153         if episode_number is None:
 154             episode_number = int_or_none(tpm.get('nbcu$airOrder'))
 155         rating = video_data.get('rating')
 156         if rating is None:
 157             try_get(tpm, lambda x: x['ratings'][0]['rating'])
 158         season_number = int_or_none(video_data.get('seasonNumber'))
 159         if season_number is None:
 160             season_number = int_or_none(tpm.get('nbcu$seasonNumber'))
 161         series = video_data.get('seriesShortTitle')
 162         if series is None:
 163             series = tpm.get('nbcu$seriesShortTitle')
 164         tags = video_data.get('keywords')
 165         if tags is None or len(tags) == 0:
 166             tags = tpm.get('keywords')
 167
 168         return {
 169             '_type': 'url_transparent',
 170             'age_limit': parse_age_limit(rating),
 171             'description': description,
 172             'episode': title,
 173             'episode_number': episode_number,
 174             'id': video_id,
 175             'ie_key': 'ThePlatform',
 176             'season_number': season_number,
 177             'series': series,
 178             'tags': tags,
 179             'title': title,
 180             'url': theplatform_url,
 181         }
 182
 183
 184 class NBCSportsVPlayerIE(InfoExtractor):
 185     _VALID_URL_BASE = r'https?://(?:vplayer\.nbcsports\.com|(?:www\.)?nbcsports\.com/vplayer)/'
 186     _VALID_URL = _VALID_URL_BASE + r'(?:[^/]+/)+(?P<id>[0-9a-zA-Z_]+)'
 187
 188     _TESTS = [{
 189         'url': 'https://vplayer.nbcsports.com/p/BxmELC/nbcsports_embed/select/9CsDKds0kvHI',
 190         'info_dict': {
 191             'id': '9CsDKds0kvHI',
 192             'ext': 'mp4',
 193             'description': 'md5:df390f70a9ba7c95ff1daace988f0d8d',
 194             'title': 'Tyler Kalinoski hits buzzer-beater to lift Davidson',
 195             'timestamp': 1426270238,
 196             'upload_date': '20150313',
 197             'uploader': 'NBCU-SPORTS',
 198             'duration': 72.818,
 199             'chapters': [],
 200             'thumbnail': r're:^https?://.*\.jpg$'
 201         }
 202     }, {
 203         'url': 'https://vplayer.nbcsports.com/p/BxmELC/nbcsports_embed/select/media/PEgOtlNcC_y2',
 204         'only_matching': True,
 205     }, {
 206         'url': 'https://www.nbcsports.com/vplayer/p/BxmELC/nbcsports/select/PHJSaFWbrTY9?form=html&autoPlay=true',
 207         'only_matching': True,
 208     }]
 209
 210     @staticmethod
 211     def _extract_url(webpage):
 212         video_urls = re.search(
 213             r'(?:iframe[^>]+|var video|div[^>]+data-(?:mpx-)?)[sS]rc\s?=\s?"(?P<url>%s[^\"]+)' % NBCSportsVPlayerIE._VALID_URL_BASE, webpage)
 214         if video_urls:
 215             return video_urls.group('url')
 216
 217     def _real_extract(self, url):
 218         video_id = self._match_id(url)
 219         webpage = self._download_webpage(url, video_id)
 220         theplatform_url = self._html_search_regex(r'tp:releaseUrl="(.+?)"', webpage, 'url')
 221         return self.url_result(theplatform_url, 'ThePlatform')
 222
 223
 224 class NBCSportsIE(InfoExtractor):
 225     _VALID_URL = r'https?://(?:www\.)?nbcsports\.com//?(?!vplayer/)(?:[^/]+/)+(?P<id>[0-9a-z-]+)'
 226
 227     _TESTS = [{
 228         # iframe src
 229         'url': 'http://www.nbcsports.com//college-basketball/ncaab/tom-izzo-michigan-st-has-so-much-respect-duke',
 230         'info_dict': {
 231             'id': 'PHJSaFWbrTY9',
 232             'ext': 'mp4',
 233             'title': 'Tom Izzo, Michigan St. has \'so much respect\' for Duke',
 234             'description': 'md5:ecb459c9d59e0766ac9c7d5d0eda8113',
 235             'uploader': 'NBCU-SPORTS',
 236             'upload_date': '20150330',
 237             'timestamp': 1427726529,
 238             'chapters': [],
 239             'thumbnail': 'https://hdliveextra-a.akamaihd.net/HD/image_sports/NBCU_Sports_Group_-_nbcsports/253/303/izzodps.jpg',
 240             'duration': 528.395,
 241         }
 242     }, {
 243         # data-mpx-src
 244         'url': 'https://www.nbcsports.com/philadelphia/philadelphia-phillies/bruce-bochy-hector-neris-hes-idiot',
 245         'only_matching': True,
 246     }, {
 247         # data-src
 248         'url': 'https://www.nbcsports.com/boston/video/report-card-pats-secondary-no-match-josh-allen',
 249         'only_matching': True,
 250     }]
 251
 252     def _real_extract(self, url):
 253         video_id = self._match_id(url)
 254         webpage = self._download_webpage(url, video_id)
 255         return self.url_result(
 256             NBCSportsVPlayerIE._extract_url(webpage), 'NBCSportsVPlayer')
 257
 258
 259 class NBCSportsStreamIE(AdobePassIE):
 260     _VALID_URL = r'https?://stream\.nbcsports\.com/.+?\bpid=(?P<id>\d+)'
 261     _TEST = {
 262         'url': 'http://stream.nbcsports.com/nbcsn/generic?pid=206559',
 263         'info_dict': {
 264             'id': '206559',
 265             'ext': 'mp4',
 266             'title': 'Amgen Tour of California Women\'s Recap',
 267             'description': 'md5:66520066b3b5281ada7698d0ea2aa894',
 268         },
 269         'params': {
 270             # m3u8 download
 271             'skip_download': True,
 272         },
 273         'skip': 'Requires Adobe Pass Authentication',
 274     }
 275
 276     def _real_extract(self, url):
 277         video_id = self._match_id(url)
 278         live_source = self._download_json(
 279             'http://stream.nbcsports.com/data/live_sources_%s.json' % video_id,
 280             video_id)
 281         video_source = live_source['videoSources'][0]
 282         title = video_source['title']
 283         source_url = None
 284         for k in ('source', 'msl4source', 'iossource', 'hlsv4'):
 285             sk = k + 'Url'
 286             source_url = video_source.get(sk) or video_source.get(sk + 'Alt')
 287             if source_url:
 288                 break
 289         else:
 290             source_url = video_source['ottStreamUrl']
 291         is_live = video_source.get('type') == 'live' or video_source.get('status') == 'Live'
 292         resource = self._get_mvpd_resource('nbcsports', title, video_id, '')
 293         token = self._extract_mvpd_auth(url, video_id, 'nbcsports', resource)
 294         tokenized_url = self._download_json(
 295             'https://token.playmakerservices.com/cdn',
 296             video_id, data=json.dumps({
 297                 'requestorId': 'nbcsports',
 298                 'pid': video_id,
 299                 'application': 'NBCSports',
 300                 'version': 'v1',
 301                 'platform': 'desktop',
 302                 'cdn': 'akamai',
 303                 'url': video_source['sourceUrl'],
 304                 'token': base64.b64encode(token.encode()).decode(),
 305                 'resourceId': base64.b64encode(resource.encode()).decode(),
 306             }).encode())['tokenizedUrl']
 307         formats = self._extract_m3u8_formats(tokenized_url, video_id, 'mp4')
 308         self._sort_formats(formats)
 309         return {
 310             'id': video_id,
 311             'title': title,
 312             'description': live_source.get('description'),
 313             'formats': formats,
 314             'is_live': is_live,
 315         }
 316
 317
 318 class NBCNewsIE(ThePlatformIE):
 319     _VALID_URL = r'(?x)https?://(?:www\.)?(?:nbcnews|today|msnbc)\.com/([^/]+/)*(?:.*-)?(?P<id>[^/?]+)'
 320
 321     _TESTS = [
 322         {
 323             'url': 'http://www.nbcnews.com/watch/nbcnews-com/how-twitter-reacted-to-the-snowden-interview-269389891880',
 324             'md5': 'cf4bc9e6ce0130f00f545d80ecedd4bf',
 325             'info_dict': {
 326                 'id': '269389891880',
 327                 'ext': 'mp4',
 328                 'title': 'How Twitter Reacted To The Snowden Interview',
 329                 'description': 'md5:65a0bd5d76fe114f3c2727aa3a81fe64',
 330                 'timestamp': 1401363060,
 331                 'upload_date': '20140529',
 332             },
 333         },
 334         {
 335             'url': 'http://www.nbcnews.com/feature/dateline-full-episodes/full-episode-family-business-n285156',
 336             'md5': 'fdbf39ab73a72df5896b6234ff98518a',
 337             'info_dict': {
 338                 'id': '529953347624',
 339                 'ext': 'mp4',
 340                 'title': 'FULL EPISODE: Family Business',
 341                 'description': 'md5:757988edbaae9d7be1d585eb5d55cc04',
 342             },
 343             'skip': 'This page is unavailable.',
 344         },
 345         {
 346             'url': 'http://www.nbcnews.com/nightly-news/video/nightly-news-with-brian-williams-full-broadcast-february-4-394064451844',
 347             'md5': '8eb831eca25bfa7d25ddd83e85946548',
 348             'info_dict': {
 349                 'id': '394064451844',
 350                 'ext': 'mp4',
 351                 'title': 'Nightly News with Brian Williams Full Broadcast (February 4)',
 352                 'description': 'md5:1c10c1eccbe84a26e5debb4381e2d3c5',
 353                 'timestamp': 1423104900,
 354                 'upload_date': '20150205',
 355             },
 356         },
 357         {
 358             'url': 'http://www.nbcnews.com/business/autos/volkswagen-11-million-vehicles-could-have-suspect-software-emissions-scandal-n431456',
 359             'md5': '4a8c4cec9e1ded51060bdda36ff0a5c0',
 360             'info_dict': {
 361                 'id': 'n431456',
 362                 'ext': 'mp4',
 363                 'title': "Volkswagen U.S. Chief:  We 'Totally Screwed Up'",
 364                 'description': 'md5:d22d1281a24f22ea0880741bb4dd6301',
 365                 'upload_date': '20150922',
 366                 'timestamp': 1442917800,
 367             },
 368         },
 369         {
 370             'url': 'http://www.today.com/video/see-the-aurora-borealis-from-space-in-stunning-new-nasa-video-669831235788',
 371             'md5': '118d7ca3f0bea6534f119c68ef539f71',
 372             'info_dict': {
 373                 'id': '669831235788',
 374                 'ext': 'mp4',
 375                 'title': 'See the aurora borealis from space in stunning new NASA video',
 376                 'description': 'md5:74752b7358afb99939c5f8bb2d1d04b1',
 377                 'upload_date': '20160420',
 378                 'timestamp': 1461152093,
 379             },
 380         },
 381         {
 382             'url': 'http://www.msnbc.com/all-in-with-chris-hayes/watch/the-chaotic-gop-immigration-vote-314487875924',
 383             'md5': '6d236bf4f3dddc226633ce6e2c3f814d',
 384             'info_dict': {
 385                 'id': '314487875924',
 386                 'ext': 'mp4',
 387                 'title': 'The chaotic GOP immigration vote',
 388                 '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.',
 389                 'thumbnail': r're:^https?://.*\.jpg$',
 390                 'timestamp': 1406937606,
 391                 'upload_date': '20140802',
 392             },
 393         },
 394         {
 395             'url': 'http://www.nbcnews.com/watch/dateline/full-episode--deadly-betrayal-386250819952',
 396             'only_matching': True,
 397         },
 398         {
 399             # From http://www.vulture.com/2016/06/letterman-couldnt-care-less-about-late-night.html
 400             'url': 'http://www.nbcnews.com/widget/video-embed/701714499682',
 401             'only_matching': True,
 402         },
 403     ]
 404
 405     def _real_extract(self, url):
 406         video_id = self._match_id(url)
 407         webpage = self._download_webpage(url, video_id)
 408
 409         data = self._search_nextjs_data(webpage, video_id)['props']['initialState']
 410         video_data = try_get(data, lambda x: x['video']['current'], dict)
 411         if not video_data:
 412             video_data = data['article']['content'][0]['primaryMedia']['video']
 413         title = video_data['headline']['primary']
 414
 415         formats = []
 416         for va in video_data.get('videoAssets', []):
 417             public_url = va.get('publicUrl')
 418             if not public_url:
 419                 continue
 420             if '://link.theplatform.com/' in public_url:
 421                 public_url = update_url_query(public_url, {'format': 'redirect'})
 422             format_id = va.get('format')
 423             if format_id == 'M3U':
 424                 formats.extend(self._extract_m3u8_formats(
 425                     public_url, video_id, 'mp4', 'm3u8_native',
 426                     m3u8_id=format_id, fatal=False))
 427                 continue
 428             tbr = int_or_none(va.get('bitrate'), 1000)
 429             if tbr:
 430                 format_id += '-%d' % tbr
 431             formats.append({
 432                 'format_id': format_id,
 433                 'url': public_url,
 434                 'width': int_or_none(va.get('width')),
 435                 'height': int_or_none(va.get('height')),
 436                 'tbr': tbr,
 437                 'ext': 'mp4',
 438             })
 439         self._sort_formats(formats)
 440
 441         subtitles = {}
 442         closed_captioning = video_data.get('closedCaptioning')
 443         if closed_captioning:
 444             for cc_url in closed_captioning.values():
 445                 if not cc_url:
 446                     continue
 447                 subtitles.setdefault('en', []).append({
 448                     'url': cc_url,
 449                 })
 450
 451         return {
 452             'id': video_id,
 453             'title': title,
 454             'description': try_get(video_data, lambda x: x['description']['primary']),
 455             'thumbnail': try_get(video_data, lambda x: x['primaryImage']['url']['primary']),
 456             'duration': parse_duration(video_data.get('duration')),
 457             'timestamp': unified_timestamp(video_data.get('datePublished')),
 458             'formats': formats,
 459             'subtitles': subtitles,
 460         }
 461
 462
 463 class NBCOlympicsIE(InfoExtractor):
 464     IE_NAME = 'nbcolympics'
 465     _VALID_URL = r'https?://www\.nbcolympics\.com/videos?/(?P<id>[0-9a-z-]+)'
 466
 467     _TEST = {
 468         # Geo-restricted to US
 469         'url': 'http://www.nbcolympics.com/video/justin-roses-son-leo-was-tears-after-his-dad-won-gold',
 470         'md5': '54fecf846d05429fbaa18af557ee523a',
 471         'info_dict': {
 472             'id': 'WjTBzDXx5AUq',
 473             'display_id': 'justin-roses-son-leo-was-tears-after-his-dad-won-gold',
 474             'ext': 'mp4',
 475             'title': 'Rose\'s son Leo was in tears after his dad won gold',
 476             'description': 'Olympic gold medalist Justin Rose gets emotional talking to the impact his win in men\'s golf has already had on his children.',
 477             'timestamp': 1471274964,
 478             'upload_date': '20160815',
 479             'uploader': 'NBCU-SPORTS',
 480         },
 481     }
 482
 483     def _real_extract(self, url):
 484         display_id = self._match_id(url)
 485
 486         webpage = self._download_webpage(url, display_id)
 487
 488         try:
 489             drupal_settings = self._parse_json(self._search_regex(
 490                 r'jQuery\.extend\(Drupal\.settings\s*,\s*({.+?})\);',
 491                 webpage, 'drupal settings'), display_id)
 492
 493             iframe_url = drupal_settings['vod']['iframe_url']
 494             theplatform_url = iframe_url.replace(
 495                 'vplayer.nbcolympics.com', 'player.theplatform.com')
 496         except RegexNotFoundError:
 497             theplatform_url = self._search_regex(
 498                 r"([\"'])embedUrl\1: *([\"'])(?P<embedUrl>.+)\2",
 499                 webpage, 'embedding URL', group="embedUrl")
 500
 501         return {
 502             '_type': 'url_transparent',
 503             'url': theplatform_url,
 504             'ie_key': ThePlatformIE.ie_key(),
 505             'display_id': display_id,
 506         }
 507
 508
 509 class NBCOlympicsStreamIE(AdobePassIE):
 510     IE_NAME = 'nbcolympics:stream'
 511     _VALID_URL = r'https?://stream\.nbcolympics\.com/(?P<id>[0-9a-z-]+)'
 512     _TESTS = [
 513         {
 514             'note': 'Tokenized m3u8 source URL',
 515             'url': 'https://stream.nbcolympics.com/womens-soccer-group-round-11',
 516             'info_dict': {
 517                 'id': '2019740',
 518                 'ext': 'mp4',
 519                 'title': r"re:Women's Group Stage - Netherlands vs\. Brazil [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$",
 520             },
 521             'params': {
 522                 'skip_download': 'm3u8',
 523             },
 524         }, {
 525             'note': 'Plain m3u8 source URL',
 526             'url': 'https://stream.nbcolympics.com/gymnastics-event-finals-mens-floor-pommel-horse-womens-vault-bars',
 527             'info_dict': {
 528                 'id': '2021729',
 529                 'ext': 'mp4',
 530                 'title': r're:Event Finals: M Floor, W Vault, M Pommel, W Uneven Bars [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
 531             },
 532             'params': {
 533                 'skip_download': 'm3u8',
 534             },
 535         },
 536     ]
 537
 538     def _real_extract(self, url):
 539         display_id = self._match_id(url)
 540         webpage = self._download_webpage(url, display_id)
 541         pid = self._search_regex(r'pid\s*=\s*(\d+);', webpage, 'pid')
 542
 543         event_config = self._download_json(
 544             f'http://stream.nbcolympics.com/data/event_config_{pid}.json',
 545             pid, 'Downloading event config')['eventConfig']
 546
 547         title = event_config['eventTitle']
 548         is_live = {'live': True, 'replay': False}.get(event_config.get('eventStatus'))
 549
 550         source_url = self._download_json(
 551             f'https://api-leap.nbcsports.com/feeds/assets/{pid}?application=NBCOlympics&platform=desktop&format=nbc-player&env=staging',
 552             pid, 'Downloading leap config'
 553         )['videoSources'][0]['cdnSources']['primary'][0]['sourceUrl']
 554
 555         if event_config.get('cdnToken'):
 556             ap_resource = self._get_mvpd_resource(
 557                 event_config.get('resourceId', 'NBCOlympics'),
 558                 re.sub(r'[^\w\d ]+', '', event_config['eventTitle']), pid,
 559                 event_config.get('ratingId', 'NO VALUE'))
 560             media_token = self._extract_mvpd_auth(url, pid, event_config.get('requestorId', 'NBCOlympics'), ap_resource)
 561
 562             source_url = self._download_json(
 563                 'https://tokens.playmakerservices.com/', pid, 'Retrieving tokenized URL',
 564                 data=json.dumps({
 565                     'application': 'NBCSports',
 566                     'authentication-type': 'adobe-pass',
 567                     'cdn': 'akamai',
 568                     'pid': pid,
 569                     'platform': 'desktop',
 570                     'requestorId': 'NBCOlympics',
 571                     'resourceId': base64.b64encode(ap_resource.encode()).decode(),
 572                     'token': base64.b64encode(media_token.encode()).decode(),
 573                     'url': source_url,
 574                     'version': 'v1',
 575                 }).encode(),
 576             )['akamai'][0]['tokenizedUrl']
 577
 578         formats = self._extract_m3u8_formats(source_url, pid, 'mp4', live=is_live)
 579         for f in formats:
 580             # -http_seekable requires ffmpeg 4.3+ but it doesnt seem possible to
 581             # download with ffmpeg without this option
 582             f['downloader_options'] = {'ffmpeg_args': ['-seekable', '0', '-http_seekable', '0', '-icy', '0']}
 583         self._sort_formats(formats)
 584
 585         return {
 586             'id': pid,
 587             'display_id': display_id,
 588             'title': title,
 589             'formats': formats,
 590             'is_live': is_live,
 591         }