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, default_ns
   7 from .adobepass import AdobePassIE
   8 from ..compat import compat_urllib_parse_unquote
   9 from ..networking import HEADRequest
  10 from ..utils import (
  11     ExtractorError,
  12     RegexNotFoundError,
  13     UserNotLive,
  14     clean_html,
  15     determine_ext,
  16     float_or_none,
  17     int_or_none,
  18     mimetype2ext,
  19     parse_age_limit,
  20     parse_duration,
  21     remove_end,
  22     smuggle_url,
  23     traverse_obj,
  24     try_get,
  25     unescapeHTML,
  26     unified_timestamp,
  27     update_url_query,
  28     url_basename,
  29 )
  30
  31
  32 class NBCIE(ThePlatformIE):  # XXX: Do not subclass from concrete IE
  33     _VALID_URL = r'https?(?P<permalink>://(?:www\.)?nbc\.com/(?:classic-tv/)?[^/]+/video/[^/]+/(?P<id>(?:NBCE|n)?\d+))'
  34
  35     _TESTS = [
  36         {
  37             'url': 'http://www.nbc.com/the-tonight-show/video/jimmy-fallon-surprises-fans-at-ben-jerrys/2848237',
  38             'info_dict': {
  39                 'id': '2848237',
  40                 'ext': 'mp4',
  41                 'title': 'Jimmy Fallon Surprises Fans at Ben & Jerry\'s',
  42                 'description': 'Jimmy gives out free scoops of his new "Tonight Dough" ice cream flavor by surprising customers at the Ben & Jerry\'s scoop shop.',
  43                 'timestamp': 1424246400,
  44                 'upload_date': '20150218',
  45                 'uploader': 'NBCU-COM',
  46                 'episode': 'Jimmy Fallon Surprises Fans at Ben & Jerry\'s',
  47                 'episode_number': 86,
  48                 'season': 'Season 2',
  49                 'season_number': 2,
  50                 'series': 'Tonight Show: Jimmy Fallon',
  51                 'duration': 237.0,
  52                 'chapters': 'count:1',
  53                 'tags': 'count:4',
  54                 'thumbnail': r're:https?://.+\.jpg',
  55             },
  56             'params': {
  57                 'skip_download': 'm3u8',
  58             },
  59         },
  60         {
  61             'url': 'http://www.nbc.com/saturday-night-live/video/star-wars-teaser/2832821',
  62             'info_dict': {
  63                 'id': '2832821',
  64                 'ext': 'mp4',
  65                 'title': 'Star Wars Teaser',
  66                 'description': 'md5:0b40f9cbde5b671a7ff62fceccc4f442',
  67                 'timestamp': 1417852800,
  68                 'upload_date': '20141206',
  69                 'uploader': 'NBCU-COM',
  70             },
  71             'skip': 'page not found',
  72         },
  73         {
  74             # HLS streams requires the 'hdnea3' cookie
  75             'url': 'http://www.nbc.com/Kings/video/goliath/n1806',
  76             'info_dict': {
  77                 'id': '101528f5a9e8127b107e98c5e6ce4638',
  78                 'ext': 'mp4',
  79                 'title': 'Goliath',
  80                 '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.',
  81                 'timestamp': 1237100400,
  82                 'upload_date': '20090315',
  83                 'uploader': 'NBCU-COM',
  84             },
  85             'skip': 'page not found',
  86         },
  87         {
  88             # manifest url does not have extension
  89             'url': 'https://www.nbc.com/the-golden-globe-awards/video/oprah-winfrey-receives-cecil-b-de-mille-award-at-the-2018-golden-globes/3646439',
  90             'info_dict': {
  91                 'id': '3646439',
  92                 'ext': 'mp4',
  93                 'title': 'Oprah Winfrey Receives Cecil B. de Mille Award at the 2018 Golden Globes',
  94                 'episode': 'Oprah Winfrey Receives Cecil B. de Mille Award at the 2018 Golden Globes',
  95                 'episode_number': 1,
  96                 'season': 'Season 75',
  97                 'season_number': 75,
  98                 'series': 'The Golden Globe Awards',
  99                 'description': 'Oprah Winfrey receives the Cecil B. de Mille Award at the 75th Annual Golden Globe Awards.',
 100                 'uploader': 'NBCU-COM',
 101                 'upload_date': '20180107',
 102                 'timestamp': 1515312000,
 103                 'duration': 570.0,
 104                 'tags': 'count:8',
 105                 'thumbnail': r're:https?://.+\.jpg',
 106                 'chapters': 'count:1',
 107             },
 108             'params': {
 109                 'skip_download': 'm3u8',
 110             },
 111         },
 112         {
 113             # new video_id format
 114             'url': 'https://www.nbc.com/quantum-leap/video/bens-first-leap-nbcs-quantum-leap/NBCE125189978',
 115             'info_dict': {
 116                 'id': 'NBCE125189978',
 117                 'ext': 'mp4',
 118                 'title': 'Ben\'s First Leap | NBC\'s Quantum Leap',
 119                 'description': 'md5:a82762449b7ec4bb83291a7b355ebf8e',
 120                 'uploader': 'NBCU-COM',
 121                 'series': 'Quantum Leap',
 122                 'season': 'Season 1',
 123                 'season_number': 1,
 124                 'episode': 'Ben\'s First Leap | NBC\'s Quantum Leap',
 125                 'episode_number': 1,
 126                 'duration': 170.171,
 127                 'chapters': [],
 128                 'timestamp': 1663956155,
 129                 'upload_date': '20220923',
 130                 'tags': 'count:10',
 131                 'age_limit': 0,
 132                 'thumbnail': r're:https?://.+\.jpg',
 133             },
 134             'params': {
 135                 'skip_download': 'm3u8',
 136             },
 137         },
 138         {
 139             'url': 'https://www.nbc.com/classic-tv/charles-in-charge/video/charles-in-charge-pilot/n3310',
 140             'only_matching': True,
 141         },
 142         {
 143             # Percent escaped url
 144             'url': 'https://www.nbc.com/up-all-night/video/day-after-valentine%27s-day/n2189',
 145             'only_matching': True,
 146         }
 147     ]
 148
 149     def _real_extract(self, url):
 150         permalink, video_id = self._match_valid_url(url).groups()
 151         permalink = 'http' + compat_urllib_parse_unquote(permalink)
 152         video_data = self._download_json(
 153             'https://friendship.nbc.co/v2/graphql', video_id, query={
 154                 'query': '''query bonanzaPage(
 155   $app: NBCUBrands! = nbc
 156   $name: String!
 157   $oneApp: Boolean
 158   $platform: SupportedPlatforms! = web
 159   $type: EntityPageType! = VIDEO
 160   $userId: String!
 161 ) {
 162   bonanzaPage(
 163     app: $app
 164     name: $name
 165     oneApp: $oneApp
 166     platform: $platform
 167     type: $type
 168     userId: $userId
 169   ) {
 170     metadata {
 171       ... on VideoPageData {
 172         description
 173         episodeNumber
 174         keywords
 175         locked
 176         mpxAccountId
 177         mpxGuid
 178         rating
 179         resourceId
 180         seasonNumber
 181         secondaryTitle
 182         seriesShortTitle
 183       }
 184     }
 185   }
 186 }''',
 187                 'variables': json.dumps({
 188                     'name': permalink,
 189                     'oneApp': True,
 190                     'userId': '0',
 191                 }),
 192             })['data']['bonanzaPage']['metadata']
 193         query = {
 194             'mbr': 'true',
 195             'manifest': 'm3u',
 196             'switch': 'HLSServiceSecure',
 197         }
 198         video_id = video_data['mpxGuid']
 199         tp_path = 'NnzsPC/media/guid/%s/%s' % (video_data.get('mpxAccountId') or '2410887629', video_id)
 200         tpm = self._download_theplatform_metadata(tp_path, video_id)
 201         title = tpm.get('title') or video_data.get('secondaryTitle')
 202         if video_data.get('locked'):
 203             resource = self._get_mvpd_resource(
 204                 video_data.get('resourceId') or 'nbcentertainment',
 205                 title, video_id, video_data.get('rating'))
 206             query['auth'] = self._extract_mvpd_auth(
 207                 url, video_id, 'nbcentertainment', resource)
 208         theplatform_url = smuggle_url(update_url_query(
 209             'http://link.theplatform.com/s/NnzsPC/media/guid/%s/%s' % (video_data.get('mpxAccountId') or '2410887629', video_id),
 210             query), {'force_smil_url': True})
 211
 212         # Empty string or 0 can be valid values for these. So the check must be `is None`
 213         description = video_data.get('description')
 214         if description is None:
 215             description = tpm.get('description')
 216         episode_number = int_or_none(video_data.get('episodeNumber'))
 217         if episode_number is None:
 218             episode_number = int_or_none(tpm.get('nbcu$airOrder'))
 219         rating = video_data.get('rating')
 220         if rating is None:
 221             try_get(tpm, lambda x: x['ratings'][0]['rating'])
 222         season_number = int_or_none(video_data.get('seasonNumber'))
 223         if season_number is None:
 224             season_number = int_or_none(tpm.get('nbcu$seasonNumber'))
 225         series = video_data.get('seriesShortTitle')
 226         if series is None:
 227             series = tpm.get('nbcu$seriesShortTitle')
 228         tags = video_data.get('keywords')
 229         if tags is None or len(tags) == 0:
 230             tags = tpm.get('keywords')
 231
 232         return {
 233             '_type': 'url_transparent',
 234             'age_limit': parse_age_limit(rating),
 235             'description': description,
 236             'episode': title,
 237             'episode_number': episode_number,
 238             'id': video_id,
 239             'ie_key': 'ThePlatform',
 240             'season_number': season_number,
 241             'series': series,
 242             'tags': tags,
 243             'title': title,
 244             'url': theplatform_url,
 245         }
 246
 247
 248 class NBCSportsVPlayerIE(InfoExtractor):
 249     _VALID_URL_BASE = r'https?://(?:vplayer\.nbcsports\.com|(?:www\.)?nbcsports\.com/vplayer)/'
 250     _VALID_URL = _VALID_URL_BASE + r'(?:[^/]+/)+(?P<id>[0-9a-zA-Z_]+)'
 251     _EMBED_REGEX = [r'(?:iframe[^>]+|var video|div[^>]+data-(?:mpx-)?)[sS]rc\s?=\s?"(?P<url>%s[^\"]+)' % _VALID_URL_BASE]
 252
 253     _TESTS = [{
 254         'url': 'https://vplayer.nbcsports.com/p/BxmELC/nbcsports_embed/select/9CsDKds0kvHI',
 255         'info_dict': {
 256             'id': '9CsDKds0kvHI',
 257             'ext': 'mp4',
 258             'description': 'md5:df390f70a9ba7c95ff1daace988f0d8d',
 259             'title': 'Tyler Kalinoski hits buzzer-beater to lift Davidson',
 260             'timestamp': 1426270238,
 261             'upload_date': '20150313',
 262             'uploader': 'NBCU-SPORTS',
 263             'duration': 72.818,
 264             'chapters': [],
 265             'thumbnail': r're:^https?://.*\.jpg$'
 266         }
 267     }, {
 268         'url': 'https://vplayer.nbcsports.com/p/BxmELC/nbcsports_embed/select/media/PEgOtlNcC_y2',
 269         'only_matching': True,
 270     }, {
 271         'url': 'https://www.nbcsports.com/vplayer/p/BxmELC/nbcsports/select/PHJSaFWbrTY9?form=html&autoPlay=true',
 272         'only_matching': True,
 273     }]
 274
 275     def _real_extract(self, url):
 276         video_id = self._match_id(url)
 277         webpage = self._download_webpage(url, video_id)
 278         theplatform_url = self._html_search_regex(r'tp:releaseUrl="(.+?)"', webpage, 'url')
 279         return self.url_result(theplatform_url, 'ThePlatform')
 280
 281
 282 class NBCSportsIE(InfoExtractor):
 283     _VALID_URL = r'https?://(?:www\.)?nbcsports\.com//?(?!vplayer/)(?:[^/]+/)+(?P<id>[0-9a-z-]+)'
 284
 285     _TESTS = [{
 286         # iframe src
 287         'url': 'https://www.nbcsports.com/watch/nfl/profootballtalk/pft-pm/unpacking-addisons-reckless-driving-citation',
 288         'info_dict': {
 289             'id': 'PHJSaFWbrTY9',
 290             'ext': 'mp4',
 291             'title': 'Tom Izzo, Michigan St. has \'so much respect\' for Duke',
 292             'description': 'md5:ecb459c9d59e0766ac9c7d5d0eda8113',
 293             'uploader': 'NBCU-SPORTS',
 294             'upload_date': '20150330',
 295             'timestamp': 1427726529,
 296             'chapters': [],
 297             'thumbnail': 'https://hdliveextra-a.akamaihd.net/HD/image_sports/NBCU_Sports_Group_-_nbcsports/253/303/izzodps.jpg',
 298             'duration': 528.395,
 299         }
 300     }, {
 301         # data-mpx-src
 302         'url': 'https://www.nbcsports.com/philadelphia/philadelphia-phillies/bruce-bochy-hector-neris-hes-idiot',
 303         'only_matching': True,
 304     }, {
 305         # data-src
 306         'url': 'https://www.nbcsports.com/boston/video/report-card-pats-secondary-no-match-josh-allen',
 307         'only_matching': True,
 308     }]
 309
 310     def _real_extract(self, url):
 311         video_id = self._match_id(url)
 312         webpage = self._download_webpage(url, video_id)
 313         return self.url_result(
 314             NBCSportsVPlayerIE._extract_url(webpage), 'NBCSportsVPlayer')
 315
 316
 317 class NBCSportsStreamIE(AdobePassIE):
 318     _VALID_URL = r'https?://stream\.nbcsports\.com/.+?\bpid=(?P<id>\d+)'
 319     _TEST = {
 320         'url': 'http://stream.nbcsports.com/nbcsn/generic?pid=206559',
 321         'info_dict': {
 322             'id': '206559',
 323             'ext': 'mp4',
 324             'title': 'Amgen Tour of California Women\'s Recap',
 325             'description': 'md5:66520066b3b5281ada7698d0ea2aa894',
 326         },
 327         'params': {
 328             # m3u8 download
 329             'skip_download': True,
 330         },
 331         'skip': 'Requires Adobe Pass Authentication',
 332     }
 333
 334     def _real_extract(self, url):
 335         video_id = self._match_id(url)
 336         live_source = self._download_json(
 337             'http://stream.nbcsports.com/data/live_sources_%s.json' % video_id,
 338             video_id)
 339         video_source = live_source['videoSources'][0]
 340         title = video_source['title']
 341         source_url = None
 342         for k in ('source', 'msl4source', 'iossource', 'hlsv4'):
 343             sk = k + 'Url'
 344             source_url = video_source.get(sk) or video_source.get(sk + 'Alt')
 345             if source_url:
 346                 break
 347         else:
 348             source_url = video_source['ottStreamUrl']
 349         is_live = video_source.get('type') == 'live' or video_source.get('status') == 'Live'
 350         resource = self._get_mvpd_resource('nbcsports', title, video_id, '')
 351         token = self._extract_mvpd_auth(url, video_id, 'nbcsports', resource)
 352         tokenized_url = self._download_json(
 353             'https://token.playmakerservices.com/cdn',
 354             video_id, data=json.dumps({
 355                 'requestorId': 'nbcsports',
 356                 'pid': video_id,
 357                 'application': 'NBCSports',
 358                 'version': 'v1',
 359                 'platform': 'desktop',
 360                 'cdn': 'akamai',
 361                 'url': video_source['sourceUrl'],
 362                 'token': base64.b64encode(token.encode()).decode(),
 363                 'resourceId': base64.b64encode(resource.encode()).decode(),
 364             }).encode())['tokenizedUrl']
 365         formats = self._extract_m3u8_formats(tokenized_url, video_id, 'mp4')
 366         return {
 367             'id': video_id,
 368             'title': title,
 369             'description': live_source.get('description'),
 370             'formats': formats,
 371             'is_live': is_live,
 372         }
 373
 374
 375 class NBCNewsIE(ThePlatformIE):  # XXX: Do not subclass from concrete IE
 376     _VALID_URL = r'(?x)https?://(?:www\.)?(?:nbcnews|today|msnbc)\.com/([^/]+/)*(?:.*-)?(?P<id>[^/?]+)'
 377     _EMBED_REGEX = [r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//www\.nbcnews\.com/widget/video-embed/[^"\']+)\1']
 378
 379     _TESTS = [
 380         {
 381             'url': 'http://www.nbcnews.com/watch/nbcnews-com/how-twitter-reacted-to-the-snowden-interview-269389891880',
 382             'md5': 'fb3dcd2d7b1dd9804305fa2fc95ab610',  # md5 tends to fluctuate
 383             'info_dict': {
 384                 'id': '269389891880',
 385                 'ext': 'mp4',
 386                 'title': 'How Twitter Reacted To The Snowden Interview',
 387                 'description': 'md5:65a0bd5d76fe114f3c2727aa3a81fe64',
 388                 'timestamp': 1401363060,
 389                 'upload_date': '20140529',
 390                 'duration': 46.0,
 391                 'thumbnail': 'https://media-cldnry.s-nbcnews.com/image/upload/MSNBC/Components/Video/140529/p_tweet_snow_140529.jpg',
 392             },
 393         },
 394         {
 395             'url': 'http://www.nbcnews.com/feature/dateline-full-episodes/full-episode-family-business-n285156',
 396             'md5': 'fdbf39ab73a72df5896b6234ff98518a',
 397             'info_dict': {
 398                 'id': '529953347624',
 399                 'ext': 'mp4',
 400                 'title': 'FULL EPISODE: Family Business',
 401                 'description': 'md5:757988edbaae9d7be1d585eb5d55cc04',
 402             },
 403             'skip': 'This page is unavailable.',
 404         },
 405         {
 406             'url': 'http://www.nbcnews.com/nightly-news/video/nightly-news-with-brian-williams-full-broadcast-february-4-394064451844',
 407             'md5': '40d0e48c68896359c80372306ece0fc3',
 408             'info_dict': {
 409                 'id': '394064451844',
 410                 'ext': 'mp4',
 411                 'title': 'Nightly News with Brian Williams Full Broadcast (February 4)',
 412                 'description': 'md5:1c10c1eccbe84a26e5debb4381e2d3c5',
 413                 'timestamp': 1423104900,
 414                 'upload_date': '20150205',
 415                 'duration': 1236.0,
 416                 'thumbnail': 'https://media-cldnry.s-nbcnews.com/image/upload/MSNBC/Components/Video/__NEW/nn_netcast_150204.jpg',
 417             },
 418         },
 419         {
 420             'url': 'http://www.nbcnews.com/business/autos/volkswagen-11-million-vehicles-could-have-suspect-software-emissions-scandal-n431456',
 421             'md5': 'ffb59bcf0733dc3c7f0ace907f5e3939',
 422             'info_dict': {
 423                 'id': 'n431456',
 424                 'ext': 'mp4',
 425                 'title': "Volkswagen U.S. Chief:  We 'Totally Screwed Up'",
 426                 'description': 'md5:d22d1281a24f22ea0880741bb4dd6301',
 427                 'upload_date': '20150922',
 428                 'timestamp': 1442917800,
 429                 'duration': 37.0,
 430                 'thumbnail': 'https://media-cldnry.s-nbcnews.com/image/upload/MSNBC/Components/Video/__NEW/x_lon_vwhorn_150922.jpg',
 431             },
 432         },
 433         {
 434             'url': 'http://www.today.com/video/see-the-aurora-borealis-from-space-in-stunning-new-nasa-video-669831235788',
 435             'md5': '693d1fa21d23afcc9b04c66b227ed9ff',
 436             'info_dict': {
 437                 'id': '669831235788',
 438                 'ext': 'mp4',
 439                 'title': 'See the aurora borealis from space in stunning new NASA video',
 440                 'description': 'md5:74752b7358afb99939c5f8bb2d1d04b1',
 441                 'upload_date': '20160420',
 442                 'timestamp': 1461152093,
 443                 'duration': 69.0,
 444                 'thumbnail': 'https://media-cldnry.s-nbcnews.com/image/upload/MSNBC/Components/Video/201604/2016-04-20T11-35-09-133Z--1280x720.jpg',
 445             },
 446         },
 447         {
 448             'url': 'http://www.msnbc.com/all-in-with-chris-hayes/watch/the-chaotic-gop-immigration-vote-314487875924',
 449             'md5': '6d236bf4f3dddc226633ce6e2c3f814d',
 450             'info_dict': {
 451                 'id': '314487875924',
 452                 'ext': 'mp4',
 453                 'title': 'The chaotic GOP immigration vote',
 454                 '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.',
 455                 'thumbnail': r're:^https?://.*\.jpg$',
 456                 'timestamp': 1406937606,
 457                 'upload_date': '20140802',
 458                 'duration': 940.0,
 459             },
 460         },
 461         {
 462             'url': 'http://www.nbcnews.com/watch/dateline/full-episode--deadly-betrayal-386250819952',
 463             'only_matching': True,
 464         },
 465         {
 466             # From http://www.vulture.com/2016/06/letterman-couldnt-care-less-about-late-night.html
 467             'url': 'http://www.nbcnews.com/widget/video-embed/701714499682',
 468             'only_matching': True,
 469         },
 470     ]
 471
 472     def _real_extract(self, url):
 473         video_id = self._match_id(url)
 474         webpage = self._download_webpage(url, video_id)
 475
 476         data = self._search_nextjs_data(webpage, video_id)['props']['initialState']
 477         video_data = try_get(data, lambda x: x['video']['current'], dict)
 478         if not video_data:
 479             video_data = data['article']['content'][0]['primaryMedia']['video']
 480         title = video_data['headline']['primary']
 481
 482         formats = []
 483         for va in video_data.get('videoAssets', []):
 484             public_url = va.get('publicUrl')
 485             if not public_url:
 486                 continue
 487             if '://link.theplatform.com/' in public_url:
 488                 public_url = update_url_query(public_url, {'format': 'redirect'})
 489             format_id = va.get('format')
 490             if format_id == 'M3U':
 491                 formats.extend(self._extract_m3u8_formats(
 492                     public_url, video_id, 'mp4', 'm3u8_native',
 493                     m3u8_id=format_id, fatal=False))
 494                 continue
 495             tbr = int_or_none(va.get('bitrate'), 1000)
 496             if tbr:
 497                 format_id += '-%d' % tbr
 498             formats.append({
 499                 'format_id': format_id,
 500                 'url': public_url,
 501                 'width': int_or_none(va.get('width')),
 502                 'height': int_or_none(va.get('height')),
 503                 'tbr': tbr,
 504                 'ext': 'mp4',
 505             })
 506
 507         subtitles = {}
 508         closed_captioning = video_data.get('closedCaptioning')
 509         if closed_captioning:
 510             for cc_url in closed_captioning.values():
 511                 if not cc_url:
 512                     continue
 513                 subtitles.setdefault('en', []).append({
 514                     'url': cc_url,
 515                 })
 516
 517         return {
 518             'id': video_id,
 519             'title': title,
 520             'description': try_get(video_data, lambda x: x['description']['primary']),
 521             'thumbnail': try_get(video_data, lambda x: x['primaryImage']['url']['primary']),
 522             'duration': parse_duration(video_data.get('duration')),
 523             'timestamp': unified_timestamp(video_data.get('datePublished')),
 524             'formats': formats,
 525             'subtitles': subtitles,
 526         }
 527
 528
 529 class NBCOlympicsIE(InfoExtractor):
 530     IE_NAME = 'nbcolympics'
 531     _VALID_URL = r'https?://www\.nbcolympics\.com/videos?/(?P<id>[0-9a-z-]+)'
 532
 533     _TEST = {
 534         # Geo-restricted to US
 535         'url': 'http://www.nbcolympics.com/video/justin-roses-son-leo-was-tears-after-his-dad-won-gold',
 536         'md5': '54fecf846d05429fbaa18af557ee523a',
 537         'info_dict': {
 538             'id': 'WjTBzDXx5AUq',
 539             'display_id': 'justin-roses-son-leo-was-tears-after-his-dad-won-gold',
 540             'ext': 'mp4',
 541             'title': 'Rose\'s son Leo was in tears after his dad won gold',
 542             'description': 'Olympic gold medalist Justin Rose gets emotional talking to the impact his win in men\'s golf has already had on his children.',
 543             'timestamp': 1471274964,
 544             'upload_date': '20160815',
 545             'uploader': 'NBCU-SPORTS',
 546         },
 547         'skip': '404 Not Found',
 548     }
 549
 550     def _real_extract(self, url):
 551         display_id = self._match_id(url)
 552
 553         webpage = self._download_webpage(url, display_id)
 554
 555         try:
 556             drupal_settings = self._parse_json(self._search_regex(
 557                 r'jQuery\.extend\(Drupal\.settings\s*,\s*({.+?})\);',
 558                 webpage, 'drupal settings'), display_id)
 559
 560             iframe_url = drupal_settings['vod']['iframe_url']
 561             theplatform_url = iframe_url.replace(
 562                 'vplayer.nbcolympics.com', 'player.theplatform.com')
 563         except RegexNotFoundError:
 564             theplatform_url = self._search_regex(
 565                 r"([\"'])embedUrl\1: *([\"'])(?P<embedUrl>.+)\2",
 566                 webpage, 'embedding URL', group="embedUrl")
 567
 568         return {
 569             '_type': 'url_transparent',
 570             'url': theplatform_url,
 571             'ie_key': ThePlatformIE.ie_key(),
 572             'display_id': display_id,
 573         }
 574
 575
 576 class NBCOlympicsStreamIE(AdobePassIE):
 577     IE_NAME = 'nbcolympics:stream'
 578     _VALID_URL = r'https?://stream\.nbcolympics\.com/(?P<id>[0-9a-z-]+)'
 579     _TESTS = [
 580         {
 581             'note': 'Tokenized m3u8 source URL',
 582             'url': 'https://stream.nbcolympics.com/womens-soccer-group-round-11',
 583             'info_dict': {
 584                 'id': '2019740',
 585                 'ext': 'mp4',
 586                 '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}$",
 587             },
 588             'params': {
 589                 'skip_download': 'm3u8',
 590             },
 591             'skip': 'Livestream',
 592         }, {
 593             'note': 'Plain m3u8 source URL',
 594             'url': 'https://stream.nbcolympics.com/gymnastics-event-finals-mens-floor-pommel-horse-womens-vault-bars',
 595             'info_dict': {
 596                 'id': '2021729',
 597                 'ext': 'mp4',
 598                 'title': r're:Event Finals: M Floor, W Vault, M Pommel, W Uneven Bars [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
 599             },
 600             'params': {
 601                 'skip_download': 'm3u8',
 602             },
 603             'skip': 'Livestream',
 604         },
 605     ]
 606
 607     def _real_extract(self, url):
 608         display_id = self._match_id(url)
 609         webpage = self._download_webpage(url, display_id)
 610         pid = self._search_regex(r'pid\s*=\s*(\d+);', webpage, 'pid')
 611
 612         event_config = self._download_json(
 613             f'http://stream.nbcolympics.com/data/event_config_{pid}.json',
 614             pid, 'Downloading event config')['eventConfig']
 615
 616         title = event_config['eventTitle']
 617         is_live = {'live': True, 'replay': False}.get(event_config.get('eventStatus'))
 618
 619         source_url = self._download_json(
 620             f'https://api-leap.nbcsports.com/feeds/assets/{pid}?application=NBCOlympics&platform=desktop&format=nbc-player&env=staging',
 621             pid, 'Downloading leap config'
 622         )['videoSources'][0]['cdnSources']['primary'][0]['sourceUrl']
 623
 624         if event_config.get('cdnToken'):
 625             ap_resource = self._get_mvpd_resource(
 626                 event_config.get('resourceId', 'NBCOlympics'),
 627                 re.sub(r'[^\w\d ]+', '', event_config['eventTitle']), pid,
 628                 event_config.get('ratingId', 'NO VALUE'))
 629             media_token = self._extract_mvpd_auth(url, pid, event_config.get('requestorId', 'NBCOlympics'), ap_resource)
 630
 631             source_url = self._download_json(
 632                 'https://tokens.playmakerservices.com/', pid, 'Retrieving tokenized URL',
 633                 data=json.dumps({
 634                     'application': 'NBCSports',
 635                     'authentication-type': 'adobe-pass',
 636                     'cdn': 'akamai',
 637                     'pid': pid,
 638                     'platform': 'desktop',
 639                     'requestorId': 'NBCOlympics',
 640                     'resourceId': base64.b64encode(ap_resource.encode()).decode(),
 641                     'token': base64.b64encode(media_token.encode()).decode(),
 642                     'url': source_url,
 643                     'version': 'v1',
 644                 }).encode(),
 645             )['akamai'][0]['tokenizedUrl']
 646
 647         formats = self._extract_m3u8_formats(source_url, pid, 'mp4', live=is_live)
 648         for f in formats:
 649             # -http_seekable requires ffmpeg 4.3+ but it doesnt seem possible to
 650             # download with ffmpeg without this option
 651             f['downloader_options'] = {'ffmpeg_args': ['-seekable', '0', '-http_seekable', '0', '-icy', '0']}
 652
 653         return {
 654             'id': pid,
 655             'display_id': display_id,
 656             'title': title,
 657             'formats': formats,
 658             'is_live': is_live,
 659         }
 660
 661
 662 class NBCStationsIE(InfoExtractor):
 663     _DOMAIN_RE = '|'.join(map(re.escape, (
 664         'nbcbayarea', 'nbcboston', 'nbcchicago', 'nbcconnecticut', 'nbcdfw', 'nbclosangeles',
 665         'nbcmiami', 'nbcnewyork', 'nbcphiladelphia', 'nbcsandiego', 'nbcwashington',
 666         'necn', 'telemundo52', 'telemundoarizona', 'telemundochicago', 'telemundonuevainglaterra',
 667     )))
 668     _VALID_URL = rf'https?://(?:www\.)?(?P<site>{_DOMAIN_RE})\.com/(?:[^/?#]+/)*(?P<id>[^/?#]+)/?(?:$|[#?])'
 669
 670     _TESTS = [{
 671         'url': 'https://www.nbclosangeles.com/news/local/large-structure-fire-in-downtown-la-prompts-smoke-odor-advisory/2968618/',
 672         'info_dict': {
 673             'id': '2968618',
 674             'ext': 'mp4',
 675             'title': 'Large Structure Fire in Downtown LA Prompts Smoke Odor Advisory',
 676             'description': 'md5:417ed3c2d91fe9d301e6db7b0942f182',
 677             'duration': 112.513,
 678             'timestamp': 1661135892,
 679             'upload_date': '20220822',
 680             'uploader': 'NBC 4',
 681             'channel_id': 'KNBC',
 682             'channel': 'nbclosangeles',
 683         },
 684         'params': {
 685             'skip_download': 'm3u8',
 686         },
 687     }, {
 688         'url': 'https://www.telemundoarizona.com/responde/huracan-complica-reembolso-para-televidente-de-tucson/2247002/',
 689         'info_dict': {
 690             'id': '2247002',
 691             'ext': 'mp4',
 692             'title': 'Huracán complica que televidente de Tucson reciba  reembolso',
 693             'description': 'md5:af298dc73aab74d4fca6abfb12acb6cf',
 694             'duration': 172.406,
 695             'timestamp': 1660886507,
 696             'upload_date': '20220819',
 697             'uploader': 'Telemundo Arizona',
 698             'channel_id': 'KTAZ',
 699             'channel': 'telemundoarizona',
 700         },
 701         'params': {
 702             'skip_download': 'm3u8',
 703         },
 704     }, {
 705         # direct mp4 link
 706         'url': 'https://www.nbcboston.com/weather/video-weather/highs-near-freezing-in-boston-on-wednesday/2961135/',
 707         'md5': '9bf8c41dc7abbb75b1a44f1491a4cc85',
 708         'info_dict': {
 709             'id': '2961135',
 710             'ext': 'mp4',
 711             'title': 'Highs Near Freezing in Boston on Wednesday',
 712             'description': 'md5:3ec486609a926c99f00a3512e6c0e85b',
 713             'duration': 235.669,
 714             'timestamp': 1675268656,
 715             'upload_date': '20230201',
 716             'uploader': '',
 717             'channel_id': 'WBTS',
 718             'channel': 'nbcboston',
 719         },
 720     }]
 721
 722     _RESOLUTIONS = {
 723         '1080': '1920',
 724         '720': '1280',
 725         '540': '960',
 726         '360': '640',
 727         '234': '416',
 728     }
 729
 730     def _real_extract(self, url):
 731         channel, video_id = self._match_valid_url(url).group('site', 'id')
 732         webpage = self._download_webpage(url, video_id)
 733
 734         nbc_data = self._search_json(
 735             r'<script>\s*var\s+nbc\s*=', webpage, 'NBC JSON data', video_id)
 736         pdk_acct = nbc_data.get('pdkAcct') or 'Yh1nAC'
 737         fw_ssid = traverse_obj(nbc_data, ('video', 'fwSSID'))
 738
 739         video_data = self._search_json(
 740             r'data-videos="\[', webpage, 'video data', video_id, default={}, transform_source=unescapeHTML)
 741         video_data.update(self._search_json(
 742             r'data-meta="', webpage, 'metadata', video_id, default={}, transform_source=unescapeHTML))
 743         if not video_data:
 744             raise ExtractorError('No video metadata found in webpage', expected=True)
 745
 746         info, formats = {}, []
 747         is_live = int_or_none(video_data.get('mpx_is_livestream')) == 1
 748         query = {
 749             'formats': 'MPEG-DASH none,M3U none,MPEG-DASH none,MPEG4,MP3',
 750             'format': 'SMIL',
 751             'fwsitesection': fw_ssid,
 752             'fwNetworkID': traverse_obj(nbc_data, ('video', 'fwNetworkID'), default='382114'),
 753             'pprofile': 'ots_desktop_html',
 754             'sensitive': 'false',
 755             'w': '1920',
 756             'h': '1080',
 757             'mode': 'LIVE' if is_live else 'on-demand',
 758             'vpaid': 'script',
 759             'schema': '2.0',
 760             'sdk': 'PDK 6.1.3',
 761         }
 762
 763         if is_live:
 764             player_id = traverse_obj(video_data, ((None, ('video', 'meta')), (
 765                 'mpx_m3upid', 'mpx_pid', 'pid_streaming_web_medium')), get_all=False)
 766             info['title'] = f'{channel} livestream'
 767
 768         else:
 769             player_id = traverse_obj(video_data, (
 770                 (None, ('video', 'meta')), ('pid_streaming_web_high', 'mpx_pid')), get_all=False)
 771
 772             date_string = traverse_obj(video_data, 'date_string', 'date_gmt')
 773             if date_string:
 774                 date_string = self._search_regex(
 775                     r'datetime="([^"]+)"', date_string, 'date string', fatal=False)
 776             else:
 777                 date_string = traverse_obj(
 778                     nbc_data, ('dataLayer', 'adobe', ('prop70', 'eVar70', 'eVar59')), get_all=False)
 779
 780             video_url = traverse_obj(video_data, ((None, ('video', 'meta')), 'mp4_url'), get_all=False)
 781             if video_url:
 782                 ext = determine_ext(video_url)
 783                 height = self._search_regex(r'\d+-(\d+)p', url_basename(video_url), 'height', default=None)
 784                 formats.append({
 785                     'url': video_url,
 786                     'ext': ext,
 787                     'width': int_or_none(self._RESOLUTIONS.get(height)),
 788                     'height': int_or_none(height),
 789                     'format_id': f'http-{ext}',
 790                 })
 791
 792             info.update({
 793                 'title': video_data.get('title') or traverse_obj(nbc_data, (
 794                     'dataLayer', (None, 'adobe'), ('contenttitle', 'title', 'prop22')), get_all=False),
 795                 'description':
 796                     traverse_obj(video_data, 'summary', 'excerpt', 'video_hero_text')
 797                     or clean_html(traverse_obj(nbc_data, ('dataLayer', 'summary'))),
 798                 'timestamp': unified_timestamp(date_string),
 799             })
 800
 801         smil = None
 802         if player_id and fw_ssid:
 803             smil = self._download_xml(
 804                 f'https://link.theplatform.com/s/{pdk_acct}/{player_id}', video_id,
 805                 note='Downloading SMIL data', query=query, fatal=is_live)
 806         subtitles = self._parse_smil_subtitles(smil, default_ns) if smil else {}
 807         for video in smil.findall(self._xpath_ns('.//video', default_ns)) if smil else []:
 808             info['duration'] = float_or_none(remove_end(video.get('dur'), 'ms'), 1000)
 809             video_src_url = video.get('src')
 810             ext = mimetype2ext(video.get('type'), default=determine_ext(video_src_url))
 811             if ext == 'm3u8':
 812                 fmts, subs = self._extract_m3u8_formats_and_subtitles(
 813                     video_src_url, video_id, 'mp4', m3u8_id='hls', fatal=is_live,
 814                     live=is_live, errnote='No HLS formats found')
 815                 formats.extend(fmts)
 816                 self._merge_subtitles(subs, target=subtitles)
 817             elif video_src_url:
 818                 formats.append({
 819                     'url': video_src_url,
 820                     'format_id': f'https-{ext}',
 821                     'ext': ext,
 822                     'width': int_or_none(video.get('width')),
 823                     'height': int_or_none(video.get('height')),
 824                 })
 825
 826         if not formats:
 827             self.raise_no_formats('No video content found in webpage', expected=True)
 828         elif is_live:
 829             try:
 830                 self._request_webpage(
 831                     HEADRequest(formats[0]['url']), video_id, note='Checking live status')
 832             except ExtractorError:
 833                 raise UserNotLive(video_id=channel)
 834
 835         return {
 836             'id': video_id,
 837             'channel': channel,
 838             'channel_id': nbc_data.get('callLetters'),
 839             'uploader': nbc_data.get('on_air_name'),
 840             'formats': formats,
 841             'subtitles': subtitles,
 842             'is_live': is_live,
 843             **info,
 844         }