yt_dlp/extractor/nbc.py

   1 import base64
   2 import json
   3 import re
   4 import urllib.parse
   5 import xml.etree.ElementTree
   6
   7 from .adobepass import AdobePassIE
   8 from .common import InfoExtractor
   9 from .theplatform import ThePlatformIE, default_ns
  10 from ..networking import HEADRequest
  11 from ..utils import (
  12     ExtractorError,
  13     RegexNotFoundError,
  14     UserNotLive,
  15     clean_html,
  16     determine_ext,
  17     float_or_none,
  18     int_or_none,
  19     join_nonempty,
  20     mimetype2ext,
  21     parse_age_limit,
  22     parse_duration,
  23     remove_end,
  24     smuggle_url,
  25     traverse_obj,
  26     try_get,
  27     unescapeHTML,
  28     unified_timestamp,
  29     update_url_query,
  30     url_basename,
  31 )
  32
  33
  34 class NBCIE(ThePlatformIE):  # XXX: Do not subclass from concrete IE
  35     _VALID_URL = r'https?(?P<permalink>://(?:www\.)?nbc\.com/(?:classic-tv/)?[^/]+/video/[^/]+/(?P<id>(?:NBCE|n)?\d+))'
  36
  37     _TESTS = [
  38         {
  39             'url': 'http://www.nbc.com/the-tonight-show/video/jimmy-fallon-surprises-fans-at-ben-jerrys/2848237',
  40             'info_dict': {
  41                 'id': '2848237',
  42                 'ext': 'mp4',
  43                 'title': 'Jimmy Fallon Surprises Fans at Ben & Jerry\'s',
  44                 'description': 'Jimmy gives out free scoops of his new "Tonight Dough" ice cream flavor by surprising customers at the Ben & Jerry\'s scoop shop.',
  45                 'timestamp': 1424246400,
  46                 'upload_date': '20150218',
  47                 'uploader': 'NBCU-COM',
  48                 'episode': 'Jimmy Fallon Surprises Fans at Ben & Jerry\'s',
  49                 'episode_number': 86,
  50                 'season': 'Season 2',
  51                 'season_number': 2,
  52                 'series': 'Tonight Show: Jimmy Fallon',
  53                 'duration': 237.0,
  54                 'chapters': 'count:1',
  55                 'tags': 'count:4',
  56                 'thumbnail': r're:https?://.+\.jpg',
  57                 'categories': ['Series/The Tonight Show Starring Jimmy Fallon'],
  58                 'media_type': 'Full Episode',
  59             },
  60             'params': {
  61                 'skip_download': 'm3u8',
  62             },
  63         },
  64         {
  65             'url': 'http://www.nbc.com/saturday-night-live/video/star-wars-teaser/2832821',
  66             'info_dict': {
  67                 'id': '2832821',
  68                 'ext': 'mp4',
  69                 'title': 'Star Wars Teaser',
  70                 'description': 'md5:0b40f9cbde5b671a7ff62fceccc4f442',
  71                 'timestamp': 1417852800,
  72                 'upload_date': '20141206',
  73                 'uploader': 'NBCU-COM',
  74             },
  75             'skip': 'page not found',
  76         },
  77         {
  78             # HLS streams requires the 'hdnea3' cookie
  79             'url': 'http://www.nbc.com/Kings/video/goliath/n1806',
  80             'info_dict': {
  81                 'id': '101528f5a9e8127b107e98c5e6ce4638',
  82                 'ext': 'mp4',
  83                 'title': 'Goliath',
  84                 '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.',
  85                 'timestamp': 1237100400,
  86                 'upload_date': '20090315',
  87                 'uploader': 'NBCU-COM',
  88             },
  89             'skip': 'page not found',
  90         },
  91         {
  92             # manifest url does not have extension
  93             'url': 'https://www.nbc.com/the-golden-globe-awards/video/oprah-winfrey-receives-cecil-b-de-mille-award-at-the-2018-golden-globes/3646439',
  94             'info_dict': {
  95                 'id': '3646439',
  96                 'ext': 'mp4',
  97                 'title': 'Oprah Winfrey Receives Cecil B. de Mille Award at the 2018 Golden Globes',
  98                 'episode': 'Oprah Winfrey Receives Cecil B. de Mille Award at the 2018 Golden Globes',
  99                 'episode_number': 1,
 100                 'season': 'Season 75',
 101                 'season_number': 75,
 102                 'series': 'The Golden Globe Awards',
 103                 'description': 'Oprah Winfrey receives the Cecil B. de Mille Award at the 75th Annual Golden Globe Awards.',
 104                 'uploader': 'NBCU-COM',
 105                 'upload_date': '20180107',
 106                 'timestamp': 1515312000,
 107                 'duration': 570.0,
 108                 'tags': 'count:8',
 109                 'thumbnail': r're:https?://.+\.jpg',
 110                 'chapters': 'count:1',
 111             },
 112             'params': {
 113                 'skip_download': 'm3u8',
 114             },
 115         },
 116         {
 117             # new video_id format
 118             'url': 'https://www.nbc.com/quantum-leap/video/bens-first-leap-nbcs-quantum-leap/NBCE125189978',
 119             'info_dict': {
 120                 'id': 'NBCE125189978',
 121                 'ext': 'mp4',
 122                 'title': 'Ben\'s First Leap | NBC\'s Quantum Leap',
 123                 'description': 'md5:a82762449b7ec4bb83291a7b355ebf8e',
 124                 'uploader': 'NBCU-COM',
 125                 'series': 'Quantum Leap',
 126                 'season': 'Season 1',
 127                 'season_number': 1,
 128                 'episode': 'Ben\'s First Leap | NBC\'s Quantum Leap',
 129                 'episode_number': 1,
 130                 'duration': 170.171,
 131                 'chapters': [],
 132                 'timestamp': 1663956155,
 133                 'upload_date': '20220923',
 134                 'tags': 'count:10',
 135                 'age_limit': 0,
 136                 'thumbnail': r're:https?://.+\.jpg',
 137                 'categories': ['Series/Quantum Leap 2022'],
 138                 'media_type': 'Highlight',
 139             },
 140             'params': {
 141                 'skip_download': 'm3u8',
 142             },
 143         },
 144         {
 145             'url': 'https://www.nbc.com/classic-tv/charles-in-charge/video/charles-in-charge-pilot/n3310',
 146             'only_matching': True,
 147         },
 148         {
 149             # Percent escaped url
 150             'url': 'https://www.nbc.com/up-all-night/video/day-after-valentine%27s-day/n2189',
 151             'only_matching': True,
 152         },
 153     ]
 154
 155     def _real_extract(self, url):
 156         permalink, video_id = self._match_valid_url(url).groups()
 157         permalink = 'http' + urllib.parse.unquote(permalink)
 158         video_data = self._download_json(
 159             'https://friendship.nbc.co/v2/graphql', video_id, query={
 160                 'query': '''query bonanzaPage(
 161   $app: NBCUBrands! = nbc
 162   $name: String!
 163   $oneApp: Boolean
 164   $platform: SupportedPlatforms! = web
 165   $type: EntityPageType! = VIDEO
 166   $userId: String!
 167 ) {
 168   bonanzaPage(
 169     app: $app
 170     name: $name
 171     oneApp: $oneApp
 172     platform: $platform
 173     type: $type
 174     userId: $userId
 175   ) {
 176     metadata {
 177       ... on VideoPageData {
 178         description
 179         episodeNumber
 180         keywords
 181         locked
 182         mpxAccountId
 183         mpxGuid
 184         rating
 185         resourceId
 186         seasonNumber
 187         secondaryTitle
 188         seriesShortTitle
 189       }
 190     }
 191   }
 192 }''',
 193                 'variables': json.dumps({
 194                     'name': permalink,
 195                     'oneApp': True,
 196                     'userId': '0',
 197                 }),
 198             })['data']['bonanzaPage']['metadata']
 199         query = {
 200             'mbr': 'true',
 201             'manifest': 'm3u',
 202             'switch': 'HLSServiceSecure',
 203         }
 204         video_id = video_data['mpxGuid']
 205         tp_path = 'NnzsPC/media/guid/{}/{}'.format(video_data.get('mpxAccountId') or '2410887629', video_id)
 206         tpm = self._download_theplatform_metadata(tp_path, video_id)
 207         title = tpm.get('title') or video_data.get('secondaryTitle')
 208         if video_data.get('locked'):
 209             resource = self._get_mvpd_resource(
 210                 video_data.get('resourceId') or 'nbcentertainment',
 211                 title, video_id, video_data.get('rating'))
 212             query['auth'] = self._extract_mvpd_auth(
 213                 url, video_id, 'nbcentertainment', resource)
 214         theplatform_url = smuggle_url(update_url_query(
 215             'http://link.theplatform.com/s/NnzsPC/media/guid/{}/{}'.format(video_data.get('mpxAccountId') or '2410887629', video_id),
 216             query), {'force_smil_url': True})
 217
 218         # Empty string or 0 can be valid values for these. So the check must be `is None`
 219         description = video_data.get('description')
 220         if description is None:
 221             description = tpm.get('description')
 222         episode_number = int_or_none(video_data.get('episodeNumber'))
 223         if episode_number is None:
 224             episode_number = int_or_none(tpm.get('nbcu$airOrder'))
 225         rating = video_data.get('rating')
 226         if rating is None:
 227             try_get(tpm, lambda x: x['ratings'][0]['rating'])
 228         season_number = int_or_none(video_data.get('seasonNumber'))
 229         if season_number is None:
 230             season_number = int_or_none(tpm.get('nbcu$seasonNumber'))
 231         series = video_data.get('seriesShortTitle')
 232         if series is None:
 233             series = tpm.get('nbcu$seriesShortTitle')
 234         tags = video_data.get('keywords')
 235         if tags is None or len(tags) == 0:
 236             tags = tpm.get('keywords')
 237
 238         return {
 239             '_type': 'url_transparent',
 240             'age_limit': parse_age_limit(rating),
 241             'description': description,
 242             'episode': title,
 243             'episode_number': episode_number,
 244             'id': video_id,
 245             'ie_key': 'ThePlatform',
 246             'season_number': season_number,
 247             'series': series,
 248             'tags': tags,
 249             'title': title,
 250             'url': theplatform_url,
 251         }
 252
 253
 254 class NBCSportsVPlayerIE(InfoExtractor):
 255     _VALID_URL_BASE = r'https?://(?:vplayer\.nbcsports\.com|(?:www\.)?nbcsports\.com/vplayer)/'
 256     _VALID_URL = _VALID_URL_BASE + r'(?:[^/]+/)+(?P<id>[0-9a-zA-Z_]+)'
 257     _EMBED_REGEX = [rf'(?:iframe[^>]+|var video|div[^>]+data-(?:mpx-)?)[sS]rc\s?=\s?"(?P<url>{_VALID_URL_BASE}[^\"]+)']
 258
 259     _TESTS = [{
 260         'url': 'https://vplayer.nbcsports.com/p/BxmELC/nbcsports_embed/select/9CsDKds0kvHI',
 261         'info_dict': {
 262             'id': '9CsDKds0kvHI',
 263             'ext': 'mp4',
 264             'description': 'md5:df390f70a9ba7c95ff1daace988f0d8d',
 265             'title': 'Tyler Kalinoski hits buzzer-beater to lift Davidson',
 266             'timestamp': 1426270238,
 267             'upload_date': '20150313',
 268             'uploader': 'NBCU-SPORTS',
 269             'duration': 72.818,
 270             'chapters': [],
 271             'thumbnail': r're:^https?://.*\.jpg$',
 272         },
 273     }, {
 274         'url': 'https://vplayer.nbcsports.com/p/BxmELC/nbcsports_embed/select/media/PEgOtlNcC_y2',
 275         'only_matching': True,
 276     }, {
 277         'url': 'https://www.nbcsports.com/vplayer/p/BxmELC/nbcsports/select/PHJSaFWbrTY9?form=html&autoPlay=true',
 278         'only_matching': True,
 279     }]
 280
 281     def _real_extract(self, url):
 282         video_id = self._match_id(url)
 283         webpage = self._download_webpage(url, video_id)
 284         theplatform_url = self._html_search_regex(r'tp:releaseUrl="(.+?)"', webpage, 'url')
 285         return self.url_result(theplatform_url, 'ThePlatform')
 286
 287
 288 class NBCSportsIE(InfoExtractor):
 289     _VALID_URL = r'https?://(?:www\.)?nbcsports\.com//?(?!vplayer/)(?:[^/]+/)+(?P<id>[0-9a-z-]+)'
 290
 291     _TESTS = [{
 292         # iframe src
 293         'url': 'https://www.nbcsports.com/watch/nfl/profootballtalk/pft-pm/unpacking-addisons-reckless-driving-citation',
 294         'info_dict': {
 295             'id': 'PHJSaFWbrTY9',
 296             'ext': 'mp4',
 297             'title': 'Tom Izzo, Michigan St. has \'so much respect\' for Duke',
 298             'description': 'md5:ecb459c9d59e0766ac9c7d5d0eda8113',
 299             'uploader': 'NBCU-SPORTS',
 300             'upload_date': '20150330',
 301             'timestamp': 1427726529,
 302             'chapters': [],
 303             'thumbnail': 'https://hdliveextra-a.akamaihd.net/HD/image_sports/NBCU_Sports_Group_-_nbcsports/253/303/izzodps.jpg',
 304             'duration': 528.395,
 305         },
 306     }, {
 307         # data-mpx-src
 308         'url': 'https://www.nbcsports.com/philadelphia/philadelphia-phillies/bruce-bochy-hector-neris-hes-idiot',
 309         'only_matching': True,
 310     }, {
 311         # data-src
 312         'url': 'https://www.nbcsports.com/boston/video/report-card-pats-secondary-no-match-josh-allen',
 313         'only_matching': True,
 314     }]
 315
 316     def _real_extract(self, url):
 317         video_id = self._match_id(url)
 318         webpage = self._download_webpage(url, video_id)
 319         return self.url_result(
 320             NBCSportsVPlayerIE._extract_url(webpage), 'NBCSportsVPlayer')
 321
 322
 323 class NBCSportsStreamIE(AdobePassIE):
 324     _VALID_URL = r'https?://stream\.nbcsports\.com/.+?\bpid=(?P<id>\d+)'
 325     _TEST = {
 326         'url': 'http://stream.nbcsports.com/nbcsn/generic?pid=206559',
 327         'info_dict': {
 328             'id': '206559',
 329             'ext': 'mp4',
 330             'title': 'Amgen Tour of California Women\'s Recap',
 331             'description': 'md5:66520066b3b5281ada7698d0ea2aa894',
 332         },
 333         'params': {
 334             # m3u8 download
 335             'skip_download': True,
 336         },
 337         'skip': 'Requires Adobe Pass Authentication',
 338     }
 339
 340     def _real_extract(self, url):
 341         video_id = self._match_id(url)
 342         live_source = self._download_json(
 343             f'http://stream.nbcsports.com/data/live_sources_{video_id}.json',
 344             video_id)
 345         video_source = live_source['videoSources'][0]
 346         title = video_source['title']
 347         source_url = None
 348         for k in ('source', 'msl4source', 'iossource', 'hlsv4'):
 349             sk = k + 'Url'
 350             source_url = video_source.get(sk) or video_source.get(sk + 'Alt')
 351             if source_url:
 352                 break
 353         else:
 354             source_url = video_source['ottStreamUrl']
 355         is_live = video_source.get('type') == 'live' or video_source.get('status') == 'Live'
 356         resource = self._get_mvpd_resource('nbcsports', title, video_id, '')
 357         token = self._extract_mvpd_auth(url, video_id, 'nbcsports', resource)
 358         tokenized_url = self._download_json(
 359             'https://token.playmakerservices.com/cdn',
 360             video_id, data=json.dumps({
 361                 'requestorId': 'nbcsports',
 362                 'pid': video_id,
 363                 'application': 'NBCSports',
 364                 'version': 'v1',
 365                 'platform': 'desktop',
 366                 'cdn': 'akamai',
 367                 'url': video_source['sourceUrl'],
 368                 'token': base64.b64encode(token.encode()).decode(),
 369                 'resourceId': base64.b64encode(resource.encode()).decode(),
 370             }).encode())['tokenizedUrl']
 371         formats = self._extract_m3u8_formats(tokenized_url, video_id, 'mp4')
 372         return {
 373             'id': video_id,
 374             'title': title,
 375             'description': live_source.get('description'),
 376             'formats': formats,
 377             'is_live': is_live,
 378         }
 379
 380
 381 class NBCNewsIE(ThePlatformIE):  # XXX: Do not subclass from concrete IE
 382     _VALID_URL = r'(?x)https?://(?:www\.)?(?:nbcnews|today|msnbc)\.com/([^/]+/)*(?:.*-)?(?P<id>[^/?]+)'
 383     _EMBED_REGEX = [r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//www\.nbcnews\.com/widget/video-embed/[^"\']+)\1']
 384
 385     _TESTS = [
 386         {
 387             'url': 'http://www.nbcnews.com/watch/nbcnews-com/how-twitter-reacted-to-the-snowden-interview-269389891880',
 388             'md5': 'fb3dcd2d7b1dd9804305fa2fc95ab610',  # md5 tends to fluctuate
 389             'info_dict': {
 390                 'id': '269389891880',
 391                 'ext': 'mp4',
 392                 'title': 'How Twitter Reacted To The Snowden Interview',
 393                 'description': 'md5:65a0bd5d76fe114f3c2727aa3a81fe64',
 394                 'timestamp': 1401363060,
 395                 'upload_date': '20140529',
 396                 'duration': 46.0,
 397                 'thumbnail': 'https://media-cldnry.s-nbcnews.com/image/upload/MSNBC/Components/Video/140529/p_tweet_snow_140529.jpg',
 398             },
 399         },
 400         {
 401             'url': 'http://www.nbcnews.com/feature/dateline-full-episodes/full-episode-family-business-n285156',
 402             'md5': 'fdbf39ab73a72df5896b6234ff98518a',
 403             'info_dict': {
 404                 'id': '529953347624',
 405                 'ext': 'mp4',
 406                 'title': 'FULL EPISODE: Family Business',
 407                 'description': 'md5:757988edbaae9d7be1d585eb5d55cc04',
 408             },
 409             'skip': 'This page is unavailable.',
 410         },
 411         {
 412             'url': 'http://www.nbcnews.com/nightly-news/video/nightly-news-with-brian-williams-full-broadcast-february-4-394064451844',
 413             'md5': '40d0e48c68896359c80372306ece0fc3',
 414             'info_dict': {
 415                 'id': '394064451844',
 416                 'ext': 'mp4',
 417                 'title': 'Nightly News with Brian Williams Full Broadcast (February 4)',
 418                 'description': 'md5:1c10c1eccbe84a26e5debb4381e2d3c5',
 419                 'timestamp': 1423104900,
 420                 'upload_date': '20150205',
 421                 'duration': 1236.0,
 422                 'thumbnail': 'https://media-cldnry.s-nbcnews.com/image/upload/MSNBC/Components/Video/__NEW/nn_netcast_150204.jpg',
 423             },
 424         },
 425         {
 426             'url': 'http://www.nbcnews.com/business/autos/volkswagen-11-million-vehicles-could-have-suspect-software-emissions-scandal-n431456',
 427             'md5': 'ffb59bcf0733dc3c7f0ace907f5e3939',
 428             'info_dict': {
 429                 'id': 'n431456',
 430                 'ext': 'mp4',
 431                 'title': "Volkswagen U.S. Chief:  We 'Totally Screwed Up'",
 432                 'description': 'md5:d22d1281a24f22ea0880741bb4dd6301',
 433                 'upload_date': '20150922',
 434                 'timestamp': 1442917800,
 435                 'duration': 37.0,
 436                 'thumbnail': 'https://media-cldnry.s-nbcnews.com/image/upload/MSNBC/Components/Video/__NEW/x_lon_vwhorn_150922.jpg',
 437             },
 438         },
 439         {
 440             'url': 'http://www.today.com/video/see-the-aurora-borealis-from-space-in-stunning-new-nasa-video-669831235788',
 441             'md5': '693d1fa21d23afcc9b04c66b227ed9ff',
 442             'info_dict': {
 443                 'id': '669831235788',
 444                 'ext': 'mp4',
 445                 'title': 'See the aurora borealis from space in stunning new NASA video',
 446                 'description': 'md5:74752b7358afb99939c5f8bb2d1d04b1',
 447                 'upload_date': '20160420',
 448                 'timestamp': 1461152093,
 449                 'duration': 69.0,
 450                 'thumbnail': 'https://media-cldnry.s-nbcnews.com/image/upload/MSNBC/Components/Video/201604/2016-04-20T11-35-09-133Z--1280x720.jpg',
 451             },
 452         },
 453         {
 454             'url': 'http://www.msnbc.com/all-in-with-chris-hayes/watch/the-chaotic-gop-immigration-vote-314487875924',
 455             'md5': '6d236bf4f3dddc226633ce6e2c3f814d',
 456             'info_dict': {
 457                 'id': '314487875924',
 458                 'ext': 'mp4',
 459                 'title': 'The chaotic GOP immigration vote',
 460                 '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.',
 461                 'thumbnail': r're:^https?://.*\.jpg$',
 462                 'timestamp': 1406937606,
 463                 'upload_date': '20140802',
 464                 'duration': 940.0,
 465             },
 466         },
 467         {
 468             'url': 'http://www.nbcnews.com/watch/dateline/full-episode--deadly-betrayal-386250819952',
 469             'only_matching': True,
 470         },
 471         {
 472             # From http://www.vulture.com/2016/06/letterman-couldnt-care-less-about-late-night.html
 473             'url': 'http://www.nbcnews.com/widget/video-embed/701714499682',
 474             'only_matching': True,
 475         },
 476     ]
 477
 478     def _real_extract(self, url):
 479         video_id = self._match_id(url)
 480         webpage = self._download_webpage(url, video_id)
 481
 482         data = self._search_nextjs_data(webpage, video_id)['props']['initialState']
 483         video_data = try_get(data, lambda x: x['video']['current'], dict)
 484         if not video_data:
 485             video_data = data['article']['content'][0]['primaryMedia']['video']
 486         title = video_data['headline']['primary']
 487
 488         formats = []
 489         for va in video_data.get('videoAssets', []):
 490             public_url = va.get('publicUrl')
 491             if not public_url:
 492                 continue
 493             if '://link.theplatform.com/' in public_url:
 494                 public_url = update_url_query(public_url, {'format': 'redirect'})
 495             format_id = va.get('format')
 496             if format_id == 'M3U':
 497                 formats.extend(self._extract_m3u8_formats(
 498                     public_url, video_id, 'mp4', 'm3u8_native',
 499                     m3u8_id=format_id, fatal=False))
 500                 continue
 501             tbr = int_or_none(va.get('bitrate'), 1000)
 502             formats.append({
 503                 'format_id': join_nonempty(format_id, tbr),
 504                 'url': public_url,
 505                 'width': int_or_none(va.get('width')),
 506                 'height': int_or_none(va.get('height')),
 507                 'tbr': tbr,
 508                 'ext': 'mp4',
 509             })
 510
 511         subtitles = {}
 512         closed_captioning = video_data.get('closedCaptioning')
 513         if closed_captioning:
 514             for cc_url in closed_captioning.values():
 515                 if not cc_url:
 516                     continue
 517                 subtitles.setdefault('en', []).append({
 518                     'url': cc_url,
 519                 })
 520
 521         return {
 522             'id': video_id,
 523             'title': title,
 524             'description': try_get(video_data, lambda x: x['description']['primary']),
 525             'thumbnail': try_get(video_data, lambda x: x['primaryImage']['url']['primary']),
 526             'duration': parse_duration(video_data.get('duration')),
 527             'timestamp': unified_timestamp(video_data.get('datePublished')),
 528             'formats': formats,
 529             'subtitles': subtitles,
 530         }
 531
 532
 533 class NBCOlympicsIE(InfoExtractor):
 534     IE_NAME = 'nbcolympics'
 535     _VALID_URL = r'https?://www\.nbcolympics\.com/videos?/(?P<id>[0-9a-z-]+)'
 536
 537     _TEST = {
 538         # Geo-restricted to US
 539         'url': 'http://www.nbcolympics.com/video/justin-roses-son-leo-was-tears-after-his-dad-won-gold',
 540         'md5': '54fecf846d05429fbaa18af557ee523a',
 541         'info_dict': {
 542             'id': 'WjTBzDXx5AUq',
 543             'display_id': 'justin-roses-son-leo-was-tears-after-his-dad-won-gold',
 544             'ext': 'mp4',
 545             'title': 'Rose\'s son Leo was in tears after his dad won gold',
 546             'description': 'Olympic gold medalist Justin Rose gets emotional talking to the impact his win in men\'s golf has already had on his children.',
 547             'timestamp': 1471274964,
 548             'upload_date': '20160815',
 549             'uploader': 'NBCU-SPORTS',
 550         },
 551         'skip': '404 Not Found',
 552     }
 553
 554     def _real_extract(self, url):
 555         display_id = self._match_id(url)
 556
 557         webpage = self._download_webpage(url, display_id)
 558
 559         try:
 560             drupal_settings = self._parse_json(self._search_regex(
 561                 r'jQuery\.extend\(Drupal\.settings\s*,\s*({.+?})\);',
 562                 webpage, 'drupal settings'), display_id)
 563
 564             iframe_url = drupal_settings['vod']['iframe_url']
 565             theplatform_url = iframe_url.replace(
 566                 'vplayer.nbcolympics.com', 'player.theplatform.com')
 567         except RegexNotFoundError:
 568             theplatform_url = self._search_regex(
 569                 r"([\"'])embedUrl\1: *([\"'])(?P<embedUrl>.+)\2",
 570                 webpage, 'embedding URL', group='embedUrl')
 571
 572         return {
 573             '_type': 'url_transparent',
 574             'url': theplatform_url,
 575             'ie_key': ThePlatformIE.ie_key(),
 576             'display_id': display_id,
 577         }
 578
 579
 580 class NBCOlympicsStreamIE(AdobePassIE):
 581     IE_NAME = 'nbcolympics:stream'
 582     _VALID_URL = r'https?://stream\.nbcolympics\.com/(?P<id>[0-9a-z-]+)'
 583     _TESTS = [
 584         {
 585             'note': 'Tokenized m3u8 source URL',
 586             'url': 'https://stream.nbcolympics.com/womens-soccer-group-round-11',
 587             'info_dict': {
 588                 'id': '2019740',
 589                 'ext': 'mp4',
 590                 '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}$",
 591             },
 592             'params': {
 593                 'skip_download': 'm3u8',
 594             },
 595             'skip': 'Livestream',
 596         }, {
 597             'note': 'Plain m3u8 source URL',
 598             'url': 'https://stream.nbcolympics.com/gymnastics-event-finals-mens-floor-pommel-horse-womens-vault-bars',
 599             'info_dict': {
 600                 'id': '2021729',
 601                 'ext': 'mp4',
 602                 'title': r're:Event Finals: M Floor, W Vault, M Pommel, W Uneven Bars [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
 603             },
 604             'params': {
 605                 'skip_download': 'm3u8',
 606             },
 607             'skip': 'Livestream',
 608         },
 609     ]
 610
 611     def _real_extract(self, url):
 612         display_id = self._match_id(url)
 613         webpage = self._download_webpage(url, display_id)
 614         pid = self._search_regex(r'pid\s*=\s*(\d+);', webpage, 'pid')
 615
 616         event_config = self._download_json(
 617             f'http://stream.nbcolympics.com/data/event_config_{pid}.json',
 618             pid, 'Downloading event config')['eventConfig']
 619
 620         title = event_config['eventTitle']
 621         is_live = {'live': True, 'replay': False}.get(event_config.get('eventStatus'))
 622
 623         source_url = self._download_json(
 624             f'https://api-leap.nbcsports.com/feeds/assets/{pid}?application=NBCOlympics&platform=desktop&format=nbc-player&env=staging',
 625             pid, 'Downloading leap config',
 626         )['videoSources'][0]['cdnSources']['primary'][0]['sourceUrl']
 627
 628         if event_config.get('cdnToken'):
 629             ap_resource = self._get_mvpd_resource(
 630                 event_config.get('resourceId', 'NBCOlympics'),
 631                 re.sub(r'[^\w\d ]+', '', event_config['eventTitle']), pid,
 632                 event_config.get('ratingId', 'NO VALUE'))
 633             media_token = self._extract_mvpd_auth(url, pid, event_config.get('requestorId', 'NBCOlympics'), ap_resource)
 634
 635             source_url = self._download_json(
 636                 'https://tokens.playmakerservices.com/', pid, 'Retrieving tokenized URL',
 637                 data=json.dumps({
 638                     'application': 'NBCSports',
 639                     'authentication-type': 'adobe-pass',
 640                     'cdn': 'akamai',
 641                     'pid': pid,
 642                     'platform': 'desktop',
 643                     'requestorId': 'NBCOlympics',
 644                     'resourceId': base64.b64encode(ap_resource.encode()).decode(),
 645                     'token': base64.b64encode(media_token.encode()).decode(),
 646                     'url': source_url,
 647                     'version': 'v1',
 648                 }).encode(),
 649             )['akamai'][0]['tokenizedUrl']
 650
 651         formats = self._extract_m3u8_formats(source_url, pid, 'mp4', live=is_live)
 652         for f in formats:
 653             # -http_seekable requires ffmpeg 4.3+ but it doesnt seem possible to
 654             # download with ffmpeg without this option
 655             f['downloader_options'] = {'ffmpeg_args': ['-seekable', '0', '-http_seekable', '0', '-icy', '0']}
 656
 657         return {
 658             'id': pid,
 659             'display_id': display_id,
 660             'title': title,
 661             'formats': formats,
 662             'is_live': is_live,
 663         }
 664
 665
 666 class NBCStationsIE(InfoExtractor):
 667     _DOMAIN_RE = '|'.join(map(re.escape, (
 668         'nbcbayarea', 'nbcboston', 'nbcchicago', 'nbcconnecticut', 'nbcdfw', 'nbclosangeles',
 669         'nbcmiami', 'nbcnewyork', 'nbcphiladelphia', 'nbcsandiego', 'nbcwashington',
 670         'necn', 'telemundo52', 'telemundoarizona', 'telemundochicago', 'telemundonuevainglaterra',
 671     )))
 672     _VALID_URL = rf'https?://(?:www\.)?(?P<site>{_DOMAIN_RE})\.com/(?:[^/?#]+/)*(?P<id>[^/?#]+)/?(?:$|[#?])'
 673
 674     _TESTS = [{
 675         'url': 'https://www.nbclosangeles.com/news/local/large-structure-fire-in-downtown-la-prompts-smoke-odor-advisory/2968618/',
 676         'info_dict': {
 677             'id': '2968618',
 678             'ext': 'mp4',
 679             'title': 'Large Structure Fire in Downtown LA Prompts Smoke Odor Advisory',
 680             'description': 'md5:417ed3c2d91fe9d301e6db7b0942f182',
 681             'duration': 112.513,
 682             'timestamp': 1661135892,
 683             'upload_date': '20220822',
 684             'uploader': 'NBC 4',
 685             'channel_id': 'KNBC',
 686             'channel': 'nbclosangeles',
 687         },
 688         'params': {
 689             'skip_download': 'm3u8',
 690         },
 691     }, {
 692         'url': 'https://www.telemundoarizona.com/responde/huracan-complica-reembolso-para-televidente-de-tucson/2247002/',
 693         'info_dict': {
 694             'id': '2247002',
 695             'ext': 'mp4',
 696             'title': 'Huracán complica que televidente de Tucson reciba  reembolso',
 697             'description': 'md5:af298dc73aab74d4fca6abfb12acb6cf',
 698             'duration': 172.406,
 699             'timestamp': 1660886507,
 700             'upload_date': '20220819',
 701             'uploader': 'Telemundo Arizona',
 702             'channel_id': 'KTAZ',
 703             'channel': 'telemundoarizona',
 704         },
 705         'params': {
 706             'skip_download': 'm3u8',
 707         },
 708     }, {
 709         # direct mp4 link
 710         'url': 'https://www.nbcboston.com/weather/video-weather/highs-near-freezing-in-boston-on-wednesday/2961135/',
 711         'md5': '9bf8c41dc7abbb75b1a44f1491a4cc85',
 712         'info_dict': {
 713             'id': '2961135',
 714             'ext': 'mp4',
 715             'title': 'Highs Near Freezing in Boston on Wednesday',
 716             'description': 'md5:3ec486609a926c99f00a3512e6c0e85b',
 717             'duration': 235.669,
 718             'timestamp': 1675268656,
 719             'upload_date': '20230201',
 720             'uploader': '',
 721             'channel_id': 'WBTS',
 722             'channel': 'nbcboston',
 723         },
 724     }]
 725
 726     _RESOLUTIONS = {
 727         '1080': '1920',
 728         '720': '1280',
 729         '540': '960',
 730         '360': '640',
 731         '234': '416',
 732     }
 733
 734     def _real_extract(self, url):
 735         channel, video_id = self._match_valid_url(url).group('site', 'id')
 736         webpage = self._download_webpage(url, video_id)
 737
 738         nbc_data = self._search_json(
 739             r'<script>\s*var\s+nbc\s*=', webpage, 'NBC JSON data', video_id)
 740         pdk_acct = nbc_data.get('pdkAcct') or 'Yh1nAC'
 741         fw_ssid = traverse_obj(nbc_data, ('video', 'fwSSID'))
 742
 743         video_data = self._search_json(
 744             r'data-videos="\[', webpage, 'video data', video_id, default={}, transform_source=unescapeHTML)
 745         video_data.update(self._search_json(
 746             r'data-meta="', webpage, 'metadata', video_id, default={}, transform_source=unescapeHTML))
 747         if not video_data:
 748             raise ExtractorError('No video metadata found in webpage', expected=True)
 749
 750         info, formats = {}, []
 751         is_live = int_or_none(video_data.get('mpx_is_livestream')) == 1
 752         query = {
 753             'formats': 'MPEG-DASH none,M3U none,MPEG-DASH none,MPEG4,MP3',
 754             'format': 'SMIL',
 755             'fwsitesection': fw_ssid,
 756             'fwNetworkID': traverse_obj(nbc_data, ('video', 'fwNetworkID'), default='382114'),
 757             'pprofile': 'ots_desktop_html',
 758             'sensitive': 'false',
 759             'w': '1920',
 760             'h': '1080',
 761             'mode': 'LIVE' if is_live else 'on-demand',
 762             'vpaid': 'script',
 763             'schema': '2.0',
 764             'sdk': 'PDK 6.1.3',
 765         }
 766
 767         if is_live:
 768             player_id = traverse_obj(video_data, ((None, ('video', 'meta')), (
 769                 'mpx_m3upid', 'mpx_pid', 'pid_streaming_web_medium')), get_all=False)
 770             info['title'] = f'{channel} livestream'
 771
 772         else:
 773             player_id = traverse_obj(video_data, (
 774                 (None, ('video', 'meta')), ('pid_streaming_web_high', 'mpx_pid')), get_all=False)
 775
 776             date_string = traverse_obj(video_data, 'date_string', 'date_gmt')
 777             if date_string:
 778                 date_string = self._search_regex(
 779                     r'datetime="([^"]+)"', date_string, 'date string', fatal=False)
 780             else:
 781                 date_string = traverse_obj(
 782                     nbc_data, ('dataLayer', 'adobe', ('prop70', 'eVar70', 'eVar59')), get_all=False)
 783
 784             video_url = traverse_obj(video_data, ((None, ('video', 'meta')), 'mp4_url'), get_all=False)
 785             if video_url:
 786                 ext = determine_ext(video_url)
 787                 height = self._search_regex(r'\d+-(\d+)p', url_basename(video_url), 'height', default=None)
 788                 formats.append({
 789                     'url': video_url,
 790                     'ext': ext,
 791                     'width': int_or_none(self._RESOLUTIONS.get(height)),
 792                     'height': int_or_none(height),
 793                     'format_id': f'http-{ext}',
 794                 })
 795
 796             info.update({
 797                 'title': video_data.get('title') or traverse_obj(nbc_data, (
 798                     'dataLayer', (None, 'adobe'), ('contenttitle', 'title', 'prop22')), get_all=False),
 799                 'description':
 800                     traverse_obj(video_data, 'summary', 'excerpt', 'video_hero_text')
 801                     or clean_html(traverse_obj(nbc_data, ('dataLayer', 'summary'))),
 802                 'timestamp': unified_timestamp(date_string),
 803             })
 804
 805         smil = None
 806         if player_id and fw_ssid:
 807             smil = self._download_xml(
 808                 f'https://link.theplatform.com/s/{pdk_acct}/{player_id}', video_id,
 809                 note='Downloading SMIL data', query=query, fatal=is_live)
 810             if not isinstance(smil, xml.etree.ElementTree.Element):
 811                 smil = None
 812         subtitles = self._parse_smil_subtitles(smil, default_ns) if smil is not None else {}
 813         for video in smil.findall(self._xpath_ns('.//video', default_ns)) if smil is not None else []:
 814             info['duration'] = float_or_none(remove_end(video.get('dur'), 'ms'), 1000)
 815             video_src_url = video.get('src')
 816             ext = mimetype2ext(video.get('type'), default=determine_ext(video_src_url))
 817             if ext == 'm3u8':
 818                 fmts, subs = self._extract_m3u8_formats_and_subtitles(
 819                     video_src_url, video_id, 'mp4', m3u8_id='hls', fatal=is_live,
 820                     live=is_live, errnote='No HLS formats found')
 821                 formats.extend(fmts)
 822                 self._merge_subtitles(subs, target=subtitles)
 823             elif video_src_url:
 824                 formats.append({
 825                     'url': video_src_url,
 826                     'format_id': f'https-{ext}',
 827                     'ext': ext,
 828                     'width': int_or_none(video.get('width')),
 829                     'height': int_or_none(video.get('height')),
 830                 })
 831
 832         if not formats:
 833             self.raise_no_formats('No video content found in webpage', expected=True)
 834         elif is_live:
 835             try:
 836                 self._request_webpage(
 837                     HEADRequest(formats[0]['url']), video_id, note='Checking live status')
 838             except ExtractorError:
 839                 raise UserNotLive(video_id=channel)
 840
 841         return {
 842             'id': video_id,
 843             'channel': channel,
 844             'channel_id': nbc_data.get('callLetters'),
 845             'uploader': nbc_data.get('on_air_name'),
 846             'formats': formats,
 847             'subtitles': subtitles,
 848             'is_live': is_live,
 849             **info,
 850         }