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 ..utils import (
  10     ExtractorError,
  11     HEADRequest,
  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             'expected_warnings': ['Ignoring subtitle tracks'],
 135             'params': {
 136                 'skip_download': 'm3u8',
 137             },
 138         },
 139         {
 140             'url': 'https://www.nbc.com/classic-tv/charles-in-charge/video/charles-in-charge-pilot/n3310',
 141             'only_matching': True,
 142         },
 143         {
 144             # Percent escaped url
 145             'url': 'https://www.nbc.com/up-all-night/video/day-after-valentine%27s-day/n2189',
 146             'only_matching': True,
 147         }
 148     ]
 149
 150     def _real_extract(self, url):
 151         permalink, video_id = self._match_valid_url(url).groups()
 152         permalink = 'http' + compat_urllib_parse_unquote(permalink)
 153         video_data = self._download_json(
 154             'https://friendship.nbc.co/v2/graphql', video_id, query={
 155                 'query': '''query bonanzaPage(
 156   $app: NBCUBrands! = nbc
 157   $name: String!
 158   $oneApp: Boolean
 159   $platform: SupportedPlatforms! = web
 160   $type: EntityPageType! = VIDEO
 161   $userId: String!
 162 ) {
 163   bonanzaPage(
 164     app: $app
 165     name: $name
 166     oneApp: $oneApp
 167     platform: $platform
 168     type: $type
 169     userId: $userId
 170   ) {
 171     metadata {
 172       ... on VideoPageData {
 173         description
 174         episodeNumber
 175         keywords
 176         locked
 177         mpxAccountId
 178         mpxGuid
 179         rating
 180         resourceId
 181         seasonNumber
 182         secondaryTitle
 183         seriesShortTitle
 184       }
 185     }
 186   }
 187 }''',
 188                 'variables': json.dumps({
 189                     'name': permalink,
 190                     'oneApp': True,
 191                     'userId': '0',
 192                 }),
 193             })['data']['bonanzaPage']['metadata']
 194         query = {
 195             'mbr': 'true',
 196             'manifest': 'm3u',
 197             'switch': 'HLSServiceSecure',
 198         }
 199         video_id = video_data['mpxGuid']
 200         tp_path = 'NnzsPC/media/guid/%s/%s' % (video_data.get('mpxAccountId') or '2410887629', video_id)
 201         tpm = self._download_theplatform_metadata(tp_path, video_id)
 202         title = tpm.get('title') or video_data.get('secondaryTitle')
 203         if video_data.get('locked'):
 204             resource = self._get_mvpd_resource(
 205                 video_data.get('resourceId') or 'nbcentertainment',
 206                 title, video_id, video_data.get('rating'))
 207             query['auth'] = self._extract_mvpd_auth(
 208                 url, video_id, 'nbcentertainment', resource)
 209         theplatform_url = smuggle_url(update_url_query(
 210             'http://link.theplatform.com/s/NnzsPC/media/guid/%s/%s' % (video_data.get('mpxAccountId') or '2410887629', video_id),
 211             query), {'force_smil_url': True})
 212
 213         # Empty string or 0 can be valid values for these. So the check must be `is None`
 214         description = video_data.get('description')
 215         if description is None:
 216             description = tpm.get('description')
 217         episode_number = int_or_none(video_data.get('episodeNumber'))
 218         if episode_number is None:
 219             episode_number = int_or_none(tpm.get('nbcu$airOrder'))
 220         rating = video_data.get('rating')
 221         if rating is None:
 222             try_get(tpm, lambda x: x['ratings'][0]['rating'])
 223         season_number = int_or_none(video_data.get('seasonNumber'))
 224         if season_number is None:
 225             season_number = int_or_none(tpm.get('nbcu$seasonNumber'))
 226         series = video_data.get('seriesShortTitle')
 227         if series is None:
 228             series = tpm.get('nbcu$seriesShortTitle')
 229         tags = video_data.get('keywords')
 230         if tags is None or len(tags) == 0:
 231             tags = tpm.get('keywords')
 232
 233         return {
 234             '_type': 'url_transparent',
 235             'age_limit': parse_age_limit(rating),
 236             'description': description,
 237             'episode': title,
 238             'episode_number': episode_number,
 239             'id': video_id,
 240             'ie_key': 'ThePlatform',
 241             'season_number': season_number,
 242             'series': series,
 243             'tags': tags,
 244             'title': title,
 245             'url': theplatform_url,
 246         }
 247
 248
 249 class NBCSportsVPlayerIE(InfoExtractor):
 250     _VALID_URL_BASE = r'https?://(?:vplayer\.nbcsports\.com|(?:www\.)?nbcsports\.com/vplayer)/'
 251     _VALID_URL = _VALID_URL_BASE + r'(?:[^/]+/)+(?P<id>[0-9a-zA-Z_]+)'
 252     _EMBED_REGEX = [r'(?:iframe[^>]+|var video|div[^>]+data-(?:mpx-)?)[sS]rc\s?=\s?"(?P<url>%s[^\"]+)' % _VALID_URL_BASE]
 253
 254     _TESTS = [{
 255         'url': 'https://vplayer.nbcsports.com/p/BxmELC/nbcsports_embed/select/9CsDKds0kvHI',
 256         'info_dict': {
 257             'id': '9CsDKds0kvHI',
 258             'ext': 'mp4',
 259             'description': 'md5:df390f70a9ba7c95ff1daace988f0d8d',
 260             'title': 'Tyler Kalinoski hits buzzer-beater to lift Davidson',
 261             'timestamp': 1426270238,
 262             'upload_date': '20150313',
 263             'uploader': 'NBCU-SPORTS',
 264             'duration': 72.818,
 265             'chapters': [],
 266             'thumbnail': r're:^https?://.*\.jpg$'
 267         }
 268     }, {
 269         'url': 'https://vplayer.nbcsports.com/p/BxmELC/nbcsports_embed/select/media/PEgOtlNcC_y2',
 270         'only_matching': True,
 271     }, {
 272         'url': 'https://www.nbcsports.com/vplayer/p/BxmELC/nbcsports/select/PHJSaFWbrTY9?form=html&autoPlay=true',
 273         'only_matching': True,
 274     }]
 275
 276     def _real_extract(self, url):
 277         video_id = self._match_id(url)
 278         webpage = self._download_webpage(url, video_id)
 279         theplatform_url = self._html_search_regex(r'tp:releaseUrl="(.+?)"', webpage, 'url')
 280         return self.url_result(theplatform_url, 'ThePlatform')
 281
 282
 283 class NBCSportsIE(InfoExtractor):
 284     _VALID_URL = r'https?://(?:www\.)?nbcsports\.com//?(?!vplayer/)(?:[^/]+/)+(?P<id>[0-9a-z-]+)'
 285
 286     _TESTS = [{
 287         # iframe src
 288         'url': 'http://www.nbcsports.com//college-basketball/ncaab/tom-izzo-michigan-st-has-so-much-respect-duke',
 289         'info_dict': {
 290             'id': 'PHJSaFWbrTY9',
 291             'ext': 'mp4',
 292             'title': 'Tom Izzo, Michigan St. has \'so much respect\' for Duke',
 293             'description': 'md5:ecb459c9d59e0766ac9c7d5d0eda8113',
 294             'uploader': 'NBCU-SPORTS',
 295             'upload_date': '20150330',
 296             'timestamp': 1427726529,
 297             'chapters': [],
 298             'thumbnail': 'https://hdliveextra-a.akamaihd.net/HD/image_sports/NBCU_Sports_Group_-_nbcsports/253/303/izzodps.jpg',
 299             'duration': 528.395,
 300         }
 301     }, {
 302         # data-mpx-src
 303         'url': 'https://www.nbcsports.com/philadelphia/philadelphia-phillies/bruce-bochy-hector-neris-hes-idiot',
 304         'only_matching': True,
 305     }, {
 306         # data-src
 307         'url': 'https://www.nbcsports.com/boston/video/report-card-pats-secondary-no-match-josh-allen',
 308         'only_matching': True,
 309     }]
 310
 311     def _real_extract(self, url):
 312         video_id = self._match_id(url)
 313         webpage = self._download_webpage(url, video_id)
 314         return self.url_result(
 315             NBCSportsVPlayerIE._extract_url(webpage), 'NBCSportsVPlayer')
 316
 317
 318 class NBCSportsStreamIE(AdobePassIE):
 319     _VALID_URL = r'https?://stream\.nbcsports\.com/.+?\bpid=(?P<id>\d+)'
 320     _TEST = {
 321         'url': 'http://stream.nbcsports.com/nbcsn/generic?pid=206559',
 322         'info_dict': {
 323             'id': '206559',
 324             'ext': 'mp4',
 325             'title': 'Amgen Tour of California Women\'s Recap',
 326             'description': 'md5:66520066b3b5281ada7698d0ea2aa894',
 327         },
 328         'params': {
 329             # m3u8 download
 330             'skip_download': True,
 331         },
 332         'skip': 'Requires Adobe Pass Authentication',
 333     }
 334
 335     def _real_extract(self, url):
 336         video_id = self._match_id(url)
 337         live_source = self._download_json(
 338             'http://stream.nbcsports.com/data/live_sources_%s.json' % video_id,
 339             video_id)
 340         video_source = live_source['videoSources'][0]
 341         title = video_source['title']
 342         source_url = None
 343         for k in ('source', 'msl4source', 'iossource', 'hlsv4'):
 344             sk = k + 'Url'
 345             source_url = video_source.get(sk) or video_source.get(sk + 'Alt')
 346             if source_url:
 347                 break
 348         else:
 349             source_url = video_source['ottStreamUrl']
 350         is_live = video_source.get('type') == 'live' or video_source.get('status') == 'Live'
 351         resource = self._get_mvpd_resource('nbcsports', title, video_id, '')
 352         token = self._extract_mvpd_auth(url, video_id, 'nbcsports', resource)
 353         tokenized_url = self._download_json(
 354             'https://token.playmakerservices.com/cdn',
 355             video_id, data=json.dumps({
 356                 'requestorId': 'nbcsports',
 357                 'pid': video_id,
 358                 'application': 'NBCSports',
 359                 'version': 'v1',
 360                 'platform': 'desktop',
 361                 'cdn': 'akamai',
 362                 'url': video_source['sourceUrl'],
 363                 'token': base64.b64encode(token.encode()).decode(),
 364                 'resourceId': base64.b64encode(resource.encode()).decode(),
 365             }).encode())['tokenizedUrl']
 366         formats = self._extract_m3u8_formats(tokenized_url, video_id, 'mp4')
 367         return {
 368             'id': video_id,
 369             'title': title,
 370             'description': live_source.get('description'),
 371             'formats': formats,
 372             'is_live': is_live,
 373         }
 374
 375
 376 class NBCNewsIE(ThePlatformIE):  # XXX: Do not subclass from concrete IE
 377     _VALID_URL = r'(?x)https?://(?:www\.)?(?:nbcnews|today|msnbc)\.com/([^/]+/)*(?:.*-)?(?P<id>[^/?]+)'
 378     _EMBED_REGEX = [r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//www\.nbcnews\.com/widget/video-embed/[^"\']+)\1']
 379
 380     _TESTS = [
 381         {
 382             'url': 'http://www.nbcnews.com/watch/nbcnews-com/how-twitter-reacted-to-the-snowden-interview-269389891880',
 383             'md5': 'cf4bc9e6ce0130f00f545d80ecedd4bf',
 384             'info_dict': {
 385                 'id': '269389891880',
 386                 'ext': 'mp4',
 387                 'title': 'How Twitter Reacted To The Snowden Interview',
 388                 'description': 'md5:65a0bd5d76fe114f3c2727aa3a81fe64',
 389                 'timestamp': 1401363060,
 390                 'upload_date': '20140529',
 391             },
 392         },
 393         {
 394             'url': 'http://www.nbcnews.com/feature/dateline-full-episodes/full-episode-family-business-n285156',
 395             'md5': 'fdbf39ab73a72df5896b6234ff98518a',
 396             'info_dict': {
 397                 'id': '529953347624',
 398                 'ext': 'mp4',
 399                 'title': 'FULL EPISODE: Family Business',
 400                 'description': 'md5:757988edbaae9d7be1d585eb5d55cc04',
 401             },
 402             'skip': 'This page is unavailable.',
 403         },
 404         {
 405             'url': 'http://www.nbcnews.com/nightly-news/video/nightly-news-with-brian-williams-full-broadcast-february-4-394064451844',
 406             'md5': '8eb831eca25bfa7d25ddd83e85946548',
 407             'info_dict': {
 408                 'id': '394064451844',
 409                 'ext': 'mp4',
 410                 'title': 'Nightly News with Brian Williams Full Broadcast (February 4)',
 411                 'description': 'md5:1c10c1eccbe84a26e5debb4381e2d3c5',
 412                 'timestamp': 1423104900,
 413                 'upload_date': '20150205',
 414             },
 415         },
 416         {
 417             'url': 'http://www.nbcnews.com/business/autos/volkswagen-11-million-vehicles-could-have-suspect-software-emissions-scandal-n431456',
 418             'md5': '4a8c4cec9e1ded51060bdda36ff0a5c0',
 419             'info_dict': {
 420                 'id': 'n431456',
 421                 'ext': 'mp4',
 422                 'title': "Volkswagen U.S. Chief:  We 'Totally Screwed Up'",
 423                 'description': 'md5:d22d1281a24f22ea0880741bb4dd6301',
 424                 'upload_date': '20150922',
 425                 'timestamp': 1442917800,
 426             },
 427         },
 428         {
 429             'url': 'http://www.today.com/video/see-the-aurora-borealis-from-space-in-stunning-new-nasa-video-669831235788',
 430             'md5': '118d7ca3f0bea6534f119c68ef539f71',
 431             'info_dict': {
 432                 'id': '669831235788',
 433                 'ext': 'mp4',
 434                 'title': 'See the aurora borealis from space in stunning new NASA video',
 435                 'description': 'md5:74752b7358afb99939c5f8bb2d1d04b1',
 436                 'upload_date': '20160420',
 437                 'timestamp': 1461152093,
 438             },
 439         },
 440         {
 441             'url': 'http://www.msnbc.com/all-in-with-chris-hayes/watch/the-chaotic-gop-immigration-vote-314487875924',
 442             'md5': '6d236bf4f3dddc226633ce6e2c3f814d',
 443             'info_dict': {
 444                 'id': '314487875924',
 445                 'ext': 'mp4',
 446                 'title': 'The chaotic GOP immigration vote',
 447                 '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.',
 448                 'thumbnail': r're:^https?://.*\.jpg$',
 449                 'timestamp': 1406937606,
 450                 'upload_date': '20140802',
 451             },
 452         },
 453         {
 454             'url': 'http://www.nbcnews.com/watch/dateline/full-episode--deadly-betrayal-386250819952',
 455             'only_matching': True,
 456         },
 457         {
 458             # From http://www.vulture.com/2016/06/letterman-couldnt-care-less-about-late-night.html
 459             'url': 'http://www.nbcnews.com/widget/video-embed/701714499682',
 460             'only_matching': True,
 461         },
 462     ]
 463
 464     def _real_extract(self, url):
 465         video_id = self._match_id(url)
 466         webpage = self._download_webpage(url, video_id)
 467
 468         data = self._search_nextjs_data(webpage, video_id)['props']['initialState']
 469         video_data = try_get(data, lambda x: x['video']['current'], dict)
 470         if not video_data:
 471             video_data = data['article']['content'][0]['primaryMedia']['video']
 472         title = video_data['headline']['primary']
 473
 474         formats = []
 475         for va in video_data.get('videoAssets', []):
 476             public_url = va.get('publicUrl')
 477             if not public_url:
 478                 continue
 479             if '://link.theplatform.com/' in public_url:
 480                 public_url = update_url_query(public_url, {'format': 'redirect'})
 481             format_id = va.get('format')
 482             if format_id == 'M3U':
 483                 formats.extend(self._extract_m3u8_formats(
 484                     public_url, video_id, 'mp4', 'm3u8_native',
 485                     m3u8_id=format_id, fatal=False))
 486                 continue
 487             tbr = int_or_none(va.get('bitrate'), 1000)
 488             if tbr:
 489                 format_id += '-%d' % tbr
 490             formats.append({
 491                 'format_id': format_id,
 492                 'url': public_url,
 493                 'width': int_or_none(va.get('width')),
 494                 'height': int_or_none(va.get('height')),
 495                 'tbr': tbr,
 496                 'ext': 'mp4',
 497             })
 498
 499         subtitles = {}
 500         closed_captioning = video_data.get('closedCaptioning')
 501         if closed_captioning:
 502             for cc_url in closed_captioning.values():
 503                 if not cc_url:
 504                     continue
 505                 subtitles.setdefault('en', []).append({
 506                     'url': cc_url,
 507                 })
 508
 509         return {
 510             'id': video_id,
 511             'title': title,
 512             'description': try_get(video_data, lambda x: x['description']['primary']),
 513             'thumbnail': try_get(video_data, lambda x: x['primaryImage']['url']['primary']),
 514             'duration': parse_duration(video_data.get('duration')),
 515             'timestamp': unified_timestamp(video_data.get('datePublished')),
 516             'formats': formats,
 517             'subtitles': subtitles,
 518         }
 519
 520
 521 class NBCOlympicsIE(InfoExtractor):
 522     IE_NAME = 'nbcolympics'
 523     _VALID_URL = r'https?://www\.nbcolympics\.com/videos?/(?P<id>[0-9a-z-]+)'
 524
 525     _TEST = {
 526         # Geo-restricted to US
 527         'url': 'http://www.nbcolympics.com/video/justin-roses-son-leo-was-tears-after-his-dad-won-gold',
 528         'md5': '54fecf846d05429fbaa18af557ee523a',
 529         'info_dict': {
 530             'id': 'WjTBzDXx5AUq',
 531             'display_id': 'justin-roses-son-leo-was-tears-after-his-dad-won-gold',
 532             'ext': 'mp4',
 533             'title': 'Rose\'s son Leo was in tears after his dad won gold',
 534             'description': 'Olympic gold medalist Justin Rose gets emotional talking to the impact his win in men\'s golf has already had on his children.',
 535             'timestamp': 1471274964,
 536             'upload_date': '20160815',
 537             'uploader': 'NBCU-SPORTS',
 538         },
 539     }
 540
 541     def _real_extract(self, url):
 542         display_id = self._match_id(url)
 543
 544         webpage = self._download_webpage(url, display_id)
 545
 546         try:
 547             drupal_settings = self._parse_json(self._search_regex(
 548                 r'jQuery\.extend\(Drupal\.settings\s*,\s*({.+?})\);',
 549                 webpage, 'drupal settings'), display_id)
 550
 551             iframe_url = drupal_settings['vod']['iframe_url']
 552             theplatform_url = iframe_url.replace(
 553                 'vplayer.nbcolympics.com', 'player.theplatform.com')
 554         except RegexNotFoundError:
 555             theplatform_url = self._search_regex(
 556                 r"([\"'])embedUrl\1: *([\"'])(?P<embedUrl>.+)\2",
 557                 webpage, 'embedding URL', group="embedUrl")
 558
 559         return {
 560             '_type': 'url_transparent',
 561             'url': theplatform_url,
 562             'ie_key': ThePlatformIE.ie_key(),
 563             'display_id': display_id,
 564         }
 565
 566
 567 class NBCOlympicsStreamIE(AdobePassIE):
 568     IE_NAME = 'nbcolympics:stream'
 569     _VALID_URL = r'https?://stream\.nbcolympics\.com/(?P<id>[0-9a-z-]+)'
 570     _TESTS = [
 571         {
 572             'note': 'Tokenized m3u8 source URL',
 573             'url': 'https://stream.nbcolympics.com/womens-soccer-group-round-11',
 574             'info_dict': {
 575                 'id': '2019740',
 576                 'ext': 'mp4',
 577                 '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}$",
 578             },
 579             'params': {
 580                 'skip_download': 'm3u8',
 581             },
 582         }, {
 583             'note': 'Plain m3u8 source URL',
 584             'url': 'https://stream.nbcolympics.com/gymnastics-event-finals-mens-floor-pommel-horse-womens-vault-bars',
 585             'info_dict': {
 586                 'id': '2021729',
 587                 'ext': 'mp4',
 588                 'title': r're:Event Finals: M Floor, W Vault, M Pommel, W Uneven Bars [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
 589             },
 590             'params': {
 591                 'skip_download': 'm3u8',
 592             },
 593         },
 594     ]
 595
 596     def _real_extract(self, url):
 597         display_id = self._match_id(url)
 598         webpage = self._download_webpage(url, display_id)
 599         pid = self._search_regex(r'pid\s*=\s*(\d+);', webpage, 'pid')
 600
 601         event_config = self._download_json(
 602             f'http://stream.nbcolympics.com/data/event_config_{pid}.json',
 603             pid, 'Downloading event config')['eventConfig']
 604
 605         title = event_config['eventTitle']
 606         is_live = {'live': True, 'replay': False}.get(event_config.get('eventStatus'))
 607
 608         source_url = self._download_json(
 609             f'https://api-leap.nbcsports.com/feeds/assets/{pid}?application=NBCOlympics&platform=desktop&format=nbc-player&env=staging',
 610             pid, 'Downloading leap config'
 611         )['videoSources'][0]['cdnSources']['primary'][0]['sourceUrl']
 612
 613         if event_config.get('cdnToken'):
 614             ap_resource = self._get_mvpd_resource(
 615                 event_config.get('resourceId', 'NBCOlympics'),
 616                 re.sub(r'[^\w\d ]+', '', event_config['eventTitle']), pid,
 617                 event_config.get('ratingId', 'NO VALUE'))
 618             media_token = self._extract_mvpd_auth(url, pid, event_config.get('requestorId', 'NBCOlympics'), ap_resource)
 619
 620             source_url = self._download_json(
 621                 'https://tokens.playmakerservices.com/', pid, 'Retrieving tokenized URL',
 622                 data=json.dumps({
 623                     'application': 'NBCSports',
 624                     'authentication-type': 'adobe-pass',
 625                     'cdn': 'akamai',
 626                     'pid': pid,
 627                     'platform': 'desktop',
 628                     'requestorId': 'NBCOlympics',
 629                     'resourceId': base64.b64encode(ap_resource.encode()).decode(),
 630                     'token': base64.b64encode(media_token.encode()).decode(),
 631                     'url': source_url,
 632                     'version': 'v1',
 633                 }).encode(),
 634             )['akamai'][0]['tokenizedUrl']
 635
 636         formats = self._extract_m3u8_formats(source_url, pid, 'mp4', live=is_live)
 637         for f in formats:
 638             # -http_seekable requires ffmpeg 4.3+ but it doesnt seem possible to
 639             # download with ffmpeg without this option
 640             f['downloader_options'] = {'ffmpeg_args': ['-seekable', '0', '-http_seekable', '0', '-icy', '0']}
 641
 642         return {
 643             'id': pid,
 644             'display_id': display_id,
 645             'title': title,
 646             'formats': formats,
 647             'is_live': is_live,
 648         }
 649
 650
 651 class NBCStationsIE(InfoExtractor):
 652     _DOMAIN_RE = '|'.join(map(re.escape, (
 653         'nbcbayarea', 'nbcboston', 'nbcchicago', 'nbcconnecticut', 'nbcdfw', 'nbclosangeles',
 654         'nbcmiami', 'nbcnewyork', 'nbcphiladelphia', 'nbcsandiego', 'nbcwashington',
 655         'necn', 'telemundo52', 'telemundoarizona', 'telemundochicago', 'telemundonuevainglaterra',
 656     )))
 657     _VALID_URL = rf'https?://(?:www\.)?(?P<site>{_DOMAIN_RE})\.com/(?:[^/?#]+/)*(?P<id>[^/?#]+)/?(?:$|[#?])'
 658
 659     _TESTS = [{
 660         'url': 'https://www.nbclosangeles.com/news/local/large-structure-fire-in-downtown-la-prompts-smoke-odor-advisory/2968618/',
 661         'info_dict': {
 662             'id': '2968618',
 663             'ext': 'mp4',
 664             'title': 'Large Structure Fire in Downtown LA Prompts Smoke Odor Advisory',
 665             'description': 'md5:417ed3c2d91fe9d301e6db7b0942f182',
 666             'duration': 112.513,
 667             'timestamp': 1661135892,
 668             'upload_date': '20220822',
 669             'uploader': 'NBC 4',
 670             'channel_id': 'KNBC',
 671             'channel': 'nbclosangeles',
 672         },
 673         'params': {
 674             'skip_download': 'm3u8',
 675         },
 676     }, {
 677         'url': 'https://www.telemundoarizona.com/responde/huracan-complica-reembolso-para-televidente-de-tucson/2247002/',
 678         'info_dict': {
 679             'id': '2247002',
 680             'ext': 'mp4',
 681             'title': 'Huracán complica que televidente de Tucson reciba  reembolso',
 682             'description': 'md5:af298dc73aab74d4fca6abfb12acb6cf',
 683             'duration': 172.406,
 684             'timestamp': 1660886507,
 685             'upload_date': '20220819',
 686             'uploader': 'Telemundo Arizona',
 687             'channel_id': 'KTAZ',
 688             'channel': 'telemundoarizona',
 689         },
 690         'params': {
 691             'skip_download': 'm3u8',
 692         },
 693     }, {
 694         # direct mp4 link
 695         'url': 'https://www.nbcboston.com/weather/video-weather/highs-near-freezing-in-boston-on-wednesday/2961135/',
 696         'md5': '9bf8c41dc7abbb75b1a44f1491a4cc85',
 697         'info_dict': {
 698             'id': '2961135',
 699             'ext': 'mp4',
 700             'title': 'Highs Near Freezing in Boston on Wednesday',
 701             'description': 'md5:3ec486609a926c99f00a3512e6c0e85b',
 702             'duration': 235.669,
 703             'timestamp': 1675268656,
 704             'upload_date': '20230201',
 705             'uploader': '',
 706             'channel_id': 'WBTS',
 707             'channel': 'nbcboston',
 708         },
 709     }]
 710
 711     _RESOLUTIONS = {
 712         '1080': '1920',
 713         '720': '1280',
 714         '540': '960',
 715         '360': '640',
 716         '234': '416',
 717     }
 718
 719     def _real_extract(self, url):
 720         channel, video_id = self._match_valid_url(url).group('site', 'id')
 721         webpage = self._download_webpage(url, video_id)
 722
 723         nbc_data = self._search_json(
 724             r'<script>\s*var\s+nbc\s*=', webpage, 'NBC JSON data', video_id)
 725         pdk_acct = nbc_data.get('pdkAcct') or 'Yh1nAC'
 726         fw_ssid = traverse_obj(nbc_data, ('video', 'fwSSID'))
 727
 728         video_data = self._search_json(
 729             r'data-videos="\[', webpage, 'video data', video_id, default={}, transform_source=unescapeHTML)
 730         video_data.update(self._search_json(
 731             r'data-meta="', webpage, 'metadata', video_id, default={}, transform_source=unescapeHTML))
 732         if not video_data:
 733             raise ExtractorError('No video metadata found in webpage', expected=True)
 734
 735         info, formats = {}, []
 736         is_live = int_or_none(video_data.get('mpx_is_livestream')) == 1
 737         query = {
 738             'formats': 'MPEG-DASH none,M3U none,MPEG-DASH none,MPEG4,MP3',
 739             'format': 'SMIL',
 740             'fwsitesection': fw_ssid,
 741             'fwNetworkID': traverse_obj(nbc_data, ('video', 'fwNetworkID'), default='382114'),
 742             'pprofile': 'ots_desktop_html',
 743             'sensitive': 'false',
 744             'w': '1920',
 745             'h': '1080',
 746             'mode': 'LIVE' if is_live else 'on-demand',
 747             'vpaid': 'script',
 748             'schema': '2.0',
 749             'sdk': 'PDK 6.1.3',
 750         }
 751
 752         if is_live:
 753             player_id = traverse_obj(video_data, ((None, ('video', 'meta')), (
 754                 'mpx_m3upid', 'mpx_pid', 'pid_streaming_web_medium')), get_all=False)
 755             info['title'] = f'{channel} livestream'
 756
 757         else:
 758             player_id = traverse_obj(video_data, (
 759                 (None, ('video', 'meta')), ('pid_streaming_web_high', 'mpx_pid')), get_all=False)
 760
 761             date_string = traverse_obj(video_data, 'date_string', 'date_gmt')
 762             if date_string:
 763                 date_string = self._search_regex(
 764                     r'datetime="([^"]+)"', date_string, 'date string', fatal=False)
 765             else:
 766                 date_string = traverse_obj(
 767                     nbc_data, ('dataLayer', 'adobe', ('prop70', 'eVar70', 'eVar59')), get_all=False)
 768
 769             video_url = traverse_obj(video_data, ((None, ('video', 'meta')), 'mp4_url'), get_all=False)
 770             if video_url:
 771                 ext = determine_ext(video_url)
 772                 height = self._search_regex(r'\d+-(\d+)p', url_basename(video_url), 'height', default=None)
 773                 formats.append({
 774                     'url': video_url,
 775                     'ext': ext,
 776                     'width': int_or_none(self._RESOLUTIONS.get(height)),
 777                     'height': int_or_none(height),
 778                     'format_id': f'http-{ext}',
 779                 })
 780
 781             info.update({
 782                 'title': video_data.get('title') or traverse_obj(nbc_data, (
 783                     'dataLayer', (None, 'adobe'), ('contenttitle', 'title', 'prop22')), get_all=False),
 784                 'description':
 785                     traverse_obj(video_data, 'summary', 'excerpt', 'video_hero_text')
 786                     or clean_html(traverse_obj(nbc_data, ('dataLayer', 'summary'))),
 787                 'timestamp': unified_timestamp(date_string),
 788             })
 789
 790         smil = None
 791         if player_id and fw_ssid:
 792             smil = self._download_xml(
 793                 f'https://link.theplatform.com/s/{pdk_acct}/{player_id}', video_id,
 794                 note='Downloading SMIL data', query=query, fatal=is_live)
 795         subtitles = self._parse_smil_subtitles(smil, default_ns) if smil else {}
 796         for video in smil.findall(self._xpath_ns('.//video', default_ns)) if smil else []:
 797             info['duration'] = float_or_none(remove_end(video.get('dur'), 'ms'), 1000)
 798             video_src_url = video.get('src')
 799             ext = mimetype2ext(video.get('type'), default=determine_ext(video_src_url))
 800             if ext == 'm3u8':
 801                 fmts, subs = self._extract_m3u8_formats_and_subtitles(
 802                     video_src_url, video_id, 'mp4', m3u8_id='hls', fatal=is_live,
 803                     live=is_live, errnote='No HLS formats found')
 804                 formats.extend(fmts)
 805                 self._merge_subtitles(subs, target=subtitles)
 806             elif video_src_url:
 807                 formats.append({
 808                     'url': video_src_url,
 809                     'format_id': f'https-{ext}',
 810                     'ext': ext,
 811                     'width': int_or_none(video.get('width')),
 812                     'height': int_or_none(video.get('height')),
 813                 })
 814
 815         if not formats:
 816             self.raise_no_formats('No video content found in webpage', expected=True)
 817         elif is_live:
 818             try:
 819                 self._request_webpage(
 820                     HEADRequest(formats[0]['url']), video_id, note='Checking live status')
 821             except ExtractorError:
 822                 raise UserNotLive(video_id=channel)
 823
 824         return {
 825             'id': video_id,
 826             'channel': channel,
 827             'channel_id': nbc_data.get('callLetters'),
 828             'uploader': nbc_data.get('on_air_name'),
 829             'formats': formats,
 830             'subtitles': subtitles,
 831             'is_live': is_live,
 832             **info,
 833         }