youtube_dl/extractor/nbc.py

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