Raphaël G. Git Repositories - youtubedl/blob - 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     js_to_json,
  14     parse_duration,
  15     smuggle_url,
  16     try_get,
  17     unified_timestamp,
  18     update_url_query,
  19 )
  20
  21
  22 class NBCIE(AdobePassIE):
  23     _VALID_URL = r'https?(?P<permalink>://(?:www\.)?nbc\.com/(?:classic-tv/)?[^/]+/video/[^/]+/(?P<id>n?\d+))'
  24
  25     _TESTS = [
  26         {
  27             'url': 'http://www.nbc.com/the-tonight-show/video/jimmy-fallon-surprises-fans-at-ben-jerrys/2848237',
  28             'info_dict': {
  29                 'id': '2848237',
  30                 'ext': 'mp4',
  31                 'title': 'Jimmy Fallon Surprises Fans at Ben & Jerry\'s',
  32                 'description': 'Jimmy gives out free scoops of his new "Tonight Dough" ice cream flavor by surprising customers at the Ben & Jerry\'s scoop shop.',
  33                 'timestamp': 1424246400,
  34                 'upload_date': '20150218',
  35                 'uploader': 'NBCU-COM',
  36             },
  37             'params': {
  38                 # m3u8 download
  39                 'skip_download': True,
  40             },
  41         },
  42         {
  43             'url': 'http://www.nbc.com/saturday-night-live/video/star-wars-teaser/2832821',
  44             'info_dict': {
  45                 'id': '2832821',
  46                 'ext': 'mp4',
  47                 'title': 'Star Wars Teaser',
  48                 'description': 'md5:0b40f9cbde5b671a7ff62fceccc4f442',
  49                 'timestamp': 1417852800,
  50                 'upload_date': '20141206',
  51                 'uploader': 'NBCU-COM',
  52             },
  53             'params': {
  54                 # m3u8 download
  55                 'skip_download': True,
  56             },
  57             'skip': 'Only works from US',
  58         },
  59         {
  60             # HLS streams requires the 'hdnea3' cookie
  61             'url': 'http://www.nbc.com/Kings/video/goliath/n1806',
  62             'info_dict': {
  63                 'id': '101528f5a9e8127b107e98c5e6ce4638',
  64                 'ext': 'mp4',
  65                 'title': 'Goliath',
  66                 '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.',
  67                 'timestamp': 1237100400,
  68                 'upload_date': '20090315',
  69                 'uploader': 'NBCU-COM',
  70             },
  71             'params': {
  72                 'skip_download': True,
  73             },
  74             'skip': 'Only works from US',
  75         },
  76         {
  77             'url': 'https://www.nbc.com/classic-tv/charles-in-charge/video/charles-in-charge-pilot/n3310',
  78             'only_matching': True,
  79         },
  80         {
  81             # Percent escaped url
  82             'url': 'https://www.nbc.com/up-all-night/video/day-after-valentine%27s-day/n2189',
  83             'only_matching': True,
  84         }
  85     ]
  86
  87     def _real_extract(self, url):
  88         permalink, video_id = re.match(self._VALID_URL, url).groups()
  89         permalink = 'http' + compat_urllib_parse_unquote(permalink)
  90         video_data = self._download_json(
  91             'https://friendship.nbc.co/v2/graphql', video_id, query={
  92                 'query': '''query bonanzaPage(
  93   $app: NBCUBrands! = nbc
  94   $name: String!
  95   $oneApp: Boolean
  96   $platform: SupportedPlatforms! = web
  97   $type: EntityPageType! = VIDEO
  98   $userId: String!
  99 ) {
 100   bonanzaPage(
 101     app: $app
 102     name: $name
 103     oneApp: $oneApp
 104     platform: $platform
 105     type: $type
 106     userId: $userId
 107   ) {
 108     metadata {
 109       ... on VideoPageData {
 110         description
 111         episodeNumber
 112         keywords
 113         locked
 114         mpxAccountId
 115         mpxGuid
 116         rating
 117         resourceId
 118         seasonNumber
 119         secondaryTitle
 120         seriesShortTitle
 121       }
 122     }
 123   }
 124 }''',
 125                 'variables': json.dumps({
 126                     'name': permalink,
 127                     'oneApp': True,
 128                     'userId': '0',
 129                 }),
 130             })['data']['bonanzaPage']['metadata']
 131         query = {
 132             'mbr': 'true',
 133             'manifest': 'm3u',
 134         }
 135         video_id = video_data['mpxGuid']
 136         title = video_data['secondaryTitle']
 137         if video_data.get('locked'):
 138             resource = self._get_mvpd_resource(
 139                 video_data.get('resourceId') or 'nbcentertainment',
 140                 title, video_id, video_data.get('rating'))
 141             query['auth'] = self._extract_mvpd_auth(
 142                 url, video_id, 'nbcentertainment', resource)
 143         theplatform_url = smuggle_url(update_url_query(
 144             'http://link.theplatform.com/s/NnzsPC/media/guid/%s/%s' % (video_data.get('mpxAccountId') or '2410887629', video_id),
 145             query), {'force_smil_url': True})
 146         return {
 147             '_type': 'url_transparent',
 148             'id': video_id,
 149             'title': title,
 150             'url': theplatform_url,
 151             'description': video_data.get('description'),
 152             'tags': video_data.get('keywords'),
 153             'season_number': int_or_none(video_data.get('seasonNumber')),
 154             'episode_number': int_or_none(video_data.get('episodeNumber')),
 155             'episode': title,
 156             'series': video_data.get('seriesShortTitle'),
 157             'ie_key': 'ThePlatform',
 158         }
 159
 160
 161 class NBCSportsVPlayerIE(InfoExtractor):
 162     _VALID_URL = r'https?://vplayer\.nbcsports\.com/(?:[^/]+/)+(?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
 180     @staticmethod
 181     def _extract_url(webpage):
 182         iframe_m = re.search(
 183             r'<iframe[^>]+src="(?P<url>https?://vplayer\.nbcsports\.com/[^"]+)"', webpage)
 184         if iframe_m:
 185             return iframe_m.group('url')
 186
 187     def _real_extract(self, url):
 188         video_id = self._match_id(url)
 189         webpage = self._download_webpage(url, video_id)
 190         theplatform_url = self._og_search_video_url(webpage).replace(
 191             'vplayer.nbcsports.com', 'player.theplatform.com')
 192         return self.url_result(theplatform_url, 'ThePlatform')
 193
 194
 195 class NBCSportsIE(InfoExtractor):
 196     # Does not include https because its certificate is invalid
 197     _VALID_URL = r'https?://(?:www\.)?nbcsports\.com//?(?:[^/]+/)+(?P<id>[0-9a-z-]+)'
 198
 199     _TEST = {
 200         'url': 'http://www.nbcsports.com//college-basketball/ncaab/tom-izzo-michigan-st-has-so-much-respect-duke',
 201         'info_dict': {
 202             'id': 'PHJSaFWbrTY9',
 203             'ext': 'flv',
 204             'title': 'Tom Izzo, Michigan St. has \'so much respect\' for Duke',
 205             'description': 'md5:ecb459c9d59e0766ac9c7d5d0eda8113',
 206             'uploader': 'NBCU-SPORTS',
 207             'upload_date': '20150330',
 208             'timestamp': 1427726529,
 209         }
 210     }
 211
 212     def _real_extract(self, url):
 213         video_id = self._match_id(url)
 214         webpage = self._download_webpage(url, video_id)
 215         return self.url_result(
 216             NBCSportsVPlayerIE._extract_url(webpage), 'NBCSportsVPlayer')
 217
 218
 219 class NBCSportsStreamIE(AdobePassIE):
 220     _VALID_URL = r'https?://stream\.nbcsports\.com/.+?\bpid=(?P<id>\d+)'
 221     _TEST = {
 222         'url': 'http://stream.nbcsports.com/nbcsn/generic?pid=206559',
 223         'info_dict': {
 224             'id': '206559',
 225             'ext': 'mp4',
 226             'title': 'Amgen Tour of California Women\'s Recap',
 227             'description': 'md5:66520066b3b5281ada7698d0ea2aa894',
 228         },
 229         'params': {
 230             # m3u8 download
 231             'skip_download': True,
 232         },
 233         'skip': 'Requires Adobe Pass Authentication',
 234     }
 235
 236     def _real_extract(self, url):
 237         video_id = self._match_id(url)
 238         live_source = self._download_json(
 239             'http://stream.nbcsports.com/data/live_sources_%s.json' % video_id,
 240             video_id)
 241         video_source = live_source['videoSources'][0]
 242         title = video_source['title']
 243         source_url = None
 244         for k in ('source', 'msl4source', 'iossource', 'hlsv4'):
 245             sk = k + 'Url'
 246             source_url = video_source.get(sk) or video_source.get(sk + 'Alt')
 247             if source_url:
 248                 break
 249         else:
 250             source_url = video_source['ottStreamUrl']
 251         is_live = video_source.get('type') == 'live' or video_source.get('status') == 'Live'
 252         resource = self._get_mvpd_resource('nbcsports', title, video_id, '')
 253         token = self._extract_mvpd_auth(url, video_id, 'nbcsports', resource)
 254         tokenized_url = self._download_json(
 255             'https://token.playmakerservices.com/cdn',
 256             video_id, data=json.dumps({
 257                 'requestorId': 'nbcsports',
 258                 'pid': video_id,
 259                 'application': 'NBCSports',
 260                 'version': 'v1',
 261                 'platform': 'desktop',
 262                 'cdn': 'akamai',
 263                 'url': video_source['sourceUrl'],
 264                 'token': base64.b64encode(token.encode()).decode(),
 265                 'resourceId': base64.b64encode(resource.encode()).decode(),
 266             }).encode())['tokenizedUrl']
 267         formats = self._extract_m3u8_formats(tokenized_url, video_id, 'mp4')
 268         self._sort_formats(formats)
 269         return {
 270             'id': video_id,
 271             'title': self._live_title(title) if is_live else title,
 272             'description': live_source.get('description'),
 273             'formats': formats,
 274             'is_live': is_live,
 275         }
 276
 277
 278 class CSNNEIE(InfoExtractor):
 279     _VALID_URL = r'https?://(?:www\.)?csnne\.com/video/(?P<id>[0-9a-z-]+)'
 280
 281     _TEST = {
 282         'url': 'http://www.csnne.com/video/snc-evening-update-wright-named-red-sox-no-5-starter',
 283         'info_dict': {
 284             'id': 'yvBLLUgQ8WU0',
 285             'ext': 'mp4',
 286             'title': 'SNC evening update: Wright named Red Sox\' No. 5 starter.',
 287             'description': 'md5:1753cfee40d9352b19b4c9b3e589b9e3',
 288             'timestamp': 1459369979,
 289             'upload_date': '20160330',
 290             'uploader': 'NBCU-SPORTS',
 291         }
 292     }
 293
 294     def _real_extract(self, url):
 295         display_id = self._match_id(url)
 296         webpage = self._download_webpage(url, display_id)
 297         return {
 298             '_type': 'url_transparent',
 299             'ie_key': 'ThePlatform',
 300             'url': self._html_search_meta('twitter:player:stream', webpage),
 301             'display_id': display_id,
 302         }
 303
 304
 305 class NBCNewsIE(ThePlatformIE):
 306     _VALID_URL = r'(?x)https?://(?:www\.)?(?:nbcnews|today|msnbc)\.com/([^/]+/)*(?:.*-)?(?P<id>[^/?]+)'
 307
 308     _TESTS = [
 309         {
 310             'url': 'http://www.nbcnews.com/watch/nbcnews-com/how-twitter-reacted-to-the-snowden-interview-269389891880',
 311             'md5': 'cf4bc9e6ce0130f00f545d80ecedd4bf',
 312             'info_dict': {
 313                 'id': '269389891880',
 314                 'ext': 'mp4',
 315                 'title': 'How Twitter Reacted To The Snowden Interview',
 316                 'description': 'md5:65a0bd5d76fe114f3c2727aa3a81fe64',
 317                 'timestamp': 1401363060,
 318                 'upload_date': '20140529',
 319             },
 320         },
 321         {
 322             'url': 'http://www.nbcnews.com/feature/dateline-full-episodes/full-episode-family-business-n285156',
 323             'md5': 'fdbf39ab73a72df5896b6234ff98518a',
 324             'info_dict': {
 325                 'id': '529953347624',
 326                 'ext': 'mp4',
 327                 'title': 'FULL EPISODE: Family Business',
 328                 'description': 'md5:757988edbaae9d7be1d585eb5d55cc04',
 329             },
 330             'skip': 'This page is unavailable.',
 331         },
 332         {
 333             'url': 'http://www.nbcnews.com/nightly-news/video/nightly-news-with-brian-williams-full-broadcast-february-4-394064451844',
 334             'md5': '8eb831eca25bfa7d25ddd83e85946548',
 335             'info_dict': {
 336                 'id': '394064451844',
 337                 'ext': 'mp4',
 338                 'title': 'Nightly News with Brian Williams Full Broadcast (February 4)',
 339                 'description': 'md5:1c10c1eccbe84a26e5debb4381e2d3c5',
 340                 'timestamp': 1423104900,
 341                 'upload_date': '20150205',
 342             },
 343         },
 344         {
 345             'url': 'http://www.nbcnews.com/business/autos/volkswagen-11-million-vehicles-could-have-suspect-software-emissions-scandal-n431456',
 346             'md5': '4a8c4cec9e1ded51060bdda36ff0a5c0',
 347             'info_dict': {
 348                 'id': 'n431456',
 349                 'ext': 'mp4',
 350                 'title': "Volkswagen U.S. Chief:  We 'Totally Screwed Up'",
 351                 'description': 'md5:d22d1281a24f22ea0880741bb4dd6301',
 352                 'upload_date': '20150922',
 353                 'timestamp': 1442917800,
 354             },
 355         },
 356         {
 357             'url': 'http://www.today.com/video/see-the-aurora-borealis-from-space-in-stunning-new-nasa-video-669831235788',
 358             'md5': '118d7ca3f0bea6534f119c68ef539f71',
 359             'info_dict': {
 360                 'id': '669831235788',
 361                 'ext': 'mp4',
 362                 'title': 'See the aurora borealis from space in stunning new NASA video',
 363                 'description': 'md5:74752b7358afb99939c5f8bb2d1d04b1',
 364                 'upload_date': '20160420',
 365                 'timestamp': 1461152093,
 366             },
 367         },
 368         {
 369             'url': 'http://www.msnbc.com/all-in-with-chris-hayes/watch/the-chaotic-gop-immigration-vote-314487875924',
 370             'md5': '6d236bf4f3dddc226633ce6e2c3f814d',
 371             'info_dict': {
 372                 'id': '314487875924',
 373                 'ext': 'mp4',
 374                 'title': 'The chaotic GOP immigration vote',
 375                 '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.',
 376                 'thumbnail': r're:^https?://.*\.jpg$',
 377                 'timestamp': 1406937606,
 378                 'upload_date': '20140802',
 379             },
 380         },
 381         {
 382             'url': 'http://www.nbcnews.com/watch/dateline/full-episode--deadly-betrayal-386250819952',
 383             'only_matching': True,
 384         },
 385         {
 386             # From http://www.vulture.com/2016/06/letterman-couldnt-care-less-about-late-night.html
 387             'url': 'http://www.nbcnews.com/widget/video-embed/701714499682',
 388             'only_matching': True,
 389         },
 390     ]
 391
 392     def _real_extract(self, url):
 393         video_id = self._match_id(url)
 394         webpage = self._download_webpage(url, video_id)
 395
 396         data = self._parse_json(self._search_regex(
 397             r'window\.__data\s*=\s*({.+});', webpage,
 398             'bootstrap json'), video_id, js_to_json)
 399         video_data = try_get(data, lambda x: x['video']['current'], dict)
 400         if not video_data:
 401             video_data = data['article']['content'][0]['primaryMedia']['video']
 402         title = video_data['headline']['primary']
 403
 404         formats = []
 405         for va in video_data.get('videoAssets', []):
 406             public_url = va.get('publicUrl')
 407             if not public_url:
 408                 continue
 409             if '://link.theplatform.com/' in public_url:
 410                 public_url = update_url_query(public_url, {'format': 'redirect'})
 411             format_id = va.get('format')
 412             if format_id == 'M3U':
 413                 formats.extend(self._extract_m3u8_formats(
 414                     public_url, video_id, 'mp4', 'm3u8_native',
 415                     m3u8_id=format_id, fatal=False))
 416                 continue
 417             tbr = int_or_none(va.get('bitrate'), 1000)
 418             if tbr:
 419                 format_id += '-%d' % tbr
 420             formats.append({
 421                 'format_id': format_id,
 422                 'url': public_url,
 423                 'width': int_or_none(va.get('width')),
 424                 'height': int_or_none(va.get('height')),
 425                 'tbr': tbr,
 426                 'ext': 'mp4',
 427             })
 428         self._sort_formats(formats)
 429
 430         subtitles = {}
 431         closed_captioning = video_data.get('closedCaptioning')
 432         if closed_captioning:
 433             for cc_url in closed_captioning.values():
 434                 if not cc_url:
 435                     continue
 436                 subtitles.setdefault('en', []).append({
 437                     'url': cc_url,
 438                 })
 439
 440         return {
 441             'id': video_id,
 442             'title': title,
 443             'description': try_get(video_data, lambda x: x['description']['primary']),
 444             'thumbnail': try_get(video_data, lambda x: x['primaryImage']['url']['primary']),
 445             'duration': parse_duration(video_data.get('duration')),
 446             'timestamp': unified_timestamp(video_data.get('datePublished')),
 447             'formats': formats,
 448             'subtitles': subtitles,
 449         }
 450
 451
 452 class NBCOlympicsIE(InfoExtractor):
 453     IE_NAME = 'nbcolympics'
 454     _VALID_URL = r'https?://www\.nbcolympics\.com/video/(?P<id>[a-z-]+)'
 455
 456     _TEST = {
 457         # Geo-restricted to US
 458         'url': 'http://www.nbcolympics.com/video/justin-roses-son-leo-was-tears-after-his-dad-won-gold',
 459         'md5': '54fecf846d05429fbaa18af557ee523a',
 460         'info_dict': {
 461             'id': 'WjTBzDXx5AUq',
 462             'display_id': 'justin-roses-son-leo-was-tears-after-his-dad-won-gold',
 463             'ext': 'mp4',
 464             'title': 'Rose\'s son Leo was in tears after his dad won gold',
 465             'description': 'Olympic gold medalist Justin Rose gets emotional talking to the impact his win in men\'s golf has already had on his children.',
 466             'timestamp': 1471274964,
 467             'upload_date': '20160815',
 468             'uploader': 'NBCU-SPORTS',
 469         },
 470     }
 471
 472     def _real_extract(self, url):
 473         display_id = self._match_id(url)
 474
 475         webpage = self._download_webpage(url, display_id)
 476
 477         drupal_settings = self._parse_json(self._search_regex(
 478             r'jQuery\.extend\(Drupal\.settings\s*,\s*({.+?})\);',
 479             webpage, 'drupal settings'), display_id)
 480
 481         iframe_url = drupal_settings['vod']['iframe_url']
 482         theplatform_url = iframe_url.replace(
 483             'vplayer.nbcolympics.com', 'player.theplatform.com')
 484
 485         return {
 486             '_type': 'url_transparent',
 487             'url': theplatform_url,
 488             'ie_key': ThePlatformIE.ie_key(),
 489             'display_id': display_id,
 490         }
 491
 492
 493 class NBCOlympicsStreamIE(AdobePassIE):
 494     IE_NAME = 'nbcolympics:stream'
 495     _VALID_URL = r'https?://stream\.nbcolympics\.com/(?P<id>[0-9a-z-]+)'
 496     _TEST = {
 497         'url': 'http://stream.nbcolympics.com/2018-winter-olympics-nbcsn-evening-feb-8',
 498         'info_dict': {
 499             'id': '203493',
 500             'ext': 'mp4',
 501             'title': 're:Curling, Alpine, Luge [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
 502         },
 503         'params': {
 504             # m3u8 download
 505             'skip_download': True,
 506         },
 507     }
 508     _DATA_URL_TEMPLATE = 'http://stream.nbcolympics.com/data/%s_%s.json'
 509
 510     def _real_extract(self, url):
 511         display_id = self._match_id(url)
 512         webpage = self._download_webpage(url, display_id)
 513         pid = self._search_regex(r'pid\s*=\s*(\d+);', webpage, 'pid')
 514         resource = self._search_regex(
 515             r"resource\s*=\s*'(.+)';", webpage,
 516             'resource').replace("' + pid + '", pid)
 517         event_config = self._download_json(
 518             self._DATA_URL_TEMPLATE % ('event_config', pid),
 519             pid)['eventConfig']
 520         title = self._live_title(event_config['eventTitle'])
 521         source_url = self._download_json(
 522             self._DATA_URL_TEMPLATE % ('live_sources', pid),
 523             pid)['videoSources'][0]['sourceUrl']
 524         media_token = self._extract_mvpd_auth(
 525             url, pid, event_config.get('requestorId', 'NBCOlympics'), resource)
 526         formats = self._extract_m3u8_formats(self._download_webpage(
 527             'http://sp.auth.adobe.com/tvs/v1/sign', pid, query={
 528                 'cdn': 'akamai',
 529                 'mediaToken': base64.b64encode(media_token.encode()),
 530                 'resource': base64.b64encode(resource.encode()),
 531                 'url': source_url,
 532             }), pid, 'mp4')
 533         self._sort_formats(formats)
 534
 535         return {
 536             'id': pid,
 537             'display_id': display_id,
 538             'title': title,
 539             'formats': formats,
 540             'is_live': True,
 541         }