_ Git - youtube-dl/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         response = self._download_json(
  91             'https://friendship.nbc.co/v2/graphql', video_id, query={
  92                 'query': '''{
  93   page(name: "%s", platform: web, type: VIDEO, userId: "0") {
  94     data {
  95       ... on VideoPageData {
  96         description
  97         episodeNumber
  98         keywords
  99         locked
 100         mpxAccountId
 101         mpxGuid
 102         rating
 103         seasonNumber
 104         secondaryTitle
 105         seriesShortTitle
 106       }
 107     }
 108   }
 109 }''' % permalink,
 110             })
 111         video_data = response['data']['page']['data']
 112         query = {
 113             'mbr': 'true',
 114             'manifest': 'm3u',
 115         }
 116         video_id = video_data['mpxGuid']
 117         title = video_data['secondaryTitle']
 118         if video_data.get('locked'):
 119             resource = self._get_mvpd_resource(
 120                 'nbcentertainment', title, video_id,
 121                 video_data.get('rating'))
 122             query['auth'] = self._extract_mvpd_auth(
 123                 url, video_id, 'nbcentertainment', resource)
 124         theplatform_url = smuggle_url(update_url_query(
 125             'http://link.theplatform.com/s/NnzsPC/media/guid/%s/%s' % (video_data.get('mpxAccountId') or '2410887629', video_id),
 126             query), {'force_smil_url': True})
 127         return {
 128             '_type': 'url_transparent',
 129             'id': video_id,
 130             'title': title,
 131             'url': theplatform_url,
 132             'description': video_data.get('description'),
 133             'tags': video_data.get('keywords'),
 134             'season_number': int_or_none(video_data.get('seasonNumber')),
 135             'episode_number': int_or_none(video_data.get('episodeNumber')),
 136             'episode': title,
 137             'series': video_data.get('seriesShortTitle'),
 138             'ie_key': 'ThePlatform',
 139         }
 140
 141
 142 class NBCSportsVPlayerIE(InfoExtractor):
 143     _VALID_URL = r'https?://vplayer\.nbcsports\.com/(?:[^/]+/)+(?P<id>[0-9a-zA-Z_]+)'
 144
 145     _TESTS = [{
 146         'url': 'https://vplayer.nbcsports.com/p/BxmELC/nbcsports_embed/select/9CsDKds0kvHI',
 147         'info_dict': {
 148             'id': '9CsDKds0kvHI',
 149             'ext': 'mp4',
 150             'description': 'md5:df390f70a9ba7c95ff1daace988f0d8d',
 151             'title': 'Tyler Kalinoski hits buzzer-beater to lift Davidson',
 152             'timestamp': 1426270238,
 153             'upload_date': '20150313',
 154             'uploader': 'NBCU-SPORTS',
 155         }
 156     }, {
 157         'url': 'https://vplayer.nbcsports.com/p/BxmELC/nbcsports_embed/select/media/_hqLjQ95yx8Z',
 158         'only_matching': True,
 159     }]
 160
 161     @staticmethod
 162     def _extract_url(webpage):
 163         iframe_m = re.search(
 164             r'<iframe[^>]+src="(?P<url>https?://vplayer\.nbcsports\.com/[^"]+)"', webpage)
 165         if iframe_m:
 166             return iframe_m.group('url')
 167
 168     def _real_extract(self, url):
 169         video_id = self._match_id(url)
 170         webpage = self._download_webpage(url, video_id)
 171         theplatform_url = self._og_search_video_url(webpage).replace(
 172             'vplayer.nbcsports.com', 'player.theplatform.com')
 173         return self.url_result(theplatform_url, 'ThePlatform')
 174
 175
 176 class NBCSportsIE(InfoExtractor):
 177     # Does not include https because its certificate is invalid
 178     _VALID_URL = r'https?://(?:www\.)?nbcsports\.com//?(?:[^/]+/)+(?P<id>[0-9a-z-]+)'
 179
 180     _TEST = {
 181         'url': 'http://www.nbcsports.com//college-basketball/ncaab/tom-izzo-michigan-st-has-so-much-respect-duke',
 182         'info_dict': {
 183             'id': 'PHJSaFWbrTY9',
 184             'ext': 'flv',
 185             'title': 'Tom Izzo, Michigan St. has \'so much respect\' for Duke',
 186             'description': 'md5:ecb459c9d59e0766ac9c7d5d0eda8113',
 187             'uploader': 'NBCU-SPORTS',
 188             'upload_date': '20150330',
 189             'timestamp': 1427726529,
 190         }
 191     }
 192
 193     def _real_extract(self, url):
 194         video_id = self._match_id(url)
 195         webpage = self._download_webpage(url, video_id)
 196         return self.url_result(
 197             NBCSportsVPlayerIE._extract_url(webpage), 'NBCSportsVPlayer')
 198
 199
 200 class NBCSportsStreamIE(AdobePassIE):
 201     _VALID_URL = r'https?://stream\.nbcsports\.com/.+?\bpid=(?P<id>\d+)'
 202     _TEST = {
 203         'url': 'http://stream.nbcsports.com/nbcsn/generic?pid=206559',
 204         'info_dict': {
 205             'id': '206559',
 206             'ext': 'mp4',
 207             'title': 'Amgen Tour of California Women\'s Recap',
 208             'description': 'md5:66520066b3b5281ada7698d0ea2aa894',
 209         },
 210         'params': {
 211             # m3u8 download
 212             'skip_download': True,
 213         },
 214         'skip': 'Requires Adobe Pass Authentication',
 215     }
 216
 217     def _real_extract(self, url):
 218         video_id = self._match_id(url)
 219         live_source = self._download_json(
 220             'http://stream.nbcsports.com/data/live_sources_%s.json' % video_id,
 221             video_id)
 222         video_source = live_source['videoSources'][0]
 223         title = video_source['title']
 224         source_url = None
 225         for k in ('source', 'msl4source', 'iossource', 'hlsv4'):
 226             sk = k + 'Url'
 227             source_url = video_source.get(sk) or video_source.get(sk + 'Alt')
 228             if source_url:
 229                 break
 230         else:
 231             source_url = video_source['ottStreamUrl']
 232         is_live = video_source.get('type') == 'live' or video_source.get('status') == 'Live'
 233         resource = self._get_mvpd_resource('nbcsports', title, video_id, '')
 234         token = self._extract_mvpd_auth(url, video_id, 'nbcsports', resource)
 235         tokenized_url = self._download_json(
 236             'https://token.playmakerservices.com/cdn',
 237             video_id, data=json.dumps({
 238                 'requestorId': 'nbcsports',
 239                 'pid': video_id,
 240                 'application': 'NBCSports',
 241                 'version': 'v1',
 242                 'platform': 'desktop',
 243                 'cdn': 'akamai',
 244                 'url': video_source['sourceUrl'],
 245                 'token': base64.b64encode(token.encode()).decode(),
 246                 'resourceId': base64.b64encode(resource.encode()).decode(),
 247             }).encode())['tokenizedUrl']
 248         formats = self._extract_m3u8_formats(tokenized_url, video_id, 'mp4')
 249         self._sort_formats(formats)
 250         return {
 251             'id': video_id,
 252             'title': self._live_title(title) if is_live else title,
 253             'description': live_source.get('description'),
 254             'formats': formats,
 255             'is_live': is_live,
 256         }
 257
 258
 259 class CSNNEIE(InfoExtractor):
 260     _VALID_URL = r'https?://(?:www\.)?csnne\.com/video/(?P<id>[0-9a-z-]+)'
 261
 262     _TEST = {
 263         'url': 'http://www.csnne.com/video/snc-evening-update-wright-named-red-sox-no-5-starter',
 264         'info_dict': {
 265             'id': 'yvBLLUgQ8WU0',
 266             'ext': 'mp4',
 267             'title': 'SNC evening update: Wright named Red Sox\' No. 5 starter.',
 268             'description': 'md5:1753cfee40d9352b19b4c9b3e589b9e3',
 269             'timestamp': 1459369979,
 270             'upload_date': '20160330',
 271             'uploader': 'NBCU-SPORTS',
 272         }
 273     }
 274
 275     def _real_extract(self, url):
 276         display_id = self._match_id(url)
 277         webpage = self._download_webpage(url, display_id)
 278         return {
 279             '_type': 'url_transparent',
 280             'ie_key': 'ThePlatform',
 281             'url': self._html_search_meta('twitter:player:stream', webpage),
 282             'display_id': display_id,
 283         }
 284
 285
 286 class NBCNewsIE(ThePlatformIE):
 287     _VALID_URL = r'(?x)https?://(?:www\.)?(?:nbcnews|today|msnbc)\.com/([^/]+/)*(?:.*-)?(?P<id>[^/?]+)'
 288
 289     _TESTS = [
 290         {
 291             'url': 'http://www.nbcnews.com/watch/nbcnews-com/how-twitter-reacted-to-the-snowden-interview-269389891880',
 292             'md5': 'cf4bc9e6ce0130f00f545d80ecedd4bf',
 293             'info_dict': {
 294                 'id': '269389891880',
 295                 'ext': 'mp4',
 296                 'title': 'How Twitter Reacted To The Snowden Interview',
 297                 'description': 'md5:65a0bd5d76fe114f3c2727aa3a81fe64',
 298                 'timestamp': 1401363060,
 299                 'upload_date': '20140529',
 300             },
 301         },
 302         {
 303             'url': 'http://www.nbcnews.com/feature/dateline-full-episodes/full-episode-family-business-n285156',
 304             'md5': 'fdbf39ab73a72df5896b6234ff98518a',
 305             'info_dict': {
 306                 'id': '529953347624',
 307                 'ext': 'mp4',
 308                 'title': 'FULL EPISODE: Family Business',
 309                 'description': 'md5:757988edbaae9d7be1d585eb5d55cc04',
 310             },
 311             'skip': 'This page is unavailable.',
 312         },
 313         {
 314             'url': 'http://www.nbcnews.com/nightly-news/video/nightly-news-with-brian-williams-full-broadcast-february-4-394064451844',
 315             'md5': '8eb831eca25bfa7d25ddd83e85946548',
 316             'info_dict': {
 317                 'id': '394064451844',
 318                 'ext': 'mp4',
 319                 'title': 'Nightly News with Brian Williams Full Broadcast (February 4)',
 320                 'description': 'md5:1c10c1eccbe84a26e5debb4381e2d3c5',
 321                 'timestamp': 1423104900,
 322                 'upload_date': '20150205',
 323             },
 324         },
 325         {
 326             'url': 'http://www.nbcnews.com/business/autos/volkswagen-11-million-vehicles-could-have-suspect-software-emissions-scandal-n431456',
 327             'md5': '4a8c4cec9e1ded51060bdda36ff0a5c0',
 328             'info_dict': {
 329                 'id': 'n431456',
 330                 'ext': 'mp4',
 331                 'title': "Volkswagen U.S. Chief:  We 'Totally Screwed Up'",
 332                 'description': 'md5:d22d1281a24f22ea0880741bb4dd6301',
 333                 'upload_date': '20150922',
 334                 'timestamp': 1442917800,
 335             },
 336         },
 337         {
 338             'url': 'http://www.today.com/video/see-the-aurora-borealis-from-space-in-stunning-new-nasa-video-669831235788',
 339             'md5': '118d7ca3f0bea6534f119c68ef539f71',
 340             'info_dict': {
 341                 'id': '669831235788',
 342                 'ext': 'mp4',
 343                 'title': 'See the aurora borealis from space in stunning new NASA video',
 344                 'description': 'md5:74752b7358afb99939c5f8bb2d1d04b1',
 345                 'upload_date': '20160420',
 346                 'timestamp': 1461152093,
 347             },
 348         },
 349         {
 350             'url': 'http://www.msnbc.com/all-in-with-chris-hayes/watch/the-chaotic-gop-immigration-vote-314487875924',
 351             'md5': '6d236bf4f3dddc226633ce6e2c3f814d',
 352             'info_dict': {
 353                 'id': '314487875924',
 354                 'ext': 'mp4',
 355                 'title': 'The chaotic GOP immigration vote',
 356                 '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.',
 357                 'thumbnail': r're:^https?://.*\.jpg$',
 358                 'timestamp': 1406937606,
 359                 'upload_date': '20140802',
 360             },
 361         },
 362         {
 363             'url': 'http://www.nbcnews.com/watch/dateline/full-episode--deadly-betrayal-386250819952',
 364             'only_matching': True,
 365         },
 366         {
 367             # From http://www.vulture.com/2016/06/letterman-couldnt-care-less-about-late-night.html
 368             'url': 'http://www.nbcnews.com/widget/video-embed/701714499682',
 369             'only_matching': True,
 370         },
 371     ]
 372
 373     def _real_extract(self, url):
 374         video_id = self._match_id(url)
 375         webpage = self._download_webpage(url, video_id)
 376
 377         data = self._parse_json(self._search_regex(
 378             r'window\.__data\s*=\s*({.+});', webpage,
 379             'bootstrap json'), video_id, js_to_json)
 380         video_data = try_get(data, lambda x: x['video']['current'], dict)
 381         if not video_data:
 382             video_data = data['article']['content'][0]['primaryMedia']['video']
 383         title = video_data['headline']['primary']
 384
 385         formats = []
 386         for va in video_data.get('videoAssets', []):
 387             public_url = va.get('publicUrl')
 388             if not public_url:
 389                 continue
 390             if '://link.theplatform.com/' in public_url:
 391                 public_url = update_url_query(public_url, {'format': 'redirect'})
 392             format_id = va.get('format')
 393             if format_id == 'M3U':
 394                 formats.extend(self._extract_m3u8_formats(
 395                     public_url, video_id, 'mp4', 'm3u8_native',
 396                     m3u8_id=format_id, fatal=False))
 397                 continue
 398             tbr = int_or_none(va.get('bitrate'), 1000)
 399             if tbr:
 400                 format_id += '-%d' % tbr
 401             formats.append({
 402                 'format_id': format_id,
 403                 'url': public_url,
 404                 'width': int_or_none(va.get('width')),
 405                 'height': int_or_none(va.get('height')),
 406                 'tbr': tbr,
 407                 'ext': 'mp4',
 408             })
 409         self._sort_formats(formats)
 410
 411         subtitles = {}
 412         closed_captioning = video_data.get('closedCaptioning')
 413         if closed_captioning:
 414             for cc_url in closed_captioning.values():
 415                 if not cc_url:
 416                     continue
 417                 subtitles.setdefault('en', []).append({
 418                     'url': cc_url,
 419                 })
 420
 421         return {
 422             'id': video_id,
 423             'title': title,
 424             'description': try_get(video_data, lambda x: x['description']['primary']),
 425             'thumbnail': try_get(video_data, lambda x: x['primaryImage']['url']['primary']),
 426             'duration': parse_duration(video_data.get('duration')),
 427             'timestamp': unified_timestamp(video_data.get('datePublished')),
 428             'formats': formats,
 429             'subtitles': subtitles,
 430         }
 431
 432
 433 class NBCOlympicsIE(InfoExtractor):
 434     IE_NAME = 'nbcolympics'
 435     _VALID_URL = r'https?://www\.nbcolympics\.com/video/(?P<id>[a-z-]+)'
 436
 437     _TEST = {
 438         # Geo-restricted to US
 439         'url': 'http://www.nbcolympics.com/video/justin-roses-son-leo-was-tears-after-his-dad-won-gold',
 440         'md5': '54fecf846d05429fbaa18af557ee523a',
 441         'info_dict': {
 442             'id': 'WjTBzDXx5AUq',
 443             'display_id': 'justin-roses-son-leo-was-tears-after-his-dad-won-gold',
 444             'ext': 'mp4',
 445             'title': 'Rose\'s son Leo was in tears after his dad won gold',
 446             'description': 'Olympic gold medalist Justin Rose gets emotional talking to the impact his win in men\'s golf has already had on his children.',
 447             'timestamp': 1471274964,
 448             'upload_date': '20160815',
 449             'uploader': 'NBCU-SPORTS',
 450         },
 451     }
 452
 453     def _real_extract(self, url):
 454         display_id = self._match_id(url)
 455
 456         webpage = self._download_webpage(url, display_id)
 457
 458         drupal_settings = self._parse_json(self._search_regex(
 459             r'jQuery\.extend\(Drupal\.settings\s*,\s*({.+?})\);',
 460             webpage, 'drupal settings'), display_id)
 461
 462         iframe_url = drupal_settings['vod']['iframe_url']
 463         theplatform_url = iframe_url.replace(
 464             'vplayer.nbcolympics.com', 'player.theplatform.com')
 465
 466         return {
 467             '_type': 'url_transparent',
 468             'url': theplatform_url,
 469             'ie_key': ThePlatformIE.ie_key(),
 470             'display_id': display_id,
 471         }
 472
 473
 474 class NBCOlympicsStreamIE(AdobePassIE):
 475     IE_NAME = 'nbcolympics:stream'
 476     _VALID_URL = r'https?://stream\.nbcolympics\.com/(?P<id>[0-9a-z-]+)'
 477     _TEST = {
 478         'url': 'http://stream.nbcolympics.com/2018-winter-olympics-nbcsn-evening-feb-8',
 479         'info_dict': {
 480             'id': '203493',
 481             'ext': 'mp4',
 482             'title': 're:Curling, Alpine, Luge [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
 483         },
 484         'params': {
 485             # m3u8 download
 486             'skip_download': True,
 487         },
 488     }
 489     _DATA_URL_TEMPLATE = 'http://stream.nbcolympics.com/data/%s_%s.json'
 490
 491     def _real_extract(self, url):
 492         display_id = self._match_id(url)
 493         webpage = self._download_webpage(url, display_id)
 494         pid = self._search_regex(r'pid\s*=\s*(\d+);', webpage, 'pid')
 495         resource = self._search_regex(
 496             r"resource\s*=\s*'(.+)';", webpage,
 497             'resource').replace("' + pid + '", pid)
 498         event_config = self._download_json(
 499             self._DATA_URL_TEMPLATE % ('event_config', pid),
 500             pid)['eventConfig']
 501         title = self._live_title(event_config['eventTitle'])
 502         source_url = self._download_json(
 503             self._DATA_URL_TEMPLATE % ('live_sources', pid),
 504             pid)['videoSources'][0]['sourceUrl']
 505         media_token = self._extract_mvpd_auth(
 506             url, pid, event_config.get('requestorId', 'NBCOlympics'), resource)
 507         formats = self._extract_m3u8_formats(self._download_webpage(
 508             'http://sp.auth.adobe.com/tvs/v1/sign', pid, query={
 509                 'cdn': 'akamai',
 510                 'mediaToken': base64.b64encode(media_token.encode()),
 511                 'resource': base64.b64encode(resource.encode()),
 512                 'url': source_url,
 513             }), pid, 'mp4')
 514         self._sort_formats(formats)
 515
 516         return {
 517             'id': pid,
 518             'display_id': display_id,
 519             'title': title,
 520             'formats': formats,
 521             'is_live': True,
 522         }