youtube_dl/extractor/nbc.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from .theplatform import ThePlatformIE
   7 from ..utils import (
   8     find_xpath_attr,
   9     lowercase_escape,
  10     smuggle_url,
  11     unescapeHTML,
  12     update_url_query,
  13     int_or_none,
  14     HEADRequest,
  15     parse_iso8601,
  16 )
  17
  18
  19 class NBCIE(InfoExtractor):
  20     _VALID_URL = r'https?://www\.nbc\.com/(?:[^/]+/)+(?P<id>n?\d+)'
  21
  22     _TESTS = [
  23         {
  24             'url': 'http://www.nbc.com/the-tonight-show/segments/112966',
  25             'info_dict': {
  26                 'id': '112966',
  27                 'ext': 'mp4',
  28                 'title': 'Jimmy Fallon Surprises Fans at Ben & Jerry\'s',
  29                 'description': 'Jimmy gives out free scoops of his new "Tonight Dough" ice cream flavor by surprising customers at the Ben & Jerry\'s scoop shop.',
  30                 'timestamp': 1424246400,
  31                 'upload_date': '20150218',
  32                 'uploader': 'NBCU-COM',
  33             },
  34             'params': {
  35                 # m3u8 download
  36                 'skip_download': True,
  37             },
  38         },
  39         {
  40             'url': 'http://www.nbc.com/the-tonight-show/episodes/176',
  41             'info_dict': {
  42                 'id': '176',
  43                 'ext': 'flv',
  44                 'title': 'Ricky Gervais, Steven Van Zandt, ILoveMakonnen',
  45                 'description': 'A brand new episode of The Tonight Show welcomes Ricky Gervais, Steven Van Zandt and ILoveMakonnen.',
  46             },
  47             'skip': '404 Not Found',
  48         },
  49         {
  50             'url': 'http://www.nbc.com/saturday-night-live/video/star-wars-teaser/2832821',
  51             'info_dict': {
  52                 'id': '2832821',
  53                 'ext': 'mp4',
  54                 'title': 'Star Wars Teaser',
  55                 'description': 'md5:0b40f9cbde5b671a7ff62fceccc4f442',
  56                 'timestamp': 1417852800,
  57                 'upload_date': '20141206',
  58                 'uploader': 'NBCU-COM',
  59             },
  60             'params': {
  61                 # m3u8 download
  62                 'skip_download': True,
  63             },
  64             'skip': 'Only works from US',
  65         },
  66         {
  67             # This video has expired but with an escaped embedURL
  68             'url': 'http://www.nbc.com/parenthood/episode-guide/season-5/just-like-at-home/515',
  69             'only_matching': True,
  70         },
  71         {
  72             # HLS streams requires the 'hdnea3' cookie
  73             'url': 'http://www.nbc.com/Kings/video/goliath/n1806',
  74             'info_dict': {
  75                 'id': 'n1806',
  76                 'ext': 'mp4',
  77                 'title': 'Goliath',
  78                 '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.',
  79                 'timestamp': 1237100400,
  80                 'upload_date': '20090315',
  81                 'uploader': 'NBCU-COM',
  82             },
  83             'params': {
  84                 'skip_download': True,
  85             },
  86             'skip': 'Only works from US',
  87         }
  88     ]
  89
  90     def _real_extract(self, url):
  91         video_id = self._match_id(url)
  92         webpage = self._download_webpage(url, video_id)
  93         theplatform_url = unescapeHTML(lowercase_escape(self._html_search_regex(
  94             [
  95                 r'(?:class="video-player video-player-full" data-mpx-url|class="player" src)="(.*?)"',
  96                 r'<iframe[^>]+src="((?:https?:)?//player\.theplatform\.com/[^"]+)"',
  97                 r'"embedURL"\s*:\s*"([^"]+)"'
  98             ],
  99             webpage, 'theplatform url').replace('_no_endcard', '').replace('\\/', '/')))
 100         if theplatform_url.startswith('//'):
 101             theplatform_url = 'http:' + theplatform_url
 102         return {
 103             '_type': 'url_transparent',
 104             'ie_key': 'ThePlatform',
 105             'url': smuggle_url(theplatform_url, {'source_url': url}),
 106             'id': video_id,
 107         }
 108
 109
 110 class NBCSportsVPlayerIE(InfoExtractor):
 111     _VALID_URL = r'https?://vplayer\.nbcsports\.com/(?:[^/]+/)+(?P<id>[0-9a-zA-Z_]+)'
 112
 113     _TESTS = [{
 114         'url': 'https://vplayer.nbcsports.com/p/BxmELC/nbcsports_share/select/9CsDKds0kvHI',
 115         'info_dict': {
 116             'id': '9CsDKds0kvHI',
 117             'ext': 'flv',
 118             'description': 'md5:df390f70a9ba7c95ff1daace988f0d8d',
 119             'title': 'Tyler Kalinoski hits buzzer-beater to lift Davidson',
 120             'timestamp': 1426270238,
 121             'upload_date': '20150313',
 122             'uploader': 'NBCU-SPORTS',
 123         }
 124     }, {
 125         'url': 'http://vplayer.nbcsports.com/p/BxmELC/nbc_embedshare/select/_hqLjQ95yx8Z',
 126         'only_matching': True,
 127     }]
 128
 129     @staticmethod
 130     def _extract_url(webpage):
 131         iframe_m = re.search(
 132             r'<iframe[^>]+src="(?P<url>https?://vplayer\.nbcsports\.com/[^"]+)"', webpage)
 133         if iframe_m:
 134             return iframe_m.group('url')
 135
 136     def _real_extract(self, url):
 137         video_id = self._match_id(url)
 138         webpage = self._download_webpage(url, video_id)
 139         theplatform_url = self._og_search_video_url(webpage)
 140         return self.url_result(theplatform_url, 'ThePlatform')
 141
 142
 143 class NBCSportsIE(InfoExtractor):
 144     # Does not include https because its certificate is invalid
 145     _VALID_URL = r'https?://www\.nbcsports\.com//?(?:[^/]+/)+(?P<id>[0-9a-z-]+)'
 146
 147     _TEST = {
 148         'url': 'http://www.nbcsports.com//college-basketball/ncaab/tom-izzo-michigan-st-has-so-much-respect-duke',
 149         'info_dict': {
 150             'id': 'PHJSaFWbrTY9',
 151             'ext': 'flv',
 152             'title': 'Tom Izzo, Michigan St. has \'so much respect\' for Duke',
 153             'description': 'md5:ecb459c9d59e0766ac9c7d5d0eda8113',
 154             'uploader': 'NBCU-SPORTS',
 155             'upload_date': '20150330',
 156             'timestamp': 1427726529,
 157         }
 158     }
 159
 160     def _real_extract(self, url):
 161         video_id = self._match_id(url)
 162         webpage = self._download_webpage(url, video_id)
 163         return self.url_result(
 164             NBCSportsVPlayerIE._extract_url(webpage), 'NBCSportsVPlayer')
 165
 166
 167 class CSNNEIE(InfoExtractor):
 168     _VALID_URL = r'https?://www\.csnne\.com/video/(?P<id>[0-9a-z-]+)'
 169
 170     _TEST = {
 171         'url': 'http://www.csnne.com/video/snc-evening-update-wright-named-red-sox-no-5-starter',
 172         'info_dict': {
 173             'id': 'yvBLLUgQ8WU0',
 174             'ext': 'mp4',
 175             'title': 'SNC evening update: Wright named Red Sox\' No. 5 starter.',
 176             'description': 'md5:1753cfee40d9352b19b4c9b3e589b9e3',
 177             'timestamp': 1459369979,
 178             'upload_date': '20160330',
 179             'uploader': 'NBCU-SPORTS',
 180         }
 181     }
 182
 183     def _real_extract(self, url):
 184         display_id = self._match_id(url)
 185         webpage = self._download_webpage(url, display_id)
 186         return {
 187             '_type': 'url_transparent',
 188             'ie_key': 'ThePlatform',
 189             'url': self._html_search_meta('twitter:player:stream', webpage),
 190             'display_id': display_id,
 191         }
 192
 193
 194 class NBCNewsIE(ThePlatformIE):
 195     _VALID_URL = r'''(?x)https?://(?:www\.)?(?:nbcnews|today)\.com/
 196         (?:video/.+?/(?P<id>\d+)|
 197         ([^/]+/)*(?P<display_id>[^/?]+))
 198         '''
 199
 200     _TESTS = [
 201         {
 202             'url': 'http://www.nbcnews.com/video/nbc-news/52753292',
 203             'md5': '47abaac93c6eaf9ad37ee6c4463a5179',
 204             'info_dict': {
 205                 'id': '52753292',
 206                 'ext': 'flv',
 207                 'title': 'Crew emerges after four-month Mars food study',
 208                 'description': 'md5:24e632ffac72b35f8b67a12d1b6ddfc1',
 209             },
 210         },
 211         {
 212             'url': 'http://www.nbcnews.com/watch/nbcnews-com/how-twitter-reacted-to-the-snowden-interview-269389891880',
 213             'md5': 'af1adfa51312291a017720403826bb64',
 214             'info_dict': {
 215                 'id': '269389891880',
 216                 'ext': 'mp4',
 217                 'title': 'How Twitter Reacted To The Snowden Interview',
 218                 'description': 'md5:65a0bd5d76fe114f3c2727aa3a81fe64',
 219             },
 220         },
 221         {
 222             'url': 'http://www.nbcnews.com/feature/dateline-full-episodes/full-episode-family-business-n285156',
 223             'md5': 'fdbf39ab73a72df5896b6234ff98518a',
 224             'info_dict': {
 225                 'id': 'Wjf9EDR3A_60',
 226                 'ext': 'mp4',
 227                 'title': 'FULL EPISODE: Family Business',
 228                 'description': 'md5:757988edbaae9d7be1d585eb5d55cc04',
 229             },
 230             'skip': 'This page is unavailable.',
 231         },
 232         {
 233             'url': 'http://www.nbcnews.com/nightly-news/video/nightly-news-with-brian-williams-full-broadcast-february-4-394064451844',
 234             'md5': '73135a2e0ef819107bbb55a5a9b2a802',
 235             'info_dict': {
 236                 'id': '394064451844',
 237                 'ext': 'mp4',
 238                 'title': 'Nightly News with Brian Williams Full Broadcast (February 4)',
 239                 'description': 'md5:1c10c1eccbe84a26e5debb4381e2d3c5',
 240             },
 241         },
 242         {
 243             'url': 'http://www.nbcnews.com/business/autos/volkswagen-11-million-vehicles-could-have-suspect-software-emissions-scandal-n431456',
 244             'md5': 'a49e173825e5fcd15c13fc297fced39d',
 245             'info_dict': {
 246                 'id': '529953347624',
 247                 'ext': 'mp4',
 248                 'title': 'Volkswagen U.S. Chief: We \'Totally Screwed Up\'',
 249                 'description': 'md5:d22d1281a24f22ea0880741bb4dd6301',
 250             },
 251             'expected_warnings': ['http-6000 is not available']
 252         },
 253         {
 254             'url': 'http://www.today.com/video/see-the-aurora-borealis-from-space-in-stunning-new-nasa-video-669831235788',
 255             'md5': '118d7ca3f0bea6534f119c68ef539f71',
 256             'info_dict': {
 257                 'id': '669831235788',
 258                 'ext': 'mp4',
 259                 'title': 'See the aurora borealis from space in stunning new NASA video',
 260                 'description': 'md5:74752b7358afb99939c5f8bb2d1d04b1',
 261                 'upload_date': '20160420',
 262                 'timestamp': 1461152093,
 263             },
 264         },
 265         {
 266             'url': 'http://www.nbcnews.com/watch/dateline/full-episode--deadly-betrayal-386250819952',
 267             'only_matching': True,
 268         },
 269         {
 270             # From http://www.vulture.com/2016/06/letterman-couldnt-care-less-about-late-night.html
 271             'url': 'http://www.nbcnews.com/widget/video-embed/701714499682',
 272             'only_matching': True,
 273         },
 274     ]
 275
 276     def _real_extract(self, url):
 277         mobj = re.match(self._VALID_URL, url)
 278         video_id = mobj.group('id')
 279         if video_id is not None:
 280             all_info = self._download_xml('http://www.nbcnews.com/id/%s/displaymode/1219' % video_id, video_id)
 281             info = all_info.find('video')
 282
 283             return {
 284                 'id': video_id,
 285                 'title': info.find('headline').text,
 286                 'ext': 'flv',
 287                 'url': find_xpath_attr(info, 'media', 'type', 'flashVideo').text,
 288                 'description': info.find('caption').text,
 289                 'thumbnail': find_xpath_attr(info, 'media', 'type', 'thumbnail').text,
 290             }
 291         else:
 292             # "feature" and "nightly-news" pages use theplatform.com
 293             display_id = mobj.group('display_id')
 294             webpage = self._download_webpage(url, display_id)
 295             info = None
 296             bootstrap_json = self._search_regex(
 297                 [r'(?m)(?:var\s+(?:bootstrapJson|playlistData)|NEWS\.videoObj)\s*=\s*({.+});?\s*$',
 298                  r'videoObj\s*:\s*({.+})', r'data-video="([^"]+)"'],
 299                 webpage, 'bootstrap json', default=None)
 300             bootstrap = self._parse_json(
 301                 bootstrap_json, display_id, transform_source=unescapeHTML)
 302             if 'results' in bootstrap:
 303                 info = bootstrap['results'][0]['video']
 304             elif 'video' in bootstrap:
 305                 info = bootstrap['video']
 306             else:
 307                 info = bootstrap
 308             video_id = info['mpxId']
 309             title = info['title']
 310
 311             subtitles = {}
 312             caption_links = info.get('captionLinks')
 313             if caption_links:
 314                 for (sub_key, sub_ext) in (('smpte-tt', 'ttml'), ('web-vtt', 'vtt'), ('srt', 'srt')):
 315                     sub_url = caption_links.get(sub_key)
 316                     if sub_url:
 317                         subtitles.setdefault('en', []).append({
 318                             'url': sub_url,
 319                             'ext': sub_ext,
 320                         })
 321
 322             formats = []
 323             for video_asset in info['videoAssets']:
 324                 video_url = video_asset.get('publicUrl')
 325                 if not video_url:
 326                     continue
 327                 container = video_asset.get('format')
 328                 asset_type = video_asset.get('assetType') or ''
 329                 if container == 'ISM' or asset_type == 'FireTV-Once':
 330                     continue
 331                 elif asset_type == 'OnceURL':
 332                     tp_formats, tp_subtitles = self._extract_theplatform_smil(
 333                         video_url, video_id)
 334                     formats.extend(tp_formats)
 335                     subtitles = self._merge_subtitles(subtitles, tp_subtitles)
 336                 else:
 337                     tbr = int_or_none(video_asset.get('bitRate') or video_asset.get('bitrate'), 1000)
 338                     format_id = 'http%s' % ('-%d' % tbr if tbr else '')
 339                     video_url = update_url_query(
 340                         video_url, {'format': 'redirect'})
 341                     # resolve the url so that we can check availability and detect the correct extension
 342                     head = self._request_webpage(
 343                         HEADRequest(video_url), video_id,
 344                         'Checking %s url' % format_id,
 345                         '%s is not available' % format_id,
 346                         fatal=False)
 347                     if head:
 348                         video_url = head.geturl()
 349                         formats.append({
 350                             'format_id': format_id,
 351                             'url': video_url,
 352                             'width': int_or_none(video_asset.get('width')),
 353                             'height': int_or_none(video_asset.get('height')),
 354                             'tbr': tbr,
 355                             'container': video_asset.get('format'),
 356                         })
 357             self._sort_formats(formats)
 358
 359             return {
 360                 'id': video_id,
 361                 'title': title,
 362                 'description': info.get('description'),
 363                 'thumbnail': info.get('thumbnail'),
 364                 'duration': int_or_none(info.get('duration')),
 365                 'timestamp': parse_iso8601(info.get('pubDate') or info.get('pub_date')),
 366                 'formats': formats,
 367                 'subtitles': subtitles,
 368             }
 369
 370
 371 class MSNBCIE(InfoExtractor):
 372     # https URLs redirect to corresponding http ones
 373     _VALID_URL = r'https?://www\.msnbc\.com/[^/]+/watch/(?P<id>[^/]+)'
 374     _TEST = {
 375         'url': 'http://www.msnbc.com/all-in-with-chris-hayes/watch/the-chaotic-gop-immigration-vote-314487875924',
 376         'md5': '6d236bf4f3dddc226633ce6e2c3f814d',
 377         'info_dict': {
 378             'id': 'n_hayes_Aimm_140801_272214',
 379             'ext': 'mp4',
 380             'title': 'The chaotic GOP immigration vote',
 381             '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.',
 382             'thumbnail': 're:^https?://.*\.jpg$',
 383             'timestamp': 1406937606,
 384             'upload_date': '20140802',
 385             'uploader': 'NBCU-NEWS',
 386             'categories': ['MSNBC/Topics/Franchise/Best of last night', 'MSNBC/Topics/General/Congress'],
 387         },
 388     }
 389
 390     def _real_extract(self, url):
 391         video_id = self._match_id(url)
 392         webpage = self._download_webpage(url, video_id)
 393         embed_url = self._html_search_meta('embedURL', webpage)
 394         return self.url_result(embed_url)