_ Git - youtube-dl/blob - 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             },
  31             'params': {
  32                 # m3u8 download
  33                 'skip_download': True,
  34             },
  35         },
  36         {
  37             'url': 'http://www.nbc.com/the-tonight-show/episodes/176',
  38             'info_dict': {
  39                 'id': '176',
  40                 'ext': 'flv',
  41                 'title': 'Ricky Gervais, Steven Van Zandt, ILoveMakonnen',
  42                 'description': 'A brand new episode of The Tonight Show welcomes Ricky Gervais, Steven Van Zandt and ILoveMakonnen.',
  43             },
  44             'skip': '404 Not Found',
  45         },
  46         {
  47             'url': 'http://www.nbc.com/saturday-night-live/video/star-wars-teaser/2832821',
  48             'info_dict': {
  49                 'id': '2832821',
  50                 'ext': 'mp4',
  51                 'title': 'Star Wars Teaser',
  52                 'description': 'md5:0b40f9cbde5b671a7ff62fceccc4f442',
  53             },
  54             'params': {
  55                 # m3u8 download
  56                 'skip_download': True,
  57             },
  58             'skip': 'Only works from US',
  59         },
  60         {
  61             # This video has expired but with an escaped embedURL
  62             'url': 'http://www.nbc.com/parenthood/episode-guide/season-5/just-like-at-home/515',
  63             'only_matching': True,
  64         }
  65     ]
  66
  67     def _real_extract(self, url):
  68         video_id = self._match_id(url)
  69         webpage = self._download_webpage(url, video_id)
  70         theplatform_url = unescapeHTML(lowercase_escape(self._html_search_regex(
  71             [
  72                 r'(?:class="video-player video-player-full" data-mpx-url|class="player" src)="(.*?)"',
  73                 r'<iframe[^>]+src="((?:https?:)?//player\.theplatform\.com/[^"]+)"',
  74                 r'"embedURL"\s*:\s*"([^"]+)"'
  75             ],
  76             webpage, 'theplatform url').replace('_no_endcard', '').replace('\\/', '/')))
  77         if theplatform_url.startswith('//'):
  78             theplatform_url = 'http:' + theplatform_url
  79         return {
  80             '_type': 'url_transparent',
  81             'url': smuggle_url(theplatform_url, {'source_url': url}),
  82             'id': video_id,
  83         }
  84
  85
  86 class NBCSportsVPlayerIE(InfoExtractor):
  87     _VALID_URL = r'https?://vplayer\.nbcsports\.com/(?:[^/]+/)+(?P<id>[0-9a-zA-Z_]+)'
  88
  89     _TESTS = [{
  90         'url': 'https://vplayer.nbcsports.com/p/BxmELC/nbcsports_share/select/9CsDKds0kvHI',
  91         'info_dict': {
  92             'id': '9CsDKds0kvHI',
  93             'ext': 'flv',
  94             'description': 'md5:df390f70a9ba7c95ff1daace988f0d8d',
  95             'title': 'Tyler Kalinoski hits buzzer-beater to lift Davidson',
  96         }
  97     }, {
  98         'url': 'http://vplayer.nbcsports.com/p/BxmELC/nbc_embedshare/select/_hqLjQ95yx8Z',
  99         'only_matching': True,
 100     }]
 101
 102     @staticmethod
 103     def _extract_url(webpage):
 104         iframe_m = re.search(
 105             r'<iframe[^>]+src="(?P<url>https?://vplayer\.nbcsports\.com/[^"]+)"', webpage)
 106         if iframe_m:
 107             return iframe_m.group('url')
 108
 109     def _real_extract(self, url):
 110         video_id = self._match_id(url)
 111         webpage = self._download_webpage(url, video_id)
 112         theplatform_url = self._og_search_video_url(webpage)
 113         return self.url_result(theplatform_url, 'ThePlatform')
 114
 115
 116 class NBCSportsIE(InfoExtractor):
 117     # Does not include https because its certificate is invalid
 118     _VALID_URL = r'https?://www\.nbcsports\.com//?(?:[^/]+/)+(?P<id>[0-9a-z-]+)'
 119
 120     _TEST = {
 121         'url': 'http://www.nbcsports.com//college-basketball/ncaab/tom-izzo-michigan-st-has-so-much-respect-duke',
 122         'info_dict': {
 123             'id': 'PHJSaFWbrTY9',
 124             'ext': 'flv',
 125             'title': 'Tom Izzo, Michigan St. has \'so much respect\' for Duke',
 126             'description': 'md5:ecb459c9d59e0766ac9c7d5d0eda8113',
 127         }
 128     }
 129
 130     def _real_extract(self, url):
 131         video_id = self._match_id(url)
 132         webpage = self._download_webpage(url, video_id)
 133         return self.url_result(
 134             NBCSportsVPlayerIE._extract_url(webpage), 'NBCSportsVPlayer')
 135
 136
 137 class NBCNewsIE(ThePlatformIE):
 138     _VALID_URL = r'''(?x)https?://(?:www\.)?nbcnews\.com/
 139         (?:video/.+?/(?P<id>\d+)|
 140         ([^/]+/)*(?P<display_id>[^/?]+))
 141         '''
 142
 143     _TESTS = [
 144         {
 145             'url': 'http://www.nbcnews.com/video/nbc-news/52753292',
 146             'md5': '47abaac93c6eaf9ad37ee6c4463a5179',
 147             'info_dict': {
 148                 'id': '52753292',
 149                 'ext': 'flv',
 150                 'title': 'Crew emerges after four-month Mars food study',
 151                 'description': 'md5:24e632ffac72b35f8b67a12d1b6ddfc1',
 152             },
 153         },
 154         {
 155             'url': 'http://www.nbcnews.com/watch/nbcnews-com/how-twitter-reacted-to-the-snowden-interview-269389891880',
 156             'md5': 'af1adfa51312291a017720403826bb64',
 157             'info_dict': {
 158                 'id': '269389891880',
 159                 'ext': 'mp4',
 160                 'title': 'How Twitter Reacted To The Snowden Interview',
 161                 'description': 'md5:65a0bd5d76fe114f3c2727aa3a81fe64',
 162             },
 163         },
 164         {
 165             'url': 'http://www.nbcnews.com/feature/dateline-full-episodes/full-episode-family-business-n285156',
 166             'md5': 'fdbf39ab73a72df5896b6234ff98518a',
 167             'info_dict': {
 168                 'id': 'Wjf9EDR3A_60',
 169                 'ext': 'mp4',
 170                 'title': 'FULL EPISODE: Family Business',
 171                 'description': 'md5:757988edbaae9d7be1d585eb5d55cc04',
 172             },
 173             'skip': 'This page is unavailable.',
 174         },
 175         {
 176             'url': 'http://www.nbcnews.com/nightly-news/video/nightly-news-with-brian-williams-full-broadcast-february-4-394064451844',
 177             'md5': '73135a2e0ef819107bbb55a5a9b2a802',
 178             'info_dict': {
 179                 'id': '394064451844',
 180                 'ext': 'mp4',
 181                 'title': 'Nightly News with Brian Williams Full Broadcast (February 4)',
 182                 'description': 'md5:1c10c1eccbe84a26e5debb4381e2d3c5',
 183             },
 184         },
 185         {
 186             'url': 'http://www.nbcnews.com/business/autos/volkswagen-11-million-vehicles-could-have-suspect-software-emissions-scandal-n431456',
 187             'md5': 'a49e173825e5fcd15c13fc297fced39d',
 188             'info_dict': {
 189                 'id': '529953347624',
 190                 'ext': 'mp4',
 191                 'title': 'Volkswagen U.S. Chief: We \'Totally Screwed Up\'',
 192                 'description': 'md5:d22d1281a24f22ea0880741bb4dd6301',
 193             },
 194             'expected_warnings': ['http-6000 is not available']
 195         },
 196         {
 197             'url': 'http://www.nbcnews.com/watch/dateline/full-episode--deadly-betrayal-386250819952',
 198             'only_matching': True,
 199         },
 200     ]
 201
 202     def _real_extract(self, url):
 203         mobj = re.match(self._VALID_URL, url)
 204         video_id = mobj.group('id')
 205         if video_id is not None:
 206             all_info = self._download_xml('http://www.nbcnews.com/id/%s/displaymode/1219' % video_id, video_id)
 207             info = all_info.find('video')
 208
 209             return {
 210                 'id': video_id,
 211                 'title': info.find('headline').text,
 212                 'ext': 'flv',
 213                 'url': find_xpath_attr(info, 'media', 'type', 'flashVideo').text,
 214                 'description': info.find('caption').text,
 215                 'thumbnail': find_xpath_attr(info, 'media', 'type', 'thumbnail').text,
 216             }
 217         else:
 218             # "feature" and "nightly-news" pages use theplatform.com
 219             display_id = mobj.group('display_id')
 220             webpage = self._download_webpage(url, display_id)
 221             info = None
 222             bootstrap_json = self._search_regex(
 223                 r'(?m)var\s+(?:bootstrapJson|playlistData)\s*=\s*({.+});?\s*$',
 224                 webpage, 'bootstrap json', default=None)
 225             if bootstrap_json:
 226                 bootstrap = self._parse_json(bootstrap_json, display_id)
 227                 info = bootstrap['results'][0]['video']
 228             else:
 229                 player_instance_json = self._search_regex(
 230                     r'videoObj\s*:\s*({.+})', webpage, 'player instance')
 231                 info = self._parse_json(player_instance_json, display_id)
 232             video_id = info['mpxId']
 233             title = info['title']
 234
 235             subtitles = {}
 236             caption_links = info.get('captionLinks')
 237             if caption_links:
 238                 for (sub_key, sub_ext) in (('smpte-tt', 'ttml'), ('web-vtt', 'vtt'), ('srt', 'srt')):
 239                     sub_url = caption_links.get(sub_key)
 240                     if sub_url:
 241                         subtitles.setdefault('en', []).append({
 242                             'url': sub_url,
 243                             'ext': sub_ext,
 244                         })
 245
 246             formats = []
 247             for video_asset in info['videoAssets']:
 248                 video_url = video_asset.get('publicUrl')
 249                 if not video_url:
 250                     continue
 251                 container = video_asset.get('format')
 252                 asset_type = video_asset.get('assetType') or ''
 253                 if container == 'ISM' or asset_type == 'FireTV-Once':
 254                     continue
 255                 elif asset_type == 'OnceURL':
 256                     tp_formats, tp_subtitles = self._extract_theplatform_smil(
 257                         video_url, video_id)
 258                     formats.extend(tp_formats)
 259                     subtitles = self._merge_subtitles(subtitles, tp_subtitles)
 260                 else:
 261                     tbr = int_or_none(video_asset.get('bitRate'), 1000)
 262                     format_id = 'http%s' % ('-%d' % tbr if tbr else '')
 263                     video_url = update_url_query(
 264                         video_url, {'format': 'redirect'})
 265                     # resolve the url so that we can check availability and detect the correct extension
 266                     head = self._request_webpage(
 267                         HEADRequest(video_url), video_id,
 268                         'Checking %s url' % format_id,
 269                         '%s is not available' % format_id,
 270                         fatal=False)
 271                     if head:
 272                         video_url = head.geturl()
 273                         formats.append({
 274                             'format_id': format_id,
 275                             'url': video_url,
 276                             'width': int_or_none(video_asset.get('width')),
 277                             'height': int_or_none(video_asset.get('height')),
 278                             'tbr': tbr,
 279                             'container': video_asset.get('format'),
 280                         })
 281             self._sort_formats(formats)
 282
 283             return {
 284                 'id': video_id,
 285                 'title': title,
 286                 'description': info.get('description'),
 287                 'thumbnail': info.get('description'),
 288                 'thumbnail': info.get('thumbnail'),
 289                 'duration': int_or_none(info.get('duration')),
 290                 'timestamp': parse_iso8601(info.get('pubDate')),
 291                 'formats': formats,
 292                 'subtitles': subtitles,
 293             }
 294
 295
 296 class MSNBCIE(InfoExtractor):
 297     # https URLs redirect to corresponding http ones
 298     _VALID_URL = r'https?://www\.msnbc\.com/[^/]+/watch/(?P<id>[^/]+)'
 299     _TEST = {
 300         'url': 'http://www.msnbc.com/all-in-with-chris-hayes/watch/the-chaotic-gop-immigration-vote-314487875924',
 301         'md5': '6d236bf4f3dddc226633ce6e2c3f814d',
 302         'info_dict': {
 303             'id': 'n_hayes_Aimm_140801_272214',
 304             'ext': 'mp4',
 305             'title': 'The chaotic GOP immigration vote',
 306             '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.',
 307             'thumbnail': 're:^https?://.*\.jpg$',
 308             'timestamp': 1406937606,
 309             'upload_date': '20140802',
 310             'categories': ['MSNBC/Topics/Franchise/Best of last night', 'MSNBC/Topics/General/Congress'],
 311         },
 312     }
 313
 314     def _real_extract(self, url):
 315         video_id = self._match_id(url)
 316         webpage = self._download_webpage(url, video_id)
 317         embed_url = self._html_search_meta('embedURL', webpage)
 318         return self.url_result(embed_url)