Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/nbc.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..compat import (
   7     compat_str,
   8     compat_HTTPError,
   9 )
  10 from ..utils import (
  11     ExtractorError,
  12     find_xpath_attr,
  13     lowercase_escape,
  14     unescapeHTML,
  15 )
  16
  17
  18 class NBCIE(InfoExtractor):
  19     _VALID_URL = r'https?://www\.nbc\.com/(?:[^/]+/)+(?P<id>n?\d+)'
  20
  21     _TESTS = [
  22         {
  23             'url': 'http://www.nbc.com/the-tonight-show/segments/112966',
  24             # md5 checksum is not stable
  25             'info_dict': {
  26                 'id': 'c9xnCo0YPOPH',
  27                 'ext': 'flv',
  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         },
  32         {
  33             'url': 'http://www.nbc.com/the-tonight-show/episodes/176',
  34             'info_dict': {
  35                 'id': 'XwU9KZkp98TH',
  36                 'ext': 'flv',
  37                 'title': 'Ricky Gervais, Steven Van Zandt, ILoveMakonnen',
  38                 'description': 'A brand new episode of The Tonight Show welcomes Ricky Gervais, Steven Van Zandt and ILoveMakonnen.',
  39             },
  40             'skip': 'Only works from US',
  41         },
  42         {
  43             'url': 'http://www.nbc.com/saturday-night-live/video/star-wars-teaser/2832821',
  44             'info_dict': {
  45                 'id': '8iUuyzWDdYUZ',
  46                 'ext': 'flv',
  47                 'title': 'Star Wars Teaser',
  48                 'description': 'md5:0b40f9cbde5b671a7ff62fceccc4f442',
  49             },
  50             'skip': 'Only works from US',
  51         },
  52         {
  53             # This video has expired but with an escaped embedURL
  54             'url': 'http://www.nbc.com/parenthood/episode-guide/season-5/just-like-at-home/515',
  55             'skip': 'Expired'
  56         }
  57     ]
  58
  59     def _real_extract(self, url):
  60         video_id = self._match_id(url)
  61         webpage = self._download_webpage(url, video_id)
  62         theplatform_url = unescapeHTML(lowercase_escape(self._html_search_regex(
  63             [
  64                 r'(?:class="video-player video-player-full" data-mpx-url|class="player" src)="(.*?)"',
  65                 r'"embedURL"\s*:\s*"([^"]+)"'
  66             ],
  67             webpage, 'theplatform url').replace('_no_endcard', '').replace('\\/', '/')))
  68         if theplatform_url.startswith('//'):
  69             theplatform_url = 'http:' + theplatform_url
  70         return self.url_result(theplatform_url)
  71
  72
  73 class NBCSportsVPlayerIE(InfoExtractor):
  74     _VALID_URL = r'https?://vplayer\.nbcsports\.com/(?:[^/]+/)+(?P<id>[0-9a-zA-Z_]+)'
  75
  76     _TESTS = [{
  77         'url': 'https://vplayer.nbcsports.com/p/BxmELC/nbcsports_share/select/9CsDKds0kvHI',
  78         'info_dict': {
  79             'id': '9CsDKds0kvHI',
  80             'ext': 'flv',
  81             'description': 'md5:df390f70a9ba7c95ff1daace988f0d8d',
  82             'title': 'Tyler Kalinoski hits buzzer-beater to lift Davidson',
  83         }
  84     }, {
  85         'url': 'http://vplayer.nbcsports.com/p/BxmELC/nbc_embedshare/select/_hqLjQ95yx8Z',
  86         'only_matching': True,
  87     }]
  88
  89     @staticmethod
  90     def _extract_url(webpage):
  91         iframe_m = re.search(
  92             r'<iframe[^>]+src="(?P<url>https?://vplayer\.nbcsports\.com/[^"]+)"', webpage)
  93         if iframe_m:
  94             return iframe_m.group('url')
  95
  96     def _real_extract(self, url):
  97         video_id = self._match_id(url)
  98         webpage = self._download_webpage(url, video_id)
  99         theplatform_url = self._og_search_video_url(webpage)
 100         return self.url_result(theplatform_url, 'ThePlatform')
 101
 102
 103 class NBCSportsIE(InfoExtractor):
 104     # Does not include https becuase its certificate is invalid
 105     _VALID_URL = r'http://www\.nbcsports\.com//?(?:[^/]+/)+(?P<id>[0-9a-z-]+)'
 106
 107     _TEST = {
 108         'url': 'http://www.nbcsports.com//college-basketball/ncaab/tom-izzo-michigan-st-has-so-much-respect-duke',
 109         'info_dict': {
 110             'id': 'PHJSaFWbrTY9',
 111             'ext': 'flv',
 112             'title': 'Tom Izzo, Michigan St. has \'so much respect\' for Duke',
 113             'description': 'md5:ecb459c9d59e0766ac9c7d5d0eda8113',
 114         }
 115     }
 116
 117     def _real_extract(self, url):
 118         video_id = self._match_id(url)
 119         webpage = self._download_webpage(url, video_id)
 120         return self.url_result(
 121             NBCSportsVPlayerIE._extract_url(webpage), 'NBCSportsVPlayer')
 122
 123
 124 class NBCNewsIE(InfoExtractor):
 125     _VALID_URL = r'''(?x)https?://(?:www\.)?nbcnews\.com/
 126         (?:video/.+?/(?P<id>\d+)|
 127         (?:watch|feature|nightly-news)/[^/]+/(?P<title>.+))
 128         '''
 129
 130     _TESTS = [
 131         {
 132             'url': 'http://www.nbcnews.com/video/nbc-news/52753292',
 133             'md5': '47abaac93c6eaf9ad37ee6c4463a5179',
 134             'info_dict': {
 135                 'id': '52753292',
 136                 'ext': 'flv',
 137                 'title': 'Crew emerges after four-month Mars food study',
 138                 'description': 'md5:24e632ffac72b35f8b67a12d1b6ddfc1',
 139             },
 140         },
 141         {
 142             'url': 'http://www.nbcnews.com/feature/edward-snowden-interview/how-twitter-reacted-snowden-interview-n117236',
 143             'md5': 'b2421750c9f260783721d898f4c42063',
 144             'info_dict': {
 145                 'id': 'I1wpAI_zmhsQ',
 146                 'ext': 'mp4',
 147                 'title': 'How Twitter Reacted To The Snowden Interview',
 148                 'description': 'md5:65a0bd5d76fe114f3c2727aa3a81fe64',
 149             },
 150             'add_ie': ['ThePlatform'],
 151         },
 152         {
 153             'url': 'http://www.nbcnews.com/feature/dateline-full-episodes/full-episode-family-business-n285156',
 154             'md5': 'fdbf39ab73a72df5896b6234ff98518a',
 155             'info_dict': {
 156                 'id': 'Wjf9EDR3A_60',
 157                 'ext': 'mp4',
 158                 'title': 'FULL EPISODE: Family Business',
 159                 'description': 'md5:757988edbaae9d7be1d585eb5d55cc04',
 160             },
 161         },
 162         {
 163             'url': 'http://www.nbcnews.com/nightly-news/video/nightly-news-with-brian-williams-full-broadcast-february-4-394064451844',
 164             'md5': 'b5dda8cddd8650baa0dcb616dd2cf60d',
 165             'info_dict': {
 166                 'id': 'sekXqyTVnmN3',
 167                 'ext': 'mp4',
 168                 'title': 'Nightly News with Brian Williams Full Broadcast (February 4)',
 169                 'description': 'md5:1c10c1eccbe84a26e5debb4381e2d3c5',
 170             },
 171         },
 172         {
 173             'url': 'http://www.nbcnews.com/watch/dateline/full-episode--deadly-betrayal-386250819952',
 174             'only_matching': True,
 175         },
 176     ]
 177
 178     def _real_extract(self, url):
 179         mobj = re.match(self._VALID_URL, url)
 180         video_id = mobj.group('id')
 181         if video_id is not None:
 182             all_info = self._download_xml('http://www.nbcnews.com/id/%s/displaymode/1219' % video_id, video_id)
 183             info = all_info.find('video')
 184
 185             return {
 186                 'id': video_id,
 187                 'title': info.find('headline').text,
 188                 'ext': 'flv',
 189                 'url': find_xpath_attr(info, 'media', 'type', 'flashVideo').text,
 190                 'description': compat_str(info.find('caption').text),
 191                 'thumbnail': find_xpath_attr(info, 'media', 'type', 'thumbnail').text,
 192             }
 193         else:
 194             # "feature" and "nightly-news" pages use theplatform.com
 195             title = mobj.group('title')
 196             webpage = self._download_webpage(url, title)
 197             bootstrap_json = self._search_regex(
 198                 r'var\s+(?:bootstrapJson|playlistData)\s*=\s*({.+});?\s*$',
 199                 webpage, 'bootstrap json', flags=re.MULTILINE)
 200             bootstrap = self._parse_json(bootstrap_json, video_id)
 201             info = bootstrap['results'][0]['video']
 202             mpxid = info['mpxId']
 203
 204             base_urls = [
 205                 info['fallbackPlaylistUrl'],
 206                 info['associatedPlaylistUrl'],
 207             ]
 208
 209             for base_url in base_urls:
 210                 if not base_url:
 211                     continue
 212                 playlist_url = base_url + '?form=MPXNBCNewsAPI'
 213
 214                 try:
 215                     all_videos = self._download_json(playlist_url, title)
 216                 except ExtractorError as ee:
 217                     if isinstance(ee.cause, compat_HTTPError):
 218                         continue
 219                     raise
 220
 221                 if not all_videos or 'videos' not in all_videos:
 222                     continue
 223
 224                 try:
 225                     info = next(v for v in all_videos['videos'] if v['mpxId'] == mpxid)
 226                     break
 227                 except StopIteration:
 228                     continue
 229
 230             if info is None:
 231                 raise ExtractorError('Could not find video in playlists')
 232
 233             return {
 234                 '_type': 'url',
 235                 # We get the best quality video
 236                 'url': info['videoAssets'][-1]['publicUrl'],
 237                 'ie_key': 'ThePlatform',
 238             }
 239
 240
 241 class MSNBCIE(InfoExtractor):
 242     # https URLs redirect to corresponding http ones
 243     _VALID_URL = r'http://www\.msnbc\.com/[^/]+/watch/(?P<id>[^/]+)'
 244     _TEST = {
 245         'url': 'http://www.msnbc.com/all-in-with-chris-hayes/watch/the-chaotic-gop-immigration-vote-314487875924',
 246         'md5': '6d236bf4f3dddc226633ce6e2c3f814d',
 247         'info_dict': {
 248             'id': 'n_hayes_Aimm_140801_272214',
 249             'ext': 'mp4',
 250             'title': 'The chaotic GOP immigration vote',
 251             '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.',
 252             'thumbnail': 're:^https?://.*\.jpg$',
 253             'timestamp': 1406937606,
 254             'upload_date': '20140802',
 255             'categories': ['MSNBC/Topics/Franchise/Best of last night', 'MSNBC/Topics/General/Congress'],
 256         },
 257     }
 258
 259     def _real_extract(self, url):
 260         video_id = self._match_id(url)
 261         webpage = self._download_webpage(url, video_id)
 262         embed_url = self._html_search_meta('embedURL', webpage)
 263         return self.url_result(embed_url)