_ Git - youtube-dl/blob - youtube_dl/extractor/nbc.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4 import json
   5
   6 from .common import InfoExtractor
   7 from ..compat import (
   8     compat_str,
   9     compat_HTTPError,
  10 )
  11 from ..utils import (
  12     ExtractorError,
  13     find_xpath_attr,
  14 )
  15
  16
  17 class NBCIE(InfoExtractor):
  18     _VALID_URL = r'http://www\.nbc\.com/(?:[^/]+/)+(?P<id>n?\d+)'
  19
  20     _TESTS = [
  21         {
  22             'url': 'http://www.nbc.com/chicago-fire/video/i-am-a-firefighter/2734188',
  23             # md5 checksum is not stable
  24             'info_dict': {
  25                 'id': 'bTmnLCvIbaaH',
  26                 'ext': 'flv',
  27                 'title': 'I Am a Firefighter',
  28                 'description': 'An emergency puts Dawson\'sf irefighter skills to the ultimate test in this four-part digital series.',
  29             },
  30         },
  31         {
  32             'url': 'http://www.nbc.com/the-tonight-show/episodes/176',
  33             'info_dict': {
  34                 'id': 'XwU9KZkp98TH',
  35                 'ext': 'flv',
  36                 'title': 'Ricky Gervais, Steven Van Zandt, ILoveMakonnen',
  37                 'description': 'A brand new episode of The Tonight Show welcomes Ricky Gervais, Steven Van Zandt and ILoveMakonnen.',
  38             },
  39             'skip': 'Only works from US',
  40         },
  41     ]
  42
  43     def _real_extract(self, url):
  44         video_id = self._match_id(url)
  45         webpage = self._download_webpage(url, video_id)
  46         theplatform_url = self._search_regex(
  47             '(?:class="video-player video-player-full" data-mpx-url|class="player" src)="(.*?)"',
  48             webpage, 'theplatform url').replace('_no_endcard', '')
  49         if theplatform_url.startswith('//'):
  50             theplatform_url = 'http:' + theplatform_url
  51         return self.url_result(theplatform_url)
  52
  53
  54 class NBCNewsIE(InfoExtractor):
  55     _VALID_URL = r'''(?x)https?://(?:www\.)?nbcnews\.com/
  56         (?:video/.+?/(?P<id>\d+)|
  57         (?:feature|nightly-news)/[^/]+/(?P<title>.+))
  58         '''
  59
  60     _TESTS = [
  61         {
  62             'url': 'http://www.nbcnews.com/video/nbc-news/52753292',
  63             'md5': '47abaac93c6eaf9ad37ee6c4463a5179',
  64             'info_dict': {
  65                 'id': '52753292',
  66                 'ext': 'flv',
  67                 'title': 'Crew emerges after four-month Mars food study',
  68                 'description': 'md5:24e632ffac72b35f8b67a12d1b6ddfc1',
  69             },
  70         },
  71         {
  72             'url': 'http://www.nbcnews.com/feature/edward-snowden-interview/how-twitter-reacted-snowden-interview-n117236',
  73             'md5': 'b2421750c9f260783721d898f4c42063',
  74             'info_dict': {
  75                 'id': 'I1wpAI_zmhsQ',
  76                 'ext': 'mp4',
  77                 'title': 'How Twitter Reacted To The Snowden Interview',
  78                 'description': 'md5:65a0bd5d76fe114f3c2727aa3a81fe64',
  79             },
  80             'add_ie': ['ThePlatform'],
  81         },
  82         {
  83             'url': 'http://www.nbcnews.com/feature/dateline-full-episodes/full-episode-family-business-n285156',
  84             'md5': 'fdbf39ab73a72df5896b6234ff98518a',
  85             'info_dict': {
  86                 'id': 'Wjf9EDR3A_60',
  87                 'ext': 'mp4',
  88                 'title': 'FULL EPISODE: Family Business',
  89                 'description': 'md5:757988edbaae9d7be1d585eb5d55cc04',
  90             },
  91         },
  92         {
  93             'url': 'http://www.nbcnews.com/nightly-news/video/nightly-news-with-brian-williams-full-broadcast-february-4-394064451844',
  94             'md5': 'b5dda8cddd8650baa0dcb616dd2cf60d',
  95             'info_dict': {
  96                 'id': 'sekXqyTVnmN3',
  97                 'ext': 'mp4',
  98                 'title': 'Nightly News with Brian Williams Full Broadcast (February 4)',
  99                 'description': 'md5:1c10c1eccbe84a26e5debb4381e2d3c5',
 100             },
 101         },
 102     ]
 103
 104     def _real_extract(self, url):
 105         mobj = re.match(self._VALID_URL, url)
 106         video_id = mobj.group('id')
 107         if video_id is not None:
 108             all_info = self._download_xml('http://www.nbcnews.com/id/%s/displaymode/1219' % video_id, video_id)
 109             info = all_info.find('video')
 110
 111             return {
 112                 'id': video_id,
 113                 'title': info.find('headline').text,
 114                 'ext': 'flv',
 115                 'url': find_xpath_attr(info, 'media', 'type', 'flashVideo').text,
 116                 'description': compat_str(info.find('caption').text),
 117                 'thumbnail': find_xpath_attr(info, 'media', 'type', 'thumbnail').text,
 118             }
 119         else:
 120             # "feature" and "nightly-news" pages use theplatform.com
 121             title = mobj.group('title')
 122             webpage = self._download_webpage(url, title)
 123             bootstrap_json = self._search_regex(
 124                 r'var\s+(?:bootstrapJson|playlistData)\s*=\s*({.+});?\s*$',
 125                 webpage, 'bootstrap json', flags=re.MULTILINE)
 126             bootstrap = self._parse_json(bootstrap_json, video_id)
 127             info = bootstrap['results'][0]['video']
 128             mpxid = info['mpxId']
 129
 130             base_urls = [
 131                 info['fallbackPlaylistUrl'],
 132                 info['associatedPlaylistUrl'],
 133             ]
 134
 135             for base_url in base_urls:
 136                 if not base_url:
 137                     continue
 138                 playlist_url = base_url + '?form=MPXNBCNewsAPI'
 139
 140                 try:
 141                     all_videos = self._download_json(playlist_url, title)
 142                 except ExtractorError as ee:
 143                     if isinstance(ee.cause, compat_HTTPError):
 144                         continue
 145                     raise
 146
 147                 if not all_videos or 'videos' not in all_videos:
 148                     continue
 149
 150                 try:
 151                     info = next(v for v in all_videos['videos'] if v['mpxId'] == mpxid)
 152                     break
 153                 except StopIteration:
 154                     continue
 155
 156             if info is None:
 157                 raise ExtractorError('Could not find video in playlists')
 158
 159             return {
 160                 '_type': 'url',
 161                 # We get the best quality video
 162                 'url': info['videoAssets'][-1]['publicUrl'],
 163                 'ie_key': 'ThePlatform',
 164             }