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 )
  14
  15
  16 class NBCIE(InfoExtractor):
  17     _VALID_URL = r'http://www\.nbc\.com/(?:[^/]+/)+(?P<id>n?\d+)'
  18
  19     _TESTS = [
  20         {
  21             'url': 'http://www.nbc.com/the-tonight-show/segments/112966',
  22             # md5 checksum is not stable
  23             'info_dict': {
  24                 'id': 'c9xnCo0YPOPH',
  25                 'ext': 'flv',
  26                 'title': 'Jimmy Fallon Surprises Fans at Ben & Jerry\'s',
  27                 'description': 'Jimmy gives out free scoops of his new "Tonight Dough" ice cream flavor by surprising customers at the Ben & Jerry\'s scoop shop.',
  28             },
  29         },
  30         {
  31             'url': 'http://www.nbc.com/the-tonight-show/episodes/176',
  32             'info_dict': {
  33                 'id': 'XwU9KZkp98TH',
  34                 'ext': 'flv',
  35                 'title': 'Ricky Gervais, Steven Van Zandt, ILoveMakonnen',
  36                 'description': 'A brand new episode of The Tonight Show welcomes Ricky Gervais, Steven Van Zandt and ILoveMakonnen.',
  37             },
  38             'skip': 'Only works from US',
  39         },
  40     ]
  41
  42     def _real_extract(self, url):
  43         video_id = self._match_id(url)
  44         webpage = self._download_webpage(url, video_id)
  45         theplatform_url = self._search_regex(
  46             '(?:class="video-player video-player-full" data-mpx-url|class="player" src)="(.*?)"',
  47             webpage, 'theplatform url').replace('_no_endcard', '')
  48         if theplatform_url.startswith('//'):
  49             theplatform_url = 'http:' + theplatform_url
  50         return self.url_result(theplatform_url)
  51
  52
  53 class NBCNewsIE(InfoExtractor):
  54     _VALID_URL = r'''(?x)https?://(?:www\.)?nbcnews\.com/
  55         (?:video/.+?/(?P<id>\d+)|
  56         (?:feature|nightly-news)/[^/]+/(?P<title>.+))
  57         '''
  58
  59     _TESTS = [
  60         {
  61             'url': 'http://www.nbcnews.com/video/nbc-news/52753292',
  62             'md5': '47abaac93c6eaf9ad37ee6c4463a5179',
  63             'info_dict': {
  64                 'id': '52753292',
  65                 'ext': 'flv',
  66                 'title': 'Crew emerges after four-month Mars food study',
  67                 'description': 'md5:24e632ffac72b35f8b67a12d1b6ddfc1',
  68             },
  69         },
  70         {
  71             'url': 'http://www.nbcnews.com/feature/edward-snowden-interview/how-twitter-reacted-snowden-interview-n117236',
  72             'md5': 'b2421750c9f260783721d898f4c42063',
  73             'info_dict': {
  74                 'id': 'I1wpAI_zmhsQ',
  75                 'ext': 'mp4',
  76                 'title': 'How Twitter Reacted To The Snowden Interview',
  77                 'description': 'md5:65a0bd5d76fe114f3c2727aa3a81fe64',
  78             },
  79             'add_ie': ['ThePlatform'],
  80         },
  81         {
  82             'url': 'http://www.nbcnews.com/feature/dateline-full-episodes/full-episode-family-business-n285156',
  83             'md5': 'fdbf39ab73a72df5896b6234ff98518a',
  84             'info_dict': {
  85                 'id': 'Wjf9EDR3A_60',
  86                 'ext': 'mp4',
  87                 'title': 'FULL EPISODE: Family Business',
  88                 'description': 'md5:757988edbaae9d7be1d585eb5d55cc04',
  89             },
  90         },
  91         {
  92             'url': 'http://www.nbcnews.com/nightly-news/video/nightly-news-with-brian-williams-full-broadcast-february-4-394064451844',
  93             'md5': 'b5dda8cddd8650baa0dcb616dd2cf60d',
  94             'info_dict': {
  95                 'id': 'sekXqyTVnmN3',
  96                 'ext': 'mp4',
  97                 'title': 'Nightly News with Brian Williams Full Broadcast (February 4)',
  98                 'description': 'md5:1c10c1eccbe84a26e5debb4381e2d3c5',
  99             },
 100         },
 101     ]
 102
 103     def _real_extract(self, url):
 104         mobj = re.match(self._VALID_URL, url)
 105         video_id = mobj.group('id')
 106         if video_id is not None:
 107             all_info = self._download_xml('http://www.nbcnews.com/id/%s/displaymode/1219' % video_id, video_id)
 108             info = all_info.find('video')
 109
 110             return {
 111                 'id': video_id,
 112                 'title': info.find('headline').text,
 113                 'ext': 'flv',
 114                 'url': find_xpath_attr(info, 'media', 'type', 'flashVideo').text,
 115                 'description': compat_str(info.find('caption').text),
 116                 'thumbnail': find_xpath_attr(info, 'media', 'type', 'thumbnail').text,
 117             }
 118         else:
 119             # "feature" and "nightly-news" pages use theplatform.com
 120             title = mobj.group('title')
 121             webpage = self._download_webpage(url, title)
 122             bootstrap_json = self._search_regex(
 123                 r'var\s+(?:bootstrapJson|playlistData)\s*=\s*({.+});?\s*$',
 124                 webpage, 'bootstrap json', flags=re.MULTILINE)
 125             bootstrap = self._parse_json(bootstrap_json, video_id)
 126             info = bootstrap['results'][0]['video']
 127             mpxid = info['mpxId']
 128
 129             base_urls = [
 130                 info['fallbackPlaylistUrl'],
 131                 info['associatedPlaylistUrl'],
 132             ]
 133
 134             for base_url in base_urls:
 135                 if not base_url:
 136                     continue
 137                 playlist_url = base_url + '?form=MPXNBCNewsAPI'
 138
 139                 try:
 140                     all_videos = self._download_json(playlist_url, title)
 141                 except ExtractorError as ee:
 142                     if isinstance(ee.cause, compat_HTTPError):
 143                         continue
 144                     raise
 145
 146                 if not all_videos or 'videos' not in all_videos:
 147                     continue
 148
 149                 try:
 150                     info = next(v for v in all_videos['videos'] if v['mpxId'] == mpxid)
 151                     break
 152                 except StopIteration:
 153                     continue
 154
 155             if info is None:
 156                 raise ExtractorError('Could not find video in playlists')
 157
 158             return {
 159                 '_type': 'url',
 160                 # We get the best quality video
 161                 'url': info['videoAssets'][-1]['publicUrl'],
 162                 'ie_key': 'ThePlatform',
 163             }