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

   1 import re
   2 import json
   3
   4 from .common import InfoExtractor
   5 from ..utils import compat_urllib_parse_urlparse, compat_urlparse
   6
   7
   8 class LivestreamIE(InfoExtractor):
   9     _VALID_URL = r'http://new.livestream.com/.*?/(?P<event_name>.*?)(/videos/(?P<id>\d+))?/?$'
  10     _TEST = {
  11         u'url': u'http://new.livestream.com/CoheedandCambria/WebsterHall/videos/4719370',
  12         u'file': u'4719370.mp4',
  13         u'md5': u'0d2186e3187d185a04b3cdd02b828836',
  14         u'info_dict': {
  15             u'title': u'Live from Webster Hall NYC',
  16             u'upload_date': u'20121012',
  17         }
  18     }
  19
  20     def _extract_video_info(self, video_data):
  21         video_url = video_data.get('progressive_url_hd') or video_data.get('progressive_url')
  22         return {'id': video_data['id'],
  23                 'url': video_url,
  24                 'ext': 'mp4',
  25                 'title': video_data['caption'],
  26                 'thumbnail': video_data['thumbnail_url'],
  27                 'upload_date': video_data['updated_at'].replace('-','')[:8],
  28                 }
  29
  30     def _real_extract(self, url):
  31         mobj = re.match(self._VALID_URL, url)
  32         video_id = mobj.group('id')
  33         event_name = mobj.group('event_name')
  34         webpage = self._download_webpage(url, video_id or event_name)
  35
  36         if video_id is None:
  37             # This is an event page:
  38             api_url = self._search_regex(r'event_design_eventId: \'(.+?)\'',
  39                                          webpage, 'api url')
  40             info = json.loads(self._download_webpage(api_url, event_name,
  41                                                      u'Downloading event info'))
  42             videos = [self._extract_video_info(video_data['data'])
  43                 for video_data in info['feed']['data'] if video_data['type'] == u'video']
  44             return self.playlist_result(videos, info['id'], info['full_name'])
  45         else:
  46             og_video = self._og_search_video_url(webpage, name=u'player url')
  47             query_str = compat_urllib_parse_urlparse(og_video).query
  48             query = compat_urlparse.parse_qs(query_str)
  49             api_url = query['play_url'][0].replace('.smil', '')
  50             info = json.loads(self._download_webpage(api_url, video_id,
  51                                                      u'Downloading video info'))
  52             return self._extract_video_info(info)