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

   1 from __future__ import unicode_literals
   2
   3 import re
   4 import json
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     compat_urllib_parse_urlparse,
   9     compat_urlparse,
  10     xpath_with_ns,
  11     compat_str,
  12 )
  13
  14
  15 class LivestreamIE(InfoExtractor):
  16     IE_NAME = 'livestream'
  17     _VALID_URL = r'http://new\.livestream\.com/.*?/(?P<event_name>.*?)(/videos/(?P<id>\d+))?/?$'
  18     _TEST = {
  19         'url': 'http://new.livestream.com/CoheedandCambria/WebsterHall/videos/4719370',
  20         'md5': '53274c76ba7754fb0e8d072716f2292b',
  21         'info_dict': {
  22             'id': '4719370',
  23             'ext': 'mp4',
  24             'title': 'Live from Webster Hall NYC',
  25             'upload_date': '20121012',
  26         }
  27     }
  28
  29     def _extract_video_info(self, video_data):
  30         video_url = video_data.get('progressive_url_hd') or video_data.get('progressive_url')
  31         return {
  32             'id': compat_str(video_data['id']),
  33             'url': video_url,
  34             'ext': 'mp4',
  35             'title': video_data['caption'],
  36             'thumbnail': video_data['thumbnail_url'],
  37             'upload_date': video_data['updated_at'].replace('-', '')[:8],
  38         }
  39
  40     def _real_extract(self, url):
  41         mobj = re.match(self._VALID_URL, url)
  42         video_id = mobj.group('id')
  43         event_name = mobj.group('event_name')
  44         webpage = self._download_webpage(url, video_id or event_name)
  45
  46         if video_id is None:
  47             # This is an event page:
  48             config_json = self._search_regex(
  49                 r'window.config = ({.*?});', webpage, 'window config')
  50             info = json.loads(config_json)['event']
  51             videos = [self._extract_video_info(video_data['data'])
  52                 for video_data in info['feed']['data'] if video_data['type'] == 'video']
  53             return self.playlist_result(videos, info['id'], info['full_name'])
  54         else:
  55             og_video = self._og_search_video_url(webpage, 'player url')
  56             query_str = compat_urllib_parse_urlparse(og_video).query
  57             query = compat_urlparse.parse_qs(query_str)
  58             api_url = query['play_url'][0].replace('.smil', '')
  59             info = json.loads(self._download_webpage(
  60                 api_url, video_id, 'Downloading video info'))
  61             return self._extract_video_info(info)
  62
  63
  64 # The original version of Livestream uses a different system
  65 class LivestreamOriginalIE(InfoExtractor):
  66     IE_NAME = 'livestream:original'
  67     _VALID_URL = r'https?://www\.livestream\.com/(?P<user>[^/]+)/video\?.*?clipId=(?P<id>.*?)(&|$)'
  68     _TEST = {
  69         'url': 'http://www.livestream.com/dealbook/video?clipId=pla_8aa4a3f1-ba15-46a4-893b-902210e138fb',
  70         'info_dict': {
  71             'id': 'pla_8aa4a3f1-ba15-46a4-893b-902210e138fb',
  72             'ext': 'flv',
  73             'title': 'Spark 1 (BitCoin) with Cameron Winklevoss & Tyler Winklevoss of Winklevoss Capital',
  74         },
  75         'params': {
  76             # rtmp
  77             'skip_download': True,
  78         },
  79     }
  80
  81     def _real_extract(self, url):
  82         mobj = re.match(self._VALID_URL, url)
  83         video_id = mobj.group('id')
  84         user = mobj.group('user')
  85         api_url = 'http://x{0}x.api.channel.livestream.com/2.0/clipdetails?extendedInfo=true&id={1}'.format(user, video_id)
  86
  87         info = self._download_xml(api_url, video_id)
  88         item = info.find('channel').find('item')
  89         ns = {'media': 'http://search.yahoo.com/mrss'}
  90         thumbnail_url = item.find(xpath_with_ns('media:thumbnail', ns)).attrib['url']
  91         # Remove the extension and number from the path (like 1.jpg)
  92         path = self._search_regex(r'(user-files/.+)_.*?\.jpg$', thumbnail_url, 'path')
  93
  94         return {
  95             'id': video_id,
  96             'title': item.find('title').text,
  97             'url': 'rtmp://extondemand.livestream.com/ondemand',
  98             'play_path': 'mp4:trans/dv15/mogulus-{0}.mp4'.format(path),
  99             'ext': 'flv',
 100             'thumbnail': thumbnail_url,
 101         }