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     orderedSet,
  13 )
  14
  15
  16 class LivestreamIE(InfoExtractor):
  17     IE_NAME = 'livestream'
  18     _VALID_URL = r'http://new\.livestream\.com/.*?/(?P<event_name>.*?)(/videos/(?P<id>\d+))?/?$'
  19     _TEST = {
  20         'url': 'http://new.livestream.com/CoheedandCambria/WebsterHall/videos/4719370',
  21         'md5': '53274c76ba7754fb0e8d072716f2292b',
  22         'info_dict': {
  23             'id': '4719370',
  24             'ext': 'mp4',
  25             'title': 'Live from Webster Hall NYC',
  26             'upload_date': '20121012',
  27         }
  28     }
  29
  30     def _extract_video_info(self, video_data):
  31         video_url = video_data.get('progressive_url_hd') or video_data.get('progressive_url')
  32         return {
  33             'id': compat_str(video_data['id']),
  34             'url': video_url,
  35             'ext': 'mp4',
  36             'title': video_data['caption'],
  37             'thumbnail': video_data['thumbnail_url'],
  38             'upload_date': video_data['updated_at'].replace('-', '')[:8],
  39         }
  40
  41     def _real_extract(self, url):
  42         mobj = re.match(self._VALID_URL, url)
  43         video_id = mobj.group('id')
  44         event_name = mobj.group('event_name')
  45         webpage = self._download_webpage(url, video_id or event_name)
  46
  47         if video_id is None:
  48             # This is an event page:
  49             config_json = self._search_regex(
  50                 r'window.config = ({.*?});', webpage, 'window config')
  51             info = json.loads(config_json)['event']
  52             videos = [self._extract_video_info(video_data['data'])
  53                 for video_data in info['feed']['data'] if video_data['type'] == 'video']
  54             return self.playlist_result(videos, info['id'], info['full_name'])
  55         else:
  56             og_video = self._og_search_video_url(webpage, 'player url')
  57             query_str = compat_urllib_parse_urlparse(og_video).query
  58             query = compat_urlparse.parse_qs(query_str)
  59             api_url = query['play_url'][0].replace('.smil', '')
  60             info = json.loads(self._download_webpage(
  61                 api_url, video_id, 'Downloading video info'))
  62             return self._extract_video_info(info)
  63
  64
  65 # The original version of Livestream uses a different system
  66 class LivestreamOriginalIE(InfoExtractor):
  67     IE_NAME = 'livestream:original'
  68     _VALID_URL = r'''(?x)https?://www\.livestream\.com/
  69         (?P<user>[^/]+)/(?P<type>video|folder)
  70         (?:\?.*?Id=|/)(?P<id>.*?)(&|$)
  71         '''
  72     _TEST = {
  73         'url': 'http://www.livestream.com/dealbook/video?clipId=pla_8aa4a3f1-ba15-46a4-893b-902210e138fb',
  74         'info_dict': {
  75             'id': 'pla_8aa4a3f1-ba15-46a4-893b-902210e138fb',
  76             'ext': 'flv',
  77             'title': 'Spark 1 (BitCoin) with Cameron Winklevoss & Tyler Winklevoss of Winklevoss Capital',
  78         },
  79         'params': {
  80             # rtmp
  81             'skip_download': True,
  82         },
  83     }
  84
  85     def _extract_video(self, user, video_id):
  86         api_url = 'http://x{0}x.api.channel.livestream.com/2.0/clipdetails?extendedInfo=true&id={1}'.format(user, video_id)
  87
  88         info = self._download_xml(api_url, video_id)
  89         item = info.find('channel').find('item')
  90         ns = {'media': 'http://search.yahoo.com/mrss'}
  91         thumbnail_url = item.find(xpath_with_ns('media:thumbnail', ns)).attrib['url']
  92         # Remove the extension and number from the path (like 1.jpg)
  93         path = self._search_regex(r'(user-files/.+)_.*?\.jpg$', thumbnail_url, 'path')
  94
  95         return {
  96             'id': video_id,
  97             'title': item.find('title').text,
  98             'url': 'rtmp://extondemand.livestream.com/ondemand',
  99             'play_path': 'mp4:trans/dv15/mogulus-{0}.mp4'.format(path),
 100             'ext': 'flv',
 101             'thumbnail': thumbnail_url,
 102         }
 103
 104     def _extract_folder(self, url, folder_id):
 105         webpage = self._download_webpage(url, folder_id)
 106         urls = orderedSet(re.findall(r'<a href="(https?://livestre\.am/.*?)"', webpage))
 107
 108         return {
 109             '_type': 'playlist',
 110             'id': folder_id,
 111             'entries': [{
 112                 '_type': 'url',
 113                 'url': video_url,
 114             } for video_url in urls],
 115         }
 116
 117     def _real_extract(self, url):
 118         mobj = re.match(self._VALID_URL, url)
 119         id = mobj.group('id')
 120         user = mobj.group('user')
 121         url_type = mobj.group('type')
 122         if url_type == 'folder':
 123             return self._extract_folder(url, id)
 124         else:
 125             return self._extract_video(user, id)
 126
 127
 128 # The server doesn't support HEAD request, the generic extractor can't detect
 129 # the redirection
 130 class LivestreamShortenerIE(InfoExtractor):
 131     IE_NAME = 'livestream:shortener'
 132     IE_DESC = False  # Do not list
 133     _VALID_URL = r'https?://livestre\.am/(?P<id>.+)'
 134
 135     def _real_extract(self, url):
 136         mobj = re.match(self._VALID_URL, url)
 137         id = mobj.group('id')
 138         webpage = self._download_webpage(url, id)
 139
 140         return {
 141             '_type': 'url',
 142             'url': self._og_search_url(webpage),
 143         }