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

   1 import re
   2 import json
   3 import xml.etree.ElementTree
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     compat_urllib_parse,
   8     find_xpath_attr,
   9     compat_urlparse,
  10 )
  11
  12 class BrightcoveIE(InfoExtractor):
  13     _VALID_URL = r'https?://.*brightcove\.com/(services|viewer).*\?(?P<query>.*)'
  14     _FEDERATED_URL_TEMPLATE = 'http://c.brightcove.com/services/viewer/htmlFederated?%s'
  15     _PLAYLIST_URL_TEMPLATE = 'http://c.brightcove.com/services/json/experience/runtime/?command=get_programming_for_experience&playerKey=%s'
  16
  17     # There is a test for Brigtcove in GenericIE, that way we test both the download
  18     # and the detection of videos, and we don't have to find an URL that is always valid
  19
  20     @classmethod
  21     def _build_brighcove_url(cls, object_str):
  22         """
  23         Build a Brightcove url from a xml string containing
  24         <object class="BrightcoveExperience">{params}</object>
  25         """
  26         object_doc = xml.etree.ElementTree.fromstring(object_str)
  27         assert u'BrightcoveExperience' in object_doc.attrib['class']
  28         params = {'flashID': object_doc.attrib['id'],
  29                   'playerID': find_xpath_attr(object_doc, './param', 'name', 'playerID').attrib['value'],
  30                   }
  31         playerKey = find_xpath_attr(object_doc, './param', 'name', 'playerKey')
  32         # Not all pages define this value
  33         if playerKey is not None:
  34             params['playerKey'] = playerKey.attrib['value']
  35         videoPlayer = find_xpath_attr(object_doc, './param', 'name', '@videoPlayer')
  36         if videoPlayer is not None:
  37             params['@videoPlayer'] = videoPlayer.attrib['value']
  38         data = compat_urllib_parse.urlencode(params)
  39         return cls._FEDERATED_URL_TEMPLATE % data
  40
  41     def _real_extract(self, url):
  42         mobj = re.match(self._VALID_URL, url)
  43         query_str = mobj.group('query')
  44         query = compat_urlparse.parse_qs(query_str)
  45
  46         videoPlayer = query.get('@videoPlayer')
  47         if videoPlayer:
  48             return self._get_video_info(videoPlayer[0], query_str)
  49         else:
  50             player_key = query['playerKey']
  51             return self._get_playlist_info(player_key[0])
  52
  53     def _get_video_info(self, video_id, query):
  54         request_url = self._FEDERATED_URL_TEMPLATE % query
  55         webpage = self._download_webpage(request_url, video_id)
  56
  57         self.report_extraction(video_id)
  58         info = self._search_regex(r'var experienceJSON = ({.*?});', webpage, 'json')
  59         info = json.loads(info)['data']
  60         video_info = info['programmedContent']['videoPlayer']['mediaDTO']
  61
  62         return self._extract_video_info(video_info)
  63
  64     def _get_playlist_info(self, player_key):
  65         playlist_info = self._download_webpage(self._PLAYLIST_URL_TEMPLATE % player_key,
  66                                                player_key, u'Downloading playlist information')
  67
  68         playlist_info = json.loads(playlist_info)['videoList']
  69         videos = [self._extract_video_info(video_info) for video_info in playlist_info['mediaCollectionDTO']['videoDTOs']]
  70
  71         return self.playlist_result(videos, playlist_id=playlist_info['id'],
  72                                     playlist_title=playlist_info['mediaCollectionDTO']['displayName'])
  73
  74     def _extract_video_info(self, video_info):
  75         renditions = video_info['renditions']
  76         renditions = sorted(renditions, key=lambda r: r['size'])
  77         best_format = renditions[-1]
  78
  79         return {'id': video_info['id'],
  80                 'title': video_info['displayName'],
  81                 'url': best_format['defaultURL'],
  82                 'ext': 'mp4',
  83                 'description': video_info.get('shortDescription'),
  84                 'thumbnail': video_info.get('videoStillURL') or video_info.get('thumbnailURL'),
  85                 'uploader': video_info.get('publisherName'),
  86                 }