youtube_dl/extractor/canalplus.py

   1 # encoding: utf-8
   2 import re
   3 import xml.etree.ElementTree
   4
   5 from .common import InfoExtractor
   6 from ..utils import unified_strdate
   7
   8
   9 class CanalplusIE(InfoExtractor):
  10     _VALID_URL = r'https?://(www\.canalplus\.fr/.*?/(?P<path>.*)|player\.canalplus\.fr/#/(?P<id>\d+))'
  11     _VIDEO_INFO_TEMPLATE = 'http://service.canal-plus.com/video/rest/getVideosLiees/cplus/%s'
  12     IE_NAME = u'canalplus.fr'
  13
  14     _TEST = {
  15         u'url': u'http://www.canalplus.fr/c-infos-documentaires/pid1830-c-zapping.html?vid=922470',
  16         u'file': u'922470.flv',
  17         u'info_dict': {
  18             u'title': u'Zapping - 26/08/13',
  19             u'description': u'Le meilleur de toutes les chaînes, tous les jours.\nEmission du 26 août 2013',
  20             u'upload_date': u'20130826',
  21         },
  22         u'params': {
  23             u'skip_download': True,
  24         },
  25     }
  26
  27     def _real_extract(self, url):
  28         mobj = re.match(self._VALID_URL, url)
  29         video_id = mobj.groupdict().get('id')
  30         if video_id is None:
  31             webpage = self._download_webpage(url, mobj.group('path'))
  32             video_id = self._search_regex(r'videoId = "(\d+)";', webpage, u'video id')
  33         info_url = self._VIDEO_INFO_TEMPLATE % video_id
  34         info_page = self._download_webpage(info_url,video_id,
  35                                            u'Downloading video info')
  36
  37         self.report_extraction(video_id)
  38         doc = xml.etree.ElementTree.fromstring(info_page.encode('utf-8'))
  39         video_info = [video for video in doc if video.find('ID').text == video_id][0]
  40         infos = video_info.find('INFOS')
  41         media = video_info.find('MEDIA')
  42         formats = [media.find('VIDEOS/%s' % format)
  43             for format in ['BAS_DEBIT', 'HAUT_DEBIT', 'HD']]
  44         video_url = [format.text for format in formats if format is not None][-1]
  45
  46         return {'id': video_id,
  47                 'title': u'%s - %s' % (infos.find('TITRAGE/TITRE').text,
  48                                        infos.find('TITRAGE/SOUS_TITRE').text),
  49                 'url': video_url,
  50                 'ext': 'flv',
  51                 'upload_date': unified_strdate(infos.find('PUBLICATION/DATE').text),
  52                 'thumbnail': media.find('IMAGES/GRAND').text,
  53                 'description': infos.find('DESCRIPTION').text,
  54                 'view_count': int(infos.find('NB_VUES').text),
  55                 }