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

   1 import re
   2 import json
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     ExtractorError,
   7 )
   8
   9 class VevoIE(InfoExtractor):
  10     """
  11     Accepts urls from vevo.com or in the format 'vevo:{id}'
  12     (currently used by MTVIE)
  13     """
  14     _VALID_URL = r'((http://www.vevo.com/watch/.*?/.*?/)|(vevo:))(?P<id>.*)$'
  15     _TEST = {
  16         u'url': u'http://www.vevo.com/watch/hurts/somebody-to-die-for/GB1101300280',
  17         u'file': u'GB1101300280.mp4',
  18         u'md5': u'06bea460acb744eab74a9d7dcb4bfd61',
  19         u'info_dict': {
  20             u"upload_date": u"20130624",
  21             u"uploader": u"Hurts",
  22             u"title": u"Somebody to Die For"
  23         }
  24     }
  25
  26     def _real_extract(self, url):
  27         mobj = re.match(self._VALID_URL, url)
  28         video_id = mobj.group('id')
  29
  30         json_url = 'http://www.vevo.com/data/video/%s' % video_id
  31         base_url = 'http://smil.lvl3.vevo.com'
  32         videos_url = '%s/Video/V2/VFILE/%s/%sr.smil' % (base_url, video_id, video_id.lower())
  33         info_json = self._download_webpage(json_url, video_id, u'Downloading json info')
  34         links_webpage = self._download_webpage(videos_url, video_id, u'Downloading videos urls')
  35
  36         self.report_extraction(video_id)
  37         video_info = json.loads(info_json)
  38         m_urls = list(re.finditer(r'<video src="(?P<ext>.*?):/?(?P<url>.*?)"', links_webpage))
  39         if m_urls is None or len(m_urls) == 0:
  40             raise ExtractorError(u'Unable to extract video url')
  41         # They are sorted from worst to best quality
  42         m_url = m_urls[-1]
  43         video_url = base_url + '/' + m_url.group('url')
  44         ext = m_url.group('ext')
  45
  46         return {'url': video_url,
  47                 'ext': ext,
  48                 'id': video_id,
  49                 'title': video_info['title'],
  50                 'thumbnail': video_info['img'],
  51                 'upload_date': video_info['launchDate'].replace('/',''),
  52                 'uploader': video_info['Artists'][0]['title'],
  53                 }