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