]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/vevo.py
Fix description of the modularization.
[youtubedl] / 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 Accecps 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
16 def _real_extract(self, url):
17 mobj = re.match(self._VALID_URL, url)
18 video_id = mobj.group('id')
19
20 json_url = 'http://www.vevo.com/data/video/%s' % video_id
21 base_url = 'http://smil.lvl3.vevo.com'
22 videos_url = '%s/Video/V2/VFILE/%s/%sr.smil' % (base_url, video_id, video_id.lower())
23 info_json = self._download_webpage(json_url, video_id, u'Downloading json info')
24 links_webpage = self._download_webpage(videos_url, video_id, u'Downloading videos urls')
25
26 self.report_extraction(video_id)
27 video_info = json.loads(info_json)
28 m_urls = list(re.finditer(r'<video src="(?P<ext>.*?):(?P<url>.*?)"', links_webpage))
29 if m_urls is None or len(m_urls) == 0:
30 raise ExtractorError(u'Unable to extract video url')
31 # They are sorted from worst to best quality
32 m_url = m_urls[-1]
33 video_url = base_url + m_url.group('url')
34 ext = m_url.group('ext')
35
36 return {'url': video_url,
37 'ext': ext,
38 'id': video_id,
39 'title': video_info['title'],
40 'thumbnail': video_info['img'],
41 'upload_date': video_info['launchDate'].replace('/',''),
42 'uploader': video_info['Artists'][0]['title'],
43 }