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

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     ExtractorError,
   8 )
   9
  10
  11 class TumblrIE(InfoExtractor):
  12     _VALID_URL = r'http://(?P<blog_name>.*?)\.tumblr\.com/((post)|(video))/(?P<id>\d*)($|/)'
  13     _TEST = {
  14         'url': 'http://tatianamaslanydaily.tumblr.com/post/54196191430/orphan-black-dvd-extra-behind-the-scenes',
  15         'file': '54196191430.mp4',
  16         'md5': '479bb068e5b16462f5176a6828829767',
  17         'info_dict': {
  18             "title": "tatiana maslany news"
  19         }
  20     }
  21
  22     def _real_extract(self, url):
  23         m_url = re.match(self._VALID_URL, url)
  24         video_id = m_url.group('id')
  25         blog = m_url.group('blog_name')
  26
  27         url = 'http://%s.tumblr.com/post/%s/' % (blog, video_id)
  28         webpage = self._download_webpage(url, video_id)
  29
  30         re_video = r'src=\\x22(?P<video_url>http://%s\.tumblr\.com/video_file/%s/(.*?))\\x22 type=\\x22video/(?P<ext>.*?)\\x22' % (blog, video_id)
  31         video = re.search(re_video, webpage)
  32         if video is None:
  33             raise ExtractorError('Unable to extract video')
  34         video_url = video.group('video_url')
  35         ext = video.group('ext')
  36
  37         video_thumbnail = self._search_regex(
  38             r'posters.*?\[\\x22(.*?)\\x22',
  39             webpage, 'thumbnail', fatal=False)  # We pick the first poster
  40         if video_thumbnail:
  41             video_thumbnail = video_thumbnail.replace('\\\\/', '/')
  42
  43         # The only place where you can get a title, it's not complete,
  44         # but searching in other places doesn't work for all videos
  45         video_title = self._html_search_regex(r'<title>(?P<title>.*?)(?: \| Tumblr)?</title>',
  46             webpage, 'title', flags=re.DOTALL)
  47
  48         return [{'id': video_id,
  49                  'url': video_url,
  50                  'title': video_title,
  51                  'thumbnail': video_thumbnail,
  52                  'ext': ext
  53                  }]