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

   1 # -*- coding: utf-8 -*-
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7
   8
   9 class TumblrIE(InfoExtractor):
  10     _VALID_URL = r'http://(?P<blog_name>.*?)\.tumblr\.com/(?:post|video)/(?P<id>[0-9]+)(?:$|[/?#])'
  11     _TESTS = [{
  12         'url': 'http://tatianamaslanydaily.tumblr.com/post/54196191430/orphan-black-dvd-extra-behind-the-scenes',
  13         'md5': '479bb068e5b16462f5176a6828829767',
  14         'info_dict': {
  15             'id': '54196191430',
  16             'ext': 'mp4',
  17             'title': 'tatiana maslany news, Orphan Black || DVD extra - behind the scenes ↳...',
  18             'description': 'md5:37db8211e40b50c7c44e95da14f630b7',
  19             'thumbnail': 're:http://.*\.jpg',
  20         }
  21     }, {
  22         'url': 'http://5sostrum.tumblr.com/post/90208453769/yall-forgetting-the-greatest-keek-of-them-all',
  23         'md5': 'bf348ef8c0ef84fbf1cbd6fa6e000359',
  24         'info_dict': {
  25             'id': '90208453769',
  26             'ext': 'mp4',
  27             'title': '5SOS STRUM ;]',
  28             'description': 'md5:dba62ac8639482759c8eb10ce474586a',
  29             'thumbnail': 're:http://.*\.jpg',
  30         }
  31     }]
  32
  33     def _real_extract(self, url):
  34         m_url = re.match(self._VALID_URL, url)
  35         video_id = m_url.group('id')
  36         blog = m_url.group('blog_name')
  37
  38         url = 'http://%s.tumblr.com/post/%s/' % (blog, video_id)
  39         webpage = self._download_webpage(url, video_id)
  40
  41         iframe_url = self._search_regex(
  42             r'src=\'(https?://www\.tumblr\.com/video/[^\']+)\'',
  43             webpage, 'iframe url')
  44         iframe = self._download_webpage(iframe_url, video_id)
  45         video_url = self._search_regex(r'<source src="([^"]+)"',
  46                                        iframe, 'video url')
  47
  48         # The only place where you can get a title, it's not complete,
  49         # but searching in other places doesn't work for all videos
  50         video_title = self._html_search_regex(
  51             r'(?s)<title>(?P<title>.*?)(?: \| Tumblr)?</title>',
  52             webpage, 'title')
  53
  54         return {
  55             'id': video_id,
  56             'url': video_url,
  57             'ext': 'mp4',
  58             'title': video_title,
  59             'description': self._og_search_description(webpage, default=None),
  60             'thumbnail': self._og_search_thumbnail(webpage, default=None),
  61         }