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 from .pornhub import PornHubIE
   8
   9
  10 class TumblrIE(InfoExtractor):
  11     _VALID_URL = r'http://(?P<blog_name>.*?)\.tumblr\.com/(?:post|video)/(?P<id>[0-9]+)(?:$|[/?#])'
  12     _TESTS = [{
  13         'url': 'http://tatianamaslanydaily.tumblr.com/post/54196191430/orphan-black-dvd-extra-behind-the-scenes',
  14         'md5': '479bb068e5b16462f5176a6828829767',
  15         'info_dict': {
  16             'id': '54196191430',
  17             'ext': 'mp4',
  18             'title': 'tatiana maslany news, Orphan Black || DVD extra - behind the scenes ↳...',
  19             'description': 'md5:37db8211e40b50c7c44e95da14f630b7',
  20             'thumbnail': 're:http://.*\.jpg',
  21         }
  22     }, {
  23         'url': 'http://5sostrum.tumblr.com/post/90208453769/yall-forgetting-the-greatest-keek-of-them-all',
  24         'md5': 'bf348ef8c0ef84fbf1cbd6fa6e000359',
  25         'info_dict': {
  26             'id': '90208453769',
  27             'ext': 'mp4',
  28             'title': '5SOS STRUM ;]',
  29             'description': 'md5:dba62ac8639482759c8eb10ce474586a',
  30             'thumbnail': 're:http://.*\.jpg',
  31         }
  32     }, {
  33         'url': 'http://naked-yogi.tumblr.com/post/118312946248/naked-smoking-stretching',
  34         'md5': 'de07e5211d60d4f3a2c3df757ea9f6ab',
  35         'info_dict': {
  36             'id': 'Wmur',
  37             'ext': 'mp4',
  38             'title': 'naked smoking & stretching',
  39             'upload_date': '20150506',
  40             'timestamp': 1430931613,
  41         },
  42         'add_ie': ['Vidme'],
  43     }]
  44
  45     def _real_extract(self, url):
  46         m_url = re.match(self._VALID_URL, url)
  47         video_id = m_url.group('id')
  48         blog = m_url.group('blog_name')
  49
  50         url = 'http://%s.tumblr.com/post/%s/' % (blog, video_id)
  51         webpage = self._download_webpage(url, video_id)
  52
  53         vid_me_embed_url = self._search_regex(
  54             r'src=[\'"](https?://vid\.me/[^\'"]+)[\'"]',
  55             webpage, 'vid.me embed', default=None)
  56         if vid_me_embed_url is not None:
  57             return self.url_result(vid_me_embed_url, 'Vidme')
  58
  59         pornhub_url = PornHubIE._extract_url(webpage)
  60         if pornhub_url:
  61             return self.url_result(pornhub_url, 'PornHub')
  62
  63         iframe_url = self._search_regex(
  64             r'src=\'(https?://www\.tumblr\.com/video/[^\']+)\'',
  65             webpage, 'iframe url')
  66         iframe = self._download_webpage(iframe_url, video_id)
  67         video_url = self._search_regex(r'<source src="([^"]+)"',
  68                                        iframe, 'video url')
  69
  70         # The only place where you can get a title, it's not complete,
  71         # but searching in other places doesn't work for all videos
  72         video_title = self._html_search_regex(
  73             r'(?s)<title>(?P<title>.*?)(?: \| Tumblr)?</title>',
  74             webpage, 'title')
  75
  76         return {
  77             'id': video_id,
  78             'url': video_url,
  79             'ext': 'mp4',
  80             'title': video_title,
  81             'description': self._og_search_description(webpage, default=None),
  82             'thumbnail': self._og_search_thumbnail(webpage, default=None),
  83         }