_ Git - youtube-dl/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 from .pornhub import PornHubIE
   8 from .vimeo import VimeoIE
   9
  10
  11 class TumblrIE(InfoExtractor):
  12     _VALID_URL = r'http://(?P<blog_name>.*?)\.tumblr\.com/(?:post|video)/(?P<id>[0-9]+)(?:$|[/?#])'
  13     _TESTS = [{
  14         'url': 'http://tatianamaslanydaily.tumblr.com/post/54196191430/orphan-black-dvd-extra-behind-the-scenes',
  15         'md5': '479bb068e5b16462f5176a6828829767',
  16         'info_dict': {
  17             'id': '54196191430',
  18             'ext': 'mp4',
  19             'title': 'tatiana maslany news, Orphan Black || DVD extra - behind the scenes ↳...',
  20             'description': 'md5:37db8211e40b50c7c44e95da14f630b7',
  21             'thumbnail': 're:http://.*\.jpg',
  22         }
  23     }, {
  24         'url': 'http://5sostrum.tumblr.com/post/90208453769/yall-forgetting-the-greatest-keek-of-them-all',
  25         'md5': 'bf348ef8c0ef84fbf1cbd6fa6e000359',
  26         'info_dict': {
  27             'id': '90208453769',
  28             'ext': 'mp4',
  29             'title': '5SOS STRUM ;]',
  30             'description': 'md5:dba62ac8639482759c8eb10ce474586a',
  31             'thumbnail': 're:http://.*\.jpg',
  32         }
  33     }, {
  34         'url': 'http://naked-yogi.tumblr.com/post/118312946248/naked-smoking-stretching',
  35         'md5': 'de07e5211d60d4f3a2c3df757ea9f6ab',
  36         'info_dict': {
  37             'id': 'Wmur',
  38             'ext': 'mp4',
  39             'title': 'naked smoking & stretching',
  40             'upload_date': '20150506',
  41             'timestamp': 1430931613,
  42         },
  43         'add_ie': ['Vidme'],
  44     }, {
  45         'url': 'http://camdamage.tumblr.com/post/98846056295/',
  46         'md5': 'a9e0c8371ea1ca306d6554e3fecf50b6',
  47         'info_dict': {
  48             'id': '105463834',
  49             'ext': 'mp4',
  50             'title': 'Cam Damage-HD 720p',
  51             'uploader': 'John Moyer',
  52             'uploader_id': 'user32021558',
  53         },
  54         'add_ie': ['Vimeo'],
  55     }]
  56
  57     def _real_extract(self, url):
  58         m_url = re.match(self._VALID_URL, url)
  59         video_id = m_url.group('id')
  60         blog = m_url.group('blog_name')
  61
  62         url = 'http://%s.tumblr.com/post/%s/' % (blog, video_id)
  63         webpage = self._download_webpage(url, video_id)
  64
  65         vid_me_embed_url = self._search_regex(
  66             r'src=[\'"](https?://vid\.me/[^\'"]+)[\'"]',
  67             webpage, 'vid.me embed', default=None)
  68         if vid_me_embed_url is not None:
  69             return self.url_result(vid_me_embed_url, 'Vidme')
  70
  71         pornhub_url = PornHubIE._extract_url(webpage)
  72         if pornhub_url:
  73             return self.url_result(pornhub_url, 'PornHub')
  74
  75         vimeo_url = VimeoIE._extract_vimeo_url(url, webpage)
  76         if vimeo_url:
  77             return self.url_result(vimeo_url, 'Vimeo')
  78
  79         iframe_url = self._search_regex(
  80             r'src=\'(https?://www\.tumblr\.com/video/[^\']+)\'',
  81             webpage, 'iframe url')
  82         iframe = self._download_webpage(iframe_url, video_id)
  83         video_url = self._search_regex(r'<source src="([^"]+)"',
  84                                        iframe, 'video url')
  85
  86         # The only place where you can get a title, it's not complete,
  87         # but searching in other places doesn't work for all videos
  88         video_title = self._html_search_regex(
  89             r'(?s)<title>(?P<title>.*?)(?: \| Tumblr)?</title>',
  90             webpage, 'title')
  91
  92         return {
  93             'id': video_id,
  94             'url': video_url,
  95             'ext': 'mp4',
  96             'title': video_title,
  97             'description': self._og_search_description(webpage, default=None),
  98             'thumbnail': self._og_search_thumbnail(webpage, default=None),
  99         }