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         'url': 'http://hdvideotest.tumblr.com/post/130323439814/test-description-for-my-hd-video',
  33         'md5': '99a84522f60972bf064a0b80f87bcbb5',
  34         'info_dict': {
  35             'id': '130323439814',
  36             'ext': 'mp4',
  37             'title': 'HD Video Testing \u2014 Test description for my HD video',
  38             'description': 'md5:97cc3ab5fcd27ee4af6356701541319c',
  39             'thumbnail': 're:http://.*\.jpg',
  40         },
  41         'params': {
  42             'format': 'sd',
  43         },
  44     }, {
  45         'url': 'http://hdvideotest.tumblr.com/post/130323439814/test-description-for-my-hd-video',
  46         'md5': '7ae503065ad150122dc3089f8cf1546c',
  47         'info_dict': {
  48             'id': '130323439814',
  49             'ext': 'mp4',
  50             'title': 'HD Video Testing \u2014 Test description for my HD video',
  51             'description': 'md5:97cc3ab5fcd27ee4af6356701541319c',
  52             'thumbnail': 're:http://.*\.jpg',
  53         },
  54         'params': {
  55             'format': 'hd',
  56         },
  57     }, {
  58         'url': 'http://naked-yogi.tumblr.com/post/118312946248/naked-smoking-stretching',
  59         'md5': 'de07e5211d60d4f3a2c3df757ea9f6ab',
  60         'info_dict': {
  61             'id': 'Wmur',
  62             'ext': 'mp4',
  63             'title': 'naked smoking & stretching',
  64             'upload_date': '20150506',
  65             'timestamp': 1430931613,
  66         },
  67         'add_ie': ['Vidme'],
  68     }, {
  69         'url': 'http://camdamage.tumblr.com/post/98846056295/',
  70         'md5': 'a9e0c8371ea1ca306d6554e3fecf50b6',
  71         'info_dict': {
  72             'id': '105463834',
  73             'ext': 'mp4',
  74             'title': 'Cam Damage-HD 720p',
  75             'uploader': 'John Moyer',
  76             'uploader_id': 'user32021558',
  77         },
  78         'add_ie': ['Vimeo'],
  79     }]
  80
  81     def _real_extract(self, url):
  82         m_url = re.match(self._VALID_URL, url)
  83         video_id = m_url.group('id')
  84         blog = m_url.group('blog_name')
  85
  86         video_urls = []
  87
  88         url = 'http://%s.tumblr.com/post/%s/' % (blog, video_id)
  89         webpage, urlh = self._download_webpage_handle(url, video_id)
  90
  91         iframe_url = self._search_regex(
  92             r'src=\'(https?://www\.tumblr\.com/video/[^\']+)\'',
  93             webpage, 'iframe url', default=None)
  94         if iframe_url is None:
  95             return self.url_result(urlh.geturl(), 'Generic')
  96
  97         iframe = self._download_webpage(iframe_url, video_id,
  98                                         'Downloading iframe page')
  99
 100         sd_video_url = self._search_regex(r'<source src="([^"]+)"',
 101                                           iframe, 'sd video url')
 102         resolution_id = sd_video_url.split("/")[-1] + 'p'
 103         if len(resolution_id) != 4:
 104             resolution_id = None
 105         video_urls.append({
 106             'ext': 'mp4',
 107             'format_id': 'sd',
 108             'url': sd_video_url,
 109             'resolution': resolution_id,
 110         })
 111
 112         hd_video_url = self._search_regex(r'hdUrl":"([^"]+)"', iframe,
 113                                           'hd video url', default=None)
 114         if hd_video_url:
 115             hd_video_url = hd_video_url.replace("\\", "")
 116             resolution_id = hd_video_url.split("/")[-1] + 'p'
 117             if len(resolution_id) != 4:
 118                 resolution_id = None
 119             video_urls.append({
 120                 'ext': 'mp4',
 121                 'format_id': 'hd',
 122                 'url': hd_video_url,
 123                 'resolution': resolution_id,
 124             })
 125
 126         # The only place where you can get a title, it's not complete,
 127         # but searching in other places doesn't work for all videos
 128         video_title = self._html_search_regex(
 129             r'(?s)<title>(?P<title>.*?)(?: \| Tumblr)?</title>',
 130             webpage, 'title')
 131
 132         return {
 133             'id': video_id,
 134             'formats': video_urls,
 135             'ext': 'mp4',
 136             'title': video_title,
 137             'description': self._og_search_description(webpage, default=None),
 138             'thumbnail': self._og_search_thumbnail(webpage, default=None),
 139         }