]> Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/tumblr.py
Prepare new release.
[youtubedl] / 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 ..utils import int_or_none
8
9
10 class TumblrIE(InfoExtractor):
11 _VALID_URL = r'https?://(?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': r'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': r're:http://.*\.jpg',
31 }
32 }, {
33 'url': 'http://hdvideotest.tumblr.com/post/130323439814/test-description-for-my-hd-video',
34 'md5': '7ae503065ad150122dc3089f8cf1546c',
35 'info_dict': {
36 'id': '130323439814',
37 'ext': 'mp4',
38 'title': 'HD Video Testing \u2014 Test description for my HD video',
39 'description': 'md5:97cc3ab5fcd27ee4af6356701541319c',
40 'thumbnail': r're:http://.*\.jpg',
41 },
42 'params': {
43 'format': 'hd',
44 },
45 }, {
46 'url': 'http://naked-yogi.tumblr.com/post/118312946248/naked-smoking-stretching',
47 'md5': 'de07e5211d60d4f3a2c3df757ea9f6ab',
48 'info_dict': {
49 'id': 'Wmur',
50 'ext': 'mp4',
51 'title': 'naked smoking & stretching',
52 'upload_date': '20150506',
53 'timestamp': 1430931613,
54 'age_limit': 18,
55 'uploader_id': '1638622',
56 'uploader': 'naked-yogi',
57 },
58 'add_ie': ['Vidme'],
59 }, {
60 'url': 'http://camdamage.tumblr.com/post/98846056295/',
61 'md5': 'a9e0c8371ea1ca306d6554e3fecf50b6',
62 'info_dict': {
63 'id': '105463834',
64 'ext': 'mp4',
65 'title': 'Cam Damage-HD 720p',
66 'uploader': 'John Moyer',
67 'uploader_id': 'user32021558',
68 },
69 'add_ie': ['Vimeo'],
70 }, {
71 'url': 'http://sutiblr.tumblr.com/post/139638707273',
72 'md5': '2dd184b3669e049ba40563a7d423f95c',
73 'info_dict': {
74 'id': 'ir7qBEIKqvq',
75 'ext': 'mp4',
76 'title': 'Vine by sutiblr',
77 'alt_title': 'Vine by sutiblr',
78 'uploader': 'sutiblr',
79 'uploader_id': '1198993975374495744',
80 'upload_date': '20160220',
81 'like_count': int,
82 'comment_count': int,
83 'repost_count': int,
84 },
85 'add_ie': ['Vine'],
86 }, {
87 'url': 'http://vitasidorkina.tumblr.com/post/134652425014/joskriver-victoriassecret-invisibility-or',
88 'md5': '01c12ceb82cbf6b2fe0703aa56b3ad72',
89 'info_dict': {
90 'id': '-7LnUPGlSo',
91 'ext': 'mp4',
92 'title': 'Video by victoriassecret',
93 'description': 'Invisibility or flight…which superpower would YOU choose? #VSFashionShow #ThisOrThat',
94 'uploader_id': 'victoriassecret',
95 'thumbnail': r're:^https?://.*\.jpg'
96 },
97 'add_ie': ['Instagram'],
98 }]
99
100 def _real_extract(self, url):
101 m_url = re.match(self._VALID_URL, url)
102 video_id = m_url.group('id')
103 blog = m_url.group('blog_name')
104
105 url = 'http://%s.tumblr.com/post/%s/' % (blog, video_id)
106 webpage, urlh = self._download_webpage_handle(url, video_id)
107
108 iframe_url = self._search_regex(
109 r'src=\'(https?://www\.tumblr\.com/video/[^\']+)\'',
110 webpage, 'iframe url', default=None)
111 if iframe_url is None:
112 return self.url_result(urlh.geturl(), 'Generic')
113
114 iframe = self._download_webpage(iframe_url, video_id, 'Downloading iframe page')
115
116 duration = None
117 sources = []
118
119 sd_url = self._search_regex(
120 r'<source[^>]+src=(["\'])(?P<url>.+?)\1', iframe,
121 'sd video url', default=None, group='url')
122 if sd_url:
123 sources.append((sd_url, 'sd'))
124
125 options = self._parse_json(
126 self._search_regex(
127 r'data-crt-options=(["\'])(?P<options>.+?)\1', iframe,
128 'hd video url', default='', group='options'),
129 video_id, fatal=False)
130 if options:
131 duration = int_or_none(options.get('duration'))
132 hd_url = options.get('hdUrl')
133 if hd_url:
134 sources.append((hd_url, 'hd'))
135
136 formats = [{
137 'url': video_url,
138 'ext': 'mp4',
139 'format_id': format_id,
140 'height': int_or_none(self._search_regex(
141 r'/(\d{3,4})$', video_url, 'height', default=None)),
142 'quality': quality,
143 } for quality, (video_url, format_id) in enumerate(sources)]
144
145 self._sort_formats(formats)
146
147 # The only place where you can get a title, it's not complete,
148 # but searching in other places doesn't work for all videos
149 video_title = self._html_search_regex(
150 r'(?s)<title>(?P<title>.*?)(?: \| Tumblr)?</title>',
151 webpage, 'title')
152
153 return {
154 'id': video_id,
155 'title': video_title,
156 'description': self._og_search_description(webpage, default=None),
157 'thumbnail': self._og_search_thumbnail(webpage, default=None),
158 'duration': duration,
159 'formats': formats,
160 }