]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/tumblr.py
debian/NEWS: Update with notes about split audio/video downloads.
[youtubedl] / youtube_dl / extractor / tumblr.py
1 import re
2
3 from .common import InfoExtractor
4 from ..utils import (
5 ExtractorError,
6 )
7
8
9 class TumblrIE(InfoExtractor):
10 _VALID_URL = r'http://(?P<blog_name>.*?)\.tumblr\.com/((post)|(video))/(?P<id>\d*)/(.*?)'
11 _TEST = {
12 u'url': u'http://tatianamaslanydaily.tumblr.com/post/54196191430/orphan-black-dvd-extra-behind-the-scenes',
13 u'file': u'54196191430.mp4',
14 u'md5': u'479bb068e5b16462f5176a6828829767',
15 u'info_dict': {
16 u"title": u"tatiana maslany news"
17 }
18 }
19
20 def _real_extract(self, url):
21 m_url = re.match(self._VALID_URL, url)
22 video_id = m_url.group('id')
23 blog = m_url.group('blog_name')
24
25 url = 'http://%s.tumblr.com/post/%s/' % (blog, video_id)
26 webpage = self._download_webpage(url, video_id)
27
28 re_video = r'src=\\x22(?P<video_url>http://%s\.tumblr\.com/video_file/%s/(.*?))\\x22 type=\\x22video/(?P<ext>.*?)\\x22' % (blog, video_id)
29 video = re.search(re_video, webpage)
30 if video is None:
31 raise ExtractorError(u'Unable to extract video')
32 video_url = video.group('video_url')
33 ext = video.group('ext')
34
35 video_thumbnail = self._search_regex(r'posters(.*?)\[\\x22(?P<thumb>.*?)\\x22',
36 webpage, u'thumbnail', fatal=False) # We pick the first poster
37 if video_thumbnail: video_thumbnail = video_thumbnail.replace('\\', '')
38
39 # The only place where you can get a title, it's not complete,
40 # but searching in other places doesn't work for all videos
41 video_title = self._html_search_regex(r'<title>(?P<title>.*?)(?: \| Tumblr)?</title>',
42 webpage, u'title', flags=re.DOTALL)
43
44 return [{'id': video_id,
45 'url': video_url,
46 'title': video_title,
47 'thumbnail': video_thumbnail,
48 'ext': ext
49 }]