]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/mit.py
debian/control: Update list of extractors.
[youtubedl] / youtube_dl / extractor / mit.py
1 import re
2 import json
3
4 from .common import InfoExtractor
5 from ..utils import (
6 clean_html,
7 get_element_by_id,
8 )
9
10
11 class TechTVMITIE(InfoExtractor):
12 IE_NAME = u'techtv.mit.edu'
13 _VALID_URL = r'https?://techtv\.mit\.edu/(videos|embeds)/(?P<id>\d+)'
14
15 _TEST = {
16 u'url': u'http://techtv.mit.edu/videos/25418-mit-dna-learning-center-set',
17 u'file': u'25418.mp4',
18 u'md5': u'1f8cb3e170d41fd74add04d3c9330e5f',
19 u'info_dict': {
20 u'title': u'MIT DNA Learning Center Set',
21 u'description': u'md5:82313335e8a8a3f243351ba55bc1b474',
22 },
23 }
24
25 def _real_extract(self, url):
26 mobj = re.match(self._VALID_URL, url)
27 video_id = mobj.group('id')
28 raw_page = self._download_webpage(
29 'http://techtv.mit.edu/videos/%s' % video_id, video_id)
30 clean_page = re.compile(u'<!--.*?-->', re.S).sub(u'', raw_page)
31
32 base_url = self._search_regex(r'ipadUrl: \'(.+?cloudfront.net/)',
33 raw_page, u'base url')
34 formats_json = self._search_regex(r'bitrates: (\[.+?\])', raw_page,
35 u'video formats')
36 formats = json.loads(formats_json)
37 formats = sorted(formats, key=lambda f: f['bitrate'])
38
39 title = get_element_by_id('edit-title', clean_page)
40 description = clean_html(get_element_by_id('edit-description', clean_page))
41 thumbnail = self._search_regex(r'playlist:.*?url: \'(.+?)\'',
42 raw_page, u'thumbnail', flags=re.DOTALL)
43
44 return {'id': video_id,
45 'title': title,
46 'url': base_url + formats[-1]['url'].replace('mp4:', ''),
47 'ext': 'mp4',
48 'description': description,
49 'thumbnail': thumbnail,
50 }
51
52
53 class MITIE(TechTVMITIE):
54 IE_NAME = u'video.mit.edu'
55 _VALID_URL = r'https?://video\.mit\.edu/watch/(?P<title>[^/]+)'
56
57 _TEST = {
58 u'url': u'http://video.mit.edu/watch/the-government-is-profiling-you-13222/',
59 u'file': u'21783.mp4',
60 u'md5': u'7db01d5ccc1895fc5010e9c9e13648da',
61 u'info_dict': {
62 u'title': u'The Government is Profiling You',
63 u'description': u'md5:ad5795fe1e1623b73620dbfd47df9afd',
64 },
65 }
66
67 def _real_extract(self, url):
68 mobj = re.match(self._VALID_URL, url)
69 page_title = mobj.group('title')
70 webpage = self._download_webpage(url, page_title)
71 self.to_screen('%s: Extracting %s url' % (page_title, TechTVMITIE.IE_NAME))
72 embed_url = self._search_regex(r'<iframe .*?src="(.+?)"', webpage,
73 u'embed url')
74 return self.url_result(embed_url, ie='TechTVMIT')