]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/ooyala.py
Merge tag 'upstream/2013.08.17'
[youtubedl] / youtube_dl / extractor / ooyala.py
1 import re
2 import json
3
4 from .common import InfoExtractor
5 from ..utils import unescapeHTML
6
7 class OoyalaIE(InfoExtractor):
8 _VALID_URL = r'https?://.+?\.ooyala\.com/.*?embedCode=(?P<id>.+?)(&|$)'
9
10 _TEST = {
11 # From http://it.slashdot.org/story/13/04/25/178216/recovering-data-from-broken-hard-drives-and-ssds-video
12 u'url': u'http://player.ooyala.com/player.js?embedCode=pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8',
13 u'file': u'pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8.mp4',
14 u'md5': u'3f5cceb3a7bf461d6c29dc466cf8033c',
15 u'info_dict': {
16 u'title': u'Explaining Data Recovery from Hard Drives and SSDs',
17 u'description': u'How badly damaged does a drive have to be to defeat Russell and his crew? Apparently, smashed to bits.',
18 },
19 }
20
21 def _extract_result(self, info, more_info):
22 return {'id': info['embedCode'],
23 'ext': 'mp4',
24 'title': unescapeHTML(info['title']),
25 'url': info['url'],
26 'description': unescapeHTML(more_info['description']),
27 'thumbnail': more_info['promo'],
28 }
29
30 def _real_extract(self, url):
31 mobj = re.match(self._VALID_URL, url)
32 embedCode = mobj.group('id')
33 player_url = 'http://player.ooyala.com/player.js?embedCode=%s' % embedCode
34 player = self._download_webpage(player_url, embedCode)
35 mobile_url = self._search_regex(r'mobile_player_url="(.+?)&device="',
36 player, u'mobile player url')
37 mobile_player = self._download_webpage(mobile_url, embedCode)
38 videos_info = self._search_regex(r'eval\("\((\[{.*?stream_redirect.*?}\])\)"\);', mobile_player, u'info').replace('\\"','"')
39 videos_more_info = self._search_regex(r'eval\("\(({.*?\\"promo\\".*?})\)"', mobile_player, u'more info').replace('\\"','"')
40 videos_info = json.loads(videos_info)
41 videos_more_info =json.loads(videos_more_info)
42
43 if videos_more_info.get('lineup'):
44 videos = [self._extract_result(info, more_info) for (info, more_info) in zip(videos_info, videos_more_info['lineup'])]
45 return {'_type': 'playlist',
46 'id': embedCode,
47 'title': unescapeHTML(videos_more_info['title']),
48 'entries': videos,
49 }
50 else:
51 return self._extract_result(videos_info[0], videos_more_info)
52