1 from __future__ import unicode_literals
3 from .common import InfoExtractor
8 from ..compat import compat_str
11 class DiscoveryIE(InfoExtractor):
12 _VALID_URL = r'''(?x)https?://(?:www\.)?(?:
14 investigationdiscovery|
22 )\.com/(?:[^/]+/)*(?P<id>[^./?#]+)'''
24 'url': 'http://www.discovery.com/tv-shows/mythbusters/videos/mission-impossible-outtakes.htm',
28 'title': 'Mission Impossible Outtakes',
29 'description': ('Watch Jamie Hyneman and Adam Savage practice being'
30 ' each other -- to the point of confusing Jamie\'s dog -- and '
31 'don\'t miss Adam moon-walking as Jamie ... behind Jamie\'s'
34 'timestamp': 1302032462,
35 'upload_date': '20110405',
38 'skip_download': True, # requires ffmpeg
41 'url': 'http://www.discovery.com/tv-shows/mythbusters/videos/mythbusters-the-simpsons',
43 'id': 'mythbusters-the-simpsons',
44 'title': 'MythBusters: The Simpsons',
46 'playlist_mincount': 10,
48 'url': 'http://www.animalplanet.com/longfin-eels-maneaters/',
52 'title': 'Longfin Eels: Maneaters?',
53 'description': 'Jeremy Wade tests whether or not New Zealand\'s longfin eels are man-eaters by covering himself in fish guts and getting in the water with them.',
54 'upload_date': '20140725',
55 'timestamp': 1406246400,
60 def _real_extract(self, url):
61 display_id = self._match_id(url)
62 info = self._download_json(url + '?flat=1', display_id)
64 video_title = info.get('playlist_title') or info.get('video_title')
68 for idx, video_info in enumerate(info['playlist']):
69 formats = self._extract_m3u8_formats(
70 video_info['src'], display_id, 'mp4', 'm3u8_native', m3u8_id='hls',
71 note='Download m3u8 information for video %d' % (idx + 1))
72 self._sort_formats(formats)
74 'id': compat_str(video_info['id']),
76 'title': video_info['title'],
77 'description': video_info.get('description'),
78 'duration': parse_duration(video_info.get('video_length')),
79 'webpage_url': video_info.get('href') or video_info.get('url'),
80 'thumbnail': video_info.get('thumbnailURL'),
81 'alt_title': video_info.get('secondary_title'),
82 'timestamp': parse_iso8601(video_info.get('publishedDate')),
85 return self.playlist_result(entries, display_id, video_title)