2 from __future__ import unicode_literals
6 from .common import InfoExtractor
14 class NFLIE(InfoExtractor):
16 _VALID_URL = r'(?x)https?://(?:www\.)?nfl\.com/(?:videos/(?:.+)/|.*?\#video=)(?P<id>\d..[0-9]+)'
17 _PLAYER_CONFIG_URL = 'http://www.nfl.com/static/content/static/config/video/config.json'
19 'url': 'http://www.nfl.com/videos/nfl-game-highlights/0ap3000000398478/Week-3-Redskins-vs-Eagles-highlights',
20 # 'md5': '5eb8c40a727dda106d510e5d6ffa79e5', # md5 checksum fluctuates
22 'id': '0ap3000000398478',
24 'title': 'Week 3: Washington Redskins vs. Philadelphia Eagles highlights',
25 'description': 'md5:56323bfb0ac4ee5ab24bd05fdf3bf478',
26 'upload_date': '20140921',
27 'timestamp': 1411337580,
28 'thumbnail': 're:^https?://.*\.jpg$',
32 def _real_extract(self, url):
33 mobj = re.match(self._VALID_URL, url)
34 video_id = mobj.group('id')
36 config = self._download_json(self._PLAYER_CONFIG_URL, video_id,
37 note='Downloading player config')
38 url_template = 'http://nfl.com{contentURLTemplate:s}'.format(**config)
39 video_data = self._download_json(url_template.format(id=video_id), video_id)
41 cdns = config.get('cdns')
43 raise ExtractorError('Failed to get CDN data', expected=True)
46 streams = video_data.get('cdnData', {}).get('bitrateInfo', [])
47 for name, cdn in cdns.items():
48 # LimeLight streams don't seem to work
49 if cdn.get('name') == 'LIMELIGHT':
52 protocol = cdn.get('protocol')
53 host = remove_end(cdn.get('host', ''), '/')
54 if not (protocol and host):
57 path_prefix = cdn.get('pathprefix', '')
58 if path_prefix and not path_prefix.endswith('/'):
59 path_prefix = '%s/' % path_prefix
61 get_url = lambda p: '{protocol:s}://{host:s}/{prefix:s}{path:}'.format(
68 if protocol == 'rtmp':
70 elif 'prog' in name.lower():
75 for stream in streams:
76 path = stream.get('path')
82 'vbr': int_or_none(stream.get('rate', 0), 1000),
83 'preference': preference,
87 self._sort_formats(formats)
90 for q in ('xl', 'l', 'm', 's', 'xs'):
91 thumbnail = video_data.get('imagePaths', {}).get(q)
97 'title': video_data.get('storyHeadline'),
99 'description': video_data.get('caption'),
100 'duration': video_data.get('duration'),
101 'thumbnail': thumbnail,
102 'timestamp': int_or_none(video_data.get('posted'), 1000),