]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/nfl.py
606e2294efb716cfe755d1b9564357dbda7f9039
[youtubedl] / youtube_dl / extractor / nfl.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 import re
5
6 from .common import InfoExtractor
7 from ..compat import (
8 compat_urllib_parse_urlparse,
9 )
10 from ..utils import (
11 ExtractorError,
12 int_or_none,
13 remove_end,
14 )
15
16
17 class NFLIE(InfoExtractor):
18 IE_NAME = 'nfl.com'
19 _VALID_URL = r'''(?x)https?://
20 (?P<host>(?:www\.)?(?:nfl\.com|.*?\.clubs\.nfl\.com))/
21 (?:.+?/)*
22 (?P<id>(?:\d[a-z]{2}\d{13}|\w{8}\-(?:\w{4}\-){3}\w{12}))'''
23 _TESTS = [
24 {
25 'url': 'http://www.nfl.com/videos/nfl-game-highlights/0ap3000000398478/Week-3-Redskins-vs-Eagles-highlights',
26 'md5': '394ef771ddcd1354f665b471d78ec4c6',
27 'info_dict': {
28 'id': '0ap3000000398478',
29 'ext': 'mp4',
30 'title': 'Week 3: Redskins vs. Eagles highlights',
31 'description': 'md5:56323bfb0ac4ee5ab24bd05fdf3bf478',
32 'upload_date': '20140921',
33 'timestamp': 1411337580,
34 'thumbnail': 're:^https?://.*\.jpg$',
35 }
36 },
37 {
38 'url': 'http://prod.www.steelers.clubs.nfl.com/video-and-audio/videos/LIVE_Post_Game_vs_Browns/9d72f26a-9e2b-4718-84d3-09fb4046c266',
39 'md5': 'cf85bdb4bc49f6e9d3816d130c78279c',
40 'info_dict': {
41 'id': '9d72f26a-9e2b-4718-84d3-09fb4046c266',
42 'ext': 'mp4',
43 'title': 'LIVE: Post Game vs. Browns',
44 'description': 'md5:6a97f7e5ebeb4c0e69a418a89e0636e8',
45 'upload_date': '20131229',
46 'timestamp': 1388354455,
47 'thumbnail': 're:^https?://.*\.jpg$',
48 }
49 }
50 ]
51
52 @staticmethod
53 def prepend_host(host, url):
54 if not url.startswith('http'):
55 if not url.startswith('/'):
56 url = '/%s' % url
57 url = 'http://{0:}{1:}'.format(host, url)
58 return url
59
60 @staticmethod
61 def format_from_stream(stream, protocol, host, path_prefix='',
62 preference=0, note=None):
63 url = '{protocol:}://{host:}/{prefix:}{path:}'.format(
64 protocol=protocol,
65 host=host,
66 prefix=path_prefix,
67 path=stream.get('path'),
68 )
69 return {
70 'url': url,
71 'vbr': int_or_none(stream.get('rate', 0), 1000),
72 'preference': preference,
73 'format_note': note,
74 }
75
76 def _real_extract(self, url):
77 mobj = re.match(self._VALID_URL, url)
78 video_id, host = mobj.group('id'), mobj.group('host')
79
80 webpage = self._download_webpage(url, video_id)
81
82 config_url = NFLIE.prepend_host(host, self._search_regex(
83 r'(?:config|configURL)\s*:\s*"([^"]+)"', webpage, 'config URL'))
84 config = self._download_json(config_url, video_id,
85 note='Downloading player config')
86 url_template = NFLIE.prepend_host(
87 host, '{contentURLTemplate:}'.format(**config))
88 video_data = self._download_json(
89 url_template.format(id=video_id), video_id)
90
91 formats = []
92 cdn_data = video_data.get('cdnData', {})
93 streams = cdn_data.get('bitrateInfo', [])
94 if cdn_data.get('format') == 'EXTERNAL_HTTP_STREAM':
95 parts = compat_urllib_parse_urlparse(cdn_data.get('uri'))
96 protocol, host = parts.scheme, parts.netloc
97 for stream in streams:
98 formats.append(
99 NFLIE.format_from_stream(stream, protocol, host))
100 else:
101 cdns = config.get('cdns')
102 if not cdns:
103 raise ExtractorError('Failed to get CDN data', expected=True)
104
105 for name, cdn in cdns.items():
106 # LimeLight streams don't seem to work
107 if cdn.get('name') == 'LIMELIGHT':
108 continue
109
110 protocol = cdn.get('protocol')
111 host = remove_end(cdn.get('host', ''), '/')
112 if not (protocol and host):
113 continue
114
115 prefix = cdn.get('pathprefix', '')
116 if prefix and not prefix.endswith('/'):
117 prefix = '%s/' % prefix
118
119 preference = 0
120 if protocol == 'rtmp':
121 preference = -2
122 elif 'prog' in name.lower():
123 preference = 1
124
125 for stream in streams:
126 formats.append(
127 NFLIE.format_from_stream(stream, protocol, host,
128 prefix, preference, name))
129
130 self._sort_formats(formats)
131
132 thumbnail = None
133 for q in ('xl', 'l', 'm', 's', 'xs'):
134 thumbnail = video_data.get('imagePaths', {}).get(q)
135 if thumbnail:
136 break
137
138 return {
139 'id': video_id,
140 'title': video_data.get('headline'),
141 'formats': formats,
142 'description': video_data.get('caption'),
143 'duration': video_data.get('duration'),
144 'thumbnail': thumbnail,
145 'timestamp': int_or_none(video_data.get('posted'), 1000),
146 }