[nfl] Support team micro-sites (fixes #3831)

This commit is contained in:
Naglis Jonaitis 2014-09-28 21:48:26 +03:00
parent dfee83234b
commit 5f4c318844
1 changed files with 100 additions and 59 deletions

View File

@ -6,6 +6,7 @@ import re
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import ( from ..utils import (
ExtractorError, ExtractorError,
compat_urllib_parse,
int_or_none, int_or_none,
remove_end, remove_end,
) )
@ -13,76 +14,116 @@ from ..utils import (
class NFLIE(InfoExtractor): class NFLIE(InfoExtractor):
IE_NAME = 'nfl.com' IE_NAME = 'nfl.com'
_VALID_URL = r'(?x)https?://(?:www\.)?nfl\.com/(?:videos/(?:.+)/|.*?\#video=)(?P<id>\d..[0-9]+)' _VALID_URL = r'''(?x)https?://
_PLAYER_CONFIG_URL = 'http://www.nfl.com/static/content/static/config/video/config.json' (?P<host>(?:www\.)?(?:nfl\.com|.*?\.clubs\.nfl\.com))/
_TEST = { (?:.+?/)*
'url': 'http://www.nfl.com/videos/nfl-game-highlights/0ap3000000398478/Week-3-Redskins-vs-Eagles-highlights', (?P<id>(?:\d[a-z]{2}\d{13}|\w{8}\-(?:\w{4}\-){3}\w{12}))'''
'md5': '394ef771ddcd1354f665b471d78ec4c6', _TESTS = [
'info_dict': { {
'id': '0ap3000000398478', 'url': 'http://www.nfl.com/videos/nfl-game-highlights/0ap3000000398478/Week-3-Redskins-vs-Eagles-highlights',
'ext': 'mp4', 'md5': '394ef771ddcd1354f665b471d78ec4c6',
'title': 'Week 3: Redskins vs. Eagles highlights', 'info_dict': {
'description': 'md5:56323bfb0ac4ee5ab24bd05fdf3bf478', 'id': '0ap3000000398478',
'upload_date': '20140921', 'ext': 'mp4',
'timestamp': 1411337580, 'title': 'Week 3: Redskins vs. Eagles highlights',
'thumbnail': 're:^https?://.*\.jpg$', 'description': 'md5:56323bfb0ac4ee5ab24bd05fdf3bf478',
'upload_date': '20140921',
'timestamp': 1411337580,
'thumbnail': 're:^https?://.*\.jpg$',
}
},
{
'url': 'http://prod.www.steelers.clubs.nfl.com/video-and-audio/videos/LIVE_Post_Game_vs_Browns/9d72f26a-9e2b-4718-84d3-09fb4046c266',
'md5': 'cf85bdb4bc49f6e9d3816d130c78279c',
'info_dict': {
'id': '9d72f26a-9e2b-4718-84d3-09fb4046c266',
'ext': 'mp4',
'title': 'LIVE: Post Game vs. Browns',
'description': 'md5:6a97f7e5ebeb4c0e69a418a89e0636e8',
'upload_date': '20131229',
'timestamp': 1388354455,
'thumbnail': 're:^https?://.*\.jpg$',
}
}
]
@staticmethod
def prepend_host(host, url):
if not url.startswith('http'):
if not url.startswith('/'):
url = '/%s' % url
url = 'http://{0:}{1:}'.format(host, url)
return url
@staticmethod
def format_from_stream(stream, protocol, host, path_prefix='',
preference=0, note=None):
url = '{protocol:}://{host:}/{prefix:}{path:}'.format(
protocol=protocol,
host=host,
prefix=path_prefix,
path=stream.get('path'),
)
return {
'url': url,
'vbr': int_or_none(stream.get('rate', 0), 1000),
'preference': preference,
'format_note': note,
} }
}
def _real_extract(self, url): def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url) mobj = re.match(self._VALID_URL, url)
video_id = mobj.group('id') video_id, host = mobj.group('id'), mobj.group('host')
config = self._download_json(self._PLAYER_CONFIG_URL, video_id, webpage = self._download_webpage(url, video_id)
config_url = NFLIE.prepend_host(host, self._search_regex(
r'(?:config|configURL)\s*:\s*"([^"]+)"', webpage, 'config URL'))
config = self._download_json(config_url, video_id,
note='Downloading player config') note='Downloading player config')
url_template = 'http://nfl.com{contentURLTemplate:s}'.format(**config) url_template = NFLIE.prepend_host(
video_data = self._download_json(url_template.format(id=video_id), video_id) host, '{contentURLTemplate:}'.format(**config))
video_data = self._download_json(
cdns = config.get('cdns') url_template.format(id=video_id), video_id)
if not cdns:
raise ExtractorError('Failed to get CDN data', expected=True)
formats = [] formats = []
streams = video_data.get('cdnData', {}).get('bitrateInfo', []) cdn_data = video_data.get('cdnData', {})
for name, cdn in cdns.items(): streams = cdn_data.get('bitrateInfo', [])
# LimeLight streams don't seem to work if cdn_data.get('format') == 'EXTERNAL_HTTP_STREAM':
if cdn.get('name') == 'LIMELIGHT': parts = compat_urllib_parse.urlparse(cdn_data.get('uri'))
continue protocol, host = parts.scheme, parts.netloc
protocol = cdn.get('protocol')
host = remove_end(cdn.get('host', ''), '/')
if not (protocol and host):
continue
path_prefix = cdn.get('pathprefix', '')
if path_prefix and not path_prefix.endswith('/'):
path_prefix = '%s/' % path_prefix
get_url = lambda p: '{protocol:s}://{host:s}/{prefix:s}{path:}'.format(
protocol=protocol,
host=host,
prefix=path_prefix,
path=p,
)
if protocol == 'rtmp':
preference = -1
elif 'prog' in name.lower():
preference = 1
else:
preference = 0
for stream in streams: for stream in streams:
path = stream.get('path') formats.append(
if not path: NFLIE.format_from_stream(stream, protocol, host))
else:
cdns = config.get('cdns')
if not cdns:
raise ExtractorError('Failed to get CDN data', expected=True)
for name, cdn in cdns.items():
# LimeLight streams don't seem to work
if cdn.get('name') == 'LIMELIGHT':
continue continue
formats.append({ protocol = cdn.get('protocol')
'url': get_url(path), host = remove_end(cdn.get('host', ''), '/')
'vbr': int_or_none(stream.get('rate', 0), 1000), if not (protocol and host):
'preference': preference, continue
'format_note': name,
}) prefix = cdn.get('pathprefix', '')
if prefix and not prefix.endswith('/'):
prefix = '%s/' % prefix
preference = 0
if protocol == 'rtmp':
preference = -2
elif 'prog' in name.lower():
preference = 1
for stream in streams:
formats.append(
NFLIE.format_from_stream(stream, protocol, host,
prefix, preference, name))
self._sort_formats(formats) self._sort_formats(formats)