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)
  20                     https?://
  21                         (?P<host>
  22                             (?:www\.)?
  23                             (?:
  24                                 (?:
  25                                     nfl|
  26                                     buffalobills|
  27                                     miamidolphins|
  28                                     patriots|
  29                                     newyorkjets|
  30                                     baltimoreravens|
  31                                     bengals|
  32                                     clevelandbrowns|
  33                                     steelers|
  34                                     houstontexans|
  35                                     colts|
  36                                     jaguars|
  37                                     titansonline|
  38                                     denverbroncos|
  39                                     kcchiefs|
  40                                     raiders|
  41                                     chargers|
  42                                     dallascowboys|
  43                                     giants|
  44                                     philadelphiaeagles|
  45                                     redskins|
  46                                     chicagobears|
  47                                     detroitlions|
  48                                     packers|
  49                                     vikings|
  50                                     atlantafalcons|
  51                                     panthers|
  52                                     neworleanssaints|
  53                                     buccaneers|
  54                                     azcardinals|
  55                                     stlouisrams|
  56                                     49ers|
  57                                     seahawks
  58                                 )\.com|
  59                                 .+?\.clubs\.nfl\.com
  60                             )
  61                         )/
  62                         (?:.+?/)*
  63                         (?P<id>[^/#?&]+)
  64                     '''
  65     _TESTS = [{
  66         'url': 'http://www.nfl.com/videos/nfl-game-highlights/0ap3000000398478/Week-3-Redskins-vs-Eagles-highlights',
  67         'md5': '394ef771ddcd1354f665b471d78ec4c6',
  68         'info_dict': {
  69             'id': '0ap3000000398478',
  70             'ext': 'mp4',
  71             'title': 'Week 3: Redskins vs. Eagles highlights',
  72             'description': 'md5:56323bfb0ac4ee5ab24bd05fdf3bf478',
  73             'upload_date': '20140921',
  74             'timestamp': 1411337580,
  75             'thumbnail': 're:^https?://.*\.jpg$',
  76         }
  77     }, {
  78         'url': 'http://prod.www.steelers.clubs.nfl.com/video-and-audio/videos/LIVE_Post_Game_vs_Browns/9d72f26a-9e2b-4718-84d3-09fb4046c266',
  79         'md5': 'cf85bdb4bc49f6e9d3816d130c78279c',
  80         'info_dict': {
  81             'id': '9d72f26a-9e2b-4718-84d3-09fb4046c266',
  82             'ext': 'mp4',
  83             'title': 'LIVE: Post Game vs. Browns',
  84             'description': 'md5:6a97f7e5ebeb4c0e69a418a89e0636e8',
  85             'upload_date': '20131229',
  86             'timestamp': 1388354455,
  87             'thumbnail': 're:^https?://.*\.jpg$',
  88         }
  89     }, {
  90         'url': 'http://www.nfl.com/news/story/0ap3000000467586/article/patriots-seahawks-involved-in-lategame-skirmish',
  91         'info_dict': {
  92             'id': '0ap3000000467607',
  93             'ext': 'mp4',
  94             'title': 'Frustrations flare on the field',
  95             'description': 'Emotions ran high at the end of the Super Bowl on both sides of the ball after a dramatic finish.',
  96             'timestamp': 1422850320,
  97             'upload_date': '20150202',
  98         },
  99     }, {
 100         'url': 'http://www.patriots.com/video/2015/09/18/10-days-gillette',
 101         'md5': '4c319e2f625ffd0b481b4382c6fc124c',
 102         'info_dict': {
 103             'id': 'n-238346',
 104             'ext': 'mp4',
 105             'title': '10 Days at Gillette',
 106             'description': 'md5:8cd9cd48fac16de596eadc0b24add951',
 107             'timestamp': 1442618809,
 108             'upload_date': '20150918',
 109         },
 110     }, {
 111         'url': 'http://www.nfl.com/videos/nfl-network-top-ten/09000d5d810a6bd4/Top-10-Gutsiest-Performances-Jack-Youngblood',
 112         'only_matching': True,
 113     }, {
 114         'url': 'http://www.buffalobills.com/video/videos/Rex_Ryan_Show_World_Wide_Rex/b1dcfab2-3190-4bb1-bfc0-d6e603d6601a',
 115         'only_matching': True,
 116     }]
 117
 118     @staticmethod
 119     def prepend_host(host, url):
 120         if not url.startswith('http'):
 121             if not url.startswith('/'):
 122                 url = '/%s' % url
 123             url = 'http://{0:}{1:}'.format(host, url)
 124         return url
 125
 126     @staticmethod
 127     def format_from_stream(stream, protocol, host, path_prefix='',
 128                            preference=0, note=None):
 129         url = '{protocol:}://{host:}/{prefix:}{path:}'.format(
 130             protocol=protocol,
 131             host=host,
 132             prefix=path_prefix,
 133             path=stream.get('path'),
 134         )
 135         return {
 136             'url': url,
 137             'vbr': int_or_none(stream.get('rate', 0), 1000),
 138             'preference': preference,
 139             'format_note': note,
 140         }
 141
 142     def _real_extract(self, url):
 143         mobj = re.match(self._VALID_URL, url)
 144         video_id, host = mobj.group('id'), mobj.group('host')
 145
 146         webpage = self._download_webpage(url, video_id)
 147
 148         config_url = NFLIE.prepend_host(host, self._search_regex(
 149             r'(?:(?:config|configURL)\s*:\s*|<nflcs:avplayer[^>]+data-config\s*=\s*)(["\'])(?P<config>.+?)\1',
 150             webpage, 'config URL', default='static/content/static/config/video/config.json',
 151             group='config'))
 152         # For articles, the id in the url is not the video id
 153         video_id = self._search_regex(
 154             r'(?:<nflcs:avplayer[^>]+data-content[Ii]d\s*=\s*|content[Ii]d\s*:\s*)(["\'])(?P<id>.+?)\1',
 155             webpage, 'video id', default=video_id, group='id')
 156         config = self._download_json(config_url, video_id, 'Downloading player config')
 157         url_template = NFLIE.prepend_host(
 158             host, '{contentURLTemplate:}'.format(**config))
 159         video_data = self._download_json(
 160             url_template.format(id=video_id), video_id)
 161
 162         formats = []
 163         cdn_data = video_data.get('cdnData', {})
 164         streams = cdn_data.get('bitrateInfo', [])
 165         if cdn_data.get('format') == 'EXTERNAL_HTTP_STREAM':
 166             parts = compat_urllib_parse_urlparse(cdn_data.get('uri'))
 167             protocol, host = parts.scheme, parts.netloc
 168             for stream in streams:
 169                 formats.append(
 170                     NFLIE.format_from_stream(stream, protocol, host))
 171         else:
 172             cdns = config.get('cdns')
 173             if not cdns:
 174                 raise ExtractorError('Failed to get CDN data', expected=True)
 175
 176             for name, cdn in cdns.items():
 177                 # LimeLight streams don't seem to work
 178                 if cdn.get('name') == 'LIMELIGHT':
 179                     continue
 180
 181                 protocol = cdn.get('protocol')
 182                 host = remove_end(cdn.get('host', ''), '/')
 183                 if not (protocol and host):
 184                     continue
 185
 186                 prefix = cdn.get('pathprefix', '')
 187                 if prefix and not prefix.endswith('/'):
 188                     prefix = '%s/' % prefix
 189
 190                 preference = 0
 191                 if protocol == 'rtmp':
 192                     preference = -2
 193                 elif 'prog' in name.lower():
 194                     preference = 1
 195
 196                 for stream in streams:
 197                     formats.append(
 198                         NFLIE.format_from_stream(stream, protocol, host,
 199                                                  prefix, preference, name))
 200
 201         self._sort_formats(formats)
 202
 203         thumbnail = None
 204         for q in ('xl', 'l', 'm', 's', 'xs'):
 205             thumbnail = video_data.get('imagePaths', {}).get(q)
 206             if thumbnail:
 207                 break
 208
 209         return {
 210             'id': video_id,
 211             'title': video_data.get('headline'),
 212             'formats': formats,
 213             'description': video_data.get('caption'),
 214             'duration': video_data.get('duration'),
 215             'thumbnail': thumbnail,
 216             'timestamp': int_or_none(video_data.get('posted'), 1000),
 217         }