Raphaël G. Git Repositories - youtubedl/blob - 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': r're:^https?://.*\.jpg$',
  76         }
  77     }, {
  78         'url': 'http://prod.www.steelers.clubs.nfl.com/video-and-audio/videos/LIVE_Post_Game_vs_Browns/9d72f26a-9e2b-4718-84d3-09fb4046c266',
  79         'md5': 'cf85bdb4bc49f6e9d3816d130c78279c',
  80         'info_dict': {
  81             'id': '9d72f26a-9e2b-4718-84d3-09fb4046c266',
  82             'ext': 'mp4',
  83             'title': 'LIVE: Post Game vs. Browns',
  84             'description': 'md5:6a97f7e5ebeb4c0e69a418a89e0636e8',
  85             'upload_date': '20131229',
  86             'timestamp': 1388354455,
  87             'thumbnail': r're:^https?://.*\.jpg$',
  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         # lowercase data-contentid
 112         'url': 'http://www.steelers.com/news/article-1/Tomlin-on-Ben-getting-Vick-ready/56399c96-4160-48cf-a7ad-1d17d4a3aef7',
 113         'info_dict': {
 114             'id': '12693586-6ea9-4743-9c1c-02c59e4a5ef2',
 115             'ext': 'mp4',
 116             'title': 'Tomlin looks ahead to Ravens on a short week',
 117             'description': 'md5:32f3f7b139f43913181d5cbb24ecad75',
 118             'timestamp': 1443459651,
 119             'upload_date': '20150928',
 120         },
 121         'params': {
 122             'skip_download': True,
 123         },
 124     }, {
 125         'url': 'http://www.nfl.com/videos/nfl-network-top-ten/09000d5d810a6bd4/Top-10-Gutsiest-Performances-Jack-Youngblood',
 126         'only_matching': True,
 127     }, {
 128         'url': 'http://www.buffalobills.com/video/videos/Rex_Ryan_Show_World_Wide_Rex/b1dcfab2-3190-4bb1-bfc0-d6e603d6601a',
 129         'only_matching': True,
 130     }]
 131
 132     @staticmethod
 133     def prepend_host(host, url):
 134         if not url.startswith('http'):
 135             if not url.startswith('/'):
 136                 url = '/%s' % url
 137             url = 'http://{0:}{1:}'.format(host, url)
 138         return url
 139
 140     @staticmethod
 141     def format_from_stream(stream, protocol, host, path_prefix='',
 142                            preference=0, note=None):
 143         url = '{protocol:}://{host:}/{prefix:}{path:}'.format(
 144             protocol=protocol,
 145             host=host,
 146             prefix=path_prefix,
 147             path=stream.get('path'),
 148         )
 149         return {
 150             'url': url,
 151             'vbr': int_or_none(stream.get('rate', 0), 1000),
 152             'preference': preference,
 153             'format_note': note,
 154         }
 155
 156     def _real_extract(self, url):
 157         mobj = re.match(self._VALID_URL, url)
 158         video_id, host = mobj.group('id'), mobj.group('host')
 159
 160         webpage = self._download_webpage(url, video_id)
 161
 162         config_url = NFLIE.prepend_host(host, self._search_regex(
 163             r'(?:(?:config|configURL)\s*:\s*|<nflcs:avplayer[^>]+data-config\s*=\s*)(["\'])(?P<config>.+?)\1',
 164             webpage, 'config URL', default='static/content/static/config/video/config.json',
 165             group='config'))
 166         # For articles, the id in the url is not the video id
 167         video_id = self._search_regex(
 168             r'(?:<nflcs:avplayer[^>]+data-content[Ii]d\s*=\s*|content[Ii]d\s*:\s*)(["\'])(?P<id>(?:(?!\1).)+)\1',
 169             webpage, 'video id', default=video_id, group='id')
 170         config = self._download_json(config_url, video_id, 'Downloading player config')
 171         url_template = NFLIE.prepend_host(
 172             host, '{contentURLTemplate:}'.format(**config))
 173         video_data = self._download_json(
 174             url_template.format(id=video_id), video_id)
 175
 176         formats = []
 177         cdn_data = video_data.get('cdnData', {})
 178         streams = cdn_data.get('bitrateInfo', [])
 179         if cdn_data.get('format') == 'EXTERNAL_HTTP_STREAM':
 180             parts = compat_urllib_parse_urlparse(cdn_data.get('uri'))
 181             protocol, host = parts.scheme, parts.netloc
 182             for stream in streams:
 183                 formats.append(
 184                     NFLIE.format_from_stream(stream, protocol, host))
 185         else:
 186             cdns = config.get('cdns')
 187             if not cdns:
 188                 raise ExtractorError('Failed to get CDN data', expected=True)
 189
 190             for name, cdn in cdns.items():
 191                 # LimeLight streams don't seem to work
 192                 if cdn.get('name') == 'LIMELIGHT':
 193                     continue
 194
 195                 protocol = cdn.get('protocol')
 196                 host = remove_end(cdn.get('host', ''), '/')
 197                 if not (protocol and host):
 198                     continue
 199
 200                 prefix = cdn.get('pathprefix', '')
 201                 if prefix and not prefix.endswith('/'):
 202                     prefix = '%s/' % prefix
 203
 204                 preference = 0
 205                 if protocol == 'rtmp':
 206                     preference = -2
 207                 elif 'prog' in name.lower():
 208                     preference = 1
 209
 210                 for stream in streams:
 211                     formats.append(
 212                         NFLIE.format_from_stream(stream, protocol, host,
 213                                                  prefix, preference, name))
 214
 215         self._sort_formats(formats)
 216
 217         thumbnail = None
 218         for q in ('xl', 'l', 'm', 's', 'xs'):
 219             thumbnail = video_data.get('imagePaths', {}).get(q)
 220             if thumbnail:
 221                 break
 222
 223         return {
 224             'id': video_id,
 225             'title': video_data.get('headline'),
 226             'formats': formats,
 227             'description': video_data.get('caption'),
 228             'duration': video_data.get('duration'),
 229             'thumbnail': thumbnail,
 230             'timestamp': int_or_none(video_data.get('posted'), 1000),
 231         }