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)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             'url': 'http://www.nfl.com/news/story/0ap3000000467586/article/patriots-seahawks-involved-in-lategame-skirmish',
  52             'info_dict': {
  53                 'id': '0ap3000000467607',
  54                 'ext': 'mp4',
  55                 'title': 'Frustrations flare on the field',
  56                 'description': 'Emotions ran high at the end of the Super Bowl on both sides of the ball after a dramatic finish.',
  57                 'timestamp': 1422850320,
  58                 'upload_date': '20150202',
  59             },
  60         },
  61     ]
  62
  63     @staticmethod
  64     def prepend_host(host, url):
  65         if not url.startswith('http'):
  66             if not url.startswith('/'):
  67                 url = '/%s' % url
  68             url = 'http://{0:}{1:}'.format(host, url)
  69         return url
  70
  71     @staticmethod
  72     def format_from_stream(stream, protocol, host, path_prefix='',
  73                            preference=0, note=None):
  74         url = '{protocol:}://{host:}/{prefix:}{path:}'.format(
  75             protocol=protocol,
  76             host=host,
  77             prefix=path_prefix,
  78             path=stream.get('path'),
  79         )
  80         return {
  81             'url': url,
  82             'vbr': int_or_none(stream.get('rate', 0), 1000),
  83             'preference': preference,
  84             'format_note': note,
  85         }
  86
  87     def _real_extract(self, url):
  88         mobj = re.match(self._VALID_URL, url)
  89         video_id, host = mobj.group('id'), mobj.group('host')
  90
  91         webpage = self._download_webpage(url, video_id)
  92
  93         config_url = NFLIE.prepend_host(host, self._search_regex(
  94             r'(?:config|configURL)\s*:\s*"([^"]+)"', webpage, 'config URL',
  95             default='static/content/static/config/video/config.json'))
  96         # For articles, the id in the url is not the video id
  97         video_id = self._search_regex(
  98             r'contentId\s*:\s*"([^"]+)"', webpage, 'video id', default=video_id)
  99         config = self._download_json(config_url, video_id,
 100                                      note='Downloading player config')
 101         url_template = NFLIE.prepend_host(
 102             host, '{contentURLTemplate:}'.format(**config))
 103         video_data = self._download_json(
 104             url_template.format(id=video_id), video_id)
 105
 106         formats = []
 107         cdn_data = video_data.get('cdnData', {})
 108         streams = cdn_data.get('bitrateInfo', [])
 109         if cdn_data.get('format') == 'EXTERNAL_HTTP_STREAM':
 110             parts = compat_urllib_parse_urlparse(cdn_data.get('uri'))
 111             protocol, host = parts.scheme, parts.netloc
 112             for stream in streams:
 113                 formats.append(
 114                     NFLIE.format_from_stream(stream, protocol, host))
 115         else:
 116             cdns = config.get('cdns')
 117             if not cdns:
 118                 raise ExtractorError('Failed to get CDN data', expected=True)
 119
 120             for name, cdn in cdns.items():
 121                 # LimeLight streams don't seem to work
 122                 if cdn.get('name') == 'LIMELIGHT':
 123                     continue
 124
 125                 protocol = cdn.get('protocol')
 126                 host = remove_end(cdn.get('host', ''), '/')
 127                 if not (protocol and host):
 128                     continue
 129
 130                 prefix = cdn.get('pathprefix', '')
 131                 if prefix and not prefix.endswith('/'):
 132                     prefix = '%s/' % prefix
 133
 134                 preference = 0
 135                 if protocol == 'rtmp':
 136                     preference = -2
 137                 elif 'prog' in name.lower():
 138                     preference = 1
 139
 140                 for stream in streams:
 141                     formats.append(
 142                         NFLIE.format_from_stream(stream, protocol, host,
 143                                                  prefix, preference, name))
 144
 145         self._sort_formats(formats)
 146
 147         thumbnail = None
 148         for q in ('xl', 'l', 'm', 's', 'xs'):
 149             thumbnail = video_data.get('imagePaths', {}).get(q)
 150             if thumbnail:
 151                 break
 152
 153         return {
 154             'id': video_id,
 155             'title': video_data.get('headline'),
 156             'formats': formats,
 157             'description': video_data.get('caption'),
 158             'duration': video_data.get('duration'),
 159             'thumbnail': thumbnail,
 160             'timestamp': int_or_none(video_data.get('posted'), 1000),
 161         }