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 ..utils import (
   8     ExtractorError,
   9     int_or_none,
  10     remove_end,
  11 )
  12
  13
  14 class NFLIE(InfoExtractor):
  15     IE_NAME = 'nfl.com'
  16     _VALID_URL = r'(?x)https?://(?:www\.)?nfl\.com/(?:videos/(?:.+)/|.*?\#video=)(?P<id>\d..[0-9]+)'
  17     _PLAYER_CONFIG_URL = 'http://www.nfl.com/static/content/static/config/video/config.json'
  18     _TEST = {
  19         'url': 'http://www.nfl.com/videos/nfl-game-highlights/0ap3000000398478/Week-3-Redskins-vs-Eagles-highlights',
  20         # 'md5': '5eb8c40a727dda106d510e5d6ffa79e5',  # md5 checksum fluctuates
  21         'info_dict': {
  22             'id': '0ap3000000398478',
  23             'ext': 'mp4',
  24             'title': 'Week 3: Washington Redskins vs. Philadelphia Eagles highlights',
  25             'description': 'md5:56323bfb0ac4ee5ab24bd05fdf3bf478',
  26             'upload_date': '20140921',
  27             'timestamp': 1411337580,
  28             'thumbnail': 're:^https?://.*\.jpg$',
  29         }
  30     }
  31
  32     def _real_extract(self, url):
  33         mobj = re.match(self._VALID_URL, url)
  34         video_id = mobj.group('id')
  35
  36         config = self._download_json(self._PLAYER_CONFIG_URL, video_id,
  37                                      note='Downloading player config')
  38         url_template = 'http://nfl.com{contentURLTemplate:s}'.format(**config)
  39         video_data = self._download_json(url_template.format(id=video_id), video_id)
  40
  41         cdns = config.get('cdns')
  42         if not cdns:
  43             raise ExtractorError('Failed to get CDN data', expected=True)
  44
  45         formats = []
  46         streams = video_data.get('cdnData', {}).get('bitrateInfo', [])
  47         for name, cdn in cdns.items():
  48             # LimeLight streams don't seem to work
  49             if cdn.get('name') == 'LIMELIGHT':
  50                 continue
  51
  52             protocol = cdn.get('protocol')
  53             host = remove_end(cdn.get('host', ''), '/')
  54             if not (protocol and host):
  55                 continue
  56
  57             path_prefix = cdn.get('pathprefix', '')
  58             if path_prefix and not path_prefix.endswith('/'):
  59                 path_prefix = '%s/' % path_prefix
  60
  61             get_url = lambda p: '{protocol:s}://{host:s}/{prefix:s}{path:}'.format(
  62                 protocol=protocol,
  63                 host=host,
  64                 prefix=path_prefix,
  65                 path=p,
  66             )
  67
  68             if protocol == 'rtmp':
  69                 preference = -2
  70             elif 'prog' in name.lower():
  71                 preference = -1
  72             else:
  73                 preference = 0
  74
  75             for stream in streams:
  76                 path = stream.get('path')
  77                 if not path:
  78                     continue
  79
  80                 formats.append({
  81                     'url': get_url(path),
  82                     'vbr': int_or_none(stream.get('rate', 0), 1000),
  83                     'preference': preference,
  84                     'format_note': name,
  85                 })
  86
  87         self._sort_formats(formats)
  88
  89         thumbnail = None
  90         for q in ('xl', 'l', 'm', 's', 'xs'):
  91             thumbnail = video_data.get('imagePaths', {}).get(q)
  92             if thumbnail:
  93                 break
  94
  95         return {
  96             'id': video_id,
  97             'title': video_data.get('storyHeadline'),
  98             'formats': formats,
  99             'description': video_data.get('caption'),
 100             'duration': video_data.get('duration'),
 101             'thumbnail': thumbnail,
 102             'timestamp': int_or_none(video_data.get('posted'), 1000),
 103         }