_ Git - youtube-dl/blob - youtube_dl/extractor/jwplatform.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..utils import int_or_none
   6
   7
   8 class JWPlatformIE(InfoExtractor):
   9     _VALID_URL = r'(?:https?://content\.jwplatform\.com/(?:feeds|players|jw6)/|jwplatform:)(?P<id>[a-zA-Z0-9]{8})'
  10     _TEST = {
  11         'url': 'http://content.jwplatform.com/players/nPripu9l-ALJ3XQCI.js',
  12         'md5': 'fa8899fa601eb7c83a64e9d568bdf325',
  13         'info_dict': {
  14             'id': 'nPripu9l',
  15             'ext': 'mov',
  16             'title': 'Big Buck Bunny Trailer',
  17             'description': 'Big Buck Bunny is a short animated film by the Blender Institute. It is made using free and open source software.',
  18             'upload_date': '20081127',
  19             'timestamp': 1227796140,
  20         }
  21     }
  22
  23     @staticmethod
  24     def _extract_url(webpage):
  25         mobj = re.search(
  26             r'<script[^>]+?src=["\'](?P<url>(?:https?:)?//content.jwplatform.com/players/[a-zA-Z0-9]{8}',
  27             webpage)
  28         if mobj:
  29             return mobj.group('url')
  30
  31     def _real_extract(self, url):
  32         video_id = self._match_id(url)
  33         json_data = self._download_json('http://content.jwplatform.com/feeds/%s.json' % video_id, video_id)
  34         video_data = json_data['playlist'][0]
  35         subtitles = {}
  36         for track in video_data['tracks']:
  37             if track['kind'] == 'captions':
  38                 subtitles[track['label']] = [{'url': self._proto_relative_url(track['file'])}]
  39
  40         formats = []
  41         for source in video_data['sources']:
  42             source_url = self._proto_relative_url(source['file'])
  43             source_type = source.get('type') or ''
  44             if source_type == 'application/vnd.apple.mpegurl':
  45                 formats.extend(self._extract_m3u8_formats(source_url, video_id, 'mp4', 'm3u8_native', fatal=None))
  46             elif source_type.startswith('audio'):
  47                 formats.append({
  48                     'url': source_url,
  49                     'vcodec': 'none',
  50                 })
  51             else:
  52                 formats.append({
  53                     'url': source_url,
  54                     'width': int_or_none(source.get('width')),
  55                     'height': int_or_none(source.get('height')),
  56                 })
  57         self._sort_formats(formats)
  58
  59         return {
  60             'id': video_data['mediaid'],
  61             'title': video_data['title'],
  62             'description': video_data.get('description'),
  63             'thumbnail': self._proto_relative_url(video_data.get('image')),
  64             'timestamp': int_or_none(video_data.get('pubdate')),
  65             'subtitles': subtitles,
  66             'formats': formats,
  67         }