_ Git - youtube-dl/blob - youtube_dl/extractor/jwplatform.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 int_or_none
   8
   9
  10 class JWPlatformBaseIE(InfoExtractor):
  11     def _parse_jwplayer_data(self, jwplayer_data, video_id, require_title=True):
  12         video_data = jwplayer_data['playlist'][0]
  13         subtitles = {}
  14         for track in video_data['tracks']:
  15             if track['kind'] == 'captions':
  16                 subtitles[track['label']] = [{'url': self._proto_relative_url(track['file'])}]
  17
  18         formats = []
  19         for source in video_data['sources']:
  20             source_url = self._proto_relative_url(source['file'])
  21             source_type = source.get('type') or ''
  22             if source_type in ('application/vnd.apple.mpegurl', 'hls'):
  23                 formats.extend(self._extract_m3u8_formats(
  24                     source_url, video_id, 'mp4', 'm3u8_native', fatal=False))
  25             elif source_type.startswith('audio'):
  26                 formats.append({
  27                     'url': source_url,
  28                     'vcodec': 'none',
  29                 })
  30             else:
  31                 formats.append({
  32                     'url': source_url,
  33                     'width': int_or_none(source.get('width')),
  34                     'height': int_or_none(source.get('height')),
  35                 })
  36         self._sort_formats(formats)
  37
  38         return {
  39             'id': video_id,
  40             'title': video_data['title'] if require_title else video_data.get('title'),
  41             'description': video_data.get('description'),
  42             'thumbnail': self._proto_relative_url(video_data.get('image')),
  43             'timestamp': int_or_none(video_data.get('pubdate')),
  44             'subtitles': subtitles,
  45             'formats': formats,
  46         }
  47
  48
  49 class JWPlatformIE(JWPlatformBaseIE):
  50     _VALID_URL = r'(?:https?://content\.jwplatform\.com/(?:feeds|players|jw6)/|jwplatform:)(?P<id>[a-zA-Z0-9]{8})'
  51     _TEST = {
  52         'url': 'http://content.jwplatform.com/players/nPripu9l-ALJ3XQCI.js',
  53         'md5': 'fa8899fa601eb7c83a64e9d568bdf325',
  54         'info_dict': {
  55             'id': 'nPripu9l',
  56             'ext': 'mov',
  57             'title': 'Big Buck Bunny Trailer',
  58             'description': 'Big Buck Bunny is a short animated film by the Blender Institute. It is made using free and open source software.',
  59             'upload_date': '20081127',
  60             'timestamp': 1227796140,
  61         }
  62     }
  63
  64     @staticmethod
  65     def _extract_url(webpage):
  66         mobj = re.search(
  67             r'<script[^>]+?src=["\'](?P<url>(?:https?:)?//content.jwplatform.com/players/[a-zA-Z0-9]{8})',
  68             webpage)
  69         if mobj:
  70             return mobj.group('url')
  71
  72     def _real_extract(self, url):
  73         video_id = self._match_id(url)
  74         json_data = self._download_json('http://content.jwplatform.com/feeds/%s.json' % video_id, video_id)
  75         return self._parse_jwplayer_data(json_data, video_id)