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