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