_ 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
  17         formats = []
  18         for source in video_data['sources']:
  19             source_url = self._proto_relative_url(source['file'])
  20             source_type = source.get('type') or ''
  21             if source_type in ('application/vnd.apple.mpegurl', 'hls'):
  22                 formats.extend(self._extract_m3u8_formats(
  23                     source_url, video_id, 'mp4', 'm3u8_native', fatal=False))
  24             elif source_type.startswith('audio'):
  25                 formats.append({
  26                     'url': source_url,
  27                     'vcodec': 'none',
  28                 })
  29             else:
  30                 formats.append({
  31                     'url': source_url,
  32                     'width': int_or_none(source.get('width')),
  33                     'height': int_or_none(source.get('height')),
  34                 })
  35         self._sort_formats(formats)
  36
  37         subtitles = {}
  38         tracks = video_data.get('tracks')
  39         if tracks and isinstance(tracks, list):
  40             for track in tracks:
  41                 if track.get('file') and track.get('kind') == 'captions':
  42                     subtitles.setdefault(track.get('label') or 'en', []).append({
  43                         'url': self._proto_relative_url(track['file'])
  44                     })
  45
  46         return {
  47             'id': video_id,
  48             'title': video_data['title'] if require_title else video_data.get('title'),
  49             'description': video_data.get('description'),
  50             'thumbnail': self._proto_relative_url(video_data.get('image')),
  51             'timestamp': int_or_none(video_data.get('pubdate')),
  52             'duration': float_or_none(jwplayer_data.get('duration')),
  53             'subtitles': subtitles,
  54             'formats': formats,
  55         }
  56
  57
  58 class JWPlatformIE(JWPlatformBaseIE):
  59     _VALID_URL = r'(?:https?://content\.jwplatform\.com/(?:feeds|players|jw6)/|jwplatform:)(?P<id>[a-zA-Z0-9]{8})'
  60     _TEST = {
  61         'url': 'http://content.jwplatform.com/players/nPripu9l-ALJ3XQCI.js',
  62         'md5': 'fa8899fa601eb7c83a64e9d568bdf325',
  63         'info_dict': {
  64             'id': 'nPripu9l',
  65             'ext': 'mov',
  66             'title': 'Big Buck Bunny Trailer',
  67             'description': 'Big Buck Bunny is a short animated film by the Blender Institute. It is made using free and open source software.',
  68             'upload_date': '20081127',
  69             'timestamp': 1227796140,
  70         }
  71     }
  72
  73     @staticmethod
  74     def _extract_url(webpage):
  75         mobj = re.search(
  76             r'<script[^>]+?src=["\'](?P<url>(?:https?:)?//content.jwplatform.com/players/[a-zA-Z0-9]{8})',
  77             webpage)
  78         if mobj:
  79             return mobj.group('url')
  80
  81     def _real_extract(self, url):
  82         video_id = self._match_id(url)
  83         json_data = self._download_json('http://content.jwplatform.com/feeds/%s.json' % video_id, video_id)
  84         return self._parse_jwplayer_data(json_data, video_id)