Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/udn.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import json
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     js_to_json,
   8     ExtractorError,
   9 )
  10 from ..compat import compat_urlparse
  11
  12
  13 class UDNEmbedIE(InfoExtractor):
  14     _VALID_URL = r'https?://video\.udn\.com/(?:embed|play)/news/(?P<id>\d+)'
  15     _TESTS = [{
  16         'url': 'http://video.udn.com/embed/news/300040',
  17         'md5': 'de06b4c90b042c128395a88f0384817e',
  18         'info_dict': {
  19             'id': '300040',
  20             'ext': 'mp4',
  21             'title': '生物老師男變女 全校挺"做自己"',
  22             'thumbnail': 're:^https?://.*\.jpg$',
  23         }
  24     }, {
  25         'url': 'https://video.udn.com/embed/news/300040',
  26         'only_matching': True,
  27     }, {
  28         # From https://video.udn.com/news/303776
  29         'url': 'https://video.udn.com/play/news/303776',
  30         'only_matching': True,
  31     }]
  32
  33     def _real_extract(self, url):
  34         video_id = self._match_id(url)
  35
  36         page = self._download_webpage(url, video_id)
  37
  38         options = json.loads(js_to_json(self._html_search_regex(
  39             r'var options\s*=\s*([^;]+);', page, 'video urls dictionary')))
  40
  41         video_urls = options['video']
  42
  43         if video_urls.get('youtube'):
  44             return self.url_result(video_urls.get('youtube'), 'Youtube')
  45
  46         try:
  47             del video_urls['youtube']
  48         except KeyError:
  49             pass
  50
  51         formats = [{
  52             'url': self._download_webpage(
  53                 compat_urlparse.urljoin(url, api_url), video_id,
  54                 'retrieve url for %s video' % video_type),
  55             'format_id': video_type,
  56             'preference': 0 if video_type == 'mp4' else -1,
  57         } for video_type, api_url in video_urls.items() if api_url]
  58
  59         if not formats:
  60             raise ExtractorError('No videos found', expected=True)
  61
  62         self._sort_formats(formats)
  63
  64         thumbnail = None
  65
  66         if options.get('gallery') and len(options['gallery']):
  67             thumbnail = options['gallery'][0].get('original')
  68
  69         return {
  70             'id': video_id,
  71             'formats': formats,
  72             'title': options['title'],
  73             'thumbnail': thumbnail
  74         }