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     IE_DESC = '聯合影音'
  15     _PROTOCOL_RELATIVE_VALID_URL = r'//video\.udn\.com/(?:embed|play)/news/(?P<id>\d+)'
  16     _VALID_URL = r'https?:' + _PROTOCOL_RELATIVE_VALID_URL
  17     _TESTS = [{
  18         'url': 'http://video.udn.com/embed/news/300040',
  19         'md5': 'de06b4c90b042c128395a88f0384817e',
  20         'info_dict': {
  21             'id': '300040',
  22             'ext': 'mp4',
  23             'title': '生物老師男變女 全校挺"做自己"',
  24             'thumbnail': 're:^https?://.*\.jpg$',
  25         }
  26     }, {
  27         'url': 'https://video.udn.com/embed/news/300040',
  28         'only_matching': True,
  29     }, {
  30         # From https://video.udn.com/news/303776
  31         'url': 'https://video.udn.com/play/news/303776',
  32         'only_matching': True,
  33     }]
  34
  35     def _real_extract(self, url):
  36         video_id = self._match_id(url)
  37
  38         page = self._download_webpage(url, video_id)
  39
  40         options = json.loads(js_to_json(self._html_search_regex(
  41             r'var options\s*=\s*([^;]+);', page, 'video urls dictionary')))
  42
  43         video_urls = options['video']
  44
  45         if video_urls.get('youtube'):
  46             return self.url_result(video_urls.get('youtube'), 'Youtube')
  47
  48         try:
  49             del video_urls['youtube']
  50         except KeyError:
  51             pass
  52
  53         formats = [{
  54             'url': self._download_webpage(
  55                 compat_urlparse.urljoin(url, api_url), video_id,
  56                 'retrieve url for %s video' % video_type),
  57             'format_id': video_type,
  58             'preference': 0 if video_type == 'mp4' else -1,
  59         } for video_type, api_url in video_urls.items() if api_url]
  60
  61         if not formats:
  62             raise ExtractorError('No videos found', expected=True)
  63
  64         self._sort_formats(formats)
  65
  66         thumbnail = None
  67
  68         if options.get('gallery') and len(options['gallery']):
  69             thumbnail = options['gallery'][0].get('original')
  70
  71         return {
  72             'id': video_id,
  73             'formats': formats,
  74             'title': options['title'],
  75             'thumbnail': thumbnail
  76         }