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