2 from __future__ import unicode_literals
7 from .common import InfoExtractor
13 from ..compat import compat_urlparse
16 class UDNEmbedIE(InfoExtractor):
18 _PROTOCOL_RELATIVE_VALID_URL = r'//video\.udn\.com/(?:embed|play)/news/(?P<id>\d+)'
19 _VALID_URL = r'https?:' + _PROTOCOL_RELATIVE_VALID_URL
21 'url': 'http://video.udn.com/embed/news/300040',
25 'title': '生物老師男變女 全校挺"做自己"',
26 'thumbnail': 're:^https?://.*\.jpg$',
30 'skip_download': True,
33 'url': 'https://video.udn.com/embed/news/300040',
34 'only_matching': True,
36 # From https://video.udn.com/news/303776
37 'url': 'https://video.udn.com/play/news/303776',
38 'only_matching': True,
41 def _real_extract(self, url):
42 video_id = self._match_id(url)
44 page = self._download_webpage(url, video_id)
46 options = json.loads(js_to_json(self._html_search_regex(
47 r'var\s+options\s*=\s*([^;]+);', page, 'video urls dictionary')))
49 video_urls = options['video']
51 if video_urls.get('youtube'):
52 return self.url_result(video_urls.get('youtube'), 'Youtube')
55 for video_type, api_url in video_urls.items():
59 video_url = self._download_webpage(
60 compat_urlparse.urljoin(url, api_url), video_id,
61 note='retrieve url for %s video' % video_type)
63 ext = determine_ext(video_url)
65 formats.extend(self._extract_m3u8_formats(
66 video_url, video_id, ext='mp4', m3u8_id='hls'))
68 formats.extend(self._extract_f4m_formats(
69 video_url, video_id, f4m_id='hds'))
71 mobj = re.search(r'_(?P<height>\d+)p_(?P<tbr>\d+).mp4', video_url)
74 # video_type may be 'mp4', which confuses YoutubeDL
75 'format_id': 'http-' + video_type,
79 'height': int_or_none(mobj.group('height')),
80 'tbr': int_or_none(mobj.group('tbr')),
82 formats.append(a_format)
84 self._sort_formats(formats)
89 } for img_type, img_url in options.get('gallery', [{}])[0].items() if img_url]
94 'title': options['title'],
95 'thumbnails': thumbnails,