yt_dlp/extractor/chilloutzone.py

   1 import json
   2
   3 from .common import InfoExtractor
   4 from .youtube import YoutubeIE
   5 from ..compat import compat_b64decode
   6 from ..utils import (
   7     clean_html,
   8     ExtractorError
   9 )
  10
  11
  12 class ChilloutzoneIE(InfoExtractor):
  13     _VALID_URL = r'https?://(?:www\.)?chilloutzone\.net/video/(?P<id>[\w|-]+)\.html'
  14     _TESTS = [{
  15         'url': 'http://www.chilloutzone.net/video/enemene-meck-alle-katzen-weg.html',
  16         'md5': 'a76f3457e813ea0037e5244f509e66d1',
  17         'info_dict': {
  18             'id': 'enemene-meck-alle-katzen-weg',
  19             'ext': 'mp4',
  20             'title': 'Enemene Meck - Alle Katzen weg',
  21             'description': 'Ist das der Umkehrschluss des Niesenden Panda-Babys?',
  22         },
  23     }, {
  24         'note': 'Video hosted at YouTube',
  25         'url': 'http://www.chilloutzone.net/video/eine-sekunde-bevor.html',
  26         'info_dict': {
  27             'id': '1YVQaAgHyRU',
  28             'ext': 'mp4',
  29             'title': '16 Photos Taken 1 Second Before Disaster',
  30             'description': 'md5:58a8fcf6a459fe0a08f54140f0ad1814',
  31             'uploader': 'BuzzFeedVideo',
  32             'uploader_id': 'BuzzFeedVideo',
  33             'upload_date': '20131105',
  34         },
  35     }, {
  36         'note': 'Video hosted at Vimeo',
  37         'url': 'http://www.chilloutzone.net/video/icon-blending.html',
  38         'md5': '2645c678b8dc4fefcc0e1b60db18dac1',
  39         'info_dict': {
  40             'id': '85523671',
  41             'ext': 'mp4',
  42             'title': 'The Sunday Times - Icons',
  43             'description': 're:(?s)^Watch the making of - makingoficons.com.{300,}',
  44             'uploader': 'Us',
  45             'uploader_id': 'usfilms',
  46             'upload_date': '20140131'
  47         },
  48     }]
  49
  50     def _real_extract(self, url):
  51         mobj = self._match_valid_url(url)
  52         video_id = mobj.group('id')
  53
  54         webpage = self._download_webpage(url, video_id)
  55
  56         base64_video_info = self._html_search_regex(
  57             r'var cozVidData = "(.+?)";', webpage, 'video data')
  58         decoded_video_info = compat_b64decode(base64_video_info).decode('utf-8')
  59         video_info_dict = json.loads(decoded_video_info)
  60
  61         # get video information from dict
  62         video_url = video_info_dict['mediaUrl']
  63         description = clean_html(video_info_dict.get('description'))
  64         title = video_info_dict['title']
  65         native_platform = video_info_dict['nativePlatform']
  66         native_video_id = video_info_dict['nativeVideoId']
  67         source_priority = video_info_dict['sourcePriority']
  68
  69         # If nativePlatform is None a fallback mechanism is used (i.e. youtube embed)
  70         if native_platform is None:
  71             youtube_url = YoutubeIE._extract_url(webpage)
  72             if youtube_url:
  73                 return self.url_result(youtube_url, ie=YoutubeIE.ie_key())
  74
  75         # Non Fallback: Decide to use native source (e.g. youtube or vimeo) or
  76         # the own CDN
  77         if source_priority == 'native':
  78             if native_platform == 'youtube':
  79                 return self.url_result(native_video_id, ie='Youtube')
  80             if native_platform == 'vimeo':
  81                 return self.url_result(
  82                     'http://vimeo.com/' + native_video_id, ie='Vimeo')
  83
  84         if not video_url:
  85             raise ExtractorError('No video found')
  86
  87         return {
  88             'id': video_id,
  89             'url': video_url,
  90             'ext': 'mp4',
  91             'title': title,
  92             'description': description,
  93         }