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

   1 from __future__ import unicode_literals
   2
   3 import re
   4 import base64
   5 import json
   6
   7 from .common import InfoExtractor
   8 from ..utils import (
   9     clean_html,
  10     ExtractorError
  11 )
  12
  13
  14 class ChilloutzoneIE(InfoExtractor):
  15     _VALID_URL = r'https?://(?:www\.)?chilloutzone\.net/video/(?P<id>[\w|-]+)\.html'
  16     _TESTS = [{
  17         'url': 'http://www.chilloutzone.net/video/enemene-meck-alle-katzen-weg.html',
  18         'md5': 'a76f3457e813ea0037e5244f509e66d1',
  19         'info_dict': {
  20             'id': 'enemene-meck-alle-katzen-weg',
  21             'ext': 'mp4',
  22             'title': 'Enemene Meck - Alle Katzen weg',
  23             'description': 'Ist das der Umkehrschluss des Niesenden Panda-Babys?',
  24         },
  25     }, {
  26         'note': 'Video hosted at YouTube',
  27         'url': 'http://www.chilloutzone.net/video/eine-sekunde-bevor.html',
  28         'info_dict': {
  29             'id': '1YVQaAgHyRU',
  30             'ext': 'mp4',
  31             'title': '16 Photos Taken 1 Second Before Disaster',
  32             'description': 'md5:58a8fcf6a459fe0a08f54140f0ad1814',
  33             'uploader': 'BuzzFeedVideo',
  34             'uploader_id': 'BuzzFeedVideo',
  35             'upload_date': '20131105',
  36         },
  37     }, {
  38         'note': 'Video hosted at Vimeo',
  39         'url': 'http://www.chilloutzone.net/video/icon-blending.html',
  40         'md5': '2645c678b8dc4fefcc0e1b60db18dac1',
  41         'info_dict': {
  42             'id': '85523671',
  43             'ext': 'mp4',
  44             'title': 'The Sunday Times - Icons',
  45             'description': 're:(?s)^Watch the making of - makingoficons.com.{300,}',
  46             'uploader': 'Us',
  47             'uploader_id': 'usfilms',
  48             'upload_date': '20140131'
  49         },
  50     }]
  51
  52     def _real_extract(self, url):
  53         mobj = re.match(self._VALID_URL, url)
  54         video_id = mobj.group('id')
  55
  56         webpage = self._download_webpage(url, video_id)
  57
  58         base64_video_info = self._html_search_regex(
  59             r'var cozVidData = "(.+?)";', webpage, 'video data')
  60         decoded_video_info = base64.b64decode(base64_video_info).decode("utf-8")
  61         video_info_dict = json.loads(decoded_video_info)
  62
  63         # get video information from dict
  64         video_url = video_info_dict['mediaUrl']
  65         description = clean_html(video_info_dict.get('description'))
  66         title = video_info_dict['title']
  67         native_platform = video_info_dict['nativePlatform']
  68         native_video_id = video_info_dict['nativeVideoId']
  69         source_priority = video_info_dict['sourcePriority']
  70
  71         # If nativePlatform is None a fallback mechanism is used (i.e. youtube embed)
  72         if native_platform is None:
  73             youtube_url = self._html_search_regex(
  74                 r'<iframe.* src="((?:https?:)?//(?:[^.]+\.)?youtube\.com/.+?)"',
  75                 webpage, 'fallback video URL', default=None)
  76             if youtube_url is not None:
  77                 return self.url_result(youtube_url, ie='Youtube')
  78
  79         # Non Fallback: Decide to use native source (e.g. youtube or vimeo) or
  80         # the own CDN
  81         if source_priority == 'native':
  82             if native_platform == 'youtube':
  83                 return self.url_result(native_video_id, ie='Youtube')
  84             if native_platform == 'vimeo':
  85                 return self.url_result(
  86                     'http://vimeo.com/' + native_video_id, ie='Vimeo')
  87
  88         if not video_url:
  89             raise ExtractorError('No video found')
  90
  91         return {
  92             'id': video_id,
  93             'url': video_url,
  94             'ext': 'mp4',
  95             'title': title,
  96             'description': description,
  97         }