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

   1 import datetime
   2 import json
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     remove_start,
   8 )
   9
  10
  11 class BlinkxIE(InfoExtractor):
  12     _VALID_URL = r'^(?:https?://(?:www\.)blinkx\.com/#?ce/|blinkx:)(?P<id>[^?]+)'
  13     _IE_NAME = u'blinkx'
  14
  15     _TEST = {
  16         u'url': u'http://www.blinkx.com/ce/8aQUy7GVFYgFzpKhT0oqsilwOGFRVXk3R1ZGWWdGenBLaFQwb3FzaWx3OGFRVXk3R1ZGWWdGenB',
  17         u'file': u'8aQUy7GV.mp4',
  18         u'md5': u'2e9a07364af40163a908edbf10bb2492',
  19         u'info_dict': {
  20             u"title": u"Police Car Rolls Away",
  21             u"uploader": u"stupidvideos.com",
  22             u"upload_date": u"20131215",
  23             u"description": u"A police car gently rolls away from a fight. Maybe it felt weird being around a confrontation and just had to get out of there!",
  24             u"duration": 14.886,
  25             u"thumbnails": [{
  26                 "width": 100,
  27                 "height": 76,
  28                 "url": "http://cdn.blinkx.com/stream/b/41/StupidVideos/20131215/1873969261/1873969261_tn_0.jpg",
  29             }],
  30         },
  31     }
  32
  33     def _real_extract(self, url):
  34         m = re.match(self._VALID_URL, url)
  35         video_id = m.group('id')
  36         display_id = video_id[:8]
  37
  38         api_url = (u'https://apib4.blinkx.com/api.php?action=play_video&' +
  39                    u'video=%s' % video_id)
  40         data_json = self._download_webpage(api_url, display_id)
  41         data = json.loads(data_json)['api']['results'][0]
  42         dt = datetime.datetime.fromtimestamp(data['pubdate_epoch'])
  43         upload_date = dt.strftime('%Y%m%d')
  44
  45         duration = None
  46         thumbnails = []
  47         formats = []
  48         for m in data['media']:
  49             if m['type'] == 'jpg':
  50                 thumbnails.append({
  51                     'url': m['link'],
  52                     'width': int(m['w']),
  53                     'height': int(m['h']),
  54                 })
  55             elif m['type'] == 'original':
  56                 duration = m['d']
  57             elif m['type'] == 'youtube':
  58                 yt_id = m['link']
  59                 self.to_screen(u'Youtube video detected: %s' % yt_id)
  60                 return self.url_result(yt_id, 'Youtube', video_id=yt_id)
  61             elif m['type'] in ('flv', 'mp4'):
  62                 vcodec = remove_start(m['vcodec'], 'ff')
  63                 acodec = remove_start(m['acodec'], 'ff')
  64                 format_id = (u'%s-%sk-%s' %
  65                              (vcodec,
  66                               (int(m['vbr']) + int(m['abr'])) // 1000,
  67                               m['w']))
  68                 formats.append({
  69                     'format_id': format_id,
  70                     'url': m['link'],
  71                     'vcodec': vcodec,
  72                     'acodec': acodec,
  73                     'abr': int(m['abr']) // 1000,
  74                     'vbr': int(m['vbr']) // 1000,
  75                     'width': int(m['w']),
  76                     'height': int(m['h']),
  77                 })
  78         formats.sort(key=lambda f: (f['width'], f['vbr'], f['abr']))
  79
  80         return {
  81             'id': display_id,
  82             'fullid': video_id,
  83             'title': data['title'],
  84             'formats': formats,
  85             'uploader': data['channel_name'],
  86             'upload_date': upload_date,
  87             'description': data.get('description'),
  88             'thumbnails': thumbnails,
  89             'duration': duration,
  90         }