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

   1 from __future__ import unicode_literals
   2
   3 import json
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     remove_start,
   8     int_or_none,
   9 )
  10
  11
  12 class BlinkxIE(InfoExtractor):
  13     _VALID_URL = r'(?:https?://(?:www\.)blinkx\.com/#?ce/|blinkx:)(?P<id>[^?]+)'
  14     IE_NAME = 'blinkx'
  15
  16     _TEST = {
  17         'url': 'http://www.blinkx.com/ce/Da0Gw3xc5ucpNduzLuDDlv4WC9PuI4fDi1-t6Y3LyfdY2SZS5Urbvn-UPJvrvbo8LTKTc67Wu2rPKSQDJyZeeORCR8bYkhs8lI7eqddznH2ofh5WEEdjYXnoRtj7ByQwt7atMErmXIeYKPsSDuMAAqJDlQZ-3Ff4HJVeH_s3Gh8oQ',
  18         'md5': '337cf7a344663ec79bf93a526a2e06c7',
  19         'info_dict': {
  20             'id': 'Da0Gw3xc',
  21             'ext': 'mp4',
  22             'title': 'No Daily Show for John Oliver; HBO Show Renewed - IGN News',
  23             'uploader': 'IGN News',
  24             'upload_date': '20150217',
  25             'timestamp': 1424215740,
  26             'description': 'HBO has renewed Last Week Tonight With John Oliver for two more seasons.',
  27             'duration': 47.743333,
  28         },
  29     }
  30
  31     def _real_extract(self, url):
  32         video_id = self._match_id(url)
  33         display_id = video_id[:8]
  34
  35         api_url = ('https://apib4.blinkx.com/api.php?action=play_video&' +
  36                    'video=%s' % video_id)
  37         data_json = self._download_webpage(api_url, display_id)
  38         data = json.loads(data_json)['api']['results'][0]
  39         duration = None
  40         thumbnails = []
  41         formats = []
  42         for m in data['media']:
  43             if m['type'] == 'jpg':
  44                 thumbnails.append({
  45                     'url': m['link'],
  46                     'width': int(m['w']),
  47                     'height': int(m['h']),
  48                 })
  49             elif m['type'] == 'original':
  50                 duration = float(m['d'])
  51             elif m['type'] == 'youtube':
  52                 yt_id = m['link']
  53                 self.to_screen('Youtube video detected: %s' % yt_id)
  54                 return self.url_result(yt_id, 'Youtube', video_id=yt_id)
  55             elif m['type'] in ('flv', 'mp4'):
  56                 vcodec = remove_start(m['vcodec'], 'ff')
  57                 acodec = remove_start(m['acodec'], 'ff')
  58                 vbr = int_or_none(m.get('vbr') or m.get('vbitrate'), 1000)
  59                 abr = int_or_none(m.get('abr') or m.get('abitrate'), 1000)
  60                 tbr = vbr + abr if vbr and abr else None
  61                 format_id = '%s-%sk-%s' % (vcodec, tbr, m['w'])
  62                 formats.append({
  63                     'format_id': format_id,
  64                     'url': m['link'],
  65                     'vcodec': vcodec,
  66                     'acodec': acodec,
  67                     'abr': abr,
  68                     'vbr': vbr,
  69                     'tbr': tbr,
  70                     'width': int_or_none(m.get('w')),
  71                     'height': int_or_none(m.get('h')),
  72                 })
  73
  74         self._sort_formats(formats)
  75
  76         return {
  77             'id': display_id,
  78             'fullid': video_id,
  79             'title': data['title'],
  80             'formats': formats,
  81             'uploader': data['channel_name'],
  82             'timestamp': data['pubdate_epoch'],
  83             'description': data.get('description'),
  84             'thumbnails': thumbnails,
  85             'duration': duration,
  86         }