]> git.bitcoin.ninja Git - youtube-dl/blob - youtube_dl/extractor/blinkx.py
Implement a different adult sites checking algorithm
[youtube-dl] / youtube_dl / extractor / blinkx.py
1 from __future__ import unicode_literals
2
3 import datetime
4 import json
5 import re
6
7 from .common import InfoExtractor
8 from ..utils import (
9     remove_start,
10 )
11
12
13 class BlinkxIE(InfoExtractor):
14     _VALID_URL = r'^(?:https?://(?:www\.)blinkx\.com/#?ce/|blinkx:)(?P<id>[^?]+)'
15     IE_NAME = 'blinkx'
16
17     _TEST = {
18         'url': 'http://www.blinkx.com/ce/8aQUy7GVFYgFzpKhT0oqsilwOGFRVXk3R1ZGWWdGenBLaFQwb3FzaWx3OGFRVXk3R1ZGWWdGenB',
19         'file': '8aQUy7GV.mp4',
20         'md5': '2e9a07364af40163a908edbf10bb2492',
21         'info_dict': {
22             "title": "Police Car Rolls Away",
23             "uploader": "stupidvideos.com",
24             "upload_date": "20131215",
25             "description": "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!",
26             "duration": 14.886,
27             "thumbnails": [{
28                 "width": 100,
29                 "height": 76,
30                 "url": "http://cdn.blinkx.com/stream/b/41/StupidVideos/20131215/1873969261/1873969261_tn_0.jpg",
31             }],
32         },
33     }
34
35     def _real_extract(self, rl):
36         m = re.match(self._VALID_URL, rl)
37         video_id = m.group('id')
38         display_id = video_id[:8]
39
40         api_url = (u'https://apib4.blinkx.com/api.php?action=play_video&' +
41                    'video=%s' % video_id)
42         data_json = self._download_webpage(api_url, display_id)
43         data = json.loads(data_json)['api']['results'][0]
44         dt = datetime.datetime.fromtimestamp(data['pubdate_epoch'])
45         pload_date = dt.strftime('%Y%m%d')
46
47         duration = None
48         thumbnails = []
49         formats = []
50         for m in data['media']:
51             if m['type'] == 'jpg':
52                 thumbnails.append({
53                     'url': m['link'],
54                     'width': int(m['w']),
55                     'height': int(m['h']),
56                 })
57             elif m['type'] == 'original':
58                 duration = m['d']
59             elif m['type'] == 'youtube':
60                 yt_id = m['link']
61                 self.to_screen(u'Youtube video detected: %s' % yt_id)
62                 return self.url_result(yt_id, 'Youtube', video_id=yt_id)
63             elif m['type'] in ('flv', 'mp4'):
64                 vcodec = remove_start(m['vcodec'], 'ff')
65                 acodec = remove_start(m['acodec'], 'ff')
66                 tbr = (int(m['vbr']) + int(m['abr'])) // 1000
67                 format_id = (u'%s-%sk-%s' %
68                              (vcodec,
69                               tbr,
70                               m['w']))
71                 formats.append({
72                     'format_id': format_id,
73                     'url': m['link'],
74                     'vcodec': vcodec,
75                     'acodec': acodec,
76                     'abr': int(m['abr']) // 1000,
77                     'vbr': int(m['vbr']) // 1000,
78                     'tbr': tbr,
79                     'width': int(m['w']),
80                     'height': int(m['h']),
81                 })
82
83         self._sort_formats(formats)
84
85         return {
86             'id': display_id,
87             'fullid': video_id,
88             'title': data['title'],
89             'formats': formats,
90             'uploader': data['channel_name'],
91             'upload_date': pload_date,
92             'description': data.get('description'),
93             'thumbnails': thumbnails,
94             'duration': duration,
95         }