]> git.bitcoin.ninja Git - youtube-dl/blob - youtube_dl/extractor/vidme.py
[SpankBang] Use python2.6 compatible string formatting spec
[youtube-dl] / youtube_dl / extractor / vidme.py
1 from __future__ import unicode_literals
2
3 from .common import InfoExtractor
4 from ..utils import (
5     int_or_none,
6     float_or_none,
7     str_to_int,
8 )
9
10
11 class VidmeIE(InfoExtractor):
12     _VALID_URL = r'https?://vid\.me/(?:e/)?(?P<id>[\da-zA-Z]+)'
13     _TEST = {
14         'url': 'https://vid.me/QNB',
15         'md5': 'f42d05e7149aeaec5c037b17e5d3dc82',
16         'info_dict': {
17             'id': 'QNB',
18             'ext': 'mp4',
19             'title': 'Fishing for piranha - the easy way',
20             'description': 'source: https://www.facebook.com/photo.php?v=312276045600871',
21             'duration': 119.92,
22             'timestamp': 1406313244,
23             'upload_date': '20140725',
24             'thumbnail': 're:^https?://.*\.jpg',
25         },
26     }
27
28     def _real_extract(self, url):
29         video_id = self._match_id(url)
30         webpage = self._download_webpage(url, video_id)
31
32         video_url = self._html_search_regex(
33             r'<source src="([^"]+)"', webpage, 'video URL')
34
35         title = self._og_search_title(webpage)
36         description = self._og_search_description(webpage, default='')
37         thumbnail = self._og_search_thumbnail(webpage)
38         timestamp = int_or_none(self._og_search_property('updated_time', webpage, fatal=False))
39         width = int_or_none(self._og_search_property('video:width', webpage, fatal=False))
40         height = int_or_none(self._og_search_property('video:height', webpage, fatal=False))
41         duration = float_or_none(self._html_search_regex(
42             r'data-duration="([^"]+)"', webpage, 'duration', fatal=False))
43         view_count = str_to_int(self._html_search_regex(
44             r'<(?:li|span) class="video_views">\s*([\d,\.]+)\s*plays?', webpage, 'view count', fatal=False))
45         like_count = str_to_int(self._html_search_regex(
46             r'class="score js-video-vote-score"[^>]+data-score="([\d,\.\s]+)">',
47             webpage, 'like count', fatal=False))
48
49         return {
50             'id': video_id,
51             'url': video_url,
52             'title': title,
53             'description': description,
54             'thumbnail': thumbnail,
55             'timestamp': timestamp,
56             'width': width,
57             'height': height,
58             'duration': duration,
59             'view_count': view_count,
60             'like_count': like_count,
61         }