]> gitweb @ CieloNegro.org - youtube-dl.git/blob - youtube_dl/extractor/shared.py
[techtalks] Relax _VALID_URL
[youtube-dl.git] / youtube_dl / extractor / shared.py
1 from __future__ import unicode_literals
2
3 import base64
4
5 from .common import InfoExtractor
6 from ..utils import (
7     ExtractorError,
8     int_or_none,
9     urlencode_postdata,
10 )
11
12
13 class SharedIE(InfoExtractor):
14     IE_DESC = 'shared.sx and vivo.sx'
15     _VALID_URL = r'https?://(?:shared|vivo)\.sx/(?P<id>[\da-z]{10})'
16
17     _TESTS = [{
18         'url': 'http://shared.sx/0060718775',
19         'md5': '106fefed92a8a2adb8c98e6a0652f49b',
20         'info_dict': {
21             'id': '0060718775',
22             'ext': 'mp4',
23             'title': 'Bmp4',
24             'filesize': 1720110,
25         },
26     }, {
27         'url': 'http://vivo.sx/d7ddda0e78',
28         'md5': '15b3af41be0b4fe01f4df075c2678b2c',
29         'info_dict': {
30             'id': 'd7ddda0e78',
31             'ext': 'mp4',
32             'title': 'Chicken',
33             'filesize': 528031,
34         },
35     }]
36
37     def _real_extract(self, url):
38         video_id = self._match_id(url)
39
40         webpage, urlh = self._download_webpage_handle(url, video_id)
41
42         if '>File does not exist<' in webpage:
43             raise ExtractorError(
44                 'Video %s does not exist' % video_id, expected=True)
45
46         download_form = self._hidden_inputs(webpage)
47
48         video_page = self._download_webpage(
49             urlh.geturl(), video_id, 'Downloading video page',
50             data=urlencode_postdata(download_form),
51             headers={
52                 'Content-Type': 'application/x-www-form-urlencoded',
53                 'Referer': urlh.geturl(),
54             })
55
56         video_url = self._html_search_regex(
57             r'data-url=(["\'])(?P<url>(?:(?!\1).)+)\1',
58             video_page, 'video URL', group='url')
59         title = base64.b64decode(self._html_search_meta(
60             'full:title', webpage, 'title').encode('utf-8')).decode('utf-8')
61         filesize = int_or_none(self._html_search_meta(
62             'full:size', webpage, 'file size', fatal=False))
63         thumbnail = self._html_search_regex(
64             r'data-poster=(["\'])(?P<url>(?:(?!\1).)+)\1',
65             video_page, 'thumbnail', default=None, group='url')
66
67         return {
68             'id': video_id,
69             'url': video_url,
70             'ext': 'mp4',
71             'filesize': filesize,
72             'title': title,
73             'thumbnail': thumbnail,
74         }