]> Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/porn91.py
debian/changelog: Annotate bugs to close.
[youtubedl] / youtube_dl / extractor / porn91.py
1 # encoding: utf-8
2 from __future__ import unicode_literals
3
4 from ..compat import compat_urllib_parse
5 from .common import InfoExtractor
6 from ..utils import (
7 parse_duration,
8 int_or_none,
9 ExtractorError,
10 )
11
12
13 class Porn91IE(InfoExtractor):
14 IE_NAME = '91porn'
15 _VALID_URL = r'(?:https?://)(?:www\.|)91porn\.com/.+?\?viewkey=(?P<id>[\w\d]+)'
16
17 _TEST = {
18 'url': 'http://91porn.com/view_video.php?viewkey=7e42283b4f5ab36da134',
19 'md5': '6df8f6d028bc8b14f5dbd73af742fb20',
20 'info_dict': {
21 'id': '7e42283b4f5ab36da134',
22 'title': '18岁大一漂亮学妹,水嫩性感,再爽一次!',
23 'ext': 'mp4',
24 'duration': 431,
25 }
26 }
27
28 def _real_extract(self, url):
29 video_id = self._match_id(url)
30 url = 'http://91porn.com/view_video.php?viewkey=%s' % video_id
31 self._set_cookie('91porn.com', 'language', 'cn_CN')
32 webpage = self._download_webpage(url, video_id, 'get HTML content')
33
34 if '作为游客,你每天只可观看10个视频' in webpage:
35 raise ExtractorError('91 Porn says: Daily limit 10 videos exceeded', expected=True)
36
37 title = self._search_regex(
38 r'<div id="viewvideo-title">([^<]+)</div>', webpage, 'title')
39 title = title.replace('\n', '')
40
41 # get real url
42 file_id = self._search_regex(
43 r'so.addVariable\(\'file\',\'(\d+)\'', webpage, 'file id')
44 sec_code = self._search_regex(
45 r'so.addVariable\(\'seccode\',\'([^\']+)\'', webpage, 'sec code')
46 max_vid = self._search_regex(
47 r'so.addVariable\(\'max_vid\',\'(\d+)\'', webpage, 'max vid')
48 url_params = compat_urllib_parse.urlencode({
49 'VID': file_id,
50 'mp4': '1',
51 'seccode': sec_code,
52 'max_vid': max_vid,
53 })
54 info_cn = self._download_webpage(
55 'http://91porn.com/getfile.php?' + url_params, video_id,
56 'get real video url')
57 video_url = self._search_regex(r'file=([^&]+)&', info_cn, 'url')
58
59 duration = parse_duration(self._search_regex(
60 r'时长:\s*</span>\s*(\d+:\d+)', webpage, 'duration', fatal=False))
61
62 comment_count = int_or_none(self._search_regex(
63 r'留言:\s*</span>\s*(\d+)', webpage, 'comment count', fatal=False))
64
65 return {
66 'id': video_id,
67 'title': title,
68 'url': video_url,
69 'duration': duration,
70 'comment_count': comment_count,
71 }