]> git.bitcoin.ninja Git - youtube-dl/blob - youtube_dl/extractor/pornhub.py
[bleacherreport] fix style issues and simplify
[youtube-dl] / youtube_dl / extractor / pornhub.py
1 from __future__ import unicode_literals
2
3 import os
4 import re
5
6 from .common import InfoExtractor
7 from ..compat import (
8     compat_urllib_parse_unquote,
9     compat_urllib_parse_unquote_plus,
10     compat_urllib_parse_urlparse,
11     compat_urllib_request,
12 )
13 from ..utils import (
14     ExtractorError,
15     str_to_int,
16 )
17 from ..aes import (
18     aes_decrypt_text
19 )
20
21
22 class PornHubIE(InfoExtractor):
23     _VALID_URL = r'https?://(?:[a-z]+\.)?pornhub\.com/(?:view_video\.php\?viewkey=|embed/)(?P<id>[0-9a-z]+)'
24     _TESTS = [{
25         'url': 'http://www.pornhub.com/view_video.php?viewkey=648719015',
26         'md5': '882f488fa1f0026f023f33576004a2ed',
27         'info_dict': {
28             'id': '648719015',
29             'ext': 'mp4',
30             "uploader": "Babes",
31             "title": "Seductive Indian beauty strips down and fingers her pink pussy",
32             "age_limit": 18
33         }
34     }, {
35         'url': 'http://www.pornhub.com/view_video.php?viewkey=ph557bbb6676d2d',
36         'only_matching': True,
37     }, {
38         'url': 'http://fr.pornhub.com/view_video.php?viewkey=ph55ca2f9760862',
39         'only_matching': True,
40     }]
41
42     @classmethod
43     def _extract_url(cls, webpage):
44         mobj = re.search(
45             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?pornhub\.com/embed/\d+)\1', webpage)
46         if mobj:
47             return mobj.group('url')
48
49     def _extract_count(self, pattern, webpage, name):
50         return str_to_int(self._search_regex(
51             pattern, webpage, '%s count' % name, fatal=False))
52
53     def _real_extract(self, url):
54         video_id = self._match_id(url)
55
56         req = compat_urllib_request.Request(
57             'http://www.pornhub.com/view_video.php?viewkey=%s' % video_id)
58         req.add_header('Cookie', 'age_verified=1')
59         webpage = self._download_webpage(req, video_id)
60
61         error_msg = self._html_search_regex(
62             r'(?s)<div class="userMessageSection[^"]*".*?>(.*?)</div>',
63             webpage, 'error message', default=None)
64         if error_msg:
65             error_msg = re.sub(r'\s+', ' ', error_msg)
66             raise ExtractorError(
67                 'PornHub said: %s' % error_msg,
68                 expected=True, video_id=video_id)
69
70         video_title = self._html_search_regex(r'<h1 [^>]+>([^<]+)', webpage, 'title')
71         video_uploader = self._html_search_regex(
72             r'(?s)From:&nbsp;.+?<(?:a href="/users/|a href="/channels/|span class="username)[^>]+>(.+?)<',
73             webpage, 'uploader', fatal=False)
74         thumbnail = self._html_search_regex(r'"image_url":"([^"]+)', webpage, 'thumbnail', fatal=False)
75         if thumbnail:
76             thumbnail = compat_urllib_parse_unquote(thumbnail)
77
78         view_count = self._extract_count(
79             r'<span class="count">([\d,\.]+)</span> views', webpage, 'view')
80         like_count = self._extract_count(
81             r'<span class="votesUp">([\d,\.]+)</span>', webpage, 'like')
82         dislike_count = self._extract_count(
83             r'<span class="votesDown">([\d,\.]+)</span>', webpage, 'dislike')
84         comment_count = self._extract_count(
85             r'All Comments\s*<span>\(([\d,.]+)\)', webpage, 'comment')
86
87         video_urls = list(map(compat_urllib_parse_unquote, re.findall(r"player_quality_[0-9]{3}p\s*=\s*'([^']+)'", webpage)))
88         if webpage.find('"encrypted":true') != -1:
89             password = compat_urllib_parse_unquote_plus(
90                 self._search_regex(r'"video_title":"([^"]+)', webpage, 'password'))
91             video_urls = list(map(lambda s: aes_decrypt_text(s, password, 32).decode('utf-8'), video_urls))
92
93         formats = []
94         for video_url in video_urls:
95             path = compat_urllib_parse_urlparse(video_url).path
96             extension = os.path.splitext(path)[1][1:]
97             format = path.split('/')[5].split('_')[:2]
98             format = "-".join(format)
99
100             m = re.match(r'^(?P<height>[0-9]+)[pP]-(?P<tbr>[0-9]+)[kK]$', format)
101             if m is None:
102                 height = None
103                 tbr = None
104             else:
105                 height = int(m.group('height'))
106                 tbr = int(m.group('tbr'))
107
108             formats.append({
109                 'url': video_url,
110                 'ext': extension,
111                 'format': format,
112                 'format_id': format,
113                 'tbr': tbr,
114                 'height': height,
115             })
116         self._sort_formats(formats)
117
118         return {
119             'id': video_id,
120             'uploader': video_uploader,
121             'title': video_title,
122             'thumbnail': thumbnail,
123             'view_count': view_count,
124             'like_count': like_count,
125             'dislike_count': dislike_count,
126             'comment_count': comment_count,
127             'formats': formats,
128             'age_limit': 18,
129         }
130
131
132 class PornHubPlaylistIE(InfoExtractor):
133     _VALID_URL = r'https?://(?:www\.)?pornhub\.com/playlist/(?P<id>\d+)'
134     _TESTS = [{
135         'url': 'http://www.pornhub.com/playlist/6201671',
136         'info_dict': {
137             'id': '6201671',
138             'title': 'P0p4',
139         },
140         'playlist_mincount': 35,
141     }]
142
143     def _real_extract(self, url):
144         playlist_id = self._match_id(url)
145
146         webpage = self._download_webpage(url, playlist_id)
147
148         entries = [
149             self.url_result('http://www.pornhub.com/%s' % video_url, 'PornHub')
150             for video_url in set(re.findall('href="/?(view_video\.php\?viewkey=\d+[^"]*)"', webpage))
151         ]
152
153         playlist = self._parse_json(
154             self._search_regex(
155                 r'playlistObject\s*=\s*({.+?});', webpage, 'playlist'),
156             playlist_id)
157
158         return self.playlist_result(
159             entries, playlist_id, playlist.get('title'), playlist.get('description'))