_ Git - youtube-dl/blob - 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 )
  12 from ..utils import (
  13     ExtractorError,
  14     sanitized_Request,
  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 = sanitized_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 PornHubPlaylistBaseIE(InfoExtractor):
 133     def _extract_entries(self, webpage):
 134         return [
 135             self.url_result('http://www.pornhub.com/%s' % video_url, PornHubIE.ie_key())
 136             for video_url in set(re.findall(
 137                 r'href="/?(view_video\.php\?.*\bviewkey=[\da-z]+[^"]*)"', webpage))
 138         ]
 139
 140     def _real_extract(self, url):
 141         playlist_id = self._match_id(url)
 142
 143         webpage = self._download_webpage(url, playlist_id)
 144
 145         entries = self._extract_entries(webpage)
 146
 147         playlist = self._parse_json(
 148             self._search_regex(
 149                 r'playlistObject\s*=\s*({.+?});', webpage, 'playlist'),
 150             playlist_id)
 151
 152         return self.playlist_result(
 153             entries, playlist_id, playlist.get('title'), playlist.get('description'))
 154
 155
 156 class PornHubPlaylistIE(PornHubPlaylistBaseIE):
 157     _VALID_URL = r'https?://(?:www\.)?pornhub\.com/playlist/(?P<id>\d+)'
 158     _TESTS = [{
 159         'url': 'http://www.pornhub.com/playlist/6201671',
 160         'info_dict': {
 161             'id': '6201671',
 162             'title': 'P0p4',
 163         },
 164         'playlist_mincount': 35,
 165     }]
 166
 167
 168 class PornHubUserVideosIE(PornHubPlaylistBaseIE):
 169     _VALID_URL = r'https?://(?:www\.)?pornhub\.com/users/(?P<id>[^/]+)/videos'
 170     _TESTS = [{
 171         'url': 'http://www.pornhub.com/users/rushandlia/videos',
 172         'info_dict': {
 173             'id': 'rushandlia',
 174         },
 175         'playlist_mincount': 13,
 176     }]
 177
 178     def _real_extract(self, url):
 179         user_id = self._match_id(url)
 180
 181         webpage = self._download_webpage(url, user_id)
 182
 183         return self.playlist_result(self._extract_entries(webpage), user_id)