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