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