youtube_dl/extractor/pornhub.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import itertools
   5 import os
   6 import re
   7
   8 from .common import InfoExtractor
   9 from ..compat import (
  10     compat_HTTPError,
  11     compat_urllib_parse_unquote,
  12     compat_urllib_parse_unquote_plus,
  13     compat_urllib_parse_urlparse,
  14 )
  15 from ..utils import (
  16     ExtractorError,
  17     int_or_none,
  18     orderedSet,
  19     sanitized_Request,
  20     str_to_int,
  21 )
  22 from ..aes import (
  23     aes_decrypt_text
  24 )
  25
  26
  27 class PornHubIE(InfoExtractor):
  28     IE_DESC = 'PornHub and Thumbzilla'
  29     _VALID_URL = r'''(?x)
  30                     https?://
  31                         (?:
  32                             (?:[a-z]+\.)?pornhub\.com/(?:view_video\.php\?viewkey=|embed/)|
  33                             (?:www\.)?thumbzilla\.com/video/
  34                         )
  35                         (?P<id>[0-9a-z]+)
  36                     '''
  37     _TESTS = [{
  38         'url': 'http://www.pornhub.com/view_video.php?viewkey=648719015',
  39         'md5': '1e19b41231a02eba417839222ac9d58e',
  40         'info_dict': {
  41             'id': '648719015',
  42             'ext': 'mp4',
  43             'title': 'Seductive Indian beauty strips down and fingers her pink pussy',
  44             'uploader': 'Babes',
  45             'duration': 361,
  46             'view_count': int,
  47             'like_count': int,
  48             'dislike_count': int,
  49             'comment_count': int,
  50             'age_limit': 18,
  51         },
  52     }, {
  53         # non-ASCII title
  54         'url': 'http://www.pornhub.com/view_video.php?viewkey=1331683002',
  55         'info_dict': {
  56             'id': '1331683002',
  57             'ext': 'mp4',
  58             'title': '重庆婷婷女王足交',
  59             'uploader': 'cj397186295',
  60             'duration': 1753,
  61             'view_count': int,
  62             'like_count': int,
  63             'dislike_count': int,
  64             'comment_count': int,
  65             'age_limit': 18,
  66         },
  67         'params': {
  68             'skip_download': True,
  69         },
  70     }, {
  71         'url': 'http://www.pornhub.com/view_video.php?viewkey=ph557bbb6676d2d',
  72         'only_matching': True,
  73     }, {
  74         # removed at the request of cam4.com
  75         'url': 'http://fr.pornhub.com/view_video.php?viewkey=ph55ca2f9760862',
  76         'only_matching': True,
  77     }, {
  78         # removed at the request of the copyright owner
  79         'url': 'http://www.pornhub.com/view_video.php?viewkey=788152859',
  80         'only_matching': True,
  81     }, {
  82         # removed by uploader
  83         'url': 'http://www.pornhub.com/view_video.php?viewkey=ph572716d15a111',
  84         'only_matching': True,
  85     }, {
  86         'url': 'https://www.thumbzilla.com/video/ph56c6114abd99a/horny-girlfriend-sex',
  87         'only_matching': True,
  88     }]
  89
  90     @classmethod
  91     def _extract_url(cls, webpage):
  92         mobj = re.search(
  93             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?pornhub\.com/embed/\d+)\1', webpage)
  94         if mobj:
  95             return mobj.group('url')
  96
  97     def _extract_count(self, pattern, webpage, name):
  98         return str_to_int(self._search_regex(
  99             pattern, webpage, '%s count' % name, fatal=False))
 100
 101     def _real_extract(self, url):
 102         video_id = self._match_id(url)
 103
 104         req = sanitized_Request(
 105             'http://www.pornhub.com/view_video.php?viewkey=%s' % video_id)
 106         req.add_header('Cookie', 'age_verified=1')
 107         webpage = self._download_webpage(req, video_id)
 108
 109         error_msg = self._html_search_regex(
 110             r'(?s)<div[^>]+class=(["\']).*?\bremoved\b.*?\1[^>]*>(?P<error>.+?)</div>',
 111             webpage, 'error message', default=None, group='error')
 112         if error_msg:
 113             error_msg = re.sub(r'\s+', ' ', error_msg)
 114             raise ExtractorError(
 115                 'PornHub said: %s' % error_msg,
 116                 expected=True, video_id=video_id)
 117
 118         # video_title from flashvars contains whitespace instead of non-ASCII (see
 119         # http://www.pornhub.com/view_video.php?viewkey=1331683002), not relying
 120         # on that anymore.
 121         title = self._html_search_meta(
 122             'twitter:title', webpage, default=None) or self._search_regex(
 123             (r'<h1[^>]+class=["\']title["\'][^>]*>(?P<title>[^<]+)',
 124              r'<div[^>]+data-video-title=(["\'])(?P<title>.+?)\1',
 125              r'shareTitle\s*=\s*(["\'])(?P<title>.+?)\1'),
 126             webpage, 'title', group='title')
 127
 128         flashvars = self._parse_json(
 129             self._search_regex(
 130                 r'var\s+flashvars_\d+\s*=\s*({.+?});', webpage, 'flashvars', default='{}'),
 131             video_id)
 132         if flashvars:
 133             thumbnail = flashvars.get('image_url')
 134             duration = int_or_none(flashvars.get('video_duration'))
 135         else:
 136             title, thumbnail, duration = [None] * 3
 137
 138         video_uploader = self._html_search_regex(
 139             r'(?s)From:&nbsp;.+?<(?:a href="/users/|a href="/channels/|span class="username)[^>]+>(.+?)<',
 140             webpage, 'uploader', fatal=False)
 141
 142         view_count = self._extract_count(
 143             r'<span class="count">([\d,\.]+)</span> views', webpage, 'view')
 144         like_count = self._extract_count(
 145             r'<span class="votesUp">([\d,\.]+)</span>', webpage, 'like')
 146         dislike_count = self._extract_count(
 147             r'<span class="votesDown">([\d,\.]+)</span>', webpage, 'dislike')
 148         comment_count = self._extract_count(
 149             r'All Comments\s*<span>\(([\d,.]+)\)', webpage, 'comment')
 150
 151         video_urls = list(map(compat_urllib_parse_unquote, re.findall(r"player_quality_[0-9]{3}p\s*=\s*'([^']+)'", webpage)))
 152         if webpage.find('"encrypted":true') != -1:
 153             password = compat_urllib_parse_unquote_plus(
 154                 self._search_regex(r'"video_title":"([^"]+)', webpage, 'password'))
 155             video_urls = list(map(lambda s: aes_decrypt_text(s, password, 32).decode('utf-8'), video_urls))
 156
 157         formats = []
 158         for video_url in video_urls:
 159             path = compat_urllib_parse_urlparse(video_url).path
 160             extension = os.path.splitext(path)[1][1:]
 161             format = path.split('/')[5].split('_')[:2]
 162             format = '-'.join(format)
 163
 164             m = re.match(r'^(?P<height>[0-9]+)[pP]-(?P<tbr>[0-9]+)[kK]$', format)
 165             if m is None:
 166                 height = None
 167                 tbr = None
 168             else:
 169                 height = int(m.group('height'))
 170                 tbr = int(m.group('tbr'))
 171
 172             formats.append({
 173                 'url': video_url,
 174                 'ext': extension,
 175                 'format': format,
 176                 'format_id': format,
 177                 'tbr': tbr,
 178                 'height': height,
 179             })
 180         self._sort_formats(formats)
 181
 182         return {
 183             'id': video_id,
 184             'uploader': video_uploader,
 185             'title': title,
 186             'thumbnail': thumbnail,
 187             'duration': duration,
 188             'view_count': view_count,
 189             'like_count': like_count,
 190             'dislike_count': dislike_count,
 191             'comment_count': comment_count,
 192             'formats': formats,
 193             'age_limit': 18,
 194         }
 195
 196
 197 class PornHubPlaylistBaseIE(InfoExtractor):
 198     def _extract_entries(self, webpage):
 199         return [
 200             self.url_result(
 201                 'http://www.pornhub.com/%s' % video_url,
 202                 PornHubIE.ie_key(), video_title=title)
 203             for video_url, title in orderedSet(re.findall(
 204                 r'href="/?(view_video\.php\?.*\bviewkey=[\da-z]+[^"]*)"[^>]*\s+title="([^"]+)"',
 205                 webpage))
 206         ]
 207
 208     def _real_extract(self, url):
 209         playlist_id = self._match_id(url)
 210
 211         webpage = self._download_webpage(url, playlist_id)
 212
 213         entries = self._extract_entries(webpage)
 214
 215         playlist = self._parse_json(
 216             self._search_regex(
 217                 r'playlistObject\s*=\s*({.+?});', webpage, 'playlist'),
 218             playlist_id)
 219
 220         return self.playlist_result(
 221             entries, playlist_id, playlist.get('title'), playlist.get('description'))
 222
 223
 224 class PornHubPlaylistIE(PornHubPlaylistBaseIE):
 225     _VALID_URL = r'https?://(?:www\.)?pornhub\.com/playlist/(?P<id>\d+)'
 226     _TESTS = [{
 227         'url': 'http://www.pornhub.com/playlist/6201671',
 228         'info_dict': {
 229             'id': '6201671',
 230             'title': 'P0p4',
 231         },
 232         'playlist_mincount': 35,
 233     }]
 234
 235
 236 class PornHubUserVideosIE(PornHubPlaylistBaseIE):
 237     _VALID_URL = r'https?://(?:www\.)?pornhub\.com/users/(?P<id>[^/]+)/videos'
 238     _TESTS = [{
 239         'url': 'http://www.pornhub.com/users/zoe_ph/videos/public',
 240         'info_dict': {
 241             'id': 'zoe_ph',
 242         },
 243         'playlist_mincount': 171,
 244     }, {
 245         'url': 'http://www.pornhub.com/users/rushandlia/videos',
 246         'only_matching': True,
 247     }]
 248
 249     def _real_extract(self, url):
 250         user_id = self._match_id(url)
 251
 252         entries = []
 253         for page_num in itertools.count(1):
 254             try:
 255                 webpage = self._download_webpage(
 256                     url, user_id, 'Downloading page %d' % page_num,
 257                     query={'page': page_num})
 258             except ExtractorError as e:
 259                 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 404:
 260                     break
 261             page_entries = self._extract_entries(webpage)
 262             if not page_entries:
 263                 break
 264             entries.extend(page_entries)
 265
 266         return self.playlist_result(entries, user_id)