Raphaël G. Git Repositories - youtubedl/blob - 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         # private video
  87         'url': 'http://www.pornhub.com/view_video.php?viewkey=ph56fd731fce6b7',
  88         'only_matching': True,
  89     }, {
  90         'url': 'https://www.thumbzilla.com/video/ph56c6114abd99a/horny-girlfriend-sex',
  91         'only_matching': True,
  92     }]
  93
  94     @classmethod
  95     def _extract_url(cls, webpage):
  96         mobj = re.search(
  97             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?pornhub\.com/embed/\d+)\1', webpage)
  98         if mobj:
  99             return mobj.group('url')
 100
 101     def _extract_count(self, pattern, webpage, name):
 102         return str_to_int(self._search_regex(
 103             pattern, webpage, '%s count' % name, fatal=False))
 104
 105     def _real_extract(self, url):
 106         video_id = self._match_id(url)
 107
 108         req = sanitized_Request(
 109             'http://www.pornhub.com/view_video.php?viewkey=%s' % video_id)
 110         req.add_header('Cookie', 'age_verified=1')
 111         webpage = self._download_webpage(req, video_id)
 112
 113         error_msg = self._html_search_regex(
 114             r'(?s)<div[^>]+class=(["\'])(?:(?!\1).)*\b(?:removed|userMessageSection)\b(?:(?!\1).)*\1[^>]*>(?P<error>.+?)</div>',
 115             webpage, 'error message', default=None, group='error')
 116         if error_msg:
 117             error_msg = re.sub(r'\s+', ' ', error_msg)
 118             raise ExtractorError(
 119                 'PornHub said: %s' % error_msg,
 120                 expected=True, video_id=video_id)
 121
 122         # video_title from flashvars contains whitespace instead of non-ASCII (see
 123         # http://www.pornhub.com/view_video.php?viewkey=1331683002), not relying
 124         # on that anymore.
 125         title = self._html_search_meta(
 126             'twitter:title', webpage, default=None) or self._search_regex(
 127             (r'<h1[^>]+class=["\']title["\'][^>]*>(?P<title>[^<]+)',
 128              r'<div[^>]+data-video-title=(["\'])(?P<title>.+?)\1',
 129              r'shareTitle\s*=\s*(["\'])(?P<title>.+?)\1'),
 130             webpage, 'title', group='title')
 131
 132         flashvars = self._parse_json(
 133             self._search_regex(
 134                 r'var\s+flashvars_\d+\s*=\s*({.+?});', webpage, 'flashvars', default='{}'),
 135             video_id)
 136         if flashvars:
 137             thumbnail = flashvars.get('image_url')
 138             duration = int_or_none(flashvars.get('video_duration'))
 139         else:
 140             title, thumbnail, duration = [None] * 3
 141
 142         video_uploader = self._html_search_regex(
 143             r'(?s)From:&nbsp;.+?<(?:a href="/users/|a href="/channels/|span class="username)[^>]+>(.+?)<',
 144             webpage, 'uploader', fatal=False)
 145
 146         view_count = self._extract_count(
 147             r'<span class="count">([\d,\.]+)</span> views', webpage, 'view')
 148         like_count = self._extract_count(
 149             r'<span class="votesUp">([\d,\.]+)</span>', webpage, 'like')
 150         dislike_count = self._extract_count(
 151             r'<span class="votesDown">([\d,\.]+)</span>', webpage, 'dislike')
 152         comment_count = self._extract_count(
 153             r'All Comments\s*<span>\(([\d,.]+)\)', webpage, 'comment')
 154
 155         video_urls = list(map(compat_urllib_parse_unquote, re.findall(r"player_quality_[0-9]{3}p\s*=\s*'([^']+)'", webpage)))
 156         if webpage.find('"encrypted":true') != -1:
 157             password = compat_urllib_parse_unquote_plus(
 158                 self._search_regex(r'"video_title":"([^"]+)', webpage, 'password'))
 159             video_urls = list(map(lambda s: aes_decrypt_text(s, password, 32).decode('utf-8'), video_urls))
 160
 161         formats = []
 162         for video_url in video_urls:
 163             path = compat_urllib_parse_urlparse(video_url).path
 164             extension = os.path.splitext(path)[1][1:]
 165             format = path.split('/')[5].split('_')[:2]
 166             format = '-'.join(format)
 167
 168             m = re.match(r'^(?P<height>[0-9]+)[pP]-(?P<tbr>[0-9]+)[kK]$', format)
 169             if m is None:
 170                 height = None
 171                 tbr = None
 172             else:
 173                 height = int(m.group('height'))
 174                 tbr = int(m.group('tbr'))
 175
 176             formats.append({
 177                 'url': video_url,
 178                 'ext': extension,
 179                 'format': format,
 180                 'format_id': format,
 181                 'tbr': tbr,
 182                 'height': height,
 183             })
 184         self._sort_formats(formats)
 185
 186         return {
 187             'id': video_id,
 188             'uploader': video_uploader,
 189             'title': title,
 190             'thumbnail': thumbnail,
 191             'duration': duration,
 192             'view_count': view_count,
 193             'like_count': like_count,
 194             'dislike_count': dislike_count,
 195             'comment_count': comment_count,
 196             'formats': formats,
 197             'age_limit': 18,
 198         }
 199
 200
 201 class PornHubPlaylistBaseIE(InfoExtractor):
 202     def _extract_entries(self, webpage):
 203         return [
 204             self.url_result(
 205                 'http://www.pornhub.com/%s' % video_url,
 206                 PornHubIE.ie_key(), video_title=title)
 207             for video_url, title in orderedSet(re.findall(
 208                 r'href="/?(view_video\.php\?.*\bviewkey=[\da-z]+[^"]*)"[^>]*\s+title="([^"]+)"',
 209                 webpage))
 210         ]
 211
 212     def _real_extract(self, url):
 213         playlist_id = self._match_id(url)
 214
 215         webpage = self._download_webpage(url, playlist_id)
 216
 217         entries = self._extract_entries(webpage)
 218
 219         playlist = self._parse_json(
 220             self._search_regex(
 221                 r'playlistObject\s*=\s*({.+?});', webpage, 'playlist'),
 222             playlist_id)
 223
 224         return self.playlist_result(
 225             entries, playlist_id, playlist.get('title'), playlist.get('description'))
 226
 227
 228 class PornHubPlaylistIE(PornHubPlaylistBaseIE):
 229     _VALID_URL = r'https?://(?:www\.)?pornhub\.com/playlist/(?P<id>\d+)'
 230     _TESTS = [{
 231         'url': 'http://www.pornhub.com/playlist/6201671',
 232         'info_dict': {
 233             'id': '6201671',
 234             'title': 'P0p4',
 235         },
 236         'playlist_mincount': 35,
 237     }]
 238
 239
 240 class PornHubUserVideosIE(PornHubPlaylistBaseIE):
 241     _VALID_URL = r'https?://(?:www\.)?pornhub\.com/users/(?P<id>[^/]+)/videos'
 242     _TESTS = [{
 243         'url': 'http://www.pornhub.com/users/zoe_ph/videos/public',
 244         'info_dict': {
 245             'id': 'zoe_ph',
 246         },
 247         'playlist_mincount': 171,
 248     }, {
 249         'url': 'http://www.pornhub.com/users/rushandlia/videos',
 250         'only_matching': True,
 251     }]
 252
 253     def _real_extract(self, url):
 254         user_id = self._match_id(url)
 255
 256         entries = []
 257         for page_num in itertools.count(1):
 258             try:
 259                 webpage = self._download_webpage(
 260                     url, user_id, 'Downloading page %d' % page_num,
 261                     query={'page': page_num})
 262             except ExtractorError as e:
 263                 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 404:
 264                     break
 265             page_entries = self._extract_entries(webpage)
 266             if not page_entries:
 267                 break
 268             entries.extend(page_entries)
 269
 270         return self.playlist_result(entries, user_id)