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     js_to_json,
  19     orderedSet,
  20     sanitized_Request,
  21     str_to_int,
  22 )
  23 from ..aes import (
  24     aes_decrypt_text
  25 )
  26
  27
  28 class PornHubIE(InfoExtractor):
  29     IE_DESC = 'PornHub and Thumbzilla'
  30     _VALID_URL = r'''(?x)
  31                     https?://
  32                         (?:
  33                             (?:[a-z]+\.)?pornhub\.com/(?:view_video\.php\?viewkey=|embed/)|
  34                             (?:www\.)?thumbzilla\.com/video/
  35                         )
  36                         (?P<id>[\da-z]+)
  37                     '''
  38     _TESTS = [{
  39         'url': 'http://www.pornhub.com/view_video.php?viewkey=648719015',
  40         'md5': '1e19b41231a02eba417839222ac9d58e',
  41         'info_dict': {
  42             'id': '648719015',
  43             'ext': 'mp4',
  44             'title': 'Seductive Indian beauty strips down and fingers her pink pussy',
  45             'uploader': 'Babes',
  46             'duration': 361,
  47             'view_count': int,
  48             'like_count': int,
  49             'dislike_count': int,
  50             'comment_count': int,
  51             'age_limit': 18,
  52             'tags': list,
  53             'categories': list,
  54         },
  55     }, {
  56         # non-ASCII title
  57         'url': 'http://www.pornhub.com/view_video.php?viewkey=1331683002',
  58         'info_dict': {
  59             'id': '1331683002',
  60             'ext': 'mp4',
  61             'title': '重庆婷婷女王足交',
  62             'uploader': 'cj397186295',
  63             'duration': 1753,
  64             'view_count': int,
  65             'like_count': int,
  66             'dislike_count': int,
  67             'comment_count': int,
  68             'age_limit': 18,
  69             'tags': list,
  70             'categories': list,
  71         },
  72         'params': {
  73             'skip_download': True,
  74         },
  75     }, {
  76         'url': 'http://www.pornhub.com/view_video.php?viewkey=ph557bbb6676d2d',
  77         'only_matching': True,
  78     }, {
  79         # removed at the request of cam4.com
  80         'url': 'http://fr.pornhub.com/view_video.php?viewkey=ph55ca2f9760862',
  81         'only_matching': True,
  82     }, {
  83         # removed at the request of the copyright owner
  84         'url': 'http://www.pornhub.com/view_video.php?viewkey=788152859',
  85         'only_matching': True,
  86     }, {
  87         # removed by uploader
  88         'url': 'http://www.pornhub.com/view_video.php?viewkey=ph572716d15a111',
  89         'only_matching': True,
  90     }, {
  91         # private video
  92         'url': 'http://www.pornhub.com/view_video.php?viewkey=ph56fd731fce6b7',
  93         'only_matching': True,
  94     }, {
  95         'url': 'https://www.thumbzilla.com/video/ph56c6114abd99a/horny-girlfriend-sex',
  96         'only_matching': True,
  97     }]
  98
  99     @staticmethod
 100     def _extract_urls(webpage):
 101         return re.findall(
 102             r'<iframe[^>]+?src=["\'](?P<url>(?:https?:)?//(?:www\.)?pornhub\.com/embed/[\da-z]+)',
 103             webpage)
 104
 105     def _extract_count(self, pattern, webpage, name):
 106         return str_to_int(self._search_regex(
 107             pattern, webpage, '%s count' % name, fatal=False))
 108
 109     def _real_extract(self, url):
 110         video_id = self._match_id(url)
 111
 112         req = sanitized_Request(
 113             'http://www.pornhub.com/view_video.php?viewkey=%s' % video_id)
 114         req.add_header('Cookie', 'age_verified=1')
 115         webpage = self._download_webpage(req, video_id)
 116
 117         error_msg = self._html_search_regex(
 118             r'(?s)<div[^>]+class=(["\'])(?:(?!\1).)*\b(?:removed|userMessageSection)\b(?:(?!\1).)*\1[^>]*>(?P<error>.+?)</div>',
 119             webpage, 'error message', default=None, group='error')
 120         if error_msg:
 121             error_msg = re.sub(r'\s+', ' ', error_msg)
 122             raise ExtractorError(
 123                 'PornHub said: %s' % error_msg,
 124                 expected=True, video_id=video_id)
 125
 126         # video_title from flashvars contains whitespace instead of non-ASCII (see
 127         # http://www.pornhub.com/view_video.php?viewkey=1331683002), not relying
 128         # on that anymore.
 129         title = self._html_search_meta(
 130             'twitter:title', webpage, default=None) or self._search_regex(
 131             (r'<h1[^>]+class=["\']title["\'][^>]*>(?P<title>[^<]+)',
 132              r'<div[^>]+data-video-title=(["\'])(?P<title>.+?)\1',
 133              r'shareTitle\s*=\s*(["\'])(?P<title>.+?)\1'),
 134             webpage, 'title', group='title')
 135
 136         flashvars = self._parse_json(
 137             self._search_regex(
 138                 r'var\s+flashvars_\d+\s*=\s*({.+?});', webpage, 'flashvars', default='{}'),
 139             video_id)
 140         if flashvars:
 141             thumbnail = flashvars.get('image_url')
 142             duration = int_or_none(flashvars.get('video_duration'))
 143         else:
 144             title, thumbnail, duration = [None] * 3
 145
 146         video_uploader = self._html_search_regex(
 147             r'(?s)From:&nbsp;.+?<(?:a href="/users/|a href="/channels/|span class="username)[^>]+>(.+?)<',
 148             webpage, 'uploader', fatal=False)
 149
 150         view_count = self._extract_count(
 151             r'<span class="count">([\d,\.]+)</span> views', webpage, 'view')
 152         like_count = self._extract_count(
 153             r'<span class="votesUp">([\d,\.]+)</span>', webpage, 'like')
 154         dislike_count = self._extract_count(
 155             r'<span class="votesDown">([\d,\.]+)</span>', webpage, 'dislike')
 156         comment_count = self._extract_count(
 157             r'All Comments\s*<span>\(([\d,.]+)\)', webpage, 'comment')
 158
 159         video_urls = list(map(compat_urllib_parse_unquote, re.findall(r"player_quality_[0-9]{3}p\s*=\s*'([^']+)'", webpage)))
 160         if webpage.find('"encrypted":true') != -1:
 161             password = compat_urllib_parse_unquote_plus(
 162                 self._search_regex(r'"video_title":"([^"]+)', webpage, 'password'))
 163             video_urls = list(map(lambda s: aes_decrypt_text(s, password, 32).decode('utf-8'), video_urls))
 164
 165         formats = []
 166         for video_url in video_urls:
 167             path = compat_urllib_parse_urlparse(video_url).path
 168             extension = os.path.splitext(path)[1][1:]
 169             format = path.split('/')[5].split('_')[:2]
 170             format = '-'.join(format)
 171
 172             m = re.match(r'^(?P<height>[0-9]+)[pP]-(?P<tbr>[0-9]+)[kK]$', format)
 173             if m is None:
 174                 height = None
 175                 tbr = None
 176             else:
 177                 height = int(m.group('height'))
 178                 tbr = int(m.group('tbr'))
 179
 180             formats.append({
 181                 'url': video_url,
 182                 'ext': extension,
 183                 'format': format,
 184                 'format_id': format,
 185                 'tbr': tbr,
 186                 'height': height,
 187             })
 188         self._sort_formats(formats)
 189
 190         page_params = self._parse_json(self._search_regex(
 191             r'page_params\.zoneDetails\[([\'"])[^\'"]+\1\]\s*=\s*(?P<data>{[^}]+})',
 192             webpage, 'page parameters', group='data', default='{}'),
 193             video_id, transform_source=js_to_json, fatal=False)
 194         tags = categories = None
 195         if page_params:
 196             tags = page_params.get('tags', '').split(',')
 197             categories = page_params.get('categories', '').split(',')
 198
 199         return {
 200             'id': video_id,
 201             'uploader': video_uploader,
 202             'title': title,
 203             'thumbnail': thumbnail,
 204             'duration': duration,
 205             'view_count': view_count,
 206             'like_count': like_count,
 207             'dislike_count': dislike_count,
 208             'comment_count': comment_count,
 209             'formats': formats,
 210             'age_limit': 18,
 211             'tags': tags,
 212             'categories': categories,
 213         }
 214
 215
 216 class PornHubPlaylistBaseIE(InfoExtractor):
 217     def _extract_entries(self, webpage):
 218         return [
 219             self.url_result(
 220                 'http://www.pornhub.com/%s' % video_url,
 221                 PornHubIE.ie_key(), video_title=title)
 222             for video_url, title in orderedSet(re.findall(
 223                 r'href="/?(view_video\.php\?.*\bviewkey=[\da-z]+[^"]*)"[^>]*\s+title="([^"]+)"',
 224                 webpage))
 225         ]
 226
 227     def _real_extract(self, url):
 228         playlist_id = self._match_id(url)
 229
 230         webpage = self._download_webpage(url, playlist_id)
 231
 232         # Only process container div with main playlist content skipping
 233         # drop-down menu that uses similar pattern for videos (see
 234         # https://github.com/rg3/youtube-dl/issues/11594).
 235         container = self._search_regex(
 236             r'(?s)(<div[^>]+class=["\']container.+)', webpage,
 237             'container', default=webpage)
 238
 239         entries = self._extract_entries(container)
 240
 241         playlist = self._parse_json(
 242             self._search_regex(
 243                 r'playlistObject\s*=\s*({.+?});', webpage, 'playlist'),
 244             playlist_id)
 245
 246         return self.playlist_result(
 247             entries, playlist_id, playlist.get('title'), playlist.get('description'))
 248
 249
 250 class PornHubPlaylistIE(PornHubPlaylistBaseIE):
 251     _VALID_URL = r'https?://(?:www\.)?pornhub\.com/playlist/(?P<id>\d+)'
 252     _TESTS = [{
 253         'url': 'http://www.pornhub.com/playlist/4667351',
 254         'info_dict': {
 255             'id': '4667351',
 256             'title': 'Nataly Hot',
 257         },
 258         'playlist_mincount': 2,
 259     }]
 260
 261
 262 class PornHubUserVideosIE(PornHubPlaylistBaseIE):
 263     _VALID_URL = r'https?://(?:www\.)?pornhub\.com/users/(?P<id>[^/]+)/videos'
 264     _TESTS = [{
 265         'url': 'http://www.pornhub.com/users/zoe_ph/videos/public',
 266         'info_dict': {
 267             'id': 'zoe_ph',
 268         },
 269         'playlist_mincount': 171,
 270     }, {
 271         'url': 'http://www.pornhub.com/users/rushandlia/videos',
 272         'only_matching': True,
 273     }]
 274
 275     def _real_extract(self, url):
 276         user_id = self._match_id(url)
 277
 278         entries = []
 279         for page_num in itertools.count(1):
 280             try:
 281                 webpage = self._download_webpage(
 282                     url, user_id, 'Downloading page %d' % page_num,
 283                     query={'page': page_num})
 284             except ExtractorError as e:
 285                 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 404:
 286                     break
 287             page_entries = self._extract_entries(webpage)
 288             if not page_entries:
 289                 break
 290             entries.extend(page_entries)
 291
 292         return self.playlist_result(entries, user_id)