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