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