_ Git - youtube-dl/blob - youtube_dl/extractor/pornhub.py

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