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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import functools
   5 import itertools
   6 import operator
   7 # import os
   8 import re
   9
  10 from .common import InfoExtractor
  11 from ..compat import (
  12     compat_HTTPError,
  13     # compat_urllib_parse_unquote,
  14     # compat_urllib_parse_unquote_plus,
  15     # compat_urllib_parse_urlparse,
  16 )
  17 from ..utils import (
  18     ExtractorError,
  19     int_or_none,
  20     js_to_json,
  21     orderedSet,
  22     # sanitized_Request,
  23     str_to_int,
  24 )
  25 # from ..aes import (
  26 #     aes_decrypt_text
  27 # )
  28
  29
  30 class PornHubIE(InfoExtractor):
  31     IE_DESC = 'PornHub and Thumbzilla'
  32     _VALID_URL = r'''(?x)
  33                     https?://
  34                         (?:
  35                             (?:[a-z]+\.)?pornhub\.com/(?:view_video\.php\?viewkey=|embed/)|
  36                             (?:www\.)?thumbzilla\.com/video/
  37                         )
  38                         (?P<id>[\da-z]+)
  39                     '''
  40     _TESTS = [{
  41         'url': 'http://www.pornhub.com/view_video.php?viewkey=648719015',
  42         'md5': '1e19b41231a02eba417839222ac9d58e',
  43         'info_dict': {
  44             'id': '648719015',
  45             'ext': 'mp4',
  46             'title': 'Seductive Indian beauty strips down and fingers her pink pussy',
  47             'uploader': 'Babes',
  48             'duration': 361,
  49             'view_count': int,
  50             'like_count': int,
  51             'dislike_count': int,
  52             'comment_count': int,
  53             'age_limit': 18,
  54             'tags': list,
  55             'categories': list,
  56         },
  57     }, {
  58         # non-ASCII title
  59         'url': 'http://www.pornhub.com/view_video.php?viewkey=1331683002',
  60         'info_dict': {
  61             'id': '1331683002',
  62             'ext': 'mp4',
  63             'title': '重庆婷婷女王足交',
  64             'uploader': 'cj397186295',
  65             'duration': 1753,
  66             'view_count': int,
  67             'like_count': int,
  68             'dislike_count': int,
  69             'comment_count': int,
  70             'age_limit': 18,
  71             'tags': list,
  72             'categories': list,
  73         },
  74         'params': {
  75             'skip_download': True,
  76         },
  77     }, {
  78         'url': 'http://www.pornhub.com/view_video.php?viewkey=ph557bbb6676d2d',
  79         'only_matching': True,
  80     }, {
  81         # removed at the request of cam4.com
  82         'url': 'http://fr.pornhub.com/view_video.php?viewkey=ph55ca2f9760862',
  83         'only_matching': True,
  84     }, {
  85         # removed at the request of the copyright owner
  86         'url': 'http://www.pornhub.com/view_video.php?viewkey=788152859',
  87         'only_matching': True,
  88     }, {
  89         # removed by uploader
  90         'url': 'http://www.pornhub.com/view_video.php?viewkey=ph572716d15a111',
  91         'only_matching': True,
  92     }, {
  93         # private video
  94         'url': 'http://www.pornhub.com/view_video.php?viewkey=ph56fd731fce6b7',
  95         'only_matching': True,
  96     }, {
  97         'url': 'https://www.thumbzilla.com/video/ph56c6114abd99a/horny-girlfriend-sex',
  98         'only_matching': True,
  99     }]
 100
 101     @staticmethod
 102     def _extract_urls(webpage):
 103         return re.findall(
 104             r'<iframe[^>]+?src=["\'](?P<url>(?:https?:)?//(?:www\.)?pornhub\.com/embed/[\da-z]+)',
 105             webpage)
 106
 107     def _extract_count(self, pattern, webpage, name):
 108         return str_to_int(self._search_regex(
 109             pattern, webpage, '%s count' % name, fatal=False))
 110
 111     def _real_extract(self, url):
 112         video_id = self._match_id(url)
 113
 114         def dl_webpage(platform):
 115             return self._download_webpage(
 116                 'http://www.pornhub.com/view_video.php?viewkey=%s' % video_id,
 117                 video_id, headers={
 118                     'Cookie': 'age_verified=1; platform=%s' % platform,
 119                 })
 120
 121         webpage = dl_webpage('pc')
 122
 123         error_msg = self._html_search_regex(
 124             r'(?s)<div[^>]+class=(["\'])(?:(?!\1).)*\b(?:removed|userMessageSection)\b(?:(?!\1).)*\1[^>]*>(?P<error>.+?)</div>',
 125             webpage, 'error message', default=None, group='error')
 126         if error_msg:
 127             error_msg = re.sub(r'\s+', ' ', error_msg)
 128             raise ExtractorError(
 129                 'PornHub said: %s' % error_msg,
 130                 expected=True, video_id=video_id)
 131
 132         tv_webpage = dl_webpage('tv')
 133
 134         encoded_url = self._search_regex(r'(var.*mediastring.*)</script>',
 135             tv_webpage, 'encoded url')
 136         assignments = encoded_url.split(";")
 137         js_vars = {}
 138
 139         def parse_js_value(inp):
 140             inp = re.sub(r'/\*[^*]*\*/', "", inp)
 141
 142             if "+" in inp:
 143                 inps = inp.split("+")
 144                 return functools.reduce(operator.concat, map(parse_js_value, inps))
 145
 146             inp = inp.strip()
 147             if inp in js_vars:
 148                 return js_vars[inp]
 149
 150             # Hope it's a string!
 151             assert inp.startswith('"') and inp.endswith('"')
 152             return inp[1:-1]
 153
 154         for assn in assignments:
 155             assn = assn.strip()
 156             if len(assn) == 0:
 157                 continue
 158
 159             assert assn.startswith("var ")
 160             assn = assn[4:]
 161             vname, value = assn.split("=", 1)
 162
 163             js_vars[vname] = parse_js_value(value)
 164
 165         video_url = js_vars["mediastring"]
 166
 167         title = self._search_regex(
 168             r'<h1>([^>]+)</h1>', tv_webpage, 'title', default=None)
 169
 170         # video_title from flashvars contains whitespace instead of non-ASCII (see
 171         # http://www.pornhub.com/view_video.php?viewkey=1331683002), not relying
 172         # on that anymore.
 173         title = title or self._html_search_meta(
 174             'twitter:title', webpage, default=None) or self._search_regex(
 175             (r'<h1[^>]+class=["\']title["\'][^>]*>(?P<title>[^<]+)',
 176              r'<div[^>]+data-video-title=(["\'])(?P<title>.+?)\1',
 177              r'shareTitle\s*=\s*(["\'])(?P<title>.+?)\1'),
 178             webpage, 'title', group='title')
 179
 180         flashvars = self._parse_json(
 181             self._search_regex(
 182                 r'var\s+flashvars_\d+\s*=\s*({.+?});', webpage, 'flashvars', default='{}'),
 183             video_id)
 184         if flashvars:
 185             thumbnail = flashvars.get('image_url')
 186             duration = int_or_none(flashvars.get('video_duration'))
 187         else:
 188             title, thumbnail, duration = [None] * 3
 189
 190         video_uploader = self._html_search_regex(
 191             r'(?s)From:&nbsp;.+?<(?:a href="/users/|a href="/channels/|span class="username)[^>]+>(.+?)<',
 192             webpage, 'uploader', fatal=False)
 193
 194         view_count = self._extract_count(
 195             r'<span class="count">([\d,\.]+)</span> views', webpage, 'view')
 196         like_count = self._extract_count(
 197             r'<span class="votesUp">([\d,\.]+)</span>', webpage, 'like')
 198         dislike_count = self._extract_count(
 199             r'<span class="votesDown">([\d,\.]+)</span>', webpage, 'dislike')
 200         comment_count = self._extract_count(
 201             r'All Comments\s*<span>\(([\d,.]+)\)', webpage, 'comment')
 202
 203         page_params = self._parse_json(self._search_regex(
 204             r'page_params\.zoneDetails\[([\'"])[^\'"]+\1\]\s*=\s*(?P<data>{[^}]+})',
 205             webpage, 'page parameters', group='data', default='{}'),
 206             video_id, transform_source=js_to_json, fatal=False)
 207         tags = categories = None
 208         if page_params:
 209             tags = page_params.get('tags', '').split(',')
 210             categories = page_params.get('categories', '').split(',')
 211
 212         return {
 213             'id': video_id,
 214             'url': video_url,
 215             'uploader': video_uploader,
 216             'title': title,
 217             'thumbnail': thumbnail,
 218             'duration': duration,
 219             'view_count': view_count,
 220             'like_count': like_count,
 221             'dislike_count': dislike_count,
 222             'comment_count': comment_count,
 223             # 'formats': formats,
 224             'age_limit': 18,
 225             'tags': tags,
 226             'categories': categories,
 227         }
 228
 229
 230 class PornHubPlaylistBaseIE(InfoExtractor):
 231     def _extract_entries(self, webpage):
 232         return [
 233             self.url_result(
 234                 'http://www.pornhub.com/%s' % video_url,
 235                 PornHubIE.ie_key(), video_title=title)
 236             for video_url, title in orderedSet(re.findall(
 237                 r'href="/?(view_video\.php\?.*\bviewkey=[\da-z]+[^"]*)"[^>]*\s+title="([^"]+)"',
 238                 webpage))
 239         ]
 240
 241     def _real_extract(self, url):
 242         playlist_id = self._match_id(url)
 243
 244         webpage = self._download_webpage(url, playlist_id)
 245
 246         # Only process container div with main playlist content skipping
 247         # drop-down menu that uses similar pattern for videos (see
 248         # https://github.com/rg3/youtube-dl/issues/11594).
 249         container = self._search_regex(
 250             r'(?s)(<div[^>]+class=["\']container.+)', webpage,
 251             'container', default=webpage)
 252
 253         entries = self._extract_entries(container)
 254
 255         playlist = self._parse_json(
 256             self._search_regex(
 257                 r'playlistObject\s*=\s*({.+?});', webpage, 'playlist'),
 258             playlist_id)
 259
 260         return self.playlist_result(
 261             entries, playlist_id, playlist.get('title'), playlist.get('description'))
 262
 263
 264 class PornHubPlaylistIE(PornHubPlaylistBaseIE):
 265     _VALID_URL = r'https?://(?:www\.)?pornhub\.com/playlist/(?P<id>\d+)'
 266     _TESTS = [{
 267         'url': 'http://www.pornhub.com/playlist/4667351',
 268         'info_dict': {
 269             'id': '4667351',
 270             'title': 'Nataly Hot',
 271         },
 272         'playlist_mincount': 2,
 273     }]
 274
 275
 276 class PornHubUserVideosIE(PornHubPlaylistBaseIE):
 277     _VALID_URL = r'https?://(?:www\.)?pornhub\.com/users/(?P<id>[^/]+)/videos'
 278     _TESTS = [{
 279         'url': 'http://www.pornhub.com/users/zoe_ph/videos/public',
 280         'info_dict': {
 281             'id': 'zoe_ph',
 282         },
 283         'playlist_mincount': 171,
 284     }, {
 285         'url': 'http://www.pornhub.com/users/rushandlia/videos',
 286         'only_matching': True,
 287     }]
 288
 289     def _real_extract(self, url):
 290         user_id = self._match_id(url)
 291
 292         entries = []
 293         for page_num in itertools.count(1):
 294             try:
 295                 webpage = self._download_webpage(
 296                     url, user_id, 'Downloading page %d' % page_num,
 297                     query={'page': page_num})
 298             except ExtractorError as e:
 299                 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 404:
 300                     break
 301             page_entries = self._extract_entries(webpage)
 302             if not page_entries:
 303                 break
 304             entries.extend(page_entries)
 305
 306         return self.playlist_result(entries, user_id)