Raphaël G. Git Repositories - youtubedl/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     remove_quotes,
  24     str_to_int,
  25 )
  26 # from ..aes import (
  27 #     aes_decrypt_text
  28 # )
  29
  30
  31 class PornHubIE(InfoExtractor):
  32     IE_DESC = 'PornHub and Thumbzilla'
  33     _VALID_URL = r'''(?x)
  34                     https?://
  35                         (?:
  36                             (?:[a-z]+\.)?pornhub\.com/(?:(?:view_video\.php|video/show)\?viewkey=|embed/)|
  37                             (?:www\.)?thumbzilla\.com/video/
  38                         )
  39                         (?P<id>[\da-z]+)
  40                     '''
  41     _TESTS = [{
  42         'url': 'http://www.pornhub.com/view_video.php?viewkey=648719015',
  43         'md5': '1e19b41231a02eba417839222ac9d58e',
  44         'info_dict': {
  45             'id': '648719015',
  46             'ext': 'mp4',
  47             'title': 'Seductive Indian beauty strips down and fingers her pink pussy',
  48             'uploader': 'Babes',
  49             'duration': 361,
  50             'view_count': int,
  51             'like_count': int,
  52             'dislike_count': int,
  53             'comment_count': int,
  54             'age_limit': 18,
  55             'tags': list,
  56             'categories': list,
  57         },
  58     }, {
  59         # non-ASCII title
  60         'url': 'http://www.pornhub.com/view_video.php?viewkey=1331683002',
  61         'info_dict': {
  62             'id': '1331683002',
  63             'ext': 'mp4',
  64             'title': '重庆婷婷女王足交',
  65             'uploader': 'cj397186295',
  66             'duration': 1753,
  67             'view_count': int,
  68             'like_count': int,
  69             'dislike_count': int,
  70             'comment_count': int,
  71             'age_limit': 18,
  72             'tags': list,
  73             'categories': list,
  74         },
  75         'params': {
  76             'skip_download': True,
  77         },
  78     }, {
  79         'url': 'http://www.pornhub.com/view_video.php?viewkey=ph557bbb6676d2d',
  80         'only_matching': True,
  81     }, {
  82         # removed at the request of cam4.com
  83         'url': 'http://fr.pornhub.com/view_video.php?viewkey=ph55ca2f9760862',
  84         'only_matching': True,
  85     }, {
  86         # removed at the request of the copyright owner
  87         'url': 'http://www.pornhub.com/view_video.php?viewkey=788152859',
  88         'only_matching': True,
  89     }, {
  90         # removed by uploader
  91         'url': 'http://www.pornhub.com/view_video.php?viewkey=ph572716d15a111',
  92         'only_matching': True,
  93     }, {
  94         # private video
  95         'url': 'http://www.pornhub.com/view_video.php?viewkey=ph56fd731fce6b7',
  96         'only_matching': True,
  97     }, {
  98         'url': 'https://www.thumbzilla.com/video/ph56c6114abd99a/horny-girlfriend-sex',
  99         'only_matching': True,
 100     }, {
 101         'url': 'http://www.pornhub.com/video/show?viewkey=648719015',
 102         'only_matching': True,
 103     }]
 104
 105     @staticmethod
 106     def _extract_urls(webpage):
 107         return re.findall(
 108             r'<iframe[^>]+?src=["\'](?P<url>(?:https?:)?//(?:www\.)?pornhub\.com/embed/[\da-z]+)',
 109             webpage)
 110
 111     def _extract_count(self, pattern, webpage, name):
 112         return str_to_int(self._search_regex(
 113             pattern, webpage, '%s count' % name, fatal=False))
 114
 115     def _real_extract(self, url):
 116         video_id = self._match_id(url)
 117
 118         def dl_webpage(platform):
 119             return self._download_webpage(
 120                 'http://www.pornhub.com/view_video.php?viewkey=%s' % video_id,
 121                 video_id, headers={
 122                     'Cookie': 'age_verified=1; platform=%s' % platform,
 123                 })
 124
 125         webpage = dl_webpage('pc')
 126
 127         error_msg = self._html_search_regex(
 128             r'(?s)<div[^>]+class=(["\'])(?:(?!\1).)*\b(?:removed|userMessageSection)\b(?:(?!\1).)*\1[^>]*>(?P<error>.+?)</div>',
 129             webpage, 'error message', default=None, group='error')
 130         if error_msg:
 131             error_msg = re.sub(r'\s+', ' ', error_msg)
 132             raise ExtractorError(
 133                 'PornHub said: %s' % error_msg,
 134                 expected=True, video_id=video_id)
 135
 136         tv_webpage = dl_webpage('tv')
 137
 138         assignments = self._search_regex(
 139             r'(var.+?mediastring.+?)</script>', tv_webpage,
 140             'encoded url').split(';')
 141
 142         js_vars = {}
 143
 144         def parse_js_value(inp):
 145             inp = re.sub(r'/\*(?:(?!\*/).)*?\*/', '', inp)
 146             if '+' in inp:
 147                 inps = inp.split('+')
 148                 return functools.reduce(
 149                     operator.concat, map(parse_js_value, inps))
 150             inp = inp.strip()
 151             if inp in js_vars:
 152                 return js_vars[inp]
 153             return remove_quotes(inp)
 154
 155         for assn in assignments:
 156             assn = assn.strip()
 157             if not assn:
 158                 continue
 159             assn = re.sub(r'var\s+', '', assn)
 160             vname, value = assn.split('=', 1)
 161             js_vars[vname] = parse_js_value(value)
 162
 163         video_url = js_vars['mediastring']
 164
 165         title = self._search_regex(
 166             r'<h1>([^>]+)</h1>', tv_webpage, 'title', default=None)
 167
 168         # video_title from flashvars contains whitespace instead of non-ASCII (see
 169         # http://www.pornhub.com/view_video.php?viewkey=1331683002), not relying
 170         # on that anymore.
 171         title = title or self._html_search_meta(
 172             'twitter:title', webpage, default=None) or self._search_regex(
 173             (r'<h1[^>]+class=["\']title["\'][^>]*>(?P<title>[^<]+)',
 174              r'<div[^>]+data-video-title=(["\'])(?P<title>.+?)\1',
 175              r'shareTitle\s*=\s*(["\'])(?P<title>.+?)\1'),
 176             webpage, 'title', group='title')
 177
 178         flashvars = self._parse_json(
 179             self._search_regex(
 180                 r'var\s+flashvars_\d+\s*=\s*({.+?});', webpage, 'flashvars', default='{}'),
 181             video_id)
 182         if flashvars:
 183             thumbnail = flashvars.get('image_url')
 184             duration = int_or_none(flashvars.get('video_duration'))
 185         else:
 186             title, thumbnail, duration = [None] * 3
 187
 188         video_uploader = self._html_search_regex(
 189             r'(?s)From:&nbsp;.+?<(?:a\b[^>]+\bhref=["\']/(?:user|channel)s/|span\b[^>]+\bclass=["\']username)[^>]+>(.+?)<',
 190             webpage, 'uploader', fatal=False)
 191
 192         view_count = self._extract_count(
 193             r'<span class="count">([\d,\.]+)</span> views', webpage, 'view')
 194         like_count = self._extract_count(
 195             r'<span class="votesUp">([\d,\.]+)</span>', webpage, 'like')
 196         dislike_count = self._extract_count(
 197             r'<span class="votesDown">([\d,\.]+)</span>', webpage, 'dislike')
 198         comment_count = self._extract_count(
 199             r'All Comments\s*<span>\(([\d,.]+)\)', webpage, 'comment')
 200
 201         page_params = self._parse_json(self._search_regex(
 202             r'page_params\.zoneDetails\[([\'"])[^\'"]+\1\]\s*=\s*(?P<data>{[^}]+})',
 203             webpage, 'page parameters', group='data', default='{}'),
 204             video_id, transform_source=js_to_json, fatal=False)
 205         tags = categories = None
 206         if page_params:
 207             tags = page_params.get('tags', '').split(',')
 208             categories = page_params.get('categories', '').split(',')
 209
 210         return {
 211             'id': video_id,
 212             'url': video_url,
 213             'uploader': video_uploader,
 214             'title': title,
 215             'thumbnail': thumbnail,
 216             'duration': duration,
 217             'view_count': view_count,
 218             'like_count': like_count,
 219             'dislike_count': dislike_count,
 220             'comment_count': comment_count,
 221             # 'formats': formats,
 222             'age_limit': 18,
 223             'tags': tags,
 224             'categories': categories,
 225         }
 226
 227
 228 class PornHubPlaylistBaseIE(InfoExtractor):
 229     def _extract_entries(self, webpage):
 230         # Only process container div with main playlist content skipping
 231         # drop-down menu that uses similar pattern for videos (see
 232         # https://github.com/rg3/youtube-dl/issues/11594).
 233         container = self._search_regex(
 234             r'(?s)(<div[^>]+class=["\']container.+)', webpage,
 235             'container', default=webpage)
 236
 237         return [
 238             self.url_result(
 239                 'http://www.pornhub.com/%s' % video_url,
 240                 PornHubIE.ie_key(), video_title=title)
 241             for video_url, title in orderedSet(re.findall(
 242                 r'href="/?(view_video\.php\?.*\bviewkey=[\da-z]+[^"]*)"[^>]*\s+title="([^"]+)"',
 243                 container))
 244         ]
 245
 246     def _real_extract(self, url):
 247         playlist_id = self._match_id(url)
 248
 249         webpage = self._download_webpage(url, playlist_id)
 250
 251         entries = self._extract_entries(webpage)
 252
 253         playlist = self._parse_json(
 254             self._search_regex(
 255                 r'(?:playlistObject|PLAYLIST_VIEW)\s*=\s*({.+?});', webpage,
 256                 'playlist', default='{}'),
 257             playlist_id, fatal=False)
 258         title = playlist.get('title') or self._search_regex(
 259             r'>Videos\s+in\s+(.+?)\s+[Pp]laylist<', webpage, 'title', fatal=False)
 260
 261         return self.playlist_result(
 262             entries, playlist_id, title, playlist.get('description'))
 263
 264
 265 class PornHubPlaylistIE(PornHubPlaylistBaseIE):
 266     _VALID_URL = r'https?://(?:www\.)?pornhub\.com/playlist/(?P<id>\d+)'
 267     _TESTS = [{
 268         'url': 'http://www.pornhub.com/playlist/4667351',
 269         'info_dict': {
 270             'id': '4667351',
 271             'title': 'Nataly Hot',
 272         },
 273         'playlist_mincount': 2,
 274     }]
 275
 276
 277 class PornHubUserVideosIE(PornHubPlaylistBaseIE):
 278     _VALID_URL = r'https?://(?:www\.)?pornhub\.com/users/(?P<id>[^/]+)/videos'
 279     _TESTS = [{
 280         'url': 'http://www.pornhub.com/users/zoe_ph/videos/public',
 281         'info_dict': {
 282             'id': 'zoe_ph',
 283         },
 284         'playlist_mincount': 171,
 285     }, {
 286         'url': 'http://www.pornhub.com/users/rushandlia/videos',
 287         'only_matching': True,
 288     }]
 289
 290     def _real_extract(self, url):
 291         user_id = self._match_id(url)
 292
 293         entries = []
 294         for page_num in itertools.count(1):
 295             try:
 296                 webpage = self._download_webpage(
 297                     url, user_id, 'Downloading page %d' % page_num,
 298                     query={'page': page_num})
 299             except ExtractorError as e:
 300                 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 404:
 301                     break
 302                 raise
 303             page_entries = self._extract_entries(webpage)
 304             if not page_entries:
 305                 break
 306             entries.extend(page_entries)
 307
 308         return self.playlist_result(entries, user_id)