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\?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
 102     @staticmethod
 103     def _extract_urls(webpage):
 104         return re.findall(
 105             r'<iframe[^>]+?src=["\'](?P<url>(?:https?:)?//(?:www\.)?pornhub\.com/embed/[\da-z]+)',
 106             webpage)
 107
 108     def _extract_count(self, pattern, webpage, name):
 109         return str_to_int(self._search_regex(
 110             pattern, webpage, '%s count' % name, fatal=False))
 111
 112     def _real_extract(self, url):
 113         video_id = self._match_id(url)
 114
 115         def dl_webpage(platform):
 116             return self._download_webpage(
 117                 'http://www.pornhub.com/view_video.php?viewkey=%s' % video_id,
 118                 video_id, headers={
 119                     'Cookie': 'age_verified=1; platform=%s' % platform,
 120                 })
 121
 122         webpage = dl_webpage('pc')
 123
 124         error_msg = self._html_search_regex(
 125             r'(?s)<div[^>]+class=(["\'])(?:(?!\1).)*\b(?:removed|userMessageSection)\b(?:(?!\1).)*\1[^>]*>(?P<error>.+?)</div>',
 126             webpage, 'error message', default=None, group='error')
 127         if error_msg:
 128             error_msg = re.sub(r'\s+', ' ', error_msg)
 129             raise ExtractorError(
 130                 'PornHub said: %s' % error_msg,
 131                 expected=True, video_id=video_id)
 132
 133         tv_webpage = dl_webpage('tv')
 134
 135         assignments = self._search_regex(
 136             r'(var.+?mediastring.+?)</script>', tv_webpage,
 137             'encoded url').split(';')
 138
 139         js_vars = {}
 140
 141         def parse_js_value(inp):
 142             inp = re.sub(r'/\*(?:(?!\*/).)*?\*/', '', inp)
 143             if '+' in inp:
 144                 inps = inp.split('+')
 145                 return functools.reduce(
 146                     operator.concat, map(parse_js_value, inps))
 147             inp = inp.strip()
 148             if inp in js_vars:
 149                 return js_vars[inp]
 150             return remove_quotes(inp)
 151
 152         for assn in assignments:
 153             assn = assn.strip()
 154             if not assn:
 155                 continue
 156             assn = re.sub(r'var\s+', '', assn)
 157             vname, value = assn.split('=', 1)
 158             js_vars[vname] = parse_js_value(value)
 159
 160         video_url = js_vars['mediastring']
 161
 162         title = self._search_regex(
 163             r'<h1>([^>]+)</h1>', tv_webpage, 'title', default=None)
 164
 165         # video_title from flashvars contains whitespace instead of non-ASCII (see
 166         # http://www.pornhub.com/view_video.php?viewkey=1331683002), not relying
 167         # on that anymore.
 168         title = title or self._html_search_meta(
 169             'twitter:title', webpage, default=None) or self._search_regex(
 170             (r'<h1[^>]+class=["\']title["\'][^>]*>(?P<title>[^<]+)',
 171              r'<div[^>]+data-video-title=(["\'])(?P<title>.+?)\1',
 172              r'shareTitle\s*=\s*(["\'])(?P<title>.+?)\1'),
 173             webpage, 'title', group='title')
 174
 175         flashvars = self._parse_json(
 176             self._search_regex(
 177                 r'var\s+flashvars_\d+\s*=\s*({.+?});', webpage, 'flashvars', default='{}'),
 178             video_id)
 179         if flashvars:
 180             thumbnail = flashvars.get('image_url')
 181             duration = int_or_none(flashvars.get('video_duration'))
 182         else:
 183             title, thumbnail, duration = [None] * 3
 184
 185         video_uploader = self._html_search_regex(
 186             r'(?s)From:&nbsp;.+?<(?:a href="/users/|a href="/channels/|span class="username)[^>]+>(.+?)<',
 187             webpage, 'uploader', fatal=False)
 188
 189         view_count = self._extract_count(
 190             r'<span class="count">([\d,\.]+)</span> views', webpage, 'view')
 191         like_count = self._extract_count(
 192             r'<span class="votesUp">([\d,\.]+)</span>', webpage, 'like')
 193         dislike_count = self._extract_count(
 194             r'<span class="votesDown">([\d,\.]+)</span>', webpage, 'dislike')
 195         comment_count = self._extract_count(
 196             r'All Comments\s*<span>\(([\d,.]+)\)', webpage, 'comment')
 197
 198         page_params = self._parse_json(self._search_regex(
 199             r'page_params\.zoneDetails\[([\'"])[^\'"]+\1\]\s*=\s*(?P<data>{[^}]+})',
 200             webpage, 'page parameters', group='data', default='{}'),
 201             video_id, transform_source=js_to_json, fatal=False)
 202         tags = categories = None
 203         if page_params:
 204             tags = page_params.get('tags', '').split(',')
 205             categories = page_params.get('categories', '').split(',')
 206
 207         return {
 208             'id': video_id,
 209             'url': video_url,
 210             'uploader': video_uploader,
 211             'title': title,
 212             'thumbnail': thumbnail,
 213             'duration': duration,
 214             'view_count': view_count,
 215             'like_count': like_count,
 216             'dislike_count': dislike_count,
 217             'comment_count': comment_count,
 218             # 'formats': formats,
 219             'age_limit': 18,
 220             'tags': tags,
 221             'categories': categories,
 222         }
 223
 224
 225 class PornHubPlaylistBaseIE(InfoExtractor):
 226     def _extract_entries(self, webpage):
 227         return [
 228             self.url_result(
 229                 'http://www.pornhub.com/%s' % video_url,
 230                 PornHubIE.ie_key(), video_title=title)
 231             for video_url, title in orderedSet(re.findall(
 232                 r'href="/?(view_video\.php\?.*\bviewkey=[\da-z]+[^"]*)"[^>]*\s+title="([^"]+)"',
 233                 webpage))
 234         ]
 235
 236     def _real_extract(self, url):
 237         playlist_id = self._match_id(url)
 238
 239         webpage = self._download_webpage(url, playlist_id)
 240
 241         # Only process container div with main playlist content skipping
 242         # drop-down menu that uses similar pattern for videos (see
 243         # https://github.com/rg3/youtube-dl/issues/11594).
 244         container = self._search_regex(
 245             r'(?s)(<div[^>]+class=["\']container.+)', webpage,
 246             'container', default=webpage)
 247
 248         entries = self._extract_entries(container)
 249
 250         playlist = self._parse_json(
 251             self._search_regex(
 252                 r'playlistObject\s*=\s*({.+?});', webpage, 'playlist'),
 253             playlist_id)
 254
 255         return self.playlist_result(
 256             entries, playlist_id, playlist.get('title'), playlist.get('description'))
 257
 258
 259 class PornHubPlaylistIE(PornHubPlaylistBaseIE):
 260     _VALID_URL = r'https?://(?:www\.)?pornhub\.com/playlist/(?P<id>\d+)'
 261     _TESTS = [{
 262         'url': 'http://www.pornhub.com/playlist/4667351',
 263         'info_dict': {
 264             'id': '4667351',
 265             'title': 'Nataly Hot',
 266         },
 267         'playlist_mincount': 2,
 268     }]
 269
 270
 271 class PornHubUserVideosIE(PornHubPlaylistBaseIE):
 272     _VALID_URL = r'https?://(?:www\.)?pornhub\.com/users/(?P<id>[^/]+)/videos'
 273     _TESTS = [{
 274         'url': 'http://www.pornhub.com/users/zoe_ph/videos/public',
 275         'info_dict': {
 276             'id': 'zoe_ph',
 277         },
 278         'playlist_mincount': 171,
 279     }, {
 280         'url': 'http://www.pornhub.com/users/rushandlia/videos',
 281         'only_matching': True,
 282     }]
 283
 284     def _real_extract(self, url):
 285         user_id = self._match_id(url)
 286
 287         entries = []
 288         for page_num in itertools.count(1):
 289             try:
 290                 webpage = self._download_webpage(
 291                     url, user_id, 'Downloading page %d' % page_num,
 292                     query={'page': page_num})
 293             except ExtractorError as e:
 294                 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 404:
 295                     break
 296             page_entries = self._extract_entries(webpage)
 297             if not page_entries:
 298                 break
 299             entries.extend(page_entries)
 300
 301         return self.playlist_result(entries, user_id)