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 re
   8
   9 from .common import InfoExtractor
  10 from ..compat import (
  11     compat_HTTPError,
  12     compat_str,
  13     compat_urllib_request,
  14 )
  15 from .openload import PhantomJSwrapper
  16 from ..utils import (
  17     ExtractorError,
  18     int_or_none,
  19     orderedSet,
  20     remove_quotes,
  21     str_to_int,
  22     url_or_none,
  23 )
  24
  25
  26 class PornHubBaseIE(InfoExtractor):
  27     def _download_webpage_handle(self, *args, **kwargs):
  28         def dl(*args, **kwargs):
  29             return super(PornHubBaseIE, self)._download_webpage_handle(*args, **kwargs)
  30
  31         webpage, urlh = dl(*args, **kwargs)
  32
  33         if any(re.search(p, webpage) for p in (
  34                 r'<body\b[^>]+\bonload=["\']go\(\)',
  35                 r'document\.cookie\s*=\s*["\']RNKEY=',
  36                 r'document\.location\.reload\(true\)')):
  37             url_or_request = args[0]
  38             url = (url_or_request.get_full_url()
  39                    if isinstance(url_or_request, compat_urllib_request.Request)
  40                    else url_or_request)
  41             phantom = PhantomJSwrapper(self, required_version='2.0')
  42             phantom.get(url, html=webpage)
  43             webpage, urlh = dl(*args, **kwargs)
  44
  45         return webpage, urlh
  46
  47
  48 class PornHubIE(PornHubBaseIE):
  49     IE_DESC = 'PornHub and Thumbzilla'
  50     _VALID_URL = r'''(?x)
  51                     https?://
  52                         (?:
  53                             (?:[^/]+\.)?(?P<host>pornhub\.(?:com|net))/(?:(?:view_video\.php|video/show)\?viewkey=|embed/)|
  54                             (?:www\.)?thumbzilla\.com/video/
  55                         )
  56                         (?P<id>[\da-z]+)
  57                     '''
  58     _TESTS = [{
  59         'url': 'http://www.pornhub.com/view_video.php?viewkey=648719015',
  60         'md5': '1e19b41231a02eba417839222ac9d58e',
  61         'info_dict': {
  62             'id': '648719015',
  63             'ext': 'mp4',
  64             'title': 'Seductive Indian beauty strips down and fingers her pink pussy',
  65             'uploader': 'Babes',
  66             'upload_date': '20130628',
  67             'duration': 361,
  68             'view_count': int,
  69             'like_count': int,
  70             'dislike_count': int,
  71             'comment_count': int,
  72             'age_limit': 18,
  73             'tags': list,
  74             'categories': list,
  75         },
  76     }, {
  77         # non-ASCII title
  78         'url': 'http://www.pornhub.com/view_video.php?viewkey=1331683002',
  79         'info_dict': {
  80             'id': '1331683002',
  81             'ext': 'mp4',
  82             'title': '重庆婷婷女王足交',
  83             'uploader': 'Unknown',
  84             'upload_date': '20150213',
  85             'duration': 1753,
  86             'view_count': int,
  87             'like_count': int,
  88             'dislike_count': int,
  89             'comment_count': int,
  90             'age_limit': 18,
  91             'tags': list,
  92             'categories': list,
  93         },
  94         'params': {
  95             'skip_download': True,
  96         },
  97     }, {
  98         # subtitles
  99         'url': 'https://www.pornhub.com/view_video.php?viewkey=ph5af5fef7c2aa7',
 100         'info_dict': {
 101             'id': 'ph5af5fef7c2aa7',
 102             'ext': 'mp4',
 103             'title': 'BFFS - Cute Teen Girls Share Cock On the Floor',
 104             'uploader': 'BFFs',
 105             'duration': 622,
 106             'view_count': int,
 107             'like_count': int,
 108             'dislike_count': int,
 109             'comment_count': int,
 110             'age_limit': 18,
 111             'tags': list,
 112             'categories': list,
 113             'subtitles': {
 114                 'en': [{
 115                     "ext": 'srt'
 116                 }]
 117             },
 118         },
 119         'params': {
 120             'skip_download': True,
 121         },
 122     }, {
 123         'url': 'http://www.pornhub.com/view_video.php?viewkey=ph557bbb6676d2d',
 124         'only_matching': True,
 125     }, {
 126         # removed at the request of cam4.com
 127         'url': 'http://fr.pornhub.com/view_video.php?viewkey=ph55ca2f9760862',
 128         'only_matching': True,
 129     }, {
 130         # removed at the request of the copyright owner
 131         'url': 'http://www.pornhub.com/view_video.php?viewkey=788152859',
 132         'only_matching': True,
 133     }, {
 134         # removed by uploader
 135         'url': 'http://www.pornhub.com/view_video.php?viewkey=ph572716d15a111',
 136         'only_matching': True,
 137     }, {
 138         # private video
 139         'url': 'http://www.pornhub.com/view_video.php?viewkey=ph56fd731fce6b7',
 140         'only_matching': True,
 141     }, {
 142         'url': 'https://www.thumbzilla.com/video/ph56c6114abd99a/horny-girlfriend-sex',
 143         'only_matching': True,
 144     }, {
 145         'url': 'http://www.pornhub.com/video/show?viewkey=648719015',
 146         'only_matching': True,
 147     }, {
 148         'url': 'https://www.pornhub.net/view_video.php?viewkey=203640933',
 149         'only_matching': True,
 150     }]
 151
 152     @staticmethod
 153     def _extract_urls(webpage):
 154         return re.findall(
 155             r'<iframe[^>]+?src=["\'](?P<url>(?:https?:)?//(?:www\.)?pornhub\.(?:com|net)/embed/[\da-z]+)',
 156             webpage)
 157
 158     def _extract_count(self, pattern, webpage, name):
 159         return str_to_int(self._search_regex(
 160             pattern, webpage, '%s count' % name, fatal=False))
 161
 162     def _real_extract(self, url):
 163         mobj = re.match(self._VALID_URL, url)
 164         host = mobj.group('host') or 'pornhub.com'
 165         video_id = mobj.group('id')
 166
 167         self._set_cookie(host, 'age_verified', '1')
 168
 169         def dl_webpage(platform):
 170             self._set_cookie(host, 'platform', platform)
 171             return self._download_webpage(
 172                 'http://www.%s/view_video.php?viewkey=%s' % (host, video_id),
 173                 video_id, 'Downloading %s webpage' % platform)
 174
 175         webpage = dl_webpage('pc')
 176
 177         error_msg = self._html_search_regex(
 178             r'(?s)<div[^>]+class=(["\'])(?:(?!\1).)*\b(?:removed|userMessageSection)\b(?:(?!\1).)*\1[^>]*>(?P<error>.+?)</div>',
 179             webpage, 'error message', default=None, group='error')
 180         if error_msg:
 181             error_msg = re.sub(r'\s+', ' ', error_msg)
 182             raise ExtractorError(
 183                 'PornHub said: %s' % error_msg,
 184                 expected=True, video_id=video_id)
 185
 186         # video_title from flashvars contains whitespace instead of non-ASCII (see
 187         # http://www.pornhub.com/view_video.php?viewkey=1331683002), not relying
 188         # on that anymore.
 189         title = self._html_search_meta(
 190             'twitter:title', webpage, default=None) or self._search_regex(
 191             (r'<h1[^>]+class=["\']title["\'][^>]*>(?P<title>[^<]+)',
 192              r'<div[^>]+data-video-title=(["\'])(?P<title>.+?)\1',
 193              r'shareTitle\s*=\s*(["\'])(?P<title>.+?)\1'),
 194             webpage, 'title', group='title')
 195
 196         video_urls = []
 197         video_urls_set = set()
 198         subtitles = {}
 199
 200         flashvars = self._parse_json(
 201             self._search_regex(
 202                 r'var\s+flashvars_\d+\s*=\s*({.+?});', webpage, 'flashvars', default='{}'),
 203             video_id)
 204         if flashvars:
 205             subtitle_url = url_or_none(flashvars.get('closedCaptionsFile'))
 206             if subtitle_url:
 207                 subtitles.setdefault('en', []).append({
 208                     'url': subtitle_url,
 209                     'ext': 'srt',
 210                 })
 211             thumbnail = flashvars.get('image_url')
 212             duration = int_or_none(flashvars.get('video_duration'))
 213             media_definitions = flashvars.get('mediaDefinitions')
 214             if isinstance(media_definitions, list):
 215                 for definition in media_definitions:
 216                     if not isinstance(definition, dict):
 217                         continue
 218                     video_url = definition.get('videoUrl')
 219                     if not video_url or not isinstance(video_url, compat_str):
 220                         continue
 221                     if video_url in video_urls_set:
 222                         continue
 223                     video_urls_set.add(video_url)
 224                     video_urls.append(
 225                         (video_url, int_or_none(definition.get('quality'))))
 226         else:
 227             thumbnail, duration = [None] * 2
 228
 229         if not video_urls:
 230             tv_webpage = dl_webpage('tv')
 231
 232             assignments = self._search_regex(
 233                 r'(var.+?mediastring.+?)</script>', tv_webpage,
 234                 'encoded url').split(';')
 235
 236             js_vars = {}
 237
 238             def parse_js_value(inp):
 239                 inp = re.sub(r'/\*(?:(?!\*/).)*?\*/', '', inp)
 240                 if '+' in inp:
 241                     inps = inp.split('+')
 242                     return functools.reduce(
 243                         operator.concat, map(parse_js_value, inps))
 244                 inp = inp.strip()
 245                 if inp in js_vars:
 246                     return js_vars[inp]
 247                 return remove_quotes(inp)
 248
 249             for assn in assignments:
 250                 assn = assn.strip()
 251                 if not assn:
 252                     continue
 253                 assn = re.sub(r'var\s+', '', assn)
 254                 vname, value = assn.split('=', 1)
 255                 js_vars[vname] = parse_js_value(value)
 256
 257             video_url = js_vars['mediastring']
 258             if video_url not in video_urls_set:
 259                 video_urls.append((video_url, None))
 260                 video_urls_set.add(video_url)
 261
 262         for mobj in re.finditer(
 263                 r'<a[^>]+\bclass=["\']downloadBtn\b[^>]+\bhref=(["\'])(?P<url>(?:(?!\1).)+)\1',
 264                 webpage):
 265             video_url = mobj.group('url')
 266             if video_url not in video_urls_set:
 267                 video_urls.append((video_url, None))
 268                 video_urls_set.add(video_url)
 269
 270         upload_date = None
 271         formats = []
 272         for video_url, height in video_urls:
 273             if not upload_date:
 274                 upload_date = self._search_regex(
 275                     r'/(\d{6}/\d{2})/', video_url, 'upload data', default=None)
 276                 if upload_date:
 277                     upload_date = upload_date.replace('/', '')
 278             tbr = None
 279             mobj = re.search(r'(?P<height>\d+)[pP]?_(?P<tbr>\d+)[kK]', video_url)
 280             if mobj:
 281                 if not height:
 282                     height = int(mobj.group('height'))
 283                 tbr = int(mobj.group('tbr'))
 284             formats.append({
 285                 'url': video_url,
 286                 'format_id': '%dp' % height if height else None,
 287                 'height': height,
 288                 'tbr': tbr,
 289             })
 290         self._sort_formats(formats)
 291
 292         video_uploader = self._html_search_regex(
 293             r'(?s)From:&nbsp;.+?<(?:a\b[^>]+\bhref=["\']/(?:(?:user|channel)s|model|pornstar)/|span\b[^>]+\bclass=["\']username)[^>]+>(.+?)<',
 294             webpage, 'uploader', fatal=False)
 295
 296         view_count = self._extract_count(
 297             r'<span class="count">([\d,\.]+)</span> views', webpage, 'view')
 298         like_count = self._extract_count(
 299             r'<span class="votesUp">([\d,\.]+)</span>', webpage, 'like')
 300         dislike_count = self._extract_count(
 301             r'<span class="votesDown">([\d,\.]+)</span>', webpage, 'dislike')
 302         comment_count = self._extract_count(
 303             r'All Comments\s*<span>\(([\d,.]+)\)', webpage, 'comment')
 304
 305         def extract_list(meta_key):
 306             div = self._search_regex(
 307                 r'(?s)<div[^>]+\bclass=["\'].*?\b%sWrapper[^>]*>(.+?)</div>'
 308                 % meta_key, webpage, meta_key, default=None)
 309             if div:
 310                 return re.findall(r'<a[^>]+\bhref=[^>]+>([^<]+)', div)
 311
 312         return {
 313             'id': video_id,
 314             'uploader': video_uploader,
 315             'upload_date': upload_date,
 316             'title': title,
 317             'thumbnail': thumbnail,
 318             'duration': duration,
 319             'view_count': view_count,
 320             'like_count': like_count,
 321             'dislike_count': dislike_count,
 322             'comment_count': comment_count,
 323             'formats': formats,
 324             'age_limit': 18,
 325             'tags': extract_list('tags'),
 326             'categories': extract_list('categories'),
 327             'subtitles': subtitles,
 328         }
 329
 330
 331 class PornHubPlaylistBaseIE(PornHubBaseIE):
 332     def _extract_entries(self, webpage, host):
 333         # Only process container div with main playlist content skipping
 334         # drop-down menu that uses similar pattern for videos (see
 335         # https://github.com/ytdl-org/youtube-dl/issues/11594).
 336         container = self._search_regex(
 337             r'(?s)(<div[^>]+class=["\']container.+)', webpage,
 338             'container', default=webpage)
 339
 340         return [
 341             self.url_result(
 342                 'http://www.%s/%s' % (host, video_url),
 343                 PornHubIE.ie_key(), video_title=title)
 344             for video_url, title in orderedSet(re.findall(
 345                 r'href="/?(view_video\.php\?.*\bviewkey=[\da-z]+[^"]*)"[^>]*\s+title="([^"]+)"',
 346                 container))
 347         ]
 348
 349     def _real_extract(self, url):
 350         mobj = re.match(self._VALID_URL, url)
 351         host = mobj.group('host')
 352         playlist_id = mobj.group('id')
 353
 354         webpage = self._download_webpage(url, playlist_id)
 355
 356         entries = self._extract_entries(webpage, host)
 357
 358         playlist = self._parse_json(
 359             self._search_regex(
 360                 r'(?:playlistObject|PLAYLIST_VIEW)\s*=\s*({.+?});', webpage,
 361                 'playlist', default='{}'),
 362             playlist_id, fatal=False)
 363         title = playlist.get('title') or self._search_regex(
 364             r'>Videos\s+in\s+(.+?)\s+[Pp]laylist<', webpage, 'title', fatal=False)
 365
 366         return self.playlist_result(
 367             entries, playlist_id, title, playlist.get('description'))
 368
 369
 370 class PornHubPlaylistIE(PornHubPlaylistBaseIE):
 371     _VALID_URL = r'https?://(?:[^/]+\.)?(?P<host>pornhub\.(?:com|net))/playlist/(?P<id>\d+)'
 372     _TESTS = [{
 373         'url': 'http://www.pornhub.com/playlist/4667351',
 374         'info_dict': {
 375             'id': '4667351',
 376             'title': 'Nataly Hot',
 377         },
 378         'playlist_mincount': 2,
 379     }, {
 380         'url': 'https://de.pornhub.com/playlist/4667351',
 381         'only_matching': True,
 382     }]
 383
 384
 385 class PornHubUserVideosIE(PornHubPlaylistBaseIE):
 386     _VALID_URL = r'https?://(?:[^/]+\.)?(?P<host>pornhub\.(?:com|net))/(?:(?:user|channel)s|model|pornstar)/(?P<id>[^/]+)/videos'
 387     _TESTS = [{
 388         'url': 'http://www.pornhub.com/users/zoe_ph/videos/public',
 389         'info_dict': {
 390             'id': 'zoe_ph',
 391         },
 392         'playlist_mincount': 171,
 393     }, {
 394         'url': 'http://www.pornhub.com/users/rushandlia/videos',
 395         'only_matching': True,
 396     }, {
 397         # default sorting as Top Rated Videos
 398         'url': 'https://www.pornhub.com/channels/povd/videos',
 399         'info_dict': {
 400             'id': 'povd',
 401         },
 402         'playlist_mincount': 293,
 403     }, {
 404         # Top Rated Videos
 405         'url': 'https://www.pornhub.com/channels/povd/videos?o=ra',
 406         'only_matching': True,
 407     }, {
 408         # Most Recent Videos
 409         'url': 'https://www.pornhub.com/channels/povd/videos?o=da',
 410         'only_matching': True,
 411     }, {
 412         # Most Viewed Videos
 413         'url': 'https://www.pornhub.com/channels/povd/videos?o=vi',
 414         'only_matching': True,
 415     }, {
 416         'url': 'http://www.pornhub.com/users/zoe_ph/videos/public',
 417         'only_matching': True,
 418     }, {
 419         'url': 'https://www.pornhub.com/model/jayndrea/videos/upload',
 420         'only_matching': True,
 421     }, {
 422         'url': 'https://www.pornhub.com/pornstar/jenny-blighe/videos/upload',
 423         'only_matching': True,
 424     }]
 425
 426     def _real_extract(self, url):
 427         mobj = re.match(self._VALID_URL, url)
 428         host = mobj.group('host')
 429         user_id = mobj.group('id')
 430
 431         entries = []
 432         for page_num in itertools.count(1):
 433             try:
 434                 webpage = self._download_webpage(
 435                     url, user_id, 'Downloading page %d' % page_num,
 436                     query={'page': page_num})
 437             except ExtractorError as e:
 438                 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 404:
 439                     break
 440                 raise
 441             page_entries = self._extract_entries(webpage, host)
 442             if not page_entries:
 443                 break
 444             entries.extend(page_entries)
 445
 446         return self.playlist_result(entries, user_id)