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     determine_ext,
  18     ExtractorError,
  19     int_or_none,
  20     orderedSet,
  21     remove_quotes,
  22     str_to_int,
  23     url_or_none,
  24 )
  25
  26
  27 class PornHubBaseIE(InfoExtractor):
  28     def _download_webpage_handle(self, *args, **kwargs):
  29         def dl(*args, **kwargs):
  30             return super(PornHubBaseIE, self)._download_webpage_handle(*args, **kwargs)
  31
  32         webpage, urlh = dl(*args, **kwargs)
  33
  34         if any(re.search(p, webpage) for p in (
  35                 r'<body\b[^>]+\bonload=["\']go\(\)',
  36                 r'document\.cookie\s*=\s*["\']RNKEY=',
  37                 r'document\.location\.reload\(true\)')):
  38             url_or_request = args[0]
  39             url = (url_or_request.get_full_url()
  40                    if isinstance(url_or_request, compat_urllib_request.Request)
  41                    else url_or_request)
  42             phantom = PhantomJSwrapper(self, required_version='2.0')
  43             phantom.get(url, html=webpage)
  44             webpage, urlh = dl(*args, **kwargs)
  45
  46         return webpage, urlh
  47
  48
  49 class PornHubIE(PornHubBaseIE):
  50     IE_DESC = 'PornHub and Thumbzilla'
  51     _VALID_URL = r'''(?x)
  52                     https?://
  53                         (?:
  54                             (?:[^/]+\.)?(?P<host>pornhub\.(?:com|net))/(?:(?:view_video\.php|video/show)\?viewkey=|embed/)|
  55                             (?:www\.)?thumbzilla\.com/video/
  56                         )
  57                         (?P<id>[\da-z]+)
  58                     '''
  59     _TESTS = [{
  60         'url': 'http://www.pornhub.com/view_video.php?viewkey=648719015',
  61         'md5': '1e19b41231a02eba417839222ac9d58e',
  62         'info_dict': {
  63             'id': '648719015',
  64             'ext': 'mp4',
  65             'title': 'Seductive Indian beauty strips down and fingers her pink pussy',
  66             'uploader': 'Babes',
  67             'upload_date': '20130628',
  68             'duration': 361,
  69             'view_count': int,
  70             'like_count': int,
  71             'dislike_count': int,
  72             'comment_count': int,
  73             'age_limit': 18,
  74             'tags': list,
  75             'categories': list,
  76         },
  77     }, {
  78         # non-ASCII title
  79         'url': 'http://www.pornhub.com/view_video.php?viewkey=1331683002',
  80         'info_dict': {
  81             'id': '1331683002',
  82             'ext': 'mp4',
  83             'title': '重庆婷婷女王足交',
  84             'uploader': 'Unknown',
  85             'upload_date': '20150213',
  86             'duration': 1753,
  87             'view_count': int,
  88             'like_count': int,
  89             'dislike_count': int,
  90             'comment_count': int,
  91             'age_limit': 18,
  92             'tags': list,
  93             'categories': list,
  94         },
  95         'params': {
  96             'skip_download': True,
  97         },
  98     }, {
  99         # subtitles
 100         'url': 'https://www.pornhub.com/view_video.php?viewkey=ph5af5fef7c2aa7',
 101         'info_dict': {
 102             'id': 'ph5af5fef7c2aa7',
 103             'ext': 'mp4',
 104             'title': 'BFFS - Cute Teen Girls Share Cock On the Floor',
 105             'uploader': 'BFFs',
 106             'duration': 622,
 107             'view_count': int,
 108             'like_count': int,
 109             'dislike_count': int,
 110             'comment_count': int,
 111             'age_limit': 18,
 112             'tags': list,
 113             'categories': list,
 114             'subtitles': {
 115                 'en': [{
 116                     "ext": 'srt'
 117                 }]
 118             },
 119         },
 120         'params': {
 121             'skip_download': True,
 122         },
 123     }, {
 124         'url': 'http://www.pornhub.com/view_video.php?viewkey=ph557bbb6676d2d',
 125         'only_matching': True,
 126     }, {
 127         # removed at the request of cam4.com
 128         'url': 'http://fr.pornhub.com/view_video.php?viewkey=ph55ca2f9760862',
 129         'only_matching': True,
 130     }, {
 131         # removed at the request of the copyright owner
 132         'url': 'http://www.pornhub.com/view_video.php?viewkey=788152859',
 133         'only_matching': True,
 134     }, {
 135         # removed by uploader
 136         'url': 'http://www.pornhub.com/view_video.php?viewkey=ph572716d15a111',
 137         'only_matching': True,
 138     }, {
 139         # private video
 140         'url': 'http://www.pornhub.com/view_video.php?viewkey=ph56fd731fce6b7',
 141         'only_matching': True,
 142     }, {
 143         'url': 'https://www.thumbzilla.com/video/ph56c6114abd99a/horny-girlfriend-sex',
 144         'only_matching': True,
 145     }, {
 146         'url': 'http://www.pornhub.com/video/show?viewkey=648719015',
 147         'only_matching': True,
 148     }, {
 149         'url': 'https://www.pornhub.net/view_video.php?viewkey=203640933',
 150         'only_matching': True,
 151     }]
 152
 153     @staticmethod
 154     def _extract_urls(webpage):
 155         return re.findall(
 156             r'<iframe[^>]+?src=["\'](?P<url>(?:https?:)?//(?:www\.)?pornhub\.(?:com|net)/embed/[\da-z]+)',
 157             webpage)
 158
 159     def _extract_count(self, pattern, webpage, name):
 160         return str_to_int(self._search_regex(
 161             pattern, webpage, '%s count' % name, fatal=False))
 162
 163     def _real_extract(self, url):
 164         mobj = re.match(self._VALID_URL, url)
 165         host = mobj.group('host') or 'pornhub.com'
 166         video_id = mobj.group('id')
 167
 168         self._set_cookie(host, 'age_verified', '1')
 169
 170         def dl_webpage(platform):
 171             self._set_cookie(host, 'platform', platform)
 172             return self._download_webpage(
 173                 'http://www.%s/view_video.php?viewkey=%s' % (host, video_id),
 174                 video_id, 'Downloading %s webpage' % platform)
 175
 176         webpage = dl_webpage('pc')
 177
 178         error_msg = self._html_search_regex(
 179             r'(?s)<div[^>]+class=(["\'])(?:(?!\1).)*\b(?:removed|userMessageSection)\b(?:(?!\1).)*\1[^>]*>(?P<error>.+?)</div>',
 180             webpage, 'error message', default=None, group='error')
 181         if error_msg:
 182             error_msg = re.sub(r'\s+', ' ', error_msg)
 183             raise ExtractorError(
 184                 'PornHub said: %s' % error_msg,
 185                 expected=True, video_id=video_id)
 186
 187         # video_title from flashvars contains whitespace instead of non-ASCII (see
 188         # http://www.pornhub.com/view_video.php?viewkey=1331683002), not relying
 189         # on that anymore.
 190         title = self._html_search_meta(
 191             'twitter:title', webpage, default=None) or self._search_regex(
 192             (r'<h1[^>]+class=["\']title["\'][^>]*>(?P<title>[^<]+)',
 193              r'<div[^>]+data-video-title=(["\'])(?P<title>.+?)\1',
 194              r'shareTitle\s*=\s*(["\'])(?P<title>.+?)\1'),
 195             webpage, 'title', group='title')
 196
 197         video_urls = []
 198         video_urls_set = set()
 199         subtitles = {}
 200
 201         flashvars = self._parse_json(
 202             self._search_regex(
 203                 r'var\s+flashvars_\d+\s*=\s*({.+?});', webpage, 'flashvars', default='{}'),
 204             video_id)
 205         if flashvars:
 206             subtitle_url = url_or_none(flashvars.get('closedCaptionsFile'))
 207             if subtitle_url:
 208                 subtitles.setdefault('en', []).append({
 209                     'url': subtitle_url,
 210                     'ext': 'srt',
 211                 })
 212             thumbnail = flashvars.get('image_url')
 213             duration = int_or_none(flashvars.get('video_duration'))
 214             media_definitions = flashvars.get('mediaDefinitions')
 215             if isinstance(media_definitions, list):
 216                 for definition in media_definitions:
 217                     if not isinstance(definition, dict):
 218                         continue
 219                     video_url = definition.get('videoUrl')
 220                     if not video_url or not isinstance(video_url, compat_str):
 221                         continue
 222                     if video_url in video_urls_set:
 223                         continue
 224                     video_urls_set.add(video_url)
 225                     video_urls.append(
 226                         (video_url, int_or_none(definition.get('quality'))))
 227         else:
 228             thumbnail, duration = [None] * 2
 229
 230         if not video_urls:
 231             tv_webpage = dl_webpage('tv')
 232
 233             assignments = self._search_regex(
 234                 r'(var.+?mediastring.+?)</script>', tv_webpage,
 235                 'encoded url').split(';')
 236
 237             js_vars = {}
 238
 239             def parse_js_value(inp):
 240                 inp = re.sub(r'/\*(?:(?!\*/).)*?\*/', '', inp)
 241                 if '+' in inp:
 242                     inps = inp.split('+')
 243                     return functools.reduce(
 244                         operator.concat, map(parse_js_value, inps))
 245                 inp = inp.strip()
 246                 if inp in js_vars:
 247                     return js_vars[inp]
 248                 return remove_quotes(inp)
 249
 250             for assn in assignments:
 251                 assn = assn.strip()
 252                 if not assn:
 253                     continue
 254                 assn = re.sub(r'var\s+', '', assn)
 255                 vname, value = assn.split('=', 1)
 256                 js_vars[vname] = parse_js_value(value)
 257
 258             video_url = js_vars['mediastring']
 259             if video_url not in video_urls_set:
 260                 video_urls.append((video_url, None))
 261                 video_urls_set.add(video_url)
 262
 263         for mobj in re.finditer(
 264                 r'<a[^>]+\bclass=["\']downloadBtn\b[^>]+\bhref=(["\'])(?P<url>(?:(?!\1).)+)\1',
 265                 webpage):
 266             video_url = mobj.group('url')
 267             if video_url not in video_urls_set:
 268                 video_urls.append((video_url, None))
 269                 video_urls_set.add(video_url)
 270
 271         upload_date = None
 272         formats = []
 273         for video_url, height in video_urls:
 274             if not upload_date:
 275                 upload_date = self._search_regex(
 276                     r'/(\d{6}/\d{2})/', video_url, 'upload data', default=None)
 277                 if upload_date:
 278                     upload_date = upload_date.replace('/', '')
 279             if determine_ext(video_url) == 'mpd':
 280                 formats.extend(self._extract_mpd_formats(
 281                     video_url, video_id, mpd_id='dash', fatal=False))
 282                 continue
 283             tbr = None
 284             mobj = re.search(r'(?P<height>\d+)[pP]?_(?P<tbr>\d+)[kK]', video_url)
 285             if mobj:
 286                 if not height:
 287                     height = int(mobj.group('height'))
 288                 tbr = int(mobj.group('tbr'))
 289             formats.append({
 290                 'url': video_url,
 291                 'format_id': '%dp' % height if height else None,
 292                 'height': height,
 293                 'tbr': tbr,
 294             })
 295         self._sort_formats(formats)
 296
 297         video_uploader = self._html_search_regex(
 298             r'(?s)From:&nbsp;.+?<(?:a\b[^>]+\bhref=["\']/(?:(?:user|channel)s|model|pornstar)/|span\b[^>]+\bclass=["\']username)[^>]+>(.+?)<',
 299             webpage, 'uploader', fatal=False)
 300
 301         view_count = self._extract_count(
 302             r'<span class="count">([\d,\.]+)</span> views', webpage, 'view')
 303         like_count = self._extract_count(
 304             r'<span class="votesUp">([\d,\.]+)</span>', webpage, 'like')
 305         dislike_count = self._extract_count(
 306             r'<span class="votesDown">([\d,\.]+)</span>', webpage, 'dislike')
 307         comment_count = self._extract_count(
 308             r'All Comments\s*<span>\(([\d,.]+)\)', webpage, 'comment')
 309
 310         def extract_list(meta_key):
 311             div = self._search_regex(
 312                 r'(?s)<div[^>]+\bclass=["\'].*?\b%sWrapper[^>]*>(.+?)</div>'
 313                 % meta_key, webpage, meta_key, default=None)
 314             if div:
 315                 return re.findall(r'<a[^>]+\bhref=[^>]+>([^<]+)', div)
 316
 317         return {
 318             'id': video_id,
 319             'uploader': video_uploader,
 320             'upload_date': upload_date,
 321             'title': title,
 322             'thumbnail': thumbnail,
 323             'duration': duration,
 324             'view_count': view_count,
 325             'like_count': like_count,
 326             'dislike_count': dislike_count,
 327             'comment_count': comment_count,
 328             'formats': formats,
 329             'age_limit': 18,
 330             'tags': extract_list('tags'),
 331             'categories': extract_list('categories'),
 332             'subtitles': subtitles,
 333         }
 334
 335
 336 class PornHubPlaylistBaseIE(PornHubBaseIE):
 337     def _extract_entries(self, webpage, host):
 338         # Only process container div with main playlist content skipping
 339         # drop-down menu that uses similar pattern for videos (see
 340         # https://github.com/ytdl-org/youtube-dl/issues/11594).
 341         container = self._search_regex(
 342             r'(?s)(<div[^>]+class=["\']container.+)', webpage,
 343             'container', default=webpage)
 344
 345         return [
 346             self.url_result(
 347                 'http://www.%s/%s' % (host, video_url),
 348                 PornHubIE.ie_key(), video_title=title)
 349             for video_url, title in orderedSet(re.findall(
 350                 r'href="/?(view_video\.php\?.*\bviewkey=[\da-z]+[^"]*)"[^>]*\s+title="([^"]+)"',
 351                 container))
 352         ]
 353
 354     def _real_extract(self, url):
 355         mobj = re.match(self._VALID_URL, url)
 356         host = mobj.group('host')
 357         playlist_id = mobj.group('id')
 358
 359         webpage = self._download_webpage(url, playlist_id)
 360
 361         entries = self._extract_entries(webpage, host)
 362
 363         playlist = self._parse_json(
 364             self._search_regex(
 365                 r'(?:playlistObject|PLAYLIST_VIEW)\s*=\s*({.+?});', webpage,
 366                 'playlist', default='{}'),
 367             playlist_id, fatal=False)
 368         title = playlist.get('title') or self._search_regex(
 369             r'>Videos\s+in\s+(.+?)\s+[Pp]laylist<', webpage, 'title', fatal=False)
 370
 371         return self.playlist_result(
 372             entries, playlist_id, title, playlist.get('description'))
 373
 374
 375 class PornHubPlaylistIE(PornHubPlaylistBaseIE):
 376     _VALID_URL = r'https?://(?:[^/]+\.)?(?P<host>pornhub\.(?:com|net))/playlist/(?P<id>\d+)'
 377     _TESTS = [{
 378         'url': 'http://www.pornhub.com/playlist/4667351',
 379         'info_dict': {
 380             'id': '4667351',
 381             'title': 'Nataly Hot',
 382         },
 383         'playlist_mincount': 2,
 384     }, {
 385         'url': 'https://de.pornhub.com/playlist/4667351',
 386         'only_matching': True,
 387     }]
 388
 389
 390 class PornHubUserVideosIE(PornHubPlaylistBaseIE):
 391     _VALID_URL = r'https?://(?:[^/]+\.)?(?P<host>pornhub\.(?:com|net))/(?:(?:user|channel)s|model|pornstar)/(?P<id>[^/]+)/videos'
 392     _TESTS = [{
 393         'url': 'http://www.pornhub.com/users/zoe_ph/videos/public',
 394         'info_dict': {
 395             'id': 'zoe_ph',
 396         },
 397         'playlist_mincount': 171,
 398     }, {
 399         'url': 'http://www.pornhub.com/users/rushandlia/videos',
 400         'only_matching': True,
 401     }, {
 402         # default sorting as Top Rated Videos
 403         'url': 'https://www.pornhub.com/channels/povd/videos',
 404         'info_dict': {
 405             'id': 'povd',
 406         },
 407         'playlist_mincount': 293,
 408     }, {
 409         # Top Rated Videos
 410         'url': 'https://www.pornhub.com/channels/povd/videos?o=ra',
 411         'only_matching': True,
 412     }, {
 413         # Most Recent Videos
 414         'url': 'https://www.pornhub.com/channels/povd/videos?o=da',
 415         'only_matching': True,
 416     }, {
 417         # Most Viewed Videos
 418         'url': 'https://www.pornhub.com/channels/povd/videos?o=vi',
 419         'only_matching': True,
 420     }, {
 421         'url': 'http://www.pornhub.com/users/zoe_ph/videos/public',
 422         'only_matching': True,
 423     }, {
 424         'url': 'https://www.pornhub.com/model/jayndrea/videos/upload',
 425         'only_matching': True,
 426     }, {
 427         'url': 'https://www.pornhub.com/pornstar/jenny-blighe/videos/upload',
 428         'only_matching': True,
 429     }]
 430
 431     def _real_extract(self, url):
 432         mobj = re.match(self._VALID_URL, url)
 433         host = mobj.group('host')
 434         user_id = mobj.group('id')
 435
 436         entries = []
 437         for page_num in itertools.count(1):
 438             try:
 439                 webpage = self._download_webpage(
 440                     url, user_id, 'Downloading page %d' % page_num,
 441                     query={'page': page_num})
 442             except ExtractorError as e:
 443                 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 404:
 444                     break
 445                 raise
 446             page_entries = self._extract_entries(webpage, host)
 447             if not page_entries:
 448                 break
 449             entries.extend(page_entries)
 450
 451         return self.playlist_result(entries, user_id)