]> git.bitcoin.ninja Git - youtube-dl/blob - youtube_dl/extractor/pornhub.py
[extractor/common] Relax interaction count extraction in _json_ld
[youtube-dl] / 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     NO_DEFAULT,
21     orderedSet,
22     remove_quotes,
23     str_to_int,
24     url_or_none,
25 )
26
27
28 class PornHubBaseIE(InfoExtractor):
29     def _download_webpage_handle(self, *args, **kwargs):
30         def dl(*args, **kwargs):
31             return super(PornHubBaseIE, self)._download_webpage_handle(*args, **kwargs)
32
33         webpage, urlh = dl(*args, **kwargs)
34
35         if any(re.search(p, webpage) for p in (
36                 r'<body\b[^>]+\bonload=["\']go\(\)',
37                 r'document\.cookie\s*=\s*["\']RNKEY=',
38                 r'document\.location\.reload\(true\)')):
39             url_or_request = args[0]
40             url = (url_or_request.get_full_url()
41                    if isinstance(url_or_request, compat_urllib_request.Request)
42                    else url_or_request)
43             phantom = PhantomJSwrapper(self, required_version='2.0')
44             phantom.get(url, html=webpage)
45             webpage, urlh = dl(*args, **kwargs)
46
47         return webpage, urlh
48
49
50 class PornHubIE(PornHubBaseIE):
51     IE_DESC = 'PornHub and Thumbzilla'
52     _VALID_URL = r'''(?x)
53                     https?://
54                         (?:
55                             (?:[^/]+\.)?(?P<host>pornhub(?:premium)?\.(?:com|net))/(?:(?:view_video\.php|video/show)\?viewkey=|embed/)|
56                             (?:www\.)?thumbzilla\.com/video/
57                         )
58                         (?P<id>[\da-z]+)
59                     '''
60     _TESTS = [{
61         'url': 'http://www.pornhub.com/view_video.php?viewkey=648719015',
62         'md5': '1e19b41231a02eba417839222ac9d58e',
63         'info_dict': {
64             'id': '648719015',
65             'ext': 'mp4',
66             'title': 'Seductive Indian beauty strips down and fingers her pink pussy',
67             'uploader': 'Babes',
68             'upload_date': '20130628',
69             'duration': 361,
70             'view_count': int,
71             'like_count': int,
72             'dislike_count': int,
73             'comment_count': int,
74             'age_limit': 18,
75             'tags': list,
76             'categories': list,
77         },
78     }, {
79         # non-ASCII title
80         'url': 'http://www.pornhub.com/view_video.php?viewkey=1331683002',
81         'info_dict': {
82             'id': '1331683002',
83             'ext': 'mp4',
84             'title': '重庆婷婷女王足交',
85             'uploader': 'Unknown',
86             'upload_date': '20150213',
87             'duration': 1753,
88             'view_count': int,
89             'like_count': int,
90             'dislike_count': int,
91             'comment_count': int,
92             'age_limit': 18,
93             'tags': list,
94             'categories': list,
95         },
96         'params': {
97             'skip_download': True,
98         },
99     }, {
100         # subtitles
101         'url': 'https://www.pornhub.com/view_video.php?viewkey=ph5af5fef7c2aa7',
102         'info_dict': {
103             'id': 'ph5af5fef7c2aa7',
104             'ext': 'mp4',
105             'title': 'BFFS - Cute Teen Girls Share Cock On the Floor',
106             'uploader': 'BFFs',
107             'duration': 622,
108             'view_count': int,
109             'like_count': int,
110             'dislike_count': int,
111             'comment_count': int,
112             'age_limit': 18,
113             'tags': list,
114             'categories': list,
115             'subtitles': {
116                 'en': [{
117                     "ext": 'srt'
118                 }]
119             },
120         },
121         'params': {
122             'skip_download': True,
123         },
124     }, {
125         'url': 'http://www.pornhub.com/view_video.php?viewkey=ph557bbb6676d2d',
126         'only_matching': True,
127     }, {
128         # removed at the request of cam4.com
129         'url': 'http://fr.pornhub.com/view_video.php?viewkey=ph55ca2f9760862',
130         'only_matching': True,
131     }, {
132         # removed at the request of the copyright owner
133         'url': 'http://www.pornhub.com/view_video.php?viewkey=788152859',
134         'only_matching': True,
135     }, {
136         # removed by uploader
137         'url': 'http://www.pornhub.com/view_video.php?viewkey=ph572716d15a111',
138         'only_matching': True,
139     }, {
140         # private video
141         'url': 'http://www.pornhub.com/view_video.php?viewkey=ph56fd731fce6b7',
142         'only_matching': True,
143     }, {
144         'url': 'https://www.thumbzilla.com/video/ph56c6114abd99a/horny-girlfriend-sex',
145         'only_matching': True,
146     }, {
147         'url': 'http://www.pornhub.com/video/show?viewkey=648719015',
148         'only_matching': True,
149     }, {
150         'url': 'https://www.pornhub.net/view_video.php?viewkey=203640933',
151         'only_matching': True,
152     }, {
153         'url': 'https://www.pornhubpremium.com/view_video.php?viewkey=ph5e4acdae54a82',
154         'only_matching': True,
155     }]
156
157     @staticmethod
158     def _extract_urls(webpage):
159         return re.findall(
160             r'<iframe[^>]+?src=["\'](?P<url>(?:https?:)?//(?:www\.)?pornhub\.(?:com|net)/embed/[\da-z]+)',
161             webpage)
162
163     def _extract_count(self, pattern, webpage, name):
164         return str_to_int(self._search_regex(
165             pattern, webpage, '%s count' % name, fatal=False))
166
167     def _real_extract(self, url):
168         mobj = re.match(self._VALID_URL, url)
169         host = mobj.group('host') or 'pornhub.com'
170         video_id = mobj.group('id')
171
172         if 'premium' in host:
173             if not self._downloader.params.get('cookiefile'):
174                 raise ExtractorError(
175                     'PornHub Premium requires authentication.'
176                     ' You may want to use --cookies.',
177                     expected=True)
178
179         self._set_cookie(host, 'age_verified', '1')
180
181         def dl_webpage(platform):
182             self._set_cookie(host, 'platform', platform)
183             return self._download_webpage(
184                 'https://www.%s/view_video.php?viewkey=%s' % (host, video_id),
185                 video_id, 'Downloading %s webpage' % platform)
186
187         webpage = dl_webpage('pc')
188
189         error_msg = self._html_search_regex(
190             r'(?s)<div[^>]+class=(["\'])(?:(?!\1).)*\b(?:removed|userMessageSection)\b(?:(?!\1).)*\1[^>]*>(?P<error>.+?)</div>',
191             webpage, 'error message', default=None, group='error')
192         if error_msg:
193             error_msg = re.sub(r'\s+', ' ', error_msg)
194             raise ExtractorError(
195                 'PornHub said: %s' % error_msg,
196                 expected=True, video_id=video_id)
197
198         # video_title from flashvars contains whitespace instead of non-ASCII (see
199         # http://www.pornhub.com/view_video.php?viewkey=1331683002), not relying
200         # on that anymore.
201         title = self._html_search_meta(
202             'twitter:title', webpage, default=None) or self._html_search_regex(
203             (r'(?s)<h1[^>]+class=["\']title["\'][^>]*>(?P<title>.+?)</h1>',
204              r'<div[^>]+data-video-title=(["\'])(?P<title>(?:(?!\1).)+)\1',
205              r'shareTitle["\']\s*[=:]\s*(["\'])(?P<title>(?:(?!\1).)+)\1'),
206             webpage, 'title', group='title')
207
208         video_urls = []
209         video_urls_set = set()
210         subtitles = {}
211
212         flashvars = self._parse_json(
213             self._search_regex(
214                 r'var\s+flashvars_\d+\s*=\s*({.+?});', webpage, 'flashvars', default='{}'),
215             video_id)
216         if flashvars:
217             subtitle_url = url_or_none(flashvars.get('closedCaptionsFile'))
218             if subtitle_url:
219                 subtitles.setdefault('en', []).append({
220                     'url': subtitle_url,
221                     'ext': 'srt',
222                 })
223             thumbnail = flashvars.get('image_url')
224             duration = int_or_none(flashvars.get('video_duration'))
225             media_definitions = flashvars.get('mediaDefinitions')
226             if isinstance(media_definitions, list):
227                 for definition in media_definitions:
228                     if not isinstance(definition, dict):
229                         continue
230                     video_url = definition.get('videoUrl')
231                     if not video_url or not isinstance(video_url, compat_str):
232                         continue
233                     if video_url in video_urls_set:
234                         continue
235                     video_urls_set.add(video_url)
236                     video_urls.append(
237                         (video_url, int_or_none(definition.get('quality'))))
238         else:
239             thumbnail, duration = [None] * 2
240
241         def extract_js_vars(webpage, pattern, default=NO_DEFAULT):
242             assignments = self._search_regex(
243                 pattern, webpage, 'encoded url', default=default)
244             if not assignments:
245                 return {}
246
247             assignments = assignments.split(';')
248
249             js_vars = {}
250
251             def parse_js_value(inp):
252                 inp = re.sub(r'/\*(?:(?!\*/).)*?\*/', '', inp)
253                 if '+' in inp:
254                     inps = inp.split('+')
255                     return functools.reduce(
256                         operator.concat, map(parse_js_value, inps))
257                 inp = inp.strip()
258                 if inp in js_vars:
259                     return js_vars[inp]
260                 return remove_quotes(inp)
261
262             for assn in assignments:
263                 assn = assn.strip()
264                 if not assn:
265                     continue
266                 assn = re.sub(r'var\s+', '', assn)
267                 vname, value = assn.split('=', 1)
268                 js_vars[vname] = parse_js_value(value)
269             return js_vars
270
271         def add_video_url(video_url):
272             v_url = url_or_none(video_url)
273             if not v_url:
274                 return
275             if v_url in video_urls_set:
276                 return
277             video_urls.append((v_url, None))
278             video_urls_set.add(v_url)
279
280         if not video_urls:
281             FORMAT_PREFIXES = ('media', 'quality')
282             js_vars = extract_js_vars(
283                 webpage, r'(var\s+(?:%s)_.+)' % '|'.join(FORMAT_PREFIXES),
284                 default=None)
285             if js_vars:
286                 for key, format_url in js_vars.items():
287                     if any(key.startswith(p) for p in FORMAT_PREFIXES):
288                         add_video_url(format_url)
289             if not video_urls and re.search(
290                     r'<[^>]+\bid=["\']lockedPlayer', webpage):
291                 raise ExtractorError(
292                     'Video %s is locked' % video_id, expected=True)
293
294         if not video_urls:
295             js_vars = extract_js_vars(
296                 dl_webpage('tv'), r'(var.+?mediastring.+?)</script>')
297             add_video_url(js_vars['mediastring'])
298
299         for mobj in re.finditer(
300                 r'<a[^>]+\bclass=["\']downloadBtn\b[^>]+\bhref=(["\'])(?P<url>(?:(?!\1).)+)\1',
301                 webpage):
302             video_url = mobj.group('url')
303             if video_url not in video_urls_set:
304                 video_urls.append((video_url, None))
305                 video_urls_set.add(video_url)
306
307         upload_date = None
308         formats = []
309         for video_url, height in video_urls:
310             if not upload_date:
311                 upload_date = self._search_regex(
312                     r'/(\d{6}/\d{2})/', video_url, 'upload data', default=None)
313                 if upload_date:
314                     upload_date = upload_date.replace('/', '')
315             ext = determine_ext(video_url)
316             if ext == 'mpd':
317                 formats.extend(self._extract_mpd_formats(
318                     video_url, video_id, mpd_id='dash', fatal=False))
319                 continue
320             elif ext == 'm3u8':
321                 formats.extend(self._extract_m3u8_formats(
322                     video_url, video_id, 'mp4', entry_protocol='m3u8_native',
323                     m3u8_id='hls', fatal=False))
324                 continue
325             tbr = None
326             mobj = re.search(r'(?P<height>\d+)[pP]?_(?P<tbr>\d+)[kK]', video_url)
327             if mobj:
328                 if not height:
329                     height = int(mobj.group('height'))
330                 tbr = int(mobj.group('tbr'))
331             formats.append({
332                 'url': video_url,
333                 'format_id': '%dp' % height if height else None,
334                 'height': height,
335                 'tbr': tbr,
336             })
337         self._sort_formats(formats)
338
339         video_uploader = self._html_search_regex(
340             r'(?s)From:&nbsp;.+?<(?:a\b[^>]+\bhref=["\']/(?:(?:user|channel)s|model|pornstar)/|span\b[^>]+\bclass=["\']username)[^>]+>(.+?)<',
341             webpage, 'uploader', fatal=False)
342
343         view_count = self._extract_count(
344             r'<span class="count">([\d,\.]+)</span> [Vv]iews', webpage, 'view')
345         like_count = self._extract_count(
346             r'<span class="votesUp">([\d,\.]+)</span>', webpage, 'like')
347         dislike_count = self._extract_count(
348             r'<span class="votesDown">([\d,\.]+)</span>', webpage, 'dislike')
349         comment_count = self._extract_count(
350             r'All Comments\s*<span>\(([\d,.]+)\)', webpage, 'comment')
351
352         def extract_list(meta_key):
353             div = self._search_regex(
354                 r'(?s)<div[^>]+\bclass=["\'].*?\b%sWrapper[^>]*>(.+?)</div>'
355                 % meta_key, webpage, meta_key, default=None)
356             if div:
357                 return re.findall(r'<a[^>]+\bhref=[^>]+>([^<]+)', div)
358
359         return {
360             'id': video_id,
361             'uploader': video_uploader,
362             'upload_date': upload_date,
363             'title': title,
364             'thumbnail': thumbnail,
365             'duration': duration,
366             'view_count': view_count,
367             'like_count': like_count,
368             'dislike_count': dislike_count,
369             'comment_count': comment_count,
370             'formats': formats,
371             'age_limit': 18,
372             'tags': extract_list('tags'),
373             'categories': extract_list('categories'),
374             'subtitles': subtitles,
375         }
376
377
378 class PornHubPlaylistBaseIE(PornHubBaseIE):
379     def _extract_entries(self, webpage, host):
380         # Only process container div with main playlist content skipping
381         # drop-down menu that uses similar pattern for videos (see
382         # https://github.com/ytdl-org/youtube-dl/issues/11594).
383         container = self._search_regex(
384             r'(?s)(<div[^>]+class=["\']container.+)', webpage,
385             'container', default=webpage)
386
387         return [
388             self.url_result(
389                 'http://www.%s/%s' % (host, video_url),
390                 PornHubIE.ie_key(), video_title=title)
391             for video_url, title in orderedSet(re.findall(
392                 r'href="/?(view_video\.php\?.*\bviewkey=[\da-z]+[^"]*)"[^>]*\s+title="([^"]+)"',
393                 container))
394         ]
395
396     def _real_extract(self, url):
397         mobj = re.match(self._VALID_URL, url)
398         host = mobj.group('host')
399         playlist_id = mobj.group('id')
400
401         webpage = self._download_webpage(url, playlist_id)
402
403         entries = self._extract_entries(webpage, host)
404
405         playlist = self._parse_json(
406             self._search_regex(
407                 r'(?:playlistObject|PLAYLIST_VIEW)\s*=\s*({.+?});', webpage,
408                 'playlist', default='{}'),
409             playlist_id, fatal=False)
410         title = playlist.get('title') or self._search_regex(
411             r'>Videos\s+in\s+(.+?)\s+[Pp]laylist<', webpage, 'title', fatal=False)
412
413         return self.playlist_result(
414             entries, playlist_id, title, playlist.get('description'))
415
416
417 class PornHubUserIE(PornHubPlaylistBaseIE):
418     _VALID_URL = r'(?P<url>https?://(?:[^/]+\.)?(?P<host>pornhub(?:premium)?\.(?:com|net))/(?:(?:user|channel)s|model|pornstar)/(?P<id>[^/?#&]+))(?:[?#&]|/(?!videos)|$)'
419     _TESTS = [{
420         'url': 'https://www.pornhub.com/model/zoe_ph',
421         'playlist_mincount': 118,
422     }, {
423         'url': 'https://www.pornhub.com/pornstar/liz-vicious',
424         'info_dict': {
425             'id': 'liz-vicious',
426         },
427         'playlist_mincount': 118,
428     }, {
429         'url': 'https://www.pornhub.com/users/russianveet69',
430         'only_matching': True,
431     }, {
432         'url': 'https://www.pornhub.com/channels/povd',
433         'only_matching': True,
434     }, {
435         'url': 'https://www.pornhub.com/model/zoe_ph?abc=1',
436         'only_matching': True,
437     }]
438
439     def _real_extract(self, url):
440         mobj = re.match(self._VALID_URL, url)
441         user_id = mobj.group('id')
442         return self.url_result(
443             '%s/videos' % mobj.group('url'), ie=PornHubPagedVideoListIE.ie_key(),
444             video_id=user_id)
445
446
447 class PornHubPagedPlaylistBaseIE(PornHubPlaylistBaseIE):
448     @staticmethod
449     def _has_more(webpage):
450         return re.search(
451             r'''(?x)
452                 <li[^>]+\bclass=["\']page_next|
453                 <link[^>]+\brel=["\']next|
454                 <button[^>]+\bid=["\']moreDataBtn
455             ''', webpage) is not None
456
457     def _real_extract(self, url):
458         mobj = re.match(self._VALID_URL, url)
459         host = mobj.group('host')
460         item_id = mobj.group('id')
461
462         page = int_or_none(self._search_regex(
463             r'\bpage=(\d+)', url, 'page', default=None))
464
465         entries = []
466         for page_num in (page, ) if page is not None else itertools.count(1):
467             try:
468                 webpage = self._download_webpage(
469                     url, item_id, 'Downloading page %d' % page_num,
470                     query={'page': page_num})
471             except ExtractorError as e:
472                 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 404:
473                     break
474                 raise
475             page_entries = self._extract_entries(webpage, host)
476             if not page_entries:
477                 break
478             entries.extend(page_entries)
479             if not self._has_more(webpage):
480                 break
481
482         return self.playlist_result(orderedSet(entries), item_id)
483
484
485 class PornHubPagedVideoListIE(PornHubPagedPlaylistBaseIE):
486     _VALID_URL = r'https?://(?:[^/]+\.)?(?P<host>pornhub(?:premium)?\.(?:com|net))/(?P<id>(?:[^/]+/)*[^/?#&]+)'
487     _TESTS = [{
488         'url': 'https://www.pornhub.com/model/zoe_ph/videos',
489         'only_matching': True,
490     }, {
491         'url': 'http://www.pornhub.com/users/rushandlia/videos',
492         'only_matching': True,
493     }, {
494         'url': 'https://www.pornhub.com/pornstar/jenny-blighe/videos',
495         'info_dict': {
496             'id': 'pornstar/jenny-blighe/videos',
497         },
498         'playlist_mincount': 149,
499     }, {
500         'url': 'https://www.pornhub.com/pornstar/jenny-blighe/videos?page=3',
501         'info_dict': {
502             'id': 'pornstar/jenny-blighe/videos',
503         },
504         'playlist_mincount': 40,
505     }, {
506         # default sorting as Top Rated Videos
507         'url': 'https://www.pornhub.com/channels/povd/videos',
508         'info_dict': {
509             'id': 'channels/povd/videos',
510         },
511         'playlist_mincount': 293,
512     }, {
513         # Top Rated Videos
514         'url': 'https://www.pornhub.com/channels/povd/videos?o=ra',
515         'only_matching': True,
516     }, {
517         # Most Recent Videos
518         'url': 'https://www.pornhub.com/channels/povd/videos?o=da',
519         'only_matching': True,
520     }, {
521         # Most Viewed Videos
522         'url': 'https://www.pornhub.com/channels/povd/videos?o=vi',
523         'only_matching': True,
524     }, {
525         'url': 'http://www.pornhub.com/users/zoe_ph/videos/public',
526         'only_matching': True,
527     }, {
528         # Most Viewed Videos
529         'url': 'https://www.pornhub.com/pornstar/liz-vicious/videos?o=mv',
530         'only_matching': True,
531     }, {
532         # Top Rated Videos
533         'url': 'https://www.pornhub.com/pornstar/liz-vicious/videos?o=tr',
534         'only_matching': True,
535     }, {
536         # Longest Videos
537         'url': 'https://www.pornhub.com/pornstar/liz-vicious/videos?o=lg',
538         'only_matching': True,
539     }, {
540         # Newest Videos
541         'url': 'https://www.pornhub.com/pornstar/liz-vicious/videos?o=cm',
542         'only_matching': True,
543     }, {
544         'url': 'https://www.pornhub.com/pornstar/liz-vicious/videos/paid',
545         'only_matching': True,
546     }, {
547         'url': 'https://www.pornhub.com/pornstar/liz-vicious/videos/fanonly',
548         'only_matching': True,
549     }, {
550         'url': 'https://www.pornhub.com/video',
551         'only_matching': True,
552     }, {
553         'url': 'https://www.pornhub.com/video?page=3',
554         'only_matching': True,
555     }, {
556         'url': 'https://www.pornhub.com/video/search?search=123',
557         'only_matching': True,
558     }, {
559         'url': 'https://www.pornhub.com/categories/teen',
560         'only_matching': True,
561     }, {
562         'url': 'https://www.pornhub.com/categories/teen?page=3',
563         'only_matching': True,
564     }, {
565         'url': 'https://www.pornhub.com/hd',
566         'only_matching': True,
567     }, {
568         'url': 'https://www.pornhub.com/hd?page=3',
569         'only_matching': True,
570     }, {
571         'url': 'https://www.pornhub.com/described-video',
572         'only_matching': True,
573     }, {
574         'url': 'https://www.pornhub.com/described-video?page=2',
575         'only_matching': True,
576     }, {
577         'url': 'https://www.pornhub.com/video/incategories/60fps-1/hd-porn',
578         'only_matching': True,
579     }, {
580         'url': 'https://www.pornhub.com/playlist/44121572',
581         'info_dict': {
582             'id': 'playlist/44121572',
583         },
584         'playlist_mincount': 132,
585     }, {
586         'url': 'https://www.pornhub.com/playlist/4667351',
587         'only_matching': True,
588     }, {
589         'url': 'https://de.pornhub.com/playlist/4667351',
590         'only_matching': True,
591     }]
592
593     @classmethod
594     def suitable(cls, url):
595         return (False
596                 if PornHubIE.suitable(url) or PornHubUserIE.suitable(url) or PornHubUserVideosUploadIE.suitable(url)
597                 else super(PornHubPagedVideoListIE, cls).suitable(url))
598
599
600 class PornHubUserVideosUploadIE(PornHubPagedPlaylistBaseIE):
601     _VALID_URL = r'(?P<url>https?://(?:[^/]+\.)?(?P<host>pornhub(?:premium)?\.(?:com|net))/(?:(?:user|channel)s|model|pornstar)/(?P<id>[^/]+)/videos/upload)'
602     _TESTS = [{
603         'url': 'https://www.pornhub.com/pornstar/jenny-blighe/videos/upload',
604         'info_dict': {
605             'id': 'jenny-blighe',
606         },
607         'playlist_mincount': 129,
608     }, {
609         'url': 'https://www.pornhub.com/model/zoe_ph/videos/upload',
610         'only_matching': True,
611     }]