X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/47d80ec0b18245caeb97018d4c1af18d0b5b972b..acbcfdb2cbb57cd2ab6a5d73341cd077fa71f6c9:/youtube_dl/extractor/pornhub.py?ds=sidebyside

diff --git a/youtube_dl/extractor/pornhub.py b/youtube_dl/extractor/pornhub.py
index 40dbe69..cb59d52 100644
--- a/youtube_dl/extractor/pornhub.py
+++ b/youtube_dl/extractor/pornhub.py
@@ -1,36 +1,57 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import functools
 import itertools
-import os
+import operator
 import re
 
 from .common import InfoExtractor
 from ..compat import (
     compat_HTTPError,
-    compat_urllib_parse_unquote,
-    compat_urllib_parse_unquote_plus,
-    compat_urllib_parse_urlparse,
+    compat_str,
+    compat_urllib_request,
 )
+from .openload import PhantomJSwrapper
 from ..utils import (
+    determine_ext,
     ExtractorError,
     int_or_none,
-    js_to_json,
     orderedSet,
-    sanitized_Request,
+    remove_quotes,
     str_to_int,
+    url_or_none,
 )
-from ..aes import (
-    aes_decrypt_text
-)
 
 
-class PornHubIE(InfoExtractor):
+class PornHubBaseIE(InfoExtractor):
+    def _download_webpage_handle(self, *args, **kwargs):
+        def dl(*args, **kwargs):
+            return super(PornHubBaseIE, self)._download_webpage_handle(*args, **kwargs)
+
+        webpage, urlh = dl(*args, **kwargs)
+
+        if any(re.search(p, webpage) for p in (
+                r'<body\b[^>]+\bonload=["\']go\(\)',
+                r'document\.cookie\s*=\s*["\']RNKEY=',
+                r'document\.location\.reload\(true\)')):
+            url_or_request = args[0]
+            url = (url_or_request.get_full_url()
+                   if isinstance(url_or_request, compat_urllib_request.Request)
+                   else url_or_request)
+            phantom = PhantomJSwrapper(self, required_version='2.0')
+            phantom.get(url, html=webpage)
+            webpage, urlh = dl(*args, **kwargs)
+
+        return webpage, urlh
+
+
+class PornHubIE(PornHubBaseIE):
     IE_DESC = 'PornHub and Thumbzilla'
     _VALID_URL = r'''(?x)
                     https?://
                         (?:
-                            (?:[a-z]+\.)?pornhub\.com/(?:view_video\.php\?viewkey=|embed/)|
+                            (?:[^/]+\.)?(?P<host>pornhub\.(?:com|net))/(?:(?:view_video\.php|video/show)\?viewkey=|embed/)|
                             (?:www\.)?thumbzilla\.com/video/
                         )
                         (?P<id>[\da-z]+)
@@ -43,6 +64,7 @@ class PornHubIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'Seductive Indian beauty strips down and fingers her pink pussy',
             'uploader': 'Babes',
+            'upload_date': '20130628',
             'duration': 361,
             'view_count': int,
             'like_count': int,
@@ -59,7 +81,8 @@ class PornHubIE(InfoExtractor):
             'id': '1331683002',
             'ext': 'mp4',
             'title': '重庆婷婷女王足交',
-            'uploader': 'cj397186295',
+            'uploader': 'Unknown',
+            'upload_date': '20150213',
             'duration': 1753,
             'view_count': int,
             'like_count': int,
@@ -72,6 +95,31 @@ class PornHubIE(InfoExtractor):
         'params': {
             'skip_download': True,
         },
+    }, {
+        # subtitles
+        'url': 'https://www.pornhub.com/view_video.php?viewkey=ph5af5fef7c2aa7',
+        'info_dict': {
+            'id': 'ph5af5fef7c2aa7',
+            'ext': 'mp4',
+            'title': 'BFFS - Cute Teen Girls Share Cock On the Floor',
+            'uploader': 'BFFs',
+            'duration': 622,
+            'view_count': int,
+            'like_count': int,
+            'dislike_count': int,
+            'comment_count': int,
+            'age_limit': 18,
+            'tags': list,
+            'categories': list,
+            'subtitles': {
+                'en': [{
+                    "ext": 'srt'
+                }]
+            },
+        },
+        'params': {
+            'skip_download': True,
+        },
     }, {
         'url': 'http://www.pornhub.com/view_video.php?viewkey=ph557bbb6676d2d',
         'only_matching': True,
@@ -94,12 +142,18 @@ class PornHubIE(InfoExtractor):
     }, {
         'url': 'https://www.thumbzilla.com/video/ph56c6114abd99a/horny-girlfriend-sex',
         'only_matching': True,
+    }, {
+        'url': 'http://www.pornhub.com/video/show?viewkey=648719015',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.pornhub.net/view_video.php?viewkey=203640933',
+        'only_matching': True,
     }]
 
     @staticmethod
     def _extract_urls(webpage):
         return re.findall(
-            r'<iframe[^>]+?src=["\'](?P<url>(?:https?:)?//(?:www\.)?pornhub\.com/embed/[\da-z]+)',
+            r'<iframe[^>]+?src=["\'](?P<url>(?:https?:)?//(?:www\.)?pornhub\.(?:com|net)/embed/[\da-z]+)',
             webpage)
 
     def _extract_count(self, pattern, webpage, name):
@@ -107,12 +161,19 @@ class PornHubIE(InfoExtractor):
             pattern, webpage, '%s count' % name, fatal=False))
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
+        mobj = re.match(self._VALID_URL, url)
+        host = mobj.group('host') or 'pornhub.com'
+        video_id = mobj.group('id')
 
-        req = sanitized_Request(
-            'http://www.pornhub.com/view_video.php?viewkey=%s' % video_id)
-        req.add_header('Cookie', 'age_verified=1')
-        webpage = self._download_webpage(req, video_id)
+        self._set_cookie(host, 'age_verified', '1')
+
+        def dl_webpage(platform):
+            self._set_cookie(host, 'platform', platform)
+            return self._download_webpage(
+                'https://www.%s/view_video.php?viewkey=%s' % (host, video_id),
+                video_id, 'Downloading %s webpage' % platform)
+
+        webpage = dl_webpage('pc')
 
         error_msg = self._html_search_regex(
             r'(?s)<div[^>]+class=(["\'])(?:(?!\1).)*\b(?:removed|userMessageSection)\b(?:(?!\1).)*\1[^>]*>(?P<error>.+?)</div>',
@@ -133,18 +194,108 @@ class PornHubIE(InfoExtractor):
              r'shareTitle\s*=\s*(["\'])(?P<title>.+?)\1'),
             webpage, 'title', group='title')
 
+        video_urls = []
+        video_urls_set = set()
+        subtitles = {}
+
         flashvars = self._parse_json(
             self._search_regex(
                 r'var\s+flashvars_\d+\s*=\s*({.+?});', webpage, 'flashvars', default='{}'),
             video_id)
         if flashvars:
+            subtitle_url = url_or_none(flashvars.get('closedCaptionsFile'))
+            if subtitle_url:
+                subtitles.setdefault('en', []).append({
+                    'url': subtitle_url,
+                    'ext': 'srt',
+                })
             thumbnail = flashvars.get('image_url')
             duration = int_or_none(flashvars.get('video_duration'))
+            media_definitions = flashvars.get('mediaDefinitions')
+            if isinstance(media_definitions, list):
+                for definition in media_definitions:
+                    if not isinstance(definition, dict):
+                        continue
+                    video_url = definition.get('videoUrl')
+                    if not video_url or not isinstance(video_url, compat_str):
+                        continue
+                    if video_url in video_urls_set:
+                        continue
+                    video_urls_set.add(video_url)
+                    video_urls.append(
+                        (video_url, int_or_none(definition.get('quality'))))
         else:
-            title, thumbnail, duration = [None] * 3
+            thumbnail, duration = [None] * 2
+
+        if not video_urls:
+            tv_webpage = dl_webpage('tv')
+
+            assignments = self._search_regex(
+                r'(var.+?mediastring.+?)</script>', tv_webpage,
+                'encoded url').split(';')
+
+            js_vars = {}
+
+            def parse_js_value(inp):
+                inp = re.sub(r'/\*(?:(?!\*/).)*?\*/', '', inp)
+                if '+' in inp:
+                    inps = inp.split('+')
+                    return functools.reduce(
+                        operator.concat, map(parse_js_value, inps))
+                inp = inp.strip()
+                if inp in js_vars:
+                    return js_vars[inp]
+                return remove_quotes(inp)
+
+            for assn in assignments:
+                assn = assn.strip()
+                if not assn:
+                    continue
+                assn = re.sub(r'var\s+', '', assn)
+                vname, value = assn.split('=', 1)
+                js_vars[vname] = parse_js_value(value)
+
+            video_url = js_vars['mediastring']
+            if video_url not in video_urls_set:
+                video_urls.append((video_url, None))
+                video_urls_set.add(video_url)
+
+        for mobj in re.finditer(
+                r'<a[^>]+\bclass=["\']downloadBtn\b[^>]+\bhref=(["\'])(?P<url>(?:(?!\1).)+)\1',
+                webpage):
+            video_url = mobj.group('url')
+            if video_url not in video_urls_set:
+                video_urls.append((video_url, None))
+                video_urls_set.add(video_url)
+
+        upload_date = None
+        formats = []
+        for video_url, height in video_urls:
+            if not upload_date:
+                upload_date = self._search_regex(
+                    r'/(\d{6}/\d{2})/', video_url, 'upload data', default=None)
+                if upload_date:
+                    upload_date = upload_date.replace('/', '')
+            if determine_ext(video_url) == 'mpd':
+                formats.extend(self._extract_mpd_formats(
+                    video_url, video_id, mpd_id='dash', fatal=False))
+                continue
+            tbr = None
+            mobj = re.search(r'(?P<height>\d+)[pP]?_(?P<tbr>\d+)[kK]', video_url)
+            if mobj:
+                if not height:
+                    height = int(mobj.group('height'))
+                tbr = int(mobj.group('tbr'))
+            formats.append({
+                'url': video_url,
+                'format_id': '%dp' % height if height else None,
+                'height': height,
+                'tbr': tbr,
+            })
+        self._sort_formats(formats)
 
         video_uploader = self._html_search_regex(
-            r'(?s)From:&nbsp;.+?<(?:a href="/users/|a href="/channels/|span class="username)[^>]+>(.+?)<',
+            r'(?s)From:&nbsp;.+?<(?:a\b[^>]+\bhref=["\']/(?:(?:user|channel)s|model|pornstar)/|span\b[^>]+\bclass=["\']username)[^>]+>(.+?)<',
             webpage, 'uploader', fatal=False)
 
         view_count = self._extract_count(
@@ -156,49 +307,17 @@ class PornHubIE(InfoExtractor):
         comment_count = self._extract_count(
             r'All Comments\s*<span>\(([\d,.]+)\)', webpage, 'comment')
 
-        video_urls = list(map(compat_urllib_parse_unquote, re.findall(r"player_quality_[0-9]{3}p\s*=\s*'([^']+)'", webpage)))
-        if webpage.find('"encrypted":true') != -1:
-            password = compat_urllib_parse_unquote_plus(
-                self._search_regex(r'"video_title":"([^"]+)', webpage, 'password'))
-            video_urls = list(map(lambda s: aes_decrypt_text(s, password, 32).decode('utf-8'), video_urls))
-
-        formats = []
-        for video_url in video_urls:
-            path = compat_urllib_parse_urlparse(video_url).path
-            extension = os.path.splitext(path)[1][1:]
-            format = path.split('/')[5].split('_')[:2]
-            format = '-'.join(format)
-
-            m = re.match(r'^(?P<height>[0-9]+)[pP]-(?P<tbr>[0-9]+)[kK]$', format)
-            if m is None:
-                height = None
-                tbr = None
-            else:
-                height = int(m.group('height'))
-                tbr = int(m.group('tbr'))
-
-            formats.append({
-                'url': video_url,
-                'ext': extension,
-                'format': format,
-                'format_id': format,
-                'tbr': tbr,
-                'height': height,
-            })
-        self._sort_formats(formats)
-
-        page_params = self._parse_json(self._search_regex(
-            r'page_params\.zoneDetails\[([\'"])[^\'"]+\1\]\s*=\s*(?P<data>{[^}]+})',
-            webpage, 'page parameters', group='data', default='{}'),
-            video_id, transform_source=js_to_json, fatal=False)
-        tags = categories = None
-        if page_params:
-            tags = page_params.get('tags', '').split(',')
-            categories = page_params.get('categories', '').split(',')
+        def extract_list(meta_key):
+            div = self._search_regex(
+                r'(?s)<div[^>]+\bclass=["\'].*?\b%sWrapper[^>]*>(.+?)</div>'
+                % meta_key, webpage, meta_key, default=None)
+            if div:
+                return re.findall(r'<a[^>]+\bhref=[^>]+>([^<]+)', div)
 
         return {
             'id': video_id,
             'uploader': video_uploader,
+            'upload_date': upload_date,
             'title': title,
             'thumbnail': thumbnail,
             'duration': duration,
@@ -208,52 +327,68 @@ class PornHubIE(InfoExtractor):
             'comment_count': comment_count,
             'formats': formats,
             'age_limit': 18,
-            'tags': tags,
-            'categories': categories,
+            'tags': extract_list('tags'),
+            'categories': extract_list('categories'),
+            'subtitles': subtitles,
         }
 
 
-class PornHubPlaylistBaseIE(InfoExtractor):
-    def _extract_entries(self, webpage):
+class PornHubPlaylistBaseIE(PornHubBaseIE):
+    def _extract_entries(self, webpage, host):
+        # Only process container div with main playlist content skipping
+        # drop-down menu that uses similar pattern for videos (see
+        # https://github.com/ytdl-org/youtube-dl/issues/11594).
+        container = self._search_regex(
+            r'(?s)(<div[^>]+class=["\']container.+)', webpage,
+            'container', default=webpage)
+
         return [
             self.url_result(
-                'http://www.pornhub.com/%s' % video_url,
+                'http://www.%s/%s' % (host, video_url),
                 PornHubIE.ie_key(), video_title=title)
             for video_url, title in orderedSet(re.findall(
                 r'href="/?(view_video\.php\?.*\bviewkey=[\da-z]+[^"]*)"[^>]*\s+title="([^"]+)"',
-                webpage))
+                container))
         ]
 
     def _real_extract(self, url):
-        playlist_id = self._match_id(url)
+        mobj = re.match(self._VALID_URL, url)
+        host = mobj.group('host')
+        playlist_id = mobj.group('id')
 
         webpage = self._download_webpage(url, playlist_id)
 
-        entries = self._extract_entries(webpage)
+        entries = self._extract_entries(webpage, host)
 
         playlist = self._parse_json(
             self._search_regex(
-                r'playlistObject\s*=\s*({.+?});', webpage, 'playlist'),
-            playlist_id)
+                r'(?:playlistObject|PLAYLIST_VIEW)\s*=\s*({.+?});', webpage,
+                'playlist', default='{}'),
+            playlist_id, fatal=False)
+        title = playlist.get('title') or self._search_regex(
+            r'>Videos\s+in\s+(.+?)\s+[Pp]laylist<', webpage, 'title', fatal=False)
 
         return self.playlist_result(
-            entries, playlist_id, playlist.get('title'), playlist.get('description'))
+            entries, playlist_id, title, playlist.get('description'))
 
 
 class PornHubPlaylistIE(PornHubPlaylistBaseIE):
-    _VALID_URL = r'https?://(?:www\.)?pornhub\.com/playlist/(?P<id>\d+)'
+    _VALID_URL = r'https?://(?:[^/]+\.)?(?P<host>pornhub\.(?:com|net))/playlist/(?P<id>\d+)'
     _TESTS = [{
-        'url': 'http://www.pornhub.com/playlist/6201671',
+        'url': 'http://www.pornhub.com/playlist/4667351',
         'info_dict': {
-            'id': '6201671',
-            'title': 'P0p4',
+            'id': '4667351',
+            'title': 'Nataly Hot',
         },
-        'playlist_mincount': 35,
+        'playlist_mincount': 2,
+    }, {
+        'url': 'https://de.pornhub.com/playlist/4667351',
+        'only_matching': True,
     }]
 
 
 class PornHubUserVideosIE(PornHubPlaylistBaseIE):
-    _VALID_URL = r'https?://(?:www\.)?pornhub\.com/users/(?P<id>[^/]+)/videos'
+    _VALID_URL = r'https?://(?:[^/]+\.)?(?P<host>pornhub\.(?:com|net))/(?:(?:user|channel)s|model|pornstar)/(?P<id>[^/]+)/videos'
     _TESTS = [{
         'url': 'http://www.pornhub.com/users/zoe_ph/videos/public',
         'info_dict': {
@@ -263,10 +398,40 @@ class PornHubUserVideosIE(PornHubPlaylistBaseIE):
     }, {
         'url': 'http://www.pornhub.com/users/rushandlia/videos',
         'only_matching': True,
+    }, {
+        # default sorting as Top Rated Videos
+        'url': 'https://www.pornhub.com/channels/povd/videos',
+        'info_dict': {
+            'id': 'povd',
+        },
+        'playlist_mincount': 293,
+    }, {
+        # Top Rated Videos
+        'url': 'https://www.pornhub.com/channels/povd/videos?o=ra',
+        'only_matching': True,
+    }, {
+        # Most Recent Videos
+        'url': 'https://www.pornhub.com/channels/povd/videos?o=da',
+        'only_matching': True,
+    }, {
+        # Most Viewed Videos
+        'url': 'https://www.pornhub.com/channels/povd/videos?o=vi',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.pornhub.com/users/zoe_ph/videos/public',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.pornhub.com/model/jayndrea/videos/upload',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.pornhub.com/pornstar/jenny-blighe/videos/upload',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
-        user_id = self._match_id(url)
+        mobj = re.match(self._VALID_URL, url)
+        host = mobj.group('host')
+        user_id = mobj.group('id')
 
         entries = []
         for page_num in itertools.count(1):
@@ -277,7 +442,8 @@ class PornHubUserVideosIE(PornHubPlaylistBaseIE):
             except ExtractorError as e:
                 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 404:
                     break
-            page_entries = self._extract_entries(webpage)
+                raise
+            page_entries = self._extract_entries(webpage, host)
             if not page_entries:
                 break
             entries.extend(page_entries)