[pornhub] Extract metadata from JSON-LD (closes #26614)

[youtube-dl] / youtube_dl / extractor / pornhub.py

diff --git a/youtube_dl/extractor/pornhub.py b/youtube_dl/extractor/pornhub.py
index b3251ccd9b2300188f7efce561c3c5d1fbff702e..529f3f7119fd4e93a8c82168d910ab4fc3d1720e 100644 (file)
--- a/youtube_dl/extractor/pornhub.py
+++ b/youtube_dl/extractor/pornhub.py
@@ -17,6 +17,7 @@ from ..utils import (
     determine_ext,
     ExtractorError,
     int_or_none,
+    merge_dicts,
     NO_DEFAULT,
     orderedSet,
     remove_quotes,
@@ -52,20 +53,21 @@ class PornHubIE(PornHubBaseIE):
     _VALID_URL = r'''(?x)
                     https?://
                         (?:
-                            (?:[^/]+\.)?(?P<host>pornhub\.(?:com|net))/(?:(?:view_video\.php|video/show)\?viewkey=|embed/)|
+                            (?:[^/]+\.)?(?P<host>pornhub(?:premium)?\.(?:com|net))/(?:(?:view_video\.php|video/show)\?viewkey=|embed/)|
                             (?:www\.)?thumbzilla\.com/video/
                         )
                         (?P<id>[\da-z]+)
                     '''
     _TESTS = [{
         'url': 'http://www.pornhub.com/view_video.php?viewkey=648719015',
-        'md5': '1e19b41231a02eba417839222ac9d58e',
+        'md5': 'a6391306d050e4547f62b3f485dd9ba9',
         'info_dict': {
             'id': '648719015',
             'ext': 'mp4',
             'title': 'Seductive Indian beauty strips down and fingers her pink pussy',
             'uploader': 'Babes',
             'upload_date': '20130628',
+            'timestamp': 1372447216,
             'duration': 361,
             'view_count': int,
             'like_count': int,
@@ -82,8 +84,8 @@ class PornHubIE(PornHubBaseIE):
             'id': '1331683002',
             'ext': 'mp4',
             'title': '重庆婷婷女王足交',
-            'uploader': 'Unknown',
             'upload_date': '20150213',
+            'timestamp': 1423804862,
             'duration': 1753,
             'view_count': int,
             'like_count': int,
@@ -121,6 +123,7 @@ class PornHubIE(PornHubBaseIE):
         'params': {
             'skip_download': True,
         },
+        'skip': 'This video has been disabled',
     }, {
         'url': 'http://www.pornhub.com/view_video.php?viewkey=ph557bbb6676d2d',
         'only_matching': True,
@@ -149,6 +152,9 @@ class PornHubIE(PornHubBaseIE):
     }, {
         'url': 'https://www.pornhub.net/view_video.php?viewkey=203640933',
         'only_matching': True,
+    }, {
+        'url': 'https://www.pornhubpremium.com/view_video.php?viewkey=ph5e4acdae54a82',
+        'only_matching': True,
     }]
 
     @staticmethod
@@ -166,6 +172,13 @@ class PornHubIE(PornHubBaseIE):
         host = mobj.group('host') or 'pornhub.com'
         video_id = mobj.group('id')
 
+        if 'premium' in host:
+            if not self._downloader.params.get('cookiefile'):
+                raise ExtractorError(
+                    'PornHub Premium requires authentication.'
+                    ' You may want to use --cookies.',
+                    expected=True)
+
         self._set_cookie(host, 'age_verified', '1')
 
         def dl_webpage(platform):
@@ -189,10 +202,10 @@ class PornHubIE(PornHubBaseIE):
         # http://www.pornhub.com/view_video.php?viewkey=1331683002), not relying
         # on that anymore.
         title = self._html_search_meta(
-            'twitter:title', webpage, default=None) or self._search_regex(
-            (r'<h1[^>]+class=["\']title["\'][^>]*>(?P<title>[^<]+)',
-             r'<div[^>]+data-video-title=(["\'])(?P<title>.+?)\1',
-             r'shareTitle\s*=\s*(["\'])(?P<title>.+?)\1'),
+            'twitter:title', webpage, default=None) or self._html_search_regex(
+            (r'(?s)<h1[^>]+class=["\']title["\'][^>]*>(?P<title>.+?)</h1>',
+             r'<div[^>]+data-video-title=(["\'])(?P<title>(?:(?!\1).)+)\1',
+             r'shareTitle["\']\s*[=:]\s*(["\'])(?P<title>(?:(?!\1).)+)\1'),
             webpage, 'title', group='title')
 
         video_urls = []
@@ -328,10 +341,10 @@ class PornHubIE(PornHubBaseIE):
 
         video_uploader = self._html_search_regex(
             r'(?s)From:&nbsp;.+?<(?:a\b[^>]+\bhref=["\']/(?:(?:user|channel)s|model|pornstar)/|span\b[^>]+\bclass=["\']username)[^>]+>(.+?)<',
-            webpage, 'uploader', fatal=False)
+            webpage, 'uploader', default=None)
 
         view_count = self._extract_count(
-            r'<span class="count">([\d,\.]+)</span> views', webpage, 'view')
+            r'<span class="count">([\d,\.]+)</span> [Vv]iews', webpage, 'view')
         like_count = self._extract_count(
             r'<span class="votesUp">([\d,\.]+)</span>', webpage, 'like')
         dislike_count = self._extract_count(
@@ -346,7 +359,11 @@ class PornHubIE(PornHubBaseIE):
             if div:
                 return re.findall(r'<a[^>]+\bhref=[^>]+>([^<]+)', div)
 
-        return {
+        info = self._search_json_ld(webpage, video_id, default={})
+        # description provided in JSON-LD is irrelevant
+        info['description'] = None
+
+        return merge_dicts({
             'id': video_id,
             'uploader': video_uploader,
             'upload_date': upload_date,
@@ -362,7 +379,7 @@ class PornHubIE(PornHubBaseIE):
             'tags': extract_list('tags'),
             'categories': extract_list('categories'),
             'subtitles': subtitles,
-        }
+        }, info)
 
 
 class PornHubPlaylistBaseIE(PornHubBaseIE):
@@ -405,7 +422,7 @@ class PornHubPlaylistBaseIE(PornHubBaseIE):
 
 
 class PornHubUserIE(PornHubPlaylistBaseIE):
-    _VALID_URL = r'(?P<url>https?://(?:[^/]+\.)?pornhub\.(?:com|net)/(?:(?:user|channel)s|model|pornstar)/(?P<id>[^/?#&]+))(?:[?#&]|/(?!videos)|$)'
+    _VALID_URL = r'(?P<url>https?://(?:[^/]+\.)?(?P<host>pornhub(?:premium)?\.(?:com|net))/(?:(?:user|channel)s|model|pornstar)/(?P<id>[^/?#&]+))(?:[?#&]|/(?!videos)|$)'
     _TESTS = [{
         'url': 'https://www.pornhub.com/model/zoe_ph',
         'playlist_mincount': 118,
@@ -473,7 +490,7 @@ class PornHubPagedPlaylistBaseIE(PornHubPlaylistBaseIE):
 
 
 class PornHubPagedVideoListIE(PornHubPagedPlaylistBaseIE):
-    _VALID_URL = r'https?://(?:[^/]+\.)?(?P<host>pornhub\.(?:com|net))/(?P<id>(?:[^/]+/)*[^/?#&]+)'
+    _VALID_URL = r'https?://(?:[^/]+\.)?(?P<host>pornhub(?:premium)?\.(?:com|net))/(?P<id>(?:[^/]+/)*[^/?#&]+)'
     _TESTS = [{
         'url': 'https://www.pornhub.com/model/zoe_ph/videos',
         'only_matching': True,
@@ -588,7 +605,7 @@ class PornHubPagedVideoListIE(PornHubPagedPlaylistBaseIE):
 
 
 class PornHubUserVideosUploadIE(PornHubPagedPlaylistBaseIE):
-    _VALID_URL = r'(?P<url>https?://(?:[^/]+\.)?(?P<host>pornhub\.(?:com|net))/(?:(?:user|channel)s|model|pornstar)/(?P<id>[^/]+)/videos/upload)'
+    _VALID_URL = r'(?P<url>https?://(?:[^/]+\.)?(?P<host>pornhub(?:premium)?\.(?:com|net))/(?:(?:user|channel)s|model|pornstar)/(?P<id>[^/]+)/videos/upload)'
     _TESTS = [{
         'url': 'https://www.pornhub.com/pornstar/jenny-blighe/videos/upload',
         'info_dict': {