yt_dlp/extractor/pornhub.py

@@ -24,7 +24,6 @@ from ..utils import (
     url_or_none,
     urlencode_postdata,
 )
-from ..utils.traversal import find_elements, traverse_obj
 
 
 class PornHubBaseIE(InfoExtractor):
@@ -138,24 +137,23 @@ class PornHubIE(PornHubBaseIE):
     _EMBED_REGEX = [r'<iframe[^>]+?src=["\'](?P<url>(?:https?:)?//(?:www\.)?pornhub(?:premium)?\.(?:com|net|org)/embed/[\da-z]+)']
     _TESTS = [{
         'url': 'http://www.pornhub.com/view_video.php?viewkey=648719015',
-        'md5': '4d4a4e9178b655776f86cf89ecaf0edf',
+        'md5': 'a6391306d050e4547f62b3f485dd9ba9',
         'info_dict': {
             'id': '648719015',
             'ext': 'mp4',
             'title': 'Seductive Indian beauty strips down and fingers her pink pussy',
-            'uploader': 'BABES-COM',
+            'uploader': 'Babes',
-            'uploader_id': '/users/babes-com',
             'upload_date': '20130628',
             'timestamp': 1372447216,
             'duration': 361,
             'view_count': int,
             'like_count': int,
+            'dislike_count': int,
             'comment_count': int,
             'age_limit': 18,
             'tags': list,
             'categories': list,
             'cast': list,
-            'thumbnail': r're:https?://.+',
         },
     }, {
         # non-ASCII title
@@ -482,6 +480,13 @@ class PornHubIE(PornHubBaseIE):
         comment_count = self._extract_count(
             r'All Comments\s*<span>\(([\d,.]+)\)', webpage, 'comment')
 
+        def extract_list(meta_key):
+            div = self._search_regex(
+                rf'(?s)<div[^>]+\bclass=["\'].*?\b{meta_key}Wrapper[^>]*>(.+?)</div>',
+                webpage, meta_key, default=None)
+            if div:
+                return [clean_html(x).strip() for x in re.findall(r'(?s)<a[^>]+\bhref=[^>]+>.+?</a>', div)]
+
         info = self._search_json_ld(webpage, video_id, default={})
         # description provided in JSON-LD is irrelevant
         info['description'] = None
@@ -500,11 +505,9 @@ class PornHubIE(PornHubBaseIE):
             'comment_count': comment_count,
             'formats': formats,
             'age_limit': 18,
-            **traverse_obj(webpage, {
+            'tags': extract_list('tags'),
-                'tags': ({find_elements(attr='data-label', value='tag')}, ..., {clean_html}),
+            'categories': extract_list('categories'),
-                'categories': ({find_elements(attr='data-label', value='category')}, ..., {clean_html}),
+            'cast': extract_list('pornstars'),
-                'cast': ({find_elements(attr='data-label', value='pornstar')}, ..., {clean_html}),
-            }),
             'subtitles': subtitles,
         }, info)
 

yt_dlp/extractor/telecinco.py

@@ -6,21 +6,20 @@ from ..networking.exceptions import HTTPError
 from ..utils import (
     ExtractorError,
     clean_html,
-    extract_attributes,
     int_or_none,
     join_nonempty,
     str_or_none,
+    traverse_obj,
     update_url,
     url_or_none,
 )
-from ..utils.traversal import traverse_obj
 
 
 class TelecincoBaseIE(InfoExtractor):
     def _parse_content(self, content, url):
-        video_id = content['dataMediaId'][1]
+        video_id = content['dataMediaId']
         config = self._download_json(
-            content['dataConfig'][1], video_id, 'Downloading config JSON')
+            content['dataConfig'], video_id, 'Downloading config JSON')
         services = config['services']
         caronte = self._download_json(services['caronte'], video_id)
         if traverse_obj(caronte, ('dls', 0, 'drm', {bool})):
@@ -58,9 +57,9 @@ class TelecincoBaseIE(InfoExtractor):
             'id': video_id,
             'title': traverse_obj(config, ('info', 'title', {str})),
             'formats': formats,
-            'thumbnail': (traverse_obj(content, ('dataPoster', 1, {url_or_none}))
+            'thumbnail': (traverse_obj(content, ('dataPoster', {url_or_none}))
                           or traverse_obj(config, 'poster', 'imageUrl', expected_type=url_or_none)),
-            'duration': traverse_obj(content, ('dataDuration', 1, {int_or_none})),
+            'duration': traverse_obj(content, ('dataDuration', {int_or_none})),
             'http_headers': headers,
         }
 
@@ -138,45 +137,30 @@ class TelecincoIE(TelecincoBaseIE):
         'url': 'http://www.cuatro.com/chesterinlove/a-carta/chester-chester_in_love-chester_edu_2_2331030022.html',
         'only_matching': True,
     }]
-    _ASTRO_ISLAND_RE = re.compile(r'<astro-island\b[^>]+>')
 
     def _real_extract(self, url):
         display_id = self._match_id(url)
         webpage = self._download_webpage(url, display_id, impersonate=True)
+        article = self._search_json(
+            r'window\.\$REACTBASE_STATE\.article(?:_multisite)?\s*=',
+            webpage, 'article', display_id)['article']
+        description = traverse_obj(article, ('leadParagraph', {clean_html}, filter))
 
-        props_list = traverse_obj(webpage, (
+        if article.get('editorialType') != 'VID':
-            {self._ASTRO_ISLAND_RE.findall}, ...,
-            {extract_attributes}, 'props', {json.loads}))
-
-        description = traverse_obj(props_list, (..., 'leadParagraph', 1, {clean_html}, any, filter))
-        main_content = traverse_obj(props_list, (..., ('content', ('articleData', 1, 'opening')), 1, {dict}, any))
-
-        if traverse_obj(props_list, (..., 'editorialType', 1, {str}, any)) != 'VID':  # e.g. 'ART'
             entries = []
 
-            for p in traverse_obj(props_list, (..., 'articleData', 1, ('opening', ('body', 1, ...)), 1, {dict})):
+            for p in traverse_obj(article, ((('opening', all), 'body'), lambda _, v: v['content'])):
-                type_ = traverse_obj(p, ('type', 1, {str}))
+                content = p['content']
-                content = traverse_obj(p, ('content', 1, {str} if type_ == 'paragraph' else {dict}))
+                type_ = p.get('type')
-                if not content:
+                if type_ == 'paragraph' and isinstance(content, str):
-                    continue
-                if type_ == 'paragraph':
                     description = join_nonempty(description, content, delim='')
-                elif type_ == 'video':
+                elif type_ == 'video' and isinstance(content, dict):
                     entries.append(self._parse_content(content, url))
-                else:
-                    self.report_warning(
-                        f'Skipping unsupported content type "{type_}"', display_id, only_once=True)
 
             return self.playlist_result(
-                entries,
+                entries, str_or_none(article.get('id')),
-                traverse_obj(props_list, (..., 'id', 1, {int}, {str_or_none}, any)) or display_id,
+                traverse_obj(article, ('title', {str})), clean_html(description))
-                traverse_obj(main_content, ('dataTitle', 1, {str})),
-                clean_html(description))
 
-        if not main_content:
+        info = self._parse_content(article['opening']['content'], url)
-            raise ExtractorError('Unable to extract main content from webpage')
-
-        info = self._parse_content(main_content, url)
         info['description'] = description
-
         return info