Start moving to ytdl-org

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

diff --git a/youtube_dl/extractor/pornhub.py b/youtube_dl/extractor/pornhub.py
index f5f3e6593a2bc788b05efd155cff2e851fab729f..3a474c17970bbd087de187bf6059e9900551f61f 100644 (file)
--- a/youtube_dl/extractor/pornhub.py
+++ b/youtube_dl/extractor/pornhub.py
@@ -16,7 +16,6 @@ from .openload import PhantomJSwrapper
 from ..utils import (
     ExtractorError,
     int_or_none,
-    js_to_json,
     orderedSet,
     remove_quotes,
     str_to_int,
@@ -24,7 +23,29 @@ from ..utils import (
 )
 
 
-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?://
@@ -128,26 +149,6 @@ class PornHubIE(InfoExtractor):
         'only_matching': True,
     }]
 
-    def _download_webpage_handle(self, *args, **kwargs):
-        def dl(*args, **kwargs):
-            return super(PornHubIE, 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
-
     @staticmethod
     def _extract_urls(webpage):
         return re.findall(
@@ -301,14 +302,12 @@ class PornHubIE(InfoExtractor):
         comment_count = self._extract_count(
             r'All Comments\s*<span>\(([\d,.]+)\)', webpage, 'comment')
 
-        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,
@@ -323,17 +322,17 @@ 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):
+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/rg3/youtube-dl/issues/11594).
+        # https://github.com/ytdl-org/youtube-dl/issues/11594).
         container = self._search_regex(
             r'(?s)(<div[^>]+class=["\']container.+)', webpage,
             'container', default=webpage)