[pornhub:playlistbase] Improve extract entries

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

diff --git a/youtube_dl/extractor/theplatform.py b/youtube_dl/extractor/theplatform.py
index 1555aa77cac30c18de3f0c2db9e13ea00cc569f6..755f816fff95bbe3567afe0168936eb6adb53095 100644 (file)
--- a/youtube_dl/extractor/theplatform.py
+++ b/youtube_dl/extractor/theplatform.py
@@ -16,11 +16,11 @@ from ..compat import (
 from ..utils import (
     determine_ext,
     ExtractorError,
-    xpath_with_ns,
-    unsmuggle_url,
-    int_or_none,
-    url_basename,
     float_or_none,
+    int_or_none,
+    sanitized_Request,
+    unsmuggle_url,
+    xpath_with_ns,
 )
 
 default_ns = 'http://www.w3.org/2005/SMIL21/Language'
@@ -84,7 +84,7 @@ class ThePlatformBaseIE(InfoExtractor):
 class ThePlatformIE(ThePlatformBaseIE):
     _VALID_URL = r'''(?x)
         (?:https?://(?:link|player)\.theplatform\.com/[sp]/(?P<provider_id>[^/]+)/
-           (?:(?P<media>(?:[^/]+/)+select/media/)|(?P<config>(?:[^/\?]+/(?:swf|config)|onsite)/select/))?
+           (?:(?P<media>(?:(?:[^/]+/)+select/)?media/)|(?P<config>(?:[^/\?]+/(?:swf|config)|onsite)/select/))?
          |theplatform:)(?P<id>[^/\?&]+)'''
 
     _TESTS = [{
@@ -204,7 +204,12 @@ class ThePlatformIE(ThePlatformBaseIE):
             smil_url = url
         # Explicitly specified SMIL (see https://github.com/rg3/youtube-dl/issues/7385)
         elif '/guid/' in url:
-            webpage = self._download_webpage(url, video_id)
+            headers = {}
+            source_url = smuggled_data.get('source_url')
+            if source_url:
+                headers['Referer'] = source_url
+            request = sanitized_Request(url, headers=headers)
+            webpage = self._download_webpage(request, video_id)
             smil_url = self._search_regex(
                 r'<link[^>]+href=(["\'])(?P<url>.+?)\1[^>]+type=["\']application/smil\+xml',
                 webpage, 'smil url', group='url')
@@ -277,8 +282,8 @@ class ThePlatformFeedIE(ThePlatformBaseIE):
         first_video_id = None
         duration = None
         for item in entry['media$content']:
-            smil_url = item['plfile$url'] + '&format=SMIL&Tracking=true&Embedded=true&formats=MPEG4,F4M'
-            cur_video_id = url_basename(smil_url)
+            smil_url = item['plfile$url'] + '&format=SMIL&mbr=true'
+            cur_video_id = ThePlatformIE._match_id(smil_url)
             if first_video_id is None:
                 first_video_id = cur_video_id
                 duration = float_or_none(item.get('plfile$duration'))