[pornhub:playlist] Improve extraction (closes #11594)

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

diff --git a/youtube_dl/extractor/amp.py b/youtube_dl/extractor/amp.py
index dcc3c97f146fc125b93a755b85a3c9c3a050b112..e8e40126baca4bad27f8593dd9bd026f16fad131 100644 (file)
--- a/youtube_dl/extractor/amp.py
+++ b/youtube_dl/extractor/amp.py
@@ -5,6 +5,8 @@ from .common import InfoExtractor
 from ..utils import (
     int_or_none,
     parse_iso8601,
+    mimetype2ext,
+    determine_ext,
 )
 
 
@@ -50,35 +52,38 @@ class AMPIE(InfoExtractor):
         if isinstance(media_content, dict):
             media_content = [media_content]
         for media_data in media_content:
-            media = media_data['@attributes']
-            media_type = media['type']
-            if media_type == 'video/f4m':
-                f4m_formats = self._extract_f4m_formats(
-                    media['url'] + '?hdcore=3.4.0&plugin=aasp-3.4.0.132.124',
-                    video_id, f4m_id='hds', fatal=False)
-                if f4m_formats:
-                    formats.extend(f4m_formats)
-            elif media_type == 'application/x-mpegURL':
-                m3u8_formats = self._extract_m3u8_formats(
-                    media['url'], video_id, 'mp4', m3u8_id='hls', fatal=False)
-                if m3u8_formats:
-                    formats.extend(m3u8_formats)
+            media = media_data.get('@attributes', {})
+            media_url = media.get('url')
+            if not media_url:
+                continue
+            ext = mimetype2ext(media.get('type')) or determine_ext(media_url)
+            if ext == 'f4m':
+                formats.extend(self._extract_f4m_formats(
+                    media_url + '?hdcore=3.4.0&plugin=aasp-3.4.0.132.124',
+                    video_id, f4m_id='hds', fatal=False))
+            elif ext == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(
+                    media_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
             else:
                 formats.append({
-                    'format_id': media_data['media-category']['@attributes']['label'],
+                    'format_id': media_data.get('media-category', {}).get('@attributes', {}).get('label'),
                     'url': media['url'],
                     'tbr': int_or_none(media.get('bitrate')),
                     'filesize': int_or_none(media.get('fileSize')),
+                    'ext': ext,
                 })
 
         self._sort_formats(formats)
 
+        timestamp = parse_iso8601(item.get('pubDate'), ' ') or parse_iso8601(item.get('dc-date'))
+
         return {
             'id': video_id,
             'title': get_media_node('title'),
             'description': get_media_node('description'),
             'thumbnails': thumbnails,
-            'timestamp': parse_iso8601(item.get('pubDate'), ' '),
+            'timestamp': timestamp,
             'duration': int_or_none(media_content[0].get('@attributes', {}).get('duration')),
+            'subtitles': subtitles,
             'formats': formats,
         }