[RedTube] Update for 2023

* URLs need host://domain default
* use traverse_obj()

youtube_dl/extractor/redtube.py

@@ -1,42 +1,36 @@
+# coding: utf-8
 from __future__ import unicode_literals
 
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_str
 from ..utils import (
     determine_ext,
     ExtractorError,
     int_or_none,
+    join_nonempty,
     merge_dicts,
     str_to_int,
+    T,
+    traverse_obj,
     unified_strdate,
     url_or_none,
+    urljoin,
 )
 
 
 class RedTubeIE(InfoExtractor):
     _VALID_URL = r'https?://(?:(?:\w+\.)?redtube\.com/|embed\.redtube\.com/\?.*?\bid=)(?P<id>[0-9]+)'
+    _EMBED_REGEX = [r'<iframe[^>]+?src=["\'](?P<url>(?:https?:)?//embed\.redtube\.com/\?.*?\bid=\d+)']
     _TESTS = [{
-        'url': 'http://www.redtube.com/66418',
-        'md5': 'fc08071233725f26b8f014dba9590005',
-        'info_dict': {
-            'id': '66418',
-            'ext': 'mp4',
-            'title': 'Sucked on a toilet',
-            'upload_date': '20110811',
-            'duration': 596,
-            'view_count': int,
-            'age_limit': 18,
-        },
-        'skip': 'private video',
-    }, {
         'url': 'https://www.redtube.com/38864951',
         'md5': '4fba70cbca3aefd25767ab4b523c9878',
         'info_dict': {
             'id': '38864951',
             'ext': 'mp4',
             'title': 'Public Sex on the Balcony in Freezing Paris! Amateur Couple LeoLulu',
-            'description': 'Watch video Public Sex on the Balcony in Freezing Paris! Amateur Couple LeoLulu on Redtube, home of free Blowjob porn videos and Blonde sex movies online. Video length: (10:46) - Uploaded by leolulu - Verified User - Starring Pornstar: Leolulu',
+            'description': 'Watch video Public Sex on the Balcony in Freezing Paris! Amateur Couple LeoLulu on Redtube, home of free Blowjob porn videos and Blonde sex movies online. Video length: (10:46) - Uploaded by leolulu - Verified User - Starring Pornstar: LeoLulu',
             'upload_date': '20210111',
             'timestamp': 1610343109,
             'duration': 646,
@@ -52,11 +46,11 @@ class RedTubeIE(InfoExtractor):
         'only_matching': True,
     }]
 
-    @staticmethod
+    @classmethod
-    def _extract_urls(webpage):
+    def _extract_urls(cls, webpage):
-        return re.findall(
+        for embed_re in cls._EMBED_REGEX:
-            r'<iframe[^>]+?src=["\'](?P<url>(?:https?:)?//embed\.redtube\.com/\?.*?\bid=\d+)',
+            for from_ in re.findall(embed_re, webpage):
-            webpage)
+                yield from_
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
@@ -87,50 +81,53 @@ class RedTubeIE(InfoExtractor):
             self._search_regex(
                 r'sources\s*:\s*({.+?})', webpage, 'source', default='{}'),
             video_id, fatal=False)
-        if sources and isinstance(sources, dict):
+
-            for format_id, format_url in sources.items():
+        def full_url(u):
-                if format_url:
+            return urljoin(url, u)
-                    formats.append({
+
-                        'url': format_url,
+        for fmt in traverse_obj(sources, (T(dict.items), {
-                        'format_id': format_id,
+                'url': (1, T(full_url)),
-                        'height': int_or_none(format_id),
+                'format_id': (2, T(compat_str)),
-                    })
+                'height': (2, T(int_or_none)), })):
-        medias = self._parse_json(
+            if 'url' in fmt:
-            self._search_regex(
+                formats.append(fmt)
-                r'mediaDefinition["\']?\s*:\s*(\[.+?}\s*\])', webpage,
+
-                'media definitions', default='{}'),
+        medias = self._search_regex(
-            video_id, fatal=False)
+            r'''mediaDefinitions?["']?\s*:\s*(\[[\s\S]+?}\s*\])''', webpage,
-        for media in medias if isinstance(medias, list) else []:
+            'media definitions', default='{}')
-            format_url = url_or_none(media.get('videoUrl'))
+        medias = self._parse_json(medias, video_id, fatal=False)
+        for fmt in traverse_obj(medias, (Ellipsis, T(dict))):
+            format_url = full_url(fmt.get('videoUrl'))
             if not format_url:
                 continue
-            format_id = media.get('format')
+            more_media = None
-            quality = media.get('quality')
+            if fmt['format'] == 'hls' or (fmt['format'] == 'mp4' and not fmt.get('quality')):
-            if format_id == 'hls' or (format_id == 'mp4' and not quality):
                 more_media = self._download_json(format_url, video_id, fatal=False)
-            else:
+            if more_media is None:
-                more_media = [media]
+                more_media = [fmt]
-            for media in more_media if isinstance(more_media, list) else []:
+            for fmt in traverse_obj(more_media, (Ellipsis, {
-                format_url = url_or_none(media.get('videoUrl'))
+                    'url': ('videoUrl', T(full_url)),
+                    'ext': ('format', T(compat_str)),
+                    'format_id': ('quality', T(compat_str)), })):
+                format_url = fmt.get('url')
                 if not format_url:
                     continue
-                format_id = media.get('format')
+                if fmt.get('ext') == 'hls' or determine_ext(format_url) == 'm3u8':
-                if format_id == 'hls' or determine_ext(format_url) == 'm3u8':
                     formats.extend(self._extract_m3u8_formats(
                         format_url, video_id, 'mp4',
-                        entry_protocol='m3u8_native', m3u8_id=format_id or 'hls',
+                        entry_protocol='m3u8_native', m3u8_id='hls',
                         fatal=False))
-                else:
+                    continue
-                    format_id = media.get('quality')
+                fmt['height'] = int_or_none(fmt.get('format_id'))
-                    formats.append({
+                fmt['format_id'] = join_nonempty('ext', 'format_id', from_dict=fmt)
-                        'url': format_url,
+                formats.append(fmt)
-                        'format_id': format_id,
-                        'height': int_or_none(format_id),
-                    })
         if not formats:
-            video_url = self._html_search_regex(
+            video_url = url_or_none(self._html_search_regex(
-                r'<source src="(.+?)" type="video/mp4">', webpage, 'video URL')
+                r'<source src="(.+?)" type="video/mp4">', webpage, 'video URL'))
+            if video_url:
                 formats.append({'url': video_url})
+
+        self._check_formats(formats, video_id)
         self._sort_formats(formats)
 
         thumbnail = self._og_search_thumbnail(webpage)