Merge 3091daa8345566b6b7b26d9b505e6fde01059333 into da7223d4aa42ff9fc680b0951d043dd03cec2d30

* improve extraction of global "useful data" Array from player JS
* also handle tv-player and add tests: thx seproDev (yt-dlp/yt-dlp#12684)

Co-Authored-By: sepro <sepro@sepr0.com>

test/test_youtube_signature.py

@@ -232,8 +232,32 @@ _NSIG_TESTS = [
         'W9HJZKktxuYoDTqW', 'jHbbkcaxm54',
     ),
     (
-        'https://www.youtube.com/s/player/91201489/player_ias_tce.vflset/en_US/base.js',
+        'https://www.youtube.com/s/player/643afba4/player_ias.vflset/en_US/base.js',
-        'W9HJZKktxuYoDTqW', 'U48vOZHaeYS6vO',
+        'W9HJZKktxuYoDTqW', 'larxUlagTRAcSw',
+    ),
+    (
+        'https://www.youtube.com/s/player/e7567ecf/player_ias_tce.vflset/en_US/base.js',
+        'Sy4aDGc0VpYRR9ew_', '5UPOT1VhoZxNLQ',
+    ),
+    (
+        'https://www.youtube.com/s/player/d50f54ef/player_ias_tce.vflset/en_US/base.js',
+        'Ha7507LzRmH3Utygtj', 'XFTb2HoeOE5MHg',
+    ),
+    (
+        'https://www.youtube.com/s/player/074a8365/player_ias_tce.vflset/en_US/base.js',
+        'Ha7507LzRmH3Utygtj', 'ufTsrE0IVYrkl8v',
+    ),
+    (
+        'https://www.youtube.com/s/player/643afba4/player_ias.vflset/en_US/base.js',
+        'N5uAlLqm0eg1GyHO', 'dCBQOejdq5s-ww',
+    ),
+    (
+        'https://www.youtube.com/s/player/69f581a5/tv-player-ias.vflset/tv-player-ias.js',
+        '-qIP447rVlTTwaZjY', 'KNcGOksBAvwqQg',
+    ),
+    (
+        'https://www.youtube.com/s/player/643afba4/tv-player-ias.vflset/tv-player-ias.js',
+        'ir9-V6cdbCiyKxhr', '2PL7ZDYAALMfmA',
     ),
 ]
 

youtube_dl/extractor/newgrounds.py

@@ -5,7 +5,6 @@ import re
 from .common import InfoExtractor
 from ..utils import (
     extract_attributes,
-    int_or_none,
     parse_duration,
     parse_filesize,
     unified_timestamp,
@@ -27,15 +26,15 @@ class NewgroundsIE(InfoExtractor):
             'duration': 143,
         },
     }, {
-        'url': 'https://www.newgrounds.com/portal/view/673111',
+        'url': 'https://www.newgrounds.com/portal/view/850292',
-        'md5': '3394735822aab2478c31b1004fe5e5bc',
+        'md5': 'bb7cacf45e1b4d648e2dac2d79284d67',
         'info_dict': {
-            'id': '673111',
+            'id': '850292',
             'ext': 'mp4',
-            'title': 'Dancin',
+            'title': 'Timeless (2021)',
-            'uploader': 'Squirrelman82',
+            'uploader': 'Kevuhn',
-            'timestamp': 1460256780,
+            'timestamp': 1657896960,
-            'upload_date': '20160410',
+            'upload_date': '20220715',
         },
     }, {
         # source format unavailable, additional mp4 formats
@@ -44,7 +43,7 @@ class NewgroundsIE(InfoExtractor):
             'id': '689400',
             'ext': 'mp4',
             'title': 'ZTV News Episode 8',
-            'uploader': 'BennettTheSage',
+            'uploader': 'ZONE-SAMA',
             'timestamp': 1487965140,
             'upload_date': '20170224',
         },
@@ -58,40 +57,46 @@ class NewgroundsIE(InfoExtractor):
 
         webpage = self._download_webpage(url, media_id)
 
+        paths = url.split('/')
+        if paths[-3] == 'audio':
+            isAudio = True
+        else:
+            isAudio = False
+
+        if isAudio:
+            media_url = self._parse_json(self._search_regex(
+                r'"url"\s*:\s*("[^"]+"),', webpage, ''), media_id)
+
+            uploader = self._html_search_regex(
+                (r'(?s)<h4[^>]*>(.+?)</h4>.*?<em>\s*Author\s*</em>',
+                 r'(?:Author|Writer)\s*<a[^>]+>([^<]+)'), webpage, 'uploader',
+                fatal=False)
+
+            formats = [{
+                'url': media_url,
+                'format_id': 'source',
+                'quality': 1,
+            }]
+
+        else:
+            media_url = 'https://www.newgrounds.com/portal/video/' + media_id
+            media = self._download_json(media_url, media_id, headers={'X-Requested-With': 'XMLHttpRequest'})
+
+            uploader = media['author']
+
+            sources = media['sources']
+
+            formats = []
+            for source in sources:
+                for i in range(len(sources[source])):
+                    formats.append({
+                        'url': sources[source][i]['src'],
+                        'format_id': source,
+                        'height': int(source[:-2])  # 1080p -> 1080
+                    })
+
         title = self._html_search_regex(
             r'<title>([^>]+)</title>', webpage, 'title')
-
-        media_url = self._parse_json(self._search_regex(
-            r'"url"\s*:\s*("[^"]+"),', webpage, ''), media_id)
-
-        formats = [{
-            'url': media_url,
-            'format_id': 'source',
-            'quality': 1,
-        }]
-
-        max_resolution = int_or_none(self._search_regex(
-            r'max_resolution["\']\s*:\s*(\d+)', webpage, 'max resolution',
-            default=None))
-        if max_resolution:
-            url_base = media_url.rpartition('.')[0]
-            for resolution in (360, 720, 1080):
-                if resolution > max_resolution:
-                    break
-                formats.append({
-                    'url': '%s.%dp.mp4' % (url_base, resolution),
-                    'format_id': '%dp' % resolution,
-                    'height': resolution,
-                })
-
-        self._check_formats(formats, media_id)
-        self._sort_formats(formats)
-
-        uploader = self._html_search_regex(
-            (r'(?s)<h4[^>]*>(.+?)</h4>.*?<em>\s*Author\s*</em>',
-             r'(?:Author|Writer)\s*<a[^>]+>([^<]+)'), webpage, 'uploader',
-            fatal=False)
-
         timestamp = unified_timestamp(self._html_search_regex(
             (r'<dt>\s*Uploaded\s*</dt>\s*<dd>([^<]+</dd>\s*<dd>[^<]+)',
              r'<dt>\s*Uploaded\s*</dt>\s*<dd>([^<]+)'), webpage, 'timestamp',
@@ -99,7 +104,6 @@ class NewgroundsIE(InfoExtractor):
         duration = parse_duration(self._search_regex(
             r'(?s)<dd>\s*Song\s*</dd>\s*<dd>.+?</dd>\s*<dd>([^<]+)', webpage,
             'duration', default=None))
-
         filesize_approx = parse_filesize(self._html_search_regex(
             r'(?s)<dd>\s*Song\s*</dd>\s*<dd>(.+?)</dd>', webpage, 'filesize',
             default=None))
@@ -109,6 +113,9 @@ class NewgroundsIE(InfoExtractor):
         if '<dd>Song' in webpage:
             formats[0]['vcodec'] = 'none'
 
+        self._check_formats(formats, media_id)
+        self._sort_formats(formats)
+
         return {
             'id': media_id,
             'title': title,

youtube_dl/extractor/youtube.py

@@ -91,12 +91,12 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
             'INNERTUBE_CONTEXT': {
                 'client': {
                     'clientName': 'IOS',
-                    'clientVersion': '19.45.4',
+                    'clientVersion': '20.10.4',
                     'deviceMake': 'Apple',
                     'deviceModel': 'iPhone16,2',
-                    'userAgent': 'com.google.ios.youtube/19.45.4 (iPhone16,2; U; CPU iOS 18_1_0 like Mac OS X;)',
+                    'userAgent': 'com.google.ios.youtube/20.10.4 (iPhone16,2; U; CPU iOS 18_3_2 like Mac OS X;)',
                     'osName': 'iPhone',
-                    'osVersion': '18.1.0.22B83',
+                    'osVersion': '18.3.2.22D82',
                 },
             },
             'INNERTUBE_CONTEXT_CLIENT_NAME': 5,
@@ -109,7 +109,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
             'INNERTUBE_CONTEXT': {
                 'client': {
                     'clientName': 'MWEB',
-                    'clientVersion': '2.20241202.07.00',
+                    'clientVersion': '2.20250311.03.00',
                     # mweb previously did not require PO Token with this UA
                     'userAgent': 'Mozilla/5.0 (iPad; CPU OS 16_7_10 like Mac OS X) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/16.6 Mobile/15E148 Safari/604.1,gzip(gfe)',
                 },
@@ -122,7 +122,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
             'INNERTUBE_CONTEXT': {
                 'client': {
                     'clientName': 'TVHTML5',
-                    'clientVersion': '7.20250120.19.00',
+                    'clientVersion': '7.20250312.16.00',
                     'userAgent': 'Mozilla/5.0 (ChromiumStylePlatform) Cobalt/Version',
                 },
             },
@@ -133,7 +133,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
             'INNERTUBE_CONTEXT': {
                 'client': {
                     'clientName': 'WEB',
-                    'clientVersion': '2.20241126.01.00',
+                    'clientVersion': '2.20250312.04.00',
                 },
             },
             'INNERTUBE_CONTEXT_CLIENT_NAME': 1,
@@ -692,7 +692,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         'invidious': '|'.join(_INVIDIOUS_SITES),
     }
     _PLAYER_INFO_RE = (
-        r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})/player',
+        r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})//(?:tv-)?player',
         r'/(?P<id>[a-zA-Z0-9_-]{8,})/player(?:_ias\.vflset(?:/[a-zA-Z]{2,3}_[a-zA-Z]{2,3})?|-plasma-ias-(?:phone|tablet)-[a-z]{2}_[A-Z]{2}\.vflset)/base\.js$',
         r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.js$',
     )
@@ -1857,7 +1857,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
     def _extract_n_function_code_jsi(self, video_id, jsi, player_id=None):
 
         var_ay = self._search_regex(
-            r'(?:[;\s]|^)\s*(var\s*[\w$]+\s*=\s*"[^"]+"\s*\.\s*split\("\{"\))(?=\s*[,;])',
+            r'(?:[;\s]|^)\s*(var\s*[\w$]+\s*=\s*"(?:\\"|[^"])+"\s*\.\s*split\("\W+"\))(?=\s*[,;])',
             jsi.code, 'useful values', default='')
 
         func_name = self._extract_n_function_name(jsi.code)