Merge ddbd5451010e00c2ce64e57ce1d03ddbcd8ab4ff 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_download.py

@@ -69,10 +69,10 @@ def _file_md5(fn):
         return hashlib.md5(f.read()).hexdigest()
 
 
-defs = gettestcases()
+ie_testcases = gettestcases()
 
 
-class TestDownload(unittest.TestCase):
+class BaseDownloadTCase(unittest.TestCase):
     # Parallel testing in nosetests. See
     # http://nose.readthedocs.org/en/latest/doc_tests/test_multiprocess/multiprocess.html
     _multiprocess_shared_ = True
@@ -91,8 +91,9 @@ class TestDownload(unittest.TestCase):
                                strclass(self.__class__),
                                ' [%s]' % add_ie if add_ie else '')
 
-    def setUp(self):
+
-        self.defs = defs
+TestDownload = unittest.TestSuite()
+
 
 # Dynamically generate tests
 
@@ -275,20 +276,42 @@ def generator(test_case, tname):
     return test_template
 
 
+cache = {}
+
 # And add them to TestDownload
-for n, test_case in enumerate(defs):
+for test_case in ie_testcases:
-    tname = 'test_' + str(test_case['name'])
+    # Get or create a sub-test for the extractor file
-    i = 1
+    module_name = test_case.get('module_name', 'unknown').rsplit('.', 1)[-1]  # type: str
-    while hasattr(TestDownload, tname):
+    extractor_file_name = str('test_%s' % module_name)
-        tname = 'test_%s_%d' % (test_case['name'], i)
+    extractor_file_suite = getattr(TestDownload, extractor_file_name, None)
+    if extractor_file_suite is None:
+        extractor_file_suite = type(extractor_file_name, (unittest.TestSuite,), {})()
+        setattr(TestDownload, extractor_file_name, extractor_file_suite)
+        TestDownload.addTest(extractor_file_suite)
+
+    # Get or create a sub-test for the info extractor
+    # This class contains the actual tests
+    extractor_class_name = str('test_%s' % test_case['name'])
+    ExtractorClass = getattr(extractor_file_suite, extractor_class_name, None)
+    if ExtractorClass is None:
+        ExtractorClass = type(extractor_class_name, (BaseDownloadTCase,), {})
+        setattr(extractor_file_suite, extractor_class_name, ExtractorClass)
+
+    i = 0
+    tname = 'test_%d' % i
+    while hasattr(ExtractorClass, tname):
+        tname = 'test_%d' % i
         i += 1
     test_method = generator(test_case, tname)
     test_method.__name__ = str(tname)
     ie_list = test_case.get('add_ie')
     test_method.add_ie = ie_list and ','.join(ie_list)
-    setattr(TestDownload, test_method.__name__, test_method)
+
-    del test_method
+    setattr(ExtractorClass, test_method.__name__, test_method)
+    extractor_file_suite.addTest(ExtractorClass(test_method.__name__))
+
+    del test_method, test_case
 
 
 if __name__ == '__main__':
-    unittest.main()
+    unittest.main(argv=['TestDownload'])

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/common.py

@@ -3263,6 +3263,7 @@ class InfoExtractor(object):
             if not include_onlymatching and t.get('only_matching', False):
                 continue
             t['name'] = type(self).__name__[:-len('IE')]
+            t['module_name'] = type(self).__module__
             yield t
 
     def is_suitable(self, age_limit):

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)