Merge 5aee29b7ca489428ca129a89b09a3a0390740466 into 2b4fbfce25902d557b86b003cf48f738129efce4

thx seproDev, bashonly: yt-dlp/yt-dlp#12748

test/test_youtube_signature.py

@@ -84,6 +84,21 @@ _SIG_TESTS = [
         '2aq0aqSyOoJXtK73m-uME_jv7-pT15gOFC02RFkGMqWpzEICs69VdbwQ0LDp1v7j8xx92efCJlYFYb1sUkkBSPOlPmXgIARw8JQ0qOAOAA',
         '0QJ8wRAIgXmPlOPSBkkUs1bYFYlJCfe29xxAj7v1pDL0QwbdV96sCIEzpWqMGkFR20CFOg51Tp-7vj_EMu-m37KtXJ2OySqa0q',
     ),
+    (
+        'https://www.youtube.com/s/player/643afba4/tv-player-ias.vflset/tv-player-ias.js',
+        '2aq0aqSyOoJXtK73m-uME_jv7-pT15gOFC02RFkGMqWpzEICs69VdbwQ0LDp1v7j8xx92efCJlYFYb1sUkkBSPOlPmXgIARw8JQ0qOAOAA',
+        'AAOAOq0QJ8wRAIgXmPlOPSBkkUs1bYFYlJCfe29xx8j7vgpDL0QwbdV06sCIEzpWqMGkFR20CFOS21Tp-7vj_EMu-m37KtXJoOy1',
+    ),
+    (
+        'https://www.youtube.com/s/player/363db69b/player_ias.vflset/en_US/base.js',
+        '2aq0aqSyOoJXtK73m-uME_jv7-pT15gOFC02RFkGMqWpzEICs69VdbwQ0LDp1v7j8xx92efCJlYFYb1sUkkBSPOlPmXgIARw8JQ0qOAOAA',
+        '0aqSyOoJXtK73m-uME_jv7-pT15gOFC02RFkGMqWpz2ICs6EVdbwQ0LDp1v7j8xx92efCJlYFYb1sUkkBSPOlPmXgIARw8JQ0qOAOAA',
+    ),
+    (
+        'https://www.youtube.com/s/player/4fcd6e4a/player_ias.vflset/en_US/base.js',
+        '2aq0aqSyOoJXtK73m-uME_jv7-pT15gOFC02RFkGMqWpzEICs69VdbwQ0LDp1v7j8xx92efCJlYFYb1sUkkBSPOlPmXgIARw8JQ0qOAOAA',
+        'wAOAOq0QJ8ARAIgXmPlOPSBkkUs1bYFYlJCfe29xx8q7v1pDL0QwbdV96sCIEzpWqMGkFR20CFOg51Tp-7vj_EMu-m37KtXJoOySqa0',
+    ),
 ]
 
 _NSIG_TESTS = [
@@ -153,7 +168,7 @@ _NSIG_TESTS = [
     ),
     (
         'https://www.youtube.com/s/player/c57c113c/player_ias.vflset/en_US/base.js',
-        '-Txvy6bT5R6LqgnQNx', 'dcklJCnRUHbgSg',
+        'M92UUMHa8PdvPd3wyM', '3hPqLJsiNZx7yA',
     ),
     (
         'https://www.youtube.com/s/player/5a3b6271/player_ias.vflset/en_US/base.js',
@@ -173,7 +188,7 @@ _NSIG_TESTS = [
     ),
     (
         'https://www.youtube.com/s/player/cfa9e7cb/player_ias.vflset/en_US/base.js',
-        'qO0NiMtYQ7TeJnfFG2', 'k9cuJDHNS5O7kQ',
+        'aCi3iElgd2kq0bxVbQ', 'QX1y8jGb2IbZ0w',
     ),
     (
         'https://www.youtube.com/s/player/8c7583ff/player_ias.vflset/en_US/base.js',
@@ -231,10 +246,6 @@ _NSIG_TESTS = [
         'https://www.youtube.com/s/player/f6e09c70/player_ias_tce.vflset/en_US/base.js',
         'W9HJZKktxuYoDTqW', 'jHbbkcaxm54',
     ),
-    (
-        'https://www.youtube.com/s/player/643afba4/player_ias.vflset/en_US/base.js',
-        'W9HJZKktxuYoDTqW', 'larxUlagTRAcSw',
-    ),
     (
         'https://www.youtube.com/s/player/e7567ecf/player_ias_tce.vflset/en_US/base.js',
         'Sy4aDGc0VpYRR9ew_', '5UPOT1VhoZxNLQ',
@@ -259,6 +270,22 @@ _NSIG_TESTS = [
         'https://www.youtube.com/s/player/643afba4/tv-player-ias.vflset/tv-player-ias.js',
         'ir9-V6cdbCiyKxhr', '2PL7ZDYAALMfmA',
     ),
+    (
+        'https://www.youtube.com/s/player/643afba4/player_ias.vflset/en_US/base.js',
+        'W9HJZKktxuYoDTqW', 'larxUlagTRAcSw',
+    ),
+    (
+        'https://www.youtube.com/s/player/363db69b/player_ias.vflset/en_US/base.js',
+        'eWYu5d5YeY_4LyEDc', 'XJQqf-N7Xra3gg',
+    ),
+    (
+        'https://www.youtube.com/s/player/4fcd6e4a/player_ias.vflset/en_US/base.js',
+        'o_L251jm8yhZkWtBW', 'lXoxI3XvToqn6A',
+    ),
+    (
+        'https://www.youtube.com/s/player/4fcd6e4a/tv-player-ias.vflset/tv-player-ias.js',
+        'o_L251jm8yhZkWtBW', 'lXoxI3XvToqn6A',
+    ),
 ]
 
 
@@ -271,6 +298,8 @@ class TestPlayerInfo(unittest.TestCase):
             ('https://www.youtube.com/s/player/64dddad9/player-plasma-ias-phone-en_US.vflset/base.js', '64dddad9'),
             ('https://www.youtube.com/s/player/64dddad9/player-plasma-ias-phone-de_DE.vflset/base.js', '64dddad9'),
             ('https://www.youtube.com/s/player/64dddad9/player-plasma-ias-tablet-en_US.vflset/base.js', '64dddad9'),
+            ('https://www.youtube.com/s/player/e7567ecf/player_ias_tce.vflset/en_US/base.js', 'e7567ecf'),
+            ('https://www.youtube.com/s/player/643afba4/tv-player-ias.vflset/tv-player-ias.js', '643afba4'),
             # obsolete
             ('https://www.youtube.com/yts/jsbin/player_ias-vfle4-e03/en_US/base.js', 'vfle4-e03'),
             ('https://www.youtube.com/yts/jsbin/player_ias-vfl49f_g4/en_US/base.js', 'vfl49f_g4'),
@@ -280,8 +309,9 @@ class TestPlayerInfo(unittest.TestCase):
             ('https://s.ytimg.com/yts/jsbin/html5player-en_US-vflXGBaUN.js', 'vflXGBaUN'),
             ('https://s.ytimg.com/yts/jsbin/html5player-en_US-vflKjOTVq/html5player.js', 'vflKjOTVq'),
         )
+        ie = YoutubeIE(FakeYDL({'cachedir': False}))
         for player_url, expected_player_id in PLAYER_URLS:
-            player_id = YoutubeIE._extract_player_info(player_url)
+            player_id = ie._extract_player_info(player_url)
             self.assertEqual(player_id, expected_player_id)
 
 
@@ -301,8 +331,8 @@ class TestSignature(unittest.TestCase):
 def t_factory(name, sig_func, url_pattern):
     def make_tfunc(url, sig_input, expected_sig):
         m = url_pattern.match(url)
-        assert m, '%r should follow URL format' % url
-        test_id = m.group('id')
+        assert m, '{0!r} should follow URL format'.format(url)
+        test_id = re.sub(r'[/.-]', '_', m.group('id') or m.group('compat_id'))
 
         def test_func(self):
             basename = 'player-{0}-{1}.js'.format(name, test_id)
@@ -335,12 +365,16 @@ def n_sig(jscode, sig_input):
 
 
 make_sig_test = t_factory(
-    'signature', signature, re.compile(r'.*(?:-|/player/)(?P<id>[a-zA-Z0-9_-]+)(?:/.+\.js|(?:/watch_as3|/html5player)?\.[a-z]+)$'))
+    'signature', signature,
+    re.compile(r'''(?x)
+        .+/(?P<h5>html5)?player(?(h5)(?:-en_US)?-|/)(?P<id>[a-zA-Z0-9/._-]+)
+        (?(h5)/(?:watch_as3|html5player))?\.js$
+    '''))
 for test_spec in _SIG_TESTS:
     make_sig_test(*test_spec)
 
 make_nsig_test = t_factory(
-    'nsig', n_sig, re.compile(r'.+/player/(?P<id>[a-zA-Z0-9_-]+)/.+.js$'))
+    'nsig', n_sig, re.compile(r'.+/player/(?P<id>[a-zA-Z0-9_/.-]+)\.js$'))
 for test_spec in _NSIG_TESTS:
     make_nsig_test(*test_spec)
 

youtube_dl/extractor/senateisvp.py

@@ -5,48 +5,52 @@ import re
 from .common import InfoExtractor
 from ..utils import (
     ExtractorError,
+    parse_duration,
     unsmuggle_url,
+    url_or_none,
 )
-from ..compat import (
-    compat_parse_qs,
-    compat_urlparse,
-)
+from ..compat import compat_parse_qs
 
 
 class SenateISVPIE(InfoExtractor):
+
+    # [committee, stream_number, stream_domain, stream_id, msl3]
     _COMM_MAP = [
-        ['ag', '76440', 'http://ag-f.akamaihd.net'],
-        ['aging', '76442', 'http://aging-f.akamaihd.net'],
-        ['approps', '76441', 'http://approps-f.akamaihd.net'],
-        ['armed', '76445', 'http://armed-f.akamaihd.net'],
-        ['banking', '76446', 'http://banking-f.akamaihd.net'],
-        ['budget', '76447', 'http://budget-f.akamaihd.net'],
-        ['cecc', '76486', 'http://srs-f.akamaihd.net'],
-        ['commerce', '80177', 'http://commerce1-f.akamaihd.net'],
-        ['csce', '75229', 'http://srs-f.akamaihd.net'],
-        ['dpc', '76590', 'http://dpc-f.akamaihd.net'],
-        ['energy', '76448', 'http://energy-f.akamaihd.net'],
-        ['epw', '76478', 'http://epw-f.akamaihd.net'],
-        ['ethics', '76449', 'http://ethics-f.akamaihd.net'],
-        ['finance', '76450', 'http://finance-f.akamaihd.net'],
-        ['foreign', '76451', 'http://foreign-f.akamaihd.net'],
-        ['govtaff', '76453', 'http://govtaff-f.akamaihd.net'],
-        ['help', '76452', 'http://help-f.akamaihd.net'],
-        ['indian', '76455', 'http://indian-f.akamaihd.net'],
-        ['intel', '76456', 'http://intel-f.akamaihd.net'],
-        ['intlnarc', '76457', 'http://intlnarc-f.akamaihd.net'],
-        ['jccic', '85180', 'http://jccic-f.akamaihd.net'],
-        ['jec', '76458', 'http://jec-f.akamaihd.net'],
-        ['judiciary', '76459', 'http://judiciary-f.akamaihd.net'],
-        ['rpc', '76591', 'http://rpc-f.akamaihd.net'],
-        ['rules', '76460', 'http://rules-f.akamaihd.net'],
-        ['saa', '76489', 'http://srs-f.akamaihd.net'],
-        ['smbiz', '76461', 'http://smbiz-f.akamaihd.net'],
-        ['srs', '75229', 'http://srs-f.akamaihd.net'],
-        ['uscc', '76487', 'http://srs-f.akamaihd.net'],
-        ['vetaff', '76462', 'http://vetaff-f.akamaihd.net'],
-        ['arch', '', 'http://ussenate-f.akamaihd.net/']
+        ['ag', '76440', 'http://ag-f.akamaihd.net', '2036803', 'agriculture'],
+        ['aging', '76442', 'http://aging-f.akamaihd.net', '2036801', 'aging'],
+        ['approps', '76441', 'http://approps-f.akamaihd.net', '2036802', 'appropriations'],
+        ['armed', '76445', 'http://armed-f.akamaihd.net', '2036800', 'armedservices'],
+        ['banking', '76446', 'http://banking-f.akamaihd.net', '2036799', 'banking'],
+        ['budget', '76447', 'http://budget-f.akamaihd.net', '2036798', 'budget'],
+        ['cecc', '76486', 'http://srs-f.akamaihd.net', '2036782', 'srs_cecc'],
+        ['commerce', '80177', 'http://commerce1-f.akamaihd.net', '2036779', 'commerce'],
+        ['csce', '75229', 'http://srs-f.akamaihd.net', '2036777', 'srs_srs'],
+        ['dpc', '76590', 'http://dpc-f.akamaihd.net', None, 'dpc'],
+        ['energy', '76448', 'http://energy-f.akamaihd.net', '2036797', 'energy'],
+        ['epw', '76478', 'http://epw-f.akamaihd.net', '2036783', 'environment'],
+        ['ethics', '76449', 'http://ethics-f.akamaihd.net', '2036796', 'ethics'],
+        ['finance', '76450', 'http://finance-f.akamaihd.net', '2036795', 'finance_finance'],
+        ['foreign', '76451', 'http://foreign-f.akamaihd.net', '2036794', 'foreignrelations'],
+        ['govtaff', '76453', 'http://govtaff-f.akamaihd.net', '2036792', 'hsgac'],
+        ['help', '76452', 'http://help-f.akamaihd.net', '2036793', 'help'],
+        ['indian', '76455', 'http://indian-f.akamaihd.net', '2036791', 'indianaffairs'],
+        ['intel', '76456', 'http://intel-f.akamaihd.net', '2036790', 'intelligence'],
+        ['intlnarc', '76457', 'http://intlnarc-f.akamaihd.net', None, 'internationalnarcoticscaucus'],
+        ['jccic', '85180', 'http://jccic-f.akamaihd.net', '2036778', 'jccic'],
+        ['jec', '76458', 'http://jec-f.akamaihd.net', '2036789', 'jointeconomic'],
+        ['judiciary', '76459', 'http://judiciary-f.akamaihd.net', '2036788', 'judiciary'],
+        ['rpc', '76591', 'http://rpc-f.akamaihd.net', None, 'rpc'],
+        ['rules', '76460', 'http://rules-f.akamaihd.net', '2036787', 'rules'],
+        ['saa', '76489', 'http://srs-f.akamaihd.net', '2036780', 'srs_saa'],
+        ['smbiz', '76461', 'http://smbiz-f.akamaihd.net', '2036786', 'smallbusiness'],
+        ['srs', '75229', 'http://srs-f.akamaihd.net', '2031966', 'srs_srs'],
+        ['uscc', '76487', 'http://srs-f.akamaihd.net', '2036781', 'srs_uscc'],
+        ['vetaff', '76462', 'http://vetaff-f.akamaihd.net', '2036785', 'veteransaffairs'],
+        ['arch', '', 'http://ussenate-f.akamaihd.net/', None, None],
+        ['uscp', None, '', '2043685', 'uscp'],
+        ['cio', None, '', '2043686', 'cio']
     ]
+
     _IE_NAME = 'senate.gov'
     _VALID_URL = r'https?://(?:www\.)?senate\.gov/isvp/?\?(?P<qs>.+)'
     _TESTS = [{
@@ -54,100 +58,120 @@ class SenateISVPIE(InfoExtractor):
         'info_dict': {
             'id': 'judiciary031715',
             'ext': 'mp4',
-            'title': 'Integrated Senate Video Player',
-            'thumbnail': r're:^https?://.*\.(?:jpg|png)$',
-        },
-        'params': {
-            # m3u8 download
-            'skip_download': True,
-        },
-    }, {
-        'url': 'http://www.senate.gov/isvp/?type=live&comm=commerce&filename=commerce011514.mp4&auto_play=false',
-        'info_dict': {
-            'id': 'commerce011514',
-            'ext': 'mp4',
-            'title': 'Integrated Senate Video Player'
-        },
-        'params': {
-            # m3u8 download
-            'skip_download': True,
+            'title': 'judiciary031715',
+            'thumbnail': 'http://www.judiciary.senate.gov/themes/judiciary/images/video-poster-flash-fit.png',
         },
+        'expected_warnings': ['Failed to download m3u8 information: HTTP Error 404: Not Found'],
     }, {
         'url': 'http://www.senate.gov/isvp/?type=arch&comm=intel&filename=intel090613&hc_location=ufi',
-        # checksum differs each time
         'info_dict': {
             'id': 'intel090613',
             'ext': 'mp4',
-            'title': 'Integrated Senate Video Player'
-        }
+            'title': 'intel090613',
+        },
+        'expected_warnings': ['Failed to download m3u8 information: HTTP Error 404: Not Found'],
     }, {
-        # From http://www.c-span.org/video/?96791-1
-        'url': 'http://www.senate.gov/isvp?type=live&comm=banking&filename=banking012715',
-        'only_matching': True,
+        'url': 'https://www.senate.gov/isvp/?comm=govtaff&type=archv&stt=975&filename=govtaff111722&auto_play=false&poster=https%3A%2F%2Fwww%2Ehsgac%2Esenate%2Egov%2Fimages%2Fvideo%2Dposter%2Dflash%2Dfit%2Epng',
+        'info_dict': {
+            'id': 'govtaff111722',
+            'ext': 'mp4',
+            'title': 'govtaff111722',
+            'thumbnail': 'https://www.hsgac.senate.gov/images/video-poster-flash-fit.png',
+        },
+    }, {
+        'url': 'https://www.senate.gov/isvp/?type=arch&comm=energy&filename=energy111722&stt=00:22:30&auto_play=false&wmode=transparent&poster=https%3A%2F%2Fwww%2Eenergy%2Esenate%2Egov%2Fthemes%2Fenergy%2Fimages%2Fvideo%2Dposter%2Dflash%2Dfit%2Epng',
+        'info_dict': {
+            'id': 'energy111722',
+            'ext': 'mp4',
+            'title': 'energy111722',
+            'thumbnail': 'https://www.energy.senate.gov/themes/energy/images/video-poster-flash-fit.png',
+        },
+    }, {
+        'url': 'https://www.senate.gov/isvp/?comm=foreign&type=archv&stt=0&filename=foreign080322&auto_play=false&wmode=transparent&poster=https%3A%2F%2Fwww%2Eforeign%2Esenate%2Egov%2Fthemes%2Fforeign%2Fimages%2Fvideo%2Dposter%2Dflash%2Dfit%2Epng',
+        'info_dict': {
+            'id': 'foreign080322',
+            'ext': 'mp4',
+            'title': 'foreign080322',
+            'thumbnail': 'https://www.foreign.senate.gov/themes/foreign/images/video-poster-flash-fit.png',
+        },
     }]
 
     @staticmethod
+    # returns url from an iframe
     def _search_iframe_url(webpage):
         mobj = re.search(
-            r"<iframe[^>]+src=['\"](?P<url>https?://www\.senate\.gov/isvp/?\?[^'\"]+)['\"]",
+            r'''<iframe\b[^>]+\bsrc\s*=\s*(['"])(?P<url>https?://www\.senate\.gov/isvp/?\?(?:(?!\1)\S)+)''',
             webpage)
         if mobj:
             return mobj.group('url')
 
+    @staticmethod
+    def __dictify(cm):
+        return dict((row[0], (row[1:] + ['', ''])[:4]) for row in cm)
+
+    # returns stream_number, stream_domain, stream_id, msl3
     def _get_info_for_comm(self, committee):
-        for entry in self._COMM_MAP:
-            if entry[0] == committee:
-                return entry[1:]
+        dict = self.__dictify(self._COMM_MAP)
+        return dict[committee][0:]
 
     def _real_extract(self, url):
+        # smuggled data may contain a forced title that should be used
         url, smuggled_data = unsmuggle_url(url, {})
-
         qs = compat_parse_qs(re.match(self._VALID_URL, url).group('qs'))
-        if not qs.get('filename') or not qs.get('type') or not qs.get('comm'):
-            raise ExtractorError('Invalid URL', expected=True)
 
-        video_id = re.sub(r'.mp4$', '', qs['filename'][0])
+        # error handling for invalid URL - specify which error
+        if not qs.get('filename'):
+            raise ExtractorError('Invalid URL. Missing filename in query parameters', expected=True)
+        if not qs.get('comm'):
+            raise ExtractorError('Invalid URL. Missing committee in query parameters', expected=True)
 
-        webpage = self._download_webpage(url, video_id)
+        committee = qs.get('comm')[0]
+        filename = qs.get('filename')[0]
+        video_id = re.sub(r'\.mp4$', '', filename)
 
-        if smuggled_data.get('force_title'):
-            title = smuggled_data['force_title']
-        else:
-            title = self._html_search_regex(r'<title>([^<]+)</title>', webpage, video_id)
-        poster = qs.get('poster')
-        thumbnail = poster[0] if poster else None
+        # there is no point in pulling the title from the webpage since it always defaults to 'Integrated Senate Player'
+        title = smuggled_data.get('force_title') or filename
 
-        video_type = qs['type'][0]
-        committee = video_type if video_type == 'arch' else qs['comm'][0]
-        stream_num, domain = self._get_info_for_comm(committee)
+        stream_number, stream_domain, stream_id, msl3 = self._get_info_for_comm(committee)
+        stream_number = int(stream_number)
+        stream_id = int(stream_id)
 
+        # the possible locations for the video: only the first has been seen in use
+        possible_manifest_urls = [
+            'https://www-senate-gov-media-srs.akamaized.net/hls/live/%d/%s/%s/master.m3u8' % (stream_id, committee, filename),
+            'https://www-senate-gov-msl3archive.akamaized.net/%s/%s_1/master.m3u8' % (msl3, filename),
+            '%s/i/%s_1@%d/master.m3u8' % (stream_domain, filename, stream_number),
+            'https://ussenate-f.akamaihd.net/i/%s.mp4/master.m3u8' % video_id,
+        ]
+
+        # we iterate through the possible locations until we find formats
         formats = []
-        if video_type == 'arch':
-            filename = video_id if '.' in video_id else video_id + '.mp4'
-            formats = [{
-                # All parameters in the query string are necessary to prevent a 403 error
-                'url': compat_urlparse.urljoin(domain, filename) + '?v=3.1.0&fp=&r=&g=',
-            }]
-        else:
-            hdcore_sign = 'hdcore=3.1.0'
-            url_params = (domain, video_id, stream_num)
-            f4m_url = '%s/z/%s_1@%s/manifest.f4m?' % url_params + hdcore_sign
-            m3u8_url = '%s/i/%s_1@%s/master.m3u8' % url_params
-            for entry in self._extract_f4m_formats(f4m_url, video_id, f4m_id='f4m'):
-                # URLs without the extra param induce an 404 error
-                entry.update({'extra_param_to_segment_url': hdcore_sign})
-                formats.append(entry)
-            for entry in self._extract_m3u8_formats(m3u8_url, video_id, ext='mp4', m3u8_id='m3u8'):
-                mobj = re.search(r'(?P<tag>(?:-p|-b)).m3u8', entry['url'])
+        for url in possible_manifest_urls:
+            entries = self._extract_m3u8_formats(
+                url, video_id, ext='mp4', m3u8_id='hls',
+                entry_protocol='m3u8_native', fatal=False)
+
+            for entry in entries:
+                mobj = re.search(r'(?P<tag>-[pb]).m3u8', entry['url'])
                 if mobj:
                     entry['format_id'] += mobj.group('tag')
                 formats.append(entry)
 
-            self._sort_formats(formats)
+            if formats:
+                break
+
+        self._sort_formats(formats)
+        thumbnail = url_or_none(qs.get('poster', [None])[-1])
+        start_time = parse_duration(qs.get('stt', [None])[-1])
+        stop_time = parse_duration(qs.get('dur', [None])[-1])
+        if stop_time is not None:
+            stop_time += start_time or 0
 
         return {
             'id': video_id,
             'title': title,
             'formats': formats,
             'thumbnail': thumbnail,
+            'start_time': start_time,
+            'stop_time': stop_time,
         }

youtube_dl/extractor/youtube.py

@@ -692,9 +692,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         'invidious': '|'.join(_INVIDIOUS_SITES),
     }
     _PLAYER_INFO_RE = (
-        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$',
-        r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.js$',
+        r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})/(?:tv-)?player',
+        r'/(?P<id>[a-zA-Z0-9_-]{8,})/player(?:_ias(?:_tce)?\.vflset(?:/[a-zA-Z]{2,3}_[a-zA-Z]{2,3})?|-plasma-ias-(?:phone|tablet)-[a-z]{2}_[A-Z]{2}\.vflset)/base\.js$',
+        r'\b(?P<id>vfl[a-zA-Z0-9_-]{6,})\b.*?\.js$',
     )
     _SUBTITLE_FORMATS = ('json3', 'srv1', 'srv2', 'srv3', 'ttml', 'vtt')
 
@@ -1626,15 +1626,13 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         """ Return a string representation of a signature """
         return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
 
-    @classmethod
-    def _extract_player_info(cls, player_url):
-        for player_re in cls._PLAYER_INFO_RE:
-            id_m = re.search(player_re, player_url)
-            if id_m:
-                break
-        else:
-            raise ExtractorError('Cannot identify player %r' % player_url)
-        return id_m.group('id')
+    def _extract_player_info(self, player_url):
+        try:
+            return self._search_regex(
+                self._PLAYER_INFO_RE, player_url, 'player info', group='id')
+        except ExtractorError as e:
+            raise ExtractorError(
+                'Cannot identify player %r' % (player_url,), cause=e)
 
     def _load_player(self, video_id, player_url, fatal=True, player_id=None):
         if not player_id:
@@ -1711,6 +1709,23 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 '    return %s\n') % (signature_id_tuple, expr_code)
         self.to_screen('Extracted signature function:\n' + code)
 
+    def _extract_sig_fn(self, jsi, funcname):
+        var_ay = self._search_regex(
+            r'''(?x)
+                (?:\*/|\{|\n|^)\s*(?:'[^']+'\s*;\s*)
+                    (var\s*[\w$]+\s*=\s*(?:
+                        ('|")(?:\\\2|(?!\2).)+\2\s*\.\s*split\(\s*('|")\W+\3\s*\)|
+                        \[\s*(?:('|")(?:\\\4|(?!\4).)*\4\s*(?:(?=\])|,\s*))+\]
+                    ))(?=\s*[,;])
+            ''', jsi.code, 'useful values', default='')
+
+        sig_fn = jsi.extract_function_code(funcname)
+
+        if var_ay:
+            sig_fn = (sig_fn[0], ';\n'.join((var_ay, sig_fn[1])))
+
+        return sig_fn
+
     def _parse_sig_js(self, jscode):
         # Examples where `sig` is funcname:
         # sig=function(a){a=a.split(""); ... ;return a.join("")};
@@ -1736,8 +1751,12 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             jscode, 'Initial JS player signature function name', group='sig')
 
         jsi = JSInterpreter(jscode)
-        initial_function = jsi.extract_function(funcname)
-        return lambda s: initial_function([s])
+
+        initial_function = self._extract_sig_fn(jsi, funcname)
+
+        func = jsi.extract_function_from_code(*initial_function)
+
+        return lambda s: func([s])
 
     def _cached(self, func, *cache_id):
         def inner(*args, **kwargs):
@@ -1856,15 +1875,9 @@ 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\("\W+"\))(?=\s*[,;])',
-            jsi.code, 'useful values', default='')
-
         func_name = self._extract_n_function_name(jsi.code)
 
-        func_code = jsi.extract_function_code(func_name)
-        if var_ay:
-            func_code = (func_code[0], ';\n'.join((var_ay, func_code[1])))
+        func_code = self._extract_sig_fn(jsi, func_name)
 
         if player_id:
             self.cache.store('youtube-nsig', player_id, func_code)
@@ -2136,7 +2149,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                     video_details = merge_dicts(*traverse_obj(
                         (player_response, api_player_response),
                         (Ellipsis, 'videoDetails', T(dict))))
-                    player_response.update(api_player_response or {})
+                    player_response.update(filter_dict(
+                        api_player_response or {}, cndn=lambda k, _: k != 'captions'))
                     player_response['videoDetails'] = video_details
 
         def is_agegated(playability):
@@ -2566,8 +2580,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         }
 
         pctr = traverse_obj(
-            player_response,
-            ('captions', 'playerCaptionsTracklistRenderer', T(dict)))
+            (player_response, api_player_response),
+            (Ellipsis, 'captions', 'playerCaptionsTracklistRenderer', T(dict)))
         if pctr:
             def process_language(container, base_url, lang_code, query):
                 lang_subs = []
@@ -2584,20 +2598,21 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             def process_subtitles():
                 subtitles = {}
                 for caption_track in traverse_obj(pctr, (
-                        'captionTracks', lambda _, v: v.get('baseUrl'))):
+                        Ellipsis, 'captionTracks', lambda _, v: (
+                            v.get('baseUrl') and v.get('languageCode')))):
                     base_url = self._yt_urljoin(caption_track['baseUrl'])
                     if not base_url:
                         continue
+                    lang_code = caption_track['languageCode']
                     if caption_track.get('kind') != 'asr':
-                        lang_code = caption_track.get('languageCode')
-                        if not lang_code:
-                            continue
                         process_language(
                             subtitles, base_url, lang_code, {})
                         continue
                     automatic_captions = {}
+                    process_language(
+                        automatic_captions, base_url, lang_code, {})
                     for translation_language in traverse_obj(pctr, (
-                            'translationLanguages', lambda _, v: v.get('languageCode'))):
+                            Ellipsis, 'translationLanguages', lambda _, v: v.get('languageCode'))):
                         translation_language_code = translation_language['languageCode']
                         process_language(
                             automatic_captions, base_url, translation_language_code,

youtube_dl/jsinterp.py

@@ -678,7 +678,7 @@ class JSInterpreter(object):
             return len(obj)
         try:
             return obj[int(idx)] if isinstance(obj, list) else obj[compat_str(idx)]
-        except (TypeError, KeyError, IndexError) as e:
+        except (TypeError, KeyError, IndexError, ValueError) as e:
             # allow_undefined is None gives correct behaviour
             if allow_undefined or (
                     allow_undefined is None and not isinstance(e, TypeError)):
@@ -1038,6 +1038,10 @@ class JSInterpreter(object):
                     left_val = self._index(left_val, idx)
             if isinstance(idx, float):
                 idx = int(idx)
+            if isinstance(left_val, list) and len(left_val) <= int_or_none(idx, default=-1):
+                # JS Array is a sparsely assignable list
+                # TODO: handle extreme sparsity without memory bloat, eg using auxiliary dict
+                left_val.extend((idx - len(left_val) + 1) * [JS_Undefined])
             left_val[idx] = self._operator(
                 m.group('op'), self._index(left_val, idx) if m.group('op') else None,
                 m.group('expr'), expr, local_vars, allow_recursion)
@@ -1204,9 +1208,10 @@ class JSInterpreter(object):
                 elif member == 'join':
                     assertion(isinstance(obj, list), 'must be applied on a list')
                     assertion(len(argvals) <= 1, 'takes at most one argument')
-                    return (',' if len(argvals) == 0 else argvals[0]).join(
-                        ('' if x in (None, JS_Undefined) else _js_toString(x))
-                        for x in obj)
+                    return (',' if len(argvals) == 0 or argvals[0] in (None, JS_Undefined)
+                            else argvals[0]).join(
+                                ('' if x in (None, JS_Undefined) else _js_toString(x))
+                                for x in obj)
                 elif member == 'reverse':
                     assertion(not argvals, 'does not take any arguments')
                     obj.reverse()
@@ -1364,19 +1369,21 @@ class JSInterpreter(object):
         code, _ = self._separate_at_paren(func_m.group('code'))  # refine the match
         return self.build_arglist(func_m.group('args')), code
 
-    def extract_function(self, funcname):
+    def extract_function(self, funcname, *global_stack):
         return function_with_repr(
-            self.extract_function_from_code(*self.extract_function_code(funcname)),
+            self.extract_function_from_code(*itertools.chain(
+                self.extract_function_code(funcname), global_stack)),
             'F<%s>' % (funcname,))
 
     def extract_function_from_code(self, argnames, code, *global_stack):
         local_vars = {}
 
+        start = None
         while True:
-            mobj = re.search(r'function\((?P<args>[^)]*)\)\s*{', code)
+            mobj = re.search(r'function\((?P<args>[^)]*)\)\s*{', code[start:])
             if mobj is None:
                 break
-            start, body_start = mobj.span()
+            start, body_start = ((start or 0) + x for x in mobj.span())
             body, remaining = self._separate_at_paren(code[body_start - 1:])
             name = self._named_object(local_vars, self.extract_function_from_code(
                 [x.strip() for x in mobj.group('args').split(',')],