Merge 20691a11e6ade1bac6f357ccb1c4808854427c16 into 2b4fbfce25902d557b86b003cf48f738129efce4

[YouTube] Support player 4fcd6e4a
thx seproDev, bashonly: yt-dlp/yt-dlp#12748
2025-05-29 17:12:41 +09:00 · 2025-03-26 07:21:52 +00:00 · 2025-03-26 02:27:25 +00:00 · 2025-03-25 22:35:06 +00:00 · 2025-03-25 22:35:06 +00:00 · 2025-03-25 22:35:06 +00:00
5 changed files with 234 additions and 106 deletions
--- a/test/test_youtube_signature.py
+++ b/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)

--- a/youtube_dl/extractor/extractors.py
+++ b/youtube_dl/extractor/extractors.py
@ -1065,7 +1065,7 @@ from .rtl2 import (
 )
 from .rtp import RTPIE
 from .rts import RTSIE
-from .rtve import RTVEALaCartaIE, RTVELiveIE, RTVEInfantilIE, RTVELiveIE, RTVETelevisionIE
+from .rtve import RTVEPlayIE, RTVEInfantilIE, RTVELiveIE, RTVETelevisionIE
 from .rtvnh import RTVNHIE
 from .rtvs import RTVSIE
 from .ruhd import RUHDIE
--- a/youtube_dl/extractor/rtve.py
+++ b/youtube_dl/extractor/rtve.py
@ -12,30 +12,30 @@ from ..compat import (
    compat_struct_unpack,
 )
 from ..utils import (
+    clean_html,
    determine_ext,
    ExtractorError,
    float_or_none,
    qualities,
-    remove_end,
-    remove_start,
    std_headers,
 )

 _bytes_to_chr = (lambda x: x) if sys.version_info[0] == 2 else (lambda x: map(chr, x))


-class RTVEALaCartaIE(InfoExtractor):
-    IE_NAME = 'rtve.es:alacarta'
-    IE_DESC = 'RTVE a la carta'
-    _VALID_URL = r'https?://(?:www\.)?rtve\.es/(m/)?(alacarta/videos|filmoteca)/[^/]+/[^/]+/(?P<id>\d+)'
+class RTVEPlayIE(InfoExtractor):
+    IE_NAME = 'rtve.es:play'
+    IE_DESC = 'RTVE Play'
+    _VALID_URL = r'https?://(?:www\.)?rtve\.es/(?P<kind>(?:playz?|(?:m/)?alacarta)/(?:audios|videos)|filmoteca)/[^/]+/[^/]+/(?P<id>\d+)'

    _TESTS = [{
        'url': 'http://www.rtve.es/alacarta/videos/balonmano/o-swiss-cup-masculina-final-espana-suecia/2491869/',
-        'md5': '1d49b7e1ca7a7502c56a4bf1b60f1b43',
+        'md5': '2c70aacf8a415d1b4e7fcc0525951162',
        'info_dict': {
            'id': '2491869',
            'ext': 'mp4',
-            'title': 'Balonmano - Swiss Cup masculina. Final: España-Suecia',
+            'title': 'Final de la Swiss Cup masculina: España-Suecia',
+            'description': 'Swiss Cup masculina, Final: España-Suecia.',
            'duration': 5024.566,
            'series': 'Balonmano',
        },
@ -47,6 +47,7 @@ class RTVEALaCartaIE(InfoExtractor):
            'id': '1694255',
            'ext': 'mp4',
            'title': 're:^24H LIVE [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
+            'description': '24H LIVE',
            'is_live': True,
        },
        'params': {
@ -54,11 +55,12 @@ class RTVEALaCartaIE(InfoExtractor):
        },
    }, {
        'url': 'http://www.rtve.es/alacarta/videos/servir-y-proteger/servir-proteger-capitulo-104/4236788/',
-        'md5': 'd850f3c8731ea53952ebab489cf81cbf',
+        'md5': '30b8827cba25f39d1af5a7c482cc8ac5',
        'info_dict': {
            'id': '4236788',
            'ext': 'mp4',
-            'title': 'Servir y proteger - Capítulo 104',
+            'title': 'Capítulo 104',
+            'description': 'md5:caae29ae04291875e611dd667fe84641',
            'duration': 3222.0,
        },
        'expected_warnings': ['Failed to download MPD manifest', 'Failed to download m3u8 information'],
@ -68,6 +70,17 @@ class RTVEALaCartaIE(InfoExtractor):
    }, {
        'url': 'http://www.rtve.es/filmoteca/no-do/not-1-introduccion-primer-noticiario-espanol/1465256/',
        'only_matching': True,
+    }, {
+        'url': 'http://www.rtve.es/alacarta/audios/a-hombros-de-gigantes/palabra-ingeniero-codigos-informaticos-27-04-21/5889192/',
+        'md5': 'ae06d27bff945c4e87a50f89f6ce48ce',
+        'info_dict': {
+            'id': '5889192',
+            'ext': 'mp3',
+            'title': 'Códigos informáticos',
+            'description': 'md5:72b0d7c1ca20fd327bdfff7ac0171afb',
+            'thumbnail': r're:https?://.+/1598856591583.jpg',
+            'duration': 349.440,
+        },
    }]

    def _real_initialize(self):
@ -86,8 +99,12 @@ class RTVEALaCartaIE(InfoExtractor):
                break
            data = encrypted_data.read(length)
            if chunk_type == b'tEXt':
-                alphabet_data, text = data.split(b'\0')
-                quality, url_data = text.split(b'%%')
+                alphabet_data, text = data.replace(b'\0', b'').split(b'#')
+                components = text.split(b'%%')
+                if len(components) < 2:
+                    components.insert(0, b'')
+                quality, url_data = components
+
                alphabet = []
                e = 0
                d = 0
@ -120,7 +137,7 @@ class RTVEALaCartaIE(InfoExtractor):

    def _extract_png_formats(self, video_id):
        png = self._download_webpage(
-            'http://www.rtve.es/ztnr/movil/thumbnail/%s/videos/%s.png' % (self._manager, video_id),
+            'http://ztnr.rtve.es/ztnr/movil/thumbnail/%s/videos/%s.png' % (self._manager, video_id),
            video_id, 'Downloading url information', query={'q': 'v2'})
        q = qualities(['Media', 'Alta', 'HQ', 'HD_READY', 'HD_FULL'])
        formats = []
@ -143,11 +160,16 @@ class RTVEALaCartaIE(InfoExtractor):
        return formats

    def _real_extract(self, url):
-        video_id = self._match_id(url)
+        groups = re.match(self._VALID_URL, url).groupdict()
+        is_audio = groups.get('kind') == 'play/audios'
+        return self._real_extract_from_id(groups['id'], is_audio)
+
+    def _real_extract_from_id(self, video_id, is_audio=False):
+        kind = 'audios' if is_audio else 'videos'
        info = self._download_json(
-            'http://www.rtve.es/api/videos/%s/config/alacarta_videos.json' % video_id,
+            'http://www.rtve.es/api/%s/%s.json' % (kind, video_id),
            video_id)['page']['items'][0]
-        if info['state'] == 'DESPU':
+        if (info.get('pubState') or {}).get('code') == 'DESPU':
            raise ExtractorError('The video is no longer available', expected=True)
        title = info['title'].strip()
        formats = self._extract_png_formats(video_id)
@ -157,17 +179,19 @@ class RTVEALaCartaIE(InfoExtractor):
        if sbt_file:
            subtitles = self.extract_subtitles(video_id, sbt_file)

-        is_live = info.get('live') is True
+        is_live = info.get('consumption') == 'live'

        return {
            'id': video_id,
            'title': self._live_title(title) if is_live else title,
            'formats': formats,
-            'thumbnail': info.get('image'),
+            'url': info.get('htmlUrl'),
+            'description': clean_html(info.get('description')),
+            'thumbnail': info.get('thumbnail'),
            'subtitles': subtitles,
            'duration': float_or_none(info.get('duration'), 1000),
            'is_live': is_live,
-            'series': info.get('programTitle'),
+            'series': (info.get('programInfo') or {}).get('title'),
        }

    def _get_subtitles(self, video_id, sub_file):
@ -179,36 +203,60 @@ class RTVEALaCartaIE(InfoExtractor):
            for s in subs)


-class RTVEInfantilIE(RTVEALaCartaIE):
+class RTVEInfantilIE(RTVEPlayIE):
    IE_NAME = 'rtve.es:infantil'
    IE_DESC = 'RTVE infantil'
    _VALID_URL = r'https?://(?:www\.)?rtve\.es/infantil/serie/[^/]+/video/[^/]+/(?P<id>[0-9]+)/'

    _TESTS = [{
-        'url': 'http://www.rtve.es/infantil/serie/cleo/video/maneras-vivir/3040283/',
-        'md5': '5747454717aedf9f9fdf212d1bcfc48d',
+        'url': 'https://www.rtve.es/infantil/serie/dino-ranch/video/pequeno-gran-ayudante/6693248/',
+        'md5': '06d3f57eec593ad93fe9dcf079fbd940',
        'info_dict': {
-            'id': '3040283',
+            'id': '6693248',
            'ext': 'mp4',
-            'title': 'Maneras de vivir',
-            'thumbnail': r're:https?://.+/1426182947956\.JPG',
-            'duration': 357.958,
+            'title': 'Un pequeño gran ayudante',
+            'description': 'md5:144ca351e31f9ee99a637ab9fc2787d5',
+            'thumbnail': r're:https?://.+/1663318364501\.jpg',
+            'duration': 691.44,
        },
        'expected_warnings': ['Failed to download MPD manifest', 'Failed to download m3u8 information'],
    }]


-class RTVELiveIE(RTVEALaCartaIE):
+class RTVELiveIE(RTVEPlayIE):
    IE_NAME = 'rtve.es:live'
    IE_DESC = 'RTVE.es live streams'
-    _VALID_URL = r'https?://(?:www\.)?rtve\.es/directo/(?P<id>[a-zA-Z0-9-]+)'
+    _VALID_URL = r'https?://(?:www\.)?rtve\.es/play/videos/directo/(?P<id>.+)'

    _TESTS = [{
-        'url': 'http://www.rtve.es/directo/la-1/',
+        'url': 'https://www.rtve.es/play/videos/directo/la-1/',
        'info_dict': {
-            'id': 'la-1',
+            'id': '1688877',
            'ext': 'mp4',
            'title': 're:^La 1 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
+            'description': 'La 1',
+        },
+        'params': {
+            'skip_download': 'live stream',
+        }
+    }, {
+        'url': 'https://www.rtve.es/play/videos/directo/canales-lineales/la-1/',
+        'info_dict': {
+            'id': '1688877',
+            'ext': 'mp4',
+            'title': 're:^La 1 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
+            'description': 'La 1',
+        },
+        'params': {
+            'skip_download': 'live stream',
+        }
+    }, {
+        'url': 'https://www.rtve.es/play/videos/directo/canales-lineales/capilla-ardiente-isabel-westminster/10886/',
+        'info_dict': {
+            'id': '1938028',
+            'ext': 'mp4',
+            'title': 're:^Mas24 - 1 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
+            'description': 'Mas24 - 1',
        },
        'params': {
            'skip_download': 'live stream',
@ -216,53 +264,77 @@ class RTVELiveIE(RTVEALaCartaIE):
    }]

    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-
-        webpage = self._download_webpage(url, video_id)
-        title = remove_end(self._og_search_title(webpage), ' en directo en RTVE.es')
-        title = remove_start(title, 'Estoy viendo ')
-
-        vidplayer_id = self._search_regex(
-            (r'playerId=player([0-9]+)',
-             r'class=["\'].*?\blive_mod\b.*?["\'][^>]+data-assetid=["\'](\d+)',
-             r'data-id=["\'](\d+)'),
+        webpage = self._download_webpage(url, self._match_id(url))
+        asset_id = self._search_regex(
+            r'class=["\'].*?\bvideoPlayer\b.*?["\'][^>]+data-setup=[^>]+?(?:"|&quot;)idAsset(?:"|&quot;)\s*:\s*(?:"|&quot;)(\d+)(?:"|&quot;)',
            webpage, 'internal video ID')
-
-        return {
-            'id': video_id,
-            'title': self._live_title(title),
-            'formats': self._extract_png_formats(vidplayer_id),
-            'is_live': True,
-        }
+        return self._real_extract_from_id(asset_id)


 class RTVETelevisionIE(InfoExtractor):
    IE_NAME = 'rtve.es:television'
-    _VALID_URL = r'https?://(?:www\.)?rtve\.es/television/[^/]+/[^/]+/(?P<id>\d+).shtml'
+    # https://www.rtve.es/SECTION/YYYYMMDD/CONTENT_SLUG/CONTENT_ID.shtml
+    _VALID_URL = r'https?://(?:www\.)?rtve\.es/[^/]+/\d{8}/[^/]+/(?P<id>\d+)\.shtml'

-    _TEST = {
-        'url': 'http://www.rtve.es/television/20160628/revolucion-del-movil/1364141.shtml',
+    _TESTS = [{
+        'url': 'https://www.rtve.es/television/20220916/destacados-festival-san-sebastian-rtve-play/2395620.shtml',
        'info_dict': {
-            'id': '3069778',
+            'id': '6668919',
            'ext': 'mp4',
-            'title': 'Documentos TV - La revolución del móvil',
-            'duration': 3496.948,
+            'title': 'Las películas del Festival de San Sebastián en RTVE Play',
+            'description': 'El\xa0Festival de San Sebastián vuelve a llenarse de artistas. Y en su honor,\xa0RTVE Play\xa0destacará cada viernes una\xa0película galardonada\xa0con la\xa0Concha de Oro\xa0en su catálogo.',
+            'duration': 20.048,
        },
        'params': {
            'skip_download': True,
        },
-    }
+    }, {
+        'url': 'https://www.rtve.es/noticias/20220917/penelope-cruz-san-sebastian-premio-nacional/2402565.shtml',
+        'info_dict': {
+            'id': '6694087',
+            'ext': 'mp4',
+            'title': 'Penélope Cruz recoge el Premio Nacional de Cinematografía: "No dejen nunca de proteger nuestro cine"',
+            'description': 'md5:eda9e6baa78dbbbcc7708c0cc8150a91',
+            'duration': 388.2,
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        'url': 'https://www.rtve.es/deportes/20220917/motogp-bagnaia-pole-marquez-decimotercero-motorland-aragon/2402566.shtml',
+        'info_dict': {
+            'id': '6694142',
+            'ext': 'mp4',
+            'title': "Bagnaia logra su quinta 'pole' del año y Márquez partirá decimotercero",
+            'description': 'md5:07e2ccb983a046cb42f896cce225f0a7',
+            'duration': 153.44,
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        'url': 'https://www.rtve.es/playz/20220807/covaleda-fest-final/2394809.shtml',
+        'info_dict': {
+            'id': '6665408',
+            'ext': 'mp4',
+            'title': 'Covaleda Fest (Soria) - Día 3 con Marc Seguí y Paranoid 1966',
+            'description': 'Festivales Playz viaja a Covaleda, Soria, para contarte todo lo que sucede en el Covaleda Fest. Entrevistas, challenges a los artistas, juegos... Khan, Adriana Jiménez y María García no dejarán pasar ni una. ¡No te lo pierdas!',
+            'duration': 12009.92,
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }]

    def _real_extract(self, url):
        page_id = self._match_id(url)
        webpage = self._download_webpage(url, page_id)

        alacarta_url = self._search_regex(
-            r'data-location="alacarta_videos"[^<]+url&quot;:&quot;(http://www\.rtve\.es/alacarta.+?)&',
+            r'data-location="alacarta_videos"[^<]+url&quot;:&quot;(https?://www\.rtve\.es/play.+?)&',
            webpage, 'alacarta url', default=None)
        if alacarta_url is None:
            raise ExtractorError(
                'The webpage doesn\'t contain any video', expected=True)

-        return self.url_result(alacarta_url, ie=RTVEALaCartaIE.ie_key())
+        return self.url_result(alacarta_url, ie=RTVEPlayIE.ie_key())
--- a/youtube_dl/extractor/youtube.py
+++ b/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,
--- a/youtube_dl/jsinterp.py
+++ b/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(',')],
Author	SHA1	Message	Date
Alba Mendez	c55e764000	Merge 20691a11e6ade1bac6f357ccb1c4808854427c16 into 2b4fbfce25902d557b86b003cf48f738129efce4	2025-03-26 07:21:52 +00:00
dirkf	2b4fbfce25	[YouTube] Support player `4fcd6e4a` thx seproDev, bashonly: yt-dlp/yt-dlp#12748	2025-03-26 02:27:25 +00:00
dirkf	1bc45b8b6c	[JSInterp] Use `,` for join() with null/undefined argument Eg: [1,2,3].join(null) -> '1,2,3'	2025-03-25 22:35:06 +00:00
dirkf	b982d77d0b	[YouTube] Align signature tests with yt-dlp thx bashonly, yt-dlp/yt-dlp#12725	2025-03-25 22:35:06 +00:00
dirkf	c55dbf4838	[YouTube] Update signature extraction for players `643afba4`, `363db69b`	2025-03-25 22:35:06 +00:00
dirkf	087d865230	[YouTube] Support new player URL patterns	2025-03-25 22:35:06 +00:00
dirkf	a4fc1151f1	[JSInterp] Improve indexing * catch invalid list index with `ValueError` (eg [1, 2]['ab'] -> undefined) * allow assignment outside existing list (eg var l = [1,2]; l[9] = 0;)	2025-03-25 22:35:05 +00:00
dirkf	a464c159e6	[YouTube] Make `_extract_player_info()` use `_search_regex()`	2025-03-25 22:35:05 +00:00
dirkf	7dca08eff0	[YouTube] Also get original of translated automatic captions	2025-03-25 22:35:05 +00:00
dirkf	2239ee7965	[YouTube] Get subtitles/automatic captions from both web and API responses	2025-03-25 22:35:05 +00:00
dirkf	20691a11e6	Really pass TestAllURLsMatching.test_no_duplicates	2022-10-10 10:54:19 +01:00
dirkf	585e806d9a	Pass TestAllURLsMatching.test_no_duplicates ?	2022-10-10 10:21:59 +01:00
Alba Mendez	e4c57418df	generalize to audios too	2022-09-17 20:24:15 +02:00
Alba Mendez	d7939e2c07	improve metadata grab metadata from the new endpoint (videos/%d.json) instead of the legacy endpoint (videos/%d/config/alacarta_videos.json), which is what is actually used for the UI now. - gives more up to date titles - adds description / URL - makes supporting audios easy (see next commit)	2022-09-17 20:24:15 +02:00
Alba Mendez	4a48a17eea	make tests up to date	2022-09-17 20:24:15 +02:00
Alba Mendez	52858d5879	generalize to other URLs	2022-09-17 20:24:15 +02:00
Alba Mendez	050b52baf9	rename rtve.es:alacarta to rtve.es:play	2022-09-17 20:24:15 +02:00
Alba Mendez	11bd5b9612	fix rtve.es:live	2022-09-17 20:24:15 +02:00
Alba Mendez	33650c8eb6	fix URLs change to ztnr.rtve.es domain doesn't seem to be required, but switch to it just in case the old route is dropped someday	2022-09-17 20:24:15 +02:00
Alba Mendez	7cc35d1051	fix decoding logic	2022-09-17 20:24:15 +02:00